We report our progress toward high-performance hydrogenated amorphous silicon (a-Si:H) solar cells fabricated in NREL's newly installed multi-chamber film Si deposition system. The a-Si: H layers are made by standard radio frequency plasma-enhanced chemical vapor deposition. This system produces a-Si: H p-i-n single-junction devices on Asahi U-type transparent conducting oxide glass with > 10% initial efficiency. The importance of the p-layer to the cell is identified: it plays a critical role in further improving cell performance. Our optimization process involves changing p-layer parameters such as dopant levels, bandgap, and thickness in cells as well as applying a double p-layer. With the optimized p-layer, we are able to increase the fill factor of our cells to as high as 72% while maintaining high open-circuit voltage.

We have developed high current density thin-film silicon n-i-p diodes for low cost and low temperature two-dimensional diode-logic memory array applications. The diodes are fabricated at temperatures below 250°C on glass, stainless steel, and plastic substrates using hot-wire chemical vapor deposition (CVD). The 0.01-mm2 standalone diodes have a forward current-density (J) of near 10 kA/cm2 and a rectification ratio over 107 at ±2 V. The 25 μm2 array diodes have J > 104 A/cm2 and rectification of 105 at ±2V. On plastic substrates, we have also used plasma-enhanced CVD to deposit 10-μm diameter diodes with J ~ 5 x 104 A/cm2. We found that the use of microcrystalline silicon µc-Si) i- and n- layers results in higher current-density diodes than with amorphous silicon. Reducing the diode area increases the forward current density by lowering the voltage drop across the external series resistances. A prototype diode array memory based on 10-micron devices was successfully demonstrated by monolithically integrating diodes with a-Si:H switching elements. High current density diodes have potential applications in a variety of large area, thin-film electronic devices, in addition to a-Si:H-based memory. This could widen the application of thin-film silicon beyond its present industrial applications in thin-film transistors, solar cells, bolometers and photo-detectors.

AlGaInP solar cells with bandgaps between 1.9 and 2.2 eV are investigated for use in next-generation multijunction photovoltaic devices. This quaternary alloy is of great importance to the development of III-V solar cells with five or more junctions and for cells optimized for operation at elevated temperatures because of the high bandgaps required in these designs. In this work, we explore the conditions for the organometallic vapor-phase epitaxy growth of AlGaInP and study their effects on cell performance. Initial efforts focused on developing ~2.0-eV AlGaInP solar cells with a nominal aluminum composition of 12%. Under the direct spectrum at 1000 W/m 2 (AM1.5D), the best of these samples had an open-circuit voltage of 1.59 V, a bandgap-voltage offset of 440 mV, a fill factor of 88.0%, and an efficiency of 14.8%. We then varied the aluminum composition of the alloy from 0% to 24% and were able to tune the bandgap of the AlGaInP layers from ~1.9 to ~2.2 eV. While the samples with a higher aluminum composition exhibited a reduced quantum efficiency and increased bandgap-voltage offset, the bandgap-voltage offset remained at 500 mV or less, up to a bandgap of ~2.1 eV.

ZnSiP2 is a potential optoelectronic material with possible application in lasers, LED's, photonic integrated circuits, and photovoltaics. The development of ZnSiP2 as a photovoltaic material could address the current technological challenge of implementing a monolithic top cell on silicon for tandem photovoltaics. In this work we present a detailed description of the growth of ZnSiP2 single crystals, which has enabled thorough optoelectronic characterization. A flux growth technique was used, under various conditions, to grow ZnSiP2 single crystals in Zn solution. The results of these growth experiments, along with analysis of previously determined phase diagrams, show that three secondary phases form as a result of the Zn flux growth technique: Zn3P2, Si, and the remaining Zn flux. Potential reasons for the formation of these particular phases are discussed, but their presence is found to be non-detrimental, and they can easily be removed. The resulting single crystals are high purity and enable the characterization of the fundamental optoelectronic properties of ZnSiP2.

The inverted metamorphic solar cell has highly tunable bandgaps, in part due to the metamorphic subcells. Using phosphide-based compositionally graded buffers, we show a wide variety of GaInAs solar cells, ranging in bandgap from 1.2 to 0.7 eV. These metamorphic subcells are all high quality and can be used for a wide variety of multijunction designs. GaInAs solar cells with 0.70 eV bandgaps are developed using an InAsP buffer that extends beyond the InP lattice constant, allowing access to an additional 2 mA/cm2 of photocurrent at AM1.5D and 25 °C. This subcell is implemented into a four-junction inverted metamorphic solar cell combined with an appropriate antireflective coating, which increases the series-connected multijunction current by 0.5 mA/cm2 with respect to designs using 0.74-eV GaInAs. However, the optimal design depends on the spectrum and operating temperature. We show how the device flexibility can be used to fine-tune the design for various spectra in order to maximize energy yield for a given operating condition. One-sun devices achieve 35.3 ± 1.2% efficiency under the AM0 spectra and 37.8 ± 1.2% efficiency under the global spectra at 25 °C. Concentrator devices designed for elevated operating temperature achieve 45.6 ± 2.3% peak efficiency under 690× the direct spectrum and 45.2 ± 2.3% efficiency at 1000× and 25 °C. Device optimization is performed for the direct spectrum on 1-sun devices with 2% shadowing, which achieve 39.8 ± 1.2% efficiency under the direct spectrum at 1 sun, highlighting the excellent performance and bandgap tunability of the four-junction inverted metamorphic solar cell.

We present recent improvements to the 4-junction inverted metamorphic solar cell. The device now includes a (Ga)InAsP buffer that transitions to lattice constants greater than InP, which allows access to GaInAs subcells with bandgaps < 0.74 eV and an additional 2 mA/cm2 of bottom junction photocurrent at AM1.5D. However, the optimal design depends on the spectrum and operating temperature. We show how the device flexibility can be used to fine-tune the design for various spectra in order to maximize energy yield for a given operating condition. 1-sun devices achieve 35.3% efficiency under the AM0 spectrum and 37.8% efficiency under the global spectrum at 25°C. Concentrator devices achieve 45.7% peak efficiency under 234x the direct spectrum and maintain over 45% efficiency at 700x at 25°C. Other device improvements include a 4-layer anti-reflection coating with low power loss, and reduced series resistance.

Multijunction solar cells can be fabricated by mechanically bonding together component cells that are grown separately. Here, we present four-junction four-terminal mechanical stacks composed of GaInP/GaAs tandems grown on GaAs substrates and GaInAsP/GaInAs tandems grown on InP substrates. The component cells were bonded together with a low-index transparent epoxy that acts as an angularly selective reflector to the GaAs bandedge luminescence, while simultaneously transmitting nearly all of the subbandgap light. As determined by electroluminescence measurements and optical modeling, the GaAs subcell demonstrates a higher internal radiative limit and, thus, higher subcell voltage, compared with GaAs subcells without the epoxy reflector. The best cells demonstrate 38.8 +/- 1.0% efficiency under the global spectrum at 1000 W/m(2) and similar to 42% under the direct spectrum at similar to 100 suns. Eliminating the series resistance is the key challenge for further improving the concentrator cells.

We present results for quadruple-junction inverted metamorphic (4J-IMM) devices under the concentrated direct spectrum and analyze the present limitations to performance. The devices integrate lattice-matched subcells with rear heterojunctions, as well as lattice-mismatched subcells with low threading dislocation density. To interconnect the subcells, thermally stable lattice-matched tunnel junctions are used, as well as a metamorphic GaAsSb/GaInAs tunnel junction between the lattice-mismatched subcells. A broadband antireflection coating is used, as well as a front metal grid designed for high concentration operation. The best device has a peak efficiency of (43.8 +/- 2.2)% at 327-sun concentration, as measured with a spectrally adjustable flash simulator, and maintains an efficiency of (42.9 +/- 2.1)% at 869 suns, which is the highest concentration measured. The V-oc increases from 3.445 V at 1-sun to 4.10 V at 327-sun concentration, which indicates high material quality in all of the subcells. The subcell voltages are analyzed using optical modeling, and the present device limitations and pathways to improvement are discussed. Although further improvements are possible, the 4J-IMM structure is clearly capable of very high efficiency at concentration, despite the complications arising from utilizing lattice-mismatched subcells.

AlGaInP solar cells with a bandgap (Eg) of ∼2.0 eV are developed for use in next-generation multijunction photovoltaic devices. This material system is of great interest for both space and concentrator photovoltaics due to its high bandgap, which enables the development of high-efficiency five-junction and six-junction devices and is also useful for solar cells operated at elevated temperatures. In this work, we explore the conditions for the Organometallic Vapor Phase Epitaxy (OMVPE) growth of AlGaInP and study their effects on cell performance. A ∼2.0 eV AlGaInP solar cell is demonstrated with an open circuit voltage (VOC) of 1.59V, a bandgap-voltage offset (WOC) of 420mV, a fill factor (FF) of 88.0%, and an efficiency of 14.8%. These AlGaInP cells have attained a similar FF, WOC and internal quantum efficiency (IQE) to the best upright GaInP cells grown in our lab to date.

Multijunction solar cells can be fabricated by bonding together component cells that are grown separately. Because the component cells are each grown lattice-matched to suitable substrates, this technique allows alloys of different lattice constants to be combined without the structural defects introduced when using metamorphic buffers. Here we present results on the fabrication and performance of four-junction mechanical stacks composed of GaInP/GaAs and GaInAsP/GaInAs tandems, grown on GaAs and InP substrates, respectively. The two tandems were bonded together with a low-index, transparent epoxy that acts as an omni-directional reflector to the GaAs bandedge luminescence, while simultaneously transmitting nearly all of the sub-bandgap light. As determined by electroluminescence measurements and optical modeling, the GaAs subcell demonstrates a higher internal radiative limit and thus higher subcell voltage, compared with GaAs subcells without enhanced internal optics; all four subcells exhibit excellent material quality. The device was fabricated with four contact terminals so that each tandem can be operated at its maximum power point, which raises the cumulative efficiency and decreases spectral sensitivity. Efficiencies exceeding 38% at one-sun have been demonstrated. Eliminating the series resistance is the key challenge for the concentrator cells. We will discuss the performance of one-sun and concentrator versions of the device, and compare the results to recently fabricated monolithic four-junction cells.

We report on a photolithographic and electro‐deposition process that results in an optimized front grid structure for high efficiency multi‐junction III–V concentrator solar cells operating under flux concentrations up to 1000 suns. Two different thick photoresists were investigated to achieve a 6 µm wide grid line with an aspect ratio of 1:1. A positive photoresist, SPR220 manufactured by Rohm and Haas was compared with a negative photoresist, nXT15 manufactured by AZ. A gold sulfite electrolyte was employed to prevent underplating as well as for environmental and safety considerations. An initial layer of nickel was discovered to be necessary to prevent delamination of the fingers during the removal of the contact layer. When deposited on a purpose grown, heavily doped GaAs contact layer, this Ni/Au contact exhibits an acceptable specific contact resistance in the low 10−4 to mid 10−5 Ohm cm2 range along with excellent adhesion without sintering. Copyright © 2014 John Wiley & Sons, Ltd.

A novel bonding approach with an interface consisting of a metal and dielectric is developed, and a pillar-array metal topology is proposed for minimal optical and electrical loss at the interface. This enables a fully lattice-matched two-terminal, four-junction device that consists of an inverted top two-junction (2J) cell with 1.85eV GaInP/1.42eV GaAs, and an upright lower 2J cell with similar to 1eV GaInAsP/0.74eV GaInAs aimed for concentrator applications. The fabrication process and simulation of the metal topology are discussed along with the results of GaAs/GaInAs 2J and (GaInP+GaAs)/GaInAs three-junction bonded cells. Bonding-related issues are also addressed along with optical coupling across the bonding interface. Copyright (c) 2014 John Wiley & Sons, Ltd.

For solar cells dominated by radiative recombination, the performance can be significantly enhanced by improving the internal optics. We demonstrate a detailed model for solar cells that calculates the external luminescent efficiency and discuss the relationship between the external and internal luminescence. The model accounts for wavelength-dependent optical properties in each layer, parasitic optical and electrical losses, multiple reflections within the cell, and assumes isotropic internal emission. For single-junction cells, the calculation leads to Voc, and for multijunction cells, the calculation leads to the Voc of each junction as well as the luminescent coupling constant. In both cases, the effects of the optics are most prominent in cells with high internal radiative efficiency. Exploiting good material quality and high luminescent coupling, we demonstrate a two-junction nonconcentrator cell with a conversion efficiency of (31.1 ± 0.9)% under the global spectrum.

Most recent gains in CdTe photovoltaic (PV) device efficiency have been in short-circuit current density (Jsc) and fill factor (FF), rather than open-circuit voltage (Voc). Because Jsc is nearing its theoretical limit, further improvements in device efficiency will require increasing Voc beyond 860 mV and increasing FF. Voc and FF may be improved by increasing both the carrier concentration and minority-carrier lifetime of the CdTe. However, Voc may be limited for other reasons, including Fermi-level pinning and surface recombination. In this study, we used doped CdTe single crystals to test whether higher carrier concentration and lifetime can overcome traditional Voc barriers. In our work to date, we have fabricated heterojunction CdTe PV cells with Voc up to 929 mV, FF of ~60%, and efficiencies of 10%.

Metal-interconnected multijunction solar cells offer one pathway toward efficiencies in excess of 50%. However, if a 3- or 4-terminal configuration is used, optical losses from the interfacial grid can be considerable. Here, we examine an alternative which provides an optimal interconnection for two-terminal bonded devices. This "pillar-array" topology is optimized by minimizing the sum of all power losses, including shadow losses and numerically computed electrical losses. Numerical modeling is used to illustrate the benefit of a pillar-array interfacial metallization for some 2-terminal configurations.

For solar cells dominated by radiative recombination, the performance can be significantly enhanced by improving the internal optics. We demonstrate a detailed model for solar cells that calculates the external luminescent efficiency and discuss the relationship between the external and internal luminescence. The model accounts for wavelength-dependent optical properties in each layer, parasitic optical and electrical losses, multiple reflections within the cell, and assumes isotropic internal emission. For single-junction cells, the calculation leads to V oc , and for multijunction cells, the calculation leads to the V oc of each junction as well as the luminescent coupling constant. In both cases, the effects of the optics are most prominent in cells with high internal radiative efficiency. Exploiting good material quality and high luminescent coupling, we demonstrate a two-junction nonconcentrator cell with a conversion efficiency of (31.1 ± 0.9)% under the global spectrum.

The ease with which screen-printed silver is deposited and its high conductivity make it an appealing choice for gridline material on CIGS-based photovoltaic devices. However, present results suggest silver diffusion into the device can cause severe reductions in efficiency after as little as 200 h at 85 °C. Dramatic reductions in fill factor, characterized by unusual inflections in the power quadrant of current-voltage curves, are observed for devices with silver gridlines but not for those with nickel or aluminum gridlines. The shape of the current-voltage curves demonstrate that the degradation mode is not simply due to changes in resistance but is consistent with the creation of a secondary barrier near the device junction.

A unique aspect of the inverted metamorphic multijunction (IMM) solar cell is the bandgap tunability of each junction, creating extremely flexible device designs. The optimal structure has subcell photocurrents that are matched for a given spectrum. However, the subcell photocurrents depend on the cell operating temperature, and therefore, the bandgaps need to be optimized for a certain range of operating conditions. In addition, imperfect material quality results in a loss of voltage and current that depends on the cell bandgap and thickness. In this case, an iterative process of multijunction design and subcell characterization is necessary to determine the optimal design. We compare two different three-junction devices to demonstrate the effect of bandgap selection and lattice-mismatched material quality on device performance at different temperatures. The triple-junction (3J)-IMM design with two lattice-mismatched junctions of perfect material quality (2MMJ) is theoretically optimal at room temperature but experimentally performs similarly to a simpler design with one mismatched junction (1MMJ) at higher temperature because of material quality tradeoffs and the temperature dependence of the designs. Significant progress in the growth, processing, and measurement has led to a 1MMJ design with (42.6 ± 2.1)% peak efficiency at 327 suns and (40.9 ± 2.0)% efficiency at 1093 suns under the direct spectrum.

The self-absorption of radiated photons increases the minority carrier concentration in semiconductor optoelectronic devices such as solar cells. This so-called photon recycling leads to an increase in the external luminescent efficiency, the fraction of internally radiated photons that are able to escape through the front surface. An increased external luminescent efficiency in turn correlates with an increased open-circuit voltage and ultimately conversion efficiency. We develop a detailed ray-optical model that calculates Voc for real, non-idealized solar cells, accounting for isotropic luminescence, parasitic losses, multiple photon reflections within the cell and wavelength-dependent indices of refraction for the layers in the cell. We have fabricated high quality GaAs solar cells, systematically varying the optical properties including the back reflectance, and have demonstrated Voc = 1.101 ± 0.002 V and conversion efficiencies of (27.8 ± 0.8)% under the global solar spectrum. The trends shown by the model are in good agreement with the data.