Analysis of vertical multijunction solar cells using a distributed circuit model

Vertical multijunction (VMJ) solar cells have generated considerable interest due to their improved performance in terms of conversion efficiency and radiation tolerance compared to the conventional planar solar cells. Fabrication of VMJ cells with junction density of 2000 junctions/cm is now possible using advanced fabrication technologies. This work describes an analysis of some of the VMJ cell structures now being fabricated—especially the ones that combine the enhanced red response and radiation tolerance of the VMJ concept and blue response of conventional planar cells. A distributed equivalent circuit model is used for analysis of complicated junction configurations—which otherwise would be very cumbersome using conventional carrier transport equations.

The VMJ cell structures were analyzed to study their device characteristics and their sensitivity to various material and fabrication parameters such as epitaxial layer resistivity and carrier lifetimes. The results show that the conversion efficiency is higher than conventional devices due to efficient carrier collection with a superior radiation tolerance. The cells, however, degrade more rapidly compared to planar cells at higher radiation levels determined by the structural parameters.     

Characteristic features of silicon multijunction solar cells with vertical p-n junctions

A relatively simple technology (without photolithography) based on diffusion welding and ion-plasma deposition of an insulating coating has been developed for fabricating multijunction silicon solar cells with vertical p-n junctions. The effective collection factor for such structures is independent of the wavelength of the incident light in the wavelength range λ=340–1080 nm. Fiz. Tekh. Poluprovodn. 31, 855–857 (July 1997)     

Modeling the efficiency of multijunction solar cells

The efficiency of multijunction solar cells (MSCs) η is calculated taking into account radiative recombination, Shockley-Read recombination, front and rear surface recombination, recombination in the space-charge regions, and recombination at heterojunctions. Calculation is performed by self-consistent solution of the photocurrent, photovoltage, and heat-balance equations. MSC cooling by increasing the numbers of cells n and improvement in the conditions of heat removal is taken into account. An effect leading to a decrease in the photocurrent with increasing n, associated with narrowing of the energy ranges of photons incident on the MSC cell, is considered. It is found that a significant increase in the MSC efficiency can be achieved by improving the heat-removal conditions, in particular, through the use of radiators and increasing the MSC grayness factor to unity. The results obtained are compared to those of other authors. It is shown that the calculated dependences η(n) are in agreement with experimental values.     

Photoelectric determination of the series resistance of multijunction solar cells

A method for determining the series resistance R _s of multijunction solar cells is suggested and sub-stantiated. The method uses the presence of a maximum in the dependence of the efficiency on the sunlight concentration ratio ??(X) or in that of the operating voltage on the photogenerated current, V _m (J _g ). The study employs the concept that, in a limited but practically important range of photogenerated currents (up to the maximum ??), the series resistance can be represented by a fixed quantity that is linear and independent of J _g . It is analytically substantiated that this resistance can be found from the formula R _S = (E/J _g )_{?? = max}, where E = AkT/q and A and J _g are local values of the ideality factor and photogenerated current at the maximum ?? (or V _m ). It is shown that the value of R _s , determined by this method, is independent of the spectral composition of the incident light, which was experimentally confirmed in a study of the photovoltaic characteristics of triple-junction InGaP/GaAs/Ge solar cells. The method is suitable for both multi- and single-junction photoelectric converters.     

Cost trade between multijunction,gallium arsenide and silicon solar cells

Multijunction (MJ),¹ gallium arsenide (GaAs) and silicon (Si) solar cells have respective test efficiencies of approximately 24%, 18.5% and 14.8%. Multijunction and gallium arsenide solar cells weigh more than silicon solar cells and cost approximately five times as much per unit power at the cell level.² A trade is performed for the Tropical Rainfall Measuring Mission (TRMM) spacecraft to determine which of these cell types would have offered an overall performance and price advantage to the spacecraft. A trade is also performed for the multijunction cells under the assumption that they will cost over ten times that of silicon cells at the cell level. The trade shows that the TRMM project, less the cost of the instrument, ground systems and mission operations, would spend approximately $552 000 per kilogram to launch and support

1 The term ‘support’ means to provide the scientific equipment (‘science’) with environmental protection from the space environ ment, to keep it at acceptable temperatures and to send the data it produces to Earth in readable form, i.e. the services provided by the spacecraft.

science in the case of the spacecraft equipped with silicon solar cells. If these cells are exch anged for gallium arsenide solar cells, an additional 31 kg of science can be launched and serviced at a price of approximately $90 000 per kilogram. The 31 kg array weight reduction is shown to derive from the smaller area of the array and hence reductio ns in the weight of the array substrate and supporting structure. If the silicon solar cells are changed out for multijunction solar cells, an additional 45 kg of science above the silicon baseline can be launched and supported at a price of approximately $58 000 per kilogram. The trade shows that even if the multijunction cells are priced over ten times that of silicon cells, a price that is much higher than projected, the additional 45 kg of science are launched and serviced at $180 000 per kilogram. This is still much less than the original $552 000 per kilogram to launch and service the science. Data and qualitative factors are presented to show that these figures are subject to a great deal of uncertainty. Nonetheless, the benefit of the higher efficiency solar cells for TRMM is far greater than the uncertainties in the analysis. © This article is a US Government work and, as such, is in the public domain in the United States of America     

Ideal efficiency of monolithic,series‐connected multijunction solar cells

Two‐terminal, monolithic tandem solar cells represent the most interesting implementation of tThe multijunction photovoltaic array capable of very high efficiency. Radiative coupling among the cells, and between the cells and the substrate, along with series interconnection are features present in this approach. Their effect on efficiency is quantified for arrays with a small number of cells in the radiative limit and compared with the best experimental cells to assess their potential improvement. Finally, it is shown that they do not lower the limiting efficiency of the infinite tandem array, which is determined by the emission losses through the illuminated face. Copyright © 2002 John Wiley & Sons, Ltd.     

Study of minority carrier diffusion lengths in photoactive layers of multijunction solar cells

A technique for determining a minority carrier’s diffusion length in photoactive III–V layers of solar cells by approximating their spectral characteristics is presented. Single-junction GaAs, Ge and multi-junction GaAs/Ge, GaInP/GaAs, and GaInP/GaInAs/Ge solar cells fabricated by hydride metal-organic vapor-phase epitaxy (H-MOVPE) have been studied. The dependences of the minority carrier diffusion length on the doping level of p-Ge and n-GaAs are determined. It is shown that the parameters of solid-state diffusion of phosphorus atoms to the p-Ge substrate from the n-GaInP nucleation layer are independent of the thickness of the latter within 35–300 nm. It is found that the diffusion length of subcells of multijunction structures in Ga(In)As layers is smaller in comparison with that of single-junction structures.     

Design of multijunction GaPNAs/Si heterostructure solar cells by computer simulation

The designs of two- and three-junction solar cells based on GaPNAs/Si lattice-matched hetero-structures are calculated. It is shown that the efficiency of two-junction solar cells constituted by a junction based on a GaPNAs solid solution with a band gap E _g of 1.78 eV and a junction based on Si may reach a value of 30.3% under AM1.5 D, 100 mW/cm², and 35.4% under AM1.5D, 20 W/cm². The maximum values of the efficiency of the three-junction solar cell constituted by top and middle junctions based on GaPNAs with E _g of 2 and 1.5 eV, respectively, and a Si-based bottom junction are 39.2% under AM1.5 D, 100 mW/cm², and 44.5% under AM1.5D, 20 W/cm². It is shown that the thickness and minority carrier lifetime of the photoactive layers affect the efficiency of solar-light conversion by the heterostructures being developed.     

Advances in amorphous silicon alloy technology—the achievement of high-efficiency multijunction solar cells and modules

Thin-film amorphous silicon alloy technology has emerged as a strong contender for providing low-cost photovoltaic products to meet the energy needs of the 21st century. World record 14·6% initial and 13·0% stable active-area conversion efficiencies have been achieved for small-area solar cells in our laboratory using a spectral-splitting, triple-junction structure. This 0·25-cm² device exhibited a total-area efficiency of 12·0%, as confirmed by the National Renewable Energy Laboratory. Key factors leading to this achievement will be discussed. We have also demonstrated a 10·2% stable module efficiency for a one-square-foot area. A state-of-the-art, continuous roll-to-roll production line with an annual capacity of 5 MW has been built and started production. Building-integrated photovoltaic applications are being demonstrated worldwide. Public awareness of environmental and ecological issues has provided a great opportunity for the maturing thin-film amorphous silicon alloy technology to demonstrate its viability as a major alternate energy source. © 1998 John Wiley & Sons, Ltd.     

Fabrication of vertical thin-GaN light-emitting diode by low-temperature Cu/Sn/Ag wafer bonding

Vertical thin-GaN LED was successfully fabricated on the GaN LED epi-layers grown on the patterned-sapphire substrate with the pyramidal pattern by low-temperature Cu/Sn/Ag wafer bonding at 150 °C. An inverted pyramidal pattern formed on the n-GaN surface after the GaN epi-layer was transferred onto Si wafer, which resulted from the pyramidal pattern on the patterned-sapphire substrate. The inverted pyramidal pattern has an equivalent function with roughening the n-GaN surface. With higher inverted pyramidal pattern coverage, the light extraction efficiency can be greatly enhanced. In addition, we found that the 4-fold increase (from 13.6% to 53.8%) in the pyramidal pattern coverage on patterned-sapphire substrate only gives the GaN LED epi-layer about 5.7% enhancement in the internal quantum efficiency.     

Development of a fabrication process for parallel multijunction thin film silicon solar cells on wafer substrates

The parallel multijunction (PMJ) cell design theoretically enables high efficiency thin film polysilicon solar cells at lower cost. Since its initial proposal in 1994 the PMJ cell has been the subject of a number of theoretical studies, however, no detailed experimental investigation has yet been reported. Any systematic study of the PMJ solar cell will require suitably designed and fabricated devices to serve as experimental test‐beds. This paper reports the successful development of a fabrication sequence for PMJ cells in CVD‐epilayers on inert single‐crystal silicon substrates, producing cells with efficiencies up to 13%. The processing sequence is based on photolithography, anisotropic wet etching, high temperature furnace steps and evaporated metallisation. Full details of the processing sequence are provided, with explanations of particular process choices, including the method of parallel electrical connection of like‐polarity layers, the use of a thick photoresist (Shipley SJR5740), avoiding pitfalls, and procedures to minimise cell shunt behaviour. The establishment of this baseline fabrication sequence for PMJ cells opens up a wealth of opportunities for systematic studies of cell performance limiting mechanisms, such as junction recombination, and the implications of various cell design and processing options, particularly those likely to be of more commercial relevance, such as laser scribing, laser doping, rapid thermal processing and electroless metal plating. Copyright © 2000 John Wiley & Sons, Ltd.     

Implications of low breakdown voltage of component subcells on external quantum efficiency measurements of multijunction solar cells

The electrical and optical coupling between subcells in a multijunction solar cell affects its external quantum efficiency (EQE) measurement. In this study, we show how a low breakdown voltage of a component subcell impacts the EQE determination of a multijunction solar cell and demands the use of a finely adjusted external voltage bias. The optimum voltage bias for the EQE measurement of a Ge subcell in two different GaInP/GaInAs/Ge triple‐junction solar cells is determined both by sweeping the external voltage bias and by tracing the I–V curve under the same light bias conditions applied during the EQE measurement. It is shown that the I–V curve gives rapid and valuable information about the adequate light and voltage bias needed, and also helps to detect problems associated with non‐ideal I–V curves that might affect the EQE measurement. The results also show that, if a non‐optimum voltage bias is applied, a measurement artifact can result. Only when the problems associated with a non‐ideal I–V curve and/or a low breakdown voltage have been discarded, the measurement artifacts, if any, can be attributed to other effects such as luminescent coupling between subcells. Copyright © 2015 John Wiley & Sons, Ltd.     

Spectral study and classification of worldwide locations considering several multijunction solar cell technologies

Multi‐junction solar cells are widely used in high‐concentration photovoltaic systems (HCPV) attaining the highest efficiencies in photovoltaic energy generation. This technology is more dependent on the spectral variations of the impinging Direct Normal Irradiance (DNI) than conventional photovoltaics based on silicon solar cells and consequently demands a deeper knowledge of the solar resource characteristics. This article explores the capabilities of spectral indexes, namely, spectral matching ratios (SMR), to spectrally characterize the annual irradiation reaching a particular location on the Earth and to provide the necessary information for the spectral optimization of a MJ solar cell in that location as a starting point for CPV module spectral tuning. Additionally, the relationship between such indexes and the atmosphere parameters, such as the aerosol optical depth (AOD), precipitable water (PW), and air mass (AM), is discussed using radiative transfer models such as SMARTS to generate the spectrally resolved DNI. The network of ground‐based sun and sky‐scanning radiometers AErosol RObotic NETwork (AERONET) is exploited to obtain the atmosphere parameters for a selected bunch of 34 sites worldwide. Finally, the SMR indexes are obtained for every location, and a comparative analysis is carried out for four architectures of triple junction solar cells, covering both lattice match and metamorphic technologies. The differences found among cell technologies are much less significant than among locations. Copyright © 2016 John Wiley & Sons, Ltd.     

Use of direct wafer bonding of silicon for fabricating solar cell structures with vertical p-n junctions

A technology based on ion implantation and the direct wafer bonding of p ⁺-p-n ⁺ structures has been developed for multijunction silicon solar cells. The internal quantum efficiency of such structures is close to unity in the wavelength range 350–900 nm. Fiz. Tekh. Poluprovodn. 32, 886–888 (July 1998)     

Optical confinement in thin-film crystalline silicon solar cells byadhesive bonding of ceramic substrate

Thin film (5-μm-thick) silicon solar cells by adhesive bonding of a near-Lambertian Al₂O₃ ceramic substrate have been fabricated, and the electrical and optical performance of the cells have been investigated. This cell structure shows a good optical confinement effect. From quantum efficiency measurements, it is found that a short circuit current that is as high as that of water-based cells can be obtained by reducing the surface reflectance. Cell and module fabrication processes can be simplified using this technique     

Evolution of silicon bulk lifetime during III–V‐on‐Si multijunction solar cell epitaxial growth

The evolution of Si bulk minority carrier lifetime during the heteroepitaxial growth of III–V on Si multijunction solar cell structures via metal‐organic chemical vapor deposition (MOCVD) has been analyzed. In particular, the impact on Si lifetime resulting from the four distinct phases within the overall MOCVD‐based III–V/Si growth process were studied: (1) the Si homoepitaxial emitter/cap layer; (2) GaP heteroepitaxial nucleation; (3) bulk GaP film growth; and (4) thick GaAs_yP_1‐y compositionally graded metamorphic buffer growth. During Phase 1 (Si homoepitaxy), an approximately two order of magnitude reduction in the Si minority carrier lifetime was observed, from about 450 to ≤1 µs. However, following the GaP nucleation (Phase 2) and thicker film (Phase 3) growths, the lifetime was found to increase by about an order of magnitude. The thick GaAs_yP_1‐y graded buffer was then found to provide further recovery back to around the initial starting value. The most likely general mechanism behind the observed lifetime evolution is as follows: lifetime degradation during Si homoepitaxy because of the formation of thermally induced defects within the Si bulk, with subsequent lifetime recovery due to passivation by fast‐diffusing atomic hydrogen coming from precursor pyrolysis, especially the group‐V hydrides (PH₃, AsH₃), during the III–V growth. These results indicate that the MOCVD growth methodology used to create these target III–V/Si solar cell structures has a substantial and dynamic impact on the minority carrier lifetime within the Si substrate. Copyright © 2015 John Wiley & Sons, Ltd.     

Studies of the Ag-In phase diagram and surface tension measurements

The phase boundaries of the Ag-In binary system were determined by the diffusion couple method, differential scanning calorimetry (DSC) and metallographic techniques. The results show that the region of the (hcp) phase is narrower than that reported previously. Thermodynamic calculation of the Ag-In system is presented by taking into account the experimental results obtained by the present and previous works, including the data on the phase equilibria and thermochemical properties. The Gibbs energies of liquid and solid solution phases are described on the basis of the sub-regular solution model, and that of the intermetallic compounds are based on the two-sublattices model. A consistent set of thermodynamic parameters has been optimized for describing the Gibbs energy of each phase, which leads to a good fit between calculated and experimental results. The maximum bubble pressure method has been used to measure the surface tension and densities of liquid In, Ag, and five binary alloys in the temperature range from 227°C to about 1170°C. ON the basis of the thermodynamic parameters of the liquid phase obtained by the present optimization, the surface tensions are calculated using Butler’s model. It is shown that the calculated values of the surface tensions are in fair agreement with the experimental data.     

Improvement of GaN-based light-emitting diode by indium-tin-oxide transparent electrode and vertical electrode

A sapphire-etched vertical-electrode nitride semiconductor (SEVENS) light-emitting diode (LED) is fabricated by means of a selective chemical wet-etching technique. The SEVENS-LED formed on sapphire substrate exhibits excellent device performance compared to a conventional NiAu lateral-electrode (LE) GaN-based LED formed on the same sapphire substrate. The integral light-output power of SEVENS-LED is /spl sim/7 mW, which is 1.75 times stronger than that of the conventional NiAu LE-LED (/spl sim/4 mW). The external quantum efficiency of SEVENS-LED is estimated to be approximately 13%.     

Phase separation-driven stratification in conventional and inverted P3HT:PCBM organic solar cells

We have used neutron reflectivity to investigate the stratification of poly(3-hexylthiophene) (P3HT) and phenyl-C₆₁-butyric acid methyl ester (PCBM) blend films. Films were spun-cast on poly(3,4-ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) and titanium oxide (TiOx) layers to mimic the procedures followed for the fabrication of conventional and inverted organic photovoltaics respectively. The resultant scattering length density profiles reveal a PCBM-rich layer is formed in the vicinity of PEDOT:PSS or TiOx, while PCBM is depleted at the free surface of the film. PCBM segregation close to the substrate is further enhanced by annealing. This stratification is considered to be favorable only for inverted devices.