A. W. Haas

Multi-terminal dual junction InGaP 2 /GaAs solar cells for hybrid system

The performance of two and three-terminal solar cells under 13-sun AM1.5G illumination is reported. The solar cells are comprised of monolithically grown InGaP2 and GaAs single junction cells. The two-terminal device consists of single junctions in series. The three-terminal device structure comprises the single junctions independently interconnected by an InGaP2 layer serving as the middle contact. Device optimization was based on modeling of the InGaP2 top junction band gap for various spectral irradiance profiles. It was found that the optimal band gap combination for a 2.6eV filtered AM1.5G spectrum is achieved with 1.84eV and 1.43 eV for the top and bottom junctions, respectively. External quantum efficiency and illuminated current-voltage (I–V) measurements of the component two and three-terminal tandem cells are discussed.

Efficiency of multijunction photovoltaic systems

In this paper, an expression for the PV system efficiency is derived that can be used in conjunction with measured device performance and detailed numerical modeling to analyze PV system performance. Such an analysis will help identify design trade-offs and also help to identify which system and cell design changes will be of greatest benefit to the enhancement of PV system performance.

Peak efficiency of multijunction photovoltaic systems

The maximum efficiency of multijunction photovoltaic systems is determined for independently connected solar cells that are optically in series. The maximum efficiency is computed as a function of solar concentration using both the Shockley-Queisser detailed balance radiative limit for the reverse saturation current density and a simple empirical expression for the reverse saturation current density obtained from published “state-of-the-art” solar cells performance characteristics. It is shown that there are an optimal number of junctions for peak efficiency after which the system efficiency will decrease as the number of junctions increases.

Numerical modeling of loss mechanisms resulting from the distributed emitter effect in concentrator solar cells

Power loss associated with lateral current flow in solar cell emitter layers is known to cause fill-factor degradation. In this work, a unique approach to modeling the lateral current flow in solar cell emitter layers is developed using MATLAB™ and current-generation models based upon one-dimensional detailed numerical cell models. An example is employed to illustrate the high degree of accuracy of this approach and that in addition to joule losses in the emitter layer and conducting electrodes, a third loss mechanism, which is the result of the potential gradient across the cell surface, also plays an important role. The relationship between these loss mechanisms is explored over a wide range of concentrations under both uniform and non-uniform illumination for the example cell.

A case study of system power efficiency loss mechanisms in a multijunction, spectral splitting, concentrator solar cell system

In multijunction solar cell concentrator systems, it is important for system and cell designers to understand the relative magnitude of each loss mechanism. This investigation of potential system power efficiency improvements for a high-efficiency concentrator system focuses on quantifying the effects that various known loss mechanisms have on the overall system performance. Each of the loss mechanisms that were investigated play a part in the degradation of the devices performance and each of these losses can be reduced by appropriate engineering. This paper will address the extensive optimization of the system power efficiency of a multijunction concentrator solar cell system for the DARPA Very High Efficiency Solar Cell (VHESC) project. These results are useful to the system and cell designers and guide the efforts to reduce losses and to maximize the system power efficiency.

Combining solar cell and optical modeling in multijunction systems

High efficiency solar cell concentrator systems, particularly non-tracking systems, may utilize optical systems that involve lenses, secondary concentrators, dichroic mirrors and other optical components in conjunction with a variety of cells that may be optically and electrically connected in series or parallel configurations. This paper describes the value of combining realistic PV models into a complex optical model, making it possible to simulate the performance of the complete PV system to simplify the task of optimizing the system design. The model also allows for the assessment of many important optics-to-PV interface effects such as angle of incidence of the rays on cell performance. The embedded PV models used are based on readily measured cell performance, which can be characterized external to the complete PV system. Agreement between model predictions and measured performance is presented.

A solar cell model for use in optical modeling of concentrating multijunction photovoltaic systems

Using embedded solar cell models in optical modeling programs has proven to be very useful for system level design optimization. These models can be used to provide quick analysis of the performance of high efficiency multijunction concentrator photovoltaic systems. They can be designed to predict the performance of the experimentally measured cells or detailed numerical models. A previous modeling approach employed by the authors used a simple empirical diode model, which limited the range of short circuit currents to a small range for which the model was valid. To more accurately model high efficiency photovoltaic systems over a wider range of short circuit currents, a new empirical curve fit model has been developed that allows model parameters to vary as a function of the short circuit current.

A distributed emitter model for solar cells: Extracting an equivalent lumped series resistance

Front-surface grid electrodes are employed in solar concentrator cells to reduce the power losses resulting from lateral current flow in the cell emitter. Because these electrodes also shadow the cell from a portion of the incident light, an optimal grid electrode layout must be determined for a particular cell design. A useful model for this optimization in the early stages of design is to simulate the distributed losses in the cell emitter using an equivalent lumped series resistance. Previously reported 1D analytical calculations showed that the equivalent resistance of one square should simply be one-third of the emitter sheet resistance. However, this calculation ignores local biasing loss. In this work, a quasi-3D model, which includes this local biasing effect, is used to extract an equivalent series resistance of a section of the emitter between grid electrodes for various sheet resistances and optical concentration.

Design of 2- and 3-terminal GaInP/GaAs concentrator cells for maximum yearly energy output

Two important factors in achieving maximum power output in a multi-junction concentrator solar cell are selecting optimal sub-cell band gaps and absorber layer thicknesses. This optimization is spectrum-dependent and does not guarantee maximum yearly energy delivery, as the distribution of spectral energy varies throughout the day and year. A simple, detailed-balance model may be used in this optimization, though detailed numerical models are ideal as the specifics of the cell structure can be incorporated. In this paper, a particular GaInP/GaAs tandem structure is optimized using ADEPT to show that maximum yearly energy delivery is achieved by optimizing under a mid-November spectrum.

A method for estimating temperature dependent short circuit current

Solar cell models that can be embedded easily into optical modeling programs have proven to be useful to study the tradeoffs inherent between the optics and solar cells in concentrating photovoltaic systems. In addition to modeling the optics and photovoltaics at 25° C, system designers would like to study the range of temperatures the system will experience during terrestrial operation. Methods for estimating temperature-dependent open-circuit voltage and fill factor have been shown. This leaves the need to accurately estimate the short-circuit current as a function of temperature. In this paper measured EQE for a GaInP/GaInAs/Ge tandem stack has been used to calculate the short-circuit current over a range of operating temperatures. This method can easily be used to conduct system level analysis in any type of solar cell system from flat plate panels to concentrator photovoltaic systems that contain complicated optics.