Search

Consolidated tables showing an extensive listing of the highest independently confirmed efficiencies for solar cells and modules are presented. Guidelines for inclusion of results into these tables are outlined, and new entries since January 2022 are reviewed. An appendix describing temporary electrical contacting of large-area solar cells approaches and terminology is also included.

We correct an error in equation (60) of the original publication.

This work presents the results of the first interlaboratory comparison of linearity measurements of short-circuit current versus irradiance that includes a wide variety of photovoltaic (PV) device types, from reference cells to full-size modules. The aim of this inter-comparison was to compare the methods employed and to collect new inputs useful for the revision of the standard IEC 60904-10, which deals with linearity measurements for PV devices. The procedures and facilities employed by the partners include the differential spectral responsivity, the white light response, the solar simulator method and the two-lamp method. The facilities are generically described and compared and their main sources of uncertainty are discussed. Comparison results show good agreement within declared uncertainties between all partners. A few minor exceptions under low-light conditions raise questions of possible uncertainty underestimation for these specific conditions. The overall outcome of the comparison also highlights the importance of considering correlations in the uncertainty budget, which can potentially improve the overall stated uncertainty. A critical review is made of the data analysis adopted in the standard IEC 60904-10 to calculate the linearity degree of the short-circuit current towards irradiance. The analysis review suggests a way to make results based on different methods more comparable and less prone to erroneous linearity assessment.

Within the EURAMET ENG55 “PhotoClass” project, several characteristics of photovoltaic (PV) devices beyond their performance at Standard Test Conditions were investigated, including measurements at varying irradiance and temperature. Four groups of PV devices of different size and technology were prepared and corresponding round-robins were run between partner laboratories with substantially different facilities and methods – namely based on spectral or integral measurements. This paper presents the outcome of the four inter-laboratory comparisons dealing with temperature coefficient measurements of the short-circuit current of PV devices, from reference-cell size to full-size commercial modules of mainly several c-Si technologies, but also with some examples of CIGS and GaAs devices. The measurement results are compared via En number assessment, hence including measurement uncertainties. The main outcome of this measurement exercise is a very good agreement of all the laboratories although completely different approaches were applied. In some cases, laboratory measurement uncertainties are even considered rather conservative and could therefore be revised. Furthermore, a comparison between bare cells and commercial modules of the same technology is made, which may represent useful information for PV manufacturers.

The determination of the spectral responsivity is an essential part of solar cell characterization. Since solar simulators only approximate the reference spectrum, a spectral mismatch correction has to be performed. This correction procedure requires spectral responsivity data. Apart from the complete differential spectral responsivity procedure, the IEC 60904-8 standard defines four simplifications. In this paper, we provide information on the variations in the spectral responsivity curves for these simplifications. We show that for nonlinear front junction cells, deviations predominantly occur at wavelengths above 700 nm and become largest around 1000 nm. While we found a maximum deviation of 30% for the simplification with lowest requirements in bias irradiance, all other simplifications yield deviations below 10%. For a nonlinear cell measured relative to a world photovoltaic scale reference cell using a class A solar simulator, this transfers to a deviation below 0.01% in the spectral mismatch factor. If one depends on the use of a simplification, we recommend using the multicolor approach. Even though the singlecolor approach might yield lower deviations, this approach requires knowledge about the maximum in the spectral responsivity, which is not generally known in advance of the measurement. Accepting a slightly higher deviation, the white bias approach is a recommendable alternative.

We analyze the influence of a variety of different contacting geometries on the fill factor (FF) of solar cell I–V measurements. For this analysis, we compare a wide variety of modeled and measured FFs of Si solar cells. We consistently find large FF differences between individual contacting geometries. These differences amount to up to 3 ${\%_{{\rm abs}}}$ for high busbar resistivities of up to 40 $\Omega/\hbox{m}$ . We analyze the contacting geometries for their sensitivity on uncontrolled variations of the contacting resistances. In this analysis, we find that using triplet rather than tandem configurations and using a larger number of test probes reduces the impact of varying contacting resistances to below 0.02 ${\%_{{\rm abs}}}$ . We propose a contacting geometry that we consider to be suitable for calibrated I–V measurements. This contacting scheme is a configuration with a total of five triplets consisting of two current probes and one sense probe. The sense probe is positioned to measure the average busbar potential between the current probes. This is the optimal contacting geometry in terms of a low sensitivity to the busbar resistivity and variations of contacting resistances. In addition, this geometry does not impose unnecessarily large mechanical stress to the cell under measurement.

Measuring the spectral irradiance of solar radiation is required in many fields of science and technology. In this work, we present an in-depth discussion of the measuring procedure and required corrections for such measurements. We also describe our measurement uncertainty analysis, which is based on a Monte-Carlo procedure in accordance with the Guide to the expression of uncertainty in measurement (JCGM, Paris, 2008). For this purpose, fifteen uncertainty sources are identified, analyzed and described analytically. As a specific application example, we describe the instrumentation and procedure for determining the spectral irradiance of a solar simulator at the ISO/IEC 17 025 accredited solar cell calibration laboratory ISFH CalTeC and the corresponding measurement uncertainty analysis. Moreover, we provide a Python implementation for this calculation along with the paper. We show that for state-of-the-art instrumentation, significant uncertainty contributions arise from the reference lamp (primary calibration standard), stray light and signal-to-noise ratio. If sharp spectral features are present (which is common, e.g. for Xenon lamps), spectral bandwidth and wavelength uncertainty also contribute significantly to the overall uncertainty.

We process a photovoltaic (PV) module with 120 half passivated emitter and rear cells that exhibits an independently confirmed power of 303.2 W and a module efficiency of 20.2% (aperture area). The cells are optimized for operation within the module. We enhance light harvesting from the inactive spacing between the cells and the cell interconnect ribbons. Additionally, we reduce the inactive area to below 3% of the aperture module area. The impact of these measures is analyzed by ray-tracing simulations of the module. Using a numerical model, we analyze and predict the module performance based on the individual cell measurements and the optical simulations. We determine the power loss due to series interconnection of the solar cells to be 1.5%. This is compensated by a gain in current of 1.8% caused by the change of the optical environment of the cells in the module. We achieve a good agreement between simulations and experiments, both showing no cell-to-module power loss.

Most methods for interpreting electroluminescence (EL) or photoluminescence (PL) images of solar cells evaluate the local diode voltages but not the local luminescence intensity itself. One exception is the Fuyuki approximation, which assumes that the local value of the luminescence signal is proportional to the local effective diffusion length. This dependence has been derived for infinitely thick solar cells and neglects self-absorption of the luminescence photons. However, for real solar cells and imaging conditions, with increasing diffusion length, the luminescence signal approaches a limiting value; hence, the Fuyuki approximation no longer holds. In this paper, we compare EL and PL images of multicrystalline solar cells using different kinds of light filtering and find that gentle shortpass filtering is useful for avoiding optical artifacts. Based on earlier calculations, a physically founded formula for the dependence of the gently shortpass-filtered luminescence signal on the bulk diffusion length, for a given rear surface recombination velocity, is presented. Since this formula only barely allows us to calculate the diffusion length from the luminescence signal, a simplified approximate formula is proposed, and its accuracy is checked. This method is tested on EL and ${V_{{{\rm oc}}}}$ PL images of solar cells. We find that for a typical industrial multicrystalline Al-backside solar cell, the obtained effective diffusion length images correlate well with such images obtained by spectral LBIC image evaluation. In addition, the saturation current density images correlate well with such images obtained by dark lock-in thermography, which show a much lower spatial resolution. The main limitation of the proposed method is that it is basically approximate and needs some fitting parameters.

We measure and discuss the complex refractive index of conventional ethylene vinyl acetate (EVA) and an EVA with enhanced UV-transmission based on spectroscopic ellipsometry, transmission and reflection measurements over the wavelength range from 300-1200 nm. Ray tracing of entire solar cell modules using this optical data predicts a 1.3% increase in short circuit current density (Jsc) at standard test conditions for EVA with enhanced UV transmission. This is in good agreement with laboratory experiments of test modules that result in a 1.4% increase in Jsc by using a UV transparent instead of a conventional EVA. Further, ray tracing simulations with realistic irradiation conditions with respect to angular and spectral distribution reveal an even larger Jsc increase of 1.9% in the yearly average. This increase is largest in the summer months with an increase of up to 2.3%. German Federal Ministry for Economic Affairs and Energy/0325641

The determination of the spectral responsivity is an essential part of solar cell calibration. Apart from the complete differential spectral responsivity procedure, which yields the most accurate results, the IEC 60904-8 defines four simplifications. We provide detailed information on the expected variations in the calculated spectral responsivity curves for the different simplifications compared to the complete procedure. For non-linear crystalline silicon front junction solar cells, we show that deviations mainly occur at wavelengths above 700 nm and become largest between 1000 to 1200 nm. Even though we found a maximum deviation in spectral responsivity of 7% for the simplification with lowest requirements in bias irradiance, all other simplifications yield deviations well below 3%. This transforms into a deviation of 0.01% in the spectral mismatch factor for an industrial PERC solar cell when using a typical world photovoltaic scale (WPVS) reference solar cell and a class A two-lamp solar simulator. If you are reliant on the use of a simplification, we recommend using the multicolour approach. Even though the singlecolour approach might yield lower deviations in specific cases, it requires knowledge about the maximum in the spectral responsivity, which is not generally known in advance of the measurement. Accepting a slightly higher deviation, the white bias approach is a recommendable alternative.

Recent reports about new cell efficiency records are highlighting the continuing development of passivated emitter and rear cells (PERC). Additionally, volume production has started, forming the basis for cutting edge solar modules. However, transferring the high efficiency of the cells into a module requires an adaptation of the conventional front metallization and of the cell interconnection design. This paper studies and compares the module output of various cell interconnection technologies, including conventional cell interconnection ribbons and wires. We fabricate solar cells and characterize their electrical and optical properties. From the cells, we build experimental modules with various cell interconnection technologies. We determine the optical and electrical characteristics of the experimental modules. Based on our experimental results, we develop an analytical model that reproduces the power output of the experimental modules within the measurement uncertainty. The analytical model is then applied to simulate various cell interconnection technologies employing halved cells, optical enhanced cell interconnectors, and multiwires. We also consider the effect of enhancing the cell-to-cell spacing. Based on the experimentally verified simulations, we propose an optimized cell interconnection for a 60-PERC module that achieves a power output of 323 W. Our simulations reveal that wires combined with halved cells show the best module performance. However, applying light-harvesting structures to the cell interconnection ribbons is an attractive alternative for upgrading existing production lines.

In a previous contribution it was found by evaluating 2-dimensional device simulation-based photoluminescence (PL) images of solar cells that the previous methods of PL image evaluation give good results for imaging the local series resistance, but lead to an erroneous distribution of an inhomogeneous saturation current density J01. This is caused by the model of independent diodes used in these evaluation methods, which does not take into account the distributed character of the series resistance and horizontal balancing currents occurring in inhomogeneous solar cells. In this contribution it is checked whether two alternative PL evaluation methods, which were already published but did not become popular yet, may overcome this problem. These are the differential luminescence imaging technique proposed by Rau et al. and the Laplacian-based method proposed by Glatthaar et al. By simulating PL images of a model device based on a 2-dimensional equivalent circuit with a known distribution of J01 and applying these PL evaluation methods, retrieved images of J01 are obtained and compared with the input J01 distribution. It is found that the differential luminescence imaging technique, at least in its presently used form, leads to wrong results, as the traditional PL evaluation techniques do. The Laplacian-based method, on the other hand, indeed has the potential to image J01 correctly. However, until now this method lacks in sensitivity. In this contribution some improvements are proposed, which may enhance the sensitivity of the Laplacian-based method. One important improvement is to enable a scaling measurement at a high instead of a low current resp. illumination intensity. Thereby the signal-to-noise ratio of the scaling measurement, which until now limited the accuracy of the whole PL evaluation, may be improved. Another proposal is that the actual PL measurement is performed under open circuit condition and not under current extraction. First experimental results show the effectiveness of this new method, but limitations still remain.

The conventional quantitative evaluation of dark lock-in thermography (DLIT), electroluminescence (EL), and photoluminescence (PL) images of solar cells is based on the model of independent diodes, where each image pixel is assumed to be connected with the terminals by an independent series resistance. In reality, however, the solar cell represents a 2-dimensional resistance-diode network. In this work solar cells containing well-defined spatial distributions of the saturation current density J01 and also containing J02-type and ohmic shunts are modeled for various externally applied biases and illumination conditions realistically as a 2-dimensional resistance-diode network. The resulting local diode voltage distributions are converted into DLIT, EL and PL images, which are further processed by conventional evaluation methods, which rely on the simple model of independent diodes. These are the so-called “Local-IV” method for the DLIT analysis, which may be supported by EL results to obtain series resistance images, and “C-DCR” for the PL analysis. This leads to calculated images of the local effective series resistance Rs and of J01. Regarding the resulting Rs images, PL shows the expected series resistance distribution and is not affected by the shunt regions. The DLIT–EL Rs images instead yield expected values only in the homogeneous regions, which are not affected by the assumed shunts. DLIT–EL determines higher values of Rs in local shunt regions and lower values around these regions and in spatially extended shunt regions. Regarding the J01 images both methods again give the expected results if J01 is distributed homogeneously. However, in the shunted regions, PL suffers from balancing currents within the emitter and DLIT from optical blurring. By comparing local and extended regions of increased J01 we find that DLIT approximates the expected J01 value better than PL, which clearly underestimates even extended local maxima of J01. For a local current analysis of silicon solar cells we recommend the use of DLIT for the determination of J01 images and PL for the determination of Rs images.

An intercomparison of terrestrial photovoltaic (PV) calibrations was performed among a number of European calibration and testing laboratories that participated in the European Metrology Research Program (EMRP) project “PhotoClass”. The purpose of this intercomparison was to evaluate the comparability of calibration and testing services within the stated uncertainties of the individual laboratories. The calibration objects were two world photovoltaic scale (WPVS)-type reference solar cells, one made from crystalline silicon and one made from GaAs. The calibration value (CV) was the short-circuit current under standard test conditions (ISTC). In conclusion, it was found that the CVs are all consistent within the stated uncertainties. This result strengthens the reliance in the calibration chain and in the PV calibration infrastructure in Europe.

In photovoltaic (PV) modules, the interconnection of solar cells is critical in terms of mechanical stability and resistive power losses. In this study, we analyze the interconnection of large-area 15.6 × 15.6 cm 2 industrial p-type passivated emitter and rear cell (PERC) solar cells in terms of resistive losses. For our analysis, we prepare a 3 × 3 minimodule from PERC solar cells with soldering pads and efficiencies up to 20.0%. We measure a significant cell-to-module (CTM) power loss of 8% at this module. For comparison, we prepare a 3 × 6 module consisting of halved 7.8 × 15.6 cm 2 PERC solar cells. Using a nanosecond laser to cut the finished solar cell in two pieces, no additional power loss is introduced by cutting. The CTM factor of 1.0 determined at the 3 × 6 module is explained using an analytical model describing the series resistance of the module interconnection. Using this model, we estimate for our current PERC cell generation and module process an output power of 275 W for 60 full-size cells and 285 W for 120 halved cells.

CCD spectrometers have become the industry-standard tools for classifying solar simulator light spectrums, but they have drawbacks. A simpler method is to measure the amount of light present in each relevant band defined by the classification standards directly, by filtering the incident light appropriately and measuring its magnitude with a basic detector. This is much less expensive, and also has some other surprising benefits. The difficult aspect of this method is quantifying and correcting for the uncertainties introduced in the process. These will be outlined in detail in order to build confidence in the accuracy of this approach.

The spectral responsivity alt;iagt;salt;/iagt;alt;subagt;?,STCalt;/subagt;(?) of solar cells is widely used for cell analysis or calibration purposes. According to the IEC60904-8:2014 standard, the reference method for the determination of alt;iagt;salt;/iagt;alt;subagt;?,STCalt;/subagt;(?) is the complete differential spectral responsivity approach. For this approach, the differential spectral responsivity alt;iagt;salt;/iagt;~(?) is measured as a function of wavelength and bias irradiance. To obtain the spectral responsivity alt;iagt;salt;/iagt;alt;subagt;?,STCalt;/subagt;(?) related to standard test conditions the IEC60904-8:2014 standard recommends to integrate 1/alt;iagt;salt;/iagt;~(?) via bias current alt;iagt;Ialt;/iagt;alt;subagt;balt;/subagt; for each wavelength. We show that this integration is wrong. It lacks analytical derivation and provides faulty curves for non-linear solar cells. We proof analytically and by means of simulations that the correct way of calculation is either the integration of alt;iagt;salt;/iagt;~(?) via the bias irradiance alt;iagt;Ealt;/iagt;alt;subagt;balt;/subagt; or the integration of alt;iagt;salt;/iagt;~(?)/alt;iagt;salt;/iagt;~alt;subagt;AMxalt;/subagt; via the bias current alt;iagt;Ialt;/iagt;alt;subagt;balt;/subagt;, with alt;iagt;salt;/iagt;~alt;subagt;AMxalt;/subagt; being the AMx-weighted (e.g. AM1.5G or AM1.5D) differential responsivity. A simulation of the differential spectral responsivity of a strongly non-linear solar cell demonstrates deviations of alt;iagt;salt;/iagt;alt;subagt;?,STCalt;/subagt;(?) up to 30 % for (the wrong) integration of 1/alt;iagt;salt;/iagt;~(?) via alt;iagt;Ialt;/iagt;alt;subagt;balt;/subagt; at some wavelengths, corresponding to a deviation in the short-circuit current of up to 3.0 %.

The fill factor determined from a measured current-voltage characteristic of a bare solar cell depends on the number and positions of the electrical contacting probes. Nine different geometries for contacting the front side busbars are used to measure the current-voltage (I-V) characteristics of a 5 busbar industrial-type passivated emitter and rear totally diffused (PERT) solar cell under standard testing conditions. The fill factors of the measured I-V characteristics vary from 78.5 % abs to 80.6 % abs . We further measure the contacting resistance of 3 different contacting probes to estimate the sensitivity of measurements with different contacting geometries on random resistance variations. The contacting resistance is 60 mΩ for nine-point probes and 80 mΩ for four- and single-point probes. We determine the magnitude of contacting resistance variations from measurements at different probe positions to be ±30 mΩ. Using this variation, we perform numerical simulations and find a larger sensitivity on random resistance variations for tandem- (pairs of current- and sense probes) compared to triplet (one sense- between two current probes) configurations. The corresponding fill factor deviation is approximately 0.1% abs for tandem configurations when the contacting resistances of up to two current probes are altered. The sensitivity for triplet configurations is negligible.

We investigate the validity of calibrated photoluminescence lifetime measurements of crystalline silicon wafers for arbitrary lifetime and injection ranges. Absolute lifetime images are obtained from steady-state photoluminescence measurements by relating the photoluminescence signal to the excess carrier density. Since the luminescence signal is expected to be related to the integral of the depth distribution of the excess carrier density, an adequate calibration of the luminescence signal requires a secondary method which yields the integral of the depth distribution of the excess carrier density in absolute units. In this paper, we investigate the applicability of steady-state photoconductance measurements for the calibration of the photoluminescence signal. We derive a generalized relation linking the photoluminescence signal with the excess carrier density, considering the impact of an inhomogeneous carrier concentration profile. We experimentally verify the impact of the carrier distribution on the photoluminescence calibration by investigating two silicon wafers with different electronic bulk properties. Finally, we propose an iterative correction procedure reducing the deviations due to an inhomogeneous carrier density profile of calibrated photoluminescence-based lifetime measurements significantly.

The collected photocurrent imaging technique measures the total amount of photocurrent that can be collected from every region of a solar cell. The technique is based on the recently introduced differential photocurrent collection efficiency imaging method, which uses luminescence images taken under varying bias conditions and a constant illumination. The differential photocurrent collection efficiency provides information about how the cell or module current reacts to local changes of the generated photocurrent. We explain in detail how certain assumptions and measurements of the photocurrent collection efficiency under different illumination conditions can extend the differential information we obtain from the differential photocurrent collection efficiency to get information about the total amount of photocurrent collected from each specific region of a solar cell. The method was demonstrated on a thin film Cu(In, Ga)Se 2 cell, showing quantitatively how series resistance reduce the collection of photocurrent.

Increasing the area of interdigitated back-contact (IBC) solar cells featuring a busbar contact geometry requires the use of longer fingers. The finger resistance will, thus, be increased if the thickness of the metallization is kept constant. In order to maintain a thin metallization, it is beneficial to increase the number of busbars per contact. However, using more than one busbar for each polarity implies an asymmetric contact geometry. As a consequence, under operation, the busbars of the same polarity carry different currents. Due to voltage drops over unavoidable electrical resistances, this may lead to significant potential differences between these busbars. Since current–voltage characteristics are usually measured using separate sense contacts for the voltage measurement, the position and number of these contacts may considerably affect the shape of the resulting current–voltage characteristic and, thus, the fill factor. By means of simulations with the circuit simulator LTSpice, we show that a permanent contacting with soldered tabs allows for a correct determination of the fill factor. A chuck used for temporary contacting should feature at least one sense contact per busbar and pin contacting resistances below 30 mΩ in order to keep the fill factor error below 0.5% absolute.

Capacitance-Voltage (C- V) measurements are an established method for determining the bulk dopant concentration Nbulk of semiconductor devices. The extraction of Nbulk requires knowledge of the surface area of the space-charge region (SCR). For textured solar cells, this surface area is enlarged, depending on texture and bias voltage. The amount of enlargement is a priori unknown, and there is no simple way of measuring directly it. In this paper, models that were previously proposed in the literature are employed to describe the SCR area. Substantial deficiencies in these models are revealed by comparison with a 3-D numerical simulation of the SCR geometry used as a benchmark. This approach leads to the development of a new analytical model that can properly describe the bias-voltage dependence of the surface area enlargement. C-V measurements over a large bias-voltage range are performed on various solar cells with different textures and resistivities. The new analytical model is employed to correct the measurement data for the influence of surface area enlargement. Thus, Nbulk can be extracted without explicit knowledge of the enlargement. Comparison with four-point probe measurements shows excellent agreement, demonstrating the power of the new method to enable the C-V technique for use in Nbulk determination for textured solar cells.

The capacitance-voltage (CV) measurement is a precise and fast method to determine base-dopant concentrations of crystalline silicon solar cells. Since available measurement equipment is usually limited in its current output, the application of CV analysis has been limited to small-area solar cells in research laboratories. We present an experimental setup that is capable of measuring CV curves with a current output of up to 2 A. Using this setup, we demonstrate the applicability of CV measurements to large-area industrial solar cells for base-dopant concentrations ranging between 6.0 × 1014 cm-3 and 4.2 t1016 cm-3. An area enhancement factor f quantifying the relation between the macroscopic cell area and the active junction area is determined for alkaline textured mono- and isotextured multicrystalline silicon solar cells. Comparing the base dopant of the CV analysis with four-point probe measurements, we achieve an agreement with an uncertainty of 10%. For alkaline textured monocrystalline silicon solar cells, we demonstrate that the area enhancement factor can be extracted from the ratio of the pyramid base length and emitter thickness.

We introduce an approach to determine the operating voltage of individual solar cells in photovoltaic (PV) modules by electroluminescence (EL) imaging. The highest EL signal of each solar cell is proportional to its operating voltage. Moreover the sum of all operating voltages equals the externally applied module voltage. Thus the operating voltage of individual solar cells is determined from the measured EL signal. The reliability of this relation is verified by measurements on specially prepared PV modules allowing us to measure the individual operating cell voltage. The experimentally measured cell voltages are deduced with an uncertainty of ±1% from an EL image. Moreover, the operating cell voltages determined from the EL image are used to calculate the module series resistance. Comparing experimentally determined values from the operating cell voltage and the total current flowing supplied to the module with calculated module series resistances using tabulated material and typical solar cell parameters, a very good correspondence is found. Copyright © 2010 John Wiley & Sons, Ltd.

For large area devices a spatial analysis of local device and material parameters is essentially important. Imaging techniques allowing a fast and contactless analysis with a high spatial resolution have become a versatile characterization tool during the last decade. We present a comprehensive overview over the existing imaging techniques for the analysis of silicon wafers and solar cells utilizing different spectral ranges of photon emission. Additionally, we report on recent studies of local junction breakdown and the emission of light from solar cells under forward and reverse bias using luminescence imaging and dark lock-in thermography. Finally we present a calibration-free dynamic infrared carrier lifetime mapping (dynamic-ILM) technique, yielding images of the carrier lifetime of multi-crystalline silicon wafers within seconds.

Improving the light trapping in a module results in an increase in the generated current. Consequently, an optimization of the front grid metallization of the cell is required for the best trade-off between series resistance, shading, and recombination losses. For this purpose, we combine ray tracing and electrical solar cell and module calculations that explicitly account for cell and module interactions. Our model bases on experimentally verified input parameters: We determine the electrical and optical properties of the front metal fingers of passivated emitter and rear cells (PERC). We show that the effective optical width of the front metal fingers in the module is significantly reduced by 54%. The optimized simulated module has 120 half-size PERC with 20.2% cell efficiency and has an output power of 295.2 W. This is achieved with an increased number of 120 front metal fingers per cell, four white-colored cell interconnection ribbons (CIR), and an increased cell spacing. Applying these optimized design changes to an experimental module we measure a module power output of 294.8 W and a cell-to-module (CTM) factor of 1.02. Measured and simulated power agree and the deviations in Voc, Isc and FF are less than 0.91%rel. We perform a module power gain analysis for the fabricated module and simulate a potential maximum module power of 374.1 W when including further improvements. German Federal Ministry for Economic Affairs and Energy/0325641

A method is proposed that enables the imaging of the photocurrent collected by a solar cell under arbitrary operating conditions. The method uses a series of luminescence images under varying illumination to derive the total photocurrent collection efficiency at a given voltage bias. The resulting total photocurrent collection image directly relates to the difference between the dark and illuminated current-voltage characteristics of the cell. A crystalline silicon solar cell is used to test the method, and the images of the total photocurrent collection efficiency are used to quantify the influence of a crack on the total collected photocurrent of the solar cell.

We have performed 2-dimensional inhomogeneous solar cell simulations using the circuit-simulation program Spice and obtained photoluminescence (PL), electroluminescence (EL), and dark lock-in thermography (DLIT) images for various illumination and biasing conditions. These images were evaluated by established methods for retrieving images of the saturation current density J01. We found that DLIT delivers blurred but quantitatively correct J01 images, but PL generally delivers too weak local maxima of J01. This is due to the used model of independent diodes, which is too simple to describe the local diode voltage in inhomogeneous cells precisely. In reality lateral balancing currents exist, which are not regarded in established PL evaluation methods. For solving this problem we have checked two alternative PL evaluation methods, which are already published but not established yet. Especially the Laplacian-based method has the potential to overcome this problem. For our simulated images the method yields correct results. However, in spite of some improvements for the determination of the calibration image, our first experimental results obtained by using this method do not agree yet with lock-in thermography based J01 results.

We present a fast and calibration-free carrier lifetime imaging technique based on photoluminescence (PL) measurements using an InGaAs camera for the examination of crystalline silicon wafers. The carrier lifetime is determined from the time dependent luminescence emission after optical excitation. A ratio, including four PL images acquired at different times during the modulated excitation, is calculated and found to depend only on the camera integration time and the effective carrier lifetime. Therefore, the carrier lifetime is unambiguously determined by this ratio without knowing any additional wafer parameter. We demonstrate the applicability of the dynamic PL technique to multicrystalline silicon wafers. (© 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

We use photoluminescence (PL) measurements by a silicon charge-coupled device camera to generate high-resolution lifetime images of multicrystalline silicon wafers. Absolute values of the excess carrier density are determined by calibrating the PL image by means of contactless photoconductance measurements. The photoconductance setup is integrated in the camera-based PL setup and therefore identical measurement conditions are realised. We demonstrate the validity of this method by comparison with microwave-detected photoconductance decay measurements. (© 2008 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim)

We present an analytical expression for modelling electroluminescence (EL) as well as photoluminescence (PL) data of samples with any combination of planar and arbitrary rough surfaces. The model also accounts for free carrier absorption in highly doped regions of the sample. It is experimentally confirmed by comparison to measured EL and PL spectra of solar cells and wafers featuring different surface geometries. Significant differences in the absorption coefficient data available from literature, which enters the model as tabulated data, are revealed. Using our model, the absorption data by Daub yields the best agreement to the measured luminescence spectra. Moreover, we present a method to determine the rear surface reflectance of samples from the peak wavelength of luminescence spectra.

Luminescence imaging has become a standard tool for solar cell analysis within the last decade. In order to understand the potential, as well as the limitations of the numerous luminescence characterization approaches, this work provides both the physical background for modeling the luminescence emission from silicon solar cells as well as a review of series resistance imaging, one of the most prominent applications of luminescence imaging. The first part addresses the measurement setup. A specific focus lies on the suitability of different cameras and the optical filters necessary to prevent the detection of reflected excitation light as well as stray light. Detection conditions with respect to capture time and averaging of images are also discussed. In the second part, we derive a general mathematical description of the spectral luminescence emission. We show that the integral over the product of the depth-dependent minority charge carrier profile and the luminescence photon detection profile fully determines the spectral luminescence emission. Moreover, we show that the luminescence photon detection profile can be obtained from the generation profile of minority charge carriers under illumination, for which well-tested expressions can be found in the literature. Based on the mathematical description, we derive a short- and a long-wavelength approximation corresponding to the spectral sensitivity of silicon and indium–gallium–arsenide detectors, widely used for luminescence measurements. While from the short-wavelength approach the local voltage of a solar cell can be determined, the long-wavelength approach yields the local collection length of the device under test. The final part describes the most prominent application of luminescence imaging: the determination of the local series resistance of wafer-based crystalline silicon solar cells. We review a variety of different approaches introduced in the past. We show that all approaches are based on the same general equation, which is a consequence of the underlying independent diode model. Based on numerical circuit simulations, we study the range of applicability of this simple model. Moreover, the most promising series resistance approaches are applied to various silicon solar cells. Resulting local series resistance and local recombination current images are compared among different methods, as well as to global values extracted from the current–voltage characteristics of the solar cell.

Since 15 years lock-in thermography (LIT) is used to investigate solar cells and modules. While LIT was used at the beginning only for shunt detection, meanwhile special LIT-techniques have been developed to image the minority carrier lifetime, also on wafers, the monochromatic cell efficiency, the local series resistance, the ideality factor, or the physical properties of breakdown sites. A general limitation of LIT is its limited spatial resolution due to thermal blurring. LIT investigation of breakdown sites takes only a fracture of a second, but investigation under forward bias usually takes some minutes. In recent years photoluminescence (PL) and electroluminescence (EL) imaging have become very popular to characterize solar wafers, cells and modules. Just as LIT, luminescence imaging allows to image the local lifetime and local series resistances (with PL even quantitatively), stronger local shunts, breakdown sites, and other faults like cracks. A big advantage of luminescence imaging is that it does not suffer from thermal blurring. Moreover, imaging times in the range of seconds are possible, and the systems are cheaper than LIT systems. The question is: Do we need LIT imaging anymore at all? It will be demonstrated that some important kinds of characterization can be done only by LIT, such as the quantitative measurement of breakdown and leakage currents (especially space charge recombination currents, which are not detectable by EL/PL), ideality factor imaging, the detection of leakage currents below grid lines and bus bars, as well as the quantitative measurement of certain breakdown parameters. Thus, LIT and luminescence imaging are complementing each other. For a complete characterization of solar cells both techniques have to be used in parallel.

We report in detail on the luminescence imaging setup developed within the last years in our laboratory. In this setup, the luminescence emission of silicon solar cells or silicon wafers is analyzed quantitatively. Charge carriers are excited electrically (electroluminescence) using a power supply for carrier injection or optically (photoluminescence) using a laser as illumination source. The luminescence emission arising from the radiative recombination of the stimulated charge carriers is measured spatially resolved using a camera. We give details of the various components including cameras, optical filters for electro- and photo-luminescence, the semiconductor laser and the four-quadrant power supply. We compare a silicon charged-coupled device (CCD) camera with a back-illuminated silicon CCD camera comprising an electron multiplier gain and a complementary metal oxide semiconductor indium gallium arsenide camera. For the detection of the luminescence emission of silicon we analyze the dominant noise sources along with the signal-to-noise ratio of all three cameras at different operation conditions. © 2011 American Institute of Physics.

The purpose of this paper is a detailed comparison of selected luminescence and lock-in thermography (LIT) results on one exemplary sample and the drawing of corresponding conclusions. Our focus is on solar cells, but some investigations on wafers will be discussed as well. The comparison will help to decide which characterization tools are needed to solve technological problems. It will be demonstrated that luminescence imaging may widely replace LIT with respect to the analysis of recombination-active bulk defects, cracks, series resistance, and junction breakdown sites. However, some important investigations can be done only by LIT. LIT allows for a quantitative analysis of different kinds of leakage currents both under forward and under reverse bias, enabling a reliable analysis of local I-V characteristics. It is shown that LIT and luminescence imaging are complementary to each other and should be used in combination.

The electroluminescence emission of crystalline silicon solar cells at near-bandgap wavelengths is investigated. We show that the intensity of the emitted luminescence at near-bandgap wavelengths is directly proportional to the collection diffusion length Lc which is a measure of bulk and rear surface recombination properties and determines the short circuit current of a solar cell illuminated with light of near-bandgap wavelengths. We provide experimental evidence for the determination of Lc by carrying out electroluminescence measurements on a set of 15 specially prepared monocrystalline silicon solar cells with different thicknesses. Moreover, we demonstrate and discuss the applicability of the proposed method to obtain images of the collection diffusion length Lc of multicrystalline silicon solar cells. The values determined by electroluminescence imaging coincide with values obtained from spectrally resolved quantum efficiency measurements with a relative accuracy of 13 %. © 2011 Published by Elsevier Ltd. Selection and/or peer-review under responsibility of SiliconPV 2011.

We study the emission of light from industrial multicrystalline silicon solar cells under forward and reverse biases. Camera-based luminescence imaging techniques and dark lock-in thermography are used to gain information about the spatial distribution and the energy dissipation at pre-breakdown sites frequently found in multicrystalline silicon solar cells. The pre-breakdown occurs at specific sites and is associated with an increase in temperature and the emission of visible light under reverse bias. Moreover, additional light emission is found in some regions in the subband-gap range between 1400 and 1700 nm under forward bias. Investigations of multicrystalline silicon solar cells with different interstitial oxygen concentrations and with an electron microscopic analysis suggest that the local light emission in these areas is directly related to clusters of oxygen. © 2009 American Institute of Physics.

We present a method to determine the effective diffusion length Leff of silicon solar cells from photoluminescence (PL) measurements carried out under two different operating conditions. Measuring the photoluminescence emission under open circuit condition (PL-oc), where the solar cell is not contacted at all, and short circuit condition (PL-sc), where the solar cell is held at zero voltage, Leff directly follows from the ratio of the PL-oc and the PL-sc signals. Detailed knowledge about the optical properties of the experimental setup is not necessary since the optical properties cancel out completely. We explain the theoretical background of our method and derive an analytical description for the PL-oc and the PL-sc luminescence emissions. The applicability of our method is demonstrated by the comparison of effective diffusion lengths from PL measurements with values determined from the analysis of internal quantum efficiency measurements.

This letter introduces a method based on electroluminescence imaging to determine mappings of the local series resistance of large area semiconductor devices such as solar cells. The method combines the local electroluminescence emission Φi(U) and its derivative Φi′(U) with respect to the applied voltage U. The combined analysis of these two quantities yields the local series resistance Rise and proves the physical validity of the used current transport model and thus the physical relevance of the determined Rise value. The method is verified on a monocrystalline silicon solar cell with local shunts and local series resistance problems.

We perform recombination current and series resistance imaging on large-area crystalline silicon solar cells using a combined analysis of camera-based dark lock-in thermography (DLIT) and electroluminescence (EL) imaging. The solar cells are imaged both by DLIT and EL under identical operating conditions. The quantitative analysis of the DLIT measurement produces an image of the local heating power and the EL picture results in an image of the local cell voltage. Combining the two images pixel by pixel allows us to calculate images of the local recombination current and the local series resistance of the solar cell.

Screen-printed passivated emitter and rear cells (PERC) require Ag pads on the rear side to enable solderable connections for module integration. These Ag pads are separated from the silicon by a dielectric layer to avoid recombination of minority charge carriers. The drawback of this configuration is an elongated transport path for the majority charge carriers generated above the pads. This results in an increase in series resistance. The strength of this effect depends on charge carrier generation above the Ag pads that critically depends on shading of the cell's front side. Ag pads are usually wider than the busbars or the interconnector ribbons and thus are only partially shaded. We build PERC test structures with various rear side configurations of Ag and Al screen printing as well as with and without laser contact openings (LCO). Using experiments and finite element simulations we investigate the impact of shading the Ag pads by the busbars and other means. While fully shaded regions do not increase the lumped solar cell's series resistance, unshaded Ag pads lead to an increase of about 37%. German Federal Ministry for Economic Affairs and Energy/0325641