Search

We present a simple yet powerful analysis of Suns-photoluminescence quantum yield measurements that can be used to determine the surface saturation current densities of thin film semiconductors. We apply the method to state-of-the-art polycrystalline perovskite thin films of varying absorber thickness. We show that the non-radiative bimolecular recombination in these samples originates from the surfaces. To the best of our knowledge, this is the first study to demonstrate and quantify non-linear (bimolecular) surface recombination in perovskite thin films.

This study assesses and improves the accuracy of commonly used expressions for the fill factor (FF). Parameters that could affect the accuracy of the revised expressions are investigated. Empirical coefficients of the commonly used analytical expressions are first recalculated using a modified fitting approach. Although the predictions of the revised expressions perfectly match the results of theoretical one-diode model simulations, gaps are observed when compared with actual measurements. The different impacts of unaccounted factors in the expressions are then explored. It is shown that adjusting the ideality factor or considering edge recombination improves the accuracy of the predictions. Moreover, the expressions can slightly overestimate the FF of cells with non-uniform implied open-circuit voltage distribution. As methods to extract electrical parameters from luminescence images continuously improve, the findings of this study can aid in developing techniques for extracting FF from luminescence images of industrial solar cells.

Luminescence imaging is a fast and non-destructive method to spatially resolve non-uniform electrical properties of solar cells. The spatial resolution of these images determines the smallest identifiable features. The higher the spatial resolution of an image, the better the capability to detect small defects. However, high-resolution cameras with high near-infrared light sensitivity that are suited to luminescence imaging of crystalline silicon are often expensive. In this study, we present a method, based on deep learning, that enhances the spatial resolution of luminescence images computationally with minimal cost. We also demonstrate the ability to overcome noise which is inevitable in any imaging system. This approach provides a simple and promising path toward reducing the cost of luminescence imaging systems and enhancing the capability of existing systems.

Identifying solar cell efficiency shortfalls in production lines is crucial to troubleshoot and optimize manufacturing processes. With the adoption of luminescence imaging as a key end-of-line characterization tool, a wealth of information is available to evaluate cell performance and classify defects, suitable for user input-free deep-learning analysis. We propose an automated reconstruction method, based on state-of-the-art generative adversarial networks, to remove defective regions in luminescence images. The reconstructed and original images are compared to estimate the efficiency loss. The method is validated on intentionally damaged cells by reconstructing defect-free images and then predicting the efficiency loss. The method can differentiate between different types of defects and pinpoint the defects that lead to the highest efficiency shortfall, enabling manufacturers to troubleshoot production lines in a fast and cost-effective manner. The proposed approach unlocks the potential of luminescence imaging as an effective end-of-line characterization tool.

Bulk defects in silicon solar cells are a key contributor to loss of efficiency. To detect and identify those defects, temperature- and injection-dependent lifetime spectroscopy is usually used, and the defect parameters are traditionally extracted using fitting methods to the Shockley-Read-Hall recombination statistics. In this study, we propose a deep learning-based extraction technique that is based on an alternative representation of the lifetime curves: lifetime mapping in the temperature and minority carrier concentration space. The deep learning approach successfully predicts all the defect parameters while addressing one of the main limitations of the traditional approach of locating the defect in the energy spectrum, which usually outputs two possible solutions. Furthermore, the approach is applied to temperature-dependent defect parameters where the traditional approach is not applicable, achieving satisfying levels of prediction of the defect parameters. Image representation and deep learning have the potential to bolster solar cell characterization techniques by extracting more insights from the characterization data.

The temperature coefficient (TC) is a critical figure of merit to accurately evaluate the performance of solar cells at various operating temperatures, and hence, enabling the comparison between different cell technologies. Recently, tunnel oxide passivated contact (TOPCon) solar cells have shown outstanding cell performance. They are likely to be adopted in production lines and deployed in the field in the near future. Therefore, knowledge of their TCs and insights into their performance at realistic operating temperatures are of significant interest. In this study, we investigate the temperature dependence of the performance of TOPCon solar cells and quantify their TCs. To gain better understanding regarding the temperature-dependent behavior of their performance, the passivation quality and the contact resistivity of polysilicon (poly-Si) passivating contacts as a function of temperature are investigated. Although an improvement of the passivation quality of these contacts with increasing temperature has been observed, it seems that this improvement weakly impacts on the open-circuit voltage TC. The cell performance at elevated temperatures is dominated by the drop in the open-circuit voltage, associated with the intrinsic carrier concentration related to band gap narrowing. The fill factor TC (TCFF) is superior to those of other cell structures reported in the literature. We attribute this favorable TCFF to the fact that some of the fill factor losses are compensated by the decrease of contact resistivity of the poly-Si passivating contacts at elevated temperatures. The relative TC of the cell efficiency of the investigated TOPCon cells is comparable to the TC of silicon heterojunction cells and it is superior to those of cell structures without passivating contacts. Moreover, we found that the investigated solar cell is more sensitive to temperature variation at lower illumination intensities.

Silicon dioxide (SiO2) has played a critical role in the development of high-efficiency silicon (Si)-based photovoltaic devices. Recently, it has experienced a renaissance as an interlayer in many of the new contact passivating structures. Studies have extensively investigated the recombination process at the Si–SiO2 interface, however, only little is known about the impact of temperature on the surface recombination. In this study, we investigate the recombination at the Si–SiO2 interface by varying the temperature, excess carrier density, and dielectric fixed charge. An improved lifetime is observed with increasing temperature. A forming gas anneal is used to improve the passivation quality, however, at higher temperatures, the hydrogenated interface passivation degrades due to increased surface state density. The degradation is stronger for corona-charged SiO2, due to the instability of the corona charge within the dielectric. Using the extended Shockley-Read Hall recombination model, the Si–SiO2 interface defects’ parameters are extracted. Most importantly, we determine the value and the temperature-dependence of the capture cross-sections at this interface.

© 2020 American Chemical Society. To keep improving the efficiency-to-cost ratio of photovoltaic solar cells, manufacturing lines must be continuously improved. Efficiency optimization is usually performed process-wise and can be slow and time-consuming. In this study, we propose a machine-learning-based method to perform simultaneous multiprocess optimization. Using the natural variation of a production line, we train machine learning models to investigate the relationship between process parameters and cell efficiency. We employ genetic algorithms to identify new process parameters in order to maximize cell efficiency. The proposed method is demonstrated on a simulated production line of monocrystalline aluminum-back surface field solar cells. Using neural networks, an accurate model is built to predict cell efficiencies from input process parameters with errors of <0.03% absolute efficiency. In five iterations, the mean cell efficiency increases from 18.07% to 19.45%. Provided strong process monitoring and accurate wafer tracking, the proposed method is directly applicable to production-type datasets, enabling the photovoltaic industry to build smart factories and join the fourth industrial revolution.

© 2011-2012 IEEE. In this article, we present a method to obtain implied open-circuit voltage images of silicon wafers and cells at different temperatures. The proposed method is then demonstrated by investigating the temperature coefficients of various regions across multicrystalline silicon wafers and cells from different heights of two bricks with different dislocation densities. Interestingly, both low and high temperature coefficients are found in dislocated regions on the wafers. A large spread of temperature coefficient is observed at regions with similar performance at 298 K. Reduced temperature sensitivity is found to be correlated with the increasing brick height and is exhibited by both wafers and cells. This may indicate that cells made from the top of the brick, although having higher defect concentration, actually suffer relatively less degradation in the performance at higher temperatures.

© 2019 Author(s). We investigate the limitations of two-photon absorption time-resolved photoluminescence to measure the low-injection bulk lifetime of different semiconductor materials under varying surface recombination. The excitation source is assumed to be a sub-bandgap pulsed laser and the localized absorption and carrier generation was modeled using a focused TEM00 Gaussian beam under the assumption of diffraction-limited performance. The subsequent carrier kinetics were simulated by applying the finite-difference time-domain method to the continuity equation. Three typical semiconductor materials were modeled: direct bandgap low-mobility material (such as CZTS), direct bandgap high mobility (such as GaAs), and indirect bandgap high mobility (such as float-zone silicon). The extracted effective lifetime as a function of surface recombination velocity was compared to the bulk lifetime and the effective lifetime calculated using an analytical 1D approximation. For the direct bandgap materials, focusing inside the material yields an effective lifetime within a few percent of the bulk lifetime, regardless of the surface recombination velocity, while for excitation close to the surface it is up to 30% lower than the bulk lifetime at high surface recombination velocities (>10 4 cm/s). For the indirect bandgap material, the effective lifetime is dominated by the surface, making the bulk lifetime inaccessible, even at surface recombination velocities of 100 cm/s. Finally, we use the 1D approximation to find under what conditions the bulk lifetime can be extracted by this method and determine that both the bulk diffusion length and the product of the bulk lifetime and surface recombination velocity must be much less than twice the device thickness.

In this work, we investigate the use of advanced hydrogenation and low-temperature diffusion processes (a 3 h 700 °C process after emitter diffusion) for the electrical neutralization of laser-induced defects for laser doped and grooved solar cells. Despite the laser doping and grooving (LDG) process being performed before silicon nitride passivation to avoid thermal expansion mismatch between the silicon and the silicon nitride layer, some crystallographic defects are still formed during the process. The application of a low-temperature diffusion process increases implied open circuit voltages by 14 mV, potentially due to phosphorus diffusion of dislocated regions induced during laser processing. Laser hydrogenation is shown to be capable of passivating the majority of the remaining laser-induced defects. Over 1% absolute improvement in efficiency is achieved on cells with a full area aluminum back surface field. Preliminary results with minimal optimization demonstrate efficiencies of over 19% with a full area Al back contact cell. The potential to achieve much higher voltages when used with a passivated rear is also demonstrated.

Most previous studies regarding the trapping effect in quasi-steady-state photoconductance measurements assume that traps are measured in a steady-state regime. In this contribution, we demonstrate that even under conventional quasi-steady-state measurement conditions, the traps can be in a transient regime. Thus, previous analyses may have led to a misinterpretation of the trap properties. This study highlights the difference between steady-state and transient regimes of the trapping effect and focuses on the analysis of traps in the transient regime. Additionally, a method to extract the trap parameters from transient photoconductance is proposed and applied to data from n -type Czochralski silicon.

The surface recombination at metallized surfaces of a solar cell is of significant interest for cell process development, considering the limiting effect such recombination has on cell efficiency. This recombination is difficult to measure accurately, because the enhanced recombination at the metal contacts causes an inhomogeneous minority carrier profile, which limits the viability of conventional photoconductance-based recombination measurements. In this study, a photoluminescence-based system is used to measure the injection-dependent effective lifetime of full area metallized samples. Several techniques to extract surface recombination parameters from these data are compared. A simulation-based approach is shown to be superior to the simplified analyses that are more commonly applied to data of this type, in the case where the depth profile of the minority carrier density is nonuniform.

Operation of photovoltaic systems in the most economical way requires monitoring the modules' performance in order to ensure that the maximum power output is obtained. Existing methods for field inspection have limited capability of detecting various electronic defects that can, however, be identified with luminescence-based methods. This study presents a fully contactless outdoor photoluminescence imaging based measurement that uses the sun as the excitation source. This is achieved by modulating the module bias and hence luminescence signal by shading a single cell. We show that this technique can identify: (a) isolated areas, by modulating the cells in a sub-string between different operating points; and (b) open-circuit bypass diode failure, which is impossible to identify in contactless fashion using conventional field-inspection methods. The main merit of this method is that it can be used when the module is under normal operation in the field, without having to change any of the electrical wirings of the photovoltaic array.

© 2018 IEEE. This paper discusses the role hydrogen plays in degradation of silicon solar cells and modules. Slightly unorthodox, it presents a high level view of the latest research findings and theories of Professor Stuart Wenham before he suddenly passed away. This paper includes 6 main parts: (1) a brief introduction including Professor Wenham's prior hydrogen related work; (2) evidence for the role of excess hydrogen in Light- and elevated Temperature- Induced Degradation (LeTID) (3) discussion of Prof. Wenham's Bucket Theory' - how hydrogen could move and change to cause LeTID or Hydrogen- Induced Degradation (HID); (4) how hydrogen can cause recombination including Prof. Wenham's realization that hydrogen can cause recombination on its own (Hydrogen- Induced Recombination); (5) demonstration that appropriate control of hydrogenation processes can enable wafers and solar cells that are stable against LeTID/HID; and (6) implications and testing.

The surface recombination at metallized surfaces of a solar cell is of significant interest to process developers, considering the limiting effect such recombination has on cell efficiency. This recombination is difficult to measure accurately because the higher recombination at metallized surfaces compared to other passivated surfaces causes an inhomogeneous minority carrier profile. In this study, a novel photoluminescence-based system is used to measure the injection dependent effective lifetime of metallized samples, and several techniques to extract surface recombination parameters are compared. A simulation-based approach is shown to be superior to the simplified analyses that are more commonly applied to data of this type, in the case where the minority carrier profile is non-uniform.

To operate photovoltaic power plants at maximum capacity, it is desirable to identify cell or module failures in the field at the earliest possible stage. Currently used field inspection methods cannot detect many of the electronic defects that can be revealed with luminescence-based techniques. In this work, photoluminescence images are acquired using the sun as the sole illumination source by separating the weak luminescence signal from the much stronger ambient sunlight signal. This is done by using an appropriate choice of optical filtering and modulation of the cells' bias between the normal operating point and open circuit condition. The switching is achieved by periodically changing the optical generation rate of at least one cell within the module. This changes the biasing condition of all other cells that are connected to the same bypass diode. This method has the advantage that it can deliver high quality images revealing electrical defects in individual cells and entire modules, without requiring any changes to the electrical connections of the photovoltaic system.

In this study, we use temperature-dependent microphotoluminescence spectroscopy to investigate non-uniform bulk defects in -type Czochralski and float-zone silicon wafers. For both wafer types, we observe defect photoluminescence (PL) with peak energy between 0.78 eV and 0.85 eV emerge as the temperature is decreased from 300 K to 80 K, along with an associated 10-100 times reduction in the band-to-band (BB) PL. This similarity in the shape of the spectral PL between the two wafer types suggests that the origin of the defects may be related. For the float-zone wafers, the defect PL is present even in regions appearing bright in conventional band-to-band PL imaging, suggesting that the bulk defect exists across the entire wafer, but is spatially inhomogeneous. Defect parameters are extracted by fitting the ratio of the integrated defect to band-to-band PL ratio using a modified Arrhenius equation derived from the ShockleyRead-Hall defect theory. We determine similar defect parameters for the FZ wafer within the bright and dark regions: A single defect species located at 150 to 160 meV from the valence band-edge and a capture cross-section temperature-dependence following an inverse power law with exponent of approximately 2. For the Cz wafer the defect is located 325 meV from the valence band-edge, as expected from the peak PL position, and a capture cross-section temperature exponent of approximately 2.5 was determined. This suggests that a different recombination process is responsible for the same defect PL band in each wafer type

Multicrystalline silicon (mc-Si) wafers show degradation due to carrier injection. This carrier-induced degradation particularly limits the performance of passivated emitter rear cell (PERC) made with mc-Si and is, therefore, gaining increasing attention from the photovoltaic community, as it is expected that PERC architecture will become dominant in the near future. In this study, we investigated the degradation behaviour of mc-Si wafers with six different silicon nitride passivation layers which were then separately fired at six different peak temperatures (a matrix of 36 conditions). We found that wafers fired at a low peak temperature (< 650 °C) had similar degradation rates independent of the silicon nitride deposition conditions, whilst the degradation rate of samples fired at higher temperature showed a clear dependence on these conditions. At high firing temperatures normally used for contact formation, it appears that there is a correlation between the degradation rate and the amount of hydrogen released from the dielectric during firing. Furthermore, we verified that no degradation of the surface passivation quality occurs, indicating that the degradation is primarily associated with a bulk-related defect.

Temperature- and injection-dependent lifetime spectroscopy (TIDLS) is employed to study the defect responsible for light- and elevated temperature-induced degradation (LeTID). In our previous analyses, titanium (Ti), molybdenum (Mo), and tungsten (W) were identified as potential candidates for LeTID. The addition of temperature dependence further constrains the defect parameters. Assuming constant defect parameters with temperature, we identify two possible sets of defect parameters: k = 23.9 ± 5.5 at E t -E i = -0.21 ± 0.06 eV and k = 23.5 ± 5.6 at E t -E i = -0.10 ± 0.07 eV. We consider our results in the context of published defect parameters identified by TIDLS in other LeTID samples, and we evaluate our results against reported defect parameter temperature dependencies for Ti and Mo. We conclude that Mo is most consistent with our measurements. Approaches beyond lifetime spectroscopy, including intentional contamination and chemical composition measurements, are required to determine the root cause of LeTID.

Photoluminescence imaging is a fast and powerful technique to extract information about the nonuniformity of the recombination properties of silicon wafers and solar cells. In conventional photoluminescence imaging, where a uniform illumination is used, lateral carrier flows within the sample lead to smeared and quantitatively inaccurate results. In this study, a modified photoluminescence imaging setup, based on an adjustable spatially nonuniform excitation light source, is presented. With this setup, a laterally uniform excess carrier concentration is achieved across samples with nonuniform recombination properties by adaptively adjusting the light intensity at each position across the sample. Uniform photoluminescence images are obtained and the sample nonuniformity is extracted from the nonuniform excitation image that is applied. Without the presence of lateral carrier flows, carrier smearing effects are largely suppressed. Meanwhile, photoluminescence imaging-based quantitative analysis methods, such as carrier lifetime images, metastable defect imaging, etc., will greatly benefit from the fact that the sample is at a uniform excess carrier density.

Illumination intensity dependent open-circuit voltage measurements, commonly known as Suns-Voc, are often used to measure the current-voltage characteristic of a solar cell without the impact of series resistance. Deviations at high illumination levels between Suns-Vocmeasurements and contactless measurements, such as injection-dependent photoluminescence, have previously been reported. These deviations are analyzed in detail in this paper and shown to cause significant errors when converting Suns-Voc data to injection-dependent minority carrier lifetimes. Numerical modeling is used to identify shading and contact recombination as main causes of these deviations. Experimental data are used to demonstrate the magnitude of this effect for a range of different cell types.

Multicrystalline silicon (mc-Si) wafers show degradation due to carrier injection. This carrier-induced degradation particularly limits the performance of passivated emitter rear cell (PERC) made with mc-Si and is, therefore, gaining increasing attention from the photovoltaic community, as it is expected that PERC architecture will become dominant in the near future. In this study, we investigated the degradation behaviour of mc-Si wafers with six different silicon nitride passivation layers which were then separately fired at six different peak temperatures (a matrix of 36 conditions). We found that wafers fired at a low peak temperature (< 650 °C) had similar degradation rates independent of the silicon nitride deposition conditions, whilst the degradation rate of samples fired at higher temperature showed a clear dependence on these conditions. At high firing temperatures normally used for contact formation, it appears that there is a correlation between the degradation rate and the amount of hydrogen released from the dielectric during firing. Furthermore, we verified that no degradation of the surface passivation quality occurs, indicating that the degradation is primarily associated with a bulk-related defect.