Search

The interest in cast mono silicon is increasing due to its lower energy consumption and resulting smaller carbon footprint, lower oxygen content and resulting less oxygen-related defects as well as easy scalability to large wafer formats like 210 × 210 mm2 full square. As a cast silicon alternative to high performance multicrystalline (hpm) silicon, which rapidly lost market share, we analyze the cell efficiency potential of cast mono silicon in a TOPCon cell structure. We show how the absence of grain boundaries and the exceptional tolerance of the material quality towards high temperature processing enable this significant increase of the cell efficiency potential compared to hpm silicon. The very effective suppression of crystal defects by the Seed Manipulation for ARtificially controlled defect Technique (SMART) results in a very low lateral variation of the high material quality. We present certified cell efficiencies of 23.3% on n-type material crystallized in our labs, which demonstrates the high efficiency potential even for our lab-type G2 crystallization. An additional crystallization experiment for 210 × 210 mm2 wafers demonstrates that SMART mono is compatible to large wafer sizes. A significant difference of the crystallization costs for Cz and cast mono crystallization as a function of electricity costs is discussed.

We propose a method of three-dimensional (3D) visualization of growth interfaces in crystalline ingots prepared by directional solidification and apply the method to cast silicon ingot for solar cells. The method consists of dispersing inclusions along with the growth interface by insertion of a quartz rod followed by measuring inclusions distribution by an infrared brick inspection system. Then, inclusions distribution can be visualized in 3D, which allows extracting growth interfaces as planes. By fitting the plane with a simple polynomial expression, its macroscopic shape and local gradient can be estimated. The method yields us the local changes in growth interfaces and thus contributes to the development of the crystal growth method for high-quality ingots.

In this contribution, we present our recent results for high efficiency multicrystalline silicon solar cells. Based on n-type high-performance multicrystalline silicon substrates in combination with the TOPCon solar cell concept featuring a full area passivating back contact and a boron-diffused emitter as well as a plasma-etched black-silicon texture at the front side, a certified conversion efficiency of 22.3% has been achieved, which is currently the world record efficiency for multicrystalline silicon solar cells. A detailed loss analysis of the record solar cell batch discloses the nature of the remaining loss mechanisms, revealing the route for further improvements. We observe an efficiency gap between the multicrystalline and the FZ reference solar cells of ~1%abs. Compared to the FZ reference cells, the mc-Si cells also feature a significantly larger scattering in Voc and Jsc as well as a fill factor loss of ~1.5%abs. We show that the scattering in Jsc correlates with the area fraction of recombination-active structural crystal defects and the scattering in Voc additionally with lateral emitter-induced inhomogeneities. The fill factor loss is attributed to the general presence of strongly recombination-active grain boundaries. A detailed loss analysis of the record mc-Si solar cell shows that the major electrical losses are due to recombination at grain boundaries (0.7%abs) and recombination in the emitter (0.6%abs). By reducing these electrical loss channels, e.g. by an improved crystallization process together with a hydrogenation of the bulk and application of an adapted emitter, we expect to reach efficiencies for mc-Si solar cells in the range of 23%.

For the optimization of the manufacturing process of multicrystalline silicon (mc-Si) for solar cells in order to reduce energy consumption and costs, it is important to know the distribution of the impurities. In our project, we analyzed the raw materials, the crucibles, and the solidified blocks (ingots) with different methods based on neutron activation. In this paper, we present the technique and the results of the boron and hydrogen determination with prompt gamma activation analysis (PGAA). We show that the distribution of boron in the ingots can be successfully described with a Scheil curve. Concerning the raw material, the most relevant information is the average elemental mass fraction. For this purpose, PGAA is well suited since it provides a bulk analysis with detection limits down to the low ng/g range for boron and µg/g range for hydrogen. In this context, the hydrogen content of fluidized bed reactor (FBR) feedstock is discussed in detail. These were the first extensive PGAA studies of the boron and hydrogen content in mc-Si.

Multicrystalline (mc) n-type silicon has proven to be a suitable substrate for the fabrication of highly efficient mc-Si solar cells. In this paper, we elaborate the impact of base material parameters on the efficiency potential of n-type mc-Si solar cells featuring a boron-diffused front side emitter and a full-area passivating rear contact (TOPCon). The electrical material quality can be significantly improved by replacing the standard crystallization process with a seed-assisted growth for crystallization of high-performance (HP) mc silicon. Using high-purity quartz crucibles or larger crucibles in combination with an optimization of the grain boundary area fraction with an adapted seed structure leads to further improvements of the material quality in terms of charge carrier lifetimes. However, not only the charge carrier lifetime, but also the base resistivity is of crucial importance for the efficiency potential depending on the cell concept. Based on experimental data and simulations, we assess the optimal range for the base resistivity and the wafer thickness for n-type mc-Si TOPCon solar cells. With the optimal material parameters, an “efficiency limiting bulk recombination analysis” (ELBA) reveals an efficiency potential in the range of 22.5% for n-type mc-Si TOPCon solar cells. Finally, we fabricated TOPCon solar cells based on optimized n-type HP mc-Si substrate and demonstrate a certified efficiency of 21.9%, which is the highest efficiency reported for multicrystalline silicon solar cells so far.

In this contribution, we demonstrate a route for efficiencies exceeding 22% with n-type multicrystalline (mc) silicon solar cells based on the TOPCon cell concept featuring a boron-diffused front side emitter and a full-area passivating rear contact. By applying a “high-performance” (HP) crystallization process with an adapted seed structure in order to obtain an optimized grain boundary area fraction, we reduce recombination losses in the HP mc-Si material to a minimum. We discuss the electrical properties of the optimized n-type HP mc-Si, which features very low material-related efficiency losses of approximately 0.5% abs and, thus, enables an efficiency potential of 22.6% with regard to a cell limit of 23.1% of the TOPCon cell concept adapted for multicrystalline silicon. Results at the device level reveal a record efficiency of 21.9%, which is the highest efficiency reported for a multicrystalline silicon solar cell. Finally, we discuss the deviations between the predicted efficiency potential and the solar cell results.

Silicon solar cells featuring the highest conversion efficiencies are made from monocrystalline n-type silicon. The superior crystal quality of high-performance multicrystalline silicon (HP mc) in combination with the inherent benefits of n-type doping (higher tolerance to common impurities) should allow the fabrication of high-efficiency solar cells also on mc silicon. In this paper, we address high-efficiency n-type HP mc solar cells with diffused boron front emitter and full-area passivating rear contact (TOPCon). n-type HP mc silicon was crystallized at Fraunhofer ISE featuring a very high average lifetime in the range of 600 μ s (i.e., diffusion length >800 μ m) after application of all high-temperature steps necessary for cell fabrication. Using a “black silicon” front texture we have achieved a weighted reflectance of ∼1% and simultaneously a very good electrical performance, i.e., J 0 e values of ≤ 60 fA/cm2 for a 90 Ω/sq emitter. The resulting n-type mc silicon solar cells show certified conversion efficiencies up to 21.9%, representing the current world record for mc silicon solar cells.

In High-Performance mc-Si [1] random grain boundaries, although being recombination active, often enhance material quality by reducing dislocations. With this work, we take a step towards statistical large-scale investigations of crystal defects via a combined analysis of different inline-measurements on as-cut wafers: photoluminescence images for the extraction of recombination-active structures and reflection and infrared transmission images for the extraction of the grain structure. The combined extraction of recombination-active structures and grain structures allows isolating dislocations from grain boundaries for all material types. To discern dislocations from other recombination-active defect structures, an image-processing-based analysis technique has been developed. By applying this separation on wafers from various bricks of our material set, typical developments of grain structure and dislocations can be identified. As a particular application, we investigate the correlation between the development of dislocations in higher parts of the brick and grain size in the lower parts. The results support theory quantitatively: Dislocation ratio in the upper brick part shows a correlation with the square root of the weighted median of the grain size in the lower brick part (R≈0.85). However, the results also show that grain size distribution, in particular grain size homogeneity, has to be considered to account for a stronger distinction between materials.

Correlations between defect-related luminescence (DRL) and recombination mechanisms of multicrystalline silicon wafers are investigated by hyperspectral photoluminescence (PL) imaging at cryogenic temperatures (∼80 K) and by PL-based techniques for charge carrier lifetime at room temperature. This unique combination of measurement techniques is used to spectrally compare the DRL in n-type and p-type wafers and to investigate the DRL as a function of block height in a p-type block. Further, the dependence of DRL on interstitial and precipitated metallic impurities has been investigated by comparison of simulated concentration profiles of interstitial and precipitated iron with the spatial distribution of DRL. Our results indicate that the origins of the dislocation-related emission lines (D-lines) are independent of the doping type and suggest that the spectral shape, rather, is determined by the dominating recombination mechanism in the material. In regions with high structural defect density, we observe increased intensities of the D-lines D1-D4 in the DRL spectrum. In regions with a high concentration of either iron or other metallic precipitates, we observe reduced emission intensities of D3 and D4. It is, thus, likely that precipitates of either iron or other impurities partly supress the D3 and D4 emission intensities.

The introduction of directional solidified cast mono silicon promised a combination of the cheaper production via a casting process with monocrystalline material quality, but has been struggling with high concentration of structural defects. The SMART approach uses functional defects to maintain the monocrystalline structure with low dislocation densities. In this work, the feasibility of the SMART approach is shown for larger ingots. A G2 sized crystal with SMART and cast mono silicon parts has been analyzed regarding the structural defects via optical analysis, crystal orientation, and etch pit measurements. Photoluminescence measurements on passivated and processed samples were used for characterization of the electrical material quality. The SMART approach has successfully resulted in a crystal with mono area share over 90% and a confinement of dislocation structures in the functional defect region over the whole ingot height compared to a mono area share of down to 50% and extending dislocation tangles in the standard cast mono Si. Cellular structures in photoluminescence measurements could be attributed to cellular dislocation patterns. The SMART Si material showed very high and most homogeneous lifetime values enabling solar cell efficiencies up to 23.3%.

Within this work, we present a method for fast characterization and statistical analysis of structural and electrical crystal properties. By this, we investigate the impact of grain size and grain shape distribution in the lower part of the brick on the defect development in the upper part of the brick. Our method is based on fast measurements on as-cut wafers, namely photoluminescence imaging and reflection or infrared transmission images. We combine the extracted information and isolate dislocation structures from recombination active grain boundaries via image-processing. Two large data sets (more than 40 bricks) of different materials allow us to quantify the assumed correlation between grain structure and dislocation development. The results confirm that the relative length of dark lines as a part of dislocation clusters in the upper brick part shows a locally linear correlation with the square root of a meaningful grain area measure in the lower brick part (Pearson coefficient R between 0.80 and 0.88). It also correlates with the inhomogeneity measures for grain area and shape (Pearson coefficient R about 0.88). As a result, changes in the crystallization process may be targeted according to the grain structure in the lower part of the brick. In the future, this method may serve to improve the prediction of solar cell results based on wafer-data.

In this study, we demonstrate the potential of multicrystalline (mc) n-type silicon for the fabrication of highly efficient mc-Si solar cells. High-quality mc n-type silicon wafers are obtained from a research ingot crystallized in a high-purity crucible, using high-purity granular silicon as seed layer in the crucible bottom and high-purity polysilicon feedstock for the block. An mc p-type silicon block crystallized under identical conditions (same seed and feedstock, crucible system, and temperature profiles) serves as reference and enables measurements of the interstitial iron and chromium concentrations by metastable defect imaging. In combination with 2-D simulations for in-diffusion and precipitation of chromium, the limitation of n-type high-performance mc silicon by these metals is assessed after different solar cell processing steps. Material-related efficiency losses are assessed by an “efficiency limiting bulk recombination analysis,” which combines injection-dependent photoluminescence imaging of minority charge carrier diffusion length with PC1D cell simulations. Finally, based on this material, boron-diffused front-junction mc n-type silicon solar cells with a full-area passivated rear contact (TOPCon) are fabricated. The record cell features an efficiency of 19.6%, which is the highest efficiency reported for an mc n-type silicon solar cell.

In general, the charge carrier lifetime in n -type silicon is less sensitive to common dissolved metals compared to p -type silicon. However, in experiments it was observed that metal impurities limit the lifetime in n -type multicrystalline (mc) silicon even if high purity feedstock was used. By evaluating Neutron Activation Analysis (NAA) and Inducitvely Coupled Plasma Mass Spectrometry (ICP-MS) at blocks grown with high purity feedstock, we identified the main currently unavoidable metal impurities present in n -type mc silicon. In addition, we measured the charge carrier lifetime on surface-passivated wafers after different solar cell processes. The measurements were compared to simulations in order to identify the limitations by different impurities and to evaluate the influence of metal precipitates. We found that Cr i is an important defect in good grains of as-grown wafers. Dissolved Co can have a severe impact on lifetime after process steps with fast cooling. The lower lifetime in the edge region of the blocks is attributed to FeSi 2 precipitates, which explains the poor gettering response of the edge region in contrast to grains in the block center. From the results obtained for FeSi 2 -precipitates, we concluded that Cu 3 Si- and NiSi 2 -precipitates located at crystal defects may be responsible for significant recombination in the block center.

This study aims to better understand the influence of crystallographic structure and impurity decoration on the recombination activity at grain boundaries in multicrystalline silicon. A sample of the upper part of a multicrystalline silicon ingot with intentional addition of iron and copper has been investigated. Correlative electron-beam-induced current, electron backscatter diffraction, and atom probe tomography data for different types of grain boundaries are presented. For a symmetric coherent Σ3 twin boundary, with very low recombination activity, no impurities are detected. In case of a non-coherent (random) high-angle grain boundary and higher order twins with pronounced recombination activity, carbon and oxygen impurities are observed to decorate the interface. Copper contamination is detected for the boundary with the highest recombination activity in this study, a random high-angle grain boundary located in the vicinity of a triple junction. The 3D atom probe tomography study presented here is the first direct atomic scale identification and quantification of impurities decorating grain boundaries in multicrystalline silicon. The observed deviations in chemical decoration and induced current could be directly linked with different crystallographic structures of silicon grain boundaries. Hence, the current work establishes a direct correlation between grain boundary structure, atomic scale segregation information, and electrical activity. It can help to identify interface–property relationships for silicon interfaces that enable grain boundary engineering in multicrystalline silicon. Copyright © 2015 John Wiley & Sons, Ltd.

In a series of crystallization experiments, the directional solidification of silicon was investigated as a low cost path for the production of silicon wafers for solar cells. Instrumental neutron activation analysis was employed to measure the influence of different crystallization parameters on the distribution of 3d-metal impurities of the produced ingots. A theoretical model describing the involved diffusion and segregation processes during the solidification and cooling of the ingots could be verified by the experimental results. By successive etching of the samples after the irradiation, it could be shown that a layer of at least 60 μm of the samples has to be removed to get real bulk concentrations.

Feeding of liquid silicon during the directional solidification process is a promising opportunity for cost reduction by increased throughput and improved material homogeneity due to constant resistivity over ingot height. In this work, a liquid feeding apparatus was developed for an industrial type directional solidification furnace. One n-type G2 sized High Performance multicrystalline ingot with liquid feeding of additional 14 kg of undoped silicon feedstock was crystallized. The resistivity was kept within a range of ±0.1 Ω cm of the target resistivity during the feeding sequence. A smaller mean grain area growth was observed during feeding, whereas the area fraction of recombination active dislocation structures was as low as in a reference ingot. Increased interstitial oxygen and substitutional carbon concentrations were measured for the ingot with liquid feeding. The measured mean bulk lifetime of 190 µs for passivated wafers in the feeding sequence can probably be increased by further pre-melting crucible improvements. For this laboratory experiment, energy reductions of 2% per wafer and time savings of 16% per wafer were realized.

This paper elaborates upon the theory of self-consistent minority carrier bulk lifetime measurements of silicon ingots via time-modulated photoluminescence and presents an experimental proof of concept. For silicon ingots, the solution of the continuity equation at harmonic time modulation of excess carrier generation is shown to generally reveal a remarkably pronounced contrast with respect to minority carrier bulk lifetime. We combine our dynamic self-consistent approach with an analysis of surface recombination velocity from photoluminescence intensity ratios upon irradiation with different laser wavelengths. This combined analysis enables an accurate simultaneous determination of bulk lifetime and surface recombination velocity. For sufficiently high or accurately known ingot surface recombination velocities, this approach could likewise be used for an accurate determination of the minority carrier diffusion coefficient and of minority carrier mobility in novel silicon materials.

This work presents recent advances in the characterisation of carrier recombination and impurities at Fraunhofer ISE. The role of iron contamination during crystallisation is analysed in more detail. Numerical simulations and comparisons to experimental data are presented which demonstrate the impact of iron from the crucible and crucible coating and show the in-diffusion of iron into the silicon melt as well as into the solid silicon during crystal cooling. Measurements of spatially resolved carrier lifetime and interstitial iron concentration on wafers after phosphorus diffusion gettering are used as input for cell efficiency modelling which reveals the specific and quantitative role of iron on cell parameters in multicrystalline silicon. A new photoluminescence based method is presented which quantitatively determines the interstitial iron concentration in finished solar cells. We finally present advances in defect characterisation with sub-micrometre resolution: We show recent progress in micro photoluminescence spectroscopy for the quantitative measurement of interstitial chromium with high spatial resolution. A further development of this setup will be discussed: By combining the principle of Light Beam Induced Current (LBIC) or voltage (LBIV) and the highly localized illumination, images of carrier recombination at local defects are presented which feature a, compared to EBIC, higher signal-to-noise ratio.

The aim of this paper is to analyze the limiting role of crucible and coating impurities on material quality of multicrystalline silicon. Both solid body diffusion and diffusion into the silicon melt are considered in this study. Two ingots of size G1 have been analyzed. One of them was crystallized in a standard crucible, whereas the other was crystallized in a quartz crucible of very high purity. Focus is put on iron and cobalt as examples of typical impurity species. Iron was found in large concentrations in standard crucibles, and cobalt was proven to be a suitable marker impurity that is mainly found in the coating. Inductively coupled plasma mass spectroscopy data are exploited for the determination of impurity concentrations in crucible, coating, and within the crystal. With higher sensitivity for low concentration, PL imaging is applied for carrier lifetime and interstitial iron concentration measurements. The different findings are compared with modeling results of iron and cobalt in-diffusion by Sentaurus Process. The analysis of silicon wafers before and after gettering steps enable a quantification of impurity-limiting cell efficiency potential. Conclusions about the role of impurities from coated crucibles in large-scale crystallization are deduced.

A fast and thorough characterization of grain structure in multicrystalline silicon (mc-Si) is crucial to improve crystal growth and thus bulk lifetime in solar cells. The presented characterization techniques are based on simple optical measurements on as-cut mc-Si wafers. An insight into the entire brick is gained by connecting 2D-information, computed via advanced pattern recognition techniques, over brick height. We identify robust statistical key parameters. Their development within typical bricks of different cast-Si techniques is compared and it is found that the distinct behavior of different materials in the lower part of the brick subsides towards the brick top where grain size distribution is similar.

We present a method for the purification of silicon ingots during the crystallization process that reduces significantly the width of the low charge carrier lifetime region at the ingot top. The back-diffusion of impurities from the ingot top is suppressed by adding a small amount of pure aluminum into the silicon melt right at the end of the solidification. We study the aluminum gettering effect by instrumental neutron activation analysis (INAA) and Fei imaging. Furthermore, we present a model for aluminum gettering of Fe in the silicon ingot that is in agreement with literature data for aluminum gettering at lower temperature. The distribution of iron in the ingots with and without aluminum is fairly well predicted by a combination of this model with a model for Fe contamination from the crucible system. A simulation with varying Al content exhibits further potential for an increased yield of silicon wafers with high charge carrier lifetime. (© 2016 WILEY-VCH Verlag GmbH &Co. KGaA, Weinheim)

Grain structure and grain competition have a strong impact on bulk lifetime in multicrystalline (mc) silicon. A fast and thorough characterization of grain structure is crucial in order to improve industrial crystal growth. This work introduces key parameters of grain structure, extracted with a newly developed image processing tool. Four bricks grown with different concepts were chosen to investigate the value of the identified key parameters and to identify characteristic developments along the brick. Optical measurements on as-cut wafers from these bricks serve as a basis to extract grain structure properties, i.e., size, shape, homogeneity and distribution of grain size. By connecting the 2D-information over brick height, a statistical insight into the entire brick is gained. Weighted percentiles of grain area offer a robust measure to characterize grain size distribution. As twinning has a large impact on grain competition, twinned grains are detected via grain shape. Additionally, regions with strong grain competition are highlighted for investigations on grain overgrowth. It is found that the share of twin grains increases with brick height in high-performance mc (HPM) silicon with fine-granular seeds from almost zero up to about 15% whereas it remains rather constant over the whole brick height in standard mc-Si. The results of the investigated bricks show clearly that towards the brick top, the material differences in grain size decrease. This suggests that an energetically favorable state may exist for grain structure development.

This paper investigates the impact of iron (Fe) and molybdenum (Mo) when they are introduced in the feedstock for mono- and multicrystalline Float-Zone (FZ) silicon (Si) growth. Neutron Activation Analysis shows that the segregation coefficient is in agreement with literature values. Lifetime maps on monocrystalline wafers show a uniform lifetime which decreases with the increase of contamination levels. Multicrystalline wafers show low lifetime areas, corresponding to grain boundaries and highly dislocated areas, which are independent from the contamination levels. Intra grain areas have a higher lifetime which changes with the contamination levels. The solar cells show a reduced diffusion length in multicrystalline uncontaminated cells compare to the monocrystalline uncontaminated. In multicrystalline cells the lowest level of Fe introduced (1012 atm/cm3) has hardly any influence, whereas in the Mo-contaminated cells the impact is visible from the lowest level (1011 atm/cm3). In monocrystalline cells the diffusion length is reduced already at the lowest contamination level of Fe.

This study aims for a quantitative investigation of the material limitations and the efficiency potential of an entire multicrystalline (mc) n-type silicon block in comparison with an mc p-type block of the same purity level in order to predict the potential of mc n-type silicon for the industrial production of solar cells. Therefore, two standard mc silicon blocks were crystallized under identical conditions (same high purity feedstock, crucible system, and temperature profiles), only differing in their type of doping. The material quality of wafers along the whole block height is analyzed after different solar cell process steps by photoluminescence imaging of the diffusion length. The bulk recombination related efficiency losses are assessed by an ""efficiency limiting bulk recombination analysis (ELBA),"" combining injection dependent lifetime images with PC1D cell simulations. The influence of the base resistivity variation along the block is considered in the PC1D cell simulations and backed up by Sentaurus Device simulations. This analysis predicts a significantly higher material-related efficiency potential after typical solar cell processes along the whole block height for mc n-type silicon compared with mc p-type silicon. In addition, the efficiency potential for mc n-type silicon depends less on block position.

Measurements of the dislocation density are compared with locally resolved measurements of carrier lifetime for p-type multicrystalline silicon. A correlation between dislocation density and carrier recombination was found: high carrier lifetimes (>100 µs) were only measured in areas with low dislocation density ( 106 cm−2) relatively low lifetimes (

The measurement of free carrier density in silicon is a crucial parameter for the characterisation of silicon solar cell material. Carrier Density Imaging (CDI) is a valuable tool to obtain spatially resolved images of the free carrier density distribution. This article describes the experimental setup of CDI for absorption mode and recently developed emission mode measurements. The theoretical dependence of the absorption and emission of infrared radiation on the free carrier density is discussed. Results of absorption and emission mode measurements are presented and the advantages of the new emission mode are elaborated.

We use photoluminescence imaging (PL) for quality control of the crystallization process at Fraunhofer ISE in order to find defects which will later limit the efficiencies of solar cells. Bricks of multicrystalline feedstock from electronic grade (EG) and upgraded metallurgical (UMG) silicon are subsequently wafered and solar cells are manufactured using a standard industrial solar cell process. PL is employed on bricks, as-cut wafers and finished cells and supplemented by additional measurement techniques such as microwave photo conductance decay (MW-PCD) or I–V curve measurements. In UMG material, the major problem is the presence of background dopants, which lead to compensation effects such as a pn-type changeover. As the cell efficiency of wafers within or beyond the type changeover drops significantly in the standard process, these wafers have to be detected reliably in the incoming test to be separated and introduced to an adapted solar cell process or discarded. To do such a separation, the position of the wafer from the brick needs to be known.

A model for the combined effect of dislocations and grain boundaries on minority carrier lifetime has been developed. Lifetime varies with dislocation density, grain boundary misorientation, and the coincidence site lattice (CSL) nature of the boundaries. Minority carrier lifetime was measured with high spatial resolution (50 mu m) using the carrier density imaging (CDI) technique on a silicon nitride passivated multicrystalline sample. Dislocation density was measured on the same sample by image recognition of optical microscope pictures of a Secco etched surface. Grain boundaries were then mapped and characterized by electron backscatter diffraction (EBSD). Lifetime was simulated based on the dislocation and grain boundary measurements. Parameters were chosen to match closely the simulated and measured maps. Very good two-dimensional (2D) correlation was obtained by assigning roughly equal importance to recombination at dislocations and grain boundaries. The value for the capture cross section, which gives the best correlation, is 4x10(-14) cm(-3). This is in the range of values reported for interstitial transition metals like, for instance, iron. It appears necessary to include also the effect of grain boundaries to explain recombination in low lifetime areas. Sub grain boundaries were particularly recombination active and are dominating the number of active grain boundaries.

The utilisation of progressively thinner wafers for solar cells leads to an increasing importance of rear surface passivation. In this work we investigate different solar cell processes for industrial rear side passivated solar cells using silicon nitride (SiN) on both sides and screen printed contacts. All solar cells were made by phosphorus emitter diffusion on p-type wafers. For a sufficient rear surface passivation, a parasitic rear emitter must be avoided. Therefore, we carefully compared different processes comprising rear SiN as barrier during phosphorus diffusion as well as parasitic emitter removal by etching after phosphorus diffusion. Furthermore two processes for rear side contacting through a silicon nitride layer were tested. The first process is firing of a screen printed aluminium grid through the backside SiN. The second process uses laser ablation to open a thick rear SiN followed by a full area screen printed rear contact. Additionally, a possible degradation of the SiN due to firing of aluminium contact fingers through the SiN layer was investigated using minority carrier lifetime topography before and after firing of the wafers. Efficiencies of up to 14.7 % were obtained on 200 mum thick wafers using SiN layers on both sides. However, these processes are still excelled by the conventional solar cell process using a screen printed full area aluminium rear contact, which lead to an efficiency of 15.3 % on a 200 mum thick neighbouring wafer

In this work we characterised recombination active defects in standard multicrystalline silicon block material by investigating the effect of a wet thermal oxidation step at 800 degC with fast temperature ramps on the effective lifetimes measured by the CDI and QSSPC methods. The average lifetime was significantly decreased by the oxidation step. An increase in the concentration of interstitial iron as well as an increase in the concentration of trapping centres after the oxidation step was observed. A detailed investigation revealed an increase in the recombination activity of grain boundaries. Modeling of measured carrier density profiles indicates that iron dissolves from precipitates in grain boundaries and diffuses into the grains during the process. It was found that iron represents a major part of the active defect concentration in the investigated multicrystalline material after the applied oxidation process

The passivation properties of phosphorus-doped amorphous silicon carbide (a-SixC1-x) layers on monocrystalline Si wafers (floating zone, 1 Omega cm) have been investigated. The cleaning and the deposition process were performed in our two source (microwave, radio frequency) plasma-enhanced chemical vapor deposition reactor. In situ plasma etching and deposition at 350 degrees C without any following annealing step led to extraordinary low surface recombination velocities of less than 5 cm/s for injection levels between 1x10(14) and 1x10(15) cm(-3). Characterization of the a-SixC1-x layer was done with quasi-steady-state photoconductance, microwave-detected photoconductance, and carrier density imaging techniques. For injection levels of more than 1x10(15) cm(-3) the effective lifetime was limited only by intrinsic Auger recombination.

The influence of the thermal treatment during the saw damage etching process on the mechanical stability of multicrystalline silicon wafers has been investigated for four different alkaline saw damage etching procedures. After the different etching procedures the mechanical stabilities of the multicrystalline wafers have been measured using the concentric ring test and the four-point-bending test. The experimental results are compared to those obtained by numerical simulations.

Carrier density imaging, a further development of the infrared lifetime mapping technique [Bail et al., Proceedings 28th IEEE-PVSC (2000)], is presented as an extremely fast, spatially resolved lifetime measurement method. It is based on the principle of absorption of infrared radiation by free carriers. The physical principles and necessary calibration procedures are discussed, showing that the steady state carrier density and thus actual carrier lifetime at low-level injection is measured, whereas many standard lifetime measurement techniques measure differential lifetimes only. A charge coupled device camera operating in the infrared is applied for measuring the infrared transmissivity of the sample. This overcomes the necessity of scanning the sample in order to receive laterally resolved information. Thus measurement times can be cut from hours to minutes or even seconds. An additional advantage is the absence of any moving parts which may cause spatial errors or even failures in the measurement. As is shown in the article the use of lock in technique and an appropriate source of illumination is essential to receive good results on production type materials. The quantitative comparison to other lifetime measurement techniques and the application to silicon samples at different process stages is discussed. Finally methods for further reduction of measurement time are discussed, showing that this method may well be suitable for in-line process control

Lifetime mappings are common tools for assessing the quality of mc-silicon for solar cell production. Nevertheless it is a difficult problem to directly relate lifetime mappings to overall solar cell parameters. This paper intends to show that this correlation is possible quantitatively. We have correlated actual low-level injection lifetimes obtained by carrier density imaging (CDI) measurements with overall cell parameters of solar cells processed on adjacent wafers. The 2D lifetime-structure is taken account for by appropriate weighing functions that include the whole information given in the frequency distribution of bulk lifetimes. Thus a one dimensional cell model (PC1D) can be applied. Good general agreement between predicted and measured cell parameters has been achieved, deviations are discussed. Further insight into the gettering behavior of block-east and Bridgman-grown mc-silicon was attained.

A method for the determination of the surface recombination velocity of cylindric vias for the application in back-contacted silicon solar cells is proposed. By applying a theoretical framework for the lifetime dependence on dislocation density, the measured minority carrier lifetime as a function of via density can be described accurately; thus the value for the surface recombination velocity at the cylinder barrel Svia is extracted. Accordingly, Svia after SiNx deposition from the front and the rear of the wafers systematically decreases if material surrounding the laser drilled via is removed prior to passivation by etching from 106to100cm∕s.

The dislocation dependence of open circuit voltage is studied based on Donolato's model for the effect of dislocations on minority carrier effective diffusion length [J. Appl. Phys. 84, 2656 (1998)]. Experimental data measured on thin-film solar cells show a strong decrease of open circuit voltage V-oc with an increase in defect density. The analysis of the recombination currents indicates that V-oc is largely reduced by space charge region recombination. For a quantitative study on the relationship between dislocation density, effective diffusion length, and V-oc the data are fitted with an extended version of Donolato's model. Taking into account the base recombination current as well as the space charge region recombination current, the modeled curves fit very well to the experimental data. However, satisfactory fitting results require that the region of high recombination is set wider than the "effective depletion region width," which is calculated from the electrical field strength in a planar p-n junction. This effect can be explained with the assumption of a geometrical enlargement of the p-n junction due to defects like dislocations.

In 1998, Donolato presented an analytical model describing the effect of dislocation density on minority carrier effective diffusion length [J. Appl. Phys. 84, 2656 (1998)]. While this analysis was derived for a "semi-infinite" specimen, our objective is the appropriate description of thin devices, such as wafer or thin-film based crystalline Si solar cells with back surface field or passivating layer on the rear side. Therefore, Donolato's model is extended for specimen of finite thickness and finite recombination velocity at the back surface. Since the associated boundary value problem does not allow a straightforward analytical solution, we derive an approximate expression, which is validated by numerical simulations. In the original work, Donolato uses a special definition of an "effective diffusion length." This definition is different from the quantity usually referred to as effective diffusion length when analyzing quantum efficiency data. Furthermore, Donolato's definition does not refer to the typical operation conditions of a solar cell. We therefore modify Donolato's model for the effect of dislocations consistently using the quantum efficiency effective diffusion length. Finally, our model is applied to the determination of dislocation recombination strength in thin-film solar cells with back surface field from effective diffusion length maps.

We present spatially resolved and injection dependent excess carrier lifetime measurements on silicon. At low level injection conditions an anomalous increase often interferes in such measurements. The origin of the anomalous increase is discussed. Assuming trapping as the origin, highly resolved images of trap parameters together with low level injection recombination lifetimes with strongly reduced trapping effects have been obtained. A theoretical model for infrared lifetime imaging based on the Hornbeck and Haynes model [J. A. Hornbeck and J. R. Haynes, Phys. Rev. 97, 311 (1955)] is presented which describes the trapping effect on the measured lifetime. This model is fitted to experimental spatially resolved data to extract trapping parameters, particularly trap density and the trap-escape ratio, i.e., the ratio between escape rate of minority carriers from the trap level into the minority band (detrapping) and trap rate from the minority band into the trap level (trapping). An upper bound for the low level injection recombination lifetimes is determined. Lifetime and trapping parameters are compared with dislocation density maps. A strong correlation is found between the total trap density and the crystal defect density.

Synchrotron-based analytical x-ray microprobe techniques were employed to study the dissolution of iron, copper, and nickel silicide precipitates at structural defects in cast multicrystalline silicon in response to rapid thermal processing (RTP). A direct correlation was observed between iron silicide precipitate dissolution, increased minority carrier recombination, and decreased device performance after high-temperature (1000°C) RTP. In contrast, iron precipitates comparable in size to as-grown material remained after lower-temperature RTP (860°C); in this case the material exhibited higher minority carrier diffusion length and better solar cell performance. RTP at both temperatures effectively dissolved nickel and copper silicide precipitates. It is concluded that iron dissolved from structural defect reservoirs detrimentally affects the cell performance, likely by forming distributed point defects and smaller precipitates. For cast multicrystalline silicon, higher performance can be expected by inhib...

Synchrotron-based, spectrally resolved x-ray beam-induced current (SR-XBIC) is introduced as a technique to locally measure the minority carrier diffusion length in semiconductor devices. Equivalence with well-established diffusion length measurement techniques is demonstrated. The strength of SR-XBIC is that it can be combined in situ with other synchrotron-based analytical techniques, such as x-ray fluorescence microscopy (μ-XRF) and x-ray absorption microspectroscopy (μ-XAS), yielding information about the distribution, elemental composition, chemical nature, and effect on minority carrier diffusion length of individual transition metal species in multicrystalline silicon. SR-XBIC, μ-XRF, and μ-XAS measurements were performed on intentionally contaminated multicrystalline silicon, revealing a strong correlation between local concentrations of copper and nickel silicide precipitates and a decrease of minority carrier diffusion length. In addition, the reduction of minority carrier diffusion length due t...

Seed-assisted growth of multicrystalline (mc) silicon ingots has led to a significant increase in mc silicon material quality (so called high performance multicrystalline silicon, “HPM Si”) and thus efficiencies of p-type mc silicon solar cells in the past years. Additionally, recent research revealed a high efficiency potential of n-type standard mc silicon due to its large tolerance to many metallic impurities. This suggests a high potential of n-type high performance mc silicon with a low density of dislocation clusters. In this study, the efficiency potential of p- and n-type high performance mc silicon, produced under identical conditions in the same laboratory crystallization furnace, is investigated after different high efficiency solar cell processing steps. The prediction of solar cell efficiencies by an “Efficiency limiting bulk recombination analysis” (ELBA) discloses a very high potential with material related losses of less than 1% abs. compared to the cell limit. In combination witha high efficiency cell concept featuring a full area passivated rear contact (TOPCon), efficiencies exceeding 20% on isotextured n-type HPM samples and close to 22% on samples with an improved texture are predicted. To support these predictions with results on actual devices, solar cells were fabricated. First results on cell level indicate high V oc and J sc values in the range predicted by ELBA simulations and an efficiency of 19.3%.