Search

Fast camera-based luminescence-imaging measurements on perovskite solar cells are presented. The fundamental correlation between the luminescence intensity and the open circuit voltage predicted by the generalised Planck law is confirmed, enabling various quantitative methods for the detection of efficiency-limiting defects to be applied to this new cell structure. Interstinegly, it is found that this fundamental correlation is valid only for light-soaked devices. Copyright © 2015 John Wiley & Sons, Ltd.

We present a method to display photoluminescence imaging results in a different way for open circuit voltage for metallised solar cells. This method virtually eliminates the influence of horizontal balancing currents, which have limited the validity of previously reported methods. It is achieved by creating a “virtual static” situation. At this virtual situation, there is no horizontal current flowing and the calculation of local open circuit voltage can be more reliable. Then effective lifetime image can be calculated more accurately. Compared to other spatially resolved lifetime measurement methods, this method does not require calibration. It is made more convenient by only using one laser wavelength and existing photoluminescence systems. The total measurement time of 2–3 min increases its practicability. Moreover, it is free of horizontal balancing current and the accuracy is experimentally verified.

We present an improved method that uses photoluminescence images to calculate the spatially-resolved efficiency in addition to other performance parameters of silicon solar cells. This new method is simpler than our previously-presented two-diode method, using only one diode with a variable ideality factor. Experimental results show that the simplified method is more tolerant of very large variations in local series resistance, a characteristic commonly seen in silicon cells. Using dark lock-in thermography techniques, we quantitatively verify the efficiency images produced by our improved method.

We present a method to separate bulk lifetime and surface recombination velocity using transient photoluminescence (PL). With the use of the analytical expression of transient PL, the asymptotic separation of two transient PL decays excited by different wavelengths can be obtained. Combined with the use of the intensity ratio of two steady-state PL intensities excited by the two different wavelengths, the lifetime separation can be achieved. In addition, detailed methods to experimentally determine the two terms are presented.

Metals are an integral part of solar cell structures—as a metal contact and/or a back reflector. However it is a less known fact that there are plasmonic losses associated with these structures and it is important to minimise such losses especially when placed in proximity to scattering media like textures on the front of the cell that can introduce an angular dependence to the light incident on the rear. This study investigates the losses in a metal reflector when placed adjacent to a dielectric layer in a typical solar cell rear geometry. The experimental measurement was realised using a novel custom built optical setup for characterising intensity of the reflected light at various internal incident angles at the Si–dielectric interface. Our results show that the thickness of the dielectric layer, the refractive index of the dielectric layer and the type of the rear metal can all affect the degree of such losses and the distribution of the angular reflection. Both measured and simulation results indicate that a 250–320 nm SiO2 along with an Ag back reflector gives the best internal rear angular reflection in Si but is dependant on the wavelength of interest. It also shows that textured or scattering front interfaces with internal incident angles greater than 28° have the potential to provide light trapping through total internal reflection and at the same time minimise plasmonic losses at the metal reflector interface.

Reflected light from textured front surfaces of a solar cell contains useful information about the surface geometry as well as the optical properties of the cell. In this study, the 2-D reflected light distributions from front surfaces of silicon cells textured in various ways are characterised by an appropriate optical setup. The results are compared to those of conventional ray tracing. For surfaces with regular inverted pyramids, we identified reflected light diffraction patterns that can be explained by the principles of geometrical optics. These reflected light patterns can be used to extract details of surface morphologies and hence be used as a tool to fine tune and monitor fabrication processes. Surfaces with random pyramids reflect light without diffraction patterns but with other distinctive features. Their reflections lie primarily in the angular range between 0° and 50°, with a maximum intensity at about 25°. The impact of the encapsulant's refractive index on the amount of total internal reflection is discussed for various types of textured surface. For both random and regularly textured surfaces, it appears possible to achieve higher total internal reflection ratio when the encapsulant's index increases from conventional 1.5 to 1.6.

Silicon solar cell technology is evolving rapidly with recent improvements in both multicrystalline and monocrystalline silicon wafers and ingots prepared by high throughput directional solidification, using both semiconductor grade and upgraded metallurgical grade silicon. Photoconductance and photoluminescence measurements are being rapidly developed to guide the development of these materials, with increasing attention being paid to combining steady-state and transient photoluminescence measurements to obtain self-consistent results. Analytical transient solutions are presented for both photoconductance and photoluminescence from silicon bricks or thick wafers which will guide this development by highlighting important interdependencies.

We present a method to image the following parameters spatially resolved of a solar cell: voltage Vxy, current density Jxy, power density Pxy, efficiency ηxy, series resistance Rs,xy, fill factor FFxy and dark saturation current densities of a two diode model J01,xy and J02,xy. The algorithm determines this set of self-consistent parameters by using a minimum of 5 electrical biased photoluminescence images. A comparison with CELLO measurements supports the presented method.

In this article, a novel rear structure using Ag nanoparticles to create surface plasmons to enhance light trapping is applied on the rear of planar high efficiency PERT (Passivated Emitter and Rear Totally Diffused) silicon wafer cells, targeting the spectrum range from 1000 nm to 1200 nm. Variations of this rear structure that combine Ag nanoparticles, dielectric layers and back metal reflectors were studied and analysed. Thickness of the rear surface passivation SiO 2 spacer layer was optimised to achieve maximum optical enhancement using surface plasmons but with minimum electronic losses due to recombination effects. The effect of the precursor evaporated Ag film thickness was also studied as a means to vary the size/shape of the nanoparticles. The measured external quantum efficiency ( EQE ) of the best performing rear reflector shows a maximum enhancement of more than 4-fold at 1160 nm. This corresponds to a 16% photocurrent increase (calculated from 900 nm to 1200 nm) compared to the cell with conventional Al rear reflector. Moreover, from the measured spectral response and optical absorption data, we successfully separated and analysed the electrical and optical properties of the novel rear light trapping designs. Light trapping features were quantified using optical parameters characterised by an effective optical path length factor Z , while electrical parameters such as surface recombination velocity S (cm/s) and effective minority charge carrier lifetime τ bulk (μs) were also extracted. Relative errors for these parameters were also calculated. For the cell with the best performing rear structure, we report a maximum Z factor enhancement of around 6-fold using Ag nanoparticles in conjunction with a detached Ag reflector, in comparison to the reference at 1200 nm.

The optimization work of high-efficient solar cells includes light-trapping schemes for the weakly absorbed near-infrared (NIR) photons. The light scattering properties of the back-reflector are here of particular interest. Novel back reflector designs that employ plasmonic and multi-stack interference effects require especially an experimental verification of the numerical simulations. Unfortunately, a good measurement of these scatter properties is nontrivial due to the refractive index of silicon and its resulting angle of total reflection. On planar wafers, light can therefore only be coupled in and out of silicon in a quite narrow angle, impossible to measure wider angles. Therefore we present an experimental setup to measure the angular resolved scatter properties of back-reflectors for IR photons.

Rear surface reflector of solar cell is designed to improve light collection capacity by allowing the low energy photons to go through multiple bounces inside the solar device before escaping. In this paper, we investigate the thickness effect of rear SiO 2 surface passivation layer on both optical and electrical properties of front-planar high efficiency PERT (Passivated Emitter and Rear Totally-Diffused) solar cells. Two kinds of metal reflectors are fabricated: the conventional planar reflectors by evaporated Al and the novel scattering reflectors by self assembled Ag nanoparticles. We find that the thickness dependence of rear SiO 2 layer (from 8 nm – 134 nm) on photocurrent shows an asymmetry for planar and scattering reflectors, moreover, the scattering reflectors perform much better than the planar reflectors under all tested SiO 2 thicknesses. A maximum current enhancement (calculated from wavelength 900 nm to 1200 nm) of 12.1% is presented for planar reflector with 134 nm SiO 2 film, and 18.4% for scattering reflector with the optimized 19 nm rear SiO 2 film. Additionally, by adding a detached metal mirror, the maximum current enhancement from scattering reflector jumps to 27.0%. Effective optical path length Z is calculated to study the light trapping (optical properties) under various SiO 2 thicknesses for both reflectors. Diffusion length L is calculated to track the electrical performance. It is shown that thicker SiO 2 is of benefit for both optical and electrical properties when planar Al reflector is used. However, for scattering reflectors, thinner SiO 2 is preferable for optical enhancement, but thicker SiO 2 is desirable for electrical gain. 19 nm SiO 2 is found to be the best choice for cells with scattering reflectors, considering both effects.

We present a method to image the following spatially resolved parameters of a solar cell: voltage V xy , current density J xy , power density P xy , efficiency η xy , series resistance R s,xy , fill factor FF xy and dark saturation current densities of a two diode model J 01,xy and J 02,xy . The algorithm determines this set of self- consistent parameters by using a minimum of five electrical biased photoluminescence images.

27th European Photovoltaic Solar Energy Conference and Exhibition; 436-438, The optimization work of high-efficient solar cells includes light-trapping schemes for the weakly absorbed near-infrared (NIR) photons. The light scattering properties of the back-reflector are here of particular interest. Novel back reflector designs that employ plasmonic and multi-stack interference effects require especially an experimental verification of the numerical simulations. Unfortunately, a good measurement of these scatter properties is nontrivial due to the refractive index of silicon and its resulting angle of total reflection. On planar wafers, light can therefore only be coupled in and out of silicon in a quite narrow angle, impossible to measure wider angles. Therefore we present an experimental setup to measure the angular resolved scatter properties of back-reflectors for IR photons. Therefore we present an experimental setup to measure the angular resolved scatter properties of back-reflectors for IR photons. It allows the free and independent selection of the illumination and detection angle at the back-reflector inside silicon, as well as its polarization effects.

The thesis presents two new imaging measurement methods to measure the effective series resistance and effective sheet resistance of a working solar cell during operation. The methods offer high spatial resolution and are as well contactless and non-destructive. Both methods use standard photoluminescence imaging setups for data acquisition and are therefore compatible with most experimental setups of this kind. Due to the high optical resolution of these systems and measurement speed, millions of data points can be measured within seconds. Novel to both presented methods is the previously unseen robustness to the influence of other electrical and material properties on the resulting data, which is usually a weak point of common optical imaging methods. Furthermore no calibration procedures or other information about the solar cell are required. This together allows measurement of solar cells with strongly changing material properties, such as multicrystalline solar cells, correctly and in high resolution. The importance and usefulness of the presented series resistance imaging algorithm are reflected in a conference poster award and the successful licensing of a patent. The presented algorithm for calculating the sheet resistance is furthermore the worlds first photoluminescence based sheet resistance imaging method. Both methods calculate the series and sheet resistance value for each camera pixel separately. Key for both methods is the usage of two different external operation conditions which result in exactly the same electrical potential for the pixel under investigation. The unchanged local potential guarantees that the influence of other electrical characteristics remains the same as they are driven by the local potential. By forming a differential the unwanted influence is eliminated. Both, the series and sheet resistance are therefore calculated in their differential form in combination with the terminal condition or their neighbour pixels.

Only recently, the quality of liquid phase crystallized silicon directly on glass substrates made a huge leap towards the quality of multi-crystalline wafers with open circuit voltages well above 600 mV. In this paper, we investigate the material quality in order to identify the factors limiting further performance improvements. We employ photoluminescence imaging on a state of the art test structure with lifetime calibration by transient photoluminescence. The resulting lifetime map is converted into an effective diffusion length map and the origin of regions with short lifetimes is investigated with electron backscattering and transmission electron microscopy. High local dislocation densities in areas with dissociated coincidence site lattice boundaries were found to be responsible for the localised quenching of the photoluminescence signal.

We present a method to simultaneously determine bulk and surface recombination properties using time-resolved photoluminescence (PL) decay. The lifetime separation algorithm makes use of the analytical expression of the asymptotic separation of two time-resolved PL decays corresponding to different excitation wavelengths as well as that of the ratio of two steady-state PL intensities excited by the two different wavelengths. Detailed experimental methods of measuring these two terms are presented and the effect of signal-to-noise ratio is discussed to determine the applicability of this algorithm.

This paper presents an accurate and practical mathematical model of time-resolved photoluminescence (PL) response from silicon wafers generated by fast repetitive excitation pulses. The model is valid under low level injection condition and takes into account the depth dependence of carrier generation, diffusion, and surface recombination. Finite element analysis is employed for the carrier density and PL computations. By comparing computational results with results obtained from PC1D (a computer program solving fully coupled nonlinear equations for quasi-one-dimensional carrier transportation in crystalline semiconductor devices, especially focusing on photovoltaic devices), the validity of this method is confirmed. Early stage application and the limitations of this method have been studied, and future work has been proposed.

A technique for fast and spatially resolved measurement of the effective series resistance of silicon solar cells from luminescence images is introduced. Without compromising the speed of existing luminescence based series resistance imaging methods, this method offers significant advantages in that it is more robust against variations in local diode characteristics. Lateral variations in the series resistance of an industrial screen printed multicrystalline silicon solar cell obtained from this method show excellent correlation with a Corescan measurement and are also shown to be unaffected by lateral variations in the diode properties.

Both dark lock-in thermography (DLIT) and photoluminescence (PL) imaging deliver information about the local electronic properties of solar cells. Special methods have been proposed for evaluating DLIT or PL images taken under various conditions with the goal to extract the local two-diode parameters. Knowing these parameters, the locally contributing efficiency data in a cell or local expectation values of the efficiency parameters Voc, FF, or η may be calculated. In this contribution these DLIT- and PL-based approaches are applied to one and the same cell and the results are compared. It is found that both methods lead to similar but not to identical results. The physical reasons for the differences between DLIT and PL based local efficiency analysis are discussed. It is found that the PL-based J02 results are still corrupted by the local series resistance, probably because the simple two-diode model is not sufficient to describe the PL process under electric load with sufficient accuracy. For DLIT this approach is less disturbing.

Photoluminescence (PL) imaging is a powerful inspection technique for research laboratories and production lines. It is used for a wide range of applications across the entire manufacturing chain from bricks and ingots to modules. However, common PL imaging systems have three main limitations: (a) Due to the uniform illumination, the acquired images are affected by lateral carrier flow, resulting in image blurring; (b) sample's nonuniformity is measured at different injection levels; and (c) images are taken at room temperatures, although there is valuable information in temperature-dependent measurements. In this paper we present a novel temperature-dependent PL imaging system that is not affected by lateral balancing currents. By adaptively adjusting the light intensity at each pixel, we set a uniform excess carrier density across the sample. Hence, the lateral currents are eliminated. The non-uniformity of the material's electrical properties and temperature characteristics can then be extracted from the excitation image. The advantages of the proposed system are demonstrated using mono and multicrystalline silicon wafers. This novel approach presents a significant improvement in accuracy and resolution compared to conventional PL imaging techniques and is therefore, expected to be beneficial for any PL-based quantitative analysis.