Martin Sever

Effect of substrate morphology slope distributions on light scattering, nc-Si:H film growth, and solar cell performance.

Thin-film silicon solar cells are often deposited on textured ZnO substrates. The solar-cell performance is strongly correlated to the substrate morphology, as this morphology determines light scattering, defective-region formation, and crystalline growth of hydrogenated nanocrystalline silicon (nc-Si:H). Our objective is to gain deeper insight in these correlations using the slope distribution, rms roughness (σ(rms)) and correlation length (lc) of textured substrates. A wide range of surface morphologies was obtained by Ar plasma treatment and wet etching of textured and flat-as-deposited ZnO substrates. The σ(rms), lc and slope distribution were deduced from AFM scans. Especially, the slope distribution of substrates was represented in an efficient way that light scattering and film growth direction can be more directly estimated at the same time. We observed that besides a high σ(rms), a high slope angle is beneficial to obtain high haze and scattering of light at larger angles, resulting in higher short-circuit current density of nc-Si:H solar cells. However, a high slope angle can also promote the creation of defective regions in nc-Si:H films grown on the substrate. It is also found that the crystalline fraction of nc-Si:H solar cells has a stronger correlation with the slope distributions than with σ(rms) of substrates. In this study, we successfully correlate all these observations with the solar-cell performance by using the slope distribution of substrates.

Design of periodic nano- and macro-scale textures for high-performance thin-film multi-junction solar cells

Surface textures in thin-film silicon multi-junction solar cells play an important role in gaining the photocurrent of the devices. In this paper, a design of the textures is carried out for the case of amorphous silicon/micro-crystalline silicon (a-Si:H/mu c-Si:H) solar cells, employing advanced modelling to determine the textures for defect-free silicon layer growth and to increase the photocurrent. A model of non-conformal layer growth and a hybrid optical modelling approach are used to perform realistic 3D simulations of the structures. The hybrid optical modelling includes rigorous modelling based on the finite element method and geometrical optics models. This enables us to examine the surface texture scaling from nano- to macro-sized (several tens or hundreds of micrometers) texturisation features. First, selected random and periodic nanotextures are examined with respect to critical positions of defect-region formation in Si layers. We show that despite careful selection of a well-suited semi-ellipsoidal periodic texture for defect-free layer growth, defective regions in Si layers of a-Si: H/mu c-Si: H cell cannot be avoided if the lateral and vertical dimensions of the nano features are optimised only for high gain in photocurrent. Macro features are favourable for defect-free layer growth, but do not render the photocurrent gains as achieved with light-scattering properties of the optimised nanotextures. Simulation results show that from the optical point of view the semi-ellipsoidal periodic nanotextures with lateral features smaller than 0.4 mu m and vertical peak-to-peak heights around or above 0.3 mu m are required to achieve a gain in short-circuit current of the top cell with respect to the state-of-the-art random texture (>16% increase), whereas lateral dimensions around 0.8 mu m and heights around 0.6 mu m lead to a > 6% gain in short-circuit current of the bottom cell.

The Two Approaches of Surface-Texture Optimization in Thin-Film Silicon Solar Cells

Textured interfaces in thin-film silicon solar cells can increase the photocurrent of the devices. However, the challenge is how to design textures (periodic or random) that could outperform the state-of-the-art random ones. Moreover, the textures should enable defect-less semiconductor layer growth, which has often been neglected in the past, resulting in the low open-circuit voltage and fill factor of the devices. In this paper, we present two approaches for the systematic optimization of surface-textures: bottom-up and top-down. Fully 3-D optical simulations, including calibrated nonconformal layer growth model were employed to predict gains in short-circuit current densities in a micromorph solar cell (bottom-up approach) and a single-junction amorphous silicon solar cell (top-down approach). In the bottom-up approach, we start with a simple sinusoidal component and change its shape systematically in the direction of broader valleys, also resulting in better conditions for the layer growth. In the top-down approach, we start from a random texture (morphology fingerprint taken from Asahi U type substrate) and modify it in a spatial frequency domain. We show the role of the presence/absence of different frequency regions as well as the important role of the phase spectrum on the optical characteristics of the device. In both approaches, improved textures have the potential to outperform state-of-the-art random ones, not only from optical point of view but in terms of conversion efficiency as well.

Optimization of advanced surface-textures for thin-film silicon solar cells

Optimization and development of (i) periodic U-like single-texture and (ii) periodic+random double-texture are carried out for micromorph silicon solar cell. Based on modeling supported by experiments more than 10% improvement in conversion efficiency is indicated.

Design challenges for light harvesting in photovoltaic devices

Device modelling and characterization are indispensable tools in the design of photovoltaic devices. In the contribution we present two challenging issues related to accurate modelling and efficient characterization of light scattering at nanotextured interfaces or other nanophotonic structures used in solar cell technologies. The model based on finite element method, which is upgraded with the Huygens’ expansion theorem is presented. It enables to calculate the angular distribution function of scattered light in the near and far field. It accounts also for the antireflection effects originating from nanoroughnesses. To characterize scattered light efficiently a camera based angular resolved spectroscopy system is presented. It captures the spatial angular distribution function in broad angular range at one shot.

Prediction and prevention of defective regions within thin-film silicon solar cells

Previously developed growth model is used to simulate occurrence of defective regions within thin-film silicon solar cells. Such procedure enables expansion of optical optimization with prediction and prevention of defective regions, resulting in optimized textures generating high short-circuit current density (JSC) while maintaining good electrical properties of the cell. The approach is applied on sinusoidal and semi-circular texture. Best predicted case for the analysed double junction thin-film silicon solar cell is the semi-circular texture with period of 1800 nm and height of 900 nm, where improvement in JSC of 2 % and 85 % is expected for top and bottom cell, respectively.

Advanced approaches in optical simulations of thin-film solar cells

In rigorous optical modeling and simulation of thin-film solar cells a few constraints and bottlenecks have been addressed recently. Two of them are related to how to include thick incoherent layers in the coherent finite element based simulations and how to describe and include non-conformal growth of layers in a thin-film solar cell. In this paper we present and apply three different solutions to speed up and improve the simulations of realistic thin-film solar cell in superstrate configuration: (i) solutions for including optically thick incoherent glass layer in superstrate type of solar cells and (ii) how to consider its thickness in millimeter range and and (iii) we present a non-conformal layer growth model. By applying these advanced features to 3-D optical modeling, it becomes possible to simulate the complete solar cell structures (incl. front glass) rigorously and time efficiently and to consider the realistic textures at all internal interfaces thus gaining in accuracy of simulations.