The 25th International Conference on Amorphous and Nano-crystalline Semiconductors
August 18–23, 2013 Toronto, Ontario Canada
a-Si Related Photovoltaics II
Chair: Mehmet Güneş, Mugla Sitki Kocman University
Mo-C2.1 (invited) 16:00–16:30
Charge Carrier Transport in Amorphous and Microcrystalline Silicon Based Materials
1. Forschungszentrum Jülich GmbH, Institute of Energy and Climate Research 5 -Photovoltaics, D-52425 Jülich, Germany
2. School of Engineering, Physics and Mathematics, University of Dundee, Dundee DD1 4HN, Scotland
The electronic transport in amorphous and microcrystalline silicon based materials have many aspects in common, but exhibits also distinct differences. Though in both inhomogeneous material systems trap controlled transport of excess carriers is typically observed which is indicated by e.g. a temperature activated mobility, the Hall effect shows a double sign reversal for the amorphous phase but for the microcrystalline phase the normal Hall coefficient is observed. As microcrystalline materials are typically phase mixtures of amorphous tissue, crystallites and voids with varying volume fractions and distributions, the relation between microstructure, density of states and fundamental transport parameters is difficult to extract without the investigation of well characterized series of samples.
We will present the results of electronic transport measurements of well characterized series of amorphous and microcrystalline silicon samples with controlled crystalline volume fraction, defect density and Fermi level position. Particularly the influence of these parameters on the carrier mobility will be discussed. Furthermore the applicability of the concept of the effective density of states will be addressed. Finally, recent results on the transport in microcrystalline silicon carbide where the polytypism can add an additional inhomogeneity will be presented.
Keywords: amorphous and microcrystalline silicon, microcrystalline silicon carbide, electronic transport, effective density of states
Lightweight Amorphous Silicon Photovoltaic Modules on Flexible Plastic Substrate
1. Electrical and Computer Engineering Department, University of Waterloo, Waterloo, Ontario N2L 3G1, Canada
2. TF TE Ioffe R&D Center, St-Petersburg, 194064, Russia
3. National Institute for Astrophysics, Optics and Electronics, Puebla, 72840, Mexico
4. Ioffe Physico-Technical Institute, St-Petersburg, 194021, Russia
Solar cells on lightweight and flexible substrates have advantages over the glass- or wafer-based photovoltaic devices in both terrestrial and space applications. Here, we report on development of amorphous silicon thin film photovoltaic modules fabricated at maximum process temperature of 150°C on 100 mm thick PEN plastic foils. Each module of 10 cm x 10 cm area consists of 72 a-Si:H n-i-p rectangular structures with ZnO top electrode and Al back electrode deposited through the shadow masks. Individual structures are connected in series forming eight rows with connection ports provided for external shunt diodes. The module yields about 132 W/kg with open circuit voltage of 51.2 V. Cell structure optimization and fabrication process details will be discussed.
Keywords: amorphous silicon, photovoltaics, flexible substrate
Variation of the Defect Density in the Absorber Layer of a-Si:H and μc-Si:H Solar Cells over Two Orders of Magnitude: Influence on Solar Cell Performance
1. IEK-5 Photovoltaik, Forschungszentrum Jülich GmbH, 52425 Jülich, Germany
2. National Science Center - Kharkov Institute of Physics & Technology, Ukraine
To investigate the limiting role of electronic defects in the absorber layer of amorphous and microcrystalline (a-Si:H, μc-Si:H) solar cells we have been able to vary the defect density in the i-layer of solar cells and in corresponding ESR samples by exposing both sample groups simultaneously to 2 MeV electron bombardment and successive stepwise annealing. Cell parameters were analyzed then as a function of NS measured in the ESR samples. Cells with a-Si:H and μc-Si:H absorber layers of different thickness illuminated through p- and n-layers (p-i-n and n-i-p configurations) were studied. Comparison of p-and n-side illuminated solar cells is performed to detect possible asymmetry in the collection of electrons and holes at increased NS. Variation of the defect density over 2 orders of magnitude provided sufficient dynamic range for observation of clear trends and further verification of device operation models. Experimental results of the dependence of the diode I-V characteristics and the quantum efficiency on the defect density were accompanied by device modeling using the ASA simulation program.
In a-Si:H cells the open circuit voltage (VOC) and short circuit current (JSC) decrease only moderately up to NS of 5x1016 cm–3 but drop steeply at higher NS. On the other hand the fill factor (FF) showed continuous decrease with an increase in NS. Generally a stronger degradation was observed in cells with thicker a-Si:H absorber and in cells illuminated through n-layer. However at NS < 1017 cm–3 both p-side and n-side illuminated cells showed very similar performance.
In μc-Si:H cells, a much more continuous decrease of both VOC and JSC is observed already at low NS values. The FF continuously decreased as NS was increasing up to 1017 cm–3 and shows considerable scatter at NS > 1017 cm–3. At NS above 1016 cm–3, I-V parameters, especially JSC, of n-side illuminated μc-Si:H solar cells were much more sensitive to an NS increase than parameters of the p-side illuminated device. The difference was remarkably stronger compared to the difference between p- and n-side illuminated a-Si:H cells. Comparison of QE of μc-Si:H cells illuminated from p- and n-side showed that while electrons are easily collected even at the highest achieved NS = 1018 cm–3, hole collection is strongly suppressed by NS increase. In the simulation with the ASA program this asymmetry in carrier collection could be reproduced by placing the defect density peak 0.2 eV above the midgap of μc-Si:H i-layer.
We conclude that the bulk defect density in a-Si:H cells is less critical for state of the art material and improvement at lower NS could be achieved mostly in FF. It suggests, for instance, that interfaces or doped layers may limit VOC, rather than defects in i-layer. For μc-Si:H absorber, contrary to a-Si:H, bulk NS in the range of 1015–1016 cm–3 (state of the art material) seems to still limit the solar cell performance. Therefore here we speculatively expect improvements in cell performance given NS of μc-Si:H absorber is eventually reduced.
Keywords: amorphous silicon, microcrystalline silicon, solar cell, defects
Influence of Post-deposition Annealing, B Grading and Ion Bombardment on Stability of a-Si Solar Cells
Iowa State University, Dept. of Electrical and Computer Engineering, Ames, Iowa 50011
Previous reports indicated that post-deposition annealing resulted in significantly fewer defects being formed in a-Si films as a result of light-induced degradation than was the case for standard films. In this study, we report on systematic investigation of device I-V curves, defect densities, hole collection properties, and subgap quantum efficiency of solar cells which were annealed at higher temperatures after deposition. Defect densities were measured using both capacitance-frequency measurements, which allow measurement of defect energy levels, and capacitance-voltage measurements at low frequency (1 Hz) and high temperature (150°C) which reveal the total defect density in the material. Hole collection properties, including hole mobility-lifetime products, were measured using quantum efficiency vs. bias techniques. We do not find any significant difference in stability or deep defect generation in post-deposition annealed cells when compared with standard cells. We next investigate alternative techniques for improving stability, namely deposition at higher temperatures combined with enhanced electric field generation using controlled B doping of the intrinsic layer in a-Si p-i-n cells. We find that ppm levels of B grading in the intrinsic layer improves the performance of cells prepared at higher temperatures. When significant ion bombardment is present, as happens at lower deposition pressures, the stability improves significantly. Measurement of defect densities after degradation confirm that the new technique, namely deposition at higher temperatures (400°C) and low pressures, combined with ppm levels of B grading in the intrinsic layer, lead to much fewer defect states being generated when the cells are exposed to light as compared to the case for standard cells. The cell efficiency drop is about 10% and saturates within ~20–30 hours. In contrast, the efficiency drop in standard cells is >20% and does not saturate up to 100 hours of exposure. The measurement of quantum efficiency under voltage bias confirms that the new technique yields significantly better hole collection than standard cells after degradation.
Keywords: amorphous Si, stability, photovoltaic devices
High Efficiency Amorphous Silicon Based Solar Cells: Towards an Objective Comparison of Various Absorber Materials
Ecole Polytechnique Fédérale de Lausanne (EPFL), Institute of Microengineering (IMT), Photovoltaics and Thin Film Electronics Laboratory, Rue Breguet 2, CH-2000 Neuchâtel, Switzerland
Major applications of amorphous silicon (a-Si:H) are in single junction, micromorph tandem, or triple junction solar cells. Different institutes have claimed over the past decades different materials (e.g. protocrystalline, polymorphous, low pressure a-Si:H) to result in most efficient solar cells that suffer least under light induced degradation, regardless of the application in single or multi-junction solar cells, or in other words, regardless of the current density required by the a-Si:H junction.
For the present study, various kinds of intrinsic a-Si:H materials deposited by plasma-enhanced chemical vapor deposition (PECVD) were characterized and integrated in single junction solar cells in superstrate configuration. By varying process pressure between 0.2 and 9 mbar, excitation frequency, hydrogen dilution, and substrate temperature, a variety of materials covering polymorphous, protocrystalline, high pressure a-Si:H (radio frequency) and industrially used deposition conditions (40 MHz) were synthetized in the same reactor, an Octopus I from Indeotec. For best bulk material quality, the lowest power density that allowed plasma ignition was chosen for each pressure.
All cells were deposited on ZnO front contacts with four different roughnesses, grown by low-pressure chemical vapor deposition (LPCVD) and subsequently treated by argon plasma to smoothen the surface. The cell design was kept constant: only the 220 nm thick intrinsic bulk layer, sandwiched between intrinsic buffer layers at p-i and i-n interfaces, was varied. This allows a direct comparison of layer bulk properties resulting from the investigated deposition conditions. Special attention was given to the photoelectric cell properties after light soaking and their correlation to layer properties measured by ellipsometry, Fourier Transform Infrared Spectroscopy (FTIR), Raman spectroscopy, and Photothermal Deflection Spectroscopy (PDS).
Suited process conditions could be defined for all different absorber materials. Even though the doped layers were kept identical for all devices, and hence not optimized for high efficiency, the best stabilized cells reach efficiencies of 8.7% without antireflective coating and fill factors above 68% for low hydrogen dilutions. For each processing pressure, cells have been obtained with relative light induced degradation well below 15%, in some cases even below 8%. The different materials and solar cells obtained will be detailed and compared for potential applications in single junction, micromorph tandem and triple-junction solar cells. For the latter case, high band-gap material near the transition to microcrystalline material will be shown to be optimum, with for instance open-circuit voltages above 940 mV (stabilized). For single and tandem cell application, lower bandgap a-Si:H material will be shown to be preferred, while limitations to reduce the bandgap and to maintain efficient solar cells will be explained. A further constraint in the choice of absorber materials is coming from substrate roughness: We show that namely the open-circuit voltage dependence on the substrate roughness increases dramatically for higher hydrogen dilutions. While this is less critical for triple junction solar cells with smoother substrates, it underlines the importance of only slightly hydrogen diluted absorbers with low band-gap for micromorph tandem devices on rougher substrates.
Keywords: amorphous silicon solar cells, thin film silicon solar cells, a-Si:H, light induced degradation, Staebler-Wronski effect
Mo-C2.6 (invited) 17:50–18:20
Towards High-Efficiency Polycrystalline Si Thin Film Solar Cells on Glass: Tailoring 3-dimensional Architectures
1. Helmholtz Zentrum Berlin für Materialien und Energie, Institute Silicon Photovoltaics, Kekulestraße 5, 12489, Berlin, Germany
2. Helmholtz Zentrum Berlin für Materialien und Energie, Si-Nanoarchitectures for Photovoltaics and Photonics, Kekulestraße 5, 12489, Berlin, Germany
3. Helmholtz Zentrum Berlin für Materialien und Energie, Institute for Nanometre Optics and Technology, Albert-Einstein-Str. 15, 12489 Berlin, Germany
4. SCHOTT AG, Hattenbergstraße 10, 55122, Mainz, Germany
5. Konrad-Zuse-Zentrum für Informationstechnik Berlin, Takustraße 7, 14195 Berlin, Germany
The development of highly efficient polycrystalline Si thin film solar cells necessitates the low-cost preparation of silicon with excellent electronic quality, which is embedded in effective light harvesting architectures. This contribution addresses our latest insights into the structural design and electronic properties of crystalline Si thin film solar cells. We will subsequently present a fabrication strategy for high-quality 3-dimensional periodic Si solar cell architectures on nano-imprinted glass, by correlating the electronic quality of polycrystalline Si thin films with their corresponding optical and structural characteristics.
The electronic quality of polycrystalline silicon is compromised by defects located inside crystal grains and at grain boundaries. By tailoring crystalline Si layers with systematically varied grain sizes ranging from 200 nm to 100 μm in a solid phase or liquid phase crystallization process, we identify the respective signatures of the intra-grain and grain boundary defects with quantitative electron paramagnetic resonance spectroscopy, and determine their influence on the solar cell performance. While solar cells based on fine-grained Si with a high defect concentration Ns of 1017 cm–3 have an open circuit voltage Voc of merely 300 mV, record Voc above 580 mV could be achieved with Si solar cells with large grains and a Ns of 1015 cm–3.
The development of advanced light harvesting architectures demands the deposition of Si on substrates with tailored structures. However, the growth mechanism of the deposited Si on these structured surfaces determines the material quality of the bulk Si. On the basis of nano-imprint lithography and the emerging high-rate deposition technique electron-beam evaporation of Si, we developed a low-cost and easily scalable fabrication process for periodic arrays of Si crystals, allowing for the precise control of the feature size and shape of the structure. By complementing electron paramagnetic resonance measurements with an optical analysis and 3-dimensional finite element method simulations, we investigate the design of such periodic 3-dimensional crystalline Si architectures on various nano-imprinted glass substrates. The complementary studies allow us to develop a solar cell concept consisting of periodic microhole arrays with high electronic quality and excellent optical broadband absorption.
Keywords: thin film solar cells, polycrystalline Si, nanostructures, defects, nano-imprint lithography