Hartung-Gorre Verlag

Inh.: Dr. Renate Gorre

D-78465 Konstanz

Fon: +49 (0)7533 97227

Fax: +49 (0)7533 97228

www.hartung-gorre.de

S

3866281579

Sven Seren

Low Cost Solar Cells

from Fast Grown
Silicon Ribbon Materials

Konstanz 2007, 144 pages/Seiten; € 64,00.
ISBN 978-3-86628-157-8

Abstract:

Within this thesis two fast grown silicon ribbon materials are analysed regarding their utilisation in photovoltaics: RGS (Ribbon Growth on Substrate) and Molded Wafer (MW). Both materials are close to commercialisation. Cost effective screen-printing based solar cell processes are developed meeting the specific material characteristics. The interstitial oxygen content plays a major role here.

Further on, the new spatially resolved measurement technique iLIT (illuminated Lock-In Thermography) was developed, which allows the contactless imaging of heat dissipating loss mechanisms in pn-structures and solar cells.

Keywords: solar cell, silicon ribbon, Ribbon Growth on Substrate, RGS, Molded Wafer, substrate, hydrogenation, oxygen, drift cell, screen-printing, lock-in thermography

Cover Picture: Photography recorded with an infrared camera system of a screen-printed 5x5 cm² RGS (Ribbon Growth on Substrate) solar cell with an open rear side metallisation held by hand.

Introduction

This thesis was written in the winter of 2006 / 2007, which was the warmest winter on record [1]. This phenomenon is not limited to Germany as the global climate shows the same trend. The expert group of the United Nations for climate IPCC1 published within its topical report [2] a forecast for the year 2100. Therein the most probable scenario anticipates a global warming between 1.7 and 4°C. Should the global warming constitute above three degrees, the mainland ice of Greenland would melt completely, resulting most probably in disastrous consequences for the coastal areas of the earth.

A correlation between the global warming and the increase of greenhouse gas concentrations within the earth’s atmosphere cannot be neglected any longer, this holds in particular for carbondioxide and methane. Todays CO2 concentration in the atmosphere is the highest it has been in the past 650.000 years. Solely the instantaneous reduction of greenhous gases can retard this progression, which cannot be stopped anymore [2].

How to accomplish this, in particular when considering the emerging markets? Politics on a global level can be factored out because from this side no effective decisions are made even on a local level. This holds in particular for the mainemitting nations. Because a reduction of the global energy consumption will not take place, a solution can be found only in a low-emission energy production, as a CO2 storage solution, for instance in the interior of the earth, is not available nor would it be sustainable. Uranium as a base for nuclear fission would wear out rapidly, if a reasonable fraction of the energy production should be shifted from fossil to nuclear energy sources. Novel techniques, such as nuclear fusion could not be utilised up to now despite the enormous effort made so far (JET2, ITER3). Instead of rebuilding the fusion reactor "sun" on earth, the energy provided by the existing sun could be used more effectively by means of light and thermal radiation, by wind and water power, biomass as well as geothermic power.

This work is attributed to the first attempt, the usage of the electromagnetic radiation of the sun by photovoltaic conversion into electric energy. For a large scale energy production solar cells produced from crystalline silicon are the dominating technique at the moment. For their production quartz sand is reduced to silicon, which has to be cleaned in the gas phase and is subsequently deposited as high-purity silicon. For the production of multicrystalline silicon the material obtained with this technique is crystallised to huge silicon blocks using an ingot casting process. The ingots were then cut down to smaller bricks, from which silicon wafers are wire-cut, the base material for solar cells. Since the diameter of the wire used for the wire sawing process approximately equals the wafer thickness, a silicon loss of roughly 50% occurs. Due to segregation- as well as contamination-based processes not the whole ingot can be wire-sawed to wafers. Areas of the ingot being in direct contact with the crucible during solidification, as well as upper and lower parts of the ingot cannot be used, which enhance the fraction of wasted silicon further above 50%.

A significant enhancement of the solar fraction of the produced global energy amount can only be reached by increasing the competitiveness, i.e. by a price reduction. Therefore, cost reduction within the production chain of solar systems has to proceed. A cost reduction as a result of mass production, however, is currently limited by a shortage of available and sufficiently pure silicon. The photovoltaic industry expanded over the last years with annual growth rates of 30-40%, whereas a broadening of production capacities for silicon dropped far behind. Until this bottleneck is overcome and for competitive photovoltaics also beyond it, the only way is to save silicon, best by using a fast producing and thus cost effective technique. This leads directly to the content of this work, which describes the characterisation and the solar cell processing of silicon wafers, produced very fast and directly from the silicon melt, i.e. without the indirection of block casting and the silicon loss linked to it.

The first part of this work presents the silicon ribbon material RGS (Ribbon Growth on Substrate). Crystallographic investigations as well as the analysis of material characteristics define the potential of the material, which is still in the R&D phase. In particular, attention is laid on the interstitial oxygen content due to its influence on the hydrogen diffusivity which directly affects the potential for material quality improvement.

For the development of a suitable solar cell process, adapted to the material quality, basic experiments are performed concerning the mechanical planarisation of uneven wafer surfaces, the reduction of cracks induced during planarisation as well as the chemical removal of defect-rich surface layers. The analysis of particular processing steps leads to a solar cell process, which avoids local shunts.

A spatially varying dopant concentration in the wafer can be used to enhance short circuit current densities of solar cells as a result of a drift-field. Wafer and cell based experiments are performed to investigate the assumed depth dependent doping concentration due to segregation and the RGS-specific crystallisation conditions. Doping with phosphorus, however, leads to n-type wafers which are characterised and processed to estimate the potential of this material.

To enhance cell efficiencies, different surface textures are investigated for this material. Further on, it is tested if scaling effects for the processing of larger RGS solar cells occur and which impact the reduction of the wafer thickness has on cell parameters and the silicon usage per output power.

Another silicon ribbon material, MW (Molded Wafer), will be presented in the second part of the work. This material is still in the R&D phase as well. As a result of the production process and combined with an annealing step at high temperatures, the comparably thick MW wafers show a broad oxygen denuded zone located in the upper wafer fraction. This wafer fraction represents the photovoltaically active zone in a solar cell process adapted to the material characteristics. The influence of the annealing step on the material quality, in particular the annealing temperature, is clarified in terms of solar cell parameters and advanced cell analysis.

The last part of this work addresses the Lock-In Thermography, a measurement technique, which allows the imaging of Joule losses in solar cells already after a very short measurement time. This is of high interest particularly for the silicon ribbon materials presented within this work due to a typical inhomogeneous lateral distribution of the material quality.

The Lock-In calculation significantly enhances besides the lateral also the thermal resolution of the measurement setup, which was built up during this work. This enables the resolution of typical temperature differences produced by shunts in the μK-range. The conventional Lock-In Thermography is advanced by a new measurement technique, the illuminated Lock-In Thermography (iLIT), which for the first time allows the contactless measurement of arbitrary pn-structures. This enables a monitoring of single solar cell processing steps without contamination.

Please pay attention to these books too:

Christopher Hebling

Die kristalline Silicium-Dünnschichtsolarzelle
auf isolierenden Substraten.

1999; VIII, 174 Seiten, € 65,45. ISBN 3-89649-455-4

Alexander Hauser

Die kristalline Siliziumsolarzelle.
Untersuchung der Einzelprozesse
und Entwicklung von Alternativen

Erste Auflage 2006; 132 Seiten; € 64,00. ISBN 3-86628-076-9

Sven Seren

Low Cost Solar Cells

from Fast Grown Silicon Ribbon Materials.

2007, 144 pages/Seiten; € 64,00. ISBN-10: 3-86628-157-9

ISBN-13: 978-3-86628-157-8