Elimination of Efficiency Degradation Mechanisms in Silicon Photovoltaic Solar Cells

Photovoltaic Solar Cells seem certain to make a significant contribution to the world's energy needs in the 21st Century. At the present time such devices only convert a small part of the 1kW per square meter of radiation from the sun which falls on the earth's surface into electricity. Typical commercial silicon devices that use un-concentrated sunlight are less than 15% efficient although laboratory devices with efficiencies as high as 24% have been made. Many very interesting ideas have been put forward for higher efficiency 3rd generation cells but most of these are a very long way from commercial realization on a large economic scale. At the present time 90% of production is based on wafered silicon (Czochralski single crystal or cast poly-crystalline). The predicted production for 2010 using existing silicon technology will have a peak output of 26GW. Unfortunately in their initial few hours of operation most silicon solar cells suffer from degradation which stabilizes after a reduction in efficiency of 10% relative ie a 20% cell becomes an 18% cell. In the context of total photovoltaic power lost on a world scale this is very significant. It is believed that the reduction in efficiency is due to a defect reaction in the silicon in which oxygen dimers diffuse to the boron dopant to form a powerful recombination centre. The problem is not restricted to the commercial devices of the next decade but is also highly relevant to some of the most promising third generation (high efficiency) cells which use silicon as part of the active structure. This research proposal aims to eliminate the degradation process by removing the reaction path from the silicon prior to the formation of the recombination centre. An essential pre-requisite to this is to achieve a detailed understanding of the defect centres and their formation.We have previously studied the oxygen dimers which are currently thought to be the precursors of the recombination centre. These can be detected using optical absorption measurements, ideally at low temperatures (~10K). Preliminary work indicates that it should be possible to develop treatments of the silicon material which reduce the concentration of the dimers to a negligible level in the finished cell and, because the dimers do not form at normal operating temperatures, so eliminate the formation of the defect. It would be quite feasible using this approach to maintain the concentration of interstitial oxygen which provides mechanical strength to the silicon with consequent yield and cost benefits. In this work the recombination centre will be studied using minority carrier Laplace Deep Level Spectroscopy and its structure determined by the application of uniaxial stress. The reaction of the oxygen dimer will be studied as the recombination centre forms, in real time, using optical absorption techniques. The work will be done in collaboration with MEMC who are one of the leading manufactures of solar silicon, the Institut fr Solarenergieforschung Hameln/Emmerthal (ISFH) in Germany who have undertaken much experimental work recently on solar cell degradation and the University of Aveiro in Portugal who will collaborate on theoretical calculations to support the Manchester work.

If the project is successful there will be an important industrial and social benefit in terms of more efficient power generation and consequent greater uptake of renewable energy. As a result of this work the absolute gain in efficiency of wafered silicon cells is anticipated to be 2%. This is very significant in the photovoltaic field (typically 10% improvement in relative terms). This type of cell constitutes more than 90% of the photovoltaic market at the present time and is expected to dominate sales for at least the next ten years despite other materials taking some of the market. This is an inevitable result of massive investment in silicon technologies for solar and recent improvement in cost effectiveness (thinner slices, cheaper poly silicon and new cell designs) If the effect of this is factored into the estimated output of the silicon solar cells susceptible to this form of degradation manufactured over the next 15 years it is equivalent to an additional 12GWp generating capacity without building any additional manufacturing capacity or generating any additional greenhouse gases. The equivalent value in today's costs of this is over twenty billion US dollars. In addition, if successful, the project outcomes are almost certain to be applicable to silicon based third generation cells which in the long term are likely to replace existing technologies. Commercial beneficiaries will be the UK and European companies to which the development is licensed who, in consequence, will have an improved competitive advantage. These are both cell manufacturers and solar materials manufactures. The general area of oxygen reactions in silicon is also one of very considerable interest to the microelectronics community as well as those specializing in photo-voltaics. The properties and structure of the recombination centres, the diffusion of the oxygen dimer are issues which are highly relevant to many semiconductor devices. Specific beneficiaries beyond solar are: a) Integrated circuits and b) Silicon power devices. Two of the world's major research establishments in Solar have provided letters of support for this proposal. The Fraunhofer-Institut fr Solare Energiesysteme ISE in Freiburg and the National Renewable Energy Laboratory, Golden, Colorado.