By Miguel Centeno Brito, University of Lisbon
It is widely accepted that solar power will play a central role in a post-carbon society. Solar radiation is abundant, clean and renewable, but for worldwide deployment, there is still a major challenge to overcome — cost. At present, a significant fraction of the final cost of solar cells is related to the substrate of the solar cell, a thin wafer of very pure crystalline silicon.
The standard process to obtain thin silicon wafers for use in solar cells consists of melting very pure silicon chunks in a quartz crucible and then letting it cool, very slowly. As the molten silicon solidifies, it grows the crystalline structure that is essential for high efficiency solar cells. In large production facilities, maintaining hundreds of kilograms of silicon at its melting temperature (1414°C) for hours or days can cost a lot of money and require a lot of energy. After cooling, the end product is a silicon ingot that can be sawn into thin slices, which then become the silicon wafers used as the substrate for solar cells. However, the thickness of the saw blade is similar to the thickness of the wafer; thus, approximately half of the costly material becomes sawdust that, typically, is discarded.
The silicon wafer industry has recognised the value of this waste for a long time. Since the 1980s, many different ideas have been put forward to avoid this waste. One such idea involved the direct growth of thin silicon substrates in the form of a ribbon, thus completely avoiding the requirement to produce an ingot. Many attempts to produce silicon ribbons met insurmountable technological difficulties, some during laboratory research and others during production. Only two ribbon methods have proven successful: the edge defined growth (EFG) silicon ribbon method, which uses a graphite mould to pull a silicon ribbon from a bath of molten silicon; and the string ribbon (SR) method, in which a silicon sheet hangs from two carbon-based vertical strings. From a technological point of view, both these methods have unwanted features in common: both require a crucible filled with molten silicon (and thus require a large energy input to keep the silicon molten during the pulling process) and the silicon material produced is contaminated by carbon (from either the mould or the strings).
Researchers at the University of Lisbon, Portugal, have been trying to develop new methods for growing silicon ribbons. One such approach is called the EZ-Ribbon technique, where EZ represents electric zone. This novel idea applies an electric current to a slab of silicon. If such a current is applied to a metal, the electric conductivity decreases with temperature and the current tends to spread throughout the entire piece of metal. However, for a semiconductor material such as silicon, the conductivity increases with temperature; thus, when current is applied to a slab of silicon, the current concentrates in the hottest region, further heating that region. This creates a runaway current-concentrating process that can be used to form a very thin line of molten silicon from an otherwise relatively cool sample. The EZ process for producing a molten line of silicon is very energy efficient since all of the required heat is delivered, via the electric current, at exactly the right location. Furthermore, the melting silicon is surrounded by silicon only and is never exposed to other materials, the result being a silicon ribbon that is virtually contamination free.
Once the EZ concept had been successfully demonstrated, a method to pull the ribbon from the molten line area. To accomplish that, small pellets of pure silicon are melted in a small silicon lake, which is connected to a downstream liquid-silicon river created by an electric current. By pulling on the top surface of the river, a thin liquid-silicon ribbon is formed by capillary forces. This low energy consumption method produces a contamination-free silicon ribbon that can then be used as a substrate for solar cells. The method is low cost because it requires relatively low amounts of energy and high quality because the product is not contaminated by other materials.
Of course, it is one thing is to develop a new concept for growing crystalline silicon ribbons, another to demonstrate that it can be done in the laboratory, and a final and important thing to make it work in a large-scale manufacturing facility. This EZ-Ribbon is presently at the final step as work is underway to make the EZ-Ribbon process robust and scalable. However, industrial success is not guaranteed as shown by commercial experiences with EFG and SR methods. In spite of overcoming technological shortcomings, and after many years of development, both EFG and SR silicon products entered the booming solar market in the early 2000s. However, these ribbons were longer and narrower than standard commercially available wafers and they required different processing equipment; as a result, these new products did not meet the industrial needs of the market. Both of these product development tales ended the same way — with the start-up companies closing their factory doors. Let us hope there is a happier end to the EZ-Ribbon tale.
Miguel Centeno Brito
University of Lisbon
www.atomiumculture.eu
Hay 0 Comentarios