"The blueprint for the production process is ready," says Wim Soppe of ECN Solar Energy. "Including the financial calculations. We are aiming at a cost per watt-peak below 70 cents - less than half the price of crystalline modules."
The new solar cell is based on a thin film instead of silicon wafers. 'Thin' is putting it mildly; a sheet of 80 grams paper is about 50 times thicker than the layers of silicon which make up the active part! Compared with ordinary crystalline cells, manufacturing the thin film type requires 100 times less silicon. Moreover, the flexible thin film can be produced using the very efficient method now being developed by ECN Solar Energy: a strip of stainless steel foil rolled off a large coil passes through a series of process chambers. There the insulating, conducting and active layers are deposited, and the cells are separated and connected in series using depth selective laser scribing and screen printing. Finally the nearly finished solar modules are wound on a second coil.
Why steel foil instead of plastic? Soppe: "Steel is attractive because it is strong and non-stretch. Also, it easily tolerates the 200 °C we need to deposit the active layers. Plastic requires somewhat lower temperatures, less suitable for silicon."
But a transparent plastic foil would allow a window to be used as a solar panel. Isn't that an advantage? "Allowing roughly half the light to pass through your solar panel means you need twice the surface area for the same amount of electrical power," says Soppe. "So that would double the price of your solar energy. Hardly the thing to do, because reducing that price has absolute top priority! We are working with Corus on various applications for thin film cells, such as roof and frontage elements."
Amorphous and microcrystalline
In monocrystalline silicon, each atom is connected to four neighbours in a single large-scale lattice. This material is crystallised at a high temperature (1400 °C). Turning it into solar cells also requires plenty of heat (up to 900 °C). By keeping the process for manufacturing thin film cells below 200 °C, costs are significantly reduced. The silicon then becomes amorphous or microcrystalline. On the nanometer-scale, there is no difference in the arrangement of atoms in amorphous and monocrystalline silicon, but the former isn't as regular at longer ranges. This causes free, 'dangling' bonds to appear, which reduce the electronic quality and thereby the efficiency of solar cells based on this material. Dangling bonds can be neutralised to a large extend by adding hydrogen.
However, amorphous silicon does have an important advantage. In layers of the same thickness it absorbs much more light than the crystalline variety - about 100 times as many electron-hole pairs are generated. The key to high efficiency is separating those negative and positive charge carriers, and keeping them apart until they leave the cell as an electric current.
Microcrystalline silicon is a promising new material which consists of very many extremely small crystals, embedded in amorphous silicon. It can be manufactured at low temperatures and has the same electronic quality as monocrystalline silicon, at least in principle. But it doesn't absorb as much light as amorphous silicon.
Roll-to-roll PECVD machine
To separate the positive and negative charge carriers, a thin film silicon solar cell has to consist of three layers: a p-type layer, an n-type layer and an 'intrinsic' layer in between. The latter is where the photons are absorbed and converted into charge carriers. The electric field generated by the positively and negatively charged p-type and n-type layers provides the separation.
In cooperation with the German company Roth & Rau, ECN has developed a machine for depositing the three layers using plasma sources: the Roll-to-roll PECVD machine. The process involves the conversion of a silane/hydrogen mixture into hydrogenated amorphous or multicrystalline silicon.
A block diagram of the new manufacturing concept, beginning with steel foil and ending with complete modules. The second illustration shows ECN's method for separation and series connection of the individual solar cells. The laser scribe P1 is exactly deep enough to separate the cells, without damaging the insulating layer. A second laser cuts the connecting and insulating P2 and P3 scribes. Next, insulating and metal pastes are pressed into P1 and P2. When the pastes are dry, the lower contact of each cell will have been connected to the top contact of its neighbour.
Even thinner?
A second way to increase the efficiency is to make a thin solar cell even thinner, so charge carriers can reach the electrical contacts faster. This gives them less chance to recombine. But a thinner layer also absorbs less light, especially when microcrystalline silicon is used - unless the backside reflects the light at a very shallow angle relative to the surface. The ECN-process uses an insulating layer to separate the cell from its steel foil base; in that same layer a special pattern is hot embossed, causing reflecting properties which trap the light to maximise absorption.
"Our partner for developing that special pattern is the University of Ljubjana in Slovenia," says Soppe, who coordinates Silicon-Light, a three-year European programme aimed at improving various aspects of thin film technology and involving eight institutes and manufacturers.
As part of Silicon-Light, a third approach is getting plenty of attention: tandem cells, consisting of one thin film cell on top of the other. Light which isn't absorbed in the top cell can be converted into electric current by the second one.
The result of depth selective laser scribing. The scribe is exactly deep enough, reaching but not damaging the transparent insulating layer (SiO2). A microscope measured that layer's thickness (2 µm) and confirmed that the non-transparent layer on top of it was completely removed.
Photo's and illustration: ECN.
Scribing with light
Another contribution by ECN is a new method to greatly simplify the production of modules: the separation of the individual solar cells and their series connection is done after all the layers have been deposited.
Soppe: "That part of the process has been tested in our laboratory. We are now looking for an industrial partner, so we can help develop a machine suitable for large scale production."
Hard to find? "We have demonstrated the various parts of the process, but not the end result, a complete module. This summer we hope to build a few." Another speed bump is caused by competing technology, like highly efficient cells based on cadmium telluride or copper indium gallium diselenide (CIGS). ECN doesn't pursue those technologies, because metals like indium are relatively scarce. "If we try to use them on a large scale, such basic materials will become unaffordable after a mere ten years or so. We want our solutions to be sustainable. That makes silicon-based technologies far more attractive. But not everyone shares our point of view, not yet anyway."
Text: Steven Bolt
Contact
Wim Soppe
ECN Solar Energy
Tel. 022 456 4087
E-mail: Wim Soppe
Info
Roll to roll fabrication process of thin film silicon solar cells on steel foil
Monolithic series interconnection for thin-film silicon solar cells on steel foil
This ECN Newsletter article may be published without permission provided reference is made to the source: www.ecn.nl/nl/nieuws/newsletter-en/
