END USER FOCUS: Solar cell future for minerals

By Mike O'Driscoll
Published: Thursday, 25 February 2010

In the manufacturing process and in cell components, minerals play a vital role in this growing market

The global drive to bring non-fossil fuel energy sources to fruition as swiftly as possible has created several new energy industries, of which solar power is very much at the forefront. Forecasts indicate that photovoltaic (PV) derived electricity could be providing up to 12% of the EU’s electricity demand by 2020.

Although they might not be immediately obvious, for industrial minerals, there are realisable opportunities in the PV manufacturing process (see panel).

PV cells

There are two main types of PV technology: crystalline silicon technology and thin film technology. The most common technology is crystalline silicon, which accounts for around 90% of the PV cells produced.

Crystalline silicon cells are made from thin slices of a single crystal (monocrystalline), or slices from a block of crystals (polycrystalline or multicrystalline).

Thin film PV cells are produced by deposition of very thin layers of photosensitive materials on to a low cost backing such as glass, stainless steel, or aluminium.

Fused silica crucibles

In the PV manufacturing process, high grade quartz is reduced to metallurgical grade silicon and converted to solar grade silicon. The Siemens process makes high purity polysilicon, which is then melted in crucibles at 1,400ºC and then crystallised out.

High quality fused silica is essential to the manufacture of these crucibles. High grade fused silica is processed and mixed into a slip/slurry, slip-cast into molds, and then dried, fired, and cooled in kilns.

There are very few fused silica raw material producers worldwide. Leading fused silica crucible manufacturers include Vesuvius (part of Cookson Plc), Belgium, and Ceradyne Thermo Materials, USA. Vesuvius produces its own fused silica raw material, while Ceradyne is supplied by subsidiary Minco Inc., Greeneville, Tennessee, which also produces fused silica for investment casting, filler, and refractory markets.

Both Vesuvius and Ceradyne in recent years have expanded plants and established Chinese facilities to meet increased demand from the solar cell market, although 2009 saw a major drop in the market.

As a result, Vesuvius closed its Hautrage, Belgium plant and reduced headcount at its Feignes, France, plant by around 30%. Although the company’s new 40,000 crucibles per annum Sunrise plant in Weiting, Jiangsu, China was completed at the end of 2008, it remains uncommissioned pending a pick up in demand.

In 2009, Jiangxi Sinoma New Solar Materials Co. Ltd, China, using gel-casting, developed the world’s largest fused silica crucible with a capacity of 800kg silicon metal for its partner LDK Solar Co. Ltd.

Simplified cross section of crystalline silicon PV cell showing use of industrial minerals

SiC wiresawing

The molten silicon metal is crystallised out using a silicon seed crystal, which is rotated to produce a cylindrical ingot. The ingot is then cut into a square, and the square ingot then is sliced into the silicon wafers destined for solar cells.

This introduces the next major industrial mineral application, that of a silicon carbide (SiC) slurry used with a wiresaw. It is thought that around 90% or more silicon wafer processors use SiC as the abrasive medium of choice.

SiC grades of choice comprise 98-99% SiC, and in FEPA sizes F500, F600, and F800, and occasionally, F1000. The key here is that the SiC grade must be ultra consistent for silicon wafer processors.

Dr Oliver Anspach, manager research and development, of PV Crystalox Solar Silicon GmbH (part of leading PV Si wafer producer PV Crystalox Solar Plc, UK), told IM: “There is still little understood about what goes on in SiC wire sawing, it’s a black box operation, so SiC grades used must be very consistent.”

Certainly, leading SiC producers have been quick to react to supplying this market by expanding more into SiC microgrits, such as Washington Mills Electro Minerals Corp. with its Carborex grades, and Saint-Gobain with its Sika grades.

Although there appeared a shortage of F600 SiC grades a couple of years ago, Anspach considers that SiC supply for Si wafer cutting is satisfactory at present.

However, Si wafer slicing is one of the most costly steps in the Si wafer production process; it is very slow, taking 5-6 hours, even with a multiple wiresaw, and losses as saw dust (kerf loss) are up to 30%.

As a result, one potential cloud on SiC’s PV market horizon might be the advent of more widespread use of diamond wiresaws in Si wafer slicing. The diamond option, which remains largely at R&D stage, could offer quicker and more efficient results, perhaps only around 1.5-6.0 metres of wiresaw per wafer, compared to 400-600km saw/wafer using SiC.

It is understood that two processors in Japan have each switched 50% of their sawing capacity to using diamond saws. However, Dr Anspach warned: “But there are problems with the look of the wafer, and it will be up to what the solar cell companies desire. The diamond wiresawing creates scratches on the wafer surface, and overall it looks ugly.”

Anspach envisages that diamond wiresawing might emerge in the second half of 2010 for monocrystalline Si wafer cutting, but much later for polycrystalline cutting.

Talking to IM at the Ceramitec fair, Munich, in October 2009, Anne Marie Moe, manager technology, Washington Mills AS, Norway, considered that diamond wiresawing was “not a big threat, no-one knows how many years it will take to develop. The sawing of Si wafers is safe as a new growth market for SiC.”

The main reasons for this view stem not only from diamond wiresawing’s perceived infancy, but also that most Si wafer processors have wiresaw plants not suitably set up for diamond wiresaws.

PV manufacturing process from quartz to solar cell system

Module mineral fillers

After cleaning and treatment, the Si wafers are then screen printed with electrical contacts, to form the solar cell. The solar cells are embedded between two glass panes and a special resin is filled between the panes, securely wrapping the solar cells on all sides to form a module. Modules are then linked together to form a system which generates a direct electrical current for a variety of applications.

Industrial minerals may find low volume, but high value niche applications in the composition of some of the above mentioned components of the solar cell modules.

Polymeric materials are used as encapsulators and backsheets (or backskins) in solar cell modules. One of the main encapsulation polymers of choice is ethylene vinyl acetate (EVA), which maybe mineral filled, and one example has used silver coated wollastonite.

For crystalline silicon solar cells, the cells can be encapsulated such that a transparent encapsulant is used between a transparent superstrate (usually glass Ð also using minerals for its manufacture) and the solar cell. In this case, a second layer of encapsulant, which may be pigmented, can be used between the solar cells and the backsheet material.

For thin film solar cell modules a single layer of encapsulant is employed.

A recent patent for encapsulation of PV cells using a silicone based material includes the use of fillers which substantially match the refractive index of the sheet material and maybe selected from one or more of wollastonite, silica, titanium dioxide, glass fibre, hollow glass spheres, and clays.

The backsheet material may additionally comprise one or more fillers to reduce weight and lower cost and to change colour or reflectivity. These may comprise one or more finely divided, reinforcing fillers such as high surface area fumed and precipitated silicas and to a certain degree calcium carbonate, or additional extending fillers such as crushed quartz, diatomaceous earths, barium sulphate, iron oxide, titanium dioxide, and carbon black, talc, and wollastonite.

Flame retardant mineral fillers are also considered for use in solar cell encapsulation materials, and would encompass alumina trihydrate, magnesium hydroxide, and wollastonite.

Building for a solar future: DuPont anticipates
the PV market to grow by 30% in 2010. Courtesy DuPont.

Fluoropolymer backsheets

Fluoropolymer films (derived from fluorochemicals, whose starting raw material is fluorspar) have played and continue to play multiple roles in the PV module package.

The backsheet provides physical protection, electrical insulation, moisture protection and in some circumstances unique colour identification all while adhering to the encapsulant over the life of the module.

Solar cell module backsheets can contain polyviny fluoride, but also in combination with a range of additives including mineral fillers and pigments, with loading levels of up to 30%.

In January 2010, DuPont, a major consumer of fluorspar, completed a $295m. investment in PV film manufacturing, with a $175m. expansion to its DuPont Circleville, Ohio facility producing Tedlar® film.

Manufacturing Tedlar® film includes producing vinyl fluoride (VF) monomer, which is converted into polyvinyl fluoride (PVF) polymer resins, and extruded into the Tedlar® film.

Tedlar® films are a main component of PV cell backsheets, and this expansion capacity is claimed to support global demand of over 10GW of PV module production.

DuPont plans to increase monomer and polymer resin capacity by more than 50%. Construction is underway for these new monomer and resin facilities at Louisville, Kentucky, and Fayetteville, North Carolina, respectively, and are scheduled to start up in mid-2010.

DuPont is clearly bullish about the long term future of PV, since it expects that overall sales of its PV products will exceed $1,000m. by 2012 (sales exceeded $500m. in 2009). DuPont anticipates that the PV market will grow by about 30% in 2010, and continue to grow rapidly over the next three years.