Thin-film solar cells

Thin-film solar cells are a relatively modern invention, often seen as a potential future replacement for conventional crystalline solar panels. Due to low cost, low space usage and large flexibility in placement and form factor, it is expected that thin-film solar will have a rapidly growing market share of approximately 20% in 2010. This is roughly equivalent to a total global capacity of 4 gigawatts. With both technologies breaking the critical $1/watt production barrier in 2009, the stage for a battle for market domination has been set. The outcome will largely be determined by whether thin-film solar will be able to exceed cost-per-performance specs of regular solar panels. If so, these thin solar panels might well rapidly take over the solar power market.

What is a thin-film solar cell?

If you ever used one of those solar-powered calculators, then you know from experience what a thin-film solar cell is. The solar cell within these calculators is relatively small and very thin; a property which gives the solar cell its descriptive name. While conventional solar cells feature photo-active layers of approximately 350 µm, this layer is only 1 µm thick in thin-film solar cells. Consequently, only very small amounts of photo-active material are needed to produce a thin-film solar cell, leading to great cost reduction.

In regular solar panels, the semiconducting layer usually consists of one more crystals. Since crystals usually have specific cleavage planes, they can become very brittle when sliced too thin. To counter this brittleness, the semiconductors in thin-film solar cells often consist of a rapidly-cooled non-crystalline variety. This type of material is commonly referred to as ‘amorphous’, meaning ‘without shape’. Since an amorphous material lacks cleavage planes, it can be bent to a much larger degree without breaking. The thinness of the semiconducting layer is however not the only reason the cell is flexible. An additional reason for the flexibility of thin-film cells is that the production of amorphous material does not require extreme temperatures. This allows deposition of the material on a plastic substrate, giving the solar cell assemblage its desired flexiblity.

What are thin-film solar cells made of?

Thin film solar cells are usually based on one of three materials: amorphous silicon (a-Si), copper indium gallium selenide (CIGS) or cadmium telluride (CdTe). The first of these materials can be directly derived from material used for conventional solar panels. The other two are based on heavy metals; they are more efficient at conversion of sunlight and relatively cheap to produce, but the materials themselves are often environmentaqlly unfriendly. It should be noted that a new, fourth, material is strongly gaining momentum: organic solar cells are currently under heavy study and results so far are very encouraging.

Amorphous silicon (a-Si)

By rapidly cooling molten silicon, crystal growth is prevented and an amorphous glass-like variety of silicon is obtained. This material lacks internal structure and is thus significantly more bendable than its crystalline counterpart, which is brittle due to cleavage planes. Amorphous silicon by itself is however full of defects, which greatly inhibit the transfer of electric carriers (see solar cells) and strongly limit the material’s photovoltaic potential. By treating amorphous silicon with hydrogen, these defects are largely corrected. The resulting alloy, referred to as hydrogenated amorphous silicon (a-Si:H), is the basis of most thin-film solar cells seen today.

The single largest disadvantage of amorphous silicon, is that the electrical structure of hydrogenated amorphous silicon is metastable, i.e. the material degenerates when exposed to sunlight. This effect, called the Staebler–Wronski effect, causes a-Si:H based solar cells to lose up to 30% of their efficiency over the first six months of operation. After this time, the efficiency stabilizes., decreasing only marginally with time.

Copper Indium Gallium Selenide (CIGS)

CIGS is a compound semiconductor material with great potential for affordable thin-film solar cells. Although not as efficient as crystalline silicon (the highest recorded efficiency for a CIGS-based cell is 19.9%), CIGS is significantly more efficient than amorphous silicon. It also does not suffer from the degeneration effect that plagues amorphous silicon solar cells. The CIGS layer is usually deposited on a substrate of molybdenum-coated glass, after which it is covered by thin layers of respectively cadmium sulfide (CdS) and zinc oxide (ZnO). These two layers make sure that a proper N-P layer junction exists, allowing the necessary charge separation. Due to the presence of many metals, CIGS cells are susceptible to oxidation; proper encapsulation of the photoactive layers is therefore absolutely necessary to guarantee a sufficiently long lifetime. The below image features a cross-section profile of a CIGS-type solar cell. Click the image for a larger, more easily readable version.

Profile of a CIGS thin-film solar cell

Cadmium Telluride (CdTe)

Research into the photovoltaic qualities of cadmium telluride already begun in the 1950s. The material has a band gap that is almost perfectly suited to sunlight and has thus always been regarded as a highly viable material for photovoltaics. Despite these seemingly brilliant qualities, the disadvantages of these cells also lie in the very materials themselves. First of all, the heavy metal cadmium is toxic and – much like lead – has the tendency to accumulate in the food chain. Second, the metal tellurium is only available in limited quanitites, limiting supply and possibly driving up the price. The highest achieved cell efficiency so far is 16.5%, although efficiencies of 20% are considered a possiblity.

How it’s made: thin-film solar cells

Despite higher material cost, thin-film solar cells based on heavy metals are relatively easy to produce. This is because the material can be applied as a special ‘solar ink’, which is ‘printed’ on the substrate. This production process is so easy, that California-based Nanosolar inc. expects to be able to produce thin-film solar cells at only $0.25/watt in the future. This price point is way below the $1/watt level that is comonly cited as the point of grid parity.

The below video clearly demonstrates the process of ‘printing’ solar cells at a manufacturing facility in San Jose, California: