The Photovoltaic Process

• Photovoltaic cells are built from semiconductors, the same class of materials used to make computer chips. Semiconductors are constructed from layers of crystalline silicon or another bulk material. The crystals normally have all their electrons "tied up," stuck close to their home atoms in a low energy state called the "valence band." When an electron absorbs energy from a photon, it gets bumped up to a higher-energy "conduction band," where it can be collected to provide electrical power.
  
  

Absorbed Light

• For a photon to be absorbed, it needs to have at least enough energy to push an electron from the valence to the conduction band. That energy difference is called the "bandgap," and it is a function of the bulk material of the crystal and the fabrication details. The most efficient absorption comes when the photon has just more than the bandgap energy --- photons lower than that energy are not absorbed at all, and as their energy gets higher than the bandgap their absorption goes down. Energy is just another way of measuring wavelength, so the bandgap determines what kind of light a solar cell needs.
  
  

Silicon (Si)

• Crystalline silicon has a bandgap of about 1 electron-Volt (eV), which corresponds to an upper wavelength limit of about 1 µ, or 1 millionth of a meter, a wavelength of invisible infrared radiation (visible light is between 0.4 to 0.7 µ). The responsivity --- a measure of how well the solar cell converts light to electricity --- drops down from its peak at 1 µ to disappear completely by about 0.4 µ. So a silicon solar cell can use wavelengths from the near infrared at 1µ down to 0.4µ-wavelength blue light.
  
  

Cadmium Telluride (CdTe)

• Crystals are made from identical building blocks assembled in a repeating pattern. In silicon semiconductors, the building blocks are individual atoms, but in another class of semiconductor the building blocks are composed of a pair of atoms. Such is the case with cadmium telluride, where one atom of cadmium and one atom of tellurium compose the building block. CdTe solar cells are much thinner and lighter than traditional Si cells, and their efficiency is comparable. The CdTe bandgap is about 1.5 eV, which corresponds to a long-wavelength limit of about 0.8 µ. CdTe responsivity cuts off at about 0.4 µ.
  
  

Copper Indium Gallium Selenide (CIGS)

• As might be expected from the name, the crystal structure of CIGS is more complex than either Si or CdTe. CIGS solar cells are composed of a combination of copper-indium-selenide (CIS) and copper-gallium-selenide. Although the crystal structure is more complex, the manufacturing process can potentially be inexpensive and the bandgap and other properties can be tuned to provide very efficient energy conversion. A "typical" CIGS cell will work from 0.9 to 0.4 µ, but different manufacturers tune the responsivity to cover different wavelengths.
  
  

Multijunction Cells

• Multijunction cells have one semiconductor built on another semiconductor, perhaps built on another. High-energy, short-wavelength light is absorbed strongly by the top crystal; lower energy, long-wavelength light absorbed by the next; then even longer wavelengths by the layer beneath. Many different materials can be used, but one common combination puts indium gallium phosphide on top of indium gallium arsenide on top of germanium. The whole point of these complex, expensive structures is to maximize the responsivity over a wide range, so these cells can accept light from about 1.8 µ to 0.4 µ.