Semiconductors and Silicon Solar Cells

There is information on two web sites that will help you understand this field.
  1. There is an excellent web site on the properties of semiconductors by a group at the University of Glasgow.
  2. Superconductors.Org has a simple and easy to understand page on the basics of superconductors.

Conductors and Insulators

In any solid, metallic (such as Zn), ionic (such as NaCl), or covalent (such as SiO2), the valence orbitals of atoms and molecules combine to form molecular orbitals. The number of molecular orbitals is equal to the number of atomic orbitals, and for any material, this is a very large number but the energy difference between highest and lowest doesn't change by much. Many orbitals separated by very small energy gaps give rise to bands of orbitals.

For sodium metal, the band is made up of molecular orbitals formed from the combination of 2s atomic orbitals and about half of the orbitals are filled. There is only a very small energy gap between the highest occupied molecular orbital (HOMO) and the lowest unoccupied molecular orbital (LUMO). When an electron goes to an unoccupied orbital with available thermal energy, it is free to travel throughout the metal. The "hole" left in the HOMO can travel too.

In silica (SiO2) there are covalent bonds between each silicon and 4 oxygens around it that are combinations of Si 3s and 3p atomic orbitals and O 2s and 2p atomic orbitals. We can think of the molecular orbitals forming similar bands but there is a large energy separation or band gap between the occupied and unoccupied levels. The electrons in the orbitals of silica are not free to travel and the material is an insulator.

In pure, crystalline silicon, there is also a band gap between the occupied and unoccupied molecular orbitals but it is smaller. Electrons in the filled valence band can be promoted to the conduction band by thermal energy or when the material aborbs light. The electrons would then be free to travel through the material and holes (electron vacancies) would travel through the valence band.

Key Points on Semiconductors

Silicon and germanium are intrinsic semiconductor that have semiconducting properties as pure materials. These elements have the same crystalline form as diamond, an insulator, but their lattice constant,a (width of the cube) is much larger. These elements each have 4 valence electrons.

What about mixtures? A mixture of 50% Ga and 50% As would also have an average of 4 valence electrons and this substance forms crystal that have the diamond lattice, just like Ge. The material gallium aresinde is also a semiconductor. Can you think of others?

Adding a very small amount of certain materials to a host material can transform it into a doped semiconductor. If the dopant has more valence electrons than the bulk, the resulting semiconductor is a n-type. For example, a crystal of Si with a few atoms of P is an n-type semiconductor. A dopant with fewer electrons than the bulk material gives a p-type semiconductor. You can make this kind of semiconductor by adding a few atoms of Al.

To begin, it is important to understand electrical circuits. If you need a review, there is an excellent and easy to understand online text on circuits and electricity: All About Circuits.

Photovoltaic Cells

When light strikes the semiconductor of a photovolatic cell energy is absorbed. Electrons are promoted to the conduction band and holes are formed in the valence band. These charge carriers could travel through an external circuit and do useful work or they could simply recombine and give off their energy as heat.

A photovoltaic cell must have two types of semiconductors in layers. (There is more information on semiconductors in the next section.) The n-type semiconductor is doped with a very small amount of an element containing an extra valence electron (such crystalline Si doped with As). The p-type semiconductor is doped with an element with one fewer valence electrons (such as crystalline Si doped with Ga).

When put together in a pn junction, an electric field forms at the junction that biases the flow of electrons and holes to opposite directions. This reduces electron-hole recombination.


The maximum effiency for a solar cell made from single crystal Si is 24.7 % but typical efficiencies are in the range of 11-15 %. Cells from amorphous silicon are much cheaper to produce but have typical efficiencies of on 5-7 % (maximum 12.7 in laboratory setting). Efficiencies greater than 25 % have been recorded for the more expensive GaAs semiconductor. Multijunction cells can have higher effiencies. One with a composition of GaInP/GaAs/Ge has an efficiency of 32 %. Factors effecting efficiency of photovoltaic cells include
  1. Light absorption by the semiconductor
      Cell thickness has an effect on light absorbed.

  2. Band Gap
      This is the energy that the semiconductor absorbs as it promotes an electron. If the light energy is greater than the band gap, the excess energy is wasted as heat. Stacking different semiconductors can work only if their crystal forms are very similar.

      Some Band Gaps
        (a) Si, 1.1 eV gap
        (b) AlGaAs, 1.7 eV
        (c) InN, 0.7 eV
        (d) GaN, 3.4

  3. Recombination
One approach to improving electricity output is to ignore efficiency but to maximize cell output by using cheap solar concentrators made from plastic.

Professor Shapley, University of Illinois, 2011