Solar cell performance is strongly related to two factors. The first factor is how much light the cell can convert into electricity, and the second factor is how much energy gets lost due to efficiency loss.
Converting Sunlight Into Electricity
When it comes to how much light the solar cell can convert into electricity, the most important thing to consider is what the band gap (or energy gap) should be.
What is this band gap? In semiconducting materials, there is a conduction band and a valance band. The band gap is the energy range between the two bands where no electron states can exist.
A high band gap means that the solar cell will produce a high voltage and low current. A low band gap means that the solar cell will product a low voltage and high current.
Since Power = Voltage x Current, there is an optimal band gap value for maximizing power. For solar cells, this value is roughly 1.34 eV which results in a 33.7% efficiency.
With current technology, we can pretty much make the band gap to be anything we want. The main trade-off is how expensive it is for us to tune our solar cell to the optimal band gap.
Ideally, we want the band gap to be slightly lower than the energy level of the light we are trying to capture. This means that the light will excite electrons just barely into the conduction band.
If the light energy is much higher than the band gap, it will still excite the electron from the valence band into the conduction band, but the electron will quickly fall to the lower energy levels in the conduction band – resulting in wasted energy. If the light energy is lower than the band gap, then the electrons cannot be excited enough for the solar cell to produce power.
As shown in the graph above, the sun is emitting light at a wide range of energy levels (the shorter the wavelength the higher the energy of the light), and each semiconductor material can only have one band gap – which means that a single semiconductor cannot be optimized for all energy levels. One method to overcome this is to overlap semiconducting materials of different band gaps on top of each other. However, this greatly increases the cost of the solar cell.
Solar Cell Efficiency Losses
When it comes to solar cell efficiency loss, the major factors to consider are:
- The fact that many photons (light) do not have enough energy to excite the electrons through the band gap.
- The high energy photons excite the electrons too high up into the conduction band. These electrons quickly drop to the lower energy levels in the band, and gives off waste heat in the process.
- Some electrons fall from the conduction band into the valence band before it gets out to the electrodes. This process is called recombination, and there are four main types to consider.
The four main types of recombination losses are:
Radiative Recombination. This process occurs when the electron falls from a semiconductor’s conduction band to the valence band and gives off its energy in the form of light. This is what we want to happen in LED’s, but not in solar panels. This recombination is not typically a problem in silicon, but it can be an issue in other semiconductors used to make solar cells.
Auger Recombination. This process occurs when an electron recombines with a hole, and gives its energy to another electron which kicks it higher in the conduction band. The electron that was kicked higher quickly relaxes and falls into a lower energy level while giving off heat. For this process to happen you need two electrons to be very close to each other (eg. a material that is heavily N-type doped).
Shockley-Read-Hall Recombination. Shockley-Read-Hall (Trap-Assisted) recombination occurs when there is an impurity which introduces an energy level within the band gap of the semiconductor. Usually, an electron or hole will first get trapped in the impurity, and the other carrier will find it. This type of recombination is usually the biggest problem in silicon solar cells.
Surface Recombination. In semiconductor crystals such as silicon, the atoms are usually bound to each other. However, on the surface of the material, there are no atoms to bond to on one side. This creates a “dangling bond”, and introduces states in the band gap much like the trap-assisted recombination.