The efficiency η of a solar panel refers to the percentage of incoming sunlight's energy converted by the panel into electrical energy. It is determined by the efficiency of the individual cells. Before discussing the actual numbers, let's briefly review how solar power conversion works. Light consists of packets of energy called photons. They have different wavelengths and different energy. When sunlight falls onto a material, some of the photons absorbed by the material increase kinetic energy of atoms and molecules, so the energy of these photons just dissipated into heat. Other photons are absorbed by electrons, which can move into a higher energy state called "excited state".

In solar cells internal p-n junctions prevent these electrons from returning to their original state. This creates hole-electron pairs resulting in a voltage between the cell terminals. This voltage can drive charges through an external load which results in electrical work.


Photons need to have a certain minimum energy to excite electrons and generate a hole-electron pair. The lowest-energy sunlight photons should have to excite electrons in a particular material are determined by the material's energy bandgap. The bandgap is the minimum amount of energy needed to free an electron from its bond. Its value is different for different semiconductors. Photons with energy below the band gap of the absorber can't excite free electrons, so the corresponding portion of the sunlight's energy is dissipated as a heat. If a photon has energy greater than the band gap, it can excite an electron, but the excess of energy will still be converterd to heat. Since only a small fraction of the photons in the light spectrum has energy close to the material bandgap, substantial power is lost through this mismatch. This puts a fundamental limit to PV efficiency. According to physics, the theoretical limit of η in a single-junction cell operating at "one sun" is about 30% for a typical band gap 1.1 eV. This limit is called Shockley-Queisser limit. In order to achieve η approaching Shockley-Queisser value, the material's energy gap must be between 1.0 and 1.7 eV.


A Belgium company called Imec has demonstrated a prototype of a single-junction GaAs cell with a record η=24.7%. In practice however, the perfomance of mass production commercial modules is about 0.8 of the best cell performance. Today's commercially available PV modules have efficiency ranging anywhere from 6 to 21%: see the list of the most efficient solar panels. Currently, the average values of η for different types of solar panels are as follows: crystalline silicon - 17-20%; crystalline silicon (ribbon) - 12%; thin-film (amorphous silicon) - 8%; thin-film other material (such as CdTe and CIGS) - 12%. A higher level of efficiency can be obtained with multi-junction cells that use multiple materials to better match the solar spectrum. In such devices, individual cells with different bandgaps are stacked on top of one another in such a way that sunlight falls first on the layer having the largest bandgap. Photons not absorbed in the first cell are transmitted to the second cell, which then absorbs the next lower-energy portion of the photons, while remaining transparent to the rest of them. These selective absorption processes continue through to the last cell, which has the lowest bandgap. The open circuit voltage of a multijunction cell is the sum of the individual cell voltages, while the peak current is slightly less than a single cell's current. Multijunction cells available commercially from Spectrolab have rated efficiency of 29%. Sharp achieved 35.8% in its research lab by using the triple-junction compound cells. A company called Semprius has demonstarted greater than 41 percent efficiency of gallium arsenide triple-junction cells. Previously, prototypes of multi-junction cells from other companies also demonstrated record efficiencies of up to 44% in the research labs. The trick is these results were obtained under heavily concentrated light up to 1000 times the normal amount of sunlight. These technologies may be useful in large-scale installations that use lenses or reflective devices to focus sunlight onto the solar arrays, but they are not very practical for homes. Also see our review of solar panel costs.


There are various types of PV cells. Original PV cells required silicon in a very pure form. As the result, polycrystalline silicon (poly-Si) used to be the main practical option. A shortage in poly-Si kept the cost of the PV devices high. There is a lot of research and development conducted around the world aimed at boosting the efficiency of photovoltaic technology and reducing its costs. Dow Corning Corp. reportedly has created solar-grade (SoG) silicon derived from metallurgical silicon that exhibits good PV performance. The Sargent Group in University of Toronto was working on special plastic solar cells by using nanotechnology. These cells are intended to utilize the sun's broad spectrum including invisible infrared rays, which carry half of solar radiated power. However, these devices reportedly exhibited less then 5% infrared power conversion efficiency and 1.8% overall efficiency. There are also attempts to make paintable solar cells that could be printed more cheaply - with a roll-to-roll printing process on a plastic substrate or stainless steel. However so far the prototypes of these cells have efficiency of only 1%. In my view, for now all these technologies present only an academic interest. In general, despite of all the media hype around solar energy, based on the recent research sponsored by US government, in the near future the efficiency of commercial PV modules in non-concentrated sunlight is not expected to exceed 24%. This means that photovoltaic electricity will remain significantly more expensive than electricity produced from traditional sources, and the solar market will continue growing primarily due to government subsidies and mandates rather than free market forces.

References and additional information:
PV technologies in competitive PV module markets;
Solar photovoltaics white paper.