To this day, the sun remains the only inexhaustible and eco-compatible vector of energy; however, the major difficulty remains converting this energy into electricity efficiently and at low cost. Photovoltaic conversion makes this possible. However, this approach remains a colossal challenge, despite the many existing solar cell technologies (inorganic, organic or dye-sensitized), which we describe below, outlining their operating principles and efficiencies. Inorganic and organic cells, with efficiencies of 22% and 10% respectively, work by absorbing a photon by a semiconductor to create an electron-hole pair, which is dissociated at the pn junction to generate electricity. Dye cells, on the other hand, copy plant photosynthesis, i.e. the creation of electrons and holes is generated by the absorption of light by a pigmented dye (the equivalent of chlorophyll in plants). The evolution of these various technologies was reviewed, including 1) the development of heterostructures, bipolar cells or concentrating cells enabling yields of up to 40% to be achieved, or 2) dye cells, or organic cells, through the development of new pigments, or based on judiciously chosen organic molecule acceptor-donor couples. The dream of higher yields may be within reach, since theoretical calculations indicate that yields in excess of 85% could be achieved.
To move in this direction, researchers are now trying to play with particle size and move up to the nanometric scale, a situation in which the electron behaves more like a wave than a particle " quantum box ". Among the new concepts being pursued, the recurring problem remains the slow cooling rate of photoexcited carriers. One approach is to collect them by reducing the phonon thermalization rate, either by recovering thermal energy through the creation of several electron-hole pairs (excitons), or by photovoltaic-phonon coupling (reducing the phonon thermalization rate). These advances mean that the cost of photovoltaic energy per kWh will catch up with that of nuclear power by the 2030s.
