engleski

Potential of thin silicon films for "third generation solar cells"

The possibility of larger contribution of solar electricity in total energy consumption became in the last decade more and more appealing task due to number of reasons among them are the global increase of energy demand, constant growth of oil prices, tendency towards independency in energy supply, care about air pollution due to increase of green house gases and many others. In order to approach this goal, the price of solar cells per unit power has to go down from today values that are 2-5 € /Wp towards few tents of euro. Historical look on the correlation between overall solar cells production and price per Wp shows gradual decrease. However, this approach contains number of parameters like marketing, economy of large scale production and others. The more interesting insight gives overview of material consumption. Simple estimation done by M.A. Green [1] show for crystalline silicon cells (90% of actual production) that the price for starting material is much too high for achieving this goal. For thin films cells, the situation is better - the low material consumption (about two orders of magnitude lower comparing to crystalline) lowers the price, while for fulfilling the expecting low price the efficiency should be much higher. This jump in efficiency is expected from new type of cells called "third generation solar cells". There are several approaches to obtain extreme high efficiency, based on minimizing the loses of energy of incoming radiation in transformation the photon energy into electric energy. The general concept is to overcome the thermodynamic limit that is characteristic for solar cells made of one type of material. In general the absorbing material is semiconductor with optical gap that defines the lowest energy of photons that can be absorbed by material. For photons with lower energy than optical gap, material is transparent. Absorption of photons with larger energy than optical gap results in creation of electron-hole par. Since solar spectrum has wide distribution, most of created tree electrons have higher energy than energy corresponding to the bottom of conduction band. These electrons are called "hot electrons". In large crystals, this excess energy is transformed in to vibration of lattice in many small steps trough process called thermalisation, in a very short time (~ 0.1 ns). The limit of efficiency due to loss of energy trough the processes of thermalisation is called "thermodynamic limit" and together with loss of part of spectrum where energy of photons are to low to create electron-hole par, define the limit of solar cell efficiency. Consequently, the best photovoltaic efficiency is expected if the optical gap of material is close to energy of photons absorbed in material. Since solar spectrum has wide distribution, high efficiency can be obtained by multilayer structure e.g. with combination of several cells stacked one after other, starting trom the cell with highest optical gap. This concept has been approved by using combination of ternary compounds like GaInAs. However, most of actual combinations contain toxic elements or are expensive or have serious problems with crystal lattice mismatch. For that reasons, silicon as abundant, low cost, stabile and non-toxic has lot of appeal. During history, amorphous silicon showed itself as a promising due to high absorption and low price. However, due to low stability, the thin film cells using single junction Si structure lowers the price per Wp but not enough. Tandem cells that use combination of amorphous Si with optical gap of 1.7 eV and microcrystalline silicon with optical gap of 1.1 eV, offer higher efficiency and folow the way towards multi-layer structures with various optical gaps. Further step in building the multi-layer structures offers nano-crystalline silicon. In this type of material it is possible to vary optical gap by variation in size of nano-crystals that build thin film by using quantum confinement effects. The variation of optical gap by changing the nanocrystal size has been confirmed by many experiments and explanation has been done in various theoretical models. However, theory and praxis work well for isolated nanocrystals in homogeneous matrix. When density of nano-particlcs increases, surrounding matrix is not homogeneous any more and the processes that appear when nano-crystals are connected are not jet fully explained and needs additional theoretical and experimental to be fully explored. Certain efforts has been also done to study other possibilities of using nano-crystalline materials for "solar cells third generation" [1, 2]. Most challenging in this respect is to use specific distribution of phonon states associated with small dimensions of nan-crystals ("quantum dots") for better economy of "hot electrons" in following two ways [3]. First approach is to increase the output voltage by extraction of "hot electrons" before thermalization. In other approach is expected that "hot electrons" produce one or more new electron-hole pars in process called impact ionization. The electron excited well above optical gap can loos its energy in one step process, and emit photon with enough energy to create one other tree electron-hole par (exciton) by absorption in the next atom. In that way the photo-current will be enhanced since one photon can generate two or more excitons. The both above mentioned processes, extraction of hot electrons before de-excitation and multi exciton generation (MEG) assume slowing the thermalization process that is extremely fast in bulk semiconductors (~ps). In quantum dots the phonon states are discrete quantized energy levels which enhance the hot carriers cooling time for several orders of magnitude (up to some ~100ps) and makes the both approaches plausible. Moreover, recently Nozik and co-workers [4] prove MEG in colloidal dispersed Si nano-crystals of some 5-10 nm. In such "large" crystals the optical gap is close to 1.2 and allows using quite large part of solar spectrum to be involved in process. Finally when consider the use of silicon thin films in "third generation solar cells", the both concepts, the multilayer structures and "hot electron economy" sound promising. The bi-layer thin film Si cells are already widely explored idea, using combination of two amorphous layers and amorphous microcrystalline combination with laboratory efficiency over 12% and for industrial application in the 7-9% range. The use of nano-crystalline silicon with variation in crystal sizes could be easy extended this tandem bi-layer structure towards multi-layers while. Multiexciton generation also works well in laboratory condition. In conclusion, with certain additional scientific and R&D efforts, even with cutTent technology level for large area thin film production, the end goal - the price of few cents per Wp using silicon as base material looks attainable in the near future. [1] M. A. Green, Third generation photovotaics: solar cells for 2020 and beyond, Physica E 14 (2002), 65-70). [2] M. A. Green, Materials Science and Engineering B 74 (2000) 118-124. [3] A. J. Nozik, Inorg. Chern. 44 (2005) 6893-6899). [4] Nano Letters, 7, No.8 (2007) 2506-2512.

thin films solar cells; nano-crystaline Si

nije evidentirano