As amorphous silicon exhibits a more random structure than crystalline silicon (see Figure 3.46), not every silicon atom is able to bond with four adjacent atoms, as is the case with crystalline silicon. Free electrons thus form dangling bonds that trap charge carriers that move freely in the semiconductor material. Amorphous silicon (a-Si) becomes a viable semiconductor material only if most of these dangling bonds are saturated with hydrogen atoms, in which case the material is often referred to as Si:H. However, in practice it is not possible for all dangling bonds to be saturated with hydrogen [3.6].
One of the main problems with amorphous silicon solar cells is the stability of their electrical properties in that, relative to baseline, the efficiency of such cells decreases by 10 to 30% during the first few months of operation, a phenomenon known as the Staebler-Wronski effect. Incident photons are able to break weak bonds between silicon atoms (e. g. because of an unduly large gap between the atoms), causing other bonds to break (see Figure 3.47). As a result, the charge carriers generated by the internal photoelectric
Figure 3.47 Principle of the Staebler-Wronski effect [3.6]: The effects of light cause weak bonds to be broken, resulting in additional dangling bonds. Reproduced by permission of Arvind Shah
glass superstrate (1.6 mm)
transparent electrode (tin oxide, 70 nm) p-layer (10 nm)
intrinsic a-Si:H (not doped, 600 nm) n-layer (50 nm)
back side contact Aluminium (1^m)
Structure of a pin solar cell made of amorphous silicon. (Based on data from Arco Solar) effect are more likely to recombine before reaching an electrode, thus provoking efficiency loss. After a time, all ‘compromised’ Si-Si bonds are broken by the action of light, and cell efficiency plateaus at a lower level. This effect is less pronounced in very thin solar cells.
The band gap energy EG values of amorphous silicon solar cells can be decreased by adding Ge, can be increased by adding C, and can be modified either way via partial saturation of broken bonds with substances other than hydrogen (e. g. F). This in turn allows for the generation of tandem or even triple solar cells with three active junctions.
Figure 3.48 shows the structure of an amorphous silicon solar cell, whereintegration of an intermediate layer consisting of non-doped (intrinsic) amorphous silicon allows for the creation of an electric field throughout virtually the entire cell. This facilitates the transport of photogenerated charge carriers and partially makes up for the fact that the semiconductor properties are of lesser quality.
Figure 3.49 shows the degradation of thin-film solar cells made of amorphous silicon. As can be seen here, most of the efficiency loss is provoked by the Staebler-Wronski effect during the first three weeks of exposure to the Sun.
Thin-film solar cells normally have a lower fill factor FF than crystalline silicon solar cells owing to the higher series resistance in the thin films and defects in the semiconductor material. Such cells also exhibit higher open-circuit voltage relative to crystalline silicon solar cells on account of the higher band gap energy of amorphous silicon (see Figure 3.50).
Adding a specific proportion of germanium (Ge) to the silicon reduces band gap energy EG in the amorphous material, thus in turn allowing for the realization of amorphous tandem and triple solar cells that are far less prone to degradation secondary to the Staebler-Wronski effect. Such cells exhibit higher and far more stable efficiency and lower initial degradation than is the case with conventional single-film amorphous cells.
Figure 3.51 shows a cross-section of a Uni-Solar triple amorphous solar cell composed of three optically and electrically series-connected ultrathin cells with varying germanium content and whose front, middle and rear cells process blue, green and red light respectively. However, the cell efficiency of
—— transparent electrode
—— ZnO layer
reflecting layer (Ag, Al)
___ stainless steel foil
solar cell products of this type is just under 9%, the triple structure notwithstanding, and is thus only half that of commercial crystalline solar cells.
Manufacturing amorphous silicon thin-film solar cells is far less energy intensive than for monocrystalline cells (see Chapter 9). But unfortunately, commercially available amorphous silicon solar cells are still very inefficient in that their efficiency plateaus at 3 to 9% following degradation secondary to the Staebler-Wronski effect, and the higher values in this range are only obtainable with triple cells. Inasmuch as the theoretical efficiency of such cells is far higher, the relatively poor performance of the currently available products must be mainly attributable to the still relatively poor semiconductor properties. The peak power prices per watt of these cells are somewhat lower than for monocrystalline and polycrystalline modules. Amorphous silicon solar cells are mainly used today as power sources in calculators, watches and other such consumer products whose efficiency and service life requirements are not as exacting as for PV systems.
Amorphous cells are now coming into increasing use for building-integrated PV systems on account of their greater visual appeal and because amorphous cell contacts are practically invisible and the consequent coloration is far more homogeneous than with crystalline cells. Moreover, at sites with elevated diffuse radiation, relative to crystalline cells, amorphous solar cells even exhibit somewhat higher specific energy yield (kWh produced per Wp of solar generator power) as they are more compatible with the diffuse radiation spectrum (see Figure 2.42). Amorphous triple cells (see Figure 3.51) are particularly well suited for building-integrated PV systems and for the manufacture of specialized hybrid products for this purpose (electricity-producing roof elements).