The choice of nanocrystal depends on several things. One, is it has to be band-gap tune-able around 1.3 eV, which is optimal for single junction solar collection. The second, is that it needs to be readily colloidally synthesized (for inexpensive applications). The third, is that it needs to be air and moisture stable (but this is not critical, as there are encapsulation methods). There is ongoing research into other potential quantum dot materials, but based on these criteria I think it's possible to account for why certain materials have not seen much use/success.

David Zhitomirsky, thank you very much for your answer. Do you think only band gap value is important. What about band structure i.e. I mean band alignment of the QDs with respect to the host semiconducting oxide?

Another question is: some oxide and nitride materials (for example: copper oxide, Indium Nitride) have band gap values in this range. Why these materials are not used in QDSCs?

The tunability in band gap is one the most important reason for it. many oxides even though they are very stable their bang gaps are not tunable. band gap tunability arises due to size quantisation effect and even if they are tunable they absorb only in uv region for e.g. ZnO. However many selenides, sulfides. tellurides like cds, cdse, cdte, pbs, pbse, pbte have a wide tunability range in some cases from nir to visible region which is highly beneficial for a solar cell type application.

Thanks Sreejith. My question is: if the particle size is ~ Debye length (~few nm in size), isn't is a suffucient condition for size quantization? Is there any other criterion that is needed?

Secondly, is only size quantization responsible for band gap tunability?

Both the Debye’s length of the particles and band gap are important as follows:

[Effect of Band Gap]: In a solar cell, light is absorbed by a semiconductor producing an electron-hole (e-h) pair which may be bound and is referred to as an excitation. It is separated by an internal electric field which is created by doping one part of semiconductor interface with atoms which act as electron donors (n-type) and another with electron acceptors (p-type) which results in a p-n junction.

Generation of an e-h pair requires that the photons of light have energy more than band gap and hence the energy of a band gap is a major factor determining the electrical conductivity of a solid. Photons with lower energies than the band gap do not get absorbed while the higher energy photons are relatively inefficient as an energetic e-h pair can quickly (within 10−13s) thermalize to the band edges and reduces the energy. The former limitation results in a lower current, while the thermalization reduces the maximum achievable cell voltage.

[Effect of Particle Size]: This thermalization ALSO depends upon the particle size because the lattice site motions of bulk of the semiconductor are generally increased with the increase of the size up to a certain extent.

Further, the solar light spectrum on the earth is spread across the visible, near-IR ranges with a very small part in the near-UV region (1.1-2.0 eV). So the semiconductors whose band gaps fall in this energy gap are best suitable. The metal oxides, carbides, nitrides and phosphides (high values of band gaps) and the metal arsenides, antimonides and phosphides (low values of band gaps), in general, are not suitable as sensitizers. Of course , metal sulphides, selenide and telurides are suitable as sensitizers.

Exceptions do occur especially in case of heavy metals (PbS,ZnS,In2S3 and CuS) which may decrease their photosensitivity.

The band gaps of some semiconductors are given as:

GaN(3.4);Ga P(2.26); Si3N4(5.0); BN(5.96); Zn S (3.68); Zn O(3.5) Si C(3.0);TiO2(3.02) ,SnO2(3.7)SiO2(9.0); Cd Te (1.5); Cd Se (1.74); SnS (1.0)Si(1.1);Cu2S(1.2) PbS (0.37); In As (0.36); In Sb (0.17)

In the above answer, it may be added (under Effect of Particle Size) that the increase in the particle size also increases the band gap of the photosensitizer.

Thanks, Manohar Sehgal.