At the onset of this blog, I interviewed Prof Aron Walsh about control defects within halide perovskites. The conclusion was that because charge carriers within this material behave as if there were no defects, doping the structure (i.e. adding some positively or negatively charged imperfections in the lattice) have no effect on the electrical properties. Thus, the creation of p-n junctions in halide perovskites would be made impossible, at that stage of the understanding of this class of materials.
A report in Nature Energy earlier this month has pushed this boundary further. Cui, Wei and co-workers report a homojunction of p-type and n-type MAPI perovskite, that is stable enough to produce a 21% efficient solar cell. They achieve the doping of the perovskite by deposition of the n-type layer with low iodine concentrations and the p-type under rich iodine condition. In this case, iodine vacancies are suppressed, giving room for cation vacancies, that are electron-acceptors.
However, it is now well established that halide perovskites (and particularly MAPI perovskites) are mixed ionic and electronic conductors. This would raise some questions about the long-term stability of such a structure, as cations or anions may be transported through the homojunction. Note that the conduction is even stronger under illumination, suggesting that the ionic mobility is greater, too.
Prior to this report, most of the perovskite solar cells were made in