Our group investigates the ferroelectric materials with perovskite structure for almost thirty years. In particular, those are KTaO3, KNbO3, etc. And I can say for sure that doping of those crystals in the bulk as well as in the nanocrystalline form is not difficult. Moreover, the vacancies (mostly, oxygen vacancies) and impurity defects (the latter one is an unavoidable impurity, which incorporates the lattice during the syntesis) are the primary defects which form even in the pure (undoped) crystals. Please read about the doping procedure and the defects structure in our publications.
Im working on the same structures. But it works easily. You should change the reaction condition or synthesis method. However, you should consider the dopants radii and charge that should be nearly same or have small different
The possibility of formation of solid solutions with the perovskite structure ABO3 isassociated with ionic radii of the dopants and is determined by Goldschmidt tolerance factor. the information about tolerance factor you can find here: http://en.wikipedia.org/wiki/Goldschmidt_tolerance_factor
And how about a popular perovskite LaGaO3 doped with strontium, magnesium, and d-elements, which is used as an oxygen conductor, therefore has oxygen vacancies? Perovskite structure is very tolerant to doping even heterovalent with resulting vacancies.