Observations on Growth of BaTiO3, Crystals from KF Solutions

The purpose of this study was to try to find out why BaTiO₃ (essentially a cubic structure) can consistently be grown from KF solutions as large flat plates when a more equant morphology is expected. It is suggested that the size, shape, and perfection of the platelike "butterfly" twin crystals can be qualitatively related to the re-entrant angles which result from {111} twinning of crystals of {100} habit. The re-entrant angles form permanent steps for both the advancing crystal tip and for the growth layers of the wings. The amount of twinning is related to the cooling rate and to the presence of certain impurities. Thus it can be shown that increasing the cooling rate increases the number of {111} twins (essentially a hexagonal stacking fault); the extreme case is the metastable formation of hexagonal barium titanate by quenching. Evidence is presented for the stabilization of the Ti₂O₉ group of the hexagonal structure by substitution of ions of lower polarizing power.