LiNbO₃ is not found in nature as a natural mineral. The crystal structure of lithium niobate (LN) crystals was first reported by Zachariasen in 1928. In 1955 Lapitskii and Simanov gave lattice parameters of hexagonal and trigonal systems of LN crystal by X-ray powder diffraction analysis. In 1958, Reisman and Holtzberg gave the pseudoelement of Li₂O-Nb₂O₅ by thermal analysis, X-ray diffraction analysis and density measurement.

The phase diagram shows that Li₃NbO₄, LiNbO₃, LiNb₃O₈ and Li₂Nb₂₈O₇₁ all can be formed from Li₂O-Nb₂O₅. Due to crystal preparation and material properties, only LiNbO₃ has been widely studied and applied. According to the general rule of chemical naming, Lithium Niobate should be Li₃NbO₄, and LiNbO₃ should be called Lithium Metaniobate. In the early stage, LiNbO₃ was indeed called Lithium Metaniobate crystal, but because the LN crystals with other three solid phases have not been widely studied, now LiNbO₃ is almost no longer called Lithium Metniobate, but is widely known as Lithium Niobate.

High-quality LiNbO3 (LN) crystal developed by WISOPTIC.com

The co-melting point of liquid and solid components of LN crystal is not consistent with its stoichiometric ratio. High quality single crystals with the same head and tail components can be easily grown by melt crystallization method only when materials with the same composition of solid stage and liquid stage are used. Therefore, the LN crystals with good solid-liquid eutectic point matching property have been widely used. The LN crystals usually unstated refer to those with the same composition, and the lithium content ([Li]/[Li+Nb]) is about 48.6%. The absence of a large number of lithium ions in LN crystal leads to a large number of lattice defects, which have two important effects: First, it affects the properties of LN crystal; Second, lattice defects provide an important basis for the doping engineering of LN crystal, which can effectively regulate the crystal performance through the regulation of crystal components, doping and valence control of doped elements, which is also one of the important reasons for the attention of LN crystal.

Different from the ordinary LN crystal, there is “near stoichiometric LN crystal” whose [Li]/[Nb] is about 1. Many of the photoelectric properties of this near stoichiometric LN crystals are more prominent than those of the ordinary LN crystals, and they are more sensitive to many photoelectric properties due to near-stoichiometric doping, so they have been extensively studied. However, since the near-stoichiometric LN crystal is not eutectic with solid and liquid components, it is difficult to prepare high-quality single crystal by conventional Czochralski method. Therefore there is still lots of works to do to prepare high-quality and cost-effective near-stoichiometric LN crystal for practical use.