Micro- to nano-domain engineering of ferroelectric single crystals aiming novel applications ---- From Frozen to Mobile Ferroele

Ferroelectric lithium niobate (LiNbO3 :LN) and lithium tantalate (LiTaO3 :LT) crystals were discovered by Matthias and Remeika in 1949. They have attracted a great deal of interest because of their large piezoelectricity, pyroelectricity, electro-optic and nonlinear optical coefficients, and transparency in a wavelength range from near ultraviolet (UV) to mid-infrared. No other materials have been investigated in so many various fields or used in a wide range of applications because of their multi-functional ferroelectric properties. One definition of ferroelectricity is that "polarization can be inverted by an external electric field." This external electric field to invert polarization is called a "coercive field". However, the inversion of polarization in these materials was considered impossible at room temperature until lately. It was reported that the coercive field for LT was ~5 MV/mm at room temperature, extrapolated from the dependence of the coercive field at temperatures from 200 to 620 ˚C for this crystal. This estimated coercive field is far in excess of the breakdown voltage of the material, meaning it was impossible to invert the polarization of LN and LT crystals at room temperature. This state was then called "frozen ferroelectricity". It turned out that polarization could be inverted by applying ~22 kV/mm to both congruent LN and LT (CLN, CLT) crystals at the end of the 1980's. Although the coercive field was still considerably high, this fact was made evident because new functions in combinations between the original ferroelectric properties and the nonlinearity induced by inverted domain structures had attracted a great deal of attention. However, the large coercive field of CLN and CLT has been making arbitrary domain patterning considerably difficult. The important fact that the coercive field strongly depends on the nonstoichometry of LN and LT crystals was discovered in 1998. The coercive field of near-stoichiometric LN (SLN) crystal is ~ 4.0 kV/mm, which is about one fifth that of CLN crystal, and that of near-stoichiometric LT (SLT) crystal is ~ 1.7 kV/mm which is about one tenth that of CLT crystal. The decrease in the coercive field is extremely advantageous in precise and micro-domain patterning. This discovery has increased the applications for domain patterning. For example, it is used to fabricate thick QPM devices for high-power applications. Moreover, the decrease in coercive field has contributed greatly to microscopic domain engineering such as nanoscale domain engineering through the use of scanning probe microscopy (SPM) since it is possible to invert polarization with a relatively low external field applied through the conductive tip of an SPM. Many issues concerning the mechanism for inverting polarization are still not clear because of the large coercive field of CLN and CLT crystals. Nanoscale domain engineering using SPM is an effective means of clarifying the fundamentals underlying the polarization inversion mechanism and of revealing new phenomena in domain engineering. This talk will summarize the relationships between various properties and nonstoichiometric defect densities of these materials and focus on the observation, fabrication and applications of the ferroelectric domains in defect-controlled LiNbO3 and LiTaO3 crystals using various technologies of SPM, revealing a lot of new findings and potentials on ferroelectric domain engineering.