Abstract
Vibrational spectra of three-component BaTiO3/SrTiO3 / CaTiO3 short-period superlattices grown by pulsed laser deposition with atomic-layer control have been investigated by ultraviolet Raman spectroscopy. Monitoring the intensity of the first-order phonon peaks in Raman spectra as a function of temperature allowed the determination of the ferroelectric phase transition temperature Tc. Raman spectra indicate that all superlattices remain in the tetragonal ferroelectric phase with out-of-plane polarization in the entire temperature range below Tc. The dependence of Tc on the relative thicknesses of ferroelectric (BaTiO3) to nonferroelectric materials (SrTiO3 and CaTiO3) has been studied. The highest Tc was found in superlattices having the largest relative amount of BaTiO3, provided that the superlattice maintains its coherency with the substrate. Strain relaxation leads to a significant decrease in the ferroelectric phase transition temperature.
| Original language | English |
|---|---|
| Article number | 054106 |
| Journal | Journal of Applied Physics |
| Volume | 105 |
| Issue number | 5 |
| DOIs | |
| State | Published - 2009 |
Funding
UV Raman spectroscopy was applied to studies of three-component BaTiO 3 / SrTiO 3 / CaTiO 3 SLs grown by atomic-scale-precision pulsed laser deposition. Raman data show that even the SLs having the ferroelectric BaTiO 3 layers as thin as 1 u cell are ferroelectric with T c near room temperature. The temperature evolution of Raman spectra indicates that all SLs remain in the tetragonal ferroelectric phase with out-of-plane polarization in the entire temperature range below T c . The latter was determined from the temperature dependence of the intensity of the first-order phonon peaks in Raman spectra. Significant variation in T c was observed; the SLs with the highest thickness of ferroelectric BaTiO 3 layers respective to nonferroelectric SrTiO 3 and CaTiO 3 layers have the highest T c , provided that BaTiO 3 layers remain highly strained. Strain relaxation causes a drastic decrease in the ferroelectric phase transition temperature. This work was partially supported by the National Science Foundation (Grant No. DMR-0705127), the U.S. Department of Energy (Grant No. DE-FG02-01ER45907), the DOE EPSCoR (Grant No. DE-FG02-04ER46142), and by the Research Corporation for Science Advancement (Grant No. 7134). H.N.L. was sponsored by the Division of Materials Sciences and Engineering, U.S. Department of Energy. Table I. Parameters of the SL samples studied. Sample label Structure d BT / d ST + d CT ratio Strain relaxation(%) T c from Raman(K) S1B1C1 ( CaTiO 3 ) 1 / ( BaTiO 3 ) 1 / ( SrTiO 3 ) 1 0.5 0 283 S4C2B4C2 ( CaTiO 3 ) 2 / ( BaTiO 3 ) 4 / ( CaTiO 3 ) 2 / ( SrTiO 3 ) 4 0.5 0 389 C2B4S2 ( SrTiO 3 ) 2 / ( BaTiO 3 ) 4 / ( CaTiO 3 ) 2 1.0 0 447 S2B4C2 ( CaTiO 3 ) 2 / ( BaTiO 3 ) 4 / ( SrTiO 3 ) 2 1.0 0 453 S1B3C1 ( CaTiO 3 ) 1 / ( BaTiO 3 ) 3 / ( SrTiO 3 ) 1 1.5 0.08 493 S2B6C2 ( CaTiO 3 ) 2 / ( BaTiO 3 ) 6 / ( SrTiO 3 ) 2 1.5 0.05 488 S2B8C2 ( CaTiO 3 ) 2 / ( BaTiO 3 ) 8 / ( SrTiO 3 ) 2 2.0 1.21 438 FIG. 1. Low-temperature Raman spectra for ( CaTiO 3 ) / ( BaTiO 3 ) / ( SrTiO 3 ) SLs and a 200 nm thick BaTiO 3 film at 30 K. FIG. 2. Temperature evolution of Raman spectra for ( SrTiO 3 ) 2 / ( BaTiO 3 ) 4 / ( CaTiO 3 ) 2 SL. Arrows indicate Raman peaks used for T c determination. FIG. 3. Temperature dependencies of normalized Raman intensities of TO 2 (solid symbols) and TO 4 (open symbols) phonons for three ( CaTiO 3 ) 2 / ( BaTiO 3 ) 4 / ( SrTiO 3 ) 2 SLs. The dashed-dotted lines are fits to a linear temperature dependence. FIG. 4. Results of T c determination from Raman data for all SLs studied. T c is plotted as a function of the thickness ratio d BT / ( d ST + d CT ) . Dashed line is a guide to the eyes.