C⁻ Outgrowths in C⁺ Thin Films of LiNbO₃ on Al₂O₃-c

Abstract

Thin-film waveguides of LiNbO3 have been grown on Al2O3-c by off-axis rf magnetron sputtering. The films have been characterized optically by prism coupling measurements, crystallographically by x-ray diffraction, and morphologically by atomic force microscopy. We find that optical losses can be dominated by scattering from large outgrowths that litter the surface of the film. These outgrowths are c− grains imbedded in a c+ matrix. Although some grains nucleate c−, others have their polarity reversed from c+ to c− after nucleation. A model will be presented to explain the preferential nucleation of c+ grains on Al2O3-c. The c− grains grow much faster than the c+ ones because of attractive coulombic forces between the c− grains and the ionized Li and Nb species in the sputter plume.

LiNbO 3 thin film growth on (0001)-GaN

“…It was found that the postannealed LiNbO 3 films that were grown on GaN were in the c+ orientation, confirming the etching results discussed above. Again, this orientation is with the spontaneous polarization vector pointing from the substrate to the surface of the film, 14,19,20 which is opposite the orientation of the spontaneous polarization orientation in GaN and AlGaN/ GaN structures grown by MOCVD (typically Ga-terminated). 7,8 It has been reported that homoepitaxial MOCVD grown LiNbO 3 films have a c+ orientation regardless of the polarity of the LiNbO 3 substrate.…”

“…5. The growth of c+ (that is, with the spontaneous polarization vector of the films pointing from the substrate to the surface of the films 14 ) on sapphire has been reported for both sputter deposited 14 and MOCVD grown films. 16 It has been well established that the c− face of LiNbO 3 etches much faster in HF and HF: HNO 3 than the c+ face.…”

“…16 It has been well established that the c− face of LiNbO 3 etches much faster in HF and HF: HNO 3 than the c+ face. 14,16,17 The outgrowths of sputter deposited LiNbO 3 films on sapphire were shown to be c− grains in a c+ matrix by etching in HF. 14 To investigate the polarity of our films, the samples were etched in HF: HNO 3 and investigated with AFM following etching.…”

Toward High‐Quality Epitaxial LiNbO3 and LiTaO3 Thin Films for Acoustic and Optical Applications

“…Different methods have been used to characterize the ferroelectric domain structure in the LN and LT films: scanning nonlinear dielectric microscope, piezoelectric force microscopy (PFM), electrostatic force microscopy (EFM), pyroelectric current measurements, confocal second harmonic microscopy, and selective chemical etching by HF solutions . It was suggested, that the grains with Z − orientation grow much faster than Z + grains during RF sputtering However, selective etching by HF ( Z − domains are etched faster than Z + domains) is not sufficient to confirm that the outgrowths had opposite polarity and they were not parasitic phases or an inhomogeneity of Li concentration. The etching rate is highly dependent on doping and Li nonstoichiometry of the LiNbO 3 phase and it differs for different crystalline phases.…”

Correlation between interfacial structure and c-axis-orientation of LiNbO3 films grown on Si and SiO2 by electron cyclotron resonance plasma sputtering

“…Substituting thin films for bulk crystal has been a long-term challenge in crystal growth technology. Radio-frequency plasma sputtering [1][2][3], metalorganic chemical vapor deposition [4], molecular beam epitaxy [5], sol-gel process [6,7], liquid-phase epitaxy [8], and pulsedlaser deposition [9][10][11][12] have all been investigated as hetero-epitaxial methods on lattice-matched achieved through electron cyclotron resonance (ECR) plasma sputtering. Silicon-based-material systems, on the other hand, have recently been attracting a great deal of attention as the base for large-scale integrated optical circuits, since the large difference in the refractive indices of Si and SiO 2 allows significantly smaller planar optical circuits [13][14][15][16].…”