Thin Solid Films 475 (2005) 283 – 286 www.elsevier.com/locate/tsf
Structural and dielectric properties of artificial PbZrO3 /PbTiO3 superlattices grown by pulsed laser deposition Taekjib Choi, Jaichan Lee* Department of Materials Science and Engineering, Center for Advanced Plasma Surface Technology, Sung Kyun Kwan University, Suwon 440-746, Korea Available online 11 September 2004
Abstract PbZrO3 (PZO)/PbTiO3 (PTO) artificial superlattices were fabricated by pulsed laser deposition, and their structural and dielectric properties were investigated. Epitaxial PbTiO3 (PTO) and PbZrO3 (PZO) layers were alternately deposited on La0.5Sr0.5CoO3 (LSCO) (100)/ MgO (100) substrate at 500 8C with various stacking periods from 1 to 100 unit cells. The (100) plane spacing (d 100) of the superlattices increased with decreasing the stacking period, and the corresponding dielectric constant was drastically improved. The dielectric constant of the superlattice reached 800 at a stacking period of 1 unit cell/1 unit cell (PZO1/PTO1). With small stacking periods (PZO1/PTO1 and PZO2/ PTO2 unit cells), superlattices exhibited the capacitance–voltage hysteresis, i.e., ferroelectric behavior. D 2004 Elsevier B.V. All rights reserved. Keywords: PZO/PTO artificial superlattices; Pulsed laser deposition; Large dielectric constant; Ferroelectric
1. Introduction Recently, perovskite-type materials have been studied for various applications in ferroelectric memories , microelectromechanical systems (MEMS)  and tunable microwave device  because of their versatile properties, such as ferroelectricity and piezoelectricity. Dielectric artificial oxide superlattices with the perovskite-type structure have received considerable attention since the superlattices have potential to create materials whose properties are improved from solid-solution film. Many studies have been performed to fabricate various superlattices such as BaTiO3/SrTiO3 [4– 6], KTaO3/KNbO3 , Bi-layered oxide  and Pb-based oxide [9,10] superlattices and to enhance their dielectric properties. Specifically, BaTiO3/SrTiO3 and Pb-based oxide superlattices are of great interest since the consisting materials, i.e., BaTiO3, SrTiO3, PbTiO3 and PbZrO3, have been extensively studied and used in solid-solution material for various applications [4,5,9]. It was reported that dielectric
* Corresponding author. Tel.: +82 31 290 7397; fax: +82 31 290 7410. E-mail address: [email protected]
(J. Lee). 0040-6090/$ - see front matter D 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.tsf.2004.07.024
constant and remanent polarization of these superlattices increased with decreasing a stacking periodicity. We have reported that large dielectric constant and extremely large nonlinearity of the dielectric constant have been obtained in BaTiO3/SrTiO3 superlattices by varying the stacking period . These dielectric behaviors were explained to be due to lattice strain-developed superlattice [4,5]. In our previous study, maximum dielectric constant of BaTiO3/SrTiO3 superlattice was achieved via strain manipulation by varying a stacking sequence and period . Those results suggest that lattice strain is an important factor in dielectric behavior of oxide artificial superlattices. Kanno et al.  have succeeded in fabricating artificial PbZrO3 (PZO)/PbTiO3 (PTO) superlattices on Pt-coated MgO substrate using a multi ion-beam sputtering technique. However, the variation in the stacking sequence was limited to more than 5 unit cells, which is not small enough to investigate the dielectric behavior of the artificial superlattice fabricated by layer-bylayer growth technique. In this study, we have fabricated epitaxial PZO/PTO superlattices with stacking period from 1 to 100 unit cells and investigated the dielectric property of the artificial superlattice with various stacking sequences including small stacking periods (PZO1/PTO1 and PZO2/ PTO2 unit cells).
T. Choi, J. Lee / Thin Solid Films 475 (2005) 283–286
2. Experimental The artificial PZO/PTO superlattices were grown on (La0.5,Sr0.5)CoO3/MgO(100) substrate by multitarget pulsed laser deposition (PLD, k=248 nm, KrF excimer laser). A laser with an energy density of 2 J/cm2 is focused on the PbZrO3 and PbTiO3 ceramic targets with excess 10 mol% of PbO alternately with pulse frequency of 1 Hz. For the formation of epitaxial PZO/PTO superlattices without thermal interdiffusion of PZO and PTO layers, the relatively low substrate temperatures are required. The epitaxial PZO and PTO layers with same number of unit cells were alternately deposited at a substrate temperature of 500 8C with an oxygen ambient of 100 mTorr. The stacking period was varied in the range from 1 unit cell of PTO layer and 1 unit cell of the PZO layer (PZO1/PTO1) to PZO100/PTO100. Total thickness of the multilayered films was fixed at about 80 nm. Fifty-nanometer-thick conductive (La0.5,Sr0.5)CoO3 was deposited at 600 8C with an oxygen ambient of 100 mTorr as top and bottom electrodes. X-ray diffraction (XRD) was carried out to examine the crystal structure of the artificial superlattice. Electrical properties were measured by an impedance analyzer (HP4194A) and a RT 66A ferroelectric test system at room temperature.
3. Results and discussion Fig. 1 shows X-ray diffraction (XRD) patterns of the [PZO10/PTO10]10 artificial lattices [i.e., the artificial lattice consisting of 10 unit cells (or called supercells) with a stacking period of PZO10/PTO10] grown on LSCO/MgO (100) at substrate temperatures ranging from 450 to 540 8C in an oxygen pressure of 100 mTorr. At a substrate temperature of 450 8C, the PZO/PTO superlattice was not obtained and amorphous phase was formed. As the substrate temperature increased, PZO/PTO superlattice was obtained
Fig. 1. XRD patterns of [PZO10/PTO10]10 films deposited on LSCO/MgO substrate at various substrate temperature.
Fig. 2. XRD patterns of PZO/PTO superlattices with various stacking period varying from 1/1 to 100/100 unit cells.
in the range of 480–540 8C. Structural feature of a superlattice is represented by satellite peaks, which are symmetric around a main Bragg reflection. Satellite peaks result from a periodic modulation in lattice spacing or composition [11,12]. The superlattice grown at 480 8C shows that there are clearly first- and second-order satellite peaks present around main peak, compared to films grown above 510 8C. As the temperature increased further, the satellite peaks began to extinct. Elevated temperatures above 510 8C seem to degrade the superlattice possibly by interdiffusion or interface roughening. Fig. 2 shows X-ray diffraction patterns of PZO/PTO superlattices with various stacking periods from PZO1/ PTO1 to PZO100/PTO100. Satellite peaks in the vicinity of main (100) peak were observed at stacking periods in the range of PZO5/PTO5 to PZO25/PTO25, indicating that the PZO/PTO superlattices were obtained. The simulation on X-ray diffraction showed that satellite peaks were increasingly distant from main peak as the stacking period decreased. However, the diffraction intensity of the satellites becomes weak. When the stacking period becomes small, i.e., 1 or 2 unit cell, the satellite peaks was too low to be observed due to very low diffraction intensity. Thus, the PZO/PTO superlattices with small stacking periods (i.e., PZO1/PTO1 and PZO2/PTO2) exhibited main (100) peak without satellite peak. When the stacking period increased (i.e., PZO50/PTO50 and PZO100/PTO100), broad peaks were observed instead of the main peak and satellite peaks. When the stacking period is large, the PZO and PTO layers become thick, leading to a multilayer heterostructure instead of the artificial superlattice. Therefore, the broad diffraction peaks were assigned to the overlap of (100) and (001) peaks of each PTO and PZO layers.
T. Choi, J. Lee / Thin Solid Films 475 (2005) 283–286
The variation of (100) plane spacing d 100 [the average of the (100) plane spacing of the PZO/PTO superlattice] with various stacking periods is illustrated in Fig. 3. The stacking period D is represented by N 1c 1+N 2c 2, where N 1 and N 2 are the number of unit cells of the PZO and PTO layers, respectively. c 1 and c 2 are the lattice constants of the PZO and PTO layer along the surface normal direction, respectively. Then, the average interplanar spacing of the PZO/ PTO superlattice (i.e., d 100) is defined to be D/(N 1+N 2). Therefore, the d 100 value was determined from the main (100) peak in the X-ray diffraction patterns. The average interplanar spacing increased from 0.4034 to 0.4078 nm with decreasing the stacking period. All d 100 values of superlattices were larger than that (0.4032 nm) of bulk Pb(Zr0.5,Ti0.5)O3 solid solution in bulk form . The increase in d 100 with the stacking period is attributed to inplane strain effect due to a lattice mismatch between the PZO and PTO layers. This in-plane strain is caused by mutual mechanical constraint of PZO and PTO layers, while the superlattice maintains lattice coherence between the PZO and PTO layers. The largest d 100 value was obtained in PZO/PTO superlattice with a smallest period (PZO1/PTO1). Fig. 4 shows dielectric constants and loss for the PZO/ PTO superlattices as a function of the stacking period. The dielectric constant of the superlattice increased from 142 to 800 as the stacking period decreased. It is noted that dielectric constant of the superlattice with stacking periods (PZO1/PTO1 and PZO2/PTO2) was larger than that of a PZT solid-solution film prepared under similar deposition conditions (i.e., the substrate temperature of 650 8C and oxygen pressure of 100 mTorr). Dielectric constant of the PZT solidsolution film was 577. It is also noted that progressive enhancement of the dielectric constant of the superlattice is obtained without substantial change in dielectric loss. Kanno et al.  suggested that the enhanced dielectric constant of PZO/PTO superlattices might be caused by the interlayer stress and/or the interaction of electric dipoles between ferroelectric PTO and antiferroelectric PZO layer. The lattice distortion was also attributed to the enhanced
Fig. 3. The (100) plane spacing of PZO/PTO superlattices as a function of stacking period in the range of 1/1–10/10 unit cells.
Fig. 4. The change of dielectric constant and loss for PZO/PTO superlattices with various stacking period.
dielectric constant and its nonlinearity of BaTiO3/SrTiO3 superlattice [4,5]. On the other hand, Catalan et al.  suggested that dielectric enhancement is not a fundamental property of artificially engineered nanometer-scale heterogeneities, but is rather an artifact with increased carrier mobility (i.e., Maxwell–Wagner effect). The dielectric loss did not increase with decreasing the stacking period or consequent lattice strain while the dielectric constant increased, as shown in Fig. 4. Therefore, it is suggested that the enhanced dielectric constant of the PZO/PTO superlattice at small stacking periods is associated with primarily the lattice strain developed in the PZO and PTO layers. The dielectric dispersion of PZO/PTO superlattice was negligible excluding the Maxwell–Wagner effect. Fig. 5 shows capacitance–voltage (C–V) curves for PZO/ PTO superlattice with the stacking period of 1/1 and 2/2 unit cells, respectively. The C–V curves measured at 1 MHz exhibit a conventional (hysteretic) buffer fly-loop due to ferroelectric polarization. These results suggest that the PZO/PTO superlattices have a good ferroelectricity and large dielectric constant at small stacking periods (i.e., PZO1/PTO1 and PZO2/PTO2 unit cells). Remanent polar-
Fig. 5. Capacitance–voltage characteristics of [PZO1/PTO1]100 and [PZO2/ PTO2]50 superlattices.
T. Choi, J. Lee / Thin Solid Films 475 (2005) 283–286
ization (2Pr) of PZO/PTO superlattices measured by polarization-electric field (P-E) hysteresis measurement was 38.7 AC/cm2 at a stacking period of PZO2/PTO2.
R&D Project for Nano Science and Technology (Project M1-0212-29-0001).
References 4. Conclusions We have prepared artificial PZO/PTO superlattices on LSCO/MgO substrate by pulsed laser deposition. We have obtained the epitaxial PZO/PTO superlattice with various stacking periods. Progressive enhancement of the dielectric constant of PZO/PTO superlattices was accompanied by the expansion of d 100 values with decreasing the stacking periodicity. The dielectric constant was significantly enhanced at stacking periods of a few unit cells: the dielectric constant reached 800 and 732 at PZO1/PTO1 and PZO2/PTO2, respectively. It is suggested that the observed improvement in dielectric constant and ferroelectric behavior may be affected by the lattice strain developed in the PZO/PTO superlattice.
Acknowledgement This work is partially supported by the Korea Ministry of Science and Technology through the National
 W. Choi, T. Sands, J. Appl. Phys. 93 (2003) 4761.  R. Maeda, Z. Wang, J. Chu, W. Lin, Proc. SPIE Int. Soc. Opt. Eng. 4397 (2001) 261.  B.H. Park, Q. Jia, Jpn. J. Appl. Phys. Part 1 41 (2002) 7222.  J. Kim, L. Kim, Y. Kim, D. Jung, J. Lee, Appl. Phys. Lett. 80 (2002) 3581.  O. Nakagawara, T. Shimuta, T. Makino, S. Arai, H. Tabata, T. Kawai, Vacuum 66 (2002) 397.  L. Kim, J. Kim, D. Jung, Y. Kim, J. Lee, Appl. Phys. Lett. 82 (2003) 2118.  J. Sigman, D.P. Norton, H.M. Christen, P.H. Fleming, L.A. Boatner, Phys. Rev. Lett. 88 (2002) 097601.  H. Tabata, H. Tanaka, T. Kawai, Jpn. J. Appl. Phys. Part 1 34 (1995) 5146.  I. Kanno, S. Hayashi, R. Takayama, T. Hirao, Appl. Phys. Lett. 68 (1996) 328.  A. Erbil, Y. Kim, R.A. Gerhardt, Phys. Rev. Lett. 77 (1996) 1628.  I.K. Schuller, Phys. Rev. Lett. 44 (1980) 1597.  Y. Ishibashi, N. Ohashi, T. Tsurumi, Jpn. J. Appl. Phys. Part 1 39 (2000) 186.  B. Noheda, D.E. Cox, G. Shirane, R. Guo, B. Jones, L.E. Cross, Phys. Rev., B 63 (2000) 014103.  G. Catalan, D. O’Neill, R.M. Bowman, J.M. Gregg, Appl. Phys. Lett. 77 (2000) 3078.