B. Noheda, D.E. Cox, G. Shirane.

Physics Department, Brookhaven National Laboratory, Upton, NY 11973-5000, US.

J. A. Gonzalo.

Dept. Fisica de Materiales. U.A.M. Cantoblanco, 28049-Madrid, Spain.

L.E. Cross ,S-E. Park.

Materials Research Laboratory, The Pennsylvania State University, PA 16802-4800, US.
 
 
 
 

A previously unreported ferroelectric phase has been revealed on a highly homogeneous sample of PbZr_0.52Ti_0.48O₃ by high-resolution synchrotron x-ray powder diffraction measurements. At ambient temperature the sample has tetragonal symmetry (a_t= 4.037 Å, c_t= 4.138 Å), and transforms below ~250 K into a phase which, unexpectedly, has monoclinic symmetry (a_m= 5.717 Å, b_m= 5.703 Å, c_m= 4.143 Å, b= 90.53^o, at 20K). The intensity data strongly indicate that the polar axis lies in the monoclinic ac plane close to the pseudo-cubic [111] direction, which would be an example of the species m3m(12)A2Fm predicted on symmetry grounds by Shuvalov.
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

The solid solution system Pb(Zr_1-xTi_x)O₃ (PZT) has a complex phase diagram containing a number of materials which exhibit useful ferroelectric and piezoelectric properties. In particular, compositions near the morphotropic phase boundary (MPB) around x» 0.45-0.5 have attracted considerable interest for many years due to their high piezoelectric response. The PZT phase diagram as now accepted, was determined by Jaffe et al.¹ (Fig.1), and nearly four decades of study on the physical and structural properties of PZT have made possible the development of a phenomenological theory to explain the stability of the phases and the properties of PZT over the entire range of the phase diagram². Nevertheless, many questions still remain. It is generally accepted that for ferroelectric compositions with rhombohedral and tetragonal symmetry on the two sides of the MPB the polar axis are [111] and [001], respectively. However, very recently, based on neutron powder diffraction data for several ferroelectric rhombohedral compositions in the range of  x» 0.12-0.40, Corker et al.³ have proposed a model in which random <100> Pb displacements are superimposed on those along the [111] direction, the rhombohedral polar axis, which allows a much better structure refinement than achieved with a normal long-range order model incorporating anisotropic temperature factors.

    The morphotropic phase boundary (MPB) which separates rhombohedral Zr-rich PZT from tetragonal Ti-rich PZT is nearly vertical along the temperature scale, and many X-ray diffraction studies have been reported over this region^4-9. This boundary is not well defined since it appears to be associated with a phase coexistence region whose width depends on the compositional homogeneity and on the sample processing conditions^7-10. Cao and Cross¹¹ have modeled the width of this coexistence region based on the free energy differences between the tetragonal and rhombohedral phases, obtaining an inverse dependence with the particle size in a polycrystalline sample.
    On the present work we have utilized high-resolution synchrotron x-ray powder techniques to study the structure of a composition close to the MPB (x= 0.48) as a function of temperature, and we report the observation of a low temperature monoclinic phase in the PZT system.
    A PZT composition with a Ti content of 48% mol. was prepared by conventional solid-state reaction techniques using appropriate amounts of reagent-grade powders of lead carbonate, zirconium oxide and titanium oxide, with chemical purities better than 99%. Pellets were pressed and heated to 1250^oC at a ramp rate of 10^oC/min, and held at this temperature for 2 hours. During sintering, PbZrO₃ was used as a lead source in the crucible to minimize volatilization of lead. The sintered product was examined on a conventional laboratory x-ray diffractometer and found to be single phase within a detection limit of < 2%.
    High resolution Synchrotron x-ray powder diffraction measurements were made at beam line X7A at the Brookhaven National Synchrotron Light Source. An incident beam of wavelength 0.6896 Å from a Ge(111) double-crystal monochromator was used in combination with a Ge(220) crystal and scintillation detector in the diffraction path. The resulting instrumental resolution is about 0.01^o on the 2q scale, an order-of-magnitude better than that of a laboratory instrument. For measurements above room temperature, the pellet was mounted on a BN sample pedestal inside a wire-wound BN tube furnace. The furnace temperature scale was calibrated with a sample of CaF₂. The accuracy of the temperature was estimated to be within 10 K, and the temperature stability was ~2 K. For measurements below room temperature, the pellet was mounted on a Cu sample holder and loaded in a closed-cycle He cryostat, with an estimated temperature accuracy of 1K and stability of 0.1 K. Coupled q -2q scans were performed over selected angular regions with a 2q step interval of 0.005 or 0.01^o depending on the peak widths. The sample was rocked 1-2^o during data collection to improve powder averaging. The diffracted intensities were normalized with respect to the incident beam monitor.
    The evolution of the (111) and (220) reflections is shown as a function of temperature in Fig. 2. At the highest temperature reached (736 K), the material is cubic and the sharpness of the peaks (full-width at half-maximum, FWHM, ~0.02^o) demonstrates the excellent quality of the sample (Fig. 2a). From a Williamson-Hall plot¹² , we estimate a particle size of ~0.7 mm and a Dd/d of about 3´ 10^-4, corresponding to a compositional inhomogeneity, Dx, better than » 0.5%.

FIG. 2 Evolution of the pseudocubic reflections ( 111) and (220) from T= 736K (Fig 2a) to T= 20K (Fig. 2e).

    On cooling, the symmetry changes to tetragonal at ~660 K and remains tetragonal down to 300 K. There is no sign of any rhombohedral component as shown by the absence of a Bragg peak at the rhombohedral (200) position. However, some of the tetragonal peaks broaden as the temperature is lowered, specially the (111) and (202) reflections. Below room temperature the tetragonal (111) and (220) reflections become distinctly asymmetric, and at 210K the latter peak is clearly split into two roughly equal peaks, while the (111) peak has a low angle shoulder (Fig.2c). As the temperature is lowered further the splitting increases until at 20K the (202) is well-resolved into two peaks, and the (111) is seen to consist of a central stronger peak with weaker shoulders on both low-and-high-angle sides. The positions and intensity ratios of the peaks are very well described by a monoclinic cell in which a_m and b_m lie along the tetragonal[`1`10] and [1`10] directions (a_m» b_m» a_tÖ2), and c_m is close to the [001] axis (c_m» c_t), as illustrated in Fig. 3. The observed monoclinic distortion has b_m as the unique axis, and the angle between a_m and c_m is ~ 90.5^o at 20K.
    The lattice parameters are plotted in Fig.4 over the entire temperature range. At the low-temperature phase transition, a_m is slightly elongated with respect to tetragonal a_tÖ2, whereas b_m» a_tÖ2 continues to decrease as the temperature is lowered, and c_m» c_t appears to reach a broad maximum around the transition.

    A direct phase transition from tetragonal to monoclinic phase is rather uncommon and the existence of a monoclinic phase is likely to be a direct consequence of the proximity of the MPB. Consequently, one might expect the monoclinic phase to exist over a relatively narrow composition region. However, because of the asymmetrical peak broadening, which occurs between 300 and 210 K, we cannot altogether rule out the possibility of an orthorhombic phase in this region, even although we feel this is unlikely.
   In the tetragonal region of PZT the space group is P4mm, with the polar axis along [001], while in the rhombohedral region, with space group R3m, the polar axis is along the pseudocubic [111]. The most plausible space group for the new monoclinic phase is Cm, which is a subgroup of P4mm and R3m and allows the polar axis to lie anywhere between the tetragonal [001] and the rhombohedral [111] axes. A preliminary check of the peak intensities indicates that the cation displacements lie close to the monoclinic [`201] direction, i.e. close to the rhombohedral [111] axis. If this is the case, monoclinic PZT would be the first example of the ferroelectric species with P_x²= P_y²¹P_z², P_x², P_y², P_z²¹0, m3m(12)A2Fm predicted from symmetry by Shuvalov¹³. A more detailed investigation of the structure is currently in progress, and the extent of the monoclinic region will be investigated with additional samples containing slightly different amounts of Ti.
 
 

   We thank E. Sawaguchi for his efforts trying to locate his 1953 sample for this work, and Evagelia Moshopoulou for her helpful comments. Support by NATO (R.C.G.970037), Spanish CICyT (PB96-0037) and U.S. Department of Energy (contract No. DE-AC 02-98 CHI0886) is also acknowledged.

FIG. 4. Lattice parameters of PZT with a composition x=0.48 as a function of temperature.

References

¹B. Jaffe, W.R. Cook, and H. Jaffe, Piezoelectric Ceramics (Academic, London, 1971), p.136.
² M.J. Haun, E. Furman, S.J. Jang and L.E. Cross. Ferroelectrics 99, 13 (1989).
³ D.L. Corker, A.M. Glazer, R.W. Whatmore, A. Stallard and F. Fauth. J. Phys.:Cond. Matter 10, 6251 (1998).
⁴G. Shirane and K. Suzuki. J. Phys. Soc. Japan 7, 333 (1952)
⁵E. Sawaguchi. J. Phys. Soc. Japan 8, 615 (1953).
⁶P. Ari-Gur and L. Benguigui. Sol. Stat. Comm. 15 , 1077 (1974)
⁷ K. Kakewaga, O. Matsunaga, T. Kato and Y. Sasaki. J. Am. Ceram. Soc. 78(41), 1071 (1995).
⁸ J.C. Fernandes, D.A. Hall, M.R.Cockburn and G.N. Greaves. Nucl. Instr. and Meth. In Phys. Res. B97, 137 (1995).
⁹ M. Hammer, C. Monty, A. Endriss and Michel J. Hoffmann. J. Am. Cerm. Soc. 81, 721 (1998).
¹⁰ A.P. Wilkinson, J. Xu, S. Pattanaik and J.L. Billinge. Chem. Mat. 10, 3611 (1998).
¹¹W. Cao and L.E. Cross. Phys. Rev. B47 (9), 4825 (1993).
¹² G.K. Williamson and W.H. Hall. Act. Metall. 1, 22 (1953).
¹³ L.A. Shuvalov. J. Phys. Soc. of Japan 28, suppl., 38 (1970).