short communications J. Appl. Cryst. (2006). 39, 601–603 doi:10.1107/S0021889806021236 601 Journal of Applied Crystallography ISSN 0021-8898 Received 22 April 2006 Accepted 4 June 2006 # 2006 International Union of Crystallography Printed in Great Britain – all rights reserved Structural investigation of tetragonally stabilized ZrO 2 in a-Al 2 O 3 –ZrO 2 composites Apurba Kanti Deb, a Partha Chatterjee b and Siba Prasad Sen Gupta a * a Department of Materials Science, Indian Association for the Cultivation of Science, Jadavpur, Kolkata 700032, India, and b Department of Physics, Vivekananda Mahavidyalaya, Haripal, Hooghly 712405, India. Correspondence e-mail: msspsg@iacs.res.in A detailed structural analysis and phase-stabilization study of tetragonal (t-) ZrO 2 in -Al 2 O 3 –ZrO 2 composites has been performed via Rietveld refinement of X-ray powder diffraction data and Fourier transform infrared (FTIR) spectroscopy. It is observed that the tetragonal distortion () of t-ZrO 2 in the composites is similar to that of pure t-ZrO 2 . Although the average Zr—O bond distances are similar to those of other cation-doped zirconias, the X-ray and FTIR study gave no indication of Al 2 O 3 –ZrO 2 solid-solution formation. It is concluded that other factors, such as oxygen vacancies, play a significant role in the stabilization process. It may be argued that size-stabilized t-ZrO 2 is actually an oxygen-deficient form. 1. Introduction The tetragonal phase of ZrO 2 is usually stable above 1473 K. At temperatures below 1473 K, it is stabilized by doping with suitable cations, e.g. Y 3+ , Ca 2+ , Ce 4+ , Mg 2+ , etc., which randomly occupy the Zr site, while charge balance is achieved by an appropriate number of vacancies at the O sites. Other mechanisms, such as controlling crystallite size (Garvie, 1978), microstrain (Fagherazzi et al., 1980; Mitsuhashi et al., 1974) or oxygen vacancy in pure t-ZrO 2 (Fabreis et al., 2002), are also known to stabilize the t-ZrO 2 phase at room temperature. The fundamental issue of stability and its atomistic origin is yet to be settled. In particular, the mechanism of ‘size’ stabilization has not been solved. It has been argued that size-stabi- lized pure t-ZrO 2 is actually an oxygen-deficient form. In a recent report, Yashima & Tsunekawa (2006) showed that t-ZrO 2 nano- particles are actually an oxygen-deficient phase. It is well known that the mechanical properties of alumina ceramics can be considerably increased by the incorporation of fine ZrO 2 particles. In an earlier work (Deb et al. , 2006) t-ZrO 2 –-Al 2 O 3 composite powders were prepared by a combustion technique. It was observed that stabilized t-ZrO 2 particles could be grown inside the pores of the -Al 2 O 3 grains. For composites with an -Al 2 O 3 content (30 wt%), a tetragonal to monoclinic (t ! m) ZrO 2 transformation was observed when the sample was annealed at 1473 K. A critical size limit of 54 nm and microstrain of 2.5  10 3 were attributed to the stabilization of the t-ZrO 2 . In this work, a detailed structural analysis has been attempted to explore the viability of other stabilization mechanisms. 2. Experimental Al 2 O 3 –ZrO 2 composites with two different weight fractions were prepared by combustion technique from commercially available aluminium nitrate and zirconyl nitrate, and urea (Deb et al. , 2006). Two composites of initial nominal weight compositions of 63:35 and 29:58 wt% (-Al 2 O 3 :t-ZrO 2 ), referred to as samples 1 and 2, respectively, and the corresponding samples annealed at 1473 K were chosen for analysis. The X-ray powder diffraction patterns of the as-prepared and annealed samples were taken at room temperature using a Philips PW 1710 diffractometer with Ni-filtered Cu K radiation. Si powder was used for measuring the instrumental profile (van Berkum et al., 1995). The data were collected in a step-scan mode with a step size of 0.02  2 and a counting time of 5 s per step. The FTIR spectra of the samples were recorded by a Nicolet 750- IR Series-II Fourier transform infrared spectrometer. The scan range varied from 4000 to 400 cm 1 at a resolution of 4 cm 1 . 3. Results and discussion Fig. 1 shows the fitted X-ray diffraction patterns for the as-prepared and annealed composites. Table 1 shows the results of Rietveld refinement. It is evident from Table 1 that for the as-prepared sample 1, the major ZrO 2 phase is tetragonal (36%), whereas for sample 2, there exists in addition to t-ZrO 2 a monoclinic phase m-ZrO 2 with a weight fraction of approximately 12 wt%. The tetragonal distortion  [expressed as  = (1/4)(c 2 /a 2  1) 1/2 ] is obtained from the refined values of the lattice parameters a and c. The value of  is 0.052 for both samples. The result is in agreement with that for pure t-ZrO 2 ( = 0.05) and size-stabilized t-ZrO 2 ( = 0.047) (Kisi & Howard, 1998). For the t-ZrO 2 phase stabilized by cation doping, the value of tetra- gonal distortion  is a function of cation concentration. For the present composite samples, there is no measurable change in tetra- gonal distortion, indicating that Al 3+ ions do not replace Zr 4+ ions in the ZrO 2 lattice. It has, however, been reported that when -Al 2 O 3 –t- ZrO 2 composites are prepared in situ (Inamura et al., 1994; Ishida et al., 1994; Hong et al., 1998; Gao et al. , 1998), there is a possibility of Al 3+ doping. But that could not be established in the present case. The ionic radius of Al 3+ for eightfold coordination is 0.68 A ˚ [extra- polated from data of Shannon & Prewitt (1969)] which is much less compared with the ionic radius of Zr 4+ (0.84 A ˚ for eightfold coordi- nation), indicating that Al 3+ -doped ZrO 2 may not be stable.