Influence of Yttrium Doping on Grain Misorientation in Aluminum Oxide Junghyun Cho, * Helen M. Chan, * Martin P. Harmer, * and J. M. Rickman * Department of Materials Science and Engineering and Materials Research Center, Lehigh University, Bethlehem, Pennsylvania 18015 Oversized dopant ions such as yttrium and lanthanum segregate to grain boundaries and reduce the tensile creep rate of -Al 2 O 3 by 2–3 orders of magnitude. One explanation for this behavior is that the oversized segregants give rise to a ‘‘site-blocking’’ effect for grain boundary diffusion. It has also been speculated that the dopant ions modify the grain boundary structure in alumina and reduce the creep rate by promoting the formation of special (e.g., coincidence site lattice (CSL)) grain boundaries. In order to test the latter hypothesis, we have used electron backscattered Kikuchi diffraction to characterize the misorientation and special grain boundary distribution for undoped and 1000-ppm-yttrium-doped alumina. The results show that the grain boundary structure in alumina (as characterized by the frequency of selected CSLs and misorientation distribution) was not significantly changed by the addition of yttrium, indicating that creep retardation results mainly from site-blocking. I. Introduction I T HAS been reported that the tensile creep rate of -Al 2 O 3 can be dramatically reduced by the addition of 1000 ppm Y 2 O 3 . 1,2 This was a surprising result since most previous studies of isovalent dopants such as Cr 3+ and Fe 3+ (whose ionic radii are similar to Al 3+ ) have revealed no improvement in the creep properties. 3,4 In addition, it is well known that yttrium (Y) doping reduces the growth rate of polycrystalline alumina scales formed by oxidation. 5,6 Segregation of Y to grain boundaries in -Al 2 O 3 , which arises from a large size (radius) mismatch between Al 3+ (0.51 Å) and Y 3+ (0.89 Å), has been clearly demonstrated in many studies. 7–10 It was suggested by Cho et al., therefore, that the observed reduction in creep rate was the result of a ‘‘site-blocking’’ effect of oversized segregant ions on grain boundary diffusion. 2 This hypothesis is consistent with recent studies on the self-diffusivity in -Al 2 O 3 which have shown that Y-doping decreases oxygen diffusivity at grain boundaries, with little effect on bulk diffusion. 11–15 Furthermore, Fang et al. observed that the densification rate of alumina was significantly reduced by the presence of Y, and the data were consistent with a grain boundary diffusion- controlled mechanism. 16 An alternative school of thought to the ‘‘site-blocking’’ model is that the presence of the dopant ions modifies the grain boundary structure. The argument is that certain types of ‘‘special’’ boundaries are favored, which play a limiting role in the transmission of grain boundary sliding due to the difficulty in accommodating lattice dislocations. 17,18 In fact, there is evi- dence that special grain boundaries in metals can decrease the grain boundary diffusivity and deformation rate; 19–21 however, there is little direct evidence of such effects in ceramics. Using transmission electron microscopy (TEM), Lartigue et al. re- ported that magnesium (Mg) doping increased the proportion of special (coincidence site lattice (CSL) and coincidence axis direction (CAD)) grain boundaries in alumina. 22 Further, the role of Y on the grain boundary structure was investigated in Mg and Y codoped alumina, where it was speculated that the proportion of near-coincidence grain boundaries increased after creep deformation, 23 consistent with the observations in me- tallic systems. 17 Thus, an alternative role of Y is that it in- creases the proportion of special grain boundaries in alumina. The aim of this study was to test this postulate by obtaining grain misorientation information for both undoped and Y- doped alumina utilizing electron backscattered Kikuchi diffrac- tion (EBKD) in the scanning electron microscope (SEM). 24,25 This technique has several advantages over the TEM; most importantly, it can characterize a large number of grains in a single scan, and in addition specimen preparation is relatively easy. II. Experimental Procedure Undoped (designated ‘‘Pure A’’) and Y-doped alumina (des- ignated ‘‘YA-UN’’) used in this study were prepared from commercial powders of -Al 2 O 3 (AKP-HP and AKP-53, re- spectively; Sumitomo Chemical America, New York, NY). 1,26 The powder was mixed with a suitable proportion of yttrium nitrate solution to achieve dopant levels of 1000 ppm (Y/Al ion). Fully dense materials were obtained by hot-pressing un- der vacuum at 50 MPa; the conditions were 30 min at 1475°C and 15 min at 1400°C for the Y-doped and undoped alumina, respectively. The resultant grain size was around 2–3 m for both materials. A specimen of the Y-doped alumina was creep- ruptured at 1300°C under a tensile stress of 50 MPa in air with a final strain of ∼5.5% (designated ‘‘YA-DEF’’). For the investigation of grain misorientation, the surfaces of bulk specimens were polished down to a 1 m finish with diamond slurry, followed by polishing with 0.02 m colloidal SiO 2 in the vibratory polisher (Vibromet II, Buehler, Lake Bluff, IL). In order to avoid charging, a thin carbon layer (<5 nm) was deposited on the polished surface. The (Kikuchi) diffraction patterns were obtained in the SEM (Camscan Series II) at an accelerating voltage of 25 kV and a working distance of 33 mm. The backscattered Kikuchi patterns were captured using a silicon intensified target (SIT) camera (Customs Camera 27, Somerset, UK) and its live image was improved by subtracting background noises through an image processor (Hamamatsu Argus 10). Subsequent analysis of the EBKD patterns was carried out with the aid of computer software (TSL, Inc., Draper, UT). The total number of grains analyzed C. A. Handwerker—contributing editor Manuscript No. 190583. Received November 10, 1997; approved July 28, 1998. Supported by the U.S. Air Force Office of Scientific Research under Contract No. F49620-94-1-0284 (monitored by Dr. A. Pechenik). * Member, American Ceramic Society. J. Am. Ceram. Soc., 81 [11] 3001–3004 (1998) J ournal 3001