In Situ HT-ESEM Observation of CeO 2 Grain Growth During Sintering Renaud Podor, † Nicolas Clavier, Johann Ravaux, Laurent Claparede, and Nicolas Dacheux ICSM-UMR 5257 CEA-CNRS-UM2-ENSCM, Site de Marcoule, Baˆt. 426, BP 17171, F-30207 Bagnols sur Ce`ze, France The CeO 2 sintering and grain growth kinetics were determined from in situ experiments performed using an environmental scanning electron microscope (ESEM) equipped with a hot stage in the temperature range 1000°C–1300°C. Videos were recorded during 8 h at various holding temperatures where the grain growth and grain elimination process were clearly observed at the grain scale with a 2–10 nm resolution. Data were extracted at both sample and grain scales to characterize the microstructural evolution (grain growth and grain elimina- tion). The grain growth kinetic was described using a parabolic law. The activation energy determined using this method for pure CeO 2 grain growth (290 ± 40 kJ/mol) was in good agree- ment with that obtained using the Dorn’s method based on dilatometric measurements (330 ± 30 kJ/mol). The microstruc- tural modifications observed using the in situ HT-ESEM experimental method were compared with computed simula- tions and revealed a good correlation between the experimental data and the models previously developed. I. Introduction T HE microstructure of ceramic materials often present a significant impact on their physico-chemical properties such as mechanical behavior, 1,2 electric, and thermal conduc- tivity, 3,4 chemical durability 5–7 or optical properties. 8 In these conditions, the monitoring of ceramic microstructure consti- tutes an important challenge to prepare optimized materials dedicated to a wide range of applications. Nevertheless, the methods traditionally available to study the final stages of sintering (i.e., elimination of open and closed porosities through grain growth mechanisms) 9 usually require a signifi- cant amount of experimental work, often based on the repet- itive annealing and observation of sintered samples. 10 However, the study of microstructural evolution using in situ techniques is currently developed and offers direct access to new information. Particularly, it is now possible to observe sintering processes at high temperature and to deduce kinetic parameters from the use of X-ray microto- mography 11,12 or transmission electron microscopy. 13–15 Also, the use of environmental scanning electron microscope coupled with a hot stage (HT-ESEM) allows the recording of in situ images of various samples and the observation of microstructure evolution over long periods. 16 This technique thus appears to be particularly suitable for the study of cera- mic sintering. 6,17–21 Indeed, it offers the possibility to follow directly the microstructural modifications of a sample from the micrometer scale—i.e., at the grain scale—to the millime- ter scale for long durations. The same group of grains can be observed during the whole experiment. As a consequence, it is possible to follow the evolution of single grain morphology with a 30 s–1 min time resolution (time that is necessary to record one image). This type of observation cannot be per- formed using the classical methods for sintering study. One must consider that HT-ESEM analysis is carried out exclu- sively at the surface and the topographic information can result differently, with respect of the material bulk, in con- trast to classical micrographic analyses that are usually car- ried out on polished cross sections of the inner material at a fair distance from the surface. The surface of the sample can present a different behavior during the process because of the sintering stress generated results unbalanced. Such effects were not considered in the present study. This work aims to develop the use of HT-ESEM for the direct observation of grain growth during CeO 2 sintering. The choice of this material was mainly motivated by its wide range of applications, rising from its potential use in SOFC technology 22,23 to its frequent use as a surrogate of PuO 2 in the nuclear industry. 24–26 In the present study, the choice of CeO 2 as a model material was driven by our interest for the development of nuclear fuel surrogates and the possibilities to easily control the ceramic microstructure that can affect significantly the behavior of the material in use, particularly during irradiation, leaching or dissolution. 6,7 The microstruc- tural modifications of dense samples were then investigated at various temperatures, leading to the rapid determination of aver- age grain sizes. Moreover, original data that can be derived from the observation of individual grains, such as individual growth or elimination rates were also determined for the first time. II. Experimental Section (1) Initial Powder Characterization Powdered CeO 2 was obtained through the initial precipita- tion of cerium(III) oxalate, Ce 2 (C 2 O 4 ) 3 10H 2 O. For this pur- pose, cerium(III) nitrate solution (99.99% pure; Aldrich, St. Louis, MO) was added dropwise to a large excess of oxalic acid. The precipitate instantly formed and was then separated by centrifugation, washed with deionized water and ethanol, dried, then finally fired at T = 500°C for 6 h. This resulted in the complete decomposition of the oxalate entities and pure CeO 2 nanopowders were formed, composed of aggregates, which were 40–50 nm in diameter (average crystallite size of 15 nm). The specific surface area of this powder, determined using a Micromeritics ASAP 2020 (Verneuil en Halatte, France) apparatus (BET method based on nitrogen adsorption at 77 K) was found to be ~47 ± 3m 2 /g. The XRD patterns of the final oxides exhibited all the characteristic lines of the FCC fluorite-type structure (space group Fm 3m) according to JCPDS file 81-0792. 27 (2) Sample Preparation Pellets of 5 mm in diameter were uniaxially pressed under 200 MPa pressure (green density = 58% of the calculated density). Approximately 200–500 lm diameter parts of this disk were considered to perform the HT-ESEM in situ exper- iments. The small size of the samples ensured a fast tempera- ture homogenization of the ceramic. Furthermore, the displacements of the sample that can be generated by its F. Wakai—contributing editor Manuscript No. 31018. Received February 09, 2012; approved July 18, 2012. † Author to whom correspondence should be addressed. e-mail: renaud.podor@ cea.fr 3683 J. Am. Ceram. Soc., 95 [11] 3683–3690 (2012) DOI: 10.1111/j.1551-2916.2012.05406.x © 2012 The American Ceramic Society J ournal