Influence of Peptization and Ethanol W ashing on the Pore-Structure Evolution of Sol–Gel-Derived Alumina Catalyst Supports Jalajakumari Nair, †,‡ Padmakumar Nair, † Jan G. Van Ommen, Julian R. H. Ross, § and Anthonie J. Burggraaf Faculty of Chemical Technology, University of Twente, 7500 AE, Enschede, The Netherlands Fujio Mizukami National Institute of Materials and Chemical Research, 1-1 Higashi, Tsukuba, Ibaraki, Japan Alkoxide-derived boehmite precipitates were subjected to different post-precipitation treatments to obtain a peptized sol, an unpeptized slurry, and an ethanol-washed gel pre- cipitate. These materials were dried further to obtain the corresponding gels. Changes in the surface area, porosity, and pore size with temperature and the phase-transforma- tion behavior from -alumina to -alumina were studied using X-ray diffractometry, differential scanning calorim- etry, and physical adsorption/desorption measurements. Gels that were obtained from the peptized sol retained the lowest porosity, in comparison to the ethanol-washed gel and the gel that was prepared from the aqueous slurry. I. Introduction O NE of the most commonly used support materials for cata- lysts is -alumina. A drastic reduction of the surface area of -alumina is reported to occur when the alumina is heated at temperatures >1000°C. One of the reasons for this reduction is the enhanced sintering that occurs during the phase transition to -alumina. 1 Boehmite gel is one of the most-studied ceramic precursors for the preparation of -alumina. 2–6 In this paper, we have investigated the influence of purely physical methods of changing the pore-structure stability of alumina by changing the texture of the boehmite gel, which is the precursor for -alumina. The major advantage of using purely physical tech- niques to stabilize the pore structure is that there is no influence on the chemistry of the system; therefore, no change in the catalytic behavior is expected. There are several studies in the literature that address the washing of oxide gels, including boehmite gel, with alcohols. 7–11 From the perspective of de- veloping alumina-based high-temperature-stable catalyst supports, phase transformation from transition alumina to -alumina is very important. However, none of the above- referenced studies have examined the influence of post- precipitation treatment on the phase transformation of transi- tion alumina to -alumina. In this paper, we compare the texture and phase-transformation behavior of ethanol-washed, water-washed, and peptized boehmite gels. II. Experimental Procedure The alumina precursor used in this work was prepared by following methods that have been described elsewhere. 12,13 Aluminium-tri-sec-butoxide was hydrolyzed by adding it drop- wise to water kept at 80°C under vigorous stirring. To prevent the hydrolysis of the alkoxide by water present in the atmo- sphere, the hydrolysis was performed under a nitrogen atmo- sphere. The water-to-alkoxide ratio was kept at 70:1. The so- lution was then heated to remove the excess butanol present and divided into three portions. One portion was dried imme- diately (hereafter called sample UP). The second portion was transferred to a three-necked round-bottomed flask and 1M nitric acid solution was added with vigorous stirring. This so- lution was then peptized by a refluxing at a temperature of 80°C for 12 h (hereafter called sample PP). The third portion was washed with ethanol and is hereafter called sample EW. All the samples were dried at 40°C and 20% relative humidity in a climate chamber (Model VTRK 300, Heraeus Votsch GmbH, Ballingen, Fommern, Germany) with an air flow rate of 0.5 m/s over the sample surfaces. Wind screens were used to create laminar flow. The final thickness of the dried gel pieces was ∼150 m. Each calcination was conducted in a furnace (Vecstar Fur- naces, Chesterfield, U.K.) in static air, using samples with masses of ∼4 g that were contained in an alumina crucible. The sample was heated to the required temperature using a heating rate of 3°C/min and maintained at that temperature for 15 h and cooled at a rate of 3°C/min. Thermal analyses were conducted in a thermal analysis sys- tem (Model STA 625 PL, Thermal Science System, Polymer Laboratories, Amherst, MA) with a heating rate of 20°C/min in flowing air (20 cm 3 /min). Alumina was used as the standard. The sample weight was 30 ± 3 mg. X-ray diffractometry (XRD) analyses were performed using an X-ray diffractometer (Philips Research Laboratories, Eind- hoven, The Netherlands) with nickel-filtered CuK radiation. Step-scan techniques were performed in the range of 30°–70° 2 and the scans were taken in steps of 0.15° 2, with a count- ing time of 10 s. The sample was rotated around the vertical axis to improve the counting statistics. The crystallite sizes (D hkl ) were calculated using the Scherrer relation: D hkl = K   B hkl cos   (1) where  is the wavelength and B hkl is the full width at half maximum, corrected for the K 1 and K 2 doublet and instru- mental broadening. The value of K varies from 0.9 to 1.4, and the value used in this study was 0.9; the value of 0.9 was used by Okubo et al. 14 to calculate the crystallite size of boehmite samples. G. L. Messing—contributing editor Manuscript No. 190530. Received December 1, 1997; approved June 16, 1998. Supported by Shell Research (Amsterdam, The Netherlands). ² Present address: National Institute of Materials and Chemical Research, 1-1 Hi- gashi, Tsukuba, Ibaraki, Japan. ‡ Author to whom correspondence should be addressed. § Present address: Department of Industrial Chemistry and Chemical Engineering, University of Limerick, Plassey Park, Limerick, Ireland. J. Am. Ceram. Soc., 81 [10] 2709–12 (1998) J ournal 2709