sults showed that the oxides, 299, 283 and 354 m2/g had relative specific surface areas of 5, 10 and 20 wt% respectively at 500˚C. According to their study, the specific surface area increased from 283 to 354 m2/g for the mixed oxides at 10 - 20 wt% of zirconia, opposite from the results obtained by Chen et al., who reported that when zirconia goes from 10% to 20%, the surface area decreased from 330 to 290 m2/g at 500˚C.
Klimova et al.  investigated a series of Al2O3-ZrO2 mixed oxides with variable amounts of ZrO2 calcined at 300˚C, 400˚C, 600˚C and 800˚C. They observed that the specific surface area of mixed oxides at 600˚C, with a molar ratio was about 389 m2/g, whereas the pure alumina sample yielded a surface area of 349 m2/g, which suggested that the incorporation of small amounts of zirconia into alumina presumably would increase the specific surface area, but the area actually decreased when more zirconium was added to the alumina. Likewise, smaller average diameter pores in the mixed oxides were obtained when the molar relation increased in comparison with the pure alumina oxide.
Dominguez et al.  prepared a series of Al2O3-ZrO2 solid solutions at different molar ratios. The results obtained by these authors showed that when calcined at 550˚C, the mixed oxides had a specific surface area of between 200 and 250 m2/g. A maximum specific surface area was obtained with a molar ratio X = 0.2. Their results agreed with those obtained by Klimova et al. .
In the synthesis and characterization of Al2O3-ZrO2 mixed oxides obtained by the sol-gel method, the supports were prepared using several solvents and different chemical precursors by modifying the process reported by Yoldas  for the synthesis of alumina by the sol-gel route.
In the present research, the objective is to investigate the effect of three different solvents on the textural and chemical properties of Al2O3-ZrO2, synthesized by the solgel method. The effect of four different water-addition rates was also investigated. The lowest value of the water-addition rates resulted in a homogeneous pore-size distribution. The research will allow the development of tailor-made supports to be used in particular hydrocarbon reactions.
In the present study, Al2O3-ZrO2 supports (1:1 molar ratio) were obtained by the sol-gel method, using a molar ratio of 80/30/0.3/1 (alcohol/water/acid/alkoxide). Aluminum tri-sec-butoxide (ATB, 97%, Aldrich) and anhydrous ethanol (94.2% J. T. Baker) were mixed in a reactor vessel with continuous stirring for 20 minutes under an inert atmosphere of nitrogen. In another reactor vessel zirconium (IV) propoxide (ZrP, 70 wt%, Aldrich) and the appropriate amount of ethanol were mixed under an inert atmosphere of nitrogen and then finally stirred for 20 minutes. Afterwards, both solutions were blended, by stirring the mixture vigorously for two hours until a clear solution at room temperature was obtained. An aqueous solution of HNO3 (65%, J. T. Baker) was added at four addition rates (0.06, 0.1, 0.15 and 0.26 mL/min) to the reactor vessel containing the solution of ethanol and alkoxides under continuous stirring. The gel was aged for 24 hours and dried in a vacuum oven. The starting heat treatment step consisted in drying the powder to 120˚C for 1 hour at a heating rate of 1˚C/min starting from room temperature. The resulting powder was finally calcined up to 700˚C for 2 hours at a heating rate of 0.5˚C/min. Figure 1 shows the schematic diagram for the experimental set-up for the preparation of Al2O3-ZrO2 supports by the sol-gel method. The same procedure was used for the mixed oxides obtained with isopropanol and isobutanol.
The mixed-oxide supports were characterized using a variety of techniques, which are mentioned below.
Figure 1. Schematic diagram of experimental set-up for the preparation of Al2O3-ZrO2 supports by the sol-gel method.
2.2.1. N2 Adsorption-Desorption
The specific surface areas of the calcined samples were calculated from N2 adsorption-desorption acquired data at liquid N2 temperature on a Micromeritics ASAP 2010 instrument, and the pore size distributions were determined by the Barrett-Joyner-Halenda (B-J-H) method applied to the desorption profile-branch of the nitrogen isotherm.
2.2.2. Infrared Spectroscopy
FT-IR spectroscopy was performed with Perkin Elmer Spectrum 100 equipment on pellets of the mixed oxides at the 400 - 4000 cm–1 wave number range. For this analysis the pellets were elaborated by the application of mechanical pressure on the powdered material formed by adding KBr to the samples to be analyzed.
2.2.3. X-Ray Diffraction
The crystalline degree of the materials was determined by a Kristalloflex Siemens D5000 instrument using monochromatic Cu-Kα radiation, scanning 2θ from 10˚ to 90˚, using standard procedure.
2.2.4. Thermogravimetric Analysis
The thermogravimetric (TGA) and differential thermal (DTA) analyses of Al2O3-ZrO2 mixed oxides were carried out with a heating rate of 10˚C/min up to 1000˚C in air at 100 mL/min, using a SDT Q600 simultaneous TGADSC instrument.
2.2.5. Scanning Electron Microscopy
In this case, the microstructure of the Al2O3-ZrO2 mixed oxides was observed in a Leica S-420s scanning electron microscope (SEM) with an acceleration voltage of 20 KeV. The samples were first covered with Au-Pd in a diode sputtering coater
3. Results and Discussions
3.1.1. Textural Properties of Supports
Table 1 gives the textural measured properties of the Al2O3-ZrO2 supports (1:1 molar relation) calcined at 700˚C. The BET surface areas of 240, 233 and 201 m2/g for ethanol, isopropanol and isobutanol, respectively, were obtained at a water addition rate of 0.06 mL/min. In Table 1, it is possible to observe that the surface area decreased with the increase of the water addition rate.
Figure 2 shows the pore size distribution of Al2O3-ZrO2 supports prepared with ethanol, isopropanol and isobutanol. The mean pore size distribution observed at the inset plot of Figure 2 corresponds to that of a mesoporous material.
The average pore diameter was observed to decrease
Table 1. Textural measured properties of Al203-ZrO2 calcined at 700˚C.
from 69 Å to 48 Å using ethanol as solvent. In the case of the isopropanol used as solvent, the average pore diameter decreased from 88 Å to 54 Å. This effect is observed when isobutanol is used as solvent as well.
The pore size distribution of the mixed oxides observed in Figure 2(c) shows a bimodal size distribution, with the first maxima around 27 Å to 35 Å, but when isobutanol alcohol is used as solvent, a second maxima is observed around 60 Å to 100 Å for all water addition rates. The pore size distribution analysis of Figures 2(a) and (b) suggests that for a water addition rate of 0.06 mL/min a bimodal pore size distribution is obtained when the mixed oxides are prepared using ethanol and isopropanol as solvents. The remaining water-addition rates for the latter solvents present only a monomodal pore size distribution.
The nitrogen adsorption-desorption isotherms of Al2O3- ZrO2 supports were prepared with ethanol, isopropanol and isobutanol, which are shown in Figure 3.
The supports show E-type hysteresis loops, which are interpreted as type IV isotherms, indicating the probable existence of tubular pores with narrow openings or tubular pores with irregular narrow openings, otherwise known as ink-well type pores.
3.1.2. FT-IR Analysis
Figure 4 shows spectra obtained by means of FT-IR analyses of Al2O3-ZrO2 supports calcined at 700˚C prepared with ethanol as solvent at two water-addition rates. Similar results were obtained using isopropyl and isobutyl alcohols as solvents.
The FT-IR spectra in the 4000 - 2900 cm–1 region show a broad band for the two water addition rates, a characteristic assigned to the presence of structural OH– groups. One band is observed at 3442 cm–1 for the water-addition rate of 0.06 mL/min is shown in Figure 4(a), while the other band appears at 3447 cm–1 for the water-addition
Figure 2. Pore size distribution of Al2O3-ZrO2 supports prepared with (a) Ethanol; (b) Isopropanol; and (c) Isobutanol, obtained using different water addition rates.
rate of 0.26 mL/min is shown in Figure 4(b). A small extra band is observed at 2900 cm–1 for the water-addition rate of 0.26 mL/min, which is assigned to C-H vibration due to an incomplete polymerization, whereas this band is not observed at the lower water-addition rate [19,23]. The band associated with the stretching vibrations for the Zr-O-Al bond appears at 2374 cm–1 at the water addition rate of 0.26 mL/min. The band decrease in the Al2O3-ZrO2 mixed oxides obtained
Figure 3. Adsorption-desorption isotherms of the Al2O3-ZrO2 supports prepared with (a) Ethanol; (b) Isopropanol; and (c) Isobutanol, obtained from different water addition rates.
from a low water-addition rate is due to lattice structural rearrangement. The bending vibrations of Zr-OH bonds appear at 1630 - 1631 cm–1 for both hydrolysis rates . The bands appearing at 1395 and 1407 cm–1 are typical for the γ-alumina support. The Al-O vibration is observed at 1371 and 1363 cm–1 for a water-addition rate of 0.06 and 0.26 mL/min respectively. Another important adsorption band can be observed at 523 cm–1, which corresponds to the vibration of the Zr-O-Al bond in the alumina-zirconia powders .
3.1.3. X-Ray Diffraction
The XRD spectra of the Al2O3-ZrO2 supports at different water-addition rates are presented in Figure 5. The spectra of the Al2O3, Figure 5(a) support showed typical diffraction peaks of γ-Alumina. This fact indicates that the synthesized alumina is apparently amorphous. Two small lumps are observed at 2q - 46˚ and 2q - 66˚ that can also be related to an incipient ordered structure.
The peak corresponding to the ZrO2 phase in supports
Figure 4. FT-IR spectra of the Al2O3-ZrO2 supports prepared with ethanol, which were calcined at 700˚C for (a) 0.06 ml/min (―) and (b) 0.26 ml/min water addition rates (- -).
Figure 5. XRD spectra of Al2O3-ZrO2 supports at different water addition rates (a) Al2O3; (b) Al2O3-ZrO2 0.06 mL/min; (c) Al2O3-ZrO2 0.1 mL/min; (d) Al2O3-ZrO2 0.15 mL/min; (e) Al2O3-ZrO2 0.26 mL/min, using ethanol as solvent, calcined at 700˚C.
spectra for Al2O3-ZrO2 is not detectable; however a moderate plateau of tetragonal ZrO2 is shown at 2q - 30˚, indicating the presence of an incipient crystalline structure in the synthesized mixed oxides [26,27]. This incipient crystalline structure is responsible for the low level definition of the bands at 523 cm–1 and 525 cm–1 in the FT-IR studies .
3.1.4. Thermogravimetric Analysis
The TGA-DTA curves of the fresh Al2O3-ZrO2 mixed oxides are shown in Figure 6 using ethanol as solvent. For a water-addition rate of 0.06 mL/min, Figure 6(a): One zone of the DTA curve in the temperature range 21˚C - 180˚C corresponds to an endothermic change leading to dehydration of Al2O3-ZrO2 powders. Likewise, the zone in the temperature range 180˚C - 250˚C is exothermic and can be attributed to the incipient combustion of organic groups. In the third decomposition zone which is exothermic, in the temperature range 250˚C - 291˚C, corresponds to the elimination of chemically bonded water [19,20,27]. The fourth decomposition zone in the temperature range 291˚C - 374˚C corresponds to the total combus-
Figure 6. TGA-DTA thermograms of Al2O3-ZrO2 supports for the water-addition rates of: (a) 0.06 mL/min; (b) 0.26 mL/min, using ethanol as solvent. TGA (- -), DTA (―).
tion of organic compounds. The exothermic small peak in the temperature range 374˚C - 481˚C is observed in the DTA curve. This signal is associated with the tetragonal phase ZrO2. A zone in the temperature range 481˚C - 637˚C corresponds to an endothermic change due to the conversion of dehydroxylated boehmite to g-Al2O3 . Finally, in the temperature range 850˚C - 1091˚C there is an exothermic change, which is caused by the transformation of the ZrO2 tetragonal phase to a monoclinic phase.
For the water addition rate of 0.26 mL/min, Figure 6(b): A first zone of the DTA curve in the temperature range 21˚C - 109˚C is due to the evaporation of adsorbed water from the Al2O3-ZrO2 powders. A second zone in the temperature range 109˚C - 234˚C is exothermic and can be attributed to a total dehydration of mixed oxides and to the incipient combustion of the organic groups. In the third decomposition zone which is exothermic, in the temperature range 234˚C - 304˚C, the peak corresponds to the elimination of chemically bonded water. The fourth decomposition zone in the temperature range 304˚C - 451˚C corresponds to the total combustion of organic compounds [19,26,27]. A fifth zone in the temperature range 451˚C - 678˚C is associated to an endothermic change due to the conversion of dehydroxylated boehmite to g-Al2O3 . Finally, in the temperature range 850˚C - 1091˚C, there is an exothermic change, which is caused by the transformation of the tetragonal into the monoclinic zirconia phase.
3.1.5. Scanning Electron Microscopy
Figure 7 shows the micrographs obtained from the scanning electron microscopy studies for the Al2O3-ZrO2 powders where an appreciable formation of agglomerates can be seen. The results show that the particle size increased by increasing the water addition rate. The agglomerates are composed of mostly homogeneous-sized particles as can be observed in Figure 7(a) for the 0.06 mL/min water-addition rate. On the other hand, Figure 7(b), shows a dispersion of non-spherical particles, where serrated edges are observed for the 0.26 mL/min water-addition rate. The lowest water-addition rate allowed for a higher extent of polymerization due to a longer reaction time which yielded more homogenous-sized particles. At the largest water-addition rate, the polymerization extent was less complete due to both the greater amount of reactant and less reaction time, which yielded particles that were less rounded and had a lesser degree of aggregation.
The main difference observed when using three different solvents was related to the textural properties of the materials. The use of ethanol as solvent achieved a narrow pore-size distribution centered at the lowest pore-diameter
Figure 7. Micrograph by SEM (20,000x magnification) of the Al2O3-ZrO2 support (calcined at 700˚C) obtained with ethanol as solvent at a water addition rate of a) 0.06 mL/min; b) 0.26 mL/min.
and with the highest surface area suitable for reactions where small molecules can participate. On the other hand, the use of isobutanol as solvent generated the highest average-pore diameter, yielding consequently the lowest surface area. The FT-IR, TGA-DTA and XRD analyses were obtained for all solvents, and the main features are described only for ethanol in this contribution, since the remaining solvents show equivalent behavior.
As the water-addition rate was increased the BET surface area decreased, this effect is confirmed by SEM micrographs where particle-agglomeration is observed, Figure 7(b). The latter concept is further confirmed by the FT-IR analysis when a water-addition rate of 0.26 mL/min is used, where a less intense peak at 3447 cm–1 associated to surface OH– groups indicates a larger agglomeration degree. The presence of an FT-IR signal at 2900 cm–1 associated to C-H vibrations indicates an incomplete polymerization assumed to be due to a short reaction time. The TGA-DTA analysis shows a larger zone at 300˚C - 450˚C with respect to the lower water-addition rate in Figure 6, which indicates the presence of more chemically-bound hydrocarbons from non-reacted alkoxides, as verified by the FT-IR analysis.
The BET surface area determined at a water-addition rate of 0.06 mL/min is larger when using ethanol as solvent. The micrograph by SEM reveals in this case the formation of uniform-sized particles with facile agglomeration and no appreciable distortions. The agglomeration of small spherical particles generates an E-type hysteresis loop, confirmed by the N2 adsorption determination. It is proposed that such configuration was achieved by a combination of variables; time is important during particle formation while the textural properties are affected by the solvent. Lower water-addition rates facilitate the growth of completely-polymerized particles. TGA-DTA analysis indicates the formation and particle agglomeration with minor incorporation of non-reacted components, which can be observed by a lesser weight loss in the range 300˚C - 450˚C than that observed at the larger water-addition rate. The FT-IR signal at 3442 cm–1 associated to OH– groups is more intense. The latter has been ascribed to a larger concentration of OH– groups, tied to a higher degree of polymerization that turns out in a lack of C-H vibrations at 2900 cm–1 as observed in Figure 4(a).
According to the results and discussions, the following conclusions are:
1) The use of ethanol as solvent produced Al2O3-ZrO2 mixed oxides with higher specific surface internal areas than employing isopropyl or isobutyl alcohols;
2) The average pore diameter is largest in this study when isobutyl alcohol is used as solvent for the synthesized Al2O3-ZrO2 supports;
3) High pore volumes were favored by using the isopropyl alcohol as solvent to the obtained Al2O3-ZrO2 mixed oxides;
4) The water-addition rate of 0.26 mL/min leads to significant structural changes as indicated by the analysis of the FT-IR spectra, which can be attributed to an incomplete polymerization in the synthesis of the supports;
5) X-ray diffraction patterns for all samples do not identify well-defined crystalline structures corresponding to ZrO2 or Al2O3. This fact suggests that the samples are mostly amorphous solids;
6) For water addition at a rate of 0.10 mL/min the specific surface area of Al2O3-ZrO2 powders is almost equal to that obtained at a water-addition rate of 0.06 mL/min, hence, this rate can be used to prepare catalytic supports.
Julio Del Angel expresses recognition to the University of Guanajuato for the financial support granted to this project and to CONACYT for the scholarship stipend.