Solvent-Induced Textural Changes of As-Synthesized Mesoporous

Langmuir 2005, 21, 2591-2597

2591

Solvent-Induced Textural Changes of As-Synthesized Mesoporous Alumina, As Reported by Spin Probe Electron Spin Resonance Spectroscopy Agneta Caragheorgheopol, Horia Caldararu, Gabriela Ionita, and Florenta Savonea I.G. Murgulescu Institute of Physical Chemistry, Romanian Academy, Spl. Independentei 202, 060021 Bucharest, Romania

Nadeˇzˇda Zˇ ilkova´, Arnosˇt Zukal, and Jirˇ´ı C ˇ ejka* J. Heyrovsky´ Institute of Physical Chemistry, Academy of Sciences of the Czech Republic, Dolejsˇ kova 3, 182 23 Prague 8, Czech Republic Received October 8, 2004. In Final Form: December 28, 2004 The effect of solvent used during the synthesis and postsynthesis treatment on textural properties of organized mesoporous aluminas was investigated and related to the behavior of spin probes studied by electron spin resonance (ESR) spectroscopy. It was found that the structure of surfactant aggregates serving in the as-synthesized precipitates as templates could be easily modified by treatment with different solvents. This treatment induces corresponding variations in surface areas, mesopore volumes, and mesopore diameters of the final products. The ESR spectrum of 5-doxyl stearic acid spin probe properly reflects the changes in template structure based on changes of the solvent used and represents an early indicator of the corresponding textural modifications of the mesoporous alumina.

Introduction Organized mesoporous molecular sieves exhibit very interesting properties from material as well as application points of view. Since successful syntheses of these materials opened new possibilities in their application, reproducible and more economic synthesis routes are being searched for.1-6 High surface aluminas are commonly used in the chemical industry as catalysts, catalyst supports and adsorbents.7-9 However, the application of commercial activated aluminas is limited due to the presence of micropores and broad mesopore size distribution. For this reason the synthesis of organized mesoporous alumina (OMA) with a narrow mesopore distribution and without micropores10 is of a great industrial importance. Several synthesis pathways leading to OMAs were recently described in the literature, and their resulting textural properties differ in a broad range based on the type of the synthesis, surfactant removal procedure, and * To whom correspondence should be addressed. E-mail: [email protected]. (1) Kresge, C. T.; Leonowicz, M. E.; Roth, W. J.; Beck, J. S. Nature 1992, 359, 710. (2) Beck, J. S.; Vartuli, J. C.; Roth, W. J.; Leonowicz, M. E.; Kresge, C. T.; Schmitt, K. D.; Chu, C. T.-W.; Olson, D. H.; Sheppard, E. W.; McCullen, S. B.; Higgins, J. B.; Schlenker, J. L. J. Am. Chem. Soc. 1992, 114, 10834. (3) Corma A. Chem. Rev. 1997, 97, 2373. (4) Ying, J. Y.; Mehnert, C. P.; Wong, M. S. Angew. Chem., Int. Ed. Engl. 1999, 38, 56. (5) Di Renzo, F.; Galarneau, A.; Trens, P.; Fajula, F. In Handbook of Porous Materials; Schu¨th, F., Sing, K., Weitkamp, J., Eds.; WileyVCH: Weinheim, 2002; pp 1311-1395. (6) Schu¨th, F. Chem. Mater. 2001, 13, 3184. (7) Misra, C. Industrial Alumina Chemicals; ACS Monograph 184; American Chemical Society: Washington, DC, 1986. (8) Dıáz, I.; Gonza´lez-Penã, V.; Ma´rques-Alvarez, C., Pe´rez-Pariente, J. Collect. Czech Chem. Commun. 2003, 68, 1937. (9) C ˇ ejka, J. Appl. Catal. 2003, A 254, 327 and references therein. (10) C ˇ ejka, J.; Zˇ ilkova´, N.; Rathousky´, J.; Zukal, A.; Jagielo, J. Langmuir 2004, 20, 7532.

calcination technique. Various structure-directing agents were used, such as fatty acids, sodium dodecyl sulfate, nonionic block copolymers, etc. An outline of these structure directors is given in ref 9. The problem with the synthesis of OMAs is usually connected with their low thermal stability. Hydrolysis of an aluminum alkoxide in an organic solvent in the presence of nonionic surfactants, particularly poly(alkylene oxide) block polymers, resulted in aluminas thermally stable only up to 500 °C.11 Further increase in thermal stability was achieved by addition of small amounts of Ce3+ or La3+ into the reaction mixture.12 It was also reported that addition of amines results in an increased thermal stability of OMAs synthesized with poly(alkylene oxide) triblock copolymer surfactants.13,14 The spin probe technique was recently used to characterize the structural and dynamic properties of aggregates of poly(alkylene oxide) type nonionic surfactants in aqueous solutions and in ternary systems with an organic solvents (xylene, decane, cyclohexane).15-17 More recently, in situ electron spin resonance (ESR) measurements were carried out to study the formation mechanism of mesoporous silica.18-20 It has been proven that ESR (11) Bagshaw, S. A.; Pinnavaia, T. J. Angew. Chem., Int. Ed. Engl. 1996, 35, 1102. (12) Zhang, W.; Pinnavaia, T. J. Chem. Commun. 1998, 1185. (13) Gonza´lez-Penã, V.; Ma´rquez-Alvarez, C.; Sastre, E.; Pe´rezPariente, J. Stud. Surf. Sci. Catal. 2001, 153, 1072. (14) Gonza´lez-Penã, V.; Dıáz, I.; Ma´rquez-Alvarez, C.; Sastre, E.; Pe´rez-Pariente, J. Microporous Mesoporous Mater. 2001, 44-45, 203. (15) Caragheorgheopol, A.; Pilar, J.; Schlick S. Macromolecules 1997, 30, 2923. (16) Caragheorgheopol, A.; Bandula, R.; Caldararu H.; Joela, H. J. Mol. Liq. 1997, 72, 105. (17) Caragheorgheopol, A.; Caldararu, H.; Dragutan, I.; Joela, H.; Brown, W. Langmuir 1997, 13, 6912. (18) Zhang, J. Y.; Luz, Z.; Goldfarb, D. J. Phys. Chem. B 1997, 101, 7087. (19) Zhang, J. Y.; Zimmerman, H.; Luz, Z.; Goldfarb, D. Stud. Surf. Sci. Catal. 1998, 117, 535.

10.1021/la047510h CCC: $30.25 © 2005 American Chemical Society Published on Web 02/18/2005

2592

Langmuir, Vol. 21, No. 6, 2005

Caragheorgheopol et al.

Table 1. Synthesis Procedures Used for the Synthesis of Different OMA Samples sample code

recipe

surfactant

solvent

washing agent

drying

OMA 1 OMA 2 OMA 3 OMA 4 OMA 5 OMA 6 OMA 7 OMA 8 OMA 9 OMA10a OMA 11b OMA 12c OMA 13 OMA 14

1 1 1 1 1 1 1 2 2 2 2 2 2 2

Pluronic P123 Pluronic P123 Pluronic P123 Pluronic P123 Pluronic PE10400 Pluronic PE10400 Pluronic PE10400 Pluronic P123 Pluronic P123 Pluronic P123 Pluronic P123 Pluronic P123 Tergitol 15-S-12 Tergitol 15-S-12

ACN ACN ACN ACN ACN ACN ACN 2-BuOH 2-BuOH 2-BuOH 2-BuOH 2-BuOH 2-BuOH 2-BuOH

ACN H2O H2O H2O ACN H2O ethanol 2-BuOH H2O 2-BuOH H2O H2O 2-BuOH H2O

RT RT 120 °C, 24 h 120 °C, 24 h RT RT RT RT RT RT RT RT RT RT

extraction

ethanol ethanol

a The same synthesis procedure as with the sample OMA 8. b The same synthesis procedure as with the sample OMA 9. c This sample was left in water overnight.

spectroscopy of spin probes can significantly contribute to detailed understanding of the local structural and dynamic parameters in micro-heterogeneous systems. Therefore, it was employed in this study to report on the changes in the surfactant organization during the synthesis and postsynthesis treatment of OMAs. The objective of this contribution is focused on new observations obtained by the spin probe technique on the intimate processes taking place during the synthesis of organized mesoporous alumina using poly(alkylene oxide) triblock copolymer surfactants as structure-directing agents and relates the use of different solvents and washing procedures to the resulting textural properties. As the textural properties of OMAs strongly depend on the conditions of washing and high-temperature treatment of as-made materials,21,22 special attention was paid to the detailed investigation of the textural parameters of OMAs after different washing procedures followed by calcination performed under well-defined conditions.

Table 2. Parameters of ESR Spectra of 5-DSA in the Starting Solutions and in As-Synthesized Aluminas Prepared by Recipe 1, Washed with Different Solventsa system ACN H2O P123 20% in H2O P123 20% in H2O/ACN ) 9:1 (g/g) P123 20% in H2O/ACN ) 8:2 (g/g) P123 20% in H2O/ACN ) 6:4 (g/g) P123 20% in ACN recipe 1/P123 as-synthesized alumina washed with ACN washed with H2O washed again with ACN recipe 1/PE10400 as-synthesized alumina washed with ACN washed with H2O

OMA code

aN (G)

1010τC (s)

14.6 15.8

0.5 2.0

S

0.47 0.41 0.29 14.9 14.6 OMA1 OMA2

OMA5 OMA6

12.8 1.3

14.6

∼27b

14.3

∼26b

14.4

∼29b

0.47

0.50

a

Experimental Section Synthesis. Two synthesis procedures were employed. In the first procedure (recipe 1) 3.2 g of Pluronic P123 (BASF), (EO)20(PO)70(EO)20, where EO and PO denote ethylene and propylene oxide units, respectively, or PE 10400 (BASF), (EO)25(PO)56(EO)25, was mixed with 22.7 g of acetonitrile (ACN), and then 1.6 g of 25 wt % water solution of NH4OH was added. The mixture was vigorously stirred at room temperature until the Pluronic was dissolved. Finally, 3.96 g of aluminum sec-butoxide was added and the mixture was stirred for 24 h. In the second procedure (recipe 2) 0.4 g of deionized water was added to 4.7 mL of anhydrous 2-butanol. Then 2.46 g of aluminum sec-butoxide in 6 mL of 2-butanol was added. In the last step, 6.91 g of Pluronic P123 or 0.74 g of Tergitol 15-S-12 (C15H31EO)12, Sigma-Aldrich) in 6 mL of 2-butanol was very slowly added. The final mixture was vigorously stirred for 3 h and allowed to stand for another 16 h of aging. The samples were recovered by filtration, washed in water, acetonitrile, ethanol, or 2-butanol (for details see Table 1), and dried at room temperature with the exception of samples OMA 3 and OMA 4, which were dried at 120 °C for 24 h. After drying, samples OMA 4 and OMA 7 were extracted with ethanol. Assynthesized aluminas were calcined in a very thin layer in a stream of air with a temperature ramp of 1 °C/min to 120 °C (120 min hold), then with 1 °C/min to 240 °C (240 min hold), and (20) Galarneau, A.; Di Renzo, F.; Fajula, F.; Mollo, L.; Fubini, B.; Ottaviani, M. F. J. Colloid Interface Sci. 1998, 201, 105. (21) C ˇ ejka, J.; Zˇ ilkova´, N.; Rathousky´, J.; Zukal, A. Phys. Chem. Chem. Phys. 2001, 3, 5076. (22) C ˇ ejka, J.; Kooyman, P. J.; Vesela´, L.; Zukal, A. Phys. Chem. Chem. Phys. 2002, 4, 4823.

For comparison, results for surfactant solutions in other solvent mixtures are also reported. b Evaluated as an approximate value with eq 1, albeit the motion rate is outside the recommended validity range.

finally with 1 °C/min to 540 °C (360 min hold). All prepared samples are listed in Table 1. ESR Measurements. Two series of samples were examined by ESR: (1) Starting solutions, representing surfactant solutions in organic solvents used for the synthesis of OMAs, to which the spin probe 5-doxyl stearic acid (5-DSA) was added to yield a 2 × 10-4 M concentration. For comparison, solutions of the same surfactant in solvent mixtures were also examined. (2) As-synthesized aluminas, obtained by the recipes 1 or 2. The spin probe 5-DSA was included into the reaction mixture before alumina precipitation. The ESR spectra of the probe were measured after different washing procedures (cf. Tables 1 and 2). The ESR spectra were recorded on a FA 100 JEOL spectrometer with 100 kHz field modulation using X-band frequency. An approximate value of the rotational correlation time, τC, was calculated according to the formula23

τC (s) ) (6.51 × 10-10)∆H(0){[h(0)/h(-1)]1/2 + [h(0)/h(1)]1/2 - 2} (1) where ∆H(0) is the line width (in Gauss) of the central line, and h(-1), h(0), and h(1) are the peak-to-peak heights of the M ) -1, 0, and +1 derivative lines, respectively. τC is connected to the (23) Stone, T. J.; Buckman, T.; Nordio, P. L.; McConell, H. M. Proc. Natl. Acad. Sci. U.S.A. 1965, 54, 1010.

Solvent-Induced Textural Changes on Alumina

Langmuir, Vol. 21, No. 6, 2005 2593

local viscosity, η, by the Debye-Stokes-Einstein equation

τC ) 4πηR3/3kT

(2)

where R is the hydrodynamic radius of the tumbling entity. The τC values obtained were used to follow, in a qualitative manner, significant changes in the microenvironment of the probes. The order parameter, S, is defined as:24 S ) (A| - A⊥)/[Azz - (Axx + Ayy)/2], where Azz, Axx, and Ayy are the principal elements of the A tensor in absence of molecular motion and A| and A⊥ are derived from experimental spectra. The order parameters, S, were calculated using the following parameters reported for doxyl probes:25 Azz ) 33.5 G, Axx ) 6.3 G and Ayy ) 5.8 G. To correct the polarity difference between the studied sample and the sample whose Azz, Axx, Ayy values are used, the S value calculated above is multiplied with the inverse ratio of the corresponding isotropic nitrogen hyperfine splitting values, aN, calculated as the arithmetical means of the tensor components.24 Characterization of Calcined Samples. The as-synthesized and calcined OMAs were characterized using X-ray powder diffraction and nitrogen adsorption isotherms. Powder X-ray diffraction patterns were collected on a Siemens D 5005 diffractometer in the Bragg-Brentano geometry arrangement with Cu KR radiation in the range 1.0-10.0 of 2θ. Adsorption isotherms of nitrogen at -196 °C on calcined samples were measured on a Micromeritics ASAP 2010 instrument. Before the adsorption measurement, all calcined samples were degassed at 350 °C overnight.

Figure 1. ESR spectra of 5-DSA in 20% P123 in (a) H2O, (b) H2O/ACN ) 8/2(g/g), (c) H2O/ACN ) 6/4(g/g), (d) ACN.

Results and Discussion Recipes 1 and 2 using nonionic surfactants (Pluronic P123, PE10400, and Tergitol 15-S-12) in nonaqueous solutions were applied for the following reaction systems: P123/ACN/water, PE10400/ACN/water, P123/2-BuOH/ water, and Tergitol 15-S-12/2-BuOH/water. ESR Results. The spin probe technique with 5-DSA as a probe was used to characterize the aggregation of the surfactant in the starting solutions and in the assynthesized materials after different washing and/or extraction treatments. 5-DSA is a long amphiphile molecule having a tendency to align with the surfactant molecules. Its molecular structure was imagined so as to present specific, easy to identify changes in the ESR spectrumsaccording to its dynamics. Starting Solutions. The P123/ACN/Water System. We have examined a series of mixed water/ACN solutions of Pluronic P123 in order to understand the role of ACN and water on the aggregation of this block copolymer. Pluronic block copolymers have large micelles, to such an extent that micelle tumbling and/or surfactant lateral diffusion are too slow to influence the ESR spectra of the spin probe.26 Thus, only the local movement of the probe is reflected by the spectrum. The ESR spectrum of the aqueous solution exhibits features specific for anisotropic rotation, i.e., the rotation rate around the long molecular axis is fast and by 1 to 2 orders of magnitudes higher than the rotation around perpendicular axis (Figure 1, curve a). For such spectra an order parameter, S, can be determined from the spectra (Table 2). When water is gradually replaced by ACN, the order degree of the surfactant chains decreases (Figure 1, curve b). After a water/ACN weight ratio of 6/4 is reached, the ordering is completely lost (Figure 1, curve c). However, the high τC value of 5-DSA unambiguously indicates the presence of P123 micelles in this solution. Thus, the spectra indicate a gradual swelling of the micelles with ACN in this series of solutions. (24) Seelig, J. J. Am. Chem. Soc. 1970, 92, 3881. (25) Gaffney, B. J. In Spin Labeling I; Berliner, L. J., Ed.; Academic Press: New York, San Francisco, London, 1976; p 567. (26) Lasic, D. D.; Hausser, H. J. Phys. Chem. 1985, 89, 2648.

Figure 2. ESR spectra of 5-DSA in 10 wt % of Pluronic P123 in (a) 2-BuOH, (b) H2O +10 wt % of 2-BuOH, (c) H2O + 6 wt % of 2-BuOH, (d) H2O.

The starting solution of recipe 1 is a solution of surfactant P123 in acetonitrile. (A small amount of water from NH4OH solution, which was added to the reaction mixture, can be neglected because it reacts with aluminum sec-butoxide.) The spectrum of three narrow, almost equal lines (Figure 1, curve d) is characteristic of rapid, isotropic rotation, as encountered in low viscosity liquids. There is only a slight increase in the rotational correlation time (τC) compared to the spectrum in pure ACN (Table 2), which could result from the increase in the macroscopic viscosity produced by the dissolution of the polymer in ACN. Thus, one cannot unambiguously decide whether there are no aggregates at all in the solution or whether they are swollen to the extent that the polar shell cannot be distinguished from a polymer solution in ACN. The P123/2-Butanol/Water System. The ESR spectrum of 5-DSA spin probe dissolved in the starting solution has been measured. For comparison the same spectrum was measured in water and several water/butanol mixtures. A solution of 10 wt % of Pluronic P123 in water consists of well-organized micelles, as results from the anisotropic ESR spectra of the 5-DSA spin probe (Figure 2, curve d). Addition of 2-butanol up to 10 wt % in the solvent (the maximum amount of 2-butanol accepted in the micellar

2594



Table 3. Parameters of ESR Spectra of 5-DSA in the Starting Solutions and in As-Synthesized Aluminas Prepared by Recipe 2, Washed with Different Solventsa system BuOH P123 10% in H2O P123 10% in H2O + 6% BuOH P123 10% in H2O + 10% BuOH P123 10% in BuOH recipe 2/P123 as-synthesized alumina washed with BuOH washed with H2O overnight in water recipe 2/Tergitol Tergitol 14% in BuOH Tergitol 20% in H2O as-synthesized alumina washed with BuOH washed with H2O

OMA code

aN (G)

1010τC (s)

14.8

6.5

S

2Azzb (G)

0.45 0.43 0.35

OMA10 OMA11 OMA12

14.8

15.1

15.1

13.5

54.6 0.45 14.8

8.4 0.43

OMA13 OMA14

57.2 0.44

a For comparison results for solutions of the surfactants in other solvent mixtures are also reported. b 2A represents the distance zz between the extreme spectral features.

Figure 3. ESR spectra of 5-DSA in as-synthesized aluminas prepared by recipe 1 with Pluronic P123, washed with (a) ACN (OMA 1), (b) H2O (OMA 2), (c) sample b washed again with ACN.

phase) leads to a significant decrease in the order parameter of the 5-DSA spectra (Figure 2, curves b and c, Table 3). 2-Butanol is not a selective solvent for any of the polymer blocks used; it is partly miscible with water and it also manifests a tendency to be localized in the interface region as a cosurfactant. In 2-butanol, the P123 micelles are swollen with the solvent and the order is lost (Figure 2, curve a). A much higher value of the rotational correlation time in this solution (τC ) 15 × 10-10 s), as compared to the probe in pure 2-butanol (τC ) 6.5 × 10-10 s) indicates the presence of aggregates in 2-butanol. In this case 2-butanol acts as a solvent but also as a cosurfactant. Tergitol micelles behave in a similar way with those of Pluronic P123: in water the micelles are tightly packed (S ) 0.43); however, in 2-butanol the micelles are swollen and τC ) 8.4 × 10-10 s is only slightly higher than the rotational correlation time in pure 2-butanol. As-Synthesized Precipitates. When the synthesis is performed in ACN (recipe 1) and the as-synthesized precipitate is washed with this solvent, the 5-DSA probe appears to be mobile. However, its mobility is strongly hindered, which indicates a very high local viscosity in the aggregate interior, in an isotropic environment without any ordering (Figure 3, curve a). This type of spectrum was not observed in Pluronic P123 solutions with different water/ACN ratios and reflects the compacting effect of the inorganic component of the as-synthesized precipitate. Brief washing of the as-synthesized mesoporous alu-

Figure 4. ESR spectra of 5-DSA in as-synthesized aluminas (by recipe 2 with P123) washed with (a) 2-BuOH (OMA 10), (b) H2O (OMA 11), and (c) left overnight in H2O (OMA 12).

mina in water has a dramatic effect on the ESR spectrum (Figure 3, curve b), which now exhibits the anisotropic features characteristic of tightly packed micelles, similar to the behavior of Pluronic P123 in water (cf. Figure 1, curve a, and Table 2). Thus, washing with water determines the substitution of ACN by more polar water molecules in Pluronic P123 micelles in the pores. The change is completely reversible, as indicated by the ESR spectrum, returning to the original form (Figure 3, curve c) after subsequent washing with ACN. The same results were obtained also with Pluronic PE10400 (OMA 5, OMA 6). Prolonged washing with ethanol led to the removal of the spin probe (OMA 4, OMA 7). This is not surprising, as refluxing in ethanol is usually applied to remove the surfactant from as-synthesized OMA samples.9 When the reaction is accomplished in 2-butanol (recipe 2), the spin probe 5-DSA appears immobilized in the assynthesized alumina, which was washed with 2-butanol (Figure 4, curve a, Table 3). In this case even the rapid rotation around the long molecular axis is hindered. After a brief washing with water, the probe appears to be more mobile and presents a three-line spectrum (Figure 4, curve b) whose parameters are very close to those of P123 micelles in 2-butanol (with aN indicating a slightly higher local polarity, Table 3). If the sample is left in water overnight, it suffers more changes and the spectrum becomes similar to the one of the spin probe in Pluronic


Langmuir, Vol. 21, No. 6, 2005 2595

Figure 5. ESR spectra of 5-DSA in as-synthesized aluminas prepared by recipe 2 with Tergitol 15-S-12 washed with (a) 2-BuOH (OMA 13) and (b) H2O (OMA 14). (c) ESR spectrum of 5-DSA in solution of 20 wt % of Tergitol in H2O.

P123 micelles in water (cf. Figure 4, curve c, and Figure 2, curve d). Regarding the two stages of 2-butanol exchange by water, one may speculate, on the basis of the spectral parameters, that they correspond to the two roles of 2-butanol, as a solvent and as a cosurfactant. The “external” solvent should be exchanged first, during the washing, while the removal of the cosurfactant (i.e., the 2-butanol molecules aligned with the surfactant in the aggregates) would require a longer exposure to water. After that the micelles are tightly packed, as are those formed in aqueous solution. Experiments using Tergitol 15-S-12 instead of Pluronic P123 in recipe 2 with 2-butanol (OMA 13 and OMA 14) showed a similar behavior. However, in this case the “water-type” spectrum appears already after brief washing and it is not necessary to leave the sample in water overnight (Figure 5). Thus, it can be inferred that the different hydrophilicity of these two surfactants influences the rate of 2-butanol exchange by water molecules. The reason the probe 5-DSA appears immobilized (being bound to alumina precursor) when the solvent is 2-butanol and becomes more mobile after washing with water could be connected with the much lower solubility of 5-DSA in water as compared to 2-butanol. Replacement of 2-butanol with water would result in probe withdrawal from the interface region to deeper micellar zones, which have a more hydrophobic character. This behavior should be, however, connected with the properties of the PEO-type surfactant used, since analogue changes were not observed by the present authors in the case of stearic acid templated synthesis of alumina in propanol. The remarkable fact is that the whole process of solvent exchange and aggregate rearrangement proceeds regardless of the presence of inorganic component (aluminum oxide/hydroxide species) of the as-synthesized precipitate. This is a result of substantially weaker interaction of nonionic surfactants with alumina species compared to the interaction of ionic surfactants, such as long-chain carboxylic acids, with mesoporous aluminas.9 On the basis of these results careful calcination of individual samples of mesoporous aluminas was carried out and textural properties were determined based on the nitrogen adsorption isotherms. The Structure of Calcined Samples. Nitrogen isotherms were measured on all the calcined OMA samples; the typical isotherms are shown in Figure 6.

Figure 6. Nitrogen adsorption isotherms on samples OMA 8 (3), OMA 10 (O), OMA 11 (4), and OMA 12 (0). Except for that on sample OMA 12, isotherms are shifted by 10 mmol/g each. Solid symbols denote desorption. Table 4. Structural Parameters of Individual OMA Samples sample code

SBET (m2/g)

VME (cm3/g)

DME (nm)

OMA 1 OMA 2 OMA 3 OMA 4 OMA 5 OMA 6 OMA 7 OMA 8 OMA 9 OMA10a OMA 11b OMA 12c OMA 13 OMA 14

272 285 284 293 286 277 303 451 385 431 450 389 285 287

0.636 0.536 0.529 0.567 0.578 0.493 0.571 1.324 1.200 1.362 1.256 0.899 0.526 0.538

7.4 4.8 4.8 4.3 6.0 5.3 5.5 8.5 7.8 8.9 7.7 5.7 6.7 5.5

a The same synthesis procedure as with the sample OMA 8. b The same synthesis procedure as with the sample OMA 9. c This sample was left in water overnight.

The BET surface areas were calculated using adsorption data in a relative pressure range from 0.05 to 0.25. The total pore volume was determined from the amount adsorbed at a relative pressure of about 0.98. These structural parameters are listed in Table 4. The distribution of mesopores was calculated from desorption branch of the hysteresis loop using the Barrett-Joyner-Halenda method.27 The resulting distribution curves are shown in Figures 7, 8, and 9. The mesopore diameters corresponding to the maximum of distribution curves are given in Table 4. The structure parameters in Table 4 show the strong influence of the solvent used during the synthesis and especially the washing procedure of as-synthesized samples on the structural properties of calcined OMA materials. The significant changes in the pore structure, induced by the changes of washing agent, are parallel to the changes in the ESR spectra of 5-DSA spin probe. This fact proves that the ESR spectral differences reflect genuine changes (27) Barrett, E. P.; Joyner, L. G.; Halenda, P. P. J. Am. Chem. Soc. 1951, 73, 373.

2596


Figure 7. Mesopore size distribution for samples OMA 1 (a), OMA 2 (b), OMA 3 (c), and OMA 4 (d).

in the template and not some possible secondary effect on the spin probe. The samples OMA 1 and OMA 2 illustrate the decisive role of the washing agent. The surface area of both samples is practically the same; however, the mesopore volume and mesopore diameter of the sample OMA 2 are smaller than analogous structural parameters of the sample OMA 1. The observed influence of water as a washing agent on the porous structure is exactly in accordance with the results of ESR investigations. The well-packed micelles, which are formed under influence of water as washing agent, bring about the tightly packed alumina texture. This results in a substantial decrease in the pore size diameter from 7.4 to 4.8 nm and mesopore volume from 0.636 to 0.536 cm3/g for OMA 1 and OMA 2, respectively. The mesopore size distribution of OMA 2 is much narrower than that of the OMA 1 (Figure 7), corresponding to more uniform micelle sizes in water, when packing is tight, compared to different sizes of swollen micelles in ACN. It should be noted that similar conclusions follow from the structural parameters of samples OMA 5 and OMA 6, analogues of OMA1 and OMA 2 prepared with Pluronic PE10400. The structure parameters of samples OMA 3 and OMA 4 show, that if washing in water was followed by drying at 120 °C for 24 h, the reduced pore diameters and narrow size distribution are maintained even after refluxing in ethanol for 2 days at 80 °C (without the heating, extraction with ethanol leads to large pores with a very broad size distribution). Presumably, the polymerization of the inorganic precursor advances to further stabilize the alumina structure. The structure parameters of the samples OMA 8, OMA 9, OMA 10, OMA 11, and OMA 12 (Table 4), prepared using 2-BuOH as a solvent, exhibit similar dependence on the washing conditions as the samples prepared in ACN. Mild washing with water brings a slight decrease in the average pore diameter. The total exchange of 2-butanol by water (the sample OMA 12 left in water overnight), results again in the decrease in the mesopore volume and mesopore size observed for the samples prepared by the recipe 1 (Table 4, Figure 8). The mesopore


Figure 8. Mesopore size distribution for samples OMA 8 (a), OMA 10 (b), OMA 11 (c), and OMA 12 (d).

Figure 9. Mesopore size distribution for samples OMA 13 (a) and OMA 14 (b).

volume of all OMA samples prepared in the 2-BuOH is more than twice larger than that of corresponding aluminas prepared in ACN. Since the pore size of both groups of samples is similar, the surface areas of samples prepared in 2-BuOH are larger than those of the samples prepared in ACN. The differences in the mesopore volumes and surface areas are probably caused by different thicknesses of the pore walls. The aluminas with thinner pore walls can be supposed to form in 2-BuOH. On the contrary, the samples prepared in ACN are characterized by a more robust structure with thicker pore walls. The structure parameters of the OMA 13 and OMA 14 prepared using Tergitol 15-S-12 depend on the washing agent analogously as the structure parameters of OMA 8, OMA 9, OMA 10, and OMA 11, i.e., the washing with water causes a decrease in the pore diameter (Table 2, Figure 9). Although, the same volumes of mesopores were found for OMA 13 and OMA 14, the pore size diameters differ in diameter more than 1 nm.


Langmuir, Vol. 21, No. 6, 2005 2597

The observed structure modifications of the as-synthesized alumina, by a simple solvent change, induced by the change of the template assembling, have important implications on the understanding of the synthesis mechanism: they show that the template is free to rearrange at this stage, if conditions require it, and that the inorganic component of the as-synthesized precipitate (aluminum oxide-hydroxides species) are bound to the surfactant (or vice versa) and “follow” it in the new structure. Heating to 120 °C brings about a certain stabilization of the structure, by more advanced polymerization of the oxidehydroxide species and solvation changes do not produce textural modifications after this treatment.

changes induce corresponding modifications in the alumina pore structure. The structure of the as-synthesized materials is already stabilized by heating at 120 °C for 24 h. The EPR spectrum of the 5-DSA spin probe is an adequate, easy to perform, early indicator of changes in template structure with the change of the solvent, pointing to corresponding changes in the alumina average pore sizes and their distribution. Presumably, this observation may prove valuable for a simple way to direct and to monitor the morphology of mesoporous alumina (or other oxides).

Conclusions

Acknowledgment. This work was carried out with the financial support of the Grant Agency of the Czech Republic (104/02/0571), Grant Agency of the Academy of Sciences of the Czech Republic (A4040411), and NATO in the framework of “Science for Peace” (SfP-974 217).

The structure of as-synthesized mesoporous alumina, prepared by the neutral pathway in organic solvents, is not stabilized. Changes in the aggregate organization of surfactant molecules can be easily produced by treatment with a different solvent at room temperature. These

LA047510H

Solvent-Induced Textural Changes of As-Synthesized Mesoporous

Recommend Documents