REACTIVE & FUNCTIONAL POLYMERS
Reactive & Functional Polymers 63 (2005) 135–141
www.elsevier.com/locate/react
Iron acetylacetonate complex anchored on silica xerogel polymer Marcia C. Brasil, Edilson V. Benvenutti *, Jose´ R. Grego´rio, Annelise E. Gerbase Instituto de Quı´mica, UFRGS, CP 15003, CEP 91501-970, Porto Alegre, RS, Brazil Received 9 March 2004; received in revised form 3 February 2005; accepted 24 February 2005 Available online 13 April 2005
Abstract Iron(III) acetylacetonate complex immobilized on silica surface was obtained by the following three steps synthesis: (i) synthesis of organic precursor (acacsil) containing the acetylacetonate (acac) group; (ii) simultaneous polycondensation of the acacsil with the tetraethylorthosilicate (TEOS) by the sol–gel method resulting in the acac/silica xerogel; and (iii) complexation of the Fe(III) on the acac sites of the xerogel surface forming a Fe-acac/silica xerogel. The xerogels, acac/silica and Fe-acac/silica, are hybrid polymers that present a covalent organic/inorganic interface between the acac groups and silica. Xerogels were characterized by using scanning electron microscopy, energy dispersive spectroscopy, infrared spectroscopy, N2 adsorption desorption isotherms (pore size distribution and surface area) and elemental analysis. The Fe-acac/silica xerogel was tested as heterogeneous catalyst for the cis-cyclooctene epoxidation, employing the same conditions of the homogeneous Mukaiyama system. 2005 Elsevier B.V. All rights reserved. Keywords: Hybrid material; Sol–gel; Acac; Iron acetylacetonate
1. Introduction The difficult recovery and recycling of homogeneous transition metal catalysts is the major drawback for their industrial use in catalysis. To overcome this difficulty several studies have been done to immobilize homogeneous catalysts on dif*
Corresponding author. Tel.: +55 51 3316 7209; fax: +55 51 3316 7304. E-mail address:
[email protected] (E.V. Benvenutti).
ferent organic or inorganic supports [1–3]. In this context, the sol–gel process is a powerful tool for the immobilization of transition metals [4,5]. This process can be described by two reactions, hydrolysis and polycondensation, starting from alkoxysilanes and organoalkoxysilanes as precursor reagents. The sol–gel synthesis can be performed at room temperature and allows to prepare pure materials with uniform distribution of organic and inorganic phases in nanometric or molecular level [6,7]. Furthermore, the resulting xerogel
1381-5148/$ - see front matter 2005 Elsevier B.V. All rights reserved. doi:10.1016/j.reactfunctpolym.2005.02.014
136
M.C. Brasil et al. / Reactive & Functional Polymers 63 (2005) 135–141
properties can be controlled by the choice of the synthetic conditions [8,9]. In recent papers, an easy procedure was presented to obtain appropriate organoalkoxysilanes to be used as sol–gel organic precursors [9,10]. The sol–gel hybrid polymers obtained, aniline/silica and biphenylaminepropylsilica/silica, were very thermally stable with a covalent organic–inorganic interface. In the present work, a new organic precursor containing oxygen chelating groups, the acetylacetonatepropyltrimethoxysilane (acacsil), was obtained. This organic precursor was polycondensed with tetraethylorthosilicate (TEOS) to obtain a new hybrid polymer containing acetylacetonate groups (acac/silica). These groups are well known ligands in coordination chemistry and are suitable to chelate different transition metals, resulting in heterogeneous transition metal acetylacetonate catalysts. When a solution of FeCl3 Æ 6H2O was mixed with the acac/silica xerogel, an iron acetylacetonate complex was formed on the silica surface, Fe-acac/silica. These new sol–gel hybrid materials, the acac/silica and the Fe-acac/silica, were characterized by infrared spectroscopy, N2 adsorption– desorption isotherms, elemental analysis, scanning electron microscopy, and energy dispersive spectroscopy. Continuing our effort in developing new environmental friendly epoxidation catalysts [11–13], the Fe-acac/silica was tested as heterogeneous catalyst for the cis-cyclooctene epoxidation, in the same conditions of homogeneous Mukaiyama system [14].
20 min, and the supernatant that contains the acetylacetonatepropyltrimethoxysilane (acacsil) was used as sol–gel chelating organic precursor in the gelation process. Afterwards, tetraethylorthosilicate (TEOS) (5 ml, 22 mmol), ethyl alcohol (5 ml), HF (0.1 ml) and water (1.6 ml, 88 mmol) were added under stirring to the precursor solution. The gelation occurs by the fluoride catalytic process, at pH ca. 4. The mixture was stored for a week, just covered without sealing, for gelation and solvent evaporation. The resulting xerogel was then extensively washed using the following solvents in this order: toluene, THF, dichloromethane, ethyl alcohol, distilled water and ethyl ether. The xerogel was finally dried for 30 min in an oven at 100 C. The resulting polymer, assigned as acac/silica, was ground in an agate mortar for subsequent analysis. 2.2. Iron complexation to acac/silica xerogel A solution of 3.3 g (0.012 mol) of iron(III) chloride hexahydrate (FeCl3 Æ 6 H2O) in 25 ml of distilled water was prepared. This solution was added to acac/silica xerogel, under stirring, for 30 min. After the complexation with iron (III) salt, the solid phase was filtered and extensively washed with water, ethyl alcohol and acetone at room temperature, until no Fe(III) has been detected using the NH4SCN solution. No iron was liberated on stirring the Fe-acac/silica with water at 45 C over a period of 30 min. The solid phase was dried at 100 C for 1 h. The resulted solid was assigned as Fe-acac/silica. 2.3. Elemental analysis
2. Experimental 2.1. Synthesis of the acac/silica xerogel Acetylacetonate (Aldrich, 0.6 ml, 6 mmol) was activated with sodium hydride (Merck, 144 mg, 6 mmol) in 20 ml of toluene:THF (1:1) solution for 30 min and then 3-chloropropyltrimethoxysilane (CPTMS) (Acros, 1.1 ml, 6 mmol) was added. The mixture was stirred under argon at solventreflux temperature for a period of 5 h. The solution was then centrifuged at 5000 rpm, for
The chelating organic phase content was obtained by CHN analysis using a CHN Perkin– Elmer analyzer, model 2400. The analysis was made three times after heating the materials at 100 C under vacuum for 1 h. 2.4. Infrared measurements A self-supporting disk of the acac/silica xerogel, with an area of 5 cm2, weighing ca. 100 mg was prepared. The disk was heated during 1 h, at a
M.C. Brasil et al. / Reactive & Functional Polymers 63 (2005) 135–141
range of temperatures from 100 to 400 C, under vacuum (10 2 Torr), using an IR cell [15]. The disk was analyzed in a Shimadzu FTIR spectrophotometer, model 8300. The IR spectra were obtained with a resolution of 4 cm 1, with 100 scans. 2.5. Scanning electron microscopy The acac/silica and Fe-acac/silica xerogels were analyzed by scanning electron microscopy (SEM) in a Jeol equipment, model JSM 5800, with 20 kV and 60,000· of magnification. The average particle sizes were obtained by choosing and manually marking the diameter of a minimum of 150 spherical particles in each image. The average diameters and the standard deviations were determined by using the Quantikov software [16].
137
acquisition time was 100 s and 500· of magnification was used. 2.9. Epoxidation of cis-cyclooctene The epoxidation reaction was carried out at room temperature, under atmospheric pressure of oxygen. Fe-acac/silica xerogel (212 mg, 0.013 mmol of Fe), cis-cyclooctene (Merck, 0.7 g, 6.3 mmol), isobutyraldehyde (Aldrich, 1.12 ml, 12.5 mmol) and 1,2-dichloroethane (15 ml) were magnetically stirred during 24 h. The mixture was then analyzed in a HP 5890 gas chromatograph equipped with a 30 m · 0.25 mm HP-1 column, using methyl stearate as internal standard. A blank reaction was carried out in the same conditions resulting in a conversion of 7%.
2.6. Pore size distribution
3. Results and discussion
The pore size distribution of acac/silica and Feacac/silica xerogels was obtained by the nitrogen adsorption–desorption isotherms. These isotherms were determined at liquid nitrogen temperature, using a homemade volumetric apparatus, connected to a turbo molecular Edwards vacuum line system, employing Hg capillary barometer. The apparatus is frequently checked with alumina standard reference. The solid was previously degassed at 150 C, in vacuum, for 2 h. The data analysis was made using the BJH (Barret, Joyner, and Halenda) method [17].
The synthesis of the acac/silica hybrid xerogel was performed in two steps. First, the reaction between acetylacetonate and CPTMS was carried out in aprotic solvent mixture toluene–THF, using NaH as base activator, as illustrated in Scheme 1. The reaction product, acetylacetonatepropyltrimethoxysilane (acacsil), was used as organic precursor. In the second step, the acacsil was polycondensed with TEOS in order to obtain the acac/silica xerogel polymer, as shown in Scheme 2. The immobilization of acac chelating group in solid matrices has already been reported [4,19]. However, in the present work, we used sodium hydride as a base activator to produce a faster reaction between acac and CPTMS. At the end of the reaction, the by-product NaCl can be easily separated from the solution by centrifugation, resulting in a high yield of the organic precursor acacsil. The presence of the organic phase in the acac/ silica and Fe-acac/silica xerogel polymers was confirmed by CHN elemental analysis. The amount of carbon corresponds to an acac/Si of 0.04 for both (Table 1). The Fe-acac/silica xerogel was formed when the acac/silica was mixed with a FeCl3 solution and extensively washed until no more iron leaching was observed. The metal immobilization
2.7. Surface area The specific surface area of the previously degassed acac/silica and Fe-acac/silica xerogels at 150 C, under vacuum, was determined by the BET (Brunauer, Emmett and Teller [18]) multipoint technique in the volumetric apparatus, cited above, using nitrogen as a probe. 2.8. EDS analysis The EDS (energy dispersive spectroscopy) analysis was made using a Jeol equipment, model JSM 5800, with a Noran detector with 20 kV. The
138
M.C. Brasil et al. / Reactive & Functional Polymers 63 (2005) 135–141
Scheme 1.
Scheme 2.
Table 1 Elemental analysis and morphological data Xerogel polymer
Organic contenta (mmol g 1)
acac/Si molar ratioa
Fe/Si molar ratioa
BET surface area (m2 g 1)
Particle sizeb (nm)
Acac/silica Fe-acac/silica
0.60 ± 5% 0.70 ± 5%
0.038 ± 7% 0.042 ± 7%
– 0.04 ± 12%
525 ± 5% 485 ± 5%
92 93
a b
Obtained from EDS or CHN analysis. Standard deviation lower than 23 nm.
was confirmed by the EDS analysis (Table 1). The Fe/Si molar ratio calculated from EDS analysis and the acac/Si molar ratio calculated from CHN analysis (Table 1) are very similar, which suggests that each iron is bonded to one acac site. This statement was supported by the pore size distribution curves shown in Fig. 1. For both xerogels, acac/silica and Fe-acac/silica, a very similar sharp pore diameter distribution was observed, with a maximum in ca. 3.3 nm. This result indicates that the iron immobilization should occur in the specific surface acac sites, without agglomeration that could result in an entrapment of the pores.
In the infrared spectra of acac/silica and Feacac/silica (Fig. 2), the presence of both components, the organic and inorganic phases, is clearly observed. The organic component can be identified by the band at 1710 cm 1 (Spectrum a), assigned to the carbonyl stretching mode of the acac group. The inorganic silica component can also be identified by the presence of the typical silica overtone bands of ca. 1860 cm 1. The spectrum of the Feacac/silica xerogel showed a displacement of the carbonyl stretching band of the acac group from 1710 to 1628 cm 1 (Spectrum e), which usually takes place when a metal complexation occurs [20].
M.C. Brasil et al. / Reactive & Functional Polymers 63 (2005) 135–141
0.08
-1
dV/dr (cm .nm )
0.06
3
(a) (b)
0.04
0.02
0.00 2
4
6
8
10
12
Pore diameter (nm) Fig. 1. Pore size distribution of (a) acac/silica xerogel and (b) Fe-acac/silica.
1710 1860
1628
Absorbance
(a) (b) (c) (d)
(e)
2100
2000
1900 1800 1700 Wavenumber (cm-1)
1600
1500
Fig. 2. Infrared spectra of acac/silica xerogel polymer obtained at room temperature, after heating treatment up to: (a) 100; (b) 200; (c) 300; (d) 350 C; (e) spectrum of Fe-acac/silica, after heating treatment up to 100 C. The bar value is 0.1.
The band area of the carbonyl stretching absorption of acac/silica xerogel (Fig. 2) was used to estimate the thermal stability of acac groups present in the polymer matrix, taking the overtone band of the silica at ca. 1860 cm 1 as a reference band. This normalization was necessary, considering the heterogeneity in disk thickness and taking into account the position changes of the infrared beam. The infrared band areas were almost constant up to 300 C. Further thermal treatment (350 C) results in the decreasing of a band area
139
(Table 2). The presence of the carbonyl stretching band after heating treatment up to 300 C, in vacuum, is evidence that the organic phase is thermally stable due to the presence of covalent bonds [21]. From the N2 adsorption–desorption isotherms, the BET surface areas were calculated and are presented in Table 1. The values founded show that the Fe-acac/silica xerogel polymer presents a suitable surface area and porosity to be used as catalyst. The scanning electron microscopy images obtained before and after the iron immobilization were also very similar (Fig. 3). The primary aggregated particles are not well defined, nevertheless it was possible to calculate the particle diameter. The average diameter of the particles was the same for acac/silica and Fe-acac/silica (Table 1). The aerobic epoxidation of alkenes with transition metal catalysts has been widely studied over the past decade. A well known and efficient method of alkene epoxidation in solution is the ‘‘Mukaiyama’’ system [14], in which an unfunctionalized alkene is epoxidized very efficiently using a transition metal b-diketonate complex as catalyst, molecular oxygen as oxidant and an aliphatic aldehyde as co-reactant. Different metal bdiketonate complexes of Ni(II), Co(II), Mn(II) and Fe(III) are effective as catalyst. The heterogenization of transition metal catalysts is an important goal concerning the development of clean technologies and several similar heterogeneous Mukaiyama systems have been studied. Usually these systems have lower performances than the homogeneous ones [22–26]. For example, modified mesoporous silica gel possessing immobilized cobalt ions used to epoxidize different substrates
Table 2 Thermal treatment of the acac/silica xerogel polymer Heat treatment (C)
Normalized CO band area
100 200 300 350
1.6 1.6 1.4 0.9
a
Band area of carbonyl group at 1710 cm 1 Band area of the silica overtone at 1860 cm
1
.
a
(cm
1
abs)
140
M.C. Brasil et al. / Reactive & Functional Polymers 63 (2005) 135–141
formance. Furthermore, under the conditions employed here, the iron ion is not leached out during use.
4. Conclusions The hybrid acac/silica xerogel polymer was easily obtained by a satisfactory way. The organic phase is thermally very stable. The acac/silica xerogel is able to bind Fe(III) ions to the acac sites in the relation Fe/acac = 1. The resulting material, Fe-acac/silica, presents appropriated morphological characteristics and promising properties to be used as heterogeneous catalyst in epoxidation reactions.
References
Fig. 3. SEM image of (A) acac/silica xerogel and (B) Fe/acac/ silica, obtained with 60,000· of magnification and 20 kV.
resulted in yields ranging from 30% (styrene) to 85% (cyclohexene) [26]. To verify the potentiality of the hybrid polymer as catalyst, we tested the Fe-acac/silica xerogel polymer in the cis-cyclooctene epoxidation using the same conditions of the homogeneous Mukaiyama system [14,22], using molecular oxygen at atmospheric pressure as oxidant and iso-butyraldehyde as a sacrificial reductant. The cis-cyclooctene was partially (39%) and selectively (100%) epoxidized to cis-cyclooctene oxide after 24 h. The experimental conditions chosen for this study (concentration of catalytic active ions, ratio of reactants, amount of catalyst, temperature, and pressure) have not been optimized for maximum conversion and selectivity. There is considerable scope for achieving improvements in catalytic per-
[1] A.P. Wight, M.E. Davis, Chem. Rev. 102 (2002) 3589. [2] J. Kramer, E. No¨llen, W. Buijs, W.L. Driessen, J. Reedijk, React. Func. Polym. 57 (2003) 1. [3] D.E. de Vos, M. Dams, B.F. Sels, P.A. Jacobs, Chem. Rev. 102 (2002) 3615. [4] Z. Lu, E. Lindner, H.A. Mayer, Chem. Rev. 102 (2002) 3543. [5] H.B. Li, Y.Y. Chen, React. Func. Polym. 55 (2003) 171. [6] L.T. Arenas, T.A.S. Aguirre, A. Langaro, Y. Gushikem, E.V. Benvenutti, T.M.H. Costa, Polymer 44 (2003) 5521. [7] K.J. Shea, D.A. Loy, Chem. Mater 13 (2001) 3306. [8] F.A. Pavan, H.S. Hoffmann, Y. Gushikem, T.M.H. Costa, E.V. Benvenutti, Mater. Lett. 55 (2002) 378. [9] F.A. Pavan, Y. Gushikem, C.C. Moro, T.M.H. Costa, E.V. Benvenutti, Colloid Polym. Sci. 281 (2003) 173. [10] S.V.M. de Moraes, J.B. Passos, P. Schossler, E.B. Carama˜o, C.C. Moro, T.M.H. Costa, E.V. Benvenutti, Talanta 59 (2003) 1039. [11] J.R. Grego´rio, A.E. Gerbase, M. Martinelli, L.D. Freitas, M.L.A. Von Holleben, M.A.M. Jacobi, P.D. Marcico, Macromol. Rapid. Commun. 21 (2000) 401. [12] A.E. Gerbase, J.R. Grego´rio, M. Martinelli, M.C. Brasil, A.N.F. Mendes, Grasas Aceites 53 (2002) 175. [13] A.E. Gerbase, J.R. Grego´rio, M. Martinelli, M.C. Brasil, A.N.F. Mendes, J. Am. Oil. Chem. Soc. 79 (2002) 179. [14] T. Takai, E. Hata, T. Yamada, T. Mukaiyama, Bull. Chem. Soc. Jpn. 64 (1991) 2513. [15] J.L. Foschiera, T.M. Pizzolato, E.V. Benvenutti, J. Braz. Chem. Soc. 12 (2001) 159. [16] L.C.M. Pinto, PhD Thesis, Universidade de Sa˜o Paulo, IPEN, 1996. [17] E.P. Barret, L.G. Joyner, P.P. Halenda, J. Am. Chem. Soc. 73 (1951) 373.
M.C. Brasil et al. / Reactive & Functional Polymers 63 (2005) 135–141 [18] S. Brunauer, P.H. Emmett, E. Teller, J. Am. Chem. Soc. 60 (1938) 309. [19] Y. Yamada, Y. Ichihashi, H. Ando, A. Ueda, H. Shioyama, T. Kobayashi, Chem. Lett. 32 (2003) 208. [20] K. Nakamoto, Infrared Spectra of Inorganic and Coordination Compounds, second ed., Wiley, New York, 1970, p. 247. [21] F.A. Pavan, S.A. Gobbi, T.M.H. Costa, E.V. Benvenutti, J. Therm. Anal. Calorim. 68 (2002) 199.
141
[22] Z.-H. Ma, T.-J. Wang, Y.-Y. Yan, M.-Y. Huang, Y.-Y. Jiang, Angew. Makrom. Chem. 243 (1996) 69. [23] E. Bouhlel, P. Laszlo, M. Levart, M.-T. Montaufier, G.P. Singh, Tetrahedron Lett. 34 (1993) 1123. [24] R. Raja, G. Sankar, J.M. Thomas, Chem. Commun. (1999) 829. [25] R.-M. Wang, H.-X. Feng, Y.-F. He, C.-G. Xia, J.-S. Suo, Y.-P. Wang, J. Molec. Catal. A: Chem. 151 (2000) 253. [26] A.J. Butterworth, J.H. Clark, P.H. Walton, S.J. Barlow, Chem. Commun. (1996) 1859.
