Silica-magnesia mixed oxides prepared by a modified Stöber route: Structural and textural aspects

Powder Technology 198 (2010) 337–346

Contents lists available at ScienceDirect

Powder Technology j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / p o w t e c

Silica-magnesia mixed oxides prepared by a modified Stöber route: Structural and textural aspects Rodrigo Brambilla a, Cláudio Radtke a, João H.Z. dos Santos a,⁎, Márcia S.L. Miranda b a b

Instituto de Química, UFRGS, Av. Bento Gonçalves, 9500, Porto Alegre 91501-970, Brazil Braskem S.A., III Pólo Petroquímico, Via Oeste, Lote 05, Triunfo 95853-000, Brazil

a r t i c l e

i n f o

Article history: Received 16 July 2009 Received in revised form 1 October 2009 Accepted 29 November 2009 Available online 3 December 2009 Keywords: Silica-magnesia Support Stöber AFM Sol–gel

a b s t r a c t Silica-magnesia mixed oxides were prepared from tetraethoxysilane (TEOS) and magnesium chloride (MgCl2) via a modified Stöber route. The TEOS/MgCl2 molar ratio and the MgCl2 addition time were varied from 1:0 to 0:1 and 0 to 1 h, respectively. The materials were characterized by a set of complementary techniques, namely, Fourier Transform infrared spectroscopy (FT-IR), X-ray photoelectron spectroscopy (XPS), matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS), X-ray diffraction (XRD), nitrogen adsorption, scanning electron microscopy-energy dispersive X-ray spectroscopy (SEM-EDX) and atomic force microscopy (AFM). Magnesium content in the silica-magnesia mixed oxides was between 3.1 and 28.6%. Increasing TEOS/MgCl2 resulted in higher crystallinity. The presence of Mg in the oxides increased specific surface area compared to the bare silica. Spherical grain morphology was observed for mixed oxide prepared with MgCl2 added after approximately 1 h of silica precipitation. © 2009 Elsevier B.V. All rights reserved.

1. Introduction Silica-magnesia mixed oxides have attracted significant research interest because of their many applications (e.g., in the manufacturing of glasses [1], the reduction of bioavailability [2], the synthesis of magnesium silicates [3] and the development of adsorbents [4] and heterogeneous catalysts [5,6]). Among the various synthetic strategies to prepare mixed oxides, the sol–gel method is most widely employed due to its ability to control material properties. This method is based on the hydrolysis of metal alkoxides and subsequent condensation, resulting in a threedimensional network. The nature of the alkoxide precursor, the molar ratio between metal and silicon alkoxides, the use of modifying agents, the nature of solvents and temperature all affect the microstructures and properties of the resulting materials [7]. Alternatively, it is possible to use a metal salt, such as metal nitrate, carbonate, or chloride, as the sol–gel precursor [8]. In previous work, we reported the preparation of silica-magnesia mixed oxides by the sol–gel method from tetraethoxysilane (TEOS) and magnesium chloride (MgCl2) under acidic conditions. The structural and textural properties were influenced by the TEOS/MgCl2 molar ratio employed in the synthesis [9]. The Stöber route is a type of sol–gel method that is based on the hydrolysis and condensation of TEOS, using ammonium hydroxide (NH4OH) as the catalyst and ethanol as the solvent. In this synthesis,

⁎ Corresponding author. Tel.: + 55 51 3316 7238; fax: + 55 51 3316 7304. E-mail address: [email protected] (J.H.Z. dos Santos). 0032-5910/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.powtec.2009.11.029

reaction conditions are tuned to avoid gelification, causing the precipitation of silica with a spherical morphology [10]. Spherical particles are important for application in polyolefin-supported catalysis, where the catalyst support commonly acts as a template for polymer particle growth, controlling size and morphology [11]. Many studies concerning the synthesis of metal-doped silica spheres have been reported. However, to our knowledge, the synthesis of silica-magnesia mixed oxides by the Stöber route has not been described in the literature. In the present study, we report the preparation of silica-magnesia mixed oxides by a modified Stöber route. The influences of the TEOS/ MgCl2 molar ratio and the MgCl2 addition time on the structure, texture and morphology of the resulting materials were investigated. The silica-magnesia mixed oxides were characterized by a set of complementary techniques: Fourier Transform infrared spectroscopy (FT-IR), matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) X-ray photoelectron spectroscopy (XPS), X-ray diffraction (XRD), nitrogen adsorption, scanning electron microscopy-energy dispersive X-ray spectroscopy (SEMEDX) and atomic force microscopy (AFM).

2. Experimental 2.1. Materials Tetraethoxysilane (TEOS) (98%, Merck), magnesium chloride (N99.9%, Sigma Aldrich), ammonium hydroxide (28% NH3, Nuclear) and ethanol (N99.8%, Merck) were used without further purification.

338

R. Brambilla et al. / Powder Technology 198 (2010) 337–346

2.2. Synthesis of silica-magnesia mixed oxides Silica-magnesia mixed oxides were prepared via a modified Stöber route from TEOS and magnesium chloride. The TEOS/MgCl2 molar ratio was varied from 1:0 to 0:1, and the MgCl2 addition time was varied between 0 and 60 min. In a typical preparation, 1.2 mL of TEOS was added to 120 mL of an alcoholic solution that was prepared by previous dilution of 20 mL of ammonium hydroxide in 100 mL of ethanol. The suspension was stirred until it precipitated (about 10 min). A solution of 251 mg of magnesium chloride diluted in 20 mL of ethanol was then added to the slurry. The mixture was stirred for 24 h. The resulting solid was dried under vacuum, washed with 5 × 50 mL aliquots of ethanol and finally dried at 150 °C for 6 h. 2.3. Characterization of silica-magnesia mixed oxides 2.3.1. Fourier Transform infrared spectroscopy (FT-IR) Samples were analyzed as pellets diluted in KBr by transmission FT-IR using a BOMEM FT-IR spectrophotometer (MB-102) at 25 °C, accumulating 32 scans at a resolution of 4 cm− 1. 2.3.2. X-ray photoelectron spectroscopy (XPS) X-ray photoelectron spectroscopy was performed in an OmicronSPHERA station using Mg/Kα/radiation (1253.6 eV). The anode was operated at 225 W (15 kV, 15 mA). Survey spectra were recorded with a 50 eV pass energy. Specific regions of interest, namely Si (2p), Mg (1s) and Mg (2p), were recorded with a higher resolution (pass energy of 10 eV). The detection angle of the photoelectrons (Θ) with respect to the sample normal (take-off angle) was fixed at 0° for all measurements. The C (1s) signal from adventitious carbon at 285 eV was used as an internal energy reference. All spectra were fitted assuming a Shirley background. Lines were fitted by 80% Gaussian + 20% Lorentzian functions with set values of full width at half maximum for each line. 2.3.3. Matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF-MS) Measurements were performed in an Applied Biosystem TOF/TOF spectrometer bearing a deionization MALDI source of Nd:YAG laser operating at 200 Hz. Different compounds were tested as matrix materials. However, clearer spectra were obtained without a matrix. 2.3.4. X-ray diffraction (XRD) Powder X-ray diffraction analysis was performed in a Diffractometer model D5000 (Siemens) using a Ni filter and CuKα (λ = 1.54 Å) radiation. The measurements were restricted to the 10–40 2θ region. 2.3.5. Nitrogen adsorption–desorption isotherms Samples were outgassed (10− 2 mbar) at 120 °C for at least 8 h. Adsorption–desorption nitrogen isotherms were measured at − 196 °C in a Gemini 2375 (Micromeritics, Norcross, GA, USA). Specific surface area was determined by the Brunauer–Emmett–Teller equation (P/P0 = 0.05–0.35). The pore size was calculated by the Barrett–Joyner–Halenda (BJH method) using the Halsey standards. 2.3.6. Scanning electron microscopy- energy dispersive X-ray spectroscopy (SEM-EDX) SEM-EDX experiments were carried out on a JEOL JSM/6060 and JEOL JSM/5800, respectively. The samples were initially fixed on a carbon tape and coated with gold by conventional sputtering techniques. A 10 kV accelerating voltage was employed for SEM. 2.3.7. Atomic force microscopy (AFM) Images were obtained using a Nanoscope IIIa atomic force microscope (Digital Instruments Co.) in contact mode with probes

of silicon nitride. WSXM4.0 software from Nanotec Electronic S.L. was used for image analysis. Samples were compressed in the form of tablets or fragments of roughly 16 mm2. 3. Results and discussion Silica-magnesia mixed oxides were made by adding magnesium chloride to an ethanol solution of growing Stöber precipitated silica at pH = 12. The influences of the TEOS/MgCl2 molar ratio and the MgCl2 addition time were investigated. The silica precipitation time was about 10 min for all systems, agreeing with data in the literature for standard Stöber silica synthesis [10]. 3.1. Influence of TEOS/MgCl2 molar ratio The Mg content in the silica-magnesia mixed oxides was measured by scanning electron microscopy-energy dispersive X-ray spectroscopy (SEM-EDX) as shown in Table 1. In the employed notation, the number that follows the silica-magnesia (SiMg) systems expresses the initial TEOS/MgCl2 molar ratio. Thus, SiMg20, for instance, indicates a system synthesized with 0.1 mol TEOS and 0.005 mol MgCl2. It is worth mentioning that such instrumental technique provides information of the external surface of the grains. According to Table 1, the Si/Mg molar ratios determined by SEMEDX were ranged from 2.2 to 27.1. These measured Si/Mg molar ratios were always higher than those initially present for all the investigated systems. This means that a certain amount of hydrolyzed magnesium chloride was removed during the washing step. The different silicamagnesia mixed oxides presented magnesium content ranging from 3.1 to 28%. For the TEOS:MgCl2 system with a molar ratio of 50 (SiMg50), the Mg content was lower than that of the detection limit. Silica-magnesia mixed oxides with Mg/Si content in the range of 4 to 7% obtained from TEOS and magnesium ethoxide have been previously reported in the literature. However, no washing step was performed [12]. Thus, the reported Mg content might be due to nonreacted and deposited Mg moieties. In addition, the magnesium contents of silica-magnesia mixed oxides prepared under basic conditions in the present paper were higher than those prepared under the acidic conditions reported in our previous work [9]. For instance, the SiMg0.5 system contained 28.6% magnesium when prepared under basic conditions. With the acidic route, the same system contained about 16% magnesium. This result could be attributed to differences between the hydrolysis and condensation rate of silica under acidic and basic conditions [7]. To determine the Mg/Si distribution in the different systems, X-ray mapping was performed. Fig. 1 shows the X-ray mapping for selected silica-magnesia systems. In Fig. 1, the Mg/Si distribution was more homogeneous for systems with lower magnesium content (i.e., SiMg20 and SiMg10 (Fig. 1a and b)). This result could be attributed to the formation of less separated silica and magnesia domains. For the SiMg2 and SiMg0.5 systems with higher Mg content (8.2% and 28.6, respectively), the Mg/ Si distribution was more heterogeneous (Fig. 1c and d), suggesting the formation of separated silica and silica-magnesia domains. This result Table 1 Magnesium content in silica-magnesia systems determined by SEM-EDX. System

Si/Mg ratio (mol/mol)

Mg/Si ratio (%)

SiMg50 SiMg20 SiMg10 SiMg5 SiMg2 SiMg0.5 SiMg0.5a

bLOD 27.1 23.8 16.9 9.7 2.2 4.5

bLOD 3.1 3.5 4.9 8.2 28.6 16.0

a

Silica-magnesia xerogel prepared under acid conditions.

R. Brambilla et al. / Powder Technology 198 (2010) 337–346

339

Fig. 1. Magnesium and silicon distribution determined by SEM-EDX for silica-magnesium systems: (a) SiMg20; (b) SiMg10; (c) SiMg5 and (d) SiMg0.5. Magnification = 3500×.

is expected because the magnesium chloride was added after the initial silica formation. Spot analysis by SEM-EDX was also performed. The Mg/Si contents in the selected spots are shown in Fig. 2. In Fig. 2a and b presenting the SiMg20 (3.1% Mg) and SiMg5 (4.9% Mg) systems, the measured Mg/Si content in the spot SEM-EDX analysis was similar to the Mg/Si content determined by total SEMEDX analysis. This proves that the systems with low magnesium content had homogeneous Mg/Si distribution, as previously shown by X-ray mapping SEM-EDX analysis. On the other hand, for SiMg0.5,

shown in Fig. 2c, the measured Mg/Si contents in the spot SEM-EDX analysis were more dispersed than the Mg/Si determined by total SEM-EDX analysis, indicating a more heterogeneous Mg/Si distribution for this system. To obtain structural information, the systems were further analyzed by infrared spectroscopy (IR). The FT-IR spectrum (not showed) of silica-magnesia mixed oxide presents a band centered at 3436 cm-1. This band can be attributed to the ν(OH) stretching of silanol groups that interact with water or residual ethanol

340

R. Brambilla et al. / Powder Technology 198 (2010) 337–346

Fig. 2. Mg/Si distribution of silica-magnesia mixed oxides measured by SEM-EDX: (a) SiMg20; (b) SiMg5 and (c) SiMg0.5.

molecules by H-bonding [13]. The band at 1640 cm− 1 can be assigned to the bending vibration of water molecules [13]. The band at 1404 cm− 1 can be attributed to the δas(C–H) asymmetric deformation of –CH2 groups corresponding to residual nonhydrolyzed ethoxy groups (–OEt) on the surface of the silicamagnesia mixed oxides [14]. The FT-IR spectrum also shows a band at 1080 cm− 1 that can be attributed to νas(Si–O) asymmetric stretching of siloxane rings into the silica network. Another band attributed to siloxane bending vibration appears in the spectrum near 790 cm− 1 [13]. The weak shoulder at 641 cm− 1 can be attributed to MgO bending vibration of magnesia [15]. The band assigned to MgOSi stretching below 500 cm− 1 cannot be seen in the FT-IR spectra due to the strong absorption of the silica backbone. The structural information obtained by FT-IR spectroscopy did not reveal if there are chemical bonds between silica and magnesia domains. Therefore, the systems were further analyzed by X-ray photoelectron spectroscopy (XPS). Fig. 3 shows the high resolution

Fig. 3. High resolution XPS spectra in the Mg 1s region of magnesia and the SiMg0.5 system.

XPS spectra in the Mg 1s region for magnesia and the SiMg0.5 system. According to Fig. 3, the XPS spectrum in the Mg 1s region of the synthesized magnesia only has a component centered at 1303.3 eV, which is in agreement with the reported binding energy (BE) of pure magnesia [16]. For the SiMg0.5 system, the deconvolution of the measured signal suggests the presence of two components: one centered at 1303.3 eV, attributed to Mg 1s from magnesia moieties, and one centered at 1304.7 eV, which can be assigned to Mg 1s from the magnesium silicate moieties. These results show that the silicamagnesia mixed oxides have a fraction of Mg atoms bound to Si atoms and a fraction bound to each other. There is no variation in the BEs of Mg 1s (not shown) among the silica-magnesia mixed oxides. To investigate the influence of magnesium content on the structural properties of silica-magnesia mixed oxides, the systems were analyzed by matrix-assisted laser desorption–ionization-mass spectroscopy (MALDI-TOF-MS). Fig. 4 shows mass spectra for bare silica and for selected silica-magnesia mixed oxides. As seen in Fig. 4, the main peaks of the mass spectra for silicamagnesia mixed oxides can be attributed to silica fragmentation. However, some peaks attributed to Mg-containing structures can be observed. For the SiMg 10 system (Fig. 4b), the silica-magnesia peaks can be observed at m/z = 149, 186, 219, 595 and 675. In the SiMg5 system (Fig. 4c), the silica-magnesia peaks appear at m/z = 170, 226, 259, 298, 611, 675 and 789. For the SiMg0.5 system (Fig. 4d), the silica-magnesia peaks are found at m/z = 149, 219 and 675. The proposed silica-magnesia structures obtained from the MALDI-TOFMS data are showed in the Scheme 1. In Scheme 1, different structures can be obtained by tuning the initial Mg content during the synthesis of the silica-magnesia mixed oxides. For the SiMg10 system, hydrolyzed silane structures bound to monomeric, dimeric and trimeric magnesia structures are observed (m/z = 149, 187 and 219, respectively). The trimeric structure contains a Cl atom bound to a Mg atom, indicating that magnesia is partially condensed in this system. Another polymeric structure containing magnesia in a silica network also can be observed (m/z = 595). m/z = 675 represents a magnesium chloride ion bound to a silica surface structure. This structure appears in all investigated systems, indicating that some of the magnesium chloride does not react to produce magnesium oxide. In the SiMg5 system, partially condensed magnesia bound to monomeric (m/z = 170 and 226), trimeric (m/z = 259 and 298) or polymeric siloxane (m/z = 611, 675 and 789) structures can be observed. No magnesia-containing structures are observed in the silica network for this system. For the SiMg0.5 system, a few structures containing Mg atoms can be observed. These structures represent partially condensed magnesia bound to monomeric (m/z = 149 and 219) or polymeric (m/z = 675) silica structures. In summary, all investigated systems contain a mixture of silica and magnesium-modified silica. Considering the magnesium-modified silica structures, increases in Mg content during synthesis result in less condensed magnesia structures bound to silane or silica structures. Lower Mg content creates silica-magnesia mixed oxides containing magnesia in the silica network. To investigate the influence of magnesium content on the crystallinity of silica-magnesia mixed oxides, X-ray diffraction analyses (XRD) were conducted. Fig. 5 shows X-ray diffractograms of selected silica-magnesia systems. According to Fig. 5, the diffractogram of bare silica (Fig. 5a) contains a halo at 23.25°. This halo indicates that the synthesized silica has an amorphous structure. For the SiMg5 system (Fig. 5b), very low intensity peaks are observed. The peak at 26.66° can be attributed to quartz [17], indicating the presence of non-reacted silica in the materials. The peaks at 29.96° and 32.67° are assigned to the magnesium silicate precursor (orthoenstatite precursor) [18]. Therefore, this system is primarily composed of silica and some crystalline

R. Brambilla et al. / Powder Technology 198 (2010) 337–346

341

Fig. 4. MALDI-TOF-MS spectra of silica-magnesia mixed oxides: (a) silica; (b) SiMg10; (c) SiMg5 and (d) SiMg0.5.

magnesium silicate. In the case of the SiMg0.5 system (Fig. 5c) with high Mg content, the same peaks can be observed in the X-ray diffractogram with much greater intensity, indicating a more crystalline system. For the magnesia (Fig. 5d) obtained by the

hydrolysis of pure MgCl2, the diffractogram contains a low intensity peak at 18.75° attributed to magnesium hydroxide with a lamellar structure [19]. The highest intensity peak at 31.8° can be attributed to the products of uncompleted MgCl2 hydrolysis, namely, Mgx(OH)yCl2,

Scheme 1. Structures of fragment ions for silica-magnesia mixed oxides proposed from MALDI-TOF-MS data.

342

R. Brambilla et al. / Powder Technology 198 (2010) 337–346

Fig. 5. X-ray diffractograms of silica-magnesia systems: (a) silica; (b) SiMg5; (c) SiMg0.5 and (d) magnesia.

but its exact stoichiometry remains unclear [19]. The peak at 38.2° can be attributed to brucite [20]. N2 adsorption analyses were performed to investigate the effects of Mg content on the textural properties of silica-magnesia mixed oxides. Fig. 6 shows a representative BET isotherm determined by N2adsorption for silica-magnesia mixed oxide systems. The N2-adsorption–desorption isotherm shown in Fig. 6 corresponds to the type IV BDDT classification, i.e., a mesoporous material [21]. The most characteristic feature of a type IV isotherm is the hysteresis loop, which is associated with pore condensation. According to the literature, silicas prepared by the Stöber route display a type II isotherm, which is characteristic of non-porous or macroporous materials [22]. According to the IUPAC classification, the hysteresis loop in the BET isotherm is type H1, which is often associated with porous materials consisting of regular pores without interconnecting channels [21]. The representative surface pore size distribution curve for silica-magnesia mixed oxides is shown in Fig. 7. According to Fig. 7, the silica-magnesia mixed oxides present a broad pore size distribution with pore diameters ranged from 30 to 120 Å. The surface areas and mean surface pore diameters determined from the BET isotherms are shown in Table 2. In Table 2, silica-magnesia mixed oxides have surface areas (SBET) in the range of 18–122 m2 g− 1. These areas are lower than those previously reported for silica-magnesia xerogels under acidic conditions [9]. However, the silica-magnesia mixed oxides have higher surface areas than bare silica. Increasing Mg content results in a decrease in surface area, which can be attributed to the formation of a

Fig. 7. Representative pore size distribution curve determined by N2 adsorption– desorption for silica-magnesia mixed oxides (SiMg5 system).

crystalline phase, as observed by XRD. The surface pore diameters (Dp) of silica-magnesia mixed oxides were in the range of 66–137 Å. According to IUPAC classification, these pore diameters are characteristic of mesopores [21]. To investigate the effect of Mg content on the morphology of silicamagnesia mixed oxides, scanning electron microscopy (SEM) analyses were carried out. Fig. 8 shows SEM images of silica-magnesia mixed oxides obtained with different TEOS/MgCl2 molar ratios. According to Fig. 8, the TEOS/MgCl2 molar ratio influences the morphology of silica-magnesia mixed oxides. For bare silica (Fig. 8a), 0.8 µm-diameter spherical particles are observed, in agreement with the literature data [10]. The addition of small amounts of MgCl2 in the case of the SiMg20 system (Fig. 8b) leads to a decrease in the spherical particle diameter. The mean particle diameter in this system is about 0.4 µm. Increasing Mg content in the silica-magnesia mixed oxides results in more aggregated small particles (Fig. 8c, d and e). For higher Mg content systems (i.e., SiMg0.5 and magnesia (Fig. 8f and g)), a granular morphology can be observed. Atomic force microscopy (AFM) analyses were performed to determine surface topography. These results are shown in Fig. 9. As seen in Fig. 9, the TEOS/MgCl2 molar ratio influences the surface topography of silica-magnesia mixed oxides. In the case of bare silica (Fig. 9a), the particles are between 200 and 800 nm in height. This broad distribution means that bare silica particles are not monodispersed and that the silica surface is rough. For the SiMg10 system (Fig. 9b), the silica particles are between 50 and 100 nm in height. This decrease in height could be attributed to MgCl2 terminating the sol– gel reaction [23]. Furthermore, the surface topography profile shows a range of heights indicating roughness in the silica-magnesia mixed oxide. Increasing the Mg content (e.g., the SiMg5 system (Fig. 9c)), results in surface heights between 50 and 250 nm. This broadening of the topography curve proves the existence of morphological magnesia

Table 2 Textural properties of silica-magnesia mixed oxides determined by N2 adsorption.

Fig. 6. Representative BET isotherm determined by N2 adsorption for silica-magnesia mixed oxides (SiMg5 system).

System

SBET (m2 g− 1)

Dp (Å)

Silica SiMg50 SiMg20 SiMg5 SiMg2 SiMg0.5 Magnesia

14 122 76 38 27 18 10

68 66 66 78 142 137 152

R. Brambilla et al. / Powder Technology 198 (2010) 337–346

343

Fig. 8. SEM images of silica-magnesia mixed oxides: (a) silica; (b) SiMg20; (c) SiMg10; (d) SiMg5; (e) SiMg2; (f) SiMg0.5 and (g) magnesia. Magnification = 5000×.

domains. In addition, the proximity of surface heights between 100 and 150 nm shows a decrease in surface roughness in relation to the SiMg10 system. For the bare magnesia system (Fig. 9d), surface heights remained mainly between 50 and 300 nm. 3.2. Influence of MgCl2 addition time We also investigated the effects of MgCl2 addition time on structural and textural properties. Fig. 10 shows the Mg/Si content as a function of MgCl2 addition time, measured by SEM-EDX analysis.

Fig. 10 shows that there is no considerable variation in Mg/Si content with increasing MgCl2 addition time. Therefore, the Mg content of silica-magnesia mixed oxides is independent of the amount of preformed silica. Mg/Si distributions for the systems prepared at different MgCl2 addition times were obtained via X-ray mapping during SEM-EDX analysis. These results are shown in Fig. 11. According to Fig. 11, the Mg/Si distribution is homogeneous for systems prepared with different MgCl2 addition times. Therefore, the Mg/Si distribution is not influenced by the MgCl2 addition time.

344

R. Brambilla et al. / Powder Technology 198 (2010) 337–346

Fig. 9. AFM images and surface topography profile of silica-magnesia mixed oxides: (a) silica; (b) SiMg10; (c) SiMg5 and (d) magnesia.

To investigate the effect of MgCl2 addition time on the structure of the resulting silica-magnesia mixed oxides, XRD analyses were performed. The X-ray diffractograms of these materials are shown in Fig. 12. In Fig. 12, the X-ray diffractogram of silica-magnesia mixed oxide prepared with a MgCl2 addition time of 0 min (Fig. 12a) has two diffraction peaks centered at 22.90 and 32.67°. These diffraction peaks can be attributed to the magnesium silicate precursor (orthoenstatite precursor). For systems prepared with a MgCl2 addition time of 30 min (Fig. 12b), the X-ray diffractogram also shows magnesium silicate precursor peaks at 22.90 and 23.34° [18]. The most intense peak at 26.66° can be attributed to quartz [17]. The same peaks can be observed in the X-ray diffractogram of mixed oxide prepared with a MgCl2 addition time of 60 min. However, the intensity of the

diffraction peaks is lower than that for the system prepared with a MgCl2 addition time of 30 min. Therefore, waiting longer to add MgCl2 decreases the crystallinity of the resulting mixed oxides. To investigate the effect of MgCl2 addition time on the morphology of silica-magnesia mixed oxides, SEM analyses were performed as shown in Fig. 13. The MgCl2 addition time considerably influenced the morphology of the silica-magnesia mixed oxides shown in Fig. 13. For systems prepared with a MgCl2 addition time of 0 and 5 min (Fig. 13a and b), a granular morphology can be observed. For systems synthesized with a MgCl2 addition time longer than 10 min (Fig. 13c, d and e), the spherical morphology is dominant. Thus, it is possible to control the morphology of silica-mixed oxides by tuning the MgCl2 addition time in the modified Stöber method.

R. Brambilla et al. / Powder Technology 198 (2010) 337–346

345

Fig. 12. X-ray diffractograms of the SiMg20 system prepared with different MgCl2 addition times: (a) 0 min; (b) 30 min and (c) 60 min. Fig. 10. Magnesium content measured by SEM-EDX as a function of MgCl2 addition time.

4. Conclusions Structural and textural properties of silica-magnesia mixed oxides were influenced by both the initial TEOS/MgCl2 molar ratio as well as

by the MgCl2 addition time. The use of low MgCl2 content in the modified Stöber synthesis yielded materials with higher surfaces areas and greater Mg/Si distribution homogeneities. In this case, the silica-magnesia mixed oxides contained Mg atoms within the silica network. Increasing the TEOS/MgCl2 molar ratio created more crystalline mixed oxides with a granular morphology. Tuning the

Fig. 11. Magnesium and silicon distribution determined by SEM-EDX for the SiMg20 system prepared with MgCl2 addition times of: (a) 0 min; (b) 30 min and (c) 60 min.

346

R. Brambilla et al. / Powder Technology 198 (2010) 337–346

Fig. 13. SEM images of silica-magnesia mixed oxides (SiMg20) prepared with different MgCl2 addition times: (a) 0 min; (b) 5 min; (c) 10 min; (d) 30 min and (e) 60 min. Magnification of 5000×.

MgCl2 addition time allowed for the formation of silica-magnesia mixed oxides with a spherical morphology. Acknowledgements Brambilla thanks CNPq for the grant. This work was financed by CNPq. The authors are thankful to LNLS (Project MAS #6406) for MALDI-TOF-MS measurements. References [1] [2] [3] [4] [5] [6] [7] [8] [9]

N.P. Bansal, J. Ceram. Soc. 71 (8) (1988) 666–672. H. Ono, S. Wada, J. Fac, Agr, Kyushu Univ. 51 (2) (2006) 437–440. T.V. Long, Ceram. Int. 34 (7) (2008) 1763–1766. K. Nobuhara, M. Kato, M. Nakamura, M. Takami, S. Kaneko, J. Chromatogr. A 704 (1995) 45–53. A.H. Sijpkes, R.W. Moonen, N. van der Puil, S.B.A. Hamid, USPTO Pat. Appl. 20070149390. T. López, R. Gomez, M.E. Llanos, E. López-Salinas, Mater. Lett. 38 (1999) 283–288. C.J. Brinker, G.W. Scherer, Sol–gel science, Academic Press, New York, 1990. S. Mitra, S. Sampath, J. Mater. Chem. 12 (2002) 2531–2537. R. Brambilla, C. Radtke, J.H.Z. dos Santos, M.S.L. Miranda, J. Sol–Gel Sci. Technol. (2009) 70–77.

[10] W. Stöber, A. Fink, E. Bohn, J. Colloid Interface Sci. 26 (1) (1968) 62–66. [11] J.H.Z. dos Santos, H.T. Ban, T. Teranishi, T. Uozumib, T. Sano, K. Soga, Appl. Catal., A Gen. 220 (2001) 287–302. [12] M.E. Manriquez, T. Lopez, R. Gomez, J. Sol–Gel, Sci. Technol. 26 (2003) 853–857. [13] E.F. Vansant, P. Van der Voort, K.C. Vrancken, Characterization and Chemical Modification of the Silica Surface, Elsevier, Amsterdam, 1995. [14] R. Al-Oweini, H. El-Rassy, J. Mol. Struct. 919 (2009) 140–145. [15] D.R.M. Brew, F.P. Glasser, Cem. Concr. Res. 35 (2005) 85–98. [16] M.A. Aramendía, V. Boráu, C. Jiménez, A. Marinas, J.M. Marinas, J.A. Navío, J.R. Ruiz, F.J. Urbano, Colloids Surf., A Physicochem. Eng. Asp. 234 (2004) 17–25. [17] L. Levien, C.T. Prewitt, D.J. Weidner, Am. Mineral. 65 (1980) (1980) 920–930. [18] A. Douy, J. Sol–Gel, Sci. Technol. 24 (2002) 221–228. [19] W. Jiang, X. Hua, Q. Han, X. Yang, L. Lu, X. Wang, Powder Technol. 191 (2009) 227–230. [20] J.B. Parise, K. Leinenweber, D.J. Weidner, K. Tan, R.B. Von Dreele, Am. Mineral. 79 (1994) 193–196. [21] S.J. Gregg, K.W. Sing, Adsorption, Surface Area and Porosity, Academic, New York, 1982. [22] F. Balas, M. Rodríguez-Delgado, C. Otero-Arean, F. Conde, E. Matesanz, L. Esquivias, J. Ramírez-Castellanos, J. Gonzalez-Calbet, M. Vallet-Regí, Solid State Sci. 9 (5) (2007) 351–356. [23] M.A.S. Pedroso, M.L. Dias, C. Azuma, C.G. Mothé, Colloid Polym. Sci. 278 (2000) 1180–1186.