ARTICLE pubs.acs.org/Langmuir
The Synthesis of Mesoporous TiO2/SiO2/Fe2O3 Hybrid Particles Containing Micelle- Induced Macropores through an Aerosol Based Process Xiangcun Li,†,‡ Vijay T. John,‡,* Jingjing Zhan,‡ Gaohong He,†,* Jibao He,§ and Leonard Spinu|| †
)
State Key Laboratory of Fine Chemicals, The R&D Center of Membrane Science and Technology, Dalian University of Technology, Dalian, China, 116012 ‡ Department of Chemical and Biomolecular Engineering, Tulane University, New Orleans, Louisiana 70118, United States § Coordinated Instrumentation Facility, Tulane University, New Orleans, Louisiana 70118, United States Advanced Materials Research Institute, University of New Orleans, Louisiana 70148, United States
bS Supporting Information ABSTRACT: Mesoporous SiO2/TiO2/Fe2O3 particles containing macropores of about 50 nm in diameter have been prepared by an aerosol process using cetyltrimethylammonium bromide (CTAB) as a templating agent. In contrast to the traditional templating effect of CTAB to form ordered mesoporous silicas, the morphology here is vastly different due to the presence of precursor iron salts. The particles have mesoporosity templated by CTAB but additionally have large voids leading to a combined macroporous and mesoporous structure. The morphology is explained through the formation of colloidal structures containing species such as CTAþX1Fe3þ colloids in the aerosol droplets, indicating of a salt bridging effect. This dual porosity has applied implications, as the macropores provide easy entry to the particle interior in potentially diffusion limited situations. Furthermore, the particles encapsulate Fe2O3 and contain TiO2 leading to the dual functional properties of magnetic response and photocatalytic activity.
’ INTRODUCTION The fabrication of mesoporous materials with tailored structures has drawn considerable attention for both practical reasons and scientific attractions.13 In recent years, SiO2, TiO2, SiO2/TiO2, and various noble-metal-embedded mesoporous inorganic materials have been synthesized through templating self-organized supramolecular assemblies of small molecules, surfactants, and block copolymers. Such porous materials, while of great interest in applications driven technologies, are also of considerable scientific interest since the concepts of templating open up the possibility of tuning pore dimensions and controlling nanostructure framework morphologies.411 An interesting and useful method of forming ordered mesoporous silica based materials is the use of aerosol containing precursors together with templating agents.12 Here, the precursor solutions are aerosolized and the resulting droplets are passed through a furnace where hydrolysis and condensation of silica occurs to form spherical particles whose pores are templated by the templating agent. The particles can be easily collected over a filter. Variations of the technology have been considerably expanded to systems containing nanoparticles embedded in mesoporous silica micrometer and submicrometer spherical particles.1315 In a recent work from this laboratory, we found that the addition of iron salts (FeCl3) to a precursor solution of r 2011 American Chemical Society
cetyltrimethylammonium bromide (CTAB, a cationic templating agent to prepare ordered mesoporous silica) and tetraethyl orthosilicate (TEOS) leads to a total loss of mesoporous structure upon aerosolizing the materials.13 In its stead, the submicrometer particles have a distinct hollow sphere morphology with a dense silica shell of about 25 nm and a hollow interior. In the absence of the iron salts, the silica particles have the traditional ordered hexagonal mesoporosity. We have rationalized this observation through the hypothesis that electrostatically induced bridging between ferric colloids and surfactant micelles locks up the surfactant from participating as a template to localize the surfactant to the interior of the particle. Silica therefore condenses on the external surface of the aerosol droplet without any induced pore structure, thereby forming the shell, while the interior contains the agglomerated surfactant and iron salts. The short residence times in the aersolization process prevent full burnoff of surfactant, allowing these agglomerated entities to stay within the droplet interior as hydrolysis and condensation of silica occurs. Subsequent calcination leads to the Received: December 31, 2010 Revised: March 19, 2011 Published: April 18, 2011 6252
dx.doi.org/10.1021/la105149p | Langmuir 2011, 27, 6252–6259
Langmuir
ARTICLE
Figure 1. SEM (a),(c) and TEM (b),(d) images of mesoporous TiO2/SiO2/F2O3 hybrid particles prepared by the aerosol based prcess (a),(b) as-synthesized particles, (c),(d) particles calcined at 500 °C for 3 h.
formation of hollow silica shells containing a single magnetic iron oxide rattle-type particle in the core.13 In this work, we exploit this finding to the synthesis of SiO2TiO2 hybrid particles to determine if similar structures can be formed when a TiO2 precursor is added to the system. Our objective was to determine if the same hollow sphere with a rattle-type magnetic particle could be generated with a SiO2TiO2 shell. We found we were not able to generate these hollow particles if the titania precursor (titanium isopropoxide) was added to the system due to the extremely rapid formation of titania. In its stead, we found a novel morphology of macropores (>30 nm) and mesopores (24 nm) containing dispersed iron oxide nanoparticles. The work describes the formation of these dual porosity materials. Such bidisperse pore structures have tremendous use as the macropores allow easy access to particle interiors, especially in cases where the reaction rate is diffusion limited. The evolution and characterization of such porous materials is the subject of this work.
’ EXPERIMENTAL SECTION Materials. Titanium isopropoxide (TIP), ferric chloride hexahydrate (FeCl3 3 6H2O), tetraethyl orthosilicate (TEOS), cetyltrimethylammonium bromide (CTAB), and all of the other chemical reagents were purchased from Sigma-Aldrich and were used without further treatment. Preparation of Mesoporous TiO2/SiO2/Fe2O3 Particles Containing Macropores. The aerosol precursor solution was prepared by mixing 6.1 mmol FeCl3 3 6H2O and 3.0 mmol CTAB in absolute ethanol (15 mL, 257.3 mmol) first, followed by sonication for 5 min. To this solution, 7.0 mmol of TIP, 14.4 mmol of TEOS, and 2.0 mL of 0.1 mol/L HCl were added, respectively. The resulting
precursor solution with a molar ratio of FeCl3 3 6H2O:TIP:TEOS: CTAB:HCl:EtOH = 1:1.1:2.4:0.5:0.03:42.2 was aged for 30 min under magnetic stirring before being passed through the aerosol apparatus, described in detail in our earlier work.13,16 The precursor solution was first atomized to form aerosol droplets, and was then sent through a heating zone where preliminary solvent evaporation and silica/titania condensation occur. The temperature of the heating zone was set at 400 °C. The aerosol particles were collected by a filter membrane which was maintained at 80 °C to avoid solvent condensation. The assynthesized particles were calcined in air at 500 °C for 3 h (the same duration was used when samples were calcined at 700 and 900 °C, respectively) to remove the surfactant and solvent, and the mesoporous TiO2/SiO2/Fe2O3 particles containing macropores were obtained. Additionally, two sets of control experiments were performed by varying the FeCl3 3 6H2O or CTAB concentration in aerosol solution in order to reveal the formation mechanism of the mesoporous materials. Characterizations. The morphology of the porous and hollow particles were characterized using field emission scanning electron microscopy (FESEM, Hitachi-4800, operated at 3 kV), and transmission electron microscopy (TEM, JEOL 2010, operated at 120 kV). The XRD patterns were recorded on a Scitantag XDS 2000 powder diffractometer at a 2θ scan rate of 2°/min (Cu KR radiation at 1.54 Å). Nitrogen adsorptiondesorption isotherms at 77K for the samples were obtained using the Micromeritics ASAP 2010. The BET specific surface area was determined by a multipoint BET method, and the pore size distribution calculated using the BarretJoynerHalendar (BJH) method. The FTIR spectra were obtained using a PerkinElmer instrument (Spectrum GX). The magnetic properties of the hollow particles were obtained using a vibrating sample magnetometer (VSM). The photocatalytic activity characteristics of each TiO2 sample was evaluated by the degradation of Rhodamine B in deionized water. The 6253
dx.doi.org/10.1021/la105149p |Langmuir 2011, 27, 6252–6259
Langmuir
ARTICLE
Figure 2. (a) High resolution TEM image of aerosol particle calcined at 500 °C, (b) TEM image of the aerosol particles calcined 900 °C, (c) element analysis of the particles calcined at 500 °C for 3 h, and (d) the element analysis of the black spots in part b (the red square). reaction was carried out in a RPR-100 Rayonet reactor (1.65 108 photons/s/cm3). The bulbs produced a strong peak centered at 254 nm. In a typical process, 10 mg of TiO2 was added to 50 mL of a 1.0 105 mol 3 L1 Rhodamine B solution and magnetically stirred in the dark for 30 min prior to irradiation, to achieve adsorption equilibrium of Rhodamine B with the catalyst. The samples were collected every 20 min and the degradation rate measured by UVvis adsorption (553.5 nm, Shimadzu UV 1700).
’ RESULTS AND DISCUSSION Figure 1 illustrates the morphologies of particles formed with the baseline concentration ratios of 1:1.1:2.4:0.5:0.03:42.2 for the precursors, FeCl3 3 6H2O:TIP:TEOS:CTAB:HCl:EtOH. Figure 1a,c shows the representative SEM images of mesoporous TiO2/SiO2/Fe2O3 particles containing macropores before and after calcination at 500 °C, respectively. The particles are spherical in shape with diameters in the range of 50800 nm,
characteristic of the polydisperse particles synthesized by the aerosol technique. The particles reveal a smooth surface but also indicate the presence of many macropores with a diameter of about 50 nm on the surface. Upon calcination (Figure 1c), the pore density on the particle surface increases compared with that of the as-synthesized particles due to the removal of solvent and surfactant. Calcination at 500 °C for 3 h decomposes CTAB molecules absorbed on the particle surface and entrapped within the particles.26 CTAB oxidation products subsequently diffuse out of the mesoporous particles. The specific structure of the mesoporous particles containing macropores is further confirmed by the TEM images in Figure 1b,d respectively, which agree with the SEM results. We propose that the macropores are templated by ferric colloid decorated CTAB micelles in the aerosol droplets, but will defer to later sections to more fully characterize the development of particle morphology. Figure 2a shows the high resolution TEM image of the calcined particle, indicating the porous structure, with mesopore 6254
dx.doi.org/10.1021/la105149p |Langmuir 2011, 27, 6252–6259
Langmuir
ARTICLE
Table 1. Isotherm Parameters and the Composition of the Aerosol Particles Calcined at Different Temperatures surface areaa [m2/g]
pore volumeb [cm3/g]
average pore diameter[c] [Å]
atomic ratio Si/Ti/Fe
500 °C
187
0.15
49
61.33/28.59/10.07
700 °C
33
0.06
77
61.63/23.15/15.22
900 °C
14
0.04
149
61.76/23.74/14.50
as-synthesized
14
0.04
170
61.58/25.22/13.20
P-25
58
0.23
153
Degussa
samples
a BET surface area. b Total pore volume, obtained from the volume of N2 adsorption at P/P0 = 0.995. [c] Average pore diameter, estimated using the adsorption branch of the isotherm.
size distributions are 23 nm and 4050 nm, respectively. The mesopores (∼2 nm) are derived from CTAB-templating and appear to be disordered. The observation is ascribed to the dispersion and growth of titania and Fe2O3 nanocrystals in the TiOSi structures.6,11,17 The titania and Fe2O3 nanocrystals are randomly embedded in the mesoporous structures when the particles are calcined at high temperature. Usually, the anisotropic crystallization and growth of crystalline particles tends to exceed the geometry of the inorganic framework resulting in distortion or deterioration of the ordered mesopore structures. In addition, the growth of nanocrystals can penetrate the pores and occupy both surface and near-surface sites, leading to strong distortion or complete disorder of the mesoporous structure. However, the Fe2O3 nanoparticles entrapped in the calcined particles cannot be easily distinguished from the porous structure. The result is due to uniform distribution of Fe2O3 nanoparticles in the framework. Elemental analysis indicates that the atomic ratio of Si/Ti/Fe is 61.33/28.59/10.07 in the particles (Figure 2c, Table 1). Upon calcination of the aerosol particles at 900 °C, many black dots can be observed (identified within the squares in Figure 2b). The dots are mainly composed of Fe and O (Cu and C are background). At such high temperatures, calcination leads to sintering of Fe2O3 and these nanoparticles become more visible through TEM. Figure 2b appears to imply that the macropores and the mesopores calcined at 900 °C are reasonably well preserved. The apparent thermal stability can be ascribed to the dispersion of Si and its interactions with the titania matrix in the post-treatment. Thus, the interactions of the Si species with Ti-oxo-hydroxo oligomers do not impact the sintering and formation of nanosized magnetic crystals but instead retards the growth of the crystalline grains when the materials are calcined at a relatively high temperature.7,1820 Consequently, mesoporous TiOSi structures with nanosized Fe2O3 are obtained. Figure 3 illustrates the FTIR spectra of the porous particles calcined at different temperatures. For the as-synthesized sample, the characteristic peaks at around 3450 cm1 and 1630 cm1 can be attributed to the stretching vibration of the OH group and the OH bending vibration of adsorbed water molecules and silica.21,22 The decrease of intensity of the two peaks with increasing the calcination temperature is due to the decrease in the amount of surface adsorbed water and hydroxyl groups. The peak at about 2950 cm1 originates from the stretching vibration of the CH band of CTAB and organic solvent. The disappearance of this band for the calcined samples demonstrates that the surfactant and the organic solvent molecules can be completely removed by calcination at temperature of 500 °C or higher. The peak at 1106 cm1 and the shoulder peak at 1200 cm1 can be attributed to asymmetric SiOSi stretching vibration, and the bands at 800 and 465 cm1 can be assigned to the symmetric stretching and deformation modes of SiOSi,
Figure 3. FTIR spectra of the mesoporous TiO2/SiO2/Fe2O3 particles calcined at different temperatures.
respectively.22,23 TiO2 and SiO2 mixed oxides are characterized by a typical band at about 960 cm1, which can be due to the asymmetric SiOTi vibration. On the basis of the FTIR analysis, we can conclude that the bands of SiOSi and SiOTi construct the porous particle. Figure 4 shows nitrogen adsorptiondesorption isotherms of the particles calcined at 500 and 900 °C. The isotherms of the particles calcined at 500 °C exhibit a combination of types I and IV (BDDT classification) with two very distinct regions (Figure 4a), indicating a bimodal pore size distribution in the particles.24 At low relative pressure (0.20.4), the isotherm exhibits high adsorbed volumes, indicating that the particles contain micro/mesopores (type I), corresponding to pores with diameter of about 23 nm in Figure 2a. At high relative pressures between 0.4 and 1.0, the curve is a H3 hysteresis loop, associated with pores of about 45 nm in diameter.25 Figure 4b illustrates the pore size distribution indicating adsorption dominated by mesopores of 23 nm. BET surface areas of the sample are 187 m2/g. For the particles calcined at 900 °C, the isotherm can be categorized as type IV, with a small hysteresis loop in the range of 0.81.0 P/P0, indicating the presence of macropores in the particles. The measurement shows that the particles do not have a clear dominance in pore sizes, and the BET surface area is 14 m2/g. The about 10-fold drop in the surface area between 500 and 900 °C can be attributed to the breakdown in mesopore dominance due to the recrystallization of the framework after calcination at high temperatures.26 We therefore conclude that calcination at high temperatures maintains the overall structure of the particle but affects the integrity of the finer pore structure. 6255
dx.doi.org/10.1021/la105149p |Langmuir 2011, 27, 6252–6259
Langmuir
ARTICLE
Figure 4. The nitrogen adsorptiondesorption curves and pore size distributions of (a),(b) the particles calcined at 500 °C for 3 h, (c),(d) of the particles calcined at 900 °C.
On the basis of these observations, we propose a mechanism for the formation of mesoporous TiO2/SiO2/Fe2O3 particles containing macropores, as shown in Figure 5. CTAB micelles are first formed upon introduction of the cationic surfactant to the aerosol precursor solution. The micelles are always positively charged around the coronas because of the dissociation of the Br ions (CTAþBr). Since the ferric species (FeO(OH)) has a point of zero charge (PZC) of 7, which is much higher than that of the silicate (PZC is ∼2), CTAB prefers to interact with the more positively charged FeO(OH) at a Hþ concentration of around 0.005 M. While the hydrolysis of FeCl3 3 6H2O generates ferric colloids that are stabilized by the electrical double layers of chloride anions,27 the ferric colloids with chloride anions on the outer surface will electrostatically adhere to the positively charged CTAB micellar corona. It is assumed that the cooperative assembly of ferric colloids with CTAB micelles relies on the (CTAþX1Fe3þ) pathway, where X1 represents the halogen anions. Kim and co-workers have proposed that CTAB-stabilized nanocrystals clusters act as seeds for the formation of spherical mesoporous silica particles with macropores of ∼50 nm in diameter by solgel reaction.11 The CTAB micelles and the clusters are responsible for the mesopores and the macropores, respectively. Following the work by Kim and co-workers, we propose that the pores of ∼50 nm are templated by clusters of
ferric chloride coagulated CTAB micelles, as shown in Figure 5. The macropores form as a result of the loss of surfactant and conversion of iron chloride to oxide which are incorporated TiO2/SiO2 framework. The mesopores (∼25 nm) originate from CTAB micelles in the aerosol solution. Therefore, the introduction of CTAB not only induces the reassembly of ferric colloids but also acts as a template to the formation of the mesopores. Let us consider the sequence of the process as shown by the schematic and the micrographs of Figure 5. On the left, we show an aerosol droplet containing ferric colloids coagulated with CTAB micelles coexisting with CTAB micelles. Indeed, the cryoTEM of the aerosol solution show (bottom left of Figure 5) shows nonuniformity in the system and the evidence of spherical microstructure approximately 500 nm in size, within which are further spherical substructures. It is very difficult to fully interpret the cryo-TEMs, but it is clear that the solution to be aerosolized is not homogeneous and may indicate the presence of these ferric colloid coagulated structures. The center panels show the assynthesized particle structure where condensation of silica and titania has occurred. In general, the hydrolysis activity of titanium in the alkoxide precursor is significantly higher than that of the silica in the related alkoxide.28 This property often results in a more rapid formation of solgel titania units (Ti(OR)4) 6256
dx.doi.org/10.1021/la105149p |Langmuir 2011, 27, 6252–6259
Langmuir
ARTICLE
Figure 5. Schematic of the formation mechanism of mesoporous TiO2/SiO2/F2O3 particles.
compared to silica units. However, in this work, the hydrolysis rate of TIP and TEOS may be comparable because the addition of hydrochloric acid can promote the TEOS hydrolysis. The center panels reveal however that the TiO2SiO2 matrix does not form uniformly in the particle and indicates the formation of the nascent macropores. The panels on the right indicate the particles after calcination with the clear formation of macropores. The ferric colloids present in the nascent macropores become Fe2O3 nanoparticles adsorbed to the walls of the TiO2SiO2 matrix. Additionally, the CTAB micelles template mesoporous structures in the TiO2SiO2 matrix. To further understand the formation mechanism of the mesoporous TiO2/SiO2/Fe2O3 particles, two sets of control experiments were carried out. The first set of samples were prepared by fixing the amount of FeCl3 3 6H2O (6.1 mmol) and gradually increasing the concentration of CTAB in the aerosol precursor solution. The TEMs in Figure 6 summarize the observations. Without the addition of CTAB, solid spheres decorated with nanoparticles on the surface are obtained (Figure 6a). Essentially there is no templating here, and a dense TiO2SiO2 matrix is formed with Fe2O3 either encapsulated in the matrix or attached to the surface. The BET surface area is only 66 m2/g, indicating the absence of meso/macropores in the particles. A similar structure is seen when the CTAB concentration is increased to 1.1 mmol, indicating too low a surfactant concentration to effectively carry out templating (Figure 6b). However, when the CTAB content was increased to 2.2 mmol, we are able to observe particles with macroporous voids (Figure 6c). The results may further validate the adsorption of ferric colloids on the micelle corona by electrostatic attraction and the mechanism of macropores templated from the cooperative assembly of ferric colloids and CTAB micelles (CTAþX1Fe3þ).
The effect of iron chloride content on the microstructure of the calcined particles was investigated by the second set of samples, where we gradually increased the FeCl3 3 6H2O loading without changing the concentrations of other species in the precursor solution (3.0 mmol CTAB). Figure 6d shows the TEM image of the calcined particles without the addition of iron chloride. The particles are solid spheres with a surface area of 199 m2/g which indicates the presence of mesopores in the particles, as also shown by the high resolution TEM image (inset in Figure 6d). Increasing the FeCl3 3 6H2O content to 1.8 mmol is insufficient to yield macroporous particles, as displayed in Figure 6e. A further increase of FeCl3 3 6H2O content to 3.6 mmol leads to the formation of macroporous particle with smooth surface, as shown in Figure 6f. The two sets of control experiments indicate that the desired amount of surfactant and ferric chloride in the aerosol solution is crucial to the synthesis of mesoporous TiO2/SiO2 particles containing Fe2O3 nanoparticles and macropores. To further investigate the effect of the TIP on the formation of the porous particles, experiments were performed to prepare SiO2/Fe2O3 composite particles without the addition of TIP and with other species at the same molar ratios. In accordance with our earlier work,13 hollow silica microspheres encapsulating ferromagnetic iron oxide nanoparticles were obtained, as shown in Figure 6g,h. The hypothesis is that ferric colloids preferentially adsorb onto CTAB micelles and coagulate the colloids to form larger clusters. During the aerosol process, a silica shell is first formed due to the preferred silicate condensation on the gasliquid interface of the aerosol droplet. Subsequent drying concentrates the ferric clusters inside the silica shell and results in a silica shell/ferric core particle. Thermal treatment of the core shell particle leads to removal of the CTAB and encapsulation of 6257
dx.doi.org/10.1021/la105149p |Langmuir 2011, 27, 6252–6259
Langmuir
ARTICLE
Figure 6. Effect of CTAB and FeCl3 3 6H2O content on the morphology of the particles calcined at 500 °C. (a) 0.0 mmol CTAB, (b) 1.1 mmol CTAB, (c) 2.2 mmol CTAB, FeCl3 3 6H2O is 6.1 mmol. (d) 0.0 mmol FeCl3 3 6H2O, (e) 1.8 mmol FeCl3 3 6H2O, (f) 3.6 mmol FeCl3 3 6H2O, CTAB is 3.0 mmol. (g) and (h) TEM images of hollow SiO2 microspheres encapsulating iron oxide nanoparticles, (i) solid TiO2/SiO2/Fe2O3 particles with a molar ratio of 1:3.0:2.4:0.5:0.03:42.2 FeCl3 3 6H2O:TIP:TEOS:CTAB:HCl:EtOH.
a single iron oxide nanoparticle inside each silica hollow microsphere. The addition of TIP therefore has a significant effect on the morphology. Instead of hollow spheres, we observe the macropore containing structures that are the subject of this paper. When TIP is added, the precursor containing mixed Ti(OR)4Si(OR)4 species forms in aerosol droplets during an aerosol process with evaporation of solvent as well as rapid hydrolysis of TIP (a white colloidal solution forms even prior to aerosolization immediately) and TEOS (under acidic conditions). Calcination in air removes the CTAB and drives the sintering/ripening of the SiOTi, SiOSi, and TiOTi bonds throughout the particles. Consequently, the rapid hydrolysis and condensation of TIP and combination with silicate actually retards the aggregation and sintering of ferric colloids to form a single large iron oxide particle in aerosol droplets. As a result, we obtain mesoporous SiOTi frameworks with a dispersion of Fe2O3 nanocrystals, rather than hollow SiO2/TiO2 microspheres encapsulating a single iron oxide nanoparticle.
Figure 6i shows the effect of high TIP loading with the composition of FeCl3 3 6H2O:TIP:TEOS:CTAB:HCl:EtOH = 1:3.0:2.4:0.5:0.03:42.2 in the aerosol solution. We observe only solid TiO2/SiO2/Fe2O3 particles (Figure 6i), representing the upper limit of titania precursor in the feed to form particles with both macropores and mesopores. Beyond these concentrations, rapid hydrolysis and condensation of TIP prevent aerosolization since the precursor solution contains solids that rapidly block the aerosolizer nozzle. The photocatalytic activity of the samples was examined by measuring the photodegradation of Rhodamine B in an aqueous suspension of the hybrid particles. The degradation rate constants are calculated to be 0.020, 0.016, and 0.003 min1 for samples calcined at 500, 700, and 900 °C respectively (Supporting Information) while the as-synthesized sample has no photoactivity because of its amorphous state. However, the activities of the studied samples are lower than that of Degussa P25 (0.0340 min1) a commercially available TiO2. This result is not 6258
dx.doi.org/10.1021/la105149p |Langmuir 2011, 27, 6252–6259
Langmuir unexpected as it is due to the weak crystallinity and low content of titania on the particle surface. The molar ratio of TIP to TEOS in the precursor solution is ∼0.5:1.0 indicating that much of the titania may exist in the particles in combination with silica (TiOSi), and the particles are mainly composed of SiOTi and SiOSi bonds, as proven by the FTIR analysis in Figure 3. Additional X-ray Diffraction data, provided in the Supporting Information indicate the presence of anatase in the sample. The EDS analysis (of the SEM sample) in Figure 2c shows that the molar ratio of Si/Ti/Fe is 61.58/25.22/13.20 (Table 1) for the synthesized particles further supporting the FTIR results. Very little of the titanium atoms exist in the TiOTi state in the particles which should be responsible for the lower photoactivity of the calcined particles. Accordingly, we can speculate that titania is well-distributed in the particles to form TiOSi frameworks. From the EDS analysis and its molar ratio in aerosol solution, Fe2O3 should be distributed throughout the particles, but mainly around the macropores with diameter of ∼50 nm. Magnetic measurements indicate no magnetization at zero field indicating these particles contain superparamagnetic iron oxide nanoparticles. The results are summarized in the Supporting Information section.
’ CONCLUSIONS This work describes the formation of TiO2SiO2 spherical particles with a macropore substructure coupled with mesoporosity. Additionally, the materials contain magnetic nanoparticles indicating responsiveness to a magnetic field. The addition of FeCl3 to the precursor solution is crucial to derive the unique microstructure of the combined TiO2/SiO2/Fe2O3 particles. The particles are microspheres reminiscent of a “Swiss cheese” structure with large macropores and intervening mesoporous solid. The ability to obtain such structures through a one-step aerosol based process and its implications are an extremely useful concept in particle design for metal catalyst incorporation in porous substrates. The macropores that originate at the particle surface may be useful to allow reactants to directly penetrate to the particle interior especially in situations where the reaction is coupled with diffusion through the mesoporous structures. In other words, these mesoporous regions may be diffusion limited, but access throughout the particle is facilitated by the macroporosity. These are materials with a bidisperse pore structure that may overcome limitations of a high Thiele Modulus based on just the mesoporosity. ’ ASSOCIATED CONTENT
bS
Supporting Information. X-ray diffraction data, results of magnetization measurements, and photocatalytic properties are provided in the Supporting Information. This material is available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION Corresponding Author
*Tel: (504) 865-5883; Fax: (504) 865-6744; E-mail:
[email protected] (V.T.J.). Tel: 86-411-84707892; Fax: 86-411-84707700; E-mail:
[email protected] (G.H.).
ARTICLE
Research Funds for the Central Universities (893123). Additional funding from the PKSFI program at the Advanced Materials Research Institute of the University of New Orleans is gratefully acknowledged.
’ REFERENCES (1) Li, G.; Kang, E. T.; Neoh, K. G.; Yang, X. Langmuir 2009, 25 (8), 4361. (2) Cao, L.; Man, T.; Kruk, M. Chem. Mater. 2009, 21 (6), 1144. (3) Li, X.; John, V. T.; He, G.; Zhan, J.; Tan, G.; McPherson, G.; Bose, A.; Sarkar, J. Langmuir 2009, 25 (13), 7586. (4) Lee, D.-W.; Park, S.-J.; Ihm, S.-K.; Lee, K.-H. J. Phys. Chem. C 2007, 111 (21), 7634. (5) Agarwal, V.; Singh, M.; McPherson, G.; John, V.; Bose, A. Colloid Surf., A 2006, 281 (13), 246. (6) Fattakhova-Rohlfing, D.; Szeifert, J. M.; Yu, Q.; Kalousek, V.; Rathousky, J.; Bein, T. Chem. Mater. 2009, 21 (12), 2410. (7) Yu, J.; Su, Y.; Cheng, B. Adv. Funct. Mater. 2007, 17 (12), 1984. (8) Qian, J.; Liu, P.; Xiao, Y.; Jiang, Y.; Cao, Y.; Ai, X.; Yang, H. Adv. Mater. 2009, 21 (36), 3663. (9) Chen, Q.; Han, L.; Gao, C.; Che, S. Micropor. Mesopor. Mater. 2010, 128 (13), 203. (10) Andersson, N.; Kronberg, B.; Corkery, R.; Alberius, P. Langmuir 2007, 23 (3), 1459. (11) Kim, J.; Lee, J. E.; Lee, J.; Yu, J. H.; Kim, B. C.; An, K.; Hwang, Y.; Shin, C.-H.; Park, J.-G.; Kim, J.; Hyeon, T. J. Am. Chem. Soc. 2005, 128 (3), 688. (12) Lu, Y.; Fan, H.; Stump, A.; Ward, T. L.; Rieker, T.; Brinker, C. J. Nature 1999, 398 (6724), 223. (13) Zheng, T.; Pang, J.; Tan, G.; He, J.; McPherson, G. L.; Lu, Y.; John, V. T.; Zhan, J. Langmuir 2007, 23 (9), 5143. (14) Suh, W. H.; Suslick, K. S. J. Am. Chem. Soc. 2005, 127 (34), 12007. (15) Prakash, A.; McCormick, A. V.; Zachariah, M. R. Chem. Mater. 2004, 16 (8), 1466. (16) Zheng, T.; Zhan, J.; Pang, J.; Tan, G.; He, J.; McPherson, G.; Lu, Y.; John, V. Adv. Mater. 2006, 18 (20), 2735. (17) Wang, K.; Morris, M. A.; Holmes, J. D. Chem. Mater. 2005, 17 (6), 1269. (18) Wang, X.; Mitchell, D. R. G.; Prince, K.; Atanacio, A. J.; Caruso, R. A. Chem. Mater. 2008, 20 (12), 3917. (19) Demir€ors, A. F.; van Blaaderen, A.; Imhof, A. Chem. Mater. 2009, 21 (6), 979. (20) Song, X.; Gao, L. Langmuir 2007, 23 (23), 11850. (21) Caruso, R. A.; Susha, A.; Caruso, F. Chem. Mater. 2001, 13 (2), 400. (22) Lee, D.-W.; Ihm, S.-K.; Lee, K.-H. Chem. Mater. 2005, 17 (17), 4461. (23) Mariscal, R.; Lopez-Granados, M.; Fierro, J. L. G.; Sotelo, J. L.; Martos, C.; Van Grieken, R. Langmuir 2000, 16 (24), 9460. (24) Yu, J.; Yu, J. C.; Leung, M. K. P.; Ho, W.; Cheng, B.; Zhao, X.; Zhao, J. J. Catal. 2003, 217 (1), 69. (25) Yu, J.; Su, Y.; Cheng, B.; Zhou, M. J. Mol. Catal. A 2006, 258 (12), 104. (26) Sakatani, Y.; Grosso, D.; Nicole, L.; Boissiere, C.; Soler-Illia, G.; Sanchez, C. J. Mater. Chem. 2006, 16 (1), 77. (27) Jolivet, J.-P.; Chaneac, C.; Tronc, E. Chem. Commun. 2004, No. 5, 477. (28) Zele nak, V.; Hornebecq, V.; Mornet, S.; Sch€af, O.; Llewellyn, P. Chem. Mater. 2006, 18 (14), 3184.
’ ACKNOWLEDGMENT The work was supported by the National Natural Science Foundation of China (20776026, 21006008) and the Fundamental 6259
dx.doi.org/10.1021/la105149p |Langmuir 2011, 27, 6252–6259
