Journal of Structural Chemistry. Vol. 54, No. 6, pp. 1034-1043, 2013 Original Russian Text © 2013 L. B. Okhlopkova, E. V. Matus, I. Z. Ismagilov, M. A. Kerzhentsev, I. P. Prosvirin, Z. R. Ismagilov
SYNTHESIS OF MESOPOROUS BIMETALLIC Pt–Sn CATALYTIC COATINGS FROM POLYNUCLEAR PRECURSORS FOR FINE ORGANIC SYNTHESIS PROCESSES L. B. Okhlopkova,1 E. V. Matus,1 I. Z. Ismagilov,1 M. A. Kerzhentsev,1 I. P. Prosvirin,1 and Z. R. Ismagilov2
UDC 544.46
A new method is developed to obtain nanosized catalytic Pt–Sn/TiO2 coatings on the inner surface of a capillary microreactor during adsorption of polynuclear carbonyl Pt–Sn complexes on mesoporous TiO2. Titanium oxide sol prepared in the presence of template (Pluronic F127 surfactant) is supported in dynamic mode. Pt–Sn bimetallic catalysts with an average particle size of 1.5-2 nm are synthesized by adsorption of − the bimetallic [Pt3(CO)3(SnCl3)2(SnCl2⋅H2O)] 2n complex followed by thermal treatment. Physicochemical n
properties of samples (thickness, structure and morphology, chemical composition of the material, electronic state, specific surface area, pore volume and size distribution) are characterized by a set of methods (HR TEM, SEM, powder XRD, XRF, XPS, low-temperature nitrogen adsorption). Conditions to prepare the uniform non-peelable Pt–Sn/TiO2 coating on the inner surface of a silica capillary with good adhesion are determined. To increase the TiO2 thickness, multilayered TiO2 films are synthesized by layerby layer deposition. The coating thickness is found to increase with an increase in the capillary diameter. The coating of a capillary with a diameter of 0.55 mm after 14-fold deposition is characterized by a thickness of 2 μm and an average pore size of 5.4 nm. The solvent effect on the adsorption of Pt–Sn carbonyl complexes into the TiO2 support is studied. The amount of the adsorbed complex increases in the following order: ethanol < acetone ∼ tetrahydrofuran. The physicochemical properties of the active component (surface concentration, dispersion, and composition) can be fine-tuned by varying the deposition method, precursor concentration in the initial solution, and temperature conditions of activation treatment. The catalyst activity in citral hydrogenation was 0.06-0.54 min–1, with the selectivity with respect to unsaturated alcohols reaching 90% at citral conversion above 95%. DOI: 10.1134/S0022476613060061 Keywords: capillary microreactors, nanostructured catalysts, mesoporous titanium oxide, sol-gel synthesis, bimetallic clusters, citral hydrogenation.
INTRODUCTION Studies in the preparation field of cost-accessible capillary microreactors are a rapidly developing branch of science. In comparison with traditional batch-type reactors, capillary microreactors have considerable advantages such as an
1
G. K. Boreskov Institute of Catalysis, Siberian Division, Russian Academy of Sciences, Novosibirsk;
[email protected]. 2Institute of Coal Chemistry and Chemical Materials Science, Siberian Division, Russian Academy of Sciences, Kemerovo. Translated from Zhurnal Strukturnoi Khimii, Vol. 54, No. 6, pp. 1003-1012, November-December, 2013. Original article submitted June 30, 2012; revised April 4, 2013. 1034
0022-4766/13/5406-1034 © 2013 by Pleiades Publishing, Ltd.
insignificant pressure drop, scalability, high heat and mass transfer rates, high selectivity, and ecological safety because small amounts of reagents and solvent are used in the process [1-3]. Residence time and heat management can be exactly tuned to avoid secondary reactions, thereby the selectivity with respect to the desired product can be enhanced for a series of consecutive reactions. This is especially important for the improvement of the ecological safety of fine organic chemistry processes with traditional bath reactors in which the amount of wastes consisting of by-products can reach 1 kg per 1 g of the target product obtained. Mesoporous thin coatings supported on microchannel walls substantially increase their geometric surface accessible to catalyst deposition. These coatings can be obtained using the sol-gel technology in combination with the evaporationinduced self assembly (EISA) method [4]. This method makes it possible to purposefully regulate the properties of the obtained material (surface structure, chemical composition and porosity) by varying its preparation conditions. The sol-gel process includes the stages of the colloidal (sol) suspension formation and the sol gel transformation into a continuous liquid phase. As a result of this process occurring at room temperature, a homogeneous oxide structure with the desired properties (hardness, chemical and thermal stability, polarity, and porosity) is formed [5, 6]. Then the solvent and surfactant are removed by drying and calcination, and thus, a strong interaction is achieved between the inorganic network and channel walls. In this work we studied the factors affecting the adhesion strength and coating thickness on the inner surface of the capillary microreactor. Polymetallic nanostructured catalyst owing to its ability to activate the necessary functional group provides the efficiency, selectivity, and a high reaction rate. However, despite numerous different methods, the regulation of the size and composition of heterogeneous catalysts by traditional methods (impregnation and deposition) remains an unsolved problem. Polynuclear clusters are the alternative precursors in which both particle composition and size are well controlled [7, 8]. Thus, when the synthesis conditions are thoroughly selected, it is possible to prepare bimetallic nanoparticles with the desired composition and structure. In the work, we synthesized a series of nanosized bimetallic Pt–Sn/TiO2 catalysts and catalytic coatings using the carbonyl Pt–Sn cluster as a precursor of the active component. The physicochemical properties of samples (thickness, structure and morphology, chemical composition of the material, electronic state, specific surface area, pore volume and size distribution) were characterized by a set of methods: high resolution transmission electron microscopy (HR TEM), scanning electron microscopy (SEM), X-ray fluorescence (XRF), and low-temperature nitrogen adsorption. The effect of the solvent nature on adsorption of Pt–Sn carbonyl complexes on TiO2 is examined. The adsorption constant and the maximum surface concentration for different solvents are calculated. The factors determining the dispersion and composition of the active component are revealed. Correlation is established between the physicochemical properties of Pt–Sn/TiO2 catalysts and their activity of the citral reaction.
EXPERIMENTAL Synthesis of Pt–Sn/TiO2 catalysts. Bimetallic carbonyl Pt–Sn complexes were synthesized by the procedure described in [9]. In order to prepare the support Pluronic F127 surfactant was dissolved with stirring in anhydrous ethanol (99.9%) to control the water concentration, then the calculated amount of H2O and HNO3 was added. Under strong stirring the calculated amount of tetraissopropoxide titanium Ti(OiPr)4 was added dropwise. The molar ratio of reagents in the sol was Ti(OiPr)4:F127SURF:EtOH:H2O:HNO3 = 1:0.003-0.009:25-40:0.7-15:0.13-1.8. Then the initial sol was stirred at room temperature for 3-24 h and used for the synthesis of the support in the form of powder and/or the formation of the TiO2 coating on the inner capillary surface. For the preparation of the powder support the sol was kept at 76% relative humidity in a desiccator over the saturated NaCl solution for 1-3 days and dried at 373 K for 2 h and calcined in multi-stage mode: heated with a rate of 1 K/min to 373 K, kept for 30 min, then heated to 473 K with keeping for 30 min, to 573 K with keeping for 30 min, and to 673 K with keeping for 120 min. Before the coating deposition on the walls the silica capillary was previously washed with 1 M NaOH at 313 K for 30 min for better adhesion of the coating. The synthesis of the TiO2 coating of the capillary was carried out by dip-coating the TiO2 sol on the inner surface of the capillary in dynamic mode in an argon flow with a 1035
withdrawal rate of the sol in the capillary of 3.6 m/h at 293 K [10]. At the ageing stage the humidity was kept at 76%. To remove the template the capillary was calcined in a muffle furnace for 2 h with stepwise temperature rise to 693 K; during layer-by-layer deposition of the sol intermediate calcination was performed in multistage mode with temperature rise up to 693 K. The Pt–Sn/TiO2 samples were synthesized by adsorption for 24 h or impregnation for 0.25 h in a СО flow with a − complex followed by drying in a CO flow for 12 h and temperature solution of the [Pt3(CO)3(SnCl3)2(SnCl2⋅H2O)] 2n n
treatment in the inert or reducing medium. For comparison we prepared samples by combined impregnation with inorganic H2PtCl6⋅6H2O and SnCl2⋅2H2O precursors. The samples are marked in accordance with the synthesis procedure: A and I correspond to the samples obtained by adsorption and impregnation respectively. The second symbol denotes the solvent: A is acetone, E is ethanol, T is tetrahydrofuran. The symbol V stands for temperature treatment in vacuum at 463 K and a pressure of 13 mbar for 2 h; OR means the annealing in air at 673 K for 2 h and in a 30% H2/Ar flow at 673 K for 2 h. The number corresponds to the Pt concentration in the initial solution (mg/ml). The Pt–Sn/TiO2 coating was synthesized by adsorption for 3 h of Pt–Sn carbonyl complexes from solutions on the ТiO2 coating in dynamic mode. After feeding the solution of Pt–Sn carbonyl complexes with a rate of 0.5 ml/h for 3 h the capillary was washed with the solvent, dried in a CO flow, and annealed in vacuum at 463 K, pressure of 13 mbar for 2 h with a heating rate of 1 K/min. Physicochemical methods to study Pt–Sn/TiO2 catalysts. The chemical analysis of the solutions under study was performed by inductively coupled plasma atom emission spectroscopy (ICP-AES). The chemical composition of the catalyst samples was determined by XRF on a VRA-30 analyzer with a Cr anode of the X-ray tube. The texture characteristics of the samples were measured by the low-temperature nitrogen adsorption method at 77 K on an ASAP 2400 Micromeritics instrument. The SEM images of the samples were taken on a JSM-6460 LV microscope (Jeol, Japan) with a resolution of 3 nm. TEM micrographs of the samples were obtained on a JEM-2010 electron microscope (JEOL, Japan) with a mesh resolution of 0.14 nm at an accelerating voltage of 200 kV. The average particle size was calculated by the formula dav = ∑nidi/∑ni for at least 200 particles. The high resolution image of periodic structures was analyzed by the Fourier method. The samples for TEM were prepared on perforated carbon substrates fixed on copper grids. Qualitative and quantitative chemical analysis was carried out using an energy dispersive spectrometer with a Si(Li) detector and a resolution of 130 eV. X-ray photoelectron spectra were measured on a MultiLab 2000 specrometer (Thermo Electron Corp, GB), MgKα radiation, hν = 1253.6 eV, 200 W). Before each cycle of experiments the binding energy scale (Ebnd) of the spectrometer was calibrated by the position of Au4f7/2 (Ebnd = 84.0 eV) and Cu2p3/2 (Ebnd = 932.7 eV) peaks. The binding energies of the measured peaks were corrected by the position of the C1s peak (Ebnd = 284.8 eV) from the internal standard corresponding to surface hydrocarbon compounds (С–С and С–Н bonds). The ratios of surface atomic concentrations were calculated from the integrated intensities of the photoelectron peaks with regard to relative sensitivity factors [11]. In addition to the survey spectrum we measured the spectral Pt4f, Sn3d, Ti2p, Cl2p, O1s, N1s, and C1s bands. The analysis of the individual spectral regions allowed us to determine the binding energies of the peaks, the ratios of the atomic concentrations of the elements on the sample surface and to identify the chemical state of the elements. For a detailed analysis of the spectra we applied the curve-fitting procedure to allow the separation of different spectral components. Studies of the activity of Pt–Sn/TiO2 catalysts in the citral hydrogenation reaction. Catalyst pellets with a particle size of 100-200 μm were fixed in a holder placed in an autoclave with a volume of 270 ml. Before the reaction the catalysts were reduced at 523 K, H2 pressure of 12 bar for 12 h. The hydrogenation reaction of 130 ml of a 0.01 M citral solution in 2-propanol was carried out at a temperature of 343 K, Н2 pressure of 12 bar, mixing rate of 1500 rev/min. Reaction products were analyzed by gas chromatography (Varian CP-3800 GC) on a CP-Sil 5 capillary column. The main reaction products were geraniol, nerol, citronellal, citronellol, and 3,7-dimethyl-1-octanol. The carbon balance with respect to carbon was closed within 99% in all experiments.
1036
RESULTS AND DISCUSSION Dynamic deposition of the mesoporous TiO2 coating. At the stage of sol formation synthesis conditions were varied to obtain the necessary degree of oligomer branching by optimizing hydrolysis rates and polycondensation. The effect of Ti:F127, Ti:Н2O, Ti:HNO3 molar ratios, the order of introduction of reaction mixture components, the sol and gel ageing time on the texture characteristics of the support was analyzed. After the calcination at 673 K for 2 h the average pore size changed from 3.9 nm to 5.1 nm, and the total pore volume was 0.06-0.16 cm3/g. The surface area increased from 82±8 m2/g to 152±15 m2/g with the Ti:F127 molar ratio increasing from 0.003 to 0.009. These parameters correspond to the literature data [12, 13]. According to the powder XRD data, TiO2 has a structure of anatase. Based on the data obtained, the optimal composition of the sol-gel mixture was determined by the concentrations of the template, acid, and water: 1Ti(O-iPr)4: 0.009F127:40ethanol:1.3H2O:0.13HNO3, which is necessary for a high specific surface area and support porosity (the data will be published later). At this sol composition mesoporous coatings on the silica substrate have a stable hexagonal structure; the cell parameter is 7 nm; porosity 25% [14]. For the synthesis of the mesoporous catalytic coating on the inner capillary surface we elaborated a synthesis procedure of the TiO2/SiO2 coating on the inner surface of the silica capillary. The effect of deposition conditions and the annealing of the TiO2 sol on the thickness and uniformity of the TiO2 coating is studied. For the formation of a uniform unpeelable TiO2/SiO2 coating with good adhesion on the inner surface of a glass capillary by layer-by-layer deposition of the TiO2 sol an intermediate annealing in the air flow is required in multistage mode: the stepwise temperature rise to 693 K with a heating rate of 1 K/min, intermediate keeping each 100 K for 30 min, finally, 120 min at 693 K (Fig. 1). At a calcination temperature of 373 K the coating peels off, which is caused by the incomplete removal of the template and a loose structure of the inorganic network. The single coating thickness is controlled by the withdrawal rate of the sol solution from the capillary [15]. A withdrawal rate of the sol of 3.6 m/h provides the coating thickness after single deposition and calcination at 693 K about 200 nm. It is found that when the number of deposition cycles and the capillary diameter are increased, the coating thickness increases (Fig. 2). The relationship between the stationary phase film thickness of the capillary column during dynamic deposition was studied by Kaiser [16] and later by Novotny with colleagues [17]. It was found that the liquid stationary phase thickness df depends on the liquid phase concentration с in the solution, the capillary radius r0, the liquid movement rate in the capillary u1, the viscosity η, and the solution surface tension σ1g df = (0.01c⋅0.5r0)⋅(ulη/σlg)0.5. From the equation it follows that the film thickness in narrow capillaries proves to be lower than that in wider ones. Thus, the TiO2 film thickness after single sol deposition is smaller in narrow capillaries. This dependence is preserved after repeated cycles of sol deposition, its drying and annealing. The TiO2 coating thickness after 14 deposition cycles decreases with a decrease in the capillary diameter (Fig. 2b).
Fig. 1. SEM image of the capillary cross-section with a diameter of 320 μm and a length of 1 m with TiO2 coating; intermediate annealing at 373 K for 1 h, 6 depositions (a), intermediate annealing at 693 K for 1 h, 9 depositions (b). 1037
Fig. 2. Dependence of the TiO2 coating thickness on the number of deposition cycles (a); capillary diameter after 14 deposition cycles (b). TABLE 1. Effect of the Number of Depositions on the Apparent density of the TiO2 Coating on the Inner Surface of the Capillary Microreactor Number of depositions
Coating thickness, μm
Weighta, mg
Apparent densityb, g/cm3
2 0.3 0.06 4 0.8 0.17 6 1.7 0.41 9 2.4 0.66 11 2.6 0.70 a Weight is determined for the silica capillary with a diameter of 1500 μm and a length of 0.08 m. b Apparent density is the weight per unit volume.
0.53 0.58 0.63 0.69 0.71
In layer-by-layer deposition with the intermediate calcination at 693 K for 2 h the coating density calculated from the coating thickness and its weight increases (Table 1). Repeated thermal treatment results in the formation of a denser structure of the coating. A typical adsorption isotherm of the TiO2 coating is shown in Fig. 3. Type IV isotherm with loop H1 at a relative pressure p/p0 > 0.8 confirms the structural mesoporosity [18]. The average pore size was 5.4 nm, the specific surface area was 130.4 m2/g, porosity 36%, coating thickness 2 μm. Synthesis of Pt–Sn/TiO2 coatings on the inner surface of the silica capillary. Bimetallic Pt–Sn catalysts supported on mesoporous TiO2 were prepared with varying the solvent nature, concentrations of bimetallic clusters in the initial solution, adsorption time, and activation conditions. For comparison by the impregnation method we synthesized samples from inorganic precursors by the impregnation method (Table 2). The metal concentration in the sample increases with an increase in the polynuclear precursor concentration in the adsorption solution in the impregnated samples. The average particle size increases from 1.5 nm to 3.2 nm when the metal concentration increases from 0.6 wt.% Pt to 3.6 wt.% Pt. Support impregnation with inorganic precursors and the annealing in air at 673 K for 2 h and in a 30% H 2/Ar flow at 673 K for 2 h result in the formation of lower dispersed crystallites and a wider particle size distribution. Adsorption isotherms of Pt–Sn and Pt carbonyl complexes obtained at room temperature are shown in Fig. 4. Adsorption isotherms of Pt–Sn and Pt carbonyl complexes were obtained at room temperature. The support contacted with the complex solution for 24 h in the argon flow (experiments on adsorption kinetics have shown that this time is enough to establish the equilibrium). Adsorption isotherms are described by the Langmuir equation Ceq/Γ(1 – Ceq) = 1/Γm⋅kexp(λΓ/RT) + Ceq⋅(kexp(λΓ/RT) – 1)/Γmkexp(λΓ/RT), where Γ, Γm, k, and Сeq are the surface concentration, the maximum concentration of adsorbed complexes, the adsorption constant, and the surface concentration of complexes in the suspension respectively. λ is E/Γm, where E is the interaction
1038
Fig. 3. Adsorption isotherm of N2 and pore size distribution of the TiO2 coating after 14 deposition cycles of the TiO2 sol. Synthesis conditions of 1 Ti(O-iPr)4:0.009 F127:40 ethanol:1.3 H2O: 0.13 HNO3, ageing of the sol for 24 h, of the gel for 24 h, capillary diameter 0.53 mm.
Fig. 4. Adsorption isotherms of carbonyl complexes at 295 K: Pt–Sn in ethanol (1), Pt–Sn in tetrahydrofuran (2), Pt–Sn in acetone (3), Pt in acetone (4).
TABLE 2. Synthesis Conditions and Properties of Pt–Sn/TiO2 Samples Sample
AE-OR-2 AE-OR-10 AE-V-2 IE-OR-2 IE-OR-33 AE-V-2-pa AE-V-10-pa AA-V-2-pa AA-V-5- pb a
Cin., mg/ml Pt
Sn
Ads. time, h
Pt, wt.%
Sn, wt.%
Pt/Sn
dav(dmax), nm
2 10 2 2
Polynuclear bimetallic precursor Pt–Sn–CO, TiO2 in the powder form 1.2 24 0.61 0.25 6 24 3.55 2.8 1.2 24 0.61 0.25 1.2 72 3.46 1.33
1.5 0.8 1.5 1.6
1.5 (5.1) 3.2 (11.3) 1.5 (2.9) –
33
Inorganic precursors, TiO2 in the powder form 20 0.25 2.16 0.96
1.4
1.8 (20.8)
2.0 7.3 2.8 4.7
Polynuclear bimetallic precursor Pt–Sn–CO, TiO2 in the coating form 1.8 3 1.6 0.4 7 3 2.3 0.6 1.9 3 1.6 1.0 2.0 3 5.4 2.08
2.4 2.3 0.9 1.6
2.8 (4.53) 3.0 (5.6) 2.6(6.09) 3.0 (8.58)
Capillary diameter 0.7 mm. Capillary diameter 0.55 mm.
b
energy of the adsorbed complexes. The calculation results of the maximum adsorbed amount (Γm) and the adsorption constant (kexp(λГ/RT)) are listed in Table 3. The use of ethanol as a solvent decreases the active component concentration, which can be due to better solvation of the Pt–Sn carbonyl complex and/or strong adsorption of ethanol on the support surface. For comparison we present the data for the Pt carbonyl complex (H2[Pt3(CO)6]n, where n = 5, 6). The introduction of tin chloride, which is a π-acceptor type ligand [19], decreases the electron density on platinum atoms, which in turn can weaken their interaction with Lewis acid centers of TiO2⋅ Long-term adsorption of the Pt–Sn polynuclear precursor causes the peeling of the coating. In the electron microscopic images of the samples prepared at longer adsorption, conglomerates with a size of up to a ten of microns are 1039
Fig. 5. Micrographs and particle size distribution of the AE-V-2-cp coating. TABLE 3. Adsorption Constant (kexp(λΓ/RT) and Maximum Surface Concentration (Γm) for Different Solvents Precursor
Solvent
kexp(λГ/RT), mol–1⋅l
Гm, μmol m–2
Pt–Sn–CO Pt–Sn–CO Pt–Sn–CO Pt–CO
Ethanol Tetrahydrofuran Acetone Acetone
6.9 11.5 8.03 16.5
1.94 2.49 2.54 4.2
observed. According to the energy dispersive analysis data their composition corresponds to the supported Pt–Sn/TiO2 coating. According to the TEM data the average particle size was 2.8 nm (Fig. 5) and increased to 3 nm with an increase in the precursor concentration (Table 2). The observed lattice spacings of 2.13 Å, 2.24 Å, and 2.32 Å do not correlate with the spacings in the cubic Pt lattice (2.265 Å) (JCPDS 4-802). These data indicate the Sn incorporation into the Pt crystal lattice and the formation of an intermetallic phase. According to the SEM-EDS and TEM-EDS chemical analysis data the atomic Pt:Sn ratio was 1:1 and corresponded to the XRF data. After deposition of the polynuclear precursor the porosity of the TiO2 coating decreases from 36% to 25.5% and the average pore size increases from 5.4 nm to 6.5 nm. A smaller pore volume for the Pt–Sn/TiO2 coating and the larger average pore size give evidence of partial blocking of narrow pores by metal nanoparticles, the availability of the active component decreases together with the transfer rate of reagents and reaction products from and to catalytic sites. Similar patterns were observed for Pd–Zn/TiO2 samples. We have shown that an increase in the catalyst pore size can be achieved by adding a cosurfactant–butanol [20]. Owing to low polarity, butanol can easily penetrate to the micelle core (PPO) through the hydrophilic-hydrophobic boundary causing the swelling of micelles and increasing the mesopore size [21, 22]. Activity of Pt–Sn/TiO2 catalysts. The catalytic activity of Pt–Sn/TiO2 samples in the citral hydrogenation reaction was measured at 343 K and a pressure of 12 bar (Table 4). The introduction of tin into the Pt/TiO2 catalyst diminishes the catalyst activity and enhances the selectivity with respect to unsaturated alcohols (UAs), which agrees with the literature data [23, 24]. Pt–Sn/TiO2 catalysts prepared from polynuclear bimetallic precursors and subjected to redox thermal treatments show the activity within 0.06-0.54 min–1 and the selectivity with respect UAs of 65-90%. At the beginning of the reaction the selectivity with respect to UAs increases with an increase in citral conversion and then it flattens. Citral is mainly conversed into UAs whereas citronellal and citronellol form in smaller amounts. The saturated product (3,7-dimethyl-1-octanol) forms at high conversion of citral. Redox treatment of the samples increases the selectivity with respect to UAs (AE-OR-2 and AEV-2 samples). The AE-OR-10 sample prepared at higher precursor concentrations in the solution (10 mgPt/ml) showed high selectivity with respect to UAs (90%) and activity (0.54 min–1). It is possible to assume the activity and selectivity of the samples to depend on the size and electronic state of metal particles. As seen from Table 4 data, the activity of the samples prepared from Pt–Sn carbonyl complexes increased with an increase in the particle size. According to the literature data [25], 1040
Fig. 6. Adsorption isotherm of N2 and pore size distribution of the TiO2 coating before (1) and after deposition of the Pt–Sn carbonyl complex (2).
Fig. 7. Pt4f (a) and Sn3d (b) X-ray photoelectron spectrum of TiO2, AE-OR-2, and IE-OR-2 samples. TABLE 4. Activity, Selectivity at 95% Conversion, Size and Electronic State of the Active Component of Pt/TiO2 and Pt–Sn/TiO2 Catalysts in the Citral Hydrogenation Reaction [27] Sample
Initial TOF, min–1
d, nm
Pt–AE–OR–2
5.6
— 1.5 1.5 — 3.2
AE-V-2 AE-OR-2 IE-OR-2 AE-OR-10
0.18 0.06 0.35 0.54
Pt4+
Sn
SnOx
Polynuclear monometallic precursor Pt–CO 3 — —
—
—
—
Polynuclear bimetallic precursor Pt–Sn–CO 39 0.0 100.0 65 70.2 21.6 80 52.1 33.9 90 — —
0.0 8.3 14.0 —
0.0 7.8 12.8 —
100.0 92.2 88.6 —
Selectivity, %
Pt0
Pt2+
the activity of Pt/TiO2 increased from 0.3 min–1 to 2.4 min–1 with an increase in the particle size from 1.1 nm to 3.8 nm. According to the XPS data, Pt and Sn are present in the samples in both oxidized and metal states (Fig. 7). The ratio of forms depends on the preparation method of catalysts (Table 4). The observed correlation between the selectivity of the samples and 1041
the metallic Sn concentration indicates that the samples with a high metallic Sn concentration are most selective in the hydrogenation reaction of the carbonyl group. The occurrence of tin in the metallic state gives evidence of the possible formation of a Pt-Sn alloy [26].
CONCLUSIONS Mesoporous catalytic coatings were synthesized on the inner surface of the silica capillary by the sol-gel method using the Pluronic F127 template. The coating thickness is controlled by layer-by-layer deposition of the sol and the capillary diameter. When the capillary diameter increases from 250 μm to 1500 μm the coating thickness increases from 1.3 nm to 2.7 nm after 10-fold deposition of the sol with the composition 1Ti(O-iPr)4:0.009F127:40ethanol:1.3H2O:0.13HNO3 and the intermediate calcination in the air flow in stepwise temperature rise mode up to 693 K. Pt–Sn nanoparticles with the average size of 2.8 nm supported on the TiO2 coating were obtained by adsorption of the polynuclear [Pt3(CO)3(SnCl3)2× − anion. It is possible to regulate the metal concentration and the average particle size distribution by varying (SnCl2⋅H2O)] 2n n
the solvent nature, precursor concentration in solution, preparation method, and thermal treatment of the samples. The selectivity with respect to UAs in the citral hydrogenation reaction increases from 39% to 80% with an increase in the metallic tin concentration in the samples; the activity of catalysts increases from 0.06 min–1 to 0.54 min–1 with an increase in the average particle size from 1.5 nm to 3.2 nm. The authors are grateful to A. V. Ishchenko for TEM and TEM-EDS studies of the samples, to Dr. V. A. Ushakov for conducting powder XRD measurements, and to E. A. Suprun for SEM and STEM-EDS studies of the coatings. The work was supported by the Presidium of the Siberian Division of the Russian Academy of Sciences (Integration project No. 76) and RFBR (project No. 12-03-00722).
REFERENCES 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 1042
P. Watts and S. J. Haswell, Chem. Soc. Rev., 34, 235 (2005). S. V. Gokhale, R. K. Tayal, V. K. Jayaraman, and B. D. Kulkarni, Int. J. Chem. React. Eng., 3, R2 (2005). M. Brivio, W. Verboom, and D. N. Reinhoudt, Lab Chip., 329 (2006). E. V. Rebrov, A. Berenguer-Murcia, B. F. G. Johnson, and J. C. Schouten, Catal. Today, 138, Nos. 3/4, 123 (2008). U.-H. Lee, H. Lee, S. Wen, S. Mho, and Y.-U. Kwon, Micropor. Mesopor. Mater., 88, 55 (2006). P. C. A. Alberius, K. L. Frindell, R. C. Hayward, E. J. Kramer, G. D. Stucky, and B. F. Chmelka, Chem. Mater., 14, 3284 (2002). C. Audo, J. F. Lambert, M. Che, and B. Didillon, Catal. Today, 65, 157 (2001). J. M. Thomas, R. Raja, B. F. G. Johnson, T. J. O’Connell, G. Sankar, and T. Khimyak, Chem. Com., 1126 (2003). N. B. Shitova, Yu. D. Perfil’ev, L. Ya. Al’t, and G. G. Savel’eva, Zh. Neorg. Khim., 46, No. 3, 438 (2001). E. V. Rebrov, A. Berenguer-Murcia, H. E. Skelton, B. F. G. Johnson, A. E. H. Wheatley, and J. C. Schouten, Lab Chip., 9 (2009). J. H. Scofield, J. Electron Spectr. Relat. Phenom., 8, 129 (1976). T.-D. Nguyen-Phan, M. B. Song, E. J. Kim, and E. W. Shin, Micropor. Mesopor. Mater., 119, 290 (2009). M. M. Yusuf, H. Imai, and H. Hirashima, J. Sol-Gel Sci. Tech., 28, 97 (2003). L. N. Protasova, E. V. Rebrov, T. S. Glazneva, A. Berenguer-Murcia, Z. R. Ismagilov, and J. C. Schouten, J. Catal., 271, 161. E. V. Rebrov and J. C. Schouten, Chem. Eng. Proc., 50, 1063 (2011). R. Kaiser, Chromatographie in der Gasphase, II Teil: Kapillarchromatographie, Bibliogr. Inst., Manheim (1961). M. Novotny, K. D. Bartle, and L. Blomberg, J. Chromatogr. Sci., 45, 469 (1969). T. R. Pauly, Y. Liu, T. J. Pinnavaia, S. J. L. Billinge, and T. P. Rieker, J. Am. Chem. Soc., 121, 8835 (1999).
19. G. W. Parshal, J. Am. Chem. Soc., 88, 704 (1966). 20. L. B. Okhlopkova, M. A. Kerzhentsev, and Z. Ismagilov, in: Topics in Chemistry and Material Science, Advanced Micro- and Mesoporous Materials, K. Hadjiivanov, V. Valtchev, S. Mintova, and G. Vayssilov (eds.), Heron Press Science Series, Bulgaria, 6, 205-216 (2011). 21. S. Y. Choi, M. Mamak, N. Coombs, N. Chopra, and G. A. Ozin, Adv. Funct. Mater., 14, 335 (2004). 22. K. Liu, H. Fu, K. Shi, B. Xin, L. Jing, and W. Zhou, Nanotechnology, 17, 3641 (2006). 23. J. C. S. Wu and W.-C. Chen, Appl. Catal. A, 289, 179 (2005). 24. A. M. Silva, O. A. A. Santos, M. J. Mendes, E. Jordao, and M. A. Fraga, Appl. Catal. A, 241, 155 (2003). 25. K. Utpal, M. Singh, and A. Vannice, Studies in Surface Science and Catalysis, 130, No. 1, 497 (2000). 26. J. Ruiz-Martínes, F. Coloma, A. Sepúlveda-Escribano, J. A. Anderson, and F. Rodríguez-Reinoso, Catal. Today, 133135, 35 (2008). 27. Z. R. Ismagilov, E. V. Matus, A. M. Yakutova, L. N. Protasova, I. Z. Ismagilov, M. A. Kerzhentsev, E. V. Rebrov, and J. C. Schouten, Catal. Today, 147, S81 (2009).
1043
