Highly selective epoxidation of styrene at ambient temperature under photo-assisted condition over bimetallic Eu and Ti-doped amino-functionalized mesoporous molecular sieve

Mesoporous molecular sieves silica materials had been functionalized employing 3-aminopropyltriethoxysilane as the surface decoration agent by a post-synthesis method and Eu and Ti with different loading ratio were synchronously anchored on amino-functionalized silica framework by incipient wetness impregnation method, respectively. The structures of the synthesized samples were evaluated by means of X-ray diffraction, Fourier transform infrared spectroscopy, ultraviolet-visible spectra, scanning electron microscope, transmission electron microscope, X-ray photoelectron spectroscopy and solid-state 29Si magic angle spinning nuclear magnetic resonance spectra. The results revealed that the highly ordered mesoporous two dimensional hexagonal structure and the mesoporous channel structure were well preserved and isolated tetrahedral surface Eu3+ and isolated [TiO4] or [HOTiO3] species were found to be grafted and highly dispersed in the amino-functionalized mesoporous framework. A significant enhancement in selectivity and conversion for the photocatalytic epoxidation of styrene has been achieved by synergistic effect based on surface ligand effect and electron density effect between Eu3+ and Ti4+ species on amino-functionalized mesoporous silica framework. The factors associated with the conversion of styrene and selectivity to styrene oxide such as the solvent, metal loading, the amount of NaOH, H2O2/Styrene, the amount of catalyst and reaction time were also investigated in details. The desired selectivity (84%) to styrene oxide and excellent conversion (32%) of styrene for the photocatalytic epoxidation of styrene with turnover frequency of 3.1 ×  10−4s−1 were obtained using aqueous H2O2 as an oxidant in the presence of acetonitrile solvent under ambient temperature.


Introduction
Styrene epoxidation was a commercially important reaction of high environmental and commercial interest since the formed oxides could be readily converted into fine chemicals and pharmaceuticals which used for the production of perfume, epoxy resins, drugs, textiles, surface coatings, plasticizers, sweeteners, anthelmintics, etc [1]. Currently, commercial production usually used the bromohydrin method, which brought about significant side effect on equipment corrosion and environmental pollution. The environmentally friendly catalytic methods for epoxidation of styrene employing clean oxidant were, therefore, investigated by many scientists. From economical and environmental points of view, aqueous hydrogen peroxide offered great opportunities for epoxidation of styrene due to its high oxygen content (47%), high purity, low cost, non-toxic, readily available, safe storage and easy treatment. In addition, the final reduction product for aqueous hydrogen peroxide was water and it was no pollution to the environment without troublesome byproduct [2]. Previously, a scarce attention had been devoted to develop a highly efficient catalyst which processed outstanding selectivity (regie-, stereoselective and chemo-) for high conversion of substrate, small cost, relative stability and greenization to environment by many industrial research. It was now considered, mesoporous materials had attracted considerable attention for their outstanding properties, for instance, well-defined pore size, large pore volume and unique high surface area, and what's more, their great practical potential were widely used in the fields such as catalysis, separation, drug delivery, adsorbents and sensors. Among all the mesoporous materials, mesoporous silica(popularly abbreviated SBA-15) was the most promising carriers on account of a higher surface area (600-1000 m 2 /g), narrower pore size distribution (5-30 nm) and larger pore diameters (5-10 nm). Besides, compared with other mesostructured silica materials, favorable thicker pore walls (3.0-6.0 nm) SBA-15 catalytic materials provided high hydrothermal stability which was suitable for in aqueous media use [3]. Due to a lack of diverse functionality, however, their applications were still limited.
In the case of the atomic composition, the framework of pure SBA-15 consisted mainly of amorphous SiO 2 , it could therefore not be applied to catalytic reaction directly. With regard to the molecular level, the surface of mesoporous silica SBA-15 contained three types of silanol groups (SiOH), i.e., single [(SiO) 3 Si-OH), Geminal ((SiO) 2 Si(OH) 2 ], hydrogen-bonded [(SiO) 3 Si-OH-OH-Si(SiO) 3 ], which could be observed. Only free silanol groups (including single-SiOH and geminal silanols = SiOH), however, have high chemical activity, which was accessible to the silylating agent, chlorotrimethylsilane. Hydrogen-bonded silanol groups had no chemical activity, but when it heated to certain degrees, it could change into free silanol groups. From the point of view of chemical, the silanol groups with chemical activity were primary mesoporous molecular sieve SBA-15 for further surface modification.
To overcome the aforementioned weaknesses, recently, researchers had drawn much attention on organic-inorganic hybrid [4] mesoporous materials due to the fact that the catalytic and adsorption function groups could easily be incorporated into the internal pore surfaces via the modification of walls with organic functional groups. In general, there were two applied approaches [5] for surface functionalization. One was direct synthesis or co-condensation and another one was post-synthesis grafting. By contrast, the latter was much more advantageous because it could obtain a much more uniform surface coverage, hydrothermally stable and selectively functionalized materials by organic groups. Amine-functionalized mesoporous silica using (3-aminopropyl) triethoxysilane (APTES) was found to be a promising applications such as base-catalyzed reactions [6], sensing [7], adsorption [8], toxic oxyanion sorbents and wastewater treatment, just to name a few.
In recent years, the catalytic performance of titanium [9] and rare earth metal La [10], Ce, Eu [11] supported mesoporous catalysts that had been achieved by impregnation procedures were tested and had been reported to be quite active in the selective oxidation of alkene at higher reaction temperature. Traditional catalysts were usually composed of a single metal component or functionalization of surface hydrophobicity of catalysts for the purpose of independence had been considered in order to improve catalytic reaction selectivity or conversion. Achievement of high selective oxidation for styrene to styrene oxide was still a major challenge in the development of new catalysts and environmentally benign chemistry research based on the twelve principles of green chemistry such as ambient and low temperature reaction, reduction in energy consumption, etc. Since the introduction of Ti could improve the oxidation and conversion and the alkali of Eu could boost the selectivity of epoxidation. The organic functionalization of carrier could increase the hydrophobic performance of SBA-15 and catalytic efficiency at catalytic site on the basis of preferable adsorption and desorption performance.
The objective of this study was to improve its activity and tune its selectivity based on synergistic effect via modification and tailoring of the local-structural environment of mesoporous SBA-15 framework architectures in model styrene epoxidation reaction and to achieve the precise photocatalytic epoxidation reaction in ambient temperature. Mesoporous molecular sieve SBA-15 was synthesized by conventional hydrothermal process employing tetraethylorthosilicate (TEOS) as Silicon source and triblock copolymer P123 as template. Then amino-functionalized mesoporous materials were prepared by a post-synthesis method using APTES as the coupling agent. In addition, Eu and Ti with the different ratios were successfully anchored on to amino-functionalized SBA-15 framework by impregnation method. It has been suggested that the synergistic effect of heteroatom Eu and Ti in the NH 2 -SBA-15 based on both the formation of novel bonds such as T-O-Si(T = Ti and Eu) with four-coordinate titanium/europium active centre for oxidation reaction through an electrophilic oxygen-transfer pathway and significant enhancement of surface density of acid sites contributed to the excellent selectivity and conversion in epoxidation reaction of styrene with hydrogen peroxide as the oxidizing agent under photocatalysis condition at ambient temperature.

Materials and chemical reagents
All chemicals used for experimental studies were analytical grade and were used without further purification. Nonionic block copolymer surfactant Pluronic P123 poly (ethylene glycol)-block-poly (propylene glycol)-block-poly (ethylene glycol) with composition EO 20

Synthesis of SBA-15
Mesoporous silica SBA-15 parent materials were synthesized following a hydrothermal method using nonionic triblock copolymer EO 20 PO 70 EO 20 (Pluronic P123) as the structure-directing agent and TEOS as the silica source under acidic conditions [12]. In a typical synthesis, 4.0 g of Pluronic P123 was dissolved in 120 g of 2 M HCl solution and 30 g of distilled water with stirring at 40 °C. After all of the surfactant dissolved 8.50 g of TEOS was added dropwise to the solution under vigorous stirring at 40 °C for 24 h. Then, the solution was transferred into the Teflon-lined autoclave and crystallized for 24 h at 100 °C under static condition. After crystallization, the solid product was recovered by filtration, washed repeatedly with distilled water and dried at 100 °C. To remove the surfactant organic template P123, the as-synthesized material was calcined from room temperature to 550 °C for 6 h, and the pure mesoporous SBA-15 with uniform hexagonal lamelliform morphology was obtained.

Synthesis of amine-functionalized SBA-15 silica using APT ES
The aminopropyl functionalization of the internal walls of SBA-15 was carried out by post-synthesis grafting process [13]. In this work, APTES was used as silylation reagent. Prior to surface modification, the support silica materials of calcined SBA-15 were dehydrated by drying the material at 150 °C for 3 h in vacuum to remove physisorbed moisture [14]. Then the freshly activated SBA-15(2.0 g) was dispersed in 100 mL of anhydrous toluene in a 250 cm 3 flask with continuously stirring, followed by addition of different quantity of 10 mL APTES depending on the desired amine loading. After that the mixture was refluxed at 110 °C for 6 h under a nitrogen atmosphere. Then the mixture was cooled to room temperature and the resultant white solid product was collected by filtration, washed thoroughly with anhydrous toluene and finally dried under a vacuum for 12 h at 100 °C. The obtained product was labeled NH 2 -SBA-15.

Synthesis of Titania-and Europia-Incorporated NH 2 -SBA-15 catalysts
Amino-functionalized SBA-15 materials modified with titanium and rare earth metal europium, referred to as Eu-Ti-NH 2 -SBA-15 catalysts, were prepared by impregnation methods. In all of the synthesis procedures, TBOT and

Characterization methods
The organic content of the percentages of carbon, nitrogen and hydrogen in the modified SBA-15 samples were estimated by combustion on a Vario EL III Elemental Analyser. The measurements were completed with twice for adequate results. The detailed test method was as follows. The sample was accurately weighed with one part per million balance (Mettler-Toledo MX5). It was then wrapped with tin foil, compacted and accurately weighed again. Finally, the sample wrapped above was placed on the tray of the automatic sampler in Vario EL III elemental analyzer. In CHN mode, acetanilide was selected as the standard sample and He was used as carrier gas (purity 99.996%). The sample was decomposed into stable gases such as CO 2 , N 2 and H 2 O on the basis of copper wire (purity 99.9%) catalytic oxidation and combustion reaction under the condition of oxygen rich (purity 99.995%) at high temperature. These gases through different special adsorption columns for separation were controlled respectively to enter the highly sensitive thermal conductivity detector(TCD) in turn for detection.
The percentages of C, N and H elements was automatically calculated on account of integral value of each element in the operating software. Wide-angle powder X-ray diffraction (XRD) patterns were obtained on at room temperature using a Bruker D8 Advance X-ray powder diffractometer with Ni-filtered Cu Kα radiation (λ = 1.5406 Å) and a scanning speed of 12°/min in a range of 20 °< 2θ < 80° at 40 kV and 40 mA.
Fourier transform infrared (FT-IR) spectroscopy was recorded on a Nicolet 6700 FT-IR spectrometer, using pressed KBr pellet technique. To prepare the pellets, the sample and KBr powder were firstly mixed together with a mass ratio of 1:100 ~ 1:200, and then the mixture powders were pressed to a pellet size suitable for the instrument. The spectrum was generated, collected16 times, and corrected for background noise in the frequency range of 4000−400 cm −1 .
The ultraviolet-visible(UV-Vis) diffuse reflectance spectroscopy (DRUV-Vis) measurements were recorded on a UV-3600(Japan, SHIMADZU) UV-Vis-NIR spectrophotometer equipped with an integrating sphere attachment. The analysis range was from 190 to 800 nm, and BaSO 4 was used as a reflectance standard.
Solid-state 29 Si cross polarization magic angle spinning nuclear magnetic resonance(CP-MAS-NMR) spectra were recorded at 79.46 MHz using a Bruker Advance 400 spectrometer. The chemical shifts were quoted in ppm from tetramethylsilane (TMS).
The XPS analysis was performed in ultrahigh vacuum (low 10 −7 Pa range) using ESCALAB spectrometer. Monochromated Al Kalph 120 W was used as X-ray source, 200 eV for survey; 30 eV for high resolution scans. The accuracy of the electron analyzer was 1 eV. The binding energies were referred to the adventitious hydrocarbon C 1s line set at 284.8 eV. The Shirley background and the Gaussian and mixed Gaussian/Lorentzian functions and a leastsquare routine were used for fitting of the peaks.
Scanning electron microscopy (SEM)-elements-mapping with a resolution of 5 nm at 30 kV scanning voltages was obtained on scanning electron microscope JEOL, JSM-6610LV equipped with energy dispersion spectroscopy (EDS) detector to observe the morphologies of products.
Transmission electron microscopy (TEM) images were performed on a JEM-2100 with an accelerating voltage of 200 kV (Japan). The powder products were dispersed in ethanol by ultrasonication for 15 min, and then added on carbon-coated copper grids. The liquid phase was evaporated before the grid was loaded into the microscope.

Catalytic performance
The epoxidation of styrene was carried out in the quartz test tube reactor with 0.05 g of catalyst, 5 mL acetonitrile, 1.6327 g of H 2 O 2 (30 wt %), 0.500 g of styrene under room temperature for 5 h.
The Photocatalytic reactor (made in Kai Feng, China) equipped with a 300 W medium pressure mercury lamp (Institute of Electric Light Source, Kaifeng) with a maximum emission at about 365 nm was employed as the light source to provide visible light irradiation. The lamp is put in an empty chamber of the annular tube, and circulating water passes through to keep the reaction at approximately room temperature. The distance between the light source and the surface of the reaction tubes was equal to 11 cm.
The internal standard substance of toluene was added to the solution, while the reaction was finished. The upper clear solution for the following quantitative test was obtained after standing.
The concentrations of the oxidation products were analyzed using a GC-2010 gas chromatograph equipped with a RTx-1 capillary column (25 m×0.32 mm×1.0 ím) and a flame ionization detector (FID). The temperatures of the injector, the column oven and the detector were 180 ℃, 80 ℃ and 230 ℃, respectively. The carrier gas was N 2 and the flow rate was 40 mL/min with 0.2 µL of the sample injection volume, respectively. Quantitative analysis of the reaction product was carried out on GC-2010 and quantified by GC using an internal standard.
The conversion of styrene and the selectivity to styrene oxide (SO) was calculated from the following expression: The conversion of styrene was equal to ratio for (n initial styrene − n final styrene ) to n initial styrene .
The selectivity to styrene oxide(SO) was equal to ratio for n SO to (n initial styrene − n final styrene ).
TON was defined as mole substrate reacted per mole catalyst and Turnover frequency (TOF) was calculated by the expression of [product]/[catalyst] × time(s −1 ).

Elemental analysis
The C, H, and N elemental analysis data by combustion on a Vario EL III Elemental Analyser were summarized in Table 1, and the amount of aminopropyl groups successful grafted on the surface of SBA-15 were estimated, which was around 1.43 mmol of NH 2 per gram of NH 2 -SBA-15.
In general, it was thought that the content of C originated in the residual carbon in the process for sintering and removal of organic template agent P123 in order to prepare mesoporous SBA-15. As for the content H and O, it was most likely to adsorb water and nitrogen in air for BSA-15 with excellent adsorption performance. The absorption of water in BSA-15 was also confirmed by its infrared spectrum. BSA-15 was a blank in order to eliminate the influence of adsorption and residual elements on the grafting effect.

X-ray diffraction
The small-angle X-ray diffraction patterns of parent SBA-15 and Eu, Ti-incorporated amino-functionalized SBA-15 were shown in Fig. 1. As seen from Fig. 1, parent SBA-15 and its catalysts samples showed three well-resolved reflection peaks(in the low-angle range) at 2θ values between 0.5 and 5°, which were indexed as the (100), (110), and (200) Bragg reflection peaks, associated with a typical two-dimensional p6mm hexagonal symmetry of the excellent textural uniformity of mesoporous material SBA-15. The diffraction patterns indicated that the inorganic wall structural integrity of the ordered mesoporous SBA-15 silica maintained after surface modifications with APTES and metal loading. In addition, with the increasing metal loading, the (110) and (200) lattice plane were gradually decreased, which was probably due to the destruction of the materials structure.
In other words, it was likely that the contrast between the pore walls and pores of amino-functionalized SBA-15 and Eu and Ti-incorporated amino-functionalized SBA-15 was lowered compared to pure SBA-15.
The wide-angle XRD spectra patterns of parent SBA-15 and Eu, Ti-incorporated amino-functionalized SBA-15 were shown in Fig. 2. According to this Fig. 2, parent SBA-15 and its catalysts samples showed a very broad peak at 2 theta angle of 22.8°, which was corresponding to amorphous silica pore walls. The absence of peaks attributed to titania or europia indicated the dispersion of these oxides were good within the lattice of the materials [14]. Comparing to SBA-15, the spectrum of amino-functionalized SBA-15 was offset at smaller angle.

FT-IR spectra
The IR spectrum was always used to characterize the sample of metal ion introduced Si-O framework. The FT-IR spectra of pure SBA-15 and Eu, Ti-incorporated amino-functionalized SBA-15 materials in the range of 400-4000 cm −1 were illustrated in Fig. 3. Typically, in the FT-IR spectra, a broad band approximately 3460 cm −1 was appeared in all materials, which was ascribed to the presence of silanol groups(Si-OH) as well as the O-H stretching vibration of the physically adsorbed water molecules or symmetric stretching vibration of -NH 2 groups [15]. In addition, the band at around 1630 cm −1 was related to the bending vibration of the adsorbed water or bending vibration of N-H [16], which was also identified the presence of adsorbed water or N-H groups. Moreover, the strong peak(principal band) was observed at about 1078 cm −1 corresponding to Si-O-Si asymmetric stretching vibration(anti-symmetric stretching) [17], and the bands appeared at around 810 cm −1 due to the typical symmetric stretching vibrations of Si-O-Si, while its bending mode with Si-O-T(T = Si or metal) linkage appeared at 466 cm −1 , which confirmed the existence both SBA-15  (5) framework and substitution of silicon by metal ions [18]. After the amine-functionalization process, the peaks at around 687 cm −1 , 1630 cm −1 and 3460 cm −1 were observed, which associated with the bending vibration of N-H, the symmetric bending vibration of -NH 2 and stretching vibration of NH 3 + [19], respectively, indicating the existence of amine in SBA-15 framework surface. Hence, the FT-IR spectra clearly showed the amino-functionalized SBA-15 were successfully obtained with aminopropyl groups. It was well known that the incorporation of titanium into the silica framework generated Lewis acid sites on the surface. These effective active sites for the selective oxidation of styrene with aqueous hydrogen peroxide could be mainly attributed to the presence of coordinatively unsaturated surface exposed Ti 4+ sites (tetrahedral-coordinated framework titanium). Ti 4+ or Eu 3+ ions in the vicinity of the hydroxyls carrying silicon could cause changes in the electron density around Si due to differences in electronegativity or to the local structure deformations resulting from the introduction of the titanium into the framework and weakening of the SiO-H bonds. It was worth noting that pure silica SBA-15 also exhibited two bands around 966 and 466 cm − 1 , which attributed to the Si-OH rocking vibration of silanol groups in SiO 2 at the defect sites [20]. Thus, this band could be interpreted in terms of the overlapping of both Si-OH groups and polarized Si -O δ− -M δ+ (M = Ti or Eu) bonds stretching vibrations [21]. As far as the peaks of both 1078 and 466 cm −1 were concerned, a small shift towards a higher wavenumber and the decrease in intensity of the wide band with the increase of content for titanium and europium could be noticed for the as-prepared Eu-Ti-NH 2 -SBA-15 compared to the pure SBA-15. These peak changes both shifts and intensities should be due to the increase in mean Si-O distance in the walls caused by the substitution of small silicon (radius 54 pm) by larger size Ti 4+ (radius 60.5 pm) and Eu 3+ (radius 94.7 pm). This was taken as an indication of a loss in the structural regularity and in agreement with the XRD observation. The lack of characteristic IR peaks of TiO 2 and Eu 2 O 3 , it might be interpreted that the TiO 2 and Eu 2 O 3 components doped in the framework of SBA-15, or equably dispersed on the surface of SBA-15. This was therefore generally considered to be a proof of incorporation of the heteroatom into framework of SBA-15.

UV-vis diffuse reflectance spectroscopy(DRUV-Vis)
It was known to all that DRUV-Vis spectra were a sensitive tool which was extensively used to distinguish the presence of framework and extraframework of titanium and europium species incorporated in the SBA-15 mesoporous molecular sieves [22]. Figure 4 exhibited DRUV-Vis spectra of as-prepared pure-silica SBA-15 and Eu, Ti-incorporated amino-functionalized SBA-15 materials. In general, intrinsic transition gave sharp increases in absorbance while impurity, scattering and interference effects or defect state transitions gave rise to Urbach tail absorption in semiconductors. Absorption tail in the visible-light region over 400-500 nm and strong absorption band from 200 to 390 nm in the UV-light region were observed, suggesting existence of impurities or defects state transitions in as-synthesized mesoporous catalysts. As could see from Fig. 4 a, it could be observed obviously that in the UV-Visible region, no absorption signal was detected of SBA-15 due to its pure silica mesoporous materials. In terms of Fig. 4c, there were two intense absorption bands in the UV region at about 220 and 393 nm, which indicated the highly dispersed isolated tetrahedral surface Eu 3+ species [23] and the formation of a conjugated system between NH 2 -SBA-15 and Eu 3+ , respectively, owing to NH 2 -SBA-15 coordinated with rare-earth ions Ln 3+ . Meanwhile, a steadier conjugated system was formed between rare earth metal and organic-inorganic networks through copolycondensation, polymerization and cohydrolysis. The UV-Vis spectra with different molar ratio of Eu and Ti were investigated on Fig. 4 [25]. It was generally believed that it was the direct proof of titanium atoms incorporated and highly dispersed into the frame-work of amino-functionalized mesoporous molecular sieve [26]. Comparing with Fig. 4c, all the Ticontaining materials peak positions were shifted to a longer wavelength (i.e. red-shift) [23] with an increase of the relative content of titanium loading, elucidating the Ti-O-Eu bond could be formed [27]. On the other hand, compared to the bulk anatase TiO 2 and Europium oxide, the lack of a characteristic absorption band of octahedral extra-framework titanium and/or Europium at about 300-330 nm in all the samples suggested that no separated titania (anatase) and Europium oxide phases were formed during the synthesis process [28]. It was demonstrated that Eu, Ti co-doped could greatly improve the absorption of visible light. In a sense, it was beneficial to the improvement of the catalytic activities and the selectivity of epoxidation.

29 Si-CP-MAS-NMR
It was well known that high-resolution solid-state CP-MAS-NMR spectroscopy had become an important tool for structural characterization of silica-based due to the fact that the extent of silence coupling agent incorporation into SBA-15 could be monitored. In general, the relative integrated intensities of the siloxane (Q n ) allowed the quantitative assessment of the incorporation degree of the organic moiety. As for Fig. 5, significant and resonance similar absorption could be observed for the siloxane compared to the literature [29]. The result showed that silence coupling agent was grafted in the silica wall structures, giving rise to local coordination chemical environment change in porous structure or mesopore wall thickness. This was consistent with XRD observations. The experimental 29 Si-MAS-NMR spectrum of Eu-Ti-NH 2 -SBA-15 could be closely fitted by a superposition of three Gaussian components due to Q 2 , Q 3 , and Q 4 sites as shown in Fig. 5. It was found that three peaks centered at − 90.4, − 101.5, and − 109.5 ppm were observed. The peak at − 90.4 ppm was attributed to silicon atoms with two siloxane bonds and two silanol groups, (SiO) 2 Si(OH) 2 (Q 2 ), and similarly the resonance at − 101.5 ppm was assigned to silicon atoms with three siloxane bonds and one silanol group, (SiO) 3 SiOH(Q 3 ), while the resonance at − 109.5 ppm corresponded to silicon atoms with four siloxane bonds, (SiO) 4 Si(Q 4 ). Thus higher peak area of Q 4 was evident from the spectra, which was very important for the catalytic activity of silica-based catalyst in liquid phase partial oxidation reactions in the presence of aqueous H 2 O 2 as oxidant, since, reduction of surface silanols introduces more hydrophobicity and thus more affinity towards organic substrates [30]. Compared to pure SBA-15 [31,32], the intensity of Q 2 and Q 3 bands was substantially increased for Eu-Ti-NH 2 -SBA-15, indicating the alternation of Si nuclear symmetric and distortion of neighboring framework geometry. As for Eu-Ti-NH 2 -SBA-15, the resonance peaks with low chemical shift at − 90.5 and − 101.5 ppm were attributed to not only Q 2 and Q 3 species but also the (SiO) 2 Si(OH) x Ti/Eu 2 − x (x = 0, 1) and (SiO) 3 SiTi/Eu species, respectively [33]. The results therefore showed that titanium and europium species were incorporated into the porous structure framework of SBA-15. Combination with the results of XRD and UV-Vis, the 29 Si-MAS-NMR analysis also supplied additional evidence that the titanium and europium ions were located in the framework of SBA-15 materials.

XPS analysis
X-ray photoelectron spectroscopy was used to determine the chemical state of the elements and their surface proportions in the fresh catalysts. Due to the fact that no an intense peak was found at high BE (531.0 eV) with shoulder at lower BE (529.0 eV) in Fig. 6c [35] were found, indicating that the amount of crystal lattice oxygen and surface hydroxyl in TiO 2 disappeared. In other words, there was no Ti-O-Ti in mesoporous SBA-15. This was also consistent with XRD analysis and HRTEM observation. The result indicated that the substitution of Si and Ti occurred when Ti was embedded in SBA-15 material and the surface hydroxyl group was effectively grafted." The values of the binding energies of the most intense peaks of europium (Eu 3d 5/2 ), titanium (Ti 2p 3/2 ), oxygen (O 1s) and silica (Si 2p) were showed in Fig. 6. As for Fig. 6a and b, the main structure of the Ti 2p spectra was defined by a Ti3/2p/ Ti1/2p doublet appearing due to spin-orbit coupling (S.O.). It was found that the peaks of Eu 3d 5/2 and Ti 2p 3/2 as well as 2p 1/2 were centered at 1136.5, 459.5 and 465.5, respectively. On the other hand, compared to the peak position of corresponding pure oxides, blue-shifted phenomenon was observed, indicating the formation of chemical bonds for both Ti-O-Si and Eu-O-Si. That was to say Titanium and Europium atoms were embedded into framework of SBA-15. As to the reason of blue-shifted phenomenon, it was believed that decrease of its density of electron cloud due to the fact that the electron cloud of Titanium and Europium atoms around shifted to the adjacent atom of oxygen, was responsible for enhancement of Coulomb attraction between nucleus of titanium or europium and optical electron. The corresponding red shifts in the Eu-Ti-SBA-15 catalyst were also confirmed by the following O and Si photoelectron spectroscopy. As soon as Fig. 6c was considered, the O 1s core level peak exhibited an asymmetric signal, which could be deconvoluted into three feature peaks at ca. 533.5, 534.9 and 535.9 eV, respectively. The main feature at 533.5 eV was unambiguously assigned to bridge-oxygen (BO) in Eu-Ti-SBA-15. Very small red-shifted phenomenon was recorded, indicating the formation of chemical bonds for both Ti-O-Si and Eu-O-Si due to produce of Si-O network defects, i.e., non-bridge-oxygen (NBO) and peroxy radical, which should be to the benefit of catalytic performance. The peroxy radical was formed, as free oxygen was likely to be captured by NBO or by NBO each other. In the case of red-shifted with low energy for the O 1s core level main peak, it was believed that increase of its density of electron cloud of NBO due to the fact that the electron cloud of Titanium and Europium atoms around shifted to the adjacent of Si-O bonds, was responsible for decrease of Coulomb attraction between nucleus of oxygen and optical electron based on Shielding effect of Coulomb interaction for atom nucleus to optical electron. The component peaks at 534.9 and 535.9 eV were attributable to NBO [31] and oxygen originated from bond coupling Eu-O and Ti-O or OH group of surface in catalysts [27]. The result was in agreement with that of IR analysis. The analogous red shift was also notarized by Si corelevel spectrum in Fig. 6 (d). The peak of Si 2p was divided into three peaks at 103.6, 104.6 and 104.8 eV, respectively through the deconvolution technique. It may be assumed that there were three kinds of chemical situation of silicon. The Si 2p core level peak of 103.6 eV was characteristic of Si-O-Si linkages of [SiO 4 ] 4− (tetrahedron structure) in SiO 2 . The results showed that the framework structure of SiO 2 was still maintained although Eu 3+ and Ti 4+ were imbedded via isomorphous substitution in framework of SBA-15. The Si 2p peaks of ca.104.6 and 104.8 eV were attributable to chemical situation silicon of Ti/Eu-O-Si and HO-Si-O, respectively [36]. No N core level peak was showed due to the fact that its concentration and sensitivity factor were lower.

Scanning electron microscopy (SEM) and element mapping analysis
The typical scanning electron microscopy (SEM) images of the as-prepared product were shown in Fig. 7a and d in order to obtain the regular microscopic morphologies and mesoporous structure. It was found that pure SBA-15 had typical uniformly rope-like type morphologies (domains), which was aggregated into wheat-like macrostructures (rod-like particles) [32]. As for the amino-functionalized SBA-15 and Eu-Ti-NH 2 -SBA-15 materials, they all had similar typical rope-like domains that aggregated into wheat-like macrostructures as commonly observed in the pure SBA-15 type materials. Grafting of aminopropyl group and doping via rare earth metal Eu and transition metal Ti almost had no effect on morphologies for asprepared mesoporous materials, which implying that the as-synthesized materials through post-synthesis grafting and impregnation method were prior to this mesoporous molecular sieve. Elemental mapping was an effective tool for the analysis of elemental distribution. The mapping of the elements for Eu (Fig. 7d, inset, right) and Ti (Fig. 7d, inset, left) indicated clearly that the elements of titanium and europium were dispersed uniformly in NN 2 -SBA-15, which was in agreement with the results of XRD.

Transmission electron microscopy (TEM) analysis
In order to obtain additional information of the pore structure and size, transmission electron microscopy (TEM) images of NH 2 -SBA-15 and Eu-Ti-NH 2 -SBA-15 were demonstrated in Fig. 8. As shown in Fig. 8, a typical long-range ordering hexagonal arrays of mesopore with one-dimensional channels as well as the two dimensional (2D) hexagonal (p6mm) pore arrays mesostructure [33] and the typical honeycomb resembling that of SBA-15 were confirmed. The images were recorded along two different crystallographic directions, with the incident electron beam parallel and perpendicular to the direction of main channels of SBA-15. When the electron beam was direction perpendicular to pore channels, the presence of the parallel highly ordered nanotubular pore structure was clearly evidenced from Fig. 8a and c [regular pore structure along (111)]. When the electron beam was parallel to mesoporous channels, a well-ordered hexagonally ordered array of mesoporous uniform channels could be observed from Fig. 8b and d [regular pore structure along (100)] [34,35]. In addition, the distance between two consecutive centers of hexagonal pores estimated by TEM image was approximately 9 nm (an inset in Fig. 8d). Besides, these results unambiguously confirmed that after the modification with APTES and the metal loading, the hexagonal pore structure of the modification SBA-15 could remain the same, indicating that the addition of Eu and Ti ions had no effect on the morphologies and structures [23].
The mesoporous molecular sieve SBA-15 could therefore provide a good matrix to support highly dispersed metal species.

Effect of different catalysts with various ratio of Eu to Ti
The results of epoxidation reaction of styrene over asprepared catalysts with different ratio of Eu to Ti were summarized in Table 2. As shown in Table 2, the epoxidation of styrene catalytic reaction over pure SBA-15 was of a low conversion and selectivity, and after amino-functionalized SBA-15, the conversion and selectivity for the epoxidation reaction were slightly increased. Comparing with pure SBA-15, the catalysis performance of catalysts with different molar ratio of Eu/Ti increased greatly for conversion and selectivity in the epoxidation of styrene reaction. The maximum styrene conversion (23.97%) and the highest selectivity to styrene oxide (84.52%) were obtained when Eu/Ti mole ratio was equal to 2. With the increase of the mole ratio of Eu to Ti, both the conversion and selectivity for the epoxidation reaction of styrene were declined. In general, the ordered mesoporous silica catalysts modified with titanium had very high activity in oxidation of styrene while its selectivity was relatively low. Modification of the Ti-SBA-15 catalysts by more incorporation of europium increased significantly the activity in epoxidation of styrene, but reduced the conversion and selectivity in the excess of incorporation of europium. It was found that Eu-Ti(40)/NH 2 (1.43)-SBA-15(2) processed more higher titanium active sites in the catalyst, which could well be responsible for its inferior catalytic performance. The excessive addition of europium was therefore highly likely that the titanium sites in the Eu-Ti-NH 2 -SBA-15 would be covered by europium oxide cluster and unable to interact efficiently with oxidant and effectively generate the active oxidant species such as metal-oxo-complex with a high electron affinity (electrophilic) for the attack of electron-rich double bond of the styrene molecule [36]. On the other hand, the introduction of a trivalent cation of europium into a silicate framework would invoke a greater negative charge than Si 4+ ; hence it will have to be compensated by a greater number of protons, creating a stronger Brӧsted acid site. The results of synergistic enhancement were consistent with previous results [37]. It was therefore concluded that the simultaneous incorporation of Eu(III) and Ti(IV) ions resulted in a subtle modification of the local structural environment around the titanium centre, which were responsible for marked improvements in catalytic activity.

Effect of the Solvent
The nature of solver had an important effect on the outcome of the reaction, i.e., on yield, by-product formation and reaction kinetics [38]. The influence of various both aprotic and protic solvents such as acetonitrile (MeCN), acetone (MeCOMe), methanol (MeOH) and dimethyl formamide(DMF) was listed in Table 3 for the epoxidation of styrene employing Eu-Ti(40)/NH 2 (1.43)-SBA-15(2) as the catalyst. As shown in Table 3, the solvents played a significant role on the epoxidation of styrene. In addition, as for the conversion of styrene, the reactivity trend in different media was in the order DMF > MeCN > MeOH > MeCOMe. It was likely to be the result of decreasing electrophilicity and increasing steric constraints of species Ti-O-H in different reaction media. The results indicated that the conversion of styrene reaction of aprotic solvents such as DMF, MeCN and MeCOMe was higher than that of the protic solvents of MeOH. Though the polarity of DMF was weaker than that of MeCN, the conversion of styrene in DMF solvent was higher than that of MeCN solvent, which probably attributed to more deep reactions in DMF under photoirradiation [39]. Moreover the polarity of MeCN was stronger than that of acetone, so the conversion in MeCN was higher than that of acetone solvent [40]. Although DMF has a higher conversion, its selectivity to styrene oxide was too low. Such a behavior could be explained that MeCN could react with hydrogen peroxide, leading to the formation of peroxyimidic acid (CH 3 -C(= NH)-O-O-H), which was known to all that it's an active oxidant agent towards many organic substrates [41]. Simultaneously, due to the presence of MeCN, the acidity of reaction system could be reduced by its slight basicity and then the open-ring reaction of styrene oxide to benzaldehyde was decreased, and then the highest selectivity to styrene oxide could be obtained [42]. The role of MeCN would inhibit the coordination of the substrate and enhance desorption of the product from the active site. The affinity of MeCN to the metal ion might promote separation of styrene oxide from the site, which resulted in the prevention of deep oxidation of the oxide produced and in the increase in the selectivity [43].

Effect of the addition amount of 3wt% NaOH
It was well known that catalytic activity for epoxidation had a close relation with alkalinity on the surface of catalysts. It was of a vital importance for both formation of epoxide and inhibition of deep oxidation or ring-opening isomerization due to the fact that the distribution of acidic sites on surface of the catalyst was changed via NH 2 -functionalized and addition of europium [44]. Figure 9 showed the conversion and the selectivity for the epoxidation reaction over Eu-Ti (40)/NH 2 (1.43)-SBA-15(2) as a function of the amount of NaOH solution. As could be seen from Fig. 7 with the increase of amount of NaOH with concentration of 3wt% the conversion of styrene and selectivity of epoxide increased significantly and then approximately trended to be stable. On the one hand, surface hydroxyl groups were generated via some reactions between Ti-O/Eu-OA in the NH 2 -SBA-15 framework and water molecules in the catalyst surfaces. The reaction between generated surface hydroxyl groups and hydrogen peroxide could generate HOO − and then it could react with acetonitrile forming imine peroxide. The reactive intermediates transferred oxygen to styrene, generating amide and phenyl epoxyethane. On the other hand, after modification SBA-15 with organic alkali and inset of Ti and Eu was conducive to amount reduction of L acid centers, thus inhibiting phenyl epoxyethane ring-opening reaction and improving effectively reaction selectivity. The optimum addition amount of the sodium hydroxide solution with the concentration of 3wt% was found to be 0.45 g. The result was similar to previous reports [45]. Figure 10 showed the conversion of and the selectivity for the epoxidation reaction as a function of the amount of asprepared catalyst of Eu-Ti (40)/NH 2 (1.43)-SBA-15 (2). As could be seen from Fig. 10, with the increase of catalyst, both the conversion and selectivity reached to the maximum and then decreased subsequently as a whole. On the one hand, an increase in the amount of catalyst was responsible for the increase of active sites on catalyst surface, improving the conversion and selectivity of epoxidation reaction. On the other hand, when the use of catalyst was more than 50 mg, the conversion and selectivity of epoxidation reaction decreased due to the fact that there was a competition reaction between styrene epoxidation reaction and decomposition of hydrogen peroxide in catalytic system. The excessive catalyst with much more active center was more likely to lead to multiple-effect such as deep oxidation of styrene to phenylacetaldehyde and decompositon of hydrogen peroxide of oxidant as well as the decrease of active sites due to the effect of overlap of the catalyst in photo-absorption, reducing the conversion and selectivity of epoxidation reaction. The optimal amount of catalyst was therefore 50 mg.

Effect of the molar ratio of H 2 O 2 /Styrene
The influence of different mmolar ratio of H 2 O 2 to styrene on the epoxidation of styrene was presented in Fig. 11. In all the cases, styrene oxide was obtained as the major product along with small amounts of benzaldehyde, mandelic acid, benzoic acid and phenylacetaldehyde as the byproducts. It could be seen that the highest selectivity of styrene oxide and higher conversion was obtained at a H 2 O 2 to styrene mmol ratio of 3:1 over Eu-Ti (40)/NH 2 (1.43)-SBA-15(2) at ambient temperature for 3 h under UV irradiation. As the mmol ratio of H 2 O 2 was increased up to 5 with standard styrene of 1 mmol, the conversion of styrene oxide increased with decreasing the selectivity of the products because the selectivity of byproducts increased via deep oxidation. As the mmol ratio of styrene was increased, the conversion of styrene and selectivity of styrene oxide decreased. This might be due to the unsaturation of reactant ratios on the catalyst surface. Thus, the optimal mmol ratio of H 2 O 2 to styrene was 3:1 and the amount of H 2 O 2 with concentration of 30% was 1.6327 g.

Effect of the reaction time
The epoxidation of styrene was carried out with different reaction time (10 min-6 h) using reactant The conversion of styrene and selectivity of styrene oxide increased slightly within 3 h due to the increase of the catalytic activity on the surface of the catalysts. When the reaction time was increased from 3 to 5 h, the conversion of styrene was found to be a maximum value of 32%, however selectivity of styrene oxide reduced continuously. Such a behavior was likely due to both the adsorption of by-products on the catalyst surface, deep oxidation reaction and polymerization of styrene oxide [46]. Thus, the optimum conditions for obtaining maximum conversion of styrene and the highest selectivity of styrene oxide could be summarized as follows: catalyst, Eu-Ti (40)/NH 2 (1.43)-SBA-15(2); radiation reaction time, 3 h; reaction medium, Acetonitrile; reaction temperature, ambient temperature; styrene to H 2 O 2 mmol ratio, 3:1. It was found that the catalytic activity (TOF = 3.1 × 10 −4 s −1 ) over Eu-Ti (40)/NH 2 (1.43)-SBA-15(2) was lower than that over polymeric chiral salen Mn (III) complex (TOF = 2.87 × 10 −3 s −1 ) in liquid phase reaction [47], which was due to the slow diffusion of the reactants to the catalytic centers in the heterogeneous catalysts.

Conclusion
Amino-functionalized mesoporous SBA-15 materials and Eu and Ti-doped amino-functionalized SBA-15 catalysts were prepared by post-grafting method and impregnation method, respectively. The effective catalysts for the epoxidation of styrene with high conversion (32%) and selectivity (84%) could be achieved with the clean and environmentally benign oxidant of hydrogen peroxide in the presence of acetonitrile under UV irradiation at ambient temperature. The catalyst structure showed that the active Ti 4+ and Eu 3+ species had been incorporated and were dispersed highly in the framework of amino-functionalized SBA-15. The presence of aminopropyl group could reduce the acidity of catalysts by its slight basicity and the rare earth metal Eu could restrain open-ring reaction, which was benefited to the formation of styrene oxide. Ti was incorporated into the silica framework mainly in the tetrahedral isolated sites and their coordination did not change upon ordering mesostructure of catalysts. At high Ti content, the broadening of the main UV-vis band with a shoulder at about 230 nm could be assigned to a higher coordination of Ti probably due to water molecules adsorbed on the catalyst as well as to the formation of some Ti-O-Ti clustering in the framework. Merging the peak shift to a longer wavelength, the Ti-O-Eu bond could be formed. The materials synthesized here showed a very good selectivity for the epoxidation of styrene using H2O2 as oxidant under photo-catalytic condition. The conversion and selectivity were strongly influenced by the structure of the catalyst, the degree of metal loading and the chemical environment around the active sites as well as surface functional synergy between redox active sites and acid and alkali active sites. The designed strategy was also readily extended to apply to other immobilization systems. More wide and in-depth applications in heterogeneous coordination catalysis were anticipated.
Author contributions GW: contributed to the conception of the study, data analyses with constructive discussions, manuscript preparation and modification. HG: performed the experiments and contributed to data analyses and wrote manuscript. QC: contributed to experiment, data analyses and plotting in the manuscript.
Funding This work was financially supported by the National Natural Science Foundation of Anhui Province (No: 1208085MB32).

Declarations
Conflict of interest I would like to declare on behalf of my co-authors that the work described was original research that has not been published previously, and not under consideration for publication elsewhere, in whole or in part. All the authors listed have approved the manuscript that is enclosed. All authors declared that: (i) no support, financial or otherwise, has been received from any organization that may have an interest in the submitted work; and (ii) there are no other relationships or activities that could appear to have influenced the submitted work. To the best of our knowledge, the named authors have no conflict of interest, financial or otherwise. No conflict of interest exits in the submission of this manuscript, and manuscript is approved by all authors for publication. The original manuscript and figures will be transferred, following the instruction by the Editorial Committee when the paper is accepted. I hope your favorable consideration for publication to your journal.