Confinement of Pt nanoparticles in cage-type mesoporous silica SBA-16 as efficient catalysts for toluene oxidation: the effect of carboxylic groups on the mesopore surface

Hung-Chi Wua, Tse-Ching Chenb, Canggih Setya Budic, Pin-Hsuan Huanga, Ching-Shiun Chen*ab and Hsien-Ming Kao*c
aCenter for General Education, Chang Gung University, Taoyuan City 33302, Taiwan, Republic of China. E-mail: cschen@mail.cgu.edu.tw
bDepartment of Pathology, Chang Gung Memorial Hospital, Taoyuan City 33302, Taiwan, Republic of China
cDepartment of Chemistry, National Central University, Taoyuan City 32001, Taiwan, Republic of China. E-mail: hmkao@cc.ncu.edu.tw

Received 3rd September 2019 , Accepted 21st October 2019

First published on 22nd October 2019


In this work, 3D cage-type mesoporous silica SBA-16, which is functionalized either with or without –COOH groups, is used to support Pt nanoparticles (NPs) and then applied for toluene oxidation. The –COOH functionalized SBA-16 exhibits higher affinity towards Pt4+ species, and as a result highly dispersed and nanosized Pt nanoparticles are formed in comparison to the case of its counterpart pure silica SBA-16 without –COOH groups. When pure SBA-16 without –COOH is used as the support, on the other hand, Pt NPs are formed outside the mesopores of SBA-16, resulting in larger Pt NPs. The catalytic activity for toluene oxidation over Pt nanoparticles deposited on SBA-16 with –COOH is significantly higher than that on pure SBA-16. The intermediate species, reaction mechanisms and active sites in the course of toluene oxidation are probed by in situ IR spectroscopy of CO and toluene adsorption. The Pt nanoparticles confined in the cage-type mesopores of SBA-16 with –COOH can induce the breakage of the strong C–C bonding between phenyl and methyl groups to form CO and carbonyl intermediates, and thus enhance the catalytic activity for toluene oxidation. The defect sites of Pt particles on SBA-16 with –COOH may play a role in the adsorption and decomposition of toluene. The low reducibility of Pt on SBA-16 results in a poor ability to degrade toluene, thus leading to a low catalytic rate for toluene oxidation.


1. Introduction

Volatile organic compounds (VOCs) are important air pollutants that can lead to environmental and health problems. Therefore, several techniques for the complete elimination of VOCs have been investigated, including catalytic oxidation, condensation, adsorption, and membrane separation.1–16 Among these techniques, the catalytic oxidation of VOCs has been recognized as an economical process for VOC elimination, even at low concentrations (250–1000 ppm).17–19 VOC substances, such as benzene, toluene, and xylene, are assumed to be endocrine-disrupting chemicals, which may have high toxicity to humans.20 In general, aromatic derivatives are widely used as solvents to dissolve organic substances in leather tanning processes, adhesives, rubbers, and paints. To date, Pt-based catalysts are frequently used for the catalytic oxidation of aromatic derivatives in the environmental control of air pollution.20–29 The effects of particle size, kinetic control, promoters and pretreatment on toluene oxidation have been widely discussed in the literature.20–29 However, enhancing the efficiency of Pt-based catalysts for the complete oxidation of toluene is still an important, attractive issue due to the high cost of Pt metal.20–29

Several materials with high surface areas have been considered as supports for incorporation of noble metals in order to obtain better metal dispersion. In recent years, ordered mesoporous MCM-41 and SBA-15 with 2D-hexagonal channels and pore diameters in the range from 5 to 30 nm have been developed to support and confine nanosized metal NPs.30–37 Mesoporous silica SBA-15 is a promising material for dispersing metal particles due to its high surface area (600–1000 m2 g−1).38–45 The use of SBA-15 mesoporous materials may improve metal dispersion and result in formation of Pt nanoparticles (NPs).31 However, SBA-15 with typical cylindrical pores may strongly favor the growth of metal particles with long spheroidal or rod-like shapes.46 Thus, pore blockage from metal particles may occur in the channel structure, influencing the molecular diffusion along the channels. Considering the disadvantages of 2D channel-like structures for their applications in catalytic reactions, the 3D cage-type mesoporous silica SBA-16 can be proposed as an alternative support system for synthesizing metal NPs.47–49 The unique characteristics of SBA-16 with a high surface area, large pore volume and tailorable pore size make it a promising support for the preparation of metal NPs. The cage-type mesoporous silica SBA-16 can effectively control the size of Ni nanoparticles to a few nanometers, regardless of high Ni loadings.47–49 SBA-16-supported Ni NPs have significantly enhanced catalytic activity for hydrogenation of CO2 and nitrophenol compared to Ni deposited on SiO2 and SBA-15 silica-based supports.47–49 In this work, 3D, mesoporous, cage-type SBA-16 is further employed to support Pt NPs for toluene oxidation. Pt NPs embedded in SBA-16, which is functionalized with and without –COOH groups, are fabricated to study the effects of organic functionality in mesoporous silicas on the particle size, metal reducibility and confinement of Pt NPs. The results from a variety of kinetic tests for toluene oxidation and IR spectroscopy are compared for the Pt NPs on SBA-16 with and without –COOH, and the data can be correlated with the active sites and reaction mechanisms on these Pt-based catalysts.

2. Experimental

2.1 Preparation of SBA-16 functionalized without and with carboxylic acid groups

The synthesis of mesoporous silica SBA-16 without –COOH groups (denoted as S16) was based on the following procedure: first, 0.93 g Pluronic F127 (Sigma-Aldrich), 0.19 g Pluronic P123 (Sigma-Aldrich), and 4 g KCl were mixed and dissolved in 60 mL of 2 M HCl aqueous solution and vigorously stirred at 631 K for 4 h. Tetraethyl orthosilicate (TEOS) was added into the solution and stirred for 20 h, and then the reaction mixture was hydrothermally treated at 373 K for 24 h. The molar ratio of the mixture for the synthesis of SBA-16 was 1[thin space (1/6-em)]:[thin space (1/6-em)]0.0016[thin space (1/6-em)]:[thin space (1/6-em)]0.0037[thin space (1/6-em)]:[thin space (1/6-em)]2.7[thin space (1/6-em)]:[thin space (1/6-em)]4.4[thin space (1/6-em)]:[thin space (1/6-em)]144 (TEOS/P123/F127/KCl/HCl/H2O). The precipitate was washed with deionized water and dried at 70 °C. The template removal was performed by calcination in air at 823 K for 6 h.

SBA-16 functionalized with –COOH groups (denoted as S16C) was synthesized by following our previous method.47–49 A premixed solution of carboxyethylsilanetriol sodium salt (CES, Gelest, 25% in water) and TEOS (Sigma-Aldrich) was added into a mixture of 0.93 g of Pluronic F127, 0.19 g of Pluronic P123, 4 g of KCl and 2 M aqueous HCl (60 mL). The content of CES with a CES/(TEOS + CES) mole ratio was 20%. The solution was vigorously stirred at 308 K for 4 h. The molar ratio of the reaction mixture for the synthesis of S16C was 0.2[thin space (1/6-em)]:[thin space (1/6-em)]0.8[thin space (1/6-em)]:[thin space (1/6-em)]0.0016[thin space (1/6-em)]:[thin space (1/6-em)]0.0037[thin space (1/6-em)]:[thin space (1/6-em)]2.7[thin space (1/6-em)]:[thin space (1/6-em)]4.4[thin space (1/6-em)]:[thin space (1/6-em)]144 for CES[thin space (1/6-em)]:[thin space (1/6-em)]TEOS[thin space (1/6-em)]:[thin space (1/6-em)]P123[thin space (1/6-em)]:[thin space (1/6-em)]F127[thin space (1/6-em)]:[thin space (1/6-em)]KCl[thin space (1/6-em)]:[thin space (1/6-em)]HCl[thin space (1/6-em)]:[thin space (1/6-em)]H2O. The resulting mixture was subjected to hydrothermal treatment at 373 K for 24 h, dried in air at 343 K after filtration and washed with deionized water. The Pluronic F127 and P123 templates were removed by acid extraction. Then, 0.35 g of the as-synthesized sample was dispersed in 125 mL of a 48 wt% H2SO4 solution with vigorous stirring at 368 K for 24 h. The final white solid product was filtered, washed with deionized water and air-dried at 343 K for 12 h and named S16C.

2.2 Preparation of Pt(x)@S16C and Pt(x)@S16 catalysts

The Pt(x)S16C and Pt(x)S16 catalysts were prepared by impregnating 0.1 g of the support with 10 mL of 1.3 × 10−3 M and 2.9 × 10−3 M aqueous solutions of H2PtCl6 for 0.5 wt% (x = 0.5) and 1.0 wt% (x = 1.0) loading, respectively. The as-impregnated Pt(x)@S16C and Pt(x)@S16 samples were pretreated using an air stream for calcination and H2 for reduction at 773 K for 5 h before being used in the reactions. The Pt concentration (wt%) was determined by inductively coupled plasma-atomic emission spectroscopy (ICP-AES).

2.3 Catalytic tests for toluene oxidation

All toluene oxidation reactions were carried out in a fixed-bed reactor (0.95 cm inner diameter) at atmospheric pressure. The reactant mixture of O2[thin space (1/6-em)]:[thin space (1/6-em)]toluene with a 22.5[thin space (1/6-em)]:[thin space (1/6-em)]1 molar ratio was obtained using a 30 mL min−1 O2 stream passing through liquid toluene at 273 K. The gaseous O2 and toluene were passed over 20 mg sample containing 1 mg catalyst and 19 mg pure support. All products were analyzed using a gas chromatograph (GC) equipped with a thermal conductivity detector (TCD). The O2, CO2, toluene and H2O molecules in the effluent stream were separated in a 12 ft Porapak-Q column. A four-way valve with an external sample loop (1 mL) was employed to sample the reactants and/or products into the GC-TCD system. The concentration of toluene in the inlet and outlet streams was quantified by injecting different volumes of liquid toluene (0.05–0.5 μL). The CO2 concentration during the oxidation process was quantified by sampling 0.1–1 mL volume of CO2. The conversion for the catalytic toluene oxidation was determined by the following equation: X% = (CinCout)/Cin, where Cin and Cout are the inlet and outlet concentrations of toluene, respectively. The conversion for the reaction was controlled below 10% to ensure that the reactions were maintained under different conditions. The turnover frequency (TOF) was calculated using the following formula: TOF = [conversion × 0.003 (mL s−1 for toluene) × 6.02 × 1023 (molecules per mol)]/[24[thin space (1/6-em)]400 (mL mol−1) × number of Pt sites]. The carbon balance for all reactions remained at approximately 95–100%.

2.4 Fourier transform infrared (FTIR) spectroscopy measurements

A Nicolet 5700 FTIR spectrometer equipped with a mercury cadmium telluride (MCT) detector was employed to perform in situ diffuse reflectance infrared Fourier transform (DRIFT) spectroscopy. All spectra were collected with a 1 cm−1 resolution for 128 scans. All experiments were performed in a DRIFT cell (Harrick) with ZnSe windows that could be heated to 773 K. A stream of CO was passed through the catalysts for 30 min at room temperature to obtain saturated adsorption. The residual gaseous CO was purged for 60 min using a He stream. Toluene adsorption onto all catalysts was performed by injection with a 5 μL liquid through a port located upstream of the DRIFT cell. The injection port was heated to 373 K to prevent the condensation of toluene.

2.5 Measurements of the Pt surface area

Saturated CO chemisorption on the Pt(x)@S16C and Pt(x)@S16 catalysts was employed to measure the Pt surface area. All measurements were performed in a glass vacuum system. First, 0.2 g of catalyst was pressed at 260 atm to obtain a sample disk, which was repeatedly reduced in 760 Torr H2 at 773 K for 2 h and evacuated at 773 K and 4 × 10−5 Torr in a glass vacuum system for 30 min. Then, 20 Torr CO was introduced to the catalyst at room temperature for 10 min to obtain uptake saturation. The surface area of the Pt catalyst was calculated assuming a CO[thin space (1/6-em)]:[thin space (1/6-em)]Pt stoichiometric ratio of 1. The average surface density of the Pt metal was 7.1 × 1018 Pt atoms per m2.

3. Results and discussion

3.1 Structural characterization of Pt(x)@S16C and Pt(x)@S16

N2 adsorption–desorption isotherms and powder XRD diffraction patterns were used to characterize the structures of S16C and S16 with and without Pt NPs. Fig. 1 shows the small-angle XRD patterns, in which three resolved peaks can be observed in the region of 2θ = 0.5–2.0°. The peaks can be assigned to the (110), (200) and (211) planes of cubic SBA-16 with cage-type mesopores (Im3m symmetry). The diffraction peak of the (110) plane at 0.65° for S16C slightly shifted to a higher angle of 0.72° for S16, indicating that the unit cell parameter of S16C is slightly larger than that of S16. When Pt NPs were deposited onto both supports, distinct changes in the XRD patterns were observed. The deposition of Pt NPs on the S16 support did not seem to lead to changes in the diffraction angles or intensities. On the other hand, the incorporation of Pt into the structure of S16C led to a decrease in the intensity of the diffraction peak, a broader d110 peak and a shift in the peak toward a higher diffraction angle. The results indicated that the deposition of Pt NPs on S16C might cause a slight contraction in the cage-type structure of S16C.
image file: c9cy01787a-f1.tif
Fig. 1 Small-angle XRD patterns of the S16, S16C, Pt(x)@S16 and Pt(x)@S16C samples, where x = 0.5 and 1.0 wt%.

The N2 adsorption–desorption isotherms and pore size distributions of S16C and Pt(x)@S16C are shown in Fig. 2, and type IV isotherms with H2 hysteresis loops could be observed for all samples. The textural properties of these samples, such as the surface area, pore volume and cage size, are listed in Table 1. During the synthesis of S16C, the template removal was conducted under acid treatment using 48 wt% H2SO4 solution at 368 K for 24 h. The small angle XRD pattern of S16C (Fig. 1) exhibits a sharp diffraction peak with a higher intensity as compared to that of the as-synthesized S16C (without template removal) (Fig. S1A, ESI), which could be indicative of the increase in the structural ordering after template removal. Consistently, the high BET surface area and pore volume of S16C of about 575 m2 g−1 and 0.71 cm3 g−1 could be achieved (Table 1), respectively, after template removal. The TEM image of S16C also reveals the presence of distinctive pores of ordered mesoporous silica S16C after template removal (Fig. S1B, ESI). Therefore, despite the template removal, there was no significant effect of such an acid treatment on the deterioration in the mesostructure of S16C. Furthermore, it has been widely accepted that the acid treatment using 48 wt% H2SO4 solution at 368 K for 24 h is an effective technique to remove the surfactant template during the preparation of surface functionalized mesoporous silica materials.


image file: c9cy01787a-f2.tif
Fig. 2 (A) N2 adsorption–desorption isotherms and (B) cage pore size distribution curves of S16C and Pt(x)@S16C samples, where x = 0.5 and 1.0 wt%.
Table 1 Textural properties of S16C, S16, Pt(x)@S16C and Pt(x)@S16
Sample ABET (m2 g−1)a Vp (cm3 g−1)b Cage sizec (nm) Entrance sized (nm)
a ABET is the surface area.b Vp is the pore volume.c Cage pore sizes were obtained from the nitrogen adsorption isotherms.d Entrance pore sizes were obtained from the nitrogen desorption isotherms.
S16 485 0.47 6.5 3.8
Pt(0.5)@S16 475 0.46 6.4 3.8
Pt(1.0)@S16 475 0.45 6.5 3.8
S16C 575 0.71 7.6 3.9
Pt(0.5)@S16C 528 0.68 7.3 3.8
Pt(1.0)@S16 516 0.63 7.2 3.8


As Pt NPs were loaded onto S16C, the width of the hysteresis loops apparently became narrower. When Pt NPs were deposited onto the S16C support, the surface area, total pore volume and cage size of S16C significantly decreased with increasing Pt loading. The same measurements for S16 and Pt(x)@S16 are presented in Fig. 3, showing that the formation of Pt NPs on the S16 support did not significantly influence the characteristic hysteresis shape of the S16 structure and its pore distribution. S16 without –COOH apparently exhibited a lower surface area and smaller cage size than S16C. The Pt NPs formed on the S16 support could cause a slight decrease in the accessible surface area and pore volume. The results indicated that the Pt NPs did not form inside the cage pores of S16. Based on the results described above, it can be reasonably deduced that the –COOH group of S16C might provide a meaningful benefit by attracting Pt4+ species to form particles that are incorporated into the cage pores. In contrast, the SiOH groups on the mesopore surface of S16 might have a poor ability to bind highly positive Pt4+ species, and thus the Pt NPs formed outside the mesopores of S16 rather than in the cage-type mesopores.


image file: c9cy01787a-f3.tif
Fig. 3 (A) N2 adsorption–desorption isotherms and (B) cage pore size distribution curves of S16 and Pt(x)@S16 samples, where x = 0.5 and 1.0 wt%.

The wide-angle XRD patterns of Pt NPs deposited on S16C and S16 are shown in Fig. 4, exhibiting diffraction peaks at 2θ values of 39.6° and 46.0° for the (111) and (200) characteristic facets, respectively. The full-width at half-maximum (FWHM) value of the (111) peak was estimated to calculate the average size of the Pt NPs using the Scherrer equation, as listed in Table 2. The particle size of Pt NPs on the S16C support was apparently smaller than that on the S16 support. The surface area and dispersion of Pt NPs for all the Pt(x)@S16C and Pt(x)@S16 catalysts were measured using saturated CO chemisorption at room temperature. The results revealed in Table 2 show that the small Pt NPs on S16C might have a high surface area and dispersion of Pt compared to those of the Pt(x)@S16 catalysts.


image file: c9cy01787a-f4.tif
Fig. 4 XRD spectra of the Pt(x)@S16C and Pt(x)S16 samples, where x = 0.5 and 1.0 wt%.
Table 2 Characterization of the Pt(x)@S16C and Pt(x)@S16 catalysts
Catalyst Pt loadinga (wt%) Pt surface areab (m2 gcat.−1) Dispersionb (%) Pt particle sizec (nm) Pt particle sized (nm)
a Estimated from ICP-AES.b Estimated from CO chemisorption.c Average particle size of Pt NPs estimated from XRD patterns.d Average particle size of Pt NPs estimated from TEM images.
Pt(0.5)@S16 0.47 0.14 5.8 12.4 9.5 ± 2.0
Pt(0.5)@S16C 0.56 0.26 13.0 4.3 6.1 ± 1.5
Pt(1.0)@S16 0.94 0.25 6.0 23.9 25 ± 16.3
Pt(1.0)@S16C 1.20 1.2 23.6 6.5 6.4 ± 1.7


3.2 Transmission electron microscopy (TEM)

Fig. 5 compares the TEM images of all Pt(x)@S16 and Pt(x)@S16C samples and their particle-size distributions as a function of the Pt loading. For the Pt(0.5)@S16 sample, the Pt NPs strongly aggregated on the surface of S16. As the Pt loading increased to 1.0 wt%, the TEM image of the Pt(1.0)@S16 sample shows extremely large particle aggregates and inhomogeneous particle distribution. When Pt NPs were formed on the S16C support, on the other hand, small and uniform Pt particles could be observed for the Pt(0.5)@S16C and Pt(1.0)@S16C samples. The average sizes of the Pt NPs for all Pt(x)@S16 and Pt(x)@S16C samples were estimated from the particle size statistics and are listed in Table 2. The size of the Pt NPs deposited on S16C could be controlled at ∼6 nm, regardless of Pt loading.
image file: c9cy01787a-f5.tif
Fig. 5 TEM images and particle size distributions of the Pt(x)@S16C and Pt(x)S16 samples, where x = 0.5 and 1.0 wt%.

3.3 H2 temperature-programmed reduction

Fig. S1 reveals the H2-TPR profiles of the Pt(x)@S16C and Pt(x)@S16 catalysts for air calcination. For the Pt(x)@S16 catalysts, two main reduction bands were observed in the range of 300–973 K. In general, the reduction of Pt oxide should occur at ∼373–473 K.50 The low-temperature band at 300–500 K may be associated with the reduction of Pt oxides. However, reports in the literature have indicated that a H2PtCl6 precursor on oxide supports calcined in air can form oxychlorinated platinum (PtOxCly) species, which may be decomposed and reduced in H2 at 673 K.51–53 Thus, the signal at 600–970 K for the H2-TPR profiles of the Pt(x)@S16 samples was attributed to the reduction of PtOxCly species. In contrast, the reduction of the Pt(x)@S16C samples appeared in the low-temperature region (300–500 K). No observable reduction of PtOxCly species at high temperature could be detected for the Pt(x)@S16C samples. In this study, all the Pt catalysts were reduced in H2 at 673 K before use. Thus, the Pt NPs on S16 should be difficult to completely reduce due to the presence of PtOxCly species. In contrast, highly reduced Pt NPs formed on the S16C support. The large difference in the reducibility of Pt NPs on both S16C and S16 might depend on the interaction between Pt4+ species and the functional groups of the supports. The structural characterization of Pt(x)@S16C clearly indicated that Pt NPs could be incorporated into the cage pores of the S16C support. The –COOH groups were reasonably deduced to be located inside the cage structure of S16C and might effectively attract Pt4+ species to form metallic Pt during calcination and reduction treatments. The strong interaction between the –COOH group and Pt4+ species might inhibit the formation of PtOxCly species. For the Pt(x)@S16 samples, Pt4+ species were not successfully impregnated into the cage structure and were likely dispersed on the S16 surface because the –OH groups of the S16 support provided weak interactions with the Pt4+ species. Therefore, PtOxCly species were likely present on Pt(x)@S16, which might cause the low reducibility of Pt NPs.

3.4 Interaction between Pt NPs and the S16C support

In this work, a wet impregnation procedure was performed by dispersing a certain amount of S16C and S16 supports in an H2PtCl6 aqueous solution under an acidic environment. The pH values of the H2PtCl6 aqueous solutions with concentrations of 1.3 × 10−3 M and 2.9 × 10−3 M (both have 10 mL) were measured to be around 2.4 and 2.3, respectively. If the H2PtCl6 salt was completely dissociated, then the theoretical pH values for both concentrations should be 2.6 and 2.2, respectively. Therefore, the measured pH values matched very well with the theoretical values, indicating that H2PtCl6 was highly soluble in water and dissociated following H2PtCl6 → 2H+ + Pt4+ + 6Cl. Meanwhile, at pH = 2–3, the surface charge of S16C is negative as confirmed by the zeta potential measurements (Fig. S3, ESI). The negative value of the surface charge of S16C might be ascribed to the partial dissociation of carboxylic acid (–COOH → –COO). On the other hand, the –SiO/–SiOH ratio might also be increased when the pH is higher than the isoelectric point of silica (IEP, pH = 2).54 The co-existence of the –COOH organic pendant along with –SiO in the S16C support might induce some possible electrostatic interactions with the highly positively charged Pt4+ species, which not only governed the effective adsorption of Pt4+, but might also control the growth of Pt NPs during thermal reduction. As a result, higher dispersion of Pt NPs is found on the S16C support than on S16 (without –COOH functionality), as observed from the TEM images shown in Fig. 5.

Fig. S4 shows the thermogravimetric analysis (TGA) and differential thermal analysis (DTA) curves of 1.0 wt% Pt4+ impregnated on S16C and S16C supports in an air stream. The weight loss at 340 K for both samples was attributed to the desorption of physisorbed water. The weight loss at 710 K for the as-impregnated Pt(1.0)@S16C might be associated with the decomposition of the carboxylic acid groups. The results presented that the residual carboxylic acid groups might still remain on the Pt(x)@S16C samples after thermal treatments in this study. In reality, the presence of –COOH as a capping agent on the silica support may control the growth of Pt NPs.46

In our previous work,48 Ni/S16C was prepared through a wet impregnation technique in which the S16C support was dispersed in an aqueous solution containing a Ni(NO3)2·6H2O precursor. Then the pH of this mixture was adjusted to 9 by dropwise addition of 0.1 M NaOH aqueous solution. Under such alkaline conditions, the deprotonation of the carboxylic acid groups resulted in the formation of –COO groups, leading to a more negative surface charge of S16C (Fig. S3, ESI). The more negative surface charge of S16C was responsible for allowing more effective adsorption of Ni2+ and its other forms (e.g., Ni(OH2)) in the cage pores of S16C. The similar light green color of the Ni(OH)2 precipitate and the Ni2+ ions adsorbed in the S16C support made the observation, whether there is simultaneous Ni(OH)2 precipitation occurring during the synthesis of Ni/S16C, more difficult. However, it should be noted that under alkaline conditions the hydroxyl species were consumed to deprotonate both carboxylic acid and silanol groups on the surface of S16C, thus generating negative charge density on the mesopore surface of S16C. It was found that increasing the solution pH from 1 to 9 gradually decreased the surface potential of S16C to a more negative value, and thus facilitated the electrostatic interaction as a driving force to adsorb Ni2+ species effectively. According to the acid–base titration, the equivalence point of the S16C powders dispersed in 0.1 M NaOH aqueous solution was obtained at pH = 7.3.48 Therefore, the optimum adsorption of Ni ions can be achieved at pH = 9. It can be deduced that there was no serious interference from the co-precipitation of Ni(OH)2 since the Ni2+ ions were quickly adsorbed on the negatively charged S16C support at pH = 9. In fact, no XRD signal of the Ni(OH)2 precipitate was observed for the Ni/S16C samples.

In the case of Pt impregnation, the H2PtCl6 precursor could be completely dissolved in water under acidic conditions. The highly positive Pt4+ species were adsorbed on the negatively charged surface of S16C. The more negative value of the surface potential of the S16C support led to more effective adsorption than that of S16 even at a low pH. The –COO groups could possibly interact with Pt4+ via an ion-exchange process (–COOH + Pt4+ → 4H+ + Pt4+(–COO)4). Therefore, high dispersion of Pt NPs with controllable particle sizes could be achieved when S16C was used as the support as compared to S16 (without –COOH functionality).

3.5 IR spectroscopy of CO adsorbed on the Pt(x)@S16C and Pt(x)@S16 catalysts

IR spectroscopy of CO adsorption was used to probe the active sites and oxidation state of Pt NPs through vibrational stretching. Fig. 6 shows the IR spectra of CO adsorbed on all Pt(x)@S16C and Pt(x)@S16 catalysts at room temperature. As shown in Fig. 6, symmetrical IR bands were observed for the Pt(x)@S16C and Pt(x)@S16 catalysts. However, the vibrational frequency of CO adsorption on the Pt(x)@S16 catalysts at ∼2092 cm−1 was slightly higher than that on the Pt(x)@S16C catalysts at ∼2073 cm−1. In general, the stretching frequency of CO adsorption on Pt surfaces is sensitive to surface sites or oxidation states. It has been reported in the literature that the ν(CO) of CO adsorbed on Pt surfaces shifts to lower wavenumbers as the Pt particle size decreases.39 The frequency peak at ∼2054–2075 cm−1 is usually assigned to CO binding at defect sites, such as edges and kinks.55–58 On the other hand, the peak centered at 2096 cm−1 can be associated with CO adsorbed on the terrace sites of close-packed surfaces or slightly oxidized Ptδ+ species.55–59
image file: c9cy01787a-f6.tif
Fig. 6 IR spectra of CO adsorbed onto reduced Pt(x)@S16C and Pt(x)S16 catalysts, where x = 0.5 and 1.0 wt%. CO adsorption was performed via exposure to a 20 mL min−1 pure CO stream at a constant pressure of PCO = 101.3 kPa for 30 min, followed by a 20 mL min−1 helium stream for 50 min to purge the CO gas. The adsorption temperature was room temperature.

In this work, the Pt NPs deposited on the S16C support apparently exhibited smaller particle sizes than the Pt(x)@S16 samples; therefore, abundant defect sites might be present on the Pt surface, inducing the ν(CO) of CO adsorption positioned at ∼2073 cm−1. For the Pt(x)@S16 catalysts, the large Pt particles have a high possibility of forming close-packed surfaces. On the other hand, the presence of PtOxCly species on the Pt(x)@S16 catalysts could also lead to Pt NPs with a slightly positive charge. These factors could influence the IR peak for CO adsorbed on Pt(x)@S16 at ∼2092 cm−1. Notably, the IR spectra of CO adsorbed on the Pt(x)@S16 samples had a lower intensity than those of the Pt(x)@S16C samples, according to the CO chemisorption results in Table 2.

3.6 Catalytic tests for toluene oxidation on the Pt(x)@S16C and Pt(x)@S16 catalysts

The total oxidation of toluene was employed to examine the catalytic activity of the Pt(x)@S16C and Pt(x)@S16 catalysts. Fig. 7 shows the temperature-dependent TOF rates and conversion for toluene oxidation on these Pt(x)@S16C and Pt(x)@S16 catalysts. CO2 and H2O were the products during toluene oxidation. The TOFs and conversion for toluene oxidation on the Pt(x)@S16C and Pt(x)@S16 catalysts followed the order Pt(1.0)@S16C > Pt(0.5)@S16C > Pt(0.5)@S16 ≈ Pt(1.0)@S16. The S16C support effectively induced a high catalytic activity for toluene oxidation due to the deposition of Pt NPs.
image file: c9cy01787a-f7.tif
Fig. 7 Comparison of the reaction rates for toluene oxidation on the Pt(x)@S16C and Pt(x)S16 catalysts as a function of temperature, where x = 0.5 and 1.0 wt%.

In this case, the particle size of Pt(0.5)@S16C was slightly smaller than that of Pt(1.0)@S16C estimated from the XRD results, but the statistics from TEM images could provide similar particle sizes for both Pt(x)S16C catalysts. It was presented that the –COOH and –OH groups of S16C might give the benefit of controlling the Pt particle size. However, it was noteworthy that the Pt(1.0)@S16C catalyst exhibited higher dispersion than the Pt(0.5)@S16C catalyst as seen in Table 2, thus it could likely lead to high catalytic activity for toluene oxidation. The –COOH and –OH groups of S16C could have sufficient capacity to adsorb Pt4+ species, and retain the high dispersion.

Fig. 8 shows the Arrhenius plots for the TOFs of the toluene oxidation reaction over all the Pt(x)@S16C and Pt(x)@S16 catalysts. The apparent activation energies were calculated from the slopes in the temperature range of 423–460 K, resulting in values of 26.2 kJ mol−1 for Pt(0.5)@S16C, 99.2 kJ mol−1 for Pt(0.5)@S16, 25.0 kJ mol−1 for Pt(1.0)@S16C and 119.5 kJ mol−1 for Pt(1.0)@S16. It is noteworthy that the toluene oxidation achieved on the Pt(x)@S16C catalysts exhibited a lower apparent activation energy compared to that on the Pt(x)@S16 catalysts. The TOF rates of toluene oxidation over all the Pt(x)@S16C and Pt(x)@S16 catalysts as a function of time at 473 K are revealed in Fig. 9. The TOF rate on Pt(1.0)@S16C could significantly increase before 25 h and then appeared to be nearly stable until 72 h. The Pt(1.0)@S16, Pt(0.5)@S16 and Pt(0.5)@S16C catalysts exhibited stable TOF rates in the period of 72 h. No deactivation could be observed for each catalyst, implying that the Pt NPs deposited on S16C and S16 provided high catalytic stability for the toluene oxidation reaction.


image file: c9cy01787a-f8.tif
Fig. 8 Arrhenius plots for toluene oxidation on the Pt(x)@S16C and Pt(x)S16 catalysts, where x = 0.5 and 1.0 wt%.

image file: c9cy01787a-f9.tif
Fig. 9 Time-dependent TOF rates of toluene oxidation on the Pt(x)@S16C and Pt(x)S16 catalysts at 473 K, where x = 0.5 and 1.0 wt%.

3.7 IR spectroscopy of toluene oxidation on the Pt(1.0)@S16C and Pt(1.0)@S16 catalysts

In this work, the high level catalytic toluene oxidation over the Pt(x)@S16C catalysts can be ascribed to the lower apparent activation energies compared to those of the Pt(x)@S16 catalysts. The large difference in activation energy for the Pt(x)@S16C and Pt(x)@S16 catalysts might be correlated to the reaction mechanism during toluene oxidation. Thus, IR spectroscopy was further used to investigate the reaction mechanism through transient measurements of toluene oxidation. Fig. 10 reveals the in situ IR spectra of transient toluene (C6H5–CH3) and isotopic toluene (C6H513CH3) oxidation achieved by injecting 5 μL of liquid onto the Pt(1.0)@S16 catalyst at 433 K in an air stream. The signals of adsorbed CO2 were clearly observed, and an oxidation reaction was generated at 4 min after injecting liquid toluene. The characteristic bands for molecular toluene, i.e., the ν(C–H) signals at 2804–3107 cm−1, and the ring vibration bands of aromatics at 1503 and 1601 cm−1 were also recorded in the toluene oxidation process. As isotopic toluene was injected onto the Pt(1.0)@S16 catalyst, new isotopic 13CO2 peaks were observed. However, a large difference in the IR spectra was observed when the same experiments were performed on the Pt(1.0)@S16C catalyst, as shown in Fig. 11. Aside from CO2 formation, the intermediates including the adsorbed CO peak at ∼2080 cm−1 and possible C[double bond, length as m-dash]O stretching of carbonyl groups at 1776 cm−1 were clearly observed as toluene was injected onto the Pt(1.0)@S16C catalyst. Notably, no detectable ν(C[double bond, length as m-dash]C) bands for the phenyl group of toluene could be observed in this figure, even if the ν(C–H) signals remained. The results indicated that the dissociation of toluene and the strong oxidation reaction might simultaneously occur on the Pt(1.0)@S16C catalyst. The adsorbed CO and carbonyl species were assumed to be the major intermediates for further oxidation to CO2. The weak intensity of ν(C–H) might result from residual fragments after toluene decomposition rather than molecular toluene. When isotopic toluene was injected onto the Pt(1.0)@S16C catalyst, the redshift of the CO peak appeared at ∼2067 cm−1 and at 1742 cm−1 for the ν(C[double bond, length as m-dash]O) groups of carbonyl species, accompanying the formation of isotopic 13CO2 peaks. Thus, it was deduced that the formation of CO and carbonyl intermediates should result from the methyl group of toluene molecules. Undoubtedly, the toluene oxidation reaction follows different reaction pathways on both Pt(1.0)@S16C and Pt(1.0)@S16 catalysts. The strong breaking of C–C bonds between phenyl and methyl groups might occur on the Pt(1.0)@S16C catalyst. However, the oxidation of methyl fragments from toluene dissociation could benefit the formation of CO2, based on the formation of CO and carbonyl species. On the Pt(1.0)@S16 catalyst, molecular toluene might be directly oxidized to form CO2. The poor ability for toluene decomposition on the Pt(1.0)@S16 catalyst is associated with the low catalytic oxidation rate.
image file: c9cy01787a-f10.tif
Fig. 10 IR spectra for 5 μL injection of toluene (C6H5–CH3) and isotopic toluene (C6H513CH3) onto the Pt(1.0)S16 catalyst under an air stream at 433 K.

image file: c9cy01787a-f11.tif
Fig. 11 IR spectra for 5 μL injection of toluene (C6H5–CH3) and isotopic toluene (C6H513CH3) onto the Pt(1.0)S16C catalyst under an air stream at 433 K.

The temperature-dependent IR spectra of isotopic toluene adsorbed on Pt(1.0)@S16C under an air stream are shown in Fig. 12 to investigate the relationship between the intermediates and temperature. The characteristic bands for molecular toluene of ν(C–H) and ν(C[double bond, length as m-dash]C) of the aromatic ring could be observed at room temperature. The initial 13CO2 and 13CO molecules were present at 373 K, and then, the carbonyl intermediates were observed as the temperature increased to 423 K. The ν(C[double bond, length as m-dash]C) bands of the phenyl group exhibited weak intensities at 373 K, but the strong signals for ν(C–H) remained. The ν(C–H) peaks might depend on the fragments during toluene dissociation. The results implied that the adsorbed toluene might undergo an initial dissociation at ∼373 K to form the CO intermediate. As the temperature increased to 423 K, the intensities of the ν(C–H) bands significantly decreased, while obvious CO and carbonyl intermediates could be found, accompanying CO2 formation. The results indicated that the dissociated fragments from toluene on Pt(1.0)@S16C might be rapidly converted to CO and carbonyl species at 423 K and then further oxidized to CO2.


image file: c9cy01787a-f12.tif
Fig. 12 Temperature-dependent IR spectra of isotopic toluene (C6H513CH3) on the Pt(1.0)S16C catalyst under an air stream. The adsorption of isotopic toluene was performed by injecting 5 μL of liquid at room temperature.

In an earlier study, the process of dehydrogenating toluene on a Pt(111) surface was considered an essential step in the catalytic oxidation process.60 The breaking of aliphatic C–H bonding is assumed to occur more rapidly than the aromatic C–H bond breaking on the Pt surface.21 The dissociation of toluene may initially result from the breaking of the C–H bond of the methyl group. In this work, the high catalytic activity for toluene oxidation on Pt(x)@S16C was associated with the breaking of the C–C bond between the phenyl and methyl groups of toluene on the Pt surface. However, the high reduction of Pt NPs is an important factor influencing the decomposition of toluene. The low reducibility of Pt NPs on S16 might result in a weak ability to break the C–C bonds between the phenyl and methyl groups, which causes a poor catalytic rate.

3.8 Active sites for toluene oxidation on Pt(1.0)@S16C and Pt(1.0)@S16 catalysts

Fig. S5 compares the IR spectra of toluene and CO co-adsorption on both Pt(1.0)@S16C and Pt(1.0)@S16 catalysts. The IR spectra of toluene adsorbed on Pt(1.0)@S16C and Pt(1.0)@S16 catalysts exhibited the characteristic bands for adsorbed toluene. When CO was adsorbed on the toluene-precovered Pt(1.0)@S16C and Pt(1.0)@S16 catalysts, the characteristic peaks of toluene on both catalysts became weaker after CO adsorption. The findings implied that CO and toluene might competitively adsorb at the same adsorption sites on the Pt(1.0)@S16C and Pt(1.0)@S16 catalysts. The results could indicate that the decomposition of adsorbed toluene might occur at the abundant defect sites of the Pt(1.0)@S16C catalyst. On the other hand, toluene adsorbed on terrace sites of close-packed surfaces or slightly oxidized Ptδ+ sites of the Pt(1.0)@S16 catalyst might be difficult to dissociate. The large difference in the site population on the Pt(x)@S16C and Pt(x)@S16 catalysts may strongly affect the catalytic efficiency of toluene oxidation.

3.9 Comparison with Pt NPs on SBA-15

In this work, catalysts of 0.5 wt% Pt deposited on SBA-15 (Pt(0.5)@S15) and SBA-15 with –COOH (Pt(0.5)@S15C) were also used and their catalytic activity for toluene oxidation was compared to that of the Pt(0.5)@S16C catalyst. The TOF rates of toluene oxidation as a function of temperature over the Pt(0.5)@S16C, Pt(0.5)@S15C and Pt(0.5)@S15 catalysts are shown in Fig. S6. As shown in Fig. S6, the Pt(0.5)@S16C catalyst could provide significantly better intrinsic TOF rates than Pt(0.5)@S15 and Pt(0.5)@S15C. In our previous studies, rod-like and spherical Pt particles could be formed inside the channels of S15 and S15C, respectively.46

However, the structural effect of SBA-16 and SBA-15 may strongly influence the catalytic efficiency of these Pt NPs. For low Pt loading deposited on SBA-15, the TOF rate for toluene oxidation may be enhanced because the sub-nanosized Pt particles (<1 nm) confined in the micropores of SBA-15 may dominate the catalytic activity.39 Increasing the Pt loading may cause the particles to aggregate inside the 2D hexagonal channel-like structure of SBA-15, which is more sensitive to the pore blockage issue. The effect of pore blockage may restrict the reactants approaching the Pt active sites. The S16C support had a large cage size, which can attract more Pt atoms into the pore structure and result in effective diffusion for the reactants in the cage space. It seems that the –COOH groups on SBA-15 have no effect as both the data of Pt(0.5)@S15 and Pt(0.5)@S15C show similar values. In general, the S15 and S15C could provide abundant mesopore and micropore structures in the long channels, thus leading to a large pore volume and high surface area, compared to the S16 and S16C supports.46 It might be highly possible that Pt4+ ions could be incorporated into the channel structure of S15 and S15C, regardless of the presence of –COOH groups. Therefore, 0.5% Pt deposited on S15 and S15C supports might cause a similar effect of pore blockage in the channel structure.

4. Conclusion

In this work, 3D cage-type mesoporous silicas SBA-16 functionalized with and without –COOH groups were used as the supports to confine Pt nanoparticles. The presence of the –COOH group on the surface of the cage-type mesopores of SBA-16 (i.e., S16C) played a key role in the efficient incorporation of Pt4+ species into the mesopores. The S16C support can induce the formation of uniform Pt nanoparticles with a particle size of ∼6 nm, regardless of Pt loading. Without the help of –COOH groups, pure silica SBA-16 (S16) cannot attract highly positive Pt4+ species effectively. As a result, the Pt NPs are formed outside the mesopores of S16 rather than being incorporated into the mesopores. The H2PtCl6 precursor impregnated in the S16 support is difficult to completely reduce to metallic Pt due to the formation of PtOxCly species during calcination treatment. The attraction of –COOH groups in S16C to Pt4+ species may weaken the interaction between Pt4+ and Cl, thus leading to the high reducibility of the Pt NPs. The high reducibility of Pt NPs of S16C can induce intensive dissociation of toluene and thus cause very high catalytic activity for toluene oxidation. The toluene oxidation reaction may follow different reaction pathways on both Pt(1.0)@S16C and Pt(1.0)@S16 catalysts. The in situ IR spectra of toluene oxidation over the Pt(1.0)@S16C catalyst supported that breaking the strong C–C bonds between phenyl and methyl groups was the essential step for enhancing the catalytic activity for toluene oxidation. The CO and carbonyl species resulting from the methyl groups of toluene molecules were considered to be the major intermediates for further oxidation to CO2. The defect sites of Pt NPs of the Pt(1.0)@S16C catalyst may play a role in the adsorption and decomposition of toluene. However, the low reducibility of Pt NPs on S16 results in a poor ability to degrade toluene, leading to a low catalytic rate for toluene oxidation.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

Financial support from the Ministry of Science and Technology (MOST 106-2113-M-182-002) and from Chang-Gung Memorial Hospital (CMRPD5J0011) is gratefully acknowledged.

Notes and references

  1. J. Ruiz, R. Bilbao and M. B. Murillo, Environ. Sci. Technol., 1998, 32, 1079–1084 CrossRef CAS.
  2. R. J. Avery, Environ. Sci. Technol., 2006, 40, 4845–4850 CrossRef CAS PubMed.
  3. Y. J. Qin, J. P. Sheth and K. K. Sirkar, Ind. Eng. Chem. Res., 2002, 41, 3413–3428 CrossRef CAS.
  4. Y. Ding, Ind. Eng. Chem. Res., 2019, 58, 5008–5017 CrossRef CAS.
  5. T. L. Corley, J. Farrell, B. Hong and M. H. Conklin, Environ. Sci. Technol., 1996, 30, 2884–2891 CrossRef CAS.
  6. P. M. Potter, S. R. Al-Abed, D. Lay and S. M. Lomnicki, Environ. Sci. Technol., 2019, 53, 4364–4370 CrossRef CAS PubMed.
  7. K. Kosuge, S. Kubo, N. Kikukawa and M. Takemori, Langmuir, 2007, 23, 3095–3102 CrossRef CAS.
  8. F. Dong, W. Han, H. Zhao, G. Zhang and Z. Tang, Nanoscale, 2019, 11, 9937–9948 RSC.
  9. H. Huang, Y. Xu, Q. Feng and D. Y. C. Leung, Catal. Sci. Technol., 2015, 5, 2649–2669 RSC.
  10. H. Wang, W. Guo, Z. Jiang, R. Yang, Z. Jiang and Y. Pan, J. Catal., 2018, 361, 370–383 CrossRef CAS.
  11. J. Kong, G. Li, M. Wen, J. Chen, H. Liu and T. An, J. Catal., 2019, 370, 88–96 CrossRef CAS.
  12. H. Huang, C. Zhang, L. Wang, G. Li, L. Song, G. Li, S. Tang and X. Li, Catal. Sci. Technol., 2016, 6, 4260–4270 RSC.
  13. S. Xie, Y. Liu, J. Deng, J. Yang, X. Zhao, Z. Han, K. Zhang and H. Dai, J. Catal., 2017, 352, 282–292 CrossRef CAS.
  14. Y. Wang, L. Guo, M. Chen and C. Shi, Catal. Sci. Technol., 2018, 8, 459–471 RSC.
  15. Q. Yang, D. Wang, C. Wang, X. Li, K. Li, Y. Peng and J. Li, Catal. Sci. Technol., 2018, 8, 3166–3173 RSC.
  16. M. Y. Kim, T. Masui and N. Imanaka, Catal. Sci. Technol., 2014, 4, 321–324 RSC.
  17. P. Papaefthimiou, T. Ioannides and X. E. Verykios, Appl. Therm. Eng., 1998, 18, 1005–1012 CrossRef CAS.
  18. D. Delimaris and T. Ioannides, Appl. Catal., B, 2009, 89, 295–302 CrossRef CAS.
  19. M. Hosseini, T. Barakat, R. Cousin, A. Aboukaïs, B. L. Su and G. De Weireld, Appl. Catal., B, 2012, 111–112, 218–224 CrossRef CAS.
  20. A. C. C. Rodrigues, Catal. Commun., 2007, 8, 1227–1231 CrossRef CAS.
  21. A. L. Marsh, D. J. Burnett, D. A. Fischer and J. L. Gland, J. Phys. Chem. B, 2004, 108, 605–611 CrossRef CAS.
  22. B. Grbic, N. Radic and A. Terlecki-Baricevic, Appl. Catal., B, 2004, 50, 161–166 CrossRef CAS.
  23. A. A. Barresi, M. Cittadini and A. Zucca, Appl. Catal., B, 2003, 43, 27–42 CrossRef CAS.
  24. M. Paulis, H. Peyrard and M. Montes, J. Catal., 2001, 199, 30–40 CrossRef CAS.
  25. H. A. E. Dole, R. J. Isaifan, F. M. Sapountzi, L. Lizarraga, D. Aubert, A. Princivalle, P. Vernoux and E. A. Baranova, Catal. Lett., 2013, 143, 996–1002 CrossRef CAS.
  26. S. M. Saqer, D. I. Kondarides and X. E. Verykios, Top. Catal., 2009, 52, 517–527 CrossRef CAS.
  27. K. J. Kim and H. G. Ahn, Appl. Catal., B, 2009, 91, 308–318 CrossRef CAS.
  28. F. W. Yan, S. F. Zhang, C. Y. Guo, F. B. Li, F. Yan and G. Q. Yuan, Catal. Commun., 2009, 10, 1689–1692 CrossRef CAS.
  29. N. Radic, B. Grbic and A. Terlecki-Baricevic, Appl. Catal., B, 2004, 50, 153–159 CrossRef CAS.
  30. H. C. Wu, T. C. Chen, N. C. Lai, C. M. Yang, J. H. Wu, Y. C. Chen, J. F. Lee and C. S. Chen, Nanoscale, 2015, 7, 16848–16859 RSC.
  31. Y. T. Lai, T. C. Chen, Y. K. Lan, B. S. Chen, J. H. You, C. M. Yang, N. C. Lai, J. H. Wu and C. S. Chen, ACS Catal., 2014, 4, 3824–3836 CrossRef CAS.
  32. C. S. Chen, Y. T. Lai, T. W. Lai, J. H. Wu, C. H. Chen, J. F. Lee and H. M. Kao, ACS Catal., 2013, 3, 667–677 CrossRef CAS.
  33. D. Wang, Y. Zhu, C. Tian, L. Wang, W. Zhou, Y. Dong, Q. Han, Y. Liu, F. Yuan and H. Fu, Catal. Sci. Technol., 2016, 6, 2403–2412 RSC.
  34. Y. Li, S. Zhao, Q. Hu, Z. Gao, Y. Liu, J. Zhang and Y. Qin, Catal. Sci. Technol., 2017, 7, 2032–2038 RSC.
  35. L. Deng, H. Miura, T. Ohkubo, T. Shishido, Z. Wang, S. Hosokawa, K. Teramura and T. Tanaka, Catal. Sci. Technol., 2019, 9, 947–956 RSC.
  36. M. C. B. Soares, F. F. Barbosa, M. A. M. Torres, A. Valentini, A. R. dos Albuquerque, J. R. Sambrano, S. B. C. Pergher, N. Essayem and T. P. Braga, Catal. Sci. Technol., 2019, 9, 2469–2484 RSC.
  37. X. Cheng, D. Wang, J. Liu, X. Kang, H. Yan, A. Wu, Y. Gu, C. Tian and H. Fu, Nanoscale, 2018, 10, 22348–22356 RSC.
  38. J. Zhu, X. Xie, S. A. C. Carabineiro, P. B. Tavares, J. L. Figueiredo, R. Schomäcker and A. Thomas, Energy Environ. Sci., 2011, 4, 2020–2024 RSC.
  39. C. H. Tu, A. Q. Wang, M. Y. Zheng, X. D. Wang and T. Zhang, Appl. Catal., A, 2006, 297, 40–47 CrossRef CAS.
  40. K. Yoshida, C. G. Arellano, R. Luque and P. L. Gai, Appl. Catal., A, 2010, 379, 38–44 CrossRef CAS.
  41. C. M. Chanquía, K. Sapag, E. Rodríguez-Castellón, E. R. Herrero and G. A. Eimer, J. Phys. Chem. C, 2010, 114, 1481–1490 CrossRef.
  42. M. S. Kumar, D. Chen, J. C. Walmsley and A. Holmen, Catal. Commun., 2008, 9, 747–750 CrossRef CAS.
  43. Z. J. Wang, Y. B. Xie and C. J. Lir, J. Phys. Chem. C, 2008, 112, 19818–19824 CrossRef CAS.
  44. H. Song, R. M. Rioux, J. D. Hoefelmeyer, R. Komor, K. Niesz, M. Grass, P. Yang and G. A. Somorjai, J. Am. Chem. Soc., 2006, 128, 3027–3037 CrossRef CAS PubMed.
  45. H. Friedrich, J. R. A. Sietsma, P. E. de Jongh, A. J. Verkleij and K. P. de Jong, J. Am. Chem. Soc., 2007, 129, 10249–10254 CrossRef CAS PubMed.
  46. C. S. Chen, Y. T. Lai, T. C. Chen, C. H. Chen, J. F. Lee, C. W. Hsu and H. M. Kao, Nanoscale, 2014, 6, 12644–12654 RSC.
  47. C. S. Budi, D. Saikia, C. S. Chen and H. M. Kao, J. Catal., 2019, 370, 274–288 CrossRef CAS.
  48. C. S. Chen, C. S. Budi, H. C. Wu, D. Saikia and H. M. Kao, ACS Catal., 2017, 7, 8367–8381 CrossRef CAS.
  49. C. S. Budi, H. C. Wu, C. S. Chen, D. Saikia and H. M. Kao, ChemSusChem, 2016, 9, 2326–2331 CrossRef CAS PubMed.
  50. P. Concepcion, A. Corma, J. Silvestre-Albero, V. Franco and J. Y. Chane-Ching, J. Am. Chem. Soc., 2004, 126, 5523–5532 CrossRef CAS.
  51. J. Arenas-Alatorre, A. Gomez-Cortes, M. Avalos-Borja and G. Diaz, J. Phys. Chem. B, 2005, 109, 2371–2376 CrossRef CAS.
  52. S. Jongpatiwut, N. Rattanapuchapong, T. Rirksomboon, S. Osuwan and D. E. Resasco, Catal. Lett., 2008, 122, 214–222 CrossRef CAS.
  53. Y. Wang, H. H. Liu, S. Y. Wang, M. F. Luo and J. Q. Lu, J. Catal., 2014, 311, 314–324 CrossRef CAS.
  54. S. H. Wu, C. Y. Mou and H. P. Li, Chem. Soc. Rev., 2013, 42, 3862–3875 RSC.
  55. C. Daniel, M. O. Clarté, S.-P. Teh, O. Thinon, H. Provendier, A. C. Van Veen, B. J. Beccard, Y. Schuurman and C. Mirodatos, J. Catal., 2010, 272, 55–64 CrossRef CAS.
  56. S. D. Ebbesen, B. L. Mojet and L. Lefferts, J. Catal., 2007, 246, 66–73 CrossRef CAS.
  57. M. J. Lundwall, S. M. McClure and D. W. Goodman, J. Phys. Chem. C, 2010, 114, 7904–7912 CrossRef CAS.
  58. P. Bazin, O. Saur, J. C. Lavalley and M. Daturi, Phys. Chem. Chem. Phys., 2005, 7, 187–194 RSC.
  59. H. Wei, X. Liu, A. Wang, L. Zhang, B. Qiao, X. Yang, Y. Huang, S. Miao, J. Liu and T. Zhang, Nat. Commun., 2014, 5, 5634 CrossRef CAS.
  60. M. C. Tsai and E. L. Muetterties, J. Am. Chem. Soc., 1982, 104, 2534–2539 CrossRef CAS.

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c9cy01787a

This journal is © The Royal Society of Chemistry 2019