Ti-containing mesoporous silica for methylene blue photodegradation

Ti-containing mesoporous silica for methylene blue photodegradation

Applied Catalysis A: General 393 (2011) 359–366 Contents lists available at ScienceDirect Applied Catalysis A: General journal homepage: www.elsevie...

1MB Sizes 0 Downloads 22 Views

Applied Catalysis A: General 393 (2011) 359–366

Contents lists available at ScienceDirect

Applied Catalysis A: General journal homepage: www.elsevier.com/locate/apcata

Ti-containing mesoporous silica for methylene blue photodegradation Juan Matos a,∗ , Andreína García a , Sang-Eon Park b a b

Engineering of Materials and Nanotechnology Centre, Venezuelan Institute for Scientific Research, I.V.I.C., Panamerica Road, Km. 11, 20632, Caracas 1020-A, Miranda, Venezuela Laboratory of Nano-Green Catalysis and Nano-Center for Fine Chemicals Fusion Technology, Chemistry Department, Inha University, Incheon 402-751, Republic of Korea

a r t i c l e

i n f o

Article history: Received 26 October 2010 Received in revised form 3 December 2010 Accepted 10 December 2010 Available online 17 December 2010 Keywords: TUD-1 Titanium Photocatalysis Methylene blue

a b s t r a c t Photodegradation of methylene blue (MB) on Ti-containing mesoporous silica prepared by microwaveassisted irradiation as a function of Si/Ti molar ratio was studied. The materials were characterized by N2 adsorption, XRD, UV–vis/DR, and TEM. All solids showed mesoporous textures with high surface areas, relatively small pore size diameters and large pore volume. XRD showed that framework of solids consists of amorphous silica. It was found that the lower the Si/Ti ratio the higher the photocatalytic activity. Under irradiation with a lamp with more photons from visible light the sample with a Si/Ti ratio equal to 10 showed a higher photoactivity than that of a commercial TiO2 photocatalyst. This result was in agreement with the UV–vis/DR spectra which showed that material with a Si/Ti = 10 has slightly higher energy band gap than that of commercial TiO2 suggesting that these materials behave as a photoactive semiconductor. © 2010 Elsevier B.V. All rights reserved.

1. Introduction A significant quantity of the total world production of dyes and azo-dyes used in textile is released into effluents with concomitant environmental hazards [1]. Different technologies for the removal of these organic molecules, among which adsorption, bio- and chemical degradation methods including advanced oxidation technologies as heterogeneous photocatalysis have been employed. TiO2 has been used in a wide range of photocatalytic reactions because of its efficient band gap energy and intrinsic properties. A combination of photocatalysis and adsorption with different supports such as silica, alumina, zeolites, clays or activated carbon [2–13] has been employed to enhance the photocatalytic efficiency of titania. An interesting review about this topic by Zou and co-workers had recently been published [14]. Also, novel synthesis of carbon-supported nano crystalline TiO2 has received an increased attention for the degradation of different azo-dyes such as Chromotrope 2R [5], Orange-II [6], Rhodamine-B [7] and Direct Blue-53 [8]. Our group has previously pointed that surface functionalization of activated carbon played an important role in the enhancement of photoactivity of TiO2 towards the degradation of aromatic molecules such as phenol and 4-chlorophenol [10–12]. It has been suggested by our group [10,12] and other groups [13] that the specific functional groups on the carbon surface could assist the photooxidation process occurring on the TiO2 surface. Also, novel nanoscale TiO2 materials such as titanate-based nanotubes have recently received an important attention [15] in photocat-

∗ Corresponding author. Tel.: +58 212 5041166; fax: +58 212 5041166. E-mail addresses: [email protected], [email protected] (J. Matos). 0926-860X/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.apcata.2010.12.020

alytic reactions mainly because they have larger surface areas than conventional TiO2 nanoparticles as well as permit higher electron transfer [16] and efficient charge separation on their surfaces which reduce the recombination of photogenerated electrons and holes [17], and a decisive role in photocatalytic activities. Ti-containing mesoporous silica materials such as Ti-MCM-41 and Ti-SBA-15 have showed high catalytic activities in oxidation reactions [18]. Recently, TUD-1 type mesoporous materials are evoking great interest [19,20] due to advantages such as high thermal stability and three-dimensional random pore structures. Moreover, TUD-1 materials are becoming potential to be applied in catalytic oxidation processes [21,22] and heterogeneous photocatalysis [23,24] because of their tunable size, high surface area, and suitable metal or metal oxides incorporation by pre or post-synthetic methods such as wet-impregnation or direct incorporation methods. Considering the above facts, the objective of the present work is to directly incorporate desired amounts of Ti into the framework of TUD-1 by microwave-assisted methodology and to study the activity of Ti-containing mesoporous silica as a function of Si/Ti ratio in the photooxidation of a model dye such as methylene blue (MB) employing two lamps with different UV and visible photon flux. Results are compared with those obtained on the commercially available P25 TiO2 from Degussa. 2. Experimental 2.1. Materials and synthesis method TUD-1 materials with different loading of titanium were synthesized as follows [19]. Scheme 1 shows a representation of the microwave-assisted synthesis of these materials but it can be sum-

360

J. Matos et al. / Applied Catalysis A: General 393 (2011) 359–366

H2O + TEA Solution-2

+

+

NH4OH

TEOS + TPOT

20ºC, 24h Aged mixture Si/Ti

Mixture Si/Ti

integration of the radiation spectrum of each lamp. Radiation was verified in a pyranometer (Solar Light PMA2100) and total photon flux was estimated by Eq. (1) which permits to convert W m−2 in photons cm−2 s−1 . The properties of lamps are shown in Table 1 (see also supplementary material). Photons = cm2 s

 W   2.77 × 1018 photons   1

m2

.

Ws

.

1 m2 (100 cm)2



(1)

2.4. Procedure

Microwave, 600W Dry mixture Si/Ti Air, 600ºC, 10h Ti-TUD-1

Scheme 1. Synthesis of microwave assisted Ti-containing TUD-1 materials.

marized as follows. Appropriate amounts of water and tetraethyl ammonium, TEA (97%, Aldrich) were stirred for a few minutes to obtain a homogeneous solution (solution-1). After this, tetraethylortosilane TEOS (98%, Aldrich) and tetrapropyl orthotitanate, TPOT (96%, Aldrich) were added and stirred to form a homogeneous solution (solution-2). This solution was added to a solution of quaternary ammonium hydroxide (Aldrich) under vigorous stirring. This mixture with Si/Ti molar ratios of 10, 20 and 30 was aged at room temperature for 24 h, and directly followed by microwave irradiation at 600 W in a Teflon vessel at 180 ◦ C for 2 h. Samples were calcined in the presence of air at 600 ◦ C for 10 h and Ti-containing TUD-1 materials were denoted as Ti-TUD-1-10, TiTUD-1-20, and Ti-TUD-1-30 where 10, 20 and 30 correspond to the nominal Si/Ti ratios, respectively. Methylene blue (MB) was of analytical grade and purchased from Sigma–Aldrich. TiO2 P25 from Degussa was employed as a standard photocatalyst for comparison purpose. It consists of mainly anatase phase (ca. 80%), non-porous polyhedral particles of ca. 30 nm mean size and a BET surface area of ca. 45 m2 g−1 . 2.2. Characterization of solids Particle size and morphology were visualized using a transmission electron microscope (TEM) from JEOL (JEM-2010, operating at 200 kV). N2 adsorption–desorption isotherms and pore size distributions were recorded at 77 K employing an adsorption instrument from Micromeritics, ASAP 2020. BET surface areas, pore volumes and pore size diameters were determined. XRD patterns were recorded in reflection mode (CuK␣ radiation) on a Rigaku Multiflex diffractometer. Ultraviolet–visible diffuse reflectance spectra (UV–vis/DRS) were performed on a Shimadzu, Solid-Spec-3700 DUV. 2.3. Photoreactor and light source A previously reported [9–12] batch photoreactor was employed, which was open to air. It consists of a cylindrical flask made of Pyrex (ca. 200 mL) with a bottom optical window of 6 cm diameter. Irradiation was provided using two lamps with different UV proportions; a mercury lamp denoted as LHg and a metal halide lamp denoted as LMH . Irradiation was filtered by a circulating water cell (thickness ca. 2.0 cm) to remove IR beams and prevent from any heating of the suspension. The UV (from 320 to 390 nm) and visible light (from 400 to 780 nm) components in each lamp were estimated by

The photocatalytic tests were performed at 25 ◦ C using 30 mg of photocatalysts under stirring in 125 mL of MB (MB) [25 ppm (78.2 ␮mol/L) initial concentration]. The samples were maintained in the dark for 60 min in order to achieve adsorption at equilibrium prior to the UV-irradiation and then the suspension was irradiated. MB aliquots were analyzed after centrifugation in an UV-spectrophotometer Perkin Elmer, Lambda 35 at 664 nm and the remaining MB concentrations were estimated using a standard calibration curve. Photoactivity tests were done in triplicate in all samples and the reproducibility of results was better than 3%. 3. Results and discussion 3.1. Characterization 3.1.1. Crystallinity and morphology Fig. 1A shows the low angle XRD patterns of Ti-containing TUD-1 materials at diffraction angles between 1◦ and 10◦ . It can be observed that the present Ti-containing TUD-1 materials prepared with different Si/Ti ratios showed a single broad peak with a maxima at about 1.5◦ indicative of a wormhole-like or a disordered mesostructure [25,26]. This peak was clearly higher than that obtained by us for mesoporous silica SBA-16 also prepared by microwave-assisted methodology [20], suggesting that incorporation of Ti atoms remarkably influenced the crystalline and structure of meso-porous silica. Fig. 1B shows the wide angle XRD patterns of Ti-containing TUD-1 materials at diffraction angles between 10◦ and 80◦ . As expected, framework of solids consists in amorphous silica [27] with a broad signal with low intensity located at 2 = 23◦ typical of siliceous materials. This signal is attributed to diffuse dispersion caused by the lack of long-range order of Si atoms located on the walls of the channels in the materials based on silica. It must be pointed out that in the present work, no peaks were found corresponding to TiO2 framework indicating the incorporation of Ti atoms to the silica framework. Careful studies of the influence of isolated molecular Ti sites in tetrahedral coordination on SiO2 based materials as TUD-1 have been performed by Hamdy and co-workers [28]. They found that Ti-TUD-1 with Si/Ti ratio equal to 2.5 showed that samples were obtained with well defined different nanoparticle sizes and in spite of the pore structure of the solids is obscured by the depth of fields in the HR-TEM image, the TiO2 crystals in the anatase phase were identified by their electron diffraction fringes. However, it must be noted that the Si/Ti ratio in that work corresponds to a higher proportion of Ti to Si atoms (0.4 Ti atoms by 1 Si atom) than those presented in the Ti-TUD-1 materials of the present work that are equal to 0.1 Ti atom, 0.05 Ti atom, and 0.03 Ti atom, by 1 Si atom for Si/Ti = 10, 20, 30, respectively. In other words, as suggested by the XRD patterns from Fig. 1B, it would be difficult to identify TiO2 crystallites in the present work. TEM images in Fig. 2 shows that Ti-containing TUD-1 materials possessed the sponge-like structure with very disordered morphology in good agreement with XRD patterns from Fig. 1. It can be inferred from TEM images that Ti-containing TUD-1 materials are constituted by a self-assembly of meso-structured nanoparticles suggesting a like-hierarchical structure [29].

J. Matos et al. / Applied Catalysis A: General 393 (2011) 359–366

361

Table 1 Properties of Hg lamp (LHg ) UV and metal halide lamp (LMH ). Lamp

Total radiation (W m−2 )

UV radiation (W m−2 )

Visible radiation (W m−2 )

Total flux (photons cm−2 s−1 )

LHg LMH

445.5 522.7

82.9 70.2

362.6 452.5

1.23 × 1017 1.44 × 1017

Fig. 1. (A) XRD patterns of Ti-TUD-1 materials at low diffraction angles between 1◦ and 10◦ . (B) XRD patterns of Ti-TUD-1 materials at wide diffraction angles between 10◦ and 80◦ . (a) Ti-TUD-1-10, (b) Ti-TUD-1-20, and (c) Ti-TUD-1-30.

3.1.2. Porosity The N2 adsorption–desorption isotherms for Ti-TUD materials are depicted in Fig. 3. These curves show typical type-IV adsorption–desorption isotherms with clear hysteresis loops, indicating the framework of the materials could be assigned principally to mesoporous materials and with important micropore contribution. Also, it can be noted that the higher the Si/Ti ratios the slower the close trend of the hysteresis loop which could be associated with differences in the pore size distributions. It is evident from Fig. 4 that the pore size distribution of Ti-TUD-1-10 and TiTUD-1-20 materials has unimodal pore size distribution with a short range of pores between 2–5 nm and 2–7 nm, respectively. By contrast, Ti-TUD-1-30 material had a bimodal pore size distribution with broad ranges of pores between 2 and 8 nm and between 10 and 22 nm for the first and second distributions, respectively. A summary of the BET surface areas, total pore volume and mean pore diameters is compiled in Table 2. This table also includes a comparison of the textural properties of the present microwave-assisted Ti-TUD-1 materials against other SiO2 mesoporous materials such as SBA-16 [20,30]. The present microwave-assisted Ti-TUD-1 developed a total pore volume of about 0.70 cm3 g−1 , 0.90 cm3 g−1 , and 0.73 cm3 g−1 for Ti-TUD1-10, Ti-TUD-1-20 and Ti-TUD-1-30, respectively, which were relatively larger than those on microwave-assisted [20] and other types [30] of SBA-16 materials (Table 2). In addition, it can be seen

from Table 2 that the higher the Si/Ti ratios the lower the BET surface areas while the mean pore sizes clearly increase with the increase of Si/Ti ratio (Table 2). The SBET surface areas of the TiTUD-1 materials are clearly higher than those obtained on SBA-16 materials prepared by microwave-assisted or by hydrothermal synthesis [20] while they are lightly higher or similar than other types of SBA-16 elsewhere reported [30]. Ti-containing TUD-1 materials showed mean pore diameters in the mesoporous range with values of 3.6 nm, 4.0 nm, and 4.8 nm for Ti-TUD-1-10, Ti-TUD-1Table 2 Textural properties of Ti-TUD-1 materials. BET surface area (SBET ), total pore volume (Vpore ), mean pore diameter (dpore ). Photocatalyst

SBET (m2 g−1 )

Vpore (cm3 g−1 )

dpore (nm)

Reference

Ti-TUD-10 Ti-TUD-20 Ti-TUD-30 SBA-16a SBA-16b SBA-16c SBA-16d

866 782 750 430–672 624 660–939 623–815

0.70 0.9 0.73 0.38–0.59 0.53 0.42–0.61 0.37–0.53

3.6 4.0 4.8 7.5–8.7 7.9 4.7–5.6 4.8–5.6

This work This work This work [20] [20] [30] [30]

a

Microwave-assisted and as a function of stirring and microwave irradiation. Hydrothermal synthesis after 24 h stirring time and 120 min microwave irradiation. c Synthesis as a function of stirring time. d Synthesis as a function of aging time. b

362

J. Matos et al. / Applied Catalysis A: General 393 (2011) 359–366

Fig. 4. Pore size distributions of Ti-TUD-1 materials.

20 and Ti-TUD-1-30, respectively. These pore diameters are much lower (Table 2) than those obtained on SBA-16 prepared by the microwave-assisted or by hydrothermal synthesis [20] and slightly lower (Table 2) than those reported for SBA-16 mesoporous silica prepared without microwave-assistance [30]. It must be noted that total pore volume in Ti-TUD-1 materials is not monotonic as a function of Si/Ti ratio but the lower mean pore diameter and larger pore volume than other mesoporous silica materials such as SBA-16 [20,30] can be associated with a more disordered structure discussed above as a consequence of the incorporation of Ti atoms in the silica structure. An alternative discussion has been given earlier by Van der Voort et al. [30] who reported that amorphous silica can develop important micropore contributions to the silica framework by means of internal microporous nanocapsules created by the controlled interaction between tetraethylortosilane (TEOS) and the surfactant during the synthesis. In the present case, the interaction of Ti source (tetrapropyl orthotitanate) with TEOS by the microwave-assisted methodology could promote the formation of a like-hierarchical structure [29], suggested by TEM images from Fig. 2. Finally, it should be noted that the typical sizes of TiO2 nanoparticle inclusion [28] correspond well to the pore diameter shown in Table 2 and therefore the co-existence of TiO2 nanoparticles with those Ti-moieties in the Ti-TUD-1 framework is possible as we discussed above.

Fig. 2. TEM images of Ti-TUD-1 materials: (A) Ti-TUD-1-10, (B) Ti-TUD-1-20, and (C) Ti-TUD-1-30.

3.1.3. UV–vis/DRS Fig. 5 shows the UV–vis/DR spectra of the Ti-TUD-1 samples. From the extrapolation of tangential line at zero absorbance in Fig. 5, shifted values at about 335 nm were observed for both TiTUD-1-20 and Ti-TUD-1-30 while that at 365 nm was observed

Fig. 3. N2 adsorption–desorption isotherms of Ti-TUD-1 materials. Fig. 5. UV–vis/DR spectra of Ti-TUD-1 materials.

J. Matos et al. / Applied Catalysis A: General 393 (2011) 359–366 8

A 1,2

TI-TUD -1-10

TITUD -1-20

7

TiO2 P25

0,8

Ct/Co

MB Adsorbed (μmol)

Photolysis Lamp Hg TI-TUD-1-30 Lamp Hg TI-TUD-1-20 Lamp Hg TI-TUD-1-10 Lamp Hg TiO2 P25 Lamp Hg

1,0

TITUD -1-30

6

363

5

0,6 0,4

4 0,2 3

0,0 0

2

50

100

150

200

250

300

350

Time (min)

B1,4

0

1,2

0

30

60

90

120

Time (min)

Fig. 6. Kinetics of MB adsorption in the dark. Conditions: 30 mg TiO2 . 125 mL MB. Co = 25 ppm (78.2 ␮mol/L).

Ln(Co/Ct)

1

TiO2 P25 Lamp Hg

Ti-TUD-1-10 Lamp Hg TI-TUD-1-20 Lamp Hg

1,0

TI-TUD-1-30 Lamp Hg

0,8 0,6 0,4 0,2

Preliminary studies of MB adsorption at 20 ◦ C were performed on Ti-TUD-1 samples and on TiO2 P25 from Degussa for comparative purpose. Fig. 6 shows the kinetics of MB adsorption in the dark. This kinetic was followed by 120 min and in all cases, most of adsorption occurred within 30 min but in order to ensure a proper equilibrium of methylene blue adsorption in the dark a longer period (60 min) was applied prior to the photodegradation experiments. Table 3 contains a summary of MB adsorption in the dark (Adsdark ) after 60 min. It can be seen from data in Table 3 that the lower the Si/Ti ratios of Ti-TUD-1 materials the larger the adsorption of MB in the dark. After 60 min about 51%, 48% and 46% adsorption was observed for Ti-TUD-1-10, Ti-TUD-120 and Ti-TUD-1-30, respectively. It must be pointed out that the adsorbed MB on Ti-TUD-1 samples was clearly higher than that on TiO2 P25 (about 30%). Kinetics of MB photodegradations is shown in Figs. 7 and 8 for the Hg lamp (LHg ) and metal halide lamp (LMH ), respectively. Figs. 7A and 8A show the kinetics for the disappearance of MB and Figs. 7B and 8B show the linear regression of kinetic data of Figs. 7A and 8A. Also, Figs. 7A and 8A indicate that direct photolysis of MB was negligible under irradiation with any of the two lamps and total disappearance of MB was achieved on TiO2 P25 from Degussa after 150 min and about 250 min of irradiation with LHg and LMH lamps, respectively. It should be noted that none of TiTUD-1 materials was as active as TiO2 P25 using the mercury lamp (Fig. 7A). This is due to the fact that this lamp emits the specific line in the UV region (see supplementary material) for the most efficient photoactivation of TiO2 (about 365 nm). In addition, in the present study, the Pyrex glass of the water-recirculation cell was able to absorb most part of UV-photons from metal-halide lamp because their wavelengths were lower than 360 nm (complementary material). However, most part of UV-range from mercury lamp was not absorbed by this filter and it showed maxima intensity at about 380 nm in the emission spectra (supplementary material). In consequence, the photoreactor contained almost all UV-photons when under irradiation with the mercury lamp while in the case of metal halide lamp the photons within the reactor were mainly from vis-

0

10

20

30

40

50

60

70

80

90

Time (min) Fig. 7. (A) Kinetic of disappearance of MB under Hg lamp. (B) Lineal regression of kinetic data.

ible light. By contrast, under irradiation with LMH lamp, Ti-TUD-1 materials were comparable or higher in photoactivities than TiO2 P25. This is particularly clear for the Ti-TUD-1-10 sample which showed a higher photoactivity up to about 120 min reaction. After this time its photoactivity was clearly inhibited. The inhibition of activity can be explained in two different manners. Firstly, even though Ti-TUD-1 samples seem not to be as active as P25 using the mercury lamp, they can be inferred that the Ti-actives sites on the Ti-TUD-1 materials are remarkably lower than those on TiO2 P25. For example, in the Ti-TUD-1-30 material, the Si/Ti ratio is equal to 30. It means that for each Ti atom there are about 30 Si atoms. In other words, the total Ti-sites in the framework of this mesoporous silica-based material are clearly lower than those in pristine TiO2 . In other words, MB adsorbed in the dark on Ti-TUD-1 materials was higher than that adsorbed on TiO2 P25 (Fig. 6 and Table 3), the proportion of MB molecules close to Ti-actives sites should be clearly lower in the Ti-TUD-1-30 and Ti-TUD-1-20. For that rea-

A 1,2

Photolysis Lamp MH

TI-TUD-1-20 Lamp MH

Ct/Co

3.2. Methylene blue photodegradation

0,0

1,0

TI-TUD-1-30 Lamp MH

0,8

TI-TUD-1-10 Lamp MH

TiO2 P25 Lamp MH

0,6 0,4 0,2 0,0 0

50

100

150

200

250

300

350

Time (min)

B 1,4

TI-TUD-1-10 Lamp MH

TI-TUD-1-20 Lamp MH

1,2

Ln(Co/Ct)

for Ti-TUD-1-10. These values correspond to band gap energies of about 3.60 eV for Ti-TUD-1-20 and Ti-TUD-1-30 and 3.35 eV for TiTUD-1-10. This value is slightly higher than that of the standard commercial TiO2 [31] employed in this study (3.22 eV for P25), indicating that the present methodology of microwave-assisted synthesis allowed the formation of a Ti-containing mesoporous TUD-1 material with similar optoelectronic properties than that of commercial TiO2 , in particular, the band gap energy.

TI-TUD-1-30 Lamp MH

1,0

TiO2 P25 Lamp MH

0,8 0,6 0,4 0,2 0,0 0

10

20

30

40

50

60

70

80

90

Time (min) Fig. 8. (A) Kinetic of disappearance of MB under MH lamp. (B) Lineal regression of kinetic data.

364

J. Matos et al. / Applied Catalysis A: General 393 (2011) 359–366

Table 3 Adsorption in the dark of MB (Adsdark ) and apparent first-order rate constants (kapp ) for MB photodegradation on Ti-TUD-1 materials. Sample

TiO2 P25 Ti-TUD-1-10 Ti-TUD-1-20 Ti-TUD-1-30 a b

Adsdark a (%)

30 51 48 46

Lamp Hg

Lamp MH

kapp (×10−4 min−1 )

Aphoto b (a.u.)

kapp (×10−4 min− )

Aphoto b (a.u.)

152.01 62.07 42.65 47.04

1.00 0.40 0.28 0.30

66.12 94.64 58.80 59.90

1.00 1.43 0.89 0.90

After 60 min of adsorption in the dark to achieve equilibrium. Aphoto defined as photocatalytic activity relative to TiO2 P25 alone (kapp−i /kapp−TiO2 ).

Table 4 Initial rate (ro ) for MB photodegradation on Ti-TUD-1 materials and vis to UV–vis relative photoactivity (ϕrel ). Sample

TiO2 P25 Ti-TUD-1-10 Ti-TUD-1-20 Ti-TUD-1-30 a b

Lamp Hg

ϕrel

Lamp MH

kapp (×10−4 min−1 )

ro a (mmol L−1 min−1 )

kapp (×10−4 min−1 )

ro a (mmol L−1 min−1 )

152.01 62.07 42.65 47.04

8.32 2.38 1.73 1.99

66.12 94.64 58.80 59.90

3.62 3.63 2.39 2.53

b

(a.u.)

0.44 1.53 1.38 1.27

ro was obtained from relation: ro = Ceq-ads kapp . ϕrel = ro-MH /ro-Hg .

son, Ti-TUD-1-10 has better photoactivity than the others because the proportion of Ti-active sites in this case is two and three times higher than that on Ti-TUD-1-20 and Ti-TUD-1-30, respectively. Secondly, it has been previously observed in Ti-containing silica or alumina photocatalysts [3] that photogenerated oxidizing species that are mainly intermediate products prevent the adsorption of new pollutant molecules from solution on the original adsorbed sites. In other words, the intermediate products compete with new MB molecules for the adsorption active sites and hence reaction was retarded or inhibited. Takeda et al. [4] concluded that optimum adsorption strength of a gas molecule like propionaldehyde on the mesoporous SiO2 adsorbent is needed to improve titania’s photoactivity. Therefore, this inhibition might be resulted from a very strong adsorption of MB on Ti-TUD-1 materials, as the color changing in Ti-TUD-10 sample observed after reaction suggests. Fig. 9B shows a remarkably bluish color on Ti-TUD-10 in comparison with white color on pristine Ti-TUD-1-10 material (Fig. 9A). The apparent first-order rate constants (kapp ) could be drawn out from the linear regression of the first 90 min kinetic data assuming a first-order reaction mechanism [1,10]. Figs. 7B and 8B show the linear regression Ln(Co /Ct ) = f(t) of kinetic data of Figs. 7A and 8A, respectively. A summary of kinetic results is compiled in Table 3 which shows the adsorption of MB in the dark (Adsdark ), the apparent first-order rate constants (kapp ) and the photocatalytic activity relative to TiO2 P25 alone, Aphoto obtained from the comparison kapp−i /kapp−TiO2 . Table 3 shows that the lower the Si/Ti ratios the higher the photocatalytic activities. In addition, it should be pointed out that the higher the total pore volume (Table 2) of Ti-TUD-1 materials the lower the photoactivity (Table 4). This trend is monotonic and reproducible in both lamps and can be explained in terms of the above discussion regarding an inhibition effect in the photoactivity consequence that too many adsorbed molecules of MB would diffuse slowly to the photoactive sites [4]. The apparent first-order rate constants (kapp ) in Table 3 indicate that Ti-TUD-1-10 was clearly more photoactive than Ti-TUD-1-20 and Ti-TUD-1-30 under irradiation with both lamps. It can be pointed out that kapp obtained on Ti-TUD1 materials were clearly higher under irradiation of more visible photons (LMH lamp) than those obtained under irradiation of more UV-photons (LHg lamp). In addition, the kapp obtained on Ti-TUD-110 sample under irradiation with the MH lamp were clearly higher than those on TiO2 P25 under the same lamp (94.6 min−1 against 66.1 min−1 ) and therefore, Ti-TUD-1-10 sample was about 1.4 times

higher in photoactivity than TiO2 P25 (Table 3). These results point out that photoactivity of the synthesized Ti-TUD-1-10 photocatalyst was better than that of the standard P25 Titania under more photons from visible light (as shown by the results of kapp and Aphoto in Table 3). Table 4 shows the initial rates (ro ) obtained by

Fig. 9. Macroscopic images of Ti-TUD-1-10 sample. (A) Pristine material and (B) After photocatalytic reaction.

J. Matos et al. / Applied Catalysis A: General 393 (2011) 359–366

(3.22 eV for Degussa P25) indicating that the microwave-irradiation permits the synthesis of a photoactive Ti-containing TUD-1, particularly having the Si/Ti ratio equal to 10, with similar optoelectronic properties than those of TiO2 P25.

A1.2 1.0

Ct/Co

365

0.8 0.6

3.3. General discussion

0.4 0.2 0.0

0

30

60

90

120

150

180

210

240

270

300

70

80

90

Time (min)

Ln(Co/Ct)

B 1.4 1.2 1.0 0.8 0.6 0.4 0.2 0.0

0

10

20

30

40

50

60

Time (min) Fig. 10. (A) Kinetic of disappearance of MB under Na lamp. (B) Lineal regression of kinetic data.

the equation [1,10] ro = Ceq-ads kapp

(2)

where Ceq-ads is the MB concentration at solution after achieving the equilibrium of adsorption in the dark and kapp are the apparent first-order rate constants. Initial rates permit to compare the relative photoactivity (ϕrel ) between both lamps obtained from: ϕrel = ro-MH /ro-Hg

(3)

where ro-MH and ro-Hg correspond to the initial rates of MB photodegradation obtained under irradiation with the metal halide lamp (LMH ) and mercury lamp (LHg ), respectively. It can be seen from Table 4 that ro values obtained on Ti-TUD-1 materials under irradiation with the lamp having more photons from visible light (LMH ) were clearly higher than those obtained under irradiation with the lamp with more photons from UV light (LHg ). By contrast, TiO2 -P25 followed an opposite behavior; ro was clearly lower under irradiation with the lamp with more photons from visible light (LMH ). In short, Table 4 shows that ϕrel is higher than unity for the Ti-TUD-1 materials in particular that the lower the Si/Ti ratio the higher the ϕrel . This seems to suggest that the higher the incorporation of Ti atoms to the mesoporous silica the higher the photoactivity of the materials. On the other hand, ϕrel for the commercial TiO2 -P25 was quite lower than unity which indicates that Ti-TUD-1 materials were more able to absorb photons from visible light than TiO2 P25. The present results are in agreement with our previous report on TiO2 -C photocatalysts which reasoned the presence of an important surface interaction between the active sites and support [31,32]. This surface interaction would be responsible for the increase in photoactivity caused by the photo-assistance [31] of the support (mesoporous silica) and/or by changes in the electron density [32] due to the partial reduction of Ti+4 to Ti+3 around the active sites (Ti atoms). Therefore, the particular photoactivity of Ti-TUD-1-10 sample under irradiation with LMH lamp cannot be ascribed simply to an adsorption effect onto the high surface area materials. As discussed above, from the extrapolation of tangential line at zero absorbance in Fig. 5, a shift at about 360 nm was obtained for Ti-TUD-1-10 which corresponds to the band gap energy of about 3.35 eV. This value was slightly larger than that of the standard commercial TiO2 employed in this study

At this point of the work, it is worthy to note several features. In spite of Ti-TUD-1-10 material that seems to be more photoactive than P25 TiO2 when excited by a metal hydride lamp, it should be pointed out that the higher activity was observed only in the first 90 min after which the catalyst seems to deactivate. This could be the consequence that the initial concentration of MB in this work is clearly much higher than those commonly employed. For example, Nakashima and co-workers [33] have reported an interesting study of the photodecomposition of methylene blue as a target molecule on titania hollow spheres. They employed only 4 ppm of MB which is clearly lower in comparison with the initial concentration employed in the present work (25 ppm). Nakashima also found that after 90 min the photoactivity of TiO2 spheres collapses in some cases after 75–80% MB conversion. In the present case, with more than 6 times higher initial MB concentration, Ti-TUD-1 materials seem to be deactivated after 40–60% and 60–80% of MB conversion under irradiation with Hg and MH lamps, respectively. In other words, we do believe that the Ti-incorporated TUD-1 materials will be able to be employed for the MB photodegradation in more diluted systems and this is the matter of our future research. Also, we can exclude the photocatalytic activity due to heating of the catalyst by absorption of light by the Ti-TUD-1 materials because only an increase of about 1 ◦ C was detected after 5 h irradiation. There are many reports [9–12,34] which suggest that suspension did not suffer from any heating during irradiation. This was due to the fact that our photocatalytic system was constructed with a water-recirculating cell composed of Pyex glass. Water inhibits the absorption of IR and then could eliminate any risk of temperature rise in the suspension. Pyrex glass is borosilicate which is a well-known filter for UV light lower than 360 nm. On the other hand, in spite of the high cost of TUD-1 materials and the deactivation phenomena discussed above which can be solved working on diluted systems, the value of the present work should be remarked. The benefit of the TUD-1 based materials as compared to other silica based supports, lies mainly in the formation of well dispersed active phases up to a loading of approximately 5 wt% for group V and group VI elements [35]. This is normally difficult to achieve with commercially available silica, or other micro-, or mesoporous materials [35]. Furthermore, at high loading (10–60 wt%), the synthesis procedure of TUD-1 allows a high level of control of the pore structure and nanoparticle size distribution for some metal oxides, ideal to investigate nanoparticle size effects in photocatalysis. In this sense, some metal-containing TUD1 materials (M-TUD-1) have been successfully studied in selective photocatalytic reactions, such as Cr-TUD-1 which is highly active in photocatalytic oxidation of propane [24,36], V-TUD-1 is highly selective in liquid cyclohexene photo-oxidation [37], and Ti-TUD-1 is highly selective for the photo-oxidation of propane to acetone the desired product [28,35]. It is also important to note that this is a preliminary study and our goal is to scale these results to a solar-plant photo-reactor. It is well-known that 5–8% of the solar spectrum is UV-light and therefore, the results obtained in the present study with the metal halide could be extrapolated to this goal. In this sense, Ti-TUD-1-10 was selected for additional experiments of methylene blue photodegradation under irradiation with a sodium lamp (LNa ). This lamp is characterized for a radiation spectrum with 99% of visible light (see supplementary material). Kinetics of MB photodegradation on TiTUD-1-10 material and TiO2 P25 under irradiation with the sodium

366

J. Matos et al. / Applied Catalysis A: General 393 (2011) 359–366

lamp is shown in Fig. 10A. As expected, direct photolysis of MB was negligible under irradiation with this lamp and total disappearance of MB was not achieved in any of two photocatalyst. The kapp obtained from the linear regression Ln(Co /Ct ) = f(t) of kinetic data (Fig. 10B) were equal to 37.38 min−1 and 43.49 min−1 for TiO2 P25 and Ti-TUD-1-10, respectively, which indicate that Ti-TUD-1-10 is more photoactive than TiO2 P25 by a lightly factor of about 1.2. It is important to remark that after 210 min both the photocatalysts showed a clear inhibition in the photoactivity; however, the fact that under pure visible light irradiation, the photocatalytic activity of Ti-TUD-1-10 is lightly higher (Fig. 10A), or in the worse of cases very similar to that of TiO2 P25, shows the valuable of Ti-TUD-1 materials. Accordingly, our present enforces are focused to employ Ti-TUD-1-10 in the treatment of diluted polluted water and in several selective photo-oxidation of hydrocarbons in liquid and gas phase as discussed above. 4. Conclusions Microwave-assisted preparation and direct incorporation of desired amounts of Ti by controlled Si/Ti ratios into the framework of TUD-1 were performed. The photocatalytic degradation of MB was studied under two lamps with different UV–vis light portions. It was found that the lower the Si/Ti ratio the higher the photocatalytic activity and photoactivity was higher under a lamp with more photons from visible light. Photoactivity of the Ti-containing TUD1 material with a Si/Ti ratio equal to 10 was slightly higher than that on a commercial TiO2 photocatalyst under visible light, which suggests that silicate composition of mesoporous Ti-TUD-1 could photo-assist the TiO2 in the photodegradation of MB. UV–vis/DR showed that material with a Si/Ti = 10 had a slightly higher band gap energy than that of conventional TiO2 suggesting that this sample could be considered as a photoactive semiconductor. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at doi:10.1016/j.apcata.2010.12.020. References [1] A. Houas, H. Lachheb, M. Ksibi, E. Elaloui, C. Guillard, J.-M. Herrmann, Appl. Catal. B: Environ. 31 (2007) 145–157. [2] J.F. Tanguay, S.L. Suib, R.W. Coughlin, J. Catal. 117 (1989) 335–347. [3] C. Minero, F. Catozzo, E. Pelizetti, Langmuir 8 (1992) 481–486. [4] N. Takeda, T. Torimoto, S. Sampath, S. Kuwabata, H. Yoneyama, J. Phys. Chem. 99 (1995) 9986–9991.

[5] W. Wang, C. Gomez Silva, J.L. Faria, Appl. Catal. B: Environ. 70 (2007) 470– 478. [6] J.M. Peralta, J. Manríquez, Y. Meas, F.J. Rodríguez, T.W. Chapman, M.I. Maldonado, L.A. Godínez, J. Hazard. Mater. 147 (2007) 588–593. [7] W. Ren, Z. Ai, F. Jia, L. Zhang, X. Fan, Z. Zou, Appl. Catal. B: Environ. 69 (2007) 138–144. [8] N. Sobana, M. Swaminathan, Sol. Energy Mater. Sol. Cells 91 (2007) 727–734. [9] T. Cordero, C. Duchamp, J.M. Chovelon, C. Ferronato, J. Matos, Appl. Catal. B: Environ. 73 (2007) 227–235. [10] J. Matos, J. Laine, J.-M. Herrmann, D. Uzcategui, J.L. Brito, Appl. Catal. B: Environ. 70 (2007) 461–469. [11] T. Cordero, C. Duchamp, J.M. Chovelon, C. Ferronato, J. Matos, J. Photochem. Photobiol. A: Chem. 191 (2007) 122–131. [12] J. Matos, A. García, T. Cordero, J.-M. Chovelon, C. Ferronato, Catal. Lett. 130 (2009) 568–574. [13] C. Lettmann, K. Hildenbrand, H. Kisch, W. Macyk, W.F. Maier, Appl. Catal. B: Environ. 32 (2001) 215–227. [14] W. Zhang, L. Zou, L. Wang, Appl. Catal. A: Gen. 371 (2009) 1–9. [15] M. Grandcolas, A. Louvet, N. Keller, V. Keller, Angew. Chem. Int. Ed. 48 (2009) 161–164. [16] Z.R. Tian, J.A. Voigt, J. Liu, B. Mckenzie, H. Xu, J. Am. Chem. Soc. 125 (2003) 12384–12385. [17] T. Tachikawa, S. Tojo, M. Fujitzuka, T. Sekino, T. Majima, J. Phys. Chem. B 110 (2006) 14055–14059. [18] Z. Shan, J.C. Jansen, L. Marchese, Th. Maschmeyer, Micropor. Mesopor. Mater. 48 (2001) 181–187. [19] Y.K. Kwang, J.-C. Chang, S.-E. Park, D.S. Kim, Y.-U. Kwon, S.H. Jhung, J.S. Kwang, M.S. Park, Angew. Chem. Int. Ed. 117 (2005) 562–566. [20] Y.K. Kwang, J.-C. Chang, Y.-U. Kwon, S.-E. Park, Micropor. Mesopor. Mater. 68 (2004) 21–27. [21] R. Anand, M.S. Hamdy, R. Parton, T. Maschmeyer, J.C. Jansen, U. Hanefeld, Appl. Catal. A: Gen. 355 (2009) 78–82. [22] R. Anand, M.S. Hamdy, R. Parton, T. Maschmeyer, J.C. Jansen, R. Glaser, F. Kapteijn, U. Hanefeld, Aust. J. Chem. 62 (2009) 360–365. [23] T. Maschmeyer, M. Che, Angew. Chem. Int. Ed. 49 (2010) 1536–1539. [24] O. Berg, M.S. Hamdy, T. Maschmeyer, J.A. Moulijn, M. Bonn, G. Mul, J. Phys. Chem. C 112 (2008) 5471–5475. [25] S.A. Bagshow, E. Prouzet, T.J. Pinnavaia, Science 269 (1995) 1242–1244. [26] H.F. Poulsen, J. Neuefeind, H.B. Neumann, J.R. Schneider, M.D. Zeidler, J. NonCryst. Solids 188 (1995) 63–74. [27] O. González, H. Pérez, P. Navarro, L.C. Almeida, J.G. Pacheco, M. Montes, Catal. Today 148 (2009) 140–147. [28] M.S. Hamdy, O. Berg, J.C. Jansen, T. Maschmeyer, J.A. Moulijn, G. Mul, Chem. Eur. J. 12 (2006) 620–628. [29] A. Corma, P. Atienzar, H. García, J.-Y. Chane-Ching, Nat. Mater. 3 (2004) 394–397. [30] P. Van der Voort, M. Benjelloun, E.F. Vansant, J. Phys. Chem. B 106 (2002) 9027–9032. [31] J. Matos, E. García-López, L. Palmisano, A. García, G. Marci, Appl. Catal. B: Environ. 99 (2010) 170–180. [32] J. Matos, A. García, P.S. Poon, J. Mater. Sci. 45 (2010) 4934–4944. [33] A. Syoufian, O.H. Satriya, K. Nakashima, Catal. Commun. 8 (2007) 755–759. [34] J. Matos, J. Laine, J.-M. Herrmann, Appl. Catal. B: Environ. 18 (1998) 281–291. [35] S. Telalovic, A. Ramanathan, G. Mul, U. Hanefeld, J. Mater. Chem. 20 (2010) 642–658. [36] M.S. Hamdy, O. Berg, J.C. Jansen, T. Maschmeyer, A. Arafat, J.A. Moulijn, G. Mul, Catal. Today 117 (2006) 337–342. [37] G. Mul, W. Wasylenko, M.S. Hamdy, H. Frei, Phys. Chem. Chem. Phys. 10 (2008) 3131–3137.