Synthesis and photocatalytic properties of nanocrystalline Au, Pd and Pt photodeposited onto mesoporous RuO2-TiO2 nanocomposites

Applied Catalysis A: General 431–432 (2012) 62–68

Contents lists available at SciVerse ScienceDirect

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

Synthesis and photocatalytic properties of nanocrystalline Au, Pd and Pt photodeposited onto mesoporous RuO2 -TiO2 nanocomposites Adel A. Ismail a,c,∗ , Detlef W. Bahnemann b , Saleh A. Al-Sayari c a b c

Advanced Materials Department, Central Metallurgical R&D Institute, CMRDI, P.O. Box: 87, Helwan 11421, Egypt Institut für Technische Chemie, Leibniz Universität Hannover, Callinstrasse 3, 30167 Hannover, Germany Centre for Advanced Materials and Nanoengineering (CAMNE), Najran University, P.O. Box: 1988, Najran 11001, Saudi Arabia

a r t i c l e

i n f o

Article history: Received 11 January 2012 Received in revised form 11 April 2012 Accepted 12 April 2012 Available online 21 April 2012 Keywords: Noble metals RuO2 -TiO2 ;Mesoporous Photooxidation Methanol UV and visible light

a b s t r a c t Noble metals (Au, Pd and Pt) have been photodeposited onto hexagonal mesoporous RuO2 -TiO2 nanocomposites to study their influence on the photocatalytic activity under UV and visible lights by determination of the formation rate of HCHO generated by photooxidation of CH3 OH in aqueous solution. X-ray diffraction (XRD) patterns and N2 sorption isotherms reveal that highly crystalline TiO2 and mesoporous structure have been formed with high surface area (150–180 m2 /g) and pore diameter ranging from 6.8 to 7.3 nm. TEM measurements show that the framework of the highly crystalline mesoporous RuO2 TiO2 is composed of anatase phase grown along [1 0 1] direction. The dependence of HCHO formation rate on the noble metals/RuO2 -TiO2 nanocomposites, behaves quite differently depending on noble metals as electrons sink. Under UV light, the findings reveal that Pd/RuO2 -TiO2 offers an improvement in term photooxidation rate of CH3 OH and photonic efficiencies over Pt/RuO2 -TiO2 and Au/RuO2 -TiO2 . However, under visible light, the photocatalytic activity of mesoporous RuO2 -TiO2 containing Au nanoparticles towards CH3 OH oxidation is remarkable and the photonic efficiency of RuO2 -TiO2 has been improved two times. Pt and Pd nanoparticles could not be observed in any improvement in photocatalytic activity of RuO2 -TiO2 under visible light. © 2012 Elsevier B.V. All rights reserved.

1. Introduction TiO2 anatase can only be excited by UV irradiation ( < 380 nm) because of its large band gap energy of 3.2 eV. Moreover, the rapid recombination of photoinduced electrons and holes greatly lowers the quantum efficiency [1]. Therefore, it is of great interest to improve the generation and separation of photoinduced electron–hole pairs in TiO2 for further applications. The manipulation of semiconductor heterostructures is one of the effective methods for photoinduced electron–hole generation and separation in recent years [2,3]. Multiple-semiconductor devices can absorb a larger fraction of the solar spectrum, which is beneficial for the excitation of the semiconductor and thus the photoinduced generation of electrons and holes. Moreover, the coupling of two different semiconductors could transfer electrons from an excited small band gap semiconductor into another attached one in the case of proper conduction band potentials [4–6]. This favors the separation of photoinduced electrons and holes and thus improves

∗ Corresponding author at: Advanced Materials Department, Central Metallurgical R&D Institute, CMRDI, P.O. Box 87, Helwan 11421, Egypt. Tel.: +20 225010643; fax: +20 225010639. E-mail address: [email protected] (A.A. Ismail). 0926-860X/$ – see front matter © 2012 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.apcata.2012.04.024

the photocatalytic efficiency of semiconductor heterostructure dramatically. The metal-semiconductor (MS) contact is one of the most widely used rectifying contacts in the electronics and photocatalysis applications [7]. When a metal and a semiconductor are joined to form MS interface, a significant redistrubtion of charge is expected to take place due to overlap of wave functions from two sides [7b]. Due to the technological importance of Schottky barrier and the most simple of the MS contact devices which are of the importance in the electron–hole separation, and, indeed, in the photocatalytic process. The nonideal behavior observed in Schottky barrier diodes has been generally attributed to the effect of interface states and the interfacial layer, which are present between metal contact and semiconductor. Thereby, the performance and reliability of Schottky barrier diodes generally depend on interface state density and their energy distribution [8]. Doping of noble metals (Au, Pd, Ag, Pt) with mesoporous TiO2 photocatalysts was proposed to enhance the photocatalytic activity due to their different Fermi levels, characterized by the work function of the metals and the band structure of the semiconductors. Upon contact, a Schottky barrier can be formed between the TiO2 and the noble metals, leading to a rectified charge carrier transfer [9–12]. 3D mesoporous TiO2 network acts as an antenna system transferring the initially generated electrons from the location of light

A.A. Ismail et al. / Applied Catalysis A: General 431–432 (2012) 62–68

absorption to a suitable interface with the noble metal catalyst where the actual electron transfer reaction will take place [11b]. Within this antenna model, it can be envisaged that the overlap of the energy bands of the nanoparticles forming this network will result in unified energy bands for the entire system enabling a quasi-free movement of the photogenerated charge carriers throughout. The antenna effect is explained by considering the mesoporous TiO2 network to behave like an antenna system for the photogenerated electrons enabling them to reach the respective TiO2 /noble metals contacts with high probability [12]. The simultaneous achievement of high crystallinity and ordered mesoporous TiO2 frameworks with high thermal stability has been reported [13]. Due to ordered large channels facilitating the diffusion within the frameworks and the thermal behavior of mesoporous titania materials is of central importance to the practical applications [13a]. Mesoporous TiO2 is an interesting material for photocatalytic applications due it is continuous, which may be beneficial compared to separated individual nanoparticles, in particular for catalyst recovery. The reason of high photocatalytic activity of mesoporous TiO2 is explained by, (i) the larger specific surface area of TiO2 mesoporous versus Degussa P25 powder; hence, there are more reactant adsorption/desorption sites for catalytic reaction. (ii) The prevention of the unwanted aggregation of the nanoparticles clusters, which is also helpful in maintaining the high active surface area. (iii) The highly porous structure, which allows rapid diffusion of various reactants and products during the reaction. (iv) The smaller crystal size, which means more powerful redox ability owing to the quantum-size effect; moreover, the smaller crystal sizes are also beneficial for the separation of the photogenerated hole and electron pairs [14–16]. Also, preparation of RuO2 -TiO2 has been investigated for enhancement electrical and photocatalytic properties under UV and visible light [4,17,18]. In the present study, we have employed CH3 OH to trap the • OH radicals produced in the photocatalytic process. CH3 OH is oxidized to HCHO, HCOOH and finally to CO2 [19]. We sought to enhance photooxidation of CH3 OH to generate HCHO by using the following strategies: (i) employ high surface area RuO2 -TiO2 mesostructured, with highly crystalline TiO2 to facilitate efficient transfer of photogenerated charge carriers to the surface species; (ii) modify the TiO2 band gap by addition 0.5 wt% RuO2 to absorb and utilize the visible portion of the solar spectrum where the bulk of the solar energy lies [4a]; (iii) distribute cocatalyst nanoparticles Au, Pd and Pt on the mesoporous RuO2 -TiO2 array surface to help the redox process. Therefore, we describe the synthesis of photodeposition of Au, Pd and Pt onto RuO2 -TiO2 nanocomposites as efficient photocatalysts.

2. Experimental Materials: The block copolymer surfactants EO106 PO70 EO106 (F 127, EO = CH2 CH2 O ,PO CH2 (CH3 ) CHO ), MW 12600 g/mol) from Sigma, ruthenium(III)acetylacetonate Ru(C5 H7 O2 )3 , tetrabutyl orthotitanate (TBOT), CH3 COOH, HCl, C2 H5 OH, CH3 OH, HAuCl4 , H2 PtCl6 and K2 PdCl4 were purchased from Aldrich and were used without further purification. RuO2 -TiO2 synthesis: 0.5 wt% RuO2 -TiO2 nanocomposites were synthesized by a simple one-step sol–gel process in the presence of F127 triblock copolymer as structure directing agent [4a]. In typical, the molar ratio of each reagent in the starting solution was fixed at TiO2 /F127/C2 H5 OH/HCl/CH3 COOH = 1:0.02:50:2:4. 1.6 g of F127, 2.3 ml of CH3 COOH and 0.74 ml of HCl were dissolved in 30 ml of ethanol and then added 3.5 ml of TBOT [20]. The calculated amount of Ru(C5 H7 O2 )3 was added to (F127 TBOT CH3 COOH) mesophase to obtain 0.5 wt% RuO2 -TiO2 nanocomposites and the mixtures were stirred vigorously for 60 min and transferred into a Petri dish. The ethanol was evaporated at 40 ◦ C with a relative humidity 40%

63

for 12 h and then it was transferred into a 65 ◦ C oven and aged for an additional 24 h. As-made samples were calcined at 450 ◦ C in air for 4 h at heating rate of 1 ◦ C/min and cooling rate of 2 ◦ C/min in air to remove the surfactant and to obtain mesoporous RuO2 -TiO2 nanocomposites. 2.1. Photodeposition Au, Pd and Pt/RuO2 -TiO2 Au, Pd and Pt/RuO2 -TiO2 photocatalysts were prepared by the photo-reduction method. 0.5 g RuO2 -TiO2 nanocomposites was mixed with HAuCl4 , H2 PtCl6 or K2 PdCl4 solution containing the equivalent amount of Au3+ , Pd2+ /or Pt4+ followed by the addition of methanol (1% v/v methanol/H2 O) to obtain 0.5 wt% noble metals/RuO2 -TiO2 . The suspension was magnetically stirred for 12 h under UV illumination. After irradiation, the samples have been centrifuged, washed then dried at 110 ◦ C overnight to obtain Au, Pd, and Pt/RuO2 -TiO2 . The synthesized samples are undoped TiO2 , RuO2 -TiO2 , Au/RuO2 -TiO2 , Pd/RuO2 -TiO2 and Pt/RuO2 -TiO2 denoted as TiO2 , RT, Au/RT, Pd/RT and Pt/RT respectively. 3. Characterization Wide angle X-ray diffraction (WXRD) data was collected on a Bruker AXS D4 Endeavour X diffractometer using Cu K␣1/2 , ␭␣1 = 154.060 pm, ␭␣2 = 154.439 pm radiation. Three thousand data points were collected with a step width of 0.02◦ in the 2 range from 20 to 80◦ . The phase analysis by the Rietveld method was carried out by using TOPAS 2.0 (Bruker AXS) software. Lattice parameters and crystallite size of all phases were refined. Structural data for the known phases were taken from ICDD PDF-2 database with PDF numbers: TiO2 (anatase) [92629]. Small angle X-ray diffraction (SXRD) patterns were recorded on a Bruker D8 advance equipped with a Göbel mirror and a secondary Ni-filter. Three hundred data points were collected with a step width of 0.02◦ in the 2 range from 0.2 to 6◦ . The nitrogen adsorption and desorption isotherms at 77 K were measured using a Quantachrome Autosorb 3B after vacuum-drying the samples at 200 ◦ C overnight. The sorption data were analyzed using the Barrett–Joyner–Halenda (BJH) model with Halsey equation [21]. FT-IR spectra were recorded with a BRUKER IF 66 spectrometer using the standard KBr pellet method. TEM measurements were conducted at 200 kV with a JEOL JEM-2100F-UHR field-emission instrument equipped with a Gatan GIF 2001 energy filter and a 1k-CCD camera in order to obtain EEL spectra. The bandgap energy of the catalysts was determined using diffuse reflectance spectroscopy (DRS). The reflectance spectra of the samples over a range of 200–700 nm were recorded with a Varian Cary 100 Scan UV–vis system equipped with a Labsphere integrating sphere diffuse reflectance accessory and using BaSO4 as reference material [22]. The reflectance data was converted to the absorption coefficient F(R) values according to the Kubelka–Munk equation [23]. The modified Kubelka–Munk function was determined using the equation as follows:

 F(R) =

(1 − R)2 × h 2R

1/2

where R is the proportion of light reflected, h is Planck’s constant, and  is the frequency of light. Plots of this parameter were used to determine the bandgap energy from the linear portion of the absorption transition using least squares regression and extrapolating to zero at the corresponding photon energy. The bandgap energies of catalysts were calculated according to the equation Eg = hc/, where Eg is the bandgap energy (eV), c the light velocity (m/s), and  the wavelength (nm).

A.A. Ismail et al. / Applied Catalysis A: General 431–432 (2012) 62–68

3.1. Photocatalytic activity tests

=

r × 100 I

where  is the photonic efficiency (%), r the photooxidation rate of methanol (mol L−1 s−1 ), and I incident photon flux. The incident photon flux in the wavelength range 300 nm ≤ I ≤ 400 nm and 400 nm ≤ I ≤ 500 were determined by ferrioxalate actinometry [26] to be 4.94 × 10−6 Ein L−1 s−1 and 3.4 × 10−7 Ein L−1 s−1 for UV and visible light, respectively. The actinometry was performed in the same photochemical reactor with the same volume of actinometric solution as the photocatalytic test, eliminating the errors associated with the influence of light reflections and reactor geometry. A 10 cm water bath and a black cut-off filter (3 mM, UG1 SCHOTT glass) were used during photon flux and photonic efficiency measurements.

RuO2-TiO2

22000

10

30000

Au/RuO2-TiO2 Pd/RuO2-TiO2

20000

Pt/RuO2-TiO2

25000

18000

Intensity/ counts

The photocatalytic oxidation activity was evaluated the formation HCHO produced from aqueous solution CH3 OH [30 mM] under UV light irradiations with wavelengths > 320 nm and visible light at wavelength >420 nm. Wavelengths were achieved using a 450 W medium pressure xenon lamp (Osram) placed inside a quartz jacket and equipped with a cooling tube. The photoreactor consisted of a quartz reactor with an effective volume of 75 ml and UV irradiation and visible light. The lamp was switched on 30 min before the beginning of the reaction to stabilize the power of its emission spectrum and the reactor was cooled by the circulation of H2 O. Reactions were carried out suspending 1 g/l of the photocatalysts and purging oxygen through the reaction vessel continuously. The suspensions were sonicated at the desired concentration before the experiment was started. As a result of photooxidation of methanol, the produced HCHO samples were withdrawn at regular intervals from the upper part of the reactor with the catalyst being removed from the liquid phase by filtration through nylon syringe filters (pore size: 0.45 ␮m). The photooxidation rate was determined by measuring the HCHO production employing the Nash method [24]. This method is based on the reaction of HCHO with acetylacetone and ammonium acetate to form a yellow colored product with a maximum of absorbance at 412 nm. Measurements were carried out using a Varian Cary 100 Scan UV–vis spectrophotometer, following an incubation time of 15 min at 60 ◦ C. The photonic efficiency was calculated for each experiment as the ratio of the photocatalytic degradation rate and the incident light intensity as given in the following equation [25].

24000

Intensity(a.u.)

64

16000 14000 12000

20000

15000

10000

20

10000 5000

8000 6000

0 0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

5.5

6.0

2θ

4000 2000 0 0.25

0.50

0.75

1.00

1.25

1.50

1.75

2.00

2.25

2.50

2.75

3.00

2θ Fig. 1. SAXS patterns of RuO2 -TiO2 , Au/RuO2 -TiO2 , Pd/RuO2 -TiO2 , Pt/RuO2 -TiO2 calcined at 450 ◦ C for 4 h and as made RuO2 -TiO2 (Inset).

formed and no reflections of any other phases of TiO2 . Interestingly, XRD data does not show any presence of either RuO2 or noble metals. The anatase intensity peaks decreased two times after RuO2 addition and no significant shifts of the anatase peaks were observed, indicating that RuO2 nanoparticles incorporated into TiO2 lattice [4a]. This is attributed to TiO2 was coated completely with Ru(III)–acetylacetonate complex in the mixed sol, and then heated at 450 ◦ C, the pre-adsorbed Ru3+ precursor decomposes to RuO2 from which certain fraction enters into the TiO2 lattice. This result could confirm the substitution of Ru cations for ˚ is very close to that Ti4+ because the ionic radius of Ru4+ (0.76 A) ˚ [27]. The calculated crystallite sizes of the anatase of Ti4+ (0.75 A) phase range from ∼13 to 16 nm (Table 1). Fig. 3 shows nitrogen adsorption–desorption isotherms and Barrett–Joyner–Halenda (BJH) pore-size distribution plots of the TiO2 , RT, Au/RT, Pd/RT and Pt/RT samples. The samples show similar type-IV isotherms, which are representative of mesoporous solids [21]. A very sharp pore size distribution at a mean value was calculated from the adsorption branch on the basis of BJH Model. They display similar nitrogen sorption isotherms with narrow pore-size distribution centered at 6.8–7.3 nm. As no systematic change in either the shape of the sorption isotherms or in the specific surface

4. Results and discussions Small angle X-ray scattering (SAXS) patterns of as-made RT and TiO2 , RT, Au/RT, Pd/RT and Pt/RT samples are shown in Fig. 1. The as-made sample show one well-resolved peak with a shoulder, which can be indexed as the (1 0) and (2 0) reflections of a two-dimensional hexagonal phase confirming an ordered mesostructure of P6 m space group [20]. The diffraction peak disappears when the treatment temperature is fixed at 450 ◦ C, showing that the long-range ordering is lost. The unit cell parameter calculated from SAXS analysis are 14.98 and 13.37 nm for as-synthesized and calcined samples, respectively. Calcination of the samples at 450 ◦ C gives rise to relatively large cell shrinkage (as much as 10%) and broadening of the diffraction peak. Wide angle X-ray diffraction (WAXRD) patterns in Fig. 2 shows TiO2 , RT, Au/RT, Pd/RT and Pt/RT samples. In order to confirm the bulk composition for each sample, the XRD patterns were compared with the JCPDS-ICDD standards for anatase (21-1272). The diffractograms for TiO2 are essentially equivalent, exhibiting peaks at 25.4◦ , 36.4◦ , 48.1◦ , 54.2◦ and 62.8◦ that are consistent with the (1 0 1), (0 0 4), (2 0 0), (2 1 1) and (2 1 3) planes associated with tetragonal anatase. The Rietveld refinements of the diffraction patterns show that only anatase was

101 200

004

(e)

211

213

(d) (c)

(b)

(a) 20

25

30

35

40

45

50

55

60

65

70

2θ (degree) Fig. 2. WAXRD patterns of TiO2 (a), RuO2 -TiO2 (b), Au/RuO2 -TiO2 (c), Pd/RuO2 -TiO2 (d), Pt/RuO2 -TiO2 (e) calcined at 450 ◦ C for 4 h.

A.A. Ismail et al. / Applied Catalysis A: General 431–432 (2012) 62–68

65

Table 1 The textural properties of mesoporous TiO2 and RuO2 -TiO2 and noble metals/RuO2 -TiO2 nanoparticles and their photocatalytic properties (initial reaction rates and photonic efficiencies of 30 mM methanol photooxidation). Photocatalysts

SBET (m2 g−1 )

PS

RuO2 -TiO2 as-made Meso- TiO2 RuO2 -TiO2 Au/RuO2 -TiO2 Pd/RuO2 -TiO2 Pt/RuO2 -TiO2 TiO2 Degussa

– 180 165 153 158 157 50.0

– 12.6 16.3 13.1 14.0 13.1 25.0

TiO2

(nm)

r × 107 (mol L−1 s−1 )

Eg (eV)

d100 (nm)

– 4.61 5.83 3.98 7. 3 6.25 3.08

– 3.11 2.86 2.95 3.01 2.94 3.20

12.96 11.34 11.57 11.67 11.67 11.60 –

Unit cell size (nm) 14.98 13.10 13.37 13.46 13.48 13.48 –

Pore wall (nm)

Vp (cm3 /g)

Dp (nm)

– 6.28 6.07 6.45 6.18 6.20 –

– 0.291 0.280 0.271 0.263 0.267 –

– 6.82 7.30 7.01 7.30 7.28 –

SBET surface area; PS TiO2 average particle size of TiO2 nanoparticle calculated from the peak of TiO2 (1 0 1) at 2 of 25.2◦ using the Lorentzian equation; r HCHO formation rate in the first 60 min;  photonic efficiency; Vp pore volume; Dp pore diameter; Eg bandgap.

area was found, the incorporation of the 0.5 wt% RuO2 nanoparticles does not appear to destroy the TiO2 mesoporous structure and to alter the sorption properties of small molecules such as N2 . In addition, the pore volume and pore diameter of RT and after photodepositing of noble metals particles Au/RT, Pd/RT and Pt/RT are almost similar ∼0.26 cm3 /g and 7 nm respectively. The surface area of undoped TiO2 (180 m2 /g) decreased slightly after synthesis of 0.5 wt% RuO2 -TiO2 (165 m2 /g), which suggested the incorporation of RuO2 nanoparticles into TiO2 lattice. TEM images of mesoporous RuO2 -TiO2 nanocomposites calcined at 450 ◦ C for 4 h are presented in Fig. 4. RuO2 -TiO2 mainly consisted of nanoscale cubes and rhombohedra, as demonstrated by the TEM image in Fig. 4a. These anatase TiO2 nanoparticles have an average size of 12 nm. Fig. 4b shows the HR-TEM images of Pd/RuO2 -TiO2 nanocomposities with randomly mesoporous channels after calcination at 450 ◦ C. Fig. 4c shows the HRTEM image of the resulting anatase TiO2 nanoparticles, one can clearly observe a 0.35 nm aligned anatase phase grown along [1 0 1] direction. The Pt, Pd and Au nanoparticles can be more easily found in dark field and bright filed TEM images (Fig. 4d–f) showing that the average Pt, Pd and Au particles diameters are 5, 15 and 50 nm respectively. RuO2 nanoparticles could not be detected at all even either HRTEM or in dark field in the pore channels and on the outer surface, implying the extremely high dispersion of the RuO2 particles, which was consistent with the above conclusion from XRD patterns. The EDS spectrum of RuO2 -TiO2 (Supporting information S1) shows that these nanocomposites are elementally composed of Ti, Ru and O, with the Cu peaks originating from the TEM grid. A detailed energy 250

Volume Adsorbed/ cc/g

200

Pore Volume/ cc/g

0.008

150

0.006

0.004

0.002

0.000 5

10

15

20

25

Pore Diameter (nm)

100

RuO2-TiO2 Au/RuO2-TiO2

4.1. Photocatalytic investigations

Pd/RuO2-TiO2 Pt/RuO2-TiO2

50

Undoped TiO22

0.0

0.2

0.4

0.6

0.8

dispersive X-ray element mapping measurement of mesoporous RuO2 -TiO2 shows uniform intensity of Ru and Ti signals throughout the particles, revealing the homogeneous distribution of Ru and Ti components within the inorganic framework. UV–visible spectra of TiO2 , RT, Au/RT, Pd/RT and Pt/RT are shown in Fig. 5. We noted that the positions of the absorption peaks of TiO2 sample suggests that it is wide band gap semiconductors, a conclusion which is consistent with previous reports [16,28]. On other hand, the Ru4+ ions-doped TiO2 show photoabsorption in the visible region, which gradually decreases with the wavelength. The results indicate that Ru4+ ions doped TiO2 show shifts in the absorption band toward visible light regions. This result suggests that Ru ions are effectively incorporated into the lattice of TiO2 by the calcination at low temperatures. It is obvious from Fig. 5 that doping 0.5% RuO2 site narrowed the band gap of TiO2 and that the absorption edges shifted to a longer wavelength region and the band gap values are varied according the noble metals deposited. The bandgaps of the TiO2 and RT, Au-RT, Pd-RT and Pt-RT can be estimated from the tangent lines in the plots of the square root of the Kubelka–Munk functions against the photon energy, as shown in Fig. 5. The tangent lines, which are extrapolated to ␣1/2 = 0, indicate the band gaps of 3.11, 2.86, 2.95, 3.01 and 2.94 eV for respectively. The possible explanation for the less-than-band gap absorption of the Ru-doped TiO2 particles; the formation of defects in the TiO2 lattice by the inclusion of Ru ions in the lattice, because the content of Ru ions very low. FT-IR data represent the characteristic absorbance of TiO2 , RT, Au/RT, Pd/RT and Pt/RT samples (See supporting information S2). For TiO2 sample, a broad absorption centered around 3400 cm−1 , and a weak sharp absorption band at about 1626 cm−1 . The absorption located at 3400 cm−1 characterizes the hydroxyl groups of Ti OH at weak surface active sites with which physisorbed water molecules are bound by weak hydrogen bonds with OH¯ groups of TiO2 surfaces [29]. On the other hand, the absorption bands associated with the isolated OH¯ group vibrations were not observed around 3665 and 3715 cm−1 , which indicates a high degree of hydration on the anatase surfaces [17a]. These results indicate that for TiO2 , OH¯ species do not occupy the lattice sites. The weak absorption at 1626 cm−1 is associated with the deformation vibration for H O H bonds of the physisorbed water and Ti OH bonds. The peaks at 910 cm−1 are contributions from the anatase phase and can be attributed to RuO2 [4a] [30].

1.0

Relative Pressure (P/Po) Fig. 3. N2 sorption isotherms and inset pore size distributions of the mesoporous of undoped TiO2 , RuO2 -TiO2 , Au/RuO2 -TiO2 , Pd/RuO2 -TiO2 , and Pt/RuO2 -TiO2 calcined at 450 ◦ C for 4 h.

The photocatalytic activities of all samples were evaluated for the photooxidation of CH3 OH to HCHO. Obviously, the production of HCHO should be proportional to that of the • OH radical. Therefore, the photonic efficiency () of HCHO can be used as an indicator of • OH production, and hence can be used to compare different photocatalysts [31]. Fig. 6 shows the photonic efficiencies of CH3 OH oxidation for noble metals (Au, Pd and Pt)/RuO2 -TiO2 compared

66

A.A. Ismail et al. / Applied Catalysis A: General 431–432 (2012) 62–68

Fig. 4. TEM images of mesoporous RuO2 -TiO2 nanocomposites calcined at 450 ◦ C for 4 h at high magnification and low demonstrates that the TiO2 nanoparticles with an average diameter of about 12 nm particles are quite uniform in size and shape (a and b). HRTEM image Pd/RuO2 -TiO2 nanocomposites observed 0.35 nm aligned anatase phase grown along [1 0 1] direction (c). Dark and bright-field TEM image of mesoporous Pt/RuO2 -TiO2 (d) Pd/RuO2 -TiO2 (e) and Au/RuO2 -TiO2 (f).

with Aeroxide Degussa P-25 under UV. The results reveal that the  of TiO2 decreases with RuO2 addition from 12.6 to 5.83% respectively. Subsequently, the  sharply increases to 14.1, 18.1 and 15.8% by adding Au, Pd and Pt nanoparticles, respectively. It is clearly seen that in case of Au/RuO2 -TiO2 and Pt/RuO2 -TiO2 nanocomposites, there was slightly improvement of  of RuO2 -TiO2 in comparison with undoped TiO2 (see by details our published paper in Ref.

[4a]). At the same time, Pd/RuO2 -TiO2 nanocomposites are highly improved the  (18.1%) of RuO2 -TiO2 which was greater 3 times than that for Degussa P-25 (4.62%) (Fig. 6). The reason why the  of RuO2 -TiO2 photocatalyst slightly increased by using both Au/RuO2 TiO2 and Pt/RuO2 -TiO2 nanocomposites and highly improved by Pd/RuO2 -TiO2 . As you see textures properties of all newly photocatalysts (Table 1), they have almost similar pore volume and

8 1.6

1.2

Absorbance

5

1.0

18 16

0.8

% Photonic efficiency

[F(R)hν]

1/2

6

20

TiO2 RT Au-RT Pd-RT Pt-RT

1.4

7

0.6 0.4

4

0.2

3

300

350

400

450

500

550

600

650

700

wavelength/nm

2

14 12 10 8 6 4

1

2 0 1.8

2.0

2.2

2.4

2.6

2.8

3.0

3.2

3.4

3.6

3.8

4.0

hν(eV) Fig. 5. UV–vis Kubelka–Munk transformed diffuse reflectance spectra of the the mesoporous of TiO2 , RuO2 -TiO2 , Au/RuO2 -TiO2 , Pd/RuO2 -TiO2 and Pt/RuO2 -TiO2 calcined at 450 ◦ C for 4 h. Inset Absorbance spectra obtained by a UV–visible spectrophotometer for the samples. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

0

P-25

TiO2

RT

Au/RT

Pd/RT

Pt/RT

Fig. 6. Dependence of photonic efficiency of mesoporous TiO2 , RuO2 -TiO2 , Au/RuO2 TiO2 , Pd/RuO2 -TiO2 , Pt/RuO2 -TiO2 calcined at 450 ◦ C for 4 h as well as commercial P25 Degussa for HCHO production. Illumination time, 60 min, photocatalyst loading, 0.5 g/l; 30 mM aqueous CH3 OH (O2 − saturated, natural pH; T = 20 ◦ C); reaction volume, 75 ml; Io = 4.49 × 10−6 Einstein L−1 s−1 (ca. >320 nm).

A.A. Ismail et al. / Applied Catalysis A: General 431–432 (2012) 62–68

67

1.0

% Photonic efficiency

0.8

0.6

0.4

0.2

0.0 RT Scheme 1. Schematic illustration of the proposed mechanism for methanol photooxidation to explain the enhanced photonic efficiency of Pd/RuO2 -TiO2 nanocomposites as photocatalyst absorption of UV light by the semiconducting nanoparticle promotes an electron from the valence band to the conduction band.

pore diameter and also surface areas. Therefore, the higher photocatalytic activity of Pd/RuO2 -TiO2 cannot be explained by the difference in texture proprieties. Ru4+ located in the interstitial site could act as an effective electron scavenger to trap the CB electrons of TiO2 . It should be noted that the redox potential of Ru(IV)/Ru(III) of +0.49 V [32] is lower than that of the TiO2 conduction band of −0.5 V at pH 7 [33], hence the thermodynamic driving force for the photocatalytic reduction is high, i.e. Ru(IV) reduced to Ru(III) (See Scheme 1). At the same time, the conduction band electrons migrate through the RuO2 -TiO2 networks until they reach Pd nanoparticles where adsorbed molecular O2 is reduced to form O2 •− radicals. On the other side, the potential of CH3 OH oxidation by a hole to an electron donating species • CH2 OH is E0 (• CH2 OH/CH2 O) = −0.95 V [33] and also Ru(IV) could be oxidized to Ru(IV) (Scheme 1). As the higher energy difference between the valence band of TiO2 (E0 (ox)) =2.94 V) and CH3 OH oxidation, CH3 OH oxidation was carried out and the excited electrons were restricted to recombination due to positively charged hole utilization by CH3 OH. Therefore, the acceleration through the electron transfer catalysis induced by the Pd nanoparticles and Ru ions was increased in the yield of the photocatalytic CH3 OH oxidation. The second explanation, the increased photonic efficiency of Pd/RuO2 -TiO2 more than either Pt/RuO2 -TiO2 or Au/RuO2 -TiO2 can be attributed to the density of states in the vicinity of the Fermi level of noble metals. They are high and consist mainly of contributions from d electrons in case of Pd [34]. In contrast, the d band in Au lies much lower, and the density of states at the Fermi level is low. This is generally believed to be one of the reasons why the HCHO formation is kinetically hindered on Au. Also the work function of bulk Pd and Pt (5.6 and 5.7 eV for Pd(1 1 1) and Pt(1 1 1)) is significantly higher than that of Au (5.3 eV for Au(1 1 1)) [35]. By reference to work function measurements, the Fermi level of Pd has been fixed 0.2 eV higher than the Fermi level of Pt (−10.8 eV) [36]. Also, Cowley and Sze [37], showed that in the presence of sufficiently small interface states the barrier height should be a linear function of the metal work function. Therefore there was a transfer of photogenerated electrons to the Pd clusters and also, the charging of the Pd clusters also enhances the HCHO formation rate. Assuming a Schottky contact between the mesoporous RuO2 -TiO2 network and the noble metal particle, the Pd particles serve as active sites for the reduction of molecular O2 , on which the trapped photogenerated electrons are transferred to O2 producing O2 •− radicals. It should be noted that this latter process is really the “bottle-neck” in most photocatalytic transformations being the rate-determining step due to

Au/RT

Pd/RT

Pt/RT

Fig. 7. Photonic efficiencies of RuO2 -TiO2 , Au/RuO2 -TiO2 , Pd/RuO2 -TiO2 and Pt/RuO2 -TiO2 photocatalysts calcined at 450 ◦ C under visible light for methanol oxidation. Photocatalyst loading, 0.5 g/l; 30 mM aqueous CH3 OH (O2 − saturated, natural pH; T = 20 ◦ C); reaction volume, 75 ml; Io = 3.34 × 10−7 Einstein L−1 s−1 (ca. >420 nm).

its very small thermodynamic driving force. Thus its acceleration through the electron transfer catalysis induced by the Pd reveals an increase in the photocatalytic efficiency of CH3 OH oxidation (Fig. 6). To examine visible-light activities of RT, Au/RT, Pd/RT and Pt/RT samples, the samples were exposed to light at wavelengths >420 nm. Our pervious published paper showed that the addition of 0.5 wt% RuO2 to TiO2 improved the photonic efficiency of TiO2 to be 0.53% under visible light [4a]. The photonic efficiencies obtained of RT, Au/RT, Pd/RT and Pt/RT are shown in Fig. 7. It is clearly seen that the photonic efficiency of RuO2 -TiO2 sample has been improved two times by addition Au nanoparticles(NPs), whereas no activity was observed with either Pt/RuO2 -TiO2 or Pd/RuO2 -TiO2 . Therefore, the higher photocatalytic activity of Au/RuO2 -TiO2 can be explained by one of the most remarkably properties of Au (See Fig. 5, green curve) is the presence of a visible band around 550 nm denoted as surface plasmon band (SPB) that arises from the collective excitation of electrons confined in the metal NPs [11b,c] [38]. The possibility to use visible light to excite this SPB of Au NPs with the injection of electrons on the TiO2 conduction band could be possible. In a certain way, the photocatalytic mechanism of Au/RuO2 -TiO2 could resemble that of the general dye sensitization mechanism of TiO2 [39] except that Au NPs are much more robust than organic dyes or metal complexes and the energy of the electrons coming from Au are much lower in energy than typical electrons in excited states of organic molecules. However, considering the low reduction energy of the conduction TiO2 band [41] and that metal NPs can behave as semiconductors [41], photoinduced electron injection from Au NPs to TiO2 promoted by visible light could be a viable process. The larger formation and more spherical of Au NPs are probably responsible for the higher photoactivity of Au/RuO2 -TiO2 under visible light. Therefore, it should be possible to apply smart molecular engineering approaches to design photocatalytic systems exhibiting considerably higher photoactivity either UV and visible light than existing systems through an improvement of the photonic efficiency using cocatalyst nanoparticles Au, Pd and Pt on the mesoporous RuO2 -TiO2 array surface to help the redox process. 5. Conclusions Noble metals (Au, Pd and Pt)/have been photodeposited onto hexagonal mesoporous RuO2 -TiO2 nanocomposites. The newly synthesized photocatalysts have been evaluated by determination of the formation rate of HCHO generated by photooxidation of

68

A.A. Ismail et al. / Applied Catalysis A: General 431–432 (2012) 62–68

CH3 OH under UV and visible lights. Under UV light, the findings reveal that Pd/RuO2 -TiO2 offers an improvement in term photooxidation rate and photonic efficiencies over Pt/RuO2 -TiO2 and Au/RuO2 -TiO2 . Therefore, the acceleration through the electron transfer catalysis induced by the Pd NPs and Ru ions were highly enhanced the photocatalytic efficiency of TiO2 for CH3 OH oxidation under UV light. However, under visible light, the photocatalytic activity of mesoporous RuO2 -TiO2 containing Au nanoparticles towards CH3 OH oxidation is remarkable and the photonic efficiency of RuO2 -TiO2 has been improved two times. Pt and Pd nanoparticles could not be observed in any improvement in photocatalytic activity of RuO2 -TiO2 . Distribute cocatalyst nanoparticles Pd and Au on the mesoporous RuO2 -TiO2 array surface have been improved the redox process under UV and visible light, respectively. This could lead to an important breakthrough in photocatalytic research since the photonic efficiency still represents the greatest bottleneck for photocatalytic systems. Acknowledgment A. A. Ismail acknowledges the Alexander von Humboldt (AvH) Foundation for granting him a research fellowship. We thank Dr. M.M. Sharifi (Institute of Physical Chemistry and Electrochemistry, Leibniz University Hannover) for N2 adsorption/desorption measurements and L. Robben (Institute of Mineralogy, Leibniz Universität Hannover) for XRD measurements. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/ j.apcata.2012.04.024. References [1] M.R. Hoffmann, S.T. Martin, W.Y. Choi, D.W. Bahnemann, Chem. Rev. 95 (1995) 69–96. [2] H. Vidal, J. Kaspar, M. Pijolat, G. Colon, S. Bernal, A. Cordon, V. Perrichon, F. Fally, Appl. Catal. B 30 (2001) 75–85. [3] J. Georgieva, S. Armyanov, E. Valova, I. Poulios, S. Sotiropoulos, Electrochem. Commun. 9 (2007) 365–370. [4] (a) A.A. Ismail, L. Robben, D.W. Bahnemann, ChemPhysChem 12 (2011) 982–991; (b) W.L. Kostedt, A.A. Ismail, D.W. Mazyck, Ind. Eng. Chem. Res. 47 (2008) 1483–1487. [5] (a) J. Choi, H. Park, M.R. Hoffmann, J. Phys. Chem. C 114 (2010) 783–792; (b) H. Irie, Y. Maruyama, K. Hashimoto, J. Phys. Chem. C 111 (2007) 1847–1852. [6] (a) D.F. Wang, J.H. Ye, T. Kato, T. Kimura, J. Phys. Chem. B 110 (2006) 15824–15830; (b) G. Li, N.M. Dimitrijevic, L. Chen, T. Rajh, K.A. Gray, J. Phys. Chem. C 112 (2008) 19040–19044; (c) M.H.H. Mahmoud, A.A. Ismail, M.M.S. Sanad, Chem. Eng. J. 187 (2012) 96–103. [7] (a) S.M. Sze, Physics of Semiconductor Devices, 2nd ed., Willey, New York, 1981; (b) R.T. Tung, Mater. Sci. Eng. R 35 (2001) 1–138. [8] A. Turut, M. Saglam, H. Efeoglu, N. Yalcln, M. Yildirim, B. Abay, Phys. B 205 (1995) 41–50. [9] (a) X. Wang, R.A. Caruso, J. Mater. Chem. 21 (2011) 20–28; (b) A.A. Ismail, D.W. Bahnemann, J. Phys. Chem. C 115 (2011) 5784–5791; (c) X. Li, J. Peng, J.-H. Kang, J.-H. Choy, M. Steinhart, W. Knoll, D.H. Kim, Soft Matter 4 (2008) 515–521.

[10] (a) X. Wang, J.C. Yu, H.Y. Yip, L. Wu, P.K. Wong, S.Y. Lai, Chem. Eur. J. 11 (2005) 2997–3004; (b) A.A. Ismail, D.W. Bahnemann, L. Robben, M. Wark, Chem. Mater. 22 (2010) 108–116; (c) A.A. Ismail, D.W. Bahnemann, J. Phys. Chem. C 115 (2011) 5784–5791. [11] (a) M. Srinvasn, T. White, Environ. Sci. Technol. 41 (2007) 4405–4409; (b) A.A. Ismail, D.W. Bahnemann, I. Bannat, M. Wark, J. Phys. Chem. C 113 (2009) 7429–7435; (c) M. Alvaro, B. Cojocaru, A.A. Ismail, N. Petrea, B. Ferrer, F.A. Harraz, V.I. Parvulescu, H. Garcia, Appl. Catal. B 99 (2010) 191–197. [12] (a) N. Lakshminarasimhan, E. Bae, W. Choi, J. Phys. Chem. C 111 (2007) 15244–15250; (b) A.A. Ismail, D.W. Bahnemann, Green Chem. 13 (2011) 428–435; (c) A.A. Ismail, Micropor. Mesopor. Mater. 149 (2012) 69–75. [13] (a) R. Liu, Y. Ren, Y. Shi, F. Zhang, L. Zhang, B. Tu, D. Zhao, Chem. Mater. 20 (2008) 1140–1146; (b) I. Sopyan, M. Watanabe, S. Murasawa, K. Hashimoto, A. Fujishima, J. Electr. Chem. 415 (1996) 183–186. [14] (a) A.A. Ismail, D.W. Bahnemann, J. Mater. Chem. 21 (2011) 11686–11707; (b) A.A. Ismail, Appl. Catal. B 117–118 (2012) 67–72. [15] (a) H. Shibata, T. Ogura, T. Mukai, T. Ohkubo, H. Sakai, M. Abe, J. Am. Chem. Soc. 127 (2005) 16396–16397; (b) A.A. Ismail, D.W. Bahnemann, ChemSusChem 3 (2010) 1057–1062. [16] J.C. Yu, J.G. Yu, W.K. Ho, Z.T. Jiang, L.Z. Zhang, Chem. Mater. 14 (2002) 3808–3816. [17] (a) C. Ducati, D.H. Dawson, J.R. Saffell, P.A. Midgley, Appl. Phys. Lett. 85 (2004) 5385–5387; (b) V. Houskova, V. Stengl, S. Bakardjieva, N. Murafa, V. Tyrpekl, Appl. Catal. B 89 (2009) 613–619. [18] (a) M. Senthilnanthan, D.P. Ho, S. Vigneswaran, H.H. Ngo, H.K. Shon, Sep. Purif. Technol. 75 (2010) 415–419; (b) T. Ohno, F. Tanigawa, K. Fujihara, S. Izumi, M. Matsumura, J. Photochem. Photobiol. A 127 (1999) 107–110. [19] F. Shiraishi, T. Nakasako, Z. Hua, J. Phys. Chem. A 107 (2003) 11072–11081. [20] J.F. Shannon, W. Boettcher, G.D. Stucky, Chem. Mater. 18 (2006) 6391–6396. [21] S.J. Gregg, K.S.W. Sing, Adsorption, Surface Area and Porosity, Academic Press, London, 1982. [22] M. Grätzel, Heterogeneous Photochemical Electron Transfer, CRC Press, Baton Rouge LA, 1988. [23] J. Tauc, R. Grigorovici, A. Vanuc, Phys. Stat. Sol. (1966) 15627–15637. [24] T. Nash, Biochem. J. 55 (1953) 416–421. [25] N. Serpone, R. Terzian, D. Lawless, P. Kennepohl, G. Sauve, J. Photochem. Photobiol. A 73 (1993) 11–16. [26] C.G. Hatchard, C.A. Parker, Proc. R. Soc. Ser. A 235 (1956) 518–536. [27] R.D. Shannon, Acta Crystallogr. A 32 (1976) 751–767. [28] H.Y. Zhu, Y. Lan, X.P. Gao, S.P. Ringer, Z.F. Zheng, D.Y. Song, J.C. Zhao, J. Am. Chem. Soc. 127 (2005) 6730–6736. [29] T. Bezrodna, G. Puchkovska, V. Shimanovska, I. Chashecnikova, T. Khalyavka, J. Baran, Appl. Surf. Sci. 214 (2003) 222–231. [30] (a) J.M. Yu, L.Z. Zhang, Z. Zheng, J.C. Zhao, Chem. Mater. 15 (2003) 2280–2286; (b) H.F. Hameka, J.O. Jensen, J.G. Kay, C.M. Rosenthal, G.L. Zimmermann, J. Mol. Spectrosc. 150 (1991) 218–221. [31] L. Sun, J.R. Bolton, J. Phys. Chem. 100 (1996) 4127–4134. [32] D.K. Atwood, T. De Vries, J. Am. Chem. Soc. 84 (1962) 2659–2661. [33] V.N. Nguyen, R. Amal, D. Beydoun, Chem. Eng. Sci. 58 (2003) 4429–4439. [34] C.G. Sanchez, E.P. Leiva, W. Schmickler, Electrochem. Commun. 5 (2003) 584–586. [35] R. C. Weast, M. J. Astle, CRC Handbook of Chemistry and Physics, 62 ed., CRC Press, Florida, 1981–1982. [36] P. Sautet, J.-F. Paul, Catal. Lett. 9 (1991) 245–260. [37] A.M. Cowley, S.M. Sze, J. Appl. Phys. 36 (1965) 3212. [38] W.L. Barnes, A. Dereux, T.W. Ebbesen, Nature 424 (2003) 824–830. [39] (a) D. Cahen, G. Hodes, M. Grätzel, J.F. Guillemoles, I. Riess, J. Phys. Chem. B 104 (2000) 2053–2059; (b) M.J. Grätzel, Photochem. Photobiol. C 4 (2003) 145–153. [41] (a) P.V. Kamat, J. Phys. Chem. B 106 (2002) 7729–7744; (b) V. Subramanian, E.E. Wolf, P.V. Kamat, J. Am. Chem. Soc. 126 (2004) 4943–4950; (c) A. Wood, M. Giersig, P. Mulvaney, J. Phys. Chem. B 105 (2001) 8810–8815.