Dispersed-Nanoparticle Loading Synthesis for Monodisperse Au

May 30, 2014 - *E-mail: [email protected]., *Tel: +81-22-795-7239, Fax: +81-22-795-7241, E-mail: [email protected]...
0 downloads 0 Views 1MB Size
Article pubs.acs.org/Langmuir

Dispersed-Nanoparticle Loading Synthesis for Monodisperse AuTitania Composite Particles and Their Crystallization for Highly Active UV and Visible Photocatalysts Takeshi Sakamoto,† Daisuke Nagao,*,† Masahiro Noba,† Haruyuki Ishii,† and Mikio Konno*,‡ †

Department of Chemical Engineering, ‡Graduate School of Engineering, Tohoku University, 6-6-07 Aoba Aramaki-aza, Aoba-ku, Sendai 980-8579, Japan S Supporting Information *

ABSTRACT: Submicrometer-sized amorphous titania spheres incorporating Au nanoparticles (NPs) were prepared in a one-pot synthesis consisting of a sol−gel reaction of titanium(IV) isopropoxide in the presence of chloroauric acid and a successive reduction with sodium borohydride in a mixed solvent of ethanol/acetonitrile. The synthesis was allowed to prepare monodisperse titania spheres that homogeneously incorporated Au NPs with sizes of ca. 7 nm. The Au NP-loaded titania spheres underwent different crystallization processes, including 500 °C calcination in air, high-temperature hydrothermal treatment (HHT), and/or low-temperature hydrothermal treatment (LHT). Photocatalytic experiments were conducted with the Au NP-loaded crystalline titania spheres under irradiation of UV and visible light. A combined process of LHT at 80 °C followed by calcination at 500 °C could effectively crystallize titania spheres maintaining the dispersion state of Au NPs, which led to photocatalytic activity higher than that of commercial P25 under UV irradiation. Under visible light irradiation, the Au NP-titania spheres prepared with a crystallization process of LHT at 80 °C for 6 h showed photocatalytic activity much higher than a commercial product of visible light photocatalyst. Structure analysis of the visible light photocatalysts indicates the importance of prevention of the Au NPs aggregation in the crystallization processes for enhancement of photocatalytic activity.



INTRODUCTION Removal of organic pollutants from water has become an issue of global concern. Photocatalytic degradation of organic pollutants is a promising method for water purification.1 Among various photocatalysts such as ZnO,2 SrTiO3,3 and WO3,4 titanium dioxide (TiO2) is regarded as the most efficient semiconductor photocatalyst, since it is environmentally benign and readily available in the water purification processes and also has strong oxidation power for organic pollutant degradation.5 Extensive effort has been devoted to expand the photoresponse of TiO2 to longer wavelength because most of the solar light corresponds to visible light. A variety of strategies, such as incorporations of metal ions,6,7 nonmetals,8 and other semiconductor components,9 have been extensively explored to enhance the photocatalytic activity of TiO2 under visible light irradiation. In addition to the substances described above, metal nanoparticles (NPs) are another candidate for the incorporation,10,11 which not only serves as an electron reservoir to localize the conduction band electrons generated under UV light irradiation,12 but also enhance the visible light absorption due to the plasmonic effect of metal NPs.13 Synthetic methods most commonly studied for preparation of metal-TiO2 composite particles are deposition of metal NPs on TiO2 surface.11,14,15 Although the synthetic method is © 2014 American Chemical Society

allowed to support various metal NPs precisely tuned on the TiO2 surface, it needs a multistep process of NP synthesis and supporting the NPs on the surface. Dissolution or desorption of metal NPs from the surface of TiO2 particles is also a possible problem in recycling processes required for maintaining the water purification efficiency. Modified methods to prevent the NP dissolution or desorption were reported by Yin et al.13 and Ye et al.10 who employed a multistep approach comprising preparation of monodisperse silica particles, decoration of the silica particles with Au NPs, and titania-coating of the Au-SiO2 composite particles. Although the modified methods have some advantages in both monodispersity of composite particles and no desorption of Au NPs from the composite particles, they require additional processes including centrifugations and sol− gel reactions to coat the Au-SiO2 composite particles. In the recycling processes for photocatalytic particles, size distributions of the particles dispersed in suspensions are another important factor in the sedimentation separation. Since the sedimentation velocity of particles strongly depends on Received: April 1, 2014 Revised: May 29, 2014 Published: May 30, 2014 7244

dx.doi.org/10.1021/la5012499 | Langmuir 2014, 30, 7244−7250

Langmuir

Article

were mainly performed to explore processes suitable for TiO2 crystallization without aggregation of Au NPs incorporated into the composite particles: (1) calcination in air at 500 °C for 3 h, (2) hydrothermal treatment at high temperature (HHT, 200 °C) in an autoclave reactor for 12 h, and (3) hydrothermal treatment at low temperature (LHT, 80 °C) in a glass container for 6−24 h. The 500 °C heat treatment (1) was chosen as a sintering temperature to increase the TiO2 crystalline size that is one of important factors for photocatalytic degradation of organic pollutants.31 The ramp rate in the calcination process was 2.8 °C·min−1. Temperatures of 200 and 80 °C were employed as severe and mild conditions for the hydrothermal treatments (2) and (3), respectively. Prior to the three crystallization processes, double centrifugations were applied to suspensions of the composite particles to remove residues. Experimental conditions for TiO2 crystallization are listed in Table 1.

their sizes in suspension, monodispersity of the sizes is essential to efficiently separate them without loss of the photocatalytic particles. Such a well-designed size distribution of photocatalytic particles will also have the possibility of increasing the light path length in TiO2, which is extensively studied these days.16,17 Preparations of monodisperse titania spheres were reported by some researchers who employed sol−gel reactions in the presence of electrolytes,18 alkylamines,19,20 or polymeric dispersants21 and sol−gel reactions in a mixed solvent of alcohol and acetonitrile.22,23 Hydrolysis of titanium glycolates was also applied to preparation of monodisperse titania spheres.24−27 These reactions allowed preparations of submicrometer-sized spherical titania particles with low polydispersity. In our report where spherical TiO2 particles were synthesized in a mixed solvent of ethanol and acetonitlile, it was indicated that calcination in a temperature range of 300−700 °C could crystallize the amorphous titania particles to anatase with maintenance of their monodispersity in sizes.22 On the base of the previous synthetic methods, we have developed a facile one-pot process to prepare UV- and visiblelight-responsive titania photocatalysts incorporating Au NPs without any desorption of Au NPs. The present method combines a sol−gel reaction of titanium(IV) isopropoxide with a successive reduction of chloroauric acid with sodium borohydride in the same solvent of an ethanol/acetonitrile mixture. For crystallization of the titania, various heat treatments were performed to explore effective ways to enhance the UV- and visible-photocatalytic activity of the titania particles while maintaining the dispersion state of Au NPs incorporated. This is the first report on providing not only a facile one-pot synthesis of monodisperse Au-TiO2 composite particles but also crystallization processes suitable for enhancing both UV- and visible photocatalytic activities with a single combination of Au and TiO2. It should be mentioned that most previous studies for the enhancement of TiO2 photocatalytic activity were focused on effects of doped species of metal ions,6 metal nanoparticles,28,29 and nonmetals.30



Table 1. Physical Properties of Photocatalysts Used for Degradation of Organic Dyes run

treatment for samples

1

Au-TiO2 particles without heat treatment Calcination 500 °C, 3 h HHT 200 °C, 12 h LHT 80 °C, 12 h LHT 80 °C, 6 h LHT 80 °C, 24 h LHT + 80 °C, 12 h + Calcination 500 °C, 3 h

2 3 4 5 6 7 P25

DAuNP [nm]

dXRD [nm] from TiO2(101)

7

amorphous

313

15−20 30−50 ≤10 ≤10 ≤15 ≤20

20 10.6 9.3 8.2 9.2 11.7

46 141 179 211 182 120



21.1

SBET [m2/g]

43.1

Characterization. Au-TiO2 composite particles with and without heat treatments were observed with a scanning transmission electron microscope (Hitachi, HD-2700) equipped with energy dispersive Xray spectroscopy (EDX, HORIBA, EMAX ENERGY) system. Volumeaveraged diameter, DV, and the coefficient of variation of particle size distribution, CV, were determined from the TEM and SEM images. Zeta potentials of Au-TiO2 composite particles were measured in water at 30 °C by electrophoretic light scattering (ELS, Otsuka Electronics, ELS-8000). The amounts of Au NPs or ions incorporated into the composite particles were measured with inductively coupled plasma atomic emission spectroscopy (ICP-AES, Seiko Instruments SPS7800). To prepare the solution for ICP-AES, the suspensions of the Au-TiO2 composite particles, before the NaBH4 reduction, were centrifuged and the sediments were dissolved with an aqueous solution of HCl. For the Au-TiO2 composite particles after the NaBH4 reduction, aqua regia was employed to dissolve Au NPs incorporated into composite particles. The crystallinities of Au-TiO2 composite particles obtained by the heat treatments were examined by X-ray diffraction (XRD, Rigaku, Ultima IV). Structural analysis of the Au-TiO2 composite particles was performed by N2 adsorption−desorption isotherms with BELSORPmini II (Bel Japan) to determine BET surface area of the composite particles. X-ray photoelectron spectroscopy (XPS) was also performed with AXIS ULTRA (KRATOS, Shimadzu) to examine chemical bonding states on surfaces of Au-TiO2 particles. Aqueous solution of methylene blue (MB) at 7.8 ppm was used for the measurements of photocatalytic activity under UV light irradiation with a xenon lamp (300 W, Asahi Spectra, MAX-303). The area of light irradiation was 12.6 cm2 and the distance between the light source and irradiation plane was 10 cm, resulting in a light irradiation of 0.48 mW·cm−2 at 355 nm. Before the light irradiation, a suspension of the Au-TiO2 composite particles at a concentration of 50 mg/L was stirred for an hour to sufficiently adsorb MB molecules on the composite particles. The concentration of MB at the 1 h adsorption is shown as C0 in Figure 4. A preparatory experiment in which an aqueous solution of MB was stirred in the presence of Au-TiO2

EXPERIMENTAL SECTION

Materials. Titanium(IV) isopropoxide (TTIP, 95%), ethanol (EtOH, 99.5%), acetonitrile (CH3CN, 99.5%), sodium borohydride, and chloroauric acid tetrahydrate were purchased from Wako Pure Chemical Industries (Osaka, Japan) and used as received for preparation of the Au-TiO2 composite particles. Dehydrated ethanol (< 10 ppm, Wako Pure Chemical Industries) was employed as a solvent when TTIP was dissolved in the preparation. Deionized water with resistivity higher than 18.2 MΩ·cm was used in all experiments. One-Pot Synthesis of Au-TiO2 Composite Particles. In a typical synthesis, TTIP dissolved in the dehydrated ethanol was added to a mixture of HAuCl4, H2O, EtOH, and CH3CN to initiate the hydrolysis and condensation of TTIP. The concentrations of TTIP, H2O, and HAuCl4 were 30 mM, 200 mM, and 0.07 mM, respectively, in the mixed solvent of EtOH and CH3CN (42 wt % CH3CN). The reaction was conducted at 30 °C for 18 h. Then, an ethanolic solution of NaBH4 was injected over a short time or continuously added to the titania suspension at 30 °C. The continuous addition was conducted for 5 min at a fixed feed rate of 2.88 μmol/min. The concentration of NaBH4 and reaction volume in the two addition ways were 0.36 mM and 40 mL, respectively, on the completion of the NaBH4 addition. The volumes of NaBH4 solution added in both processes were oneeighth as large as that of titania suspension. Crystallization of Au-TiO2 Composite Particles. The Au-TiO2 composite particles obtained in the continuous addition process were crystallized by different heat treatments. The following heat treatments 7245

dx.doi.org/10.1021/la5012499 | Langmuir 2014, 30, 7244−7250

Langmuir

Article

Figure 1. TEM images of particles formed in hydrolysis and condensation of titanium alkoxide at 0.07 mM HAuCl4 (A) and in the absence of HAuCl4 (B).

Figure 2. TEM images of TiO2 composite particles formed by different NaBH4 additions to suspensions of Au ion-loaded TiO2 particles. The images in the left and right sides are TEM and Z-contrast ones, respectively. The ion-loaded TiO2 were reduced by a continuous addition (A) or a one-shot addition of NaBH4 (B). The continuous addition was conducted for 5 min at a fixed feed rate of 2.88 μmol/min. composite particles without UV light irradiation confirmed that the adsorption of MB molecules on the particles was within 2% of the amount of MB molecules added to the suspension in degradation tests. To evaluate the photocatalytic activity under visible light irradiation, aqueous solution of methyl orange (MO) at 7.8 ppm was employed. The visible light in a wavelength range longer than 400 nm was supplied with a UV-filter system (Asahi Spectra, UV 400 nm φ25) from the same xenon lamp.

of the spherical particles formed were negative and their absolute values were increased with the HAuCl4 concentration (see Figure S1 in SI). The increase in the HAuCl 4 concentration decreased the size of particle formed. It was reported that titania particles formed in sol−gel reactions were stabilized with addition of electrolytes such as NaCl and HCl.18,32 Probably, the HAuCl4 added to the present reaction system also had a similar effect on stabilization of titania particles. ICP-AES measurement for Figure 1(A) showed that 28% of AuCl4− added to the reaction system was incorporated into titania particles, indicating that the rest of 72% AuCl4− was still dissolved in the solution to be reduced by NaBH4 in the next process. Different reduction methods with continuous and one-shot additions of NaBH4 were performed for the suspension of particles shown in Figure 1A. The total amounts of NaBH4 added to the suspension were the same in the two different methods of NaBH4 addition. In the continuous addition, Au



RESULTS AND DISCUSSION 1. Sol−Gel Synthesis of Monodisperse Titania Spheres Incorporating Gold Nanoparticles. Figure 1 shows TEM images of particles formed by sol−gel reaction of TTIP in the mixed solvent of ethanol and acetonitrile with and without addition of HAuCl4. Spherical monodisperse particles with a DV value of 431 nm and a CV value of 9.0% were formed in the presence of HAuCl4 (Figure 1A), whereas aggregation of submicrometer-sized spherical particles was observed in the absence of HAuCl4 (Figure 1B). Zeta potentials 7246

dx.doi.org/10.1021/la5012499 | Langmuir 2014, 30, 7244−7250

Langmuir

Article

Figure 3. EM images of Au-TiO2 composite particles obtained by the three different heat treatments. The TEM (A) and SEM (B) images for the composite particles were obtained by sintering at 500 °C for 3 h. The images of (C) and (D) were for hydrothermal treatment at 200 °C, and the images of (E) and (F) were for hydrothermal treatment at 80 °C.

2. Crystallization of Au-TiO2 Particles for UV and Visible Photocatalysts. EM images of Au-TiO2 particles obtained by the three different heat treatments described in the Experimental Section are shown in Figure 3. In the calcination process at 500 °C, as shown in Figure 3A,B, the sizes of Au NPs incorporated into TiO2 particles were increased to 15−20 nm, which might be caused by Ostwald ripening or coalescence of Au NPs at the high calcination temperature.33,34 Although the calcination process could crystallize the titania particles to anatase phase with a crystalline size of 20 nm estimated from the TiO2 (101) peak in XRD, it caused both a drastic decrease in BET surface area to 46 m2/g and a roughening in particle surface (see Figure 3B). In the HHT at 200 °C, much lower than the calcination temperature, unfortunately, Au NPs were more distinctly aggregated, as shown in TEM and its SEM images (Figure 3C,D) where the sizes of Au NPs were in the range of 30−50 nm. The high pressure in the autoclave (∼1.4 MPa) probably promoted Ostwald ripening or coalescence of AuNPs in the TiO2 particles. On the other hand, in the LHT at 80 °C and ambient pressure for 12 h, Au NPs incorporated were scarcely

NPs with an average size of 7 nm were uniformly incorporated into the titania particle shown in TEM and Z-contrast images of Figure 2A. In the one-shot addition (see Figure 2B), inhomogeneous incorporation of Au NPs into titania particles was observed. The highly uniform incorporation of Au NPs in the continuous addition was confirmed by elemental mappings with EDX (see Figure S2). ICP-AES measurement for the titania particles of Figure 2A indicated that 90 ± 5% of AuCl4− added to the synthetic system was incorporated into the particles. Characterizations with XRD and N2 adsorption revealed that the Au NP-TiO2 composite particles had no crystalline phase and BET surface area of 313 m2/g (see Figures S3-A and S4-A). A possible reason to explain the superiority of continuous addition over one-shot addition is homogeneity in NaBH4 concentration in the reduction system. Since the reduction rate to form AuNPs is probably much higher than diffusion of AuCl4− to be homogeneous in the solution, it is possible that Au NPs were preferentially precipitated in a region where NaBH4 solution was shot at a high concentration. Thus, the local AuNPs precipitation induced by the one-shot addition lowered uniformity in AuNP incorporation into titania spheres. 7247

dx.doi.org/10.1021/la5012499 | Langmuir 2014, 30, 7244−7250

Langmuir

Article

processes of HHT and LHT (Runs 3 and 4). Then, the additional heating process of calcination at 500 °C was applied to the Au-TiO2 particles obtained with the LHT. Figure 5 shows EM images of Au-TiO2 particles treated with the combined heating process of LHT and calcination. Although morphologies of Au-TiO2 particles obtained by the additional heating of calcination (Figure 5A) are similar to those without the additional process (Figure 3F), the surface of particles obtained seems to be slightly rougher than the one before the calcination. As seen from the titania crystalline sizes and BET surface areas between Run 4 and 7 in Table 1, the additional calcination following the LHT increased the crystalline size without a drastic decrease in BET surface area, whereas direct sintering without the LHT (Run 2) drastically decreased BET surface area during the titania crystallization. The maintenance of the Au NP dispersion state in the titania particles was clearly indicated by elemental mappings with EDX (see Figure S5), although local aggregation of AuNP was partially observed in the Au mapping. The photocatalytic activity of Au-TiO2 particles obtained by the combined heating process is shown in Figure 4. It was remarkable that the activity was highly enhanced to exceed the commercial P25 activity. As seen in Table 1, the combined process effectively prevented reduction in the BET surface area of Au-TiO2 particles. This was also confirmed with AuNP-free TiO2 particles that were treated with the combined process or the direct calcination. Since the present reaction condition of Figure 1 could not prepare AuNP-free TiO2 particles without their coagulation, the previous method22 was applied to the preparation of AuNP-free TiO2 particles. The BET surface area of AuNP-free TiO2 particles crystallized with the combined process was much larger than that with the direct calcination. The photocatalytic test with the AuNP-free TiO2 particles prepared with the combined process indicated that the particles had photocatalytic activity similar to that of P25 (Figure S6). Thus, the combined process is an effective way to prepare highly photocatalytic TiO2 particles under UV light irradiation. Since it is reported that MB molecules are directly photolyzed under visible light irradiation without photocatalyst,36 we employed a dye of methyl orange (MO)37−39 to examine photocatalytic activity under visible light irradiation. The MO degradation in the presence of particles obtained in the combined process is presented in Figure 6 together with that for the commercial P25. The Au-TiO2 particles for the combined process, however, had a considerably low activity under the visible light irradiation.

aggregated and could maintain their particle sizes (