Materials Letters 65 (2011) 966–969
Contents lists available at ScienceDirect
Materials Letters j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / m a t l e t
Photodeposition of Pt nanoparticles on TiO2–carbon xerogel composites Helder T. Gomes a,b, Bruno F. Machado a, Adrián M.T. Silva a, Goran Dražić c, Joaquim L. Faria a,⁎ a Laboratório de Catálise e Materiais (LCM), Laboratório Associado LSRE/LCM, Departamento de Engenharia Química, Faculdade de Engenharia, Universidade do Porto, Rua Dr. Roberto Frias s/n, 4200-465 Porto, Portugal b Departamento de Tecnologia Química e Biológica, Escola Superior de Tecnologia e Gestão, Instituto Politécnico de Bragança, Campus de Santa Apolónia, 5300-857 Bragança, Portugal c Department of Nanostructured Materials, Jozef Stefan Institute, Jamova 39, SI-1000 Ljubljana, Slovenia
a r t i c l e
i n f o
Article history: Received 16 September 2010 Accepted 16 December 2010 Available online xxxx Keywords: Carbon xerogel Titanium dioxide Platinum Photodeposition
a b s t r a c t Carbon xerogels synthesized from polycondensation of resorcinol with formaldehyde, having specific surface areas in the range 650 to 990 m2 g−1 and variable degrees of surface oxidation, are used to prepare TiO2– carbon xerogel composites by sol–gel methods. These composite materials are used to support Pt nanoparticles (5 wt.%) by the photodeposition technique. After a high temperature reduction treatment at 773 K, the obtained materials were characterized in order to assess the interactions between the phases Pt, TiO2 and carbon xerogel. It is observed that the carbon xerogel acts as an adhesive agent of the TiO2 and Pt particles, enhancing the interaction between the metal and the composite support. © 2010 Elsevier B.V. All rights reserved.
1. Introduction The photodeposition of noble metals over semiconductor materials is a simple and easy method of obtaining a well dispersed reduced metal phase [1]. If the support is a composite with a more elaborated structure and composition, advantages can be gained from the catalytic point of view, especially in what concerns synergistic interaction between phases. Since the first report by Pekala et al. [2], resorcinol–formaldehyde carbon xerogel (CX) materials have received considerable attention due to state of the art applications in energy generation fuel cells (using Pt and Ru metals) [3], advanced oxidation technologies [4], and fine chemical applications [5]. In this work, CX in different amounts and with varying surface chemistry are incorporated in a TiO2 matrix to produce different TiO2–CX composites on which Pt nanoparticles are photodeposited. This procedure has not been reported to our knowledge, but Pt/TiO2–carbon composites have known important applications in selective CO oxidation [6] and fuel cells [7], among others.
2. Experimental CX materials were prepared by polycondensation of resorcinol with formaldehyde (1:2), following the procedure described elsewhere [8]. Oxygen-containing functional groups were introduced on the surface of CX: (a) by gas-phase treatment using a 5% O2
⁎ Corresponding author. Tel.: + 351 225081645; fax: +351 225081449. E-mail address:
[email protected] (J.L. Faria). 0167-577X/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.matlet.2010.12.028
atmosphere, diluted in N2, at 673 K, with a burn-off of 44% (CXO2); and (b) by hydrothermal liquid-phase treatment with nitric acid 0.3 M at 413 K (CXNA) [8]. The preparation of TiO2/CX composites (TiO2–CX, TiO2–CXO2 and TiO2–CXNA) involved an acid catalyzed sol–gel method [9]. The weight fraction of carbon material in the composite was 1 or 5 wt.% (per ex. TiO 2 –1CX or TiO2 –5CX, respectively). In the case of Pt supported materials, 5 wt.% of Pt was supported on the TiO2/CX composites by photochemical deposition [10]. Pure TiO2 and Pt supported on pure TiO2 were also prepared for comparison purposes. Textural characterization was based on N2 adsorption-desorption isotherms, determined at 77 K with a Quantachrome NOVA 4200e multi-station apparatus [8]. Specific surface areas (SBET) were calculated for all samples. In the case of CX materials the micropore volumes (VMIC) and the non-microporous surface areas (mainly mesoporous, SMES) were also determined by the t-method. Temperature-Programmed Desorption (TPD) spectra were obtained with a fully automated Altamira AMI-200 instrument [8]. The peak assignment and deconvolution procedures described by Figueiredo et al. were applied to the TPD spectra [11,12]. TEM observations were performed on a JEOL JEM-2100, operating at 200 kV. SAED was performed on several particles in order to find their crystalline structure. Using HAADF in scanning transmission mode (STEM) Zcontrast images were collected on a JEOL JEM-2010 F microscope equipped with a field emission gun. XRD patterns were recorded in the 2θ range of 20–80° in steps of 0.017° on a Philips X'Pert MPD rotatory target diffractometer (CuKα radiation with λ = 0.15406 nm, 40 kV, 50 mA) in order to identify the crystallographic phases present and to calculate the crystallite size by using the modified Scherrer equation.
H.T. Gomes et al. / Materials Letters 65 (2011) 966–969
967
Table 1 Textural properties of the prepared CX materials and corresponding amounts of CO and CO2 released by TPD. Sample
SBET (± 10 m2 g−1)
SMES (± 10 m2 g−1)
VMIC (± 0.01 m3 g−1)
CO (± 20 μmol g−1)
CO2 (± 20 μmol g−1)
CX CXO2 CXNA
650 990 830
190 360 220
0.18 0.27 0.25
680 9420 7020
140 1740 2620
3. Results and discussion The texture and surface chemistry characterization of the prepared CX materials (CX, CXO2 and CXNA) is summarized in Table 1, together with concentration of CO and CO2 released by TPD. The results show that the hydrothermal treatment with nitric acid (CXNA samples) increases the porosity of the original CX material, increasing the specific surface area by 30%, and the mesoporous surface area and the microporous volume by ca. 15% and 40%, respectively. Gas-phase treatment with O2 (CXO2 samples) is found to significantly develop the specific surface area to 990 m2 g-1, corresponding to an increase of ca. 50% relatively to the original CX. The gas phase treatment increases both the mesoporous surface area and the micropore volume to a higher extent than that observed with the hydrothermal treatment, respectively 90% and 50% with comparison to the parent CX. The TPD (not shown) clearly marks the difference in the amount of surface groups attached to the CX surface before and after the treatments with O2 and nitric acid. Depending on the treatment, carbon materials often develop large amounts of surface groups, namely carboxylic acids (released by TPD at 573 K) and lactones (1000 K) evolved as CO2, whilst anhydrides (800 K), phenols (950 K), and carbonyl/quinones (1100 K) are detected as CO [11]. The SBET of the prepared TiO2 support was found to be 100 m2 g−1. The incorporation of 1 wt.% of carbon xerogel in the structure of TiO2,
regardless of the surface chemistry, resulted in a marginal 10% increase of the SBET (110 m2 g−1 for TiO2–1CXO2). Further incorporation of carbon xerogel (5 wt.%) provided an additional increase in the SBET of the composite (130 m2 g−1 for TiO2–5CXO2). These results are in good agreement with the calculated ratios obtained from the weight fraction of the composites (106 and 128 m2 g−1, respectively). HRTEM observations on pure TiO2 show that the sample is very uniform, both morphologically and in particle size, consisting of somewhat spherical particles with 8–10 nm in size (Fig. 1a). The SAED performed on selected areas on the sample show that TiO2 is present as anatase (Fig. 1b). XRD results confirmed the predominance of TiO2 crystallites in anatase form with a calculated TiO2 particle size of 9.7 nm, value well in agreement with the HRTEM results. HRTEM observations of the TiO2-CXO2 composites (Fig. 1c) show a thin layer of carbon surrounding the TiO2 particles (marked with an arrow), being concluded that when CX is incorporated on the TiO2 matrix, it acts as an adhesive agent to the TiO2 particles. The carbon appears in the form of amorphous carbon, while TiO2, with particle sizes around 8–10 nm, is found to be mainly present as anatase, as confirmed by SAED analysis (Fig. 1d). TEM analysis of the Pt metal particles supported on TiO2–CX composites and on pure TiO2 was performed in order to determine the Pt particle sizes and shapes which will help to understand how the Pt interacts with the support. Low magnification observations of the Pt/TiO2 material revealed the presence of a large quantity of Pt particles
Fig. 1. HRTEM micrograph of pure TiO2 (a) and respective SAED pattern (b); HRTEM of TiO2–1CXO2 (c) and respective SAED pattern (d).
968
H.T. Gomes et al. / Materials Letters 65 (2011) 966–969
Fig. 2. TEM micrograph of Pt/TiO2 at low magnification (a) and particle size distribution (b); TEM micrograph of Pt/TiO2 at high magnification (c) and microdiffraction pattern of a cubic Pt particle in [101] zone axis (d).
with sizes in the range 10–20 nm, well dispersed over a large TiO2 area (Fig. 2a). The distribution of the platinum particle sizes, as observed at this low magnification, is shown in Fig. 2b. In higher magnification observations, the presence of smaller spherical Pt particles (4–10 nm) was also confirmed, but in less amount than the large particles (Fig. 2c). SAED patterns recorded on the Pt particles reveal that Pt is in a cubic crystalline form (Fig. 2d). The mean size of the Pt particles was inferred by XRD measurements and found to be 17.2 nm. HRTEM observations of the Pt metal particles supported on the TiO2–CX composites, regardless of the CX used (CX, CXO2 or CXNA), show that amorphous carbon is a key component of the composite matrix acting as an adhesive agent to the TiO2 and Pt particles (Fig. 3). As observed in the TiO2–CX composites, a thin layer of amorphous carbon covers the TiO2 particles, independently from the surface chemistry of the carbon xerogel used. In the same fashion as found for Pt/TiO2, low magnification observations of the Pt/TiO2–1CX composite revealed the presence of a large quantity of Pt particles with sizes in the range 10–20 nm, well dispersed over the support (Fig. 3a). The distribution of the platinum particle sizes, as observed at this low magnification, is shown in Fig. 3b. Notwithstanding, at higher magnification observations, a higher quantity of Pt particles in the range of 4–5 nm are found, when compared with high magnification observations of the Pt/TiO2. This can be explained by the presence of the CX structure (and associated surface chemistry) in the composite, which favors the anchoring of the platinum complex during the deposition process [13], thus producing smaller Pt particles than those obtained in the pure TiO2 support. The mean size of the Pt particles in the Pt/TiO2–1CX composite was inferred by XRD measurements and found to be 15.9 nm. SAED patterns of TiO2 particles reveal the presence of TiO2 mainly in anatase form. SAED patterns of Pt particles confirm the cubic structure found previously in the Pt/TiO2 catalyst.
Regardless of the CX surface chemistry in the composite material, the Pt particles form strong interfaces with the composite supports (Fig. 3c and d). Comparing the samples Pt/TiO2–1CXO2 (Fig. 3c) and Pt/TiO2–5CXO2 (Fig. 3d), which only differ in the CX content, it is observed that the later has more amorphous carbon phase present than the former, as expected. The presence of a layer of amorphous carbon joining the platinum particles and the bulk TiO2 material is also seen in Fig. 3e and confirmed by EDXS (Fig. 3f and g). Bulk EDXS analysis on selected zones also confirms that the amorphous carbon acts as an adhesive agent between all the components of the composite material, which is mainly composed of TiO2 particles with platinum and carbon dispersed over the surface. 4. Conclusion Carbon xerogels synthesized from polycondensation of resorcinol with formaldehyde were successfully combined with TiO2 to produce composite materials by sol–gel methods and with 5 wt.% load of Pt impregnated by photodeposition. The carbon xerogel is found to act as an adhesive agent between the TiO2 and Pt particles, enhancing the metal/support interaction and reducing the size of the Pt particles. Acknowledgements The authors acknowledge the joint financial support from Fundação para a Ciência e a Tecnologia (FCT), Portugal and the Slovenian Ministry of Higher Education, Science and Technology. Additional support was provided by LSRE/LCM LA financing from “Programa de Financiamento Plurianual de Unidades de I&D/Laboratórios Associados” by FCT, and from the MEC-FEDER (CTQ200761324). A.M.T.S. acknowledges the financial support from POCI/N010/ 2006.
H.T. Gomes et al. / Materials Letters 65 (2011) 966–969
969
Fig. 3. TEM micrograph of Pt/TiO2–1CX at low magnification (a) and particle size distribution (b); HRTEM micrograph of Pt/TiO2–1CXO2 (c) and Pt/TiO2–5CXO2 (d); HAADF/STEM image (Z-contrast image) of Pt/TiO2–1CX focusing on a platinum particle covered with a thin layer of amorphous carbon (e) and EDXS analysis performed for Pt (f) and TiO2 (g).
References [1] Zhang F, Chen J, Zhang X, Gao W, Jin R, Guan N, et al. Langmuir 2004;20:9329–34. [2] Pekala RW, Alviso CT. Novel Forms Carbon 1992;270:3–14. [3] Arbizzani C, Beninati S, Manferrari E, Soavi F, Mastragostino M. J Power Sources 2007;172:578–86. [4] Gomes HT, Machado BF, Ribeiro A, Moreira I, Rosário M, Silva AMT, et al. J Hazard Mater 2008;159:420–6. [5] Machado BF, Gomes HT, Serp P, Kalck P, Figueiredo JL, Faria JL. Catal Today 2010;149:358–64. [6] Zheng T, Dong Y-R, Nishiyama N, Egashira Y, Ueyama K. Appl Catal, A 2006;308: 210–5.
[7] [8] [9] [10]
Hirakawa K, Inoue M, Abe T. Electrochim Acta 2010;55:5874–80. Silva AMT, Machado BF, Figueiredo JL, Faria JL. Carbon 2009;47:1670–9. Wang WD, Serp P, Kalck P, Silva CG, Faria JL. Mater Res Bull 2008;43:958–67. Silva AMT, Machado BF, Gomes HT, Figueiredo JL, Dražić G, Faria JL. J Nanopart Res 2010;12:121–33. [11] Figueiredo JL, Pereira MFR, Freitas MMA, Órfão JJM. Carbon 1999;37:1379–89. [12] Figueiredo JL, Pereira MFR, Freitas MMA, Órfão JJM. Ind Eng Chem Res 2007;46: 4110–5. [13] Solhy A, Machado BF, Beausoleil J, Kihn Y, Gonçalves F, Pereira MFR, et al. Carbon 2008;46:1194–207.
