Ru bimetallic nanoparticles in high-temperature and high-pressure fluids

Ru bimetallic nanoparticles in high-temperature and high-pressure fluids

Journal of Colloid and Interface Science 322 (2008) 358–363 www.elsevier.com/locate/jcis Note Synthesis of Pt/Ru bimetallic nanoparticles in high-te...

369KB Sizes 0 Downloads 24 Views

Journal of Colloid and Interface Science 322 (2008) 358–363 www.elsevier.com/locate/jcis

Note

Synthesis of Pt/Ru bimetallic nanoparticles in high-temperature and high-pressure fluids Masaki Ueji a , Masafumi Harada b , Yoshifumi Kimura a,∗ a Department of Chemistry, Graduate School of Science, Kyoto University, Kyoto 606-8502, Japan b Department of Health Science and Clothing Environment, Faculty of Human Life and Environment, Nara Women’s University, Nara 630-8506, Japan

Received 28 December 2007; accepted 27 February 2008 Available online 4 March 2008

Abstract A high-temperature and high-pressure flow-reactor system was applied to the synthesis of monometallic ruthenium (Ru) nanoparticles and platinum/ruthenium (Pt/Ru) bimetallic nanoparticles using the thermal reduction of ruthenium ion (Ru(III)) and the mixture of platinum (Pt(IV)) and ruthenium ions in water and ethanol mixture in the presence of poly(N-vinyl-2-pyrrolidone). Monometallic Ru nanoparticles with an average diameter of ca. 2 nm were synthesized above 200 ◦ C at 30 MPa. The monometallic Ru nanoparticles tended to make large aggregates in colloidal dispersions. By the reduction of the mixture solution of Pt(IV) and Ru(III) in water and ethanol above 200 ◦ C at 30 MPa, Pt/Ru bimetallic nanoparticles with an average diameter of ca. 2.5 nm were synthesized with relatively small size distribution. The EXAFS spectra for the Pt/Ru bimetallic particles indicated that the particle possesses metallic bonds between Pt and Ru atoms in contrast to the case of the nanoparticles produced by thermal reduction under ambient pressure at 100 ◦ C [M. Harada, N. Toshima, K. Yoshida, S. Isoda, J. Colloid Interface Sci. 283 (2005) 64], and that the Pt/Ru bimetallic particle has a Pt-core/Ru-shell structure. © 2008 Elsevier Inc. All rights reserved. Keywords: Platinum/ruthenium bimetallic nanoparticles; High-temperature and high-pressure synthesis; EXAFS

1. Introduction Noble metal nanoparticles have attracted interests both academically and industrially [1], because of their uniqueness of size-dependent optical, catalytic, magnetic, and electronic properties. A primary goal of the synthesis of nanometer-sized particles is the regulation of function and structure including their sizes and compositions, together with the establishment of the efficient and effective synthetic methods. Among many synthetic methods, chemical reduction methods in a liquid phase have been examined intensively in a laboratory scale because of its mass productivity and size controllability [2–5]. At the same times, the utilization of a wide variety of protective reagents such as polymers [2,4,6–9], mercaptans [10–14] and microemulsions [15], which are used to prevent nanoparticles produced in the liquid phase from aggregation, is an advantage * Corresponding author. Fax: +81 75 753 4000.

E-mail address: [email protected] (Y. Kimura). 0021-9797/$ – see front matter © 2008 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2008.02.056

of this method. For example, Toshima et al. successfully synthesized Pt-core/Pd-shell bimetallic particles which are under 2 nm in size covered with poly(N -vinyl-2-pyrrolidone) (PVP) by the reduction of metallic ions in alcoholic solution under ambient pressure [6]. On the other hand, the effective use of hightemperature and high-pressure fluids such as supercritical water (ca. 400 ◦ C) to synthesize small particles was demonstrated by Adschiri and co-workers [16]. By using a more moderate experimental condition near the critical temperature of ethanol (e.g., near 250 ◦ C), Kimura et al. recently demonstrated that the chemical reduction of platinum ion (Pt(IV)) under the existence of PVP in water and ethanol mixture could produce a colloidal dispersion of Pt nanoparticles of 2.6 ± 0.5 nm within a few seconds [17]. The same method has also been applied successfully to the case of rhodium (Rh) nanoparticles [18], palladium (Pd) nanoparticles, and Pt/Rh, Pt/Pd [19], and Au/Rh [20] bimetallic nanoparticles. The moderately high synthetic temperature (200 to 300 ◦ C) has merits of the rapid reduction rate and of keeping the stability of the polymer which prevents the aggregation [17,18]. The size and microstructure of nanoparticles could be

M. Ueji et al. / Journal of Colloid and Interface Science 322 (2008) 358–363

controlled by the solvent species and the reaction temperature of the synthesis. In this work, the high-temperature and high-pressure synthetic method has been applied to the production of monometallic ruthenium (Ru) nanoparticles and bimetallic nanoparticles of Pt and Ru. The combination of Pt and Ru is expected to give effective catalytic properties in activities for polymer electrolyte fuel cell [21–24]. Previously the structure of Pt/Ru bimetallic nanoparticles produced by the reflux of Pt(IV) and Ru(III) mixture solution of water and ethanol under normal pressure at 100 ◦ C has been analyzed [25]. According to this analysis, the nanoparticles produced by the reduction were not alloyed and composed of segregated particles below 6 nm containing monometallic Pt and Ru nanoparticles. On the other hand, Shimazaki et al. have reported that Pt/Ru nanoparticles (ca. 2 nm) produced by the chemical reduction using sodium borohydride in the presence of a capping reagent such as citric acid were alloyed according to the analysis of the XRD, although the detail structure of the alloyed Pt/Ru nanoparticles was not presented [26]. Here, we demonstrate that Pt/Ru bimetallic nanoparticles have been successfully synthesized using the high-temperature and high-pressure synthetic method, and that the bimetallic nanoparticles is found to have a core– shell structure based on the EXAFS analysis. 2. Experimental

359

centration of PVP were adjusted to be 30 mM and 30 g dm−3 , respectively. The detail experimental setups are given in supplementary information (Fig. S1 in Supplementary material). 2.3. Characterization of the colloidal dispersions of metal nanoparticles The size of metal nanoparticles was estimated from TEM images taken by JEM-2000FX (JEOL, acceleration voltage: 200 kV). The histogram of the particle size distribution was obtained by measuring about 300 particles in arbitrarily chosen areas on the enlarged photograph. The EXAFS measurements of Pt-LIII edge and Ru-K edge of colloidal dispersions were performed at Photon Factory, High Energy Accelerator Research Organization (KEK-PF), using BL-10B, BL-7C, and BL-9C stations. The details of the EXAFS measurement are given elsewhere [2,25,28,29]. EXAFS data were analyzed by REX2000 vers. 2.0.7 program (Rigaku Co.). The details of the analysis are given in supplementary information. The coordination number and the bond distance were obtained by the fit of the EXAFS oscillation χ(k) to the following function [30]:    k 3 χ(k) = Nj Fj (kj )kj2 exp −2σj2 kj2 j

  × sin 2kj rj + φj (kj ) rj2 ,

2.1. Materials

kj = (k 2 − 2mE0j /h¯ 2 )1/2 ,

Hexachloroplatinate (IV) hexahydrate (Nacalai Tesque, Guaranteed Reagent) and ruthenium chloride (III) n-hydrate (Nacalai Tesque, Guaranteed Reagent) were used as received. Poly(N -vinyl-2-pyrrolidone) (PVP), whose average molecular weight was about 40,000, was purchased from Nacalai Tesque and used as received. Ethanol (Nacalai Tesque, Guaranteed Reagent, 99.5%) was used without further purification. Water purified by a Milli-Q system (>10 M cm, Millipore, Milli-Q SP UF) was used.

where k = [2m(E − E0 )/h¯ 2 ]1/2 , E and E0 are the X-ray beam energy and the threshold energy, respectively. Nj denotes the coordination number (CN), rj the bond distance, E0j the difference between theoretical and experimental threshold energies, and σj the Debye–Waller factor of the j th coordination shell, respectively. In the fitting, we employed empirically derived phase shifts (φj (k)) and amplitude functions (Fj (k)) evaluated from the reference samples Pt–Pt (Pt foil), Pt–Ru (Pt/Ru 1/9 alloyed powder), Ru–Ru (Ru powder) and Ru–Pt (Ru/Pt 1/9 alloyed foil), where X–Y denotes a pair of absorbing atom and scattering atom, by assuming that the scattering from Y atoms contributes the observed EXAFS spectrum of each reference sample [2,6].

2.2. Synthetic procedure of the colloidal dispersions of metal nanoparticles The reaction system used here was similar to the flow reactor system used for the hydrothermal reaction [16] and is described elsewhere [27]. Briefly, the thermal reduction was performed by mixing the preheated solvent (ca. 400 ◦ C) and the solution containing reactant ions and PVP in a high-pressure and hightemperature reactor whose temperature was regulated within ±2 ◦ C. The solvent and the solution were flowed by two different HPLC pumps (JASCO Corp.). The mixed solution after the reactor was immediately cooled by mixing solvent using another HPLC pump. The flow rates of the preheated solvent and the sample solution were typically 1.0 and 0.5 cm3 min−1 , respectively, and the reaction time was less than a few seconds. Synthesizes were performed using the 1/1 volume mixture of water and ethanol at 30 MPa and 200, 260, and 300 ◦ C, respectively. The total concentration of metal ions and the con-

3. Results and discussion 3.1. Structure of monometallic Ru nanoparticles Before presenting the results of the Pt/Ru colloidal dispersions, we briefly mention the results of the monometallic Ru nanoparticles. In the synthesis, the flow rates of the samples and preheated solvents were adjusted to be 0.5 and 1.5 cm3 min−1 , respectively. At all the temperatures studied here, we got black-brown solutions, which were stable at room temperature for more than a few months. Fig. 1a shows a TEM image of colloidal dispersions of monometallic Ru nanoparticles synthesized at 30 MPa and 300 ◦ C. It is found that the

360

M. Ueji et al. / Journal of Colloid and Interface Science 322 (2008) 358–363

(a)

(b)

(c) Fig. 1. TEM images and histograms of the size distributions of colloidal dispersions of (a) monometallic Ru nanoparticles produced by the reduction of Ru(III) salt in water and ethanol 1/1 mixture with PVP at 30 MPa and 300 ◦ C, and (b, c) Pt/Ru(1/1) bimetallic nanoparticles produced by the reduction of both Pt(IV) and Ru(III) salts in water and ethanol 1/1 mixture with PVP at 30 MPa and 200 ◦ C (b) and 300 ◦ C (c).

particles tend to form large aggregates. The estimated average particle sizes for the samples synthesized at different temperatures (200, 260, and 300 ◦ C) are around or less than 2.0 nm (see Table 1), which is smaller than the size of monometallic Pt [19] and Rh [20] nanoparticles produced in a similar way. The particles of these samples are confirmed to be metallic and not oxidized by the EXAFS, and the details are given in supplementary information (Fig. S2 in Supplementary material).

3.2. Structure of Pt/Ru bimetallic nanoparticles Synthesis of the Pt/Ru bimetallic nanoparticles was performed at the flow rates of the sample and preheated solvents of 1.0 and 1.0 cm3 min−1 , respectively. At all the temperatures studied here, we got similar black-brown solutions as in the cases of colloidal dispersions of monometallic Ru nanoparticles. Figs. 1b and 1c show the TEM images of the colloidal dispersions of Pt/Ru(1/1) bimetallic nanoparticles produced at

M. Ueji et al. / Journal of Colloid and Interface Science 322 (2008) 358–363

361

Table 1 Average particle diameters evaluated by the TEM for monometallic Ru and Pt/Ru(1/1) bimetallic nanoparticles, and the results of EXAFS analysis and the expected coordination numbers of model structures T (◦ C)

[Pt]/[Ru]a,b

dc

CN

Pt-LIII RPt–Pt e

200 260 300

0/1 0/1 0/1

1.7 ± 0.5 2.0 ± 0.5 2.3 ± 0.6

6.9 ± 2.0 7.8 ± 2.0

200 260 300

1/1 1/1 1/1

2.5 ± 0.8 2.8 ± 0.6 2.5 ± 0.7

8.5d 8.7 d 9.3 d

Random Pt-core/Ru-shell Cluster-in-cluster

9.4f

2.74 2.74 2.75

Ru-K NPt–Pt

5.8 ± 1.6 5.9 ± 2.2 7.3 ± 2.5 4.7 8.4 7.3

RPt–Ru e

2.72 2.71 2.73

NPt–Ru

3.0 ± 1.0 3.7 ± 1.4 3.6 ± 1.2 4.7 3.1 2.6

RRu–Ru e

NRu–Ru

2.67 2.66

6.9 ± 2.0 7.8 ± 2.0

2.66 2.65 2.66

3.2 ± 1.2 3.0 ± 1.2 2.5 ± 0.8 4.7 4.1 6.3

RRu–Pt e

NRu–Pt

2.68 2.69 2.72

4.9 ± 1.5 4.8 ± 2.0 5.2 ± 2.0 4.7 3.2 2.6

Note. RX–Y means the bond distance between X and Y atoms. N X–Y means the coordination number (CN) of Y atoms around X atom. The error bars of the coordination number were estimated by the change when the values of Debye–Waller factor (fixed to 0.065 Å in the fitting) were varied by ±0.01. a The mole ratio of Pt to Ru in the ionic solution as source. b [Pt] + [Ru] = 30 mM, [PVP] = 30.0 g dm−1 . c The average diameter d (Å) and its standard deviation of nanoparticles estimated from the corresponding TEM images. d Average coordination numbers of the Pt/Ru(1/1) bimetallic nanoparticle. CN = (1/2)(N Pt–Pt + NPt–Ru ) + (1/2)(NRu–Ru + NRu–Pt ). e The estimated error bar in the R values is ±0.03 Å. f The values of CN for the random, Pt-core/Ru-shell, and cluster-in-cluster are for N = 9.4 [7].

200 and 300 ◦ C, respectively. The average particle sizes are around 2.5 nm (as shown in Table 1), which is slightly larger than the case of monometallic Ru nanoparticles. However, in the present case, these nanoparticles are well dispersed in contrast to the case of the monometallic Ru nanoparticles, as shown in Fig. 1a. Figs. 2a and 2c show Fourier transforms of EXAFS oscillations k 3 χ(k) on both Pt-LIII and Ru-K edges for the colloidal dispersions of Pt/Ru(1/1) bimetallic nanoparticles produced at different temperatures, respectively. Fourier transforms of Pt foil and Pt/Ru(1/9) alloyed powder on Pt-LIII edge and of Ru powder and Ru/Pt(1/9) alloyed foil on Ru-K edge are also shown as references in Figs. 2b and 2d, respectively. For the Pt-LIII edge, the spectra of the Pt/Ru colloidal dispersions show double peaks of nearly equal intensities at the R range between 2 and 3 Å, while the spectrum due to the Pt–Ru bond (attributed from Pt/Ru(1/9) alloyed powder) shows only a single peak and the spectrum due to the Pt–Pt bond (attributed from Pt foil) shows double peaks with different peak intensities at the R range between 2 and 3 Å. The spectra of the colloidal dispersions of Pt/Ru(1/1) bimetallic nanoparticles strongly suggest that both the Pt–Pt and the Pt–Ru bonds exist in Pt/Ru nanoparticles. Almost the same story is true to the case of Ru-K edge for the colloidal dispersions of Pt/Ru(1/1) bimetallic nanoparticles. The present results indicate that there are Pt–Ru bonds in individual nanoparticles in contrast to the case of nanoparticles synthesized under ambient pressure at 100 ◦ C [25]. Table 1 summarizes the results of TEM and EXAFS analysis. In order to estimate an appropriate structure of the Pt/Ru (1/1) bimetallic nanoparticle, we compare our results with three model structures which have similar total coordination number (N = 9.4) given in Ref. [7] for the Au/Pd(1/1) bimetallic clusters. In these models, the bimetallic nanoparticles consist of the centered six atoms and three layered atoms with fcc structure around six atoms, totally containing 248 atoms. The occupa-

tion of Au and Pd atoms in the particle is different from one another in three models: Au-core/Pd-shell model, cluster-incluster model, and random model. The correspondence to the present Pt/Ru system should be taken if we read Au as Pt and Pd as Ru. The average CNs of scattered components (Pt and Ru) around each absorbing atom are shown in Table 1. Comparing the average CNs (NPt–Pt , NPt–Ru , NRu–Ru , and NRu–Pt ) calculated for three models with the values estimated by the EXAFS analysis, the Pt/Ru(1/1) bimetallic nanoparticle produced here is found to have the structure mostly like the Pt-core/Ru-shell structure, although the contribution of NRu–Pt is slightly large regardless of synthetic temperatures at which the colloidal dispersions were formed. It is an interesting issue that under the high-temperature and high-pressure condition Pt-core/Ru-shell nanoparticles are produced in contrast to the case of the nanoparticles synthesized under the ambient pressure at 100 ◦ C where segregated nanoparticles were produced [25]. The most significant difference between these two methods is the reaction time. At the high temperature above 200 ◦ C, the reduction of the metal ions proceeds very rapidly within a few seconds, which might be related to the formation of bimetallic nanoparticle. In fact, the reduction by the sodium borohydride [26], which can produce the alloyed Pt/Ru particles, also proceeds very rapid in comparison with the thermal reduction in alcohols under normal pressure at 100 ◦ C (more than hours). The Pt-core/Ru-shell structure of the bimetallic particles is similar to the cases of the bimetallic particles of Pt/Rh [2] and Pt/Pd [6] produced by the thermal reduction in alcoholic solution under ambient pressure. 4. Conclusion In summary, well-dispersed Pt/Ru bimetallic nanoparticles were produced for the case of the Pt/Ru(1/1) colloidal dispersions, and the average particle diameter was about 2.5 nm. The

362

M. Ueji et al. / Journal of Colloid and Interface Science 322 (2008) 358–363

Fig. 2. (a) Temperature-dependence of Fourier transforms of EXAFS oscillation of Pt-LIII edge for the colloidal dispersions of Pt/Ru(1/1) bimetallic nanoparticles produced by the reduction of the mixture solutions of Pt(IV) and Ru(III) salts with PVP at 30 MPa, and (b) Fourier transforms of EXAFS oscillation of Pt-LIII edge for the Pt foil and Pt/Ru(1/9) alloyed powder. (c) Temperature-dependence of Fourier transforms of EXAFS oscillation of Ru-K edge for the colloidal dispersions of Pt/Ru(1/1) bimetallic nanoparticles produced by the reduction of the mixture solutions of Pt(IV) and Ru(III) salts with PVP at 30 MPa, and (d) Fourier transforms of EXAFS oscillation of Ru-K edge for the Ru powder and Ru/Pt(1/9) alloyed foil.

EXAFS spectra for the Pt/Ru(1/1) colloidal dispersions indicated that the bimetallic nanoparticle has a Pt-core/Ru-shell structure. We consider that this work would demonstrate the usefulness of the high-temperature and high-pressure synthetic method for the various bimetallic particles of noble metals. Acknowledgments We are grateful to Professor S. Isoda, Professor H. Kurata, Dr. T. Nemoto, and Dr. K. Yoshida (Institute of Chemical Research, Kyoto University) for the TEM observations. We appreciate very much for the approval of the Photon Factory Advisory Committee (PAC) (Proposal No. 2005G033) at the High Energy Accelerator Research Organization (KEK) for the EXAFS measurements. This work is partially supported by the fund from Tokuyama Science Foundation.

Supplementary material The online version of this article contains additional supplementary material. Please visit DOI: 10.1016/j.jcis.2008.02.056. References [1] [2] [3] [4] [5]

G. Schmid (Ed.), Nanoparticles, Wiley–VCH, Weinheim, 2004. M. Harada, K. Asakura, N. Toshima, J. Phys. Chem. 98 (1994) 2653. D.G. Duff, P.P. Edwards, B.F.G. Johnson, J. Phys. Chem. 99 (1995) 15934. T. Teranishi, M. Miyake, Chem. Mater. 10 (1998) 594. G. Viau, R. Brayner, L. Poul, N. Chakroune, E. Lacaze, Chem. Mater. 15 (2003) 486. [6] N. Toshima, M. Harada, T. Yonezawa, K. Kushuhashi, K. Asakura, J. Phys. Chem. 95 (1991) 7448. [7] N. Toshima, M. Harada, Y. Yamazaki, K. Asakura, J. Phys. Chem. 96 (1992) 9927. [8] N. Toshima, T. Yonezawa, New J. Chem. (1998) 1179.

M. Ueji et al. / Journal of Colloid and Interface Science 322 (2008) 358–363

[9] K. Niesz, M. Grass, G.A. Somorjai, Nano Lett. 5 (2005) 2238. [10] A.C. Temleton, W.P. Wuelfing, R.W. Murray, Acc. Chem. Res. 33 (2000) 27. [11] M. Brust, C.J. Kiely, Colloids Surf. A 202 (2002) 175. [12] Y.-S. Shon, E. Cutler, Langmuir 20 (2004) 6626. [13] C.J. Ackerson, P.D. Jadzinsky, R.D. Kornberg, J. Am. Chem. Soc. 127 (2005) 6550. [14] M. Büttner, T. Belser, P. Oelhafen, J. Phys. Chem. B 109 (2005) 5464. [15] H. Ohde, F. Hunt, C.M. Wai, Chem. Mater. 13 (2001) 4130. [16] See, e.g., T. Adschiri, K. Kanazawa, K. Arai, J. Am. Ceram. Soc. 75 (1992) 1019. [17] Y. Kimura, D. Abe, T. Ohmori, M. Mizutani, M. Harada, Colloids Surf. A 131 (2003) 231. [18] M. Harada, D. Abe, Y. Kimura, J. Colloid Interface Sci. 292 (2005) 113. [19] Y. Kimura, D. Abe, E. Maru, M. Ueji, M. Harada, in: M. Nakahara, N. Matubayashi, M. Ueno, K. Yasuoka, K. Watanabe (Eds.), Water, Steam, and Aqueous Solutions for Electric Power, 2005, p. 345.

363

[20] M. Ueji, M. Harada, Y. Kimura, Chem. Lett. 34 (2005) 2001. [21] W. Vogel, P. Britz, H. Bönnemann, J. Rothe, J. Hormes, J. Phys. Chem. B 101 (1997) 11029. [22] T.J. Schmidt, M. Noeske, H.A. Gasteiger, R.J. Behm, P. Britz, W. Brijoux, H. Bönnemann, Langmuir 13 (1997) 2591. [23] U.A. Paulus, U. Endruschat, G.J. Feldmeyer, T.J. Schmidt, H. Bönnemann, R.J. Behm, J. Catal. 195 (2000) 383. [24] X. Wang, I. Hsing, Electrochim. Acta 47 (2002) 2981. [25] M. Harada, N. Toshima, K. Yoshida, S. Isoda, J. Colloid Interface Sci. 283 (2005) 64. [26] Y. Shimazaki, Y. Kobayashi, S. Yamada, T. Miwa, M. Konno, J. Colloid Interface Sci. 292 (2005) 122. [27] M. Harada, M. Ueji, Y. Kimura, Colloids Surf. A 315 (2008) 304. [28] M. Nomura, A. Koyama, KEK Report, 1989, p. 89. [29] M. Nomura, A. Koyama, M. Sakurai, KEK Report, 1991, p. 91. [30] B.K. Teo, EXAFS Basic Principles and Data Analysis, Inorganic Chemistry Concepts, vol. 9, Springer-Verlag, Berlin, 1986.