NANO LETTERS
Dumbbell-like Pt-Fe3O4 Nanoparticles and Their Enhanced Catalysis for Oxygen Reduction Reaction
2009 Vol. 9, No. 4 1493-1496
Chao Wang,† Hideo Daimon,‡ and Shouheng Sun*,† Department of Chemistry, Brown UniVersity, ProVidence, Rhode Island 02912, and Fundamental Technology Center, DeVelopment & Technology DiVision, Hitachi Maxell, Ltd., 1-1-88 Ushitora, Ibaraki-shi, Osaka, 567-8567, Japan Received November 16, 2008; Revised Manuscript Received January 28, 2009
ABSTRACT Monodisperse dumbbell-like Pt-Fe3O4 nanoparticles are synthesized by epitaxial growth of Fe onto Pt nanoparticles followed by Fe oxidation. The nanoparticle size in the structure is tunable from 2 to 8 nm for Pt and 6 to 20 nm for Fe3O4. Pt nanoparticles in the Pt-Fe3O4 structure show a 20-fold increase in mass activity toward oxygen reduction reaction compared with the single component Pt nanoparticles and the commercial 3 nm Pt particles. The work proves that it is possible to maximize catalytic activity of the Pt nanoparticle catalyst through the control not only of Pt size and shape but also of its interaction with Fe3O4 nanoparticles.
Developing nanoparticle (NP) catalysts with controlled size and surface structure is key to achieving high catalytic activity. Of numerous NP catalysts studied, Pt NPs have attracted particular interest due to their superior catalysis for many chemical reactions,1-5 especially for hydrogen oxidation and oxygen reduction in fuel cells.6-8 However, the need to limit the Pt usage has promoted the search for Pt-based catalysts with higher activity9-16 or for non-Pt catalysts with comparable catalytic property.16-18 Here we report that in conjunction with Fe3O4 NPs, Pt NPs exhibit high catalytic activity for oxygen reduction reaction (ORR) and this enhancement is dependent on the size of Fe3O4 NPs. We present a general strategy for synthesizing Pt-Fe3O4 NPs with Pt tunable from 2 to 8 nm and Fe3O4 from 6 to 20 nm. In the composite structure, the Pt links epitaxially to the Fe3O4 and becomes electron rich. We demonstrate that this electron-rich Pt offers much enhanced catalysis for ORR in both alkaline (0.5 M KOH) and acid media (0.5 M H2SO4), showing up to 20-fold increase in mass activity compared with the single-component Pt NP catalysts. The work proves that it is possible to maximize catalytic activity of Pt catalyst through control not only on Pt NP size and shape but also on its junction and interaction with metal oxide NPs. The Pt-Fe3O4 NPs were synthesized by epitaxial growth of iron onto Pt NPs followed by Fe oxidation. Pt NPs were prepared via the reduction of platinum acetylacetonate, Pt(acac)2, at 200 °C in the presence of a trace amount (∼1%) * Corresponding author,
[email protected]. † Brown University. ‡ Hitachi Maxell, Ltd. 10.1021/nl8034724 CCC: $40.75 Published on Web 03/04/2009
2009 American Chemical Society
of Fe(CO)5.19 The size of the Pt NPs was tuned by the temperature at which the trace amount of Fe(CO)5 was added. Adding Fe(CO)5 at 120, 160, and 180 °C produced 7, 5, and 3 nm Pt NPs, respectively. Mixing Pt NPs with Fe(CO)5 in 1-octadecene solvent in the presence of oleic acid and oleylamine and heating the mixture to 300 °C, followed by room temperature oxidation under air, led to Pt-Fe3O4 NPs. The Fe3O4 NP size (6-20 nm) was controlled by adjusting the ratio of Fe(CO)5 to Pt NPs. More Fe(CO)5 led to larger Fe3O4 NPs.20 Figure 1 shows the transmission electron microscopy (TEM) images of representative 3-7 nm (Figure 1a), 3-10 nm (Figure 1b), 5-12 nm (Figure 1c), and 5-17 nm (Figure 1d) Pt-Fe3O4 NPs with the black dots being Pt NPs. The epitaxial relation between Pt and Fe3O4 is seen in a highresolution transmission electron microscopy (HRTEM) image of a 3-10 nm Pt-Fe3O4 NP (Figure 1e). The distance between two lattice fringes in 10 nm Fe3O4 measured from the HRTEM image is 0.22 nm, close to (400) plane spacing (0.212 nm) in inverse spinel structured Fe3O4. The interfringe distance in 3 nm Pt is 0.20 nm, corresponding to (200) plane spacing (0.196 nm) in face centered cubic (fcc) Pt. The epitaxial relation between Pt and Fe3O4 is further confirmed by the controlled growth of Fe on the cubic Pt seeds. Figure 1f shows the TEM image of the 7-10 nm Pt-Fe3O4 NPs obtained from the growth of Fe on the 7 nm Pt nanocubes followed by air oxidation of Fe. It can be seen that the Fe3O4 grows on one face of the cube and adopts cubelike morphology. The crystal structure of Pt-Fe3O4 NPs was studied by X-ray diffraction (XRD). Figure 2a is the XRD of Pt-Fe3O4
Figure 1. TEM images of (a) 3-7 nm, (b) 3-10 nm, (c) 5-12 nm, and (d) 5-17 nm Pt-Fe3O4 NPs. (e) HRTEM image of a 3-10 nm Pt-Fe3O4 NP. (f) TEM image of 7-10 nm Pt-Fe3O4 NPs with Pt nanocubes as seeds.
images, indicating that both Pt and Fe3O4 are single crystals in the Pt-Fe3O4structure. The controlled nucleation and growth of only one Fe3O4 on each Pt seeding NP in the current synthetic condition is attributed to electron transfer between Pt and Fe. In the growth process, Fe(CO)5 decomposes to Fe that nucleates on Pt NPs. Once the Fe nucleus is formed on Pt, the free electrons from Fe tend to flow across the junction to Pt due to the higher Fermi energy of Fe (11.1 eV)21 than Pt (8.8 eV).22 As a result, Pt in Pt-Fe becomes electron rich, unsuitable for multinucleation of Fe and giving only the Pt-Fe structure. When exposed to air, Fe is oxidized to Fe3O4, forming Pt-Fe3O4 NPs.
Figure 2. Structural analysis of Pt-Fe3O4 NPs. (a) XRD patterns of Pt-Fe3O4 NPs of various sizes. (b) Pt4f XPS spectra of 5 nm Pt NPs, 5-12 nm Pt-Fe3O4 NPs, and 20 nm Pt film on Si. The binding energy for bulk metallic Pt is also indicated.
NPs with fcc Pt from 3 to 7 nm and inverse spinel structured Fe3O4 from 7 to 15 nm. The increase in peak sharpness reflects the size increase for both Pt and Fe3O4 NPs. The sizes of Pt and Fe3O4 NPs estimated from Scherrer’s formula are close to what are measured in the representative TEM 1494
The electronic structure of Pt in Pt-Fe3O4 was qualitatively characterized by X-ray photoelectron spectroscopy (XPS) measurements. Figure 2b shows the electron binding energy of Pt4f measured from XPS for 5 nm Pt NPs, 5-12 nm Pt-Fe3O4 NPs, and 20 nm Pt film on Si substrate prepared by sputtering. Compared with the 5 nm Pt NPs and the 20 nm Pt film, the 5-12 nm Pt-Fe3O4 NPs have a ∼0.4 eV decrease in electron binding energy. Considering that the reference level (C 1s, set to 284.5 eV for calibration)23 has no change in the measurements, the change of binding energy is likely due to the electron transfer from Fe3O4 to Pt and may also be explained by the work function difference24 in 5.93 eV25 for Pt(111) and 5.52 eV26 for Fe3O4(111). Such observation of Pt chemical shift by XPS has also been reported in thin film studies of Pt/oxide interface, including both negative (receiving electrons)27 and positive shift (losing electrons).28 The charge transfer from Fe3O4 to Pt is further proved by electrochemistry behavior of Pt-Fe3O4 NPs. Cyclic voltammograms (CVs) under N2 bubbling in 0.5 M KOH were recorded for 5 nm Pt, 5-12 nm Pt-Fe3O4, 5-17 nm Nano Lett., Vol. 9, No. 4, 2009
Figure 3. TEM images (a) 5 nm Pt and (b) 5-17 nm Pt-Fe3O4 NPs on the amorphous carbon support. (c) I-V curves for Pt and Pt-Fe3O4 catalysts obtained from the RDE measurements in 0.5 M KOH with the rotation speed at 1600 rpm and sweeping rate at 10 mV/s. (d) Mass activities of Pt and Pt-Fe3O4 catalysts.
Pt-Fe3O4, 3 nm commercial Pt (TEC10E50E from Tanaka Noble Metal Ltd.), and 17 nm Fe3O4 NPs (Figures S1a-S5a in Supporting Information). Unlike the CV curve for Pt catalyst, Fe3O4 NPs and Pt-Fe3O4 NPs have two redox peaks at -0.4 V/-1.0 V and at -0.6 V/-0.9 V, respectively. These two peaks correspond to the Fe(II)/Fe(III) redox couple.29 Compared with Fe3O4 NPs, the shift of redox potential for Pt-Fe3O4 NPs is due to the charge transfer from Fe to Pt that causes Fe3O4 in Pt-Fe3O4 to be more difficultly oxidized but more easily reduced. The slight electronic structure change on Pt in Pt-Fe3O4 induces significant improvement in Pt catalysis. We demonstrate that Pt NPs in Pt-Fe3O4 are more active catalysts than the single component Pt NPs for ORR. Parts a and b of Figure 3 show the TEM images of the 5 nm Pt and 5-17 nm Pt-Fe3O4 NPs on the carbon support used for the test. Rotating disk electrode (RDE) measurements in 0.5 M KOH at various rotation speeds (rounds per minute, rpm) were performed under O2 bubbling.20 The polarization curves from 5 nm Pt, 5-12 nm Pt-Fe3O4, 5-17 nm Pt-Fe3O4, 3 nm commercial Pt and 17 nm Fe3O4 NPs are given in Figures S1b-S5b of the Supporting Information. Figure 3c shows the oxygen reduction currents (I) versus the potentials (V) applied to the electrode for 5 nm Pt, 5-12 nm Pt-Fe3O4, and 5-17 nm Pt-Fe3O4 catalysts measured at 1600 rpm. It can be seen the half-wave potentials for the 5-12 nm Pt-Fe3O4 and 5-17 nm Pt-Fe3O4 NPs are shifted positively by about 20 and 50 mV, respectively. The shift indicates that O2 is more readily reduced on Pt-Fe3O4 (or on 5-17 nm Pt-Fe3O4) than on Pt (or on 5-12 nm Pt-Fe3O4). At -0.1 V (vs Hg/HgO), a potential commonly used for testing Nano Lett., Vol. 9, No. 4, 2009
Figure 4. ORR mass activities for 5 nm Pt and 5-17 nm Pt-Fe3O4 NP catalysts measured in 0.5 M H2SO4 with the RDE rotation speed at 1600 rpm and sweeping rate at 10 mV/s.
ORR in alkaline media, the current for the 5-17 nm Pt-Fe3O4 catalyst reaches -0.1 mA, about 1.4 times that of the current measured for Pt catalyst (-0.07 mA). Note that the Fe3O4 NPs do not have any significant contribution to the reduction current at this potential, as shown in Figure S6 of the Supporting Information. Figure 3d shows the mass activities of Pt and Pt-Fe3O4 catalysts obtained by plotting I versus V over 1 mg of Pt. For 0.01 mg of NP catalyst used for catalysis test, the amount of Pt in the 5-17 nm Pt-Fe3O4 is 0.00066 mg. (The weight ratio of Pt in 5-17 nm Pt-Fe3O4 was measured to be 6.6% by the inductively coupled plasma atomic emission spectroscopy.) At -0.1 V, the mass activity for 5-17 nm Pt-Fe3O4 catalyst reaches 142.4 mA/mg of Pt, about 20 times of that of the 5 nm Pt NP catalyst (7.2 1495
mA/mg of Pt). The reduction current for the commercial 3 nm Pt catalyst is almost the same as that of 5 nm Pt NPs (Figure S4b of the Supporting Information). The data indicate that in conjunction with Fe3O4 NPs, Pt NPs are catalytically more active than the single component Pt and this catalysis can be further enhanced by larger Fe3O4 NPs. The Pt-Fe3O4 NPs also show enhanced catalytic activity for ORR in acid media. In a separate test, Pt and Pt-Fe3O4 NPs were loaded on the GC electrode and stabilized by Nafion.20 The catalytic activity for ORR was evaluated by RDE in 0.5 M H2SO4sa medium used commonly to mimic the conditions in proton exchange membrane fuel cells. Figure 4 shows the mass activities from the 5 nm Pt and 5-17 nm Pt-Fe3O4 NPs. It can be seen that the Pt in Pt-Fe3O4 NPs have a 5-fold enhancement in mass activity at the potential 0.5 V (vs Ag/AgCl) used for ORR. Note that Fe3O4 has only limited acid resistance and the dissolution of Fe3O4 by H2SO4 can only be neglected in the first 10 rounds of the measurements after which the activity drops gradually to that of the 5 nm Pt NPs. This further proves that the enhanced Pt catalysis comes from the interaction between Pt and Fe3O4 in the Pt-Fe3O4 structure. The present work demonstrates that the dumbbell-like Pt-Fe3O4 NPs are readily made from epitaxial growth of Fe onto Pt followed by air oxidation of Fe. The interaction between Pt and Fe3O4 results in higher electron population on Pt, making it catalytically more active toward ORR than the single component Pt NP catalyst. The Pt-Fe3O4 catalyst reduces the overpotential for ORR in 0.5 M KOH by up to 50 mV and shows a 20-fold increase in mass activity compared with either Pt NPs of similar size or the commercial 3 nm Pt catalyst. The composite NP catalyst also exhibits a 5-fold increase in mass activity for ORR in 0.5 M H2SO4 over Pt NPs. The work proves that catalytic activity of Pt catalyst can indeed be improved through control not only on Pt NP size and shape but also on its junction and interaction with Fe3O4 NPs. With these nanoscale tuning capabilities, we expect that highly active Pt or non-Pt NP catalysts in junction with other NPs can be developed for fuel cell and other important reaction applications. Acknowledgment. The work was supported by NSF/ DMR 0606264, the Brown University Seed Fund, and a scholarship from Hitachi Maxell, Ltd. We thank Mr. Anthony McCormick for the help on HRTEM.
1496
Supporting Information Available: Dumbbell-like NPs synthesis and electrochemistry study. These materials are available free of charge via the Internet at http://pubs.acs.org. References (1) Davis, R. J.; Derouane, E. G. Nature 1991, 349, 313. (2) Ahmadi, T. S.; Wang, Z. L.; Green, T. C.; Henglein, A.; EI-Sayed, M. A. Science 1996, 272, 1924. (3) Studer, M.; Blaser, H.-U.; Exner, C. AdV. Synth. Catal. 2003, 345, 45. (4) Hagen, J. Industrial Catalysis: a Practical Approach; Wiley-VCH: New York, 2006. (5) Bratlie, K. M.; Lee, H.; Komvopoulos, K.; Yang, P.; Somorjai, G. A. Nano Lett. 2007, 7, 3097. (6) Steele, B. C. H.; Heinzel, A. Nature 2001, 414, 345. (7) Larminie, J.; Dicks, A. Fuel Cell Systems Explained; John Wiley & Sons: New York, 2003. (8) Brandon, N. P.; Skinner, S.; Steele, B. C. H. Annu. ReV. Mater. Res. 2003, 33, 183. (9) Toda, T.; Igarashi, H.; Uchida, H.; Watanabe, M. J. Electrochem. Soc. 1999, 146, 3750. (10) Zhang, J.; Vukmirovic, M. B.; Xu, Y.; Mavrikakis, M.; Adzic, R. R. Angew. Chem., Int. Ed. 2005, 44, 2132. (11) Stamenkovic, V. R.; Fowler, B.; Mun, B. S.; Wang, G.; Ross, P. N.; Lucas, C. A.; Markovic, N. M. Science 2007, 315, 493. (12) Stamenkovic, V. R.; Mun, B. S.; Arenz, M.; Mayrhofer, K.; Lucas, C. A.; Wang, G.; Ross, P. N.; Markovic, N. M. Nat. Mater. 2007, 6, 241. (13) Tian, N.; Zhou, Z.-Y.; Sun, S.-G.; Ding, Y.; Wang, Z. L. Science 2007, 316, 732. (14) Zhang, J.; Sasaki, K.; Sutter, E.; Adzic, R. R. Science 2007, 315, 220. (15) Lee, E. P.; Peng, Z.; Cate, D. M.; Yang, H.; Campbell, C. T.; Xia, Y. J. Am. Chem. Soc. 2007, 129, 10634. (16) Gasteiger, H. A.; Kocha, S. S.; Sompalli, B.; Wagner, F. T. Appl. Catal., B 2005, 56, 9. (17) Bashyam, R.; Zelenay, P. Nature 2006, 443, 63. (18) Baker, W. S.; Pietron, J. J.; Teliska, M. E.; Bouwman, P. J.; Ramaker, D. E.; Swider-Lyons, K. E. J. Electrochem. Soc. 2006, 153, A1702. (19) Wang, C.; Daimon, H.; Onadera, T.; Koda, T.; Sun, S. Angew. Chem., Int. Ed. 2008, 47, 1. (20) Supporting Information. (21) Lide, D. R. CRC Handbook of Chemistry and Physics, 84th ed.; CRC Press: New York, 2003. (22) Smith, N. V. Phys. ReV. B 1974, 9, 1365. (23) Riviere, J. C.; Myhra, S. Handbook of Surface and Interface Analysis: Methods for Problem-SolVing; Marcel Dekker, Inc.: New York, 1998. (24) Qin, Z.-H.; Lewandowski, M.; Sun, Y.-N.; Shaikhutdinov, S.; Freund, H.-J. J. Phys. Chem. C 2008, 112, 10209. (25) Weast, R. C.; Astle, M. J. CRC Handbook of Chemistry and Physics, 63rd ed.; CRC Press, Boca Rotan, FL, 1982. (26) Weiss, W.; Ranke, W. Prog. Surf. Sci. 2002, 70, 1. (27) Bahl, M. K.; Tsai, S. C.; Chung, Y. W. Phys. ReV. B 1980, 21, 1344. (28) Murgai, V.; Raaen, S.; Strongin, M.; Garrett, R. F. Phys. ReV. B 1986, 33, 4345. (29) Hang, B. T.; Hayashi, H.; Yoon, S.-H.; Okada, S.; Yamaki, J. J. Power Sources 2008, 178, 393.
NL8034724
Nano Lett., Vol. 9, No. 4, 2009
