- Email: [email protected]
Microemulsion synthesis and electrocatalytic properties of carbon-supported Pd–Co–Au alloy nanoparticles
Journal of Colloid and Interface Science 367 (2012) 337–341
Contents lists available at SciVerse ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Microemulsion synthesis and electrocatalytic properties of carbon-supported Pd–Co–Au alloy nanoparticles Ch. Venkateswara Rao ⇑, B. Viswanathan ⇑ National Centre for Catalysis Research, Department of Chemistry, Indian Institute of Technology Madras, Chennai 600 036, India
a r t i c l e
i n f o
Article history: Received 26 May 2011 Accepted 9 October 2011 Available online 7 November 2011 Keywords: Pd–Co–Au alloy Oxygen reduction Water-in-oil microemulsion Cyclic voltammetry
a b s t r a c t Carbon-supported Pd and Pd–Co–Au alloy electrocatalysts are prepared by water-in-oil microemulsion technique using water as aqueous phase, non-ionic Triton-X-100 as surfactant, and cyclohexane as oil phase. The materials are characterized using XRD, TEM, and EDX. Face-centered cubic structure of Pd and presence of respective elements with controllable composition is evident from the analysis of XRD and EDX. The role of alloying elements on the redox behavior as well as catalytic activity of Pd is studied by cyclic voltammetry and ultraviolet photoelectron spectroscopy. Electrochemical measurements performed by rotating disk electrode voltammetry indicate the good ORR activity of Pd–Co–Au/C compared to Pd/C. Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction A variety of electrode materials have been investigated for ORR over the years. To date, Pt-based materials are the most successful electrocatalysts, even though the kinetics of ORR on the precious metal is sluggish and causes performance losses in energy conversion devices [1–4]. In recent years, there has been great interest in the search and development of electrode materials alternative to platinum for ORR. Pd is one of the efficient electrodes for oxygen reduction reaction (ORR) in acidic medium. But the intermediate oxygen species covered on the Pd surface at the technically relevant potential region of 0.7–0.9 V vs NHE significantly reduces ORR activity [5–9]. To enhance ORR activity, various bimetallic and trimetallic Pd alloys such as Pd–Co, Pd–Cr, Pd–Ni, Pd–Fe, Pd–Ti, Pd–Co–Mo, and Pd–Co–Au were investigated [10–18]. The results implied the comparable performance of Pd–Co–M/C (M = Mo and Au) with that of commercial Pt/C [14–17]. The origin of the good ORR activity of Pd alloys has been undertaken, and various hypotheses have been proposed to explain the enhanced activity of Pd alloys compared to that of Pd [10–16,19,20]. It is assumed that the observed enhancement in the ORR activity for the Pd alloys might be related to the electronic property of Pd like the Pt alloys. But still the reason for enhanced ORR activity of Pd alloys is not clear. In the present work, carbon-supported Pd and Pd–Co–Au alloy nanoparticles are prepared by employing water/Triton-X-100/ propanol-2/cyclohexane microemulsion system and characterized. ⇑ Corresponding authors. Fax: +91 44 2257 4202. E-mail addresses: [email protected] [email protected] (B. Viswanathan).
(Ch.
Venkateswara
0021-9797/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2011.10.020
Rao),
The role of alloying elements toward the oxygen reduction activity of Pd catalysts is studied by cyclic voltammetry and ultraviolet photoelectron spectroscopy (UPS).
2. Experimental section 2.1. Synthesis of carbon-supported Pd and Pd–Co–Au alloy nanoparticles The microemulsion system used in this study consisted of nonionic Triton-X-100 as a surfactant, propanol-2 as a co-surfactant, cyclohexane as the continuous oil phase, and either the Pd–Co– Au precursor solution or hydrazine solution as the dispersed aqueous phase. Stock solutions of 7 mM H2PdCl4, 2 mM CoCl2, 1 mM HAuCl4, and 1 M hydrazine are used to prepare microemulsions. Microemulsions 1 and 2 are prepared separately by mixing by volume 10% surfactant, 35% cyclohexane, 40% propanol-2, and 15% of the aqueous phase. The amount of hydrazine is in stoichiometric excess compared with the equal volume of metal precursors in the microemulsion 1. The two stable microemulsions were then mixed together and stirred for 2 h. Subsequently, an appropriate amount of carbon black (CDX975, received from Columbian Chemicals Company, USA; Surface area 300 m2/g) was added to the mixture to give a metal(s)/C weight ratio of 20:80. The resultant slurry was stirred for 2 h, filtered, washed with acetone followed by de-ionized water, and dried in an air oven at 75 °C for 2 h. Pd/C was also prepared in the similar manner. The as-synthesized Pd–Co–Au/C was heat-treated (HT) at different temperatures 700, 800, and 900 °C in a flowing mixture of 10% H2–90% Ar for 1 h to promote alloy formation.
338
Ch. Venkateswara Rao, B. Viswanathan / Journal of Colloid and Interface Science 367 (2012) 337–341
2.2. Characterization techniques XRD measurements were performed on a Rigaku Miniflex X-ray diffractometer using a Cu Ka source at a scan rate of 0.025 s1. Scanning electron microscope equipped with EDX (FEI, Model: Quanta 200) was used to observe the composition and distribution of elements in the catalysts. Transmission electron microscopic images were obtained using JEOL2100 system. Ultraviolet photoemission spectroscopy (UPS) measurements of the catalysts was carried out with Omicron nanotechnology instrument using an Al monochromatic X-ray (hm = 1486.6 eV) and operated under the base pressure of 62 109 mbar. The powder samples were mounted onto the holder, and the catalyst surface was scraped in situ by repeated cycles of Ar-ion sputtering to remove any surface contamination that could arise from atmospheric components like water and CO2. The energy resolution of the spectrometer was set at 1.1 eV at pass energy of 80 eV. The binding energy (BE) was calibrated with respect to Ag 3d core level. For the high-resolution valence-band photoemission spectra, the total energy resolution, including the photon energy resolution, was less than 0.05 eV. 2.3. Electrochemical measurements Electrocatalytic activity of the materials was investigated by cyclic voltammetry using a potentiostat (BAS 100 electrochemical analyzer). All experiments were performed in a one-compartment cell assembled with Ag/AgCl as reference, Pt foil as counter, and glassy carbon disk as working electrodes, respectively. A thin-film rotating disk glassy carbon electrode (Pine Instruments) technique is employed to investigate the ORR activity of the materials. The working electrode was fabricated as follows: 5 mg of catalyst was dispersed in 5 ml of isopropanol by ultrasonication for 15 min. An aliquot of 20 ll catalyst ink was pipetted onto the mirror-polished glassy carbon substrate and dried in flowing argon at room temperature. Then, 10 ll of diluted 5 wt.% NafionÒ solution was pipetted on the electrode surface in order to attach the catalyst particles onto the GC disk and dried. For oxygen reduction measurements, linear sweep voltammograms (LSVs) were recorded between +0.0 and +1.2 V vs NHE at a scan rate of 5 mV/ s and rotation rate of 1600 rpm in O2-saturated 0.5 M H2SO4. Current densities are normalized to the geometric area of the glassy carbon substrate in the text. 3. Results and discussion Microemulsion method was employed as a suitable method to generate metal colloids and/or clusters on the nanoscale with controllable composition. The diameter and morphology of nanoparticles can be controlled by varying the concentration and type of surfactant, concentration of precursor and reducing agents, type of solvent, and ultrasonication [21–23]. Under the experimental conditions, Pd–Co–Au nanoparticles were formed upon contact between the precursor containing droplets and the hydrazine containing droplets. The reductions of H2PdCl4, CoCl2, and HAuCl4 are as follows:
2H2 PdCl4 þ 7N2 H4 H2 O ! 2Pd þ 8NH4 Cl þ 3N2 þ 7H2 O CoCl2 þ 2N2 H4 H2 O ! Co þ 2NH4 Cl þ N2 þ 2H2 O HAuCl4 þ 4N2 H4 H2 O ! Au þ 4NH4 Cl þ 2N2 þ 4H2 O Powder XRD patterns recorded for the carbon-supported Pd and Pd–Co–Au alloy materials are shown in Fig. 1. Both the Pd/C and as-synthesized Pd–Co–Au/C exhibited diffraction peaks at 2h values around 40°, 47°, 68°, 82°, and 87° corresponding to the (1 1 1),
Fig. 1. Powder XRD pattern of (a) Pd/C, (b) as-synthesized Pd–Co–Au/C; heattreated Pd–Co–Au/C at (c) 700, (d) 800, and (e) 900 °C, respectively.
(2 0 0), (2 2 0), (3 1 1), and (2 2 2) planes of face-centered cubic structure of Pd (JCPDS No. 88-2335). The broad peaks indicate that the particles are in nanocrystalline range. The diffraction peak at 2h value around 25° corresponds to the (0 0 2) plane of the hexagonal carbon support. No appreciable change in the XRD pattern of the as-synthesized Pd–Co–Au/C compared to the Pd/C indicates that the Pd/Co/Au metal nanoparticles are not transformed into alloy. In order to form alloy, the as-synthesized Pd–Co–Au/C is heat-treated at 700, 800, and 900 °C. With increasing temperature, the Pd diffraction peaks are shifted to higher angles in Pd–Co–Au/C compared to the Pd/C. It indicates the contraction of Pd lattice due to the alloy formation. The extent of shift increases and the lattice parameter decreases with increasing heat-treatment temperature, suggesting an increase in the degree of alloying. By using lattice parameter values, degree of alloying DA is calculated by using the equation: fa a g DA ¼ a Pda alloyðTÞ 100; where aPd, aalloy(T), and aalloy(900) refer to f Pd alloyð900Þ g the cell parameter values of Pd, Pd–Co–Au alloy heat-treated at various temperatures (T = 700 and 800 °C), and Pd–Co–Au alloy heattreated at 900 °C, respectively [18]. The DA values are given in Table 1. It indicates the increase in degree of alloying with heattreatment temperature. TEM images of the Pd/C and heat-treated Pd–Co–Au/C catalysts are shown in Fig. 2. The images depict a relatively narrow size distribution of Pd nanoparticles over carbon support in the case of Pd/C whereas a relatively broader distribution of Pd alloy nanoparticles over carbon support in the case of Pd–Co– Au/C catalysts. The average particle sizes obtained from the measurement of about 200 particles found in an arbitrarily chosen area of enlarged micrographs by assuming spherical shape of particles are given in Table 1. Fig. 2a shows the good dispersion of spherical-shaped Pd nanoparticles of average size 2.1 nm with a standard deviation of 0.2 nm. The estimated particle sizes for the heat-treated Pd–Co–Au/C at 700, 800, and 900 °C are 23 ± 8, 31 ± 13, and 42 ± 17 nm, respectively. EDX confirms the presence of Pd and C in Pd/C and Pd, Co, Au, and C in Pd–Co–Au/C, respectively. The calculated elemental composition in the heat-treated Pd–Co–Au/C catalysts is corresponding to the approximate atomic ratio of 7:2:1 (Table 1). Scanning electron microscopic images with EDX mapping for the Pd/C and heat-treated Pd–Co–Au/C at 800 °C are shown in Fig. 3. The EDX mapping images depicted the homogeneous distribution of respective elements in the catalysts.
339
Ch. Venkateswara Rao, B. Viswanathan / Journal of Colloid and Interface Science 367 (2012) 337–341 Table 1 Degree of alloying, average particle size (from TEM), EDX composition, onset potential, and ORR activity of the catalysts. Catalyst
Degree of alloying (%)
Particle size (nm)
EDX composition Pd:Co:Au (wt.%)
Onset potential (mV vs NHE)
Activity at +0.8 V (mA/ cm2)
As-syn Pd/C As-syn Pd–Co–Au/C HT Pd–Co–Au/C at 700 °C HT Pd–Co–Au/C at 800 °C HT Pd–Co–Au/C at 900 °C
– – 59
2.1 – 23
100:–:– – 70.2:11.2:18.6
+845 – +885
0.15 – 0.53
79
31
71.3:11.6:17.1
+920
1.32
100
42
71.8:11.3:16.9
+915
0.92
(HT = heat-treated).
(a)
(b)
50 nm
20 nm
(c)
50 nm
(d)
100 nm
Fig. 2. TEM images of (a) Pd/C; heat-treated Pd–Co–Au/C at (b) 700, (c) 800, and (d) 900 °C, respectively.
Cyclic voltammograms (CVs) of the Pd/C and Pd–Co–Au/C catalysts in deaerated 0.5 M H2SO4 are shown in Fig. 4. The redox behavior of the Pd/C is found to be consistent with literature reports [24,25,14]. The redox peaks observed in the +0.15 0.25 V region correspond to the adsorption/desorption of hydrogen on the Pd surface and formation of the Pd hydride phase. The redox peaks in the 0.6–1.2 V region corresponding to the formation and reduction of surface Pd oxides (Pd–OH or PdOx species). It is known that the PdOx species at the metallic Pd surface is inactive toward oxygen reduction [26]. For the as-synthesized Pd–Co–Au/C catalyst, anodic peak was shifted to lower potentials of about 0.14 V compared to Pd/C. It shows the modification of the electronic structure of Pd by the presence of Co and Au. In the case of heattreated Pd–Co–Au/C, hydrogen adsorption/desorption peaks are appeared at potentials of 0.04 V showed a similar behavior to that of Pt [24,27]. This is due to the interruption of the dissolution of
hydrogen into the Pd lattice by alloying elements. Moreover, the formation of surface oxy species, which are the poisoning species for oxygen reduction, was not observed, and this may be beneficial to the oxygen adsorption at low overpotential and thus the ORR kinetic enhancement. Linear sweep voltammograms (LSVs) of the Pd/C and Pd–Co– Au/C catalysts in both Ar- and O2-saturated 0.5 M H2SO4 is shown in Fig. 5. When the potential was swept from +1.2 to +0.2 V, single oxygen reduction peak was observed in the potential region of 1.0– 0.6 V. The calculated ORR activity of the catalysts is given in Table 1. The results indicate the good ORR performance of Pd–Co–Au/C catalysts compared to Pd/C. This is due to the suppression of (hydr)oxy species on Pd electrode surface by the presence of alloying elements. Among the heat-treated Pd–Co–Au/C catalysts, maximum activity was observed for the heat-treated Pd–Co–Au/C at 800 °C compared to 700 and 900 °C. The initial increase is due
340
Ch. Venkateswara Rao, B. Viswanathan / Journal of Colloid and Interface Science 367 (2012) 337–341
Pd
(a)
300 nm Pd
(b)
300 nm Co
Au
Fig. 3. EDX mapping of elements in the catalysts: (a) Pd/C and (b) heat-treated Pd–Co–Au/C at 800 °C.
Fig. 4. Cyclic voltammograms (CVs) of carbon-supported Pd and Pd–Co–Au catalysts in Ar-saturated 0.5 M H2SO4; scan rate—25 mV/s.
to an increase in the degree of alloying while the decrease at higher temperatures is due to an increase in crystallite size. Also, ORR proceeded in a relatively positive potential region for the Pd–Co–Au/C catalysts compared to Pd/C. For each of the investigated materials, the onset potential for ORR – the potential where the current density increases to 20 lA/cm2 – is given in Table 1. The overpotential for ORR on the best Pd–Co–Au alloy was
Fig. 5. Linear sweep voltammograms (LSVs) of O2 reduction on carbon-supported Pd and Pd–Co–Au catalysts in 0.5 M H2SO4; scan rate—5 mV/s and rotation rate— 1600 rpm.
ca.75 mV less compared to Pd. The differences in the onset potential for oxygen reduction at various heat-treated Pd–Co–Au/ C catalysts may be due to differences in the surface activation, which are related to the size and distribution of the metallic nanoparticles.
Ch. Venkateswara Rao, B. Viswanathan / Journal of Colloid and Interface Science 367 (2012) 337–341
341
performance. The good activity of the Pd–Co–Au/C compared to Pd/C makes it as applicable cathode for fuel cell applications. Acknowledgments The authors wish to acknowledge Columbian Chemicals Company, USA, for financial support and also Department of Science and Technology (DST), India, for creating National Centre for Catalysis Research at IIT Madras. References
Fig. 6. Valence-band spectra of (a) Pd/C and (b) heat-treated Pd–Co–Au/C at 800 °C.
The voltammetric behavior of carbon-supported Pd and Pd–Co– Au alloys indicates the change in the electronic structure of Pd by alloying elements. To investigate the surface electronic properties of Pd, valence-band spectra of Pd/C and heat-treated Pd–Co–Au/C at 800 °C is recorded and shown in Fig. 6. It depicts the shift of Pd d-band center from 2.27 eV on Pd/C to 3.04 eV on Pd–Co–Au/ C. A similar downshift was observed for a variety of bimetallic Pt alloys [28]. The downshift of d-band center indicates the decreased density of states at the Fermi level of Pd in the Pd–Co–Au/C. As a result, the formation of Oads or OHads on Pd surface was inhibited (evident from CV behavior shown in Fig. 4) and leads to the good ORR activity (evident from LSV behavior shown in Fig. 5). 4. Conclusions Carbon-supported Pd–Co–Au alloys were synthesized by microemulsion method and characterized. XRD showed the single phase face-centered cubic structure of Pd in the catalysts. EDX analysis confirmed the atomic composition of 7:2:1 in the heat-treated Pd–Co–Au/C catalysts. Voltammograms and valence-band spectra measurements revealed the inhibition of (hydr)oxy species on Pd surface by the presence alloying elements enhanced the ORR
[1] B. Viswanathan, M. Aulice, Scibioh (Eds.), Fuel Cells – Principles and Applications, Universities Press (India) Private Limited, 2006. [2] H.A. Gasteiger, N.M. Markovic, Science 324 (2009) 48–49. [3] Ch. Venkateswara Rao, B. Viswanathan, J. Phys. Chem. C 113 (2009) 18907– 18913. [4] Ch. Venkateswara Rao, B. Viswanathan, J. Phys. Chem. C 114 (2010) 8661– 8667. [5] D.B. Sepa, M.V. Vojnovic, A. Damjanovic, Electrochim. Acta 26 (1981) 781–793. [6] Lj.M. Vracˇar, D.B. Sepa, A. Damjanovic, J. Electrochem. Soc. 133 (1986) 1835– 1839. [7] Lj.M. Vracˇar, D.B. Sepa, A. Damjanovic, J. Electrochem. Soc. 136 (1989) 1973– 1977. [8] F.G. Will, C.A. Knorr, Z. Electrochem. 64 (1960) 258–269. [9] K. Kinoshita, Electrochemical Oxygen Technology, John Wiley & Sons, New York, 1992. [10] J.L. Fernández, D.A. Walsh, A.J. Bard, J. Am. Chem. Soc. 127 (2005) 357–365. [11] M.H. Shao, K. Sasaki, R.R. Adzic, J. Am. Chem. Soc. 128 (2006) 3526–3527. [12] Y.X. Wang, P.B. Balbuena, J. Phys. Chem. B 109 (2005) 18902–18906. [13] K. Lee, O. Savadogo, A. Ishihara, S. Mitsushima, N. Kamiya, K.I. Ota, J. Electrochem. Soc. 153 (2006) A20–A24. [14] Ch. Venkateswara Rao, B. Viswanathan, Electrochim. Acta 55 (2010) 3002– 3007. [15] J.L. Fernández, J.M. White, Y. Sun, W. Tang, G. Henkelman, A.J. Bard, Langmuir 22 (2006) 10426–10432. [16] J.L. Fernández, V. Raghuveer, A. Manthiram, A.J. Bard, J. Am. Chem. Soc. 127 (2005) 13100–13101. [17] V. Raghuveer, A. Manthiram, A.J. Bard, J. Phys. Chem. B 109 (2005) 22909– 22912. [18] V. Raghuveer, P.J. Ferreira, A. Manthiram, Electrochem. Commun. 8 (2006) 807–814. [19] B. Hammer, J.K. Nørskov, Adv. Catal. 45 (2000) 71–129. [20] Y. Guo, L. Zhuang, J. Lu, Angew. Chem. Int. Ed. 46 (2007) 2862–2864. [21] K. Holmberg, J. Colloid Interface Sci. 274 (2004) 355–364. [22] R. Najjar, C. Stubenrauch, J. Colloid Interface Sci. 331 (2009) 214–220. [23] Ch. Venkateswara Rao, B. Viswanathan, J. Phys. Chem. C 111 (2007) 16538– 16543. [24] M. Enyo, T. Yamazaki, K. Kai, K. Suzuki, Electrochim. Acta 28 (1983) 1573– 1579. [25] A.J. Bard, Electroanalytical Chemistry, Marcel Dekker Inc., New York, 1976. [26] J. Lipkowski, P.N. Ross, Electrocatalysis, Wiley-VCH, New York, 1998. [27] J.P. Hoare, S. Schuldiner, J. Phys. Chem. 62 (1958) 229–233. [28] B.S. Mun, M. Watanabe, M. Rossi, V. Stamenkovic, N.M. Markovic, P.N. Ross Jr., J. Chem. Phys. 123 (2005) 204717–204721.
Contents lists available at SciVerse ScienceDirect
Journal of Colloid and Interface Science www.elsevier.com/locate/jcis
Microemulsion synthesis and electrocatalytic properties of carbon-supported Pd–Co–Au alloy nanoparticles Ch. Venkateswara Rao ⇑, B. Viswanathan ⇑ National Centre for Catalysis Research, Department of Chemistry, Indian Institute of Technology Madras, Chennai 600 036, India
a r t i c l e
i n f o
Article history: Received 26 May 2011 Accepted 9 October 2011 Available online 7 November 2011 Keywords: Pd–Co–Au alloy Oxygen reduction Water-in-oil microemulsion Cyclic voltammetry
a b s t r a c t Carbon-supported Pd and Pd–Co–Au alloy electrocatalysts are prepared by water-in-oil microemulsion technique using water as aqueous phase, non-ionic Triton-X-100 as surfactant, and cyclohexane as oil phase. The materials are characterized using XRD, TEM, and EDX. Face-centered cubic structure of Pd and presence of respective elements with controllable composition is evident from the analysis of XRD and EDX. The role of alloying elements on the redox behavior as well as catalytic activity of Pd is studied by cyclic voltammetry and ultraviolet photoelectron spectroscopy. Electrochemical measurements performed by rotating disk electrode voltammetry indicate the good ORR activity of Pd–Co–Au/C compared to Pd/C. Ó 2011 Elsevier Inc. All rights reserved.
1. Introduction A variety of electrode materials have been investigated for ORR over the years. To date, Pt-based materials are the most successful electrocatalysts, even though the kinetics of ORR on the precious metal is sluggish and causes performance losses in energy conversion devices [1–4]. In recent years, there has been great interest in the search and development of electrode materials alternative to platinum for ORR. Pd is one of the efficient electrodes for oxygen reduction reaction (ORR) in acidic medium. But the intermediate oxygen species covered on the Pd surface at the technically relevant potential region of 0.7–0.9 V vs NHE significantly reduces ORR activity [5–9]. To enhance ORR activity, various bimetallic and trimetallic Pd alloys such as Pd–Co, Pd–Cr, Pd–Ni, Pd–Fe, Pd–Ti, Pd–Co–Mo, and Pd–Co–Au were investigated [10–18]. The results implied the comparable performance of Pd–Co–M/C (M = Mo and Au) with that of commercial Pt/C [14–17]. The origin of the good ORR activity of Pd alloys has been undertaken, and various hypotheses have been proposed to explain the enhanced activity of Pd alloys compared to that of Pd [10–16,19,20]. It is assumed that the observed enhancement in the ORR activity for the Pd alloys might be related to the electronic property of Pd like the Pt alloys. But still the reason for enhanced ORR activity of Pd alloys is not clear. In the present work, carbon-supported Pd and Pd–Co–Au alloy nanoparticles are prepared by employing water/Triton-X-100/ propanol-2/cyclohexane microemulsion system and characterized. ⇑ Corresponding authors. Fax: +91 44 2257 4202. E-mail addresses: [email protected] [email protected] (B. Viswanathan).
(Ch.
Venkateswara
0021-9797/$ - see front matter Ó 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.jcis.2011.10.020
Rao),
The role of alloying elements toward the oxygen reduction activity of Pd catalysts is studied by cyclic voltammetry and ultraviolet photoelectron spectroscopy (UPS).
2. Experimental section 2.1. Synthesis of carbon-supported Pd and Pd–Co–Au alloy nanoparticles The microemulsion system used in this study consisted of nonionic Triton-X-100 as a surfactant, propanol-2 as a co-surfactant, cyclohexane as the continuous oil phase, and either the Pd–Co– Au precursor solution or hydrazine solution as the dispersed aqueous phase. Stock solutions of 7 mM H2PdCl4, 2 mM CoCl2, 1 mM HAuCl4, and 1 M hydrazine are used to prepare microemulsions. Microemulsions 1 and 2 are prepared separately by mixing by volume 10% surfactant, 35% cyclohexane, 40% propanol-2, and 15% of the aqueous phase. The amount of hydrazine is in stoichiometric excess compared with the equal volume of metal precursors in the microemulsion 1. The two stable microemulsions were then mixed together and stirred for 2 h. Subsequently, an appropriate amount of carbon black (CDX975, received from Columbian Chemicals Company, USA; Surface area 300 m2/g) was added to the mixture to give a metal(s)/C weight ratio of 20:80. The resultant slurry was stirred for 2 h, filtered, washed with acetone followed by de-ionized water, and dried in an air oven at 75 °C for 2 h. Pd/C was also prepared in the similar manner. The as-synthesized Pd–Co–Au/C was heat-treated (HT) at different temperatures 700, 800, and 900 °C in a flowing mixture of 10% H2–90% Ar for 1 h to promote alloy formation.
338
Ch. Venkateswara Rao, B. Viswanathan / Journal of Colloid and Interface Science 367 (2012) 337–341
2.2. Characterization techniques XRD measurements were performed on a Rigaku Miniflex X-ray diffractometer using a Cu Ka source at a scan rate of 0.025 s1. Scanning electron microscope equipped with EDX (FEI, Model: Quanta 200) was used to observe the composition and distribution of elements in the catalysts. Transmission electron microscopic images were obtained using JEOL2100 system. Ultraviolet photoemission spectroscopy (UPS) measurements of the catalysts was carried out with Omicron nanotechnology instrument using an Al monochromatic X-ray (hm = 1486.6 eV) and operated under the base pressure of 62 109 mbar. The powder samples were mounted onto the holder, and the catalyst surface was scraped in situ by repeated cycles of Ar-ion sputtering to remove any surface contamination that could arise from atmospheric components like water and CO2. The energy resolution of the spectrometer was set at 1.1 eV at pass energy of 80 eV. The binding energy (BE) was calibrated with respect to Ag 3d core level. For the high-resolution valence-band photoemission spectra, the total energy resolution, including the photon energy resolution, was less than 0.05 eV. 2.3. Electrochemical measurements Electrocatalytic activity of the materials was investigated by cyclic voltammetry using a potentiostat (BAS 100 electrochemical analyzer). All experiments were performed in a one-compartment cell assembled with Ag/AgCl as reference, Pt foil as counter, and glassy carbon disk as working electrodes, respectively. A thin-film rotating disk glassy carbon electrode (Pine Instruments) technique is employed to investigate the ORR activity of the materials. The working electrode was fabricated as follows: 5 mg of catalyst was dispersed in 5 ml of isopropanol by ultrasonication for 15 min. An aliquot of 20 ll catalyst ink was pipetted onto the mirror-polished glassy carbon substrate and dried in flowing argon at room temperature. Then, 10 ll of diluted 5 wt.% NafionÒ solution was pipetted on the electrode surface in order to attach the catalyst particles onto the GC disk and dried. For oxygen reduction measurements, linear sweep voltammograms (LSVs) were recorded between +0.0 and +1.2 V vs NHE at a scan rate of 5 mV/ s and rotation rate of 1600 rpm in O2-saturated 0.5 M H2SO4. Current densities are normalized to the geometric area of the glassy carbon substrate in the text. 3. Results and discussion Microemulsion method was employed as a suitable method to generate metal colloids and/or clusters on the nanoscale with controllable composition. The diameter and morphology of nanoparticles can be controlled by varying the concentration and type of surfactant, concentration of precursor and reducing agents, type of solvent, and ultrasonication [21–23]. Under the experimental conditions, Pd–Co–Au nanoparticles were formed upon contact between the precursor containing droplets and the hydrazine containing droplets. The reductions of H2PdCl4, CoCl2, and HAuCl4 are as follows:
2H2 PdCl4 þ 7N2 H4 H2 O ! 2Pd þ 8NH4 Cl þ 3N2 þ 7H2 O CoCl2 þ 2N2 H4 H2 O ! Co þ 2NH4 Cl þ N2 þ 2H2 O HAuCl4 þ 4N2 H4 H2 O ! Au þ 4NH4 Cl þ 2N2 þ 4H2 O Powder XRD patterns recorded for the carbon-supported Pd and Pd–Co–Au alloy materials are shown in Fig. 1. Both the Pd/C and as-synthesized Pd–Co–Au/C exhibited diffraction peaks at 2h values around 40°, 47°, 68°, 82°, and 87° corresponding to the (1 1 1),
Fig. 1. Powder XRD pattern of (a) Pd/C, (b) as-synthesized Pd–Co–Au/C; heattreated Pd–Co–Au/C at (c) 700, (d) 800, and (e) 900 °C, respectively.
(2 0 0), (2 2 0), (3 1 1), and (2 2 2) planes of face-centered cubic structure of Pd (JCPDS No. 88-2335). The broad peaks indicate that the particles are in nanocrystalline range. The diffraction peak at 2h value around 25° corresponds to the (0 0 2) plane of the hexagonal carbon support. No appreciable change in the XRD pattern of the as-synthesized Pd–Co–Au/C compared to the Pd/C indicates that the Pd/Co/Au metal nanoparticles are not transformed into alloy. In order to form alloy, the as-synthesized Pd–Co–Au/C is heat-treated at 700, 800, and 900 °C. With increasing temperature, the Pd diffraction peaks are shifted to higher angles in Pd–Co–Au/C compared to the Pd/C. It indicates the contraction of Pd lattice due to the alloy formation. The extent of shift increases and the lattice parameter decreases with increasing heat-treatment temperature, suggesting an increase in the degree of alloying. By using lattice parameter values, degree of alloying DA is calculated by using the equation: fa a g DA ¼ a Pda alloyðTÞ 100; where aPd, aalloy(T), and aalloy(900) refer to f Pd alloyð900Þ g the cell parameter values of Pd, Pd–Co–Au alloy heat-treated at various temperatures (T = 700 and 800 °C), and Pd–Co–Au alloy heattreated at 900 °C, respectively [18]. The DA values are given in Table 1. It indicates the increase in degree of alloying with heattreatment temperature. TEM images of the Pd/C and heat-treated Pd–Co–Au/C catalysts are shown in Fig. 2. The images depict a relatively narrow size distribution of Pd nanoparticles over carbon support in the case of Pd/C whereas a relatively broader distribution of Pd alloy nanoparticles over carbon support in the case of Pd–Co– Au/C catalysts. The average particle sizes obtained from the measurement of about 200 particles found in an arbitrarily chosen area of enlarged micrographs by assuming spherical shape of particles are given in Table 1. Fig. 2a shows the good dispersion of spherical-shaped Pd nanoparticles of average size 2.1 nm with a standard deviation of 0.2 nm. The estimated particle sizes for the heat-treated Pd–Co–Au/C at 700, 800, and 900 °C are 23 ± 8, 31 ± 13, and 42 ± 17 nm, respectively. EDX confirms the presence of Pd and C in Pd/C and Pd, Co, Au, and C in Pd–Co–Au/C, respectively. The calculated elemental composition in the heat-treated Pd–Co–Au/C catalysts is corresponding to the approximate atomic ratio of 7:2:1 (Table 1). Scanning electron microscopic images with EDX mapping for the Pd/C and heat-treated Pd–Co–Au/C at 800 °C are shown in Fig. 3. The EDX mapping images depicted the homogeneous distribution of respective elements in the catalysts.
339
Ch. Venkateswara Rao, B. Viswanathan / Journal of Colloid and Interface Science 367 (2012) 337–341 Table 1 Degree of alloying, average particle size (from TEM), EDX composition, onset potential, and ORR activity of the catalysts. Catalyst
Degree of alloying (%)
Particle size (nm)
EDX composition Pd:Co:Au (wt.%)
Onset potential (mV vs NHE)
Activity at +0.8 V (mA/ cm2)
As-syn Pd/C As-syn Pd–Co–Au/C HT Pd–Co–Au/C at 700 °C HT Pd–Co–Au/C at 800 °C HT Pd–Co–Au/C at 900 °C
– – 59
2.1 – 23
100:–:– – 70.2:11.2:18.6
+845 – +885
0.15 – 0.53
79
31
71.3:11.6:17.1
+920
1.32
100
42
71.8:11.3:16.9
+915
0.92
(HT = heat-treated).
(a)
(b)
50 nm
20 nm
(c)
50 nm
(d)
100 nm
Fig. 2. TEM images of (a) Pd/C; heat-treated Pd–Co–Au/C at (b) 700, (c) 800, and (d) 900 °C, respectively.
Cyclic voltammograms (CVs) of the Pd/C and Pd–Co–Au/C catalysts in deaerated 0.5 M H2SO4 are shown in Fig. 4. The redox behavior of the Pd/C is found to be consistent with literature reports [24,25,14]. The redox peaks observed in the +0.15 0.25 V region correspond to the adsorption/desorption of hydrogen on the Pd surface and formation of the Pd hydride phase. The redox peaks in the 0.6–1.2 V region corresponding to the formation and reduction of surface Pd oxides (Pd–OH or PdOx species). It is known that the PdOx species at the metallic Pd surface is inactive toward oxygen reduction [26]. For the as-synthesized Pd–Co–Au/C catalyst, anodic peak was shifted to lower potentials of about 0.14 V compared to Pd/C. It shows the modification of the electronic structure of Pd by the presence of Co and Au. In the case of heattreated Pd–Co–Au/C, hydrogen adsorption/desorption peaks are appeared at potentials of 0.04 V showed a similar behavior to that of Pt [24,27]. This is due to the interruption of the dissolution of
hydrogen into the Pd lattice by alloying elements. Moreover, the formation of surface oxy species, which are the poisoning species for oxygen reduction, was not observed, and this may be beneficial to the oxygen adsorption at low overpotential and thus the ORR kinetic enhancement. Linear sweep voltammograms (LSVs) of the Pd/C and Pd–Co– Au/C catalysts in both Ar- and O2-saturated 0.5 M H2SO4 is shown in Fig. 5. When the potential was swept from +1.2 to +0.2 V, single oxygen reduction peak was observed in the potential region of 1.0– 0.6 V. The calculated ORR activity of the catalysts is given in Table 1. The results indicate the good ORR performance of Pd–Co–Au/C catalysts compared to Pd/C. This is due to the suppression of (hydr)oxy species on Pd electrode surface by the presence of alloying elements. Among the heat-treated Pd–Co–Au/C catalysts, maximum activity was observed for the heat-treated Pd–Co–Au/C at 800 °C compared to 700 and 900 °C. The initial increase is due
340
Ch. Venkateswara Rao, B. Viswanathan / Journal of Colloid and Interface Science 367 (2012) 337–341
Pd
(a)
300 nm Pd
(b)
300 nm Co
Au
Fig. 3. EDX mapping of elements in the catalysts: (a) Pd/C and (b) heat-treated Pd–Co–Au/C at 800 °C.
Fig. 4. Cyclic voltammograms (CVs) of carbon-supported Pd and Pd–Co–Au catalysts in Ar-saturated 0.5 M H2SO4; scan rate—25 mV/s.
to an increase in the degree of alloying while the decrease at higher temperatures is due to an increase in crystallite size. Also, ORR proceeded in a relatively positive potential region for the Pd–Co–Au/C catalysts compared to Pd/C. For each of the investigated materials, the onset potential for ORR – the potential where the current density increases to 20 lA/cm2 – is given in Table 1. The overpotential for ORR on the best Pd–Co–Au alloy was
Fig. 5. Linear sweep voltammograms (LSVs) of O2 reduction on carbon-supported Pd and Pd–Co–Au catalysts in 0.5 M H2SO4; scan rate—5 mV/s and rotation rate— 1600 rpm.
ca.75 mV less compared to Pd. The differences in the onset potential for oxygen reduction at various heat-treated Pd–Co–Au/ C catalysts may be due to differences in the surface activation, which are related to the size and distribution of the metallic nanoparticles.
Ch. Venkateswara Rao, B. Viswanathan / Journal of Colloid and Interface Science 367 (2012) 337–341
341
performance. The good activity of the Pd–Co–Au/C compared to Pd/C makes it as applicable cathode for fuel cell applications. Acknowledgments The authors wish to acknowledge Columbian Chemicals Company, USA, for financial support and also Department of Science and Technology (DST), India, for creating National Centre for Catalysis Research at IIT Madras. References
Fig. 6. Valence-band spectra of (a) Pd/C and (b) heat-treated Pd–Co–Au/C at 800 °C.
The voltammetric behavior of carbon-supported Pd and Pd–Co– Au alloys indicates the change in the electronic structure of Pd by alloying elements. To investigate the surface electronic properties of Pd, valence-band spectra of Pd/C and heat-treated Pd–Co–Au/C at 800 °C is recorded and shown in Fig. 6. It depicts the shift of Pd d-band center from 2.27 eV on Pd/C to 3.04 eV on Pd–Co–Au/ C. A similar downshift was observed for a variety of bimetallic Pt alloys [28]. The downshift of d-band center indicates the decreased density of states at the Fermi level of Pd in the Pd–Co–Au/C. As a result, the formation of Oads or OHads on Pd surface was inhibited (evident from CV behavior shown in Fig. 4) and leads to the good ORR activity (evident from LSV behavior shown in Fig. 5). 4. Conclusions Carbon-supported Pd–Co–Au alloys were synthesized by microemulsion method and characterized. XRD showed the single phase face-centered cubic structure of Pd in the catalysts. EDX analysis confirmed the atomic composition of 7:2:1 in the heat-treated Pd–Co–Au/C catalysts. Voltammograms and valence-band spectra measurements revealed the inhibition of (hydr)oxy species on Pd surface by the presence alloying elements enhanced the ORR
[1] B. Viswanathan, M. Aulice, Scibioh (Eds.), Fuel Cells – Principles and Applications, Universities Press (India) Private Limited, 2006. [2] H.A. Gasteiger, N.M. Markovic, Science 324 (2009) 48–49. [3] Ch. Venkateswara Rao, B. Viswanathan, J. Phys. Chem. C 113 (2009) 18907– 18913. [4] Ch. Venkateswara Rao, B. Viswanathan, J. Phys. Chem. C 114 (2010) 8661– 8667. [5] D.B. Sepa, M.V. Vojnovic, A. Damjanovic, Electrochim. Acta 26 (1981) 781–793. [6] Lj.M. Vracˇar, D.B. Sepa, A. Damjanovic, J. Electrochem. Soc. 133 (1986) 1835– 1839. [7] Lj.M. Vracˇar, D.B. Sepa, A. Damjanovic, J. Electrochem. Soc. 136 (1989) 1973– 1977. [8] F.G. Will, C.A. Knorr, Z. Electrochem. 64 (1960) 258–269. [9] K. Kinoshita, Electrochemical Oxygen Technology, John Wiley & Sons, New York, 1992. [10] J.L. Fernández, D.A. Walsh, A.J. Bard, J. Am. Chem. Soc. 127 (2005) 357–365. [11] M.H. Shao, K. Sasaki, R.R. Adzic, J. Am. Chem. Soc. 128 (2006) 3526–3527. [12] Y.X. Wang, P.B. Balbuena, J. Phys. Chem. B 109 (2005) 18902–18906. [13] K. Lee, O. Savadogo, A. Ishihara, S. Mitsushima, N. Kamiya, K.I. Ota, J. Electrochem. Soc. 153 (2006) A20–A24. [14] Ch. Venkateswara Rao, B. Viswanathan, Electrochim. Acta 55 (2010) 3002– 3007. [15] J.L. Fernández, J.M. White, Y. Sun, W. Tang, G. Henkelman, A.J. Bard, Langmuir 22 (2006) 10426–10432. [16] J.L. Fernández, V. Raghuveer, A. Manthiram, A.J. Bard, J. Am. Chem. Soc. 127 (2005) 13100–13101. [17] V. Raghuveer, A. Manthiram, A.J. Bard, J. Phys. Chem. B 109 (2005) 22909– 22912. [18] V. Raghuveer, P.J. Ferreira, A. Manthiram, Electrochem. Commun. 8 (2006) 807–814. [19] B. Hammer, J.K. Nørskov, Adv. Catal. 45 (2000) 71–129. [20] Y. Guo, L. Zhuang, J. Lu, Angew. Chem. Int. Ed. 46 (2007) 2862–2864. [21] K. Holmberg, J. Colloid Interface Sci. 274 (2004) 355–364. [22] R. Najjar, C. Stubenrauch, J. Colloid Interface Sci. 331 (2009) 214–220. [23] Ch. Venkateswara Rao, B. Viswanathan, J. Phys. Chem. C 111 (2007) 16538– 16543. [24] M. Enyo, T. Yamazaki, K. Kai, K. Suzuki, Electrochim. Acta 28 (1983) 1573– 1579. [25] A.J. Bard, Electroanalytical Chemistry, Marcel Dekker Inc., New York, 1976. [26] J. Lipkowski, P.N. Ross, Electrocatalysis, Wiley-VCH, New York, 1998. [27] J.P. Hoare, S. Schuldiner, J. Phys. Chem. 62 (1958) 229–233. [28] B.S. Mun, M. Watanabe, M. Rossi, V. Stamenkovic, N.M. Markovic, P.N. Ross Jr., J. Chem. Phys. 123 (2005) 204717–204721.