Contents lists available atScienceDirect
Journal of Alloys and Compounds
j o u r n a l h o m e p a g e :w w w . e l s e v i e r . c o m / l o c a t e / j a l l c o mEffect of anodic dissolution in multi-element nanoparticles on
methanol electro-oxidation
Yi-Fan Hsieh, Kung-Yu Yeh, Pu-Wei Wu
∗, Yu-Chi Hsieh, Pang Lin
Department of Materials Science and Engineering, National Chiao Tung University, 1001 Ta Hsueh Road, Hsin-Chu 30010, Taiwan, ROC
a r t i c l e i n f o
Article history:Received 6 March 2009
Received in revised form 6 July 2009 Accepted 7 July 2009
Available online 16 July 2009
Keywords:
Electrode materials Metals
Electrochemical reactions
a b s t r a c t
Radio frequency sputter deposition was used to deposit multi-element nanoparticles on carbon clothes for methanol electro-oxidation. Results from energy dispersive spectroscope confirmed the composition was Pt35Fe12Co12Ni20Cu12Ag9. X-ray diffraction pattern suggested a fcc phase. An electrochemical
de-alloying treatment was conducted in an acidic electrolyte with multiple potential sweeps at a scan rate of 20 mV/s for 0–0.3, 0–0.5, and 0–0.7 V, respectively. Images from scanning electron microscope after anodic dissolutions demonstrated considerable variations in surface morphologies. Measurements on hydrogen desorption also revealed substantial increase in electrochemical active surface area. Current responses from cyclic voltammetry indicated the sample of 0–0.5 V anodic dissolution with notable enhancement. In mass activities, the same sample exhibited a comparable value to that of Pt43Ru57.
© 2009 Elsevier B.V. All rights reserved.
1. Introduction
Materials to promote methanol electro-oxidations have been investigated intensively in the past decades because their develop-ments are critical for direct methanol fuel cells (DMFCs)[1,2]. Since electrolytes in DMFCs are corrosive in nature, conventional candi-dates for electrocatalysis are noble metals or oxides because their recognized chemical stabilities ensure reasonable lifetime[3–5]. Among them, the Pt and alloys of Pt have received considerable attentions, as evidenced by extensive reports covering a variety of compositions and morphologies with impressive electrochem-ical behaviors[6–8]. In general, the electro-oxidation of methanol involves successive de-hydrogenation steps, leaving CO chemically bond to the Pt surface. Therefore, secondary elements responsible for oxidizing the adsorbed CO are necessary. Currently, it is estab-lished that Ru is effective in removing the CO, and nanoparticles of PtRu reveal the highest catalytic ability for methanol electro-oxidations[9,10]. Unfortunately, due to excessive cost for Pt and Ru, search of less expensive elements with comparable performances is still in pursuit.
The simplest way to improve catalytic performances is to increase the catalyst loading. However, this approach often leads to undesirable material costs. An alternative route is to synthe-size electrocatalysts with large active areas. One of the methods under studies is the electrochemical de-alloying process in which elements with relatively more positive redox potentials are
electro-∗ Corresponding author. Tel.: +886 3 5131227; fax: +886 3 5724727.
E-mail address:[email protected](P.-W. Wu).
chemically removed under deliberate anodic polarizations, leaving elements with less positive redox potentials mostly intact[11,12]. For example, Liu et al. electrodeposited a PtCu film and dissolved the Cu potentiostatically at 0.35 V, rendering a nanoporous Pt film with significantly enhanced surface area[13]. A similar process is adopted by Koh and Strasser where nanoparticles of Pt25Cu75were converted to Pt79Cu21by repeated cyclic voltammetric scans in an acidic electrolyte[14]. Substantial activity enhancements for the oxygen reduction reaction were observed, which are attributed to a favorable structural arrangement of Pt atoms at the particle surface. The de-alloying process becomes rather complicated once nanoparticles contain three or more elements. It is because their redox potentials vary, and distinct morphologies and compositions are possible contingent on the anodizing potentials imposed. For instance, Srivastava et al. developed a three-step procedure to de-alloy Cu from nanoparticulate PtCoCu, and reported impressive oxygen reduction behaviors[15]. Formation of metals with multi-ple principal elements was first demonstrated by Yeh et al. with the objective to develop high-entropy alloys[16]. On the other hand, the synthesis of multi-element nanoparticles has received less atten-tion. Recently, in our group we employed a radio frequency sputter deposition to prepare multi-element nanoparticles (PtFeCoNiCuAg) in various compositions, and studied their catalytic abilities on methanol electro-oxidations[17,18]. We realize that due to sub-stantial difference in their redox potentials, further improvements in electrocatalysis are likely if the de-alloyed process is properly carried out.
In this work, we fabricated multi-element nanoparticles and conducted de-alloying treatments at various potentials to enhance active surface areas for methanol electro-oxidation.
0925-8388/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jallcom.2009.07.055
2. Experimental
2.1. Deposition of the multi-element nanoparticles
Multi-element nanoparticles were deposited on carbon clothes (E-TEK) using a radio frequency sputter. The as-received carbon cloth was coated with carbon slurry composed of 70 wt% Shawinigan Acetylene Black (Chevron) and 30 wt% poly-tetrafluoroethylene (Dupont) to reach a total dry weight of 22 mg/cm2. To prepare
sputter target, metallic powders including Fe (10.3m), Co (1.4 m), Ni (2.5 m), Cu (45.0m), and Ag (0.8 m) were mixed at a molar ratio of 24:22:21:15:18. After dry tumbling for 24 h under N2, the mixture underwent a hydraulic pressing at 2500 psi
for 30 s to make a disk with 3.0 in. in diameter. Next, Pt foils (99.9 wt%, 1 cm× 1 cm) were positioned atop the disk to add the Pt atoms into the sputtered flux. The size for the carbon cloth was 4 cm× 4 cm, and the sputter deposition lasted 5 min. The carbon cloth underwent a rotating motion to obtain uniform coverage during sput-ter deposition. Details processing steps and a schematic for the experimental setup were reported previously[17].
2.2. Materials characterizations for the multi-element nanoparticles
Identification for relevant phase present in the as-prepared multi-element film was conducted by a XRD with Cu K␣ radiation ( = 1.5418 Å) (Siemens D5000). The specimen was subjected to a sputter deposition for 120 min to ensure sufficient loading. SEM (JSM-6500F) was used to observe morphologies of the multi-element nanoparticles before and after anodic dissolutions. An energy dispersive X-ray spec-troscope (EDX) was employed to obtain the molar ratio for the multi-element nanoparticles. In mass activity determination, an inductively coupled plasma mass spectrometry (ICP-MS; SCIEX ELAN 5000) was adopted where the samples were dissolved in a solution containing HCl, HNO3, and HF at a 2:2:1 volume ratio.
2.3. Anodic dissolutions and electrochemical analysis
Electrochemical characterizations were performed using a Solartron SI 1287 at 26◦C in a three-electrode configuration in which a Pt foil (8 cm2) and Ag/AgCl were
used as counter and reference electrodes, respectively. The area for the working electrode was 1 cm2. Anodic dissolutions for the multi-element nanoparticles were
carried out in 1 M H2SO4aqueous solution in linear voltage sweeps for 100 cycles at
a scan rate of 20 mV/s. The voltage ranges selected were 0–0.3, 0–0.5, and 0–0.7 V. Upon completion, the samples were immersed in an aqueous solution consisted of 0.5 M H2SO4and 1 M CH3OH to study their catalytic abilities on the methanol
electro-oxidation. Before measurements, the electrolyte was purged with N2for 15 min and
waited for 30 min to stabilize the open circuit voltage. Then, cyclic voltammetry (CV) was adopted in a voltage range of 0–0.95 V at a scan rate 50 mV/s. In addition, we obtained the values for electrochemically active surface area (ESA) by conducting hydrogen desorption analysis. The ESA was obtained by a CV scan for a voltage range of−0.2 to 1.0 V in a 0.5 M H2SO4aqueous solution at a scan rate of 10 mV/s.
3. Results and discussion
3.1. Characterizations of multi-element nanoparticles
Since the sputtering yield varies considerably in each element, the atomic ratio for the multi-element nanoparticles can only be determined empirically. Results from multiple EDX analysis on rel-atively large area SEM samples confirmed the composition was Pt35Fe12Co12Ni20Cu12Ag9. We realized that the sputter deposition technique might produce minute local variation in the deposit composition. But the EDX provided the statistical average for the as-prepared samples so the composition of Pt35Fe12Co12Ni20Cu12Ag9 should be reliable. The XRD pattern on the multi-element nanopar-ticles after 120 min of sputter deposition is demonstrated inFig. 1. Due to interference from the carbon cloth, the diffraction pattern contained moderate noises. Nevertheless, we were able to identify diffraction signals from Pt, C, and polytetrafluoroethylene (PTFE). Notably, the Pt existed in a fcc lattice with relative intensity in proper order. The positions for the Pt (1 1 1), Pt (2 0 0), Pt (2 2 0), and Pt (3 1 1) were located at 39.8◦, 46.2◦, 67.5◦, and 81.3◦, respectively. These angles were slightly shifted as opposed to the unalloyed Pt lattice. Since the atomic ratio for Fe, Co, Ni, and Cu are smaller than that of Pt, the resulting lattice parameter for the multi-element nanoparticles was reduced as expected following Vegard’s law.
Fig. 2 exhibits the SEM images for the carbon cloth and as-prepared multi-element nanoparticles. As shown inFig. 2(a), the carbon cloth was decorated with carbon particles in irregular shape,
Fig. 1. X-ray diffraction pattern for the multi-element nanoparticles after 120 min
of sputter deposition.
and their average size was approximately 40 nm. After deposition for 5 min, as shown inFig. 2(b), there appeared surface nodules on individual carbon particles and the average particle size was increased to 80 nm. The sputter rate for the multi-element nanopar-ticles was estimated at 4 nm/min.
3.2. Effects of anodic dissolutions
Anodic dissolutions conducted in various voltage regimes are expected to produce distinct de-alloying behaviors because the redox potentials for individual elements differ considerably. In an acidic electrolyte, elements with a positive redox potential are
Fig. 2. SEM images for the carbon cloth (a) before sputter deposition, and (b) 5 min
Table 1
Values for standard redox potentials for Pt, Fe, Co, Ni, Cu, and Ag. The H2/H+represents 0 V.
Reaction E0(V) Pt(s)→ Pt2+(aq)+ 2e −1.20 Ag(s)→ Ag+(aq)+ e −0.80 Cu(s)→ Cu2+(aq)+ 2e −0.34 H2(g)→ 2H+(aq)+ 2e 0.00 Ni(s)→ Ni2+(aq)+ 2e 0.25 Co(s)→ Co2+(aq)+ 2e 0.28 Fe(s)→ Fe2+(aq)+ 2e 0.44
susceptible to spontaneous dissolutions without external stimuli. The tendency for dissolution becomes more pronounced when the redox potential becomes more positive. On the other hand, elements with a negative one infer chemical stability unless the anodizing voltage sufficiently compensates the negative redox value.Table 1lists the redox potentials for the Pt, Fe, Co, Ni, Cu, Ag, and H2. Among the elements present, the order for anodic disso-lutions is Fe, Co, Ni, Cu, Ag, and Pt. Since we performed the anodic dissolution in multiple linear sweeps, a larger voltage sweep not only suggested stronger anodizing power but also indicated longer immersion time in the electrolyte.
The compositions for the multi-element nanoparticles after anodic dissolutions are provided in Table 2. Notably, all sam-ples became Pt-rich. For samsam-ples undergoing anodic dissolution in 0–0.3 V, we observed complete loss of Fe and Co, with the resulting composition confirmed as Pt48Ni18Cu17Ag17. For samples experi-encing anodic dissolution in 0–0.5 V, additional Cu and Ag were dissolved and the composition was Pt86Ni14. Further anodization for samples in 0–0.7 V resulted in more Ni loss as the composition became Pt89Ni11. The presence of Ni after anodic dissolutions was likely due to its preponderance in the as-deposited state.
Fig. 3 demonstrates the SEM images for the multi-element nanoparticles after anodic dissolutions of 0–0.3, 0–0.5, and 0–0.7 V, respectively. Apparently, the size for the multi-element nanoparti-cles became smaller when the anodic voltage was increased. For the voltage ranges in 0–0.3 and 0–0.5 V, the particle size was reduced to 70 and 60 nm, respectively. Interestingly, distinct morphological changes were observed when the anodizing voltage was raised to 0.7 V where the as-prepared nodular structure was transformed to textural surface with moderate coalescences. However, the exact nature for this phenomenon is still unclear.
Electrodes with altered morphologies often can be confirmed by ESA measurements. Fig. 4 exhibits the ESA curves for the as-prepared nanoparticles and samples after anodic dissolutions. According to Navessin et al. and Liu et al., appearance of anodic and cathodic spikes in−0.2 to 0 V are attributed to desorption and absorption of hydrogen[19,20]. Since the hydrogen desorp-tion takes place exclusively on surface Pt sites, the ESA of Pt is estimated by integrating the coulombic charge in the desorption regime. The as-prepared multi-element nanoparticles revealed a ESA of 15.5 cm2. Remarkably, once anodic dissolutions were per-formed, the resulting ESA was increased significantly. Their values were 26.1, 36, and 34.2 cm2for samples undergoing 0–0.3, 0–0.5, and 0–0.7 V treatments. Despite the notable difference in surface morphologies, the ESA from anodization of 0–0.7 V was similar Table 2
Compositions from EDS in molar ratios for multi-element nanoparticles with and without anodic dissolutions.
Pt Fe Co Ni Cu Ag Total (%)
As-prepared 35 12 12 20 12 9 100
0–0.3 V 48 0 0 18 17 17 100
0–0.5 V 86 0 0 14 0 0 100
0–0.7 V 89 0 0 11 0 0 100
Fig. 3. SEM images for the Pt35Fe12Co12Ni20Cu12Ag9after anodic dissolutions in (a)
0–0.3, (b) 0–0.5, and (c) 0–0.7 V, respectively.
Fig. 5. CV curves in apparent current densities for the Pt35Fe12Co12Ni20Cu12Ag9after
anodic dissolutions of 0–0.3, 0–0.5, and 0–0.7 V, respectively.
to that of 0–0.5 V. This indicated anodic dissolution conducted at 0–0.5 V is adequate for ESA enhancement without unnecessary material loss.
3.3. Characterizations on methanol electro-oxidation
Fig. 5presents the CV profiles for the multi-element nanopar-ticles after anodic dissolutions of 0–0.3, 0–0.5, and 0–0.7 V, respectively. In each sample, there appeared current responses during forward and backward scans. According to literature, cur-rents from the forward scan are due to the oxidation of methanol while those of backward scan are attributed to the oxidation of car-bonaceous species produced from the forward scan[21]. The peak current densities during forward and backward scans are defined as ifand ib. Its ratio, if/ib, indicates the catalytic ability to remove CO [22]. As shown in the CV profiles, the sample undergoing 0–0.5 V anodic dissolution revealed the highest current responses. In con-trast, the one with 0–0.3 V showed the lowest one. Their if/ibratios were between 1.10 and 1.29, which is consistent with what was reported earlier on Pt-based electrocatalysts[23].
Because anodic dissolutions inevitably resulted in weight loss, true catalytic abilities can only be evaluated by determining their mass activities (mA/mg). For this purpose, we also conducted simi-lar sputter depositions to prepare Pt and Pt43Ru57-catalyzed carbon clothes.Fig. 6provides the CV profiles in mass activities for the multi-element nanoparticles in as-prepared state and after anodic dissolution of 0–0.5 V. The curves of Pt and Pt43Ru57 were also shown here for comparison purpose[18]. Relevant electrochem-ical parameters are listed inTable 3. Among them, the Pt43Ru57 exhibited the highest mass activity with a if/ibratio of 1.57. These behaviors confirmed the superiority of PtRu as an effective catalyst
Fig. 6. CV curves in mass activities for the Pt35Fe12Co12Ni20Cu12Ag9in as-prepared
state, and anodic dissolution of 0–0.5 V, as well as Pt and Pt43Ru57.
Table 3
Electrochemical parameters from the CV scans in mass activities.
Pt loading (mg) Forward Backward if/ib Vfa(mV) ifb (mA/Pt mg) Vbc(mV) ibd (mA/Pt mg) As-prepared 0.11 0.69 168.57 0.53 160.29 1.05 0–0.5 V 0.04 0.66 346.92 0.48 315.38 1.10 Pt 0.16 0.66 100.32 0.59 122.13 0.82 Pt43Ru57 0.07 0.67 368.00 0.50 234.00 1.56
aPeak potential in forward scan. bPeak mass activity in forward scan. c Peak potential in backward scan. dPeak mass activity in backward scan.
for methanol electro-oxidation, a fact well-established in litera-ture. Interestingly, the sample after anodic dissolution of 0–0.5 V exhibited a comparable anodic mass activity. However, its if/ibratio was still inferior to that of Pt43Ru57. On the other hand, samples of as-prepared multi-element nanoparticles and Pt demonstrated reduced catalytic behaviors in both mass activities and if/ibratios. These results clearly demonstrate the positive effects of anodic dissolutions in enhancing the catalytic abilities of multi-element nanoparticles.
4. Conclusions
Sputter deposition was employed to prepare multi-element nanoparticles for methanol electro-oxidation. EDS determined the composition as Pt35Fe12Co12Ni20Cu12Ag9 and XRD pattern indicated a fcc phase. Electrochemical de-alloying process was performed by multiple voltage sweeps in an acidic electrolyte at potentials of 0–0.3, 0–0.5, and 0–0.7 V. After anodic dissolutions, the surface morphologies were altered with substantial increase in ESA. In mass activities, the sample after 0–0.5 V anodic dissolution revealed a comparable mass activity to that of Pt43Ru57, albeit with a lower if/ib. Nevertheless, its catalytic abilities were improved sig-nificantly over those of as-prepared multi-element nanoparticles and Pt.
Acknowledgements
The authors would like to express their gratitude to Profes-sor Chiun-Hsun Chen of Mechanical Engineering Department for equipment supports. Discussions over multi-element alloy forma-tions with Professor Cheun-Guang Chao of Materials Science and Engineering Department is appreciated.
References
[1] B.C.H. Steele, A. Heinzel, Nature 414 (2001) 345.
[2] Z.B. Wang, G.P. Yin, P.F. Shi, J. Alloys Compd. 420 (2006) 126.
[3] H.N. Dinh, X. Ren, F.H. Garzon, P. Zelenay, S. Gottesfeld, J. Electroanal. Chem. 491 (2000) 222.
[4] N. Ren, A.G. Dong, W.B. Cai, Y.H. Zhang, W.L. Yang, S.J. Huo, Y. Chen, S.H. Xie, Z. Gao, Y. Tang, J. Mater. Chem. 14 (2004) 3548.
[5] H.J. Ahn, J.S. Jang, Y.E. Sung, T.Y. Seong, J. Alloys Compd. 473 (2009) L28. [6] H.J. Ahn, H.S. Shim, W.B. Kim, Y.E. Sung, T.Y. Seong, J. Alloys Compd. 471 (2009)
L39.
[7] J.F. Whitacre, T. Valdez, S.R. Narayanan, J. Electrochem. Soc. 152 (2005) A1780. [8] Y.J. Song, S.B. Han, J.M. Lee, K.W. Park, J. Alloys Compd. 473 (2009) 516. [9] T. Yajima, H. Uchida, M. Watanabe, J. Phys. Chem. B 108 (2004) 2654. [10] P. Waszczuk, G.Q. Lu, A. Wieckowski, C. Lu, C. Rice, R.I. Masel, Electrochim. Acta
47 (2002) 3637.
[11] J. Erlebacher, M.J. Aziz, A. Karma, N. Dimitrov, K. Sieradzki, Nature 410 (2001) 450.
[12] Z. Liu, S. Koh, C. Yu, P. Strasser, J. Electrochem. Soc. 154 (2007) B1192. [13] H. Liu, P. He, Z. Li, J. Li, Nanotechnology 17 (2006) 2167.
[14] S. Koh, P. Strasser, J. Am. Chem. Soc. 129 (2007) 12624.
[16] J.W. Yeh, S.K. Chen, S.J. Lin, J.Y. Gan, T.S. Chin, T.T. Shun, C.H. Tsau, S.Y. Chang, Adv. Eng. Mater. 6 (2004) 299.
[17] C.F. Tsai, P.W. Wu, P. Lin, C.G. Chao, K.Y. Yeh, Jpn. J. Appl. Phys. 47 (2008) 5755.
[18] C.F. Tsai, K.Y. Yeh, P.W. Wu, Y.F. Hsieh, P. Lin, J. Alloys Compd. 478 (2009) 868. [19] T. Navessin, M. Eikerling, Q. Wang, D. Song, Z. Liu, J. Horsfall, K.V. Lovell, S.
Holdcroft, J. Electrochem. Soc. 152 (2005) A796.
[20] Z. Liu, X.Y. Ling, B. Guo, L. Hong, J.Y. Lee, J. Power Sources 167 (2007) 272. [21] Y.J. Gu, W.T. Wong, Langmuir 22 (2006) 11447.
[22] J. Bagchi, S.K. Bhattacharya, J. Power Sources 163 (2007) 661.
[23] W. Jingjie, T. Haolin, P. Mu, W. Zhaohui, M. Wentao, Electrochim. Acta 54 (2009) 1473.
