Electrocatalysis of oxygen reduction reaction on polyaniline-derived nitrogen-doped carbon nanoparticle surfaces in alkaline media

Electrocatalysis of the Oxygen Reduction Reaction on Pd0.5M0.4Pt0.1 (M=Cu,Ni,Fe,Co) Nanoparticles

F. Godínez-Salomóna,b, J.M Hallen Lópezb, O. Solorza-Feriaa

a Centro de Investigación y Estudios Avanzados del IPN, Depto. Química, Av. IPN 2508.

14-740, 07360 D.F. México, México. b Departamento de Ingeniería Metalúrgica, IPN-ESIQIE, UPALM Edif. 7, Zacatenco,

07738 D.F. México, México.

This work presents the study of the electrocatalytic activity of the oxygen reduction reaction (ORR) in 0.5M H2SO4 on dispersed nanoparticles of different ternary Pd0.5M0.4Pt0.1 (M = Cu, Ni, Fe and Co) alloys, through thin-film rotating disk electrode (TF-RDE). Transmission electron microscopy (TEM) showed homogeneous distribution of nanoparticles between 17-45 nm in sizes. The catalysts containing Co, Ni or Fe, show equivalent behavior to Pd alone, however, the Pd0.5Cu0.4Pt0.1 catalyst showed similar specific activity (µA cm2 real) but twice specific mass activity (mA mg-1) compared to Pt synthesized under the same experimental condition.

Introduction Electrocatalysis with nanoparticles continues to attract widespread attention in many scientific fields including on site applications. The development of novel catalysts with high activity for the oxygen reduction reaction, ORR, continues to be an important area of research in polymer electrolyte membrane fuel cells (PEMFCs) (1-2). The electrocatalytic activity associated to a sluggish of ORR is one of the hindrances in developing of fuel cells as energy sources. Pt and its alloys, due to their electrocatalytic characteristics continue to be the best choice to promote the ORR, however their use is limited by their cost and natural resources. These problems have motivated considerable interest in the developing of new alternatives of nanocatalysts for the ORR. Although an outstanding progress has been observed for the synthesis of novel catalysts without noble metals content (3), these kinds of materials present reduced activity and stability compared with analogous of Pt. An alternative is replacing Pt for another less expensive noble metal which presents considerable catalytic activity and electrochemical stability in acid solutions. Pd is the second most active metal for the ORR (4), however its mass activity is around 5 times less active than the elemental Pt. Some binary PdM (where M = Co, Ni, Fe, Cu, W, etc.) alloys, have been identified as possible candidates to be used in PEMFCs, owing to higher activity and stability comparing with Pd alone (5-6). The present study reports the synthesis, physical characterization and electrocatalytic evaluation of ternary Pd-based alloys with very low Pt content (i.e., Pd0.5M0.4Pt0.1 , M = Cu, Ni, Fe and Co), for the oxygen reduction reaction in 0.5M H2SO4. This approach therefore addresses one of several key requirements for the development of polymer electrolyte membrane fuel cells, fed with hydrogen and oxygen as a source of clean energy generation.

ECS Transactions, 36 (1) 541-548 (2011)10.1149/1.3660649 © The Electrochemical Society

541Downloaded 18 Jan 2012 to 148.247.62.2. Redistribution subject to ECS license or copyright; see http://www.ecsdl.org/terms_use.jsp

Experimental procedure

Synthesis of the Catalysts The catalysts preparation was carried out using a modified version of a colloidal

method technique reported elsewhere (7). Briefly, acetylacetonates (Acac) of: Pd, Pt, Ni, Fe, Cu and CoCl2 (Sigma-Aldrich Co.) were used as metallic precursors. The tetra-butyl ammonium bromide (TBAB) and sodium borohydride were used as capping and reducing agents respectively. 100 mg of Acac-Pd (2.54x10-4 mol), 20 mg of Acac-Pt (5.08x10-5 mol) and 54 mg of Acac-Ni, 71.7 mg of Acac-Fe, 53.1 mg of Acac-Cu and 26.4 mg of CoCl2, all corresponding to 2.03x10-4 mol respectively, were dissolved in 100 mL of THF, immediately, 163.3 mg of TBAB (5.07x10-4 mol) was added and the solution maintained under magnetic stirring. Over the course of 3 h, 153.4 mg (4.05x10-3 mol) of NaBH4 was added to the solution, and it was brought under a refluxing condition for additional 5 h. In order to remove excess reducing agent as well capping agent, the catalysts were successively washed with water, ethanol and acetone. The catalysts’ nanoparticles were recovered by centrifugation and dried in an oven at 40 °C under N2 flow Physical Characterization

Particle size, morphology and distribution of the material were assessed from

transmission electron microscopy (TEM) in a JEOL model 1011 instrument operated at an accelerated voltage of 80 kV, with a filament of LaB6. The specimens were prepared by dropping the catalysts/acetone ink on a copper mesh covered by a carbon film. Electrochemical Characterization The electrochemical measurements were conducted in a classical three-electrodes glass cell in 0.5 M H2SO4 electrolyte solution at room temperature. Disk potential was controlled with a potentiostat/galvanostat (PARSTAT model 2273). A Pt mesh was used as counter-electrode and the potentials were determined using a freshly prepared reversible hydrogen electrode (RHE). The working electrode was prepared according to Schmidt et al. (8). Briefly, a volume of 1-5 µL of catalysts previously dispersed in ethanol by ultrasound was pipeted onto a polished glassy carbon substrate (GC), geometrical surface area 0.196 cm2, that lead to a nominal Pt loading between 25-130 µg cm-2. After ethanol evaporation, 5 µL of a diluted Nafion® solution (5 wt.% sigma-Aldrich) were pipeted on the electrode surface to assure a Nafion® film thickness of ca. 0.1 µm (the ratio of Ethanol/Nafion® solution was 100/1) (9). The newly prepared electrode was then transferred to the electrochemical cell and immersed under potential control at 0.2 V in a nitrogen-saturated 0.5 M H2SO4 solution. Owing to slight contamination from Nafion® solution, the potential was then continuously cycled between 0.05 V and 1.2 V until a stable voltammogram was recorded (around 10-20 cycles). Quantitative kinetic measurements for the ORR were made using thin-film rotating disk electrode (TF-RDE) method (10), this technique allows the accurate determination of kinetic data such as Tafel slopes, reaction orders and apparent activation enthalpies in the absence of mass transport effect. The acid electrolyte was purged with oxygen and the measurement was performed in a backward manner from 1.0 V to 0.3 V at 20 mV s-1,

ECS Transactions, 36 (1) 541-548 (2011)


100 nm a)

100 nm b)

100 nm c) 100 nm d)

maintaining the starting potential during 50s before scanning. The electrochemical parameters, specific current density (Jk) and mass current density (Jm), were determined at 0.85V and 0.8 V. Electrochemical surface area was determined by CO stripping techniques according to Markovic et al. (11).

Results and Discussion

Transmission electron microscopy The Figure 1 shows a set of micrographs and histograms of the different Pd0.5M0.4Pt0.1 catalysts using low resolution. As can be observed from the TEM micrographs, the Pd0.5M0.4Pt0.1 nanoparticles showed morphology similar to popcorn with distribution rather uniform. Histograms of the particle size distribution, which included analysis of several different regions of the catalysts, established an average sizes between 17 nm to 45 nm. Figure 1. Electron microscopy images of non-supported catalysts: (a) Pt, (b) Pd0.5Co0.4Pt0.1, (c) Pd0.5Fe0.4Pt0.1 and (d) Pd0.5Ni0.4Pt0.1.



Electrochemical Characterization The Figure 2 shows the first and second polarization curves in presence (dashed line) and absence of CO adsorbed on surface (solid line). The CO oxidation reaction was used to calculated electrocatalytic surface area, assuming a charge of 420 µC cm-2 necessary to oxidized a CO monolayer previously adsorbed (12). It is interesting to mention that CO adsorption on alloy catalysts was much difficult to attain compared to Pt. On Pt surface, CO adsorption blocks the surface completely for H-atom electrosorption and suppresses the hydrogen electrode reaction between 0.05 and 0.4 V. However, on catalysts alloys a very small current was detected in the Hupd region which may suppose that some active sites remain unblocked. The voltammogram from Pt nanoparticles used as reference was congruent with polycrystalline surface (13), where the three characteristics region were observed: i) hydrogen adsorption (Hupd) between 0.05 V and 0.4 V, ii) double layer between 0.4 V and 0.7 V and finally, iii) oxides formation after 0.7 V. On the other hand, the Pd0.5M0.4Pt0.1 (M = Co, Fe y Ni) catalysts features are largely different from those for the Pt nanoparticles but closer to polycrystalline palladium (14).

0.0 0.2 0.4 0.6 0.8 1.0 1.2

0.0 0.2 0.4 0.6 0.8 1.0 1.2

PtLcat = 38 μg c

0.1 mA cm-2geo

0.5 mA cm-2geo

Pd0.5

Cu0.4

Pt0.1

Lcat

= 45 μg cm-2geo

E (Volts/RHE) Pd0.5Co0.4Pt0.1

Lcat = 45 μg cm-2geo

E (Volts/RHE)

Pd0.5

Ni0.4

Pt0.1

Lcat

= 36 μg cm-2geo

0.6 mA cm-2geo

0.4 mA cm-2geo

Pd0.5Fe0.4Pt0.1

Lcat = 41 μg cm-2geo

0.3 mA cm-2geo

Figure 2. Polarization curves with and without CO adsorption on nano-electrocatalyst surface, evaluated in a 0.5 M H2SO4 electrolyte solution at room temperature at 20 mV s-1. The solid curves show the second sweep after CO oxidation. On these alloys both Hupd and oxides formation/reduction potentials were practically the same ones. The catalyst Pd0.5Cu0.4Pt0.1 unlike those alloyed with M = Co, Fe and Ni, showed significant differences. Here, the peak associated to hydrogen adsorption and evolution around 0.06 V does not appear, instead only a wide peak around 0.2 V could be observed, in addition, the oxides reduction potential shifted to positive values since an easier reduction of the oxides appears, approaching to Pt catalysts behavior. On the



Pd0.5M0.4Pt0.1 (M = Co, Fe y Ni) catalysts, the electrochemical behavior might suggest a higher Pd surface enrichment which promote almost a complete Pd feature. On the other hand, the catalyst Pd0.5Cu0.4Pt0.1 could present a better metal distribution where Pt and Cu are decorating the surface. The last agree with the observation of a wide peak around 0.3 V to 0.4 V during the first sweep, which reduces successively to a small peak around 0.65 V after several sweeps and which is ascribed to the presence of Cu on surface. Oxygen reduction Reaction Kinetics Figure 3a shows a set of negatively polarization curves from 1.0 V to 0.3 V for the ORR at 1600 rpm after parameter optimization (15). The experimental parameters optimizing is a very important step before getting confident data; the use of high catalyst loading can promote the formation of agglomerates or thick film, avoiding a complete catalyst distribution on the working electrode, besides it can avoid the complete catalyst utilization or interfere in normal hydrodynamic processes of RDE. On the contrary, very small loading cannot be enough to completely cover the working geometrical surface area, which can let to inactive areas (15). The averaged results of Koutecky-Levich plots obtained at 0.3 and 0.5 V, as well as the Tafel plots are presented in Figure 3b and 3c, respectively. Results of kinetic analysis are summarized in table 1. For a mixing reaction process, as for the oxygen reduction reaction, the total current density from a thin-film rotating disk electrode is the sum of reciprocal of kinetic current (1/Jk), limiting current (1/Jm) and thin-film diffusion current through Nafion® film (1/Jf). However, Jf contribution can be neglected whether Nafion® layer is enough thin (< 0.2 µm) (8). The last was considered during the working electrode preparation. Considering that the limiting current is a linear function of square root of rotation rate (ω1/2) according to the follow [1]:

[1] where n is the electron number (generally 4 for ORR), F is the Faraday constant (96 485 C mol-1), CO2 is the saturating oxygen concentration (1.12x10-6 mol cm-2), DO2 is the oxygen diffusion coefficient (1.8x10-5 cm2 s-1), v is the kinematic viscosity (1x10-2 cm2 s-

1) and ω is electrode rotation rate (rpm), the linearity between 1/Jl vs w-1/2 suggest a diffusion control. The parallelism of the Koutecky-Levich slops support a first order kinetics (Figure 3b). The averaged slopes value agree with theoretical slope calculated considering four electron transfer mechanism, calculated by using parameters earlier reported (B=12.6x10-2 mA cm-2 rpm-1/2) (16). This suggests that the overall oxygen reduction mechanism proceeds through four electrons transfer: O2 + 4H+ + 4e- 2H2O. The kinetic data were obtained from Tafel slops after mass transfer correction (17). The slops calculated between 0.9-0.8 V shows values from 60 to 80mV dec-1 (Figure 3c) supporting the first electron transfer as the limiting reaction step which agree with reported in literature (18). The kinetics data (Table 1), revealed an enhanced activity of the Pd0.5Cu0.4Pt0.1 catalyst in relation to Pd0.5M0.4Pt0.1 (M = Co, Fe y Ni) catalysts, this becoming as high as that obtained with platinum. Adzic et al. (19) reported a specific activity of 131 µA cm-

2Pd for Pd at 0.85 V. This result agrees with those of Pd0.5M0.4Pt0.1 (M = Co, Fe y Ni) but

two times lower than Pd0.5Cu0.4Pt0.1 catalyst obtained in this work. However is important to mention that Adzic used 0.1 M HClO4 as electrolyte, where the absence of spectator



ion co-adsorption phenomenon, like bisulfate ions in sulfuric acid electrolyte, would expect a higher activity (20). On the other hand, mass activity from Pd0.5Cu0.4Pt0.1, which has a direct relation with cost and production/application from an economic point of view, showed two times more activity than Pt, making this alloy an interesting candidate to be used as cathode electrode in PEM fuel cell with in on-site applications.

0.4 0.6 0.8 1.0

-4

-3

-2

-1

0

0.20

0.25

0.30

0.35

0.40

0.45

0.50

0.550.020 0.025 0.030 0.035 0.040 0.045 0.050 -1.6 -1.2 -0.8 -0.4 0.0 0.4

0.78

0.80

0.82

0.84

0.86

0.88

0.90

0.92

a)

J (m

A c

m-2

ge

o)

E (Volts/RHE)

Pt Pd

0.5Fe

0.4Pt

0.1

Pd0.5

Ni0.4

Pt0.1

Pd0.5

Co0.4

Pt0.1

Pd0.5

Cu0.4

Pt0.1

20 mV s-1 @ 1600 rpm

bo = mV dec-1

w1/2 (rpm-1/2)

11.5 x10-2 ± 0.18 10.4 x10-2 ± 0.84 11.2 x10-2 ± 0.60 11.2 x10-2 ± 1.90 10.6 x10-2 ± 1.50

J -1 (m

A-1 c

m2 ge

o)

Bo (mA cm-2 rpm-1/2)

b) c)

Log Jk (mA cm-2real)

E (V

olts

/RH

E)

80 64 ± 6.4 72 ± 1.8 72 ± 3.6 82 ± 28

Figure 3. (a) Set of polarization curves on Pt and Pd0.5M0.4Pt0.1 (M = Co, Cu, Fe y Ni) alloys with different specific loadings in an O2 saturated 0.5 M H2SO4 0.5 M, at 20 mV/s and 1600 rpm. (b) Average Koutecky-Levich data at 0.3 and 0.5 V. (c) Tafel slopes at low overpotential from 0.9 V to 0.8 V.

TABLE I. Kinetic parameters, jk y jm deduced at different electrode potentials (0.9-0.8 V/RHE), for the ORR on Pt and Pd0.5M0.4Pt0.1 (M = Co, Fe Cu, Ni) catalysts.

Catalysts Jk (μA cm-2real) Jm (mA mg-1

PdPt) 0.9 0.85 0.80 0.9 0.85 0.80

Pt 187 ± 109 660 ± 270 1900 ± 595 5.8 ± 2.9 21.3 ± 7.3 61.6 ± 11 Pd0.5Cu0.4Pt0.1 82 ± 28 269 ± 160 1354 ± 1507 6.9 ± 5.0 20.1 ± 14 102 ± 12.2 Pd0.5Ni0.4Pt0.1 34 ± 4.7 128 ± 22 821 ± 171 2.0 ± 0.2 9.3 ± 0.8 60 ± 8.2 Pd0.5Co0.4Pt0.1 47 ± 25 175 ± 139 1082 ± 620 3.1 ± 0.3 13.2 ± 1.6 72 ± 45 Pd0.5Fe0.4Pt0.1 29 ± 6.3 120 ± 40 904 ± 137 1.5 ± 0.07 6.1 ± 0.6 49 ± 17



Conclusions

The colloidal synthesis method lets obtaining new ternary alloys with very low Pt loading. According to transmission microscopy, particles between 17 nm and 45 nm with homogeneous distribution and popcorn morphology are obtained. An important feature of this synthesis is that thermal treatment for surfactant stripping was not required. The catalysts Pd0.5M0.4Pt0.1 (M = Co, Ni, Fe) showed specific activities similar to Pd, associated to a produced highly segregated structure, statement which is now in route in order to be prove. The specific activities from catalysts Pd0.5Cu0.4Pt0.1 was two times higher than that from some other PdCu alloy previously reported (21) and similar to Pt, but two times more mass activity than the last one.

Acknowledgments The authors gratefully acknowledge M.C. Angélica González Maciel, M.C. Rafael Reynoso R. from Instituto Nacional de Pediatría, and especially to Ph.D. Rocio George Tellez for their invaluable assistance in carrying TEM measurements. The authors acknowledge the support of the Mexico’s National Polytechnique Institute for the doctoral fellowships under grant SIP-2010-1029. This work was partially supported by CONACYT (grant 101537).

References

1. M. Oezaslan, M. Strasser, J. Power Sources, 196, 5240 (2011). 2. K. Sasaki, K. Kuttiviel, A. Kurian, D. Su, R.R. Adzic, Electrocatalysis, 2, 134

(2011). 3. Y. Wang, K.S. Chen, J. Mishler, S.C. Cho, X.C. Adroher, Appl. Energy, 88, 981

(2011). 4. L. Jiang, A. Hsu, D. Chu, R. Chen, J. Electrochem. Soc., 156, B643 (2009). 5. O. Savadogo, K. Lee, S. Mitsushima, N. Kamiya, K. I. Ota, J. New Mat.

Electrochem. Systems, 7, 77 (2004). 6. M. H. Shao, K. Sasaki, R. R. Adzic, J. Am. Chem. Soc., 128, 3526 (2006). 7. V. H. Bönnemann, W. Briyoux, R. Brinkmann, E. Dinjus, T. Jouβen, Korall;

Angew. Chem., 103, 1344 (1991). 8. T.J. Schmidt, H.A. Gasteiger, G.D. Stab, P.M. Urban, D.M. Kolb, R.J. Behm, J.

Electrochem. Soc., 145, 2354 (1998). 9. U. A. Paulus, T. J. Schmidt, H. A. Gasteiger, R. J. Behm, J. Electroanal. Chem.,

495, 134 (2001). 10. T.J. Schmidt, H.A. Gasteiger, G.D. Sta¨b, P.M. Urban, D.M. Kolb, R.J. Behm, J.

Electrochem. Soc., 145, 2354 (1998). 11. K.J.J. Mayrhofer, D. Strmcnik, B.B. Blizanac, V. Stamenkovic, M. Arenz, N.M.

Markovic, Electrochim. Acta, 53, 3181 (2008). 12. H. Ye, J.A. Crooks, R.M. Crooks, Langmuir, 23, 11901 (2007). 13. J. Clavilier, J. M. Orts, R. Gomez, J. M. Feliu, A. Aldaz, J. Electroanal. Chem.,

404, 281 (1996). 14. G. Ramos-Sanchez, H.Y. Madeira, O. Solorza-Feria, Int. J. hydrogen Energy, 33,

3596 (2008).



15. K. J. J. Mayrhofer, D. Strmcnik, B. B. Blizanac, V. Stamenkovic, M. Arenz, N. M. Markovic, Electrochim. Acta, 53, 3181 (2008)].

16. C. Shengli, A. Kucernak, J. Phys. Chem. B., 108, 3262 (2004). 17. U.A. Paulus, A. Wokaun, G.G. Scherer, T.J. Schmidt, V. Stamenkovic, N.M.

Markovic, P.N. Ross, Electrochim. Acta, 47, 3787 (2002). 18. S. Thanasilp, M. Hunsom, Renewable Energy, 36, 1795 (2011). 19. M. H. shao, K. Sasaki, R. R. Adzic, J. Am. Chem. Soc., 128, 3526 (2006). 20. N. M. Markovic, T. J. Schmidt, V. Stamenkovic, P. N. Ross; Fuel Cells, 1, 105

(2001). 21. N.N. Kariuki, X. Wang, J.R. Mawdsley, M.S. Ferrandon, S.G. Niyogi, J.T.

Vaughey, D.J. Myers, Chem. Mater., 22, 4144 (2010).



Electrocatalysis of oxygen reduction reaction on polyaniline-derived nitrogen-doped carbon nanoparticle surfaces in alkaline media

Documents