CO monolayer oxidation on Pt nanoparticles: Further insights into the particle size effects

Dedicated to Professor David J. Schiffrin on the occasion of his retirement
https://doi.org/10.1016/j.jelechem.2006.02.024Get rights and content

Abstract

This paper provides further insights into the particle size effects in CO monolayer oxidation. Strong particle size effects are confirmed in the size range from 1.8 to 5 nm. The discrepancies in the literature concerned with the particle size effects in CO monolayer oxidation are reconciled by exploring the influence of the experimental conditions on the stripping voltammograms and chronoamperograms. Evidence supporting the contribution of slow non-electrochemical step to the overall mechanism of CO oxidation is presented. The particle size effects in CO monolayer oxidation are attributed to the size-dependent COads + OHads interaction as well as to the size-dependent COads surface diffusion coefficient.

Introduction

For over a decade, research into fuel cells has grown in line with the worldwide demand for energy, particularly with global warming concerns and the consequent requirements to reduce consumption of fossil fuels (petrol, coal, gas), which produce greenhouse gases. Polymer electrolyte membrane (PEM) fuel cells are a promising solution to provide electricity for mobile and portable applications. Thus, a growing interest develops in understanding the electrocatalytic activity of finely dispersed metallic electrocatalysts in the form of nanoparticles anchored to conducting substrates. According to the definition of Boudart [1], most of PEM fuel cell reactions are structure-sensitive when Pt is the electrocatalyst. The activity of Pt for oxygen electroreduction [2], [3], [4], [5], oxygen electroreduction in methanol-containing electrolyte [6], methanol [3], [7], [8], [9], [10] and CO [9], [11], [12], [13], [14] electrooxidation is strongly dependent on the particle size.

Different approaches have been employed in order to unveil the influence of the particle size and structure on electrocatalytic activities. One of them rests on the utilisation of stepped (high index) single crystals as the means to control the amount of low coordination sites (steps, kinks) on the surface. However, recent investigations have shown unambiguously that the (electro)catalytic behaviour of edge and corner atoms on nanoparticles may be drastically different from that exhibited by steps and kinks on single crystalline surfaces [14], [15], [16].

In previous papers, we have demonstrated strong particle size-effects on the kinetics of CO electrooxidation. By decreasing the particle size from 3.1 to 1.7 nm, the CO oxidation onset and the stripping peak shift positive by ca. 90 mV and pronounced tailing towards positive potentials develops [12], [13]. Chronoamperometric measurements show peaked current transients [11], [12]. The position of the peak is strongly particle size dependent shifting to longer times with the decrease of the particle size. Not only the positions of the transients on the time scale, but also their shapes are strongly influenced by the particle size. In contrast to single crystalline surfaces exhibiting symmetric current transients, which can be reasonably described by the Mean Field model, Pt nanoparticles show asymmetric transients, the tailing of the current after the maxima increasing with the decrease in the particle size. The origin of such tailing was attributed to the transition from activation-controlled to diffusion-controlled kinetics [11], [12]. By using a simple mathematical model, we proposed that the COads diffusion coefficient is strongly particle size dependent (restricted COads mobility at Pt nanoparticles below ca. 2 nm size) and is controlling the CO electrooxidation kinetics for sizes below 2 nm [12]. The paper was presented at the Faraday Discussion meeting organised by David Schiffrin in 2003 in Liverpool, and provoked heated debate. Later the particle size effect was confirmed by FTIR spectroscopy which evidenced that CO2 formation on small (ca. 1.7 nm) particles is shifted to positive potentials compared to large particles [14].

Recently, Mayrhofer et al. [17], [18] reported conflicting data that the position of the CO stripping peak on carbon-supported Pt particles is almost particle size independent in the range from 1 to 5 nm and that current transients obtained on 2 nm particles are composed of two peaks contrary to what is observed on 1 nm particles. On the other hand, they claimed that although the onset potential of CO oxidation is almost independent of the particle size, the rate of CO2 production is strongly dependent on the size (i.e. 1 < 2 < 5 < 30 nm).

In this paper, we report novel data confirming the existence of the particle size effects in CO monolayer oxidation. The influence of experimental parameters such as sweep rate, CO coverage, electrooxidation potential on the CO oxidation is studied using stripping voltammetry and chronoamperometry. Thus, we attempt to resolve literature discrepancies on the influence of the size of Pt particles on the rate of CO monolayer oxidation on their surfaces. The experimental results are compared to the literature data related to CO oxidation on Pt single crystals and nanoparticles. Further evidence supporting the contribution of slow non-electrochemical step to the overall mechanism of CO oxidation is presented. Pros and cons low COads surface mobility are put on the table and discussed.

Section snippets

Nanoparticle preparation and characterisation

Glassy carbon (GC) rods (Sigradur G from Hochtemperatur-Werkstoffe GmbH, 7 mm diameter, 0.385 cm2 face area) were polished to a mirror finish with 0.25 μm diamond paste and cleaned repeatedly with methanol, acetone and water in an ultrasonic bath before the electrode preparation. Pt nanoparticles supported onto the surface of GC were prepared using wet chemical deposition (WCD) method described in Refs. [9], [19]. In order to activate the surface before Pt deposition, GC was oxidised in 0.1 M H2SO4

CO stripping voltammetries: the effect of the sweep rate

Fig. 2 represents the effect of the potential sweep rate on CO stripping voltammograms obtained for Pt/GC#a with d¯=2.8nm (see Table 1). We start with the CO stripping voltammogram acquired at 5 mV s−1. The stripping peak is very symmetric in agreement with that reported by Arenz et al. [18] for “CO-annealed” carbon-supported Pt nanoparticles with the average particle size from 1 to 5 nm. The onset of CO oxidation is located at ca. 0.7 V vs. RHE in agreement with the previous observations obtained

Conclusions

Since COads electrooxidation is a multistep process, depending on the experimental conditions employed such as the electrode potential, the potential scan rate, the CO coverage different reaction steps may become limiting, hence the discrepancies between different research groups. CO monolayer electrooxidation is strongly dependent on the particle size and structure, its rate decreasing with the decrease of the particle size. Both stripping voltammetry as well as chronoamperometry reveal the

Acknowledgements

Financial support by the Deutsche Forschung Gemeinschaft (DFG) under contract Sti74/8-4 is gratefully acknowledged. We thank Dr. Bernhard Andreaus and Prof. Michael Eikerling from Simon Fraser University, Canada for fruitful and stimulating discussions. We thank Dr. Marianne Hanzlik and Siegfried Schreier for the TEM measurements and Prof. Sevil Weinkauf for providing the access to the electron microscope.

References (71)

  • M. Boudart

    Adv. Catal.

    (1969)
  • Y. Takasu et al.

    Electrochim. Acta

    (1996)
  • A. Kabbabi et al.

    J. Electroanal. Chem.

    (1994)
  • A. Gamez et al.

    Electrochim. Acta

    (1996)
  • L. Genies et al.

    Electrochim. Acta

    (1998)
  • F. Maillard et al.

    Electrochim. Acta

    (2002)
  • K. Yahikozawa et al.

    Electrochim. Acta

    (1991)
  • Y. Takasu et al.

    Electrochem. Commun.

    (2000)
  • O.V. Cherstiouk et al.

    Electrochim. Acta

    (2003)
  • T. Frelink et al.

    J. Electroanal. Chem.

    (1995)
  • O.V. Cherstiouk et al.

    J. Electroanal. Chem.

    (2003)
  • C.R. Henry

    Appl. Surf. Sci.

    (2000)
  • K.J.J. Mayrhofer et al.

    Electrochim. Acta

    (2005)
  • L. Palaikis et al.

    Surf. Sci.

    (1988)
  • C. Mc Callum et al.

    J. Electroanal. Chem.

    (1976)
  • A.V. Petukhov et al.

    Surf. Sci.

    (1998)
  • N.P. Lebedeva et al.

    J. Electroanal. Chem.

    (2002)
  • A.B. Anderson

    Electrochim. Acta

    (2002)
  • Y. Takasu et al.

    Appl. Surf. Sci.

    (1997)
  • J.M. Feliu et al.

    J. Electroanal. Chem.

    (1990)
  • S.C. Chang et al.

    Surf. Sci.

    (1990)
  • N.P. Lebedeva et al.

    J. Electroanal. Chem.

    (2000)
  • J.H. Jiang et al.

    J. Electroanal. Chem.

    (2002)
  • M. Gustavsson et al.

    J. Electroanal. Chem.

    (2004)
  • H. Gleiter

    Nanostruct. Mater.

    (1992)
  • C. Mottet et al.

    Surf. Sci.

    (1996)
  • S. Mukerjee et al.

    J. Electroanal. Chem.

    (1998)
  • J. Solla-Gullon et al.

    Electrochem. Commun.

    (2006)
  • P. Rodriguez et al.

    Electrochim. Acta

    (2005)
  • A. Cuesta

    Surf. Sci.

    (2004)
  • F. Maillard et al.

    Faraday Discuss.

    (2004)
  • F. Maillard et al.

    Phys. Chem. Chem. Phys.

    (2005)
  • F. Maillard et al.

    J. Phys. Chem. B

    (2004)
  • C.R. Henry

    Surf. Sci. Rep.

    (1998)
  • M. Arenz et al.

    J. Am. Chem. Soc.

    (2005)
  • Cited by (0)

    View full text