As a full professor of Mechanical Engineering and director of the Center for Sustainable Energy Technology at The Hong Kong University of Science and Technology (HKUST), Tim S. Zhao has been working on fuel cells for more than a decade. Zhen-Xiang Liang is full professor at the School of Chemistry and Chemical Engineering, South China University of Technlogy, Guangzhou, China. He has been working on direct alcohol fuel cells for almost a decade and has a proven track record in electrochemistry, particularly in the area of electrocatalyst developments.

Catalysts for Alcohol-Fuelled Direct Oxidation Fuel Cells

By Z. X. Liang, T. S. Zhao

The Royal Society of Chemistry

Copyright © 2012 The Royal Society of Chemistry
All rights reserved.
ISBN: 978-1-84973-405-9

Contents

Chapter 1 Electrocatalysis of Alcohol Oxidation Reactions at Platinum Group Metals Claude Lamy and Christophe Coutanceau, 1,
Chapter 2 Nanoalloy Electrocatalysts for Alcohol Oxidation Reactions Jun Yin, Bridgid Wanjala, Bin Fang, Jin Luo, Rameshowri Loukrakpam, Lefu Yang, Shiyao Shan, Ming Nie and Chuan-Jian Zhong, 71,
Chapter 3 Theoretical Studies of Formic Acid Oxidation Wang Gao and Timo Jacob, 97,
Chapter 4 Gold Leaf Based Electrocatalysts Rongyue Wang and Yi Ding, 129,
Chapter 5 Nanocatalysts for Direct Borohydride Oxidation in Alkaline Media Christophe Coutanceau, Stève Baranton and Mário Simõesg, 158,
Chapter 6 Bioelectrocatalysis in Direct Alcohol Fuel Cells H. A. Reeve and K. A. Vincent, 206,
Chapter 7 Challenges and Perspectives of Nanocatalysts in Alcohol-Fuelled Direct Oxidation Fuel Cells E. H. Yu, X. Wang, X. T. Liu and L. Li, 227,
Subject Index, 250,

CHAPTER 1

Electrocatalysis of Alcohol Oxidation Reactions at Platinum Group Metals

CLAUDE LAMY AND CHRISTOPHE COUTANCEAU

1.1 Introduction

Electrocatalysis, i.e. the heterogeneous catalysis of electrochemical reactions occurring at the electrode/electrolyte interface, has mainly concerned technological investigations related either to energy storage (e.g. water electrolysis) or to energy conversion (e.g. fuel cells). However, except for the hydrogen electrode, which is now well known, and the oxygen electrode, which has been extensively studied, other electrodes of practical interest, such as soluble fuel electrodes, need much more investigation. Among them, alcohol electrodes are particularly suitable for use in a direct oxidation fuel cell (DOFC) because of several favourable features, such as a high theoretical energy density (4–9 kWh kg-1 compared to 33 kWh kg-1 for molecular hydrogen) and a great facility of handling.

Moreover, alcohols, which may be produced from the biomass, are very interesting fuels due to a lot of advantages: high solubility in aqueous electrolytes, relatively high reactivity, ease of storage and supply, low toxicity (except for methanol). They can be directly electro-oxidized in a direct alcohol fuel cell (DAFC).This explains why many fundamental investigations were undertaken in the last three decades on the electrochemical oxidation of several alcohols: methanol, ethanol, ethylene glycol, glycerol, propanol and butanol, and also on related compounds: formic acid, formaldehyde, carbon monoxide, etc. Until now, the most promising and most studied fuels for application in a DOFC, with the direct oxidation of the organic molecule, are alcohols such as methanol and ethanol.

However, a lot of electrocatalytic problems still arise due to the relative complexity of the reaction mechanisms. These include the effect of the nature of the reaction products, the structure of the adsorbed intermediates, the nature and structure of the electrode material, the molecular structure of the organic compounds, the pH and the anions of the supporting electrolyte, and the role of the water adsorption residues.

Furthermore, the catalyst structure, such as the particle size, the composition and the degree of alloying, are also of great importance, since most of the alcohol oxidation reactions are structure sensitive and because it is of great interest to reduce the platinum group metal loading (either by metal dispersion or by synthesis of multimetallic catalysts) in order to reduce the cost of the DAFC system. These topics will be illustrated mainly with results obtained in this laboratory.

1.2 Thermodynamics and Kinetics of Alcohol Oxidation Reactions

In a DOFC the total electro-oxidation to CO2 of an aliphatic oxygenated compound CxHyO containing one oxygen atom (mono-alcohols, aldehydes, ketones, ethers, etc.) involves the participation of water (H2O) or of its adsorbed residue (OHads) provided by the cathodic reaction (electro-reduction of dioxygen).

The overall electro-oxidation reaction in acid medium to reject the carbon dioxide produced can thus be written as follows:

CxHyO+(2x – 1)H2O->xCO2+nH++ne- (1.1)

with n = 4x + y – 2. Such an anodic reaction is very complicated from a kinetics point of view since it involves multielectron transfers and the presence of different adsorbed intermediates and several reaction products and by- products. However, from thermodynamic data it is easy to calculate the reversible anode potential, the cell voltage under standard conditions, the theoretical energy efficiency and the energy density.

1.2.1 Thermodynamic Data

According to reaction (1.1) the standard Gibbs energy change -ΔG1°, allowing calculation of the standard anode potential, [MATHEMATICAL EXPRESSION OMITTED], can be evaluated from the standard energy of formation ΔGfi of reactant i:

[MATHEMATICAL EXPRESSION OMITTED] (1.2)

In the cathodic compartment the electro-reduction of oxygen occurs, as follows:

1/2O2 + 2H+ + 2e- -> H2O (1.3)

with [MATHEMATICAL EXPRESSION OMITTED], leading to a standard cathodic potential, E°2:

[MATHEMATICAL EXPRESSION OMITTED] (1.4)

where SHE is the standard hydrogen electrode, acting as a reference electrode.

In the fuel cell the electrical balance corresponds to the complete combustion of the organic compound in the presence of oxygen, as follows:

[MATHEMATICAL EXPRESSION OMITTED] (1.5)

with [MATHEMATICAL EXPRESSION OMITTED], leading to the equilibrium standard cell voltage:

[MATHEMATICAL EXPRESSION OMITTED] (1.6)

Then it is possible to evaluate the specific energy We in kWh kg-1:

[MATHEMATICAL EXPRESSION OMITTED] (1.7)

with M the molecular mass of the compound. Knowing the enthalpy change ΔH° from thermodynamic data:

[MATHEMATICAL EXPRESSION OMITTED] (1.8)

one may calculate the reversible energy efficiency under standard conditions:

εrev = ΔG°r/δH°r (1.9)

For the oxidation of methanol and ethanol, the corresponding equations are:

[MATHEMATICAL EXPRESSION OMITTED] (1.10)

[MATHEMATICAL EXPRESSION OMITTED] (1.11)

whereas the electro-reduction reaction of molecular oxygen occurs at the cathode:

[MATHEMATICAL EXPRESSION OMITTED] (1.12)

where E°i are the standard electrode potentials vs. SHE.

This corresponds to the overall combustion reaction of alcohols in oxygen:

[MATHEMATICAL EXPRESSION OMITTED] (1.13)

[MATHEMATICAL EXPRESSION OMITTED] (1.14)

where the cell voltages are calculated under standard conditions.

For higher alcohols, such as n-propanol, taken as an example, the following calculations can be made:

C3H7OH + 5H2O -> 3CO2 + 18H+ + 18e- (1.15)

[MATHEMATICAL EXPRESSION OMITTED] (1.16)

[MATHEMATICAL EXPRESSION OMITTED] (1.17)

C3H7OH + 9/2O2 -> 3CO2 + 4H2O (1.18)

[MATHEMATICAL EXPRESSION OMITTED] (1.19)

The standard cell voltage is thus:

[MATHEMATICAL EXPRESSION OMITTED] (1.20)

and the specific energy is:

We = 1963/3600 × 60 = 9.09 kWh kg-1 (1.20a)

The enthalpy change of reaction (1.18) is:

ΔH°r = -3×395.5-4×285.8+302.6=-2027 kJ mol-1 (1.21)

so that the reversible energy efficiency is:

[MATHEMATICAL EXPRESSION OMITTED] (1.22)

Similar calculations can be made with many oxygenated fuels, including polyols (ethylene glycol, glycerol), propargyl alcohol, ethers and polyethers [dimethyl ether, CH3OCH3, ethyl methyl ether, CH3OC2H5, diethyl ether, C2H5OC2H5, dimethoxymethane, (CH3O)2CH2, trimethoxymethane, (CH3O)3CH, trioxane, (CH2O)3].

The energy density of the fuel, We, the cell voltage of the cell at equilibrium, E°eq, and the reversible energy efficiency of the cell, εrev, for several alcohols can be calculated under standard conditions (25 °C, liquid phase). The results are summarized in Table 1.1.

For all oxygenated compounds listed in Table 1.1, the cell voltage varies from 1.2 to 1.0 V, which is very similar to that of a hydrogen/oxygen fuel cell (E°eq = 1.23 V). The energy density varies between half to one of that of gasoline (10–11 kWh kg-1), so these compounds are good alternative fuels to hydrocarbons. Furthermore, the reversible energy efficiency εrev is close to 1, while that of the H2/O2 fuel cell is 0.83 at 25 °C (standard conditions). From these data, it appears that amongst the mono-alcohols, methanol and ethanol lead to higher cell voltages and reversible energy efficiency under standard conditions.

1.2.2 Kinetics Problems

The electro-oxidation of aliphatic oxygenated compounds, even the simplest one, i.e. methanol, involves the transfer of many electrons (n = 6 for methanol). The reaction mechanism is thus complex, the oxidation reaction occurring through many successive and parallel paths involving many adsorbed intermediates and by-products. The oxidation reaction needs a convenient electrocatalyst to increase the reaction rate and to modify the reaction pathway in order to reach more rapidly the final step, i.e. the production of carbon dioxide. The relative slowness of the reaction, and the difficulty to break the C–C bond at low temperatures (25–80 °C), lead to high anodic overvoltages ηa, which will greatly reduce the operating cell voltage (Figure 1.1).

Thus, the practical electrical efficiency of a fuel cell is dependent on the current that is delivered by the cell and is lower than that of the reversible efficiency due to the irreversibility of the electrochemical reactions involved at the electrodes. The practical efficiency of a fuel cell can be expressed as follows:

[MATHEMATICAL EXPRESSION OMITTED] (1.23)

with the cell voltage [MATHEMATICAL EXPRESSION OMITTED] at a current density j and Re the electrolyte and interfacial specific resistances.

From eqn (1.23) it follows that the increase of the practical fuel cell efficiency can be achieved by increasing the voltage efficiency (εE = E(j)/E°eq) and the faradic efficiency (εF = nexp/n), the reversible yield being fixed by the thermodynamic data.

For a given electrochemical system the increase of the voltage efficiency is directly related to the decrease of the overpotentials of the oxygen reduction reaction, ηc, and alcohol oxidation reaction, ηa, which needs to enhance the activity of the catalysts at low overpotentials and low temperature, whereas the increase of the faradic efficiency is related to the ability of the catalyst to oxidize completely (or not) the fuel into carbon dioxide, i.e. it is related to the selectivity of the catalyst. Indeed, in the case of ethanol, taken as an example, acetic acid and acetaldehyde are formed at the anode, which corresponds to the number of electrons involved as 4 and 2, respectively, against 12 for the complete oxidation of ethanol to carbon dioxide. The enhancement of both these efficiencies is a challenge in electrocatalysis.

In most cases, ηa is at least 0.5 V for a reasonable current density (100 mA cm-2) so that the cell voltage, including an overvoltage ηc = 0.3 V for the cathodic reaction, will be of the order of 0.4 V, and the voltage efficiency will be εE = 0.4/1.2 = 0.33 under the operating conditions. Such a drawback of the direct alcohol fuel cell can only be removed by improving the kinetics of the electro-oxidation of the fuel. This needs a relatively good knowledge of the reaction mechanisms, particularly of the rate-determining step, and to search for electrode materials (Pt–X binary and Pt–X–Y ternary electrocatalysts) with improved catalytic properties.

1.3 Preparation and Physicochemical Characterization of Platinum-Based Nanocatalysts

Several methods, developed to synthesize platinum-based nanocatalysts, with control of the particle size, atomic composition and metal loading, are shown in Figure 1.2: chemical methods (impregnation–reduction, colloidal, carbonyl complex route), physicochemical methods [electrochemical pulse deposition, metal organic chemical vapour deposition (MOCVD)] and physical methods [plasma enhanced physical vapour deposition (PVD), laser ablation], etc. Some of these methods will be developed further.

1.3.1 Synthesis by Chemical Methods

1.3.1.1 Impregnation–Reduction Methods

The co-impregnation–reduction method is extensively used in the domain of heterogeneous catalysis for the synthesis of supported metallic nanoparticles. In the case of fuel cell anodes, a carbon support (Vulcan XC-72 previously heated at 400 °C for 4 h under a pure nitrogen atmosphere) is used. This support is first oxidized by contact with aqua regia (solution of 1/3HNO3 + 1/3HCl + 1/3H2O) for 5 min in order to form surface functional groups (hydroxyl and carboxylic acid groups) which will be further used to disperse the metal by exchanging protons with the metal salts. The carbon powder is then put in an aqueous solution (ultrapure water from Millipore, MilliQ 18 MV cm) containing chloroplatinic acid (H2PtCl6 from Alfa Aesar) and other metal salts (SnCl4 and ReCl3 from Sigma, as examples), and the pH is adjusted to about 1. After 24 h, the metallic catalysts are deposited on the carbon substrate, and the mixture is dried overnight at 70 °C in an oven.

The resulting powder is then calcined in air for 4 h at 300 °C and reduced in a pure hydrogen atmosphere for 4 h at 300 °C. PtSn/C and PtRe/C catalytic powders with different atomic ratios can be prepared in this way with a metal loading close to 30 wt% on carbon.

1.3.1.2 Colloidal Methods

1.3.1.2.1 Bönnemann method

The procedure described by Bönnemann et al. was slightly modified and adapted to the preparation of mono- and multimetallic catalyst precursors. The synthesis is carried out under a controlled atmosphere (argon) free of oxygen and water, with non-hydrated metal salts (99.9% PtCl2, RuCl3, SnCl2, etc.). The first step consists of the preparation of a tetraalkyl triethylborohydride reducing agent (Nalk4)+(BEt3H)-, which will also act as a surfactant after metal reduction, preventing any agglomeration of the metallic particles:

[MATHEMATICAL EXPRESSION OMITTED] (1.24)

The colloidal precursors are then dispersed on a carbon support (e.g. Vulcan XC-72) and calcined at 300 °C for 1 h under an air atmosphere to remove the organic surfactant. The Bönnemann method allows different possibilities for the synthesis of multimetallic supported catalysts:

• Synthesis of multimetallic catalysts with a controlled atomic ratio by mixing different metal salts before the reduction step and formation of the precursor colloid (co-reduction method):

[MATHEMATICAL EXPRESSION OMITTED] (1.25)

• Synthesis of catalysts with controlled atomic ratios by codeposition of different metal colloids before the calcination step and formation of the catalytic powder (codeposition method):

[MATHEMATICAL EXPRESSION OMITTED] (1.26)

• Synthesis by simply mixing the different catalytic powders (ball milling method): xPt/C + yM/C.

1.3.1.2.2 Water-in-oil microemulsion method

The “water in oil” microemulsion method is derived from that developed by Boutonnet et al. The catalysts are prepared by mixing NaBH4 (99% from Acros Organics), as reducing agent, with a microemulsion carrying the specific reactants dissolved in an aqueous phase (MilliQ Millipore ultrapure water, 18.2 MΩ cm). Poly(ethylene glycol dodecyl ether) (Brij30 from Fluka) was chosen as surfactant and the organic phase was n-heptane (99% from Acros). Desired amounts of the metal salts are dissolved in ultrapure water in order to obtain metallic nanoparticles with controlled size and compositions after the reduction process with NaBH4. The following concentrations of reactants are used: [water]/[Brij30] molar ratio = 5, Brij30 = 16.5 wt%, [metal salts] = 0.205 mol L-1, NaBH4/PtIV molar ratio = 15.

Carbon (Vulcan XC-72), previously treated under N2 at 400 °C for 4 h, is added directly to the colloidal solution to obtain the desired metal loading and the mixture is kept under stirring for 2 h. In the present work, all the catalysts were synthesized in order to obtain 40 wt% metal loading. The mixture was filtered on a 0.22 µm Durapore membrane filter (Millipore). The resulting powder was abundantly rinsed with ethanol, acetone and ultrapure water. The carbon-supported catalysts were dried overnight in an oven at 75 °C. This method leads to small particles (2–4 nm diameter) with a narrow size distribution.

(Continues…)Excerpted from Catalysts for Alcohol-Fuelled Direct Oxidation Fuel Cells by Z. X. Liang, T. S. Zhao. Copyright © 2012 The Royal Society of Chemistry. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.