Electrocatalyst oxidation of formic acid on Pt nanoparticles/Ti electrode

Solutions, chemicals and electrochemical measurement

All chemicals used were of analytical grade. All electrochemical experiments were carried out at room temperature. Distilled water was used throughout. The electrochemical experiments were performed in a three-electrode cell arrangement. A platinum sheet was used as counter electrode, while all potentials were measured with respect to a commercial saturated calomel reference electrode. Electrochemical experiments were carried out using a Princeton Applied Research, EG&G PARSTAT 2263 Advanced Electrochemical system run by Powersuite software.

Preparation of Pt-NPs/Ti electrodes

Titanium discs were cut from titanium plates and mounted using polyester resin. The titanium electrodes was first mechanically polished and then chemically etched by immersing in a mixture of HF/HNO₃ 1∶3 solution for 1 min. Before electrodeposition, titanium samples were degreased by sonicating in acetone and ethanol followed by rinsing with distilled water. The Pt-NPs were electrochemically deposited at the surface of titanium substrate (Pt-NPs/Ti) from 1 mM H₂PtCl₆ in aqueous 0·1M H₂SO₄ solution as the supporting electrolyte. The deposition conditions were a current density of 10 mA cm⁻² for 5 min, and the temperature is maintained at 45°C.

Surface morphology of electrodes

For characterising the morphology of Pt-NPs deposited on titanium substrates, a scanning electron microscope (Model XL30, Philips, The Netherlands) was employed with an accelerating voltage 15 kV. To identify the element composition, an energy dispersive X-ray was employed with an accelerating voltage 16 kV.

Morphology of Pt-NPs/Ti electrodes

Figure 1a illustrates the SEM images of Pt-NPs deposited on the titanium plate. It can be seen that Pt-NPs with diameters of ∼20–40 nm are distributed in an almost homogeneous manner at the surface of the titanium plate. Figure 1b shows the EDX spectrum of Pt-NPs/Ti electrodes, and the result confirms the presence of Pt-NPs on the titanium palates.

OPEN IN VIEWER

Figure 1

a surface morphology of Pt-NPs deposited on titanium plate and b EDX of Pt-NPs/Ti electrodes

Characterisation of Pt-NPs/Ti electrode surface

To determine whether the electrodeposition procedure had resulted in the removal of the oxide layer, thereby ensuring good electrical contact between the platinum deposit and the underlying substrates, the Pt-NPs/Ti were tested as electrodes using a one-electron redox couple. Figure 2 shows the voltammetric curves for the reduction in K₃Fe(CN)₆ on smooth platinum, Pt-NPs/Ti and titanium electrodes. The voltammogram for the Pt-NPs/Ti electrodes shows the expected reversible behaviour for the reduction on a bulk platinum electrode. In comparison, the voltammogram obtained with one titanium electrode shows increased peak separation and peak widths. This is probably attributable to a passivating surface film, most likely the oxide layer present on the surface of the titanium electrode. The presence of such a passivating layer would shift the position of the peaks for the redox reaction of the ferri/ferrocyanide to greater overpotentials, increasing the peak separation and reducing the peak heights, as observed. The lack of such resistances and overpotentials observed on repeated redox cycling of Pt-NPs/Ti electrodes indicates that there is no significant resistive film between the underlying titanium and the deposited platinum film. It suggests that the adhesion and electrical contact property of the deposited platinum film with titanium is quite satisfactory, and because of the high surface area of Pt-NPs/Ti electrode, the current density is much more than the smooth platinum electrode.

OPEN IN VIEWER

Figure 2

Cyclic voltammograms for Pt-NPs/Ti (1×1 cm), smooth platinum (1×1 cm) and bare titanium (1×1 cm) electrodes recorded solution containing 10 mM K₃[Fe(CN)₆]+1M KCl at 25°C with scan rate of 100 mV s⁻¹

Electrocatalytic activity of Pt-NPs/Ti and smooth platinum electrodes for formic acid electro-oxidation

In order to compare the Pt-NPs/Ti electrode with the smooth platinum electrode, the cyclic voltammetry method was used to estimate the electrocatalytic behaviour of the electrodes. Figure 3 shows the comparison of oxidation of formic acid on the smooth platinum electrode and the Pt-NPs/Ti electrode. It can be seen from Fig. 3 that the cyclic voltammogram of Pt-NPs/Ti electrode (curve b) shows the usual characteristics of smooth Pt electrode (curve a) except that for both forward and reverse scan directions the oxidation currents of formic acid on the Pt-NPs/Ti electrode are significantly higher than on the smooth Pt electrode, indicating that the surface area of the Pt-NPs/Ti electrode was enlarged by the dispersion Pt-NPs on titanium plates. Figure 3, inset A, shows the cyclic voltammograms of 1 cm² smooth platinum electrode in 0·1M H₂SO₄ aqueous solution at a scan rate of 100 mV s⁻¹ without formic acid (dash line) and at present 0·1M formic acid (solid line). Cyclic voltammetry data were recorded for 1 cm² Pt-NPs/Ti electrode in 0·1M H₂SO₄ aqueous solution at a scan rate of 100 mV s⁻¹ without formic acid (dash line) and at present 0·1M formic acid (solid line), as shown in Fig. 3 (inset B). It can be seen from the cyclic voltammetry of formic acid oxidation on the Pt-NPs/Ti electrode that the reaction commences in the hydrogen region and proceeds slowly in the positive direction, and then reaches a plateau at about −0·20 V. At potentials with more than ∼0·00 V, the reaction becomes accelerated, and the maximum rate at ∼0·78 V occurs. Another increase in current at potentials more than ∼1·10 V is assigned to the reaction intermediates and poisonous adsorbed species to CO₂ and oxygen evolution reaction. Upon reversing the potential sweep, a very steep increase in the reaction rate at ∼0·48 V develops, and a maximum current is observed at ∼0·42 V. After that, the current gradually decreases, but the reaction rate is still faster than in the forward scan. This large anodic peak in the reverse scan is attributed to the removal of the incompletely oxidised carbonaceous species formed in the forward scan.28

OPEN IN VIEWER

Figure 3

Cyclic voltammograms for (a) Pt-NPs/Ti and (b) smooth Pt electrodes in 0·1M H₂SO₄+0·1M formic acid aqueous solution at 25°C with scan rate of 100 mV s⁻¹: (A) cyclic voltammogram for smooth platinum and (B) cyclic voltammograms for Pt-NPs/Ti in 0·1M H₂SO₄ without (dash line) and with (solid line) 0·1M formic acid

It has been widely accepted in the literature that formic acid in the forward scan is oxidised to CO₂ via a dual pathway mechanism, i.e. direct and indirect oxidation processes involving dehydrogenation and dehydration. In the direct pathway mechanism, i.e. dehydrogenation pathway, the HCOOH adsorbed on Pt sites is converted to CO₂ without the formation of poisonous CO_ads species (1) (2) The indirect reaction pathway via dehydration is proposed in the following (3) (4) The poisonous CO_ads is removed via a reaction involving the OH_ads species originating from water dissociation. While the net reaction is the same in both pathways, the water dissociation reaction is rather difficult. The OH_ads formed by the dissociation of water molecules on the Pt surface aids in removing the adsorbed surface poison CO_ads by oxidising it. However, indeed, this process is very intricate, as a higher potential is required for water activation on the Pt surfaces. Consequently, the electrode surface will be blocked by large amounts of CO_ads species, thereby hindering the further adsorption of other formic acid molecules on the electrode surface. This drawback will make only a few number of formic acid molecules to be oxidised as the electrode poison CO_ads remains on the electrode surface for a long time occupying active catalyst sites, thereby reducing the overall activity of formic acid oxidation. Hence, the rate of formic acid oxidation primarily depends on the amount of CO_ads removed from the electrode surface.28, 29

Cyclic voltammetric data under different scan rates were recorded on the Pt-NPs/Ti electrodes in a solution containing 0·1M formic acid–0·1M H₂SO₄, as shown in Fig. 4. It is seen from Fig. 4 that with the increase in the scanning rates, anodic currents increase, and a linear relationship of the anodic peak currents against the square root of scan rate ν^1/2 (inset) ensures that the reaction is a diffusion controlled process.

OPEN IN VIEWER

Figure 4

Cyclic voltammograms of formic acid oxidation on Pt-NPs/Ti electrode in 0·1M H₂SO₄+0·1M formic acid aqueous solution at different scan rates: (inset) dependence of formic acid oxidation peak current on square root of scan rates

The long term stability of Pt-NP/Ti was examined in 0·1M H₂SO₄ aqueous solution containing 0·1M formic acid (Fig. 5). It can be observed that the anodic current remains constant with an increase in the scan number at the initial stage and then starts to decrease after 50 scans. The peak current of the 200th scan is ∼96% than that of the first scan. After the long term cyclic voltammetry (CV) experiments, the Pt-NP/Ti was stored in water for a week; then, formic acid oxidation was carried out again by CV, and excellent catalytic activity towards formic acid oxidation was still observed. This indicates that the Pt-NP/Ti electrodes prepared in our experiment have good long term stability and storage properties.

OPEN IN VIEWER

Figure 5

Plot of anodic peak current in electro-oxidation of 0·1M formic acid as function of scan number in cyclic voltammetric method at scan rate of 100 mV s⁻¹ (long term stability of Pt-NP/Ti electrode)