Properties of Pt/C nanoparticle catalysts synthesized by electroless deposition for proton exchange membrane fuel cell

Vulcan XC-72 carbon with particle size ∼40 nm serving as support was purchased from Fuel Cell Earth Co. (USA). All the chemicals were of analytical grade. Chloroplatinic acid hexahydrate (H₂PtCl₆·6H₂O), sodium borohydride (NaBH­₄), ethylene glycol (EG), potassium hydroxide (KOH), hydrochloric acid (HCl) (all from Merck) were used.

Pt/C catalysts particles were synthesized by the following route: at first Vulcan XC-72 carbon supports were pre-treated by heating at 773 K for 2 h. Afterward, the carbon supports were filled, washed with a huge amount of deionized (DI) water, and then dried for one hour at 337 K. In the next stage, the carbon supports were dispersed into the solvent (DI water with and without EG). The precursor H₂PtCl₆·6H₂O in concentration 10 g l⁻¹ was poured with an amount calculated to get the content of Pt to be 20 wt% on the supports. After mixing, the mixture was ultrasonicated for 1 h. An excess amount of reduction agent NaBH₄ 0.01 M was added and then this mixture was heated to 80 °C at 4 h. Finally, the synthesized catalyst particles Pt/C were filled, washed by DI water, and dried for 12 h at 120 °C [14, 17–19].

Chemical composition analysis of catalysts was performed by EDS method with a JED 2300 analyzer associated with a scanning electron microscope (SEM). The size of the Pt catalyst particles and the carbon supports was estimated by transmission electron microscopy (TEM). The TEM pictures of the Pt/C catalyst were captured by the JEM 1010 apparatus made by JEOL Corporation (Japan) with different magnifications at National Institute of Hygiene and Epidemiology. The sizes of the 100 Pt catalyst particles were measured directly on the TEM pictures in order to determine the distribution and mean diameter of the synthesized catalyst particles.

The catalyst ink was prepared by ultrasonication dispersing 6 mg of Pt/C 20 wt% in 6.2 ml H₂O; 2.0 ml isopropyl alcohol; 16.2 μl nafion 10% solution. After 1 h, a homogeneous solution ink was obtained.

The sample used in electrochemical measurements was a high-density carbon disc having a diameter of 15 mm. The sample surface was ground up to 2000 grid paper and then finished by chromium particles. Afterward, the sample was cleaned by ultrasonication in water and alcohol for 10 min. The catalyst layer was prepared by pipetting the catalyst ink on the sample surface. After reaching the Pt loading, the sample was dried in atmosphere at room temperature.

The cyclic voltammetry (CV) measurement was the most popular method to evaluate the activity and the durability of the catalyst Pt/C. The sample was placed into a teflon holder with a controlled working area of approximately 1 cm². A three-electrodes cell was used in which the counter electrode was platinum and the reference electrode was saturated calomel. In the CV measurements, the potential values were converted automatically to normal hydrogen electrode (NHE). The CV curves were conducted in H₂SO₄ 0.5 M solution and measured by the potentiostast PARSTAT2273 (EG&G —USA).

2.3.1. Activity of catalyst.

To investigate the catalyst's activity, the CV measurements were swept in the potential range from 0 to 1.3 V (NHE) with scan rate 50 mV s^–1 at ambient temperature. Figure 1 is a typical CV curve of Pt/C catalyst in H₂SO₄ 0.5 M solution. From the CV curve, it was easy to calculate the charge of the electrochemical processes which occur on the catalyst particles by integrating. As a result, the electrochemical surface area (ESA) showing the activity of the catalyst was estimated by the following formula [20–24]:

$$\begin{eqnarray*}&&{\rm ESA} = \frac{{Q_{\rm H}}}{{0.21m_{{\rm Pt}} }},\end{eqnarray*}\noindent$$

where Q_H is the average charge integrated from the voltammogram of the adsorbtion/desorbtion hydrogen process on the CV curve (mC), constant 0.21 shows the charge in theoretical calculation to oxidate a single hydrogen layer adsorbed on bright platinum (mC), m_Pt is the platinum loading on the surface sample (g cm⁻²).

Zoom In Zoom Out Reset image size

2.3.2. Durability of catalyst.

To examine the durability of the catalyst Pt/C, the CV characteristic of the sample was measured in the potential range 0.5–1.2 V (NHE) with 1000 cycles and scan rate 100 mV s⁻¹. In this potential range, Pt would be dissolved becoming ions in the forward scan. Therefore, the Pt loading in the sample would be reduced leading to the decrease of ESA value. In addition, in the reverse scan, Pt ion would be sintered so that the size of Pt catalyst particles increases, which would also lead to the decrease of ESA value of the catalyst. Prior to each CV measurement, the sample was polarized at the start potential value for 60 s.

The content of Pt in the catalysts is determined by energy dispersive x-ray spectroscopy (EDS) analysis. Figure 2 shows the EDS result of the commercial catalyst which indicates that the content of Pt in this catalyst is approximately 19.86%. Besides the two main elements C and Pt, there is a little amount of oxygen in the catalyst. Appearance of oxygen may be caused by contaminants during EDS analysis preparation. The contents of Pt in three catalysts are shown in table 1.

Zoom In Zoom Out Reset image size

Figure 3 is a TEM image of the commercial catalyst with Pt grains in black and carbon support in gray. It shows that the Pt particles have very small diameter spread homogeneously on spherical carbon grains. The histograms of the particle sizes show that the particle sizes range from 2 to 6 nm for Pt and concentrated in a range 3–4 nm.

Zoom In Zoom Out Reset image size

For the sample catalyst synthesized without supported EG, the TEM image is shown in figure 4. The TEM image also proves that the distribution in size is absolutely heterogeneous. The catalyst particles trends to become bigger due to agglomerate. The synthesized catalyst particles have big size, changing in a large range 4–16 nm. In contrast, when EG is used, the size of the synthesized Pt particles is reduced significantly in the range 1–4 nm and concentrated in the range 2–3 nm (figure 5). The distribution of Pt particles on the carbon powder is uniform. From this it may be interpreted that EG plays a role like an immediate bridge between carbon supports and precursors PtCl₆²⁻ and the capacity of adsorption of PtCl₆²⁻ precursors on carbon supports increases. As a result, the distribution of Pt particles is more uniform and the size of catalyst is reduced after the reduction process by NaBH₄. Therefore, the catalyst properties are improved considerably when EG is used.

Zoom In Zoom Out Reset image size

Zoom In Zoom Out Reset image size

The distribution and difference in the size of the catalyst particles strongly influence the activity of the Pt/C catalysts. As mentioned above, the electrochemical surface area (ESA) reflects the intrinsic electrocatalytic activity of a catalyst and its values can be calculated from the CV curve by integrating. Therefore, the CV method is used to evaluate this property. Figure 6 shows CV curves in H₂SO₄ 0.5 M solution with scan rate 50 mV s⁻¹ of the catalyst samples synthesized with and without EG in comparison with a commercial catalyst. On the CV curves, there are the electrochemical peaks corresponding to the different electrochemical reactions on the samples' surface. In the potential range 0–0.4 V, the peaks express the adsorb/desorb processes of hydrogen on Pt crystal. The mechanism of these processes may take place by two stages. In the forward scan, the potential range from 0.4 to 0.8 V corresponds to the charge of the double layer by the oxygenated groups on the carbon support surface. At the potential 0.85 V, the oxidation process of Pt metal happens to form Pt oxides. Corresponding with this process, there is a reduction peak at 0.7 V of reduction process of Pt–O in the reverse scan [3, 20, 22]. Among the peaks of the catalyst samples, those without EG are the smallest. This means that the activity of these samples is relatively low. In comparison with the commercial sample, the activity of the catalyst synthesized with EG might be the highest.

Zoom In Zoom Out Reset image size

The calculated ESA values of the catalyst samples are shown in table 2. The difference in ESA value may be due to the difference in the size of Pt particles. With the biggest size of particles, the ESA value of the sample without EG is the lowest. However, although possessing the smallest size, the ESA value of the synthesized sample with EG is significantly higher than the ESA value of the commercial sample. Thus, the catalysts with smaller particle size give a higher activity due to larger surface area.

Evaluating the durability of the Pt/C catalysts is based on declining of the ESA values after testing 1000 CV cycles. After each 200 cycles test, the sample activity is evaluated in order to observe the changes of ESA values of the catalyst. Figure 7 shows the CV curves to evaluate the activity of the catalyst sample synthesized with EG before and after the durability test. After each 200 cycles, the size of the peaks on the CV curve decreases significantly, leading to the decrease of the ESA value of the sample, and the activity of the catalyst is decreased considerably. There are two processes during the test which effect the decrease of the durability of the catalyst Pt/C. At first, the Pt particles are dissolved to the ions according to the following equations:

$$\begin{eqnarray*}&&{\rm{Pt }} \to {\rm{ Pt}}^{{\rm{2}} + } + {\rm{ 2}}e, \end{eqnarray*}\noindent$$

$$\begin{eqnarray*}&&{\rm{PtO }} + {\rm{ 2H}}^ + \to {\rm{ Pt}}^{{\rm{2}} + } + {\rm{ H}}_{\rm{2}} {\rm{O}}.\end{eqnarray*}\noindent$$

This process happens to directly affect the Pt catalysts' loading [8, 12, 25]. In addition, the durability is affected by the second process which is the corrosion of carbon material. In PEMFC, the carbon supports are corroded according to the following equations:

$$\begin{eqnarray*}&&{\rm{C }} \to {\rm{ C}}^ + + {\rm{ }}e, \end{eqnarray*}\noindent$$

$$\begin{eqnarray*}&&{\rm{C}}^ + + {\rm{ H}}_{\rm{2}} {\rm{O }} \to {\rm{ CO }} + {\rm{ 2H}}^ + + {\rm{ }}e, \end{eqnarray*}\noindent$$

$$\begin{eqnarray*}&&{\rm{CO }} + {\rm{ H}}_{\rm{2}} {\rm{O }} \to {\rm{ CO}}_{\rm{2}} + {\rm{ 2H}}^ + + {\rm{ 2}}e. \end{eqnarray*}$$

The resulting reaction

$$\begin{eqnarray*}&&{\rm{C }} + {\rm{ 2H}}_{\rm{2}} {\rm{O }} \to {\rm{ CO}}_{\rm{2}} + {\rm{ 4H}}^ + + {\rm{ 4}}e\end{eqnarray*}$$

is taken place at the potential 1.2 V.

Zoom In Zoom Out Reset image size

Corrosion of the carbon supports leads to the fall of Pt particles. Furthermore, corrosion of carbon supports also causes the poisoning phenomena of the catalyst on forming CO product [3, 24, 25].

Figure 8 shows the decrease of the ESA values of the catalyst samples after durability tests. After testing, the ESA values of all catalyst samples decrease. The ESA value of the commercial catalyst decreases the most. After 1000 cycles, the ESA values of the commercial catalyst decrease approximately 41.38%, while the ESA value of the catalyst synthesized with EG decreases around 28.3%. The synthesized catalyst without EG just has a decreased ESA value of 15%. As this catalyst has the biggest size, consequently it has the highest durability although the lowest activity. The catalyst synthesized with EG has high activity and durability.

Zoom In Zoom Out Reset image size