Chemical Synthesis of Platinum-Rare Earth Alloy Catalysts for the Oxygen Reduction Reaction

Renewable energies are the ultimate solution to the ever-growing concern of the depletion of fossil fuels, global warming and subsequent climate changes. Hydrogen as an energy carrier has been suggested as an effective means of the renewable energy storage and conversion via electrochemical devices i.e. water electrolyzers and fuel cells(FCs), as the evolution and oxidation of hydrogen is electrochemically reversible processes with the highest efficiency of the energy conversion.

Using proton exchange membrane (PEM) as electrolyte a major challenge for the fuel cell technology is the slow kinetics of the oxygen reduction reaction on the cathode. Platinum nanoparticles supported on high surface area carbons is the benchmark catalyst for PEMFCs due to its high activity. Further enhancement of the catalytic activity and stability can be achieved by alloying of platinum with other metals in core-shell structures. Among the alloying metals, which are often the later transition metals, rare earth (RE) metals have been shown to possess superior ORR performance in forms of polycrystalline alloys and model nanoparticles. Due to the very negative reduction potentials and strong oxygen affinity of rare earth elements, the chemical synthesis of Pt-RE nanoalloys on carbon supports has been a challenge especially with respect to the control of the crystalline phases and particle sizes.

A unique chemical approach was recently developed to prepare Pt−RE nanoalloys from a mixture of solid-state precursors consisting of metal salts, a suitable nitrogen-rich compound and a carbon support. The synthesis involves in situ formation of a carbon-nitrogen network, onto which the two metals are atomically coordinated via the nitrogen sites. During the following heat-treatment in a dilute hydrogen atmosphere, the collapse of the network releases the metal ions, which are reduced and alloyed on the surface of the carbon support. The negative alloying free energy thermodynamically drives the reduction of rare earth metals. This thesis is devoted to further explore and optimize the method aiming at tuning the size and its distribution of the nanoalloy particles.

The alloy formation process was studied at 650 °C when a single alloy phase was obtained. At this temperature platinum is first reduced to Pt nanoparticles while the rare earth metal may disperse as intermediate species. Alloying of the rare earth metal with platinum occurs either on the already existing or simultaneously formed platinum nanoparticles. The alloy formation may be completed within a period of up to 30 minutes while any further heating results in excessive growth of the alloy particles. A key step for the size control seems to be the reduction of Pt sizes, which can be from the metal salt precursor or from the per-formed Pt articles on carbon.

The metal loading on top of carbon support does not seem to influence the alloy particles, when it is adjusted a range from 20 to 36 wt%. Carbon supports of varied specific surface areas play a critical role of the alloy particle growth. Carbon supports of high BET surface areas e.g. Black Pearl 2000, favors formation of small alloy particles but the particle size distribution is still an issue.

Instead of using the Pt salt precursor, pre-formed Pt nanoparticles on carbon support were ex situ prepared and explored as the platinum and carbon precursor. The formation of nearly pure crystal phase of nanoalloys has been demonstrated feasible.

For this purpose, a facile polyol method was employed to synthesize the Pt/C powder with controlled metal particle sizes from 1.5 to 8.8 nm by tuning the initial pH value of the polyol slurry. Using thus obtained Pt/C powders as precursor, Pt-RE alloys with tunable size and a relatively narrow size distribution have been synthesized. With Pt/C powders of initial Pt particle size of 2.5 ± 0.4, 5.0 ± 0.8, and 8.8 ± 2.0 nm, Pt₅Ce nanoparticless were successfully prepared having a mean alloy particle size of 4.5 ± 1.1, 6.1 ± 1.7 and 11.8 ± 4.6 nm, respectively. A uniform growth of the precursor Pt particles is observed during the alloy formation, which opens the opportunity to tailor the alloy particle size by varying the Pt/C precursor.

A low alloying temperature is desirable for the alloy particle size control. It has been shown that an alloying temperature as low as 550 °C is viable to obtain a pure Pt₅Ce alloy phase. It is the lowest temperature that has ever been reported in literature for complete alloy formation with no residual Pt particles identified.

The C-N network precursors have also been explored based on cyanamide and melamine, the latter being much less costly and toxic. The molar ratio of these two nitrogen precursors to the respective metal precursor is investigated, showing a minimum excess of the cyanamide or melamine in the precursor mixture exists in order to complete the alloy formation from the Pt/C powder and avoid additional growth of the alloy particles. The best samples prepared in this work are 30 wt% Pt₅Ce/C powders with a mean alloy particle size of 5.8 ± 1.8 nm from melamine and 6.4 ± 2.2 nm from cyanamide. The ORR performance of selected Pt₅Ce/C catalysts were evaluated by the RDE test. The best sample shows an ECSA of 30-40 m² g_Pt^-1, a specific activity of 2945 μA cm_Pt^-2 and a mass activity of 957 mA mg_Pt^-1, about 3.0 times and 1.7 times better than those for the reference Pt/C, respectively. Further enhancement of the ORR performance is expected by optimization of the alloy particle size and particularly the size distribution.