Deposition of platinum catalyst by plasma sputtering for fuel cells: 3D simulation and experiments

Plasma sputtering is one of the most promising methods for reducing the amount of platinum catalyst in porous electrodes for low temperature fuel cells. Here, a simulation of the platinum deposition by radio frequency plasma sputtering has been developed and compared with experimental results to allow optimization of the deposition process. In the simulation, the transport of sputtered atoms through the argon plasma is obtained using a 3D Monte Carlo model called SPaTinG (Sputtered Particles Transport in Gas). The Yamamura formula provides the Pt sputtering yield on the target, and the initial energy distribution of sputtered atoms is given by the Thompson distribution. A 1D hybrid model is used to estimate the mean energy of argon ions impinging onto the platinum target. Experimentally, platinum is deposited on silicon in two plasma sputtering chambers with different geometries. The deposition rate is measured by Rutherford backscattering spectroscopy. The angular distribution of the Pt atoms ejected from the target surface and the condensation coefficient of the Pt atoms on silicon are calculated by adjusting the simulated and experimental deposition rates at 0.5 Pa. A good agreement between the simulation and the experiment is observed as a function of the target-substrate distance for the two system geometries at low pressure (0.5 Pa).