The electrodeposition of noble metals, particularly gold, on semiconducting surfaces is a common step in the development of optoelectronic and catalytic devices. Theoretical models of electrodeposition transients can provide mechanistic insights on nucleation and growth of metal particles. Here we study to what extent these models can be applied to the growth of gold nanoparticles on semiconductor photoelectrodes, with the ultimate purpose of being able to control in two dimensions the electrodepostion process. We have examined current transients for the reduction of aurate salts at Si(100) photocathodes and have made adjustments both to the experimental parameters as well as to the available models so as to account for parallel adsorption steps. We have observed to what extent these models hold predictive power for a nucleation process on semiconducting photoelectrodes. We found that the hydrogen evolution reaction is significant even at very basic pH values, leading to a poor match between the modelled and actual outcomes in electrodeposition experiments. We have concluded that the catalytic activity of gold particles and semiconductor photoeffects make it difficult to rely on current transients alone to refine the experimental conditions for the growth of gold particles on Si(100) photocathodes. Specifically, it is proposed that hydrogen evolution causes turbulence leading to the displacement of particles and significant aggregation on the surface. A solution to this problem is to electrodeposit metals with high overpotentials for hydrogen evolution, such as copper, which allows us to control nucleation and ultimately to use illumination patterns to spatially address nanoparticle deposition on semiconducting surfaces.
Read full abstract