The growth of supported Pt nanoparticles at room temperature employing a three-step atomic layer deposition (ALD) process, involving exposures to MeCpPtMe3, O2 plasma, and H2 plasma, has been investigated. From spectroscopic ellipsometry and transmission electron microscopy measurements it has been established that up to 300 cycles of ALD nanoparticles are formed by island formation and island growth. In situ infrared spectroscopy has been used to obtain more insight into the surface chemistry by determining which species are present at the surface during the different stages of nucleation as well as within one ALD cycle. After precursor exposure the surface is covered with a carbonaceous layer, originating from the precursor ligands or (de)­hydrogenated fragments thereof. Also adsorbed CO is present, which is already formed in the preceding H2 plasma step. The O2 plasma removes both the carbonaceous layer and the CO. Furthermore, the surface region of the nanoparticle is oxidized by the O2 plasma; i.e., PtO x and Pt–OH are formed at the surface. The subsequent H2 plasma converts the PtO x back into Pt and removes the Pt–OH. The oxidizing and reducing properties of the O2 and H2 plasma have also been observed through changes in free-carrier absorption. Overall, the experiments resulted in a refined understanding of the reaction mechanism of Pt nanoparticles grown by ALD at room temperature.

The surface reactions during atomic layer deposition (ALD) of SiNx were studied using in situ attenuated total reflection Fourier transform infrared spectroscopy. Specifically, di(sec‐butylamino)silane (DSBAS) and bis(diethylamino)silane (BDEAS) were used as the silicon precursors with N 2 plasma as the nitrogen source for SiN x ALD over a temperature range of 225–375°C. The infrared spectra recorded during each ALD half‐cycle provide unambiguous experimental evidence that surface secondary amines (>NH) are the primary reactive sites for chemisorption of DSBAS and BDEAS on a SiN x surface that was exposed to an N 2 plasma. Based on these observations, we predict that most aminosilane precursors will primarily react with surface >NH groups: This observation is contrary to most atomistic‐level simulations for this reaction that predict a high activation energy barrier.