Scientific content of the collaboration:
Area-selective atomic layer deposition (AS-ALD) allows nanostructures of arbitrary shape and composition to be built with atomic precision. Most current approaches for S-ALD are based on local deactivation or activation. However, these blocking/activation layers have limited lateral resolution or are not compatible with self-aligned fabrication schemes. For this reason, exploiting inherent differences in nucleation on different substrates is of crucial importance for developing future S-ALD processes. We are focussing on the silicon-based materials Si, Si3N4, SiO2 and SiC, which are low-permittivity dielectrics used in electronics for various barrier applications. In order to see if there is selectivity towards one of the surfaces exposed to aminosilane precursor, we calculate the reactivity of this precursor with each surface using density-functional theory (DFT) approach. This project is funded by the company Lam Research, which is a member of HERALD.
We were interested in deposition on Si, SiC, SiO2 and Si3N4 starting surfaces by ALD using the aminosilane precursor DSBAS and compare the nucleation behaviour on these surface with our DFT results. To this end, PhD student (Ekaterina Filatova) from Tyndall National Institute, Cork, Ireland visited Dr. Adrie Mackus in the Plasma & Materials Processing (PMP) research group, at Eindhoven University of Technology, Netherlands. During this scientific visit, nucleation curves of SiNx ALD on amorphous a-Si:H and SiO2 starting surfaces were determined by measuring the film thickness using in-situ spectroscopic ellipsometry. Moreover, in order to analyze the DSBAS adsorption reaction on Si:H and SiO2:OH surfaces prior to a N2 plasma dose, in-situ ellipsometry measurements were performed after every half cycle.
In addition, as a part of a collaboration with Dr. Adrie Mackus’s group, Ekaterina performed computational screening using DFT to evaluate various precursors for area-selective deposition.