Data underlying the publication: Computational Investigation of Precursor Blocking during Area-Selective Atomic Layer Deposition Using Aniline as a Small-Molecule Inhibitor
doi: 10.4121/0bb03d8b-d1ac-49bc-8010-a498a981cf3e
This dataset includes raw and processed data published in manuscript "Computational Investigation of Precursor Blocking during Area- Selective Atomic Layer Deposition Using Aniline as a Small-Molecule Inhibitor". The study focuses on area-selective atomic layer deposition (ASD) using small-molecule inhibitors (SMIs) to achieve selective growth on specific surfaces. Aniline (C6H5NH2) was identified as an effective SMI for ASD of TiN, enabling selective growth on SiO2 while inhibiting growth on Ru and Co non-growth areas. We used density functional theory (DFT) and random sequential adsorption (RSA) simulations to understand how aniline achieves this inhibition. The DFT calculations confirmed that aniline selectively adsorbs on Ru and Co non-growth areas, while its adsorption on the SiO2 growth area is limited to physisorption. The RSA simulations showed that having two stable adsorption configurations of aniline allows for a high surface inhibitor coverage on both Co and Ru surfaces. Overall, the study provides insights into the mechanism behind area-selective atomic layer deposition using aniline as a small-molecule inhibitor, which can be crucial for developing industry-compatible approaches for selective growth in microfabrication processes. This dataset includes the originals of the published figures as well as raw DFT and RSA data to generate those images and plots.
- 2023-08-07 first online, published, posted
- European Research Council (ERC) under the European Union’s Horizon 2020 Research and Innovation Programme (Grant Agreement 949202)
- Dutch Research Council (NWO) Vidi Project 18363
- ANID/ ACT210059
Universidad Técnica Federico Santa María, Department of Chemical and Environmental Engineering, Santiago, Chile
DATA
- 608 bytesMD5:
637678160bf576033d18e7e0f44d87c0
README.txt - 310,603,554 bytesMD5:
e0da8ad3a7674d2a0575b56d13b39aa9
2023-Computational_investigation_of_precursor_blocking_during_ASALD.zip -
download all files (zip)
310,604,162 bytes unzipped