Continuum simulation surrogate dataset for AR 1000 PillarHall structure - Trained surrogate models for parameter to depth profile mapping - Models include Thermal ALD, PEALD, non-self-limiting growth and desorption effects.

Supplementary Dataset for the Master Thesis of Freek Klabbers, titled: "Physics-Informed Bayesian Estimation of Physical Parameters for Atomic Layer Deposition". This repo includes the following: Data used to train surrogate models for the continuum simulations of profile growth in an AR 1000 PillarHall trench structure. The data is split into "Dynamic" and "Flat" ranges. The flat data is uniformly sampled over the entire range and can often result in a fully saturated or unsaturated profile. To give the neural network something to train on, the dynamic range was heuristicaly determined which almost always give a curve from 1 to 0 in the valid range. Weights for trained surrogate models. The surrogate models that have been trained have been added to save on compute and for convience The Python script on how to load the models - Without knowing the arcitecture you cant use the models. The python file contains this. What is not added in the repo, is the literature data which has been used to validate the model. This is due to copyright reasons. The data can be extracted from the respective sources: [1] E. Haimi, ‘Atomic layer deposition of zinc oxide films on lateral high-aspect-ratio test structures using diethylzinc and water as precursors | Journal of Vacuum Science & Technology A | AIP Publishing’. Accessed: Mar. 17, 2026. [Online]. Available: https://pubs.aip.org/avs/jva/article/44/2/022410/3382222/Atomic-layer-deposition-of-zinc-oxide-films-on [2] F. Gao, S. Arpiainen, and R. Puurunen, ‘Microscopic silicon-based lateral high-aspect-ratio structures for thin film conformality analysis’, Journal of Vacuum Science & Technology A: Vacuum, Surfaces, and Films, vol. 33, p. 010601, Jan. 2015, doi: 10.1116/1.4903941. Note that data labelled Purunen was taken from Zenedo [3] was published by Yim [4]: [3] R. Puurunen and O. Ylivaara, ‘ALD Al2O3 thickness profile in microscopic rectangular channel, TMA-water 300 degC 500 cycles, PillarHall(TM) LHAR3-1b, V0001’. Zenodo, May 31, 2021. doi: 10.5281/zenodo.3974438. [4] J. Yim et al., ‘Saturation profile based conformality analysis for atomic layer deposition: aluminum oxide in lateral high-aspect-ratio channels’, Oct. 2020, doi: 10.1039/D0CP03358H. [5] Utriainen, ‘Optical metrology of 3D thin film conformality by LHAR chip assisted method’. Accessed: Jan. 19, 2026. [Online]. Available: https://www.spiedigitallibrary.org/conference-proceedings-of-spie/12008/120080D/Optical-metrology-of-3D-thin-film-conformality-by-LHAR-chip/10.1117/12.2609643.full [6] M. Ylilammi, O. M. E. Ylivaara, and R. L. Puurunen, ‘Modeling growth kinetics of thin films made by atomic layer deposition in lateral high-aspect-ratio structures’, J. Appl. Phys., vol. 123, no. 20, p. 205301, May 2018, doi: 10.1063/1.5028178. [7] K. Arts et al., ‘Sticking probabilities of H2O and Al(CH3)3 during atomic layer deposition of Al2O3 extracted from their impact on film conformality’, Journal of Vacuum Science & Technology A: Vacuum, Surfaces, and Films, vol. 37, no. 3, p. 030908, May 2019, doi: 10.1116/1.5093620.