No Cover Image

E-Thesis 301 views

Modelling and improving the vapour deposition efficiency of a novel PVD-based galvanising process / JAMES CARLOS

Swansea University Author: JAMES CARLOS

  • Redacted version - open access under embargo until: 16th July 2026

DOI (Published version): 10.23889/SUthesis.59492

Abstract

Physical vapour deposition (PVD) is being developed by Tata Steel as a continuous coating process for a steel substrate. Typically, the process is undertaken in low to medium vacuum conditions (0.001-0.1 Pa) to reduce contamination of the Zinc vapour and minimise impedance to the vapour flow towards...

Full description

Published: Swansea 2021
Institution: Swansea University
Degree level: Doctoral
Degree name: EngD
Supervisor: Lavery, N. ; Penney, D. ; Zoestbergen, E. ; Commandeur C.
URI: https://cronfa.swan.ac.uk/Record/cronfa59492
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract: Physical vapour deposition (PVD) is being developed by Tata Steel as a continuous coating process for a steel substrate. Typically, the process is undertaken in low to medium vacuum conditions (0.001-0.1 Pa) to reduce contamination of the Zinc vapour and minimise impedance to the vapour flow towards the substrate. This project centers around using computational models developed in OpenFOAM and ANSYS Fluent, to simulate PVD coating process with the aim of finding and optimising process parameters to enhance deposition yield. The lid-driven cavity benchmark model showed that the software packages mentioned, produced matching results to each other and against the benchmark data. Another benchmark, a vacuum interrupter model, was used to validate OpenFOAM’s DsmcFoam, which is a stochastic particle-based Direct Simulation Monte Carlo (DSMC) model. It allowed modelling of low pressure (high Knudsen number) flows and results showed similar deposition trends to the benchmark data. The inlet profile models focused on the coating uniformity and vapour distribution width. The models showed that an array of nozzles and slits produced wider deposition distribution when compared to a wide slit profile and, generally, nozzles produced a more uniform deposition distribution across the substrate when compared to slits. Additionally, for a specific inlet to substrate distance, nozzle spacing needs to be configured accordingly to ensure uniform deposition. For a gap of 20 mm, a nozzle spacing of either 6 and 10 mm; and for a 10 mm gap a spacing of 6 mm is suggested. Finally, the inlet to substrate models predicted coating yields of over 99% at low vacuum conditions (1-10 Pa). Generally, pressure is inversely proportional to the yield and in terms of reducing waste coating material, 10 Pa vacuum pressure is recommended. Moreover, the accretion model showed that coating chamber pressure is inversely proportional to the velocity of the vapour exiting the nozzle if the pressure of the vapour distribution box is kept constant. Importantly, the combination of these two models allowed the PVD coating process to be simulated through the transition to the continuum Knudsen regime.
Item Description: A selection of third party content is redacted or is partially redacted from this thesis due to copyright restrictions.
College: Faculty of Science and Engineering