Numerical analysis of an inline metal-organic chemical vapour deposition process based on sliding-mesh modelling

Xiaosong Zhou, Yiyi Wu, Xiaogang Yang, Chaowen Huang

Research output: Journal PublicationArticlepeer-review

4 Citations (Scopus)

Abstract

The flow behaviour under the influence of susceptor moving speed is a key factor for the fabrication of high-quality cadmium telluride (CdTe) thin films during the inline metal-organic chemical vapour deposition (MOCVD) process. The main purpose of this paper is to find a method to study the real-time dynamics of transport phenomena inside the reactor. The sliding mesh method is thus proposed and its feasibility is evaluated using computational fluid dynamics (CFD) modelling. A computational grid with 173,400 hexahedral cells is adopted through a grid sensitivity test validation. The simulations show that comparing to 2D modelling, the results of 3D modelling are found to be in good agreement with the experimental data for the temperature range of 628–728 K. Based on the velocity field, the temperature field and distribution of species concentration under different sampling time intervals of 60, 180 and 300 s, the thin film uniformity on both edges of the substrate is found to be influenced by the side effect of the baffle plate. The mass deposited on the substrate is further investigated under different susceptor moving speeds from 0.75 to 2.25 cm/min, and a moving speed between 0.75 to 1.13 cm/min is found to be effectively beneficial to the deposition process.

Original languageEnglish
Article number1198
Pages (from-to)1-16
Number of pages16
JournalCoatings
Volume10
Issue number12
DOIs
Publication statusPublished - Dec 2020

Keywords

  • CFD simulation
  • CdTe thin film
  • Fluid flows
  • MOCVD
  • Sliding mesh

ASJC Scopus subject areas

  • Surfaces and Interfaces
  • Surfaces, Coatings and Films
  • Materials Chemistry

Fingerprint

Dive into the research topics of 'Numerical analysis of an inline metal-organic chemical vapour deposition process based on sliding-mesh modelling'. Together they form a unique fingerprint.

Cite this