Back to results

University of Tennessee at Chattanooga

Modeling reactive rarefied flows in Chemical Vapor Infiltration using Direct Simulation Monte Carlo

Abstract

dc:description.abstract

Chemical Vapor Infiltration (CVI) is a key method for fabricating silicon carbide (SiC) matrix composites. Gas-phase precursors flow into a porous fiber, react, and deposit to form the ceramic matrix. Deposition quality and rate are strongly influenced by surface reactions, depending on temperature, gas flow, and reactor pressure. This study applies a computational model to better understand rarefied gas behavior and surface chemistry during CVI. We use the Direct Simulation Monte Carlo method to simulate gas flow and chemical reactions around the fibers. Simulations span Knudsen numbers 0.001–20 and temperatures 1000–1600 K (near‑continuum to free‑molecular). As rarefaction increases, deposition rate decreases, yet temperature remains significant. In forced‑flow CVI, deposition becomes uneven: the fiber's inlet side grows more, the opposite side less. To improve gas‑phase chemistry accuracy, we adapt a Quantum‑Kinetic model for methyltrichlorosilane and chlorine reactions. Overall, it clarifies how rarefied gas dynamics and activation energy control CVI.

Degree

thesis:*
Grantor dc:publisher
University of Tennessee at Chattanooga

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • EK, Ege C
Contributors dc:contributor
  • Barisik, Murat
  • Sreenivas, Kidambi; Ranjan, Reetesh
  • College of Engineering and Computer Science

Subjects

dc:subject × 5

Rights

dc:rights
Language dc:language
English, eng

Identifiers

dc:identifier.*
Repository record dc:identifier
https://scholar.utc.edu/theses/1021
OAI identifier oai:identifier
oai:scholar.utc.edu:theses-2207

Chain of custody

source
Harvested from
University of Tennessee - Chattanooga
Base URL
scholar.utc.edu/do/oai/
Last updated
2026-07-24
Source record
OAI-PMH GetRecord
citation

EK, Ege C. Modeling reactive rarefied flows in Chemical Vapor Infiltration using Direct Simulation Monte Carlo. University of Tennessee at Chattanooga, https://scholar.utc.edu/theses/1021