Investigation of nanoscale mechanisms driving erosion and pitting in graphitic carbon

Edward, Sharon

Permalink

https://hdl.handle.net/2142/132497 Copy

Description

Title

Investigation of nanoscale mechanisms driving erosion and pitting in graphitic carbon

Author(s)

Edward, Sharon

Issue Date

2025-11-21

Director of Research (if dissertation) or Advisor (if thesis)

Johnson, Harley T.

Doctoral Committee Chair(s)

Johnson, Harley T.

Committee Member(s)

• Admal, Nikhil C.
• Ertekin, Elif
• Panerai, Franceso

Department of Study

Mechanical Sci & Engineering

Discipline

Mechanical Engineering

Degree Granting Institution

University of Illinois Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

Kinetic Monte Carlo (KMC), Carbon oxidation, Graphite oxidation, Gas-surface interaction, Direct Simulation Monte Carlo (DSMC), Gas-structure coupling, Stochastic simulation, Ablation modelling

Abstract

Understanding the fundamental mechanisms that causes oxidative pitting in carbon materials is essential to accurately predict ablation rates of Thermal Protection Systems (TPS) during hypersonic reentry. The extreme heat and unique gas phase chemistry encountered during hypersonic reentry necessitates an analysis of the oxidation mechanisms at these conditions. Despite extensive research, the nanoscale mechanisms that facilitate ablation in carbon materials used for TPS are largely unexplored. During reentry, carbon fiber materials used for TPS undergo oxidative pitting that significantly deteriorates the structural integrity of the materials. At the nanoscale, these carbon fibers are composed of graphitic units at various orientations. The current thesis is focused on the understanding of reaction kinetics and gas-phase behavior on hyperthermal oxidative pitting and erosion of graphite at the nanoscale. The results demonstrate the consequences of atomic-scale influences and interplay between competing reactions on pitting and erosion at larger scales. An atomistic kinetic Monte Carlo (AKMC) model is developed to understand pitting in multi-layer graphene, initiated by adsorbed oxygen surface groups on the graphene basal plane. A carbon-oxygen reaction model from prior studies is modified and used, with the main modification being the addition of reactions that result in defect initiation from an initially pristine graphene surface. The AKMC model allows for the adsorption of O and O2 with CO being the main product of gasification, while a number of surface groups containing epoxies, carbonyls, lactones and ethers are formed throughout the pitting process. This AKMC model serves as the foundation for four key investigations in this thesis. In the first investigation, the AKMC model is first used to examine oxidative pitting behavior in graphite under system parameters such as temperature (1300 K to 2200 K) and pressure (0.1 to 10 kPa). Simulations reveal a significant dependence of the pitting behavior on temperature and pressure. The pit shapes change from hexagonal to circular to branched with temperature increase due to competing etching rates on zigzag and armchair edge sites. The rates of carbon removal increase with both temperature and pressure, where the increase in this rate with temperature is found to be more significant than the increase in the rate with pressure. Epoxy diffusion is identified to play a crucial role in the pitting process, specifically in transporting chemisorbed oxygen initially adsorbed on the basal plane towards edge sites for gasification into CO molecules. Furthermore, results show that the contribution of epoxy and its diffusion towards the pitting rate varies with temperature and the state of graphite erosion. In the second investigation, time-varying data from the AKMC simulations are next exploited to reveal information about the relationship between reaction kinetics and the evolving graphite structure. The aim of this work is to determine the time varying influence of two adsorption mechanisms, that is, adsorption of oxygen on the basal plane as epoxy groups and adsorption of oxygen directly on the edges of defects. Simulations reveal that the influence of basal plane oxygen adsorption relative to that of defect edge oxygen adsorption on pitting decreases as pitting progresses. This is due to a reduction in basal plane sites and an increase in defect edge sites as carbon atoms are removed. Furthermore, it is observed that the majority of CO formation results from oxygen adsorbed on the basal plane, which subsequently diffuses towards defect edges. This is due to the greater availability of basal sites relative to defect edges throughout most of the simulation despite its decrease, and the rapid diffusion of epoxy groups towards defect edges at the studied temperatures (1700-2200 K). Since the AKMC model uses rate constants mainly obtained from first principles, the third investigation focuses on the propagation of uncertainty from first principle simulations to stochastic simulations for a carbon-nitrogen system. CN formation and nitrogen adsorption are identified to be the major contributors towards nitridative pitting in graphene from AKMC simulations. A sensitivity analysis is performed to understand how the uncertainty in the rate constants of these two reactions affect the uncertainty in pitting rates from the AKMC simulation. The uncertainty in the reaction rates comes from the various density functional formulations that can be used to compute energy barriers in DFT, or uncertainty in the measurement of parameters used in the rate constant such as gas density in the N adsorption rate. The sensitivity analysis reveals that pitting is governed by a competition between N adsorption and CN formation, where the uncertainty in the pitting rate is more sensitive to the uncertainty of the limiting reaction. Additionally, it is observed that under identical relative uncertainties in the CN formation activation energy or N gas density, the uncertainty in the pitting rate scales more significantly with uncertainties in the CN formation activation energy than with those in N gas density. The fourth investigation is a comprehensive work to account for the gas-phase behavior of the oxygen gas towards pitting in graphite. Additionally this behavior is investigated in graphite oriented at various angles of tilt, representing the differently aligned graphite units in the internal structure of the carbon fiber. A one-way coupling is introduced between the particle-based gas flow method, direct simulation Monte Carlo (DSMC), and the material evolution method, AKMC, at the nanoscale. Simulated gas particles in DSMC have a one-to-one mapping with real atoms or molecules, and the two methods are spatially coupled such that particles from DSMC can directly trigger adsorption events in AKMC. The results show that the coupled method more accurately captures gas behavior on pit geometries in comparison to results with a pure uncoupled AKMC method. This is due to a realistic oblique incidence angle distribution for gas particle impact on the material which is typically oversimplified or approximated in regular uncoupled AKMC. Comparison between results from the coupled and uncoupled methods reveals a discrepancy between their predicted pit geometries, especially at lower angles of graphite tilt where the regular uncoupled AKMC overpredicts the depth or narrowness of the pit relative to its width. Since the pit geometry significantly influences the structural stability of carbon fibers, this discrepancy is important to acknowledge. The results also report the trend in pit geometry under realistic gas incidence across various angles of graphite tilt. In addition, an exploratory analysis is performed in DSMC to determine if CO molecules emitted from the oxidizing graphite affect the oxygen number density local to the graphite, indicating the need for a two-way coupling between DSMC and AKMC. Several challenges associated with the analysis and coupling of the AKMC method with other methods due to its stochastic nature are also addressed in the present work. The thesis work is a contribution towards the nanoscale understanding of the mechanisms that influence pitting in graphite, and demonstrates that AKMC serves as a useful method to study nanoscale pitting at relatively long timescales. The completed work provides a basis for continued investigation in this area, with the overarching aim of improving the understanding of ablation in Thermal Protection Systems for hypersonic reentry.

Graduation Semester

2025-12

Type of Resource

Thesis

Handle URL

https://hdl.handle.net/2142/132497

Copyright and License Information

Copyright 2025 Sharon Edward

Owning Collections