Enabling the Statistical Calibration of Active Nitridation of Graphite by Atomic Nitrogen

Abstract

A workflow is presented for using PDE-based models to simulate an experimental setup for inferring properties associated with carbon nitridation. The eventual goal is the statistical calibration of the parameters associated with the active nitridation of graphite carbon; this work is a stepping stone in this direction. Simple catalycity models are used to illustrate the the calibration workflow. This study builds from previous work where it was determined that a one-dimensional model used was inadequate to represent the physical mechanisms at work. The experimental work is briefly recapitulated and the mathematical model used for data reduction is presented. Details of the statistical calibration procedure and all assumptions are discussed. Results of the calibration procedure for the simple catalytic model will be presented and discussed. Extensions underway are highlighted.

Computational Modeling of Laser Absorption in Reacting Flows

“…(3)] and through the broadening mechanisms [Eq. (7)] in the line shape calculation. Thus, although it is clear that pressure is an important parameter, the overall trend is not clear a priori.…”

“…(25)(26)(27)(28)(29). For simplicity, we do not attempt to account for the sample holder, as was done in previous work [7]. Otherwise, we have replicated the experimental conditions as closely as possible.…”

“…We demonstrate an example application of this framework. Second, we wish to study the possibility of using this modeling approach together with TDLAS data to compute the solution of inverse problems for parameters in catalytic reactions, extending previous work in this direction [4,6,7]. The feasibility of using TDLAS for inverse problems is evaluated by showing that the variation of the computed laser absorption due to parameters in the catalytic reaction is much higher than variations due to other factors in an experimental setup, including pressure, laser angle, and laser position.…”

GRINS: A Multiphysics Framework Based on the libMesh Finite Element Library

“…Here, an extended Kalman filter approach was used to update the damage predictions based on measured strain data; the underlying forward model in GRINS supplied to forward model predictions derivatives. The final application is the inference of surface chemistry parameters [12]. This case uses a two-dimensional reacting flow model to study the consumption of carbon by dissociated nitrogen from which surface chemistry between carbon and nitridation can be inferred.…”