Files and links (1)
Free to read and download, Open Access
Dissertation
Multi-scale modeling of heterogeneous energetic materials in three roles: prediction, deduction and induction
University of Iowa
Doctor of Philosophy (PhD), University of Iowa
Autumn 2023
DOI: 10.25820/etd.006924
Abstract
Formulations of energetic materials rely on intrinsic (molecular) structure of a particular nitramine (CHNO) compound to engineer their responses to shocks. Based on decades of empirical data some of the most commonly used energetic materials (EMs) can be arranged in a hierarchy of sensitivity; from the most insensitive TATB, useful to prevent accidental initiation in storage, handling and transportation to the most sensitive compound PETN, used to trigger charges. Book-ended by these two compounds are a variety of other CHNO species with varying degrees of sensitivity. In multi-scale predictive models of shock-to-detonation transition (SDT), microstructure-aware burn models are employed to connect meso-scale hotspot dynamics with macro-scale energy release. Meso-scale calculations of hotspot ignition and growth rely on chemical reaction rates expressed in Arrhenius forms; these chemical kinetics models therefore are foundational to the construction of physics-based, simulation-derived meso-informed closure (reactive burn) models. However, even for well-studied energetic materials such as HMX, the reaction time scales and pathways provided by available reaction kinetic schemes vary widely. For practical, predictive calculations of SDT in HMX-based energetic materials (EM), uncertainties in reaction rate parameters can amplify to produce large uncertainties in predicted initiation envelopes, such as James curves or Pop-plots.
In this thesis, the multiscale model or the meso-informed ignition and growth (MES-IG) model which allows for connecting meso-scale hotspot dynamics to macro-scale shock-to-detonation transition (SDT) was used to perform a forward prediction of sensitivity to bound reaction chemistry models. Three global Arrhenius-form rate models that span the range of reaction time scales, namely the Tarver 3-equation, the Henson 1-equation and the Menikoff 1-equation models. The ability of the three reaction models to reproduce experimentally observed sensitivity is assessed by comparing the predicted criticality envelope (Walker-Wasley curve) with experimental data for pressed HMX Class V microstructures. The forward model tells you the chemistry models for HMX should be in the vicinity of Henson1 Tarver3 models. To better pin down reaction rate parameters for HMX, we perform an inverse analysis (deduction) to extract the rate constants in an Arrhenius model (assuming a single reaction global kinetics model) that best predicts an experimentally obtained criticality (James) envelope for a neat-pressed HMX material for which the microstructure is well characterized. Interestingly, the estimated global kinetic scheme is shown to conform to a previously proposed global kinetics model, i.e., the well-known Henson 1-equation model. To establish the validity of the inverse analysis, we further carry out a forward analysis using the determined reaction rate law and demonstrate good predictions of the Pop-plot for a different class of pressed HMX material. While the inverse analysis has been applied herein to HMX, this presents a general route to arrive at global kinetics models that can be applied to a wide class of EM species. We can explore the possibility of developing techniques to accelerate the calculation of sensitivity envelopes for the general class of CHNO materials. Such a technique would be highly beneficial in the prediction and design of EMs and their formulations. Therefore, we use the knowledge gained from single energetic species (HMX) for Induction: generalizing surrogate models for energy localization across CHNO species through Transfer Learning. In order to guide the accelerated development of meso-informed multiscale models for other HE species a multi output convolution gaussian process technique was used which uses the metamodel generated for neat pressed HMX as the foundation to obtain the generic response hypersurfaces F_CHNO. These low fidelity F_CHNO are then corrected using sparse DNS simulations in order to develop a high fidelity metamodel for each CHNO, eg. TATB and RDX. This process of transfer learning provides a route for rapid estimations of shock sensitivity for a wide range of species across the CHNO family.
Details
- Title: Subtitle
- Multi-scale modeling of heterogeneous energetic materials in three roles: prediction, deduction and induction
- Creators
- Prarthana Parepalli
- Contributors
- H. S. Udaykumar (Advisor)Chao Wang (Committee Member)Jia Lu (Committee Member)Hongtao Ding (Committee Member)Xuan Song (Committee Member)
- Resource Type
- Dissertation
- Degree Awarded
- Doctor of Philosophy (PhD), University of Iowa
- Degree in
- Mechanical Engineering
- Date degree season
- Autumn 2023
- Publisher
- University of Iowa
- DOI
- 10.25820/etd.006924
- Number of pages
- xiv, 123 pages
- Copyright
- Copyright 2023 Prarthana Parepalli
- Grants
- FA9550-15RWCOR123, United States Air Force Office of Scientific Research (United States, Arlington) - AFOSRFA955018-18RWCOR108, United States Air Force Office of Scientific Research (United States, Arlington) - AFOSR
- Grant note
- Multi-scale modeling of heterogeneous energetic materials in three roles : prediction, deduction and induction by Parepalli, Prarthana I would like to gratefully acknowledge financial support from the Air Force Research Laboratory Munitions Directorate (AFRL/RWML), Eglin AFB, under contract number FA8651- 16-1-0005 (Program manager: Dr. D. Barrett Hardin). Portions of this work are based upon work supported by the Air Force Office of Scientific Research under award number FA9550- 15RWCOR123 and FA955018-18RWCOR108.
- Language
- English
- Date submitted
- 11/09/2023
- Description illustrations
- illustrations, tables, graphs
- Description bibliographic
- Includes bibliographical references (pages 119-123).
- Public Abstract (ETD)
- Heterogenous energetic materials (HEs) commonly used in applications such as mining, munitions and propellants comprise of complex micro-structures. The sensitivity of heterogeneous energetic materials (HE) depends on their chemical (molecular) and physical (micro-) structure. For a wide range of energetic materials, the primary energetic components are organic CHNO crystals. The plastic bonded (PBX) or pressed explosives are packed with defects in the form of voids, cracks and interfaces in their meso-structures. When the HEs are subjected to an external impact or shock, localized heated zones or hotspots are generated at the defects. The ignition and growth of the hotspots at the meso-scale play a vital role to predict the response of HEs at the macro scale. The overall macroscopic sensitivity of HEs depends on a complex coupling of the molecular reaction chemistry and microstructural dynamics, due to the localization of energy at hotspots in the microstructure. Reactions triggered at hotspots advance into the unreacted sample, leading to shock-to-detonation (SDT) scenarios. The decomposition of the HE material from solid energetic crystals to gaseous products is modeled using global Arrhenius form of chemical reaction model. In this work, we perform multi-scale simulations to investigate the effect of uncertainties in the chemical kinetics parameters for the decomposition of the HE material on the rate of deposition of energy at the macro-scale. Ensembles of high-resolution reactive void collapse simulations are performed by varying the global Arrhenius parameters (representing a wide class of HE materials, ranging from insensitive TATB to highly sensitive PETN) to construct meso informed surrogate models for energy localization. Then macro-scale computations of shock-to detonation transition are performed using the meso-informed Ignition and Growth (MES-IG) model. The performance of the HE at the macro-scale is evaluated via the critical energy required for initiation in the Walker-Wasley/James space. The predicted critical energy envelopes are compared with experimental data. The results quantify the effects of uncertainties in the chemical kinetics parameters on the macro-scale sensitivity predictions. This study will guide the development of reaction kinetics models to reliably predict macro-scale sensitivity for a wide range of CHNO materials.
- Academic Unit
- Mechanical Engineering
- Record Identifier
- 9984546542302771
Metrics
1 File views/ downloads
11 Record Views