Reactive Dynamics Modeling and Simulations to Predict Structures, Physicochemical Properties, Initiation Kinetics, and Detonation Performance of Novel Energetic Materials and Composites
Abstract
The overarching goal of this project is to provide guidance to help experimentalists make the most rapid progress in developing new generations of EM for future naval force applications To do this we propose to build on recent progress in quantum mechanics (QM) and in reactive dynamics methods that retain the accuracy of QM for large scale (millions of atoms) models ofrealistic models of EM composites to develop a new generation of first principles based simulation tools that predict accurately crystal structures, density, performance, stability, and sensitivity prior to experiment. We propose to then use these tools to help navy chemists identify the most promising materials worthy of synthesis and characterization prior to experiment. We also propose to use high throughput in silico combinatorial chemistry to discover new promising leads for experimental study.Prior to experimental synthesis and characterization, we intend to predict for proposed materials:a) crystal structures, density, heat of formation, constitutive and chemical properties ofb) performance stability, and initiation sensitivity,c) compatibility with major binders, solvents and additives used in composite formulations.Thus we propose to build a new generation of force fields for large scale simulations of up to 100nm (100 million atoms), that retain the accuracy of valence interactions for quantum mechanics (QM) while accurately describing van der Waals, hydrogen bonding, and electrostatic interactions for molecular, polarizable, and ionic solids and composites. We propose advancesampling methodology to extend the simulation timescale to microseconds and longer. We expect these developments to enable dramatic improvement in predicting the morphology, thermodynamics and kinetics of EM and composites.Also, we intend to use chemometric and machine learning techniques to extract correlations in complex properties to address phenomena too complex for straightforward simulations.This project proposes the following tasks:1. Build next generation of Reactive Force Fields (ReaxPoN) for EM by including our recent advances in ReaxPoN methodology and expand its training for improved description of hydrogen, ionic, and polar bonds (including acids and salts), and nitrogen-rich heterocycles.2. Optimize and validate our procedures for predicting crystal structures and polymorphs of new EM, using multi-objective fitness functions3. Develop simulation protocols with QM and FF methods to predict thermodynamic and physicochemical properties important for synthesis.4. Develop protocols to predict EM performance, such as shock and detonation velocities, Chapman~Jouguet (CJ) pressure and temperature, including cluster.5. Predict properties of polymer-crystal interfaces for new EM composites.6. Develop protocols for to predict dynamic shear initiation and sensitivity under shock compression.7. Use new first-principles based ReaxPoN to predict new classes co-crystals and 2D-composite explosives of extended solid structures with potentially game-changing properties.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Feb 07, 2019
- Source ID
- N000141912081
Entities
People
- William Andrew Goddard III
Organizations
- California Institute of Technology
- Office of Naval Research
- United States Navy