UNCLASSIFIED U S Army Research Development and Engineering

  • Slides: 19
Download presentation
UNCLASSIFIED U. S. Army Research, Development and Engineering Command Intermolecular Forces in Army Research

UNCLASSIFIED U. S. Army Research, Development and Engineering Command Intermolecular Forces in Army Research De. Carlos E. Taylor August 2, 2012 UNCLASSIFIED

EM Design Candidate EM PROPERTIES Computer Model Quantum Mechanics • Structure • Heat of

EM Design Candidate EM PROPERTIES Computer Model Quantum Mechanics • Structure • Heat of formation • Density • Decomposition pathways • Mechanical properties • Shock response • Thermal stability • Sensitivity EM performance is proportional to crystal density! Detonation Velocity D = a + b * ρ Detonation Pressure P ~ ρ * D 2 3% Rao Surapenani, ARDEC

Octanitrocubane 20 YEARS! Millions of $$$ Idea: Fully nitrate cubane! • High energy content

Octanitrocubane 20 YEARS! Millions of $$$ Idea: Fully nitrate cubane! • High energy content • High density SUPER EXPLOSIVE!!! Low density! Approved for public release: distribution unlimited Performance expectations were not met!

QM Methods in Army Research Semiempirical NDDO (AM 1, PM 3, MNDO/d) Perturbation Theory

QM Methods in Army Research Semiempirical NDDO (AM 1, PM 3, MNDO/d) Perturbation Theory MBPT(2), SAPT Quantum Mechanics Coupled Cluster Theory CCSD, CCSD(T), EOM UNCLASSIFIED Density Functional Theory

Computational Research in EMs Virtual Design • New Energetic Materials • Energetic Formulations •

Computational Research in EMs Virtual Design • New Energetic Materials • Energetic Formulations • “Computational Toolkit” Multiscale Response of Energetic Materials • Development of meso-scale models of heterogeneous EM • Development of models relating hot spot dynamics to microstructure • Bottom-up meso-particle dynamics models • Virtual testing of EM in munitions Disruptive Energetics ·QM characterization of non-conventional energetics ·Release of stored energy in ND ·Dynamic response of shocked poly-N Approved for public release: distribution unlimited

Computational toolbox Crystal Densities • Using quantum mechanics, we have derived correlations to solid

Computational toolbox Crystal Densities • Using quantum mechanics, we have derived correlations to solid phase heats of formation and crystalline densities for both neutral and ionic energetic materials Neutral materials rms: 3. 6% Ionic materials rms: 4. 0% • Correlations require calculations only on single molecule (not bulk material) CL 20 Mapping out e Density Neutral molecules refit to improve the RMS error to 2. 9%! + Heats of Formation Neutral materials rms < 6 kcal/mol Electrostatic Potential e rich Ionic materials rms error (kcal/mol) Comments 25. 0 Requires knowledge of crystal structure 24. 0 Requires QM information of single molecule! e poor E. F. C. Byrd and B. M. Rice, “Improved Prediction of Heats of Formation of Energetic Materials Using Quantum Mechanical Calculations”, Journal of Physical Chemistry A (2006) 110, 1005 -1013; ibid (2009) 113, 5813. B. M. Rice and E. F. C. Byrd, “Accurate predictions of crystal densities using quantum mechanical molecular volumes”, Journal of Physical Chemistry A (2007) 111(42), 10874 -1087 Approved for public release. Distribution is unlimited. Presented under the auspices of DEA 1060

Computational Toolbox David Chavez (LANL) BNDD Phil Leonard (LANL) r(g/cc) = 1. 914 (1.

Computational Toolbox David Chavez (LANL) BNDD Phil Leonard (LANL) r(g/cc) = 1. 914 (1. 870) DHs (kcal/mol) = 145. 1 (142) Phil Pagoria (LLNL) Damon Parrish (NRL) Jeff Deschamps (NRL) Ripu Malholtra (SRI) David Tevaul (ECBC) 3, 6 -Bis(4 -nitro-1, 2, 5 -oxadiazol-3 -yl)-1, 4, 2, 5 -dioxadiazene (BNDD): A Powerful Sensitive Explosive. Leonard, Philip W. Pollard, Colin J; Chavez, David E. ; Rice, Betsy M. Parrish, Damon A. Tom Klapotke (LMU) SYNLETT 14 , 2097 (2011). Al Stern (NSWC) Matt Sherrill (ARL) Reddy Damavarapu (ARDEC) “Have I told you lately how much I LOVE the new script! It makes Michael Miller (ARDEC) everything so much easier! Thanks! You are the best!” 12/12/2011 Non-expert User Toolkit (Designed by Ed Byrd) 1. Simple Input (xyz) 2. Submit one job, answers pop out. 3. Write paper and wow your friends with your theoretical acumen. We have a special DSRC account for you to use (“The Sandbox”)! Matt is our control. If it works for Matt, it will work for anyone

Disruptive Energetics Polymeric CO Polymeric Nitrogen Nanodiamonds Surface reconstruction 15 GPa New polymeric crystalline

Disruptive Energetics Polymeric CO Polymeric Nitrogen Nanodiamonds Surface reconstruction 15 GPa New polymeric crystalline phase! High Velocity Collisions

UNCLASSIFIED Hexanitrobenzene Fox-7 DATB Trinitrobenzene Trinitrotoluene Energetic Molecular Crystals TATB • Large unit cells

UNCLASSIFIED Hexanitrobenzene Fox-7 DATB Trinitrobenzene Trinitrotoluene Energetic Molecular Crystals TATB • Large unit cells • Condensed phase (periodicity) • Temperature and stress is important (MD) Trinitroaniline

DFT – Energetic Molecular Crystals 3% Maximum Allowable Error UNCLASSIFIED

DFT – Energetic Molecular Crystals 3% Maximum Allowable Error UNCLASSIFIED

Dispersion - TATB SAPT Interaction Energy Electrostatic=-4. 89 Induction=-5. 301 Dispersion=-16. 47 Total=-10. 05

Dispersion - TATB SAPT Interaction Energy Electrostatic=-4. 89 Induction=-5. 301 Dispersion=-16. 47 Total=-10. 05 SAPT Interaction Energy Electrostatic=-5. 43 Induction=-7. 87 Dispersion=-22. 47 Total=-11. 65 SAPT Interaction Energy Electrostatic=-3. 87 Induction=-2. 16 Dispersion=-5. 12 Total=-4. 80 UNCLASSIFIED SAPT Interaction Energy Electrostatic=0. 44 Induction=-2. 68 Dispersion=-8. 52 Total=-3. 60 SAPT Interaction Energy Electrostatic=-4. 63 Induction=-3. 08 Symmetry Adapted Perturbation Theory Dispersion=-7. 57 Total=-5. 93 • Fit intermolecular potential (exp-6) • 900 ab initio data points SAPT Interaction Energy Electrostatic=-4. 07 • Minima on fitted surface analyzed with ab Induction=-1. 84 initio SAPT(DFT) Dispersion=-4. 06 Total=-4. 47 SAPT Interaction Energy Electrostatic=-4. 47 Induction=3. 71 Dispersion=-9. 22 Total=-6. 09

Fox-7 SAPT Potential Experiment Maximum Error Edge lengths: 0. 83% Cell angles: 0. 24%

Fox-7 SAPT Potential Experiment Maximum Error Edge lengths: 0. 83% Cell angles: 0. 24% Density: 0. 37% UNCLASSIFIED

Dispersion Corrections SOFTWARE • CP 2 K • VASP Dispersion Corrected Atom-centered Potentials (DCACPS)

Dispersion Corrections SOFTWARE • CP 2 K • VASP Dispersion Corrected Atom-centered Potentials (DCACPS) DFT-D* (Grimme) T) ( SD CC Fitted to CCSD(T) interaction energies UNCLASSIFIED

DCACPS – Energetic Molecular Crystals RDX TATB HMX PETN UNCLASSIFIED

DCACPS – Energetic Molecular Crystals RDX TATB HMX PETN UNCLASSIFIED

% error in density Ambient pressure ionic high nitrogen EMs UNCLASSIFIED

% error in density Ambient pressure ionic high nitrogen EMs UNCLASSIFIED

DFT-D Energetic Molecular Crystals TATB HMX PETN UNCLASSIFIED

DFT-D Energetic Molecular Crystals TATB HMX PETN UNCLASSIFIED

Motivation Quantum Mechanics Is Foundation Of Our Program! Continuum “What is the state of

Motivation Quantum Mechanics Is Foundation Of Our Program! Continuum “What is the state of the art? ” Polycrystal • The number of different approaches Single crystal Subgrain New density functionals Virtual orbital approaches Pseudopotential methods Empirical Corrections Atomistic • Different benchmark systems

Coupled Cluster Benchmarks Water Imidazole Nitromethane Methyl Formate Nitrobenzene Benzene-Methane EDNA 6. 5 8.

Coupled Cluster Benchmarks Water Imidazole Nitromethane Methyl Formate Nitrobenzene Benzene-Methane EDNA 6. 5 8. 5 -1 -2 CCSD(T) -3 CBS -4 -5 2 Million CPU Hours 10 12 -2 -4 CCSD(T) -6 CBS -8 -10 -6 8 0 10. 5 0 -7 Fox-7 6 Energy(kcal/mol) 4. 5 Ethanol R(Ang) R (Ang) 2. 5 Benzene-Water -12 -14

Discussion What is “best” option for advancing the needs of the Army? • Density

Discussion What is “best” option for advancing the needs of the Army? • Density Functionals vs. Empirical Corrections C 6 R-6 corrections do not change electronic structure “…all non-empirical attempts to introduce van der Waals interaction in DFT will finally end up with methods that will be at least at compex as the simplest wavefunction methods. ” • Perspectives on application to large systems