Session K1: EM-7: Detonation

A Continuum Theory for Shock Induced Heating of Metalized Explosive

A well-developed continuum field theory for Deflagration-to-Detonation Transition (DDT) in granular explosive is generalized to account for the existence of an arbitrary number of condensed phases and a gas product phase. Formulation of the more generic theory is motivated by a desire to model both the low and high pressure impact response of metalized explosive for which the metal and explosive grains may have distinct average densities, velocities, temperatures, and sizes. The theory is consistent with the strong form of the dissipation inequality and allows for flexible partitioning of dissipation between phases. The theory is applied to inert impact of aluminized HMX in the limit of low gas pressure. Emphasis is placed on characterizing the spatial structure of planar deformation waves and its dependence on impact speed and initial metal mass fraction.   

JAGUAR Procedures for Detonation Behavior of Explosives Containing Boron

The JAGUAR product library was expanded to include boron and boron containing products. Relationships of the Murnaghan form for molar volumes and derived properties were implemented in JAGUAR. Available Hugoniot and static volumertic data were analyzed to obtain constants of the Murnaghan relationship for solid boron, boron oxide, boron nitride, boron carbide, and boric acid. Experimental melting points were also utilized with optimization procedures to obtain the constants of the volumetric relationships for liquid boron and boron oxide. Detonation velocities for HMX - boron mixtures calculated with these relationships using JAGUAR are in closer agreement with literature values at high initial densities for inert (unreacted) boron than with the completely reacted metal. These results indicate that boron mixtures may exhibit eigenvalue detonation behavior, as observed by aluminized combined effects explosives, with higher detonation velocities than would be achieved by a classical Chapman-Jouguet detonation. Analyses of calorimetric measurements for RDX - boron mixtures indicate that at high boron contents the formation of side products, including boron nitride and boron carbide, inhibits the energy output obtained from the detonation of the formulation.   

Steady non-ideal detonations

Theories for determining the velocity of detonation (VoD) in highly non-ideal explosives, e.g. commercial explosives used in mining, are discussed. Such explosives have critical charge diameters of several centimetres. An analysis of the interaction between detonations and confining materials along the explosive-confiner interface reveals there a two main types of interaction. In the first (denoted here by case 1) the detonation drives an oblique shock into the confiner. For the second (case 2), a wave propagates in the confiner ahead of the detonation in the explosive. Shock polar interactions are examined for commercial explosives and rocks, which shows that a significant proportion of problems are case 2 in mining. For case 1, numerical simulations show that for a given explosive model there is a unique relationship (valid for all charge diameters and confinements) between the VoD and the curvature of the detonation shock at the charge axis. This relationship is shown to be well predicted by a quasi-one-dimensional type analysis. A simple detonation shock dynamics method which uses this relationships predicts well the VoD even in highly non-ideal cases, provided the explosive is sufficiently confined (usually the case in mining), but which is inaccurate in the limit of an unconfined charge. Preliminary results of a novel variational method for solving the unconfined situation are also discussed. Numerical simulations are performed to investigate the coupling mechanisms in case 2 situations, including the influence on diameter effects. It is shown that, in agreement with an approximate theory, the detonation is driven up to VoDs above the confiner's sound speed, and the wave in the confiner weakly pre-compresses the explosive ahead of the detonation front.   

Incorporation of a Chemical Kinetics Model for Composition B in a Parallel Finite-Element Algorithm

A thermal degradation model for Composition B (Comp B) explosive is being evaluated for incorporation into a finite-element algorithm [1]. The RDX component of Comp B dominates the thermal degradation since its decomposition process occurs at lower temperatures than TNT. The model assumes that solid and liquid RDX decompose by the same mechanisms, but along different reaction pathways [2, 3]. A steady-state approximation is applied to the gaseous intermediates and is compared to the full transient analysis for the entire reaction scheme. The parallel finite-element algorithm is used to predict the pressure increase on the interior of the metal casing of confined Comp B due to the production of gases during thermal decomposition. \\ \parindent=0pt \textbf{References} [1] E. M. Kallman, ``Scalable Cluster-Based Galerkin Analysis for Kinetics Models of Energetic Materials,'' SIAM CSE, March 2-6, 2009. [2] D. K. Zerkle, ``Composition B Decomposition and Ignition Model,'' 13th International Detonation Symposium, July 23-28, 2006. [3] J. M. Zucker, A. J. Barra, D. K. Zerkle, M. J. Kaneshige and P. M. Dickson, ``Thermal Decomposition Models for High Explosive Compositions,'' 14th APS Topical Conference on Shock Compression of Condensed Matter, July 31-August 5, 2005.   

Reactive Burn Modelling at Temperature Extremes Using CREST

The CREST reactive burn model uses entropy-dependent reaction rates to simulate behaviour in plastic bonded explosives. A CREST model for the TATB-based explosive PBX 9502, described at the last conference, was shown to be able to predict a range of shock initiation and detonation data at ambient temperature. However, it is well known that the behaviour of PBX 9502 varies significantly with initial temperature. Modelling the change in response that occurs upon heating or cooling the explosive, without having to modify the equation of state (EOS) and reaction rate parameters, is a significant challenge for reactive burn models. An important feature of CREST is that the initial state of the explosive can be incorporated without having to change the reference EOS or reaction rate model. In this paper, CREST is applied to PBX 9502 shock initiation data at temperature extremes. It is shown that the model can account for the variation in shock sensitivity with initial temperature using one set of parameters.   

Session K2: MD-3: Molecular Dynamics III

Analysis of $\alpha $-phase RDX vibrational lattice modes under hydrostatic pressure

Calculations employing density functional theory are performed on $\alpha $-phase RDX using the all-electron CRYSTAL06 program. The lowest frequency infrared active lattice modes are investigated as a function of hydrostatic pressure from ambient conditions up to 3 GPa. The strength of coupling between lattice and molecular modes as a function of pressure is examined. The anharmonic deviation of each mode from simple harmonic behavior as a function of pressure is also illustrated.   

A Theoretical Study of Vibrational Spectroscopy in Hydrostatically Compressed Nitromethane Crystal Using an Empirical Form Molecular Dynamics Force Field

Molecular dynamics simulations have been used with an unreactive but vibrationally accurate force field [Sorescu et al., J. Phys. Chem. B \textbf{104}, 8406 (2000)] to investigate the effects of isothermal hydrostatic stress on the vibrational spectrum of crystalline nitromethane. Power spectra corresponding to the classical vibrational density of states were obtained as Fourier transforms of the mass-weighted velocity-velocity autocorrelation functions at 298 K for seven hydrostatic pressures between 0.0 and 13.5 GPa. The goal is to provide an explanation for the pressure-induced shifts, splittings, and appearance/disappearance of bands in the infrared and Raman spectra observed in recent experimental [Ouillon et al., J. Raman Spectrosc. \textbf{39}, 354 (2008); Citroni et al., J. Phys. Chem. B \textbf{112}, 1095 (2008)] and electronic structure-based [Liu et al., J. Chem. Phys. \textbf{124}, 124501 (2006)] studies.   

Molecular Dynamics Simulations of Shock-Induced Defect Healing in Silicon

Molecular dynamics (MD) simulations of the interaction of planar shock waves with point defects (interstitials and vacancies, or Frenkel pairs) have been performed to investigate the possibility defect reduction in Si resulting from substantial mechanical stress behind the shock wave front. The MD shock experiments were run in Si samples containing Frenkel pairs of varying concentration and composition. The defect dynamics behind the shock wave front were studied as a function of the shock wave intensity and the crystallographic orientation of its propagation. We also simulated shock unloading that returns the compressed samples to their uncompressed state. The overall effectiveness of shock-induced defect healing was studied as well. Such an unusual application of the shock compression of solids might be useful in the microelectronics industry where such defects produced by ion implantation are considered a serious obstacle towards the further size reduction of Si CMOS devices.   

Density Functional Theory (DFT) Simulations of Shocked Liquid Xenon

Xenon is not only a technologically important element used in laser technologies and jet propulsion, but it is also one of the most accessible materials in which to study the metal-insulator transition with increasing pressure. Because of its closed shell electronic configuration, Xenon is often assumed to be chemically inert, interacting almost entirely through the van der Waals interaction, and at liquid density, is typically modeled well using Leonard-Jones potentials. However, such modeling has a limited range of validity as Xenon is known to form compounds at normal conditions and likely exhibits considerably more chemistry at higher densities when hybridization of occupied orbitals becomes significant. In this talk, we present DFT-MD simulations of shocked liquid Xenon with the goal of developing an improved equation of state. Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the United States Department of Energy's National Nuclear Security Administration under contract DE-AC04-94AL85000.   

Microscopic theory and kinetic model of fracture of liquids

Fracture kinetic model of liquids based on molecular dynamics simulations is presented. Stretched liquid appears as a result of large energy deposition to condensed matter, for example, under laser processing or shock-wave loading of materials. The kinetic model of fracture includes two processes: nucleation and growth of voids (NAG approach). The rates of nucleation and growth of voids are evaluated separately from molecular dynamics simulations on the example of Lennard-Jones liquid. Pressure and temperature dependences of nucleation rate can be approximated in the form of classical nucleation theory. The kinetics of void growth is shown to satisfy the hydrodynamic Rayleigh-Plesset equation. The fracture kinetics and spall strength are determined by means of the proposed model. The results of calculations show good agreement with the experimental data. This work was supported by the RAS programs {\#} 11, 12, and SNL under the US DOE/NNSA ASC program.   

Session K3: ED-2a: PDV Applications

Study on Dynamic Compression Properties of K9 Glass with Doppler Pins Array Measurements

K9 glass is one of archetypal brittle materials for studies of dynamic fracture, failure wave, and so on. This paper presented the dynamic compression properties of K9 glass under uniaxial strain condition. Experimental sample is K9 glass with internal pre-existed defects, and the shape of pre-existed defects is disc with less than 0.5 mm diameter. All tests were conducted by power gun with 37 mm diameter chamber. Doppler Pins array with high space-time resolutions, which consists of sixteen pins in range of 2 mm line length, were applied to measure the particle velocity histories in different positions at the sample rear surface, and the space-resolution is 127 $\mu $m, Experimental results show failure waves initiate at internal micro-surfaces of the sample under shock loading, and the dynamic stress concentration is likely attributed to be a physical mechanism of the initiation of the failure wave. These defects that by the controlled laser irradiation in advance are some internal micro-surfaces. Meanwhile, the experimental results show that internal micro-surfaces of the sample have influence on the elastic precursor wave decay.   

Improved Bar Impact Tests using a Photonic Doppler Velocimeter

Bar impacts were used to measure the dynamic strength of glasses. The conventional bar technique has been greatly improved through use of a photonic Doppler velocimeter (PDV) to measure free surface motion. The PDV records a compression pulse corresponding to compressive failure of the impact zone and a spall signal corresponding to tensile failure of the distal end. Best results were obtained using polished free surfaces, as opposed to retroreflective tape. Use of a graded density film had little effect on strain rate but reduced the peak transmitted stress. The experiments were interpreted with the aid of EMU (peridynamics) calculations. Indications are that the impact end of the bar fails in compression. The bar separates into two sections when the reflected tensile wave arrives at the zone of impact damage. The rear of the bar fails from an inward propagating failure wave that originates at surface flaws.   

Ultrafast Dynamics of Coherent Acoustic Phonons in the GaMnAs/GaAs Heterostructure System

Pronounced oscillations were found in the reflectivity curves of ferromagnetic GaMnAs/GaAs heterostructure using pump-probe spectroscopy that are caused by coherent acoustic phonons propagating through the sample. The difference in the oscillations period, damping and amplitude as the phonons travel across the GaMnAs/GaAs interface reflect sin electronic structures and optical properties of these materials. Analysis of the oscillation amplitude indicates that this method provides a novel, non-invasive, and non-destructive way to depth profiling.   

Line-imaging ORVIS measurements of interferometric windows under quasi-isentropic compression

A line-imaging optically recording velocity interferometer system (ORVIS) has been implemented on the Veloce pulsed power generator to enable measurement of spatially resolved velocity histories of materials under dynamic compression. Interferometric windows are regularly used to maintain the high-pressure state of shock and ramp (quasi-isentropic) loaded materials. Although imaging through a shock or rapid ramp ($\le $ 10 ns) loaded transparent window material has been reasonably successful, for slower ramp loading ($\sim $ 440 ns) experiments, the elastic-plastic yielding of the window has an adverse effect on return light to the line-imaging ORVIS. The results of quasi-isentropic loading experiments with various interferometric windows such as LiF, NaCl, SiO2, PMMA, and sapphire are presented. Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the U.S. Department of Energy's National Nuclear Security Administration under Contract No. DE-AC04-94AL85000.   

Session K4: ED-2b: Isentropic Compression

Quasi-isentropic compression of materials using the magnetic loading technique

The Isentropic Compression Experiment (ICE) technique has proven to be a valuable complement to the well-established method of shock compression of condensed matter. The magnetic loading technique using pulsed power generators was first developed about a decade ago on the Z Accelerator, and has matured significantly. The recent development of small pulsed power generators have enabled several key issues in ICE, such as panel {\&} sample preparation, uniformity of loading, and edge effects to be studied. Veloce is a medium-voltage, high-current, compact pulsed power generator developed for cost effective isentropic experiments. The machine delivers up to 3 MA of current rapidly ($\sim $ 440-530 ns) into an inductive load where significant magnetic pressures are produced. Examples of recent material strength measurements from quasi-isentropic loading and unloading of materials will be presented. In particular, the influence that the strength of interferometer windows has on wave profile analyses and thus the inferred strength of materials is examined. Sandia is a multiprogram laboratory operated by Sandia Corporation, a Lockheed Martin Company, for the U.S. Department of Energy's National Nuclear Security Administration under Contract No. DE-AC04-94AL85000.   

Sample Preheating Capabilities for Shock and Isentropic Loading Experiments at the DICE Facility

A system for preheating test specimens prior to shock or isentropic loading was developed at Sandia's Dynamic Integrated Compression Experimental (DICE) Facility. A dual-output, proportional-integral-derivative (PID) controller using feedback from thermocouples regulated power supplied independently to one or two resistive heaters so as to achieve the desired temperature(s) at selected position(s) within the test assembly. Thermal isolation features validated by finite element heat-transfer analyses afforded temperature uniformity across samples mounted in electrode panels for the Veloce pulsed electromagnetic driver. The preheat system was demonstrated during Veloce experiments on samples (e.g., tin) preheated up to 200 C, and during gas-gun tests. Temperatures exceeding 600 C for Veloce tests are possible, pending identification and qualification of an appropriate high-temperature insulator for the gap between the electrode panels.   

Design of a Sample Recovery Assembly for Magnetic Ramp-Wave Loading

Characterization of material behavior under dynamic loading requires studies at strain rates ranging from quasi-static to the limiting values of shock compression. For completeness, these studies involve complementary time-resolved data, which define the mechanical constitutive properties, and microstructural data, which reveal physical mechanisms underlying the observed mechanical response. Well-preserved specimens must be recovered for microstructural investigations. Magnetically generated ramp waves produce strain rates lower than those associated with shock waves, but recovery methods have been lacking for this type of loading. We adapted existing shock recovery techniques for application to magnetic ramp loading using 2-D and 3-D ALEGRA MHD code calculations to optimize the recovery design for mitigation of undesired late-time processing of the sample due to edge effects and secondary stress waves. To assess the validity of our simulations, measurements of sample deformation were compared to wavecode predictions.   

Explosive loading liner-type devices for generation of loading pulses having short durations

To investigate the phenomena of short-time softening in metals under effect of planar shock waves having amplitudes of 35 and 16 GPa when loading duration is less then 1 ms, two series of explosive loading devices are developed. In these devices, impactors are accelerated in the regime of sliding detonation. Loading devices with thicknesses of copper impactors of 1.0, 0.5, and 0.2 mm are presented in each series. The characteristic size of the investigated samples can be up to 90 mm in diameter. The paper includes the basic characteristics of the devices and results of their verifications.   

Shockless Compression Studies of HMX-Based and TATB-Based Explosives

Several HMX-based and TATB-based explosive samples along with their constituent binders were subjected to shockless compression to determine the material response at high stresses. A Velocity Interferometer System for Any Reflector (VISAR) was used to measure the transmitted wave profiles. The measured wave profiles were compared to calculated profiles generated from backward and forward analysis procedures using optimization methods. These results were used to determine the constitutive and equation of state (EOS) properties of the explosives and binders.   

Session K5: DC-1: Chemistry of Energetics and Reactive Metals

Radiation-induced Precursors in Crystalline Energetic Composites

We present new experimental evidence that demonstrates the origination of precursors of the major reaction front at SDT in PBXs, based on the results of wedge tests of the HMX/Epoxy 77/23 (wt. {\%}) and HMX/Water composites. The precursors were spatially resolved in the modified wedge tests performed with the Multi-Channel Optical Analyzer -- MCOA by means of the simultaneous registration of the reaction radiance transmitting through the explosive bulk at the SDT and the stress field, which is induced by the reaction zone in the optical monitor. Experimental evidence, obtained at a wide variation of the HMX particle sizes (1.64 $\mu$m $<$ d50 $<$ 960 $\mu$m), point to the fact that the precursor is arisen as a result of the radiation heating due to the photon absorption, as the reaction radiation is scattered within the bulk of the crystalline explosive material. Within the precursor layer, thickness of which depends on both, temperature localization and radiation intensity in the major DRZ as well as on optical and kinetic properties (the photon absorption and further reactivity of the explosive particles), the explosive particles undergo thermal expansion, phase transformation and partial decomposition. Such a mechanism implies that the photo-excitation and energy localization due to radiation of the shock front play a crucial role in starting decomposition process.   

Time-Resolved Temperature Measurements of Shock Initiation in Heterogeneous Exothermic Mixtures

Because the onset of reaction in shock-initiated exothermic powder compositions is difficult to observe, few dynamic measurements that could provide information about the initiation delay or the reaction mechanism have been reported. A method has been developed to experimentally measure the delay between the time of shock arrival and the time when most of the reactions have taken place using embedded thermocouples. The powder mixtures used in the tests were Ni-Al, Mn-S, Ti-Si, Ti-C and Ti-B. The test samples were placed in planar recovery ampoules containing thermocouples and a strong shock was delivered via the detonation of an explosive charge. A sharp temperature rise was measured, providing a reliable measurement of the time at which an exothermic reaction had occurred in the bulk mixture. The delay time before the temperature rise provided an upper bound of the initiation delay time, as well as information regarding the reaction mechanism. The results for all mixtures tested showed that bulk temperature starts to rise 10's of milliseconds after the mixture was shocked, which indicates that most of the reaction did not take place on the microsecond timescale.   

In-situ measurement of shock-induced reactive flow in a series of related hydrocarbons

Understanding of the chemistry that occurs under extreme, high-pressure, high-temperature shock environments poses both a significant scientific challenge, due to the difficulty of direct experimental observations, and an opportunity for discovery of new materials and bonding constructs. The combined high pressure, high temperature conditions induced by shock loading results in prompt reactions that may include dynamic bond breaking, dimerization and polymerization, and dissociation to small molecules. Detonating high explosives represent a case where the exothermicity of the chemical reactions drives a steady reactive flow field. Understanding of the evolution of different reaction pathways as a function of shock input remains a significant challenge, due to both the very short shock timescales, and difficulty in measurement of reaction intermediates and products. We have used \textit{in-situ} electromagnetic gauges to measure mechanical variables (such as multiple shock waveforms) that result from the chemistry occurring in the shock. This allows us to gain some understanding of the nature of the input conditions necessary to start the reaction. Among the materials studied are benzene, toluene, phenyl acetylene, benzonitrile. This work has led to a systematic study of shock-induced chemistry as a function of chemical structure.   

The Effect of Charge Reactive Metal Cases on Air Blast

Experiments were conducted in a 23~m$^{3}$ closed chamber using explosive encased in a cylindrical reactive metal case to study the effect on air blast from the case fragments. Parameters varied included explosive material, case material, case/explosive mass ratio and charge internal diameter, which ranged from 7.62 to 12.7~cm. The pressure histories measured on the chamber wall showed a double-shock front structure with an accelerating precursor shock followed by the primary shock, suggesting the early-time reaction of small case fragments. During the early reflections on the chamber wall, the pressure rise achieves a factor of 1.6 versus the steel-cased and a factor of 1.2-1.4 versus the bare charges, indicating combustion of a large amount of small case particles generated by secondary fragmentation. The analysis of explosion pressures and recovered fragments and solid products showed that the burnt case mass increases with detonation pressure and case/explosion mass ratio over a test range from 0.29 to 1.75 in a quadratic function. The influences of charge diameter and various reactive metal cases on the burnt case mass are further investigated.   

Effect of Particle Morphology on Critical Conditions for Shock-Initiated Reactions in Titanium-Silicon Powder Mixtures

The effect of titanium particle morphology on the shock sensitivity of titanium-silicon powder mixtures has been investigated experimentally. The powder mixtures were tested in a planar recovery capsule, with the shock loading produced by a high explosive Tetryl booster charge placed on top of the capsule and a PMMA attenuator. Reactions were not observed for stoichiometric mixtures of large (75 -- 106 $\mu $m), spherical Ti particles with fine ($<$ 44 $\mu $m) Si particles for incident peak shock pressures of up to 23 GPa, estimated with LS-DYNA. In contrast, mixtures with fine ($<$ 45 $\mu $m) spherical Ti particles or irregularly-shaped fine ($<$ 20 $\mu $m) Ti particles had critical shock pressures for reaction initiation of 7$\pm $3 GPa and 5$\pm $2 GPa, respectively. Microscopy and spectroscopy were used to identify the degree of intermixing between the particles for shock loading just below the reaction threshold. For the largest spherical Ti particles, little particle intermixing was evident. However, differential thermal analysis carried out demonstrated that even for the large Ti particles, shock loading of the samples generated microstructural effects which lowered the temperature for the onset of exothermic reaction of the shocked sample by about 80$^{\circ}$C.   

Experimental study on shock-induced doping of titania photocatalysts

Titania is a most effective photo-functional material and is widely used. But since the band gap of titania is large (Eg=3.2 eV), it is only active in the ultraviolet region, which accouts only 3{\%}-5{\%} of the overall solar intensity. Therefore, it is very important to enhance the visible light activity of the titania photocatalyst. In this study, the nitrogen-doping of titania photocatalysts were induced by shock waves, which were generated through detonation-driven flyer impact. The samples were shocked at different flyer impact velocities and recovered successfully. Two nitrogen resources containing hexamethylene tetramine(HMT) and dicyandiamide were considered. The phase composition, light absorption spectra and N doping status of the recovered samples under different shock conditions were characterized. The absorption edge of the N-doped titania photocatalysts by shock wave was extended to 450nm corresponding to visible light region. The photocatalytic degradation to rhodamine B of the samples doped with dicyandiamide increased with the increase of the flyer velocity due to the higher N doping concentration and wider response to visible light.