Session A03: Materials in Extremes: Energetic Materials: Initiation of Detonation, Hotspots and Sensitivity

Time resolving the loss of crystallinity during detonation in a secondary solid explosive

There are still significant uncertainties in our ability to predict and control detonation in secondary solid explosives which has serious implications for the safety and performance of explosives. One reason is for this uncertainty is that while chemical kinetics are well understood in gases and liquids, much less is known about how chemistry proceeds within a crystalline lattice. Secondly, events like detonation, where a bulk material can go from ambient conditions to pressures of Gigapascals (GPa) and temperatures of about 4000 kelvin (K) within nanoseconds (ns), are extremely difficult to directly observe. To better understand the role of the loss of crystallinity and how this affects temperature and chemical kinetics during a detonation, we developed a technique using visible laser scattering to probe morphology changes on a nanosecond time scale before and during a detonation. We will present our results applying this to several common secondary solid explosives, PETN, HMX and TATB, and considering steady detonation, initiation of detonation, and failure scenarios. These measurements reveal when during a detonation wave, and how fast that the initial crystals change into a less scattering dense product fluid giving new insight into the microscopic mechanism of a detonation in solid explosives.
  

A Hotspot’s Better Half: A Characterization of the Local Potential Energy Rise in Mechanically Induced Hotspots

Shock loading of high explosives leads to energy localization into hotspots, which are thought to govern the initiation of detonation. Hotspots are typically characterized in terms of their size and temperature. Criticality of a hotspot depends on a competition between thermal diffusivity and endothermic reactions that tend to quench the hotspot against exothermic reactions that can transform the hotspot into a deflagration wave. However, this view ignores the role of potential energy (PE) as a descriptor of energy localization and criticality. We show through large-scale molecular dynamics simulations of TATB pore collapse that more energy is localized in PE than in kinetic energy (KE). Furthermore, the spatial extent and diffusivity of the PE and KE hotspots are significantly different, and far from expectations based on equipartition. An analysis of the MD trajectories reveals the molecular origin of this puzzling observation. Prepared by LLNL under Contract DE-AC52-07NA27344. Approved for unlimited release, LLNL-ABS-794457.
  

Void collapse in shocked β-HMX single crystals across scales

Heat generation in the vicinity of a void during shock compression plays a critical role on the initiation of detonation in high explosives (HE). Atomistic simulations of under shock compression have shown that the void collapse regime transitions from visco-plastic to hydrodynamic jetting as the shock strength increases in many energetic materials. However, atomistic simulations are limited to nanometer size voids. On the other hand, void collapse experiments have been performed in micron size samples. Here, we present a mesoscale model informed from atomistic simulations to study the anisotropic response of shocked β-HMX single crystals that bridges nanometer to micrometer scales. The shock response of an β-HMX single crystal containing a void is studied with finite element simulations that include plasticity and heat transport. The effect of crystal orientation over a range of impact velocities are discussed. The continuum model is calibrated with non-reactive molecular dynamics simulations of planar socks. The simulations are compared with both atomistic simulations and gas gun experimental results of β-HMX containing a single void.   

Mechanisms and Size Effects of Hotspot Formation due to Shock-Induced Collapse of Pores and Cracks

The shock to detonation transition in heterogeneous high energy density materials starts with the spatial localization of mechanical energy into hotspots due to the interaction of the mechanical wave with microstructural features and defects. We present large-scale molecular dynamics simulations of hotspot formation in HMX crystals following the collapse of pores of various shapes and sizes for impact velocities ranging from 0.5 to 2.5 km/s. Hotspots resulting from cracks elongated along the shock direction show significantly higher sensitivity to both shock strength and defect size. Elongated cracks 80 nm in length result in temperatures almost three times higher that voids 80 nm diameter and reach values corresponding to the ideal case of isentropic recompression of a gas. The MD trajectories reveal the atomic origin of this contrasting behavior. While circular voids undergo a transition from viscoelastic pore collapse to a hydrodynamic regime with increasing shock strength, shock focusing in elongated cracks results in jetting and vaporization, which upon recompression leads to increased heating.   

Time-resolved x-ray imaging of void collapse at micron length scales

Pulsed x-ray imaging can provide substantial insight into a wide range of initiation-related phenomena, particularly the in situ imaging of dynamically compressed voids, which are thought to play a fundamental role in explosive initiation. Current models of the dynamic compression behavior of inhomogeneous materials are empirically calibrated to bulk, aggregate experimental data. The development of more fundamental models depends on detailed measurements and corresponding simulations which resolve single void collapse events. Further, since material strength depends on scale, experiments at the scale of actual voids (10 μm) are preferred. Since the field of view for these experiments is relatively small (~100s μm) and void collapse occurs at low pressure, these experiments can be performed with a small scale (100 mJ) laser in a portable experimental setup. Here we present the results of x-ray imaging experiments at the Advanced Photon Source on voids embedded in TNT and silicon using both explosive and laser-driven shocks, which approach the spatial scale of void collapse in actual explosives, 1-10 μm.   

Evaluation of the roles of crystal plasticity and hydrodynamic jetting in hot-spot formation in heterogeneous energetic materials under shocks

Heterogeneous energetic materials are crucial components in munitions and energy-delivery systems. Under shocks, localized hot-spots are formed near voids and microstructural defects in these materials. Hydrodynamic jetting due to void collapse and strain localization due to plastic work are two important mechanisms for hot-spot formation in the material. This work evaluates the relative contribution of these two mechanisms in the formation of hot-spots under shocks in energetic materials. To this end, an Eulerian computational framework is used to study the response of HMXs comprising a single void under shock loading. To study the role of hydrodynamic jetting, the voids are resolved sharply using level-sets and the hot-spot temperatures are tracked at different stages of void collapse. To model heating due to plastic work, an anisotropic crystal plasticity model based on a power-law slip-rate with hardening and thermal softening is used; the localized temperatures in the slip/shear bands formed in the material under shocks are also tracked. Computations are performed for different crystal orientations and shock strength. The presentation will discuss the combined roles of jetting and plastic heating in hot-spot formation in HMXs.   

Measurements of State Variables During Exploding Bridgewire Detonator Function

We have studied the behavior of exploding bridgewire detonators using a variety of observables for measuring temperature and density during detonator function. Continuous density movies have been collected using either proton or x-ray radiography and continuous light emission videos simultaneously measured using ultra-high speed video cameras. Spectrally resolved and single broadband pyrometric measures of temperature are also measured. We attempt to provide a description of the mechanism of function for EBW detonators based on the full suite of observations made and comparisons to studies in the literature. In this talk, we will show the results of measurements on PETN and HMX based detonators run in a variety of conditions.
  

Internal shock structure and thermal response in the function of Exploding Bridgewire Detonators

We have recently shown that the input shock, subsequent initiation and detonation propagation in an Exploding Bridgewire (EBW) detonator exhibits complex internal structure and temporal behavior. Using flash radiography we have observed the prompt emanation of a relatively weak shock wave (3000 m/s) from the region of the bridgewire at the time of vaporization. Abel inversion of the images reveals a highly symmetric, hemishperical structure in the density that propagates through the initial pressing. Visible imaging of the cylindrical surface of the detonator reveals a luminous wavelike structure that appears radially symmetric but lacking the exact symmetry of the density feature. In this talk these structures will be directly compared spatially and temporally for a number of EBW detonators of different sizes and materials. The relationship between shock and detonation like features will be discussed in the context of the mechanism of function of EBW detonators and recent progress in modeling the chemistry of energy release in these applications.
  

Comparing small-scale detonation simulation to experimental data and multi-dimensional initiation sensitivity study

We are interested in modeling exploding foil initiator (EFI) detonators to help with design timelines and understanding the fundamental science of detonators. We will discuss efforts to model small-scale detonations in the absence of calibration data that take into account the ignition and growth in PETN sub-centimeter pellets. Several reaction rate models along with parameter studies were performed to show variation in the breakout time and wave shape at the end of the pellet using the hydrocode FLAG. Since 1D initiation does not wholly describe the initiation in these small-scale detonations of interest, we will explore the effect of multi-dimensional ignition on the outcome. We will show several comparisons of simulation data to experimental data and discuss the results.
  

(U) Introducing HEDONIST- A Low Explosive Mass Experiment That Attains Very High Pressures

The proton radiography facility at LANL offers unique experimental capabilities,
particularly the planned Pu@pRad line. Pu@pRad limitations however preclude the usage
of large quantities of high explosives. We present the hydrocode analysis of a small-scale
system (under 30g TNT equivalent) capable of reaching pressures in excess of 2 Mbar.
This system utilizes a novel multipoint initiation system to establish converging
detonation waves which deliver a strong shock wave onto a sample cell. HEDONIST has
been designed to present a low areal density to the pRad beam thereby providing ample
signal-to-noise ratio to discern shock waves in sample cell materials from other shocks
present. Results from the first two pRad experiments and the comparison to hydrocode
simulations will be discussed.   

Modeling and Experimental Analysis of Shaped Charge Jet Characteristics

The U.S. Army CCDC Armaments Center at Picatinny Arsenal, NJ uses a variety of computational and analytical tools to predict and validate shaped charge warhead performance. In this work, an 81mm copper lined conical shaped charge loaded with the HMX-based explosive LX-14 is modeled in the hydrocode ALE3D and tested using flash radiography. Jet characterization x-ray films are digitized and analyzed using a series of computer programs developed by Miers and Pham. Experimentally determined jet parameters such as velocity profile and accumulated length/mass/energy are compared with hydrocode predictions. Additionally, the Walsh theory for plastic instability and particulation in stretching metal jets is utilized to predict jet breakup time from hydrocode models, and results are compared with experiment. The resulting effects on armor penetration performance predictions are discussed.