Bulletin of the American Physical Society
68th Annual Meeting of the APS Division of Fluid Dynamics
Volume 60, Number 21
Sunday–Tuesday, November 22–24, 2015; Boston, Massachusetts
Session R2: Detonation and Explosion |
Hide Abstracts |
Chair: Ju Zhang, Florida Institute of Technology Room: 101 |
Tuesday, November 24, 2015 12:50PM - 1:03PM |
R2.00001: Detonation Initiation with Thermal Deposition due to Pore Collapse in Energetic Materials - Towards the Coupling between Micro- and Macroscale Ju Zhang, Thomas Jackson Initiation of detonation through thermal power deposition due to pore collapse in energetic materials (such as HMX) is studied numerically by solving the reactive Euler equations. The thermal power deposition model is partially based on previous results of direct simulations of pore collapse. The thermal deposition time scales obtained from the pore collapse model are significantly longer than acoustic time scale. It is found here that a critical size of hot spots exists, and when hot spots exceed the critical size, direct initiation of detonation upon ignition seems independent of power input, and is achieved even with low power input. On the other hand, when hot spots are below the critical size, the ignition does not lead to detonation. However, if the thermal deposition time scale is decreased, a scenario different than pore collapse, such that it is on the acoustic time scale, detonation does arise, a scenario corresponding to the so-called ``explosion in explosion''. A time scale criterion for direct initiation of detonation is then proposed and demonstrated with numerical simulations. It is proposed that if the chemical reaction time scale is shorter than the acoustic time scale at ignition, the ignition will lead to a direct initiation of detonation. [Preview Abstract] |
Tuesday, November 24, 2015 1:03PM - 1:16PM |
R2.00002: Experimental investigation of turbulent mixing in post-explosion environment Josh Smith, Michael Hargather Experiments are performed to investigate the turbulent mixing of product gases and the ambient environment in a post-explosion environment. The experiments are performed in a specially constructed shock tunnel where thermite-enhanced explosions are set off. The explosives are detonated at one end of the tunnel, producing a one-dimensional shock wave and product gas expansion which moves toward the open end of the tunnel. Optical diagnostics are applied to study the shock wave motion and the turbulent mixing of the gases after the detonation. Results are presented for schlieren, shadowgraph, and interferometry imaging of the expanding gases with simultaneous pressure measurements. An imaging spectrometer is used to identify the motion of product gas species. Results show varying shock speed with thermite mass and the identification of turbulent mixing regions. [Preview Abstract] |
Tuesday, November 24, 2015 1:16PM - 1:29PM |
R2.00003: Exhaust Gas Emissions from a Rotating Detonation-wave Engine Kazhikathra Kailasanath, Douglas Schwer Rotating detonation-wave engines (RDE) are a form of continuous detonation-wave engines. They potentially provide further gains in performance than an intermittent or pulsed detonation--wave engine (PDE). The overall flow field in an idealized RDE, primarily consisting of two concentric cylinders, has been discussed in previous meetings. Because of the high pressures involved and the lack of adequate reaction mechanisms for this regime, previous simulations have typically used simplified chemistry models. However, understanding the exhaust species concentrations in propulsion devices is important for both performance considerations as well as estimating pollutant emissions. Progress towards addressing this need will be discussed in this talk. In this approach, an induction parameter model is used for simulating the detonation but a more detailed finite-chemistry model including NOx chemistry is used in the expansion flow region, where the pressures are lower and the uncertainties in the chemistry model are greatly reduced. Results show that overall radical concentrations in the exhaust flow are substantially lower than from earlier predictions with simplified models. The performance of a baseline hydrogen/air RDE increased from 4940 s to 5000 s with the expansion flow chemistry, due to recombination of radicals and more production of H2O, resulting in additional heat release. [Preview Abstract] |
Tuesday, November 24, 2015 1:29PM - 1:42PM |
R2.00004: Investigation of detonation propagation through an array of random discrete energy sources using the reactive Burgers' analog Giuseppe Di Labbio, Charles Basenga Kiyanda, XiaoCheng Mi, Andrew Jason Higgins, Nikolaos Nikiforakis, Hoi Dick Ng For a homogeneous reactive medium such as a combustible gaseous mixture, the detonation wave is nearly always observed to propagate at a velocity predicted by the Chapman-Jouguet (CJ) condition. Although the CJ condition was originally formulated for a wave propagating in homogeneous media at constant velocity, it has been posited that this condition may also determine the average detonation velocity in heterogeneous media. This work aims to test the applicability of the CJ condition to heterogeneous media on the one-dimensional reactive Burgers’ equation, a tractable analog to the reactive Euler equations, with the reaction governed by an Arrhenius rate law. In this study, heterogeneity is modeled using discrete energy sources, of random energy content, randomly distributed throughout space such that the total energy release is equivalent to that of a homogeneous medium with constant energy density. The equations are solved using a second-order finite volume approach with an exact Riemann solver. The evolution of the discrete detonation is tracked over a long duration and its average propagation velocity is computed. In all cases, the average detonation velocity was found to be in agreement with the velocity predicted by the CJ condition for the equivalent homogeneous system. [Preview Abstract] |
Tuesday, November 24, 2015 1:42PM - 1:55PM |
R2.00005: Numerical investigation of the density effect in modeling detonation propagation in high explosives Carlos Chiquete, Chad D. Meyer, Mark Short Detonation Shock Dynamics (DSD) is an asymptotically-derived detonation propagation model used in engineering models of high explosive (HE) performance. The method is based on the limit where the detonation reaction zone length and time scales are small in relation to the much larger geometry in which the HE is embedded. The intrinsic DSD propagation law (functionally relating the surface normal velocity and curvature) for each HE is typically calibrated to simplified geometry tests where steady-state front velocities and shapes are measured. This relationship is necessarily a function of the experimental conditions and is thus limited in scope. For HE’s with variable pressing or casting density, a particular need exists for calibrations sensitive to this variability. However, there is little constraint on how the density effect is specifically incorporated into the fitting procedure. To investigate this issue, shock-attached calculations in simple slab or cylindrical geometries are performed for varying initial density for a “numerical” explosive model with a realistic equation of state. The steady-state detonation velocities, front shapes and the resulting DSD calibration of this generated data are analyzed as function of the applied HE density. [Preview Abstract] |
Tuesday, November 24, 2015 1:55PM - 2:08PM |
R2.00006: Dynamics of galloping detonations: inert hydrodynamics with pulsed energy release Matei I. Radulescu, Joseph E. Shepherd Previous models for galloping and cellular detonations of Ulyanitski, Vasil'ev and Higgins assume that the unit shock decay or cell can be modeled by Taylor-Sedov blast waves. We revisit this concept for galloping detonations, which we model as purely inert hydrodynamics with periodically pulsed energy deposition. At periodic time intervals, the chemical energy of the non-reacted gas accumulating between the lead shock and the contact surface separating reacted and non reacted gas is released nearly instantaneously. In between these pulses, the gas evolves as an inert medium. The resulting response of the gas to the periodic forcing is a sudden gain in pressure followed by mechanical relaxation accompanied by strong shock waves driven both forward and backwards. It is shown that the decay of the lead shock in-between pulses follows an exponential decay, whose time constant is controlled by the frequency of the energy deposition. More-over, the average speed of the lead shock is found to agree within 2 percent to the ideal Chapman-Jouguet value, while the large scale dynamics of the wave follows closely the ideal wave form of a CJ wave trailed by a Taylor expansion. When friction and heat losses are accounted for, velocity deficits are predicted, consistent with experiment. [Preview Abstract] |
Tuesday, November 24, 2015 2:08PM - 2:21PM |
R2.00007: Detonation Propagation through Nitromethane Embedded Metal Foam Brandon Lieberthal, Warren R. Maines, D. Scott Stewart There is considerable interest in developing a better understanding of dynamic behaviors of multicomponent systems. We report results of Eulerian hydrodynamic simulations of shock waves propagating through metal foam at approximately 20\% relative density and various porosities using a reactive flow model in the ALE3D software package. We investigate the applied pressure and energy of the shock wave and its effects on the fluid and the inert material interface. By varying pore sizes, as well as metal impedance, we predict the overall effects of heterogeneous material systems at the mesoscale. In addition, we observe a radially expanding blast front in these heterogeneous models and apply the theory of Detonation Shock Dynamics to the convergence behavior of the lead shock. [Preview Abstract] |
Tuesday, November 24, 2015 2:21PM - 2:34PM |
R2.00008: Theory of weakly nonlinear multidimensional detonations Luiz Faria, Aslan Kasimov, Rodolfo Rosales We derive an asymptotic model for the dynamics of weakly nonlinear multi-dimensional detonations from the compressible reactive Navier-Stokes equations. It is assumed that activation energy is large, heat release is small, evolution is slow, and $\gamma-1$ is small. The resultant model in 2D in dimensionless form is given by \begin{eqnarray*} u_{t}+uu_{x}+v_{y} & = &-\frac{1}{2}T_{x}+\mu u_{xx}\\ v_{x} & = &u_{y}\\ \lambda_{x} & = &-k(1-\lambda) e^{\theta T}-d\lambda_{xx}\\ \kappa T_{x}+T & = &u+q\lambda+qd\lambda_{x}. \end{eqnarray*} where $u, v$ is the velocity field, $T$ is the temperature, and $\lambda \in[0,1]$ is the reaction progress variable, $q$ heat release, and $\mu$, $\kappa$, $d$ are coefficients of viscosity, heat conduction, and diffusion, respectively. This system is a generalization of the models of small disturbance unsteady transonic flow, weakly nonlinear acoustics (Zabolotskaya-Khokhlov (ZK) equation), and water waves (dispersionless Kadomtsev-Petviashvili (KP) equation). The model predicts regular and irregular multi-dimensional patterns, and in 1D exhibits transition from steady and stable traveling waves to oscillatory traveling waves through a Hopf bifurcation as $\theta$ is increased. Period-doubling bifurcations leading to chaos are also observed. [Preview Abstract] |
Tuesday, November 24, 2015 2:34PM - 2:47PM |
R2.00009: On the quasi-one dimensional structure of the cellular detonation in a two-dimensional duct C.M. Uyeda, M. Kurosaka, A. Ferrante We performed numerical simulations of cellular detonations in a 2D duct to establish the validity of the one-dimensional ZND model. The detonation was simulated by solving the Euler equations with a WENO-TCD numerical method using adaptive mesh refinement and a detailed chemical reaction mechanism. The results show that the properties of the ZND model of a 2H$_2$-O$_2$-7Ar reaction are very close to the results of the simulations initiated using three different methods for the area-averaged properties and the properties of particles tracked along their pathlines. Disagreements between the particle properties and the ZND model are greatest near the detonation front where the transverse wave and Mach stem introduce larger jumps in the flow properties than the ZND model predicts. The particle pathlines also exhibit a quasi one-dimensional motion downstream from the detonation front which is supported by the quick decay in the particles' velocity ratio of the vertical to horizontal velocity components, in the reference frame attached to the detonation front. These findings show the quasi one-dimensional nature of 2D detonations and the applicability of the ZND model. [Preview Abstract] |
Tuesday, November 24, 2015 2:47PM - 3:00PM |
R2.00010: Flame Acceleration and Transition to Detonation in Channels Gabriel Goodwin, Ryan Houim, Elaine Oran Two-dimensional numerical simulations of a confined, homogeneous, chemically reactive gas were used to compute and catalog interactions leading to deflagration-to-detonation transition (DDT). The geometrical configuration was a long rectangular channel with regularly spaced obstacles and adiabatic boundary conditions on all of the surfaces. The channel contained a stoichiometric mixture of ethylene-oxygen at 300 K and one atm that was ignited with a circular flame. The reactive Navier-Stokes equations were solved on an adapting grid by a high-order Godunov algorithm. The channel height was fixed at 0.32 cm and obstacle heights created blockage ratios ranging from 0.8 to 0.05, where the blockage ratio is defined as the obstacle height divided by the channel height. The computations show the development of a turbulent flame, the creation of shocks, shock-flame interactions, and a host of fluid and chemical-fluid instabilities. The result is an accelerating flame and eventual DDT in unburned, but shock-heated, material. Several DDT mechanisms were observed; these will be shown and discussed, with an emphasis on several new observations related to shock interactions. [Preview Abstract] |
Tuesday, November 24, 2015 3:00PM - 3:13PM |
R2.00011: Liquid explosions induced by X-ray laser pulses Claudiu Stan, Hartawan Laksmono, Raymond Sierra, Trevor McQueen, Despina Milathianaki, Jason Koglin, Thomas Lane, Marc Messerschmidt, Garth Williams, Matt Hayes, Serge Guillet, Sabine Botha, Karol Nass, Ilme Schlichting, Robert Shoeman, Howard Stone, Sébastien Boutet Sudden generation and release of enough energy to vaporize matter are encountered in systems that range from supernovae explosions and asteroid impacts to applications in fusion energy generation, materials processing, and laser surgery. Understanding these strong explosions is important to both fundamental science and technical applications. We studied a new type of microexplosion, induced by absorption of X-ray pulses from a free-electron laser in micron-sized drops and jets of water. These explosions are related to, but different from, those observed in experiments performed with optical lasers. Unlike explosions caused by optical lasers, X-ray laser explosions produce symmetric expansion patterns that are simpler to rationalize. The release of energy initially concentrated in a small region inside drops and jets leads to ballistic vapor flow and inertial liquid flow. The kinematics of these flows indicates that the conversion of the energy deposited by X-rays into flow has a scaling that is similar to the one encountered in shock waves. [Preview Abstract] |
Follow Us |
Engage
Become an APS Member |
My APS
Renew Membership |
Information for |
About APSThe American Physical Society (APS) is a non-profit membership organization working to advance the knowledge of physics. |
© 2024 American Physical Society
| All rights reserved | Terms of Use
| Contact Us
Headquarters
1 Physics Ellipse, College Park, MD 20740-3844
(301) 209-3200
Editorial Office
100 Motor Pkwy, Suite 110, Hauppauge, NY 11788
(631) 591-4000
Office of Public Affairs
529 14th St NW, Suite 1050, Washington, D.C. 20045-2001
(202) 662-8700