Cylinder Expansion Test Research Articles

Numerical Simulations of 122 mm M-21OF Missile Fragments Propulsion

This paper presents numerical simulations of 122 mm M-21OF rocket fragments propulsion. A simplified geometry of the rocket was constructed and analysed with two main numerical FE approaches: a mesh-based lagrangian-eulerian FSI method and a coupled meshless SPH–lagrangian approach. The resulting fragment velocities were compared with those obtained from an analytical approach (Gurney formula). The shell fragmentation process was also successfully depicted. Both numerical approaches differed from the analytical velocity results to a comparable degree. Considering that the Gurney energy of a high-explosive is obtained from cylinder expansion tests, results obtained with the Gurney method can be treated as a reference point and validation method for natural fragmentation warhead models.

Validation of a detonation product equation of state for an insensitive high explosive via slab geometry expansion tests

Slab expansion (SLABEX) tests are conducted to validate a process for calibrating the detonation products equation of state (EOS) of a high explosive (HE). The SLABEX tests use rectangular slabs of PBX 9502, a polymer-bonded HE formulation consisting of 95 wt.% 1,3,5-triamino-2,4,6-trinitrobenzene bound with Kel F-800, a co-polymer of chlorotrifluoroethylene and vinylidene-fluoride. Three PBX 9502 slab thicknesses are examined, each confined symmetrically by two rectangular copper (Cu) plates approximately one-tenth the thickness of the HE slab. For the duration of each experiment, the detonation flow along the central axis of the PBX 9502 slab remains two-dimensional. The lateral flow velocity component of the outer surfaces of the expanding Cu plates is measured, along with the steady axial detonation speed along the central axis of the SLABEX. Hydrodynamic simulations of the Cu plate expansion in the SLABEX geometry, driven by the energy stored in the detonation products by the detonation combustion event, are conducted using a Jones–Wilkins–Lee EOS for the detonation products. This EOS form was recently parameterized for PBX 9502 in the cylinder expansion test geometry using a newly developed calibration technique [Voelkel et al., Combust. Flame 246, 112373 (2022)]. Good agreement between the experiment and prediction is found in each SLABEX test, demonstrating that the detonation product EOS calibration technique produces EOSs that are predictive when applied to other geometries.

GammaV EOS for Detonation Products

The standard way to start the derivation of an equation of state for detonation products is by defining a trial function Ps(V) along the principal isentrope of the products (where Ps is pressure and V is specific volume), with parameters to be determined from a cylinder expansion test. If that specific function turns out not to be good enough, a different function needs to be tried. We propose here a different approach which is to use instead the function gs(V), where gs is the adiabatic gamma along the principal isentrope, as defined by Eq. (2) below. By choosing gs we don’t have to assume the form of the function from the outset. Instead, we define the function by a set of points (n≥4), and we’re able to define the amplitudes of these points recursively. In what follows we first write down the equations for determining the values of gsi for the specific volumes Vi, and then work out an example with n=4. We show that even for n=4 the reproduction of test data is quite good, and also that the results are not sensitive to the exact choice of the values of Vi.

Effect of lot microstructure variations on detonation performance of the triaminotrinitrobenzene (TATB)-Based insensitive high explosive PBX 9502

PBX 9502 is an important insensitive high explosive due to its combination of safety properties and detonation performance. It is a polymer-bonded formulation consisting of 95 wt.% 1,3,5-triamino-2,4,6-trinitrobenzene (TATB) as the high explosive crystal, bound with Kel F-800 (FK-800), a co-polymer of chlorotrifluoroethylene and vinylidene-fluoride. Two different types are used, one known as virgin that uses only pristine manufactured TATB, and a second known as recycled that has 50 wt.% of its TATB reclaimed from machining scraps of previously pressed virgin PBX 9502. Recycled lots have a higher percentage of fine particles compared to virgin lots, due to the fracturing and damage sustained by TATB crystals during pressing. We examine the influence that this TATB microstructure difference between virgin and recycled PBX 9502 has on detonation performance properties. New rate-stick geometry diameter effect, detonation front shape and cylinder expansion test data are obtained for two previously uncharacterized virgin lots and one recycled lot of PBX 9502. This is combined with previously published data for one virgin and one recycled lot for evaluation purposes. Detonation shock dynamics model calibrations are conducted on each of the five lots to provide an assessment of the detonation timing characteristics of virgin versus recycled PBX 9502 lots. For two of the virgin lots, detonations propagate slower in the rate-stick geometry than those in the two recycled lots. However, the other virgin lot tested has propagation rates comparable to that of the recycled lots for larger diameter rate-sticks. The latter result, though, is shown to be geometry dependent and depends on the range of detonation curvatures accessed in different geometries. New copper-confined, cylinder expansion tests are conducted on each of the three virgin and two recycled lots to obtain detonation product Jones-Wilkins-Lee equations of state, enabling an assessment of the metal push capabilities for each of the five lots. We find that the metal push capabilities, characterized by the evolution of the heat of detonation with volume, are similar between the virgin and recycled lots. Thus, changes in microstructure between different PBX 9502 lots seemingly affect the rate of reaction, but not the overall energy content.

Extended JCZ3 Equation of State and JAGUAR Implementation for High Explosives

AbstractAn extended JCZ3 based equation of state for the modelling of high explosive detonation products has been implemented in the JAGUAR computer model. This development includes the parameterization of EXP‐6 potentials for individual products species, as well as mixing rules for the combination of species. Non‐linear variable metric optimization was used for the species parameterization, as well as for the solution of free energy minimized states. Explosive detonation and products expansion are typically modelled for JWL and JWLB high explosives thermodynamic equation of state development. An analytic c ylinder test model was developed for the comparison to experimental cylinder expansion test data and for the characterization of explosive products expansion and work output. Examples and comparisons to data are presented. Historically, streak photography was used for cylinder testing, whereas photon‐doppler velocimetry (PDV) is now standard. This has resulted in some changes in cylinder testing data reduction. A JAGUAR MSIAC web‐based application is currently under development for use by the MSIAC member nations.

Detonation Reaction Zone and Acceleration Ability of a TKX‐50 Based Polymer Bonded Explosive

AbstractPrevious studies on detonation of the Dihydroxylammonium 5,5′‐bistetrazole‐1,1′‐diolate (TKX‐50), an attractive energetic ionic salt (EIS), were mainly based on thermodynamic codes or small‐scale tests, the available experimental data are not adequate to fully understand detonation characteristics of this new kind of energetic material. This article gives an experimental investigation on detonation reaction zone (DRZ) and acceleration ability of a TKX‐50 based Polymer Bonded eXplosive (PBX), PBX‐TKX‐50. The detonation properties of two traditional explosives: (i)PBX‐HMX, an HMX‐based PBX, (ii) PBX‐RDX, an RDX‐based PBX, both of which contain the energetic compounds and binder in the same mass ratio as PBX‐TKX‐50, are also measured for a comparative analysis. The results show that PBX‐TKX‐50 has the highest detonation velocity and detonation pressure, nevertheless the slowest reaction rate within detonation reaction zone, the longest reaction time and DRZ length among the three PBX samples. Despite its prominent detonation velocity and detonation pressure, the acceleration ability of PBX‐TKX‐50 is lower than PBX‐RDX. The Jones‐Wilkins‐Lee (JWL) equation of state for detonation products calibrated by the cylinder expansion test has severely overestimated the axial acceleration ability of PBX‐TKX‐50 in the numerical simulation of Disc Acceleration eXperiment (DAX), which may due to the fact that remarkable confinement effect and/or geometry effect exist in detonation of the EIS based sample. The mechanisms behind these detonation behaviors are still under investigation in our ongoing studies.

Energy Performance of HMX‐Based Aluminized Explosives Containing Polytetrafluoroethylene (PTFE)

AbstractInvestigation of the energy performance of HMX‐based aluminized explosives added with polytetrafluoroethylene (PTFE) oxidizer was undertaken and compared with the PTFE‐free aluminized explosive of similar formulation. Firstly, the heat of explosion was measured in a calorimetric bomb, next the underwater explosion was conducted, then the cylinder expansion test was performed, and finally, the detonation reaction zone parameters were determined by the interface particle velocity experiment. Compared with the PTFE‐free aluminized explosive, the addition of PTFE into aluminized explosives increases the heat of explosion significantly from the calorimetric data. The PTFE‐containing aluminized explosive also yields the longer first bubble oscillation time in the underwater explosion than the PTFE‐free and ammonium perchlorate‐containing counterparts. From the cylinder test data, the wall velocity and Gurney energy were determined. The aluminized HMX containing PTFE shows an inferior acceleration ability to PTFE‐free aluminized HMX, but exhibits a stronger afterburning potential. The addition of PTFE to aluminized HMX decreases detonation velocity and CJ pressure and increases the detonation reaction zone time and length. It has reasons to believe that the PTFE partially decomposes in detonation reaction zone. Conclusions on the potential use and suggestion for future research of PTFE as an oxidizer in aluminized explosives are drawn.

Properties and detonation performance of tetraamminecopper(II) nitrate (TACN) – a prospective green explosive

ABSTRACT The simplest ligand that can be present in energetic complex compounds is ammonia. It has a lone pair of electrons on the nitrogen atom that can form a coordinate bond with the metal cation. An example of such a compound is tetraamminecopper(II) nitrate (TACN). TACN is considered in the literature as an energy material with potential use in ammunition and civil applications. In the work, the physicochemical properties and detonation performance of TACN were investigated. The compound was identified using differential thermal analysis coupled with thermogravimetry (DTA/TG) and infrared spectroscopy (IR). The sensitivity of TACN to mechanical stimuli (friction and impact) was determined. The thermal stability was tested. The critical diameter of the charge with bulk density and the ability to detonation for charges of different densities initiated by an electric detonator were determined. The velocity of detonation was also measured for various diameters and densities of charges and the method of their initiation. The cylinder expansion tests were performed for TACN charges of various densities, and the acceleration ability, detonation pressure and energy as well as the equation of state of detonation products were determined. The calorimetric heats of combustion and detonation were also measured.

Characterisation and validation of the JWL equation of state parameters for PE4

This investigation focused on determining the Jones-Wilkins-Lee (JWL) equation of state (EoS) parameters for the Australian made explosive PE4-MC. Four cylinder expansion (CYLEX) experiments were conducted using the T-20 methodology with measurements of the time of arrival of the copper cylinder on the perspex plate made using optical fibres connected to an optical pulse timer. Three different model parameter fits were then generated from the CYLEX test data. These were based on; (1) Traditional least-squares fit of CYLEX data to JWL EoS and (2) Inverse modeling method using LS-DYNA and LS-OPT (two different fits). The three parameter sets were then compared in terms of the accuracy of their predictions against two sets of blast loading experiments. The first set of blast loading experiments evaluated the different parameter fits at a scaled distance (Z) of 1.77–2.55 m/kg1/3, representing a mid-field blast loading scenario while the second set of blast loading experiments represented a near-field blast loading scenario (Z = 0.08–0.1 m/kg1/3). The JWL EoS parameters which were derived using the standard fitting method from the CYLEX test data (CYLEX) were shown to provide accurate predictions (within 5% of experiments) in the near-field but under-predicted the peak reflected pressure in the mid-field by more than 20%. The best predictions across both the near and mid-field blast loading scenarios was provided by the two parameters sets (LS-OPT Final and LS-OPT Select), which were derived using an inverse modeling approach, with the LS-OPT Select set of parameters being the preferred option. Additionally, comparisons between the experimental data and predictions using the JWL EoS parameters for the US made C4 explosive indicated that the blast loading in the mid-field was similar between C4 and PE4-MC, which was consistent with the previous literature. However, in the near-field C4 was predicted to provide higher blast loadings. This highlights the importance of validating an explosives EoS across a range of different scaled distances.

Predicting the Gurney velocity of chemical high explosives: thermochemical code calculations and studies on the detonation products isentrope

ABSTRACT The simplicity and versatility of the Gurney model has led to its widespread adoption for the calculation of peak fragment velocity of explosively-driven munitions such as fragmentation warheads and artillery projectiles. The Chemical Equilibrium with Applications (CEA) thermochemical code provides several relevant parameters that were found to be important for the construction of a mathematical model for estimating the Gurney velocity of pure and mixed chemical high explosives. Compared to other methods, the proposed model demonstrated its accuracy in predicting the Gurney velocity of sensitive and insensitive explosive compositions. Models based on detonation velocity were found to be particularly inadequate to predict the Gurney velocity of insensitive explosives, which is strong proof that the detonation velocity does not necessarily correlate with the Gurney velocity for all types of high explosives. Results from the cylinder expansion test provide some evidence that the low to moderate metal acceleration ability of insensitive explosives is due to their unique and steep expansion isentropes rather than charge diameter effects or non-ideal detonation behavior.

Detonation performance experiments and modeling for the DAAF-based high explosive PBX 9701

Abstract Detonation performance experiments and modeling are reported for the explosive PBX 9701, which is composed of 97% 3,3’-diamino-4,4’-azoxyfurazan (DAAF) and 3% vinylidene fluoride-chlorotrifluoroethylene copolymer (Kel-F 800) binder by weight. PBX 9701 is a newly developed reduced-sensitivity explosive with increased performance relative to the triaminotrinitrobenzene or TATB-based PBX 9502 while still retaining low sensitivity to mechanical insult. The first detonation performance measurements for this formulation are presented, including fron curvature rate sticks and cylinder expansion test data. Prior shock initiation data is also reviewed. These data are used to develop programmed burn (PB) and reactive burn (RB) calibrations for existing, commonly used, performance models which allow engineering calculations with PBX 9701. The calibration process involves several enhancements relative to conventional approaches, including the use of an analytical scaling correlation to speed the equation of state (EOS) calibration process. It uses a PB hydrocode-based approach and development of a new methodology to improve the consistency between the PB and RB model calibrations and associated calculations. This link is achieved by populating the RB products EOS in direct reference to the PB release isentrope and Chapman-Jouguet state and in calibrating the timing components of each model using an equivalent procedure, all in order to facilitate comparison between the two modeling approaches. Overall, PBX 9701 is seen to exhibit improved performance relative to insensitive explosives, with a trinitrotoluene (TNT) equivalence of 1.24. The detonation performance properties are found to be well captured by existing models.

Experimental measurement of energy release from an initiating layer in an insensitive explosive

When subjected to a shock of insufficient strength to trigger prompt reaction, heterogenous condensed phase explosives can form regions where significant amounts of the explosive remain unreacted for times much greater than the reaction time of the detonating explosive. This phenomena is observed for the explosive PBX 9502 (95 wt% TATB) both for planar and oblique input shocks. In this work, we build on previous results by performing cylinder expansion (CYLEX) tests where the explosive charge is comprised of a faster core of PBX 9501 (95 wt% HMX) inside a slower annulus of PBX 9502. The detonation in the faster PBX 9501 drives an oblique shock into the adjacent PBX 9502, and an annular transverse initiating layer (IL) results. In the test geometry, the IL travels steadily down the length of the test after a short run distance. At radial positions beyond the IL, an annular region of detonating PBX 9502 is observed. Using standard CYLEX test diagnostics, we infer the total energy release of this experiment. By making the assumptions that (1) the combined energy release is comprised of contributions from detonating PBX 9501, detonating PBX 9502, and the IL in the PBX 9502 and (2) mass-specific energy release for the detonating explosives is approximately the same as typically observed for each explosive, the IL energy release and reaction efficiency can be computed. Results are compared to prior results for a similar geometry, and indicate that while shock deadened PBX 9502 does not detonate promptly, it does eventually release a significant portion of its chemical potential energy over longer timescales on the order of 10 µs.

Detonation performance of the CL-20-based explosive LX-19

The generation of constitutive detonation performance model components for high explosives (HEs) invariably involves reference to experiment, as reliable first-principles determinations of these models are beyond our current capability. Whatever its form or complexity, the detonation performance model must be able to accurately capture the detonation wave timing and the energy release that it triggers upon arrival. Specifically, the HE products equation-of-state (EOS), which largely determines the detonating HE’s ability to do useful work on its surroundings, is typically inferred from cylinder expansion tests where metal-confined HE cylinders are detonated and the ensuing outer confiner wall-expansion trajectory is recorded. Expensive, iterative comparisons to multimaterial hydrodynamic (or “hydrocode”) simulations of these experiments are then used to constrain the parameters of the chosen EOS form. Here, we report on new detonation performance experiments produced for the highly-ideal, plastic-bonded explosive and CL-20-based LX-19 which are used to produce a new sub-scale detonation performance model for the explosive. This includes new products EOS and a new Detonation Shock Dynamics front propagation law. We also confirm the capability of two new, non-hydrocode-based products EOS generation techniques to accelerate the HE model parameterization process. This latter development is particularly significant for detonation performance modeling of new HE formulations.

Detonation Characteristics of an Aluminized Explosive Added with Boron and Magnesium Hydride

AbstractInvestigation of the detonation characteristics of an aluminized explosive added with boron (B) and magnesium hydride (MgH2) were undertaken and compared with the aluminized explosive of similar formulation. Firstly, the explosion heat was measured in a vacuumed calorimetric bomb, then the detonation pressure was determined by the interface velocity experiment, and finally the cylinder expansion test was performed. From the calorimetric data, the addition of B and MgH2 into an aluminized explosive increases the heat of explosion slightly. From the detonation pressure data, the addition of B and MgH2 appears to have no significant influence to the detonation pressure. From the cylinder test data, the detonation velocity, wall velocity, Gurney energy and detonation energy were determined. The aluminized explosive added with B and MgH2 shows weak acceleration ability, but exhibits a stronger afterburning potential. Finally the equation of state of the detonation products was determined for different casing conditions. Conclusions about how to use B and MgH2 in aluminized explosives to generate optimal effects are drawn.

The dependence of Ammonal detonation performance on cylinder test scale

For a detonation to propagate steadily, the lead shock must be supported by energy release from the chemical reaction zone. Flow divergence cools the flow behind the compression shock, slowing chemical reactions and reducing the available energy to drive the detonation forward. This results in the diameter effect and eventually detonation failure as charge size is reduced. Similarly, the energy available to perform work on adjacent materials in the product flow tends to decrease with charge size. The performance of non-ideal explosives such as ammonium nitrate blended with aluminum powder (Ammonal) are particularly sensitive to flow divergence. In this study, the effect of flow divergence on Ammonal performance is investigated through application of an analytic method to analyze cylinder expansion (CYLEX) test wall velocity profiles for tests with inner diameters of 12.7 mm up to 76.2 mm. For these tests, the detonation velocity, detonation product isentropes, and heat of detonation are reported. In addition, analysis of the velocity profiles is shown to reveal an experimental measurement of the Rayleigh line, which agrees well with the theoretical Rayleigh line for all experiments. Reaction zone times are inferred using this feature.

Direct measurement of energy loss due to aging effects in the condensed phase explosive PBX 9404

The explosive performance of PBX 9404, a condensed phase explosive with HMX and nitrocellulose reactive ingredients, was evaluated using the detonation cylinder expansion test after it had been naturally aged for over 46 years. Nitrocellulose is known to chemically degrade with age, but the corresponding effect on explosive performance is currently unknown. Two new cylinder tests were fielded with the oldest known PBX 9404 explosive (>46 years) and the data was compared with prior PBX 9404 cylinder test data. A method for comparing wall motion data collected by streak camera and laser interferometry diagnostics was also introduced. Analysis of the cylinder motion indicated that the aged explosive exhibits decreased performance, which varied with cylinder expansion radius and product specific volume. This energy decrement was found to be 0.8% of the total initial explosive energy per decade of age at a product volume of 7.0 cc/g. The measured energy decrement in explosives older than 37 years exceeds the chemical energy content of nitrocellulose, indicating that nitrocellulose decomposition radicals are likely degrading the normally stable HMX molecules during the aging process.

Combined Mini‐Cylex & Disk Acceleration Tests in Type K Copper

AbstractWe combined the miniature cylinder expansion test (Mini‐Cylex), with the Disk Acceleration Test (DAX) using Type K copper, Picatinny Liquid Explosive, and photonic Doppler velocimetry. We estimated the CJ state using plate reverberation methods at the test cap. We extracted velocities at 2, 7, and 10 volume expansions to fit Jones‐Wilkins‐Lee Equation of State and estimated Gurney velocity at the tube wall. The test also provides an additional method to estimate reaction products Hugoniot through knowledge of the copper test cap. Our experiments and simulations are within expected uncertainty. The test and the analysis effectively reduce costs while keeping or increasing fidelity.

Evaluation of the Detonation Performance of Insensitive Explosive Formulations Based on 3,3ʹ Diamino-4,4ʹ-Azoxyfurazan (DAAF) and 3-Nitro-1,2,4-Triazol-5-One (NTO)

ABSTRACTTwo energetic materials identified for relatively high energy, but little to no response to impact, spark or friction stimuli are 3-nitro-1,2,4-triazole-5-one (NTO), and 3,3ʹ diamino-4,4ʹ-azoxyfurazan (DAAF). More of an outlier in performance versus sensitivity, DAAF illustrates insensitivity by small-scale sensitivity tests, yet has a failure diameter estimated to be 1.25 mm and a short run length to detonation. Because of this unusual behavior, DAAF is an ideal material to formulate with NTO to obtain tailored shock sensitivity and critical diameter, with detonation velocities and pressures higher than PBX 9502. Here, we present detonation properties of Kel-F® bonded formulations with ratios of 20–70 wt.-% DAAF added to NTO. All formulations were evaluated for detonation velocity, aluminum flyer acceleration at jump-off, and via the cylinder expansion test.

Is the detonation “Dead Zone” really dead?

This paper explores the Dead Zone (DZ) phenomenon in solid High Explosives (HEs), whereby weaker regions of a detonation shock fail to trigger prompt reaction, leaving behind them isolated pockets of substantially unburned explosive. A key unanswered question is the extent to which DZs react in the following flow, and whether they contribute a significant amount of energy on timescales relevant to system deconsolidation. This paper comprises two parts. The first surveys the DZ phenomenon and discusses (1) the multiple contexts in which it arises, (2) what is observed and believed about it, and (3) how it may be broadly categorized. This general perspective sets the stage for the second part, which examines in detail one particular DZ variant called transverse initiation. This case was not chosen because it is the most common or important, but because it is the most amenable to inquiry. (DZ quantification is quite difficult, and the presented experiment is apparently the first conceived to quantify DZs of any type.) The technique uses a modified Cylinder Expansion (CYLEX) test, wherein a faster HE core (PBX 9501, 95wt% HMX) is surrounded by a slower HE annulus (PBX 9502, 95wt% TATB). The fast HE drives detonation in the slow HE faster than the latter would naturally propagate; however, in doing so a largely unreacted initiation layer is left in the slow HE adjacent to the fast HE. Combining the usual CYLEX diagnostics with detonation front curvature measurements and invoking a novel analysis based on the Gurney and Taylor methods, the energy release vs time (and distance) is inferred. The result is a ∼0.3μs induction time (~3mm induction lag), after which the energy release rises in ∼2μs (15mm) to half its asymptotic value: ∼86% of the energy released by a PBX 9502 detonation.

Silicon Fuel in High Performance Explosives

AbstractIn an effort to improve the insensitive munition (IM) response but maintain performance of aluminized formulations, silicon was investigated as a possible replacement for aluminum. An RDX‐based silicon explosive was developed in which nearly 90 % reaction of silicon to silicon dioxide was realized by 7 volume expansions as measured by the 2.54 cm diameter copper cylinder expansion test. In spite of the low nitramine loading in the formulation (79 wt.‐%), the corresponding Gurney constant for the explosive was 2.81±0.02 km s−1, which is superior to Composition A‐3 under the same experimental conditions (91 % RDX, 2.69±0.02 km s−1). Energy calculations from detonation calorimetry also indicate reaction of the silicon, which was further confirmed by both silicon metal and silicon dioxide in the analyzed residue. The energy release, despite it being equivalent to a highly loaded explosive, was found to lag behind the rate of A‐3. This indicates silicon oxidation may occur sometime after lighter gas reactions in the reaction front, but is fast enough to impart work in the copper cylinder test.