Shock-induced vaporization in metals

Results of well-controlled experiments on shock-induced vaporization studies in zinc, indium, and aluminum are presented. A titanium alloy impact at a velocity of 10.4 km/s will melt these materials totally. The expansion products upon release will consist of liquid–vapor mixtures. The ratio of liquid to vapor in the mixture depends on the material and also on the degree of expansion upon release. The impact generated debris propagates a gap dimension up to 125 mm before it stagnates against a stationary witness plate. The non-uniform spatial loading on the witness plate is determined using multiple velocity interferometers. Radiographic measurements of the debris cloud are also taken before it stagnates against the witness plate. Both radiographic and the velocity interferometer measurements suggest lateral and axial expansion. We have identified that the kinetics of the vaporization process can be related to the energy of the material shocked to the high-pressure state. In particular, the energy E of the material in the shocked state is expressed in units of the energy E_v required to vaporize a gram of material from room temperature. Results of these experiments indicate that the rate of vaporization is strongly dependent on E/E_v as it is increased by an order of magnitude from 1 to 10.     

Impact experiments on low-temperature bumpers

This paper presents the results of hypervelocity impact experiments that were carried out at CISAS Impact Facility onto aluminum bumpers cooled down to −120 °C with liquid nitrogen and to −60 °C with solid carbon dioxide. The thickness of the targets was 0.8, 1, 2 and 3.17 mm, the diameter of the spherical projectiles was 1.5, 1.9, 2.3 and 2.9 mm and the impact velocity did span between 4 and 5 km/s. To establish if any temperature dependence exists in the bumpers’ impact response, two different features were analyzed: the hole size and the bumper protection capabilities. The latter property, that is related to the bumper capacity of producing debris cloud composed of fragments as fine and slow as possible, was assessed through observation of the damage patterns on witness plates and through measurements of the debris cloud tip velocity. Moreover, qualitative analyses of high-speed shadowgraphs representing the debris cloud evolution were performed. On one hand, it was found that low temperature has only minor influence on the hole diameter. On the other hand, the examination of shadowgraphs showed that the debris cloud structure varies with bumper temperature, even though it was not proved that such differences correspond to significant dissimilarities between damage patterns recorded onto witness plates.     

Fragmentation of targets during ballistic penetration events

An analytical target fragmentation model is developed to predict the number of armor debris fragments produced in a ballistic penetration event. The model employs an energy balance to estimate the energy available to propagate cracks in the target material and thus produce an estimate of the mean number of behind-armor debris fragments. The expected number of behind-armor debris fragments agrees well with the results of conventional ordnance velocity target penetration behind-armor debris tests. This analytical fragmentation model and test results were used as the basis for the development of a means of calibrating the parameters of a Weibull statistical distribution function to predict the probability distribution of the number of debris fragments produced in a ballistic impact. It was concluded that the armor fragmentation model was a good predictor of the number of fragments produced in a ballistic penetration event deserving further investigation.     

Quantitative analysis of debris from SiC-fiber-reinforced aluminum-alloy targets impacted by spherical projectiles

Silicon-carbide-continuous-fiber-reinforced aluminum matrix composite targets were impacted with duralumin projectiles at velocities from 2.9 km/s to 4.3 km/s. The debris from the composite targets was monitored by flash, soft x-ray radiography. The spatial distribution of the leading half of the debris were quantified in terms of mass, velocity, and kinetic energy and compared with that of debris from monolithic aluminum-alloy targets. A material effect on the debris production was found through the experiment so that fragments from the composite targets were smaller in mass and size, but more in number than those from the monolithic bumper. An increase of the impact velocity brought the enhancement of fragmentation in the leading edge part of the debris produced from both the composite and the monolithic targets in comparison with lower velocity impacts. The 4.3 km/s impact for the composite gave the spatial densities of debris mass and kinetic energy biased toward the periphery of the debris cloud, while other lower velocity impacts gave the different densities biased toward a gun axis. Such peripheral distribution of debris was found at an impact velocity of 3.5 km/s for the monolithic target. In the present velocity range, the composite debris always exhibited its larger peripheral distribution than the monolithic one did.     

Projectile density, impact angle and energy effects on hypervelocity impact damage to carbon fibre/peek composites

This paper explores the effects of projectile density, impact angle and energy on the damage produced by hypervelocity impacts on carbon fibre/PEEK composites. Tests were performed using the light gas gun facilities at the University of Kent at Canterbury, UK, and the NASA Johnson Space Center two-stage light gas gun facilities at Rice University in Houston, Texas. Various density spherical projectiles impacted AS4/PEEK composite laminates at velocities ranging from 2.71 to 7.14 km/s. In addition, a series of tests with constant size aluminum projectiles (1.5 mm in diameter) impacting composite targets at velocities of 3, 4, 5 and 6 km/s was undertaken at incident angles of 0, 30 and 45 degrees. Similar tests were also performed with 2 mm aluminum projectiles impacting at a velocity of approximately 6 km/s. The damage to the composite was shown to be independent of projectile density; however, debris cloud damage patterns varied with particle density. It was also found that the entry crater diameters were more dependent upon the impact velocity and the projectile diameter than the impact angle. The extent of the primary damage on the witness plates for the normal incidence impacts was shown to increase with impact velocity, hence energy. A series of tests exploring the shielding effect on the witness plate showed that a stand-off layer of Nextel fabric was very effective at breaking up the impacting debris cloud, with the level of protection increasing with a non-zero stand-off distance.     

A numerical comparison of the ballistic performance of unitary rod and segmented-rods against stationary and moving oblique plates

The purpose of this paper is to investigate the ballistic performance of segmented-rods against stationary or moving oblique plates. To do this, a series of three-dimensional numerical simulations for the impact characteristics of segmented-rods (5 of L/D=1) into stationary or moving oblique thin-plate targets is conduced. To provide a base line data, an L/D=5 unitary rod projectile which has the same mass and kinetic energy is also considered. The ballistic characteristics of the projectiles are evaluated by examining the crater profile in a thick witness target that is placed behind the oblique plate. The impact velocities considered are 1400, 1800 and 2200 m/s. The results for the test range show that the unitary rod projectile shows better performance in the moving oblique target than the stationary one and the segmented-rods always show slightly better performance in the stationary target. From the impact velocity of 2200 m/s, the outstanding penetration performance of the segmented-rods can be observed. This trend is due to the interaction between the reactive plate and projectile. The extent of the interaction relies on the relative velocities of the plate and projectiles, the plate angle and extended total length of the segmented-rods     

Impact flash and debris cloud expansion of high-pure metal foils

Results of 2 mm aluminum spheres perforating Al, Cu, Mo, Au, Sn, and Zn metal foils of a purity > 99.9 % with thicknesses between 0.1 mm and 2.0 mm, densities of up to 20 g/cm³, melting temperatures of 500 – 3000 K and specific heats of fusion of 20 – 350 kJ/kg at impact velocities between v_p = 4.5 km/s and v_p = 9 km/s are presented. The influence of target thickness, target material properties and impact velocity on the perforation hole diameter, impact flash duration and expansion velocity, fragmentation and debris cloud formation at nearly constant areal density is demonstrated. The dependence of impact crater pattern at witness plates on target material density, thickness, impact velocity and areal density ratio between projectile and target material is discussed. For tin and lead evidence is given for the ability of digital scanning electron microscope analysis as an effective tool for indicating change of aggregation from solid into liquid and for the determination of relative projectile and target material quantities.     

A quantitative analysis of computed hypervelocity debris clouds

This paper presents quantitative analyses of computed hypervelocity debris clouds due to aluminum spheres, rods and disks impacting aluminum bumper plates. The computations were performed using an algorithm to convert distorted Lagrangian finite elements to meshfree particles. The analyses were performed using a new postprocessing algorithm. The combination of this computational approach and this postprocessing algorithm is also used to characterize behind-armor debris due to tungsten rods impacting steel plates at ballistic velocities, and the results are compared to test data. The quantitative analyses are an extension of previous qualitative comparisons to radiographs of hypervelocity debris clouds.     

Metallic witness packs for behind-armour debris characterisation

For the experimental characterisation of behind-armour debris so-called metallic witness packs can be used. A metallic witness pack consists of an array of metallic plates interspaced by polystyrene foam sheets. To quantify the fragment mass and velocity from the corresponding hole area and position in the individual witness plates, perforation threshold curves must be available to provide the ballistic limit velocity for each plate as a function of the fragment mass. Furthermore, the relation between the fragment impact velocity and hole size should be known. This paper presents the results of a collaborative study between Defence Research Establishment, Valcartier (DREV) and TNO Prins Maurits Laboratory, The Netherlands. The first phase of this project is an experimental study to develop perforation threshold and hole-growth curves for two metallic witness pack configurations. In addition, it is shown that a hydrocode can be succesfully used to predict perforation threshold velocities.     

Hypervelocity impact damage to composites

This paper presents the results of hypervelocity impact tests conducted on graphite/PEEK laminates. Both flat plate and circular cylinders were tested using aluminum spheres of varying size, travelling at velocities from 2–7 km/s. The experiments were conducted at several facilities using light gas guns. Normal and oblique angle impacts were investigated to determine the effect of impact angle, particle energy and laminate configuration on the material damage and ejecta plumes. Correlations were established between an energy parameter and the impact crater size, spallation damage and debris cone angle. Secondary damage resulting from the debris plume on adjacent composite structures was studied using high-speed photography and witness plates. It was observed that for hypervelocity impacts, the debris plume particles have sufficient energy to penetrate adjacent structures and cause major structural damage as well.     

Debris cloud distributions at oblique impacts

The purpose of this study is to obtain mass, spray angle and velocity distributions of fragments in debris cloud generated by oblique impacts on an aluminum alloy plate. Hypervelocity impact tests were performed with a two-stage light gas gun at Kyushu Institute of Technology. The impact angles were changed to 0°, 15°, 30°, 45° and 60°. The projectile impacted on the targets at 2 km/s. After the impact, the debris cloud was taken with flash X-rays and an ultra high-speed video camera. The fragments were then captured in a stack of polystyrene sheets. As a result, the projectile was broken up into smaller fragments by oblique impacts with the larger impact angles. Lower velocity fragments dispersed in wider spray angles according to the increase of the impact angles.     

Analysis of debris clouds produced by impact of aluminum spheres with aluminum sheets

Results of two-stage light gas gun testing of two diameters of aluminum spheres impacting 0.5 mm and 1.0 mm thickness aluminum plates were described in this paper. Impact velocities for these tests were between 3.16 km/s and 5.17 km/s. The components of debris cloud and damage patterns in the witness plate were described. The morphologic features of debris clouds such as shape, axial velocity, and diametral velocity were discussed. The size and number of fragments in the internal structure of debris cloud were not evaluated quantitatively, but described qualitatively. As a result, the shape of the leading face of the internal structure of debris cloud appeared to be sensitive to impact velocity, but not t/D ratio (bumper-thickness-to-projectile diameter ratio). The point at which the maximum diameter of the external bubble of debris cloud occurred had a same half spray angle of 30 degree and the last fragments ejected from bumper had a same half spray angle of 42 degree for each test. Fragments after the point mentioned above in the external bubble of debris cloud were ejected as several chains, the number of which is sensitive to impact velocity, but not t/D ratio. The changes in normalized velocity of the measurement points at debris cloud appeared the same trend as conclusions presented by Piekutowski except for the normalized internal structure expanding velocity. A certain value of t/D ratio, at two sides of which, the normalized internal structure expanding velocity appeared different variety trend existed.     

Debris fragment characterization in oblique hypervelocity impacts

A case history in debris characterization is presented for oblique impacts of chunky tungsten projectiles against thin plates. The integrated approach of scaled experiments and hydrocode simulations led to a semi-analytic model of behind the plate debris fragment distributions. This debris distribution model agreed quite well with the experimental fragment distributions derived from witness plate measurements. The 1/4 scale test program included three projectile masses, two target geometries (single and dual plates), a velocity range of 4–7 km/s and a strike angle range of 15–55 degrees. Close correlation of measured and predicted fragment distributions encouraged the extension of the model to higher velocities not currently obtainable in the laboratory.

The paper also includes discussions of critical features of debris in oblique hypervelocity impact, the scalability of fragment data, and the utilization of the derived fragment models in semi-analytic damage assessment codes.     

The influence of confinement on the penetration of ceramic targets by KE projectiles at 1.8 and 2.6 km/s

This paper presents the results of scale size experiments using a tungsten-alloy long-rod projectile fired against 97.5% Al₂O₃ ceramic targets at 1.8 and 2.6 km/s. Two targets were used, one having lateral steel confinement; the other without. The projectile overmatched the target, and residual projectile length and velocity were recorded using ballistic-syncro photography. Flash radiography was used during penetration of the unconfined target to obtain the penetration velocity. Manganin pressure gauges were also used to obtain additional data on the response of the ceramic target during penetration. Results from the eight experiments indicate that the confinement reduced the residual energy of the projectile at both impact velocities. Expressed in terms of the projectile impact energy, 55–56% was lost in the unconfined target at 2.6 km/s compared with 60% for the confined design. The same trend was found at 1.8 km/s with 68% and 72–73% for the unconfined and confined, respectively. Predictions using the QinetiQ GRIM2D hydrocode and a simplified form of the Johnson–Holmquist ceramic material model agreed well with the experiments for three out of the four test configurations. The predicted projectile erosion and retardation against the confined target at 1.8 km/s was excessively high. Analytical predictions using the Tate modified Bernoulli equation also gave reasonably accurate predictions for three of the tests, but values for the Tate target ‘strength’ extracted from experiments using a different target configuration were not accurate for the target design used in this paper.     

The influence of phase changes on debris-cloud interactions with protected structures

The physical state of the debris cloud generated by the interaction of a projectile with a thin target depends on the energy balance associated with the impact event. At impact velocities well above the sound speeds of the materials involved, the cloud is expected to consist of material in solid, liquid, and vapor phases. A series of numerical calculations using the multi-dimensional finite-difference hydrocode CTH has been used to evaluate the effect of phase changes (i.e., different vapor fractions) on these clouds, and on their subsequent interaction with backwall structures. In the calculations, higher concentrations of vapor are achieved either by (1) increasing the initial temperature of both the projectile and the thin shield while keeping the impact velocity constant, or (2) by actually increasing the impact velocity. The nature of the debris cloud and its subsequent loading on the protected structure depend on both its thermal and physical state. This interaction can cause rupture, spallation, or simply bulging of the backwall. These computational results are discussed and compared with new experimental observations obtained at an impact velocity of 10 km/s. In the experiment, the debris cloud was generated by the impact of a plate-shaped titanium projectile with a thin titanium shield.     

An engineering model to predict perforation damage to plates impacted by high velocity debris clouds

This paper describes experiments and the development of a model to predict damage to metallic plates impacted by high velocity, multi-particle debris clouds. The experiments involved single steel spheres fired at a steel shatter plate at speeds near 1.5 and 2.0 km/sec to generate the debris clouds. In each series of tests, the impact velocity was controlled, and a witness plate was placed at increasing distances behind the shatter plate to observe the effects of debris particle dispersion on plate damage. This paper focuses on the variations, with plate spacing, in the size of the central region removed from the witness plates. The central hole size model compares the post impact kinetic energy distribution in a witness plate impacted by a debris cloud to the free impact residual kinetic energy in an equivalent plate impacted by an L/D=1 steel cylinder, at the ballistic limit velocity. This approach permits extension of the model to other plate materials through utilization of existing ballistic limit velocity data.     

Debris clouds behind double-layer targets

It is demonstrated that the impedance mismatch and the order of the layers in two-layer sandwiches strongly influences the crater hole size formed in the target, the down-range debris cloud peak velocity, the fragment number and size, and the angles of downrange and uprange debris. Full and half scale test series with aluminum spheres of 10 mm and 5 mm diameter are performed with two-stage light gas guns against glued sandwiches of two layers at about equal areal density and different as well as equal shock impedances in the velocity range of 3–8 km/s. In the case of the titanium/tungsten plate sequence the transmitted shock wave is much stronger than for the tungsten/titanium target. This leads to a higher degree of fragmentation of the participated materials. For titanium/tungsten the hole diameter formed in the titanium layer is distinctly larger than in the tungsten layer for tungsten/titanium. For the titanium/tungsten target the larger crater diameter on the impact side is in agreement with the lower maximum debris cloud velocity.     

Phase three penetration

Phase three penetration is defined as the target penetration that occurs after a high-velocity penetrator has fully eroded. Phase 3 penetration is due to one or the combination of after-flow and secondary penetration of the target by the eroded penetrator debris. Recent experimental data for long tungsten rods penetrating confined boron carbide, aluminum nitride, and silicon carbide targets are used to investigate phase 3 penetration. It is found for these three ceramic targets that the onset impact velocity for the occurrence of phase 3 penetration is very roughly 2 km/s. The phase 3 penetration increases with impact velocity approximately as v². For these experiments the phase 3 penetration appears to be due almost entirely to secondary penetration.     

Measuring hypervelocity impact velocity from micrometeoroid crater geometry

Guided by half-space computer simulations showing hypervelocity impact crater formation for an iron particle impacting an aluminum target and characteristic crater geometry changes with impact velocity over the range 8–40 km s⁻¹, we examined normal surface crater views and cross-sectional views through craters (>0.5 mm diameter) from samples retrieved from the NASA LDEF satellite and examined in the scanning electron microscope (SEM). While geometrical features suggested in the computer simulations were indeed observed for micrometeoroid craters in 6061-T6 aluminum targets and 303 stainless steel targets, there was no consistent estimate for impact velocities in any of the experimental samples, and velocity estimates based on measuring ratios of ejecta width/crater diameter and ejecta height/crater depth as well as ejecta height/crater diameter varied from 8 to 42 km s⁻¹; over the same range simulated. These results point to the need to create reference data from actual hypervelocity impact experiments in the laboratory, and systematic observation of residual crater geometries in the SEM. These experiments also demonstrate the uncertainty in assuming a fixed impact velocity for all impact craters in space materials as well as an apparent futility in attempting to correlate impacting particle velocity with post-mortem characteristics of a given crater.