real-time, high-speed dynamic observations can in principle provide even better indications of failure modes. However, it is difficult to simultaneously achieve both high spatial and high temporal resolution. Future advances in instrumentation will bring new insights to the complex interplay of deformation and failure mechanisms during penetration.

Partially penetrated targets are particularly useful for determining failure mechanisms. A close examination of areas where the damaged material remains in place and of polished cross sections taken on a plane containing the shot line demonstrates how damage varies with distance from the side and distance ahead of the penetrating object. Such observations also suggest how damage evolves, thereby providing notions for equations describing damage development. The next section illustrates the failure mechanisms invoked by a penetrator by presenting damage observations in penetrated and partially penetrated targets of metals and alloys, ceramics and glasses, and polymeric materials. This is followed by a short discussion on the damage mechanisms in cellular materials invoked by blast loads.


Consider the case of a rod impacting a steel plate (Figure 3-1). If the plate is relatively soft compared to the rod, perforation may occur by homogeneous plastic flow of the plate, with little or no damage to the rod. A hardened plate, on the other hand, may fail by shear banding and consequent liberation of a plug of material pushed out by the projectile (Figure 3-1). Reflected stress waves from the rear surface of the plate may produce tensions large enough to nucleate, grow, and coalesce voids or microcracks, causing spallation. Thus, the result of an encounter between a rod and plate is determined by microscopic failure processes such as homogeneous plastic flow, shear banding, and tensile fracture in the plate and in the rod. The impact conditions and properties of both plate and rod determine which failure processes operate. Typically, however, it is a combination of simultaneously active failure modes that governs the outcome of the encounter.


FIGURE 3-1 Impact on steel plate. Rod impacting a plate at 90 degrees (left), and cross section showing simple plugging of rolled homogeneous armor by shear instabilities and back surface spalling by nucleation, growth, and coalescence of voids and cracks (right). SOURCE: Erlich, D.C., L. Seaman, D.A. Shockey, and D.R. Curran. 1980. Development and Application of a Computational Shear Band Mode. Menlo Park, Calif.: SRI International.

Other failure modes may be invoked at higher velocities. Figure 3-2 shows a polished and etched cross section through the crater in a 1-in.-thick steel plate that has been impacted at 6 km/s by a 12.7-mm-diameter polycarbonate sphere.2,3 Adiabatic4 shear bands can be seen as white-etching bands of hard, untempered martensite extending into the plate (1), surfaces of strain localization that look like bands when seen edge-on. The path of the bands is followed by brittle cracks (2), which intersect with other cracks and liberate fragments. Just below the crater are spherical voids (3), a manifestation of ductile tensile failure; these are linked by shear bands. Homogeneous plastic flow (4) is made clear by the deviation of the process rolling lines from the horizontal. Ultimately, the hemispherical volume of dark-etching material just below the point of experienced α↔ε polymorphic phase change brought about by pressure (5). The grain size is refined and the transformed material is significantly hardened. The boundary of the dark-etching material is a 130 kbar isobar. Thus, five failure modes operated at once, with the stress relaxation effect of each mode affecting the behavior of the others.

The damage beneath the crater in Figure 3-2 is complex and seems at first nearly impossible to interpret, yet it reveals how the material is failing. Such damage “hieroglyphics” must nevertheless be read and understood in order to predict penetration behavior and design microstructures with enhanced protective capabilities. Key to developing a deeper understanding are laboratory experiments that isolate each specific damage mechanism. This would allow each failure mode to operate under a range of well-controlled rate, temperature, and stress state conditions, providing the opportunity to study and quantitatively describe its evolution by means of real-time observation or post-test analysis of tests interrupted at various stages of damage development.

Finding 3-1. Ballistic penetration of metals can occur by five failure modes—adiabatic shear bands, cracks, voids, plastic deformation, and phase changes—more than one or all of which can occur simultaneously.


2Shockey, D.A., D.R. Curran, and P.S. DeCarli. 1975. Damage in steel plates from hypervelocity impact, I: Physical changes and effects of projectile material. Journal of Applied Physics 46(9): 3766-3775.

3Bertholf, L.D., L.D. Buxton, B.J. Thorne, R.K. Byers, A.L. Stevens, and S.L. Thompson. 1975. Damage in steel plates from hypervelocity impact II: Numerical results and spall measurement. Journal of Applied Physics 46(9): 3776-3783.

4“Adiabatic” refers to any process which occurs without heat transfer.

The National Academies of Sciences, Engineering, and Medicine
500 Fifth St. N.W. | Washington, D.C. 20001

Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement