Search

The primary objective of this research is to investigate, through fundamental experiments, the dynamic multiaxial deformation, failure and strength degradation mechanisms that govern individual ballistic fiber failure. Predicting ballistic impact performance of armor systems requires an understanding of fiber failure under complex multiaxial loading conditions. This study examines the failure behavior of ultrahigh molecular weight polyethylene (UHMWPE) Dyneema® SK76 single fibers under dynamic transverse impact as a function of varying loading rates and projectile geometry. A novel single fiber transverse impact experiment is developed by modifying the Kolsky bar to characterize failure of fibers to create the foundation for a failure model. Experiments are performed with load cells at the fiber ends and with high speed imaging for determining average stresses and strains. Post-test microscopy imaging of failure surfaces are compared to determine the dominant fiber failure modes for each experimental group.

Bone specimens were collected from the frontal and parietal bones of 4 adult, human skulls. The microstructure was characterized using microcomputed tomography (micro- CT) at about 6-μm resolution to map the change of porosity as a function of the depth, P(d), from the inner surface nearest to the brain to the outer surface nearest to the skin. A quantifiable method was developed using the measured P(d) to objectively distinguish between the three layers of the skull: the outer table, diploe , and inner table. The thickness and average porosity of each of the layers were then calculated from the measured porosity distributions, and a Gaussian function was fit to the P(d) curves. Morphological parameters were compared between the two bone types (frontal and parietal), while accounting for skull-to-skull variability. Parietal bones generally had a larger diploe accompanied by a thinner inner table. The arrangement of the porous vesicular structure within the outer table was also obtained with micro-CT scans with longer scan times, using enhanced parameters for higher resolution and lower noise in the images. From these scans, the porous structure of the bone appeared to be randomly arranged in the transverse plane, compared to the porous structure of the human femur, which is aligned in the loading direction.

Full thickness skull specimens were extracted from the human crania, with both the inner and outer surfaces intact. The BVF-morphology (bone volume fraction) of these specimens had been previously characterized in detail and reported, with high-resolution micro-computed tomography at ~5 μm resolution. A subset of these specimens was loaded in the direction normal to the outer surface in quasi-static compression. In contrast to many previous mechanical characterization studies of skulls, following two additional procedures were used in this study. (1) Fresh skull specimens were used, which were stored refrigerated before mechanical loading, instead of using embalmed or dried specimens. (2) Furthermore, using digital image correlation, non-contact full-field inhomogeneous strain measurements were made using the speckled specimen surfaces and the compression platens, also avoiding possible errors in strain measurements from machine compliance and due to irregularities in the loading surfaces of the specimen. The averaged far-field compressive mechanical response was obtained from these local full-field measurements on the composite bone specimens. Assuming a layered structure for the skull bone, using the local averaged full-field strain measurements of each layer, a power law was used to represent the relationship between initial mechanical response and the averaged BVF of the layers. Using the measured porosity maps of the rest of the non-compressed specimens, this relationship was used to predict the modulus-depth dependency of the skull bone and the variabilities associated with the structure. The mechanical properties and density as a function of the normalized thickness of the skull are presented for use in finite element simulations to model the skull with the desired degrees of complexities, also based on the region of action, depending on the goals of the computer simulation of the impact: either as a single homogenous layer, three-layer sandwich, multilayer heterogeneous or continuous elemental structure. In addition, a power law was derived relating the compressive failure strength and bone volume fraction (BVF) for the skull bone.

The Gottingen minipig has been used as a surrogate in impact experiments designed to better understand the mechanisms by which mechanical loading induces traumatic brain injury (TBI). However, the relationship between mechanical response and structural morphology of the minipig cranium must be understood relative to the human skull in order to accurately scale any quantitative results, such as injury thresholds, from non-human TBI experiments to the human anatomy. In this study, bone specimens were dissected from the crania of adolescent Gottingen minipigs. These specimens were small cubes that contained the entire thickness of the skull. The microstructure of these skull specimens was quantified at the micron-length scale using micro-computed tomography (micro-CT). The skull was found to be highly porous near the skin-side surface and became less porous nearer the brain-side surface. The skull specimens were then loaded in quasi-static compression to obtain their mechanical response. The surface strain distribution on the specimen face was measured during loading using digital image correlation (DIC). The 2-D strain field formed a gradient of iso-strain bands along the thickness (depth) dimension from the skin-most to brain-most sides of the skull. The variation of the minipig microstructure along the thickness differed significantly from that of the adult human skull; thus the mechanical load transmission through the minipig skull is expected to be quite different from that of the human skull. The objective was to develop the methodology of relating the microstructure, as quantified by the bone volume fraction (BVF), to the mechanical response. The specimen was modeled by discretizing the depth dimension into a series of layers, which enabled the calibration of a power law relating the depth-dependent BVF to the depth-varying modulus. The relationship was used to predict moduli values for the adolescent minipig skull to provide updated, biofidelic parameters for finite element simulations at varying levels of complexity. Moreover, the methodology outlined in this paper can be applied to other skulls with different structural variations, such as the human.

Blunt impacts to the head can transfer energy to the skull and brain, which can cause injuries such as skull fracture, and traumatic brain injury (TBI). The presence of a cranial fracture has been shown to be associated with a higher incidence of intracranial lesions, neurological deficit, and poorer outcome. Therefore, it is important to understand the effect of blunt impact on skull deformation and fracture. In this study, human skull specimens were subjected to blunt loading at high rate in a controlled setting to quantify the deformation, loading, and fracture mechanisms that occur. The microstructure of the specimens were characterized using high-resolution CT (Northstar XRD 1620, at 25 μm resolution). After scanning, the skull specimens were loaded using a high-rate indentation technique. Digital image correlation was used to quantify the skull deformation. Here, the experimental setup and preliminary results are discussed.

Under extreme environments, such as a blast or impact event, the human body is subjected to high-rate loading, which can result in damage such as torn tissues and broken bones. The ability to numerically simulate these events would help improve the design of protective gear by iterating different configurations of protective equipment to reduce injuries. Computer codes capable of simulating these events require accurate rate-dependent material models representing the material deformation and failure (or injury) to properly predict the response of human body during simulation. Therefore, the high-rate material response must be measured to allow for simulation of high-rate events. This study seeks to quantify the high-rate mechanical response of human femoral cortical bone for use in high fidelity human anatomical models. Cortical bone compression specimens were extracted from the longitudinal and transverse directions relative to the long axis of the femur from three male donors, ages 36, 43, and 50. The compressive behavior of the cortical bone was studied at quasi-static (0.001/s), intermediate (1/s), and dynamic (~1000/s) strain rates using a split-Hopkinson pressure bar to determine the strain rate dependency and anisotropic effect on the strength of bone. The results indicate that cortical bone is anisotropic and stronger in the longitudinal direction compared to the transverse direction. The human cortical bone compressive response was also rate dependent in both directions, demonstrating significant increase in strength with increase in strain rate. Additionally, as the strain rate increased from intermediate to dynamic, a decrease in the elongation at transverse orientation was observed, which would indicate the bone becomes more brittle.

The Gottingen miniature pig is a useful surrogate to understand mechanisms of traumatic brain injury (TBI) in the human. However, the mechanical response of the minipig skull has not been previously reported. In this study, cranial samples were extracted from the skulls of adolescent minipigs (six months of age, average weight of 13.8 kg). The microstructure was first characterized using high-resolution μCT. A highly gradient structure was observed, with the bone volume fraction (BVF) almost doubling in the through-thickness (depth) dimension. These specimens were then mechanically loaded in quasi-static compression. The surface strain distribution along the loading direction was measured during the experiments using digital image correlation (DIC). Depth-dependent moduli were derived from the measured DIC strains rather than machine displacement, due to the large gradient in morphology. An elasticity-morphology relationship from literature was extended to represent the modulus variation in the functionally gradient skull structure (BVF), by calibrating the relationship with the experimentally derived local moduli. The model enables the prediction of local moduli based solely on the morphological parameter BVF measured with μCT, and also provided an estimation of the modulus of the bony phase of the cranium.

The failure behavior of cross-linked polymer epoxies with different glass transition temperatures (Tg) was investigated under Mode I fracture at high loading rate using a novel experimental method with in situ observation of the fracture process. By varying the monomer choices, the properties of the epoxies can be tailored to achieve greater resistance to cracking and higher impact toughness. For these experiments, a unique four-point bending specimen was used. High rate experiments were conducted on a modified split Hopkinson pressure bar with pulse-shaping. High speed digital imaging was used to visualize failure initiation. The images were also used with digital image correlation to optically measure the crack opening displacement and crack propagation velocity. The experimental results were used to calculate the energy required to initiate fracture at high loading rate. The results indicate that the critical energy required to initiate fracture at high loading rate was higher for epoxies with lower Tg values, up to an optimum Tg. This dependence of critical energy on the Tg of the epoxy was similar to that which has been previously measured for the epoxy’s impact resistance. In this paper, the experimental methods and results are discussed.

It is well known that amorphous polymers, such as polycarbonate (PC), will exhibit adiabatic heating due to the large plastic work that occurs when undergoing significant plastic deformation. However the extent of adiabatic heating has not been investigated with respect to strain rate, with full-field temperature measurements performed on the specimen during deformation. In this study, ASTM tensile dog bone PC specimens were used to investigate the rate-dependent mechanical response from quasi-static to intermediate (~10 s−1) strain rates using a traditional servo-hydraulic load frame. To determine the variations in yield and post-yield response at different locations of the gage area of the specimen, digital image correlation was used to measure the full-field surface strains. In addition, an InSb thermal camera was used concurrently to measure the full-field temperature distribution in the gage area during the deformation. The material experienced non-uniform heating as high as 50–70 °C, and showed significant rate sensitive mechanical response. In this paper, the experimental techniques and results are discussed.

The failure behavior of cross-linked polymer epoxies were investigated under Mode I fracture at low and high loading rates. By varying the monomer choices, the properties of the epoxies can be tailored to achieve a greater resistance to cracking and high impact toughness. For these experiments, a unique four-point bending specimen was developed. High rate experiments were conducted on a modified Split Hopkinson Pressure Bar (SHPB) with pulse-shaping. High speed digital imaging was used to determine the failure initiation, and allow for use of digital image correlation (DIC) to optically measure the crack opening displacement (COD) and crack propagation velocity. The experimental results are used to obtain the energy required to initiate fracture at various loading rates. The critical energy required to initiate fracture for this epoxy shows limited rate dependence from quasi-static to dynamic loading rates, but becomes rate sensitive when loaded at dynamic rates. The experimental procedures and results are discussed in this paper.

High performance fibers such as Kevlar KM2 used in protective vests are loaded primarily in tension at high loading rates. During the weaving and crimping process of fibers/yarns to make fabrics, individual fiber tensile strength may be reduced due to pre-straining of fibers by twisting. To systematically investigate the loading rate effects under different pre-twisted stress states, a miniaturized Kolsky tension bar is used to study single Kevlar KM2 fibers subject to varying levels of shear strain at high strain rates. Low and intermediate strain rate responses of the fibers are investigated using a Bose Electroforce loading system. In this paper, tensile strength as a function of pre-loaded shear strain on a single fiber is presented for different strain rates of loading. Fracture morphology of the fibers is also investigated using a high resolution scanning electron microscope to understand the mechanisms of failure.

During injury causing events such as impact, blunt trauma, penetration or blast, the human body is subjected to high rate loading. These events will deform and damage human tissue, frequently causing tearing in soft tissues and cracking/fracture in hard tissue (bone) due to the inherent brittleness of bone. For the development of accurate fracture models for computer simulation of bone fracture and also to understand the failure mechanisms of hard tissues during blast and impact events, it is necessary to investigate the failure behavior (fracture) of hard tissues at different loading rates, including high loading rates. In this study, the Mode I fracture behavior of cortical bone from the human femur, perpendicular to the long axis, is investigated as a function of loading rate, including dynamic loading rates which are accomplished with a modified Kolsky bar. The bone fracture specimens were extracted from three different human donors, ages 36, 50, and 43. For all three donors, the fracture toughness increases as loading rate increases from quasi-static to intermediate. As loading rate increases from intermediate to dynamic rates, the initiation fracture toughness decreased on average, while remaining higher than at quasi-static rate. The observed variations in initiation fracture toughness with donor age for each loading rate did not show any specific trends. The initiation fracture toughness of cortical bone is dependent on the path of the crack relative to the internal bone structure.

The strain rate response of PMMA was investigated under uniaxial compression at room temperature at strain-rates ranging from 0.0001/sec to about 4300/sec. In addition, the temperature response of PMMA was investigated at strain-rates of 1/sec and 0.001/sec at temperatures ranging from 0°C to 115°C (below Tg). High rate experiments at room temperature (greater than 1/sec rates) were conducted using a split-Hopkinson Pressure bar (SHPB) with pulse-shaping. This is necessary to induce a compressive loading on the specimen at a constant strain rate to achieve dynamic stress equilibrium. Results conducted at room temperature show that PMMA is strain rate sensitive from quasi-static to dynamic loading. Additionally, the stress-strain response exhibits a decrease in the flow stress with an increase in temperature. These experimental data are being used to develop constitutive behavior models of PMMA.

Blunt impact experiments were conducted on single and multi-layered Kevlar KM2 (Style 706) fabric at various velocities. To complement previous experiments with multi-ply fabric layers without resins, multi-ply laminates were fabricated from single or multiple plies of Kevlar fabric hot-pressed with a blended polyolefin (PO) resin consisting primarily of low molecular weight polyethylene to form a relatively rigid panel while maintaining a high fiber volume. During impact, full-field, back surface displacements were obtained using digital image correlation (DIC) technique using a pair of high-speed digital cameras configured stereoscopically. This allowed the full-field measurement of the fabric panel in three dimensions. These experimental results are to be used to evaluate different types of fabric and composite laminate models and simulation methodologies that are in literature. In this paper, experimental measurements are presented.

For decades, ballistic gelatin has been used as a tissue surrogate to test and evaluate bullets and firearms due to its similar viscosity to natural tissue. However, the high water content in ballistic gelatin makes it unstable at room temperature, and therefore causes it to have a poor shelf life. The development of polymer-based gels has shown promise as an alternative tissue surrogate. Polymer gels such as Perma-Gel are stable at room temperature and can be stored for long periods of time. Gels often fail due to tensile stresses during penetration. The failure behavior in tension is highly influenced by the presence of defects, such as cracks and voids, in the bulk material. A mode I experimental method was developed to obtain tensile failure criteria for the initiation and propagation behavior of these types of soft materials. Digital image correlation is used to determine the full-field surface strains around the crack tip to obtain a quantitative measure of the critical strain-field required for initiation and propagation of failure due to a defect. This systematic study utilizes these experimental techniques to determine the critical criteria for crack growth initiation and crack propagation of ballistic gelatins and a polymer gel as a function of loading rate. This paper presents experimental methodologies and results from Mode I fracture experiments including measured critical energy and strain-based criteria for failure initiation and growth, as well as their dependence on the rate of loading.

The commercially available ZylonTM single fiber has been investigated to understand the effect of strain rate on the tensile response. Experiments were conducted on single Zylon fibers across a range of strain rates ranging from 0.001/s to ~1000/s. The low rate uniaxial tensile experiments were performed using a Bose TestBenchTM system. The high rate dynamic tensile experiments were performed using a modified Kolsky Bar setup. This paper discusses the experimental setup and presents the effect of strain rate on the tensile deformation and failure behavior of this single fiber. These experimental data will be used to develop continuum and micro-mechanics based fabric models and computational simulation methodologies for impact of these fabrics by projectiles; this will allow engineers to better understand the impact behavior of composite structures created with fabrics made of this fiber.

Polycarbonate (PC) and polymethyl-methacrylate (PMMA) are commonly used materials for transparent protection. For increased effectiveness against impact, PC and PMMA are typically sandwiched and bonded in multiple layers of varying thicknesses to create a composite laminate. To develop high fidelity simulation methodologies for impact behavior for this type of laminated construction, panels were fabricated with different layers of PC and PMMA, on which blunt impact experiments were conducted. A high speed digital image correlation (DIC) technique was used to obtain full-field deformation measurements including out-of-plane displacement and surface strain. The experimental results are used to evaluate constitutive models and simulation methods for these various configurations of PC/PMMA composite laminates. In this paper, experimental technique and results are presented.