Search

The phase transition of crystalline and amorphous silica materials to a stishovite-like high-pressure phase has received considerable interest. While most work has focused on pure SiO2 glass (fused silica), studies identifying a similar high-pressure phase transformation in soda-lime glass have not been performed. Marked differences in the network structure and compressibility of soda-lime glass as compared to fused silica raise questions as to how a stishovite transition might manifest in these glasses. In this work, new plate impact experiments on soda-lime glass have been performed to ∼110 GPa to measure the Hugoniot response in the high-pressure regime. Results of these tests were unable to find evidence of a transition to a high-pressure polymorph under shock loading. A glass constitutive model is proposed to capture this newly measured high-pressure response. Good agreement between time resolved measurements and simulation results from the CTH hydrocode with this model implemented are shown, along with additional insights into the apparent glass behavior.

The work presented in this paper details both an experimental program and an associated numerical modeling effort to characterize and predict the ballistic response of S-2 glass/SC15 epoxy composite panels. The experimental program consisted of ¼ inch diameter soft carbon steel spheres impacting ¼ and ½ inch thick flat composite panels at velocities ranging from 220 to 1570 m/s. High speed cameras were used to capture the impact event and resulting residual velocity of the spheres for each test configuration. After testing, each panel was inspected both visually and with ultrasonic C-scan techniques to determine the extent and depth of damage imparted on the panel by the impactor. The numerical modeling efforts utilized the anisotropic multi-constituent composite model (MCM) within the CTH shock physics hydrocode. The MCM model allows for evaluation of damage at the constituent level through continuum averaged stress and strain fields. The model also accounts for the inherent coupling of the equation of state and strength response that occurs in anisotropic composite materials. Finally, the simulation results are compared against the experimentally measured residual velocity as a quantitative metric and against the measured damage extent and patterns as a qualitative metric. The comparisons show good agreement in residual velocity and damage extent.

This article details the implementation and application of the glass-specific computational constitutive model by Holmquist and Johnson (J Appl Mech 78:051003, 2011) to simulate the dynamic response of soda-lime glass under high rate and high pressure shock conditions. The predictive capabilities of this model are assessed through comparison of experimental data with numerical results from computations using the CTH shock physics code. The formulation of this glass model is reviewed in the context of its implementation within CTH. Using a variety of experimental data compiled from the open literature, a complete parameterization of the model describing the observed behavior of soda-lime glass is developed. Simulation results using the calibrated soda-lime glass model are compared to flyer plate and Taylor rod impact experimental data covering a range of impact and failure conditions spanning an order of magnitude in velocity and pressure. The complex behavior observed in the experimental testing is captured well in the computations, demonstrating the capability of the glass model within CTH.

Composite materials are used as alternatives to conventional metallics in a multitude of applications including military ground vehicles, aircraft, space launch and re-entry vehicles and even personnel protection where weight savings are critical. In application, these materials are susceptible to high velocity impacts from various threats and it is essential that the response of these materials, under relevant conditions, be understood in order to provide optimized effective designs. This work details an on-going effort to validate the anisotropic multiple constituent model (MCM) within the CTH hydrocode. Within the CTH framework, the anisotropic MCM model is coupled with an equation of state (EOS) and provides continuum averaged stress and strain fields for each constituents (fiber and resin) of a composite microstructure from which progressive damage evaluations can be performed. In this paper we focus on recent validation efforts where woven S2/SC15 (glass/epoxy) composite panels were impacted with steel spheres at various impact velocities and angles of obliquity. The experimental testing was performed at the Shock Thermodynamics Applied Research (STAR) Facility at Sandia National Laboratories to provide data for further validation of the MCM model under oblique impact conditions. Oblique impacts result in stress fields which exercise the anisotropy of the strength model and the EOS coupling of the MCM model more robustly. Results are presented for both the CTH MCM model predictions and the experimental testing. The primary comparison metrics evaluated are the predicted and observed damage extent, overall damage pattern, and residual velocity of the projectile.

In many structural applications the use of composite material systems in both retrofit and new design modes has expanded greatly. The performance benefits from composites such as weight reduction with increased strength, corrosion resistance, and improved thermal and acoustic properties, are balanced by a host of failure modes whose genesis and progression are not yet well understood. As such, structural health monitoring (SHM) plays a key role for in-situ assessment for the purposes of performance/operations optimization, maintenance planning, and overall life cycle cost reduction. In this work, arrays of fiber Bragg grating optical strain sensors are attached to glass-epoxy solid laminate composite specimens that were subsequently subjected to specific levels of fully reversed cyclic loading. The fatigue loading was designed to impose strain levels in the panel that would induce damage to the laminate at varying numbers of cycles. The objectives of this test series were to assess the ability of the fiber Bragg grating sensors to detect fatigue damage (using previously developed SHM algorithms) and to establish a dataset for the development of a prognostic model to be applied to a random magnitude of fully reversed strain loading. The prognostic approach is rooted in the Failure Forecast Method, whereby the periodic feature rate-of-change was regressed against time to arrive at a failure estimate. An uncertainty model for the predictor was built so that a probability density function could be computed around the time-of-failure estimate, from which mean, median, and mode predictors were compared for robustness.

On-going research into the complexities of polyurea behavior under shock loading has led to some breakthroughs in the predictive simulation of how this nominally soft polymer responds under high velocity impact conditions. This work expands upon a previously reported modified pressure-dependent viscoelastic constitutive model for polyurea and its performance under ballistic impact. Specifically, we present recent enhancements to the model including nonlinearites in the Hugoniot and improvements in the high-temperature viscoelastic behavior, which substantially improves accuracy and extends the model’s range of applicable conditions. These improvements are demonstrated through correlation of computations for a suite of normal and pressure-shear plate impact experiments well documented in the open literature. Additionally, microparticle impact experiments were performed on polyurea using a laser-induced particle impact test (LIPIT) technique. High-speed imaging of the impact mechanics revealed elastic particle rebound at low velocity but penetration at high velocity. Simulation of these LIPIT experiments demonstrates good accuracy of the polyurea model under these conditions as well as provides insight into the mechanisms governing the results observed.

This paper details a numerical modeling and experimental test program focused on high energy ballistic impacts on composite rib-stiffened structures. The numerical model used in this paper is designated as the Multiconstituent Composite Model (MCM), which is implemented within the CTH shock physics code for simulation of ballistic impact damage on composite structures. The presented work utilized a building block approach, where component level flat panels were studied first, followed by sub-structure level T-joint specimens, both fabricated from a carbon/epoxy composite material. All numerical studies and experimental tests utilized the military grade 0.50 caliber M2 Ball round as a projectile. Following testing, each panel was inspected both visually and using Ultrasonic C-Scan techniques to determine the extent of damage sustained upon impact. Finally, comparisons of the experimental data with the numerical predictions are presented and discussed. Both the flat panel and T-joint damage predictions were in good agreement with the experimental data.

Impact loading response of unidirectional and plain weave fiber reinforced polymer composite materials is typically modeled using a Lagrangian method such as the finite element method. However, these methods often lack a coupled equation of state. In anisotropic materials, the pressure (equation of state) and deviatoric (strength) portions of the stress tensor are coupled: a shear stress can produce a volumetric response and a volumetric strain can produce a shear stress. High-velocity impacts of composite materials instigate a coupled pressure and stress response, so an equation of state is important in determining the non-uniform stress response of the material. A new composite model, which couples the pressure response to the constitutive response of the material, has been implemented in an Eulerian large deformation, strong shock wave, solid mechanics code. Experiments of steel projectiles perforating composite targets were numerically simulated to begin to validate this new composite model. This paper will discuss the coupled equation of state and strength response, and compare the results of these experiments with the results predicted by the model.

Failure of composite materials often results from damage accumulation in the individual constituents (fiber and matrix) of the composite. At times, damage may even be limited to a single constituent. The ability to accurately predict not only ultimate strength values but also intermediate constituent level failures is crucial to the success of introducing composite materials into demanding structural applications. In this paper, we develop two progressive failure models for the analysis of a plain weave composite material. The formulations are based on treating the weave as consisting of separate but linked continua representing the warp fiber bundles, fill fiber bundles, and pure matrix pockets. Retaining constituent identities allows one to access constituent (phase averaged) stress fields that are used in conjunction with both a stress based and damage based failure criterion to construct a nonlinear progressive failure algorithm for the woven fabric composite material. The MCT decomposition and the nonlinear progressive failure algorithm are incorporated within the framework of a traditional finite element analysis. The constituent based progressive failure algorithm combined with both the stress based and damage based failure criteria are compared against experimental data for a plain weave, woven fabric composite under various loading conditions. The analytical results from the damage based approach show a marked improvement over the stress based predictions and are in excellent agreement with the experimental data.

Multicontinuum theory (MCT) is based on the two-constituent decomposition developed by Hill [Hill R. Elastic properties of reinforced solids: some theoretical principles. J Mech Phys Sol 1963;11:357–72] to calculate the constituent level stress/strain fields of a composite material from the ‘smeared’ composite stress/strain fields. The work presented in this paper expands previous isothermal MCT work to include thermal effects related to large temperature gradients caused by fires. Analytical fire conditions are simulated using a 1-D heat diffusion equation, which provides time-dependent temperature gradients through the thickness of the composite. Results illustrate that when the composite materials are subjected to extreme thermal conditions, the thermal residual stresses resulting from the fiber and matrix coefficient of thermal expansion mismatch are significant. Progressive failure simulations are performed using the thermal–mechanical MCT algorithm, in which an increased rate of failure is observed for a composite material when it is subjected to large external temperature gradients similar to those experienced in fire environments.

An accurate assessment of accumulative damage of a composite ship structure subjected to fire is strongly reliant on the accurate characterization of the time dependent temperature distribution within the composite system. Current state-of-the-art analysis tools assume that thermal–mechanical properties of composite structures are independent of temperature change and mass loss. In this paper, a temperature and mass dependent heat diffusion model is developed to characterize the temperature and mass dependent heat conduction, energy consumption resulting from the decomposition, and the energy transfer associated with vaporous migration. The temperature dependent thermal conductivity and specific heat capacity are determined for the composite at a given resin decomposition stage using a recently developed small-scale test apparatus. Given the temperature dependent thermal properties of the fiber and resin materials, the resulting temperature dependent thermal properties of a woven fabric composite are computed from the newly developed thermal–mechanical analysis tool (TMAT). To assess the effect of using single heating rate based mass loss on the composite fire response prediction, the kinetic parameters are obtained from the post-processing of thermogravimetric analysis (TGA) test data conducted at various heating rates. The effects of temperature dependent thermal conductivity, specific heat capacity, and kinetic parameters determined at different heating rates are explored through the application of the temperature and mass dependent fire model to a composite plate subjected to a hydrocarbon fire. The accuracy of the temperature and mass dependent thermal model is demonstrated by comparing its response prediction with available experimental data.

The inability to accurately predict the onset of failure in modern fabric-reinforced composite materials has hindered the implementation of these materials into the mainstream applications. Excessive qualification and structural testing must be performed on composite members as a result of uncertainty in the performance of the material, resulting in significantly higher expenses associated with these materials. This situation can only be improved through the rigorous development of both improved failure and material response prediction capabilities and improved experimental verification. The current study addresses these issues, as well as explores recent developments in numerical predictions and experimental techniques. A combined numerical investigation of damage initiation mechanics and experimental verification of predicted results for a woven glass–vinyl ester composite material is performed. More specifically, 18-oz biased (5 warp/4 fill rovings) plain weave E-glass–vinyl ester laminate with warp rovings oriented in [0/90]s and [0/90/±45]s configurations are investigated. Experimental data for the E-glass–vinyl ester [0/90]s laminate is summarized in a two-dimensional biaxial failure envelope. The feasibility of using thickness-tapered cruciform specimens for generating the experimental biaxial test data is also addressed. Finally, numerical predictions of failure of the fabric-reinforced laminate are developed and compared against the experimental failure envelope for the cross-ply laminate.

An ongoing challenge within the structural analysis community is to accurately predict damage and progressive failure in large-scale structural components composed of composite materials. Multicontinuum technology (MCT) provides a means to produce constituent (fiber and matrix) level stress and strain information within the framework of commercial finite element analysis. Constituent level stress/strain information forms the basis for a progressive failure algorithm that has successfully been used to predict coupon failure in a wide range of composite materials and laminate configurations. In this paper, MCT is used to analyze the failure of rib-stiffened panels associated with advanced composite grid-stiffened structure (AGS) designs. More specifically, MCT is used to predict and analyze the separation of the rib to skin interface for comparison with tee pulloff and tee bend test data. Predictions for the initial and final separation of the rib to skin interfaces are shown to be in good agreement with experimental test data. The results also lend insight into design and manufacturing considerations that are key to the strength and performance of the rib to panel interface.

Multicontinuum theory (MCT) refers to the use of phase averaged constituent stress/strain fields for predicting failure in composite structural analysis. Given the composite material mechanical properties as well as those of the constituents, well known closed form algebraic expressions exist to decompose the composite stress/strain fields down to the constituent level. Recent research indicates constituent based failure algorithms show a great deal of promise in predicting material failure when coupled to nonlinear finite element codes. A limitation of MCT is that the traditional constituent decomposition is only valid for materials composed of two constituents. In this paper, the MCT decomposition is generalized to handle composite materials composed of three constituents. The application of interest is a woven fabric composite material. The three constituents consist of the warp bundles, fill bundles, and pure matrix pockets. Numerical results are presented for the proposed three-constituent decomposition and are shown to be in good agreement with phase averaged stresses obtained from direct volume averaging of finite element micromechanics models.

An alternative composite material failure model utilizing multicontinuum theory (MCT) is compared against experimental dynamic testing results. The investigation examines the advantages of the failure model, along with current model limitations. The experimental procedure used as the control data involved testing solid laminate glass reinforced plastic (GRP) composite specimen of varying thickness under a similar dynamic load. The testing subjected the solid laminate specimen to an air-backed, dynamic pressure load environment. All of the composite specimens were fitted with strain gauges on both the dry and wet surface of the specimen. Post test experimental analysis consisted of nondestructive ultrasonic (UT) evaluations to determine the damaged regions of the specimens. Experimental strain results and UT failure regions are compared with the composite material failure model.

The composite material research and development performed over the last year has greatly enhanced the capabilities of CTH for non-isotropic materials. The enhancements provide the users and developers with greatly enhanced capabilities to address non-isotropic materials and their constitutive model development. The enhancements to CTH are intended to address various composite material applications such as armor systems, rocket motor cases, etc. A new method for inserting non-isotropic materials was developed using Diatom capabilities. This new insertion method makes it possible to add a layering capability to a shock physics hydrocode. This allows users to explicitly model each lamina of a composite without the overhead of modeling each lamina as a separate material to represent a laminate composite. This capability is designed for computational speed and modeling efficiency when studying composite material applications. In addition, the layering capability also allows a user to model interlaminar mechanisms. Finally, non-isotropic coupling methods have been investigated. The coupling methods are specific to shock physics where the Equation of State (EOS) is used with a nonisotropic constitutive model. This capability elastically corrects the EOS pressure (typically isotropic) for deviatoric pressure coupling for non-isotropic materials.

The rapidly growing acceptance of Elastic Memory Composite (EMC) materials for deployable space structures illustrates the fact that EMC materials offer substantial performance improvements over existing spaceflight-qualified materials. This growing acceptance has, however, revealed a pressing need for further advancement of models to accurately predict key, non-classical, aspects of the material’s behavior. A critical deformation mechanism dominating the bending response of EMC materials at elevated temperatures is fiber microbuckling occurring within the softened resin. This paper will focus on analysis efforts aimed at predicting the geometric nonlinear response of EMC materials loaded in bending including fiber microbuckling. The analysis is based on a finite strain (Lagrangian) micromechanics analysis of a lamina in bending with an initial fiber waviness. Comparison of analytical predictions to experimental results for structural applications are provided.

This study details and demonstrates a strain-based criterion for the prediction of polymer matrix composite material damage and failure under shock loading conditions. Shock loading conditions are characterized by high-speed impacts or explosive events that result in very high pressures in the materials involved. These material pressures can reach hundreds of kbar and often exceed the material strengths by several orders of magnitude. Researchers have shown that under these high pressures, composites exhibit significant increases in stiffness and strength. In this work we summarize modifications to a previous stress based interactive failure criterion based on the model initially proposed by Hashin, to include strain dependence. The failure criterion is combined with the multi-constituent composite constitutive model (MCM) within a shock physics hydrocode. The constitutive model allows for decomposition of the composite stress and strain fields into the individual phase averaged constituent level stress and strain fields, which are then applied to the failure criterion. Numerical simulations of a metallic sphere impacting carbon/epoxy composite plates at velocities up to 1000 m/s are performed using both the stress and strain based criterion. These simulation results are compared to experimental tests to illustrate the advantages of a strain-based criterion in the shock environment.

In this paper, we examine a modified mechanics of materials based approach for predicting average interlaminar shear stress adjacent to free edges in laminated composite structures. Finite element models are constructed to study the free edge Interlaminar Shear Stress (ILSS) response of laminates under both tensile and bending loading conditions. Based on the finite element model results it was observed that the average ILSS for a bending condition was less than that for a pure membrane loading condition for the same local in-plane strain. The results also showed that the location of the lamina interface relative to the neutral axis influenced the difference between the bending and membrane induced free edge ILSS. Motivated by the work of Joo and Sun 1 who treated the membrane loading case only, an analytical expression is proposed for predicting free edge ILSS due to bending. Unlike the expression developed by Joo and Sun, the expression proposed here for critical interface ILSS for bending is not a simple linear function of the extension-shear coupling coefficients. The agreement with finite element results is very good.

The inability to accurately predict the onset of failure in modern fabric reinforced composite materials has hindered the implementation of these materials into mainstream applications. Excessive qualification and structural testing must be performed on composite members as a result of uncertainty in the performance of the material, resulting in significantly higher expenses associated with these materials. This situation will only be improved through the rigorous development of both improved failure and material response prediction capabilities and improved experimental verification. The current study addresses these issues, as well as explores recent developments in numerical predictions and experimental techniques. A combined numerical investigation of damage initiation mechanics and experimental verification of predicted results for a woven glass/vinyl ester composite material was performed. More specifically, 18-oz biased (5 warp/4 fill rovings) plain weave E-glass/vinyl ester laminate with warp rovings oriented in [0/90] s and [0/90/±45]s configurations were investigated. Experimental data for the E-glass/vinyl ester [0/90] s laminate is summarized in a two-dimensional biaxial failure envelope. The feasibility of using thicknesstapered cruciform specimens is also addressed. Numerical predictions of failure of the fabric reinforced laminate are developed and compared against the experimental failure envelope for the cross-ply laminate.