Improved knowledge connecting the chemistry, structure, and properties of polymers is necessary to develop advanced materials in a materials-by-design approach. Molecular dynamics (MD) simulations can provide tremendous insight into how the fine details of chemistry, molecular architecture, and microstructure affect many physical properties; however, they face well-known restrictions in their applicable temporal and spatial scales. These limitations have motivated the development of computationally-efficient, coarse-grained methods to investigate how microstructural details affect thermophysical properties. In this dissertation, I summarize my research work in structure-based coarse-graining methods to establish the link between molecular-scale structure and macroscopic properties of two different polymers. Systematically coarse-grained models were developed to study the viscoelastic stress response of polyurea, a copolymer that segregates into rigid and viscous phases, at time scales characteristic of blast and impact loading. With the application of appropriate scaling parameters, the coarse-grained models can predict viscoelastic properties with a speed up of 5-6 orders of magnitude relative to the atomistic MD models. Coarse-grained models of polyethylene were also created to investigate the thermomechanical material response under shock loading. As structure-based coarse-grained methods are generally not transferable to states different from which they were calibrated at, their applicability for modeling non-equilibrium processes such as shock and impact is highly limited. To address this problem, a new model is developed that incorporates many-body interactions and is calibrated across a range of different thermodynamic states using a least square minimization scheme. The new model is validated by comparing shock Hugoniot properties with atomistic and experimental data for polyethylene. Lastly, a high fidelity coarse-grained model of polyethylene was constructed that reproduces the joint-probability distributions of structural variables such as the distributions of bond lengths and bond angles between sequential coarse-grained sites along polymer chains. This new model accurately represents the structure of both the amorphous and crystal phases of polyethylene and enabling investigation of how polymer processing such as cold-drawing and bulk crystallization affect material structure at significantly larger time and length scales than traditional molecular simulations.

Stress corrosion cracking (SCC) is a materials degradation phenomena resulting from a combination of stress and a corrosive environment. Among the alphabet soup of proposed mechanism of SCC the most important are film-rupture, film-induced cleavage and hydrogen embrittlement.

This work examines various aspects of film-induced cleavage in gold alloys for which the operation of hydrogen embrittlement processes can be strictly ruled out on thermodynamic grounds. This is so because in such alloys SCC occurs under electrochemical conditions within which water is stable to hydrogen gas evolution. The alloy system examined in this work is AgAu since the corrosion processes in this system occur by a dealloying mechanism that results in the formation of nanoporous gold. The physics behind the dealloying process as well as the resulting formation of nanoporous gold is today well understood.

Two important aspects of the film-induced cleavage mechanism are examined in this work: dynamic fracture in monolithic nanoporous gold and crack injection. In crack injection there is a finite thickness dealloyed layer formed on a AgAu alloy sample and the question of whether or not a crack that nucleates within this layer can travel for some finite distance into the un-corroded parent phase alloy is addressed. Dynamic fracture tests were performed on single edge-notched monolithic nanoporous gold samples as well as “infinite strip” sample configurations for which the stress intensity remains constant over a significant portion of the crack length. High-speed photography was used to measure the crack velocity. In the dynamic fracture experiments cracks were observed to travel at speeds as large as 270 m/s corresponding to about 68% of the Raleigh wave velocity. Crack injection experiments were performed on single crystal Ag77Au23, polycrystalline Ag72Au28 and pure gold, all of which had thin nanoporous gold layers on the surface of samples. Through-thickness fracture was seen in both the single crystal and polycrystalline samples and there was an indication of ~ 1 μm injected cracks into pure gold. These results have important implications for the operation of the film-induced cleavage mechanism and represent a first step in the development of a fundamental model of SCC.

A new critical plane-energy model is proposed in this thesis for multiaxial fatigue life prediction of homogeneous and heterogeneous materials. Brief review of existing methods, especially on the critical plane-based and energy-based methods, are given first. Special focus is on one critical plane approach which has been shown to work for both brittle and ductile metals. The key idea is to automatically change the critical plane orientation with respect to different materials and stress states. One potential drawback of the developed model is that it needs an empirical calibration parameter for non-proportional multiaxial loadings since only the strain terms are used and the out-of-phase hardening cannot be considered. The energy-based model using the critical plane concept is proposed with help of the Mroz-Garud hardening rule to explicitly include the effect of non-proportional hardening under fatigue cyclic loadings. Thus, the empirical calibration for non-proportional loading is not needed since the out-of-phase hardening is naturally included in the stress calculation. The model predictions are compared with experimental data from open literature and it is shown the proposed model can work for both proportional and non-proportional loadings without the empirical calibration. Next, the model is extended for the fatigue analysis of heterogeneous materials integrating with finite element method. Fatigue crack initiation of representative volume of heterogeneous materials is analyzed using the developed critical plane-energy model and special focus is on the microstructure effect on the multiaxial fatigue life predictions. Several conclusions and future work is drawn based on the proposed study.

Polymer matrix composites (PMCs) are attractive structural materials due to their high stiffness to low weight ratio. However, unidirectional PMCs have low shear strength and failure can occur along kink bands that develop on compression due to plastic microbuckling that carry strains large enough to induce nonlinear matrix deformation. Reviewing the literature, a large fraction of the existing work is for uniaxial compression, and the effects of stress gradients, such as those present during bending, have not been as well explored, and these effects are bound to make difference in terms of kink band nucleation and growth. Furthermore, reports on experimental measurements of strain fields leading to and developing inside these bands in the presence of stress gradients are also scarce and need to be addressed to gain a full understanding of their behavior when UDCs are used under bending and other spatially complex stress states.

In a light to bridge the aforementioned gaps, the primary focus of this work is to understand mechanisms for kink band evolution under an influence of stress-gradients induced during bending. Digital image correlation (DIC) is used to measure strains inside and around the kink bands during 3-point bending of samples with 0°/90° stacking made of Ultra-High Molecular Weight Polyethylene Fibers. Measurements indicate bands nucleate at the compression side and propagate into the sample carrying a mixture of large shear and normal strains (~33%), while also decreasing its bending stiffness. Failure was produced by a combination of plastic microbuckling and axial splitting. The microstructure of the kink bands was studied and used in a microstructurally explicit finite element model (FEM) to analyze stresses and strains at ply level in the samples during kink band evolution, using cohesive zone elements to represent the interfaces between plies. Cohesive element properties were deduced by a combination of delamination, fracture and three-point bending tests used to calibrate the FEMs. Modeling results show that the band morphology is sensitive to the shear and opening properties of the interfaces between the plies.

Characterization and modeling of deformation and failure in metallic materials under extreme conditions, such as the high loads and strain rates found under shock loading due to explosive detonation and high velocity-impacts, are extremely important for a wide variety of military and industrial applications. When a shock wave causes stress in a material that exceeds the elastic limit, plasticity and eventually spallation occur in the material. The process of spall fracture, which in ductile materials stems from strain localization, void nucleation, growth and coalescence, can be caused by microstructural heterogeneity. The analysis of void nucleation performed from a microstructurally explicit simulation of a spall damage evolution in a multicrystalline copper indicated triple junctions as the preferred sites for incipient damage nucleation revealing 75% of them with at least two grain boundaries with misorientation angle between 20-55°. The analysis suggested the nature of the boundaries connecting at a triple junction is an indicator of their tendency to localize spall damage. The results also showed that damage propagated preferentially into one of the high angle boundaries after voids nucleate at triple junctions. Recently the Rayleigh-Taylor Instability (RTI) and the Richtmyer-Meshkov Instability (RMI) have been used to deduce dynamic material strength at very high pressures and strain rates. The RMI is used in this work since it allows using precise diagnostics such as Transient Imaging Displacement Interferometry (TIDI) due to its slower linear growth rate. The Preston-Tonks-Wallace (PTW) model is used to study the effects of dynamic strength on the behavior of samples with a fed-thru RMI, induced via direct laser drive on a perturbed surface, on stability of the shock front and the dynamic evolution of the amplitudes and velocities of the perturbation imprinted on the back (flat) surface by the perturbed shock front. Simulation results clearly showed that the amplitude of the hydrodynamic instability increases with a decrease in strength and vice versa and that the amplitude of the perturbed shock front produced by the fed-thru RMI is also affected by strength in the same way, which provides an alternative to amplitude measurements to study strength effects under dynamic conditions. Simulation results also indicate the presence of second harmonics in the surface perturbation after a certain time, which were also affected by the material strength.

CubeSats offer a compelling pathway towards lowering the cost of interplanetary exploration missions thanks to their low mass and volume. This has been possible due to miniaturization of electronics and sensors and increased efficiency of photovoltaics. Interplanetary communication using radio signals requires large parabolic antennas on the spacecraft and this often exceeds the total volume of CubeSat spacecraft. Mechanical deployable antennas have been proposed that would unfurl to form a large parabolic dish. These antennas much like an umbrella has many mechanical moving parts, are complex and are prone to jamming. An alternative are inflatables, due to their tenfold savings in mass, large surface area and very high packing efficiency of 20:1. The present work describes the process of designing and building inflatable parabolic reflectors for small satellite radio communications in the X band.

Tests show these inflatable reflectors to provide significantly higher gain characteristics as compared to conventional antennas. This would lead to much higher data rates from low earth orbits and would provide enabling communication capabilities for small satellites in deeper space. This technology is critical to lowering costs of small satellites while enhancing their capabilities.

Principle design challenges with inflatable membranes are maintaining accurate desired shape, reliable deployment mechanism and outer space environment protection. The present work tackles each of the mentioned challenges and provides an



understanding towards future work. In the course of our experimentation we have been able to address these challenges using building techniques that evolved out of a matured understanding of the inflation process.

Our design is based on low cost chemical sublimates as inflation substances that use a simple mechanism for inflation. To improve the reliability of the inflated shape, we use UV radiation hardened polymer support structures. The novelty of the design lies in its simplicity, low cost and high reliability. The design and development work provides an understanding towards extending these concepts to much larger deployable structures such as solar sails, inflatable truss structures for orbit servicing and large surface area inflatables for deceleration from hypersonic speeds when re-entering the atmosphere.

The objective of this research is to develop robust, accurate, and adaptive algorithms in the framework of the extended finite element method (XFEM) for fracture analysis of highly heterogeneous materials with complex internal geometries. A key contribution of this work is the creation of novel methods designed to automate the incorporation of high-resolution data, e.g. from X-ray tomography, that can be used to better interpret the enormous volume of data generated in modern in-situ experimental testing. Thus new algorithms were developed for automating analysis of complex microstructures characterized by segmented tomographic images.

A centrality-based geometry segmentation algorithm was developed to accurately identify discrete inclusions and particles in composite materials where limitations in imaging resolution leads to spurious connections between particles in close contact.To allow for this algorithm to successfully segment geometry independently of particle size and shape, a relative centrality metric was defined to allow for a threshold centrality criterion for removal of voxels that spuriously connect distinct geometries.

To automate incorporation of microstructural information from high-resolution images, two methods were developed that initialize signed distance fields on adaptively-refined finite element meshes. The first method utilizes a level set evolution equation that is directly solved on the finite element mesh through Galerkins method. The evolution equation is formulated to produce a signed distance field that matches geometry defined by a set of voxels segmented from tomographic images. The method achieves optimal convergence for the order of elements used. In a second approach, the fast marching method is employed to initialize a distance field on a uniform grid which is then projected by least squares onto a finite element mesh. This latter approach is shown to be superior in speed and accuracy.

Lastly, extended finite element method simulations are performed for the analysis of particle fracture in metal matrix composites with realistic particle geometries initialized from X-ray tomographic data. In the simulations, particles fracture probabilistically through a Weibull strength distribution. The model is verified through comparisons with the experimentally-measured stress-strain response of the material as well as analysis of the fracture. Further, simulations are then performed to analyze the effect of mesh sensitivity, the effect of fracture of particles on their neighbors, and the role of a particles shape on its fracture probability.

Al 7075 alloys are used in a variety of structural applications, such as aircraft wings, automotive components, fuselage, spacecraft, missiles, etc. The mechanical and corrosion behavior of these alloys are dependent on their microstructure and the environment. Therefore, a comprehensive study on microstructural characterization and stress-environment interaction is necessary. Traditionally, 2D techniques have been used to characterize microstructure, which are inaccurate and inadequate since the research has shown that the results obtained in the bulk are different from those obtained on the surface. There now exist several techniques in 3D, which can be used to characterize the microstructure. Al 7075 alloys contain second phase particles which can be classified as Fe-bearing inclusions, Si-bearing inclusions and precipitates. The variation in mechanical and corrosion properties of aluminum alloys has been attributed to the size, shape, distribution, corrosion properties and mechanical behavior of these precipitates and constituent particles. Therefore, in order to understand the performance of Al 7075 alloys, it is critical to investigate the size and distribution of inclusions and precipitates in the alloys along with their mechanical properties, such as Young's modulus, hardness and stress-strain behavior. X-ray tomography and FIB tomography were used to visualize and quantify the microstructure of constituent particles (inclusions) and precipitates, respectively. Microscale mechanical characterization techniques, such as nanoindentation and micropillar compression, were used to obtain mechanical properties of inclusions. Over the years, studies have used surface measurements to understand corrosion behavior of materials. More recently, in situ mechanical testing has become more attractive and advantageous, as it enables visualization and quantification of microstructural changes as a function of time (4D). In this study, in situ X-ray synchrotron tomography was used to study the SCC behavior of Al 7075 alloys in moisture and deionized water. Furthermore, experiments were performed in EXCO solution to study the effect of applied stress on exfoliation behavior in 3D. Contrary to 2D measurements made at the surface which suggest non-uniform crack growth rates, three dimensional measurements of the crack length led to a much more accurate measurement of crack growth rates.

In this thesis, a FORTRAN code is rewritten in C++ with an object oriented ap-

proach. There are several reasons for this purpose. The first reason is to establish

the basis of a GPU programming. To write programs that utilize GPU hardware,

CUDA or OpenCL is used which only support C and C++. FORTRAN has a feature

that lets its programs to call C/C++ functions. FORTRAN sends relevant data to

C/C++, which in turn sends that data to OpenCL. Although this approach works,

it makes the code messy and bulky and in the end more difficult to deal with. More-

over, there is a slight performance decrease from the additional data copy. This is

the motivation to have the code entirely written in C++ to make it more uniform,

efficient and clean. The second reason is the object oriented feature of the C++. The

“abstraction”, “inheritance” and “run-time polymorphism” features of C++ provide

some form of classes and objects, the ability to build new abstractions, and some

form of run-time binding, respectively. In recent years, some of popular codes has

been rewritten in C++ which were initially in FORTRAN. One of these softwares is

LAMMPS.

In this code the level set equation is solved by RLSG method to track the interface in

two phase flow. In gas/fluid flows, the surface tension is important and only exists at

the interface. Therefore, the location and some geometric features of interface need

to be evaluated which can be achieved by solving the level set equation.

Fracture phenomena have been extensively studied in the last several decades. Continuum mechanics-based approaches, such as finite element methods and extended finite element methods, are widely used for fracture simulation. One well-known issue of these approaches is the stress singularity resulted from the spatial discontinuity at the crack tip/front. The requirement of guiding criteria for various cracking behaviors, such as initiation, propagation, and branching, also poses some challenges. Comparing to the continuum based formulation, the discrete approaches, such as lattice spring method, discrete element method, and peridynamics, have certain advantages when modeling various fracture problems due to their intrinsic characteristics in modeling discontinuities.

A novel, alternative, and systematic framework based on a nonlocal lattice particle model is proposed in this study. The uniqueness of the proposed model is the inclusion of both pair-wise local and multi-body nonlocal potentials in the formulation. First, the basic ideas of the proposed framework for 2D isotropic solid are presented. Derivations for triangular and square lattice structure are discussed in detail. Both mechanical deformation and fracture process are simulated and model verification and validation are performed with existing analytical solutions and experimental observations. Following this, the extension to general 3D isotropic solids based on the proposed local and nonlocal potentials is given. Three cubic lattice structures are discussed in detail. Failure predictions using the 3D simulation are compared with experimental testing results and very good agreement is observed. Next, a lattice rotation scheme is proposed to account for the material orientation in modeling anisotropic solids. The consistency and difference compared to the classical material tangent stiffness transformation method are discussed in detail. The implicit and explicit solution methods for the proposed lattice particle model are also discussed. Finally, some conclusions and discussions based on the current study are drawn at the end.