James Guilkey

A component-based parallel infrastructure for the simulation of fluid–structure interaction

The Uintah computational framework is a component-based infrastructure, designed for highly parallel simulations of complex fluid–structure interaction problems. Uintah utilizes an abstract representation of parallel computation and communication to express data dependencies between multiple physics components. These features allow parallelism to be integrated between multiple components while maintaining overall scalability. Uintah provides mechanisms for load-balancing, data communication, data I/O, and checkpoint/restart. The underlying infrastructure is designed to accommodate a range of PDE solution methods. The primary techniques described here, are the material point method (MPM) for structural mechanics and a multi-material fluid mechanics capability. MPM employs a particle-based representation of solid materials that interact through a semi-structured background grid. We describe a scalable infrastructure for problems with large deformation, high strain rates, and complex material behavior. Uintah is a product of the University of Utah Center for Accidental Fires and Explosions (C-SAFE), a DOE-funded Center of Excellence. This approach has been used to simulate numerous complex problems, including the response of energetic devices subject to harsh environments such as hydrocarbon pool fires. This scenario involves a wide range of length and time scales including a relatively slow heating phase punctuated by pressurization and rupture of the device.

Extracellular Matrix Density Regulates the Rate of Neovessel Growth and Branching in Sprouting Angiogenesis

Angiogenesis is regulated by the local microenvironment, including the mechanical interactions between neovessel sprouts and the extracellular matrix (ECM). However, the mechanisms controlling the relationship of mechanical and biophysical properties of the ECM to neovessel growth during sprouting angiogenesis are just beginning to be understood. In this research, we characterized the relationship between matrix density and microvascular topology in an in vitro 3D organ culture model of sprouting angiogenesis. We used these results to design and calibrate a computational growth model to demonstrate how changes in individual neovessel behavior produce the changes in vascular topology that were observed experimentally. Vascularized gels with higher collagen densities produced neovasculatures with shorter vessel lengths, less branch points, and reduced network interconnectivity. The computational model was able to predict these experimental results by scaling the rates of neovessel growth and branching according to local matrix density. As a final demonstration of utility of the modeling framework, we used our growth model to predict several scenarios of practical interest that could not be investigated experimentally using the organ culture model. Increasing the density of the ECM significantly reduced angiogenesis and network formation within a 3D organ culture model of angiogenesis. Increasing the density of the matrix increases the stiffness of the ECM, changing how neovessels are able to deform and remodel their surroundings. The computational framework outlined in this study was capable of predicting this observed experimental behavior by adjusting neovessel growth rate and branching probability according to local ECM density, demonstrating that altering the stiffness of the ECM via increasing matrix density affects neovessel behavior, thereby regulated vascular topology during angiogenesis.

Simulation of Soft Tissue Failure Using the Material Point Method

Understanding the factors that control the extent of tissue damage as a result of material failure in soft tissues may provide means to improve diagnosis and treatment of soft tissue injuries. The objective of this research was to develop and test a computational framework for the study of the failure of anisotropic soft tissues subjected to finite deformation. An anisotropic constitutive model incorporating strain-based failure criteria was implemented in an existing computational solid mechanics software based on the material point method (MPM), a quasi-meshless particle method for simulations in computational mechanics. The constitutive model and the strain-based failure formulations were tested using simulations of simple shear and tensile mechanical tests. The model was then applied to investigate a scenario of a penetrating injury: a low-speed projectile was released through a myocardial material slab. Sensitivity studies were performed to establish the necessary grid resolution and time-step size. Results of the simple shear and tensile test simulations demonstrated the correct implementation of the constitutive model and the influence of both fiber family and matrix failure on predictions of overall tissue failure. The slab penetration simulations produced physically realistic wound tracts, exhibiting diameter increase from entrance to exit. Simulations examining the effect of bullet initial velocity showed that the anisotropy influenced the shape and size of the exit wound more at lower velocities. Furthermore, the size and taper of the wound cavity was smaller for the higher bullet velocity. It was concluded that these effects were due to the amount of momentum transfer. The results demonstrate the feasibility of using MPM and the associated failure model for large-scale numerical simulations of soft tissue failure.

Mixing kinematics of moderate Reynolds number flows in a T-channel

An experimental study of water flow in a T-shaped channel with rectangular cross section (20×20 mm2 inlet ID and 20×40 mm2 outlet ID) has been conducted for a Reynolds number Re range based on inlet geometry of Re=56–422. Dynamical conditions and T-channel geometry of the current study are applicable to the microscale. This study supports a large body of numerical work, and resolution and the interrogation region are extended beyond previous experimental studies. Laser induced fluorescence (LIF) permits a detailed look at the flow fields that evolve in the outlet channel over the broad range of Re where realistic T-channels operate. Scalar structures previously unresolved in the literature are presented. Unsteady flow regimes numerically predicted to occur at higher Re are characterized, and simultaneous planar and discrete-point LIF measurements relate the development of oscillatory behavior in the outlet channel to flow structure in the junction. Further, the development of an unsteady symmetric topolog...

Mechanical Interaction of Angiogenic Microvessels With the Extracellular Matrix

Angiogenesis is the process by which new blood vessels sprout from existing blood vessels, enabling new vascular elements to be added to an existing vasculature. This review discusses our investigations into the role of cell-matrix mechanics in the mechanical regulation of angiogenesis. The experimental aspects of the research are based on in vitro experiments using an organ culture model of sprouting angiogenesis with the goal of developing new treatments and techniques to either promote or inhibit angiogenic outgrowth, depending on the application. Computational simulations were performed to simulate angiogenic growth coupled to matrix deformation, and live two-photon microscopy was used to obtain insight into the dynamic mechanical interaction between angiogenic neovessels and the extracellular matrix. In these studies, we characterized how angiogenic neovessels remodel the extracellular matrix (ECM) and how properties of the matrix such as density and boundary conditions influence vascular growth and alignment. Angiogenic neovessels extensively deform and remodel the matrix through a combination of applied traction, proteolytic activity, and generation of new cell-matrix adhesions. The angiogenic phenotype within endothelial cells is promoted by ECM deformation and remodeling. Sensitivity analysis using our finite element model of angiogenesis suggests that cell-generated traction during growth is the most important parameter controlling the deformation of the matrix and, therefore, angiogenic growth and remodeling. Live two-photon imaging has also revealed numerous neovessel behaviors during angiogenesis that are poorly understood such as episodic growth/regression, neovessel colocation, and anastomosis. Our research demonstrates that the topology of a resulting vascular network can be manipulated directly by modifying the mechanical interaction between angiogenic neovessels and the matrix.

Verification tests in solid mechanics

Code verification against analytical solutions is a prerequisite to code validation against experimental data. Though solid-mechanics codes have established basic verification standards such as patch tests and convergence tests, few (if any) similar standards exist for testing solid-mechanics constitutive models under nontrivial massive deformations. Increasingly complicated verification tests for solid mechanics are presented, starting with simple patch tests of frame-indifference and traction boundary conditions under affine deformations, followed by two large-deformation problems that might serve as standardized verification tests suitable to quantify accuracy, robustness, and convergence of momentum solvers used in solid-mechanics codes. These problems use an accepted standard of verification testing, the method of manufactured solutions (MMS), which is rarely applied in solid mechanics. Body forces inducing a specified deformation are found analytically by treating the constitutive model abstractly, with a specific model introduced only at the last step in examples. One nonaffine MMS problem subjects the momentum solver and constitutive model to large shears comparable to those in penetration, while ensuring natural boundary conditions to accommodate codes lacking support for applied tractions. Two additional MMS problems, one affine and one nonaffine, include nontrivial traction boundary conditions.

Stochastic reconstruction of the microstructure of equilibrium form snow and computation of effective elastic properties

Three-dimensional geometric descriptions of microstructure are indispensable to obtain the structure-property relationships of snow. Because snow is a random heterogeneous material, it is often helpful to construct stochastic geometric models that can be used to model physical and mechanical properties of snow. In the present study, the Gaussian random field-based stochastic reconstruction of the sieved and sintered dry-snow sample with grain size less than 1 mm is investigated. The one- and two-point correlation functions of the snow samples are used as input for the stochastic snow model. Several statistical descriptors not used as input to the stochastic reconstruction are computed for the real and reconstructed snow to assess the quality of the reconstructed images. For the snow samples and the reconstructed snow microstructure, we also estimate the mechanical properties and the size of the associated representative volume element using numerical simulations as additional assessment of the quality of the reconstructed images. The results indicate that the stochastic reconstruction technique used in this paper is reasonably accurate, robust and highly efficient in numerical computations for the high-density snow samples we consider.

Mixing mechanisms in turbulent pipe flow

An experimental investigation of passive scalar mixing in turbulent pipe flow is carried out using a new non-intrusive scalar initialization technique. The measurements support a recently predicted similarity scaling of concentration spectra in flows that are unbounded in one direction. Reflecting this scaling, the scalar variance exhibits a power-law rather than exponential decay, indicating that the traditional plug-flow reactor picture of turbulent pipe-flow mixing omits key physical mechanisms.

A case study: visualizing material point method data

The Material Point Method is used for complex simulation of solid materials represented using many individual particles. Visualizing such data using existing polygonal or volumetric methods does not accurately encapsulate both the particle and macroscopic properties of the data. In this case study we present various methods used to visualize the particle data as spheres and explain and evaluate two methods of augmenting the visualization using silhouette edges and advanced illumination such as ambient occlusion. We also present informal feedback received from the application scientists who use these methods in their workflow.

Use of caged fluorescent dyes for the study of turbulent passive scalar mixing

The non-intrusive initialization of a flow field with distinct and spatially segregated scalar components represents a significant experimental difficulty. Here a new technique is described which makes possible the non-intrusive initialization of a spatially binary passive scalar field in a laminar or turbulent flow field. This technique uses photoactivatable (caged) fluorescent dyes dissolved in the flow medium. The scalar field within the flow field is tagged or initialized by “uncaging” the appropriate regions with an ultraviolet excimer laser. Mixing between the tagged and untagged regions is quantified using standard laser induced fluorescence techniques. The method is currently being used to study mixing in a turbulent pipe flow.