Thermo-mechano-chemical modeling and computation of thermosetting polymers used in post-installed fastening systems in concrete structures.

As thermoset polymers find frequent implementation in engineering design, their application in structural engineering is rather limited. One key reason relies on the ongoing curing process in typical applications such as post-installed adhesive anchors, joints by structural elements or surface-mounted laminates glued by adhesive polymers. Mechanochemistry including curing and aging under thermal as well as mechanical loading causes a multiphysics problem to be discussed. For restricting the variety of material models based on empirical observations, we aim at a thermodynamically sound strategy for modeling thermosets. By providing a careful analysis and clearly identifying the assumptions and simplifications, we present the general framework for modeling and computational implementation of thermo-mechano-chemical processes by using open-source codes.

Computational Modeling of Polymer Matrix Based Textile Composites.

A simple approach to the multiscale analysis of a plain weave reinforced composite made of basalt fabrics bonded to a high performance epoxy resin L285 Havel is presented. This requires a thorough experimental program to be performed at the level of individual constituents as well as formulation of an efficient and reliable computational scheme. The rate-dependent behavior of the polymer matrix is examined first providing sufficient data needed in the calibration step of the generalized Leonov model, which in turn is adopted in numerical simulations. Missing elastic properties of basalt fibers are derived next using nanoindentation. A series of numerical tests is carried out at the level of yarns to promote the ability of a suitably modified Mori-Tanaka micromechanical model to accurately describe the nonlinear viscoelastic response of unidirectional fibrous composites. The efficiency of the Mori-Tanaka method is then exploited in the formulation of a coupled two scale computational scheme, while at the level of textile ply the finite element computational homogenization is assumed, the two-point averaging format of the Mori-Tanaka method is applied at the level of yarn to serve as a stress updater in place of another finite element model representing the yarn microstructure as typical of FE2 based multiscale approach. Several numerical simulations are presented to support the proposed modeling methodology.

Mechanical Properties of Porous Structures for Dental Implants: Experimental Study and Computational Homogenization.

A combined experimental and numerical study on titanium porous microstructures intended to interface the bone tissue and the solid homogeneous part of a modern dental implant is presented. A specific class of trabecular geometries is compared to a gyroid structure. Limitations associated with the application of the adopted selective laser melting technology to small microstructures with a pore size of 500 µm are first examined experimentally. The measured effective elastic properties of trabecular structures made of Ti6Al4V material support the computational framework based on homogenization with the difference between the measured and predicted Young's moduli of the Dode Thick structure being less than 5%. In this regard, the extended finite element method is promoted, particularly in light of the complex sheet gyroid studied next. While for plastic material-based structures a close match between experiments and simulations was observed, an order of magnitude difference was encountered for titanium specimens. This calls for further study and we expect to reconcile this inconsistency with the help of computational microtomography.

Modeling Adhesive Anchors in a Discrete Element Framework.

In recent years, post-installed anchors are widely used to connect structural members and to fix appliances to load-bearing elements. A bonded anchor typically denotes a threaded bar placed into a borehole filled with adhesive mortar. The high complexity of the problem, owing to the multiple materials and failure mechanisms involved, requires a numerical support for the experimental investigation. A reliable model able to reproduce a system's short-term behavior is needed before the development of a more complex framework for the subsequent investigation of the lifetime of fasteners subjected to various deterioration processes can commence. The focus of this contribution is the development and validation of such a model for bonded anchors under pure tension load. Compression, modulus, fracture and splitting tests are performed on standard concrete specimens. These serve for the calibration and validation of the concrete constitutive model. The behavior of the adhesive mortar layer is modeled with a stress-slip law, calibrated on a set of confined pull-out tests. The model validation is performed on tests with different configurations comparing load-displacement curves, crack patterns and concrete cone shapes. A model sensitivity analysis and the evaluation of the bond stress and slippage along the anchor complete the study.