Recent advancements at TU Delft have investigated the potential of geometrically enhanced profiles to improve the performance of concrete-to-concrete interfaces, with one study focusing on the tensile behavior of single-tab Strain Hardening Cementitious Composites (SHCC)-to-SHCC interfaces through experimental testing. SHCC, an advanced fiber-reinforced cement-based material, is known for its distributed cracking and strain-hardening capabilities, offering superior ductility and crack control compared to traditional concrete. Building on this foundation, this thesis models the tensile behavior of geometrically profiled SHCC-to-SHCC interfaces using a continuum smeared cracking model in Abaqus and a lattice discrete cracking model.
The research aims to assess the predictive capability of these numerical models in capturing the behavior of geometrically profiled SHCC-to-SHCC interfaces. It focuses on understanding how interface parameters, such as strength and geometric profiles, influence tensile performance, particularly in enhancing ductility, controlling fracture response, and shifting failure modes from brittle interface failure to ductile SHCC material failure.

To evaluate the ability of numerical models to represent the tensile behavior of geometrically profiled SHCC-to-SHCC interfaces and assess the benefits of geometric enhancements for improving tensile performance, a systematic methodology was adopted. This included an analytical analysis to identify underlying failure mechanisms and develop a simplified model for quantifying their force-displacement response. A continuum smeared cracking model in Abaqus was employed to replicate experimental behaviors and investigate the influence of interface parameters through a parametric study. Finally, a lattice model was implemented to capture fracture responses in detail, enabling an in-depth analysis of the effects of interface strength and interface geometry on strength, ductility and fracture response.

The numerical models provided insights into the tensile behavior of geometrically profiled SHCC-to-SHCC interfaces, revealing the influence of interface parameters and geometric characteristics on failure mechanisms. The continuum smeared cracking model demonstrated the ability to simulate various failure mechanisms by adjusting interface parameters, but it underestimated SHCC’s strain-hardening behavior during local material-dominated failure due to the delayed introduction of damage. Conversely, the lattice discrete cracking model effectively captured distributed cracking patterns but exhibited overly brittle responses in interface-dominated failure mechanisms due to assumptions of brittle interface elements and the neglect of frictional resistance. Parametric studies confirmed that changing interface geometry and optimizing interface strength could shift failure modes from brittle, interface-dominated failure to ductile SHCC material-dominated failure.

This study highlights the capabilities and limitations of numerical models in optimizing the tensile behavior of geometrically profiled SHCC-to-SHCC interfaces. The continuum smeared cracking model in Abaqus identified critical parameters, such as tensile strength and fracture energy, influencing failure mechanisms. However, it underestimated SHCC hardening during tab failure due to delayed damage initiation, which should occur immediately after the elastic limit to accurately capture distributed cracking behavior.
The lattice discrete cracking model effectively simulated distributed cracking but exhibited overly brittle responses in interface-dominated failure mechanisms. This limitation stemmed from brittle interface assumptions and the omission of friction, resulting in a significant underestimation of displacement capacity.

Parametric studies using the analytical and lattice models revealed that modifying the interface geometry significantly reduces the interface strength required for ductile failure. For instance, the analytical model demonstrated that adjusting the geometry reduced the interface strength needed to transition from interface-dominated to material-dominated failure from 20% to 7%. In the lattice model, at low interface strengths (≤ 25%), a change in geometry of 1% enhanced SHCC damage by 3% to 6%. At higher strengths (≥ 30%), interface-dominated failure mechanisms diminished, with full SHCC material activation observed beyond 150% strength, ensuring global rather than localized failure.

These findings confirm that optimizing interface strength and geometry enables the transition from brittle, interface-driven failure to ductile, material-driven failure. Nonetheless, the models' limitations must be addressed, particularly the insufficient representation of strength-geometry-friction interactions in the analytical model and the lack of interface ductility and friction in the lattice model. Despite these challenges, the results provide a valuable foundation for refining numerical models and guiding experimental validation to optimize SHCC-to-SHCC interfaces.

The concrete bridges constructed in the 1960s and 1970s lack shear reinforcement, rendering them non-compliant with current design regulations. To potentially enhance their shear capacity, the proposed approach involves applying a layer of ultrahigh-performance fiber-reinforced concrete (UHPFRC) on both sides of the normal strength concrete (NSC) beams supporting the bridge deck. A crucial aspect of this method is ensuring a high-quality connection between the UHPFRC layer and the NSC beam to transfer shear stresses effectively. Therefore, the report explores the feasibility of employing active infrared thermography (IRT) as a non-destructive testing (NDT) method to identify delaminated areas within this interface.
The primary objective of this report is to answer the question: "Is it possible to detect and quantify delaminated areas, within the connection between a NSC beam and a layer of UHPFRC of a certain thickness in an objective way, with data gathered by an infrared camera in a lab environment?"
The report outlines the chosen approach for detecting delaminated areas, establishes a standardized testing methodology, and formulates an analytical model to process the data gathered during experiments conducted based on the established standards. The data collection process involves utilizing an infrared camera to measure the temperature of every pixel in each frame of a recording.
From literature review and a qualitative comparison of different methods to detect delaminated areas, the signal-to-noise ratio (SNR) method is chosen in this report. The SNR method compares the temperature in a specific area to that of a sound area while considering the standard deviation of temperature in sound areas, accounting for environmental factors.
The analytical model developed in the report identifies delaminated areas when infrared camera recordings are available. To ensure objectivity, a standardized approach for heating and cooling the specimens is established. Heating is achieved with a heat flux of 1750 W/m² on the UHPFRC layer's surface for 1500 seconds, followed by recording the cooling process with an infrared camera. The analytical model selects the optimal frame for analysis based on this data together with data gathered from a finite element analysis performed in the software COMSOL Multiphysics 5.6.
The report distinguishes between two situations: one where the location of a sound area is known, and another where it is not. The analytical model produces promising results for both scenarios, with simulated delaminated areas visible in the test samples. However, in cases where the location of sound areas is unknown, the model may underestimate the size or extent of delamination, especially in nearly fully delaminated structures. Further research is recommended to assess the accuracy and applicability of the method in situations where sound area locations are unknown and to determine the model's minimum detectable delaminated area.