Design of thick bonded joints for the strengthening of metallic structures – assessment of thermomechanical behavior and fatigue performance

Structural bonding offers promising solutions for the repair and strengthening of metallic structures (Lepretre et al., 2018; Chataigner et al., 2018), provided that issues related to implementation, design, and long-term durability are properly addressed. Structural bonding consists in joining different components using a polymer adhesive, generally thermosetting, which allows the association of materials with different mechanical properties without resorting to invasive processes on the existing structure. Several application techniques exist (manual application with or without reaction catalysis, infusion processes, etc.), and various types of adhesives can be used.

Bonded solutions are particularly well suited for fatigue strengthening, as they limit damage associated with drilling or thermal treatments required when adding reinforcing elements. This is especially relevant in the context of aging metallic infrastructures, such as bridges, and in strategies aiming to extend the service life or reuse existing structural components through strengthening operations.

For metallic structures, two main families of bonded strengthening solutions can be identified:

• Thin bonded joints, with bondline thicknesses below 2 mm, typically using relatively stiff structural adhesives such as epoxies or acrylates. The objective is to ensure mechanical continuity between the substrates while minimizing the interface thickness. This approach is commonly used in externally bonded composite reinforcements.
• Thick bonded joints, with bondline thicknesses greater than 10 mm, generally based on more compliant polymers such as polyurethanes. In this case, the bonded assembly behaves as a sandwich structure made of two stiff substrates separated by a thick adhesive layer. The joint thickness provides a lever arm that directly affects the global mechanical response of the system. Such solutions are used, for example, in offshore applications involving resin injection between steel plates, and may also be considered for strengthening metallic bridge decks.

Most existing studies have focused on the behavior of thin bonded joints. This has led to the development of well-established design approaches and to initial investigations of mechanical and environmental durability. In particular, fracture mechanics-based methods have been developed for thin bonded joints (Sourisseau, 2022) and later extended to account for environmental effects (Fawaz, 2025). However, these approaches rely on properties averaged over the bond-line thickness, an assumption that is reasonable for thin joints but questionable for thick bonded joints. In addition, relatively few studies have addressed the mechanical behavior of thick bonded joints (Savio Calabrese et al., 2025), especially under cyclic loading and variable temperature conditions.

Moreover, the polymers used in thick bonded joints exhibit a strongly non-linear and viscoelastic behavior, which is sensitive to temperature and may lead to:

• progressive stress redistribution over time (creep and stress relaxation);
• energy dissipation under cyclic loading;
• local selfheating, which may alter the mechanical properties of the adhesive.

These effects may significantly influence the fatigue performance and long-term durability of strengthening solutions based on thick bonded joints, yet they are still rarely accounted for in current design approaches.

The objective of the proposed PhD work is to improve the understanding of the mechanical behavior of thick bonded joints, considering the effects of temperature and cyclic loading, and to assess the conditions under which concepts developed for thin bonded joints—particularly fracture mechanics-based approaches (Sourisseau, 2022; Fawaz, 2025)—can be extended to thick bonded joints.

The aim is not limited to the analysis of failure mechanisms, but also to link local mechanisms, whether volumetric or interfacial, to the long-term stability of the mechanical efficiency of the strengthening system, with a particular focus on fatigue performance.

Proposed PhD Work plan :

After a literature review, the PhD work will be structured as follows:

Step I – Characterization of adhesives for thick bonded joints

A limited number of structural adhesives representative of thick bonded joint applications (polyurethanes) will be selected. Their mechanical and viscoelastic properties will be characterized, including:

• the influence of temperature (three temperature levels);
• loading and unloading cycles to identify nonlinear behavior;
• an initial assessment of energy dissipation and selfheating under cyclic loading (Katunin et al., 2025).

These tests will be used to define a rheological model suitable for thick bonded joints.

Step II – Experimental investigation at the thick bonded joint scale

An experimental program on steel–steel thick bonded joints will be defined to:

• identify dominant damage mechanisms (volumetric versus interfacial);
• evaluate the applicability of design concepts developed for thin bonded joints to thick bonded joints at a given temperature;
• study the influence of joint thickness and temperature;
• analyze the evolution of mechanical properties under cyclic loading.

Step III – Finite element modeling of thick bonded joints

Finite element models will be developed to describe the mechanical behavior of steel–steel thick bonded joints. The models will account for the viscoelastic behavior of the adhesive and for temperature effects. Numerical results will be compared with experimental observations at both material and joint scales to analyze load transfer and damage mechanisms. When relevant, fracture mechanics-based approaches developed for thin bonded joints will be explored for thick joints.

Step IV – Qualification methodology on representative structural elements

Experimental tests will be carried out on structural elements representative of configurations encountered in practice. These tests will be used to:

• quantify the benefit of strengthening in terms of stress redistribution and fatigue performance (Teixeira de Freitas et al., 2013);
• analyze the evolution of this benefit over time under thermomechanical loading;
• compare experimental results with predictions from finite element models.

Expected outcomes :

The PhD work will:

• clarify the conditions under which design approaches developed for thin bonded joints can be applied to thick bonded joints;
• assess the influence of joint thickness, temperature, and time on the mechanical behavior and fatigue performance of thick bonded joints;
• improve the understanding of nonlinear behavior, energy dissipation, and damage mechanisms in these systems;
• provide a basis for qualification procedures adapted to strengthening solutions using thick bonded joints.

References :

Bureau Veritas (2015). Adhesive joints and patch repair, Guidance Note NI 613 DT R00 E.

Chataigner S., Benzarti K., Foret G., Caron J.F., Gemignani G., Brugiolo M., Calderon I., Pinero I., Birtel V., Lehmann F. (2018). Design and evaluation of an externally bonded CFRP reinforcement for the fatigue reinforcement of old steel structures, Engineering Structures, Vol. 177, pp.556-565.

El Andaloussi H. A., Mouton L., Ahmad F.S., Maherault-Mougin S., Paboeuf S., Errotabehere X. (2019), World first fatigue S-N curve for bonded reinforcements for FPSO application, Proceedings of OMAE 2019, Glasgow, Scotland, June 2019.

Fawaz A. (2025), Etude de l’évolution des lois cohésives d’interface en mode II pour un assemblage collé en milieu marin sous charge, Phd Thesis, Université Gustave Eiffel.

Katunin A., Rogala T., Amraei J., Wachla D., Bilewicz M., Krzeminski L., Reis P. N. B. (2025), Fatigue response and fracture mechanims of polymer matrix composites under dominance of the self-heating effect, Composites Structures, Vol. 365, 119207.

Lepretre E., Chataigner S., Dieng L., Gaillet L. (2018). Fatigue strengthening of cracked old steel plates with CFRP laminates, Composite Structures, Vol. 174, pp. 421-432.

Savio Calabrese A., Vassilopoulos A.P. (2025), On the fatigue behavior of thin and thick adhesively bonded composite joints, International Journal of fatigue, Vol.199, 109065.

Sourisseau Q. (2022), Evaluation de stratégies de dimensionnement de renforcements composites collés sur structures métalliques offshore – Strength Bond Offshore, phd Thesis, Nantes Université-IFSTTAR.

Teixeira de Freitas S., Kolstein H., Bijlaard F. (2013), Fatigue behavior of bonded and sandwich systems for strengthening orthotropic bridge decks, Composite Structures, Vol. 97, pp. 117-128.

The PhD project will be carried out within the Structures Métalliques et à Câbles (SMC-MAST) laboratory at Université Gustave Eiffel (Nantes campus). The laboratory has experimental facilities for the mechanical and fatigue characterization of steel–steel bonded joints, including mechanical testing machines, bending test setups, and a thermal chamber. Various instrumentation techniques are available (strain gauges, laser measurements, LVDTs, load cells, and digital image correlation). With the support of the Structure et Instrumentation Intégrée (SII-COSYS) laboratory, additional fiber optic instrumentation will also be implemented.

The laboratory also provides computing resources and Abaqus and MSC Marc Mentat licenses for finite element modeling. The project will be conducted within a scientific environment with established expertise in structural bonding, fatigue testing, and modeling of bonded joints.

The PhD project requires the following prerequisites:

• solid background in structural mechanics and strength of materials;
• knowledge of polymer mechanics, particularly non-linear and viscoelastic behavior;
• knowledge of fatigue of structures and fracture mechanics;
• knowledge of adhesive bonding (structural bonding);
• strong interest in experimental work, including specimen manufacturing, instrumentation, testing, and data analysis;
• good knowledge of finite element (FE) modeling, preferably using Abaqus©, required for the development and exploitation of thermomechanical models.

These prerequisites are consistent with the content of the PhD project, which combines experimental investigations, mechanical analysis, and FE modeling.