Empirically derived Charpy energy to fracture toughness (J-integral) correlations are often used to estimate the fracture toughness of steels from Charpy tests due to the higher testing costs and time associated with direct fracture toughness tests, but analytical insight into these correlations is lacking. Accounting for differences in the strain rates and stress states in these tests to simulate the correct response in both while keeping model complexity and calibration effort manageable presents an obstacle to a numerical approach for this problem. This paper hence establishes a modelling and calibration approach that could be used to contribute mechanics-based insight into the correlations between the Charpy energy, J-integral, yield-to-tensile strength ratio and tensile test fracture elongation. A phenomenological rate-dependent plasticity model coupled with damage and temperature effects is developed by implementing the strain-based modified Mohr–Coulomb damage-softening model with Johnson–Cook thermal softening in a thermodynamically consistent Cowper–Symonds viscoplasticity model. The validity of the modelling framework is shown by its ability to simultaneously model the tensile test, the Charpy V-notch test and the precracked single-edge notched bending test. This is demonstrated for two steels, AH36 and S690QL, capturing the force–displacement responses and the characteristic ductile fracture mechanism of slant fracture in all three tests. Accounting for thermal softening due to adiabatic heating proves to be important for the accurate simulation of ductile Charpy tests involving high impact energies. Capitalising on weak triaxiality dependence in the middle-to-high triaxiality ranges in the given materials and adopting a triaxiality-independent assumption is found to be effective for reducing the damage model complexity while maintaining its ability to simulate the mechanical response in key tests covering an important range of stress states. The importance of the role of the Lode angle in ductile fracture modelling in weakly triaxiality-dependent regimes is further substantiated. Key similarities in the fracture behaviour of the Charpy and single-edge notched bending tests are identified: they span a similar range of stress states over a large range of their response despite the initial notched versus cracked difference—an insight that could be used to reduce the calibration effort of damage mechanics models for these tests, assuming that the key differentiating factors of rate-dependence and adiabatic heating are correctly accounted for.

Ductile fracture in steels relevant to the offshore and maritime industry is often characterized by the occurrence of slant fracture, which is the development of fracture surfaces that are slanted relative to the original surface of the material. The modeling of this phenomenon is important for describing ductility and fracture toughness accurately in ductile fracture simulations. This work uses a consistency model for viscoplasticity with damage softening within a strain-based framework to investigate the effect of variations in strain hardening and Lode dependence on the slant fracture area and impact energy in Charpy tests. The model is first calibrated to uniaxial tensile, single-edge-notched-bending fracture toughness and instrumented Charpy tests performed on an S690QL steel, and then a parametric study varying the strain hardening and the Lode-dependence is performed. It is seen that an increase in the yield-to-tensile-strength ratio (equivalent to a decrease in the strain hardening exponent) leads to a decrease in the impact energy and negligible difference in the percentage slant fracture area when the damage and rate parameters are kept constant. It is found that the Charpy impact energy is not sensitive to the maximum strains in the fracture strain locus and is mainly affected by the minimum strains in the locus. Finally, the rate-dependent consistency plasticity model with a strain-based damage-softening formulation is capable of simulating slant fracture behavior even in cases where the fracture initiation strain is stress-state-independent and constant.

This article describes a dataset used to calibrate a finite element model of a thick circular hollow section (CHS) with varying d/t (diameter to thickness) ratio under cyclic loading which may be used as a computational model validation benchmark by researchers working on similar problems in structural and mechanical engineering. The test data consists of seven cold-formed S335J2H steel CHS tube specimens tested to buckling failure in low-cycle fatigue under a three-point bending arrangement, instrumented with discrete strain gauges, displacement transducers and string potentiometers together with continuous surface deformation fields obtained by two pairs of digital image correlation (DIC) cameras. ‘Half-cycle’ material data from the uniaxial tensile testing of dog-bone coupons is also provided. Comparisons between measured and simulated entities such as midspan forces, moments, displacements and mean curvatures can be obtained with MATLAB processing scripts. Complete ABAQUS model input files are also provided to aid in benchmarking.

High-strength steel beams are known to have less plastic rotation capacity than beams with lower yield strengths. This has been related to the decreased strain-hardening ability of high-strength steels, and various rules and standards for steel structures stipulate maximum limits on the allowable yield-to-tensile strength ratio ((Formula presented.)), which indirectly acts as a measure of strain hardening. While the literature suggests that there is an interdependence between strain hardening ability, yield strength, cross-sectional slenderness and rotation capacity, the presently prescribed limits on (Formula presented.) (e.g. 0.91, 0.94, 0.95) are typically constant for a given material regardless of the other parameters mentioned. This computational study hence investigates how the rotation capacity is simultaneously dependent on yield strength, strain hardening ability and cross-sectional slenderness, and how each parameter affects the relationship between the others. The results show that, with the geometrical aspect kept constant through the use of normalised slenderness parameters, a higher yield strength leads to higher rotation capacity for a given (Formula presented.), while the well-known decrease of rotation capacity with higher (Formula presented.) is confirmed. This suggests the possibility of more efficient use of high-strength steels with high (Formula presented.) when the interdependence of all the variables are accounted for. The results also suggest the importance of accounting for the relative slendernesses of the web and the flange and whether the buckling behaviour is web- or flange-dominated, since a switch between a web- and flange- dominated buckling response could lead to a reverse in the trend between the rotation capacity and the overall cross-sectional slenderness.

Application of high-strength steels in the maritime and offshore industry is currently limited by rules governing the ratio of the yield to tensile strength (the Y/T ratio). To better understand the physical basis for these rules, the nature and extent of the plastic stress/strain field in the vicinity of a stress concentration (a circular hole) in a structure made from high-strength steel are analyzed. This is done through analytical models of the stress field and the extent of plasticity in the vicinity of a hole based on classical methods. These analytical methods are validated through FEA models that are in turn validated by published experimental data. This paper concludes that a high Y/T ratio leads to a lower plastic SCF and a higher local strain in the vicinity of a hole. The extent of the plastic zone is not affected by different values of Y/T ratio for different values of σ_nom/σ_y.

Upper limits on the ratio of the yield strength to the tensile strength (σ_y/σ_u ratio) and lower limits on the fracture elongation ε_f are present in various offshore, maritime and civil engineering rules, standards and specifications for steel structures as a provision for the minimum material ductility and toughness which ensures sufficient structural ductility. In other instances, the design yield stress to be adopted in strength calculations is reduced from its nominal value if the σ_y/σ_u ratio exceeds a certain limit. Such requirements deter the use of high strength steels (nominal σ_y higher than 690 MPa), which inherently have a high σ_y/σ_u ratio. To guide subsequent efforts towards optimised and scientifically grounded σ_y/σ_u limits and wider application of high strength steels, this paper first presents an overview of the current provisions in engineering practice relating to the σ_y/σ_u ratio and structural ductility, and it then discusses the key underlying failure mechanisms to which these ductility requirements are relevant: tensile strain localization, yielding and localization precipitated by stress concentrations, localization of plastic bending hinges and ductile fracture. The reasoning behind the current provisions, the findings of previous research concerning the requirements, and the key potential areas for future research are highlighted.