The experimental data of Ref prove that

The experimental data of Ref. [32] prove that the yield point is particularly strongly dependent on hydrogen concentration. A linearly elastic material may be logically assumed to have a free surface area that, for small strains, is proportional to the deformation itself. Consequently, the α and β coefficients must linearly depend on deformation. Then for small deformations we may adopt a linear dependence for the coefficient ratio:
The latter relationship depends the change in the material properties during hydrogen redistribution in it, allowing to obtain a good approximation of the experimental data. Fig. 3 shows the dependence of the maximum tensile stresses on the initial diffusely mobile hydrogen concentration for steel. The calculated dependence was compared to the experimental data. This data was obtained in Ref. [33] for the AISI 4135 steel and is represented by the squares in Fig. 3. The proposed approach was used to study the effect of hydrogen on pipeline wall stresses [34] and on fatigue strength of metals [35,36].
Discussion of the results The discrepancy between the theoretical and the experimental data (see Fig. 3) observed for a low initial concentration of diffusely mobile hydrogen can be attributed to the fracture mechanisms that are not connected to hydrogen effect and, consequently, not described by the model. The rest of the experimental points show a good agreement with the curve, which proves that the model is adequate for describing the examined processes. As this model describes material fracture, it Rapalink-1 does not make any assumptions about microcracks existing in the material or of a specific concentration or orientation of dislocations [8]. This approach is also different from modeling hydrogen embrittlement by introducing a crack resistance parameter [15]. The HELP model [7] used a physical mechanism of hydrogen effect that manifests itself as changes in the local mechanical properties of a metal but only for quantitative ratios of about 1 : 1 between hydrogen particles and metal atoms. It proves impossible to obtain these mean hydrogen concentrations to find the specific parameter values of the constitutive equation, and to subsequently set an experiment or carry out a calculation based on the physical mechanisms of the interaction between hydrogen and metal. Such an experiment or a calculation is actually hard to imagine from a physical standpoint, as solid-state hydrogen has a lattice constant that is 1.5 times higher than the corresponding value for most metals. The main advantage of the proposed constitutive equations and two-component model equations is that spindle apparatus can be applied at a macrolevel. The micromechanisms of hydrogen effect were included into the rheological model. The parameters α, β, EH, k0 and k1 must be determined within the scope of macrovalues, such as experimental stress–strain state diagrams. Therefore, hydrogen concentrations in materials and its volume distributions must be determined accurately in order to find the characteristics of a rheological model. This involves certain difficulties, and the majority of the researchers find these parameters indirectly by measuring cathodic current and charging time. As a result, the obtained data is unsuitable for determining model parameters, as hydrogen is in this case located near sample surface. For example, there is as yet no unique relationship established between ultimate tensile strength and hydrogen-charging time [37]. A descending part present on the σ(ɛ) curve of a hydrogen-containing material indicates the instability of the material under stress. The failure under an actual load will occur when the point of the maximum stress is reached in the respective curve. This point can be interpreted as the ultimate tensile strength of the material due to its hydrogen saturation.
Conclusions
Acknowledgment The study was sponsored by a Russian Science Fund Grant (project no. 15-19-00091).