Search

This paper evaluates the brittle fracture risk for a C–Mn weld in the upper shelf of the brittle-to-ductile transition: the criterion considered is based on a critical stress σth, with the failure probability related to the volume around the crack where the maximum principal stress exceeds σth. The weld shows a complex microstructure with two types of melted zone. SEM observations showed that the main cleavage sites are located in the coarse grain zone, near an inclusion. In test analyses, the material’s heterogeneity in the weld metal is considered to estimate local fields at the weld pass scale.

International audience; In this paper, polycrystalline bi-phased microstructure calculations were performed using finite element (FE) method and compared to mechanical fields measured experimentally using a digital image correlation (DIC) technique. From scanning electron microscopic (SEM) observations and electron back-scattered diffraction (EBSD) crystal orientation measurements performed on an austenitic-ferritic stainless steel test specimen, a quasi-2D numerical polycrystalline microstructure was constructed. Mechanical behaviors of austenitic (FCC) and ferritic (BCC) grains were modeled by two crystal plasticity laws, based on crystallographic slips and dislocation densities. FE mechanical fields were calculated using two types of boundary conditions: the displacements measured by DIC, and the average of the displacements measured by DIC. A better description of intraphase and intragranular strain heterogeneities is obtained using as boundary conditions the displacements measured instead of the average of the displacements measured.

Low cycle fatigue of SAF 2507 duplex stainless steel has been studied at different temperatures in two thermal conditions: unaged (as-received) and aged (100 hours at 475°C). The corresponding cyclic flow stress components, the friction, σF, and the back stresses, σB, have been analyzed in relation to the observed microstructure. σF attains higher values at room temperature than in the intermediate range of temperatures in both thermal conditions due to the influence of the lattice friction stress on screw dislocations in the ferrite phase. The cyclic softening observed at all temperatures in aged samples is caused by the decrease of the back stress σB. The increase of σB in unaged samples at 300°C is consistent with the high dislocation density observed in the ferrite phase ascribed as a typical manifestation of dynamic strain aging.

International audience; This paper focuses on the identification of activated slip systems in low-cycle fatigue (Δεt/2 = 5 × 10−3) in a duplex stainless steel. From electron backscattered diffraction measurements and scanning electron microscopy observations, the slip systems and their associated Schmid factor are analyzed in both constitutive phases. In austenitic grains, one or two slip systems are activated with Schmid factors greater than 0.25. While in the ferritic grains, several slip systems are activated, with a variety of Schmid factors.

The cyclic behavior of type DIN 1.4460 duplex stainless steels in as-received and aged conditions was investigated at room temperature and at 500°C. At room temperature, the aged samples showed embrittlement effects such as loss of toughness and reduction of fatigue life. At 500°C, beneficial effects of the synergy between temperature and strain cycling were observed. It is proposed that at high temperature in the ferrite the strain cycling can decompose the chemical composition fluctuations, promoting a demodulation of the spinodal decomposition formed by aging.

In the framework of an extensive investigation regarding crack initiation and propagation during low-cycle fatigue in Duplex Stainless Steels (DSS), the present study was focused on the fatigue surface damage in a low nitrogen DSS. The Coffin Manson curve is presented for DIN W Nr. 1.4460 DSS and compared with that of a high nitrogen steel. The surface damage has been observed by optical and electronic microscopy and analysed in relation to the underlying dislocation structure by transmission electron microscopy. The results for the present low nitrogen steel DIN W Nr. 1.4460 show that, at Δεt = 0.8 × 10−2, the preferred site for short crack initiation seems to be the intense slip marks formed mostly in the ferrite and less frequently in phase boundary. On the contrary, at Δεt = 1.2 × 10−2, short cracks form mostly in phase boundary.

Various components of nuclear power plants are submitted to very sharp multiaxial thermomechanical loadings, due for instance to the incomplete mixing of flows at different temperatures. As an example, thermal fatigue damage has been detected in auxiliary loops of the primary cooling circuits of Pressurized Water Reactors. In particular, crack networks were observed in in-service pipes submitted to thermomechanical loading resulting from cyclic temperature gradients across the wall-thickness of components in 304 L type austenitic stainless steel. The thermal fatigue behaviour of AISI 304 L type steel has been studied using a specific thermal fatigue test, called Splash, developed in order to reproduce experimentally such thermomechanical biaxial loading in the thickness of parallelepipedic specimens. All tests have been performed at a maximum temperature of 320°C, but with different minimum temperatures. First, the morphological characteristics of the growing networks were analysed, in surface and in depth. Crack initiation is multiple and occurs on sliding lines or at material defects. Crack network stabilization is observed after 400 000 cycles at a temperature of 150°C. The maximum depth is 2.5 mm. Secondly, the stability of the thermal-fatigue cracknetworks previously obtained was investigated under additional isothermal mechanical loading (four-point bend tests). Selection mechanisms of a dominating crack are observed, showing a great influence of shielding effects, branching and tortuous path. Comparison of the dominating crack behaviour with one having a single crack initiated at a notch tip reveals a significant delay effect.

In order to model the mechanical behaviour of an austenitic-ferritic duplex stainless steel thanks to "composite" micromechanical non-linear models, its microstructure was analysed and the mechanical behaviour of each phase was characterised. The microstructural morphology of this steel was studied by selective dissolution of the austenitic phase. The microstructure consists of unconnected austenite islands dispersed in a ferritic matrix. Nano-indentation tests were carried out on each phase. These tests allowed to obtain the hardness and the Young modulus of each phase. A non-linear homogenization approach (secant and incremental formulations) was implemented and the results were compared to the monotonous macroscopic tensile tests carried out at constant strain rate. It allowed us to evaluate the relevance of the non-linear homogeneization models for the description of the elasto-plastic behaviour of the studied duplex stainless steel.

Austenitic-ferritic duplex stainless steel has been subjected to uniaxial and biaxial nonproportional cyclic loading with the same equivalent strain amplitude. The dislocation structures in specimens fatigued to fracture using both types of loadings were studied and compared. Uniaxial cyclic loading, both in austenitic and in ferritic grains, produces simple structures due to activation of predominantly one slip system. Non-proportional cyclic loading results in formation of cell and wall structures and thus in higher stress response of the material.

The early stages of the formation of dislocation microstructures in low strain fatigue are analysed,using three-dimensional discrete dislocation dynamics modelling (DDD). A detailed analysis of the simulated microstructures provide a detailed scheme for the persistent slip band formation, emphasizing the crucial role of cross-slip for both the initial strain spreading inside of the grain and for the subsequent strain localization in the form of slip bands. A new ad-hoc posttreatment tool evaluates the surface roughness as the cycles proceed. Slip markings and their evolutions are analysed, in relation to the dislocation microstructure. This dislocation-based study emphasizes the separate contribution of plastic slip in damage nucleation. A simple 1D dislocation based model for work-hardening in crystal plasticity is proposed. In this model, the forest dislocations are responsible for friction stress (isotropic work-hardening), while dislocation pile-ups and dislocation trapped in Persistent Slip Bands (PSB) produce the back stress (kinematic workhardening). The model is consistent with the stress-strain curves obtained in DDD. It is also consistent with the stress-strain curves experimentally obtained for larger imposed strain amplitudes.

The low-cycle fatigue behavior of a duplex stainless steel, 60 % ferrite - 40 % austenite, is studied under tension-compression/torsion loading at room temperature. The influences of loading direction and loading path are analyzed. It is shown that the duplex stainless steel has an isotropic behavior under tension-torsion loading in monotonic as well as in cyclic conditions. The loading path induces an over-hardening on cyclic hardening of duplex stainless steel, but lower than the one on austenitic stainless steels. The effect of loading history is studied in terms of strain amplitude, mean strain and loading path. It is shown that only histories in strain amplitude and loading path have a small effect on the stabilized stress.

Cyclic tests performed in the temperature range 300773 K on duplex stainless steel DIN 1.4460 show that the cyclic stressstrain behaviour of this steel is strongly temperature dependent. At 300 and 473 K an almost constant peak tensile stress stage, is followed by a slight softening that continues up to failure in the case of 300 K, but by a secondary hardening at 473 K. Pronounced initial cyclic hardening followed by secondary hardening was the main feature of the temperature range between 573 and 723 K. At 773 K, after a weak hardening stage, a strong softening continues up to failure. The mechanical behaviour and the evolution of the microstructure were analysed, and the internal and the effective stresses were studied. It was found that the internal stress is responsible for the strong hardening that occurs in the intermediate temperature range and for the softening at 773 K.

Stress–strain hysteresis loops of a ferritic stainless steel during cyclic straining with increasing strain amplitudes were recorded and the shapes of the hysteresis loops were analyzed. An extra strain around zero stress as a result of magnetoelastic effect was identified. Its effect on the determination of the effective and internal stresses in cyclic straining is discussed.

The low-cycle fatigue behaviour of a cast metal matrix composite has been studied experimentally and numerically. Tension–compression cyclic tests have been carried out in situ in a scanning electron microscope to observe damage mechanisms of a AS10U3NG alloy based composite reinforced with 10 vol.% SiCp. Micro-cracks always initiate on the specimen surface, on the biggest casting defects. Then the small cracks mainly propagate in the eutectic zone by precipitate fractures or by either particle/dendrite or particle/precipitate debondings. One of these cracks becomes the main crack. When the main crack reaches a ‘critical length', it tends to grow in mode I through the dendrites and by particle fracture in front of the crack. As far as the numerical study is concerned, a finite element analysis models the composite behaviour under tension–compression cyclic loading, before crack initiation, until a stabilised cycle is reached. Numerical and experimental stress–strain hysteresis loops are compared. Numerical and experimental cyclic stress–strain curves are in a good agreement.

In constant amplitude cyclic straining the hysteresis loops of an austenitic and a ferritic steel were recorded and their shape was analysed. The generalised statistical theory of the hysteresis loop based on the Masing hypothesis that considers the effective stress and the distribution of the critical internal stresses of the elementary volumes was employed. The second derivative of the hysteresis half-loop yields information on the effective stress and the probability density function of the critical internal stresses and its evolution in cyclic straining and in dependence on the strain amplitude. The changes of the effective modulus are also reported. The total cyclic stress can be separated into the effective and internal stress components. The high effective stress in ferritic steel is connected with the difficult motion of screw dislocations in the b.c.c. structure. The evolution of the probability density function of the austenitic and ferritic steels during cyclic straining is discussed in terms of changes of the internal dislocation arrangement.

The hysteresis loop of austenitic and austenitic–ferritic duplex stainless steels, both with different levels of nitrogen, was analyzed using a generalised statistical theory. The plot of the second derivative yields information on the average effective stress and the probability density function of the critical internal stresses, its evolution in cyclic straining and its dependence on the strain amplitude. The probability density function in duplex steels has two peaks, which correspond to the two-phase structure. The position of these peaks plotted versus the fictive stress corresponds very well to the position of the peaks in the individual phases. The changes of the effective moduli and the evolution of the effective and internal components of the cyclic stress are reported. Nitrogen takes a strong effect on the probability density function of the critical internal stresses in austenitic steel. The suppression of the second peak formed in high amplitude straining is discussed in terms of the appearance of planar dislocation structures, which develop due to the presence of nitrogen.

The internal structure of AISI 316LN austenitic stainless steel cyclically strained at liquid nitrogen temperature has been studied using transmission electron microscopy and electron diffraction. High amplitude cyclic straining promotes the transformation of austenite with face centred cubic (f.c.c.) structure into e-martensite with hexagonal close packed (h.c.p.) structure and α′-martensite with distorted base centred cubic (b.c.c.) structure. Thin plates containing e-martensite were identified in all grains. α′-martensite nucleates at the intersection of the plates in grains with two or more systems of plates and can grow in the bands. The orientation of transformed phases follows the Shoji–Nichiyama and Kurdjumov–Sachs relations. Mechanisms of low temperature cyclic straining are discussed.

The hysteresis loop of a 316 austenitic stainless steel was recorded and analysed in order to separate the contributions of the effective and internal stresses. The difficulties of application of the Kuhlmann-Wilsdorf-Laird method are discussed and a new method based on the statistical approach to analysis of the hysteresis loop (SAP method) is proposed. In contrast to the Kuhlmann-Wilsdorf-Laird method, this method allows evaluation of the effective elastic moduli and the effective stresses in tension and in compression without ambiguity. The SAP method was applied to cyclic loading of 316 austenitic stainless steel and the effective elastic moduli and effective and internal stress components of the cyclic stress in cyclic loading and their dependence on the number of cycles and on the applied strain amplitude was studied. The effective elastic moduli decrease appreciably with the strain amplitude. The effective stress represents an important fraction of the total stress (approximately 30%) and for most of the life it is constant. The hardening-softening curves and the cyclic stress-strain curve are determined mostly by the internal stress.

The local approach method is used to calculate the fatigue crack initiation/early crack growth lives (N i ) in high strength structural steel weldments. Weld-toe geometries, welding residual stresses and HAZ (heat affected zone) cyclic mechanical properties are taken into account in the N i estimation procedure. Fatigue crack initiation lives are calculated from either a Basquin type or a Manson-Coffin type equation. The local (HAZ) stress and strain amplitudes and the local mean stress are determined from an analysis based on the Neuber rule and the Molski-Glinka energy approach. The accuracy of the different methods is evaluated and discussed. Finally the previous methods are used with HAZ cyclic mechanical properties estimated from hardness measurements.

International audience; As other metallic materials, in low-cycle fatigue, duplex stainless steels (DSS) exhibit a cyclic hardening, followed by a cyclic softening, before stabilization of the stress. In order to simulate the cyclic hardening/softening curves in low-cycle fatigue of an austenitic–ferritic or duplex stainless steel (DSS), a new polycrystalline model is proposed. The polycrystalline model developed by (Cailletaud, 1992) and (Pilvin, 1990) and modified by Hoc and Forest (2001) was previously extended in Evrard et al. (2008) in order to take into account the bi-phased character of the DSS. This model correctly accounts for the cyclic hardening, but it is not able to simulate the cyclic softening, consequently, stresses at the stabilized state are overestimated. TEM observations of the dislocation structures built during a cyclic uniaxial tension/compression test show that, during the cyclic hardening, planar arrangements are observed in austenitic grains and no significant evolution is observed during the subsequent cyclic softening and stabilization stage. On the contrary, in ferritic grains, dislocations are homogeneously distributed during cyclic hardening, and the microstructure evolves during the subsequent cyclic softening and stabilization stage. Dislocation structures build progressively, consisting of hard zones or walls, separated by soft zones or channels. We propose to model the cyclic softening through dislocation structure evolution within ferritic grains. The single crystal law used by Hoc and Forest (2001) is modified in order to take into account the heterogeneous distribution of dislocations in the ferrite. Numerical simulations are compared with experimental data. A good agreement is observed between experimental and calculated hardening/softening curves and stabilized hysteresis loops.

This second part aims to propose an extension of the bi-phased polycrystalline model shown in a previous contribution [1], in order to simulate the cyclic softening observed in an austenitic-ferritic stainless steel under uniaxial and non-proportional biaxial cyclic loading [2]. Assuming that the cyclic softening for uniaxial and biaxial non-proportional cyclic loading is due to a rearrangement of dislocations in hard (rich in dislocations) and soft (poor in dislocations) zones in ferritic grains [3], the ferritic single crystal law is modified. The new model parameters are identified from TEM observations [3] and by inverse procedure. Finally, the new model is validated for other cyclic loading paths. A good agreement is observed between simulation and experimental data.

This study analyses the plastic flow behaviour of types 316L and 316LN austenitic stainless steels. The following features are shown. 1. (i) The yield stresses vary quasi-linearly with the nitrogen content. This is attributed to the linear variation in the thermal component with the interstitial concentration, according to the Friedel model. 2. (ii) The monotonic work hardening of these steels can be modelled by the modified Ludwik model proposed by Ludwigson: σ = ϰ1en1 + Δ, which takes into account the positive deviation from the Ludwik relation, which appears below a transient strain eL. This transition corresponds to a change in the dislocation slip mode, essentially planar below eL and multiple above eL. 3. (iii) The influence of nitrogen on Ludwigson's modelling parameters is adequately explained by the fact that the nitrogen favours planar dislocation slip in the austenitic stainless steels.

International audience; The present paper proposes to modify the single crystal law [Hoc, T., Forest, S., 2001. Polycrystal modelling of IF-TI steel under complex loading path. Int. J. Plast. 27, 65–85] used in polycrystalline model developed by [Pilvin, P., 1990. Approches multiéchelles pour la prévision du comportement anélastique des métaux. Ph.D. Thesis, Université Pierre et Marie Curie; Cailletaud, G., 1992. A micromechanical approach to inelastic behavior of metals. Int. J. Plast. 8, 55–73], in order to account for the cyclic softening observed in some metallic materials during fully reversed low-cycle fatigue tests under total strain control. Experimental observations [Mayama, T., Sasaki, K., 2006. Investigation of subsequent viscoplastic deformation of austenitic stainless steel subjected to cyclic preloading. Int. J. Plast. 22, 374–390] show that the cyclic softening is partially due to a rearrangement of dislocations in hard and soft zones within the grains during cyclic straining. We propose thus to modify the initial single crystal law by including a composite model of hard and soft zones on each slip system within a grain. The softening is modeled through the evolution of soft zones. The model is applied to a FCC polycrystal. The self-consistent scheme is used in order to obtain the polycrystalline macroscopic behavior. The results show that the new model correctly describes an initial cyclic hardening, a subsequent cyclic softening and finally a stabilized state.

The aim of the present paper is to propose a polycrystalline approach in order to model the elastic–plastic behavior of an austenitic–ferritic stainless steel. In order to take into account the specific character of the steel, the multi-scale polycrystalline approach proposed by Cailletaud–Pilvin [Pilvin, P., 1990. Approches multiechelles pour la prevision du comportement anelastique des metaux, PhD Thesis, Universite Pierre et Marie Curie; Cailletaud, G., 1992. A micromechanical approach to inelastic behavior of metals. International Journal of Plasticity 8, 55–73; Cailletaud, G., Pilvin, P., 1994. Utilisation des modeles polycristallins pour lecalcul par elements finis. Revue Europeenne des Elements Finis 3 (4), 515–541] is extended to bi-phased material. In particular, two interaction laws and two local behaviors, based on the crystallographic slip and the dislocation density evolution, are simultaneously considered. After an identification of the model parameters on simple tests (monotonous tension, tension-compression), we propose an evaluation of the predictive capabilities of the multi-scale approach for non-proportional loading paths. A good agreement is observed between simulation and experimental data.

Strain field measurements by digital image correlation today offer new possibilities for analysing the mechanical behaviour of materials in situ during mechanical tests. The originality of the present study is to use this technique on the micro-structural scale, in order to understand and to obtain quantitative values of the fatigue surface damage in a two-phased alloy. In this paper, low-cycle fatigue damage micromechanisms in an austenitic-ferritic stainless steel are studied. Surface damage is observed in real time, with an in situ microscopic device, during a low-cycle fatigue test performed at room temperature. Surface displacement and strain fields are calculated using digital image correlation from images taken during cycling. A detailed analysis of optical images and strain fields measured enables us to follow precisely the evolution of surface strain fields and the damage micromechanisms. Firstly, strain heterogeneities are observed in austenitic grains. Initially, the austenitic phase accommodates the cyclic plastic strain and is then followed by the ferritic phase. Microcrack initiation takes place at the ferrite/ferrite grain boundaries. Microcracks propagate to the neighbouring austenitic grains following the slip markings. Displacement and strain gradients indicate probable microcrack initiation sites.

The cyclic hardening–softening response, the cyclic stress–strain curve and the substructure evolution of a high nitrogen duplex stainless steel S32750 have been evaluated and the results compared with reference to low and medium nitrogen duplex stainless steels, S32205 and S32900 grades, respectively. The beneficial effects of nitrogen on the cyclic properties of most modern alloys have been analyzed in terms of the flow stress components, i.e. the internal and the effective stress.

Under fatigue loading, the number of cycles to failure and its associated scatter increase when the loading level decreases. The High-Cycle Fatigue (HCF) regime is thus characterized by a large scatter in the number of cycles to failure [1]. Cracks initiation represents an important part of the lifetime of the structures. A stochastic method is used to study the fatigue crack initiation prediction in the 316L austenitic stainless steel. The present work proposes to show that this scatter can be attributed to the random orientation of individual grains, which influences the crack initiation localization. The stresses in grains are determined by finite element computations (FEM [2]), using a configuration representative of a polycrystalline aggregate. This approach takes into account the crystallographic orientations of the grains in the aggregate as well as the deformation incompatibilities between neighbouring grains due to crystalline anisotropic elasticity and elasticplasticity [3]. Then, the scatter of the number of cycles to crack initiation is derived from the FEM stress fields using two fatigue crack initiation criteria: an usual one, Mura’s criterion [4] and a more recent one [5], based on Discrete Dislocation Dynamics (DDD) simulations and taking into account plastic slips, cross slip and stress tensor components.

The present work studies the effect of the microstructure and of the stress gradient on the early fatigue crack growth in a two-phase austenitic-ferritic stainless steel. Fatigue tests were performed on two geometries of double edge notched specimens. Direct optical observations of the surface are used to measure the surface crack growth rates. The results are presented and the effects of stress and strain gradients on the crack propagation kinetics are discussed.

Dislocation structures developed in the austenitic-phase of SAF 2507 duplex stainless steel during fatigue at total strain ranges of 0.8% and 1.2% have been investigated. According to this research, at total strain ranges higher than 0.8%, persistent slip bands (PSBs) can be formed within twins in the austenitic grains. Depending on the size of the twins, the ladder-like structure associated with PSBs can occupy the entire twin or part of them. Besides, a zone affected by the constraint of PSB at grain boundary could also be disclosed. Observations of the fracture surface for the highest total strain range have revealed the presence of intergranular secondary cracks, which were correlated with the PSB structure developed in narrow twins in the bulk of the material.

The statistical approach method was applied to the analysis of the hysteresis loops in low-cycle fatigue of stainless steels with different crystalline structures. This method, based on Masing hypotheses, allows to determine the effective stress, the effective elastic modulus and the ‘probability density function’ of the internal critical stresses. The dependence of these quantities on the applied strain amplitude was investigated in austenitic, ferritic and duplex stainless steels. The ‘probability density function’ in austenitic stainless steel exhibits a single peak. It changes appreciably with strain amplitude. This dependence on the strain amplitude was less important in the ferritic stainless steel. A double peak form was found on the ‘probability density function’ in duplex stainless steel, which is believed to correspond to the plastic deformations of the two phases, γ and α, respectively.

Bauschinger effect in an AISI 316L austenitic stainless steel at room temperature was studied as a function of the plastic prestrain. Typical uni-axial Bauschinger tests in tension–compression and in compression–tension were carried out. Internal and effective stresses were evaluated from the experimental stress–strain curves using the method proposed by Kuhlmann-Wilsdorf and Laird. Four constitutive relations derived from a thermodynamical formulation were identified using an inverse identification method. The modelled stress–strain curves were then compared to the experimental ones. Bauschinger effect was characterised using Bauschinger indicators proposed in the literature in terms of strain, stress and energy, respectively. These indicators were evaluated from the experimental stress–strain curves on the one hand, from the modelled ones on the other hand. Their variations as functions of prestrain were then compared. The best constitutive relation uses two non-linear kinematic work-hardening terms.

A thermal fatigue device––called SPLASH––similar to the facility developed by Marsh [Fatigue crack initiation and propagation in stainless steels subjected to thermal cycling, International Conference on Mechanical Behaviour and Nuclear Applications of Stainless Steels at Elevated Temperature, 1981] has been built in CEA/SRMA in 1985. Since then, it was used mostly on austenitic stainless steels to assess the initiation and growth of thermal fatigue crack networks. In 1998, a leak appeared in an auxiliary loop of the primary circuit of a pressurized water nuclear plant in Civaux (France). Thermal fatigue was suspected and studies began on AISI 304 L type austenitic stainless steel. They were eventually compared to results obtained earlier on AISI 316 L(N). First, the initiation conditions were determined and the damage before initiation was qualitatively observed. Then, some crack networks parameters were chosen and quantitatively determined by image analysis. This part of the study was done at the surface, during crack growth, and at the end of the tests, in depth. Finally, the stability of the crack networks obtained by thermal fatigue was tested under isothermal load controlled four point bending fatigue test, and some conclusions were drawn on the mechanisms of propagating crack selection.

Unidirectional and multilayer composites are usually subjected to bending, and often fail in fatigue by interlaminar shearing. This work studies new experimental conditions in fatigue, using a three-point-bending test with predominant shearing, while tensile/compressive stresses are significant. The experimental device geometry is validated under monotonic loading by comparison between finite element results and experimental observations. The deflection amplitude vs. fatigue life curve is used as a pseudo-Wöhler curve. An approximate drawing of maximum deformation vs. fatigue life curve is proposed using Fournier's model. Experimental results in fatigue are well accounted for by this model.

Tensile and low-cycle fatigue tests were performed on a 316LN austenitic stainless steel at 300 and 77 K. The tensile and low-cycle fatigue properties were obtained and analysed in terms of influence of temperature on the plastic deformation process, and particularly on the strain-induced martensite formation. The martensite content was measured by a magnetic-at-saturation method. No martensite was detected at 300 K. On the contrary, strain-induced martensite transformation is responsible for the higher tensile elongation at 77 K and for the secondary hardening observed on softening-hardening curves in low-cycle fatigue at 77 K. The induced martensite content in tensile tests is a function of the strain according to Angel's model, and in low-cycle fatigue it is a function of the strain level and of the accumulated plastic strain.