Flow and fracture of austenitic stainless steels at cryogenic temperatures

Abstract

In this paper, we have characterized the microstructural evolution and the plastic flow and fracture behaviours of AISI 304L and AISI 316LN stainless steel grades at liquid nitrogen temperature (77 K) and at liquid helium temperature (4 K). Uninterrupted tensile experiments, where the sample is continuously deformed under quasi-static loading conditions until fracture, have been carried out with a Single-Section Sample to obtain the stress-strain characteristics of the two grades. Interrupted tensile experiments, in which the sample is unloaded before fracture, have been performed with a novel Double-Section Sample to later characterize the strain-induced martensitic transformation at different levels of deformation. The content of martensite has been determined post-mortem, using magnetic induction, electron backscatter diffraction and quantitative light optical micrography. The results obtained with the three methods show quantitative agreement, and reveal that the martensitic transformation in AISI 304L occurs faster and to a greater extent than in AISI 316LN both at 77 K and at 4 K. To the authors' knowledge, in this paper we provide the first experimental results for the evolution of the content of strain-induced martensite in AISI 304L and AISI 316LN samples tested at liquid helium temperature. In addition, the experimental data for the evolution of the martensite volume fraction with the strain have been used to identify the temperature-dependent parameters of the martensitic transformation kinetic models proposed by Olson and Cohen (1975) and Garion and Skoczeń (2002). Moreover, Mode I fracture tests with fatigue-precracked Compact Samples have been carried out to determine the fracture properties of the two investigated materials using the "resistance curve procedure" (ASTM-E1820-20a, 2020). The crack-growth resistance curves have been obtained with four different methods here referred to as ASTM Compliance Method, W-N Compliance Method, Modified W-N Compliance Method and ASTM Normalization Method, which is an original methodological contribution of this paper. While the four approaches yield similar results for the fracture toughness, only the W-N Compliance Method and the Modified W-N Compliance Method, the latter being proposed in this paper, fulfil all the requirements of the standard ASTM-E1820-20a (2020) so that the calculated fracture toughness can be accepted as a material property. The comparison of results for both materials and testing temperatures shows that the AISI 316LN displays higher fracture toughness than the AISI 304L. Moreover, post-mortem microstructural analysis of the Compact Samples near the fracture surface has revealed that the content of martensite is greater in AISI 304L than in AISI 316LN. Furthermore, for AISI 304L more martensite is formed in the sample tested at 77 K because the plastic deformation near the crack is greater than at 4 K.

Tensile Tests at 77 K and 4 K on 316L Stainless Steel Welded Plates

“…This trend is consistent with previous studies on AISI 316L steel (base metal) [9]. In addition, serrated yielding and serrated flow stresses were observed in tests conducted at 4 K, which is consistent with previous studies on AISI 316L [10] and is a result of changes in dislocation character, sometimes referred to as low temperature plastic instabilities [11]. As temperature decreased, all welds showed a diminution in total elongation and reduction of area, which is consistent with work by Tobler et al [12].…”

Modeling strain-induced martensitic transformation in austenitic stainless steels subjected to cryogenic temperatures using incremental mean-field homogenization schemes

“…The constitutive model is calibrated to describe the mechanical behavior of two different austenitic steel grades that are widely used in superconducting magnet system applications: AISI 304L and AISI 316LN. The experiments performed by Fernández-Pisón et al (2021) are the basis for the identification of the constitutive parameters. The data set includes uniaxial tensile stress-strain curves and the corresponding evolution of the volume fraction of martensite with the strain for the two aforementioned grades at room temperature (300 K), liquid nitrogen temperature (77 K), and liquid helium temperature (4 K); see Figs.…”

“…For these two material parameters, identical values for the two individual phases are assumed for the two grades over the whole temperature range analyzed in this work. For the rest of the parameters, an explicit temperatureevolution function is considered, and the coefficients of the function are determined by best-fit operation with the aforementioned experimental data presented in Fernández-Pisón et al (2021). The idea is to cover the entire analyzed temperature range, unlike conventional approaches that independently determine the material parameters at a certain temperature.…”

“…( 27), which is similar to the one used by Hosseini et al (2015) for a Chaboche kinematic hardening model. Moreover, the same functions proposed by Fernández-Pisón et al (2021), which are based on the experimental data provided by Olson and Cohen (1975), are adopted here with regard to the material parameters involved in the kinetics of martensitic transformation, as shown in Eq. ( 28).…”

“…The 𝛾-𝛼 ′ transformation may lead to a considerable evolution of material properties (e.g., strong hardening) and to ferromagnetic behavior of the bi-phase material, which is problematic for the above-mentioned application (i.e., for superconducting magnet systems). Moreover, the transformation rate increases with decreasing temperature, so a nearly fully or fully transformed 𝛼 ′ -martensitic material may be obtained at low temperatures and high levels of deformation (Fernández-Pisón et al, 2021). For these reasons, it is necessary to formulate a specific constitutive model for cryogenic applications that covers the entire transformation range and resolves the inherent complexity of the evolving microstructure of austenitic steels.…”