Hydrogen-related phase transformations in austenitic stainless steels

The effect of hydrogen and stress (strain) on the stability of the austenite phase in stainless steels was investigated. Hydrogen was introduced by severe cathodic charging and by elevated temperature equilibration with high pressure H₂ gas. Using X-ray diffraction and magnetic techniques, the behavior of two “stable” type AISI310 steels and an “unstable” type AISI304 steel was studied during charging and during the outgassing period following charging. Transformation from the fcc γ phase to an expanded fcc phase, γ*, and to the hcp ε phase occurred during cathodic charging. Reversion of the γ* and e phases to the original γ structure and formation of the bcc α structure were examined, and the kinetics of these processes was studied. The γ* phase was shown to be ferromagnetic with a subambient Curie temperature. The γ⇆ε phase transition was studied after hydrogen charging in high pressure gas, as was the formation of a during outgassing. These results are interpreted as effects of hydrogen and stress (strain) on the stability of the various phases. A proposed psuedo-binary phase diagram for the metal-hydrogen system was proposed to account for the formation of the γ* phase. The relation of these phase changes to hydrogen embrittlement and stress corrosion cracking of stainless steel is discussed.     

Simulation of the precipitation of sigma phase in duplex stainless steels

The precipitation of sigma phase within the ferrite component of a duplex stainless steel has been simulated using a two-dimensional computer model which takes into account the partitioning of alloy elements between ferrite and austenite. The model is based on a cellular automaton and, despite having a rather simple set of transition rules, is able to simulate changes in the volume fractions of the austenite, ferrite, and sigma phases. The microstructures produced are similar in appearance to those in the real system. Comparison of the model and the real system may assist in the assessment of the various phenomena occurring. Use is made of the model to examine many of the factors that might conceivably be harnessed to retard precipitation of the sigma phase in duplex stainless steels.     

Hydride formation and decomposition in electrolytically charged metastable austenitic stainless steels

An investigation of phase transformations in hydrogen-charged metastable austenitic stainless steels was carried out. Solution-annealed, high-purity, ultralow-carbon Fel8Crl2Ni (305) and laboratory-heat Fel8Cr9Ni (304) stainless steels were examined. The steels were cathodically charged with hydrogen at 1, 10, and 100 mA/cm², at room temperature for 5 minutes to 32 hours, in an lN H₂SO₄ solution with 0.25 g/L of NaAsO₂ added as a hydrogen recombination poison. Changes in microstructure and hydrogen damage that resulted from charging and subsequent room-temperature aging were studied by X-ray diffraction (XRD) and transmission electron microscopy (TEM). Hydrides from hydrogen charging (hep ε* in 305 SS and fcc γ* and hcp ε* in 304 SS) were observed. The evidence suggests the following mechanisms for hydride formation during charging: (1)γ → ε → ε* hydride and (2) γ → γ* hydride. These hydrides were found to be unstable and decomposed during room-temperature aging in air by the following suggested mechanisms: (1)ε* hydride (hcp) → expanded ε (hcp) phase →α′ (bcc) phase and (2) γ* hydride →γ phase. The transformation from ε* toα′, however, was incomplete, and a substantial fraction of ε was retained. A kinetics model for hydride decomposition and the accompanying phase transformation during aging is proposed.     

Modelling and simulation of dispersed phase formation in steels

Modelling has proven to be an efficient way of cutting time and costs associated with the investigation of materials properties. Two examples of thermodynamic modelling are presented which describe different phenomena in steels: the precipitation of carbonitrides and Ti-based inclusions in IF steels, and the coarsening of dispersed phases in HSLA steels. The underlying mode of describing the phases is the Calphad method, which has been combined with diffusion calculations to give results from both microscopic and macroscopic statistical points of view. The results of the calculations show a good agreement with the experimental data available in literature, and the methodology may be considered as illustrative and extendible to a variety of similar cases.     

Void formation during tensile testing of dual phase steels

The effects of martensite volume fraction (MVF) and strain state on necking behavior, post-uniform elongation, and the nucleation and growth of voids in thin sheet dual phase steel, strained in tension, were investigated. Steel containing, in weight percent, 0.08C, 1.45Mn, and 0.21Si, was cold rolled 50 pct and intercritically annealed to produce dual phase microstructures. The effects of MVF were evaluated with a series of constant geometry tensile samples with martensite volume fractions between 5 and 40 pct. The effects of strain state within the neck were evaluated with a series of constant thickness samples with 20 pct MVF and with width variations between 3 and 25 mm. A transition from diffuse to localized necking, as well as a decrease in post-uniform elongation, occurred with both an increase in MVF and sample width. Metallographic analysis of deformed samples revealed that the void nucleation occurs primarily at martensite particles by three distinct mechanisms. The void size and density in the necked region increased toward the fracture surface in all samples and the void density was significantly higher for the samples which exhibited localized necking. However, independent of neck geometry, voids were nucleated uniformly throughout the samples, and were associated with the martensite. The difference in void size and density between the samples with different necking behavior indicates that void growth is a consequence of the strain gradient while the shape of the voids depends on both the strain state and strain gradient. The implications of the void structure analysis are interpreted based on the dual phase microstructure. Formerly Graduate Research Assistant, Colorado School of Mines.     

High temperature phase chemistries and solidification mode prediction in nitrogen-strengthened austenitic stainless steels

Nitronic 50 and Nitronic 50W, two nitrogen-strengthened stainless steels, were heat treated over a wide range of temperatures, and the compositions of the ferrite and austenite at each temperature were measured with analytical electron microscopy techniques. The compositional data were used to generate the (γ + δ phase field on a 58 pct Fe vertical section. Volume fractions of ferrite and austenite were calculated from phase chemistries and compared with volume fractions determined from optical micrographs. Weld solidification modes were predicted by reference to the Cr and Ni contents of each alloy, and the results were compared with predictions based on the ratios of calculated Cr and Ni equivalents for the alloys. Nitronic 50, which contained ferrite and austenite at the solidus temperature of 1370 °C, solidified through the eutectic triangle, and the weld microstructure was similar to that of austenitic-ferritic solidification. Nitronic 50W was totally ferritic at 1340 °C and solidified as primary delta ferrite. During heat treatments, Nitronic 50 and Nitronic 50W precipitated secondary phases, notably Z-phase (NbCrN), sigma phase, and stringered phases rich in Mn and Cr.     

Grain structures in gas tungsten-arc welds of austenitic stainless steels with ferrite primary phase

The grain structures were investigated in full penetration gas tungsten-arc (GTA) welds in sheets of 304 and 321 austenitic stainless steels for a range of welding conditions. In type 321 steel welds, fine equiaxed ferrite dendrites were observed in the ferrite phase. The equiaxed structure was ascribed to heterogeneous nucleation of ferrite on Ti-rich cuboidal inclusions present in this steel, since these inclusions were observed at the origin of equiaxed dendrites. In type 304 welds, the ferrite grains were columnar, except in less complete penetration specimens, where a few coarse equiaxed dendrites appeared to originate from the weld surface. The secondary austenitic grain structure was columnar in both steels. In type 304 steel, the columnar austenitic grain structure did not necessarily correspond to the primary ferrite grains. In type 321 steel, the secondary austenite was columnar despite the equiaxed structure of the primary ferrite. Factors which affect the columnar-to-equiaxed transition (CET) are discussed. The failure to form equiaxed austenitic grains in type 321 steel is ascribed to austenite growing across the space between ferrite grains instead of renucleating on the primary equiaxed ferrite.     

Embrittlement of ferritic stainless steels

The mechanical properties and microstructures of commercial 11 to 29 pct Cr ferritic steels were examined as functions of aging times to 1000 h at 371, 482, and 593°C. Of the properties evaluated, changes in impact transition temperatures were the best measure of embrittlement. Embrittlement at 482°C occurs most rapidly in the 29 pct Cr alloy and somewhat more slowly in the stabilized 26 pct Cr alloy. The stabilized 18 pct Cr alloy embrittles much more slowly while little, if any, embrittlement was detected in a stabilizedll pct Cr alloy. Embrittlement at 482°C was characterized by a rapid change in properties followed by a plateau region and then further property changes. The early property change is attributed to precipitation of interstitial compounds and the later change to classic 475°C embrittlement. The onset of 475°C embrittlement in the two highest Cr alloys was accompanied by clustering of Cr atoms along {100} planes indicative of spinodal decomposition. Concurrent with clustering there was also a change from turbulent slip to a more planar slip along {110} planes. Some embrittlement was observed after longer exposures at 371°C which was attributed to a combination of 475°C embrittlement and the precipitation of interstitial compounds. Two of the alloys also embrittled at 593°C, accompanied by optically observable precipitates. The precipitate in the stabilized 18 pct Cr alloy was identified as Laves (Fe₂Ti) phase. One of the precipitates in the 29 pct Cr alloy was identified as sigma phase. Formerly with Allegheny Ludlum Steel Corporation.     

Creep-fatigue interaction in austenitic stainless steels

Low cycle fatigue failures occur by the initiation and controlled growth of a surface crack. The development of crack propagation models, based on continuum mechanics, have enabled successful predictions of fatigue life at both room and elevated temperatures. This paper attempts to extend such models to cover the situations in which creep damage, introduced during periods of stress relaxation, influences the rate of growth of the surface fatigue crack. Equations predicting fatigue life as a function of hold period are in good agreement with experimental data, for Type 304 stainless steel, Type 316 stainless steel and Incoloy-800.     

Instabilities in stabilized austenitic stainless steels

The effect of aging on the precipitation of grain boundary phases in three austenitic stainless steels (AISI 347, 347AP, and an experimental steel stabilized with hafnium) was investigated. Aging was performed both on bulk steels as well as on samples which were subjected to a thermal treatment to simulate the coarse grain region of the heat affected zone (HAZ) during welding. Aging of the bulk steels at 866 K for 8000 hours resulted in the precipitation of Cr₂₃C₆ carbides, σ, and Fe₂Nb phases; the propensity for precipitation was least for the hafnium-stabilized steel. Weld simulation of the HAZ resulted in dissolution of the phases present in the as-received 347 and 347AP steels, leading to grain coarsening. Subsequent aging caused extensive grain boundary Cr₂₃C₆ carbides and inhomogeneous matrix precipitation. In addition, steel 347AP formed a precipitate free zone (PFZ) along the grain boundaries. The steel containing hafnium showed the best microstructural stability to aging and welding. Formerly with Exxon Research and Engineering Company.     

Fe-Cr-Mn microduplex ferritic-martensitic stainless steels

An alloy development program has been undertaken with the aim of identifying an Fe-Cr-Mn stainless steel with ferritic-martensitic microduplex phase balance of sufficient stability to produce moderate strength and ductility, good impact resistance and acceptable as-welded properties. A microduplex, low C and N, Ti stabilized composition of Fe-11.5 pct Cr-3 pct Mn has been found to provide a yield strength of ⋍550 MPa, a tensile strength of ≃650 MPa, tensile elongation of 20 pct, a CVN impact transition temperature of-115°C (at 0.33 cm gage) and good weldability as determined by bend, impact, and intergranular corrosion testing. The alloy possesses general corrosion resistance roughly comparable to T405 and T430 ferritic stainless steels. The impact resistance achieved with the mixture of ferrite and martensite is inconsistent with previous concepts of second phase toughening in microduplex alloys, with the mixture apparently being significantly tougher than either of its components in bulk form. J.R. WOOD formerly with Allegheny Ludlum Steel Corporation, Brackenridge, PA     

An experimental and theoretical study of heat-affected zone austenite reformation in three duplex stainless steels

Three duplex grades, one molybdenum-free, one 22Cr type, and one super duplex grade, have been subjected to weld simulation treatments, and the resulting microstructures have been quantified by automatic image analysis techniques. Substantial differences between the duplex grades were observed with an increased ability to reform austenite with increased alloying content. A theoretical model has been applied, based upon the paraequilibrium concept elaborated by Hillert, and the paraequilibrium compositions of individual phases were calculated as a function of temperature using the THERMOCALC database. A model based on Cahns theory of grain boundary nucleated reactions has also been utilized to calculate the kinetics of the reaction. By using this model, the grain size effects could be included in the treatment. The results of the calculations were compared with experimental data, and the experimental results were reproduced using the same parameter set for the three materials, with the exception of the diffusion coefficient values which had to be adjusted. This adjustment has in a later study been verified experimentally. The results validate the model used and the physical relevance of using the paraequilibrium model. The appropriateness of a paraequilibrium approach is also supported by experimental evidence from weld metal compositions. It is shown that the nitrogen content of the alloys plays an important role, and a higher nitrogen content results in more efficient austenite reformation. This implies that the alloy nitrogen compositions should lie close to the upper specification limits for these materials and nitrogen losses should be avoided on welding since the material properties, both mechanical and corrosive, are strongly related to the austenite-ferrite phase ratio.     

Stress corrosion cracking of stainless steels

The similarities and differences in the stress corrosion cracking response of ferritic and austenitic stainless steels in chloride solutions will be examined. Both classes of materials exhibit a cracking potential: similar transient response (to loading) of the potential in open circuit tests or the current in potentiostatic tests and similar enrichment of chromium and depletion of iron in the film associated with localized corrosion processes. The ferritic steels are more resistant to localized corrosion than are the austenitic steels, which is responsible for the difference in the influence of prior thermal and mechanical history on cracking susceptibility of the two types of steel. Similarities in the fractography of stress corrosion cracks and those produced by brittle delayed failure during cathodic charging of the ferritic steels indicate that hydrogen embrittlement is involved in the failure process.     

Phase distributions in rapidly solidified stainless steels

Phase distributions and the internal magnetic fields have been determined in rapidly solidified stainless steels (Fe-nCr-8Ni-0.05C, Fe-nCr-5Ni, and Fe-nCr withn in the range of 10 to 24) by transmission and conversion electron Mössbauer spectroscopy (TMS and CEMS). Based on these results, a modification of the phase boundaries in the Schaeffler diagram is suggested to account, in particular, for rapidly solidified stainless steels. The suggested modification is primarily an expansion of the austenite field toward higher Cr and lower Ni equivalent contents. Combining CEMS and TMS makes it possible to determine the phase distributions both in the near surface region (outmost 300 nm) and in the bulk of the ribbons. For the low-Cr alloys, the content of the bcc phase (martensite) in the surface region is higher than in the sample as a whole. In the high-Cr alloys, the content of the bcc phase (ferrite) is lower in the surface than in the bulk. This disparity is ascribed to the different mechanisms of formation of martensite (diffusionless) and ferrite (nucleation and growth) in relation to the higher cooling rates of the surface layers. The determinations of the internal magnetic field are in good agreement with earlier investigations on conventionally processed Fe-Cr steels, where it was found that the internal magnetic field decreases with increasing Cr content.