Phase transformation of austenitic stainless steels as a result of cathodic hydrogen charging

The effects of cathodic hydrogen charging and aging on surface phase transformations were studied in solution treated and cold worked specimens of two austenitic stainless steels. Quantitative phase evaluation using an X-ray technique has shown that cathodic hydrogen charging and aging can result in a considerable amount of surface transformation toε andα ′ martensites. The extent of this surface transformation differs significantly from deformation-induced transformation at the same temperature, and abnormally high volume fractions ofε martensite are produced by the charging process. A minimum charging current density is necessary to induce transformation. In cold-worked samples, further surface transformation due to hydrogen charging and aging is inhibited by high volume fractions of pre-existing martensite. A. P. BENTLEY, formerly with the Department of Metallurgy and Materials Science, University of Cambridge     

The thermoelastic martensitic transformation in β′ Ni-Al alloys: II. Electron microscopy

The substructure of martensite in a 63.4 at. pct Ni-Al alloy was studied by transmission electron microscopy with particular attention to the “mixed” martensite morphologies observed in Ni-Al alloys, {111} <11?2>_fct transformation twinning was the only mode of lattice-invariant shear found and the measured twin thicknesses lead to an inhomogeneous shear value,m ₂, in excellent agreement with that predicted from the crystallographic theory. The internal twins in the martensite plates forming characteristic mixed morphologies were found to be at definite angles to one another and these angles were carefully measured to be 96 deg and 125 deg for the “fork” type or “spear” type martensite morphologies. The corresponding angles as predicted from the Bowles-Mackenzie theory are 94.2 deg and 124.4 deg. The interface plane between the martensite variants that form the spear type morphology was found to be a {011}_β' type plane with the internal twins in each variant twin related to each other across this mirror plane.     

Hydrogen transport in nickel-base alloys

The electrochemical permeation technique has been used to characterize hydrogen transport and trapping in pure nickel and in alloys 600, X-750, and 718 at a temperature of 80 °C. The “effective diffusivity ” of hydrogen atoms in alloy 600 is reduced by a factor of about 5 compared to pure nickel. This is attributed to both compositional changes and the presence of [(Ti, Nb)C] carbides. Aging of alloy 600, with subsequent M₂₃C₆ carbide precipitation, does not significantly influence the measured “effective diffusivity,” which is explained by the dominant effect of preexisting [(Ti, Nb)C] carbides. The “effective diffusivity” of hydrogen atoms in solution-annealed alloy X-750 is reduced by a factor of about 9 compared to that of pure nickel. This is also attributed to compositional changes and [(Ti,Nb)C] carbides. Aging of alloy X-750, which causes precipitation of γ′[Ni₃(Al, Ti)], reduces the “effective diffusivity” by an additional factor of 5 or more. Double aging at 885 °C/24 hours, 704 °C/20 hours following hot working yields the greatest reduction in “effective diffusivity.” Analysis of permeation transients using a diffusion- trapping model indicates a binding energy associated with trapping due to the γ phase of be- tween -31 and -37 kJ/mol. The “effective diffusivity” of hydrogen in alloy 718 is about 40 pct greater than for alloy X-750 for the same double and direct aging treatments. The average “effective diffusivities” of the double-aged and direct-aged alloy 718 are comparable, but the permeation transients for the double-aged treatment are significantly steeper. The double-aged treatment with predominantly δ phase (orthorhombic Ni₃Nb) yields a binding energy of about -30 kJ/mol. Analysis of the direct aged-treated 718, which contains predominantly γ″ phase (body-centered tetragonal Ni₃Nb) gave a binding energy between -23 and -27 kJ/mol. Seg- regation of hydrogen atoms to the γ matrix interface, combined with a large volume fraction of γ′ at grain boundaries, provides the most likely explanation for the enhanced cracking as- sociated with the double-aging treatment in alloy X-750.     

A study of electrical resistivity,internal friction and shear modulus on an aged Ti49Ni51 alloy

The 400°C aged Ti₄₉Ni₅₁ alloy can exhibit the transformation sequence of B2 →r premartensite R-phase →r martensite. In the early aging stage, only the premartensitic transformation is observed due to the M_s point being deeply depressed by the coherent stress of Ti₁₁Ni₁₄ precipitates. In the later aging stage, internal friction peaks associated with premartensitic and martensitic transformations are all observed on both heating and cooling. The sharp peaks associated premartensitic transformation on heating is believed to be related to the “bias” effect of the coherent stress induced by the Ti₁₁Ni₁₄ precipitates. The serrations of internal friction appearing significantly in the temperature around −30 to −80°C are found to be caused by the stress induced accomodation of R-phase or martensite variants, and are not associated with the transformation. The Ti₁₁Ni₁₄ precipitates can enhance the amount of martensite formed by unit of temperature or time during the martensitic transformation.     

Microstructural evolution of modified 9Cr-1Mo steel

The tempering and subsequent annealing of modified 9Cr-lMo steel have been investigated to determine the influence of trace amounts of V and Nb on the sequence of precipitation processes and to identify the basis for the enhanced high-temperature strength compared to the standard 9Cr-lMo composition. Air cooling (normalizing) from 1045 °C results in the precipitation of fine (Fe, Cr)₃C particles within the martensite laths. Additional carbide precipitation and changes in the dislocation structure occur during the tempering of martensite at 700 °C and 760 °C after normalizing. The precipitation of M₂₃C₆ carbides occurs preferentially at lath interfaces and dislocations. The formation of Cr₂C was detected during the first hour of tempering over the range of 650 °C to 760 °C but was replaced by V₄C₃ within 1 hour at 760 °C. During prolonged annealing at 550 °C to 650 °C, following tempering, the lath morphology remains relatively stable; partitioning of the laths into subgrains and some carbide coarsening are evident after 400 hours of annealing at 650 °C, but the lath morphology persists. The enhanced martensite lath stability is attributed primarily to the V₄C₃ precipitates distributed along the lath interfaces and is suggested as the basis for the improved performance of the modified 9Cr-lMo alloy under elevated temperature tensile and creep conditions.     

The morphology and substructure of {225}F martensite in an Fe−Mn−Cr−C alloy

A transmission electron microscopy and diffraction study of martensite plates in an Fe-3 pct Mn-3 pct Cr-1 pct C alloy was carried out with particular attention to details of the martensite substructure. A corresponding optical metallographic study of plate morphology was made. A variability in martensite substructure was observed from plate to plate, although a (252)_F ^* plate was generally associated with (112)_B transformation twins and {111}_F stacking faults. The particular (111)_F fault variant gave rise to a wedge-shaped plate morphology. Planar {101}_B inhomogeneities were frequently observed in the martensite, and most of these appear to be derived from austenite stacking faults. In general, more than one type of inhomogeneity was observed in a single martensite plate and the “typical” plate substructure was rather difficult to characterize, although the habit plane was found invariably to be (252)_F.     

High-Temperature oxidation of cathodically hydrogen-charged two-phase (Ti3Al,TiAl) titanium aluminides

Ti-42A1, Ti-45A1, and Ti-5OA1 (at. pct) titanium aluminides, which were cathodically hydrogen charged in a 5 pct H₂SO₄ solution for charging times between 1.8 ks (0.5 hours) and 14.4 ks (4 hours), were oxidized in a static air under atmospheric pressure at temperatures between 1170 K (897 °C) and 1350 K (1077 °C). All the hydrogen-charged alloys, as well as alloys without hydrogen charging, followed parabolic oxidation kinetics. The weight gains of the alloys after hydrogen charging for normally less than 3.6 ks (1 hour) were 20 to 30 pct less than those without hydrogen charging. In the alloys charged with hydrogen for more than 7.2 ks (2 hours), the weight gains increased with increasing the charging time. The activation energies of oxidation indicated that the oxidation-controlling factor would change after a charging time of 7.2 ks (2 hours) in all the alloys. The decrease in the activation energies with charging time was more drastic in the Ti-5OA1 alloy, which suggested that hydrogen damage, such as cracking, was more severe in the Ti-50Al alloy than in the Ti-42A1 or Ti-45A1 alloys. The formation of cracks during hydrogen charging provides titanium-diffusion paths and accelerates formation of rutile (TiO₂) scale on the surface of the alloys. The TiO₂ on the alloys after hydrogen charging formed at a comparatively lower temperature than that on the alloys without charging.     

Concentration dependence of hydrogen diffusivity in amorphous Fe40Ni38Mo4B18 alloy

Hydrogen permeation in amorphous Fe₄₀Ni₃₈Mo₄B₁₈ alloy was studied using an electrochemical technique in the temperature range 303–348 K. Hydrogen diffusivities were calculated by several methods. They increased as the charging current density increased. The surface, concentration, and trapping effects on hydrogen diffusion were carefully examined. It was suggested that the surface impedance was not existent and diffusivity was affected by the “intrinsic” concentration and trapping effects. When diffusivities were correlated with hydrogen concentration, the effective activation energy decreased from 36.8 to 34.8 kJ/mol as the surface hydrogen concentration increased from 1000 to 6000 mol H/m³. Under the same charging current density, hydrogen concentration in the alloy decreased as temperature increased. Therefore, lower effective activation energies would be obtained if diffusivities were correlated with the charging current density. The concentration dependence of hydrogen diffusivity was rationalized on the basis of the structural features of amorphous alloys.     

The effect of aging on hydrogen embrittlement of a nickel alloy

The effect of aging at 500° C on the hydrogen embrittlement tendency of a cold worked Ni-base superalloy was investigated in a series of experiments which included hydrogen charging studies, mechanical tests in hydrogen and in air, and fractographic and slip line investigations. Embrittlement tendency increased (time-to-failure decreased) markedly during the first hour of aging and then remained constant until about 1000 h aging time, whereupon it increased rapidly again. The short-time embrittlement could be accounted for either by a mechanism involving segregation of P to grain boundaries or by one involving planar slip induced by short-range order. The hydrogen charging studies indicated that hydrogen uptake decreases during aging, a result which is not consistent with the P segregation hypothesis. The increase in embrittlement at long aging times is most readily explained in terms of planar slip induced by long range order. Tensile tests over a range of strain rates suggested that accelerated transport of hydrogen by dislocation dragging of hydrogen atmospheres is involved in embrittlement. Formerly Graduate Assistant, Dept. of Met. Eng. & Mat. Sci., University of Notre Dame     

Stress- and strain-induced formation of martensite and its effects on strength and ductility of metastable austenitic stainless steels

The effects of deformation-induced formation of martensite have been studied in metastable austenitic stainless steels. The stability of the austenite, being the critical factor in the formation of martensite, was controlled principally by varying the amounts of carbon and manganese. The formation of martensite was also affected by different test and rolling temperatures, rolling time, and various reductions in thickness. The terms “stress-induced” and “strain-induced” formation of martensite are defined. Experimental results show that low austenite stability resulted in stress-induced formation of martensite, high work-hardening rates, high tensile strengths, low “yield strengths,” and low elongation values. When the austenite was stable, plastic deformation was initiated by slip, and the work-hardening rate was too low to prevent early necking. A specific amount of strain-induced martensite led to an “optimum” work-hardening rate, resulting in high strengthand high ductility. For best results processing should be carried out aboveM _d and testing betweenM _d andM _s. Mechanical working aboveM _d had a negligible effect on the yield strength betweenM _d andM _s when the austenite stability was low, but its effect increased as the austenite became, more stable. Serrations appeared in the stress-strain curve when martensite was strain induced.     

The influence of austenite stability on the hydrogen embrittlement and stress- corrosion cracking of stainless steel

A study has been made of the HE and SCC of a type 304 and a type 310 austenitic stainless steel, and the results correlated with the presence or absence of α′ martensite, determined by means of a ferrite detector. Hydrogen induced slow crack growth (SCG) was observed at room temperature when type 304 was stressed i) in 1 psig (∼10⁵ N/m²) gaseous hydrogen, ii) after high temperature charging, and iii) while undergoing cathodic charging. The fracture surfaces corresponding to SCG were primarily transgranular and cleavage-like, and were found to be associated with α′. Conditions i) to iii) did not produce SCG in the type 310 steel, in which α′ martensite was not detected, nor did SCG occur when type 304 was stressed in gaseous hydrogen above the M_D temperature (∼110°C). These observations indicated that the formation of the martensitic phase was a prerequisite for SCG under these test conditions. Stressing of type 310 while it was undergoing cathodic charging at room temperature was found to produce shallow, nonpropagating cracks, confirming earlier reports that austenite can be embrittled by hydrogen in the absence of α′. SCC occurred in both alloys in boiling aqueous MgCl₂ (154°C) with no evidence for α′ formation. The results are discussed in terms of the mechanisms of HE and SCC. Formerly Research Associate, Department of Metallurgy and Mining Engineering, University of Illinois. Formerly Corrosion-Control Analyst with the Physical Plant at the University of Illinois.     

Influence of austenite and martensite strength on martensite morphology

The martensite morphology and austenite flow strength have been determined in a variety of ferrous alloys chosen so that the austenites were paramagnetic, ferromagnetic, substitutional strengthened, and interstitial strengthened. It is demonstrated that two of the most important variables in determining the habit plane (and thus morphology) of martensite in a given alloy are the resistances to dislocation motion in austenite and in ferrite (i. e., martensite). In the wide variety of alloys where martensite with a {259}_γ habit plane was observed, the austenite flow strength atM _s is greater than 30,000 psi. At lower austenite strengths, either {225}_γ or {111}_γ habit planes are found depending on the resistance to dislocation motion in ferrite. Thus, {225} martensites are not always found as part of the spectrum between {111} and {259} martensites but only in the cases (e. g., interstitial strengthening) where ferrite is preferentially strengthened relative to austenite. All of the observations are consistent with the idea that the habit plane observed in a given alloy is the one involving the minimum plastic work for the lattice invariant shear.     

The shape memory effect and pseudoelasticity in polycrystalline Cu-Zn alloys

The shape memory effect associated with the reverse transformation of deformed martensite, pseudoelastic behavior involved in stress-induced martensite formation and the reversion of strained martensite after an applied stress is relaxed aboveA _f have been studied. Grain size and specimen geometry effects have been related to the above phenomena. Although recoverable strains as high as 10.85 pct were observed in coarse-grained (“bamboo” type) specimens, the shape memory effect is restricted in fine-grained specimens because of permanent grain boundary deformation and intergranular fracture which occurs at relatively low strains. A fine grain size also acts to suppress pseudoelastic behavior because permanent, localized deformation is generated concurrent with the formation of stress-induced martensite which inhibits reversion of the latter upon release of stress. The apparent plastic deformation of martensite belowM _f can be restored by transforming back to the original parent phase by heating toA _f (shape memory) or alternatively, can be recovered belowM _f by applying a small stress of opposite sign. Martensite deformed belowM _f with the same stress maintained while heating persists aboveA _f, but reverts to the parent phase in a pseudoelastic manner when the stress is relieved. The athermal thermoelastic martensite, which forms in groups composed of four martensite plate variants, undergoes several morphology changes under deformation. One of the variants within a plate group cluster may grow with respect to the others, and eventually form a single crystalline martensitic region. At a later stage pink colored deformation bands form in the same area and join up with increasing stress, resulting in thermally irreversible kinks. The clusters of plate groups may expand like grain growth or contract as a whole during deformation, or act as immobile “subgrains” which lead to permanent deformation at their boundaries. Stress-induced martensite usually forms as one variant of parallel plates which join up with increasing stress to form single crystalline regions. Further stress leads to pink colored deformation bands, similar to those in the deformed athermal martensite. Other similarities and differences between the stress-induced and athermal martensite have been investigated and are discussed.     

Effect of Si on δ-Ferrite/γ-Austenite Transformation and M23C6 Precipitation During Solidification of X10CrAlSi18 Ferritic Heat-Resistant Stainless Steel

Herein, the δ-ferrite/γ-austenite transformation and the precipitation behavior of M₂₃C₆ carbides in X10CrAlSi18 ferritic heat-resistant stainless steel (FHSS) with various Si contents at a cooling rate of 100 °C min⁻¹ using confocal scanning laser microscopy (CSLM) are investigated. The findings reveal that γ-austenite preferentially forms along the δ-ferrite phase boundaries, and it progressively precipitates into the δ-ferrite phase as the temperature decreases. The increase in the Si content reduces the δ-ferrite/γ-austenite transformation temperature. It also inhibits the martensite transformation in the subsequent cooling process, decreasing the volume fraction of γ-austenite/martensite. M₂₃C₆ carbides are mostly found at the δ-ferrite and γ-austenite/martensite phase boundaries. Meanwhile, the nucleation of M₂₃C₆ carbides becomes more difficult as the volume fraction of γ-austenite/martensite decreases. Furthermore, the complex solidification mechanism of the nucleus is addressed.     

Effect of High-Pressure Torsion Processing and Annealing on Hydrogen Embrittlement of Type 304 Metastable Austenitic Stainless Steel

The effect of high-pressure torsion (HPT) and annealing on hydrogen embrittlement (HE) of a type 304 stainless steel was studied by metallographic characterization and tensile test after hydrogen gas charging. A volume fraction of ~78 pct of the austenite transformed to α′ martensite by the HPT processing at an equivalent strain of ~30. Annealing the HPT-processed specimen at a temperature of 873 K (600 °C) for 0.5 hours decreased the α′ martensite to ~31 pct with the average grain size reduced to ~0.43 μm through the reverse austenitic transformation. Hydrogen charge into the HPT-processed and the HPT+annealed specimens in the hydrogen content of ~10 to 20 ppm led to no severe HE but appeared in the solution-treated specimen. Especially the 873 K (600 °C) annealed specimen had the ~1.4 GPa tensile strength and the ~50 pct reduction of area (RA) despite the hydrogenation.     

Validation and Analysis of the Parameters for Reconstructing the Austenite Phase from Martensite Electron Backscatter Diffraction Data

This work focuses on the validation of a method for reconstructing the fcc crystallographic data from martensite orientation electron backscatter diffraction (EBSD) maps based on the “γ nuclei identification” and “γ nuclei spreading strategy.” To that end, an Fe-30Ni alloy was employed. The martensite transformation start temperature (M _s ) of this material is close to or below room temperature; therefore, during hot deformation and after water quenching, it presents an fcc austenitic microstructure, while after subzero quenching, austenite-to-martensite transformation takes place. Accordingly, the reconstruction procedure can be applied to the martensitic EBSD crystallographic data, and the morphological and orientation results of the reconstruction can be validated by comparison with the original crystallographic fcc data. Torsion tests were performed to produce recrystallized and deformed austenite microstructures. Although applying the Kurdjumov–Sachs orientation relationship (OR) resulted in reconstructed area fractions larger than 75 pct, the reconstruction quality improved significantly when other ORs closer to the Greninger–Troiano OR were applied. The analysis carried out on the recrystallized microstructure shows that the method is robust against variation in the different parameters involved in the reconstruction. Good angular and morphological reconstruction results were obtained in both recrystallized and deformed microstructures, including the ability to reconstruct twins.     

Influence of Partitioning Effects on the Retained Austenite Content and Properties of Martensitic Stainless Steel

The effect of a quenching and partitioning (Q&P) heat treatment with a quenching temperature (T_Q) range from 20 to 190 °C is investigated for two martensitic stainless tool steels X40Cr14 and “X25CrN13”, focusing on microstructural evolution, hardness, and toughness. The influence on the retained austenite (RA) content, when replacing part of carbon with nitrogen, is of core interest. The amount of RA is analyzed by X-ray diffraction and is additionally proved with electron backscatter diffraction, and the RA content is thermodynamically calculated. Subsequently, the effect of the microstructure on toughness and hardness is investigated. For both steels, the toughness maximum is reached in the region of the RA maximum. The “X25CrN13” attains higher toughness at higher RA contents. Higher RA contents do not benefit X40Cr14. Furthermore, the effect of double tempering at higher tempering temperatures after Q&P on the steels is investigated. Besides RA contents and hardness, dilatometer curves are used to evaluate the formation of fresh martensite in the microstructure. The secondary hardness maximum of “X25CrN13” is reached at 500 °C and that of X40Cr14 is at 480 °C. For double tempering temperature at 520 °C, T_Q has little effect on toughness, and “X25CrN13” shows better values.     

Martensite lattice changes during tempering

An X-ray diffractometer study of martensite formed in an 18 wt pct nickel, 0.98 wt pct carbon austenite single crystal yields the shapes, positions, and integrated intensities of 200, 020, and 002 peaks. Martensite, which forms below ? 60 °C, was tempered at successively higher temperatures from ?45 to 450 °C. The results show that after subambient aging, during which C atoms in c-oriented octahedral sites have clustered, carbide precipitation starts and small regions (～30Å in the [001]) with negative tetragonality appear. Upon subsequent tempering these are augmented by larger regions which have small positive tetragonality. In this process the “c” lattice parameter changes markedly but the “a” and “b” lattice parameter increase very little. These results indicate the formation of carbon depleted martensite which is coherently strained by the carbide particles. At and just above 100 °C the 200, 020, and 002 peaks all become doublets as the martensite matrix discontinuously breaks free of coherency and becomes highly imperfect ferrite. This change also occurs during the so-called “first stage of tempering.” Further tempering decreases the defect content of this ferrite. The lattice of the martensite is extensively reoriented during tempering just above room temperature. These reorientations probably accommodate the lattice parameter changes described above and may be carried out by movement of twin boundaries.     

Strength and fatigue properties in medium carbon duplex steels

The effect of microstructure on strength and fatigue properties has been investigated in two medium carbon alloy steels (BS 817M40 and BS 835M30) by developing dual-phase, ferritic-martensitic microstructures. Hardness-strength relationships and fatigue resistance at comparatively high strength levels were investigated by producing various microstructures. Conventional quenching and tempering, intercritical annealing and step quenching were used to vary the proportion, morphology and distribution of the ferrite and martensite phases. The results of the present study show that both hardness and strength increase with increasing proportion of martensite and/or hardness of the second phase. The relationship between hardness or strength and martensite percent is not in good agreement with a simple “law of mixtures” but is compatible with a more rapid strength increase at high martensite contents. The dual phase microstructures from the present study show superior near threshold ΔK_TH values than normal tempered martensite. The results also show a high degree of correlation between Paris equation m values and fracture toughness K_IC, showing that for high m values K_IC is low and vice versa. The present experiments show that although crack initiation resistance in dual-phase steels is excellent crack propagation rates are higher than in quenched and tempered microstructures for a given ΔK.     

Trap-governed hydrogen diffusivity and uptake capacity in ultrahigh-strength AERMET 100 steel

The hydrogen-uptake capacity and mobility in ultrahigh-strength AERMET 100 are characterized for various electrochemical charging and baking conditions. From thermal desorption spectroscopy, the apparent hydrogen diffusivity (D _H < 3 × 10⁻⁸ cm²/s at 23 °C) is over tenfold less than the values typical of tempered martensitic steels such as AISI 4130. The value of D _H decreases with decreasing temperature below 200 °C, with a relatively high apparent activation energy for diffusion of 17.7 to 18.8 ± 0.2 kJ/mol at the 95 pct confidence level. The value of D _H also decreases with decreasing diffusible H concentration from less-severe charging or increased baking. Potentiostatic charging in saturated Ca(OH)₂ produced total and diffusible H concentrations in AERMET 100 which increase with (H⁺/H) overpotential and are significantly higher than results for AISI 4130 steel under the same conditions. A significant H concentration was produced by zero overpotential deposition. These characteristics are explained by extensive reversible and irreversible H trapping involving at least three unique trap states in the ultrafine AERMET 100 microstructure. The former likely include coherent M₂C carbides, soluble Ni, or precipitated austenite, and the latter include larger incoherent M _x C _y or martensite lathed-packet interfaces. Baking at 23 °C and 200 °C removes H from the lowest binding-energy sites, but results in reduced D _H levels to prolong outgassing time. Additionally, substantial H was retained in stronger trap states. These trapping effects are pertinent to hydrogen embrittlement of AERMET 100 steel.