A critical but non exhaustive review of the aqueous corrosion fatigue mechanisms is proposed with a particular emphasis on the corrosion-deformation interactions, The limits of the classical approaches of corrosion fatigue crack initiation and propagation predictions from the electrochemical point of view are pointed out. It is shown how the cyclic plastic deformation can locally modify the electrochemical reactions and how corrosion (anodic dissolution and/or hydrogen effects) can change the local plasticity and microfracture. The experimental methods to detect the localized corrosion fatigue damage are detailed. The crack propagation mechanisms near the fatigue threshold are underlined, with a particular attention for the anodic dissolution and hydrogen effects and the interaction between corrosion fatigue and stress corrosion cracking. Finally recent numerical simulations of corrosion fatigue damage and lifetime at mesoscopic scales are proposed and trends for future researches are given.

View full abstract

Cold model studies have been carried out to characterize the drop-size distribution in liquid-liquid emulsions formed by bottom gas injection. The emulsion was formed by injecting compressed air through the bottom of a tank containing water and kerosene. A polycondensation technique was used to take samples of the dispersed water drops inside the plume, using a specially designed pneumatic trap. The drop samples were photographed and size analysis was done using an image analysis software.
The drop-size distribution closely obeyed the Gaudin-Schuhmann equation. Correlations have been developed between the drop size distribution characteristics (the Sauter diameter and standard deviation) and appropriate dimensionless numbers characterizing the operating conditions. The experimental conditions covered the industrial range of these dimensionless numbers. Typical drop diameters were in the range of 0.5 to 8 mm. The values of the Sauter diameter were between 4 and 6 mm and those for the standard deviation were between 0.8 and 1.6 mm.

View full abstract

A laboratory scale fine particles-gas conveyed system was utilized to measure the hydrogen reduction rates of liquid wustite containing gangue such as 14.5 mol%CaO, 18.2 mol%SiO₂, or 5.15 mol%A₂O₃ at 1773 K. The N₂-H₂ mixture having various flow rates and compositions was flowed downward through a cylindrical reactor maintained isothermally and a batch of spherical wustite particles (mean dia.; 58 μm) was concurrently fed into the reactor at a small constant rate and reduced in a hot zone.
The reduction process was found to proceed in such a manner that a single metallic iron sphere was formed in the center of a wustite droplet. Rate analysis was made according to one dimensional mass balance equations for particles and gas in an isothermal steady moving bed. Under relatively small reducing potentials, it was concluded that the major fraction of overall reaction resistance is attributable to chemical reaction. The obtained chemical reaction parameters were found to be two orders of magnitudes larger than those in previous CO reduction. CaO raised the parameter, SiO₂ lowered it, and Al₂O₃ did not affect it, relative to gangue-free one. Under higher reducing potentials, the reduction process was estimated to include an appreciable diffusion resistance in the liquid particle.

View full abstract

The equilibrium distribution ratio of manganese between BaO-BaF₂-MnO slag and Mn(62-73%)-Fe-C_sat melt has been determined at 1573-1673 K. The effects of slag composition, oxygen partial pressure and temperature on the manganese distribution ratio have been investigated.
The equilibrium manganese distribution ratio was found to increase with oxygen partial pressure and BaO content in the slags. Addition of BaF₂ to BaO-MnO slag provided a beneficial condition to cut down the losses of manganese and enabled to dephosphorize ferro-manganese at a relatively high oxygen partial pressure. At a certain oxygen partial pressure, the molar fraction ratio of BaO to MnO decreased with the increased BaO content in a slag equilibrated with carbon-saturated ferro-manganese.
A high temperature resulted in a low manganese distribution ratio. For a BaO (50%)-BaF₂(47%)-MnO(3%) slag, the effect of temperature on the manganese capacity can be expressed as:
log C_Mn²⁺ = 21670/T−3.73
The heat of reaction
Mn(1)+1/2O₂=(MnO)
in the investigated temperature interval was estimated to be –415 kJ/mol.

View full abstract

The phosphorus distribution ratios between MnO (8-16 mass%)-containing CaO-SiO₂-Fe_tO slags and liquid iron were measured at 1873 K in a MgO crucible. The Addition of MnO was found to reduce the phosphorus distribution ratio at a given total Fe content and (mass%CaO)/(mass%SiO₂) ratio. The equilibrium quotient of dephosphorization reaction, (mass%P)/{[mass%P]·(mass%T.Fe)^5/2} the activity coefficient of P₂O₅, and that of Fe_tO in logarithmic form were formulated as a linear function of slag compositions. Based on these empirical formulas, the dephosphorization in hot metal and steel was thermodynamically assessed and the characteristics of double slag treatments were discussed from the mass balance relation between slag and metal with respect to phosphorus.

View full abstract

The manganese distribution ratios between CaO-SiO₂-Fe_tO-P₂O₅ (≤0.9 mass%)-MnO (8-16 mass%) slags and liquid iron were determined at 1873 K in a MgO crucible. The activity coefficient of MnO and the values for k'_Mn=(mass%MnO)/{[mass%Mn]·(mass%T.Fe)} and C_Mn=(mass%MnO)/{a_Mn·a_O} were formulated as functions of slag composition and temperature. By using these formulas, the manganese distribution ratios at 1623 and 1923 K were estimated as functions of total Fe content and (mass%CaO)/(mass%SiO₂) ratio. The manganese contents in hot metal and steel were also estimated as functions of manganese distribution ratio, slag volume and amount of manganese ore by using the mass balance relation.

View full abstract

In order to clarify the hydration characteristics of free CaO phase, the contents of divalent metal oxides in crystallized CaO phase in BOF, remelted BOF and synthetic slags were determined as functions of slag composition and cooling rate. The contents of divalent metal oxides increased by slow cooling to 1474 K, followed by rapid cooling so as not to precipitate the calcium ferrite phase in crystallized CaO phase. As a result, free CaO with the divalent metal oxides of more than 20 mass% was not corroded with ethylene glycol.

View full abstract

In order to investigate the equilibrium between yttrium and oxygen in liquid iron and nickel, the deoxidation equilibrium of yttrium in liquid iron and nickel saturated with yttria has been measured at the temperature ranges from 1600 to 1700°C using the sampling technique.
The equilibrium constant for the deoxidation reaction:
Y₂O₃(s)=2Y+3O
in liquid iron was found to be:
logK_Y(Fe)(K_Y(Fe)=a_Y²a_O³/a_Y2O3)=–36160/T+7.33    1600–1700°C
while the deoxidation product, K^'_Y(Fe)(=[%Y]²[%O]³) was expressed as follows:
logK'_Y(Fe)=logK_Y(Fe)–(–17350/T+3.14)(3[%Y]+11.1[%O])    0.02<[%Y]<0.11
by the use of the interaction parameter, e_O(Fe)^Y=–17350/T+3.14.
The equilibrium constant for the deoxidation of liquid nickel, log K_Y(Ni), was determined as the expression:
logK_Y(Ni)=–36250/T+6.36    1600–1700°C
Besides, the deoxidation product of nickel was represented to be:
logK'_Y(Ni)=logK_Y(Ni)+6.47(3[%Y]+11.1[%O])    0.01<[%Y]<0.11
with the interaction parameter, e_O(Ni)^Y=–6.47 at the temperature between 1600 and 1700°C.

View full abstract

The effect of the oxygen content in steel on the size distribution of oxides was investigated using Ti deoxidized steels. The number and size of oxides in the continuously cast steel slabs have been observed and the volume fraction of the fine oxides and the amount of oxygen precipitated as oxides during cooling and solidification have been examined. The results obtained are as follows.
The number and diameter of the fine complex oxides composed of mainly Ti₂O₃, Al₂O₃ and MnO increase with increasing the oxygen content in the steel. Almost all the oxygen in the steel is existing as oxides, namely the oxygen content of the steel is almost equal to the amount of the oxides.
The oxides precipitated during the cooling and solidification correspond to about 70% of the total oxide amount in the steel and are mostly smaller than 10 μm in size. Most of oxides present in molten steel before casting are also fine because these fine oxides can not float and separate from the molten steel and are resultantly suspended in the molten steel. The oxides smaller than around 10 μm occupy a large proportion of the total oxide content in the steel and govern the oxygen content in the steel.

View full abstract

Flow velocities and free surface deformation in an electromanetically stirred Wood's metal system were investigated using both computational and experimental methods. Local flow velocities were measured using an electromagnetic Vives probe. Computations were performed using a commercial computational fluid dynamics package modified with user subroutines to take into account the electromagnetic forces induced by the rotary electomagnetic stirrer. Results of the two methods were found to be in good agreement for the flow velocities, total free surface deformation, and the time scale to reach the final flow velocity and final shape of the free surface. The presented paper is reporting the results of the second part of the author's previously reported research.

View full abstract

Impact studies on hydrogen precharged and uncharged charpy specimens of Fe-0.45C-17Mn-2.8Al steel were conducted in the temperature range 77 to 300 K. For the uncharged specimens, it was observed that values of total absorbed energy and crack propagation energy decreased with lowering of temperature. The crack initiation energy followed the same trend in the temperature range 77-230 K but decreased with increase of temperature above 230 K. Hydrogen precharging resulted in decrease of these energy values and also the value of dynamic yield stress at all temperatures. Maximum decrease was noted in the temperature range 220-300 K.
Analysis of fracture surfaces, using the X-ray diffraction technique (XRD) and scanning electron microscopy (SEM) studies, suggested that the stable austenitic Fe-Mn-Al steel is prone to hydrogn degradation even at low temperatures and high strain rates. These observations are rationalised on the basis of an atomistic model for hydrogen embrittlement which does not essentially require redistribution of hydrogen concentration or martensitic transformation during the test.

View full abstract

A 9Cr-1Mo-V-Nb-N steel was subjected to the thermo-mechanical control process (TMCP), more specifically the direct quenching and tempering process, and the influences of heating temperature (Ts) and finish-rolling temperature (Tf) on the mechanical properties, including the creep rupture strength (CRS), were examined. The results were analyzed by observation of the substructure and precipitates, then the reasons for the improvement in CRS were discussed.
Raising Ts and lowering Tf improve CRS for separate reasons. Raising Ts augments the coherency strain around VN through lattice expansion owing to the increased solution of Nb to VN. This increase in coherency strain is the reason for the improvement of CRS. On the other hand, lowering Tf disperses VN more finely by serving dislocations as nucleation sites for VN, resulting in improved CRS through decreased interprecipitate distance.
When TMCP is applied, thin disc-like (V, Nb)N which nucleates on dislocations becomes a dominant type of VN precipitation, instead of NbN/VN complex precipitate which prevails when the steel is normalized and tempered. Furthermore, successive rolling from sufficiently high temperature is important to avoid coarse precipitation of VN and to exert the precipitation hardening during tempering to its full extent.

View full abstract

The effect of W addition on creep rupture strength (CRS) of high Cr ferritic steel was evaluated with three kinds of steels; base steel of 9Cr-1Mo-V-Nb-N, 0.7%W addition to base steel, and 1.7%W addition with decreased No addition to 0.5%. The latter two steels have the same values of Mo+0.5W viz. Mo equivalent. Further, the reasons for the beneficial effect of W was analyzed.
The CRS at 600°C for 1000 h increases with W addition at a rate of 35 MPa/%W. As the hardness remains almost constant during the creep test, the enhancement of CRS could not be attributed to the formation of precipitate or cluster during the creep test. The W addition was found to increase the partition of Nb to VN, resulting in the lattice expansion of VN and enhancing the CRS through coherency strain around VN. This effect accounts for about half of the increase of CRS by W addition.
The W addition also causes the precipitation of a film-like Laves phase along the subgrain boundary. Inconsequently, Cr₂C precipitates simultaneously. The Laves phase contributes to the other half of the CRS increase, possibly through suppressing the growth of subgrain during the creep test.

View full abstract

It has been long recognized that steels with high formability and strength can be produced with the aid of the deformation induced transformation of retained austenite to martensite. Selecting the steel composition and microstructure to obtain significant amounts of retained austenite has been the first step in designing TRIP steels. Thus, it is now well established that a significant amount of retained austenite can be obtained in steels containing high concentrations (>1%) of Si, where bainite is one of the main microconstituents. The effects of different intercritical annealing conditions on various aspects of Si-Mn TRIP steels have been extensively studied in other investigations. The present work was, however, conducted to investigate the influence of thermomechanical processing on the general behaviour of two Si-Mn TRIP steels, including a microalloyed Nb-bearing grade. The effects of microstructural features produced by dynamic recrystallization and pancaking of austenite, along with that of finishing temperature below tha T_nr (the austenite no-recrystallization temperature) on the subsequent transformations of the parent austenite, and the state of retained austenite at room temperature, were examined. It was found that, for microstructures comprised of polygonal ferrite, bainite and retained austenite, the variation of the retained austenite volume fraction with strain in the no-recrystallization region exhibits a maximum. Furthermore, the dynamically recrystallized austenite substructure, when retained to transformation temperatures, tends to retain more austenite.

View full abstract