Creep Resistant Ferritic Steels Research Articles

A first step towards computational design of W-containing self-healing ferritic creep resistant steels

ABSTRACT In this work, we combine a generic alloy-by-design model with a novel concept, the nucleation barrier for the formation of Laves phase to fill the creep cavities, in order to develop multi-component creep resistant steels with kinetically tuned self-healing behaviour. In the model the high-temperature long-term strength is estimated by integrating precipitation strengthening due to M23C6 carbides and solid solution strengthening, while the optimized compositional solutions are determined by employing the coupled thermodynamic and kinetic principles. W-containing Laves phase herein is selected as the self-healing agent to autonomously fill the grain boundary cavities, so as to prolong the creep lifetime. To achieve the effective healing reaction, the nucleation time for Laves precipitates are expected to coincide simultaneously with which creep cavities start to form or reach a healable size. Using experimental data from literature, an empirical relationship to estimate the incubation time for Laves phase formation has been constructed, from which the thermodynamic driving force for onset of precipitation as a function of temperature and intended precipitate nucleation time was derived. Three sample alloys have been selected among the desirable solutions, which are predicted to have the same strength but widely different Laves phase nucleation times. The calculations are also performed for different use temperatures to explore the compatibility between high temperature strength and timely cavity filling behaviour. In its current form the model is not expected to yield the truly optimal composition but to demonstrate how the kinetics of the healing reaction can affect the predicted optimal alloy compositions.

Creep Damage Tolerance Factor λ of Selected Creep-Resistant Steels

The creep damage tolerance factor λ as an important outcome of the continuum damage mechanics approach has been used to asses the creep fracture mode and the susceptibility of material to localized cracking at strain concentrations. In this work, using sets of our earlier published creep data of three advanced ferritic creep-resistant steels (T23 low alloy steel, P91 and P92 chromium steels) are analysed in terms of the creep damage tolerance factor λ. It was found that the value of the creep damage factor λ is not constant and depends on the creep loading conditions. The data analysis is followed by fractographic investigations, which is used to identify the creep fracture mode (s) experimentally.

Effect of rhenium on the microstructure and mechanical behavior of Fe–2.25Cr–1.6W–0.25V-0.1C bainitic steels

A new ferritic creep resistant steel has been developed by eliminating Nb and adding 1.5 mass % Re to a ferritic steel grade T/P23 with the aim of enhancing its mechanical properties at high temperature. Cast ingots of both steels, new grade and ASTM T/P 23, were hot rolled at 900°C and then submitted to a thermal treatment consisting of solubilization at 1050°C and tempering at 700°C. Tempered bainitic microstructures obtained contain second phases reinforcing carbide particles, mainly M6C and M23C6 at the boundaries of both, prior austenite grains and bainitic ferrite laths, as well as MC within the grains. Mechanical properties at temperatures ranging from 540 to 600°C were studied by strain-rate-change tests in compression at strain rates between 10−7 and 10−4s−1. These tests showed high stress exponents (n≥20) and activation energies (Q≈400kJ/mol) for both alloys, which were associated with a dislocation movement mechanism with a strong interaction between dislocations and precipitates. On the other hand, a creep exponent of 5 was derived for the stress dependence of minimum creep rate from conventional-type creep tests at 600°C. Although this stress exponent is usually related to a dislocation climb controlled creep mechanism, remarkable microstructural degradation observed with increasing creep time makes difficult to elucidate the true deformation mechanism controlling creep.

Charting the ‘composition–strength’ space for novel austenitic, martensitic and ferritic creep resistant steels

We report results of a large computational 'alloy by design' study, in which the ‘chemical composition–mechanical strength’ space is explored for austenitic, ferritic and martensitic creep resistant steels. The approach used allows simultaneously optimization of alloy composition and processing parameters based on the integration of thermodynamic, thermo-kinetics and a genetic algorithm optimization route. The nature of the optimisation depends on both the intended matrix (ferritic, martensitic or austenitic) and the desired precipitation family. The models are validated by analysing reported strengths of existing steels. All newly designed alloys are predicted to outperform existing high end reference grades.

Creep behavior of dissimilar metal weld joints between P91 and AISI 304

Dissimilar welds between ferritic creep-resistant steels and austenitic stainless steels in power plant service are known to fail prematurely due to a large difference in their thermal coefficient of expansion. To address this problem, in the current work, modified 9Cr-1Mo steel (P91) and austenitic stainless steel (AISI 304) transition joints were produced using friction welding employing three interlayers: Inconel 625, Inconel 600, and Inconel 800H. These interlayers were friction welded one over another on P91 alloy, which was finally friction welded to AISI 304. The joints produced with single and three interlayers were subjected to creep rupture tests at different temperatures and stress levels. The creep test results demonstrated that the proposed approach can help realize striking improvements in creep performance of P91/AISI 304 dissimilar metal welds. Further, analysis of creep resulted in a stress exponent (n) of 3 by incorporating the concept of threshold stress which suggested the creep deformation mechanism to be viscous glide due to solute drag. Creep rupture data was also analyzed using Monkman-Grant relationship and creep damage tolerance factor (λ). Results indicated that the damage mechanism is due to cavity growth by the combined effect of power law and diffusion creep.

Hydrogen induced cold cracking of creep resistant ferritic P91 steel for different diffusible hydrogen levels in deposited metal

Hydrogen-assisted cracking (HAC) is a serious issue in P91 steel welds because of unwanted martensitic metallurgical-transformation in the heat affected zone (HAZ). In the present research, different electrode conditions have been employed during welding of P91 steel to study their effects on HAC. The HAC susceptibility of cast and forged (C&F) P91 steel welds produced by shielded metal arc welding (SMAW) process has been evaluated through implant test with respect to the different level of diffusible hydrogen content in the deposited metal. The weld implant test results such as applied stress vs time to failure, the lower critical stress (LCS), embrittlement index (EI) and normalized critical stress ratio (NCSR) were evaluated with respect to the different level of hydrogen content in the deposited metal. The microstructure of coarse grain heat affected zone (CGHAZ) and weld metal in different condition have been characterized by using scanning electron microscope (SEM). The fracture morphology of implant specimen was characterized by using the SEM. The paper also describes the effect of hydrocarbon contamination (oil or grease) in low-hydrogen basic SMAW electrode for P91 steel. It can be concluded that P91 steel welded by the electrode having higher contamination (high hydrogen level) is more susceptible to HAC. The LCS value was observed to be decreasing with the increase in the diffusible hydrogen content in deposited metal and minimum value was measured to be 180 MPa for 12.54 ml of diffusible hydrogen per 100 g of deposited metal.

Non-instantaneous growth characteristics of martensitic transformation in high Cr ferritic creep-resistant steel

Microstructural observation and high-resolution dilatometry were employed to investigate kinetics of martensitic transformation in high Cr ferritic creep-resistant steel upon different quenching/cooling rates. By incorporating the classical athermal nucleation and impingement correction, a non-instantaneous growth model for martensitic transformation has been developed. The developed model describes austenite/martensite interface mobility during martensite growth. The growth rate of martensite is found to be varied from 1 × 10−6 to 3 × 10−6 m/s. The low interface mobility suggests that it is not appropriate to presume the instantaneous growth behavior of martensite. Moreover, based on the proposed model, nucleation rate of martensite under different cooling rates is found to be nearly the same, while the growth rate of martensite is promoted by increasing the cooling rate.

Improving the high-temperature creep strength of 15Cr ferritic creep-resistant steels at temperatures of 923–1023 K

The precipitation strengthening of creep-resistant 15Cr ferritic steels at temperatures of 923–1023K was improved by nickel addition. The addition of nickel, an austenite-stabilizing element, enhanced the formation of martensitic grains in the ferritic grain boundaries and changed the precipitation behavior during creep at elevated temperatures. The precipitates produced in the creep-ruptured 15Cr steels were intermetallic compounds (Laves phase and χ-phase), carbide (Cr23C6), and Z phase (Cr(V,Nb)N). An increase in the amounts of intermetallic compounds and the preferential precipitation of Cr23C6 carbide on the grain boundaries between the ferritic and martensitic grains were accelerated by the nickel addition. The 15Cr steel after 10,000h at 973 and 1023K demonstrated approximately double the creep rupture strength of 9Cr steel (ASME Grade T92) with a tempered martensitic microstructure. At temperatures of up to 1023K, the 15Cr steel had over 10 times the creep rupture lifetime of T92 steel, primarily because of increased precipitation of intermetallic compounds and Cr23C6 carbide enhanced by nickel addition.

The impact of intended service temperature on the optimal composition of Laves and M23C6 precipitate strengthened ferritic creep resistant steels

Recently, the Laves phase precipitate family has been identified as a promising alternative for Precipitation Hardening (PH) in ferritic creep resistant steels owing to its significant contribution to creep strength. A computational alloy design approach to create novel ferritic creep resistant steels strengthened by optimized Laves phase is presented. This novel approach aims at a simultaneous optimization of the steel’s chemical composition involving 9 chemical elements to obtain a ferritic steel optimally strengthened by Laves phases with an ideal combination of a high volume fraction and a reduced coarsening rate at the intended service temperature. Using this approach, three alloys possessing ideal PH contributions at their intended service temperatures, i.e. 650, 700 and 750°C, respectively, are designed. This approach is validated by analyzing PH contributions in existing 15Cr ferritic steels at different service temperatures. Good agreement between reported experimental performance and model predictions is found. The newly designed alloys are predicted to have a substantially higher PH contribution at corresponding service temperature than those of the existing alloy. While Laves phases generally precipitate slowly during service and have a more significant strengthening effect at the late stage, the precipitation of the alternative M23C6 precipitates can be used to increase the initial strength of ferritic steel. Hence, our computational approach was extended to aim for a high strengthening contribution at both initial and late stages at the intended service temperature.

On the selection of halide activators for the formation of hybrid Ni-aluminide/Ni coatings on creep resistant ferritic steels by low temperature pack cementation process

Thermochemical calculations were made and experiments were subsequently conducted to verify the thermochemical predictions in an attempt to establish a basis for selecting suitable halide activators for forming Ni-aluminide/Ni hybrid coatings on creep resistant steels by pack cementation at temperatures below 700 °C. The pack powder mixtures analysed and used in subsequent experiments consisted of Al, Al2O3 and one of chloride or fluoride salts which included NH4Cl, AlCl3 (anhydrous), NaCl, NH4F and AlF3. Pack Al content was varied from 0.5 to 16 wt% to show its effects on aluminising kinetics at constant temperature and time. The aluminised layers were analysed using XRD, SEM and EDS. It was demonstrated that the partial pressure of AlX (X = Cl and F) vapour species generated in packs can be used to measure the activating strengths of different chloride and fluoride salts. This is particularly true in the case of chloride salt activated packs in which aluminising kinetics was shown to be higher in packs activated by salts with higher activating strengths. Fluoride salts are not suitable activators in the low temperature range concerned because a solid AlF3 phase layer would deposit on the surface, which would then prematurely terminate the aluminising process.

Creep Behavior of Ferritic Steels after Heat-treated

The creep resistant ferritic steels have a microstructure with fine stable alloy carbides that impede the movement of the dislocations; however it is inevitable that during long periods of service or very critical conditions, microstructural changes that are responsible for the loss of material strength, occur.This type of steel is used for pressurized pipes, boiler tubes and heat exchangers at temperatures of up to 863K. It exhibits good mechanical strength at high temperature due to the precipitation of carbides but for the low content of chromium present small corrosion resistance at temperatures around 823K.The objective of this work is to study the creep behavior of 1Cr 0.5Mo steel and compare the resistance when -prior to service- it is subjected to different heat treatments that enhance its conditions of service.Tensile creep tests are performed at stresses between 131 and 205MPa and temperatures between 843 and 923K.The microstructural change is analyzed in original material, after heat treatment and after test creep. Samples are characterized with light optical microscopy (LOM) and scanning electron microscopy (SEM). The micro hardness of the phases present in the different specimens is also measured.

A Material Genomic Design of Advanced High Performance, Non-Corroding Steels for Ambient and High Temperature Applications

This work presents an artificial intelligence based design of a series of novel advanced high performance steels for ambient and high temperature applications, following the principle of the materials genome initiative, using an integrated thermodynamics/kinetics based model in combination with a genetic algorithm optimization routine. Novel steel compositions and associated key heat treatment parameters are designed both for applications at the room temperature (ultra-high strength maraging stainless steel) and at high temperatures (ferritic, martensitic and austenitic creep resistant steels). The strength of existing high end alloys of aforementioned four types are calculated according to the corresponding design criteria. The model validation studies suggest that the newly designed alloys have great potential in outperforming existing grades.

Halide Activators Suitable for Low Temperature Pack Aluminisation of Electroplated Nickel on Creep Resistant Ferritic Steels

This study aims to identify suitable activators for pack aluminising electroplated Ni in order to provide processing flexibility for forming Ni-aluminide/Ni hybrid coatings on creep resistant ferritic steels at temperatures below 700 °C. Experiments are made using NH4Cl, AlCl3(anhydrous), NaCl and NH4F activated packs prepared from Al and Al2O3powders. Both NH4Cl and AlCl3(anhydrous) are found to be suitable activators. NaCl may be used as an activator if the amount of Al added is sufficiently high. But, the fluoride salts cannot be used as activators in the low temperature range concerned here due to their strong tendency to deposit AlF3.

Influence of Al on scale formation and growth kinetics of 10 wt.% Cr creep resistant ferritic steels at 650 °C in air

Two 10wt.% Cr steels containing 1.90 and 2.33wt.% Al respectively were oxidised in air at 650°C. The scales formed on the steel with 2.33wt.% Al were θ-Al2O3 and α-Al2O3 with differing parabolic rate constants. The scales formed on the steel containing 1.90wt.% Al consisted of multiple layers of Fe-oxides and the amount of oxidation followed logarithmic kinetics Δmt=klln(t/t0+1). An empirical model was obtained for calculating the minimum amount of Al needed for forming an external Al2O3 scale on Fe–10Cr–xAl (wt.%) alloys in air in 650–1050°C.

Low temperature pack aluminising kinetics of nickel electroplated on creep resistant ferritic steel

A systematic study was made to determine the low temperature (<700°C) growth kinetics of the Ni2Al3 layer formed by the pack aluminising process on the surface of nickel electroplated on creep resistant ferritic steels. The aluminising powder packs were prepared using Al powder as a source for depositing Al, Al2O3 powder as inert filler and anhydrous AlCl3 as activator. The microstructures of the coatings formed were characterized by means of SEM/EDS and XRD. The results obtained were used to quantify the effects of aluminising temperature (T), time (t) and amount of Al powder (W) added into the pack powders on the growth kinetics of the Ni2Al3 layer (h). It was confirmed that the thickness of the Ni2Al3 layer formed can be related to the processing parameters by h=k0W1/2t1/2exp[−Ea/(RT)]. The values of the pre-exponential constant k0 and the activation energy Ea were determined. Additional experiments were made to demonstrate the effects of purity of the Al2O3 powder and amount of halide activator (AlCl3) added into the powder packs on the growth kinetics of the Ni2Al3 layer.

Thermal stability enhancement of hybrid Ni2Al3/Ni coatings on creep resistant ferritic steels by a mechanism of thermodynamically constrained interdiffusion

Based on the thermodynamic analysis, experimental work is undertaken to investigate the thermal stability of the hybrid Ni2Al3/Ni coating with a thin inner Ni layer formed on creep resistant ferritic steel. The results showed that once the coating is transformed by interdiffusion to a NiAl/Al-rich-Fe-base-alloy layer structure, thermal stability of the newly formed outer NiAl layer at 650°C can be enhanced by the constraint imposed on inward Al diffusion by Al activity difference at the outer/inner layer interface. The Fe diffusion into the newly formed outer NiAl layer is restrained by its solubility limit in the newly formed outer NiAl layer.

Air and Wet Air Oxidation of 9Cr-3Co-3W Creep Resistant Ferritic Steel at 650 °C

The new experimental creep resistant ferritic steel of the 9Cr-3Co-3W type was oxidised at 650 °C in air and wet air. The oxidation kinetics was measured by intermittent weight measurement. The scales formed were analysed using techniques of XRD, SEM and EDS. The results showed that the oxidation rate was more than a magnitude faster in wet air than in air. The oxidation kinetics in air obeyed the parabolic rate law of oxidation only in a limited oxidation period of up to 1726 h whereas it did not follow any power rate law of oxidation in wet air. The steel cannot form a protective Cr2O3 scale either in air or in wet air at 650 °C. Instead, the scale formed in air consisted of an outer (Fe0.6Cr0.4)2O3 layer and an inner Cr-rich (Fe,Cr)2O3 layer containing Cr2O3 particles, but in wet air it consisted of an outer Fe3O4 layer and an inner (Fe,Cr)3O4 layer.

Observations of native oxides on ion beam polished heterogeneous Cr–Mo weld TEM foil specimens

Native oxide films formed on argon ion beam polished transmission electron microscopy (TEM) specimens have been characterised. TEM studies of heterogeneous 9Cr –1Mo (P91) and 2.25Cr –1Mo (P22) creep resistant ferritic steel welds reveal the existence of oxide layers formed on thin TEM specimen foils after short term exposure to laboratory air. These films which are grown on as-polished specimens form a rim of oxide a few nanometres wide around the perforation edges of both P91 parent alloy and P22 weld metal samples. Selected area diffraction patterns indicate an epitaxial relationship exists between oxide layers and the steel substrates which conform to the Nishiyama– Wassermann orientation relationship. As a result, diffraction and mass thickness contrast and moiré patterns are observed in transmission electron micrographs. The confusing and obstructive nature of the additional contrast can lead to the loss of, or potential misinterpretation of, important information in images. Efforts to minimise the occurrence of these oxide layers during sample preparation and transport between preparation equipment and the microscope have been made.

Effect of Pack Al Content on Growth Kinetics at 650 °C of Ni2Al3 Coating Layer on Nickel-Electroplated Creep Resistant Ferritic Steels

The hybrid Ni2Al3/Ni coating was formed on creep resistant ferritic steels by firstly nickel electroplating and then partially aluminising the Ni layer at 650 °C by pack cementation process using powder mixtures of Al, AlCl3and Al2O3. The effect of pack Al content (W) on growth kinetics of the outer Ni2Al3layer of the coating was investigated by varying it from 2 to 10 wt% whilst keeping the pack AlCl3content constant at 2 wt% and aluminising conditions at 650°C/4h. It was revealed that, once W was above a minimum level, the growth of the outer Ni2Al3layer thickness depended linearly on W1/2. The possible reasons for such growth kinetics were discussed.

A phenomenological model for lifetime design of Ni2Al3/Ni hybrid coating formed on creep resistant ferritic steels

The phase layer transformation kinetics in the Ni2Al3/Ni hybrid coating formed on creep resistant steel P92 has been studied via a series of prolonged isothermal annealing experiments at 650 °C. All the intermediate phase layers of NiAl, Ni5Al3 and Ni3Al formed in the coating by interdiffusion during isothermal annealing process. The phase layers of NiAl and Ni3Al formed at the very beginning of isothermal annealing at the interface between Ni2Al3 and Ni, but the Ni5Al3 phase layer formed at the interface between the NiAl and Ni3Al phase layer only at an annealing time at which the outer Ni2Al3 phase layer was completely consumed. The growth and consumption of the Ni2Al3 and NiAl phase layers and the growth of the Ni3Al5 phase layer were all parabolic, but the growth of the Ni3Al phase layer obeyed the power rate law d = kt1/n. The growth kinetics of an intermediate phase layer was found to be faster than the kinetics of its subsequent consumption. The rate constants in both the growth and consumption kinetics need to be determined for each of the intermediate phase layers at a particular temperature. The lifetime of the coating with an outer Ni2Al3 phase layer of any specified initial thickness and a sufficiently thick inner Ni layer can then be estimated using the lifetime design model delineated in this study.