A decrease in number density of MX carbonitrides due to Z phase precipitation is the most serious factor for the degradation in creep strength of Gr.92 steel at long times. The degradation in RA of Gr.92 steel at long times is caused by the boron nitride particles and by very short normalising time. The rupture ductility is further reduced in steam. The reduction of boundary hardening due to poor M23C6 carbides along PAGBs in HAZ is the most important for the significant decrease in time to rupture of HAZ in welded joints of Gr.92 steel. The formation of protective Cr2O3-rich scale is achieved on the surface of 9Cr-3WVNb steel by the pre-oxidation treatment in argon gas containing small amount of oxygen. The application of pre-oxidation treatment in argon gas to Gr.92 steel is a future challenge.

Stabilized austenitic stainless steels (ASS) are critical for high-temperature applications in power plants. While their creep resistance depends on precipitation strengthening, the influence of grain size (GS) remains a subject of debate, with conflicting reports on its linear or non-linear effects. In this study, Nb-stabilized 25Cr-20Ni ASS (HR3C) with three grain size levels, including a fine-grain state and those meeting the code requirements (25 & micro;m, 53 & micro;m, and 80 & micro;m), was investigated under dislocation-dominated creep conditions (700 degrees C/150 MPa and 750 degrees C/ 110 MPa). The results reveal that the high-temperature tensile strength follows an inverse linear relationship with GS; however, the creep rupture life exhibits a non-monotonic trend, with the 53 & micro;m specimen consistently showing the longest life at both temperatures. Microstructural characterization reveals that increasing GS reduces the fraction of primary MX (NbC) and Z-phases (NbCrN). This suppression of primary phases enriches the matrix nitrogen (N) concentration, which in turn facilitates the dense precipitation of nanoscale secondary Zphase, providing superior synergistic strengthening through dislocation shearing and Orowan pinning. However, excessive nitrogen availability in the coarsest grains (80 & micro;m) triggers accelerated coarsening and clustering of the secondary Z-phase. This morphological degradation significantly increases the local inter-particle spacing, leading to a drastic loss of Orowan resistance and a transition in dislocation gliding mode. Furthermore, while fine-grained specimens (25 & micro;m) suffer from premature grain boundary embrittlement due to continuous M23C6 chains, the performance of coarse-grained specimens is limited by the loss of intragranular strengthening efficiency. This study clarifies the underlying mechanisms for grain size-dependent creep behavior and emphasizes that an optimal grain size is essential to balance grain boundary integrity with the stability of intragranular phases, providing a critical criterion for the microstructural design of high-performance stabilized austenitic steels.

The stress exponent n1 of the minimum creep rate ε̇ min for Gr.91 steel with tempered martensitic microstructure is evaluated to be n1 ≳ 10 and n1 ≈ 4 to 7 at high and low stresses, respectively, at 823 (550) to 923 K (650 °C). The time to minimum creep rate tm increases with decreasing stress, while the strain to minimum creep rate εm is constant of 2.2 pct at high stresses of 160 to 140 MPa and decreases with decreasing stress below 140 MPa at 873 K (600 °C). The constant εm suggests homogeneous creep deformation, while the decrease in εm with decreasing stress suggests localized creep deformation, presumably in the vicinity of prior austenite grain boundaries (PAGBs). The n1 is evaluated to be 11 and 4 in the high and low stress regions, respectively, at 823 K (550 °C) for 2.25Cr-1Mo steel specified as ASTM A542/A542M with tempered martensitic microstructure. There is no sign of reduction of n1 in the low stress region in 2.25Cr-1Mo steel specified as JIS STBA 24 with ferrite/pearlite microstructure containing low dislocation density. The localized creep deformation is responsible for the reduction of n1 and apparent activation energy Q for the ε̇ min in the low stress region, although the creep deformation is due to the mechanism of dislocation creep in both the high and low stress regions. High density of dislocations in the tempered martensitic microstructure promotes the recovery of dislocations in the vicinity of PAGBs during creep, resulting in localized creep deformation.

The reasonable procedures for estimating long-term creep rupture strength are investigated for ASME grade 91 steel using creep rupture data in the National Institute for Materials Science (NIMS) creep data sheet. The target was set to the estimation of 500,000 h creep rupture strength at 550 degrees C, taking a strong demand for fast reactor applications of this steel into account, and the estimation of 100,000 h creep rupture strength at 600 degrees C for ultrasupercritical (USC) applications. A regression analysis is carried out using rupture data located inside the time-temperature-precipitation (TTP) curve for Z phase precipitation and below the nose temperature of the TTP curve. Inside the TTP curve, the degradation takes place. Below the nose temperature of the TTP curve, the precipitation of the Z phase is accelerated with increasing temperature. The second-order regression analysis by the Larson-Miller method gives us the estimated values of 100.1 MPa and 73.4 MPa at 550 degrees C and 500,000 h and at 600 degrees C and 100,000 h, respectively. These values are compared with those by the analysis using rupture data above 500 h, those by the Region splitting analysis method, and those by the Multiregion analysis method.

Better correlation of the creep life tr with the minimum creep rate $\dot \varepsilon $epsilon(center dot)min is obtained by (tr/epsilon m) = 2.4 /$\dot \varepsilon $epsilon(center dot)min as a modified Monkman-Grant relation for both Gr.91 and Gr.92, where epsilon m is the strain to minimum creep rate. Using the creep deformation parameters, the tr is described as tr = g tm, where g is a constant slightly depending on stress and tm is the time to minimum creep rate, and as tr = 1.5/[$\dot \varepsilon $epsilon(center dot)min (dln $\dot \varepsilon $epsilon(center dot)/d $\varepsilon $epsilon)], where dln $\dot \varepsilon $epsilon(center dot)/d $\varepsilon $epsilon is an increase in creep rate by strain in the accelerating region. Using the equation tr = g tm, the creep life is predicted by evaluating the tm after carrying out a short-term creep test of about 30% of the tr without any stress extrapolation. The equation tr = g tm also enables us to predict the degradation in creep rupture strength and the heat-to-heat variation in tr.

The controlling factors for the creep rates in the transient region and for the relating parameters, such as the time to minimum creep rate tm, strain to minimum creep rate epsilon m and minimum creep rate epsilon(center dot)min, have been investigated for Gr.91 steel at 550, 600 and 650 degrees C. The creep rates at 0.2 h, an early stage of transient region, decrease with increasing 0.2 % proof stress. The increase in tm with decreasing stress becomes less significant at low stresses, similar as the time to rupture. The epsilon m is approximately a constant of 2.2 % at high stresses above a critical stress, while it decreases with decreasing stress below the critical stress by the localization of creep deformation, presumably near prior austenite grain boundaries. The strong heats exhibit large tm but small epsilon m. At high stress level above the critical stress, the epsilon(center dot)min is inversely proportional to the tm as epsilon(center dot)min = 1.2 x 10- 2/tm. At stresses below the critical stress, both the tm and epsilon m change with stress and the epsilon(center dot)min is given by epsilon(center dot)min = 0.54 (epsilon m/tm).

The influence of hydrogen produced during steam oxidation on the creep deformation and rupture ductility has been investigated for Gr.92 at 650 degrees C by comparing creep data in steam with those in air. The total strain epsilon r is definitely smaller in steam than in air, while the time to minimum creep rate t m , strain to minimum creep rate epsilon m , minimum creep rate epsilon (center dot) min and time to rupture t r are approximately the same between in steam and in air. Hydrogen introduced into the specimens from the surface oxide scale formed during creep test in steam is expected to be always uniformly distributed throughout the creep specimens with 5 mm gauge diameter during creep tests at 650 degrees C because of its high diffusion rate. The smaller epsilon r in steam than in air and substantially the same epsilon m between in steam and in air are reasonably explained by the mechanism of hydrogen-enhanced deformation-induced vacancy formation model. Large strains in the later stage of accelerating creep region, together with the presence of hydrogen during creep in steam, enable the formation of high density of vacancies and microvoids, which reduces the total strain epsilon r . The amount of strain seems to be too small in the transient creep region, suggesting that vacancies scarcely form and hence no effect of steam environment on the t m and epsilon m .

The degradation of the long-term rupture strength of ASME Grade 122 steel occurs earlier than that of Grade 92 steel. To investigate the reasons for this phenomenon, the long-term creep curves of Grade 122 steel pipe, plate, and tube product forms were analyzed by applying an exponential law to the temperature, stress, and time parameters. The activation energy (Q), activation volume (V), and Larson–Miller constant (C) were obtained as functions of creep strain. All Q, V, and C (QVC) decreased simultaneously with an increase in creep strain during the transient creep in a data group (Gr.IIIa), where an unexpected drop in the long-term rupture strength was experienced. Metallurgical considerations of the variations in QVC meant that “heterogeneous recovery and heterogeneous deformation” (HRHD) should occur during the simultaneous decreases in QVC. The Z-phase is easily formed by the consumption of the strengthening particles of MX in the HRHD zone, which causes the degradation of the long-term strength of Grade 122 steel. The higher hardness of Grade 122 steels promotes the coarsening of the Laves phase particles and, in addition to this, the amount of MX inside the subgrains is estimated to be less than Grade 92 steel, which cause severe HRHD and the resultant degradation in rupture strength compared to Grade 92 steel. In a data group subjected to lower stresses than those of Gr.IIIa, the degradation rate is mitigated, and a deformation mechanism was proposed. The improvement in the long-term rupture strength of Grade 122 steel was also discussed.

A modified Monkman-Grant equation, which can provide a more accurate means of predicting creep rupture life than the standard Monkman-Grant formula, has been investigated for Gr.92 using creep data in the NIMS Creep Data Sheets at 550 to 750 degrees C. The t(r) versus minimum creep rate $$\dot \varepsilon $$e(min) plot, which is called the Monkman-Grant relation, deviates downward at low stresses and long times. Better correlation of the t(r) with the $$\dot \varepsilon $$e(min) is obtained by the replacement of t(r) with (t(r) /e(r)), where e(r) is the total or rupture strain. The (t(r) /e(r)) is inversely proportional to the $$\dot \varepsilon $$e(min) over a wide range of stress, temperature and test duration, and the magnitude of data scattering is only a little bit even at low stresses and long times. The creep life of Gr.92 can be predicted by evaluating the $$\dot \varepsilon $$e(min), together with the e(r) evaluated from the stress and or $$\dot \varepsilon $$e(min) dependence.

ABSTRACT The role of BN, AlN and MnS particles on the degradation in creep life and rupture ductility has been investigated for 9 to 12Cr martensitic steels and 1Cr bainitic steel mainly at 550 to 650 oC. The BN particles form in Gr.92 and Gr.122 during normalising at around 1100 oC. The BN particles have nothing to do with the degradation in creep life. The AlN particles precipitate in the high-Al heats of 12Cr-1Mo-1 W-0.3 V steel during creep, reducing dissolved nitrogen and fine nitrides beneficial for the creep strength and degrading the creep life. The MnS particles have nothing to do with the degradation in creep life of 1Cr-1Mo-0.25 V steel. The BN, AlN and MnS particles are responsible for the degradation in reduction of area of the steels by accelerating the formation of creep voids at interfaces between the particles and alloy matrix.

A modified version of the Monkman-Grant equation, which can provide a more accurate means of predicting creep rupture life than the standard Monkman-Grant formula, has been investigated for Gr.91 using creep data in the NIMS Creep Data Sheets at 450-725 degrees C. The maximum time to rupture tr was 1.2 x 105 h. The tr versus minimum creep rate emin plot, which is called the Monkman-Grant relation, deviates downward at low stresses and long times. The difference between the maximum and minimum tr at a same emin becomes more significant with decreasing stress and increasing test duration. Better correlation of the tr with the emin is not obtained by the replacement of tr with (tr/er), where er is the total or rupture strain. The magnitude of data scattering is larger in the (tr/er) versus emin plot than in the tr versus emin plot. The (tr/em), where em is the strain to minimum creep rate, is inversely proportional to the emin over a wide range of stress, temperature and test duration and the magnitude of data scattering is only a little bit even at low stresses and long times. The (tr/em) versus emin plot gives us more reliable relation for Gr.91 than the tr versus emin and (tr/er) versus emin plots.

Equations for predicting creep rupture life of grade 91 steel have been revised officially in Japan and European Union. First, the revised equations are examined with a large creep rupture database on the steel. The database consists of 2066 data points, and its longest creep rupture life is 232,833 h crept at 600 degrees C. The equations deviate from data points of the database in a long-term region beyond 30,000 h. Decrease in activation energy for creep rupture life in the long-term region is the cause of the deviation. A multi-region analysis of the database can be a probable way of reducing the deviation. However, some modifications of the analysis are necessary when applying it to the large database containing data points of multiple heats. A trial of the modifications is made in the second part of this paper. The equations obtained through the trial can well represent all the data points including the ones in the long-term region.

New models for creep life equation based on creep strain analysis have been proposed for Gr.91. Boundary hardening is the most important strengthening mechanism for long-term creep of martensitic 9 to 12Cr steels and austenitic steels. The addition of boron stabilizes fine distributions of M23C6 carbides at and near prior austenite grain boundaries and improves long-term creep rupture strength of 9Cr steel at 650°C. Mechanisms responsible for creep strength loss at long times have extensively been investigated for 9 to 12Cr steels. New alloy-design concepts have been proposed to develop advanced 9 to 12Cr steels and advanced austenitic steels.

Soluble boron stabilises fine distributions of M23C6 carbides near grain boundaries (GBs) in 9Cr steel during creep, enhancing GB precipitation hardening, and suppresses the Type IV fracture in welded joints at 650oC. Excess addition of boron and nitrogen produces BN inclusions at high temperature, degrading creep strength and rupture ductility. A 9Cr-3W-3Co-0.2V-0.05Nb steel with 130 ppm boron and 80 ppm nitrogen, designated martensitic 9Cr steel strengthened by boron and nitrides (MARBN), exhibits not only the much higher creep rupture strength of base metal than Gr.92 but also no Type IV fracture in welded joints at 650oC. The boron and nitrogen concentrations in MARBN are located just inside the phase boundary for BN at normalising heat treatment temperature in the phase diagram, where we can obtain enough soluble boron and soluble nitrogen but no BN inclusion. The oxidation resistance is improved by the pre-oxidation treatment in argon gas.

To investigate the formation process of the Z-phase, which lowers the long-term rupture strength of high-Cr martensitic steel, the creep curves of Grades T91, T92, and P92 were analyzed along with the experimental steels of 9Cr-1W and 9Cr-4W by applying an exponential law to the temperature, stress, and time parameters. The activation energy (Q ), activation volume (V ), and Larson-Miller constant (C ) were obtained as functions of creep strain. At the beginning of creep, sub-grain boundary strengthening occurs due to dislocations that are swept out of the sub-grains, which is followed by strengthening due to the rearrangement of M23C6 and the precipitation of the Laves phase. After Q  reaches a peak, heterogeneous recovery and subsequent heterogeneous deformation begin at an early stage of transient creep in the vicinity of several of the weakest boundaries due to coarsening of the precipitates. This activity triggers an unexpected degradation in strength due to the accelerated formation of the Z-phase. Stabilization of M23C6 and the Laves phase is important for mitigating the degradation of the long-term rupture strength of high-strength martensitic steel. The stabilization of the Laves phase is especially important for the Cr-Mo systems because Fe2Mo is easily coarsened at ~600 °C as compared to Fe2W. Lowering the hardness and Si content also prevents excess hardening due to the Laves phase, which also mitigates the degradation. The online monitoring of creep curves and the QVC  analysis render it possible to detect signs of long-term degradation under targeted conditions within a relatively short period.

ABSTRACT The creep rupture ductility has been investigated for Gr.122 at 550 to 650 oC for up to 112,247.8 h. The obtained results are compared with those for Gr.91 and Gr.92. The degradation in reduction of area (RA) in Gr.122 is similar to that in Gr.92 but more significant than that in Gr.91. The boron nitride inclusions are responsible for the degradation in RA in Gr.92 and Gr.122. The creep ductility parameter λ of Gr.122 decreases with increasing test duration to small values less than 5 at 20,000 h to 30,000 h and then increases steeply to about 20 at long times at 650 oC. The RA and total elongation of Gr.122 decrease with increasing test duration to small values but then turn to increase at 625 oC and 650 oC and long times.

The creep deformation behaviour and its effect on the creep life and rupture ductility are investigated for W-Mo-balanced 9Cr steels at 600 degrees C to 700 degrees C. The creep life is described as t(r) = 1.5/[(min) (dln/d)], where (min) and dln/d reflect the creep deformation behaviour in the transient and acceleration creep regions, respectively. The creep life is also correlated with the time to minimum creep rate t(m) as t(r) = gt(m), where g slightly increases with decreasing stress. The logarithm of creep rate linearly increases with strain in the initial stage of acceleration creep region. This is followed by a decrease in dlnd with strain at high stresses, resulting in ductile failure at large strains, while it is followed by an increase in dln/d with strain at low stresses, resulting in rupture at small strains.

The influence of boron nitride (BN) particles, W and Mo concentrations and creep test conditions of stress, temperature and test duration on the creep ductility parameter lambda in UK R5 has been investigated for five heats of W-Mo-balanced 9Cr steel with 1.5% Mo equivalent at 600, 650 and 700(o)C. With increasing time to rupturet(r), the lambda at first decreases only a little bit, then increases for up to a time showing a maximum peak and finally significantly decreases. The BN particles are responsible for the decreases in lambda at long times by accelerating the formation of creep voids. The evaluation of creep rupture ductility based on the lambda should be restricted to the long-time region at low stresses after reaching a maximum peak in lambda, where the lambda decreases with increasingt(r)mainly by a decrease in total elongation.

ABSTRACT The creep rupture ductility has been investigated for Gr.91 and Gr.92 at 550 to 700°C for up to 118,648.8 h. The degradation in the reduction of area (RA) is more significant in Gr.92 than in Gr.91 at 550°C to 700°C. The boron nitride particles are responsible for the degradation in RA in Gr.92 at long times by accelerating the formation of creep voids. The creep ductility parameter λ decreases to small values below 5 for Gr.92 at long times at 650°C and 700°C, while that is larger than 5 for Gr.92 at 550°C and 600°C and for Gr.91 at 550°C to 700°C. The data exhibiting λ > 5 and λ < 5 in the λ versus time diagram correspond to the data in the necking and creep void dominant regions, respectively, in the RA versus total elongation diagram.