Impact of Temperature on Low-Cycle Fatigue Characteristics of the HR6W Alloy

This paper presents the results of tests conducted on the HR6W (23Cr-45Ni-6W-Nb-Ti-B) alloy under low-cycle fatigue at room temperature and at 650 °C. Fatigue tests were carried out at constant values of the total strain ranges. The alloy under low-cycle fatigue showed cyclic strengthening both at room temperature and at 650 °C. The degree of HR6W strengthening described by coefficient n’ was higher at higher temperatures. At the same time, its fatigue life Nf at room temperature was, depending on the range of total strain adopted in the tests, several times higher than observed at 650 °C.


Introduction
In Poland, electricity is mainly produced in conventional power plants, i.e., those in which the steam needed to drive the generators comes from coal-fired or lignite-fired boilers. These power plants emit significant amounts of CO 2 to the atmosphere. With European policy aiming to reduce CO 2 emissions into the atmosphere, the Polish energy sector needs to adapt to European regulations. The reduction in CO 2 emissions can be achieved, among others, by increasing the operating parameters of supercritical boilers with constantly improved advanced materials which have better properties at elevated temperatures [1][2][3][4][5][6][7]. Such materials include the HR6W alloy, which is mainly used for superheater tubes in power boilers, thick-walled tubular elements, reactors, as well as in chemical plants.
What is required from materials under high temperatures and pressures, which occur in boilers under increased operating parameters, is high creep resistance and thermalmechanical and low-cycle fatigue resistance as well as high corrosion resistance . Unfortunately, in contrast to the large number of generally available data in the field of research on basic mechanical properties at room temperature and elevated temperature, creep, and corrosion resistance of the HR6W alloy [1][2][3][4][5][6][7][8][9][10][11][12][13][15][16][17][18][19], there is less data on thermomechanical and low-cycle fatigue testing of this alloy [2,6,14,20]. The benefits of using data obtained in low-cycle fatigue tests at elevated temperature for prediction of durability under thermo-mechanical fatigue loading for the HR6W alloy were demonstrated by Noguchi et al. [20]. The authors of the paper [20] investigated the relationship between the fatigue life of HR6W in isothermal fatigue tests and the fatigue life obtained in thermomechanical fatigue tests. The tests were carried out in the temperature range between 100 and 700 • C. Researchers pointed out a similarity between these types of tests for this alloy. Therefore, it seems justified to estimate the strength of the alloy under thermomechanical fatigue based on less complex, more feasible isothermal fatigue tests. In the work presented, a study of low-cycle fatigue at room temperature and at 650 • C constituted a primary research experiment. This research is meant as a contribution to the existing knowledge on the cyclic behavior of the HR6W alloy with the properties determined at 650 • C. Therefore, the test results obtained can be useful for more precise forecasting of the fatigue life of power plant components.

Materials and Methods
The material for low-cycle fatigue testing was the as-delivered HR6W (23Cr-45Ni-6W-Nb-Ti-B) alloy taken from a fragment of a thick-walled pipe. The chemical composition of the alloy was determined on the basis of the X-ray energy spectrum as presented in Figure 1a. A microanalysis of the chemical composition of the alloy was performed using the EDS Oxford LINK ISIS-300 X-ray (Manufacture: Oxford Instruments plc) microanalyzer coupled with a JEOL JSM 5800 LV scanning electron microscope (Manufacture: Oxford Instruments plc).

Materials and Methods
The material for low-cycle fatigue testing was the as-delivered HR6W (23Cr-45Ni-6W-Nb-Ti-B) alloy taken from a fragment of a thick-walled pipe. The chemical composition of the alloy was determined on the basis of the X-ray energy spectrum as presented in Figure 1a. A microanalysis of the chemical composition of the alloy was performed using the EDS Oxford LINK ISIS-300 X-ray (Manufacture: Oxford Instruments plc) microanalyzer coupled with a JEOL JSM 5800 LV scanning electron microscope (Manufacture: Oxford Instruments plc). In the as-delivered condition, the material has an austenite structure, which is characterized by a non-uniform grain size. Numerous precipitations of M23C6 carbides were observed at the grain boundaries ( Figure 1b).
Fatigue test samples ( Figure 2) were made of material taken from the ϕ360 mm × 75 mm pipe. Mechanical properties were tested at room temperature and at 650 °C. The test results are presented in Table 1. In the as-delivered condition, the material has an austenite structure, which is characterized by a non-uniform grain size. Numerous precipitations of M 23 C 6 carbides were observed at the grain boundaries ( Figure 1b).

Carbides M23C6
Fatigue test samples ( Figure 2) were made of material taken from the φ360 mm × 75 mm pipe. constituted a primary research experiment. This research is meant as a contribution to the existing knowledge on the cyclic behavior of the HR6W alloy with the properties determined at 650 °C. Therefore, the test results obtained can be useful for more precise forecasting of the fatigue life of power plant components.

Materials and Methods
The material for low-cycle fatigue testing was the as-delivered HR6W (23Cr-45Ni-6W-Nb-Ti-B) alloy taken from a fragment of a thick-walled pipe. The chemical composition of the alloy was determined on the basis of the X-ray energy spectrum as presented in Figure 1a. A microanalysis of the chemical composition of the alloy was performed using the EDS Oxford LINK ISIS-300 X-ray (Manufacture: Oxford Instruments plc) microanalyzer coupled with a JEOL JSM 5800 LV scanning electron microscope (Manufacture: Oxford Instruments plc). In the as-delivered condition, the material has an austenite structure, which is characterized by a non-uniform grain size. Numerous precipitations of M23C6 carbides were observed at the grain boundaries ( Figure 1b).
Fatigue test samples ( Figure 2) were made of material taken from the ϕ360 mm × 75 mm pipe. Mechanical properties were tested at room temperature and at 650 °C. The test results are presented in Table 1. Mechanical properties were tested at room temperature and at 650 • C. The test results are presented in Table 1. Fatigue tests for the low-cycle HR6W alloy were carried out on the MTS-810 testing machine ( Figure 3).The tests were run in strain control (extensometer MTS-632-11C-20 temp. room, extensometer MTS-632-14B-05 temp. 650 • C). Sinusoidal loading cycles were applied with a cycle asymmetry ratio R = −1 and load frequency f = 0.1 Hz. The fatigue tests were carried out at room temperature and at 650 • C. The tests were carried out for five ranges of total deformation ∆ε t = 0.6; 0.7; 0.8; 1.0 and 1.2%. The samples were induction heated using a Hüttinger TIG-300 heater ( Figure 4). Low-cycle fatigue tests at elevated temperatures were carried out with the use of a FLIR SC6000 thermal imaging camera, which allowed continuous temperature control during the tests. Fatigue tests for the low-cycle HR6W alloy were carried out on the MTS-810 testing machine ( Figure 3).The tests were run in strain control (extensometer MTS-632-11C-20 temp. room, extensometer MTS-632-14B-05 temp. 650 °C). Sinusoidal loading cycles were applied with a cycle asymmetry ratio R = −1 and load frequency f = 0.1 Hz. The fatigue tests were carried out at room temperature and at 650 °C. The tests were carried out for five ranges of total deformation Δεt = 0.6; 0.7; 0.8; 1.0 and 1.2%. The samples were induction heated using a Hüttinger TIG-300 heater ( Figure 4). Low-cycle fatigue tests at elevated temperatures were carried out with the use of a FLIR SC6000 thermal imaging camera, which allowed continuous temperature control during the tests.

Results and Discussion
During the fatigue tests, changes in stress amplitude (σan) and strain (εt) versus the number of cycles (N) were constantly recorded. On this basis, hysteresis loops characteristic for selected stages of the fatigue process were developed, i.e., the initial phase, the saturation phase (characterized by a fixed value of the stress amplitude σan), and the fracture phase for the sample. Examples of such graphs for the selected strain range ∆εt = 0.6% are shown in Figures 5 and 6. In low-cycle tests, fatigue life Nf was defined as the number of cycles to failure for the sample.  Fatigue tests for the low-cycle HR6W alloy were carried out on the MTS-810 testing machine ( Figure 3).The tests were run in strain control (extensometer MTS-632-11C-20 temp. room, extensometer MTS-632-14B-05 temp. 650 °C). Sinusoidal loading cycles were applied with a cycle asymmetry ratio R = −1 and load frequency f = 0.1 Hz. The fatigue tests were carried out at room temperature and at 650 °C. The tests were carried out for five ranges of total deformation Δεt = 0.6; 0.7; 0.8; 1.0 and 1.2%. The samples were induction heated using a Hüttinger TIG-300 heater ( Figure 4). Low-cycle fatigue tests at elevated temperatures were carried out with the use of a FLIR SC6000 thermal imaging camera, which allowed continuous temperature control during the tests.

Results and Discussion
During the fatigue tests, changes in stress amplitude (σan) and strain (εt) versus the number of cycles (N) were constantly recorded. On this basis, hysteresis loops characteristic for selected stages of the fatigue process were developed, i.e., the initial phase, the saturation phase (characterized by a fixed value of the stress amplitude σan), and the fracture phase for the sample. Examples of such graphs for the selected strain range ∆εt = 0.6% are shown in Figures 5 and 6. In low-cycle tests, fatigue life Nf was defined as the number of cycles to failure for the sample.

Results and Discussion
During the fatigue tests, changes in stress amplitude (σ an ) and strain (ε t ) versus the number of cycles (N) were constantly recorded. On this basis, hysteresis loops characteristic for selected stages of the fatigue process were developed, i.e., the initial phase, the saturation phase (characterized by a fixed value of the stress amplitude σ an ), and the fracture phase for the sample. Examples of such graphs for the selected strain range ∆ε t = 0.6% are shown in Figures 5 and 6. In low-cycle tests, fatigue life N f was defined as the number of cycles to failure for the sample.  The process of cyclic hardening of the HR6W alloy is clearly visible in the above hysteresis loops obtained in low-cycle fatigue tests at room temperature and at elevated temperature (650 °C) for the total strain range of 0.6%. It is manifested in an increase between the stress values read at the tops of the hysteresis loop corresponding to the beginning of the test and the saturation state. A particularly notable amplification occurs for samples tested at 650 °C, where the stress value increases more than twice, from a level of about 160 MPa for the initial loop to 400 MPa corresponding to the saturation loop. Moreover, the saturation stresses recorded in the tests carried out at 650 °C were higher than those recorded at room temperature ( Figure 7b).
Both at room temperature and at elevated temperature, a distinct change in the shape of the hysteresis loop was observed ( Figures 5-7), which is linked directly with material  The process of cyclic hardening of the HR6W alloy is clearly visible in the above hysteresis loops obtained in low-cycle fatigue tests at room temperature and at elevated temperature (650 °C) for the total strain range of 0.6%. It is manifested in an increase between the stress values read at the tops of the hysteresis loop corresponding to the beginning of the test and the saturation state. A particularly notable amplification occurs for samples tested at 650 °C, where the stress value increases more than twice, from a level of about 160 MPa for the initial loop to 400 MPa corresponding to the saturation loop. Moreover, the saturation stresses recorded in the tests carried out at 650 °C were higher than those recorded at room temperature (Figure 7b).
Both at room temperature and at elevated temperature, a distinct change in the shape of the hysteresis loop was observed ( Figures 5-7), which is linked directly with material The process of cyclic hardening of the HR6W alloy is clearly visible in the above hysteresis loops obtained in low-cycle fatigue tests at room temperature and at elevated temperature (650 • C) for the total strain range of 0.6%. It is manifested in an increase between the stress values read at the tops of the hysteresis loop corresponding to the beginning of the test and the saturation state. A particularly notable amplification occurs for samples tested at 650 • C, where the stress value increases more than twice, from a level of about 160 MPa for the initial loop to 400 MPa corresponding to the saturation loop. Moreover, the saturation stresses recorded in the tests carried out at 650 • C were higher than those recorded at room temperature (Figure 7b). was especially evident at elevated temperature (650 °C). Cyclic plastic strains occurring in the internal structure of the material may also cause different stress states in differentlyoriented grains. This can trigger plastic deformation mechanisms, which results in the Bauschinger effect. This effect is caused by both mechanical strengthening, presumably related to the reconstruction of dislocation systems and the increase in their density, as well as to the precipitation processes caused by prolonged exposure to elevated temperature. Such behavior of the HR6W alloy is linked to precipitation strengthening occurring at its grain boundaries, responsible for blocking the slip bands. In the case of this alloy, precipitation processes are intensified as a result of high temperatures. Therefore, in order to determine the reasons for the hardening of the HR6W alloy at 650 °C, metallographic tests of the material taken from the samples submitted to the low-cycle fatigue tests were carried out.
It can also be seen that the maximum stress in the first load cycle at room temperature is higher than the maximum stress in the first cycle at 650 °C. This behavior of the material which differs in the saturated state in relation to the first load cycle results from the simultaneous impact of the temperature and the number of load cycles on the strengthening process. On the one hand, an increased temperature affects the weakening of the material; on the other hand, its long-term effect favors precipitation processes that lead to strengthening [12].
The changed behavior of the material in the saturated state results from the simultaneous influence of the temperature and the number of load cycles on the material strengthening process. On the one hand, the increase in temperature affects the weakening of the material, yet, on the other, its long-term effect favors precipitation processes, thus leading to the strengthening of the material.  Both at room temperature and at elevated temperature, a distinct change in the shape of the hysteresis loop was observed ( Figures 5-7), which is linked directly with material hardening in the process of cyclic deformation. The shape and the dimensions of the hysteresis loop indicated a very large increase in the maximum and the minimum stress. This was especially evident at elevated temperature (650 • C). Cyclic plastic strains occurring in the internal structure of the material may also cause different stress states in differentlyoriented grains. This can trigger plastic deformation mechanisms, which results in the Bauschinger effect. This effect is caused by both mechanical strengthening, presumably related to the reconstruction of dislocation systems and the increase in their density, as well as to the precipitation processes caused by prolonged exposure to elevated temperature.
Such behavior of the HR6W alloy is linked to precipitation strengthening occurring at its grain boundaries, responsible for blocking the slip bands. In the case of this alloy, precipitation processes are intensified as a result of high temperatures. Therefore, in order to determine the reasons for the hardening of the HR6W alloy at 650 • C, metallographic tests of the material taken from the samples submitted to the low-cycle fatigue tests were carried out.
It can also be seen that the maximum stress in the first load cycle at room temperature is higher than the maximum stress in the first cycle at 650 • C. This behavior of the material which differs in the saturated state in relation to the first load cycle results from the simultaneous impact of the temperature and the number of load cycles on the strengthening process. On the one hand, an increased temperature affects the weakening of the material; on the other hand, its long-term effect favors precipitation processes that lead to strengthening [12].
The changed behavior of the material in the saturated state results from the simultaneous influence of the temperature and the number of load cycles on the material strengthening process. On the one hand, the increase in temperature affects the weakening of the material, yet, on the other, its long-term effect favors precipitation processes, thus leading to the strengthening of the material.  Based on the analysis of the obtained specimens, coarse-grained, high-nickel austenite with visible twins and streaked precipitates of carbides M23C6 was found. These precipitates were distributed in the direction of plastic forming. Carbides M23C6 distributed on the boundaries of austenite grains create a local continuous shells. In addition, this alloy was shown to form both MC carbides. All these structural precipitates are in turn responsible for such strong strengthening of the HR6W material as observed in low-cycle fatigue testing at elevated temperature.
The mechanical characteristics of HR6W determined in the low-cycle fatigue tests at room temperature and elevated temperature are summarized in Tables 2 and 3. Based on these data, cyclic stress-strain curves for this alloy were plotted as presented in Figures  10 and 11, which were described by a mathematical model in the form of σ = A • ε and coupled with a curve determined in the static tensile test.
As can be seen from the comparison of the curves, the HR6W alloy under the fatigue test conditions was characterized by cyclic hardening both at room temperature and at 650 °C. In addition, mathematical models of cyclic strengthening were developed by σ = K • ε (Figure 12), which were described by the expression, where K′-is the cyclic strength coefficient, and n′-the cyclic weakening coefficient.  Based on the analysis of the obtained specimens, coarse-grained, high-nickel austenite with visible twins and streaked precipitates of carbides M23C6 was found. These precipitates were distributed in the direction of plastic forming. Carbides M23C6 distributed on the boundaries of austenite grains create a local continuous shells. In addition, this alloy was shown to form both MC carbides. All these structural precipitates are in turn responsible for such strong strengthening of the HR6W material as observed in low-cycle fatigue testing at elevated temperature.

Carbides M23C6
The mechanical characteristics of HR6W determined in the low-cycle fatigue tests at room temperature and elevated temperature are summarized in Tables 2 and 3. Based on these data, cyclic stress-strain curves for this alloy were plotted as presented in Figures  10 and 11, which were described by a mathematical model in the form of σ = A • ε and coupled with a curve determined in the static tensile test.
As can be seen from the comparison of the curves, the HR6W alloy under the fatigue test conditions was characterized by cyclic hardening both at room temperature and at 650 °C. In addition, mathematical models of cyclic strengthening were developed by σ = K • ε (Figure 12), which were described by the expression, where K′-is the cyclic strength coefficient, and n′-the cyclic weakening coefficient. Carbides M23C6 Figure 9. Microstructure of the HR6W alloy. Light microscopy Mi30Fe, a sample material subjected to a total strain of 0.7% at 650 • C, was etched.
Based on the analysis of the obtained specimens, coarse-grained, high-nickel austenite with visible twins and streaked precipitates of carbides M 23 C 6 was found. These precipitates were distributed in the direction of plastic forming. Carbides M 23 C 6 distributed on the boundaries of austenite grains create a local continuous shells. In addition, this alloy was shown to form both MC carbides. All these structural precipitates are in turn responsible for such strong strengthening of the HR6W material as observed in low-cycle fatigue testing at elevated temperature.
The mechanical characteristics of HR6W determined in the low-cycle fatigue tests at room temperature and elevated temperature are summarized in Tables 2 and 3. Based on these data, cyclic stress-strain curves for this alloy were plotted as presented in Figures 10 and 11, which were described by a mathematical model in the form of σ an = A·ε ac b and coupled with a curve determined in the static tensile test. where ε ael -elastic strain amplitude, ε apl -plastic strain amplitude, σ an -saturation stress amplitude, N f -fatigue life as the number of cycles to fracture.  where εael-elastic strain amplitude, εapl-plastic strain amplitude, σan-saturation stress amplitude, Nf-fatigue life as the number of cycles to fracture. where εael-elastic strain amplitude, εapl-plastic strain amplitude, σan-saturation stress amplitude, Nf-fatigue life as the number of cycles to fracture.  As can be seen from the comparison of the curves, the HR6W alloy under the fatigue test conditions was characterized by cyclic hardening both at room temperature and at 650 • C. In addition, mathematical models of cyclic strengthening were developed by σ an = K ·ε apl n (Figure 12), which were described by the expression, where K -is the cyclic strength coefficient, and n -the cyclic weakening coefficient.
The analysis of the characteristics obtained indicates cyclic strengthening of the HR6W alloy both at room temperature and elevated temperatures. At 650 • C, the strengthening process is significantly greater. A greater increase in the value of the stress σ an is visible with the increase in the amplitude of the strain ε apl . In this case, the value of the cyclic strength coefficient K' = 1403.9 MPa and the cyclic weakening coefficient n' = 0.1709 are significantly higher than those determined for room temperature, i.e., K' = 717.62 MPa and n' = 0.0992 ( Figure 12). The increase in stresses at room temperature from the initial state to  The analysis of the characteristics obtained indicates cyclic strengthening of the HR6W alloy both at room temperature and elevated temperatures. At 650 °C, the strengthening process is significantly greater. A greater increase in the value of the stress σan is visible with the increase in the amplitude of the strain εapl. In this case, the value of the cyclic strength coefficient K' = 1403.9 MPa and the cyclic weakening coefficient n' = 0.1709 are significantly higher than those determined for room temperature, i.e., K' = 717.62 MPa and n' = 0.0992 ( Figure 12). The increase in stresses at room temperature from the initial state to the saturation state was found to be on average 80 MPa, while the temperature of 650 °C was on average 240 MPa.
Evidence for a much higher intensity of strengthening at 650 °C can also be seen in the characteristics of cyclic deformation (Figures 13 and 14). The process is particularly dynamic in the initial phases of LCF tests, which was observed for all ranges of total strain. The effect observed here is analogous to the characteristics shown in Figures 10 and 11.  The analysis of the characteristics obtained indicates cyclic strengthening of the HR6W alloy both at room temperature and elevated temperatures. At 650 °C, the strengthening process is significantly greater. A greater increase in the value of the stress σan is visible with the increase in the amplitude of the strain εapl. In this case, the value of the cyclic strength coefficient K' = 1403.9 MPa and the cyclic weakening coefficient n' = 0.1709 are significantly higher than those determined for room temperature, i.e., K' = 717.62 MPa and n' = 0.0992 ( Figure 12). The increase in stresses at room temperature from the initial state to the saturation state was found to be on average 80 MPa, while the temperature of 650 °C was on average 240 MPa.
Evidence for a much higher intensity of strengthening at 650 °C can also be seen in the characteristics of cyclic deformation (Figures 13 and 14). The process is particularly dynamic in the initial phases of LCF tests, which was observed for all ranges of total strain. The effect observed here is analogous to the characteristics shown in Figures 10 and 11. Evidence for a much higher intensity of strengthening at 650 • C can also be seen in the characteristics of cyclic deformation (Figures 13 and 14). The process is particularly dynamic in the initial phases of LCF tests, which was observed for all ranges of total strain. The effect observed here is analogous to the characteristics shown in Figures 10 and 11.  Materials 2021, 14, x FOR PEER REVIEW 9 of 13  Based on the results obtained (Tables 2 and 3), Figure 15 summarizes values for fatigue life Nf versus total strain range Δεt.  Based on the results obtained (Tables 2 and 3), Figure 15 summarizes values for fatigue life Nf versus total strain range Δεt. From the analysis of the data, it can be concluded that, at room temperature, a strainrange (∆ε t ) dependent fatigue life N f of the HR6W alloy increased several times (from approx. 3 to approx. 4) as compared to its durability at 650 • C. On the other hand, when analyzing the results of testing the material at 650 • C, it can be noticed that, with a double increase in the total strain range ∆ε t , for example from 0.6% to 1.2%, the number of cycles to failure N f decreased four times from 3377 to 834, and at room temperature up to four times from 13,890 to 2630. Based on the mechanical characteristics in Tables 2 and 3, the fatigue life graphs for HR6W were also developed (Figures 16 and 17