Effect of Temperature on Foreign Object Damage Characteristics and High Cycle Fatigue Performance of Nickel-Based Superalloy GH4169

Abstract

High-speed ballistic impact tests were conducted at room temperature and 500 °C on nickel-based superalloy GH4169 simulated blade specimens containing leading-edge features. The microscopic characteristics of the impact notch at room temperature versus 500 °C were observed by electron backscatter diffraction (EBSD), and it was found that the grains on the notched subsurface were ruined, while in more distant regions, the impact energy was mainly absorbed by grain boundaries. Internal damage is more concentrated in the notched subsurface region at 500 °C compared to room temperature. The high cycle fatigue strength of the damaged specimens under different conditions was tested. The results showed that the high cycle fatigue strength of the damaged specimens increased with the increase in the notch depth, and the fatigue strength of the damaged specimens at 500 °C was higher than the fatigue strength at room temperature. Both the 48 h post-impact holding time at 500 °C and the preload during impact at 500 °C increased the fatigue strength of the damaged specimens.

1. Introduction

Aerospace turbine engines are likely to inhale foreign objects such as gravel, rocks, or metal fragments during takeoff, landing, or low-level flight. These hard objects may impact the high-speed rotating compressor blades, resulting in foreign object damage (FOD) [1,2]. Notch-type damage represents the most prevalent form of foreign body damage observed on aero-engine fan/compressor blades [3,4]. It is important to consider the effect of foreign object damage on blade resistance to high cycle fatigue (HCF) during the design of structural integrity and the formulation of serviceable limits for fan/compressor blades [5,6,7].

FOD notches differ from mechanical machining notches in that the geometric appearance and residual stresses of FOD notches are highly randomized. The effect of FOD on the high cycle fatigue strength of blades encompasses not only stress concentration resulting from irregular damage geometry but also microstructural damage, including microcracks and voids caused by impact deformation and material loss, as well as tensile and compressive residual stresses resulting from residual plasticity [8,9,10]. How to reproduce FOD injury is the primary problem. Nicholas’s research shows that despite the similarity in fatigue strengths, quasi-static and low-velocity pendulum indents result in damage mechanisms that differ from those observed in equivalent depth craters produced by ballistic impacts [11]. Therefore, light air guns are currently the principal method of reproducing FODs.

A significant number of experimental and numerical studies have been conducted on the FOD of titanium blades, which are commonly used in fans and low-pressure compressors. Ruschau found that off-angle impacts relative to head-on (0°) impacts were more detrimental to Ti-6Al-4V airfoil specimens [12]. In studying the effect of residual stress on the strength of HCFs from dent-type FODs at a 90° impact, Chen et al. found that FOD leads to a reduction in the critical crack size of HCF by approximately 60% [13]. Bao et al. demonstrated that the strength degradation of HCF caused by notch-type FOD was more significant than that caused by dent-type FOD through the performance of notch-type and dent-type FOD impact tests on TC4 titanium alloy flat plate specimens with different impact angles [14]. In a series of experiments conducted by Ruschau, it was observed that the fatigue strength of damaged samples increased due to stress relief [10]. In the study of fatigue life of notched components, Liao and He further developed the critical distance theory [15,16]. Jia conducted a series of simulations to investigate the effects of FOD on TC4 specimens at varying angles. The findings revealed a decline in the fatigue strength of notch-type damage with both an increase in damage length and maximum damage depth [17]. Furthermore, Jia conducted simulations of notch-type damage under varying preloads, observing the macro-micro characteristics of the resulting notches. Additionally, finite element simulations were performed to assess the influence of stress concentration factors and residual stresses. It was discovered that the preload has the potential to enhance the residual compressive stresses resulting from the impact, which in turn can suppress the initiation of cracks and improve fatigue strength [18]. Zhang developed an FE model of aeroengine blades that considered the initial centrifugal force [19]. Zhang also performed a large number of impacts on rectangular-sectioned thin specimens and developed a spring-mass system based on Winkler’s elastic-plastic foundation theory to model FOD [20]. The modeling results were in good agreement with the experimental measurements. In light of the rapid development of high-speed railway development, the issue of foreign object axle damage has become a subject of increasing scrutiny [21,22,23,24,25]. It has damage characteristics similar to the FODs on blades. Furthermore, laser impact strengthening has been the subject of continuous investigation as a potential method for enhancing the fatigue resistance of blade FOD [26,27,28].

Despite the relatively low probability, high-pressure compressors face the same FOD problem as fans and low-pressure compressors. However, there is a paucity of studies on the FOD characteristics of nickel-based high-temperature alloy rotor blades for high-pressure compressors and their effect on HCF strength. Farahani conducted a simulation of FODs on the surface of flat specimens, varying two parameters: the shape of the object nose and the impact angle [29]. The results indicated that flat-nosed projectiles with an impact angle of 45° exhibited the greatest number of induced microcracks and stress concentration, while spherical-nosed projectiles with an impact angle of 90° exhibited the least. Jia investigated the foreign object damage characteristics and their effects on the high cycle fatigue properties of a GH4169 nickel-based superalloy, but the study was limited to the impact characteristics and HCF behavior at room temperature and did not examine those at the operating temperature of GH4169 [30].

In this paper, a kind of simulated blade specimen containing leading-edge features is designed based on the geometric characteristics of the vulnerable position of the leading edge of the high-pressure compressor blade. Notch-type FODs of various depths at room and high temperatures were simulated using a high-speed ballistic impact device. High cycle fatigue tests at various stress ratios were performed on damaged specimens using the step loading test technique. The macroscopic and microscopic characteristics of the notch-type foreign object damage and the effect of temperature on the variation rule of high cycle fatigue strength with damage depth are revealed. Finally, the study reveals the effect of holding time after high-temperature impact and preload with high-temperature impact on the variation rule of high cycle fatigue strength with damage depth of notched FOD.

2. Experiments

2.1. Materials and Specimens

GH4169 is a forged nickel-based superalloy commonly used for aero-engine high-pressure compressor blades. The chemical composition of GH4169 used in this paper is shown in

Table 1. The chemical composition of GH4169.

Element	Cr	Ni	Mo	Al	Ti	Nb	Fe
Mass fraction (%)	19.64	54.32	2.90	0.45	0.93	5.36	15.67

. The microstructure is shown in Figure 1.

The hourglass specimens used for the material fatigue strength tests are shown in Figure 2. Since the notch-type FOD mostly occurs at the leading edge of the blade, the simulated blade specimen containing leading-edge features was designed, as shown in Figure 3. The simulated blade specimens were machined by CNC milling and polished in the axial direction.

2.2. Simulated FOD Impact Test

Figure 4 shows the high-speed ballistic impact system used for the simulated FOD impact test. For the test, GCr15 bearing steel balls with a diameter of 3 mm were used as the impactor. The simulated impact velocity was set at 300 m/s, and the impact angle was set at 60° based on the linear velocity of the compressor blade and the most dangerous impact angle. The impact temperature for the specimen was 500 °C. The electromagnetic induction equipment used for the test has a maximum power of 45 kW and employs infrared induction PIC feedback control. Heating the specimen to 500 °C takes approximately 1 min. Due to the fast heating speed and small thickness of the specimen, the holding time before impact was set to 1 min. The depth of impact was controlled by the coaxial laser and the displacement scales of the multi-axial test stand.

2.3. Microscopic Characteristics Observation

In order to reveal the effect of temperature on the microscopic characteristics of GH4169 notch-type FOD, three sets of impact tests with different target depths were conducted, and the notches were dissected and observed after impact.

Table 2. Impact test plan of microscopic observation.

Specimen ID	Target Depth	Impact Depth (Maximum Depth)	Impact Temperature	Holding Time at 500 °C
#1-N	0.6 mm	0.556 mm	Room temperature	0 h
#2-H		0.696 mm	500 °C	0 h
#3-H-48		0.621 mm	500 °C	48 h
#4-N	1.1 mm	1.141 mm	Room temperature	0 h
#5-H		1.120 mm	500 °C	0 h
#6-H-48		1.075 mm	500 °C	48 h
#7-N	1.6 mm	1.59 mm	Room temperature	0 h
#8-H		1.658 mm	500 °C	0 h
#9-H-48		1.572 mm	500 °C	48 h

shows the impact test plan of microscopic observations. ‘N’ means the impact temperature is room temperature, and ‘H’ shows the impact temperature is 500 °C; ‘48’ represents the holding time of 48 h at 500 °C after impact. Each notch was cut, cold mounted, and grinded to the center cross-section, as shown in Figure 5.

Optical microscope observation was used to observe the metallography of the cross-section of the center of the notch. The corrosion layer was removed by grinding and polishing after optical observation, and the specimens were characterized by electron backscatter diffraction (EBSD) using Gemini 500 (Gemini, New York, NY, USA). The accelerating voltage was 30 kv, and the scanning step was 2 μm. The EBSD data were post-processed using AZtecCrystal 2.1 software to obtain inverse pole figures (IPFs) and geometrically necessary dislocation density (GND) figures.

2.4. High Cycle Fatigue Test

The HCF strength is specified as the life of 10⁶ cycles. The HCF tests were conducted at room temperature (about 20 °C) and 500 °C, and the loading form was axial loading in load control mode with a frequency of 100~120 Hz. Since different locations of the real blade have different stress ratios, the stress ratio ranges from about −1 to 0.8, and the stress ratio of the damage location-prone area is about 0.1; the stress ratios of −1, 0.1, and 0.5 were selected as the experimental stress ratios. Meanwhile, to investigate the effect of sustained high temperature on HCF strength, one specimen group was selected for an HCF strength test with a stress ratio of 0.1 after being held at 500 °C for 48 h after impact. Another set of specimens was subjected to a preload stress of 30% of the yield strength applied by a tensile tester to simulate the centrifugal load received by the blade during impact to investigate the effect of the centrifugal load on the HCF strength at 500 °C. The simulated impact test and HCF test plan are shown in

Table 3. Simulated FOD impact test and HCF test.

Specimen ID	Impact Velocity	Impact Angle	Stress Ratio	Impact Condition	Damage Depth
N1.1~N1.5	300 m/s	60°	0.1	Room temperature	0~1.6 mm
N2.1~N2.5			0.5
N3.1~N3.5			−1
H1.1~H1.5	300 m/s	60°	0.1	500 °C	0~1.6 mm
H2.1~H2.6			0.5
H3.1~H3.4			−1
HP1.1~HP1.4	300 m/s	60°	0.1	500 °C with 48 h post-damage holding time	0~1.6 mm
HL1.1~HL1.6	300 m/s	60°	0.1	500 °C with preloading (30% yield strength)	0~1.6 mm

.

Since it is difficult to obtain exactly the same damage, the traditional HCF experimental methods, such as the group method and the lift method, which require a large number of identical specimens for testing, are not suitable for fatigue strength testing of notch-type FOD specimens. In this paper, step-loading test technology was used to obtain the HCF strength of specimens with different damages. In this method, an initial cyclic stress lower than the estimated fatigue strength is first applied to the specimen for a target number of cycles (10⁶ cycles). If the specimen fractures under the initial cyclic load, the test results are invalid. If the specimen does not fracture, the test continues for the target number of cycles by adding a small amount of load (generally 5% of the initial load) to the previous load level. The increase is repeated until the specimen fractures within the target cycle. The fatigue strength of the specimen is determined by the last two stages of loading versus the number of cycles using the following formula:

$${\mathsf{\sigma }}_{\mathrm{es}}={\mathsf{\sigma }}_{\mathrm{pr}}+{\displaystyle \frac{N_{\mathrm{f}}}{{10}^6}}\left({\mathsf{\sigma }}_{\mathrm{f}}-{\mathsf{\sigma }}_{\mathrm{pr}}\right)$$

(1)

where ${\mathsf{\sigma }}_{\mathrm{f}}$ is the maximum cyclic stress of the final loading block with failure of the specimen, $N_{\mathrm{f}}$ (<10⁶ cycles) is the number of cycles of the final loading block, and ${\mathsf{\sigma }}_{\mathrm{pr}}$ is the maximum cyclic stress of the preceding load block.

3. Results and Discussion

3.1. Macroscopic Characteristics Observation Results

Figure 6 and Figure 7 show the notch-type FOD observed by optical observation at the same magnification, including the incident side view and the exit side view. At room temperature, the macroscopic morphology of the notch-type FOD after impact by a 3 mm diameter steel ball shows a “C” shape. For notches with shallow impact depths, material loss is minimal and almost invisible from the incident side, resembling a crater. As the impact depth increases, material loss increases, and a small amount of extruded burr appears on both sides of the notch. As the impact depth increases, more pronounced bending deformation occurs on both sides of the notch. In all cases, a layer of wrinkles forms on the inside of the notches, indicating that the impact not only cut the material but also affected the surrounding area to some extent. At 500 °C, the macroscopic morphology of the FOD notch and its variation with depth are not significantly different from those at room temperature.

Figure 8 shows the relationship between the incident and exit side damage lengths with the incident and exit side depths (maximum depth) at room temperature and 500 °C. It can be seen that the incident side damage length increases with damage depth at room temperature and approaches the diameter of the steel ball. The damage length on the exit side is slightly larger than that on the incident side, while the damage depth on the exit side is significantly larger. In addition, as the damage depth increases, the maximum damage length on the exit side can exceed the diameter of the steel ball. This is due to the fact that the exit side experiences additional material loss, not just steel ball-cutting material. Figure 6 also supports this analysis. Figure 8c,d show that at 500 °C, the damage length and depth on the incident side still maintain a good relationship, but the damage length on the exit side increases compared to that at room temperature. Meanwhile, the relationship between the damage length and the damage depth on the exit side is significantly more scattered compared to that at room temperature. It can be concluded that the impact at 500 °C is likely to cause more material loss on the exit side.

3.2. Microscopic Characteristics Observation Results

Figure 9 shows the macrograph of the notches used for microscopic observation. Figure 10 shows the cross-sectional metallography at the center of each notch. The impact produced not only a notch but also cracks below the notch. When the notch size was small (#1-N), cracking occurred on the exit side, and the angle of the notch cross-section was similar to the impact angle. This is because the impact energy absorbed by the leading edge was small, and the material was not completely stripped by the shear cracks formed by the impact. As the notch size increased, the cracks on the exit side disappeared and instead appeared on the incident side (#3-H-48, #4-N, #6-H-48); then, the notch cross-section angle increased, which was greater than the impact angle. This is due to the increase in impact energy absorbed by the leading edge of the blade; the material on the exit side was completely stripped, and the material on the incident side was also extruded by the impact, forming cracks under the surface of the incident side. As the notch continued to grow, the notch root angle approached 90°, and the incident side material was completely stripped along the impact crack (#7-N). At the same time, the notched surface was flatter and smoother under a 500 °C impact than under a room temperature impact. When the notch size was large (#8-H, #9-H-48), cracks were still visible below the surface of the incident side of the 500 °C impact.

Figure 11 shows the inverse pole figure (IPF) and geometrically necessary dislocation density (GND) results for each notch center section. From the figure, it can be seen that the grains below the impact surface are severely damaged, and the grain orientation is unrecognizable. Furthermore, it is observed that the dislocations generated by the impact extend from the grain boundaries to the inside of the grains in the severely damaged regions, such as in the vicinity of the notch, while in the rest of the affected regions, they are mainly concentrated at the grain boundaries. It can be concluded that when the impact occurs, the impact energy is mainly absorbed by the grains in the vicinity of the impact notch, and dislocations are generated inside the grains. When the dislocations reach the threshold value, the grains are destroyed. In the region slightly away from the impact notch, the impact energy is mainly absorbed by the grain boundaries. By comparing the GND figures of notches of different depths, it can be seen that the damage depth is greater on the incident and exit surfaces than on the inside. At the same time, the surface damage depth of the exit side is greater than that of the incident side. When the notch depth is small (such as #1-N), the microscopic damage depth of the incident surface and the exit surface are similar. As the notch depth increases, the damage depth on the exit side becomes larger than that on the incident side. When the notch depth is large (such as #7-N), the surface damage depth of the incident side becomes minimal, while the grain damage occurs at a deeper position on the exit side. Temperature comparison of the impact damage shows that at 500 °C, the damage is more concentrated in the notch than it is at room temperature. Meanwhile, the stress in the part away from the notch can be recovered for the specimens held at 500 °C for 48 h after impact, which reduces the size of the zone affected by impact.

3.3. High Cycle Fatigue Test Results

To enhance the intuitiveness of the results, the test results were normalized using the fatigue strength of hourglass specimens with the same temperature and stress ratio (Equation (2)). The fatigue strength test results of the hourglass specimens are shown in

Table 4. High cycle fatigue strength of hourglass specimens.

Temperature	High Cycle Fatigue Strength (MPa)
	R = 0.1	R = 0.5	R = −1
Room temperature	493.3	742.8	217.7
500 °C	492.4	743.6	267.7

. Figure 12 shows the variation in HCF strength with damage depth of notch-type specimens at room temperature and 500 °C for three different stress ratios (R = 0.1, R = 0.5, and R = −1). From the figure, it can be seen that the relative strengths of the notch-type FOD specimens increase in the order of R = 0.1, R = 0.5, and R = −1 for the three stress ratios at room temperature. When the notch is shallow, the relative strength of the stress ratio R = 0.5 is similar to that of R = 0.1, and when the notch is deep, it is similar to that of the stress ratio R = −1. At 500 °C, the relative strengths of the three stress ratios still increase in the order of 0.1, 0.5, and −1. However, the relative strengths of the stress ratios R = 0.5 and R = −1 are closer to each other while being far away from the relative strength of the stress ratio R = 0.1. Meanwhile, in the case of shallow notches, the relative strengths of R = 0.5 and R = −1 would be more than 1, which means they are higher than the fatigue strength of the hourglass-shaped specimens. This is due to the difference between the simulated blade specimens containing leading-edge features used for notch-type FOD simulation and the hourglass specimen, although both have the same stress concentration factor of Kt = 1.01.

$$\mathrm{Relative} \mathrm{strength}={\displaystyle \frac{\mathrm{HCF} \mathrm{strength} \mathrm{of} \mathrm{FOD} \mathrm{specimen}}{\mathrm{HCF} \mathrm{strength} \mathrm{of} \mathrm{hourglass} \mathrm{specimen}}}$$

(2)

The fractography of notch-type damage specimens with different depths at three stress ratios at room temperature was observed using scanning electron microscopy (SEM), as shown in Figure 13. It can be seen that all the specimens began to show HCF fractures from the location of the damage, and some of the notches showed friction ablation. It is difficult to determine the location of crack initiation from the fractography. Some of the notches showed cracks under the impact surface (N1.2, N2.4, and N2.5 in Figure 13). The fractography of notch-type damage specimens with different depths at three stress ratios at 500 °C is shown in Figure 14. Compared with the fractography at room temperature, it appears flatter at high temperature. It is still difficult to determine the location of crack sprouting, but the tendency for cracks to sprout from the surface of the notch can be vaguely seen. Combined with microscopic observations, it can be supposed that notch-type damage HCF fractures originate from the stress concentration generated by the notch and multi-source crack sprouting in the grain fragmentation region of the notched subsurface.

Figure 15 displays the relationship between the relative strength of notch-type damaged specimens and the notch sizes for the stress ratio R = 0.1 at room temperature, 500 °C, 500 °C with 48 h post-damage holding time, and 500 °C with preloading. Due to the small amount of valid data at 500 °C with 48 h post-damage holding time, one specimen with the HCF fracture location deviating from the small impact notch (less than 0.06 mm at incident side and 0.3 mm at exit side) was considered an undamaged specimen and plotted in Figure 15. The relative strengths varied in a similar trend with notch sizes for all four conditions, increasing in the following order: room temperature, 500 °C, 500 °C with 48 h post-damage holding time, and 500 °C with preloading. In the case of the 500 °C with 48 h post-damage holding time, the relative strengths were increased by approximately 20.4% compared to 500 °C. In the case of the 500 °C with preloading, the relative strengths were increased by 38.1% in comparison to 500 °C. This suggests that 500 °C, 500 °C with 48 h post-damage holding time, and 500 °C with preloading can improve the ability of GH4169 material blades to resist post-impact HCF fracturing.

In summary, the impact conditions at 500 °C reduced the extent of the decrease in HCF strength after FOD. Combined with the macro- and micro-features of the FOD notch, it can be hypothesized that this may be due to the fact that the damage is more concentrated on the notch subsurface and the size of the microscopic damage region is smaller when FOD occurs at 500 °C. The increase in HCF strength by 48 h post-damage holding time may be due to the elimination of dangerous residual stresses, and the increase in HCF strength under preloading may be due to the effect of compressive residual stresses generated by the preloading impact [19]. These indicate that notch-type FOD of high-pressure compressor blades is more dangerous at low temperatures and within the moment of damage occurrence. Meanwhile, notch-type FOD under preloading is safer. This conclusion can provide a reference for the design of high-pressure compressor blades against FOD.

4. Conclusions

(1)

The macroscopic morphology of notch-type FOD is a C-shaped type at both room temperature and 500 °C, and a layer of wrinkles forms on the inside of all notches, indicating that the impacts have affected the inside of the notch to some extent. The damage length of the FOD notch increases as the damage depth increases. The impact causes additional material loss at both the incident and exit sides, resulting in a notch that is larger than the material cut by the path of movement of the steel ball. This damage is further exacerbated at 500 °C compared to room temperature, with the exit side notch size becoming more dispersed.

(2)

When FOD occurred, the grains on the notched subsurface near the notched region were ruined, while in more distant regions, the impact energy was mainly absorbed by grain boundaries. Impacts caused more damage on both the incident and exit side surfaces than on the interior. Notched subsurface microscopic damage was more concentrated in the notch at 500 °C compared to room temperature, and the holding time after impact allowed the distal damage to be restored.

(3)

It was observed that sub-surface cracks were present at the root of notch-type FODs. Furthermore, the fractography of fatigue fractures was rougher at room temperature and flatter at 500 °C. The HCF fracture at both room temperature and 500 °C was found to originate from the notch stress concentration and multi-source crack sprouting in the grain fragmentation region of the notched subsurface.

(4)

The decrease in HCF strength after FOD at 500 °C is smaller than that after FOD at room temperature under the three stress ratios. For stress ratio R = 0.1, the relative fatigue strengths at 500 °C with 48 h post-damage holding time and 500 °C with preloading increased. The data indicate that for GH4169 blades, notch-type FOD damage at low temperatures and the moment of damage occurrence are the most dangerous. Conversely, notch-type FOD damage under preload is safer. This is a significant reference in the field of blade design to resist foreign object damages.