Optimization and fracture mechanism analysis of TC17 titanium alloy simulated-blade with two-sided laser shock processing

X. Nie; Y. Li; W. He; L. Zhou    Science and Technology on Plasma Dynamics Laboratory, Air Force Engineering University, P.R. China

1 INTRODUCTION

Laser shock processing (LSP) is an innovative surface treatment technology, which can improve the fatigue resistance of metals and alloys (1). Compared with conventional shot peening (SP), LSP has some special advantages, such as deeper compressive residual stress, small cold work rate, lower surface roughness and better controllability. Beacause of the above advantages, it becomes more and more popular in the surface treatment field. What’s more, LSP has been successfully applied on fan/compressor blades of military aero-engines to improve HCF performance and foreign object damage resistance. Many studies have been carried out to discuss the effects of LSP on fatigue property for different metals and alloys (2,3). And some reseaches focused on the effects on the fatigue crack initiation and growth (4-6). In our previous work, LSP was successfully applied on some titanium alloys to improve the fatigue strength, and the strengthening mechanism was also discussed(7,8).

However, the above experimental studies just investigated the effects of a specified LSP procedure, always with a simple laser-peened path, on the fatigue performance. But there were few experimental studies about the effects of different LSP procedures including laser parameters and laser shocked-path. In this paper, LSP procedure for TC17 simulated-blade was optimized according to the work stress state. And the simulated-blades were treated with two LSP procedures. The fatigue limits with different LSP procedures were compared by the up-and-down method during fatigue tests. The fracture mechanisms of fatigue crack initiation and growth with different LSP procedures were compared.

2 EXPERIMENTAL PROCEDURE

2.1 Laser shock processing

The LSP process utilizes laser pulse irradiated at the target surface covered by opaque ablating layer and transparent confining layer. When the laser beam passes through the transparent layer and strikes the surface, the ablating layer absorbs the laser and immediately vaporized into the plasma. The rapidly expanding plasma leads to the formation of shock wave which strikes the material and propagates into the material with an intensity of several GPa. If the shock pressure is greater than the dynamic field strength, plastic deformation will be generated with compressive residual stress and microstructure changes in the surface. Because of the symmetrical structure, the simulated-blades were treated by two-sided LSP shown in reference (7).

2.2 Material and specimen

TC17 titanium alloy is widely used in Chinese aviation field, such as aero-engine fan and compressor blade. In order to simulate the work stress state of aero-engine compressor blade, the TC17 titanium alloy specimen was designed and machined into the special structure, namely simulated-blade shown in Fig.1(a). The simulated-blades were machined by Gas-turbine research institute of China. Fig.1(b),(c) is the first-order modal displacement contour and equivalent effective stress contour by finite element analysis (FEM). The first-order modal is a cantilever vibration modal. The greatest vibration amplitude locates at the blade-tip. The calculated first-order frequence is 338 HZ, which is well consistent with the factual resonance frequence in vibratory fatigue tests, 332HZ ~ 345HZ. In addition, the maxiam vibration stress locates at the blade-root rounding near the R transition line, especially the center region. In response to the stress analysis results, the LSP region was confirmed as the blade-root rounding area, the dashed region in Fig.2(a), which is 30 mm × 14.4 mm.

3 RESULTS AND DISCUSSION

3.1 LSP procedure optimization and fatigue performance

The simple snaky laser shocked-path was applied in first LSP procedure (Fig.2(a)) with detailed laser parameters as following: laser energy/6 J, laser duration/20 ns, laser spot diameter/3 mm, laser fluence/4.24GW/cm2, overlapping rate/50%, 1 impact. Fifteen simulated-blades were treated by LSP. After LSP, thirty simulated-blades without and with LSP were used to conduct vibration fatigue tests on D-300-3 electric vibration system shown in reference (7). In the fatigue test, the test stress was measured by a strain gage which is sticked on the simulated-blade root with highest work stress as shown in Fig.1(c). At the same time, the amplitude of simulated-blade tip in the fatigue test was measured by a laser displacement sensor. Then, the relataionship between the highest work stress and amplitude of simulated-blade tip was established. And we just monitored the amplitude of simulated-blade tip in test. The fatigue test results were processed by the fatigue up-and-down method.

The original fatigue limit of TC17 titanium alloy simulated-blade is 405.7 MPa. However, the fatigue limit with first procedure is only 360 MPa, decreased by 11.3%. According to the fluorescent test results, it is found that the fatigue crack initiates at the blade edge within the LSP region, far from the blade-root rounding. In order to analyze the the cause of the fatigue crack initiation, a numerical simulation work of the first LSP procedure was conducted. The simulation results indicate that the maximal equivalent effective plastic strain locates at blade edge, where there is a material protrusion resulted from the accumulated plastic deformation. The accumulated plastic deformation may result in the complicated residual stress field generated. Even more, tensile residual stresses may be generated in the material protrusion area, which leads to fatigue crack initiation. Thus, the main cause of fatigue crack initiation is not the vibration stress, and it may be induced by the unbenefited residual stresses.

According to the cause of the fatigue performance deterioration with the first LSP procedure, the LSP procedure was optimized as shown in Fig.2(b). In order to advoid the accumulated plastic deformation generated at blade edge, the center region undergone great work stress was designed to be laser peened with a great intensity, but with a low intensity for the simulated-blade edge region. The design target is to induce greater residual compressive stresses in the center region and reduce the accumulated plastic strain at the blade edge. In the previous research work, it is found that there is direct relationship between the laser fluence and laser induced plastic deformation. The greater laser fluence is, the greater plastic deformation induced by LSP is. Therefore, an optimized distinctive LSP procedure was confirmed with four LSP regions, region 1 (3 J/20 ns/Φ2.4 mm/50%/1 impact) with high laser fluence (3.32GW/cm2) and region 2/3/4 (2 J/20 ns/Φ2.4 mm/50%/1 impact) with low laser fluence (2.21GW/cm2). And the treatment sequence is region 1 first, then region 2, region 3 and region 4. The great compressive residual stresses are produced in region 1 for the resistance to the greatest work stress. And lower compressive stresses are generated in region 2/3/4 for the transition of great compressive stresses in region 1, preventing stress concentration. Fifteen simulated-blades treated with the optimized LSP procedure were used to conduct fatigue tests. The fatigue limit with the optimized LSP procedure is 462.9 MPa, 14.1% incresed by compared with the orginal fatigue limit. And the fatigue crack initiates at the center region, not the blade edge. In summary, the first LSP procedure can induce great compressive stresses in the LSP region, but with great accumulated plastic deformation, even tensile residual stress at blade edge resulting in fatigue cracking. In contrast, the optimized LSP procedure can induce great compressive stresses in region 1and realize the gradual transition of the compressive stresses in region2/3/4, without great accumulated plastic deformation generated at blade edge.

3.2 Fracture mechanism analysis

The fractographys of simulated-blades treated by two LSP procedures were observed by a JEOL/JSM-6360LV scanning electron microscope (SEM). Before the observation, the raptured simulated-blades were cleaned with ethanol in the ultrasound cleaning machine. Fig.3(a) is the typical fractography with first procedure. It shows that the fatigue crack initiates at the blade edge and propogates into the center/depth. In the period of fatigue crack initiation and...