Thermal relaxation of residual stresses in shot peened surface layer of (TiB+TiC)/Ti–6Al–4V composite at elevated temperatures
Lechun Xie,Chuanhai Jiang,Vincent Ji
DOI: https://doi.org/10.1016/j.msea.2011.04.075
2011-01-01
Abstract:As an effective and important surface treatment method, shot peening can introduce high residual compressive stress and microstructure variation at near surface deformation layers. In this work, residual stresses relaxation behaviors of the shot peened layer of (TiB + TiC)/Ti–6Al–4V composite were investigated during thermal exposure, and the microstrain was calculated according to the integral breadth after isothermal annealing. The microstrain decreased fast and reached the minimum at 500 °C, which resulted from the thermal recovery and dynamic recrystallization. At elevated temperatures, the residual compressive stresses were relaxed in the whole deformation layers, which were caused by the thermally activated gliding of dislocations. The processes of relaxation can be described using a Zener–Wert–Avrami function and the activation energy of the residual stresses relaxation was higher than that of titanium self diffusion, which was ascribed to the hindrance effects of reinforcements as sink sources of dislocations during annealing. Keywords Thermal relaxation Residual stresses Microstrain Shot peening (TiB + TiC)/Ti–6Al–4V 1 Introduction As one kind of metal matrix composites, titanium matrix composites have been widely concerned because of their excellent physical and mechanical properties [1,2] . However, in the process of manufacturing and subsequent heat treatments, the tensile stresses always generated and deteriorate the fatigue properties [3] . In order to improve their fatigue strength and fatigue life, crack initiation and growth at surface layers must be suppressed using surface mechanical treatments [4] . As an effective and important surface treatment method, shot peening (SP) can introduce high residual compressive stress (RCS) and microstructure variation at near surface layers, which can enhance their fatigue properties compared to un-peened materials. The process of SP involves the bombardment of spherical balls of a hard material against the surface of components, which induces the strong elastic–plastic deformation at the surface and sub-surface regions. In the deformation layers, high RCS and microstructure refinements are introduced after SP. However, RCS could relax significantly under thermo-mechanical loadings, thermal exposure, static loading and cyclic loading [5,6] . The stability of the RCS on near-surface regions against thermal and mechanical loading is crucial to the fatigue properties improvement [7] . The relaxation behaviors in traditional alloys and metals have been extensively investigated [8–12] . But few investigations have been made on the relaxation behaviors in shot peened titanium matrix composites at elevated temperatures. Therefore it is significant to understand the residual stresses relaxation behaviors of titanium matrix composites reinforced with TiB and TiC. The residual stresses relaxation involves mechanical and thermal relaxation mechanisms [8] . In this work, the thermal residual stresses relaxations of the shot peened (TiB + TiC)/Ti–6Al–4V have been studied and thermal relaxation mechanisms have been discussed in detail. Besides, the residual stress distribution and microstrain variations have also been investigated after isothermal annealing. 2 Experimental 2.1 Preparation of materials The samples of (TiB + TiC)/Ti–6Al–4V (TiB:TiC = 1:1 (vol%)) were fabricated via in situ technology [13–15] , and the volume fraction of reinforcements (TiB + TiC) was 8%. Stoichiometric raw materials of sponge titanium, B 4 C and graphite powder were melted homogeneously in a consumable vacuum arc remelting (VAR) furnace to produce titanium matrix composites via self-propagation high-temperature synthesis. In order to obtain the even composition, the process of melt was three times. The samples were cut directly from ingots and then polished. The shape of samples liked disk, diameter is 15 mm and thickness is 3 mm. Young's modulus of the samples is 133 GPa, and the yield strength is 1150 MPa. The microstructure of Ti–6Al–4V and (TiB + TiC)/Ti–6Al–4V (Vol. 8%) have been shown in Fig. 1 . Fig. 1 (a) and (b) represent unetched surfaces of Ti–6Al–4V and (TiB + TiC)/Ti–6Al–4V respectively, and it can be found that the existence of reinforcements are obvious in Fig. 1 (b) compared to Fig. 1 (a). In order to observe the microstructure of reinforcements clearly, the microstructure of etched surface of (TiB + TiC)/Ti–6Al–4V has been obtained and displayed in Fig. 1 (c), and the results reveal that the reinforcements of TiB and TiC are distributed uniformly in the titanium matrix composites. The morphologies of reinforcements mainly exhibit needle-shaped, short fibre-shaped and particle-shaped. The needle-shaped and short fibre-shaped reinforcements are TiB, while the particle-shaped reinforcements are TiC. The differences in morphologies ascribe to their solidification paths and crystal structure. 2.2 SP and isothermal annealing treatments The SP treatments were performed using an air blast machine (Carthing Machinery Company, Shanghai). The first and second SP treatments were carried out and shot media were cast steel balls and Al 2 O 3 ceramic balls respectively. Cast steel ball with hardness 610 HV and Al 2 O 3 ceramic ball with hardness 700 HV were used on all samples. The diameter of peening nozzle was 15 mm and the distance between nozzle and samples was 100 mm. The coverage of SP was 100% in all peening steps. According to previous investigations [16] , the detailed SP parameters were given as follows: 0.3 + 0.2 MPa of jet pressure, 0.3 + 0.15 mmA of SP intensity, 0.6 + 0.3 mm of average diameter of cast steel balls and Al 2 O 3 ceramic balls respectively, 0.5 + 0.3 min of first and second SP times. In this work, the main aim of second SP is making the surface smooth after the first SP, so lower peening intensity and smaller ball diameter were carried out on the second SP treatment. After SP, based on the references [12,17] , isothermal annealing treatments were carried out at 350 °C, 400 °C, 450 °C and 500 °C, respectively. While annealing, the samples were sunk in the alumina powders for even thermal environment. About the suitable SP intensity, it is determined from the improvement of surface RCS and the variation of surface roughness. According to the Fig. 3 in references [16] , when the Almen intensity is increased from 0.3 + 0.15 mmA to 0.45 + 0.15 mmA, the improvement of near surface residual compressive stresses is not obvious. Besides, according to the Fig. 2 in this paper, after the first SP with 0.3 mmA using cast steel balls, the surface roughness of the sample is remarkable, if improving the Almen intensity, the surface roughness will be increased rapidly and result in crack initiation and growth in surface, which will deteriorate the fatigue properties. Consequently, the Almen intensity of 0.3 mmA is suitable in this work. 2.3 Measurements and calculation The X-ray diffraction (XRD) patterns were obtained on Rigaku Ultima IV X-ray diffractometer, which was operated at 40 kV/40 mA with Cu Kα radiation ( λ = 1.54056 Å). The residual stresses and integral breadth of Ti (213) peak were investigated by X-ray Stress Analyzer (LXRD, Proto, Canada) with Cu Kα radiation, voltage of 30 kV, current of 25 mA, Ni filter. About the residual stresses measurement, the range of tilting angles (psi) is 0°–045° The elastic diffraction constants of Young's modulus ( E ) and Poisson ratio ( ν ) are 130 GPa and 0.31 respectively, and which represents the constants of (TiB + TiC)/Ti–6Al–4V and obtained from the X-ray Stress Analyzer according to the fitted diffraction peak when measuring the residual stresses. According to the residual stress calculation method, the planar stresses are determined finally. Considering the integral breadth, the instrumental effect has been eliminated from the measured profile and the relationship has been shown in formula (1) [18–20] , (1) β G h 2 = β G f 2 + β G g 2 , where subscripts G denotes the Gaussian components, and superscripts h, f, g denote the measured line profile, the structurally broadened profile and the instrumental profile respectively. After deconvolution, the structurally breadth of β G f is in fact mainly determined by the sample's microstrain. So the microstrain is given by formula (2) [18–20] . (2) ε = β G f 4 tan ( θ ) According to β G f of Ti (2 1 3) peak, the microstrain ( ɛ ) were calculated via formula (2) . In the process of microstrain calculation, the reflection of (2 1 3) peak was selected. Compared with other peaks, the reflection of (2 1 3) plane is a representative high angle reflection and the 2 θ of (2 1 3) reflection is about 139.4°, which can embody the variation of interplaner spacing ( d ) markedly. So the reflection of (2 1 3) plane is used for calculating the residual stresses via X-ray Stress Analyzer, and using the reflection of (2 1 3) to calculate the microstrain is reasonable. In order to obtain the depth distribution of above values, the thin top surface layers were removed one by one via the method of chemical etch with the solution of water, nitric acid, and hydrofluoric acid in proportion of 31:12:7. In order to obtain the relaxation behavior of RCS during isothermal annealing, the peened samples were etched to the subsurface layer with higher RCS, and then annealed. According to the method of chemical etching, and in the process of chemical etching, the structure of deep surface layers will not be destroyed, and the residual stress relaxation caused by etching is very weak. In addition, the residual stresses obtained by X-ray Stress Analyzer are planar stresses. Therefore, the residual stress relaxation resulted from chemical etching can be ignored. All the measurements were carried out at room temperature. 3 Results and discussion 3.1 The surface roughness and microhardness after the first and second SP The surface roughness of titanium matrix composites after first and second SP have been measured and shown in Fig. 2 . From the figure, it reveals that after the second SP, the surface roughness decline obviously and the second SP makes the surface smooth after first SP. In addition, the microhardness at the surface after first and second SP has also been tested, and the values are 600.8 HV, 654.4 HV respectively. Compared the microhardness, it can be found that although the microhardness is improved, the increase is gently and unconspicuous. Therefore, in this work the main aim of the second SP is making the surface smooth after first SP. 3.2 XRD analyses of the samples before and after SP The XRD patterns of 8% (TiB + TiC)/Ti–6Al–4V composite before and after SP have been shown in Fig. 3 . It can be seen that the diffraction peaks of Ti, TiB and TiC are shown obviously. After SP, there is no new phase generates and the locations of all peaks are the same as the sample before SP. In addition, it is shown that the diffraction peaks of peened sample become wider, which results from the refined crystal domain sizes. In the process of SP, a great amount of small balls with high kinetic energy impacting on the surface of samples causes elastic and plastic deformation, and crystal domain sizes refinement at surface layers [21–25] . 3.3 Analyses of residual stresses relaxation and microstrain variations after isothermal annealing Fig. 4 represent the depth distributions of the residual stresses of the shot peened 8% (TiB + TiC)/Ti–6Al–4V composites after isothermal annealing about 2 h at 350 °C, 400 °C, 450 °C and 500 °C, respectively. Before annealing, the samples are etched to the subsurface layer (about 25 μm) with higher RCS. From Fig. 4 , it can be found that the RCS are relaxed in the whole deformation layers. And the higher the temperature, the more obvious the stress relaxation is. When the annealing temperature reaches 500 °C, the RCS almost are completely relaxed at the top surface. The maximum residual compressive stresses (MRCS), the depths of maximum residual compressive stresses (DMRCS) and the total depths of residual compressive stresses (DRCS) reduce gradually, which have been displayed in Table 1 . These results are ascribed to thermal recovery process and dynamic recrystallization of deformation layers at elevated temperatures [26,27] . Besides, the surface residual compressive stresses (SRCS) have also been shown in Table 1 . The depth distributions of microstrain of the shot peened composites after isothermal annealing have been shown in Fig. 5 . With the increase of depth, the microstrain decline gradually, and it is obvious that the microstrain of all samples are most severe at the top surface under different annealing temperatures. Compared to as-peened sample, after isothermal annealing, the microstrain decrease rapidly, and the higher the temperature, the more obvious the microstrain decline is, which result from the thermal recovery and dynamic recrystallization at elevated temperature. The higher the temperature, the more serious of thermal recovery and dynamic recrystallization are. In order to investigate the relationship between residual stresses relaxations and annealing time during isothermal annealing, residual stresses relaxation behaviors of peened samples are observed and shown in Fig. 6 . Before annealing, the samples are etched to the subsurface layers (about 25 μm) with higher RCS, and the annealing times are 1, 2, 4, 8, 16, 32, 64, 128 min. From Fig. 6 , it can be seen that the relaxations depend on both time and temperature. Before annealing, the surface RCS values are about −850 MPa by measuring the etched samples. After annealed at 350 °C, 400 °C, 450 °C and 500 °C for 128 min respectively, the RCS are relaxed to −484 MPa, −285 MPa, −120 MPa and −71 MPa, and the decline rates are about 43.1% 66.5%, 85.9%, and 91.6%. So with the temperature increased and the annealing time prolonged, the residual stresses relaxed conspicuously. The RCS of the sample annealed at 500 °C is relaxed almost completely after 128 min, which is identical to the result showed in the Fig. 4 . The microstrain of the shot peened 8% (TiB + TiC)/Ti–6Al–4V during isothermal annealing at different time have been displayed in Fig. 7 . With the prolongation of annealing time, the microstrain decrease quickly, then decline gently and reach a minimum. It also can be revealed that the higher annealing temperature, the faster of decline rate is. Therefore when the annealing temperature at 500 °C, the values of microstrain decline sharply and reach the minimum, which are attributed to the effects of annealing time and temperature, especially elevated temperatures. Consequently, the annealing time and temperature are the main reasons for the thermal recovery and dynamic recrystallization [12,26,27] . In terms of microstrain in Figs. 5 and 7 , the marked difference between the microstrain distributions published in reference [16] and in this paper, which is ascribe to the isothermal annealing treatments. After isothermal annealing, the residual stresses are relaxed and the thermal recovery occurs, therefore, the depth of deformation layer become lower and the values of microstrain decrease remarkably. However, in reference [16] , the treatment is only SP and the microstrain distributions are different from the depth and values. 3.4 Discussion on the relaxation mechanism and the activation energy of residual stresses relaxation According to above analysis, the thermal relaxations of residual stress are mainly influenced by the annealing time and temperature. When the annealing temperature is fixed, the fast rate of thermal relaxations takes place in the initial stage of annealing time (see Fig. 6 ), which is according with Refs. [5,12] . Thermal relaxations of residual stresses are controlled by thermally activated mechanism and which can be investigation by a Zener–Wert–Avrami function as Eq. (3) [8–10,28] , (3) σ T , t R S σ o R S = exp − ( A t ) m where σ o R S is the initial residual stress before annealing, σ T , t R S is the residual stress during temperature T and time t . m is a numerical parameter depending on the dominant relaxation mechanism. A is a function depending on the material and temperature according to Eq. (4) , (4) A = B exp − Δ H k T where B is the material constant, k is the Boltzmann constant, Δ H is the activation enthalpy for the relaxation process. Based on Eq. (3) , in order to obtain the value of numerical parameter m , the diagram of log ( − ln ( σ T , t R S / σ o R S ) ) as the function of log ( t ) has been shown in Fig. 8 according to the measured data. As shown in Fig. 8 , it is revealed that the relationship between log ( − ln ( σ T , t R S / σ o R S ) ) and log ( t ) is an approximate linear relationship at a constant annealing temperature and the slope of straight line is m . And at different annealing temperatures, the slopes are almost the same and m = 0.4483. According to Eqs. (3) and (4) and the date in Fig. 8 , the activation enthalpy of the relaxation process has been obtained through regression analysis and Δ H = 2.92 eV. And the activation energy of residual stresses relaxation Q RS can also be calculated as 282 kJ/mol, which is bigger than the self diffusion activation energy of both α-Ti and β-Ti ( Q α-Ti = 170 kJ/mol, Q β-Ti = 153 kJ/mol) [29] . According to above discussion, thermal relaxation of the residual stress indicates that the relaxation mechanism is controlled by the thermal recovery and dynamic recrystallization process [9,12] . At the temperature of 500 °C, dislocation movements and rearrangement are adequate for relaxations of RCS, and the dislocation density in the deformation layer are reduced significantly. In the process of relaxation, the reinforcements play very important roles. One hand, the reinforcements act as sink sources of dislocations and prevent the dislocation gliding during annealing. On the other hand, high density dislocations around the reinforcement particles promote recrystallization [30–35] . Consequently, thermal relaxation of the residual stress is ascribed to the thermally activated gliding of dislocations in the thermal recovery and recrystallization process. The higher activation energy of residual stress relaxation compared to that of titanium self diffusion results from the hindrance effects of the reinforcements during dislocation gliding. 4 Conclusions The residual stress relaxation behaviors and the variations of microstrain in the shot peened layer of (TiB + TiC)/Ti–6Al–4V composite were investigated during isothermal annealing. The results revealed that the residual stresses were relaxed in the whole deformation layers. At different annealing temperature, the RCS relaxed and the higher temperature, the more obvious of relaxations were. The maximum stress relaxation rate occurred mainly in the initial stage of annealing. Based on the discussion, the relaxation of residual stresses was ascribed to the thermally activated gliding of dislocations in the thermal recovery and recrystallization process. The thermal relaxation process of residual stresses was analyzed by applying the Zener–Wert–Avrami function. According to the function, the numerical parameter depending on the dominant relaxation mechanism ( m ), the activation enthalpy (Δ H ) and the activation energy ( Q RS ) of residual stresses relaxation were obtained as 0.4483, 2.92 eV and 282 kJ/mol respectively. It was found that Q RS was bigger than the self diffusion activation energy of both α-Ti and β-Ti. The main reason of the activation energy increment was the hindrance effects of the reinforcements during dislocation gliding. Acknowledgement The authors are indebted to Professor Weijie Lu for the kindly guidance. References [1] A.K. Kuruvilla K.S. Prasad V.V. Bhanuprasad Y.R. Mahajan Scr. Metall. Mater. 24 1990 873 878 [2] S. Ranganath J. Mater. Sci. 32 1997 1 16 [3] M.E. Fitzpatrick P.J. Withers A. Baczmanski M.T. Hutchings R. Levy M. Ceretti A. Lodini Acta Mater. 50 2002 1031 1040 [4] L. Wagner Mater. Sci. Eng. A 263 1999 210 216 [5] A. Evans S.B. Kim J. Shackleton G. Bruno M. Preuss P.J. Withers Int. J. Fatigue 27 2005 1530 1534 [6] H. Lee S. Mall Mater. Sci. Eng. A 366 2004 412 420 [7] I. Nikitin I. Altenberger H.J. Maier B. Scholtes Mater. Sci. Eng. A 403 2005 318 327 [8] P. Juijerm I. Altenberger Scr. Mater. 55 2006 1111 1114 [9] I. Nikitin M. Besel Scr. Mater. 58 2008 239 242 [10] M.C. Berger J.K. Gregory Mater. Sci. Eng. A 263 1999 200 204 [11] W. Luan C. Jiang V. Ji Mater. Trans. 50 2009 1499 1501 [12] B.X. Feng X.N. Mao G.J. Yang L.L. Yu X.D. Wu Mater. Sci. Eng. A 512 2009 105 108 [13] S.C. Tjong Z.Y. Ma Mater. Sci. Eng. R 29 2000 49 113 [14] S. Ranganath M. Vijayakumar J. Subrahmanyam Mater. Sci. Eng. A 149 1992 253 257 [15] X. Zhang W. Lu D. Zhang R. Wu Y. Bian P. Fang Scr. Mater. 41 1999 39 46 [16] L. Xie C. Jiang W. Lu K. Zhan Y. Chen Mater. Sci. Eng. A 528 2011 3423 3427 [17] X. Zhang D. Liu Int. J. Fatigue 31 2009 889 893 [18] S. Vives E. Gaffet C. Meunier Mater. Sci. Eng. A 366 2004 229 238 [19] J.I. Langford J. Appl. Crystallogr. 11 1978 10 14 [20] W. Luan C. Jiang H. Wang J. Panardie D. Chen Mater. Sci. Eng. A 480 2008 1 4 [21] G. Liu J. Lu K. Lu Mater. Sci. Eng. A 286 2000 91 95 [22] H.W. Zhang Z.K. Hei G. Liu J. Lu K. Lu Acta Mater. 51 2003 1871 1881 [23] Y. Lin J. Lu L. Wang T. Xu Q. Xue Acta Mater. 54 2006 5599 5605 [24] T. Roland D. Retraint K. Lu J. Lu Scr. Mater. 54 2006 1949 1954 [25] T. Roland D. Retraint K. Lu J. Lu Mater. Sci. Eng. A 445–446 2007 281 288 [26] R.Z. Wang Mater. Mech. Eng. 5 1988 19 23 [27] Y.K. Gao Rare Metal Mater. Eng. 33 2004 1209 1212 [28] P. Juijerm I. Altenberger B. Scholtes Int. J. Fatigue 29 2007 1374 1382 [29] F. Dyment H. Kimura O. Izumi Proceedings of the Fourth International Conference on Titanium, TMS Warrendale PA 1980 519 528 [30] F.J. Humphreys P.N. Kalu Acta Metall. Mater. 35 1987 2815 2829 [31] M. Ferry P.R. Munroe Compos. Part. A: Appl. Sci. Manuf. 35 2004 1017 1025 [32] P. Cavaliere E. Evangelista Compos. Sci. Technol. 66 2006 357 362 [33] W. Luan C. Jiang V. Ji Y. Chen H. Wang Mater. Sci. Eng. A 497 2008 374 377 [34] T. Ungár S. Ott P.G. Sanders A. Borbely J.R. Weertman Acta Mater. 46 1998 3693 3699 [35] W. Luan C. Jiang V. Ji Mater. Sci. Eng. A 504 2009 124 128