Microtextural effects on mechanical properties of duplex microstructures in (α+β) titanium alloys

Microtextural effects on mechanical properties of duplex microstructures in (α+β) titanium alloys

Materials Science and Engineering A263 (1999) 137 – 141 Microtextural effects on mechanical properties of duplex microstructures in (a+b) titanium al...

1MB Sizes 0 Downloads 26 Views

Materials Science and Engineering A263 (1999) 137 – 141

Microtextural effects on mechanical properties of duplex microstructures in (a+b) titanium alloys J. Lindemann *, L. Wagner Technical Uni6ersity of Brandenburg at Cottbus, 03013 Cottbus, Germany

Abstract Duplex microstructures with 20% ap content were produced in Ti – 6Al – 7Nb by a variety of thermomechanical treatments including variations in deformation mode with or without intermediate short time b annealing, and variations in cooling rate from the duplex anneal. The fatigue results indicate a correlation between facetted fatigue fracture origins, microtexture and fatigue strength at R =0.1. © 1999 Elsevier Science S.A. All rights reserved. Keywords: Duplex microstructures; Thermomechanical treatments; Microtextural effects

1. Introduction

2. Experimental

Duplex microstructures (primary a in lamellar (a+ b) matrix) are superior to fully lamellar microstructures regarding tensile ductility and resistance to microcrack growth. Unlike fully lamellar structures, duplex structures suffer from a so-called anomalous mean stress sensitivity, i.e. poor high-cycle fatigue strength at low tensile mean stresses [1,2]. Earlier work has shown that crystallographic texture and loading direction have a predominant influence on the mean stress sensitivity [3]. Normal mean stress sensitivity was found only on unidirectionally rolled material with a basal/transversal type of texture if loaded in rolling direction (RD). The fracture surfaces of fatigue specimens of this condition exhibited a fine structure at the crack nucleation site while all other conditions showed large facets of the order of a few hundred microns in size [4]. The present investigation was performed to understand the origin of these large facets and to determine methods of preventing their occurrence.

The investigation was performed on the surgical implant (a+ b) titanium alloy Ti–6Al–7Nb. Duplex microstructures with 20% ap volume fractions and 15 mm ap grain sizes were produced by different thermomechanical treatments (Fig. 1). For the conventional processing route A (Fig. 1a), the deformation mode was varied by using either unidirectional rolling or swaging in the (a+ b) phase field. In addition, the cooling rate from the duplex anneal was varied to change the strength of the lamellar component. In route B (Fig. 1b), (a+ b) deformation was interrupted by an intermediate short time b anneal. For short time b annealing, DC resistance heating (1500 A, 5 V) was used. Details of the thermomechanical treatments are given elsewhere [5,6]. Optical microscopy on unetched mechanically polished samples was performed using polarized light to study qualitatively the occurrence and sizes of similarly oriented regions in the duplex microstructures. Tensile and fatigue testing of the unidirectionally rolled material were performed in transverse direction (TD). Axial fatigue tests were conducted on hour glass shaped electrolytically polished specimens using a servohydraulic testing machine. Testing was performed at frequencies of roughly 60 Hz and stress ratios of R=

* Corresponding author.

0921-5093/99/$ - see front matter © 1999 Elsevier Science S.A. All rights reserved. PII: S 0 9 2 1 - 5 0 9 3 ( 9 8 ) 0 1 1 7 2 - 1

138

J. Lindemann, L. Wagner / Materials Science and Engineering A263 (1999) 137–141

Fig. 1. Thermomechanical treatment for producing duplex microstructures in (a + b) titanium alloys. (a) Coarse b grain size processing (route A). (b) Fine b grain size processing (route B).

Table 1 Tensile properties of duplex microstructures in Ti–6Al–7Nb (a+b) working*

**

s0.2 (MPa)

UTS (MPa)

El (%)

RA (%)

Rolled

AC WQ AC AC

920 1030 855 880

995 1120 965 1020

13.5 14.7 16.0 16.5

39.1 41.9 33.6 36.2

Swaged Intermediate short time b annealed+rolled

* Prior treatment: 0.5 h 1050°C/AC. ** Duplex anneal: 1 h 1000°C (AC: air cooled; WQ: water-quenched).

− 1 and R=0.1. Fracture surfaces were studied by SEM.

3. Results and discussion Duplex microstructures of Ti – 6Al – 7Nb are shown in Fig. 2 comparing material which was air cooled (Fig. 2a) with material which had been water-quenched (Fig. 2b) from the duplex anneal. While no effect of deformation mode or intermediate rapid b annealing was found on microstructure, the cooling rate from the duplex anneal clearly affects the fineness of the lamellar

matrix (compare Fig. 2a with Fig. 2b). Tensile properties of the various conditions are shown in Table 1. As seen in Table 1, the cooling rate from the duplex anneal has the most marked effect on yield stress and tensile strength. This effect is clearly related to the strength of the lamellar portion in the duplex microstructures since the strength of the ap phase is not affected by cooling rate [6]. The S–N curves of the rolled and swaged conditions of processing route A are shown in Fig. 3. Swaging leads to a markedly higher fatigue strength at R= 0.1 than unidirectional rolling (Fig. 3a) while the fatigue strengths at R= − 1 are hardly affected by deformation mode. Thus, the mean stress sensitivity of

J. Lindemann, L. Wagner / Materials Science and Engineering A263 (1999) 137–141

Fig. 2. Duplex microstructures in Ti–6Al–7Nb. (a) AC. (b) WQ.

the fatigue strength is normal for swaging but anomalous for rolling. The improvement of the fatigue

139

strength by changing the cooling rate from the duplex anneal from AC to WQ is very pronounced as seen on rolled material in Fig. 3b. Fracture surfaces of fatigue specimens (R=0.1) can be seen in Fig. 4. Large facets in the regions of crack nucleation in rolled material (Fig. 4a) are observed independent of cooling rate from the duplex anneal while such facets are absent or less significant in swaged material (compare Fig. 4a with Fig. 4b). Similarly oriented regions or microtextures (Fig. 5) are fairly coarse in rolled material (Fig. 5a), but much reduced in size in swaged material (Fig. 5b). In contrast to the coarse prior b grain sizes (Fig. 6a) after a conventional solution anneal (route A, Fig. 1a), very fine prior b grains (Fig. 6b) are present after short time b annealing of (a+b) deformed material (route B, Fig. 1b). Consequently, subsequent (a+ b) working leads to pancake grains which are coarse for conventional b annealing (Fig. 7a) or fine for intermediate short time b annealing (Fig. 7b). As seen in polarized light in Fig. 8, these dimensions apparently carry over after the duplex anneal as was also reported in Ref. [7]. This is indicated by similarly oriented regions which are locally fairly large for coarse prior b grains (Fig. 8a) and small for fine prior b grains (Fig. 8b). The changes in dimensions of the b grains owing to unidirectional

Fig. 3. S –N curves of duplex microstructures in Ti – 6Al – 7Nb. (a) (b) WQ.

Fig. 4. Fracture surfaces (R= 0.1) at fatigue crack nucleation site. (a) Rolled. (b) Swaged.

140

J. Lindemann, L. Wagner / Materials Science and Engineering A263 (1999) 137–141

Fig. 5. Microtextures (polarized light) of duplex microstructures in Ti – 6Al – 7Nb. (a) Rolled. (b) Swaged.

Fig. 6. Prior b grain sizes. (a) Conventional solution anneal. (b) Intermediate short time b anneal.

Fig. 7. Deformed microstructures after rolling. (a) Coarse prior b grain size. (b) Fine prior b grain size.

Fig. 8. Microtextures (polarized light) of duplex microstructures in Ti – 6Al – 7Nb. (a) Coarse prior b grain size. (b) Fine prior b grain size.

deformation (8 = −1.4) was estimated using the assumptions that: (1) grains were initially equiaxed; (2) volume remains constant and; (3) the change in grain shape is geometrically similar to the change in shape of the deformed work piece. According to this analysis, the starting equiaxed grains with average grain diameters of 600 mm deform into pancakes with lengths of L =1500 mm, widths of W= 650 mm and heights of H =150 mm. These estimates were found to be in good

agreement with the experimentally determined values of the largest microtextural regions. Locally coarse microtextures appear to affect the resistance to fatigue crack nucleation at R =0.1 in the HCF regime (Fig. 9), since the fatigue strength for the fine textured material (Fig. 9b) is higher than for the coarse textured material (Fig. 9a). Intermediate short time b annealing is clearly seen to prevent the facets on the fatigue fracture surfaces (Fig. 10, compare Fig. 10b

J. Lindemann, L. Wagner / Materials Science and Engineering A263 (1999) 137–141

141

Fig. 9. S – N curves (R =0.1) of duplex microstructures in Ti – 6Al – 7Nb. (a) Coarse prior b grain size. (b) Fine prior b grain size.

Fig. 10. Fatigue crack nucleation (R =0.1) of duplex microstructures in Ti – 6Al – 7Nb. (a) Coarse prior b grain size. (b) Fine prior b grain size.

with Fig. 10a). Thus, it is concluded that refining the microtextural unit size by deformation mode or intermediate short time b annealing can improve the HCF strength by increasing the resistance to fatigue crack nucleation.

[2]

Acknowledgements

[4]

Part of this investigation was supported by the Deutsche Forschungsgemeinschaft (DFG). The authors would like to thank Peter Brehm Chirurgie-Mechanik, Weisendorf for providing the titanium alloy. They would also like to thank Dr P.E. Jones for helpful comments on the manuscript.

[3]

[5]

[6] [7]

References [1] R.K. Steele, A.J. McEvily, The effect of mean stress on fatigue

.

behavior of Ti – 6Al – 4V alloy, in: J.C. Williams, A.F. Belov (Eds.), Titanium and titanium alloys, Plenum Press, NY, 1982, p. 589. A. Atrens, W. Hoffelner, T.W. Duerig, J.E. Allison, Scr. Metall. 17 (1983) 601. S. Adachi, L. Wagner, G. Lu¨tjering, Influence of microstructure and mean stress on fatigue strength of Ti – 6Al –4V, in: G. Lu¨tjering, U. Zwicker, W. Bunk (Eds.), Titanium science and technology, DGM, Oberursel, 1985, p. 2139. S. Adachi, L. Wagner, G. Lu¨tjering, Influence of mean stress on fatigue strength of Ti – 6Al – 4V, in: H.J. Mc Queen, J.-P. Bailon, J.I. Dickson, J.J. Jonas, M.G. Akben (Eds.), Proc. ICSMA 7, Pergamon Press, Oxford, 1985, p. 2117. U. Holzwarth, J. Kiese, L. Wagner, Improvement of the fatigue life of the surgical implant titanium alloy Ti – 6Al – 7Nb through shot peening and roller-burnishing, In: Mechanical properties of implant materials, DVM, 1998, p. 75 (in German). J. Lindemann, L. Wagner, Mat. Sci. Eng. A234–236 (1997) 1118. A.P. Woodfield, M.D. Gorman, R.R. Corderman, J.A. Sutliff, B. Yamrom, Effect of microstructure on dwell fatigue behavior of Ti-6242, in: P.A. Blenkinsop, W.J. Evans, H.M. Flower (Eds.), Titanium 95-Science and Technology, The Institute of Materials, London, 1996, p. 1116.