- Email: [email protected]
Crystallographic texture and lattice strain evolution during tensile load of swaged brass
Materials Science & Engineering A 711 (2018) 149–155
Contents lists available at ScienceDirect
Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea
Crystallographic texture and lattice strain evolution during tensile load of swaged brass
MARK
⁎
Nowfal Al-Hamdanya, , Heinz-Günter Brokmeiera, Weimin Ganb a b
Institut für Werkstoffkunde und Werkstofftechnik, TU Clausthal, Agricolastr. 6, D-38678 Clausthal-Zellerfeld, Germany German Engineering Materials Science Center at MLZ, Helmholtz-Zentrum Geesthacht, D-85748 Garching, Germany
A R T I C L E I N F O
A B S T R A C T
Keywords: Crystallographic texture Lattice strain Swaging Brass Neutrons diffractions
Evolutions of texture and lattice strain of swaged brass samples were investigated by neutron diffraction at STRESS-SPEC under tensile deformation using a unique tension/compression rig. The two phased sample BS1 (61% α-brass and 39% β-brass) became 100% α-brass after 400 °C annealing (sample BS2). The starting texture of the as-received material BS1 was the typical < 111 > , < 200 > double fiber. This texture develops firstly by in-situ tension to a moderate strengthening. After annealing (BS2) the < 111 > fiber survives with surprisingly high strength and develops by in-situ tension a very strong < 111 > fiber of 39 mrd. Line broadening and lattice strain behaviour shows the development of the elastic strain and plastic strain.
1. Introduction Swaging is a highly used manufacturing process to improve materials properties, such as fatigue behaviour, strength and corrosion properties. Different types of swaging are in use, which is basically forging. Most of the swaging types are ideal for mass production. One in particular, rotary swaging (Fig. 1) can be used to reduce the cross section of rods and tubes and also to increase the strength [1]. An example of this type of swaging can be applied to the weight reduction of bike axles or high pressure tubes. The property change after swaging is based on grain refinement, strengthening of the crystallographic texture and generation of compressive stresses on the surface. It should be noted that most swaging processes are cold working processes, which reduces the costs. Rotary swaging in ideal conditions belongs to uniaxial deformation, same as wire and rod drawing. Uniaxial deformation is known for the high texture symmetry, which is defined by ideal fiber texture with a preferred fiber axis in transport direction and a rotational freedom around this fiber axis [2,3]. From Fig. 1, it can be deduced that due to the forging matrix which can consist of 2, 3, 4 or 6 segments a little variation from the ideal fiber texture can be observed. The texture symmetry is related to the process but the fiber type depends on the crystal symmetry and also on the stacking fault energy of the swaged material. FCC metals typically develop double fibers of < 111 > and < 100 > under uniaxial tension. The ratio depends on the stacking fault energy [2,3]. BCC metals show typically a < 110 > fiber [2,3]. HCP metals tend to have either the < 10.0 > or
⁎
the < 11.0 > fiber sometimes a double fiber < 10.0 > , < 11.0 > generates. A mixture of materials in case of precipitations or composites needs a co-deformation during swaging. In general it is known that a second phase influences texture development during deformation and recrystallization often in a decrease of the texture sharpness as well as the final mechanical properties [4]. Neutron diffraction allows the texture determination of the average texture over the whole cross section of rotary swaged samples, which is ideal for technical applications. This is possible because of the high penetration power of neutrons against standard X-rays. Moreover, the possibility to get complete pole figures makes the error of texture sharpness low, so that the expected variations in texture sharpness can be obtained in high quality [5–7]. Kalu et al. [8] have investigated the texture evolution in swaged Cu wire and Cu-Nb/Ti composite. Both materials developed < 111 > and < 100 > fiber texture. Cu-Nb/Ti composite had generally less texture sharpness than those of pure Cu. Hupalo et al. [9] have studied the evolution of texture and microstructure during cold swaging and recrystallization of Oligocrystalline INCOLOY® MA 956. They have used light optical microscopy, transmission electron microscopy, X-ray diffraction, electron back scattered diffraction (EBSD) and Vickers microhardness testing. A sharp < 110 > fiber texture was developed during plastic deformation after starting from a strong < 111 > fiber. They observed that recrystallization did not change the texture significantly. Gue et al. [10] have evaluated the texture during recrystallization of cold swaged Ti-Nb-Ta-Zr-O alloy (TNTZO). The texture measurements showed a pronounced < 110 > fiber along the
Corresponding author. E-mail address: [email protected] (N. Al-Hamdany).
https://doi.org/10.1016/j.msea.2017.11.047 Received 15 June 2017; Received in revised form 2 November 2017; Accepted 13 November 2017 Available online 14 November 2017 0921-5093/ © 2017 Elsevier B.V. All rights reserved.
Materials Science & Engineering A 711 (2018) 149–155
N. Al-Hamdany et al.
Fig. 3. Dimensions of the brass tensile sample.
and 500 °C before the ductility then recovered at higher temperatures. STRESS-SPEC is equipped with a tensile/compression machine having a rotated rig (χ tilt and ϕ rotation) with a maximum load of 50 kN. Different sample geometries are possible to be measured by this machine. An additional light furnace can be installed around the loaded sample with temperature up to 1000 °C [14,15]. Commercial Cu-36%Zn brass (Fig. 2) was chosen firstly due to expected activation of deformation twins during deformation and secondly because of the mixture of α- and β-brass. This alloy has excellent cold workability. Typical applications of Cu-36%Zn are radiator cores, tanks, lamp fixtures, fasteners, locks, hinges, plumbing accessories, pins, rivets [16]. The aim of this experiment was to investigate the influence of the existing preferred orientation on texture development of α brass during tensile load for α+β brass and α brass. Additionally, the evolution of line broadening (defect evolution) and line position (lattice strain) of both samples BS1 and BS2 were investigated during tensile loading.
Fig. 1. Scheme of the swaging process with 4 segments.
deformation axis in the cold swaged TNTZO alloy which is the typical fiber texture for bcc materials after drawing or swaging. The < 110 > fiber texture was gradually replaced by random orientations by increasing the annealing time. Gan et al. [11,12] have investigated the bulk and local texture evolution of pure Mg processed by rotary swaging. They started from diameter of 10 mm going down to 4.5 mm which is corresponding to a deformation degree of 1.58. A main {00.2} basal fiber texture in rotary swaging processed Mg was observed. Basal planes tend to be more homogeneous with the increase of the process pass. Texture was relatively inhomogeneous from the surface to the center at rods which indicate a non-uniform deformation. Bache et al. [13] have characterized an advanced Nickel based superalloy RR1000 post cold work by swaging. Cold work of 30% was subjected to cylindrical bars of 27 mm diameter. Beside the reduction in grain size, a relatively strong < 111 > fiber texture parallel to the longitudinal bar axis has been confirmed by EBSD. The effect of cold work on bulk mechanical properties was observed from the increase in the ultimate tensile strength (UTS) and yield tensile strength (YTS), but also resulted in reduced ductility. The ductility reduced between room temperature
2. Materials and experimental procedure Two samples were produced from commercial Cu-36%Zn brass. One sample (BS1) was swaged up to a deformation degree of φ = 3 having a reduction in cross section from 32 mm in diameter to 7 mm. Sample 2 (BS2) was additionally annealed for 2 h at 400 °C after swaging. During this annealing β-brass was solved and BS2 consists only of α-brass. Phase analysis was carried out with hard X-rays at the high energy beamline HEMS@Petra III. The microstructure was examined for both Fig. 2. Phase diagram of Cu-Zn alloy [17].
150
Materials Science & Engineering A 711 (2018) 149–155
N. Al-Hamdany et al.
Fig. 4. BS1 tensile sample (a) installed in the loading machine (b) installed on the Eulerian cradle at STRESS-SPEC.
STRESS-SPEC (Fig. 4a). Both samples were tensioned until failure, sample BS1 up to 10.5 kN and sample BS2 up to 5.7 kN). The loading rate was 10 N/s. Two different types of experiments were performed, in-situ lattice strain investigations and ex-situ texture investigations. For lattice strain investigations, diffraction patterns were collected during tensile load. The loading direction was parallel to the transport direction during swaging. The neutron measurements were carried out with a wavelength of 1.5 Å obtained from Germanium (Ge (311)) monochromator. An area detector of 300 mm × 300 mm was used with a sample to detector distance of 1065 mm. The detector position of 2θ = 45° allows the simultaneous measurement of the reflections (111) and (200). The beam was guided by primary slit of 5 mm in diameter and on the detector side by at FWHM = 5 mm radial collimator to get the average information over the whole sample thickness. The exposure time was 5 min. Under load control diffraction patterns were collected for BS1 until 8 kN in steps of 1 kN and in steps of 0.5 kN from 8 kN until ultimate load. For BS2, diffraction patterns were collected until 2 kN in steps of 0.5 kN and in steps of 0.25 kN from 2 kN until 5.5 kN. For pole figure measurements, the samples were screwed on a plate and installed on the Eulerian cradle (Fig. 4b). The pole figure measurements were carried out using neutrons wavelengths of 1.66 Å and 1.1 Å obtained from pyrolytic graphite (PG) monochromator using (0004) and (0006) planes. Dual wavelength was chosen to save counting time by using only one detector position. An area detector of 300 mm × 300 mm was used with a sample to detector distance of 1065 mm. Three pole figures were measured with the detector position of 2θ = 50° for (111), (200) and (220). According to the sample to the detector distance and the detector position, six sample tilts in χ were needed. For each χ-position a continuous scanning around φ was performed. An exposure time of 10 s per 2.5° continuous φ rotation gives sufficient intensities. 3. Results Quantitative phase composition was calculated using MAUD [19]. The sample BS1 consists of 61% α-brass and 39% β-brass (Fig. 5), while sample BS2 after annealing had only α-brass. Fig. 6 shows the microstructure of BS1and BS2 before loading. The microstructure of BS1 consists of deformed α-grains in light colour and deformed β-grains in dark colour. BS2 has a recrystallized α-brass microstructure. The stress strain curves for BS1 and BS2 are presented in Fig. 7 giving the experimental values for load in kN and for tension in mm. In Table 1 the mechanical properties of both samples are summarized. Red dots describe the positions for in-situ lattice strain measurements.
Fig. 5. Diffraction pattern obtained from hard X-rays for (a) BS1 and (b) BS2.
BS1 and BS2 before and after loading. The microstructure samples were etched after grinding and polishing with a solution of 100 ml distilled water 20 ml hydrochloric acid and 5 g iron (III) chloride [18] Two standard tensile samples were machined following DIN 50515 having d0 = 4.03 mm, L0 = 20 mm, d1 = M6, Lc = 24 mm and Lt = 40 mm (Fig. 3). The tensile tests have been carried out using the tensile machine at
151
Materials Science & Engineering A 711 (2018) 149–155
N. Al-Hamdany et al.
Fig. 6. Microstructure of BS1 (a) and (BS2) (b) before and after loading.
12000
a
Loading curve Single shot
6000
Load (N)
Load (N)
5000
8000
4000 2000 0 0.0
b
6000
10000
4000 3000
Loading curve Single shot
2000 1000
0.2
0.4
0.6
0.8
0
1.0
Deformation (mm)
0
1
2
3
4
5
6
7
Deformation (mm)
Fig. 7. Stress strain curve and single shots for BS1 (a) and BS2 (b).
defects produced by the swaging process. The reduction in the number of defects after heat treatment can be indicated from the reduction of the full width of half maximum (FWHM) at 0 kN (Fig. 8a). In principal one can estimate the defect density from the slope of the WilliamsonHall Plot [20]. For 0 kN it can be seen that the slope of BS1 is much higher than of BS2, (Fig. 8b). Moreover, BS1 has two phases, the FCC αbrass and the BCC β-phase. The β phase is the harder phase which also participates to improve the mechanical properties. Fig. 8a shows the evolution of the line broadening during tensile test. In BS1 one can see a small increase in the line broadening of the (111) planes starting after 8 kN while for (200) planes nearly no change is visible. During elastic strain no increase of the line broadening was
Table 1 Mechanical properties of BS1 and BS2. Sample
Young's modulus (GPa)
Yield stress (MPa/kN)
Ultimate tensile strength (MPa/ kN)
Elongation (%)
BS1 BS2
123 104
770/9.6 233/2.9
820/10.5 455/5.7
2.4 35
The yield stress in BS1 is higher than in BS2 as well as the ultimate tensile stress. BS2 alternatively has a higher percentage of elongation than BS1 as a result of the heat treatment of BS2 releasing most of the 152
Materials Science & Engineering A 711 (2018) 149–155
N. Al-Hamdany et al.
Fig. 8. Line broadening and lattice strain; (a) FWHM against load, (b) FWHM against 2θ obtained from hard X-rays diffraction (Williamson-Hall Plot) (c) Lattice strain against applied load.
a1
(111)
a2
Pmax= 22 mrd
(200)
b1
Pmax= 4.4 mrd
(111)
Pmax= 29 mrd
b2
(200)
Pmax= 6.5 mrd
Fig. 9. BS1 (111) and (200) pole figures of α-brass; a) before load (only swaged), b) after ultimate load.
Fig. 8c shows the evolution of the lattice strain versus the load. The difference in the slops in the elastic region before the yield point of the strain stress curve is based on the elastic anisotropy. Different crystal lattice planes have different elastic properties [21]. For BS1 the experimental value for α-brass, calculated from Fig. 8c, for E111 is 176.43 GPa and for E200 is 135.86 GPa. For BS2 E111 is 196.81 GPa and E200 is 114.63 GPa. The experimental elastic properties are different to the theoretical values calculated for alpha brass (30% Zn) after the Kröner model. The theoretical elastic properties for alpha brass (30% Zn) are (E111 is 147.75 GPa and E200 is 88.93 GPa) [22]. The reasons for these differences are the alloying elements, the crystallographic texture
expected while during plastic strain new defects are generated with hardening. One also has to take into account that the swaging process up to deformation degree of φ = 3 causes a high number of defects. The (111) oriented grains start deforming earlier than (200) oriented grains. Line broadening in sample BS2 shows a much different behaviour. For (111) planes a moderate increase of the line broadening start at 2.5 kN while for (200) oriented grains a rapid increase is seen. This agrees with the Williamson-Hall Plot in Fig. 8b, showing a strong increase of the defect density in sample BS2. Continuous hardening behaviour during in-situ loading takes place, which also correlates with the stress strain curve. 153
Materials Science & Engineering A 711 (2018) 149–155
N. Al-Hamdany et al.
a1
(111)
Pmax= 21 mrd
a2
(200)
b1
Pmax= 4.7 mrd
(111)
Pmax= 39 mrd
b2
(200)
Pmax= 6.1 mrd
Fig. 10. BS2 (111) and (200) pole figures of α-brass; a) before load (swaged + annealed), b) after ultimate load. Fig. 11. ODF of α-brass for BS1; (a) before load (only swaged), (b) after ultimate load.
direction (in uniaxial tension) moves towards alignment with the direction of the applied stress [2,3]. To analyse the fibers separately the orientation distribution function (ODF) has been calculated. The ODFs for all samples have been calculated from three pole figure using the iterative series expansion method with Lmax = 35 as degree of series expansion. Due to the stronger texture BS2 after load was calculated with Lmax = 55 [25]. Figs. 11 and 12 show the ODF-section φ2 = 0° which includes both the < 100 > and the < 111 > fiber. The < 100 > fiber axis is at φ1 = 90°, Φ = 0°, φ2 = 0°; the fiber itself runs at φ1 = 0° and φ2 = 0° from Φ = 0° to 90°. For the < 111 > fiber only the cross section of the fiber can be seen at φ2 = 0°, φ1 = 55°, Φ = 45°. In general, it is known that the relative concentrations of the two fiber components depend on the stacking fault energy. Other influences on the texture evolution, together with the stacking fault energy, are the process temperature, the deformation rate, the initial texture, the grain size, and the purity of the material [2].
and the grain morphology [23]. The plastic behaviour of (111) and (200) oriented grains in both samples BS1 and BS2 (Fig. 8c) is different to the one from the macroscopic stress strain curve (Fig. 7). The (200) oriented grains starts to yield earlier than (111) oriented grains in both BS1 and BS2 (Fig. 8c). The reason is different critical resolved shear stress is required for the initiation of glide on a substantial scale [24] The pole figures of α-brass are shown in Fig. 9 (BS1 before loading and after ultimate load and in Fig. 10 (BS2 before loading and after ultimate load). The BS1 swaging texture shows the typical strong < 111 > and weaker < 100 > double fibers texture which is strengthened during tensile load. The (111) pole figure sharpness increases from 22 mrd to 29 mrd (mrd – multiples of random distribution). BS2 has only a < 111 > single fiber which survives after the heat treatment of BS1 at 400 °C for 2 h. With heat treatment the β phase dissolved and α phase recrystallized to a preferred < 111 > orientation. BS2 has similar texture sharpness before load than BS1, which is unexpected that 400 °C annealing has no influence on the texture sharpness of the < 111 > fiber. But tensile load of only 5.7 kN (455 MPa) leads to an unexpected increase of (111) pole figure sharpness to 39 mrd. During the slip process, the crystal lattice also rotates so that the active slip
4. Conclusions The crystallographic texture and lattice strain evolution of swaged brass during tensile load using neutrons diffractions has been 154
Materials Science & Engineering A 711 (2018) 149–155
N. Al-Hamdany et al.
Fig. 12. ODF of α-brass for BS2; (a) before load (swaged + annealed), (b) after ultimate load.
References
investigated. The diffraction information was obtained from the complete cross section of standard tensile test sample after DIN-norm. The sample BS1 consists of 61% α-brass and 39% β-brass while BS2 (heat treated sample) has only α-brass. The swaged sample BS1 showed higher strength (820 MPa) and lower elongation (2.4%) compared to BS2 (455 MPa and 35%). This was due to the removal of many defects produced by the swaging process as a result of heat treatment. Moreover, the BS1sample had two phases with the harder β-phase leading to increased strength. Line broadening studies have shown the formation or removal of defects within the material. During tensile testing, the BS1 sample had a minor increase in line broadening due to already containing a high number of defects. The BS2 sample showed a rapid increase in line broadening especially for the (200) plane. The difference in the elastic properties of the α-brass (111) and (200) planes in the BS1 and BS2 specimens was due to differences in crystallographic texture and grain morphology. Grains oriented with < 200 > parallel to the loading directions started to yield earlier than grains oriented with < 111 > parallel to loading directions in both BS1 and BS2. This was due to different critical resolved shear stresses for the different planes after which glide on a substantial scale was initiated. Both the BS1 and BS2 samples showed the expected increase in texture sharpness of the < 111 > fiber during tensile test. An unexpectedly strong increase in texture sharpness of the BS2 sample to approximately 39 mrd was measured. During the slip process, the crystal lattice also rotates so that the active slip direction (in uniaxial tension) moves towards alignment with the direction of the applied stress. More investigation will be made to understand the strong increase in the texture sharpness of the BS2 sample.
[1] ASM, Forming and Forging 14 ASM international, Metal Park, Ohio, 1993. [2] G. Wassermann, J. Grewen, Texturen metallischer Werkstoffe, Springer Verlag, Berlin, 1962. [3] H. Hu, Texture of metals, Texture 1 (1974) 233. [4] H.-G. Brokmeier, Experimental textures of Al-Pb, Al-Cu, Fe-Cu metal-metal composites, Mater. Sci. Eng. A 175 (1994) 131. [5] H.-J. Bunge, Z. Met. 76 (1985) 457. [6] H.-G. Brokmeier, Neutron diffraction texture analysis of multi-phase systems, Textures Microstruct. 10 (1989) 325. [7] H.-G. Brokmeier, Texture analysis by neutron diffraction, Mater. Sci. Forum 157–162 (1994) 59. [8] P.N. Kalu, L. brandao, F. Ortiy, O. Egungwu, F. Ige, On the texture evolution in swaged Cu-based wires, Scr. Mater. 38 (12) (1998) 1755. [9] M.F. Hupalo, A.F. Padiha, H.R.Z. Sandim, texture and microstructure evolution during cold swaging and recrystallization of oligocrystalline INCOLAY® MA 956, Solid State Phenom. 105 (2005) 315. [10] W.Y. Gue, H. Xing, J. Sun, X.L. Li, J.S. Wu, R. Chen, Evolution of microstructure and texture during recrystallization of cold-swaged Ti-Nb-Ta-Zr-O alloy, Metall. Mater. Trans. A 39A (2008) 672. [11] W.M. Gan, Y.D. Huang, R. Wang, Z.Y. Zhong, N. Hort, K.U. Kainer, N. Schell, H.G. Brokmeier, A. Schreyer, Bulk and local texture of pure magnisium processed by rotary swaging, J. Magnes. Alloy. 1 (2013) 341. [12] W.M. Gan, Y.D. Huang, R. Wang, A. Srinivasan, H.-G. Brokmeier, N. Schell, K.U. Kainer, N. Hort, Microstructures and mechanical properties of pure Mg processed by rotary swaging, Mater. Des. (2014) 83. [13] M.R. Bache, J. O׳Hanlon, P.J. Withers, D.J. Child, M.C. Hardy, Characterization of an advanced nickel based superalloy post cold work by swaging, Metals (2016), http://dx. doi.org/10.3390/met6030054. [14] M. Hofmann, R. Schneider, G.A. Seidle, J. Rebelo-Kornmeier, R.C. Wimpory, U. Garbe, H.G. Brokmeier, The new materials science diffractometer STRESS-SPEC at FRM-II, Physica B 385–386 (2006) 1035. [15] H.-G. Brokmeier, W.M. Gan, C. Randau, M. Völler, J. Rebelo-Kornmeier, M. Hofmann, Texture analysis at neutron diffractometer STRESS-SPEC, Nucl. Instrum. Methods Phys. Res. Sect. A 642 (2011) 87. [16] ASM, Properties and Selection: Nonferrous alloys and Special-Purpose Materials 2 ASM international, Metal Park, Ohio, 1990. [17] ASM, Alloy Phase Diagrams 3 ASM international, Metal Park, Ohio, 1992. [18] G. Petzow, Metallographisches, keramographisches, plastographiches Ätzen, Stuttgart, 1994. [19] M. Ferrari, L. Lutterotti, Method for the simultaneous determination of anisotropic residual stresses and texture by X-ray diffraction, J. Appl. Phys. 76 (11) (1994) 7246. [20] G. Williamson, W. Hall, X-ray line broadening from filed aluminum and wolfram, Acta Metall. 1 (1953) 22. [21] M.T. Hutchings, P.J. Withers, T.M. Holden, T. Lorentzen, Introduction to the Characterization of Residual Stress by Neutron Diffraction, CRC Press Taylor & Francis Group, Boca Raton, 2005. [22] H.H. Landolt, R Börnstein, Numerical data and functional relationships in science and technology, in: K.-H. Hellwege, A.M. Hellwege (Eds.), New Series, 11 1979. [23] M. Kamaya, A procedure for estimating Young's modulus of textured polycrystalline materials, Int. J. Solids Struct. 46 (2009) 2642. [24] E. Schmid, ‘Yield Point’ of Crystals, Critical Shear Stress Law, Proceedings Internat. Congr. App. Mech., Delft, p. 342. [25] H.-J. Bunge, Texture Analysis in Materials Science: Mathematical Methods, Butterworths, London, 1982.
Acknowledgment This work is based upon experiments performed at the STRESSSPEC instrument operated by HZG and TUM at the Heinz MaierLeibnitz Zentrum (MLZ), Garching, Germany. The authors would like to thank S. Rekier for the measurement of the mechanical properties. The authors would like to thank Dr. J. Paul and Dr. S. Gavras for their help in the English grammar.
155
Contents lists available at ScienceDirect
Materials Science & Engineering A journal homepage: www.elsevier.com/locate/msea
Crystallographic texture and lattice strain evolution during tensile load of swaged brass
MARK
⁎
Nowfal Al-Hamdanya, , Heinz-Günter Brokmeiera, Weimin Ganb a b
Institut für Werkstoffkunde und Werkstofftechnik, TU Clausthal, Agricolastr. 6, D-38678 Clausthal-Zellerfeld, Germany German Engineering Materials Science Center at MLZ, Helmholtz-Zentrum Geesthacht, D-85748 Garching, Germany
A R T I C L E I N F O
A B S T R A C T
Keywords: Crystallographic texture Lattice strain Swaging Brass Neutrons diffractions
Evolutions of texture and lattice strain of swaged brass samples were investigated by neutron diffraction at STRESS-SPEC under tensile deformation using a unique tension/compression rig. The two phased sample BS1 (61% α-brass and 39% β-brass) became 100% α-brass after 400 °C annealing (sample BS2). The starting texture of the as-received material BS1 was the typical < 111 > , < 200 > double fiber. This texture develops firstly by in-situ tension to a moderate strengthening. After annealing (BS2) the < 111 > fiber survives with surprisingly high strength and develops by in-situ tension a very strong < 111 > fiber of 39 mrd. Line broadening and lattice strain behaviour shows the development of the elastic strain and plastic strain.
1. Introduction Swaging is a highly used manufacturing process to improve materials properties, such as fatigue behaviour, strength and corrosion properties. Different types of swaging are in use, which is basically forging. Most of the swaging types are ideal for mass production. One in particular, rotary swaging (Fig. 1) can be used to reduce the cross section of rods and tubes and also to increase the strength [1]. An example of this type of swaging can be applied to the weight reduction of bike axles or high pressure tubes. The property change after swaging is based on grain refinement, strengthening of the crystallographic texture and generation of compressive stresses on the surface. It should be noted that most swaging processes are cold working processes, which reduces the costs. Rotary swaging in ideal conditions belongs to uniaxial deformation, same as wire and rod drawing. Uniaxial deformation is known for the high texture symmetry, which is defined by ideal fiber texture with a preferred fiber axis in transport direction and a rotational freedom around this fiber axis [2,3]. From Fig. 1, it can be deduced that due to the forging matrix which can consist of 2, 3, 4 or 6 segments a little variation from the ideal fiber texture can be observed. The texture symmetry is related to the process but the fiber type depends on the crystal symmetry and also on the stacking fault energy of the swaged material. FCC metals typically develop double fibers of < 111 > and < 100 > under uniaxial tension. The ratio depends on the stacking fault energy [2,3]. BCC metals show typically a < 110 > fiber [2,3]. HCP metals tend to have either the < 10.0 > or
⁎
the < 11.0 > fiber sometimes a double fiber < 10.0 > , < 11.0 > generates. A mixture of materials in case of precipitations or composites needs a co-deformation during swaging. In general it is known that a second phase influences texture development during deformation and recrystallization often in a decrease of the texture sharpness as well as the final mechanical properties [4]. Neutron diffraction allows the texture determination of the average texture over the whole cross section of rotary swaged samples, which is ideal for technical applications. This is possible because of the high penetration power of neutrons against standard X-rays. Moreover, the possibility to get complete pole figures makes the error of texture sharpness low, so that the expected variations in texture sharpness can be obtained in high quality [5–7]. Kalu et al. [8] have investigated the texture evolution in swaged Cu wire and Cu-Nb/Ti composite. Both materials developed < 111 > and < 100 > fiber texture. Cu-Nb/Ti composite had generally less texture sharpness than those of pure Cu. Hupalo et al. [9] have studied the evolution of texture and microstructure during cold swaging and recrystallization of Oligocrystalline INCOLOY® MA 956. They have used light optical microscopy, transmission electron microscopy, X-ray diffraction, electron back scattered diffraction (EBSD) and Vickers microhardness testing. A sharp < 110 > fiber texture was developed during plastic deformation after starting from a strong < 111 > fiber. They observed that recrystallization did not change the texture significantly. Gue et al. [10] have evaluated the texture during recrystallization of cold swaged Ti-Nb-Ta-Zr-O alloy (TNTZO). The texture measurements showed a pronounced < 110 > fiber along the
Corresponding author. E-mail address: [email protected] (N. Al-Hamdany).
https://doi.org/10.1016/j.msea.2017.11.047 Received 15 June 2017; Received in revised form 2 November 2017; Accepted 13 November 2017 Available online 14 November 2017 0921-5093/ © 2017 Elsevier B.V. All rights reserved.
Materials Science & Engineering A 711 (2018) 149–155
N. Al-Hamdany et al.
Fig. 3. Dimensions of the brass tensile sample.
and 500 °C before the ductility then recovered at higher temperatures. STRESS-SPEC is equipped with a tensile/compression machine having a rotated rig (χ tilt and ϕ rotation) with a maximum load of 50 kN. Different sample geometries are possible to be measured by this machine. An additional light furnace can be installed around the loaded sample with temperature up to 1000 °C [14,15]. Commercial Cu-36%Zn brass (Fig. 2) was chosen firstly due to expected activation of deformation twins during deformation and secondly because of the mixture of α- and β-brass. This alloy has excellent cold workability. Typical applications of Cu-36%Zn are radiator cores, tanks, lamp fixtures, fasteners, locks, hinges, plumbing accessories, pins, rivets [16]. The aim of this experiment was to investigate the influence of the existing preferred orientation on texture development of α brass during tensile load for α+β brass and α brass. Additionally, the evolution of line broadening (defect evolution) and line position (lattice strain) of both samples BS1 and BS2 were investigated during tensile loading.
Fig. 1. Scheme of the swaging process with 4 segments.
deformation axis in the cold swaged TNTZO alloy which is the typical fiber texture for bcc materials after drawing or swaging. The < 110 > fiber texture was gradually replaced by random orientations by increasing the annealing time. Gan et al. [11,12] have investigated the bulk and local texture evolution of pure Mg processed by rotary swaging. They started from diameter of 10 mm going down to 4.5 mm which is corresponding to a deformation degree of 1.58. A main {00.2} basal fiber texture in rotary swaging processed Mg was observed. Basal planes tend to be more homogeneous with the increase of the process pass. Texture was relatively inhomogeneous from the surface to the center at rods which indicate a non-uniform deformation. Bache et al. [13] have characterized an advanced Nickel based superalloy RR1000 post cold work by swaging. Cold work of 30% was subjected to cylindrical bars of 27 mm diameter. Beside the reduction in grain size, a relatively strong < 111 > fiber texture parallel to the longitudinal bar axis has been confirmed by EBSD. The effect of cold work on bulk mechanical properties was observed from the increase in the ultimate tensile strength (UTS) and yield tensile strength (YTS), but also resulted in reduced ductility. The ductility reduced between room temperature
2. Materials and experimental procedure Two samples were produced from commercial Cu-36%Zn brass. One sample (BS1) was swaged up to a deformation degree of φ = 3 having a reduction in cross section from 32 mm in diameter to 7 mm. Sample 2 (BS2) was additionally annealed for 2 h at 400 °C after swaging. During this annealing β-brass was solved and BS2 consists only of α-brass. Phase analysis was carried out with hard X-rays at the high energy beamline HEMS@Petra III. The microstructure was examined for both Fig. 2. Phase diagram of Cu-Zn alloy [17].
150
Materials Science & Engineering A 711 (2018) 149–155
N. Al-Hamdany et al.
Fig. 4. BS1 tensile sample (a) installed in the loading machine (b) installed on the Eulerian cradle at STRESS-SPEC.
STRESS-SPEC (Fig. 4a). Both samples were tensioned until failure, sample BS1 up to 10.5 kN and sample BS2 up to 5.7 kN). The loading rate was 10 N/s. Two different types of experiments were performed, in-situ lattice strain investigations and ex-situ texture investigations. For lattice strain investigations, diffraction patterns were collected during tensile load. The loading direction was parallel to the transport direction during swaging. The neutron measurements were carried out with a wavelength of 1.5 Å obtained from Germanium (Ge (311)) monochromator. An area detector of 300 mm × 300 mm was used with a sample to detector distance of 1065 mm. The detector position of 2θ = 45° allows the simultaneous measurement of the reflections (111) and (200). The beam was guided by primary slit of 5 mm in diameter and on the detector side by at FWHM = 5 mm radial collimator to get the average information over the whole sample thickness. The exposure time was 5 min. Under load control diffraction patterns were collected for BS1 until 8 kN in steps of 1 kN and in steps of 0.5 kN from 8 kN until ultimate load. For BS2, diffraction patterns were collected until 2 kN in steps of 0.5 kN and in steps of 0.25 kN from 2 kN until 5.5 kN. For pole figure measurements, the samples were screwed on a plate and installed on the Eulerian cradle (Fig. 4b). The pole figure measurements were carried out using neutrons wavelengths of 1.66 Å and 1.1 Å obtained from pyrolytic graphite (PG) monochromator using (0004) and (0006) planes. Dual wavelength was chosen to save counting time by using only one detector position. An area detector of 300 mm × 300 mm was used with a sample to detector distance of 1065 mm. Three pole figures were measured with the detector position of 2θ = 50° for (111), (200) and (220). According to the sample to the detector distance and the detector position, six sample tilts in χ were needed. For each χ-position a continuous scanning around φ was performed. An exposure time of 10 s per 2.5° continuous φ rotation gives sufficient intensities. 3. Results Quantitative phase composition was calculated using MAUD [19]. The sample BS1 consists of 61% α-brass and 39% β-brass (Fig. 5), while sample BS2 after annealing had only α-brass. Fig. 6 shows the microstructure of BS1and BS2 before loading. The microstructure of BS1 consists of deformed α-grains in light colour and deformed β-grains in dark colour. BS2 has a recrystallized α-brass microstructure. The stress strain curves for BS1 and BS2 are presented in Fig. 7 giving the experimental values for load in kN and for tension in mm. In Table 1 the mechanical properties of both samples are summarized. Red dots describe the positions for in-situ lattice strain measurements.
Fig. 5. Diffraction pattern obtained from hard X-rays for (a) BS1 and (b) BS2.
BS1 and BS2 before and after loading. The microstructure samples were etched after grinding and polishing with a solution of 100 ml distilled water 20 ml hydrochloric acid and 5 g iron (III) chloride [18] Two standard tensile samples were machined following DIN 50515 having d0 = 4.03 mm, L0 = 20 mm, d1 = M6, Lc = 24 mm and Lt = 40 mm (Fig. 3). The tensile tests have been carried out using the tensile machine at
151
Materials Science & Engineering A 711 (2018) 149–155
N. Al-Hamdany et al.
Fig. 6. Microstructure of BS1 (a) and (BS2) (b) before and after loading.
12000
a
Loading curve Single shot
6000
Load (N)
Load (N)
5000
8000
4000 2000 0 0.0
b
6000
10000
4000 3000
Loading curve Single shot
2000 1000
0.2
0.4
0.6
0.8
0
1.0
Deformation (mm)
0
1
2
3
4
5
6
7
Deformation (mm)
Fig. 7. Stress strain curve and single shots for BS1 (a) and BS2 (b).
defects produced by the swaging process. The reduction in the number of defects after heat treatment can be indicated from the reduction of the full width of half maximum (FWHM) at 0 kN (Fig. 8a). In principal one can estimate the defect density from the slope of the WilliamsonHall Plot [20]. For 0 kN it can be seen that the slope of BS1 is much higher than of BS2, (Fig. 8b). Moreover, BS1 has two phases, the FCC αbrass and the BCC β-phase. The β phase is the harder phase which also participates to improve the mechanical properties. Fig. 8a shows the evolution of the line broadening during tensile test. In BS1 one can see a small increase in the line broadening of the (111) planes starting after 8 kN while for (200) planes nearly no change is visible. During elastic strain no increase of the line broadening was
Table 1 Mechanical properties of BS1 and BS2. Sample
Young's modulus (GPa)
Yield stress (MPa/kN)
Ultimate tensile strength (MPa/ kN)
Elongation (%)
BS1 BS2
123 104
770/9.6 233/2.9
820/10.5 455/5.7
2.4 35
The yield stress in BS1 is higher than in BS2 as well as the ultimate tensile stress. BS2 alternatively has a higher percentage of elongation than BS1 as a result of the heat treatment of BS2 releasing most of the 152
Materials Science & Engineering A 711 (2018) 149–155
N. Al-Hamdany et al.
Fig. 8. Line broadening and lattice strain; (a) FWHM against load, (b) FWHM against 2θ obtained from hard X-rays diffraction (Williamson-Hall Plot) (c) Lattice strain against applied load.
a1
(111)
a2
Pmax= 22 mrd
(200)
b1
Pmax= 4.4 mrd
(111)
Pmax= 29 mrd
b2
(200)
Pmax= 6.5 mrd
Fig. 9. BS1 (111) and (200) pole figures of α-brass; a) before load (only swaged), b) after ultimate load.
Fig. 8c shows the evolution of the lattice strain versus the load. The difference in the slops in the elastic region before the yield point of the strain stress curve is based on the elastic anisotropy. Different crystal lattice planes have different elastic properties [21]. For BS1 the experimental value for α-brass, calculated from Fig. 8c, for E111 is 176.43 GPa and for E200 is 135.86 GPa. For BS2 E111 is 196.81 GPa and E200 is 114.63 GPa. The experimental elastic properties are different to the theoretical values calculated for alpha brass (30% Zn) after the Kröner model. The theoretical elastic properties for alpha brass (30% Zn) are (E111 is 147.75 GPa and E200 is 88.93 GPa) [22]. The reasons for these differences are the alloying elements, the crystallographic texture
expected while during plastic strain new defects are generated with hardening. One also has to take into account that the swaging process up to deformation degree of φ = 3 causes a high number of defects. The (111) oriented grains start deforming earlier than (200) oriented grains. Line broadening in sample BS2 shows a much different behaviour. For (111) planes a moderate increase of the line broadening start at 2.5 kN while for (200) oriented grains a rapid increase is seen. This agrees with the Williamson-Hall Plot in Fig. 8b, showing a strong increase of the defect density in sample BS2. Continuous hardening behaviour during in-situ loading takes place, which also correlates with the stress strain curve. 153
Materials Science & Engineering A 711 (2018) 149–155
N. Al-Hamdany et al.
a1
(111)
Pmax= 21 mrd
a2
(200)
b1
Pmax= 4.7 mrd
(111)
Pmax= 39 mrd
b2
(200)
Pmax= 6.1 mrd
Fig. 10. BS2 (111) and (200) pole figures of α-brass; a) before load (swaged + annealed), b) after ultimate load. Fig. 11. ODF of α-brass for BS1; (a) before load (only swaged), (b) after ultimate load.
direction (in uniaxial tension) moves towards alignment with the direction of the applied stress [2,3]. To analyse the fibers separately the orientation distribution function (ODF) has been calculated. The ODFs for all samples have been calculated from three pole figure using the iterative series expansion method with Lmax = 35 as degree of series expansion. Due to the stronger texture BS2 after load was calculated with Lmax = 55 [25]. Figs. 11 and 12 show the ODF-section φ2 = 0° which includes both the < 100 > and the < 111 > fiber. The < 100 > fiber axis is at φ1 = 90°, Φ = 0°, φ2 = 0°; the fiber itself runs at φ1 = 0° and φ2 = 0° from Φ = 0° to 90°. For the < 111 > fiber only the cross section of the fiber can be seen at φ2 = 0°, φ1 = 55°, Φ = 45°. In general, it is known that the relative concentrations of the two fiber components depend on the stacking fault energy. Other influences on the texture evolution, together with the stacking fault energy, are the process temperature, the deformation rate, the initial texture, the grain size, and the purity of the material [2].
and the grain morphology [23]. The plastic behaviour of (111) and (200) oriented grains in both samples BS1 and BS2 (Fig. 8c) is different to the one from the macroscopic stress strain curve (Fig. 7). The (200) oriented grains starts to yield earlier than (111) oriented grains in both BS1 and BS2 (Fig. 8c). The reason is different critical resolved shear stress is required for the initiation of glide on a substantial scale [24] The pole figures of α-brass are shown in Fig. 9 (BS1 before loading and after ultimate load and in Fig. 10 (BS2 before loading and after ultimate load). The BS1 swaging texture shows the typical strong < 111 > and weaker < 100 > double fibers texture which is strengthened during tensile load. The (111) pole figure sharpness increases from 22 mrd to 29 mrd (mrd – multiples of random distribution). BS2 has only a < 111 > single fiber which survives after the heat treatment of BS1 at 400 °C for 2 h. With heat treatment the β phase dissolved and α phase recrystallized to a preferred < 111 > orientation. BS2 has similar texture sharpness before load than BS1, which is unexpected that 400 °C annealing has no influence on the texture sharpness of the < 111 > fiber. But tensile load of only 5.7 kN (455 MPa) leads to an unexpected increase of (111) pole figure sharpness to 39 mrd. During the slip process, the crystal lattice also rotates so that the active slip
4. Conclusions The crystallographic texture and lattice strain evolution of swaged brass during tensile load using neutrons diffractions has been 154
Materials Science & Engineering A 711 (2018) 149–155
N. Al-Hamdany et al.
Fig. 12. ODF of α-brass for BS2; (a) before load (swaged + annealed), (b) after ultimate load.
References
investigated. The diffraction information was obtained from the complete cross section of standard tensile test sample after DIN-norm. The sample BS1 consists of 61% α-brass and 39% β-brass while BS2 (heat treated sample) has only α-brass. The swaged sample BS1 showed higher strength (820 MPa) and lower elongation (2.4%) compared to BS2 (455 MPa and 35%). This was due to the removal of many defects produced by the swaging process as a result of heat treatment. Moreover, the BS1sample had two phases with the harder β-phase leading to increased strength. Line broadening studies have shown the formation or removal of defects within the material. During tensile testing, the BS1 sample had a minor increase in line broadening due to already containing a high number of defects. The BS2 sample showed a rapid increase in line broadening especially for the (200) plane. The difference in the elastic properties of the α-brass (111) and (200) planes in the BS1 and BS2 specimens was due to differences in crystallographic texture and grain morphology. Grains oriented with < 200 > parallel to the loading directions started to yield earlier than grains oriented with < 111 > parallel to loading directions in both BS1 and BS2. This was due to different critical resolved shear stresses for the different planes after which glide on a substantial scale was initiated. Both the BS1 and BS2 samples showed the expected increase in texture sharpness of the < 111 > fiber during tensile test. An unexpectedly strong increase in texture sharpness of the BS2 sample to approximately 39 mrd was measured. During the slip process, the crystal lattice also rotates so that the active slip direction (in uniaxial tension) moves towards alignment with the direction of the applied stress. More investigation will be made to understand the strong increase in the texture sharpness of the BS2 sample.
[1] ASM, Forming and Forging 14 ASM international, Metal Park, Ohio, 1993. [2] G. Wassermann, J. Grewen, Texturen metallischer Werkstoffe, Springer Verlag, Berlin, 1962. [3] H. Hu, Texture of metals, Texture 1 (1974) 233. [4] H.-G. Brokmeier, Experimental textures of Al-Pb, Al-Cu, Fe-Cu metal-metal composites, Mater. Sci. Eng. A 175 (1994) 131. [5] H.-J. Bunge, Z. Met. 76 (1985) 457. [6] H.-G. Brokmeier, Neutron diffraction texture analysis of multi-phase systems, Textures Microstruct. 10 (1989) 325. [7] H.-G. Brokmeier, Texture analysis by neutron diffraction, Mater. Sci. Forum 157–162 (1994) 59. [8] P.N. Kalu, L. brandao, F. Ortiy, O. Egungwu, F. Ige, On the texture evolution in swaged Cu-based wires, Scr. Mater. 38 (12) (1998) 1755. [9] M.F. Hupalo, A.F. Padiha, H.R.Z. Sandim, texture and microstructure evolution during cold swaging and recrystallization of oligocrystalline INCOLAY® MA 956, Solid State Phenom. 105 (2005) 315. [10] W.Y. Gue, H. Xing, J. Sun, X.L. Li, J.S. Wu, R. Chen, Evolution of microstructure and texture during recrystallization of cold-swaged Ti-Nb-Ta-Zr-O alloy, Metall. Mater. Trans. A 39A (2008) 672. [11] W.M. Gan, Y.D. Huang, R. Wang, Z.Y. Zhong, N. Hort, K.U. Kainer, N. Schell, H.G. Brokmeier, A. Schreyer, Bulk and local texture of pure magnisium processed by rotary swaging, J. Magnes. Alloy. 1 (2013) 341. [12] W.M. Gan, Y.D. Huang, R. Wang, A. Srinivasan, H.-G. Brokmeier, N. Schell, K.U. Kainer, N. Hort, Microstructures and mechanical properties of pure Mg processed by rotary swaging, Mater. Des. (2014) 83. [13] M.R. Bache, J. O׳Hanlon, P.J. Withers, D.J. Child, M.C. Hardy, Characterization of an advanced nickel based superalloy post cold work by swaging, Metals (2016), http://dx. doi.org/10.3390/met6030054. [14] M. Hofmann, R. Schneider, G.A. Seidle, J. Rebelo-Kornmeier, R.C. Wimpory, U. Garbe, H.G. Brokmeier, The new materials science diffractometer STRESS-SPEC at FRM-II, Physica B 385–386 (2006) 1035. [15] H.-G. Brokmeier, W.M. Gan, C. Randau, M. Völler, J. Rebelo-Kornmeier, M. Hofmann, Texture analysis at neutron diffractometer STRESS-SPEC, Nucl. Instrum. Methods Phys. Res. Sect. A 642 (2011) 87. [16] ASM, Properties and Selection: Nonferrous alloys and Special-Purpose Materials 2 ASM international, Metal Park, Ohio, 1990. [17] ASM, Alloy Phase Diagrams 3 ASM international, Metal Park, Ohio, 1992. [18] G. Petzow, Metallographisches, keramographisches, plastographiches Ätzen, Stuttgart, 1994. [19] M. Ferrari, L. Lutterotti, Method for the simultaneous determination of anisotropic residual stresses and texture by X-ray diffraction, J. Appl. Phys. 76 (11) (1994) 7246. [20] G. Williamson, W. Hall, X-ray line broadening from filed aluminum and wolfram, Acta Metall. 1 (1953) 22. [21] M.T. Hutchings, P.J. Withers, T.M. Holden, T. Lorentzen, Introduction to the Characterization of Residual Stress by Neutron Diffraction, CRC Press Taylor & Francis Group, Boca Raton, 2005. [22] H.H. Landolt, R Börnstein, Numerical data and functional relationships in science and technology, in: K.-H. Hellwege, A.M. Hellwege (Eds.), New Series, 11 1979. [23] M. Kamaya, A procedure for estimating Young's modulus of textured polycrystalline materials, Int. J. Solids Struct. 46 (2009) 2642. [24] E. Schmid, ‘Yield Point’ of Crystals, Critical Shear Stress Law, Proceedings Internat. Congr. App. Mech., Delft, p. 342. [25] H.-J. Bunge, Texture Analysis in Materials Science: Mathematical Methods, Butterworths, London, 1982.
Acknowledgment This work is based upon experiments performed at the STRESSSPEC instrument operated by HZG and TUM at the Heinz MaierLeibnitz Zentrum (MLZ), Garching, Germany. The authors would like to thank S. Rekier for the measurement of the mechanical properties. The authors would like to thank Dr. J. Paul and Dr. S. Gavras for their help in the English grammar.
155