ÓASM International 1059-9495/$19.00

JMEPEG DOI: 10.1007/s11665-016-2354-3

A Novel Cu-10Zn-1.5Ni-0.34Si Alloy with Excellent Mechanical Property Through Precipitation Hardening Wei Chen, Mingpu Wang, Zhou Li, Qiyi Dong, Yanlin Jia, Zhu Xiao, Rui Zhang, and Hongchun Yu (Submitted April 7, 2016; in revised form September 18, 2016) A novel Cu-10Zn-1.5Ni-0.34Si alloy was designed to replace the expensive tin-phosphor bronze in this paper. The alloy had better comprehensive mechanical properties than traditional C5191 alloy. The aged Cu-10Zn-1.5Ni-0.34Si alloy had a hardness of 220 HV, electrical conductivity of 28.5% IACS, tensile strength of 650 MPa, yield strength of 575 MPa and elongation of 13%. Ni2Si precipitates formed during aging, and the crystal orientation relationship between matrix and precipitates was: (001)a//(001)d, and [110]a//[100]d. Ductile fracture surface with deep cavities was found in the alloy. Solid solution strengthening, deformation strengthening and precipitation strengthening were found to be core strengthening mechanisms in the alloy. Keywords

copper, heat treatment, microstructure, property

1. Introduction Copper-based elastic alloys have been widely used for conductive elastic components in instruments and meters, owing to their high electrical conductivity, thermal conductivity and good mechanical properties (Ref 1-4). They are classified into aging precipitation strengthening copper alloy (APSCA) and solid solution strengthening copper alloy (SSSCA) according to the different strengthening mechanisms. Cu-Be alloy is a widely used APSCA because of their high strength and elasticity (Ref 5). However, the addition of Be in these alloys is a significant, severe hazard to the environment and human health. Tin-phosphor bronze is a typical SSSCA, which is most widely applied as ductile materials in civil areas. But it is limited by the low cold working performance as high strain-hardening (Ref 6). Further, Sn-P bronze is expensive due to the shortage of Sn, which limited their application. With the rapid development of the conductive elastic alloys, copper-based elastic alloys have been requested with high performance, low cost, high reliability and no pollution. Cu-Zn alloy is economically more efficient than tin-phosphor bronze (i.e., C5191 alloy) not only in terms of composition but also from the view point of a working process and further of a shorter process. It is believed that Wei Chen, Mingpu Wang, Qiyi Dong, Zhu Xiao, and Rui Zhang, School of Materials Science and Engineering, Central South University, Changsha 410083, People’s Republic of China; Zhou Li, School of Materials Science and Engineering, Central South University, Changsha 410083, People’s Republic of China; and State Key Laboratory of Powder Metallurgy, Changsha 410083, PeopleÕs Republic of China; Yanlin Jia, School of Metallurgical Science and Engineering, Central South University, Changsha 410083, PeopleÕs Republic of China; Hongchun Yu, School of Physics and Microelectronics, Hunan University, Changsha 410082 Hunan, PeopleÕs Republic of China. Contact e-mails: [email protected] and [email protected].

Journal of Materials Engineering and Performance

Cu-Zn alloy has excellent mechanical properties due to the solid solution strengthening effect. However, its conductivity is relatively low (Ref 7). Effects of Ni and Si on the microstructure and properties of copper matrix have been reported (Ref 8-11). Results show that the comprehensive properties of the copper alloy can be effectively improved by the precipitation hardening of Ni2Si. Thus, by adding the element Ni and Si, enhanced strength in Cu-Zn alloy could be obtained. In this study, a novel Cu-10Zn-1.5Ni-0.34Si alloy was designed and investigated. This developed copper-based elastic alloy has low cost and good comprehensive performance. Thus, it has great potential to replace the expensive tin-phosphor bronze.

2. Materials and Experimental Procedure Cu-10Zn and Cu-10Zn-1.5Ni-0.34Si ingots were prepared using an intermediate frequency furnace. After surface defects were removed, all the casting ingots were homogenized at 900 °C for 2 h, then hot-rolled from a thickness of 30 to 5 mm at 880 °C, followed by quenching into cold water. The hotrolled plate was then cold-rolled with 80% reduction at room temperature, followed by isothermal aging at different temperatures in a salt-bath furnace. Vickers hardness (HV) was carried out on a HV-5 type micro-hardness tester, with 2 kg load and 10 s loading time according to ASTM E384-11 Standard (Ref 12). Electrical conductivity was measured at 20 °C using an Eddy Conductivity Instrument, taking the average of 7 values to minimize errors. Tensile tests were performed at 20 °C using an Instron 8019 tester machine with a constant strain rate of 103 s1. Specimens were etched in a solution containing FeCl3, hydrochloric acid and alcohol, and optical microstructures were observed on a Leica optical microscope. Transmission electron microscopy (TEM) samples were prepared by double jet electro-polishing techniques using a 20% nitric acid in methanol solution at about 30 °C. TEM observations were carried out on a TECNAI G2 F20 transmission electron microscope with operation voltage of 200 kV.

3. Results and Discussion 3.1 Thermo-Mechanical Treatment Figure 1 shows the microstructures of Cu-Zn and Cu-Zn-NiSi alloys. The as-cast Cu-Zn alloy had a microstructure of pure a-Cu phase with no obvious dendrite. Typical dendrite structure appeared in as-cast Cu-Zn-Ni-Si alloy (Fig. 1b), with the average grain size about 180 lm. It is much smaller than that of as-cast Cu-Zn alloy, which have an average grain size of more than 500 lm. After homogenizing treatment at 900 °C for 2 h, dendrites disappeared and grains slightly coarsened (about 450 lm). This indicated that the addition of Ni and Si element refined the grains of Cu-Zn alloy effectively. The Cu-Zn-Ni-Si ingot was hot-rolled at 880 °C with total deformation of 80%. Figure 2 shows the microstructure of the hot-rolled Cu-Zn-Ni-Si alloy. Complete recrystallization occurred in the designed alloy, and the mean recrystallization grain size was about 30 lm.

3.2 Aging Treatment Figure 3 shows the variations of hardness and electrical conductivity of the homogenized and cold-rolled Cu-Zn alloys after annealing at 200, 250, 300 and 350 °C, respectively. The

Fig. 1 alloy

hardness of Cu-Zn alloy increased as annealing temperature below 250 °C due to annealing hardening effect. Then, the hardness decreased rapidly with the annealing duration when annealing above 300 °C, which could be attributed to the nucleation of recrystallization. This indicated that the starting temperature of recrystallization is about 300 °C. The corresponding electrical conductivity of the alloy increased with

Rolling direcon

100 m

Fig. 2 Optical microstructure of designed Cu-Zn-Ni-Si alloy after hot-rolled by 80% at 880 °C

Optical microstructure of designed alloys. (a) as-cast Cu-Zn alloy; (b) as-cast Cu-Zn-Ni-Si alloy; (c) homogenizing state Cu-Zn-Ni-Si

Journal of Materials Engineering and Performance

Fig. 3

Variations of the hardness and electrical conductivity of Cu-Zn alloys during annealing. (a) hardness; (b) electrical conductivity

Fig. 4

Variations of the hardness and electrical conductivity of Cu-Zn-Ni-Si alloys during annealing. (a) hardness; (b) electrical conductivity

annealing time at all applied annealing temperature, but the increment is less lower than 5% IACS. According to the previous study (Ref 13-15), the major enhancement of electrical conductivity during annealing should not be explained by short range ordering. It was concluded that the segregation of solute atoms at dislocations is the only consistent interpretation for the change of conductivity. And the decrease in lattice defects (i.e., dislocations) yielded the increase in conductivity. The hardness of Cu-Zn alloy had big changes after annealing at 300 and 350 °C, but the corresponding conductivity had almost no change. As a type of solid solution strengthening copper alloy, Cu-Zn alloy has no strengthening phases precipitated during annealing treatment,and logically no alloying elements dissolved which can change conductivity apparently. The slight variation of conductivity (about 3-4% IACS) of Cu-Zn alloy during annealing treatment at 300 and 350 °C is mainly because of the segregation of solute atoms, whereas the significant reduction of hardness after annealing at temperatures above 300 °C is attributed to recovery and recrystallization.

Journal of Materials Engineering and Performance

Figure 4 shows the variations of hardness and electrical conductivity of the Cu-Zn-Ni-Si aged at 350, 400, 450 and 500 °C, respectively. Obviously, lower hardness peak value and shorter peak-aging time appeared on higher aging temperatures (Fig. 4a). Hardness peaks of samples occurred at 120, 30, 5 and 1 min after aging at 350, 400, 450 and 500 °C, respectively. The peak value and the corresponding electrical conductivity were 222, 220, 215, 205 HV and 25.9, 28.5, 26.2, 24.5% IACS, respectively. The conductivity increased with the aging duration at all applied aging temperatures, and the conductivity arising was more obvious at higher temperatures. However, when the temperature exceeded 400 °C, the stationary conductivity decreased with the temperature elevated, indicating an optimal aging temperature of 400 °C for this alloy. The dominant strengthening mechanism of this alloy transformed from solid solution hardening to precipitation hardening by adding Ni and Si. The optimal treatment of the designed alloy was as follow: solution treatment at 900 °C for 2 h, hot rolling followed by quick water-quenching, cold rolling by 80% and

then aging treatment at 400 °C for 30 min. The tensile stressstrain curve of the designed alloy is shown in Fig. 5. The designed alloy showed excellent mechanical property (elongation is 13%, tensile strength is 650 MPa and yield strength is 575 MPa) and good electrical conductivity (28.5% IACS).

Fig. 5

Tensile stress-strain curve of designed Cu-Zn-Ni-Si alloy

3.3 Precipitation Figure 6 shows the TEM micrographs and selected-area diffraction patterns (SADP) of cold-rolled Cu-Zn-Ni-Si alloy aged at 400 °C for different times. Nanoscale strain-field contrast dispersed after aging for 1 min, while distributed nano-scale precipitates appeared when the alloy was aged for 5 min. Zero-contrast line can be seen clearly (marked by arrow in Fig. 6b), indicating that the precipitates were coherent with the matrix. After being aged for 16 h, lots of rod-like and elliptical precipitates (marked by arrows) were found in the alloy (Fig. 6c, d). The rod-like precipitates had two directions perpendicular to each other, and the length was approximately 50 nm. The elliptical precipitates seem to be embedded in the matrix. The corresponding SADP with electron beam parallel to [001]a is shown in Fig. 6(e). The diffraction spots from d-Ni2Si precipitates were observed, and the indexation results show that the crystal orientation relationship between matrix and d-Ni2Si precipitates is (001)a// (001)d, and [110]a//[100]d. Teplitskiy et al. (Ref 16) reported that the disk-shaped or elliptical precipitates were generated at {110}a in Cu-Ni-Si alloy during aging process. There were six equivalent [110]a directions, and only two of them were parallel to the electrical beam due to the face-centered cubic structure. Thus, rod-like precipitates in two mutually perpendicular

Fig. 6 TEM micrographs and SADPs of designed CuZnNiSi alloys that underwent solution treatment and cold rolling by 80% and followed by aging treatment at 400 °C for different durations: (a) 5 min, bright-field micrograph; (b) 30 min, bright-field micrograph; (c) 16 h, bright-field micrograph; (d) 16 h, dark-field micrograph; (e) 16h, beam direction of SADP along [001]a

Journal of Materials Engineering and Performance

directions ([110]a and [1-10]a) are observed in Fig. 6(d). The disk-shaped precipitates in other four {110}a were presented as an elliptical shape on the image plane.

3.4 Fracture The fracture morphologies of peak-aged specimen are shown in Fig. 7. Tear ridges, cleavage planes and cavities were found in the fracture surface, suggesting a typical ductile fracture. Hard precipitation particles, which have a bad impact on elongation that might be broken away from the matrix grains (Ref 17), cannot be observed in the fracture. The cleavage planes of grain in the specimen would not be located on the fracture surface when tested under embrittling conditions, such as low temperature and high stress. As a result, ductile fracture with tear ridges and deep cavities appeared around the periphery of the tensile-ruptured surfaces in the Cu-Zn-Ni-Si alloy.

3.5 Strengthening Mechanism As the designed alloy was aged at 400 °C, there were two major strengthening mechanisms: solid solution hardening and precipitation hardening. The solid solution hardening was attributed to the lattice distortion and strong covalent bond between Cu and Zn (Ref 18). Precipitation hardening was the main strengthening mechanism responsible for the good strength of the Cu-Zn-Ni-Si alloy. Increase in strength by the nanoscale precipitations can be described by the Orowan mechanism, and increment of DrOrowan can be expressed as Ref 19, 20

k ¼ dp

pffiffiffiffiffiffiffiffiffiffiffiffiffi 3p=2fv =2

ðEq 2Þ

where fv is the volume fraction of particles. For simplification, we assumed that all the Si atoms were precipitated in the form of d-Ni2Si precipitates. Thus, volume fraction of dNi2Si precipitates can be calculated as about 0.027, and increment of DrOrowan depending on dp can be expressed as follows: DrOrowan ¼ ð1034 ln dp þ 1413Þ=dp

ðEq 3Þ

Figure 8 shows the relationship between DrOrowan and dp, according to Eq 3. When the designed alloy was aged at 400 °C for 30 min, dp was about 7 nm (Fig. 6b), and the DrOrowan was about 489 MPa. When the dp was about 50 nm in Fig. 6(d) (for the specimen aged at 400 °C for 16 h), the DrOrowan was 109 MPa. This indicated that the mean particle size had a significant effect on the increment of DrOrowan when it was <50 nm. The precipitates were small than 7 nm in the alloy aged for 30 min, while coarsened to be about 50 nm with aging time extended to 16 h. The decreasing of micro-hardness of

DrOrowan ¼ 0:81 MGb  lnðdp =bÞ=2pð1  mÞ1=2 ðk  dp Þ ðEq 1Þ where M is Taylor factor (for Cu is 3.1), G is the shear modulus of matrix (for Cu is 45.5 Gpa), dp is the mean diameter of d-Ni2Si particles (here, we measured the value from mass TEM micrographs of samples as different treatment), b is the Burgers vector (0.255 nm for Cu), and m is the PoissonÕs ratio (0.34 for Cu). k is the mean particle plane square lattice spacing (i.e., apparent particles spacing), which can be given by Ref 21

Fig. 7

Fracture morphologies of designed Cu-Zn-Ni-Si alloy

Journal of Materials Engineering and Performance

Fig. 8 Schematic of Orowan strengthening attributed to average size of precipitates particles

Fig. 9 Comparisons of mechanical properties and electrical conductivity of C5191 and designed Cu-Zn-Ni-Si alloy. (a) Tensile strength; (b) Electrical conductivity and Elongation

specimens with the aging time was due to the coarsening of precipitates.

3.6 Properties Comparison Figure 9 shows the comparison of mechanical properties and electrical conductivity between commercial phosphor bronze C5191 alloy and the designed Cu-Zn-Ni-Si alloys. C5191 alloy has good mechanical properties, but its electrical conductivity is relatively low (less than 14% IACS). Cu-Zn-NiSi alloys have a good tensile strength and elongation which is similar to C5191. However, the electrical conductivity (28.5% IACS) is over two times than that of C5191. During aging treatment, precipitation of Ni2Si phase strengthened the alloy and also enhanced the conductivity through the dissolution of the alloying elements. It could be expected that Cu-Zn-Ni-Si alloy would have a bright future in replacing phosphor bronze as for its better property and lower cost.

4. Conclusions 1. The Ni2Si precipitates formed in the Cu-Zn-Ni-Si alloy during aging process and the crystal orientation relationship between the matrix and precipitates was: (001)a// (001)d, [110]a//[100]d. 2. After homogenization treatment at 900 °C for 2 h, hot rolling by 80%, cold rolling by 80%, and aging treatment at 400 °C for 30 min, Cu-Zn-Ni-Si alloy showed excellent properties (tensile strength is 650 MPa, elongation is 13%, and electrical conductivity is 28.5% IACS). 3. The core strengthening mechanisms of Cu-Zn-Ni-Si alloy were precipitation strengthening and solute solution strengthening. 4. The designed Cu-Zn-Ni-Si alloy showed a quasi-cleavage fracture with deep dimples.

Acknowledgments The authors are pleased to acknowledge the financial supply supported by the Project supported by the National Natural Science Foundation of China (51271203) and the Project of Innovationdriven Plan in Central South University.

References 1. Y. Zhang, B.H. Tian, A.A. Volinsky, H.L. Sun, Z.P. Liu, X.H. Chen, and Y. Liu, Microstructure and PrecipitateÕs Characterization of the CuNi-Si-P Alloy, J. Mater. Eng. Perform., 2016, 25, p 1336–1341 2. Y. Zhang, P. Liu, B.H. Tian, Y. Liu, R.Q. Li, and Q.Q. Xu, Hot Deformation Behavior and Processing Map of Cu-Ni-Si-P Alloy, Trans. Nonferrous Met. Soc. China, 2013, 23, p 2341–2347 3. Y. Zhang, A.A. Volinsky, Q.Q. Xu, Z. Chai, B.H. Tian, P. Liu, and H.T. Tran, Deformation Behavior and Microstructure Evolution of the Cu2Ni-0.5Si-0.15Ag Alloy During Hot Compression, Metall. Mater. Trans. A, 2015, 46, p 5871–5876 4. Y. Jia, M. Wang, C. Chen, Q. Dong, S. Wang, and Z. Li, Orientation and Diffraction Patterns of d-Ni2Si Precipitates in Cu-Ni-Si Alloy, J. Alloys Compd., 2013, 557, p 147–151 5. H. Tsubakino, R. Nozato, and A. Yamamoto, Precipitation Sequence for Simultaneous Continuous and Discontinuous Modes in Cu-Be Binary Alloys, Mater. Sci. Technol., 1993, 9, p 288–294 6. M. Miyafuji, Shirnonoseki-shi, Copper Alloy with Excellent Stress Relaxation Resistance Property and Production Method Therefor, United States patent US 2003/0047259 A1, 2003 7. R.M. Hussein and O.I. Abd, Influence of Al and Ti Additions on Microstructure and Mechanical Properties of Leaded Brass Alloys, Indian J. Mater. Sci., 2014, 2014, p 1–5 8. Q. Lei, Z. Li, C. Dai, J. Wang, X. Chen, J.M. Xie, W.W. Yang, and D.L. Chen, Effect of Aluminum on Microstructure and Property of CuNi-Si Alloys, Mater. Sci. Eng. A, 2013, 572, p 65–74 9. H. Fujiwara, T. Sato, and A. Kamio, Effect of Alloy Composition on Precipitation Behavior in Cu-Ni-Si Alloys, J. Jpn. Inst. Met., 1998, 62, p 301–309 10. D.M. Zhao, Q.M. Dong, P. Liu, B.X. Kang, J.L. Huang, and Z.H. Jin, Structure and Strength of the Age Hardened Cu-Ni-Si Alloy, Mater. Chem. Phys., 2003, 79, p 81–86 11. F. Huang, J. Ma, H. Ning, Y. Cao, and Z. Geng, Precipitation in Cu-NiSi-Zn Alloy for Lead Frame, Mater. Lett., 2003, 57, p 2135–2139 12. ASTM Standard, E384-11, Standard Test Method for Knoop and Vickers Hardness of Materials, ASTM International, West Conshohocken, 2011 13. S. Nestorovic´, D. Markovi´c, and I. Markovi´c, Influence of Thermal Cycling Treatment on the Anneal Hardening Effect of Cu-10Zn Alloy, J. Alloys Compd., 2010, 489, p 582–585 14. M. Bader, G.T. Eldis, and H. Warlimont, The Mechanisms of Anneal Hardening in Cu-Al Alloys, Metall. Trans. A, 1976, 7, p 249–255 15. S. Nestorovic´, B. Milicevic´, and D. Markovic´, Anneal Hardening Effect in Sintered Copper Alloys, Sci. Sinter., 2002, 34, p 169–174 16. M.D. Teplitskiy, A.K. Nikolayev, N.I. Revina, and V.M. Rozenberg, The Structure and Properties of Aging Copper-Nickel Based Alloys, Fiz. Metal. Metalloved., 1975, 40, p 1240–1243 17. P. Virtanen and T. Tiainen, Stress Relaxation Behaviour in Bending of High Strength Copper Alloys in the Cu-Ni-Sn System, Mater. Sci. Eng. A, 1997, 238, p 407–410

Journal of Materials Engineering and Performance

18. F. Li, X. Zhao, X. Li, and Y.H. Jin, Analysis of Valence Electron Structures of Cu-Zn Alloy Solid Solutions, Ordnance Mater. Sci. Eng., 2009, 32, p 23–25 (in chinese) 19. M. Mabuchi and K. Higashi, Strengthening Mechanism of Mg-Si Alloy, Acta Mater., 1996, 44, p 4611–4618

Journal of Materials Engineering and Performance

20. R.M. Aikin and L. Christodoulou, The Role of Equiaxed Particles on the Yield Stress of Composites, Scr. Metall. Mater., 1991, 25, p 9–14 21. G. Le Roy, J.D. Embury, G. Edward, and M.F. Ashby, A Model of Ductile Fracture Based on the Nucleation and Growth of Voids, Acta Metal., 1981, 29(8), p 1509–1522