Fabrication of Dual-Phase Strengthened Cu–Ti Alloy Sheets

Abstract

The advancement of electronic devices and their capabilities has driven the demand for specific material property combinations, such as mechanical strength and electrical conductivity, in materials pertinent to device fabrication. To this end, the microstructure and properties of Cu–4.2 at% Ti alloy sheets, produced through a multi-step process involving over-aging and severe cold rolling, were investigated. The microstructure of the over-aged alloy prior to cold rolling consisted of cellular components with laminated plates of terminal copper solid solution (Cu_ss) and β–Cu₄Ti. When the over-aged alloy was severely cold rolled for a 99% reduction in thickness, a hierarchical double-phase microstructure was formed parallel to the cold-rolling direction, with Cu_ss bands and two-phase bands containing small β–Cu₄Ti pieces stacked within Cu_ss phase. The strength of the over-aged alloy sheet increased steadily during increasing degrees of cold rolling, caused by a large volume fraction and fine dispersion of hard β–Cu₄Ti pieces and high dislocation density in the Cu_ss matrix. The electrical conductivity decreased in the later stages of cold rolling; however, the conductivity was higher than that of the alloy sheet prepared by peak aging and cold rolling. Eventually, the balance between strength and electrical conductivity of this Cu–Ti alloy was significantly improved by over-aging and severe cold rolling compared to conventional peak-aging and cold rolling processes.

 

This Paper was Originally Published in Japanese in J. Japan Inst. Copper 62 (2023) 68–72. Figures 4 and 5 were slightly modified.

(Left) Tensile strength and electrical conductivity in Cu–4.2 at% Ti alloy sheets, prepared by solid solution treatment (squares), peak-aging (circles), or over-aging (circles), then cold-rolled from a 10 mm thickness to 0.3, 0.2, and 0.1 mm. (Right) Cross-sectional FE-SEM images of Cu–4.2 at% Ti alloys after over-aging and cold rolling from a thickness of 10 mm to 0.3 mm. The brighter and darker regions correspond to the β–Cu₄Ti and Cu phases, respectively. The alloy sheet exhibits a hierarchical double-phase microstructure consisting of laminated Cu bands (with darker contrasts) and two-phase bands containing small β–Cu₄Ti particles suspended in the Cu phase.

1. Introduction

Owing to recent innovations in electronic instruments and devices, the improvement in mechanical and electrical properties of copper and copper alloys is earnestly desired for applications in conductive parts, such as lead frames and contact leaf springs. Among the various commercial copper alloys, age-strengthened Cu–Ti alloys have superior mechanical properties, including yield and tensile strengths, stress relaxation, fatigue properties, and bending workability.^1,2) However, the electrical conductivity of Cu–Ti alloys is inferior to that of other Cu alloys, such as age-strengthened Cu–Be and Cu–Ni–Si alloys.^3–6) Therefore, numerous studies have been conducted to design and improve the strength characteristics and the electrical conductivity of Cu–Ti alloys.^7–9)

Age-strengthened Cu–Ti alloys, which typically contain 3–5 at% titanium, are generally fabricated with a solid solution treatment at a high temperature (>800°C), followed by isothermal aging at a moderate temperature (420–500°C). In an early stage of aging, modulation of solute titanium concentration in the copper solid solution (Cu_ss) matrix phase occurs. Then a fine needle-like metastable β′–Cu₄Ti phase precipitates with lengths of several tens of nanometers, continuously forming in the parent grains.^2,10–12) Upon further aging, cellular components, composed of terminal Cu_ss and stable β–Cu₄Ti plates, laminate and are discontinuously formed at the grain boundaries. In a later stage of aging, the developing cellular components replace the metastable β′–Cu₄Ti needles, and eventually, the cellular components occupy the entire microstructure.¹¹⁾ The strength of the Cu–Ti alloys increases and reaches a maximum by isothermal aging for an optimized period (i.e., peak-aged condition); this occurs when the highest number of β′–Cu₄Ti needles are finely dispersed within the grains before the cellular components develop.¹²⁾ However, when the Cu–Ti alloys are aged over the peak-aging period (i.e., over-aging), the age-induced strengthening degrades because of the development of the cellular components containing coarse β–Cu₄Ti plates, instead of the fine metastable β′–Cu₄Ti needles. On the other hand, the electrical conductivity increases during aging due to reduced solute titanium content in the copper matrix. This results from the nucleation and growth of the Ti-enriched precipitates, such as β′–Cu₄Ti needles and β–Cu₄Ti plates. Thus, although the strength of the over-aged Cu–Ti alloys is not superior to that of the peak-aged alloys, the electrical conductivity is improved. From the viewpoint of industrial utilization, peak-aged Cu–Ti alloys, rather than over-aged alloys, have been a more attractive research and development subject, particularly for applications in conductive components of electronic devices.

Semboshi et al. successfully fabricated Cu–Ti alloy wires with strengths and electrical conductivities superior to those of conventional peak-aged alloy wires.^13,14) The Cu–Ti alloy wires were processed via over-aging and then severe cold-drawing; during over-aging, cellular components occupied the specimen (i.e., fully cellular microstructure), and the subsequent cold-drawing transformed the β–Cu₄Ti plates in the cellular components into nanofibers. The microstructural evolution of the over-aged and severely cold-drawn Cu–Ti alloy wires led to a significant improvement in their strength. Note that the over-aged and cold-drawn alloy wires have a higher electrical conductivity than that of the peak-aged and cold-drawn alloy wires. Therefore, the over-aged and cold-drawn Cu–Ti alloy wires exhibit excellent strength and conductivity. The present study proposed that the procedure of over-aging and severe cold drawing described in the previous study can be applied to producing thin sheets via cold rolling. Essentially, if over-aged Cu–Ti alloys with a fully cellular microstructure are properly deformed into sheets by cold rolling, sheets with an improved balance of strength and conductivity compared with those of the conventional peak-aged alloy sheets can be formed. These materials show promise for applications such as lead frames and contact leaf springs. In this study, an over-aged Cu–Ti alloy featuring a fully cellular microstructure underwent severe cold rolling with a 99% thickness reduction, resulting in the fabrication of high-strength and high-conductivity sheets. The microstructural evolution and variations in the strength and electrical conductivity of the over-aged alloy during cold rolling were systematically investigated and compared with those of the peak-aged alloy. Based on the findings of this study, the effect of the microstructural evolution in the over-aged alloy during cold rolling on its strength and conductivity is also discussed.

2. Experimental Procedure

Solid solution-treated Cu–4.2 at% Ti alloy blocks with a length, width, and thickness of 200, 50, and 10 mm, respectively, were supplied by DOWA METALTECH Co., LTD. The alloy block was aged in multiple steps: 600°C for 3 h, 550°C for 3 h, 500°C for 3 h, and then 450°C for 12 h (21 h in total). According to literature, alloys aged in this manner exhibit a fully cellular microstructure, similar to that of an alloy over-aged at 450°C for 480 h.¹³⁾ In this study, an alloy aged over multiple steps is defined as an “over-aged alloy”. Another alloy block was isothermally peak-aged at 450°C for 12 h to compare with the microstructure and properties of the over-aged alloy. The three blocks of solid solution treated, peak-aged, and over-aged alloys were cold rolled down from 10 mm to 0.1 mm with a reduction of less than 0.5 mm/step. In this study, the degree of deformation caused by cold rolling is represented by the equivalent strain ε, as given in the following equation.

  

\begin{equation} \varepsilon = \ln(t_{0}/t), \end{equation} (1)

where t₀ and t are the thicknesses of the alloy before and after cold rolling, respectively.

The microstructures of the alloy blocks and sheets prepared in this study were observed by field-emission scanning electron microscopy (FE-SEM) using a JEOL JSM-7001F at an acceleration voltage of 15 kV, attached to an electron backscatter diffraction (EBSD) system from TSL solutions, and by transmission electron microscopy (TEM) using a JOEL 2000EX at an acceleration voltage of 200 kV. For FE-SEM observation, the cross-section surface of the alloy sheets was polished using SiC abrasive paper, followed by chemical etching with a solution of 40% nitric acid and 60% pure water. For TEM observation, the requisite thin specimens were prepared by mechanically polishing the cross-section of the alloy to less than 30 µm in thickness and then finished by ion milling. The volume fractions of the age-induced precipitates were measured during the extraction procedure using a solution of 40% nitric acid and 60% pure water. The details of the extraction procedure have been described in a previous study.¹⁵⁾ The dislocation density in the copper matrix was directly analyzed using X-ray diffraction (XRD), employing the diffraction peak width and the Williamson–Hall method with a correction for elastic constants.¹⁶⁾ XRD measurements were performed with a Bragg–Brentano diffractometer utilizing a Cu Kα₁ source.

The Vickers hardness of the alloy sheets was measured using a Mitsutoyo HM-101, applying a load of either 1.96 or 2.94 N; the average hardness values were determined from more than ten indentations. Tensile tests were performed at an initial strain rate of 1.67 × 10⁻⁴ s⁻¹ using a Shimadzu Autograph AG-IS. The electrical conductivity and resistivity were measured via eddy-current testing at 20°C for the specimens with a thickness greater than 1.0 mm or the four-probe method with a direct current of 500 mA or 1 A for the specimens with a thickness less than 1.0 mm. In both methods, the electrical conductivity (resistivity) was measured in the direction parallel to the cold-rolling plane.¹⁷⁾

3. Results and Discussions

3.1 Microstructural evolution during cold rolling

Figure 1 shows the inverse pole figure (IPF) map obtained from the EBSD analysis and FE-SEM images for the peak-aged and over-aged Cu–4.2 at% Ti alloys before cold rolling. We can see that the solid solution-treated alloy has a single phase consisting of a copper solid solution (Cu_ss) with a grain size of approximately 20 µm (Fig. 1(a)). In the peak-aged alloy, fine needle-shaped precipitates with 50–80 nm lengths were homogeneously dispersed in the Cu_ss matrix (Fig. 1(b)). According to the previous reports,^2,11) the needle-shaped precipitates were identified as β′–Cu₄Ti with a tetragonal structure (prototype: Ni₄Mo, lattice parameter: a = 0.586 nm, c = 0.365 nm). In the over-aged alloy, the cellular components with lamellar Cu_ss (fcc, a = 0.361 nm) and β–Cu₄Ti form an orthorhombic structure (Au₄Zr, a = 0.452 nm, b = 0.434 nm, c = 1.292 nm; Fig. 1(c)). The lamination orientation of each cellular component was random. The Cu_ss phase has a solute titanium content of approximately 0.4 at%, corresponding with the solubility limit at 450°C in the Cu–Ti binary system.^2,11)

Fig. 1

Micrographs of Cu–4.2 at% Ti alloys after (a) solid solution treatment, (b) peak-aging at 450°C for 12 h, and (c) over-aging at 600°C for 3 h, 550°C for 3 h, 500°C for 3 h, and 450°C for 12 h. Schematic unit cells of tetragonal β′–Cu₄Ti and orthorhombic β–Cu₄Ti are illustrated in the inset of (b) and (c), respectively.

Figure 2 shows a high-resolution HAADF-STEM image taken from the interface between β–Cu₄Ti and Cu_ss plates in a cellular component in an over-aged alloy.¹⁸⁾ The crystal orientation relationship between the β–Cu₄Ti and Cu_ss plates in the cellular microstructure is described as {010}_β // {111}_Cu, ⟨001⟩_β // $\langle \bar{1}10\rangle_{\text{Cu}}$. The misfit strain between the β–Cu₄Ti and Cu_ss at the boundary is 7%. The β–Cu₄Ti plate has a partially coherent interface with the copper plate (the dotted line in Fig. 2 encloses an incoherent interface). The volume fractions of fine β′–Cu₄Ti needles and coarse β–Cu₄Ti plates in the peak-aged alloy were 1.8% and 17%, respectively, measured using the extraction method. These values are similar to those measured previously.^11,14)

Fig. 2

High-resolution TEM (HAADF-STEM) image of the boundary interface region between lamellae within a cellular component in an over-aged Cu–4.2 at% Ti alloy, as reported by Saito et al.¹⁸⁾ The left half is the orthorhombic β–Cu₄Ti phase viewed along the [001] axis, and the right half is the fcc copper phase viewed along the [011] axis. The atomic structure profiles of the corresponding phases are superimposed. The dotted rectangle shows the location of an incoherent interface site.

The solid solution-treated and peak-aged Cu–4.2 at% Ti alloys were cold-rolled from a thickness t of 10 mm. First, slip bands were introduced into the grains. As cold rolling proceeded, the grains elongated in the rolling direction, producing a typical deformation microstructure for copper and copper alloys.^19,20) In the peak-aged alloy, fine β′–Cu₄Ti needles in the Cu_ss matrix were elongated and broken by cold rolling. In the final stage of cold rolling, at an equivalent strain of ε = 4.6 (t = 0.1 mm), the volume fraction of β′–Cu₄Ti precipitates decreased to approximately 0.3%, according to analysis via the extraction method. This is primarily owing to the partial re-dissolution of the β′–Cu₄Ti precipitates into the Cu_ss matrix during the cold rolling, similar to the case of wire drawing.¹³⁾

The over-aged Cu–4.2 at% Ti alloy was also deformed from a thickness of 10 mm to 0.1 mm or less (ε > 4.6) by cold rolling, as shown in Fig. 3. Notably, the over-aged alloy, containing a large volume fraction (17%) of coarse, brittle β–Cu₄Ti intermetallic plates, was able to deform into a thin sheet via a significant strain of ε > 4.6, despite the presence of several edge cracks in the alloy sheets. The deformability of the over-aged alloy can be attributed to the brittle β–Cu₄Ti plates being surrounded by ductile Cu_ss plates with a partially coherent interface, as illustrated in Fig. 2.²¹⁾

Fig. 3

Appearance of Cu–4.2 at% Ti alloy sheets with a thickness of 0.1 mm, which were prepared by over-aging and then cold rolling to a 99% reduction in thickness.

Figure 4 illustrates the microstructural evolution of the over-aged alloy during the cold-rolling process. The FE-SEM images were obtained from a cross-section perpendicular to the rolling direction. In the initial stage of deformation (ε < 1.0 (t > 3.6 mm)), the β–Cu₄Ti plates (with brighter contrasts in Fig. 4) appeared wavy, generally aligned with the rolling direction, and broke into smaller pieces. The β–Cu₄Ti lamellae thickness and inter-lamellar distance gradually decreased with further cold rolling. In the later stage of deformation (ε = 3.5 (t = 0.3 mm)), the coarse β–Cu₄Ti plates were torn into submicron-sized pieces, eventually forming a hierarchical double-phase microstructure. This microstructure consisted of laminated Cu_ss bands (with darker contrasts) and two-phase bands containing small β–Cu₄Ti pieces suspended in the Cu_ss phase (Figs. 4(c) and (d)). The volume fraction of β–Cu₄Ti in the sheet cold-rolled to a strain of ε = 4.6 decreased to approximately 11% due to the severe cold rolling. This finding implies that the β–Cu₄Ti phase partially redissolves into the Cu_ss matrix during severe cold rolling, as observed in the cold drawing of the wire-form of the same Cu–Ti alloy^13,14) and pearlite steel wires.^22,23)

Fig. 4

Cross-sectional FE-SEM images of Cu–4.2 at% Ti alloys after over-aging and cold rolling from a thickness of 10 mm to (a) 6.0 mm (ε = 0.5), (b) 3.6 mm (ε = 1.0), (c) 1.4 mm (ε = 2.0), and (d) 0.3 mm (ε = 3.5), together with a (d′) high magnification FE-SEM image and (d′′) TEM image. The brighter and darker regions correspond to the β–Cu₄Ti and Cu_ss phases, respectively.

Figure 5 shows the dislocation density in the Cu_ss phase for the solid solution-treated and over-aged alloy sheets during cold rolling, which was determined from XRD measurements. In the case of the solid solution-treated alloy, the dislocation density was already saturated at approximately 5 × 10¹⁵ m⁻² after the initial stage of cold rolling (ε = 1.0 (t = 3.6 mm)). However, when the over-aged alloy was cold-rolled, the dislocation density steadily increased with an increasing degree of deformation, where it exceeded 1 × 10¹⁶ m⁻² with a strain of ε = 4.6 (t = 0.1 mm). When the over-aged alloy with a fully cellular microstructure, containing laminated ductile copper and brittle β–Cu₄Ti plates, was severely cold rolled, a high density of dislocations was accumulated in the copper phase, directly adjacent to the neighboring β–Cu₄Ti phase. In contrast to the solid solution-treated alloy, this shows an efficient accumulation of dislocations in the copper phase. This is similar to the case of two-phase layered composites fabricated via rolling.^24–27)

Fig. 5

Dislocation densities in the Cu_ss matrix for Cu–4.2 at% Ti alloy sheets prepared by a solid solution treatment (squares) or over-aging then cold rolling (circles), as a function of equivalent strain by cold rolling. The dislocation density was estimated by Direct fitting/modified Williamson-Hall Method from XRD measurement.

3.2 Strength and electrical conductivity of cold-rolled sheets

Figure 6 shows the Vickers hardness and electrical conductivity of the solid solution-treated, peak-aged, and over-aged Cu–4.2 at% Ti alloy sheets. When the solid solution-treated Cu–4.2 at% Ti alloy was cold rolled, the hardness initially increased markedly, then plateaued at approximately 280 HV after ε ≧ 1.0 (t ≦ 3.6 mm). This hardening behavior is explained by the deformation hardening effect; the hardening of the solid solution-treated alloy during cold rolling is principally due to the accumulation of dislocations, as shown in Fig. 5. However, the electrical conductivity decreased slightly during cold rolling. This demonstrates that the electrical conductivity (or resistivity) is not very sensitive to deformation strain by dislocation accumulation. It is well known that the conductivity of copper alloys is primarily dependent on impurities or solute elements in the matrix and only slightly dependent on structural defects, such as dislocations and grain boundaries.^28,29)

Fig. 6

Plots of (a) Vickers hardness and (b) electrical conductivity for Cu–4.2 at% Ti alloy sheets, prepared by solid solution treating (diamonds), peak-aging (circles), or over-aging (squares), and then cold rolling, as a function of equivalent strain and sheet thickness.

In the peak-aged alloy, the hardness before cold rolling was 290 HV. It increased gradually during cold rolling, reaching 350 HV when the equivalent strain reached ε = 4.6 (t = 0.1 mm). This is also principally due to the deformation hardening effect. The electrical conductivity was approximately 15% IACS before cold rolling. (Note that the unit “% IACS” is the electrical conductivity relative to that of an annealed pure copper standard measured at 298 K (25°C), which is equal to 5.8 × 10⁷ Ω⁻¹ m⁻¹ and 5.8 × 10⁷ S m⁻¹.) It decreased monotonically during cold rolling and reached a value of less than 6% IACS at ε = 4.6 (t = 0.1 mm). Here, the electrical conductivity of the peak-aged Cu–Ti alloys is useful for estimating the solute titanium content in the Cu_ss matrix. This is done using Nordheim’s equation, ignoring a small fraction of precipitates:^7,12,30)

  

\begin{align} 1/\sigma_{\text{alloy}} &\approx 1/\sigma_{\text{Cu matrix}} = \rho_{\text{Cu matrix}} \\ &= \rho_{\text{Cu}} + AC_{\text{Ti}}(100 - C_{\text{Ti}})/100, \end{align} (2)

where, σ_alloy and σ_{Cu matrix} are the electrical conductivities for the Cu–Ti alloy and the copper matrix in the alloy, respectively; ρ_Cu and ρ_{Cu matrix} are the resistivities for the pure copper (1.724 × 10⁻⁸ Ωm at 20°C) and copper matrix in the alloy, respectively; A is a constant for the Cu–Ti binary system (≈10.2 × 10⁻⁸ Ωm/at%); and C_Ti is the solute titanium content in the copper matrix. According to eq. (2), the amount of titanium in the matrix, C_Ti, in the peak-aged alloy is estimated to increase from 1.0 at% to 2.6 at% Ti after cold rolling. This is caused by the dissolution of the fine β′–Cu₄Ti precipitates during cold rolling, similar to when Cu–Ti alloy wires were fabricated by peak-aging and severe cold-drawing.^13,14)

Finally, the hardness of the over-aged alloy before cold rolling was 155 HV, which is lower than that of the peak-aged alloy (290 HV). However, it increased steadily during cold rolling, eventually exceeding 320 HV when the strain reached ε > 4.6 (t < 0.1 mm). The increase in hardness after cold rolling to an equivalent strain ε = 4.6 was 155 HV, which is greater than the 140 HV increase in the solution heat-treated alloy and the 65 HV increase in the peak-aged alloy. One reason for the significant increase in hardness of the over-aged alloy during cold-rolling is the transformation from the cellular microstructure containing the coarse β–Cu₄Ti plates (volume fraction of 17%) to the hierarchical double-phase microstructure, as shown in Figs. 4(d). This microstructure induces dual-phase reinforcement through the fine dispersion of β–Cu₄T particles in the two-phase bands and deformation hardening due to the efficient accumulation of dislocations in the copper matrix, as illustrated in Fig. 5.

The electrical conductivity of the over-aged alloy before cold rolling was 29% IACS, which was greater than those of the solid solution-treated and peak-aged alloys. It increased slightly to 31% IACS during initial cold rolling (ε = 1.0 (t = 3.6 mm)) and then decreased gradually to approximately 18% IACS for the alloy cold-rolled to ε = 4.6 (t = 0.1 mm). Despite the decrease, this value is still much higher than those of the solid solution-treated and peak-aged alloys. The slight increase in conductivity during the initial stage of cold rolling can be explained by the increasingly parallel arrangement of the β–Cu₄Ti plates to the rolling direction.¹³⁾ The gradual decrease in conductivity during the later stages of cold rolling is primarily due to three co-existing factors. First, the titanium content in the solution increased due to the partial dissolution of β–Cu₄Ti; the variation in solute content in the copper matrix is estimated to increase from 0.4 at% before cold rolling (which is close to the solubility of titanium in the Cu_ss phase at 450°C^12,31)) to 0.8 at% after cold rolling, from eq. (2). Second, the decrease in conductivity is related to the accumulation of dislocations. Third, the interface area between Cu_ss and small platelet β–Cu₄Ti increased, as demonstrated in Fig. 3. The variation in the conductivity of the over-aged alloy as a function of the deformation strain is similar to that of the over-aged and cold-drawn alloy wires.^13,14)

Figure 7 summarizes the ultimate tensile strength and electrical conductivity of Cu–4.2 at% Ti alloy sheets, as fabricated with either a solid solution treatment, peak-aging, or over-aging, followed by cold rolling to three different stages ((1) ε = 3.5 (t = 0.3 mm), (2) ε = 3.9 (t = 0.2 mm), and (3) ε = 4.6 (t = 0.1 mm)). It is well-known that the strength and conductivity of peak-aged and cold-rolled alloy sheets are enhanced compared to those of solid solution-treated alloy sheets. In this study, over-aged and cold-rolled alloy sheets demonstrated a superior combination of strength and conductivity compared to peak-aged and cold-rolled alloy sheets. The electrical conductivity was more than three times that of peak-aged and cold-rolled alloy sheets, even though the ultimate tensile strength decreased by approximately 5%. Thus, alloy sheets prepared by over-aging and severe cold rolling have a strength-conductivity balance not achieved in Cu–Ti alloys prepared by conventional peak-aging and cold-rolling. In the future, the strength-conductivity balance is expected to be widely controlled by optimizing alloy composition and processing conditions, such as the reduction in thickness and deformation temperature.

Fig. 7

Tensile strength as a function of electrical conductivity in Cu–4.2 at% Ti alloy sheets, prepared by a solid solution treatment (squares), peak-aging (circles), or over-aging (circles), then cold-rolled from 10 mm in thickness to 0.3, 0.2, and 0.1 mm.

4. Summary

This study prepared Cu–4.2 at% Ti alloy sheets with a favorite combination of high strength and electrical conductivity through severe cold rolling an over-aged alloy with degraded strength and high electrical conductivity. Subsequently, the microstructural evolution and variations in strength and electrical conductivity during cold rolling were investigated. The following conclusions were drawn from this study.

1. (1)    When cold rolling an over-aged alloy was initially entirely occupied by cellular components of the terminal copper solid solution (Cu_ss) and β–Cu₄Ti laminates, the microstructure progressed in the following sequence. (i) Arrangement of β–Cu₄Ti laminates parallel to the cold-rolling direction, (ii) reduction in the lamellae thickness and inter-lamellar distance, (iii) tearing of the laminates into submicron-sized pieces, and (iv) continuous refinement and dissolution of the pieces, thereby leading to an increase in the solubility of titanium in the Cu_ss matrix. Moreover, the fine β–Cu₄Ti pieces promote an increase in dislocation density of the Cu_ss phase, thereby leading to a significant increase in strength, which is not seen in the case of cold rolling a solid solution-treated alloy.
2. (2)    When cold rolling the over-aged alloy, the strength steadily increased, and the hardness eventually exceeded 310 HV, with an ultimate tensile strength of 1200 MPa, at an equivalent strain of ε = 4.6 (99% reduction in thickness). The electrical conductivity slightly increased during initial cold rolling but dropped during more severe cold rolling. However, the conductivity was always greater than that of the peak-aged alloy sheets. In summary, the balance between strength and conductivity was enhanced by employing an over-aging procedure followed by cold rolling instead of the conventional peak-aging and cold-rolling processes.

Acknowledgments

The authors thank Prof. K. Saito of Akita University for permission to use the HAADF-STEM images (Fig. 2). They also thank Mr. S. Ito and Mr. E. Aoyagi of the Institute for Materials Research (IMR) of Tohoku University for their assistance with the experiments. This work was supported by the cooperative program of the Collaborative Research and Development Center for Advanced Materials (CRDAM) at the IMR (No. 202212-CRKEQ-0403). The authors gratefully acknowledge financial support from the Japan Society for the Promotion of Science via a Grant-in-Aid for Scientific Research (B) (22H01825) and from the Japan Copper and Brass Association.

REFERENCES

• 1)   S.  Nagarjuna,  K.  Balasubramanian and  D.S.  Sarma: J. Mater. Sci. 34 (1999) 2929–2942. doi:10.1023/A:1004603906359
• 2)   W.A.  Soffa and  D.E.  Laughlin: Prog. Mater. Sci. 49 (2004) 347–366. doi:10.1016/S0079-6425(03)00029-X
• 3)   T.  Sakai,  H.  Miura and  N.  Muramatsu: Mater. Trans. JIM 36 (1995) 1023–1030. doi:10.2320/matertrans1989.36.1023
• 4)   X.  Guoliang,  W.  Qiangsong,  M.  Xujun,  X.  Baiqing and  P.  Lijun: Mater. Sci. Eng. A 558 (2012) 326–330. doi:10.1016/j.msea.2012.08.007
• 5)   Q.  Lei,  Z.  Li,  T.  Xiao,  Y.  Pang,  Z.Q.  Xiang,  W.T.  Qiu and  Z.  Xiao: Intermetallics 42 (2013) 77–84. doi:10.1016/j.intermet.2013.05.013
• 6)   M.  Gholami,  J.  Vesely,  I.  Altenberger,  H.-A.  Kuhn,  M.  Janecek,  M.  Wollmann and  L.  Wagner: J. Alloy. Compd. 696 (2017) 201–212. doi:10.1016/j.jallcom.2016.11.233
• 7)   S.  Semboshi and  T.J.  Konno: J. Mater. Res. 23 (2008) 473–477. doi:10.1557/JMR.2008.0050
• 8)   M.  Sobhani,  A.  Mirhabibi,  H.  Arabi and  R.M.D.  Brydson: Mater. Sci. Eng. A 577 (2013) 16–22. doi:10.1016/j.msea.2013.03.063
• 9)   H.  Wei,  Y.  Cui,  H.  Cui,  Y.  Wei and  L.  Hou: Mater. Sci. Eng. A 707 (2017) 392–398. doi:10.1016/j.msea.2017.09.072
• 10)   T.K.  Vaidyanathan and  K.  Mukherjee: J. Mater. Sci. 10 (1975) 1697–1710. doi:10.1007/BF00554931
• 11)   S.  Semboshi,  S.  Sato,  M.  Ishikuro,  K.  Wagatsuma,  A.  Iwase and  T.  Takasugi: Metall. Mater. Trans. A 45 (2014) 3401–3411. doi:10.1007/s11661-014-2293-2
• 12)   S.  Semboshi,  S.  Amano,  J.  Fu,  A.  Iwase and  T.  Takasugi: Metall. Mater. Trans. A 48 (2017) 1501–1511. doi:10.1007/s11661-016-3949-x
• 13)   S.  Semboshi,  Y.  Kaneno,  T.  Takasugi and  N.  Masahashi: Metall. Mater. Trans. A 49 (2018) 4956–4965. doi:10.1007/s11661-018-4816-8
• 14)   S.  Semboshi,  Y.  Kaneno,  T.  Takasugi,  S.Z.  Han and  N.  Masahashi: Metall. Mater. Trans. A 50 (2019) 1389–1396. doi:10.1007/s11661-018-5088-z
• 15)   S.  Semboshi,  M.  Ishikuro,  S.  Sato,  K.  Wagatsuma and  T.  Takasugi: Mater. Charact. 82 (2013) 23–31. doi:10.1016/j.matchar.2013.04.016
• 16)   D.  Akama,  T.  Tsuchiyama and  S.  Takaki: J. Soc. Mater. Sci., Japan (Japanese) 66 (2017) 522–527. doi:10.2472/jsms.66.522
• 17)   M.  Rosen: Metall. Trans. A 20 (1989) 605–610. doi:10.1007/BF02667578
• 18)   K.  Saito,  M.  Suzuki,  Y.  Hayasaka,  K.  Takenaka and  S.  Semboshi: Mater. Charact. 189 (2022) 111911. doi:10.1016/j.matchar.2022.111911
• 19)   A.  Yoshida,  Y.  Miyajima and  S.  Onaka: Mater. Trans. 53 (2012) 26–29. doi:10.2320/matertrans.MD201116
• 20)   H.T.  Jeong and  W.J.  Kim: J. Mater. Sci. Technol. 111 (2022) 152–166. doi:10.1016/j.jmst.2021.08.038
• 21)   M.  Zelin: Acta Mater. 50 (2002) 4431–4447. doi:10.1016/S1359-6454(02)00281-1
• 22)   W.J.  Nam,  C.M.  Bae,  S.J.  Oh and  S.-J.  Kwon: Scr. Mater. 42 (2000) 457–463. doi:10.1016/S1359-6462(99)00372-3
• 23)   N.  Guelton and  M.  François: Metall. Mater. Trans. A 51 (2020) 1602–1613. doi:10.1007/s11661-020-05640-4
• 24)   J.T.  Avallone,  T.J.  Nizolek,  T.M.  Pollock and  M.R.  Begley: Mater. Sci. Eng. A 761 (2019) 138016. doi:10.1016/j.msea.2019.06.026
• 25)   H.  Seyedrezai,  A.K.  Pilkey and  J.D.  Boyd: Mater. Sci. Eng. A 594 (2014) 178–188. doi:10.1016/j.msea.2013.11.034
• 26)   D.J.  Savage,  I.J.  Beyerlein,  N.A.  Mara,  S.C.  Vogel,  R.J.  McCabe and  M.  Knezevic: Int. J. Plast. 125 (2020) 1–26. doi:10.1016/j.ijplas.2019.08.015
• 27)   J.S.  Carpenter,  C.  Miller,  D.J.  Savage,  D.R.  Coughlin,  E.L.  Tegtmeier and  A.P.  Wintter: Metall. Mater. Trans. A 53 (2022) 2208–2213. doi:10.1007/s11661-022-06662-w
• 28)   J.W.  Rutter and  J.  Reekie: Phys. Rev. 78 (1950) 70–71. doi:10.1103/PhysRev.78.70
• 29)   Y.  Abe,  S.  Semboshi,  N.  Masahashi,  S.H.  Lim,  E.-A.  Choi and  S.Z.  Han: Metall. Mater. Trans. A 54 (2023) 928–938. doi:10.1007/s11661-022-06938-1
• 30)   S.  Nagarjuna,  K.  Balasubramanian and  D.S.  Sarma: Mater. Sci. Eng. A 225 (1997) 118–124. doi:10.1016/S0921-5093(96)10578-5
• 31)   M.A.  Turchanin,  P.G.  Agraval and  A.R.  Abdulov: Powder Metall. Met. Ceramics 47 (2008) 344–360. doi:10.1007/s11106-008-9026-2