Effect of Annealing on Mechanical Properties and Nanoscale Lamellar ...

Materials Transactions, Vol. 50, No. 3 (2009) pp. 570 to 578 #2009 The Japan Institute of Metals

Effect of Annealing on Mechanical Properties and Nanoscale Lamellar Structure in Co-Cu Alloy Motohiro Yuasa1; *1 , Hiromi Nakano2 , Yoshiaki Nakamoto1; *2 and Mamoru Mabuchi1 1 2

Department of Energy Science and Technology, Graduate School of Energy Science, Kyoto University, Kyoto 606-8501, Japan Electron Microscope Laboratory, Faculty of Science and Technology, Ryukoku University, Ostu 520-2194, Japan

An electrodeposited Co-Cu alloy with a nanoscale lamellar structure was annealed at 673–973 K. The effects of annealing on the mechanical properties and the microstructures of the alloy such as the lamellar area ratio and the lamellar spacing were investigated. The Young’s modulus of the Co-Cu alloy increased by annealing. This is mainly due to a decrease in the lamellar area ratio. The as-deposited specimens exhibited a low value of the activation volume for plastic deformation; however, the activation volume was increased by annealing. This appears to be related to an increase in lamellar spacing caused by annealing. The origins of the lamellar spacing dependence of the activation volume are discussed from the viewpoint of the emission of dislocations from the lamellar boundary. [doi:10.2320/matertrans.MRA2008367] (Received October 8, 2008; Accepted December 5, 2008; Published January 28, 2009) Keywords: cobalt, nanoscale lamellar structure, hardness, activation volume, microstructure-property relation

1.

Introduction

Multifunctional materials are often required for the development of advanced functional systems and machinery. For example, high-strength materials exhibiting unique electric and/or magnetic properties can contribute to advances in small-size electric device and machine systems. Recently, it was reported that nanoscale growth twins with coherent boundaries induce both high strength and high electrical conductivity in Cu.1) To date, many studies have been conducted on nanocrystalline metals whose grain boundaries are not coherent, and a deep understanding of the roles of grain boundaries in deformation mechanisms has been obtained.2–5) For example, in the range of 10 nm < d < 100 nm, where d is the grain size, the grain boundaries are not the sources of dislocation pileup, but are sources of dislocation emission and absorption; on the other hand, in the range of d < 10 nm, the grain boundaries are not even the sources of dislocation emission and absorption.6–8) Recently, it was reported that nanoscale twins are sites for the pileup and emission of dislocations, resulting in high strength and high ductility in Cu.9–11) Zhu et al.12) showed from slip transfer reactions mediated by twin boundaries, that high ductility can be attributed to the hardening of twin boundaries, namely, high ductility is a result of twin boundaries losing coherency during deformation. Another characteristic in nano-twinned Cu is the low activation volume of 12-22b3 ,9) where b is the Burgers vector of a dislocation. Asaro and Suresh13) attributed the low activation volume for plastic deformation to the emission of a partial or perfect dislocation from nanoscale twins. The deformation mechanisms in nano-twinned metals depend on the planar fault energy,14,15) suggesting that the roles of twins depend on the materials. Sort et al.16) showed that high hardness is attained in Co due to the presence of *1Graduate

Student, Kyoto University Student, Kyoto University. Present address: Toyota Motor Corporation, Toyota 471-8571, Japan

*2Graduate

twins. In addition, it was recently reported that electrodeposited Co-Cu alloy containing a high-density in-growth nanoscale lamellar structure with a narrow spacing of 3 nm exhibits high strength and high ductility.17) It was noted in this work that the activation volume of the electrodeposited Co-Cu alloy was only 3:3b3 . Futhermore, electrodeposited Co-Cu alloy exhibits unique soft magnetic behavior.18) Clearly, the distinguishing characteristics of electrodeposited Co-Cu alloys are related to the presence of the nanoscale lamellar structure. However, there are too few data available to determine the relationship between the nanoscale lamellar structure and the properties of the alloy. It is worthwhile investigating the microstructure-property relationship in nanoscale lamellar Co-Cu alloy with the ultimate aim of realizing more magnetic/mechanical applications. The microstructure of electrodeposited metals strongly depends on the processing conditions such as the substrate, current density, additives, and annealing temperature.19,20) In the present work, the nanoscale lamellar Co-Cu alloy is annealed at 673–973 K to change its microstructure, and the relationships between the annealing temperature and the microstructure such as the lamellar spacing are examined. In addition, the effects of the nanoscale lamellar structure on the mechanical properties of Co-Cu alloy are investigated. 2.

Experimental

Two types of Co-Cu alloy samples with an in-growth nanoscale lamellar structure were processed by electrodeposition: one (sample A) was processed with no additives at a current density of 40 mA/cm2 , and the other (sample B) was processed with an additive of 0.015 g/L saccharin and a current density of 15 mA/cm2 . The electrolyte composition was CoSO4 7H2 O (1 M) and CuSO4 5H2 O (0.025 M). The pH of the electrolyte was adjusted to 5.0 using H2 SO4 . The bath temperature was maintained at 292 K. The film specimens with the thickness of 30 mm were electrodeposited on an amorphous Fe substrate plate. The deposited specimens of samples A and B were annealed at 673, 773, 873, and 973 K for 1 h in an Ar + 5% vol H2 atmosphere.

Effect of Annealing on Mechanical Properties and Nanoscale Lamellar Structure in Co-Cu Alloy

(a)

(b)

fcc-Co (111) Cu (200) Cu (220) Cu (111) fcc-Co (200)

hcp-Co (110) fcc-Co (220)

Intensity (a. u.)

Intensity (a. u.)

annealed at 773 K

hcp-Co (110) fcc-Co (220) annealed at 973 K

annealed at 973 K

annealed at 873 K

571

anneald at 873 K

annealed at 773 K

annealed at 673 K annealed at 673 K As-deposited

As-deposited

30

40

50 60 70 Difraction Angle, 2θ /deg

Fig. 1

80

90

40

50 60 70 Diffraction Angle, 2θ /deg

80

90

X-ray diffraction patterns of the Co-Cu alloy: (a) sample A and (b) sample B.

The microstructure of the Co-Cu alloy specimens was investigated by transmission electron microscopy (TEM). The TEM observation was carried out with a JEOL JEM2010 at an operating voltage of 200 kV and a JEOL JEM3000F at an operating voltage of 300 kV. The specimens observed by TEM were thinned using a dimple grinder and Ar ion milling. From the large amount of data obtained by TEM observation, the grain size, the lamellar area ratio, and the lamellar spacing were measured. The lamellar area ratio was defined as the area of the observed nanoscale lamellar structure, divided by the total observed area. The lamellar spacing was defined as the average distance between the lamellar boundaries in the lamellar area. Energy-dispersive X-ray (EDX) analyses were carried out at 200 kV by TEM using EDX equipment (Noran Instruments VANTAGE) to investigate the chemical composition of the Co-Cu alloy specimens. Also, X-ray diffraction (XRD) experiments were performed on the Co-Cu alloy specimens using Cu-K radiation. The hardness and Young’s modulus were measured at room temperature from indentation data following the procedure outlined in Refs. 21–23). A Shimadzu DUHW201 hardness tester equipped with a diamond Berkovich tip was used. The hardness tests were carried out at loading rates of 13.24, 1.324, and 0.378 mN/s. The tests were performed 10 times at each loading rate. The tip was brought into contact with the specimen, and then the specimen was indented at a constant loading rate to a depth of 600 nm. The load was kept constant at its maximum value for 10 s, and finally the specimen was unloaded. 3.

30

Results

The XRD patterns of the Co-Cu alloy specimens are shown in Fig. 1. It appears that in the as-deposited specimen of sample A, there are hcp-Co (110) and/or fcc-Co (220) peaks, although it is difficult to distinguish an hcp-Co (110) peak from an fcc-Co (220) peak in the XRD profile. The coexistence of hcp and fcc Co phases in the Co-Cu alloy has been observed by high-resolution TEM.24) In general, polycrystalline Co is in the hcp phase at room temperature, and it transforms to the fcc phase at 723 K. However, the fcc

Co phase is stable even at room temperature in nanocrystalline Co.25) Hence, the coexistence of hcp and fcc Co phases in the Co-Cu alloy may be due to the high stability of the nanoscale fcc Co phase. In sample B, not only peaks for Co, but also a peak for Cu (220) was observed in the as-deposited specimen. The intensity of the Cu (220) peak increased with the annealing temperature. Other Cu peaks were also found in the annealed specimens of sample B. On the other hand, for sample A, no peaks for Cu phase were found even after annealing. However, small Cu phases that cannot be detected by XRD analysis may have precipitated in sample A. The solid solubility limit of Cu in Co is almost 0% in the equilibrium state at room temperature. Hence, the Cu must be forced to dissolve into the Co in the Co-Cu alloy because electrodeposition tends to cause the nonequilibrium state. As shown in Fig. 1, the precipitation of the Cu phase was enhanced by the addition of saccharin, suggesting that the addition of saccharin makes the dissolved Cu atoms or the nanoscale lamellar structure unstable. Transmission electron micrographs of the Co-Cu alloy specimens are shown in Fig. 2, where (a) is the as-deposited specimen (sample A), (b) is the specimen annealed at 773 K (sample A), (c) is the specimen annealed at 973 K (sample A), (d) is the as-deposited specimen (sample B), (e) is the specimen annealed at 773 K (sample B), (f) is the specimen annealed at 973 K (sample B). The average grain size was 110 nm for the as-deposited specimen of sample A and 111 nm for the as-deposited specimen of sample B. The grain size increased with increasing annealing temperature. The nanoscale lamellar structure was observed in almost all grains for the as-deposited specimens of both samples A and B. It should be noted that the nanoscale lamellar structure was even found in many grains of the specimens annealed at 773 K, indicating that the nanoscale lamellar structure is thermally stable. A high-resolution transmission electron micrograph of the specimen of sample A annealed at 773 K is shown in Fig. 3. It is clear from the figure that fcc/fcc twins exist. The nanoscale lamellar structure consists of the twin boundaries and hcp/fcc Co boundaries.24) The relationship between the grain size and the annealing temperature is shown in Fig. 4. The grain size gradually

572

M. Yuasa, H. Nakano, Y. Nakamoto and M. Mabuchi

(c)

(b)

(a)

100 nm

500 nm

200 nm

(f)

(e)

(d)

200 nm

100 nm

500 nm

Fig. 2 Transmission electron micrographs of the Co-Cu alloy: (a) as-deposited specimen (sample A), (b) specimen annealed at 773 K (sample A), (c) specimen annealed at 973 K (sample A), (d) as-deposited specimen (sample B), (e) specimen annealed at 773 K (sample B), and (f) specimen annealed at 973 K (sample B).

111 002

002T

111T

111 000 111T

111

002

002T

20nm

Fig. 3 Transmission electron micrograph of the specimen of sample A annealed at 773 K.

600

Grain Size, d / nm

500

Sample A Sample B

400 300 200 100 0 200

400

600

800

1000

Annealing Temperature, T / K Fig. 4 Relationship between the grain size and the annealing temperature for the Co-Cu alloy.

increased with annealing temperature up to 773 K, and then it rapidly increased from 773 to 873 K. This trend was independent of the addition of saccharin. Wu et al.26) showed that grain refinement was enhanced for Ni-Co alloy by the addition of saccharin. In the present work, however, the addition of saccharin addition did not necessarily induce grain refinement of the Co-Cu alloy. The variation of the lamellar area ratio as a function of annealing temperature is shown in Fig. 5(a). For sample A, the lamellar area ratio gradually decreased with annealing temperature up to 773 K, and then it rapidly decreased from 773 to 873 K. For sample B, the lamellar area ratio gradually decreased with annealing temperature up to 673 K, then it moderately decreased from 673 to 873 K, and finally it rapidly decreased above 873 K.


(b) 50

0.8

40

Lamellar Spacing, s / nm

(a) 1

Lamellar Area Ratio

Sample A Sample B

0.6 0.4 0.2 0 200

400

600

800

573

Sample A Sample B

30 20 10 0 200

1000

400

Annealing Temperature, T / K

600

800

1000

Annealing Temperature, T / K

Fig. 5 Variation of lamellar area ratio and spacing as a function of annealing temperature for the Co-Cu alloy: (a) lamellar area ratio and (b) lamellar spacing.

40

40

(a)

(b)

13.24 (mN/s)

13.24 (mN/s) 30

1.324 (mN/s) Load, F / mN

Load, F / mN

30

0.378 (mN/s) 20

10

0.378 (mN/s) 20

10

0 0

0.2 0.4 Displacement, h / µm

30

0 0

0.6

0.6

(d) 13.24 (mN/s) 30


Load, F / mN


40

(c)

20

1.324 (mN/s)

0.378 (mN/s)

10

as deposited 673 K 773 K 883 K 993 K

20

10

0 0


0.6

0 0


0.6

Fig. 6 Load-displacement curves obtained from the hardness tests for sample A: (a) as-deposited specimen, (b) specimen annealed at 773 K, (c) specimen annealed at 973 K, and (d) constant loading rate of 13.24 mN/s.

The variation of the lamellar spacing as a function of annealing temperature is shown in Fig. 5(b). Note that the variation of lamellar spacing with temperature for sample A was much greater than that for sample B. For sample A, the lamellar spacing gradually increased with annealing temperature up to 873 K, and then it rapidly increased from 873 to 973 K. As shown in Fig. 5(a), the lamellar area ratio rapidly decreased from 773 to 873 K. Thus, the temperature range for the large change in lamellar spacing was different from

that for the large change in the lamellar area ratio for sample A. On the other hand, for sample B, the lamellar spacing gradually increased with annealing temperature up to 673 K, then it moderately increased from 673 to 873 K and rapidly increased above 873 K. This trend for the temperature dependence of lamellar spacing was similar to that for the temperature dependence of the lamellar area ratio for sample B. The load-displacement curves obtained from the hardness

574


40

40

(a)

(b) 13.24 (mN/s)

13.24 (mN/s) 30


Load, F / mN

30

0.378 (mN/s) 20

10

0.378 (mN/s) 20

10

0 0

40

1.324 (mN/s)


0 0

0.6

40

(c)


(d) as deposited 673 K 773 K 873 K 973 K

13.24 (mN/s) 30


Load, F / mN

30

0.378 (mN/s) 20

10

0.6

20

10

0 0


0 0

0.6


0.6

Fig. 7 Load-displacement curves obtained from the hardness tests for sample B: (a) as-deposited specimen, (b) specimen annealed at 773 K, (c) specimen annealed at 973 K, and (d) constant loading rate of 13.24 mN/s.

190

(a)

Young's Modulus, E/ GPa

Young's Modulus, E / GPa

190

180

170

160

150 200

Fig. 8

400 600 800 Annealing Temperature, T / K

1000

(b)

180

170

160

150 200

400 600 800 Annealing Temperature, T / K

1000

Variation of Young’s modulus as a function of annealing temperature: (a) sample A and (b) sample B.

tests are shown in Fig. 6 for sample A and Fig. 7 for sample B, where (a) shows the curves at three loading rates for the asdeposited specimen, (b) shows the curves at three loading rates for the specimen annealed at 773 K, (c) shows the curves at three loading rates for the specimen annealed at 973 K, and (d) shows the curves at a loading rate of 13.24 mN/s for the as-deposited specimen and the specimens annealed at 673, 773, 873, and 973 K, respectively. Clearly, the hardness decreased with increasing annealing temperature. Note that the as-deposited specimen showed greater loading rate dependence of hardness than the annealed specimens for both samples A and B.

4.

Discussion

The variation of Young’s modulus as a function of annealing temperature is shown in Fig. 8(a) for sample A and Fig. 8(b) for sample B. For nanocrystalline Co with a grain size of 12 nm, its strength was 2–3 times higher than that of conventional polycrystalline Co, while its Young’s modulus was comparable to that of polycrystalline Co.27) For the nanoscale lamellar Co-Cu alloy, however, the Young’s moduli of the as-deposited specimens were significantly lower than that of polycrystalline Co (¼ 212{223 GPa27)), and the Young’s moduli of the Co-Cu alloy specimens were

Effect of Annealing on Mechanical Properties and Nanoscale Lamellar Structure in Co-Cu Alloy 6

6

(b) Hardness, H / GPa

(a) Hardness, H / GPa

575

5

4

3

5

4

3

10-1

100 . 101 Loading Rate, h / mN/s

102


102

10-1


102

6 Hardness, H / GPa

(c) 5

4

3 10-1

Fig. 9 Variation of hardness as a function of loading rate for sample A: (a) as-deposited specimen, (b) specimen annealed at 773 K, and (c) specimen annealed at 973 K.

increased by annealing. It can be seen from Fig. 8(a) that the Young’s modulus of sample A gradually increased up to 773 K and then rapidly increased above 773 K. This trend for the variation of Young’s modulus for sample A is similar to that for the variation of the lamellar area ratio, namely, the lamellar area ratio gradually decreased with annealing temperature up to 773 K then rapidly decreased from 773 to 873 K, as shown in Fig. 5(a). For sample B, the Young’s modulus rapidly increased above 873 K. This trend for the variation of Young’s modulus for sample B is also similar to that for the variation of the lamellar area ratio. Therefore, it is likely that the Young’s modulus of the Co-Cu alloy is closely related to the lamellar area ratio. However, the low Young’s modulus for the Co-Cu alloy cannot be explained simply by the presence of a nanoscale lamellar structure because the Young’s moduli of the specimens annealed at 973 K, whose lamellar area ratios are only 10%, are lower than that of polycrystalline Co. As shown in Fig. 1, Cu phases were precipitated during annealing for sample B, and their precipitation probably also occurred for sample A. The Young’s modulus of Cu (¼ 136 GPa) is much lower than that of Co. Hence, Cu precipitates as well as the nanoscale lamellar structure may contribute to a reduction of the Young’s modulus for the Co-Cu alloy. The variation of hardness as a function of loading rate is shown in Fig. 9 for sample A and Fig. 10 for sample B, where (a) shows the results for the as-deposited specimen, (b) shows the results for the specimen annealed at 773 K and (c) shows the results for the specimen annealed at 973 K. The activation volume for plastic deformation is given by13) pffiffiffi @ ln "_ ð1Þ v ¼ 3kT @ where v is the activation volume, "_ is the strain rate, is the

flow stress (which is generally assumed to be one-third of the hardness), k is the Boltzmann constant, and T is the absolute temperature. The strain-rate sensitivity and activation volume obtained from the results in Figs. 9, and 10 are listed in Table 1. Note that the activation volume for sample A more rapidly increased with increasing annealing temperature than that for sample B, although the effect of the nanoscale lamellar structure on the activation volume was minor for the specimens annealed at 973 K because of their low lamellar area ratio. As shown in Fig. 5(b), the variation of lamellar spacing with temperature for sample A was much greater than that for sample B. The trend for the variation of lamellar spacing with annealing temperature is in agreement with the trend for the variation of activation volume with annealing temperature. Therefore, the activation volume is likely to be related to the lamellar spacing. A large strain-rate sensitivity and low activation volume have been also obtained in nanocrystalline metals.28–30) The grain-boundary-affected zone model has been proposed to explain the high strain-rate sensitivity of nanocrystalline metals.28) Recently, Dao et al.31) revealed the origin of the high strain-rate sensitivity of nano-twinned Cu using analogous model called the twin-boundary-affected zone. This model can explain the twin spacing dependence of the strainrate sensitivity. However, the rate-controlling process should be considered to understand the origins of the high strain-rate sensitivity and the low activation volume. Lattice dislocations, which play a critical role in plastic deformation, are emitted from twin boundaries.9,11,32,33) Therefore, the ratecontrolling process of plastic deformation in nano-twinned metals is the emission of dislocations from the boundaries. Asaro and Suresh13) analyzed the activation volume for nanocrystalline and nano-twinned metals, assuming that the rate-controlling process is the emission of partial or perfect

576

M. Yuasa, H. Nakano, Y. Nakamoto and M. Mabuchi 6

6

(b) Hardness, H / GPa

Hardness, H / GPa

(a) 5

4

5

4

3

3 10-1


102


102

10-1


102

6 Hardness, H / GPa

(c) 5

4

3 10-1

Fig. 10 Variation of hardness as a function of loading rate for sample B: (a) as-deposited specimen, (b) specimen annealed at 773 K, and (c) specimen annealed at 973 K. Table 1 Strain-rate sensitivity, m, and activation volume for plastic deformation, v , in the Co-Cu alloy. Strain-rate sensitivity, m

Activation volume, v

as-deposited

0.055

3.3b3

annealed at 773 K

0.033

10b3

annealed at 973 K

0.009

44b3

Specimen Sample A

Sample B

as-deposited

0.026

12b3

annealed at 773 K

0.017

19b3

annealed at 973 K

0.012

29b3

lattice dislocations due to stress concentrations at the boundary. According to their result, the critical athermal radius of a perfect dislocation emitted from a grain boundary in a nanocrystalline metal, rc , is given by 1 5 5 dflnðrc =rÞ þ 1g ð2Þ bflnðrc =rÞ þ 1g rc 16 G 16 where is the shear stress resolved along the direction of a dislocation and G is the shear modulus. An increase in activation volume with increasing grain size can be explained from eq. (2) because the radius increases with increasing grain size. In nano-twinned metals, a similar situation may occur when dislocations are emitted from a twin boundary. Another feature in nano-twinned metals is the pileup of dislocations by the twins.9,34,35) Therefore, the stress concentrations caused by the pileup of dislocations are suggested to enhance the emission of dislocations, resulting in the boundary spacing dependence of the activation volume. It was reported that the hardness of Cu/steel multilayers increased with decreasing boundary spacing from 300 to 50 nm and it saturated at the boundary spacing of a few tens

of nm.36) Also, it was shown that the yield stress of nanotwinned Cu follows the empirical Hall-Petch relationship in the boundary spacing range of 15–96 nm,10) indicating that the pileup of dislocations against the twins can occur at the boundary spacing of 15 nm. However, when the boundary spacing is 15–20 nm, only 2 dislocations pile up at the boundary.37) Hence, when the boundary spacing is too narrow, for example, less than 10 nm, no pileup of dislocations occurs at the boundary and the transmission of a single dislocation plays a critical role in deformation. The experimental results in the present work demonstrate that even when the boundary spacing is less than 10 nm, the strain-rate sensitivity and activation volume depend on the boundary spacing. This cannot be explained by the enhancement of the dislocation emission by the pileup of dislocations. The EDX spectrum of the specimen annealed at 973 K in sample A is shown in Fig. 11. It was found that from the EDX analyses that the Co content was 93 mass% and the Cu content was 7 mass% for the as-deposited specimen.17) On the other hand, the Co content was 95 mass% and the Cu content was 5 mass% for the specimen annealed at 973 K. Because the solid solubility limit of Cu in Co is almost 0% in the equilibrium state at room temperature, the Cu is forced to dissolve into the Co in the Co-Cu alloy. No nanoscale lamellar structure was formed in electrodeposited pure Co processed under the same conditions as those for the nanoscale lamellar Co-Cu alloy. This indicates that the dissolved Cu atoms play an important role in the generation of the nanoscale lamellar structure. Hence, the dissolved Cu atoms may be located at the lamellar boundaries due to a reduction of the boundary energy. In such a case, the dissolved Cu atoms can pin the dislocations emitted from the boundaries, namely, the critical radius of a dislocation emitted from the boundary may depend on the distance


(a)

Intensity (a. u.)

1200

A

(b)

577

Co

900 600 300

Co Cu

Co Cu

200 nm

0 0

2

4 6 Energy, E / keV

8

10

Fig. 11 Energy-dispersive X-ray spectrum of sample A: (a) transmission electron micrograph of the specimen annealed at 973 K and (b) energy-dispersive X-ray profile at point A in (a).

between the dissolved Cu atoms located at the boundary. The number of Cu atoms dissolved in the Co matrix was decreased by annealing. Hence, in the low spacing range of less than approximately 10 nm, the increase in activation volume by annealing may be attributed to an increase in the distance between the dissolved Cu atoms located at the boundary. Another possibility for the increase in activation volume by annealing is a variation in stacking fault energy of Co by dissolution of Cu. It is known that the stacking fault energy affects the deformation characteristics of nanocrystalline metals such as the dislocation emission from the boundaries.38) The stacking fault energy of Co may be decreased by dissolution of Cu, suggesting that an increase in stacking fault energy of Co due to precipitation of the dissolved Cu atoms is responsible for the increase in activation volume by annealing. Anyway, it is suggested that the dissolved Cu atoms play an important role in deformation characteristics in the small-boundary-spacing range where no pileup of dislocations at the boundary occurs. Further research is needed to understand the deformation characteristics in the small-boundary-spacing range. 5.

Conclusions

A Co-Cu alloy containing a nanoscale lamellar structure was fabricated by electrodeposition. The Co-Cu alloy was annealed at different temperatures, and the relationships between the annealing temperature and the microstructure such as the lamellar spacing were examined. On the basis of the results, the effects of the nanoscale lamellar structure on the mechanical properties of the Co-Cu alloy were investigated. The following main conclusions were reached. (1) The lamellar area ratio decreased and the lamellar spacing increased with increasing annealing temperature. The nanoscale lamellar structure was relatively stable even at 773 K; however, the lamellar area ratio decreased to about 10% by annealing at 973 K. (2) The Young’s modulus of the Co-Cu alloy was increased by annealing. This is mainly due to a decrease in the lamellar area ratio by annealing.

(3) The as-deposited specimens showed a low value of the activation volume for plastic deformation, which was increased by annealing. The activation volume appears to be related to the lamellar spacing. (4) It is suggested that the lamellar spacing dependence of the activation volume is related to the enhancement of the dislocation emission by the pileup of dislocations at the boundary. On the other hand, in the small-boundaryspacing range where no pileup of dislocations occurs at the boundary, the dissolved Cu atoms may play an important role in deformation characteristics. REFERENCES 1) L. Lu, Y. Shen, X. Chen, L. Qian and K. Lu: Science 304 (2004) 422–426. 2) H. Gleiter: Acta Mater. 48 (2000) 1–29. 3) K. S. Kumar, H. Van Swygenhoven and S. Suresh: Acta Mater. 51 (2003) 5743–5774. 4) D. Wolf, V. Yamakov, S. R. Phillpot, A. Mukherjee and H. Gleiter: Acta Mater. 53 (2005) 1–40. 5) M. A. Meyers, A. Mishra and D. J. Benson: Prog. Mater. Sci. 51 (2006) 427–556. 6) H. Van Swygenhoven and P. M. Derlet: Phys. Rev. B 64 (2001) 224105(1)–(9). 7) D. Torre and K. W. Jacobsen: Nature 391 (1998) 561–563. 8) V. Yamakov, D. Wolf, S. R. Phillpot, A. K. Mukherjee and H. Gleiter: Nat. Mater. 3 (2004) 43–47. 9) L. Lu, R. Schwaiger, Z. W. Shan, M. Dao, K. Lu and S. Suresh: Acta Mater. 53 (2005) 2169–2179. 10) Y. F. Shen, L. Lu, Q. H. Lu, Z. H. Jin and K. Lu: Scr. Mater. 52 (2005) 989–994. 11) Y. F. Shen, L. Lu, M. Dao and S. Suresh: Scr. Mater. 55 (2006) 319–322. 12) T. Zhu, J. Li, A. Samanta, H. G. Kim and S. Suresh: PNAS 104 (2007) 3031–3036. 13) R. J. Asaro and S. Suresh: Acta Mater. 53 (2005) 3369–3382. 14) A. G. Fro¨seth, P. M. Derlet and H. Van Swygenhoven: Appl. Phys. Lett. 85 (2004) 5863–5865. 15) A. G. Fro¨seth, H. Van Swygenhoven and P. M. Derlet: Acta Mater. 52 (2004) 2259–2268. 16) J. Sort, A. Zhilyaev, M. Zielinska, J. Nogue´s, S. Surinãch, J. Thibault and M. D. Bar: Acta Mater. 51 (2003) 6385–6393. 17) Y. Nakamoto, M. Yuasa, Y. Chen, H. Kusuda and M. Mabuchi: Scr. Mater. 58 (2008) 731–734. 18) M. Yuasa, H. Nakano, Y. Nakamoto, M. Hakamada and M. Mabuchi:

578


Mater. Trans. 50 (2009) 419–422. 19) B. Z. Cui, K. Han, Y. Xin, D. R. Waryoba and A. L. Mbaruku: Acta Mater. 55 (2007) 4429–4438. 20) M. Hakamada, Y. Nakamoto, H. Matsumoto, H. Iwasaki, Y. Chen, H. Kusuda and M. Mabuchi: Mater. Sci. Eng. A 457 (2007) 120–126. 21) M. Qasmi and P. Delobelle: Surf. Coat. Technol. 201 (2006) 1191– 1199. 22) M. Dao, N. Chollacoop, K. J. Van Vliet, T. A. Venkatesh and S. Suresh: Acta Mater. 49 (2001) 3899–3918. 23) S. Guicciardi and G. Pezzotti: J. Ceram. Soc. 115 (2007) 186–190. 24) M. Yuasa, H. Nakano and M. Mabuchi: unpublished data. 25) C. Wen, B. Huang, Z. Chen and Y. Rong: Mater. Sci. Eng. A 438–440 (2006) 420–426. 26) B. Y. C. Wu, P. J. Ferreira and C. A. Schuh: Metall. Mater. Trans. A 36 (2005) 1927–1936. 27) A. A. Karimpoor, U. Erb, K. T. Aust and G. Palubbo: Scr. Mater. 49 (2003) 651–656. 28) R. Schwaiger, B. Moser, M. Dao, N. Chollacoop and S. Suresh: Acta Mater. 51 (2003) 5159–5172.

29) J. Chen, Y. N. Shi and K. Lu: J. Mater. Res. 20 (2005) 2955–2959. 30) Y. J. Li, J. Mueller, H. W. Ho¨ppel, M. Go¨ken and W. Blum: Acta Mater. 55 (2007) 5708–5717. 31) M. Dao, L. Lu, Y. F. Shen and S. Suresh: Acta Mater. 54 (2006) 5421– 5432. 32) K. Konopka, J. Mizera and J. W. Wyrzykowski: J. Mater. Proc. Technol. 99 (2000) 255–259. 33) D. P. Field, B. W. True, T. M. Lillo and J. E. Flinn: Mater. Sci. Eng. A 372 (2004) 173–179. 34) E. Ma, Y. M. Wang, Q. H. Lu, M. L. Sui, L. Lu and K. Lu: Appl. Phys. Lett. 85 (2004) 4932–4934. 35) X. Zhang, A. Misra, H. Wang, M. Nastasi, J. D. Embury, T. E. Mitchell, R. G. Hoagland and J. P. Hirth: Appl. Phys. Lett. 84 (2004) 1096–1098. 36) X. Zhang, A. Misra, H. Wang, T. D. Shen, M. Nastasi, T. E. Mitchell, J. P. Hirth, R. G. Hoagland and J. D. Embury: Acta Mater. 52 (2004) 995–1002. 37) R. J. Asaro and Y. Kulkarni: Scr. Mater. 58 (2008) 389–392. 38) H. V. Swygenhoven, P. M. Derlet and A. G. Frøseth: Nat. Mater. 3 (2004) 399–403.