Scholarly article on topic 'Dynamic and atomic-scale understanding of the twin thickness effect on dislocation nucleation and propagation activities by in situ bending of Ni nanowires'

Dynamic and atomic-scale understanding of the twin thickness effect on dislocation nucleation and propagation activities by in situ bending of Ni nanowires Academic research paper on "Materials engineering"

CC BY
0
0
Share paper
Academic journal
Acta Materialia
OECD Field of science
Keywords
{" In situ " / "Atomic scale" / "Plastic deformation" / "Metallic nanowires" / Twin}

Abstract of research paper on Materials engineering, author of scientific article — Lihua Wang, Yan Lu, Deli Kong, Lirong Xiao, Xuechao Sha, et al.

Abstract Although their mechanical behavior has been extensively studied, the atomic-scale deformation mechanisms of metallic nanowires (NWs) with growth twins are not completely understood. Using our own atomic-scale and dynamic mechanical testing techniques, bending experiments were conducted on single-crystalline and twin-structural Ni NWs (D =∼40nm) using a high-resolution transmission electron microscope (HRTEM). Atomic-scale and time-resolved dislocation nucleation and propagation activities were captured in situ. A large number of in situ HRTEM observations indicated strong effects from the twin thickness (TT) on dislocation type and glide system. In thick twin lamella (TT >∼12nm) and single-crystalline NWs, the plasticity was controlled by full dislocation nucleation. For NWs with twin thicknesses of ∼9nm< TT <∼12nm, full and partial dislocation nucleation occurred from the free surface, and the dislocations glided on multiple systems and interacted with each other during plastic deformation. For NWs with twin thicknesses of ∼6nm< TT <∼9nm, partial dislocation nucleation from the free surface and the gliding of those dislocations on the plane that intersected the twin boundaries (TBs) were the dominant plasticity events. For the NWs with twin thicknesses of ∼1nm< TT <∼6nm, the plasticity was accommodated by a partial dislocation nucleation process and glide parallel to the TBs. When TT <∼1nm, TB migration and detwinning processes resulting from partial dislocation nucleation and glide adjacent to the TBs were frequently observed.

Academic research paper on topic "Dynamic and atomic-scale understanding of the twin thickness effect on dislocation nucleation and propagation activities by in situ bending of Ni nanowires"

CrossMark

Available online at www.sciencedirect.com

ScienceDirect

Acta Materialia 90 (2015) 194-203

www.elsevier.com/locate/actamat

Dynamic and atomic-scale understanding of the twin thickness effect on dislocation nucleation and propagation activities by in situ bending of Ni

nanowires

Lihua Wang,a'* Yan Lu,a Deli Kong,a Lirong Xiao,a Xuechao Sha,a Jialin Sun,b Ze Zhanga'c and

Xiaodong Hana'*

aInstitute of Microstructure and Property of Advanced Materials, Beijing Key Lab of Microstructure and Property of Advanced Materials, Beijing University of Technology, Beijing 100124, China bDepartment of Physics and Key Lab of Atomic and Molecular Nanoscience of Education Ministry, Tsinghua University,

Beijing 100084, China

cState Key Laboratory of Silicon Materials, Zhejiang University, Hangzhou 310008, China

Received 21 September 2014; revised 31 January 2015; accepted 1 February 2015 Available online 16 March 2015

Abstract—Although their mechanical behavior has been extensively studied, the atomic-scale deformation mechanisms of metallic nanowires (NWs) with growth twins are not completely understood. Using our own atomic-scale and dynamic mechanical testing techniques, bending experiments were conducted on single-crystalline and twin-structural Ni NWs (D = ~40 nm) using a high-resolution transmission electron microscope (HRTEM). Atomic-scale and time-resolved dislocation nucleation and propagation activities were captured in situ. A large number of in situ HRTEM observations indicated strong effects from the twin thickness (TT) on dislocation type and glide system. In thick twin lamella (TT > ~12 nm) and single-crystalline NWs, the plasticity was controlled by full dislocation nucleation. For NWs with twin thicknesses of ~9 nm < TT < ~12 nm, full and partial dislocation nucleation occurred from the free surface, and the dislocations glided on multiple systems and interacted with each other during plastic deformation. For NWs with twin thicknesses of ~6 nm < TT < ~9 nm, partial dislocation nucleation from the free surface and the gliding of those dislocations on the plane that intersected the twin boundaries (TBs) were the dominant plasticity events. For the NWs with twin thicknesses of ~1nm< TT < ~6nm, the plasticity was accommodated by a partial dislocation nucleation process and glide parallel to the TBs. When TT < ~1 nm, TB migration and detwinning processes resulting from partial dislocation nucleation and glide adjacent to the TBs were frequently observed.

© 2015 Acta Materialia Inc. Published by Elsevier Ltd. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/ 4.0/).

Keywords: In situ; Atomic scale; Plastic deformation; Metallic nanowires; Twin

1. Introduction

Twin-structured nanomaterials have attracted great interest because they exhibit ultrahigh strength and ductility [1-5]. For twin-structured nanocrystalline (NC) metals, the inverse Hall-Petch effect is suggested to occur with decreasing twin thickness, which limits the increase in the strength of these metals [2,3]. Twin-structured nanowires (NWs) are a new family of nanomaterials [6-12], which are expected to exhibit unusual mechanical properties relative to those of twin-free NWs. It has been demonstrated that twin-structured metallic NWs exhibit strong Hall-Petch effects with a decrease in twin thickness (TT) [6,8,13-15], with a nearly ideal strength achieved by decreasing the TT [6,13-15]. These findings indicate that

* Corresponding authors; e-mail addresses: wlh@bjut.edu.cn; xdhan@bjut.edu.cn

the strength of twin-structured metallic NWs can be significantly affected by the TT [16,17]. However, whether the TT has an effect on dislocation behavior and how twin-structured NWs accommodate plastic deformation remain unclear [6,17,18-29]. Recently, MD simulations of twin-structured NC materials have shown that there exists a transition in the deformation mechanism at a critical TT; at this point, the plastic deformation, governed by partial dislocations that intersect with TBs, switches to the nucle-ation and motion of partial dislocations parallel to the TBs [5]. However, it remains uncertain whether this transition can be directly extrapolated to twin-structured NWs in which grain boundaries are replaced by free surfaces. If so, what are the implications?

MD simulations have revealed that only partial dislocations are responsible for plastic deformation in twin-structured NW, with researchers claiming that twin boundary-surface (TB-S) intersections are favorable as dislocation nucleation sites during plastic deformation

http://dx.doi.org/10.1016/j.actamat.2015.02.002

1359-6462/© 2015 Acta Materialia Inc. Published by Elsevier Ltd.

This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

[6-8,13,14,16,17]. However, do full dislocations occur in such twin-structured NWs during plastic deformation? If so, how does the TT affect the full dislocation behavior? Compared to the free surfaces of these NWs, does the TB-S truly have priority in nucleating dislocations? Considering that most studies in this research area have been performed on metallic NWs with low stacking fault (SF) energy [6-11,13-25,30], whether these results are valid for NWs with high SF energies (such as Ni) remains an unresolved issue. Finally, previous studies have only focused on NWs with twins that are perpendicular to the wire axis [6-10,13-17]. Indeed, direct atomic-scale experimental observations of metallic NWs with growth twins that are parallel to the wire axis are currently lacking.

To address the foregoing questions, we present the results of a series of in situ bending tests performed on single-crystalline and twin-structured Ni NWs using highresolution transmission electron microscopy (HRTEM). We observed the atomic-scale and time-resolved dislocation dynamics of the Ni NWs, which revealed the plastic deformation mechanisms that occur in the NWs under bending

strain, following our previous studies [12]. Our experiments focused on NWs with (111) twins that were nearly parallel to the growth direction, i.e., the longitudinal direction of the NWs. The atomic-scale in situ images revealed a strong TT effect on the types of dislocations and glide systems. Although these dislocations occur on different glide systems, they all have edge components with the Burgers vector along the strain direction, which can contribute to large plastic strains.

2. Experimental section

Ni NWs were synthesized using an electrochemical deposition method that employs an anodic aluminum oxide template. The details of the experiment are described elsewhere [31]. The atomic-scale in situ bending experiments of the individual NWs were realized by using our recently developed method [32-38]. The NWs were scattered on a broken transmission electron microscopy (TEM) grid that was covered by colloidal thin films (CTFs). Next, the

Fig. 1. (a) TEM image showing the typical morphologies of single-crystalline Ni NWs. Selected area electron diffraction patterns (SAEDP) along the [110] direction are shown in the insets. (b) HRTEM image of a single-crystalline Ni nanowire. (c) TEM image showing the typical morphologies of twin-structured Ni NWs. (d) SAEDP captured along the [110] direction. (e) HRTEM image of a twin-structured Ni nanowire.

scattered NWs were bent or deformed in tension by the CTFs at low strain rates (~ 10~4/s). Using this method, the specimen could be tilted at large angles along two orthogonal directions (±20°) without a special specimen holder or mechanical tensile attachments. Thus, the bending and axial tensile deformations of the individual NWs were recorded in situ at the atomic scale. The experiments were conducted using a JEOL HRTEM with a field-emission gun (JEOL 2010 F) and a point resolution of 0.19 nm.

3. Results

The synthesized Ni NWs measured approximately 50 im in length and ~40 nm in diameter (d). The two types of NWs observed were single-crystalline NWs and NWs with growth twins. Fig. 1a shows the typical morphologies of the single-crystalline NWs. The corresponding selected electron diffraction pattern (SAEDP) captured along the [1 10] direction is shown as an inset. The growth direction of the NWs was generally along the [11 2] direction (as Fig. 1a shows). As shown in Fig. 1b, the atomic-scale image reveals that the single-crystalline NWs were well crystalline without growth dislocations or twins. Fig. 1c shows typical examples of twin-structured Ni NWs, which consist of parallel (1 1 1) twin lamellas. Fig. 1d presents the corresponding SAEDP, which was captured along the ^ 10] direction. This pattern corresponds to a typical twinned diffraction pattern and indicates that the growth direction of the twinned NWs was approximately along the [11 2] direction (as Fig. 1c shows). A large number of HRTEM images show growth twins with thicknesses of 1-14 nm.

Fig. 1e shows a typical atomic-scale image indicating that the twin-structured NWs were well crystalline. The TBs appear as perfectly flat interfaces without pre-existing dislocations.

Fig. 2 provides serial TEM images, which show the continuous bending process of a twin-structured Ni NW. According to the traditional formula ebent = r/(r + R)% [39], where R is the bending curvature and r is the radius of a NW (ebent is the largest bending strain), the maximum strain in the Ni NW increased from 1.9% (Fig. 2a) to approximately 14.5 % (Fig. 2d) and the strain rate was approximately 5.4 x 10~4/s. During bending, the dynamic atomic-scale structural evolution of the bent Ni NW and the related dislocation processes were recorded in situ and in real time. Fig. 2(e) shows a time- and position-resolved atomic-scale image of the bent Ni NW. The highly bent, arc-shaped (1 1 1) lattices and TBs are highlighted by a dotted red line. For the bent, strained NW, there was a neutral axis. The strain near the neutral axis was nearly zero. The regions above this neutral line, i.e., along the longitudinal direction of the NWs, sustained tensile strains, whereas the areas below this neutral axis line were subjected to com-pressive strains. In this case, our in situ observations focused on the region that sustained tensile strains. The electron beam dose was maintained at 5 x 1019 e cm 2 s \ which led to a local temperature that was only a few degrees warmer than room temperature. Such a small temperature increase should not stimulate dislocation events [32,35,40].

In twin lamellae with TT < ~ 1 nm, partial dislocation nucleation and glide on the plane adjacent to the TBs was frequently observed. Fig. 3 shows typical in situ

Fig. 2. (a-d) A series of low-magnification TEM images that show the bending process of a twin-structured Ni NW. The strain sustained on the Ni NW increased from 1.9 % to ~14.5. (e) Atomic-scale image of the bent Ni nanowire. The red dotted line indicates highly bent, arc-shaped (111) lattices and twin boundaries (TBs). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3. In situ observations of detwinning in a twin with a thickness of (TT)<~1 nm. (a)-(c) HRTEM images showing the detwinning process resulting from a bundle of adjacent partial dislocation nucleations and glide on the adjacent plane of the TBs.

HRTEM images (taken along the [I 1 0] direction) of the detwinning process of the thinner twins. In Fig. 3a, the highly bent, arc-shaped TBs are highlighted by a dotted white line. Upon further straining, the plasticity was accompanied by detwinning, which resulted from a bundle of adjacent partial dislocations that were nucleated from the TB-S intersections and glide on the plane adjacent to the TBs (Fig. 3b). With extensive bending, the detwinning process that led to the left side of the TBs vanished (Fig. 3c). In addition, SFs that resulted from partial dislocations were also observed. The glide of these dislocations is on the (1 1 1) plane parallel to the strain direction and their Burgers vectors possessed edge components along the longitude direction of the NWs (strain direction), which could contribute to large plastic strain.

Fig. 4 shows typical in situ HRTEM images of the TB migration process. As indicated by the arrows in Fig. 4a,

SFs that resulted from partial dislocation emission were frequently observed in twin lamellae with a TT < ~1 nm. These dislocations nucleated from the free surface and glided on the (1 1 1) plane parallel to the TBs. In addition, it was also observed that the partial dislocations that nucleated from the TB-S intersection, which glide on the plane adjacent to the TBs, led to TB migration (Fig. 4b). Unlike the process shown in Fig. 3, the partial dislocations that were emitted layer-by-layer from the TB-S intersection led to TB migration without causing the TBs to vanish. In this process, the free surface and the TB-S intersections acted as the dislocation source.

In twin lamellae with TT ~ 2.4 nm, partial dislocation nucleation from the free surface and glide on the plane parallel to the TBs were the dominant plastic events. Fig. 5a and b shows typical in situ HRTEM images of partial dislocation emissions. In Fig. 5a, no partial dislocations

Fig. 4. In situ observations of the migration of TBs in a twin with TT < ~1 nm. (a, b) HRTEM images showing the partial dislocations that were emitted from the TB-S intersection layer-by-layer and that led to TB migration.

Fig. 5. (a, b) In situ HRTEM images showing partial dislocation nucleation and glide parallel to the TBs in a twin with TT = 2.4 nm. (c) Another image showing partial dislocations (as indicated by the arrows) gliding parallel to the TBs in a twin with TT = 2.6 nm.

can be observed in the 2.4 nm thick twin. As shown in Fig. 5b, three partial dislocations nucleated from the free surface and glided on the plane parallel to the TBs

(indicated by the arrows) as the bending strain increased. Fig. 5c shows another typical example of the partial dislocation nucleation and glide on the plane parallel to the

Fig. 6. In situ observations of partial dislocation nucleation in a twin with TT = 4.2 nm. (a) No partial dislocations were observed at low strain. (b) Two stacking faults (as indicated by arrows) resulting from partial emission and glide parallel to the twin boundary.

TBs. As indicated by the arrows, a high density of SFs resulted from the partial dislocations. Partial dislocations emitted from the TB-S intersections were rarely observed when TT > ~ 1 nm.

Fig. 6 shows typical in situ HRTEM images of the partial dislocation behavior of a twin measuring ~4 nm in thickness. In Fig. 6a, no partial dislocations were detected over the range of elastic strain. As the bending strain increased to exceed the elasticity limits, two partial dislocations nucleated from the free surface and glided on the plane parallel to the TBs (Fig. 6b). Fig. 7 shows another typical in situ HRTEM image of partial dislocation emission. As indicated by the arrows in Fig. 7b, partial dislocation nucleation occurred from the free surface and glided on the plane parallel to the TBs. Dislocations intersecting with twin planes were rarely observed.

For a TT of approximately 6.5 nm, partial dislocations intersecting with TBs were frequently observed in the

plastically deformed NWs. Fig. 8 shows a typical HRTEM image that was captured after plastic deformation. In this case, partial dislocation nucleation and intersection with TBs were directly observed. This dislocation behavior is frequently observed in twins with TT > ~6nm. Often, partial dislocations on different {111} slip planes form SF cross-structures. As shown in Fig. 9a, the partial dislocation density in the twin was low. As the strain increased, partial dislocations both intersecting and parallel to the TBs were observed. As shown in Fig. 9b, partial dislocations nucleated on different {1 1 1} planes to form SF cross-structures. Although these dislocations glided on the plane inclined toward the strain direction, the edge component of their Burgers vectors was along the strain direction, which could contribute to large plastic strain.

For TT ~ 9 nm, both partial and full dislocations were observed. Fig. 10a-c shows dislocation dynamics involving

Fig. 9. Partial dislocations intersecting the TBs of a twin with 7T~6.5 nm. (a, b) HRTEM image captured at different stages of the straining process, and partial dislocation nucleation on different {111} planes observed in situ.

%mï a

ife •

• 1 ' \V.

k WWM'i i i i ;llU)ll|)l!Hlii|i|! i»';

' *"vHilMt)ll] ;

Fig. 10. Partial and full dislocations in a twin with 7T~9 nm. (a) HRTEM image captured at low strain. (b, c) In situ image of full dislocations (marked with "T"), nucleation and motion during the bending process.

the nucleation of both partial and full dislocations from the surfaces of NWs. As indicated by the arrows in Fig. 10b and c, dislocations intersecting the TBs were frequently observed. In addition, perfect 60° dislocations with a Burgers vector of 1/2 (110) (indicated by "T" in Fig. 10b and c) were observed to nucleate and glide. For example, as indicated in Fig. 10b, full dislocation nucle-ation occurred. This nucleated full dislocation escaped during the deformation process (Fig. 10c), and many other full dislocations (marked with "7") were blocked by the SFs. The partial dislocation emission caused the SFs to act as obstacles against dislocation motion even under large stress [41]. This process led to high-density dislocation accumulation in the NWs, which played an important role in accommodating plastic deformation.

As the twin thickness increased to ~12nm, the partial-dislocation-controlled plasticity switched to full-dislocation-controlled deformation. Fig. 11 shows an HRTEM image of a twin with a thickness of ~12nm for which a high density of full dislocations was observed. As indicated by "T", full dislocations with Burgers vectors of b = 1/2[011] were the dominant plastic events. Partial dislocation (as indicated by arrows) played a much smaller part in accommodating plastic deformation.

In addition to the above-described MD simulation predictions in which the TB-S intersection was identified as the most favorable dislocation nucleation site [68,10,42,43], several of our in situ TEM images showed that the free surface of the NWs could act as an important dislocation source. In addition, a strong twin thickness effect

Fig. 11. High density of full dislocations (marked with "T") with a Burgers vector of b = 1/2[011] observed in a twin with TT~12 nm.

on the types of dislocations and glide systems that arose was also observed. Fig. 12 presents a schematic view illustrating the twin thickness effects on the dislocation glide systems. In the thinner twin lamellae (TT< ~6 nm), the plasticity is accommodated by partial dislocation nucle-ation and glide parallel to the TBs (Fig. 12a). In the thicker twin lamellae (TT> ~6 nm), partial or full dislocation nucleation from the free surface and glide on the multiple glide systems is shown to have occurred (Fig. 12b).

Finally, single-crystalline Ni NWs (D = 40 nm) subjected to large bending strain were investigated by HRTEM. Fig. 13 shows an HRTEM image captured from the red framed region of the inset. A high-density of full dislocation accumulation was observed in the NWs (the NWs were defect-free before they reached their plasticity limits). The full dislocations with a Burgers vector of b = 1/2[011] (marked with "T") were the dominant plastic events. Few partial dislocations were observed (as marked by arrows); thus, full dislocations played a dominant role in accommodating plastic deformation in the

Fig. 13. HRTEM image captured from the red framed region of the inset. Full dislocations (marked with "T") with a Burgers vector of b = 1/2[0 1 1] were the dominant plastic events in the single-crystalline NWs. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

single-crystalline NWs. Fig. 14 provides another HRTEM image that shows the high-density of full dislocation accumulation in the NWs, in which partial dislocation was rarely observed.

4. Discussion

In our experiments, the NW with growth twins were nearly parallel to the nanowire axis. The dislocations that

Fig. 12. Schematic view of dislocations that were parallel and intersected TBs.

glided parallel to or inclined to the TBs could be significantly affected by the twin thicknesses. These dislocations' Burgers vectors possessed an edge component along the strain direction, which could contribute large plastic strain without fracture nucleation. In this case, the switch from partial dislocation glides that were parallel to the TBs to partial dislocation glides that intersected the TBs could be understood by the repulsive force that the TBs exerted on the dislocations [42,43]. The critical resolved shear stress for dislocation emissions is

Sc = S0 + Sjb (1)

Here, s0 is the critical resolved shear stress for single-crystalline NWs, with s0 ~ 1 GPa for a NiNW with a diameter of ~40 nm [44]. In addition, sTB is the shear stress of the TBs that acts on the dislocations and can be written as

, ib sin 0 f b /8A1

Stb = k4P1b(T^{1 + 2r lnU)j (2)

Here, k is approximately 0.6 [45], the shear modulus i is 93.2 GPa for Ni [46], 0 is the angle between the TB and the {111} slip plane, v is Poisso n ' s ratio, b is the Burgers vector of dislocations and r = VDb cos 0. In this case, partial dislocations for which the glide planes are parallel to the TBs, sTB=0, and the critical resolved shear stress for partial dislocation emission is sC =~ 1GPa.

For TT =2nm, tib ~ 520MPa for partial dislocations that intersect TBs and sC ~ 1.52GPa. This stress is much higher than that of dislocations that are parallel to TBs and makes nucleation more difficult for partial dislocations that intersect TBs. In addition, partial dislocation parallel to the TBs is preferred. For TT = 6.5 nm and tjb ~ 160MPa, sC ~ 1.16 GPa. In this case, the stress is relatively low and is comparable with that of the dislocations that are parallel to the TBs. Thus, partial dislocations that intersect the TBs can be observed.

Our TEM observations revealed distinct trends indicating that only partial dislocations were observed in thin twin

lamellae with a thickness of less than 6.5 nm and full dislocations were preferred in thick twin lamellae and single-crystal NWs. This transition is similar to the NC case in which there was a switch from full dislocations to partial dislocations as the grain size decreased below a critical value [47-51]. For Ni NWs with an SF energy between ~0.128 and 0.24 J/m [52,53] and a shear modulus of 93.2 GPa[46], the critical grain size ranged from ~11 to 22 nm. When considering the twin thickness as the grain size, it is reasonable that only partial dislocations were observed when the TT was less than 6.5 nm. In this case, both partial and full dislocations appeared in twins measuring ~9 nm in thickness, and full dislocations were the dominant plastic events in the thicker twins (~12 nm) and single-crystalline NWs (D = 40 nm).

5. Conclusions

In summary, more than eight Ni NW bending processes were investigated in situ at the atomic scale. Our in situ HRTEM observations indicate that the free surface was an important dislocation source for the experimentally synthesized nanowires. In addition, TB-S intersections were not prioritized in dislocation nucleation. The NWs with growth twins that were parallel to the wire axis could sustain large plastic strain. A large number of HRTEM observations showed that there was a transition in the dislocation nucleation and glide systems that occurred at a certain twin thickness. At this point, the plasticity that was controlled by partial dislocation pile-up and cutting through twin planes switched to partial dislocation glide parallel to TBs. In addition, a transition in dislocation type occurred at a critical twin thickness. At this point, the partial-dislocation-controlled plasticity switched to full-dislocation-controlled deformation. The results of this study demonstrate the possibility of fabricating nanostructures with a desirable combination of ultrahigh strength and high ductility by optimizing the twin structure.

Acknowledgments

This work was supported by the Key Project of C-NSF (50831001), the NSF (11234011, 11127404, 10102001201304, 11102001201001) and the Beijing Municipal Natural Science Foundation (1112004), the Beijing Nova Program, the Beijing PXM201101420409000053 and Beijing 211 Project. Specialized Research Fund for the Doctoral Program of Higher Education of China (3C102001201301). The Project of Construction of Innovative Teams and Teacher Career Development for Universities and Colleges Under Beijing Municipality (IDHT20140504).

References

[1] L. Lu, Y.F. Shen, X.H. Chen, L.H. Qian, K. Lu, Science 304

(2004) 422.

[2] L. Lu, X. Chen, X. Huang, K. Lu, Science 323 (2009) 607.

[3] Y.T. Zhu, X.Z. Liao, X.L. Wu, Prog. Mater. Sci. 57 (2012) 1.

[4] X.Z. Liao, F. Zhou, E.J. Lavernia, D.W. He, Y.T. Zhu, Appl. Phys. Lett. 83 (2003) 5062.

[5] X.Y. Li, Y.J. Wei, L. Lu, K. Lu, H.J. Gao, Nature 464 (2010) 878.

[6] B. Wu, A. Heidelberg, J.J. Boland, J.E. Sader, X.M. Sun, Y.D. Li, Nano Lett. 6 (2006) 468.

[7] A.J. Cao, Y.G. Wei, Phys. Rev. B 74 (2006) 214108.

[8] A.J. Cao, Y.G. Wei, S.X. Mao, Appl. Phys. Lett. 90 (2007) 151909.

[9] D. Jang, X. Li, H. Gao, J.R. Greer, Nat. Nanotechnol. 7 (2012) 594.

[10] C. Deng, F. Sansoz, Nano Lett. 9 (2009) 1517.

[11] T. Zhu, J. Li, A. Samanta, H.G. Kim, S. Suresh, Proc. Natl. Acad. Sci. U.S.A. 104 (2007) 3031.

[12] L.H. Wang, P. Liu, P.F. Guan, M.J. Yang, J.L. Sun, Y.Q. Cheng, et al., Nat. Commun. 4 (2013) 2413.

[13] K.A. Afanasyev, F. Sansoz, Nano Lett. 7 (2007) 2056.

[14] C. Deng, F. Sansoz, ACS Nano 3 (2009) 3001.

[15] J.W. Wang, F. Sansoz, J.Y. Huang, Y. Liu, S.H. Sun, Z. Zhang, S.X. Mao, Nat. Commun. 4 (2013) 1742.

[16] C. Deng, F. Sansoz, Acta Mater. 57 (2009) 6090.

[17] L.H. Wang, X.D. Han, Z. Zhang, NPG Asia Mater. 5 (2013) e40.

[18] Y. Zhu, H.D. Espinosa, Proc. Natl. Acad. Sci. U.S.A. 102

(2005) 14503.

[19] Y. Lu, Y. Ganesan, J. Lou, Exp. Mech. 50 (2010) 47.

[20] Y. Ganesan, Y. Lu, C. Peng, H. Lu, R. Ballarini, J. Lou, J. Microelectromech. Syst. 19 (2010) 675.

[21] D.F. Zhang, J.M. Breguet, R. Clavel, L. Philippe, I. Utke, J. Michler, Nanotechnology 20 (2009) 365706.

[22] H. Zheng, A. Cao, C.R. Weinberger, J.Y. Huang, K. Du, J.B. Wang, Y.Y. Ma, Y.N. Xia, S.X. Mao, Nat. Commun. 1 (2010) 144.

[23] B. Wu, A. Heidelberg, J.J. Boland, Nat. Mater. 4 (2005) 525.

[24] J.H. Seo, Y.D. Yoo, N.Y. Park, S.W. Yoon, H. Lee, S. Han, S.W. Lee, et al., Nano Lett. 11 (2011) 3499.

[25] Y. Lu, J. Song, J.Y. Huang, J. Lou, Adv. Funct. Mater. 21 (2011) 3982.

[26] E.W. Wong, P.E. Sheehan, C.M. Lieber, Science 277 (1997) 1971.

[27] M.F. Yu, O. Lourie, M.J. Dyer, K. Moloni, T.F. Kelly, R.S. Ruoff, Science 287 (2000) 637.

[28] P. Poncharal, Z.L. Wang, D. Ugarte, W.A. De Heer, Science 283 (1999) 1513.

[29] R.P. Gao, Z.L. Wang, Z.G. Bai, W.A. De Heer, L.M. Dai, M. Ga, Phys. Rev. Lett. 85 (2000) 622.

[30] N. Agrait, J.G. Rodrigo, S. Vieira, Phys. Rev. B 47 (1993) 12345.

[31] X.C. Zhao, J.L. Sun, J.L. Zhu, Appl. Phys. Lett. 93 (2008) 152107.

[32] K. Zheng, C.C. Wang, Y.Q. Cheng, Y.H. Yue, X.D. Han, Z. Zhang, et al., Nat. Commun. 1 (2010) 1.

[33] K. Zheng, X.D. Han, L.H. Wang, Y.F. Zhang, Y.H. Yue, Y. Qin, X.N. Zhang, Z. Zhang, Nano Lett. 9 (2009) 2471.

[34] X.D. Han, Y.F. Zhang, K. Zheng, X.N. Zhang, Z. Zhang, Nano Lett. 7 (2007) 452.

[35] L.H. Wang, K. Zheng, Z. Zhang, X.D. Han, Nano Lett. 11 (2011) 2382.

[36] L.H. Wang, X.D. Han, Y.F. Zhang, K. Zheng, P. Liu, Z. Zhang, Acta Mater. 59 (2011) 651.

[37] X.D. Han, K. Zheng, Y.F. Zhang, X.N. Zhang, Z. Zhang, Z.L. Wang, Adv. Mater. 19 (2007) 2112.

[38] Y.F. Zhang, X.D. Han, K. Zheng, Z. Zhang, X.N. Zhang, J.Y. Fu, Y. Ji, Y.J. Hao, X.Y. Guo, Z.L. Wang, Adv. Func. Mater. 17 (2007) 3435.

[39] L.D. Landau, E.M. Lifshitz, Theory of Elasticity, Pergamon Press, New York, 1986.

[40] T. Zhu, J. Li, A. Samanta, A. Leach, K. Gall, Phys. Rev. Lett. 100 (2008) 025502.

[41] S.J. Zhou, D.L. Preston, P.S. Lomdahl, D.M. Beazley, Science 279 (1998) 1525.

[42] C. Denga, F. Sansoz, Scripta Mater. 63 (2010) 50.

[43] C. Deng, F. Sansoz, Appl. Phys. Lett. 95 (2009) 091914.

[44] C. Peng, Y. Ganesan, Y. Lu, J. Lou, J. Appl. Phys. 111 (2012) 063524.

[45] Z.M. Chen, Z.H. Jin, Phys. Rev. B 75 (2007) 212104.

[46] D.H. Warner, W.A. Curtin, S. Qu, Nat. Mater. 6 (2007) 876.

[47] L.H. Wang, X.D. Han, P. Liu, Y.H. Yue, Z. Zhang, E. Ma, Phys. Rev. Lett. 105 (2010) 135501.

[48] M.W. Chen, E. Ma, K.J. Hemker, H.W. Sheng, Y.M. Wang, X.M. Cheng, Science 300 (2003) 1275.

[49] Y.H. Yue, P. Liu, Q.S. Deng, E. Ma, Z. Zhang, X.D. Han, Nano Lett. 12 (2012) 4045.

[50] L.H. Wang, Z. Zhang, E. Ma, X.D. Han, Appl. Phys. Lett. 98 (2011) 051905.

[51] L.H. Wang, J. Teng, P. Liu, A. Hirata, E. Ma, Z. Zhang, M.W. Chen, X.D. Han, Nat. Commun. 5 (2014) 4402.

[52] L.E. Murr, Interfacial Phenomena in Metals and Alloys, Addison Wesley, Reading, MA, 1975.

[53] P.S. Dobson, P.J. Goodhew, R.E. Smallman, Philos. Mag. 16 (1967) 9.