Effect of thermal annealing on nanoimprinted Cu–Ni alloys using molecular dynamics simulation

(1)

Effect of thermal annealing on nanoimprinted Cu–Ni alloys using molecular

dynamics simulation

Te-Hua Fang

a

, Cheng-Da Wu

b

, Win-Jin Chang

c,

*

, Sung-Shui Chi

a

Institute of Mechanical and Electromechanical Engineering, National Formosa University, Yunlin 632, Taiwan

b_{Department of Mechanical Engineering, National Cheng Kung University, Tainan 701, Taiwan} c_{Department of Mechanical Engineering, Kun Shan University, Tainan 710, Taiwan}

1. Introduction

Nanoimprinting lithography (NIL) is a revolutionary approach and can provide a simple operation, high replication, and lower cost technique through molding and printing processes under proper temperature and pressure conditions[1–3]. Over the past decade, it has been developed for various applications such as nano-ﬂuidic devices, nano-optical devices, nano-electronic devices, and high-density data storage systems[4–6].

There have been many experimental investigations on the NIL in order to study the structural deformation of specimen[7–9]. On the other hand, using molecular dynamics (MD) simulation is also a useful tool for understanding the deformation mechanisms of nanoimprinting process and can provide preliminary information for the design of tooling and determination of processing conditions [10–12]. Therefore, molecular dynamics simulation has a growing and important role on the study of nanoimprinting process. Recently, the plastic behaviors of nanoimprinting thin ﬁlm was achieved by considering the effects of temperature and velocity using molecular dynamics simulation based on Lennard-Jones[13]and Morse potential[14]. Furthermore, Pei et al.[15]

utilized the embedded atom method to study the materials

deformation, dislocation movement and imprint forces in the nanoimprinting process of copper with a diamond mold.

In nanoimprinting, the residual stress is induced in specimen, and it is inevitable. The stress is not stable and can result in deformation of the material in use. Therefore, the study of how to reduce the residual stress is important and worth in nanoimprint-ing process. However, only a limited portion of the literature is concerned with the aspect. In this paper, the effects of thermal annealing on the residual stress of imprinted Cu–Ni alloys are studied through molecular dynamics simulation with tight-binding (TB) potential. The Cu–Ni alloys are adopted in this simulation because they were widely used for industry applica-tions due to their excellent resistance to corrosion, high inherent resistance to biofouling and good fabricability.

2. Methodology

The nanoimprinting process between a tungsten punch and a Cu 80%–Ni 20% thin ﬁlm at an isothermal state of 300 K was studied using molecular dynamics computational simulation as shown inFig. 1 [14]. The punch is assumed as ideally rigid and the pattern of the punch can be pressed into the specimen due to the impression of the punch onto the specimen. A Cartesian coordinate system was used in the proposed system and the original point was set at the center of the lower left atom of the specimen. There are a total of 6102 tungsten atoms with a width and height in a punch Applied Surface Science 255 (2009) 6043–6047

A R T I C L E I N F O Article history:

Received 18 December 2008

Received in revised form 21 January 2009 Accepted 21 January 2009

Available online 31 January 2009 Keywords: Thermal annealing Nanoimprint Molecular dynamics Residual stress Cu–Ni alloys A B S T R A C T

The mechanical behaviors of nanoimprinted Cu–Ni alloys before and after annealing were studied using molecular dynamics simulation with a tight-binding potential. The results showed that when the punch is advancing, the punching force obtained from the simulation with a tight-binding potential is lower than with the Morse potential. During and after withdrawing the punch from the specimen, the adhesive phenomena are observed and the large residual stress in the Cu–Ni alloys is induced. During the annealing process, the internal energy of Cu–Ni alloys decreased with increasing the temperature and the component of Cu. In addition, comparing the maximum residual stress in the Cu–Ni alloys with and without annealing treatment, the stress is signiﬁcantly released after annealing, especially in the higher component of Ni.

* Corresponding author.

E-mail address:[email protected](W.-J. Chang).

Contents lists available atScienceDirect

Applied Surface Science

j o u r n a l h o m e p a g e : w w w . e l s e v i e r . c o m / l o c a t e / a p s u s c

(2)

Cu–Ni alloys decrease with increasing time steps. The reason is that a small energy is required for plastic deformation to occur when the material temperature is higher. It can also be seen that the energy remains constant for 100% Cu since the temperature of specimen is kept at 300 K. In addition, the energies of alloys increase with increasing Ni component. This is because the potential energy of Ni is larger than that of Cu[19].

Fig. 6(a and b) shows the atomic conﬁgurations of the plastic pattern of the pure copper specimen in xy-plane before and after annealing process, respectively. The lower stress exhibited far away from the imprinted contact region. A high stress and affected-zone occurred around the surface region and plastic

groove can be identiﬁed inFig. 6(a). After the annealing process the whole stress decreased and more uniform distribution than that before annealing. The slightly wavelike proﬁles of the stress in the specimen occurred due to a higher degree of the elastic recovery and structural relaxation. The elastic behavior would dominate the imprinted structure and deformation characteristic around the relaxation region after annealing. The maximum residual stress for different Cu–Ni alloys before and after thermal annealing is shown inFig. 7. The residual stress is observed to be correlated with the components of alloys before annealing. The stress in the Cu–Ni alloy is very high, especially in a higher component of Ni. This is because the yield strength of Ni is higher than that of Cu. In addition, the maximum residual stress is obviously released as a result of microstructural change after annealing process. 4. Conclusions

In this article, the effects of the thermal annealing on the nanoimprinting process of the Cu–Ni alloy were studied using molecular dynamics simulation with a tight-binding potential model. The following results were obtained:

(1) During the nanoimprinting process, the punching force obtained in the simulation with a tight-binding potential is lower than with the Morse potential.

(2) The adhesive phenomena were observed during withdrawing the punch from the specimen, and the large residual stress in the Cu–Ni alloys was induced after unloading.

(3) During the annealing process, the internal energy of Cu–Ni alloys decreased with increasing the temperature and the component of Cu.

(4) The maximum residual stress in the Cu–Ni alloys was obviously released after annealing process, especially in the higher component of Ni.

Acknowledgement

This work was partially supported by the National Science Council of Taiwan, under grant no. NSC 96-2628-E-150-005-MY3. References

[1] Z.W. Liu, Q.H. Wei, X. Zhang, Nano Lett. 5 (2005) 957.

[2] J.S. Gwag, J. Fukuda, M. Yoneya, H. Yokoyama, Appl. Phys. Lett 91 (2007) 073504. [3] J.Y. Kim, D.G. Choi, J.H. Jeong, E.S. Lee, Appl. Surf. Sci. 254 (2008) 4793. Fig. 6. Atomic conﬁgurations of nanoimprinted pattern (a) before and (b) after

annealing. Stress unit: GPa.

Fig. 7. The relationship between the maximum residual stress and different Cu–Ni alloys before and after thermal annealing.

Fig. 5. The relationship between of internal energy of specimen and time steps for different Cu–Ni alloys during the annealing process for different Cu–Ni alloys.

T.-H. Fang et al. / Applied Surface Science 255 (2009) 6043–6047 6046

(3)

[4] S. Zankovych, I. Maximov, I. Shorubalko, J. Seekamp, M. Beck, S. Romanov, D. Reuter, P. Schafmeister, A.D. Wieck, J. Ahopelto, C.M.S. Torres, L. Montelius, Microelectron. Eng. 67–68 (2003) 214.

[5] A. Han, N.F. de Rooij, U. Staufer, Nanotechnology 17 (2006) 2498. [6] B.D. Terris, T. Thomson, G. Hu, Microsyst. Technol. 13 (2007) 189. [7] H.C. Scheer, H. Schulz, Microelectron. Eng. 56 (2001) 311. [8] I. Park, J. Cheng, A.P. Pisano, Appl. Phys. Lett. 90 (2007) 093902.

[9] W. Zhou, J. Zhang, Y. Liu, X. Li, X. Niu, Z. Song, G. Min, Y. Wan, L. Shi, S. Feng, Appl. Surf. Sci. 255 (2008) 2885.

[10] W.J. Chang, T.H. Fang, J. Phys. Chem. Solids 64 (2003) 1279.

[11] D.L. Chen, T.C. Chen, Nanotechnology 16 (2005) 2972. [12] T.H. Fang, W.J. Chang, S.L. Lin, Appl. Surf. Sci. 253 (2006) 1649. [13] Q.C. Hsu, C.D. Wu, T.H. Fang, Comput. Mater. Sci. 34 (2005) 314. [14] T.H. Fang, C.D. Wu, W.J. Chang, Appl. Surf. Sci. 253 (2007) 6963. [15] Q.X. Pei, C. Lu, Z.S. Liu, K.Y. Lam, J. Phys. D: Appl. Phys. 40 (2007) 4928. [16] T.H. Fang, W.J. Chang, C.I. Weng, Mater. Sci. Eng. A 430 (2006) 332. [17] F. Cleri, V. Rosato, Phys. Rev. B 48 (1993) 22.

[18] M.A. Karolewski, Radiat. Eff. Defects Solids 153 (2001) 239.

[19] K. Giesen, F. Hage, F.J. Himpsel, H.J. Riess, W. Stinmann, Phys. Rev. B 35 (1987) 971.