• 沒有找到結果。

Effects of Ag buffer layer on the microstructure and magnetic properties of nanocomposite FePt/Ag multilayer films

N/A
N/A
Protected

Academic year: 2021

Share "Effects of Ag buffer layer on the microstructure and magnetic properties of nanocomposite FePt/Ag multilayer films"

Copied!
3
0
0

加載中.... (立即查看全文)

全文

(1)

Effects of Ag buffer layer on the microstructure and magnetic properties

of nanocomposite FePt/ Ag multilayer films

S. C. Chena兲

Department of Mechanical Engineering, De Lin Institute of Technology, Taipei 236, Taiwan and Center for Nanostorage Research, National Taiwan University, No.1, Sec. 4, Roosevelt Road, Taipei 106, Taiwan

P. C. Kuo, C. Y. Chou, and A. C. Sun

Institute of Materials Science and Engineering and Center for Nanostorage Research, National Taiwan University, Taipei 106, Taiwan

共Presented on 9 November 2004; published online 16 May 2005兲

The face-centered-tetragonal L10FePt films with 共001兲 preferred orientation has perpendicular

coercivity 共Hc兲 of about 2462 Oe that can be achieved by stacking a structure of 共FePt 4 nm/Ag 2 nm兲5 multilayer films on the 5 nm thick MgO underlayer and annealing at

600 ° C for 30 min. It is found that both the perpendicular anisotropy and coercivity of 共FePt 4 nm/Ag 2 nm兲5 multilayer films are enhanced by introducing an Ag buffer layer 共⬉20 nm兲 between the 共FePt 4 nm/Ag 2 nm兲5 films and the MgO underlayer. When introducing

an Ag buffer layer of 20 nm thickness, the Hc value of the MgO 5 nm/ Ag 20 nm/ 共FePt 4 nm/Ag 2 nm兲5multilayer films is observed to increase from 2462 to 4731 Oe, which has

significant potential as perpendicular magnetic recording media for high-density recording. © 2005

American Institute of Physics.关DOI: 10.1063/1.1850386兴

I. INTRODUCTION

Due to the large magnetocrystalline anisotropy共Ku⬃7 ⫻107erg/ cm31

of the ordered L10FePt phase, a small

face-centered-tetragonal 共fct兲 FePt grain about 3 nm can over-come thermal instability that results in superparamagnetism.2 On the other hand, perpendicular magnetic recording media has a narrow transition region between recording bits,3which leads to a higher recording density than that of longitudinal recording. Therefore, the L10FePt film has attracted much

attention to be applied as perpendicular magnetic recording media for ultra high-density recording.4

Because the highest atom density plane 共close-packed plane兲 of the FePt alloy is the 共111兲 plane, the L10FePt film normally has 共111兲 preferred orientation, which means the easy axis 关001兴 is tilted 35° away from the film plane. In order to obtain共001兲 preferred orientation, i.e., the easy axis 关001兴 is perpendicular to the film plane, FePt films grown on single crystal MgO 共100兲 substrate has been used.5,6 How-ever, large lattice misfit at the interface between the MgO 共100兲 and FePt 共001兲 planes is about 9%, the introduction of an Au buffer layer between the MgO共100兲 and FePt 共001兲 planes is observed to decrease the lattice misfit, leading to the enhancement of the perpendicular anisotropy of FePt layer.7

An Ag underlayer is found not only to induce epitaxial growth of FePt films but also to decrease the FePt ordering temperature.8 Furthermore, the Ag is immiscible with either Fe or Pt. Instead, it tends to segregate at a grain boundary of FePt,9,10and increases the grain boundary energy, which can change the preferred orientation of FePt film.11In this study, the effects of an Ag buffer layer on the microstructures,

mag-netic properties, and orientation of the FePt easy axis of the nanocomposite MgO /共FePt/Ag兲5multilayer films are

inves-tigated.

II. EXPERIMENT

In order to obtain maximum intensity of the MgO共200兲 peak, the MgO underlayer of 5 nm thickness is deposited onto naturally oxidized Si 共100兲 substrates by rf magnetron sputtering at ambient temperature under an 共Ar+N2兲

pres-sure of 10 mTorr.12The flow rate ratio of N2to Ar is 2:5. An

Ag buffer layer with thickness of 0 – 30 nm and 共FePt 4 nm/Ag 2 nm兲5multilayer films are deposited

subse-quently by dc magnetron sputtering onto the MgO under-layer. The as-deposited films are annealed at 600 ° C for 30 min in vacuum better than 5⫻10−7Torr. The composi-tion of the FePt film determined by x-ray energy disperse spectrum共EDS兲 is Fe50.2Pt49.8.

The microstructures of the film are investigated by a Philips Tecnai F30 field emission gun transmission electron microscopy共TEM兲 and by x-ray diffractometer 共XRD兲 with Cu-K radiation. Compositions of the films are determined by EDS. The magnetic properties of the films are measured using a vibrating sample magnetometer.

III. RESULTS AND DISCUSSION

Figure 1 shows the x-ray diffraction patterns of various MgO 5 nm/ Ag t nm/共FePt 4 nm/Ag 2 nm兲5 multilayer

films 共where t=0–30 nm兲 after annealing at 600 °C for 30 min. It is found that the共001兲 and 共002兲 peaks of the FePt films are enhanced by introducing an Ag buffer layer 共⬉20 nm兲 between the 共FePt 4 nm/Ag 2 nm兲5films and the

MgO underlayer. This implies that introducing an Ag buffer

a兲Electronic mail: sscchh@ms28.hinet.net

JOURNAL OF APPLIED PHYSICS 97, 10N107共2005兲

0021-8979/2005/97共10兲/10N107/3/$22.50 97, 10N107-1 © 2005 American Institute of Physics

(2)

layer共⬉20 nm兲 is beneficial for increasing the perpendicular anisotropy of the FePt films. However, as the Ag buffer layer is increased to 30 nm, the fct FePt 共111兲 and 共200兲 peaks appeared, indicating the preferred orientation of the FePt films is changed from 共001兲 plane to random. The worse 共001兲FePttexture when the Ag layer increases to 30 nm may

be attributed to the epitaxial growth of the共200兲Agplane on

the共200兲MgOplane which cannot keep as the Ag layer

thick-ness is larger than 20 nm. This results in the共001兲FePtplane

not stacking well on the共200兲Agplane.

Figure 2 shows the high-resolution TEM cross-sectional lattice image of Si/ MgO 5 nm/共FePt 4 nm/Ag 2 nm兲5 multilayer films after annealing at 600 ° C for 30 min. The enclosed area clearly shows a misfit dislocation at the MgO / FePt interface due to a large lattice misfit 共⬃9%兲 at the interface. Therefore, the interface between the FePt

mag-netic layer and the MgO underlayer is a semicoherent inter-face, which has low strain energy. However, the lattice misfit will be decreased to about 4.5% as introducing an Ag buffer layer between the FePt magnetic layer and the MgO under-layer, and a coherent interface associated with high strain energy is obtained at the interface between the FePt magnetic layer and the Ag buffer layer, as shown in the enclosed area of Fig. 3.

Figure 4 shows the variations of the in-plane coercivity 共Hc储兲 and perpendicular coercivity 共Hc兲 with Ag buffer

layer thickness of various MgO 5 nm/ Ag t nm/ 共FePt 4 nm/Ag 2 nm兲5multilayer films共where t=0–30 nm兲

after annealing at 600 ° C for 30 min. It is found that both Hc储 and Hc of the film increase with increasing the

thick-ness of the Ag buffer layer. For the MgO 5 nm/ 共FePt 4 nm/Ag 2 nm兲5 multilayer films without an Ag

buffer layer, the Hc储and Hcare 2190 and 2460 Oe,

respec-tively. When introducing an Ag buffer layer of 20 nm

thick-FIG. 1. The x-ray diffraction patterns of various MgO 5 nm/ Ag t nm/共FePt 4 nm/Ag 2 nm兲5 multilayer films 共where t = 0 – 30 nm兲 after annealing at 600 °C for 30 min.

FIG. 2. The high-resolution cross-sectional lattice image of Si/ MgO 5 nm/共FePt 4 nm/Ag 2 nm兲5 multilayer films after annealing at 600 ° C for 30 min.

FIG. 3. The high-resolution cross-sectional lattice image of Si/ MgO 5 nm/ Ag 20 nm/共FePt 4 nm/Ag 2 nm兲5multilayer films after an-nealing at 600 ° C for 30 min.

FIG. 4. Variations of the Hc储and Hcwith an Ag buffer layer thickness of

various MgO 5 nm/ Ag t nm/共FePt 4 nm/Ag 2 nm兲5 multilayer films 共where t=0–30 nm兲 after annealing at 600 °C for 30 min.

10N107-2 Chenet al. J. Appl. Phys. 97, 10N107共2005兲

(3)

ness, the Hc储 and Hc⬜are observed to increase to 2960 and 4730 Oe, respectively. When the thickness of the Ag buffer layer is further increased to 30 nm, although the Hc储 and

Hc both increase to 5420 and 5300 Oe, respectively, the Hcvalue is almost the same as the Hc储. This indicates that

both the perpendicular anisotropy and coercivity of the FePt films are enhanced by introducing an Ag buffer layer 共⬉20 nm兲 between the 共FePt 4 nm/Ag 2 nm兲5films and the

MgO underlayer. However, the preferred orientation of the FePt films will be changed from the共001兲 plane to random as introducing an Ag buffer layer of 30 nm thickness. This is consistent with the XRD observed共see Fig. 1兲.

The gain in Hcand Hcof the FePt films expands with increasing the thickness of the Ag buffer layer may be as-cribed to the following reasons:共1兲 In the film with the Ag buffer layer, a coherent interface associated with high strain energy is obtained at the interface between the FePt magnetic layer and the Ag buffer layer. During annealing, the higher strain energy is stored in the film, and decreases the energy barrier of phase transformation of FePt from fcc disordered phase to fct ordered phase, resulting in an increase in the coercivity; 共2兲 introduction of the Ag buffer layer will pre-vent the interdiffusion of the MgO underlayer and 共FePt 4 nm/Ag 2 nm兲5 multilayer films; and 共3兲 the Ag is

immiscible with either Fe or Pt. Instead, Ag tends to segre-gate at the grain boundary of FePt, and increases the grain boundary energy, which results in the enhancement of the coercivity and changes the preferred orientation of the FePt film.

Figure 5 shows the variations of the saturation magneti-zation共Ms兲 and perpendicular squareness 共S兲 with the Ag buffer layer thickness of various MgO 5 nm/ Ag t nm/ 共FePt 4 nm/Ag 2 nm兲5multilayer films共where t=0–30 nm兲

after annealing at 600 ° C for 30 min. The results show that the Ms value of the film decreases with increasing the thickness of the Ag buffer layer. For

MgO 5 nm/共FePt 4 nm/Ag 2 nm兲5multilayer films without the Ag buffer layer, the Ms values is 800 emu/ cm3, and it

will decrease to 620 emu/ cm3 as the Ag buffer layer of

20 nm thickness is introduced. The decrease in magnetiza-tion of the FePt layer with an increasing Ag buffer layer thickness is mainly due to the increase of ordered L10FePt phase content in the film13 and the dilution of no-magnetic element Ag. On the other hand, the Svalue is in the range of 0.7–0.8, when introducing an Ag buffer layer thickness less than 20 nm, where Sis defined as the ratio of perpen-dicular remnant magnetization共Mr兲 to Ms. However, as the Ag buffer layer thickness increases to 30 nm, the Svalue is decreased dramatically to about 0.33. The result is also con-sistent with the observation of XRD 共see Fig. 1兲 and the coercivity measurement共see Fig. 4兲.

IV. CONCLUSION

Both the perpendicular anisotropy and coercivity of FePt films are enhanced by introducing an Ag buffer layer 共⬉20 nm兲 between the 共FePt 4 nm/Ag 2 nm兲5films and the

MgO underlayer. When introducing a 20 nm thick Ag buffer layer, the Hc value of the MgO 5 nm/ Ag 20 nm/ 共FePt 4 nm/Ag 2 nm兲5 multilayer films can be increased

from 2462 to 4731 Oe, which reveals that it is a promising candidate for high-density perpendicular magnetic recording media application.

ACKNOWLEDGMENTS

This work was supported by the National Science Coun-cil and Ministry of Economic Affairs of Taiwan through Grant Nos. NSC 92-2216-E-002-020 and 93-EC-17-A-08-S1-0006, respectively.

1

S. C. Chen, P. C. Kuo, A. C. Sun, C. T. Lie, and C. C. Chiang, IEEE Trans. Magn. 39, 584共2003兲.

2

D. Weller, A. Moser, L. Folks, M. E. Best, W. Lee, M. F. Toney, M. Schwickert, J.-U. Thiele, and M. F. Doerner, IEEE Trans. Magn. 36, 10 共2000兲.

3

T. Suzuki, IEEE Trans. Magn. 20, 675共1984兲. 4

K. Kang, Z. Zhang, and T. Suzuki, IEEE Trans. Magn. 39, 2317共2003兲. 5

M. H. Hong, K. Hono, and M. Watanabe, J. Appl. Phys. 84, 4403共1998兲. 6

J.-U. Thiele, L. Folks, M. F. Toney, and D. K. Weller, J. Appl. Phys. 84, 5686共1998兲.

7

T. Seki, T. Shima, K. Takanashi, Y. Takahashi, E. Matsubara, Y. K. Taka-hashi, and K. Hono, J. Appl. Phys. 96, 1127共2004兲.

8

Y. N. Hsu, S. Jeong, D. E. Laughlin, and D. N. Lambeth, J. Appl. Phys.

89, 7068共2001兲.

9

Z. L. Zhao, J. Ding, K. Inaba, J. S. Chen, and J. P. Wang, Appl. Phys. Lett.

83, 2196共2003兲.

10

S. S. Kang, D. E. Nikles, and J. W. Harrell, J. Appl. Phys. 93, 7178 共2003兲.

11

Y. Z. Zhou, J. S. Chen, G. M. Chow, and J. P. Wang, J. Appl. Phys. 93, 7577共2003兲.

12

Y. Misaki, M. Mikawa, T. Ishiguro, and K. Hamasaki, J. Vac. Sci. Technol. A 15, 48共1997兲.

13

T. Katayama, T. Sugimoto, Y. Suzuki, M. Hashimoto, P. dc Haan, and J. C. Lodder, J. Magn. Magn. Mater. 104–107, 1002共1992兲.

FIG. 5. Variations of the Ms and Swith an Ag buffer layer thickness of various MgO 5 nm/ Ag t nm/共FePt 4 nm/Ag 2 nm兲5 multilayer films 共where t=0–30 nm兲 after annealing at 600 °C for 30 min.

10N107-3 Chenet al. J. Appl. Phys. 97, 10N107共2005兲

數據

Figure 2 shows the high-resolution TEM cross-sectional lattice image of Si/ MgO 5 nm/ 共FePt 4 nm/Ag 2 nm兲 5
Figure 5 shows the variations of the saturation magneti- magneti-zation 共Ms兲 and perpendicular squareness 共S ⬜ 兲 with the Ag buffer layer thickness of various MgO 5 nm/ Ag t nm/

參考文獻

相關文件

substance) is matter that has distinct properties and a composition that does not vary from sample

Recommendation 14: Subject to the availability of resources and the proposed parameters, we recommend that the Government should consider extending the Financial Assistance

Wang, Solving pseudomonotone variational inequalities and pseudocon- vex optimization problems using the projection neural network, IEEE Transactions on Neural Networks 17

Then, we tested the influence of θ for the rate of convergence of Algorithm 4.1, by using this algorithm with α = 15 and four different θ to solve a test ex- ample generated as

Particularly, combining the numerical results of the two papers, we may obtain such a conclusion that the merit function method based on ϕ p has a better a global convergence and

Then, it is easy to see that there are 9 problems for which the iterative numbers of the algorithm using ψ α,θ,p in the case of θ = 1 and p = 3 are less than the one of the

volume suppressed mass: (TeV) 2 /M P ∼ 10 −4 eV → mm range can be experimentally tested for any number of extra dimensions - Light U(1) gauge bosons: no derivative couplings. =>

Let T ⇤ be the temperature at which the GWs are produced from the cosmological phase transition. Without significant reheating, this temperature can be approximated by the