- Email: [email protected]
X trilayer
Available online at www.sciencedirect.com
Physics Procedia 16 (2011) 36–41
The 9th Perpendicular Magnetic Recording Conference
Effect of third element for FePt ordered alloy thin films with perpendicular magnetic anisotropy fabricated from Fe/Pt/X trilayer Yuji Ogataa, Yasuharu Imaia and Shigeki Nakagawaa,* Department of Physical Electronics, Tokyo Institute of Technology,2-12-1 O-okayama, Meguro-ku, Tokyo 152-8552, Japan
Abstract L10-FePt ordered alloy thin films with (001) preferential orientation were fabricated by annealing Fe/Pt/X (X: Cu and Au) trilayers to reduce Curie temperature and ordering temperature. FePtCu film exhibited Curie temperature of around 280 °C and steep change of perpendicular coercivity to a measurement temperature |dHc/dT| at around 200 °C. In addition, process temperature decreased to 500 °C by adding Au atom to FePt film and perpendicular coercivity was as high as 5.8 kOe even though annealing temperature was as low as 500 °C. © 2011 Published by Elsevier B.V. Selection and/or peer-review under responsibility of the Organising Committee of the 9th Perpendicular Magnetic Recording © 2010 Published by Elsevier B.V. Conference Pacs: 75.70.-i; 75.50.Ss Keywords: FePt; third element; trilayer; perpendicular magnetic anisotropy; heat assisted magnetic recording
1. Introduction L10-FePt ordered alloy thin film is one of the promising materials for future magnetic recording media, especially, heat assisted magnetic recording (HAMR) media because of its high uniaxial magnetocrystalline anisotropy (7.0×107 erg/cc) [1]. Previously, we successfully fabricated L10-FePt thin films with high perpendicular magnetic anisotropy by annealing Fe(3 nm)/Pt(3 nm) films under hydrogen atmosphere without any seed layer [2]. To apply FePt to HAMR media, the Curie temperature should be below 300 °C and perpendicular coercivity is required to decrease sufficiently by heating. Additionally, drastic decline of coercivity around 200 °C is also required for HAMR media to assure recorded information. Because the Curie temperature of the FePt film is very high, it is an effective way to add third element such as Ag, B and Cu to FePt films [3]-[5]. Previously, FePtCu ordered alloy thin films was fabricated from PtCu/Fe films and Curie temperature was controllable by adjusting Cu ratio of PtCu alloy [6]. However, their magnetic anisotropy was lower and their c-axis dispersion of the FePt crystallites was larger than that of FePt film. In this study, we have fabricated FePtCu ordered alloy thin films from Cu/Pt/Fe trilayers to improve magnetic property and control Cu ratio of FePtCu film more easily. In our previous paper, annealing temperature of 550 °C is required to attain FePt (001) diffraction peak [2] and such high temperature processes may enhance grain growth which will induce strong exchange interaction among
* Corresponding author. Tel.: +81-03-5734-3564; fax: +81-03-5734-2513. E-mail address: [email protected]
1875-3892 © 2011 Published by Elsevier B.V. Selection and/or peer-review under responsibility of the Organising Committee of the 9th Perpendicular Magnetic Recording Conference doi:10.1016/j.phpro.2011.06.104
37
Yuji Ogata et al. / Physics Procedia 16 (2011) 36–41
the grains [7]. Several methods have been attempted to reduce process temperature, multilayer configuration of Fe and Pt thin films [8], doping of third element such as Au [9] [10]. In this study, FePtX (X: Cu and Au) films were fabricated by annealing Fe/Pt/X trilayers in order to attain magnetization characteristics which are suitable for HAMR media as well as to reduce process temperature for ordering FePt ordered alloy thin films. 2. Experimental Facing target sputtering (FTS) system was used to prepare Fe/Pt bilayer. Each Fe bottom layer was deposited directly on crystallized glass substrate (OHARA: TSCZ) at room temperature under Ar atmosphere. Background pressure and Ar gas pressure were below 3.0×10-7 Torr and 0.4 mTorr, respectively. All the deposition rates of Fe, Pt, Cu and Au were about 0.5 nm/min. All the films were annealed for 120 minutes under hydrogen atmosphere to remove internal oxygen and enhance the inter-diffusion among the thin layers. Annealing temperature was varied from 500 to 600 °C. Crystallographic structure was evaluated by X-ray diffractometry (XRD). M-H characteristics were measured by vibrating sample magnetometer (VSM) at the maximum field of 20 kOe. Composition distributions were evaluated by using energy dispersive spectroscopy (EDS) of scanning transmission electron microscope (STEM).
(a)
(b)
(c)
Pt 3nm
Cu 0.5nm Pt 3nm
Au 0.5nm Pt 3nm
Fe 3nm
Fe 3nm
Fe 3nm
glass sub.
glass sub.
glass sub.
Fig. 1. Schematic images of (a) Fe/Pt bilayer, (b) Fe/Pt/Cu trilayer and (c) Fe/Pt/Au trilayer.
FePt (002)
FePt (111)
FePt (001)
Intensity (a.u.)
3. Results and Discussion Fig. 1 shows schematic configurations of films (a) Fe(3 nm)/Pt(3 nm) bilayer, (b) Fe(3 nm)/Pt(3 nm)/Cu(0.5 nm) trilayer and (c) Fe(3 nm)/Pt(3 nm)/Au(0.5 nm) trilayer prepared in this study. Fig. 2 shows XRD diagrams of Fe/Pt, Fe/Pt/Cu and Fe/Pt/Au films after annealed under hydrogen atmosphere at 600 °C. The diffraction lines corresponding to the TSCZ substrates were eliminated from the XRD diagrams. All films showed FePt(001) diffraction peak which implied formation of L10-FePt ordered phase with c-axis orientation perpendicular to the film plane. Particularly, the FePt(001) diffraction peak of FePtCu and FePtAu films exhibited larger diffraction intensity than that of FePt film. Fig.3 shows FePt(001) peak position and full width at half maximum (FWHM) of FePt(001) diffraction line. The FePt(001) peak of FePtCu was shifted to higher angle than that of FePt. This result suggests that Fe atom is substituted by Cu atom which has smaller atomic radius than Fe atom. On the other hand, the FePt(001) diffraction peak of FePtAu film was shifted to lower angle than that of FePt. Since Au atom has low solubility with both Fe and Pt atoms and doesn’t substitute for both atoms [11], the addition of Au to FePt layer may release internal stress within the film and (001) peak may shift to lower angle. Additionally, the FWHM of FePtAu is much smaller than those of FePtCu and FePt, which corresponding to increase of the crystallite diameter. Fig. 4 shows M-H curves of Fe/Pt bilayer, Fe/Pt/Cu trilayer and Fe/Pt/Au trilayer after annealed at 600 °C under hydrogen atmosphere. All films showed less in-plane hysteresis loss which means low dispersion of crystalline orientation. Saturation magnetization of FePtCu film was a little smaller than that of FePt film. This result was due to an effect of simple Fe dilution caused by substitution of Cu atom for Fe atom [12]. The large perpendicular coercivity of FePtAu film around 13.6 kOe was related to the effect of domain wall pinning effect caused by diffusion of Au atom to boundaries of FePt grains [13].
FePtAu
FePtCu FePt
20
30
40 2θ (deg.)
50
Fig. 2. XRD diagrams of Fe/Pt/Au, Fe/Pt/Cu and Fe/Pt films after annealed at 600 ºC.
38
Yuji Ogata et al. / Physics Procedia 16 (2011) 36–41
1
FWHM (deg.)
Bulk value (23.966)
24
23.5
(b) 0.91 0.69
0.5
0.36
0
FePt
FePtCu
FePt
FePtCu
FePtAu
Fig. 3. (a) FePt(001) peak position and (b) Full width at half maximum of Fe/Pt/Au, Fe/Pt/Cu and Fe/Pt films after annealed at 600 ºC. 20 20
20
4πM (kG)
FePtAu
(a)
10 0
//
10
(b)
0
4πM (kG)
23
4πM (kG)
FePt(001) peak (deg.)
(a) 24.5
//
-10
-10 -20 -20 -10 0 10 Hext (kOe)
20
-20 -20 -10 0 10 Hext (kOe)
20
10 0
(c)
//
-10 -20 -20 -10 0 10 Hext (kOe)
20
Fig. 4. M-H curves of (a) Fe/Pt, (b) Fe/Pt/Cu and (c) Fe/Pt/Au films after annealed at 600 ºC.
900 450 0
1000
500 550 600 Annealing temperature (°C)
FePt(002)
1350
(b) FePt FePtCu FePtAu
FePt(111)
1800
Intensity ( a.u )
Integrated intensity of FePt(001) peak (a.u.)
(a)
FePt(001)
Fig. 5 shows the change of integrated diffraction intensity of FePt(001) as a function of annealing temperature and XRD diagram of FePtAu after annealed at 500 °C. The integrated intensities of FePtCu film were higher than those of FePt at 600 °C, but they decreased at less than 550 °C. On the other hand, FePtAu film showed higher integrated intensities than those of other films at this temperature range. From XRD diagram of FePtAu after annealed at 500 °C, the large diffraction intensities of corresponding to (001) and (002) diffraction lines of L10-FePt alloy were observed even though the annealing temperature was as low as 500 °C. In other words, ordering temperature of FePt decreased to 500 °C by addition of Au atom. Fig. 6 shows M-H curves of Fe/Pt bilayer, Fe/Pt/Cu trilayer and Fe/Pt/Au trilayer after annealed at relatively low temperature as low as 500 °C under hydrogen atmosphere. FePtAu film exhibited larger coercivity around 5.8 kOe than that of FePt film and high squareness ratio in perpendicular magnetization loop and high anisotropy field observed in the in-plane loop. These results denote that the addition of Au as a third element is effective way to attain c-axis oriented FePtAu film at lower process temperature.
Sub. Film + Sub.
500
20
30
40 2(deg.)
Fig. 5. Diffraction intensity of FePt(001) as a function of annealing temperature and XRD diagram of FePtAu after annealed at 500 ºC.
50
39
Yuji Ogata et al. / Physics Procedia 16 (2011) 36–41
0 -10
20
//
-20 -20 -10 0 10 Hext (kOe)
20
10
(b)
20
4πM (kG)
10
(a)
4πM (kG)
4πM (kG)
20
0 -10
//
-20 -20 -10 0 10 Hext (kOe)
10
(c)
0 // -10 -20 -20 -10 0 10 Hext (kOe)
20
20
Fig. 6. M-H curves of (a) Fe/Pt, (b) Fe/Pt/Cu and (c) Fe/Pt/Au films after annealed at 500 ºC. Fig. 7 shows dark filed TEM images of FePtAu films after annealed at 600 °C and 500 °C. Cross-sectional image of the film after annealed at 600 °C became island like structure which was caused by Au addition and film thickness increased anomalously, but that of after annealed at 500 °C was very flat and smooth. Fig. 8 shows composition distribution along the film thickness direction of FePtAu after annealed at 600 °C and 500 °C. Although Fe and Pt atoms distributed homogeneously along the thickness direction of the film, Au atom diffused to the top and the bottom regions of the film after annealed at 600 °C. On the other hand, Fe, Pt, Au atoms distributed homogeneously in the film after annealed at 500 °C.Fig. 9 shows EDS profile along the film thickness direction of FePtCu film after annealed at 600 °C. Since the Cu signal is much smaller than those of Fe and Pt, it is expanded in 5th times to compare its profile to them. It is observed that Fe, Pt and Cu atom are diffused monotonically throughout the film thickness direction.
(a)
(b) Film
Film
Sub.
Sub.
Fig. 7. TEM images of FePtAu film after annealed (a) at 600 ºC and (b) at 500 ºC.
(a)
Fe
(b)
Pt
Count (a.u.)
Count (a.u.)
Pt
Si
Fe Si Au
Au Distance (nm) Film
Distance (nm) Sub.
Film
Sub.
Fig. 8. EDS analysis along thickness direction of FePtAu film after annealed (a) at 600 ºC and (b) at 500 ºC.
40
Yuji Ogata et al. / Physics Procedia 16 (2011) 36–41
Cu Pt
0.5 nm
Fe
3 nm
3 nm
Glass sub.
Sub.
Film
Fig. 9. EDS analysis along thickness direction of FePtCu film after annealed at 600 ºC.
(a)
8 4 0 0
FePt FePtCu
100 200 300 Temperature (°C)
8
Hc (kOe)
4πMs (kG)
12
(b)
6 4 2 0
FePt FePtCu
0
100 200 300 Temperature (°C)
|dHC/dT| (Oe/°C)
Fig. 10(a) shows the change of saturation magnetization of the FePtCu film as a function of measurement temperature from room temperature to 300 °C. Spontaneous magnetization of the FePtCu film disappeared at lower temperature than that of FePt film. That is to say, Curie temperature of FePt decreased to about 280 °C by adding Cu atom because Cu addition weakened the exchange coupling of FePt ordered alloy. Fig. 10(b) shows the change of perpendicular coercivity of the FePtCu film as a function of measurement temperature. Perpendicular coercivity of FePtCu film decreased to around 1 kOe at 200 °C though that of FePt film was in excess of 2 kOe. As shown in Fig. 10(c), both films indicated different gradient of perpendicular coercivity to a measurement temperature |dHc/dT|, which was caused by deterioration of exchange coupling by addition of Cu atom.
40
(c)
30 20 10 0 0
FePt FePtCu
100 200 300 Temperature (°C)
Fig. 10. Measurement temperature dependence of (a) the saturation magnetization, (b) perpendicular coercivity and (c) the gradient of perpendicular coercivity to temperature of FePt and FePtCu films. 4. Conclusion FePtX films with (001) preferential orientation were successfully fabricated by annealing Fe/Pt/X trilayers. It was confirmed that the Curie temperature of FePtCu was decreased to about 280 ºC, suitable for HAMR media. FePtAu which was annealed at 600 ºC showed perpendicular coercivity of remarkably high value around 13.6 kOe and (001) oriented L10-FePt film was formed even though annealing temperature was as low as 500 ºC. Fabrication of FePt from trilayer configuration hold enormous potentialities to attain (001) oriented FePt ordered alloy thin films which are suitable for next generation perpendicular magnetic recording media. Acknowledgment Authors would like to express appreciation to OHARA Inc. and TDK Corporation. A part of this work has been sponsored by Storage Research Consortium (SRC). References 1. T. Suzuki, N. Honda and K. Ouchi, J. Appl. Phys. 85 (1999) 4301-4303. 2. S. Nakagawa, and T. Kamiki, J. Magn. Magn. Mater. 287 (2005) 204-208. 3. T. Konagai, Y. Kitahara, T. Itoh, T. Kato, S. Iwata and S. Tsunashima, J. Magn. Magn. Mater. 310 (2007) 26622664.
Yuji Ogata et al. / Physics Procedia 16 (2011) 36–41
41
4. C. W. Chang, H. W. Chang, C. H. Chiu and W. C. Chang, J. Appl. Phys. 97 (2005) 10N117. 5. T. Maeda, A. Kikitsu, T. Kai, T. Nagase, H. Aikawa and J. Akiyama, IEEE Trans. Magn. 38 (2002) 2796-2798. 6. J. Ikemoto, Y. Imai and S. Nakagawa, IEEE Trans.Magn, 44(11) (2008) 3543-3546. 7. C. L. Platt, K. W. Wierman, E. B. Svedberg, R. van de Veerdonk. and J. K. Howard, J. Appl. Phys. 92 (2002) 10. 8. S. Imada, A. Yamasaki, S. Suga, T. Shima, and K. Takanashi, Appl. Phys. Lett. 90 (2007) 132507. 9. C. Feng, B. H. Li, Y. Liu, J. Teng, M. H. Li, Y. Jiang, G. H. Yu, J. Appl. Phys. 103 (2008) 023916. 10. C. Feng, Q. Zhan, B. Li, J. Teng, M. Li, Y. Jiang and G. Yu, Appl. Phys. Lett. 93 (2008) 152513. 11. F. T. Yuan, S. K. Chen, W. C. Chang and Lance Horng, Appl. Phys. Lett. 85 (2004) 15. 12. M. L. Yan, Y. F. Xu and D. J. Sellmyer, J. APPL. PHYS. 99 (2006) 08G903. 13. F. T. Yuan, S. K. Chen, W. M. Liao, C. W. Hsu, S. N. Hsiao and W. C. Chang, J. Magn. Magn. Mater. 304 (2006) 109-111.
Physics Procedia 16 (2011) 36–41
The 9th Perpendicular Magnetic Recording Conference
Effect of third element for FePt ordered alloy thin films with perpendicular magnetic anisotropy fabricated from Fe/Pt/X trilayer Yuji Ogataa, Yasuharu Imaia and Shigeki Nakagawaa,* Department of Physical Electronics, Tokyo Institute of Technology,2-12-1 O-okayama, Meguro-ku, Tokyo 152-8552, Japan
Abstract L10-FePt ordered alloy thin films with (001) preferential orientation were fabricated by annealing Fe/Pt/X (X: Cu and Au) trilayers to reduce Curie temperature and ordering temperature. FePtCu film exhibited Curie temperature of around 280 °C and steep change of perpendicular coercivity to a measurement temperature |dHc/dT| at around 200 °C. In addition, process temperature decreased to 500 °C by adding Au atom to FePt film and perpendicular coercivity was as high as 5.8 kOe even though annealing temperature was as low as 500 °C. © 2011 Published by Elsevier B.V. Selection and/or peer-review under responsibility of the Organising Committee of the 9th Perpendicular Magnetic Recording © 2010 Published by Elsevier B.V. Conference Pacs: 75.70.-i; 75.50.Ss Keywords: FePt; third element; trilayer; perpendicular magnetic anisotropy; heat assisted magnetic recording
1. Introduction L10-FePt ordered alloy thin film is one of the promising materials for future magnetic recording media, especially, heat assisted magnetic recording (HAMR) media because of its high uniaxial magnetocrystalline anisotropy (7.0×107 erg/cc) [1]. Previously, we successfully fabricated L10-FePt thin films with high perpendicular magnetic anisotropy by annealing Fe(3 nm)/Pt(3 nm) films under hydrogen atmosphere without any seed layer [2]. To apply FePt to HAMR media, the Curie temperature should be below 300 °C and perpendicular coercivity is required to decrease sufficiently by heating. Additionally, drastic decline of coercivity around 200 °C is also required for HAMR media to assure recorded information. Because the Curie temperature of the FePt film is very high, it is an effective way to add third element such as Ag, B and Cu to FePt films [3]-[5]. Previously, FePtCu ordered alloy thin films was fabricated from PtCu/Fe films and Curie temperature was controllable by adjusting Cu ratio of PtCu alloy [6]. However, their magnetic anisotropy was lower and their c-axis dispersion of the FePt crystallites was larger than that of FePt film. In this study, we have fabricated FePtCu ordered alloy thin films from Cu/Pt/Fe trilayers to improve magnetic property and control Cu ratio of FePtCu film more easily. In our previous paper, annealing temperature of 550 °C is required to attain FePt (001) diffraction peak [2] and such high temperature processes may enhance grain growth which will induce strong exchange interaction among
* Corresponding author. Tel.: +81-03-5734-3564; fax: +81-03-5734-2513. E-mail address: [email protected]
1875-3892 © 2011 Published by Elsevier B.V. Selection and/or peer-review under responsibility of the Organising Committee of the 9th Perpendicular Magnetic Recording Conference doi:10.1016/j.phpro.2011.06.104
37
Yuji Ogata et al. / Physics Procedia 16 (2011) 36–41
the grains [7]. Several methods have been attempted to reduce process temperature, multilayer configuration of Fe and Pt thin films [8], doping of third element such as Au [9] [10]. In this study, FePtX (X: Cu and Au) films were fabricated by annealing Fe/Pt/X trilayers in order to attain magnetization characteristics which are suitable for HAMR media as well as to reduce process temperature for ordering FePt ordered alloy thin films. 2. Experimental Facing target sputtering (FTS) system was used to prepare Fe/Pt bilayer. Each Fe bottom layer was deposited directly on crystallized glass substrate (OHARA: TSCZ) at room temperature under Ar atmosphere. Background pressure and Ar gas pressure were below 3.0×10-7 Torr and 0.4 mTorr, respectively. All the deposition rates of Fe, Pt, Cu and Au were about 0.5 nm/min. All the films were annealed for 120 minutes under hydrogen atmosphere to remove internal oxygen and enhance the inter-diffusion among the thin layers. Annealing temperature was varied from 500 to 600 °C. Crystallographic structure was evaluated by X-ray diffractometry (XRD). M-H characteristics were measured by vibrating sample magnetometer (VSM) at the maximum field of 20 kOe. Composition distributions were evaluated by using energy dispersive spectroscopy (EDS) of scanning transmission electron microscope (STEM).
(a)
(b)
(c)
Pt 3nm
Cu 0.5nm Pt 3nm
Au 0.5nm Pt 3nm
Fe 3nm
Fe 3nm
Fe 3nm
glass sub.
glass sub.
glass sub.
Fig. 1. Schematic images of (a) Fe/Pt bilayer, (b) Fe/Pt/Cu trilayer and (c) Fe/Pt/Au trilayer.
FePt (002)
FePt (111)
FePt (001)
Intensity (a.u.)
3. Results and Discussion Fig. 1 shows schematic configurations of films (a) Fe(3 nm)/Pt(3 nm) bilayer, (b) Fe(3 nm)/Pt(3 nm)/Cu(0.5 nm) trilayer and (c) Fe(3 nm)/Pt(3 nm)/Au(0.5 nm) trilayer prepared in this study. Fig. 2 shows XRD diagrams of Fe/Pt, Fe/Pt/Cu and Fe/Pt/Au films after annealed under hydrogen atmosphere at 600 °C. The diffraction lines corresponding to the TSCZ substrates were eliminated from the XRD diagrams. All films showed FePt(001) diffraction peak which implied formation of L10-FePt ordered phase with c-axis orientation perpendicular to the film plane. Particularly, the FePt(001) diffraction peak of FePtCu and FePtAu films exhibited larger diffraction intensity than that of FePt film. Fig.3 shows FePt(001) peak position and full width at half maximum (FWHM) of FePt(001) diffraction line. The FePt(001) peak of FePtCu was shifted to higher angle than that of FePt. This result suggests that Fe atom is substituted by Cu atom which has smaller atomic radius than Fe atom. On the other hand, the FePt(001) diffraction peak of FePtAu film was shifted to lower angle than that of FePt. Since Au atom has low solubility with both Fe and Pt atoms and doesn’t substitute for both atoms [11], the addition of Au to FePt layer may release internal stress within the film and (001) peak may shift to lower angle. Additionally, the FWHM of FePtAu is much smaller than those of FePtCu and FePt, which corresponding to increase of the crystallite diameter. Fig. 4 shows M-H curves of Fe/Pt bilayer, Fe/Pt/Cu trilayer and Fe/Pt/Au trilayer after annealed at 600 °C under hydrogen atmosphere. All films showed less in-plane hysteresis loss which means low dispersion of crystalline orientation. Saturation magnetization of FePtCu film was a little smaller than that of FePt film. This result was due to an effect of simple Fe dilution caused by substitution of Cu atom for Fe atom [12]. The large perpendicular coercivity of FePtAu film around 13.6 kOe was related to the effect of domain wall pinning effect caused by diffusion of Au atom to boundaries of FePt grains [13].
FePtAu
FePtCu FePt
20
30
40 2θ (deg.)
50
Fig. 2. XRD diagrams of Fe/Pt/Au, Fe/Pt/Cu and Fe/Pt films after annealed at 600 ºC.
38
Yuji Ogata et al. / Physics Procedia 16 (2011) 36–41
1
FWHM (deg.)
Bulk value (23.966)
24
23.5
(b) 0.91 0.69
0.5
0.36
0
FePt
FePtCu
FePt
FePtCu
FePtAu
Fig. 3. (a) FePt(001) peak position and (b) Full width at half maximum of Fe/Pt/Au, Fe/Pt/Cu and Fe/Pt films after annealed at 600 ºC. 20 20
20
4πM (kG)
FePtAu
(a)
10 0
//
10
(b)
0
4πM (kG)
23
4πM (kG)
FePt(001) peak (deg.)
(a) 24.5
//
-10
-10 -20 -20 -10 0 10 Hext (kOe)
20
-20 -20 -10 0 10 Hext (kOe)
20
10 0
(c)
//
-10 -20 -20 -10 0 10 Hext (kOe)
20
Fig. 4. M-H curves of (a) Fe/Pt, (b) Fe/Pt/Cu and (c) Fe/Pt/Au films after annealed at 600 ºC.
900 450 0
1000
500 550 600 Annealing temperature (°C)
FePt(002)
1350
(b) FePt FePtCu FePtAu
FePt(111)
1800
Intensity ( a.u )
Integrated intensity of FePt(001) peak (a.u.)
(a)
FePt(001)
Fig. 5 shows the change of integrated diffraction intensity of FePt(001) as a function of annealing temperature and XRD diagram of FePtAu after annealed at 500 °C. The integrated intensities of FePtCu film were higher than those of FePt at 600 °C, but they decreased at less than 550 °C. On the other hand, FePtAu film showed higher integrated intensities than those of other films at this temperature range. From XRD diagram of FePtAu after annealed at 500 °C, the large diffraction intensities of corresponding to (001) and (002) diffraction lines of L10-FePt alloy were observed even though the annealing temperature was as low as 500 °C. In other words, ordering temperature of FePt decreased to 500 °C by addition of Au atom. Fig. 6 shows M-H curves of Fe/Pt bilayer, Fe/Pt/Cu trilayer and Fe/Pt/Au trilayer after annealed at relatively low temperature as low as 500 °C under hydrogen atmosphere. FePtAu film exhibited larger coercivity around 5.8 kOe than that of FePt film and high squareness ratio in perpendicular magnetization loop and high anisotropy field observed in the in-plane loop. These results denote that the addition of Au as a third element is effective way to attain c-axis oriented FePtAu film at lower process temperature.
Sub. Film + Sub.
500
20
30
40 2(deg.)
Fig. 5. Diffraction intensity of FePt(001) as a function of annealing temperature and XRD diagram of FePtAu after annealed at 500 ºC.
50
39
Yuji Ogata et al. / Physics Procedia 16 (2011) 36–41
0 -10
20
//
-20 -20 -10 0 10 Hext (kOe)
20
10
(b)
20
4πM (kG)
10
(a)
4πM (kG)
4πM (kG)
20
0 -10
//
-20 -20 -10 0 10 Hext (kOe)
10
(c)
0 // -10 -20 -20 -10 0 10 Hext (kOe)
20
20
Fig. 6. M-H curves of (a) Fe/Pt, (b) Fe/Pt/Cu and (c) Fe/Pt/Au films after annealed at 500 ºC. Fig. 7 shows dark filed TEM images of FePtAu films after annealed at 600 °C and 500 °C. Cross-sectional image of the film after annealed at 600 °C became island like structure which was caused by Au addition and film thickness increased anomalously, but that of after annealed at 500 °C was very flat and smooth. Fig. 8 shows composition distribution along the film thickness direction of FePtAu after annealed at 600 °C and 500 °C. Although Fe and Pt atoms distributed homogeneously along the thickness direction of the film, Au atom diffused to the top and the bottom regions of the film after annealed at 600 °C. On the other hand, Fe, Pt, Au atoms distributed homogeneously in the film after annealed at 500 °C.Fig. 9 shows EDS profile along the film thickness direction of FePtCu film after annealed at 600 °C. Since the Cu signal is much smaller than those of Fe and Pt, it is expanded in 5th times to compare its profile to them. It is observed that Fe, Pt and Cu atom are diffused monotonically throughout the film thickness direction.
(a)
(b) Film
Film
Sub.
Sub.
Fig. 7. TEM images of FePtAu film after annealed (a) at 600 ºC and (b) at 500 ºC.
(a)
Fe
(b)
Pt
Count (a.u.)
Count (a.u.)
Pt
Si
Fe Si Au
Au Distance (nm) Film
Distance (nm) Sub.
Film
Sub.
Fig. 8. EDS analysis along thickness direction of FePtAu film after annealed (a) at 600 ºC and (b) at 500 ºC.
40
Yuji Ogata et al. / Physics Procedia 16 (2011) 36–41
Cu Pt
0.5 nm
Fe
3 nm
3 nm
Glass sub.
Sub.
Film
Fig. 9. EDS analysis along thickness direction of FePtCu film after annealed at 600 ºC.
(a)
8 4 0 0
FePt FePtCu
100 200 300 Temperature (°C)
8
Hc (kOe)
4πMs (kG)
12
(b)
6 4 2 0
FePt FePtCu
0
100 200 300 Temperature (°C)
|dHC/dT| (Oe/°C)
Fig. 10(a) shows the change of saturation magnetization of the FePtCu film as a function of measurement temperature from room temperature to 300 °C. Spontaneous magnetization of the FePtCu film disappeared at lower temperature than that of FePt film. That is to say, Curie temperature of FePt decreased to about 280 °C by adding Cu atom because Cu addition weakened the exchange coupling of FePt ordered alloy. Fig. 10(b) shows the change of perpendicular coercivity of the FePtCu film as a function of measurement temperature. Perpendicular coercivity of FePtCu film decreased to around 1 kOe at 200 °C though that of FePt film was in excess of 2 kOe. As shown in Fig. 10(c), both films indicated different gradient of perpendicular coercivity to a measurement temperature |dHc/dT|, which was caused by deterioration of exchange coupling by addition of Cu atom.
40
(c)
30 20 10 0 0
FePt FePtCu
100 200 300 Temperature (°C)
Fig. 10. Measurement temperature dependence of (a) the saturation magnetization, (b) perpendicular coercivity and (c) the gradient of perpendicular coercivity to temperature of FePt and FePtCu films. 4. Conclusion FePtX films with (001) preferential orientation were successfully fabricated by annealing Fe/Pt/X trilayers. It was confirmed that the Curie temperature of FePtCu was decreased to about 280 ºC, suitable for HAMR media. FePtAu which was annealed at 600 ºC showed perpendicular coercivity of remarkably high value around 13.6 kOe and (001) oriented L10-FePt film was formed even though annealing temperature was as low as 500 ºC. Fabrication of FePt from trilayer configuration hold enormous potentialities to attain (001) oriented FePt ordered alloy thin films which are suitable for next generation perpendicular magnetic recording media. Acknowledgment Authors would like to express appreciation to OHARA Inc. and TDK Corporation. A part of this work has been sponsored by Storage Research Consortium (SRC). References 1. T. Suzuki, N. Honda and K. Ouchi, J. Appl. Phys. 85 (1999) 4301-4303. 2. S. Nakagawa, and T. Kamiki, J. Magn. Magn. Mater. 287 (2005) 204-208. 3. T. Konagai, Y. Kitahara, T. Itoh, T. Kato, S. Iwata and S. Tsunashima, J. Magn. Magn. Mater. 310 (2007) 26622664.
Yuji Ogata et al. / Physics Procedia 16 (2011) 36–41
41
4. C. W. Chang, H. W. Chang, C. H. Chiu and W. C. Chang, J. Appl. Phys. 97 (2005) 10N117. 5. T. Maeda, A. Kikitsu, T. Kai, T. Nagase, H. Aikawa and J. Akiyama, IEEE Trans. Magn. 38 (2002) 2796-2798. 6. J. Ikemoto, Y. Imai and S. Nakagawa, IEEE Trans.Magn, 44(11) (2008) 3543-3546. 7. C. L. Platt, K. W. Wierman, E. B. Svedberg, R. van de Veerdonk. and J. K. Howard, J. Appl. Phys. 92 (2002) 10. 8. S. Imada, A. Yamasaki, S. Suga, T. Shima, and K. Takanashi, Appl. Phys. Lett. 90 (2007) 132507. 9. C. Feng, B. H. Li, Y. Liu, J. Teng, M. H. Li, Y. Jiang, G. H. Yu, J. Appl. Phys. 103 (2008) 023916. 10. C. Feng, Q. Zhan, B. Li, J. Teng, M. Li, Y. Jiang and G. Yu, Appl. Phys. Lett. 93 (2008) 152513. 11. F. T. Yuan, S. K. Chen, W. C. Chang and Lance Horng, Appl. Phys. Lett. 85 (2004) 15. 12. M. L. Yan, Y. F. Xu and D. J. Sellmyer, J. APPL. PHYS. 99 (2006) 08G903. 13. F. T. Yuan, S. K. Chen, W. M. Liao, C. W. Hsu, S. N. Hsiao and W. C. Chang, J. Magn. Magn. Mater. 304 (2006) 109-111.