- Email: [email protected]
Investigation of FePt-SiO2-C granular film in columnar structure
Journal of Alloys and Compounds 788 (2019) 600e603
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: http://www.elsevier.com/locate/jalcom
Investigation of FePt-SiO2-C granular film in columnar structure L. Zhang a, *, C. Li b, Y.J. Zou a, H.L. Xu a, K. Liao a, X.L. Zhang a a b
College of Mechanical and Electrical Engineering, Pingxiang University, 211 Pinganbei Ave., Pingxiang, 337055, People's Republic of China Physical Science and Technology College, Yichun University, 576 Xuefu Rd., Yichun, 336000, People's Republic of China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 December 2018 Received in revised form 22 February 2019 Accepted 23 February 2019 Available online 25 February 2019
We worked on a FePt granular film with a columnar structure, for the purpose of ultrahigh density perpendicular magnetic recording media. In experiments, a Fe53Pt47-SiO230%-C25% (10 nm) film was deposited on a silicon substrate with a MgO (12 nm) interlayer at 510 C. The perpendicular coercivity of the film is 29 kOe, with a squareness of 1. Bright-field TEM images show that the FePt granular film has small and uniform grains of 8.3 ± 1.9 nm with a columnar structure. More work of high-resolution TEM imaging shows excellent L10 ordering inside the FePt grains. The measurement of remnant coercivity proves that it has an energy barrier of 290kBT at room temperature, meaning excellent thermal stability. We demonstrate FePt granular film to be a qualified candidate for high-density perpendicular magnetic recording media, and both SiO2 and C addition helps form columnar structure. © 2019 Elsevier B.V. All rights reserved.
Keywords: FePt granular film Columnar structure Magnetic recording media
1. Introduction In modern computers, magnetic recording method is extensively applied as the basic storage architecture. A typical case is hard-disk drive (HDD). In order to improve the HDD performance, an essential task in the research and development work is to increase the recording density, or reduce the dimension of a single storage unit. However, with the further increasing density, the super-paramagnetic effect [1] cannot be neglected any more, and it could induce thermal relaxation in each magnetic grain in the storage units. In order to get excellent thermal stability of recorded information for at least 5 years, the normal lift-time of a HDD, we should increase the coercivity and anisotropy of the recording materials, i.e. making the magnetic material harder and harder. Thus, people in magnetic recording industry work hard to search qualified magnetic materials for high-density recording purpose. After years' investigation, among all possible magnetic materials for recording media, L10-phase FePt granular thin films are almost the best candidates as perpendicular magnetic recording (PMR) media with the recording density of 1 Terabits/in2 or above, because it has a substantially high magneto-crystalline anisotropy (Ku ~ 6 107 erg/cc), which can assist maintain the thermal stability at room temperature for the small grain size of even below 4 nm [2e10]. On the other hand, hard recording medium is difficult in
* Corresponding author. E-mail address: [email protected] (L. Zhang). https://doi.org/10.1016/j.jallcom.2019.02.263 0925-8388/© 2019 Elsevier B.V. All rights reserved.
writing at room temperature because no pole material can create strong field. Nowadays this trouble could be solved by heating the recording medium during heat-assisted magnetic recording (HAMR) [2,11e14]. In early efforts of other groups, they achieved nanometer grain size FePt thin films with the addition of Ag and C with high coercivity and thermal stability [6]. In their work, carbon is used as the spacer to separate FePt nanograins, and Ag is used to reduce the L10 ordering temperature. However, early work on FePt granular films always shows poor squareness of the perpendicular MH loop due to the small grain size and also thickness of the film. In addition to carbon, some oxides can also act as spacer in FePt thin films. Chen et al. worked on FePt granular film by addition of TiO2 and Ta2O5 [8]. Those oxides can help form columnar structure, in which we believe that it could improve the switching behavior of nanograins. However, their work did not show convincing improvement on the magnetic properties. On the other hand, for the currently-applied CoCrPt thin film in magnetic recording media in HDD industries, SiO2 is a successful spacer to separate CoCrPt grains with excellent magnetic behavior [15]. However, the addition of SiO2 alone into FePt thin films does not improve both the microstructure and magnetic properties [16]. The combination of both oxides and carbon has been few reported. In this paper, we studied a FePt granular film with the addition both of SiO2 and C. Then, we displayed detailed work to study the microstructure of FePt grains through high-resolution transmission electron microscopy (TEM) imaging. Finally, we measured the remnant coercivity of the film in different time delay to study the thermal stability of this FePt granular film.
L. Zhang et al. / Journal of Alloys and Compounds 788 (2019) 600e603
601
2. Experiments We deposit a 10 nm thick Fe53Pt47-SiO230%-C25% film on a bare silicon substrate. At first, we grow a 12 nm MgO interlayer on the top of the silicon substrate, then, FePt-SiO2-C layer on the top of MgO interlayer, in a high-vacuum magnetron-sputtering chamber. The volume fraction of carbon is 25%, and that of SiO2 is 30% in the film. The MgO layer was grown at room temperature, while the FePt-SiO2-C layer at 510 C. During the film growth, the Ar pressure is 1.1 Pa for MgO, and 0.55 Pa for FePt-SiO2-C, respectively. The deposition rate of MgO is about 0.06 nm/s, and FePt-SiO2-C is about 0.025 nm/s. The composition of Fe, Pt, SiO2, and C is based on the growth rate of each single material. The total FePt-SiO2-C film was formed by co-sputtering method. In experiments, the base pressure of the vacuum chamber is about 2 106 Pa. After the thin film fabrication, we use XRD (x-ray diffractor) to study the texture of the FePt-SiO2-C film with MgO underlayer. We use SQUID (superconducting quantum interferometer device) MPMS (magnetic property measurement system) to measure the magnetic properties. The field range of SQUID is 55 ~ þ55 kOe. Then, we use a TEM to study the microstructure of the film, in two modes: bright-field and high-resolution. 3. Results and discussion In Fig. 1, we display the XRD pattern of the FePt-SiO2-C/MgO/Si film. The index of both MgO and FePt peaks are labeled, while FePt peaks are marked by numbers only. We observed a distinct MgO(200) peak at 2q ¼ 43.1 and the L10-ordered fct-FePt(001) (002) peaks, basically induced by the MgO(200) texture from the underlayer [6]. Both fcc-FePt(200) and FePt(111) peaks are suppressed. Based on the theoretical investigation by Yang et al., we calculated the degree of the L10 order of this FePt-SiO2-C granular film, to be 0.89 [17]. The above facts indicate excellent L10 order in the FePt-SiO2-C film layer. Next, we work on the microstructure of the FePt-SiO2-C film by TEM. Fig. 2(a) shows a bright-field TEM image of the plane-view for the film, with the insets of selected area electron diffraction (SAED) pattern and statistic diagram of the grain size distribution. From the image, we can see that the grains in the FePt-SiO2-C thin film are uniformly separated and distributed. After statistical calculation, we obtained the average grain size is 8.3 nm, with a standard deviation of 1.9 nm. In SAED pattern, counted from center to edge, the rings of FePt(001), FePt(110) MgO(200), and FePt(002) peaks are clearly displayed. It agrees with the data in Fig. 1 (XRD pattern). It
means that the excellent L10 ordering is formed in this FePt-SiO2-C granular film. Fig. 2(b) shows the bright-field TEM image of the cross-sectional view of the FePt-SiO2-C granular film. It is clear that those FePt grains form a single-grain layer columnar structure in the film, much improved compared with early FePt-C granular films [6]. After that, we did high-resolution TEM imaging on the FePtSiO2-C granular film, so that we can understand more details on the microstructure. Fig. 3 displays a plane-view high-resolution TEM image of the FePt grains. We observed some MgO matrix below the L10-ordered FePt nanograins. We also can clearly find the boundary of MgO grains in the underlayer. Next, Fig. 4 displays a cross
Fig. 1. XRD pattern of FePt-SiO2-C granular film.
Fig. 3. High-resolution plane-view TEM image of FePt-SiO2-C granular film.
Fig. 2. Bright-field TEM image of (a) the plane-view, with the inset of selected area electron diffraction pattern, and statistics of grain size distribution, (b) the crosssectional view of FePt-SiO2-C granular film.
602
L. Zhang et al. / Journal of Alloys and Compounds 788 (2019) 600e603
E f ¼ f0 exp b kB T
Fig. 4. High-resolution cross-sectional view of FePt-SiO2-C granular film.
sectional high-resolution TEM image of the FePt grains. It seems difficult to find the proof of direct relationship between the FePt grains and the MgO grains below. However, we are able to find the epitaxial relationship between FePt and MgO grains, i.e. (001)FePt// (002)MgO. It demonstrates that the MgO polycrystalline underlayer aids FePt grains aligned to (001) orientation normal to the film plane, getting the L10-order in the FePt granular film. In addition, the strong order of FePt L10-phase can be demonstrated through the measurement of magnetic properties. We use MPMS SQUID to work on this FePt-SiO2-C granular film at room temperature. In Fig. 5, we see both perpendicular and in-plane M-H loops of this film. From those two curves, this film has high perpendicular coercivity of 29 kOe, with perfect squareness of 1 at 0 field, demonstrating excellent L10 order of the FePt-SiO2-C layer. Compared to early FePt-C granular film [6], the shape of M-H loops was much better improved. It is consistent with the XRD data in Fig. 1, and the degree of L10 order. Furthermore, the in-plane M-H plot is almost linear, with a small coercivity of 8 kOe, meaning excellent recording performance as HDD recording media [12]. As the last step, we study the thermal stability of the FePt-SiO2-C granular film, to prove that it is qualified for magnetic recording media. In small magnetic particle system, the magnetization of the grain can be flipped by thermal fluctuation kBT at a non-zero absolute temperature T. The probability of magnetic switching observes the Arrhenius-Neel's Law [1]:
Fig. 5. Both perpendicular and in-plane M-H loops of FePt-SiO2-C granular film.
(1)
Where f0 is the attempt frequency, typically ~109 Hz, and Eb is the energy barrier of those grains. If the grains are completely isolated, we can apply the assumption that Eb ¼ KuV, where V is the volume of one grain. When the recording density is increased, the grain becomes smaller and smaller while the thermal energy becomes more and more distinct thus can no more be trivial. As a typical threshold, when Eb/kBT < 40, the magnetization of the grain will be flipped randomly by thermal fluctuation, which is defined as the super-paramagnetic effect [1]. By this effect, the recorded information in the storage bits will decay shortly, thus the recording process is destroyed. In order to delay the super-paramagnetic effect so that the recording bit is thermally stable for more than 10 years, the energy barrier Eb of the magnetic grain system in recording media must satisfy Eb/kBT > 40. Therefore, it is important to measure the Eb value of this FePt-SiO2-C granular film before future application in commercial hard disk drives. On the other hand, in the case of no field, i.e. archive storage, the variation of magnetization for commercial recording media caused by thermal decay will not happen until extremely long time like thousands of years. In order to observe the magnetic time effects in short-time experiments, we apply a field oppositely oriented to the initial magnetization. As a result, the energy barrier is decreased due to negative contribution of Zeeman energy, and the magnetic moment of the recording media will be reversed in a short time, making it possible to be measured in lab. In theory, the time-dependence of remnant coercivity HC(t) observes the Sharrock Equation [18].
1=2 k T f0 t HC ðtÞ ¼ HC0 1 B In Eb 0:693
(2)
In the above equation, the remnant coercivity is defined as the field that switches half of the magnetic particles in the system duration of t. For normal measurement of MH loops by our MPMS SQUID system, the waiting time at each data point is about 1 s after the field application. In order to observe the magnetic decay with time, the waiting time is increased to 10e1000 s. First, we add a huge positive field þ55 kOe; then the field was reduced to negative 24 ~ 32 kOe fields with a step of 2 kOe around the coercive point 29 kOe of this FePt-SiO2-C film, then stay there for certain waiting time; second, the field is set to zero, and the instrument measures the magnetic moment of the film, same as
Fig. 6. Magnetization decay with field in different time delay.
L. Zhang et al. / Journal of Alloys and Compounds 788 (2019) 600e603
603
addition of both SiO2 and C helps form columnar structure, which improves the magnetic behavior of the film. High-resolution TEM observation shows an epitaxial growth of FePt grains on the top of MgO underlayer, forming excellent the L10-ordering of FePt in the film. The perpendicular MH loop of the film has an excellent squareness behavior with the coercivity of 29 kOe. The investigation of remnant coercivity shows that this film has an energy barrier of 290kBT, meaning excellent thermal stability.
References
Fig. 7. Remnant coercivity as a function of the time delay.
normal operation. In Fig. 6, we display the relationship of magnetization and field in different waiting time. We observed a distinct decay of magnetization with increasing waiting time. For each t point, we apply linear fitting for M(H) curve, and get the remnant coercivity at zero field point, shown in Fig. 7. By fitting with Eq. (2), we obtained Eb ¼ 7.3 eV ~ 290kBT at room temperature T ¼ 300 K (kBT ¼ 0.025 eV). So we prove that the film is thermally stable to support data storage for more than 10 years. 4. Conclusion We successfully achieved FePt-SiO2-C granular film with a columnar structure, suitable for ultrahigh density perpendicular recording media higher than 1 Tbits/in2. We grow a 10 nm thick Fe53Pt47-SiO230%-C25% film on silicon substrates at 510 C with a 12 nm MgO interlayer. The FePt grains are well-separated by both SiO2 and C spacers with an average size of 8.3 ± 1.9 nm, which meets the requirement of 1 Tbits/in2 recording density. The
[1] P.L. Lu, S.H. Charap, IEEE Trans. Magn. 31 (1995) 2767. [2] Y. Kubota, Y. Peng, Y. Ding, E.K.C. Chang, L. Gao, F. Zavaliche, T.J. Klemmer, S. Zhu, X. Zhu, P.W. Huang, A.Q. Wu, H. Amini, S. Granz, T. Rausch, C.J. Rea, J. Qiu, H. Yin, M.A. Seigler, Y. Chen, G. Ju, J.U. Thiele, IEEE Trans. Magn. 54 (2018), 3201206. [3] D.A. Gilbert, L.W. Wang, T.J. Klemmer, J.U. Thiele, C.H. Lai, K. Liu, Appl. Phys. Lett. 102 (2013), 132406. [4] T. Ono, T. Moriya, M. Hatayama, K. Tsumura, N. Kikuchi, S. Okamoto, O. Kitakami, T. Shimatsu, Appl. Phys. Lett. 110 (2017), 022402. [5] A. Ghoreyshi, R.H. Victora, J. Appl. Phys. 115 (2014) 17B719. [6] L. Zhang, Y.K. Takahashi, A. Perumal, K. Hono, J. Magn. Magn. Mater. 322 (2010) 2658. [7] A. von Reppert, L. Willig, J.-E. Pudell, M. Roessle, W. Leitenberger, M. Herzog, F. Ganss, O. Hellwig, M. Bargheer, Appl. Phys. Lett. 113 (2018), 123101. [8] J.S. Chen, B.C. Lim, Y.F. Ding, J.F. Hu, G.M. Chow, G. Ju, J. Appl. Phys. 105 (2009) 07B702. [9] C. Feng, M.Y. Yang, X.J. Li, K. Gong, Y.M. Zhu, Q. Zhan, B.H. Li, Y. Jiang, G.H. Yu, Mater. Lett. 99 (2013) 46. [10] J.Y. Shi, M. Tang, Z. Zhang, L. Ma, L. Sun, C. Zhou, X.F. Hu, Z. Zheng, L.Q. Shen, S.M. Zhou, Y.Z. Wu, L.Y. Chen, H.B. Zhao, Appl. Phys. Lett. 112 (2018), 082403. [11] J.J.M. Ruigrok, R. Coehoorn, S.R. Cumpson, H.W. Kesteren, J. Appl. Phys. 87 (2000) 5398. [12] Y. Zeng, X. Huang, W. Zhou, S. Yu, Int. J. Heat Mass Transf. 60 (2013) 322. [13] L. Zhang, X.-T. Zu, J.-G. Peng, Z. Yan, Z.-Y. Zhong, Nanotechnology 19 (2008), 085714. [14] R.E. Rottmayer, S. Batra, D. Buechel, W.A. Challener, J. Hohlfeld, Y. Kobota, L. Li, B. Lu, C. Mihalcea, K. Mountfield, K. Pelhos, C.B. Peng, T. Rausch, M.A. Seigler, D. Weller, X.M. Yang, IEEE Trans. Magn. 42 (2006) 2417. [15] Y. Hirayama, I. Tamai, I. Takekuma, R. Nakatani, J. Phys. Conf. Ser. 165 (2009), 012033. [16] T.O. Seki, Y.K. Takahashi, K. Hono, J. Appl. Phys. 103 (2008), 023910. [17] E. Yang, D.E. Laughlin, J.G. Zhu, IEEE Trans. Magn. 48 (2012) 7. [18] M.P. Sharrock, J. Appl. Phys. 76 (1994) 6413.
Contents lists available at ScienceDirect
Journal of Alloys and Compounds journal homepage: http://www.elsevier.com/locate/jalcom
Investigation of FePt-SiO2-C granular film in columnar structure L. Zhang a, *, C. Li b, Y.J. Zou a, H.L. Xu a, K. Liao a, X.L. Zhang a a b
College of Mechanical and Electrical Engineering, Pingxiang University, 211 Pinganbei Ave., Pingxiang, 337055, People's Republic of China Physical Science and Technology College, Yichun University, 576 Xuefu Rd., Yichun, 336000, People's Republic of China
a r t i c l e i n f o
a b s t r a c t
Article history: Received 6 December 2018 Received in revised form 22 February 2019 Accepted 23 February 2019 Available online 25 February 2019
We worked on a FePt granular film with a columnar structure, for the purpose of ultrahigh density perpendicular magnetic recording media. In experiments, a Fe53Pt47-SiO230%-C25% (10 nm) film was deposited on a silicon substrate with a MgO (12 nm) interlayer at 510 C. The perpendicular coercivity of the film is 29 kOe, with a squareness of 1. Bright-field TEM images show that the FePt granular film has small and uniform grains of 8.3 ± 1.9 nm with a columnar structure. More work of high-resolution TEM imaging shows excellent L10 ordering inside the FePt grains. The measurement of remnant coercivity proves that it has an energy barrier of 290kBT at room temperature, meaning excellent thermal stability. We demonstrate FePt granular film to be a qualified candidate for high-density perpendicular magnetic recording media, and both SiO2 and C addition helps form columnar structure. © 2019 Elsevier B.V. All rights reserved.
Keywords: FePt granular film Columnar structure Magnetic recording media
1. Introduction In modern computers, magnetic recording method is extensively applied as the basic storage architecture. A typical case is hard-disk drive (HDD). In order to improve the HDD performance, an essential task in the research and development work is to increase the recording density, or reduce the dimension of a single storage unit. However, with the further increasing density, the super-paramagnetic effect [1] cannot be neglected any more, and it could induce thermal relaxation in each magnetic grain in the storage units. In order to get excellent thermal stability of recorded information for at least 5 years, the normal lift-time of a HDD, we should increase the coercivity and anisotropy of the recording materials, i.e. making the magnetic material harder and harder. Thus, people in magnetic recording industry work hard to search qualified magnetic materials for high-density recording purpose. After years' investigation, among all possible magnetic materials for recording media, L10-phase FePt granular thin films are almost the best candidates as perpendicular magnetic recording (PMR) media with the recording density of 1 Terabits/in2 or above, because it has a substantially high magneto-crystalline anisotropy (Ku ~ 6 107 erg/cc), which can assist maintain the thermal stability at room temperature for the small grain size of even below 4 nm [2e10]. On the other hand, hard recording medium is difficult in
* Corresponding author. E-mail address: [email protected] (L. Zhang). https://doi.org/10.1016/j.jallcom.2019.02.263 0925-8388/© 2019 Elsevier B.V. All rights reserved.
writing at room temperature because no pole material can create strong field. Nowadays this trouble could be solved by heating the recording medium during heat-assisted magnetic recording (HAMR) [2,11e14]. In early efforts of other groups, they achieved nanometer grain size FePt thin films with the addition of Ag and C with high coercivity and thermal stability [6]. In their work, carbon is used as the spacer to separate FePt nanograins, and Ag is used to reduce the L10 ordering temperature. However, early work on FePt granular films always shows poor squareness of the perpendicular MH loop due to the small grain size and also thickness of the film. In addition to carbon, some oxides can also act as spacer in FePt thin films. Chen et al. worked on FePt granular film by addition of TiO2 and Ta2O5 [8]. Those oxides can help form columnar structure, in which we believe that it could improve the switching behavior of nanograins. However, their work did not show convincing improvement on the magnetic properties. On the other hand, for the currently-applied CoCrPt thin film in magnetic recording media in HDD industries, SiO2 is a successful spacer to separate CoCrPt grains with excellent magnetic behavior [15]. However, the addition of SiO2 alone into FePt thin films does not improve both the microstructure and magnetic properties [16]. The combination of both oxides and carbon has been few reported. In this paper, we studied a FePt granular film with the addition both of SiO2 and C. Then, we displayed detailed work to study the microstructure of FePt grains through high-resolution transmission electron microscopy (TEM) imaging. Finally, we measured the remnant coercivity of the film in different time delay to study the thermal stability of this FePt granular film.
L. Zhang et al. / Journal of Alloys and Compounds 788 (2019) 600e603
601
2. Experiments We deposit a 10 nm thick Fe53Pt47-SiO230%-C25% film on a bare silicon substrate. At first, we grow a 12 nm MgO interlayer on the top of the silicon substrate, then, FePt-SiO2-C layer on the top of MgO interlayer, in a high-vacuum magnetron-sputtering chamber. The volume fraction of carbon is 25%, and that of SiO2 is 30% in the film. The MgO layer was grown at room temperature, while the FePt-SiO2-C layer at 510 C. During the film growth, the Ar pressure is 1.1 Pa for MgO, and 0.55 Pa for FePt-SiO2-C, respectively. The deposition rate of MgO is about 0.06 nm/s, and FePt-SiO2-C is about 0.025 nm/s. The composition of Fe, Pt, SiO2, and C is based on the growth rate of each single material. The total FePt-SiO2-C film was formed by co-sputtering method. In experiments, the base pressure of the vacuum chamber is about 2 106 Pa. After the thin film fabrication, we use XRD (x-ray diffractor) to study the texture of the FePt-SiO2-C film with MgO underlayer. We use SQUID (superconducting quantum interferometer device) MPMS (magnetic property measurement system) to measure the magnetic properties. The field range of SQUID is 55 ~ þ55 kOe. Then, we use a TEM to study the microstructure of the film, in two modes: bright-field and high-resolution. 3. Results and discussion In Fig. 1, we display the XRD pattern of the FePt-SiO2-C/MgO/Si film. The index of both MgO and FePt peaks are labeled, while FePt peaks are marked by numbers only. We observed a distinct MgO(200) peak at 2q ¼ 43.1 and the L10-ordered fct-FePt(001) (002) peaks, basically induced by the MgO(200) texture from the underlayer [6]. Both fcc-FePt(200) and FePt(111) peaks are suppressed. Based on the theoretical investigation by Yang et al., we calculated the degree of the L10 order of this FePt-SiO2-C granular film, to be 0.89 [17]. The above facts indicate excellent L10 order in the FePt-SiO2-C film layer. Next, we work on the microstructure of the FePt-SiO2-C film by TEM. Fig. 2(a) shows a bright-field TEM image of the plane-view for the film, with the insets of selected area electron diffraction (SAED) pattern and statistic diagram of the grain size distribution. From the image, we can see that the grains in the FePt-SiO2-C thin film are uniformly separated and distributed. After statistical calculation, we obtained the average grain size is 8.3 nm, with a standard deviation of 1.9 nm. In SAED pattern, counted from center to edge, the rings of FePt(001), FePt(110) MgO(200), and FePt(002) peaks are clearly displayed. It agrees with the data in Fig. 1 (XRD pattern). It
means that the excellent L10 ordering is formed in this FePt-SiO2-C granular film. Fig. 2(b) shows the bright-field TEM image of the cross-sectional view of the FePt-SiO2-C granular film. It is clear that those FePt grains form a single-grain layer columnar structure in the film, much improved compared with early FePt-C granular films [6]. After that, we did high-resolution TEM imaging on the FePtSiO2-C granular film, so that we can understand more details on the microstructure. Fig. 3 displays a plane-view high-resolution TEM image of the FePt grains. We observed some MgO matrix below the L10-ordered FePt nanograins. We also can clearly find the boundary of MgO grains in the underlayer. Next, Fig. 4 displays a cross
Fig. 1. XRD pattern of FePt-SiO2-C granular film.
Fig. 3. High-resolution plane-view TEM image of FePt-SiO2-C granular film.
Fig. 2. Bright-field TEM image of (a) the plane-view, with the inset of selected area electron diffraction pattern, and statistics of grain size distribution, (b) the crosssectional view of FePt-SiO2-C granular film.
602
L. Zhang et al. / Journal of Alloys and Compounds 788 (2019) 600e603
E f ¼ f0 exp b kB T
Fig. 4. High-resolution cross-sectional view of FePt-SiO2-C granular film.
sectional high-resolution TEM image of the FePt grains. It seems difficult to find the proof of direct relationship between the FePt grains and the MgO grains below. However, we are able to find the epitaxial relationship between FePt and MgO grains, i.e. (001)FePt// (002)MgO. It demonstrates that the MgO polycrystalline underlayer aids FePt grains aligned to (001) orientation normal to the film plane, getting the L10-order in the FePt granular film. In addition, the strong order of FePt L10-phase can be demonstrated through the measurement of magnetic properties. We use MPMS SQUID to work on this FePt-SiO2-C granular film at room temperature. In Fig. 5, we see both perpendicular and in-plane M-H loops of this film. From those two curves, this film has high perpendicular coercivity of 29 kOe, with perfect squareness of 1 at 0 field, demonstrating excellent L10 order of the FePt-SiO2-C layer. Compared to early FePt-C granular film [6], the shape of M-H loops was much better improved. It is consistent with the XRD data in Fig. 1, and the degree of L10 order. Furthermore, the in-plane M-H plot is almost linear, with a small coercivity of 8 kOe, meaning excellent recording performance as HDD recording media [12]. As the last step, we study the thermal stability of the FePt-SiO2-C granular film, to prove that it is qualified for magnetic recording media. In small magnetic particle system, the magnetization of the grain can be flipped by thermal fluctuation kBT at a non-zero absolute temperature T. The probability of magnetic switching observes the Arrhenius-Neel's Law [1]:
Fig. 5. Both perpendicular and in-plane M-H loops of FePt-SiO2-C granular film.
(1)
Where f0 is the attempt frequency, typically ~109 Hz, and Eb is the energy barrier of those grains. If the grains are completely isolated, we can apply the assumption that Eb ¼ KuV, where V is the volume of one grain. When the recording density is increased, the grain becomes smaller and smaller while the thermal energy becomes more and more distinct thus can no more be trivial. As a typical threshold, when Eb/kBT < 40, the magnetization of the grain will be flipped randomly by thermal fluctuation, which is defined as the super-paramagnetic effect [1]. By this effect, the recorded information in the storage bits will decay shortly, thus the recording process is destroyed. In order to delay the super-paramagnetic effect so that the recording bit is thermally stable for more than 10 years, the energy barrier Eb of the magnetic grain system in recording media must satisfy Eb/kBT > 40. Therefore, it is important to measure the Eb value of this FePt-SiO2-C granular film before future application in commercial hard disk drives. On the other hand, in the case of no field, i.e. archive storage, the variation of magnetization for commercial recording media caused by thermal decay will not happen until extremely long time like thousands of years. In order to observe the magnetic time effects in short-time experiments, we apply a field oppositely oriented to the initial magnetization. As a result, the energy barrier is decreased due to negative contribution of Zeeman energy, and the magnetic moment of the recording media will be reversed in a short time, making it possible to be measured in lab. In theory, the time-dependence of remnant coercivity HC(t) observes the Sharrock Equation [18].
1=2 k T f0 t HC ðtÞ ¼ HC0 1 B In Eb 0:693
(2)
In the above equation, the remnant coercivity is defined as the field that switches half of the magnetic particles in the system duration of t. For normal measurement of MH loops by our MPMS SQUID system, the waiting time at each data point is about 1 s after the field application. In order to observe the magnetic decay with time, the waiting time is increased to 10e1000 s. First, we add a huge positive field þ55 kOe; then the field was reduced to negative 24 ~ 32 kOe fields with a step of 2 kOe around the coercive point 29 kOe of this FePt-SiO2-C film, then stay there for certain waiting time; second, the field is set to zero, and the instrument measures the magnetic moment of the film, same as
Fig. 6. Magnetization decay with field in different time delay.
L. Zhang et al. / Journal of Alloys and Compounds 788 (2019) 600e603
603
addition of both SiO2 and C helps form columnar structure, which improves the magnetic behavior of the film. High-resolution TEM observation shows an epitaxial growth of FePt grains on the top of MgO underlayer, forming excellent the L10-ordering of FePt in the film. The perpendicular MH loop of the film has an excellent squareness behavior with the coercivity of 29 kOe. The investigation of remnant coercivity shows that this film has an energy barrier of 290kBT, meaning excellent thermal stability.
References
Fig. 7. Remnant coercivity as a function of the time delay.
normal operation. In Fig. 6, we display the relationship of magnetization and field in different waiting time. We observed a distinct decay of magnetization with increasing waiting time. For each t point, we apply linear fitting for M(H) curve, and get the remnant coercivity at zero field point, shown in Fig. 7. By fitting with Eq. (2), we obtained Eb ¼ 7.3 eV ~ 290kBT at room temperature T ¼ 300 K (kBT ¼ 0.025 eV). So we prove that the film is thermally stable to support data storage for more than 10 years. 4. Conclusion We successfully achieved FePt-SiO2-C granular film with a columnar structure, suitable for ultrahigh density perpendicular recording media higher than 1 Tbits/in2. We grow a 10 nm thick Fe53Pt47-SiO230%-C25% film on silicon substrates at 510 C with a 12 nm MgO interlayer. The FePt grains are well-separated by both SiO2 and C spacers with an average size of 8.3 ± 1.9 nm, which meets the requirement of 1 Tbits/in2 recording density. The
[1] P.L. Lu, S.H. Charap, IEEE Trans. Magn. 31 (1995) 2767. [2] Y. Kubota, Y. Peng, Y. Ding, E.K.C. Chang, L. Gao, F. Zavaliche, T.J. Klemmer, S. Zhu, X. Zhu, P.W. Huang, A.Q. Wu, H. Amini, S. Granz, T. Rausch, C.J. Rea, J. Qiu, H. Yin, M.A. Seigler, Y. Chen, G. Ju, J.U. Thiele, IEEE Trans. Magn. 54 (2018), 3201206. [3] D.A. Gilbert, L.W. Wang, T.J. Klemmer, J.U. Thiele, C.H. Lai, K. Liu, Appl. Phys. Lett. 102 (2013), 132406. [4] T. Ono, T. Moriya, M. Hatayama, K. Tsumura, N. Kikuchi, S. Okamoto, O. Kitakami, T. Shimatsu, Appl. Phys. Lett. 110 (2017), 022402. [5] A. Ghoreyshi, R.H. Victora, J. Appl. Phys. 115 (2014) 17B719. [6] L. Zhang, Y.K. Takahashi, A. Perumal, K. Hono, J. Magn. Magn. Mater. 322 (2010) 2658. [7] A. von Reppert, L. Willig, J.-E. Pudell, M. Roessle, W. Leitenberger, M. Herzog, F. Ganss, O. Hellwig, M. Bargheer, Appl. Phys. Lett. 113 (2018), 123101. [8] J.S. Chen, B.C. Lim, Y.F. Ding, J.F. Hu, G.M. Chow, G. Ju, J. Appl. Phys. 105 (2009) 07B702. [9] C. Feng, M.Y. Yang, X.J. Li, K. Gong, Y.M. Zhu, Q. Zhan, B.H. Li, Y. Jiang, G.H. Yu, Mater. Lett. 99 (2013) 46. [10] J.Y. Shi, M. Tang, Z. Zhang, L. Ma, L. Sun, C. Zhou, X.F. Hu, Z. Zheng, L.Q. Shen, S.M. Zhou, Y.Z. Wu, L.Y. Chen, H.B. Zhao, Appl. Phys. Lett. 112 (2018), 082403. [11] J.J.M. Ruigrok, R. Coehoorn, S.R. Cumpson, H.W. Kesteren, J. Appl. Phys. 87 (2000) 5398. [12] Y. Zeng, X. Huang, W. Zhou, S. Yu, Int. J. Heat Mass Transf. 60 (2013) 322. [13] L. Zhang, X.-T. Zu, J.-G. Peng, Z. Yan, Z.-Y. Zhong, Nanotechnology 19 (2008), 085714. [14] R.E. Rottmayer, S. Batra, D. Buechel, W.A. Challener, J. Hohlfeld, Y. Kobota, L. Li, B. Lu, C. Mihalcea, K. Mountfield, K. Pelhos, C.B. Peng, T. Rausch, M.A. Seigler, D. Weller, X.M. Yang, IEEE Trans. Magn. 42 (2006) 2417. [15] Y. Hirayama, I. Tamai, I. Takekuma, R. Nakatani, J. Phys. Conf. Ser. 165 (2009), 012033. [16] T.O. Seki, Y.K. Takahashi, K. Hono, J. Appl. Phys. 103 (2008), 023910. [17] E. Yang, D.E. Laughlin, J.G. Zhu, IEEE Trans. Magn. 48 (2012) 7. [18] M.P. Sharrock, J. Appl. Phys. 76 (1994) 6413.