Perpendicular Magnetic Anisotropy in FePt Patterned Media Employing a CrV Seed Layer
- Hyunsu Kim†1,
- Jin-Seo Noh†1,
- Jong Wook Roh1,
- Dong Won Chun2,
- Sungman Kim2,
- Sang Hyun Jung3,
- Ho Kwan Kang3,
- Won Yong Jeong2 and
- Wooyoung Lee1Email author
© Kim et al. 2010
Received: 10 July 2010
Accepted: 10 August 2010
Published: 26 August 2010
A thin FePt film was deposited onto a CrV seed layer at 400°C and showed a high coercivity (~3,400 Oe) and high magnetization (900–1,000 emu/cm3) characteristic of L 10 phase. However, the magnetic properties of patterned media fabricated from the film stack were degraded due to the Ar-ion bombardment. We employed a deposition-last process, in which FePt film deposited at room temperature underwent lift-off and post-annealing processes, to avoid the exposure of FePt to Ar plasma. A patterned medium with 100-nm nano-columns showed an out-of-plane coercivity fivefold larger than its in-plane counterpart and a remanent magnetization comparable to saturation magnetization in the out-of-plane direction, indicating a high perpendicular anisotropy. These results demonstrate the high perpendicular anisotropy in FePt patterned media using a Cr-based compound seed layer for the first time and suggest that ultra-high-density magnetic recording media can be achieved using this optimized top-down approach.
Conventional planar magnetic recording methods have been facing difficulties in reducing the thickness of a magnetic film and the average grain size in it, which is required for the high bit density [1, 2]. Furthermore, these methods showed a bit density limit of about 100 Gbit/in2 due to the magnetic moment instability termed 'superparamagnetism' in very small grains and the magnetic exchange interaction between adjacent grains [1–4]. To overcome this limit, a perpendicular magnetic recording was introduced, where magnetic moments are aligned perpendicular to the film plane [5, 6]. However, a bit loss still occurs by exchange interaction between neighboring grains. Patterned magnetic media have emerged as a means to prevent this intergranular exchange interaction, thus to achieve the ultra-high density of magnetic recording. To realize the very fine patterned media, a proper material stack and well-optimized fabrication process should be chosen to retain the magnetization in the perpendicular direction with a high perpendicular anisotropy (K u ).
FePt is a magnetic material that has been intensively investigated due to its high coercivity (H c = 1–10 kOe) [7–11] and high magnetocrystalline anisotropy (K c = 7.0 × 107 erg/cm3) [8, 10, 12]. This material undergoes a transition from chemically disordered face-centered cubic phase (FCC, A 1 phase) to ordered face-centered tetragonal phase (FCT, L 10 phase) at a specific temperature, and the transition temperature and perpendicular anisotropy are known to depend on the buffer layer and process employed. A variety of buffer layers have been introduced on Si or glass substrates to grow high quality FCT structures at low temperatures, including Pt, Au, Ag, Ti, and MgO [13–16]. Although L 10 FePt films on these buffer layers demonstrated an increase in coercivity with respect to the buffer-free films, the ratio of out-of-plane to in-plane coercivities has generally been smaller than 3. Other than these rather conventional buffer layers, Cr-based compounds such as CrW  and CrRu  have also been examined as underlayers since (200) planes of a body-centered cubic (BCC) Cr were likely to stimulate (001) texture formation of the FCT FePt and to facilitate the FCC-to-FCT transition in FePt layer by forcing the tensile stress to a0 side of the original FCC FePt [17, 18], achieving the ratio of out-of-plane to in-plane coercivities larger than 5 at a relatively low temperature (400°C) . To our knowledge, however, no works have successfully demonstrated the high perpendicular anisotropy in FePt fine-patterned media employing a Cr compound, presumably due to the difficulty in optimal process design.
In this work, we fabricated magnetic recording media by a combination of E-beam lithography and either dry etching (deposition-first process) or lift-off (deposition-last process), where magnetic nano-columns were regularly arranged with a fixed spacing. The magnetic properties and crystal structures were investigated at important steps of the fabrication of the patterned media. The high perpendicular anisotropy is demonstrated in the fine-patterned media, suggesting the feasibility of achieving the ultra-high-density recording media through a well-designed fabrication process.
As an alternative process, a lift-off process (deposition-last process) was employed to fabricate the patterned media, as shown in Figure 1b. For this process, a type of positive ER was coated on CrV layer and patterned undergoing E-beam exposure and development steps, leaving behind a regular array of holes of a fixed size (typically, 100 nm). Then, a 7-nm-thick FePt layer was deposited by sputtering at room temperature, followed by a lift-off. The final FePt patterns (the last panel of Figure 1b) were subsequently annealed at 400°C for 1 h to induce a phase transformation from A 1 to L 10 phase.
To analyze the crystal structures of as-grown films and patterned media, conventional θ–2θ X-ray diffraction (XRD) was performed using Cu K α radiation. Magnetic properties were investigated at room temperature, using a superconducting quantum interference device (SQUID) with a sensitivity of 1 × 10-6 emu. Microstructures of the film stacks and top-views of the fabricated patterned media were observed using transmission electron microscopy (TEM) and scanning electron microscopy (SEM), respectively.
Results and Discussion
However, the coercivities (Hc,pattern = 450–900 Oe) of the patterned medium appear to be 4 to sevenfold smaller than Hc,film both in film plane and normal to plane, although its saturation magnetizations (Ms,pattern) are similar to Ms,film. In addition, the ratio (Mr,pattern/Ms,pattern = 0.4–0.7) of Mr,pattern to Ms,pattern for the medium is smaller than that of the as-grown film. Recollecting that the coercivity and Mr/Ms ratio are more structure-sensitive than the saturation magnetization, these results suggest that the chemically ordered FCT structure was destroyed and replaced by the chemically disordered FCC structure at least partially during ICP Ar etching. To verify this presumption, we carried out the XRD analysis on the patterned medium. Indeed, it is shown from Figure 3c that the FCT (002) peak is weak and instead, a FCC (002) peak is clearly developed around 2θ = 44.5°, justifying the propriety of the above presumption. We think that the large decrease in coercivity for the patterned medium originated from the relaxation of magnetocrystalline anisotropy (K c ) due to the chemical disordering in the FePt patterns [7, 24, 25]. This is because shape anisotropy (K d α α, where α is the demagnetization factor) strengthens the perpendicular alignment of magnetic moments, and magnetoelastic anisotropy (K where is the magnetostriction constant and is the stress in film) remains almost unchanged via patterning . The Ar-ion penetration into the FePt film and a large momentum delivered from impinging Ar ions may be primary sources for the collapse of the FCT structure. The drastic decrease in coercivity was also observed in other patterned media with different pattern size, as shown in Figure 3b. It is seen from this figure that both coercivities and Mr/Ms ratios for patterned media are significantly reduced from the values of the as-grown film irrespective of pattern size, reflecting the FCT structure was collapsed for samples undergoing Ar plasma etching as confirmed by the XRD result in Figure 3c.
Comparing the out-of-plane coercivity of this patterned medium with that of the as-grown film prepared by the deposition-first process, there exists a small difference of about 400 Oe. We believe that this magnitude of difference is reasonable since the surface migration of adatoms during film growth at elevated temperature (400°C) is easier compared to solid-state diffusion of constituents during post-annealing at the same temperature. Qiu et al. also fabricated FePt patterned media with underlayers such as Ag and MgO, employing a similar deposition-last process . In their media, however, the FCC-to-FCT phase transition was retarded to higher temperatures and no perpendicular anisotropy was observed. Assuming that the magnetocrystalline anisotropy is a primary source of our perpendicular anisotropy as explained above, the perpendicular anisotropy is proportional to the coercivity and saturation magnetization in the out-of-plane direction, K u Hc,outMs,out. In our 100-nm-sized FePt patterns fabricated by the deposition-last process, the values are measured to be 3,000 Oe and 870 emu/cm3, respectively. These values are comparable to those of the previously reported FePt thin films on other Cr-based compounds such as CrW  and CrRu , demonstrating the high efficiency of the CrV seed layer in fabricating patterned media with a high perpendicular anisotropy. Besides this, our results disclose important implications: (1) a root cause of the magnetic property degradation of FePt patterned media fabricated by a conventional deposition-first process is chemical disordering incurred by ion plasma etching. (2) The deposition-last process is desirable for implementing ultra-high-density patterned media, and the post-annealing temperature can be maintained low by the support of an appropriate seed layer.
We fabricated FePt-based perpendicular patterned media using a selective combination of E-beam lithography and either Ar plasma etching (deposition-first process) or FePt lift-off (deposition-last process). A FePt film on a CrV seed layer grown at 400°C showed a high perpendicular anisotropy indicating L 10 phase of FCT structure formed during deposition, whereas the anisotropy was collapsed in patterned media fabricated from the film stack. We employed the deposition-last process to avoid chemical and structural disordering by impinging Ar ions. For a patterned medium with 100 nm patterns made by this process, the out-of-plane coercivity was measured to be fivefold larger than its in-plane value and the out-of-plane M-H curve exhibited a perfect squareness, indicating a high perpendicular anisotropy. To our knowledge, this is the first demonstration of a high perpendicular anisotropy in patterned media using a Cr-based compound seed layer. Furthermore, the deposition-last process may be a promising way to achieve ultra-high-density patterned media due to its maintainability of perpendicular anisotropy and controllability of pattern size and shape.
This research was supported by a grant from the Fundamental R&D Program for Core Technology of Materials funded by the Ministry of Knowledge Economy, Republic of Korea, and the Priority Research Centers Program (2009-0093823) funded by the National Research Foundation of Korea (NRF).
- Moser A, Takano K, Margulies DT, Albrecht M, Sonobe Y, Ikeda Y, Sun S, Fullerton EE: J Phys D Appl Phys. 2002, 35: R157. 10.1088/0022-3727/35/19/201View ArticleGoogle Scholar
- Albrecht M, Rettner CT, Best ME, Terris BD: Appl Phys Lett. 2003, 83: 4363. 10.1063/1.1630153View ArticleGoogle Scholar
- Bertram HN, Zhou H, Gustafson R: IEEE Trans Magn. 1998, 34: 1845. 10.1109/20.706722View ArticleGoogle Scholar
- Acharya BR, Inomata A, Abarra EN, Ajan A, Hasegawa D, Okamoto I: J Magn Magn Mater. 2003, 260: 261. 10.1016/S0304-8853(02)00284-6View ArticleGoogle Scholar
- Iwasaki S: IEEE Trans Magn. 1980, 16: 71. 10.1109/TMAG.1980.1060546View ArticleGoogle Scholar
- Grundy PJ: J Phys D Appl Phys. 1998, 31: 2975. 10.1088/0022-3727/31/21/001View ArticleGoogle Scholar
- Terris BD, Weller D, Folks L, Baglin JEE, Kellock AJ, Rothuizen H, Vettiger P: J Appl Phys. 2000, 87: 7004. 10.1063/1.372912View ArticleGoogle Scholar
- Hasegawa T, Pei W, Wang T, Fu Y, Washiya T, Saito H, Ishio S: Acta Mater. 2008, 56: 1564. 10.1016/j.actamat.2007.12.008View ArticleGoogle Scholar
- Chun DW, Kim SM, Kim GH, Jeung WY: J Appl Phys. 2009, 105: 07B731. 10.1063/1.3075981View ArticleGoogle Scholar
- Hong MH, Hono K, Watanabe M: J Appl Phys. 1998, 84: 4403. 10.1063/1.368662View ArticleGoogle Scholar
- Zhang B, Soffa WA: IEEE Trans Magn. 1990, 26: 1388. 10.1109/20.104386View ArticleGoogle Scholar
- Sun S, Murray CB, Weller D, Folks L, Moser A: Science. 2000, 287: 1989. 10.1126/science.287.5460.1989View ArticleGoogle Scholar
- Chen JS, Yingfan Xu, Wang JP: J Appl Phys. 2003, 93: 1661. 10.1063/1.1531817View ArticleGoogle Scholar
- Zhu Yun, Cai JW: Appl Phys Lett. 2005, 87: 1–032504.Google Scholar
- Qiu LJ, Ding J, Adeyeye AO, Yin JH, Chen JS, Goolaup S, Singh N: IEEE Trans Magn. 2007, 43: 2157. 10.1109/TMAG.2007.893135View ArticleGoogle Scholar
- Chen SC, Kuo PC, Kuo ST, Sun AC, Lie CT, Chou CY: Mater Sci Eng. 2003, B98: 244. 10.1016/S0921-5107(03)00048-5View ArticleGoogle Scholar
- Cao J, Cai J, Liu Y, Yang Z, Wei F, Xia A, Han B, Bai J: J Appl Phys. 2006, 99: 1–08F901.Google Scholar
- Xu Y, Chen JS, Wang JP: Appl Phys Lett. 2002, 80: 3325. 10.1063/1.1476706View ArticleGoogle Scholar
- Murayama N, Soeya S, Takahashi Y, Futamoto M: J Magn Magn Mater. 2008, 320: 3057. 10.1016/j.jmmm.2008.08.053View ArticleGoogle Scholar
- Chun DW, Kim SM, Jeung WYTo be published To be publishedGoogle Scholar
- Breitling A, Goll D: J Magn Magn Mater. 2008, 320: 1449. 10.1016/j.jmmm.2007.12.003View ArticleGoogle Scholar
- Zhong H, Tarrach G, Wu P, Drechsler A, Wei D, Yuan Jun: Nanotechnology. 2008, 19: 095703. 10.1088/0957-4484/19/9/095703View ArticleGoogle Scholar
- Lairson BM, Visokay MR, Sinclair R, Clemens BM: Appl Phys Lett. 1993, 62: 639. 10.1063/1.108880View ArticleGoogle Scholar
- Jaafar M, Sanz R, Vázquez M, Asenjo A, Jensen J, Hjort K, Flohrer S, McCord J, Schäfer R: Phys Stat Sol (A). 2007, 204: 1724. 10.1002/pssa.200675342View ArticleGoogle Scholar
- Kavita S, Raghavendra Reddy V, Gupta A, Amirthapandian S, Panigrahi BK: Nucl Instrum Meth B. 2006, 244: 206. 10.1016/j.nimb.2005.11.070View ArticleGoogle Scholar
- Jeong JR, Kim YS, Shin SC: J Appl Phys. 1999, 85: 5762. 10.1063/1.370118View ArticleGoogle Scholar
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.