Abstract
Magnetic recording media are composed of magnetic thin films consisting of magnetically isolated crystallites. For practical use of magnetic particles as recording media, it will be necessary to realize high coercivity by fabricating nanocrystalline grains and forming grain boundaries with the nonmagnetic phase. In this study, a high-coercivity FePt–Ag nanocomposite magnet was synthesized by means of block copolymer-templated self-assembly. Precursors of Fe, Pt, and Ag were introduced into a polymer block, and the resulting material was oxidized and then reduced to form a nanocomposite consisting of FePt nanoparticles surrounded by a matrix of Ag. X-ray diffraction analysis revealed that the introduction of Ag did not significantly affect the crystalline ordering of the FePt. The addition of Ag increased the coercivity by 53% (from 11.1 to 17.0 kOe). Our results suggest that the grain boundaries of the nonmagnetic Ag metal acted as pinning sites, disrupting magnetic coupling between individual FePt nanocrystallites and hindering domain wall motion at an external magnetic field.
1. Introduction
Along with recent improvements in the processing capability of computers, the capacity of the hard disk drive as an external storage device has also dramatically increased. Magnetic recording media are rapidly becoming popular not only for personal computers but also for consumer electronics such as video recorders, and demand for such media will likely increase as high-resolution digital television broadcasting becomes more common. In addition, as the IoT develops, further data densification will be required so that digital data can be recorded for longer periods of time and so that the portability of related equipment using magnetic recording media can be increased by means of miniaturization.
Magnetic recording media are composed of magnetic thin films consisting of magnetically isolated crystallites [1–6]. Increasing the recording density requires a reduction in the medium area for a single bit, which in turn necessitates a proportional decrease in the volume of the magnetic particles constituting the medium. However, decreasing the particle volume results in deterioration of resistance to thermal disturbance and in magnetization reversal due to thermal fluctuation, and consequently information cannot be recorded on the medium [7–9].
The thermal fluctuation problem can be minimized by using magnetic recording media with a high magnetic anisotropy () [10, 11]. The ordered L10 FePt phase has of at least 10 times that of the currently used CoCr alloy system [12], and this material is the most promising candidate for a high- medium. However, for practical use of L10 ordered alloys as recording media, it will be necessary to realize high coercivity by fabricating nanocrystalline grains, controlling (001) orientation, and forming grain boundaries with the nonmagnetic phase [10, 13].
Sputtering is the most commonly used method for fabricating FePt nanostructures [14–18], but this method, which requires high vacuum and high energy, is too expensive for industrial applications. Fabrication of FePt nanostructures by chemical synthesis is a promising alternative, but achieving high coercivity with chemical synthesis methods has been difficult [19–22]. The problem has been attributed to the difficulty of controlling nanocomposite structures.
In this study, we synthesized FePt–Ag nanoparticles by using self-assembled block copolymer (BCP) nanostructures as templates to control the nanocomposite structure. Precursors of Fe, Pt, and Ag were introduced into one polymer block of the BCP, and then oxidation and reduction were used to remove the BCP template and crystallize the FePt nanocrystallites. We also investigated how the addition of nonmagnetic Ag affected the structure and magnetic properties of the FePt–Ag nanoparticles.
2. Materials and Methods
2.1. Sample Preparation
In a typical procedure, 0.13 g of tricarbonyl(cyclooctatetraene)iron (>96%), 0.54 g of dimethyl(1,5-cyclooctadiene)platinum(II) (99%), 0.57 g of tris(acetylacetonate)iron(III) (99%), and 0.13 g of silver trifluoroacetate (97%) were dissolved in a 0.5 wt% solution of polystyrene-b-poly(4-vinylpyridine) (PS-P4VP, kg mol–1, kg mol–1, and polydispersity index = 1.09) in 45 mL of toluene (>99.5%). After stirring for 3 h at room temperature, the solution was transferred to a Petri dish and calcined at 180°C for 6 h, 450°C for 6 h, and 650°C for 1 h under a flow of 5% H2 in Ar. Then the sample was oxidized at 550°C for 6 h in air and reduced at 450°C for 6 h under a flow of 5% H2 in Ar. A reference sample was synthesized by means of the same protocol, but without silver trifluoroacetate.
2.2. Characterization
Scanning transmission electron microscopy (STEM) images and energy-dispersive spectrometry (EDS) results were obtained with a JEOL JEM-2010FEF(HR) transmission electron microscope operated at 200 keV. X-ray diffraction (XRD) patterns were collected with Cu Kα radiation on a Rigaku RINT-TTR diffractometer operated at 40 kV and 50 mA. To measure magnetic properties, samples were saturated in a 100 kOe pulsed magnetic field produced by a pulsed high-field magnetometer (TPM-1-15, Toei Industry Co.), and the magnetic properties were measured at 300 K with a vibrating sample magnetometer (VSM-3S-15, Toei Industry Co.).
3. Results and Discussion
Dark-field and bright-field STEM images of a sample synthesized with Ag and heat-treated at 180°C showed the formation of spherical BCP-precursor composite mesophases with diameters of 40–100 nm (Figures 1(a) and 1(e)). The dark areas in the bright-field image correspond to more-electron-dense areas, and the contrast observed in the image suggests that electron-dense elements were homogeneously distributed in the areas. The EDS mapping images in Figures 1(b)–1(d) show that the Fe, Pt, and Ag precursors were introduced into the P4VP blocks and that the Fe precursor was introduced into the PS matrix, which is consistent with previous reports [23, 24]. Spherical mesophase formation was driven by the selective dissolution of the PS block in the toluene-based solvent [25], which resulted in dissolution of the Fe, Pt, and Ag precursors into the P4VP block.
(a)
(b)
(c)
(d)
(e)
EDS spot analysis of a sample synthesized with Ag and heat-treated at 180°C revealed that large portions of the Fe, Pt, and Ag precursors were introduced into the spheres, with smaller portions being introduced into the matrix (Figure 2 and Table 1).
SEM and EDS mapping images of oxidized and reduced sample synthesized with Ag clearly showed the distributions of Fe, Pt, and Ag after removal of the polymer templates. A dark-field STEM image shows condensed nanoparticles with diameters of 20–60 nm (Figure 3(a)), which corresponds almost exactly to the dimensions of the P4VP polymer blocks into which the Fe, Pt, and Ag precursors were introduced (Figure 1(a)). EDS mapping images show that these nanoparticles consisted of Fe and Pt and that they were surrounded by a matrix of Ag.
(a)
(b)
(c)
(d)
Peaks for an L10 tetragonal hard magnetic phase of FePt were observed in the XRD patterns of an oxidized and reduced sample synthesized without Ag (Figure 4(a)). The XRD profile of a sample synthesized with Ag confirmed the presence of two crystalline phases (Figure 4(b)): a L10 tetragonal hard magnetic phase of FePt containing Ag and a nonmagnetic phase of the Ag metal. The L10 tetragonal phase peaks were slightly shifted to a lower angle, indicating that the FePt was alloying with Ag. This result corroborates the EDS results; that is, a large portion of Ag was present at the grain boundary and a small portion of Ag was present within the FePt nanoparticles (Figure 3).
(a)
(b)
For bulk L10 FePt, a- and c-lattice parameters of 3.85 and 3.71 Å, respectively, have been reported [26]. Calculating the a- and c-lattice parameter values from the position of the (200) and (001) XRD peaks for the oxidized and reduced sample synthesized with Ag gives = 3.87 Å and = 3.74 Å, which are a little larger than those reported for bulk L10 FePt. This difference in parameter values is likely due to expansion of the lattice through incorporation of Ag atoms and the altered FePt composition resulting from the reaction of Fe with a small amount of Ag.
These results indicate that the Pt, Fe, and Ag precursors introduced into the spherical polymer blocks (Figure 1) crystallized into L10 FePt containing Ag. In the formation of FePt nanoparticles, a large portion of the Ag within the spheres of the polymer template was extruded or diffused to the grain boundary. The spheres containing large amounts of Ag ultimately became FePt nanoparticles containing a small amount of Ag. We hypothesize that the diffusion distance of Ag was less than several tens of nanometers during the heat treatment up to 650°C.
We evaluated the degree of chemical ordering in the samples by considering the uniaxial magnetocrystalline anisotropy of the L10 FePt phase. The degree of chemical ordering is based on the crystalline ordering of the face-centered tetragonal structure and is described by the chemical order parameter . This dimensionless parameter is given by the equation , where and are the integrated intensities of the (001) and (002) peaks in the XRD spectrum, respectively [27]. The values of for calcined samples synthesized without Ag and with Ag were 0.944 and 0.955, respectively. These values suggest that the introduction of Ag did not significantly affect the crystalline ordering of FePt. Ag diffusion promotes atomic diffusion, leading to a change in the chemical ordering of the L10 tetragonal phase after reduction. However, Ag diffusion had little effect on the promotion of this chemical ordering change. We hypothesize that the amount of Ag (13%) was insufficient to promote a change in the chemical ordering [19].
Magnetization curves of oxidized and reduced samples synthesized with Ag and without Ag (Figure 5) revealed that saturation magnetization of the sample synthesized with Ag (31.7 emu/g) was 24% less than that of the sample synthesized without Ag (41.9 emu/g). This result suggests that nonmagnetic species, possibly including Ag metal, lowered the saturation magnetization by decreasing the amount of magnetic substances in the sample.
The coercivity was increased by 53% (from 11.1 to 17.0 kOe) by the addition of Ag. The coercivity of the sample synthesized with Ag was much higher than the reported values for chemically synthesized FePt nanocomposites [19–22]. This suggests that the increase in the coercivity was caused by the addition of Ag to the FePt nanocrystals and that the grain boundaries of the nonmagnetic Ag metal acted as pinning sites, disrupting magnetic coupling between individual FePt nanocrystallites and hindering domain wall motion at an external magnetic field. This result is ascribed to the formation of well-controlled nanocomposite structures of hard magnetic FePt nanoparticles surrounded by nonmagnetic Ag metal, under the influence of templating by self-assembled BCP nanostructures.
4. Conclusions
Using self-assembled BCP templates, we synthesized a high-coercivity FePt–Ag nanocomposite magnet consisting of FePt nanoparticles surrounded by a matrix of Ag. XRD analysis revealed that the introduction of Ag did not significantly affect the crystalline ordering of the FePt. The coercivity increased by 53% (from 11.1 to 17.0 kOe) owing to the addition of Ag.
Conflicts of Interest
The authors declare no competing financial interests.
Acknowledgments
The authors greatly appreciate assistance from Ms. A. Mori with chemical synthesis and the transmission electron microscopy.