Alignment Of Fept

IEEE TRANSACTIONS ON MAGNETICS, VOL. 41, NO. 2, FEBRUARY 2005

665

Chemically Synthesized FePt Nanoparticle Material for Ultrahigh-Density Recording Hiroyoshi Kodama, Satoru Momose, Toshio Sugimoto, Takuya Uzumaki, and Atsushi Tanaka

Abstract—We have examined the magnetic anisotropy of the “heat-treated FePt nanoparticles” annealed in a magnetic field. The magnetic easy axis of the “heat-treated FePt nanoparticles” is found to be three-dimensional (3-D) random and a partial ordering fct structure is observed before annealing in the presence of a magobtained is 0.5. After annealing netic field. The value of in the presence of a magnetic field, the – loop indicates that the easy axis is oriented preferably in the perpendicular direction than along the in-plane direction. The value of ( ) ( ) at 10 K is 0.62 (1410 Oe/2250 Oe). The value of ( ) is 0.58 at 10 K larger than the value of ( ). Therefore, a weak magnetic easy axis orientation is fundamentally possible on the chemically synthesized FePt nanoparticles. We have studied the recording characteristics of a 3-D random nanoparticle medium using a GUZIK spinstand and observed the recorded patterns for the medium by imaging with a magnetic force microscopy. Index Terms—Easy axis, heat-treated FePt nanoparticles, magnetic field, – loops, surface reduction.

I. INTRODUCTION

S

ELF-ASSEMBLED FePt nanoparticles are promising materials for future ultrahigh-density magnetic recording media, due to their high anisotropy and their ability to form small and uniform grains [1]. Self-organized magnetic array (SOMA) media can serve as: 1) conventional media with reduced dispersions; 2) bit-patterned media with bit-transitions defined by rows of particles; and 3) single-particle-per-bit recording [2]. In the case of 3), the recording density reaches 40 Tbit/in according to a simple estimation if one bit can be recorded in a single particle which is separated from the nearest erg/cm , , bit by 4 nm ( nm). Some important challenges in using the FePt nanoparticles for recording media are finding suitable deposition technique, preventing the coalescence and inducing the perpendicular magnetic anisotropy. The Langmuir–Blodgett method [3], the polymer-mediated method [4] and the [3-(2-minoethlyamino) propyl] trimethoxysilane (APTS) method [5] are already reported for depositing nanoparticles. We have proposed a spin-coating technique for fabricating the FePt nanoparticles media [6]. Our method was shown to be capable of making FePt self-assemble on an entire 2.5-in disk substrate. Moreover, the number of

Manuscript received June 14, 2004. The authors are with Fujitsu Laboratories Ltd., Atsugi, Kanagawa 243-0197, Japan (e-mail: [email protected]). Digital Object Identifier 10.1109/TMAG.2004.838050

Fig. 1.

Our idea of controlling the easy axis of the FePt nanoparticle.

layers of FePt nanoparticles can be controlled with the proposed method. The FePt nanoparticles, as synthesized, have a chemically disordered face-centered cubic (fcc) crystalline structure and are superparamagnetic at 300 K. After annealing at over 500 C, the FePt nanoparticles were transformed into the face-centered crystal structure. This annealing process tetragonal (fct) could lead to nanoparticles coalescence because of its high temperature [7], [8]. We have solved this problem by adding stabilizer (oleic acid and oleyl amine) into FePt solution and have -FePt structure [9]. confirmed the To have practical applications, the orientation of the easy axis of the FePt nanoparticles must be controlled as well. This appears to be one of the most difficult and unsolved problems today. Previously, we have examined the magnetic anisotropy of the fcc FePt nanoparticles annealed in a magnetic field. However, the magnetic easy axis of the fcc FePt nanoparticles is found to be three-dimensional (3-D) random. In this paper, we examine the heat-treated FePt nanoparticles, annealed in the presence of a magnetic field and its effect on improving the degree of easy axis orientation. The read-write (R/W) measurements and imaging with a magnetic force microscopy (MFM) was also carried out on the 3-D random FePt nanoparticle medium in order to check the magnetization switching in the medium.

II. CONCEPT OF INDUCING ANISOTROPY Fig. 1 shows our idea of controlling the easy axis of the FePt nanoparticle. It was thought that easy axis can be controlled by an external magnetic field if it was fct-FePt which had been dispersed in hexane. It is expected that the c-axis align along the external magnetic field if the nanoparticle was able to move freely in the stabilizer.

0018-9464/$20.00 © 2005 IEEE

666

IEEE TRANSACTIONS ON MAGNETICS, VOL. 41, NO. 2, FEBRUARY 2005

Fig. 2. HR-SEM image of heat-treated Fe Pt nanoparticles deposited on thermally oxidized silicon substrate using simple deposition method.

?

MH

Fig. 4. Perpendicular ( ) and in-plane (//) – loops of the heat-treated FePt nanoparticles measured at (a) 10 K and (b) 300 K using SQUID magnetometer. Fig. 3.

XRD patterns of (a) as-synthesized and (b) heat-treated nanoparticles.

III. PROPERTY OF HEAT-TREATED FePt NANOPARTICLE A. Synthesis of FePt Nanoparticles The high-temperature solution phase decomposition of Fe CO and reduction of Pt acac was used to produce mono-dispersed FePt nanoparticles [1]. After synthesis, the FePt nanoparticles were dispersed in hexane. The particle size and dispersion were observed using transmission electron microscopy (TEM) and high-resolution scanning electron microscopy (HR-SEM). B. Preparation of Heat-Treated FePt Nanoparticle The preparation of the heat-treated Fe Pt nanoparticles was carried out using the surface reduction and the heat-treatment after synthesis and the dispersion of the products in hexane. After synthesis, we added oleyl amine (5 L), octyl ether (1 mL) and sodium hydride (NaH) and heated the mixture to about 300 C for 30 min. Fig. 2 shows the HR-SEM image of heat-treated FePt nanoparticles deposited on thermally oxidized silicon substrate. Though the resolution of this image was not too good, we have confirmed the self assembly of nanoparticles. Fig. 3 shows the X-ray diffraction (XRD) (Rigaku-ATX-G) spectrum, using CuK radiation, of as-synthesized and after heat-treated films. The heat-treated FePt nanoparticles indicated a partial ordering face centered tetragonal (fct) FePt structure inferred from the shift of (111) diffraction peak to the wide angle side. We believe that the oxide layer on the surface

of the nanoparticle is reduced by NaH and this reduction acts as a trigger for the ordering. Moreover, the (111) diffraction peak from the heat-treated FePt nanoparticles is narrower than that of the as synthesized particles. One of the reasons for the smaller width of the peak is that the average grain size of heat-treated FePt nanoparticle is relatively larger than that of as synthesized particles. A shoulder peak appears on the low angle degrees of (111) peak for the heat-treated film. side This peak could not be assigned to any of the known diffraction peaks from FePt, Fe, Pt Fe O , Fe O , or from possible impurities. Fig. 4 shows the in-plane and perpendicular – loops of the heat-treated FePt nanoparticles measured at 10 K and at 300 K using a superconducting quantum interference device (SQUID) magnetometer. The in-plane and perpendicular coercivity values at 10 K were 1900 and 1890 Oe, respectively. From Figs. 3 and 4 it is clear that the magnetic easy axis of the heat-treated FePt nanoparticles is found to be 3-D random. A partial ordering fct phase is also present in the sample. The and were 0.5 at 10 K. It values of indicates that each nanoparticle was isolated magnetically. C. Annealing Procedures A thermally oxidized silicon wafer was cut into 3.4-mmsquare sizes using dicing saws (DISCO) and were used as the substrates. The heat-treated FePt nanoparticles were deposited onto this substrate, without applying any external magnetic field. The sample was set in furnace (maximum temperature is 400 C) and the atmosphere in the chamber was changed to

KODAMA et al.: CHEMICALLY SYNTHESIZED FePt NANOPARTICLE MATERIAL FOR ULTRAHIGH-DENSITY RECORDING

667

Fig. 6. XRD patterns of heat-treated nanoparticle after annealing in the magnetic field.

MH

Fig. 5. Perpendicular and in-plane – loops of the heat-treated FePt nanoparticles measured at (a) 10 K and at (b) 300 K after annealing in the presence of a magnetic field using SQUID magnetometer.

Ar H (3%) gas mixture at 1 atm pressure. The magnetic field of 50 kOe was applied along the perpendicular direction of the film using superconducting quantum magnet. The reason of choosing the above mentioned maximum annealing temperature is to prevent the agglomeration of heat-treated FePt nanoparticle. If agglomeration occurs, there is a possibility of formation of chained particles which may induce the shape anisotropy. The sample was annealed continuously by 2 step process, at 200 C for 30 min and at 400 C for 30 min. The magnetic field was applied during the annealing and cooling. In-plane and perpendicular – loops were measured at 10 K and 300 K with SQUID magnetometer. D. Perpendicular Magnetic Anisotropy Fig. 5 shows the perpendicular and in-plane – loops of the heat-treated FePt nanoparticles measured at 10 K and 300 K, after annealing in the presence of a magnetic field. The magnetic moment was normalized with the value of moment at 50 kOe. The – loop indicates that the easy axis is oriented preferably in the perpendicular direction than along the in-plane diis 0.58 at 10 K lager than the rection. The value of . The above observations indicate that the value of sample possess a magnetic anisotropy along the perpendicular direction. This direction is the same as the direction of the magnetic field during the annealing. The value of at 10 K is 0.62 (1410 Oe/2250 Oe). From the cross-section TEM

image, we have confirmed that the FePt nanoparticles are isolated and not a chain of spheres, after annealing in the magnetic field. Thus, a weak magnetic easy axis orientation is fundamentally possible on the chemically synthesized FePt nanoparticles media. We believe the origin of the magnetic anisotropy due to the crystal orientation of fct-phase. However, value is to be reduced to 0.1 or less to use it as a perpendicular recording media. We have analyzed the crystal structure of the heat-treated FePt nanoparticles which were annealed in a magnetic field using XRD. Fig. 6 shows the XRD spectrum, using CuK radiation after annealing in the magnetic field. Initial analysis of the samples annealed in a magnetic field indicated a partial ordering fct FePt structure inferred from the shift of (111) diffraction peak to the wide angle side. Moreover, a very weak (001) peak is observed after annealing in the magnetic field. However, the peak intensity of (001) diffraction was very small and noisy. Since the sample (substrate) size was small (3.4 mm ), we could not make a reliable structural characterization. The evidence of improved crystal anisotropy is under investigation. The magnetic anisotropy of our samples after annealing in a magnetic field is very weak. This may be due to the partial ordering of the heat-treated FePt nanoparticles. We believe the origin of perpendicular magnetic anisotropy in the annealing in the magnetic field is due to the improvement of crystal orientation of fct-phase. By further improving the ordering of the nanoparticle, annealed in a magnetic field, we expect a stronger perpendicular magnetic anisotropy. IV. READ-WRITE MEASUREMENT Our initial trials using GUZIK tester to carry out the R/W test on the FePt nanoparticle media was difficult due to the smaller [6]. read out signals and due to larger media noise and higher R/W demonstration of FePt nanoparticle have been performed using a static R/W tester by Sun et al. [1]. The thickness of the media used was 120 nm and read-out signal detected up to 197 kFCI (5000 flux change/mm). We think that it is necessary to perform R/W measurement using a thinner film of FePt nanoparticles using a spinstand to demonstrate the feasibility of

668

Fig. 7.

IEEE TRANSACTIONS ON MAGNETICS, VOL. 41, NO. 2, FEBRUARY 2005

M –H loop of R/W media measured at 300 K.

high density recording. Magnetic properties of the nanoparticle media were adjusted to achieve a detectable read-out signal. The high-temperature solution phase decomposition of Fe CO and reduction of Pt acac was used to produce mono-dispersed FePt nanoparticles. We did not use the heat-treated nanoparticle for recording performance studies. The FePt nanoparticle was deposited on thermally oxidized silicon substrate using a spin-coating method [6]. The media structure used for the study is carbon (4.5 nm)/Fe Pt -nanoparticle (20 nm)/CrMo (4 nm)/SiO -Si (2.5-in disk substrate). Before annealing, the average nanoparticle diameter is found to be 4.4 nm, and the is 8% obtained from the TEM grain size mean dispersion micrographs. The media was annealed at 700 C for 30 min in vacuum without the presence of an external magnetic field. Fig. 7 shows the in-plane – loop at 300 K of this media. The magnetization of this media is found to be 3-D random. The in-plane coercivity of the media is 2200 Oe measured using a SQUID. The saturation field of the – loop of this media is about 20 kOe. The value is 0.56 which indicates a weak magnetic interaction of each nanoparticle. R/W measurements were carried out using a GUZIK 1601 RWA (upgrade 320 Mb/s) and LS90S spinstand. The write head used is a ring-type head, and the read head uses the GMR sensor. mm. The flying The rotational speed was 5725 rpm at height was about 20 nm. Fig. 8 shows the track average amplitude (TAA) and the overwrite (O/W) characteristics with different writing currents for the nanoparticle media. The low-frequency (LF) and high-frequency (HF) linear densities were set at 36.8 kFCI and 294 kFCI, respectively. The O/W value is improved from 16 dB with the increase in the write current and is about 26 dB at write currents greater than 60 mA. The O/W value is considerably improved from the initial R/W experiments on nanoparticle media [6]. The TAA output voltage is improved for both LF and HF with the write current, and the improvement being very significant for the former case. However, LF is not saturated even at a higher write current of 60 mA. The maximum TAA output voltage at LF reached on about 1 mV. The thickness of this media is found to be 20 nm from the

Fig. 8. The saturation characteristics of FePt nanoparticle media. The linear densities of LF (square) and HF (circle) were set at 36.8 kFCI and 294 kFCI.

Fig. 9. The MFM image of write patterns of nanoparticle media. Scan area of this image is 10 m.

cross-section TEM observation. By increasing the packing densities, we believe the output voltage level and hence signal to noise ratios could be increased as well. MFM imaging was carried out with the system from digital instruments (Nanoscope III). Fig. 9 shows the MFM image of the written transitions on the above mentioned medium at different recording densities. The write current used for this purpose is 80 mA. The linear densities used were 20, 40, 60, 80, 100, 120, 140, 160, 200, 250, 300, and 400 kFCI. The recorded pattern was clearly observed up to at 250 kFCI. It is possible to image the recorded bit patterns up to 400 kFCI. The recorded bits at 300 and 400 kFCI are not well resolved in comparison to the patterns at lower densities. We believe the larger medium noise at high densities and larger dependence on the 3-D random magnetization makes it difficult to image the high density patterns. Moreover, the write field is not sufficient enough to enable switching of all the magnetic clusters in a single bit. This is partially due to the lower exchange between the nanoparticles. V. CONCLUSION We have examined the magnetic anisotropy of the heat-treated FePt nanoparticles annealed in a magnetic field. A weak magnetic easy axis orientation in the direction of the

KODAMA et al.: CHEMICALLY SYNTHESIZED FePt NANOPARTICLE MATERIAL FOR ULTRAHIGH-DENSITY RECORDING

applied magnetic field was obtained. We believe that the reason for the observed anisotropy may be due to the presence of a partial ordering fct FePt phase. We have measured the R/W properties of low anisotropic, 3-D random FePt nanoparticle media. The recording patterns were imaged using MFM and bit patterns up to 400 kFCI could be observed. Future work involves the recording performance of the FePt nanoparticle media using heat assisted magnetic recording (HAMR), and to study the noise mechanisms and magnetic cluster sizes in these media. ACKNOWLEDGMENT The authors would like to thank Dr. A. Ajan of Fujitsu Laboratory for fruitful discussions and suggestions. REFERENCES [1] S. Sun, C. B. Murray, D. Weller, L. Folks, and A. Moser, “Monodisperse FePt nanoparticles and ferromagnetic FePt nanocrystal superlattices,” Science, vol. 287, pp. 1989–1992, Mar. 2000.

669

[2] T. Klemmer and D. Weller, “SOMA and nanomagnetics for ultra-high density storage,” presented at the 7th Perpendiculor Magnetic Recording Conf. (PMRC2004), Sendai, Japan, May 2004, Paper 02pC-04. [3] Q. Guo, X. Teng, S. Rahman, and H. Yang, “Patterned Langmuir-Blodgett films of monodisperse nanoparticles of iron oxide using soft lithography,” J. Amer. Chem. Soc., vol. 125, pp. 630–631, 2003. [4] S. Sun, S. Ansers, H. F. Hamann, J.-U. Thiele, J. E. E. Baglin, T. Thomson, E. E. Fullerton, C. B. Murray, and B. D. Terris, “Polymer mediated self-assembly of magnetic nanoparticles,” J. Amer. Chem. Soc., vol. 124, pp. 2884–2885, 2002. [5] A. C. C. Yu, M. Mizuno, Y. Sasaki, M. Inoue, H. Kondo, I. Ohta, D. Djayaprawira, and M. Takahashi, “Fabrication of monodispersive FePt nanoparticle films stabilized on rigid substrate,” Appl. Phys. Lett., vol. 82, pp. 4352–4354, Jun. 2003. [6] H. Kodama, S. Momose, N. Ihara, T. Uzumaki, and A. Tanaka, “Disk substrate deposition techniques for monodisperse chemically synthesized FePt nanoparticle media,” Appl. Phys. Lett., vol. 83, pp. 5253–5255, Dec. 2003. [7] H. Zeng, S. Sun, T. S. Vedantam, J. P. Liu, Z.-R. Dai, and Z.-L. Wang, “Exchange-coupled FePt nanoparticle assembly,” Appl. Phys. Lett., vol. 80, pp. 2583–2585, Apr. 2002. [8] S. Kang, J. W. Harrell, and D. E. Nikles, “Reduction of the fcc to L1 ordering temperature for self-assembled FePt nanoparticles containing Ag,” Nano Lett., vol. 2, pp. 1033–1036, Oct. 2002. [9] S. Momose, H. Kodama, N. Ihara, T. Uzumaki, and A. Tanaka, “L1 -FePt nanoparticles in a magnetically isolated state,” Jpn. J. Appl. Phys., vol. 42, pp. L1252–L1254, Oct. 2003.