Atomic Composition Effect In Fept

APPLIED PHYSICS LETTERS

VOLUME 85, NUMBER 25

20 DECEMBER 2004

Atomic composition effect on the ordering of solution-phase synthesized FePt nanoparticle films Andrew C. C. Yu,a) M. Mizuno, Y. Sasaki, and H. Kondo Sony Corporation, Sendai Technology Center, 3-4-1 Sakuragi, Miyagi 985-0842, Japan

(Received 6 July 2004; accepted 19 October 2004) FePt nanoparticle monolayer films were fabricated with the nanoparticles stabilized on organic-coupling-layer coated Si substrates. The as-prepared films were nonmagnetic. In order to transform the nanoparticle phase from chemically disordered face-centered-cubic to chemically ordered L10, the films were annealed at 800° C under nitrogen atmosphere for 30 min. The annealed films showed different degrees of sintering depending on the organic coupling layer materials used. At room temperature, sintered films exhibited high coercivity up to 2.4⫻ 104 Oe, while monodispersive films with insignificant sintering showed a low coercivity of 1.1⫻ 102 Oe. Such results can be explained by a large atomic composition distribution among the nanoparticles together with the size effect on ordering. Improvement of the nanoparticle atomic composition homogeneity is essential for applying the nanoparticle films for future ultrahigh-density data storage applications. © 2004 American Institute of Physics. [DOI: 10.1063/1.1835998] High anisotropy L10 type nanoparticles, such as FePt and CoPt, have attracted much attention due to their potential for the development of not only ultrahigh-density magnetic recording media, but also various novel memory devices, nanotechnological designs, and bio-medical applications.1–5 It is important to fabricate the nanoparticle heterostructures on the substrates of interest that are either rigid or flexible in nature in order to apply the heterostructures for various commercial product development. We have reported a technique for stabilizing monodispersive chemically synthesized nanoparticle films on Si wafer substrates by means of introducing organosilane coupling layers deposited between the substrates and the nanoparticle dispersions.6 As the assynthesized nanoparticles were nonmagnetic with chemically disordered fcc phase, annealing should therefore be carried out in order to transform the nanoparticles to chemically ordered L10 phase such that the nanoparticles will become magnetic.1–3,7 Recently, we have observed that with the application of different coupling layer materials, the annealed monodispersive nanoparticle films exhibited different degrees of nanoparticle sintering resulting in showing significantly different coercivity values for the films. The aim of this letter is to report a chemical analysis investigation of the nanoparticles in order to understand the atomic composition effect on the coercivity variation of the films that show different degree of sintering. FexPt100−x nanoparticles were chemically synthesized using a similar method reported by Sun et al.8 In order to transform the FexPt100−x nanoparticles from nonmagnetic chemically disordered fcc phase to magnetic chemically ordered L10 phase, the nanoparticle films were annealed at 800° C for 30 min under N2 atmosphere (note: the annealing chamber was first evacuated to around 10−6 Torr before fluxing in N2 aimed at avoiding serious oxidation issue on the nanoparticles during annealing). We observed that with the application of different coupling layer materials, the annealed nanoparticle films exhibited different degrees of sina)

Author to whom correspondence should be addressed; electronic mail: [email protected]

tering resulting in different sintered particle sizes and size distribution.9 Here we compared two representative series of annealed samples (note: same FexPt100−x nanoparticle dispersion but different coupling layers were used for the fabrication of the two samples), one with serious sintering issue (sample 1) while the other one showed insignificant sintering (sample 2). Coercivity, Hc, of the nanoparticle films was measured using a superconducting quantum interferometer device. High resolution electron microscopy and high resolution scanning electron microscopy were employed to characterize the nanoparticle structure as well as the nanoparticle film heterostructure. Chemical analyses of the nanoparticles were done by field emission electron nano-beam energy dispersive spectrometry (EDS) and x-ray microanalysis (XMA). Table I summarizes the Hc values, average nanoparticle size, nanoparticle size distribution, and XMA data of the two samples. The order of the Hc value of sample 1 was 102 times higher than that of sample 2 although the XMA data indicated that both samples possessed the same atomic composition. Superparamagnetic effect could be one of the causes for the low Hc value of sample 2 at room temperature, however at 5 K, the Hc value of sample 2 was only 770 Oe, which was still much smaller than the 3.0⫻ 104 Oe Hc value of sample 1 at 5 K. It is therefore expected that some other factors could lead to the difference in the Hc values. Systematic chemical analysis of the nanoparticles using electron nano-beam EDS equipped with a 300 kV field-emission transmission electron microscope was carried out. The elecTABLE I. Summary of Hc, particle size, particle size distribution, and XMA data for samples 1 and 2. Sample Hc共Oe兲 at room temperature Hc共Oe兲 at 5 K Average particle size before annealing 共nm兲 Average particle size after annealing 共nm兲 Particle size distribution after annealing 共␴ / nm兲 XMA

1

2

2.4⫻ 104 3.0⫻ 104 3.0 28 11 Fe51Pt49

1.1⫻ 102 7.7⫻ 102 3.0 3.2 0.86 Fe51Pt49

0003-6951/2004/85(25)/6242/3/$22.00 6242 © 2004 American Institute of Physics Downloaded 13 Jan 2005 to 129.93.16.3. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp

Yu et al.

Appl. Phys. Lett., Vol. 85, No. 25, 20 December 2004 TABLE II. Summary of EDS data for different number of FexPt100−x nanoparticles being detected for each measurement. No. of nanoparticles detected per EDS measurement (order) 104 103 102 1 – 10

Fe (average at.%)

Pt (average at.%)

49.68 49.71 46.55 47.52

50.32 50.29 53.45 52.48

tron beam size was varied in order to measure the average output signals generated by different numbers of nanoparticles, while the smallest nano-beam size of around 5 nm in diameter was used to detect the EDS signal emitted from individual nanoparticles. Table II summarizes the average FexPt100−x atomic compositions with different numbers of nanoparticles detected per EDS measurement (note: each average data was a mean value of over 30–50 raw data points obtained at different locations of the nanoparticle dispersion). According to the average atomic composition data, the atomic composition distribution of the nanoparticles was small while it was reasonably consistent with the XMA result. We then carried out a series of EDS measurements on individual nanoparticles. More than 100 individual nanoparticles were measured from each dispersion sample and a total of five identical samples were examined.10 Based on the experimental data obtained, we found that the atomic composition of each individual nanoparticle was different. Only around 29.0% of the FexPt100−x nanoparticles were within the composition of 40⬍ x ⬍ 60, while approximately 40.5% and 30.5% of the nanoparticles were Pt-rich (i.e., x ⬍ 40) and Fe-rich (i.e., x ⬎ 60), respectively. According to the FexPt100−x phase diagram, only when 40⬍ x ⬍ 60 could the FexPt100−x phase transform to the L10 phase after annealing resulting in high coercivity for the system.11 We selected (from over 500 raw data points) six individual nanoparticles with different atomic compositions as an example to demonstrate the large atomic composition variation in the nanoparticle dispersion, Table III. Particles 1 and 2 were Fe-rich, on the contrary, particles 5 and 6 were Pt-rich. Only particles 3 and 4 fulfilled the phase transformation requirement, but the other four nanoparticles might not be able to transform completely to the L10 phase even after high temperature annealing. Since there existed a large atomic composition distribution among the nanoparticle dispersion, the nanoparticles in the annealed films with insignificant sintering could then have different degrees of L10 phase transformation. There-

6243

fore, in sample 2 only a small coercivity was observed while a broad range of anisotropy field was expected among the nanoparticle dispersion.12 On the other hand, in sample 1 nanoparticles sintered forming different sizes of nanoparticle clusters, thus the atomic compositions of the clusters became the average atomic composition values of the sintered nanoparticles, which could fulfill the 40⬍ x ⬍ 60 condition. Therefore, the sintered films exhibited high coercivity as L10 phase transformation of the sintered nanoparticle clusters was possible. In addition to the atomic composition effect, size effect is also another important factor affecting the ordering of nanoparticles.13–16 It was reported that the fcc phase of 4 nm chemically synthesized FePt nanoparticles was stable up to 560° C as long as coalescence did not occur.13 However, based on our experimental results, we observed a coercivity of 1.1⫻ 102 Oe for the 800° C-annealed nonsintered films (Table I), while the as-prepared films were nonmagnetic at room temperature. The existence of the small coercivity could imply the occurrence of a very small degree of ordering in the nonsintered film, possibly resulting from the relatively high annealing temperature (no coercivity was observed when the films were annealed at 600° C or less). Nevertheless, it is believed that the size effect could significantly (but possibly not completely) hinder the ordering of the nonsintered nanoparticles even for those which had nearly equiatomic composition. Therefore, together with the effect of atomic composition variation, the nonsintered annealed films could not reveal large coercivity as the sintered films did. In conclusion, besides size effect, atomic composition distribution of the nanoparticle dispersion also has significant influence on the fcc to L10 phase transformation. In order to apply L10 FexPt100−x nanoparticles for ultrahigh-density data storage, it is important to control the nanoparticle atomic composition in the range of 40⬍ x ⬍ 60 such that thermodynamical phase transformation of the nanoparticles can easily occur. Meanwhile, a small atomic composition distribution is favorable for the consequent achievement of small bit size made up of only one or just a few nanoparticles. Taking size effect into concern, we suggest that FePt (as well as CoPt) particle size around 6 – 8 nm would be suitable for various magnetic product development. The authors would like to thank M. Takahashi for the provision of magnetic measurement facilities. A.C.C.Y. and K.H. acknowledge partial support of the Nanotechnology Support Project by the Ministry of Education, Culture, Sports, Science and Technology, Japan. 1

S. Sun, E. E. Fullerton, D. Weller, and C. B. Murray, IEEE Trans. Magn. 37, 1239 (2001). 2 D. Weller, A. Moser, L. Folks, M. E. Best, W. Lee, M. F. Toney, M. Schwickert, J.-U. Thiele, and M. F. Dorner, IEEE Trans. Magn. 36, 10 (2000). 3 A. C. C. Yu, M. Mizuno, Y. Sasaki, H. Kondo, and K. Hiraga, Appl. Phys. Nanoparticle Fe (at.%) Pt (at.%) Lett. 81, 3768 (2002). 4 U. Häfeli, W. Schutt, J. Teller, and M. Zborowski, Scientific and Clinical 1 66.89 33.11 Applications of Magnetic Carriers (Plenum, New York, 1997). 2 65.82 34.18 5 D. G. Mitchell, J. Magn. Reson Imaging 7, 1 (1997). 6 3 54.07 45.93 A. C. C. Yu, M. Mizuno, Y. Sasaki, M. Inoue, H. Kondo, I. Ohta, D. 4 46.52 53.48 Djayaprawira, and M. Takahashi, Appl. Phys. Lett. 82, 4352 (2003). 7 S. Kang, J. W. Harrell, and D. E. Nikles, Nano Lett. 2, 1033 (2002). 5 33.24 66.76 8 S. Sun, C. B. Murray, D. Weller, L. Folks, and A. Moser, Science 287, 6 21.92 78.08 1989 (2000). Downloaded 13 Jan 2005 to 129.93.16.3. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp

TABLE III. Atomic composition of six selected FexPt100−x nanoparticles demonstrating the atomic composition variation issue among individual nanoparticles. EDS measurement was carried out on each individual nanoparticle.

6244 9

Appl. Phys. Lett., Vol. 85, No. 25, 20 December 2004

M. Mizuno, A. C. C. Yu, and Y. Sasaki, Langmuir (in press). It is practically very difficult, if not impossible, to measure all the individual nanoparticles in each sample. Indeed, it is believed that the over 500 raw data points obtained could give at least a crude approximation as well as a general image of the atomic composition distribution of the nanoparticle dispersion. 11 T. B. Massalski, H. Okamoto, P. R. Subramanian, and L. Kacprzak, Binary Alloy Phase diagrams (ASM International, Materials Park, OH, 1990). 12 Y. Sasaki, M. Mizuno, A. C. C. Yu, M. Inoue, K. Yazawa, I. Ohta, M. 10

Yu et al. Takahashi, B. Jeyadevan, and K. Tohji, J. Magn. Magn. Mater. 282, 122 (2004). 13 B. Stahl, J. Ellrich, R. Theissmann, M. Ghafari, S. Bhattacharya, H. Hahn, N. S. Gajbhiye, D. Kramer, R. N. Viswanath, J. Weissmuller, and H. Gleiter, Phys. Rev. B 67, 014422 (2003). 14 D. H. Ping, M. Ohnuma, K. Hono, M. Watanabe, T. Iwasa, and T. Masumoto, J. Appl. Phys. 90, 4708 (2001). 15 H. Yasuda and H. Mori, Z. Phys. D: At., Mol. Clusters 37, 181 (1996). 16 R. Poduri and L.-Q. Chen, Acta Mater. 44, 4253 (1996).

Downloaded 13 Jan 2005 to 129.93.16.3. Redistribution subject to AIP license or copyright, see http://apl.aip.org/apl/copyright.jsp