ARTICLE IN PRESS

Journal of Crystal Growth 266 (2004) 500–504

Magnetic properties improvement in Fe3O4 nanoparticles grown under magnetic fields Jun Wang, Kai Zhang, Zhenmeng Peng, Qianwang Chen* Department of Materials Science & Engineering, Structure Research Laboratory, University of Science & Technology of China, Hefei 230026, China Received 27 December 2003; accepted 5 March 2004 Communicated by T. Hibiya

Abstract Magnetite nanoparticles have been synthesized under magnetic fields by the method of co-precipitation. The asprepared samples were characterized by X-ray diffraction and transmission electron microscopy, and their magnetic properties were evaluated on a vibrating sample magnetometer. It is shown that the Fe3O4 samples produced by coprecipitation method under a magnetic field of 1 T have much higher saturation magnetization (15.3 emu/g) than that without magnetic field (6.59 emu/g). It is suggested that the well-crystallized Fe3O4 grains formed under magnetic fields should be responsible for the improved magnetic properties of nanosized Fe3O4 powders. r 2004 Elsevier B.V. All rights reserved. Keywords: A1. X-ray diffraction; B1. Nanomaterials; B2. Magnetic materials

1. Introduction Over the past several years, the preparation and characterization of nanoscale magnetic materials have attracted much attention as the nanomaterials would allow investigating the fundamental aspects of magnetic-ordering phenomena in magnetic materials with reduced dimensions and could lead to new potential applications such as data storage technology [1]. However, it is reported that the saturation magnetization of nanoparticles was much lower than that of correspondent bulk sample and decreased along with the reduction of *Corresponding author. Tel./fax: +86-551-3607292. E-mail address: [email protected] (Q. Chen).

the particle size [2–6]. These would hinder some applications of nanosized magnetic materials in some technique fields, such as high-density perpendicular magnetic and magnetooptic recording media. Nanoparticles exhibit a large surface to volume ratio and experimental evidences for surface spin disorder have been reported in several previous researches [7–9]. A model of nonmagnetic surface layer was further proposed to interpret anomalous magnetic properties of ferrite nanoparticles [6,10]. Additionally, the magnetic properties of the magnetic nanoparticles are related to the crystallinity [11]. Therefore, it is essential to find an available synthesis method to improve crystallinity of magnetic nanoparticles for studying and

0022-0248/$ - see front matter r 2004 Elsevier B.V. All rights reserved. doi:10.1016/j.jcrysgro.2004.03.034

ARTICLE IN PRESS J. Wang et al. / Journal of Crystal Growth 266 (2004) 500–504

improving the magnetic properties of magnetic nanomaterials. Because of the unique physical properties of magnetic materials, magnetic field is recognized not only as a parameter for studying the physical properties but also as a tool to control the microstructure and morphology of the materials during magnetic materials synthesis [12]. For instance, in our previous experiment, we utilized magnetic fields to synthesize Fe3O4 nanowires [13]. Arborescent shape aggregates of Fe were obtained in electrochemical growth of iron under a 0.6 T magnetic field [14], and Matsuda et al. reported the Co-ferrite produced by co-precipitation method in a magnetic field have a better crystallinity than that produced without magnetic field [15], and de Rango et al. found that the crystal orientation occurred in the melt growth process of YBCO in magnetic fields [16]. So, the crystallinity of the Fe3O4 formed in a magnetic field is expected to change with varying magnetic field magnitude. In this paper, we exhibit the effects of magnetic field on growth of the Fe3O4 nanoparticles during synthesis processes and discuss the relationship between the magnetic properties and magnetic field in the magnet range (0–1 T). It is expected that this process could also be a promising technique to improve the magnetic properties of other magnetic materials.

501

reaction was measured using digital pH meter. All the chemicals used (FeCl2  4H2O, FeCl3  6H2O and NaOH) were analytical grade. Distilled water was used for the preparation of aqueous solution with 5 mM Fe(II) ions and 10 mM Fe(III) ions. To prepare aqueous solution of Fe(II) ions, distilled water was degassed with nitrogen gas by passing through it. During the experiment using the Fe(II) solution, nitrogen gas was kept passing through the solution to prevent the oxidation of the solution. 2.2. Procedure The nitrogen gas passed continuously through the reactor in order to stir the solution during reaction. The pH of reaction solution was adjusted with the 0.5 M NaOH solution, and the reaction vessel was placed in the center of magnetic field. The reaction suspension was prepared by adding alkaline solution (0.5 M NaOH solution) into the degassed the Fe(II)–Fe(III) solution. The Fe3O4 formation reactions were carried out for 10 min in magnetic fields. The products of the Fe3O4 particles were collected by magnetic separation, and washed with distilled water several times. After drying in vacuum oven at 25
C for 12 h, the products were weighed, and their phases were identified by XRD. The four syntheses are carried out under the same parameters except the strength of the magnetic field (0, 0.4, 0.6, 1 T).

2. Experiment 2.1. Apparatus and chemicals

3. Results and discussion

To apply the external magnetic field, a 1 T electromagnet was used. The samples, which were recorded at a scanning rate of 0.05
per second with the 2y range from 10
to 70
, were characterized by X-ray diffraction (XRD) using an X-ray diffractometer with high-intensity CuKa ( radiation (l ¼ 1:5418 A). Transmission electron microscopy (TEM) images were taken with a Hitachi model H-800 transmission electron microscope, using an accelerating voltage of 200 kV. The magnetic properties were measured on a BHV-55 vibrating sample magnetometer (VSM) at room temperature. The pH of the solution during the

Fig. 1 shows XRD patterns of the co-precipitates of iron ions obtained under magnetic field 0, 0.4, 0.6, 1 T, respectively. As can be seen in Fig. 1, all the diffraction peaks can be indexed by the cubic structure of Fe3O4 (JCPDS card no. 85– 1436). No peaks for impurities are detected. However, the XRD pattern cannot provide enough evidence to confirm the formation of Fe3O4, since there is little difference between the XRD patterns of Fe3O4 and that of g-Fe2O3. The black color of the samples suggests the formation . of Fe3O4. The Mossbauer spectrum (not shown) consists of two hyperfine magnetic sextets, one for

ARTICLE IN PRESS 502

J. Wang et al. / Journal of Crystal Growth 266 (2004) 500–504

Fig. 1. XRD patterns for samples derived by a co-precipitation process under magnetic fields at 0 T (A), 0.4 T (B), 0.6 T (C), and 1 T (D), respectively.

Fe3+ at tetrahedral sites and the other for the mixed valence Fe2.5+ at octahedral sites. This result also indicates that the sample is not g-Fe2O3 but Fe3O4 [17]. The reflection peaks (Fig. 1) become sharper and narrower along with the increasing of magnetic field strength, indicating the improvement of crystallinity. The particle sizes were calculated using the Debye–Scherrer formula [18] from the reflection peak of (3 1 1). The average particle sizes of Fe3O4 prepared under magnetic field (0, 0.4, 0.6 and 1 T) are 4, 5, 9 and 10 nm, respectively. The results reveal that the particle size increases gradually and crystallinity is improved along with the increasing strength of magnetic field, which are consistent with the results measured from TEM images. As shown in Fig. 2, the shape of particles derived by a co-precipitation process under magnetic fields at 1 T is quadrangle in sample B, while quasi-sphere polyhedron in sample A. The insets of Fig. 2 show electron diffraction patterns (ED) from the two samples using the electron diffraction technique in conjunction with TEM measurements. The ED patterns disclose distinct difference of crystallinity of two samples and prove that synthesis under magnetic field could produce nanoscale powders consisted of well-crystallized Fe3O4 particles. The

Fig. 2. TEM micrograph of two samples derived by coprecipitation process under magnetic fields at 0 T (A), and 1 T (B), respectively. The insets show the electron diffraction patterns of two samples, respectively.

Fig. 3. Magnetic hysteresis curves for the samples measured at room temperature. The samples were obtained under magnetic fields at 0 T (A), 0.6 T (B) and 1 T (C), respectively.

same result that application of a magnetic field during the process can improve crystallinity of sample has been reported in other papers [15,19–21]. The magnetic properties for Fe3O4 obtained at different strength of magnetic field are shown in Fig. 3. The sample derived under magnetic field (1 T) exhibits saturation magnetization (Ms) of B15.3 emu/g, which is much higher compared

ARTICLE IN PRESS J. Wang et al. / Journal of Crystal Growth 266 (2004) 500–504

with that of the sample obtained under magnetic field (0 and 0.6 T). It is found that the increase tendency of Ms is consistent with the improvement of crystallinity. The particle size has been reported to influence the magnetic properties of materials. The crystallite size, however, should not be mainly responsible for the saturation magnetization difference observed in the above samples. Firstly, because the sample derived under 1 T magnetic field, consisting of particles with average size of 10 nm, possesses saturation magnetization (Ms ) of B15.3 emu/g, significantly higher than that of the sample with average size of 9 nm under 0.6 T magnetic field. Secondly, average particles size of all samples is below the critical size of Fe3O4 particles (single domain: B54 nm [22]). Furthermore, it is reported that Fe3O4 nanoparticles samples with same size (10 nm) only possess saturation magnetization (Ms ) of B1.25 emu/g at room temperature [23]. Combined with their XRD and TEM results, we suggest that their difference in crystallinity aroused the significant difference of their magnetic properties of the above two samples. For nanoparticles, the main reasons resulting in decrease of Ms may come from the nature of ultrafine particles, surface disorder and cation distribution, etc. For example, surface disorder layer could serve as a non-magnetic layer and decrease saturation magnetization. On the other hand, Fe3O4 crystallizes in cubic structure with iron in two valence states. The Fe3O4 can be written in the form of FeO,  Fe2O3 with Fe(II) as FeO and Fe(III) as Fe2O3. The Fe(III) occupies the tetrahedral sites, and half the octahedral sites, with the Fe(II) occupying the other half. The magnetic moments on the octahedral sites are antiferromagnetic, while on the tetrahedral sites they are ferro-magnetically aligned. The differences in crystallization process could influence the distribution of Fe(II)-octahedral and Fe(III)-octahedral and then the superexchange interaction between ferric ions. The well-crystallized particles have a thinner non-magnetic surface layer and less superamagnetic relaxation, which can be used to explain the increase of saturation magnetization in hydrothermal derived particles, and to interpret mag-

503

netic properties variation from different conditions in hydrothermal process because they would lead to different crystallinity of particles. And, better crystallinity in Fe3O4 nanoparticles would result in less crystal defects. Therefore, the superexchange interaction of Fe–O–Fe was strengthened, and the magnetic properties of Fe3O4 nanoparticles were improved. Indeed, strong correlation between the crystallinity and the magnetic properties has been reported in barium ferrite powder [24,25]. For example, Chen et al. [24] found that the improved crystallinity of the magnetic materials, not the particle size increasing, is the main contributing factor to the observed increase of MS and HC in BaFe12O19. However, the mechanism of the magnetic field effects on the crystal and microstructures of magnetic materials are not clear yet.

4. Conclusions Growth of Fe3O4 nanoparticles under magnetic fields was investigated. XRD and TEM studies reveal that higher strength of magnetic field in the process tends to result in formation of larger grains with better crystallinity. The magnetic measurements show that the powders obtained under magnetic field at 1 T possess excellent magnetic properties with saturation magnetization of 15.3 emu/g, much higher than 6.59 emu/g for the nanoparticles formed without a magnetic field applied. The improved magnetic properties were suggested to be originated from the thinner nonmagnetic layer and strengthened superexchange interactions of Fe–O–Fe due to magnetic field making Fe3O4 nanoparticles well crystallinity. The results reveal a possibility that magnetic properties of the magnetic materials can be improved through applying a magnetic field during processes of synthesis.

Acknowledgements This work was supported by the National Natural Science Foundation under the contract nos. 20125103 and 90206034.

ARTICLE IN PRESS 504

J. Wang et al. / Journal of Crystal Growth 266 (2004) 500–504

References [1] (a) S. Sun, C.B. Murray, D. Weller, L. Folks, A. Moser, Science 287 (2000) 1989; (b) V. Skumryev, S. Stoyanov, Y. Zhang, G. Hadjipanayis, D. Givord, J. Nogues, Nature 423 (2003) 850. [2] G. Visalakshi, G. Venkataswaran, S.K. Kulshreshtha, P.N. Moorthy, Mater. Res. Bull. 28 (1993) 829. [3] S.Z. Wang, H.W. Xin, Y.T. Qian, Mater. Lett. 33 (1997) 113. [4] D. Vollath, D.V. Szabo, J. Mater. Res. 12 (1997) 2175. [5] Z.H. Zhou, J. Wang, X. Liu, H.S.O. Chan, J. Mater. Chem. 11 (2001) 1704. [6] J.P. Chen, C.M. Sorensen, K.J. Klabunde, G.C. Hadjipanayis, E. Delvin, A. Kostikas, Phys. Rev. B 54 (1996) 9288. [7] H.G. Zheng, J.H. Zeng, H.M. Yu, J. China Univ. Sci. Technol. 29 (1999) 722. [8] R.H. Kodama, A.E. Berkowitz, E.J. McNiff Jr., S. Foner, Phys. Rev. Lett. 77 (1996) 394. [9] L. Rezlescu, E. Rezlescu, P.D. Popa, N. Rezlescu, J. Magn. Magn. Mater. 193 (1999) 288. [10] D.H. Han, J.P. Wang, H.L. Luo, J. Magn. Magn. Mater. 136 (1994) 176. [11] X.Y. Qi, B.J. Stadler, J.H. Judy, R.H. Victora, J. Appl. Phys. 93 (2003) 7849. [12] A.P. Chiriac, C.I. Simionescu, J. Polym. Sci. 34 (1996) 567.

[13] J. Wang, Q.W. Chen, C. Zeng, B.Y. Hou, Adv. Mater. 16 (2004) 137. [14] S. Boden, L. Vignon, M.R.P. Ballou, Phys. Rev. Lett. 83 (1999) 2612. [15] K. Matsuda, Y. Kawada, N. Shima, T. Makino, I. Kayama, Nippon Kagakkai-shi 6 (1981) 947. [16] P. de Rango, M. Lee, P. Lejay, A. Sulpice, R. Tournier, M. Ingold, P. Germi, M. Pernet, Nature 349 (1991) 770. [17] R. Bauminger, S.G. Cohen, A. Marinov, S. Ofer, E. Segal, Phys. Rev. 122 (1961) 1447. [18] P. Gallezot, Catalysis Science and Technology, Vol. 5, Springer, Berlin, Heidelberg, New York, 1983. [19] Y.W. Ma, K. Watanabe, S. Awaji, M. Motokawa, J. Crystal Growth 233 (2001) 483. [20] S. Awaji, Y. Ma, W.P. Chen, H. Maeda, K. Watanabe, M. Motokawa, Curr. Appl. Phys. 3 (2003) 391. [21] Y.W. Ma, K. Watanabe, S. Awaji, M. Motokawa, Physica C 357 (2001) 337. [22] A. Aharoni, J.P. Jakubovics, IEEE Trans. Magn. 24 (1988) 1892. [23] R. Vijayakumar, Y. Koltypin, I. Felner, A. Gedanken, Mater. Sci. Eng. A 286 (2000) 101. [24] M.S. Chen, Z.X. Shen, X.Y. Liu, J. Wang, J. Mater. Res. 15 (2000) 483. [25] C.A. Barrero, R.E. Vandenberghe, E. De Grave, Hyperfine Interactions 122 (1999) 39.