Prevention Of Nano Particle Coalescence Under High Temperature Annealing

Langmuir 2004, 20, 11305-11307

11305

Prevention of Nanoparticle Coalescence under High-Temperature Annealing Mikihisa Mizuno,* Yuichi Sasaki, Andrew C. C. Yu, and Makoto Inoue Sony Corporation, Sendai Technology Center, 3-4-1 Sakuragi, Tagajo, Miyagi 985-0842, Japan Received July 21, 2004. In Final Form: October 4, 2004 An effective method of employing 3-aminopropyldimethylethoxysilane linker molecules to stabilize 4.4 nm FePt nanoparticle monolayer films on a SiO2 substrate as well as to prevent coalescence of the particles under 800 °C annealing is reported. As-deposited FePt nanoparticle films in chemically disordered facecentered-cubic phase transform to mostly chemically ordered L10 structure after annealing, while the nanoparticles are free from serious coalescence. The method may fulfill the pressing need to prevent nanoparticle coalescence under high-temperature annealing for the development of FePt nanoparticle based products, such as ultrahigh-density magnetic recording media and novel memory devices.

Introduction Well-dispersed nanoscale metallic particles coated with organic surfactants prepared via solution-phase synthesis have attracted much attention due to their application potential as fundamental building blocks for a wide range of nanotechnological devices and products, such as active catalysts with large effective surface area,1 bright luminescent quantum dots with size-dependent optical properties,2 ultrahigh-density magnetic recording media,3 highfrequency devices based on ferromagnetic resonance,4 and sensitive markers for biomolecule detection.5 Solutionphase synthesis provides a useful means of producing various nanoparticle systems with tunable size and narrow size distribution. Focusing on ultrahigh-density magnetic recording media and/or memory applications, hard magnetic nanoparticles, such as FePt and CoPt, are excellent candidates.3,6-8 Two of the main technological issues that have to be tackled are nanoparticle stabilization on the substrate of interest and anticoalescence of the particles upon heat treatment. It is highly desirable to stabilize the magnetic nanoparticles homogeneously on extended planar surfaces into thin film form with controlled film thickness and morphology, while the nanoparticles are generally separated from each other.9 Nanoparticles can exhibit high local ordering via self-assembly; however, it is currently difficult to form assemblies with controlled assembly thickness and morphology over a large area by simply depositing the particles on a bare substrate.10,11 A new technique for nanoparticle film fabrication is spontaneous absorption of particles onto a solid surface by means of a linker* To whom correspondence should be addressed. Phone: +81 22 367 2564. Fax: +81 22 367 2783. E-mail: Mikihisa.Mizuno@ jp.sony.com. (1) Son, S. U.; Jang, Y.; Park, J.; Na, H. B.; Park, H. M.; Yun, H. J.; Lee, J.; Hyeon, T. J. Am. Chem. Soc. 2004, 126, 5026. (2) Alivisatos, A. P. Science 1996, 271, 933. (3) Sun, S.; Murray, C. B.; Weller, D.; Folks, L.; Moser, A. Science 2000, 287, 1989. (4) Toneguzzo, P.; Viau, G.; Acher, O.; Fie´vet-Vincent, F.; Fie´vet, F. Adv. Mater. 1998, 10, 1032. (5) Xu, C.; Xu, K.; Gu, H.; Zhong, X.; Guo, Z.; Zheng, R.; Zhang, X.; Xu, B. J. Am. Chem. Soc. 2004, 126, 3392. (6) Yu, A. C. C.; Mizuno, M.; Sasaki, Y.; Kondo, H.; Hiraga, K. Appl. Phys. Lett. 2002, 81, 3768. (7) Chen, M.; Nikles, D. E. Nano Lett. 2002, 2, 211. (8) Shevchenko, E. V.; Talapin, D. V.; Schnablegger, H.; Kornowski, A.; Festin, O ¨ .; Svedlindh, P.; Haase, M.; Weller, H. J. Am. Chem. Soc. 2003, 125, 9090. (9) Weller, D.; Moser, A. IEEE Trans. Magn. 1999, 35, 4423. (10) Pileni, M. P. J. Phys. Chem. B 2001, 103, 3358.

molecule layer, resulting in a macroscopic two-dimensional structure.12,13 The linker-molecule layer surface bears chemical groups that are capable of binding nanoparticles either through ligand-exchange reaction or electrostatic interaction.14-16 Prevention of nanoparticle coalescence is very important particularly for materials which require high-temperature annealing for magnetic phase transformation. For instance, in the formation of hard magnetic FePt nanoparticle self-assembled films, annealing at temperatures above 530 °C is required to transform the FePt nanoparticles from their as-synthesized, superparamagnetic, facecentered-cubic (fcc) phase to the tetragonal (L10) one with high magnetocrystalline anisotropy. During annealing, however, coalescence could occur due to thermal decomposition of the surfactants, resulting in the formation of large aggregates with broad size distribution and intergrain exchange coupling.17,18 These features are not desirable for the high-density recording media and memory device applications. Some recent efforts have been devoted to reduce coalescence using FePt nanoparticle self-assembled films by lowering the onset temperature for the fcc to L10 phase transformation down to around 400 °C either by means of doping elements such as Ag and Au in FePt nanoparticles19,20 or the rapid thermal annealing technique.21 However, it is still a great challenge to obtain well-transformed L10 FePt nanoparticle films with insignificant coalescence. (11) Tanase, M.; Nuhfer, N. T.; Laughlin, D. E.; Klemmer, T. J.; Liu, C.; Shukla, N.; Wu, X.; Weller, D. J. Magn. Magn. Mater. 2003, 266, 215. (12) Sun, S.; Anders, S.; Hamann, H. F.; Thiele, J.-U.; Baglin, J. E. E.; Thomson, T.; Fullerton, E. E.; Murray, C. B.; Terris, B. D. J. Am. Chem. Soc. 2002, 124, 2884. (13) Yu, A. C. C.; Mizuno, M.; Sasaki, Y.; Inoue, M.; Kondo, H.; Ohta, I.; Djayaprawira, D.; Takahashi, M. Appl. Phys. Lett. 2003, 82, 4352. (14) Colvin, V. L.; Goldstein, A. N.; Alivisatos, A. P. J. Am. Chem. Soc. 1992, 114, 5221. (15) Kotov, N. A.; De´ka´ny, I.; Fendler, J. H. J. Phys. Chem. 1995, 99, 13065. (16) Shipway, A. N.; Katz, E.; Willner, I. ChemPhysChem 2000, 1, 18. (17) Dai, Z. R.; Sun, S.; Wang, Z. L. Nano Lett. 2001, 1, 443. (18) Wang, S.; Kang, S. S.; Nikles, D. E.; Harrell, J. W.; Wu, X. W. J. Magn. Magn. Mater. 2003, 266, 49. (19) Kang, S.; Harrell, J. W.; Nikles, D. E. Nano Lett. 2002, 2, 1033. (20) Kang, S.; Jia, Z.; Nikles, D. E.; Harrell, J. W. IEEE Trans. Magn. 2003, 39, 2753. (21) Zeng, H.; Sun, S.; Sandstrom, R. L.; Murray, C. B. J. Magn. Magn. Mater. 2003, 266, 227.

10.1021/la0481694 CCC: $27.50 © 2004 American Chemical Society Published on Web 11/19/2004

11306

Langmuir, Vol. 20, No. 26, 2004

Letters

Figure 1. Schematic diagrams of the PEI/FePt (a) and APS/FePt (b) heterostructures on SiO2 substrates. Plane-view FE-SEM images of the as-deposited PEI/FePt (c) and APS/FePt (d) films; bright dots correspond to the Fe51Pt49 nanoparticles. Plane-view FE-SEM images of the 800 °C annealed PEI/FePt (e) and APS/FePt (f) films.

Previously, we reported the stabilization of macroscopic FePt nanoparticle monolayer films using [3-(2-aminoethylamino)propyl]trimethoxysilane13 and 3-aminopropyldimethylethoxysilane (APS)22 as linker molecules. In this paper, we discuss the mechanism of the function of linker molecules for preventing coalescence upon hightemperature annealing. And, we propose, with experimental evidence, the effective application of APS as a promising candidate for the above-mentioned function. Different degrees of anticoalescence were observed depending on the linker molecules used. For comparison purposes, we present two types of FePt nanoparticle monolayer films with one deposited on polyethylenimine (PEI) and the other one deposited on APS linker-molecule layers. Experimental Section Ethanol and hexane were reagent grade and were used as received. The water used was prepared by using a membrane purification system (GSR-500, Advantec). PEI (average Mw of ca. 25 000) and APS (>95%) were purchased from Aldrich Chemical Co., Inc., and Gelest, Inc., respectively, and used without further purification. An Fe51Pt49 nanoparticle hexane dispersion stabilized with oleic acid and oleylamine was synthesized according to the method reported by Sun et al.3 Based on transmission electron microscopy (TEM) observations, the Fe51Pt49 nanoparticles were spherical in shape and mostly single crystalline, with an average dimension of 4.4 nm, while showing a relatively small size distribution of 17%. Electron diffraction of the particles revealed a typical fcc pattern. The nanoparticle film fabrication process consisted of two steps: first, substrate surface modification with corresponding linker molecules, and second, exchanging surfactants around the nanoparticles with (22) Sasaki, Y.; Mizuno, M.; Yu, A. C. C.; Inoue, M.; Yazawa, K.; Ohta, I.; Takahashi, M.; Jeyadevan, B.; Tohji, K. J. Magn. Magn. Mater. 2004, 282, 122.

amino-functional groups of the linker-molecule layer surfaces. The PEI/FePt film was fabricated as follows: an oxidized silicon wafer was immersed into PEI aqueous solution (25 mg/mL) for 10 min and then rinsed with water and dried; subsequently, the PEI-derivative substrate was treated with a hexane dispersion of the Fe51Pt49 nanoparticles (1 mg/mL) for 10 min, followed by rinsing with hexane and drying, yielding a PEI/(Fe51Pt49 nanoparticle monolayer) bilayer heterostructure. For the preparation of APS/FePt film, an oxidized silicon wafer was immersed into APS ethanol/H2O ()4/1) solution overnight and then rinsed with ethanol and dried; subsequently, the APS-derivative substrate was treated with a hexane dispersion of the Fe51Pt49 nanoparticles (1 mg/mL) for 10 min, followed by rinsing with hexane and drying, yielding an APS/(Fe51Pt49 nanoparticle monolayer) bilayer heterostructure. The film morphology was characterized by using a Hitachi S-5200 field emission scanning electron microscope (FE-SEM). In-plane X-ray diffraction (XRD) spectra of the films were collected on a Rigaku ATX-G under Cu KR radiation (λ ) 0.154 184 nm), and the X-ray incident angle was set at 0.25° from the plane.

Results and Discussion The schematic diagrams of the PEI/FePt and APS/FePt heterostructures are shown in panels a and b of Figure 1, respectively. Plane-view FE-SEM images of the resulting PEI/FePt and APS/FePt films are shown in panels c and d of Figure 1, respectively. The particles were randomly distributed on both surfaces. It has been estimated that the average packing densities for the PEI/ FePt and APS/FePt films were 1.1 × 1012 and 1.0 × 1012 particles/cm2, respectively. In addition, the average areal packing fractions estimated were 17% for the PEI/FePt film and 15% for the APS/FePt film in their as-deposited states. Considering the maximum possible packing fraction of 25%, both surfaces were not fully covered by the particles. Nevertheless, the particles were separated from each other by a few nanometers in both cases, which is

Letters

Figure 2. In-plane XRD spectra for the 800 °C annealed PEI/ FePt (- - -) and APS/FePt (s) films. The diffraction peaks for L10 structure are indexed.

useful for magnetic recording media regarding the exchange coupling concern. The two films were then annealed at 800 °C for 30 min under a N2 atmosphere. Panels e and f of Figure 1 show the plane-view FE-SEM images of the 800 °C annealed PEI/FePt and APS/FePt films, respectively. For the PEI/ FePt case, particles seriously coalesced. The resulting average particle aggregate diameter was 28 nm (640% of the as-deposited particle size) with a broad size distribution of 39%. The average areal packing fraction for the annealed PEI/FePt film was estimated to be less than 17% due to possible formation of multilayer aggregates. Conversely, for the APS/FePt system, coalescence was not significant. The average annealed particle size and the average areal packing fraction were comparable to those values for the as-deposited film. These results suggest that the APS linker molecule can prevent coalescence effectively. In-plane XRD measurements for the 800 °C annealed PEI/FePt and APS/FePt films revealed that both nanoparticle films transformed to L10 phase (Figure 2). The c/a ratios estimated were 0.971 and 0.987 for the PEI/ FePt and APS/FePt cases, respectively. Comparing with the bulk L10 Fe50Pt50 value of 0.964,23 the APS/FePt film was not nearly as well transformed as the PEI/FePt film. The observed different degree of ordering was possibly due to the dependence of phase stability on particle size24 and atomic composition.22 The magnetic properties of the APS/FePt film are discussed in detail elsewhere.22 (23) Hansen, M. Constitution of Binary Alloys; McGraw-Hill: New York, 1958; p 699. (24) Takahashi, Y. K.; Koyama, T.; Ohnuma, M.; Ohkubo, T.; Hono, K. J. Appl. Phys. 2004, 95, 2690.

Langmuir, Vol. 20, No. 26, 2004 11307

Let us discuss the APS effectiveness for prevention of nanoparticle coalescence compared to the PEI case. Both PEI and APS thermally decomposed after the 800 °C annealing. And it is believed that the decomposition behaviors of PEI and APS were different from each other. PEI molecules shrank or stretched on the substrate surface during annealing due to their long molecular length and weak absorption onto the substrate surface via electrostatic interaction,25 resulting in physical movement of the particles on the substrate surface over a relatively long range of area. However, such particle movement resulting from the short-chain APS molecules that were covalently bound to the substrate surface25 could only be localized and thus insignificant. According to the Fourier transform infrared spectroscopy measurements, PEI decomposed into carbon while APS decomposed into a mixture of carbon and Si oxide. Cross-sectional TEM observations of the 800 °C annealed APS/FePt films suggested that each nanoparticle was partly covered and hence protected by the mixture of carbon and Si oxide. Therefore, the annealed particles were prevented from serious coalescence. Conclusions In conclusion, APS and PEI linker molecules are both useful for stabilizing metallic nanoparticles on a substrate; however, APS is a better candidate than PEI for preventing particle coalescence upon high-temperature annealing. This work can be extended for the development of nanoparticle arrays with a combination of nanolithographic patterning techniques (research in active progress).26-28 We expect that our results can provide a solid support for future development of both magnetic and nonmagnetic nanoparticle based nanodevices, especially for those materials or devices that require high-temperature annealing. Acknowledgment. We acknowledge M. Takahashi for supporting X-ray measurements and Japan MEXT for partial support of the Nanotechnology Support Project. LA0481694 (25) Tang, Z.; Wang, Y.; Kotov, N. A. Langmuir 2002, 18, 7035. (26) Liu, G. Y.; Xu, S.; Qian, Y. Acc. Chem. Res. 2000, 33, 457. (27) Finnie, K. R.; Haasch, R.; Nuzzo, R. G. Langmuir 2000, 16, 6968. (28) Liu, S.; Maoz, R.; Sagiv, J. Nano Lett. 2004, 4, 845.