J Sol-Gel Sci Techn (2006) 40:317–323 DOI 10.1007/s10971-006-9321-7
Fe3 O4 –SiO2 nanocomposites obtained via alkoxide and colloidal route A. Jitianu · M. Raileanu · M. Crisan · D. Predoi · M. Jitianu · L. Stanciu · M. Zaharescu
Published online: 22 August 2006 C Springer Science + Business Media, LLC 2006
Abstract The magnetic nanocomposite materials represent an important class of nanomaterials extensively studied nowadays due to their varied applications from medical diagnostic to storage information. The iron oxides in silica matrix systems are highly investigated. The sol-gel method is a suitable way of preparation of Fe3 O4 -SiO2 nanocomA. Jitianu () Department of Materials Science and Engineering, Rutgers, The State University of NJ, 607 Taylor Road, Piscataway, NJ, 08854, USA e-mail:
[email protected] A. Jitianu · M. Raileanu · M. Crisan · M. Jitianu · M. Zaharescu Institute of Physical Chemistry, 202 Splaiul Independentei, 060021 Bucharest, Romania M. Raileanu e-mail:
[email protected] M. Crisan e-mail:
[email protected]
posite materials, since this method allowed the preparation of nanocomposite materials with narrow size distribution of magnetite in silica matrix. In the present work, nanocomposite materials in the Fe3 O4 -SiO2 system were prepared by solgel method via alkoxide and aqueous route. As SiO2 sources, tetraethoxysilan (TEOS) for the alkoxide route, as well as silica sol Ludox (30%) for the aqueous route, were used. This study shows the influence of the type of silica matrix on the structure, size, and distribution of the Fe3 O4 nanoparticles in the Fe3 O4 -SiO2 systems. The gels were annealed at 550◦ C in order to consolidate the matrices. The structural characterization of the obtained materials via the two preparation routes was performed by DTA/TGA analysis, X-ray diffraction, IR and M¨ossbauer spectroscopy, Transmission Electron Microscopy (TEM) and Selected Area Electron Diffraction (SAED). Keywords Fe3 O4 -SiO2 nanocomposites . Magnetite . Sol-gel method . FT-IR . TEM . M¨ossbauer spectra . Magnetic materials
M. Jitianu e-mail:
[email protected] M. Zaharescu e-mail:
[email protected] D. Predoi National Institute for Physics of Materials, P.O. Box. MG 07, 77125 Bucharest Magurele, Romania e-mail:
[email protected] M. Jitianu Clarkson University, CAMP, 8 Clarkson Ave., Box 5814 Potsdam, NY 13699-5814, USA L. Stanciu School of Materials Engineering, Purdue University, 501 Northwestern Ave., West Lafayette, IN 47907-2044, USA e-mail:
[email protected]
1. Introduction The nanocomposite materials that exhibit new properties and applications have become an important target of investigation in the last decade. The great interest on nanomaterials is based on the excellent study published by Smith [1], which made a connection between physical properties and grain size of materials. The Fex Oy -SiO2 materials prepared by solgel may display specific magnetic, electrical, as well as catalytic properties both in bulk and in film forms [2]. Yoshida et al. [3] have accomplished the first study on the Fe2 O3 -SiO2 amorphous magnetic composite systems in 1981. Magnetic materials as γ -Fe2 O3 (maghemite) embedded in SiO2 were Springer
318
extensively studied by Piccaluga et al. [4–6]. Morales et al. [7] were the first who studied the magnetic properties of γ -Fe2 O3 embedded in silica matrices obtained by either alkoxide or aqueous route. The magnetic materials for example nanoparticles of magnetite (Fe3 O4 ), pure or encapsulate in inert matrices such as silica (SiO2 ) or anatase (TiO2 ), have been the subject of both technological and theoretical interest [8]. Recently, Fe3 O4 nanoparticles coated with gold [9] were used with great success in biologic and catalytic applications. The core-shell particles [10, 11] based on Fe3 O4 coated with silica and than with titanium oxide by sol-gel method were introduced as a new material for photocatalytic decomposition of organic residues. The presence of Fe3 O4 core facilitates the magnetic separation of the catalyst after the photocatalytic processes, which represent a very important technological achievement, due to the difficulty of this process. The preparation of Fe3 O4 -SiO2 core-shell nanoparticles [12, 13] presents a particular interest since silica shell could screen the magnetic dipolar attraction between Fe3 O4 particles and as a consequence very uniform dispersions in liquid medium of these types of particles could be achieved. On the other hand the silica shells act as an efficient barrier against the magnetic particles leaching in acid medium. These kinds of materials are usually prepared by sol-gel method [12], inverse miniemulsion processes [13] or reverse-micelle technique [14]. By using the reverse-micelle technique [14] new composite materials were obtained with applications as biosensors devices and/or as drug delivery materials. However, the most important applications of the Fe3 O4 based materials are in medicine. By using superparamagnetic nanoparticles as Fe3 O4 coated and uncoated with silica a new class of contrast agents for magnetic resonance imaging (MRI) and infrared imaging (IRI) were involved in cancer diagnosis [15– 17]. Nevertheless, the cancer therapy is another important application of these nanocomposite materials. These particles could generate heat in an alternating magnetic field and these acts as hyperthermic agent which could destroy tumor tissue [16, 17]. Beside the application of the magnetic particles in modern medicine also the monolithic magnetic gels in Fe3 O4 -SiO2 systems were used with real success for magnetophoresis and for other biophysical applications [18]. On the other hand the Fe3 O4 -SiO2 nanocomposites were found to be very important, especially as magnetic data-recording media [19, 20]. This type of nanocomposites was found to have good applications in the field of magneto-optical sensors and magnetic devices due to their attractive physical and chemical properties [21]. In our previous papers [2, 22], the optimization of the synthesis of the nanocomposites in the Fe2 O3 -SiO2 systems was studied, by using tetraethoxysilane (TEOS) and methyltriethoxysilane (MTEOS), as silica sources, respectively. On the other hand, the nanocomposites in Fe2 O3 -SiO2 systems Springer
J Sol-Gel Sci Techn (2006) 40:317–323
were studied from the structural and textural point of view [2] and tested for magnetical intensified adsorption of arsenic ions from wastewater [23]. For this study, two different Fe3 O4 -SiO2 systems were prepared using TEOS and silica sol (Ludox 30%) as silica sources. The aim of this study was to describe a new simple way to obtain nanocomposite materials with magnetic properties, to outline the influence of the type of silica matrix on the structure, size, and distribution of the Fe3 O4 nanoparticles in the Fe3 O4 -SiO2 system. 2. Experimental 2.1. Preparation of the samples All samples were prepared by two sol-gel methods in order to obtain nanocomposites with 3% wt final iron content. The silica sources were tetraethoxysilane (TEOS) from Merck, in the alkoxidic route, and colloidal silica sol Ludox SM 30 type (Aldrich), in the colloidal route. The Fe3 O4 source, laboratory prepared, was a common one for both routes of the sol-gel method employed. The Fe3 O4 was prepared by dissolving of two iron salts Fe(NO3 )2 and Fe(NO3 )3 (Merck) with molar ratio Fe2+ : Fe3+ = 2 : 3 in 1 liter of distilled water. Ammonium hydroxide (25%) was added dropwise to the so-prepared solution under vigorous magnetic stirring until pH = 10 was reached. The system was kept under constant stirring at room temperature for 24 h, followed by the separation of Fe3 O4 powder that was then washed with distilled water until a neutral pH was achieved. The Fe3 O4 -SiO2 nanocomposites have been prepared using the colloidal and the alkoxide routes. For both sol-gel pathways, initially an aqueous dispersion of Fe3 O4 have been obtained by sonication for 15 min, followed by admix of NH4 OH until the value of pH = 9 was reached. The amount of Fe3 O4 used for suspension preparation was calculated as well as the final material SiO2 -Fe3 O4 to contain 3% iron. The syntheses of the nanocomposites Fe3 O4 -SiO2 materials were performed as follows: – In the alkoxide route, the gel was obtained by mixing a solution of TEOS in ethanol (TEOS : EtOH = 1 : 4) with the so prepared aqueous (TEOS : H2 O = 1 : 13.375) dispersion of Fe3 O4 . The reaction mixture was refluxed at 65◦ C under continuous stirring until the gelation occurred. During the gelation process, the pH of the reaction solution decreased up to value presented in Table 1. – In the colloidal route, the gel was obtained by mixing a 25% SiO2 diluted colloidal silica sol (destabilized with a 25% HNO3 solution from Merck) with a Fe3 O4 aqueous suspension. The reaction mixture was kept at 65◦ C, under
J Sol-Gel Sci Techn (2006) 40:317–323 Table 1 Composition of starting solutions and experimental conditions for Fe3 O4 -SiO2 nanocomposites preparation
319
Molar ratio Sample Precursors 1 2
TEOS + Fe3 O4 Colloidal silica + Fe3 O4
continuous stirring, until the gelation has occurred. The gelation was judged by examination by the naked eye. All the prepared nanocomposites have been dried at 70◦ C for 12 h in air. The composition of the starting solutions and the experimental conditions used for preparation of Fe3 O4 -SiO2 gels are presented in the Table 1. The obtained gels were thermally treated at 550◦ C, 2 h in air. 2.2. Samples characterization The thermal stability of the nanocomposites was study with a Differential Thermal Analyzer DTA-7 and with Thermogravimetric Analyzer TGA-7 from Perkin-Elmer at 5◦ C/min heating rate in the temperature range between 50–800◦ C under air flow (20 ml/min). The XRD analysis was performed using a Scintag Diffractometer XDS 200 with Cu-Kα radiation. IR spectra in the range 4000–200 cm−1 were recorded using a Carl-ZeissJena M80 spectrometer. The microstructure of samples was evidenced by conventional bright field imaging method on a JEOL 1200EX Transmission Electron Microscope (TEM) operating at 120 kV and by Selected Area Electron Diffraction (SAED).
Reaction conditions
ROH SiO2
H2 O SiO2
Fe SiO2
NH4 OH SiO2
pH
T (◦ C)
tgel (h)
3.981 –
13.375 24.250
0.033 0.033
0.008 0.008
4.0 9.5
65 65
1.5 22
M¨ossbauer absorption spectra were recorded in standard transmission geometry, by using a source of 57 Co in rhodium (37 MBq). Calibration was performed using a 25 µm thick natural iron foil. The measurements were carried out at room temperature on powdered samples kept in a Plexiglas holder. The surface density of the absorbers ranges from 105 to 145 mg/cm3 . Hyperfine parameters such as the isomer shift and quadrupole shift have been determined by the NORMOS program [24] and α–Fe at 300 K was used to calibrate isomer shift and velocity scale. 3. Results and discussion The DTA-TGA curves are displayed in Fig. 1. The main effects for sample 1 were assigned to the water and alcohol removal (endothermic effect at 120–190◦ C), hydroxyl groups evolution (exothermal effect at 305◦ C) and to combustion of the residual organics (exothermal effect at 460◦ C). For sample 2, the main effects were assigned to the water removal (endothermic effect at 145◦ C) and hydroxyl evolution (exothermal effect at 265◦ C). Both samples presents an exothermal effect at 740◦ C and 720◦ C, respectively, on DTA curves, assigned to the magnetite oxidation. The oxidation of magnetite was evidenced only
Fig. 1 The TGA-DTA curves for samples 1 (Fe3 O4 + TEOS) and 2 (Fe3 O4 + Colloidal silica), respectively, and in insert a detail of TGA curve of sample 1
Springer
320
in TG analysis of sample 1 (Fig. 1 insert) which show an increasing weight effect at ∼760◦ C. The TG curves of these samples present slow slopes, which can be connected with hydroxyl group evolution. The total weight loss has been evaluated to be 8% and 6% for samples 1 and 2, respectively. According to the DTA-TGA data, the samples were annealed at 550◦ C in air with a heating rate of 1◦ C/min. This temperature was chosen since this temperature was high
Fig. 2 The XRD patterns of the pure magnetite used for nanocomposites preparation and for the studied samples thermally treated at 550◦ C
Fig. 3 The IR spectra of the nanocomposite materials thermally treated at 550◦ C
Springer
J Sol-Gel Sci Techn (2006) 40:317–323
enough to allow the elimination of all organic groups, but rather low to avoid the oxidation of magnetite, preserving the magnetic properties of the gels. Figure 2 shows the XRD patterns of samples thermally treated at 550◦ C, and of the pure Fe3 O4 used for the nanocomposites preparation. For all patterns presented in Fig. 2, the characteristic diffraction lines of Fe3 O4 were identified in good agreement with the reference file, JCPDS
J Sol-Gel Sci Techn (2006) 40:317–323
321
19-0629. From the diffraction lines intensity it could be concluded that for sample 2, prepared with colloidal silica, Fe3 O4 has a lower degree of crystallization, comparatively with Fe3 O4 for sample 1, prepared, with TEOS. The amorphous character of sample 2 can be the result of a partial dissolution of the magnetite during the sol-gel reaction since the gelation processes occurred slower comparatively to sample 1. These results are in good agreement with the IR data that are presented in Fig. 3. Both IR spectra present the characteristic bands for the Si–O–Si bonds (ν as Si–O–Si LO at 1210 cm−1 , ν as Si–O–Si TO at 1080 cm−1 , ν as Si–O(H) at 980 cm−1 , ν s Si–O–Si at 790 cm−1 , and δ Si–O–Si at 450 cm−1 ) reported by Bertoluzza et al. [25]. The Fe3 O4 presents a specific inverse spinel structure FeIII (FeII FeIII )O4 [26] in which the iron ions are arranged as follows: – half of Fe (III) ions are placed in octahedral holes as FeIII O6 and another half are placed in tetrahedral holes as FeIII O4 ; – the Fe (II) ions are placed in octahedral holes as FeII O6 . In the IR spectrum of sample 1 three characteristic vibrations due to the presence of Fe3 O4 (ν Fe–O from isolated tetrahedral FeIII O4 at 636 cm−1 , from condensed octahedral FeIII O6 at 570 cm−1 and from condensed octahedral FeII O6 at 390 cm−1 ) were identified. These characteristic vibrations were not identified for sample 2, prepared with colloidal silica, due to the partial dissolution of the Fe3 O4 in the silica matrix. This supposition is sustained by the shift at 1130 cm−1 of the ν as Si–O–Si TO of sample 2. The shift can be explained by the modification of the silica network by interaction with the iron oxides species [2]. The presence of the nano-sized iron oxides particles in the silica matrices and their crystalline degree were evidenced by TEM and SAED measurements, respectively, presented in Fig. 4. The size of the iron oxide particles in different silica matrices ranges between 15–20 nm for sample 1 and 8–15 nm for sample 2. It can be observed that for sample 2 prepared with colloidal silica, lower size of Fe3 O4 nanoparticle was obtained on the surface of silica. In addition, it can be observed that for this sample, an agglomeration of particles appeared due to the longer time of gelation necessary to obtain the final gel. This fact confirms the partial dissolution of Fe3 O4 during the sol-gel reaction. The SAED images confirm that sample prepared with TEOS has a higher degree of crystallinity comparatively with sample 2 prepared with colloidal silica. The SAED picture of sample 1 could be indexed as (111), (220), (311), (400), (522) and (440) planes of Fe3 O4 . The M¨ossbauer spectra of the studied samples in the range 80–300 K were recorded. The M¨ossbauer spectra for samples 1 and 2 at 80 K are displayed in Figs. 5–7, respectively. A
Fig. 4 Representative TEM micrographs of sample 1 (a) and 2 (b) thermally treated at 550◦ C and their SAED analyses in inserts
Springer
322
Fig. 5 The M¨ossbauer spectra for samples 1 and 2 recorded at 80 K
J Sol-Gel Sci Techn (2006) 40:317–323
The spectra at room temperature, for both samples, were fitted using two magnetic components of hyperfine fields Bhyp = 49.5(1) and 46.0(1) T, corresponding to Fe3+ ions in inverse spinel structure B(AB)O4 of magnetite (Fe3 O4 ) [26] at sites A (Fe2+ Fe3+ ) and ions at site B, respectively, with nearly null quadrupole splitting (QS) [29]. The spectrum at room temperature matches very well with reported stoichiometric Fe3 O4 [30, 31]. Previous M¨ossbauer studies at 80 K (below the Verwey transition) have evidenced rather complex spectra which present two main components as can be seen in Fig. 7. They were fitted with two distributions of hyperfine fields [31], corresponding to Fe3+ ions at A sites and to Fe2+ and Fe3+ at B sites. On the other hand, other studies show that the spectrum M¨ossbauer at 80 K was decomposed into two sextets [29]. For the studied samples the superparamagnetic blocking temperature was not observed up to 300 K, indicating the effect of agglomeration, which has also been also confirmed by TEM measurements. 4. Conclusions
Fig. 6 The M¨ossbauer spectra of sample 1 recorded at 80 K
Fig. 7 The M¨ossbauer spectra of sample 2 recorded at 80 K
significant difference was observed for the sample 2, namely the paramagnetic part (the doublet in the central part), which does not appear for the spectrum of sample 1. This modification could indicate that sample 2 have particles of different sizes including such one are superparamagnetic at 80 K. The fraction of the small particles is 12%. For both samples the spectra presented in Figs. 6 and 7 display two components represented by a magnetic sextet with 52.6 T and 51.1 T magnetic fields which has been assigned to magnetite [27, 28].
Springer
New Fe3 O4 -SiO2 nanocomposites have been prepared by employing the alkoxide and aqueous routes of the sol-gel method, using two silica sources, TEOS and colloidal silica, respectively. The nanocomposite materials were characterized from structural and textural point of view. It was established that the silica source plays a major role in the future properties of the so obtained materials. The M¨ossbauer spectra indicated that the spherical magnetite nanoparticles were not oxidized after the thermal treatment at 550◦ C for 2 h. Taking into account the magnetic properties of Fe3 O4 based materials, these nanocomposites will be further evaluated as possible new contrast media (CM) for magnetic resonance imaging (MRI). Acknowledgment The authors would like to thank to Dr. V.E. Kuncser from the National Institute of Materials Physics, Bucharest for valuable discussions.
References 1. Smith WF (1986) Principles of materials science engineering. MacGraw Hill Book. Singapore, p 128 2. Jitianu A, Crisan M, Meghea A, Rau I, Zaharescu M (2002) J Mater Chem 12:1–8 3. Yoshio T, Kawaguchi C, Kanamaru F, Takahashi K (1981) J Non-Cryst Solids 43:129 4. Ennas G, Masinu A, Piccaluga G, Zedda D, Gatteschi D, Sangregorio C, Stanger JL, Concas G, Spano (1998) Chem Mater 10:495
J Sol-Gel Sci Techn (2006) 40:317–323 5. Cannas C, Gatteschi D, Masinu A, Piccaluga G, Sangregorio C (1998) J Phys Chem B 102:7721 6. Bruni S, Cariati F, Casu M, Lai A, Musinu A, Piccaluga G, Solinas S (1999) Nano Struct Mater 11:573 7. Morales MP, Munoz-Aguado MJ, Garcia-Placios JL, Lazaro FJ, Serna CJ (1998) J Magn Magn Mater 183:232 8. Meldrum FC, Kotov NA, Fendler JH (1994) J Phys Chem 98:4506 9. Wang L, Luo J, Fan Q, Suzuki M, Suzuki IS, Engelhard MH, Lin Y, Kim N, Wang JQ, Zhong C-J (2005) J Phys Chem B 109:21593 10. Watson S, Beydoun D, Amal R (2002) J Photochem Photobio A: Chem 148:303 11. Watson S, Scott S, Beydoun D, Amal R (2005) J Nanopart Res 7:691 12. Deng YH, Wang CC, Hu JH, Yang WL, Fu SK (2005) Colloids Surf A: Physiochem Eng Aspects 262:87 13. Xu Z, Wang CC, Yang WL, Fu SK (2005) J Mat Sci 40:4667 14. Yang HH, Zhang SQ, Chen XL, Zhuang ZX, Xu JG, Wang XR (2004) Anal Chem 76:1316 15. Shieh DB, Cheng FY, Su CH, Yeh CS, Wu MT, Wu YN, Tsai CY, Wu CL, Chen DH, Chou CH (2005) Biomat 26:7183 16. Ito A, Shinkai M, Honda H, Kobayashi T (2005) J Biosci Bioeng 100:1 17. Kalambur VS, Han B, Hammer BE, Shield TW, Bischof JC (2005) Nanotech 16:1221 18. Jun YN, Dabbs DM, Aksay IA, Erramilli S (1994) Langmuir 10:3377
323 19. Bate G (1991) J Magn Magn Mater 100:413 20. Homola AM, Lorenz MR, Mastrangelo CJ, Tilbury TL (1986) IEEE Trans Magn 22:716 21. Ll. Casas, Roig A, Rodriguez E, Molins E, Tejada J, Sort J (2001) J Non-Cryst Solids 285:37 22. Zaharescu M, Crisan M, Jitianu A, Crisan D, Meghea A, Rau I (2000) J Sol-Gel Sci Tech 19:631 23. Peleanu I, Zaharescu M, Rau I, Crisan M, Jitianu A, Meghea A (2000) J Radioanal Nucl Chem 246:557 24. Brand RA (1987) Nucl Instr Methods Phys Res B 28: 398 25. Bertoluzza A, Fagnano C, Morelli MA, Gottaradi V, Guglielmi M (1982) J Non-Cryst Solids 48:117 26. Wells AF (1984) Structural inorganic chemistry, 5th edn. Clarendon Press, Oxford, p. 551 27. Mørup S, Dumisec JA, Topsoe H (1980) Magnetic microcristals in application of m¨ossbauer spectroscopy, Vol. II. In: Cohen RL (ed) 28. Szab´o D, Czak´o-Nagy I, Zrinyi M, V´ertes A (2000) J Colloid Interf Sci 221:166 29. Goya GF, Berqu´o TS, Fonseca FC, Morales MP (2003) J Appl Phys 94: 30. Visalakski G, Venkateswaran G, Kulshreshtha SK, Moorthy PH (1993) Mater Res Bull 28:829 31. Simmons GW, Leidheiser H Jr (1976) In: Cohen RL (ed) Applications of M¨ossbauer spectroscopy, vol. I. Academic Press, New York
Springer
