Synthesis of Magnetic Silica-Based Nanocomposites Containing Fe₃O₄ Nanoparticles.

By Victor Matsura,^a Yannick Guari,^a* Joulia Larionova,^a* Christian Guérin,^a Andrea Caneschi,^b Claudio Sangregorio,^b Emmanuelle Lancelle-Beltran,^a Ahmad Mehdi,^a Robert J. P. Corriu^a

 

^aLaboratoire de Chimie Moléculaire et Organisation du Solide (LCMOS), UMR 5637, Université Montpellier II, Place E. Bataillon, 34095 Montpellier cedex 5, France. Fax: (33) 4 67 14 38 52, e-mail: This email address is being protected from spambots. You need JavaScript enabled to view it.; This email address is being protected from spambots. You need JavaScript enabled to view it.

^bINSTM Research Unit-Dipartimento di Chimica, Università di Firenze, via della Lastruccia 3, 50019 Sesto F.no, Firenze, Italy.

Abstract

A new mesostructured hybrid organic-inorganic silica host with internal anchored acetylacetonate groups has been used as a matrix for the growing and organisation of  Fe₃O₄ nanoparticles.  The approach used consists in the impregnation and the subsequent organic solution-phase decomposition of the molecular precursor Fe[(OC(CH₃))₂CH]₃ into the hybrid silica pores. The magnetic nanocomposite material obtained was fully characterised using transmission electron microscopy (TEM), infrared spectroscopy (IR), nitrogen physisorption, X-Ray diffraction techniques and magnetic measurements. These measurements reveal the presence of uniform-sized pure magnetite nanoparticles with a narrow size distribution of 3-4.5 nm exclusively inside the silica matrix. The data demonstrate that the covalent anchoring of the molecular precursor in the silica plays a crucial role in the fabrication of nanocomposites presenting a homogeneous spatial distribution of nanoparticles.

Key words: nanocomposites, hybrid silica, nanoparticles, iron oxide, magnetite

1. Introduction

In the recent years, the synthesis and study of nanometer-scale magnetic particles have been the focus of intense fundamental and applied research with special emphasis on their size-dependent properties.¹ Among magnetic nanoparticles, iron oxides (Fe₂O₃ and Fe₃O₄) have attracted technological interest due to their magnetic, catalytic, conducting and biological properties which could find applications in many fields including magnetic storage devices,² ferrofluids,³ magnetic refrigeration systems,⁴ magnetic carriers for drug targeting,⁵ separation of biochemical products⁵ and catalysis.⁶ Most of these applications require well-dispersed chemically stable nanoparticles having uniform size and shape. In this respect, the synthesis of nanocomposite materials where the magnetic nanoparticles are integrated into organic or inorganic matrixes is particularly promising since it allows not only the control over the particles growth but, in some cases, also their spatial arrangement. Indeed, polymer or copolymer templates,⁷ glass⁸ or ceramic materials⁹ have been extensively used as matrixes in the preparation of nanocomposites containing iron oxide nanoparticles. Alternatively, ordered mesostructured silica have also been used as hosts, not only providing the stabilisation of the nanoparticles and avoiding their aggregation but also providing their size control and their organization in one-, two- and three-dimensional arrays. Studies concerning the growth of a- and g-Fe₂O₃ nanoparticles inside mesoporous silicas MCM-41 and MCM-48 have been reported.¹⁰ Recently, using a supercritical fluid technique with a high temperature and a high pressure, Fe₃O₄ nanowires have also been obtained into SBA-type mesoporous hexagonal silicas.¹¹ 

In the current few years, an additional advantage in the synthesis and the structuring of metal nanoparticles has been offered by the employment of hybrid mesostructured silica as a matrix. In this respect, the discovered synthetic approach consists in the thermal or chemical decomposition of metal organic or molecular precursors of nanoparticles, which are covalently anchored into the hybrid mesostructured silica by appropriate functionalities. Thus, the nature and the number of organic functionalities control host – guest interactions and the uniform distribution of these precursors into the silica matrix that provides the retention and the homogeneous dispersion of the nanoparticles into the matrix. This method has successfully been used by some of us and others groups in growing and organizing various metal or metal oxide nanoparticles or nanorods inside of hybrid hexagonal silicas.¹²

In this context, the present work reports on a new and simple method to growth and structure magnetic Fe₃O₄ nanoparticles using decomposition of the molecular precursor Fe[(OC(CH₃))₂CH]₃ incorporated into hybrid mesostructured silica containing acetylacetonate functionalities. The choice of the molecular precursor has been determined on the basis of the following reasons. First of all, it has already been proved to be an efficient starting material for the synthesis of Fe₃O₄ nanoparticles stabilized by different combinations of surfactants and/or stabilizing agents as “mortar”.¹³ Secondly, the acetylacetonate ligands of this precursor may be exchanged with acetylacetonate functionalities of the hybrid silica that open the way for covalent anchoring and homogeneous dispersion of the precursor into the hybrid silica. The article is organised as follows: first, we report on the preparation of the parent hexagonal silica SBA-15 functionalised with internal anchored acetylacetonate groups; then we describe the growing and organisation of Fe₃O₄ (magnetite) nanoparticles by the impregnation and subsequent organic solution-phase decomposition of the molecular precursor Fe[(OC(CH₃))₂CH]₃ inside this hybrid silica matrix; finally the magnetic properties of the as synthesised nanostructured materials are presented.

 

2. Experimental Section

2.1. Synthesis

Tetraethylorthosilicate (TEOS), sodium fluoride (NaF), iron(III) acetylacetonate (Fe[(OC(CH₃))₂CH]₃) and 1-amino-9-octadecene (oleylamine) were purchased from Acros. (3-chloropropyl)triethoxysilane, Pluronic 123 (PEO₂₀PPO₇₀PEO₂₀ with PEO = poly (ethylene oxide) and PPO = poly (propylene oxide)), 1,2-hexadecanediol and cis-9-Octadecenoic acid (Oleic acid) were purchased from Aldrich. Potassium tert-butoxide was purchased from Lancaster. All of the solvents used in these experiments were analytical grade.

The mesoporous non-functionalised silica SBA-15 was synthesised as previously described.¹⁴ Its specific surface area (S_BET) of 851 m²g^-1, pore diameter of 5.4 nm (calculated from the desorption branch) and pore volume of 1.08 cm³g^-1 were determined from nitrogen physisorption data (see Supporting Materials). d₁₀₀ = 10.0 nm, d₁₁₀ = 17.7 nm and d₂₀₀ = 20.0 nm were determined from the powder XRD diffraction pattern.

Mesoporous functionalised hybrid silica [CH₃C(O)]₂C(CH₂)₃SiO_1.5/9SiO₂ 1.

Compound [CH₃C(O)]₂C(CH₂)₃Si(OCH₂CH₃)₃ was synthesised from I(CH₂)₃Si(OCH₂CH₃)₃ as previously described with slight modifications.¹⁵

Synthesis of I(CH₂)₃Si(OCH₂CH₃)₃. (3-chloropropyl)triethoxysilane (30.0 g, 124.7 mmol) and sodium iodide (28.1 g, 187.1 mmol) were allowed to react at reflux in acetone (200 mL) for 72 hours and the solvent removed under vacuum. Pentane was added and the resulting suspension filtered in order to remove the salts. The solvent was removed under vacuum and the product distilled under reduced pressure to afford 34.9 g (105.1 mmol, 83%) of a colourless oil. b.p. 85 °C at 3.3 mmHg. Anal. calc. for C₉H₂₁O₃ISi (%): C, 32.53; H, 6.37. Found (%): C, 32.50; H, 6.37. ¹H NMR (400.13 MHz, CDCl₃, 293K, δ): 0.70 (m, 2H, CH₂Si), 1.20 (t, 9H, ³J_HH = 7.0 Hz, CH₃CH₂OSi), 1.90 (m, 2H, CH₂CH₂Si), 3.19 (m, 2H, ICH₂), 3.82 (q, 6H, ³J_HH = 7.0 Hz, CH₃CH₂OSi). ¹³C{¹H} NMR (100.71 MHz, CDCl₃, 293K, δ): 11.06 (s, CH₂Si), 12.66 (s, CH₂CH₂Si), 18.68 (s, CH₃CH₂OSi), 27.97 (s, ICH₂), 58.86 (s, CH₃CH₂OSi).

Synthesis of [CH₃C(O)]₂C(CH₂)₃Si(OCH₂CH₃)₃. To a solution of ^tBuOK (11.2 g, 100 mmol) in ^tBuOH (100 mL), 15.0 g of acetyl acetone (150 mmol) were added at room temperature. After stirring for 10 minutes, 33.2 g (100 mmol) of iodopropyltriethoxysilane were added. The resulting mixture was stirred and heated under reflux for 20 hours. The solvent was removed under vacuum and the residue dissolved in pentane (200 mL). After filtration under argon, the filtrate was concentrated. The residual clear yellow liquid was distilled to afford 22.1 g of [CH₃C(O)]₂C(CH₂)₃Si(OCH₂CH₃)₃ (70 mmol, 70%) as a colourless liquid. b.p. 115 °C at 0.01 mmHg. As determined by Gas Chromatography analysis, the O-alkylated product is present as an impurity (4%). IR (cm^-1, CCl₄): 2976, 2932, 2888 (w, ν_CH); 1730 and 1703 (ν_CO, keto form), 1616 (ν_CO, enol form). ¹H NMR (400.13 MHz, CDCl₃, 293K, δ): 0.58 (m, 2H, CH₂Si), 1.18 (t, 9H, ³J_HH = 7.0 Hz, CH₃CH₂OSi), 1.32 and 1.46 (t, CH₂CH₂Si), 1.80 (m, CH₂CH₂CH₂Si for the keto form), 2.0 (s, CH₃C(O) for the enol form), 2.10 (s, C(O)CH₃ for the keto form), 2.16 (m, CH₂CH₂CH₂Si for the enol form), 3.68 (t, ³J_HH = 7.0 Hz, CH₂CH[C(O)CH₃]₂ for the keto form), 3.74 (q, 6H, ³J_HH = 7.0 Hz, CH₃CH₂OSi). ¹³C{¹H} NMR (100.71 MHz, CDCl₃, 293K, δ): 10.7 (s, CH₂Si), 18.6 (s, CH₃CH₂OSi), 21.4, 23.1, 24.3, 29.3, 31.7, 58.7 (s, CH₃CH₂OSi), 69.0 (s, CHCH₂CH₂CH₂Si for the keto form), 110 (s, CCH₂CH₂CH₂Si for the enol form), 191.5 (s, C(O)CH₃ for the enol form), 204.7 (s, C(O)CH₃ for the keto form). ²⁹Si{¹H} NMR (39.75 MHz, CDCl₃, 293K, δ): - 46.0. FAB MS (NBA): m/z: 305 [M+H]⁺, 25 %; m/z: 259 [M-OEt]⁺, 100 %.

Mesoporous hybrid silica [CH₃C(O)]₂C(CH₂)₃SiO_1.5/9SiO₂ 1. Functionalised silica 1 was prepared using the so-called direct method via co-condensation of the primary building blocks ([CH₃C(O)]₂C(CH₂)₃Si(OCH₂CH₃)₃ and TEOS) in the presence of the surfactant. Pluronic 123 (PEO₂₀PPO₇₀PEO₂₀ with PEO = poly(ethylene oxide) and PPO = poly(propylene oxide)) (1.06 g) was dissolved in an aqueous solution of HCl (pH=1.5). The resulting clear solution was then added to a mixture of [CH₃C(O)]₂C(CH₂)₃Si(OCH₂CH₃)₃ (0.37 g, 1.2 mmol) and tetraethylorthosilicate (TEOS) (2.36 g, 11.3 mmol)). The molar composition of the mixture for 1.0 g of triblock copolymer is 0.015NaF:0.25TEOS:0.03[CH₃C(O)]₂C(CH₂)₃Si(OCH₂CH₃)₃:0.03HCl:55H₂O. The mixture was vigorously stirred for 3 hours at room temperature until a transparent solution appears. The solution was transferred in a hot bath at 60 °C and NaF (19 mg, 0.76 mmol) was immediately added. After aging under regular stirring for 3 days at 30 °C, the resulting powder was filtered off and the surfactant was selectively removed by soxhlet extraction over ethanol for 24 hours. After drying at 120 °C under vacuum (0.1 mmHg), a white powder was obtained. [CH₃C(O)]₂C(CH₂)₃SiO_1.5/9SiO₂, yield 0.9 g (98 %). The content of functional groups was determined by thermogravimetric analysis. The weight loss of 18 % corresponds to the content of 1.36 mmol.g^-1 of the acetylacetonate groups. IR (cm^-1, KBr disk): 1726 and 1699 (ν_CO, keto form), 1634 (ν_CO, enol form). ¹³C NMR CP-MAS (75.42 MHz, d): 10.7 (br, CH₂Si), 20.9 (br), 28.7 (br), 57.9 (br, CH₃CH₂OSi), 67.8 (br, CHCH₂CH₂CH₂Si for the keto form), 110 (br, CCH₂CH₂CH₂Si for the enol form), 190.1 (br, C(O)CH₃ for the enol form), 205.6 (br, C(O)CH₃ for the keto form). ²⁹Si NMR CP-MAS (59.62 MHz, d): -66.2 (T³), -101.2 (Q³), -110.3 (Q⁴).

Growing of Fe₃O₄ nanoparticles into the hybrid silica.

Incorporation of Fe[(OC(CH₃))₂CH]₃ into the hybrid silica [CH₃C(O)]₂C(CH₂)₃SiO_1.5/9SiO₂. Composite 2 was prepared by incorporation of the precursor Fe[(OC(CH₃))₂CH]₃ into the mesoporous functionalised silica 1. A typical experiment was performed as follows: 150 mg of the silica powder 1 were suspended in 5 mL of an ethanolic solution of Fe[(OC(CH₃))₂CH]₃ (0.3 g, 0.84 mmol) and the mixture was stirred for 5 hours at room temperature. Then, the suspension was filtered off and the red iron-containing silica 2 was washed copiously until the filtrate became colourless and dried at room temperature. Elemental analysis, calc. for Fe[(OC(CH₃))₂CH]₂[CH₃C(O)]₂C(CH₂)₃SiO_1.5/9SiO₂ (%): Fe, 5.65; Si, 28.42. Found (%): Fe, 4.95; Si, 28.52. EDAX analyses: Fe/Si atomic ratio 8.3/90. Composite 2 will be formulated as Fe[(OC(CH₃))₂CH]₂[CH₃C(O)]₂C(CH₂)₃SiO_1.5/9SiO₂ considering the closeness between theoretical and experimental values for elemental analyses.

Growing of Fe₃O₄ nanoparticles into the hybrid silica. The nanocomposite Fe₃O₄@SiO₂ 3 was prepared by suspending material 2 (100 mg) in phenyl ether (20 mL), then Fe[(OC(CH₃))₂CH]₃ (1.06 g, 3 mmol), 1,2-hexadecanediol (2.88 g, 10 mmol), oleic acid (1.88 g, 6 mmol) and oleylamine (1.78 g, 6 mmol) were added and the resulting mixture heated under nitrogen at 265 °C for 30 minutes. After cooling to room temperature, ethanol was added and the solution filtered off. The resulting precipitate was suspended in hexane, filtered and washed copiously with hexane. Elemental analysis, found (%): Fe, 6.04; Si, 29.33. EDAX analyses: Fe/Si atomic ratio 9.7/87.

Growing of Fe₃O₄ nanoparticles into the non-functionalized silica. The attempt to growth Fe₃O₄ nanoparticles into the mesoporous non-functionalized silica SBA-15 having a pore size of 5.4 nm was performed using the same procedure consisting in the impregnation of the non-functionalized silica with the precursor Fe[(OC(CH₃))₂CH]₃ and its organic solution-phase decomposition into the silica matrix. Elemental analysis, found (%): Fe, 0.94; Si, 37.98. EDAX analyses: Fe/Si atomic ratio 1.5/97.

 

2.2. Physical Measurements

NMR Spectra were recorded on a Bruker ADVANCE DRX 400 (at 400.13 MHz for ¹H and at 100.71 MHz for ¹³C) and on a Bruker ADVANCE DPX 200 (at 39.75 MHz for ²⁹Si) operating on the Fourier transform mode. The ratio C-alkylated/O-alkylated product was determined by Gas chromatography analysis on a Hewlett Packard 6890 GC system using a DB5 phase capillary column (30 m x 0.25 mm, 0.25 μm). Mass spectrometry analyses were performed on a JEOL 9MS-SX apparatus. Cross-polarization magic angle spinning (CP MAS) ²⁹Si NMR spectra were recorded on a Bruker FTAM 300 as well as CP MAS ¹³C NMR spectra. In both cases, the repetition time was 5 and 10 seconds with contact times of 5 and 2 ms. Qⁿ and Tⁿ notations are given respectively for ((SiO)_nSiO_4-n) and ((SiO)_n(R)SiO_3-n) environments. IR spectra were recorded on a Perkin Elmer 1600 spectrometer. Thermogravimetric analyses (TGA/DTA) were performed on a NETZSCH STA 409 PC LUXX instrument. Specific surface areas were determined by the Brunauer-Emmett-Teller (BET) method on a Micromeritics ASAP 2010 analyzer. Elemental analyses were performed by the Service Central d'Analyse (CNRS, Vernaison, France). Powder X-ray diffraction patterns were measured on a Bruker D5000 diffractometer equipped with a rotating anode (Institut Européen des Membranes, UMR ENSCM-UMII-CNRS 5625, Montpellier, France). Transmission Electron Microscopy (TEM) observations were carried out at 100 kV on a JEOL 1200 EXII microscope. Samples for TEM measurements were prepared using ultramicrotomy techniques and then deposited on copper grids. Extractive replicas were prepared in order to obtain the nanoparticle size distribution histogram, which was determined using enlarged TEM photo images taken with a magnification of  X100K. Three concordant histograms with a minimum of 200 nanoparticles each were completed in order to obtain a size distribution with good statistics. Magnetic measurements were performed with a Cryogenic S600 SQUID magnetometer. The field dependence of the magnetization was measured on an Oxford Instruments Vibrating Sample Magnetometer (VSM) equipped with a super-conducting magnet capable of producing field up to 12 T. The data were corrected for the sample holder and the diamagnetic contribution calculated from the Pascal’s constants.¹⁶

 

3. Results and Discussion

Scheme 1 represents the method we used in order to form and organise the Fe₃O₄ nanoparticles into the mesostructured silica matrix. The first step consists in the preparation of hybrid mesostructured silica functionalised with acetylacetonate groups (Scheme 1a). The second step involves the incorporation of the molecular precursor Fe[(OC(CH₃))₂CH]₃ into the hybrid silica and the third step concerns the intrapore growth of nanosized Fe₃O₄ by organic solution-phase decomposition of this precursor into the silica pores (Scheme 1b). The different characterisations allowed us to monitor the evolution of our material at each step of the synthesis.

 

3. 1. Synthesis.

Hybrid silica functionalized with acetylacetonate groups. The parent highly organised hybrid silica containing acetylacetonate functionalities 1 was prepared by a one-pot synthesis as shown in Scheme 1a. Firstly, the organic building block [CH₃C(O)]₂C(CH₂)₃Si(OCH₂CH₃)₃ was synthesised by treating the iodopropyltriethoxysilane with potassium acetylacetonate. Secondly, the co-condensation reaction of [CH₃C(O)]₂C(CH₂)₃Si(OCH₂CH₃)₃ and tetraethylorthosilicate (TEOS) in the presence of the structure-directing agent Pluronic 123 (PEO₂₀PPO₇₀PEO₂₀ with PEO = poly(ethylene oxide) and PPO = poly(propylene oxide)) using NaF as catalyst in acidic media affords the formation of the hybrid silica containing acetylacetonato functions [CH₃C(O)]₂C(CH₂)₃SiO_1.5/9SiO₂ 1. The use of the one-pot synthesis allows to control the localisation of propylacetylacetonato groups of the hybrid silica in the hydrophobic core of the micelle during the synthesis that permits the preparation of a mesostructured material with the organic functions within the pore channels.¹⁷ Furthermore, this synthesis allows to control the amount of organic functions in the resulting hybrid organic-inorganic mesoporous material.

The solid state ¹³C NMR spectrum of 1 presents signals at d = 67.8 ppm (CH₃C(O)CH) and d  = 205.6 (CH₃C(O)CH) characteristics of the acetylacetonato group. ²⁹Si NMR display resonance for siloxane centres at –101.2 (Q³) and –110.3 ppm (Q⁴), and for organosiloxane centres at -66.2 ppm (T³). These results demonstrate the stability of Si-C bonds and the covalent attachment of organic groups to the silica wall. The Infrared spectrum of 1 clearly shows the C=O group stretching vibrations of acetylacetonato groups observed at 1726, 1699 and 1634 cm^-1 for the keto and enol forms respectively and SiO₂ vibration bands at 1080, 948; 798 and 459 cm^-1 (Figure 2Sa, Supplementary Materials).

The content of acetylacetonate groups in the hybrid silica 1 was determined by thermogravimetric analysis. The thermogravimetric curve of 1 performed under argon atmosphere from room temperature up to 700 °C exhibits two well-pronounced weight loss steps with an inflexion point at 50 and 330 °C which corresponds to the weight losses of 3 and 18 %, respectively (Figure 1S, Supplementary Materials). The first step (3%) corresponds to the loss of the solvent molecules situated in the pores of the silica and the residual of the polycondensation process. It should be noted that the same weight loss step was observed on the thermogravimetric curve of the non-functionalised silica SBA-15. The second weight loss (18%) corresponds to the loss of organic moieties of hybrid silica 1 and allows a determination of the content of acetylacetonato groups in 1, which is equal to 1.36 mmol.g^-1 giving the stoechiometric formula [CH₃C(O)]₂C(CH₂)₃SiO_1.5/9SiO₂.

 

Growing of Fe₃O₄ nanoparticles inside of the hybrid silica. The growth of Fe₃O₄ nanoparticles into the mesoporous silica was performed using the organic solution-phase decomposition of the precursor Fe[(OC(CH₃))₂CH]₃ inside the hybrid silica material, as shown in Scheme 1b. The key stage of our approach consists in the incorporation of Fe[(OC(CH₃))₂CH]₃ into the pristine hybrid silica 1. A substitution reaction between the silica functionalities ([CH₃C(O)]₂C(CH₂)₃-) and the acetylacetonate ligand of the iron complex allows its covalent anchorage to the silica. The insertion was performed by adding of the silica powder to an ethanolic solution of Fe[(OC(CH₃))₂CH]₃ at room temperature. The maximal insertion of 4.95 wt % of iron was obtained. The composite material formed can be formulated as Fe[(OC(CH₃))₂CH]₂[CH₃C(O)]₂C(CH₂)₃SiO_1.5/9SiO₂ 2. It should be noted that the increasing of Fe[(OC(CH₃))₂CH]₃ concentration or the repeating of the cluster insertion into the silica does not allow increasing the amount of inserted precursor. The IR spectrum of 2 (Figure 2Sb, Supplementary Materials) clearly shows the appearance of two new absorption bands at 1570 and 1526 cm^-1 attributed to the C=O stretching vibrations of the acetylacetonate groups coordinated to iron.¹⁸ In addition, the strong decrease of the intensity of the bands at 1726, 1699 and 1634 cm^-1 observed in the IR spectrum of 1 is indicative of a ligand substitution between the acetylacetonate ligands of Fe[(OC(CH₃))₂CH]₃ and acetylacetonate functionalities of the hybrid silica (Figure 2S, Supplementary Materials). The intensive washing of 2 by re-suspension in organic solvents did not lead to a decrease of the amount of the inserted precursor. In comparison, the incorporation of Fe[(OC(CH₃))₂CH]₃ into the non-functionalised silica SBA-15 (pore size of 5.4 nm) allows the incorporation of only 0.94 wt % of iron. All these results confirm that iron has been anchored into the pores by covalent interactions.

The third stage of our approach consists in the intrapore formation of the Fe₃O₄ nanoparticles by the decomposition of the anchored iron complex (Scheme 1b). The growth of the nanoparticles was performed by the reaction of 2 in phenyl ether in the presence of 1,2-hexadecanediol, Fe[(OC(CH₃))₂CH]₃, oleic acid and oleylamine at 265 °C leading to the formation of the nanocomposite Fe₃O₄@SiO₂ 3. The iron content in 3 determined by elemental analysis is equal to 6.04 wt % that corresponds to 8.6 wt % of Fe₃O₄ (vide infra) into the silica.

Attempt to grow of Fe₃O₄ nanoparticles inside of a non-functionalized silica. In order to prove the role of the functionalised silica in the growing and the structuring of Fe₃O₄ nanoparticles, the impregnation and the subsequent organic solution phase decomposition of Fe[(OC(CH₃))₂CH]₃ was performed into the non-functionalised silica SBA-15 presenting a comparable pore size (5.4 nm). In this case, we observed the formation of Fe₃O₄ phase exclusively at the surface of the silica grains with an iron content of 0.94 wt % as inferred from elemental analyses. The TEM image of the material resulting from the reaction clearly shows out-of-pore aggregated Fe₃O₄ nanoparticles (Figure 3S, Supplementary Materials).

 

3. 2. Textural and physical characterisations of the pristine hybrid silica and of iron-containing nanocomposite materials.

Powder X-Ray diffraction. The X-Ray powder diffraction pattern of the hybrid silica 1 clearly demonstrates the hexagonal symmetry of the space group P6m with a sharp peak due to the (100) reflection along with two weak peaks attributable to the (110) and (200) reflections (Figure 1a, Table 1). The high degree of long-range order can be assumed according to the reflections at higher scattering angles 2q.

The XRD patterns at 2q (0 – 10°) of the nanocomposite materials 2 and 3 do not show any change of the peak positions in comparison to the XRD pattern of pristine hybrid silica 1, clearly indicating that the mesoporous silica keeps its hexagonal structure. However, after the incorporation of Fe[(OC(CH₃))₂CH]₃ and particularly after its transformation in Fe₃O₄ nanoparticles, a significant reduction of the X-ray peak intensities with respect to material 1 was observed (Figure 1). The intensity of the Bragg reflections originates from the difference in the scattering power between the silica walls and the empty pores. Due to the impregnation of the pores with the scattering material, the amount of scattering power within the pores is increased, resulting in overall loss of intensity due to phase cancellation between the pore walls and the guest species. This is a well-known phenomenon described in the literature.¹⁹ It should be noted that in the case of 3, a complete absence of long-range order reflections (110) and (200) was observed. This feature can be explained in terms of the presence of Fe₃O₄ nanoparticles randomly dispersed in the pore channels, which would lower the periodicity due to increasing destructive interferences between the network forming material and the guest particles within the pores.²⁰

The powder X-Ray diffraction pattern within the range of 2q (30 – 60°) of 3 exhibits diffraction peaks indexed as (311), (400), (422) and (511) reflections of the cubic structure of magnetite Fe₃O₄ (Figure 2).

Nitrogen Physisorption. The nitrogen physisorption isotherm and the pore-size distribution of 1 are presented in Figure 3, Table 1. Typical adsorption-desorption isotherm of type IV with an H1 hysteresis loop is observed. The very high Brunauer-Emmett-Teller (BET) surface area and pore volume are indicative of mesostructured silica. The pore size calculated from the adsorption/desorption branches of the isotherm using the Barret-Joyner-Hellenda (BJH) formula is equal to 9.9/7.6 nm (Table 1).

The nitrogen physisorption isotherms and the pore-size distributions of nanocomposites 2 and 3 are also shown in Figure 3 and Table 1. No change in the isotherm type is observed after the incorporation of Fe[(OC(CH₃))₂CH]₃ and its transformation into Fe₃O₄ nanoparticles proving the conservation of the cylindrical pore system in both cases. The amount of adsorbed nitrogen as well as the BET surface area is strongly reduced with increasing pore filling, i.e. incorporation of Fe[(OC(CH₃))₂CH]₃ and particularly after the intrapore formation of Fe₃O₄ nanoparticles. However, the mesoporosity can still be found in both cases, which proves again the preservation of the mesoporous host structure and leads to the conclusion that the precursor Fe[(OC(CH₃))₂CH]₃ in the case of 2 and the Fe₃O₄ nanoparticles in the case of 3 are situated inside the pores.

            The mean pore diameter decreases progressively after the incorporation of Fe[(OC(CH₃))₂CH]₃ and the intrapore formation of Fe₃O₄ as indicated by the shift of the peak of the pore-size distribution to smaller values. The amount of adsorbed nitrogen is also decreasing in both cases and mainly after the particle formation. These results clearly demonstrate a filling of the pores with the guest species in both cases. The total pore volumes calculated at p/p₀ = 0.9 is 1.15 cm.g^-1 for the pristine hybrid silica which becomes 0.62 cm.g^-1 after the nanoparticles formation. This finding indicates that the degree of filling of the pores in 3 is over 46 %. The observed nitrogen physisorption effects are known in the literature for intrapore formation inside different host-guest systems.²¹

Transmission Electron Microscopy. A representative TEM image of the pristine host silica 1 was taken with the electron beam direction perpendicular to the channel direction (Figure 4a) and parallel to the channel direction (Figure 4a, insert).

The TEM micrograph performed for the host structure after incorporation of Fe[(OC(CH₃))₂CH]₃ (Figure 4S, Supporting Materials) indicates that the mesostructure of the parent silica is still retained after incorporation. No visible particles were observed out of silica. A distinctive signal for the iron is detected by the energy dispersive spectroscopy (EDS) analysis with an atomic ratio Fe/Si of 8.3/90 proving that the precursor is situated into the pores of the silica.    

Figure 4b shows the TEM image of the host material after the intrapore formation of Fe₃O₄ nanoparticles. The organisation of the host and the pore structure are not damaged during the synthesis. No external bulk magnetite phase was observed at the surface of the silica grains and the presence of Fe₃O₄ nanoparticles with a narrow size distribution inside the pore channels is clearly visible. Removal of silica from the material 3 using an HF treatment was performed in order to determine the nanoparticles size distribution. A representative TEM micrograph of the nanoparticles obtained after the extractive replica is presented in Figure 5a. A single distribution with a mean size value of 3.8 nm is observed (Figure 5b) that is slightly smaller than the pore channel mean diameter of 5.7 nm obtained from BET measurements. Obviously, the pores of the silica matrix preclude further growth or aggregation of the formed nanoparticles.

Magnetic Measurements. The magnetic properties of the nanocomposite 3 were investigated by measuring the temperature dependence of the zero field cooled (ZFC) and field cooled (FC) magnetisations in the 2-300 K range. The curves, shown in Figure 6, display the thermal irreversibility characteristic of assemblies of single domain Fe₃O₄ nanoparticles,²² arising from the blocking-unblocking process of the particles magnetic moment when the temperature is varied. The ZFC curve exhibits a broad maximum at ca. 45 K and never collapses into the FC curve. The maximum of the ZFC magnetisation is related, even if in a complex way, to the blocking of particles with average energy barrier, while the temperature at which the ZFC and FC curves separate corresponds to the blocking of the largest particles. The observed behaviour indicates that at room temperature, a fraction of particles is still in the blocked state and thus that a broad distribution of the anisotropy energy barriers for the reorientation of the magnetic moments characterises the whole sample. A second, smaller peak is observed in the ZFC curve at 5 K. This peak can be ascribed to the blocking of a fraction of particles with a very small diameter (ca. 2-3 nm). The field dependence of the magnetisation was measured up to 12 T at 2.5 and 300 K and is shown in Figure 7a. At 2.5 K the magnetisation saturates at high field to 98 A×m²×kg^-1, close to the value expected for bulk magnetite.²³ At 200 mT the loop opens and an hysteretic behaviour appears with a remanence to saturation magnetisation ratio of 0.29 and a coercive field of ca. 38 mT (Figure 7b). At room temperature, the sample does not reach saturation even in the largest external field of 12 T where M = 71 A×m²×kg^-1. However even at this temperature a small hysteresis is observed, (H_c » 6 mT). Such hysteretic behaviour can be ascribed to the presence of a part of particles, which, even at this temperature, are still in the blocked state as indeed suggested by ZFC/FC magnetisations measurements.

The dynamics of the magnetization was also investigated through a.c. susceptibility measurements. Both, the in-phase, c’, and out-of-phase, c”, components of the a.c. susceptibility exhibit a single relatively sharp maximum in the investigated temperature range. The position of the peaks is frequency dependent shifting toward low temperatures as the frequency decreases. The average relaxation time, t, can be extracted from the c” curves assuming that t = 1/2pn, at the temperature of the maximum. The data, plotted as ln(t) vs. 1/T_max are shown in Figure 8. They can be satisfactorily fit to a straight line, as indeed expected for an assembly of randomly oriented non-interacting particles for which the relaxation time of the magnetisation follows the Néel-Brown model.²⁴ According to this model the temperature dependence of the relaxation time is given by t=t₀exp(D/k_BT), where D is the energy barrier and is equal to KV, K being the magnetic anisotropy constant and V the particle volume, t₀ is a time constant which is usually of the order of 10^-9-10^-12 s, and k_B is the Boltzmann constant. The best parameters of the linear fit we found were t₀ = 3×10^-10 s and D=KV/k_B = 495 K. The t₀ value is in the range usually found for non-interacting nanoparticles suggesting that interparticle interactions do not significantly affect the dynamics of relaxation.²⁵ From the obtained value for the energy barrier and using the average diameter obtained by TEM measurements, we evaluated K = 6×10^-4 J×m^-3, which again is close to what commonly observed for magnetite nanoparticles for which the surface contribution causes an increase of the total magnetic anisotropy.²⁶

 

                 4. Conclusion

In summary, the concept of using the well-organised pore system of mesoporous hybrid silica acting as a nanoreactor for the growth and the organisation of nanoparticles appears to be a promising alternative on the route to nanocomposites materials. In this article, we prepared and organised Fe₃O₄ nanoparticles by organic solution-phase decomposition of the precursor Fe[(OC(CH₃))₂CH]₃ incorporated into a hybrid mesostructured silica functionalised with acetylacetonate groups. We obtained uniform-sized nanoparticles of Fe₃O₄ with a mean size value of 3-4.5 nm located exclusively inside of the silica matrix. The Fe₃O₄ content in this nanocomposite was shown to be 8.6 wt %. XRD, nitrogen physisorption, TEM and magnetic measurements confirm the formation of the magnetite nanoparticles. It is important to note that the silica functionalities play a crucial role in the synthesis of Fe₃O₄-containing nanocomposites. Indeed, the use of non-functionalised silica in the synthesis leads to a material, which presents aggregates of Fe₃O₄ nanoparticles at the grain surface with an iron content of only 0.94%. Another point to note is that considering the huge quantity of metal-acetylacetonate complexes known, this method opens the way to a large variety of new nanocomposite materials.

 

                 Acknowledgements

                 The authors thank Dr. Arie van der Lee (IEM, UMR ENSCM-UMII-CNRS 5625, Montpellier, France) for XRD measurements, the CNRS and the Université Montpellier II for financial support.

 

References

 

1          (a) A. P. Alivisatos, Science, 1996, 271, 933; (b) H. Weller, Angew. Chem. Int. Ed. Engl. 1993, 32, 41; (c) Nanoscale Materials in Chemistry, ed. K. J. Klabunde, Wiley-Interscience, New-York, 2001; (d) M. R Diehl, J.-Y. Yu, J. R. Heath, G. A. Doyle, S. Sun, C. B. Murray, J. Phys. Chem. B 2001, 105, 7913; (e) D. L. Leslie-Pelescky, R. D. Rieke, Chem. Mater. 1996, 8, 1770.

2          (a) D. D. Awschalom, D. P. Di Vicenzo, Phys. Today  1995, 4, 43; (b) K. Raj, R. J. Moskowitz, Magn. Magn. Mater. 1990, 85, 233; (c) M. L. Billas, A. Chatelain, W. A. de Heer, Science 1994, 265, 1682; (d) R. C. O’Handley, Modern Magnetic Materials, Wiley, New York, 1999.

3          L. Shen, P. E. Laibinis, T. A. Hatton, Langmuir 1999, 15, 447.

4          R. D. McMichael, R. D. Shull, L. J. Swartzendruber, L. H. Bennet, R. E. Watson, J. Magn. Magn. Mater, 1992, 111, 29.

5          B. R. Peters, R. A. Williams, C. Webb, Magnetic Carrier Technology, Butterworth-Heinemann, Ltd.: Oxford, England, 1992.

6          B. M. Berkovsky, V. F. Medvedev, M. S. Krakov, Magnetic Fluids: Engineering Applications, Oxford University Press, Oxford, 1993.

7          (a) D. Rabelo, E. C. D. Lima, A. C. Reis, W. C. Nunes, M. A. Novak, V. K. Garg, A. C. Oliveira, P. C. Morais, Nano Letters 2001, 1, 105; (b) P. C. Morais, R. B. Azevedo, D. Rabelo, E. C. D. Lima, Chem. Mater., 2003, 15, 2485; (c) A. A. Novakova, V. Yu. Lanchinskaya, A. V. Volkov, T. S. Gendler, T. Yu. Kiseleva, M. A. Moskvina, S. B. Zezin, J. Magn. Magn. Mater. 2003, 5, 976; (d) L. A. Harris, J. D. Goff, A. Y. Carmichael, J. S. Riffle, J. J. Harburn, T. G. ST. Pierre, M. Saunders, Chem. Mater. 2003, 15, 1367 ; (e) M. Breulmann, H. Gölfen, H.-P. Hentze, M. Antonietti, D. Walsh, S. Mann, Adv. Mater. 1998, 10, 237; (f) X. Yan, G. Liu, F. Liu, B. Z. Tang, H. Peng, A. B. Pakhomov, C. Y. Wong, Angew. Chem. Int. Ed. 2001, 40, 3593; (g) D. Szabo, G. Szeghy, M. Zrinyi, Macromolecules 1998, 31, 6541; (h) C. Castro, J. Ramos, A. Millan, J. Gonzalez-Calbet, F. Palacio, Chem. Mater. 2000, 12, 3681; (i) M. Zrinyi, Colloid. Polym. Sci. 2000, 278, 98.

8          R. S. Sapiezko, E. Matijevic, Colloid Interface Sci., 1980, 74, 405.

9          (a) G. Ennas, A. Musinu, G. Piccaluga, D. Zedda, D. Gatteschi, C. Sangregorio, J. L. Stanger, G. Concas, G. Spano, Chem. Mater. 1998, 10, 495; (b) J. Rose, M. M. Cortalezzi-Fidalgo, S. Moustier, C. Magnetto, C. D. Jones, A. R. Barron, M. R. Wisener, J.-Y. Bottero, Chem. Mater. 2002, 14, 621; (c) N. Leventis, I. A. Elder, G. J. Long, D. R. Rolison, Nano Letters 2002, 2, 63; (d) A. Bourlinos, A. Simopoulos, D. Petridis, H. Okumura, G. Hadjipanays, Adv. Mater. 2001, 13, 289.

10        (a) R. Köhn, D. Paneva, M. Dimitrov, T. Tsoncheva, I. Mitov, C. Minchev, M. Fröba, Micropor. Mesopor. Mater. 2003, 63, 125; (b) M. Iwamoto, T. Abe, Y. Tachibana, J. Mol. Catal. A: Chem., 2000, 155, 143; (c) Z. Y. Yuan, S. Q. Liu, T. H. Chen, J. Z. Wang, H. X. Li, Chem. Commun., 1995, 973; (d) T. Abe, Y. Tachibana, T. Uematsu, M. Iwamoto, Chem. Commun., 1995, 1617; (e) M. Fröba, R. Köhn, G. Bouffaud, Chem. Mater., 1999, 11, 2858; R. Köhn, M. Fröba, Catalysis Today, 2001, 68, 227.

11        T. A. Crowley, K. J. Ziegler, D. M. Lyons, D. Erts, H. Olin, M. A. Morris, J. D. Holmes, Chem. Mater. 2003, 15, 3518.

12        (a) Y. Guari, C. Thieuleux, A. Mehdi, C. Reyé, R. J. P. Corriu, S. Gomez-Gallardo, K. Philippot, B. Chaudret, R. Dutartre, Chem. Commun., 2001, 1374; (b) Y. Guari, C. Thieuleux, A. Mehdi, C Reyé, R. J. P. Corriu, S Gomez-Gallardo, K. Philippot, B. Chaudret, Chem. Mater., 2003, 15, 2017; (c) Y. Guari, K. Soulantica, K. Philippot, C. Thieuleux, A. Mehdi, C. Reyé, B. Chaudret, R. J. P. Corriu, New J. Chem., 2003, 7, 1029; (d) B. Folch, J. Larionova, Y. Guari, Ch. Guérin, A. Mehdi, C. Reyé, J. Mater. Chem. 2004, accepted ; (e) H. Wellmann, J. Rathousky, M. Wark, A. Zukal, G. Schylz-Ekloff, Micropor. Mesopor. Mater., 2001, 44-45, 419; (f) C. R. Patra, A. Ghosh, P. Mukherjee, M. Sastry, R. Kumar, Stud. Surf. Sci. Catal., 2002, 141, 641 ; (g) Y. S. Cho, J. C. Park, B. Lee, Y. Kim, J. Yi, Catal. Letters, 2002, 81, 89; (h) C.-H. Yang, P.-H. Liu, Y.-F. Ho, C.-Y. Chiu, K.-J. Chao, Chem. Mater., 2002, 15, 275 ; (h) L.-X. Zhang, J.-L. Shi, J. Yu, Z.-L. Hua, X.-G. Zhao, M.-L. Ruon, Adv. Mater., 2002, 14, 1510 ; (i) C.-M. Yang, H.-S. Sheu, K.-J. Chao, Adv. Funct. Mater., 2002, 12, 143.

13        (a) Sh. Sun, H. Zeng, J. Am. Chem. Soc., 2002, 124, 8204 ; (b) Y. Hou, J. Yu, S ; Gao, J. Mater. Chem., 2003, 13, 1983 ; (c) Z. Li, H. Che,, H. Bao, M. Gao, Chem. Mater., 2004, 16, 1391 ; (d) Y. Hou, S. Gao, T. Ohta, H. Kondoh, Eur. J. Inorg. Chem., 2004, 1169.

14        D. Zhao, J. Feng, Q. Huo, N. Melosh, G. H. Fredrickson, B. F. Chmelka, G. D. Stucky, Science, 1998, 279, 548; D. Zhao, Q. Huo, J. Feng, B. F. Chmelka, G. D. Stucky, J. Am. Chem. Soc., 1998, 120, 6024.

15        W. Rupp, N. Hüsing, U. Schubert, J. Mater. Chem., 2002, 12, 2594.

16        E. A. Boudreaux, L. N. Mulay, Theory and Applications of Molecular Paramagnetism, Eds. John Wiley Sons, New-York, 1976.

17        A Sayari, S. Hamoudi Chem. Mater. 2001, 13, 3151.

18        K. Nakamoto, Infrared and Raman Spectra, John Wiley and Sons Inc.: New York, 1986.

19        (a) W. Hammond, E. Prouzet, S. D. Mahanti, T. Pinnavaia, Microporous Mesoporous Mater. 1999, 27, 19 ; (b) M. Froba, R. Köhn, G. Bouffaud, Chem. Mater., 1999, 11, 2858 ; (c) F. J. Brieler, M. Fröba, L. Chen, P. J. Klar, W. Heimbrodt, Chem. Eur. J. 2002, 8, 185; (d) F. J. Brieler, P. Grundmann, M. Fröba, L. Chen, P. J. Klar, W. Heimbrodt, H.-A. K. von Nidda, T. Kurz, A. Loidl, J. Am. Chem. Soc. 2004, 126, 797.

20        B. J. Aronson, Ch. F. Blanford, A. Stein, J. Phys. Chem. B 2000, 104, 449; B. Marler, U. Oberhagemann, S. Vortmann, H. Gies, Micropor. Mater. 1996, 6, 375; B. J. Aronson, C. F. Blanford, A. Stein, J. Phys. Chem. B 2000, 104, 449.24.

21                    R. Kön, M. Fröba, Catal. Today 2001, 68, 227.

22        B. Lindlar, M. Boldt, S. Eiden-Assmann, G. Maret, Adv. Mater., 2002, 14, 1656.

23        R. M. Cornell, U. Schwertmann, The Iron Oxides VCH Weinheim, 1996.

24        L. Néel, Ann. Geophys. 1949, 60, 661.

25        J. L. Dormann, D. Fiorani, R. Tronc, Advances in Chemical Physics, Vol. XCVIII Eds. I. Prigogine, S. A. Rice Wiley & Sons Inc.: New York, 1997.

26        X. Batle, A. Labarta, J. Phys. D 2002, 35, R15.

Table

Table 1. Some relevant characteristics for the samples 1, 2 and 3: X-Ray diffraction powder peaks, nitrogen physisorption data and iron loading, wt %.

 

Scheme

Scheme 1. a) Synthesis of hybrid silica functionalized with acetylacetonato groups (^tBuOK = Potassium tert-butoxide, ^tBuOH = tert-Butanol and P123 = (PEO)₂₀(PPO)₇₀(PEO)₂₀; b) Representation of the used approach for the preparation of Fe₃O₄ nanoparticles into the hybrid silica.

 

Figure Captions

Figure 1. Powder X-ray diffraction patterns within the 2q range of 0.5-4 for materials 1 (-·-), 2 (-o-) and 3 (-¨-). Insert: Magnification of the powder X-ray diffraction patterns showing the (110) and (200) reflections.

Figure 2. Powder X-Ray diffraction pattern within the range of 2q 35 – 60° for nanocomposite 3.

Figure 3. Nitrogen adsorption/desorption isotherms for materials 1 (-¨-), 2 (-o-) and 3 (--). Insert : Barret-Joyner-Hellenda (BJH) pore size distribution calculated from the desorption branch of the nitrogen isotherms for 1 (-¨-), 2 (-o-) and 3 (--).

Figure 4. Transmission electron microscopy (TEM) images for: (a) the pristine hybrid silica 1. The insert shows the hexagonal ordering of the pores ; (b) the nanocomposite 3 containing Fe₃O₄ nanoparticles into the silica; Scale bars = 100 nm.

Figure 5. a) Transmission electron microscopy (TEM) images for Fe₃O₄ nanoparticles obtained from the nanocomposite 3 after removal of silica using the HF treatment (replicas); Scale bar = 50 nm; b) Histogram of Fe₃O₄ nanoparticles size distribution for the nanocomposite 3 showing the correlation between the size of nanoparticles and the pore diameter of the hybrid silica used (solid line for desorption data from N₂ studies).

Figure 6. Temperature dependence of the  ZFC (○) and FC (●) magnetisations for 3. Curves were measured with an applied magnetic field equals to 3.0 mT.

Figure 7. a) Field dependence of the magnetization at 2.5 and 300 K for sample 3 in a magnetic field up to 12 T; b) Hysteresis loop of 3 measured at 2.5 K.

Figure 8. Temperature dependence of the relaxation time of the magnetisation estimated from a.c. susceptibility measurements. Experiments were made in the 1.81-181 Hz frequency range.

 

            Supporting Materials

Figure 1S. Thermogravimetric analysis (ATG) and ATD curves for the hybrid silica 1.

Figure 2S. Infrared spectra of a) the mesoporous silica (CH₃C(O))₂C(CH₂)₃SiO_1.5/9SiO₂ 1. Insert : Magnification in the range 1400-1800 cm^-1 ; b) the nanocomposite 2. Insert : Magnification in the range 1400-1800 cm^-1.

Figure 3S. Transmission electron microscopy (TEM) image for the material obtained after the incorporation of Fe₃O₄ nanoparticles into the non-functionalized silica SBA-15 (pore size of 5.4 nm) showing the extrusion of the iron oxide out of the silica pores. Scale bar = 100 nm.

Figure 4S. Transmission electron microscopy (TEM) image for the composite material 2. Scale bar = 100 nm.

 

 

 

 

 

Table 1. Some relevant characteristics for the samples 1, 2 and 3: X-Ray diffraction powder peaks, nitrogen physisorption data and iron loading, wt %.

 

 

^aDetermined from TGA. ^bDetermined from elemental analysis. ^cDetermined as the maximum of the Barret-Joyner-Hellenda (BJH) pore size distribution calculated from the desorption branch of the nitrogen isotherm. ^dCalculated by a₀ - pore size [a₀ : 2d₁₀₀/3 ^(1/2)]

 

 

 

 

 

 

 

 

a)

b)

 

Scheme 1. a) Synthesis of hybrid silica functionalized with acetylacetonato groups (^tBuOK = Potassium tert-butoxide, ^tBuOH = tert-Butanol and P123 = (PEO)₂₀(PPO)₇₀(PEO)₂₀; b) Representation of the used approach for the preparation of Fe₃O₄ nanoparticles into the hybrid silica.

 

Figure 1. Powder X-ray diffraction patterns within the range 2q of 0.5-4 for materials 1 (-·-), 2 (-o-) and 3 (-¨-). Insert: Magnification of the powder X-ray diffraction patterns showing the (110) and (200) reflections.

 

 

 

 

 

 

 

 

 

 

Figure 2. Powder X-Ray diffraction pattern within the range of 2q 35 – 60° for nanocomposite 3.

 

 

 

 

3

2

1

3

2

1

Figure 3. Nitrogen adsorption/desorption isotherms for materials 1 (-¨-), 2 (-o-) and 3 (--). Insert : Barret-Joyner-Hellenda (BJH) pore size distribution calculated from the desorption branch of the nitrogen isotherms for 1 (-¨-), 2 (-o-) and 3 (--).

 

 

 

 

 

 

 

 

 

 

a)

 

b)

 

Figure 4. Transmission electron microscopy (TEM) images for: (a) the pristine hybrid silica 1. The insert shows the hexagonal ordering of the pores ; (b) the nanocomposite 3 containing Fe₃O₄ nanoparticles into the silica; Scale bars = 100 nm.

 

Figure 5. a) Transmission electron microscopy (TEM) images for Fe₃O₄ nanoparticles obtained from the nanocomposite 3 after removal of silica using the HF treatment (replicas); Scale bar = 50 nm; b) Histogram of Fe₃O₄ nanoparticles size distribution for the nanocomposite 3 showing the correlation between the size of nanoparticles and the pore diameter of the hybrid silica used (solid line for desorption data from N₂ studies).

 

Figure 6. Temperature dependence of the  ZFC (○) and FC (●) magnetisations for 3. Curves were measured with an applied magnetic field equals to 3.0 mT.

 

 

 

 

 

 

 

 

 

 

 

 

 

 

a)                                                                    b)

Figure 7. a) Field dependence of the magnetization at 2.5 and 300 K for sample 3 in a magnetic field up to 12 T; b) Hysteresis loop of 3 measured at 2.5 K.

 

 

 

 

 

 

 

 

 

 

 

 

 

Figure 8. Temperature dependence of the relaxation time of the magnetisation estimated from a.c. susceptibility measurements. Experiments were made in the 1.81-181 Hz frequency range.