J Nanopart Res (2012) 14:894 DOI 10.1007/s11051-012-0894-5
RESEARCH PAPER
Liquid-phase syntheses of cobalt ferrite nanoparticles Katalin Sinko´ • Enik} o Manek • • ´ Aniko Meiszterics Ka´roly Havancsa´k Ulla Vainio • Herwig Peterlik
•
Received: 3 October 2011 / Accepted: 28 April 2012 Ó Springer Science+Business Media B.V. 2012
Abstract The aim of the present study was to synthesize cobalt-ferrite (CoFe2O4) nanoparticles using various liquid phase methods; sol–gel route, co-precipitation process, and microemulsion technique. The effects of experimental parameters on the particle size, size distribution, morphology, and chemical composition have been studied. The anions of precursors (chloride and nitrate), the solvents (water, n-propanol, ethanol, and benzyl alcohol), the precipitating agent (ammonia, sodium carbonate, and oxalic acid), the surfactants (polydimethylsiloxane, ethyl acetate, citric acid, cethyltrimethylammonium bromide, and sodium dodecil sulfate), their concentrations, and heat treatments were varied in the experiments. The smallest particles (around 40 nm) with narrow polydispersity and spherical shape could be achieved by a simple, fast sol–gel technique in the
K. Sinko´ (&) E. Manek A. Meiszterics Institute of Chemistry, L. Eo¨tvo¨s University, Budapest 1117, Hungary e-mail:
[email protected] K. Havancsa´k Institute of Physic, L. Eo¨tvo¨s University, Budapest 1117, Hungary U. Vainio DESY, 22607 Hamburg, Germany H. Peterlik Faculty of Physics, University of Vienna, Vienna 1090, Austria
medium of propanol and ethyl acetate. The size characterization methods have also been investigated. Small-angle X-ray scattering (SAXS), dynamic light scattering (DLS), and scanning electron microscopy (SEM) provide the comparison of methods. The SAXS data correspond with the sizes detected by SEM and differ from DLS data. The crystalline phases, morphology, and chemical composition of the particles with different shapes have been analyzed by X-ray diffraction, SEM, and energy dispersive X-ray spectrometer. Keywords Cobalt ferrite nanoparticles Sol–gel method Co-precipitation DLS SEM SAXS
Introduction Recently, nanostructured transition metal oxides have attracted a lot of attention because of their outstanding properties and various applications. The properties (e.g., magnetic, optic, catalytic, and electronic) of nanomaterials depend strongly on their microstructural features such as morphology, crystallite size, and porosity. Magnetite nanoparticles may behave as single magnets when the domain size is as large as the particle. Cobalt ferrite (CoFe2O4) is a well-known hard magnetic material with high coercivity and
123
Page 2 of 14
moderate magnetization. The magnetic properties of the ferrites, MFe2O4, are accompanied with the cation configuration of the spinel lattice. CoFe2O4 can be usually characterized by an inverse spinel structure. CoFe2O4 nanoparticles have a broad prospect of applications, e.g., in electronic devices, ferrofluids, and high-density information storage (Laurent et al. 2008; Sun and Zeng 2004). The magnetite nanoparticles could have many applications in the medical diagnostics and therapy; targeted drug delivery (Jain et al. 2005; Chourpa et al. 2005); magnetic resonance imaging (MRI) as contrast agents (Bulte 2006; Burtea et al. 2005; Boutry et al. 2006); and tissue repair and cell separation (Gupta and Gupta 2005). Such magnetic nanoparticles can bind to drugs, proteins, enzymes, antibodies, or nucleotides and can be directed to an organ, tissue, or tumor using an external magnetic field (Chastellain et al. 2004). In contrast to the metal nanoparticles, the nano ferrites are very stable in different chemical environments, which provide the ferrites with great importance (Laurent et al. 2008) in the biomedical research. All these biomedical applications require that the nanoparticles have high magnetization values, a size smaller than 100 nm, and a narrow particle size distribution. Many different synthesis techniques give access to nanomaterials with a well-defined crystallite size. The liquid-phase syntheses offer a good technique and control for tailoring the structures, the compositions, and the morphological features of nanomaterials. The liquid-phase routes include the co-precipitation, the hydrolytic as well as the nonhydrolytic sol–gel processes, the hydrothermal or solvothermal methods, the template synthesis, and microemulsion-based processes. In the sol–gel synthesis, soluble precursor molecules are hydrolyzed and condensed to form a dispersion of nano-sized particles. In the preparation of cobalt ferrite powders, inorganic cobalt and ferric salts are subjected to hydrolysis and condensation generally in ethanol at 50–80 °C (Silva et al. 2005; Lee et al. 1998; Meron et al. 2005). Further heat treatments are needed to develop the final crystalline state. The drying process and the heat treatment of wet sol solutions have a strong effect on the surface area, pore volume, crystallinity, particle structure, and corresponding electrochemical properties (Laurent et al. 2008; Ennas et al. 1998; Brinker and Sherrer 1990). Modifying agents are often applied in the sol–gel technique; e.g.,
123
J Nanopart Res (2012) 14:894
citric acid (Liu and Zhang 2009) or polyvinyl alcohol (Pramanik et al. 2004). Some experiments on the sol– gel synthesis performed in benzyl alcohol have been recently reported (Murray and Agan 2000; Pinna and Niederberger 2008; Bilecka et al. 2008). According to the published results, additional surfactant is not needed in the preparation of metal oxides in benzyl alcohol. The benzyl alcohol is acting as solvent, ligand, and reactant in the synthesis. The co-precipitation is a simple and rapid technique. This method provides several possibilities to modify the particle size, surface, and shape. However, the control of the particle size and distribution is difficult. A commonly used procedure for preparing ferrite particles has been the co-precipitation of M2? and Fe3? ions by a base, usually NaOH or NH3, in an aqueous solution (Zhang et al. 1998; Neveu et al. 2002; Olsson and Salazar-Alvarez 2005; Chinnasamy et al. 2003). There are few examples for the application of hexamethylene tetramin (Liu et al. 2008) or tetraalkyl ammonium hydroxides (Paike et al. 2007; Gyergyek and Makovec 2010) as a precipitating agent. Nanoparticles can also be formed on liquid–liquid interfaces. Among the chemical methods, the microemulsion process involving reverse micelles has been demonstrated as a versatile method for obtaining a wide variety of nanocrystalline oxides (Pillai and Shah 1996; Ahn et al. 2001; Moumen and Pileni 1996; Han et al. 2004). For example, cobalt ferrite nanoparticles could be prepared by microemulsion method from a mixture of Co(II) and Fe(III) dodecylsulfates treated with an aqueous solution of methylamine (Moumen and Pileni 1996). Monodisperse CoFe2O4 nanocrystals have been synthesized using normal and reverse micelle microemulsion methods and by combining a non-hydrolytic process and seed-mediated growth (Han et al. 2004). The hydrothermal technique has been commonly used to prepare ferrite nanoparticles (Li and Xu 2010; Rebolledo et al. 2008; Komarneni et al. 1998; Bilecka and Niederberger 2010). Most of these preparations involve a combination of co-precipitation and hydrothermal synthesis (Rebolledo et al. 2008). An innovation to the hydrothermal method is the application of microwaves during the hydrothermal synthesis (Komarneni et al. 1998; Bilecka and Niederberger 2010). The aim of the present study was to prepare ferromagnetic cobalt ferrite (CoFe2O4) nanoparticles by different liquid-phase syntheses; sol–gel, co-
J Nanopart Res (2012) 14:894
precipitation, and microemulsion techniques combined with thermal decomposition. We have monitored the effect of the synthesis route, the type of precursors and solvents, the chemical compositions, the concentration of the initial materials, the application of surfactants, and the heat treatments on the size, and the chemical and crystalline compositions of the particles. The research study also concentrated on a comparison of the various size characterization methods. The particle sizes have been determined by dynamic light scattering (DLS), scanning electron microscope (SEM), and small angle X-ray scattering (SAXS). The identification and characterization of the phases and the morphology of products have been performed by X-ray diffraction (XRD) and scanning electron microscope (SEM). The processes and the weight loss occurred by heat treatments have been recorded by thermal analysis in a controlled atmosphere.
Experimental methods Preparation methods In the experiments of the sol–gel method, CoCl26H2O (Aldrich, a.r.) and Co(NO3)26H2O (Aldrich, a.r.) were provided as cobalt precursor, FeCl36H2O (Aldrich, a.r.) and Fe(NO3)39H2O (Aldrich, a.r.) as iron precursor. The Co2? and Fe3? ions were allowed to hydrolyze for 1–2 h at 50 °C in first step and for 2–6 h at 85 °C in the second step in ethanol or 1-propanol solutions with (citric acid, PDMS, or ethyl acetate) or without any surfactant. The chemical compositions of the preparation experiments are summarized in Table 1. By partial evaporation of the solvent, a precipitate formed. The sol–gel technique produced mixed basic Co- and Fe-containing precipitates. The precipitates were centrifuged and dried at 80 °C. The heat treatment was carried out at different temperatures under oxidative atmosphere to obtain cobalt ferrite particles. The initial Co- and Fecontaining solutions were treated either in a common system or separately. In order to avoid the usual problem of ferrite preparations, e.g., the formation of hematite, the alcoholic solution of ferric salts was slowly added with a dropping rate of 0.5 cm3 min-1 into the alcoholic solution of cobalt(II) salts at 80 °C. After 5-h reflux of mixture at 80 °C, a gelatinous
Page 3 of 14
precipitate formed. The average yield was very low (5–6 %) in the case of chloride precursor and the use of nitrate salts yielded 45–60 % of the theoretical mass (Table 1). In the experiments of the co-precipitation method, cobalt and ferric chloride were the initial materials. The aqueous solution of precipitating agents (sodium carbonate, oxalic acid, and ammonia) was dropped into the aqueous solution of precursors. Polydimethylsiloxane (PDMS, Aldrich, 550 g mol-1, 5600 g mol-1); polyethylene glycol tert-octylphenyl ether (Triton X-100, Aldrich); sodium dodecyl sulfate (NaDS, Merck); sodium dodecylbenzene sulfonate (NaDBS, Aldrich); and tetradodecylammonium bromide (TDAB, Aldrich) were used as surfactants in the concentration of 0–10 w/w%. Precipitates formed directly during the addition of precipitating agents. The suspensions were stirred for 2 h at room temperature with or without a surfactant. The particles were separated by centrifugation. The dried precipitates were subjected to heat treatment at various temperatures. The average yield was 66 % of the theoretical mass by carbonate precipitator and 49 % by oxalate or ammonia agents. The system of water-in-oil (W/O) microemulsion method consists of sec. buthylalcohol as an oil phase, CTAB, NaDS, and PDMS as surfactans and an aqueous phase of cobalt and ferric salts. Aqueous solution was prepared by dissolving stoichiometric amounts of cobalt and ferric chloride in deionized water. Sodium carbonate was taken as a precipitating agent. The precipitating agent was separately dissolved in water. The aqueous solution of precursors and after that the aqueous solution of precipitating agent were dropped into the surfactant-containing oil phase during intensive mixing. The microemulsion synthesis of the nanoparticles could be carried out with yield of 20–30 %. Characterization methods Dynamic light scattering (DLS) measurements were performed by means of a DLS equipment (Brookhaven) consisting of a BI-200SM goniometer and a BI9000AT digital correlator. An argon-ion laser (Omnichrome, model 543AP) operating at 488-nm wavelength and emitting vertically polarized light was used as the light source. The signal analyzer was used in real-time ‘‘multi tau’’ mode. In this mode, the time
123
Page 4 of 14
J Nanopart Res (2012) 14:894
Table 1 Chemical compositions in preparation experiments Methods
Sol gel Sol gel Sol gel
Molar ratios
Yield w/w%
Co-precursor
Fe-precursor
Solvent
Precipitator
Additives
Nitratea
Nitrateb
n-propanol
–
Ethyl acetate 80
50–60
1
1
70
Nitratea
Nitrateb
Ethanol
–
Ethyl acetate 10–100
45–55
1
1
70
Nitratea
Nitrateb1
–
Citric acid 0.1–1.0
10–15
Sol gel
–
PDMS 550 0.1–5
10–25
Sol gel
–
–
1–5
Sol gel
–
Ethyl acetate 80
5–6
Sol gel
–
–
1-5
Ethanol 70
1 Nitratea
Nitrateb
Ethanol
1
1
70
Nitratea
Nitrateb
Benzyl alcohol
1
1
18–280
Chloridec
Chlorided
n-propanol
1
1
70
Chloridec
Chlorided
Benzyl alcohol
1
1
18–280
Co-precipitation
Chloridec 1
Chlorided 1
Water 40–200
Na2CO3 2.5–3
PDMS 550 0–10e
60–70
Co-precipitation
Chloridec
Chlorided
Water 80
Na2CO3 2.5–3
PDMS 5600 0–10e
50–70
1
1
Co-precipitation Co-precipitation
Chloridec 1 c
Chloride
Chlorided 1
Water 50
(COO)2 2.5–3
PDMS 550 0–10e
40–50
d
Chloride 1
Water 50
(COO)2 2.5–3
PDMS 5600 0-10e
40–50
Chlorided
Water 80
NH3 5–7
PDMS 5600 0–10e
40–60
1 Co-precipitation
Chloridec 1
1 Microemulsion
Chloridec 1
Chlorided 1
Water 40–100 secbutanol 100
Na2CO3 2.5–3
PDMS 5600 5, 10e
25–30
Microemulsion
Chloridec
Chlorided
Na2CO3 2.5–3
CTAB 5, 10e
25–30
Microemulsion
1 Chloridec
1 Chlorided
Water 40–100 secbutanol 100
Na2CO3 2.5–3
NaDS 5, 10e
20–30
1
1
Water 40–100 secbutanol 100
a
Co(NO3)26H2O; bFe(NO3)39H2O; cCoCl26 H2O; dFeCl36H2O; ew/w%
axis was logarithmically spaced over an appropriate time interval and the correlator used 218 time channels. The pinhole size was 100 lm. The particles were generally dispersed in ethanol for DLS measurements instead of water to avoid the aggregation of the particles in water. The number-weighted particle size distribution was detected by DLS. The particle size and morphology were studied by a FEI Quanta 3D FEG SEM. The SEM images were prepared by the Everhart–Thornley secondary electron detector (ETD), its ultimate resolution is 1–2 nm.
123
Since the conductance of the particles investigated is high enough to remove the electric charge accumulated on the surface, the SEM images were performed in high vacuum without any coverage on the specimen surface. For the best SEM visibility, the particles were deposited on a HOPG (graphite) substrate surface. SEM combined with energy disperse X-ray spectroscopy (EDX) is mainly applied for spatially resolved chemical analysis of bulk samples. SAXS experiments were conducted on several instruments. The laboratory equipment was operated
J Nanopart Res (2012) 14:894
with a 5.4 kW rotating anode X-ray generator (Nanostar from Bruker AXS, Karlsruhe), a pinhole camera with variable sample to detector distance (25–108 cm), and a 2D position sensitive detector (Bruker AXS). The gels were covered in vacuum tight foil. The 2D spectra were corrected for parasitic pinhole scattering, as well as for the foil scattering. Simultaneous small-angle and wide-angle X-ray scattering experiments (SAXS and WAXS) were also recorded on the JUSIFA beamline of HASYLAB at DESY in Hamburg (8 keV photon energy; 925-, and 3625-mm sample-to-detector distances). The SAXS intensities were fitted with a form factor from spheres with a Gaussian size distribution. In the case of the small particles, the fit can be slightly improved by an additional structure factor using the local monodisperse approximation. However, as the tendency for agglomeration is small (described by a low value for the hard-sphere volume factor), the structure factor was set to one in all samples for an easier comparison of the data. The WAXS curves were evaluated by means of the standard PDF cards. The XRD measurements were carried out by means of a Philips (PW1130) X-ray generator set up with a Guinier-chamber. The chamber has a diameter of 100 mm, and the patterns were recorded on FUJI Imaging Plates (BAS MS2025). The XRD data were collected over the 2h range of 9–90° with a step size 0.005°. The identification of phases was carried out by comparing the diffraction patterns with the standard PDF cards. Thermogravimetric analysis (thermogravimetry— TG; and differential thermal analysis—DTA) was used to investigate the processes that occurred during the heat treatment. TG and DTA curves were recorded using Derivatograph-C System (MOM, Hungary) under air or nitrogen flow at a heating rate of 6 °C min-1 on crushed bulk specimens from room temperature to 1,000 °C.
Results Particle size measurements The particle sizes have been determined by various techniques: DLS, SAXS, and SEM. For the exact comparison of various methods, a sample having
Page 5 of 14
nearly monodisperse size distribution was chosen. The results, which are summarized in Table 2, show considerable differences between different characterization techniques. The sizes (diameters) obtained by SEM and SAXS are consistent with each other. The size derived from DLS is two or three times larger. SEM delivers direct images of the size and shape of solid nanoparticles, and the photograph is taken under vacuum. Dried powders have also been measured under vacuum in the SAXS experiments. In DLS technique, the particles are dispersed in a solvent. The nanoparticles can be hydrated or solvated in polar solutions. The difference in the sizes might be attributed to the hydration/solvation shells. In order to verify the influence of hydration/solvation shells on the particle size, we measured the size of nanoparticles by DLS in various solvents. The sizes obtained in aqueous solution are significantly bigger than those detected in ethanol solutions (Table 2). The dipole moment of water is larger than that of ethanol resulting in a stronger connection of solvent molecules to the particle surfaces. Tobler et al. provide further explanation for the size difference (2009). The highly hydrous and open-structured particles (e.g., silica) can collapse because of the dehydration and relaxation processes under high vacuum (Tobler et al. 2009). In the DLS measurements, surfactants were used to hinder the aggregation of nanoparticles in their aqueous dispersions. The usually applied ionic surfactants proved to be ineffective against aggregation (Table 2). The particle sizes identified by various methods are listed in Table 3. The size means diameter. The particles were synthesized by different routes and Table 2 Particle size determination by various techniques Method
SEM
Average sizea (nm) 40
Size-range (nm) 30–60
SAXS
40
42–54
DLS in ethanol
86
53–143
DLS in water
120
85–135
DLS in CTABb
153
120–180
DLS in NaDSc
205
140–255
a
Number-weighted average values
b
10 mM aqueous solution of CTAB
c
10 mM aqueous solution of NaDS
123
Page 6 of 14
J Nanopart Res (2012) 14:894
Table 3 Particle size and distribution of cobalt ferrite particlesa synthesized by different methods Preparation technique
DLSb Average size (nm)
SEM Size-range (nm)
Average size (nm)
SAXS Size-range (nm)
Average size (nm)
Mod. sol gel nitrates, ethyl acetate
86
53–143
40
30–60
40 ± 6
Sol gel nitrates, ethyl acetate
72
26–139
41
23–67
40 ± 4
Sol gel chlorides, ethyl acetate
100
80–168
58
45–98
–
Sol gel nitrates, citric acid
156
95–210
–
–
–
Sol gel nitrates, PDMS 5600
102
76–142
–
–
–
Co-precipitation carbonate
155
90–205
52
12–54
55 ± 10
Co-precipitation carbonate, 5 % PDMS 5600
122
47–157
43
26–79
40 ± 6
Co-precipitation carbonate, 5 % PDMS 550
139
50–166
40
22–68
–
Co-precipitation oxalate
141
86–209
58
24–140
–
Co-precipitation oxalate, 5 % PDMS 5600
89
53–135
79
44–118
–
Microemulsion carbonate, 5 % PDMS 5600 Microemulsion carbonate, 5 % CTAB
447 392
222–476 264–579
31, 1400 27, 1550
30–2500 20–2200
– –
a
The precipitates were heated at 600 °C under air
b
DLS measurements were carried out in ethanol.
–, No data
dried at 80 °C for 2 h, and heat treated at 600 °C also for 2 h. The sizes measured by DLS are generally two or three times larger than those derived from SEM or SAXS. For the correct explanation of these differences, an additional reason must be taken into account apart from those detailed above. That is, by the variation of precipitated particles, not only cobalt ferrite is formed in the preparation, but even a small amount of larger particles can also significantly modify the average particle size and size distribution in DLS method. SAXS technique is less susceptible to the presence of larger aggregates, and SEM is capable of distinguishing between the particle types. The difference between the sizes depends on the hydrophobicity, the roughness of the particle surface, and the particle shape. The amorphous shape (e.g., sol–gel derived products) and the rough or porous surface (e.g., sol–gel-derived and carbonate-co-precipitated products) thicken the hydration/solvation layers. The smallest cobalt ferrite nanoparticles could be achieved by an uncomplicated, fast sol–gel method starting from nitrate salts (40–41 nm, SEM) and by coprecipitation with carbonate in the presence of 5 w/w% PDMS (40–43 nm, SEM). The co-precipitation with oxalate acid yielded slightly bigger (58 nm) particles. Among the surfactants, the PDMS of 550 or 5,600 g mol-1 proved to be the most effective in the
123
reduction of the size and size distribution in the series of co-precipitation. The particle size in the function of PDMS concentration shows a minimum (Fig. 1). The smallest particles (122 nm, DLS) with less polydispersity (47–157 nm, DLS) could be obtained in the solution of 5.0 w/w% for both PDMS 550 and 5,600 g mol-1 in the co-precipitation series with carbonate (Fig. 1). A minimum can also be observed at 5.0 w/w% PDMS in the size of particles prepared with oxalate co-precipitation. In the series of sol–gel technique, the use of ethyl acetate in large amount (about 80 mol ethyl acetate/ Co2?) produced the smallest particles (40–41 nm, SEM). The application of microemulsion technique yielded the widest size distribution (20–2200 nm, SEM) because of the several species. The modifying agents decreased consistently the size in the function of surfactant concentration. For example, raising the volume of surfactants from 5 to 10 w/w% the mean size reduced by 190–221 nm in every case, from 347 to 141 nm by NaDS; from 392 to 171 nm by CTAB; from 447 to 250 nm by PDMS. Shape and morphology of nanoparticles The shape and the morphology of nanopowders were controlled by SEM and XRD. The nanoparticles were produced by various synthesis techniques, dried at
J Nanopart Res (2012) 14:894
Page 7 of 14
Particle size (nm)
by addition of a surfactant. The microemulsion products consist of many types of particles; fine nanoparticles with amorphous shape (cobalt ferrite, XRD, and EDX), octahedral crystals with average size of 1.55 lm (hematite, XRD, EDX), plate-like aggregates of 1.0–2.2 lm (NaCl, XRD, and EDX), and rodlike aggregates of 1.0–2.0 lm (iron oxide, and EDX). Effect of heating process
carbonate 550 carbonate 5600 oxalate 5600
PDMS (w/w%)
Fig. 1 Particle sizes of carbonate and oxalate precipitates versus concentration of PDMS (550 and 5600 g mol-1)
80 °C for 2 h and heat treated at 600 °C also for 2 h. The sol–gel process yields spherical cobalt ferrite nanoparticles from nitrate salts and ethyl acetate (Figs. 2, 3). Without a slow addition of Fe(NO3)3 solution, a small amount of hematite is precipitated from the common solution of nitrate salts and large amount from chloride salts (Fig. 4). Octahedral crystals of large size represent the hematite phase (Fig. 3). XRD identifies only cobalt ferrite crystalline phase using a slow addition of ferric nitrate solution (Fig. 4). The cobalt ferrite phase can be readily detected in the gels obtained by sol–gel technique and dried at 80 °C. By other preparation routes, hematite (Fe2O3) always forms over cobalt ferrite. Co-precipitation with carbonate without any surfactant produces inhomogeneous particles; nanoparticles with amorphous shape (cobalt ferrite, verified by XRD and EDX), plate-like aggregates of 1.5–2 lm (NaCl, XRD, and EDX), octahedral crystals with average size of 320 nm (hematite, XRD, and EDX) (Figs. 3, 5). The inhomogeneity and the size of particles reduce by the effect of PDMS, the volume of ferrite phase increases, and the particle shape is cubic rather than amorphous (Figs. 3, 6). In the product of the co-precipitation with oxalic acid, nanoparticles with amorphous shape (cobalt ferrite, XRD, and EDX) and octahedral crystals with average size of 97 nm (hematite, EDX) can be revealed (Figs. 3, 5). The size and its dispersion change slightly
The nanoparticles obtained by various synthetic routes were dried at 80 °C for 2 h to evaporate the main part of solvents. The nanopowders dried at 80 °C have been investigated with thermoanalysis and XRD. The processes of weight loss finish by \300 °C in the samples of the sol–gel technique starting from nitrate salts (by 260–285 °C) and the co-precipitation method using oxalic acid agent (by 210 °C) (Figs. 7, 8). The processes of weight loss continue until 500–700 °C in the other samples (Figs. 7, 8). The weight loss in the range from 25 to 100–150 °C is generally 5–7 %, and it can be attributed to the evaporation of residual solvents (e.g., water and n-propanol). The temperature range of 100–190 °C belongs to the volatilization of bonded water content (e.g., crystalline water, molecules of hydration layers) in precipitates, its weight loss changes between 5 and 20 %. The nitrate content, the N-containing organic molecules derived from the reaction of 1-propanol and nitrate ions, and the organic molecules connected around the metal ions escape between 180 and 280 °C in two or three steps. The combustion of organic molecules is indicated by the exothermic changes on the DTA curves. The decarbonization of carbonate precipitates occurs between 150 and 300 °C. The chloride ions may decompose above 300–400 °C (Figs. 7, 8). The sol–gel derived precipitate dried at 80 °C proved to be amorphous (XRD) basic nitrate/chloridecontaining salts (TA). The precipitate obtained by a slow addition of ferric nitrate solution has a much lower nitrate content than that of precipitate produced by a regular sol–gel route (DTA) and includes already some cobalt ferrite ordering (XRD). The small basic chloride-containing residue consists mostly of CoCl22H2O (XRD). The products of co-precipitation with sodium carbonate are amorphous (XRD) basic carbonate salts (TA). By oxalic acid, CoC2O42H2O precipitates (XRD). The samples produced by microemulsion technique contain many compounds proved
123
Page 8 of 14
J Nanopart Res (2012) 14:894
Fig. 2 SEM image of the nanopowders synthesized by modified sol–gel method from nitrate salts
Fig. 3 SEM images of the nanopowders prepared by various techniques. The samples were prepared by 1 modified sol–gel method from nitrate salts; 2 sol–gel method from nitrate salts; 3 sol–gel method from chloride salts; 4 microemulsion; 5 coprecipitation with carbonate and 5 % PDMS of 5,600 g mol-1;
123
6 co-precipitation with carbonate; 7 co-precipitation with oxalate and 5 % PDMS of 5,600 g mol-1; 8 co-precipitation with oxalate. The samples were heated at 600 °C. With exceptions (4, 6), the magnification is 9100,000
J Nanopart Res (2012) 14:894
Fig. 4 X-ray diffraction patterns of sol–gel derived nanopowders heated at 600 °C. The samples were prepared by 1 modified sol–gel method from nitrate salts; 2 sol–gel method from nitrate salts; 3 sol–gel method from chloride salts
by TA (Fig. 8) and SEM (Fig. 3). The main component is an amorphous cobalt carbonate salt with less OH groups (XRD). The effect of heat treatment was investigated by in situ, small and wide angle X-ray scattering (SAXS, WAXS). Figs. 9 and 10 represent these measurements; the SAXS (Fig. 9) and WAXS (Fig. 10) investigations were carried out on the best sample obtained by modified sol–gel method, i.e., using a slow addition of the ferric nitrate solution. The SAXS curves indicate particle sizes of \15 nm in the temperature of 20–400°C (Fig. 9). The size of particles grows significantly up to 40–44 nm above 400 °C. Above 600 °C, a further growth can be observed (82 nm). The particle sizes at C800 °C cannot be detected by SAXS because the size is too large ([100 nm) for SAXS range. A significant change can also be monitored by WAXS between 400 and 500 °C (Fig. 10). WAXS identifies only some ferrite ordering in the samples heated at B400 °C. The real crystalline cobalt ferrite phase may be detected in the nanopowders heat treated at C500 °C. Thus, the crystallization results in a dramatic change in the particle size above 400 °C.
Page 9 of 14
Fig. 5 X-ray diffraction patterns of nanopowders prepared by co-precipitation and heated at 600 °C. The samples were precipitated by 4 carbonate without any surfactant; 5 carbonate and 5 % PDMS of 5,600 g mol-1; 6 oxalate and 5 % PDMS of 5,600 g mol-1; 7 carbonate using microemulsion technique
The co-precipitated nanopowders are also very fine (8–10 nm, SAXS) after a drying at 80 °C. The particle sizes grow continuously with the temperature of heat treatment. The sizes obtained by carbonate precipitation in the presence of PDMS are around 20 nm at 400°C; &30 nm at 500 °C; and 40–45 nm at 600 °C. WAXS as well as SAXS indicate a slow structural transformation between 300 and 400 °C. WAXS identifies already crystalline phases (NaCl, cobalt ferrite, and hematite) in the sample of 400 °C; however, the well-developed crystals of cobalt ferrite appear only at 600–700 °C.
Discussion Sol–gel method A new, simple, fast way of sol–gel method has been developed for preparation of cobalt ferrite nanoparticles. The new sol–gel route has nitrate salts reacted in
123
Page 10 of 14
J Nanopart Res (2012) 14:894
Fig. 6 SEM image of the nanopowders precipitated by carbonate and 5 % PDMS of 5,600 g mol-1
Fig. 7 Thermoanalysis of sol–gel derived nanopowders dried at 80 °C. The samples were prepared by 1 modified sol–gel method from nitrate salts; 2 sol–gel method from nitrate salts; 3 sol–gel method from chloride salts
Fig. 8 Thermoanalysis of nanopowders prepared by co-precipitation and dried at 80 °C. The samples were precipitated by 4 carbonate without any surfactant; 5 carbonate and 5 % PDMS of 5,600 g mol-1; 6 oxalate and 5 % PDMS of 5,600 g mol-1; 7 carbonate using microemulsion technique
1-propanol in the presence of ethyl acetate. The nitrate salts proved to be a more efficient precursor for the sol– gel technique than chloride. The application of chloride precursors yields very small amount of particles: 5–6 % of theoretical mass. In the case of nitrate salts, the average yield is 45–60 %. In the solution of
nitrates, the condensation reactions are more intensive. The hydrolysis of metal ions produces OH groups which make the condensation possible. A part of nitrate content escapes as nitrous gases during the reactions increasing the pH that also supports the condensation. The decomposition of nitrate ions depends on the
123
J Nanopart Res (2012) 14:894
Page 11 of 14
Fig. 9 SAXS patterns of the nanopowders synthesized by modified sol–gel method from nitrate salts in the function of temperature
Fig. 10 WAXS patterns of the nanopowders synthesized by modified sol–gel method from nitrate salts in the function of temperature
polarity of the medium. The lower the polarity, the more intensive the decomposition. Thus, the results of experiments in 1-propanol are more impressive than in ethanol. The chloride ions can not escape during the gelation at 80 °C. The sol–gel method also needs any of the surfactants to obtain nanoparticles. Citric acid, PDMS of 550
and 5,600 g mol-1 molecular weights, and ethyl acetate were applied as surfactants in the sol–gel procedures. The smallest particle size could be achieved in the presence of ethyl acetate (Table 3). Application of ethyl acetate in the concentration of 40 w/w% yielded spherical cobalt ferrite nanoparticles of average diameter of 40 nm (SEM) with narrow polydispersity (30–60 nm). If the common solution of precursors is subjected to reaction and heating, then hematite (Fe2O3) always forms. (See Fig. 4) The formation of hematite can be avoided by a slow addition of alcoholic solution of Fe(NO3)3 to the solution of Co(NO3)3. The slow addition of ferric nitrate solution results in the finest particles and the lowest temperature for the reactions. The decomposition and the combustion of organic compounds and the bonded nitrate content occur in one step between 175 and 260 °C. That proves that the nitrate content is less bonded in the particles and escapes mostly as nitrous gases during the gelation. The precipitate contains a smaller amount of nitrate ions than the product made by regular sol–gel route of a common precursor solution. Treating a common precursor solution, the processes of weight loss are carried out in three steps until 285 °C. The gelation using a slow addition of ferric nitrate solution produces an amorphous basic nitrate-containing salt with some cobalt ferrite
123
Page 12 of 14
ordering and with average size of \10 nm at 80 °C. The particle sizes increase significantly to 40–44 nm (SAXS) above 400 °C accompanied by the appearance of the crystalline ferrite phase. The use of benzyl alcohol instead of aliphatic alcohol resulted in only a small amount of inhomogeneous precipitates; however, the initial materials (chloride and nitrate), the ratios of solvent/precursor (18–280 molar ratios), and the reaction time (2–48 h) were widely varied in the experiments. Surfactant-assisted precipitation method In the study on the surfactant-assisted precipitation techniques, the precipitation agents, the surfactants, their concentration, and the temperature of heat treatment were varied. The precipitations with sodium hydroxide or ammonia yield coarse and large particles ([100 nm) in aqueous solutions. Thus, the experiments concentrated on the application of carbonate and oxalate precipitators. The sizes of particles obtained by oxalate precipitators are larger (58 nm) and more polydisperse (24–140 nm, SEM) than that of carbonate precipitates (12–54 nm, SEM) (Table 3). The precipitator ratio has only a slight influence on the size and distribution above 1 molar ratio of precipitator/metal ion. Co-precipitation with carbonate without any surfactant yields inhomogeneous particles; cobalt ferrite nanoparticles (52 nm) with amorphous shape, NaCl plate-like aggregates (1.5–2 lm), and hematite octahedral crystals with average size of 320 nm (Figs. 3, 5). PDMS of 550 or 5,600 g mol-1 proved to be the most effective surfactant considering the size and size distribution of the particles synthesized with assistance of several surfactants (PDMS, Triton X-100, NaDS, NaDBS, and TDAB) in the coprecipitation series. The ionic surfactants hinder the aggregation less than PDMS. The particle size prepared with both precipitators (carbonate and oxalate) represents a minimum in the function of the PDMS concentration (Fig. 1). The smallest cobalt ferrite nanoparticles could be obtained by a carbonate precipitator in the presence of 5 w/w% PDMS. By PDMS, the volume of ferrite phase increases and the amorphous particles assume a cubic shape (Figs. 3, 6). The co-precipitation with sodium carbonate and PDMS results in amorphous basic carbonate salts at room temperature, from those cobalt ferrite and small amount of hematite can be evolved at around 600 °C.
123
J Nanopart Res (2012) 14:894
During the heat treatment, particles of 8–10 nm (SAXS) grow up to 40–45 nm (SAXS). By oxalic acid, CoC2O42H2O precipitates at room temperature, which increases the inhomogeneity of the product. The size and its dispersion change slightly by addition of a surfactant. Microemulsion technique In the microemulsion preparation, sodium carbonate serves as precipitating agent and sec. buthylalcohol as an oil phase. The preparation conditions were similar to the surfactant-assisted co-precipitation procedure to compare the methods and to study the effect of the microemulsion technique. The microemulsion technique resulted in a much more inhomogeneous product than the co-precipitation. This route of the microemulsion technique yielded the widest size distribution (20–2200 nm, SEM) owing to the several species. However, the cobalt ferrite amorphous particles are very fine—25–31 nm (SEM, the powders treated with PDMS or CTAB surfactant and heated at 600 °C). The ‘‘nanocontainers,’’ i.e., the emulgated droplets control the growth of amorphous ferrite particles rather than that of crystalline species. The components of the microemulsion-derived products are cobalt ferrite fine nanoparticles with amorphous shape, hematite octahedral crystals with average size of 1.55 lm, NaCl plate-like aggregates of 1.0–2.2 lm, and iron oxide rod-like aggregates of 1.0–2.0 lm. The application of the microemulsion requires any modifying agent. The surfactants (CTAB, NaDS, and PDMS) reduced significantly (on the average by about 200 nm) the particles size, especially in the presence of a large amount agent.
Conclusions In the present study, cobalt ferrite nanoparticles were synthesized by various liquid-phase methods, namely, by coprecipitation process, sol–gel route, and microemulsion technique combined with thermal decomposition. The cobalt ferrite nanoparticles can be used as components of polymer nanocomposites in medical diagnosis and targeted drug delivery. The effects of experimental parameters on the particle size, size distribution, morphology, and chemical composition have been studied. The preparation
J Nanopart Res (2012) 14:894
experiments were carried out by varying the parameters such as the anions of precursors (chloride and nitrate), the solvents (water, n-propanol, ethanol, and benzyl alcohol), the surfactants (polydimethylsiloxane, ethyl acetate, citric acid, cethyltrimethylammonium bromide, and sodium dodecil sulfate), the concentration of the surfactant (0–10 m/m%), the precipitating agents (sodium carbonate and oxalic acid), the temperatures of the hydrolysis (room temperature, 50, and 80 °C), and the thermal treatment (80–1,000 °C). A new, simple, and fast way of sol–gel method has been developed for preparation of cobalt ferrite nanoparticles. The smallest particles (40 nm Ø, SEM) and the best dispersion (30–60 nm) could be achieved by this sol–gel route starting from nitrate salts. The nitrate salts were reacted in the mixture of 1-propanol and ethyl acetate at 80 °C. The lower polarities of propanol and ethyl acetate support the decomposition of nitrate ions. The escape of nitrous gases increases the pH, which promotes the hydrolysis and condensation reactions of metal ions. In order to avoid the usual problem of the ferrite synthesis, i.e., the formation of iron oxide (hematite), the iron precursor must be slowly added to the excess of cobalt solution during mixing at 80 °C. The fine precipitate synthesized with slow addition of ferric nitrate solution in the presence of 1-propanol and ethyl acetate contains significantly less nitrate ions than that obtained by other surfactants in ethanol and requires the lowest temperature for its reactions. The use of chloride precursors in the sol–gel technique produces inhomogeneous products (cobalt-ferrite and ironoxide), a very low yield (5–6 %), and some larger sizes (58–80 nm, SEM). In the surfactant-assisted precipitation techniques, the basic precipitators (sodium hydroxide or ammonia) produce coarse and large particles ([100 nm) in aqueous solutions. The precipitates derived from oxalate precipitation are inhomogeneous and polydisperse in nature (24–140 nm). The application of the carbonate-precipitating agent yields a very fine ferrite powder (40–43 nm) in the presence of PDMS of 550 or 5,600 g mol-1 used in 5.0 w/w%. The particle size shows a minimum in the function of the PDMS concentration. The particles prepared by carbonate precipitator contain not only a cobalt ferrite phase but a small amount of sodium chloride and iron oxides, too. The co-precipitation carried out in a
Page 13 of 14
microemulsion generates polydisperse and polymorph particles; several shapes (spherical, octahedral, and rod-like), sizes (from 27 nm to 1–2 lm), and different chemical compositions (cobalt-ferrite, iron oxide, sodium chloride, etc.). The size characterization techniques (SAXS, SEM, and DLS) have been compared. The SAXS data are consistent with the sizes determined by SEM and differ from the DLS data. The size derived from DLS is two or three times larger. Dried powders are measured under vacuum in the SAXS and SEM experiments. The nanoparticles dispersed in a polar solution can be hydrated or solvated in the DLS technique and the hydration/solvation shells may result in the difference in the sizes. The effect of the polar solvent has been proved by DLS measurements in aqueous and alcoholic solutions. Acknowledgments This study has been supported I-04-009 EU in HASYLAB, DESY and OTKA NK 101704 funds. The European Union and the European Social Fund have provided financial support to the project under the grant agreement no. ´ MOP 4.2.1./B-09/KMR-2010-0003. TA
References Ahn Y, Choi EJ, Kim S, Ok HN (2001) Magnetization and Mo¨ssbauer study of cobalt ferrite particles from nanophase cobalt iron carbonate. Mater Lett 50:47–52 Bilecka I, Niederberger M (2010) Microwave chemistry for inorganic nanomaterials synthesis. Nanoscale 2:1358–1374 Bilecka I, Djerdj I, Niederberger M (2008) One-minute synthesis of crystalline binary and ternary metal oxide nanoparticles. Chem Commun 886–888 Boutry S, Laurent S, Elst VL, Muller RN (2006) Specific E-selectin targeting with a superparamagnetic MRI contrast agent. Contrast Med Mol Imaging 1:15–22 Brinker CJ, Sherrer GW (1990) Sol–gel science. Academic Press, New York Bulte JW (2006) Intracellular endosomal magnetic labeling of cells. Methods Mol Med 124:419–439 Burtea C, Laurent S, Roch A, Elst VL, Muller RN (2005) C-MALISA (cellular magnetic-linked immunosorbent assay), a new application of cellular ELISA for MRI. J Inorg Biochem 99:1135–1144 Chastellain M, Petri A, Gupta A, Rao KV, Hofmann H (2004) Superparamagnetic silica-iron oxide nanocomposites for application in hyperthermia. Adv Eng Mater 6:235–241 Chinnasamy CN, Senoue M, Jeyadevan B, Perales-Perez O, Shinoda K, Tohji K (2003) Synthesis, characterization and thermal analysis of polyimide–cobalt ferrite nanocomposites. J Col Interface Sci 263:80–83 Chourpa I, Douziech-Eyrolles L, Ngaboni-Okassa L, Fouquenet JF, Cohen-Jonathan S, Souce M, Marchais H, Dubois P (2005) Molecular composition of iron oxide
123
Page 14 of 14 nanoparticles, precursors for magnetic drug targeting, as characterized by confocal Raman microspectroscopy. Analyst 130:1395–1403 Ennas G, Musinu A, Piccaluga G, Zedda D, Gatteschi D, Sangregorio C, Stanger JL, Concas G, Spano G (1998) Characterization of iron oxide nanoparticles in an Fe2O3–SiO2 composite prepared by a sol–gel method. Chem Mater 10:495–502 Gupta AK, Gupta M (2005) Synthesis and surface engineering of iron oxide nanoparticles for biomedical applications. Biomaterials 26:3995–4021 Gyergyek S, Makovec D (2010) Influence of synthesis method on structural and magnetic properties of cobalt ferrite nanoparticles. J Nanoparticle Res 12:1263–1273 Han M, Vestal CR, Zhang ZJ (2004) Quantum couplings and magnetic properties of CoCrxFe2-xO4 (0 \ x \ 1) spinel ferrite nanoparticles synthesized with reverse micelle method. J Phys Chem B 108:583–587 Jain TK, Morales MA, Sahoo SK, Leslie-Pelecky DL, Labhasetwar V (2005) Iron oxide nanoparticles for sustained delivery of anticancer agent. Mol Pharm 2:194–205 Komarneni S, D’Arrigo MC, Leonelli C, Pellacani GC, Katsuki H (1998) Microwave-hydrothermal synthesis of nanophase ferrites. J Am Ceram Soc 81:3041–3043 Laurent S, Forge D, Port M, Robic C, Elst VL, Muller RN (2008) Magnetic iron oxide nanoparticles: synthesis, stabilization, vectorization, physicochemical characterizations, and biological applications. Chem Rev 108:2064–2110 Lee JG, Park JY, Kim CS (1998) Growth of ultra-fine cobalt ferrite particles by a sol–gel method and their magnetic properties. J Mater Sci 33:3965–3968 Li X-H, Xu CL (2010) Synthesis and magnetic properties of nearly monodisperse CoFe2O4 nanoparticles through a simple hydrothermal condition. Nanoscale Res Lett 5:1039–1044 Liu Y, Zhang Y (2009) Dependence of magnetic properties on crystallite size of CoFe2O4 nanoparticles synthesised by auto-combustion method. J Exp Nanosci 4:159–168 Liu Z, Li X, Leng Y, Lian J, Liu S, Xiu Z, Huo D, Li J, Sun X (2008) Homogeneous precipitation synthesis and magnetic poperties of cobalt ferrite nanoparticles. J Nanomat 2008:4, Article ID 921654 Meron T, Rosenberg Y, Lareah Y, Markovich G (2005) Synthesis and assembly of high-quality cobalt ferrite nanocrystals prepared by a modified sol–gel technique. J Magn Magn Mater 292:11–16
123
J Nanopart Res (2012) 14:894 Moumen N, Pileni MP (1996) New syntheses of cobalt ferrite particles in the range 2–5 nm: comparison of the magnetic properties of the nanosized particles in dispersed fluid or in powder form. Chem Mater 8:1128–1134 Murray CB, Agan CRC (2000) Synthesis and characterization of monodisperse nanocrystals and close-packed nanocrystal assemblies. Ann Rev Mat Sci 30:545–610 Neveu S, Bec A, Robineau M, Talbol D (2002) Size-selective chemical synthesis of tartrate stabilized cobalt ferrite ionic magnetic fluid. J Colloid Interface Sci 255:293–298 Olsson RT, Salazar-Alvarez G (2005) Controlled synthesis of near-stoichiometric cobalt ferrite nanoparticles. Chem Mater 17:5109–5118 Paike V, Niphadkar P, Bokade V, Joshi P (2007) Synthesis of spinel CoFe2O4 via the co-precipitation method using tetraalkyl ammonium hydroxides as precipitating agents. J Am Ceram Soc 90:3009–3012 Pillai V, Shah DO (1996) Synthesis of high-coercivity cobalt ferrite particles using water-in-oil microemulsions. J Magn Magn Mat 163:243–248 Pinna N, Niederberger N (2008) Surfactant-free nonaqueous synthesis of metal oxide nanostructures. Angew Chem Int Ed 47:5292–5304 Pramanik NC, Fujii T, Nakanishi M, Takada J (2004) Effect of Co2? ion on the magnetic properties of sol-gel cobalt ferrite thin film. Mater Chem 14:3328–3332 Rebolledo AF, Fuertes AB, Gonzalez-Carreno T, Sevilla M, Balde-Solis T, Tartaj P (2008) Signatures of clustering in superparamagnetic colloidal nanocomposites of an inorganic and hybrid nature. Small 4:254–261 Silva JB, Diniz CF, Viana AP, Mohallem ND (2005) Characterization of porous nanocomposites formed by cobalt ferrites dispersed in sol–gel silica matrix. J Sol Gel Sci Technol 35:115–122 Sun S, Zeng H (2004) Monodisperse MFe2O4 (M=Fe, Co, Mn) nanoparticles. J Am Chem Soc 126:273–279 Tobler DJ, Shaw S, Benning LG (2009) Quantification of initial steps of nucleation and growth of silica nanoparticles: an in situ SAXS and DLS study. Geochim Cosmochim Acta 73:5377–5393 Zhang ZJ, Wang ZL, Chakoumakos BC, Yin JS (1998) Temperature dependence of cation distribution and oxidation state in magnetic Mn–Fe ferrite nanocrystals. J Am Chem Soc 120:1800–1804
