- Email: [email protected]
Effect of divalent metal hydroxide solubility product on the size of ferrite nanoparticles
Materials Letters 61 (2007) 4545 – 4548 www.elsevier.com/locate/matlet
Effect of divalent metal hydroxide solubility product on the size of ferrite nanoparticles G. Gnanaprakash, John Philip ⁎, Baldev Raj Metallurgy and Materials Group, Indira Gandhi Centre for Atomic Research, Kalpakkam 603 102, Tamilnadu, India Received 27 September 2006; accepted 21 February 2007 Available online 1 March 2007
Abstract We study the effect of divalent metal hydroxide solubility product on the size and magnetic properties of nanoparticles formed during coprecipitation. We synthesized ferrite nanoparticles by varying the solubility product from 10− 13 to 10− 17 by using different divalent cations of Mn, Co, Fe and Zn, where the average particle size decreased from 29.1 to 8.9 nm. The Mn, Co and Fe ferrites were magnetic in nature with saturation magnetization of 44.6, 47.38 and 56.19 emu/g respectively, whereas the Zn ferrite was paramagnetic. The increase in particle size observed with increasing solubility product of divalent metal hydroxide is in agreement with the nucleation theory. © 2007 Elsevier B.V. All rights reserved. Keywords: Co-precipitation; Solubility product; Ferrite nanoparticles; Particle growth
1. Introduction Due to several technological applications of magnetic nanoparticles, researchers have been working on new and cost effective synthesis methods [1–7]. The cubic spinel structured MO·Fe2O3 (M = Fe, Mn, Co, Ni, Zn, etc.) has been among the most frequently used systems to understand fundamental aspects of nanomagnetism [8]. It also has several applications in various fields such as ferrofluids [9], recording media [10], biomedical [8], electronic devices [8] and fundamental studies [11]. The synthesis techniques used, in recent years, to produce nanoparticles are mechano-chemical [5], laser pyrolysis [7], co-precipitation [6], sol–gel [2], solvo-thermal [3], hydrothermal [4] and very recently bacterial aerobic synthesis techniques [1]. Each synthesis procedure has its own merits and de-merits and is useful to prepare certain nanoparticles with specific properties. Among the above preparation techniques, one of the most convenient and versatile techniques is coprecipitation of M 2+ and Fe3+ by an alkali in an aqueous solution [6] or in reverse micelle [12] or using liquid foam as template [13]. Co-precipitation reaction occurs at low temper⁎ Corresponding author. E-mail address: [email protected] (J. Philip). 0167-577X/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.matlet.2007.02.048
atures, has biocompatible yield with low impurity levels and is cost effective. Size of particles can be tuned by optimizing the reaction parameters such as temperature, pH and ionic strength imposed by non-complexing salt [6,14]. Precipitation reactions consist of a nucleation step followed by particle growth stages and these two simultaneous processes govern the product morphology. Due to the difficulties in isolating nucleation and growth processes, the fundamental mechanism of precipitation reactions is still not fully understood. When the product contains more than two elements, the process becomes more complex, as multiple elements must be precipitated simultaneously. In co-precipitation reactions, it is necessary to understand the fundamental aspects of the process and the effects of physical and chemical properties of the individual elements used in the system to manipulate the growth of nanoparticles to the desired size and shape. In the ferrite system, the critical radius (r⁎) of the nucleated particle size and the structure depends on the individual properties of both the metal ions. Moreover, the pathways from soluble Fe3+ and M2+ to thermodynamically stable metal ferrite involve intermediate products which have a vital role in the growth process. Here, we investigate the effect of divalent metal hydroxide solubility product on the size and magnetic properties of nanoparticles formed during co-precipitation.
4546
G. Gnanaprakash et al. / Materials Letters 61 (2007) 4545–4548
2. Experimental 2.1. Preparation of magnetic nanoparticles The synthesis of ferrite nanoparticles was carried out by precipitating freshly prepared (GR grade) M(II) and Fe(III) salt solutions of stoichiometric proportions in alkaline medium at constant stirring, where M = Fe, Mn, Co, Zn. On vigorous stirring, the solution pH was rapidly increased to 12, by the addition of 6 N NaOH at 90 °C. At a constant rate, the addition of alkali was finished in 10 s for all the samples and they were left for 60 min for digestion. Then, the pH of the dispersion was adjusted to ∼6–7, with diluted HCl. As protonation was done, the particles were separated from the dispersion. The top water layer with salts was discarded. The obtained particles have been washed with triply distilled water at 60 °C on vigorous stirring, until the washed water was free of ionic impurities (checked for chloride ions in top water layer with AgNO3 solution). Later, the water-washed particles were treated with acetone and particles were dried at 35 °C for 48 h in inert atmosphere. Acetone washes were carried out to avoid the particle drying process at higher temperatures and to eliminate water content. Especially, in the case of magnetite (FeO·Fe2O3), Fe2+ is more sensitive to aerial oxidation and hence higher temperature enhances the oxidation process. The details of the preparation procedure are discussed elsewhere [15]. 2.2. Characterization of magnetic nanoparticles 2.2.1. XRD studies A MAC Science MXP18 X-ray diffractometer was used for crystal structure and average particle size analysis. 2θ values were taken from 11 to 80° using Cu Kα radiation with step size of 0.04°. The XRD patterns were verified by comparison with the JCPDS data. The broadening of the peak at half height was related to the average diameter (d) of the particle according to Scherrer's formula, i.e. d = 0.9λ / Δ cosθ where λ is the X-ray wavelength, Δ is the line broadening measured at half-height
Fig. 2. The TEM image of the Fe3O4 sample. The average size of the particles is about 9 nm.
and θ is the Bragg angle of the particles. The average particle size is obtained from the most intense peak, corresponding to (311) reflection by using the Debye–Scherrer formula. 2.2.2. Morphological evaluations The TEM instrument used was a JEOL 2011 with an acceleration voltage of 200 kV. The average particle size obtained from the Debye–Scherrer formula was in good agreement with the TEM results. 2.3. Magnetic properties A vibrating sample magnetometer (EG and G Princeton, Model: 4500) was used for the magnetization measurements. These measurements were taken from 0 to ± 7 kOe field at room temperatures. 3. Results and discussion 3.1. Crystal structure and particle morphology On increasing pH, Fe3+ ions precipitate as ferrihydrite at pH of 3, whereas divalent metal ions precipitate at higher pH values. The minimum pH values required to precipitate divalent metal hydroxides, when a single cation is used, are 7, ∼ 8, 8 and 9 for Fe2+, Zn2+, Co2+ and Mn2+, respectively [16]. In the ferrite precipitation process, while increasing the pH of the salt solution, ferrite nanoparticles are nucleated by the simultaneous reaction of hydrolysis and dehydration. The overall chemical reaction can be written as, M2þ þ 2Fe3þ þ 8OH− →MO:Fe2 O3 þ 4H2 O
Fig. 1. The XRD patterns of the samples prepared with varying divalent metal ions of Mn, Fe, Co and Zn.
Fig. 1 shows the XRD patterns of the samples prepared with varying divalent metal ions of Mn, Fe, Co and Zn. MnFe2O4, Fe3O4 and CoFe2O4 have an inverse spinel structure, where ZnFe2O4 has a normal spinel structure. Size, covalent bonding effects and crystal field stabilization energies of M2+ seem to influence the site preference of ions. In inverse spinel structure, divalent metal ions occupy octahedral
G. Gnanaprakash et al. / Materials Letters 61 (2007) 4545–4548
4547
Table 1 M
M(OH)2, KSP
M2+ ionic radii (Å)
Av. size (nm)
Lattice const. (nm)
Ms a (emu/g)
Hc b (kOe)
Hr c (emu/g)
Kbulk d (erg/cm3) × 105
SPD e (nm)
Mn Co Fe Zn
2 × 10− 13 1.09 × 10− 15 4.9 × 10− 17 7.71 × 10− 17
0.91 0.82 0.83 0.83
29.1 14.5 9.9 8.9
0.85 0.837 0.838 0.844
44.6 47.38 56.19 –
0.083 0.782 0.038 –
7.4 19.2 2.8 –
0.3 20 1.1 –
40.2 9.9 26.1 –
a b c d e
Ms: saturation magnetization. Hc: coercivity. Hr: remanence. Kbulk: anisotropy constant of the bulk sample. SPD: minimum diameter required to exhibit superparamagnetism.
sites. The calculated lattice parameter values from the XRD pattern are 0.850, 0.838, 0.837 and 0.844 nm for MnFe2O4, Fe3O4, CoFe2O4 and ZnFe2O4, respectively. These values are in agreement with earlier reports [17,18]. The lattice parameter is dependent on the ionic radii of the divalent metal ion. Since the ionic radii of Mn2+, Fe2+ and Co2+ are in decreasing order (0.91, 0.83 and 0.82 Å respectively), the lattice parameter is also expected to follow the same order. Though, the ionic radii of Zn2+ (0.83 Å) are almost the same as those of Co2+ and Fe2+ ions, the Zn2+ ions prefer to be in the tetrahedral sites leading to normal spinel structure. Usually in a close packing structure, the radius of the tetrahedral site is about 0.866 times smaller than the octahedral site [19]. The observed larger lattice parameter (0.844 nm) in ZnFe2O4 seems to be due to the strong inter-atomic interactions compared to cobalt or iron ferrite. The calculated average size of the MnFe2O4, Fe3O4, CoFe2O4 and ZnFe2O4 particles is 29.1, 9.9, 14.5 and 8.9 nm respectively. Though all these samples are synthesized under identical conditions (reaction temperature, pH, solvent, etc.), the large variation at the average size of the particles shows that the divalent metal ions have a strong influence on the crystallization process. Fig. 2 shows the TEM image of the Fe3O4 sample. In the case of magnetite formation, earlier studies revealed that on increasing pH, an intermediate complex of Fe2+–ferrihydrite is formed first. A further increase in pH leads to the conversion of this complex to spinel structured magnetite due to condensation [20]. The magnetite precipitation process is proceeded by two simultaneous competing pathways, i.e. solid state reaction and dissolution–crystallization. In the former, the process is topotactic and spinel ordering occurs with dehydration where electron transfer between Fe2+ and Fe3+ plays a fundamental role. In the latter case, dissolution of Fe2+–Fe3+complexes from the surface, followed by crystallization and this process, causes considerable particle growth [14]. However, with the other divalent
Fig. 3. The variation of particle size with ln(KSP).
cations, inter-valence transfers are generally negligible and spinel ferrite forms only by the dissolution–crystallization process [20]. 3.2. Solubility product Solubility products are usually calculated from the Gibb's free energies of the substances as solids and those of the aqueous ions at their standard states of unit molality. For a reaction Mm Xn ðSÞ X mMnþ ðaqÞ þ nXm− ðaqÞ DG0 ¼ mDG0f ðmnþ ; aqÞ þ nDG0f ðXm− ; aqÞ−DG0f ðMm Xn ; sÞ Where MnXn is the slightly soluble substance, Mn+ and Xm− are the two ions produced in the solution by the dissociation of MnXn. Thus the solubility product constant, KSP, is calculated by using the equation lnðKSP Þ ¼
−DG0 RT
In precipitation reactions, as nucleation begins in a supersaturation solution, depending on various parameters, the equilibrium critical radius (r⁎) is given by [21], r⁎ ¼
S¼
2vr kTlnðSÞ
aa ab Ksp
ð1Þ
ð2Þ
σ is the surface energy, v is the molecular volume of the precipitated embryo, k is the Boltzmann constant, T is the temperature and S is the
Fig. 4. Magnetization curves of different ferrite particles.
4548
G. Gnanaprakash et al. / Materials Letters 61 (2007) 4545–4548
supersaturation, αa is the activities of Fe2+ and Fe3+ ions and αb is the activity of OH− ions. KSP is solubility product constant. If the nucleated particle radius is greater than the critical radius, then the particle will continue to grow and the particle radius, which is smaller than the critical radius, will dissolve. In the case of MFe2O4, first M2+ and Fe3+ ions simultaneously undergo the hydrolysis process leading to the formation of sparingly soluble hydroxides, which later condense to thermodynamically stable oxides. Therefore, it is expected that the KSP values of both elements can influence the precipitation process. Since the ferrihydrite (KSP ∼ 10− 39) formation is common in all ferrite ions, the size variation can be attributed to KSP values of M(OH)2. Later, the particle growth process (by Oswald ripening) also depends on various solubility parameters such as pH, temperature, KSP of respective ions (Table 1) and polarity of solvent. Fig. 3 shows the plot of ln(KSP) of M(OH)2 as a function of the average particle size. As expected by theory, the experimental values of ln(KSP) and the average particle size showed a linear dependence. 3.3. Magnetic properties Fig. 4 shows the magnetization curves of the different ferrite particles obtained by co-precipitation. Saturation magnetization (Ms), remanence (Hr) and coercivity (Hc) values obtained from VSM measurements are given in Table 1. Mn and Co ferrites show a ferromagnetic nature with a hysteresis loop, whereas the magnetite sample exhibits superparamagnetic nature with negligible remanence and coercivity. The Zn ferrite shows a paramagnetic nature. However, the bulk MnFe2O4, Fe3O4 and CoFe2O4 materials show ferrimagnetic nature whereas ZnFe2O4 is paramagnetic at room temperatures [19]. The minimum particle volume (Vp) required for exhibiting superparamagnetic behavior is 25kBT / K, where kB is Boltzmann constant, K is anisotropic constant and T is temperature (300 K) [17]. The reported anisotropy constants (K) for MnFe2O4, Fe3O4 and CoFe2O4 are 0.3 × 105, 1.1 × 105 and 20 × 105 erg/cm3, respectively [17]. If particles are assumed as spherical, the minimum diameter required for the superparamagnetic behavior (SPD) is 40.2, 26.1 and 9.9 nm for MnFe2O4, Fe3O4 and CoFe2O4 nanoparticles, respectively. For MnFe2O4, considerable coercivity is shown and it may be attributed to an increase in K value in the nano-regime and the polydispersity of the sample. The observed value of K for MnFe2O4 particles with an average size of 7.4 nm is 0.056 J/cm3, which is 20 times more than that of the bulk material [22]. This enhancement is attributed to the possibility of strong interactions between the magnetic core and the shell of the nanoparticles. Among the ferrites, CoFe2O4 has the largest K value due to the strong spin-orbital coupling at Co2+ lattice sites. Therefore, the SPD value of CoFe2O4 is much lower than that of the other ferrites. Hence, though CoFe2O4 particles are prepared at identical conditions in co-precipitation, these particles are not superparamagnetic. In brief, our experimental results suggest that by varying the divalent metal ion (M2+), the magnetic nature of MO·Fe2O3 can be tuned.
4. Conclusions We have studied the effect of divalent metal ion on the size and magnetic behavior of precipitated nanoparticles. When the solubility product is varied from 10− 13 to 10− 17, by using
different divalent cations of Mn, Co, Fe and Zn, the average particle size decreased from 29.1 to 8.9 nm. The Mn, Co and Fe ferrites were magnetic in nature with saturation magnetization of 44.6, 47.38, 56.19 emu/g respectively, whereas the Zn ferrite was paramagnetic. The experimental values of ln(KSP) and the average particle size showed a linear dependence, in agreement with nucleation theory. Acknowledgements We thank Dr. P.R. Vasudeva Rao, Mr. P. Kalyanasundaram and Dr. T. Jayakumar for support and encouragements and Mr. Mahadevan for XRD data. JP would like to thank Prof. Bernard Binks of University of Hull, U.K. for extending the TEM facilities. References [1] A. Bharde, A. Wani, Y. Shouche, P.A. Joy, B.L.V. Prasad, M. Sastry, J. Am. Chem. Soc. 127 (2005) 9326. [2] J.W. Long, M.S. Logan, C.P. Rhodes, E.E. Carpenter, R.M. Stroud, D.R. Rolison, J. Am. Chem. Soc. 126 (2004) 16879. [3] S. Sun, H. Zeng, D.B. Robinson, S. Raoux, P.M. Rice, S.X. Wang, G. Li, J. Am. Chem. Soc. 126 (2004) 273. [4] J. Wang, Q. Chen, C. Zeng, B. Hou, Adv. Mater. 16 (2004) 137. [5] C.-R. Lin, Y.-M. Chu, S.-C. Wang, Mater. Lett. 60 (2006) 447. [6] Y. Qu, H. Yang, N. Yang, Y. Fan, H. Zhu, G. Zou, Mater. Lett. 60 (2006) 3548. [7] S. Veintemillas-Verdaguer, M.P. Morales, C.J. Serna, Mater. Lett. 35 (1998) 227. [8] K.M. Krishnan, A.B. Pakhomov, Y. Bao, P. Blomqvist, Y. Chun, M. Gonzales, K. Griffin, X. Ji, B.K. Roberts, J. Mater. Sci. 41 (2006) 793. [9] K. Raj, B. Moskowitz, R. Casciari, J. Magn. Magn. Mater. 149 (1995) 174. [10] D.E. Speliotis, J. Magn. Magn. Mater. 193 (1999) 29. [11] J. Philip, G.G. Prakash, T. Jayakumar, P. Kalyanasudaram, B. Raj, Phys. Rev. Lett. 89 (2002) 268301; Macromolecules 36 (2003) 9230. [12] J.A.L. Perez, M.A.L. Quintela, J. Mira, J. Rivas, S.W. Charles, J. Phys. Chem., B 101 (1997) 8045. [13] T. Bala, C.R. Shankar, M. Baidakova, V. Osipov, T. Enoki, P.A. Joy, B.L.V. Prasad, M. Sastry, Langmuir 21 (2005) 10638. [14] L. Vayssieres, C. Chanaec, E. Tronc, J.P. Jolivet, J. Colloid Interface Sci. 205 (1998) 205. [15] G. Gnanaprakash, S. Ayyappan, T. Jayakumar, J. Philip, B. Raj, Nanotechnology 17 (2006) 5851. [16] Z.X. Tang, C.M. Sorensen, K.J. Klabunde, G.C. Hadjipanayis, J. Colloid Interface Sci. 146 (1991) 38. [17] B.D. Cullity, Introduction to Magnetic Materials, Addison-Wesley Publishing Company, 1972. [18] P.M.G. Nambissan, C. Upadhyaya, H.C. Verma, J. Appl. Phys. 93 (2003) 6320. [19] A.R. West, Solid State Chemistry and Its Applications, John Wiley & Sons Ltd., 1984. [20] J.-P. Jolivet, E. Tronc, C. Chaneac, C.R. Geosci. 338 (2006) 488. [21] J.W. Mullin, Crystallisation, Butterworth & Co (Publishers) Ltd., London, 1972. [22] A.J. Rondinone, C. Liu, Z.J. Zhang, J. Phys. Chem., B 105 (2001) 7967.
Effect of divalent metal hydroxide solubility product on the size of ferrite nanoparticles G. Gnanaprakash, John Philip ⁎, Baldev Raj Metallurgy and Materials Group, Indira Gandhi Centre for Atomic Research, Kalpakkam 603 102, Tamilnadu, India Received 27 September 2006; accepted 21 February 2007 Available online 1 March 2007
Abstract We study the effect of divalent metal hydroxide solubility product on the size and magnetic properties of nanoparticles formed during coprecipitation. We synthesized ferrite nanoparticles by varying the solubility product from 10− 13 to 10− 17 by using different divalent cations of Mn, Co, Fe and Zn, where the average particle size decreased from 29.1 to 8.9 nm. The Mn, Co and Fe ferrites were magnetic in nature with saturation magnetization of 44.6, 47.38 and 56.19 emu/g respectively, whereas the Zn ferrite was paramagnetic. The increase in particle size observed with increasing solubility product of divalent metal hydroxide is in agreement with the nucleation theory. © 2007 Elsevier B.V. All rights reserved. Keywords: Co-precipitation; Solubility product; Ferrite nanoparticles; Particle growth
1. Introduction Due to several technological applications of magnetic nanoparticles, researchers have been working on new and cost effective synthesis methods [1–7]. The cubic spinel structured MO·Fe2O3 (M = Fe, Mn, Co, Ni, Zn, etc.) has been among the most frequently used systems to understand fundamental aspects of nanomagnetism [8]. It also has several applications in various fields such as ferrofluids [9], recording media [10], biomedical [8], electronic devices [8] and fundamental studies [11]. The synthesis techniques used, in recent years, to produce nanoparticles are mechano-chemical [5], laser pyrolysis [7], co-precipitation [6], sol–gel [2], solvo-thermal [3], hydrothermal [4] and very recently bacterial aerobic synthesis techniques [1]. Each synthesis procedure has its own merits and de-merits and is useful to prepare certain nanoparticles with specific properties. Among the above preparation techniques, one of the most convenient and versatile techniques is coprecipitation of M 2+ and Fe3+ by an alkali in an aqueous solution [6] or in reverse micelle [12] or using liquid foam as template [13]. Co-precipitation reaction occurs at low temper⁎ Corresponding author. E-mail address: [email protected] (J. Philip). 0167-577X/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.matlet.2007.02.048
atures, has biocompatible yield with low impurity levels and is cost effective. Size of particles can be tuned by optimizing the reaction parameters such as temperature, pH and ionic strength imposed by non-complexing salt [6,14]. Precipitation reactions consist of a nucleation step followed by particle growth stages and these two simultaneous processes govern the product morphology. Due to the difficulties in isolating nucleation and growth processes, the fundamental mechanism of precipitation reactions is still not fully understood. When the product contains more than two elements, the process becomes more complex, as multiple elements must be precipitated simultaneously. In co-precipitation reactions, it is necessary to understand the fundamental aspects of the process and the effects of physical and chemical properties of the individual elements used in the system to manipulate the growth of nanoparticles to the desired size and shape. In the ferrite system, the critical radius (r⁎) of the nucleated particle size and the structure depends on the individual properties of both the metal ions. Moreover, the pathways from soluble Fe3+ and M2+ to thermodynamically stable metal ferrite involve intermediate products which have a vital role in the growth process. Here, we investigate the effect of divalent metal hydroxide solubility product on the size and magnetic properties of nanoparticles formed during co-precipitation.
4546
G. Gnanaprakash et al. / Materials Letters 61 (2007) 4545–4548
2. Experimental 2.1. Preparation of magnetic nanoparticles The synthesis of ferrite nanoparticles was carried out by precipitating freshly prepared (GR grade) M(II) and Fe(III) salt solutions of stoichiometric proportions in alkaline medium at constant stirring, where M = Fe, Mn, Co, Zn. On vigorous stirring, the solution pH was rapidly increased to 12, by the addition of 6 N NaOH at 90 °C. At a constant rate, the addition of alkali was finished in 10 s for all the samples and they were left for 60 min for digestion. Then, the pH of the dispersion was adjusted to ∼6–7, with diluted HCl. As protonation was done, the particles were separated from the dispersion. The top water layer with salts was discarded. The obtained particles have been washed with triply distilled water at 60 °C on vigorous stirring, until the washed water was free of ionic impurities (checked for chloride ions in top water layer with AgNO3 solution). Later, the water-washed particles were treated with acetone and particles were dried at 35 °C for 48 h in inert atmosphere. Acetone washes were carried out to avoid the particle drying process at higher temperatures and to eliminate water content. Especially, in the case of magnetite (FeO·Fe2O3), Fe2+ is more sensitive to aerial oxidation and hence higher temperature enhances the oxidation process. The details of the preparation procedure are discussed elsewhere [15]. 2.2. Characterization of magnetic nanoparticles 2.2.1. XRD studies A MAC Science MXP18 X-ray diffractometer was used for crystal structure and average particle size analysis. 2θ values were taken from 11 to 80° using Cu Kα radiation with step size of 0.04°. The XRD patterns were verified by comparison with the JCPDS data. The broadening of the peak at half height was related to the average diameter (d) of the particle according to Scherrer's formula, i.e. d = 0.9λ / Δ cosθ where λ is the X-ray wavelength, Δ is the line broadening measured at half-height
Fig. 2. The TEM image of the Fe3O4 sample. The average size of the particles is about 9 nm.
and θ is the Bragg angle of the particles. The average particle size is obtained from the most intense peak, corresponding to (311) reflection by using the Debye–Scherrer formula. 2.2.2. Morphological evaluations The TEM instrument used was a JEOL 2011 with an acceleration voltage of 200 kV. The average particle size obtained from the Debye–Scherrer formula was in good agreement with the TEM results. 2.3. Magnetic properties A vibrating sample magnetometer (EG and G Princeton, Model: 4500) was used for the magnetization measurements. These measurements were taken from 0 to ± 7 kOe field at room temperatures. 3. Results and discussion 3.1. Crystal structure and particle morphology On increasing pH, Fe3+ ions precipitate as ferrihydrite at pH of 3, whereas divalent metal ions precipitate at higher pH values. The minimum pH values required to precipitate divalent metal hydroxides, when a single cation is used, are 7, ∼ 8, 8 and 9 for Fe2+, Zn2+, Co2+ and Mn2+, respectively [16]. In the ferrite precipitation process, while increasing the pH of the salt solution, ferrite nanoparticles are nucleated by the simultaneous reaction of hydrolysis and dehydration. The overall chemical reaction can be written as, M2þ þ 2Fe3þ þ 8OH− →MO:Fe2 O3 þ 4H2 O
Fig. 1. The XRD patterns of the samples prepared with varying divalent metal ions of Mn, Fe, Co and Zn.
Fig. 1 shows the XRD patterns of the samples prepared with varying divalent metal ions of Mn, Fe, Co and Zn. MnFe2O4, Fe3O4 and CoFe2O4 have an inverse spinel structure, where ZnFe2O4 has a normal spinel structure. Size, covalent bonding effects and crystal field stabilization energies of M2+ seem to influence the site preference of ions. In inverse spinel structure, divalent metal ions occupy octahedral
G. Gnanaprakash et al. / Materials Letters 61 (2007) 4545–4548
4547
Table 1 M
M(OH)2, KSP
M2+ ionic radii (Å)
Av. size (nm)
Lattice const. (nm)
Ms a (emu/g)
Hc b (kOe)
Hr c (emu/g)
Kbulk d (erg/cm3) × 105
SPD e (nm)
Mn Co Fe Zn
2 × 10− 13 1.09 × 10− 15 4.9 × 10− 17 7.71 × 10− 17
0.91 0.82 0.83 0.83
29.1 14.5 9.9 8.9
0.85 0.837 0.838 0.844
44.6 47.38 56.19 –
0.083 0.782 0.038 –
7.4 19.2 2.8 –
0.3 20 1.1 –
40.2 9.9 26.1 –
a b c d e
Ms: saturation magnetization. Hc: coercivity. Hr: remanence. Kbulk: anisotropy constant of the bulk sample. SPD: minimum diameter required to exhibit superparamagnetism.
sites. The calculated lattice parameter values from the XRD pattern are 0.850, 0.838, 0.837 and 0.844 nm for MnFe2O4, Fe3O4, CoFe2O4 and ZnFe2O4, respectively. These values are in agreement with earlier reports [17,18]. The lattice parameter is dependent on the ionic radii of the divalent metal ion. Since the ionic radii of Mn2+, Fe2+ and Co2+ are in decreasing order (0.91, 0.83 and 0.82 Å respectively), the lattice parameter is also expected to follow the same order. Though, the ionic radii of Zn2+ (0.83 Å) are almost the same as those of Co2+ and Fe2+ ions, the Zn2+ ions prefer to be in the tetrahedral sites leading to normal spinel structure. Usually in a close packing structure, the radius of the tetrahedral site is about 0.866 times smaller than the octahedral site [19]. The observed larger lattice parameter (0.844 nm) in ZnFe2O4 seems to be due to the strong inter-atomic interactions compared to cobalt or iron ferrite. The calculated average size of the MnFe2O4, Fe3O4, CoFe2O4 and ZnFe2O4 particles is 29.1, 9.9, 14.5 and 8.9 nm respectively. Though all these samples are synthesized under identical conditions (reaction temperature, pH, solvent, etc.), the large variation at the average size of the particles shows that the divalent metal ions have a strong influence on the crystallization process. Fig. 2 shows the TEM image of the Fe3O4 sample. In the case of magnetite formation, earlier studies revealed that on increasing pH, an intermediate complex of Fe2+–ferrihydrite is formed first. A further increase in pH leads to the conversion of this complex to spinel structured magnetite due to condensation [20]. The magnetite precipitation process is proceeded by two simultaneous competing pathways, i.e. solid state reaction and dissolution–crystallization. In the former, the process is topotactic and spinel ordering occurs with dehydration where electron transfer between Fe2+ and Fe3+ plays a fundamental role. In the latter case, dissolution of Fe2+–Fe3+complexes from the surface, followed by crystallization and this process, causes considerable particle growth [14]. However, with the other divalent
Fig. 3. The variation of particle size with ln(KSP).
cations, inter-valence transfers are generally negligible and spinel ferrite forms only by the dissolution–crystallization process [20]. 3.2. Solubility product Solubility products are usually calculated from the Gibb's free energies of the substances as solids and those of the aqueous ions at their standard states of unit molality. For a reaction Mm Xn ðSÞ X mMnþ ðaqÞ þ nXm− ðaqÞ DG0 ¼ mDG0f ðmnþ ; aqÞ þ nDG0f ðXm− ; aqÞ−DG0f ðMm Xn ; sÞ Where MnXn is the slightly soluble substance, Mn+ and Xm− are the two ions produced in the solution by the dissociation of MnXn. Thus the solubility product constant, KSP, is calculated by using the equation lnðKSP Þ ¼
−DG0 RT
In precipitation reactions, as nucleation begins in a supersaturation solution, depending on various parameters, the equilibrium critical radius (r⁎) is given by [21], r⁎ ¼
S¼
2vr kTlnðSÞ
aa ab Ksp
ð1Þ
ð2Þ
σ is the surface energy, v is the molecular volume of the precipitated embryo, k is the Boltzmann constant, T is the temperature and S is the
Fig. 4. Magnetization curves of different ferrite particles.
4548
G. Gnanaprakash et al. / Materials Letters 61 (2007) 4545–4548
supersaturation, αa is the activities of Fe2+ and Fe3+ ions and αb is the activity of OH− ions. KSP is solubility product constant. If the nucleated particle radius is greater than the critical radius, then the particle will continue to grow and the particle radius, which is smaller than the critical radius, will dissolve. In the case of MFe2O4, first M2+ and Fe3+ ions simultaneously undergo the hydrolysis process leading to the formation of sparingly soluble hydroxides, which later condense to thermodynamically stable oxides. Therefore, it is expected that the KSP values of both elements can influence the precipitation process. Since the ferrihydrite (KSP ∼ 10− 39) formation is common in all ferrite ions, the size variation can be attributed to KSP values of M(OH)2. Later, the particle growth process (by Oswald ripening) also depends on various solubility parameters such as pH, temperature, KSP of respective ions (Table 1) and polarity of solvent. Fig. 3 shows the plot of ln(KSP) of M(OH)2 as a function of the average particle size. As expected by theory, the experimental values of ln(KSP) and the average particle size showed a linear dependence. 3.3. Magnetic properties Fig. 4 shows the magnetization curves of the different ferrite particles obtained by co-precipitation. Saturation magnetization (Ms), remanence (Hr) and coercivity (Hc) values obtained from VSM measurements are given in Table 1. Mn and Co ferrites show a ferromagnetic nature with a hysteresis loop, whereas the magnetite sample exhibits superparamagnetic nature with negligible remanence and coercivity. The Zn ferrite shows a paramagnetic nature. However, the bulk MnFe2O4, Fe3O4 and CoFe2O4 materials show ferrimagnetic nature whereas ZnFe2O4 is paramagnetic at room temperatures [19]. The minimum particle volume (Vp) required for exhibiting superparamagnetic behavior is 25kBT / K, where kB is Boltzmann constant, K is anisotropic constant and T is temperature (300 K) [17]. The reported anisotropy constants (K) for MnFe2O4, Fe3O4 and CoFe2O4 are 0.3 × 105, 1.1 × 105 and 20 × 105 erg/cm3, respectively [17]. If particles are assumed as spherical, the minimum diameter required for the superparamagnetic behavior (SPD) is 40.2, 26.1 and 9.9 nm for MnFe2O4, Fe3O4 and CoFe2O4 nanoparticles, respectively. For MnFe2O4, considerable coercivity is shown and it may be attributed to an increase in K value in the nano-regime and the polydispersity of the sample. The observed value of K for MnFe2O4 particles with an average size of 7.4 nm is 0.056 J/cm3, which is 20 times more than that of the bulk material [22]. This enhancement is attributed to the possibility of strong interactions between the magnetic core and the shell of the nanoparticles. Among the ferrites, CoFe2O4 has the largest K value due to the strong spin-orbital coupling at Co2+ lattice sites. Therefore, the SPD value of CoFe2O4 is much lower than that of the other ferrites. Hence, though CoFe2O4 particles are prepared at identical conditions in co-precipitation, these particles are not superparamagnetic. In brief, our experimental results suggest that by varying the divalent metal ion (M2+), the magnetic nature of MO·Fe2O3 can be tuned.
4. Conclusions We have studied the effect of divalent metal ion on the size and magnetic behavior of precipitated nanoparticles. When the solubility product is varied from 10− 13 to 10− 17, by using
different divalent cations of Mn, Co, Fe and Zn, the average particle size decreased from 29.1 to 8.9 nm. The Mn, Co and Fe ferrites were magnetic in nature with saturation magnetization of 44.6, 47.38, 56.19 emu/g respectively, whereas the Zn ferrite was paramagnetic. The experimental values of ln(KSP) and the average particle size showed a linear dependence, in agreement with nucleation theory. Acknowledgements We thank Dr. P.R. Vasudeva Rao, Mr. P. Kalyanasundaram and Dr. T. Jayakumar for support and encouragements and Mr. Mahadevan for XRD data. JP would like to thank Prof. Bernard Binks of University of Hull, U.K. for extending the TEM facilities. References [1] A. Bharde, A. Wani, Y. Shouche, P.A. Joy, B.L.V. Prasad, M. Sastry, J. Am. Chem. Soc. 127 (2005) 9326. [2] J.W. Long, M.S. Logan, C.P. Rhodes, E.E. Carpenter, R.M. Stroud, D.R. Rolison, J. Am. Chem. Soc. 126 (2004) 16879. [3] S. Sun, H. Zeng, D.B. Robinson, S. Raoux, P.M. Rice, S.X. Wang, G. Li, J. Am. Chem. Soc. 126 (2004) 273. [4] J. Wang, Q. Chen, C. Zeng, B. Hou, Adv. Mater. 16 (2004) 137. [5] C.-R. Lin, Y.-M. Chu, S.-C. Wang, Mater. Lett. 60 (2006) 447. [6] Y. Qu, H. Yang, N. Yang, Y. Fan, H. Zhu, G. Zou, Mater. Lett. 60 (2006) 3548. [7] S. Veintemillas-Verdaguer, M.P. Morales, C.J. Serna, Mater. Lett. 35 (1998) 227. [8] K.M. Krishnan, A.B. Pakhomov, Y. Bao, P. Blomqvist, Y. Chun, M. Gonzales, K. Griffin, X. Ji, B.K. Roberts, J. Mater. Sci. 41 (2006) 793. [9] K. Raj, B. Moskowitz, R. Casciari, J. Magn. Magn. Mater. 149 (1995) 174. [10] D.E. Speliotis, J. Magn. Magn. Mater. 193 (1999) 29. [11] J. Philip, G.G. Prakash, T. Jayakumar, P. Kalyanasudaram, B. Raj, Phys. Rev. Lett. 89 (2002) 268301; Macromolecules 36 (2003) 9230. [12] J.A.L. Perez, M.A.L. Quintela, J. Mira, J. Rivas, S.W. Charles, J. Phys. Chem., B 101 (1997) 8045. [13] T. Bala, C.R. Shankar, M. Baidakova, V. Osipov, T. Enoki, P.A. Joy, B.L.V. Prasad, M. Sastry, Langmuir 21 (2005) 10638. [14] L. Vayssieres, C. Chanaec, E. Tronc, J.P. Jolivet, J. Colloid Interface Sci. 205 (1998) 205. [15] G. Gnanaprakash, S. Ayyappan, T. Jayakumar, J. Philip, B. Raj, Nanotechnology 17 (2006) 5851. [16] Z.X. Tang, C.M. Sorensen, K.J. Klabunde, G.C. Hadjipanayis, J. Colloid Interface Sci. 146 (1991) 38. [17] B.D. Cullity, Introduction to Magnetic Materials, Addison-Wesley Publishing Company, 1972. [18] P.M.G. Nambissan, C. Upadhyaya, H.C. Verma, J. Appl. Phys. 93 (2003) 6320. [19] A.R. West, Solid State Chemistry and Its Applications, John Wiley & Sons Ltd., 1984. [20] J.-P. Jolivet, E. Tronc, C. Chaneac, C.R. Geosci. 338 (2006) 488. [21] J.W. Mullin, Crystallisation, Butterworth & Co (Publishers) Ltd., London, 1972. [22] A.J. Rondinone, C. Liu, Z.J. Zhang, J. Phys. Chem., B 105 (2001) 7967.