Hyperfine Interact (2009) 191:87–93 DOI 10.1007/s10751-009-9957-0
Characterization of magnetite nanoparticles supported in sulfonated styrene-divinylbenzene mesoporous copolymer A. F. R. Rodriguez · J. A. H. Coaquira · J. G. Santos · L. B. Silveira · E. M. Marmolejo · W. Trennepohl · D. Rabelo · A. C. Oliveira · V. K. Garg · P. C. Morais
Published online: 25 March 2009 © Springer Science + Business Media B.V. 2009
Abstract The chemical co-precipitation process was used to synthesize (in situ) spherical iron-oxide nanoparticle in sulfonated styrene-divinylbenzene polymeric template. X-ray diffraction technique supports the magnetite phase formation with a mean particle diameter of about 19 nm. The analysis of Mössbauer spectra is consistent with two magnetic splitting patterns assigned to A- and B-iron sites of magnetite, with no visible magnetic relaxation effect even at 297 K. Considering the different experimental time window between Mössbauer spectroscopy and DC magnetization, the results obtained from both techniques are in very good agreement. Magnetic data suggest hosting magnetite nanoparticles interacting antiferromagnetically. Keywords Mössbauer spectroscopy · Magnetite nanoparticles · Sulfonated styrene-divinylbenzene template · Magnetic nanocomposites · Magnetization
A. F. R. Rodriguez · J. G. Santos · L. B. Silveira · E. M. Marmolejo · W. Trennepohl Departamento de Física, Fundação Universidade Federal de Rondônia, Ji-Paraná RO 78961-970, Brazil J. A. H. Coaquira · A. C. Oliveira · V. K. Garg · P. C. Morais Núcleo de Física Aplicada, Instituto de Física, Universidade de Brasília, Brasília DF 70910-900, Brazil D. Rabelo Instituto de Química, Universidade Federal de Goiás, Goiânia GO 74001-970, Brazil Present Address: A. F. R. Rodriguez (B) Departamento de Física, Universidade Federal do Acre-UFAC, Rio Branco AC 69915-900, Brazil e-mail:
[email protected]
88
A.F.R. Rodriguez et al.
1 Introduction The design and the synthesis of nanometer-scaled magnetic structures have recently attracted much attention in response to their potential use in nanotechnology, with focus in areas ranging from biological/medical applications to data storage [1–3]. These material systems are quite interesting not only due to their enhanced and unusual properties but particularly because the size-dependence they present allows the tunability required by the technological applications. As far as the nanosize and size-dispersity control are concerned the tendency of supported isolated magnetic nanostructures for aggregating into bigger clusters during in situ synthesis process, driven by particle-particle interaction and/or by the reduction of energy associated to the high surface-to-volume ratio, has represented a critical obstacle for fine properties tunability. Magnetic polymer-based spheres have been considered as an important material for the biotechnological industry, as for instance in cell separation [4] and DNA extraction [5]. In particular, mesoporous polymeric templates could be produced as micron-sized spheres that allow in situ chemical synthesis of nanosized ferrite particles with tunable magnetic properties and mass density [6]. The fine tuning of the physical parameters, e.g. net magnetic moment and density of the composite, can be achieved by performing several cycles (N) of chemical synthesis. In the present study Mössbauer spectroscopy and magnetic measurements were used to characterize magnetite nanoparticle synthesized in mesoporous sulfonated styrenedivinylbenzene (Sty-DVB) microspheres.
2 Experimental The composite preparation uses a four-step experimental procedure [7]. In the first step micron-sized sulfonated Sty-DVB spheres are immersed in ferrous sulfate (FeSO4 ) aqueous solution while stirred at room temperature, providing the polymeric template the ion exchange capacity of about 5 mmol H+ /g. Secondly, the polymer particles are separated by filtration and washed thoroughly with water until no iron is further detected in the eluent. Third, the oxidation of the ferrous ion was performed in alkaline medium at about pH 13 following a standard recipe [8]. The as-produced black composite was filtered, washed to neutral pH and dried at 60◦ C, in air. For a particular milimolar concentration (mM) of the ferrous solution, the above-described procedure can be performed N times in order to obtain the N-cycle composite. The present investigation is focused on samples fabricated by treating the polymeric template with a 2 mmol/L (2 mM) ferrous aqueous solution while using different chemical cycles (N = 1, 3, 4, 5, and 6). It is well known that by increasing the number N of chemical cycles one increases the nanoparticle concentration dispersed in the polymeric template with small changes in the average particle diameter [9]. Transmission electron microscopy (TEM), X-ray diffraction (XRD), magnetic measurements and Mössbauer spectroscopy were used to characterize the following samples: 1-cycle (1C2M), 3-cycles (3C2M), 4-cycles (4C2M), 5-cycles (5C2M), and 6-cycles (6C2M). Transmission Mössbauer spectra were recorded at temperatures ranging from 20 to 297 K, using a constant acceleration spectrometer with a 57 Co/Rh source, whereas magnetic experiments were carried out using both a Vibrating
Magnetite nanoparticles in Sty-DVB mesoporous copolymer Fig. 1 XRD patterns of quoted samples. The numbers represent the (hkl) reflections of the cubic Fe3 O4 phase
89
(311)
(220)
(400)
Intensity
6C2M
5C2M
4C2M
25
30
35
40
45
50
2θ (degrees)
Sample (VSM) and SQUID magnetometers in the temperature range of 4.2 to 300 K and magnetic fields up to 5 T.
3 Results and discussion X-ray diffraction (XRD) patterns show Bragg reflections which are consistent with the magnetite phase (see Fig. 1). Broadened XRD peaks indicate the formation of small particles of the magnetite phase, (Fe3+ )[Fe2+ ,Fe3+ ]O4 , where the iron ion in parenthesis occupies the tetrahedral A-site, whereas the iron ions in square brackets occupy the B-site. In order to estimate the nanocrystal size the Scherrer octahedral relation Dv = 0.9λ . cos 0 was used [10]. The full-width at half-maximum of the more intense XRD peak centered at 2θ ∼ 35◦ , corresponding to the (311) reflection, was used to determine the average magnetite diameter (D = 18.7 nm) in the 4C2M composite sample. The mean nanocrystal size remains practically the same for samples with different number of chemical cycles (N), although the XRD patterns present better signal-to-noise ratios as N increases, suggesting composite samples with higher nanoparticle concentration. The mean nanocrystal size values obtained from the Scherrer relation are consistent with those obtained from the TEM image analyses (data not shown). Figure 2 shows typical Mössbauer spectra of sample 4C2M, recorded at different temperatures. Raw experimental data indicate the presence of well defined six-line magnetic splitting and a central two-line pattern in the whole range of temperature. The spectrum obtained at T = 20 K is well resolved using one doublet with ≈10% of relative spectral area and two sextets with hyperfine magnetic fields of ∼54 and ∼52 T, which tentatively could be assigned to iron ions located in the tetrahedral site (A-site) and octahedral site (B-site), respectively. It is worth noticing that the observed hyperfine magnetic field values obtained from the analysis of 4C2M sample are above the commonly reported values for bulk magnetite [11] and are slightly higher than those reported for polymercoated magnetite nanoparticles [12]. The doublet component, with isomer shift (IS)≈0.12 mm/s and quadruple splitting (QS)≈0.50 mm/s, may be associated with the
90
A.F.R. Rodriguez et al.
1.00
0.99
54 1.00
52
0.99
Bhf (T)
Relative Transmission
20 K
84 K
ε=3/2 50
ε=2
1.00 site 1 site 2
48 0.99
Bhf (T)=Bhf(0)(1- αTε)
297 K -12
-8
-4
0
4
8
12
46 0
100
Velocity (mm/s)
200
300
Temperature (K)
Fig. 2 Left panel Mössbauer spectra of sample 4C2M obtained at different temperatures. The solid line represents the fit carried out using two six-line and one central doublet patterns. Right panel Hyperfine field as a function of the temperature for both iron sites. Solid lines represent the T2 whereas the dotted ones represent the T3/2 -dependence
0.19
FC 0.18
6C2M H=500 Oe 0.186
0.17
C
ZF
0.16
M (emu)
Magnetization (emu)
Fig. 3 ZFC and FC magnetization curves of sample 6C2M obtained applying a magnetic field of 500 Oe. The inset shows the maximum observed in the ZFC curve
0.185
Tm=242K 0.184
0.15
T1
200
250
0
300
T (K)
0.14 100
200
300
Temperature (K)
primary oxidation product (Fe3+ ) of the reagent (FeSO4 ) [13] or related to extrinsic paramagnetic impurities. The two six-line patterns remain as the main feature of the Mössbauer spectra until room temperature, with any remarkable increase in
Magnetite nanoparticles in Sty-DVB mesoporous copolymer
91
Table 1 Hyperfine parameters obtained from Mössbauer spectra of sample 4C2M at 20, 84 and 297K T (K)
Site
Bhf (T)
IS (mm/s)
QS (mm/s)
Rel. area (%)
20
Sextet 1 Sextet 2 Doublet Sextet 1 Sextet 2 Doublet Sextet 1 Sextet 2 Doublet
54.0(2) 52.3(2) – 54.2(1) 52.3(1) – 51.4(1) 50.0(−) –
0.49(2) 0.43(1) 0.11(6) 0.48(7) 0.41(6) 0.06(2) 0.32(2) 0.36(2) 0.24(6)
−0.01(3) −0.04(3) +0.50(9) −0.01(1) −0.03(1) +0.54(3) +0.05(5) −0.00(4) +0.68(12)
30.8 58.6 10.6 38.3 50.9 10.8 27.9 58.8 13.3
84
297
Numbers in parenthesis represent the uncertainties
the spectral area of the central doublet (see Table 1). This finding indicates that if thermal relaxation phenomenon is present in our samples it will be developed above room temperature. Table 1 shows the 4C2M sample’s Mössbauer parameters obtained at some temperatures. It is worth noting that at low temperatures, the IS value for the sextet 1 is slightly larger than that one for the sextet 2, whereas at room temperature the IS value of sextet 2 is larger. The hyperfine fields for both iron ion sites show a monotonic reduction as the temperature is increased. Considering the relationship between the hyperfine field (Bhf ) and the saturation magnetization (MS ), Bhf (T) ∝ MS (T), and taking into account that in the thermodynamic equilibrium the low energy collective excitations (spin waves) result in the decrease of the spontaneous magnetization as the temperature increases, the hyperfine field is expected to follow the relation [14]: Bhf (T) = Bhf (0) (1 − B0 T ε ) ,
(1)
where ε has a value between 0.3 and 2.0 for fine particles [15] and Bhf (0) is a parameter related to the magnetic coupling constant (B0 ∼ 1/J ε ). Figure 2 (right-hand side panel) shows the experimental data for both magnetite iron sites. Data points are better described for the curve corresponding to ε = 2, which is in good agreement with results reported for magnetite nanoparticles. Values of B0 ∼ 6.5–7.0 × 10−7 K−2 were determined from our data, suggesting that the value of J remains essentially the same as in bulk magnetite [14]. Zero field-cooled (ZFC) and field-cooled (FC) curves of sample 6C2M are presented in Fig. 3. The main features are: (1) the onset of irreversibility between ZFC and FC curves is observed at temperatures around 290 K and (2) a clear shoulder at T1 ∼ 20 K together with a broad maximum at Tm = 242 K. While the onset of irreversibility at temperatures above the maximum of ZFC curve indicates the presence of interacting particles the shoulder at ∼20 K seems to be related to an additional magnetic phase (probably FeSO4 or related phase), which has Néel temperature TN ∼ 21 K [16]. The maximum at Tm = 242 K is assigned to the blocking temperature (TB ) of the system. This result seems to be inconsistent with Mössbauer results, but if we take into account the experimental time window of both techniques the blocking temperature determined from Mössbauer experiments is expected to mag be TBMoss ≈ 10TM , which can explain our results. Assuming the Tm value as the
92
A.F.R. Rodriguez et al.
representative blocking temperature of the system one has Keff V kB Tm = ln ( f0 t) ∼ 25, where f0 ∼ 109 Hz, Keff is the effective anisotropy constant, and V is the particle volume. Considering spherically-shaped nanoparticles and using the particle size obtained from both the XRD data and from the TEM analysis an effective anisotropy constant of Keff = 2.6×104 J/m3 is estimated. This value is in very good agreement with the values reported for magnetite nanoparticles and is slightly above the value of the magneto-crystalline anisotropy of bulk magnetite [12], suggesting that the effect of an extra contribution to the effective anisotropy (surface or exchange) has little influence in the magnetic behavior of our composite samples. This finding may indicate that the hosting template contributes to the reduction of the surface anisotropy, probably due to the interaction between the metallic ion at the nanoparticles surface and the chemical groups (sulfonate moieties) provided by the polymer hosting matrix. Room temperature magnetization vs. magnetic field curve (not shown here) indicates samples with almost full saturation state for fields up to ∼0.5 T. The analyses of hysteresis loops provide coercive fields (HC ) with higher values for samples with smaller number of chemical cycles (N). At T = 4.2 K the value of HC ∼ 120 Oe was determined for the 6C2M sample and this value decreases down to HC ∼ 68 Oe for measurements carried out at room temperature. This non-vanishing coercive field value at room temperature strongly indicates the occurrence of particle-particle interaction. The remanence-to-saturation ratio r = MR MS determined from the 4.2 K-data gives a value of about 0.12. Similar value was determined from the room temperature curve. This value is smaller than the expected value for randomly oriented spherical grains (r = 0.5), thus suggesting the presence of magnetite nanoparticles interacting via antiferromagnetic coupling [17]. Recent calculation has also pointed out the possibility of magnetically-ordered phase resulting from particleparticle interaction [18]. Acknowledgements The authors acknowledge the financial support of the Brazilian agencies ELETRONORTE, FINATEC, CTPETRO/FINEP, and MCT/CNPq-UNIR-SEPLAN-DCR-RO.
References 1. Menon, A.K., Gupta, B.K.: Nanostruct. Mater. 11, 965 (1999) 2. Thomas, L., Lionti, F., Ballou, R., Gatteschi, D., Sessoli, R., Barbara, B.: Nature. 383, 145 (1996) 3. Macaroff, P.P., Oliveira, D.M., Ribeiro, K.F., Lacava, Z.G.M., Lima, E.C.D., Morais, P.C., Tedesco, A.C.: IEEE Trans. Magn. 41, 4105 (2005) 4. Badia, A., Cuccia, L., Demers, L., Morin, F., Lennox, R.B.: J. Am. Chem. Soc. 119, 2682 (1997) 5. Leslie-Pelecky, D.L., Rieke, R.D.: Chem. Mater. 8, 1770 (1996) 6. Morais, P.C., Azevedo, R.B., Rabelo, D., Lima, E.C.D.: Chem. Mater. 15, 2485 (2003) 7. Rabelo, D., Lima, E.C.D., Reis, A.C., Nunes, W.C., Novak, M.A., Garg, V.K., Oliveira, A.C., Morais, P.C.: Nano Lett. 1, 105 (2001) 8. Couling, S.B., Mann, S.: J. Chem. Soc., Chem. Commun. 23, 1713 (1985) 9. Morais, P.C., Azevedo, R.B., Silva, L.P., Rabelo, D., Lima, E.C.D.: Phys. Status Solidi, A Appl. Res. 187, 203 (2001) 10. Patterson, A.L.: Phys. Rev. 56, 978 (1939) 11. Evans, B.J., Hafner, S.S.: J. Appl. Phys. 40, 1411 (1969) 12. Lima, E. Jr, Brandl, A.L., Arelaro, A.D., Goya, G.F.: J. Appl. Phys. 99, 083908-1 (2006) 13. Cohen, R.L.: Applications of Mössbauer Spectroscopy, Vol, 1. Academic Press, New York, 1976, p. 226
Magnetite nanoparticles in Sty-DVB mesoporous copolymer
93
14. Thakur, M., De, K., Giri, S., Si, S., Kotal, A., Mandal, T.K.: J. Phys.: Condens. Matter. 18, 9094 (2006) 15. Kodama, R.H.: J. Magn. Magn. Mater. 200, 359 (1999) 16. Ok, H.N.: Phys. Rev. B 4, 3870 (1971) 17. Hadjipanayis, G., Sellmyer, D.J., Brandt, B.: Phys. Rev. B 23, 3349 (1981) 18. Bakuzis, A.F., Morais, P.C.: J. Magn. Magn. Mater. 285, 145 (2005)
