- Email: [email protected]
Properties of nanocomposites of α-Fe and Fe3O4
Journal of Magnetism and Magnetic Materials 246 (2002) 162–168
Properties of nanocomposites of a-Fe and Fe3O4 P. Brahmaa, S. Banerjeeb, D. Dasc, P.K. Mukhopadhyayd, S. Chatterjeee, A.K. Nigame, D. Chakravortyf,* a Gurudas College, Calcutta 700 054, India Department of Physics, University of Calcutta, Calcutta 700 009, India c IUC-DAEF, India d S.N. Bose Centre for Basic Sciences, India e Tata Institute of Fundamental Research, Mumbai, India f Indian Association for the Cultivation of Science, 8/1, Clarke Street, Calcutta 700 032, India b
Received 25 October 2001; received in revised form 18 December 2001
Abstract Composites of a-Fe and Fe3O4 having dimensions in the range 10–20 nm have been prepared by subjecting micrometer-sized a-Fe2O3 powders to a reduction treatment in hydrogen at a temperature 683 K for a duration from 5 to 40 min. The specimens have been characterized by Mossbauer spectroscopy. The latter shows the presence of a substantial amount of superparamagnetic particles. The fraction of these particles increases as the reduction treatment is enhanced. The incorporation of hydrogen atoms during the reduction process appears to break down the precursor oxide particles. Magnetic susceptibility shows a maximum at around 125 K which arises due to Verwey transition in Fe3O4 particles. Magnetization measurements on zero-field cooled and field cooled specimens indicate a blocking temperature close to 300 K. Coercivity variation with temperature shows an unusual behaviour. Exchange coupling between the ferro- and ferri-magnetic particles seems to be the reason behind this. The presence of such nearest neighbouring pairs of a-Fe/Fe3O4 is confirmed by the DC resistivity data which show electron tunnelling between metal islands separated by Fe3O4 granules. r 2002 Elsevier Science B.V. All rights reserved. Keywords: Nanocomposites; a-Fe; Mossbauer spectroscopy; Superparamagnetic particles
1. Introduction Nanostructured magnetic materials have been a subject of great scientific and technological interest [1–3] in recent years. Some of the interesting properties reported are superparamagnetism [4], *Corresponding author. Tel.: +91-33-473-4688; fax: +9133-473-2805. E-mail address: [email protected] (D. Chakravorty).
enhanced coercivity [5], quantum tunnelling of magnetization [6] and giant magnetoresistance [7]. A number of techniques, both physical [8–11] and chemical [12–14], have been used to make magnetic nanoparticles. Heterogeneous systems comprising both ferromagnetic and ferrimagnetic phases have been reported to show intriguing magnetic properties [15]. We have used a simple chemical reduction method to prepare a heterogeneous composite of nanosized a-Fe and Fe3O4 particles. These have been characterized by X-ray
0304-8853/02/$ - see front matter r 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 8 8 5 3 ( 0 2 ) 0 0 0 4 4 - 6
P. Brahma et al. / Journal of Magnetism and Magnetic Materials 246 (2002) 162–168
diffraction, Mossbauer spectroscopy, magnetization and electrical measurements. The results indicate the importance of exchange interaction between spins on ferro- and ferri-magnetic particles. The details are reported in this paper.
2. Experimental The starting material was analytical reagent grade a-Fe2O3 of average particle diameter 5 mm as procured from Aldrich Chemicals, USA. Measured amount (few grams) of a-Fe2O3 was taken in an alumina boat which was inserted in a furnace. The latter consisted of a quartz tube with heating coil surrounding it. The sample temperature was raised to 683 K while argon gas was passed through the quartz tube. After equilibration at this temperature hydrogen was passed through the furnace chamber at a predetermined rate. Batches of samples were reduced for durations extending from 5 to 40 min. After the reduction treatment the samples were cooled slowly to room temperature while keeping the hydrogen flow on. The reduced powder was characterized by taking X-ray diffractograms in a Seifert XRD 3000 P diffractometer using Cu Ka radiation. The crystallite sizes of the magnetic phases grown by reduction of a-Fe2O3 powder were estimated from the broadening of the prominent diffraction lines using the Scherrer equation [16] d¼
0:89l ; b cos y
163
33.33 Hz in a magnetic field of 100 Oe. Magnetization vs. field measurements in the temperature range 2–300 K were carried out in a vibrating sample magnetometer. For electrical measurements pellets of 1 cm diameter were prepared by taking the reduced powder in a mould of that diameter and cold pressing the same by a load of 1 tonne. Silver paint (supplied by Acheson Colloiden, The Netherlands) was applied on the two faces of the sample and DC electrical resistivity measured in the temperature range 80–300 K using a Keithley 617 Electrometer.
3. Results and discussion Fig. 1 is the X-ray diffractogram obtained for the sample subjected to a reduction treatment at 683 K for 5 min. The phases present are found to be a-Fe and Fe3O4. This is typical for the other samples. Taking the most prominent lines for these
ð1Þ
where l is the wavelength of radiation, b is the line broadening (in radians), y is the angle of diffraction and d is the particle size. Mossbauer spectra of the samples were recorded in a conventional set-up operating in a constant acceleration mode. The source used was a 10 mCi 57Co in Rhodium matrix. A 25 mm thick highpurity natural iron foil was used to calibrate the velocity drive. The experimental data were fitted with a least-squares programme assuming Lorentzian lineshape. Magnetic susceptibility of the different specimens was measured over the temperature range 80–300 K by a double balance susceptibility coil at
Fig. 1. X-ray diffractogram for specimen (a-Fe2O3) subjected to a reduction treatment in hydrogen at 683 K for 5 min: (J) Fe3O4; and (&) a-Fe.
P. Brahma et al. / Journal of Magnetism and Magnetic Materials 246 (2002) 162–168
164
Table 1 Average particle sizes for a-Fe and Fe3O4 in different samples produced by the reduction of a-Fe2O3 from X-ray line broadening data Sample no. 1 2 3
Reduction treatment 683 K for 5 min. 683 K for 12 min. 683 K for 40 min.
Average particle size d (nm) a-Fe
Fe3O4
22.6
19.6
20.8
13.8
10.6
19.6
phases as indicated in the figure we estimated the average particle size from the observed linewidth using Eq. (1). The d values estimated from the X-ray diffraction data for a-Fe and Fe3O4 phases in different samples are summarized in Table 1. It is evident that as the reduction treatment is increased the average size of a-Fe decreases. In the case of Fe3O4 the size decreases first and then it shows an increasing trend. The Mossbauer data (see below) are also found to be consistent with these results. The lowering of the particle size is believed to arise due to breaking up of the grain size of original a-Fe2O3 particles as a result of hydrogen atom incorporation into it during the reduction treatment [17]. The reduction of a-Fe2O3 evidently occurs by a two-step process, viz., a-Fe2O3 to Fe3O4 followed by Fe3O4 to a-Fe. It appears therefore that the change of particle size of the two phases Fe3O4 and a-Fe, respectively, with enhancement of reduction treatment is a consequence of the competition between the two processes delineated as above. Fig. 2 shows the Mossbauer spectra obtained in the case of different specimens. Figs. 2(a–c) give the spectra for the samples subjected to a reduction treatment at 683 K for 5, 12 and 40 min, respectively. Each of these spectra could be fitted to three sextets and a doublet. The sextet I with almost zero isomer shift and internal magnetic field (Hint ) B330 kOe is identified as arising due to the presence of a-iron [18]. Sextet II with Hint B495 kOe is attributed to A-site of Fe3O4 and Sextet III with Hint B463 kOe arises due to B-site of Fe3O4 [19]. The strong doublet at the central
region is attributed to superparamagnetic relaxation of the ultrafine particles of a-iron and Fe3O4, respectively, present in the specimen. Fig. 2(d) shows the Mossbauer spectra obtained at a temperature 70 K in the case of specimen subjected to a reduction treatment at 683 K for 40 min. It is evident that the intensity of the central doublet is drastically reduced thereby confirming that this arises due to the presence of superparamagnetic particles. Assuming the Lamb Mossbauer factors to be identical for the constituents the relative concentrations of the different phases were calculated from the corresponding resonance areas. The values are summarized in Table 2. It is evident that the amount of superparamagnetic particles increases as the reduction treatment is enhanced. This result is consistent with the data shown in Table 1 viz., the particle size of a-Fe decreases as the reduction time is raised. In Fig. 3 the variation of magnetic susceptibility (w) as a function of temperature for different specimens is shown. The data taken on pure Fe3O4 powder supplied by Aldrich Chemicals are also shown in the figure. All the samples show a peak at around 125 K. This arises due to the Verwey transition [20–22]. An order–disorder transformation involving occupation of octahedral and tetrahedral sites by Fe2+ and Fe3+ ions takes place at this temperature. This causes a peak in the susceptibility temperature plot as seen in the present case. The lowering of the peak value of w as the reduction time is increased from 5 to 40 min is consistent with the reduction of the amount of Fe3O4 in the sample concerned as shown in Table 2. Fig. 4 gives the variation of magnetization as a function of temperature under field cooled (FC) and zero-field cooled (ZFC) conditions in the temperature range 2–300 K for specimen no. 1. The magnetic field used for the FC experiment was 50 Oe. The peak observed at around 120 K is due to the Verwey transition as discussed earlier. The effective blocking temperature cannot be determined from the present data because the maximum measurement temperature was 300 K. The FC and ZFC curves are close to convergence at 300 K but lack of further data beyond 300 K makes it difficult to delineate the exact value of
P. Brahma et al. / Journal of Magnetism and Magnetic Materials 246 (2002) 162–168
165
Fig. 2. Mossbauer spectra for different specimens: (a) specimen no. 1, (b) specimen no. 2, (c) specimen no. 3, and (d) specimen no. 3 at 70 K. Table 2 Relative concentrations of different phases as estimated from Mossbauer data Specimen no.
a-Fe (%)
Fe3O4 (%)
Superparamagnetic particles (a-Fe and Fe3O4) (%)
1 2 3 3 (at 70 K)
25 25 32 64
49 45 31 36
26 30 37
blocking temperature for the present sample system. However, the trend of data is consistent with the fact that the presence of superparamag-
netic particles in this sample has been confirmed by the Mossbauer data as shown in Fig. 2. In Fig. 5 the variation of coercivity (HC ) as a function of temperature for specimen no. 1 is shown. The dots are the data points and the solid line is drawn to guide the eye. The data indicate that the variation is different from what is expected in the case of a system of non-interacting nanosized ferromagnetic particles [23]. In the present system, the latter condition is evidently not satisfied. It should be apparent from the mechanism of nanoparticle formation here that the system consists of ferromagnetic (a-Fe) and ferrimagnetic (Fe3O4) nanoparticles forming a composite structure. This means that there would
166
P. Brahma et al. / Journal of Magnetism and Magnetic Materials 246 (2002) 162–168
Fig. 5. Variation of coercivity (HC ) as a function of temperature for specimen no. 1. Fig. 3. Variation of magnetic susceptibility as a function of temperature for different specimens: pure Fe3O4, specimen no. 1, specimen no. 2, and specimen no. 3.
could be due to the presence of (i) neighbouring nanoparticles of the types a-Fe and Fe3O4 and (ii) nanoparticles of identical variety viz., a-Fe/a-Fe or Fe3O4/Fe3O4. At this stage, it is not possible to delineate these mechanisms. However, presence of pairs like a-Fe/Fe3O4 and Fe3O4/Fe3O4 or a-Fe/ a-Fe is confirmed by the electrical resistivity data as discussed below. Fig. 6 gives the resistivity variation as a function of inverse temperature for all the three specimens. It is evident that the resistivity variation is controlled by more than one activated process. We have fitted the data to the following equation which incorporates two activated mechanisms: 1 f ð1 f Þ ¼ þ ; ð2Þ r r1 r2
Fig. 4. Variation of magnetization as a function of temperature under field cooled (FC) and zero-field cooled (ZFC) conditions in the case of specimen no. 1. Measuring field: 50 Oe.
be exchange coupling between the a-Fe and Fe3O4 particles. There appears to be two mechanisms present which control the HC variation. These
where r is the over all resistivity r1 ; r2 are the resistivities due to the two processes, and f is the volume fraction contributing to the process represented by r1 : We also represent r1 ; and r2 by the following relations: r1 ¼ r01 expðW1 =kTÞ;
ð3Þ
r2 ¼ r02 expðW2 =kTÞ;
ð4Þ
P. Brahma et al. / Journal of Magnetism and Magnetic Materials 246 (2002) 162–168
Fig. 6. Variation of DC resistivity as a function of inverse temperature for different specimens.
Table 3 Extracted parameters by fitting of resistivity data to Eq. (3) Specimen no.
1
2
3
r01 W1 r02 W2 f
6.3 102 0.06 6.1 0.18 0.60
6.0 102 0.06 5.5 0.11 0.61
4.7 0.06 4.1 0.12 0.60
(O cm) (eV) (O cm) (eV)
where, r01 ; and r02 are the pre-exponential factors, W1 and W2 are the two activation energies, k is the Boltzmann constant and T is the temperature. The theoretically fitted curves are shown in Fig. 7 by the solid lines. The agreement between the theoretical fits and the experimental data is satisfactory. The parameters obtained by this fitting procedure are summarized in Table 3. The activated process represented by W1 (=0.06 eV) is ascribed to an electron tunnelling mechanism between metal nanoparticles. The activation energy for such tunneling is given by [25] 1:44 1 1 f¼ eV; ð5Þ e r rþs
167
where e is the dielectric constant of the intervening medium, r is the particle radius and s the interparticle separation both expressed in nanometers. Considering r ¼ 5 nm; s ¼ 15 nm and eB10; we estimate a value of fB0:02 eV. This is of the same order of magnitude as that estimated by curve fitting described above. The value of W2 is B0.11 eV. This is of the same order as that exhibited by Fe3O4 [24]. We therefore conclude that an assembly of nanosized a-Fe and Fe3O4, respectively, present in the specimen system give rise to the resistivity variation as shown in Fig. 7. In summary, nanocomposites of a-Fe and Fe3O4 have been synthesized by subjecting micron sized a-Fe2O3 powders to a reduction treatment in hydrogen at a temperature of 683 K for durations varying from 5 to 40 min. The particle dimensions have been found from X-ray line broadening analysis to be around 10–20 nm. Mossbauer spectroscopy analysis of these specimens indicate the presence of superparamagnetic particles the fraction of which increases as the reduction treatment is enhanced. The incorporation of hydrogen atom is believed to cause the breaking up of the precursor oxide particles. Magnetic susceptibility as a function of temperature shows a maximum at around 125 K. This arises due to Verwey transition in Fe3O4 particles. Magnetization measurements on ZFC and FC specimens indicate a blocking temperature close to 300 K. Coercivity (HC ) variation with temperature exhibits an increase with lowering of temperature but with a curvature opposite to that shown by noninteracting ferromagnetic particles. Exchange coupling between neighbouring particles probably causes this behaviour. The DC resistivity data have been analysed on the basis of existence of neighbouring pairs of a-Fe/ Fe3O4 and Fe3O4/ Fe3O4, respectively.
References [1] R.H. Kodama, A.E. Berkowitz, E.J. McNiff Jr., S. Foner, Phys. Rev. Lett. 77 (1996) 394. [2] F.T. Parker, F.F. Spada, T.J. Cox, A.E. Berkowitz, J. Magn. Magn. Mater. 162 (1996) 122. [3] D. Zhang, K.J. Klabunde, C.M. Sorensen, G.C. Hadjipanayis, Phys. Rev. B 58 (1998) 14167.
168
P. Brahma et al. / Journal of Magnetism and Magnetic Materials 246 (2002) 162–168
[4] I.S. Jacobs, C.P. Bean, in: G.T. Rado, H. Suhl (Eds.), Magnetism, Vol. 3, Academic Press, New York, 1963 (Chapter 6). [5] E.F. Kneller, F.E. Luborsky, J. Appl. Phys. 34 (1963) 656. [6] E.M. Chudnovsky, L. Gunther, Phys. Rev. Lett. 60 (1988) 661. [7] J.Q. Xiao, J.S. Jiang, C.L. Chien, Phys. Rev. Lett. 68 (1992) 3745. [8] A.S. Edelstein, et al., Studies of Magnetic Properties of Fine Particles, Elsevier Science, New York, 1992, p. 47. [9] H. Gleiter, Prog. Mater. Sci. 33 (1998) 223. [10] S.S. Parkin, N. More, K.P. Roche, Phys. Rev. Lett. 64 (1990) 2304. [11] C.C. Koch, Nanostructured Mater. 9 (1997) 13. [12] J.P. Chen, et al., Phys. Rev. B 51 (1995) 11527. [13] J.A. Becker, et al., J. Chem. Phys. 103 (1995) 2520. [14] D. Zhang, G. Glavee, K.J. Klabunde, G.C. Hadjipanayis, C.M. Sorensen, High Temp. Mater. Sci. 36 (1996) 93. [15] C. Prados, M. Multigner, A. Hernando, J.C. Sanchez, A. Fernandez, C.F. Conde, A. Conde, J. Appl. Phys. 85 (1999) 6118.
[16] H.P. Klug, L.E. Alexander, X-ray diffraction procedures for polycrystalline and amorphous materials, 2nd Edition, Wiley Interscience, New York, 1974, p. 689. [17] M. Pal, D. Das, S.N. Chintalapudi, D. Chakravorty, J. Mater. Res. 15 (2000) 683. [18] A. Chatterjee, D. Das, D. Chakravorty, K. Choudhury, Appl. Phys. Lett. 57 (1990) 1360. [19] E. Murad, J.H. Johnston, Iron Oxides and Hydroxides, in: G.J. Long (Ed.), Mossbauer Spectroscopy Applied to Inorganic Chemistry, Vol. 2, Plenum Press, New York, 1987, p. 514. [20] E.J.W. Verwey, P.W. Haayman, Physica 8 (1941) 979. [21] E.J.W. Verwey, P.W. Haayman, F.C. Romeijn, J. Chem. Phys. 15 (1947) 181. [22] Liang Wang, Jianmin Li, Weiping Ding, Tiejun Zhou, Bin Liu, Wei Zhong, Jian Wu, Youwei Du, J. Magn. Magn. Mater. 207 (1999) 111. [23] S. Banerjee, S. Roy, J.W. Chen, D. Chakravorty, J. Magn. Magn. Mater. 219 (2000) 45. [24] J. Smit, H.P.J. Wijn, Ferrites, Wiley, New York, 1959, p. 235. [25] C.A. Neugebauer, M.B. Webb, J. Appl. Phys. 33 (1962) 74.
Properties of nanocomposites of a-Fe and Fe3O4 P. Brahmaa, S. Banerjeeb, D. Dasc, P.K. Mukhopadhyayd, S. Chatterjeee, A.K. Nigame, D. Chakravortyf,* a Gurudas College, Calcutta 700 054, India Department of Physics, University of Calcutta, Calcutta 700 009, India c IUC-DAEF, India d S.N. Bose Centre for Basic Sciences, India e Tata Institute of Fundamental Research, Mumbai, India f Indian Association for the Cultivation of Science, 8/1, Clarke Street, Calcutta 700 032, India b
Received 25 October 2001; received in revised form 18 December 2001
Abstract Composites of a-Fe and Fe3O4 having dimensions in the range 10–20 nm have been prepared by subjecting micrometer-sized a-Fe2O3 powders to a reduction treatment in hydrogen at a temperature 683 K for a duration from 5 to 40 min. The specimens have been characterized by Mossbauer spectroscopy. The latter shows the presence of a substantial amount of superparamagnetic particles. The fraction of these particles increases as the reduction treatment is enhanced. The incorporation of hydrogen atoms during the reduction process appears to break down the precursor oxide particles. Magnetic susceptibility shows a maximum at around 125 K which arises due to Verwey transition in Fe3O4 particles. Magnetization measurements on zero-field cooled and field cooled specimens indicate a blocking temperature close to 300 K. Coercivity variation with temperature shows an unusual behaviour. Exchange coupling between the ferro- and ferri-magnetic particles seems to be the reason behind this. The presence of such nearest neighbouring pairs of a-Fe/Fe3O4 is confirmed by the DC resistivity data which show electron tunnelling between metal islands separated by Fe3O4 granules. r 2002 Elsevier Science B.V. All rights reserved. Keywords: Nanocomposites; a-Fe; Mossbauer spectroscopy; Superparamagnetic particles
1. Introduction Nanostructured magnetic materials have been a subject of great scientific and technological interest [1–3] in recent years. Some of the interesting properties reported are superparamagnetism [4], *Corresponding author. Tel.: +91-33-473-4688; fax: +9133-473-2805. E-mail address: [email protected] (D. Chakravorty).
enhanced coercivity [5], quantum tunnelling of magnetization [6] and giant magnetoresistance [7]. A number of techniques, both physical [8–11] and chemical [12–14], have been used to make magnetic nanoparticles. Heterogeneous systems comprising both ferromagnetic and ferrimagnetic phases have been reported to show intriguing magnetic properties [15]. We have used a simple chemical reduction method to prepare a heterogeneous composite of nanosized a-Fe and Fe3O4 particles. These have been characterized by X-ray
0304-8853/02/$ - see front matter r 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 4 - 8 8 5 3 ( 0 2 ) 0 0 0 4 4 - 6
P. Brahma et al. / Journal of Magnetism and Magnetic Materials 246 (2002) 162–168
diffraction, Mossbauer spectroscopy, magnetization and electrical measurements. The results indicate the importance of exchange interaction between spins on ferro- and ferri-magnetic particles. The details are reported in this paper.
2. Experimental The starting material was analytical reagent grade a-Fe2O3 of average particle diameter 5 mm as procured from Aldrich Chemicals, USA. Measured amount (few grams) of a-Fe2O3 was taken in an alumina boat which was inserted in a furnace. The latter consisted of a quartz tube with heating coil surrounding it. The sample temperature was raised to 683 K while argon gas was passed through the quartz tube. After equilibration at this temperature hydrogen was passed through the furnace chamber at a predetermined rate. Batches of samples were reduced for durations extending from 5 to 40 min. After the reduction treatment the samples were cooled slowly to room temperature while keeping the hydrogen flow on. The reduced powder was characterized by taking X-ray diffractograms in a Seifert XRD 3000 P diffractometer using Cu Ka radiation. The crystallite sizes of the magnetic phases grown by reduction of a-Fe2O3 powder were estimated from the broadening of the prominent diffraction lines using the Scherrer equation [16] d¼
0:89l ; b cos y
163
33.33 Hz in a magnetic field of 100 Oe. Magnetization vs. field measurements in the temperature range 2–300 K were carried out in a vibrating sample magnetometer. For electrical measurements pellets of 1 cm diameter were prepared by taking the reduced powder in a mould of that diameter and cold pressing the same by a load of 1 tonne. Silver paint (supplied by Acheson Colloiden, The Netherlands) was applied on the two faces of the sample and DC electrical resistivity measured in the temperature range 80–300 K using a Keithley 617 Electrometer.
3. Results and discussion Fig. 1 is the X-ray diffractogram obtained for the sample subjected to a reduction treatment at 683 K for 5 min. The phases present are found to be a-Fe and Fe3O4. This is typical for the other samples. Taking the most prominent lines for these
ð1Þ
where l is the wavelength of radiation, b is the line broadening (in radians), y is the angle of diffraction and d is the particle size. Mossbauer spectra of the samples were recorded in a conventional set-up operating in a constant acceleration mode. The source used was a 10 mCi 57Co in Rhodium matrix. A 25 mm thick highpurity natural iron foil was used to calibrate the velocity drive. The experimental data were fitted with a least-squares programme assuming Lorentzian lineshape. Magnetic susceptibility of the different specimens was measured over the temperature range 80–300 K by a double balance susceptibility coil at
Fig. 1. X-ray diffractogram for specimen (a-Fe2O3) subjected to a reduction treatment in hydrogen at 683 K for 5 min: (J) Fe3O4; and (&) a-Fe.
P. Brahma et al. / Journal of Magnetism and Magnetic Materials 246 (2002) 162–168
164
Table 1 Average particle sizes for a-Fe and Fe3O4 in different samples produced by the reduction of a-Fe2O3 from X-ray line broadening data Sample no. 1 2 3
Reduction treatment 683 K for 5 min. 683 K for 12 min. 683 K for 40 min.
Average particle size d (nm) a-Fe
Fe3O4
22.6
19.6
20.8
13.8
10.6
19.6
phases as indicated in the figure we estimated the average particle size from the observed linewidth using Eq. (1). The d values estimated from the X-ray diffraction data for a-Fe and Fe3O4 phases in different samples are summarized in Table 1. It is evident that as the reduction treatment is increased the average size of a-Fe decreases. In the case of Fe3O4 the size decreases first and then it shows an increasing trend. The Mossbauer data (see below) are also found to be consistent with these results. The lowering of the particle size is believed to arise due to breaking up of the grain size of original a-Fe2O3 particles as a result of hydrogen atom incorporation into it during the reduction treatment [17]. The reduction of a-Fe2O3 evidently occurs by a two-step process, viz., a-Fe2O3 to Fe3O4 followed by Fe3O4 to a-Fe. It appears therefore that the change of particle size of the two phases Fe3O4 and a-Fe, respectively, with enhancement of reduction treatment is a consequence of the competition between the two processes delineated as above. Fig. 2 shows the Mossbauer spectra obtained in the case of different specimens. Figs. 2(a–c) give the spectra for the samples subjected to a reduction treatment at 683 K for 5, 12 and 40 min, respectively. Each of these spectra could be fitted to three sextets and a doublet. The sextet I with almost zero isomer shift and internal magnetic field (Hint ) B330 kOe is identified as arising due to the presence of a-iron [18]. Sextet II with Hint B495 kOe is attributed to A-site of Fe3O4 and Sextet III with Hint B463 kOe arises due to B-site of Fe3O4 [19]. The strong doublet at the central
region is attributed to superparamagnetic relaxation of the ultrafine particles of a-iron and Fe3O4, respectively, present in the specimen. Fig. 2(d) shows the Mossbauer spectra obtained at a temperature 70 K in the case of specimen subjected to a reduction treatment at 683 K for 40 min. It is evident that the intensity of the central doublet is drastically reduced thereby confirming that this arises due to the presence of superparamagnetic particles. Assuming the Lamb Mossbauer factors to be identical for the constituents the relative concentrations of the different phases were calculated from the corresponding resonance areas. The values are summarized in Table 2. It is evident that the amount of superparamagnetic particles increases as the reduction treatment is enhanced. This result is consistent with the data shown in Table 1 viz., the particle size of a-Fe decreases as the reduction time is raised. In Fig. 3 the variation of magnetic susceptibility (w) as a function of temperature for different specimens is shown. The data taken on pure Fe3O4 powder supplied by Aldrich Chemicals are also shown in the figure. All the samples show a peak at around 125 K. This arises due to the Verwey transition [20–22]. An order–disorder transformation involving occupation of octahedral and tetrahedral sites by Fe2+ and Fe3+ ions takes place at this temperature. This causes a peak in the susceptibility temperature plot as seen in the present case. The lowering of the peak value of w as the reduction time is increased from 5 to 40 min is consistent with the reduction of the amount of Fe3O4 in the sample concerned as shown in Table 2. Fig. 4 gives the variation of magnetization as a function of temperature under field cooled (FC) and zero-field cooled (ZFC) conditions in the temperature range 2–300 K for specimen no. 1. The magnetic field used for the FC experiment was 50 Oe. The peak observed at around 120 K is due to the Verwey transition as discussed earlier. The effective blocking temperature cannot be determined from the present data because the maximum measurement temperature was 300 K. The FC and ZFC curves are close to convergence at 300 K but lack of further data beyond 300 K makes it difficult to delineate the exact value of
P. Brahma et al. / Journal of Magnetism and Magnetic Materials 246 (2002) 162–168
165
Fig. 2. Mossbauer spectra for different specimens: (a) specimen no. 1, (b) specimen no. 2, (c) specimen no. 3, and (d) specimen no. 3 at 70 K. Table 2 Relative concentrations of different phases as estimated from Mossbauer data Specimen no.
a-Fe (%)
Fe3O4 (%)
Superparamagnetic particles (a-Fe and Fe3O4) (%)
1 2 3 3 (at 70 K)
25 25 32 64
49 45 31 36
26 30 37
blocking temperature for the present sample system. However, the trend of data is consistent with the fact that the presence of superparamag-
netic particles in this sample has been confirmed by the Mossbauer data as shown in Fig. 2. In Fig. 5 the variation of coercivity (HC ) as a function of temperature for specimen no. 1 is shown. The dots are the data points and the solid line is drawn to guide the eye. The data indicate that the variation is different from what is expected in the case of a system of non-interacting nanosized ferromagnetic particles [23]. In the present system, the latter condition is evidently not satisfied. It should be apparent from the mechanism of nanoparticle formation here that the system consists of ferromagnetic (a-Fe) and ferrimagnetic (Fe3O4) nanoparticles forming a composite structure. This means that there would
166
P. Brahma et al. / Journal of Magnetism and Magnetic Materials 246 (2002) 162–168
Fig. 5. Variation of coercivity (HC ) as a function of temperature for specimen no. 1. Fig. 3. Variation of magnetic susceptibility as a function of temperature for different specimens: pure Fe3O4, specimen no. 1, specimen no. 2, and specimen no. 3.
could be due to the presence of (i) neighbouring nanoparticles of the types a-Fe and Fe3O4 and (ii) nanoparticles of identical variety viz., a-Fe/a-Fe or Fe3O4/Fe3O4. At this stage, it is not possible to delineate these mechanisms. However, presence of pairs like a-Fe/Fe3O4 and Fe3O4/Fe3O4 or a-Fe/ a-Fe is confirmed by the electrical resistivity data as discussed below. Fig. 6 gives the resistivity variation as a function of inverse temperature for all the three specimens. It is evident that the resistivity variation is controlled by more than one activated process. We have fitted the data to the following equation which incorporates two activated mechanisms: 1 f ð1 f Þ ¼ þ ; ð2Þ r r1 r2
Fig. 4. Variation of magnetization as a function of temperature under field cooled (FC) and zero-field cooled (ZFC) conditions in the case of specimen no. 1. Measuring field: 50 Oe.
be exchange coupling between the a-Fe and Fe3O4 particles. There appears to be two mechanisms present which control the HC variation. These
where r is the over all resistivity r1 ; r2 are the resistivities due to the two processes, and f is the volume fraction contributing to the process represented by r1 : We also represent r1 ; and r2 by the following relations: r1 ¼ r01 expðW1 =kTÞ;
ð3Þ
r2 ¼ r02 expðW2 =kTÞ;
ð4Þ
P. Brahma et al. / Journal of Magnetism and Magnetic Materials 246 (2002) 162–168
Fig. 6. Variation of DC resistivity as a function of inverse temperature for different specimens.
Table 3 Extracted parameters by fitting of resistivity data to Eq. (3) Specimen no.
1
2
3
r01 W1 r02 W2 f
6.3 102 0.06 6.1 0.18 0.60
6.0 102 0.06 5.5 0.11 0.61
4.7 0.06 4.1 0.12 0.60
(O cm) (eV) (O cm) (eV)
where, r01 ; and r02 are the pre-exponential factors, W1 and W2 are the two activation energies, k is the Boltzmann constant and T is the temperature. The theoretically fitted curves are shown in Fig. 7 by the solid lines. The agreement between the theoretical fits and the experimental data is satisfactory. The parameters obtained by this fitting procedure are summarized in Table 3. The activated process represented by W1 (=0.06 eV) is ascribed to an electron tunnelling mechanism between metal nanoparticles. The activation energy for such tunneling is given by [25] 1:44 1 1 f¼ eV; ð5Þ e r rþs
167
where e is the dielectric constant of the intervening medium, r is the particle radius and s the interparticle separation both expressed in nanometers. Considering r ¼ 5 nm; s ¼ 15 nm and eB10; we estimate a value of fB0:02 eV. This is of the same order of magnitude as that estimated by curve fitting described above. The value of W2 is B0.11 eV. This is of the same order as that exhibited by Fe3O4 [24]. We therefore conclude that an assembly of nanosized a-Fe and Fe3O4, respectively, present in the specimen system give rise to the resistivity variation as shown in Fig. 7. In summary, nanocomposites of a-Fe and Fe3O4 have been synthesized by subjecting micron sized a-Fe2O3 powders to a reduction treatment in hydrogen at a temperature of 683 K for durations varying from 5 to 40 min. The particle dimensions have been found from X-ray line broadening analysis to be around 10–20 nm. Mossbauer spectroscopy analysis of these specimens indicate the presence of superparamagnetic particles the fraction of which increases as the reduction treatment is enhanced. The incorporation of hydrogen atom is believed to cause the breaking up of the precursor oxide particles. Magnetic susceptibility as a function of temperature shows a maximum at around 125 K. This arises due to Verwey transition in Fe3O4 particles. Magnetization measurements on ZFC and FC specimens indicate a blocking temperature close to 300 K. Coercivity (HC ) variation with temperature exhibits an increase with lowering of temperature but with a curvature opposite to that shown by noninteracting ferromagnetic particles. Exchange coupling between neighbouring particles probably causes this behaviour. The DC resistivity data have been analysed on the basis of existence of neighbouring pairs of a-Fe/ Fe3O4 and Fe3O4/ Fe3O4, respectively.
References [1] R.H. Kodama, A.E. Berkowitz, E.J. McNiff Jr., S. Foner, Phys. Rev. Lett. 77 (1996) 394. [2] F.T. Parker, F.F. Spada, T.J. Cox, A.E. Berkowitz, J. Magn. Magn. Mater. 162 (1996) 122. [3] D. Zhang, K.J. Klabunde, C.M. Sorensen, G.C. Hadjipanayis, Phys. Rev. B 58 (1998) 14167.
168
P. Brahma et al. / Journal of Magnetism and Magnetic Materials 246 (2002) 162–168
[4] I.S. Jacobs, C.P. Bean, in: G.T. Rado, H. Suhl (Eds.), Magnetism, Vol. 3, Academic Press, New York, 1963 (Chapter 6). [5] E.F. Kneller, F.E. Luborsky, J. Appl. Phys. 34 (1963) 656. [6] E.M. Chudnovsky, L. Gunther, Phys. Rev. Lett. 60 (1988) 661. [7] J.Q. Xiao, J.S. Jiang, C.L. Chien, Phys. Rev. Lett. 68 (1992) 3745. [8] A.S. Edelstein, et al., Studies of Magnetic Properties of Fine Particles, Elsevier Science, New York, 1992, p. 47. [9] H. Gleiter, Prog. Mater. Sci. 33 (1998) 223. [10] S.S. Parkin, N. More, K.P. Roche, Phys. Rev. Lett. 64 (1990) 2304. [11] C.C. Koch, Nanostructured Mater. 9 (1997) 13. [12] J.P. Chen, et al., Phys. Rev. B 51 (1995) 11527. [13] J.A. Becker, et al., J. Chem. Phys. 103 (1995) 2520. [14] D. Zhang, G. Glavee, K.J. Klabunde, G.C. Hadjipanayis, C.M. Sorensen, High Temp. Mater. Sci. 36 (1996) 93. [15] C. Prados, M. Multigner, A. Hernando, J.C. Sanchez, A. Fernandez, C.F. Conde, A. Conde, J. Appl. Phys. 85 (1999) 6118.
[16] H.P. Klug, L.E. Alexander, X-ray diffraction procedures for polycrystalline and amorphous materials, 2nd Edition, Wiley Interscience, New York, 1974, p. 689. [17] M. Pal, D. Das, S.N. Chintalapudi, D. Chakravorty, J. Mater. Res. 15 (2000) 683. [18] A. Chatterjee, D. Das, D. Chakravorty, K. Choudhury, Appl. Phys. Lett. 57 (1990) 1360. [19] E. Murad, J.H. Johnston, Iron Oxides and Hydroxides, in: G.J. Long (Ed.), Mossbauer Spectroscopy Applied to Inorganic Chemistry, Vol. 2, Plenum Press, New York, 1987, p. 514. [20] E.J.W. Verwey, P.W. Haayman, Physica 8 (1941) 979. [21] E.J.W. Verwey, P.W. Haayman, F.C. Romeijn, J. Chem. Phys. 15 (1947) 181. [22] Liang Wang, Jianmin Li, Weiping Ding, Tiejun Zhou, Bin Liu, Wei Zhong, Jian Wu, Youwei Du, J. Magn. Magn. Mater. 207 (1999) 111. [23] S. Banerjee, S. Roy, J.W. Chen, D. Chakravorty, J. Magn. Magn. Mater. 219 (2000) 45. [24] J. Smit, H.P.J. Wijn, Ferrites, Wiley, New York, 1959, p. 235. [25] C.A. Neugebauer, M.B. Webb, J. Appl. Phys. 33 (1962) 74.