On the structural, magnetic and electrical properties of sol–gel derived nanosized cobalt ferrite

On the structural, magnetic and electrical properties of sol–gel derived nanosized cobalt ferrite

Journal of Alloys and Compounds 485 (2009) 711–717 Contents lists available at ScienceDirect Journal of Alloys and Compounds journal homepage: www.e...

2MB Sizes 21 Downloads 143 Views

Journal of Alloys and Compounds 485 (2009) 711–717

Contents lists available at ScienceDirect

Journal of Alloys and Compounds journal homepage: www.elsevier.com/locate/jallcom

On the structural, magnetic and electrical properties of sol–gel derived nanosized cobalt ferrite E. Veena Gopalan a , P.A. Joy b , I.A. Al-Omari c , D. Sakthi Kumar d , Yasuhiko Yoshida d , M.R. Anantharaman a,∗ a

Department of Physics, Cochin University of Science and Technology, Cochin 682022, Kerala, India Physical Chemistry Division, National Chemical Laboratory, Pune 411008, India Department of Physics, College of Sciences, P O Box 36, Sultan Qaboos University, PC 123 Muscat, Oman d Bio-Nano Electronics Research Centre, Department of Applied Chemistry, Toyo University, Kawagoe, Saitama 350-8585, Japan b c

a r t i c l e

i n f o

Article history: Received 25 May 2009 Received in revised form 5 June 2009 Accepted 6 June 2009 Available online 12 June 2009 Keywords: Nanostructures Magnetically ordered materials Sol–gel synthesis Scanning and transmission electron microscopy X-ray diffraction Electronic Transport Magnetic measurements

a b s t r a c t Nanoparticles of cobalt ferrite were synthesized by sol gel method. These particles were structurally characterized by using X-Ray Diffraction and Transmission Electron Microscopy, High Resolution Transmission Electron Microscopy, Energy Dispersive Spectrum and Inductively Coupled Plasma Analysis and the results confirmed the formation of spherically shaped nanoparticles of cobalt ferrite having a size lying in the range of 13–14 nm. The as prepared sample was sintered at 800 ◦ C and the structural, magnetic and dielectric properties were measured. The dielectric properties were studied and analyzed as a function of temperature and frequency. The ac and dc conductivity studies were carried out to delve into the conduction mechanism. The existing models based on quantum mechanical tunneling were effectively employed to explain the frequency dependent conductivity. © 2009 Elsevier B.V. All rights reserved.

1. Introduction Ferrite nanoparticles are of interest for both technological and fundamental reasons. From an applied standpoint, these materials are promising as high density storage media, as spin dependent electron transport devices and for therapeutic or diagnostic medical functions [1–3]. From a fundamental research perspective, ferrite nanoparticles offer a unique system to study magnetism at the nanolevel. Cobalt ferrite is an important member of the ferrite family and is characterized by its high coercivity, moderate magnetization and very high magnetocrystalline anisotropy. Cobalt ferrite finds innumerable applications in stress sensors, as precursors for making ferrofluids and also magnetic refrigerants [4,5]. Normally, ferrites become superparamagnetic at room temperature for nanoparticles below 10 nm [6,7]. However, cobalt ferrite nanoparticles do not show superparamagnetic behaviour unless they are made significantly smaller [8]. The critical size for the transition to superparamagnetic state appears to be around 5 nm. This is due to the large crystalline anisotropy associated with cobalt con-

∗ Correspondence author. E-mail address: [email protected] (M.R. Anantharaman). 0925-8388/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.jallcom.2009.06.033

taining compounds. Strong electron coupling at Co2+ lattice sites leads to large magnetic anisotropy which in turn gives rise to such magnetic property [9,10]. For bulk CoFe2 O4 , cubic anisotropy is known to dominate since the orbital contribution is not quenched by the lattice. The electrical and magnetic properties of bulk ferrites are found to be sensitive to a number of factors namely grain size, grain structure, porosity and distribution of the metal cations among the lattice sites in the spinel structure [11]. Bulk particles of cobalt ferrite exhibit an inverse spinel structure with one half of the Fe3+ ions in the A sites and the remaining half of Fe3+ ions and Co2+ ions in the B sites, represented as Fe3+ A [Co2+ , Fe3+ ]B . Nanocrystalline cobalt ferrite particles are found to be exhibiting interesting structural and magnetic properties as compared to their micron sized counterparts[12]. The presence of cobalt ions in the tetrahedral sites coupled with the existence of multiple oxidation states of cobalt (Co2+ /Co3+ ) can give rise to altogether different structural and magnetic properties with respect to their bulk counterparts. It has been reported recently that spin disorder on the surface of the cobalt ferrite nanoparticles leads to interesting magnetic phenomena [13]. The electrical properties exhibited by ferrites in the micron regime are normally characteristic of hopping of electrons or holes or polarons between cationic sites [14]. Thus the cation

712

E.V. Gopalan et al. / Journal of Alloys and Compounds 485 (2009) 711–717

distribution as well as the oxidation states of the cations also determines the dielectric polarisation and electrical conduction. Jonker studied the electrical properties of a series of bulk Co3−x Fex O4 and observed that the conductivity is minimum for the x = 2 composition and that is a p-type conductor for x less than x = 2 and n-type conductivity for x greater than x = 2 [15]. The p-type behaviour was explained as due to the enhanced contribution of hole hopping between Co3+ /Co2+ ions in cobalt rich phase.Though the magnetic properties of cobalt ferrite in the nanoregime have been extensively investigated, reports on the electrical properties of nanosized cobalt ferrite are not very abundant in literature. The role of cations, the cation redistribution, if any in the nanoregime, the presence of more than one valence state for the cation states and their influence on the electrical properties are seldom understood when the particle sizes approach the nanometric dimensions. This is an area where there is scope for understanding the basic mechanism of conduction in cobalt ferrite in the nanoregime. Understanding of the mechanism of conduction both in the ac and dc regime will be abundantly useful in tailoring the properties of these materials for specific applications. Such a study on cobalt ferrite nanoparticles needs the synthesis of phase pure CoFe2 O4 . For this, pristine sample was prepared using sol–gel method and was thermally treated at 800 ◦ C in order to have phase pure compound. They were characterized both structurally and magnetically. The ac and dc conductivities were also measured and analyzed. The properties of the sintered as well as the as prepared samples are compared wherever needed. 2. Experimental 2.1. Synthesis of cobalt ferrite nanoparticles Fine particles of Cobalt ferrite were synthesized by sol–gel combustion method. For this, AR grade chemical precursors of cobalt nitrate and ferric nitrate were used with ethylene glycol as the solvent. Cobalt nitrate and ferric nitrate were dissolved in ethylene glycol in the molar ratio of 1:2 at 40 ◦ C to form the sol. This sol is then heated slowly at 60 ◦ C to obtain a wet gel. The gel was then dried at 90 ◦ C. This resulted in the self ignition of the gel producing a highly voluminous and fluffy product. The product is then ground to form fine powders of cobalt ferrite and pelletised. They were sintered in the furnace at 800 ◦ C. The unsintered and sintered samples are designated as CP and C8. 2.2. Characterization The samples were characterized by using X-Ray Powder Diffractometer (Rigaku D max C) using Cu-K␣ radiation ( = 1.5405 Å). Lattice parameter was calculated assuming cubic symmetry. The average crystallite size was determined from the measured width of their diffraction curves by using Debye Scherrer formula. The particle size of the pristine sample was determined by subjecting the sample to Transmission electron microscopy (TEM), (Joel JEM-2200 FS). High Resolution Transmission Electron Microscopy Images (HRTEM) and Energy Dispersive X-ray Spectra (EDS) were also obtained. Thermo Electron Corporation, IRIS INTRPID II XSP model was used for Inductively Coupled Plasma Analysis measurement. Hysteresis loop parameters at room temperature and low temperature were evaluated by using a Vibrating Sample Magnetometer. (DMS 1660 VSM) with an external magnetic field varying from −13.5 kOe to 13.5 kOe. Scanning Electron Microscopy was employed to check the morphology of the samples (JSM-6335 FESEM). The dc conductivity of the sample was measured using a Keithley 236 Source meter (Pico ammeter). The resistance R was measured for different temperatures and the conductivity was estimated. Dielectric measurements were carried out on these samples using a home made dielectric cell and an HP 4285 LCR meter in the frequency range of 100 kHz to 8 MHz over a temperature of 303–403 K. The principle of parallel plate capacitor was employed for the evaluation of permittivity. From the measured loss factor (tan ı) and real part of the dielectric constant (ε ), ac conductivity ( ac ) of these samples were evaluated. The data acquisition was automated by interfacing the Picoammeter and LCR meter with virtual instrumentation package, LabVIEW (National Instruments).

3. Results and discussions 3.1. Structural characterization The XRD patterns of nanoparticles of pristine and sintered sample are depicted in Fig. 1 and are typical of spinel structure. Additional peaks of ␣-Fe2 O3 were found in the pristine sample.

Fig. 1. XRD diffraction patterns of cobalt ferrite. As prepared (CP) and sintered (C8).

The average size of the crystallite, calculated using Debye Scherrer formula was found to be 13.8 nm. The lattice parameter ‘a’ for the pristine sample was found to be 8.413 Å. It should be noted that the lattice parameter of the bulk cobalt ferrite is 8.391 Å (ICDD file no: 22-1086). A sharp increase in the crystalline nature of cobalt ferrite powder is observed as the firing temperature was increased to 800 ◦ C which is recorded as a decrease in the broadening of the peaks in the diffraction pattern. This clearly indicates that the grain size has increased with sintering. The sintered sample showed no traces of ␣-Fe2 O3 . The grain size was evaluated using Debye Scherer formula and found to be 40 nm. The details are given in Table 1. The TEM micrograph of the pristine sample (Fig. 2(a)) shows the formation of uniform spherical shaped ferrite nanoparticles. The mean particle size determined from TEM, 13 nm and is in agreement with the grain size obtained using XRD analysis (13.8 nm). The HRTEM image clearly depicts the crystal planes corresponding to cobalt ferrite (Fig. 2(b)). The EDS spectrum (Fig. 3) of pristine cobalt ferrite sample indicate that Co:Fe ratio is 1:2 and hence the loss of ferric or cobalt ion is ruled out. Scanning electron micrographs (SEM) of the samples are depicted in Fig. 4(a) and (b). The grain growth at higher sintering temperature for C8 is clearly evident from the SEM images. 3.2. Magnetic characterization Fig. 5(a) and (b) show the hysteresis curves for the CoFe2 O4 particles measured at room temperature and at low temperature (100 K). The loop parameters like saturation magnetization and coercivity are estimated and are given in Table 1. It can be seen that the samples are not superparamagnetic at room temperature. The hysteresis curves for the pristine samples (CP) at room temperature does not saturate even at an applied magnetic field of 13.5 kOe. High magnetocrystalline anisotropy associated with cobalt ferrite will be contributing to this type of behaviour [16]. The presence of antiferromagnetic impurities will also impede the sample from being saturated. The loop opened up at 100 K and the coercivity was found to be increasing at low temperatures. Magnetisation values of the samples were found to be lower than the corresponding bulk value of 80.8 emu/g. However an increase in magnetization (Ms ) was observed for the sample fired at 800 ◦ C because of its monophasic nature. The coercivity observed in the case of sintered sample was around 700 Oe at 300 K which is less than that of pristine sample. The variation of Hc with particle size can be explained on the basis of domain structure, critical diameter and the anisotropy of the crystal. The decrease in coercivity with increase in particle size is attributed to its change from the single domain characteristics to multidomain nature. The ratio of Mr /Ms throws light on the exchange interactions and magnetocrystalline anisotropy associated with the ferrite nanoparticles [17].

E.V. Gopalan et al. / Journal of Alloys and Compounds 485 (2009) 711–717

713

Table 1 Structural and magnetic parameters of cobalt ferrite nanoparticles. CoFe2 O4

Grain size (nm)

Porosity (%)

Ms (emu/g) at 300 K

Mr /Ms at 300 K

Mr /Ms at 100 K

Hc (Oe) at 300 K

Hc (Oe) at 100 K

CP C8

14 40

31 9

48 56

0.41 0.33

0.66 0.82

1080 700

8500 4500

According to Stoner Wohlfarth model [18], a theoretical value of Mr /Ms is 0.5 for non interacting uniaxial single domain particles with the easy axis being randomly oriented. Mr /Ms value observed is found to be 0.41 for the pristine samples. Hence CP exhibits uniaxial anisotropy. However the squareness ratio at room temperature decreases in the case of C8 indicating the formation of multidomain particles. There is an increase in Mr /Ms values measured at 100 K with firing temperature. This points towards the increase in magnetocrystalline anisotropy associated with decreased temperature. The maximum Mr /Ms value is found to be 0.82 indicating the enhanced contribution from cubic anisotropy (higher order terms of magnetocrystalline anisotropy) at lower temperature. Hence a large coercivity can be expected at lower temperatures.

Fig. 3. EDS pattern of pristine cobalt ferrite.

Fig. 2. (a) TEM and (b) HRTEM images of cobalt ferrite nanoparticles.

Fig. 4. SEM images of (a) CP and (b) C8.

714

E.V. Gopalan et al. / Journal of Alloys and Compounds 485 (2009) 711–717

3.3. Electrical characterization 3.3.1. dc conductivity studies The temperature dependence of dc conductivity in cobalt ferrite is studied for different grain sizes. The dc conductivity for the pristine sample is found to be 10−6 S/m while that of C8 is 10−5 S/m. The variation of conductivity with temperature shows the semiconducting nature of the ferrites (Fig. 6). One of the popular models for explaining the conductivity in ferrites is the electron hopping model [19]. In this model, it is presumed that the conductivity is mainly due to the hopping of electrons between Fe2+ and Fe3+ ions present at octahedral B sites. In addition to this the presence of Co3+ ions if any at B sites may also initiate hole hopping between Co3+ and Co2+ ions and thus may have small contribution to conductivity. Since Fe2+ and Co3+ ions have a strong preference for B sites, they can produce n-type and p-type conductivity respectively. Hence the conductivity will depend on the availability of charge carriers and their mobility. However, not much increase is obtained in our sample with sintering. This may be due to unavailability of charge carriers for conduction and also due to reduced mobility of the charge carriers. It must be noted here that the presence of a large number of traps on the grain boundaries will also adversely affect the conductivity. The activation energy for conduction calculated from the Arrhenius plot (inset in Fig. 6) is found to be very high 0.9 eV for pristine sample and 0.63 eV for the sintered one indicating the high resistive nature of cobalt ferrite. Large values of activation energy points towards a polaron conduction [20]. Anharmonic electron–phonon interactions play a significant role in the transport

Fig. 5. Hysteresis curves for (a) CP (b) C8 at 300 and 100 K.

Fig. 6.  dc vs temperature (log  dc vs 1000/T given as inset).

properties of the Fe–Co systems. Gruhn et al. reported that photoinduced optical second harmonic generation in Fe–Co metallic spin glasses [21]. 3.3.2. Dielectric properties The variation of dielectric permittivity with frequency is depicted in Fig. 7(a) and (b). It can be seen that the dielectric per-

Fig. 7. Dielectric dispersion in cobalt ferrite (a) CP and (b) C8.

E.V. Gopalan et al. / Journal of Alloys and Compounds 485 (2009) 711–717

mittivity exhibit an inverse dependence on frequency as reported in a number of ferrites. It decreases with increase in frequency and remains a constant at higher frequencies. In our frequency regime of measurements (100 kHz to 8 MHz), it is the interfacial and dipolar polarization which contribute to the overall dielectric properties of the sample [22]. According to Maxwell Wagner Theory, the dielectric polarization has its origin in the heterogeneous structure of ferrites with grains and grain boundaries [23,24]. Hence the effect of grain interfaces are more pronounced at lower frequencies where we observe larger values of ε . The space charge polarization occurring at the interfaces at lower frequencies can also contribute to the enhanced ε values at lower frequencies. As temperature increases, at lower frequencies the accumulation of charges on the grain boundaries increases which causes an increase in the interfacial polarization. Therefore the dielectric polarization increases causing a marked difference in ε with temperature at lower frequencies. Thus at lower frequencies the rapid increase in dielectric constant with temperature is mainly due to interfacial which is strongly temperature dependent. If one invokes the Rezlescue Model [25], the polarization can be explained based on conduction process which involves the electron hopping between Fe2+ and Fe3+ and hole hopping between Co2+ and Co3+ in the octahedral sites. At higher frequencies, the frequency of electron/hole exchange will not be able to follow the applied electric field thus resulting in a decrease in polarization. Consequently the dielectric permittivity attains a constant value at higher frequencies. It has been reported that the conductivity of a polycrystalline material in general increases with increasing particle size. Smaller grains imply smaller grain to grain surface contact area and therefore a reduced electron flow [26]. Thus the observed increase in dielectric permittivity of sintered sample can be attributed to the increase in particle size with respect to the as prepared sample. The variation of tan ı and ε are depicted in Fig. 8(a) and (b) and Fig. 9(a) and (b). It can be seen that although a relaxation is seen in the tan ı vs frequency curves, a corresponding relaxation is absent in the case of dielectric absorption. The variation of ac conductivity with frequency and temperature is depicted in Fig. 10(a) and (b). It is observed that the ac conductivity first increases with frequency, reaches a maximum and then decreases. As the frequency of the applied field increases, the hopping of charge carriers also increases thereby increasing the conductivity. But at higher frequencies, the hopping of charge carriers could not follow the applied field frequency and it lags behind the applied frequency resulting in a decrease in the ac conductivity values. The sintered samples exhibited higher conductivity than the pristine sample. When the grain size increases, intergranular porosity decreases, eventually resulting in an increase in conductivity as observed in the case of sintered sample. However heat treatment results in chemical homogeneity which may also affect the transport properties. The formation of Co3+ ions during sintering at higher temperatures can also influence the conductivity to a great extent. The results of ac conductivity can be explained on the basis of the assumption that the real ac electrical conductivity consists of two terms [27,20] ac (T ) = 1 (T ) + 2 (ω, T )

(1)

The first term is the temperature dependent dc conductivity which is related to the drift of electric charge carriers and follows an Arrhenius relation given by



1 (T ) = 0 exp −

Ea kB T

 (2)

715

Fig. 8. Variation of tan ı in cobalt ferrite (a) CP and (b) C8.

where Ea is the activation energy for electric conduction.  0 the pre-exponential factor. The activation energy can be found out by plotting ln  ac vs temperature as in Fig. 11 (from Eq. (2)). From the slope of the linear region, activation energy (Ea ) for electrical conduction was estimated and is found to be 0.273 eV and 0. 256 e V for CP and C8 respectively. The activation energy is found to be smaller than for dc conductivity and this may be due to the frequency and temperature dependence of ac conduction. The second term in Eq. (1) is the temperature and frequency dependent part which is related to the dielectric relaxation caused by the localized electric charge carriers. The power law is given by [28] 2 (ω, T ) = B(T )ωn (T )

(3)

where B is the parameter having the unit of conductivity and n is a dimensionless parameter. The log ω vs log  ac plots of CP and C8 are given in Fig. 12(a) and (b). It can be found that the parallel straight lines were obtained for CP while the slope increases with temperature in C8. For CP, the frequency exponent n is found to be almost temperature dependent and lies around 0.6 while for the other sample it varied between 0.5 and 0.98 for the temperature variation (Fig. 13). The temperature dependence of n is used to explain the mechanism of ac conduction in the ferrite [29,30,20]. In that case where

716

E.V. Gopalan et al. / Journal of Alloys and Compounds 485 (2009) 711–717

Fig. 9. Dielectric loss (ε ) in cobalt ferrite (a) CP and (b) C8. Fig. 10.  ac variation with frequency and temperature of (a) CP and (b) C8.

n is temperature independent, quantum mechanical tunneling is expected [31]. If n increases with the temperature, small polaron tunneling is the predominant mechanism [32]. By simple quantum mechanical tunneling model, the expression for n is given by the equation [33] n=1−

4 ln(1/ω)

(4)

samples sintered at higher temperatures. The formation of small polaron may be due to the defect levels or oxygen vacancies created during sintering. Thus the conduction mechanisms in the two samples are explained based on the quantum mechanical tunneling models. Quantum mechanical tunneling is found suitable for the pristine sample while small polaron conduction is the proposed mechanism in the sintered sample.

where  the relaxation time and ω the angular frequency This equation gives a temperature independent value for n. Hence quantum mechanical tunneling is the most probable mechanism in the pristine sample where we have observed a temperature independent behaviour of n. In the case of sintered sample we have got a temperature dependent n. A temperature dependent frequency component n can be obtained within the frame work of quantum mechanical tunneling model in the pair approximation by assuming that the charge carriers are non overlapping small polarons. In this case the frequency exponent n becomes [34] n=1−

4 ln(1/ω0 ) − WH /kB T

(5)

where kB is the Boltzmann’s constant, T the temperature. WH is the barrier height for infinite site separation,  0 the relaxation time and ω the angular frequency. The model predicts a temperature dependent frequency component and by Eq. (5) n increases with increasing temperature. This type of behaviour is obtained for

Fig. 11. ln  dc vs 1000/T.

E.V. Gopalan et al. / Journal of Alloys and Compounds 485 (2009) 711–717

717

as a function of temperature and frequency has been explained on the basis of Maxwell Wagner theory of interfacial polarization and also by Rezlescue model. Activation energies were calculated from the temperature dependence of ac conductivity and the frequency component n is estimated from the relation  ac = Bωn . The variation of n with temperature suggest a quantum mechanical tunneling conduction for pristine sample. The theory of small polaron tunneling which is a modification of quantum mechanical tunneling was found to be suitable for explaining the ac conductivity in samples sintered at higher temperature. Hence it can be concluded that the reduced number of hopping charge carriers and reduced mobility of the localized charge carriers in nanosized cobalt ferrite resulted in tunneling conduction. Acknowledgements EVG acknowledges Cochin University of Science and Technology for the Research Fellowship. Al-Omari would like to thank the Sultan Qaboos University for the support under Grant number IGSCI-PHYS-07-05. MRA acknowledges AICTE, Government of India (‘Centre for Ferrofluids’ File No.8023/RID/RPS-73/2004-05. Dated 29/03/2005) for the financial assistance. References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] Fig. 12. log  ac vs log ω (a) CP and (b) C8.

[11] [12] [13] [14] [15] [16] [17] [18] [19] [20]

[21] [22] [23] [24] [25] [26] [27] Fig. 13. Variation of n with temperature in CP and C8. [28] [29]

4. Conclusions The structural, magnetic and electrical properties of sol–gel derived cobalt ferrite nanoparticles are presented. The temperature variation of dc conductivity indicated the semiconducting behaviour of the ferrite. The variation in the dielectric properties

[30] [31] [32] [33] [34]

M.H. Kryder, MRS Bull. 21 (1996) 17–19. C.C. Berry, A.S.G. Curtis, J. Phys. D: Appl. Phys. 36 (2003) R198–R206. S.D. Bhame, P.A. Joy, J. Phys. D: Appl. Phys. 40 (2007) 3263–3267. S. Kückelhaus, A.C. Tedesco, D.M. Oliveira, P.C. Morais, G.R. Boaventura, Z.G.M. Lacava, J. App. Phys. 97 (2005), 10Q910-1–10Q910-3. S. Hariharan, J. Gass, Rev. Adv. Mater. Sci. 10 (2005) 398–402. Joong-Hee Nam, Won Ki Kim, Sang Jin Park, Phys. Status Solidi (a) 201 (2004) 1838–1841. C. Liu, Z. John Zhang, Chem. Mater. 13 (2001) 2092–2096. Q. Song, Z. John Zhang, J. Phys Chem. B 110 (2006) 11205–111205. A.J. Rondinone, A.C.S. Samia, Z.J. Zhang, J. Phys. Chem. B 103 (1999) 6876–6880. L.D. Tung, V. Kolesnichenko, D. Caruntu, N.H. Chou, C.J. O’Connor, L. Spinu, J. Appl. Phys. 93 (2003) 7325–7327. B. Vishwanathan, V.R.K. Moorthy, Ferrite Materials: Science and Technology, Springer Verlag, 1990, pp. 10–14. N. Sivakumar, A. Narayanasamy, K. Shinoda, C.N. Chinnasamy, B. Jeyadevan, J.M. Greneche, J. Appl. Phys. 102 (2007) 013916-1–013916-8. C.T. Hsieh, J.T. Lue, Phys. Lett. A 316 (2003) 329–335. A.J. Bosmann, C.C. Creve, Phys. Rev. 144 (1966) 763–770. G.H. Jonker, J. Phys. Chem. Solids 9 (1959) 165–175. B.G. Toksha, Sagar, E. Shrisath, S.M. Patange, K.M. Jadhav, Solid State Commun. 147 (2008) 479–483. C.N. Chinnasamy, B. Jeyadevan, K. Shinoda, K. Tohji, D.J. Djayaprawira, M. Takahashi, Appl. Phys. Lett. 83 (2003) 2862–2864. E.C. Stoner, E.P. Wohlfarth, Philos. Trans. R. Soc. A 240 (1948) 599–642. M. Jamet, W. Wernsdorfer, C. Thirion, V. Dupuis, P. Mélinon Alain Pérez, D. Mailly, Phys. Rev. B 69 (2004) 24401-1–24401-12. E. Veena Gopalan, K.A. Malini, S. Saravanan, D. Sakthi Kumar, M. Yasuhiko Yoshida, R. Anantharaman, J. Phys. D: Appl. Phys. 41 (2008) 185005-1–1850059. I.V. Gruhn, S. Kityk, Benet, Mater. Lett. 55 (2002) 158–164. Mathew George, S.S. Nair, K.A. Malini, P.A. Joy, M.R. Anantharaman, J. Phys. D: Appl. Phys. 40 (2007) 1593–1602. K.W. Wagner, Ann. Phys. 40 (1913) 817–855. C.G. Koops, Phys. Rev. 83 (1951) 121–124. N. Rezlescu, E. Rezlescu, Phys. Status Solidi (a) 23 (1974) 575–582. G.P. Kramar, Ya.I. Panova, V.V. Passynkov, Phys. Status Solidi (a) 76 (1983) 95–100. M.A. Ahmed, M.A. El Hiti, M.K. El Nimr, A.M. Amer, J. Magn. Magn. Mater. 152 (1996) 391–395. M.A. El Hiti, J. Phys. D: Appl. Phys. 29 (1996) 501–505. M.K. Fayek, M.F. Mostafa, F. Sayedahmed, S.S. AtaAllah, M. Kaiser, J. Magn. Magn. Mater. 210 (2000) 189–195. S.R. Elliot, Adv. Phys. 36 (1987) 135–217. I.G. Austin, N.F. Mott, Adv. Phys. 19 (1969) 41–102. T.M. Meaz, S.M. Attia, A.M. Abo El Ata, J. Magn. Magn. Mater. 257 (2003) 296–305. K.K. Som, B.K. Chaudari, Phys. Rev. B 41 (1990) 1581–1591. S. Bhattacharya, A. Ghosh, Phys. Rev. B 68 (2003) 224202-1–224202-5.