γ -Fe 2 O 3 nanoparticles dispersed in porous Vycor glass: A magnetically diluted integrated system Danielle Cangussu, Wallace Castro Nunes, Heberton Luis da Silva Corrêa, Waldemar Augusto de Almeida Macedo, Marcelo Knobel, Oswaldo Luiz Alves, Antônio Gomes Souza Filho, and Italo Odone Mazali Citation: Journal of Applied Physics 105, 013901 (2009); doi: 10.1063/1.3054173 View online: http://dx.doi.org/10.1063/1.3054173 View Table of Contents: http://scitation.aip.org/content/aip/journal/jap/105/1?ver=pdfcov Published by the AIP Publishing Articles you may be interested in Physical limits of pure superparamagnetic Fe3O4 nanoparticles for a local hyperthermia agent in nanomedicine Appl. Phys. Lett. 100, 092406 (2012); 10.1063/1.3689751 Formation and evolution of magnetic nanoparticles in borate glass simultaneously doped with Fe and Mn oxides J. Appl. Phys. 104, 103917 (2008); 10.1063/1.3021289 Magnetic properties of ( γ Fe 2 O 3 ) 80 Ag 20 nanocomposites prepared in reverse micelles J. Appl. Phys. 97, 10G101 (2005); 10.1063/1.1847331 Fragmentation of Fe 2 O 3 nanoparticles driven by a phase transition in a flame and their magnetic properties Appl. Phys. Lett. 83, 4842 (2003); 10.1063/1.1632534 Magnetic nanoparticles of Fe 2 O 3 synthesized by the pulsed wire evaporation method J. Appl. Phys. 93, 7196 (2003); 10.1063/1.1558234
[This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP: 143.106.201.17 On: Mon, 26 Jan 2015 20:41:49
JOURNAL OF APPLIED PHYSICS 105, 013901 共2009兲
␥-Fe2O3 nanoparticles dispersed in porous Vycor glass: A magnetically diluted integrated system Danielle Cangussu,1 Wallace Castro Nunes,2 Heberton Luis da Silva Corrêa,3 Waldemar Augusto de Almeida Macedo,3 Marcelo Knobel,4 Oswaldo Luiz Alves,1 Antônio Gomes Souza Filho,5 and Italo Odone Mazali1,a兲 1
Laboratório de Química do Estado Sólido, Instituto de Química, Universidade Estadual de Campinas-UNICAMP, Campinas, P.O. Box 6154, 13083-970, São Paulo, Brazil 2 Departamento de Física, Centro de Física da Matéria Condensada, Universidade de Lisboa, 1749-016, Lisboa, Portugal 3 Laboratório de Física Aplicada, Centro de Desenvolvimento da Tecnologia Nuclear, Belo Horizonte, Minas Gerais, 31270-901, Brazil 4 Departamento de Física da Matéria Condensada, Instituto de Física Gleb Wataghin, Universidade Estadual de Campinas-UNICAMP, Campinas, P.O. Box 6165, 13083-970, São Paulo, Brazil 5 Departamento de Física, Universidade Federal do Ceará, P.O. Box 6030, 60450-900, Fortaleza, Ceará, Brazil
共Received 28 April 2008; accepted 11 November 2008; published online 5 January 2009兲 An investigation of the effect of interparticle interaction and particle size distribution has been carried out on iron oxide nanoparticles dispersed into porous Vycor glass. ␥-Fe2O3 nanoparticles dispersed into monoliths of Vycor glass were obtained using impregnation-decomposition cycles through the single-source metallo-organic decomposition process. Magnetic properties were investigated by ac magnetic susceptibility measurements, as a function of temperature at different frequencies, by measuring zero-field-cooled and field-cooled magnetization curves and by constructing hysteresis loops at different temperatures. A log-normal size distribution of monodomain nanoparticles has been deduced from the analysis of the magnetization curves. 57Fe Mössbauer spectroscopy was also employed for investigating the magnetic behavior as a function of nanoparticle size. The systems exhibit typical superparamagnetic behaviors with a wide particle size distribution that can be changed without significantly affecting the interparticle interaction. The experimental data are discussed in terms of the evolution of the particle size distribution with the number of impregnation-decomposition cycles used for preparing the nanoparticles. © 2009 American Institute of Physics. 关DOI: 10.1063/1.3054173兴 I. INTRODUCTION
Research on magnetic nanoparticles 共NPs兲 has grown significantly in recent years, owing to their unique physical properties and their potential for applications in future nanobased technologies. The goal of nanotechnology is to exploit changes in a system induced by its low dimensionality, thus allowing the design of materials with superior performances compared with their bulk counterparts. However, for the proper use of such systems in devices, a better understanding and modeling of the relationship between their magnetic properties and their nanoscopic structural features is needed.1 The preparation of magnetic NPs dispersed into solid matrices is nowadays attracting much scientific and technological interest. The physical and chemical properties of NP are sometimes dramatically different from those of their bulk counterpart. These materials often exhibit size-induced effects, for example, below a critical size magnetic particles become single domain particles, as opposed to multidomain species in the bulk structure. In addition, small magnetic particles exhibit unique phenomena such as superparamagnetism and quantum tunneling of magnetization.2,3 Magnetic a兲
Author to whom correspondence should be addressed. Electronic mail:
[email protected].
0021-8979/2009/105共1兲/013901/7/$23.00
NPs have been dispersed into polymer matrices,4,5 silica gels,6–8 and grown into porous colloidal silica particles.9 Nanocomposites consisting of iron oxide NPs dispersed into polymer matrices10,11 and into porous Vycor glass 共PVG兲12–15 have also been prepared. The NPs have been prepared by using different methods where one of the main goals is to achieve control of the particle size distribution as well as the homogeneous dispersion of these NP into the matrices. In particular, an interesting synthesis technique for nanostructured materials has been labeled the single-source precursor 共SSP兲 approach and has been increasingly employed in the preparation of semiconductor nanocrystals and thin films.16–18 A major challenge for this research field is to achieve a fine-controlled synthesis of nanostructures and to assemble them into chemically integrated systems where their performance can be engineered. In this work, we investigated the magnetic properties of ␥-Fe2O3 NPs dispersed into porous Vycor glass synthesized by using the single-source metallo-organic decomposition process. Samples obtained with different impregnationdecomposition cycles 共IDCs兲 were prepared in order to modify the size distributions of the NP and to study their influence on the macroscopic magnetic properties of the resulting ␥-Fe2O3/porous glass nanocomposites.
105, 013901-1
© 2009 American Institute of Physics
[This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP: 143.106.201.17 On: Mon, 26 Jan 2015 20:41:49
013901-2
Cangussu et al.
J. Appl. Phys. 105, 013901 共2009兲
II. EXPERIMENTAL PROCEDURE A. Porous Vycor glass monolith
The PVG, code 7930, was purchased from Corning Glass Co. Glass plates were cut with dimensions of approximately 共10⫻ 10⫻ 1兲 mm3 and polished. The PVG was cleaned and activated as described by the manufacturer.19 PVG has an open porous structure of fundamentally pure silica with interconnecting pores ranging from 2 to 20 nm. The volume of pores represents approximately 28% of the total volume.20 The porous surface contains silanol groups, with acidic hydrogen atoms 共pKa ⬃ 9兲, which can be used as active sites for incorporation of several compounds,14–18 such as the metallo-organic compound employed in this paper. B. Synthesis of ␥-Fe2O3 NPs dispersed in PVG
As the precursor of iron oxide the single-source metalloorganic compound iron 共III兲 2-ethylhexanoate Fe关OOCCH共C2H5兲C4H9兴3 was employed and synthesized as described elsewhere.21 The ␥-Fe2O3 / PVG nanocomposites were prepared by impregnation of monoliths of PVG in a hexane solution of Fe关OOCCH共C2H5兲C4H9兴3 共25 ml, 1.0 mol l−1兲 for 8 h at room temperature, followed by a thermal treatment at 873 K for 8 h under a synthetic air flow. ␥-Fe2O3 / PVG with different crystallite sizes were obtained by employing different IDCs.16–18 We have prepared samples with 2, 4, 6, 8, and 10 IDCs. All the monoliths, independent of the number of IDC, were submitted to ten thermal treatments completing a total of 80 h at 873 K. C. Techniques
X-ray measurements were performed on a Shimadzu XRD-6000 powder diffractometer with Cu K␣ radiation, in the step scan mode 共0.01° with 10 s for counting per step兲. Transmission electron microscopy 共TEM兲 images were obtained using a Carl Zeiss CEM 902 microscope equipped with a Castaing–Henry energy filter spectrometer within the column and a Proscan Slow Scan charge-coupled device camera, both controlled by a microcomputer running the AnaluSis 3.0 system. ␥-Fe2O3 / PVG powdered samples were dispersed in de-ionized water by an ultrasonic process and a drop of the suspension was placed onto a carbon coated copper grid. The excess liquid was removed using a paper wick and the deposit was dried in the air prior to imaging. The zero-field-cooled 共ZFC兲 and field-cooled 共FC兲 magnetization measurements were performed in a superconducting quantum interference device magnetometer 共Quantum Design MPMS XL7兲. The dynamics of the magnetization was studied by measuring the temperature dependence of both in-phase 共⬘兲 and out-of-phase 共⬙兲 components of the ac susceptibility on a Physical Properties Measuring System 共Quantum Design PPMS兲 at different frequencies 共from = 10 Hz to 10 kHz兲 of the alternating ac field. 57Fe Mössbauer spectroscopy experiments were performed at low temperature using a conventional constant-acceleration spectrometer with a 57Co / Rh source and a closed-cycle cryostat.
FIG. 1. XRD patterns for 共a兲 Fe2O3 / PVG obtained with ten IDCs and 共b兲 free Fe2O3 共dotted line= pristine PVG兲.
III. RESULTS AND DISCUSSION A. Structural and morphological properties
The thermal decomposition of free Fe关OOCCH共C2H5兲C4H9兴3 共outside the porous glass matrix兲 under a synthetic air flow was performed at 873 K for 8 h and leads to the formation of ␣-Fe2O3 as evidenced by the peaks in the powder x-ray diffraction 共XRD兲 spectrum 关see Fig. 1共b兲兴 being indexed to those of ␣-Fe2O3.22 The average crystallite size of 47 nm is estimated from x-ray line broadening analysis using the standard Scherrer’s equation23 considering the 共104兲, 共110兲, 共024兲, and 共116兲 diffraction peaks. In comparison, the XRD pattern for the Fe2O3 / PVG composites as shown in Fig. 1共a兲 for the ten IDC sample is composed of a broad peak near 2 = 22°, which arises from the noncrystalline PVG matrix and three smaller broad peaks at higher angles that can be indexed to either Fe3O4 共magnetite24兲 or ␥-Fe2O3 共maghemite25兲 since their XRD patterns are essentially indistinguishable. Unfortunately, neither subsequent IR nor Mössbauer spectroscopy measurements on the nanocomposite samples were able to provide conclusive evidence as to the iron oxide phase. Due to the low concentration of the iron oxide and its dispersion within the matrix, the IR spectrum of the composite was identical to the IR spectrum of the pristine PVG. Similarly the Mössbauer measurements were unable to resolve the tetrahedral 关A兴 and octahedral 关B兴 sites for the Fe atoms. Nevertheless, it is more probable that the resulting iron oxide phase is ␥-Fe2O3 since maghemite can form from hematite in the 473–523 K temperature range26,27 well below the temperature of 873 K used to decompose the organic compounds in the starting materials. Moreover, the average crystallite diameter estimated to be 6 nm from the width of the 共311兲 peak for the ten IDC sample is much smaller than the diameter sizes of thermally stable crystallites of hematite. The maghemite to hematite phase transition 共hereafter indicated as ␥ → ␣ transition兲 is strongly dependent on crystallite size.27–29 ␥-phase has a surface energy smaller than ¯ values the ␥ → ␣ transition ␣-phase.27 With decreasing D
[This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP: 143.106.201.17 On: Mon, 26 Jan 2015 20:41:49
013901-3
Cangussu et al.
J. Appl. Phys. 105, 013901 共2009兲
FIG. 2. Cumulative mass gain of ␥-Fe2O3 / PVG as a function of the number of IDC.
rate will increase because of the increase in the specific surface area. Consequently the transformation rate will increase, therefore decreasing the temperature of the ␥ → ␣ transition. According to Multani,30 ␥-Fe2O3 is a thermodynamically stable phase up to diameters of 30 nm. For larger sizes, the thermodynamically stable phase is ␣-Fe2O3. The ␥ → ␣ transition temperature occurs in the 523–873 K temperature range, depending on the prior history.28 Gnanaprakash et al.29 obtained ␥-Fe2O3 powders through the coprecipitation ¯ ⬃ 5 nm and verified that after thermal treatmethod with D ment at 873 K for 1 h, the conversion of ␥ → ␣ transition was ¯ complete and the crystallites of ␣-Fe2O3 presented D ⬃ 35 nm. The stabilization of the ␣-Fe2O3 phase needs the coalescence of the ␥-Fe2O3 particles that is induced by the diffusion of cations during the thermal treatment, resulting in increasing the average crystallite size.27,28 The critical size ¯ ⬃ 15 nm. for the ␥ → ␣ transition to take place is around D ¯ Despite the reduction in D to increase the growth rate and ¯ values decrease the ␥ → ␣ transition temperature, smaller D inside the pores were found here. This result is significant in the sense that the final ␥-Fe2O3 / PVG was submitted to ten IDCs, completing a total of 80 h of thermal annealing at 873 K. The inhibition of crystallite growth and the consequent decrease in the ␥ → ␣ transition was associated with the presence of crystallites dispersed in the porous structures that prevent their growth through coalescence processes.29 The IDC methodology showed a linear mass increment inside the PVG 共Fig. 2兲 leading to ␥-Fe2O3 / PVG with different crystallite sizes 共see TEM data below兲 but keeping the average distance between the NP almost constant. The linear mass increase in Fig. 2 is the result of a mean diameter size increase in the spherelike shaped NP via layer-by-layer assembly. The IDC methodology has been shown to be a versatile bottom-up nanofabrication technique. This statement is confirmed by the TEM results and magnetic analysis in this paper and by our other manuscripts about CeO2,18 TiO2,17,31 CdS,16 and PbS 共Ref. 16兲 in PVG that show particle size increments as a function of the number of cycles. However,
FIG. 3. TEM bright field image of ␥-Fe2O3 / PVG obtained with 共a兲 two and 共b兲 eight IDCs. The insets show histograms of the particle size distributions.
the correlation between the mass increase and particle size increase with different numbers of the IDCs is not linear because the particle size distribution width increases with the number of IDC. For low diameter nanocrystals 共low numbers of IDCs兲 both the poor signal-to-noise ratio and the overlap with the broad peak from PVG make the analysis of crystallite size very difficult. TEM bright field images of the ␥-Fe2O3 / PVG obtained with two and eight IDCs are shown in Fig. 3. An estimate of nanocrystal size was performed by using the image tool package where the size of the nanocrystals is measured by a scale on the computer screen. Typical histograms are shown in the inset of Fig. 3. From TEM images spherelike shaped NP with large size distributions can be inferred and the av-
[This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP: 143.106.201.17 On: Mon, 26 Jan 2015 20:41:49
013901-4
J. Appl. Phys. 105, 013901 共2009兲
Cangussu et al.
cating the existence of a broad distribution of NP 共Fig. 4兲. It is also remarkable that the peak in the ZFC curves, related to the average blocking temperature, increases with an increasing number of IDC. Such increasing of blocking temperature ¯ 共Ref. 34兲 and/or increases in can be related to increases in D magnetic interactions between the particles 共due to particle concentration兲, as observed in other experimental and theoretical papers.34–36 In order to distinguish between these two possibilities, other experimental studies were carried out. The ZFC curve of a noninteracting system constituted by single domain NPs can be calculated considering the distribution of particle volumes 关f共V兲兴 as37,38 M ZFC共H,T兲 M CS H 1 = MS 3K ⍀ FIG. 4. Temperature dependence of the ZFC 共solid lines兲 and FC 共dotted lines兲 magnetization curves for ␥-Fe2O3 / PVG with different numbers of IDC.
erage nanocrystal sizes were found to be about 3.1⫾ 0.7 and 5.2⫾ 1.1 nm for two and eight IDCs, respectively. The crystallinity degree of the ␥-Fe2O3 nanocrystals is further confirmed through dark field imaging 共not shown here兲 from which it was possible to observe images from the diffracted electrons. Therefore, the TEM images show the presence of ␥-Fe2O3 NPs finely dispersed into porous Vycor glass, thus indicating that the applied synthesis procedure inhibits their growth through coalescence processes. B. Magnetic properties
Figure 4 shows the ZFC-FC curves for the samples with different numbers of IDCs. At higher temperatures ZFC and FC curves coincide, suggesting that the samples behave as typical superparamagnetic NP. The study of such system has been performed based on the Néel relaxation time,32
= 0eKV/kBT ,
共1兲
where the characteristic time constant 0 is normally taken in the range of 10−11 – 10−9 s, kB is the Boltzmann constant, K is the uniaxial anisotropy constant, and V is the particle volume. KV represents the energy barrier between two easy directions. According to Bean and Livingston,33 at a given observation time obs there is a critical temperature called the blocking temperature 共TB兲 above which the system behaves as a superparamagnet. On the contrary, below TB the systems are said to be blocked.33 However, in real systems there is a distribution of particle sizes, which gives rise to a distribution of TB. This distribution is characterized by an average value and width. Therefore, the ZFC curve of an assembly of monodomain NP exhibits a maximum at the average blocking temperature 共TB兲 as a consequence of the progressive unblocking of NP as the temperature is increased, while their magnetization decreases due to thermal fluctuations. The ZFC and FC curves converge at the temperature corresponding to the blocking of the biggest NP that herein will be called the irreversibility temperature 共Tirr兲. For our samples, the irreversibility temperatures are rather higher than the temperature where ZFC curves exhibit the maximum, indi-
+
冕 冕 ⬁
f共V兲dV
Vb共T兲
M CS H 1 3kBT ⍀
Vb共T兲
Vf共V兲dV,
共2兲
0
where ⍀ = 兰⬁0 Vf共V兲dV, Vb共T兲 = 25kBT / K, H is the applied field, M CS is the saturation magnetization of the NP, M S is the sample saturation magnetization, and T is the absolute temperature. The first term in Eq. 共2兲 comes from the blocked particle contributions, while the second comes from the superparamagnetic ones. The FC magnetization curve can be calculated by using an expression similar to Eq. 共2兲.37 We have calculated the ZFC and FC curves of the samples studied by considering a log-normal distribution of volumes M CS = 420 emu/ cm3 共bulk value兲 and using the value of K determined by a procedure described elsewhere36 共K = 5.9 ⫻ 105 erg/ cm3兲. Figure 4 shows, using solid and dotted lines, respectively, the calculated magnetization curves for ZFC and FC for different samples. The parameters of the log-normal distribution resulting in the best fits to the data are shown in Table I. As can be seen in Fig. 4, the ZFC curves are well described by superparamagnetic behavior, even for the sample with a large number of IDC; thus suggesting that particle agglomeration is not significant in such samples. We find a rather large particle size distribution width 共兲 共Table I兲 and, ¯ value increases with the increasing as expected, that the D number of IDC in agreement with the previous analysis by ¯ determined TEM. For ␥-Fe2O3 / PVG with eight IDCs, the D by TEM and magnetic analysis shows good agreement with both around 5.2 nm. Furthermore, the results of the simulation show that the main effect of the number of IDC occurs in the width of particle size distribution, with the exception of the ␥-Fe2O3 / PVG with ten IDCs 共Table I兲 where a more pronounced change in magnetic behavior occurs, as will be discussed later. In order to obtain more information on the magnetic properties of these systems, the field dependence of the magnetization up to 60 kOe at different temperatures was also investigated. The coercive fields obtained from M共H兲 curves measured at different temperatures are shown in Fig. 5. The inset in Fig. 5 shows the magnetization loop measured at 2 K for the ␥-Fe2O3 / PVG with ten IDCs, which displays a narrow hysteresis with a coercive field of 931 Oe.
[This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP: 143.106.201.17 On: Mon, 26 Jan 2015 20:41:49
013901-5
J. Appl. Phys. 105, 013901 共2009兲
Cangussu et al.
TABLE I. Best parameters of the size analysis distribution used in the ZFC and FC simulation of the experi¯ determined by mental data, irreversibility temperature 共Tirr兲, and the coercive field 共HC兲 values at 3 K and D TEM for the ␥-Fe2O3 / PVG samples studied. Best parameter of ZFC simulation
No. of IDC
V
¯ D 共nm兲
Tirr 共K兲
HC 共Oe兲 at 3 K
¯ D 共nm兲 by TEM
⌬T M / T M ⌬ log
2 4 6 8 10
¯ 0.70共10兲 0.69共5兲 0.62共5兲 0.58共5兲
¯ 4.7共3兲 4.8共2兲 5.2共2兲 6.1共2兲
¯ 180共10兲 170共10兲 125共10兲 150共10兲
¯ 746共5兲 843共5兲 873共5兲 848共5兲
3.1共7兲 ¯ ¯ 5.2共11兲 ¯
¯ 0.090共3兲 0.087共3兲 0.083共3兲 0.082共3兲
where TB is the average blocking temperature. It can be inferred that the temperature behavior of the samples studied follows Eq. 共3兲, at least over the low temperature range. These results further confirm the existence of single-particle blocking behavior in these systems. An increase in the coercive field with increasing particle size of a single-magnetic domains is expected, starting from zero when the particle size becomes so small that the thermal energy is sufficient to flip the magnetization direction over the energy barrier associated with the anisotropy of the particle, and reaching a maximum value 共2K / M S兲 when the particle size approaches the single-domain limit.38 Table I shows that, as the number of IDC increases from four to eight, a gradual increase in the coercive field at low temperature is observed. Further increases in the IDC induce a decline in the coercive value. This result can be related to a considerable demagnetizing role played by dipolar interactions.36 Dipolar interaction can make a collection of magnetic moments of individual par-
ticles behave like a collective magnetic system. Such collective magnetic systems have some magnetic properties similar to those observed in noninteracting superparamagnetic materials but with an effective volume and anisotropy determined by interplay between the individual particles and their interaction.40 Therefore, although we have observed, for the sample with ten ICDs, a good agreement between the ZFC and FC experimental data and that calculated based on the framework of noninteracting superparamagnetic approach, ¯ obtained from this procedure might be rather the value of D larger than the actual value. Magnetic susceptibility is strongly sensitive to interactions between particles and thus can be used to better understand the magnetic behavior of the ␥-Fe2O3 / PVG composites prepared with ten IDCs. Figure 6 shows the thermal dependence of the in-phase 共⬘兲 and out-of-phase 共⬙兲 ac magnetic susceptibility measured for different frequencies under a zero dc field and an ac field of 4 Oe for ␥-Fe2O3 / PVG with eight IDCs. For both ⬘ and ⬙, the peak position shifts toward higher temperatures as the frequency is increased. Curves with similar profiles were measured for the other samples studied 共data not shown兲. The calculated relaxation time 共 = 1 / 兲 at TB of the samples follows the characteristics of a thermally activated mechanism given by
FIG. 5. Coercive field vs T1/2 for ␥-Fe2O3 / PVG with four, eight, and ten IDCs. The inset shows the hysteresis loop of ␥-Fe2O3 / PVG with ten IDCs at T = 2 K.
FIG. 6. Thermal dependence of in-phase 共⬘兲 and out-of-phase 共⬙兲 ac magnetic susceptibility measured for different frequencies under a zero dc field and an ac field of 4 Oe for ␥-Fe2O3 / PVG with eight IDCs.
The temperature dependences of coercive field HC for systems of identical noninteracting particles with random anisotropic axes follow the well known relation,39 HC = 0.96
冋 冉 冊册
T K 1− MS TB
1/2
,
共3兲
[This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP: 143.106.201.17 On: Mon, 26 Jan 2015 20:41:49
013901-6
J. Appl. Phys. 105, 013901 共2009兲
Cangussu et al.
FIG. 7. The log共1 / 兲 vs 1 / TB plot for a wide range of measured frequencies of the ␥-Fe2O3 / PVG samples studied.
the Arrhenius law 关Eq. 共1兲兴, with the pre-exponential factor values being in the range from 0 = 1.6⫻ 10−14 to 5 ⫻ 10−13 s and the energy barriers for the reversal of the magnetization in the range of KV / kB = 950– 1800 K 共Fig. 7兲. Such magnitudes of 0 and KV / kB are in line with observed values for other NP system presenting interactions between particles.35 A useful parameter that can be obtained from ac susceptibility is the measurement of the temperature shift in TB, usually obtained from ⌬TB / TB per decade of frequency , i.e., ⌬T M / T M ⌬ log . This parameter has been used as a criteria to distinguish a spin-glass-like transition from a singleparticle blocking effect. For our samples this parameter is in the range of 0.082–0.090 共Table I兲, which falls within the range observed for interacting magnetic NP.35,41 Therefore, although the dipolar interactions play an important role in the magnetic behavior of the ␥-Fe2O3 / PVG with ten IDCs, the ac susceptibility results suggest that the ZFC and FC curves of such samples are dominated by single-blocking of NP.
FIG. 9. 57Fe Mössbauer spectra of ␥-Fe2O3 / PVG with six IDCs as a function of temperature.
Moreover, these results support the analysis of the ZFC and FC curves for most of our samples based on the superparamagnetic approach, indicating nucleation and coarsening of ␥-Fe2O3 NP for increasing numbers of IDCs. Figure 8 shows the 57Fe Mössbauer spectra at 23 K for samples with different numbers of IDC. The measurements do not result in any well resolved spectra, but show the superposition of a superparamagnetic component 共broad doublet兲 with a broad magnetically ordered component, behavior typical of superparamagnetic ␥-Fe2O3 NPs distributed in size.42 The broad doublets, attributed to the smallest particles in the samples, dominate the spectra for two and four IDCs, and the broad six-line component, due to larger magnetically ordered particles, is dominant starting from six IDCs. The superparamagnetic behavior of our samples is further illustrated in Fig. 9 where Mössbauer spectra for the sample ␥-Fe2O3 / PVG prepared with six IDCs at 23, 35, and 50 K are shown. The fast collapse in the magnetic order 共decrease in the broad six-line components兲 of the NP with a relatively small increase in the temperature is quite evident.
IV. CONCLUSION
FIG. 8. 57Fe Mössbauer spectra at 23 K for ␥-Fe2O3 / PVG with different numbers of IDC.
In conclusion, we have shown that the SSP approach is an appropriate method for preparing ␥-Fe2O3 NP dispersed in porous Vycor glass with suitable superparamagnetic properties. All experimental results suggest that the particle size distribution is the parameter that most significantly changes as a function of different numbers of IDC. The magnetic properties of the samples with different particle size distributions follow well-established models for noninteracting particle systems. Although the interparticle interactions start to be important to the magnetic behavior of the nanocomposite after a large number of IDC, progressive blocking of individual NP is observed for all samples studied.
[This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP: 143.106.201.17 On: Mon, 26 Jan 2015 20:41:49
013901-7
ACKNOWLEDGMENTS
The authors are grateful to FAPESP, CNPq, and CAPES for financial support and to Professor C. H. Collins 共IQUNICAMP, Brazil兲 for English revision. 1
J. L. Dormann, D. Fiorani, and E. Tronc, Adv. Chem. Phys. 98, 283 共1997兲. E. M. Chudnovsky and L. Gunther, Phys. Rev. Lett. 60, 661 共1988兲. 3 J. K. Vassiliou, V. Mehrotra, M. W. Russell, E. P. Giannelis, R. D. McMichael, R. D. Shull, and R. F. Ziolo, J. Appl. Phys. 73, 5109 共1993兲. 4 D. Rabelo, E. C. D. Lima, A. C. Reis, W. C. Nunes, M. A. Novak, V. K. Garg, A. C. Oliveira, and P. C. Morais, Nano Lett. 1, 105 共2001兲. 5 C. R. Mayer, V. Cabuil, T. Lalot, and R. Thouvennot, Adv. Mater. 共Weinheim, Ger.兲 12, 417 共2000兲. 6 C. Chanéac, E. Tronc, and J. P. Jolivet, J. Mater. Chem. 6, 1905 共1996兲. 7 C. Cannas, D. Gatteschi, A. Musino, G. Piccaluga, and C. Sangregorio, J. Phys. Chem. B 102, 7721 共1998兲. 8 S. Solinas, G. Piccaluga, A. P. Morales, and C. J. Serrna, Acta Mater. 49, 2805 共2001兲. 9 A. Bourlinos, A. Simpoulos, D. Petridis, H. Okumura, and G. Hadjipanayis, Adv. Mater. 共Weinheim, Ger.兲 13, 289 共2001兲. 10 R. F. Ziolo, E. P. Giannelis, B. A. Weinstein, M. P. O’Horo, B. N. Ganguly, V. Mehrotra, M. W. Russel, and D. R. Huffman, Science 257, 219 共1992兲. 11 P. P. Vaishnava, U. Senaratne, E. C. Buc, R. Naik, V. M. Naik, G. M. Tsoi, and L. E. Wenger, Phys. Rev. B 76, 024413 共2007兲. 12 M. Zayat, F. Monte, M. P. Morales, G. Rosa, H. Guerrero, C. J. Serna, and D. Levy, Adv. Mater. 共Weinheim, Ger.兲 15, 1809 共2003兲. 13 M. Iwamoto, T. Abe, and Y. Tachibana, J. Mol. Catal. A: Chem. 155, 143 共2000兲. 14 D. Sunil, H. D. Gafney, M. H. Rafailovich, J. Sokolov, R. J. Gambino, and D.M. Huang, J. Non-Cryst. Solids 319, 154 共2003兲. 15 M. C. Schnitzler, A. S. Mangrich, W. A. A. Macedo, J. D. Ardisson, and A. J. G. Zarbin, Inorg. Chem. 45, 10642 共2006兲. 16 I. O. Mazali, R. Romano, and O. L. Alves, Thin Solid Films 495, 64 共2006兲. 17 I. O. Mazali, A. G. Souza Filho, B. C. Viana, J. Mendes Filho, and O. L. Alves, J. Nanopart. Res. 8, 141 共2006兲. 18 I. O. Mazali, B. C. Viana, O. L. Alves, J. Mendes Filho, and A. G. Souza 2
J. Appl. Phys. 105, 013901 共2009兲
Cangussu et al.
Filho, J. Phys. Chem. Solids 68, 622 共2007兲. Available in www.corning.com/docs/specialtymaterials/pisheets/ Vycor%207930.pdf. 20 B. M. Volf, Technical Approach to Glass 共Elsevier, Amsterdam, 1990兲. 21 R. W. Vest and S. Singaram, Mater. Res. Soc. Symp. Proc. 60, 35 共1986兲. 22 JCPDS File No. 33-664. 23 R. Jenkins and R. L. Snyder, Introduction to X-ray Powder Diffractometry 共Wiley, London, 1966兲. 24 JCPDS File No. 19-629. 25 JCPDS File No. 39-1346. 26 I. Chamritski and G. Burns, J. Phys. Chem. B 109, 4965 共2005兲. 27 I. V. Chernyshova, M. F. Hochela, and A. S. Madden, Phys. Chem. Chem. Phys. 9, 1736 共2007兲. 28 T. Belin, N. Millot, N. Bovet, and M. Gailhanou, J. Solid State Chem. 180, 2377 共2007兲. 29 G. Gnanaprakash, S. Ayyappan, T. Jayakumar, J. Philip, and B. Raj, Nanotechnology 17, 5851 共2006兲. 30 M. S. Multani, Condens. Matter News 1, 25 共1991兲. 31 I. O. Mazali and O. L. Alves, J. Phys. Chem. Solids 66, 37 共2005兲. 32 L. Néel, Ann. Geophys. 共C.N.R.S.兲 5, 99 共1949兲. 33 C. Bean and J. D. Livingston, J. Appl. Phys. 30, S120 共1959兲. 34 J. M. Vargas, L. M. Socolovsky, M. Knobel, and D. Zanchet, Nanotechnology 16, S285 共2005兲. 35 J. L. Dormann, D. Fiorani, and E. Tronc, J. Magn. Magn. Mater. 202, 251 共1999兲. 36 W.C. Nunes, F. Cebollada, M. Knobel, and D. Zanchet, J. Appl. Phys. 99, 08N705 共2006兲. 37 M. Respaud, J. M. Broto, H. Rakoto, A. R. Fert, L. Thomas, B. Bárbara, M. Verelst, E. Snoeck, P. Lecante, A. Mosset, J. Osuna, T. O. Ely, C. Amiens, and B. Chaudret, Phys. Rev. B 57, 2925 共1998兲. 38 M. F. Hansen and S. Morup, J. Magn. Magn. Mater. 203, 214 共1999兲. 39 E. F. Kneller and F. E. Luborsky, J. Appl. Phys. 34, 656 共1963兲. 40 W. C. Nunes, L. M. Socolovsky, J. C. Denardin, F. Cebollada, A. L. Brandl, and M. Knobel, Phys. Rev. B 72, 212413 共2005兲. 41 J. A. Mydosh, Spin Glass: An Experimental Introduction 共Taylor & Francis, London, 1993兲. 42 E. Tronc, A. Ezzir, R. Cherkaoui, C. Chaneac, M. Nogues, H. Kachkachi, D. Fiorani, A. M. Testa, J. M. Greneche, and J. P. Jolivet, J. Magn. Magn. Mater. 221, 63 共2000兲. 19
[This article is copyrighted as indicated in the article. Reuse of AIP content is subject to the terms at: http://scitation.aip.org/termsconditions. Downloaded to ] IP: 143.106.201.17 On: Mon, 26 Jan 2015 20:41:49
