Abstract

We succeeded in the preparation of CoFe2O4/CeO2 nanocomposites with very high specific surface area (up to 264 g/m2). First, highly crystalline nanoparticles (NPs) of CoFe2O4 (4.7 nm) were prepared by hydrothermal method in water-alcohol-oleic acid system. The oleate surface coating was subsequently modified by ligand exchange to citrate. Then the NPs were embedded in CeO2 using heterogeneous precipitation from diluted Ce3+ sulphate solution. Dried samples were characterized by Powder X-Ray Diffraction, Energy Dispersive X-Ray Analysis, Scanning and Transmission Electron Microscopy, Mössbauer Spectroscopy, and Brunauer-Emmett-Teller method. Moreover, detailed investigation of magnetic properties of the bare NPs and final composite was carried out. We observed homogeneous embedding of the magnetic NPs into the CeO2 without significant change of their size and magnetic properties. We have thus demonstrated that the proposed synthesis method is suitable for preparation of extremely fine CeO2 nanopowders and their nanocomposites with NPs. The morphology and magnetic nature of the obtained nanocomposites make them a promising candidate for magnetoresponsive catalysis.

1. Introduction

Nanocomposites composed of nanoparticles (NPs) incorporated in functional matrix, which combine more than one functional physical and/or chemical property, are suggested as promising candidates for applications in various fields such as biomedicine, catalysis, and data storage [14]. In the last decade, several nanomaterials with magnetic and catalytic functionalities have been developed. Catalytic materials in form of NPs exhibit increase of the catalytic activity and number of active sites accessible for the reaction in comparison to the bulk counterparts due to the large surface-to-volume ratio and enhanced reactivity of the molecular orbitals of the surface species [5]. By incorporation of magnetically active components, the nanomaterials could be used as magnetically separable catalysts for various applications requiring effective and rapid separation of the active substance from liquid phase [59]. This is very beneficial for the solution of the environmental problems such as the wastewater treatment [9].

Cerium oxide, CeO2, is one of the most important catalytic materials [1012]. It exhibits relatively high catalytic activity under UV and visible light irradiation, sufficient chemical stability, and nontoxicity. Also its two special features—high reactivity mediated by the redox couple and exceptional oxygen storage capacity [13]—make this material unique. CeO2-containing materials are often used in industry [12] as components of commercial catalysts for reduction of the CO, hydrocarbons and emissions from gasoline engines [14], and fuel cell technologies [15], in wastewater treatment [16], and in cadmium [17] or lead [18] removal. Creation of the nanocomposite systems comprising the CeO2 and magnetic NPs could extend possible applications of this material [19]. For example, the CuFe2O4/CeO2 nanocomposite is a promising material in energy storage [20]; Fe3O4/CeO2 was identified as magnetically separable catalyst [21, 22].

The use of magnetite as a magnetic carrier may be problematic by its instability against topotactic oxidation to maghemite [23, 24] and transformation to hematite at elevated temperatures [25, 26], which are often unavoidable for proper functionality of the catalyst. Therefore, CoFe2O4 is a suitable magnetic material due to its high magnetic anisotropy, relatively high saturation magnetization (80 Am2 kg−1 [27]) and coercivity (up to 2 T at low temperatures for NPs in the matrix [28]), good mechanical hardness, and excellent chemical stability against oxidation [28, 29]. Use of the CoFe2O4 as catalyst [30, 31] or magnetic carrier for magnetically separable catalysts based on activated carbon [9], TiO2 [32], or ZnO [33] was reported recently. CoFe2O4/CeO2 nanocomposite should be a very promising candidate as a magnetic separable catalyst.

Up to date, only little work has been done on the preparation and characterization of the CoFe2O4-CeO2 system. Wetchakun et al. report on successful preparation of the CoFe2O4/CeO2 composite by coupling homogeneous precipitation with hydrothermal method [34]. The particle size and specific surface area were controlled by CoFe2O4 to CeO2 ratio.

Recently, we have introduced a general approach for the preparation of multifunctional nanocomposites of monodisperse NPs embedded in oxide matrices [35]. This method enabled homogenous embedding of the CoFe2O4 NPs into various oxide matrices without change of the NP diameter and size distribution and their magnetic response. Unfortunately, this method was not successful in the case of the CeO2 matrix due to the problematic hydrolysis of the cerium alkoxides. Therefore, we have modified this approach by excluding the alkoxides and successfully prepared the CoFe2O4/CeO2 nanocomposites by embedding the NPs with the help of heterogeneous precipitation of the CeO2 matrix at basic pH.

Using the modified approach, we have prepared a series of CoFe2O4/CeO2 nanocomposite samples varying the conditions of the incorporation of the NPs into the matrix. We have paid a special attention to identify the exact specific preparation conditions leading to the minimal change of the embedded CoFe2O4 NPs properties together with the best potential catalytic activity of the CeO2 matrix. The successful embedding of the CoFe2O4 NPs into the CeO2 matrix was confirmed for all samples. Magnetic response of the matrix-free modified sample and nanocomposites was compared with that of the bare CoFe2O4 NPs, which enabled classifying any changes of the NP properties after their surface modification and insertion into the matrix. The role of the preparation conditions in the structure and magnetic response of the individual CoFe2O4 NPs has been studied. Also the structure, morphology, and specific surface areas of the nanocomposites were examined carefully. The proposed preparation route can be easily scaled up and modified for systems with more components or doped materials.

2. Materials and Methods

2.1. Preparation

We designed a two-step synthesis for preparation of the CoFe2O4/CeO2 nanocomposites. First, the oleic acid coated CoFe2O4 NPs were obtained by the hydrothermal synthesis in the presence of oleic acid as was reported previously [29, 35]. The reference as-prepared matrix-free sample was obtained by drying the ferrofluid at 75°C and labeled as CoFe_free.

The hydrophobic NPs (dispersed in hexane) are unsuitable for the reaction in the aqueous media, which is essential for their incorporation into the CeO2 matrix. Hence the surface modification of the NPs was done by citric acid. After that, the NPs were incorporated into the matrix. Preparation conditions, such as the final pH or method of its adjustment, were varied for each sample in order to reach optimal morphology of the nanocomposite.

2.1.1. Citrate Surface Modification of the CoFe2O4 NPs

Surface modification of the CoFe2O4 NPs was inspired by methods described in [3638]. First, 100 mL of acetone was added to the 10 mmol CoFe2O4-hexane solution. The precipitate was magnetically separated and the liquid was decanted afterwards—this procedure was repeated two times. The separated precipitate was dispersed in 100 mL of water, and then 1.5 mL of concentrated nitric acid together with 40 mmol of citric acid was added. The reaction mixture was heated at 80°C under simultaneous sonication in ultrasonic bath (at 80 W transferred power) for 90 min. After cooling the mixture to room temperature (RT), the modified CoFe2O4 NPs were purified by combination of resuspension in acetone (3x) and diethylether (2x) and magnetic separation of the NPs. The purified NPs were dispersed in 100 mL of water and sonicated for 15 h. The final water dispersion was stable for more than 6 months at RT. We also tested pH-influence on stability of the citrate modified CoFe2O4 NPs in water solution. The sample was stable in the pH range of 2.4–13.0 for few hours. Amount of the citric acid coating was determined gravimetrically by Thermogravimetry-Differential Thermal Analysis (TG-DTA). The reference matrix-free modified sample, labeled as CoFe_mod, was obtained by drying the ferrofluid at 45°C.

2.1.2. Embedding of the (CoFe2O4) NPs into the (CeO2) Matrix

General procedure starts with 5 mmol solution of cerium sulphate in 150 mL of distilled water with 2 mL of concentrated sulfuric acid (solution A). This solution was diluted to 750 mL with distilled water and then pH was set to 2.6 by addition of 0.5 M sodium hydroxide solution. In the next step, solution of 0.5 mmol citrate modified CoFe2O4 NPs was added to 250 mL of water (solution B). Solution B was stirred and sonicated (at 200 W transferred power) for 30 min. Afterwards, pH was set to 3.5 by addition of diluted sulfuric acid (1 : 5). Solutions A and B were mixed together and stirred for 15 min under continuous sonication (at 20 W transferred power, 10% duty cycle, 5 min).

Furthermore, 0.5 M sodium hydroxide solution with the final pH value was added and method of pH adjustment varied for each prepared sample as summarized in Table 1. During neutralization, the mixture became turbid at pH around 5. After targeting the final pH value, the samples were heated at 60°C under intensive overnight stirring. pH slightly decreased during the heating; thus it was necessary to add small amount of sodium hydroxide for several times to keep the final pH value constant.

After cooling to RT the mixtures were decanted and then purified according to the following procedure: samples were redispersed in 750 mL of water by sonication, then decanted, and centrifuged. Afterwards, the samples were dispersed in 60 mL of water and centrifuged for three more times. The same procedure was performed with ethanol (2x) and finally with diethylether (1x). The samples kept under vacuum were finally heated at a rate of 0.2°C/min to 200°C and kept at final temperature for 2 h. The labels of the samples and details of preparation conditions are summarized in Table 1.

2.2. Sample Characterization

All samples were characterized using the RT Powder X-Ray Diffraction (PXRD) performed on the Philips X’Pert PRO MPD X-ray diffraction system equipped with the X’Celerator detector working with the Cu-Kα (λ = 1.5418 Å; matrix-free samples) or Co-Kα (λ = 1.7889 Å; nanocomposite samples) radiation. The high-temperature measurements were done in the same equipment (with the Cu-Kα radiation) accompanied by the high-temperature chamber in the temperature range of 200–500°C. The particle size, , was determined using the Rietveld refinement procedure implemented within the FullProf software [39].

The content of citric acid was determined gravimetrically by TG-DTA using a TG-DTA device by SETARAM. The Transmission Electron Microscopy (TEM) and High-Resolution TEM (HR-TEM) using TEM FEI Tecnai G2 Spirit and HR-TEM JEOL JEM 3010 (matrix-free samples, v02 and v06 nanocomposite samples) were performed in order to study the morphology of the prepared nanocomposites and possible determination of the particle size of the CoFe2O4 NPs. TEM analysis was supplemented with selected area electron diffraction (TEM/SAED) and energy dispersive analysis of X-rays (TEM/EDX). The experimental TEM/SAED patterns were processed by program ProcessDiffraction [40] and compared with the theoretical PXRD patterns of CoFe2O4 and CeO2 using Crystallography Open Database [41] and program PowderCell [42] as described elsewhere [43].

The chemical composition of the samples was verified by Energy Dispersive X-Ray Analysis (EDX) using the Bruker Quantax EDS detector implemented within the Scanning Electron Microscope (SEM) Tescan Mira 3 LMH. The specific surface area of the samples was measured using Brunauer-Emmett-Teller (BET) nitrogen adsorption using the Quantachrome Instrument NOVA 4200e apparatus. Before the measurements, the samples were degassed at 100°C for 24 h.

The Mössbauer Spectroscopy measurement was done in transmission mode with 57Co diffused in the Rh matrix as the source moving with constant acceleration. The spectrometer (Wissel, Germany) was calibrated by standard α-Fe foil and the isomer shift was related to this standard at 293 K. The resulting parameters were determined in the NORMOS program.

The magnetic measurements were carried out using the SQUID magnetometer (MPMS7XL, Quantum Design). The zero-field-cooled (ZFC) and field-cooled (FC) curves were recorded as follows: at first, the sample was cooled down to 10 K; then the magnetic field of 0.01 T was applied and the temperature dependence of the magnetization was measured up to 400 K (ZFC curve). Afterwards, the sample was cooled down to 10 K in the applied field and the FC curve was obtained. The magnetization isotherms (field dependence of the magnetization, curves) were measured at 10, 300, and 400 K up to the 7 T in both polarities. Distributions of magnetic moment values, , of the CoFe2O4 NPs were determined by analyzing the unhysteretic magnetization isotherms in the MINORIM software [44].

3. Results and Discussion

3.1. Characterization of the Structure, Phase Composition, and Morphology

The phase composition of the samples was studied using PXRD. Also the particle size of the CoFe2O4 (for the matrix-free NPs in the CoFe_free and CoFe_mod samples only) and CeO2 was determined. The diffraction patterns of the CoFe_free and CoFe_mod samples exhibit only reflections corresponding to the CoFe2O4 spinel phase (PDF database, card number 00-022-1086), with the lattice parameter  Å for both samples (Figure 1(a)). The crystalline size (size of coherently diffracted domain) determined by Rietveld refinement is  nm and  nm for the CoFe_free and CoFe_mod samples, respectively. Slight decrease of the in comparison with can be explained as lowering of the NP crystallinity in the surface layer of the NP (a shell) and hence reduction of the size of coherently diffracting domain after the citric acid treatment.

In the case of the CoFe2O4/CeO2 nanocomposite samples, the reflections corresponding to the CoFe2O4 phase possess very low intensity (comparable with the background) and most of them are completely overlapped by the CeO2 reflections (Figure 1(b)). Therefore, neither the presence of the CoFe2O4 NPs nor the NP diameters is determinable by the PXRD. It has been observed that the CeO2 matrix is nanocrystalline in all nanocomposite samples (Figure 1(b)); resulting values of the crystallite sizes, , and lattice parameters, , are summarized in Table 1. It is obvious that decreases with increasing final pH. The determined values are in good agreement with the bulk tabulated value,  Å.

In order to study the temperature stability of the CoFe2O4/CeO2 nanocomposites, in situ PXRD study of the v04 sample was carried out up to 500°C. The typical patterns for three different annealing temperatures, , are shown in Figure 1(c). No additional phase formed during annealing process. increases with increasing from () nm at 25°C to () nm at 500°C (see inset of Figure 1(c)). The observed increase of diameter of the CeO2 nanocrystallites is consistent with decrease of the width of the CeO2 reflections with increasing , as is shown in Figure 1(c). The obtained results indicate that the nanocomposite is thermally stable up to the 500°C.

The EDX measurements confirmed the expected elemental composition of the samples; results are summarized in Table S1 in the Supplementary Material available online at http://dx.doi.org/10.1155/2016/7091241. Fe and Co contents for the CoFe_free sample correspond within the error to the Fe/Co ratio expected for the stoichiometric CoFe2O4. However, the CoFe_mod sample contains higher amount of the Fe than the CoFe_free sample with the Fe/Co ratio equal to 3.7 instead of expected stoichiometric 2. This implies that some Co is missing in the CoFe_mod sample. One of the possible explanations is modification by citric acid. This can result in the formation of a Co-depleted shell.

If we assume the limiting case in which the modified NPs have the core-shell structure, with the stoichiometric CoFe2O4 crystalline core and disordered Co-depleted shell, we can easily calculate the volume attributed to the shell. Let us assume that both the CoFe_free and CoFe_mod samples have the same apparent size of the NPs, , and apparent volume, . For the NPs in the CoFe_free sample, corresponds to the CoFe2O4 from 100%. If we compare the Fe/Co ratio for the CoFe_free and CoFe_mod samples (2.4 versus 3.7), then the NPs in the CoFe_mod sample contain 66% of the CoFe2O4. If this CoFe2O4 forms the NP core, its diameter corresponds to 82% of the CoFe_free diameter. This agrees well with the results of the PXRD, where corresponds only to the 94% of the value. Hence the partial removal of the Co from the NP shell is reasonable explanation of the changed Fe/Co ratio after the citric acid treatment.

In the case of the nanocomposite samples, the Fe/Co ratios are within the error close to the value for the CoFe_mod sample (3.6–4.2, Table S1), indicating that the elemental composition of the NPs was not affected by the incorporation of the NPs into the matrix.

The NPs were directly observed by TEM. In the case of matrix-free CoFe_free and CoFe_mod samples, the individual NPs can be clearly distinguishable (Figures 2(a) and 2(b)) and, as a result, the particle size distributions of these two samples were obtained (insets of the Figures 2(a) and 2(b)). The mean physical particle size determined from the particle size histograms is = () nm and = () nm for the CoFe_free and CoFe_mod samples, respectively. It can be observed that is slightly larger than as a consequence of the surface modification by citric acid. Chemical treatment connected with the surface modification probably causes partial dissolution of the NPs resulting in the observed difference between NP sizes of both samples. The morphology of the all nanocomposite samples observed by TEM and HR-TEM is comparable; representative TEM micrographs of the v02 and v06 samples are shown in Figures 2(c) and 2(d) (micrographs for other nanocomposite samples are given in Figure S1). In the nanocomposite samples, the CoFe2O4 NPs are surrounded by smaller CeO2 NPs which makes their observation and hence determination of the particle size distributions and mean rather difficult and imprecise.

TEM/SAED experiments (Figure 3) confirmed the results of PXRD: the matrix-free NPs (Figures 3(a) and 3(b)) correspond to the CoFe2O4 spinel structure (Crystallography Open Database [41], record ID = 5910063). In the case of CoFe2O4/CeO2 nanocomposite samples (Figures 3(c) and 3(d)), the diffractograms are dominated by the broad diffractions of CeO2 cubic phase (Crystallography Open Database [41], record ID = 9009008). The broad diffractions in Figures 3(c) and 3(d) are associated with the smaller size of CeO2 crystallites (width of the peak is inversely proportional to the crystallite size). Weak CoFe2O4 diffractions form just weak shoulders of dominating CeO2 peaks, which is due to the low concentration of the CoFe2O4 in the composite.

The BET measurements of nanocomposite samples were carried out in order to determine the specific surface area, , as one of the important characteristics of the nanocomposite materials. The results are summarized in Table 1. It is obvious that increases with increasing final pH, which is consistent with the evolution of values with the preparation conditions. The resulting values of the are significantly higher than the values previously reported for the CeO2-based materials [10, 12, 13, 22, 4547]. It is also clear that of the v06 sample is more than 25% higher than that of the other samples. These results are in good agreement with the different morphology of the v06 sample as observed by SEM, with larger surface area. Due to very high values, the high catalytic activity of our samples is expected [45]. A test of catalytic performance by means of oxidation of butane to carbon dioxide in oxygen/argon atmosphere was carried out; for details see Supplementary Material.

Finally, the nature of the spinel phase in the NP samples was investigated using Mössbauer Spectroscopy. The Mössbauer spectra of the CoFe_free and CoFe_mod samples (Figure 4) reveal that most of the NPs are in the blocked state at RT, as is observed from the presence of sextet with asymmetric absorption peaks corresponding to the small particle size with nonzero size distribution. The doublet and singlet in the spectra correspond to the NPs in superparamagnetic (SPM) regime and are close to the blocking temperature, , respectively. No significant changes in the spectra and Mössbauer parameters of the CoFe_mod sample in comparison with the CoFe_free sample indicate that the modification of the sample is not connected with remarkable structural or phase changes.

Quantitative information about samples was extracted refining all spectra; parameters of the fit and resulting Mössbauer parameters are summarized in the Supplementary Material (Table S2). The convolution of the Gaussian distribution and Lorentzian profile function was used to refine the sextet spectra. Determined smaller values of the hyperfine fields (43.0 T for the sites and 47.2 T for the sites of the CoFe_free and CoFe_mod samples) are consistent with small CoFe2O4  . The isomer shifts,  mm·s−1, and zero quadrupole shifts are attributed to the ferrite spinel structure [25].

Slight change in the relative area of the doublet (13.6% versus 23.4% for CoFe_free and CoFe_mod samples, resp.) can be interpreted as the change in the NP surface layer (increase of disorder of the surface) in the CoFe_mod sample with respect to the CoFe_free sample after citric acid treatment.

3.2. Magnetic Properties

Magnetic measurements were carried out in order to unambiguously confirm the presence of the CoFe2O4 NPs in the nanocomposites; advanced analysis of the data was applied in order to examine changes in the magnetic response of the NPs associated with their surface modification and incorporation into the CeO2 matrix.

The ZFC-FC curves of the CoFe_free sample exhibit behavior expected for the SPM system with the interparticle interactions [48, 49] that are manifested by the saturation of the low temperature part of the FC curve (Figure 5(a)). The temperature of the maximum of the ZFC curve, , reaches () K. The ZFC-FC curves of the CoFe_mod and nanocomposite samples are comparable to that of the CoFe_free sample (Figure 5(b)); difference is the broader maximum at the ZFC curves of the nanocomposite samples. The values of are summarized in Table 2. There is observable increase of value for the CoFe_mod sample and larger increase for the nanocomposite samples in comparison with the CoFe_free sample. No features indicating presence of another magnetic phase except the SPM CoFe2O4 were observed.

The ZFC curves were refined using theoretical relationship for the ZFC magnetization of the NP system [50] (details are in the Supplementary Material) in order to estimate distribution of and to evaluate the effect of the particle size and effective anisotropy distributions. The resulting values of the mean blocking temperature, , are summarized in Table 2. Compared with the values of for the CoFe_free sample, increases after the NP surface modification (the CoFe_mod sample) and further after the incorporation of the NPs into the matrix (the nanocomposite samples). Focusing on the details of preparation, the increase of the is larger for dropwise addition of the NaOH (the v02–v04 samples, ~230 K) than for the fast addition of the NaOH (the v05 and v06 samples, ~220 K).

The distribution for the CoFe_mod and nanocomposite samples is wider in comparison with the CoFe_free sample, but the individual distributions are comparable for the nanocomposite samples prepared with the similar rate of the NaOH addition (comparing the v02–v04 versus v05 and v06 samples, Figure 5(c)). Hence we can conclude that distribution width for the nanocomposite samples and corresponding shift of are given by that rate of the NaOH addition.

Focusing on the mechanism behind distribution widening for the CoFe_mod and nanocomposite samples, we can explain observed effects as follows: the process of the NP surface modification by citric acid and embedding of the NPs into the matrix can vary both the NP volume as well as its effective anisotropy constants, , and hence distribution of as a result. The NP is given both by the particle volume and by the surface anisotropy constant term [48] which is significantly influenced by any changes of the NP spin structure that can arise during the modification and embedding process. In our case, the spin structure and hence the can be varied by at least one of the three different mechanisms: () Change of the particle volume and particle size distribution, which also alter due to the finite-size effect [48]. Partial dissolution of the NPs may be the consequence of the addition of citric acid/NaOH. In the case of the nanocomposite samples, the different NaOH addition regimes theoretically lead to the different volume changes for individual samples. () Chemical treatment during modification and incorporation process (e.g., addition of the citric acid/NaOH) which causes surface disruptions and hence alters the surface spin structure [51, 52]. In the nanocomposite samples, each regime of the NaOH addition can lead to different degrees of surface disruptions and as a result affects and distributions. Chemical treatment of the as-prepared NPs during their incorporation into the matrix should affect just the NP surface/shell layer, not the ion distribution within the NPs. Comparing our results with the literature, Aneesh Kumar et al. [53, 54] reported study of the role of the pH value during chemical reaction on the magnetic properties of the spinel ferrite structure. Adjusting of the pH value led to the change of the cation distribution in the spinel structure. However, in our work, the as-prepared CoFe2O4 nanoparticles were exposed to the different pH values during the embedding process to the matrix, not during the formation of the individual nanoparticles as in the Aneesh Kumar works. Thus the degree of inversion in our CoFe2O4 nanoparticles should not be significantly modified by changing the pH during the embedding step, as it is not expected to change cation distribution in the whole volume of the as-prepared nanoparticles by the subsequent pH treatment unlike the case of nanoparticles exposed to the pH treatment during their formation. () Change of the interparticle interactions can also lead to the changes in the alignment of the surface spins (the so-called spin canting). Surface modification of the NPs causes raising of the strength of the dipolar interactions via decreasing of the interparticle distances between NPs with the citric acid coatings in comparison with the oleic acid coated NPs of the CoFe_free sample [55]. On the other hand, dispersion of the NPs in the CeO2 matrix in the case of nanocomposite samples leads to the weakening of the dipolar interparticle interactions. Both effects result in the changes of the spin structure of the NPs.

All three mechanisms leading to different width of the distribution can be present simultaneously and their individual contributions are not clearly distinguishable, especially if we have no information about apparent or crystallite NP size in the nanocomposite samples. However, experimentally observed change of the NP crystalline size for the CoFe_mod sample in comparison with the CoFe_free sample suggests that the first and second mechanisms are reasonable for explanation of the effects after exposure of the NPs to citric acid.

The of all samples measured at 10 K are symmetric for both polarities of the magnetic field. curves measured at 400 K are unhysteretic (Figure 6) which indicates that the NPs are in the SPM regime. The determined magnetic parameters (coercivity at 10 K, , and saturation magnetization at 10 and 400 K, and ) are summarized in Table 2. The CoFe_free sample possesses very similar value as was previously reported for the NPs prepared by hydrothermal method [29]. The CoFe_mod sample exhibits nearly the same hysteretic behavior as the reference CoFe_free sample (Figure 6(a)), with the similar value of indicating presence of no other magnetic phase except the CoFe2O4.

All nanocomposite samples exhibit similar large values, lower than the value for the CoFe_free sample. Decrease of the value can be explained by the varied value, modified single NP properties, or weakening of the interparticle interactions due to incorporation of the NPs into the matrix, as was already discussed in the description of the curves and distributions above.

curves were refined in the MINORIM software [44] to determine the distribution of the magnetic moments (superspins), , belonging to the individual CoFe2O4 NPs. Resulting fits are depicted as solid lines in Figures 6(b) (the CoFe_free and CoFe_mod samples) and 6(d) (nanocomposite samples). Resulting distributions are shown in Figure 7. The unimodal distribution was found for all samples; values of the mean magnetic moments, , are summarized in Table 2. The mean magnetic diameters of the NPs, , were calculated from the values in the same way as is described in [56]. Taking the average per formula unit of the half inverse CoFe2O4 equal to 5  [57] and considering that the unit cell contains eight formula units, the per unit cell, , was taken as 40  for all samples. The resulting values of the are summarized in Table 2.

Analysis of the distributions for individual samples revealed that the distribution changes with the surface modification of the NPs (when the CoFe_free and CoFe_mod samples are compared) and then again with incorporation of the NPs into the matrix (Figure 7 and Table 2). The maximum of the distribution as well as and corresponding values are shifted to the lower values after the citric acid and subsequently the NaOH treatments. Decrease of the value for the CoFe_mod sample in comparison with the CoFe_free sample is consistent with decreased and proposed core-shell structure of the modified NPs, with the Co-free (deficit) shell. The largest reduction of the values was observed for the samples exposed to dropwise addition of the NaOH (the v02–v04 sample series) with the highest pH value (the v04 sample). We assume that higher final pH may cause larger changes in the structure and spin order in the NP surface layer during the incorporation process resulting into the reduction of the value, whereas fast addition of NaOH leads to smaller changes of the NP surface layer as the reaction is generally faster (the v05 and v06 samples). We also observed that variation of the through the nanocomposite sample series is comparable with variation of the values. All together finally suggest that not only the NP size and size distribution, but also the surface spin alignment and hence are slightly modified after chemical treatment of the NPs.

of the CoFe2O4 NPs was estimated from the values of and using the relationship for of the SPM NPs [48], resulting in the equation for :Overall evolution of the values through the sample series (Table 2) is consistent with the so-called finite-size effect [48]— increases with decreasing particle size, . Its value is larger than the bulk value, = 1.8–2.7 × 105 J/m3 [27, 58], for all samples and corresponds with the value expected for the small NP (with diameter reaching approximately 4 nm). increases with the modification by citric acid and subsequently with the NaOH treatment, reaching the largest values for dropwise addition of NaOH under large final pH, which is consistent very well with our previous findings. These results finally confirmed previous assumptions deduced from the and distributions that (especially its surface term) is altered after the surface modification and incorporation of the NPs into the matrix.

4. Conclusion

We succeeded in the preparation of the CoFe2O4/CeO2 nanocomposites by a novel two-step preparation route, which is based on embedding of the CoFe2O4 NPs as magnetically active carriers into a catalytically active CeO2 matrix. Specific parameters of the preparation route, as the final pH of the reaction and method of its adjustment, were varied through the series in order to determine the most efficient conditions for preparation of the nanocomposite with the highest surface area—a prerequisite for large catalytic activity. It has been observed that the microstructure of the CeO2 matrix evolves with preparation conditions; increase of the final pH and fast addition of the NaOH led to decrease of the crystallite diameter and consequently increase of the specific surface area.

The unambiguous presence of the CoFe2O4 NPs in nanocomposite samples was confirmed by magnetic property measurements. The experiments revealed only slight changes in magnetic size and magnetic properties of the CoFe2O4 NPs connected with the incorporation process. The best structural and magnetic parameters were observed for the nanocomposite sample prepared by fast addition of NaOH with the final pH equal to 12. This sample exhibits one of the largest values of the specific surface area reported for the CeO2-based catalyst (264 g/m2) and extremely low crystallite size of 2 nm. Our approach is thus very promising for production of ultrafine powders and nanocomposites with large specific surface area based on CeO2 and different magnetic carriers. Moreover, extension of the preparation route enables embedding of more than one type of nanoparticles, which open door for facile preparation of multicomponent catalytic materials.

Competing Interests

The authors declare that they have no conflict of interests.

Acknowledgments

This research was supported by the Grant Agency of the Czech Republic (Project no. 15-01953S). Magnetic measurements were performed in MLTL (http://mltl.eu/), which is supported within the program of Czech Research Infrastructures (Project no. LM2011025). Electron microscopy at the Institute of Macromolecular Chemistry was supported by Project POLYMAT LO1507 (Ministry of Education, Youth and Sports of the CR, program NPU I). The authors thank Mrs. Jirina Hromadkova for technical assistance.

Supplementary Materials

Supplementary Material contains: Additional Figures and Tables for EDX, SEM, TEM and Mössbauer Spectroscopy measurements. Detailed description of ZFC curve refinement and description of the test of the catalytic activity of the samples by oxidation of butane to carbon dioxide.

  1. Supplementary Material