Aqueous Synthesis of Sub-11 nm Fe-Cu Oxides and Alloy Nanostructures: Structural and Morphological studies

INTRODUCTION

Bimetallic nanoparticles have demonstrated to possess better properties than monometallic nanoparticles mainly due to the additive properties of the two metal components. Iron and copper oxides nanoparticles have been synthesized for several years, and their interesting properties and applications have been reported. Magnetite (Fe₃O₄), hematite (α-Fe₂O₃), and maghemite (γ-Fe₂O₃) are magnetic materials which have been applied as nanomaterials in areas such as catalysis, biomedicine, data storage devices, magnetic fluids, gas sensors, magnetic resonance imaging, among others [1,2]. Copper and copper oxides nanoparticles are recognized for their high electrical conductivity, with potential applications in fields including electronics, optics, and catalysis [3,4]. Thus, the combination of Fe and Cu and/or their oxides in a nanostructure can provide very interesting properties that widens the potential applications for that functional nanomaterial. For instance, bimetallic copper-iron oxide nanoparticles have been synthesized throughout a simple chemical precipitation methodology; and these bimetallic nanoparticles have been employed to add electrical and magnetic properties to leather for lighting applications [5]. Furthermore, bimetallic iron-copper oxide nanoparticles supported on nanometric diamond have been employed as an efficient and stable sunlight-assisted Fenton photocatalyst [6]. Additionally, Fe-Cu oxides nanostructures were obtained using Virginia Creeper (Parthenocissus quinquefolia) leaf extract in the presence of oxalic acid; these nanomaterials were used for the removal of green malachite from water [7]. Moreover, nanoalloys have drawn increasing research interest owing to their structural studies [8-12] and their potential applications in catalysis [13-15], surface plasma band energy [16,17], optoelectronics, information storage, adsorption process, magnetic properties [18,19], among others. The study of transition metal clusters and metallic alloys has been also developed both theoretically and experimentally, with great interest in their applications for the removal of contaminants [11,14,15]. These kinds of applications are closely related to the sui generis properties of each metal forming the alloy and determined, also, by their shape, size distribution and composition [20]. The growing interest in Fe-Cu nanoalloy has risen due to its bulk immiscibility, the intrinsic magnetic properties, and the synergistic effects of two-metal redox couples between iron and copper [21-24]. This nanoalloy has been used in electronic, chemical, and environmental research [25], due to their magnetic and catalytic properties. Individually, iron nanoparticles have many attractive applications [26] mainly in optics [27], magnetism [28], electrical devices [29], electrocatalysis [30], and environmental remediation [31]. Likewise, copper nanoparticles have been used in optical [32], magnetic [33] and sensor devices [34], catalysis [35] and environmental remediation [36] as well as antifungal and bacteriostatic agents [37]. Several methods have been applied for the preparation of bimetallic nanoparticles, including alcohol reduction [13], citrate reduction [17,38], polyol processes [39], borohydride reduction [40], solvent extraction–reduction [16,41,42], sonochemical methods [43], photolytic reduction [44,45], radiolytic reduction [46,47], laser ablation [48,49], and biological programming [50]. Mechanical alloying, using the high energy ball milling (BM), was also proved to be useful for synthesizing various phases [13,14,20]. Thus, chemical reduction methods have confirmed to be relatively simple and useful when small size nanoparticles are required for catalytic applications; these particles usually have 10 to 1000 times greater reactivity compared with particles obtained by other methods [51].

In this work, a series of FeCuNS were prepared by an aqueous chemical reduction methodology, utilizing sodium borohydride as a reducing agent and iron and copper chlorides as metal precursors. The FeCuNS were thoroughly analyzed by XRD, SEM, EDS, TEM and XPS to obtain their chemical, structural and morphological characteristics.

Experimental

Materials and Methods

All chemical reagents were of analytical grade, commercially acquired, and used without further purification. All solutions were prepared with deionized water (with ρ 18.2 MΩ cm) obtained from a Mili-Q System (Millipore, USA).

In a typical preparation of FeCuNS, 1 M FeCl₂∙4H₂O (Fermont) and 1 M CuCl₂∙2H₂O (J.T. Baker) aqueous solutions were combined in a 1:1 ratio. The mixture was stirred vigorously for one hour, followed by reduction using a 4 M NaBH₄ aqueous solution (98%, Aldrich). Nitrogen gas was bubbled during synthesis to evacuate oxygen and to prevent the complete oxidation of Fe-Cu nanoparticles. A fine black precipitate was obtained, which was filtrated and washed two times with deionized water. Further syntheses of the nanoalloy were performed with three different Fe:Cu mass ratios (75:25, 50:50, 25:75 wt.% FeCuNS, respectively), following the same reaction conditions described above. The crystalline phases of the Fe-Cu samples were identified by X-ray diffraction (XRD) patterns, which were collected on a Bruker D8 Advance powder diffraction system using Cu-Kα radiation (λ = 0.15406 nm) in 2θ range from 5 to 70°, operating at 35 kV and 35 mA, with the scanning speed of 0.02° s^-1. X-ray photoelectron spectra (XPS) for all samples were collected using a JEOL JPS-9200, equipped with a Mg X-ray source (1253.6 eV) at 200 W over an analysis area of 1 mm² under vacuum on the order of 1×10⁻⁸ Torr. Survey and narrow spectra were recorded and scannedwith an energy pass of 50 and 20 eV, respectively. The spectra were analyzed using the Specsurf™ software included with the instrument. The surface chemical composition was determined from the corresponding peak area. Corrections of charging of the scanned elements were carried out taking the carbon signal (C1s) at 284.5 eV as the reference point. The Shirley method was used for the background subtraction, whereas curve fitting was conducted with the Gauss-Lorentz method. SEM observations were carried out on a JEOL JSM-6510LV operated at 20 kV accelerating voltage. Each FeCuNS sample was attached to aluminum stub using conductive double-stick carbon tape and observed without coating. Elemental characterization was performed using an OXFORD Energy-Dispersive X-Ray spectroscope with a resolution of 137 eV anchored to the Scanning Electron Microscope. A FEG Hitachi S-5500 ultra-high-resolution electron microscope (0.4 nm at 15 kV) with a BF/DF Duo STEM detector attached to an EDS-probe (from Bruker) was employed. Transmission electron microscopy (TEM) micrographs and SAED patterns were collected in a JEOL JEM-2100 microscope operated at 200 kV with LaB₆ filament. Samples were prepared in the following way:  each FeCuNS was dispersed in isopropyl alcohol, and then one drop of this suspension was placed in a copper grid coated with carbon. Finally, the grid was allowed to air-dry. Brunauer–Emmett–Teller (BET) surface areas were determined by standard multipoint techniques of nitrogen adsorption, using an Autosorb iQ Station 1 of Quantachrome Instruments. Samples were heated at 200°C for 2 h, before measuring the specific surface areas.

RESULTS AND DISCUSSION

The chemical reduction of FeCl₂∙4H₂O and CuCl₂∙2H₂O aqueous solutions, using NaBH₄ as a reducing agent, yielded a black fine precipitate. Fig. 1 shows pictures of the precipitate, with the characteristic blackened color of iron oxides. Additionally, a magnetic response of the powder can be seen when a magnet approaches the glass vial. This behavior was more obvious, as expected, in the FeCuNS samples with 75:25 and 50:50 wt%.

XRD patterns of the as-prepared FeCuNS (75:25, 50:50 and 25:75 wt. %) are shown in Fig. 2. Three different phases of FeCuNS: Cu_xFe_1-x, CuFe₂O₄, and FeCu₄ can be detected in the samples and confirmed by the cards JCPDS 49-1399, 25-0283 and 65-7002, respectively. The peak obtained at 2θ = 43.33° shows the formation of the Fe-Cu nanoalloy (JCPDS No.065-7002 like FeCu₄) in the 50:50 wt.% and 25:75 wt.% samples, respectively, which can be indexed on the basis of a face−centered cubic structure (FCC) [52]. Figs. 2b-c show diffraction peaks at 2θ = 35°, which can be associated to copper ferrite (CuFe₂O₄) and are assigned to the (103) plane. In addition, the 75:25 wt. % sample (Fig. 2a) illustrates the peak at 2θ = 30.1°. The presence of CuFe₂O₄ (JCPDS No.25-0283) diffraction peaks is evidence of the strong bimetallic interaction, which can lead to the formation of a spinel structure at nanoalloys. XRD analysis indicates that initial CuO was reduced to Cu₂O/Cu⁰ in all the samples. The main peak of cuprite (Cu₂O) is oriented at (111), showing at 2θ = 36.43°, 35.45° and 36.53° for the 25:75 wt. %, 50:50 wt. % and 75:25 wt. % samples, respectively, as well as at 2θ = 42.32° in the 25:75 wt. % and 50:50 wt. % samples, oriented at (200); these peaks are in accordance with the reported values by [52]. The Fe-Cu nanoalloy showed distinct peaks at around 2θ = 43.30 and 50.5 (Figs. 2b-c), which could be attributed to Cu (JCPDS No.04-0836) or FeCu₄ (JCPDS No.65-7002). Since Cu and FeCu₄ have similar peaks in XRD patterns, it is hard to distinguish between these two phases. According to these results, it was found that there is an interaction between iron and copper oxides, promoting the dispersion of both species. In all samples, the absence of metallic copper or cuprous oxide signals suggested that either they are well dispersed or their peaks are overlapped with that coming from Fe₂O₃ at 2θ = 35.6°. In addition, a broadening of the diffraction peaks occurs gradually with increasing the mass concentration of Fe [52].

The elemental content of the superficial layers of the FeCuNS was obtained by XPS. The survey spectra, scanned from 0 to 1100 eV binding energy (BE), are shown in Fig. 3. To further confirm the existence of the Fe-Cu nanoalloy in the nanoparticles obtained, iron and copper elements were analyzed by deconvolution of their high resolution XPS spectra. In the three FeCuNS samples (Fig. 4), the different peaks observed correspond to different iron (FeO₂; Fe(OH)O; Fe₂O₃; Fe₃O₄) or iron-copper (FeCuO₂) oxide species. Only in the 25:75 wt.% FeCuNS sample, there is a peak at 706.7 eV, which can be attributed to the binding energy of the Fe-Cu nanoalloy (Fe 2p_3/2). These results indicate that the Fe and Cu particles form a layer of diverse iron and copper oxides, which is due to the easily oxidizing zero-valent metals when exposed to air [51-53]. Furthermore, the presence of FeCu₄, showed by XRD, was confirmed by the signal at 706.7 eV, which can be assigned to Fe-Cu in the 25:75 wt.% FeCuNS sample. The presence of FeCuO₂ at 713.6 eV and from 714.1 to 716.9 eV in the 75:25 wt.% and 50:50 wt.% FeCuNS samples, respectively (Figs. 4a and b), confirms the presence of CuFe₂O_4, as obtained by XRD studies.

XPS spectra of Cu 2p_3/2 are shown in Fig. 5. Two main photoelectron peaks at 932 to 932.9 eV can be observed in the 75:25 wt.%, 50:50 wt.% and 25:75 wt.% FeCuNS samples [54]; the peak at 932 eV confirmed that CuO was reduced to Cu₂O and metallic copper in all samples. O1s energetic distributions were adjusted with three Gaussian curves with FWHM = 1.4 ± 0.1 eV, as can be seen in Fig. 6. Regarding Fe 2p_3/2 and Cu 2p_3/2, each curve was assigned to a state or combination of states, according to its binding energy. The signal at 529.8 eV in Fig. 5c corresponds to CuFe₂O₄. Other peaks can be correlated to the different iron and copper oxides: Fe₂O₃, Cu₂O [55], in the 50:50 wt.% FeCuNS (Fig. 6b), and the iron oxides Fe₃O₄ and Fe₂O₃ in the 75:25 wt.% FeCuNS sample (Fig. 6a).

The morphology of the FeCuNS was analyzed by SEM. Fig. 7 shows the images of the 75:25 wt. %, 50:50 wt. % and 25:75 wt. % FeCuNS (Figs. 1a, b and c, respectively), where it is observed that the nanomaterials form agglomerates having rough surfaces [56]. EDS spectra were collected in two different regions of interest with a magnification of 2,000x for each FeCuNS. The results of chemical analysis by EDS are shown in Table 1, where similar proportions of 75:25, 50:50 and 25:75 wt.% FeCuNS were observed. Nevertheless, there is not an exact match between experimental and theoretical concentration of Fe and Cu due to the presence of oxygen associated with the formation of oxides during the synthesis.

The carbon content can be associated to the conducting C-tape, where the samples were mounted to perform the SEM-EDS analyses. Otherwise, the high oxygen content, observed in Table 1, can be explained due to the oxidation process of the bimetallic samples.

FESEM and STEM-BF images of the FeCuNS are shown in Figs. 8a and b, respectively. Their morphology can be described as irregular quasi-spheroidal particles with sizes smaller than 100 nm. EDS mapping (Figs. 8c-e) shows a homogeneous distribution of the Cu and Fe on the FeCuNS. From FESEM-BF image (Fig. 8b), it can be observed that the particles tend to mainly form agglomerates, but foil-like structures are also present.

It is known that different concentrations of the metal precursor solutions when forming the nanoalloy and the difference in the standard reduction potentials of Fe and Cu generate different bimetallic structures [9]. Moreover, there are reports of particles where Fe is at the center of the structure of the nanoalloy, forming the core of the particle, and Cu and Fe oxides are present too but in the shell [52]. Thus, the morphology of the obtained FeCuNS can be observed by a transmission electron microscopy (TEM) analysis. The images in Fig. 9 confirm the irregular quasi-spheroidal morphology of the FeCuNS. Size distribution histograms (depicted in the lower part of Fig. 9) of the FeCuNS were built from the measurement and analysis of 200 nanoparticles for each sample using representative TEM images. The particle size in the 75:25 wt.% FeCuNS sample shows the unimodal distribution of the synthesized nanostructures (Fig. 9a and bottom left of the histogram), with an average size of 7.7 nm (±3.4 nm). 50:50 wt. % and 25:75 wt. % FeCuNS samples demonstrate asymmetric distribution (Figs. 9b and c and bottom middle and right of the histograms, respectively), with an average size of 11 nm and (±5.5 nm) and 6.5 nm (±2.0 nm), respectively. HRTEM images (Fig. 10) show lattice spaces of 2.123 nm, 1.32 nm, and 1.035 nm between the (400), (531) and (553) planes, respectively, consistent with a Fe-Cu nanoalloy (JCPDF No.49-1399, as Cu_xFe_1-x). The HRTEM image in Fig. 10a reveals a fringe spacing of 1.887 nm in agreement with lattice spacing of the (331) plane of Fe₃O₄. The HRTEM micrograph in Fig. 10c shows an interplanar distance of 2.063 nm, which can be indexed as the (220) plane associated with copper iron oxide (CuFe₂O₄). Furthermore, the SAED patterns confirm the presence of magnetite as the diffraction rings can be indexed based on the JCPDF No.79-0416 card.

N₂ adsorption–desorption isotherms of the FeCuNS samples are illustrated in Fig. 11. All samples exhibit typical III N₂ adsorption isotherms. Specifically, the inflections at P/P₀ = 0.4–0.6 reveal the narrow pore size distribution of mesoporous materials [20]. Table 2 shows the textural properties of Fe-Cu nanoparticles at the different metal ion concentrations used in the synthesis. The BET surface area increases in the 75:25 wt.% FeCuNS sample, while it decreases in 25:75 wt.% FeCuNS sample, which can be attributed to the saturation of nanostructures pores due, perhaps, to the increasing amount of Cu in the samples. However, this needs to be studied further.

CONCLUSIONS

FeCuNS were prepared in three different Fe:Cu ratios, namely 75:25, 50:50, and 25:75 wt. %, by an aqueous reduction process using NaBH₄, under N₂ atmosphere. The main phases in these nanostructures, as determined by XRD, were Fe₂O₃, Fe₃O₄ and CuFe₂O₄. The main peak obtained at 2θ = 43.33° also revealed the formation of the nanoalloy Fe-Cu (JCPDS No.065-7002, as FeCu₄) in the 50:50 wt.% and 25:75 wt.% FeCuNS samples, which can be indexed as a face-centered cubic structure (FCC). According to the TEM results, the obtained FeCuNS tend to form agglomerates, which can be associated to the magnetic nature of some metal oxides present in the nanomaterials, and, remarkably, their particle average size ranges from 6.5 to 11 nm. The XPS analysis of the FeCuNS further confirms the existence of FeCu₄ (Fe 2p_3/2 at 706.7 eV), along with the presence of the other main iron and copper oxides phases.

ACKNOWELDGMENTS

The authors are thankful to Dr. Uvaldo Hernández Balderas, Dr. Gustavo López-Tellez, and L. I. A. María Citlalit Martínez Soto (CCIQS UAEM-UNAM) for X-ray powder diffraction analyses, XPS analyses and computing assistance, respectively. RZL thanks CONACyT for PhD scholarship (273831).

FUNDING

This work was funded by Universidad Autónoma del Estado de México.

CONFLICT OF INTEREST

The authors declare no conflict of interest.