Sonochemically Synthesized Spin-Canted CuFe2O4 Nanoparticles for Heterogeneous Green Catalytic Click Chemistry.

Heterogeneous green catalysis by using magnetically separable nanometal-oxide catalysts has become a subject of prime focus recently. PXRD (powder X-ray diffraction), FESEM (field emission scanning electron microscopy), and HRTEM (high-resolution tunneling electron microscopy) with IR and Raman spectroscopy are applied to analyze the structural and microstructural properties of nanosized (∼15.3 nm) CuFe2O4 synthesized by both sonochemical and mechanochemical processes. The sonochemical process provides a better uniformity of sizes of the nanoparticles (NPs). Rietveld refinement with the PXRD pattern reveals the inverse spinel-like architecture of CuFe2O4 NPs. The Raman spectra also indicate the phase purity of the synthesized material. The static magnetic measurements are performed at different magnetic fields and temperature ranges from 300 to 5 K, which confirms the existence of the ferrimagnetic phase mixed with some finer superparamagnetic (SPM) nanophases within the sample. Unsaturated magnetization is observed even at an applied 5 T magnetic field for the presence of spin-canting nature in the partially inverted copper ferrite phases at the surfaces of the nanoparticles. Now, these coupled magnetic CuFe2O4 NPs are used as a heterogeneous catalyst for three-component Huisgen 1,3-dipolar cycloaddition click reaction in aqueous media. By this catalyst system, we were able to couple alkyl halide, epoxide, or boronic acid with alkynes efficiently to furnish 1,4-disubstituted 1,2,3-triazoles in excellent yields within very short reaction time. The test for heterogeneity, reusability, and reproducibility of the catalyst has also been performed successfully without prominent decrease in yield up to the fifth cycle.

Introduction

Heterogeneous green catalytic click chemistry is very important from several perspectives: (a) green chemistry: the reaction proceeds without using any solvent only in the presence of the catalyst, (b) easy separation: the catalyst can be separated from the mixture by simple filtration, (c) click chemistry: click chemistry provides the maximum conversion.[1] Catalytic click chemistry is a unique approach for the conversion of reactant molecules to a particular desired product in a single step or consecutive steps. Several types of heterogeneous catalysts such as MOFs,[2] metal nanoparticles (NPs),[3] and metal–oxide NPs[4] are used for such click chemistry, while usage of magnetically separable spinel ferrites for such application is relatively scarce.[5] On the other hand, copper-catalyzed azide–alkyne cycloaddition (CuAAC) is an important “click chemistry” reaction which has been extensively applied in chemical biology, medicinal chemistry, and materials science. The CuAAC reaction of terminal alkynes provides a mild and efficient synthesis of 1,4-disubstituted 1,2,3-triazoles. However, such reaction in the case of internal alkynes to afford trisubstituted triazoles, is still very challenging.

In recent times, the nanosized spinel ferrites have been used as a potent applicant in the field of materials science not only due to their widespread technological applications but also to understand the fundamentals of nanomagnetism.[6] Interestingly, the physical properties of such nanometric substances differ enormously from their bulk properties. Thus, understanding and discovering a technique for controlling superparamagnetism, collective magnetic excitation, spin canting effect, and spin-glass (SG)-type nature of ferrite nanomaterials require considerable thought to expand their broad scale application in electronic and magnetic devices (microwave, radiofrequency, optoelectronic devices, etc.), ferrofluid technology, magnetic resonance imaging, and hyperthermia for cancer treatment.[7] According to the phase diagram proposed by Villain, Poole, and Farach, with the introduction of nonmagnetic ions in place of magnetic ions at tetrahedral (A) and octahedral [B] sites, different magnetic orders evolve in the spinel ferrite systems, and this is mainly due to the modifications in the inter-sublattice A–O–B (JAB) and intra-sublattice A–O–A (JAA) and B–O–B (JBB) exchange interactions.[8]

Among various spinel ferrites, nanosized copper ferrite (CuFe2O4) has drawn significant attention because it is easy to synthesize, gives high yield at low cost, and is mechanically hard and chemically stable.[9] Furthermore, it is an attractive magnetic material extensively used in a variety of applications, such as catalysis, sensor technology, high-density data storage, magnetic resonance imaging, and magnetically guided drug delivery.[10] Magnetic and electrical properties of copper ferrite nanostructures are sharply guided by the cation distribution in their specified magnetic sites.[11] Copper ferrite occurs naturally in two crystallographic spinel structures: the high-temperature cubic phase (c-CuFe2O4) and the low-temperature tetragonal phase (t-CuFe2O4). It is ferrimagnetic (Néel temperature approximately 780 K) at room temperature (RT). The magnetization of sublattice-A is antiparallel to sublattice-B, and both sublattices (A and B) are ferromagnetically coupled. The overall μeff of copper ferrite arises as a result of uncompensated magnetic moments of eight Cu2+ ions in sublattice-B. The relatively small energy difference of Cu2+ ions in A and B sites helps to originate five types of cation redistributions, and this is strongly dependent on annealing temperature as well as the cooling rate.[12] Thus on an A site, a single Cu2+ ion per unit cell doubles the magnetic moment.

Under this background, we have synthesized monodispersed, uniform-sized CuFe2O4 NPs by the sonochemical process and characterized them by powder X-ray diffraction (PXRD), high-resolution transmission electron microscopy (HRTEM), dynamic light scattering, selected area electron diffraction (SAED), and energy-dispersive X-ray spectroscopy techniques to investigate its various physical properties. Rietveld analysis reveals that CuFe2O4-NPs are crystallized in the cubic Fd3̅m space group, and the corresponding JCPDS file number is 77-0010. The NPs were also characterized by IR and Raman spectral analyses. The magnetic study indicates that CuFe2O4-NPs show strong ferromagnetic interaction with spin-canting behavior. These CuFe2O4 NPs are employed for the heterogeneous catalytic conversion of Huisgen 1,3-dipolar cycloaddition click reaction in water. Through this article, we want to report in situ generation of alkyl or aromatic azides from corresponding halide/boronic acid/epoxide and sodium azide and subsequently their coupling with terminal and internal alkynes in the presence of CuFe2O4 NPs. Leaching is a major limitation in the use of heterogeneous catalysts, and it can completely cripple application. However, we observed negligible amount of leaching in our synthesized catalyst. The click reaction of internal alkynes with azides, producing trisubstituted triazoles, is extremely challenging.[13] However in our hand, from the point of reactivity, there is no difference between terminal and internal alkynes. This result is very promising because there are only few reports to couple internal alkynes with azides.[14]

Results and Discussions

Structural and Microstructural Analyses

Phase purity of the bulk material has been analyzed by the PXRD pattern (Figure ) of the synthesized material at the solid state, and also a detailed structural analysis has been carried out by using Rietveld analysis. Initially, the PXRD pattern of our sample was indexed by statistical analysis using DICVOL06[15] and TREOR90 of the Fullprof.2k package.[16] The corresponding Miller indices and peak positions are perfectly matched with the standard PXRD pattern of spinel ferrites.[17] Structural and microstructural properties, phase purity along with the crystal structure, bond lengths, and bond angles were investigated by the Rietveld refinement of the PXRD pattern for our sample using the GSAS program[18] with the EXPGUI interface.[19] The results are shown in Figure . The refinement parameters, metal–oxide bond lengths, and bond angles are listed in Table S1 (Supporting Information), and the fractional coordinate as well as occupancy of different ions obtained from the refinement by the GSAS program are provided in Table S2 (Supporting Information). According to the literature, the Cu2+ ions have B-sites affinity rather than A-sites.[20] We have fitted the PXRD pattern of the sample with three different concentrations of Cu2+ ions occupying [A] and [B] sites mentioned in Lakhani et al.[21] It has been found that “Chi-square” (χ2) values obtained for this (Cu0.20Fe0.80)A[Cu0.80Fe1.20]BO4 cation distribution suggest good refinement of the PXRD pattern of the sample.

Figure 1

XRD pattern of CuFe2O4 NPs.

Figure 2

Structural diagram of CuFe2O4 NP.

The average crystallite size of the sample is calculated from the broadening of well-defined peaks of the sample in the (XRD) pattern by using the Debye–Scherrer equationwhere ⟨D⟩ is the average crystallite size, “λ” is the wavelength of the incident X-ray radiation, and “θ” is the Bragg angle. Here, “β1/2” is the full-width at half-maximum of the XRD peak. The average estimated crystallite size of the sample is ∼9 nm which indicates the nanocrystalline nature of the sample and in the range of superparamagnetic limit. The lattice parameter of the sample is calculated by taking consideration of all peaks in the XRD pattern and is given by ∼8.2 nm.

FESEM and HRTEM Study

To have a deep insight into the morphology of the nanocrystalline CuFe2O4, HRTEM images were taken and shown in Figure . During the HRTEM observation of the SAED patterns, some representative micrographs and typical lattice fringe patterns recorded are shown in Figure . The representative pattern of the micrograph having distribution of particle size of the CF sample is shown in Figure a. From this study, it is observed that the NPs are monodispersed of more or less spherical shape with uniform size. The average crystallite size obtained from the TEM micrograph is ∼9 nm. The SAED pattern of CF is shown in Figure b. The lattice planes corresponding to CF were detected in the SAED pattern, and these planes are also assigned in Figure b. The clear rings with different diameters (i.e., different lattice planes of CF) in Figure b point out the polycrystalline nature of the CF sample. A clear lattice fringe pattern of CF is shown in Figure c which indicates the well-crystalline nature of the sample. In Figure c, the separations between the consecutive parallel fringes are assigned and marked according to the calculation, and the calculated values are 2.53 and 2.10 Å and these correspond to the (311) and (400) lattice planes of CF. The HRTEM study also ruled out the existence of the unnecessary impurity phase in our CF material (Figure b,c) which is consistent with XRD analysis.

Figure 3

(a) HRTEM image, (b) SAED pattern, and (c) fringe pattern of CF NPs.

IR and Raman Spectroscopy

We gathered information on formation of the desired phase, existence of any impurity, structural defects, crystallinity, and so forth from the analysis of Raman active modes. Figure shows the Raman spectra of CF NPs at RT indicating 10 prominent peaks at 154, 219, 282, 338, 399, 471, 555, 604, 645, and 686 cm–1. All these peaks are assigned to the cubic inverse-spinel CuFe2O4. The bands around 215, 278, 481, 586, and 656 cm–1 are assigned to F2g (1), Eg, F2g (2), F2g (3), and A1g, respectively, corresponding to the cubic ferrite structure. The appearance of additional Raman peaks may be due to the breakdown of the momentum conservation rule because the excitation wavelength (632 nm) is much larger compared to particle diameter (∼9 nm).

Figure 4

Raman spectrum of CF NPs.

Therefore, the number of Raman phonon modes found is incompatible with cubic symmetry. From the observed Raman spectra of CF, the nonappearance of the Raman mode corresponding to α-Fe2O3 and Fe clusters implies that no unnecessary impure phase is present in the sample. In the IR spectra of CuFe2O4, there are two strong absorption bands at about 490 cm–1 which correspond to M–O stretching vibration and O–M–O bending vibration of CuFe2O4, respectively.

Magnetic Study

To know about the magnetic property of our sample, SQUID analysis under field-cooled (FC) and zero field-cooled (ZFC) conditions was recorded in the quantum design superconducting quantum interference device (SQUID) at a constant dc magnetic field starting from 300 K down to 5 K. The M versus T curve is displayed in Figure . The corresponding ZFC–FC magnetization curves in the temperature range of 5–300 K were recorded during heating after cooling in the presence of an applied magnetic field of 250 Oe. In addition, another comparatively higher field of 250 Oe was applied to obtain a considerable difference between the ZFC and FC magnetizations because this difference was used for the evaluation of the average particle size as well as the spin-glass-like phases present in the sample. The ZFC–FC curves of CF are bifurcated below RT and branched out prominently with decreasing temperature. However, with decreasing temperature below 150 K, the magnetic moment related to the ZFC curve decreases while FC curve increases, and this forms a clear hump in ZFC–FC curves. Below the freezing temperature, the system proceeds into a glassy state and a strong irreversibility between the ZFC and FC curve is shown in the M versus T curve, a typical superparamagnetism (SPM) relaxation phenomenon of a SG- or cluster-like system.[22] Because our synthesized NPs have size below 10 nm, it conversely indicates that the hump observed in the ZFC and FC curves is indicating the presence of the SPM blocked state of the NPs. This temperature corresponding to this hump is called blocking temperature. For monodispersed NPs, the maximum point in the ZFC curve and the bifurcation point of FC and ZFC curves are very closely spaced. In monodispersed NPs, usually the blocking condition is defined by single temperature, but two temperatures (TB and TP) are required to explain the blocking condition phenomenon for distributed particles, where TB is considered as the highest temperature at which the ZFC and FC magnetizations are bifurcated corresponding to the larger group of NPs present. Tp is the temperature corresponding to the peak obtained in thermal variation of the ZFC magnetization. In fact, the ZFC magnetization decreases below TP for the existence of tiny particles along with lower TP than TB for such systems. In this analysis, we have found TB is ∼300 K and that TP is near about 150 K (Figure ). Possibly, this is due to the size distribution of CF NPs.

Figure 5

Temperature-dependent magnetization ZFC–FC curves for CF NPs.

The magnetic hysteresis behavior of the sample was recorded in temperatures 300, 150, and 5 K as shown in Figure . This figure exhibits a clear “S”-like symmetry. The opening of the MH loop at lower temperatures results in the ferromagnetic character. Apparently, the magnetic system exhibits the coexistence of ferromagnetism and SG. The magnetization becomes a nonlinear function of applied magnetic field and shows ferromagnetic hysteresis nature in the lower magnetic field range. The magnetization value of CF increases as the applied magnetic field increases. The maximum magnetization that is saturation magnetization (MS) of CF (at 5 T) is quite high as compared to other copper ferrite NPs.[23] As the temperature decreases, the saturation magnetization increases, and the rise in magnetization may be attributed to a variety of factors, namely, freezing of uncompensated surface spins, enhancing ferrimagnetic fraction by lowering the temperature, decreasing nanocrystalline anisotropy, and so forth. The saturation magnetizations of CF at 300, 150, and 5 K are found to be 6.6, 7.7, and 9.3 emu g–1, respectively. The high value of saturation magnetization authenticates that our sample can be emerged as a proper material for applications in soft magnetic devices. Here, very low coercivity at RT confirms soft magnetic nature of our sample. Also, with decreasing temperature, the coercive field increases accordingly and is found to be ∼2800 Oe at 5 K temperature. At relatively low temperature, this huge enhancement of coercivity is mainly due to the blocking of single-domain NPs present in the soft magnetic system.

Figure 6

Magnetic hysteresis loops of CF-NPs at 300, 150, and 5 K temperatures.

Heterogeneous Catalytic Click Chemistry

Huisgen 1,3-dipolar cyclo-addition reaction between terminal alkynes with organic azides to afford 1,4-regioisomers of 1,2,3-triazoles is an example of copper(I)-catalyzed click reaction and is known as CuAAC. The coupling product 1,4-diaryl-1,2,3-triazoles have been recognized for its anticancer activity.[24] For the click reaction of alkynes and azides, a number of homogeneous copper catalysts have been employed though these were not applicable in practical cases. In spite of important productivity, improved yield, high selectivity, and simplicity of optimization of the reported homogeneous catalysts, they have problems of product separation and purification from the catalyst, which creates environmental barriers to expand their scope, thus limiting their wide range of use in a number of applications. Trace amounts of catalyst removal from the end products is very important as metal contamination can create toxicity, especially in the pharmaceutical industry. Even by means of careful utilization of several methods such as chromatographic or distillation techniques, the extraction or elimination of trace amounts of the catalyst remains a challenge. Again, the utilization of heterogeneous and recyclable catalysts and eco-benign solvents for the coupling of three or more components in a single step is highly important and exciting in green chemistry. The starting reagents need not be completely soluble for the successful reaction. The end product can be filtered from the solution easily in many cases, as only the purification step is required. The use of water as an economical and safer solvent has a lot of advantages over expensive, combustible, and toxic organic solvents, diminishing atmospheric pollution and ease of separation of final products because most of the organic compounds are insoluble in water. Under such conditions, the heterogenization of click catalysis is an attractive alternative because of the benefits of easy removal, recovery, and reusability.

Metal NPs have received much attention as dynamic catalysts for various organic conversions and, in particular, because of enormous availability of copper(II) NPs, they have gained very good impact in catalysis and biology. In recent times, magnetically separable metal NPs have been industrialized to show potential for immobilization with the applications in catalytic transformations as readily available, very strong, and high-surface area heterogeneous catalysts due to their ease of separation after catalytic conversions to overcome time-consuming and difficult separation steps. Additionally, these catalysts can be used to give the desired products in aqueous conditions in very short reaction times with high yields.

Copper(I)-catalyzed azide–alkyne cycloaddition (CuAAC) is describe here with in situ generation of alkyl or aromatic azides from corresponding halide/boronic acid/epoxide and sodium azide and subsequently their coupling with terminal and internal alkynes in the presence of CuFe2O4 NPs. The click reaction of internal alkynes with azides, to afford trisubstituted triazoles, is very challenging.[13] However, in our hand, from the point of reactivity, there is no difference between terminal and internal alkynes. They are showing very similar reactivity with respect to yield and time.

To evaluate the scope of synthetic applicability and standardization, we selected benzyl bromide, sodium azide, and phenylacetylene as coupling partners for azide–alkyne cycloaddition (AAC) reactions. To optimize this facile AAC process, we struggled a lot and took several attempts. A typical optimization for coupling of phenylacetylene with benzyl azide is described here. To optimize the catalyst, we used commercial Fe2O3 and sonochemically prepared nano CuFe2O4. We found that efficient coupling can take place only by CuFe2O4 (Table ). Yield of the coupling product was found to be maximum in aqueous solvent rather than miscible organic solvent. Our initial success began with the reaction of benzyl bromide (4 mmol) and phenylacetelene (4 mmol) as reactants and CuFe2O4 as the catalyst (0.4 mmol, 10 mol %). In the first step, benzyl bromide (4 mmol), sodium azide (4 mmol), and CuFe2O4 (0.4 mmol) were mixed in water at RT and stirred for 4 h. After consumption of benzyl bromide, phenylacetylene (4 mmol) was added to the reaction mixture and stirred until it is completely consumed [thin layer chromatography (TLC) monitoring, hexane/ethyl acetate = 9:1]. The overall yield of the reaction is found to be 80% (Table , entry 3). A control reaction conducted in the same reaction condition but in the absence of the catalyst gave no yield of the product despite prolonged reaction time and higher temperature (Table , entry 1). When CuFe2O4 is replaced by Fe2O3 in similar reaction conditions, trace amounts of yield of the product could be obtained (Table , entry 2). Organic solvents such as dichloromethane, tetrahydrofuran, diethyl ether, ethanol and so forth proved to be less effective or ineffective. When the temperature of the reaction is increased to 80 °C, yield of the reaction is increased to 90% with reduced time (Table , entry 4). This reaction condition is found to be optimized. In the optimized reaction condition, we are pleased to choose phenylboronic acid, tolyl boronic acid, styrene oxide, allyl bromide and 4-bromo benzyl bromide, and so forth as coupling partners (Table ). Phenylacetylene can be replaced by 4-tert-butyl phenylacetylene, diphenylacetylene, o-nitrophenyl propargyl ether, propargyl alcohol, amylacetylene, or 1 heptynyltrimethylsilyl acetylene.

Table 1

Results for Different Conditions on the AACa

entry	catalyst	solvent	temp (°C)	time (h)	yield (%)
1		H2O	RT to 90	12
2	Fe2O3	H2O	RT to 90	12
3	CuFe2O4	H2O	RT	10	80
4	CuFe2O4	H2O	90	3	90
5	CuFe2O4	C2H5OH	70	10	30
6	CuFe2O4	CH3CN	90	10	15
7	CuFe2O4	toluene	110	10	10
8	CuFe2O4	acetone	70	10	trace
9	CuFe2O4	CH3OH	70	10	40
10	CuFe2O4	DCM	50	12	trace
11	CuFe2O4	THF	80	12	trace
12	CuFe2O4	Dioxane	90	12	trace
13	CuFe2O4	Et2O	70	12	trace

Conditions: halide (A) 4 mmol, NaN3 4 mmol, solvent 4 mL, catalyst (commercial Fe2O3/sonochemically prepared nano CuFe2O4) 0.4 mmol, acetylene (B) 4 mmol.

Table 2

Substrate Scope of Catalyzed AACa

Conditions: boronic acid/epoxide/halide 4 mmol, NaN3 4 mmol, solvent 4 mL H2O, CuFe2O4 0.4 mmol, acetylene 4 mmol, 80 °C.

Recycling the Catalyst and Leaching

Recycling of the catalyst can be easily achieved for this AAC reaction. The magnetically separated and recovered catalyst was washed with water and dichloromethane sequentially, dried, and used for the next cycle. No significant loss of activity was found up to the fifth cycle (Table ).

Table 3

Recycling of the Reagentsa

sl. no.	cycle no.	yield (%)
1	1st	90
2	2nd	88
3	3rd	86
4	4th	85
5	5th	84

Conditions: halide (A) 4 mmol, NaN3 4 mmol,H2O 4 mL, CuFe2O4 0.4 mmol, acetylene (B) 4 mmol.

A model reaction of phenylacetylene, benzyl bromide, and NaN3 was designed to investigate catalyst leaching during the reaction time. The reaction was carried out in optimized conditions. After half of the reaction time (90 min), the catalyst was separated magnetically. The rest of the reaction mixture (in the absence of the catalyst) was allowed to stir for another 90 min. As seen, no triazole was produced after catalyst removal. These results indicate that the catalyst is perfectly heterogeneous and catalyst leaching is negligible under these reaction conditions. In another experiment, the water extract of the reaction mixture after removal of the catalyst and organic layer was used for abovementioned model reaction. However, trace amounts of the desired triazole product were obtained. These experiments prove that there is negligible amount of leaching of our catalyst.

Conclusions

In this endeavor, we have presented the synthesis and characterization of bifunctional spin-canted CuFe2O4 NPs for use in magnetically separable heterogeneous catalytic green click chemistry. Phase pure, highly uniformed CuFe2O4 NPs have been synthesized by the sonochemical method and identified by PXRD and Raman studies. The crystal structure and inverse nature of this spinel have been determined by the Rietveld refinement method. Field emission scanning electron microscopy (FESEM) and HRTEM agree with the particle size of ∼15.3 nm evaluated from the Debye–Scherrer formula. The CuFe2O4 NPs show spin-canted behavior with the presence of both superparamagnetic and ferrimagentic phases, and it shows magnetic unsaturation even at 5 T. Because of its strong magnetization, we have successfully applied the material for magnetically separable heterogeneous catalysis for the three-component click chemistry in aqueous medium. A simple and efficient three-component green procedure has been developed by using the magnetically separable heterogeneous CuFe2O4 catalyst to synthesize 1,4 disubstituted 1,2,3-triazoles in excellent yield.

Experimental Details

Materials and Methods

Copper(II) chloride, dehydrate, ferric(III) chloride, and hexahydrate were purchased from Merck Chemical Company. All other chemicals used were of AR grade. Raman spectra in the wavenumber range of 100–800 cm–1 of the spinel ferrite were recorded at RT using a T64000 Raman spectrophotometer (J.Y. HORIBA) fitted with an argon ion laser using 514.5 nm as the exciting radiation in the Raman spectrometer that interfaced with the computer in the photon counting mode. The nanocrystalline CuFe2O4 sample was characterized by the PXRD technique, using a Bruker D8 ADVANCE diffractometer, with Cu Kα radiation (λ = 1.5406 Å) in the range of 2θ = 10°–80° in steps of 0.02°. To have an idea of microstructural information, the samples were characterized by HRTEM (JEOL, JEM-2011, operating at 200 kV, Japan with resolution = 1.9 Å). Static magnetic measurements, namely, magnetic hysteresis loops and temperature-dependent magnetization curves under ZFC and FC conditions were carried out by a SQUID magnetometer (Quantum Design, MPMS XL-7) in the temperature range of 5–300 K. Reactions were performed in oven-dried glassware. 1H and 13C{1H} NMR spectra were recorded in Bruker AVANCE III 400 and 300 MHz NMR spectrometers. Chemical shifts (δ) are expressed in ppm using the residual proton resonance of the solvent as an internal standard (CHCl3: δ = 7.26 ppm for 1H spectra, 77.36 ppm for 13C{1H} spectra, CHDCl2/CH2Cl2, 5.32 ppm; all coupling constants (J) are expressed in hertz and only given for 1H–1H couplings. The following abbreviations were used to indicate multiplicity: s (singlet), d (doublet), t (triplet), dd (doublet of doublets), dt (doublet of triplets), m (multiplet), bs (broad singlet), and bm (broad multiplet).

Synthesis of CuFe2O4 NPs

Two separately prepared 100 mL solutions of CuCl2·2H2O and FeCl3 were mixed together by intense ultrasonication for 30 min at RT. Now, 6 N NaOH solution was added dropwise to the homogeneous mixture to maintain pH ≈ 9 for precipitation. After 2 h, complete coprecipitation was obtained, and then, the coprecipitated particles were rigorously stirred for another 2 h. Finally, the coprecipitated particles were filtered by the vacuum filtration method, and the residue was washed several times by deionized triple distilled water and propanol to neutralize the pH and remove extra ions. The filtered coprecipitated particles were dried at 80 °C for 24 h in a hot chamber and then it was annealed at 600 °C for 6 h in vacuum to obtain the desired pure crystalline phase.

General Procedure for Catalytic 1,3-Dipolar Cycloaddition

The catalytic reactions were carried out in a glass batch reactor according to the following procedure. To a stirred solution of the desired halide/epoxide/boronic acid (4 mmol) in 4 mL water, 4 mmol sodium azide and 0.4 mmol CuFe2O4 were added, and the reaction mixture was stirred at 80 °C for 1 h and monitored continuously by TLC (hexane/ethyl acetate = 9:1). The reactions were performed in open air. After consumption of the starting material, acetylene (4 mmol) was added dropwise and stirred for 1–3 h. After completion of the reaction mixture, the catalyst was separated and recovered by a strong magnet, and the remaining mixture was diluted with water and extracted with ethyl acetate thrice. The combined organic layer was dried over anhydrous Na2SO4, filtered, and concentrated to afford the crude material. The crude material was then purified by column chromatography (silica 100–200 mesh, ethyl acetate–hexane 5–10%) to afford the desired triazole product (yield = 55–95%). The magnetically separated and recovered catalyst was washed with water and dichloromethane sequentially, dried, and used for the next cycle.