The CuFe2O4@SiO2@ZrO2/SO42−/Cu nanoparticles: an efficient magnetically recyclable multifunctional Lewis/Brønsted acid nanocatalyst for the ligand- and Pd-free Sonogashira cross-coupling reaction in water

Herein, the synthesis and application of copper-incorporated sulfated zirconium oxide supported on CuFe2O4 NPs (CuFe2O4@SiO2@ZrO2/SO42−/Cu NPs) as a novel Lewis/Brønsted acid nanocatalyst were studied for the Sonogashira C–C cross-coupling reaction. The fabricated CuFe2O4@SiO2@ZrO2/SO42−/Cu catalyst exhibited efficient activity for a large variety of aryl iodides/bromides and, most importantly, aryl chlorides in water and in the presence of NaOH as a base in short reaction times. The catalyst was fully characterized by FTIR, TG-DTG, VSM, XRD, EDX, FE-SEM and TEM analyses. A synergetic effect could be considered to have arisen from the various Lewis acid and Brønsted acid sites present in the catalyst. The efficient incorporation of copper into zirconia provided a robust highly stable hybrid, which prevented any metal leaching, whether from the magnetite moiety and/or Cu sites in the reaction mixture. Moreover, the catalyst was successfully recovered from the mixture by a simple external magnet and reused for at least 9 consecutive runs. Zero metal leaching, stability, consistency with a variety of substrates, fast performance, cost-effectiveness, environmental friendliness, and preparation with accessible and cheap materials are some of the advantages and highlights of the current protocol.


Introduction
The Sonogashira C-C cross-coupling reaction is one of the most applicable types of C-C cross-couplings, which involves the coupling of vinyl or aryl halides or triates with terminal alkynes (C sp 2 -C sp ); 1,2 since its vital application for the construction of complex biological and pharmaceutical molecules from simple precursors, the Sonogashira reaction has had signicant importance in the eld of synthetic organic chemistry. 3 The reaction was rst developed using Pd and Cu as a catalyst and a co-catalyst, respectively (Scheme 1). Since the discovery of this reaction, various methodologies have been developed to resolve its impediments. Palladium is a toxic, rare and expensive transition metal that is mainly used along with toxic and air-sensitive phosphine ligands. 4 More importantly, the presence of copper as a co-catalyst promotes the Glaser-type homo-coupling of terminal acetylenes to generate a by-product (Scheme 1); thus, various attempts have been developed to perform the reaction under copper-free and Pdfree conditions in a mild, safe, ecofriendly, and cost-effective manner; in this regard, one strategy is the use of cheaper and safer alternative transition metals including Ni, 5 Cu, 6 Fe, 7 and Co. 8 Among these, the potential of Cu for application in the C-N as well as C-C cross coupling reactions is well-known; 9 moreover, several achievements have been reported for the Cu-catalyzed Sonogashira reaction; the recent examples include the use of Cu 2 O/RGO, 10 Cu/ Mn bimetallic, 9 CuI/PPh 3 /K 2 CO 3 , 11 CuI/K 3 PO 4 /1,4-dioxane, 12 and Au$CuFe 2 O 4 @silica as catalysts for this reaction. 2 Recently, Sun and coworkers 13 have reported the application of a Cu-MOF derived from two-phase Cu/Cu 2 O-rGO as an efficient catalyst for the Sonogashira reaction.
However, various drawbacks, including harsh reaction conditions, long reaction times, use of expensive and toxic materials, lack of selectivity, lack of environmental sustainability, and low reaction yields, especially for aryl chlorides, are still present in most of the reported protocols; therefore, the development of a promising alternative method is required.
Zirconia is one of the most well-known promising solid acids with signicant catalytic activity. It is widely used as an efficient acid catalyst in oil reneries and petrochemical industries for processes such as hydrocarbon conversion, alkylation, cracking, Friedel-Cras acylation, esterication and isomerization; 14,15 moreover, the activity of zirconia can be largely promoted by its treatment with sulfate groups, and as a result, sulfated zirconium oxide (ZrO 2 /SO 4 2À ) is obtained. The high thermal stability, outstanding catalytic activity, high acidity, stability in various organic solvents, and durability under harsh reaction conditions are some of the notable and applicable properties of sulfated zirconia that make it a suitable support for more modications (an objective of this study) and/or catalytic aspects. 16 Various catalytic activities, such as towards benzylation, 16 multicomponent reactions, 17 and synthesis of dioxane, 18 of ZrO 2 /SO 4 2À have been reported in the literature; in addition, heterogeneous solid supports can be magnetized by magnetic nanoparticles (MNPs) to make these supports magnetically recoverable; 19 moreover, due to their high aspect ratio, MNPs can strongly improve the catalytic activity of a catalyst. 3 In this study, we introduced copper-incorporated sulfated zirconium oxide supported on CuFe 2 O 4 nanoparticles as an efficient, recyclable and durable magnetic nanocatalyst for the rst time for the C-C cross coupling reaction of phenylacetylene with aryl iodides, aryl bromides and aryl chlorides under mild reaction conditions. The present system not only benets from the durable ZrO 2 /SO 4 2À solid support, but also the magnetic CuFe 2 O 4 magnetic core in the catalyst provides suitable recyclability to the catalyst via an external magnet.

Instrumentation and materials
All chemicals were freshly purchased from Sigma and Merck or Fluka Chemical Companies with no further purication. All solvents were distilled under a N 2 atmosphere and dried before use. The reaction progress was monitored by thin layer chromatography (TLC). The FTIR spectra were obtained via the JASCO FT/IR 4600 spectrophotometer using KBr pellets. The 1 H NMR (250 MHz) and 13 C NMR (62.9 MHz) spectroscopies were performed by the Bruker Avance DPX-250 spectrometer in CDCl 3 and DMSO-d 6 as solvents, respectively. TMS was used as an internal standard. Mass spectrometry was performed using the Thermolyne 79300 model tube furnace equipped with the MKS gas analyzer coupled to a quadrupole mass selective detector. The scanning electron microscopy images (FE-SEM) were obtained by the TESCAN MIRA3 apparatus. Transmission electron microscopy (TEM) was conducted using the Philips EM208 microscope at 100 kV. The magnetic behavior of the samples was investigated using the Lake Shore Cryotronics 7407 vibrating sample magnetometer (VSM) at room temperature. EDX spectroscopy was performed using a eld-emission scanning electron microscope (FESEM, JEOL 7600F), equipped with an X-ray energy dispersive spectrometer obtained from Oxford instruments. /Cu magnetic nanocatalyst (0.005 g, 0.3 mol% Cu), NaOH (1.0 mmol) and water (1.0 mL) was stirred at 60 C in an oil bath. The reaction progress was monitored by TLC. Upon completion of the reaction, the catalyst was magnetically removed, and the mixture was extracted with 10 mL of Et 2 O. The combined organic layers were dried over MgSO 4 , and then, the solvent was removed under reduced pressure. The desired pure coupling product was obtained by ash chromatography of the crude product.

Catalyst characterization
The FTIR spectra of Zr(OH) 4  /Cu are shown in Fig. 1A(a). The broad and resolved peak at 3403 cm À1 related to the O-H stretching vibration conrmed the hydration of zirconium chloride using ammonia. Moreover, the absorption peaks related to Zr-O-Zr could be seen at 640-750 cm À1 (Fig. 1A(a)). The FTIR spectrum of ZrO 2 /SO 4 2À demonstrated characteristic peaks at 1143, 1044, and 994 cm À1 (as a shoulder), which corresponded to the asymmetric or symmetric stretching vibrations of the S]O or S-O bonds ( Fig. 1A(b)). 23 These vibrations were characteristic for the bidentate sulfate ions coordinated to a metal cation. The series of peaks at 467-747 cm À1 were assigned to the Zr-O-Zr asymmetric stretching vibrations. 24 Moreover, the broad peak at 3421 cm À1 and the medium peak at 1636 cm À1 were assigned to the O-H stretching and bending vibrations of the adsorbed and/ or coordinated water by the sulfate groups, respectively. 25 A sharp peak near 500 cm À1 was attributed to the Cu-O stretching vibration, demonstrating that the incorporation of the Cu ions took place through the oxygen atoms of the sulfated ions in ZrO 2 /SO 4 2À /Cu; moreover, this subsequently conrmed the successful coordination of the Cu cations to the catalyst framework ( Fig. 1A(c)). 26 The stretching vibrations related to Zr-O-Zr were covered due to this strong absorption.
In the CuFe 2 O 4 FTIR spectrum, two absorptions at 1629 and 3435 cm À1 represented the H-O-H bending and free O-H stretching vibrations, respectively, due to the water molecules adsorbed on the surface of the CuFe 2 O 4 NPs with high aspect ratio. 27 The two absorption bands at 476 and 590 cm À1 were assigned to the Cu-O and Fe-O stretching vibrations, respectively ( Fig. 1A(d)). 27,28 The strong absorption at 1093 cm À1 (Si-O vibrations) conrmed the successful coating of the CuFe 2 O 4 NPs with a silica shell (Fig. 1A(e)).
The presence of vibration bands at 421, 575, and 870-1148 cm À1 , which were due to Fe-O, Cu-O, and Si-O-Si, respectively, demonstrated that ZrO 2 /SO 4 2À /Cu was successfully supported on CuFe 2 O 4 @SiO 2 ( Fig. 1A(f)). In addition, the presence of several bands with medium intensity in the 1361-1641 cm À1 region was allocated to the ZrO 2 /SO 4 2À stretching vibrations ( Fig. 1A(f) 1B(a)). 21,29 More precisely, a mixture of the monoclinic and tetragonal phases was observed in the spectrum ( Fig. 1B(a), stars and diamonds represent a tetragonal and monoclinic structure, respectively) that was in agreement with the reported ZrO 2 /SO 4 2 crystal structure. 21,29 The peaks with lower intensities at 2q ¼ 24.1 and 28.3 were assigned to the monoclinic structure of zirconia. Note that the presence of sulfated groups does not lead to a phase change of zirconia; this is may be due to strong interactions between zirconia and the sulfate ions. 29 (113), respectively, which corresponded to the thermally prepared CuO powder structure ( Fig. 1B(b)). 30,31 Furthermore, the peaks with much lower intensities near to baseline indicating the crystal structure of the sulfated zirconium oxide ( Fig. 1B(b)). The preparation of ZrO 2 /SO 4 2À was conrmed by the presence of the Cu, Zr, O, and S elements, which were detected by EDX analysis (Fig. 2a). Moreover, the presence of these elements in the catalyst was investigated and conrmed by EDX analysis. As shown in Fig. 2b, the elements Zr, Cu, O, S, Fe, and Si were detected in the catalyst. The thermal behavior of ZrO 2 /SO 4 2À and the catalyst is shown in Fig. 3a. ZrO 2 /SO 4 2À showed a signicant thermal stability, and only a 7.5% weight loss was observed in the temperature range of 25-1000 C (Fig. 3a). This degradation occurred in four steps, where the rst and second steps were assigned to the loss of the physically adsorbed water from the catalyst surface (0.26% weight loss at 210 C), and the escape of the trapped water in the catalyst network by sulfate groups (1.19% weight loss at 350 C), respectively. The third weight loss in the temperature range of 530-665 C was related to the oxidation of copper and the formation of CuO. 32 The weight loss that appeared in the temperature range of 680-860 C was due to the decomposition of sulfate as well as structural OH À groups. [33][34][35] The decoration of ZrO 2 /SO 4 2À /Cu on CuFe 2 O 4 @SiO 2 improved the thermal stability of the catalyst, and only 6.5% weight loss was observed until 1000 C. The rst weight loss with a mild slope, which lasted till 780 C, was due to the loss of the adsorbed water in the crystal structure of the catalyst. The subsequent weight loss was attributed to the decomposition of incorporated Cu and sulfate groups with a 4.0% weight loss (Fig. 3b). The magnetic properties of CuFe 2 O 4 and CuFe 2 O 4 @SiO 2 @-ZrO 2 /SO 4 2À /Cu were studied by VSM analysis (Fig. 4). As shown in Fig. 4, the samples represented a superparamagnetic behavior with no hysteresis loops in their spectra. The saturation magnetization for CuFe 2 O 4 was found to be 24.6 emu g À1 (Fig. 4a). This amount was largely reduced to 10.1 emu g À1 for CuFe 2 O 4 @SiO 2 @ZrO 2 /SO 4 2À /Cu; this strongly conrmed its surface functionalization (Fig. 4b) Fig. 5a and b). 36 By comparing their TEM images ( Fig. 5c and d), this agglomeration was more clearly observed. According to Fig. 5c

Optimization of reaction parameters
To nd premium reaction conditions for the Sonogashira C-C coupling reaction, the reaction of iodobenzene with phenylacetylene was chosen as the model reaction. The effects of the reaction parameters, such as the type of base, reaction temperature, solvent, and catalyst amount, were studied. The results are presented in Table 1. The reaction obviously proceeded in polar-protic solvents such as EtOH, MeOH and water (Table 1, entries 1, 8, and 12). These results were in agreement with the structure of the catalyst containing hydrophilic groups as well as the mechanism proposed in the next section. Other   Paper solvents provided low to moderate yields. The highest efficiency was obtained in water aer reaction for 30 min in the presence of 0.005 g of catalyst (entry 12, 92%). There was no satisfactory conversion under solvent-free conditions ( Table 1, entry 11). NaOH and KOH were found to be efficient bases for this transformation (Table 1, entries 12 and 13). Moreover, 60 C and 0.005 g of the catalyst were the premium temperature and catalyst amount for the model reaction, respectively (Table 1, entries 12, 21-26).
The scope of the reaction was investigated and extended with a variety of aryl halides and phenylacetylene in the presence of CuFe 2 O 4 @SiO 2 @ZrO 2 /SO 4 2À /Cu NPs as the catalyst under the previously obtained optimum conditions. As shown in Table 2, the method tolerated various substrates bearing either electrondonating and/or electron-withdrawing substituents, and highto-excellent yields were obtained for all substrates (Table 2). Generally, the substrates with electron-withdrawing substituents provide higher efficiencies than others in terms of time and yield ( Table 2, for example entries 5, 7, and 13). Moreover, iodide as a leaving group accelerated the reaction than Br or Cl. The results are in agreement with an oxidative addition/ reductive elimination mechanism, which has been discussed hereinaer.

Mechanistic study
At   . The coordinated Cu and zirconium were efficient active Lewis sites. Moreover, water was coordinated through an interconversion reaction between free sulfate groups on the catalyst, and this provided active Brønsted acid sites. The presence of water as a solvent promoted the active Brønsted acid sites (Scheme 3); this explained the high catalytic activity of the catalyst with water as a solvent. Due to the presence of these catalytic active sites in CuFe 2 O 4 @SiO 2 , ZrCl 4 , and ZrO 2 /SO 4 2À , a synergetic effect could be speculated for this catalyst, arising from the Cu sites, Zr sites, sulfate groups, 38 coordinated water, 21 and CuFe 2 O 4 . 2 A plausible structure for the catalyst is shown in Scheme 4, which is in agreement with the characterization data as well as the structure proposed in literature. 2,21,38 In the rst step of the proposed mechanism, Cuacetylide (Scheme 4, intermediate I) was formed via oxidative addition with the participation of a base. This addition could be mediated by electron transfer from zirconium to copper (from Cu I to Cu II for example, see Scheme 3). To prove this claim, the Sonogashira reaction was performed in the presence of CuSO 4 under the same reaction conditions. No coupling products were found in the mixture. However, it could be concluded that the presence of zirconium in the catalyst was mandatory for electron transfer. A water molecule was formed during this transformation. The hydrophilic nature of the catalyst surface arising from the sulfate groups increased the solubility of the base. Due to interconversion between sulfate groups (Schemes 3 and 4),   According to the proposed catalyst structure shown in Scheme 4, the catalyst can provide a suitable medium for conducting the reaction in water. As shown in Scheme 5, with respect to the nanocomposite structure of the catalysts, organic compounds were introduced into the catalyst by removing water, which contained catalytic active sites including copper and zirconium; as the concentration and number of effective collisions increased, the reaction proceeded with high efficiency. This structure not only addressed the concerns about the transfer of mass in an aqueous medium, but was also consistent with the high efficiencies achieved for the Sonogashira compounds in this study. In the end, the desired product was removed from the medium. This rigid intermediate also prevented the formation of diyne by-products, which were produced by the coupling of two equivalents of Cu-acetylide in the presence of molecular oxygen (Scheme 1, Glaser type reaction). 6

Recoverability studies
Stability, durability and, consequently, recyclability of a heterogeneous catalyst are prominent and important factors from economical, energy saving, and environmental points of view; 39-41 the rigid inorganic structure of the sulfated zirconium oxide solid support along with the magnetic properties of the CuFe 2 O 4 moieties made the catalyst recoverable and reusable and minimized any metal leaching. The recyclability of the catalyst was investigated in the Sonogashira cross-coupling reaction of phenylacetylene and iodobenzene in the presence of NaOH at 60 C. The catalyst was recovered in each cycle, washed with EtOH (2 Â 5 mL) and reused in the next run without any purication or pre-activation. Fig. 6a shows the corresponding results for nine consecutive runs, and an insig-nicant loss in efficiency (catalyst yield and reaction yield) was observed.
The yield of the coupling product reached 89% (3% loss) aer the 9 th run. There was also a very intransigent increase in the reaction time until the 9 th cycle. The results suggested a rigid and durable structure for CuFe 2 O 4 @SiO 2 @ZrO 2 /SO 4 2À / Cu as a heterogeneous nano-catalyst. Furthermore, to elucidate the chemical structure as well as stability of the catalyst, the catalyst recovered aer the 9 th run was studied by FTIR, FE-SEM and TEM analyses (Fig. 6b-d). Aer comparing the FTIR spectrum of the recovered catalyst with the corresponding FTIR spectrum of the fresh catalyst, it was determined that the structure of the catalyst remained intact during the recycles (Fig. 6b). Moreover, the FE-SEM and TEM images of the recovered catalyst revealed that the morphology of the nanoparticles was the same as revealed in the corresponding images of the fresh catalyst ( Fig. 6c and d), respectively. No agglomeration or increase in the particle size was observed even aer nine consecutive recycles. Note that the catalyst did not show any detectable metal leaching even aer the 9 th run. ICP analysis of the residue obtained from the mixture aer the 9 th run was performed to separately investigate the presence of Fe, Cu, and Zr; for each experiment, a negligible amount of these elements was detected, which conrmed the heterogeneous nature as well as durability of the catalyst during the reactions (Table S1 †).
The heterogeneous nature of the catalyst was studied by a hot ltration test. 42 The aforementioned model reaction was applied for this test. The catalyst was magnetically removed aer 10 min of the reaction (30% yield, GC analysis). The reaction was allowed to proceed, and the conversion was investigated aer 2 h by GC. The reaction conversion reached 33%, which conrmed that CuFe 2 O 4 @SiO 2 @ZrO 2 /SO 4 2À /Cu operated heterogeneously in the mixture, and no metal leaching took place during the reaction.
We compared the catalytic activity of CuFe 2 O 4 @SiO 2 @ZrO 2 / SO 4 2À /Cu with those reported for the Sonogashira coupling reaction of phenyl acetylene with 4-Me-iodobenzene, 4-NO 2bromobenzene, and 4-MeO-iodobenzene. As shown in Table 4, the present methodology was superior to all the reported catalytic systems in terms of time, catalyst amount and yield of the reaction. Evidently, the reaction conditions were very mild, and the heterogeneous catalyst compromised some advantages such as easy preparation and recycling, minimum metal contamination and economic friendliness.

Conclusion
Herein, copper was incorporated into sulfated zirconium oxide (ZrO 2 /SO 4 2À /Cu) supported on copper ferrite nanoparticles (CuFe 2 O 4 NPs); the resultant compound was found to be an efficient magnetically durable catalyst for the Sonogashira reaction in water. The catalyst demonstrated high efficiency not only for aryl iodides but also for aryl bromides and aryl chlorides. Note that the catalytic activity of the modied sulfated zirconium oxide in the organic synthesis has been rarely studied. The catalyst has a monoclinic-tetragonal mixed crystal structure, high thermal stability until 1000 C, and a 10 emu g À1 saturation magnetization with a 40 nm average size and an irregular shape. The catalyst was further characterized by the EDX and FTIR analyses. This magnetic nanocatalyst could be recycled for at least 9 consecutive runs without any notable loss in activity. The study on the recovered catalyst revealed the high stability and durability of the proposed catalyst. The control experiments completely rule out the synergetic effect of ZrO 2 / SO 4 2À /Cu and CuFe 2 O 4 @SiO 2 , which leads to the incredible catalytic activity of the proposed catalyst; in literature, the interference of CuFe 2 O 4 @SiO 2 in the Sonogashira reaction has also been demonstrated. An electron-transfer between Cu and Zr metal sites could be responsible for the proposed oxidative addition and reductive elimination mechanism, in agreement with literature. Furthermore, an interconversion between the sulfate ions on the catalyst surface mediated/facilitated the function of the base in water via adsorption of the cation. The current methodology can indeed replace the expensive Pdbased catalytic systems with highly toxic and expensive phosphine ligands to catalyze the Sonogashira cross-coupling reactions. The use of water as a solvent, short reaction time, high efficiency and absence of by-products are other advantages of the abovementioned catalyst.

Conflicts of interest
There are no conicts to declare.