Introduction

Nanoparticles (NPs) have attracted great attention because of their unique optical, electrochemical, medical and catalytic properties1,2,3,4,5,6,7,8. Indeed, the capacities and applications of NPs, as sorbent, sensor, heterogeneous catalyst, etc, are limited due to their low sustainability and dispersibility9,10,11, deactivation or constant leaching, and low recyclability12. Fortunately, the properties of NPs are affected seriously by support materials such as inorganic porous materials mainly silica or alumina13, organic polymers or biopolymers14,15, dendrimers16, carbon nanotubes17, graphene oxide18, etc., which are essential for their applications specially in heterogeneous catalysis19,20,21. Because of natural abundance, low toxicity compared to other transition metals and redox potential, copper is an appropriate metal used in nature for many oxidation reactions22,23,24. Therefore, the immobilization of Cu(II) and magnetite NPs on multifunctional group supports can be considered as a feasible way to address above mentioned problems and sustainable chemistry principles11,25. In this way, supramolecular chelation of Cu(II) species onto the surface of bifunctional modified chitosan with enhanced catalytic activity would be very desirable. This is a successful example of the combination of chemo- and bio-drived materials catalysis for mimicing biocatalysis as well as sustainable and one pot multistep synthesis26,27,28,29,30,31,32. Indeed, one-pot multistep reactions which are also known as cascade reactions, do not require the isolation of intermediates and reduce the solvent wastes. Hence, cascade reactions have attracted a great deal of attention in both academia and industry in recent decades29,33,34,35,36. These reactions afford desired products in high to excellent yeilds and hence properly adress green and sustainable chemistry principles. To date, most of the reported catalysts for one-pot cascade reactions are homogeneous, which typically suffer from product pollution and poor recyclability. Therefore, designing and introducing multifunctional catalysts with different catalytically active sites for one-pot cascade reactions is still a serious challenge.

Cyanocinamonitriles, which are prepared by the Knoevenagel condensation of a carbonyl group and C–H acids, serve as key intermediates in the synthesis of many important fine chemicals37,38,39,40,41,42. Nowadays, the tandem oxidation of alcohols and Knoevenagel condensation reaction of the obtained aldehydes and active methylene compounds, as an economic and eco-friendly procedure towards cyanocinamonitriles, has also received a great attention from green and sustainable stand points of view35,43,44,45,46,47. Traditionally, stoichiometric amounts of oxidants such as permanganates48, chromium reagents49 or the Dess-Martin periodinane40 have been used for oxidation reactions with significant environmental impacts due to their toxicity and producing large amounts of waste. To address the aformentioned concerns, we were interested to modify chitosan, as a naturally abundant biopolymer with suitable ligands to afford multiple active sites50. In the previous repots, we succeeded in covalentlly modification of chitosan with melamine and exploring of the activity of obtained Cs-Pr-Me as a bifunctional bio-derived organocatalyst51,52. To our delight, the Cs-Pr-Me materials, with plentiful amino and hydroxyl groups in a proper geometry, was found to be a suitable support for the immobilization of Cu(II) and magnetic nanoparticles. Due to the economic and sustainable benefits of magnetic copper-based catalysts25 and in continuation of our ongoing efforts towards developing efficient and novel heterogeneous catalysts15,51,52,53,54,55,56, we wish herein to report preparation and characterization of the supramolecular Cu(II) and magnetite nanoparticles decorated melamine-functionalized chitosan (Cs-Pr-Me-Cu (II)-Fe3O4), as a recyclable catalyst, for green cascade oxidation/Knoevenagel condensation. Furthermore, the use of TBHP, as a green oxidant, is a reasonable choice55 (Scheme 1).

Scheme 1
scheme 1

Oxidation/Knoevenagel condensation of different benzyl alcohol derivatives with malononitrile in the presence of the Cs-Pr-Me-Cu(II)-Fe3O4 (1).

Full size image

Results and Discussion

The Cs-Pr-Me materials were prepared by grafting of melamine to the chitosan backbone using 1,3-dibromopropne, as an appropriate linker, according to our published procedures51,52. In the next step, immobilization of the Cu(II) species on the Cs-Pr-Me backbone was achieved using an aqueous solution of Cu(OAc)2. Then, the surface of Cs-Pr-Me were decorated in situ by magnetite nanoparticles. The obtained Cs-Pr-Me-Cu(II)-Fe3O4 materials were characterized in details with various analysis techniques and methods such as Fourier transform infrared (FTIR) spectroscopy, field emission scanning electron microscopy (FESEM), transmission electron microscopy (TEM), inductively coupled plasma (ICP), X-ray diffraction (XRD), energy dispersive X-ray (EDX) mapping analysis, and vibrating sample magnetometer (VSM).

The FTIR spectra of Cs-Pr-Me-Cu(II)-Fe3O4 (1) demonstrated the evidence for existence of Cu(II) and magnetite nanoparticles on the melamine-functionalized chitosan backbone. Fig. S1 illustrates the FTIR spectra of Cs-Pr-Me (a), fresh Cs-Pr-Me-Cu(II)-Fe3O4 (1, b) and Cs-Pr-Me-Cu(II)-Fe3O4 after four times recycling (1, c), respectively (See Electronic Supplementary Information). As shown in Fig. S1a, the Cs-Pr-Me showed the adsorbtion bands at 3567, 3453, 3450–3100 and 1390 cm−1, which are attributed to the asymmetric and symmetric stretching vibrations of N–H bonds and the stretching vibration of O–H bonds of chitosan. Furthermore, the signals at 1616 and 1593 cm−1 are assigned to the NH2 and NCN bending vibrations of melamine. The OH bending vibration of chitosan corresponds to the broad signal in the range of 1470–1330 cm−1. After formation of Cs-Pr-Me-Cu(II)-Fe3O4 (1), the above band intensities were decreased as shown in Fig. S1b, which represent Cu(II) and magnetite nanoparticles have been coordinated by the amino and hydroxyl functional groups of the Cs-Pr-Me materials. In addition, the characteristic bands for Fe-O and Cu-O stretching vibrations were observed at 560 and 430 cm−1, respectively.

The morphological features and size of Cu(II) and magnetic iron oxide NPs were examined by FESEM and TEM experiments (Figs. 1 and 2). The FESEM images of the commercial chitosan and Cs-Pr-Me befor modification have been shown in Fig. 1a,b, respectively. Furtthermore, Fig. 1c,d show the FESEM images of Cs-Pr-Me-Cu(II)-Fe3O4 (1) materials. The Cu(II) and Fe3O4 nanoparticles are approximately spherical and scattered with an average diameter of about 23–51 nm. On the other hand, TEM images (Fig. 2) obviously demonstrate almost uniform decoration of Cu(II) species and magnetic nanoparticles on the melamine-functionalized chitosan support. Also, according to ICP analysis, the percentage of chelated copper and iron in the fresh Cs-Pr-Me-Cu(II)-Fe3O4 (1) was found to be 1.38 wt% (Cu) and 25.72 wt% (Fe), respectively. The EDX spectrum of the Cs-Pr-Me-Cu(II)-Fe3O4 (1) is shown in Fig. 3. The EDX spectrum indicates that the introduced catalyst 1 materials are composed of Cu, Fe, O, N and C elements. In addition, EDX mapping was performed to observe distributions of the elements in the Cs-Pr-Me-Cu(II)-Fe3O4 (1) nanocomposite. As it can be seen in Fig. 3, the indicated elements, especially Cu and Fe, demonstrate uniform distributions. These obtained data strongly confirm successful immobilization of the Cu(II) species and iron oxide nanoparticles on the Cs-Pr-Me backbone. On the other hand, the magnetic properties of the Cs-Pr-Me-Cu(II)-Fe3O4 (1) was measured using vibrating sample magnetometer (VSM) at room temperature. As it was seen in Fig. S5, the value of the saturation magnetization was 67 emu/g for the Cu(II) species and magnetic nanoparticles decorated melamine-functionalized chitosan (1, See Electronic Supplementary Information).

Figure 1
figure 1

FESEM images of the commercial chitosan (a), Cs-Pr-Me materials befor modification (b), and Cs-Pr-Me-Cu(II)-Fe3O4 (1) materials at 500 (c) and 200 nm (d) scales.

Full size image
Figure 2
figure 2

TEM images of the Cs-Pr-Me-Cu(II)-Fe3O4 (1) nanomaterials.

Full size image
Figure 3
figure 3

Energy dispersive spectroscopy (EDX) pattern and elemental mapping of the Cs-Pr-Me-Cu(II)-Fe3O4 (1) nanocomposite.

Full size image

Also, Fig. 4 shows the XRD patterns of the fresh Cs-Pr-Me-Cu(II)-Fe3O4 (1) materials and the recycled sample. The XRD patterns of the melamine and chitosan are also illustrated for comparison as offset patterns. The diffraction peaks at 2θ values of 43.30, 50.40 and 74.15° can be attributed to the reflections of cubic Cu ((JCPDS No. 04–0836, marked with ●). Moreover, the peaks at 2θ values of 30.20, 35.39, 36.89, 53.31, 56.98, 73.91° can be assigned to the reflections of cubic Fe3O4 (JCPDS No. 019-0629, marked with ▲). On the other hand, the well-defined high intensity diffraction signals (2θ) at 13.41, 17.95, 21.65, 22.25, 26.28, 28.90 and 29.91° are in accordance with the monoclinic crystal system of melamine (JCPDS no. 024-1654). Furthermore, the lower intensity for the diffraction peaks of Cu and Fe3O4 may be due to their lower weights compared to that of Cs-Pr-Me support in the structure of nanocomposite 1.

Figure 4
figure 4

XRD patterns of the fresh Cs-Pr-Me-Cu(II)-Fe3O4 materials (1, a), recycled sample after five runs (b), melamine (c) and commercial chitosan (d).

Full size image

Cs-Pr-Me-Cu(II)-Fe3O4 nanomaterials-promoted cascade oxidation/Knoevenagel condensation for the synthesis of α,β-unsaturated nitriles 4a-f

To evaluate the catalytic activity of Cs-Pr-Me-Cu(II)-Fe3O4 (1), the one-pot oxidation/Knoevenagel condensation between benzyl alcohol (2a) and malononitrile (3) in CH3CN was chosen as the model reaction. As the information in Table 1 show, in the absence of the catalyst 1 and using different oxidants including TBHP, H2O2, O2, and air, no condensation product, 2-benzylidinemalononitrile (4a), was formed at 80 °C. However, 58%, 75% and 20% conversion to benzoic acid 6a was observed when TBHP, H2O2, and O2 were used, respectively. On the other hand, only trace amounts of bezaldehde 5a and benzoic acid 6a were formed when the reaction mixture was subjected to an air flow (Table 1, entries 1–4). Interestingly, 23% of the desired benzylidinemalononitrile 4a was observed by employing 10 mg of Cs-Pr-Me-Cu(II)-Fe3O4 (1), as a catalyst, without any oxidant (Table 1, entry 5). Among TBHP and H2O2, the former was found to be more effective for the reaction (Table 1, entries 6–7). Indeed, H2O2 afforded lower yeild of the desired product 4a compared to TBHP. In fact, the hydroxyl radical produced by H2O2 is a more powerful oxidant compared to the t-BuOO radical generated by TBHP during the reaction. It has been reported before that the OH radical reacts with polysaccharides such as chitosan to depolymerize them and forming chitosan chains with lower molecular weights57. Therefore, the lower yeild of the desired product 4a can be attributed to the higher tendency of OH radicals to depolymerize the chitosan cahins in the structure of the catalyst 1 rather than oxidation of benzyl alcohol (Table 1, entry 7). Hence, TBHP was used in the next optimization experiments. To our delight, by reducing the reaction tempereture, the desired product 4a was formed in the same yeilds at 60 °C and room tempereture, however longer times were required (entries 8,9). Hence, room temperature was chosen as a sustainable conditions for the reaction although it requires a longer time. Furthermore, the nature of the solvent showed a significant impact on the oxidation of benzyl alcohol (2a) and subsequent Knoevenagel condensation. For instance, tolouene and H2O afforded the desired product 4a in lower yields compared to CH3CN under the same catalyst loading even after longer times (Table 1, entries 10,11). Upon increasing of the catalyst loading from 5 to 20 mg, the conversion of benzyl alcohol 2a to 2-benzylidinemalononitrile (4a) considerably increased (Table 1, entries 7, 12–14). It is noteworthy that the Cs-Pr-Me-Cu(II) materials afforded a lower yeild compared to the Cs-Pr-Me-Cu(II)-Fe3O4 (1) under the same conditions (entries 14, 15). This can be attributed to the involvement of Fe(III) species in the catalytic cycle (See Scheme 2).

Table 1 Screening of different conditions on the oxidation/Knoevenagel condensation products of benzyl alcohol (2a)a.
Full size table
Scheme 2
scheme 2

Plausible mechanism for the oxidation/Knoevenagel condensation of different benzyl alcohols 2a-f catalyzed by Cs-Pr-Me-Cu(II)-Fe3O4 (1).

Full size image

In the next step, the effect of TBHP oxidant equivalent against benzyl alcohol (2a) was systemically investigated in 4 h intervals. The results have been shown in Fig. 5. The results of this part of our study showed that one equivalent of TBHP oxidant afforded the desired oxidation/Knoevenagel product 4a with 100% conversion after 8 h. On the other hand, higher or lower equivalents of TBHP oxidant produced lower yeilds of the desired oxidation/Knoevenagel product 4a. Encouraged by these results, the substrate scope of this oxidation/Knoevenagel condensation was studied in the next step under the optimal conditions. Table 2 shows the summerized results.

Figure 5
figure 5

The effect of different TBHP oxidant: benzyl alcohol (2a) mole ratios on the yield of oxidation/Knoevenagel condensation product 4a.

Full size image
Table 2 Scope of the cascade oxidation/Knoevenagel condensation of different benzyl alcohols 2a-f catalyzed by Cs-Pr-Me-Cu(II)-Fe3O4 (1) under optimized conditionsa,b.
Full size table

As revealed in Table 2, the reaction conditions were compatible with both electron-withdrawing and electron-donating substituents at p- as well as o- positions of the aromatic ring. Intrestingly, 2-benzylidenemalononitrile (2a) could be obtained from corresponding substrate in 100% conversion. Furthermore, alcohols such as p-hydroxybenzyl alcohol (2d) and p-nitrobenzyl alcohol (2e), reacted slowly to form the corresponding aldehydes in good conversions. On the other hand, 2-pyridylmethanol (2f) did not afford oxidation/Knoevenagel condensation product 4f under optimized reaction conditions. However, it was partially converted to its corresponding carboxylic acid 6f when two or more equivalents of TBHP was used. This observations can be attributed to fast conversion of substrate 2f to its corresponding N-oxide which rearrange subsequently to the corresponding 2-pyridinecarboxylic acid (6f)58,59.

According to the above observations, a plausible free radical mechanism, as shown in Scheme 2, can be proposed for the cascade oxidation/Knoevenagel condensation of different benzyl alcohols 2 by TBHP in the presence of Cs-Pr-Me-Cu (II)-Fe3O4 (1). First, THBP is broken down into t-butylproxide radical and proton by reduction of Cu+2 and Fe+3 ions. Abstraction of a hydrogen radical from benzyl alcohol derivatives 2 affords corresponding benzyl radicals (I) which can combine later with t-butyloxide radical to form corresponding benzaldehydes (II). Next, the obtained aromatic aldehydes and malononitrile (3) are activated by the Lewis acidic and basic sites of multifunctional catalyst 1, respectively via a typical Knoevenagel condensation route. Finally, elimination of one molecule of water affords desired products 4(a-f)60,61. It should be noted that the hygroscopic nature of the chitosan backbone of the supramolecular catalyst 1 can additionally adsorb water molecules on its surface and hence promote smoothly the Knoevenagel condensation15,51,62.

Furthermore, reusability of a heterogeneous catalyst is an important feature for its efficiency and future industrial application. Consequently, we studied the reusability of the Cs-Pr-Me-Cu(II)-Fe3O4 (1) up to fifth cycle (Fig. 6). Therefor, the catalyst 1 was magnetically separated from the reaction mixture, washed with acetone and hexane to remove any organic impurities and dried in an oven. After drying, it was again used for oxidation/Knoevenagel condensation following the same procedure as mentioned in the experimental section. As data in Fig. 6 show, the decrease of catalytic activity of the nanocomposite 1 from the first run to the second run was slight (about 2%). However, more decrease (about 12%) was observed in the next runs with the recycled catalyst 1. According to ICP analysis results, the percentage of copper and iron in the fresh Cs-Pr-Me-Cu(II)-Fe3O4 (1) was found to be 1.38 wt% (Cu) and 25.72 wt% (Fe), respectively. On the other hand, the percentage of copper and iron in the recycled Cs-Pr-Me-Cu(II)-Fe3O4 (1) after five runs was observed to be 1.26 wt% (Cu) and 25.16 wt% (Fe), respectively. This means that relative percentage of copper loss (8.8 wt%) in the recycled catalyst after five runs is higher than iron loss (2.2 wt%) compared to the fresh sample. These observations can be interpreted to more loss of melamine units, as a more probable chelating agent of Cu(II) species, compared to chitosan monomers with more tendency to chelate Fe3O4 nanoparticles. Indeed, some parts of the covalent bonds between melamine and chitosan may be broken during mechanical stirring and heating required for reaction or recycling. On the other hand, bridge methylene groups in the 1,3-propylene linker are more labile to be partially oxidised and subsequently facilitate break down of the melamine units from the modified polymer backbone through hydrolysis during reaction, separation or recycling63. On the other hand, Fig. 4(b) shows the XRD patterns of the recycled catalyst 1 after five runs used in the model reaction. As can be seen, there is very good coincidenc between the powder XRD signals of the fresh Cs-Pr-Me-Cu(II)-Fe3O4 (1) and the recycled sample.

Figure 6
figure 6

Reusability of the heterogeneous nanocatalyst Cs-Pr-Me-Cu(II)-Fe3O4 (1) for the synthesis of 4a.

Full size image

To demonstrate the efficiency and merits of the Cs-Pr-Me-Cu(II)-Fe3O4 supramolecular catalyst (1) for the cascade oxidation/Knoevenagel condensation of different benzyl alcohols, it has been compared with some of the recently catalytic systems. The comparison has been summarized in Table 3. It is obvious that the present catalytic system requires low loading of a nontoxic and inexpensive transition metal species working at room temperature to afford one-pot oxidation/Knoevenagel condensation products in one step.

Table 3 Comparison of the catalytic activity of Cs-Pr-Me-Cu(II)-Fe3O4 (1) with other reported catalysts for the synthesis of the 2-benzylidenemalononitrile (2a).
Full size table

Experimental Section

General Information

All reagents and solvents were obtained from commercial suppliers and used without further purification. Chitosan (MW = 100000–300000 Da) was purchased from Acros Organics. 1H NMR spectra were recorded at 500 MHz using a Bruker DRX-500 Avance spectrometers in DMSO-d6 or CDCl3 as the solvent. Characterization of the catalyst 1 was carried out using FESEM TESCAN-MIRA3, EDX Numerix DXP-X10P, VSM (BHV-55, Riken, Japan), Shimadzu FT-IR-8400S and TEM Philips CM30. The analytical thin layer chromatography (TLC) experiments were performed using Merck 0.2 mm silica gel 60 F-254 Al-plates.

General procedure for preparation of the Cs-Pr-Me-Cu(II)-Fe3O4 (1)

The melamine-functionalized chitosan (Cs-Pr-Me) was first prepared according to the procudure described in our previous works51,52. Next, the Cs-Pr-Me (1.0 g) was suspended in 50 mL of distilled water. To this suspension, Cu(OAc)2 (0.5 g) was added and stirring was continued for 12 h. The final dispersed solution was centrifuged and the obtained solid was dried under vacuum for 1 h. Then, Fe3O4 nanoparticles were fabricated by in-situ coprecipitation as follows: Iron(III) chloride hexahydrate (4.6 g, 0.017 mol) and iron(II) chloride tetrahydrate (2.2 g, 0.011 mol) were dissolved in distilled water. The prepared Cs-Pr-Me-Cu(II) was then added into the obtained aqueous solution and heated to 50 °C under N2 atmosphere. Then, 25% aqueous ammonia (10 mL) was slowly added to the obtained mixture under vigorous stirring. After 30 min, the precipitate was collected from the solution by an external magnet and washed three times with distilled water (3 × 5 mL). Finally, the obtained brown solid was dried in an oven at 60 °C for 2 h before using.

Typical procedure for the synthesis of α,β-unsaturated nitriles (4a-f) through cascade oxidation/Knoevenagel condensation catalyzed by the Cs-Pr-Me-Cu(II)-Fe3O4 (1)

In a round-bottomed flask, benzyl alcohol (2, 1.0 mmol), TBHP (1.0 mmol) and Cs-Pr-Me-Cu(II)-Fe3O4 (1, 20 mg) were mixed in CH3CN (2.0 mL) and stirred at room temperature. Then, malononitrile (3, 1.1 mmol) was added to the reaction mixture and the mixture was stirred for the appropriate times reported in Table 2. After completion of the reaction, the solvent was evaporated. Then, EtOAc (3 mL) was added to the mixture and the catalyst 1 was separated by an external magnet. Afterwards, n-hexane was added drop wise into the solution untill benzylidinemalononitriles 4 were completely precipitated. The obtained mixture was filtered off and the precipitate were washed with n-hexane and then dried in an oven at 70 °C for 1 h. Alternatively, the products were extracted by EtOAc and the crude reaction mixture after evaporation of the solvent was analyzed by 1H NMR (Figs. S6–S10, See Electronic Supplementary Information). The recycled catalyst 1 was washed with acetone and hexane (1 mL), respectively and then dried at 50 °C for 2 h and stored for another run.

Leaching test of the Cs-Pr-Me-Cu(II)-Fe3O4 (1)

The Cs-Pr-Me-Cu(II)-Fe3O4 (1) was separated from the model reaction mixture after 4 h. The reaction was continued with the filtrate in the absence of nanocatalyst 1 for an extra 4 h. No further increase in the conversion of benzyl alcohol (2a) was observed, which approves the catalytically active sites for this oxidation/Knoevenagel condensation are located on the Cs-Pr-Me-Cu(II)-Fe3O4 (1) and do not disperse into the reaction mixture.

Conclusions

In summary, we have developed a cost-effective and practical route for fabrication of uniform Cs-Pr-Me-Cu(II)-Fe3O4 nanomaterials for the first time. The new Cs-Pr-Me-Cu(II)-Fe3O4 supramolecular catalyst was found to be, a valuable magnetically reusable catalyst for sustainable cascade oxidation/Knoevenagel condensation of benzyl alcohols to their corresponding α,β-unsaturated nitriles under truely mild reaction conditions. Significant advantages of the present work are as follows: high to quantitative conversion of benzyl alcohols, mild reaction conditions, scalable synthesis, simple separation of the catalyst using an external magnet and reusability of the catalyst for at least five cycles. Based on this points, the present work can be considered as a valuable complementary study in the field of cascade oxidation/Knoevenagel condensation. Furthermore, this supramolecular catalyst is a successful example of the combination of chemo- and bio-drived materials catalysis for mimicing biocatalysis and sustainable and one pot multistep synthesis.