Cu(II) and magnetite nanoparticles decorated melamine-functionalized chitosan: A synergistic multifunctional catalyst for sustainable cascade oxidation of benzyl alcohols/Knoevenagel condensation

The uniform decoration of Cu(II) species and magnetic nanoparticles on the melamine-functionalized chitosan afforded a new supramolecular biopolymeric nanocomposite (Cs-Pr-Me-Cu(II)-Fe3O4). The morphology, structure, and catalytic activity of the Cs-Pr-Me-Cu(II)-Fe3O4 nanocomposite have been systematically investigated. It was found that Cs-Pr-Me-Cu(II)-Fe3O4 nanocomposite can smoothly promote environmentally benign oxidation of different benzyl alcohol derivatives by tert-butyl hydroperoxide (TBHP) to their corresponding benzaldehydes and subsequent Knoevenagel condensation with malononitrile, as a multifunctional catalyst. Interestingly, Fe3O4 nanoparticles enhance the catalytic activity of Cu(II) species. The corresponding benzylidenemalononitriles were formed in high to excellent yields at ambient pressure and temperature. The heterogeneous Cs-Pr-Me-Cu(II)-Fe3O4 catalyst was also very stable with almost no leaching of the Cu(II) species into the reaction medium and could be easily recovered by an external magnet. The recycled Cs-Pr-Me-Cu(II)-Fe3O4 was reused at least four times with slight loss of its activity. 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 synthesis.

The FTIR spectra of Cs-Pr-Me-Cu(II)-Fe 3 O 4 (1) demonstrated the evidence for existence of Cu(II) and magnetite nanoparticles on the melamine-functionalized chitosan backbone.  cs-pr-Me-cu(ii)-fe 3 o 4 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)-Fe 3 O 4 (1), the one-pot oxidation/Knoevenagel condensation between benzyl alcohol (2a) and malononitrile (3) in CH 3 CN 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, H 2 O 2 , O 2 , 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, H 2 O 2 , and O 2 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)-Fe 3 O 4 (1), as a catalyst, without any oxidant (Table 1, entry 5). Among TBHP and H 2 O 2 , the former was found to be more effective for the reaction (Table 1, entries 6-7). Indeed, H 2 O 2 afforded lower yeild of the desired product 4a compared to TBHP. In fact, the hydroxyl radical produced by H 2 O 2 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 weights 57 . 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 H 2 O afforded the desired product 4a in lower yields compared to CH 3 CN www.nature.com/scientificreports www.nature.com/scientificreports/ 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][13][14]. It is noteworthy that the Cs-Pr-Me-Cu(II) materials afforded a lower yeild compared to the Cs-Pr-Me-Cu(II)-Fe 3 O 4 (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).
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.
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)-Fe 3 O 4 (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, www.nature.com/scientificreports www.nature.com/scientificreports/  www.nature.com/scientificreports www.nature.com/scientificreports/ 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 condensation 15,51,62 .
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)-Fe 3 O 4 (1) up to fifth cycle  www.nature.com/scientificreports www.nature.com/scientificreports/ (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)-Fe 3 O 4 (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)-Fe 3 O 4 (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 Fe 3 O 4 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  www.nature.com/scientificreports www.nature.com/scientificreports/ subsequently facilitate break down of the melamine units from the modified polymer backbone through hydrolysis during reaction, separation or recycling 63 . 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)-Fe 3 O 4 (1) and the recycled sample.
To demonstrate the efficiency and merits of the Cs-Pr-Me-Cu(II)-Fe 3 O 4 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. 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. 1 H NMR spectra were recorded at 500 MHz using a Bruker DRX-500 Avance spectrometers in DMSO-d6 or CDCl 3 as the solvent. General procedure for preparation of the cs-pr-Me-cu(ii)-fe 3 o 4 (1). The melamine-functionalized chitosan (Cs-Pr-Me) was first prepared according to the procudure described in our previous works 51,52 . 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, Fe 3 O 4 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 N 2 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)-fe 3 o 4 (1). In a round-bottomed flask, benzyl alcohol (2, 1.0 mmol), TBHP (1.0 mmol) and Cs-Pr-Me-Cu(II)-Fe 3 O 4 (1, 20 mg) were mixed in CH 3 CN (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 1 H 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.