Synthesis of N,N´-Alkylidenebisamides in the Presence of Fe3O4@Nano-Cellulose/B(III) as a Natural Based Super Paramagnetic Nanocatalyst

INTRODUCTION
Bis-amides are an important class of organic compounds because of these groups can be easily transformed into other functionalities (such as gem-diaminoalkyl and aminoalkyl group) and also, used for the synthesis of pharmacological materials such as peptidomimetic compounds [1]. The first synthesis of gem-bis-amides was achieved via condensation reaction between an aldehyde and an amide by Noyes et al in 1933 [2]. Various catalysts, such as montmorillonite K10 [3], silica supported calcium chloride (SiO2-CaCl2) [4], hydroxyapatite [5], ZnCl2/SiO2 [6], acetyl chloride [7], zinc chloride [8], phosphotungstic acid,[9] nano copper ferrite (CuFe2O4) [10], silica supported polyphosphoric acid (SiO2-PPA) [11] and silica-bonded S-sulfonic acid nanoparticles (SBSSANPs) [12] have been applied for the preparation of N,N’-alkylidenebisamides.
In this study, Fe3O4@nano-cellulose/B(III) was synthesized as natural based super paramagnetic nano-catalyst. The catalyst was characterized by FT-IR, VSM, XRD, XRF, BET, FESEM, TEM, DSC(DSX) and TGA techniques. Fe3O4@nano-cellulose/B(III) was successfully applied to the synthesis of N,N’-alkylidenebisamides derivatives via one-pot three-component condensation reaction of various aldehydes and amides.

MATERIAL AND METHODS
All compounds were purchased from Aldrich, Merck, and Fluka chemical companies. Nano-cellulose and Fe3O4@nano-cellulose were synthesized via our previously reported methods [13-16]. FT-IR spectra were run on a Bruker, Equinox 55 spectrometer. A Bruker (DRX-400 Avance) NMR was used to record the 1H NMR and 13C NMR spectra. The X-ray diffraction (XRD) pattern was obtained by a Philips Xpert MPD diffract meter equipped with a Cu Ka anode (k = 1.54 A°) in the 2θ range from 10 to 80°. XRF analysis was done with Bruker, S4 Explorer instrument. VSM measurements were performed by using a vibrating sample magnetometer (Meghnatis Daghigh Kavir Co. Kashan, Iran). Melting points were determined by a Buchi melting point B-540 B.V.CHI apparatus. Field emission scanning electron microscopy (FESEM) was obtained on a Mira 3-XMU. Transmission electron microscopy (TEM) was obtained using a Philips CM120 with a LaB6 cathode and accelerating voltage of 120 kV. Quantitative elemental information (EDS) of Fe3O4@nano-cellulose/B(III) was measured by an EDS instrument and Phenom pro X. Thermal gravimetric analysis (TGA) was conducted using “STA 504” instrument. Brunauer–Emmett–Teller (BET) surface area of catalyst was done with Micromeritics, Tristar II 3020 analyzer.

Preparation of Fe3O4@nano-cellulose/B(III) 
In a beaker, 10 ml of dichloromethane was added to a Fe3O4@nano-cellulose (0.5 g) and stirred at room temperature. Then, 0.5 ml BF3.OEt2 was added dropwise to the mixture and stirred for 1 hour at room temperature. After this time, the mixture was filtered, washed with dichloromethane, dried at room temperature and Fe3O4@nano-cellulose/B(III) catalyst was obtained.

General procedure for synthesis of N,N′-alkylidenebisamide derivatives
A mixture of aldehyde (1.0 mmol) and amide (2.0 mmol) was stirred at 70 °C in the presence of Fe3O4@nano-cellulose/B(III) (0.06 g). After completion of the reaction (monitored by TLC, n-hexane: EtOAc, 80:20), the reaction mixture was dissolved in ethanol and the catalyst was removed by an external magnet and reaction mixture was filtered. Then, by adding water to the filtrate, the pure products were obtained.

RESULTS AND DISCUSSION
In this work, Fe3O4@nano-cellulose/B(III) was prepared in several steps. At first, nano-cellulose was obtained from the acid hydrolysis of cotton. By this step, the free OH groups in nano-cellulose have been increased and could be used to synthesis of nano-cellulose supported catalysts. Fe3O4@nano-cellulose, were obtained simply through co-precipitation of ferric and ferrous ions with ammonium hydroxide in an aqueous solution containing nano-cellulose. At the end, preparation of Fe3O4@nano-cellulose/B(III) was done via mixing of Fe3O4@nano-cellulose and BF3.OEt2 at room temperature. The OH groups in the nano-cellulose act as nucleophile and B–O–C bonds were formed by the interaction between the BF3 and the OH groups (Fig. 1).  The morphology and structure of the prepared Fe3O4@nano-cellulose/B(III) composite was characterized through FT-IR, VSM, XRD, XRF, BET, FESEM, TEM, EDS and TGA techniques.
The FT-IR spectra of Fe3O4, nano-cellulose, Fe3O4@nano-cellulose and Fe3O4@nano-cellulose/B(III) are shown in Fig.2.
The FT-IR spectrum of Fe3O4@nano-cellulose/B(III) shows the stretching vibrations of the OH group at 3334 cm-1 and the stretching vibrations of the C–O group appears at around 1032 and 1160 cm-1. In addition to the above mentioned bands, a broad band at 567 cm-1 corresponds to Fe/O stretching vibrations in Fe3O4 Lattice. 
 Fig. 3 represents the FESEM and TEM images of Fe3O4@nano-cellulose/B(III) which was applied to investigate the particle size and surface morphology. This image indicates the Fe3O4@nano-cellulose/B(III) particles are on average below 20 nm.
The X-ray diffraction (XRD) patterns of Fe3O4, Fe3O4@nano-cellulose and Fe3O4@nano-cellulose/B(III) are shown in Fig. 4. According to XRD pattern of Fe3O4@nano-cellulose/B(III) (Fig. 4b), signal in 2θ equal to 23 confirmed the existence of cellulose in its structure. The signals in 2θ= 30, 36, 43, 54, 57 and 63 demonstrate the existence of Fe3O4.  
The magnetic properties of Fe3O4 and Fe3O4@nano-cellulose/B(III) were investigated at room temperature (300 K) by a vibrating sample magnetometer (VSM) studies (Fig. 5) no hysteresis loop and no remanence was detected and also the coercivity value is zero for two samples, suggesting typical super-paramagnetic property at room temperature. The amounts of saturation magnetization for Fe3O4 and Fe3O4@nano-cellulose/B(III) are 50 and 23 emu g-1, respectively. Despite this significant decrease, the catalyst can still be easily separated from the reaction mixture by using an external magnet.
To investigate the elemental component of Fe3O4@nano-cellulose/B(III), XRF analysis of catalyst was done by comparison of its kilo counts per second (KCPS) of elements in the catalyst with pure samples (NaF, H3BO3), (Table 1). The amounts obtained for B and F are 3.496 g (0.32 mol) and 1.112 g (0.059 mol), respectively. Thus, the B:F ratio, in the catalyst is approximately 5:1.
The specific surface area of catalyst was measured by Brunauer–Emmett–Teller (BET) theory (Fig. 6a). The single point surface area at P/Po = 0.0685 is 65.9890 m2 g-1 and BET surface area is 74.2480 m2 g-1. The N2 adsorption isotherm of the catalyst is depicted in Fig. 6b.
TGA-DTA analysis was performed to estimate thermal stability of the Fe3O4@nano-cellulose/B(III) in the temperature range of 26–814 °C (Fig. 7). The TGA curve illustrates three stages of weight loss. The first weight loss in 93 °C (5% weight loss) is related to the removal of moisture from the catalyst. The next weight loss (52%) appears in the range of 100–670 °C and relates to the decomposition of cellulosic units in the nano-composite. Finally, the main weight loss (6%) is observed in the range of 670–700 °C. 
The EDS spectrum of Fe3O4@nano-cellulose/B(III) (Fig. 8), shows the presence of the elements O, Fe, F, and B with the corresponding weight percentages (37.90, 26.76, 27.20 and 4.21 %).

Catalytic activity of Fe3O4@nano-cellulose/B(III)
After characterization of catalyst, The catalytic activity of Fe3O4@nano-cellulose/B(III) was investigated for the synthesis of N,N′-alkylidenebisamides derivatives. For optimization of N,N′-(phenylmethylene) dibenzamide the via reaction of benzamide (2.0 mmol) and benzaldehyde (1.0 mmol) as the model reaction. To optimize the reaction conditions, the effect of several parameters such as catalyst amount, solvent, temperature reaction time and laboratory conditions was studied (Table 2). According to Table 1, the best condition for the reaction is a solvent-free condition using 0.06 g of Fe3O4@nano-cellulose/B(III) as an efficient catalyst at 70 °C (Table 2, entry 6).
Based on optimal reaction conditions, N,N′-alkylidenebisamides derivatives were synthesized using the three-component condensation of various aldehydes (1.0 mmol) and various amides (2.0 mmol) (Table 3). As the results in Table 3 show, all these reactions occurred smoothly affording high yields. The structures of these products were characterized by melting point, FT-IR, 1H- and 13C-NMR spectroscopy. To indicate the capability of the present method and efficiency of our catalyst in comparison with the reported methods to prepare N,N′-(phenylmethylene) dibenzamide, we compared our results with other methods in the literature for the model reaction (Table 4). we have observed good and high yields of products in very mild and green conditions using Fe3O4@nano-cellulose/B(III). The reusability of the catalyst was also investigated on the model reaction. The magnetic nature of the catalyst allowed its facile recovery by simple separation by an external magnet, then washed with ethanol, dried at room temperature to be used in the subsequent run of the reaction with fresh reactants under similar conditions. It was observed that the recovered biocatalyst could be used at least four times with marginal decrease in their catalytic activity (Fig. 9). The proposed mechanism for the synthesis of N,N′-alkylidenebisamides derivatives in the presence of Fe3O4@nano-cellulose/B(III) is shown in Fig. 10. 

CONCLUSION
In summary, we have demonstrated the preparation and characterization of Fe3O4@nano-cellulose/B(III) as highly efficient, magnetically recyclable, cheap and bio-based catalysts. The catalytic activity of the prepared catalyst was investigated in the synthesis of N,N′-alkylidenebisamides derivatives reaction via the condensation reaction of various aldehydes and various amides under mild reaction conditions. This method offers several advantages including easy work-up, excellent yields, short reaction time, reusability of the catalyst and environmental friendliness.

ACKNOWLEDGEMENT
The Research Council of Yazd University is gratefully acknowledged for the financial support for this work.

CONFLICT OF INTEREST
The authors declare that there is no conflict of interests regarding the publication of this manuscript.