Reusable (MWCNTs)-COOH/Fe3O4-CaO Magnetic Nanostructure Promoted Three-Component Synthesis of Hexahydroacridine Derivatives

INTRODUCTION
In recent years, the combination of carbon nanotubes (CNTs) and inorganic nanoparticles have formed nanostructures that have superior properties [1]. Multi-walled carbon nanotubes have gained more interest than others, based on their great potentialities in various technological fields such as controlled drug release, wide surface area, chemical, electrical, thermal performance, and heterogeneous catalysis [2–6]. The greatest advantages of heterogeneous catalysts are the ease of separation from the reaction mixture and reuse [7,9]. However, the most important problem concerning MWCNTs lies within their separation from aqueous solution [10]. To overcome the poor processability and dispersibility, the functionalization and solubilization of CNTs have received much attention. The decoration of CNTs with inorganic compounds through covalent and non-covalent bonds can give them new properties and potential for various new applications [11]. The reaction of bimetal oxides like Fe3O4-CaO nanoparticles and multi-walled carbon nanotubes to form new type of hybrid nanostructure materials will exploit even more diverse applications [12].
Fe3O4 magnetic nanoparticles were used in most studies due to high saturation magnetization, high magnetic susceptibility, chemical stability, and biocompatibility. Fe3O4 MNPs have been used in gene delivery, cell therapy, drug delivery, recording material, sensor, and catalyst [13-18]. These magnetic nanoparticles have been used as an efficient catalyst in many organic reactions [19]. Fe3O4 MNPs tend to aggregate to form the bulk metal oxide, giving rise to a dramatic decrease in surface area. To prevent this undesirable metal oxide aggregation, magnetic cores have been surrounded by functional ligands such as ligands that contain terminal amine, phosphoric acid, thiol, or carboxylic acid groups [20,21]. They can bind with the surface of magnetic nanoparticles. Magnetic metal oxide NPs immobilized acid functionalized multi-walled carbon nanotubes (MWCNTs)-COOH have shown superior catalytic activities for the synthesis of organic reactions, due to the presence of a very active site on the large surface of (MWCNTs–NPs) hybrid structures [22,23]. On the other hand, CaO was often used as a heterogeneous catalyst because of its low toxicity, regeneration, high basicity, and high catalytic activity [24]. However, the shortcomings possessed by CaO are low thermal stability and low mechanical strength, so it needs to be impregnated with other oxides [25]. CaO can be easily derived from the environment, waste sources such as ashes, crab shells, sand, oyster shells, animal bones, snail shells, and also eggshells [26-30]. The most chemical component of the calcined waste eggshell is CaO (about 97%), which can be obtained from calcium carbonate in the eggshell under high temperatures (in the range of 700–1000 oC) [31]. Because of the above reasons and due to the presence of the high surface area of MWCNT–COOH/Fe3O4-CaO hybrid, it can be employed as alternative catalyst support, because of their high surface area resulting in high catalyst loading capacity, high dispersion, outstanding stability, and convenient catalyst recycling. We proposed that the MWCNT–COOH/Fe3O4-CaO hybrid can increase the catalytic properties for the synthesis of hexahydroacridine-1,8-dione derivatives. hexahydroacridine-1,8-dione derivatives have been extensively studied due to their wide range of biological activities and pharmaceutical properties, such as anticancer, antimicrobial, anti-Alzheimer, antibiotic, antileishmanial, and antimalarials agents [32,33]. There are various reports in the literature on the three-component Hantzsch-type synthesis of acridines involving condensation of aromatic aldehydes, anilines, and dimedone via conventional methods [34]. However, many of these methods suffered some limitations.  So in this article, we present a simple, cheap, and eco-friendly synthesis of hexahydroacridine-1,8-dione derivatives using (MWCNTs)-COOH/Fe3O4-CaO hybrid as an effective catalyst under mild reaction conditions and good yields.

MATERIALS AND METHODS
Chemicals and Instrumentation
Solvents and chemicals were purchased from Aldrich and Merck. (MWCNTs)-COOH (OD: 20-30 nm) was purchased from US Research Nanomaterials, Inc. (MWCNTs)-COOH/Fe3O4-CaO was distinguished by powder X-ray diffraction (XRD) PW 3040/60 X’Pert PRO diffractometer system, using Cu Ka radiation with (λ = 1.5418 Å) in the range of 2θ = 20–80° at room temperature. The morphology and sizes of NPs were measured using a transmission electron microscope (TEM, 150 kV, and Philips-CM 10) and a scanning electron microscope (SEM) by Daypetronic Company-Iran. FT-IR measurements were recorded on a Shimadzu 8400s spectrometer with KBr plates. The NMR spectra were determined on Bruker XL 400 (400 MHz) instruments; Mass-spectrometric measurements were made on an Agilent 6890 N Network GC system. The elemental analysis was performed by the microanalytical service of the Daypetronic Company. Melting points were obtained on an Electrothermal 9100 without further corrections.  

Preparation of (MWCNTs)-COOH/Fe3O4-CaO hybrid
Waste quail eggshells were thoroughly cleaned and air-dried after the removal of the inner membrane layer. Cleaned eggshells were crushed into small pieces and dried at 80°C for 24 h in the oven. The functionalize (MWCNTs)-COOH (0.3 g), the dried eggshells (0.1 g), and FeSO4 (0.1 g) were added to 50 ml of acetic acid in a flask. The mixture was kept in an ultrasonic bath for 30 min and then slowly stirred outside the ultrasonic device for another 2 hours, under reflux conditions. The solvent was evaporated and The resulting precipitate was calcined at 250°C for 3 hours to obtain magnetically (MWCNTs)-COOH/Fe3O4-CaO (Fig. 1).

General procedure for the synthesis of compounds 1-5 in the presents of (MWCNTs)-COOH/Fe3O4-CaO 
Raw materials and (MWCNTs)-COOH/Fe3O4-CaO (7 mol%) were mixed and reacted in ethanol (10 ml) under reflux conditions. The completion of the reaction was determined by TLC using n-hexane: ethyl acetate (2:1) and appeared by a UV lamp (254 & 366 nm). In the end, the catalyst was separated by an external magnet, filtered, washed with ethanol and water, dried at 80 °C for 1h, and reused for the same reaction. The residue of the reaction mixture was evaporated, and the crude product was purified by short-column chromatography on silica gel (CHCl3: MeOH / 10:1). This column chromatography operation was repeated to give pure compounds (1-5) as colorless, viscous oils. The products were determined by CHN analyses, NMR, and FT-IR spectra.

10-(4-bromophenyl)-9-((R)-((2S,3S,4R)-3,4-dihydroxytetrahydrofuran-2-yl)(hydroxy) methyl)-3,3,6,6-tetramethyl-3,4,6,7,9,10-hexahydroacridine-1,8(2H,5H)-dione 1 
Reaction of d-mannose (1 mmol), dimedone (2 mmol), and 4-bromoaniline (1 mmol), yellow powder. FT-IR spectrum, ν, cm–1: 3420.8 (OH), 3061.7 (CHaro), 2955.4 (CHaliph), 2869.2 (CHaliph), 1740.1 (C=O), 1622.0 (C=C), 1489.5 (C=C), 1398.8 (C=C), 1262.6 (C-O), 1223.1 (C-O), 1771.5 (C-O), 1146.5 (C-O), 1070.4 (C-O), 1036.1 (C-O). 1H NMR spectrum (400 MHz, DMSO-d6), δ ppm (J, Hz): 7.11 (2H, d, J = 8.8 Hz, CHaro), 6.50 (2H, d, J = 8.4 Hz, CHAro), 5.25 (2H, br, OH), 4.28 (1H, t, J = 7.6 Hz, CH), 4.15 (1H, d, J = 6.4 Hz, CH), 3.65 (1H, d, J = 8.4 Hz, CH), 3.57 (1H, d, J = 8.4 Hz, CH2), 3.34 (2H, t, J = 9.2 Hz, CH), 2.23 (1H, d, J = 16.0 Hz, CH2), 2.16 (1H, d, J = 16.0 Hz, CH2), 2.09 (1H, d, J = 14.4 Hz, CH2), 2.02 (1H, d, J = 14.4 Hz, CH2), 1.94 (3H, t, J = 16.2 Hz, CH2), 1.09 (3H, s, CH3), 1.00 (3H, s, CH3), 0.94 (6H, s, 2CH3). 13C NMR spectrum (100 MHz, DMSO-d6), δ, ppm: 193.01, 184.97, 171.56, 145.09, 138.05, 132.02, 124.98, 116.18, 113.59, 110.91, 93.49, 81.09, 70.03, 64.21, 51.60, 49.25, 34.51, 28.81. Found, %: C, 60.47; H, 6.21; N, 2.53. C28H34BrNO6 (559.16). Calculated, %: C, 60.00; H, 6.11; N, 2.50.

10-(4-bromophenyl)-9-((S)-((2S,3S,4R)-3,4-dihydroxytetrahydrofuran-2-yl) (hydroxy)methyl)-3,3,6,6-tetramethyl-3,4,6,7,9,10-hexahydroacridine-1,8(2H,5H)-dione 2
Reaction of d-galactose (1 mmol), dimedone (2 mmol), and 4-bromoaniline (1 mmol), yellow powder. FT-IR spectrum, ν, cm–1: 3404.1 (OH), 3369.8 (OH), 3012.4 (CHaro), 2953.9 (CHaliph), 2874.8 (CHaliph), 1699.7 (C=O), 1619.9 (C=C), 1489.2 (C=C), 1402.0 (C=C),1260.6 (C-O), 1223.7 (C-O), 1172.4 (C-O), 1069.6 (C-O), 1031.7 (C-O). 1H NMR spectrum (400 MHz, DMSO-d6), δ ppm (J, Hz): 7.12 (2H, d, J = 8.8 Hz, CHaro), 6.50 (2H, d, J = 8.4 Hz, CHAro), 5.25 (2H, br, OH), 4.66 (1H, d, J = 8.0 Hz, CH), 4.37 (1H, d, J = 7.2 Hz, CH), 3.72 (1H, t, J = 5.2 Hz, CH), 3.45 (1H, t, J = 9.6 Hz, CH2), 3.40 (1H, br, CH), 2.24 (1H, d, J = 16 Hz, CH2), 2.13 (1H, d, J = 16 Hz, CH2), 1.98 (3H, br, CH2), 1.93 (1H, d, J = 8.8 Hz, CH2), 1.86 (1H, d, J = 10.0 Hz, CH2), 1.06 (3H, s, CH3), 1.00 (3H, s, CH3), 0.93 (6H, s, 2CH3). 13C NMR spectrum (100 MHz, DMSO-d6), δ, ppm: 192.45, 186.64, 175.25, 149.11, 146.77, 131.69, 121.97, 116.19, 115.62, 111.86, 110.72, 101.69, 90.53, 70.86, 69.99, 63.52, 51.60, 34.20, 31.88, 28.36. Found, %: C, 60.36; H, 6.15; N, 2.47. C28H34BrNO6 (559.16). Calculated, %: C, 60.00; H, 6.11; N, 2.50.
10-(4-bromophenyl)-9-((2S,3S,4R)-3,4-dihydroxytetrahydrofuran-2-yl)-3,3,6,6-tetramethyl-3,4,6,7,9,10-hexahydroacridine-1,8(2H,5H)-dione 3   
Reaction of d-arabinose (1 mmol), dimedone (2 mmol), and 4-bromoaniline (1 mmol), yellow powder. FT-IR spectrum, ν, cm–1: 3417.8 (OH), 3015.9 (CHaro), 2955.4 (CHaliph), 2873.3 (CHaliph), 1705.2 (C=O), 1616.9 (C=C), 1485.0 (C=C), 1402.6 (C=C),1262.4 (C-O), 1224.4 (C-O), 1171.9 (C-O),1147.3 (C-O), 1078.1 (C-O), 1035.3 (C-O). 1H NMR spectrum (400 MHz, DMSO-d6), δ ppm (J, Hz): 7.11 (2H, d, J = 8.8 Hz, CHaro), 6.50 (2H, d, J = 8.8 Hz, CHAro), 5.25 (2H, br, OH), 4.62 (1H, t, J = 5.6 Hz, CH), 4.38 (1H, d, J = 8.0 Hz, CH), 3.57 (1H, d, J = 8.4 Hz, CH), 3.38 (2H, m, CH2), 3.21 (1H, d, J = 8.4 Hz, CH), 2.19 (2H, d, J = 11.6 Hz, CH2), 1.94 (3H, t, J = 5.6, CH2), 1.89 (2H, s, CH2), 1.06 (3H, s, CH3), 1.00 (3H, s, CH3), 0.92 (6H, s, 2CH3). 13C NMR spectrum (100 MHz, DMSO-d6), δ, ppm: 192.60, 186.96, 176.41, 148.53, 131.77, 116.20, 116.05, 109.87, 106.42, 90.51, 72.68, 71.54, 64.33, 51.61, 49.47, 37.97, 35.29, 34.19, 31.83, 29.40, 28.98, 28.51. Found, %: C, 61.28; H, 6.17; N, 2.69. C27H32BrNO5 (530.46). Calculated, %: C, 61.14; H, 6.08; N, 2.64. 

10-(4-bromophenyl)-9-((2S,3S,4R)-3,4-dihydroxytetrahydrofuran-2-yl)-3,3,6,6-tetramethyl-3,4,6,7,9,10-hexahydroacridine-1,8(2H,5H)-dione 4
Reaction of d-ribose (1 mmol), dimedone (2 mmol), and 4-bromoaniline (1 mmol), yellow powder. FT-IR spectrum, ν, cm–1: 3393.5 (OH), 3019.0 (CHaro), 2954.6 (CHaliph), 2876.3 (CHaliph), 1710.7 (C=O), 1600.8 (C=C), 1474.3 (C=C), 1395.9 (C=C), 1220.5 (C-O), 1171.6 (C-O), 1147.9 (C-O), 1082.1 (C-O), 1035.5 (C-O). 1H NMR spectrum (400 MHz, DMSO-d6), δ ppm (J, Hz): 7.11 (2H, d, J = 8.8 Hz, CHaro), 6.50 (2H, d, J = 8.8 Hz, CHAro), 5.26 (2H, br, OH), 4.60 (1H, dd, J = 4.4, 7.2 Hz, CH), 4.31 (1H, d, J = 7.2 Hz, CH), 3.57 (1H, dd, J = 4.4, 6.4 Hz, CH), 3.42-3.35 (2H, m, CH), 2.25 (1H, d, J = 16.8 Hz, CH2), 2.01 (4H, s, CH2), 1.94 (2H, d, J = 18.2, CH2), 1.85 (1H, d, J = 14.8 Hz, CH2), 1.05 (3H, s, CH3), 1.00 (3H, s, CH3), 0.93 (6H, s, 2CH3). 13C NMR spectrum (100 MHz, DMSO-d6), δ, ppm: 192.54, 174.95, 131.76, 116.20, 115.42, 111.50, 91.59, 73.17, 71.82, 63.64, 51.50, 48.73, 37.80, 34.67, 34.18, 31.89, 29.49, 28.79, 28.26. Found, %: C, 61.33; H, 6.02; N, 2.68. C27H32BrNO5 (530.46). Calculated, %: C, 61.14; H, 6.08; N, 2.64.
 
9-((S)-hydroxy((2S,3S,4R)-4-hydroxy-3-(((2R,3R,4S,5S,6R)-3,4,5-trihydroxy-6-(hydroxyl methyl)tetrahydro-2H-pyran-2-yl)oxy)tetrahydrofuran-2-yl)methyl)-3,3,6,6-tetramethyl-10-phenyl-3,4,6,7,9,10-hexahydroacridine-1,8(2H,5H)-dione 5
Reaction of maltose (1 mmol), dimedone (2 mmol), and aniline (1 mmol), yellow powder. FT-IR spectrum, ν, cm–1: 3415.6 (OH), 3017.4 (CHaro), 2955.5 (CHaliph), 2928.8 (CHaliph), 1728.8 (C=O), 1603.2 (C=C), 1473.1 (C=C), 1404.2 (C=C), 1268.5 (C-N), 1145.8 (C-O), 1120.7 (C-O), 1071.7 (C-O), 1033.0 (C-O). 1H NMR spectrum (400 MHz, DMSO-d6), δ ppm (J, Hz): 7.35 (2H, t, J = 7.6 Hz, CHaro), 7.17 (2H, d, J = 7.6 Hz, CHaro), 7.13 (1H, t, J = 8.0 Hz, CHAro), 4.90 (4H, br, OH), 4.19 (1H, d, J = 8.0 Hz, CH), 4.13 (1H, br, OH),  3.70 (2H, d, J = 4.8 Hz, CH), 3.57 (2H, d, J = 18.8 Hz, CH), 3.42-3.49 (6H, br, CH), 3.13 (1H, t, J = 6.0 Hz, CH), 3.04 (1H, t, J = 9.2 Hz, CH), 2.20 (2H, q, J = 12.4 Hz, CH2), 2.05 (2H, s, CH2), 1.95 (2H, d, J = 12.4 Hz, CH2), 1.90 (4H, br, CH2), 1.35 (1H, m, CH2), 1.28 (2H, br, CH2), 1.05 (3H, s, CH3), 0.99 (3H, s, CH3), 0.90 (6H, s, 2CH3). 13C NMR spectrum (100 MHz, DMSO-d6), δ, ppm: 192.45, 179.16, 167.45, 132.16, 129.65, 129.12, 116.20, 100.46, 91.86, 79.62, 73.87, 73.64, 72.94, 70.39, 63.42, 51.72, 38.52, 34.16, 31.64, 30.24, 29.27, 28.81, 28.57, 23.69, 22.86. Found, %: C, 63.39; H, 7.15; N, 2.29. C34H45NO11 (503.22). Calculated, %: C, 63.44; H, 7.05; N, 2.18.

RESULTS AND DISCUSSION
The present paper reports the results of research aimed to verify the activity of the (MWCNT)-COOH/Fe3O4-CaO hybrid as an effective catalyst in the synthesis of hexahydroacridine-1,8-dione derivatives. The possible interaction between eggshells, FeSO4, and (MWCNTs)-COOH was investigated using TGA/DTA, XRD, TEM, SEM, EDX, and FT-IR spectroscopy.

(MWCNTs)-COOH/Fe3O4-CaO characterization
The typical FT-IR spectra of (MWCNTs)-COOH/Fe3O4-CaO can be clarified briefly.
the vibration of the carbon skeleton of the carbon nanotubes was shown at around 1300-1550 cm-1. The bands at about 1630-1750 and 1000-1300 cm-1 indicate the existence of C=O groups of (MWCNTs)–COOH [35]. The bands at about 2000-2450 cm-1 are belonged to the C=C double bonds stretch vibration from the surface of the MWCNTs [36]. The weak peaks around 3500-3900 cm-1 can be assigned to the stretching vibrations of OH groups [37]. The absorption band at 400–700 cm−1 is related to the Fe-O and Ca-O, which confirms the formation of the Fe3O4-CaO MNPs. 
X-ray diffraction (XRD) is normally used to study and characterize the crystallization and average size of (MWCNTs)-COOH/Fe3O4-CaO. In Fig. 2, the XRD pattern of (MWCNTs)-COOH/Fe3O4-CaO demonstrates seven intense peaks in the whole spectrum of 2θ values ranging from 5˚ to 80˚. The presence of eight distinct high diffraction peaks at 2θ values of 23.08˚, 43.15˚, 47.84˚, and 57.37˚ for carbon, 29.52˚, 35.94˚, 39.44˚, 48.67˚ for CaO, and 30.36˚, 35.76˚, 43.47˚, 57.51˚, and 63.16˚ for Fe3O4 respectively, (JCPDS Number. C: 00-026-1080, Fe3O4: 01-075-0449, and JCPDS Number. CaO: 98-000-5337) [38,39] confirmed that the (MWCNTs)-COOH/Fe3O4-CaO had been formed. The other diffraction peaks could be due to some chemical compounds and crystals on the surface of the nanoparticle. The wide X-ray diffraction peaks around their bases show that the (MWCNTs)-COOH/Fe3O4-CaO is in nano sizes. With the XRD pattern, the average diameter which can be calculated from the Scherrer equation [40] (D=Kλ/βcosθ, where β is the peak width at half maximum, λ is X-ray wavelength, and K is constant) is obtained at about 13.7 nm for Fe3O4-CaO NPs. The crystallite size was calculated based on the diffraction peak at a 2θ value of 35.75 These nanoparticles have been fixed on the different layers of carbon nanotubes and increased the outer diameter of the nanotubes (about 70-100 nm, Fig. 2).
The morphology and size of (MWCNTs)-COOH/Fe3O4-CaO were studied using transmission electron microscopy (TEM) in Fig. 3. The TEM image indicates that the Fe3O4-CaO nanoparticles are well bonded to the surface of multi-wall carbon nanotubes. On the other hand, TEM values are in good agreement with XRD.
Fig. 4 shows the SEM images of (MWCNTs)-COOH/Fe3O4-CaO. The outside diameter (OD) of (MWCNTs)-COOH was 20-30 nm but after modification, it was changed to 70-100 nm. It is shown that Fe3O4-CaO nanoparticles have grown as nanoparticles on the surface and inside of the (MWCNTs)-COOH. 
In Fig. 5, EDX analysis was performed to confirm the elements present in the resulting (MWCNTs)-COOH/Fe3O4-CaO. For using SEM/EDS to analyse the composition of a sample, usually a heavy metal such as Au (Au-Pd) was coated the sample to make it conductive before insert it into FE-SEM. Therefore, there is a signal of coating metal (Au) in EDX. In addition,  the analysis reveals the presence of Fe, Ca, O, and C which emphasizes the success of the decoration process with Fe3O4-CaO nanoparticles.
Fig. 6 shows the saturation magnetization (Ms) values of the magnetic (MWCNTs)-COOH/Fe3O4-CaO and pure Fe3O4 NPs measured. All of the samples exhibited typically superparamagnetic features with negligible of coercivity and remanence. As shown in Fig. 6, the Ms of the (MWCNTs)-COOH/Fe3O4-CaO was weakened to a large extent when compared with that of pure Fe3O4 NPs. (MWCNTs)-COOH, the magnetic Fe3O4 NPs, and CaO NPs can cause the coating effect, which leads to the reduction of the magnetic responsiveness. However, this value is high enough for the nanostructure to be separated from the reaction mixture by an external magnet.
The catalytic activity of (MWCNTs)-COOH/Fe3O4-CaO and heterocyclic compounds characterization
 (MWCNTs)-COOH/Fe3O4-CaO (7 mol%) was used as an efficient catalyst for the synthesis of hexahydroacridine-1,8-dione derivatives. Because of its excellent capacity, the exceedingly simple workup and good yields (MWCNTs)-COOH/Fe3O4-CaO was proved to be a good catalyst for these reactions.
In the preliminary stage of the investigation, the model reaction of 4-bromoaniline, arabinose, and dimedone (Fig. 7) was carried out by using various amounts of (MWCNTs)-COOH/Fe3O4-CaO in various solvents and solvent-free conditions. As shown in Table 1, the optimum amount of catalyst was 7 mol%. Decreasing the amount of catalyst leads to a decrease in the yield of the reaction while increasing the amount of catalysts, the optimum amount of (MWCNTs)-COOH/Fe3O4-CaO was 7 mol% as shown in Table 2. Increasing the amount of the catalyst to more than 7 mol%  does not improve the yield of the product any further. 
In the absence of (MWCNTs)-COOH/Fe3O4-CaO, the result of the reaction on the TLC plate even after 4h of the reaction wasn’t good. The best yield of the product was acquired with 7 mol% of (MWCNTs)-COOH/Fe3O4-CaO in EtOH under mild reaction conditions (Table 1, Entry 9). It is important to note that, under the same conditions, (MWCNTs)-COOH or Fe3O4-CaO NPs displayed almost no activity. D-arabinose did not react with dimedone and the yield of the reaction did not exceed more than 10% even after 6 h. It is clear that modification of (MWCNTs)-COOH with Fe3O4-CaO remarkably increased its catalytic activity.
The inductively coupled plasma–atomic emission spectroscopy (ICP-AES) analysis was performed to determine the amount of Fe and Fe3O4-CaO loading in (MWCNTs)-COOH/Fe3O4-CaO before (6.74 mg/g) and after (6.62 mg/g) the reaction. 
The leaching of the catalyst has been measured by using a hot filtration method. The reaction mixture has been filtered out the catalyst ((MWCNTs)-COOH/Fe3O4-CaO) from the reaction mixture at the stage of 50% conversion. We did not observe further progress of the reaction after filtration which indicates there was no leaching to confirm the absence of Fe3O4-CaO NPs and stability of the catalyst. 
We extended our study to different organic reactions to evaluate the scope and potential limitations of this methodology (Table 2, entries 1–5). In almost all cases, the reactions proceeded smoothly within 2-3 hours, providing the corresponding products in good isolated yields. The products were isolated, purified, and analysed by NMR and IR. For example, the 1H NMR spectrum of 10-(4-bromophenyl)-9-((2S,3S,4R)-3,4-dihydroxytetrahydrofuran-2-yl)-3,3,6,6-tetramethyl-3,4,6,7,9,10-hexahydroacridine-1,8(2H,5H)-dione 3 (Fig. 8) shows singlets at 5.25 and 4.62 ppm for OH protons, 3.57, 3.38, and 3.21 ppm for CH protons. Three singlet signals attributed to Me protons have appeared at 1.06, 1.00, and 0.92 ppm, respectively. 
In the 13C NMR spectrum (Fig. 9), the resonances related to CH3 and CH2 carbon groups of 3 were appeared at 37.97, 35.29, 34.19, 31.83, 29.40, 28.98, and 28.51 ppm, respectively. The signals attributed to CO carbon groups and their enol forms have appeared at 192.60, 186,96, 176.41, and 148.53 ppm. The C, CH and CHaro carbon bonds have appeared at 131.77, 116.20, 116.05, 109.87, 106.42, 90.51, 72.68, 71.54, 64.33, 51.61, and 49.47 ppm. 
Thereafter, we carried out the synthesis of hexahydroacridine-1,8-dione derivatives with 7 mol% of (MWCNTs)-COOH/Fe3O4-CaO in ethanol (Table 2). 
A plausible mechanism for the reaction of 4-bromoaniline, dimedone, and sugar is envisaged in (Fig. 10). It is proposed that CO group of sugar is first activated by (MWCNTs)-COOH/Fe3O4-CaO. Nucleophilic addition of dimedone to the activated oxygen followed by the loss of H2O generates intermediate I, which is further activated by (MWCNTs)-COOH/Fe3O4-CaO. Then, the 1,4-nucleophilic addition of a second molecule of dimedone on the activated intermediate I, in the Michael addition fashion, affords the intermediate II, which undergoes nitrogen attack of 4-bromoaniline to give 10-(4-bromophenyl)-9-((2S,3S,4R)-3,4-dihydroxytetrahydrofuran-2-yl)-3,3,6,6-tetramethyl-3,4,6,7,9,10-hexahydroacridine-1,8(2H,5H)-dione 3.   
To investigate the efficiency of the (MWCNTs)-COOH/Fe3O4-CaO, we compared some other metal oxide NPs for the synthesis of compound 3 and the results were summarized in Table 3. The metal oxide NPs were synthesized according to the previously reported procedures [41-45]. As shown in Table 3, the best catalyst for the synthesis of compound 3 is (MWCNTs)-COOH/Fe3O4-CaO, using this metal oxide as a catalyst offers several advantages such as excellent yields, short reaction times, a simple procedure, and using ethanol as a green solvent in contrast with other metal oxides. 
In order to show the advantages of this method over previously reported ones. Some of our results are compared in Table 4 to those of some other methods. These results showed that the yield, time and ratio of the present method are better or comparable to the other reported results for the synthesis of hexahydroacridin derivatives.
The catalyst was simply separated by an external magnet, washed with ethanol and water, and dried at 100 °C for 2 h. The recovered catalyst was then re-entered to a fresh reaction mixture under the same conditions and recycled 5 times without considerable loss of activity (Table 5). More recycling of the nano catalyst led to a gradual reduction during the recovering and washing steps.

CONCLUSION
In conclusion, hybrids of (MWCNTs)-COOH and Fe3O4-CaO NPs have been successfully fabricated in acetic acid to produce (MWCNTs)-COOH/Fe3O4-CaO. The structure, morphological magnetic, and surface were evaluated in details. The presence of nanoparticles and CNTs were confirmed via EDX, XRD, FT-IR. TEM and SEM. (MWCNTs)-COOH/Fe3O4-CaO was used as a reusable efficient catalyst for synthesis of hexahydroacridine-1,8-dione derivatives in ethanol. We have described a rapid and very efficient one-pot three component reaction between dimedone, unprotected sugars and aniline or 4-bromoaniline for the synthesis of hexahydroacridine-1,8-dione derivatives. We have demonstrated that eco-friendly, low-cost and high- yielding synthetic route towards hexahydroacridine-1,8-dione derivatives using (MWCNTs)-COOH/Fe3O4-CaO. This strategy will not only give practical synthetic methods but also assures the expansion of the versatility of clean organic reactions ethanol. In addition to the intrinsic properties of nano catalysts, (MWCNTs)-COOH/Fe3O4-CaO hybrid showed high catalytic activity in organic reactions and increased the rate of the reaction without pollution. In addition, this study provides a new alternative to the poultry waste from the eggshell for its use in the biosynthesis of organic and heterocyclic compounds.

ACKNOWLEDGMENT
The authors wish to thanks Islamic Azad University in Qaemshahr Branch for the institutional support.

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