Imine reduction to the corresponding amine is a widely used and important functional group transformation reaction in synthetic organic chemistry. Many amines are used in the pharmaceutical and agricultural industries ^{[1, 2, 3, 4, 5, 6, 7]}. Two approaches are commonly used to convert imines to amines. One is catalytic hydrogenation, which requires high pressure. The other is hydride reduction, which uses metal hydrides such as LiAlH₄ or NaBH₄ ^[8]. LiAlH₄ is normally used in a dry organic solvent in a protective inert atmosphere ^{[9, 10]}; in contrast, NaBH₄ is an inexpensive, safe, and environmentally friendly reducing agent. Various Lewis acids such as LiCl ^[11], TiCl₄ ^[12], TiCl₃ ^[13], Ni₂B ^[14], I₂ ^[15], ZrCl₄ ^[16], and CoCl₂ ^[17] have been used to reduce C=N bonds in the presence of NaBH₄, but they are not always effective.

Spinel oxides play important roles in various catalytic applications, e.g., gaseous pollutant decomposition ^[18] and the water-gas-shift reaction ^[19]. In normal spinels, AB₂O₄, A is generally a divalent cation occupying tetrahedral sites, and B is a trivalent cation occupying octahedral sites. In inverse spinels, half of the B cations occupy tetrahedral sites, and the formula is B(AB)O₄.

Spinels are prepared using various methods, including hydrothermal synthesis ^[20], modified oxidation processes ^[21], forced hydrolysis ^[22], and ball-milling ^[23], solid-state ^[24], and microemulsion methods ^[25]. Sol-gel methods are preferred because they are effective and convenient techniques and use less energy and materials ^[26].

In this study, as part of our research on organic transformations using simple reagents, catalysts, and reaction media ^[27], we investigated the direct reduction of imines to the corresponding secondary amines by NaBH₄ in the presence of a catalytic amount of nano-sized NiLa₂O₄ spinel (Scheme 1).

Scheme 1. Imines to the corresponding secondary amines by NaBH4 and NiLa2O4.

Citric acid (18 mmol, 3.50 g) was gradually added to a solution of LaCl₃ (4 mmol, 1.48 g) and Ni(NO₃)₂·4H₂O (2 mmol, 0.58 g) in deionized water (10 mL), with constant magnetic stirring. The gel was heated at 80 °C for 3 h and dried at 120 °C for 24 h. It was then ground to a powder in an agate mortar and fired at 750 °C in air for 4 h.

Decomposition of the dried polymeric gel precursor was investigated using differential thermal analysis (DTA; NETZSCH, Germany) at a heating rate of 10 °C/min in the range 25-1000 °C. Crystal structures were examined by X-ray diffraction (XRD; Xpert 200, Philips, Cu K_α radiation) at a scanning rate of 0.02°/s in the 2θ range 0° to 70°. The morphologies and dimensions of the nanoparticles were examined using transmission electron microscopy (TEM; LEO 912AB, accelerating voltage 120 kV). The morphologies of the calcined powders were examined using scanning electron microscopy (SEM; LEO 1450VP, accelerating voltage 120 kV).

NaBH₄ (0.37 mmol, 0.014 g) was added at room temperature to a stirred mixture of the appropriate imine (0.5 mmol) and nano-sized spinel NiLa₂O₄ (0.3 mol%, 6.0 mg) in CH₂Cl₂ (1 mL). Methanol (1 mL) was added slowly under constant magnetic stirring. On completion of the reaction, which was monitored using thin-layer chromatography (TLC; n-hexane:ethyl acetate = 8:2), the catalyst was removed by centrifugation and the solvent was evaporated under reduced pressure. Addition of water (10 mL) gave an oily liquid; the mixture was extracted with CH₂Cl₂ (2 × 10 mL). The combined organic layers were dried over anhydrous Na₂SO₄ and the solvent was evaporated under reduced pressure to afford the product. The sample purity was confirmed using TLC and ¹H nuclear magnetic resonance (NMR) spectroscopy.

Figure 1 shows the DTA curve of the NiLa₂O₄ gel precursor obtained using the sol-gel method. The DTA curve has a large endothermic peak at around 100 °C, which is related to the evaporation of water. The combustion of organic compounds such as citric acid causes the exothermic peak at around 200 °C. The small exothermic peak at 350 °C is attributed to the decomposition of nitrate, which decomposes completely at around 420 °C, giving a large exothermic peak ^[28]. The exothermic peak at 750 °C corresponds to crystallization of the NiLa₂O₄ phase. No obvious change is observed above 800 °C. It is therefore concluded that the optimum calcination temperature is 750 °C.

Fig. 1. DTA curve of NiLa2O4 gel precursor obtained using the sol-gel method.

Figure 2 shows the XRD pattern of spinel-type NiLa₂O₄, which was formed when the gel was calcined at 750 °C for 4 h. No secondary phase is observed. The good agreement between the experimental and calculated data confirms that the sample consists of a single phase. Rietveld refinement showed that the material had a tetragonal unit cell with the I₄/mmm space group. The refined unit cell parameters were a = 3.861205 Å and c = 12.6793 Å. The unit cell volume was 189.035 ų, which is in close agreement with the results reported in the literature ^[29].

Fig. 2. XRD pattern of NiLa2O4 nanopowder sintered at 750 °C.

Figure 3 shows a TEM image of the NiLa₂O₄ nanoparticles prepared at 750 °C. It shows that nanoparticles were homogeneous and consisted of uniform quasi-spherical crystallites of average size 40 nm. This is in agreement with the results obtained using XRD ^[30].

Fig. 3. TEM image of nano-sized NiLa2O4 spinel calcined at 750 °C.

The nanoparticle size affects many properties of nanomaterials and is a major indicator of quality and performance. Figure 4 shows the nano-NiLa₂O₄ spinel particle size distribution; this was determined from the TEM image based on more than 100 particles.

Fig. 4. Particle size distribution for calcined nano-sized NiLa2O4 spinel catalyst.

The SEM image in Fig. 5 of the NiLa₂O₄ nanoparticles prepared using the sol-gel method shows the typical morphology of the obtained powder. It shows that the nanoparticles had regular shapes with well-defined crystal faces ^[31].

Fig. 5. SEM image of nano-sized NiLa2O4 prepared by a sol-gel method.

The imines were synthesized by treatment of the corresponding carbonyl compounds with the appropriate amines, using a previously reported method ^{[32, 33]}. N-(4- chlorobenzylidene)aniline was used as a model compound for optimization of the catalytic conditions for reductive hydrogenation.

Various solvents were screened for the reduction of N-(4-chlorobenzylidene)aniline with NaBH₄ using NiLa₂O₄ spinel nanoparticles as a catalyst (Table 1). The results show that CH₂Cl₂ was the best solvent in terms of activity. The polarity of CH₂Cl₂ is appropriate for dissolving the reactant, and appropriate interactions between the hydrogenated surface of the nanocatalyst and the substrate are triggered. Furthermore, CH₂Cl₂ is a solvent with a low moisture content, toxicity, and boiling point, and is used in many organic transformations.

Table 1 Solvent optimization for chemoselective reduction of N-(4- chlorobenzylidene)aniline to N-(4-chlorobenzyl)aniline. Solvent Time (min) Conversion a (%) DMF 30 90 THF 30 88 CH2Cl2 5 97 CHCl3 5 96 n-Hexane 5 85 Reaction conditions: N-(4-chlorobenzylidene)aniline 1 mmol, nano- sized NiLa2O4 spinel 1 mg, solvent 2 mL, NaBH4 4 eq., room temperature. a Based on 1H NMR spectroscopic integrations.

Table 1 Solvent optimization for chemoselective reduction of N-(4- chlorobenzylidene)aniline to N-(4-chlorobenzyl)aniline.

The catalytic role of the nano-sized NiLa₂O₄ spinel was clarified by performing the model reaction in the absence of a nanocatalyst and in the presence of the micro-NiLa₂O₄ spinel catalyst (Table 2, entries 1 and 2). Neither reaction was complete after 60 min. The reduction of N-(4- chlorobenzylidene)aniline was then performed using different molar ratios of the nanocatalyst to identify the optimum reaction conditions. The results show that the minimum amount of nanocatalyst needed to significantly affect the reduction process was 0.3 mol%, with 0.75 equiv. of NaBH₄ (Table 2, entry 9). The results also show that the use of 0.15 mol% of nanocatalyst gave a low conversion because of the lack of appropriate surface active sites at low nanocatalyst concentrations (Table 2, entry 10).

Table 2 Data for reduction of N-(4-chlorobenzylidene)aniline to N-(4- chlorobenzyl)aniline. Entry Nano-sized NiLa2O4 (mol%) NaBH4 (eq.) Time (min) t (°C) Conversion a (%) 1 None 4 60 25 40 2 Micro-NiLa2O4 (1 mol %) 4 60 25 63 3 1 4 10 25 95 4 1 2 10 25 96 5 1 1 10 25 98 6 1 0.75 10 25 97 7 1 0.50 10 25 50 8 0.6 0.75 10 25 97 9 0.3 0.75 10 25 95 10 0.15 0.75 60 25 50 11 0.3 0.75 35 0 73 12 0.3 0.75 6 40 96 a According to TLC monitoring and 1H NMR spectroscopic integration.

Table 2 Data for reduction of N-(4-chlorobenzylidene)aniline to N-(4- chlorobenzyl)aniline.

Table 3 shows the general applicability of the current method in the synthesis of a wide variety of secondary amines. The reactions were performed by mixing the synthesized imines with NaBH₄ in methanol in an ambient atmosphere until TLC showed the complete disappearance of the imine and formation of the corresponding secondary amine. In all cases, excellent yields of secondary amines with different aryl groups, including those bearing electron-withdrawing or electron- donating groups, were obtained in short reaction times under the optimum reaction conditions.

Table 3 TEM image of nano-sized NiLa2O4 spinel calcined at 750 °C.

Table 3 TEM image of nano-sized NiLa2O4 spinel calcined at 750 °C.

The mechanism of heterogeneous hydrogenation is unclear, partly because the reactive sites on the metal surface cannot be as precisely described as can small molecule reagents in solution. However, it is probable that hydrogen is adsorbed onto the surface of the catalyst, forming metal-hydrogen bonds; this type of intermediate has been implicated in the hydrogenation of unsaturated double bonds, particularly on the nanoscale ^{[34,35,36]. A suggested mechanism for the reduction of imines to their corresponding secondary amines over the nano-sized NiLa₂O₄ spinel catalyst in the presence of NaBH₄ is shown schematically in Scheme 2.}

Scheme 2. Proposed mechanism for the reduction of imines to secondary amines.

The nano-sized NiLa₂O₄ spinel catalyst is insoluble in organic solvents, therefore it can easily be recovered. The catalyst reusability was examined under the optimum reaction conditions, using N-(4-chlorobenzylidene)aniline reduction as a model reaction. In each cycle, the catalyst was recovered, washed with ethyl acetate, and dried before use in the next reaction. It is important to note that the catalyst was reused four times without significant loss of activity and selectivity, as shown in Fig. 6.

Fig. 6. Recycling of NiLa2O4 spinel nanopowder in selective reductive hydrogenation of imines to corresponding secondary amines.

The SEM and TEM images of the nano-sized NiLa₂O₄ spinel after the fourth cycle are shown in Fig. 7. The surface roughness of the used catalyst increased, which could be attributed to carbon deposition on the catalyst surface, but its activity did not decrease significantly.

Fig. 7. SEM (a) and TEM (b) images of nano-sized NiLa2O4 spinel after fourth cycle.

A literature survey was performed to compare this method with some previously reported methods, including NaBH₄ reductive hydrogenation of imines. As shown in Table 4, the nano-sized NiLa₂O₄ spinel is an efficient catalyst for this selective reduction.

Table 4 Comparison of some other catalysts with nano-sized NiLa2O4 spinel. Catalyst Amount of catalyst (eq.) Temperature (°C) Time (min) Amount of NaBH4 (eq.) Yield (%) Ref. H3BO3 1 r.t. 40 1 80 [37] Al2O3 1.4 r.t. 5−15 3 80−98 [38] Rany Ni 27−55 mol% 40−60 10−30 3.2 75−93 [39] Nano-sized NiLa2O4 0.75 mol% r.t. 2−20 0.75 90−98 This work

Table 4 Comparison of some other catalysts with nano-sized NiLa2O4 spinel.

We developed a novel, highly efficient, and easy protocol for the synthesis of secondary amines from the corresponding imines in the presence of NaBH₄. The procedure has several advantages over other methods, including improved yields, cleaner reactions, a simple experimental procedure, and the use of easily prepared and inexpensive starting materials. This approach is therefore a useful and attractive strategy in synthetic organic chemistry.

The authors gratefully acknowledge Ferdowsi University of Mashhad and Qeshm Branch, Islamic Azad University for partial support of this project (Grant No. 3/28973).