Palladium(II) complexes assembled on solid materials: as catalysts for the –NO₂ (nitro) to –NH₂ (amine) reactions

Abstract

Herein, a new series of [PdCl₂(L)₂] complexes where ligands are monodentate amine ligands bearing sulfonamide groups were synthesized, characterized using various techniques such as NMR, FT-IR, UV–Vis, and sc-XRD and investigated for their catalytic performance for the reduction of nitroarenes (2-nitroaniline, 4-nitroaniline, and nitrobenzene) in the presence of NaBH₄ in water under heterogeneous conditions. Because the results show that the synthesized complexes are very efficient catalysts, materials using the selected palladium(II) complex supported by multiwall carbon nanotubes, silicon dioxide, and iron(II,III) oxide (Fe₃O₄) were fabricated by a simple-impregnation methodology, characterized by FT-IR, BET, TEM, and XRD techniques and investigated for their catalytic performance for the same reaction. Thus, a series of supported catalysts was designed with the aim of both enhancing catalytic activity and reducing noble-metal contents. Our findings serve to develop simple catalytic systems and this system can be easily used for catalytic reduction reactions which are the cornerstone of the production of important chemicals.

Graphic abstract

Introduction

Nitro compounds are widely regarded as contaminants found in industrial waste and they pose a serious problem for both living organisms and the ecosystem. Also, amine compounds formed by the hydrogenation of nitro compounds are used in many areas as qualified chemicals; for example, as chelating agents in pharmaceuticals, polymers, pesticides, explosives, fibers, dyes, and cosmetics, etc. [1,2,3,4,5,6,7]. The most suitable method to produce aromatic amines which belong to the group of compounds that are more useful than nitrobenzene derivatives is an efficient, environmentally friendly, catalytic system operating in mild conditions. In this case, many processes have been developed, including adsorption/desorption, photoactive catalysts, electrochemical treatment, the e-Fenton method, electrocoagulation, hydrogenation, etc. [8,9,10,11,12,13]. The development of a catalytic system undergoes several reaction optimizations steps. Among these, substrate-related screening tests, as well as the development of suitable catalysts and, studies about increasing the activity of existing catalysts can be quite valuable. In general, the catalytic hydrogenation of nitrobenzene derivatives is made possible by catalysts containing many metals. For example, palladium, platinum, gold, and silver are the most common ones, and their catalytic performances are excellent. However, the high price of noble metals used during the catalytic process is the main disadvantage of these conventional noble-metal catalysts and limit their large-scale applications. The catalysts including other metals may also lose catalytic performance. Therefore, the idea of adding materials that can add a synergistic effect to catalysts consisting of noble metals has also been investigated by researchers. Of course, many parameters come to the fore in these catalytic systems; for example, solvent, hydrogen source, amount of substrate, amount of catalyst, etc. [14,15,16,17].

Also, avoiding the use of organic solvents and the improvement of suitable processes for hydrogenation of nitrobenzenes in aqueous media under mild conditions are still necessary. Likewise, some quite remarkable studies have been carried out with compounds bearing palladium for the hydrogenation of nitro-compounds [18,19,20,21,22,23,24,25,26,27].

Complexes or materials containing palladium are also widely used during the application of reduction processes [28], C–C or C–N bond formation reactions [29, 30], oxidation works [31, 32], hydrogenation studies [33], anti-cancer drugs [34, 35], as biosensors [36], in solar cells [37], etc. Especially, palladium complexes or materials have been the focus of intense research in catalyst chemistry as a homogeneous/heterogeneous catalyst [38,39,40]. From most of the research about palladium complexes in the literature, it is seen that at least bidentate ligands were used, because electronic and steric parameters are one of the most important cases in catalysis chemistry. The research about Pd(II) complexes with monodentate ligands is rare. On the other hand, the preparation conditions for Pd(II) complexes bearing N-donor ligands generally does not require a specific atmosphere such as argon or nitrogen gas. Recently, palladium-based catalysts, in particular, those supported by carbon [41], zeolite [42], silica [43], alumina [44], and polymer [45], have been fabricated by different methods (such as deposition–precipitation, impregnation and co-precipitation, immobilization, etc.) and used as catalysts in many organic transformations (C–C bond formation, hydrogenation, oxidation, etc.).

The studies of combining molecular and material structures with immobilization methods have been frequently seen in recent years [26, 37, 46,47,48]. In this way, the materials containing different characteristic properties are combined with a simple method. These new materials can also show synergistic effects in different application areas.

Because of all these reasons, there is increasing attention in the literature about the preparation of simple, novel, effective, and reusable catalysts for catalytic reduction of nitro compounds in aqueous media. For these purposes, herein, we report a new series of synthesized Pd(II) complexes C₁–C₆ bearing monodentate N-(2-aminophenyl)benzenesulfonamides ligands which can easily interact with supporting materials via secondary chemical bonding and tested as catalysts for the reduction of nitrobenzenes. Herein the sulfonamides ligands are easily synthesizable compounds, and the sulfonamide compounds support the immobilization methodology with the functional groups. In addition, sulfonamide derivative ligands and their complexes are often used in catalytic reactions [48,49,50].

Then, to reduce noble-metal (Pd) content, the MWCNT, SiO₂, and Fe₃O₄-supported C₄ materials M₁–M₃ were also prepared and screened for their efficiency as catalysts in the reduction of nitroarenes. The catalytic results for the supported materials clearly show that although the palladium content is reduced, the catalytic efficiencies are higher than for the complexes.

Results and discussion

The details of analytical methods

The synthesis of monoamine aromatic sulfonamide ligands L₁–L₆ was performed in a simple one-step reaction, starting from commercially available arylsulfonyl chlorides and 1,2-diaminobenzene. Their amine-coordinated Pd(II) complexes C₁–C₆ were isolated as yellow solids with high yields and were stable in the solid-state and solution. The structures of Pd(II) complexes C₁–C₆ were elucidated by FT-IR, NMR, and UV–Vis spectrophotometer.

In ¹H NMR spectra recorded from a Bruker 400 NMR spectrometer at 297 K, the methoxy and 2,4,6-trimethyl protons for C₂ and C₄ were assigned at 3.81 ppm as singlet for –OCH₃ group, at 2.24 ppm as singlet for -p-CH₃ and 2.38 ppm as singlet for o-CH₃ for 2,4,6-trimethyl protons, respectively. Also, the aromatic ring protons as multiplet peaks were located at 6.72–7.99 ppm for C₁, 6.49–7.60 ppm for C₂, 6.46–8.37 ppm for C₃, 6.45–7.09 ppm for C₄, 6.75–8.02 ppm for C₅, and 6.75–7.95 ppm for C₆. In the ¹³C NMR spectra for the N-coordinate Pd(II) complexes C₁–C₆, the peaks belonging in (c) position to the -p-OCH₃, -p-CH₃ carbons were located at 55.6 ppm and 22.6 ppm, respectively. Additionally, the carbons belonging to aromatic rings were obtained at 117.1–147.2 ppm for C₁, 113.8–162.3 ppm for C₂, 124.2–149.7 ppm for C₃, 116.7–147.3 ppm for C₄, 115.7–147.0 ppm for C₅, and 121.0–146.9 ppm for C₆. All NMR chemical shifts and integrations of signals were consistent with the proposed structures (Fig. S_1–8). FT-IR was measured with a Perkin-Elmer Spectrum 400 FTIR system. For the palladium(II) complexes C₁–C₆, the –N–H stretching frequency peaks belonging to the sulfonamide groups appeared at 3098 cm⁻¹, 3131 cm⁻¹, 3147 cm⁻¹, 3153 cm⁻¹, 3109 cm⁻¹, and 3152 cm⁻¹ and -NH₂ stretching frequency peaks were observed at 3252–3148 cm⁻¹, 3279–3165 cm⁻¹, 3300–3216 cm⁻¹, 3269–3185 cm⁻¹, 3221–3153 cm⁻¹, and 3247–3185 cm⁻¹, respectively.

The UV–Vis spectra of Pd(II) complexes C₁–C₆ in dimethyl sulfoxide (DMSO) (1 × 10^–4 mol dm⁻³) solvent were recorded within the 260–560 nm range, and a representative spectrum is shown in Fig. S9. In the spectra of the complexes, there are two main transition bands excluding C₆ complex. The first bands (280–301 nm) are attributed to the π → π* transition and the second bands in the range of 326–335 nm are attributed to the n → π* transition of the Pd(II) complexes.

Description of the crystal structure of C₁

X-ray data were collected with an STOE IPDS II diffractometer at room temperature using graphite-monochromated Mo Kα radiation by applying the ω-scan method. Data collection and cell refinement were carried out using X-AREA, while data reduction was applied using X-RED32 (X-AREA Version 1.18 and X-RED32 Version 1.04, Stoe & Cie, Darmstadt, Germany, 2002). The structure was solved by direct methods using SHELXS-2013 and refined with full-matrix least-squares calculations on F² using SHELXL-2014 [51, 52] implemented in WinGX [53] program suit. H atoms bonded to C atoms were assigned C–H distances of 0.93 Å, with U_iso(H) = 1.2U_eq(C). The nitrogen-bound hydrogens were in a difference Fourier map and refined freely. Crystal data, data collection and structure refinement details are summarized in Table 1, while relevant geometric parameters are collected in Table 2. PLATON was used for the structure analysis. Molecular graphics were generated using ORTEP-3 [53].

Table 1 Crystal data and structure refinement parameters for C₁

Table 2 Selected geometric parameters for C₁

The molecular structure of complex C₁ was confirmed using the X-ray crystallography. The compound crystallizes in the triclinic space group P$\stackrel{-}{1}$ with a single molecule in the unit cell and is shown in Fig. 1.

Fig. 1

Molecule of C₁ showing the atom-labeling scheme. Displacement ellipsoids are drawn at the 30% probability level. Hydrogen bonds are shown as dashed lines. For the sake of clarity, only hydrogen atoms involved in hydrogen bonding have been included

The asymmetric unit of the compound contains one-half of the molecule, because the palladium atom occupies a special position in the crystallographic center of inversion. The Pd(II) cation is coordinated to two chlorido ligands and the nitrogen atoms of two N-(2-aminophenyl)-benzenesulfonamide ligands, in an almost perfect square-planar geometry. The molecule adopts the trans disposition, which is commonly observed for bis(amino)-dihalopalladium(II) complexes [54, 55]. The N−Pd−Cl cis angles [89.66(4)–90.34(4)°] are very close to the optimal angle of 90°, while the Cl−Pd−Cl and N−Pd−N trans angles are exactly equal to 180°. The bond distances for Pd−N [2.0623(14) Å] and Pd−Cl [2.2850(4) Å] are comparable with those published in the literature for related complexes of general formula PdCl₂L_2, where L denotes the amino-ligands [54,55,56,57,58,59,60,61]. The square plane makes dihedral angles of 86.22(5)° and 65.53(6)° with the planes of the C1–C6 and C7–C12 benzene rings, while the dihedral angle between the two benzene rings is 64.30(7)°. In the benzenesulfonamide ligand, the N1−C7 and N2−C12 bond distances of 1.414(2) and 1.442(2) Å agree well with the single bond value, and the S1−O1 and S1−O2 distances of 1.4351(19) and 1.4213(19) Å are consistent with S=O double bonding. Atom S1 has a distorted tetrahedral configuration which is evident from the angles changing from 105.21(10)° to 120.49(13)°.

In the molecular structure of the complex, there are two intramolecular interactions of N−H···Cl type (Fig. 1), forming seven-membered rings with graph-set descriptor S(7) [62]. The 2D supramolecular structure of the complex can readily be analyzed in terms of two one-dimensional substructures. In the first substructure, amino atom N2 in the molecule at (x, y, z) acts as a hydrogen-bond donor, via atom H2B, to sulfonyl atom O1 in the molecule at (x − 1, y, z), thereby forming an ${R}_{2}^{2}(18)$ ring. The propagation of this hydrogen-bonding motif generates a chain of rings running parallel to the [100] direction (Fig. 2a). In the second substructure, benzene carbon atom C10 in the molecule at (x, y, z) acts as a hydrogen-bond donor to chlorido ligand Cl1 in the molecule at (− x − 1, − y + 1, − z + 1), so forming an ${R}_{2}^{2}(14)$ ring. Propagation of this hydrogen-bonding motif generates another chain of rings running parallel to the $\left[1\stackrel{-}{1}0\right]$ direction (Fig. 2b). Full details of the hydrogen-bonding geometry are given in Table 3.

Fig. 2

a Part of the crystal structure of C₁ showing the formation of a chain of ${R}_{2}^{2}(18)$ rings along [100]. b Part of the crystal structure of C₁, showing the formation of a chain of ${R}_{2}^{2}(14)$ rings along $\left[1\stackrel{-}{1}0\right]$. For the sake of clarity, H atoms not involved in the motif shown have been omitted

Table 3 Hydrogen-bonding geometry for C₁

Materials’ characterization

The infrared spectrum of multiwall carbon nanotubes (MWCNTs)–, SiO₂–, and Fe₃O₄– supported M₁–M₃ materials and C₄ are shown in Fig. 3. For M₁ (Fig. 3a), the carbonyl (C=O), the C=C stretching, and the –OH bending absorption bands were associated with ≈ 1800 and 1100–1300 cm⁻¹, ≈ 1500–1600 cm⁻¹ and ≈ 3500–3800 cm⁻¹ and were compatible with bare MWCNTs-COOH, respectively. For M₂ (Fig. 3b), the bare SiO₂ has two major peaks at 794 and 1047 cm⁻¹ which are attributed to Si–O bending and Si–O–Si asymmetric stretching, respectively and the related major peaks were also observed in the FT-IR spectra of M₂ material. For M₃, the absorption band at 529 cm⁻¹ belongs to the Fe–O vibration (magnetic phase). Hereby, the formation of peaks belonging to the C₄ complex does not appear in the initial curve of the support materials (MWCNT, SiO₂, Fe₃O₄).

Fig. 3

FT-IR spectra prepared materials

The bare MWCNT, SiO_2, and Fe₃O₄ support appear as (a), (b), and (c) in Fig. 3, respectively, and the formation of peaks belonging to Pd(II) complex C₄ is seen in Fig. 3. In brief, it was found that the Pd(II) complex C₄ is supported by MWCNT, SiO₂, and Fe₃O₄ after the use of the impregnation method.

Transmission electron microscopy (TEM) images were taken by the JEOL TEM-1400-EDX model device. Figure 4a–c and S10 show a representative TEM image of the prepared M₁–M₃ materials, respectively. When investigating the TEM image, it can be concluded that the characteristic shapes belonging to MWCNT, SiO₂, and Fe₃O₄ were preserved after impregnation. Also, the EDX images are given in Fig. S11 as supporting information.

Fig. 4

Representative TEM images of M₁ (a), M₂ (b), and M₃ (c)

The Barrett–Joyner–Halenda (BJH) pore volume (cm³ g⁻¹), Brunauer–Emmett–Teller (BET) surface area (m² g⁻¹), and BJH pore width (Å) were calculated by the N₂ adsorption–desorption (BET) isotherms for M₁–M₃ using a Micromeritics Gemini VII Surface Area and Porosity system. The results indicate that the BET isotherms of M₁–M₃ materials were identified as typical type IV isotherms according to IUPAC (Fig. 5). The BET surface areas (S_BET) of M₁–M₃ were found as 125.59 m² g⁻¹, 2.989 m² g⁻¹, and 54.815 m² g⁻¹, respectively. The S_BET surface area of M₁ is high compared to M₂ and M₃ (Table 4). Table 4 shows the corresponding BJH pore widths of 531.375 Å, 484.197 Å, and 152.692 Å, respectively. Also, all the N₂ adsorption–desorption (BET) isotherms data is give in the supporting information as Table S1. As expected, the surface area of the carbon nanotube structure (MWCNTs) was observed higher than other solid support materials. It is well known that the surface area is an important parameter in catalyst systems and has catalytic efficiency increasing effect. Likewise, the surface area of the Fe₃O₄ solid support is well and it is predicted that it contributes positively to the catalytic yield.

Fig. 5

BET isotherms of M₁ (a), M₂ (b), and M₃ (c)

Table 4 Adsorption–desorption characteristics of M₁–M₃ materials

X-ray powder diffraction (XRD) analyses were measured with a Bruker AXS D8 Advance Model diffractometer which was run at 20–60 kV and 6–80 mA, 2Ө = 10°–90°, and a step of 0.002° using CuKα X-ray. Powder XRD studies were performed to analyze the formation of patterns. The XRD spectra demonstrated that all the Pd(II)-impregnated MWCNT, SiO₂, and Fe₃O₄ materials produced characteristic peaks similar to those of MWCNT [JCDPS card no: 75–1621, 26.29° (002), 43.12° (200), SiO₂ and Fe₃O₄ [JCDPS card no: 19-0629, 30.1° (220), 35.5° (311), 43.1° (400), 57.0° (511), 62.9° (440)] in Fig. 6. Likewise, all the XRD data have new peaks, in addition to the data for pure supporting material (MWCNT, SiO₂, Fe₃O₄). The structure confirms the effectiveness of the simple wet impregnation method for the synthesis MWCNT-, SiO₂-, and Fe₃O₄-supported C4 materials M₁–M₃.

Fig. 6

XRD patterns of M₁ (a), M₂ (b), and M₃ (c)

Catalytic studies

We investigated the catalytic activity of C₁–C₆ complexes using the hydrogenation of 2-nitroaniline (2-NA) to o-phenylenediamine in the presence of BH₄⁻ ion in the water at ambient temperature as a precursor reaction. The functional groups of the synthesized complexes are designed to compare steric, electron-donating and electron-withdrawing groups.

The reaction was easily followed spectrophotometrically due to the reactant (2-NA) and product having different absorption bands such as λ_max = 410 nm [63, 64] (–NO₂ group of 2-NA). Initially, the 2-NA (2.5 × 10^–4 mol dm⁻³) mixture has a yellow color which is the color of the adsorption band belonging to the nitro group, this color gradually vanished because of the formation of the o-phenylenediamine product. Briefly, the absorption peak at 410 nm gradually decreased in intensity during this reaction. The percent conversions were monitored at different times between 30 and 90 s. It is well known that when the times were increased from minimum time to maximum time. The catalytic efficiencies of C₁–C₆ found as 93.2%, 94.4%, 93.8%, 94.9%, 97.0%, and 94.1% at the end of 90 s, respectively (Fig. 7). It is seen that all catalytic reactions were almost complete in 90 s, and so the synthesized Pd(II) complexes C₁–C₆ were showed excellent catalytic activity under these conditions. A difference in the functional groups could not be determined in all the catalytic activity results obtained. However, the positive contribution of the -F group, an electron-withdrawing group, was detected. Also, the C₄ complex, which shows an average activity in complex selection, was chosen as the representative of the synthesis group. In recent years, the catalytic activity of hybrid materials has been remarkable in the literature [65,66,67,68,69,70]. In light of this information, we envisaged that the catalytic activity of the tested palladium complexes C₁–C₆ can be combined with a different type of solid support material such as MWCNTs, SiO₂, and Fe₃O₄ and the performance of these new hybrid materials will be high for the reduction of nitroarenes. Thus, we selected one of the effective palladium complexes (bis[N-(2-aminophenyl)-2,4,6-trimethylbenzenesulfonamide]-di-chloro-palladium(II), C₄) as randomly (not having the best activity) for the impregnation method into MWCNTs, SiO₂, and Fe₃O₄ materials. Herein, the solid support materials selected were selected from different types and their potentials were compared, and the C₄ was chosen for the preparation of solid-supported catalysts, because the supporting material generally increases the efficiency of catalysts synergistically and we also want to clearly demonstrate this effect in this paper. The materials were chosen SiO₂ as the most common material, Fe₃O₄ as a metal oxide with magnetic properties, and MWCNTs as a carbon based and commonly used material.

Fig. 7

Time-dependent UV–Vis absorption spectra of the 2-nitroaniline reduced by NaBH₄ catalyzed by the C₁–C₆ complexes

To demonstrate the success of this methodology, the catalytic performances of bare solid support materials (MWCNTs, SiO₂, and Fe₃O₄) were tested as catalysts under the same conditions which were followed by UV–Vis spectrophotometer in the presence of NaBH₄ in water with 2-NA (Fig. 8). When the time-dependent spectrophotometer measurement results are evaluated, the catalytic performances were 50.2% (5 min), 51.4% (10 min), and 58.2% (15 min) for MWCNTs, 37.1% (5 min), 41.3% (10 min), and 54.7% (15 min) for SiO₂, 6.7% (5 min), 16.1% (10 min), and 20.9% (15 min) for Fe₃O₄. In light of these results, it can be said clearly that the catalytic conversion of bare Fe₃O₄ material is lower than other selected solid support materials. However, as known, catalysts bearing Fe₃O₄ can be recovered easily due to the magnetic properties of Fe₃O₄ (via a simple magnet). Therefore, efforts to improve the catalytic performance power belonging to Fe₃O₄ material can be quite valuable.

Fig. 8

Time-dependent UV–Vis absorption spectra of the 2-nitroaniline reduced by NaBH₄ catalyzed by the MWCNT, SiO₂, and Fe₃O₄ supported materials

Based on the above results, we tested the catalytic activity of M₁–M₃ nanomaterials for the catalytic hydrogenation of 2-nitroaniline (2-NA). As shown in Fig. 9, the specific absorption band belonging to 2-nitroaniline is observed 410 nm and when M₁–M₃ nanomaterials were added to the reaction media, the corresponding peak intensity showed a quick decline. The catalytic performances of M₁–M₃ nanomaterials for the hydrogenation of 2-NA were 96.8% (30 s), 96.9% (60 s), 97.7% (90 s); 94.5% (30 s), 96.7% (60 s), 97.6% (90 s); and 93.7% (30 s), 96.0% (60 s), 96.1% (90 s), respectively. A significant improvement in the catalytic performance of the nanomaterials M₁–M₃ was observed compared to bare materials. Interestingly, the best-enhanced catalyst was recorded as M₃ compared to bare Fe₃O₄. These results are compatible with the literature [71]. However, the success of hybrid materials formed by molecular metal complexes is limited [63]. When the catalytic performances of C₄, the bare solid support materials and the hybrid nanomaterials M₁–M₃ are examined in detail, it is evident that the hybrid nanomaterials have higher catalytic conversions than both the palladium(II) complex C₄ and the solid support materials (MWCNTs, SiO₂, Fe₃O₄). This situation may be related to the synergistic effect.

Fig. 9

Time-dependent UV–Vis absorption spectra of the 2-nitroaniline reduced by NaBH₄ catalyzed by the M₁–M₃

We also studied the hydrogenation of 4-nitroaniline (4-NA) using M₁–M₃ under the same conditions. The characteristic absorption band of the pure 4-nitroaniline arises at 380 nm from the –NO₂ group. After the addition of catalysts M₁–M₃, the characteristic band of 4-NA at 380 nm gradually decreased in intensity and the color of the solution gradually vanished as the reaction proceeded. The catalytic activities were found to be 96.7% (30 s), 97.8% (60 s), 98.1% (90 s) for M₁, 96.0% (30 s), 97.1% (60 s), 97.5% (90 s) for M₂, and 96.0% (30 s), 96.8% (60 s), 97.4% (90 s) for M₃ (Fig. 10).

Fig. 10

Time-dependent UV–Vis absorption spectra of the 4-nitroaniline reduced by NaBH₄ catalyzed by the M₁–M₃

Moreover, M₁–M₃ nanomaterials were used as catalysts for the hydrogenation of nitrobenzene under the optimized conditions. The band disappeared after hydrogenation at 265 nm from the –NO₂ group. The performance of catalysts (M₁–M₃) were 89.7% (30 s), 90.3% (60 s), 93.3% (90 s) for M₁, 89.2% (30 s), 90.4% (60 s), 90.6% (90 s) for M₂, and 90.3% (30 s), 90.5% (60 s), 91.3% (90 s) for M₃ (Fig. 11).

Fig. 11

Time-dependent UV–Vis absorption spectra of the nitrobenzene reduced by NaBH₄ catalyzed by the M₁–M₃

Furthermore, the magnetic catalyst M₃ could be recovered from the reaction mixture by filtration with the help of a magnet and used again in the next cycle. According to Fig. 12, the recovered M₃ magnetic catalyst exhibited good catalytic efficiency for at least five cycles in the hydrogenation of 2-NA with the conversion of 96.1–83.5% during the 90 s.

Fig. 12

Reusability of M₃ magnetic catalyst for the hydrogenation of 2-NA with NaBH₄ (at 90 s intervals). % Conversion = [(A_o − A_t)/A_o] × 100, A_o is the absorbance at time t = 0

Herein, the kinetic equation for the reduction of nitrobenzenes to anilines can be represented as ln(C_t/C₀) = − kt, where t is time for the reaction and, k is the apparent first‐order rate constant (s⁻¹) in Table 5. Also, the k′ = k/M parameter (M: the amount of the catalyst) is introduced for quantitative comparison and the parameter is defined as the ratio of the rate constant k to the weight of the catalyst added [72]. In Table 5, the catalytic activity rate constant parameters were compared for the palladium(II) complexes C₁–C₆ and the nanomaterials M₁–M₃ and summarized. The table showed that the parameters k and k' for the nanomaterials M₁–M₃ are higher than the corresponding palladium(II) complexes and this also supports the synergistic effect during the reduction reaction of nitrobenzenes.

Table 5 Catalytic activity rate constant of catalysts

As a general concept, the catalytic activity in any reaction depends on both the composition-type of catalyst and substrate which is detailed in the literature. A catalyzed process is controlled by different parameters in the catalytic system such as particle size (micro or nano), recovery, active surface area, reusability, surface morphology, active functional groups, the redox and electron charge potential of metal center [73]. Materials containing metal ion can appear to be especially efficient catalysts for the reduction of nitroarene or nitrobenzene compounds to aminobenzenes [15, 17, 74, 75]. The hydrogenation or reduction of nitrobenzenes has different mechanisms or pathways and potential intermediates [25, 76]. The catalyzed process/cycle began immediately by adding M₁–M₃ nanomaterials to the reaction mixture. The nitrobenzenes and ions belonging to NaBH₄ are diffused from the reaction solution to the active surface of M₁–M₃ nanomaterials and especially, around the Pd(II) ion.

Hence, the metal center (palladium(II)) and other regions (ligand and solid material) act as a tool for electron transfer from hydride ions to the nitro group which achieves the formation of the amino group. With this route, the H⁺ ions are exposed in the reaction medium for the reduction of nitrobenzenes to aminobenzenes and as the cycle rate increases, the reaction rate constant also rises. With this operation, it is clear that the catalytic cycle was rapid with the M₁–M₃ nanomaterials compared to bare MWCNTs, SiO₂, Fe₃O₄ materials. There was also a difference in catalytic activity according to the type of substrate. In this reaction medium and conditions, we would recommend that these types of hybrid materials (M₁–M₃) are preferable for the hydrogenation of nitro groups (nitrobenzenes, etc.)

Conclusion

Herein, the fabrication of a series of simple N-coordinate palladium(II) complexes C₁–C₆ and [PdCl₂(L₄)₂]@MWCNTs (M₁), [PdCl₂(L₄)₂]@SiO₂ (M₂), and [PdCl₂(L₄)₂]@Fe₃O₄ (M₃) nanomaterials and their use as catalysts for the hydrogenation of nitrobenzenes to aminobenzenes was defined. Using a basic impregnation methodology, palladium(II) complexes were immobilized on the surface of MWCNTs, SiO₂, and Fe₃O₄ and detailed structural analysis of the complex and materials was carried out using NMR, UV–Vis, FT-IR, single-crystal X-ray diffraction for C₁, BET, TEM, and XRD methods. The materials M₁–M₃ are highly efficient at room temperature for catalysis of the reduction of nitroanilines (> 95% conversion in < 1 min) with NaBH₄ as the hydrogen source in water. Moreover, the M₃ catalyst is highly durable under the optimized conditions for at least five cycles without a significant loss of activity. The catalytic studies carried out with this method have the advantages of high yields, simple synthesis methodology, and easy workup. This type of catalyst is simple and stable, and it can be recovered by an external magnet.

Experimental

All chemicals (reagents and solvents) were purchased from chemical companies (Sigma-Aldrich, Merck, and Alfa Aesar) and used as received unless otherwise stated. Ligands L₁–L₃ [32] and L₄ [77] were synthesized according to the literature.

The 400 MHz ¹H NMR and 100.56 MHz ¹³C NMR spectra were recorded at ambient temperature on a Bruker 400 NMR spectrometer. A Perkin-Elmer Spectrum 400 FT-IR system with universal ATR sampling accessory was used to obtain the FT-IR spectra. For melting point determination, an Electrothermal 9100 instrument was used with open capillary tubes. X-ray diffraction (Bruker AXS D8 Advance Model) was used to confirm for immobilization methodology. The SEM–EDX analysis employed a Leo 440 Computer Controlled Digital for surface morphological characterization. The UV–Vis spectrophotometer (Perkin-Elmer Lambda 25 UV/Vis spectrophotometers) measurements were used for monitoring the reduction of nitrophenols and dyes.

General procedure for the synthesis of ligands L₅ and L₆

A solution of 1,2-diaminobenzene (1 mmol) and triethylamine (2 mmol) in 10 cm³ tetrahydrofuran (THF) was stirred at ambient temperature for 30 min. To this, a solution of aromatic sulfonyl chlorides (1 mmol) in 5 cm³ THF was added slowly under ambient temperature overnight. The resulting solution was filtered, and the volatiles were removed in vacuo. The material was taken up in 20 cm³ dichloromethane (DCM), washed with water (3 × 5 cm³) and brine (2 × 5 cm³), dried over MgSO₄, and concentrated under reduced pressure. The product was recrystallized from DCM/diethyl ether (Scheme 1).

N-(2-Aminophenyl)-4-fluorobenzenesulfonamide (L ₅ , C ₁₄ H ₁₅ FN ₂ S)

Color: orange; yield: 75%; m.p.: 96–97 °C; ¹H NMR (400 MHz, CDCl₃): δ = 4.00 (br., 2 H, –NH₂), 6.52 (d, 1H, J = 8 Hz, –H₁), 6.51 (t, 2H, J = 8 Hz, –H₃), 6.73 (d, 1H, J = 8 Hz, –H₄), 7.05 (t, 1H, J = 8 Hz, –H₂), 7.12 (t, 2H, J = 8 Hz, –H_b), 7.77 (d, 2H, J = 8 Hz, –H_a) ppm; ¹³C NMR (100 MHz, CDCl₃): δ = 116.1, 117.1, 118.6, 120.7, 128.5, 129.1, 130.3, 134.9, 144.5, 165.3 (d, J = 260 Hz) (–C–F) ppm; FT-IR (ATR): $\overline{v}$  = 3445, 3338, 3275, 3103, 3071, 3038, 2988, 1621, 1590, 1491, 1463, 1406, 1383, 1330, 1312, 1292, 1261, 1228, 1211, 1164, 1151, 1124, 1088, 1030, 1013, 930, 899, 834, 817, 787, 745, 726, 708, 685, 670 cm⁻¹.

N-(2-Aminophenyl)-4-chlorobenzenesulfonamide (L ₆ , C ₁₄ H ₁₅ ClN ₂ S)

Color: brown; yield: 77%; m.p.: 106–107 °C; ¹H NMR (400 MHz, CDCl₃): δ = 3.90 (br., 2 H, –NH₂), 6.50 (d, 1H, J = 8 Hz, –H₁), 6.56 (t, 2H, J = 8 Hz, –H₃), 6.76 (d, 1H, J = 8 Hz, –H₄), 7.07 (t, 1H, J = 8 Hz, –H₂), 7.44 (d, 2H, J = 8 Hz, –H_b), 7.70 (d, 2H, J = 8 Hz, –H_a) ppm; ¹³C NMR (100 MHz, CDCl₃): δ = 117.2, 118.7, 120.6, 128.5, 129.0, 129.2, 129.3, 137.3, 139.6, 144.4 ppm; FT-IR (ATR): $\overline{v}$  = 3477, 3379, 3255, 3094, 3074, 2998, 2954, 2904, 1616, 1587, 1574, 1498, 1476, 1464, 1387, 1335, 1308, 1280, 1253, 1214, 1159, 1124, 1084, 1028, 1014, 1007, 973, 941, 930, 902, 840, 814, 755, 744, 726, 703, 667, 649, 621, 567, 555, 534, 486, 469 cm⁻¹.

General procedure for the synthesis of Pd(II) complexes C₁–C₆

A solution of PdCl₂ (0.2 mmol) in 5 cm³ methanol was slowly transferred to a solution of N-(2-aminophenyl)benzenesulfonamides (0.4 mmol) in 5 cm³ methanol. The mixture was slowly warmed to 50 °C for 30 min, cooled to ambient temperature, and stirred overnight before the removal of the volatiles in vacuo. The crude product was washed with diethyl ether (3 × 10 cm³) and dried under reduced pressure to give C₁–C₆ (Scheme 1).

Bis[ N -(2-aminophenyl)benzenesulfonamide]-dichloro-palladium(II) (C ₁ , C ₂₈ H ₃₂ Cl ₂ N ₄ PdS ₂ )

Color: light yellow; yield: 79%; m.p.: 310–311 °C; ¹H NMR (400 MHz, DMSO-d₆): δ = 6.58 (s, 4H, –NH₂), 6.72 (t, 2H, J = 8 Hz, –H₃), 6.94 (t, 2H, J = 8 Hz, –H₂), 7.07 (d, 2H, J = 8 Hz, –H₁), 7.46 (d, 2H, J = 8 Hz, –H₄), 7.51 (d, 4H, J = 8 Hz, –H_b), 7.99 (d, 4H, J = 8 Hz, –H_a) ppm; ¹³C NMR (100 MHz, DMSO-d₆): δ = 117.1, 121.1, 126.8, 128.8, 129.9, 131.5, 133.8, 138.1, 143.0, 147.2 ppm; FT-IR (ATR): $\overline{v}$ = 3252 (–NH₂), 3148 (–NH₂), 3098 (-NH), 3030, 1608, 1570, 1488, 1454, 1445, 1303, 1286, 1264, 1243, 1208, 1174, 1161, 1141, 1117, 1107, 1081, 1039, 1022, 999, 978, 958, 862, 848, 818, 773, 749, 719, 703, 686 cm⁻¹; UV–Vis (DMSO, c = 1 × 10^–4 mol dm⁻³): λ_max (ε) = 280 (9831), 327 (5611) nm (mol⁻¹ dm³ cm⁻¹).

CCDC 1,483,076 contains supplementary crystallographic data for the compound C₁ reported in this article. These data can be obtained free of charge via https://www.ccdc.cam.ac.uk/conts/retrieving.html, or from the Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: (+ 44) 1223-336-033; or e-mail: deposit@ccdc.cam.ac.uk.

Bis[ N -(2-aminophenyl)-4-methoxybenzenesulfonamide]-dichloro-palladium(II) (C ₂ , C ₃₀ H ₃₆ Cl ₂ N ₄ O ₂ PdS ₂ )

Color: yellow; yield: 71%; m.p.: 308–309 °C; ¹H NMR (400 MHz, DMSO-d₆): δ = 3.43 (br., 2H, –NH), 3.82 (s, 6H, –OCH₃), 6.50 (t, 2H, J = 8 Hz, –H₃), 6.58 (s, 4H, -NH₂), 6.64 (d, 2H, J = 8 Hz, –H₁), 6.70 (d, 2H, J = 8 Hz, –H₄), 6.92 (t, 2H, J = 8 Hz, –H₂), 7.07 (d, 4H, J = 8 Hz, –H_b), 7.60 (d, 4H, J = 8 Hz, –H_a) ppm; ¹³C NMR (100 MHz, DMSO-d₆): δ = 55.6 (–OCH₃), 113.8, 114.1, 126.9, 127.4, 128.7, 129.0, 131.5, 135.2, 141.9, 162.3 ppm; FT-IR (ATR): $\overline{v}$ = 3279 (–NH₂), 3216 (–NH₂), 3165 (–NH₂), 3131 (–NH), 3072, 2996, 2974, 1593, 1574, 1496, 1463, 1445, 1406, 1332, 1311, 1282, 1261, 1248, 1182, 1149, 1129, 1088, 1030, 1009, 982, 945, 906, 835, 802, 794, 761, 750, 724, 669 cm⁻¹; UV–Vis (DMSO, c = 1 × 10^–4 mol dm⁻³): λ_max (ε) = 300 (12,902), 326 (6934) nm (mol⁻¹ dm³ cm⁻¹).

Bis[ N -(2-aminophenyl)-4-nitrobenzenesulfonamide]-dichloro-palladium(II) (C ₃ , C ₂₈ H ₃₀ Cl ₂ N ₆ O ₄ PdS ₂ )

Color: dark yellow; yield: 79%; m.p.: 311–312 °C; ¹H NMR (400 MHz, DMSO-d₆): δ = 6.99 (s, 4H, –NH₂), 6.48 (t, 2H, J = 8 Hz, –H₃), 6.68 (d, 2H, J = 8 Hz, –H₁), 6.99 (d, 2H, J = 8 Hz, –H₄), 7.16 (t, 2H, J = 8 Hz, –H₂), 7.91 (d, 4H, J = 8 Hz, –H_b), 8.37 (d, 4H, J = 8 Hz, –H_a) ppm; ¹³C NMR (100 MHz, DMSO-d₆): δ = 124.2, 124.4, 126.4, 127.1, 127.5, 128.0, 128.5, 141.3, 145.7, 149.7 ppm; FT-IR (ATR): $\overline{v}$   = 3300 (–NH₂), 3246 (–NH₂), 3216 (–NH₂), 3147 (-NH), 3103, 3072, 1607, 1578, 1527, 1498, 1478, 1466, 1414, 1401, 1367, 1351, 1340, 1318, 1303, 1292, 1280, 1249, 1238, 1167, 1117, 1088, 1044, 1031, 1014, 949, 938, 914, 853, 832, 796, 764, 748, 738, 716, 699, 682, 663 cm⁻¹; UV–Vis (DMSO, c = 1 × 10^–4 mol dm⁻³): λ_max (ε) = 295 (1668), 336 (6314) nm (mol⁻¹ dm³ cm⁻¹).

Bis[ N -(2-aminophenyl)-2,4,6-trimethylbenzenesulfonamide]-dichloro-palladium(II) (C ₄ , C ₃₄ H ₄₄ Cl ₂ N ₄ PdS ₂ )

Color: yellow; yield: 75%; m.p.: 256–257 °C; ¹H NMR (400 MHz, DMSO-d₆): δ = 2.24 (s, 6H, -p-CH₃), 2.38 (s, 12H, o-CH₃), 6.45 (d, 2H, J = 8 Hz, –H₁), 6.60 (t, 2H, J = 8 Hz, –H₃), 6.75 (d, 2H, J = 8 Hz, –H₄), 6.81 (t, 2H, J = 8 Hz, –H₂), 6.98 (s, 4H, –H_b) ppm; ¹³C NMR (100 MHz, DMSO-d₆): δ = 20.5, 22.6, 116.7, 119.8, 127.7, 131.6, 134.2, 137.1, 138.8, 140.4, 141.7, 147.3 ppm; FT-IR (ATR): $\overline{v}$  = 3269 (–NH₂), 3185 (–NH₂), 3153 (–NH), 3070, 2987, 2940, 1599, 1494, 1465, 1453, 1408, 1393, 1382, 1332, 1299, 1274, 1246, 1191, 1161, 1138, 1117, 1105, 1052, 1043, 1038, 1023, 958, 894, 864, 853, 792, 753, 716, 672, 660 cm⁻¹; UV–Vis (DMSO, c = 1 × 10^–4 mol dm⁻³): λ_max (ε) = 289 (13,725), 335 (6553) nm (mol⁻¹ dm³ cm⁻¹).

Bis[ N -(2-aminophenyl)-4-fluorobenzenesulfonamide]-dichloro-palladium(II) (C ₅ , C ₂₈ H ₃₀ Cl ₂ F ₂ N ₄ PdS ₂ )

Color: dark yellow; yield: 73%; m.p.: 260–261 °C; ¹H NMR (400 MHz, DMSO-d₆): δ = 6.56 (s, 4H, –NH₂), 6.75 (t, 2H, J = 8 Hz, –H₃), 6.96 (t, 2H, J = 8 Hz, –H₂), 7.07 (d, 2H, J = 8 Hz, –H₁), 7.35 (4H, –H_b), 7.50 (d, 2H, J = 8 Hz, –H₄), 8.02 (4H, –H_a) ppm; ¹³C NMR (100 MHz, DMSO-d₆): δ = 115.7, 120.9, 125.9, 126.6, 129.6, 135.7, 139.5, 141.3, 147.0, 163.6 (d, J = 240 Hz) (–C–F) ppm; FT-IR (ATR): $\overline{v}$  = 3221 (–NH₂), 3153 (–NH₂), 3109 (–NH), 3081, 3013, 2829, 2781, 1615, 1590, 1574, 1539, 1492, 1455, 1403, 1306, 1296, 1289, 1272, 1252, 1235, 1216, 1192, 1166, 1144, 1127, 1112, 1098, 1081, 1042, 1029, 1012, 981, 960, 858, 843, 830, 813, 763, 743, 709, 701, 679, 664 cm⁻¹; UV–Vis (DMSO, c = 1 × 10^–4 mol dm⁻³): λ_max (ε) = 301 (13,291), 335 (7698) nm (mol⁻¹ dm³ cm⁻¹).

Bis[ N -(2-aminophenyl)-4-chlorobenzenesulfonamide]-dichloro-palladium(II) (C ₆ , C ₂₈ H ₃₀ Cl ₄ N ₄ PdS ₂ )

Color: light yellow; yield: 71%; m.p.: 331–332 °C; ¹H NMR (400 MHz, DMSO-d₆): δ = 6.55 (s, 4H, –NH₂), 6.75 (t, 2H, J = 8 Hz, –H₃), 6.97 (t, 2H, J = 8 Hz, –H₂), 7.06 (d, 2H, J = 8 Hz, –H₁), 7.51 (d, 2H, J = 8 Hz, –H₄), 7.58 (d, 4H, J = 8 Hz, –H_b), 7.95 (d, 4H, J = 8 Hz, –H_a) ppm; ¹³C NMR (100 MHz, DMSO-d₆): δ = 121.0, 121.4, 125.9, 126.5, 128.6, 128.8, 135.7, 136.1, 142.0, 146.9 ppm; FT-IR (ATR): $\overline{v}$  = 3247 (–NH₂), 3185 (–NH₂), 3152 (–NH), 3102, 2908, 1612, 1574, 1489, 1476, 1456, 1394, 1306, 1293, 1266, 1240, 1208, 1169, 1142, 1115, 1107, 1080, 1037, 1014, 983, 949, 865, 847, 821, 811, 774, 748, 709, 648 cm⁻¹; UV–Vis (DMSO, c = 1 × 10^–4 mol dm⁻³): λ_max (ε) = 287 (12,236) nm (mol⁻¹ dm³ cm⁻¹).

General procedure for the preparation of MWCNT, SiO₂, and Fe₃O₄-supported C₄ (M₁–M₃)

C₄@MWCNTs (M₁), C₄@SiO₂ (M₂), and C₄@Fe₃O₄ (M₃) materials were prepared by modifying the published procedure [37, 63, 78]. Raw MWCNTs (698,849 Aldrich) was refluxed with a magnetic stirrer in a mixture of HNO₃ (78%) and H₂SO₄ (98%) (1/3 v/v) at 130 °C. This process ensured the formation of the carboxylic acid (R–COOH) groups on the sidewalls of MWCNTs. The final product MWCNTs–COOH was filtered and washed with pure water (5  cm³, five times), and dried under reduced pressure at 40 °C [37].

In a 20 cm³ glass tube, 90 mg of MWCNTs, SiO₂, or Fe₃O₄ (from FeCl₃·6H₂O and FeSO₄·4H₂O with stoichiometric ratios in the presence of ammonia in water [71]) was dispersed in 10 cm³ of diethyl ether via an ultrasonic device, and then 10 mg of C₄ complex as a model complex (this weight ratio was selected to set the amount of catalyst used in each catalytic experiment) was added to the pot. The final mixture was sonicated for 3 h to form a stable suspension at ambient temperature. Then, the volatiles were evaporated in vacuo and dried under reduced pressure. The prepared materials were structurally analyzed by FT-IR, TEM, BET, and XRD.

Catalytic reduction of nitroarenes

The catalytic activities of C₁–C₆ and M₁–M₃ were examined for the reduction of nitroarenes (2-nitroaniline, 2-NA, 4-nitroaniline, 4-NA, nitrobenzene, NB) to aminoarenes in the presence of NaBH₄ as a hydrogen source in the water at ambient temperature.

In a typical reduction reaction, 2.5 mg of the catalysts were added to nitroarenes (2.5 × 10^–4 mol dm⁻³) and NaBH₄ (0.015 M, freshly) in aqueous solution (10 mol dm⁻³) and stirred at ambient temperature for a period of the desired time. It is important that the reaction started instantly after the addition of the catalyst. At the end of the period followed, the samples are taken from the reaction and filtered through the micro column. The progress of the reaction was monitored using an ultraviolet–visible (UV–Vis) spectrophotometer (Perkin-Elmer Lambda 25). The catalytic activities of all the materials were monitored by comparing the bands which appeared and disappeared after reduction on the UV–Vis spectrum.