Synthesis, characterization and biological activity of tin(IV) macrocyclic complexes derived from 1,2-diphenylethane-1,2-dione dihydrazone

Introduction

The relevant transition metal complexes are of great interest in coordination chemistry. Although this subject has been extensively studied (Keypour et al., 2015a,b), the synthesis and complexing properties of the macrocycles deserve special attention as in some cases these macrocycles have been used for the construction of the corresponding polyaza macrocyclic complexes (Romain et al., 2015). The importance of macrocyclic complexes has provided a motivation for the investigation of the metal ion chemistry of biological systems as well as of cyclic ligand systems (Zhang et al., 2013). Schiff-base ligands contain an N=C=N structural unit and form a strong chelate ring with different metal ions or anionic species, with significant potential in areas such as catalysis, modeling of metalloenzymes and molecular recognition (Khan et al., 1990; Chandra and Sangeetika, 2004; Dong et al., 2014) which may affect the nature of the complex formed (Keypour et al., 2015a,b). Recently, the template synthesis of various imine macrocyclic complexes has been reported (Adams et al., 1999; Hong et al., 2005; Daszkiewicza et al., 2010) and has a long-term interest in producing imine macrocyclic complexes of a wide variety. The template condensation reaction lies at the heart of macrocyclic chemistry and has been widely used for the synthesis of macrocyclic complexes. Many of the synthetic routes to macrocyclic ligands involve the use of a metal ion template to orient the reacting groups of the ligand in the desired conformation for optimum ring closure and thereby promote the cyclization reaction (Curtis, 1968; Niasari and Davar, 2006). The extensive developments over recent decades in the use of organotin compounds as reagents or intermediates in organic synthesis prompted the preparation of many new organotin compounds (Ghani and Deo, 2014). A considerable number of organotin macrocyclic complexes with nitrogen donors have been reported over the past few decades (Chaudhary et al., 2006; Zhang et al., 2016). Herein, we report a new series of tetraimine macrocyclic complexes of the types [MLX₂Sn(CH₃)₂] and [CuLSn(CH₃)₂]X₂ [M=Mn(II), Co(II), Ni(II) and Zn(II); X=Cl or NO₃] obtained by the template condensation of dimethyl tin(IV) dichloride with 1,2-diphenylethane-1,2-dione dihydrazone (DPEDDH) in the presence of metal ions in a 2:2:1 molar ratio.

Results and discussion

The tertaiminetetraamide macrocyclic complexes [MLX₂] and [CuL]X₂ (M=Mn(II), Co(II), Ni(II) and Zn(II); X=Cl or NO₃) have been prepared by the condensation reaction of DPEDDH with dimethyl tin(IV) dichloride followed by metal templation in a 2:2:1 molar ratio (Scheme 1).

Scheme 1:

Formation and suggested structure of tin(IV) macrocyclic complexes.

The complexes examined show poor solubility in solvents such as chloroform, carbon tetrachloride and water. However, they are freely soluble in dimethylsulfoxide (DMSO) and warm dimethylformamide. The purity of the product was monitored by thin-layer chromatography using silica gel. The elemental analysis results agree well with the mononuclear structure. The molar conductivities in DMSO indicate (Geary, 1971; Khan et al., 2010) that copper(II) complexes are electrolytic while others are non-electrolytic in nature. The IR spectra of all the complexes are similar but strikingly different from the spectra of the free ligand. The spectra of the free ligand (DPEDDH) show two strong, medium-intensity prominent bands at 3290–3210 cm⁻¹, assignable to ν(N-H) vibrations of amino groups which occur at slightly lower frequencies than are usually associated with ν(N-H) vibrations of primary amino groups (Khan and Ghani, 2005). This result is expected because of the high degree of conjugation in the ligand as the interaction of the amino group lone pair with conjugated groups weakens the N-H bond. The ν(C=N) band vibration appeared at 1615 cm⁻¹, which is found to be slightly negatively shifted than that usually found for the azomethine linkage, further confirms the delocalization of electrons in the free ligand.

The reported macrocyclic complexes exhibit a strong intensity band in the range of 1580–1595 cm⁻¹ assignable (Nelson et al., 1988; Thoonen et al., 2006) to the coordinated ν(C=N). There is an observed increase in the ν(C=N) band intensity which may be due to the possible transformation of the trans structure of the free ligand (DPEDDH) into cis configuration upon complexation. The cis structure is stabilized by the chelate ring, lowering the symmetry which consequently enhances the ν(C=N) band intensity. The bands at 425–470 cm⁻¹ originate from (M-N) vibration. In addition, the IR spectra of nitrato complexes exhibit bands at 1280, 1040 and 850 cm⁻¹ which are consistent with the monodentate nature (Ball and Blake, 1969) of this group. However, the phenyl ring vibrations appear at their expected positions.

The ¹H NMR spectra of benzildihydrazone (DPEDDH) and the corresponding zinc complexes have been recorded and compared. Benzildihydrazone shows a multiplet at 7.18 ppm and a slightly broad signal at 4.92 ppm assignable (Chaudhary and Singh, 2001) to the phenyl ring (C₆H₅, 10 H) and amino (N-NH₂, 4 H) protons of the hydrazine moiety. Both the zinc complexes showed the appearance of a band at 8.42 and 8.45 ppm, reasonably ascribed (Parker et al., 1994) to the (N-NH, 4 H) protons. A multiplet at 6.90 and 6.97 ppm in both complexes is characteristic of phenyl ring (C₆H₅, 20 H) protons. The absence of characteristic peaks corresponding to the protons of hydrazone moieties further confirms that cyclization has indeed taken place.

The complexes gave a sharp signal at δ −578 ppm and δ −576 ppm in the ¹¹⁹Sn NMR spectra which is at par with the reported values for a four-coordinated geometry (Ghani and Deo, 2014).

The room temperature electron paramagnetic resonance (EPR) spectra of both polycrystalline Cu(II) complexes exhibit a similar single broad absorption band. However, no complex is found to give hyperfine splitting. The calculated g_║ and g_⊥ values appeared in the region 2.31, 2.34 and 2.08, 2.09, respectively, which support the contention that d_x²−d_y² may be the ground state of the unpaired electron. Both the complexes show g_║>2.3, indicating that they exhibit appreciable ionic character (Khan et al., 2006). The g values are related by the expression G=(g_║−2)/(g_⊥−2), and the present complexes [CuLSn(CH₃)₂]X₂ (X=Cl or NO₃) gave the axial symmetry parameter G in the 3.875, 3.775 region, indicating (Proctor et al., 1968) very small exchange interaction among copper(II) ions in these complexes, as the G values are < 4.

The electronic spectra of manganese gave two bands in the 22 400–22 550 cm⁻¹ and 18 650–18 850 cm⁻¹ regions which may be assigned to the ⁶A_1g→⁴T_2g(F) and ⁶A_1g→⁴T_1g(P) transitions, respectively, corresponding to an octahedral environment around the Mn(II) ion. The cobalt complexes exhibit two ligand field bands in the 21 400–21 700 and 15 950–16 200 cm⁻¹ regions which may reasonably correspond to the ⁴T_1g(F)→⁴T_1g(P) and ⁴T_1g(F)→⁴A_2g(F) transitions, respectively, consistent with octahedral geometry (Lever, 1984) around the Co(II) ion. All the nickel complexes exhibit two electronic bands in the 21 200–24 350 and 18 200–18 750 cm⁻¹ regions which may be ascribed to the ³A_2g(F)→³T_1g(P) and ³A_2g(F)→³T_1g(F) transitions, respectively, suggesting (Kimura, 1986) an octahedral geometry around the metal ion. Magnetic moment values are in close agreement with their electronic spectral data which further support the proposed geometry.

The observed magnetic moments of Cu(II) complexes in the range 1.87–1.92 Bohr Magneton (B.M.) are consistent with the square planar Cu(II) system (Inba et al., 2013). The electronic spectra of these complexes show a broad band maximum at 16 050 cm⁻¹ which may be assigned to the ²B_1g → ²A_1g transition. However, two weak shoulders appearing in the regions 21 500–21 750 and 11 750–12 450 cm⁻¹ may be attributed to the ²B_1g→²E_g and ²B_1g→²B_2g transitions, respectively, suggesting (Jain et al., 1977) a square-planar geometry around the Cu(II) ion.

The antifungal activities of the synthesized complexes were tested against Aspergillus fumigatus and Trichoderma harzianum by the radial growth method (Bansal and Singh, 2000) using Czapek’s agar medium. The complexes were directly mixed with the medium in 50, 100 and 200 ppm concentrations in methanol. The controls were run and three replicates were used in each case. The linear growth of the fungus was obtained by measuring the diameter of the fungal colony after 5 days. The amount of growth inhibition in all the replicates was calculated by the following equation: percent inhibition=(C−T)/C×100, where C is the diameter of the fungal colony in the control plate and T is the diameter of the fungal colony in the test plate. The experimental data of antifungal screening reveal that the metal complexes are more fungi toxic than the starting materials. The fungi toxicity of precursors and their complexes decreases on lowering the concentration. The results recorded from the biocidal activity were also compared with the standard fungicide Folicur (Table 1). The increased fungi toxicity of the complexes may be due to the chelation as the polarity of the metal ion in the complexes which reduces considerably due to the partial sharing of its positive charge with the donor groups and possible π-electron delocalization over the whole chelate ring system. This in turn increases its permeation through the lipoid layers of the fungal membrane (Sharma et al., 2001). The enhanced activities of the metal complexes relative to free macrocycles can be attributed to the increased lipophilic nature of these complexes arising due to chelation (Ibrahim et al., 2006).

It is rational to assume that more lipophilic compounds are more active simply because they enter the lipoid layers of cell membranes more briskly.

Conclusions

The tertaimine macrocyclic complexes of Mn(II), Co(II), Ni(II) and Zn(II) have been synthesized by the condensation reaction of DPEDDH with dimethyl tin(IV) dichloride followed by metal templation in an appropriate molar ratio which directs the course of the reaction. The attempts to prepare metal-free macrocyclic ligands were not successful as there was formation of oily products which could not be isolated and characterized. Analytical data, electronic spectra, magnetic susceptibility, IR, ¹H NMR and EPR data reveal mononuclear nature of all the complexes. Except for the copper(II) complexes which are electrolytic, the others are non-electrolytic in nature. The complexes exhibit moderate antifungal activity on Aspergillus fumigatus and Trichoderma harzianum.

Experimental section

All chemicals were purchased from Merck (Kenilworth, NJ, USA) and were used without further purification. Benzildihydrazone (DPEDDH) was prepared by the reaction of benzyl and hydrazinehydrate in methanol in 1:2 molar ratios. Elemental analyses were carried out on a Perkin-Elmer 240 elemental analyzer (Waltham, MA, USA). ¹H NMR spectra in DMSO-d₆ (Sigma-Aldrich, St. Louis, MO, USA) were recorded on a JEOL-FX-100-FT-NMR spectrometer (Cranford, NJ, USA), with MeSi₄ (Sigma-Aldrich, St. Louis, MO, USA) as an internal standard, and ¹¹⁹Sn spectra were recorded on Bruker ACF300 (Billerica, MA, USA) (at 300 MHz) with SnMe₄ as an external standard. The estimation of halogen was done gravimetrically (Vogel, 1961) and the metals were estimated by titrating with standard EDTA solution (Reilley et al., 1959). The FT-IR spectra (4000–400 cm⁻¹) were recorded as KBr disks on an IR 408 Shimadzu spectrophotometer (Columbia, MD, USA). The electronic spectra of the compounds were recorded on a Pye-Unicam 8800 spectrophotometer (Cambridge, UK) at room temperature. The EPR spectra of solid complexes were recorded on a JEOL JES RE2X EPR spectrometer (Arvada, CO, USA). Magnetic susceptibility measurements were carried out using a Faraday balance at 25°C. Conductance measurements were performed using a Hanna HI 8820 conductivity meter (Carrollton, TX, USA).