Introduction

In the last decade, Schiff bases have received much attention, mainly because of their extensive application in the field of synthesis and catalysis [1, 2]. Their complexes have also been amongst the most widely studied coordination compounds in the past few years [3–5]. This is due to the fact that Schiff bases offer opportunities for inducing substrate chirality, tuning the metal-centered electronic factors, and enhancing the solubility and stability of either homogeneous or heterogeneous catalysts [6, 7]. They are also becoming increasingly important as biochemical, analytical, and antimicrobial reagents [8]. These ligands are readily available, versatile, and depending on the nature of the starting materials (primary amines and carbonyl precursors), they exhibit various functionalities. Moreover, the number, the nature, and the relative position of the donor atoms of Schiff base allow a good control over the stereochemistry of the metallic centers, as well as over the number of the metal ions within homo-and heteropolynuclear complexes [9, 10]. Several complexes coordinated through the Schiff base ONNO donors have been studied as oxygen carriers and they are useful models for bioinorganic processes [11]. The cobalt(II) and manganese(II) complexes with tetradentate Schiff bases which coordinate through N2O2 donor atoms have been extensively studied as oxygen carriers and also as catalysts for water-splitting systems [12]. Further, significant advances have also been achieved in the field of straightforward preparation of new chiral ligands such as chiral diamines and complexes for asymmetric catalysis [13]. For example, complexes based on chiral diamines have been employed for chiral hydrogenations and transfer hydrogenations [14]. Both homogeneous and heterogeneous diamine complexes have proved highly effective in enantioselective catalysis [15, 16].

Chiral metal-catalyzed reactions are potent tools for the synthesis of useful chiral compounds. Thus, various chiral metal complexes have been synthesized and used as asymmetric catalysts [17]. Some of them show high asymmetric induction and can be applied to a wide range of asymmetric reactions. Chiral complexes of the well-known Schiff base salen constitute one such class of complexes: a wide variety of these complexes have been synthesized and successfully used as catalysts for a range of asymmetric reactions, such as epoxidation, aziridination, cyclopropanation, and so on [18, 19]. Although high enantioselectivity has been achieved in these transformations, some of the reactions are still unsatisfactory in terms of atom efficiency and environmental considerations, improvement of which is a key issue to realize a sustainable society from the synthetic viewpoint.

In view of the versatile importance of Schiff bases, their complexes and their use in homogeneous and heterogeneous catalysis [20] and in continuation of our recent research [21–23] on synthesis and characterization of new Schiff base complexes and their use as catalysts for cyclopropanation of olefins, we describe herein the synthesis and characterization of the tetradentate Schiff bases 6,6′-(1E,1′E)-(4,5-dimethyl-1,2-phenylene)bis(azan-1-yl-1-ylidene)bis(methan-1-yl-1-ylidene)bis(7-hydroxy-5-methoxy-2-methyl-4H-chromen-4-one) H2L1 and 6,6′-(1E,1′E)-cyclohexane-1,2-diylbis(azan-1-yl-1-ylidene)bis(methan-1-yl-1-ylidene)bis(7-hydroxy-5-methoxy-2-methyl-4H-chrome n-4-one) H2L2 and their Co(II), Cu(II), Ni(II), and Fe(III) complexes. The coordination behavior of these Schiff bases toward transition metals was investigated via the IR, molar conductance, and magnetic moment measurements. 13C NMR spectra were also obtained to determine the structures of the ligands and some of their complexes. The catalytic activity of complexes [L1Cu].H2O (3), [L2Cu].2H2O (6) and [L2*Co].2H2O (9) toward the decomposition of ethyl diazoacetate (EDA) in the presence of olefins to yield cyclopropane products has also been investigated. Although various methodologies are available today, the cycloaddition of carbenoids to C=C double bonds is practically the most important one to construct this class of compounds. Stereochemical control of these reactions, such as differentiation of the enantiotopic faces of the double bond or cis–trans selectivity imposed on the cyclopropanation reaction by the presence of a substituent at the carbenoid carbon are the main issues [24, 25].

Experimental

Materials

All the reagents employed for the preparation of the ligands and their complexes were of the best grade available and used without further purification. 7-hydroxy-5-methoxy-2-methyl-4-oxo-4H-chromene-6-carbaldehyde used for the synthesis of the ligands was prepared by oxidation of visnagin according to the reported method [26]. Unless otherwise stated, all catalytic tests were carried out under an atmosphere of purified dinitrogen using modified Schlenk techniques. Solvents were dried and distilled before use by standard methods. Benzene, cyclohexene, and 1-octene were distilled over sodium; styrene and α-methyl styrene were distilled over calcium hydride and stored under dinitrogen.

Physical measurements

The ligands and their metal complexes were analyzed for C, H, N, and M contents at the Microanalytical Laboratory, Faculty of Science, Cairo University, Egypt. Analytical and physical data of the ligands H2L1and H2L2 and their metal complexes are reported in (Table 1). The metal ion contents of the complexes were also determined [27–29] by the previously reported methods [30, 31]. IR spectra of the ligands and their metal complexes were measured using KBr discs with a Jasco FT/IR 300E Fourier transform infrared spectrophotometer covering the range 400–4,000 cm−1 and in the 500–100 cm−1 region using polyethylene-sandwiched Nujol mulls on a Perkin Elmer FT-IR 1650 spectrophotometer. 1H and 13C NMR spectra were obtained on Brucker Avance 300-DRX or Avance 400-DRX spectrometers. Chemical shifts (ppm) are reported relative to TMS. The mass spectra were run at 70 eV with HP MODEL: MS. 5988A and/or GCMS. Cap 1000 EX Shimadzu spectrometer using electron impact technique. The electronic spectra of the ligands and their complexes were obtained in Nujol mulls and in saturated DMSO solutions using a Shimadzu UV–240 UV–Visible recording spectrophotometer. Molar conductivities of the metal complexes in DMSO (10−3 M) were measured using a dip cell and a Bibby conductimeter MC1 at room temperature. The resistance measured in Ohms and the molar conductivities were calculated according to the equation: Λ = V × K × M w/g × Ω, where Λ is the molar conductivity (Ohm−1 cm2 mol−1), V is the volume of the complex solution (mL), K is the cell constant 0.92 cm−1, Mw is the molecular weight of the complex, g is the weight of the complex, and Ω is the resistance measured in Ohms. Magnetic moments at 298 K were determined using the Gouy method with Hg[Co(SCN)4] as calibrant. The solid ESR spectra of the complexes were recorded with an ELEXSYS E500 Brucker spectrometer in 3-mm Pyrex tubes at 298 K. Diphenylpicrylhydrazide (DPPH) was used as a g-marker for the calibration of the spectra. Mass spectra of the solid ligand were recorded using a JEUL JMS-AX-500 mass spectrometer.

Table 1 Analytical and physical data of the ligands H2L1, H2L1 and their metal complexes
Full size table

Synthesis of the Schiff bases

Synthesis of H2L1

A solution of 4,5-dimethylbenzene-1,2-diamine (0.15 g, 1.1 mmol) in ethanol (10 mL) was added dropwise to a hot (75 °C) solution of 7-hydroxy-5-methoxy-2-methyl-4-oxo-4H-chromene-6-carbaldehyde (0.5 g, 2.1 mmol) in ethanol (15 mL) and refluxed for 2 h. The reaction mixture was condensed to (10 mL) and the brown precipitate was then filtered off, washed with cold methanol, and dried under vacuum over anhydrous CaCl2. 1H NMR (300 MHz, CDCl3): δ = 14.74 ppm [S, 2H, H(5) and H(22)], 9.11 ppm [S, 2H, H(16) and H(33)], 7.10 ppm [S, 2H, H(35) and H(38)], 6.74 ppm [S, 2H, H(1) and H(18)], 6.00 ppm [S, 2H, H(10) and H(27)], 4.01 ppm [S, 6H, –OCH3 (15) and –OCH3 (32)], 2.39 ppm [S, 6H, CH3 (43) and CH3 (44)], 2.33 ppm [S, 2H, CH3 (14) and CH3 (31)]. 13C NMR (300 MHz, CDCl3, 300 K): C(2) and C(19) 166.2, C(11) and C(28) 164.2, C(36) and C(37) 159.1, C(16) and C(33) 158.9, C(4) and C(21) 140.0, C(6) and C(23) 137.7, C(35) and C(38) 121.2, C(10) and C(27) 111.8, C(1) and C(18) 101.4, C(15) and C(32) 64.7, C(9) and C(26) 58.9, C(14), C(31), C(43), and C(44) 19.9 and 20.5 (Scheme 1).

Scheme 1
scheme 1

Schematic representation for the formation of the Schiff base ligand H2L1 and the numbering scheme for ligand H2L1

Full size image

Synthesis of H2L2

A solution of (±)-trans-1,2-diaminocyclohexane (in case of ligand (±)H2L2) or (1R,2S)-cyclohexane-diamine (in case of ligand H2L2*) (0.24 g, 2.1 mmol) in ethanol (5 ml) was added dropwise to a hot (75 °C) solution of 7-hydroxy-5-methoxy-2-methyl-4-oxo-4H-chromene-6-carbaldehyde (1.00 g, 4.3 mmol) in ethanol (20 ml) and refluxed for 2 h. The yellow precipitate was then filtered off, washed with cold methanol, and dried under vacuum over anhydrous CaCl2. 1HNMR (300 MHz, DMSO): δ = 15.21 [S, broad, H(5) and H(22)], 8.72 [S, 2H, H(16) and H(33)], 6.50 [m, 2H, H(1) and H(18)], 5.92 [S, 2H, H(10) and H(27)], 3.69 [S, 6H, –OCH3(15) and –OCH3(32)], 3.69 [S, 2H, H(39) and H(40)], 2.23 [S, 2H, CH3(14) and CH3(31)], 1.75 [S, 6H, H cyclohexyl]. 1H NMR (300 MHz, CDCl3): δ = 14.94 ppm [S, broad, H(5) and H(22)], 8.66 ppm [S, 2H, H(16) and H(33)], 6.59 ppm [m, 2H, H(1) and H(18)], 5.91 ppm [S, 2H, H(10) and H(27)], 3.76 ppm [S, 6H, –OCH3(15) and –OCH3(32)], 3.46 ppm [t, 2H, J = 7.6, H(39) and H(40)], 2.25 ppm [S, 2H, CH3(14) and CH3(31)], 1.85 ppm [S, 6H, H cyclohexyl]. 13C NMR (300 MHz, DMSO, 300 K): C(2) and C(19) 168.1, C(11) and C(28) 164.4, C(16) and C(33) 162.7, C(4) and C(21) 145.4, C(6) and C(23) 142.6, C(10) and C(27) 110.9, C(1) and C(18) 101.4, C(39) and C(40) 69.8, C(15) and C(32) 63.7, C(35), C(38), C(36) and C(37) 24.2 and 32.3, C(14) and C(31) 20.2. αD = +169.9 (Scheme 2).

Scheme 2
scheme 2

Schematic representation for the formation of the Schiff base ligand H2L2 and the numbering scheme for ligand H2L2

Full size image

Synthesis of the metal complexes

The metal complexes of the ligands were prepared by mixing a hot (60 °C) methanolic solution of the metal salt with the required amount of a hot (75 °C) ethanolic solution of the ligand to form 1:1 (metal/ligand) complexes, as shown in (Table 2). The two cobalt complexes 5 and 9 were prepared and filtered off under dinitrogen atmosphere, since their solutions are sensitive to air but stable when they become dry. The reaction mixture was then refluxed for a time depending on the transition metal salt used. The precipitates formed were filtered off, washed with ethanol, then with diethyl ether and dried under vacuum over anhydrous CaCl2.

Table 2 Amounts of the reactants used in the formation of the metal complexes
Full size table

Typical procedure for the catalytic cyclopropanation of olefins

EDA (see captions of Tables 5, 6 for quantities for [L2*Co].2H2O (9)) was added to a suspension of the complex 3, 6, or 9, the olefin and 1-methylimidazol (NMI) in dichloroethane (10 mL) (only in case of complex 9). The resulting pale violet or pale orange solution was heated at the required temperature under stirring. The reaction was followed until total consumption of the EDA (IR absorbance, υmax = 2114 cm−1, <0.025) was observed. The final solution was analyzed by GC–MS after the addition of 2,4-dinitrotoluene as internal standard.

Results and discussion

The elemental and physical data of the ligands H2L1 and H2L2 and their complexes (Table 1) showed that the stoichiometry of the complexes obtained is 1:1 (metal:ligand).

Mass spectra of the ligands

The mass spectra of the Schiff bases H2L1 and H2L2 revealed the molecular ion peaks at m/e 569 and 547, which are coincident with the formulae weights (568.6) and (546.6), respectively, for these ligands and support the identity of their structures. The fragmentation pathway pattern of the mass spectrum of the ligands is depicted in Schemes 3 and 4.

Scheme 3
scheme 3

The pathway fragmentation pattern of the mass spectrum of H2L1

Full size image
Scheme 4
scheme 4

The pathway fragmentation pattern of the mass spectrum of H2L2

Full size image

Optical activity

Ligand H2L2 was prepared either from the racemic mixture of the aldehyde (±)-trans-1,2-diaminocyclohexane to give the racemic mixture (±)H2L2 or from (1R,2S)-1,2-diaminocyclohexane to give the optically active H2L2*, which showed optical rotation of αD = +169.9.

Infrared spectra

The IR data for the Schiff bases H2L1 and H2L2 and their complexes are listed in (Table 3). The IR spectra of the complexes are compared with those of the free ligand in order to determine the coordination sites that may be involved in chelation. The IR spectra of the free ligands showed a strong band in the region 1,642–1,662 cm−1, which is characteristic of the azomethine group. Coordination of the Schiff bases to the metal through the nitrogen atom is expected to reduce the electron density and so the azomethine frequency. The band due to azomethine nitrogen υ(C=N) shows a modest decrease in the stretching frequency for the complexes, being shifted to around 1,614–1,630 cm−1, which indicates the coordination of the azomethine nitrogen [32, 33]. The bands at 1,218 and 1226 cm−1 for [H2L1], 1,222 cm−1 for [H2L2] are ascribed to the phenolic C–O stretching vibrations. These bands are shifted upward due to O–metal coordination [34]. This assignment is further supported by the disappearance of the υ(OH) band in the range 3,246–3,250 cm−1 for all the complexes, indicating the subsequent deprotonation of the phenolic proton prior to coordination [35]. The bands in the 510–540 and 460–475 cm−1 regions are ascribed to the formation of M–O and M–N bonds, respectively [36, 37], which further supports the coordination of the azomethine nitrogen and the phenolic oxygen. The presence of coordinated water in the complexes [38] is indicated by a broad band in the region 3,420–3,445 cm−1.

Table 3 IR frequencies of the bands (cm−1) of the ligands H2L1, H2L1, and their metal complexes and their assignments
Full size table

NMR studies of diamagnetic Ni(II) complexes

Comparison of 1HNMR spectral data of the diamagnetic Ni(II) complexes, [L1Ni].3H2O and [L2Ni].H2O further supplements the conclusion drawn from IR data. The 1H NMR spectra of the Schiff bases revealed signals from azomethine hydrogen at 9.11 and 8.72 ppm for H2L1 and H2L2, respectively, and singlets at 14.74 ppm (s, 2H) for H2L1 and 15.21 ppm (s, 2H) for H2L2 from OH resonances, which is clearly connected with the intramolecular hydrogen bonds system in H2L1 and H2L2. Upon coordination, in the 1H NMR spectra of H2L1 and H2L2 signals from OH, in the range 14.74–15.21 for the free ligands were absent, indicating deprotonation of the hydroxyl group due to Ni–O bond formation. For [L2Ni].H2O complex, the multiplet from H(55) and H(56) in the cyclohexyl ring is centered at 2.19 ppm. This signal is thus shifted toward lower frequencies, giving a coordination shift ∆coord: −1.5 ppm, which is in favor of linkage via N=CH [39, 40]. The chemical shifts of H(17) and H(34) were found at 8.70 and 8.07 ppm for complexes: [L1Ni]·3H2O and [L2Ni]·H2O, respectively. These protons exhibited large coordination shifts, because they are located three bonds from the central ion and charge distortion on these hydrogens is most noticeable [41]. The mode of bonding suggested above has also been confirmed by the 13C NMR spectroscopic data. The comparison of the 13C NMR spectra of the complexes and ligands lead to a similar conclusion and indicate changes in the charge redistribution in the whole molecule induced by the complexation. The azomethine (C=N) carbon signal occurring as singlet at δ 158.9–162.7 ppm in the ligands moves upfield, to δ 150.4–154.4, upon complexation, which provides further support for the coordination of the ligand through the azomethine nitrogens to metal ions. Also, the occurrence of the C=N carbon as one singlet indicates the equivalence of the two C=N carbon atoms both in the Schiff bases and their complexes.

Electronic and magnetic susceptibility data

The two bands observed at 270–320 and 360–490 nm in the free ligands are reasonably assigned to π → π* and n → π* transitions for the phenolic-OH and azomethine moieties, respectively. The blue or red shifts of these bands with respect to the ligand depend on the type of metal ions coordinated to the ligand and indicate coordination of phenolic-OH and azomethine moieties to the metal ions.

The electronic spectrum of the Co(II) complex [L2Co(H2O)2]·2H2O (5) gives three bands, two of them at 800 and 531 nm. [40] These are assigned to the transitions 4T1g(F) → 4T2g(F) and 4T1g(F) → 4A2g(F), respectively, suggesting that there is an octahedral geometry around Co(II) ion [40]. The third one at 396 nm refers to the charge transfer band. The magnetic susceptibility measurements lie at 4.80 B.M. (normal range for octahedral Co(II) complexes is 4.3–5.2 B.M.), indicative of octahedral geometry [42]. The cobalt(II) complex [L2*Co]·2H2O (9) exhibits absorption bands at 1,285 and 735 nm which can be assigned to 2B2g ← 2Eg and 2A1g ← 2Eg transitions, respectively [43, 44], in a square planar field. This cobalt complex displayed a magnetic moment of 2.68 B.M. which corresponds to a low spin square planar geometry [28].

The Cu(II) complexes register a broad band at 546 nm due to 2B1g → 2Eg transition of a square planar Cu(II) structure [44]. The magnetic moment of these Cu(II) complexes is 1.8–1.9 B.M. indicating the presence of one unpaired electron. The electronic spectral results clearly indicated square planar geometry for the Cu(II) system, which should contain one unpaired electron, where μeff value would be in the range 1.8–2.1 B.M. [45].

The nickel(II) complexes [L1Ni]·3H2O and [L2Ni]·H2O show two shoulders in the ranges 539–542 nm and 440–442 nm, which are characteristic of square planar diamagnetic Ni(II) complexes [31], and resemble the absorption bands observed in the spectrum of an analogous Ni(II) complex reported by Bouwman [46]. A third band which is expected to appear at a higher energy is apparently hidden under the charge transfer band.

The electronic spectrum of the Fe(III) complex showed broad bands at 1,182 and 470 nm. The former band may be due to the spin forbidden transition 6A1g → 4T2g(G), which may gain intensity as a result of the vibronic mechanism in the octahedral field around ferric ion. The second band may be attributed to 6A1g → 4T1(G) transitions. In addition, a third absorption band with high intensity observed at 470 nm can be assigned to a charge transfer transition. The observed magnetic moment of the Fe(III) complex is 5.75 B.M. Thus, these complexes have octahedral geometry [47] (Table 4).

Table 4 The electronic absorption spectral bands (nm) and magnetic moment (B.M) for the ligands H2L1, H2L2 and their Metal complexes
Full size table

Molar conductivities of the complexes

The molar conductivities Λm of the metal complexes dissolved in DMF or DMSO at 10−3 M were found to be in the range 8–15 Ω−1cm2mol−1. These low values indicate that all these complexes are non-electrolytes due to the absence of any counter ions in their structures [48, 49] and that the [H2L1] and [H2L2] Schiff bases are coordinated to the metal ions as doubly negatively charged anions. Therefore, it seems that the two phenolic-OH have been deprotonated and bonded to the metal ions as oxygen anions [50].

Catalytic activity

The catalytic activity of the complexes [L1Cu]·H2O (3), [L2Cu]·2H2O (6) and [L2*Co]·2H2O (9) in cyclopropanation reactions has been investigated. As a model reaction we chose the cyclopropanation of α-methyl styrene by EDA (EDA = ethyl diazoacetate) (Scheme 5). Catalytic reactions were run by adding the EDA to a stirred solution containing the olefin and the metal complex in dichloroethane under dinitrogen.

Scheme 5
scheme 5

Schematic representation of the model reaction

Full size image

Both copper complexes 3 and 6 have low catalytic activity as cyclopropanation catalysts and at 75 °C in 1,2-dichloroethane at a [Cu]/EDA/olefin ratio of 1:20:150 we observed a conversion of less than 1.5% based on EDA after 8 h [21–23, 51–53].

The cobalt complex [L2*Co]·2H2O (9) showed a very high selectivity as cyclopropanation catalyst and at 75 °C in dichloroethane at a [Co]/EDA/NMI/olefin ratio of 1:20:0:150 we observed a conversion of 99% based on EDA after 110 min. To evaluate the catalytic activity of [L2*Co]·2H2O (9), we used this complex as a catalyst in our model reaction. Conditions of the catalytic reaction were examined by varying the temperature and the relative ratios of [Co]/EDA/NMI/olefin. The results are summarized in Table 5.

Table 5 Cyclopropanation of α-methyl styrene with EDA catalyzed by [L2*Co]·2H2O (9)a
Full size table

The cobalt complex [L2*Co]·2H2O (9) exhibited an excellent capability toward the decomposition of ethyl diazoacetate and the subsequent transfer of the carbene moiety to the C=C double bond with a very low tendency to form by-products.

According to the results shown in Table 5, it can be seen that [L2*Co]·2H2O (9) can act as a catalyst for the cyclopropanation reaction of unactivated olefins with a very high selectivity. The maximum level of diastereoselectivity reached, however, is not very satisfactory (maximum ratio trans:cis = 74:26) if compared to the results reported some years ago by the group of Katsuki, using Co(II) Schiff bases complexes [54, 55]. That is the reason why we did not try to investigate further the optimization of the catalyst loading; our [Co]/EDA ratio of 1:30 is still far from the one reported [Co]/EDA ratio of 1:200 [56–59] by employing chiral Co(II)-porphyrin complexes. The catalyst showed a very low tendency to form the well-known coupling by-products (maleate and fumarate) normally obtained in cyclopropanation reactions (Table 5). The temperature played an important role in these reactions: the catalytic rate of the catalyst increases very much with increasing the temperature (compare: entry 2 with entry 1 and entry 7 with entry 8 in Table 5). The best result in terms of cyclopropane yield (99% based on EDA) was obtained at 75 °C at a complex (9)/EDA/olefin ratio of 1:20:150 (entry 2, Table 5). The presence of NMI improved the cis:trans ratio but with a decrease in the yield depending on the concentration of NMI (compare: entry 1 with entry 4, entry 1 with entry 5, and entry 3 with entry 8 Table 5).

To examine the generality of the complex [L2Co]·2H2O* (9) as cyclopropanation catalyst some other alkenes were employed to determine the substrate scope of the complex [L2Co]·2H2O* (9). At a complex 9/EDA/olefin ratio of 1:20:150 at 75 °C, the complex catalyzed the cyclopropanation of a range of substrates. The results are summarized in Table 6.

Table 6 Cyclopropanation of olefins with EDA catalyzed by [L2*Co]·2H2O (9)
Full size table

When styrene was employed as substrate, we observed a slight decrease in the quantitative conversion of the starting EDA to what observed in the case of α-methyl styrene with a decrease in the conversion rate (compare entry 1 in Table 6 and entry 2 in Table 5). However in this case, the diastereoselectivity is improved (cis:trans = 36:64). Similar yields in cyclopropane products, but with slightly longer reaction times, are obtained when electron-donating substituents are present in the para position of the aromatic ring (entry 2 in Table 6). If 4-chloro-α-methylstyrene is employed as substrate both the catalytic rate and the yield decrease.

Conclusions

We report here a study of the coordination capability of two new potentially tetradentate (ONNO) Schiff bases diamine ligands [H2L1] and [H2L2] toward Co(II), Cu(II), Ni(II), and Fe(III). Conductivity measurements show that both Schiff bases are coordinated to the metal ions, from two phenolic oxygen atoms and two azomethine nitrogen atoms, as doubly negatively charged anions. Therefore, it seems two phenolic-OH have been deprotonated and bonded to the metal ions as oxygen anions. The proposed structures of the complexes are shown in Chart 1 and 2.

Chart 1
figure 1

The proposed structures of metal complexes of the ligand H2L1. Complex 2: [L1Ni]·3H2O: M = Ni, n = 3, and Complex 3:[L1Cu]H2O: M = Cu, n = 1

Full size image
Chart 2
figure 2

The proposed structures of metal complexes of the ligand H2L1. Complex 5, [L2Co(H2O)2]·2H2O: M = Co, X1 = X2 = H2O, n = 2, Complex 6, [L2Cu]·2H2O: M = Cu, X1 = X2 = none, n = 2, Complex 7, L2Fe(H2O)Cl: M = Fe, X1 = H2O, X2 = Cl, n = none, Complex 8, [L2Ni]H2O: M = Ni, X1 = X2 = none, n = 1, and Complex 9, [L2*Co].2H2O: M = Co, X1 = X2 = none, n = 2

Full size image

It is to be noted that the complex [L2*Co]·2H2O (9) acted as a catalyst in cyclopropanation reactions of unactivated olefins by ethyldiazoacetate with very high selectivity (up to 99%), although Co(II) complexes have seldom been used as catalysts for cyclopropanation reactions. Further investigations by changing the ligand skeleton will be made in the future to improve the diastereo and enantio-selectivities reported in this article.