Ni(II)-Aroylhydrazone Complexes as Catalyst Precursors Towards E ﬃ cient Solvent-Free Nitroaldol Condensation Reaction

: The aroylhydrazone Schi ﬀ bases 2-hydroxy-(2-hydroxybenzylidene)benzohydrazide and (2,3-dihydroxybenzylidene)-2-hydroxybenzohydrazide have been used to synthesize the bi-and tri-nuclear Ni(II) complexes [Ni 2 (L 1 ) 2 (MeOH) 4 ] ( 1 ) and [Ni 3 (HL 2 ) 2 (CH 3 OH) 8 ] · (NO 3 ) 2 ( 2 ). Both complexes have been characterized by elemental analysis, spectroscopic techniques [IR spectroscopy and electrospray ionization-mass spectrometry (ESI-MS)], and single-crystal X-ray crystallography. The coordination behavior of the two ligands is di ﬀ erent in the complexes: The ligand exhibits the keto form in 2 , while coordination through enol form was found in 1 . Herein, the catalytic activity of 1 and 2 has been compared with the nitroaldol condensation reaction under various conditions. Complex 2 exhibits the highest activity towards solvent-free conditions. for the β -nitroalkanol formation in the Ni(II) catalyzed nitroaldol reaction. The resultant mixture was stirred for 15 min at 50 ◦ C to obtain a dark green solution. A clear solution was obtained by ﬁltering the mixture and the solvent was then allowed to evaporate slowly. After 1 d, green X-ray quality single crystals were isolated. The isolated compound was then washed 3 times with cold methanol and dried in open air.

The design of the catalyst plays a crucial role in controlling the diastereo-and enantioselectivity of the products, leading to a serious task for researchers to search for efficient and selective catalysts from synthetic, economic, and environmental perspectives. Di and polynuclear metal complexes including coordination polymers can efficiently catalyze the nitroaldol reaction between aldehydes and nitroalkanes [26][27][28]. Solvent-free organic synthesis has received significant interest from the chemist worldwide due to its advantage over chemical wastes in terms of sustainability and the requirements of green chemistry [29]. The use of solvent-free condition is also applied in the nitroaldol reaction [30,31]. Some nickel complexes showed good catalytic activity towards nitroaldol reactions under various conditions such as in homogeneous reaction states [32,33], under solvent-free microwave irradiation [25], in ionic liquids [34,35], or under heterogeneous [36][37][38] conditions. Aroylhydrazone Schiff bases form highly stable complexes with transition metals in various oxidation states, coordination numbers, and nuclearities, and can be tuned easily by changing different carbonyl derivatives [39][40][41][42][43][44][45][46].
In this study, we describe the synthesis and characterization of one dinuclear and one trinuclear Ni(II) complex derived from two different aroylhydrazones, and their activity as catalyst precursors towards nitroaldol (Henry) reactions to achieve desired functionalized products under different homogeneous catalytic reactions conditions. Solvent-free conditions are found most efficient in our catalytic system. This is probably due to more accessibility of substrate molecules to the metal centre under solvent-free condition than the presence of solvent molecule; 94%-97% yields are observed in our case, which is relatively higher than the yield found in other di or polynuclear catalytic system [26].

Syntheses and Characterizations
In this study, we have used two different aroylhydrazone Schiff bases, namely, 2-hydroxy-(2-hydroxybenzylidene)benzohydrazide (H 2 L 1 ) and (2,3-dihydroxybenzylidene)-2hydroxybenzohydrazide (H 3 L 2 ) [47,48], to synthesize two different (one dinuclear and one trinuclear) Ni(II) complexes. In the dinuclear complex [Ni 2 (L 1 ) 2 (MeOH) 4 ] (1), the ligand exhibits the enol form with the loss of all (two) acidic hydrogens whereas H 3 L 2 undergoes coordination with the loss of two acidic hydrogens (out of three) and one remains protonated to form the trinuclear [Ni 3 (HL 2 ) 2 (CH 3 OH) 8 ]·(NO 3 ) 2 (2) (Scheme 1). In both complexes, the phenolate oxygen at the orthoposition (from the aldehyde moiety of the ligand) forms a phenoxido bridge between two Ni(II) ions. Characterizations of 1 and 2 have been carried out by elemental analysis, spectroscopic methods (IR spectroscopy, ESI-MS), and X-ray diffraction (single crystal) techniques. Beside similar characteristic stretching signals of the ligand, a band at 1609 cm −1 appears in the IR spectrum of 2 that corresponds to the C=O stretching frequency [47,48]. The m/z value of 2 suggests the loss of two noncoordinate nitrate ions and one acidic proton from one of the two ligands present in 2 (see Experimental). The catalytic properties of 1 and 2 were investigated towards solvent-free nitroaldol condensation reaction and their activities were compared.

General Description of the Crystal Structures
Crystals of [Ni 2 (L 1 ) 2 (DMF) 4 ] [Ni 2 (L 1 ) 2 (DMF) 2 (H 2 O) 2 ]·2DMF (1A) suitable for X-ray diffractions were obtained upon re-crystallization of 1 from DMF-water and slow evaporation of 2 from the methanolic solution, at ambient temperature. The crystallographic data are summarized in Table 1, representative molecular structures are displayed in Figures 1 and 2, and selected dimensions are presented in Table 2.
The asymmetric unit of the compound 1A comprises a half unit of two different molecules and one solvent DMF. One half unit of one molecule contains the nickel(II) cation with one coordinated ligand, one DMF, and one water molecule. Another half consists of nickel(II) cation with one coordinated ligand and two DMF molecules. All the Ni(II) centers exhibit a distorted octahedral coordination environment. The structure of 1A contains crystallographically generated inversion centres in the middle of the Ni1-Ni1 i or Ni2-Ni2 ii bonds, therefore in the heart of the respective Ni 2 O 2 planes. The aroylhydrazones act as dianionic and tetradentate ONOO chelating equatorial ligands, binding to one of the Ni(II) centers via the enolate oxygen, the imino nitrogen, and the deprotonated phenolate oxygen, which is connected to other metal cation. Thus, the structure displays two µ-O bridges in each molecular unit, which connect the two nickel(II) centers. In the axial positions, there are two DMF oxygens atoms in one unit and DMF and water in the other unit. The N−N bond distances of 1.385 (3)

General Description of the Crystal Structures
Crystals of [Ni2(L 1 )2(DMF)4] [Ni2(L 1 )2(DMF)2(H2O)2]·2DMF (1A) suitable for X-ray diffractions were obtained upon re-crystallization of 1 from DMF-water and slow evaporation of 2 from the methanolic solution, at ambient temperature. The crystallographic data are summarized in Table 1, representative molecular structures are displayed in Figures 1 and 2, and selected dimensions are presented in Table 2.
The asymmetric unit of the compound 1A comprises a half unit of two different molecules and one solvent DMF. One half unit of one molecule contains the nickel(II) cation with one coordinated ligand, one DMF, and one water molecule. Another half consists of nickel(II) cation with one coordinated ligand and two DMF molecules. All the Ni(II) centers exhibit a distorted octahedral coordination environment. The structure of 1A contains crystallographically generated inversion centres in the middle of the Ni1-Ni1 i or Ni2-Ni2 ii bonds, therefore in the heart of the respective Ni2O2 planes. The aroylhydrazones act as dianionic and tetradentate ONOO chelating equatorial ligands, binding to one of the Ni(II) centers via the enolate oxygen, the imino nitrogen, and the deprotonated phenolate oxygen, which is connected to other metal cation. Thus, the structure displays two µ-O bridges in each molecular unit, which connect the two nickel(II) centers. In the axial positions, there are two DMF oxygens atoms in one unit and DMF and water in the other unit. The N−N bond distances of 1.385 (3) and 1.397 (3) Å for the coordinated ligand indicate their single bond hydrazino character. The ODMF-Ni-ODMF or ODMF-Ni-Owater groups are nearly linear (170.47 (9)° and 172.52 (7)°) and in the Ni2(μ-O)2 cores, the Ni-O-Ni angles are ca. 100°. The Ni-Ni contact distances are about 3.10 Å.
Compound 2 crystallizes in the triclinic P¯1 space group. Its unit cell contains one molecule of the compound and two nitrate anions. Compound 2 is trinuclear with H3L 2 coordinating to the Ni(II) cations in the dianionic (HL 2 ) 2− form using both phenolate O atoms, the keto O atom, and the imine N atom. The phenolate oxygen at the ortho position of the aldehyde moiety exhibits µ-O bridges with the other metal cation, which is located at the inversion centre. The asymmetric units of 2 contains half of the molecules, i.e., one and a half nickel cations, one (HL 2 ) 2− ligand, four methanol molecules, and one nitrate anion. All the axial positions are occupied by methanol molecules. The terminal nickel cations exhibit distorted octahedral N1O5 coordination environment but the Ni(II) at center of Compound 2 crystallizes in the triclinic P¯1 space group. Its unit cell contains one molecule of the compound and two nitrate anions. Compound 2 is trinuclear with H 3 L 2 coordinating to the Ni(II) cations in the dianionic (HL 2 ) 2− form using both phenolate O atoms, the keto O atom, and the imine N atom. The phenolate oxygen at the ortho position of the aldehyde moiety exhibits µ-O bridges with the other metal cation, which is located at the inversion centre. The asymmetric units of 2 contains half of the molecules, i.e., one and a half nickel cations, one (HL 2 ) 2− ligand, four methanol molecules, and one nitrate anion. All the axial positions are occupied by methanol molecules. The terminal nickel cations exhibit distorted octahedral N1O5 coordination environment but the Ni(II) at center of inversion displays more regular octahedral geometry. The Ni-Ni contact distances are 3.764 Å, longer than the distances found in 1A.

Catalytic Studies
The catalytic activity of complexes 1 and 2 for the nitroaldol (or Henry) reaction was evaluated using benzaldehydes and nitroethane as models (Scheme 2, Tables 3 and 4). An excess nitroethane was used to obtain maximum conversion of benzaldehydes into products.
The selectivity for the β-nitroethanol products was 100% for all the experiments (the only compounds found besides syn-and anti-β-nitroethanols were non-reacted substrates).

Catalytic Studies
The catalytic activity of complexes 1 and 2 for the nitroaldol (or Henry) reaction was evaluated using benzaldehydes and nitroethane as models (Scheme 2, Tables 3 and 4). An excess nitroethane was used to obtain maximum conversion of benzaldehydes into products. The catalytic activity of complexes 1 and 2 for the nitroaldol (or Henry) reaction was evaluated using benzaldehydes and nitroethane as models (Scheme 2, Tables 3 and 4). An excess nitroethane was used to obtain maximum conversion of benzaldehydes into products.
The selectivity for the β-nitroethanol products was 100% for all the experiments (the only compounds found besides syn-and anti-β-nitroethanols were non-reacted substrates).

Scheme 2. Nitroaldol condensation reaction of nitroethane and benzaldehydes.
Trinuclear complex 2 showed a higher activity than the dinuclear one, at the same temperature and under similar conditions (Table 3). Consequently, the reaction conditions (temperature, reaction time, amount of catalyst, and type of solvent) were optimized using a model nitroethanebenzaldehyde system with catalyst 2 (Table 3). Thus, the optimal experimental conditions found are exhibited in entry 10 of Table 3, leading to a β-nitroethanol yield of 94% (TOF = 3.9 h −1 ) and a good selectivity syn:anti (77:23). Table 3. Catalytic activity of 1 and 2 in the model nitroaldol condensation reaction a of nitroethane and benzaldehyde.
To determine the catalytic performance of 2 in the nitroaldol condensation reaction, a series of aromatic and aliphatic aldehydes were selected to screen as starting materials (Table 4). In general, the reactivity of the substituted benzaldehydes is less than benzaldehyde itself, possibly the presence of steric hindrance. In the case of substituted aromatic aldehydes having electron-withdrawing substituents in para-position, higher yields (entries 5 and 7, Table 4) are obtained, which may be attributed to the higher electrophilicity of the substrate compared to the aldehydes bearing electrondonating moieties. A maximum yield of 97% with a syn: anti diastereoselectivity ratio of 80:20 was observed for the aldehyde, which is para-nitro-substituted (entry 7, Table 4). For the aliphatic aldehydes, acetaldehyde and propionaldehyde (entries 8 and 9, Table 4), a maximum of 78.2% yield was obtained for propionaldehyde (entry 9, Table 4) with a syn: anti diastereoselectivity ratio of 74:26. nitroaldol reaction also did not occur appreciably (9% yield of nitroethanol) by using Ni(OAc)2 instead of the catalyst precursor 1 or 2 (entry 13, Table 3).
To determine the catalytic performance of 2 in the nitroaldol condensation reaction, a series of aromatic and aliphatic aldehydes were selected to screen as starting materials (Table 4). In general, the reactivity of the substituted benzaldehydes is less than benzaldehyde itself, possibly the presence of steric hindrance. In the case of substituted aromatic aldehydes having electron-withdrawing substituents in para-position, higher yields (entries 5 and 7, Table 4) are obtained, which may be attributed to the higher electrophilicity of the substrate compared to the aldehydes bearing electrondonating moieties. A maximum yield of 97% with a syn: anti diastereoselectivity ratio of 80:20 was observed for the aldehyde, which is para-nitro-substituted (entry 7, Table 4). For the aliphatic aldehydes, acetaldehyde and propionaldehyde (entries 8 and 9, Table 4), a maximum of 78.2% yield was obtained for propionaldehyde (entry 9, Table 4) with a syn: anti diastereoselectivity ratio of 74:26. The catalytic activity of compound 2 was also compared in the reactions using various substituted aromatic aldehydes with different nitroalkanes, producing the corresponding β - nitroaldol reaction also did not occur appreciably (9% yield of nitroethanol) by using Ni(OAc)2 instead of the catalyst precursor 1 or 2 (entry 13, Table 3).
To determine the catalytic performance of 2 in the nitroaldol condensation reaction, a series of aromatic and aliphatic aldehydes were selected to screen as starting materials (Table 4). In general, the reactivity of the substituted benzaldehydes is less than benzaldehyde itself, possibly the presence of steric hindrance. In the case of substituted aromatic aldehydes having electron-withdrawing substituents in para-position, higher yields (entries 5 and 7, Table 4) are obtained, which may be attributed to the higher electrophilicity of the substrate compared to the aldehydes bearing electrondonating moieties. A maximum yield of 97% with a syn: anti diastereoselectivity ratio of 80:20 was observed for the aldehyde, which is para-nitro-substituted (entry 7, Table 4). For the aliphatic aldehydes, acetaldehyde and propionaldehyde (entries 8 and 9, Table 4), a maximum of 78.2% yield was obtained for propionaldehyde (entry 9, Table 4) with a syn: anti diastereoselectivity ratio of 74:26. The catalytic activity of compound 2 was also compared in the reactions using various substituted aromatic aldehydes with different nitroalkanes, producing the corresponding β - nitroaldol reaction also did not occur appreciably (9% yield of nitroethanol) by using Ni(OAc)2 instead of the catalyst precursor 1 or 2 (entry 13, Table 3).
To determine the catalytic performance of 2 in the nitroaldol condensation reaction, a series of aromatic and aliphatic aldehydes were selected to screen as starting materials (Table 4). In general, the reactivity of the substituted benzaldehydes is less than benzaldehyde itself, possibly the presence of steric hindrance. In the case of substituted aromatic aldehydes having electron-withdrawing substituents in para-position, higher yields (entries 5 and 7, Table 4) are obtained, which may be attributed to the higher electrophilicity of the substrate compared to the aldehydes bearing electrondonating moieties. A maximum yield of 97% with a syn: anti diastereoselectivity ratio of 80:20 was observed for the aldehyde, which is para-nitro-substituted (entry 7, Table 4). For the aliphatic aldehydes, acetaldehyde and propionaldehyde (entries 8 and 9, Table 4), a maximum of 78.2% yield was obtained for propionaldehyde (entry 9, Table 4) with a syn: anti diastereoselectivity ratio of 74:26. The catalytic activity of compound 2 was also compared in the reactions using various substituted aromatic aldehydes with different nitroalkanes, producing the corresponding β - nitroaldol reaction also did not occur appreciably (9% yield of nitroethanol) by using Ni(OAc)2 instead of the catalyst precursor 1 or 2 (entry 13, Table 3).
To determine the catalytic performance of 2 in the nitroaldol condensation reaction, a series of aromatic and aliphatic aldehydes were selected to screen as starting materials (Table 4). In general, the reactivity of the substituted benzaldehydes is less than benzaldehyde itself, possibly the presence of steric hindrance. In the case of substituted aromatic aldehydes having electron-withdrawing substituents in para-position, higher yields (entries 5 and 7, Table 4) are obtained, which may be attributed to the higher electrophilicity of the substrate compared to the aldehydes bearing electrondonating moieties. A maximum yield of 97% with a syn: anti diastereoselectivity ratio of 80:20 was observed for the aldehyde, which is para-nitro-substituted (entry 7, Table 4). For the aliphatic aldehydes, acetaldehyde and propionaldehyde (entries 8 and 9, Table 4), a maximum of 78.2% yield was obtained for propionaldehyde (entry 9, Table 4) with a syn: anti diastereoselectivity ratio of 74:26. The catalytic activity of compound 2 was also compared in the reactions using various substituted aromatic aldehydes with different nitroalkanes, producing the corresponding βnitroalkanols (Scheme 3), leading to yields ranging from 38% to 94% (Table 5). nitroaldol reaction also did not occur appreciably (9% yield of nitroethanol) by using Ni(OAc)2 instead of the catalyst precursor 1 or 2 (entry 13, Table 3).
To determine the catalytic performance of 2 in the nitroaldol condensation reaction, a series of aromatic and aliphatic aldehydes were selected to screen as starting materials (Table 4). In general, the reactivity of the substituted benzaldehydes is less than benzaldehyde itself, possibly the presence of steric hindrance. In the case of substituted aromatic aldehydes having electron-withdrawing substituents in para-position, higher yields (entries 5 and 7, Table 4) are obtained, which may be attributed to the higher electrophilicity of the substrate compared to the aldehydes bearing electrondonating moieties. A maximum yield of 97% with a syn: anti diastereoselectivity ratio of 80:20 was observed for the aldehyde, which is para-nitro-substituted (entry 7, Table 4). For the aliphatic aldehydes, acetaldehyde and propionaldehyde (entries 8 and 9, Table 4), a maximum of 78.2% yield was obtained for propionaldehyde (entry 9, Table 4) with a syn: anti diastereoselectivity ratio of 74:26. The catalytic activity of compound 2 was also compared in the reactions using various substituted aromatic aldehydes with different nitroalkanes, producing the corresponding βnitroalkanols (Scheme 3), leading to yields ranging from 38% to 94% (Table 5). nitroaldol reaction also did not occur appreciably (9% yield of nitroethanol) by using Ni(OAc)2 instead of the catalyst precursor 1 or 2 (entry 13, Table 3).
To determine the catalytic performance of 2 in the nitroaldol condensation reaction, a series of aromatic and aliphatic aldehydes were selected to screen as starting materials (Table 4). In general, the reactivity of the substituted benzaldehydes is less than benzaldehyde itself, possibly the presence of steric hindrance. In the case of substituted aromatic aldehydes having electron-withdrawing substituents in para-position, higher yields (entries 5 and 7, Table 4) are obtained, which may be attributed to the higher electrophilicity of the substrate compared to the aldehydes bearing electrondonating moieties. A maximum yield of 97% with a syn: anti diastereoselectivity ratio of 80:20 was observed for the aldehyde, which is para-nitro-substituted (entry 7, Table 4). For the aliphatic aldehydes, acetaldehyde and propionaldehyde (entries 8 and 9, Table 4), a maximum of 78.2% yield was obtained for propionaldehyde (entry 9, Table 4) with a syn: anti diastereoselectivity ratio of 74:26. The catalytic activity of compound 2 was also compared in the reactions using various substituted aromatic aldehydes with different nitroalkanes, producing the corresponding βnitroalkanols (Scheme 3), leading to yields ranging from 38% to 94% (Table 5). The selectivity for the β-nitroethanol products was 100% for all the experiments (the only compounds found besides synand antiβ-nitroethanols were non-reacted substrates).
Trinuclear complex 2 showed a higher activity than the dinuclear one, at the same temperature and under similar conditions (Table 3). Consequently, the reaction conditions (temperature, reaction time, amount of catalyst, and type of solvent) were optimized using a model nitroethane-benzaldehyde system with catalyst 2 (Table 3). Thus, the optimal experimental conditions found are exhibited in entry 10 of Table 3, leading to a β-nitroethanol yield of 94% (TOF = 3.9 h −1 ) and a good selectivity syn:anti (77:23).
A blank test was performed with neat benzaldehyde and nitroethane in the absence of any metal catalyst, at 60 • C. No β -nitroalkanol was detected after 24 h of reaction time (entry 12, Table 3). The nitroaldol reaction also did not occur appreciably (9% yield of nitroethanol) by using Ni(OAc) 2 instead of the catalyst precursor 1 or 2 (entry 13, Table 3).
To determine the catalytic performance of 2 in the nitroaldol condensation reaction, a series of aromatic and aliphatic aldehydes were selected to screen as starting materials (Table 4). In general, the reactivity of the substituted benzaldehydes is less than benzaldehyde itself, possibly the presence of steric hindrance. In the case of substituted aromatic aldehydes having electron-withdrawing substituents in para-position, higher yields (entries 5 and 7, Table 4) are obtained, which may be attributed to the higher electrophilicity of the substrate compared to the aldehydes bearing electron-donating moieties. A maximum yield of 97% with a syn: anti diastereoselectivity ratio of 80:20 was observed for the aldehyde, which is para-nitro-substituted (entry 7, Table 4). For the aliphatic aldehydes, acetaldehyde and propionaldehyde (entries 8 and 9, Table 4), a maximum of 78.2% yield was obtained for propionaldehyde (entry 9, Table 4) with a syn: anti diastereoselectivity ratio of 74:26.
The catalytic activity of compound 2 was also compared in the reactions using various substituted aromatic aldehydes with different nitroalkanes, producing the corresponding β-nitroalkanols (Scheme 3), leading to yields ranging from 38% to 94% (Table 5).  In general, the yield of β-nitroalkanol decreased in the order of nitromethane > nitroethane > 1nitropropane for both aldehydes, and the same molecular size dependent behavior was found for the aldehydes (Table 5). A proposed catalytic cycle promoted by Ni(II) centre is presented in Scheme 4 showing the reaction pathway towards the formation of the β-nitroalkanols.   In general, the yield of β-nitroalkanol decreased in the order of nitromethane > nitroethane > 1nitropropane for both aldehydes, and the same molecular size dependent behavior was found for the aldehydes (Table 5). A proposed catalytic cycle promoted by Ni(II) centre is presented in Scheme 4 showing the reaction pathway towards the formation of the β-nitroalkanols.  In general, the yield of β-nitroalkanol decreased in the order of nitromethane > nitroethane > 1nitropropane for both aldehydes, and the same molecular size dependent behavior was found for the aldehydes ( In general, the yield of β-nitroalkanol decreased in the order of nitromethane > nitroethane > 1-nitropropane for both aldehydes, and the same molecular size dependent behavior was found for the In general, the yield of β-nitroalkanol decreased in the order of nitromethane > nitroethane > 1nitropropane for both aldehydes, and the same molecular size dependent behavior was found for the aldehydes (Table 5). A proposed catalytic cycle promoted by Ni(II) centre is presented in Scheme 4 showing the reaction pathway towards the formation of the β-nitroalkanols. Scheme 4. Proposed catalytic cycle for the β-nitroalkanol formation in the Ni(II) catalyzed nitroaldol reaction.

Materials and Methods
Scheme 4. Proposed catalytic cycle for the β-nitroalkanol formation in the Ni(II) catalyzed nitroaldol reaction.

Materials and Methods
The syntheses for this study were performed in air and reagents and solvents (commercially available) that were used as received, without further purification. The metal source for the synthesis of complexes was Ni(NO 3 ) 2 ·6H 2 O. Elemental analyses (C, H, and N) were carried out by the Microanalytical Service of the Instituto Superior Técnico. Infrared spectra (4000-400 cm −1 ) were recorded on a Bruker Vertex 70 instrument in KBr pellets (Bruker Corporation, Ettlingen, Germany); wavenumbers are in cm −1 . The 1 H NMR spectra were recorded on a Bruker Avance II + 400.13 MHz (UltraShieldTM Magnet) spectrometer at room temperature. The internal reference was tetramethylsilane. The chemical shifts are reported in ppm in the 1 H NMR spectra. Mass spectra were recorded in a Varian 500-MS LC Ion Trap Mass Spectrometer (Agilent Technologies, Amstelveen, The Netherlands) equipped with an electrospray (ESI) ion source. For electrospray ionization, the drying gas and flow rate were optimized according to the particular sample with 35 p.s.i. nebulizer pressure. Scanning was carried out from m/z 100 to 1200 in methanol solution. The compounds were observed in the positive and negative mode (capillary voltage = 80-105 V).

X-ray Measurements
Single crystals of complexes 1A and 2 appropriate for X-ray diffraction analysis were immersed in cryo-oil, mounted in Nylon loops, and measured at 296 K. A Bruker AXS PHOTON 100 diffractometer with graphite monochromated Mo-Kα (λ 0.71073) radiation was used to collect the intensity data. Data collections were recorded using omega scans of 0.5 • per frame and full sphere of data were obtained. Cell parameters were retrieved using Bruker SMART [49] software and the data were refined using Bruker SAINT [49] on all the observed reflections. Absorption corrections were done using SADABS [49]. Structures were solved by direct methods by applying SIR97 [50] and refined with SHELXL2014 [51]. Calculations were carried out using WinGX v2014.1 [52]. All non-hydrogen atoms were refined anisotropically. Those H-atoms bonded to carbon were placed in the model at geometrically calculated positions and refined using a riding model. U iso (H) were defined as 1.2U eq of the parent carbon atoms for phenyl and methyne residues and 1.5U eq of the parent carbon atoms for the methyl groups. Least square refinements were employed with anisotropic thermal motion parameters for all the non-hydrogen atoms and isotropic for the remaining atoms.

Catalytic studies
All catalytic tests were run under atmospheric ambiance under with the following conditions for each essay: 1.0-5.0 mol% (0.1-0.5 µmol) of the catalyst precursor 1 or 2 (usually 1 mol%) under solvent-free condition contained 2 mmol of nitroethane and 1 mmol of aldehyde, in that order, or added solvent (2 mL) for reaction in a particular solvent. The reaction mixture was stirred at the particular temperature for the required duration. The solvent was then evaporated, the residue was dissolved in CDCl 3 , and analyzed by 1 H NMR. In the case of solvent-free conditions, the reaction mixture was dissolved in CDCl 3 and analyzed by 1 H NMR. A previously reported method was employed to determine the yield of the β-nitroalkanol product (relatively to the aldehyde) by 1 H NMR [47,48]. To verify the adequacy of the procedure, a number of 1 H NMR analyses were performed in the presence of 1,2-dimethoxyethane as an internal standard, added to the CDCl 3 solution, giving yields similar to those obtained by the above method. Moreover, the internal standard method also confirmed the absence of other side products formed. The ratio between the syn and anti isomers was also determined by 1 H NMR spectroscopy. The values of vicinal coupling constants (for the β-nitroalkanol products), in the 1 H NMR spectra, between the α-N-C-H, and the α-O-C-H protons identify the isomers, being J = 7-9 or 3.2-4 Hz for the syn or anti isomers, respectively [53,54].