2.1 Materials and instrumentation

Chemicals were purchased from Merck and Aldrich Chemicals, and were used without further purification. Elemental analyses were performed using a Thermo Finnigan Flash Elemental Analyzer 1112EA. FT–IR spectra were recorded on a FT–IR 8400-SHIMADZU spectrophotometer as KBr discs in the 400–4000-cm⁻¹ range. Molar conductance measurements were made by means of a Metrohm 712 Conductometer in MeOH. ¹H-NMR spectra were recorded at 25 °C using a Bruker BRX 100 Avance spectrometer. Electronic spectra in MeOH solutions of the complexes were recorded with a Shimadzu model 2550 UV–Vis spectrophotometer. Diffraction data were measured using a Bruker APEX II CCD area-detector diffractometer.

2.2 2-(((5-chloro-2-hydroxyphenyl)imino)methyl)phenol (H₂L)

H₂L have been prepared according to our previous report [18]. A solution of 2-amino-4-chlorophenol (0.3 g, 2 mmol) in 5 mL of ethanol was added to an ethanolic solution (5 mL) of 2-hydroxy-benzaldehyde (0.2 g, 2 mmol) under vigorous stirring. After heating for 15 min, the solution was cooled and an orange precipitate was separated by filtration, washed with cold ethanol and dried in a desiccator over silica gel.

Yield: 0.41 g, 83%. m.p.: 156 °C. Anal. calcd. for C₁₃H₁₀ClNO₂ (247.68 g·mol⁻¹): C, 63.04; H, 4.07; N, 5.66. Found: C, 63.18; H, 4.01; N, 5.73%. FTIR (KBr) cm⁻¹: ν(OH) 3047wb, ν(NH) 2099sb, ν(CN) 1627s, ν(CC_ring) 1496s, ν(CO) 1307 m, ν(CCl) 655 m. ¹H-NMR (100 MHz, DMSO-d₆, 25 °C, ppm): δ = 13.4 (s, 1H; OH_aldehyde), 10 (s, 1H; OH_amin), 9.0 (s, 1H; CHN), 6.7–7.8 (m, 7H, rings). UV/Vis (methanol) λ_max, nm (logɛ, L·mol⁻¹·cm⁻¹): 226 (4.49), 268 (4.32), 350 (4.27), 438 (3.53).

2.3 Methanol (2-((5-chloro-2-hydroxyphenylimino)methyl)phenolato)dioxidomolybdenum(VI)·[MoO₂(L)(MeOH)] (1)

The solid Schiff-base ligand H₂L (0.01 mmol, 0.03 g) was dissolved in 3 mL of methanol. This solution was added dropwise to a methanolic solution (3 mL) of MoO₂(acac)₂ (0.01 mmol, 0.04 g) under constant stirring. The mixture was refluxed for 1 h. The resulting yellow solution was then left undisturbed at room temperature. After two days, yellowish block-shaped XRD-quality single crystals appeared. These were dried in the air at room temperature and stored in a desiccator over silica gel.

Irregular red, yield: 0.031 g, 77.5%. m.p.: > 300 °C. Molar conductance (10⁻³ M, MeOH): 4.4 Ω⁻¹·cm²·mol⁻¹. Anal. calcd. for C₁₄H₁₂ClMoNO₅ (405.64 g·mol⁻¹): C, 41.45; H, 2.98; N, 3.45. Found: C, 41.33; H, 2.91; N, 3.52%. IR (KBr) cm⁻¹: ν(OH) 3371w, ν(CN) 1604s, ν(MoO₂) 817, 925s, ν(CC_ring) 1550 m, ν(CO) 1288 m, ν(CCl) 694s. ¹H-NMR (100 MHz, DMSO-d₆, ppm): δ = 9.3 (s, 1H; CN), 6.8–8.1 (m, 7H; rings), 4.2 (q, 1H; OH), 3.2 (d, H; CH₃). UV/Vis (MeOH) λ_max, nm (logɛ, L·mol⁻¹·cm⁻¹): 218 (4.70), 240 (4.56), 302 (4.52), 354 (4.12), 426 (3.99).

2.4 Imidazole(2-((5-chloro-2-hydroxyphenylimino)methyl)phenolato) nickel(II) [Ni(L)(Im)].MeOH (2)

A methanolic solution (3 mL) of Ni(OAc)₂·4H₂O (0.1 mmol, 0.02 g) was added to a solution of H₂L (0.1 mmol, 0.03 g) in 2 mL of methanol, and the reaction mixture was refluxed for 30 min. Then, 3 mL of a methanolic solution of imidazole (0.2 mmol, 0.02 g) was added to the reaction mixture and the reflux was continued for an additional 2 h. By slow evaporation of the resulting solution, red crystals appeared, which were separated, dried in the air at room temperature and stored in a desiccator over silica gel.

Irregular red, crystals. Yield: 0.027 g, 67%. m.p.: 208 °C. Molar conductance (10⁻³ M, MeOH): 3.9 Ω⁻¹ cm² mol⁻¹. Anal. calcd. for C₁₇H₁₆ClN₃NiO₃ (404.49 g·mol⁻¹): C, 50.48; H, 3.99; N, 10.39. Found: C, 50.57; H, 3.85; N, 10.46%. IR (KBr) cm⁻¹: ν(OH) 3139w, ν(CN) 1604s, ν(NH) 3062s, ν(CC_ring) 1527s, ν(CO) 1280 m, ν(CCl) 655s. ¹H-NMR (100 MHz, DMSO-d₆, ppm): δ = 12.9 (s, 1H, NH), 9.8 (s, 1H; CHN), 6.4–8.1 (m, 10H; rings), 3.8 (q, H; OH), 3.2 (d, 3H; CH₃). UV/Vis (MeOH) λ_max, nm (logɛ, L·mol⁻¹·cm⁻¹): 206 (4.65), 234 (4.36), 294 (3.95), 428 (4.28), 452 (4.21).

2.5 Crystal structure determination

The data were collected at room temperature with a Bruker APEX II CCD area-detector diffractometer using Mo Kα radiation (λ = 0.71073 Å). Data collection, cell refinement, data reduction and absorption correction were performed using multiscan methods with Bruker software [19]. The structures were solved by direct methods using SIR2004 [20]. The non-hydrogen atoms were refined anisotropically by full-matrix least-squares on F² using SHELXL [21]. All hydrogen atoms were placed at calculated positions and constrained to ride on their parent atoms. Details concerning collection and analysis are reported in Table 1.

2.6 Computational

Full geometry optimization of the ground-state (S₀) complexes was carried out using DFT-B3LYP and PBE methods [22,23] with 6–31 + G(d,p) basis set for C, N, O, Cl, H and LANL2DZ for Mo and Ni atoms using the G03 program [24]. Crystal structures were used as starting points for DFT calculations.

To validate the optimization of the structures, frequency calculations were performed, and the results showed no negative (imaginary) frequency. Furthermore, the determination of the origin of structural behaviors has been done with Natural Band Orbital (NBO) analysis at the same level. UV–Vis spectra, electronic transitions, absorbance and oscillator strengths are computed with the time-dependent DFT (TD-DFT) method at the B3LYP/6–31 + G(d,p) level. Solvent (MeOH) was considered as a uniform dielectric constant 32.6 and Polarized Continuum Model (PCM). GAUSSSUM 2.2 with FWHM 0.3 eV have been used to calculate the DOS diagram and to analyze the contribution percentage of groups and atoms to the molecular orbitals [25].

3.1 Characterization

According to our previous work [6], this ligand has two tautomeric forms, enol (benzenoid) or keto (quinoid) forms. FT–IR spectral studies are performed by comparison of the free ligand spectrum with those of its Ni(II) and Mo(VI) complexes. A v(CN) strong band at 1627 cm⁻¹ in the free ligand spectrum shows a red shift to 1604 cm⁻¹ (23 cm⁻¹) after coordination, indicating the coordination of the azomethine nitrogen atom to the metal center. In the IR spectrum of the ligand, the band observed at 3047 cm⁻¹ is due to the NH vibration. Its absence in the spectra of 1 and 2 is another evidence that demonstrates the coordination of the azomethine nitrogen atom to the metal center [26]. Moreover, a shift of the CO bond energy of H₂L (1307 cm⁻¹) to lower frequencies at 1288 and 1280 cm⁻¹ for 1 and 2, respectively, is another reason for the coordination of the ligand to the metal through the phenolic oxygen atom. Dioxidomolybdenum(VI) complex with a cis-MoO₂ moiety exhibits two strong bands at 817 and 925 cm⁻¹ that can be assigned to symmetrical and asymmetrical v(OMoO) vibrations respectively [27]. The OH band vibrations of coordinated and non-coordinated methanol in 1 and 2 appear at about 3100–3400 cm⁻¹.

The ¹H-NMR spectrum of the ligand displayed singlet signals at 13.4 and 10 ppm, attributed to the OH_alhehyde and OH_amine respectively. After complexation, these signals disappeared, indicating that both phenolic oxygen atoms are connected to the metal atom.

The H₂L spectrum shows a singlet signal at 9 ppm, which can be assigned to H_CH=N. This signal appeared at 9.3 and 9.8 ppm for 1 and 2, respectively. Therefore, the high-field shift of these protons proved the coordination of the azomethine nitrogen to the metal center.

In the Mo(VI) and Ni(II) complexes, the signals at 4.1 and 3.2 ppm correspond to the coordinated methanol hydroxyl and the lattice methanol, respectively. Also, protons of the rings revealed at 6.8–8.1 ppm in both complexes.

In compound 2, the NH signal of the imidazole moiety was at 12.9 ppm and the protons of the azomethine and of the rings were disclosed to 9.8 and 6.4–8.1 ppm, respectively. Likewise, the OH and methyl groups of methanol appeared at 3.8 and 3.2 ppm, respectively.

3.2 X-ray structures

ORTEP representations of compounds 1 and 2 are shown in Figs. 1 and 2; selected bond lengths and angles are listed in Table 2.

The molybdenum complex has a distorted octahedral geometry with the ONO deprotonated ligand connected via two phenolic oxygen and iminic nitrogen atoms in the equatorial plane, two oxido groups and a methanol molecule. The molybdenum atom has a 0.316 Å deviation from the equatorial mean plane (N1O1O2O3). Comparison of the MoN [28,29] and MoO [29–32] band distances with the other analogues compounds showed some similarities.

The Mo1O2 distance (1.983(4) Å) is longer than the Mo1O1 distance (1.922(4) Å), which is probably due to the O5H···O2 hydrogen bond. Two adjacent Mo-complexes bind to each other to form a dimmer via an O5H···O2 hydrogen bond (Fig. 3). This dimer made a one-dimensional chain with C9H9···O4 (3.398(7) Å) hydrogen bonds.

These chains join each other through C3H3···C14H14 C interactions to form a two-dimensional lattice. Finally, C7H7···O4 (3.342(6) Å), C5H5···O4 (3430(7) Å), C14H14···C9 (3.70(1) Å) and C14H14B···O3 (3.388(8) Å) hydrogen bonding interactions complete the third dimension in the lattice.

As it can be seen, the Ni ion has a distorted square planar coordination sphere in 2 and a lattice methanol molecule. The double deprotonated tridentate ONO ligand is coordinated to the nickel ion via two phenolic oxygen and iminic nitrogen atoms. The fourth coordination position is occupied by the imidazole molecule. In the center of 2, the nickel atom has a 0.019-Å deviation from the best plane formed by O1N1O2N2; also, the O1Ni1O2 (177.70(7)°) and N1Ni1N2 (175.81(7)°) angles which are close to the ideal value of 180° confirmed the nearly square planar geometry. The Ni1O1 (1.824(2)°) bond length is equal to that in the analogous compound [33], and the Ni1N2 (1.905(2)°) bond length is the smallest one reported in these nickel square planar complexes [33–36]. Two methanol molecules form N3H3···O3 and O3H···O2 intermolecular hydrogen bonds with adjacent complexes to make supra-molecular dimer with an R4416 graph set (Fig. 4). According to the hydrogen bonding classification [37,38], these intramolecular and intermolecular hydrogen bonds are moderate electrostatic interactions.

These dimmers link to each other through C1Cl interaction and C5H5···O1 intermolecular hydrogen bonds along a and b axis respectively, and form the three-dimensional lattice.

3.3 Computational

3.3.2 Electronic spectra

UV–Vis spectra of the title compounds were recorded in a MeOH solution. The calculated and experimental diagrams of absorption spectra for 1 and 2 are shown in Fig. 5. The spectra of methanolic solution of H₂L and the related complexes show four and five absorption bands in the UV–Vis region.

In the electronic spectrum of H₂L, two bands at 226 nm and 268 nm are assigned to π→π* transitions of aromatic rings. The azomethine π→π* and n→π* transitions are viewed at 350 nm and 438 nm, respectively [6]. For both complexes, π→π* and n→π* transitions showed hypsochromic shifts. In the Mo(VI) complex, the band that appeared at 426 nm may be due to a LMCT (O(p) →Mo(d)) transition [39].

According to d⁰ electronic configuration (+6 formal oxidation state) of molybdenum, the Mo-complex showed no d–d transition. In the d⁸ square planar nickel complex, it is expected to show three spin-allowed d–d transition bands corresponding to ¹A_1g→¹A_2g, ¹A_1g→¹B_1g and ¹A_1g→¹E_g, while the tailing of CT bands that appeared at 428 nm and 452 nm toward the visible region cause the disappearance of these bands [40]. In the calculated spectra for 2, these bands are located at 527.28 nm, 498.8 nm and 443.04 nm due to transitions of HOMO–2 and HOMO–5 to LUMO+1. Also, two bands at 384.4 and 342.72 can be related to LMCT and ILCT charge transfer transitions that are centered on 428 nm in the experimental spectra. Other bands that appeared in the theoretical spectrum can be described as an ILCT charge transfer that is mainly due to transitions from lower MOs to LUMO. Three bands at 294, 234 and 206 nm are correlated to this.

For more investigations, the observed and computed electronic transitions based on TD-DFT calculation for Mo(VI) and Ni(II) complexes are listed in Tables 3 and 4, respectively. The B3LYP level is selected due to more accurate results in comparison with those obtained using the PBE method. The calculated values of λ_max are in good agreement with the experimental ones. In the calculated spectra of 1, the transitions between 395.68 and 276.72 nm are assigned to an admixture of ILCT and LMCT. These bands mainly arise from HOMO to LUMO (λ = 395.68 nm), HOMO to LUMO+1 (λ = 349.92 nm), HOMO–1 to LUMO (λ = 324.0 nm), HOMO to LUMO+2 (λ = 315.76 nm), HOMO to LUMO+3 (λ = 297.52 nm), HOMO–1 to LUMO+1 (λ = 292.08 nm) and HOMO–2 to LUMO (λ = 276.72 nm), respectively. The intense band that is located at 240 nm in the experimental spectra can be assigned to more than one excitation. In the calculated spectra, the related excitations are viewed at energies of 3.70 eV (λ = 267.92 nm) and 3.78 eV (λ = 262.16 nm), respectively. The last intense absorption band at 218 nm is attributed to ILCT, whereas appeared at 253.2 nm in simulated theoretical spectra.

Table 3

Main calculated optical transition with composite ion in terms of molecular orbital contribution of the transition, vertical excitation energies and oscillator strength for [MoO₂(L)(MeOH)].

Composition	Weight (%)	E (eV)	Oscillatory strength (f)	λmax (calcd.)	λmax (exp.)	Assignment
H → L	70	2.51	0.0222	395.68	426	LMCT, ILCT
H → L + 1	70	2.83	0.0002	349.92	354	LMCT, ILCT
H-1 → L	58	3.06	0.0196	324.0		LMCT, ILCT
H → L + 2 H-1 → L H → L + 3	53 33 26	3.14	0.0556	315.76	302	LMCT, ILCT
H → L + 3 H-2 → L H-1 → L + 1 H-1 → L	49 29 26 11	3.33	0.1125	297.52		LMCT, ILCT
H-1 → L + 1	64	3.39	0.0310	292.08		LMCT, ILCT
H-2 → L H → L + 2	58 19	3.58	0.2877	276.72		LMCT, ILCT
H-1 → L + 4 H-2 → L H-1 → L + 3	61 21 21	3.70	0.1347	267.92	240	ILCT
H-1 → L + 4	63	3.78	0.0392	262.16		ILCT
H-4 → L H-4 → L + 1 H-4 → L + 2 H-7 → L	53 15 18 11	3.92	0.0035	253.2	218	ILCT

Table 4

Main calculated optical transition with composition in terms of molecular orbital contribution of the transition, vertical excitation energies and oscillator strength for [Ni(L)(Im)]·MeOH.

Composition	Weight (%)	E (eV)	Oscillatory strength (f)	λmax (calcd.)	λmax (exp.)	Assignment
H-2 → L + 1 H-5 →L + 1	52 17	1.88	0.0001	527.28		d–d
H-2 →L + 1	71	1.99	0.0012	498.8		d–d
H-5 →L + 1 H-3 →L + 1 H-6 →L + 1	43 44 13	2.24	0.0038	443.04	452	LMCT, d–d
H-7 →L + 1 H-14 →L + 1	52 23	2.58	0.0024	384.4	428	LMCT, ILCT
H →L + 1	67	2.89	0.2962	342.72		LMCT, ILCT
H-1 →L H →L	67 15	3.02	0.1400	328.64	294	ILCT
H →L H-5 →L + 1 H-3 →L + 1 H-1 →L + 1	61 19 21 18	3.11	0.0022	319.184		ILCT
H-2 →L	70	3.34	0.0003	297.12		ILCT
H-3 →L H-6 →L H-5 →L	66 13 14	3.81	0.0729	260.56	234	ILCT
H-1 →L H-6 →L H-3 →L	44 43 12	3.93	0.0025	222.64	206	ILCT

3.3.3 Analysis of frontier molecular orbitals

Frontier molecular orbitals, especially HOMO and LUMO, have a significant effect on electric properties such as UV–Vis spectra or NLO behaviors and on the prediction of the chemical reactivity of the material [41]. The surface plots of the HOMO and LUMO orbitals of Ni(II) and Mo(VI) complexes are depicted in Fig. 6.

The HOMO–LUMO energy gap correlates with the stability of the compounds. According to Fig. 7, the HOMO–LUMO gap in [MoO₂(L)(MeOH)] is higher than that in [Ni(L)(Im)]·MeOH complex. Hence, the Ni(II) complex is chemically more active than 1. Furthermore, details of the frontier molecular orbitals in complexes 1 and 2 are presented in Table 5.

Table 5

Orbital energies and atomic orbital contributions for the frontier molecular orbitals for Mo(VI) and Ni(II) complexes.

	Ni(II)	Mo(VI)
	E (eV)	Ni (%)	Im (%)	L (%)	Met (%)	E (eV)	Mo (%)	Oxo%	L (%)	Meth (%)
LUMO+4	–0.08	6	50	28	16	–0.92	4	2	93	1
LUMO+3	–0.15	5	13	76	5	–2.11	59	27	13	1
LUMO+2	–0.84	10	2	4	84	–2.3	57	18	25	0
LUMO+1	–1.3	63	6	30	1	–2.35	50	22	27	1
LUMO	–1.88	3	0	97	0	–2.91	18	12	70	0
HOMO	–5.17	4	0	96	0	–6.26	1	3	96	0
HOMO−1	–5.5	18	0	82	0	–6.83	1	3	96	0
HOMO−2	–6.26	100	0	0	0	–7.25	0	0	100	0
HOMO−3	–6.35	24	3	73	0	–8.09	0	25	74	1
HOMO−4	–6.75	28	42	30	0	–8.15	0	63	32	5

The HOMO and LUMO orbitals in both complexes are mainly localized on non-metal atoms. For the Mo(VI) complex, the value of the contribution for non-metals in HOMO/HOMO–2 is equal to 99–100% and for LUMO/LUMO+3, it is equal to 82, 50, 43 and 41%, respectively. In the higher LUMOs of the Mo center, the contributions of the d orbitals have been calculated as follows: LUMO+1: 50%, LUMO+2: 57% and LUMO+3: 59%. The L⁻² has the major contribution to the frontier orbitals. The L⁻² constructs 96% and 70% of HOMO and LUMO, respectively. In [MoO₂(L)(MeOH)], coordinated methanol to the central atom in HOMO–3/–8 played a 26–62% role, whereas it varies between 18–27% in LUMO+1/+3.

The d orbitals of Ni in the [Ni(L)(Im)]·MeOH have a significant contribution (18–27%) to lower HOMO (HOMO–1/–7), also 63 and 14% for LUMO+1 and LUMO+2. The ligand's contributions to the HOMO and LUMO are equal to 96 and 81%. The minimum contribution in HOMO and lower MOs belongs to MeOH (0%), but in orbitals higher than HOMO this value has increased up to 84% for LUMO+2. The DOS diagrams of complexes 1 and 2 are presented in Fig. 8.

These diagrams display the orbital composition characteristics of molecular orbitals with respect to particular fragments or atoms. According to these diagrams, the ligand plays a significant role in the frontier orbitals, except for central atoms that have no importance in the frontier molecular orbitals (especially in HOMO). However, the central metal ion contributes to a wide range of LUMO orbitals (Fig. 8).

Based on these data, the first transition state from HOMO to LUMO in the Ni(II) complex can be assigned to an intra-ligand charge transfer (ILCT), and for the Mo(VI) complex it can be predicted as an admixture of ILCT and LMCT.