A naphthalene-based and hydroxyamide-functionalized hybrid Schiff base and its dimeric and dinuclear Zn(II) complex: synthesis, structural and theoretical characterization, and optical, DPPH scavenging and in silico ADME properties

Abstract

A novel hybrid ligand 2b (3,5-bis((E)-((2-hydroxynaphthalen-1-yl)methylene)amino)-N-(2-hydroxyphenyl)benzamide) was obtained from the amidation reaction between Schiff base 1b (3,5-bis((E)-((2-hydroxynaphthalen-1-yl)methylene)amino)benzoic acid) and 2-hydroxyaniline by using dicyclohexylcarbodiimide. The coordination of 2b with zinc ions resulted in the formation of a neutral dinuclear and non-electrolyte complex of 3b [Zn₂L₂(H₂O)₄]·2H₂O at a stoichiometric 2 : 2 mol ratio, where each zinc(II) cation is six-coordinate in a distorted octahedral geometry with two coordinated water molecules. They were characterized by IR, NMR and UV-vis spectroscopy both experimentally and theoretically. Thermal and mass analyses provided information regarding the presence of water molecules. DFT simulations were performed to obtain the lowest-energy structures of the compounds. The calculated frontier molecular orbital gaps correlated well with the optical bandgaps. 1b–2b emitted strong greenish-blue and green emissions in dilute DMSO and DMF solutions. The emission color of 3b shifted to bluish-green/green. Their emission properties were evaluated in binary aqueous mixtures with varying water content (f_w = 0–90%). Upon increasing the water fraction, the intensity was enhanced by changing the spectral profiles of the media. 1b–3b possess lower DPPH scavenging activity than ascorbic acid. The absorption, distribution, metabolism, and elimination (ADME) parameters were calculated for the ligands.

---

Introduction

Zinc(II) complexes are the most commonly investigated metal–organic frameworks (MOFs) because of the possibility of having a wide variety of coordination arrangements resulting from the d¹⁰ electronic configuration. These complexes exhibit promising functional properties owing to the combination of the advantages of both metal ions and organic ligands. Many zinc complexes emit in the range of blue-green to red emission of the electromagnetic spectrum for application in photo- and electroluminescent materials.^1,2 In the area of photochemistry, the fluorescence emission of luminophores can be hampered upon aggregation in high concentration solution, which is known as the ‘aggregation-caused quenching (ACQ)’ effect, due to the existence of π–π stacking interaction between the independent molecules.^3–5 Aggregation-induced emission (AIE) materials display weak or almost no fluorescence in dilute solutions, but exhibit enhanced fluorescence intensity in the solid state or aggregated state.^6,7 On the other hand, metal complexes can exhibit better fluorescence performance than the corresponding ligands because of their high photo- and thermal stabilities.⁸

Schiff bases with imine/azomethine groups are an important branch of inorganic multidentate ligands. The 2-hydroxy Schiff bases obtained by the condensation of salicylaldehyde/2-hydroxy-1-naphthylaldehyde with aromatic amines are phenol (OH) and keto (NH) tautomers in the crystalline state and solution.⁹ The NH tautomer can also exist in zwitterionic form.¹⁰ The phototautomerization from enol to keto tautomer is connected to an intramolecular proton transfer (IPT) reaction in a six-membered chelate ring between the phenolic oxygen and imine nitrogen atoms, which requires a small amount of energy. Energy can be obtained by light or temperature changes, causing these Schiff bases to exhibit photochromic and thermochromic behaviors.¹¹ Amido-linked Schiff bases consisting of imine (C=N) and amide (C(O)–N) functional groups are among the well-known classes of chelating ligands used in coordination chemistry to obtain new transition metal complexes. They are routinely produced by reacting hydrazides/hydrazones with aldehydes/ketones.¹² A number of these compounds are of interest for use in fluorometric and colorimetric sensing of metal ions^13,14 or inorganic anions.¹⁵ Such complexes have been reported to have potential industrial applications in the disciplines of supramolecular chemistry, corrosion,¹⁶ and catalysis,¹⁷ as well as pharmacological activities, including antimicrobial, antioxidant, and anticancer activities.^18,19

This study focused on combining and improving the properties of Schiff bases and amides in the structure of one molecule, which is one of the most promising scaffolds in the field of biology and a promising class of fluorescent materials. Moreover, the amide (C=O) bond may be a key functional group in the formation of intermolecular hydrogen-bonded aggregates for fundamental photochemistry studies. This paper reports the synthesis and characterization of a new hydroxyamide-functionalized Schiff base (2b) and its zinc complex (3b) for potential medicinal and optoelectronic applications in the future. N,N′-Dicyclohexylcarbodiimide has been utilized in the acylation of the carboxyl group of the diimine-Schiff base (1b) for the condensation of 2-hydroxyaniline. A dimeric Zn(II) complex (3b) was synthesized in a 2 : 2 (L : M) ratio, in which 2b behaves as a deprotonated O₂N₂ tetradentate ligand and binds to metal cations using naphtholic oxygen and imine nitrogen atoms. A dinuclear complex consisting of two Zn(II) ions indicates that each zinc centre is surrounded by NO donor atoms of two ligands and two water molecules, resulting in a distorted octahedral geometry. Such bridging modes with zinc(II) complexes have been published rarely in the literature, and this is suggested to be the first investigation related to amido-Schiff base ligands. Their structures were identified via IR, NMR, mass, and UV-vis spectroscopies, elemental analysis, and density functional theory (DFT) calculations. 3b exhibited good thermal stability in the solid state. The IPT and chromaticity characteristics of the compounds were investigated in DMSO and DMF solutions using UV-vis and fluorescence spectrometry. These synthesized compounds act as ACQ active molecules in highly concentrated solutions, whereas they are AIE-fluorophores in a solvent/H₂O mixture. They have also been evaluated for their antioxidant properties in DMSO using the DPPH assay. Theoretical absorption, distribution, metabolism, and elimination (ADME) parameters were calculated in silico for 1b and 2b.

The molecular framework of the new hybrid Schiff base 2b was designed to incorporate an extended π-conjugated system composed of phenyl and naphthyl rings together with potential hydrogen-bonding sites, which are expected to stabilize a more rigid molecular conformation. This electronic modification may lead to noticeable shifts in emission wavelengths. Based on the structural features, it is anticipated that the IPT process in the excited-state will play an important role in determining whether this amido-Schiff base exhibits ACQ or AIE characteristics in solution and in the aggregated state. Complexation with Zn²⁺ ions is expected to rigidify the molecular framework through chelation involving the imine nitrogen and hydroxyl oxygen atoms. This coordination restricts intramolecular motions and minimizes non-radiative decay pathways, giving rise to chelation-enhanced fluorescence (CHEF). In combination with aggregation-induced restriction, this effect may lead to a synergistic enhancement of fluorescence intensity. The molecular design of the amido-Schiff base also suggests a plausible mechanism for radical scavenging activity. The conjugated π-system formed by the naphthyl and phenyl rings enables efficient delocalization of unpaired electrons, thereby stabilizing radical intermediates, while the phenolic and naphtholic OH groups can donate hydrogen atoms to neutralize reactive radicals. Consequently, this compound represents a suitable candidate for antioxidant evaluation using the DPPH assay. Furthermore, the presence of a conjugated amide group may favorably influence the ADME properties, potentially improving the balance between solubility and lipophilicity as well as metabolic stability, which may enhance bioavailability. The hydrogen-bonding capability of the molecule may also affect protein binding and tissue distribution, contributing to a potentially favorable pharmacokinetic profile. In summary, the rational design of the hybrid Schiff base 2b and zinc complex 3b integrates three key aspects:

(i) Structure–photophysical relationship governing the ACQ-to-AIE transition,

(ii) Metal coordination effects enhancing fluorescence through the CHEF mechanism,

(iii) Structure–activity relationships associated with potential antioxidant and ADME properties.

Experimental

Materials and instrumentation

3,5-Diaminobenzoic acid (Aldrich), 2-hydroxy-1-naphthaldehyde (Aldrich), N,N′-dicyclohexylcarbodiimide (DCC) (Sigma-Aldrich), 2-hydroxyaniline (Aldrich), zinc nitrate hexahydrate (Zn(NO₃)₂·6H₂O) (Acros Organics), L-ascorbic acid (Carlo Erba), and 1,1-diphenyl-2-picrylhydrazyl (DPPH) (Aldrich) were used as received. Absolute ethyl alcohol (EtOH) (Sigma-Aldrich), dichloromethane (DCM) (Riedel-de-Haen), dimethyl sulfoxide (DMSO) (Merck), and N,N-dimethylformamide (DMF) (Riedel-de-Haen) were of spectroscopic purity.

Melting points were determined using a Stuart SMP30 apparatus and uncorrected. Carbon, hydrogen, and nitrogen microanalyses were performed using a Thermo Scientific Flash 2000 elemental analyzer. Infrared absorption (ATR) spectra were recorded at room temperature in the wavenumber range from 4000 to 400 cm⁻¹ using a Thermo Scientific Nicolet iD5 spectrometer as a powder, at a resolution of 4 cm⁻¹ accumulating 20 scans per spectrum, supplied by OMNIC software. Nuclear magnetic resonance (NMR) spectra were established in deutero dimethylsulfoxide on a Bruker Avance Neo spectrometer (500 MHz), operating at 500.13 MHz (¹H) and 125.75 MHz (¹³C), employing Me₄Si (TMS) as an internal standard (abbreviations: s, singlet; d, doublet; t, triplet). Typical spectral parameters were utilized for ¹H-NMR: spectral width 11.90–12.50 kHz, per point digital resolution 0.363–0.381 Hz, acquisition time 2.62–2.75 s, relaxation delay 1 s, pulse width 8.00–8.64 µs, temperature 296.1 ± 0.1 K; for ¹³C-NMR: spectral width 30.12 kHz, per point digital resolution 0.919 Hz, acquisition time 1.08 s, relaxation delay 2 s, pulse width 10 µs, temperature 299.2 ± 0.3 K. The high resolution mass spectrum (HRMS) was recorded on an Agilent LC Q-TOF mass spectrometer at high cone voltage (150.0 V) using electrospray (ESI) positive ionization mode. All the solutions were prepared in methanol and subjected to spectrometry. An acetonitrile-water mixture containing 0.1% formic acid was used as the mobile phase. The conductivity was measured in DMSO solvent (C = 1.0 × 10⁻³ M) on a Bante Instrument 950 conductivity meter equipped with a DJS-0.1 conductivity electrode. TGA-DTA thermograms were collected on a SII7300 ExstarDTA instrument under a nitrogen atmosphere within the range of 25–750 °C and under an O₂ atmosphere between 750 and 800 °C at a heating rate of 10 °C min⁻¹. Ultraviolet-visible (UV-vis) absorption spectra were captured in DMSO and DMF (C = 100 and 50 µM) with quartz cuvettes of 1 cm path length in the 250–900 nm range on an Analytik Jena Specord 200 spectrophotometer (slit 1 nm, lamp change at 300, Δλ 1 nm, speed 50 nm s⁻¹). The purity of the compounds was determined using thin-layer chromatography (TLC, silica gel F₂₅₄).

Synthesis

Compound 1b [3,5-bis((E)-((2-hydroxynaphthalen-1-yl)methylene)amino)benzoic acid]. Diimine-Schiff base was synthesized according to a previously reported procedure.²⁰ 3,5-Diaminobenzoic acid (5 mmol) was dissolved in absolute ethanol (50 mL) with stirring and heating. A red-colored solution of 2-hydroxy-1-naphthaldehyde (10 mmol) in EtOH (40 mL) was slowly added to a light-yellow solution of amine. The color of the solution initially turned yellow and then orange. The resulting dark orange mixture was refluxed for 3 h with stirring. After cooling to room temperature, the resulting solid was filtered and dried in air. It was purified by crystallization in absolute ethanol. Orange solid, yield: 84%, decomposition point: 233 °C. IR (ATR, ν (cm⁻¹)): 3053 (Ar–H), 2870 (C–H), 2763 (O–H, carboxyl), 2502 (O–H, naphthol), 1702 (C=O), 1620 (C=N), 1587/1527 (C=C), 1492, 1469, 1446, 1418, 1405, 1351, 1301 (C–O, carboxyl), 1254, 1221, 1203 (C–O, naphthol), 1135 (C–N). ¹H NMR (500 MHz, DMSO-d₆, δ (ppm)): 7.071–7.089 (d, J = 9.164 Hz, 2H, Ar–H), 7.285 (s, 1H, Ar–H), 7.375–7.404 (t, J = 7.133 Hz, 2H, Ar–H), 7.572–7.605 (t-d, J = 7.073 Hz, 2H, Ar–H), 7.822–7.837 (d, J = 7.495 Hz, 2H, Ar–H), 7.951–7.993 (d, J = 9.174 Hz, 4H, Ar–H), 8.258–8.266 (t, J = 3.874 Hz, 2H, Ar–H), 8.584–8.601 (d, J = 8.462 Hz, 2H, CH=N), 9.576 (s, OH, enol tautomer), 9.829 (s, 1H, COOH), 15.608/15.787 (s, NH, keto tautomer). ¹³C NMR (125 MHz, DMSO-d₆, δ (ppm)): 113.69, 115.56, 118.85, 119.51, 120.48, 124.00, 124.45, 125.87, 126.99, 127.74, 128.41, 128.86, 129.60, 130.50, 132.14, 134.00, 136.10, 136.56, 142.68, 149.86 (Ar–C), 151.02 (Ar = C–N, keto tautomer), 154.33 (CH=N, enol tautomer), 166.96 (COOH), 170.92 (C=O, keto tautomer). Anal. calculated for C₂₉H₂₀N₂O₄ (460.48 g mol⁻¹): C, 75.64; H, 4.38; N, 6.08. Found: C, 74.14; H, 4.71; N, 6.56. MS (ESI⁺): m/z calculated for [M + H]⁺: 461.1501; observed: 461.1489.

Compound 2b [3,5-bis((E)-((2-hydroxynaphthalen-1-yl)methylene)amino)-N-(2-hydroxyphenyl)benzamide]. The amido-Schiff base was synthesized according to the published method.²¹ Diimine 1b (1 mmol) was dissolved in a mixture of DCM and EtOH (120 mL, 1 : 2 v/v) under magnetic stirring. A DCM solution (10 mL) of DCC (1 mmol) was added dropwise to a solution of diimine. After stirring the mixture at 0 °C for 30 min, it was treated with 2-hydroxyaniline (1 mmol). The dark yellow mixture was stirred at 0–5 °C for 3 h and room temperature for 24 h. The reaction mixture was left to complete at room temperature for 3 days. The obtained solid was then filtered and dried at room temperature. The product was recrystallized from dichloromethane. Yellow solid; yield: 50%; decomposition point: 311 °C. IR (ATR, ν (cm⁻¹)): 3321 (N–H), 3067 (Ar–H), 2929 (C–H), 2852, 2659 (O–H, naphthol), 2590 (O–H, phenol), 1667/1660 (C=O, amide I), 1622 (C=N), 1606 (C=C), 1558 (amide II), 1483, 1445, 1327 (C–O, phenol), 1308, 1244, 1234 (C–O, naphthol), 1142 (C–N). ¹H NMR (500 MHz, DMSO-d₆, δ (ppm)): 5.758 (s, 1H, NH), 7.087–7.105 (d, J = 9.116 Hz, 2H, Ar–H), 7.389–7.418 (t, J = 7.320 Hz, 3H, Ar–H), 7.501 (s, 1H, Ar–H), 7.587–7.617 (t, J = 7.422 Hz, 3H, Ar–H), 7.842–7.857 (d, J = 7.829 Hz, 4H, Ar–H), 7.994–8.012 (d, J = 9.106 Hz, 4H, Ar–H), 8.118–8.141 (d, J = 11.342 Hz, 2H, Ar–H), 8.557–8.574 (d, J = 8.419 Hz, 2H, CH=N), 9.776 (s, 1H, OH), 15.533 (s, 2H, NH, keto tautomer). ¹³C NMR (125 MHz, DMSO-d₆, δ (ppm)): 112.73, 115.89, 118.30, 119.25, 120.20, 122.23, 124.10, 124.69, 125.95, 127.09, 127.68, 128.05, 128.76, 129.19, 129.72, 130.68, 132.18, 134.21, 135.71, 140.55, 142.84, 146.91 (Ar–C), 150.14 (C–OH), 151.02 (Ar = C–N, keto tautomer), 155.16 (CH=N, enol tautomer), 167.67 (C=O, amide), 170.85 (C=O, keto tautomer). Anal. calculated for C₃₅H₂₅N₃O₄ (551.59 g mol⁻¹): C, 76.21; H, 4.57; N, 7.62. Found: C, 76.69; H, 4.30; N, 7.73. MS (ESI⁺): m/z calculated for [M + H]⁺: 552.1923; observed: 552.1910.

Compound 3b. Amido-Schiff base 2b (0.67 mmol) and Zn(II) nitrate hexahydrate (1.34 mmol) were separately dissolved in EtOH by heating. Then, a colorless solution of the metal cation was added to the reaction flask bearing a yellow solution of 2b. The mixture was then stirred and refluxed for 3 h. Upon cooling to room temperature, the obtained product was filtered, washed with warm water and EtOH, and dried in air. Yellow solid, yield: 87%, decomposition point: 308 °C. IR (ATR, ν (cm⁻¹)): 3320 (N–H), 3061 (Ar–H), 2928 (C–H), 2852, 2573 (O–H), 1667/1660 (C=O, amide I), 1611 (C=N), 1583/1564 (C=C), 1543 (amide II), 1469, 1442, 1328, 1298 (C–O, naphthol), 1148 (C–N), 641 (Zn–O), 608 (Zn–N). ¹H NMR (500 MHz, DMSO-d₆, δ (ppm)): 5.759 (s, 1H, NH), 7.087–7.105 (d, J = 9.116 Hz, 2H, Ar–H), 7.389–7.418 (t, J = 7.320 Hz, 3H, Ar–H), 7.501 (s, 1H, Ar–H), 7.587–7.617 (t, J = 7.422 Hz, 3H, Ar–H), 7.842–7.857 (d, J = 7.829 Hz, 4H, Ar–H), 7.994–8.012 (d, J = 9.106 Hz, 4H, Ar–H), 8.118–8.141 (d, J = 11.342 Hz, 2H, Ar–H), 8.513 (s, 2H, CH=N), 9.776 (s, 1H, OH), 11.794 (s, 2H, H₂O). Anal. calculated for [Zn₂(C₃₅H₂₃N₃O₄)₂(H₂O)₄]·2H₂O (1338.00 g mol⁻¹): C, 62.84; H, 4.37; N, 6.28; Zn, 9.77. Found: C, 63.03; H, 4.37; N, 6.10; Zn, 10.62. MS (ESI⁺): m/z calculated for [M]⁺: 1334.2594; observed: 1334.4074. Λ = 1.003 µS (C = 1.0 mM, DMSO) (cell constant = 9.84 cm⁻¹, Λ_DMSO = 2.87 µS, Λ_H2O = 2.26 µS).

Specific reaction details for these obtained compounds are given in Scheme 1.

Scheme 1 The synthesis routes of compounds 1b–3b: (i) ethanol, reflux (3 h), yield: 84%. (ii) DCC, ethanol, DCM, 0 °C, 30 min. (iii) 2-Hydroxyaniline, 0–5 °C (3 h), at room temperature (24 h), yield: 50%. (iv) Zn(NO₃)₂·6H₂O, ethanol, reflux (3 h), yield: 87%.

Computational methodology

All molecular calculations were performed using the Gaussian 16 Revision C.02 software. The geometries of all the molecular systems and the property (IR, UV, and NMR) calculations were performed using the Becke three-parameter Lee–Yang–Parr-based (B3LYP) hybrid functional, along with a 6-311++G** (for C, H, N, and O elements) and 3-21G* basis set for Zn to achieve a balance between computational efficiency and accuracy. Harmonic vibrational frequency calculations were performed to confirm that the optimized geometries represented true minima on the potential energy surface (PES). In addition, stability calculations were performed to verify the stability of the wave functions.

Emission measurements

The emission data were obtained at room temperature using a Hitachi F-4500 fluorescence spectrophotometer equipped with a xenon lamp (150 W). An excitation wavelength (λ_exc) of 365 nm was used with slit widths of 5 and 10 nm (exc./em.) at 0.5 s response time. For the determination of ‘concentration quenched emission,’ DMSO and DMF solutions (C = 100 µM and 50 µM) were used and the observations were recorded using fluorimetry. For the investigation of ‘aggregation,’ DMSO or DMF solution (100 µM) was taken in a quartz cell (l = 1 cm) and increasing volumes (0–4.5 mL) of water were added into the tube by keeping the total volume at 5 mL. The volume ratio of the water in fractions was varied from 0% to 90%. The fluorescence spectra of these fractions were collected 1 day later between 370 nm and 900 nm.

DPPH radical scavenging assay

The radical scavenging activity of the molecules was measured at room temperature using the DPPH method via UV-vis spectroscopy. An appropriate volume of the sample (C = 100 µM, DMSO) was mixed with 1 mL of DPPH solution (C = 200 µM, DMSO), and the total volume was added to 4 mL of pure solvent. The mixtures were then incubated in the dark for 30 min. All measurements were performed in triplicate. The activity was calculated using eqn (1):²²where A₀ is the absorbance of the positive control (DPPH solution, DMSO) and A₁ is the absorbance of the test solution at 517 nm. Ascorbic acid was used as a standard.

Results and discussion

Chemistry

The synthesis procedure for these compounds is illustrated in Scheme 1. The formation of diimine-Schiff base 1b was easily achieved through the direct single condensation of 3,5-diaminobenzoic acid and 2-hydroxy-1-naphthaldehyde in a 1 : 2 mol ratio, owing to its simple and cost-effective preparation. Next, the carboxyl group of 1b was activated using DCC to generate an intermediate product (3,5-bis((E)-((2-hydroxynaphthalen-1-yl)methylene)amino)benzoic(Z)-N,N′-dicyclohexylcarbamimidic anhydride). The treatment of the corresponding acyl lactim with 2-hydroxyaniline led to the amidation of the carboxylic acid and the synthesis of the novel target amido-Schiff base (2b). Finally, the zinc complex 3b was produced from a typical one-pot complexation reaction. The amido-Schiff base (2b) has various binding sites and potential N- and O-donor atoms. Therefore, neutral 2b was reacted with Zn(NO₃)₂·6H₂O in ethanol at a 1 : 2 molar ratio.

The prepared Schiff bases 1b–2b exist in enol-imine ↔ keto-amine tautomeric equilibrium between the enol and keto forms in the solid state and solution (Scheme S1), which is due to intramolecular O⋯H⋯N type hydrogen bonding in the six-membered chelated ring. It is expected that intramolecular proton transfer is shifted towards the keto tautomer for these compounds derived from 2-hydroxy-1-naphthaldehyde. As depicted in Schemes S2 and S3, they also exhibit (E)–(Z) isomerization around the imine bond. The Z-conformer of the enol tautomer and the E-conformer of the keto tautomer are less stable species because of having a non-planar structure related to the twist of the naphthyl rings.

These molecules were characterized by FT-IR, ¹H and ¹³C NMR, UV-vis and mass spectroscopies, elemental analysis, thermal analysis, and conductance measurements and supported by density functional theory (DFT) calculations. The C, H, and N analyses of the Schiff bases were in good agreement with the calculated values. The elemental analysis data agreed with the [Zn₂(C₃₅H₂₃N₃O₄)₂(H₂O)₄]·2H₂O formulation of new dimeric complex 3b. The charge balance exhibits that 2b acts as a dianionic ligand to chelate with cations in the complex. For this molecular structure, both centrosymmetrical Zn ions are coordinated by the NO donor set of the two ligands, in which the octahedral coordination environment at each zinc atom is completed by two H₂O molecules. The IR, NMR, and mass spectral data were consistent with the assigned dinuclear structures. Complex formation was established in the solid state and solution from the absorption band positions and resonance signal shifts in the IR and ¹H NMR spectra, indicating that ligand 2b bonded with the zinc cation through CH=N and two naphtholate groups. A dinuclear zinc(II) complex was formed at a 2 : 2 (M : L) stoichiometric ratio, which is compatible with some dimeric complex structures in the literature.²³ New complex 3b is a dimer with a Zn₂N₂O₆ core in which the chelating naphtholate oxygen and imine nitrogen atoms from the dianionic tetradentate ligand in the enol form (Scheme S1) and two water molecules. HRMS (ESI⁺) analysis confirmed the presence of a dimeric dinuclear complex in the gas phase, as evidenced by the characteristic [M]⁺ signal at m/z = 1334.4077. The compounds were stable in air and soluble in DMSO and DMF. The specific conductivity (κ) of 3b was calculated as 9.87 × 10⁻⁶ S cm⁻¹ in DMSO (C = 1.0 × 10⁻³ M) at room temperature. The molar conductivity (Λ_m) was found to be 9.87 S cm² mol⁻¹, which signifies that it is a non-electrolyte.²⁴

DFT analysis and computational results

No single crystals were obtained for any of the synthesized compounds. Therefore, to support and validate the experimental observations, density functional theory (DFT) calculations were performed for compounds 1b–3b. All computations were carried out at the B3LYP/6-31+G* level of theory using the Gaussian 16 suite to obtain the lowest-energy geometries.²⁵ The geometries were optimized in the gas phase and confirmed as true minima using frequency analysis. Additionally, stability calculations were performed to verify the stability of the wave functions.

The optimized ground-state structures are depicted in Fig. 1. The optimized geometries of 1b–3b clearly indicate the presence of intramolecular N⋯H–O hydrogen bonding, which supports the experimental conclusion that these compounds exist primarily in the enol tautomeric form in the solid state. These intramolecular H-bonds play a crucial role in stabilizing the molecular conformation and are consistent with the IR spectral signatures and downfield shifts in the NMR spectra. The enol and keto tautomers of 1b and 2b were examined to evaluate the stable form. It was obtained that the enol structure has lower energy than the keto form. The enol tautomer is found to be energetically favorable in the gaseous phase, which is the major structure to coordinate with the metal ion. Based on the results, the six-coordinate distorted octahedral geometry is energetically stable for 3b, where each zinc ion is linked to two coordinated water molecules. Ligand 2b chelates the metal cation via the imine (C=N) nitrogen and hydroxyl oxygen atoms.

Fig. 1 Optimized geometries of molecules 1b–3b, with relative energies in parentheses (kcal mol⁻¹).

Comparisons of the experimental and theoretical IR spectra of the compounds are given in Fig. S1–S3. The DFT-calculated IR spectra (keto-tautomer) reproduced the key vibrational bands of the experimental spectra. Specifically, the imine ν(C=N) stretching vibrations were observed at ∼1620–1611 cm⁻¹, closely matching the experimental values (1620 cm⁻¹ for 1b, 1622 cm⁻¹ for 2b and 1611 cm⁻¹ for 3b). Similarly, the calculated ν(C–H) and ν(O–H) stretching frequencies were in agreement with the IR results, confirming the tautomeric form and coordination environment.

The computed chemical shifts for ¹H (Fig. S4–S6) and ¹³C NMR (Fig. S7–S9) align well with the experimental spectra, further validating the proposed structures of the keto-tautomer. For 1b–2b, the signal at ∼15.8 and 15.6 ppm is attributed to the naphtholic NH protons of the keto tautomer. A broad peak within the range of 6.0–9.0 ppm was assigned to the enamine (=CH–N) and aromatic protons. The signal at 4.6 ppm corresponded to the NH proton of the amide moieties. For 3b, the imine and aromatic ring protons are observed at ∼6.0–10.0 ppm. In the ¹³C NMR spectra, the peaks at ∼172 ppm and ∼154 ppm correspond to (C=O) and (C=N) carbon atoms of the enol and keto tautomers for 1b–2b. Two different signals were at ∼177 and ∼180 ppm belonging to naphtholic (C–O) carbons in the zinc complex. The theoretical downfield shifts observed for the imine protons support their involvement in the tautomerism and coordination with Zn(II). The calculated data confirmed that 1b and 2b existed in equilibrium between the enol and keto forms, with the keto form stabilized in polar media, as reflected in both the experiment and computation.

TD-DFT calculations provided UV-vis absorption spectra in the gas phase for the keto-tautomer (Fig. S10–S12), which closely matched the experimental λ_max values. The simulated absorption bands at 432 nm (1b) and 427 nm (2b) were in excellent agreement with the experimental results, supporting the assignments for intraligand π → π* and n → π* transitions for the keto tautomer. The spectrum of the zinc complex exhibits an absorption maximum at 396 nm. The observed spectral change confirmed that this band was shifted to a higher energy region upon complexation.

The calculated optical gaps due to the highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) also corroborate the experimentally derived optical band gaps from the Tauc and ASF methods. The DFT-derived gaps were 3.188 eV (1b), 3.202 eV (2b), and 3.054 eV (3b) (Fig. 2 and Fig. S13), which closely aligned with the experimental direct band gap values of ∼4.34–4.46 eV. These DFT-calculated gaps are lower than the corresponding experimentally estimated optical bandgaps, which is expected, as DFT, particularly with hybrid functionals such as B3LYP, is well known to underestimate bandgap energies due to the inherent approximations in the exchange–correlation functionals.

Fig. 2 Frontier MOs of (a) 1b and (b) 2b (keto tautomer).

The HOMO of the ligands is mainly localized on the aromatic π-system and phenolic oxygen, while the LUMO resides on the imine π* orbital and adjacent aromatic rings. Upon complexation in 3b, the HOMO–LUMO distribution shifts to include the Zn coordination sphere, indicating enhanced conjugation and intramolecular charge transfer (ICT) characteristics. Nonetheless, the computational trend in gap energies correctly mirrors the experimental results, validating the electronic structure and photophysical behavior.

Spectral analysis

IR spectra. In the infrared spectrum of 1b (Fig. S1), the broad absorption band centered at 2763 cm⁻¹ is attributed to the COOH frequency. The weak band at 2502 cm⁻¹ is related to the naphtholic ν(O–H) stretching vibration. The ν(C=O) frequency of the carboxyl group was 1702 cm⁻¹. The imine ν(C=N) and ν(C–H) vibrations were observed at 1620 and 2870 cm⁻¹ for 1b, existing mainly in the enol tautomer with N⋯H–O interactions in the solid state. The ν(C–H) and ν(C=C) frequencies of the aromatic ring appeared at 3053 cm⁻¹, 1587, and 1527 cm⁻¹. The band at 1301 cm⁻¹ is attributed to the carboxylic ν(C–O) frequency. The naphtholic ν(C–O) frequency was observed at 1203 cm⁻¹, indicating the existence of an enol tautomer. For 2b (Fig. S2), the ν(O–H) carboxyl band disappeared, as expected. The naphtholic ν(O–H) stretching vibration frequency is observed at 2659 cm⁻¹. The ν(N–H) and ν(C=O) frequencies of the amide A and I bands appear at 3321 and 1667/1660 cm⁻¹,²⁶ indicating the formation of the proposed skeleton structure. The amide II band, due to the combination of ν(C–N) stretching and δ(N–H) bending vibrations, was observed at 1558 cm⁻¹. The ν(O–H) frequency in the weak intramolecular hydrogen bonds (O–H⋯NH) due to the new phenyl ring was observed at 2590 cm⁻¹. The imine ν(C=N) and (C–H) bands were centered at 1622 and 2929 cm⁻¹, confirming that 2b forms the enol tautomer. The aromatic ring ν(C–H) and ν(C=C) bands peaked at 3067 cm⁻¹ and 1606 cm⁻¹. ν(C–O) frequencies were observed at 1327 cm⁻¹ and 1234 cm⁻¹. For 3b (Fig. S3), the amide ν(N–H) frequency appeared at 3320 cm⁻¹. There were no appreciable changes in the bands corresponding to amide I and II vibrations. These bands were observed at 1667/1660 cm⁻¹ and 1543 cm⁻¹, indicating that the amide carbonyl functional group was not bound to metal ions. The ν(O–H) frequency of the phenol group appeared at 2573 cm⁻¹. Upon complex formation, the naphtholic (O–H) band was absent, and the ν(C–O) frequency shifted to 1298 cm⁻¹ compared to that of the free ligand 2b, supporting the deprotonation of the hydroxyl groups. The ν(C=N) frequency was shifted to 1611 cm⁻¹. A significant negative shift (11 cm⁻¹) in this stretching mode indicated that the imine nitrogen atom was coordinated to zinc. The aromatic ν(C–H) and ν(C=C) vibrations peaked at 3061 cm⁻¹ and 1583/1564 cm⁻¹. The new absorption bands at 641 cm⁻¹ and 608 cm⁻¹ were assigned to ν(Zn–O) and ν(Zn–N), respectively.

NMR spectra. ¹H NMR data of the compounds were recorded in DMSO-d₆. For 1b (Fig. S14), the chemical shift of the carboxyl OH proton was observed at δ_H 9.829 ppm as a singlet. The singlet signals at δ_H 15.608 and 15.787 ppm were assignable to the naphtholic NH protons, which confirms the existence of the keto tautomer as a significant contributor in the equilibrium. A low-integral singlet at δ_H 9.576 ppm and a doublet in the range of δ_H 8.584–8.601 ppm were attributed to the signals of the OH and imine protons of the enol structure, respectively. The ³J_(NH,CH) coupling constant of the imine signal was recorded as 8.462 Hz, suggesting a proton transfer from the hydroxyl group to the imine-N atom. Aromatic protons were observed between δ_H 7.071 ppm and 8.266 ppm. According to these spectral results, 1b exists as a mixture of enol and keto forms in the solution. In 2b (Fig. 3a), the carboxyl–OH signal disappeared. The signal of the naphtholic NH protons was observed at δ_H 15.533 ppm, which gave an upfield shift with respect to that of the corresponding signal of the diimine. In addition, the imine protons shifted slightly downfield compared with that of 1b and were observed in the range of δ_H 8.557–8.574 ppm as a doublet. The obtained results confirmed the shift in the intramolecular proton transfer equilibrium from the enol tautomer to the keto form in DMSO. Phenol OH and amide NH protons appeared at δ_H 9.776,²⁷ and 5.758 ppm, respectively. The aromatic protons were signaled within the range of δ_H 7.087–8.141 ppm. In the ¹H NMR spectrum of 3b (Fig. 3b), the naphtholic NH protons of the free ligand 2b were absent. The singlet signal at δ_H 11.794 ppm was assigned to coordinated H₂O molecules. The signal due to the azomethine protons underwent an upfield shift to δ_H 8.513 ppm. This may demonstrate the coordination nature of 2b via imino N-atoms and deprotonated hydroxyl O-atoms upon binding with metal ions. Phenolic OH and amidic NH protons were observed at δ_H 9.776 and 5.759 ppm, respectively, in the complex structure, indicating that these groups did not participate in complexation. 3b exhibited approximately the same peaks for aromatic protons, similar to those of 2b.

Fig. 3 ¹H-NMR spectra of (a) 2b and (b) 3b in DMSO-d₆.

The proton-decoupled ¹³C NMR spectrum of 1b displayed 24 peaks in DMSO-d₆ (Fig. S15). The signal at δ_C 166.96 ppm belonged to the carboxyl carbon. The signal at δ_C 170.92 ppm was attributed to the carbonyl carbon of the keto tautomer. The peaks at δ_C 154.33 and 151.02 ppm may be assigned to the (C=N) and (=C–N) carbons of the enol and keto tautomers, respectively. Other aromatic carbons appeared between δ_C 113.69 ppm and 149.86 ppm. The ¹³C NMR spectrum of 2b showed 27 peaks (Fig. S16). The carboxyl carbon peak was absent with the observation of a new peak at δ_C 167.67 ppm due to the carbonyl carbon atom of the amide band. The other carbonyl carbon assigned to the keto tautomer was seen at δ_C 170.85 ppm. The imine and the enamine carbons were detected at δ_C 155.16 and 151.02 ppm. The peak at δ_C 150.14 ppm may be due to the phenolic carbon of new benzene ring. Aromatic carbons resonated in the range of δ_C 112.73 and 146.91 ppm. The ¹³C NMR spectrum of 3b could not been obtained because of its solubility problem in deuterated-DMSO.

Mass spectra. Q-TOF mass spectral patterns of the newly synthesized compounds confirmed the proposed empirical formula. The detected molecular masses, fragment ions, and retention times (RT) are listed in Table S1. Diimine 1b exhibited abundant molecular ions [(C₂₉H₂₀N₂O₄) + H]⁺ at m/z = 461.1489 with an 8.2 a retention time. The peak at m/z 921.2901 is consistent with the theoretical mass value of the dicarboxylic acid dimer structure of diimine, which is related to the formation of intermolecular H-bonds between carboxylic OH groups. An important fragment was observed at m/z 214.0861, which may be related to the loss of the (C₁₁H₈O), (C₄H₄), and (COOH) components of the diimine structure (Fig. S17). The mass spectrum of 2b showed a molecular ion [(C₃₅H₂₅N₃O₄) + H]⁺ at m/z = 552.1910 as a base peak within a 10.8 min retention time (Fig. 4). The m-phenylenediamine-based Schiff bases can create mononuclear, homo- and hetero-dinuclear complexes in ligand : metal ratios of 1 : 1, 2 : 1, 1 : 2, and 2 : 2, or oligonuclear metal complexes because they behave as bridges that link metal ions.^28,29

Fig. 4 TOF-MS spectrum of 2b.

The correlations of counts vs. acquisition time (min) and mass-to-charge (m/z) for 3b are given in Fig. 5. The presence of dimer species 2 : 2 [M₂L₂] of complex 3b was confirmed through mass spectrometry. The component ion at m/z = 1334.4077 was consistent with symmetrical dinuclear/dimeric complex formation, which shows the stability of the dimeric ion in the gas phase. The ions’ stability in the gas phase depends on the intensity of the m/z signals, which is responsible for relative ion abundance. In the mass spectrum, the most abundant peak was found to be a monomeric [ML]⁺ species. The amount of the fragments due to the dimeric (1334.4077 m/z) and monomeric (667.7077 m/z) structures of 3b is ∼50% and 100% abundance. The coordinated water molecules stabilize dimeric structures due to the formation of strong intra- and intermolecular hydrogen bonds. But aromatic rings are not close to planar in complex 3b, while the naphthalene rings show a nearly planar conformation in the free ligand. Decreased planarity of the conjugated π-system and the presence of steric hindrance of the N-(2-hydroxyphenyl)formamide group may lead to a decrease in the stability of the dimeric structure at L : M ratio 2 : 2. Therefore, the equilibria [M₂L₂] ↔ 2[ML] are shifted to the right to increase the percentage of the monomeric species. The other fragment at m/z 214.2086 was attributable to the following molecule 2-(((3-aminophenyl)amino)methyl)phenol (chemical formula C₁₃H₁₄N₂O).

Fig. 5 TOF-MS spectrum of 3b.

TGA-DTA thermograms

Thermograms of zinc complex 3b are given in Fig. 6. All thermoanalytical results are summarized in Table 1. The TGA curve displayed a first decomposition reaction in the temperature range of 28 °C and 255 °C under a nitrogen atmosphere, where the average mass loss is 19.71%. This loss may belong to elimination of coordinated and uncoordinated H₂O molecules and a (C₁₁H₇O) group. In the 2nd step, the mass loss was 45.70% over 255–464 °C. The 3rd decomposition reaction was observed between temperatures of 464 °C and 764 °C with a slow decrease in the mass (21.37%). The obtained total mass loss (86.78%) was due to degradation of organic ligand molecules, where the residue may be assumed as a mixture of zinc oxide and zinc hydroxide in the remaining part. The DTA curve of 3b showed that the endothermic peaks at 247, 316 and 463 °C overlapped with the decomposition reactions in the 1st and 2nd step in the TGA thermogram. The exothermic peak was observed at 361 °C, which is related to other mass loss in the TGA curve. An exothermic peak was seen at 755 °C under an open air O₂ atmosphere.

Fig. 6 TGA-DTA thermograms of 3b.

Table 1 Thermal decomposition of dimeric Zn(II) complex 3b

Step	ΔT (°C)	DTAmax (°C)	DTGmax (°C)	DTG (µg min^−1)	Mass loss (%)		Assignment	Residue (%)
					Found	Calc.		Found	Calc.
1	28.78–254.91	247.03 (endo)	244.56	1538.203	19.71	19.67	Loss of 6 mol H2O
							1 mol (C11H7O)
2	254.91–463.97	316.17 (endo)	362.69	1078.125	45.70	44.92	Loss of 1 mol (C35H23N3O3)	1 mol ZnO
		361.58 (exo)					1 mol (C4H4O)	1 mol Zn(OH)2
		463.88 (endo)						13.22 13.51
3	463.97–764.24	757.68 (exo)	757.76	658.734	21.37	21.89	Loss of 1 mol (C20H10N3)

---

Optical properties

Electronic absorption of the compounds. Enol–keto tautomerism is affected by either the solvent nature or light. Polar solvents stabilize the keto-amine form in the ground state, which can be explained by the more polar electronic structure of this tautomer.³⁰ The existence of the NH form is manifested by absorption above 400 nm.³¹ The absorption spectra of the synthesized compounds were measured at room temperature in DMSO and DMF at various concentrations. As listed in Table S2, it was noted that the spectra showed nearly no difference between these solutions. In DMSO (C = 100 µM), diimine 1b has a maximum at 320 nm, a shoulder at 391 nm, and a dual absorption band at 444 and 467 nm, while four maxima were observed at 322, 394, 443, and 468 nm for 2b (Fig. S18). In DMF (C = 100 µM), 1b exhibited intense absorption at 320 nm and 390 nm and dual absorption at 442 and 465 nm (Fig. S19). In the presence of 2b, bands appeared at 320 nm, 393 nm, 440 nm, and 465 nm. The highest energy absorption peaks were ascribed to π → π* and n → π* transitions of the (C=N) bond of the azomethine group, respectively, in which the band at 320–322 nm is characteristic of the enol form. The lowest energy dual maxima were assigned to n → π* transitions of the (C=O) bond of the carbonyl group. The results showed that both 1b and 2b may be mixtures of enol and keto tautomers at room temperature. The keto form was favored in polar aprotic DMSO and DMF solutions, which was in accordance with the NMR results. These molecules lie on a two-fold rotation around the imine C=N bond axis,³² which are in the trans(E) and cis(Z) configuration for the enol–keto tautomers of 1b and 2b, respectively (Schemes S2 and S3).

Diamagnetic Zn(II) ions with a d¹⁰ configuration can exhibit various coordination numbers and geometries. The polarizing power, size, and steric requirements of the ligand molecules influence the stereochemistry of the zinc complexes. At a concentration of 100 µM, complex 3b contained absorption bands at 321, 395, 444, and 468 nm in DMSO and 322, 391, 439, and 465 nm in DMF. The absorption bands of ligand 2b were not significantly affected by chelation, but the λ_max values due to the imino bond showed strong intensity bands, confirming the coordination with metal ions. The complex geometry d–d transition bands were not observed in the UV-vis spectrum. Accordingly, an octahedral geometry may be suggested for the coordination sphere around the Zn(II) ion in 3b on the basis of the spectral results in the literature.^33–35 When the concentration was lower than 100 µM, the absorption of 1b–3b complied with Beer's law, indicating a lack of aggregation.

The chemical stability of the compounds was evaluated over time. UV-vis spectra were recorded a few months later (Fig. S20). In the case of 1b, two absorption maxima were observed at 320 and 355 nm in DMSO at 100 and 50 µM, respectively. The observation of a broad tail instead of a dual band above 400 nm may be attributed to the presence of intermolecular H-bonds with polar protic solvent molecules. Absorption bands appeared at 320, 353, 443, and 468 nm, and 320, 354, 442, and 467 nm for 2b and 3b, respectively. The absence of any new peaks confirmed that 1b–3b were stable in the solution over time. Intramolecular interactions may play a decisive role in stabilization. The stability of the molecules was studied using the quantum chemical parameters of the molecular orbital structures.

Optical band gap energy. The optical bandgap of photonic materials is an important parameter for their potential use in optoelectronic applications. The band gap energy (E_g) between the valence and conduction bands was determined using the UV-vis absorption spectrum. The Tauc method describes the optical bandgap as a function of the absorption coefficient (α) and incident photon energy (hν). The absorption coefficient is defined by eqn (2),³⁶ where l is the thickness of the quartz cuvette (10 mm).The E_g (eV) value is the cut-off point at the energy axis (αhν = 0) of the straight line of the Tauc plot in eqn (3).³⁷ Here, A is an energy-independent constant (1 for an ideal situation). The n″ constant depending on the electron transition processes can take 1/2, 2, 3/2, and 3 values for the calculation of the allowed direct and indirect, forbidden-direct, and indirect transitions, respectively.

(αhν) = A(hν − E_g)ⁿ (3)

The absorbance fitting method (ASF) describes the optical band wavelength (λ_g), which is the cut-off point at the λ⁻¹ axis of the straight part of eqn (4).³⁸ λ_g was determined by extrapolating the curve of (Aλ⁻¹)^1/n as a function of (λ⁻¹). Here, B₁ and B₂ are constants and A(λ) is the absorbance at λ. E_g (eV) was calculated using eqn (5):³⁹The aligned direct (E_d) and indirect (E_i) transition band gaps of the newly synthesized compounds obtained by the Tauc and ASF methods are shown graphically in Fig. 7 and Fig. S21–S23. At a 100 µM concentration, the E_d gap energy of 1b–3b was calculated between 4.34 and 4.46 eV in DMSO, while it was found to be in the range of 4.41–4.49 eV in DMF. At a concentration of 50 µM, the E_d energy value was within the range of 4.36–4.46 eV in these solutions. In DMSO, the E_i energy of the compounds was found to be between 4.03–4.18 eV and 4.06–4.25 eV at 100 µM and 50 µM concentrations. The E_i value was determined between 4.23–4.33 eV and 4.09–4.30 eV in DMF.

Fig. 7 Tauc's plots for the direct allowed transition of 1b–3b in (a) DMSO and (b) DMF.

Metal ions can decrease the energy gap of the HOMO level and π → π* molecular orbitals of the ligand, because of increasing conformational rigidity in the structure. Despite this, new zinc complex 3b displayed similar E_d and E_i values as the corresponding ligand 2b. It was noticed that all compounds have optical band gaps of ∼4 eV, which may be due to their semiconductor behavior.⁴⁰ On the other hand, the obtained results of the Tauc and ASF methods were found to be very close to each other. As listed in Table 2, the E_d energies were generally determined to be higher than the E_i.

Table 2 The optical band gap energies of 1b–3b

Media	1b		2b		3b
	Tauc method	ASF method	Tauc method	ASF method	Tauc method	ASF method
	Ed (eV)	Ed (eV)	Ed (eV)	Ed (eV)	Ed (eV)	Ed (eV)
	Ei (eV)	Ei (eV)	Ei (eV)	Ei (eV)	Ei (eV)	Ei (eV)
DMSO (100 µM)	4.43	4.34	4.46	4.39	4.45	4.44
	4.10	4.03	4.16	4.09	4.18	4.07
DMSO (50 µM)	4.41	4.36	4.46	4.43	4.44	4.38
	4.10	4.08	4.25	4.14	4.20	4.06
DMF (100 µM)	4.46	4.44	4.49	4.44	4.48	4.41
	4.32	4.23	4.28	4.28	4.32	4.33
DMF (50 µM)	4.45	4.41	4.45	4.41	4.46	4.38
	4.23	4.09	4.21	4.14	4.30	4.24

---

Emission of the compounds. Solvent polarity affects the fluorescent behavior of fluorophores belonging to the nature of the formation of intramolecular charge (ICT) and photoinduced electron (PET) transfers in such molecules or the occurrence of intermolecular solute–solvent interactions and charge transfer complex.⁴¹ Intramolecular proton transfer reaction in the excited state (ESIPT) can favour and stabilize the keto form for most of the naphthalidene-aniline ligands,⁴² which is only the light-emissive tautomer in nonpolar solvents.⁴³ The excited state of the fluorescent Z-keto tautomer returns back to the ground state by a thermal process or displaying emission or undergoing geometric cis–trans isomerization. Upon chelating with metal ions, the hydroxyl protons of the ligands are deprotonated and the ESIPT phenomenon is hindered. Coplanarity and conformational rigidity of the molecular structure increase as a result of the complexation-induced inhibition of rotation around the C=N functional group. This hampers the non-radiative decay route, such as degradation and vibrational relaxation,⁴⁴ and so fluorescence enhancement can be observed through the combination of chelation-enhanced fluorescence (CHEF), the complexation-induced internal charge transfer (ICT) or PET mechanisms.⁴⁵

The fluorescence behavior of Schiff bases 1b–2b is strongly influenced by π-conjugation, conformational rigidity, and intermolecular interactions. The E/Z photoisomerization arising from free rotation around the imine bond, together with the flexibility of the naphthyl rings, may facilitate non-radiative relaxation pathways, resulting in relatively weak fluorescence that is consistent with ACQ-type behavior in the aggregated state. In contrast, the restriction of intramolecular rotational and vibrational motions (RIM) suppresses non-radiative decay and promotes radiative transitions, leading to enhanced fluorescence characteristics of AIE-type behavior. Metal ion coordination provides an additional strategy to modulate the photophysical properties of the hybrid Schiff base 2b. Zn²⁺ chelation disrupts the intramolecular O–H⋯N hydrogen bonds, thereby suppressing the ESIPT process to favor the enol tautomer of the imine moiety. Coordination of Zn²⁺ via the imine nitrogen and hydroxyl oxygen rigidifies the molecular framework, producing chelation-enhanced fluorescence (CHEF). This restricts intramolecular rotations and inhibits photoinduced electron transfer (PET), which often quenches the fluorescence in free ligands. The diamagnetic d¹⁰ configuration of Zn²⁺ can modify both the electronic structure and molecular dynamics of the fluorophore. In particular, processes such as intramolecular charge transfer (ICT) and metal–ligand charge transfer (MLCT) may become more prominent. These electronic effects can minimize non-radiative decay pathways, thereby suppressing ACQ and activating AIE-related emission pathways.

The fluorescence spectra of the newly synthesized compounds are shown in Fig. 8 and 9, respectively. All compounds were yellow in color in DMSO and DMF solutions. Once excited with a xenon lamp (λ_exc = 365 nm), the compounds emitted green or bluish-green/greenish-blue fluorescence at room temperature. This emission appeared visibly brighter in the DMSO solution. For diimine-Schiff base 1b, the emission maxima were located at 426 and 504 nm, originating from intraligand charge-transfer (ILCT) transitions. This highest-energy emission was attributed to the high π* level of the former. In the case of amido-Schiff base 2b, the emission band at 426 nm underwent a 33 nm red shift with respect to that of 1b (Table S2). The maximum emission was also observed at 504 nm in DMSO (C = 100 µM), which may have been caused by enol–keto tautomerism in the excited state (ESIPT). The emission intensity of 1b was found to be almost 1.06-fold higher than that of 2b, indicating that strong intramolecular interactions may be responsible for the intensive emission of 1b. The lower fluorescence intensity of 2b may be attributed to the poor coplanarity of this molecule, which is caused by the formation of amide bonds and attachment of a new benzene group in the skeleton structure.⁴⁶ This phenyl ring may not conjugate very well with the π-system of the corresponding naphthyl groups and may not show π⋯π interactions with naphthalene moieties in 2b because it is impossible to locate in the same plane. Compared to the corresponding ligand 2b, the resulting complex 3b exhibited almost no shift in pure DMSO (C = 100 µM). The Zn complex had an emission wavelength of 505 nm, which is neither ligand-metal nor metal–ligand charge transfer. The light emission of 3b may arise from π → π* transitions within the ligand or interligand. Complex 3b exhibited relatively weak fluorescence, with an intensity that was 0.5-fold lower than that of 2b. This may be due to the weaker binding of zinc ions to the ligand, which reduces the rigidity of the complex and decreases the rate of the radiative pathway of electronic excitation. In a dilute DMSO solution (C = 50 µM), the fluorescence maxima were centred in the same spectral range of 425–452 and 500–503 nm with a significant enhancement (∼1.23–2.27-fold) in the fluorescence intensity for all the compounds, probably ascribable to excimer formation.

Fig. 8 Fluorescence spectra of 1b–3b in DMSO. Inset: The photographs of the compounds after excitation.

Fig. 9 Fluorescence spectra of 1b–3b in DMF. Inset: The photographs of the compounds after excitation.

In the spectrum of these compounds in DMF (C = 100 µM), the emission maxima exhibited a small blue shift (Δλ_max = 2–5 nm) compared to that of DMSO. 1b emitted at 426 and 499 nm in solution, and the emission bands red-shifted to 457 and 500 nm for 2b. The emission strength of 1b is 1.49-fold higher than that of 2b. For the Zn complex, the emission maxima were observed at 453 and 501 nm in pure DMF, with a 0.89-fold weaker fluorescence intensity than that of the ligand. When the concentration was varied to 50 µM, enhanced intensities (1.69-fold, 1.32-fold, and 1.31-fold for 1b–3b) were obtained. The emission maxima of these compounds ranged from 427 to 453 nm and from 495 to 499 nm in dilute DMF solutions.

According to the fluorescence results, the emission intensity of 1b–3b decreased with increasing concentration, indicating aggregation-caused quenching (ACQ) properties. At higher concentrations, strong intermolecular π–π stacking interactions between these molecules may result in aggregate formation.⁴⁷ In Schiff bases, intramolecular proton transfer from the hydroxyl group to the imine bond occurs in either the ground (IPT) or excited (ESIPT) electronic state. The keto tautomer is more stable than the enol form in the excited state because of stabilization by intermolecular hydrogen bonding with the polar solvent molecules. The presence of a keto tautomer may also stabilize the aggregated state.⁴⁸ After photoexcitation, the most stable (E)-enol tautomer in the ground state absorbed UV light and reached the excited state (S_n) (Scheme S4). Subsequently, cis–trans conversion and excited state tautomerization occur to produce the keto structure.⁴⁹ When the (Z)-keto form of the molecule absorbs light, it emits fluorescence without changing its configuration in the excited singlet state.⁵⁰ The excited state energy may transfer non-radiatively to the ground state to reduce the emissions. The ground-state structures of 1b and 2b were observed to be present as a mixture of (E)-enol and (Z)-keto forms in pure DMSO and DMF. In 1b and 2b, the observed normal and tautomeric emission bands with different fluorescent behaviors may be attributed to the excited state of the Z-keto forms.⁵¹ These results were in agreement with previous work,²¹ which provides important information about a hydroxyamide-functionalized salen-type hybrid Schiff base 3,5-bis((E)-((2-hydroxybenzylidene)amino)-N-(2-hydroxyphenyl)benzamide).

AIE and ACQ behavior of the compounds. AIE is a phenomenon in which fluorophores emit weak fluorescence in dilute solution but exhibit strong emission upon aggregation. Aqueous-organic mixed solvents are frequently used to trigger AIE behavior. In water-rich mixtures:

(i) Molecular solubility decreases: AIE-active molecules are generally well soluble in organic solvents but poorly soluble in water. As the water content increases, the solubility of the molecules decreases, and they tend to aggregate. This aggregation initiates the AIE effect.

(ii) Intramolecular motion becomes restricted: the aqueous environment promotes aggregation, which restricts molecular motion. The fundamental mechanism of AIE is the restriction of intramolecular rotational and vibrational motions. When molecules undergo free rotation in solution, the excited-state energy is dissipated as heat rather than light, resulting in weak fluorescence. In the aggregated state, these motions are restricted, and the energy is released as photons, producing strong fluorescence.

(iii) Fluorescence intensity increases: due to aggregation and restricted intramolecular motion, the fluorescence intensity increases significantly.

(iv) The emission wavelength may change: aggregation can alter the electronic interactions between molecules, which may cause shifts in the emission wavelength.

In fluorescent molecules that initially exhibit ACQ behavior, the transformation from ACQ to AIE can occur through metal-ion complexation. This process modifies the electronic structure of the molecule and restricts its intramolecular motions. As a result, complexation can suppress ACQ behavior and activate the AIE mechanism. In molecules showing ACQ characteristics, aggregation leads to strong π–π stacking interactions between the aromatic units. These interactions promote non-radiative decay of the excited state, resulting in fluorescence quenching. Consequently, the molecule may exhibit fluorescence in its monomeric state, whereas fluorescence is significantly quenched in the aggregated state. The main processes responsible for ACQ include: π–π stacking interactions, excimer formation, energy transfer processes and intramolecular free rotation. When the fluorescent ligand forms a complex with a metal ion, several important changes may occur:

(i) Complexation limits rotational freedom at the coordination sites, making the molecular framework more rigid. This rigidity activates the RIM mechanism, which suppresses non-radiative decay pathways. Consequently, fluorescence intensity increases.

(ii) Metal coordination can introduce steric hindrance and alter the molecular packing arrangement, making it more difficult for molecules to stack through π–π interactions. As a result, the π–π stacking interactions responsible for ACQ are reduced.

(iii) Upon complexation, molecules may adopt different aggregation patterns, such as J-type aggregates or loosely packed aggregates. These aggregation modes are less likely to cause fluorescence quenching and may even enhance emission in AIE-active systems.

(iv) Metal complexation can modify the electronic transitions of the system such as ICT (intramolecular charge transfer) and MLCT (metal–ligand charge transfer). These electronic effects can increase the probability of radiative transitions, thereby improving the fluorescence quantum yield.

Fluorophores can exhibit different emission behaviors in water owing to their high polarity and ability to form H-bonds. Therefore, the AIE and ACQ emission properties of the compounds were investigated in polar protic aqueous solutions. For this, DMSO and DMF solutions of 1b–3b with water fractions (f_w) of 10–90% were prepared, and their emission was measured after overnight incubation to ensure that the formation of a stable aggregate was fully completed. As shown in Fig. 10, 1b exhibited two fluorescence maxima at 428 and 498 nm at 10% water content in DMSO. These emission bands disappeared with the appearance of a new broad emission band in the range of 443–448 nm until 80% fraction, while two maxima were observed again at 420 and 442 nm at 90% fraction. 1b displayed very strong fluorescence in the DMSO–H₂O binary solvent mixture, although it had much less intense fluorescence in pure DMSO. Its emission intensity increased with increasing proportion of water, indicating aggregation-induced emission in an aqueous DMSO solution.⁵² When the volume ratio of water was 30%, the intensity reached the maximum value.

Fig. 10 Fluorescence spectra of 1b–3b in DMSO–H₂O mixtures at different water fractions after overnight incubation. Plots of fluorescence intensity (I/I₀) versus water fraction (f_w) Inset: The photographs of the samples with increasing f_w illuminated under 365 nm.

The fluorescence maxima of 2b exhibited small shifts to 458–449 nm and 502–504 nm in the range of 10–60% fraction, which depends on the polarity of the medium. Upon increasing the solvent polarity (10–50% fraction), a slight increase was obtained in the emission intensity as compared to that in pure DMSO solution. This was followed by a weak emission intensity shoulder (λ_em = 417–418 nm) and maximum (λ_em = 511–509 nm) until 90% content. In complex 3b, the maximum of the fluorescence band was seen at 504 nm at 10% water fraction and then shifted to the red area by 3–12 nm. A new hump was seen at 456–458 nm within the 20–40% fraction range, and blue shifted to 417–418 nm in 50–90% aqueous DMSO solution. The maximum fluorescence intensity was obtained at 40% water fraction for both 2b and 3b, because of maximal aggregate formation.

Similar results were obtained for the aqueous DMF solution. 1b–3b exhibited intensive emission in the binary solvent mixture, which indicates that the polarity of the medium has the same effect on these molecules. The maximum emission intensity was observed at 30%, 50%, and 60% water fraction for 1b–3b, respectively. As depicted in Fig. 11, the fluorescence maxima of 1b were around at 428–436 and 500–492 nm for two water fractions of DMF (10–20%). These bands were absent, and a new broad maximum was observed between 451 and 446 nm in the region of 30–80% fraction, while a new maximum appeared at 422 nm at 90% water fraction. The ligand and Zn(II) complex had two emission maxima (λ_em = 447–456 and 496–510 nm for 2b, and 445–458 and 504–511 nm for 3b) in all water fractions of DMF. As well, they emitted a weak intensity band (λ_em = 416–420 nm) at 80–90% fraction.

Fig. 11 Fluorescence spectra of 1b–3b in a DMF–H₂O mixture at different water fractions after overnight. Plots of fluorescence intensity (I/I₀) versus water fraction (f_w) Inset: The photographs of the samples with increasing f_w illuminated under 365 nm.

The above results confirmed that 1b–3b show AIE behavior in a solvent–H₂O binary mixture, due to the existence of strong intermolecular H-bonding interactions. When f_w increases, the molecules form aggregates by electrostatic interactions. Increasing the solvent polarity may cause the transformation of these compounds from the excited state to the twisted intramolecular charge transfer state.⁵³ Increasing water content increases the restriction of intramolecular rotation in the aggregated state of these compounds, because of decreasing their solubility.⁵⁴ The isolated dimer formation in aqueous solution can lead to an increase in the loss of energy via the radiative path and enhance their fluorescence intensity.⁵⁵ In Schiff bases, the C=N bond isomerization and intramolecular rotation inhibit the ESIPT process. The intramolecular motions, including free rotation and vibration of the C=N group and aromatic ring in the molecular structure, promote non-radiative decay, which dominates leading to low emission. The aggregated state activates the ESIPT because of the restriction of intramolecular rotation, which contributes to strong fluorescence enhancement. In summary, an aqueous–organic medium decreases the solubility of compounds 1b–3b, which promotes aggregation and restricts intramolecular motion. As a result, the AIE fluorescence was significantly enhanced in this medium. For zinc complex 3b, the more rigid molecular structure leads to restricted intramolecular motions, decreased π–π stacking interactions and reduced non-radiative energy dissipation. As a result, the initially ACQ-dominant system was transformed into an AIE-active system, leading to enhanced fluorescence in the aggregated state (Scheme 2). The AIE-fluorescent behavior of 3b was also compared with those Zn-MOFs, and the data are listed in Table 3. This property is particularly important for several potential applications:

Scheme 2 Proposed mechanistic pathway for ESIPT and AIE behaviours of 2b.

Table 3 Comparison of 3b with other Zn-MOFs

MOFs structure	Ligand	Media	Emission-characteristics	Ref.
LLCT: ligand-to-ligand charge transfer; TICT: twisted intramolecular charge transfer.
[Zn(L^1)2Cl2]	1-Methylbenzimidazo-2-yl-methyleneaniline	Solid	Green luminophores, red luminophore	1
[Zn(L^2)2Cl2]	1-Ethyl-2-(4-methoxyphenyl)azobenzimidazole
[Zn(Titpe)(bcpf)·(DMF)]n	1,1,2,2-Tetrakis(4-(1H-imidazol-1-yl)phenyl)ethane	Solid	Blue fluorescence, LLCT	56
	4,4′-Sulfonyldibenzoic acid
[KZn(3-MeOsalpn)(µ1,1-NCS)]2	N,N′-Bis(2-hydroxy-3-methoxybenzylidene)-propane-1,2-diamine	Solid	Weak-fluorescent	57
		DMSO	Fluorescent, CHEF
[Zn(L)]n	3,6-Bis(1,10-phenathroline-[5,6-d]imidazole-2-yl)carbazole	Solid	Orange fluorescence	58
		DMSO	Blue fluorescence CHEF
[Zn5(L)2]·2H2O	N,N-Bis(3-ethoxysalicylidene)hydrazine	Solid	Red fluorescence, AIE	59
		MeOH	Non-emissive
[ZnL2Cl2]	(3-(4-5-(4-Chlorophenyl)diazenyl)-2-hydroxybenzylideneamino)	DMSO	Fluorescent, CHEF	60
	((Phenylimino)methyl)-4H-chromen-4-one
[Zn(L)2]	2-{(E)-[(3,4-Dimethylphenyl)imino]methyl}-5-	DMF	Fluorescent, CHEF	61
	[(E)-Phenyldiazenyl]phenol
[ZnLCl(Ts)]	N-[2-(5,6-Dihydrobenzimidazo[1,2-c]quinazoline-6-yl)phenyl]-4-methylbenzolsulfamide	Solid	Fluorescent	62
	p-Tolylsulfonyl	CH3CN EtOH	Weak-fluorescent
[Zn(L)2][N2O6·2(H2O)]	(E)-4-(((8-(4-Chlorophenyl)-3-isobutyl-3H-imidazo	MeOH	Fluorescent	63
	[4′,5′:3,4]Benzo[1,2-c]isoxazol-5-yl)imino)methyl)phenol
[Zn2(L)2(H2O)2]·2DMF	3-[(5-Chloro-2-hydroxy-benzylidene)-amino]-4-hydroxy-chromen-2-one	DMSO	Fluorescent, CHEF	64
[Zn2(L)2·(4,4′-bpy)2]n	4,4′-Bipyridine		Weak-fluorescent
[(Zn2(o-van-gly)(phen)OAc)2]·3MeOH	Potassium 2-((2-hydroxy-3-methoxybenzylidene)amino)acetate	Solid	Fluorescent, LLCT	65
	1,10-Phenanthroline
[Zn(L)Cl2]	4-Methoxy-N-(pyridin-2-ylmethylene)aniline	Solid	Fluorescent, LLCT
[Zn(L)Cl2]·CH3CN	4-Methoxy-N-(quinolin-8-ylmethylene)aniline	CH3CN	TICT	66
[ZnDyL(NO3)2(OAc)(CH3OH)](CH3OH)3	Bis(5-bromine-3-methoxysalicylidene)-3-oxapentane-1,5-diamine	Solid	Fluorescent, CHEF	67
[Zn2L2(μ1,1-N3)(N3)]	(((E)-2-Ethoxy-6-((quinolin-8-ylimino)methyl)phenol))	DMF	Weak-fluorescent	68
[ZnL(NCO)(H2O)]
[ZnL(NCS)(H2O)]
[ZnL2]	Potassium(E)-3-hydroxy-2((2-hydroxy-4(tetradecycloxy)benzylidene)amino)butanoate	DMSO	Fluorescent, CHEF	69
[ZnL2]	(E)-1-(4-Fluorophenyl)-N-(1,10-phenanthrolin-5-yl)methanimine	EtOH	Fluorescent	70
[Zn2L2(H2O)2]·2H2O	3,5-Bis((E)-((2-hydroxynaphthalen-1-yl)methylene)amino)-N-(2-hydroxyphenyl)benzamide	DMSO	Bluish-green/green fluorescence, AIE	This work
		DMF

---

(i) Fluorescent sensing of metal ions: the prepared Schiff bases 1b–2b can act as a selective metal ion receptor due to the strong enhancement of emission upon chelation.

(ii) Bioimaging: bright AIE in aqueous or aggregated environments can allow cell or tissue imaging with reduced background fluorescence compared to ACQ-prone fluorophores.

(iii) Optoelectronic materials: the RIR- and ESIPT-mediated emission can be leveraged in solid-state lighting or AIE-active OLEDs.

CIE 1931 chromaticity diagrams. Chromaticity diagrams were obtained using the color calculator software.⁷¹ As presented in Table 4, the chromaticity coordinates (x/y) of 1b were 0.1765/0.3757 and 0.1661/0.2632, falling between the bluish-green and greenish-blue regions of the diagram at 100 µM and 50 µM concentrations in DMSO (Fig. 12a). The coordinates (x/y = 0.1699/0.3445 and 0.1653/0.2614) lie in the bluish-green and greenish-blue regions in DMF (Fig. 12b). In these solutions, 1b has a moderate correlated color temperature (CCT) value (10 000–20 000 K) at a 100 µM concentration, but it displays very high CCT > 40 000 K at a 50 µM concentration, classified as a ‘cool’ light source.⁷² The coordinates (x/y = 0.1879/0.5831 and 0.1769/0.4889) of 2b are close to the green region of the diagram in DMSO. These coordinates are observed in the green and greenish-blue regions in DMF. 2b has a moderate CCT < 10 000 K at 100 µM concentration of DMSO and DMF, which may be described as a ‘warm’ light source.⁷³ The correlated color temperature value increased with a decrease in concentration, and was calculated to be in the range of ∼10 000 K to 41 000 K. 3b has the coordinates (x/y = 0.1867/0.6042 and 0.1692/0.3568) in DMSO, which fall in the green and bluish-green regions of the diagram, while it appears in the green region in DMF. The CCT of 3b ranging from 8300 K to 15 700 K, significantly increased with a reduction in the concentration in DMSO. A CCT ∼ 10 000 K may be referred to as a ‘cool’ light source.

Table 4 The chromaticity diagram parameters of compounds 1b–3b

Media	1b		2b		3b
	Color coordinates		Color coordinates		Color coordinates
	x	CCT (K)	X	CCT (K)	x	CCT (K)
	y		Y		y
DMSO (100 µM)	0.1765	13 789.67	0.1879	8501.71	0.1867	8352.8
	0.3757		0.5831		0.6042
	Bluish-green		Green		Green
DMSO (50 µM)	0.1661	41 067.32	0.1769	10 028.77	0.1692	15 689.55
	0.2632		0.4889		0.3568
	Greenish-blue		Green		Bluish-green
DMF (100 µM)	0.1699	16 751.15	0.1918	9878.91	0.1822	10 056.01
	0.3445		0.4675		0.4773
	Bluish-green		Green		Green
DMF (50 µM)	0.1653	42 861.63	0.1711	41 029.34	0.1778	10 603.59
	0.2614		0.2609		0.4598
	Greenish-blue		Greenish-blue		Green

---

Fig. 12 CIE-1931 chromaticity diagrams of 1b–3b in (a) DMSO and (b) DMF.

The color purity of 1b–3b was obtained using eqn (6):⁷⁴

Here (x_i/y_i) are the coordinates of the illuminant, (x_s, y_s) are the coordinates of the sample, and (x_d/y_d) are the coordinates of the dominant emission wavelength.

Table S3 shows that diimine 1b had the lowest color purity values (∼33–40%) at a 100 µM concentration in DMSO and DMF. The color purity reached to ∼38–45% with a decrease of the concentration. Amido-Schiff base 2b has a moderate color purity at a 100 µM concentration in DMSO and DMF, which ranges from ∼65–66% to ∼49–51%. The purity values of 2b decreased to 58–49% at a 50 µM concentration. The color purity of complex 3b was found to be ∼68–67% at a 100 µM concentration in DMSO, while it decreased sharply to 38% at a 50 µM concentration. The purity values were calculated to be ∼51–52% and ∼53–54% in DMF. Consequently, it was concluded that the color purity values increased in the order 1b < 2b < 3b in both DMSO and DMF.

DPPH radical scavenging activity. DPPH is a stable free nitrogen radical that displays a purple color in solution. When the DPPH radical accepts a hydrogen atom from the scavenger molecule, its color turns yellow, with a reduction in absorbance at 517 nm.⁷⁵

Free radical scavenging was studied in DMSO at a set concentration of DPPH (50 µM). Absorbance was measured at 517 nm after incubation for 30 min in the dark. Ascorbic acid was used as the standard for comparison. As seen in Fig. S24, 1b–3b showed dose-dependent scavenging activity in solution. A lower concentration of the compounds in the free radical solution displayed low antioxidant potential (1.39–3.94%), but DPPH radicals were scavenged in the range of percentage 9.38–28.91% upon increasing the concentration from 12.5 to 75 µM. According to the results, the inhibition decreased in the order 2b > 1b > 3b. 1b and 2b exhibited relatively higher potency than 3b, which may be due to the presence of active hydroxyl groups for DPPH scavenging. 2b was found to be more active than diimine 1b, whose formula lacks an extra hydroxyl group in the phenyl ring. Moreover, all compounds showed weak activity compared to ascorbic acid.

In silico ADME results. Some physicochemical parameters, such as molecular weight (MW), number of heavy atoms, ratio of sp³ hybridized carbon atoms (F. C_sp³), number of rotatable bonds, hydrogen bond acceptors (HBA) and donors (HBD), molar refractivity (MR), topological polar surface area (TPSA), lipophilicity and water solubility, and absorption, distribution, metabolism, and excretion (ADME) properties of molecules can be obtained using the in silico SwissADME web browser at https://www.swissadme.ch.^76,77 The partition coefficient (log P_o/w) was used as a descriptor for lipophilicity (o: n-octanol, w: water). Solubility is one of the main factors affecting absorption. The log S scale was as follows: insoluble < −10 < poorly < −6 < moderately < −4 < soluble < −2 very < 0 < highly. The in silico SwissADME method provides free access to evaluate the drug-likeness potency of molecules because of the relationship between physicochemical and pharmacokinetic parameters according to the directions of the Lipinski, Ghose, Veber, Egan, and Muegge filters. Lipinski's rule of five: (i) MW ≤ 500, (ii) M log P ≤ 5, (iii) HBA ≤ 10, (iv) HBD ≤ 5. This rule is called the Rule of 5′, because the border values are 500, 5, 2 × 5, and 5.⁷⁸ Ghose filter: (i) 160 ≤ MW ≤ 480; (ii) −0.4 ≤ W log P ≤ 5.6; (iii) 40 ≤ MR ≤ 130; (iv) 20 ≤ atoms ≤ 70. Veber filter: (i) rotatable bonds ≤ 10, (ii) TPSA ≤ 140. Egan filter: (i) W log P ≤ 5.88, (ii) TPSA ≤ 131.6. Muegge filter: (i) MW ≤ 500, (ii) M log P ≤ 4.15, (iii) HBA ≤ 10, (iv) HBD ≤ 5. Ghose filter: (i) 200 ≤ MW ≤ 600; (ii) −2 ≤ X log P ≤ 5; (iii) TPSA ≤ 150; (iv) Num. rings ≤ 7; (v) Num. carbon > 4; (vi) Num. heteroatoms > 1 and (vii) Num. Rotatable bonds ≤ 15; (viii) HBA ≤ 10; (ix) HBD ≤ 5.

The bioavailability radar contains physicochemical properties, such as flexibility (rotatable bonds), insaturation (C_sp³), insolubility (log S), polarity (TPSA), size (MW), and lipophilicity (XLOGP3). The pink area of the radar includes the optimal range for each property (−0.7 < XLOGP3 < +5.0; 150 < MW < 500; 20 Ų < TPSA < 130 Ų; 0 < log S < 6; 0.25 < C_sp³ < 1; 0 < rotatable bonds < 9).⁷⁹ The BOILED-Egg graphical model deals with human gastrointestinal absorption (HIA) and blood–brain barrier (BBB) permeation, which are according to the white and yellow (yolk) regions, respectively. This method uses XLOGP3 and TPSA as the lipophilicity and polarity descriptors, respectively. The white region showed a high probability of passive absorption by the gastrointestinal tract, whereas the yolk region exhibited a high probability of brain penetration. The blue and red points are related to the P-glycoprotein substrate (PGP⁺) and non-substrate (PGP⁻), respectively.

As shown in the ADME profiles in Tables S4 and S5, compounds 1b–2b have a number of HBA groups 6 and HBD groups 3–4. The MR value was found to be within the range of 139 and 169, while the TPSA was calculated to be less than 115 Ų. Lipophilicity was found to be in the following order: 2b > 1b, owing to the consensus log P_o/w. All compounds are poorly soluble in water and have low HIA properties, but cannot pass through the blood–brain barrier. Organic compounds 1b and 2b were found to be non-substrates of P-glycoprotein (PGP⁻) and non-inhibitors of cytochrome P450 (CYP) liver isoenzymes. These compounds agreed with Veber's and Lipinski's rules, except for molecular weight (>500 for 2b). Violation is related to the physicochemical properties of the molecules, which are outside the range of filters. It has been suggested that molecules violating more than one of these rules may have problems with bioavailability. If the molecule provides at least three out of the five rules, its GI absorption is likely to improve. For 1b–2b, the bioavailability score was found to be 0.55–0.56, indicating the probability (%10) for use as an oral drug in rats with respect to drug-likeness. PAINS and Brenk filters are chemically reactive, metabolically unstable, and toxic fragments of the molecule. The obtained synthetic accessibility scores (3.31–3.69) implied a relatively difficult synthesis of compounds 1b–2b.

The Bioavailability Radar is shown in Fig. S25a–c. As shown in this figure, some descriptors fell outside the pink area of the radar. These compounds are not suggested to be orally bioavailable because of their lipophilic, insoluble, and unsaturated properties. The molecular size of 2b was higher than 500 g mol⁻¹. The BOILED-Egg graphs show that they fall outside this range, indicating no GI absorption and BBB penetration ability. Organic compounds 1b and 2b are non-substrates for P-glycoprotein (PGP⁻).

The computational ADME analysis suggests that these compounds exhibit limited water solubility and reduced gastrointestinal (GI) absorption, which may impede their oral bioavailability. Several deviations from commonly used drug-likeness criteria, including the Ghose, Egan, and Muegge rules, were also observed, indicating that the molecule does not fully conform to standard physicochemical parameters typically associated with orally active drugs. These limitations are likely a consequence of the large molecular size and extended π-conjugated system, which increase lipophilicity and hinder solubility in aqueous media. Nonetheless, such deviations are not uncommon for structurally complex or highly conjugated molecules, many of which retain significant biological activity despite low drug-likeness scores. Optimization strategies, such as structural modification or advanced formulation techniques, may be necessary to improve solubility, absorption, and overall pharmacokinetic performance.

Conclusions

In this work, an amido-Schiff base, 3,5-bis((E)-((2-hydroxynaphthalen-1-yl)methylene)amino)-N-(2-hydroxyphenyl)benzamide (2b), was produced by the amidation reaction between Schiff base 1b and 2-hydroxyaniline. Dimeric and homo-dinuclear complex 3b has a 2 : 2 molar ratio of the ligand Zn²⁺. Their structures were established using infrared (IR) spectroscopy, nuclear magnetic resonance (NMR) spectroscopy, mass and UV-vis spectroscopies, and elemental analyses. According to the microanalysis and mass data, dinuclear zinc complex 3b possesses a [Zn₂L₂(H₂O)₄]·2H₂O composition with a distorted octahedral geometry. The IR and NMR spectral data showed the coordination of naphtholic oxygen and imino nitrogen atoms with the central metal ion. TGA and molar conductance measurements revealed the thermal stability and nonelectrolytic nature of 3b.

The experimental results indicated that organic compounds 1b–2b were present in the enol form in the solid state, whereas their structures were more stable in the keto form in solution. DFT analysis strongly supports the experimental findings. The calculated IR, NMR, and UV-vis data confirmed the structural features and photophysical properties of the molecules. The presence of N⋯H–O and N–H⋯O hydrogen bonds in the enol and keto forms, validated by geometry optimization, underpins the observed spectral characteristics and absorption properties. Furthermore, the correlation between the calculated and experimental band gaps solidifies our understanding of the electronic behavior of 1b–3b and provides insights into the tuning of the optical properties through metal complexation.

Upon irradiation at 365 nm, 1b–3b was fluorescent in both DMSO and DMF solutions at room temperature. These compounds are green or bluish-green/greenish-blue fluorophores, respectively. Their emissive states are attributed to intraligand charge transfer transitions. The position and intensity of the emission bands were evaluated depending on their structures and concentrations, which influence the release of absorbed energy through nonradiative decay. The fluorescence emission intensity of 1b was higher than that of 2b in the pure solvents. The dual emission of the free molecules exhibited typical ESIPT characteristics in the solution. The emission brightness of ligand 2b decreased upon coordination with the zinc ions. The ACQ effect was observed at high concentrations, whereas the intensity increased in a mixture of solvent–H₂O with varying water fractions (f_w = 0–90%) because of their AIE properties.

The physicochemical and pharmacokinetic ADME (absorption, distribution, metabolism, and excretion), drug-likeness, medicinal chemistry, and related parameters of the synthesized compounds were calculated using the SwissADME login-free website. 1b–2b had high lipophilicity but poor solubility in water. They have low human intestinal absorption and are unable to cross the blood–brain barrier (BBB). Ligands cannot act as P-glycoprotein substrates or inhibit cytochrome P450 enzymes. They provide Veber's rules (rotatable bonds and TPSA), but violate some rules (Lipinski, Ghose, Egan and Muegge). Their bioavailability score (0.55–0.56) possesses good drug-likeness properties.

Author contributions

Conceptualization, Ö. G. and S. K.; methodology, Ö. G.; investigation, Ö. G.; visualization, Ö. G.; validation, Ö. G.; software, G. J. and S. K.; writing – original draft, Ö. G., G. J. and S. K.; writing – review and editing, Ö. G. All authors have read and agreed to the published version of the manuscript.

Conflicts of interest

The authors declare no conflicts of interest.

Data availability

Additional documents are included in the supplementary information (SI). This supplementary information file contains theoretically calculated IR, NMR, and UV-GB spectra of compounds in the gas phase, as well as spectra recorded in solution, Tauc and ASF graphs, antioxidant activity and bioavailability radar plots, and some tables and schemes related to enol–keto tautomerism. See DOI: https://doi.org/10.1039/d6nj00512h.

Acknowledgements

Özlem Güngör is grateful to Prof. Metin Gürü, Prof. Fatih Akkurt, Prof. Atilla Murathan and Assist. Prof. Levent Nuralın for allowing her to use a fluorescence spectrophotometer. This paper is dedicated to the 103rd anniversary of the Republic of Turkey.

References

1. S. E. Korolenko, K. P. Zhuravlev, V. I. Tsaryuk, A. S. Kubasov, V. V. Avdeeva, E. A. Malinina, A. S. Burlov, L. N. Divaeva, K. Y. Zhizhin and N. T. Kuznetsov, J. Lumin., 2021, 237, 118156,  DOI:10.1016/j.jlumin.2021.118156.
2. A. S. Burlov, V. G. Vlasenko, Yu. V. Koshchienko, N. I. Makarova, A. A. Zubenko, Yu. D. Drobin, L. N. Fetisov, A. A. Kolodina, Ya. V. Zubavichus, A. L. Trigub, S. I. Levchenkov and D. A. Garnovskii, Polyhedron, 2018, 154, 65–76,  DOI:10.1016/j.poly.2018.07.034.
3. M. Arshad, J. R. At, V. Joseph and A. Joseph, J. Lumin., 2023, 258, 119818,  DOI:10.1016/j.jlumin.2023.119818.
4. A. B. Velappan, A. Husain, N. Rajendran, B. Ghazal and S. Makhseed, J. Mol. Liq., 2022, 359, 119273,  DOI:10.1016/j.molliq.2022.119273.
5. K. Bamnavat, V. Bhardwaj, T. Anand, S. K. A. Kumar and S. K. Sahoo, Dyes Pigm., 2021, 184, 108844,  DOI:10.1016/j.dyepig.2020.108844.
6. Z. Wu, J. Xu, Z. Wu, R. Zhao and L. Hou, J. Photochem. Photobiol., A, 2024, 453, 115668,  DOI:10.1016/j.jphotochem.2024.115668.
7. V. Bhardwaj, K. Bhardwaj and S. K. Sahoo, J. Fluoresc., 2023, 33, 1157–1164,  DOI:10.1007/s10895-022-03138-3.
8. B. Kupcewicz, A. Kaczmarek-Kedziera, K. Lux, P. Mayer and E. Budzisz, Polyhedron, 2013, 55, 259–269,  DOI:10.1016/j.poly.2013.03.035.
9. T. Dziembowska, M. Szafran, A. Katrusiak and Z. Rozwadowski, J. Mol. Struct., 2009, 929, 32–42,  DOI:10.1016/j.molstruc.2009.04.001.
10. A. K. Satapathya, S. K. Behera, A. Yadav, L. N. Mahour, C. V. Yelamaggad, K. L. Sandhya and B. Sahood, J. Lumin., 2019, 210, 371–375,  DOI:10.1016/j.jlumin.2019.02.056.
11. A. Mandal, D. Fitzmaurice, E. Waghorne, A. Koll, A. Filarowski, S. Quinn and S. Mukherjee, Spectrochim. Acta, Part A, 2004, 60, 805–813,  DOI:10.1016/S1386-1425(03)00305-6.
12. P. Acosta-Guzmán, A. Ojeda-Porras and D. Gamba-Sáncheza, Adv. Synth. Catal., 2023, 365, 4359–4391,  DOI:10.1002/adsc.202301018.
13. A. K. Manna, M. Sahu, K. Rout, U. K. Das and G. K. Patra, Microchem. J., 2020, 157, 104860,  DOI:10.1016/j.microc.2020.104860.
14. A. K. Manna, J. Mondal, K. Rout and G. K. Patra, J. Photochem. Photobiol., A, 2018, 367, 74–82,  DOI:10.1016/j.jphotochem.2018.08.018.
15. S. Ghosh, Md. A. Alam, A. Ganguly and N. Guchhait, Spectrochim. Acta, Part A, 2015, 149, 869–874,  DOI:10.1016/j.saa.2015.04.025.
16. M. Mishra, K. Tiwari, A. K. Singh and V. P. Singh, Inorg. Chim. Acta, 2015, 425, 36–45,  DOI:10.1016/j.ica.2014.10.026.
17. M. Nemakal, Giddaerappa, Shantharaja, V. A. Sajjan and L. K. Sannegowda, J. Electroanal. Chem., 2021, 898, 115657,  DOI:10.1016/j.jelechem.2021.115657.
18. S. Abbas, S. Afzal, H. Nadeem, D. Hussain, P. Langer, J. Sevigny, Z. Ashraf and J. Iqbal, Bioorg. Chem., 2022, 118, 105456,  DOI:10.1016/j.bioorg.2021.105456.
19. D. Śmiłowicz and N. Metzler-Nolte, J. Inorg. Biochem., 2020, 206, 111041,  DOI:10.1016/j.jinorgbio.2020.111041.
20. P. ReddyPrasad and T. Imae, J. Taiwan Inst. Chem. Eng., 2017, 72, 194–199,  DOI:10.1016/j.jtice.2017.01.009.
21. Ö. Güngör, Inorg. Chim. Acta, 2025, 574, 122410,  DOI:10.1016/j.ica.2024.122410.
22. A. Benhassine, M. I. Chouiter, M. Kara-Ali, N. Kacem-Chaouche, H. Merazig and A. Belfaitah, J. Mol. Struct., 2023, 1294, 136261,  DOI:10.1016/j.molstruc.2023.136261.
23. Ö. Güngör and P. Gürkan, Arabian J. Chem., 2019, 12, 2244–2256,  DOI:10.1016/j.arabjc.2015.02.009.
24. S. U. Rehman, R. Nawaz and A. Faiz, Synth. React. Inorg., Met.-Org., Nano-Met. Chem., 2012, 42, 1053–1059,  DOI:10.1080/15533174.2011.614314.
25. M. J. Frisch, G. W. Trucks, H. B. Schlegel, G. E. Scuseria, M. A. Robb, J. R. Cheeseman, G. Scalmani, V. Barone, G. A. Petersson, H. Nakatsuji, X. Li, M. Caricato, A. V. Marenich, J. Bloino, B. G. Janesko, R. Gomperts, B. Mennucci, H. P. Hratchian, J. V. Ortiz, A. F. Izmaylov, J. L. Sonnenberg, D. Williams-Young, F. Ding, F. Lipparini, F. Egidi, J. Goings, B. Peng, A. Petrone, T. Henderson, D. Ranasinghe, V. G. Zakrzewski, J. Gao, N. Rega, G. Zheng, W. Liang, M. Hada, M. Ehara, K. Toyota, R. Fukuda, J. Hasegawa, M. Ishida, T. Nakajima, Y. Honda, O. Kitao, H. Nakai, T. Vreven, K. Throssell, J. A. Montgomery, Jr., J. E. Peralta, F. Ogliaro, M. J. Bearpark, J. J. Heyd, E. N. Brothers, K. N. Kudin, V. N. Staroverov, T. A. Keith, R. Kobayashi, J. Normand, K. Raghavachari, A. P. Rendell, J. C. Burant, S. S. Iyengar, J. Tomasi, M. Cossi, J. M. Millam, M. Klene, C. Adamo, R. Cammi, J. W. Ochterski, R. L. Martin, K. Morokuma, O. Farkas, J. B. Foresman and D. J. Fox, Gaussian 16, Revision C.01, Gaussian, Inc., Wallingford CT, 2016.
26. S. Ahsen, I. Tanvir, N. Uddin, T. Yasmeen, S. Abbas, S. Naz, A. Haider, S. Bhattacharya, U. Kortz, F. Rashid, J. Iqbal and S. Ali, Inorg. Chim. Acta, 2024, 564, 121965,  DOI:10.1016/j.ica.2024.121965.
27. A. Fetoh, K. A. Asla, A. A. El-Sherif, H. El-Didamony and G. M. Abu El-Reash, J. Mol. Struct., 2019, 1178, 524–537,  DOI:10.1016/j.molstruc.2018.10.066.
28. A. Mederos, S. Dominguez, R. Hernandez-Molina, J. Sanchiz and F. Brito, Coord. Chem. Rev., 1999, 193–195, 857–911.
29. A. A. El-Sherif and M. S. Aljahdali, J. Coord. Chem., 2013, 66, 3423–3468,  DOI:10.1080/00958972.2013.839027.
30. H. Joshi, F. S. Kamounah, C. Gooijer, G. van der Zwan and L. Antonov, J. Photochem. Photobiol., A, 2002, 152, 183–191.
31. S. R. Salman and F. S. Kamounah, Spectrosc. Lett., 2002, 35, 327–335.
32. R. Chandra, A. K. Manna, M. Sahu, K. Rout and G. K. Patra, Inorg. Chim. Acta, 2020, 499, 119192,  DOI:10.1016/j.ica.2019.119192.
33. B. Samanta, J. Chakraborty, S. Shit, S. R. Batten, P. Jensen, J. D. Masuda and S. Mitra, Inorg. Chim. Acta, 2007, 360, 2471–2484,  DOI:10.1016/j.ica.2006.12.019.
34. M. Kumar, A. K. Singh, A. K. Singh, R. K. Yadav, S. Singh, A. P. Singh and A. Chauhan, Coord. Chem. Rev., 2023, 488, 215176,  DOI:10.1016/j.ccr.2023.215176.
35. C. P. Matos, Y. Addis, P. Nunes, S. Barroso, I. Alho, M. Martins, A. P. A. Matos, F. Marques, I. Cavaco, J. C. Pessoa and I. Correia, J. Inorg. Biochem., 2019, 198, 110727,  DOI:10.1016/j.jinorgbio.2019.110727.
36. N. Ghobadi, Int. Nano Lett., 2013, 3, 2–4,  DOI:10.1186/2228-5326-3-2.
37. K. Mariselvam and J. Liu, Opt. Mater., 2021, 114, 110997,  DOI:10.1016/j.optmat.2021.110997.
38. Ö. B. Mergen and E. Arda, Synth. Met., 2020, 269, 116539,  DOI:10.1016/j.synthmet.2020.116539.
39. I. Kaya, B. Ayten and D. Şenol, Opt. Mater., 2018, 78, 421–431,  DOI:10.1016/j.optmat.2018.02.057.
40. J. P. Wu, Y. C. Cheng, L. Lu, J. Wang and S. B. Qiao, Acta Crystallogr., Sect. C: Cryst. Struct. Commun., 2021, C77, 257–261,  DOI:10.1107/S2053229621004642.
41. E. N. Nuraneeva, G. B. Guseva, E. V. Antina, M. B. Berezin and A. I. V’yugin, J. Lumin., 2016, 170, 248–254,  DOI:10.1016/j.jlumin.2015.10.061.
42. T. Wang, Q. Pang, Z. Tong, M. Wang and N. Xiao, Spectrochim. Acta, Part A, 2021, 250, 119249,  DOI:10.1016/j.saa.2020.119249.
43. M. Mukhopadhyay, D. Banerjee, A. Koll, A. Filarowski and S. Mukherjee, Spectrochim. Acta, Part A, 2005, 62, 126–131,  DOI:10.1016/j.saa.2004.12.015.
44. Y. P. Huo, S. Z. Zhu and S. Hu, Tetrahedron, 2010, 66, 8635–8640,  DOI:10.1016/j.tet.2010.09.039.
45. P. Mohanty, R. Behura, V. Bhardwaj, P. P. Dash, S. K. Sahoo and B. R. Jali, Trends Environ. Anal. Chem., 2022, 34, e00166,  DOI:10.1016/j.teac.2022.e00166.
46. X. Ma, W. Chi, X. Han, C. Wang, S. Liu, X. Liu and J. Yin, Chin. Chem. Lett., 2021, 32, 1790–1794,  DOI:10.1016/j.cclet.2020.12.031.
47. X. Liu, Y. Sun, Y. Gao, X. Zhang, X. Li, W. Zheng, M. Liu, T. Zhao, X. A. Yuan, M. Yue and Z. Liu, J. Inorg. Biochem., 2025, 262, 112767,  DOI:10.1016/j.jinorgbio.2024.112767.
48. K. Ogawa and J. Harada, J. Mol. Struct., 2003, 647, 211–216,  DOI:10.1016/S0022-2860(02)00526-4.
49. A. Ohshima, A. Momotake and T. Arai, J. Photochem. Photobiol., A, 2004, 162, 473–479,  DOI:10.1016/S1010-6030(03)00388-5.
50. N. Klinhom, N. Saengsuwan, S. Sriyab, P. Prompinit, S. Hannongbua and S. Suramitr, Spectrochim. Acta, Part A, 2019, 206, 359–366,  DOI:10.1016/j.saa.2018.08.010.
51. V. I. Minkin, A. V. Tsukanov, A. D. Dubonosov and V. A. Bren, J. Mol. Struct., 2011, 998, 179–191,  DOI:10.1016/j.molstruc.2011.05.029.
52. Y. Zhao, K. Chen, B. Yu, Q. Wan, Y. Wang, F. Tang and X. Li, J. Mol. Struct., 2024, 1317, 139126,  DOI:10.1016/j.molstruc.2024.139126.
53. T. Jiang, Y. Fan, J. H. Lu, C. Huang and B. X. Zhu, Spectrochim. Acta, Part A, 2024, 322, 124827,  DOI:10.1016/j.saa.2024.124827.
54. M. J. Tian, H. Q. Tian, T. Jiang, H. H. Yu, C. Huang, B. X. Zhu and C. Zhu, Tetrahedron, 2024, 167, 134288,  DOI:10.1016/j.tet.2024.134288.
55. G. Zhang, S. Yu, J. Peng, Y. Xiong, L. Hu and W. Lai, Trends Anal. Chem., 2025, 183, 118098,  DOI:10.1016/j.trac.2024.118098.
56. J. J. Shao, J. L. Ni, Y. Liang, X. D. Xu and F. M. Wang, J. Mol. Struct., 2021, 1244, 130975,  DOI:10.1016/j.molstruc.2021.130975.
57. D. Majumdar, S. Dey, D. Das, D. K. Singh, S. Das, K. Bankura and D. Mishra, J. Mol. Struct., 2019, 1185, 112–120,  DOI:10.1016/j.molstruc.2019.02.092.
58. A. He, C. Zhong, H. Huang, Y. Zhou, Y. He and H. Zhang, J. Lumin., 2008, 128, 1291–1296,  DOI:10.1016/j.jlumin.2007.12.031.
59. E. Liu, Y. Zhuo Zhang, J. Tan, C. Yang, L. Li, J. A. Golen, A. L. Rheingold and G. Zhang, Polyhedron, 2015, 102, 41–47,  DOI:10.1016/j.poly.2015.07.074.
60. C. Anitha, C. D. Sheela, P. Tharmaraj and S. J. Raja, Spectrochim. Acta, Part A, 2012, 98, 35–42,  DOI:10.1016/j.saa.2012.08.022.
61. F. Purtas, K. Sayin, G. Ceyhan, M. Kose and M. Kurtoglu, J. Mol. Struct., 2017, 1137, 461–475,  DOI:10.1016/j.molstruc.2017.02.065.
62. Ya. N. Albrekht, V. F. Plyusnin, E. M. Glebov, M. S. Milutka, A. S. Burlov, Y. V. Koshchienko, V. G. Vlasenko, V. A. Lazarenko and L. D. Popov, J. Lumin., 2024, 266, 120286,  DOI:10.1016/j.jlumin.2023.120286.
63. N. Ahmad and R. Shirin, Iran. J. Chem. Chem. Eng., 2019, 38, 79–90.
64. S. Z. Zhang, G. Guo, W. M. Ding, J. Li, Y. Wu, H. J. Zhang, J. Q. Guo and Y. X. Sun, J. Mol. Struct., 2021, 1230, 129627,  DOI:10.1016/j.molstruc.2020.129627.
65. M. Liu, H. Yang, D. Li, Q. Yao, H. Wang, Z. Zhang and J. Dou, Inorg. Chim. Acta, 2021, 522, 120384,  DOI:10.1016/j.ica.2021.120384.
66. J. Cao, Z. Jia, W. Chen, Y. Song, Z. Yu, Y. Dong and Y. Ren, Inorg. Chim. Acta, 2025, 577, 122494,  DOI:10.1016/j.ica.2024.122494.
67. J. Dong, T. Wan, K. Li, X. Kong, Q. Shen and H. Wu, J. Mol. Struct., 2022, 1264, 133340,  DOI:10.1016/j.molstruc.2022.133340.
68. S. Dalbera, S. S. Saha, H. A. Sparkes and S. Mohanta, J. Mol. Struct., 2025, 1319, 139537,  DOI:10.1016/j.molstruc.2024.139537.
69. H. Pyngrope, J. Chetia, B. Kharpan, R. Nandi, A. K. Pradhan, P. C. Paul, H. A. R. Pramanik and D. Kumar, Inorg. Chem. Commun., 2025, 173, 113770,  DOI:10.1016/j.inoche.2024.113770.
70. G. Keser Karaoğlan, J. Mol. Struct., 2022, 1256, 132534,  DOI:10.1016/j.molstruc.2022.132534.
71. E. H. H. Hasabeldaim, H. C. Swart and R. E. Kroon, RSC Adv., 2023, 13, 5353–5366,  10.1039/d2ra07897j.
72. C. C. Vidyasagar, B. M. Muñoz Flores, V. M. Jiménez-Pérez and P. M. Gurubasavaraj, Mater. Today Chem., 2019, 11, 133–155,  DOI:10.1016/j.mtchem.2018.09.010.
73. F. S. M. Canisares, A. G. Bispo-Jr, A. M. Pires and S. A. M. Lima, Optik, 2020, 219, 164995,  DOI:10.1016/j.ijleo.2020.164995.
74. L. E. Lopez-Ruiz, Ch. J. Salas-Juárez, I. Garduño-Wilches, H. I. Beltran, U. Orozco-Valencia, R. I. López-Esquivel, J. C. Guzman-Olguin, M. Centeno-Alvarez and J. Guzman-Mendoza, J. Lumin., 2023, 263, 120020,  DOI:10.1016/j.jlumin.2023.120020.
75. N. Ganji, A. Rambabu, N. Vamsikrishna, S. Daravath and Shivaraj, J. Mol. Struct., 2018, 1173, 173–182,  DOI:10.1016/j.molstruc.2018.06.100.
76. A. Daina, O. Michielin and V. Zoete, J. Chem. Inf. Model., 2014, 54, 3284–3301,  DOI:10.1021/ci500467k.
77. A. Daina and V. Zoete, ChemMedChem, 2016, 11, 1117–1121,  DOI:10.1002/cmdc.201600182.
78. Z. H. Chohan, S. H. Sumrra, M. H. Youssoufi and T. B. Hadda, Eur. J. Med. Chem., 2010, 45, 2739–2747,  DOI:10.1016/j.ejmech.2010.02.053.
79. A. Daina, O. Michielin and V. Zoete, Sci. Rep., 2017, 7, 42717,  DOI:10.1038/srep42717.

---