New Tools in Heavy Metal Detection: Synthesis, Spectroscopic, and Quantum Chemical Characterization of Selected Water-Soluble Styryl Derivatives of Quinoline and 1,10-Phenanthroline

Abstract

A series of water-soluble molecules based on 8-isopropyl-2-methyl-5-nitroquinoline and 1,10-phenanthroline core were designed by introducing a π-conjugated bridge, vinyl unit –CH=CH–. We present the selective conversion of methyl groups located on the C2 and C9 positions in the constitution of selected quinoline or 1,10-phenanthroline derivatives, respectively, into vinyl (or styryl) products by applying Perkin condensation. The two groups of ligands differ in the presence of one or two arms. The structure of the molecule ((1E,1′E)-(1,10-phenanthroline-2,9-diyl)bis(ethene-2,1-diyl))bis(benzene-4,1,3-triyl) tetraacetate was determined by single-crystal X-ray diffraction measurements. The X-ray, NMR, and DFT computational studies indicate the influence of rotation (rotamers) on the physical properties of studied styryl molecules. The results show that the styryl molecules with the vinyl unit –CH=CH– exhibit significant static and dynamic hyperpolarizabilities. Quantum chemical calculations using density functional theory and B3LYP/6-311++G(d,p) with Grimme’s dispersion correction approach predict the existence and relative stability of different spatial cis(Z)- and trans(E)-conformers of styryl derivatives of quinoline and 1,10-phenanthroline, which exhibit different electronic distribution and conjugation within the molecular skeleton, dipole moments, and steric interactions, leading to variations in their photophysical behavior and various applications. Our studies indicate that the rotation and isomerization of aryl groups can significantly influence the electronic and optical properties of π-conjugated systems, such as vinyl units (–CH=CH–). The rotation of aryl groups around the single bond that connects them to the vinyl unit can lead to changes in the effective π-conjugation between the aryl group and the rest of the π-conjugated system. The rotation and isomerization of aryl groups in π-conjugated systems significantly impact their electronic and optical properties. These changes can modify the efficiency of π-conjugation, affecting charge transfer processes, absorption properties, light emission, and electrical conductivity. In designing optoelectronic materials, such as organic dyes, organic semiconductors, or electrochromic materials, controlling the rotation and isomerization of aryl groups can be crucial for optimizing their functionality.

1. Introduction

Quinoline and 1,10-phenanthroline derivatives represent highly versatile photoactive compounds that demonstrate distinct properties in coordination chemistry [1,2,3,4,5], medicinal chemistry [6,7,8,9,10,11,12,13,14,15], chemical engineering [16,17], electrochemistry [18,19], industrial chemistry [20,21], etc., [22,23]. Despite their importance, water-soluble ligands based on quinoline or 1,10-phenanthroline core are relatively poorly studied compounds. Developing efficient and convenient syntheses for such ligands could open new avenues for functional material and complex design [24,25].

Our studies describe novel and practical ways to introduce a polyphenol group under mild reaction conditions. In particular, we apply Perkin condensation to synthesize a vinyl (or styryl) analog of 1,10-phenanthroline and quinoline derivatives with a phenol function. This reaction also demonstrates a new, simple, and efficient strategy for converting methyl derivatives of 1,10-phenanthroline or quinoline into more reactive styryl derivatives [24]. We anticipate that the new way of converting methyl will find wide application in chemical synthesis.

The synthesized styryl derivatives of quinoline and 1,10-phenanthroline possess a specific π-conjugated structure, containing a vinyl that functions as a bridge between an azoquinoline or phenanthroline moiety and an additional phenyl ring. Such extended π-conjugation increases the stability of the molecule and ensures efficient π-delocalization throughout the whole structure.

Quantum chemical calculations using density functional theory (DFT) support the existence of multiple spatial cis(Z)- and trans(E)-conformers of these styryl derivatives of quinoline and 1,10-phenanthroline, which exhibit different electronic distribution and conjugation within the molecular skeleton, dipole moments, and steric interactions, leading to variations in their photophysical behavior. The presence of nitrogen atoms within the molecular system enables it to bind to specific ions selectively. As a result, styryl derivatives of quinolines and 1,10-phenanthrolines are utilized in chemical sensors for detecting heavy metal ions, such as copper and mercury, which is significant in environmental protection and chemical analysis [26,27,28,29,30].

For example, Aatif and Kumar designed a novel phenanthroline-benzothiazole fluorescent sensor (L) for detecting Hg²⁺ in both biological and environmental samples [27]. Fernandes et al. developed a self-contained, portable iron measurement system based on spectroscopic absorption using the phenanthroline complex for the detection of Fe²⁺ in water [31]. The device demonstrated high sensitivity (2.5 µg Fe²⁺/L) and a linear response in the 25–1000 µg Fe²⁺/L range [31]. Mula et al. reported a BODIPY-phenanthroline conjugate that allows highly selective, fluorescence turn-off detection of Cu²⁺ ions [32]. The sensor demonstrated 1:1 complex formation with Cu²⁺ and was successfully applied in live-cell imaging and serum sample analysis [32]. Riela et al. developed a hectorite/phenanthroline nanohybrid material as a selective and stable fluorescent sensor for Zn²⁺ detection in aqueous solution [33]. This material exhibited strong interaction with Zn²⁺ ions and was characterized by combined spectroscopic, microscopic, and theoretical techniques to confirm its sensing potential and structural stability [33]. Zhou et al. presented a dual-function fluorescence sensor based on the inner filter effect between NaYF₄:Yb,Er@NaYF₄@PAA and the Fe(II)-1,10-phenanthroline complex for the sensitive detection of Sn(II) and ascorbic acid (AA) [34]. The system exhibited excellent selectivity and linear response with detection limits of 1.08 μM for Sn(II) and 0.97 μM for AA, and was successfully applied to tap and spring water samples [34]. Soleymani et al. reported a dual-emission ratiometric fluorescent sensor based on terbium-1,10-phenanthroline–nitrogen-doped-graphene quantum dots for the sensitive detection of metformin in biological samples [35].

In this paper, we demonstrate how our newly synthesized water-soluble ligands align with these design principles and explore their photophysical and coordination properties in depth. In this context, a series of spectroscopic investigations were carried out to evaluate the response of the ligands toward various metal ions, focusing on changes in absorbance behavior and fluorescence intensity (quenching). Among the tested ions, Cu²⁺ induced the most significant spectral changes, highlighting the high specificity of the phenanthroline derivative.

2. Results and Discussion

2.1. Synthesis and Structural Characterization

Recently, we reported the synthesis of a vinyl analog of 1,10-phenanthroline based on Perkin condensation [24]. In the present study, we successfully obtained a similar molecule 2,2′-((1E,1′E)-(1,10-phenanthroline-2,9-diyl)bis(ethene-2,1-diyl))diphenol, a vinyl analog of 1,10-phenanthroline, which, like our previously published molecule, has a phenolic group. This prompted us to obtain an analog with a polyphenol group, providing better solubility in water. For this purpose, we used a previously developed procedure in which we replaced 2-hydroxybenzaldehyde with 2,4-dihydroxybenzaldehyde. We received the product with 91% efficiency (Scheme 1). Since quinolines, like 1,10-phenanthrolines, are commonly used as ligands, we have expanded our research work to include this group of molecules. We chose the less commonly used, i.e., molecules 2-methyl-8-(trifluoromethyl)quinoline (1b) and 8-isopropyl-2-methyl-5-nitroquinoline (1c) as models. An analog with a trifluoromethyl group can be used as an NMR probe. The second model, on the other hand, refers to the isopropyl group of the first one, and the nitro group was chosen because, after reduction, the amino analogs will allow for further functionalization of the obtained compounds.

2.2. Single-Crystal X-Ray Data and Hirshfeld Surface Analysis

((1E,1′E)-(1,10-Phenanthroline-2,9-diyl)bis(ethene-2,1-diyl))bis(benzene-4,1,3-triyl) tetraacetate (2a)(Figure 1a) crystallizes in the monoclinic system, P2₁/c, with unit cell dimensions of a = 9.5864(6) Å, b = 36.0174(19) Å, c = 8.6135(5) Å, cell volume V = 2973.01ų, and Z = 4; the monoclinic angle (β) is 91.514(6), and the unit cell contains four molecules (Figure 1b–d). There is one molecule in the asymmetric part of the unit cell. The atom numbering of the selected atoms is shown in Figure 1a. The selected intra- and intermolecular interactions in 2a presented in Figure 1b,c affect molecular conformation and crystal packing. The shortest intramolecular distance, 2.512 Å, belongs to the H···H contact, indicating very close proximity between the two hydrogen atoms in the 2a molecule. The 5.164 Å and 4.466 Å distances correspond to long-range CH···N intramolecular contacts in 2a. These contacts stabilize the molecular conformation and crystal packing, albeit at relatively weak interaction distances. The intermolecular N···N distances are 4.050 Å and 4.704 Å (Figure 1c).

Crystals of 2a consist of closely spaced continuous columns of molecules (Figure 1e) linked together by numerous CH···O intermolecular interactions, which can be classified as weak hydrogen contacts according to the characteristics proposed by Desiraju and Steiner [36]. Each column consists of mutually parallel molecules with plane-to-plane distances between the centroids in parallel phenanthroline fragments of 3.954 Å and 4.007 Å. Although these distances are slightly larger than the optimal range for strong π···π interactions (3.3–3.8 Å), they still suggest weak π···π interactions due to the overlap of p-orbitals between the closely aligned phenanthroline cores in a face-to-face arrangement (Figure 1d). The detailed analysis of intermolecular interactions was performed with Hirshfeld d_norm surfaces of selected dimer configurations within the crystal of 2a (Figure 1f). The red spots on d_norm indicate numerous CH···O and CH···N intermolecular interactions, which involve acetate groups and phenanthroline core (Figure 1g–k); these contacts contribute mainly to stability and well-ordered crystal packing motif of 2a

The shape index map (Figure 1l) uses concave (red) and convex (blue) regions to represent areas where molecules fit into each other, similar to a lock-and-key principle. The presence of the red and blue triangles displays the areas where the π-system of one molecule intercalates or overlaps with another through van der Waals forces. It confirms the weak π···π stacking interactions between phenanthroline cores placed one above the other in neighboring molecules of 2a.

The curvedness map in Figure 1m displays areas with different packing densities, where green regions represent flat molecular surfaces with low curvature and correspond to the π–π stacking regions in 2a.

2.3. Structural Characterization of the Quinoline Derivatives 3b, 3c, and 3d

The B3LYP/6-311G++G(d,p), including Grimme’s dispersion correction optimized ground state structures of the trans-, cis-isomers, and the transition states (TS) with indicated intramolecular O···H, and N···H interactions for 3b, 3c, and 3d are presented in Figure 2. The selected angles and energy barriers are summarized in

Table 1. Selected structural parameters of trans- and cis-isomers of 3b, 3c, and 3d and their corresponding energy differences ΔE (reported in kcal mol⁻¹).

Compound	A(C2C3C4)	A(C3C4C5)	D(C1C2C3C4)	D(C2C3C4C5)	D(C3C4C5N6)	ΔE
3b (trans)	129.54	125.38	−0.002	−180.00	179.99	3.81
3b (cis)	130.54	129.84	−137.36	5.68	−157.61
3c (trans)	129.74	125.29	0.32	−179.87	−178.35	1.61
3c (cis)	128.87	128.57	138.76	−8.83	−18.77
3d (trans)	126.91	126.41	1.44	−179.78	−178.17	3.61
3d (cis)	129.07	129.21	143.64	−7.33	149.47

The energy differences (ΔE) reported in Table 1 correspond to the B3LYP/6-311++G(d,p) with Grimme’s dispersion correction calculated energy differences relative to the most stable trans-isomer. Angles are reported in degrees.

and

Table 2. Selected structural parameters and energy barrier of the transition states (TS) of 3b, 3c, and 3d.

TS	A(C2C3C4)	A(C3C4C5)	D(C1C2C3C4)	D(C2C3C4C5)	D(C3C4C5N6)	ΔEZPE
3b	123.25	126.65	83.22	179.89	−175.09	13.56
3c	123.35	126.60	84.33	179.91	−175.91	4.72
3d	123.42	126.67	85.73	179.92	−176.12	5.43

The energy barrier of TSs (ΔE_ZPE: TS—cis) is in kcal mol⁻¹ (zero-point energies (ZPE) correction included). Angles are reported in degrees. Atom numbering corresponds to Figure 3.

. The structure of 3b, 3c, and 3d trans-isomers is planar with dihedral angles D(C1C2C3C4) and D(C2C3C4C5) close to 0° and 180°, respectively (Figure 3, Table 1). Planarity facilitates π-conjugation of 3b, 3c, and 3d structures, increasing their stability. The 3b, 3c, and 3d cis-isomers exhibit significant geometrical distortions with non-planar D(C1C2C3C4) and D(C2C3C4C5) dihedral angle values (Table 1). The bond angles (A(C2C3C4), A(C3C4C5) vary slightly between the trans- and cis-isomers, but remain close to 120°, reflecting the sp² hybridization of these carbon atoms.

The structure of the 3b trans-isomer is stabilized by weak intramolecular O···H and N···H interactions with distances 2.205 Å and 2.468 Å (Figure 2). In the cis-isomer of the 3b, the O···H distance increases to 2.985 Å, indicating a weaker interaction compared to the trans form. The 3c and 3d trans- and cis-isomers are also stabilized by intramolecular O···H and N···H interactions (Figure 2). All the compounds 3b, 3c, and 3d possess structural trend that trans form exhibits relatively shorter and stronger O···H and N···H interactions compared to the corresponding cis-isomer. The energy differences (ΔE) between the cis- and trans-isomers indicate that the trans-isomers are energetically more stable, consistent with their planar and less strained structure. The energy differences relative to the trans-isomer are 3.81 kcal mol⁻¹ for 3b, 1.61 kcal mol⁻¹ for 3c, and 3.61 kcal mol⁻¹ for 3d (Table 1).

The TS structures of 3b, 3c, and 3d exhibit distorted geometries with intermediate O···H and N···H distances, consistent with partial bond breaking/forming during the isomerization process. The cis–trans isomerization reaction proceeds primarily via an inversion pathway, as indicated by the significant increase in the dihedral angle D(C1C2C3C4) in the TS. As one can see from Table 1 and Table 2, the D(C1C2C3C4) changes from nearly 0° in the trans-isomers to around ±140° in the cis-isomers, and to ~83–86° in the TS. This structural feature supports the predominance of inversion over a pure rotation mechanism around the C=C bond. However, the rotation process plays a role in the changes in other torsional and bond angles (Table 1 and Table 2), which suggests that the process also involves a rotational component, consistent with a combined inversion-rotation mechanism.

The TS energy barriers for 3b, 3c, and 3d vary significantly among the compounds; 3b shows the highest energy barrier (13.56 kcal mol⁻¹, Table 2), which reflects the higher stability of the trans-isomer. Energy barriers for 3c and 3d are much lower (4.72 kcal mol⁻¹ and 5.43 kcal mol⁻¹, respectively, Table 2) due to weaker intramolecular interactions and increased torsion in the TS. The lower energy barriers for 3c and 3d suggest a higher probability of cis–trans interconversion, leading to nonradiative decay pathways that quench photoluminescence.

2.4. Conformational Analysisof 3a

We simulated six structural trans-, cis-, and cis–trans rotamers 1–6 for 3a (Figure 4 and Figure S2), and their highest occupied molecular orbital (HOMO) and lowest unoccupied molecular orbital (LUMO) are shown in Figure 5, and Table S1. The trans rotamer 1 of 3a is energetically the most stable, due to minimized steric hindrance, and a centroid-to-centroid distance of 7.411 Å between the two terminal benzene rings with OH groups placed in opposite directions. The HOMO–LUMO gap for this rotamer 1 is 3.35eV.

The other two cis rotamers, 4 and 5 of 3a, are only 1.28 and 1.29 kcal·mol⁻¹ higher in energy than trans rotamer 1. Rotamer 4, with the unidirectional orientation of the terminal substituents, is stabilized by an intramolecular O···H interaction 1.940 Å, which appears between the OH groups on the terminal benzene rings. It exhibits a HOMO–LUMO gap of 3.56 eV which suggests a more localized electronic distribution and slightly reduced π-conjugation compared to trans rotamer 1 of 3a. Rotamer 5, with bidirectional branches, adopts a helical shape, with a centroid-to-centroid distance of 3.611 Å between the terminal benzene rings. This rotamer is stabilized by a network of intramolecular O···H and N···H interactions 2.064 Å, 2.090 Å, 2.387 Å, and 3.424 Å (Figure 4). Rotamer 5 has a HOMO–LUMO gap of 3.55 eV, similar to that of cis rotamer 4, indicating the same π-delocalization principle.

Trans rotamer 2, with unilaterally placed OH groups on the terminal benzene rings, lies higher by 1.39 kcal·mol⁻¹ compared to trans rotamer 1. While the HOMO–LUMO gap of rotamer 2 is only slightly larger (3.36 eV) than that of rotamer 1 (3.35 eV), this minor energy difference between the two rotamers originates primarily from subtle steric and conformational effects related to the orientation of the OH groups, rather than from differences in electronic delocalization, as evidenced by the similar shapes of the HOMO and LUMO shown in Figure 5.

Cis-trans rotamer 3 lies 2.47 kcal·mol⁻¹ higher in energy than trans rotamer 1; the structure adopts an intermediate conformation in which one terminal OH-substituted benzene ring is oriented in a cis fashion (toward the phenanthroline core), while the other is trans-oriented (away from the phenanthroline core). This mixed conformation leads to a more compact geometry than fully trans rotamers 1 and 2. The HOMO–LUMO gap (3.32 eV) is narrower than that of rotamers 1 (3.35 eV) and 2 (3.36 eV), indicating slightly better π–π overlap in this intermediate geometry (Figure 5). The slightly higher HOMO energy (–5.17 eV), compared to rotamers 1 (−5.23 eV) and 2 (−5.25 eV), suggests reduced stabilization due to increased intramolecular steric strain, although its electronic properties are relatively similar.

Cis rotamer 6 is higher in energy than trans rotamer 1, with a relative energy difference of 4.81 kcal·mol⁻¹ (Figure 4). This rotamer adopts a fully cis conformation, where both OH-substituted terminal benzene rings are oriented toward the phenanthroline core. This geometry leads to distortion of the π-conjugated system, which reduces planarity and increases molecular strain. The HOMO–LUMO gap of rotamer 6 is 3.32 eV, the same as for cis–trans rotamer 3 (Figure 5). This suggests that despite the steric strain and non-planarity in rotamer 6, the π–π* interactions remain moderately preserved, due to local orbital overlap between adjacent aromatic fragments. The HOMO energy level of rotamer 6 is higher (–5.10 eV) compared to the other rotamers 1–5 in Figure 5, which suggests reduced stabilization of the HOMO. This supports that intramolecular repulsion and conformational strain in rotamer 6 dominate electronic delocalization.

2.5. Spectroscopic Characterizationof the Phenanthroline Derivative 3a and of the Quinoline Derivatives 3c and 3d

The derivatives 1,10-phenanthroline (3a) and quinolines (3c and 3d) were spectroscopically characterized, and their photoluminescence properties, such as absorption maximum, molar extinction coefficient, excitation, emission, and luminescence decay time, were investigated. The data obtained are summarized in

Table 3. Spectroscopic characterization of 3a, 3c, and 3d with different solutions.

	λabs [nm]	ε [M−1cm−1]	λem[nm]	Stokes Shift [nm]	Φf [%]	τ [ns]
3a	463 a 351 a	18,700 a 14,400 a	550a	87 a	<1 a	2.75 a
	387 c 346 c	19,400 c 18,600 c	500c	113 c	8.17 c	0.99 c
3c	350 b	21,400 b	-	-	-	-
	362 c	20,300 c	-	-	-	-
	429 d	27,400 d	-	-	-	-
3d	346 b	25,900 b	-	-	-	-
	359 c	23,800 c	-	-	-	-
	323 d	118,00 d	-	-	-	-

^a EtOH, ^b MeOH, ^c DMSO, ^d 0.1 M KOH.

and Tables S1–S3. In addition, the optimized geometries using different functionals, absorption, excitation, and emission spectra of the studied compounds 3a, 3c,and 3d are presented in Figure 6, Figure 7, Figure 8, Figure 9 and Figure 10 and Figures S3–S5. The 1,10-phenanthroline derivative (3a) shows two absorption bands in ethanol (463 nm and 351 nm) and dimethylsulfoxide (DMSO) (387 nm and 346 nm) and exhibits fluorescence with an emission maximum in the 500–550 nm range (Table 3).

The solvent’s polarity significantly affects the spectroscopic properties of compound 3a, with a bathochromic shift in the absorption and emission maxima observed as the solvent polarity decreases, indicating that the excited state of 3a is stabilized more effectively in less polar solvents. In ethanol, 3a shows the emission peak at 550 nm, resulting in a moderate Stokes shift of 87 nm. This indicates minimal structural reorganization between the ground singlet state (S₀) and the first excited singlet state (S₁), which is confirmed by quantum-chemical calculations (Figure S2); 3a efficiently relaxes from the S₁ state to the S₀ state before emission. In DMSO, the emission peak shifts to 500 nm, reducing the emission wavelength and increasing the Stokes shift to 113 nm. This larger shift reflects solvent polarity effects, as DMSO stabilizes the excited state differently than ethanol. Depending on the solvent used, the fluorescence quantum yield (Φ_f) of compound 3a ranges from less than 1% to 8.17%. The obtained fluorescence intensity decays were fitted by biexponential and triexponential models for DMSO and EtOH, respectively (Figure S7). In ethanol, the excited-state lifetime of 3a is 2.75 ns, consistent with weak fluorescence. In DMSO, the lifetime decreases to 0.99 ns despite the higher Φ_f. This indicates faster decay pathways in DMSO due to a more stabilized S₁ state and enhanced radiative decay rates. The noticeable shifts in absorption and emission maxima for 3a with changes in solvent polarity demonstrate its solvatochromic behavior, making it potentially useful for sensing applications.

The quinoline derivative (3c) has an absorption band with a near-ultraviolet maximum at 350 nm in methanol and 362 nm in DMSO, with slightly higher ε values than 3a. The presence of a third hydroxyl group in the benzene ring of the quinoline derivative (3d) did not cause any significant changes in the absorption spectra in the presence of MeOH and DMSO (Figure 6).

The quantum-chemical time-dependent (TD) DFT/B3LYP/6-31G(d,p) calculations of the absorption spectra for the trans and cis rotamers of 3c and 3d in DMSO confirm the experimental observations, with absorption maxima in the near-UV region calculated at 364 nm for 3c and 366 nm for 3d in the convolution profiles (Figure 7). This band corresponds to the HOMO→LUMO+1 transition for both trans and cis forms of 3c and 3d (Tables S1 and S2). The first low-intensity S₀→S₁ transition is due to HOMO→LUMO configuration and was calculated at 474 nm (f =0.1051) for trans3c, 482 nm (f = 0.0283) for cis3c, 476 nm (f =0.1078) for trans3d, 467 nm (f =0.0699) for cis3d (Table S2).

The alkaline environment strongly affects the electronic charge distribution of 3c and 3d, as can be seen from the UV-Vis spectra (Figure 6), as the hydroxyl groups deprotonate to form phenolate anions. For the two hydroxyl groups in the structure of compound 3c, a shift in the absorption maximum towards longer wavelengths is observed (Figure 6), since the two phenolate anions increase the charge delocalization in the aromatic system, thus reducing the energy difference between HOMO and LUMO. In contrast, in the case of three ionized hydroxyl groups, there is an excessive accumulation of negative charge, leading to a destabilization of the excited state and an associated increase in the difference between HOMO and LUMO. The effect is a shift in the absorption maximum towards shorter wavelengths (Figure 6).

The frontier molecular orbitals (FMO) of 3c and 3d and their ionized anionic forms 3c²⁻ and 3d³⁻ are presented in Figure 8. For the neutral 3c and 3d, the HOMO is mainly delocalized over the conjugated aromatic system with additional electron density extending to the hydroxyl groups. This reflects the π-conjugation across the whole system of 3c and 3d involving electron-donating hydroxyl groups. The LUMO is localized on the quinoline fragment and extends to the electron-withdrawing nitro group, which reflects its ability to accept electron density in electron-deficient (electrophilic) regions of the molecule (Figure 8). Such spatial separation between the HOMO (on electron-rich region) and LUMO (on electron-deficient region) suggests a pronounced intramolecular charge transfer (ICT) character in the excited state for neutral molecules of 3c and 3d.

After deprotonation of both hydroxyl groups in 3c, the HOMO becomes significantly more localized, with electron density strongly concentrated on the negatively charged oxygen atoms. In contrast, the LUMO exhibits a broadened distribution across the molecular system of 3c², primarily extending over the electron-deficient nitro group, as well as the quinoline core and phenolate ring. This spatial extension of the LUMO reflects enhanced charge delocalization in the 3c²⁻ excited state. The increased separation of charge between the localized HOMO and delocalized LUMO reduces the HOMO-LUMO energy gap and improves the ICT. This decrease in excitation energy correlates directly with the experimentally observed red shift in the UV-visible absorption spectrum of 3c^2– compared to the neutral 3c.

For the three ionized hydroxyl derivative of 3d, the HOMO becomes more localized, and its electron density is asymmetric due to the excess negative charge from the deprotonated phenolate groups (Figure 8); this leads to destabilization of the HOMO energy level. The LUMO also undergoes a significant dual localization effect, being concentrated on the ionized hydroxyl group and the electron-withdrawing nitro group (Figure 8). Such spatial distribution, which is affected by the Coulomb repulsion and the altered electrostatic potential throughout the molecular system, leads to a more destabilized LUMO than the HOMO. This localization effect affects the orbital overlap and the nature of electronic transitions, contributing to the observed blue shift in the experimental absorption spectrum of 3d in the alkaline environment.

Quinoline derivatives 3c and 3d with the nitro group do not exhibit photoluminescent properties that are typical for nitroaromatic compounds due to the nitro group effect, which introduces non-radiative decay mechanisms and distorts the electronic structure [37,38,39,40,41].

For a detailed analysis of the observed spectrum of 3a in Figure 9, we performed TD DFT calculations of vertical electronic transitions simulated in DMSO for six rotamers 1–6, which can coexist and affect the absorption and emission spectra of 3a. The weak absorption observed experimentally in the 520–430 nm region in Figure 9a corresponds to the S₀→S₁ transitions calculated at 428 nm for trans rotamer 1 (oscillator strength, f = 0.3689), 422 nm for trans rotamer 2 (f = 0.5993), 433 nm for cis–transrotamer 3 (f = 0.1561), 414 nm for cis rotamer 4 (f = 0.0838), 444 nm for cis rotamer 5 (f = 0.0637), and 395 nm for cis rotamer 6 (f = 0.2308). Among these, the trans rotamer 2 makes the largest contribution to this absorption. For rotamers 1–3, the S₀→S₁ transition predominantly involves the HOMO→LUMO excitation, whereas for rotamers 4–6, it involves a mixture of HOMO→LUMO and HOMO−1→LUMO transitions (Figure 5).

The observed absorption band at 387 nm in DMSO (Table 3, Figure 9a) is well reproduced by the TDDFT-calculated convolution line (black line in Figure 10) with a maximum at 383 nm. This band (383 nm) has a complex nature and corresponds to the absorption mainly from the trans rotamers 1 and 2, with a minor contribution from the cis–trans rotamer 3, and the cis rotamers 4–6 (Figure 10). For the trans rotamers of 3a, this absorption arises from the S₀→S₂ transition calculated at 385 nm for rotamer 1 and 382 nm rotamer 2 (Figure 10). The observed absorption band at 346 nm (Figure 9a, Table 3) corresponds to the calculated maximum at 335 nm in the convolution line (Figure 10). This band arises mainly from transition S₀→S₄ at 346 nm of trans rotamer 2 and transition S₀→S₅ at 332 nm of trans rotamer 1, with trans rotamer 2 being the major contributor.

The fluorescence of 3a corresponds to the S₁→S₀ transition from the LUMO to the HOMO and was calculated for rotamers 1–6. The emission maxima were found at 527 nm (f = 0.5685) for trans rotamer 1, 521 nm (f = 0.8771) for trans rotamer 2, 568nm (f = 0.1271) for cis–trans rotamer 3, 539nm (f = 0.1578) for cis rotamer 4, 536nm (f = 0.2937) for cis rotamer 5, 525nm (f = 0.5939) and for cis rotamer 6 (Table S3). These results confirm that all six rotamers 1–6 contribute to the emission profile, with trans rotamer 2 showing the most intense fluorescence due to the highest oscillator strength.

2.6. Spectroscopic Evaluation of Phenanthroline Derivative 3a Towards Metal Ions

The chemical reactivity of chemical probe 3a towards different ions was measured in a DMSO: H₂O (9:1, v/v) solution using colorimetric and fluorimetric methods. A number of changes were observed in the absorption spectra in the presence of test ions, as shown in Figure 11. Chemical probe 3a shows characteristic absorption peaks at 345 and 386 nm, which are shifted towards longer wavelengths (bathochromic effect) and enhanced absorption intensity (hyperchromic effect) in the presence of Hg²⁺, Ni²⁺, and Ag⁺ ions. Only a bathochromic shift in the absorption maximum is observed for Cd²⁺ and Zn²⁺ ions. However, a shift in the maximum towards shorter wavelengths, combined with a hyperchromic effect, is observed for Fe²⁺ and Pb²⁺ ions. In contrast, Cu²⁺ ions show a bathochromic shift with a decrease in absorption intensity (hypochromic effect).

The bathochromic shift in spectra of complexes of 3a with Hg²⁺, Ni²⁺, Ag⁺, Cd²⁺, Zn²⁺, and Cu²⁺ is due to the stabilization of the LUMO energy level, which reduces the HOMO-LUMO energy gap and thus lowers the energy of electronic transitions. This shift is influenced by the electronic properties of the metal ions, including their charge density, coordination behavior, and ability to participate in π-bonds or accept electron density from a ligand. For example, as shown in Figure 11, the absorption spectra of 3a and its complexes with Li⁺, K⁺, Mg²⁺, and Ca²⁺ ions are nearly identical, i.e., there are no significant shifts in band positions or changes in spectral shape. These experimental data are supported by TDDFT calculations using the B3LYP/6-31G(d,p) method performed for the most stable trans-isomer of 3a with Li⁺, K⁺, Mg²⁺, and Ca²⁺ ions (Figure S6). For the complexes 3a-Li⁺ and 3a-K⁺, the first absorption band, calculated in the 450–420 nm range, corresponds to the S₀→S₁ transition and arises from a single HOMO→LUMO excitation, similar to the free trans-isomer of 3a. In contrast, for the divalent metal complexes 3a-Mg²⁺ and 3a-Ca²⁺, the S₀→S₁ transition exhibits mixed character, arising predominantly from the HOMO→LUMO+1 with an additional HOMO−1→LUMO contribution (Figure 12). This orbital mixing reflects the influence of Mg²⁺ and Ca²⁺ coordination on the electronic structure of 3a, through electrostatic interaction and polarization effects rather than direct involvement of the metal orbitals in the π-π* system.

Similar spectral profiles suggest that the coordination of these metal ions does not substantially change the electronic structure of 3a, which indicates the absence of effective electronic interaction between the Li⁺ or K⁺ metal cations and the π-system of 3a orweak complexation in the case of Mg²⁺ and Ca²⁺ cations. Such behavior is typical of electronically inert ions that form complexes without significant orbital overlap with the ligand (Figure 12). For the 3a-Li⁺, 3a-K⁺, 3a-Mg²⁺ and 3a-Ca²⁺ complexes, the HOMO is localized on the OH-substituted benzene fragments, vinyl groups, and particularly involves phenanthroline moiety. The LUMO for 3a-Li⁺ and 3a-K⁺ complexes and LUMO+1 for 3a-Mg²⁺ and 3a-Ca²⁺ complexes is concentrated on a phenanthroline core (Figure 12). The molecular orbitals (MOs) involved in the electronic transitions of 3a-Li⁺, 3a-K⁺, 3a-Mg²⁺ and 3a-Ca²⁺ show a similar spatial distribution to those of the free trans-isomer of 3a (Figure 5), which explains the similarity and minimal spectral shift observed in their absorption spectra (Figure 11).

Transition metals are well known for their ability to form polynuclear complexes with chelating ligands such as phenanthroline. As suggested in Scheme 2, such coordination can lead to cage-like or extended structures. Although TDDFT calculations of absorption spectra of large polynuclear complexes are resource-intensive, we have simulated representative bidentate (3a)₂-Zn²⁺ and (3a)₂-Cu²⁺ complexes, where coordination involves the two nitrogen atoms from two phenanthroline units (Figure 12). These simplified systems serve to illustrate the influence of the transition metal center on the charge transport and electronic structure of the complex. For the (3a)₂-Zn²⁺ complex, the HOMO is localized on the OH-substituted benzene rings and vinyl groups, while the LUMO is concentrated on phenanthroline core, especially around the nitrogen atoms (Figure 12). In the case of the (3a)₂-Cu²⁺, the chelation mode facilitates metal-to-ligand charge transfer (MLCT) and/or ligand-to-metal charge transfer (LMCT) interactions. These interactions lower the energy of π→π^* transitions, which contributes to the bathochromic shift in the spectrum of the (3a)₂-Cu²⁺ complex (Figure 11). Cu²⁺, being a paramagnetic d⁹ metal center, introduces significant spin–orbit coupling (SOC). This SOC perturbs the electronic structure and promotes non-radiative decay pathways, like intersystem crossing and vibrational relaxation. This results in both a reduction in absorption intensity (hypochromic effect) and effective fluorescence quenching, consistent with the role of Cu²⁺ as a strong fluorescence quencher [42].

The fluorescence response of the derivative 3a toward various metal ions was investigated in solution to evaluate its potential as a selective chemosensor. Figure 13a presents the emission spectra of 3a in the absence and presence of different metal ions (100 μM), whereas the corresponding fluorescence intensities at the emission maximum are summarized in the bar chart (Figure 13b). Of all the tested cations, Cu²⁺ caused the most significant fluorescence quenching. This effect is attributed to strong coordination between Cu²⁺ and ligand 3a, which may involve the nitrogen atoms of the phenanthroline core and the phenolic oxygen atoms. Notably, the Ag⁺ ion also induced significant quenching of fluorescence, although to a lesser extent than Cu²⁺. This may be attributed to Ag⁺ forming coordination interactions with the ligand via nitrogen and/or oxygen donor atoms. Therefore, the response to Ag⁺ should be considered in applications where silver contamination may be relevant. Other transition metal ions, such as Fe²⁺, Hg²⁺, and Ni²⁺, also reduced the emission intensity to a lesser extent, indicating a moderate binding affinity. In contrast, alkali and alkaline earth metals (e.g., K⁺, Ca²⁺, Mg²⁺, and Li⁺) had a negligible influence on fluorescence. These results confirm that compound 3a exhibits high selectivity and sensitivity towards Cu²⁺, consistent with the Irving–Williams series [43], where Cu²⁺ typically forms the most stable complexes with first-row transition metal ions.

In addition, the changes observed in the absorption and emission spectra of compound 3a in the presence of various metal ions are visible to the naked eye (Figure 14A) and under UV illumination at 365 nm (Figure 14B). The results obtained suggest that compound 3a can be used as a new tool for detecting copper ions.

In the presence of the analytes tested, no changes were observed in the absorption spectra of the quinoline 3c and 3d derivatives.

The limits of detection (LOD) and quantification (LOQ) for the selected metal ions (Cu²⁺, Ag⁺, Hg²⁺, and Ni²⁺) were determined based on changes observed in the emission spectra. LOD and LOQ were calculated using the following equations, with the standard deviation of the blank signal (σ) and the slope of the calibration curve (s) substituted in

LOD = 3.3 σ/s → → LOQ = 10 σ/s

The results are summarized in

Table 4. The limit of detection and quantification for derivative 3a with various metal ions.

Metal Ion	LOD [µM]	LOQ [µM]
Cu2+	0.97	2.96
Ag+	2.68	8.12
Hg2+	28.8	87.3
Ni2+	29.8	90.4

, and the corresponding calibration plots of fluorescence intensity versus metal ion concentration are presented in Figure S10. The LOD and LOQ values indicate that derivative 3a exhibits the highest sensitivity toward Cu²⁺ and Ag⁺ ions, as evidenced by their significantly lower detection and quantification limits compared to Hg²⁺ and Ni²⁺. In particular, the low LOD values for Cu²⁺ (0.97 µM) and Ag⁺ (2.68 µM) suggest that derivative 3a is well-suited for the detection of these metal ions at trace levels in solution. In contrast, the markedly higher LOD and LOQ values observed for Hg²⁺ (28.8 µM and 87.3 µM, respectively) and Ni²⁺ (29.8 µM and 90.4 µM) indicate a substantially lower sensitivity toward these ions.

Further studies focused exclusively on Cu²⁺ ions because they caused complete fluorescence quenching of derivative 3a at the lowest concentration of all the tested metal analytes. Furthermore, Cu²⁺ ions yielded the lowest limits of detection (LOD) and quantification (LOQ), establishing them as the most promising analyte.

To determine whether the mechanism of fluorescence quenching of derivative 3a by Cu²⁺ ions is dynamic (collision) or static (complex formation), studies of the changes in fluorescence intensity of derivative 3a towards Cu²⁺ ions at different concentrations were carried out (Figure 15a). A Stern–Volmer plot was made from the data obtained, which was linear in the concentration range 0.2 to 1.8 µM and was described by this equation:

F₀/F = 1 + K_SV [Cu²⁺]

where F₀ and F are the fluorescence intensities in the absence and presence of Cu²⁺, respectively, and [Cu²⁺] is the quencher concentration. From the slope of the linear region, the Stern–Volmer constant was determined to be K_SV = 6.0 × 10¹¹ M⁻¹ (Figure 15b).

At higher concentrations of Cu²⁺, a deviation from linearity was observed, with the plot exhibiting an exponential increase (see inset in Figure 15b), suggesting a change in the quenching mechanism. Such a deviation is characteristic of static quenching, likely caused by the formation of a ground-state non-fluorescent complex between 3a and Cu²⁺. This hypothesis is further supported by changes in the absorption spectrum (Figure 11), which are commonly associated with complex formation in static quenching [44].

To further evaluate the nature of the quenching, the fluorescence lifetime (τ) of the 3a derivative was considered. Using the Stern–Volmer constant and the known fluorescence lifetime of 3a (τ = 0.99 ns), the bimolecular quenching rate constant k_q was calculated as follows:

k_q = K_SV/τ = 6.06 × 10²⁰ M⁻¹s⁻¹

This value greatly exceeds the diffusion controlled quenching rate in aqueous media (typically k_diff ≈ 10⁹–10¹⁰ M⁻¹s⁻¹), clearly ruling out a purely dynamic (collisional) quenching mechanism. Therefore, the quenching observed for the 3a-Cu²⁺ system is best explained by a static quenching mechanism involving the formation of a non-emissive ground-state complex between 3a and Cu²⁺. The formation of this complex is facilitated by photoinduced electron transfer (PET) or ligand-to-metal charge transfer (LMCT) processes, wherein electron density from the donor-rich ligand system of 3a is transferred to the redox-active Cu²⁺ ion, effectively quenching fluorescence.

The modified logarithmic Stern–Volmer plot was used to calculate the complex formation constant between derivative 3a and Cu²⁺ ions (Figure S7). The obtained value of the constant was 9.73 × 10⁸ M⁻¹, indicating the high affinity of the chemosensor for copper ions and the stability of the formed complex.

The saturation method was used to determine the stoichiometry of the complex by analyzing the dependence of the fluorescence intensity on the Cu²⁺ concentration (Figure S8). The maximum fluorescence quenching was observed at a Cu²⁺ mole fraction of 0.4, which corresponds to a molar ratio of 1:2 (3a:Cu²⁺). This result suggests that one ligand 3a binds two copper ions, leading to the formation of a complex with a stoichiometry of 1:2.

Based on the literature reports for polynuclear phenanthroline-metal complexes [45], we propose a cage-like coordination structure (Scheme 2), in which both the nitrogen atoms of the phenanthroline core and the hydroxyl oxygen atoms act as donor sites. The two Cu²⁺ ions are stabilized through bridging interactions, resulting in a rigid architecture that efficiently quenches fluorescence via static quenching.

3. Materials and Methods

3.1. Materials

All experiments were carried out in an atmosphere of dry argon, and flasks were flame-dried. Solvents were dried by usual methods (diphenyl ether, diethyl ether, and THF over benzophenone ketyl, CHCl₃, and CH₂Cl₂ over P₄O₁₀, hexane, and pyridine over sodium-potassium alloy) and distilled. Chromatographic purification was carried out on silica gel 60 (0.15–0.3 mm, Macherey-Nagel GmbH & Co. KG, Dueren, Germany). 2,9-Dimethyl-1,10-phenanthroline (neocuproine), 2-hydroxybenzaldehyde, 2,4-dihydroxybenzaldehyde, and 3,4,5-trihydroxybenzaldehyde were purchased from Sigma–Aldrich (Poznań, Poland), and were used without further purification.

3.2. Instrumentation

NMR spectra were obtained with Avance 400 and 500 spectrometers (Bruker, Billerica, MA, USA) operating at 500.2 or 400.2 MHz (¹H), 125.8 or 100.6 MHz (¹³C), and 470.5 MHz (¹⁹F) at 21 °C. Chemical shifts referenced to ext. TMS (tetramethylsilane) (¹H, ¹³C), CFCl₃ (¹⁹F) or using the residual CHCl₃ signal (δ_H 7.26 ppm), and CDCl₃ (δ_C 77.1 ppm) as internal references, and ext. DSS for ¹H and ¹³C-NMR, respectively. Coupling constants are given in Hz. The LCMS-IT-TOF analysis was performed on an Agilent 1200 Series binary LC system coupled to a micrOTOF-Q system mass spectrometer (BrukerDaltonics, Bremen, Germany). High-resolution mass spectrometry (HRMS) measurements were performed using a Synapt G2-Si mass spectrometer (Waters, New Castle, DE, USA) equipped with an ESI source and quadrupole-time-of-flight mass analyzer. To ensure accurate mass measurements, data were collected in centroid mode, and mass was corrected during acquisition using leucine enkephalin solution as an external reference (Lock-Spray) (Waters, New Castle, DE, USA). The measurement results were processed using the MassLynx4.1 software (Waters, Milford, MA, USA) incorporated within the instrument. A Nicolet iS50 FTIR spectrometer was used to record spectra in the IR range of 4000–400 cm⁻¹. FTIR spectra were recorded on a Perkin Elmer (Schwerzenbach, Switzerland) spectrophotometer in the spectral range of 4000–450 cm⁻¹ with the samples in the form of KBr pellets. Elementary analysis was performed using a Vario EL III apparatus (Elementar, Langenselbold, Germany). Differential scanning calorimetry (DSC) measurements were performed using a Q2000 calorimeter (TA Instruments, New Castle, DE, USA) in a nitrogen stream at a scanning rate of 10 °C/min. Samples were analyzed in aluminum pans in the temperature range of 50 to 350 °C. Melting points were determined on an MPA100 OptiMelt melting point apparatus (Stanford Research Systems, Sunnyvale, CA, USA) and were uncorrected. Absorption spectra were recorded using a JASCO V-670 spectrophotometer (Jasco, Japan). The determination of the molar extinction coefficient was verified by linear least squares fitting of the values obtained from five independent solutions at different concentrations ranging from 2 µmol/L to 20 µmol/L. Fluorescence spectra were measured on an FLS-920 spectrofluorometer (Edinburgh Instruments, Livingston, UK) in a 10 mm quartz cuvette. The slit width was 1.0 nm. All samples for steady-state measurements were excited in optical dilution. Absorbance values at the selected excitation wavelengths were less than or equal to 0.1.

3.3. Photophysical Measurements

The stock solutionof chemosensor3a was prepared in DMSO at a concentration of 1 mM. All cation solutions were prepared in distilled water at a concentration of 10 mM. All spectroscopic experiments were carried out in DMSO:H₂O (9:1, v/v). The concentration of probe 3a was 10 µM in all spectroscopic studies. An excitation wavelength of 388 nm was used to record the fluorescence emission spectra. UV-Vis absorbance and fluorimetric experiments were performed on chemosensor3a in the presence of different ions. The concentrations were 1 mM for Hg²⁺, Co²⁺, Ni²⁺, Ca²⁺, Cd²⁺, Mn²⁺, Mg²⁺, Zn²⁺, Li⁺, K⁺, Ag⁺ and 0.1 mM for Cu²⁺, Fe²⁺, Pb²⁺. In addition, the optical properties of 3a were investigated in the presence of 0 to 10 µM Cu²⁺ to evaluate the performance of the chemical sensor.

3.4. Photochemical Experiments

Fluorescence quantum yields of 3a were calculated using quinine sulfate (in 0.1 M H₂SO₄ solution) as a standard (Φ_ref = 0.54) [46] quantum yield was calculated accordingly to Equation (1), in which Φ_ref is the absolute quantum yield of the reference, η is the refractive index of the solvent, and Grad is the gradient obtained from the plot of the integrated fluorescence intensity as a function of the absorbance. The subscripts s and ref refer to the sample and reference, respectively.

$${\Phi }_f={\Phi }_{ref}\left({\displaystyle \frac{{Grad}_s}{Gra{d}_{ref}}}\right)\times \left({\displaystyle \frac{{\eta }_s^2}{{\eta }_{ref}^2}}\right)$$

(1)

Fluorescence lifetime measurements were made using a time-correlated single photon counting system (TCSPC) with a pulsed picosecond LED (EPLED-380) as the excitation source. Collecting scattered light from a Ludox silica suspension measured the instrument response function. Fluorescence lifetimes were calculated from intensity decay analyses. Fluorescence decay was acquired at a level of 1 × 10⁴ counts at the peak and was fitted to a sum of exponentials:

$$I(t)=\sum _{i=1}^n{\alpha }_i\mathit{exp}(-t/{\tau }_i)$$

(2)

with amplitudes, α_i, and decay lifetimes τ_i. Average lifetimes (τ) for multiexponential fluorescence decay were calculated from decay times and pre-exponential factors using the following:

$$(\tau )={\displaystyle \frac{\sum {\alpha }_i{\tau }_i^2}{\sum {\alpha }_i{\tau }_i}}$$

(3)

3.5. X-Ray Diffraction Experiments

The colorless crystals of 2a were mounted on a SuperNova 4-circle X-ray diffractometer equipped with an Atlas CCD detector and used for data collection. X-ray intensity data were collected with mirror monochromated CuKα radiation (λ = 1.54184 Å) at a temperature of 150(1) K with ω scan mode.

Controlling the measurement procedure and data reduction were performed using the CrysAlis^Pro software Version 1.171.38.41q [47]. The same program was used to determine and refine the lattice parameters.

The structure was solved by direct methods and subsequently completed by the difference Fourier methods. All the non-hydrogen atoms were refined using a full-matrix, least-squares technique with anisotropic displacement parameters. The hydrogen atoms were treated as riding on their parent carbon atoms with isotropic displacement parameters (equal to 1.2 or 1.5 times the value of carbon displacement parameters). The SHELXS-2013 and SHELXL-2019/2 [48] programs were used for all the calculations.

One of the carbonyl oxygens from the carboxylic group is disordered over two positions in 0.796(6):0.204(6) ratio.

Details concerning crystal data and refinement are gathered in Table 1.

CCDC 2355007 contains the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/structures (or from the CCDC, 12 Union Road, Cambridge CB2 1EZ, UK; Fax: +44 1223 336033; E-mail: deposit@ccdc.cam.ac.uk).

3.6. Quantum Chemical Computations

The molecular structures of the 3a, 3b, 3c, and 3d in the ground singlet state (S₀) were optimized using density functional theory (DFT), and the B3LYP/6-311++G(d,p) [49,50,51] with Grimme’s D3 dispersion correction approach [52,53]. We have also tested the range-separated hybrid cam-B3LYP-GD3 functional with Grimme’s D3 dispersion correction, long-range corrected hybrid density functional wB97XD [54], which includes empirical dispersion corrections (D2), as well as the highly parametrized hybrid meta-GGA functional M06-2X [55], which incorporates 54% Hartree–Fock (HF) exchange with the 6-311++G(d,p) basis set (Figure S3). The minimal structural changes (0.010–0.015Å) found during testing of these functionals confirm that the B3LYP/6-311++G(d,p) level of theory is adequate to describe the key structural features of our systems.

Transition states (TS) for 3b, 3c, and 3d were found using the QST3 [56] and Berny [57] optimization methods realized in Gaussian 16 [58]. The presence of one imaginary frequency in the TS indicates that the structure corresponds to a saddle point on the potential energy surface. This imaginary frequency represents the bond rotation of the C2-C3 and C3-C4 bonds responsible for the cis–trans isomerization process. For the TSs, intrinsic reaction coordinate (IRC) analysis was further performed to ensure that the obtained TS connects the minima. The TSs were calculated using the same B3LYP/6-311++G(d,p) [49,50,51] with Grimme’s D3 dispersion correction approach [52,53].

The first excited singlet (S₁) state and the UV-visible absorption spectra of 3a, 3c, and 3d were calculated using the time-dependent (TD) DFT formalism with the polarizable continuum model (PCM), employing TDDFT/B3LYP/6-31G(d,p) level of theory [59,60]. For the UV-visible absorption spectra of 3a, 3c, and 3d, the lowest 20 vertical electronic transitions were computed. We tested the effect of diffuse functions on absorption spectra using the trans-isomer 3a as an example. The absorption spectrum for the trans rotamer of 3a was calculated with B3LYP functional using both the 6-31G(d,p) and 6- 31+G(d,p) basis sets (Figure S4). Both spectra have the same overall shape. Small shifts (~2–7 nm, red shift, Figure S4) were found, which are within the typical range of basis set effects when adding diffuse functions. This demonstrates that the diffuse functions only introduce minor corrections to the calculated electronic excitation energies. We have also calculated the optimized geometry of the trans rotamer of 3a, both in the gas phase and with DMSO solvent effects included (Figure S5). The comparison shows only negligible changes in bond lengths (0.001Å, Figure S5), which supports the decision to perform geometry optimizations in the gas phase.

To analyze the intermolecular interactions in 2a, Hirshfeld surfaces were calculated using Crystal Explorer 21.5 [61]. The procedure follows the approach detailed in Refs [62,63].

3.7. Synthesis of Vinyl Derivatives of 1,10-Phenanthroline (Neocuproine) 1a

Neocuproine 1a (1.04 g, 5.0 mmol) was dissolved in Ac₂O (100 mL), followed by the addition of 2,4-dihydroxybenzaldehyde (2.76 g, 20.0 mmol). The reaction was heated under reflux for 24 h. Then, the volatiles were removed in vacuo (16 mmHg). The crude product was purified by crystallization from a mixture of THF/Et₂O and dried over P₄O₁₀ to yield precipitates as follows:

((1E,1′E)-(1,10-phenanthroline-2,9-diyl)bis(ethene-2,1-diyl))bis(benzene-4,1,3-triyl) tetraacetate (2a) beige; DSC m.p. = 188.54 °C; 3.0 g (4.8 mmol, 96%), ¹H-NMR (DMSO-d6; 500.2 MHz) δ = 2.31 (s, 6H, CH₃), 2.45 (s, 6H, CH₃), 7.15 (d, ³J_H,H = 2.3 Hz, 2H, aromatic), 7.20 (dd, ³J_H,H = 8.5 Hz, ⁴J_H,H = 2.4 Hz, 2H, aromatic), 7.65 (d, ³J_H,H = 16.3 Hz, 2H, vinyl), 7.95 (s, 2H, aromatic), 7.95 (d, ³J_H,H = 16.3 Hz, 2H, vinyl), 8.06 (d, ³J_H,H = 8.7 Hz, 2H, aromatic), 8.12 (d, ³J_H,H = 8.4 Hz, 2H, aromatic), 8.48 (d, ³J_H,H = 8.3 Hz, 2H, aromatic); ¹³C{¹H}-NMR (DMSO-d6; 125.8 MHz) δ = 20.8, 20.9, 117.1, 120.2, 121.7, 125.9, 126.2, 126.6, 127.2, 128.0, 131.2, 136.9, 145.2, 148.8, 150.8, 154.7, 168.9, 169.1; HRMS (ESI TOF): m/z Calcd for C₃₆H₂₉N₂O₈ (M + H)⁺ = 617.1924, Found 617.1921. CCDC 2355007.

2a (3.08 g, 5.0 mmol) was added to a water solution of hydrochloric acid (10%) and stirred under further reflux for 16 h. Next, a water solution of sodium hydroxide (10%) was added, and the red precipitate was filtered off and purified by crystallization from methanol and dried over P₄O₁₀ to yield precipitates as follows:

4,4′-((1E,1′E)-(1,10-Phenanthroline-2,9-diyl)bis(ethene-2,1-diyl))bis(benzene-1,3-diol) (3a) red 1.9 g (4.6 mmol, 91.0%), DSC m.p. = 264.26 °C; ¹H-NMR (DMSO-d6/KOD; 500.2 MHz) δ = 6.31 (d, ³J_H,H = 8.4 Hz, 2H, aromatic), 6.36 (d, ³J_H,H = 2.4 Hz, 2H, aromatic), 7.42 (d, ³J_H,H = 16.4 Hz, 2H, vinyl), 7.50 (d, ³J_H,H = 8.4 Hz, 2H, aromatic), 7.78 (s, 2H, aromatic), 7.92 (d, ³J_H,H = 8.5 Hz, 2H, aromatic), 8.02 (d, ³J_H,H = 16.4 Hz, 2H, vinyl), 8.32 (d, ³J_H,H = 8.3 Hz, 2H, aromatic); ¹³C{¹H}-NMR (DMSO-d6/KOD; 125.8 MHz) δ = 103.4, 108.2, 115.5, 121.1, 125.4, 126.1, 127.8, 129.3, 130.8, 137.3, 145.6, 157.2, 158.0, 160.0; HRMS (ESI TOF): m/z Calcd for C₂₈H₂₁N₂O₄ (M + H)⁺ = 449.1501, Found 449.1496.

3.8. Synthesis of Vinyl Derivatives of Quinoline

The synthesis of vinyl quinolines followed our procedure described in the literature [24,64,65]:

2-Methyl-8-(trifluoromethyl)quinoline (1b) or 8-isopropyl-2-methyl-5-nitroquinoline (1c) (10.0 mmol) was partially dissolved in Ac₂O (100 mL), followed by the addition of 2-hydroxybenzaldehyde, 2,4-dihydroxybenzaldehyde or 3,4,5-trihydroxybenzaldehyde (20.0 mmol). The reaction was heated under reflux for 16 h. Then, the volatiles were evaporated from the resulting solution, and the residue was acidified by a water solution of hydrochloric acid (10%), and the reaction was stirred for 16 h, followed by the addition of CHCl₃, and was filtered off. The crude product was dissolved in a water solution of NaOH (10%), and next was neutralized with an aqueous solution of hydrochloric acid (10%), and the solid was filtered. The crude product was purified by crystallization from methanol and dried over P₄O₁₀ to yield residues as follows:

(E)-2-(2-(8-(Trifluoromethyl)quinolin-2-yl)vinyl)phenol (3b) Yield: 2.14 g (6.8 mmol, 68.0%); DSC = 276.65 °C; ¹H-NMR (DMSO–d6; 400.2 MHz) δ = 6.86 (td, ³J_H,H = 7.5 Hz, ⁴J_H,H = 1.2 Hz, 1H, aromatic), 7.00 (dd, ³J_H,H = 8.1 Hz, 4JH,H = 1.2 Hz, 1H, aromatic), 7.19 (ddd, ³J_H,H = 8.1 Hz, ³J_H,H = 7.2 Hz, ⁴J_H,H = 1.7 Hz, 1H, aromatic), 7.49 (d, ³J_H,H = 16.3 Hz, 1H, vinyl), 7.60 (t, ³J_H,H = 7.7 Hz, 1H, aromatic), 7.66 (d, ³J_H,H = 7.8 Hz, ⁴J_H,H = 1.6 Hz, 1H, aromatic), 7.89 (d, ³J_H,H = 8.6 Hz, 1H, aromatic), 8.10 (d, ³J_H,H = 7.4 Hz, 1H, aromatic), 8.16 (d, ³J_H,H = 7.6 Hz, 1H, aromatic), 8.17 (d, ³J_H,H = 16.1 Hz, 1H, vinyl), 8.41 (d, ³J_H,H = 8.7 Hz, 1H, aromatic), 10.12 (s, 1H, OH); ¹³C{¹H}-NMR (DMSO–d6; 100.6 MHz) δ = 116.1, 119.5, 121.1, 122.9, 123.3, 124.6, 125.6 (q, ¹J_C,F = 28.6 Hz CF₃), 127.3, 127.7, 127.8, 128.3 (q, ²J_C,F = 5.5 Hz CF₃), 130.1, 131.4, 132.8, 137.0, 144.0, 156.1, 157.0; ¹⁹F{¹H}-NMR (DMSO–d6; 470.55 MHz) δ = −58.48; ¹⁹F-NMR (DMSO–d6; 470.55 MHz) δ = −58.48; LCMS-IT-TOF [M+H]⁺ = 316.0950 (100%), HRMS (IT TOF): m/z Calcd for C₁₈H₁₃F₃NO [M+H]⁺: 316.0949, Found 316.0950.

(E)-4-(2-(8-Isopropyl-5-nitroquinolin-2-yl)vinyl)benzene-1,3-diol (3c) red Yield: 2.75 g (7.9 mmol, 78.6%); DSC = 213.12 °C; ¹H-NMR (DMSO–d6; 500.2 MHz) δ = 1.37 (d, ³J_H,H = 6.9 Hz, 6H, 2CH₃), 4.40 (hept, ³J_H,H = 6.9 Hz, 1H, CH), 6.70 (s, 2H, aromatic), 7.13 (d, ³J_H,H = 16.2 Hz, 1H, vinyl), 7.69 (d, ³J_H,H = 16.1 Hz, 1H, vinyl), 7.78 (d, ³J_H,H = 8.1 Hz, 1H, aromatic), 8.08 (d, ³J_H,H = 9.2 Hz, 1H, aromatic), 8.30 (d, ³J_H,H = 8.1 Hz, 1H, aromatic), 8.78 (d, ³J_H,H = 9.1 Hz, 1H, aromatic), 9.61 (bs, 1H, OH); ¹³C{¹H}-NMR (DMSO–d6; 100.6 MHz) δ = 23.0, 27.6, 102.7, 107.7, 114.4, 118.8, 122.3, 123.4, 123.7, 124.3, 129.2, 132.0, 132.2, 143.4, 144.8, 153.9, 156.7, 157.7, 159.9; LCMS-IT-TOF [M+H]⁺ = 351.1340 (100%), HRMS (IT TOF): m/z Calcd for C₂₀H₁₉N₂O₄ [M+H]⁺: 351.1345, Found 351.1340.

(E)-5-(2-(8-isopropyl-5-nitroquinolin-2-yl)vinyl)benzene-1,2,3-triol (3d) black Yield: 0.9 g (24.6%); DSC = 213.12 °C; ¹H-NMR (DMSO–d6; 500.2 MHz) δ = 1.38 (d, ³J_H,H = 7.0 Hz, 6H, 2CH₃), 4.40 (hept, ³J_H,H = 6.9 Hz, 1H, CH), 6.69 (s, 2H, aromatic), 7.12 (d, ³J_H,H = 16.2 Hz, 1H, vinyl), 7.68 (d, ³J_H,H = 16.3 Hz, 1H, vinyl), 7.78 (d, ³J_H,H = 8.1 Hz, 1H, aromatic), 8.07 (d, ³J_H,H = 9.2 Hz, 1H, aromatic), 8.30 (d, ³J_H,H = 8.1 Hz, 1H, aromatic), 8.78 (d, ³J_H,H = 9.1 Hz, 1H, aromatic), 9.61 (bs, 1H, OH); ¹³C{¹H}-NMR (DMSO–d6; 125.8 MHz) δ = 23.01, 27.7, 106.9, 118.9, 122.3, 123.6, 124.4, 124.5, 126.4, 132.1, 135.2, 136.9, 143.4, 144.8, 146.3, 154.0, 156.0; LCMS-IT-TOF [M-H]⁻ = 365.1142 (100%), HRMS (IT TOF): m/z Calcd for C₂₀H₁₇N₂O₅ [M-H]⁻: 365.1137, Found 365.1142.

4. Conclusions

Biological activity is believed to arise from the metal-chelating properties of the compounds toward various metal cations. This study tested the ability to chelate metal ions using selected water-soluble ligands based on the styryl derivatives of quinoline and 1,10-phenanthroline. These ligands were synthesized via standard Perkin condensation. Hydrophilic properties were introduced by incorporating phenol or polyphenol groups. The proton transfer reactions between the pyridine and phenol functional groups, or the generally simple protonation/deprotonation of pyridine rings in the 1,10-phenanthroline constitution, could be used as a pH indicator. The absorption spectra of compounds 3c and 3d at a concentration of 20 µM in various solvents confirmed that their spectral properties are influenced by basic media. For example, compound 3d decomposed in CD₃OD/KOD under aerobic conditions, due to oxidation.

The studied compounds were further characterized using analytical methods and rationalized based on quantum chemical DFT and TDDFT calculations using the B3LYP functional. The TDDFT/B3LYP/6-31G(d,p) calculations performed in DMSO confirm the experimental absorption and emission behavior of compounds 3a, 3c, and 3d. For 3c and 3d, both trans and cis rotamers show absorption maxima in the near-UV region (364–366 nm), with the dominant transition being HOMO→LUMO+1. A weak, longer-wavelength absorption band corresponds to a low-intensity HOMO→LUMO transition (S₀→S₁). In the case of 3a, the absorption and fluorescence properties are more complicated due to the presence of six rotamers. The observed weak absorption in the 520–430 nm region arises primarily from S₀→S₁ transitions in rotamers 1–3, especially trans rotamer 2, which shows the strongest oscillator strength (f = 0.5993). The emission of 3a originates from the S₀→S₁ transition (LUMO→HOMO) and is contributed by all six rotamers. Trans rotamer 2 exhibits the most intense fluorescence (λ_em = 521 nm, f = 0.8771), making it the dominant species in the emission spectrum. Overall, the DFT and TDDFT calculations highlight the role of molecular conformation (rotamerism) in tuning the photophysical properties of 3a–3d and explain the complex nature of their UV-Vis and fluorescence spectra. The strong correlation between computed and experimental results validates the use of TDDFT for designing and understanding such metal-sensing systems.

((1E,1′E)-(1,10-phenanthroline-2,9-diyl)bis(ethene-2,1-diyl))bis(benzene-4,1,3-triyl)tetraacetate (2a) was structurally characterized by the single-crystal X-ray diffraction method. The X-ray crystal structure analysis of 2a showed the presence of the weak π···π stacking interactions between phenanthroline cores placed one above the other in neighboring molecules of 2a. The ¹H and ¹³C NMR spectra of the presented compounds displayed distinctively diagnostic signals from the vinyl (styryl) functional group. The analysis of the value of coupling constants for vinyl protons confirmed the sole presence of the E conformer.

In this work, we also developed a tandem UV-Vis and fluorescence spectroscopy method to detect metal ions. Compound 3a, in DMSO:H₂O (9:1, v/v), formed complexes with various metal cations, enabling their detection using colorimetric and fluorometric responses. This approach allowed sensitive detection of Hg²⁺, Ni²⁺, Cu²⁺, and Ag⁺ ions, with detection limits down to 1 mM, and 0.1 mM for Cu²⁺.