Synthesis, Structure, and Properties of a Copper(II) Binuclear Complex Based on Trifluoromethyl Containing Bis(pyrazolyl)hydrazone

A new complex of copper(II) with methyl-5-(trifluoromethyl)pyrazol-3-yl-ketazine (H2L) was synthesized with the composition [Cu2L2]∙C2H5OH (1). Recrystallization of the sample from DMSO yielded a single crystal of the composition [Cu2L2((CH3)2SO)] (2). The coordination compounds were studied by single-crystal X-ray diffraction analysis, IR spectroscopy, and static magnetic susceptibility method. The data obtained indicate that the polydentate ligand is coordinated by both acyclic nitrogen and heterocyclic nitrogen atoms. The cytotoxic activity of the ligand and complex 1 was investigated on human cell lines MCF7 (breast adenocarcinoma), Hep2 (laryngeal carcinoma), A549 (lung carcinoma), HepG2 (hepatocellular carcinoma), and MRC5 (non-tumor lung fibroblasts). The complex was shown to have a pronounced dose-dependent cytotoxicity towards these cell lines with LC50 values in the range of 0.18–4.03 μM.

The functionalization of the pyrazole core with fluorinated substituents has led to the development of many bioactive molecules, as shown in Figure 1 [13,[15][16][17][18].Currently, almost a quarter of all drugs in the pharmacological market contain fluorine in their composition.Owing to the high electronegativity, small size, and low polarizability of fluorine, its introduction into the structure of organic compounds leads to an increase in their lipophilicity, membrane permeability, and metabolic stability, which determines pronounced biological activity, including anticancer [19].The use of pyrazoles in the synthesis of metal complexes represents a prom avenue in the design of novel antitumor agents [20].This strategy enables the com tion of pyrazole bioactivity with their coordination properties as N-ligands [21 Complexes based on the biocompatible copper(II) ion represent an alternative to exi anticancer metallodrugs [20].Depending on the nature of substituents and reaction ditions, pyrazoles behave both as mono-and bidentate ligands, forming complexes copper ions of various oxidation states and nuclearity [20][21][22][23][24][25][26][27][28][29].In particular, the pre of trifluoromethyl groups in the pyrazole ring leads to an increase in the N-H acid the heterocycle and the stability of complexes with transition metals [13,30].In add copper(II) pyrazolates are of interest because of their magnetic, catalytic, and p physical properties [31][32][33][34][35].
(Figure 2).The structure of H2L contains six nitrogen-donor atoms capable of coordination with transition metal ions.Although the potential of pyrazole derivatives in coordination chemistry is well documented, the number of fluorinated bis-pyrazoles used as ligands remains low [13].

General Procedure for Synthesis of Compounds
Complex 1 was synthesized by varying the metal: ligand ratios (2:1, 1:1, 1:2, 1:3) and using solutions of different copper(II) salts-nitrate, chloride, and sulfate.In all cases, the same dark green color complex was isolated.Elemental analysis and IR spectra of the complexes obtained with different copper(II) salts showed that they have the same composition [Cu2L2]•C2H5OH (Scheme 1).
Upon recrystallization of portion 1 from DMSO and prolonged standing of the solution, we were able to obtain a single crystal of the complex [Cu2L2((CH3)2SO)] (2) suitable for X-ray diffraction analysis (Scheme 2).

General Procedure for Synthesis of Compounds
Complex 1 was synthesized by varying the metal: ligand ratios (2:1, 1:1, 1:2, 1:3) and using solutions of different copper(II) salts-nitrate, chloride, and sulfate.In all cases, the same dark green color complex was isolated.Elemental analysis and IR spectra of the complexes obtained with different copper(II) salts showed that they have the same composition [Cu 2 L 2 ]•C 2 H 5 OH (Scheme 1).
Int. J. Mol.Sci.2024, 25, x FOR PEER REVIEW 3 of 16 (Figure 2).The structure of H2L contains six nitrogen-donor atoms capable of coordination with transition metal ions.Although the potential of pyrazole derivatives in coordination chemistry is well documented, the number of fluorinated bis-pyrazoles used as ligands remains low [13].

General Procedure for Synthesis of Compounds
Complex 1 was synthesized by varying the metal: ligand ratios (2:1, 1:1, 1:2, 1:3) and using solutions of different copper(II) salts-nitrate, chloride, and sulfate.In all cases, the same dark green color complex was isolated.Elemental analysis and IR spectra of the complexes obtained with different copper(II) salts showed that they have the same composition [Cu2L2]•C2H5OH (Scheme 1).
Upon recrystallization of portion 1 from DMSO and prolonged standing of the solution, we were able to obtain a single crystal of the complex [Cu2L2((CH3)2SO)] (2) suitable for X-ray diffraction analysis (Scheme 2).Upon recrystallization of portion 1 from DMSO and prolonged standing of the solution, we were able to obtain a single crystal of the complex [Cu 2 L 2 ((CH 3 ) 2 SO)] (2) suitable for X-ray diffraction analysis (Scheme 2).

Crystallography
X-ray phase analysis data indicate the crystallinity of powdered sample 1.However, we were unable to grow single crystals of 1 suitable for analysis.To determine the coordination ability of the new ligand, we obtained and studied a single crystal of complex 2. It should be noted that the comparison of diffractograms indicates that (despite the same Scheme 2. Synthesis of complex 2.

Crystallography
X-ray phase analysis data indicate the crystallinity of powdered sample 1.However, we were unable to grow single crystals of 1 suitable for analysis.To determine the coordination ability of the new ligand, we obtained and studied a single crystal of complex 2. It should be noted that the comparison of diffractograms indicates that (despite the same Cu:ligand ratio) complexes 1 and 2 are not isostructural (Figure S1).The differences should be explained, among other things, by their different composition and packaging of molecules.
Single-crystal X-ray diffraction analysis of complex 2 indicates that the [Cu 2 L 2 ((CH 3 ) 2 SO)] phase crystallizes in a triclinic crystal system.The crystal packing is molecular (Figures 3 and 4    The corrugated pseudo-layers of molecular binuclear complex particles in the bc plane, distorted according to the elongated shape, could be isolated in the package (Figure 4).Further packing of the layers takes place one above the other without displacement, according to the corrugation.
It should be noted that the coordination ability of the fluorinated ligand differs significantly from that assumed for the non-fluorinated analog [41] (Scheme 3).It should be noted that the coordination ability of the fluorinated ligand differs significantly from that assumed for the non-fluorinated analog [41] (Scheme 3).Scheme 3. The expected type of cage metal complexes for 3D metals with non-fluorinated ligands [41].It should be noted that the coordination ability of the fluorinated ligand differs significantly from that assumed for the non-fluorinated analog [41] (Scheme 3).Scheme 3. The expected type of cage metal complexes for 3D metals with non-fluorinated ligands [41].
Scheme 3. The expected type of cage metal complexes for 3D metals with non-fluorinated ligands [41].

Spectroscopy
In the high-frequency region of the infrared spectra of the ligand (Figure S2), (N-H) vibration bands near 3250 cm −1 are observed.In the spectrum of complex 1 (Figure S3), (O-H) bands are present within the broad range of 3500-3000 cm −1 .The (C-H) bands of methyl groups and pyrazole rings are in the interval 3090-2830 cm −1 .For H 2 L, at 1630 cm −1 , a band of valence-deformation vibration of the acyclic bond -C=N-is observed, which is very sensitive to coordination.In the spectrum of the complex, this band appears at 1608 cm −1 .In the spectrum of H 2 L in the region of 1550-1500 cm −1 , the bands of valence-deformation vibrations of pyrazole rings are observed, which are split or shifted by ~30 cm −1 in the spectrum of the complex (Table 2).The changes in the position and shape of the bands are caused by the coordination of nitrogen atoms to Cu(II) [42].

Magnetic Measurements
The temperature and field dependences of the magnetic susceptibility were measured for a polycrystalline sample of complex 1 and corrected for temperature-independent Langevin diamagnetism to single out the paramagnetic component of the susceptibility χ p (T) associated with copper ions.The resulting χ p (T) dependences were insensitive to the thermo-magnetic prehistory and exhibited a broad peak at T m ∼ = 34 K, independent of the magnitude of the applied magnetic field (Figure 5a).The drop in magnetic susceptibility upon cooling below Tm is unambiguous evidence of the predominance of antiferromagnetic (AF) interactions between copper ions, while the absence of phase transition and the smoothness of χ p change indicate a reduced dimensionality of the magnetic subsystem, reduced to the level of chains or polynuclear clusters.
In the high-temperature region, the obtained χ p (T) data can be formally approximated by the Curie-Weiss dependence χ p (T) = N A µ 2 eff /3k B (T − θ) with values of the effective momentum µ eff ≈ 1.80 µ B and the Weiss constant θ ≈ −26 K (Figure 5b).The µ eff value is characteristic of copper ions Cu 2+ (S = 1/2) and corresponds to an average value of the g-factor g ≈ 2.08, exceeding the spin-only value g ≈ 2 due to the contribution of orbital moments.In turn, in the region of the lowest temperatures, after passing the peak, a susceptibility increase is observed, typical for most samples of chain and polynuclear complexes and associated with the inevitable presence of a small number of mononuclear impurities.To determine the percentage of monomer impurities, their contribution to the temperature dependence of the susceptibility was approximated by the Curie-Weiss dependence and the contribution to the field dependence of the magnetization M(H) was approximated by the Brillouin function.The analysis showed that the monomer fraction includes ~1.2% of copper ions.The magnetic susceptibility of the main phase can be determined by subtracting the contribution of mononuclear impurities from the experimental data.As can be seen in Figure 6, the magnetic susceptibility of the main phase of complex 1 after passing the peak at Tm decreases almost to zero, which clearly indicates the formation of a gap in the spectrum of spin excitations.The most obvious explanation for this is related to the presence of magnetic dimers in the structure of the complex described by the Hamiltonian  = − ⃗ ⋅  ⃗ [43], where J is the exchange interaction between Cu 2+ ions inside the dimer (J < 0 for the AF interaction), which is consistent with the structural data for the related complex 2. However, the model of isolated dimers, in which the interaction between dimers J′ = αJ is assumed to be zero, does not allow us to achieve any acceptable quality description of the experimental data χp(T) (Figure 6).The position of the magnetic susceptibility peak at Tm ~ 34 K in the isolated dimer model corresponds to the exchange interaction within the dimers J/kB ≈ 50 K; however, when such a value of J is chosen, the model calculation significantly exceeds the experimental values of χp (dashed-double dotted curve in Figure 6).In turn, the behavior of the magnetic susceptibility within the range of 80-300 K in the same model requires a value of J/kB ≈ 70 K (dashed-dotted line).This can be explained by the presence of an additional AF interaction between the dimers, J′ = αJ ≠ 0. As can be seen in Figure 6, the magnetic susceptibility of the main phase of complex 1 after passing the peak at T m decreases almost to zero, which clearly indicates the formation of a gap in the spectrum of spin excitations.The most obvious explanation for this is related to the presence of magnetic dimers in the structure of the complex described by the [43], where J is the exchange interaction between Cu 2+ ions inside the dimer (J < 0 for the AF interaction), which is consistent with the structural data for the related complex 2. However, the model of isolated dimers, in which the interaction between dimers J ′ = αJ is assumed to be zero, does not allow us to achieve any acceptable quality description of the experimental data χ p (T) (Figure 6).The position of the magnetic susceptibility peak at T m ~34 K in the isolated dimer model corresponds to the exchange interaction within the dimers J/k B ≈ 50 K; however, when such a value of J is chosen, the model calculation significantly exceeds the experimental values of χ p (dashed-double dotted curve in Figure 6).In turn, the behavior of the magnetic susceptibility within the range of 80-300 K in the same model requires a value of J/k B ≈ 70 K (dashed-dotted line).This can be explained by the presence of an additional AF interaction between the dimers, J ′ = αJ ̸ = 0.
One of the limiting cases of strong AF interaction between dimers is the homogeneous chain, which also exhibits a broad peak in magnetic susceptibility [44].However, the homogeneous chain lacks a gap in the magnetic excitation spectrum, and the magnetic susceptibility drops at low temperatures by only ~30% relative to the peak value rather than to zero, as observed experimentally.An intermediate variant is a partially dimerized chain described by the Hamiltonian [44][45][46], when a strong AF interaction in pairs of ions, J, alternates with a weaker J ′ = αJ interaction between pairs.The availability of the theory of partially dimerized chains [44,45] allowed us to perform simulations that showed that by adjusting the degree of dimerization α (dashed and solid lines in Figure 6), a reasonably good agreement of the model with the experimental data can be achieved.The optimal fitting result was obtained for α ≈ 0.5, which means a sufficiently strong interaction between dimers, only two times weaker than the AF interaction within the dimer, J ′ = J/2.One of the limiting cases of strong AF interaction between dimers is the homogeneous chain, which also exhibits a broad peak in magnetic susceptibility [44].However, the homogeneous chain lacks a gap in the magnetic excitation spectrum, and the magnetic susceptibility drops at low temperatures by only ~30% relative to the peak value rather than to zero, as observed experimentally.An intermediate variant is a partially dimerized chain described by the Hamiltonian  = − ∑ ( ⃗ ⋅  ⃗ +  ⃗ ⋅  ⃗ ) / [44][45][46], when a strong AF interaction in pairs of ions, J, alternates with a weaker J′ = αJ interaction between pairs.The availability of the theory of partially dimerized chains [44,45] allowed us to perform simulations that showed that by adjusting the degree of dimerization α (dashed and solid lines in Figure 6), a reasonably good agreement of the model with the experimental data can be achieved.The optimal fitting result was obtained for α ≈ 0.5, which means a sufficiently strong interaction between dimers, only two times weaker than the AF interaction within the dimer, J′ = J/2.
Detailed structural data (Figure 3) were obtained only for complex 2 since it was possible to grow single crystals for it.However, certain conclusions can be drawn from the structure of complex 1, which differs from complex 2 only by the absence of the (CH3)2SO group coordinated with the copper(II) ion.As can be seen in Figure 3, the two copper ions in the structure of complex 2, and obviously complex 1, are united in a stable metallocycle Cu2N4, which leads to the appearance of magnetic dimers.In the structure of complex 2, dimers are spatially separated by groups (CH3)2SO, while in complex 1, these groups are absent, which leads, according to the magnetic data, to the formation of a chemical bond between the dimers, capable of transferring between them strong enough AF exchange interaction.It should be noted that the magnetic susceptibility data, while demonstrating the presence of interaction between magnetic Cu 2+ -Cu 2+ dimers, still do not allow us to conclude whether linear chains of dimers, ladder, or other similar structures are formed upon the packing of molecules in the crystal structure.Temperature dependencies of the magnetic susceptibility χ p of complex 1, after subtracting the contribution from the monomer impurity.The lines show approximations of the experimental data by theoretical models: the model of non-interacting dimers (dashed-dotted line, dashed-double dotted line) and the model of dimerized chains (dashed and solid lines) with the interaction parameters within dimers J and between dimers J ′ = αJ indicated in the figure.The first curve (dashed-double dotted line) corresponds to the optimization of the model description in the region of the magnetic susceptibility peak; for the other curves, the fitting of the J parameter was carried out in the temperature region 80-300 K.
Detailed structural data (Figure 3) were obtained only for complex 2 since it was possible to grow single crystals for it.However, certain conclusions can be drawn from the structure of complex 1, which differs from complex 2 only by the absence of the (CH 3 ) 2 SO group coordinated with the copper(II) ion.As can be seen in Figure 3, the two copper ions in the structure of complex 2, and obviously complex 1, are united in a stable metallocycle Cu 2 N 4 , which leads to the appearance of magnetic dimers.In the structure of complex 2, dimers are spatially separated by groups (CH 3 ) 2 SO, while in complex 1, these groups are absent, which leads, according to the magnetic data, to the formation of a chemical bond between the dimers, capable of transferring between them strong enough AF exchange interaction.It should be noted that the magnetic susceptibility data, while demonstrating the presence of interaction between magnetic Cu 2+ -Cu 2+ dimers, still do not allow us to conclude whether linear chains of dimers, ladder, or other similar structures are formed upon the packing of molecules in the crystal structure.

Cytotoxic Activity
The effect of H 2 L ligand, Cu(NO 3 ) 2, and complex 1 on human cell viability was evaluated on the tumor cell lines MCF7 (breast adenocarcinoma), Hep2 (laryngeal carcinoma), A549 (lung carcinoma), HepG2 (hepatocellular carcinoma) and non-tumor MRC5 lung fibroblasts.The results of the study are presented in Table 3 in the form of the LC50 parameter, which is calculated as the concentration of the substance at which cell death is 50%, and Figures 7 and S4.The H2L ligand and copper (II) salt have no cytotoxic effect on all cell lines after 48 h of exposure in the concentration range from 1 to 100 µM, whereas the [Cu2L2]•C2H5OH complex exhibits strong dose-dependent cytotoxic activity (Table 3, Figure 7).Complex 1 exhibited the highest activity on MCF7 cells (LC50 = 0.18 ± 0.03 µM) and the lowest activity on A549 cells (LC50 = 43.6 ± 2.6 µM).

Compound
Based on the obtained data on cytotoxic activity, the selectivity indices (SI, it is a ra-  The H 2 L ligand and copper (II) salt have no cytotoxic effect on all cell lines after 48 h of exposure in the concentration range from 1 to 100 µM, whereas the [Cu 2 L 2 ]•C 2 H 5 OH complex exhibits strong dose-dependent cytotoxic activity (Table 3, Figure 7).Complex 1 exhibited the highest activity on MCF7 cells (LC 50 = 0.18 ± 0.03 µM) and the lowest activity on A549 cells (LC 50 = 43.6 ± 2.6 µM).
Based on the obtained data on cytotoxic activity, the selectivity indices (SI, it is a ratio of the LC 50 value for non-tumor MRC5 fibroblasts to the LC 50 value for tumor cells) of the action of complex 1 against tumor cells were calculated, and the results are presented in Table 3.According to the literature, compounds with SI > 3 (or >9 in some sources) are considered promising for further studies as potential antitumor agents [47].The investigated copper(II) complex shows high selectivity of action against HepG2 (SI = 5.1) and MCF7 (SI = 22.4) cells; for Hep2 cells, the activity of the complex is comparable to the activity on non-tumor fibroblasts MRC5, and for A549 cells SI < 1.

Synthetic Procedures
Synthesis of [Cu 2 L 2 ]•C 2 H 5 OH (1).A 0.176 g (0.50 mmol) portion of H 2 L was dissolved on heating in 15 mL of ethanol, and 0.060 g (0.25 mmol) of Cu(NO 3 ) 2 •3H 2 O was dissolved in 1 mL of H 2 O with the addition of 1 drop of 1N HNO 3 .The copper(II) solution was added to the hot ligand solution, and a dark green solution was formed, from which a precipitate of the same color quickly formed.The precipitate was filtered off the next day, washed several times with ethanol, and air-dried.The yield was 0.08 g (73% based on Cu). For

Materials and Techniques
Commercially available reagents and solvents were used for the synthesis without additional purification.Methyl-5-(trifluoromethyl)pyrazol-3-yl-ketazine (H 2 L) was prepared according to the published method [40].
Elemental analysis for C, H, N was performed in the Laboratory of Microanalysis of the N.N.Vorozhtsov Novosibirsk Institute of Organic Chemistry SB RAS.Copper(II) content was analyzed complexometrically with EDTA disodium salt solution and murexide as an indicator after decomposition of the complex samples in concentrated acids (H 2 SO 4 + HClO 4 ).
Single-crystal X-ray diffraction data were collected using the graphite monochromatized MoKα-radiation (λ = 0.71073 Å) at 150(2) K on an X8APEX Bruker Nonius diffractometer equipped with a 4K CCD area detector.The φ-scan technique was employed to measure intensities.Absorption corrections were applied empirically using the SADABS program [48].Structures were solved by the direct methods of the difference Fourier synthesis and further refined by the full-matrix least squares method using the SHELXTL package(Veesion 2014/7) [49].Atomic thermal parameters for non-hydrogen atoms were refined anisotropically (Table 4).The positions of hydrogen atoms were calculated corresponding to their geometrical conditions and refined using the riding model.The disordering of the CF 3 group was introduced statistically with the restriction of C-F bonds with a length of 1.32(1) Å and a population of 80.8/19.2%.The introduction of the second position of the fluorine atoms also required a slight limitation of the thermal displacement parameters (ISOR).

Spectroscopy
IR absorption spectra were taken on an IRAfinity-1S spectrometer (Shimadzu, Kyoto, Japan) 4000-400 cm −1 at room temperature, and samples were prepared in KBr.

Magnetic Measurements
Magnetic properties were measured on a Quantum Design MPMS-XL SQUID magnetometer within the temperature range of 1.77-300 K at magnetic fields H = 0-10 kOe.To determine the paramagnetic component of the molar magnetic susceptibility χ p (T), the contributions of core diamagnetism χ d and possible ferromagnetism of micro-impurities χ FM were subtracted from the measured values of the total susceptibility χ = M/H (M = magnetization).The temperature-independent contribution χ d was calculated according to Pascal's additive scheme.To check the presence of ferromagnetic contribution

Figure 4 .
Figure 4. Two orientations of the pseudo-layer of 2 in the bc plane (a) and along the b-axis (b).

Figure 4 .
Figure 4. Two orientations of the pseudo-layer of 2 in the bc plane (a) and along the b-axis (b).

Table 1 .Figure 4 .
Figure 4. Two orientations of the pseudo-layer of 2 in the bc plane (a) and along the b-axis (b).

Figure 5 .
Figure 5. (a) Temperature dependences of paramagnetic component of magnetic susceptibility χp of complex 1 measured in magnetic fields H = 1, 10 kOe.(b) Temperature dependences of the inverse susceptibility 1/χp and effective magnetic moment µeff calculated in the approximation of non-interacting ions (θ = 0).

Figure 5 .
Figure 5. (a) Temperature dependences of paramagnetic component of magnetic susceptibility χ p of complex 1 measured in magnetic fields H = 1, 10 kOe.(b) Temperature dependences of the inverse susceptibility 1/χ p and effective magnetic moment µ eff calculated in the approximation of non-interacting ions (θ = 0).

16 Figure 6 .
Figure 6.Temperature dependencies of the magnetic susceptibility χp of complex 1, after subtracting the contribution from the monomer impurity.The lines show approximations of the experimental data by theoretical models: the model of non-interacting dimers (dashed-dotted line, dashed-double dotted line) and the model of dimerized chains (dashed and solid lines) with the interaction parameters within dimers J and between dimers J′ = αJ indicated in the figure.The first curve (dashed-double dotted line) corresponds to the optimization of the model description in the region of the magnetic susceptibility peak; for the other curves, the fitting of the J parameter was carried out in the temperature region 80-300 K.

Figure 6 .
Figure 6.Temperature dependencies of the magnetic susceptibility χ p of complex 1, after subtracting the contribution from the monomer impurity.The lines show approximations of the experimental data by theoretical models: the model of non-interacting dimers (dashed-dotted line, dashed-double dotted line) and the model of dimerized chains (dashed and solid lines) with the interaction parameters within dimers J and between dimers J ′ = αJ indicated in the figure.The first curve (dashed-double dotted line) corresponds to the optimization of the model description in the region of the magnetic susceptibility peak; for the other curves, the fitting of the J parameter was carried out in the temperature region 80-300 K.

Table 2 .
Wave numbers (frequencies, cm -1 ) of absorption band maxima in the IR spectra of H 2 L and complex 1.

Table 3 .
Cytotoxic activity of the ligand and complex 1 expressed in terms of LC 50 parameter and selectivity index (SI).