A tetranuclear copper (II) complex with pyrazole-3,5-dicarboxylate ligands: synthesis, characterization and electrochemical properties

2.1 Materials and methods

All reagents [CuI (Sigma-Aldrich, Steinheim, Germany) and H₃pdc.H₂O (Alfa-Aesar, Karlsruhe, Germany)] and solvents [dimethylformamide (DMF), Merck, Darmstadt, Germany] used were of analytical grade. IR spectra were recorded on a Perkin-Elmer 100 FTIR spectrophotometer (PerkinElmer Inc., Buckinghamshire, UK) as ATR between 4000 and 650 cm⁻¹. The elemental analyses (C, H, and N) were determined with a LECO, CHNS-932 instrument (Leco Corp., St. Joseph, MI, USA). TG-DTA curves were obtained using a Seiko Exstar 6200 thermogravimetry/differential thermal analysis (TG/DTA) thermal analyzer (SII NanoTechnology Inc., Tokyo, Japan) under a dynamic air atmosphere. A sample size of 3–5 mg and a heating rate of 20 °C/min in the temperature range of 30–1000 °C were used. Room temperature magnetic susceptibility measurement was carried out using a Sherwood Scientific MKI model Gouy magnetic balance (Sherwood Scientific Ltd., Cambridge, UK). The electrochemical studies were performed using a CH Instruments system (model 1140B; CH Istruments Inc., Austin, TX, USA) with a Pt disc as working electrode, a Pt spiral as counter electrode, and Ag/AgCl as the reference electrode. Before the experiments, the working electrode was polished, rinsed with triply distilled water, cleaned in the ultrasonic bath, washed with DMF and dried.

2.2 Synthesis of [Me₂NH₂]₄[Cu₄(pdc)₄]

A 5-mL DMF suspension of CuI (0.19 g, 1 mmol) was mixed with a 5-mL DMF solution of 0.087 g (0.5 mmol) H₃pdc·H₂O. The stock mixture distributed in five glass vials was placed in an oven at 100 °C. Block-shaped deep-blue single crystals of the complex were obtained within 4 days (0.021 g, yield 16%, based on H₃pdc·H₂O). C₂₈H₃₆Cu₄N₁₂O₁₆ (1050.839): calcd. C 32.00, H 3.45. N 15.99; found C 31.64, H 3.90, N 16.18. IR (ATR) (cm⁻¹): ν = 3098m, 3047m (C–H_arom); 2935w (CH₃); 2789, 2484 [ν(N⁺–H)]; 1619vs, 1602vs, 1573vs (ν_as C=O); 1384s (ν_s C=O).

2.3 Crystal structure determination

Single-crystal data collection was carried out on a STOE IPDS2 diffractometer. The structure was solved by Direct Methods with shelxs and refined with shelxl [18]. Where possible, H atom positions were calculated, and a riding model was applied with U_eq = 1.2/1.5 U_eq (parent atom). The highest peak on the final difference Fourier map of 0.41 e/ų was located on the C4–C5 bond. The basic crystallographic data are collected in Table 1.

CCDC 1058512 contains the supplementary crystallographic data for this paper. These data can be obtained free of charge from The Cambridge Crystallographic Data Centre viawww.ccdc.cam.ac.uk/data_request/cif.

3.1 Description of the crystal and molecular structure

Each complex anion consists of two nearly planar dinuclear [Cu^II₂(pdc)₂] units, where pdc is the triply deprotonated pyrazole-3,5-dicarboxylate (pdc³⁻) anion (Fig. 1). The dinuclear [Cu^II₂(pdc)₂] units are stacked together to form a tetranuclear complex. The stacking of the [Cu^II₂(pdc)₂] units occurs with a small offset with respect to each other, resulting in additional axial Cu–N and Cu–O bonds and expansion of the coordination sphere of the Cu²⁺ ions from four to five in a square-pyramidal arrangement. The entire tetranuclear aggregate has crystallographic inversion symmetry.

Fig. 1:

Molecular structure of the complex [Me₂NH₂]₄[Cu₄(pdc)₄]. Symmetry-independent non-hydrogen and non-carbon atoms are labeled. Displacement ellipsoids are drawn at the 30% probability level.

Interestingly, two dinuclear Cu²⁺ complexes containing solely the planar [Cu^II₂(pdc)₂] unit have been synthesized and structurally characterized before ([Et₃NH]₂[Cu₂(pdc)₂(H₂O)₂] [6] and [Me₂NH₂]₂[Cu₂(pdc)₂(H₂O)₂]·H₂O [20]). In these dinuclear complexes, additional water molecules are coordinated to the axial position of the square-pyramidal Cu²⁺ ions, thus preventing the formation of the tetranuclear complex that we obtained in our project. [Me₂NH₂]₂[Cu₂(pdc)₂(H₂O)₂]·H₂O [20] was synthesized using copper(II) chloride as metal source in aqueous dimethylamine and DMF solutions after a month. [Et₃NH]₂[Cu₂(pdc)₂(H₂O)₂] [6] was synthesized starting from Cu(NO₃)₂·3H₂O under hydrothermal conditions at 165 °C for 16 h.

Although no temperature was reported for the synthesis of [Me₂NH₂]₂[Cu₂(pdc)₂(H₂O)₂]·H₂O [20], regarding the long crystallization time, we assume that reaction conditions probably milder than ours yield the dinuclear complex. Meanwhile, in the case of [Et₃NH]₂[Cu₂(pdc)₂(H₂O)₂] [6], much harsher reaction conditions likewise yielded only a dinuclear species. Thus, it is not clear what exactly caused the formation of [Me₂NH₂]₄[Cu₄(pdc)₄], but it is reasonable to assume that the complex redox reactions leading to the tetranuclear complex also played a role.

The anions chelate each of the Cu²⁺ cations through one of the pyrazole N atoms and two carboxylate O atoms (Fig. 1). Within each planar unit, the Cu–N and Cu–O bond lengths are at 1.908(2)–1.924(2) and 1.961(2)–1.988(2) Å, respectively (Table 2). The units comprising two Cu²⁺ ions and the organic pdc³⁻ anion are essentially planar, especially the unit with the N3–N4 ligand. The uncoordinated carboxylate atoms O4 and O8 are displaced from these planes by 0.40(1) and 0.12(1) Å, respectively. The two units are stacked through Cu–O/N bonds (Cu2–N4 at –x, –y, 1 – z of 2.850(3) Å; Cu1–O5 at –x, –y, 1 – z of 2.876(3) Å). Similar planar units arranged in stacks were reported for the related compounds of various transition metals [6, 17].

The Cu1···Cu2 distance within each planar dinuclear complex unit is 3.957(1) Å, which is similar to the distance previously observed in the complex [Et₃NH]₂[Cu₂(pdc)₂(H₂O)₂] (3.962 Å [6]). The shortest Cu···Cu distance between the two stacked dinuclear units is 3.458(2) Å (Cu1···Cu2 at –x, –y, 1 – z). Between the pairs of stacked dinuclear units, the shortest Cu···Cu distance is 4.716(3) Å (Cu1···Cu2 at 1 – x, –y, 1 – z).

The anions form columns in the crystal structure, extending along [100] (Fig. 2). In the voids between these columns, the organic cations are located. Dimethylammonium cations act as donors in N–H···O hydrogen bonds with carboxylate O atoms as acceptors to create a three-dimensional hydrogen-bonded network (Table 3 and Fig. 2).

Fig. 2:

Columns of the complex anions extending along [100] in the crystal structure of [Me₂NH₂]₄[Cu₄(pdc)₄]. H atoms are omitted and C atoms are shown as sticks for clarity.

3.2 Infrared spectra

The IR spectra of H₃pdc and of the [Cu₄] complex are given in Figs. S1 and S2, respectively, in the Supplementary Information available online (see note at the end of the paper for availability). As expected, characteristic changes are observed. The C–H stretching modes of the aromatic ring and of the CH₃ group of the cations are observed in the region around 3000 cm⁻¹. The medium-weak bands at 2789 and 2484 cm⁻¹ are attributed to the NH₂⁺ stretching band of dimethylammonium cations [21]. The strong and sharp bands centered at 1602 cm⁻¹ with two shoulders at 1384 cm⁻¹ are assigned as the asymmetric (ν_as) and the symmetric (ν_s) stretching modes of the COO groups. The value of Δν (ν_as–ν_s) is 216 cm⁻¹, indicating a monodentate coordination mode of the carboxylate groups [22]. Other characteristic peaks are at 1474 and 1285 cm⁻¹, which can be attributed to C=N and C=C stretching vibrations of the pyrazole linker. The parent acid spectrum shows a strong absorption band at 1700 cm⁻¹, which is due to the stretching vibration of the COOH groups. The absence of this peak confirms the deprotonation of the ligand in the [Cu₄] complex.

3.3 Thermal analysis

The thermal behavior of the complex was studied by TG-DTA under dynamic air atmosphere. The TG-DTA curves are illustrated in Fig. 3. The mass loss at the first stage, between 30 and 100 °C, is attributed to the loss of one dimethylamine molecule. The experimental mass loss of 4.3 % is consistent with the calculated value of 4.4 %. The exothermic elimination of the remaining dimethylamine molecules and of the ligands occurs in the range of 231–563 °C, with mass loss of 66.9 % (calcd. 65.4 %). The final decomposition product is CuO (exp. 28.8 %, calcd. 30.3 %).

Fig. 3:

TG/DTA curves of [Me₂NH₂]₄[Cu₄(pdc)₄].

3.4 Electrochemical studies

Cyclic voltammograms of H₃pdc, Cu(II) nitrate and of the [Cu₄] complex (Fig. 4) were obtained at room temperature in 100 mm tetrabutylammonium perchlorate (TBAP) solutions in DMF, at the potential range of 1.5 to –1.5 V using Pt disc electrode versus an Ag/AgCl reference electrode.

Fig. 4:

Cyclic voltammograms of (a) H₃pdc and (b) Cu(II) nitrate solution. (c) The complex (0.50 mm) in DMF containing 100 mm TBAP at a scan rate of 100 mV/s.

The Cu(II) nitrate solution displays two reduction peaks at –0.298 V (reduction of Cu²⁺ to Cu⁺) and –0.759 V (reduction of Cu⁺ to Cu⁰) and two oxidation peaks at 0.723 V (oxidation of Cu⁰ to Cu⁺) and 1.027 V (oxidation of Cu⁺ to Cu²⁺). The complex displays two reduction peaks at –0.508 and –1.251 V, as well as one oxidation peak at 1.227 V. As a result of complexation, one oxidation peak disappeared. This may be caused by the chemical oxidation of copper during the complex formation (i.e. in complexation reaction the Cu(I) species is transformed to Cu²⁺ and the ligand is reduced). Voltammograms of the ligand solution show one reduction peak at –1.081 V and one oxidation peak at 0.827 V. The complex shows a lower current than observed for the Cu(II) salt solutions. This may lead to the following conclusions: (a) the diffusion coefficient of the complex is lower than that of the Cu(II) nitrate solution (i.e. the mass transfer rate for Cu²⁺ is higher than the mass transfer rate of the complex) and (b) the conductivity of the copper film and the rate of charge transfer on the electrode surface formed in Cu(II) nitrate solution is higher than that for the solution containing the [Cu₄] complex.

When the scan direction of potential was changed, we obtained similar results, which means that the initial oxidation state of the species present in the electrochemical cell does not affect the electrochemical behavior.

Cyclic voltammograms of the [Cu₄] compound at different scan rates are shown in Fig. S3 (Supplementary Information). The peak currents of all reduction and oxidation processes change linearly with the square root of the scan rate, indicating the diffusion characteristics of the electrochemical reactions. Furthermore, the logarithm of the peak current changes linearly with the logarithm of the scan rate by a slope value very close to 0.5, supporting the diffusion effect on the electrochemical mechanism [23]. In cyclic voltammogram studies, successive cycles were measured, and each cycle displays nearly the same peak current, indicating the diffusion control in the mechanism (Figs. S4–S6; Supplementary Information).