(I)-Olefin Complexes Bearing a 1,3,4-Oxadiazole

Using the alternating-current electrochemical technique, four new π -complexes, namely [Cu 2 (C 11 H 10 N 2 OS) 2 Br 1.91 Cl 0.09 ] ( 1 ), [Cu(C 11 H 10 N 2 OS)NO 3 ] ( 2 ), [Cu 2 (C 11 H 10 N 2 OS) 2 (H 2 O) 2 ](BF 4 ) 2 ( 3 ) and [Cu 2 (C 11 H 10 N 2 OS) 2 (H 2 O) 2 ](ClO 4 ) 2 ( 4 ), were obtained using copper(II) salts and the 2-(allylthio)-5-phenyl-1,3,4-oxadiazole (C 11 H 10 N 2 OS) ligand. The metal and halogen centers in 1 form Cu 2 X 2 dimers; the N-atom from the oxadiazole ring and the C=C bond of the allyl group from the same ligand complete the copper coordination environment, giving [Cu(C 11 H 10 N 2 OS)X] 2 isolated fragments. The ligand plays the same chelating role in 2 , whereas the O (NO 3 ) atom occupies the third position in the copper atom’s equatorial plane. Two more elongated Cu–O(NO 3 ) contacts associate the Cu(C 11 H 10 N 2 OS)NO 3 fragments into 1D chains. The geometries of the [Cu(C 11 H 10 N 2 OS)] 22+ cationic units in 3 and 4 are affected by the position of two water molecules in the coordination spheres of the copper atoms with respect to the plane containing the oxadiazole rings and the copper atoms. The molecular structures and Raman spectra of the compounds were computed using the DFT/B3LYP/cc-pVDZ level of theory. The results are compared with the experimental data obtained and used for vibrational band assignment.


Introduction
Complexes involving transition metals and olefins have received considerable theoretical and practical interest because of remarkable advances in the development of various catalytic processes.This means they have also played an important role in organometallic chemistry in recent decades [1][2][3][4].Among them, considerable attention has been paid to the investigation of Cu(I) πcomplexes with allyl derivatives of heterocyclic organic compounds, since the combination of the allyl radical and heterocyclic cores (both of which, according to hard soft acid base (HSAB) theory, act as soft bases) efficiently contributes to the stabilization of inorganic fragments that occur only extremely rarely [5].This is caused by a certain deficiency in the electron density in the d x2-y2 orbital (due to the π-dative (Cu→C=C) π component of the Cu(I)-olefin bond) and the difference in electron donation from the atoms of the competing heterocylic molecule and the present inorganic (or organic) anions.For example, despite the "hard character" of the SiF 6 2-anion, its fluorine atoms in the presence of π,σ-coordinated 1-allylbenzotriazole were first observed to be bound with a "soft" Cu(I) center in the corresponding crystalline compound [6].A specific contribution of the allyl radical attached to the 2-amino-5-phenyl-1,3,4-thiadiazole fragment through an amino group was also observed in an acetonitrile solution, where the organic ligand, while reacting with CuNO 3, undergoes deprotonation and gives a new, tetra-nuclear [Cu I (L)] 4 (L = ligand) complex of the azanide type [7].
Among a number of nitrogen-containing, five-membered aromatic heterocycles with an electron-deficient nature, 1,3,4-oxadiazole derivatives have been intensively studied in the fields of organic light-emitting diodes, biological activity, as well as crystal engineering [8][9][10][11].Despite the huge advances in the synthesis and the possible applications of 1,3,4-oxadiazole derivatives, the data concerning their use as co-ligands in metal-olefine π-coordination is still limited: only 16 corresponding entries have been found in the Cambridge Crystallographic Database (CCD) [12], while the analogous Cu(I) π-compounds appear to be structurally unstudied.Moreover, the coordination of CuCl, CuNO 3 and CuClO 4 with a 1,3,4-oxadiazole core has not been observed in the CCD references.To reveal the main features of copper-olefin π-coordination in the presence of a 1,3,4-oxadiazole core we present the synthesis and structural characterization of four new π-

Materials and instrumentation
Unless mentioned otherwise, all the chemicals were obtained from commercial sources and used without further purification.The NMR experiments, i.e.NMR spectra.Diffraction data were collected on an Agilent Gemini A four-circle diffractometer equipped with an Atlas CCD detector.

Single crystal X-ray diffraction studies.
Single-crystal data were collected on a Gemini A diffractometer equipped with an Atlas CCD detector, using graphite monochromated CuKa radiation for crystals 1 and 2 or graphite monochromated MoKa radiation for compounds 3 and 4. The collected diffraction data for 1-4 were processed with the CrysAlis PRO program [22].The structures were solved using direct methods with SHELXS-97 and refined using the least-squares method on F 2 with SHELXL-2013 and using the graphical interface of OLEX2 [23][24][25].The atomic displacements for the nonhydrogen atoms were refined using an anisotropic model.The hydrogen atoms (except the water molecules) were placed in ideal positions and refined as riding atoms with relative isotropic displacement parameters.In 2 and 3, the one allylsulfanyl group (C3 and C4 atoms) is disordered over two sites with an occupancy ratio of 0.886(6):0.114(6)and 0.874(4):0.126(4)respectively.
Similar behaviour of the allylsulfanyl group was observed in 4. Also, the ClO 4 -anion (O3, O4 and O6 atoms) is disordered in 4 over two sites with an occupancy ratio of 0.798(6):0.202(6).The crystal parameters, data collection and the refinement are summarized in Table 1.
Table 1.Selected crystal data and structure refinement parameters of 1-4. [a]   Crystal

Raman spectroscopy
Raman spectra were measured in the air on crystals of 1-4 and the ligand powder with a Horiba Jobin-Yvon LabRAM HR spectrometer using the 632.81-nm excitation line of a He-Ne laser with a power of 17 mW.The spectrometer was calibrated using a Si polycrystalline plate as a standard with a characteristic band at 520.6 cm -1 .

Computational details
All the calculations were performed at the DFT/B3LYP/cc-PVDZ level [26,27] using the GAMESS(US) program package [28].A basis set with effective core potentials (ECPs) was applied for the copper atoms.The initial geometries, derived from the crystal structure, were assumed to have C 1 symmetry and were utilized for the geometry optimization.The equilibrium geometries obtained were used for the Forces Constants matrix calculation.A polarizability tensor was then calculated and the resulting Raman activities (S i ) were converted to Raman intensities (I i ) using the following relationship from the intensity theory of Raman scattering [29,30].(dimethyl(vinyl)silyl)pyridine [36], allylacetone oxime [37], and 2-(allylthio)benzimidazole [15].The Cu(I) atom in the structure of 2 adopts close to a trigonal-pyramidal geometry (τ 4 = 0.79) [42], with one oxadiazole N atom, the allylic C=C bond and two O atoms of two symmetryrelated NO 3 -anions.The coordination of the apically bonded O3 i atom is intriguing, since the Cu-O3 i distance value of 2.598(2) Å is very close to the limiting distance of 2.63 Å for a Cu-O ap interaction in the structures of Cu(I) π-complexes with allyl derivatives of heterocycles [5,12].On the other hand, the fact that the copper atom is located in the plane of the equatorial ligands was never reported in the analogous compounds and indicates a significant deficiency of the electron

Computed geometries
The optimized structures are shown in Figures S1-S3, Supplementary Material.Selected geometrical parameters for every compound in comparison to the experimental ones are provided below each figure.
All the compounds were computed with C 1 symmetry.For compound 1 halogen atoms were set to be Br, since the fraction of C 1 is low in the experimental structure.The geometrical parameters at the equilibrium point appeared to be close to the experimental values.
For compound 2 the calculation was on an anion with two NO 3 − groups to keep the coordination surrounding the Cu atom.As expected, a slight distortion of the NO 3 − group's position appeared due to the absence of lattice influence, i.e., the O3 atom is pushed back and the Cu atom has only two oxygen atoms within the coordination sphere.This change affected the Cu−O distances, and the Cu-O3 i bond at the equilibrium point has a length of 2.0539 Å, while the same bond in the experimental structure was significantly longer at 2.598(2) Å.
For compounds 3 and 4 the geometry optimization faced some difficulties: in both cases the oxygen atom from the water molecules in the coordination sphere of the Cu atom tends to be equidistant from both metal centers.This leads to the exclusion of a second water molecule in the case of compound 3, so it was decided to use the dimer structure without water molecules.The optimized geometry appeared to be close to the experimental geometry.All the optimized structures are available in the Supplementary Material in XYZ format.
The equilibrium geometries were applied for Raman spectra calculations.In the computed spectra no imaginary frequencies appeared.

Raman spectra
The Raman spectra of compounds 1-4 are shown in Figure 7.The band assignments were made on the basis of the computed Raman spectra, see Tables S1- S3.The spectrum of the pure ligand is given in Figure S4.
The most intensive band in all the spectra appeared at 1604-1610 cm −1 and is due to ν(CC) from the ring of the Atphod ligand.The most interesting peculiarity of the spectra is related to the bands from the allyl group.The ν(C=C) band from the allyl group is a commonly used "fingerprint" for such compounds; it appeared at 1639 cm -1 in the spectrum of the pure ligand and was observed in the spectra of compounds 1-4 at 1561, 1565, 1562 and 1566 cm -1 , respectively.The computed spectrum for the cationic unit [Cu 2 (Atphod) 2 ] 2+ appearing in the structures of 3 and 4 indicated that this band is of zero intensity and is expected to appear at approximately 1570 cm -1 .Presumably, the suppression of the ν(C=C) allyl band is caused by the contribution of the Cu coordination surroundings to the resulting polarizability tensor, whereas the presence of water molecules in the structures of 3 and 4 indicated distortions, so that the band becomes Raman active.

3. Results and discussion 3 . 1
Crystal structures.The structures 1-4 demonstrate the first examples of copper(I)-olefin π-complexes with an active 1,3,4-oxadiazole core bonded to Cu(I) ion.The atphod molecule in compounds 1 and 2 acts as a bidentate chelating π,σ-ligand, being attached to the Cu(I) ion by means of the C=C bond of the allyl group and one the most nucleophilic N1 atoms of the oxadiazole ring (Figs.1 and 2).The copper(I) atom in 1 adopts close to a trigonal pyramidal geometry (τ 4 =0.85,τ 4 -four-coordinate geometry index), including the N atom, the C=C bond and one of the halogen X atoms in the basal plane of the metal polyhedron.The second halogen atom, that is apically bonded to the Cu(I) ion, is at a distance of 2.836(2) Å from the metal center, which is significantly shorter than the sum of the VdW radii of the Cu and Cl (3.15 Å) or the Cu and Br (3.23 Å) atoms, reported by Bondi,[31,32] and very much shorter than the corresponding sum of 4.20 Å for the Cu and Cl atoms, recently discovered by Alvarez[33].As a result, in 1 the bridging halogen atoms connect the organometallic [Cu(Atphod)] fragments into a centrosymmetric [Cu 2 (Atphod) 2 Br 1.91 Cl 0.09 ] dimer.A similar [-N,(C=C)Cl 2 -] 2 coordination topology for CuX (X = Cl or Br) was previously found in the structure of seven copper(I) chloride and three copper(I) bromide complexes with olefin-containing ligands,
atom and the other one to the O2 ii atom of another NO 3 -moiety (Cu-O2 ii ) 3.469(3) Å.The above statement might suggest that the Cu atom in structure 2 is five-coordinated and adopts a geometry that occurs for the Ag(I) ion in its π-coordination with olefins [43,44].Thus, bridging NO 3 -anions, by means of weak Cu-O3 i bonds in 2, connect [Cu(Atphod)] units into infinite chains in the 010 direction.

Figure 3 .
Figure 3. Coordination polymer in the structure of 2. ellipsoids are shown at 50 % probability.In contrast to 1 and 2, in complexes 3 and 4 the coordination behavior of the Atphod ligand, with regard to the Cu(I) ion, is reminiscent of 2-allylamino-5-methyl-1,3,4-thiadiazole [19] and 5allylthio-1-phenyl-tetrazoles [17, 45] in terms of dimer formation.The Atphod molecule in 3 and 4 acts as a tridentate chelate-bridging ligand, being attached to the metal centers by both oxadiazole N atoms (Figs. 4 and 5).A close to trigonal pyramidal Cu(I) environment in both structures (for 3 τ4 = 0.83 and 0.85; for 4 τ4 = 0.80) involves an allylic C=C bond in an equatorial position and two different N atoms of two neighboring oxadiazole rings, while the apical position of the metal polyhedron is occupied by a water molecule.Thus, a pair of copper atoms connects two Atphod

Figure 7 .
Figure 7. Raman spectra recorded on single crystals of 1, 2, 3 and 4. The region 1700-2700 cm -1 is removed due to the absence of bands.
1H NMR (500 MHz) and13C[ 1 H] NMR (125 MHz), were recorded on a Bruker Advance 500-MHz NMR spectrometer.The chemical shifts are reported in ppm relative to the residual peak of deuterated CDCl 3 for the 1 H and 13 C[1 H]