Crystal structure of a trifluoromethyl benzoato quadruple-bonded dimolybdenum complex

The quadruple-bond complex, [Mo2(p-O2CC6H4CF3)4·2THF], crystallizes in the triclinic space group P with intercalated pentane/THF lattice solvent molecules. The electron-withdrawing group on the paddlewheel carboxylate together with the axial THF molecules lead to a slight lengthening of the metal–metal bond, as predicted by Cotton.


Chemical context
The 2 4 2 quadruple bond has contributed prominently to the elucidation of the single most distinguishing feature of the discipline of chemistry -the two-electron bond (Lewis, 1916). As originally defined with the inception of valence and molecular orbital bonding models (Heitler & London, 1927;Pauling, 1928;Lennard-Jones, 1929;Mulliken, 1932;James & Coolidge, 1933;Coulson & Fischer, 1949), the two-electron bond forms from pairing two electrons in two orbitals. Remarkably, the four states that characterize the two-electron bond remained undefined experimentally for over 60 years owing to the dissociative nature of the and antibonding orbitals. This experimental challenge was overcome with the two-electron bond of quadruple-bonded metal-metal complexes. Anchored by a 2 4 framework and sterically locking ligands, the four states of the two-electron bond, 1 , 3 *, 1 * and 1 **, were experimentally defined for dimolybdenum quadruple-bond complexes (Engebretson et al. 1994(Engebretson et al. , 1999Cotton & Nocera, 2000). Within the group of dimolybdenum quadruple-bond complexes, the tetraacetates are exemplars. The initial structure of Mo 2 (O 2 CCH 3 ) 4 by Lawton & Mason (1965) established the existence of the quadruple bond in the now familiar paddlewheel arrangement of acetates. The dimolybdenum bond distance of 2.11 Å in this structure was subsequently refined nearly a decade later to be 2.0934 (8) (Cotton et al., 1974). Intriguingly, many subsequent structures have shown that the inductive effect of the R group on the carboxylic acid does not perturb the Mo-Mo bond distance, indicating the robustness of the two-electron bond within a quadruple-bond architecture. It has been postulated ISSN 2056-9890 that the strength of the Mo-Mo quadruple bond may be perturbed, but only in cases where R is a strong electronwithdrawing group and there is a good axial donor ligand (Cotton et al., 1978). To add further to an understanding of Mo 2 (II,II) quadruple bond distances, we examined a dimolybdenum core ligated by trifluoromethylbenzoate with THF axial donor ligands. We now report the synthesis and X-ray crystal structure of tetrakis(-4-trifluoromethylbenzoato-2 O:O 0 )dimolybdenum(II) 0.762-pentane 0.238-tetrahydrofuran solvate [Mo 2 (p-O 2 CC 6 H 4 CF 3 ) 4 Á2THF]Á0.762C 5 H 12 Á0.238C 4 H 8 O. The presence of the CF 3 electron-withdrawing group on the bridging benzoate ligands, together with the donor THF axial ligands, results in a slightly elongated metal-metal bond distance as compared to its benzoate congener, Mo 2 (O 2 CC 6 H 5 ) 4 .

Structural commentary
The dimolybdenum complex, [Mo 2 (p-O 2 CC 6 H 4 CF 3 ) 4 Á2THF] (Fig. 1), was characterized by using single-crystal X-ray diffraction. Half of the molecule (Fig. 1) resides in the asymmetric unit, with the complete molecule generated by inversion about the quadruple-bond inversion center. The fluorine atoms of the trifluoromethyl groups are rotationally disordered and the highest occupancy positions are shown in Fig. 1. The crystallization solvents, THF and pentane, are disordered (0.238:0.762) (Fig. 2).
Selected bond metrics for Mo 2 (p-O 2 CC 6 H 4 CF 3 ) 4 Á2THF are listed in Table 1. Complete lists of the structural metrics for the compound are presented in the Supporting information. The Mo-Mo bond length is 2.1098 (7) Å . Whereas the bond distance is within the typical range of dimolybdenum quadruple bond lengths of 2.06-2.17Å (Cotton et al., 2002), it is slightly longer than what is observed for dimolybdenum cores bridged by carboxylates. As a comparison, the dimolybdenum bond distance in the Mo 2 (O 2 CC 6 H 5 ) 4 congener, is 2.096 (1) Å . Thus, with the addition of a CF 3 group in the 4-position of benzoate, the Mo-Mo bond length increases by 0.014 (2) Å .
A similar trend is observed for the bond distances in the primary coordination sphere. The minimum Mo-O bond distance decreases by 0.008 (5) Å , and the maximum Mo-O bond distance decreases by 0.011 (5) Å as compared to Mo 2 (O 2 CC 6 H 5 ) 4 . The most significant decrease in bond metrics is observed for the Mo-O1S axial ligand distance, which results in a decrease of 0.033 (4) Å for the axial coordinated oxygen atom of THF as compared to the axially coordinated oxygen in Mo 2 (O 2 CC 6 H 5 ) 4 . However, we note for this compound that the oxygen is provided from a carboxylate ligand of a neighboring molecule as opposed to an axially coordinated solvent molecule. Consequently, as proposed by Cotton (Cotton et al., 1978), the presence of ligands about the dimolybdenum center that are electron withdrawing and donating in the axial position is needed to perturb the overall bonding within a quadruple-bond framework. To this point, the metrics of [Mo 2 (p-O 2 CC 6 H 4 CF 3 ) 4 ÁTHF] are indistinguishable from those of Mo 2 (O 2 CC 6 F 5 ) 4 ÁTHF (Han, 2011). The electron-withdrawing nature of the fluoro-substituted benzoates is established by their pK a s as compared to that of benzoate (pK a = 1.75, 3.77 and 4.20 for C 6 F 5 COOH, p-CF 3 C 6 H 4 COOH and C 6 H 5 COOH, respectively; Rumble, 2021;Boiadjiev & Lightner, 1999). That an electron-withdrawing group alone is insufficient to perturb the dimolybdenum bond distance is indicated by a comparison of the structures for Mo 2 (O 2 CCH 3 ) 4  Symmetry code: (i) Àx; Ày þ 1; Àz þ 1.

Figure 1
Ellipsoid plot of the dimolybdenum complex. The CF 3 groups are rotationally disordered, therefore the highest occupancy positions are shown for each atom. Hydrogen atoms and unbound solvent are omitted for clarity.

Supramolecular features
The structure was solved in the triclinic space group P1 with a half of an Mo-dimer per asymmetric unit and one full molecule per unit cell (Fig. 2) Sluis & Spek, 1990). The adjacent pairs of symmetry-related benzene rings (C10-C16) in the p-O 2 CC 6 H 4 CF 3 ligands interact through aromaticstacking interactions with a face-to-face distance of 3.7856 (9) Å ( Fig. 2b) and form a one-dimensional chain. In addition, the trifluoromethyl group of a p-O 2 CC 6 H 4 CF 3 ligand (for C10-C16 and F4-F6) is perpendicular to the aromatic plane of a neighboring p-O 2 CC 6 H 4 CF 3 ligand (C1-C7 and F1-F3) with weak C-FÁ Á Á interactions (Kawahara et al., 2004) [the distances between the F atoms and the C2-C8 plane are 3.024 (2)-3.430 (1) (Cotton & Norman, 1971) is TFACMO.

Purification and crystallization
The overall synthetic scheme is shown in the reaction scheme. Molybdenum hexacarbonyl, 4-(trifluoromethyl) benzoic acid, THF, and 1,2-dichlorobenzene were purchased from Sigma-Aldrich. Mo(CO) 6 and 4-(trifluoromethyl)benzoic acid were combined in a flask with THF and anhydrous 1,2-dichlorobenzene. The reaction was heated under reflux for 24 h at 413 K under nitrogen (Pence et al., 1999). The reaction mixture was cooled, the solution was filtered and the collected residue was washed with dichloromethane and hexanes.
The crystallization was set up in a glove box. The crude product was dissolved in THF and recrystallized by vapor diffusion of pentane using a 6 by 50 mm borosilicate glass crystallization tube housed within a 20 mL glass vial. The assembly was allowed to stand at 238 K for 24 days. Orange rectangular crystals were observed and harvested for X-ray diffraction analysis.

Refinement
Crystal data, data collection and structure refinement details are included in Table 2. Hydrogen atoms on C atoms were placed at idealized positions and refined using a riding model. The isotropic displacement parameters of all hydrogen atoms were fixed to 1.2 times the atoms to which they are linked (1.5 times for methyl groups). Rotational and positional disorder for one trifluoromethyl substituent containing C1 and C13 was modeled. The overlapping solvent molecules (assigned as THF and pentane based on solvent crystallization conditions and apparent arrangement of electron-density peaks) were disordered adjacent to an inversion center (special position). The restraints on bond lengths and constraints of the atomic displacement parameters on each pair of disorder fragments (SADI/SAME and EADP instructions of SHELXL2014) as well as the restraints of the atomic displacement parameters (SIMU/RIGU instructions of SHELXL2014) were applied for the disorder refinement (Zheng et al., 2008). Crystallographic refinement details, including disorder modeling and the software employed, are given in the crystallographic information file (*.cif). To stabilize the refinement model, 713 restraints (SADI/SAME and RIGU/SIMU) were applied to accommodate the disordered trifluoromethyl group, the coordinated THF molecules, as well as the THF/pentane solvent molecules in the channel as detailed by Mü ller et al. (2006) to furnish a data+restraint-to-parameter ratio of 9.75. This ratio increases to 11.6 if the disordered THF/pentane solvent molecules in the channel are squeezed out of the structure.

Computing details
Data collection: APEX2 (Bruker, 2015); cell refinement: SAINT (Bruker, 2015); data reduction: SAINT (Bruker, 2015); program(s) used to solve structure: SHELXT2014 (Sheldrick, 2015a); program(s) used to refine structure: SHELXL2014 (Sheldrick, 2015b); molecular graphics: SHELXTL (Sheldrick, 2008); software used to prepare material for publication: SHELXTL (Sheldrick, 2008), and PLATON (Spek, 2020). Special details Experimental. A single orange plate (0.297 mm × 0.132 mm × 0.056 mm) was chosen for single crystal X-ray diffraction using a Bruker three-circle platform goniometer equipped with an Photon100 CMOS detector. Data were collected as a series of φ and/or ω scans. Data integration down to 0.84 Å resolution was carried out using SAINT V8.37A with reflection spot size optimization. Absorption corrections were made with the program SADABS 2016/2 (Krause et al., 2015). Space group assignments were determined by examination of systematic absences, E-statistics, and successive refinement of the structures. The structure was solved by the Intrinsic Phasing methods and refined by least squares methods also using SHELXT-2014 and SHELXL-2014 with the OLEX 2 (Dolomanov et al., 2019) interface. The program PLATON (Spek, 2020) was employed to confirm the absence of higher symmetry space groups. All non-H atoms, including the disorder fragment, were located in difference Fourier maps, and then refined anisotropically. Outlier reflections were omitted from refinement when appropriate. Geometry. All esds (except the esd in the dihedral angle between two l.s. planes) are estimated using the full covariance matrix. The cell esds are taken into account individually in the estimation of esds in distances, angles and torsion angles; correlations between esds in cell parameters are only used when they are defined by crystal symmetry. An approximate (isotropic) treatment of cell esds is used for estimating esds involving l.s. planes. Refinement. Refinement of F 2 against ALL reflections. The weighted R-factor wR and goodness of fit S are based on F 2 , conventional R-factors R are based on F, with F set to zero for negative F 2 . The threshold expression of F 2 > 2sigma(F 2 ) is used only for calculating R-factors(gt) etc. and is not relevant to the choice of reflections for refinement. R-factors based on F 2 are statistically about twice as large as those based on F, and R-factors based on all data will be even larger. All non-H atoms, as well as the disordered atoms were located in difference-Fourier maps, and then refined anisotropically. The restraints on bond lengths and constraints of the atomic displacement parameters on each pair of disorder fragments (SADI/SAME and EADP instructions of SHELXL-2014) as well as the restraints of the atomic displacement parameters (SIMU/RIGU instructions of SHELXL-2014), if necessary, have been applied for the disorder refinement.