I. INTRODUCTION
In selective oxidation processes, it is a technological challenge to use atmospheric oxygen directly under ambient conditions of pressure and temperature to replace highly polluting oxidizing agents normally used, avoiding excess reagents and reducing the formation of byproducts (Anastas and Kirchhoff, Reference Anastas and Kirchhoff2002; Hermans et al., Reference Hermans, Spier, Neuenschwander, Turrà and Baiker2009; Roduner et al., Reference Roduner, Kaim, Sarkar, Urlacher, Pleiss, Gläser, Einicke, Sprenger, Beifuß, Klemm, Liebner, Hieronymus, Hsu, Plietker and Laschat2013; Hone and Kappe, Reference Hone and Kappe2019). The activation of molecular oxygen and more specifically the action of depositing an oxygen atom on a substrate is a process that in nature is carried out by enzymes such as oxotransferases or hydroxylases, where the molybdenum-oxygen unit (Mo = O) is present as the active center (Arzoumanian, Reference Arzoumanian1998; Hille, Reference Hille2002; Hille et al., Reference Hille, Nishino and Bittner2011; Heinze, Reference Heinze2015).
The study of these natural systems called molybdenum-enzymes has been the source of inspiration to synthesize biomimetic structures with the dioxo-Mo unit as the active center, compounds which have been used as catalysts selectively in oxygen atoms transfer (OAT) process to substrates such as phosphines, alcohols, arylalkanes, and alkenes using different oxo-donors in solution (Martínez et al., Reference Martínez, Cáceres, Martínez, Páez-Mozo, Valange, Castellanos, Molina, Barrault and Arzoumanian2016; Martinez Q et al., Reference Martinez Q, Amaya, Paez-Mozo, Martinez O and Valange2020; Kargar et al., Reference Kargar, Fallah-Mehrjardi, Behjatmanesh-Ardakani, Munawar, Ashfaq and Tahir2021; Julião et al., Reference Julião, Gomes, Cunha-Silva, Pillinger, Gonçalves and Balula2022).
Molybdenum's ability to selectively generate oxygenated products has been explained by its natural property to incorporate oxygen atoms into its coordination sphere in high oxidation states (>4) (Dupé et al., Reference Dupé, Judmaier, Belaj, Zangger and Mösch-Zanetti2015; Zwettler et al., Reference Zwettler, Ehweiner, Schachner, Dupé, Belaj and Mösch-Zanetti2019). On the other hand, it has been shown that its oxotransfer capacity can be increased by factors such as the presence of a second oxo ligand in the Mo = O unit and/or the presence of ligands with a high electronic density that can provide charge density and populate the metal center to reduce the bond order and facilitate oxygen donation (Arzoumanian et al., Reference Arzoumanian, Lopez and Agrifoglio1994; Kühn et al., Reference Kühn, Santos and Abrantes2006).
Different studies of OAT catalyzed by dioxo-Mo complexes have been reported using bidentate dithiocarbamates, bipyridines, Schiff bases, salen-type ligands, or combinations of these, and the results have shown their high selectivity and reactivity as homogeneous and heterogenous catalysts (Arzoumanian et al., Reference Arzoumanian, Bakhtchadjian, Agrifoglio, Atencio and Briceño2006; Bakhtchadjian et al., Reference Bakhtchadjian, Tsarukyan, Barrault, Martinez, Tavadyan and Castellanos2011; Castellanos et al., Reference Castellanos, Martínez, Páez-Mozo, Ziarelli and Arzoumanian2012, Reference Castellanos, Martínez, Lynen, Biswas, Van Der Voort and Arzoumanian2013, Reference Castellanos, Martínez Q, Martínez O, Leus and Van Der Voort2021; Qin et al., Reference Qin, Wang, Li, Wu and Chen2019; Bouzari et al., Reference Bouzari, Bezaatpour, Babaei, Amiri, Boukherroub and Szunerits2021; Nunes et al., Reference Nunes, Gomes, Neves, Mendes, Almeida Paz, Lopes, Pillinger, Gonçalves and Valente2023). Oxygen atom transfer processes using dioxo-Mo complexes with phenanthroline-type ligands are less frequent in the literature although this polypyridine ligand is considered one of the main symmetric platforms to obtain transition metal complexes. Its relevance lies in the donor capacity of nitrogen atoms, which can be modulated by the incorporation of different substituents on its periphery, influencing the catalytic activity of the metal center (Shen and Patrick Sullivan, Reference Shen and Patrick Sullivan1995; Bencini and Lippolis, Reference Bencini and Lippolis2010; Camargo et al., Reference Camargo, Henao and Castellanos2016).
In this work, we report the synthesis of a new dioxo-molybdenum (VI) complex using the bidentate ligand 1,10-phenanthroline disubstituted with methyl groups. The results obtained by NMR, elemental analysis, and infrared spectroscopy allowed the respective molecular characterization. The structural determination of the dichloridodioxido-[(4,7-dimethyl)-1,10-phenanthroline]molybdenum(VI) compound as a biomimetic complex of molybdenum-enzymes was solved from X-ray powder diffraction (XRPD) data analysis using the simulated annealing method with a subsequent refinement using the Rietveld method. The atomic coordinates were determined using the EXPO2014 program and validated using the PLATON multipurpose software.
II. EXPERIMENTAL SECTION
4,7-dimethyl-1,10-phenanthroline was obtained from Alfa Aesar and MoO2Cl2 was purchased from Sigma-Aldrich and used without further purification. Commercial grade solvents were dried and deoxygenated by refluxing for at least 12 h over appropriate drying agents under argon atmosphere and were freshly distilled prior to use. IR (KBr) was recorded with a Perkin-Elmer 1720XFT. 1H and 13C NMR were performed with Bruker Avance 400 spectrometer. The CHN elemental analysis was performed on a Thermo Scientific Flash 2000 CHNS/O analyzer equipped with a TCD detector.
A. Synthesis of dichloridodioxido-[(4,7-dimethyl)-1,10-phenanthroline]molybdenum(VI)
0.208 g (1.0 mmol) of 4,7-dimethyl-1,10-phenanthroline ligand dissolved in acetonitrile (20 mL) was slowly added under a nitrogen atmosphere to a solution of acetonitrile (20 mL) containing 0.200 g (1.0 mmol) of MoO2Cl2 and the reaction mixture was stirred overnight at reflux temperature (85 °C). After that, the reaction mixture was cooled to room temperature, the solid obtained separated from the reaction medium by filtration under N2 atmosphere, and the resulting solid was washed three times with ethyl ether (3 × 20 mL) to obtain a light pink solid. (0.310 g; %R = 75.9) IR (cm−1) KBr: 3089 (=CH), 2991 (CH), 1622 (C = C), 1427 (C = C), 935 (Mo = Oasym), 899 (Mo = Osym). 1H NMR (400 MHz, DMSO) δ 9.17 (d, J = 5.1 Hz, 2H), 8.47 (s, 2H), 8.10 (d, J = 5.1 Hz, 2H), 2.99 (s, 6H). Elemental analysis calculated for C14H12Cl2MoN2O2 (407.93): C 41.30, H 2.97, N 6.88. Found: C 41.44, H 2.99, N 6.39 (Figure 1).
Figure 1. Synthesis of dichloridodioxido-[(4,7-dimethyl)-1,10-phenanthroline]molybdenum(VI).
B. Powder diffraction data collection
XRPD measurement of the dichloridodioxido-[(4,7-dimethyl)-1,10-phenanthroline]molybdenum(VI) complex (Figure 2, blue line) was carried out at 298 K using an X'Pert Pro MPD PANalytical diffractometer with a Cu anode (λ = 1.5418 Å) and a nickel filter. The diffractometer employed the Bragg–Brentano configuration and a high-speed PIXcel solid-state detector for data acquisition with a primary and secondary goniometer radius of 240 mm. A small fraction of the sample was ground and homogenized in an agate mortar and then mounted on a stainless-steel sample holder using the face-fill technique. The diffraction data were collected using a step size of 0.0263° 2θ and a counting time of 97.920 s with a No. of points of 2856. From the powder diffractogram data obtained, crystallographic characterization (indexing and space group determination), whole powder profile decomposition (WPPD), structural elucidation using the simulated annealing method, and refinement using the Rietveld method were then carried out. Finally, the quality of the results obtained was evaluated using the PLATON and DIAMOND programs.
Figure 2. Experimental XRPD diffractogram of the dichloridodioxido-[(4,7-dimethyl)-1,10-phenanthroline]molybdenum(VI) using Cu Kα radiation (blue solid line) and the refined pattern obtained using WPPD procedure (red solid line).
III. RESULTS AND DISCUSSION
The pattern obtained by XRPD was indexed on a monoclinic unit cell with a = 12.9287 Å, b = 9.7604 Å, c = 11.9717 Å, β = 101.688°, unit-cell volume V = 1479.376 Å3 using TOPAS v.5 (Coelho and Kern, Reference Coelho and Kern2004; Bruker AXS GmbH, 2017a, 2017b) (Figure 2, blue line). Analysis of the systematic absences using EXPO2014 (Altomare et al., Reference Altomare, Cuocci, Giacovazzo, Moliterni, Rizzi, Corriero and Falcicchio2013) suggested the space group C2/c (No. 15) with figures of merit: M(20) = 20.7 and F(30) = 31.3 (Table I). The refinement of microstructural parameters of polycrystalline material was performed with TOPAS v.5 (Pawley, Reference Pawley1981) fitting using the WPPD procedure employing fundamental parameters ( Cheary and Coelho, Reference Cheary and Coelho1992; Le Bail, Reference Le Bail2005; Bruker AXS GmbH, 2017b) (Figure 2, red line). To generate the emission profile, the CuKα5_Berger profile option was selected, and a Chebyshev Polynomial of order 5 was used to fit the background and to obtain the refined unit-cell parameters. Residual values from Pawley fit obtained from the refinement process allowed us to validate the experimental data obtained (R exp = 1.23, R wp = 4.65, Rp = 4.65, 
$R^{\prime}_{{\rm wp}}$ = 17.96, 
${R}^{\prime}_p$ = 19.05, and GoF = 3.78). It is important to mention that six weak peaks (2θ = 10.342°, 10.784°, 16.120°, 17.379°, 19.432°, and 27.034°) observed in the experimental diffractogram were excluded from the indexing process and were associated with the presence of remaining impurities of the precursors used in the synthesis which were not possible to eliminate in the purification process, and it was not possible to attribute directly to another specific phase present in the sample.
TABLE I. X-ray powder diffraction data obtained for the dichloridodioxido-[(4,7-dimethyl)-1,10-phenanthroline]molybdenum(VI) complex.
The d-values were calculated using Cu Kα 1 radiation (λ = 1.5405981 Å).
Once the crystallographic characterization was completed, the structural elucidation was carried out using EXPO2014 program applying the simulated annealing method (Altomare et al., Reference Altomare, Camalli, Cuocci, Giacovazzo, Moliterni and Rizzi2009, Reference Altomare, Cuocci, Giacovazzo, Moliterni, Rizzi, Corriero and Falcicchio2013). The angular range used for the structural determination was 5.0° to 45.0° 2Theta with a Chebyshev-type polynomial background noise modeling of order 19. To solve the structure and obtain the best fit between the observed profile and the calculated profile, it was necessary to work with the dynamic occupation model, in turn refining the torsions between the atoms C4:N1:Mo1:O1, C2:C3:C13: H5, and C1:C2:C6:H10. The cost function obtained was 9.719 and the atomic positions of the refined atomic positions are given in Table II.
TABLE II. Atomic coordinates obtained for the compound dichloridodioxido-[(4,7-dimethyl)-1,10-phenanthroline]molybdenum(VI)
Finally, the structure obtained was refined using the Rietveld method (least squares) with the TOPAS program, and the function of fundamental parameters was used as a profile modeling function and for peak corrections, the Zero error variable whose value obtained was 0.0646(8). Additionally, the scale factor, cell constants, and possible preferential orientations were refined using the mathematical model of spherical harmonics of order 4. Background was treated using a Chebyshev-type model of order 7, and the microstructural factors corrections using Lorentzian-type Double-Voigt mathematical approximation (Cry size). The signals of the identified impurities are highlighted in blue in Figure 3 and were included in the modeling using the fundamental parameter function refining their position, intensity, and peak width. The final crystallographic parameters obtained are presented in Table III.
Figure 3. Rietveld refinement result of the dichloridodioxido-[(4,7-dimethyl)-1,10-phenanthroline]molybdenum(VI) (red line). The blue diffractogram corresponds to the experimental profile and the refinement residual is the gray baseline. The perpendicular solid blue lines correspond to the reflections of impurities identified in the material.
TABLE III. Experimental data for dichloridodioxido-[(4,7-dimethyl)-1,10-phenanthroline]molybdenum(VI)
Figure 4 shows the molecular structure generated using the Mercury program. The planarity of the phenanthroline rings and the bidentate coordination of the nitrogen atoms with the central metal of molybdenum (VI) were confirmed. The asymmetric unit corresponds to half of the molecule (Z′ = 0.5) and the other half is reproduced by symmetry, which is why the labels of the atoms are repeated. It is so the chlorine, nitrogen, oxygen, and molybdenum atoms form a distorted octahedral coordination geometry with a volume of 12.307 Å3 and a quadratic elongation of 1.063 (Figure 5).
Figure 4. Representation of the molecular structure of the dichloridodioxido-[(4,7-dimethyl)-1,10-phenanthroline]molybdenum(VI) with labeling of atoms and rings. Cg2 and Cg3 refer to the centroids of the aromatic rings.
Figure 5. Octahedral coordination of the cis-dioxo-molybdenum (VI) complex coordinated with chlorine and the bidentate nitrogenous ligand 4,7-dimethyl-1,10-phenanthroline.
Previous research has established that the chlorine atoms are in -trans- position, while the oxygen and nitrogen atoms of the phenanthroline ligand are in the cis-configuration, generating greater stability in the molecule as a product of their respective electronic contributions. This preferential configuration (-cis-oxo, trans-Halogen, cis-N–N) has been previously studied by both experimental and theoretical measurements in bioinspired complexes of molybdenum or tungsten using ligands such as the chlorine, bromine fluorine, and bidentate ligands such as bipyridines and Schiff bases (Pietsch and Hall, Reference Pietsch and Hall1996; Barea et al., Reference Barea, Lledos, Maseras and Jean1998; Hofmann, Reference Hofmann2006; Agustin et al., Reference Agustin, Bibal, Neveux, Daran and Poli2009; Roy et al., Reference Roy, Biswal, Pramanik, Drew, Paul, Kachhap, Haldar and Chakrabarti2021).
A. Intermolecular hydrogen bonds
The geometry of the different intermolecular interactions and the π–π interactions between the centroid rings CgI and CgJ (I take the values of 2 and J take the values of 2 or 3) were determined through the PLATON program (Spek, Reference Spek2003, Reference Spek2015, Reference Spek2020), where it was specified that Cg2 and Cg3 are the centroids of rings 2 and 3 as described in Figure 4. The value of d refers to the distance between the centroids CgI⋯CgJ, α corresponds to the dihedral angle between planes I and J, β is the angle between the vector Cg(I) → Cg(J) and the normal to plane I, γ is the angle between the vector Cg(I) → Cg(J) and the normal to the plane J, CgI_Perp is the perpendicular distance of the centroid CgI on the ring J, and CgJ_Perp is the perpendicular distance of the centroid CgJ on the ring I. Table IV summarizes the most relevant geometric parameters generated by the variety of interactions between hydrogen bond donor (D) and acceptor (A) atoms.
TABLE IV. Most relevant geometric parameters of the dichloridodioxido-[(4,7-dimethyl)-1,10-phenanthroline]molybdenum(VI)
Cg2 is the centroid of the ring formed by atoms C1 to C5 and N1. These atoms are reproduced through symmetry to obtain the other aromatic ring, also with the centroid denoted Cg2 present in the other half of the molecule.
Cg3 is the centroid of the ring formed by the atoms C4, C3, C13, C4*, C3* y C13*.
a Atoms with atomic positions that are reproduced by symmetry. D is the Donor atom; A is the acceptor atom.
The structural elucidation process revealed the presence of four intermolecular hydrogen bonds. The C1–H1⋯Cl1 (2 interactions throughout the molecule) which in a projection on the c axis, form zigzag chains along the a axis [Figure 6(a)], and in the projection along the b axis are organized forming chains diagonally along the a and c axes [Figure 6(b)]. Furthermore, the other two intermolecular hydrogen atoms are present in the C13–H5⋯O1 bonds form linear chains along the b axis whose distance between the hydrogen atom and the oxygen atom corresponds to 2.534 Å (Figure 7).
Figure 6. Representation of the intermolecular interactions of the C1–H1⋯Cl1 atoms in the crystalline packing of dichloridodioxido-[(4,7-dimethyl)-1,10-phenanthroline]molybdenum(VI).
Figure 7. Representation of the intermolecular interactions of the C13–H5⋯O1 atoms in the crystalline packing of dichloridodioxido-[(4,7-dimethyl)-1,10-phenanthroline]molybdenum(VI).
Additionally, the analysis of the structure also indicates the existence of two intramolecular hydrogen bonds (C5–H2⋯O1) that help in the stability of the bonds of the octahedral system formed by the central molybdenum atom. In Figure 8, these interactions are observed in a projection along the a axis.
Figure 8. Representation of the intramolecular interactions between C5–H2⋯O1 atoms in the crystalline packing of dichloridodioxido-[(4,7-dimethyl)-1,10-phenanthroline]molybdenum(VI).
B. π–π interactions
In addition to hydrogen bond interactions, connectivity between the molecules was also determined through π–π interactions. Figure 9 shows the π–π interactions generated between the centroids Cg2⋯Cg2, at distances of 4.623 and 4.569 Å. These interactions are related to the symmetry operations −X, 2−Y, 2−Z, and 1/2−X, 3/2−Y, 2−Z (Table IV), evidencing the formation of chains diagonally along the axes a and c.
Figure 9. π–π interactions of the Cg2 centroids between neighborhood molecules (a) along the axes a (b) along the axes c.
The π–π interaction obtained at a longer distance (d = 4.795 Å) was determined between the centroid of ring 2 (Cg2) and the centroid of ring 3 (Cg3) of its neighboring molecule. This π–π interaction is associated with the X, 2−Y, 2−Z; X, 2−Y, 1/2 + Z; X, 2−Y, −1/2 + Z symmetry operation. and is presented in a projection along the a axis in Figure 10.
Figure 10. π–π interactions between the centroid Cg2 and the centroid Cg3 along the axes a.
C. Hirshfeld surface analysis
The calculated Hirshfeld surface mapped over the normalized contact distance dnorm and the associated two-dimensional fingerprint plots of the dioxo-molybdenum complex (VI) were calculated using CrystalExplorer21 (Spackman et al., Reference Spackman, Turner, McKinnon, Wolff, Grimwood, Jayatilaka and Spackman2021). Hirshfeld surfaces mapped on dnorm use the function of normalized distances from a given point to the closest atom inside (di) and outside (de) the surface (Spackman and Jayatilaka, Reference Spackman and Jayatilaka2009). The blue, white, and red color conventions used for Hirshfeld surfaces show long, van der Waals, and short interatomic contacts, respectively. Figure 11 shows the short-range interactions between the atoms H⋯O (H5⋯O1) and C⋯C (C1⋯C5 y C5⋯C1) with a distance between atoms of 2.534 and 2.890 Å, respectively.
Figure 11. Analysis of the Hirshfeld surface (dnorm surface) of the dichloridodioxido-[(4,7-dimethyl)-1,10-phenanthroline]molybdenum(VI). The boxes show the atoms of the surrounding molecule with which the interaction is generated.
The shape index and the curvature surfaces provide a greater chemical understanding of the packing of the complex. Figure 12(a) presents the shape index obtained, where the interactions of the atoms present in the aromatic rings (π⋯π interactions) of the complex are highlighted. The region, atom, or zone that acts as a donor corresponds to blue color and the acceptor is represented with a red color. It is important to mention that surface shape is only a qualitative analysis of the donor–acceptor interactions. Additionally, in Figure 12(b), the curvature surface is presented from which it is possible to define that there are no flat areas in the structure, which is an indication of the absence of stacking of one structure on top of another in the packing.
Figure 12. Hirshfeld surface analysis of the structure of the dioxo-molybdenum (VI) complex, (a) shape index and (b) curvature.
Finally, Figure 13 presents the fingerprint plots of the contributions corresponding to the most important interactions on the surface of the dichloridodioxido-[(4,7-dimethyl)-1,10-phenanthroline]molybdenum(VI) complex structure. In the crystalline packing, the greatest contributions were obtained by the interactions H⋯Cl/Cl⋯H, H⋯C/C⋯H, H⋯O/O⋯H, and H ⋯H whose percentages are 30.2%, 23.4%, 20.3%, and 16.5%, respectively. Additionally, the C⋯Cl/Cl⋯C, C⋯C, N⋯H/H⋯N contacts contribute in a percentage of 3.3%, 2.3%, and 2.2%, respectively. Some other interactions with contributions to a lesser degree reached a percentage of 1.8%.
Figure 13. Fingerprints and their contributions in percentage for the most relevant interactions on the surface of the dichloridodioxido-[(4,7-dimethyl)-1,10-phenanthroline]molybdenum(VI) complex.
IV. DEPOSITED DATA
The CIF (Chemical Information File) of the compound was deposited with ICDD. The data can be requested from the Managing Editor at pdj@icdd.com.
ACKNOWLEDGEMENTS
This work was financially supported by the Facultad de Ciencias at Universidad Nacional de Colombia by the internal Projects code 57674. NJC appreciates the collaboration from the X-Ray Diffraction laboratory for the support with the analysis of the samples and Geology Department for the access to the EVA and TOPAS programs.
CONFLICTS OF INTEREST
The authors have no conflicts of interest to declare.
