Crystallographic coincidence of two bridging species in a dinuclear CoIII ethynylbenzene complex

In the title compound, trans,trans-[μ-(m-phenylene)bis(ethyne-1,2-diyl)]bis[chlorido(1,4,8,11-tetraazacyclotetradecane)cobalt(III)]–trans,trans-[μ-(5-bromo-m-phenylene)bis(ethyne-1,2-diyl)]bis[chlorido(1,4,8,11-tetraazacyclotetradecane)cobalt(III)]–tetraphenylborate–acetone (0.88/0.12/2/4), [Co2(C12H4)Cl2(C10H24N4)2]0.88[Co2(C10H3Br)Cl2(C10H24N4)2]0.12(C24H20B)2·4C3H6O, with the exception of the acetylene and bromine groups, all atomic postitions are the same in the two compounds and are modeled at full occupancy. The CoIII ions are six-coordinate with acetylide and chloride ligands bound to the axial sites and the N atoms from the cyclam rings coordinated at the equatorial positions. N—H⋯O and N—H⋯Cl hydrogen-bonding interactions help to consolidate the crystal packing.


Comment
From a technological standpoint, metallodendrimers are of interest for their unique catalytic and optical properties (Mery & Astruc, 2006;Onitsuka & Takahashi, 2003). A particular subset of metallodendrimers based on ethynylbenzene have been pursued because of their structural rigidity and topological anisotropy. Although a variety of Pt(II)-and Ru(II)-containing dendrimers based on a 1,3,5-triethynylbenzene (H 3 TEB) linkage have been reported (Onitsuka et al., 2004;McDonagh et al., 2003), we are interested in the properties of first row transition metal TEB complexes for potential applications in molecular magnetism (Weyland et al., 1998). For elaboration to higher nuclearity species, the inclusion of an axially coordinated anionic ligand that is poized for substitution is vital.
The synthesis of these macromolecules can be accomplished by divergent or convergent pathways; regardlesss, each strategy hinges upon the isolation of structurally characterized "building blocks" prior to dendrimer assembly. The preparation of complexes that contain first-row metals is synthetically challenging because of their high kinetic lability relative to their second and third row counterparts. In that respect, our initial synthetic targets contain Co III because of its relative inertness.
The combination of H 3 TEB with two equivalents of trans-[(cyclam)CoCl 2 ]Cl produces the dinuclear Co III arylacetylide-bridged complex 1 in good yield ( Figure 1). Initial refinement attempts on high quality X-ray data did not converge satisfactorily, as the third aromatic substituent showed apparent disorder of the alkynyl group. However, structure refinement proceeds smoothly if compositional disorder is invoked. The accepted method for the preparation of 1,3,5-triethynylbenzene involves Sonogashira coupling between 1,3,5-tribromobenzene and trimethylsilylacetylene (Weber et al., 1988). On one occasion, following the protocol resulted in a batch of TEB containing a sizeable amount of 1-bromo-3,5-diethynylbenzene (Robinson et al., 1998), indicating incomplete substitution. The impurity was carried through several purification steps, eventually affording a mixture of the ethynyl-(1) and bromo-(2) substituted complexes. The crystal structure revealed that both components of the ligand mixture were incorporated into metal complexes and the atomic sites were superimposed.
During structure refinement, the compositional disorder at the aromatic 1 position was modeled with a free variable. Final site occupancy factors indicate that the two ligand components are present in an 88:12 1:2 ratio. This compares favorably with subsequent 1 H NMR analysis of the batch of "H 3 TEB" ligand, which shows resonance integrations in an 87:13 H 3 TEB:H 2 BrTEB ratio.
The molecular structure of the complex cations in 1 and 2 are shown in Figure 1. Each pseudo-octahedral Co III center coordinates four nitrogen atoms from the cyclam rings at the equitorial positions with an average Co-N bond length of 1.9767 (11) Å, which is only slightly longer than the corresponding bond length from the reported structure of trans-[(cyclam)CoCl 2 ]Cl (1.9741 (12); Ivaniková et al., 2006). Anionic chloride and acetylide ligands occupy the axial Co III coordination sites with average metal-ligand distances of 2.3076 (4) and 1.8770 (14) respectively. The former bond length is significantly longer than the average Co-Cl distance in trans-[(cyclam)CoCl 2 ]Cl, suggesting that the arylacetylide ligand supplementary materials sup-2 imparts a stronger trans influence than chloride. The cationic charge is balanced by the presence of two tetraphenylborate anions, and the asymmetric unit includes four molecules of acetone. Figure 2, the crystal packing in 1 and 2 is influenced by several weak hydrogen bonding interactions. Notably, the complex cations experience a dimeric interaction through pairwise N-H···Cl contacts with a complex in a neighboring unit cell. Furthermore, three of the four acetone molecules participate in hydrogen bonds through the cyclam N-H groups.

Shown in
In summary, a mixture of H 3 TEB and 1-bromo-3,5-diethynylbenzene combined with trans-[(cyclam)CoCl 2 ]Cl to yield a co-crystallized mixture of 1 and 2. The compounds are superimposed in the solid state with the exception of the 5-position acetylene and bromine groups. Using a free variable to model the compositional disorder, we conclude that the two compounds are present in a 88:12 ratio. The first coordination sphere for each Co III ion includes an axially replaceable chloride ligand, which is a necessary condition for future metallodendrimer assembly. This result exemplifies the key role of crystallographic analysis in organometallic synthesis development.
Triethylamine was purchased from Sigma-Aldrich and was distilled prior to use. Preparation of 1 and 2: Triethylamine (0.34 ml, 2.42 mmol) was added to a 100 ml round-bottomed flask containing a green methanolic (10 ml) solution of [(cyclam)CoCl 2 ]Cl (233 mg, 0.637 mmol) and freshly sublimed mixture (45.5 mg) of 1,3,5-triethynylbenzene (87% by 1 H NMR) and 1-bromo-3,5-diethynylbenzene (13% by 1 H NMR). The flask was fitted with a condenser tube and the solution was refluxed for 24 h, during which time the solution turned orange-brown. The solvent was removed by rotary evaporation, and the resulting red-brown residue was washed with 10 ml of absolute ethanol, causing an orange solid to precipitate. The solid was isolated by filtration, washed with ethanol (3 × 3 ml) and diethyl ether (3 × 3 ml) and dried in air to afford 92.1 mg of an orange solid. The orange solid was dissolved in methanol (10 ml) and a solution of excess sodium tetraphenylborate in methanol (5 ml) was added, causing a salmon-colored solid to precipitate.

Refinement
Displacement parameters for all non-hydrogen atoms were refined anisotropically. Hydrogen atoms were assigned to ideal positions and were refined using a riding model where the displacement parameters were set at 1.2 times those of the attached carbon or nitrogen atoms (1.5 times for methyl protons). Fig. 1. Structure of the superimposed complex cations present in 1 and 2 with atomic numbering scheme and thermal ellipsoids rendered at 40° probability. Orange, green, blue, gray, and red ellipsoids represent cobalt, chlorine, nitrogen, carbon, and bromine atoms respectively. With the exception of the acetylenic hydrogen (H1A, represented by a gray shaded sphere), hydrogen atoms have been omitted for clarity. Fig. 2. Hydrogen bonding interactions present in the solid state structures of 1 and 2. Thermal ellipsoids are rendered at 40% probability. Red ellipsoids represent oxygen atoms. Otherwise, the color scheme is identical to that found in Figure 1. Tetraphenylborate anions, the acetone molecule that includes O4 (which does not participate in H-bonding), the bromine substituent present in 2, and hydrogen atoms that do not participate in H-bonding have been omitted.

Special details
Experimental. Although we cannot explain the source of the Hirshfield tests that give rise to the B-and C-level alerts, there is no evidence of substitutional disorder at the atomic sites mentioned in the alerts. The reason for the presence of a non-integer number of atoms is due to substitutional disorder between bromine and acetylene substituents as descibed in the text. Four reflections were omitted from refinement due to beamstop interference. Probable reasons for the missing cusp of data include beamstop interference and data truncation at resolutions higher than 0.70 Å during the initial stages of refinement. The low "solvent" U eq in C88 C91 (the central C atoms in two of the acetone molecules) compared to neighboring atoms cannot be explained by substitutional disorder or incorrect atom type. However, we note that the differences in U eq are relatively minor. The four D-H groups on the cyclam rings do not interact with acceptors. This has been checked and the exception is apparently common for N-H groups. One of the tetraphenylborate anions and one of the acetone molecules do not have their centers of gravity within the unit cell. Since neither molecule is the main species, there is no cause for alarm. The s.u. values for the unit cell angles have been checked, and the fact that all angles have the same s.u. is purely coincedental. The long C(sp2)-C(sp1) bonds noted for C5-C9 and C7-C11 appear to be real. Since these bonds include an aromatic carbon, this may be a false alarm. 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 > σ(F 2 ) is used only for calculating Rfactors(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.