Self assembly is common place in nature [1Whitesides, GM; Boncheva, M Beyond molecules: Self-assembly of mesoscopic and macroscopic components Proc. Natl. Acad. Sci. U S A, 2002, 99, 4769-4774.
[http://dx.doi.org/10.1073/pnas.082065899] ] and coordination chemists have provided a new dimension to this natural phenomenon through the driven self assembly of various metal ions with multidentate ligands. For six co-ordinate complexes, the octahedral geometry is overwhelming more common than the trigonal prismatic arrangement. However such arrangements have received intermittent interest in comparison to their octahedral counterparts and recently, they have been investigated for their usefulness to generate molecular assemblies ranging from the macroscale [2Hu, WJ; Liu, LQ; Ma, ML; Zhao, XL; Liu, YA; Mi, XQ; Jiang, B; Wen, K A Trigonal Prismatic Ligand in the Metal-Mediated Self-Assembly of One- and Two-Dimensional Metallosupramolecular Polymers Inorg. Chem, 2013, 52, 9309-9319.] to nano scale [3Company, A; Roques, N; Güell, M; Mugnaini, V; Gómez, L; Imaz, I; Datcu, A; Solà, M; Luis, JM; Veciana, J; Ribas, X; Costas, M Nanosized trigonal prismatic and antiprismatic CuII coordination cages based on tricarboxylate linkers Dalton. Trans, 2008, 37(13), 1679-1682.].

In order to rationalize why specific six-coordinate complexes are biased towards trigonal prismatic character, we generally consider repulsive ligand-ligand interactions, sterochemical predispositions of a ligand as well as the stereo-electronic preferences of metal ions [4Gillum, WO; Huffman, JC; Streib, WE; Wentworth, RAD Trigonal Prismatic Co-ordination in the cis,cis-1,3,5-Tris(pyridine-2 carboxaldimino)cyclohexanezinc(II) Ion J. Chem. Soc, 1969, (15), 843-844.].

Pioneering work of L. Pauling and R. G. Dickinson’s on the study of mineral molybdenite [5Dickinson, RG; Pauling, L The crystal structure of molybdenite J. Am. Chem. Soc, 1923, 45, 1466-1471.
[http://dx.doi.org/10.1021/ja01659a020] ], in which sulfur atoms surrounds the molybdenum in a trigonal prismatic coordination arrangement, initially invoked interest in the subject. Consequently, a range of analogous compounds were investigated and shown to exhibit TP or pseudo TP arrangements [6(a) Eisenberg, R; Ibers, JA Trigonal Prismatic Coordination. The Molecular Structure of Tris(cis-1,2-diphenylethene-1,2-dithiolato)rhenium J. Am. Chem. Soc, 1965, 87, 3776-3778.(b)Eisenberg, R; Ibers, JA Trigonal Prismatic Coordination. The Crystal and Molecular Structure of Tris(cis-1,2-diphenylethene-1,2-dithiolato)rhenium Inorg. Chem, 1966, 5, 411-416., 7(a) Eisenberg, R; Stiefel, EI; Rosenberg, RC Six-Coordinate Trigonal-Prismatic Complexes of First-Row Transition Metals J. Am. Chem. Soc, 1966, 88, 2874-2876.
[http://dx.doi.org/10.1021/ja00964a061] (b) Eisenberg, R; Gray, HB Trigonal-prismatic coordination. Crystal and Molecular Structure of Tris (cis-1,2-diphenylethylene-1,2-dithiolato)vanadium Inorg. Chem, 1967, 6, 1844-1849.
[http://dx.doi.org/10.1021/ic50056a018] ]. Several other structurally rigid and non rigid ligand systems were also reported. The structurally rigid systems utilised a variety of ligands, such as the bimacro-cyclic ligands, bicapped-TRENCAM (BCT) and bicapped-TPTCAM (BCTPT) [8Karpishin, TB; Stack, TDP; Raymond, KN Octahedral versus trigonal prismatic geometry in a series of catechol macrobicyclic ligand-metal complexes J. Am. Chem. Soc, 1993, 115, 182-192.
[http://dx.doi.org/10.1021/ja00054a025] ], N,N,N’,N’-tetramethyle-thylenediamine (tmeda)[9Morse, PM; Girolami, GS Are d⁰ ML₆ complexes always octahedral? The x-ray structure of trigonal-prismatic [Li(tmed)]₂[ZrMe₆] J. Am. Chem. Soc, 1989, 111, 4114-4116.
[http://dx.doi.org/10.1021/ja00193a061] ], (py)₃tach, (py)₃tame and clathro-chelates [10(a) Parks, JE; Wagner, BE; Holm, RH Three-dimensional macrocyclic encapsulation reactions. I. Synthesis of six-coordinate complexes with nonoctahedral stereochemistry J. Am. Chem. Soc, 1970, 92, 3500-3502.(b) Parks, JE; Wagner, BE; Holm, RH Three-dimensional macrocyclic encapsulation reactions. II. Synthesis and properties of nonoctahedral clathro chelates derived from tris(2-aldoximo-6-pyridyl)phosphine and boron trifluoride or tetrafluoroborate Inorg. Chem, 1971, 10, 2472-2478.
[http://dx.doi.org/10.1021/ic50105a021] (c) Churchill, MR; Reis, AH Structural studies on clathro chelate complexes. I. Trigonal-prismatic coordination of d8 nickel(II) in crystalline [fluoroborotris(2-aldoximo-6-pyridyl)phosphine]nickel(II) tetrafluoroborate Inorg. Chem, 1972, 11, 1811-1818.
[http://dx.doi.org/10.1021/ic50114a014] ], to name but a few. More recently, tris(pyrazolyl)borate (Tp^py) based ligand systems were coordinated with Co (II) by Ward and McCleverty et al. [11Paul, RL; Amoroso, AJ; Jones, PL; Couchman, SM; Reeves, ZR; Rees, LH; Jeffery, JC; McCleverty, JA; Ward, MD Effects of metal co-ordination geometry on self-assembly: a monomeric complex with trigonal prismatic metal co-ordination vs. tetrameric complexes with octahedral metal co-ordination J. Chem. Soc, 1999, 28(10), 1563-1568.], and the trigonal prismatic conformation was observed, however other metal ions yielded tetrameric complexes containing octahedrally co-ordinated metals.

Recently, we have investigated the metal driven self assembly of a six-coordinate tripodal ligand, tris(2,2’-bipyrid-6-yl)methanol with series of transition metals [12Knight, JC; Alvarez, S; Amoroso, AJ; Edwards, PG; Singh, N A novel bipyridine-based hexadentate tripodal framework with a strong preference for trigonal prismatic co-ordination geometries Dalton. Trans, 2010, 39(16), 3870-3883.
[http://dx.doi.org/10.1039/b924651g] ]. Furthermore, we have focussed our attention in investigating the relationship between octahedral and trigonal prismatic character while varying d electron configuration. It was established that unlike the monoanionic Tp^Py, the bipy based ligand naturally favours monomeric complexes as the formation of tetrameric complexes will induce unfavourable electrostatic repulsion. It was concluded that configurationally restrained ligand framework was responsible for enforcing this comparatively rare geometry with respect to an octahedron. In this work we wished to investigate the solid state and solution state structure of the ligand without a metal ion co-ordinated to it. Single crystals of this ligand have been isolated for the first time in the protonated form as a perchlorate salt. We have determined the structure of this species, as well as obtaining the ¹H NMR spectra of the ligand and its protonated form. These finding will be discussed below.

All materials were of reagent grade and used as received from commercial sources. Solvents were distilled before use. Ligand L¹ (Fig. 1) was prepared as described in our previous work [12Knight, JC; Alvarez, S; Amoroso, AJ; Edwards, PG; Singh, N A novel bipyridine-based hexadentate tripodal framework with a strong preference for trigonal prismatic co-ordination geometries Dalton. Trans, 2010, 39(16), 3870-3883.
[http://dx.doi.org/10.1039/b924651g] ]. L¹ (50 mg, 0.10 mmol) was dissolved in methanol (15 mL) and then added dropwise to an aqueous solution of praseodymium(III) perchlorate (110 mg, 101 mmol; 40 wt % aqueous solution). The solution was stirred for 20 minutes and left at 4°C for 2 days. A light brown precipitate was filtered off and the remaining filtrate was left to slowly evaporate at room temperature, yielding rectangular shaped crystals suitable for X-ray diffraction (30%). ¹H NMR (400 MHz; CD3CN): 8.63 (d, 3H, J = 8.64 Hz, py-H8), 8.40 (d, 3H, J = 8.53 Hz, py-H4), 8.35 (d, 3H, J = 8.5 Hz, py-H5), 8.13 (t, 3H, J = 8.38 Hz, py-H3), 8.05 (d, 3H, J = 8.35 Hz, py-H2), 8.00 (t, 3H, J = 8.32 Hz, py-H6), 7.55 ( t, 3H, J = 8.09 Hz, py-H7).

Fig. (1) Ligand L 1 showing the proton numbering.

All single crystal X-ray data was collected at 150 K on a Bruker/Nonius Kappa CCD diffractometer using graphite monochromatic Mo-Kα radiation (λ = 0.71073 Å), equipped with an Oxford Cryostream cooling apparatus. Crystal parameters and details of the data collection, solution and refinement are presented in Table 1. The data was corrected for Lorentz and polarisation effects and for absorption using SORTAV [13Blessing, RH An empirical correction for absorption anisotropy Acta. Crystallogr. A, 1995, 51, 33-38.
[http://dx.doi.org/10.1107/S0108767394005726] ]. Structure solution was achieved by direct methods (Sir- 92 program system [14Altomare, A; Cascarano, G; Giacovazzo, C; Guagliardi, A Completion and refinement of crystal structures with SIR92 J. Appl. Crystallogr, 1993, 26, 343-350.
[http://dx.doi.org/10.1107/S0021889892010331] ] and refined by full-matrix least-squares on F² (SHELXL-97 [15Sheldrick, GM SHELXL-97: Program for the Refinement of Crystal Structures University of Göttingen: Göttingen, Germany, 1997. ] with all non-hydrogen atoms assigned anisotropic displacement parameters. Hydrogen atoms attached to carbon atoms were placed in idealised positions and allowed to ride on the relevant carbon atom. In the final cycles of refinement, a weighting scheme that gave a relatively flat analysis of variance was introduced and refinement continued until convergence was reached. Molecular structures in the figures were drawn with ORTEP 3.0 for Windows (version 1.08) [16Farrugia, LJ Ortep-3 for Windows - A Version of ORTEP-III with a Graphical User Interface (GUI) J. Appl. Crystallogr, 1997, 30, 565.
[http://dx.doi.org/10.1107/S0021889897003117] ]. Crystallographic data for 1 in CIF format have been deposited with the Cambridge Crystallographic Data Centre (CCDC 783585). Copies of these later data may be obtained free of charge via www.ccdc.cam.ac.uk/data_request/Cif, by emailing data_request@ccdc.cam.ac.uk, or by contacting the CCDC, 12 Union Road, Cambridge CB2 IEZ, UK; Fax: +44-1223-336033.

Table 1

Crystal data and structure refinement for 1.

The crystal structure of 1 is depicted in Fig. (2). The ligand consists of three bipy groups linked via a methine bridge. One of these bipy groups is monoprotonated and is held in a cisoid configuration by an intramolecular hydrogen bond (Table 2). The dihedral angle between the least-squares planes of these two pyridyl units is just 9.27°. The two remaining bipy groups are not configurationally restrained by any such intramolecular interactions and each exhibits a transoid configuration around the interannular bond. Accordingly, the dihedral angle for the N3/N4-containing bipy group is 22.27° and for the N5/N6-containing bipy group this value is larger at 36.54°. The three bipy arms adopt an oblique configuration (Fig. 3), spreading further apart as the sterical influence conferred from the methine bridge is reduced. The C-N and mean C-C bond lengths for the protonated pyridine are 1.355(3) Å and 1.384(3) Å respectively. The corresponding values for the neutral pyridyl units are statistically identical at 1.344(3) Å and 1.385(3) Å. These bond lengths are in excellent agreement with literature values for both 2,2’-bipyridine [17Merritt, Jr LL; Schroeder, ED The Crystal Structure of 2,2'-Bipyridine Acta. Crystallogr, 1956, 9, 801-804.
[http://dx.doi.org/10.1107/S0365110X56002175] ] and 2,2’-bipyridinium (bis)perchlorate [18Ma, G; Ilyukhin, A; Glaser, J 2,2'-Bipyridinium bis(perchlorate) Acta. Crystallogr. C, 2000, 56, 1473-1475.
[http://dx.doi.org/10.1107/S0108270100012452] ].

Fig. (2) Perspective view of the asymmetric unit. Displacement ellipsoids are shown at the 50% probability level. H atoms are represented by circles of arbitrary size. Three water molecules within the lattice have been excluded for clarity.

Fig. (3) Highlighting the distorted trigonal prism formed between pyridyl centroids in which the upper triangular face in close proximity to the methine bridge is significantly smaller than the lower triangle formed by the more distant centroids.

Table 2

H-bonding geometry (Å, º) for 1.

Another feature of interest is the orientation of the pyridine rings directly bonded to the bridgehead carbon of the tripod (the proximal pyridine donors). The crystallographic data for tris(2-pyridyl)methanol [19Keene, FR; Snow, MR; Tiekink, ERT Tris(2-pyridyl)methanol Acta. Crystallogr. C, 1988, 44, 937-938.
[http://dx.doi.org/10.1107/S0108270188000411] ] shows these rings positioned such that all three lone pairs on the nitrogens are pointing away from each other.

However, when protonated, two rings interact with the proton, leading to a N-N distance of 2.656 Å [20Boggess, RK Monoprotonated perchlorate salt of tris(2-pyridyl)methanol Acta. Crystallogr. C, 1992, 48, 1327-1329.
[http://dx.doi.org/10.1107/S0108270191013860] ]. In contrast, protonated tris(2,2’-bipyrid-6-yl) methanol has all three lone pairs of the proximal pyridine donors pointing towards the proton. In addition, one distal pyridine is also hydrogen bonding to the proton. This distal pyridine (N2-C7/C11) is part of the bipyridyl unit to which the proton is most strongly associated. The N-N distances between the pyridinium moiety (N1-C2/C6) and the two other proximal pyridines is 2.786(3) Å (N1-N5) and 2.852(3) Å (N1-N3), while the remaining N-N distance between these proximal donors is 3.048(3) Å (N3-N5). The N-N distance between the proximal pyridinium unit and its distal pyridine is 2.667(3) Å (N1-N2).

The molecular packing is consolidated by various intermolecular interactions, such as π-stacking between N1-C2/C6 and N2-C7/C11 (symmetry code: 2-x, 1-y, -z; Fig. 4) and several H-bonds involving the perchlorate counterion and three water molecules. The alcohol group (O1) is also acting as an H-bond donor to a water molecule (O8) within the lattice.

Fig. (4) Fragment of the title structure, viewed along the c-axis showing the π-stacking interactions (gold dotted lines) between symmetrically related molecules (symmetry code: 2-x, 1-y, -z) which support the molecular packing. Hydrogen atoms have been excluded for clarity.

The ¹H NMR 2-D COSY spectra of both the mono-protonated and neutral forms of 1 were measured in order to elucidate the coupling behaviour within the aromatic region and thus unambiguously assign the resonances (Fig. 5 and Fig. 6, respectively). First, one can note that the addition of a single proton to the tripod structure does not destroy the C₃ symmetry of the molecule and all bipyridyl moieties are still equivalent. This would be consistent with fluxional behaviour of the proton, faster than the NMR timescale. Also, whilst the proton resonance frequencies for both the mono-protonated and neutral ligand exhibit similar ordering, those of the protonated form are shifted slightly downfield as one would expect from the deshielding effect of adding a proton to the pyridine ring [21Proton NMR data for the neutral ligand (400 MHz; CDCl3): δ 8.55 (d, 3H, J = 7.8 Hz), 8.28 (d, 3H, J = 7.5 Hz), 8.12 (d, 3H, J = 7.2 Hz), 7.89 – 7.72 (m, 6H), 7.67 (t, 3H, J = 7.5 Hz), 7.21 (t, 3H, J = 7.2 Hz), 5.21 (s, 1H), ].

Fig. (5) A COSY map showing the expanded aromatic region of the neutral unprotonated ligand taken at 293K in CD3CN.

Fig. (6) A COSY map showing the expanded aromatic region of the protonated ligand taken at 293K in CD3CN.

In both cases, there exists strong coupling between vicinal protons around each of the pyridine groups. This allow us to unambiguously group the resonances for each pyridyl ring ((H_2, H_3, H_4,) vs (H_5, H_6, H_7, H₈)), when considering the cross peaks. By then assigning the peak which the highest ppm to H₈ (ortho to the nitrogen atom), this allows the full assignment of all peaks. This is particularly useful in discriminating between H₄ and H₅ as well as H₃ and H₆. This allows us to determine that the relative ordering of all resonances is unchanged between the protonated and free ligand and suggests there is no preferencial localisation of the proton on one of the two different pyridyl rings.

The protonated ligand also reveals evidence of long range coupling between the two pyridines, which are in a cisoid arrangement, specifically through H₄ and H₅ (labelled a).

In conclusion, we have shown that the solid state structure of the protonated form of the ligand L¹ has a structure consisting of the three bipyridyl moieties connected to a central carbon. The pyridyl rings connected directly to the central carbon all having their nitrogen donor pointing to the centre of the molecule, perhaps suggesting they all interact with the proton. The other three nitrogen donors on the outermost pyridines vary in their behaviour, with two having a transoid arrangement within the bipyridine moieties and one being cis. However, this asymmetric arrangement is not observed in the solution state as ¹H NMR identifies C₃ symmetry in the complex. COSY NMR has allowed us to unambiguously assign the proton resonances. Interestingly, the similar ordering of the resonances, in the free ligand compared to its protonated form, suggests little difference in the relative electron density within a compound.

The authors confirm that this article content has no conflict of interest.