1 2 3b (exo) 3a (endo) (b) (a) X-Ray structure of the 1 : 1 complex of a tripodal receptor and cis-cyclohexane-1,3,5-tricarboxylic acid Pablo Ballester, a Antoni Costa, a Pere M. Dey` a, a Ghislain Deslongchamps, b Daniel Mink, b Andreas Decken, b Rafael Prohens, a Salvador Tom` as a and Manuel Vega a a Departament de Qu´ ımica, Universitat de les Illes Balears, 07071 Palma de Mallorca, Spain b Department of Chemistry, University of New Brunswick, Fredericton, N.B., E3B 6E2, Canada The X-ray crystal structure of the 1:1 complex of a tripodal abiotic receptor and cis-cyclohexane-1,3,5-tricarboxylic acid is reported; the 1:1 complex is devoid of C 3 -symmetry and packs into a multi-columnar self-assembly. Two years ago, some of us reported 1 the synthesis and binding properties of the tripodal receptor 1, designed to bind tri- carboxylic acids. 2 The receptor 1 shows high selectivity 3 for binding cis-cyclohexane-1,3,5-tricarboxylic acid 2 (Fig. 1) and forms a very stable complex with it in chloroform solution. Molecular modeling studies 4 suggested that two different C 3 -symmetric geometries, 3a and 3b, were possible for the binding of triacid 2 by receptor 1. In both cases, six intermolecular hydrogen bonds could form between the three carboxylic acids of the guest and the three amidopyridine ‘arms’ of the receptor. ROESY 51 H NMR studies of the complex in CHCl 3 revealed intermolecular contacts that agreed with a predominance in solution of the ‘endo’† supramolecular complex 3a. Slow diffusion of cyclohexane into a 1 : 1 solution of 1 and 2 in CH 2 Cl 2 produced colourless plates that were relatively unstable when removed from the mother liquor. After numerous attempts to obtain a suitable dataset, the X-ray crystal structure was solved,‡ revealing that 2 is docked into the receptor via six intermolecular hydrogen bonds (Fig. 2). This structure shows a complex geometry which is in general agreement with the solution structure 3a proposed previously based on NMR data. However, the triacid complex does not possess the anticipated C 3 -symmetry axis. The O···N distances for all six hydrogen- bonded interactions are different, ranging from 2.612 (0.010) to 2.739 (0.010) Å (N···HO) and from 2.854 (0.010) to 2.910 (0.010) Å (O···HN). The dihedral angles about the carboxy– cyclohexyl single bonds vary by as much as 94°. The asymmetry is also evident in the three propyl chains, where two are in different anti conformations while the third chain is gauche. Deviation from C 3 -symmetry undoubtedly arises from the presence of disordered cyclohexane, which was found to be present at an occupancy less than 1 in the difference map. Further inspection of the crystal packing reveals a columnar motif. The 1 : 1 complexes stack on top of each other in an alternating top-to-top and bottom-to-bottom fashion and form columnar supramolecular arrangements (Fig. 3). The bottom- to-bottom interaction involves p–p interaction 6 between the two central benzene rings of the receptor in a perfect face-to- face geometry separated by ca. 3.9 Å. The absolute stereo- chemistry of the complexes alternates within a single column. Thus, any two adjacent complexes within a column (whether top-to-top or bottom-to-bottom) form a perfectly centrosym- metric dimer, a direct consequence of the C 2 /c space group in which the complex crystallizes. The columnar structures pack in parallel fashion to form palisade-like sheets, which are stabilized by antiparallel p–p Fig. 1 Possible binding geometries for the 1 : 1 complex of 1 and 2 Fig. 2 X-Ray structure of 1 : 1 complex between 1 and 2: (a) side view (only polar hydrogens in calculated positions are shown, dotted lines indicate intermolecular hydrogen bonds) and (b) top view of a CPK representa- tion Chem. Commun., 1997 357 Published on 01 January 1997. Downloaded on 27/10/2014 14:03:53. View Article Online / Journal Homepage / Table of Contents for this issue