From intermolecular forces, via crystal engineering, to mechanical flexibility in the organic solid-state

Abstract

To better understand the structural influence of competing hydrogen- and halogen-bond interactions, we have investigated the solid-state landscape of a family of amide-substituted pyridines, belonging to four series; (N-(pyridin-2-yl)benzamides (Bz), N-(pyridin-2-yl)picolinamides (2Pyr), N-(pyridin-2-yl)nicotinamides (3Pyr) and N-(pyridin-2-yl) isonicotinamides (4Pyr) functionalized with three different halogen atoms (chlorine, bromine and iodine). We analyzed crystal structures of these sixteen compounds and identified their primary intermolecular interactions. The calculated molecular electrostatic potential (MEP) of the halogen atoms increased in the order of chlorine< bromine<iodine. Halogen bonding, either competitive or complementary (in comparison with the unhalogenated parent) was only shown by the iodinated target and this was consistent with its higher MEP. Next we explored the influence of an unactivated halogen atom in co-crystal assembly with targets from the three series N-(pyridin-2-yl)picolinamides (2Pyr-X), N-(pyridin-2-yl)nicotinamides (3Pyr-X), N-(pyridin-2-yl)isonicotinamides (4Pyr-X) with ten aliphatic dicarboxylic acids. A total of 100 experiments were carried out. The 4Pyr-I target was the only compound to show competitive halogen bonding both in homomeric assemblies as well as in heteromeric co-crystals. To investigate halogen-bond binding preferences, we co-crystallized sixteen targets from four families (Bz, 2Pyr, 3Pyr and 4Pyr) with two well halogen-bond donors 1,4-diiodotetrafluorobenzene(DITFB) and 1,3,5-trifluoro-2,4,6-triiodobenzene(TITFB). A total of 13 co-crystals were analyzed by single-crystal X-ray diffraction (SCXRD). The intermolecular hydrogen bonding present in the homomeric targets (those belonging to Bz, 3Pyr and 4Pyr) were broken to enable co-crystal formation in 9/10 cases. In the three co-crystals from 2Pyr series, where an intramolecular hydrogen bond (HB) is present, a halogen bond (XB) occurred to either a vacant (oxygen or nitrogen) binding site. Overall, the acceptor sites selected by the halogen-bond donors were distributed as follows; N(pyr)=77%, O=C (19%) or [pi] (4%). Next, we deliberately increased the strength of the halogen-bonding sites (Cl, Br and I) of the Bz, 2Pyr, 3Pyr and 4Pyr targets by activating the halogen atoms via an sp-hybridized carbon atom. The analogues ethynyl hydrogen functionalized targets were also synthesized. SCXRD data for fourteen of sixteen compounds were obtained. Upon activation, the MEP of chlorine is in the range of the unactivated iodine atom. Within each series (Bz, 2Pyr, 3Pyr and 4Pyr) the [sigma]-hole values rank in the order of chloroethynyl~iodine<bromoethynyl<iodoethynyl. However, the halogen-bond forming frequency varies in the order of chloroethynyl<iodine~bromoethynyl<iodoethynyl. Chloroethynyl, iodine and bromoethynyl typically formed halogen bonds to [pi] or nitrogen acceptors. The iodoethynyl moiety emerged as the superior XB donor forming halogen bonds 4/4 times, and the interaction was always to a pyridyl nitrogen. The ethynyl hydrogen-bond donor had a MEP between halogens of bromoethynyl and iodoethynyl, and selected a carbonyl oxygen as an acceptor site in all three cases. We next investigated the self-aggregation of N-(pyridin-2-yl)alkylamides (substituted at the 5 position), which may be considered as the alkyl version of the previously explored Bz targets. These molecules can primarily aggregate through chain or dimer formation. We synthesized twenty targets with substituents Cl, Br, I, H or methyl combined with different alkyl chains (acetyl, propyl, butyl, and pentyl). IR and SXCRD were used to assess the mode of assembly and we found that IR data alone could be used to elucidate which of the two primary motifs were present in the solid state. Broadly speaking chains were more common with the two shorter alkyl chains (7/10) and dimers with the two longer alkyl chains (7/10). As a strategy for altering the assembly of compounds carrying shorter alkyl chains, we introduced an iodoethynyl and ethynyl hydrogen groups as potential structural ‘disruptors’. We successfully switched the assembly from chains to dimers 4/4 times. In order to synthesize new (and very rare) examples of ternary co-crystals, we developed assembly strategies that combined NH-N and iodoethynyl moieties to act as recognition sites for a carboxylic acid and a pyridyl nitrogen acceptor site respectively. Two target molecules, N-(5-(iodoethynyl)pyridin-2-yl)acetamide and N-(5-(iodoethynyl)pyridin-2-yl)benzamide, were synthesized and screened against three aromatic acids and either five or seven halogen bond acceptors. A total of 36 attempted syntheses of ternary co-crystals were carried out, and based on IR and DSC data, we identified 29 positive hits corresponding to a remarkable success rate 80%. We also managed to grow a single-crystal of one of these co-crystals which demonstrated that our postulated strategy did indeed deliver the intended binding modes. Finally, we discovered that one of our co-crystals, N-(5-iodopyridin-2-yl) propionamide:benzoic acid displayed remarkable mechanical flexibility, and given the broad interest in understanding how flexibility of crystalline materials is related to structure (packing features), we began to explore the structure-property relationships of similar co-crystals (i.e. formed by N-(pyridin-2-yl)alkylamides and carboxylic acids). A total of ten co-crystals were obtained. The three co-crystals obtained from the target bearing an acetyl chain were brittle, while the seven co-crystals obtained from a target bearing a propyl or butyl chain were flexible. Five of these displayed elastic deformation while two displayed plastic deformation. The three different responses (brittle, plastic, and elastic deformation) to external mechanical stimuli, could be explained by specific intermolecular interactions and packing features in each crystal structure.