A molecular receptor embedded in a 'poor-solvent' receiving phase, such as a fluorous phase, should offer the ideal medium for selective extraction and sensing. The limited solubility of most solutes in fluorous phases enhances selectivity by reducing the extraction of unwanted matrix components. Thus, receptor-doped fluorous phases may be ideal extraction media. Unfortunately, sufficient data do not exist to judge the capability of this approach. The solubilities of very few nonfluorous solutes are known. As far as we are aware, such important quantities as the strength of a hydrogen bond in a fluorous environment are not known. Thus, it is currently impossible to predict whether a particular receptor/solute complex based on a particular set of noncovalent interactions will provide enough thermodynamic stabilization to extract the solute into the fluorous phase. In this work, fluorous carboxylic acids (a carboxylic acid-terminated perfluoropolypropylene oxide called Krytox and perfluorodecanoic acid (PFDA)) were used as receptors and substituted pyridines as solutes to show that the fluorous receptor dramatically enhances the liquid-liquid extraction of the polar substrates from chloroform into perfluorohexanes. The method of continuous variations was used to determine the receptor-pyridine complex stoichiometry of 3:1. The free energies of formation of the 3:1 complexes from one pyridine and 3/2 H-bonded cyclic dimers of the fluorous carboxylic acid are -30.4 (Krytox) and -37.3 kJ mol-1 (PFDA). The free energy required to dissociate the dimer in perfluorohexanes is +16.5 kJ mol-1 (Krytox). The crystal structure of the complex showed a 1:1 stoichiometry with a mixed strong-weak hydrogen-bonded motif. Based on the stoichiometry, crystal structure, and UV and IR spectroscopic shifts, we propose that the 3:1 complex has four hydrogen bonds and the carboxylic acid transfers a proton to pyridine. The resulting pyridinium carboxylate N+H-O- hydrogen bond is accompanied by a weak pyridine ring CH-O bond and is supported by two more carboxylic acid H-bond donors. We estimate that the free energy of formation of this complex from a free acid, pyridine, and a carboxylic acid dimer to be approximately -39 kJ mol-1; this is the first reported hydrogen bond strength in a fluorous environment.

There are only a few studies of noncovalent association in fluorous solvents and even fewer that are quantitative. A full understanding, particularly of stoichiometry and binding strength of noncovalent interactions in fluorous solvents could be very useful in improved molecular-receptor-based extractions, advancements in sensor technologies, crystal engineering, and supramolecular chemistry. This work investigates hydrogen bonding between heterocyclic bases and a perfluoropolyether with a terminal carboxylic acid group (Krytox 157FSH (1)), chiefly in FC-72 (a mixture of perfluorohexanes). In particular, we were interested in whether or not proton transfer occurs, and if so, under what conditions in H-bonded complexes. Continuous variations experiments show that in FC-72 weaker bases (pyrazine, pyrimidine, and quinazoline) form 1:1 complexes with 1, whereas stronger bases (quinoline, pyridine, and isoquinoline) form 1:3 complexes. Ultraviolet and infrared spectral signatures reveal that the 1:1 complexes are molecular (B · HA) whereas the 1:3 complexes are ionic (BH + · A -HAHA). Infrared spectra of 1:3 ionic complexes are discussed in detail. Literature and experimental data on complexes between N-heterocyclic bases and carboxylic acids in a range of solvents are compiled to compare solvent effects on proton transfer. Polar solvents support ionic hydrogen bonds at a 1:1 mol ratio. In nonpolar organic solvents, ionic hydrogen bonds are only observed in complexes with 1:2 (base/acid) stoichiometries. In fluorous solvents, a larger excess of acid, 1:3, is necessary to facilitate proton transfer in hydrogen bonds between carboxylic acids and the bases studied.

No abstract

Porphyrins have found application in a remarkable variety of areas such as sensors, ion selective electrodes, photodynamic therapy, and energy-transfer systems. Here, we demonstrate the extraction of 5,10,15,20-tetraphenylporphyrin (TPhP) and 5,10,15,20-tetra(4-pyridyl)porphyrin (TPyP) into a mixture of perfluorohexanes (FC-72) through noncovalent interactions with Krytox (1), a carboxylic acid terminated perfluoropolyether. We found that 1 transfers two protons to the TPhP tetrapyrrole ring to create the porphyrin dication (H2TPhP2+) in FC-72 while up to six protons are transferred to the TPyP pyridyl and tetrapyrrole nitrogens to create a hexavalent cation macrocycle in the fluorous phase. The total charge on TPyP is controlled by adjusting the concentration of 1 in the fluorous phase. In addition, we observed extraction of ZnTPyP from CDCl3 with 1/FC-72, while ZnTPhP is not extracted by 1/FC-72. We prepared the Zn salt of 1 and found that it extracts (from CDCl3) and metalates TPyP but not TPhP. Competitive binding between the porphyrins and an ethanol cosolvent hinders the extraction of both TPhP and TPyP and inhibits the formation of the TPyP hexacation in FC-72. By controlling the concentration of porphyrin, 1, and ethanol, it is possible to reversibly solubilize TPyP in the fluorous phase through noncovalent interactions between the pyridyl moieties and 1 while leaving the tetrapyrrole ring available to interact with metals or other substrates. In addition, both porphyrins and ZnTPyP are easily recovered from the fluorous phase using commercially available fluorous solid-phase extraction cartridges. Understanding noncovalent interactions in fluorous matrices should lead to development of more robust devices for sensing and energy transfer.

Three fluorous resorcin [4]arenes (1a-c) were synthesized and characterized for their potential application in separation science. A crystal structure of 1b confirmed its ring size and conformation. The fluorous resorcin [4]arenes are soluble in hybrid fluorous solvents. Preliminary studies demonstrated that 1a immobilized in a fluorous bulk membrane is capable of transporting tetramethylammonium cation from an aqueous source phase into an aqueous receiving phase.