Development of Catalysts for Green Chemistry Transformations

Abstract

The development of new ligands is crucial to the discovery of effective catalysts for homogeneous reactions. By synthesising ligands with new donor groups and then coordinating these ligands to appropriate metal centres, new complexes may be developed for the catalysis of specific reactions. In this thesis, investigations into the synthesis of a new macrocyclic ligand that contains pyridinium amides as the primary donor groups are reported. Studies of the spectroscopic, structural, and catalytic properties of various transition metal complexes of this new ligand are also reported. Although the first pyridinium amide was synthesised over a century ago, the coordination chemistry of pyridinium amides has only recently been investigated. While some of these complexes have been shown to be effective catalysts for a number of reactions, catalysis by these complexes remains relatively little explored. The structure of the target macrocyclic pyridinium amide ligand, H2Lm ((5E,18aE)-8,12,12,16- tetramethyl-8,10,14,16-tetrahydro-11H-benzo[e]dipyrido[3,4-b:4’,3’-h][1,4,7,10]tetraazacyclotridecine- 11,13(12H)-dione), is related to the structure of tetraamido macrocyclic ligands (TAMLs), porphyrins, and dibenzotetramethyltetraaza[14]annulenes (TMTAAs). Like these ligands, the Lm 2- ligand is expected to donate strongly to metal centres. While it has been shown that metal complexes of TAML, porphyrin, and TMTAA ligands are effective catalysts for oxidation reactions and for small molecule activation reactions, metal-pyridinium amide complexes have not been investigated as catalysts for these reactions. Consequently, metal complexes of the Lm 2- ligand were studied in this thesis as potential catalysts for the oxidation of dye substrates by hydrogen peroxide and as potential catalysts for the activation of small molecules (such as dihydrogen, carbon monoxide, and acetylene) to produce organic products. Upon deprotonation, H2Lm was found to coordinate to a number of transition metals to give complexes that include FeIII(Lm)Cl, CoIII(Lm)Br, NiII(Lm), PdII(Lm), [RuIII(Lm)]2[BF4]2, and Na[[RhII(Lm)]2Cl], which were all fully characterised. The data obtained strongly indicates that [RuIII(Lm)]2[BF4]2 and Na[[RhII(Lm)]2Cl] have direct unsupported metal-metal bonds. A paramagnetic copper complex tentatively formulated as CuIII(Lm)(OH)(H2O) was also obtained. The complexes FeIII(Lm)Cl and CoIII(Lm)Br were paramagnetic, whereas NiII(Lm), PdII(Lm,), [RuIII(Lm)]2[BF4]2, and Na[[RhII(Lm)]2Cl] were diamagnetic. The diamagnetic nature of the latter two complexes could be explained on the basis of simplified orbital diagrams for the metalmetal bonding in these complexes. An acyclic pyridinium amide ligand, H2La (N1,N2-bis(1-methyl-2-(p-tolylimino)-1,2- dihydropyridin-3-yl)oxalamide), was also synthesised and was coordinated to palladium, iron and cobalt. Although the palladium complex, PdII(La), was fully characterised, high resolution mass spectrometry suggested that a mixture of iron(III) products and cobalt(III) products was formed in the reaction of H2La with iron(II) and cobalt(II) salts, respectively. These mixtures were not successfully separated. The high resolution positive ion mass spectra of these products agreed with formulations of [(MIII)n(La)n+1 + xH+](n+x-2)+, where n = 1 to 4 and x = 1 to 4. FeIII(Lm)Cl and CoIII(Lm)Br were selected for study as potential catalysts for dye oxidation reactions with hydrogen peroxide because it has been shown that related FeIII-TAML and CoIIITAML complexes are highly effective catalysts for this reaction. Since Orange II dye has been widely used as a substrate in these types of studies, the ability of FeIII(Lm)Cl and CoIII(Lm)Br to behave as oxidation catalysts was determined by monitoring the decolourisation of Orange II dye in the presence of hydrogen peroxide using time-dependent UV-visible absorption spectroscopy. The initial rate of dye oxidation was obtained from these spectra and by varying the reaction conditions, reaction kinetics could be estimated. The typical conditions used in these experiments were: 25 °C, 0.25-10 μmol L-1 FeIII(Lm)Cl or CoIII(Lm)Br, 45 μmol L-1 Orange II dye, and 0.25-10 mmol L-1 hydrogen peroxide in pH 7.5-10.8 aqueous solutions containing 0.01 mol L-1 carbonate or phosphate buffers. However, even when the ratio of the concentrations of the catalyst to the Orange II dye was relatively high (up to 1:4.5), no significantly acceleration of dye bleaching was observed compared to the blank with no metal compound present. This indicates that FeIII(Lm)Cl and CoIII(Lm)Br are very poor oxidation catalysts under these conditions. In contrast, both of these complexes were found to catalyse the disproportionation of hydrogen peroxide to water and dioxygen. Na[[RhII(Lm)]2Cl] was investigated as a catalyst for the activation of small molecules, because it has been reported that related rhodium(II)-porphyrin and rhodium(II)-TMTAA dimers react in interesting ways with a wide range of small molecules, such as carbon monoxide, dihydrogen, acetylene, alkyl halides, and triarylphosphines. Na[[RhII(Lm)]2Cl] and selected derivatives of this compound were found to react stoichiometrically but not catalytically with most of these molecules. A number of organometallic and coordination complexes were synthesised and characterised through these reactions including Na[[RhII(Lm)]2(Cl)(pyridine)], Na[[RhII(Lm)]2(Cl)(4-picoline)], [RhIII(Lm)(PPh3)][PF6], Na[RhI(Lm)], RhIII(Lm)Me, RhIII(Lm)Et, RhIII(Lm)Bn, RhIII(Lm)(C(=O)OR), and RhIII(Lm)(CH=CHOR). The OR groups of the latter two complexes depended on the alcohol solvent used for the reaction. Reactivity studies were investigated to obtain some information about possible reaction pathways for the syntheses of [RhIII(Lm)(PPh3)]+, RhIII(Lm)(C(=O)OR), and RhIII(Lm)(CH=CHOR) from the reaction between Na[[RhII(Lm)]2Cl] and triphenylphosphine, carbon monoxide, or acetylene, respectively, in alcohol solvents under aerobic conditions. From these studies, it was postulated that coordination of the reagent (triphenylphosphine, carbon monoxide, or acetylene) to Na[[RhII(Lm)]2Cl] facilitates heterolytic cleavage of the rhodiumrhodium bond. Of the rhodium(I) and rhodium(III) complexes formed in this process, it is the rhodium(III) species that coordinates the reagent, and in the case of carbon monoxide and acetylene, activates it to undergo further reaction with the alcohol solvent. In a parallel process, the rhodium(I) complex is oxidised under the aerobic conditions to form more of the rhodium(III) complex, which also coordinates reagent. In this way, more of the same ultimate product is formed. This proposed reaction pathway explains the observation that essentially only one product was obtained in very good yield from each of these reactions.