Preparation of model complexes of non-haem oxygen carrier proteins and oxidase enzymes and evaluation of their catalytic behaviour

Elcoate, Curtis Jay (2010) Preparation of model complexes of non-haem oxygen carrier proteins and oxidase enzymes and evaluation of their catalytic behaviour. PhD thesis, James Cook University.

[img] PDF (Thesis) - Submitted Version
Download (5Mb)

Abstract

A primary objective of the work described within this dissertation was to model the structures of the heterogeneic active sites of dinuclear metalloproteins such as haemerythrin with metal complex analogues and to evaluate whether these species exhibited oxidase catalytic activity. Two avenues were adopted in exploring the preparation of such complexes.

The first approach adopted was the attempted merging of two mononuclear copper complexes into a dimer in which the two Cu centres adopt different coordination modes with respect to each other. The methodology involved the reaction between one mononuclear complex that possesses a potential bridging ligand bound terminally to the metal centre (e.g. acetato) and a second monomeric Cu complex with chloro ligands that could either be displaced or precipitated with Ag+. This approach was ultimately unsuccessful in obtaining the desired heterogeneic Cu2 complexes, but numerous mono-, di- and oligomeric Cu complexes were isolated and characterised by single crystal X-ray diffraction and, in some cases, their spectroscopic and magnetic properties were investigated further.

A noteworthy complex that was isolated from these reactions was [{Cu(4,4’- tBu2bipy)}2(μ-OAc)2(μ-DMF)](ClO4)2, which was obtained from reaction mixtures in DMF (used to suppress the lability of Cu). This complex is a rare example of a homometallic transition metal coordination complex in which the metals are bridged by N,Ndimethylformamide. [{Cu(tBu2bipy)}2(μ-OAc)2(μ-DMF)](ClO4)2 crystallises in the orthorhombic space group P212121 (a = 15.16, b = 17.38 and c = 18.87 Å) with four molecules in the unit cell. The magnetic susceptibility of [{Cu(tBu2bipy)}2(μ-OAc)2(μ-DMF)](ClO4)2 was determined in the range 4 – 300 K, and an antiferromagnetic (J = - 38.8 cm-1) exchange coupling found between the Cu2+ centres affording an overall S = 0 ground state. EPR spectra were collected of [{Cu(tBu2bipy)}2(μ-OAc)2(μ-DMF)](ClO4)2 as single crystal, powder and in frozen methanol / ethanol and acetonitrile solution, revealing peaks for the S = 1 excited state and the spin forbidden ΔMs ± 2 transitions. Analysis of the single crystal EPR spectra reveals a strong angular dependence on the S = 1 signal suggesting that the DMF bridge does not contribute to exchange between the Cu2+ ions; the antiferromagnetic exchange mediated by the acetato ligands is the dominant interaction.

Attempts to prepare other triply bridged Cu2 complexes with the 4,4’-di-tert-butyl-2,2’-bipy ligand failed, but [Cu4(μ3-OH)4(tBu2bipy)4](ClO4)4, a distorted cubane, was crystallised. The complex crystallised in the tetragonal space group I 4 2m (a = b = 21.53 Å, c = 11.27 Å) with 2 cubane molecules in the unit cell. The magnetic susceptibility of [Cu4(μ3-OH)4(tBu2bipy)4](ClO4)4 was determined in the 4 – 300 K range. The data were fit to a 3J distorted tetrahedral (rhomboidal-like) coupling model to give J1 = -11.2 cm-1, J2 = -8.1 cm-1, and J3 = +0.6 cm-1 affording an overall S = 0 ground state.

As the direct merging approach proved unsuccessful in generating heteroleptic Cu dimers a new approach to the problem had to be explored. This led to the design and synthesis of five macrocyclic ligands, possessing two coordination pockets differing in the number of the nitrogen donors (2 vs. 3 and 3 vs. 4) in each pocket and separated by meta- or para- phenylene or phenoxy linkers. The notion was that the macrocycle would fix the number of N- donors on each metal enforcing heterogeneous coordination environments on the two metal centres; the remaining coordination sites could be made up by addition of suitable bridging ligands such as carboxylates.

Of all the complexes prepared with these “heterodentate” macrocyclic ligands, only one crystal structure of a metal complex was obtained, and this was with FeIII 2 complex of the bis(phenoxy) bridged macrocycle (M2). [{FeIII 2(M2)}2(μ-O)2]Cl4.6H2O crystallises in the monoclinic P21/c space group (a = 13.13 Å, b = 18.13 Å, c = 13.30 Å and β = 113.4°) with two molecules in the unit cell. The preparation of this complex from iron(II) chloride shows that the Fe(II) complex oxidises in the presence of atmospheric O2. Mass spectral evidence was obtained for a putative peroxo bound intermediate {[Fe2(M2)(O2)]Cl2}, in which one O2 molecule has bound to the {Fe2(M2)} complex, and which converts to the product by approach of a second {Fe2(M2)} complex followed by a second 2 electron reduction of O2 and O-O bond rupture to form the bis(μ-oxo) complex. The magnetic susceptibility of [{FeIII 2(M2)}2(μ-O)2]Cl4.6H2O was determined in the temperature range 4 - 300 K and fits were attempted using a 2J coupling model. While the susceptibility data show the dominant superexchange pathway to be consistent with antiferromagnetism, the fit did not afford the expected S = 0 ground state at 2 K. Magnetisation studies confirmed a non S = 0 ground state and the presence of many near degenerate energy states. It is likely that this is the result of an impurity in the sample.

The third area of synthesis in the current work was the preparation of tripodal tetradentate ligands, based on tris(2-pyridylmethyl)amine, and their Fe and Cu complexes. The key feature in the design of the tripodal ligands was to incorporate pyridyl, phenoxy and / or carboxylato donor groups and therefore replicate the donor sites found in mononuclear non-haem iron oxidase enzymes (NHIO’s). In general the Fe(II) complexes were prepared in situ in catalysis experiments, but one Fe(III) complex, in which the ligand pbpa (that with pyridyl, benzoato, phenoxy and amine donors) was bound was crystallised and its X-ray crystal structure and magnetic properties determined. [Fe2Cl2(pbpa)2] crystallised in the monoclinic space group C2/c with a = 24.14 Å, b = 11.36 Å and c =15.91 Å and β = 111.4º. The complex exists as a dimer in which the carboxylato group of each pbpa ligand bridges to the other Fe atom. Magnetic susceptibility measurements show weak antiferromagnetic coupling between the Fe(III) centres (J = -2.12 cm-1), affording an overall S = 0 ground state, but with a number of excited states lying at low energy.

As the dinuclear O2 carrier proteins, haemerythrin and haemocyanin bear close structural homology to related oxidase enzymes, it was of interest to determine whether model complexes of these species also exhibit oxidase activity. A series of catalysis experiments were conducted; the oxidation of olefins to oxygenated species with H2O2 (e.g., epoxides and diols) were investigated with cyclohexene and cyclooctene. In addition to the above mentioned heterodentate macrocyclic complexes a series of symmetrical macrocyclic and mononuclear complexes designed to mimic the isolated compartments of the macrocycles were investigated. It was found that the heterodentate Fe triflate macrocyclic complexes were superior catalysts (with respect to TONs) relative to both the symmetrical macrocyclic complex counterparts and the open compartmental analogues (that did not afford catalytic TONs). The dominant oxidation products were for cyclohexene, the allylic alcohol, ketone or peroxide, whilst with cyclooctene the epoxide (cyclooctene oxide) was the major oxidation product and cyclooctane- 1,2-diol was a minor product. The results suggest that, at least with Fe, two metals are necessary for catalytic activity and that catalyst asymmetry facilitates catalytic activity. Of the NHIO model complexes prepared in situ, only the Fe(II) triflato complex with pbpa gave rise to catalytic turnovers (TON = 5.7 with cyclohexene and 2.9 with cyclooctene), the other species being insufficiently soluble in acetonitrile to facilitate oxidation. Similar product distributions were obtained to the experiments using macrocycle based catalysts.

Oxidase / hydroxylase catalysis experiments were also carried out using a series of oxo- centred trinuclear metal carboxylate, [M3O(O2CR)6L3]+, complexes of Fe(III), Cr(III) and Mn(III). These complexes were found not to catalyse the oxidation of cyclohexene or cyclooctene by H2O2, but rather to catalyse the oxidation of these alkenes by atmospheric oxygen. With all three metals it was found that significantly higher catalytic turnovers (TONs) were achieved by the benzoato complexes than their acetato analogues ([Fe3O(O2CPh)6Py3]+ gave a TON of 20 for the autoxidation of cyclohexene whilst with [Fe3O(O2CCH3)6Py3]+ this was 3.5). Furthermore, in the case of cyclohexene, [Mn3O(O2CPh)6Py3]+ gave rise to cyclohexene oxide (epoxide) rather than the normally preferred allylic autoxidation products obtained with the Fe and Cr analogues, and the complexes already discussed above. Furthermore, [Fe3O(O2CPh)6Py3]+ was found to slowly facilitate oxidation of cyclohexane in the presence of atmospheric oxygen.

Item ID: 14927
Item Type: Thesis (PhD)
Keywords: inorganic chemistry, iron, copper, heterodentate macrocycles, alkenes, catalysis, coordination chemistry, metalloproteins, ferromagnetism, chemical synthesis, bioinorganic chemistry
Additional Information:

Data appendices are not available through this repository.

Date Deposited: 12 Nov 2010 04:45
FoR Codes: 03 CHEMICAL SCIENCES > 0302 Inorganic Chemistry > 030201 Bioinorganic Chemistry @ 34%
03 CHEMICAL SCIENCES > 0302 Inorganic Chemistry > 030207 Transition Metal Chemistry @ 33%
03 CHEMICAL SCIENCES > 0306 Physical Chemistry (incl Structural) > 030606 Structural Chemistry and Spectroscopy @ 33%
SEO Codes: 97 EXPANDING KNOWLEDGE > 970103 Expanding Knowledge in the Chemical Sciences @ 100%
Downloads: Total: 1948
Last 12 Months: 197
More Statistics

Actions (Repository Staff Only)

Item Control Page Item Control Page