Some aspects of the chemistry of iron sulphur complexes

A Mössbauer Effect study of a number of complexes of iron with sulphur containing ligands has been carried out, in order to further the understanding of iron-sulphur chemistry and hence eventually provide suitable model complexes for the study of the non-haem iron proteins. The complexes investigated have been considered in three main sections, some dialkyldithiocarbamate complexes, some complexes that have been proposed as models for the non-haem iron proteins, and some complexes of some 2-substituted pyridines. The iron(III) N,N dialkyldithiocarbamates exhibit anomalous magnetic properties which have been ascribed by other workers to a high-spin, low-spin equilibrium. All the iron(III) N,N dialkyldithiocarbamates that we have studied showed a single Mössbauer spectrum at all temperatures, so that the rate of exchange between the high-spin and low-spin states must be greater than 1.5 × 10+7 sec.−1. The isomer shifts observed were large for low-spin ferric ions and suggest that the metal s-electrons are screened from the nucleus by electrons donated from the ligands. The isomer shifts and quadrupole splittings for all the complexes, except the high-spin iron(III) pyrrolidyldithiocarbamate, exhibited anomalous dependences on temperature due to the existence of the spin equilibrium. Iron(III) pyrrolidyldithiocarbamate exhibited long electron spin relaxation times at 4.2°K. The other compounds, which are low-spin at low temperatures, also showed relaxation effects, though with shorter times. The magnetic hyperfine interactions in the Mössbauer spectrum of iron(III) N,N dimethyldithiocarbamate gave information on the nature of the ground state of the complex, which was found to consist of almost equal amounts of dxz and dyz holes, with slightly less of dxy. Mössbauer Effect and epr measurements on samples of this complex diluted in cobalt(III) N,N dimethyldithiocarbamate confirmed this result. Estimates for the orbital separations were obtained as Δdxz, dyz = 1.2 cm.−1 and Δdyz, dxy = 91 cm.−1 The dependence of the Mössbauer spectrum of iron(III) pyrrolidyldithiocarbamate upon temperature and magnetic field has been interpreted in terms of a model with a large negative zero field splitting of the 6S state. The magnitude of the zero field splitting (4D) has been estimated as −9.3°K. Bis(N,N diethyldithiocarbamato)iron(III) chloride and bis(N,N diisopropyldithiocarbamato)iron(III) chloride exhibited similar Mössbauer spectra to iron(III) pyrrolidyldithiocarbamate at low temperatures. We have confirmed that the former compound is ferromagnetic, while the diisopropyl derivative is paramagnetic with the Kramer's doublet |±3/2〉 lying lowest. Another complex of iron with dialkyldithiocarbamate ligands that has been studied is the formally iron(I) complex, nitrosyliron bis(N,N diethyldithiocarbamate). A high isomer shift was again observed. The quadrupole splitting showed little dependence on temperature and the application of an external magnetic field showed the sign of the major component of the electric field gradient to be positive. Magnetic hyperfine interactions arc observed in large external magnetic fields at low temperatures. The variation of the effective field at the nucleus with H/T followed the Brillouin function for S = 1/2, and the value of the saturation hyperfine field was found to be −110 ±5kG. The orbital ground state of the iron atom could not be deduced from the sign of the quadrupole splitting, due to large contributions from the lattice and bonding electrons. From the magnetic hyperfine interactions it was deduced to be dz2. The model complexes Fe(TTD)2(DTT) and Fe(TTD)(DTT)2 which contain an Fe-S-S- linkage, as well as the related complex Fe(DTT)3 have been studied. (TTD = thio-p-toluoyldisulphide, DTT = dithio-p-toluate) The isomer shifts have been interpreted in terms of the relative ability of the TTD and DTT ligands to delocalise electrons from the metal. The sign of the major component of the electric field gradient has been found to be negative for Fe(TTD)2(DTT) and Fe(TTD)(DTT)2 but positive for Fe(DTT)3. The Mössbauer results have been used in conjunction with epr measurements to obtain information on the ground state of the complexes containing ligands. The ground state is deduced to be a dxy hole in both cases, but the agreement between the Mössbauer and epr data is not very satisfying. Mössbauer Effect measurements on the model Fe(sacsac)2Cl4, where sacsac is dithioacetylacetonate, have shown that the complex contains tetrahedral FeCl42− ions, and does not have the structure originally proposed in which the iron is coordinated by the dithioacetylacetonate groups. The related complex (Ph2C3MS2)2FeCl4 contains FeCl42− ions in two different environments at high temperatures. Tristetraphenylphosphonium trisdicyano-1,2-dithiolene iron was proposed as a model complex on the basis of its epr spectrum, which is qualitatively similar to that of the non-haem iron proteins. Mössbauer Effect investigations have been made on this complex and are reported here. The temperature dependence of the line widths and the relative line intensities has been interpreted in terms of the relative magnitudes of the spin-lattice and spin-spin relaxation times. An attempt to correlate the temperature dependence of the quadrupole splitting with that of the magnetic moment was only partially successful. Measurements in externally applied magnetic fields showed the sign of the major component of the electric field gradient to be negative; lattice contributions to the electric field gradient are not important in determining its sign. The ground state of the complex was deduced to be a dxy hole well separated from higher states, from both the Mössbauer and epr measurements. The complex (π-C5H5Fe(CO)SMe)2SbF6 has been investigated, and is proposed as a model complex for the non-haem iron proteins, on the basis of similarities in its structure and epr spectrum. The Mössbauer Effect measurements have shown that the single unpaired electron is equally shared by the two iron atoms, which results in an anomalously small value for the hyperfine field at the iron nuclei. The epr spectrum has been interpreted in terms of a model in which a high-spin ferric ion and a high-spin ferrous ion are antiferromagnetically coupled to give a total spin S of 1/2. Estinates of the relative energies of some of the orbitals of the ferrous ion have been made. A number of complexes of iron with 2-substituted pyridine ligands have been prepared and studied. The preparation of complexes of thiopicolinic acid anilide, thiopicolinamide, pyridine-2-aldehyde thiosemicarbazone, 2-mercaptomethylpyridine, o-aninobenzenethiol and 2,2′-dipicolyldisulphide are reported. A number of these complexes are described and characterised for the first tine. With one exception all the complexes contain ferrous ion; oxidation of the ligand usually occurred when the preparation of ferric complexes was attempted. The Mössbauer parameters of these complexes are reported and discussed. The values of the isomer shifts of the low-spin ferrous complexes of thiopicolinic acid anilide, thiopicolinamide and pyridine-2-aldehyde thiosemicarbazone are interpreted in terms of the relative σ-donor powers of the ligands. The complexes of o-amino-benzenethiol and 2-mercaptomethylpyridine were found to contain high-spin ferrous ions. The former complex was found to be antiferromagnetic at low temperatures, whilst at 77°K it exhibited the largest quadrupole splitting yet reported for an iron compound, 4.06 mm.sec.−1 The latter complex also chowed a large quadrupole splitting at 77°K. The Mössbauer spectrum observed at low temperatures suggested the presence of some weak antiferromagnetism. The single ferric complex of these ligands that was prepared, is a complex of 2,2′-dipicolyldisulphide and is the first complex of iron with a disulphide ligand to be reported. Further work is necessary before it is fully characterised. Finally, the value of the information obtained from Mössbauer spectroscopy and epr in the elucidation of molecular and electronic structure is discussed. A correlation between the magnitude of the Fermi contact contribution to the hyperfine field is considered in the light of the data obtained during this work.