Summary: | This thesis describes the preparation of a number of polydentate ligends having nitrogen, or nitrogen and phosphorus as the donor atoms. Complexes of the ligends with several transition metals were studied and their preparations and properties are described. Deductions of the stereochemistry of the resulting complex ions are made where possible. The attempted preparation of the ligand 1:2 dipentyl ethylene diamine is described and the preparations of 1.2 diphenyl ethylene diamine (m-stien) and the para-para' isopropyl derivative (p-stien) are discussed. The following ligands were prepared with both phosphorus and nitrogen onor atoms: tris(e-dimethylaminophenyl)phosphine, PTN; bis(e-dimethylaminophenyl)phenyl phosphine, PDN; and e-dimethylaminophenyl diphenyl phosphine, PN. They were synthesised by the action of a suitable chlorophenyl phosphine on the ortho-litho derivative of dimethyl aniline. M-stien complexes of copper II were prepared and studied. The general formula was [m-stien<sub>2</sub>CuX<sub>2</sub>JnH<sub>2</sub>O, (where X=Cl, n=1; x=Br, n=1,0; X=I, n=0; X=SCN, n=1; X=NO<sub>3</sub>, n=0; X=ClO<sub>4</sub>, n=0; X=½SO<sub>4</sub>, n=1; X=acetate, n=1; X=oxalate, n=1). The compounds varied between violet and green in colour and were almost all insoluble in nitromethane. Magnetic and spectral studies of the m-stien copper complexes showed them to be tetragonal with four nitrogen atoms in a square plane. The ligand field of water in m-stien<sub>2</sub>CuBr<sub>2</sub>.H<sub>2</sub>O causes it to be violet whilst the anhydrous material, with a weaker ligand field, is green. Where deductions could be made the infra-red spectra confirms that there is some interaction between the metal and anionic ligends. Some m-stien complexes of palladium II were prepared. The general formula was m-stien<sub>2</sub>PdX<sub>2</sub> where X=Cl, Br, SCN, NO<sub>3</sub>). These compounds were less well characterised than the copper ones. Physical studies on these compounds showed them to be essentially square planar; they were not sufficiently soluble in non-donor solvents to make further studies possible and so tri- and tetradentate ligands of this type were not prepared. The compounds [p-stien<sub>2</sub>NiX<sub>2</sub>nH<sub>2</sub>O were then studied (where X=Cl, n=4, O; X=I, n=2, O). In the chloride complex, removal of the water of hydration changes the yellow colour to green and changes the magnetic moment from 0 to 2.68 BM. This behaviour is similar to that of the Lifschits complexes of nickel with m-stien. However, it is unusual to observe the formation of a paramagnetic species on dehydration. The spectral and magnetic evidence of structure in these compounds is compared with some parallel cases, and it is deduced that the yellow compound is essentially square planer whilst the green compound is a mixture of 'octahedral' and square planar stereochemistries. Compounds of the type p-stien<sub>2</sub>PdX<sub>2</sub>.2H<sub>2</sub>O were also prepared (where X=Cl, Br, NO<sub>3</sub>). These white compounds were similar to the m-stien analouges and the para-isopropyl groups on the ligand seemed to confer no special advantages over m-stien with palladium. Thus neither system was suitable for the study of complexes of tri- or tetradentate ligands of this type. The mixed phosphorus-nitrogen ligands were, therefore, prepared and their complexes studied. The platinum and palladium complexes of PTN, PDN and PN were isolated. They were of the formula [LMX<sub>2</sub>] (where L=PTN, PDN, PN; M=Pt, Pd; and X=Cl, Br, I). These compounds were more soluble than the stien derivatives in non-donor solvents and were extensively studied. Magnetic, spectral and NMN. evidence showed that the compounds were all cis-square planar, the metal being bound to two halide ions, one phosphorus and one nitrogen atom. It is of great interest that although PTN is formally analogous to the ligand tris(diphenylareinophenyl)-arsine, QAS, it does not form the trigonal bipyramidal complexes well known for the latter ligand with palladium and platinum. Instead the more familiar square planar stereochemistry is retained in the d<sup>8</sup> ions. It is possible that a considerable amount of π-bonding is required to stabilise the five-coordinate ion. Compounds of the three ligands with nickel were prepared. They had the formula [LNiX<sub>2</sub>] (where L=PTN, PDN, PN, X=Cl, Br, I; and L=PTN, X=SCN). These complexes were less soluble than the palladium and platinum ones and were paramagnetic. The spectral and magnetic evidence suggests that, in these compounds too, the organic ligand is bound to the metal by one nitrogen and one phosphorus atom in each case. The evidence is not conclusive but does strongly suggest that the nickel is in an essentially tetrahedral environment, though tetragonally distorted. Differences in the spectra are then attributed to interactions, wither steric or electronic, between the metal and the 'free' nitrogen atoms in PDN and PTN. The cobalt compounds of the formula [LCoN<sub>2</sub>] (where L-PTN, PDN and X=Cl, Br) were prepared. In this case the iodide complexes could not be islated. These compounds had quartet ground states and ultraviolet spectra suggested that these compounds were probably tetrahedral. interactions with the 'free' nitrogen atoms explained small differences in both the spectra and the magnetic moments. Finally some complexes of copper I were prepared, of the formula [LCuK] (where L=PTN, PDN, X=Cl, Br, I and L=PN, X=Cl) and [L<sub>2</sub>CuK] (where L=PN, X=Br, I). No firm deductions could be made on the available evidence of the stereochemistries of these compounds, although the infra-red spectra suggest that all the ligands may be bound only through phosphorus. Thus the work in the present thesis suggests that, with the metals studied, the mixed phosphorus-nitrogen ligands owe such of their coordinating ability to the phosphorus atom with its π-bonding potential. the ligand PTN does not form the five-coordinate analogues of QAS despite its chelating ability and rigid stereochemistry.
|