Ronald A. Prados

Some factors influencing mixed complex formation

Abstract The tendency toward mixed complex formation may be expressed quantitatively in two different ways. The most common approach is to compare the stability of the mixed complex with the two pure 2:1 complexes where two like ligands are bound to a single metal ion. It is shown that extraneous factors that enter into formation of the pure 2:1 complexes, such as steric effects, may give a misleading indication of the stability of a mixed complex. Moreover, unless both ligands are of the same charge, there is an appreciable electrostatic contribution to mixed complex formation that is ionic strength dependent. The alternative formulation that relates the stability of the mixed complex to 1:1 complexes provides a better measure of the intrinsic tendency toward mixed complex formation. When one of the ligands is uncharged as is the case with amines, there is no electrostatic contribution in this formulation. For most ligands the visible circular dichroism magnitudes of mixed Cu(II) complexes containing l -alanine or R -1,2-diaminopropane as one ligand are not half that of the 2:1 complex with one of the optically active ligands. This result suggests profound electronic interactions within the tetragonal plane. In addition four aromatic amines, bipyridyl, phenanthroline, bipyridylamine and histamine, when present as the other ligand in a mixed Cu(II) complex invert the sign of the visible CD from that of the 2:1 complex of the optically active ligand. Steric effects are ruled out and transmission of electronic effects through π systems is suggested to account for the inversions. Most Cu(II) complexes with two bidentate ligands are at least five-coordinate in solution, but axially bound water is not specifically considered in mixed complex formation. So-called binary complexes are actually ternary complexes with water as the second ligand. Viewed in this way ratios of the first to second stability constants provide an untapped fund of data on tendencies toward formation of mixed complexes.

Lanthanide complexes of amino acids

Abstract Even in the presence of ten-fold excess ligand, amino acids and lanthanide ions titrate 2·4–2·8 equivalents of base per mole of Ln(III) associated with hydrolysis of the metal ion in neutral solutions. Ammonium groups titrate as if unbound, and no ionization from the hydroxy group of serine appears. With all fifteen ligands studied except α,β-dicarboxylic acids, the number of equivalents due to lanthanide ion hydrolysis are the same whether or not ligand is present. The non-integral number of equivalents and the narrowness of the pH range over which their addition takes place suggest formation of polynuclear complexes in the hydrolysis region. Probably because they serve as bi- and terdentate ligands, aspartic and malic acids titrate 2·0 equivalents of base per mole of Ln(III) regardless of the ligand to lanthanide ion molar ratio. At greater than equimolar ratios malic acid yields a bipartite titration curve in the hydrolysis region, the first part of which is due to addition of 1·3–1·4 and the second to 0·6–0·7 equivalents of base. Especially for the simpler optically active amino acids, extensive increases in complexity and magnitude of circular dichroism occur through the hydrolysis region of the lanthanide ion complexes.

Fe(III) and Cu(II) conalbumin visible circular dichroism spectra.

The single polypeptide chain of conalbumin strongly binds two Fe(III) or two Cu(II) ions to yield intense absorption in the visible region similar to that shown by the related protein transferrin. Comparison of the metal-ion-binding sites in the two proteins is made by exploiting the sensitivity to ligand geometry of circular dichroism (CD). For the Fe(III) proteins strong similarities of the CD spectra outweigh marginal differences. For Cu(II) conalbumin an additional negative extremum near 506 nm appears between two positive ones at 634 and 410 nm suggesting greater subtraction of oppositely signed CD components leading to lesser magnitudes for the two positive peaks than are found in Cu(II)-transferrin. The two Fe(III)-binding sites within conalbumin are compared by noting the strong similarities of the CD and MCD of proteins with Fe(III) in one site and Ga(III) in the other site, and vice versa, with the protein containing Fe(III) in both sites. Due to features of the amino acid sequences of the single protein chains, the four strong metal ion binding sites in conalbumin and transferrin cannot be identical in all particulars, yet CD spectra of their metal ion complexes are closely similar. From a study of model phenolate complexes and the wavelength maxima of visible absorption in the Fe(III), Cu(II), and Co(III) proteins near 465, 440, and 405 nm, respectively, these strong absorption bands are identified as ligand to metal ion electron-transfer transitions. It is suggested that tyrosyl residues are the donors in the electron transfer transitions and that they lock in the metal ions after being keyed into position by binding of bicarbonate or other anions.

Identification of stereoisomers of mixed heteropoly anions. Mixed valence and triplet state electron spin resonance spectra of vanadium(IV)

The existence of several isomers of the Keggin anions PVxW12 –xO40(3 +x)– and PVxMo12 –xO40(3 +x)–, (x= 1–4), is demonstrated by 31P n.m.r., e.s.r., and i.r. spectroscopy.