X-ray absorption fine structure (XAFS) spectra have been measured for a series of structurally characterized zinc model complexes that mimic the zinc sites found in metalloproteins. These include both inorganic zinc coordination complexes and small zinc binding peptides. These data have been analyzed to determine the extent to which Zn XAFS can be used to determine reliably the ligation environment of the zinc. Because Zn−N and Zn−S XAFS oscillations are nearly out of phase over the accessible energy range, it is difficult to determine the relative number of sulfur and nitrogen scatterers, and in some cases, it is even difficult to determine whether a low-Z (N or O) ligand is bound in the presence of high-Z (S) ligands. We describe a protocol that, by controlling the number of variable parameters, can be used to obtain an accurate quantitation of the number of low-Z ligands. We also show that two of the variables that are often treated as freely adjustable parameters, the scale factor and shift in the threshold energy, can lead to erroneous results if not carefully controlled.

Reactions of NO and CO with Fe(II) complexes of the tripodal trithiolate ligands NS3 and PS3* yield trigonal-bipyramidal (TBP) complexes with varying redox states and reactivity patterns with respect to dissociation of the diatomic ligand. The previously reported four-coordinate [Fe(II)(NS3)](-) complex reacts irreversibly with NO gas to yield the S = 3/2 {FeNO}(7) [Fe(NS3)(NO)](-) anion, isolated as the Me(4)N(+) salt. In contrast, the reaction of NO with the species generated by the reaction of FeCl(2) with Li(3)PS3* gives a high yield of the neutral, TBP, S = 1 complex, [Fe(PS3*)(NO)], the first example of a paramagnetic {FeNO}(6) complex. X-ray crystallographic analyses show that both [Fe(NS3)(NO)](-) and [Fe(PS3*)(NO)] feature short Fe-N(NO) distances, 1.756(6) and 1.676(3) A, respectively. However, whereas [Fe(NS3)(NO)]- exhibits a distinctly bent FeNO angle and a chiral pinwheel conformation of the NS3 ligand, [Fe(PS3*)(NO)] has nearly C(3v) local symmetry and a linear FeNO unit. The S = 1 [Fe(II)(PS3)L] complexes, where L = 1-MeIm, CN(-), CO, and NO(+), exhibit a pronounced lengthening of the Fe-P distances along the series, the values being 2.101(2), 2.142(1), 2.165(7), and 2.240(1) A, respectively. This order correlates with the pi-backbonding ability of the fifth ligand L. The cyclic voltammogram of the [Fe(NS3)(NO)](-) anion shows an irreversible oxidation at +0.394 V (vs SCE), apparently with loss of NO, when scanned anodically in DMF. In contrast, [Fe(PS3*)(NO)] exhibits a reversible {FeNO}(6)/{FeNO}(7) couple at a low potential of -0.127 V. Qualitatively consistent with these electrochemical findings, DFT (PW91/STO-TZP) calculations predict a substantially lower gas-phase adiabatic ionization potential for the [Fe(PS3)(NO)](-) anion (2.06 eV) than for [Fe(NS3)(NO)](-) (2.55 eV). The greater instability of the {FeNO}(7) state with the PS3* ligand results from a stronger antibonding interaction involving the metal d(z(2)) orbital and the phosphine lone pair than the analogous orbital interaction in the NS3 case. The antibonding interaction involving the NS3 amine lone pair affords a relatively "stereochemically active" dz2 electron, the z direction being roughly along the Fe-N(NO) vector. As a result, the {FeNO}(7) unit is substantially bent. By contrast, the lack of a trans ligand in [Fe(S(t)Bu)3(NO)](-), a rare example of a tetrahedral {FeNO}(7) complex, results in a "stereochemically inactive" d(z(2)) orbital and an essentially linear FeNO unit.