OPTICAL AND THERMAL ELECTRON-TRANSFER ACTIVATION OF DIOXYGEN BY VIOLOGEN DITHIOLENE METALATES

Photoinduced electron transfer activation of dioxygen by redoxactive c harge-transfer ion pairs of the type {A(2+)[Pt(mnt)(2)](2-)} (A(2+)=2, 2'-,4,4'-bipyridinium or cycloalkylated biimidazolium dication; mnt(2- )=maleonitriledithiolate) occurs through an optical electron transfer within an ion pair. This affords the primary redox products A(.+) and [Pt(mnt)(2)](-) as indicated by laser flash photolysis. Under argon th e transients recombine by fast second-order kinetics. Under dioxygen a different behavior is observed. In the case of accepters with a first reduction potential more positive than -0.6 V back electron transfer prevails. When the potential is more negative however, A(.+) reduces O -2 by pseudo-first-order kinetics to generate O-2(.-), while [Pt(mnt)( 2)](-) accumulates in the solution. Quantum yields increase with decre asing excitation wavelength. This suggests that internal conversion of the initially populated excited state to the photoreactive ion pair c harge-transfer state is more efficient upon excitation to the interlig and (pi,pi) state (334 nm) than to the metal-to-ligand charge-transfe r state (437 or 580 nm). In the latter cases competitive radiationless deactivation via metal-centered states occurs. The corresponding Ni a nd Pd complexes do not exhibit any reactivity due to their very short excited state lifetimes. Formation of O-2(.-) was proved by ESR spin-t rapping techniques. Accumulation of [Pt(mnt)(2)](-) occurs also when i nstead of irradiating, the reaction is performed in the dark at about 160 degrees C. The activation energy of 108 +/- 10 kJ/mol as obtained for the thermal electron transfer from [Pt(mnt)(2)](2-) to A(2+) corre sponds well to the value calculated from the Hush-Marcus model.