Metal complex dyads, M-D, comprise a transition metal chromophore (M) covalently linked to an organic electron donor (D).1-3 Photochemical excitation of these assemblies produces a charge-separated state, M•--D+•, which has a lifetime that is controlled by the dynamics of the highly exothermic (inverted region) charge recombination reaction.2,3 In previous studies it has been shown that the lifetime of the charge-separated state in M-D assemblies increases with the driving force for charge recombination2b,3c or the separation distance between M and D.3b Herein we report a new approach to increasing the lifetime of a charge-separated state in a metal complex dyad which relies on the reversible ring opening of the radical cation of a piperazine electron donor. Thus, metal complex dyad c-1 contains the (bpy)Re(CO)3(Re) chromophore7 covalently linked to a 2,3-diaryl-1,4-dimethylpiperazine electron donor. Intramolecular electron transfer from piperazine to photoexcited Re produces a charge-separated state in which Re is reduced and the piperazine is a cation radical. Carbon-carbon bond fragmentation in the piperazine cation radical produces a new charge-separated state in which the donor exists as an openchain distonic cation radical. The net result is the production of a charge-separated state having a lifetime which is 5-10fold longer than that observed in structurally related metal complex dyads. The isomeric complexes c-1 and t-1 were synthesized and fully characterized.8 The near-UV absorption of both compunds is dominated by the dπ (Re) f π* (bpy) metal-to-ligand charge transfer (MLCT) transition of the Re chromophore.9 Irradiation of a solution of c-1 in air-saturated or Ar-degassed CH3CN at 366 nm affords t-1 as the sole chemical product with high quantum efficiency (Φc-1ft-1 ) 0.46 ( 0.05 in Ar-degassed solution).10 A mechanism for c-1 f t-1 photoisomerization is provided in Scheme 1. Near-UV excitation produces MLCT excited state c-1*, which subsequently relaxes either by radiative and nonradiative decay to c-1 (step 1) or by forward electron transfer (step 2) to afford charge-separated state c-3 in which the piperazine donor is oxidized and the bpy acceptor ligand is reduced (i.e., bpy•-). The occurrence of step 2 in c-1* is indicated by the fact that the MLCT emission lifetime of c-1 (τem ) 55 ns) is suppressed compared to that of model complex 2 (τem ) 235 ns). The emission lifetimes of c-1 and 2 allow estimation of k2 (1.4 × 107 s-1) and the efficiency for formation of charge-separated state c-3 (φ2 ) 0.76).11 Piperazine isomerization is believed to occur via fragmentation of the 2,3-C-C bond (step 3) to form a new charge-separated state in which the piperazine exists as a distonic cation radical with two possible structures (4a or 4b). Carbon-carbon bond fragmentation in radical cations is well precedented in acyclic systems.12,13 Indeed, recent studies indicate that for acyclic 1,2-diaryl-1,2diaminoethanes C-C bond fragmentation is exoergonic and occurs with k > 108 s-1.13e Since C-C bond fragmentation in 4 is probably exoergonic,14 the reverse process (e.g., 4 f 3) cannot occur, and therefore isomerization occurs via charge recombination (step 4) followed by coupling of 1,6-diradical 5 (step 6).16 A significant question is why t-1 f c-1 photoisomerization does not occur. Quite remarkably, the MLCT emission lifetime of t-1 (τem ) 233 ns) is not suppressed substantially from that of model 2, which indicates that forward electron transfer is too slow to compete with normal decay of the MLCT excited state, t-1*. The lack of electron transfer quenching in t-1* precludes formation of charge-separated state t-3 (not shown in Scheme 1), thereby closing off the pathway to t-1 f c-1 isomerization via fragmentation of the piperazine cation radical.
Read full abstract