Construction of Artificial Photosynthetic Reaction Centers on a Protein Surface:  Vectorial, Multistep, and Proton-Coupled Electron Transfer for Long-Lived Charge Separation

Artificial photosynthetic reaction centers have been constructed on a protein surface by cofactor reconstitution, which mimic the function of photosynthetic organisms to convert light energy to chemical potential in the form of long-lived charge-separated states. They feature a ruthenium tris(2,2‘-bipyridine) moiety as the sensitizer, which is mechanically linked (i.e., in catenane-type) with a cyclobis(paraquat- p -phenylene) unit (BXV 4+, acceptor) and covalently linked with a protoheme or Zn−protoporphyrin (donor) located in the myoglobin pocket. Their cofactors 1 and 2, which are tris(heteroleptic) Ru−bipyridine complexes, were synthesized by sequential coordination of the two different functionalized bipyridine ligands with a readily obtainable precursor [Ru(4,4‘-dimethyl-2,2‘-bipyridine)Cl 3 ] n followed by metal insertion; this represents a new efficient synthetic method for tris(heteroleptic) Ru(II) complexes of bidentate polypyridine ligands. Reconstitution of apo-myoglobin (Mb) with 1 and 2 affords the two Mb-based artificial triads, Mb(Fe III OH 2 )−Ru 2+ −BXV 4+ and Mb(Zn)−Ru 2+ −BXV 4+ . Laser flash photolysis of the Ru(bpy) 3 moiety of Mb(Fe III OH 2 )−Ru 2+ −BXV 4+ in an aqueous solution yields an initial charge-separated state, Mb(Fe III OH 2 )−Ru 3+ −BXV 3+ •, via noncovalent electron transfer, followed by dark electron transfer to generate an intermediate consisting of porphyrin cation radical, Mb(Fe III • OH 2 )−Ru 2+ −BXV 3+ • . Mb(Fe III • OH 2 )−Ru 2+ −BXV 3+ • thus generated is subsequently converted, via a proton-coupled process and with a quantum yield of 0.005, into the final charge-separated state, Mb(Fe IV O)−Ru 2+ −BXV 3+ •, which bears an energy more than 1 eV above the ground state and a lifetime (τ > 2 ms) comparable to that of natural photosynthetic reaction center. Photoexcitation of Mb(Zn)−Ru 2+ −BXV 4+ also gives rise to a vectorial two-step electron-transfer relay with the intermediate CS state, Mb(Zn)−Ru 3+ −BXV 3+ •, for the main pathway leading to the final CS state, Mb(Zn + )−Ru 2+ −BXV 3+ •, in a yield of 0.08. Although the driving forces for the recombination of Mb(Fe IV O)−Ru 2+ −BXV 3+ • and Mb(Zn + )−Ru 2+ −BXV 3+ • are similar (Δ G ≈ 1.30 eV), the recombination rate of the former is at least 10 2 −10 3 -fold slower than that of the latter. By analogy with a related system reported previously, it was considered that back ET from BXV 3+ • to Mb(Fe IV O) might be coupled to the protonation of Mb(Fe IV O) and governed by the slow interconversion between the metal−oxo form and the proton-activated species, rendering the CS state Mb(Fe IV O)−Ru 2+ −BXV 3+ • specially long-lived. Control experiments clearly demonstrated that partial incorporation of the triads into the protein matrix plays a crucial role in regulating the electron-transfer pathway and stabilizing the charge separation state.