Hydrogen production from model complexes of the [FeFe]- and [NiFe]-hydrogenase active sites

Hydrogenase;Hydrogen;Proton reduction

Barton, Bryan E.

关键词: Hydrogenase;    Hydrogen;    Proton reduction;   

Others  :  https://www.ideals.illinois.edu/bitstream/handle/2142/18280/Barton_Bryan.pdf?sequence=1&isAllowed=y

美国|英语

来源: The Illinois Digital Environment for Access to Learning and Scholarship

【 摘 要 】

Facing a global energy crisis, many chemists have envisioned molecular hydrogen as an efficient and environmentally friendly fuel of the future. However, for thehydrogen economy to become viable, new hydrogen-processing catalysts are needed toreplace platinum, which is costly and of limited supply. Nature offers direction towardthis goal, as enzymes called hydrogenases evolved several billion years ago to utilizemolecular hydrogen as a fuel. The [FeFe]-hydrogenases predominately function toproduce hydrogen from protons and electrons, while the [NiFe]-hydrogenases functionto oxidize hydrogen. The active sites of both enzymes contain first-row transition metalsand biologically exotic ligands. When assayed for their rates and efficiencies ofhydrogen processing, the hydrogenase enzymes are directly comparable to platinum.Unfortunately, despite several crystal structures and a wealth of spectroscopictechniques, the mechanism of hydrogen processing remains speculative. Our goals assynthetic chemists have focused on the reactivity of models for the [FeFe]- or [NiFe]-hydrogenases in hopes of understanding of how Nature tunes these first-row transitionmetals into phenomenal catalysts.Interestingly, active site models for the [FeFe]-hydrogenases were unknowinglypresent before the first crystal structure in 1999. In fact, the structurally similar diirondithiolate hexacarbonyls had been investigated since the 1920s and had well establishedchemistry. However, unique compared to all other diiron dithiolates, theactive-site structure of [FeFe]-hydrogenase features a rotated diiron dithiolate core,exposing a vacant terminal position. Carbon monoxide binds to this terminal position and inhibits catalysis. Thus, the mechanism of hydrogen processing is thought to occurby substrate (H2, H+) binding in the terminal position. To properly model the biologicalmechanism of [FeFe]-hydrogenase, we sought terminal hydrides of diiron dithiolates.After the first terminal hydride complex, [HFe2(edt)(CO)2(PMe3)4]+, was published fromour group others quickly followed. This new class of diiron dithiolate terminal hydrideswas derived by the biologically relevant pathway of protonation of a Fe(I)Fe(I) precursor.Unfortunately, the terminal hydrides derived in this fashion were unstable, and whenwarmed above –80 °C quickly isomerized to isomeric bridging hydrides.To understand and control the selective formation of terminal hydride speciesand subsequent isomerization pathway, a series of diiron dithiolates were investigated.All diiron dithiolates explored showed the kinetic formation of a terminal hydride speciesthat subsequently isomerized via a series of turnstile rotations to bridging hydrides. Welearned that these turnstile rotations were controlled by a combination of electronic andsteric effects, as the 1,3-propanedithiolate derivatives were vastly more stable than theircorresponding 1,2-ethanedithiolate derivatives. Additionally, more phosphine ligandsgenerally resulted in a more stable terminal hydride. Thus, protonation ofFe2(pdt)(CO)2(dppv)2 provided the terminal hydride complex [(t-H)Fe2(pdt)(CO)2(dppv)2]+, which isomerized at room temperature (t1/2 ~ 10 min) to thebridging hydride [(μ-H)Fe2(pdt)(CO)2(dppv)2]+. With a pseudo-stable terminal hydridecomplex in hand, we sought to explore the catalytic mechanism of proton reduction viathe terminal hydride. To our surprise, although the mechanism was very similar to thatproposed in biology, the catalytic efficiency suffered greatly when compared to thebridging hydride complex. We continued focusing our research efforts on proton relay.The active site of [FeFe]-hydrogenase is speculated to contain a 2-azapropane-1,3-dithiolate as the bridging dithiolate ligand, although the exact identity of thebridgehead atom could be either carbon, nitrogen, or oxygen. Recent work onmononuclear nickel phosphines led to the impression that an azadithiolate (adt) couldfunction as a proton relay lowering the kinetic barrier of proton transfers to and from theterminal hydride position. Due to the significant amount of steric congestion in[(H)Fe2(pdt)(CO)2(dppv)2]+, the iron hydride does not deprotonate withtetramethylguanidine (pKa = 26) and requires the strongacid HBF4•Et2O (pKa = -2) forits formation. Upon incorporation of the proposed azadithiolate cofactor,[(H)Fe2(adt)(CO)2(dppv)2]+ was observed to have a significantly smaller barrier forproton transfer to and from the terminal position. In addition, [(H)Fe2(adt)(CO)2(dppv)2]+was observed to be a remarkably fast and efficient catalyst for proton reduction, withturnover frequencies approaching that of the enzyme.Unlike the [FeFe]-hydrogenases, model complexes for the [NiFe]-hydrogenaseswere unknown prior to the crystal structure in 1996. However, most synthetic effortsfocused on structural models for the active site, and neglected the catalyticallyimperative hydride ligand. Thus, we sought a nickel-iron hydride complex to explore therelevant reactivity of the first (μ-H)Ni(μ-SR)2Fe complex. We found that the previouslyreported (dppe)Ni(μ-pdt)Fe(CO)3, a Ni(I)Fe(I) complex, reacted with acid to provide[(dppe)Ni(μ-H)(μ-pdt)Fe(CO)3]+, the first nickel-iron hydride. After protonation, thehydride complex is amenable to substitution chemistry at the Fe(CO)3 subunit. Furtherderivatives altering the Ni(diphosphine)(SR)2 subunit have been achieved through analternative synthetic procedure to the Ni(I)Fe(I) complex. All nickel-iron hydridesinvestigated are active catalysts for the reduction of protons. As the catalyticmechanism of [NiFe]-hydrogenase is widely speculative, the reactivity of this new classof nickel-iron hydrides offers powerful insights into Nature’s catalytic mechanism.