Small molecule mimics of hydrogenase enzymes: synthesis, protonation, and electrocatalysis

Carroll, Maria

Permalink

https://hdl.handle.net/2142/45281 Copy

Description

Title

Small molecule mimics of hydrogenase enzymes: synthesis, protonation, and electrocatalysis

Author(s)

Carroll, Maria

Issue Date

2013-08-22T16:34:34Z

Director of Research (if dissertation) or Advisor (if thesis)

Rauchfuss, Thomas B.

Doctoral Committee Chair(s)

Rauchfuss, Thomas B.

Committee Member(s)

• Girolami, Gregory S.
• Gewirth, Andrew A.
• Murphy, Catherine J.

Department of Study

Chemistry

Discipline

Chemistry

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• Hydrogenase
• Iron
• Nickel
• Cobalt
• Hydrogen Production

Abstract

In nature, H2 is processed by enzymes called hydrogenases, which catalyze the reduction of protons to dihydrogen, as well as the reverse reaction. The active sites of the two most prevalent hydrogenases contain NiFe or FeFe cores, bound to thiolates, cyanide, and carbon monoxide ligands. These enzymes are also rich in Fe-S clusters to allow the necessary redox chemistry of hydrogen oxidation and production. Both enzymes operate at rates and overpotentials comparable with the best synthetic Pt catalysts. Due to growing concern over the climate effects of burning fossil fuels, there is a push to replace these fuels with carbon free alternatives, one option being H2. However, this would require catalysts for H2 production that are based on cheap, easily accessible metals. This problem inspired extensive research on the development of functional small molecule mimics of the hydrogenase enzymes. The work presented herein is motivated by the goal of understanding the mechanism of hydrogenase enzymes, in order to design better catalysts for hydrogen production. Chapter 1 presents an overview of current methods for the production of H2, including methods used in industry, as well as heterogeneous and homogeneous metal catalysts. The most unique features of the active site of [FeFe]-H2ases are the amino-dithiolate cofactor that bridges the two Fe centers and the rotated geometry of one Fe center, leaving an open coordination site adjacent to the amine, which is the proposed site of substrate binding. Many active site models have been prepared, many of which undergo protonation to form hydride complexes and catalyze proton reduction. However, in all cases, the thermodynamic product of protonation is a bridging hydride, which is not biologically relevant. A number of phosphine substituted diiron complexes have been found to form terminal hydrides at very low temperatures. Chapter 2 describes the protonation of complexes of the type Fe2(xdt)(CO)4(dppv)2 (xdt= pdt, 1,2-propanedithiolate, or adtNH= azadithiolate; dppv= cis-1,2-bis(diphenylphosphino)ethylene, which form terminal hydrides that are stable at 0 °C for ~ 30 minutes and then isomerize to the corresponding bridging hydrides. Fe2(adtNH)(CO)4(dppv)2 undergoes protonation with weak acids, whereas, the pdt analogue requires strong acid; the difference being attributed to the presence of a pendant base in Fe2(adtNH)(CO)4(dppv)2, which is initially protonated and then relays the proton to the Fe center. Additionally, in the presence of excess acid, Fe2(adtNH)(CO)4(dppv)2 sustains double protonation to form a terminal hydride ammonium species. The complex, [t-HFe2(adtNH2)(CO)4(dppv)2](BF4)2, is the first example of a crystallographically characterized terminal hydride produced by protonation. The most significant feature of the structure is the NH--HFe distance of 1.88 Å, which indicates dihydrogen bonding. The molecule is positioned to release H¬2, and represents a key intermediate in the mechanism of proton reduction catalysis. Chapter 3 describes the redox and catalytic properties of the terminal and bridging hydrides of Fe2(xdt)(CO)2(dppv)2. For both the adtNH and the pdt derivatives, the terminal hydride species are reduced at ~150 mV more mild potentials than the corresponding bridging hydrides. The voltammetry of [t-H Fe2(adtNH)(CO)2(dppv)2]+ is strongly affected by relatively weak acids and proton reduction catalysis proceeds at 5000 s-1 with an overpotential (the deviation from the thermodynamic reduction potential of the acid) of 0.7 V. The ammonium-hydride [t-H Fe2(adtNH2)(CO)2(dppv)2]2+ is a faster catalyst, operating at 58,000 s-1 and an overpotential of 0.5 V. When the adtNH cofactor is replaced with the less biologically relevant ligand pdt or the terminal hydride species are replaced with isomeric bridging hydrides, catalysis is significantly slower (TOF= 5- 20 s-1) and inefficient (overpotential= 0.9- 1.3 V), indicating that hydrogen evolution by biomimetic diiron dithiolates is accelerated by the amine cofactor, to which the hydride ligand must be adjacent, and inclusion of features found in the enzyme results in fast and efficient catalysis. Unlike models for [FeFe]-hydrogenase, hydride containing models of [NiFe]-hydrogenase were unknown until, in 2009, our group reported the complex [(dppe)Ni(pdt)(-H)Fe(CO)3]BF4¬ (dppe= 1,2-bis(diphenylphosphino)ethane), which was found to undergo substitution of CO with phosphine ligands. Most importantly, these new NiFe hydrides were found to catalyze proton reduction at mild overpotentials. Chapter 4 describes the development of a new synthesis of complexes of the type (diphosphine)Ni(xdt)(CO)3, in which Fe(CO)4I2 condensed with Ni(xdt)(diphosphine), forming a NiIIFeII -iodide intermediate, which is then reduced to form the neutral NiFe complex. With this new synthetic method in hand, we synthesized of new derivatives, varying in the identity of the dithiolate, the diphosphine, and the ligands on the Fe center. Having a range of derivatives, we probed the individual roles of Ni vs Fe in catalysis. The acidity of the hydride appears to be more strongly affected by changes at the Fe center (pKaMeCN of 4 for L= PPh3 vs L= CO) than changes at the Ni center (pKaMeCN of 2.5 for R = Ph vs Cy). The reduction potential appears to be more strongly affected by changes at Ni (Ered¬ of 250 mV for R = Ph vs Cy) than at Fe (Ered¬ of 200 mV for L = CO vs PPh3). Changes in the dithiolate have a minimal effect on the reduction potentials of the hydrides, although the rates of hydrogen evolution are strongly affected by the dithiolate. The catalysts operate at overpotentials of 0.4 V and the rates up to 300 s-1, which are good by the standards of model studies although modest by the standards of the enzymes. Chapters 5 and 6 focus on the development of new methods for the synthesis of bimetallic dithiolato complexes, as well as the effect of changing the metal centers to metals that are not found in hydrogenases. Chapter 5 describes the synthesis of new ferrous dicarbonyl dithiolato diphosphine complexes containing chelating diphosphine and dithiolate ligands. A new building block method for the synthesis of substituted diiron complexes of the type Fe2(xdt)(CO)4(diphosphine), by reaction of Fe(xdt)(CO)2(diphosphine) complexes with an Fe(0) tricarbonyl source, is described. Additionally, this building block approach is extended to the synthesis of Mn-Fe analogues of the FeFe complexes. The new bimetallic complexes, [(CO)3Mn(xdt)Fe(CO)2(diphosphine)]BF4 are synthesized by the reaction of Fe(xdt)(CO)2(diphosphine) with the manganese tricarbonyl transfer reagent, [(acenaphthene)Mn(CO)3]BF4. These cationic complexes undergo decarbonylative reduction to form neutral MnFe complexes that are models for the FeIFeII Hox state of the enzyme. Synthesis of the bridging hydride complex Mn(CO)3(pdt)(-H)Fe(CO)(dppe) is described. The hydride complexes been characterized crystallographically, and can be oxidized reversibly, reactivity that is not seen in NiFe or FeFe models that contain hydride ligands. Chapter 6 describes the synthesis of bimetallic CpCo complexes of the type, (C5H5)Co(xdt)Co(C5H5) (xdt= pdt, 1,2-propanedithiolate; edt, 1,2 ethanedithiolate, and tdt= 3,4-toluenedithiolate), in an effort to synthesize more electron rich model complexes, by replacing the Fe(CO)3 unit with CpCo. These complexes undergo protonation to form bridging hydride species, which catalyze the reduction of protons, albeit at modest rates and fairly high overpotentials.

Graduation Semester

2013-08

Permalink

http://hdl.handle.net/2142/45281

Copyright and License Information

Copyright 2013 Maria Carroll

Owning Collections