3
Fundamental Aspects of Bioinspired Chemistry for Energy

Marcetta Darensbourg, Thomas Rauchfuss, Michael Wasielewski, and Charles Dismukes presented examples of the fundamental research being done at their institutions.

HYDROGENASES AS INSPIRATION

Marcetta Darensbourg of Texas A&M University began by explaining the motivation for research on hydrogenases in the 1930s. Marjory Stephenson was the first person to discover hydrogenases in microorganisms that were responsible for the pollution in the River Ouse in Cambridge, United Kingdom. She looked at reactions, such as the fermentation of glucose, and concluded that there was an enzyme in the cells that produced hydrogen and enzymes that absorbed hydrogen in processes important to the production of methane. In one of Stephenson’s papers she noted, “Bacterial coli has been shown to catalyze the oxidation of dihydrogen to two protons, releasing two electrons in a completely reversible way. The hydrogenase system is the most negative reversible oxidation reduction as yet described in living cells.” Later, this activation was found to proceed from the heterolytic splitting of dihydrogen (i.e., to a proton and a hydride). Each decade following the initial discovery of hydrogenases brought tremendous advances, but in this century, Darensbourg noted another motivation for research: applying them to biotech challenges.

Darensbourg described how Fraser Armstrong (University of Oxford) and Antonio DeLacey (Universidad Autonoma de Madrid) are making enzyme electrodes in order to establish the hydrogen uptake and hydrogen production ability of the enzymes. Armstrong has calculated that on a graphite electrode, nickel-iron hydrogenase catalyzes hydrogen oxidation at a diffusion-controlled rate that matches the rate achieved by platinum. However, said Darensbourg, there are problems associated with the enzyme-electrode technology: (1) the enzymes are derived from air-sensitive extremophiles; (2) while some can be tailored to be less air sensitive, most are of questionable robustness; and (3) they weigh a lot more than platinum. Hence, the bioinorganic chemistry approach to these problems involves preparation of small-molecule synthetic analogs that have the essence of the enzyme’s metal-containing active site while reducing the amount of biomatter surrounding it. Darensbourg’s research goals and those of others around the world substitute abiological ligands to engender the exact electronic environment of the metal that would allow it to have the same function it has in the complicated protein matrix. Ultimately they would like to attach them to carbon electrodes.

Hydrogenase Structures and Functions

Darensbourg then described the structures and functions of the hydrogenases (Figure 3.1), pointing out the common factors of the active sites of two classes of iron-sulfur cluster containing hydrogenases (Figure 3.2), which are genetically distinct and evolved along different pathways, yet ended up at the same point in terms of function. She looked more closely at the all-iron hydrogenase, pointing out the iron-sulfur clusters that serve as a molecular wire into the hydrogen cluster that are a built-in electron-delivery system. She pointed out a second coordination sphere effect on the active site, which, if modified, will reduce the activity of that enzyme. With such complexities, the obvious question posed by Darensbourg was: “Will a small molecule modeled solely on the two-iron subsites be an electrocatalyst for hydrogen production or hydrogen uptake in oxidation?”

Darensbourg, Tom Rauchfuss, and Chris Picket saw a simple diiron organometallic molecule as an obvious mimic of the iron hydrogenase active site. Darensbourg said that this comparison was a good starting point for modeling



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 15
3 fundamental Aspects of Bioinspired Chemistry for Energy Marcetta Darensbourg, Thomas Rauchfuss, Michael (1) the enzymes are derived from air-sensitive extremophiles; Wasielewski, and Charles Dismukes presented examples of (2) while some can be tailored to be less air sensitive, most the fundamental research being done at their institutions. are of questionable robustness; and (3) they weigh a lot more than platinum. Hence, the bioinorganic chemistry approach to these problems involves preparation of small-molecule HYDROGENASES AS INSPIRATION synthetic analogs that have the essence of the enzyme’s Marcetta Darensbourg of Texas A&M University began metal-containing active site while reducing the amount of by explaining the motivation for research on hydrogenases biomatter surrounding it. Darensbourg’s research goals in the 1930s. Marjory Stephenson was the first person to and those of others around the world substitute abiological discover hydrogenases in microorganisms that were respon- ligands to engender the exact electronic environment of the sible for the pollution in the River Ouse in Cambridge, metal that would allow it to have the same function it has in United Kingdom. She looked at reactions, such as the the complicated protein matrix. Ultimately they would like fermentation of glucose, and concluded that there was an to attach them to carbon electrodes. enzyme in the cells that produced hydrogen and enzymes that absorbed hydrogen in processes important to the produc- Hydrogenase Structures and functions tion of methane. In one of Stephenson’s papers she noted, “Bacterial coli has been shown to catalyze the oxidation Darensbourg then described the structures and functions of dihydrogen to two protons, releasing two electrons in a of the hydrogenases (Figure 3.1), pointing out the com- completely reversible way. The hydrogenase system is the mon factors of the active sites of two classes of iron-sulfur most negative reversible oxidation reduction as yet described cluster containing hydrogenases (Figure 3.2), which are in living cells.” Later, this activation was found to proceed genetically distinct and evolved along different pathways, yet from the heterolytic splitting of dihydrogen (i.e., to a proton ended up at the same point in terms of function. She looked and a hydride). Each decade following the initial discovery more closely at the all-iron hydrogenase, pointing out the of hydrogenases brought tremendous advances, but in this iron-sulfur clusters that serve as a molecular wire into the century, Darensbourg noted another motivation for research: hydrogen cluster that are a built-in electron-delivery system. applying them to biotech challenges. She pointed out a second coordination sphere effect on the Darensbourg described how Fraser Armstrong (Univer- active site, which, if modified, will reduce the activity of that sity of Oxford) and Antonio DeLacey (Universidad Autonoma enzyme. With such complexities, the obvious question posed de Madrid) are making enzyme electrodes in order to estab- by Darensbourg was: “Will a small molecule modeled solely lish the hydrogen uptake and hydrogen production ability on the two-iron subsites be an electrocatalyst for hydrogen of the enzymes. Armstrong has calculated that on a graphite production or hydrogen uptake in oxidation?” electrode, nickel-iron hydrogenase catalyzes hydrogen oxi- Darensbourg, Tom Rauchfuss, and Chris Picket saw a dation at a diffusion-controlled rate that matches the rate simple diiron organometallic molecule as an obvious mimic achieved by platinum. However, said Darensbourg, there are of the iron hydrogenase active site. Darensbourg said that problems associated with the enzyme-electrode technology: this comparison was a good starting point for modeling 5

OCR for page 15
 BIOINSPIRED CHEMISTRY FOR ENERGY A B S. pasteurianum Sugars Methanogens CO2 CH4 Acetogens CO2 Acetate Fddx 2H+ Pyruvate Sulfate reducers SO H2S 2– 1 H2ase Iron reducers Fe Fe 2+ 3+ Acetate + CO2 H2 Fdied Denitrifiers NO3 N2 Nitrogen fixers N2 NH4+ CO2 Photosynthesizers Sugars Protons Acetate + CO2 Chemolithotrophs FE4S4 C FE4S4 H cluster H2 FE4S4 Electrons D FE2S2 Hydrogenase FIGURE 3.1 Schematic of hydrogen metabolism and the hydrogenase active site. (A) The cell of C. pasteurianum whose metabolism involves the oxidation of sugars and evolution of hydrogen by the 3-1.eps iron-only hydrogenase designated as a hexagon. (B) The range of organisms that use hydrogen as a reductant and use the nickel-iron uptake hydrogenase. (C) Schematic of the iron-only hydrogenase enzyme showing paths for electron and proton transfer converging at the H center. (D) Schematic of the H center showing the six-iron cluster with a two-iron subcluster bound to five CO or CN- ligands. SOURCE: Adams, M.W.W. and E. I. Stiefel. 1998. Biological Hydrogen Production: Not So Elementary. Science 282(5395): 1842-1843. [FeFe]-Hydrogenase [NiFe]-Hydrogenase H2 e- e- X Cys Cys S Cys S [4Fe4S] Cys S S O C Fe Fe S S OC C C N Hδ+ Ni Fe C C C H2 NL H N O O S δ+ H CN Cys LH H+ H+ X = CH2 , NH, O?? A B FIGURE 3.2 Examples of two of the main iron-sulfur cluster containing hydrogenase active sites. SOURCE: Modified from presentation of Marcetta Darensbourg, Texas A&M University, based on crystal structures derived by (A) J.W. Peters and coworkers, 1998. X-ray crystal structure of the Fe-only hydrogenase from Clostridium pasteurianum to 1.8 Angstrom Resolution. Science 282(5395): 1853-1858; and (B) J.C. Fontecilla-Camps and coworkers, 1995. Crystal structure of the nickel-iron hydrogenase from 3-2.eps Desulfovibrio gigas. Nature 373: 580-587. studies in conjunction with vibrational spectroscopy of feature can be monitored by nuclear magnetic resonance diatomic ligands. This could be used to match properties (NMR) spectroscopy. Darensbourg noted that additional of the enzyme active site with the small molecule models. flexibility is in the Fe(CO)3 units on each end of the diiron The diiron organometallic molecule has several attractive complex, which shows intramolecular CO site exchange features, such as the flexibility associated with the iron also detectable in variable temperature NMR experiments. dithiacyclohexane ring as it switches between chair-boat As a result of studying the fundamental properties of the forms, flipping the bridge-head carbon in the process. This molecule, Darensbourg and others found that there are still

OCR for page 15
 FUNDAMENTAL ASPECTS OF BIOINSPIRED CHEMISTRY FOR ENERGY some key differences between the structures of the enzyme’s geometry of the iron(II)iron(I)oxidized species, which is active site and the synthetic model. She claimed the big- the same as adding an electron to the species. The HOMO gest difference is the orientation of the diiron sites, which of the reduced species in a rotated orientation has electron are composed of sulfur-bridged square pyramids. In the delocalization over the bridging carbonyl and a large amount synthetic analogue, these square pyramids are symmetrical of electron density at that open site. The significance of these with respect to each other; the apexes of the pyramids are rotated forms, explained Darensbourg, is that the HOMO is pointed out and away. However, in the enzyme, one square located on the accessible face of the iron and is poised to pyramid is inverted or rotated with respect to the other. take up a second electron and a proton to make dihydrogen When that rotation occurs, said Darensbourg, it positions a in the reduced form. In the oxidized form, the SOMO (singly carbon monoxide underneath the iron-iron vector, which is occupied molecular orbital) is also on the open face, the very important in preventing reactions that are dead ends for accessible face, and is poised to give up an electron. If this the catalytic cycle. Darensbourg and others found that the configuration is maintained, there would be two iron (II)s following tweaks to the molecular models are necessary to and the species would be amenable to binding of H2. In other produce a more precise synthetic mimic: words, said Darensbourg, this is the inactive or resting form of the potential catalyst and it matches the resting form of the • Multiple and asymmetric substitutions; enzyme that was isolated and structurally characterized. • Redox active ligand; Darensbourg asked, “Does our molecule, with that • A rotated structure that yields a bridging or semi- seemingly open face, do anything? Will it bind dihydrogen or bridging CO ligand; and will it bind CO?” If we change this highly sterically hindered • Access to a stable Fe(II)Fe(I) complex. N-heterocyclic carbene ligand to the dimethyl N-heterocyclic carbene, we see CO binding and we see a stable carbonyl adduct. Other questions posed by Darensbourg included: Asymmetric Model Compounds “What are the radical properties of this molecule? Will it bind Darensbourg explained that early work found facile hydrogen atoms? Will intermolecular CO exchange occur routes to asymmetric model compounds that demonstrated rapidly?” She said that they are exploring the molecule’s electrocatalytic hydrogen production in the presence of stability under carbon monoxide and then will look at CO added aliquots of acetic acid. However, platinum is still a exchange with added 13C-labeled CO. much better catalyst under the same conditions. Darensbourg The key conclusions of Darensbourg’s presentation then discussed how graduate student Tianbiao Liu explored were: derivatives of diiron carbonyl complexes with multiple CO • The unusual ”rotated” structure of the enzyme ligand substitutions. Liu saw that in the cyclic voltamogram of an N-heterocyclic carbine(NHC)/trimethylphosphine active site is achieved in a mixed-valent Fe(II)Fe(I) complex, complex there are reduction events at –2 volts, close to the which uses the unique orientation of a bulky NHC ligand to solvent window. The oxidation, however, was fully reversible protect the open site on the rotated iron. • The odd electron is on the open face of the rotated at –.47 volts. As this oxidation wave is removed from every- thing else, Liu concluded that he might be able to isolate the Fe(I). • The 17-electron Fe(I) promotes CO exchange with oxidized product, and indeed he did. The oxidized product can be reduced back to the original compound, making it a exogeneous 13CO. • This structure mimics the resting state of the fully reversible redox event, chemically and electrochemi- cally. There was a reorientation of the NHC side of the com- [FeFe]hydrogenase active site. The enzyme holds this con- pound, while the iron dicarbonyl phosphine side remained formation in position throughout proton/electron coupling/ the same. This mixed-valent Fe(I)Fe(II) complex looked very decoupling reactions. • What sort of synthetic matrix or solid support might much like the active site of the all-iron hydrogenase. Darensbourg posed the question: Which is iron (I) and restrict reverse rotation in such “rotated” structures? which is iron (II)? Mike Hall and Christine Thomas (Texas A&M University) are looking at this structure using density New Chemistry functional theory (DFT). The HOMO-1 and HOMO-2 are Thomas Rauchfuss, University of Illinois at Urbana localized to the IMes NHC ligand. From the DFT study various parameters can be extracted, including unpaired introduced his presentation by explaining that there is a lot spin density. Hall and Thomas found that the unpaired spin of organometallic chemistry occurring in nature. In addition, density lies primarily on the rotated iron, rather than the he said the country’s future is likely to be tied to synthetic unrotated one, implying that the oxidation, the iron (II), is gas, so there is a need for research on bioinspired syngas-like on the latter while the open-site iron is iron (I). The HOMO chemistry. Some of the key actors discovered so far include of the starting material is the iron-iron bond. Thomas has the aerobic CODH (CO dehydrogenase). He noted that taken the iron(I)iron(I) precursor and twisted it to match the hydrogenases are remarkable; they evolved independently

OCR for page 15
 BIOINSPIRED CHEMISTRY FOR ENERGY three times over 3 billion years and each time produced an minal, it picks up protons and makes H2. If it is left alone, iron carbonyl. it isomerizes and gives a bridging hydride, which does not Rauchfuss’ research team is interested in connecting H2 form H2. Rauchfuss sees this as an incredibly versatile and to iron to activate hydrogenation in fuel cells. There are only robust system. three crystal structures on the hydrogenase compounds, and Next, Rauchfuss explained how his team has used they are extremely precious, he said. The structure of the CO- various chelating ligands to manipulate the symmetry of inhibited structure has been revised based on IR data. diiron models. The introduction of chelating agents changes He presented the design of the hydrogenases and pointed the relative basicity and electronic asymmetry of the diiron out that one of the major problems is that these systems are models. His team has determined that electron-rich diiron subject to dynamic equilibria. All of the substrates, electrons complexes, made possible using chelating diphoshine included, are transported in and out of the active site in a very ligands, are both redox active and Lewis basic (at the specific way based on numerous studies. oxygen of the CO ligands). The redox chemistry of the Greg Kubas of Los Alamos National Laboratory has diiron complex is sensitive to the presence of substrate and determined how hydrogen interacts with metals. The impor- inhibitors used to determine what to bind to the extra posi- tant part of his work is that hydrogen, a substrate that is tion. Rauchfuss’s team has examined a one-electron oxidized redox inactive substrate and not Brønsted acidic, transforms diiron model and is now wondering what happens if they upon complexation whereupon the coordinated H2 becomes doubly oxidize it. acidic. The deprotonation of a metal dihydrogen complex The preparation of models for the Hox state of the hydrog- generates oxidizable species and in this way, H2 is connected enase is an important breakthrough, noted Rauchfuss, and his to electrons and heterolytic activation. Rauchfuss explained group would never have considered this target without the that Kubas’ discovery has helped guide his team’s effort to guidance provided by structural biology. He also showed connect H2 binding to this redox-active iron metal. some of the other types of reactivity for the mixed-valence Rauchfuss presented the catalytic cycle for the diiron complex and explained that the power of synthetic hydrogenases and pointed out two states that are most likely organometallic chemistry is contributing new concepts in stopping points in the cycle (Figure 3.3). In his models, hydrogen-activation and hydrogen-relevant chemistry. instead of using a complicated dithiolate, he is using a simple ethane dithiolate and replacing complicating cyanides and Noninnocent Ligands the iron-4 sulfur-4 cluster with phosphine ligands. It is a robust system, and very ordinary old Wilkinson-style ligands Rauchfuss’s team is working on noninnocent ligands, are used to support this complicated chemistry. After a lot a family of ligands in which the oxidation state is unclear. of work, weak ligands can be installed on one iron and then A quinone is a typical noninnocent ligand that has not been subsequently replaced by a hydride. If the hydride is ter- used very much in hydrogen activation. Using such non- FIGURE 3.3 Catalytic cycle for FeFe-hydrogenases, with two most likely stopping points in the cycle highlighted in red. SOURCE: Presentation of Thomas Rauchfuss, University of Illinois, Urbana-Champagne. 3-3.eps bitmap image

OCR for page 15
 FUNDAMENTAL ASPECTS OF BIOINSPIRED CHEMISTRY FOR ENERGY innocent ligands, the team is working on simulating the role of the iron-4, S4 cluster in the hydrogenases. The team is trying to address the following questions: What happens when you put this system on a metal that might activate hydrogen? Can you use this design to encour- age hydrogen activation? Rauchfuss thinks the results look promising. The ligand is active and it binds to virtually any substrate provided. If a base and hydrogen are provided to these complexes, the hydrogen is oxidized to protons. This is a system in which a metal that is otherwise uninteresting is “turned on” due to a ligand-based redox. The challenging part of the fuel cell is O2 reduction. Rauchfuss’s team is interested in whether bringing an organometallic perspective to that type of reactivity. The team is wondering whether the old reaction of O2 plus H2 will work. Part of the problem, explained Rauchfuss, is that most compounds used to hydrogenate oxygen in a potential fuel cell application would produce hydroxides. A new family of hydrogen activating species is coming online in the near future. There may be a role for heterolytic activa- tors of hydrogen in oxygen reduction for fuel cell develop- ment. Progress with these activators was illustrated with an experiment conducted in an NMR tube, a Knallgas reaction. Nickel-iron hydrogenases carry out and effect a similar H2 + O2 reaction to provide energy for certain bacteria. Rauchfuss concluded by restating his main point: Syn- thesis enables translation of mechanistic insights into cataly- sis and is a critical component of the overall bioinspired effort. Even though redox chemistry and hydrogen seem FIGURE 3.4 Light-harvesting peripheral antenna complex from quite old, the field is wide open for new discoveries. green-sulfur bacteria consisting of self-assembled arrays of chloro- phyll molecules. ARTIfICIAL PHOTOSYNTHETIC SYSTEMS fOR SOURCE: A. R. Holzwarth, Max Planck Institute. SOLAR ENERGY CONVERSION At Northwestern University Michael Wasielewski and his team are trying to understand various biological processes relevant to energy, especially photosynthesis. They hope to each case chlorophyll’s basic asymmetry gives one transition achieve different protein environments and different juxta- dipole moment, which is oriented in a particular direction. positions of the cofactors relative to one another to elicit a The coupling of the transition dipole moments are critical to specific function of bioinspired and biomimetic systems. the function of chlorophyll in spectral forms and the energy This is critical, said Wasielewski, since society depends transfer properties gleaned from them. heavily on frontline synthesis. Wasielewski stressed the importance of avoiding groups In a biomimetic study being done by his team, the in the building block that could interfere with those positions. peripheral antenna complex from green-sulfur bacteria was His team developed a new functionalization strategy for the investigated (Figure 3.4). The bacteria are unique because 20 positions of chlorophyll so that there is a hook to attach chlorophyll is associated with protein. The metal ligand other species and to use without getting in the way of self- bonds to protein, which bonds the chlorophylls to the protein. assembly points of interest. The particular antenna complex that he presented is unique Wasielewski then focused on a particular ring structure because it relies on chlorophyll-chlorophyll interactions to found in antenna proteins to see how a system could be produce a micellar structure. Wasielewski’s team has been developed based on chlorophyll that mimics some of these able to use the ability of chlorophyll to glom onto itself features. It turns out that the spectral shift does not explain to study some of the issues pertaining to energy transfer. anything. He said that there needs to be a structural tool to Magnesium requires five ligands, so the fifth ligand can be present specific information. The Advanced Photon Source, one of the oxygen atoms of a corresponding nearby chloro- the brightest X-ray source in the country, at Argonne National phyll, such as the carbonyl group or a bridging ligand. In Laboratory, is currently being used for this purpose.

OCR for page 15
0 BIOINSPIRED CHEMISTRY FOR ENERGY Wasielewski is also working on determining the struc- Dismukes, the use of platinum as the electrode material for ture of a four-fold symmetric cyclic ligamer that is forming oxidizing water requires a large overpotential to drive the a structure spontaneously in solution. His team is looking at reaction and thus is not ideal. Nonprecious metal catalysts spectroscopy to excite the system and identify any energy are needed for water oxidation and integration into cells that transfer and is putting two excitations in the molecule to use light in terms of the overall reaction. study energy transfer. The result of the four fold symmetric system is that energy transfer occurs incredibly fast, at about Photosystem II Water-Oxidizing Complex a picosecond. This demonstrates that energy transfer in the self-assembled system is faster than most porphyrin systems Dismukes explained how photosynthesis splits water, involving covalent linkages and it is almost as fast as some produces oxygen, and extracts electrons and protons to of the quickest natural systems. Use of robust components undergo fixation of carbon dioxide. The enzyme that carries in bioinspired systems is one that has become a major theme out the water oxidation process is called photosystem II in Wasielewski’s group. water-oxidizing complex (WOC). There have been devel- The team is now interested in electron transfer in a opments in the last three years on the crystallography of stacked, noncovalently linked system, with an electron proteins involved in the process. Prior to that, chemistry and photochemically pumped in. If four of these molecules spectroscopy have provided much data about functions of are placed around a porphyrin, explained Wasielewski, the the enzyme. Three crystal structures has been reported for system self-assembles into a large aggregate and because of this enzyme, which has evolved for about 3 billion years to the side groups in the system, an interlead aggregate results achieve its current catalytic efficiency. Dismukes said that where every other layer is missing a porphyrin. Synthesis there is only one example of this evolved enzyme, which is of the building block occurs, leading to a new type of self found in all terrestrial plants, green and red algae, and bac- assembly. Side groups were eliminated and some long-tail teria. There is no variation of this blueprint across the entire end groups were substituted to aid solubility. range of oxygenic phototrophs. Given a differential recombination of charge when an ion Dismukes described how coworker Gennady Ananyev pair is formed and the recombination rates are different, the and students in Ananyev’s group are characterizing oxygenic direction of charge transport can be controlled by choosing phototrophs that can operate at pH 0 up to pH 12 in every which direction the electron enters. This is called a chlorophyll redox environment and many toxic metal environments. The mimic. It has the same oxidation potential of chlorophyll and emergence of a single enzymatic blueprint for catalyzing absorbs in the same place that chlorophyll does. water oxidation chemistry is a critical clue that Dismukes Wasielewski noted the importance of compartmentaliza- says should not be overlooked in the design of engineered tion in generating hydrogen and oxygen by splitting water. catalysts. These enzymes constitute nature’s optimal design He discussed a paper that his team recently published 1 achieved through combinatorial biosynthesis. The bio- describing how a specifically tailored perylene diimide-type inspired approach relies on adopting nature’s blueprint for system can build a nanotube. Wasielewski concluded that this catalyzing the lowest-energy five-bond rearrangement neces- kind of approach shows what the next step of bioinspiration sary for the water splitting reaction: will be in developing systems for artificial photosynthesis. 2 H 2O → O 2 + 2 H 2 WATER SPLITTING BY BIOINSPIRED CATALYSTS Working with graduate students Jyotishman Dasgupta Charles Dismukes of Princeton University focused and Rogier Van Willigen, Dismukes described their proposed his presentation on one reaction that splits water to create mechanism for how the native WOC enzyme catalyzes oxygen. He highlighted new developments in bioinspired oxygen production from water (Figure 3.5). After accumula- catalysis that mimic the active site of the water-splitting tion of four holes and release of protons into solution, the enzyme of green plants and other oxygenic phototrophs. highest oxidation state of the cluster is reached. The oxida- Drivers for this research include economics, gasoline prices, tion states represented are not unequivocally established political issues, and the environment. Dismukes said, “I like by any of the spectroscopy thus far. A chemist looking at to think that we have a golden opportunity right now because that structure will think it is intrinsically unstable based on of the motivation that many of the young people are experi- the structure of the bridging tetrahedral oxygen atom. The encing from these forces.” oxygen prefers to rearrange into a coplanar arrangement The availability of platinum is a limitation to hydrogen with three manganese atoms. In other words, said Dismukes, production at the anode and for oxygen reduction at the the oxygen will sacrifice a weak single bond to a calcium cathode of fuel cells. Unlike the anode reaction, explained ion in favor of forming a multiple bond to three manganese atoms. In this view of the mechanism, a tetrahedral oxygen atom would go coplanar, cleaving the bond and allowing the Sinks, Rybtchinski, Jones, Goshe, Zuo, Tiede, Li, and Wasielewski, 1 calcium to move over to bind to a peroxide intermediate that 2005. Chem Mater 17: 6295-6303.

OCR for page 15
 FUNDAMENTAL ASPECTS OF BIOINSPIRED CHEMISTRY FOR ENERGY FIGURE 3.5 One of the postulated pathways for the O2 release step of the WOC. The naturally occurring WOC of photosystem II is able to efficiently photooxidize water in a sustainable manner using visible light according to the reaction: 2 H 2O → O2 + 4 H+ + 4 e–. SOURCE: Presented by Charles Dismukes. 3-5.eps includes bitmap image, arrow only is a vector object forms between what were formerly two oxo bridges. This no net photoreaction occurs. In the condensed phase the is the proposed slowest step and represents the activation barrier to O2 release is too large to surmount rapidly such barrier to forming the highest energy intermediate of the that the phosphinate does not dissociate or rebinds so fast reaction. Subsequent release of O2 by a further two-electron that it prevents the O2 from forming. Thus, if an open-face transfer reaction from the peroxide to manganese is thermo- cubane with a lower barrier to O2 release could be prepared, dynamically favored and occurs spontaneously. the system could in principle be used in a catalytic cycle to convert water into O2 + 4 H+ and 4e–. All of the gas phase work has been published. Manganese-oxo Cubane Dismukes spoke about his team’s recent efforts with Dismukes showed a manganese-oxo cluster sharing Australian collaborators from Monash University (Robin structural features with the WOC, which exhibits O2 forma- Brimblecombe and Leone Spiccia) and Commonwealth tion by an analogous pathway. These manganese-oxo cubane Scientific and Industrial Research Organisation (Gerhard molecules possess the Mn4O4 core type and were unknown Swiegers) to examine methods to achieve a catalytic water core types in inorganic coordination complexes until first oxidation cycle by doping cationic cubanes compound 1+ synthesized by graduate student Wolfgang Ruettinger (com- (Figure 3.6) in the aqueous channels of proton-conducting plex 1 in Figure 3.6). membranes like Nafion®. Nafion is a fluorinated polymer with Dismukes discussed the oxygen-evolving chemistry ionizable sulfonic acid head groups that remain hydrated in in the gas phase (Figure 3.6). When the molecules are water and form aqueous channels that are about 20 nm in vaporized into the gas phase using UV-visible light, most diameter. The channels are permeable to cations but not of them release oxygen (60-100 percent depending on anions since they are “lined” with sulfonate groups whose choice of phosphinate derivative). This is a laser desorp- charge is balanced by mobile cations (H+ or Na+). Nafion is tion ionization-mass spectrometry experiment (LDI-MS). readily deposited as a thin layer upon electrode surfaces. The In the gas phase they can either thermalize to form the cationic cubane species was doped in thin Nafion films by ion unmodified cubane in its ground state or dissociate by exchange in acetonitrile. Replacement of the CH3CN solvent releasing a phosphinate ligand and an oxygen molecule. by water traps the hydrophobic cubane in the channels of Importantly, said Dismukes, the only other product is the the film. Voltammetry detected the redox transition 1↔1+, intact “butterfly” compound 3, L5Mn4O2+ (Figure 3.6), as thereby unequivocally confirming the presence of the cubane was shown in the positive-ion LDI-MS. Dismukes found that in the Nafion film. It also established that direct electron removal of a single phosphinate is required to achieve the transfer occurred readily between the immobilized cubane flexibility needed to form and release O2 in the gas phase. and the underlying electrode. Calculations by Princeton colleagues Filippo DeAngelis When polarized at a potential of 1 V (vs Ag/AgCl), and Roberto Car showed that the activation barrier for O2 the resulting electrode assembly generates a transient dark release from the resulting L5Mn4O4+ intermediate is much current corresponding to the complete oxidation of 1/1+. smaller at 23 kcal/mol. This is called the jack-in-the-box Subsequent illumination by UV-visible light generates a mechanism for oxygen release. When the photolysis is car- large increase in this current, which systematically and ried out in condensed phases, either solid state or in organic reproducibly tracks with the duration of the light interval solvents that dissolve the cubane, there is no O2 release and as it is switched on and off. Gas bubbles form at the

OCR for page 15
 BIOINSPIRED CHEMISTRY FOR ENERGY FIGURE 3.6 Redox reactions and photochemistry of Mn4O4(Ph2PO2)6. 3-6.eps SOURCE: Presented by Charles Dismukes. bitmap image photoanode, and analysis by GC-MS confirmed this to arise light is necessary to knock off the ligand. As they investigate from isotopically enriched 36O2 (produced using 18O water). the possibility of using catalysts that can accept weaker-field Several experiments illustrated the wavelength dependence, phosphinate ligands, they can replace the stronger phosphi- the pH dependence, and the solvent dependence, all of which nates with weaker-binding phosphinates or possibly other confirm that water is the oxidizing reactant for O2 produc- ligands. His group also thinks they can make some hetero- tion. The electrochemical conversion of charge was shown cubanes by fusing two classes of metal dimers. Dismukes to match the volumetric yield of O2. Dismukes thinks these said that it would be helpful to try to put alkaline earth ions in results hold promise to an exciting new approach for water there. They plan to look at other proton exchange membranes oxidation based on the principles inspired by nature. This on the market. He also said that some conducting polymers work has been submitted for publication. would be very helpful. They think they are processing only At the end of his presentation Dismukes talked about the the first 50 nanometers or so the catalyst, so they want to next steps for his research. His group does not want to use include polymers that would allow them to be transferred to UV light to activate the system, but they understand that the a much farther distance.

OCR for page 15
 FUNDAMENTAL ASPECTS OF BIOINSPIRED CHEMISTRY FOR ENERGY DISCUSSION won’t have patience for our excitement if it’s wrong. So we should hash this out among ourselves and then present only During the discussion period, some participants were to the best of our ability what we know is right.” Thomas skeptical of Charles Dismukes’s research. Thomas Moore Rauchfuss of the University of Illinois agreed with Nocera’s of Arizona State University asked Dismukes whether he point and said that a good example is the overenthusiasm for could show quantitatively that a molecule of cubane really hydrogen. He thinks people are under the impression that turns over multiple times. Dismukes said that the electro- when hydrogen is produced by electrolysis a major problem chemistry data shows that after a cycle of the system, has been solved. However, Rauchfuss feels the bigger barrier electrochemical potentials can be seen for the 2, 2+couple. is integration of hydrogen into the energy infrastructure. He His team is trying to take the material that has been cycled also explained how ethics comes into play when scientists and do X-ray spectroscopy on it. They have also looked at want to promote their work to the public in order to receive photodegradation products like manganese dioxide. Gary more funding. On the other hand, Judy Raper of the National Brudvig of Yale University pointed out that Dismukes claims Science Foundation said that it is important to get students hundreds of thousands of turnovers, which is much higher interested in engineering and science, so there has to be a than a natural system. Dismukes backed up his research balance between ethics and informing the public about the and explained that the molecules produced oxygen in the research being done. gas phase with very high quantum yield and with no other Sharon Haynie closed the fundamental aspects session photochemistry. by pointing out the promise and perils of the bioinspired Daniel Nocera of MIT raised an important issue about research. Haynie thinks there is promise because the public how scientists approach the public with regard to the details understands the current energy crisis; however, over- of their research. Nocera warned the group that they can lose promising the research can lead to peril by disillusioning the public’s confidence if they are overenthusiastic and do and alienating the public. not back up their research results, stating that, “The public

OCR for page 15