Bioinspired Chemistry for Energy: A Workshop Summary to the Chemical Sciences Roundtable (2008)

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

4 Robust Implementation of Bioinspired Chemistry for Energy During this session of the workshop, Thomas Moore, the membrane potential drops by a factor of two, the cell dies. G. Tayhas Palmore, and Mark Emptage presented examples The high efficiency of energy transduction by mitochondria of how bioinspired chemistry for energy is being imple- is the fundamental difference between standard fuel cells and mented within their organizations. These projects utilize mitochondria (Figure 4.1). some of the key fundamentals described in Chapter 3 to Moore talked about the opportunity to produce electricity present the big picture of how bioinspired chemistry can with photovoltaics, but noted that a really important advance improve the energy field. would be to use sustainably produced electricity to produce fuels for the transportation sector. Nature has the catalysts that direct reactions such as the reduction of CO2 to reduced, Mimicking biological Energy Transduction energy-rich carbon compounds. These catalysts operate at Thomas Moore of Arizona State University began room temperature and yield essentially pure products. The by discussing how nature has adapted for 3.4 billion years problem is that these natural catalysts do not “recognize” elec- through fierce evolutionary competition. He said that tromotive force and cannot use it efficiently to drive synthetic researchers need to follow nature’s example, but accept reactions. The research challenge is to focus on effectively the fact that in meeting human energy needs there may be “wiring” enzymes into sources of electromotive force. He ways to improve on nature. He defined the purpose of the said that discovery is a key to achieving sustainable energy workshop as a way to explore how to improve the natural production and use. Arizona State University, for example, has process of photosynthesis and incorporate the kinetically and a large metagenomics program in which researchers travel to thermodynamically successful features in human-engineered various locations looking for microorganisms with catalysts constructs. that can carry out useful reactions. Moore compared the technological branch of solar energy Moore also compared water splitting by photosynthetic conversion, essentially photovoltaics, with the ­ biological systems with a human engineered system consisting of three branch. He explained how a standard fuel cell that operates PV cells operating in series driving a commercially available on oxygen and hydrogen produces water and electromotive electrolyzer (Figure 4.2). Series operation is necessary for force. A typical human-engineered fuel cell operates at 50- Si-based PV cells to provide the voltage necessary to oxidize 60 percent power conversion efficiency and uses platinum or water and reduce protons to hydrogen in the electrolyzer. He other noble metals as catalysts. pointed out that photosynthesis uses two photons per electron Moore then explained how mitochondria are biological and has a threshold for absorption of light at about 700 nm. fuel cells. The oxygen reduction taking place in a mito- The PV cells need three photons per electron but, because chondrion is exactly the same as in a standard fuel cell. their threshold is about 1100 nm, which means they gather Using several enzymes and only earth-abundant elements, about twice as many of the available photons, they are prob- the mitochondrion converts electrochemical potential to ably more efficient at water oxidation than photosynthesis. biochemical work with efficiency greater than 90 percent. Moore highlighted several projects at Arizona State Uni- This is a steady-state process in which protons are pumped versity that are using biological and bioinspired chemistry across the membrane to maintain its electrical potential. If for energy discoveries. 25 

26 BIOINSPIRED CHEMISTRY FOR ENERGY M H+ e H+ m H+ b H+ H+ H+ r a H+ n H+ e H+ H+ H+ H+ Mitochondrion as a fuel cell H+ H2/air fuel cell 2NADH + O 2 → 2NAD+ + 2H2O 2H2 + O2 → 2H2O ~20H+in → ~20H+out Conversion of electrochemical Conversion of electrochemical potential potential to biochemical work to human-directed work with modest with high (> 90%?) efficiency. efficiency (~ 60%). Uses Pt Input: ~ 1 V redox; Output: pmf, Input: ~ 1.1 V redox; Output: emf ~20 H+ across ~ 200 mV Figure 4.1  Comparison of mitochondria and engineered fuel cells. SOURCE: Presented by Thomas Moore, Arizona State University. 4-1.eps includes 3 bitmap images sized for oblong to preserve definition on that smallest bitmap Photosynthesis Silicon Photovoltaic λ threshold = 680 nm λ threshold = 1100 nm 2NADPH 2H2 - - Pt Si Si 4H+ 2NADP+ + + cyt PSI - PSII Si + 2H2O O2 + 4H+ 2H2O O2 + 4H+ - 2 photons / e 3 photons / e- AM 1.5 Solar 280 nm to 680 nm AM1.5 Solar 280 nm to 1100 nm = 1.19 x 1017 photons/sec/cm2 = 2.71 x 1017 photons/sec/cm2 Assume 100% LHE and QY Assume 100% LHE and QY therefore 5.95 x 1016 e-/sec/cm2 therefore 9 x 1016 e-/sec/cm2 = 9.5 mA/cm2 = 14.5 mA/cm2 With leftover λ = 700 to 1100 nm photons Figure 4.2  Comparison of photosynthesis and a silicon-based photovoltaic system. SOURCE: Presented by Thomas Moore, Arizona State University. 4.2 new 

ROBUST IMPLEMENTATION OF BIOINSPIRED CHEMISTRY FOR ENERGY 27 Biodiesel to Fuel a Large Power Plant. Researchers providing the opportunity to produce reduced material. By at ASU’s Center for Bioenergy and Photosynthesis have using a platinum carbon cathode, hydrogen can be produced. calculated that a 25 × 25 km field of bioreactors using cyano- Essentially, every two electrons produce one hydrogen from bacteria to fix carbon could uptake all of the carbon dioxide this process. In collaboration with a group at the National produced by a 1.6 GW power plant and subsequently provide Renewable Energy Laboratory (NREL), scientists at Arizona the biomass as lipid to fuel the power plant. The parameters State University have been putting hydrogenase enzymes necessary to achieve this goal are a seven percent power on a carbon electrode to successfully produce hydrogen. conversion efficiency for photosynthesis, 40 percent con- Hydrogenase enzymes use only iron and nickel metals to version efficiency of biomass to fuel, 50 percent conversion carry out the catalytic process. efficiency of fuel to electricity, and 80 percent conversion efficiency of land area covered by the bioreactors. This sys- Failed Water Oxidation Experiment. Although direct tem would then be carbon neutral in operation and produce water oxidation using an artificial photosynthetic device about 1.6 GW of electrical power. The key to making this is not possible because the necessary catalyst does not yet feasible is to achieve a seven percent power conversion effi- exist, Moore is developing an approach to “push the enve- ciency for cyanobacteria. Moore noted that the area required lope”. His group is working to create a hybrid system that to produce a specified amount of energy scales directly with would use a naturally occurring photosystem II to oxidize the energy conversion efficiency of the system or device. water. Moore explained that the anticipated electron flow would be from illuminated PS II particles, oxidizing water, Tubes in the Desert. This project offers one approach to the photoanode, where the process described above would to the design of bioreactors to produce biodiesel using the produce hydrogen. Thus two photons would be used to photosynthetic capabilities of microalgae or cyanobacteria. move an electron from the redox level of water to the level Solar energy conversion by photosynthetic organisms offers of ­ hydrogen—much in the same way that natural photo­ the almost inestimable advantage of self assembly, self synthesis uses two photons to move an electron over roughly replication and self repair of the key photoconverter. Cyano­ the same redox span. The output of photosystem II is a bacteria in bioreactors do not require arable land, which, reduced quinone, a quinol. The quinols carrying the electrons Moore says, avoids the conflict of using agricultural land for from water would then be the input into the artificial system energy production. The bacteria grows quickly year-round to produce hydrogen. Moore noted that the aim was to create with a doubling time of 0.5-1.0 day. The reactors do not use a hybrid system to explore ways to couple two photosystems a large amount of water, which is good for dry locations, such together and to investigate the thermodynamic and kinetic as Arizona. There is a high output of biomass: The practical constraints involved. Unfortunately, this experiment has not range is 40-100 tons/ha/yr dry biomass and the optimal level worked to date. is 500 tons/ha/yr. The output can be a high-value liquid fuel Moore explained that the experiment failed because created with genetic and engineering controls. The process using the quinols as electron donors to the photoanode is not cellulosic or lignocellulosic, explained Moore. recombination limited the output potential to values that were not negative enough to reduce protons to hydrogen. When Hydrogenase on a Carbon Electrode. Michael ­Graetzel NADH is the electron donor, photo currents for hydrogen designed a system that permits nanoparticulate material to appear. When NADH is oxidized, a radical cation with a pKa absorb light using a dye. After it absorbs the light, the dye of -4 remains. This cation immediately drops the proton off, injects electrons into a conduction band. This creates electro­ generating a species that donates a second electron. There motive force through an interface between a molecular is no recombination reaction and no recombination current system and a semiconductor. This experiment involved a from the mediator to the anode with that mediator. When Graetzel-type photo anode being sensitized by a porphyrin a quinone mediator is used, the first electron is removed that upon excitation puts an electron into a wire that connects from the quinone and the radical species formed, and vari- to a cathode. If the cathodic reaction is positive and there is ous semiquinones are in the electrochemical region where a good catalyst, Moore explained that electromotive forces recombination reactions from the electrode can occur. The will be produced. But if the cathodic reaction is negative in rate of this recombination limits the potential the cathode electrochemical potential, such as proton reduction, energy- can reach. Moore talked about trying to use an electrode rich fuel is produced. After the electron injection, a radical with a blocking layer to solve the problem. He concluded cation of the dye is left. The cation then oxidizes a coenzyme, that the experiment raised some important questions about NADH, for which there are many different dehydrogenase efficiency, redox span, and the limitations that must be over- enzymes that can oxidize many different bioorganic ­materials come in wiring enzymes to electrical circuits. While nature and reduce the NAD. NAD is thus a mediator or carrier of obviously “got it right” for nature’s purposes, there is room electrons between the bioorganic material and the photo- for improvement when using bioinspired processes to meet anode. This type of electrochemical cell is nonregenerative, human energy needs. 

28 BIOINSPIRED CHEMISTRY FOR ENERGY Biofuel Cells what is used in the anode and what is used in the cathode, G. Tayhas Palmore of Brown University presented a which is oxygen. diagram of the four components of energy conversion: bio- • The current density is also important and depends mimetics, purified or engineered enzymes, synthetic biology, on the rate of catalysis and the movement of ions across the and bacterial isolates. Palmore gave a brief tutorial on how membrane. fuel cells work (Figure 4.3). She explained how they convert • It is important to consider pH, temperature, and chemical energy into electrical energy by separating the pressure since conventional fuel cells use catalysts that func- oxidation and reduction reactions into two separate cham- tion at very low pH and high temperatures. bers using a membrane. The anode is the electrode where hydrogen is oxidized to protons. The oxidation of hydrogen Palmore continued her talk with some examples of releases electrons drives an electrical load, which is coupled research that has been done in the field. with the movement of ions through the membrane that drives the reduction of oxygen to water. For a hydrogen-oxygen fuel Itamar Willner and Eugenii Katz’s Glucose Fuel. cell, the open circuit voltage is 1.2 volts. The three-phase Wilner and Katz (Hebrew University of Jerusalem) used interface between a catalyst breaks the bonds. A current col- an enzyme-based biofuel cell that had direct contact (i.e., lector then takes the electrons that are released from those tethered to an electrode) to produce glucose as a fuel. The bonds away from the active site. Finally, said Palmore, an experiment showed the power of chemical control in orga- ion conductor moves the protons or ions released away from nizing the assembly on the surface. Wilner and Katz took the catalytic site to repeat the process. advantage of modified thiol-gold chemistry, a PQQ mediator Palmore also explained that the biomimetic approach with the thiol, and added a reactive carboxylic acid that could uses inspiration from biology to develop new chemistry be reacted with the FAD cofactor enzyme. Using a cathode to perform catalysis. Palmore focused on using biology to reaction, they also demonstrated a truncated peroxidase that do the energy conversion, both in the free form as purified was tethered to the surface. They demonstrated that open- enzyme extract and where new organisms have been isolated circuit voltage could reach approximately 140 microamps and coupled with electrode surfaces (Figure 4.4). per centimeter squared. The maximum power was at about According to Palmore, the following parameters are 32 microwatts. The voltage was low because peroxide was important for both bio- and conventional fuel cells: used as the oxidant instead of oxygen. • Fuels, such as methanol, ethanol, glucose, and Adam Heller’s Glucose Fuel. Another example of higher-level organics, can also be considered since they all enzyme-based glucose oxidation was done by Heller’s group have potentials similar to hydrogen. The fuel will depend on (University of Texas at Austin). They worked with redox- active polymers, which are molecular wires. Heller’s team used the osmium-bisbipyridine complexes and tuned them with their ligands to adjust the potential relative to an anodic fuel like glucose or a cathodic fuel like oxygen. The com- plexes were then immobilized with the enzymes to catalyze the conversion of glucose to gluconolactone and oxygen to water. The system demonstrated the following: • It is a membrane system, physically separating the anodic and cathodic reactions. • It survived about seven days before the current began to diminish. Palmore pointed out how difficult it is to reduce oxygen to water, and explained that it can be done using a number of enzymes in the copper oxidase family. She has used laccase because it is a monomeric protein but is unlike dehydroge- nases that depend on a cofactor and a binding site. Laccase has a redox center of four copper ions that are reduced from 4-3.eps Figure 4.3  A fuel cell converts chemical energy into electrical  I. Willner, E. Katz, F. Patolsky, and A. F. Buckmann. 1998. Journal of bitmap image energy. H2/O2: open circuit voltage ~1.2 V. the Chemical Society, Perkin Transactions 2, 1817-1822. SOURCE: Image courtesy of G. Tayhas Palmore, Brown  N. Mano, F. Mao, and A. Heller. 2003. Journal of the American Chemi- U ­ niversity. cal Society 125: 6588. 

ROBUST IMPLEMENTATION OF BIOINSPIRED CHEMISTRY FOR ENERGY 29 Figure 4.4  Microbial and Enzymatic Biofuel Cells. SOURCE: Image courtesy of G. Tayhas Palmore, Brown University. 4-4.eps bitmap image copper (II) to copper (I) with subsequent binding of oxygenfor oblong from biomass Initiatives at DuPont sized BioFuels to reduce it to water. Mark Emptage of DuPont stated that biomass is the Palmore presented some recent examples of simple only carbon-based renewable resource for fuel and that there chemistry done on the surface of proteins that can have a really needs to be a focus on biofuels. However, he believes tremendous effect on their overall stability. Jungbae Kim there is a limit to grain ethanol in the United States, and to and Jay Grate of Pacific Northwest National Laboratory used produce more ethanol, cellulosics need to be used. Emptage simple chemistry to modify lysine residues on enzyme sub- explained that there are many ways of converting biomass strates and then tethered them to siloxane, allowing it to gel to a variety of energy needs. He focused his discussion on and harden. This example combines chemistry, nanotechnol- hydrolysis of biomass to fermentable sugars and converting ogy, and biotechnology, and shows how chemists can affect the sugars into biofuels. Emptage focused on ethanol, but how we look at catalysts. explained that the technology developed for ethanol can be Palmore also talked about examples of microbial applied to other fuels as well. He believes biorefining infra- systems. Derek Lovley’s laboratory at University of Massa­ structure should be an add-on to current infrastructure. Next, chusetts, Amherst, is looking at doing electrochemistry in Emptage discussed the by-products of producing ethanol: ocean or bay sediments. New organisms are being discovered The lignin and biomass that result from separating out etha- through their research and the use of a microbial fuel cell nol can be used as fuel for the entire facility, and the glucose in which carbohydrates are taken to carbon dioxide using can be used for products other than fuel. Rhodoferax ferrireducens have been demonstrated. Emptage listed DuPont’s guiding principles for cellu- Palmore concluded by highlighting her wish list for losic ethanol process development: future research, which is detailed in chapter 6. 

30 BIOINSPIRED CHEMISTRY FOR ENERGY xanthophyl. Moore’s team is working on a model system that • Keep it simple with minimal unit operations and responds to high light and quenches excited states releasing separations and minimal capital investment. the energy as heat, and then as the light intensity comes back • Target a single agricultural residue at the beginning down again, the system shuts itself off. He also explained (corn stover is the largest source of agricultural residues in that the system needs to respond to the membrane potential the United States). and pH gradient across membranes. This can be done with a • Maximize titer. potential sensitive sensor. • Life-cycle assessment—need to understand overall R. David Britt of the University of California, Davis, process to make it sustainable. asked about the limits to purely biological approaches. • Process integration—need to have high efficiency Thomas Moore said he thinks natural photosynthesis needs to fit all the pieces together. to be reengineered to double or triple its power of conver- • Risk mitigation—take small steps within the current sion efficiency. He said that solar will ultimately solve the infrastructure. problem. Moore then called for research focused on fuels by photosynthesis created by cyanobacteria grown on nonarable Emptage explained the steps involved in converting land and photovoltaics for electricity. sugar to ethanol: milling, pretreatments, saccharification, Douglas Ray from Pacific Northwest Laboratories conversion of biomass to fermentable sugars, fermentation, asked G. Tayhas Palmore whether the process of engineering and separation. He said that the key technology is fermenta- enzymes needs to be improved. Palmore said that not much tion since it can be applied to butanol or other products. He is known about engineering proteins but she hopes that it can also explained exactly how corn is harvested and that the two be solved using a Brown University database. Mark Emptage main sugars that can be extracted from biomass are glucan said that DuPont has worked with Diversa (non Verenium and xylene. Corp.), which has put together a set of technologies to allow Emptage explained that Dupont is taking a new approach high-throughput screening and enzyme evolution to be done. using Zymomonas mobilis, a natural ethanol producer found He said that there is still a need for more basic understanding in the sap of agave plants in the tropics. It has a higher yield about how the enzymes operate on the complex structures. and productivity than yeast and has the potential to be a ­better Charles Dismukes of Princeton asked Mark Emptage organism than yeast. Emptage announced a collaboration how DuPont plans on solving the two major problems that between DuPont and POET, the largest dry-grind producer of he said need to be addressed: costs for removal of the etha- ethanol in the United States with over a billion-gallon ethanol nol distillation and acetic acid inhibition. To solve the first capacity. They are working together to develop a pilot plant problem, Emptage explained that consolidated bioprocessing in South Dakota using the new technology. He said, “This will be necessary. That technology has not yet been devel- isn’t a revolutionary program. This is really an evolutionary oped, so it is important to figure out what to do in the near program, just adding on to the current infrastructure.” term. DuPont has looked into thermopiles, but they have Emptage believes that the key remaining challenges are very low ethanol yields. DuPont is now seeking an organism solids handling, having an infrastructure to collect, transport, that maximizes yield, which is why they chose to work with and store biomass effectively and efficiently with its low- Zymomonas. Emptage explained that one way to solve the bulk density. Another challenge is the cost of enzymes. The acetic acid problem is to adjust fermentation conditions to goal is to make the handling of the Z. mobilis derived biomass the highest pH level tolerable. DuPont has developed more cost competitive with grain ethanol. acetate-tolerant strains. In DuPont’s process with ammonia, acetamide is a by-product with ammonolysis competing with hydrolysis of the acetyl groups, which lowers the total Discussion concentration of acetic acid in the process. Michael Wasielewski of Northwestern University asked Douglas Ray asked whether biobased approaches are Thomas Moore about the type of light fluxes being used going to scale. Emptage said that scaling in fermentation to investigate the solar flux. He also asked, “Since we all is straightforward, and that tanks can be scaled almost as know that photosynthesis has control mechanisms that actu- large as sugar and fermentative organisms. Daniel Nocera ally modify electron flow, based on light flux, what kind of of Massa­chusetts Institute of Technology stated that he is p ­ rospectus or perspective do we have for control mechanisms worried about the long-term scaling issue for energy, which in such systems?” Moore explained that one of the factors is why he supports solar. However, advances in solar energy that seems to limit natural photosynthesis is the diffusion of involve discovery research that is 50 years out. Nocera went carbon dioxide into the system for fixing, so it is important on to say that the energy problem is a basic science problem, in photosynthesis to throttle back the powerful oxidant when not an engineering problem, and people should stop focusing carbon dioxide is limiting. There is a control mechanism on the complex engineering to find a solution. called nonphotochemical quenching that is related to the