Dr. Albert,

Thank you for your question. I believe that bio-solar-photovoltaic devices (BSPVs) are feasible within the next decade. At the present, the power density of the best BSPV model system is about an order of magnitude lower (0.875 mW/cm^2, Leblanc et. al, 2013) than that of the average PV panels in the market (approximately 10 mW/cm^2). So clearly, significant enhancements to systems based on the photosynthetic proteins or whole cells need to be made in order for BSPVs to be comparable to organic or inorganic PVs, especially since these technologies keep getting more efficient and cheaper. However, recent results in the area (Kane Jennings, personal communication) indicate that BSPVs are still in the pre-development phase and have been realizing order of magnitude increases in power density over the past four years. So it is conceivable that a prototype field scale device will be in place in the next 10 or even five years.

One of the major impediments of using photosynthetic proteins, such as photosystem I, is their inability to absorb solar energy across a broad spectrum, contributing to their low solar absorbing and conversion capacity and low efficiency, though photosystem I itself has quantum efficiency near unity. Photosynthetic pigments are only able to absorb visible light, while solar cells are capable of absorbing energy from infrared to ultraviolet. Applying genetic engineering, we could introduce pigments from different organisms to enhance photosystem I or another similar reaction center to capture ultraviolet or infrared light. This could potentially make bio-solar-photovoltaic cells with efficiencies on par with current solar cells more feasible. Also using a compatible surface, such as silicon, which is able to absorb solar energy and its band alignment is appropriate for enhancing electron transfer with photosystem I.

In addition, biologically derived materials are more labile than inorganic or synthetic organic materials. On the other hand, they are constructed of earth abundant materials and can be renewed at a higher frequency at a much lower cost. Lastly, most proteins must be hydrated to function. A panel that must retain a high degree of water saturation will be more difficult to maintain than one that can operate independent of water concentration.

Dr. Shin,

Thank you for your question. The most prominent finding of my study so far is seeing that under the condition where we use sortase-mediated ligation of the sortase-recognition sequence tag containing protein, in the presence of the sortase enzyme, and the peptide decorated surface, we get the highest current density (10-15x greater) for our system compared to all of our control conditions. This shows that using our optimal condition yields oriented proteins on the surface, enhanced by the attachment scheme we used. Other systems with similar current densities do not allow preferential orientation. This single improvement, the unidirectional orientation of all protein molecules, will increase the power density by about 30% since there will not be effective cancelation of the currents from bidirectional orientation.

Currently, systems with the highest current density result from multilayers of the photosystem I protein. The next thing we would try is stacking several layers of oriented protein on the surface using different sortase enzymes and linkers to enhance these multi-layer systems. Current systems are assembled by several depositions of protein in a non-specific manner, without uniform directionality, but cleverly overcome this impediment by using band alignment between the surface and the protein. However, I think coupling a stacking scheme for uniform protein layers for a multilayer system will greatly enhance what has already been achieved with multilayers.

Dr. Buneo,

Thank you for your question. You bring up an excellent point. Simulations by Ciesielski et al (2011) showed that photocurrents produced by a PSI monolayer cancels unless 80% of the proteins are oriented in a uniform orientation. Their work suggests that “orders of magnitude” more photocurrent will be produced in comparison with monolayers of PSI that are oriented randomly. Small net photocurrents are produced when only 20-80% of PSIs are oriented similarly. Our results show 10-15x greater current density in the system when we use our sortase-mediated ligation system compared to non-specifically bound protein.

Based on the locations of exposed lysines on the protein complex, where the protein could covalently bind to a gold surface, 70% of the time the photosystem I will, on average, orient itself stromal side down, that is with its iron-sulfur clusters near the surface for electrons donating to the surface, while 30% of the time the protein will orient itself with the P700 complex oriented near the surface such that the surface would provide electrons which would be transferred through the complex to an electron acceptor in solution. So if we assume that in a non-specifically oriented surface, 50% are oriented in the desired orientation, either stromal side up or stromal side down, while in our sortase-attached system the proteins are orientated at least 80% in the desired orientation that we can expect to see at least a 30% improvement in current density.

Dr. McDonald,

Thank you for your comments and questions. We do not need to recapitulate the entire system for producing electron flow since our system would be a closed circuit. We extract the photosystem I from the thylakoid membrane, working with the protein in vitro, bypassing the intermediary steps, thereby eliminating the need to include the need for the other players of photosynthesis (PSII, cytochrome, etc). We do however require surrogates for the electron donor/mediator and electron acceptor – in our case our electron donor/mediator is ferro- or ferricyanide in place of the electrons donated by water splitting from PSII, and the terminal electron acceptor is the gold surface rather than ferrodoxin, which would eventually donate to FNR. Alternatively, to mimic the natural system of photosystem I, one could potentially use recombinantly expressed cytochrome and ascorbate as the electron mediator and sacrificial electron donor, respectively, which we have used in the past and has worked well with our hydrogen production system with PSI. We have not tried the cytochrome and ascorbate redox system yet, but will likely use it in the future to optimize mediator solutions for improved current density. We currently use the ferrocyanide system since it has been well-studied and much easier to track electrochemically. The redox potential of ferrocyanide also matches well with cytochrome and is therefore compatible with the PSI system.

Admittedly, our current density is not practically relevant, at least not for large scale applications such as powering a home. At the moment, our current density, as well as other bio-based cells like ours is four orders of magnitude below that of the best multilayer PSI PV cells, which may be comparing apples to oranges, but even that system produces 10 times less current per unit area than the average PV solar cell. We have a lot of catching up to do for this technology to be useful. I am hopeful in the coming years there will be advancements in these technologies that will be competitive with current solar cells.

Dr. Pfromm,

Thank you for your question. We have not done a temperature study with the proteins used in this system, but that is certainly something we will do in the future. The organism we work with is a mesophile, so the photosystem I protein that we extract from it is most active at 25°C – 35°C, but is stable up to 50°C. However, there is a thermophilic analog that is typically grown at 55°C, but can be grown at higher temperatures, and has been found to grow in conditions near 100°C. We have seen that the photosystem I protein from the thermophilic cyanobacterium Thermosynechococcus elongatus is stable at much higher temperatures, up to 80°C. Photosystem I from the thermophile could easily replace the one we currently use and be more thermally stable. We are using 6803 as a model organism primarily because much of the cloning work in the literature is done with this organism.

The average PV panels produce 10 W/ft^2 or 10.695 mW/cm^2. The best case multilayer PSI system produces 875 micro-amps/cm^2 (LeBlanc et. al, 2012). Assuming that the sortase-mediated ligation attachment scheme enhances current density by 30%, yielded a 1137.5 micro-amps/cm^2 multiplied by the 1 V potential of the PSI complex = 1.1375 mW/cm^2. So currently, the power output of the best bio-photo-based solar cell is an order of magnitude below that of the average photovoltaic cell. There is room for improvement, but I am hopeful that technologies like the one presented will be feasible and in the market in the near future.

Thank you, Mr. Blackwell!

Thank you, Ms. Evans!

Thank you for viewing, Luis!

Thank you, Dr. O’Neill!

Hi Dr. Westwood. Thank you for your question.

We have not done longevity experiments with this system, but from previous experience, the photosystem I protein we use is a robust protein complex and unlike most proteins will stay active after sitting in a refrigerator for over 6+ months, if not longer. This protein is also active after drying and reconstituting and after sitting at room temperature for long times (at least several weeks, though we always use the freshest protein we can for our studies). I believe these proteins will stay active for several months before they need to be washed off and replaced with fresh protein, which is a straight forward and simple process.

We currently do not have a complete and functioning cell built yet. For electrochemical studies, our protein is attached to our conductive surface and a known area of this surface is exposed to our redox mediator. We have not done longevity studies on these surfaces, which is something we will look into in the future, but these surfaces are still active after several weeks. Unfortunately I do not have data for current production longer than that, since we clean off our surfaces and re-use them to test several different conditions.

Hello, Christine! Thank you for checking out my presentation.

The organism we work with is a mesophile, because it is model system and making modifications to the protein is simpler compared to other similar organisms. The photosystem I protein we use is most active between 25°C – 35°C, but is stable up to 50°C. We have not done a temperature study with these proteins with this particular system, but it would definitely be interesting to find out whether or not this system would function well under real and extreme conditions outside of our controlled laboratory conditions.

Our lab has also worked with a thermophilic cyanobacterium with a photosystem I protein up to 80°C, but not for this particular application. This thermophilic PSI could easily replace the one we currently use, potentially making our system more thermally stable, at least in warmer climates. I think one would have to use organic or inorganic PV cells in extreme climates, as labile protein systems would not be appropriate for those conditions.

Dr. Chris, thank you for viewing my presentation! I’m glad you allowed me to show you what I’ve been working on. I think one of the most important aspects of this competition is to share research projects like these with the public so they are aware of potential future technologies and to encourage people to dream up creative ideas that can one day materialize and be implemented.