Storing Solar Energy by Taking a Cue from Green Plants
Photons to Fuels
- Fig. 1: Schematic depiction of a DSPEC for water splitting into hydrogen and oxygen. It illustrates the separate steps for water splitting: 1) Light is absorbed by the chromophore, generating a high energy electron; 2) The electron is injected into the semiconductor; 3) The electron is transported through the semiconductor to a transparent conducting oxide (TCO) collector electrode where it enters an external circuit; 4) The electron is transferred to the cathode for reduction of water or protons to hydrogen; 5) An electron is transferred from the catalyst to the chromophore beginning the process of catalyst activation; 6) Steps 1-4 are repeated four times driving water oxidation and oxygen evolution at the photoanode.
Is storing solar energy the solution? The increasing global demand for energy is one of the most pressing problems facing the global community. The sun provides 10,000 times our current energy needs but, it is diffuse. To become a primary energy source its input would have to be stored on massive scales. Inspired by natural photosynthesis and advances in semiconductor technology, we describe here an approach for the direct conversion of solar energy into fuels by Dye Sensitized Photoelectrosynthesis Cells (DSPEC).
The DSPEC concept has been developed at the University of North Carolina at Chapel Hill Energy Frontier Research Center (UNC EFRC) in Solar Fuels with the mantras "Keep it simple" and "Let the molecules do the work" as guides. Recent results show that in these devices, molecules can do the work of both absorbing light and using its input to carry out water splitting into hydrogen and oxygen when combined with core/shell oxide semiconductors.
What to Do When The Sun is Not Shining?
The growing prosperity of the world economy with the emergence of the third world is increasing the global demand for energy. Reliance on finite energy resources, with >85% of the world's energy provided by hydrocarbons, is unsustainable in the long term and is having increasingly obvious impacts on the global climate. Solutions are needed which provide renewable, carbon-free or carbon-integrated technologies with sufficient energy density to power the planet. The sun provides more than enough energy in an hour to meet the world's energy needs for a year and is the only renewable source that can provide energy at the needed scale. It does, however, suffer from a serious drawback when it sets at night or disappears on a cloudy day.
Developing a method to convert and store solar energy for use when the energy is needed is the central theme of the UNC EFRC. Inspired by natural photosynthesis, we adopted an approach called "Artificial Photosynthesis". Natural photosynthesis occurs in the mitochondria of green plants where sunlight is absorbed and used to power the complex molecular machinery for converting carbon dioxide and water into carbohydrates.
We use many of the same molecular phenomena but integrated with solid-state oxide materials in Dye Sensitized Photoelectrosynthesis Cells (DSPEC) [1,2]. To address this challenge we adopted a modular approach, forming teams of experts who focus on the development and characterization of the separate parts of a DSPEC, which are integrated into final device architectures. Initial results speak for themselves, with a recent report of the first functioning DSPEC for sustained water splitting and hydrogen production appearing in the Proceedings of the National Academy of Sciences .
Overcoming Limitations of Artificial Photosynthesis
At the heart of the DSPEC is a "chromophore-catalyst assembly". The chromophore absorbs light and the catalyst oxidizes water. For water splitting, the assembly is tethered to the surfaces of TiO2 nanoparticles sintered together in 5-10 µm thick films. TiO2 is an oxide semiconductor used as a pigment in white paints and in sunscreens for UV protection. Light is absorbed by the chromophore creating a high-energy electron in the molecule. The high-energy electron is rapidly injected into the semiconductor and then transferred through an external circuit to a separate cathode where it reduces protons to hydrogen gas. Loss of electrons from the assembly activates the catalyst toward water oxidation and, after four electrons are lost, oxygen is evolved. The overall process is illustrated in Figure 1.
Although the initial concept was first described in 1999, two limitations had to be overcome before a working DSPEC could be realized . Phosphonate-surface bonding was used to link the assemblies to the semiconductor oxide surface and was only stable in acidic solutions. A higher degree of surface stability was desirable, especially in basic solutions with added buffer bases where rate enhancements of up to a million had been observed for water oxidation. A second problem came from the thickness of the 5-10 µm thick TiO2 nanoparticle films. Following injection into TiO2, the injected electron found its way back to the oxidized assembly before it could diffuse through the oxide to the underlying transparent electrode for transfer to the cathode where hydrogen is produced.
Both limitations were overcome by applying a technique called atomic layer deposition (ALD) in a collaboration with the group of Professor Greg Parsons at North Carolina State University. ALD functions by exposing a reactive, oxide precursor gas, e.g., Al(CH3)3, TiCl4, to the surface of the TiO2 nanoparticle film. After reacting with the surface and being exposed to water in a second step, a thin layer of Al2O3 or TiO2 builds up on the surface. By repeating the process, thin layers of the oxide form and build up on the surface in controllable thicknesses.
Protective Layers Enhance Stability
ALD was applied to surface stabilization by first binding the molecular assembly to TiO2, and then adding an overlayer of Al2O3 or TiO2. The overlayer stabilized the surface toward hydrolysis even in strongly basic solutions. Surface stabilization allowed experiments to be performed at pH 12 with high concentrations of added PO43- with the rate of water oxidation enhanced by a factor of ~106 compared to pH 1 !
The second problem was also solved with ALD. Thin 2-4 nm layers of TiO2 were deposited on the outside of 3 µm thick nanoparticle films of the transparent conducting oxide ITO (tin-doped indium oxide). This formed a "core/shell" structure with a transparent conducting oxide as the core and the semiconductor oxide as a shell on the outside. The molecular assembly was then tethered to the surface of the semiconductor and the DSPEC in Figure 1 constructed by connecting the core/shell assembly photoanode to a Pt cathode for reduction of protons to hydrogen. The core/shell structure was the key to success. Following excitation, the injected electron rapidly reached the transparent conducting oxide core for transmission to the cathode. With application of a 0.2 V bias, sustained water splitting was observed with O2 appearing at the photoanode and H2 at the cathode. The absorbed photon current efficiency was only 4.4% but later improvements are leading to promising results and devices that, in the end, should be able to carry out solar water splitting with 10-15% solar yields .
These results provide a significant breakthrough for solar fuels and a promise in the future for scale up and, ultimately, practical devices. Immediate targets for future research are enhanced solar efficiencies, long term stability, and tandem cells which split the solar spectrum driving water oxidation to oxygen at a photoanode and CO2 reduction to formate or syngas, 2:1 H2:CO mixtures, at the photocathode. The essential pieces are in place, early results are promising, stay tuned...
This research is wholly supported as part of the UNC EFRC: Center for Solar Fuels, an Energy Frontier Research Center funded by the U.S. Department of Energy, Office of Science, Office of Basic Energy Sciences under Award Number DE-SC0001011.
 Alibabaei L. et al.: Journal of Materials Chemistry A 1, 4133 (2013)
 House R.L. et al.: PV Magazine, pp 87 (2013)
 Alibabaei L. et al.: PNAS 110, 20008 (2013)
 Treadway J.A. et al.: Inorganic Chemistry 38, 4386 (1999)
 Vannucci A.K. et al.: PNAS 110, 20918 (2013)
Ralph L. House
Thomas J. Meyer
Department of Chemistry
University of North Carolina at Chapel Hill, USA