Let the molecules do the work

Energy – its production, impact on the world’s economy, and environmental impact – will be among the major challenges of the 21st century. Although per capita consumption is stabilizing in the developed economies of the U.S., Europe, and Japan, accelerated use is occurring in China, India, and other parts of Asia driven by economic growth. Currently, hydrocarbons account for 85 to 90 percent of the world’s energy supply. Recent technological breakthroughs in hydraulic fracturing (fracking) and horizontal drilling are opening a wellspring of natural gas in the U.S. and, with it, a cheaper, cleaner hydrocarbon alternative to coal and oil.
Nonetheless, a stable energy supply in the long term and the environmental impact of burning fossil fuels remain important issues for both the USA and the world at large.

Solar fuels

The ultimate solution to the world’s need for renewable, environmentally friendly energy is all around us. The sun provides about 10,000 times our current daily energy needs, but it will always be limited until two significant challenges are met. In the U.S. alone, a 10% solar efficient device would require a collection area of approximately 1.6×1011 square meters (roughly the land area of North Carolina) to meet the country’s current power needs. Equally daunting is the fact that the sun goes down at night. To be practical, utilization of solar energy requires energy storage on massive scales, far greater than any available based on existing technology.
Natural photosynthesis provides an inspiration and biomass a partial solution, but not for meeting the vast power density requirements of urban centers or industrial complexes. Useful working devices will require much higher efficiencies, lower cost, and simplicity of design and maintenance.
The only practical approach at the required scale is “Artificial Photosynthesis” with “solar fuels” as the product.
Solar fuels are high-energy molecules like carbohydrates or hydrogen with the energy of the sun stored in chemical bonds. Target reactions are water splitting into hydrogen and oxygen and light-driven reduction of CO2 to CO or other reduced forms of carbon.
These reactions are made up of component “half reactions” that can be carried out in the separate compartments of an electrolysis cell.
For water splitting in acid solutions, the half reactions are: water oxidation 2H2O ? O2 + 4 H+ + 4e- (in acid), and proton reduction 2H+ + 2e- ? H2. An ultimate target is liquid hydrocarbons that could utilize our existing energy infrastructure.

Background

In the early 1970s, Fujishima and Honda provided an opening by showing that band gap excitation of anatase TiO2 with an applied bias in a photoelectrochemical cell (PEC) resulted in water splitting. Considerable research on PECs has followed. They are classified as wired or wireless.
In wired cells, wires are used to transmit current between physically separated electrodes where the half reactions take place.
In single light absorber devices, absorbed light drives oxidation at a “photoanode” or reduction at a “photocathode.” In tandem cells, the solar input is split between the two wired layers in a stacked configuration. Here, light absorption at the two photoelectrodes splits the solar spectrum to drive the separate half reactions. This necessarily halves the overall efficiency since two photons are used, but the range of useful solar collection is extended into lower energy regions of the spectrum and strongly oxidizing and reducing catalytic centers can be activated to speed up the rates of the solar fuel half reactions. In the wireless configuration, current is passed through an ionic conductor (electrolyte) instead of a wire. This approach has been described by Nocera and coworkers at MIT based on a triple-junction amorphous Si cell coated with electrodeposited catalysts such as NiMoZn for water reduction, and cobalt oxide for water oxidation. This configuration is similar to one used earlier by Turner and coworkers at NREL but with Pt catalysts in a wired cell.
In the Nocera example, the output of three Si cells were connected in series and current matched to provide voltages in excess of 2.0 V. Licht at Technion has also reported a wireless dual band gap tandem PV device and Lewis and coworkers, at the U.S. Department of Energy-sponsored JCAP Solar Hub, are developing a wireless approach in which at least one of the photoelectrodes incorporates semiconductor nanowires functionalized with catalysts. While attractive at the laboratory scale, the performance of wireless devices could prove challenging for scale up.

A molecular approach

A different approach to solar fuels has been taken by the U.S. DOE-funded Energy Frontier Research Center at the University of North Carolina at Chapel Hill. The UNC EFRC’s target is to generate solar fuels from sunlight by water splitting or carbon dioxide reduction based on dye-sensitized photoelectrosynthesis cells (DSPEC).
This research was inspired by results in the 1970s which showed that once excited, molecular excited states can initiate solar fuel half reactions by electron transfer, much as in natural photosynthesis.
Molecules do most of the work, absorbing light, transferring electrons, and catalyzing reactions. This is in contrast to direct band gap excitation of semiconductors, where the semiconductor is asked to do it all: absorb the light, separate photo-produced holes and electrons, and provide the catalysts, all without degradation or decomposition.
The DSPEC design benefits from a modular approach, allowing the separate parts to be synthesized, evaluated, and improved in an iterative manner. figure 1 (p. 88) illustrates a tandem cell configuration with a photoanode for light-driven water oxidation on the left and a photocathode for CO2 reduction on the right. At the heart of the photoanode is a chromophore-catalyst assembly.
The chromophore absorbs light and the resulting excited state injects an electron into the conduction band of a mescopic, nano-structured film of a high band gap semiconductor, typically TiO2.
The oxidative equivalent (hole) left behind is transferred to the catalyst by intra-assembly electron transfer. Oxygen is released following the build-up of the four oxidative equivalents required for water oxidation.
The right hand electrode is a photocathode based on NiO or another p-type semiconductor. Excitation of the chromophore-quencher assembly on the semiconductor surface is followed by electron transfer from the semiconductor to the chromophore excited state. Intra-molecular electron transfer from the reduced chromophore activates an external catalyst toward water or CO2 reduction.

Challenges in design

Letting the “molecules do the work” has appeal as a strategy but brings with it significant challenges. These mainly come from the mismatch between the single photon/single electron nature of the activation process and the multiple electron/multiple proton transfer solar fuel half reactions, such as water oxidation. The resulting challenges are illustrated in figure 2 (p. 88):

• Achieving appropriate band gap and conduction (Ecb) or valence band (Evb) potentials in the semiconductor; electron/hole collector dynamics in the semiconductor film.
• The chemical link to the surface, which must provide long-term stability and an electronic connection for electron transfer to occur.
• Light absorption with high absorptivity throughout the visible into the near infrared spectrum for high efficiencies.
• Maximizing electron injection and minimizing back electron transfer from the semiconductor to the oxidized assembly.
• Water oxidation and CO2/water reduction catalysis need to be more rapid than back electron transfer with catalysts that are indefinitely stable.
• Device prototype evaluation by application of transient spectroscopies and photocurrent measurements.
• Device scale up in practical applications.

The DSPEC approach has been criticized as relying on molecular components, which may have limited lifetimes and stabilities. The long-term operating lifetimes achieved by commercial dye-sensitized cells are notable in this regard.
Further, a great strength of the molecular approach and surface binding is ease of fabrication. Preparation of surfaces modified by molecular assemblies, like the one shown in figure 2, is accomplished simply by exposing the surface to dilute solutions of the assembly. Even with decomposition and loss from the surface, a next generation of surface-bound assemblies is available after a simple, on-demand, soaking cycle.
The UNC EFRC is tackling these challenges by adopting a broad multidisciplinary approach, drawing on expertise in chemistry, physics and materials sciences.
The feasibility of using DSPEC discoveries will require translating laboratory-scale observations into engineered devices guided by economic analyses to gauge market feasibility. The initial steps in this process are underway through a partnership with chemical engineers at the Research Triangle Solar Fuels Institute.

Economic feasibility

Compared to the development of photovoltaic modules over the last 30 years, progress in photoelectrochemical devices has been far slower. But viable alternatives are needed: systems that can be either renewed continuously (self healing) or during downtimes (night, cloudy days).
The DSPEC approach could be key given the ease of fabrication and simple procedures for regeneration without physically removing solid electrodes, which would be cost prohibitive.
In a major step in developing the DSPEC concept, the University of North Carolina EFRC/RTSFI team has been evaluating electrocatalysts for water splitting to decrease over potentials, investigating the stabilization of surface-bound catalysts, and exploring new electrode configurations with the eventual goal of engineering an economically viable DSPEC device.
The authors
Ralph House, Leila Alibabaei, Christopher Bonino, Paul Hoertz, James Trainham and Thomas Meyer – Department of Chemistry, University of North Carolina at Chapel Hill (www.efrc.unc.edu); Research Triangle Solar Fuels Institute (www.solarfuels.org)