As the article on electrofuels points out, solar fuels are an important goal (C&EN, Nov. 28, 2011, page 36).
Unfortunately, the article includes some simplistic ideas from the community that is chasing them. Those proposing biofuels need to pay more attention to old engineering concepts: scale and throughput.
The article quotes Eric J. Toone, deputy director of ARPA-E, on high-risk investment in fuels based on microbes and bacteria, ideas that have the same problems as fuel produced by algae. In his lecture at the National Academy of Engineering annual meeting in October, Arun Majumdar, the director of ARPA-E, highlighted his agency’s investment in algae for liquid fuel. But he did not mention the many square miles of glass-covered or huge plastic-bag structures that would be required for even modest production volume or the issues in collecting and processing the product.
C&EN also extensively describes the proposals of Sun Catalytix, the company founded by Massachusetts Institute of Technology chemistry professor Daniel G. Nocera that has received support from ARPA-E for its artificial leaf cell that Nocera claims could power a house. He includes individual houses in both the developed and developing worlds, noting that people in the latter are off the electricity grid, separating them according to energy usage.
Let’s take a detailed look at what Nocera said, ignoring the unattractive thought of having a hydrogen compressor, high-pressure tanks, and an expensive fuel cell in every house.
The artificial leaf cell for the average U.S. home is said to be the size of a door, about 2 m2. We’ll make an optimistic assumption for average insolation of 0.7 sun for 12 hours per day, much more than what you would get in Phoenix, a very sunny city (apricus.com/html/solar_collector_insolation.htm).
We also assume a horizontal cell, ignoring the obvious problem of gas separation, which is much easier in a vertical but less absorbent cell. Then factor in an optimistic 10% artificial leaf and current 70% (max) fuel-cell efficiencies (hydrogen.energy.gov). We ignore the commercially available electronics involved for all these systems.
Such a system would provide a small amount of energy:
700 W/m2 × 2 m2 × 12 hours × 10% × 70% = 1.2 kWh per day
That’s only enough to run just a large U.S. refrigerator (1 kWh/day; frigidaire.com) and a few small loads. The average American home uses >20 kWh/day (eia.gov/tools/faqs/faq.cfm?id=97&t=3). There is a reason why residential solar-cell installations (15% efficiency) cover most of the roof.
If you are determined to make hydrogen from sunlight—not a bad idea—pay attention to the KISS principle and consider conventional polysilicon solar cells (15% efficiency) that feed power into a centrally located industrial-scale electrolyzer (70% efficiency; NREL/FS-560-36705, September 2004). This uses off-the-shelf low-maintenance equipment with a lifetime of at least a couple of decades. These solar cells will provide hydrogen generation that is greater than 10% efficient.
And if you live in much of the developed world, you won’t have to worry about the water in the cells on your roof freezing in the winter and breaking them. People off the grid already use a simple, cheap, and efficient storage device for solar-cell power. A typical deep-cycle battery has a usable capacity of 0.5–1.0 kWh (solarelectric.com/cosuagmba.html).
By Edwin A. Chandross
Murray Hill, N.J.