Apr 9, 2014

dredmorbius comment on "Navy lab makes gasoline from seawater, as low as $3 per gallon"

dredmorbius: Since this story is making rounds but the reportage is so godawful terrible, I've dug into the source Naval Research Lab papers and press reports for a bit more context on this.
  • The US Navy's study is looking at producing 100,000 gallons/day of aviation fuel. Technically, that's a high-grade kerosene, not gasoline. For comparison, an F-18 Super Hornet has a fully-loaded capacity of about 430 gallons, so this would supply around 230 flights/day. It's about 1/8400 the total petroleum consumption of the US (20 million barrels/day, more later).

  • The process consumes energy which must be input from elsewhere. The benefit is obtaining liquid hydrocarbon fuel, which is useful, versatile, and very energy dense. Power requirements for 100k gal/day is around 240 MW, which happens to be roughly the output of a Nimitz-class carrier's reactors.

  • There are three parts to the process:

    1. Electrolysis of hydrogen, which consumes most of the energy, and is, on an energy-delivered basis, about 60% efficient. That is, you're losing 40% of your input energy, but you get storable energy in the form of hydrogen, which is converted to hydrocarbons later.
    2. Carbon dioxide is present in seawater at about 140x greater volumetric densities than in the atmosphere, in the form of dissolved gas (2-3%) and as bicarbonate and carbonate (97-98%). This can be extracted via partial vaccuum and pH changes which, conveniently, accompany hydrogen electrolysis.
    3. Fischer-Tropsch process long-chain hydrocarbon synthesis. This is a well-understood, long-established chemical process developed in the 1920s. Most liquid petrochemical fuels are long-chain hydrocarbons, the Navy's goal is a chain length of eleven.
  • The process will require a lot of seawater: 8.8 billion liters/day. That's 3520 Olympic-sized swimming pools worth. I strongly suspect that pumping and water-handling costs and energy usage will be significant.

  • Reactor size based on existing electrolysis vessels would be roughly 24,500 m3 , or a cube 157 m on a side. This exceeds the practical dimensions of a carrier -- we're not talking about a system which could fit within an existing ship's hull dimensions. Instead the Navy is considering a fixed-location plant, and possibly a floating platform which could accompany a carrier task group. More below.

  • Supplied energy for a fixed-site plant is seen coming from an ocean thermal energy conversion (OTEC) facility as a renewable source. Alternatively, an aircraft carrier's reactor output could supply a floating platform, though this would involve tethering a carrier to another vessel to supply power. Reports suggesting that existing tender and supply logistics would be eliminated are, as far as I can tell, entirely bogus. You've still got fuel which requires transfer. The logistics will change but they'll still exist.

  • On a rough back-of-the-envelope basis, scaling this to national levels is at least within the realm of possibility, though it would be tremendously expensive. The plant size could be as small as a few square kilometers (compared with tens of thousands to a few million km2 for biomass algae or crop fuel scenarios), electric power would be a couple of terawatts (a 180 km square of solar output), and cost would be around $8 trillion (about half of annual GDP of the US), for the fuel synthesis alone. Solar generation costs would likely double that.

  • The technology should be carbon-neutral. It's extracting carbon dioxide from seawater, and the fuel is then burned, returning it to the atmosphere. This doesn't reduce net biosphere carbon, but it doesn't increase it either. This is contrasted with both fossil fuel use, in which ancient carbon stored underground is released to the atmosphere (which increases biospheric CO2) and carbon sequestration, in which carbon is removed (by technical or biological means) and buried (decreasing biospheric CO2).

State of the project is that some very small-scale (relative to the proposed size) tests have been conducted, with reasonably positive results. I tend not to get very excited about much in the way of alternative energy / fuel suggestions, but this at least seems like it's based on solid technical foundations. I suspect the eventual costs will be higher than what's currently projected ($3-$6/gallon fuel). The military has a huge interest in energy and fuel, and navies in particular have driven a number of fuel revolutions: adoption of coal in the mid-19th century, oil during WWI, and nuclear power following WWII.