An overhaul of the battery will jump-start a shift to renewable energy

A partial list of people working on it:

* Amyris

* Solazyme

* KiOR

* Algenol

* Sapphire

* Aurora

* Synthetic Genomics

* Cellana

* PetroSun

Notably, most of the companies on that list have taken hundreds of millions of dollars from domestic and foreign oil companies hoping to create their own supply. Maybe more notably, several of the companies on the list went public, then lost 80-90% of their market cap. Most are pursuing other, higher value products as a first-order business, with promises of biofuels down the road.

The world uses 100 million barrels of petroleum daily, the US about 20 million (slightly less). Even small uses of petroleum such as aviation (about 2% of the total) represent vast amounts of fuel. And with projected population and energy utilization growth, demand will increase above this.

The basic maths are laid out in a paper I've found to be a tremendously useful reference, Jeffrey Dukes, "Burning Buried Sunshine": http://globalecology.stanford.edu/DGE/Dukes/Dukes_ClimChange...

Briefly: present human consumption of net primary productivity (plant growth) is about 14% of Earth's total. The fossil fuels consumed in 1997 (the date analyzed by the paper) comprise the equivalent of another 21% of NPP (and the actual period over which the fossil fuels accumulated is far greater: 400 years of growth, accumulated over ~5 million years, due to inefficiencies in the process of coal, oil, and gas formation and accumulation).

You can look at biofuels on an energy per unit area basis. There's roughly 1 KW/m^2 of incident sunlight. Current PV solar cells convert around 20% of this, and tend to capture the equivalent of 8 hours of sunlight per day, which is to say, a 30% capacity factor. Maximum PV efficiency (with far higher costs of production) is around 85%. But with present commercially available means, you can capture 20% of 1 kW for 8 hours each day, or roughly 1.6 kWh per day. A kilowatt (kW) is a unit of power, a kilowatt-hour is a unit of energy.

US electrical production of roughly 4,000 TWh would require 7800 km^2 of area to produce, or a square 88 km on a side.

That's for electricity.

Plants and algae have an efficiency of from 1-10%, with conventional crops and plants generally in the 1-3% range. We can do the same area calculation as above, or work from yields expressed as gallons of fuel per acre. A high-yield crop such as canola is generally given as 100 gal/acre, hemp advocates claim as much as 300 gal/acre (though this is generally disputed), and algae biofuels as much as 1000 gal/acre.

Given the 7.3 billion barrels of oil consumed, lets work out acreage requirements for algae. GNU Units is one of my favorite tools for this:

    You have: 7.3 billion barrels / (1000 gallons/acre)
    You want: million km^2
	* 1.2407662

There are presently slightly more than 400 million acres of land under agricultural production in the US. Three quarters of that would have to go to fuel production under one of the most productive biofuel processes we have.

Moreover, algae require ponds, which is to say, a LOT of water, pumping pesticide, fertilizer, and a lot of processing.

Say you figured out a way to move all of this off-shore, and could set up grow-ponds in the oceans off the Atlantic and Pacific coasts? They're roughly 1000 miles long each, so to figure on the width of our algae grow structure, we divide 479,000 by 2000 and find ... the ponds would have to extend only 239 miles off-shore. That's from Key West, FL, to Eastport, ME, from San Diego, CA, to Port Angeles, WA. And they'd have to survive freezing and nor'easters and hurricanes and ...

Yeah, sounds kinda tough.

If the US population and per-capita energy utilization were far less, say maybe 10% of present levels, this might be conceivable, but as things stand, not really. And then there's the rest of the world.

Even more modest proposals to, say, supply aviation fuel from halophyte + aquaponics farms, described by Boeing as the biggest breakthrough in biofuels, really doesn't add up: http://redd.it/1wo2hl

(Another interesting fact I discovered in researching that was that the US hit Peak Aviation Fuel in 1999, and present flight activity is more than 30% below what projections of that time suggested. Passenger miles are up somewhat given increased load factors -- more bumped flights and less legroom due to cramming more people onto fewer flights. Kopits points this out also in his presentation, confirming my observations.)

Which is why I find the NRL's project so interesting. The magnitude of its plant and scale requirements is far, far less than competing biofuel alternatives.

Electrifying rail is certainly possible and it's the mode under which much European freight rail operates. The challenge for other regions: the US, Canada, Russia, China, and Australia in particular, is that distances are far greater, meaning far more km to energize. There's also existing traction (railroad engines) which would have to be replaced. The costs are not insignificant, though I'd really like to see long-term plans.

Electrifying drayage (that is, non-fixed-track overland cargo) is a significantly different prospect. The advantages of trucking over rail are both spatial and temporal flexibility: trucks can go anywhere there are paved roads (and some places there aren't), and don't generally need to wait for tracks to clear or switching points to move. This is at a cost of much, much higher costs expressed as dollars per ton-mile.

If you're going to fully electrify trucking, you've either got to fully electrify every mile on which trucks travel, and there's vastly more road mileage than track, or you've got to provide for some level of local energy storage on the truck. That storage has to be able to re-charge quickly, provide sufficient energy to meet transport demands, not impose excessive weight and space demands, and offer high price/performance characteristics.

The usual suspects all have some pretty significant issues: batteries, flywheels, capacitors, etc. Not that it's impossible, but being able to dump a few hundred gallons of long-chain hydrocarbons in a tank and use them as needed has major benefits.

Then there are other off-the-grid uses, though most of these tend to be a relatively small fraction of current petroleum use: small equipment (from chainsaws and leaf blowers to mowers and tractors), construction equipment, ag use, remote power (generators, etc.), shipping (electrifying the oceans would be a significant engineering challenge), aircraft (ditto skies), etc. Even where many of the specific instances might be substituted otherwise, there are current edge cases (and in the case of general aviation, pretty much the whole case) which are really predicated on liquid hydrocarbons. Not to mention industrial processes reliant on petroleum/fossil inputs (chemicals, plastics, steel)

In noting that it's plausible for all existing petroleum utilization to be substituted by synthesis, I'm not saying that it's either necessary or desirable. I do see much present petroleum use being substituted, likely by direct electric power + storage (batteries do work for many instances).

But if you can swap out all of it, then the prospect of substituting smaller portions of utilization, say at the order of 5-25%, is also feasible. And that gets you a long way.

And no, that would obviate the benefits of their new technology.

Existing grids have benefited from the fact that though any given individual demand rate might vary widely, averaged over a large number of consumers, trends are far more predictable.

On the generating side this hasn't typically been as much of an issue, with highly reliable and dispatchable generating plants. With intermittent renewable/sustainable sources, the fact that local variations in solar or wind capacity can be mitigate by transmitting surpluses, or importing necessary capacity, also helps. In an advanced grid with attached storage, again, large-number averaging should, with effective management, help balance supply and demand.

Trying to do this at the scale of individual households means far more variability which must be dealt with on he immediate premises. This is among the reasons I strongly doubt that visions of a "gridless future" will emerge, at least so long as there's not widespread collapse of all social and technical infrastructure.