[ScottBurson]

In short, it sounds like you could have a car with solar panels on the roof, and they could drive the creation of liquid fuel that could then be used to power the car’s engine.

Sure. Leave your car in the sun for a week and you can drive it three miles.

People need to get how diffuse solar energy is.

(Oh, and could someone fix the title? "Bacteria" is plural.)

[ScottBurson]

Okay downvoters, let's run the numbers.

Suppose you have about 1 square meter of solar cells on the roof of your car. From Wikipedia, a typical solar PV installation in the US or Europe gets 1kWh/sqm/day (depending on latitude, of course). How far will 1kWh take you? Let's see, gasoline contains about 37 kWh/gal (US), which at 40mpg is a little over 1 mile/kWh.

So if your electricity-to-fuel conversion process is 50% efficient, you'll get about half a mile on a day's charge.

[WiseWeasel]

Exactly right. The premise that this technology would be integrated into cars is ridiculous, and my guess is it was added by the "journalist". This technology would most likely be used in places where feeds of concentrated CO2 are already available, such as coal power plants, water treatment, waste management, compositing, etc., and the fuel would be sent off to be distributed through fuel stations just as it is now.

The only impact this might possibly have on automobile design would be to try to recuperate the carbon from the exhaust using some of the power generated from the combustion to convert the exhaust CO2 into formic acid, to be exchanged for fuel in the next refueling. My guess is it's not going to pay to do that though.

[gus_massa]

It is difficult to get hard numbers, but probably the limiting resource is the energy in form of electricity, not the availability of carbon dioxide. So, if this is a good idea, the producing plant should be near a cheap electricity plant.

[mchannon]

Good point, but it does end up depending on the vehicle. A 6 m^2 15% efficient array parked outside in a desert climate could generate 5.4kWh a day and over 37kWh a week. GM's EV-1 had efficiencies of over 6mi/kWh. As such, a vehicle could be built that could go 32 miles a day off sunlight alone, using ho-hum $1/W solar cells and without any tracking.

Of course, at that point, you don't need to futz with bacteria and internal combustion, when batteries store and release that energy much more efficiently.

[ScottBurson]

32 miles a day... in Arizona... in a car 80% of whose upper surface area is solar cells, leaving little room to see out. I can imagine some enthusiast building one, but I don't think it's a mass-market product.

None of this, I hasten to add, is to impugn the idea of using bacteria to fix CO2. It's just that the idea of doing it in your car, driven by solar cells on the roof, is silly. The numbers just don't pencil out.

(And as you point out, it's doubly silly since you would use batteries anyway.)