Taking a look at the full-cycle EROEI is something I'd like to do, but haven't. It's not necessary for any given element of a power cycle to be EROEI-positive. In fact any given energy transform will represent a net energy loss. But for the total effective cycle you've got to get more than you give. Note that at present agriculture in the US represents a 10:1 energy cost -- you get 1 unit of energy for every ten units of fossil-fuel energy you input. In Europe it's about a 1:5 ratio. Again, negative EROEI. The saving grace is that fossil fuel energy has such a high EROEI.
On marginal cost of energy: if the alternative is to discard the generating potential entirely, then the marginal cost is zero. If you're building excess capacity specifically to provide fuel synthesis capabilities, you do have real costs. The US Naval Research Lab's estimate is $3-6/gallon for aviation fuel, though I'm not sure if they assume a gratis reactor. I've specced out $9/gallon with solar input. Not cheap, but a long-term stable price, vs. constantly rising fossil fuel costs.
As for input energy: my assumption is generally for solar + wind -- they're simply the largest available long-term sustainable energy sources we've got (I've got my doubts on how long nuclear fuel will last, and terrestrial fusion's still unproven). The initial scheme for large-scale F-T synthesis was based on a presumption of nuclear energy input. M. King Hubbert proposed this in 1964, and the idea was picked up by Meyer Steinberg of BNL pretty much immediately. The initial proposal was that CO2 come from, e.g., limestone, but seawater was identified as a reservoir pretty early on. Steinberg's a nuclear engineer, and most of his work assumes nuclear power supplying electricity.
Incidentally: in a nuclear economy, you'd still need liquid fuels, though most proposals focus on hydrogen alone. Given difficulties with its chemistry, I doubt this will prove effective.
Compared with alternatives, I really see solar as the backbone of any future energy system. The only questions are how large the supported population will be, and how advanced its technology. Solar energy is what humans relied on before finding fossil fuels. And there's no assurance that we'll retain our present tech levels.
Nuclear's problems are not merely political, though that's a significant hurdle of its own.
On marginal cost of energy: if the alternative is to discard the generating potential entirely, then the marginal cost is zero. If you're building excess capacity specifically to provide fuel synthesis capabilities, you do have real costs. The US Naval Research Lab's estimate is $3-6/gallon for aviation fuel, though I'm not sure if they assume a gratis reactor. I've specced out $9/gallon with solar input. Not cheap, but a long-term stable price, vs. constantly rising fossil fuel costs.
As for input energy: my assumption is generally for solar + wind -- they're simply the largest available long-term sustainable energy sources we've got (I've got my doubts on how long nuclear fuel will last, and terrestrial fusion's still unproven). The initial scheme for large-scale F-T synthesis was based on a presumption of nuclear energy input. M. King Hubbert proposed this in 1964, and the idea was picked up by Meyer Steinberg of BNL pretty much immediately. The initial proposal was that CO2 come from, e.g., limestone, but seawater was identified as a reservoir pretty early on. Steinberg's a nuclear engineer, and most of his work assumes nuclear power supplying electricity.
Incidentally: in a nuclear economy, you'd still need liquid fuels, though most proposals focus on hydrogen alone. Given difficulties with its chemistry, I doubt this will prove effective.
Compared with alternatives, I really see solar as the backbone of any future energy system. The only questions are how large the supported population will be, and how advanced its technology. Solar energy is what humans relied on before finding fossil fuels. And there's no assurance that we'll retain our present tech levels.
Nuclear's problems are not merely political, though that's a significant hurdle of its own.