There is nothing theoretical about the energy released from fission. Fission of one atom of U-235 releases 202.5 MeV of energy in total. 89% of that is prompt, i.e. at the instant of fission, which is all transformed into heat in a reactor. The remainder is from beta decay and gamma emission. Beta decay produce antineutrinos which are very weakly interacting and thus escape into space, which mean that about 4.3% of the total energy they represent can not be captured. Realistically we won't capture all the decay heat since about 10% is on the order of 200k years or more. But 83% of all fission products decay within 10 years, so in that time frame we will be able to utilize ~94.5% of all fission energy, which is still 2.65 million times more than burning the same mass of coal.

Some simple number crunching for a kg of U-235: U-235 has an atomic mass of 235.043929918, and a kg of it would thus equal about 4.25452382603 mole. Multiplying that with Avogadro constant we get that a kg of U-235 consists of of ~2.562134*10^24 atoms. Each atom when fissioned releases 202.5 MeV giving a total of 5.18832135*10^26 MeV, or about 23.09 GWh.

Coal on the other hand is about 8.136 kWh per kg, once again showing that for the the same mass, fission of U-235 gives in the ballpark of 2.8 million times more energy than coal.

From a nucleus standpoint the reaction might be energy positive, but the energy required to accelerate the particles to induce fission mean that you could end up with a negative EROEI. I've had a very difficult time though finding actual numbers on how much energy is released from the fission of elements lighter than thorium, so it's hard to estimate how useful it would be for energy generation.

You using a 1000 fold increase in energy usage as an argument is laughable. There are thermodynamic limits to how much energy we can use on earth before the waste heat starts affecting the climate. 1000 fold increase is way past that, and would increase global temperature to that of our body temperature. We would have to leave earth and colonize the solar system way before that, which would result in the resources of the solar system opening up for us, including all the fissile material out there.

A five fold increase in energy use would be more realistic, which come from the assumption that the world would increase to and hold steady at a population of 10 billion, while per capita consumption rate rises to US levels of 10 kW, giving a global energy consumption rate of 100 TW. Even more realistically though would be that it only increases to 60-70 TW given that a lot of the US consumption is pure waste. UK for example manage with just over 5 kW of per capita consumption.

As long as population does not increase, there is very little that is going to dramatically increase the amount of energy humanity uses.

There will never be a reactor that can burn anything below thorium, as the closest elements that could be fissile are unstable and decay too fast, while elements further down are too stable to fission.

And the energy density of nuclear fuel is not theoretical. Combustion of one metric ton of coal gives 8.136 MWh of thermal energy, while fission of one metric ton of fissile fuel yields 22800000 MWh of thermal energy, or roughly 2.8 million times more energy.

The numbers are back of the envelope calculations done by Alvin Weinberg, co-inventor of the Pressurized Water Reactor and Director of Oak Ridge National Labs. It's from a paper called "Energy as an ultimate raw material" which was published in 1959. The 30B number is actually based on just the amount of extractable thorium available in the earths crust, when projecting for a population of 7 billion at western levels of energy consumption. You can find the paper here: http://www.the-weinberg-foundation.org/downloads/Weinberg_EnergyRawMaterial.pdf

Something he brings up in the paper is the mining needed. The numbers he's using assumes 3 grams of uranium and thorium per metric ton of rock. The mining required to provide enough fuel for 40 TW heat (we use ~18 TW today) would then be about 10 million tons of rock a day, which is comparable to the 6 million tons mined a day in for coal and lignite in 1953.

The number for the concentration of nuclear fuel in rocks is quite low though. The crustal average is ~13 grams of uranium and thorium. But it still mean that to supply the entire world today, we would need to mine less rocks than what coal mining 60 years ago did.

This also does not take into account that there exists highly concentrated sources available (monazites with thorium content in whole percents) that will be used first, and vast quantities of less dense sources like granites that still have 25-100 g/ton of thorium.

All your fission numbers are for once-though fuel cycles. Using closed fuels cycles and breeders to consume 100% of the fuel, there is enough uranium and thorium, on earth, to power civilization for over 30 billion years.

Not by a long shot. Fission of U-235 releases 202.5 MeV in total, 89% of which is directly at the time of fission. Only 11% is from further decay, and over half of that is antineutrinos that almost never interact with matter.

Those enrichment plants use gaseous diffusion, which is a very inefficient way of enriching uranium. Centrifuge enrichment is much more efficient and use less energy. The only reason gaseous diffusion plants still operate is because the capital cost is paid off.

If SILEX enrichment, which is vastly more efficient yet again, is approved and deployed, maybe the gaseous diffusion plants can finally be shut.

Uranium won out in the beginning because it was the easiest to begin with, and because the LWR was the chosen design by Admiral Rickover for the Nautilus. To get civilian power started, it was much easier to begin with the design shown to work in the Nautilus. Going for a better civilian design as a first reactor would have delayed deployment of nuclear energy generation by many years. Almost everyone saw LWRs as a stepping stone to better reactors and did not envision them still being used today, and as our primary reactor type even.

Many errors here. Thorium is not borderline fissile, it will never fission in a thermal reactor. Though it can fission in a fast reactor, but then only 2/3 of the time, just like Pu-239 in a thermal reactor. But thorium is fertile, because it can be transmuted into fissile U-233. You don't need enriched uranium or plutonium to start it every time it stops. What you need is enough fissile in the reactor to initiate a chain reactor. This is not a problem as thorium is a net breeder even in a thermal reactor. It can generate more fuel than is consumed, so you will never need to bring in fissile fuel from external sources to sustain it.

That have absolutely nothing to do with thorium. It is a function of the design of the reactor. If the reactor is not designed to load follow, it will most likely struggle a bit when trying to do it. Generally, the reactor type used in France is not very good at responding to demand, but the design was modified slightly to allow this, making it possible for France to have such a high percentage of nuclear. The reactors normally talked about when discussion Thorium nowadays is Molten Salt Reactors, which the LFTR is, which are excellent at responding to demand.

The technology required did exist but, the entire process was not demonstrated on a large scale, and the MSR program at Oak Ridge was shut down before it could be done. But there is no problem with processing the fuel stream. The processes that are needed are high temperature to begin with, and much of the chemical processing that needs to be done is well known, and used in many industries. And no, reprocessing is not needed for proliferation resistance. It is inherent from the production of U-232 from neutron interaction with either the intermediary isotope Pa233, or on the U-233 itself.