It would take about 6000 3 GW(th) plants to supply the world's current average rate of primary energy consumption (18 TW). And if these were today's thermal reactors we'd run out of the current estimate of the global resource of U (at < $260/kg) in less than a decade.
Increase in price would increase amount of mineable U. Nobody even prospected it in last 50 years. There is enough of of known deposits, why trying to find new ones?
That is like peak oil stories. Increase price twice or thrice and suddenly there is enough place where it is worth mining. Nobody even talks about peak oil anymore. It's not like U is significant part of a cost/kWh. We could get U even from seawater.
Sea mining as an alternative to renewables + storage is an absolutely riotous joke. Take a look at the results of practical attempts and see if you can find the punchlines (there are at least three).
Powering a single 1000 MW(e) LWR reactor with sea water uranium would require an adsorber field covering 170 square kilometers of continental shelf. The average power/area would be considerably worse than solar.
Also (from second reference): "The predicted cost to extract uranium from seawater ranged from $610/kg U to $830/kg U." Ouch.
You missed the really funny bit where to reduce costs to a "mere" ~$400/kg you attach the sea mining equipment to supply 5MW to an offshore wind turbine which in 2022 will produce around 10MW net. Then it produces enough vanadium to add about 1hr or storage in flow batteries for the wind turbine each year.
It also requires a constant supply of polymer that caps the EROI around 10.
Good thing we only extract 10% of the energy from the Uranium today. Also, you wouldn't go 100% nuclear with today's PWRs. Gen 4 reactors will be much more responsive and efficient with the fuel.
Yes, that's why I'm saying breeders would be required. To scale from existing operating breeder reactors to powering the world would require increasing the number of such reactors by a factor of 5000. Scaling up renewables and storage is a much smaller multiplier, in comparison.
> To scale from existing operating breeder reactors to powering the world would require increasing the number of such reactors by a factor of 5000
That's a very silly way to frame it because there's only like 3 commercial breeder reactors today. Why is that the case? Because energy has been cheap, we know how to do PWRs so there's little reason to do the work necessary to use Uranium more efficiently. It's only in the last 3 or 4 years that the world has collectively realized the extent of our problems.
> Scaling up renewables and storage is a much smaller multiplier, in comparison.
I feel that you don't appreciate the sheer number of machines we'd have to build to accomplish a renewable only primary energy design. How much mining and refining we'd have to do. Things that we cannot do at scale today without fossil fuels. Nuclear energy's main advantages are the sheer energy density you get, on demand. 6000 power plants to replace global primary energy is a bargain compared to the hundreds of millions of machines we'd have to build to do that with solar/wind. If solar/wind weren't intermittent, it might be doable. But we really suck at storing energy today.
Let's look at a large scale off shore wind farm in the US. We use offshore wind because they get the highest capacity factors and therefore the most bang for the buck. The Dominion offshore wind farm is a 112,000 acre 2.6 GW (peak) project planning to utilize 180 state-of-the-art Siemens Gamesa 15 MW turbines for a cost of 10 Billion USD. Let's say it gets an amazing 50% capacity factor because it is 27 miles off the Virginia coast. That means, it will produce 1.3 GW on average. But wait, we need battery storage because the output can randomly go from 0 to 100% in several minutes. Let's eyeball it and say we want to be able to provide an hour of buffer. So 2.6 GWh of battery storage. That is the same as the combined capacity of all the grid batteries that existed in the US in 2020. Source: https://sandia.gov/ess-ssl/gesdb/public/statistics.html
Or we could build a single APR-1400 nuclear plant for 1.35 GW on demand. The UAE just built 4 of them in 10 years, for 6B each. And each of them will last at least 2 times as long as the turbines.
> That's a very silly way to frame it because there's only like 3 commercial breeder reactors today. Why is that the case? Because energy has been cheap, we know how to do PWRs so there's little reason to do the work necessary to use Uranium more efficiently. It's only in the last 3 or 4 years that the world has collectively realized the extent of our problems.
There are zero commercial breeders today. There are zero demonstration breeders that use Pu239 or U233 as their fissile fuel. There is one reactor doing something approximating breeding by playing a shell game with MOX and has made no public claim of producing more Pu239 than it consumses under normal operation.
Your argument about not being needed does, however, apply to pumped hydro which has geography for thousands of >50GWh sites all around the world. Only a tiny handful of them are needed.
> I feel that you don't appreciate the sheer number of machines we'd have to build to accomplish a renewable only primary energy design. How much mining and refining we'd have to do. Things that we cannot do at scale today without fossil fuels. Nuclear energy's main advantages are the sheer energy density you get, on demand. 6000 power plants to replace global primary energy is a bargain compared to the hundreds of millions of machines we'd have to build to do that with solar/wind. If solar/wind weren't intermittent, it might be doable. But we really suck at storing energy today.
"Big number scary" is not an argument. What matters is raw materials and cost. These both favor monocrystalline solar by such a large margin that there's no contest. A mix with onshore wind using modular foundations has reached concrete parity with Gen III reactors or an EPR and offshore is not far behind. Nuclear reactors require less iron, but far more chromium. Only the latter is in anything approximating short supply and iron mining for either pales in comparison to the tens of billions of tonnes of ore that need to be mined (or leached where geography or apathy towards the locals allows) to fuel the reactors.
The overwhelming majority of Uranium ore is an absolutely terrible fuel barely more energy dense than coal if used in a PWR (but requiring much more processing), and its only redeeming features are low carbon emissions and portability once you mill and refine it.
> Or we could build a single APR-1400 nuclear plant for 1.35 GW on demand. The UAE just built 4 of them in 10 years, for 6B each. And each of them will last at least 2 times as long as the turbines.
You forgot the mandatory service contract to KEPCO for another $20 billion, that its net power is 1.1GW per reactor even when nothing goes wrong (Korea's reactors historically have about a 7% unplanned outage rate on top of the 15% planned) and that O&M which is on top of the service contract costs as much as replacing the entire wind turbine (rather than just repowering which is commensurable with normal maintenance on a thermal generator assuming there isn't a massive upgrade available for lower than the original price).
