[evilos]

The report that the presentation is covering goes over this in fine detail: https://tupa.gtk.fi/raportti/arkisto/42_2021.pdf

> False. If it's a problem for PV it's a much worse problem for existing nuclear plants.

There's basically unlimited quantities of uranium in sea water. Plus you can breed it from thorium if you want to.

> But once you include all the concrete and steel and low level waste that needs decades of storage it's about the same size as a 4hr battery for the entire country.

I highly doubt it but I'd love to see the math on that.

The author of the report I linked concludes that fission cannot be main power source of the future because of the limits of mineable uranium. However he completely ignores the ocean as a source of uranium, which is basically inexhaustible. We don't get it from there today because demand is low and it's cheaper to get it from the ground but ocean uranium capture has been demonstrated.

Solar panels by themselves don't require a lot of rare material but they do require tons of high heat and carbon to "bake". A large part of how cheap they are today is due to the fact that they are made using coking coal in China. But of course the material constraints of the storage needed for solar/wind is the main obstacle. Until we demonstrate cheap storage at scale, wind/solar won't cut it.

[Schroedingersat]

> I highly doubt it but I'd love to see the math on that.

It was very rough hyperbole/fermi estimate. Can't find figures for the US, so using europe.

Europe has 2.5 million m^3 of low level waste and 1.5 million m^3 that's fairly imminent from refurbishments/decommisioning and replacement (an array of 6 by 6 football fields stacked 20 yards). This is only about ~50 years, so it will go up over time.

Europe uses about 2660TWh/yr or 300GW

At 500Wh/L (high but existing) that's 4 hours for the 2.5. At 300Wh/L it's 4 hours at 4 million m^3

A lot of that waste probably doesn't need containment after a decade or two, so it's only a ballpark. (and a real battery can't be that densely packed without overheating). Conversely 5% or so of it needs multiple centuries or millenia so it would accumulate much higher.

[Schroedingersat]

> The author of the report I linked concludes that fission cannot be main power source of the future because of the limits of mineable uranium. However he completely ignores the ocean as a source of uranium, which is basically inexhaustible.

In addition to ignoring sea uranium mining which does not exist in any meaningful way he also ignores a bunch of things which actually do exist like LiFePO4 batteries, the last 10 years of PV research, trains, LEVs and affordable offshore wind.

It also ignores things that are much more likely to exist than uranium sea mining like prussian blue batteries, AlS, iron air, perovskite solar panels and affordable tidal power (all of which are presently undergoing large scale industrialisation).

[evilos]

It doesn't exist industrially because there's no need for it. Uranium has been incredibly cheap for decades. But the cost of fuel is such a small factor in the cost of nuclear energy that you could 10x the fuel costs and the LCOE wouldn't move much at all.

LiFePO4 batteries still need lithium, and any wind turbine > 1MW requires literally tons of copper. These solutions are simply too material intensive, ie not energy dense enough.

[Schroedingersat]

> It doesn't exist industrially because there's no need for it. Uranium has been incredibly cheap for decades. But the cost of fuel is such a small factor in the cost of nuclear energy that you could 10x the fuel costs and the LCOE wouldn't move much at all.

That's a great way of saying 'I believe the capital and operational costs for nuclear are at least 5x as much as renewables and will stay that way'.

Fuel is $1660/kg currently using fairly generous assumptions https://world-nuclear.org/information-library/economic-aspec...

This is $5/MWh in an AP1000 and $10/MWh in lower burnup models. Once the cheap Uranium goes, that doubles. If Rio Tinto or Kazatomprom actually cleaned up their mess properly, that doubles again. Your proposed 100000km^2 sea mining operation is unlikely to be cheaper. There are subsidy free solar farms already at around $13/MWh with more normal prices being in the $30-50 range.

A vestas 660kW turbine uses around ~500kg or 3t per MW net of copper (less than a nuke plant) in the nacelle, busbars and cabinet https://www.copper.org/environment/green/casestudies/wind_en... everything else including transmission can be made from Aluminum. Bigger turbines are more mass efficient as are ones placed offshore. If you wanted to scare monger about wind, focus on the permanent magnets as that's not solved commercially yet.

PV already uses less than a tenth of this. A TopCon or Perc cell uses about 5-10x as much silver as an AP 1000, but silver plated copper metallization reduces this by a factor of 10 and is already being used in demo plants. There are also Aluminum foil backed cells being trialled commercially.

> LiFePO4 batteries still need lithium,

SIB, Fe-Air, AlS and NaS batteries do not, and supply chains for them are better developed than those for PWRs or for still-fictional mining methods.

[evilos]

> That's a great way of saying 'I believe the capital and operational costs for nuclear are 5x as much as renewables and will stay that way'.

Yes, and it still ends up being a better energy source IMO because you don't have to build millions of individual machines. Further, I'd argue that those costs can come down much further and easier than trying to reduce material cost of renewables.

This source (https://help.leonardo-energy.org/hc/en-us/articles/360010919...) puts offshore wind at 6 tons per MW and nuclear at 0.7 tons per MW. I'm not sure if that is accounting for the fact that the location of wind farms are determined by where the wind is vs nuclear plants can be placed wherever. SMRs will help place generation even closer to demand.

Time for napkin math. If we want to convert half of our 490 exajoules of global annual fossil fuel energy to offshore wind energy (let's say all offshore so we can use a generous 50% capacity factor). Then we need 5.22 million offshore wind turbines (3MW each), which will require 94 million tons of copper. That's 11% of all the copper we think exists on Earth. For one generation of wind turbines that will maybe last 25 years. Maybe. This is before we build ANY battery storage, which also needs large amounts of copper, not to mention lithium. It just doesn't make sense to me.

The same calculation for nuclear says we need 6 million tons of copper to replace that same 245 EJ, at a 95% capacity factor. Just 6% of the copper compared to wind. And those plants can go for 60+ years.

Solar PV is even worse than wind for utility scale energy because of its abysmal capacity factor. The main drawback for PV will be the amount of storage or overcapacity needed. So unless we get a miracle breakthrough in energy storage soon, I don't see how wind/solar can scale to replace most of our fossil fuel energy. I'm not against using wind/solar in numbers that make sense. They will definitely be part of the solution, just not the main components IMO.

[Schroedingersat]

> Time for napkin math. If we want to convert half of our 490 exajoules of global annual fossil fuel energy to offshore wind energy (let's say all offshore so we can use a generous 50% capacity factor). Then we need 5.22 million offshore wind turbines (3MW each), which will require 94 million tons of copper. That's 11% of all the copper we think exists on Earth. For one generation of wind turbines that will maybe last 25 years. Maybe. This is before we build ANY battery storage, which also needs large amounts of copper, not to mention lithium. It just doesn't make sense to me.

State of the art offshore turbines are 15MW each (so about 5-8MW net). There are commercial 3MW turbines which only have a relatively tiny amount of copper in the nacelle. 6t/MW is high for the turbines alone, but low if including substations and cabling, so I don't know where they drew the boundary. Worldnuclear uses 27t per MW net.

Here's a commercial wind turbine with no copper windings or permanent magnets https://electriccity2021.windeurope.org/sites/default/files/...

Bus and internal cabling is sometimes Al in smaller turbines as well.

Connecting cables and transformers can also be done with Aluminum. It doesn't happen very often yet because the wind industry lives in the real world where a $1/MWh O&M increase means something won't happen unless it has a commensurable reduction in capital. The second copper rises in price by 50% or so the industry will swap. Al coil substation transformers are already viable.

This cuts out about 80-90% of the copper in onshore wind and 50% or so in offshore (transmission is extremely copper intensive)

Subsea Al transmission cable exists in demo projects but has maintenance issues. It has been tried at scale but would increase costs substantially.

> Solar PV is even worse than wind for utility scale energy because of its abysmal capacity factor. The main drawback for PV will be the amount of storage or overcapacity needed. So unless we get a miracle breakthrough in energy storage soon, I don't see how wind/solar can scale to replace most of our fossil fuel energy

We've had the 'miracle' in the form of sodium ion batteries in about 2018 -- niche safety critical applications for aqueous chemistries are on the market and multi GWh/yr scale factories are presently being built. $60/kWh cell cost is a reasonable expectation (at which point 48hr storage is viable and all PV+battery production is cost competive with nuclear at ~45 degrees north). There are also various projects to radically reduce the cost of brine mining.

Fe-Air is highly promising for 100hr storage but still building out a demo project. AlS and molten salt NaS are undergoing initial commercialisation steps but there have been other technologies at this level that never materialised.

Re. The 480EJ of fossil fuels. There are only 800EJ (2400EJ thermal -- which has relevance for heating and hydrogen) of commercially viable Uranium ore for use in a PWR (even if all the lower efficiency ones are decommisioned immediately). BWRs are a little more efficient, as are Candus (which can use about 4kg of natural uranium where a PWR would use 1kg of enriched). Even in the existing FNR plants (which have resource constraints, cost, reliability, and safety issues to work through before they can be commercial), a closed fuel cycle has never been demonstrated. Reprocessing via MOX and existing designs only adds another 200EJ (600EJ thermal).

Sea mining might be possible, but a world where we can extract the 8kg of natural uranium for 1kg of fuel is a world where we get several thousand tonnes of copper from the same water.

The problem is immense, and the role nuclear energy can provide is only small. Reduction is the primary tool we have.