> Global battery production is...

Now repeat the same math for global production of new nuclear power plants, in GW/year.

While writing your reply about how this isn't the right way to think about the future of nuclear power, realize that those exact same arguments apply to batteries, too.

No, the same arguments do not apply. Lithium ion battery production is essentially a chemical industry. It cannot become cheaper than its input materials. Most of the cobalt produced goes to lithium ion batteries [1]. Eventually, the battery industry hits the ceiling of what extraction industries can supply and shortages are already on the horizon [2].

Nuclear power plants are mostly steel, concrete, and copper. They need to be manufactured to precise tolerances, but the cost of a nuclear plant isn't driven by the cost of concrete and steel as rae inputs. The nuclear industry accounts for a negligible share of steel, copper, and concrete consumption. The lion's share of the cost is in the design, approval, and construction of the plants. Unlike batteries, these are factors that benefit from building the same repeated design. A serial run of 40 heat exchangers is a lot cheaper than a one-off production of 3 or 4 heat exchangers. This is why nuclear plants were so much cheaper in the 1970s and 80s when the same designs were built repeatedly.

1. https://www.statista.com/statistics/1143399/global-cobalt-co...

2. https://www.reuters.com/business/energy/shortages-flagged-ev...

There are two flaws with your argument. First, it’s actually somewhat unlikely that lithium ion batteries are going to be the cheapest grid storage. Second, natural resources almost always get cheaper to extract as demand increases, especially when they’re not actually scarce (there’s tons of Lithium out there).

> First, it’s actually somewhat unlikely that lithium ion batteries are going to be the cheapest grid storage

Then please share what storage mechanism is going to be cheaper.

> Second, natural resources almost always get cheaper to extract as demand increases, especially when they’re not actually scarce (there’s tons of Lithium out there).

Incorrect, as easily accessible reserves get depleted more and more remote reserves must be tapped. This increases cost of extraction.

You’d think so but in reality extraction costs just drop as technology improves. For example, lithium..

Lithium prices have more than doubled over the last decade: https://www.metalary.com/lithium-price/

They've increased more than sixfold since 2000.

"Past performance does not indicate future returns."

All you're doing is repeating the fact that we've been under-investing in lithium (and other energy minerals[0]) for the past decade+. Everyone already knows that.

Nothing indicates lithium mining & refining is at its physical or technology limits. Far from it: we see large low-hanging fruit for process improvements in both.

Ion exchange and/or membrane and/or electrochemical separation vs ponds, lithium carbonate intermediate vs lithium hydroxide, clay vs spodumene, etc. Lots of exciting stuff going on!

[0] https://youtu.be/Keja23R2N7g

Lithium ion is just one type of battery. There are many including quite a few that don't involve exotic materials. For grid storage, energy density is a lot less important.

Even so, we're on track to produce li on batteries for cars by the twh per year. And most of those will have a second life in grid storage. That's before considering the effects of vehicle to grid storage.

There are no plans for nuclear anywhere close to the same scale. Changing that will require a debate about cost. This article does a great job of avoiding that topic entirely.

Why are you comparing cost with industrial output?

There is absolutely no need for batteries to provide for storage, so that is a misleading assertion.

And what, pray tell, would we be using for storage instead? Surely you're not going to suggest some untested and unproven storage mechanism like hydrogen storage, ammonia, thermal batteries, etc. Those technologies have never been built at anything remotely close to grid-scale, and thus we have no idea what they'd actually cost when we attempt to build them at grid scale.

Batteries and hyrodelectric storage are the only energy storage mechanisms for which we have any meaningful cost history.

> what ... would we be using for storage instead?

You are already well aware of numerous alternatives. Insisting they are "untested" is not fooling anybody.

Yes I'm well aware of the numerous alternatives you repeatedly insist are incredibly cheap and scalable, despite the fact that no such systems have been deployed.

Are thermal batteries unproven? I thought they were at the "available for purchase from Siemens and GE" stage of development.

They exist in prototypes, I am not aware of any grid scale energy storage facilities using thermal batteries.

Are there any >1GWh thermal battery facilities? Remember the US uses 8GWh of electricity per minute.

> Are there any >1GWh thermal battery facilities?

I don't know about that, but for comparison, the "world’s biggest battery storage system" opened last year in California[0] at 400MW / 1,600MWh.

[0] https://www.energy-storage.news/expansion-complete-at-worlds...

Which is a lithium ion battery facility. Not a thermal batter facility.

Yes, I realise that. That's why I said "I don't know" and "for comparison".

My point was that GWh-scale thermal battery proposals would be competing with already existing (and possibly cheaper) chemical batteries, and that energy storage on this scale is not unheard of.

The point is that current energy storage facilities are too small for fully renewable grid (or arguably even 60% renewable grid).

Even as a nuclear supporter I would welcome more renewables if we had cheap scalable storage, which no one has it seems. (at least no one pointed out such project in the entire thread)

A minor caveat: some regions have large hydroelectric generation, which doubles as a great energy storage site. Countries seem to be able to manage a roughly 1:1 ratio between intermittent sources and hydroelectric generation. Sweden is managing to decarbonize it's grid with ~1/3rd each of nuclear, intermittent sources, and hydroelectric.

But for much of the world what doesn't have this option, nuclear remains the most viable energy source.

Yes, that's what I meant by "scalable". You can't just continue picking low hanging fruit forever.

Btw even the Sweden example uses nuclear for decarbonization of the grid ...

Certainly there are >1GWh thermal storage systems. I'm not sure if any are set up to generate electricity though.

The only deployed thermal storage systems are in solar thermal arrays. Those have costs that aren't not competitive with nuclear, and they're still subject to seasonal output variability due to weather and Earth's inclination.

And district heating systems. Where a 1-2GWh thermal accumulator will be used.

District heating systems distribute heat, usually scavenged from a power plant. They don't store heat really - they do in a pedantic sense in that heat is "stored" in the distribution pipelines, but not in the sense of a battery. And they don't transfer heat at gradients large enough to generate electricity effectively.

They store heat in the sense that one of the components is a giant thermal accumulator.