[manfredo]

The factor that drives up price for grid level storage is scale. Only ~300 GWh worth of batteries is produced globally each year. The world uses 2.5 TWh of electricity each hour. If anyone tries to install battery storage at a significant scale, demand will vastly outstrip supply and drive prices up.

[tzs]

That's lithium batteries, isn't it? For storage to balance out fluctuations in renewable sources you shouldn't need to use lithium.

It's used in cars and consumer devices because it can store a lot of energy for its size and weight and you don't have to mollycoddle it to avoid memory effects.

Those are much less important concerns for this application. You'd build you battery facilities somewhere outside your cities, perhaps near where you build your solar farms, and you don't need the batteries to move. Batteries that take up more room and/or weigh more than lithium batteries for a given capacity should be fine.

[Manfredo_1]

Right. Lithium batteries won't cut it. That leaves geographically-dependent hydroelectricity, which isn't so easy to build. And then proposed solutions that are still in the prototyping phase, and aren't commercially available.

[VBprogrammer]

I've really enjoyed reading your contribution to this discussion. To the extent I kind of wished there was a private message function.

Partially this is because we have similar views on a lot of the challenges facing a move to renewables. I think sometimes this comes across as being sceptical of the progress of renewables.

In my case, and I suspect in yours, that's not really the case. In fact I'm excited and interested in how we will solve these problems in a variety of different ways.

I think we are in agreement that lithium isn't going to be the answer to energy storage at grid scale. If for no other reason than being in direct competition with the electrification of transportation isn't ideal.

Personally I'm hopeful that Ambri's liquid metal battery will materialize.

What developments do you have your eye on?

[jeremysalwen]

Assuming there aren't economies of scale. Demand for solar had gone way up in recent decades (e.g. in germany, before it was cheap), and the price subsequently went down.

[mlyle]

It's not clear batteries will do the same. While there's been effort to make batteries less reliant on scarce natural resources and mining, there's no guarantee we really get there. If we don't, price can be expected to go up, not down, with scale.

I would bet on price going down slightly with scale, but one can't really tell now what will happen: it might go up a lot, it might go down a lot, or it might stay flat.

[pfdietz]

There are thousands of different chemistries for batteries. The nuclear stans are betting that all of them fail.

[imtringued]

These discussions are always super boring. There are dozens of technologies that can scale to the level needed but everyone goes goes lithium ion and pumped hydro as if everything else didn't exist.

Sure cell batteries might not work, we can try out flow batteries, we can try liquid metal batteries, we can try hydraulic hydro storage, we can try out hydrogen, we can try compressed air, we can try electrolyzing iron or aluminum, we can try another dozen different things and it is highly likely that at least 3 will work out just fine.

[Manfredo_1]

None of those other battery chemistries are seeing the massive growth that lithium ion batteries have experienced. The nuclear "stans" are just pointing out that these are potential solutions, not actual solutions. If iron oxide batteries, or some other chemistry, suddenly becomes cheap and easily deployed at the TWh scale, great. But until then they're not a solution.

[pfdietz]

Yes, you all are engaging in the "nothing can be invented" argument. It's profoundly reactionary, and also hypocritical, because nuclear itself is dead without great improvement. Uranium quickly runs out if the world is powered by burner reactors and known uranium resources, so either massive seawater uranium extraction or breeding cycles would be needed.

Batteries have the advantage of being explorable at a small scale. Now that the potential market has become so clear this is happening, in many companies.

[mlyle]

> Yes, you all are engaging in the "nothing can be invented" argument.

No, we're engaging in the "this has been resistant to being invented so far, so let's not bet everything on it showing up tomorrow" argument.

> Uranium quickly runs out if the world is powered by burner reactors and known uranium resources

You could quadruple the present rate of uranium use, representing in a major contribution to mankind's energy use, and have 35 years of supply, just using known reserves and no breeding.

And if you were using that much uranium, more reserves would be quickly proven. Do you think we've found all the uranium we'll ever find, even if market prices go up significantly?

And breeding is possible, and understood. Yes, there's proliferation concerns, but that's not the end of the world.

And seawater extraction is practical without much increase in cost.

No one is saying "no renewables" or "no battery storage" or "no pumped storage". Or "no power to gas to power". We need all of these things. And we need the diversity of having nuclear in the mix, too.

[Manfredo_1]

It's not "nothing can be invented". It's "come back to me after it's invented, not before".

And uranium seawater extraction already exists: https://www.forbes.com/sites/jamesconca/2016/07/01/uranium-s...

It's more expensive than mined uranium, but since fissile material is so energy-dense that increase in fuel cost amounts to hardly any change in overall cost.

[imtringued]

There is very little demand for grid storage. The biggest problem with energy storage is that it's a red herring until the 30s and 40s.