Hacker News new | past | comments | ask | show | jobs | submit login

> Solar and wind already are cheaper than anything else out there

Yes if you conveniently choose to ignore the availability problems (in the night and in no wind). And that no scalable battery/storage solution exists yet?

How are batteries, power-to-gas, and solar-thermal with molten salt not scalable?

Because they can provide nowhere near the amount of storage required. For example, one of the flagship solar storage facilities here in California has a capacity of 300 MWh. By comparison, the Diablo Canyon nuclear power plant generates 2,200 Megawatts. That storage facility can store less than _ten minutes_ worth of energy that the nuclear plant outputs.

The German natural gas infrastructure can in fact store several hundreds of TWh.

What do you mean natural gas infrastructure? Pneumatic storage?

Lots of German households and industries use natural gas for heating, hence we've got something like 400,000 km of pipeline as well as storage facilities sitting around. Instead of buying natural gas from Russia, we could fill them with synthetically produced methane and/or hydrogen, generated during off-peak hours.

Synthetic methane production has a theoretical maximum efficiency of 30-40%. So we'd need to build 3x as much power and is actually needed, and a bunch of electrolysis and Sabatier reaction plants on top of that.

Or we can just follow France's example and build nuclear plants.

Those numbers are outdated. Round-trip efficiencies of up to 80% have been claimed:


If that's legit and can be made cost-effective, the case for nuclear is largely gone...

Efficiencies of over 100% have been claimed. Over 50% of the US patent office's applications are for entropy-reversing devices.

Come back to me once these claimed figures are actually implemented.

Reverse fuel cells aren't magic, though. I agree that 80% seems rather ambitous...

...in the form of methane. How do you store wind and solar?

Uhm, well, electrolysis and Sabatier reaction, and then you clean it up and compress it or something. Unfortunately, you need a source of carbon dioxide (no, not air, extracting a trace gas is rather impractical), and then the whole process has a round trip efficiency of certainly no better than 20%. Looks like we need a 4x or so overbuild of unrealiables so that methanation can keep the lights on in winter.

What confuses me is that there are much more practical chemical storage methods nobody talks about. Ammonia comes to mind. It's easier to make and easier to store. I can't help but think that the whole methanation idea is a PR stunt by the gas industry, intended to positively associate renewables with fossil gas in the minds of the unwashed masses.

German Wikipedia lists round-tip efficiencies of 30-38%, and 43–54% if you cogenerate heat.

I've already linked a paper which makes promises of efficiencies of up to 80% using reverse fuel cells.

Here's an older one that promises 'only' 70% efficiency, using caverns for CO2 and CH4 storage:


Not sure how much of these claims will survive after contact with reality...

Heat cogeneration doesn't really improve efficiency of the methane generation process itself. Cogeneration refers to using the waste heat of a thermal plant to assist in some other facility. Using the waste heat of a gas plant to heat water in desalination is an example of cogeneration. So it does improve overall energy efficiency, but it assumes that there's a convenient source of heat next door. Wind and photovoltaics don't generate any significant amount of heat though, so there's no opportunity for cogeneration.

Cogeneration happens at the gas-to-power side of things. Doesn't help you with electricity generation, but that's ok as the goal is reduction of emissions across all sectors.

The source calculates somewhat optimistically. They assume storage at 80bar, while mentioning that actual storage is at 200bar. They also assume 60% efficient conversion from methane to electricity, while using 55% in other parts of the paper. There is no accounting for transmission losses or the energy needed to procure the CO2. Cogeneration is again creative accounting. We're talking about supplying electricity, and heat isn't electricity.

Those reversible fuel cells... I'll believe in them when I can buy them. And a round trip efficiency of 80% is unbelievable when simple electrolysis of water, which is only half the round trip, isn't that efficient.

They probably can scale, but haven't scaled yet. The only proven solution today is pumped storage, which is not available everywhere. Pumped storage accounts for >95% of world's installed energy storage capacity.

Not being needed to scale yet and not being able to scale are two completely different things though.

Simple: because at high levels of intermittent penetration, you have to build so much "non-productive" storage technology that the Energy Return on Investment is below the level thought necessary to sustain industrial civilization. For example, this was just published last week:

[1] https://www.sciencedirect.com/science/article/pii/S2211467X1...

(Iñigo Capellán-Pérez, Carlos de Castro, Luis Javier Miguel González, Dynamic Energy Return on Energy Investment (EROI) and material requirements in scenarios of global transition to renewable energies, Energy Strategy Reviews, Volume 26, 2019, 100399, ISSN 2211-467X, https://doi.org/10.1016/j.esr.2019.100399.)

I accidentally clicked get PDF on the recommended paper from the page, this one:

"Will EROI be the Primary Determinant of Our Economic Future? The View of the Natural Scientist versus the Economist"


Which actually answers why the point you bring up is not relevant:

"In a recent meeting of scientists and economists in London, economists raised eight points as to why it was not necessary to consider EROI in determining future energy availability or policy."

Now the above paper tries to refute those points, but using invalid logic, e.g. arguing that because EROI and costs are linked in oil and gas, then the same observation must hold across categories to renewables.

That paper has problems.

The red flag that stood out to me is that they show huge increases in tellurium, gallium, and indium demand in Figure 10. Those materials are only required for thin film solar technologies. But according to Table 2, their scenario includes one PV technology: fixed-tilt arrays of silicon PV. Where are the increased demand for tellurium, gallium and indium coming from? It reads like they copy-pasted information from prior studies without paying attention to their own scenario parameters.

People are not only powering up their "macbook pros" to watch netflix. Try to run steel mill or any factory on batteries, solar or molten salt, even decent size server farm is not going to run well on solar or wind.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact