|
|
Subscribe / Log in / New account

The "missing piece" in renewable energy

The "missing piece" in renewable energy

Posted Feb 27, 2025 11:57 UTC (Thu) by farnz (subscriber, #17727)
In reply to: No wonder all commercial efforts have failed by PeeWee
Parent article: Building an open-source battery

Disclaimer: this is my field of work, so I'm not unbiased here - I work on renewable generation and lithium-ion battery storage systems.

Setting the scene, we have two factual statements:

  1. Over a 12 month period, and assuming everyone's energy demand increases to match the most demanding users today, renewables can provide all the energy we need, just not necessarily when we need it. In particular, some areas have a surplus in summer, and a deficiency in winter, while others have enough energy most weeks, but randomly have a week of insufficient renewable supply.
  2. Lithium-ion batteries are good enough and cheap enough to fill the short-term storage need (24 hours), but self-discharge means that they're not suitable for storage over a scale of months. Sodium-ion is likely to be cheaper than lithium, but not improve the self-discharge rate by enough to change niche. This means that arguments about "storing solar for night-time" are already resolved, but not "storing September's energy for January".

This leaves us with a niche to fill; we need something that can absorb energy when we have excess renewables available for more than 24 hours, and then release it months later when renewables hit what the Germans call a "dunkelflaute" - insufficient insolation and no wind over a long period, causing a shortfall in energy - and this niche is where flow batteries are interesting.

Something that fills this niche does not need high power potential - you'll still have your short-term batteries available, so you're filling those steadily while energy consumers discharge them in bursts - but does need to scale in terms of energy storage. It can also be placed somewhere where size and weight is not a significant concern; we happily store tens of thousands of tonnes of supplies for biomass, nuclear and fossil fuel plants, so storing similar amounts for long-term storage is fine. It's also OK to have a slow rate of change of power; you've got the short term storage to cover you as these things switch from charging to discharging, and thus it's fine for them to have response times comparable to coal or nuclear (18 to 36 hours to go from full charge rate to full discharge is fine).

However, what it does need is low self-discharge rates, since the goal is to charge whenever we have more energy available than we can use, and discharge when there's not enough energy. For example, charge on a cool and windy summer's day, when solar is providing the full energy needs over 24 hours, and the wind is "wasted", discharge in a dunkelflaute during a cold snap where we need lots of energy to keep people warm.

The three current top contenders for this niche are:

  • Flow batteries
  • Green hydrogen and fuel cells.
  • Synthetic hydrocarbons made using atmospheric CO2 and green hydrogen.

And of these contenders, note that the two that involve hydrogen run a substantial human risk; it's easy to substitute grey hydrogen (made by steam reformation of fossil fuels, releasing the CO2 into the atmosphere) for green hydrogen, such that there's no actual benefit in terms of the climate, but saving money for the people who do this.

There's also the possibility of a miracle in terms of low self-discharge sodium-ion or lithium-ion batteries, or a completely new tech that nobody's thought of yet. But it's unwise to consider a miracle likely :-)


to post comments

The "missing piece" in renewable energy

Posted Feb 27, 2025 13:04 UTC (Thu) by PeeWee (guest, #175777) [Link] (12 responses)

> renewables can provide all the energy we need

I have yet to see a serious and honest calculation for *all* the demand. The easiest mistake is to just take current *electricity* demand and calculate how much renewables are needed to satisfy it. But that is only a fraction of what we actually need if we want to get rid of all fossil fuels - and no, burning trees is not green ("renewable" being the latest lie that just *sounds* like "green") and not sustainable; people should have listened instead of sleeping through History lessons in school.
Just take electric cars which easily blow the electricity demand of a household to at least double of the status quo, so much even that in Germany an electric utility company had a field test, what would happen if all cars in a test neighborhood were electric, only to find out that the grid cannot deliver that additional demand without expensive load balancing solutions.
Next is heating and not only households but on an industrial scale, which makes households pale in comparison every time. Guess why Island is the prime supplier of Aluminium; because they have geothermal electricity to burn, quite literally. Now you do the math, what happens if and when all industrial scale energy guzzlers, like Aluminium/Steel/Concrete/etc. works run on electricity instead of coal or gas or whatever non-sustainable source.

And *then* we can start thinking about efficiencies and the price (also and most importantly: energetically) of storage solutions and materials.

Yes, renewables can power all human energy requirements

Posted Feb 27, 2025 13:22 UTC (Thu) by jjs (guest, #10315) [Link] (10 responses)

Current power striking the earth from the sun is 170,000 TW (actually 173,000 - https://phys.org/news/2011-10-vast-amounts-solar-energy-e... - but I'll use 170,000 for easy math). That makes total annual solar energy striking the earth 1,490,220,000 TWh/year. Total human annual energy need is 180,000 TWh/year (https://ourworldindata.org/energy-production-consumption). So if we capture 0.012% of the solar flux, we could provide all power needed for humans. Assuming we only capture on land (30% of earth's surface), and a 20% efficiency (roughly what current solar cells do), that means we need to capture 0.20% of the solar flux on land to power all human energy needs.

Yea, renewables can do it.

Also, remember that roughly 2/3 of primary power is thrown away as waste heat. We won't be doing that as we electrify, since we don't have the inefficiencies of fossil fuel burning, so it's even better.

Yea, renewables can do it.

Yes, renewables can power all human energy requirements

Posted Feb 27, 2025 13:41 UTC (Thu) by farnz (subscriber, #17727) [Link]

Just to extend the calculations further; a bit over 1.5% of the land area is built-up urban land. We need about 0.2% of the land area to have current solar panels over it to produce as much electricity as we currently get from all sources, including fossil sources - that means that we need about 15% of built-up land to be covered in solar panels, ignoring wind (as the other significant renewable source). And that's enough that solar alone, averaged over 12 months, covers all our current energy production; wind and increased coverage of built-up land lets us go for more profligate uses of energy.

Technologically, this is entirely possible; the limiting factors are economic and political. And note that wind power ends up capturing some of the insolation over the sea, since the energy from that insolation drives air pressure shifts.

I'm also completely ignoring geothermal (where practical) and biomass (since determining how much biomass is genuinely renewable, and how much is unsustainable, is a hard problem). These do contribute something, but we don't need them to be confident that we can meet our needs on renewables.

Yes, renewables can power all human energy requirements

Posted Feb 27, 2025 13:56 UTC (Thu) by malmedal (subscriber, #56172) [Link]

Some nice resources like https://globalsolaratlas.info/map and https://globalwindatlas.info/en/ show a detailed view.

Looks like just about every country can get the energy they need with renewables within their own borders.

Yes, renewables can power all human energy requirements

Posted Feb 27, 2025 14:12 UTC (Thu) by PeeWee (guest, #175777) [Link] (7 responses)

Until you realize that we need the raw materials to build solar cells and wind mills. The latter are basically hazardous waste and cannot be recycled since they consist of compound materials that cannot be put back into a pristine state to rinse and repeat by any stretch of the imagination. Ask Denmark who just dismantled their pioneer wind park.

And that's not counting the concurrency with other uses of the land, like such minor matters as agriculture, which also still relies heavily on industrially produced fertilizers from finite resources with high energy inputs. Once those resources run out there will only be one way to scale food production and that is by increasing the area demand - guess why one Bill Gates is investing heavily in farm land. And yes, I know that certain researchers claim that solar cells can be built atop agriculture, in total disregard of feasibility.

There are also huge swaths of land that are simply inaccessible to humans, let alone endeavors like solar or wind parks, e.g. mountains, the (Ant)arktis, Siberia whose permafrost regions are thawing, making the ground miry which prohibits any building activity, huge swaths of Canada, etc. And that's just the northern hemisphere; rainforests and the likes are also off limits, if one does not want to throw out the baby with the bath water or cast out the devil with Satan's help. I think it is safe to say that the remainder of the Earth's surface has been captured already for other uses, like living and supporting human life, e.g. agriculture. As I said, I want an honest, preferably worst case calculation, of what is actually achievable and not some rosy number juggling. I remember an Energy Technology professor of mine - yes, I know a thing or two about the matter as well - gushing about that solar Gigapark - IIRC 100km * 100km - in the Sahara that would cover all human energy needs. Look how that turned out; keyword: Desertec. Plus, that is just another example of neocolonialism where Africa and the likes are just about good enough to serve *our* needs and can sod off when it comes to *theirs*.

Yes, renewables can power all human energy requirements

Posted Feb 27, 2025 14:24 UTC (Thu) by farnz (subscriber, #17727) [Link] (3 responses)

We only need to cover all roofs in the world with current solar panels to meet global energy needs, and build up electricity grids to move energy from where it's being produced to where it's being demanded; we benefit from solar panels over farmland because they protect from weather (although you inherently have to reduce the areal efficiency of the solar farm by spreading the panes out since you need to ensure that enough light gets past the panels to the ground below).

We also need to fix the political and social mess created by the idea that energy supply needs to be under the control of big businesses, and not diffuse; one of the reasons we have such a huge problem is that fossil fuels are inherently not evenly distributed and accessible, and the entities that extract them have gained political power from doing so. Desertec is a typical example of not trying to fix this - because a big entity that makes money from Africa is more politically palatable than every home generating some power, and paying to share resources with each other.

Yes, renewables can power all human energy requirements

Posted Feb 27, 2025 14:35 UTC (Thu) by paulj (subscriber, #341) [Link] (2 responses)

Solar and arable farming are surely mutually exclusive. Crops do not grow well under large panels. And what you can grow is then harder to efficiently harvest.

Yes, renewables can power all human energy requirements

Posted Feb 27, 2025 14:41 UTC (Thu) by farnz (subscriber, #17727) [Link]

Depends on the crop; solar panels over farmland should be implemented as a "farmland first" sort of deal, where you place the panels to shade areas that get too much sunlight, and to direct rainfall to where the plants can make most use of it.

Maximally efficient use of land for solar power is incompatible with farmland, since that needs large panels; fortunately, we can get all the energy we need by only using rooftops for power, so for farmland, you're using the panels where you would otherwise have put up structures anyway.

Yes, renewables can power all human energy requirements

Posted Feb 27, 2025 14:45 UTC (Thu) by corbet (editor, #1) [Link]

There is a growing "solar garden" movement here trying to prove that assertion wrong.

But all of this is getting pretty far off the original topic. There are plenty of places to discuss these issues, but I don't think this is what most folks come to LWN for, so maybe we can wind it down, please?

Yes, renewables can power all human energy requirements

Posted Feb 27, 2025 14:34 UTC (Thu) by jjs (guest, #10315) [Link] (2 responses)

ANY energy conversion means we need to use raw materials to build it, so that's a wash argument. Yes, CURRENT wind turbine blades have limited recyclability (although there are other uses for them, and no, they're not HAZARDOUS waste - that has specific definitions). That's being worked on. All the rest of the material for renewable energy is recyclable.

Concurrency of use - solar panels on roofs (residential and commercial solar), agrivoltaics (solar on farms, which actually can provide benefits to the farming side, as well as a source of income for the farmers), most wind farms are on actual farm land - the owner continues to farm/ranch around the wind turbines, because the majority of the land for a wind farm is to space out the turbines so they don't interfere with each other. Again, a source of guaranteed income to the farmers.

Yes, there's areas that are inaccessible. Simple answer - don't put your power production there. When you need less than 1% of the land to power the world, you have a lot of options.

Yes, renewables can power all human energy requirements

Posted Feb 27, 2025 14:51 UTC (Thu) by farnz (subscriber, #17727) [Link] (1 responses)

Especially in the context that over 1.5% of land is built-up (cities and towns); we're in the fortunate position that everything other than solar panels on the roof is not needed to meet our annual energy needs from solar, "merely" to provide diversity of supply and reduce the need for energy storage, and thus you can use it in places where the energy generation is secondary.

For example, wind farms are also used to provide wind breaks for structures in the area; one of the things that the climate crisis has caused is stronger winds, and if you can extract energy from that wind (turning into electricity), you get storms that match historical norms, instead of stronger storms, reducing the damage done. The only issue then is accepting that the wind farms (and solar panels on farmland, and so on) are set up with electricity generation as a secondary thing, not as their primary goal; it's something you get "for free" when you solve the primary problem (like needing to shade part of a field, or protect a village from storms), and thus being bad at electricity generation is A-OK.

Yes, renewables can power all human energy requirements

Posted Feb 27, 2025 15:41 UTC (Thu) by Wol (subscriber, #4433) [Link]

It's all about diversity, and things work out well. So wind turbines reduce wind - obvious. So does hedging - plant a coppice hedge and bingo you've got a plentiful supply of renewable bio-crop to burn ... yes growing wood and shipping it thousands of miles as a "green fuel" is daft...

And then coppiced woodland (and many rotational farming schemes) are incredibly efficient and bio-diverse, both at the same time ...

Cheers
Wol

The "missing piece" in renewable energy

Posted Feb 27, 2025 23:34 UTC (Thu) by intgr (subscriber, #39733) [Link]

> Guess why Island is the prime supplier of Aluminium; because they have geothermal electricity to burn, quite literally.

This is a common misconception. The majority of Iceland's energy comes from hydropower (~80%), not geothermal.

Long term storage options

Posted Feb 27, 2025 14:50 UTC (Thu) by jjs (guest, #10315) [Link]

Pumped hydro - there's plenty of places for closed-loop pumped hydro - we only need to be able to develop 1% of it (https://re100.eng.anu.edu.au/pumped_hydro_atlas/)

From that:
"Our Atlases currently contain 820,000 non-overlapping sites with 86 million GWh of energy storage potential which is equivalent to about 2 trillion EV batteries. Most of the Atlas sites are off-river and do not require any new dams on rivers. Land and water use is very low. PHES constitutes >90% of electricity storage worldwide because of its low cost. Batteries are preferred for storage of seconds to hours, and PHES for overnight and longer."

Their types of pumped hydro:
Type Unprotected Area Atlas Protected Area Atlas Explanation
Greenfield Unprotected Areas Protected Areas Two new reservoirs
Bluefield Unprotected Areas Protected Areas At least one existing reservoir
Brownfield Unprotected Areas Protected Areas Repurpose mining sites for pumped hydro reservoirs
Ocean Unprotected Areas Protected Areas Use the ocean for the lower reservoir
Seasonal Unprotected Areas Protected Areas Store water for longer periods
Turkey’s Nest Unprotected Areas Protected Areas Create PHES reservoirs on flat ground, for more siting options

Other: Interconnect to get around the area. Example: When was the last time the entirety of North America saw no wind or sun for weeks? I don't think it's happened.

Re hydrogen: I doubt it will be a factor in long term. Hydrogen is a terrible material for energy storage, due to the small size of the molecule, the poor volumetric energy density, and the fact it has to be created (i.e. the inefficiency of electrolysis, pumping, storage, and then conversion back to electricity via fuel cell). Also, all those steps have maintenance requirements, which are not small. Example: Fuel Cell bus maintenance costs around twice that of diesel bus maintenance costs for the same miles traveled.

You need to either do massive compression (which costs energy), cool it down to 20K (which takes energy), or use large volumes. And regardless of your storage, that small molecule gets out. NASA, who are experts with hydrogen, took 5 wet dress rehearsals to successfully fuel the Artemis mission. There's a reason all the commercial launch vehicles are going to either methalox or kerolox, instead of hydrolox.

Add in that it's a potent greenhouse gas (indirectly), and losses estimated around 1% for every connection in the pipeline, it's just not a great idea for energy. It's very useful for chemistry and chemical engineering. For energy? Not in my opinion.

The "missing piece" in renewable energy

Posted Feb 27, 2025 15:28 UTC (Thu) by Wol (subscriber, #4433) [Link]

> And of these contenders, note that the two that involve hydrogen run a substantial human risk;

Dunno how easy this is - a caustic solution and artificial chlorophyll. Pump surplus electric in, get methanol out. Given that we invented oil cracking not that long ago and made petrol as cheap as diesel using zeolites, is there any way we can bubble CO2 through a charged zeolite to give a similar effect?

Cheers,
Wol

The "missing piece" in renewable energy

Posted Feb 27, 2025 19:49 UTC (Thu) by sfeam (subscriber, #2841) [Link] (1 responses)

"This leaves us with a niche to fill; we need something that can absorb energy when we have excess renewables available for more than 24 hours, and then release it months later when renewables hit what the Germans call a "dunkelflaute" - insufficient insolation and no wind over a long period, causing a shortfall in energy - and this niche is where flow batteries are interesting."

I may be missing something, and my perspective is probably skewed because I live in an area where most (>80%) of the electric power is hydroelectric, but isn't that niche already nicely filled by storing water in a reservoir in the winter when rain is plentiful and sunshine is not?

The "missing piece" in renewable energy

Posted Feb 27, 2025 20:48 UTC (Thu) by malmedal (subscriber, #56172) [Link]

Not all locations have suitable mountains/hills for reservoirs.

Lithium-ion batteries self-discharge something like 0.3 to 3% per month, so you could store energy for quite some time.

At current prices, $50/kWh, they are very profitable when you get paid for a discharge every day. If you charged your batteries in summer and sold in winter you'd only get money once per year.

With flow-batteries, the reactants and tank-space promise to be less than a dollar for an extra kWh so it becomes economically feasible.


Copyright © 2025, Eklektix, Inc.
Comments and public postings are copyrighted by their creators.
Linux is a registered trademark of Linus Torvalds