Solving renewable energy’s sticky storage problem

When the Sun doesn't shine and the wind is calm, humankind still needs power.

See full article...

 

[Frank C.]

Storage is storage, but when you need 1000x the current capacity you start running into limits about cost, physical space and sourcing materials. If you can find something that's 10-100x cheaper then the numbers stack up somewhat better.

You must have missed the part of ever cheaper and the current trajectory of implementation. It’s getting done. All of the battery installs in my part of country went onto already utility owned land that wasn’t being used for anything. Regardless, installations aren’t that big and the issue of physical space is not a concern. Sourcing materials? The number of global manufacturers and the growing battery output from them tells me sourcing materials is not an issue.

 

[andreas-motzek]

about the dreaded battery storage issue. Look, California is getting it done

During the last dunkelflaute, Germany had to import around 1 TWh. This led to a price spike on the electricity market and some unpleasant comments from politicians from Sweden and Norway (where electricity is imported from and where prices also rose).

Building a battery storage facility for 1 TWh would cost about as much as 10 to 20 EPR nuclear reactors (at 13 billion euros each) and use the entire global production of lithium-ion batteries for a year. Of course, it is economically better to let electricity prices spike. But to become CO2 neutral, some smart and cost-effective solutions are missing.

 

[plectrum]

You must have missed the part of ever cheaper and the current trajectory of implementation. It’s getting done. All of the battery installs in my part of country went onto already utility owned land that wasn’t being used for anything. Regardless, installations aren’t that big and the issue of physical space is not a concern. Sourcing materials? The number of global manufacturers and the growing battery output from them tells me sourcing materials is not an issue.

So can you build a battery to store 10TWh economically? That's the UK requirements for a two week dunkelflaute on current demand (most heating being from natural gas).

Let me put that into perspective, that's about 720 million Tesla powerwalls. There are 24 million homes in the UK, so that's 30 each, at a retail cost of about £180k per home. With electrification (heatpumps, EVs) that figure is only going to rise substantially.

(Edit: yes I know the Powerwall at £6k for 14kWh is a pretty awful deal, but it's a handy ballpark as it's an all in one package, rather than pricing up batteries / inverter / installation / etc separately)

 

[afidel]

Sure. It's a minor change to move natural gas plants that mostly burn methane to hydrogen.

But it's a lot cheaper to transport and store liquid hydrocarbons than hydrogen. And if you're using the sun to provide most of the energy for your biofuel, you're investing a lot less grid power in your feed stock.

You're also competing with food stock for land, fertilizer, and water if you want to do it economically. I mean the US can probably do it by continuing to produce ethanol from corn after we transition to EVs for transportation, not sure if our annual production is sufficient to power Europe's backup needs?

 

[saltenjoyer]

When the Sun doesn't shine and the wind is calm

It’s shining and blustery as hell someplace else. Also the tide and waves are always going and geothermal vents don’t take days off. That’s why you put a whole bunch of things in a network and then some of them are always making plenty of power right?

 

[mozbo]

Nuclear is and has been the answer to most of our energy problems for decades. There's too much pearl clutching about nuclear fuel in the US that makes us generate and hang onto gobs of nuclear waste. Build fast breeder reactors and tell the public to get bent. The cats out of the bag plutonium is everywhere. If india and Russia can run them without them going boom the US can as well.

On an article focused on cost ... saying the most expensive option by orders of magnitude is "the solution" is clown-level hilarious.

 

[grimmm]

As battery costs become cheaper and cheaper, there’s absolutely no reason to not install a surplus 3x, 5x or more of actual on-paper needed capacity. Any situation should be readily covered.

The article also doesn't mention the potential for V2G charging from EV batteries, which for a large portion of the population spend most of their time at home or parked at work. If we could financially incentivize the use of, say, 20-30% of pack capacity in those vehicles, it could be used in several different ways. Even this amount of vehicular capacity would be equivalent to 1-2 Powerwalls each and Tesla has demonstrated great results with their virtual power plant solutions in several states.

1) Charge in work parking lots and garages during wind/daylight hours, releasing that energy back to the grid overnight at home (same day storage)

2) Charge to a higher level automatically to offset a projected shortfall and defer charging for up to a couple days in the future, maximizing the potential for grid storage solutions to stretch the inclement period (same week storage)

3) Near instantaneous grid load shaving or ramping to smooth out supply/demand peaks, reducing the need for the utility to invest in BESS or flywheel storage, so they can focus on building long duration energy storage (solving the problems outlined in this article)

Really a win-win-win once the required inverter hardware can be mass produced and mass deployed at the required scale.

 

[randomname19937]

It’s shining and blustery as hell someplace else. Also the tide and waves are always going and geothermal vents don’t take days off. That’s why you put a whole bunch of things in a network and then some of them are always making plenty of power right?

The problem here is political rather than technical - but nonetheless very real. The "someplace else" might be unwilling to sell you the energy when you need it - or hike the prices, or whatever.
They have the power (pun intended) and whole regions aren't very keen on putting themselves on somebody else's mercy. Note that we're talking about very short timescales here - weeks! If you can't secure your country's energy for weeks, you're screwed. Sure, many countries do import oil - but they also have months of reserves to secure new sources should the international situation change. It would seem rather prudent to have the same kind of security when it comes to grid electricity.
Then there's war and sabotage.
Can Europe put underwater cables and build power plants in Morocco? Perhaps. Can Russian ships drag anchors across those cables - the hell they can. Let Nord Stream 2 explosion serve as a memento.

 

[BuckoA51]

Tony Seba, who has a good history of accurate predictions as a futurist, says the solution to the problem of renewable to energy storage can mostly be extrapolated from trends. Solar keeps getting much more efficient at lower light levels - it used to be worthless when it's cloudy, now the output falls a lot but is still significant. This keeps improving. Plus solar keeps getting much cheaper, really fast. By his calculations, for most places, for most needs, large grids to send power around + massively overbuilt solar is the cheapest solution to provide enough energy all the time. Not even taking CO2 emissions into account - it will simply be the cheapest, even compared to fossil fuels. Storage only needs sufficient to handle nights, not long stretches of clouds.
This solution also has the upsides that if you want to build something to capitalize on intermittent massive energy excesses, you're going to have those a lot, if the system is so overbuilt as to provide for base load when you have long stretches of clouds.
I don't know if he's right, but it's interesting.

There are 100s of studies saying he's right, what's needed more than storage are grids that can get electricity where it's needed more efficiently and reliably than they do now. It's pretty rare to have a completely still, windless night across a whole country. Don't get me wrong storage has a part to play, particularly as you say to keep solar farms going through the night, but this idea we need billions of dollars of batteries or other storage, or dozens of nuke plants to deal with intermittency is just wrong.

 

[1Zach1]

I don't think this is going to be a one size fits all solution, we are going to have a mix of overnight battery's, long duration batteries, and other technologies for those extreme examples. I'm personally not sold on the magic of hydrogen storage, but if someone can make the economics work and not rely on fossil fuel derives hydrogen then I would welcome it to the mix. This combined with better long distance interconnections are what is going to solve the problem of moving to a renewable future, but this also is going to take decades to implement.

 

[redtomato]

Space elevators. The answer is always space elevators.

When electricity is cheap, send mass into geosynchronous orbit. When electricity is expensive, bring it back down.

I believe the minimum energy required is about 100MJ / KG to take mass from surface to geosynchronous, costing maybe $10 / KG in electricity required. In other words, about $10,000/ton. Could be a lot more or a lot less. Then you can recover that electricity by bringing the mass back down, on a 35,000 mile trip from geosync back to the Earth's surface.

If we have a zero-gravity marshalling yard / storage at the top of the elevator, say, able to store a megaton (1 million tons) of mass, that's roughly $10 billion of electricity storage available at the top.

Plus there may be one or two other rather nice things possible when you have a space elevator(s), but I leave that as an exercise for the advanced reader.

 

[adamsc]

I would think the solution would be more R&D into nuclear energy.

I’d love it if we had France’s power blend, too, but we don’t and the problem isn’t R&D but time. Sadly even a gigawatt won’t let us go back to 1970 and start building nuclear power plants so any nuclear advocacy should start with how you can it online before 2050. It’s popular to blame environmentalists for this but I’ll note that there are an awful lot of “environmentalists” who drive huge SUVs to suburban homes with immaculate lawns, eat mostly beef, and otherwise show little sign of being environmentally conscious except when it comes to blocking things like nuclear plants or transit projects. I mention that because they aren’t going to go away or be shamed by message board chiding because that wasn’t a deep principled objection but rather a convenient tactic to prevent change or “preserve property values”, so you really need a plan for dealing them.

 

[athlon11]

That was addressed in the article. That only works well in mountainous areas. With droughts expected to be more common in the future, one cannot expect to use the massive volumes of water for energy storage that would be required without a large elevation change.

Works great along lakes as well. https://en.wikipedia.org/wiki/Ludington_Pumped_Storage_Power_Plant

 

[airscottdenning]

Storage has a role to play for sure, in "moving" energy across TIME (from when we have it to when we want it). But I'm quite surprised that the article doesn't discuss the MUCH more cost-effective strategy of moving energy across SPACE (from where we have it to where we want it).

The "dunkelflauten" idea of "dark calms" is naive. Dark calms are geophysical phenomena caused by meteorology and astronomy, and quite predictable. Most importantly, they have a characteristic radius of about 1000 km in the middle latitudes of planet Earth. To the extent that the power the grid is larger than this characteristic size, dark calms have less and less impact on the balance between supply and demand.

Balancing power supply and demand (a.k.a. "load") on a 21st century grid is accomplished by a combination of
1) Mixing generating supply with complementary time properties (wind vs solar vs hydro vs geothermal vs nuclear)
2) Peak shaving (demand management via smart grid tech)
3) Long distance transmission via high-voltage direct current and
4) storage (including batteries and all the other innovations discussed in this article)

Certainly storage is an important component here, but it's best seen as a very expensive "last resort," along with nuclear and advanced geothermal ("clean, firm power"). Only by integrating all four strategies do we get a power grid that's clean, reliable, and affordable. ALL THREE!

Jumping straight to storage or nukes implies that "affordable" doesn't matter.

 

[fenris_uy]

I grew up somewhere with LOTS of hydro dams. I have a soft spot for them! But the reality is, building a hydro dam is a bit of an ecological disaster. Pumped storage just makes that worse, really, because it's all the downsides of expense and ecological destruction with only storage capability, none of the new power generating capacity.

As far as "somebody just has to pay," I mean, yeah. If everything were free storage would be easy.

Looking at how several of the bigger dams in the US have their lakes under their designed levels, you probably have some cheap storage already available, just let those dams get back to their designed levels. Either by using them less during the day, when the winds is blowing. Or by allowing them to cut the flow a little more for weeks to downriver.

 

[Unthinkable537]

On an article focused on cost ... saying the most expensive option by orders of magnitude is "the solution" is clown-level hilarious.

You can have cheap or you can have reliable. They don't go hand in hand. Anyone that says otherwise is clown level hilarious. There's no reason not to boost renewables. However, there's also a need for continuous reliable power. Unless you like rolling blackouts?

 

[odikweos]

Why no commentary on using excess power to create biofuels to feed the existing power plants? It seems that if you've already got a massive infrastructure for boilers and power generation that converting them to operate on bio-derived fuels is a much lower infrastructure option than any of the ones mentioned.

I don't think this is a good idea. Didn't downvote it because it's something worth discussing.. but:

1. microparticles from burning things damages lungs and the environment generally
2. the losses due to efficiency aren't trivial.

Overbuilding renewables and then overbuilding storage would probably be a lot more efficient, and we're already doing so.. so a lot of your argument that "we're already doing it" applies there too.

 

[astack]

Is lead actually cheap, when you start needing to store GWh or TWh? Won't supply and demand raise the price when you buy that much?

Yup! Not just cheap, but plentiful and economically efficient to mine: modern lead mining recovers more than 90% of the lead in the ore. Compare that to rare earths where the recovery is something like half to two thirds of the REEs in the ore.

It's so cheap, that's why they used to make pipes out of it and put it as an additive in gasoline. Also it is easy to weld and soft, which makes it easy to machine, which is partly why it is so toxic because it is volatile and requires extra care to avoid exposure.

 

[OrvGull]

Looking at how several of the bigger dams in the US have their lakes under their designed levels, you probably have some cheap storage already available, just let those dams get back to their designed levels.

The problem is most of them are below design level due to drought. There's no water to refill them.

 

[OrvGull]

You can have cheap or you can have reliable. They don't go hand in hand. Anyone that says otherwise is clown level hilarious. There's no reason not to boost renewables. However, there's also a need for continuous reliable power. Unless you like rolling blackouts?

Nuclear plants haven't proven especially reliable in recent years. They're always offline for one reason or another, meanwhile ratepayers are still paying the bill for their construction costs.

It also takes 30 years to build a nuclear plant. I was a nuclear supporter but I've had to admit the timeline and the cost per watt just don't work out.

 

[HalfAHero]

Each flywheel can store 32 kilowatt-hours of energy, close to the daily electricity demand of an average American household.

How does the cost for one of these compare to a home battery, and can I get one buried in my backyard?

 

[DaiMacculate]

Space elevators. The answer is always space elevators.

When electricity is cheap, send mass into geosynchronous orbit. When electricity is expensive, bring it back down.

I believe the minimum energy required is about 100MJ / KG to take mass from surface to geosynchronous, costing maybe $10 / KG in electricity required. In other words, about $10,000/ton. Could be a lot more or a lot less. Then you can recover that electricity by bringing the mass back down, on a 35,000 mile trip from geosync back to the Earth's surface.

If we have a zero-gravity marshalling yard / storage at the top of the elevator, say, able to store a megaton (1 million tons) of mass, that's roughly $10 billion of electricity storage available at the top.

Plus there may be one or two other rather nice things possible when you have a space elevator(s), but I leave that as an exercise for the advanced reader.

I am also a huge fan of space elevators, the main issue is we don't currently have a cable material strong enough. But yes, having a space elevator (or more ideally, a series of them) anchored to a massive asteroid covered in solar panels and recovering the vast majority of energy required to send the attached containers/passenger vehicles up to the anchor point when sending a counter weight down would solve a lot of issues

 

[MST2.021K]

Each flywheel can store 32 kilowatt-hours of energy, close to the daily electricity demand of an average American household.

Flywheels for everyone!! Just stick 'em up in the attic and orient it so it won't slice anyone in half during catastrophic failure.

 

[plectrum]

Yup! Not just cheap, but plentiful and economically efficient to mine: modern lead mining recovers more than 90% of the lead in the ore. Compare that to rare earths where the recovery is something like half to two thirds of the REEs in the ore.

It's so cheap, that's why they used to make pipes out of it and put it as an additive in gasoline. Also it is easy to weld and soft, which makes it easy to machine, which is partly why it is so toxic because it is volatile and requires extra care to avoid exposure.

A quick back of the envelope. There were 4.5 million tonnes of lead mined in 2023. Let's say it takes 10kg to make a 1kWh battery (12V 83Ah). The annual global production of lead would make 450GWh of batteries, which is enough to run the UK for 15 hours. That's not nothing, but the UK would need about 10x as much, and other countries might have similar demands.

What could be interesting though is lead batteries where the lead is in foil rolls like lithium cells, rather than heavy plates. I wonder if that would substantially reduce the amount of lead needed per kWh. (IANA electrochemist)

 

[andygates]

Sure. It's a minor change to move natural gas plants that mostly burn methane to hydrogen.

But it's a lot cheaper to transport and store liquid hydrocarbons than hydrogen. And if you're using the sun to provide most of the energy for your biofuel, you're investing a lot less grid power in your feed stock.

Hasn't "just make some hydrocarbons" turned out to be hugely expensive? I'm thinking F1 and aviation, where money is sloshing around, and it's still too pricey to scale.

 

[Panama]

What about Green Ammonia or biodiesel from industrial hemp? If we only need enough storage for a few weeks a year, these ( or similar dual use products) could play valuable roles due to their energy density and cheap long term storage.

 

[cbf123]

We've had the answer to non-fossil fuel based energy and non-greenhouse gas emitting energy for 70 years: nuclear power. The modern environmental movement has done immense damage to their own cause by their irrational hatred of nuclear power

I'm hopeful about nuclear, but so far the data I've seen shows that it has been very expensive compared to other power sources. And historically it has only made sense for larger power plants.

Standardized designs and SMRs might help with that, but those have yet to be commercially proven.

 

[1Zach1]

How does the cost for one of these compare to a home battery, and can I get one buried in my backyard?

It appears to be 2-3x the cost of current home battery systems of similar capacity.

 

[cbf123]

As battery costs become cheaper and cheaper, there’s absolutely no reason to not install a surplus 3x, 5x or more of actual on-paper needed capacity. Any situation should be readily covered.

When you need to store two weeks worth of energy, chemical batteries are very expensive compared to things like compressed air or hydrogen or gravity-based storage.

 

[OrvGull]

It's baffling how this is even a controversial opinion. We need nuclear, and we need it yesterday.

That's the problem. We need it yesterday and building it takes 30 years. Then it's shut down a lot of the time for maintenance and recertification.

 

[andygates]

Looking at how several of the bigger dams in the US have their lakes under their designed levels, you probably have some cheap storage already available, just let those dams get back to their designed levels. Either by using them less during the day, when the winds is blowing. Or by allowing them to cut the flow a little more for weeks to downriver.

That's very much an engineer's solution, ignoring all the current technical, legal, social and so or reasons why that's not being done already.

 

[cbf123]

It’s shining and blustery as hell someplace else. Also the tide and waves are always going and geothermal vents don’t take days off. That’s why you put a whole bunch of things in a network and then some of them are always making plenty of power right?

Two years ago we had no wind for roughly a week across a thousand km of Canadian prairie. (And it was winter so the days were short and power consumption was near the annual maximum.) You'd need either very long transmission lines or very large amounts of energy storage to cover that.

 

[Veritas super omens]

Why no commentary on using excess power to create biofuels to feed the existing power plants? It seems that if you've already got a massive infrastructure for boilers and power generation that converting them to operate on bio-derived fuels is a much lower infrastructure option than any of the ones mentioned.

We need to rapidly lower atmospheric CO2. Not add to it.

 

[saltenjoyer]

Two years ago we had no wind for roughly a week across a thousand km of Canadian prairie. (And it was winter so the days were short.) You'd need either very long transmission lines or very large amounts of energy storage to cover that.

Wouldn’t Canada need pretty long transmission lines anyway?

 

[OrvGull]

We need to rapidly lower atmospheric CO2. Not add to it.

Presumable the carbon in bio derived fuels would have already come from the atmosphere -- that's where the bio part comes in.

 

[Veritas super omens]

Where does Pumped Storage Hydropower fit into the picture? It's already in some use in a number of countries, including a number in the US. I guess one big question is whether there are good geographic locations to build out more of them.

I also think the authors missed this one. If you are building massive, expensive infrastructure, such as compressed air storage can't you build pumped hydro in a similar manner?

 

[Oafish]

Isn't the only benefit of hydrogen as an energy storage medium that you can transport it? Like, if Egypt has an excess of energy they can store it in hydrogen and sell it to Germany.

Throwing some DC cables between Egypt and the European grid would probably be more efficient in the long term.

EDIT: It's 400 miles between Egypt and mainland Greece. I don't know of Crete is connected to the main EU grid or not. But that could be a side benefit of this.

AI says…”The maximum length of a DC cable depends heavily on the gauge of the wire and the current draw, but generally, a DC cable should be kept under 100 feet to avoid significant voltage drop, with longer runs requiring thicker gauge wires for higher current applications; for low current applications, a longer cable may be possible with a smaller gauge wire.”

I stand corrected.

 

[schlock]

The simplest solution is to decentralize how power is stored by moving the generation to households (alongside meaty subsidies) where applicable. If each household were to buffer 100KWh of battery (alongside solar) this would almost remove the need for a power grid altogether, as well as greatly reducing the inefficiencies related to distance.