A supercapacitor made from cement | Hacker News Reader

A supercapacitor made from cement

spectrum.ieee.org

89 pointsrbanffy50 comments

Previously: https://news.ycombinator.com/item?id=36958531

I was hoping this article would have info about the structural strength at large scale, but it seems to be basically the same press-release-level info... Real pictures though, that's nice.

"The team also developed button-size capacitors with different ratios of cement to carbon black but found that while adding more carbon black (above 10 percent by volume) to the mixture increased its storage capacity, this came at the cost of the cement’s structural integrity. And for the use cases that Ulm and colleagues have in mind, the structural strength of the cement is essential."

From this it seems that below a certain amount of carbon black being used, the structural strength is not affected.

As we discussed in the thread I linked, that doesn't change the fact that the concrete is already in multiple pieces, due to the membranes, thus structurally compromised.

Thanks! Macroexpanded:

Low-cost additive turns concrete slabs into super-fast energy storage - https://news.ycombinator.com/item?id=36964735 - Aug 2023 (6 comments)

Researchers have come up with a new way to store electricity in cement - https://news.ycombinator.com/item?id=36958531 - Aug 2023 (57 comments)

MIT engineers create an energy-storing supercapacitor from ancient materials - https://news.ycombinator.com/item?id=36951089 - Aug 2023 (89 comments)

Cement's future could be a combination of carbon capture and electrification - https://news.ycombinator.com/item?id=36339253 - June 2023 (90 comments)

The article states that storing 10kWh of electricity with this will take 45m3 of concrete, weighing 110 megagrams.

One Tesla Powerwall takes 0.127m3, weighs 0.114 megagrams, and stores 13.5 kWh.

Now of course the attraction of this energy storage device is that the materials are cheap, but I have to wonder if storing energy in concrete is cheaper and easier by moving it up and down a gravity field rather than trying to use it as a capacitor.

For example, take a look at this 100MWh gravity storage facility being built in China: https://www.energyvault.com/project-cn-rudong

Isn’t the idea that with this cement mixture your house’s foundation acts as energy storage “for free”?

I think the application is for cases where you’re already using cement as a building material, so you might as well mix in some carbon and also get energy storage out of it.

Is it free? They don't mention home energy storage as a potential application. I would have no trouble believing that the extra electrical work and additions to the foundation beyond what's involved in a traditional cement foundation cost a whole lot less than a lithium battery.

I'd also be concerned about problems with the foundation. Foundations crack and suffer water damage. My mom just had her foundation repaired. What risk is there that the electricity can discharge?

Like if it's as simple as running some inexpensive cables through the cement mix and plugging in a little box, awesome. But I suspect this is hardly the case, and I suspect the sheer volume of cement needed to be practical/useful is quite high for residential use.

It appears that the benefit of this is that the concrete can be used structurally as well as storing electricity

Batteries get noticeably hot when charging and discharging. I wonder how significant this effect will be for the concrete and if it will have a meaningful impact on its effective lifespan.

When I glanced at the paper in a previous posting, it mentioned cycle life on the order of 100,000 charge-discharge cycles IIRC. Saying that being a capacitor it doesn't have the limited lifetime of a battery.

Weight is simply not a significant factor for stationary large-scale energy storage. It's arguably weird to be using the same cells you'd put in a car as a "powerwall".

WRT gravity storage that's going to have moving parts and require regular maintenance like a hydroelectric dam.

I'm not sure what counts as a charge-discharge cycle, but if it happens once a day that's about 277 years. Significantly longer than any battery I've heard of, most of which are warranted in the single-digit years.

On the other hand, the cement manufacturing process has high carbon emissions.

So 45 cubic meters for 10KW/h means roughly 100 tons concrete. A Tesla power wall with a little more capacity has a weight of maybe 130kg. So roughly a factor of 1000 less capacity.

Thanks. That's the important number. Really hard to find in that overhyped article.

This is a generic problem with battery announcements. I want to see some publication like Electrek publish 1, 5 and 10 years ago in battery PR.

Did you read the article? Nobody is suggesting replacing batteries with this, but 50,000 square miles in the united states is dedicated to roads. What about concrete foundations? Think about how much energy we can store, which is desperately needed for renewable mass adoption.

That's the bullshit part of the article. Cement in an uncontrolled environment is not going to work as a capacitor, even if it works in a controlled environment. Some water will get in. There's expansion and contraction as temperature changes, which causes cracking. The road battery idea belongs in the same category as the road and sidewalk generator idea [1][2], which, in the end, was just an advertising stunt.[3]

The paper jumps directly from showing a tiny effect in a tiny sample to changing the world. This seems to be a disease of battery-related articles.

Also, cement is not the same as concrete. Concrete contains cement as a binder, but is mostly sand and rock. It also absorbs water, which is kind of a problem for a capacitor.

[1] https://www.sciencedirect.com/science/article/abs/pii/S03062...

[2] https://cait.rutgers.edu/generating-power-every-time-you-hit...

[3] https://www.archdaily.com/911965/sidewalks-that-generate-ene...

Touché. I guess I'm cautiously optimistic that it can scale (goodness we need good news), but you bring up plenty of valid points. Thanks for putting me in my place.

> Cement in an uncontrolled environment is not going to work as a capacitor, even if it works in a controlled environment. Some water will get in.

Concrete always has water. It's part of it's composition. What exactly led you to believe it doesn't?

Also, there's a technology that can turn concrete waterproof which is called paint. In harsh environments it's also mundate to coat rebar with epoxy paints to protect against corrosion.

> There's expansion and contraction as temperature changes, which causes cracking.

Thermal stress in monolithic structures does not cause cracking. At best, it can happen in hyper static structures when thermal strain imposes displacements beyond design limits. This is countered by using technologies such as prestress.

What can cause cracking in concrete is delamination from frosting, where water penetrating through pores expands and detaches concrete blades at the surface. This is prevented by applying a technology called paint.

I saw nothing in your comment that put into question the usage of concrete-based super capacitors. You pointed to known failure modes that are addressed in a very mundane way.

What’s the cost difference though? Could it be the foundation of your house? 40x20x2 feet is about 45 cubic meters.

Two feet seems very thick for a foundation. That’s probably thick enough to be an airport runway.

40x20’ is also rather small as far as American houses go

Usual foundations (though it varies in different parts of the USA) would be 8 inches around the perimeter with 6 inches in the "middle" part of the floor. Of course for commercial or multi-story buildings you would see an increase in the thickness.

My house is 80x32 with a six inch poured floor and 9’ poured walls 8” thick. If i punched in my numbers correctly that’s a little over 70 cubic meters.

Energy density is the word you're looking for. It's a important metric for many application (smartphones, EV, etc.) but not for everything: for instance when it comes to grid storage, the TCO trumps everything. And even on cost it doesn't necessarily make sense to compare innovations against the incubant, because lithium-based battery have had lots of R&D over several decades to make them “cheap”, and it's impossible to know in advance who's gonna win in the end: innovation necessarily involves multiple failures for every success.

Slight nitpick: if we're talking about kg then it's specific energy.

"Energy density" = energy per volume

"Specific energy" = energy per weight

They're usually quite correlated though, although confusing them makes hydrogen fuel a pain to talk about.

Thanks, didn't know that.

Yes, however since they are maintaining the cements structural ability consider that they could replace your houses foundation with a capacitor that could power your house all night long if your solar capacity could both charge it and run your house during the day.

Similarly would building facades providing power storage for lighting or other uses during the "off hours"

Build roads from such?

Shouldnt it just be an inherant regulated requirement to build roads from materials such as this - but require modular units so as upgrades happen...

Having it in the roads would be interesting but difficult to exploit. One idea I heard pitched was to put coils in the roads connected to nichrome (thermal) wire and then have trucks mount magnets under their trailers. As they drove down the road it would warm the pavement which would de-ice it in the winter. (make it hotter in the summer though, so not particularly practical without a switch).

What if you were to do the same, but with subterannian trains travelling in opposite directions (?) - such that the cars north are met with the subway south and they interact for great justice? (Im serious on the idea, but I dont know the outcome?)

(I am talking about huge magnets. if that wasnt clear.)

(Symbiotic asymetrical train propusion through shared interior corridor walls) (easy concept... hard execution)

Roads are tear and wear.

They are everything but not stable.

Also why solar freaking roadways garbage is.

Any large version of this capacitor has to consist of fine layers of material between electrodes and separators. Its structural ability will be very different from bulk cement.

That is precisely correct. The article went to lengths to describe how they maintain cement's compression strength which is the primary stress axis for house foundations. It would definitely make constructing a foundation more time consuming (not the pump concrete out of a truck and fill up the forms that is is today) but there isn't anything in the article or their paper that suggests this wouldn't be a reasonable application. (again, maybe not cost effective, but that is a different axis of trade-offs)

They don't mention = they weren't concerned. Yes normally it's primarily the vertical axis but you don't want these layers slipping sideways in case of light earthquake. Or water from outside getting in not only messing the electrolyte but also causing whole thing to swell. There are plenty of potential problems, needs much more r&d.

The voltage seems key here. From the article:

> they fabricated button-size capacitors capable of holding 1 volt ... As the next step, Ulm says the team is now focused on developing a 12-V supercapacitor using these materials.

If I remember my physics correctly, charge is proportional to voltage, so if they can achieve 12 volts max voltage, shouldn't it result in 12X energy density?

If so, that would reduce the required volume (for 10 kilowatt-hours of storage) from 45 to 3.75 cubic meters.

Apparently concrete slab foundations are typically 4-6 inches (10-15 cm) thick[1], so in a single-story 1000 square foot (92 m^2) house, the volume of the foundation would be 9.4 to 14.1 cubic meters.

Which means you could store around 25 to 35 kilowatt-hours in that foundation.

So it seems like it's in the right ballpark for what they're aiming for.

EDIT: Another article[2] says this:

> Having proved the principle, they now plan to build a series of larger versions, starting with ones about the size of a typical 12-volt car battery, then working up to a 45-cubic-meter version to demonstrate its ability to store a house-worth of power.

Given the main article is not very technical, and both articles mention 12 volts but in different ways, I'm now less sure what the researchers are aiming to do. Maybe they are trying to improve the max voltage of this type of capacitor or maybe not.

---

[1] https://wilsonexteriors.com/a-comprehensive-guide-to-concret...

[2] https://news.mit.edu/2023/mit-engineers-create-supercapacito...

Energy in a capacitor is (1/2)CV^2 - so 12x the voltage would be 144x the energy.

If they can keep C at the same value in a higher voltage version.

Yep. Trivially, one would expect the capacitance to decrease linearly with voltage.

There's also a question about current leakage.

Is the electrolyte concrete? Otherwise I don't see it working too well for building.

The Tesla Powerwall is (despite it's name) NOT a wall! From the sound of things, this new supercapacitor might be. I can very reasonably imagine there being many orders of magnitude more structure vs things that hang from that structure.

I also suspect there may be a slight cost difference between adding some soot to the concrete you were already planning on using vs whatever they put in a Powerwall.

I guess my point is there's probably usages for both.

It seems actually close to classical capacitors. And these are so cheap you can use them as bricks already :)

> They estimate that a 45 cubic meter sample of their supercapacitor could hold 10 kilowatt-hours of energy

In other words, 0.22Wh/L or 0.09Wh/kg. For comparison, the lower end for li-ion batteries is 250Wh/L and 100Wh/kg. Cute trick, but considering the carbon emissions inherent in concrete production this seems like a horrible idea - even if they manage to make them two or three orders of magnitude better.

It might be usable if you can essentially turn buildings into capacitors with zero additional change required, but as soon as it even remotely impacts the construction process it essentially instantly becomes nonviable.

Poured concrete almost always needs reinforcement, which is typically done with steel bars (rebar), which would presumably mess with the capacitance. Maybe making cinder blocks (breezeblocks) out of this would be more interesting

Depending on what's being built and local code, reinforcement can be done with fibers of various types mixed into the concrete. Anything in a commercial building needing to support heavy loads is probably still steel rebar.

This feels like another annoying fluff piece from the world's best tech hype institution. I'm sure MIT is a great school with amazing researches, but it seems like they get a paper out every few months that promises to "Revolutionize the whole world with this one simple trick!" To their credit, they get it right every once and a while and produce something really impressive (the Apollo Guidance Computer comes to mind). But more often than not, they're just really good at fluffing up their research.

> For instance, carbon black was used to pen the Dead Sea Scrolls in the 3rd century BCE. History aside, carbon black is also a highly conductive material.

Sounds like the perfect ink to write down circuit schematics.

What happens if one of these roads gets hit by lightning?

If you're standing on it when it happens I'm pretty sure you turn into a super hero.

you'll get sent back to the year 1955. Whatever you do, don't bump into your parents.

Two things I don't get from the articles: 1) How to make connections to the supercapacitor. 2) Does it work with rebar in the concrete?

Another HN post on cement+carbon as capacitor https://news.ycombinator.com/item?id=36951089

Reply on news.ycombinator.com