TDK claims solid state battery breakthrough
arstechnica.com
arstechnica.com
Older TDK story from 2020.[2]
TDK claims "a next-generation solid-state battery with an energy density of 1,000 Wh/L, approximately 100 times greater than the energy density of TDK’s conventional solid-state battery." Their "conventional solid state battery" is a tiny thing used in meat thermometers, a ceramic device with, like most ceramic devices, good high temperature tolerance.
Lithium-ion batteries are around 250–693 W⋅h/L. So this is maybe 2x existing lithium-ion technology. That's about what everybody else is claiming for next-generation solid state batteries.
Incidentally, gasoline is around 9,500 Wh/L, although only about half of that reaches the driveshaft.
End result: longer cell phone battery life, and an end to "bulging" battery failures.
[1] https://www.tdk.com/en/news_center/press/20240617_01.html
Isn't that more like around 35% on a good day? There are some pure ICE systems that can approach 50%, but those aren't in common passenger cars. Hybrid cars do considerably better, but even then 50% is an achievement.
More typical: 28% engine thermal efficiency, 75% transmission efficiency means 21% overall efficiency.
So yes, gasoline energy equivalency is pretty meaningless unless you multiply it by 0.2 first.
Gas mileage figures from the EPA and others don't account at all for the time a vehicle spends idling before/after a trip - or even in heavy traffic, just waiting at traffic lights. For example, time parents spend sitting in their cars idling waiting to pick up their kids, time spend idling in coffee and fast food drive-through lines, etc. Start-stop systems help, but a lot of people disable them.
The F1 hybrid design is pretty amazing. The motor–generator is driven by and drives the turbocharger. This allows the turbo to be optimized: When the turbo wants to spin too fast electricity is generated and when the turbo would otherwise not deliver enough pressure (lag) the motor–generator augments the turbo speed. Excess stored power goes to the drive train.
This effectively solves turbo charging, recovering waste heat through all operation phases, eliminating lag and delivering high efficiency. F1 chose not to field it, and I don't know why. I don't know if it will ever be seen in normal applications.
I've been exposed to a fair amount of different start-stop systems because every rental is different, and some of them are reasonable/good and some of them are awful. My main car is a PHEV, so I'm used to limited idling and that's fine, the big question is what triggers the start and how long does it take to start.
I've been in cars that consistently aren't ready to move when I've pressed the accelerator. Those get disabled everytime I start the car; hopefully I wouldn't buy a broken car like this, but I'd definitely figure out how to disable it if I did. The better ones will start when I reduce brake pressure, so by the time my foot gets to the accelerator, it's ready to move. Those get to stay enabled. In the middle, I might disable them sometimes when they're particularly bothersome, but otherwise leave them be.
I haven’t seen a modern car that let’s you actually disable this versus turn it off for the current ride only, fwiw.
Are manual transmission's only 80%?
Some places claim the most fuel efficient vehicles are about 40% (again, on a good day in testing conditions).
EVs on the other hand are up around 87% or higher.
You can make it a bit more efficient by optimizing an engine for a specific singular output, sometimes up to 50% I think is what nissan claims. One automaker, I forget who, was researching doing a hybrid drive train like the volt had, but where the engine could run in that one single speed and charge a battery.
Seems a lot more complicated than just electrifying the system, but I'm not a auto researcher.
Nissan's PR about it: https://www.nissan-global.com/EN/INNOVATION/TECHNOLOGY/ARCHI...
Not sure if this is what you're thinking of but Honda's eHEV platform drives the wheels with an electric motor and small battery†, with the engine kicking in as needed to generate current and to charge the battery (as well as regen).
†Until you get to high speeds, at which point a clutch engages and the engine drives the wheels directly, via a single fixed gear.
[1] https://en.wikipedia.org/wiki/Diesel–electric_powertrain
Fortunately.[1][2] British Rail tried in the 1950s. Four 12-cylinder Diesel engines driving the the wheels through three differentials and four hydraulic clutches. The engines were geared such that the speeds added. As speed increased, more engines were turned on. Since startup only used one engine, starting pull was rather low for the engine power installed.
Only one was ever built.
But if you drive like a leadfoot that doesn't really work cuz then usually the hybrid systems fall back on the gasoline engine for additional torque and it goes to the normal gas cycle when it does that and you lose the efficiency
Where the challenge comes for smaller vehicles, is the more power you are trying to give, the more you have to build everything up; not just the engine but whatever that engine is inputting into, also trains have somewhat more predictable speed profiles over a trip vs a car.
This is why some manufacturers have vehicles that are PHEVS but really BEVs with an 'oh crap' extender that can struggle to keep up a good grade.
I gotta give TRD/Toyota/Ford credit, the HSD style drive train does work around a lot of this stuff for PHEVS; in lockup mode you have a 1:1 ratio with minimal losses and at that point it's up to the ability of the engine to adjust profile.
Works very well in practice, and far simpler than the setups of others.
Automotive motor controllers now use similar technology, but the locomotives had it first.
Coal plants are about 33% efficient and natural gas plants are about 45%.
If you want to say that this factor doesn't count: In what sense should the oft-quoted factor in the final step count? (That is, the loss in converting from petrol to rotational motion in an ICE, or from electric potential to rotational motion in an EV.) I think the only real utility that number has is in estimating the total amount of stored energy in a typical car of each type -- this could be used to estimate the amount of damage that would be caused by the vehicle catching on fire.
Other claims strike me as meaningless.
We care about the efficiency of converting resources we have a limited amount of, sunlight is unlimited on humanity's scale, so is wind, but how much solar or wind power we can extract from them and transmit has constraints and that is the step where we should start caring about efficiency
https://en.m.wikipedia.org/wiki/Carnot%27s_theorem_(thermody...
Toyota and Idemitsu Kosan (part of the same METI grant that TDK got for SSB development 20 years ago) are going to commercialize Solid-State Batteries for EVs by 2028 [0][1]
The new Toyota Battery factory in NC is part of that push [2]
[0] - https://www.reuters.com/business/autos-transportation/toyota...
[1] - https://www.reuters.com/business/autos-transportation/toyota...
[2] - https://www.reuters.com/business/autos-transportation/toyota...
The problem is mostly driver mindset.
You just can't convince them that
a)most of their charging will happen at home while the car sits in their garage / driveway
b)When they do need to fast charge, the 20 minutes it takes for a number of current EVs to get to 80% charge isn't much longer than what you'd spend at a highway service area by the time you get done with fueling the car, going to the bathroom, chasing down everyone who was in the car, maybe buying a drink and snack, etc
c)For the rare occasion they need a vehicle with more range or are going into an area without good charging infrastructure, they can rent a car. This is how things are done in Europe - you take public transit most of the time, but for a trip where public transit isn't convenient, you rent - often times after taking a train to get closer to the area you're going to be in.
Drivers still buy giant 7-passenger SUVs and hulking pickups that spend most of their service life with one, maybe two people in them and little or no cargo.
Making car rentals much less of a hassle would help, as would mandating maximum passenger vehicle heights, and tax penalties on noncommercial vehicles over a certain weight.
This is a very different story around the world. Something like ~60%~ [it’s actually 35%, I got it backwards] of UK homes have no off-street parking. You'd need streetside chargers or built into lampposts, which is being trialled in some areas but is very small scale and almost certainly would not be as cheap as charging via your home's electricity.
> the 20 minutes it takes for a number of current EVs to get to 80% charge isn't much longer than what you'd spend at a highway service area by the time you get done with fueling the car, going to the bathroom, chasing down everyone who was in the car, maybe buying a drink and snack
The problem with this is that it assumes the car will always need a break at the same I do. That works if I can charge easily at home and will only need to fast charge on long journeys, but as I said above it's just not practical for many people. It's also best case scenario insofar as it assumes both that your car can charge quickly, and that you have close-by access to a fully functioning fast charger.
It's not a huge hurdle (and I do believe that by say 2034 this stuff won't be a concern at all), but the additional planning required is more than a lot of people are willing to consider, and I don't think they're wrong in thinking that. Especially because even now electric cars are still very much out of price range for lots of people (again, going by UK prices and salaries).
https://www.racfoundation.org/media-centre/cars-parked-23-ho...
> However, with 18 million (65%) of Britain’s 27.6 million households having – or with the potential to have – enough off-street parking to accommodate at least one car or van there is a huge opportunity for charging electric vehicles at home.
Granted, lack of parking is correlated with living in big cities, but I live in what was once primarily a mining, steelworks, and ceramics town; there is a hell of a lot of terraced housing where even parking close to your house is a nightmare, let alone charging.
Depending on where you live (speed limits, weather, driving habits, topology, etc...) impacts real-world range greatly. If you drive 80 mph on a flat freeway you might actually get 50-70% of the rated range. That brings a respectable 300 rated miles down to somewhere around 200 miles or less.
If the rated range was 400 for a cheaper vehicle and 600-700 for premium vehicles, almost all range issues would be solved overnight.
People buy cars thinking about the worst case, not the realistic or average one? What if I go on that trip across the US next year?
In Chicago (Pareto optimal for flattest/biggest metro area in the USA), we have two “going speeds” on the highway: 5 mph and 85+ mph. I drive fast (not going to go on the record here, but use your imagination - I mainly buy German cars that excel on the Autobahn) and I’m routinely only in the top quartile if I’m not in an actual hurry.
(I know EVs are much better in stop-and-go traffic than ICE but I can’t imagine 5mph with five-to-ten second bursts of 30 mph is great for the range, either.)
It is sort of sad to think the future of Apple Watch and AirPod is only getting at best 2x battery life, and that is assuming they dont make it even smaller.
So, that energy density is 100 times greater than whatever TDK's previous solid-state battery was, not necessarily 100 times other battery technologies.
Also, note that they're measuring the density in Wh/L, or Watt-hours per Liter. That is, they're measuring energy density by volume, not by weight. According to my Perplexity search, lithium-ion batteries have a "volumetric energy density ranging from 250 to 680 Wh/L". So these TDK solid state batteries will have maybe twice that energy density by volume.
That press release doesn't say anything about the weight of these batteries, which is probably why they're not proposing these for vehicles. If these were lighter than lithium-ion batteries, electric car makers would be all over them, looking for ways to get volume production up to lower costs. The fact that they don't mention the energy density by mass suggests that they're no better than lithium-ion.
So this is neat development, if not a major tectonic shift. The use cases TDK proposes, wireless earphones, hearing aids and smartwatches, are applications where size is a more important consideration than weight (below a certain threshold). Good for them! And if TDK can manufacture these cheaply and reliably, I'm sure engineers will come up with other clever uses for this technology.
EDIT: Hearing aids were the first electronic products with transistors, so that is a historically auspicious precedent. Asianometry did a video on transistors in hearing aids here: https://youtu.be/3ykz4JAO91g
If that works, then you could just equip cars with swappable Zinc-Air cells: you go to a "Zinc Air" station and in the time it takes to fill up a tank of gas, your car gets a new Zinc Air cell and you're good to go for 1500 miles(?).
Yes. 100x better density than current LiPo, LiMH, or something like that would immediately enable electric airliners. Like tomorrow.
> The ceramic material used by TDK means that larger-sized batteries would be more fragile, meaning the technical challenge of making batteries for cars or even smartphones will not be surmounted in the foreseeable future, according to the company.
Still, doubling an Apple Watch’s battery life or 1.5x but making it smaller would still be great.
Cars don't have one huge battery, and smartphones are starting to have multiple smaller ones to fit around the other components (and in folding phones).
You wouldn't need a huge array of coin cell batteries; one coin cell might well be enough.
[1] https://en.wikipedia.org/wiki/Static_random-access_memory#De...
From the TFA: "Solid-state batteries are safer, lighter and potentially cheaper and offer longer performance and faster charging than current batteries relying on liquid electrolytes."
The article to me is worse than click bait. They keep mentioning Apple supplier TDK as if this battery is currently being used in Apple devices. The only link this battery has to Apple is that they use other batteries from the company. So, yes, you're correct in that nothing indicates this as a direct replacement. It is shitty journo looking for relevancy in SEO
[0]: https://en.wikipedia.org/wiki/TDK#Sponsorship_and_advertisin... [1]: https://www.tdk.com/en/news_center/press/aah33300.html
Looking at their product directory, it's all B2B
The big open question is whether and how this can be scaled up to larger battery sizes in a safe and functional way. Still exciting to see!
It could still end up making a major difference.
> The battery technology is designed to be used in smaller-sized cells, replacing existing coin-shaped batteries found in watches and other small electronics.
> The ceramic material used by TDK means that larger-sized batteries would be more fragile, meaning the technical challenge of making batteries for cars or even smartphones will not be surmounted in the foreseeable future, according to the company.
Still extremely interesting.
This is not in production environment. We don't know how big these batteries can be, what temperature they can operate at, how much they cost to produce, if they can even be mass produced, what their output could be, etc...
Every week we have a claim like this one made by some reporter.
Yes. But not the converse.
“Kevin Shang, senior research analyst at Wood Mackenzie, a data and analytics firm, said that “unfavorable mechanical properties,” as well as the difficulty and cost of mass production, are challenges for moving the application of solid-state oxide-based batteries into smartphones.”
“The group plans to start shipping samples of its new battery prototype to clients from next year and hopes to be able to move into mass production after that.”
So they don’t know yet if they’ll be able to pull off mass production, and it’ll be at least a couple years off.
Breakthroughs like these are an important step. They are not nothing but they are not the end of the journey either.
a less blunt critique would've specifically identified why this breakthrough is similar to the last one that didn't make it past the lab
surely there's a large enough sample set of past failures to choose from since there are claims like this every week
For good reason, I think.
I imagine the plight of a poor tech reporter. The upside of real advances in battery tech are huge, so a scoop would get a lot of clicks. But in the absence of step-change inventions, we have incremental improvements in boring areas like “recharge cycles” which make terrible news articles. So our intrepid tech reporter must ignore all that real but unsexy stuff.
There is a source of zingy articles though: pr releases about unproven tech which are likely meant to pump up investment activity.
- https://news.ycombinator.com/item?id=40701402 (3 comments)
- https://news.ycombinator.com/item?id=40703316 (19 comments)
A laptop/phone that needs to be recharged every few weeks… heaven.
We eat up our new efficiencies the moment we invent them.
looks pretty compatible to me!
Of course it's also possible that being perpetually on guard for limits and risks is part of how we avoid them. The catastrophes don't happen because people thought they would and took actions such as investing in next generation energy R&D to try to avoid them.
It's probably some of both.
They market it for use-cases like RTC backup batteries and solar powered BLE beacons.
So, not 100 times greater than lion, but 2-4x (250–693 W⋅h/L [0]) is a lot.
Agreed. Unfortunately it seems like increases in battery density are soaked up by either increased energy consumption (screen resolution/brightness, processor) or decreased battery volume to make a thinner device.
I have a smartwatch that doesn't require wall charging (a Garmin Instinct with solar) and I really like the combination of solar cells, low-power processor, monochrome screen, and frugal radios.
That got me curious about a low-powered laptop and I've been eyeing an e-ink tablet with keyboard (e.g. Boox Tab or Remarkable 2) but I think it'd be a little too limited for my laptop usecases. And all the super-energy-efficient phones (e.g. e-ink) I've seen simply shrink the battery volume to offset any efficiency gains.
"Yeah you're not getting that, you'll just get a smaller battery that'll give you the same 6 hours and you're gonna like it."
- every laptop manufacturer ever
It’s a myth that energy density is a reason to worry about batteries or any other form of energy storage. How much energy is stored doesn’t say anything about how easy it is to release that energy in an uncontrolled way.
For Li-ion batteries it’s not really the electric energy stored you need to worry about at all. It’s the flammable electrolyte.
Solid state batteries are often very safe since they usually don’t have a flammable electrolyte. And when the electrolyte doesn’t burn it’s much harder to get a short and thermal runaway as well
The new tesla 4680 cells seem to be in the ballpark:
Cell volumetric energy density = 622 to 650 Wh/litre
Good to know.
Making these devices work charged for weeks, and a charging case for Airpods that can keep them charged for months (years?) would be extraordinary.
The subtitle includes “Apple supplier” as does first paragraph.
Edit: Holy shit yes they are https://product.tdk.com/en/index.html