Donut Lab alleged Solid State Battery launched at CES 2026

[Siwastaja]

One interesting thing to note is how all the discharge tests were only at 1C. Why?

One thing this effectively does is mask the round-trip efficiency to look better than it is. For example, if you calculate round-trip efficiency for the 11C test as 82%, you would quickly assume that oh, at 11C, that means 91% one-way efficiency (because 0.91*0.91 = 0.82). But discharge was not at 11C, but 1C only. So the efficiency for discharge was better, and for charge, it is worse, so the actual efficiency at 11C is closer to 82% than 91%. Maybe 85%.

Seeing that a thin cell needs a massive heatsink on at least one side, producing whopping ~170W (286A*4V*15%) of heat, means this 11C rating is not very practical.

It doesn't need to be, of course; EV market specifically does not need a cell capable of 11C.

Now the question is, how do the best existing commercial off-the-shelf high-power li-ion cells behave at 11C charge? I know these are routinely charged at 4C-5C no problem. And they behave very similarly to Donut's sample when tested at 5C. So could an existing cell be abused at 11C charge for a few cycles, to produce identical results?

[Siwastaja]

In the video there seem to be two different cells tested. The cell at 4:36 looks different than the cell at 5:01.

Looks like it's the same cell just turned upside down. See how + and - are swapped, and the small piece of kapton tape is in the same place but opposite side.

[rteodor]

In the video there seem to be two different cells tested. The cell at 4:36 looks different than the cell at 5:01.

Looks like it's the same cell just turned upside down. See how + and - are swapped, and the small piece of kapton tape is in the same place but opposite side.

Yes, that seems right. I guess I'm getting tired of the subject. Back to work.

[Siwastaja]

I am curious about the heat part of the charging. Is it getting hot because its charging fast?

Yes.

Is that heat (energy) loss linear across charging amps?

It usually is linear-ish, for first-order estimate; given constant R, losses are I^2 * R, so energy efficiency drops linearly with power. Then again R is not really a constant, rising temperature increases efficiency (until some point when it doesn't anymore).

If there's heat being lost during charging, how much joules in heat vs how much joules staying inside the device? If I can charge 150kWh in 1min but also waste about the same, 50%eff is not so great

That can be easily calculated from VTT's numbers which I already did above, so 82% round-trip efficiency for the combination of 11C charge and 1C discharge. That would probably mean something like 85% efficiency for the 11C charge and 97% for the 1C discharge, or so. So 82% of joules inside the device and 18% lost as heat.

Eyeballing from the results, the efficiency is OK but not exceptionally great, seems to match state-of-the-art high-power li-ion easily available. For an EV you would never really charge at 11C, you would do most charging at home at 0.1C and sometimes quick charge at maybe 3C-4C, for that 90% efficiency is fine, but needs some active thermal management (basically liquid cooling) like in any other EV.

It walks like a good li-ion cell, it quacks like a good li-ion cell...

[Siwastaja]

Quite interestingly, for a high-power lithium ion cell, it would be perfectly normal to see a test cycle like this:

1) Discharge at 11C
2) Charge at 1C

and see peak temperature of approx 90degC during that 11C discharge.

Now for Donut's sample, it seems they have simply reversed discharge and charge.

Li-ion behaves nearly symmetrically (if you allow damage), but li-ion is normally asymmetrically limited: charge power must be lower to prevent lithium plating. Charge temperature range is more limited, too; 90degC may be acceptable during discharge as maximum peak, but for charge the absolute limit would be e.g. 60-70degC.

Now the key question, which I probably repeat the third time, is what happens if you just take a usual li-ion cell and simply swap charge and discharge limits? The slower discharge then allows for an implicit cooling period and increases average efficiency during the round-trip cycle. At 11C charge, some lithium plating is going to happen. But it does not immediately lead to thermal runaway, just accelerated loss of capacity, which could be potentially dangerous if done 50 or 100 times. And some of that lithium plating could be reversible on the next discharge cycle, acting like a secondary lithium metal battery. (Emphasis on the word "some" - that's exactly why we don't do secondary li-metal batteries beyond lab tests, where they are fine for a few cycles). But if you only care about doing 10 cycles, then...

It would be quite easy to rule this sort of scam ("abuse standard li-ion cell") out by doing a cycle life test - I'm sure problems would start popping up well before cycle #500, which you could do, with these C rates, in just two weeks.

And no one really needs the 100 000 cycles. Show us the 400Wh/kg and 500 cycles at 11C, or even 5C charge and we will believe you.

[tszaboo]

if anybody notices the chart
there is a thermal lag time due to the thermal mass and distance to sensor
which means there is a even higher temperature at the core
maybe 100 ? 110 ?

Nice to see that they used the ACT0550 to test it, I was one of the designers of that equipment. Specifically the output driver, we called it the powerboard. Each current channels has a PT100/PT1000/ Thermocouple input, so they likely use that for the measurements.

The charge and discharge rate of a battery is determined by it's ESR. Because ESR will dissipate your power, and that has to go somewhere, or that energy will be used to damage your cell.
I would've liked to see ESR measurements, but we can estimate that from the published data.
At 26A charge discharge, the cell has 90Wh/110Wh = 81% efficiency, 20% losses. So it dissipates  20% of the input power in the ESR, 80% goes into the cell. The input power is 286A*~4V=1136W, and the test is 6 minutes long. So in 6 minutes it dissipates 20Wh, or 200W continuously. We can see why heatsinking is required. P=I*I*R => R = P/I*I=200W/284*284= 2.4 milliohm.
This is for a 26Ah cell. let's compare it with some 18650 cells:
https://lygte-info.dk/review/batteries2012/Intl-outdoor%20NCR18650BD%203200mAh%20%28Black%29%20UK.html

This is a 3Ah cell, you need ~8 of them in parallel to reach the same capacity, that pack would have ESR of ~1 mOhm. You can abuse those cells with 11C charging for a test, and it will produce similar charts. I don't see anything out of the ordinary, which solid state batteries would suggest.

In fact I'm going to say that anyone who say they came out with a solid state "cell" is doing something wrong.
At least a decade ago this was the wisdom.
The benefit of the internal construction of a solid state battery is that you can stack the cells just like in a lead acid battery. So you don't make individual cells, like with lithium. You save a lot on the housing. But you save a lot more on the battery formation, which is a lengthy process, that is expensive, and takes hours. It's much less expensive if you can connect 24V blocks to the machine, as opposed to 4V cells.
The savings were estimated 40-50% a decade ago, maybe the economics of this has changed a bit.
This is also where the weight and size and cost savings come from. That instead of individually housing 6 cells, you make a 6 times bigger block with the same thickness of housing. And less balancing, more cost saving.
So why would you even try to make a cell?

[mikeselectricstuff]

..not to mention the lack of any evidence that what's been tested is actually solid-state....

[Siwastaja]

actually solid-state....

That is a moot point anyway. "Solid state" isn't any sort of rigorous technical or scientific term. It's a marketing label only. "Actually sold-state" is thus meaningless. You can say that with tiny amount of liquid electrolyte within the pores of the plastic separator, possibly even in gel-like form, as in the standard li-ion cell, is "solid state" enough, if you just decide so.

Reputable companies do the exact same shit. When they realized in early 2000's that they can't make Lithium Polymer or LiPo work, they just started to apply the term on their standard pouch format li-ion offering; in big picture, this terminology "scam" did nothing else except create endless confusion, still to this day.

Solid state battery could be the next such term, with or without Donut Lab. But now that Donut Lab has this much press coverage, other li-ion manufacturers are definitely looking into starting to call their cells "solid state" as well, with nothing actually changing.

If Donut Lab actually has a working product, others need to answer to it by saying "we have solid state battery too". And if this is just a scam, others still have an incentive to say "they didn't have it, but we do, look".

[rteodor]

And no one really needs the 100 000 cycles. Show us the 400Wh/kg and 500 cycles at 11C, or even 5C charge and we will believe you.

Well that depends on what application one has in mind. I am on tho opposite side, I don't care about density (as long as its not significantly less than LFP) but I want those kilocycles and long lifetime. Build the battery together with the house with comparable lifetime for each. I don't want to replace it every 10...15 years or so. I will keep dreaming for now it seems.

[rteodor]

The benefit of the internal construction of a solid state battery is that you can stack the cells just like in a lead acid battery. So you don't make individual cells, like with lithium. You save a lot on the housing. But you save a lot more on the battery formation, which is a lengthy process, that is expensive, and takes hours. It's much less expensive if you can connect 24V blocks to the machine, as opposed to 4V cells.
The savings were estimated 40-50% a decade ago, maybe the economics of this has changed a bit.
This is also where the weight and size and cost savings come from. That instead of individually housing 6 cells, you make a 6 times bigger block with the same thickness of housing. And less balancing, more cost saving.
So why would you even try to make a cell?

The guy in the lab talks at 2:50 about "typical swelling and shrinking of the SSB ... which can be up to 15-20% of its size". Well apart from the fact that nobody has seen SSB batteries in the wild yet, I suppose this amount of swelling would significantly reduce the number of cells that can be packed together. Of course DL batteries do not have this problem which would amount to yet another brave claim. Gosh this battery must be the best ..

Joke aside, how would these SSB's  be used ? Would they need permanent spring (or other material) compression ?

[LogicDude]

In an earlier posting there is a mention of connecting lithium-ion batteries in parallel to configure a battery pack.  Is that doable for lithium-ion batteries?  I see people mentioning hooking batteries in parallel all the time on various YouTube clips.

I feel quite old-school because I wince whenever I hear about hooking batteries of any kind in parallel.  I don't really follow developments in batteries so I don't really know.

Whenever I see batteries hooked up in parallel I feel stress!  I imagine huge currents flowing as one battery attacks another battery and they both end up burning up in the fight.

[Siwastaja]

In an earlier posting there is a mention of connecting lithium-ion batteries in parallel to configure a battery pack.  Is that doable for lithium-ion batteries?

Of course. It's one of the most basic arrangements. It fundamentally does not differ much from having just the same amount of active materials inside one package.

Fundamental requirements are equal enough voltages during the initial connection to avoid excessive currents, and permanent, reliable connection so that the cells are permanently in parallel, acting exactly like one bigger cell. After that, current (and charge level) is shared between the cells just like in one bigger cell. Maybe, in some cases, even better.

[ejeffrey]

Well that depends on what application one has in mind. I am on tho opposite side, I don't care about density (as long as its not significantly less than LFP) but I want those kilocycles and long lifetime. Build the battery together with the house with comparable lifetime for each. I don't want to replace it every 10...15 years or so. I will keep dreaming for now it seems.

I would of course take extra cycles but if there is any kind of tradeoff it's rare that more cycles are the highest priority.  Todays LiFePo4 cells can last a long time of you treat them nicely.

But more to the point: anyone who actually needs 11C charging would almost certainly be willing to settle for 500 cycles or even fewer.  Just probably not 10 cycles.

[tszaboo]

The benefit of the internal construction of a solid state battery is that you can stack the cells just like in a lead acid battery. So you don't make individual cells, like with lithium. You save a lot on the housing. But you save a lot more on the battery formation, which is a lengthy process, that is expensive, and takes hours. It's much less expensive if you can connect 24V blocks to the machine, as opposed to 4V cells.
The savings were estimated 40-50% a decade ago, maybe the economics of this has changed a bit.
This is also where the weight and size and cost savings come from. That instead of individually housing 6 cells, you make a 6 times bigger block with the same thickness of housing. And less balancing, more cost saving.
So why would you even try to make a cell?

The guy in the lab talks at 2:50 about "typical swelling and shrinking of the SSB ... which can be up to 15-20% of its size". Well apart from the fact that nobody has seen SSB batteries in the wild yet, I suppose this amount of swelling would significantly reduce the number of cells that can be packed together. Of course DL batteries do not have this problem which would amount to yet another brave claim. Gosh this battery must be the best ..

Joke aside, how would these SSB's  be used ? Would they need permanent spring (or other material) compression ?

I'm not a chemist, but I think swelling is always bad. It's physical movement, the last thing you want to have with a long lasting durable object. There are two ways I know about. One is gas buildup, this is when you need to safely dispose of a battery, or be ready to call a fire brigade. The other is heat expansion. Like a LiFePO4 will swell because the current was high, it heated up, and it's airtight.
None of this sounds good to me, it sounds to me that you are at minimum compromising the lifetime of the battery with foolery, like charging it at 11C.

[booscrawl]

I'm not a chemist, but I think swelling is always bad. It's physical movement, the last thing you want to have with a long lasting durable object. There are two ways I know about. One is gas buildup, this is when you need to safely dispose of a battery, or be ready to call a fire brigade. The other is heat expansion. Like a LiFePO4 will swell because the current was high, it heated up, and it's airtight.
None of this sounds good to me, it sounds to me that you are at minimum compromising the lifetime of the battery with foolery, like charging it at 11C.

Basically all SSB expand and contract because the lithium metal is either on the anode (lots of volume of plated metal) when charged or intercalates into the cathode (negligible expansion because the cathode is so porous) when discharged.

[Siwastaja]

I'm not a chemist, but I think swelling is always bad. It's physical movement, the last thing you want to have with a long lasting durable object. There are two ways I know about. One is gas buildup, this is when you need to safely dispose of a battery, or be ready to call a fire brigade. The other is heat expansion. Like a LiFePO4 will swell because the current was high, it heated up, and it's airtight.
None of this sounds good to me, it sounds to me that you are at minimum compromising the lifetime of the battery with foolery, like charging it at 11C.

Basically all SSB expand and contract because the lithium metal is either on the anode (lots of volume of plated metal) when charged or intercalates into the cathode (negligible expansion because the cathode is so porous) when discharged.

And this is one of the most weird aspects in this terminology soup: it appears that many battery professionals / scientists equate:

solid state battery = lithium metal battery

which makes absolutely zero sense at all, but, I guess it's some sort of semi-official convention to some. But expect others not follow it!

Oh well. Lithium metal battery at least has a well-defined meaning! Lithium metal is one of the electrodes, just like in primary lithium batteries (non-rechargeable).

But solid state is like "low noise" - it refers to some fluffy feeling of "no moving parts", or in case of batteries, "no liquids". But it is quite ironical that the things classified as "solid state" batteries in labs exactly suffer from things moving! (Volume changes, making them not so solid.) Then again, similarly, solid state cooling devices (Peltier elements) also die out of mechanical failure, unlike the mechanical compressors which easily last 20 years.

[rteodor]

[...]

And this is one of the most weird aspects in this terminology soup: it appears that many battery professionals / scientists equate:

solid state battery = lithium metal battery

which makes absolutely zero sense at all, but, I guess it's some sort of semi-official convention to some. But expect others not follow it!

Oh well. Lithium metal battery at least has a well-defined meaning! Lithium metal is one of the electrodes, just like in primary lithium batteries (non-rechargeable).

But solid state is like "low noise" - it refers to some fluffy feeling of "no moving parts", or in case of batteries, "no liquids". But it is quite ironical that the things classified as "solid state" batteries in labs exactly suffer from things moving! (Volume changes, making them not so solid.) Then again, similarly, solid state cooling devices (Peltier elements) also die out of mechanical failure, unlike the mechanical compressors which easily last 20 years.

"Solid state" seems like a beloved attribute major technology shifts: first transistorized radios were "solid state". And there were other uses that I can not recall now but for sure other people on this forum will hop in to mention.

[Analog Kid]

Wouldn't it be amazing if all the sceptics were wrong and their claims were right. It'll be interesting to see how this plays out.

Steve, I hope you're not going all @Electrodynamic on us here!
I.e., don't let that pendulum swing too far away from its skeptical position.

[TimFox]

Solid, liquid, and gaseous states refer to bulk matter, not necessarily to mechanisms.
My automobile contains all three in their proper places.

[SteveThackery]

Wouldn't it be amazing if all the sceptics were wrong and their claims were right. It'll be interesting to see how this plays out.

Steve, I hope you're not going all @Electrodynamic on us here!
I.e., don't let that pendulum swing too far away from its skeptical position.

   Don't worry, my friend! That's the last thing I would do. No, I was just feeling whimsical and toying with the idea, and actually trying to think of an analogous circumstance where a newcomer has delivered a breakthrough that all the experts in the world said couldn't be done.

Scepticism has served the world of science brilliantly for centuries and I'm proud to be one of the many millions of torchbearers for the principles embodied in the scientific method. I think those principles didn't just underpin science; I think concepts like rationalism, logic, reason, open-minded-but-sceptical, burden of proof, and so on, have permeated into all aspects of society and politics to the enormous benefit of us all.

By casting off the trappings of religion, where "truth" was established by revelation and assertion and resided entirely with the powerful, we have unleashed the power of reason, and reasonableness. For 1600 years Britain was a largely Christian country, and after all that time we were still savages: beheading people, torturing people, enslaving people, and having people be hanged, drawn and quartered.

Then came the Scientific Revolution, followed by the Enlightenment, and look at us now: a secular democracy with the rule of law, sovereignty exercised through a democratic Parliament, freedom of religion and belief, legal equality of all persons, and freedom of expression.

So yes: I subscribe wholeheartedly to scientific and Enlightenment principles and values. Unfortunately significant parts of the world are still where the West was 500 years ago: where religion dominates, "truth" is the domain of the powerful, dogmatism rules over reason, and inequality and oppression is baked into the very fabric of society.

EDIT: Wuh-hey! Who would have thought solid state batteries could lead to a rant like that!

[oPossum]

tldr:  Ordinary Li-Ion cell. Estimated IR of 2 mohm.

[Analog Kid]

Wouldn't it be amazing if all the sceptics were wrong and their claims were right. It'll be interesting to see how this plays out.

Steve, I hope you're not going all @Electrodynamic on us here!
I.e., don't let that pendulum swing too far away from its skeptical position.

   Don't worry, my friend! That's the last thing I would do. No, I was just feeling whimsical and toying with the idea, and actually trying to think of an analogous circumstance where a newcomer has delivered a breakthrough that all the experts in the world said couldn't be done.

Well, if you (or anyone else here) can think of any such examples, it'd be interesting to read about them.
I suppose stranger things have happened ...

[tszaboo]

I'm not a chemist, but I think swelling is always bad. It's physical movement, the last thing you want to have with a long lasting durable object. There are two ways I know about. One is gas buildup, this is when you need to safely dispose of a battery, or be ready to call a fire brigade. The other is heat expansion. Like a LiFePO4 will swell because the current was high, it heated up, and it's airtight.
None of this sounds good to me, it sounds to me that you are at minimum compromising the lifetime of the battery with foolery, like charging it at 11C.

Basically all SSB expand and contract because the lithium metal is either on the anode (lots of volume of plated metal) when charged or intercalates into the cathode (negligible expansion because the cathode is so porous) when discharged.

Their main differentiator is solid electrolyte.
The anode or cathode can be whatever. There were many different types already showcased.

[Kalvin]

Seems like Donut Lab is running out of money, and they are now in the "fake it till you can make it" - mode, desperately trying to get new inverstors' money by performing these marketing stunts. Just like Theranos.

[LogicDude]

tldr:  Ordinary Li-Ion cell. Estimated IR of 2 mohm.

He made a great analysis clip.

Reading between the lines: Donut Lab has no battery cell and they just sourced a quality "white box" lithium-ion cell from a supplier in China and gave it to the testing lab to test.