I'm scanning the paper really quickly. I'm not a chemist but I do know a thing or two about batteries and the standard caveats apply here:
When they say 3x volumetric energy density, that is the actual energy density, which is energy per liter (normal density is mass per liter). Normally people use energy density to refer to energy per kg. Because this is a solid state battery, it is much denser than normal batteries (which are roughly as dense as water). Solid state batteries are smaller but much heavier and this is no exception. It is 33% the size of a lithium battery, but for the same energy it's about 2.5x heavier. Weight is still a much bigger problem for batteries than size- batteries are much smaller than the exhaust, engine and transmission of a car, but also much heavier.
The main limit on specific energy(kwh/kg) for this battery and for solid state batteries in general is voltage. Li-ion is 3.7v nominal, this battery is 2.5v nominal.
1,200 cycles may seem low, but it is actually very good; around 3x the life of current batteries. This cycle life is the time to degrade to 80% maximum storage, at a certain discharge depth and speed. Current batteries only last 300-400 cycles at their specs, but last tens of thousands at 30% depth of discharge.
Problem with the above: in this particular battery, the chemistry breaks down very strongly after it reaches the end of life. Normal lithium does this too, but not as strongly. This stuff may potentially last longer, but it fails much less gracefully. Not in a dangerous way, but in the same way as a normal car often does; once its broken it'll just work worse and worse until it is barely limping.
The temperature capabilities may seem irrelevant, but they are actually a decent problem for li-ion and are the reason lead acid is still used in cars.
Another interesting possibility for glass solid state lithium batteries is that recycling would be very easy. In organic batteries the electrolyte burns or reacts pretty much no matter what you do, but with glass you can plate and unplate cells. Unfortunately due to specific energy, polymer solid state electrolytes are much more likely than glass (also much cheaper).
Edit: IMPORTANT NOTE: this is NOT a fundamentally new type of li-ion battery! Solid state batteries have been around a while (glass, ceramic and polymer), and have specific advantages but low specific energy and power. This particular implementation is a bit higher power and possibly lower cost, but it's just a little blip of progress. Solid state batteries are a good candidate for the future, but they aren't there yet.
If this thing really has significantly higher volumetric energy efficiency than existing Li-ion batteries, then it's already good enough to be used in cellphones, at least according to me. I don't care if it's 2.5x heavier; I want more energy storage in my phone so it lasts longer between charges. I really, really don't care if that means the phone weighs an ounce more. I'm sure the Otterbox case on it adds just as much extra weight, and I don't see many people complaining about those.
Yeah, for cars, the weight is a big issue, because that hurts fuel (well, battery charge in this case) economy significantly, and EV cars use a LOT of energy storage compared to a phone. But for phones, I don't see the problem.
As for this battery failing harder, no problem: make sure the battery is easily replaceable by users, just like the Samsung phones up to the S5.
> I don't care if it's 2.5x heavier; I want more energy storage in my phone so it lasts longer between charges
I'm with you on this! But I'd clarify that I want more energy in any form. I would also not mind a phone that was a few millimeters thicker. But it seems the average consumer is swayed by a very thin phone, so more weight it is.
Its a super hard topic to get more energy into a battery of similar weight and dimension.
In relation to that - it would be super easy to create a User Defending App that guns down rogue energy using Background Apps (GPS Location, homecalls every minute, etc.) but hey - all hail the mighty data kraken getting his prioritys straight.
No battery maker will keep up, with the energy demands of the surveillance industries.
> Otterbox case on it adds just as much extra weight, and I don't see many people complaining about those
Irks me that not one single phone manufacturer dares to ship a product as indestructible as an iphone in an Otterbox. Sure it would be the clunkiest handset in the store, but compare it to other phones + case, there's room to make a superior final product. But no. Apple keeps shaving millimeters and Otterbox keeps slathering them back on.
You can definitely buy more rugged phones. Often they lag behind in specs compared to popular flagship models, but they're not bad. For instance, I currently use a CAT S60 phone: http://www.catphones.com/en-gb/phones/s60-smartphone
After a normal day of use I'll still have 60% of battery life. It's normally water resistant down to 6 feet for 30 minutes, flipping two switches makes it water resistant down to 15 feet for 30 minutes (and it even comes with a "drying" app that helps it to dry faster by, I think, vibrating the speakers.)
This particular phone comes with FLIR imaging, which probably isn't useful to the general population...
But a few have done what you suggest. At least slightly. The Galaxy s6 came in an active version that had like 30% more battery and better water resistance and a slightly tougher default shell.
I don't know how their sales were for the 2 models but I hope they were good enough to continue the trend
Yeah... I'm not sure what that is. They also conflate mA and mAh a few times. It's kind of confusing.
I think that they are using the weight of lithium metal rather than the actual battery. That's the only thing that makes sense, as the specific energy of PURE LITHIUM is barely above that 10.5 Wh/g figure. The 10.5 must come from the energy drop to sulfur intercalation, which is at .4v. The 8.5 Wh/g figure comes from the reversable voltage range. It's either that or their numbers are off by 100x.
Yes, but its not a useful measure. Most batteries will have similar specific energy when compared by lithium directly. How effectively the lithium is used is also almost completely irrelevant to the cost, weight and size of the battery. It's just a really weird thing to measure and they don't actually specify what the grams are of- lithium or sulfur are just the only things that make sense.
Yes, but on first read I thought that they were specifying for the battery using lithium metal as they were also talking about sodium and nonmetallic lithium. It's just so random I didn't read it correctly.
It seems like the biggest advance of this one is that the design allows the use of a metallic lithium anode rather than NMC/LMO/LFP etc. They say in the introduction:
"The organic-liquid electrolyte of the lithium-ion battery has an energy-gap window Eg ≈ 3 eV, but its LUMO is below the Fermi level of an alkali-metal anode, it is flammable, and it is not wet by an alkali-metal anode..."
The actual design is lithium metal plated directly onto lithium-doped glass, placed onto sulfur ink on copper as an electrode.
My translation is "The organic-liquid electrolyte will start to spark above 3eV, which is lower than the electron energy if lithium were directly touching it as a metal. Plus, hey, the electrolyte won't wet the metallic lithium anyway so contact is poor."
The glass has a much higher band gap, wets lithium (and sodium, which they also talk about) metal, diffuses sodium and lithium pretty fast, and manufacturability is super high since now you can use all kinds of standard metal working techniques rather than powder handling and sintering -- lithium even melts at 180C, so processing temperatures might be way lower and less costly.
In any case, in this context, the weight density is probably just normalized to anode mass. That would be better to explain clearly what they meant, but it seems fair since the energy density really is higher if you have more lithium in there and that's often the energy density limiting component.
Thank you for the awesome summary. I still think super capacitors for batteries would be a better future. If we need it to last longer we can carry solid state battery power bank.
The issue is how long does it take to charge as opposed to how long for most instances if phones only took 20 seconds to charge but lasted hours, especially if it was wireless it would be more practical.
Advantages of Super Capacitors:
Long Life - Supercapacitors have many advantages. For instance, they maintain a long cycle lifetime—they can be cycled hundreds of thousands times with minimal change in performance. A supercapacitor’s lifetime spans 10 to 20 years, and the capacity might reduce from 100% to 80% after 10 or so years.
- Temperature performance is also strong, delivering energy in temperatures as low as –40°C.
And then the TOO GOOD TO BE TRUE UCF story "You could charge your mobile phone in a few seconds and you wouldn't need to charge it again for over a week," says UCF postdoctoral associate Nitin Choudhary.
Anywhere you need a small but longer lasting battery and don't care about mass. The primary issue with a larger mass is accelerating and decelerating it. However with the larger capacity, it would be less efficient but perhaps could still lead to longer range in the car if the extra energy stored overcomes the added cost of moving it around. Probably not a good thing but could be useful then in something like an electric race car or something.
Submersibles (which may not care about weight due to ballast anyhow), Batteries for Solar Storage (price dependent likely bad at first), Pacemakers, Hearing aids, once it's invented if robotics outpace man engineered human biology, maybe smart blood.
Perhaps also Nanobots, Active 3d Glasses, Cell phones, Laptops, Tablets, Smart Watches that go for smaller space vs weight.
Outside of that perhaps home appliances like a portable bread machine, or a hand-held wireless electric apple peeler, very small things like a rock that has a hidden microphone / camera in it for spycraft, perhaps a heated travel mug to keep your cider warm, or an electric bowl for keeping great-gravy from coagulating on the dining room table, maybe a warm pie plate because who doesn't like warm cherry pie? Perhaps some sort of autonomous mud tunneling microbots, rather have a large boring machine just let these small guys go and be patient and in a decade or two you'll have your tunnel, electric candles in churches could cut down on unnecessary wax use, and helmet lamps used for stuff like lead mining could also find a use (added neck strain not withstanding). Not sure if the added weight would be too much for a target-practice duck that swam around in a lake or not, but if not, that might be the right answer.
I think that in many cases, smaller but heavier with more power is a good trade-off. Think cell phones, cameras, bike lights, camping lights, laptops, etc. while lightweight is a feature for all of these, they're also useless without a charge, so IMO the battery life and size together are more important than weight as long as it's still reasonable. Glass isn't light at all, yet they use that a ton in phones. The iPhone 6S Plus is really heavy and I haven't heard anyone complain about it. The efforts around consumer electronics I've seen in the past 20 years have mainly been about reducing size, with reducing weight as a nice typical side-effect.
This would be really great if it ever hits production, but given past battery progress vs what's actually available, I'm not holding my breath.
> The iPhone 6S Plus is really heavy and I haven't heard anyone complain about it
A significant amount of that weight is the battery, so making the battery 2.5x heavier would make the phone substantially heavier, even with the same overall charge. You'd be hard pressed to sell someone on 'This new phone we made is great, you get the same battery life as before but it weighs 40% more!'
I suppose the question is, in which cases would people be willing to accept a lower charge capacity if their phone could charge to full in, say, minutes rather than hours? I feel as though those USB battery packs that are everywhere these days would be a great candidate; plug it into the wall for a few minutes and you're fully charged, and you can charge anything else from there.
Are you sure that wasn't to dampen vibrations which might affect (or even themselves be caused by) spinning the disk and movements of the read head? (Same reason there's concrete blocks in the base of a washing machine)
> electric candles in churches could cut down on unnecessary wax us
I know your comment is a bit tongue in cheek, but an LED connected to an AA battery is enough to last a few days - IKEA sell a nice set that even flickers. I don't go to church, but I suspect the flame (fire) itself is symbolic.
I wish I had a more pithy comment with which to reply, but I'll have to make do with just saying that was just hysterical. I hope you meant it to be so.
>The primary issue with a larger mass is accelerating and decelerating it
It's actually friction, for electric cars. Regen is pretty effective at recapturing acceleration, so the energy lost in the wheels is the biggest factor.
> It's actually friction, for electric cars. Regen is pretty effective at recapturing acceleration, so the energy lost in the wheels is the biggest factor.
Mass counts much. It's not just about accelerating a still object (car) out of friction. But it's also about working against gravity (g). If the car's acceleration vector is at some angle with gravity (ie, when not perpendicular), more energy will be required to accelerate (when climbing slops), and to break (down the slope).
This is very much clear for aircrafts, and other flying objects.
Edit: Inertia is also a reason for energy loss in cars (and other moving objects), which increases with mass.
Regeneration recovers a certain percentage of any energy put into speed or altitude. No matter how long you spend climbing you get a good fraction of that energy back. In practice this is pretty quick, and even in stop and go you spend more time at around the same speed than accelerating and slowing.
Rolling friction loss is related to speed and is occurring constantly as you drive. That energy is just gone immediately, unlike inertia, which is stored. Say you spend 50% of your time accelerating and braking, and 50% keeping speed. If you regen 80% of the energy and accelerate with 2.5x the power you use to cruise, you're using twice as much energy on friction as accelerating.
The parent mentioned regen and that electric cars are good at it. That means: you don't accelerate infinitely. At some point you decelerate, and then you can regenerate the electric energy. However long you drive uphill, eventually comes the downhill of equal length and you regenerate.
I don't know how good exactly electric cars are here, but considering that a Tesla can work as a taxi in the city, it must be pretty good.
Except for the risks inherent in large quantities of mechanical inertia stored in a rotating mass, flywheels might seem an attractive option for rapid energy exchange.
Not really. Acceleration is related to torque so jerk is the rate of change of current in the motor. Current in an electric motor goes from 0-100% ~10k times per second (PWM), so the motor doesn't notice much difference.
>Current in an electric motor goes from 0-100% ~10k times per second (PWM)
That's just the voltage across the terminals. The reason why this works is precisely because current doesn't go from 0-100% with the voltage. PWM is relying on the inductance of the motor to keep the current the same as it would be if the average voltage was applied across the terminals.
I'm curious if these would work well in uninterruptible power supplies (whether under your desk or in the electrical room in your datacenter). As I understand, the potential explosive danger of Li-ion batteries makes them unfeasible for UPS applications due to the costs of meeting the hazmat regulations upon them, in addition to higher cost of lithium itself. If this solid state battery can be handled much safer than Li-ion batteries, and be made significantly cheaper, then perhaps we'll start seeing them in applications where LA batteries are usually used.
I'm no expert on this, but as a consumer, I wouldn't mind having a heavier phone if it meant a longer battery life. In fact, I much prefer a device with some heft.
It sounds like it would _not_ have longer battery life, in the sense of "how many days can I go in between having to charge"
It sounds like the battery can withstand more charge cycles before getting crappy and needing replacement (eg, your phone might take 4 years instead of 2 before you have to replace the battery) -- way less useful
If you get the same energy in 1/3 that space (but 2.5x the weight), you could have a phone that is the same size as today but with 3x the time on a single charge in exchange for a battery that is 7.5x heavier. Phones are light enough that I'd probably make that trade.
Why? It seems that phones are space constrained, not weight, and this proposed battery is 3x better by space but 2.5x worse by weight. That would allow a 1 day phone to go 3 days right?
Even "space constrained" is arguable, depending on the context. We've had powerful, feature-packed phones no taller/wider than the screen and thin enough to fit in a tight pocket for years, and that's all anyone reasonably needs. But developers and gadgetphiles fetishize thinness to the point that devs are now actually removing important features, like headphone jacks and (on Android) removable batteries, to squeeze out a few more useless millimeters.
As long as those priorities remain, any power-per-volume improvements will be twisted around to serve pointless shrinkage rather than actually improving anything.
Some of this relates to the size of the phone in a case which add a lot of bulk and seem overly popular. But, another factor is larger phones need to be very thin to be comfortable in many pockets.
Do you really think the audio jack was removed because of thinness?
Have you seen the inside of a smartphone? Its not just about thinness. There is very little space for anything besides the battery.
Also, removable batteries and the audio jack get in the way of dust/water proofing.
You may think shrinkage is pointless, but that's clearly not what the market or the top designers think. Why do you want a giant, useless, 100+ year old analog port on your 2017 device? I don't. Its ugly.
> Why do you want a giant, useless, 100+ year old analog port on your 2017 device? I don't. Its ugly.
Arguing about aesthetics is futile, but I want the port because I find it useful, and don't measure utility by invention-age. Seems similar to judging a beer by the bartender's height.
>Also, removable batteries and the audio jack get in the way of dust/water proofing.
My Samsung Galaxy S5 proves you wrong on this.
>Why do you want a giant, useless, 100+ year old analog port on your 2017 device?
3.5mm jacks aren't 100 years old. 1/4" ones are. And I want one because my $350 Sennheiser headphones use a 3.5mm jack, and Bluetooth headphones sound like crap and Beats headphones are utter garbage.
While i agree that the port is useful and i want a phone with one, your comment about Bluetooth headphones sounding crap is ridiculous.
The Bower and Wilkins ones are decent, the BW P7 Wireless is the best portable headphone i ever heard, the integrated bluetooth DAC is actually better than the DAC integrated in my iphone 5s and huawei p9 phones.
I had the sennheiser momentum 2 headphones before them, also different AKG Headphones and the BW7 Wireless beat them on all aspects when it comes to the sound, in Bluetooth mode (they also have an optional cable included - they sound better with the bluetooth dac tho!)
Maybe that's true, but my experience from the other end of the price scale is that cheap bluetooth headphones are absolutely awful compared to cheap wired headphones. I'd love to be proven wrong on that, as I don't like the wires, but even when I've had bluetooth headphones that sounded tolerable, switching back to even the cheapest wired headphones felt like a revelation.
Personally, I care just enough about the audio quality that this matters, but that I'm not willing to pay for an expensive pair like the BW P7 (more like 1/10th of that price...), and in my price range.
Bluetooth won't be a good replacement until they can compete on quality across that whole price range.
And the water faucets! And the drains! And the toilets! And the windows! And doorknobs! And doors! Heck who needs doors - we can just crawl in the windows.
Sometimes things invented 100 years ago (or 1000) are all we need. People don't evolve that fast.
> Do you really think the audio jack was removed because of thinness?
That and to shill overpriced, easy-to-lose Bluetooth earbuds, yeah.
> You may think shrinkage is pointless, but that's clearly not what the market or the top designers think. Why do you want a giant, useless, 100+ year old analog port on your 2017 device? I don't. Its ugly.
How on earth is a tiny, inconspicuous round hole uglier than a big dangly headphone adapter?
That aside, this is exactly the problem I'm talking about. The "top designers" have decided that phones are fashion accessories first and working tools second. I like good design as much as the next guy, but good design is functional. A pretty tool that doesn't work properly is useless.
That's fine, but that has nothing to do with whether or not the presence of a 3.5mm jack makes wires mandatory. Whether or not the quality of the bluetooth stack improves with the removal of the 3.5mm jack is highly debatable.
There is very little space for anything because of the insistence on making the phone thin. Without the thinness fetish there'd be plenty of space without sacrificing the jack. Add even 1mm, and e.g. the area covered by a battery with the same capacity is drastically reduced. Just look at some older (thicker) phone batteries for comparison to the pancake batteries in newer phones.
> Also, removable batteries and the audio jack get in the way of dust/water proofing.
The same method used to dust-proof/water proof the USB apply exactly as much to the jack. And they also apply just as well for the battery compartment - nobody cares about making the battery compartment dust proof, after all.
I can say water proofing is never something I've looked for in any phone beyond being able to handle the occasional splash. Even then, if you look at the inside of devices with removable batteries vs. devices where the batteries are not intended to be, the main difference is that in the former there is usually an inner shell covering the rest of the components - overall the internals of the phone tends to be better protected. If you want to water-proof that inner compartment, then worst case you get shorts. Again, this is about thinness and some extent cost-custting - dust proofing and water proofing is a total canard since you still have wires crossing the boundary, and two more sets is not going to make a major difference.
> You may think shrinkage is pointless, but that's clearly not what the market or the top designers think.
We don't know what the market thinks, because there are no high end devices that sacrifice thinness for e.g. battery capacity as none of the big brands dare to even try to be different.
I've just gotten a Umi phone that uses a 4000 mAh battery (vs. more typically 2400-2600 for most of the mid-range MTK based phones) and it's fantastic to have that extra battery capacity, but it's a Chinese phone with little presence in Europe/US.
Give it a few more years and we may see if any of the smaller brands manage to grow based on their larger battery capacity.
There are a few others - you can get a decent quality Android phone with 6000 mAh, and one with 10,000 mAh. The downside is that the latter was built with the assumption that anyone wanting that much care more about battery lifetime than anything else, and so it sacrifices the screen quality for a lower power one too, and makes various other sacrifices.
> Why do you want a giant, useless, 100+ year old analog port on your 2017 device? I don't. Its ugly.
I found this hysterical after your appeal to the authority of "the market or the top designers". Outside of Apple, pretty much only HTC have taken that leap. The vast majority of "the market or the top designers" so far still insist on keeping the headphone plug. Maybe that will change, but I note that on my commute now, pretty much everyone I see with a new iPhone also drag along ugly adapters so they can use their old headphones.
I'd like to wager there'll be all kinds of phones trying to discover what the customer wants. I'm guessing, for the majority, it'll be neither of the extremes - heavy with max battery, or light weight with min battery - but somewhere in between those extremes
It has higher energy density than Lithium, so theoretically yes it could increase the battery life of your phone. For equivalent energy the size is 33%, so one could fit 3x the energy capacity into a cell of the same size.
The real reason is that if you make several million phones that a certain percentage is always going to explode, either by manufacturing defect or a user pierces the cell through accident. This might mean the difference between some light leg burns or your death and phone makers are not going to take the negative press of the latter scenario. Their only answer is better power usage.
That would depend mainly on price. I imagine at this early stage pricing would be entirely speculative, but new technologies don't usually start out cheaper than established ones.
That's a matter of cost, not size. Mass is more important than volume for cost because transport depends on the weight primarily and because fundamentally more mass means more stuff. In the end you still have to pull x tonnes out of the ground.
Maybe electric bikes and mopeds/scooters? Space seems to be a bigger issue for those than cars. And in the early days (and in some cases today) lead acid batteries were used. Lead acid had similar weight issues, but are much worse in terms of size and power, so seems like if folks were able to make that work, then a bike or scooter with this solid state battery could work.
I went on amazon to look at some product weights. There is an e-bike battery that weighs 6.5 lbs[1], and an e-bike that claims to have a shipping weight of 60 lbs[2]. So assuming 3x battery weight for roughly the same power, that means an extra battery weight of 13 lbs and a total bike weight of 73 lbs for a battery that could possibly fit in the frame (as opposed to be mounted on the down tube). So that is definitely getting on the heavy side for a bike but the bike in question is also relatively 'over-powered' as far as e-bikes go (25 mi self-powered range, 350 watt motor). A lot of the sleeker/higher end e-bikes are 'pedal-assist only' and already hide the battery in the frame. These bikes can get in the sub 40 lb range, and probably use 1/2 the battery that the one in the link does. So an extra 10-15 lbs is probably manageable and it would result in a bit more range while still being able to hide the battery in the frame (assuming a li-ion battery weight of ~3lbs, you would add an extra 6lbs to 'break even', or an extra 15 to double the power).
For scooters/mopeds, the math probably works out even better, as these are already over 100 lbs, and an extra 30-40 lbs won't really matter that much. And if you want to build a electric scooter or motorcycle that can go highway speeds, that extra weight can help with stability.
It's not; it has practical consequences. Everything in a phone runs at a spectacularly low voltage compared to most electronics because the battery only has one cell. At small sizes it's simply not practical to make more cells.
Smart phones typically run lower than 3.7V so it's still not a big deal. The weight is a possible show stopper for cars but probably not smart phones (where size trumps weight) and of course they may be able to get the weight down by some kind of chemistry or materials magic (glass foam?)
Phones turn off automatically if the battery dips below 3 V, and this battery tops out at 2.7 V. It's totally possible to make it work, it just isn't terribly simple.
As for chemistry and glass foam, the voltage of a cell is determined by the electronegativity of the reaction in the battery. With lithium, it can never be very much above 5 V- atoms just don't pull on each other that hard. This already uses lithium, and the chemistry won't be able to improve much bast 2.5 V.
For any given chemistry, you can't increase the voltage without putting cells in series, because the atoms decide the voltage. Any tweaks to construction, electrolyte etc. can only increase the current, by giving it more places to flow.
"It's totally possible to make it work, it just isn't terribly simple."
Please, it's as simple as a three-component boost converter - one IC and two resistors. Same thing I use to bump 1.2V up to 3.6Vto run LEDs. It takes zero space (7x7mm total.)
That's 4 mm x 4 mm (more with passives, but not too bad), and hits a solid 95% efficiency at the normal load of an iphone, and is above 90% all peak loads. Heat dissipation will be fine, all parts cost about $1.20. Low EMI.
However, its still a dozen new components. It's a ~10% tradeoff somewhere and while that may be totally worth it, its not simple. Phones do not have square millimeters to spare.
Its like... adding a turbocharger to a car. Yes, it is possible. There are even kits, after a fashion. But it is still not simple.
Also, its not even at the stage of a hypothetical question yet. This battery would last about two weeks in an iphone under these load conditions. The only reason it "looks" possible is if you're considering a battery that is not this one.
If it's a true boost converter, it's also going to need an inductor somewhere. If it's really just driving an LED, it could be a charge pump, which uses capacitors instead. But charge pumps probably can't handle the current level that a phone requires.
"If it's a true boost converter, it's also going to need an inductor somewhere."
We use resistors as inductors. Plenty of wound resistors in teeny-tiny packages exist, that's how we have metal detectors with credit card-sized boards, the literal bulk of the unit is the frame and coil assembly and board housing + adjustment controls.
Older gas charge pumps can handle it, but they're bulky and they are lossy and generate a lot of heat.
Modern boost converters have efficiency from 80% to 95%. They also operate at high frequencies – 1MHz or even 3MHz, which allows to make inductors incredibly small. Boost converters aren't a problem for a long time now.
Remember that '3.7V' nominal cells range from 4.2V down to a low voltage cutoff about 2.8V (and in practice will not be let too close to that DoD). So stacking two of these '2.5V' cells will probably give a 6-4V usable range.
Remember also that the electronics is in three groups: 1.8V (most of the processor), 3.3V (peripherals), and raw battery voltage (RF front end output power amplifier). The former two will already be driven by switchmode regulators and the latter is much more flexible.
Laptops, maybe? The voltage is too low for phones, as is the power. That's why solid state batteries aren't used yet, they (non-research) can't match the power that phones and computers require.
If I can use a small corner of my garage for batteries rather than half of it, that's an advantage. We can always use more power, especially when the sun provides it for free.
> The main limit on specific energy(kw/kg) for this battery and for solid state batteries in general is voltage. Li-ion is 3.7v nominal, this battery is 2.5v nominal.
Forgive my limited electronics knowledge, but would something simple like throwing a boost converter on the end of this not be sufficient for many applications? Is the maximum current output too low to reach sufficient voltages too?
Op's point is that the lower nominal voltage is the culprit in the battery's lower energy density. If you and I each have a 1 kg battery that stores 1 amp hour, but yours puts out 3.7v and mine puts out 2.5, then your battery is storing more total energy than mine. We can both manipulate that output voltage (via boost converter, or more commonly just using multiple batteries in serial) to get whatever we need, but given the same load, my battery's just going to drain faster than yours.
You may be leaping to conclusions here about the energy density. I have seen no info that says the battery is 2.5 times heavier for a given energy capacity. The article from UT only said that the volumetric energy was higher. There is no information about the specific energy (wh/kg). Throughout the rest of the article they only refer to energy density without specifying Wh/liter or Wh/kg. I think it is a huge leap to deduce that this technology will be 2.5 times heavier for a given energy density.That does not make sense. A glass electrolyte will be very thin and is not likely to add much weight to the battery. The weight is much more likely to be dependent on the electrode materials and the weight of any necessary containment materials. Some battery chemistries require steel plates compressing every cell to prevent electrode expansion.
Unfortunately solid state batteries can't provide enough power for a smartphone. I would absolutely buy a phone that browsed slower for longer battery life + the ability to work below freezing though.
The solid state battery does not have enough power to meet average power demands. The phone would work for a short amount of time, then shut off until the small battery charged again.
Thank you for spotting the trick. The article says:
> The researchers demonstrated that their new battery cells have at least three times as much energy density as today’s lithium-ion batteries. A battery cell’s energy density gives an electric vehicle its driving range, so a higher energy density means that a car can drive more miles between charges.
I believe this is misleading enough that the article should be flagkilled, since the "energy density" they are talking about, volumetric energy density, is not what gives an electric vehicle its driving range, and on the metric that really matters - mass energy density - it does worse than regular Li-Ion batteries.
I think part of the confusion comes from the paper doing all of its energy and power comparisons to the mass of pure lithium in the battery- that leads to a lot of numbers being 10-15 times what they should be. Reading the paper is kind of confusing because of it. They also have a couple things that appear to be switched up and finally it doesn't help when they say things like this:
>Replacement of a host insertion compound as cathode by a redox center for plating an alkali-metal cathode provides a safe, low-cost, all-solid-state cell with a huge capacity giving a large energy density and a long cycle life suitable for powering an all-electric road vehicle or for storing electric power from wind or solar energy.
> finally it doesn't help when they say things like this:
> "Replacement of a host insertion compound as cathode by a redox center for plating an alkali-metal cathode provides a safe, low-cost, all-solid-state cell with a huge capacity giving a large energy density and a long cycle life suitable for powering an all-electric road vehicle or for storing electric power from wind or solar energy."
I don't understand. That was surprisingly easy to understand, compared to the bulk of HN comments.
Why do you say that if it's 3x the energy density per volume and "only" 2.5x heavier? That should give it at the same volume as Li-Ion 82.5% of the weight.
I agree this doesn't solve the weight problem for EVs, except only a little it seems, but if these batteries can be made competitive on price with Li-Ion (Eventually), and unless we find a much better alternative weight-wise, then this sort of battery would still be very useful for EVs because of the safety they give.
"batteries are much smaller than the exhaust, engine and transmission of a car, but also much heavier."
Excuse me, have you tried lifting an actual engine, lately? I want you to find me a car battery that weighs more than even the engine block alone, no pistons, cams, heads, nada.
My vehicle curb weight is 4,374 lbs with engine installed and 2,686 without it. You can subtract another 280 for the transmission for a non-mobile weight of 2,406.
With the additional equipment I've got atop the engine, the vehicle breaks 4,500.
He is referring to the battery packs used to drive EVs. Those are in the range of 100kg to past 400kg. Auto manufacturers have been desperately trying to shed weight from cars. it is one reason why most are slowly moving if at all into pure EV and still moving on hydrogen or other fuel cell.
plus many small engines stripped of power accessories can be carried on carry on luggage
Wait...is this type of battery at all related to EDLC supercapacitors? Those have the same upper limit of 2.5-2.7V, which seems like an odd coincidence.
The greatest travesty in the world right now is there doesn't exist a multi-government 10xManhattan sized project to develop a really really good battery.
It would enable a huge reduction in CO2 emissions(bye gasoline), allow developing countries have stable electrical sources and low cost renewable distributed generation, cut costs(good includes cheap), and enable renewable energy sources to make up a larger mix of our energy production.
Always exciting read about new developments, wish I knew more about chemistry/physics, hopefully we get there sometime soon!
People don't have great ideas on what they'd do with new monies that would come a 10xManhattan size project. Tesla, VW, Toyota, Dept of Energy, DARPA, NSF, Energy Companies, Industrial Conglomerates like GE, Panasonic, etc. all work on battery research. (& many more)
The actual Manhattan project could spend the funds by building reactors with desirable end-products (uranium etc.) and building a lot of centrifuges to refine it further - along with engineering teams to design, build, and operate these things, various bomb and delivery mechanisms etc. It was the very beginnings of nuclear science, so there were many fundamental ideas to be discovered and tested out - it meant sense to spend a lot of money to travel and explore this newly discovered branch of the tree of knowledge.
Efforts are underway. Unfortunately, it's a tedious slog.
There's a book by Steve LeVine, The Powerhouse, that describes the current environment of battery research and some of the main players (Argonne national lab, DoE, Startups, Korea, Japan, etc.). (It features Goodenough.)
GP's post is oversimplistic, but I think there's something there. Maybe it's better stated as, "Governments tend toward invention of things that, in the long term, are destructive and divisive rather than peaceful and collaborative."
If "technology" is defined (or construed) to more closely follow successful human evolution (which I think is reasonable), then weapons of mass destructive are a differently (albeit obviously related) phenomenon and not "technology."
It depends on how you view the age-old political question of means and ends. If the latter grows out of the former, and if governments, as monopoly-holders on initiation of violence, use force as their means, then naturally we can expect things that come from government to grow toward those ends.
Silicon Valley itself wouldn't have existed in the form it does without the huge government investments in the area, so even if that happened - and you can speculate all you want but facts remain they didn't -, it'd still be a win for government R&D funding.
Actually, you could not be more wrong. If you read The Entrepreneurial State from Mariana Mazzucato[1]. It argues that most inventions are coming from public funding. Companies are the ones that customize to the users need and then monetize it.
He's also well known in solid-state physics for the "Goodenough rules" [1] (more commonly called the Goodenough-Kanamori rules, but that has less of a ring to it).
What's really astounding is reading that, and looking at the dates. He first formulated this rule in 1955. And he's still working and making real contributions!
We really need to work on greatly extending human lifespans, if for no other reason than to keep people like this alive.
On the other hand, an engineer needs to know when to say "it's good enough" and release it. An invention in the hands of a perfectionist may never be released.
The press release says "the researchers’ cells have demonstrated more than 1,200 cycles with low cell resistance." That's nice, but rising cell resistance is just one way the battery might cycle poorly over time.
Skimming the actual paper, I don't think they demonstrate 1200 cycles for any parameter. Eyeballing the graphs, it looks like they did charge/discharge testing for maybe 1200 hours (Figure 3a), but with really slow charge/discharge cycles of 10 hours each. Anyone publishing in this space would highlight 1200 cycles of stable cycling in the actual paper, if they had data to demonstrate it. They'd also show off faster cycles if high-rate performance looked good.
Looking at the data the authors did not present in the actual paper, I'm guessing that this battery doesn't handle rapid charge/discharge cycles well. (Their cycling test is at only 0.1C, paper claims "acceptable charge/discharge rates" but does not further quantify it... implication is "not a strength of this design.") It may not have great capacity retention either. I don't see any graphics specifically highlighting capacity retention vs. cycles. So at present I'd call this a solid research effort, but even if it could be commercialized immediately it's not clear that it would be a winner. The demonstrated charge/discharge rate is too slow to be practical for EVs or portable electronics. The demonstrated cycling stability is too low to be attractive for grid tied storage.
People who found this paper interesting may also be interested in this related publication about solid state sodium ion batteries that Goodenough was also involved with: https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5269650/
I really wish they'd publish more of the limitations as well in the paper. I wonder how much of it got cut out in editing.
Publishing limitations and negative results that lead to dead ends can help other scientists when they attempt to replicate and improve on tech like this.
I agree that publishing negative results would help the whole enterprise of science, but the prisoner's dilemma is pretty clear. I would guess that the authors have been "trained" well enough by publication incentives that they didn't bother to highlight limitations even in their original manuscript. That kind of information shows up in the paper only by absences and implication; it's only informal chats with peers where people talk honestly and thoroughly about the negative results and limitations of their approaches. It's a shame that scientific publication ended up this way. It's one of the reasons I left research to write software.
I have put a note in my calendar a year from now, and a year after that to read the story about how these laboratory curiosities could not be made in production quantities.
Something I would really love to see would be a solid state battery that exploited the fact that we can draw silicon features at 7nm. How about a couple trillion equivalents of a FLASH cell which we can drain in rows or fill in rows. Sort of a bucket brigade of capacitors at that point but it would not have any recharge issues and since its just charge flying about no dendrites to speak of. At some point I predict that will become a useful way to build energy storage devices.
It is more battery breakthrough fatigue. Last time I looked there was something like 17 - 18 battery "breakthroughs" over the last 8 years of which exactly one made it into production for a relatively small (10%) improvement in charging performance.
These laboratory curiosities usually fail because either you can't reliably manufacture the precursor material or reliably make the structures needed at production rates (which translates into the cost of the resulting battery)
It came up in a discussion with some Tesla enthusiasts on the ground breaking of the Gigafactory where I wondered if there would be a break through that would make the batteries the Gigafactory would produce obsolete before they finished building the factory. That didn't happen :-).
As a result I recognize its a really hard slow slog through chemistry which is well understood to change batteries. And you're correct there have been a number of improvements in manufacturing which have resulted in incrementally better batteries and that is all good, but so far, starting with a basic battery and making a new battery that is significantly better, has been disappointing.
A comment that I think is relevant in a community of startups that often, explicitly or de-facto practice age discrimination:
This engineer is 94 years old. Few, if any, SCV startups would have hired him. Yet, here we are, he might have just developed the key technology to push electric transportation through the hockey stick curve past the inflection point.
Three times the energy density, many times more cycles, no shorting, high charge and discharge rates. Yeah, this is more than just about laptops and cars, this is about planes, boats, trucks and ships.
Think about that before you reject a 50+ year old. Experience has real value.
I don't want to take anything away from Dr. Goodenough, but there is a slight caveat with respect to academia. Older, established, and respected researchers are able to attract large teams of of highly talented people. They are also able to secure funding. It is usual that this leader is given top billing on all papers and press interactions, even if they are mostly coordinating the work. This is good for everybody because the lab can continue to attract good talent and especially it can attract funding. Brilliant young researchers working by themselves in obscurity can't really make much headway because they have neither the man power, nor the money to tackle big problems.
So, while I agree with you in principle, older researchers offer more in academia than just their experience.
> Braga began developing solid-glass electrolytes with colleagues while she was at the University of Porto in Portugal.
Slightly orthogonal to the content: this is why it's important for the US to allow people to come to this country. Continuing research and development with the worlds best gives the US a leg up. Past ease of immigration is the reason the US has led in so many areas, constant flow of new and innovative ideas.
This is why it's important for the US to allow scientists and engineers to come to the US. Immigration is not an all or nothing thing. You can pick and choose who you let in.
Literally no one is arguing to the contrary. The problems people have with immigration into the US typically involve the sheer scale of it, the fact that many people who come to the United States do so illegally, difficulty with integrating, and difference in values from native citizens. Nobody is arguing against small-scale immigration of highly skilled scientists, there aren't even enough of them to make a demographic difference.
I'd say the current stories about people enforcing poorly written rules at the border prove you wrong.
People say yes to scientists, but not if they're from certain countries. On top of that we have scientists from Europe being stopped at the border, making it unlikely that they or their compatriots may try to make the trip in the future. I know of at least one computer conference that has discussed moving the venue outside the US to try and mitigate the issues at the border.
While you may be right that no one debates elite people in their respective fields from moving or coming to the US, the enforcement is causing exactly the opposite to occur.
So, I respectfully disagree with you that what you suggest would be acceptable.
A few days ago some Indian engineer was shot dead because he "looked Iranian". There is a very strong political reason why people like this killer do what they do.
Perhaps I should have clarified: by "literally nobody" I meant "virtually nobody", as in "the number of people who believe that we should accept no immigrants whatsoever is very small and statistically negligible".
Interesting footnote: "The UT Austin Office of Technology Commercialization is actively negotiating license agreements with multiple companies engaged in a variety of battery-related industry segments."
I recommend the book "The Powerhouse: America, China, and the Great Battery War"[1], it goes into some detail about the legal issues around (one) li-ion design and Thackary's work with/without Goodenough and attribution.
Basically, the battery has a lithium (Li), sodium (Na), or potassium (K) metallic anode, a solid glass electrolyte, and an "ink" cathode of sulfur (S8), ferrocene (Fe(C5H5)), or manganese oxide (MnO2) on a copper (Cu) current collector.
So it's not entirely solid. The cathode may have a tiny bit of liquid electrolyte between the glass and the copper.
I would guess that microfractures in the glass would impede the movement of charge carriers, even if it didn't shatter outright.
It wouldn't be good for the battery to hit it with a hammer, I think, but that really applies to any high-energy-density battery. Not all glasses are created equal, either. It all depends on the composition and the cooling regime. Some glasses can shrug off small-caliber bullets, and others will crumble under your fingertips.
I agree the battery density is the main story here, but you also cannot help but be inspired by the mans age. Just today ironically enough, I was questioning if my age might have meant I was not as sharp as I was, and then this story pops up with a man over double my age, still very active and at the top of his field. Cannot help, but be inspired.
Wouldn't it be great if this could get to market. Almost sounds to good to be true. Maybe I missed it but are there any working prototypes out there in a vehicle or some kind of device or is this just in a lab?
You would think there would be folks all over this if it were the case. I wonder what is holding it up?
It's taken 25 years for Li-ion batteries to get from the lab to the current level of industrial production. If this technology pans out it will probably take a similar time to do the same - maybe a bit faster given the greater interest in batteries now.
I imagine (hope!) that the much larger industry and greater need for large scale commodity batteries (electric cars) can speed this up considerably, and not just a bit.
This may be a silly question to ask, but forgive me, I dont know much about EE or battery technology:
Its my understanding that current batteries found in mobile phones, laptops, etc make use of rare earth minerals which are limited and expensive and only available from big players like China. Does anybody know if this technology also makes use of rare earth minerals?
NiMH batteries used lanthanum, but we don't use NiMH any more. Li-ion batteries use heavy metals, specifically cobalt, of which China is the largest buyer. Almost all cobalt comes from Congo though.
Cobalt is widely available everywhere, as are rare earths. Congo just happens to have a freakishly high amount of it plus cheap everything, so it is widely mined there. Unlike other technology mining easily transfers and it doesn't take much time (a few years) or money to open new mines. This battery doesn't use cobalt at all, and most newer batteries use significantly less cobalt or none at all (LiFePO4). Phones and batteries still use a lot of cobalt, and are major consumers, but the biggest consumer is machine tools.
There's already a Wikipedia article about solid state batteries. I found it interesting to read that in such batteries, ions are still allowed to move around.
Hmm, the article mentioned nothing about the weight. For a medium sized quadcopter, 3 times more energy (given the same weight) means as long as 3 hours of flight time which would make all these drone delivery dream come true.
Unfortunately, as hwillis explains in another comment (https://news.ycombinator.com/item?id=13779143), it's not 3x energy for the same weight, it's 3x energy for the same size. It seems to actually be about 2.5x heavier for the same energy.
No, solid-state usually refers to a component that undergoes no mechanical or chemical change while in normal operation; conventional batteries undergo redox reactions while charging and discharging.
Turns out that occasionally an inventor appears who gains more lifetime experience than over 99.99 percent of their "peers".
This always requires more decades than most people expect.
And some of these technology creators are the kind that can go from nothing to something in only a couple years, not unlike the way a software developer can sometimes go from learning basics to an incredibly effective breakthrough app in just a couple years themselves.
Then you give them decades and it's amazing.
With some of them it's so natural that they get better every year, until there is no way anyone else can compare in only just a single decade or two.
These are the operators who if properly capitalized can create the most jobs.
TL, DR:
This tech utilizes solid-glass electrolytes vs today's liquid electrolytes. This apparently brings us:
"at least three times as much energy density as today’s lithium-ion batteries"
"greater number of charging and discharging cycles, which equates to longer-lasting batteries, as well as a faster rate of recharge (minutes rather than hours)."
The team lead is Dr. Goodenough, making things better!
(sorry, couldn't resist. that's really his name)
If true, that first point especially sounds awesome!
These new discoveries and inventions are years away from market, but to say, that they never come to market is just not true. Why do you think our phones are getting thinner with longer battery life? They are certainly coming to market, but to introduce something this new in your consumer grade products would increase price by likely $1000's, to very few people you'll need to build plants, and production lines.
Yes, but not to the extent that the hyped tech over the years would have you believe. It's hard to tell how much improvement is due to new tech vs. simple incremental improvements.
"Faster rate of recharge (minutes rather than hours)" sounds like a big win for some applications.
It's such a shame that these batteries are so heavy, because IMHO fast recharge is one of the things still holding back electric car adoption. Even the Tesla fast-charge stations take 30-40 min to reach 80% charge, while a gasoline-powered car can hit 100% in only a minute or two.
(Though OTOH, I've always been surprised that fleet vehicles like city buses can't surmount that issue with modular battery backs that can be physically swapped out at the terminal stop and then charged offline before swapping back into another bus...)
I borrowed and electric car for a week, charging overnight on a 15 amp circuit. And occasionally on a middle and top end circuits as a test. Charge times come into play when you dont own your home or easily have a way to run power to your car. That is it. I don't think is the biggest blocker, cost is the blocker.
I guess it depends on how hard it is to swap out a battery but on a vehicle like a bus that can probably be made very simple. Now that I think about it a bus must carry a large quantity of fuel, I wonder how long it takes to fill one up. Swapping a battery could actually be faster.
Depends on battery charging rates but in general, a swap will be faster at large energy densities. BAIC in China are currently experimenting with this in taxis and buses.
I don't understand why so many people are ignoring that it uses sodium instead of lithium. Even if it was WORSE on metrics it would still be better due to the cost advantage. Amazing.
I was watching a Nova episode on batteries and they showed a new type of Lithium Metal battery that used a plastic material that separated the positive and negative halves. The unique thing about this new Lithium Metal battery was that it was not prone to the existing problems of Lithium Metal batteries and did not catch fire when punctured. They even started cutting portions of the battery off with a scissor and it continued powering an iPad.
Yup, same principle. The problem all solid state batteries share is that the solid electrolyte isn't a good conductor- plastic isn't great for that. Despite that it's the current favorite for reasons of cost. NB that the plastic is very very special.
I don't want to diminish what this guy has done, but I hope all the other people involved get the credit they deserve.
The sole inventor is for super hero movies. In reality, there are teams, sometimes just 6 ~ 12, but sometimes dozens of engineers and scientists, grad and undergrad students, that work together to figure out all the little parts to make technology like this work and make it viable.
It's never just the first author, but the next six or so authors on a paper that helped make a project what it is.
All that being said, it's still cool to see research institutions working to make stuff like this.
I don't understand, the first 2 lines of the first two paragraphs:
>>>A team of engineers led by 94-year-old John Goodenough, professor in the Cockrell School of Engineering
>>>Goodenough’s latest breakthrough, completed with Cockrell School senior research fellow Maria Helena Braga
Further down:
>>>Braga began developing solid-glass electrolytes with colleagues while she was at the University of Porto in Portugal. About two years ago, she began collaborating with Goodenough and researcher Andrew J. Murchison at UT Austin. Braga said that Goodenough brought an understanding of the composition and properties of the solid-glass electrolytes that resulted in a new version of the electrolytes that is now patented through the UT Austin Office of Technology Commercialization.
I understand the sentiment, but I don't think it was needed in the context of the article. And while you don't want to dimish from what that guy has done, in a way you diminish the message of the article by singling out a name that the article did not.
Hope this doesn't come across as overly antagonistic, just thought it was odd considering the article seemed pretty good.
Racism does not get any better when you use fancy words. Your exact line of thoughts went through the killer that shot that Indian engineer. The difference is that he was less eloquent.
I don't think it's right to equate expressing unpopular opinions with murder, no matter what you believe. If he's wrong, prove it, there's no need to resort to such ad-hominem attacks.
I did not equate actions. I equated thoughts. Two people think the same thing; one chooses to curse, the other chooses to murder. Trump has not killed anyone himself, but his supporters have. Are you telling me there is no link between their thought processes?
Even today, my physics lab has multiple computers running Windows XP or 2000. It's the easiest way to keep running old and sometimes proprietary software. We keep them off the network, of course.
(Coincidentally, our work overlaps with John Goodenough's. He's a legend, and impressively prolific even in old age.)
It's not just the software, it's also the instruments themselves.
The most worthwhile function of Windows turns out to be its backward compatibility.
This became most apparent after Microsoft leveraged their software engineering leadership by pivoting into an anti-recycling company to make most of their money.
There are incredible numbers of irreplaceable laboratory instruments made years ago which depend completely on the version of Windows that was "supported" when the instrument was issued.
Since Microsoft has failed at their truly most worthwhile function, it is now necessary to put incredible effort into preserving older versions of Windows and even more difficult to preserve the older PC's. These cheap office machines, which were built to spend as little useful time as possible before progressing into their primary target duty as landfill, as electronics go were built especially crummy, just for you.
Anyway, I'll be restoring a mass spectrometer shortly to operate as designed on Windows 3.1.
Yes, it will cost a few thousand for the PC effort alone, but we're not going to spend 100x that amount to get a new(er) spectrometer just so it will operate using a more modern Windows version like XP or W7.
There is also a great possibility that a brand new instrument compatible with Windows 10 will be completely useless in less than 20 years due to this Windows deficiency alone.
I noticed that too. Also expected 'but we keep them off the net and it's secret software and we cannot change things' posts. I wonder when history will stop repeating itself.
It's not that we can't change things, is that doing so sometimes costs $10s of thousands of dollars. We often have instruments running on XP, that work completely fine, but the software will not run on 7, 8, or 10. When we contact the company to ask about an update they say "sure, but you'll need a software maintenance contract, with a re-certification fee, etc. etc, for $40k." No faculty member wants to pay $40k to move something working to a new OS when we can put it on our unrouted internal network and keep it running on XP for $0.
When they say 3x volumetric energy density, that is the actual energy density, which is energy per liter (normal density is mass per liter). Normally people use energy density to refer to energy per kg. Because this is a solid state battery, it is much denser than normal batteries (which are roughly as dense as water). Solid state batteries are smaller but much heavier and this is no exception. It is 33% the size of a lithium battery, but for the same energy it's about 2.5x heavier. Weight is still a much bigger problem for batteries than size- batteries are much smaller than the exhaust, engine and transmission of a car, but also much heavier.
The main limit on specific energy(kwh/kg) for this battery and for solid state batteries in general is voltage. Li-ion is 3.7v nominal, this battery is 2.5v nominal.
1,200 cycles may seem low, but it is actually very good; around 3x the life of current batteries. This cycle life is the time to degrade to 80% maximum storage, at a certain discharge depth and speed. Current batteries only last 300-400 cycles at their specs, but last tens of thousands at 30% depth of discharge.
Problem with the above: in this particular battery, the chemistry breaks down very strongly after it reaches the end of life. Normal lithium does this too, but not as strongly. This stuff may potentially last longer, but it fails much less gracefully. Not in a dangerous way, but in the same way as a normal car often does; once its broken it'll just work worse and worse until it is barely limping.
The temperature capabilities may seem irrelevant, but they are actually a decent problem for li-ion and are the reason lead acid is still used in cars.
Another interesting possibility for glass solid state lithium batteries is that recycling would be very easy. In organic batteries the electrolyte burns or reacts pretty much no matter what you do, but with glass you can plate and unplate cells. Unfortunately due to specific energy, polymer solid state electrolytes are much more likely than glass (also much cheaper).
Edit: IMPORTANT NOTE: this is NOT a fundamentally new type of li-ion battery! Solid state batteries have been around a while (glass, ceramic and polymer), and have specific advantages but low specific energy and power. This particular implementation is a bit higher power and possibly lower cost, but it's just a little blip of progress. Solid state batteries are a good candidate for the future, but they aren't there yet.