Making Safer Lithium-Ion Batteries 77
itwbennett writes "Exploding iPhones may be a thing of the past. Researchers at Taiwan's Industrial Technology Research Institute have developed a new polymer, STOBA (that's self-terminated oligomers with hyper-branched architecture to you and me), that is added to the cathode material inside a lithium-ion battery to keep them from overheating. 'Fires or explosions in these batteries are caused by short circuits,' said Wu Hung-chun, a researcher at ITRI, explaining that even minor mishandling such as dropping the handset could result in damage causing a short circuit. 'The technology is ready for lithium-ion batteries used in electronic devices, mobile phones, laptops,' said Wu. And ITRI has started testing STOBA on electric car batteries."
Half the fun is in the danger (Score:2, Interesting)
A few years ago, you could crack open the older Lithium batteries and extract a ribbon of pure Lithium, which of course was fun to douse in water and other stuff to make explosions and other shenanigans. I don't necessarily want safe.
I also want the old liquid mercury thermostats and thermometers... mercury is fun to play with as long as you don't eat it.
Aww, no more fall down go boom? (Score:2, Interesting)
"drop phone watch it overheat" is the latter-day version of the halt and catch fire [wikipedia.org] "instruction" of days gone by.
Battery-related paranoia (Score:5, Interesting)
Step 1 (Score:3, Interesting)
Step 1: Stop manufacturing them in China
Nearly all Li-Ion battery failures (going out with a boom, pop, or fizzle) are a result of inferior materials being substituted in the manufacturing process.
Li-Ion battery cells (the individual cells containing your delicious electrons - millions in a single manufactured cell, several of those in a single packaged battery) are expected to pop.
Over-charge them? Pop.
Drain them too fast? Pop.
Result? Slightly diminished capacity.
Over time, the capacity gets lower and lower.
The trick is they're isolated, and you don't get enough of them popping at once to cause a noticeable failure (flame, explosion, etc.).
But when you have shitty charging circuitry, shitty components measuring and regulating the current and voltage, and shitty material (like fucking paper) inside the thing, yeah, shit's gonna go up in flames.
Ni-MH is the superior fucking choice. But the self-discharge rate is too high for the plebes to accept. They've got ones that sacrifice capacity for a lower self-discharge rate (such as Sanyo's Eneloop design), but Li-Ion is firmly entrenched, unfortunately.
Not about energy density (Score:5, Interesting)
I'll post this preemptively since usually when battery fires are discussed some people insist this is unavoidable if you want a high energy density, but this is not true. Whether batteries can fail catastrophically or not is mainly down to two things:
a)Whether the energy released when a cell fails is sufficient to cause nearby fails to fail, thereby causing a cascade of failed cells.
b)Whether the materials the battery is made of can react violently with materials it is likely to come into contact with when it does fail.
For traditional Lithium ion batteries the answer to both these questions is yes. The temperature necessary to cause a cell to fail is easily within the range of what is generated when a nearby cell fails. Furthermore the lithium batteries and their electrolyte burn quite well upon contact with air, adding even more energy to the reaction.
There's however no principal reason why this has to be the case. As an example if the heat capacity and conductivity of the battery is good enough it is possible to design batteries so that the failure of one cell won't heat nearby cells enough to cause them to fail. Different chemistries also have different activation energies, as an example lithium iron phosphate batteries are much safer for this reason. It is also quite plausible that one might be able to create a battery from a chemistry that doesn't react violently with oxygen.
Many batteries that use a water-based electrolytes qualify for both these criteria. Water has a high heat capacity and doesn't burn in oxygen. Unfortunately such batteries have other drawbacks. In particular while water itself won't burn it is susceptible to electrolysis at typical battery voltages, producing flammable hydrogen.
Why haven't we moved to lifepo4 yet? (Score:4, Interesting)
In the rc car world, the two major types of batteries in use are lipo (lithium polymer) and nimh. Nimh has less energy density, self discharge, and requires some rest after discharge still to retain full capacity when charged. I run nimh due to reasons I won't go into, but I have my eye on lifepo4, or lithium iron phosphate. They are not only more robust than traditional li cells, they go off in much the same fashion as the batteries mentioned in the article. The disadvantage to them, and why they aren't 'the thing' in rc cars is that they have a voltage disadvantage. Given the strict regulation of motors in spec class racing, a voltage disadvantage is a huge issue. In other applications, where you could pick whatever voltage and number of cells to use, these batteries are awesome. In rc, their voltage makes you pick arranging them in series at a voltage level that is a disadvantage or adding another cell and making yourself have a huge advantage -- ie, their acceptance isn't based on technical merit but existing standards in racing.
Re:Half the fun is in the danger (Score:3, Interesting)
Eh, rob a high school. I've never heard of a high school chemistry teacher without a good sized chunk of sodium in a jar in the classroom somewhere.
Re:Step 1 (Score:5, Interesting)
Actually, you're thinking of NiCad [wikipedia.org], which does suffer from memory effect and actually has a lower ampHr/kg ratio.
NiMH have no memory effect and can now come in a low-discharge (15% a YEAR) variant. They're pretty nice actually since they are several times cheaper than the equivalent li-ion. The Li-Ion/Li-Poly's come in handy since their ampHr/kg is much higher and therefore can fit in tiny spots when scaled against the larger Ni-MH. Nasty stuff to not charge a liion though.
Re:Step 1 (Score:2, Interesting)
But if they weren't made in China, they would cost twice as much. That is a non-starter.
That's not true. There are countries that have the Rule of Law and are not much more expensive than China. Taiwan or Malaysia being my favorites.
Plus a lot of electronics is not labor intensive if you do it right. Most of the labor intensiveness comes from people that make stuff badly so that the failure rate is high and then employ huge numbers of people to screen out the failures. Still if you get it right and source the components from people you can sue or at least not pay if they send you bad components this won't be necessary.
In fact I've met people who manufactured PCI cards in Australia. Basically once you've got it right you end up with a machine about the size of a photocopier that you feed in boards and components and it will apply solder paste, pick and place the components and then heat the board up. He reckoned said he had one in his office and it would occasionally beep when he needed to reload supplies. This is what the factory in China did. He'd designed the ASIC and bought the boards in. Since his design was stable and the components were sourced from reputable suppliers that cared about their brands, he wasn't screening.
In the China case it seems cheap at the start. But the workers are basically serfs who don't care about quality. The suppliers don't have brands so if they can sell you fake components they will. Even if they don't the quality levels will be terrible. So you end up with things coming off the assembly line with a very high failure rate. Then you employ a load of people to screen out the failures. Go to any Chinese factory and you'll see a few people running a production line and many, many more sorting. Basically they're sorting to find the minority of machines that don't have fatal defects. In a sane world the bugs would get fixed and the sorting would not be necessary. Still China isn't really set up for this - the factory provides a load of unskilled, underpaid labor to screen and the suppliers will get leaned on to provide more.
Re:Step 1 (Score:2, Interesting)
Re:Step 1 (Score:3, Interesting)
I know there are plenty of sites that claim that NiMH doesn't have a memory effect, but some manufacturers apparently claim otherwise in their data sheets. I found this discussion [electronicspoint.com] initiated by a man who was testing some Sanyo NiMH batteries; the Sanyo data sheet definitely did claim they had a memory effect, and his tests confirmed this. The effect is small but apparently measurable, and apparently also easy to undo with a normal discharge cycle.