Lithium-air batteries: Their time has come | Hacker News Reader

Lithium-air batteries: Their time has come

economist.com

134 pointsmdani87 comments

Lithium-air potentially has great energy density. The theoretical limit is near that of gasoline. So there's been a lot of work in this area, going back decades. There are lots of problems. This is a possible solution to one of about six major problems with this chemistry. It's a step forward, though.

This research project used iridium, which is a problem. Total world production of iridium is only about 10 metric tons a year, and it sells for about $14,000/kg. A production technology is going to need something cheaper.

Do you have a reference to the usage of iridium?

"Iridium is one of the nine least abundant stable elements in Earth's crust, having an average mass fraction of 0.001 ppm in crustal rock; gold is 40 times more abundant, platinum is 10 times more abundant, and silver and mercury are 80 times more abundant." - https://en.wikipedia.org/wiki/Iridium#cite_note-greenwood-5

Silver is only double as abundant as Gold?

But 20x less demand dating back thousands of years. Weird huh?

There was a great article I read about this question a few years ago, I couldn't find it though.

Basically the question was why gold and not silver, platinum, etc., as the desireable metal for commerce/jewelry. The answer lies in golds uniqueish properties, it is rare (not the rarest), it is unique in color (platinum is hard to distinguish from other metals by sight), where gold is confused for other things or to determine its quality a simple bite test will determine its quality, without needing to melt it down (unlike silver colored metals), it's soft enough to be easily formed into jewelry.

Not an article, but here's a podcast that goes into that: http://www.npr.org/sections/money/2011/02/07/131363098/the-t...

(multiple estimates since they don't exactly line up squarely)

...

[PDF] http://minerals.usgs.gov/minerals/pubs/commodity/platinum/my...

On page 4, it says 2014 global consumption of Ir was 6100kg. (3rd last para).

...

https://en.wikipedia.org/wiki/Iridium

Wikipedia article para 2 says global consumption and production is three tonnes.

...

http://www.bloomberg.com/news/articles/2014-02-19/iridium-cl...

This (2014) article estimates global production as 129,000 ounces to 322,000 ounces, which is about 3500 kg to 9000 kg.

Sorry, I meant usage of iridium for this battery. The article only mentions cobalt.

Oh, in that case, I don't see any mention of it either, or in the paper, http://li.mit.edu/Archive/Papers/16/Zhu16KushimaNatureEnergy...

I know iridium is sometimes used in spark plugs.

http://densoautoparts.com/iridium-power-spark-plug-configura...

You're thinking of the Argonne research from Jan, that used iridium to create a closed lithium air battery. They used the iridium to prevent lithium peroxide from forming and clogging the pores of the electrode.

http://www.anl.gov/articles/stable-superoxide-opens-door-new...

This research from MIT is entirely different. It doesn't use iridium at all.

It's actually quite interesting how they solved this problem (in comparison to Argonne's approach). To quote the press release:

The secret to the new formulation is creating minuscule particles, at the nanometer scale (billionths of a meter), which contain both the lithium and the oxygen in the form of a glass, confined tightly within a matrix of cobalt oxide. The researchers refer to these particles as nanolithia. In this form, the transitions between LiO2, Li2O2, and Li2O can take place entirely inside the solid material, he says. The nanolithia particles would normally be very unstable, so the researchers embedded them within the cobalt oxide matrix, a sponge-like material with pores just a few nanometers across. The matrix stabilizes the particles and also acts as a catalyst for their transformations.

http://news.mit.edu/2016/new-lithium-oxygen-battery-greatly-...

http://li.mit.edu/Archive/Papers/16/Zhu16KushimaNatureEnergy...

Also this battery isn't Lithium-air in spite of the the Economist having that as a title.

Apparently, "lithium-air" is a term for lithium-oxygen batteries, not solely those relying on ambient oxygen. While the MIT news writeup only refers to other batteries using "lithium-air", the ANL press release refers to it as "a lithium-air battery". [1] [2]

So while lithium-oxygen is possibly more correct, both seem to be valid usage.

[1] - https://news.mit.edu/2016/new-lithium-oxygen-battery-greatly...

[2] - http://www.anl.gov/cnm/articles/lithium-air-battery-based-li...

The ANL only has "air" in the title "A Lithium-Air Battery Based on Lithium Superoxide" which you could read as an abbreviation of "a Lithium-Air like battery".

Right. This doesn't need iridium, and may be more cost-effective.

The MIT paper has more useful info. The materials used were only lithium, cobalt, iron, and potassium. All of those are abundant. They claim the battery is stable during overcharging, and only loses 1-2% of capacity in 130 cycles.

If this is real, it's a big deal.

Talk to me when you can make 10,000 of them and sell them at a profit at only 10% more than equivalently sized LiPo chemistries.

Seriously battery breakthroughs have a reputation of being nice tricks in the lab that are infeasible in production. So before I get excited I really want to see a mass quantity of batteries made.

All that aside, I think finding a way to finesse the reaction one proton at a time like cells do would be a bigger win. I want a battery that runs on ATP.

You are a battery that runs on ATP

Combined with a form of fusion, we have all the energy we would ever need

"Thermodynamics doesn't work that way!"

"Where did you learn physics, Neo?"

http://hpmor.com/chapter/64

No, YOU are a battery that runs on ATP.

Wait, what are we doing?

Soylent green

There's a reason slavery was* one of the most energy efficient tecnologies, until it was banned on obvious ethical grounds.

* although "Estimates of the number of slaves today range from around 21 million-29 million to 46 million." https://en.wikipedia.org/wiki/Contemporary_slavery

Battery, solar and windpower have a very bad ratio of technology announcements:products.

They seem to be doing better than nuclear fusion...

Not to mention perpetual motion...

In this forum, we obey the laws of thermodynamics!

Are you joking?

Batteries have improved like 100 times, solar like 10 times and windpower like 1000 times over time.

I have an amazing Ultrafire flashlight with lithium battery that last three months since last time I charged it. I use it every single day.

Solar tech are getting 40, 50% efficiency. The first one I got had like 2%.

Near my house in central Europe there is a 3MegaWatts wind turbine.

It looks people have difficulties understanding exponential growth. If some technology improves 10% each year like batteries have done and it looks more or less the same, in reality it makes enormous change over time.

I'm not denying there are real breakthroughs. But I am noting that for every 100 breakthrough press releases only a very very small number eventually make it to market. And that's besides the outright scams (solar panels made from human hair?).

This is certainly true of wind turbines where you see designs that lack an elemetary understanding of the physics involved. The real development happens at an industry level.

Yep. I've done a fair bit of community service on a renewable energy forum and the number of designs that are just laughably wrong is depressing. A number of them were clearly investor scams (windbelt, windtree).

https://en.wikipedia.org/wiki/Windbelt

http://www.dailymail.co.uk/sciencetech/article-2855155/Now-e...

Solar Roadways!!!

(Google it for some of the most entertaining debunking videos and articles)

Probably not happening within a decade, if ever.

"Most researches working in the field of Li-O2 batteries do not consider this technology as a good investment opportunity."

- https://en.wikipedia.org/wiki/Lithium%E2%80%93air_battery#Fu...

"The possibility of buying off the shelf Li–air batteries within 10–20 years does not seem realistic at the moment."

- Balaish, Kraytsberg et al. 2014

"Dr Li hopes, within a year, to turn the prototype into something that might be manufactured. This is an ambitious goal but Dr Hardwick agrees that, from an engineering perspective, the challenges are similar to conventional lithium-ion batteries, so rapid development is possible."

Here's to hoping they succeed!

Yeah, it looks promising, I want to believe this.

> Dr Li

Nominative determinism in action!

When I participated in the Local Motors LITECAR challenge I believe a good 50% of the entries said "USE GRAPHENE!" as a premise for their entries. Sure, graphene shows merit in a lab in perfect conditions, but when I suggested using organic based composites to rival aluminum for chassis &/or body panels / components I got jumped on for being impractical. Yes, I'll admit I'm still sore about the whole thing.

Sometimes those competitions (hackathons, science fairs, and grant proposals too) are more about picking the most popular buzzwords than they are about actual science and engineering. Next time just let a Markov chain write your submission for you.

Oh I feel you, I really put a lot of time and effort into watching and participating in all the Q&A sessions and what was claimed were not defining factors ended up being totally shot when the results came in. Places and prizes went, suspiciously IMO, to college teams and questionable rubric evaluations. I mentioned the body panels of my submission but it was 1 of 3 innovative approaches to make a holistic concept...as they required...

To wit: They said "just replacing the panels / bodywork for a vehicle is not innovative in the way we are evaluating things" and then one of the top 3 prize winners was just that - a type of sandwich material that was only beneficial if it replaced panels and bodywork. I'm still sore because I hate being lied to especially when there's money at stake. In hindsight the whole process was flawed but that don't change what was said versus what was done.

They might try to argue but I studied their TOC & shit so extensively that when they already had a fully developed winner announcement video that debuted on the date they said they would notify winners, I smelled something rotten. I can't prove anything but if I had forensic level access to their communications I certainly wouldn't hesitate to invest my time in finding out what the fuck actually happened.

I have also wondered why Duroplast (stamped panels made up of leftover cotton and resin) never made it outside of the Trabant to the wider automotive world.

If anything, the bodies lasted too long.

Whoa, I'd never even heard of that! This kind of learning is why I keep coming back and talking honest. Thanks for mentioning that.

There's something to be said for understanding manufacturer profit objectives, but in the context of the contest reaching for something "iffy" like what won didn't make sense when a reasonable alternative could show statistically significant benefits.

Basically I pointed out we could "grow our own" panels with further research and the judges picked an unproven but Uni backed submission instead. Made me feel like Galileo for a hot minute there.

The price of new technology can be much higher than the conventional alternative. Products, where the higher energy density battery allows for the manufacturer to make the money back through higher volumes or price, will start switching over.

That's a really high standard. Let's settle for making 10,000 and worry about the price later. There are plenty of people that will pay extra for batteries with better energy density.

Absolutely. I'm not sure how much higher density these get, but if it's something like 2x that'd be enough to warrant maybe a 100x price or more - for certain applications.

Apparently current experiments are at 5x if Wikipedia is to be believed. 5x! For something trivial like an ultrabook that's going from 8 hours to 40 hours of battery life! From workday to work week.

https://en.m.wikipedia.org/wiki/Lithium%E2%80%93air_battery

It is a standard I've heard from lots of people, it goes "How do I know if my breakthrough will be successful?" Answer, "When it has twice the performance and can be sold for only 10% more."

It speaks to the question of moving someone from a known good technology to an unknown technology. I believe the first place I heard it was in a talk by Geoffrey Moore on evaluating startups but since then I've heard it repeated again and again. What is more I've seen it in action (or inaction) where companies of "better" products which didn't move the price/performance needle enough, didn't get enough traction to succeed.

Sounds very promising. There is more detail in the MIT News article:

http://news.mit.edu/2016/new-lithium-oxygen-battery-greatly-...

Sounds quite stable:

>“With a typical battery, if you overcharge it, it can cause irreversible structural damage or even explode,” Li says. But with the nanolithia battery, “we have overcharged the battery for 15 days, to a hundred times its capacity, but there was no damage at all.”

>In cycling tests, a lab version of the new battery was put through 120 charging-discharging cycles, and showed less than a 2 percent loss of capacity, indicating that such batteries could have a long useful lifetime.

> Overall, the new battery system is “very scalable, cheap, and much safer” than lithium-air batteries, Li says.

There are more dangerous combinations of chemicals than lithium superoxide mixed intimately with lithium, but there aren't many.

Is it worse than gasoline? If so, by how much?

My guess is: gasoline doesn't actually contain oxygen which means it needs it from external source to explode. Here you have everything prepackaged by desing.

*design

Which means we are on the right track towards standard science fiction "power packs," which notoriously double as improvised bombs.

With 'safety limiters' that you can reprogram or disable in about 5 seconds with no other tools.

The place to watch re battery tech, in my opinion, is the RC hobby scene. Eg hobby king will sell you, for $10, a lithium graphene battery which will pump out 60C (eg for a 2 amp-hour battery, it'll give you 120A!). It's incredible whats out there for the non-risk-averse hobbyists.

http://www.rcgroups.com/forums/showthread.php?t=2592234

maybe it's easier to innovate in the rc space due to some factors. such as hobbyist takes warnings and limitations seriously (e.g. charging with specialized chargers and use "blast-proof" charging bags), and sizes are quite standardised for use. having that much caution for mass-produced item usage is quite hard.

Interestingly, NASA just announced Li-Air batteries are one of its long shot research programs:

http://www.nasa.gov/aero/five-new-ideas-to-be-explored-by-na...

Am I the only one who noticed that the author's name is Li, which is also the chemical symbol for lithium?

Finding a way to so dramatically reduce energy loss during recharges and also retain capacity so well over 100+ cycles is pretty significant. I was not expecting to read about that level of advance.

> The new battery, brainchild of Ju Li of the Massachusetts Institute of Technology, is some way from commercialisation, but its design is such that commercialising it should not be hard.

Well, here's hoping. I always feel skeptical of technologies that just can't leave the lab, though.

So LiO2 has only been isolated at temperatures up to 40K, says Wikipedia. Oxygen is a solid at that temperature. Is this prototype also at 40K?

Article explains that they don't use pure oxygen molecules. Basically two solids oscillating between themselves.

My point of bringing up the melting point of oxygen is that if the LiO2 is only stable at temperatures where O2 is solid and the advantages of the method are that it avoids the state change in O2, then why not just operate a 'normal' Li-air battery at that temperature? This would also avoid the state change. I'm sure there are good reasons, and just pointing out that the article's motivation is a little lacking.

A matrix of cobalt oxide?

Isn't that going to make it heavy?

I don't get it. It contains its own oxygen in an electrolyte matrix. So it isn't any lighter than existing lithium batteries? What's the point?

>...prototype stores twice as much juice as the lithium-ion cells the Gigafactory will turn out, and which could eventually do better than that.

Twice as much on a per-volume basis? Or per-weight?

Energy density is measured in J/L, so volume.

Specific energy is measured all sorts of ways. J/L is not very helpful for cars at all. So this battery will probably not be very interesting for transportation?

Just because the energy density is measured in a certain way, it doesn't then follow that the energy density when measured in a certain way is going to be significantly worse.

J/L is actually one of the more important ways to measure battery efficiency for transportation. In many discussions around this theme you'll find this to be used in comparison with a gas tank for size and that ultimately translates into range, arguably the most important feature of an electric car.

Higher energy density means higher range, all other things being equal, and with the weight of cars being what they already are a battery weighing a few hundred pounds would be perfectly acceptable.

So J/L is actually very helpful.

Per-weight is almost double current lithium batteries. But the much bigger issue it solves seems to be the energy loss versus other types of lithium-air batteries which can lose up to 30% to heat, versus 8% in the lithium-oxygen battery.

Did we not read the same article?

Lithium Air batteries could offer up to 4x the energy density, this one is 2x.

Part of the reason this hasn't been done before is that previously the oxygen introduced to the battery was from the air and other components of air damaged and degraded the battery rapidly.

This battery is sealed with all the Oxygen it needs as a LiO mixture stored in a cobalt matrix to stabilize it. Thus is doesn't degrade.

The MIT press release has more details, but the cathodes are half the weight of the current Lithium batteries and so can store almost double the energy for a given weight.

Please warn us if something is paywalled.

Prove that it can survive in the wild without becoming a firebomb, and we'll talk.

This is an unreasonable standard that doesn't hold true for gasoline or other fuels.

The byproduct of having something that has a high energy density is that it has a lot of energy to dissipate when things fail.

Yes and no. Yes it is an unreasonable standard of safety for cars (and maybe planes) today. No, it's unreasonable to allow things to blow up in close quarters like on a laptop.

I don't get how a laptop exploding is less acceptable than a car or an airplane exploding.

Because a car won't explode due to malfunction, and not on your reproductive organs.

That makes no sense. Why do you think motorcycles are called "crotch rockets"?

For their speed/weight ratio? Certainly not because they typically don't explode... Outside of Schwarzenegger films.

Contrary to what Hollywood tells us, mopeds, bicycles, cars, trucks, motorcycles, and boats almost never explode (Airplanes occasionally explode. Rockets frequently explode.) They certainly don't explode as frequently as your average laptop battery.

Anytime someone blithely talks about including an oxidizer (or oxygen) into the mix, it's good to remember all of what you just said. Highly reactive elements and oxygen are far more scary in general, and given enough oxygen pretty much everything becomes highly reactive.

That being said, while a car won't explode, they sure as hell do burn.

In cars the flammable bits are normally fairly separated from the passengers so if the engine goes up in flames you can just get out and watch. Laptops less so.

Of course, you may find that the impact which caused the fire has also rearranged those fairly separated parts. Still, that's predicated on a crash, which is at the very least, fair warning that more things are likely to be going wrong. "My laptop seems war- OH GOD", less so.

Cars are an absolute necessity of what we consider to be the modern world, while laptops with lithium-air batteries could not possibly be considered that critical?

Gasoline alone does not contain extractable chemical energy. Gasoline mixed with oxygen does, though, and that mixture makes a pretty good bomb if you have any amount of it, like most strong reducers mixed with strong oxidizers. Liquid fuels may burn if they are exposed to air above their flash point and a flame exists, but they will not explode like a bomb.

Unless you mix a bit of it with a lot of air, gasoline is surprisingly quiescent.

Reply on news.ycombinator.com