Designing better batteries for electric vehicles

jillesvangurp 2021-08-18 07:08:29 +0000 UTC [ - ]

Iterating on battery design is important but it's important to realize that existing batteries are already pretty good. More than good enough to be cost competitive with ICE vehicles over the life time of the vehicle in most classes of road transport (e-bikes to heavy trucks) and trending towards being cost competitive on purchase price as well in the next years and production price soon after (purchase price typically involves tax incentives).

Most of that is just relentless optimization by companies that have gone from small factories to operating multiple so-called Giga factories, i.e. factories producing multiple GWH worth of battery capacity. Tera factories are coming next. Tera factories are essential to be able to actually satisfy demand. Currently any battery that is produced finds its way to a product that probably has some kind of waiting list. Demand is that strong. Driving prices down below 100$/kwh is important for this as most cars don't need more than 50-70 kwh (i.e. 5000-7000$) worth of batteries. Tesla does not disclose their battery cost but they've been rumored to have long broken through that price point. Chinese manufacturers are selling cars below 5000$ in their domestic market already.

Solid state batteries will have to be cost competitive to be able to make a dent in that market. That will likely happen eventually but probably not in the next year. However, demand won't be an issue even at an initial premium price point. There are plenty of markets where it makes sense to pay a premium for a lighter, more efficient, safer battery. Prices will come down as that market matures. But that might take a decade or so.

Cthulhu_ 2021-08-18 08:21:16 +0000 UTC [ - ]

There's some practical concerns with current generation batteries though; use of rare earth materials, lithium, and fire hazards and risks in accidents.

I mean they're not the end of the world and they are acceptable compared to e.g. the environmental damage done by the production and consumptions of petrol and diesel, but I feel like they still iterate on them a lot.

I mean the ideal here is not to just create an alternative to ICE's, but to come out on top. For a good while the only reason people went for an electric car besides idealism was subsidies. But if an electric car ends up being superior in every way - range, handling, cost - of an ICE car, they will compete ICE cars out of the market without subsidies or force.

Because at the moment it feels forced. Car manufacturers have already announced they will stop producing ICE cars in a decade or so.

jillesvangurp 2021-08-18 09:16:28 +0000 UTC [ - ]

ICE cars catch fire all the time. But when an EV catches news, it's major news. It happens relatively rarely. Nowhere near enough for this to be a concern.

Interestingly, some rare earths used for EVs are also used for ICE cars and fuel production. For example cobalt is used for getting rid of sulfur in pretty large quantities as part of the oil refining process. The difference is that with EVs, recycling is both feasible and already common and expected to become a multi billion dollar business as actual volume of batteries ready to recycle becomes larger.

The article mentions battery yield as an issue without mentioning that e.g. Tesla recycles all their discarded cells. Why wouldn't they? It contains valuable materials.

ICE car demand is going to collapse before the EV market is ready to supply replacements. The reason is that people will anticipate the approaching point where that market is collapsing and will stop buying new ones because they expect the second hand value is going to be poor and because EVs are obviously better/cheaper. That's already happening. It's going to be a rough ride for manufacturers. That's why essentially all of them are ramping up investments in battery production right now: it's a survival strategy for when their existing revenue is drying up.

That in turn drives down battery prices; thus speeding the whole process up. Manufacturers are hoping for 2030 but planning for 2025.

lithium_throw 2021-08-18 12:28:03 +0000 UTC [ - ]

>For example cobalt is used for getting rid of sulfur in pretty large quantities as part of the oil refining process.

I don't know why this piece of misleading information is recently circulating so much.

By far the most common usage (63%, in 2020, and having grown from 20% in 2006)[1] of cobalt is for lithium battery electrodes (and that is despite the fact that EVs make up a miniscule percentage of cars on the road).

Cobalt is used as a catalyst in the petroleum industry, however its depletion rate is of the order of kg per millions of litres of fuel.

>Tesla recycles all their discarded cells. Why wouldn't they?

I'd like a source for that. The best I can find is that "Tesla say that 100% of their cells can be recycled".

That is true (as it is of any lithium-ion cell), but not in a way that makes any economic or practical sense.

[1] https://www.globalenergymetals.com/cobalt/cobalt-demand/

jillesvangurp 2021-08-18 13:28:59 +0000 UTC [ - ]

It's not nothing; so it isn't misinformation. Also there are many other rare earths used for cars of course. The point here is that ICE proponents get all green and hippy like when it comes to EVs but are not willing to talk rare earth sourcing when it comes to ICE engines, fuel production, etc. It's called selective bias. The reality is that ICE vehicles contain a wide variety of materials. Ice vehicle recycling is mostly focused on extracting steel and aluminium however.

Cobalt specifically is also becoming less relevant for EVs in any case because more recent batteries use less of it; or none at all in some cases. Like for example the solid state batteries discussed in the article.

Regarding recycling, Tesla is working with third parties as well as working on in house expertise:

https://electrek.co/2019/04/16/tesla-battery-recycling-syste...

You need economies of scale for battery recycling to become more significant. One minor challenge here is that batteries seem to last a lot longer than people expected a few years ago. Supply of used up batteries is simply not there yet. Mostly the industry is still focused on recycling batteries from laptops, phones, etc. So, not an urgent issue for Tesla to be working on yet. But there are of course plenty of startups in this space as it is such an obvious thing to start doing in the next decade. J.B. Straubel, the Tesla co-founder, just got a nice investment of 700$ million for one of those companies. So, very practical and lucrative apparently.

lithium_throw 2021-08-18 14:49:51 +0000 UTC [ - ]

>It's not nothing; so it isn't misinformation.

I could nitpick over your definition of "a very large amount of cobalt is used in petroleum refining", but the fact that an EV owner creates orders of magnitude more demand for cobalt than an ICE car owner speaks for itself.

>Ice vehicle recycling is mostly focused on extracting steel and aluminium however.

What other rare earths are present in an ICE car's engine? They are almost entirely comprised of alu and steel, which is why recycling them is so easy. Platinum is also very widely extracted from old catalytic converters.

Obviously modern cars also include a huge amount of (almost entirely un-necessary) electronics, but if you want to start talking about that, then we'll also have to include all of the other (almost entirely un-necessary) electronics we own.

Nice article on Tesla and battery recycling. Exactly which part supports your previous assertion that "Tesla recycles all of its batteries"?

>You need economies of scale for battery recycling to become more significant.

I get a feeling we'll be hearing this line for a long time.

guerby 2021-08-18 14:01:59 +0000 UTC [ - ]

"kg per millions of litres of fuel" we consume lots of fuel, if 1kg per million liter that's between 5 and 6 thousands metric tons of cobalt per year. number two cobalt producer is russia with 6 thousand metric ton per year (first is DR Congo with 100 thousand metric ton).

Also rare earth are present in fossil fuel cars (Catalytic converter amongst other pieces):

https://engineering.stackexchange.com/questions/13757/what-r...

But nobody ever talked about these before EV were a thing.

lithium_throw 2021-08-18 14:43:57 +0000 UTC [ - ]

Apparently, a typical EV battery contains 0.13kg cobalt per kwh. That's 13kg cobalt in a 100kwh battery.

Over ten years, I might drive 100,000 miles, using 12,500 litres of petrol. That's 13g cobalt, in the form of sulphur-removing catalyst.

Very obviously, widespread adoption of EVs is going to cause a tremendous increase in the demand for cobalt.

You don't do EV advocacy a favour when you try to distract from these things.

guerby 2021-08-18 18:43:50 +0000 UTC [ - ]

I did the computation because I just learned about this cobalt use for fossil cars (thanks to HN !) and wanted to check if it was significant or not.

On your comment:

- Cobalt (and CO2 and other chemicals) from fossil car is lost as pollution, cobalt from car batteries will be recycled (cheaper than mining) - even if recycling losses will likely be near in magnitude to fossil losses in the case of Cobalt.

- Typical EV batteries are not 100 kWh

- Tesla Model 3 SR+ MIC have 0 cobalt since use LFP chemistry. Tesla said it will use LFP for all entry level cars in the near future, and no longer use cobalt

- You didn't address the point about rare earth media coverage for fossil cars vs EV

My take on the last one: given that the company currently owning 75% of the BEV market in the USA doesn't pay mass media for advertisement at all, given that fossil car maker spend lots and lots of money for advertisement (in the 10% of car value range), mass media has massive incentives to paint EV in a negative way and it what's they're doing since the beginning.

lithium_throw 2021-08-19 08:47:41 +0000 UTC [ - ]

Sure, cobalt can be recycled from EV batteries (it's one of the few things that is currently economical to extract, and even then, only if the batteries are essentially incinerated first). At around 98% recovery, that means you'd lose about as much cobalt as an ICE-driving person would consume over that same period. However, it doesn't help the fact that every EV currently needs around 10kg of the stuff in the first place, so that all has to come from somewhere.

>Typical EV batteries are not 100 kWh

For these order-of-magnitude calculations, it doesn't matter if we're talking about a 100kwh battery or a 75kwh battery.

>Tesla Model 3 SR+ MIC have 0 cobalt since use LFP chemistry.

LFP might have some nice properties, but it has around 2/3rds the energy density, and poor temperature performance. I'm sure you'll tell me "it'll get better", but then it looks to me like we're going to continue buying many generations of EVs with the promise that "the next one will fix everything!".

>You didn't address the point about rare earth media coverage for fossil cars vs EV

Why is that an issue? ICE cars don't use much in the way of rare earths, and certainly not until de-sulphurisation became a thing (which is predominately for diesel, anyway), and catalytic converters became mandated.

Once we're done talking about cobalt, we can move on to all the rare earths in an EV's traction motor...

inglor_cz 2021-08-18 10:23:47 +0000 UTC [ - ]

"It happens relatively rarely. Nowhere near enough for this to be a concern."

You are too dismissive. Not every fire is created equal and putting out battery fires is much harder than putting out gas/diesel fires.

Also, EVs sometimes go up in flames when at rest, which, short of deliberate arson, is not a failure mode in ICE cars. ICE cars will generally catch fire underway, far enough from other cars, thus limiting the damage to one car. (A deadly exception might happen in a tunnel.)

You might have seen this video from China.

https://www.scmp.com/video/china/3136069/electric-bus-bursts...

If something like that happened in a car park full of EVs, for example, underneath a shopping mall, the consequences would be pretty bad.

pjc50 2021-08-18 11:31:02 +0000 UTC [ - ]

Catching fire at rest is a risk for ICE cars: https://www.autosafety.org/bmw-recalls-1-million-vehicles-fi...

qzw 2021-08-18 13:26:38 +0000 UTC [ - ]

Yup, had an ICE car that was part of a recall for risk of fire when parked. The letter I received suggested not parking inside a garage until the repairs had been completed.

Thlom 2021-08-18 10:58:47 +0000 UTC [ - ]

An airport parking facility burned to the ground in Norway a couple of years ago because an ICE car went up in flames while parked.

magicalhippo 2021-08-18 13:15:29 +0000 UTC [ - ]

According to the evaluation report[1], the fire started after the car was started, while stationary. So I think it's unfair to call it "while parked".

However the report also mentions in section 4.1 a study which concluded there's no noticeable difference in the chance of catching fire and the severity of the fire between ICE vehicles and electric vehicles.

And in 4.2 they say that in this fire there was not any noticeable difference between the two types of cars. No any evidence was found to suggest that any of the EVs experienced thermal runaway.

[1]: https://dibk.no/globalassets/02.-om-oss/rapporter-og-publika...

pigeonhole123 2021-08-18 13:05:24 +0000 UTC [ - ]

And ignited the impossible to extinguish electric car batteries in its vicinity

happosai 2021-08-18 13:46:01 +0000 UTC [ - ]

Extinguishing EV fire is not hard, you just just need a lot of water. The problem is that it will re-ignite after a while. So you can't just leave it after that, you need to move the wreck to a safe location and have some firefighters wait until if reignites, drown it in water and repeat.. The alternative is to drop the car in a container of water, but has it's own drawbacks.

It really depends on what kind of risks we willing to accept. Since people sleep well in homes with pipelines of explosive gas

https://www.dallasnews.com/news/2021/07/20/plano-home-explos...

I don't think EV fires are going to stop their adoption.

paul_f 2021-08-18 10:01:32 +0000 UTC [ - ]

It is not the frequency of EV fires that is a concern, but the difficulty and danger associated with putting them out. From this article, an ICE fire requires 300 gallons of water to extinguish - and for electric vehicle fire, 28,000 gallons and 7 hours. https://www.nbcnews.com/business/autos/federal-regulators-wa...

rcMgD2BwE72F 2021-08-18 08:33:33 +0000 UTC [ - ]

>rare earth materials

Which?

>But if an electric car ends up being superior in every way - range, handling, cost - of an ICE car, they will compete ICE cars out of the market without subsidies or force.

This is already the case.

Check BMW Series 3 sales (or any entry luxury sedan) against the Tesla Model 3. And how sales of Model Y are growing against comparable ICE midsized SUV. When Tesla release their compact cars (expected around 2023), it is game over. If you don't trust Tesla, see Porsche Taycan sales against their Panamera.

guerby 2021-08-18 09:01:28 +0000 UTC [ - ]

On the last point:

https://electrek.co/2020/09/22/taycan-electric-car-porsches-...

https://insideevs.com/news/517601/us-porsche-taycan-sales-20...

_ZeD_ 2021-08-18 09:08:08 +0000 UTC [ - ]

> Iterating on battery design is important but it's important to realize that existing batteries are already pretty good.

are you serius? do you really think it's "pretty good" the range electric vehicles have, compared to ICE ones, looking also at the added weight and the time needed to recharge / refuel?

Electric vehicles are not a fad, but the battery energy density it's still... in the "tolerable, at most" phase of the tecnology

just FYI I just buyed a plug-in hybrid car mostly because I don't think a electric only vehicle it's enough for today.

adrianN 2021-08-18 09:15:28 +0000 UTC [ - ]

EVs can cover almost all usecases for a large number of people. I think that counts as "pretty good", the main issue seems to be cost and lack of charging infrastructure for people without a garage.

h4vot 2021-08-18 10:41:12 +0000 UTC [ - ]

Husband got a Jag i-Pace a few months ago. 400 km range. Haven't even bothered to get the fast charging equipment installed yet at home just give it 240 volts at night.

Its never been under 50%. I think for the vast majority of city and suburban folk, 99/100 trips are not an issue.

sokoloff 2021-08-18 14:19:26 +0000 UTC [ - ]

I've driven (and overall enjoyed) a Nissan LEAF (pure EV) for the last 6.5 years. It's true that 99/100 of trips are not an issue, because many trips are short. Back/forth to work, back/forth to lunch, back/forth to sports, back/forth to the post office or grocery store.

Those quickly add up the 99% of trips (as measured by getting into the car and going somewhere). Those 1% of trips really impact your overall satisfaction with the car, though. I don't remember all the times the car effortlessly and nearly silently took me to work, but I do remember when I couldn't go to visit a friend or play in a softball tournament because my wife needed to use the ICE car for something and the LEAF couldn't do the trip. Or where we had to carefully plan who was going to take which kid to which activity and who was going to take the dog to her chemo appointment and in which car. If you "plan it wrong" in an ICE or hybrid, you lose 10 minutes. If you "plan it wrong" in a BEV, you lose an hour minimum (and more in a LEAF; mine has been DC fast charged twice since new, once before delivery and only once have I found and used a CHAdeMO in the wild).

I liked the LEAF (as the person who maintains our cars, I absolutely treasure it), but I suspect my next car will not be a pure electric. Teslas are more money than I choose to spend on a car. (Taycan is way more.) I'll probably end up with a 10-year old ICE car that will do everything I need for 20% of the purchase price, 30% of the insurance, and 150% of the marginal cost per mile, which is fine as we drive each car only about 3-5K/yr.

gameswithgo 2021-08-18 14:35:34 +0000 UTC [ - ]

You might be able to get a used Tesla by then. Shops are proliferating that can repair battery packs for way less than replacement which offsets a lot of the risk.

sokoloff 2021-08-18 14:40:32 +0000 UTC [ - ]

Tesla's got their service information and parts supply far too locked up for someone like me to ever own a used Tesla.

Could be a good option for people who prefer paying someone else to maintain their cars (which is most people, of course).

gameswithgo 2021-08-18 13:10:07 +0000 UTC [ - ]

Very close to good enough, if the Tesla tabless stuff works out, I think we will be at a point where it is fine for close to everyone when it comes to range/charge time.

But teslas are premium, will be some time for things like a Nisan leaf to get the ~400 mile range + fast charge times.

baybal2 2021-08-18 09:27:25 +0000 UTC [ - ]

> Iterating on battery design is important but it's important to realize that existing batteries are already pretty good.

Seconding this. The current mainstream will hover at around 200-220 Wh/kg for a long time. It already does.

Cell energy densities above >200 Wh/kg were available for a long time, but even quite major car brands intentionally chose to not to use premium priced cells, and stayed on commodity grade ones.

Cells sold on the open market still have a quite big double digit markup. Only direct sales to EV makers go below $100 per kWh.

And yes, small cars for people only driving in the city make the most of cars sold around the world. Here, your observation is completely correct.

In China, even lead acid battery powered cars find buyers in small towns, and villages.

Dropping 100kg in lead, and structure from small cars with around 10kWh packs will make a big difference.

It's not Tesla whom the big automakers are afraid of, it's Wuling MiniEVs — a car for a city driver which petrol powered cars can't beat at $5000 pricepoint.

blunte 2021-08-18 10:10:19 +0000 UTC [ - ]

What happened to the arrays of nano capacitors which were a promising alternative?

csours 2021-08-18 14:10:39 +0000 UTC [ - ]

Are such solid state batteries suitable for cell phones and laptops, power tools (drills, saws) etc?

Are there conflicting engineering goals for batteries of small consumer appliances and vehicles?

gameswithgo 2021-08-18 14:37:17 +0000 UTC [ - ]

Probably only a question of cost and fire risk. A solid state would be better for a phone, but the cost/weight/safety trade off might not be worth it there, by the time it is in a car (if it ever gets there)

Notable that electric car players are still innovating on non solid state, so I guess solid state isn't imminent.

baybal2 2021-08-18 09:06:41 +0000 UTC [ - ]

Note, the mass of an anode is at most 15% of mass of the modern battery cell.

All "metallic" lithium cells still need some mechanical structure to hold lithium. So, at most you will get a half of the mass of graphite, given that graphite is already a quite lightweight material.

For the cost, the second most expensive part in the cell after the cathode is the separator. Close to a third of cell's cost can be the separator. Japanese currently have a near monopoly on high-end separators, with Chinese having the rest of the market.

csours 2021-08-18 14:00:11 +0000 UTC [ - ]

From where do the Korean battery manufacturers source their separators?

baybal2 2021-08-18 14:31:55 +0000 UTC [ - ]

I have no idea honestly. One may think they may be trying to avoid Japanese suppliers, but who knows. The entirety of Korean semi-industry sits on Japanese consumables for example. Times change.