Battery breakthrough could change everything | Hacker News Reader

Battery breakthrough could change everything

wired.com

53 pointssteven12 comments

Okay, this is a little more information than we had a couple of weeks ago [0].

Sounds like these could beat lithium batteries on cost, even if their energy density and power density aren't at the same levels. (Not saying they won't be, but we haven't seen any numbers yet.)

[0] https://news.ycombinator.com/item?id=14941624

I personally think that, for electric cars at least, there are two main things in battery tech holding them back:

Capacity and charging time

Basically, if we could get a battery cell that was no bigger than say, the lithium cells used by Tesla (sub-C size? I can't recall what they used), and with about the same mass, but had double the capacity - in theory, Tesla could swap out one for the other and instantly double the range of their vehicles. That would instantly put them in line with the range of a gasoline powered vehicle (to be honest, though, they are already there or close enough - provided you aren't too aggressive with the pedal).

But extra range is just one component. The battery also needs to be able to be charged in a reasonable amount of time. I would personally like to see 5 minutes tops to a full 100% charge. I want to be able to travel cross country, and if I don't want to do an extended stop, I'd like to get out, charge the car to 100% in 5 minutes, and then drive on. I don't want to have to wait, I want to be able to do the same as I can do today with a gasoline powered vehicle.

I honestly don't know if such a scheme is even possible within the laws of physics, while still being safe for the consumer. It might be a case where a different form of infrastructure needs to be created to handle charging, due to likely high currents and/or higher voltages involved. It doesn't seem possible to offer this as a "plug in" style recharging system, but perhaps something you drive into, and large contacts or something from the bottom rise up and "plug in" to ports on the underside of the vehicle could work? There are also the "battery swap" schemes, but this seems like something that could take longer than 5 minutes. Regardless, just about anything will ultimately require some kind of standards for car manufacturers to adhere to so it all works (imagine if there were different size nozzles for each manufactured vehicle for gasoline).

Note that I don't expect for "overnight charging" to happen quickly; existing charger designs (or similar) could likely handle this for home or on-the-road use (when there isn't another option, or for perhaps plugging in while staying at a hotel or something).

Until these two items are solved, though, I don't see electric cars being as widespread among consumers as many think they will be.

I see no major problems with overnight charging in the long run. We do it for our mobile phones and it's not a major hassle, the only thing holding it back for cars is availability of charging points. It's a bit of a 'chicken or the egg' problem, but some parts of the world appear to have reached the critical mass of electric cars and charging points to have steady growth in both, it's just a matter of time before the rest of the world catches up.

To charge an 85 kWh in 5 minutes would be ~1MW which seems beyond home charging capacity.

Even more out of reach if you want to double the capacity

If, let's say 95% of energy goes into the battery, and 5% becomes heat - what would the cooling system need to look like?

5% of 1MW is obviously 20kW. That's rougly comparable to the "large" 68000BTU/hour space heaters.

On the other hand, it's well within the range of a conventional car engine cooling system: http://what-when-how.com/automobile/forced-circulation-water...

It depends somewhat on how hot the hot side can afford to get. The "weak link" will be getting the heat out of the cells themselves - it's no good if you can cool the exterior to 20C if the interior exceeds the safe limit of the cell.

> To charge an 85 kWh in 5 minutes would be ~1MW which seems beyond home charging capacity

It would be pointless to attempt this directly from the grid, but IMHO charging a home battery slowly and using that to charge a car quickly when needed, will become practical soon. The grid would only need to provide as much power per time as you could spend driving your 1-2 cars, so perhaps 50-100KW.

Hailable fleets of self driving cars could get around the range and charging times, with current technology.

Charging times: fleet is driving most of the time. There's a home base that is also charging batteries simultaneously. When one starts to run low, they head home and swap out the battery. Even if the swap takes 10 or 15 minutes, that wouldn't affect utilization rates too much. Though, depending on the economics of utilization, having the home base just provide Tesla-style superchargers could also be fine.

Range: this is a bit trickier, but not insurmountable. You can have transfer stations every couple hundred miles. And perfectly timed transfers wouldn't be too hard. Though even there a self driving electric bus would have cut out the most expensive component (labor) and would have significantly more range than a regular electric car. I wonder if SF to LA would be doable on a single trip.

Summary of tech: But in this battery, you have no liquid. You have just a plastic, a polymer, that replaces the liquid, so it’s solid. To be solid instead of liquid is something people have been striving for for 100 years. Michael Zimmerman was the expert on a certain class of polymers. He invented a new ionic conduction mechanism.

My main concern with the electric vehicle hockey stick growth curve is sourcing the cobalt required in Li-ion battery chemistry (although there are non cobalt versions LiFePo with worse metrics). One of the basic problems is how does a cobalt miner know whether or not something like this article's battery doesn't pop into existence just as soon as they put in billions of dollars to get a cobalt mine going. I have a suspicion that Tesla is waiting to be more clearly stable financially and then they will do something to derisk cobalt miners (like promise the purchase of a boatload of cobalt at a certain price).

Aside from a little bit of mainstream tech journalist misinformation (Bill Joy being co-inventor of Java? News to me, and presumably the inventors of Java listed on Wikipedia), glad to see this article, the development being described is very promising. I have a bit more hope in this than in most battery breakthroughs, the fundamental idea seems sound, will be interested to see how it develops.

Together with James Gosling and Guy Steele, he wrote the first Java spec: http://titanium.cs.berkeley.edu/doc/java-langspec-1.0/

Reply on news.ycombinator.com