This looks closer to an industrial product than many new battery technology announcements. The cathode chemistry isn't exotic. The efficiency is high and stable (supplementary table 3). The rate capability is good. The specific energy is quite good. The cycling stability is pretty good.
The trickiest part looks like the chemical vapor deposition of graphene onto SiO2 nanoparticles. CVD is a slow growth process that I normally see applied to creating precise, thin layers on flat substrates. I think it would be hard to scale this up to industrial (tonne per day) quantities of coated particles. Is it possible to replace that process with something like a fluidized bed reactor? I'm out of my depth here regarding paths to scale-up -- I have a chemistry background, but I'm not qualified to comment on most chemical engineering.
Wow, this might be one of those rare instances where new research is gonna proceed rapidly into industry. The paper[1] isn't shy about it either. This is great on all fronts: increases cycle life, charge speed, and even marginally increases capacity. They're very optimistic about integrating it into production lines and it seems cost-effective. Cheap, even. The inputs are methane and fumed silica into a 1000 C furnace- you can practically buy those at a hardware store and then it just gets mixed into the r2r slurry.
I think it's pretty likely that charge speeds are about to increase handily. Fig. 4 shows the battery with additives charging at 5 C compared to virgin chemistry at 1 C. That's about 5 minutes to charge the middle 50% of a battery- incredible. Still remains to be seen if this is compatible with standard additives and SEI conditioning, but I'd be surprised if it didn't work out fine.
I feel the same about the CVD but it looks like it was fast and easy from the paper (as much as I recall right now). Certainly way less exotic than most CVD.
I'm out of my depth here regarding paths to scale-up -- I have a chemistry background, but I'm not qualified to comment on most chemical engineering.
I applaud you, sir, on your awareness of self and awareness of scale. If only all programmers were as self aware and aware of the difference between a background in programming and software engineering at scale, I'd smile a little more.
CVD in general is slow and time consuming, but because graphene is just one atom thick growing graphene specifically might not be too bad. Also the process gas/chemistry requirements are relatively very simple.
For reference, I used to see ~4 hour cycle times for growing graphene on copper sheets. I think most of that was in the heat up and cool down, maybe order of 30min actual growth time. These numbers might be off by a factor of 2-5, it's been a few years since that job, and I was a equipment design engineer not a process guy as such.
>This looks closer to an industrial product than many new battery technology announcements.
Compared to other Battery tech advance from University Research and Startup which are trying to hype and gain new funding, Samsung doesn't need that. And my guess it is at least small scale production ready before they make such announcement. ( Or they knew a competitor which has a similar product announcement soon and step up ahead of them )
Hopefully we see this in shipping product by 2020.
Can you help parse the specific energy in terms of Wh/Kg? I feel like I’m doing something wrong since I keep arriving at numbers about an order of magnitude better than current lithium tech.
They claim a full size battery has a "possibility" to reach 800 Wh/L. I'll just use 800 for illustration. They don't report the specific gravity of a full battery, but for other lithium ion batteries a specific gravity of 2.5 - 3.0 might be reasonable.
800 / 2.5 would be 320 Wh/kg.
800 / 3.0 would be 267 Wh/kg.
Both numbers are quite good, as batteries go, but not an order of magnitude higher than what's available now.
An order of magnitude might be really hard and may be impossible to reach *1. Price, charging speed, and lifetime have more space. 5X charging speed is huge for EV.
pause the screen at 2:49- he's calculating the specific energy for a very specific chemistry, lithium cobalt oxide with a carbon anode. EVs don't even use that chemistry. It's more appropriate to use a more general formula.
Here's the most general possible one: only lithium. Each lithium atom gives up one electron, at some voltage. The standard electrode reference voltage of lithium is 3.04 volts. That works out to 26.8 amp-hours per mole, and 81.47 Wh/mole. A mole of lithium weighs 6.941 grams. The end result is 11.74 kWh/kg. That's the absolute, utter maximum energy density for a closed system battery (which is why li-air can exceed that figure).
I am continually surprised by how quickly capacity keeps increasing towards that. Battery capacity will easily double with tech quite similar to current, and 10x would not surprise me within my lifetime.
The trickiest part looks like the chemical vapor deposition of graphene onto SiO2 nanoparticles. CVD is a slow growth process that I normally see applied to creating precise, thin layers on flat substrates. I think it would be hard to scale this up to industrial (tonne per day) quantities of coated particles. Is it possible to replace that process with something like a fluidized bed reactor? I'm out of my depth here regarding paths to scale-up -- I have a chemistry background, but I'm not qualified to comment on most chemical engineering.