Great writeup. I think ebikes are not as widely covered in US media because most of the USA doesn't have good bike infrastructure. Try biking to work in Houston from the suburbs or Phoenix and you'll see what I mean. Maybe as part of general densification and the revival of urbanism we can see proper bike infrastructure.
Even in a place like Houston there are probably several hundred thousand people who live within a few miles of work and with a bayou bike trail that covers a significant fraction of that distance. Once biking itself doesn’t make you any hotter and sweatier than being outside (and the ebike means you even get a bit of a breeze) this could start to be big.
I ebike to work periodically on mostly greenway in Wyoming. Its 5.5 miles but I don’t sweat on turbo going to work (and I don’t care about sweat coming back).
The problems: my preferred shopping is not easy to chain trip (why I skipped today though I biked yesterday), it doubles my commute time (from 15 minutes to 35 ish each way), inclement weather (including summer thunderstorms), and darkness--I refuse to ride in the dark.
I can do it a lot in spring and fall and have been building up a winter kit to ride in cold mornings but since I have to have a car its always more of a hassle.
I relied on them a lot when I was in Sacramento for the state fair. At the time, Uber had a monopoly on the e-bike and e-scooter market in the city, but as a result charged much lower rates than in cities where they charged the market rate. For the e-bikes, the rate was $1 for the first 15 minutes. (In that time, I could go from near Cal State to the Tower Bridge).
The bikes provided great mobility and were widely used. Also, bicycling was pleasant in downtown and midtown. It was 100 degrees that weekend, but Sacramento streets are sheltered by a thick tree canopy so it only felt like the high 70s when biking.
Worth noting the US just plowed a ton of money into their battery supply chains as part of the IRA. Literally every segment, plus R&D, factories, and spurred massive demand. Tens of billions. Countries are taking this v seriously as part of industrial policy.
In addition, the recyclability of metals is insane. Like 99%+, this is such a change in circularity compared to use once fossil fuels. BFD, great article!
Permitting will be a huge issue too. After nuclear power this could be the next big wedge in the enviro movement with the ecomodernists on one side and the neo-pastoralists/degrowthers on the other.
That's true-ish. The investment is there, it's the permitting and project development that is lagging. We may see something similar to the solar industry in the mid aughts. We had a severe polysilicon shortage.
There was a stalling out of the price curves for like 4 years. Interestingly, once that bottleneck was resolved, the solar cost curve accelerated and caught up to longterm cost curve because the manufacturing gains were hidden by elevated commodity costs.
The other interesting impact of these very high prices is that it changes what are "economic reserves" at these prices - that opens the door to things like lithium clays. If prices remain elevated, there will be material substitution and economic optimization, especially at the scale of investments that we are talking about here.
Finally, there is a lot of talk about how lithium has to 40x while other materials have to 2x. This is true, but also lithium is coming from a much lower baseline- it was used as a niche substance for ceramics and grease and small ball stuff. Not trivial to expand, but in comparison to the level of extraction we do for other materials, well within reasonable bounds.
Sort of - the fundamental problem is that it just takes a long time to build mines anywhere. They can't be spun up in just a couple years in most cases across the world, the US is just particularly gnarly. The time from discovery to scaled up production can be up to a decade long. This could drive some pretty major supply shortfalls in the near-term, but likely it will be resolved by more mining if demand stays tight.
That's great data, and I hope to find time to pore over some of your sources. Long ago (1990 - 2000), I plotted up energy density vs year of introduction (on a log scale plot of course), starting about 1900, with the Voltaic cell. In the early years it took about 20 years for each new technology to come on stream, and over the previous century the doubling time stayed about 10 years. It has increased, so that now capacities triple in 10 years or less. I suspect the acceleration comes in bursts as manufacturing scales up with much greater total delivered capacity. It has still taken over 20 years for lithium battery systems to dominate, so the total amount of the technology delivered has to be a factor in the faster recent growth. If you run across estimates of the total delivered capacity, it may give a more accurate characterization of the potential growth in batteries going forward, since we can estimate the battery content implied by various product adoption rates.
Incidentally, handheld power tools, I suspect, have driven a big part of this increase. They come into wide use in the 1990s, and at first were the last big use of the previous, Ni-metal hydride batteries, which could provide higher currents. And then digital cameras exploded, but are now seeing their niche filled by the smartphone.
100% agree that batteries and autonomous machines go together and love to see it highlighted. Electrification not only makes industrial machines clean and quiet but also dramatically more robust without all the moving parts required by ICE power. It also gives far better self-diagnostics with all the sensors that can be trivially implanted.
You mention mowers and robots separately, so I thought I’d shamelessly plug Scythe https://ScytheRobotics.com where we’re leveraging all those battery-electric advantages to build the next generation of off-road power equipment to be clean, quiet and autonomous. Strong believer in the decade of the battery _and_ AI since they’re so symbiotic.
One thing I realized the other day is how well batteries deal with the intermittency of renewable generation. We could well get a situation where electricity has substantially different prices over the course of a day and a season, maybe with difficult to predict fluctuations over the week. But if all of your home appliances have batteries, then (for instance) your air conditioner could fill up its battery on a cheap electricity day rather than contributing to peak demand while the wind has stopped blowing and the sun is covered. I’ve gotten used to putting my dishwasher on delay to run in the middle of the night when electricity demand is low, but with batteries, you can run whenever you want, and just charge when demand is low, particularly with devices that are either stationary, or plugged in whenever you’re not using them.
It made me realize how useful a home-scale battery could be! My family in India had those when we visited in 2003, because at that time, power generation was intermittent even in big cities. But now, the battery is much cheaper, and could thus be used just to arbitrage against daily fluctuation in electricity prices (due to some combination of daily fluctuation in supply and demand) rather than just dealing with blackouts.
And actually, it seems quite plausible that a car that is plugged in most of the day could serve a significant fraction of the role of a home battery.
They can also serve as generators. Hybrid trucks have power outlets for workers to charge tools. Rivian also boasted of this feature for its electric trucks, which are out on the roads as of this year.
Check this out. Not for cars or devices but for multi GW power grids (and will free up more resources for the other categories. First project is underway and opens in NV in October. Www.aresnorthamerica.com
Great to see e-bikes get a shout out. Also, ramping up manufacturing and supply chain will be a big lift. It's a great opportunity for SE Asia and India.
Hard tech, deep tech are really the future. It’s pieces like these that help get people focused on the big problems we face, and start the process of building solutions towards it.
I’d like to see the innovation and development state side.
But we will need to really focus (re-focus?), be patient and invest (over invest?).
Govt will have to support with more CHIPS like programs.
Electronic devices have been pretty pervasive as is, but I'm sure having higher capacity smaller batteries will make them even more widespread. AR glasses? Health monitoring clothes? Personal drone? So many potential uses.
Yeah. I'll try to get a battery engineer to do a guest post on the specific performance characteristics that batteries need in order to enable all these things.
Back when computing was still young, they used to produce analyses like that all the time. How fast would a computer have to be to solve various problems or work in various niches? How small and light would it have to be? How much long term storage would be necessary to serve as a particular kind of archive?
It would be nice to see something like this for battery storage.
Aug 30, 2022·edited Aug 30, 2022Liked by Noah Smith
Nice post. Finally got me to be a paid subscriber. I'm curious if Noah has any thoughts on iron flow batteries, like the ones ESS is said to be making? It's essentially made of iron, salt and water, so it isn't dependent on mining the ocean floor or mining in unfriendly countries for lithium, cobalt, nickel, etc. The materials are readily available and the technology would be workable for energy scale storage.
I followed the links at the end of the post, but didn't see any mention of it. Peter Coy mentioned it in his newsletter a few months back. Maybe is the future is the Iron Age?
Yeah! Iron-flow batteries look like a great cheap, scalable option for grid storage. But that storage will only work for maybe 1 day, so it's not a solution to the seasonal storage problem...
Great writeup. I think ebikes are not as widely covered in US media because most of the USA doesn't have good bike infrastructure. Try biking to work in Houston from the suburbs or Phoenix and you'll see what I mean. Maybe as part of general densification and the revival of urbanism we can see proper bike infrastructure.
Even in a place like Houston there are probably several hundred thousand people who live within a few miles of work and with a bayou bike trail that covers a significant fraction of that distance. Once biking itself doesn’t make you any hotter and sweatier than being outside (and the ebike means you even get a bit of a breeze) this could start to be big.
I ebike to work periodically on mostly greenway in Wyoming. Its 5.5 miles but I don’t sweat on turbo going to work (and I don’t care about sweat coming back).
The problems: my preferred shopping is not easy to chain trip (why I skipped today though I biked yesterday), it doubles my commute time (from 15 minutes to 35 ish each way), inclement weather (including summer thunderstorms), and darkness--I refuse to ride in the dark.
I can do it a lot in spring and fall and have been building up a winter kit to ride in cold mornings but since I have to have a car its always more of a hassle.
I'd love to see e-bikes catch on.
I relied on them a lot when I was in Sacramento for the state fair. At the time, Uber had a monopoly on the e-bike and e-scooter market in the city, but as a result charged much lower rates than in cities where they charged the market rate. For the e-bikes, the rate was $1 for the first 15 minutes. (In that time, I could go from near Cal State to the Tower Bridge).
The bikes provided great mobility and were widely used. Also, bicycling was pleasant in downtown and midtown. It was 100 degrees that weekend, but Sacramento streets are sheltered by a thick tree canopy so it only felt like the high 70s when biking.
Worth noting the US just plowed a ton of money into their battery supply chains as part of the IRA. Literally every segment, plus R&D, factories, and spurred massive demand. Tens of billions. Countries are taking this v seriously as part of industrial policy.
In addition, the recyclability of metals is insane. Like 99%+, this is such a change in circularity compared to use once fossil fuels. BFD, great article!
Thanks! Yeah, the IRA is really encouraging.
The one problem I'm hearing about is that there's not yet enough investment in the mining sector.
Permitting will be a huge issue too. After nuclear power this could be the next big wedge in the enviro movement with the ecomodernists on one side and the neo-pastoralists/degrowthers on the other.
That's true-ish. The investment is there, it's the permitting and project development that is lagging. We may see something similar to the solar industry in the mid aughts. We had a severe polysilicon shortage.
There was a stalling out of the price curves for like 4 years. Interestingly, once that bottleneck was resolved, the solar cost curve accelerated and caught up to longterm cost curve because the manufacturing gains were hidden by elevated commodity costs.
The other interesting impact of these very high prices is that it changes what are "economic reserves" at these prices - that opens the door to things like lithium clays. If prices remain elevated, there will be material substitution and economic optimization, especially at the scale of investments that we are talking about here.
Finally, there is a lot of talk about how lithium has to 40x while other materials have to 2x. This is true, but also lithium is coming from a much lower baseline- it was used as a niche substance for ceramics and grease and small ball stuff. Not trivial to expand, but in comparison to the level of extraction we do for other materials, well within reasonable bounds.
The permitting problem is mostly within the U.S., right?
Sort of - the fundamental problem is that it just takes a long time to build mines anywhere. They can't be spun up in just a couple years in most cases across the world, the US is just particularly gnarly. The time from discovery to scaled up production can be up to a decade long. This could drive some pretty major supply shortfalls in the near-term, but likely it will be resolved by more mining if demand stays tight.
Can you say more about the recyclability, or provide some good things to read?
That's great data, and I hope to find time to pore over some of your sources. Long ago (1990 - 2000), I plotted up energy density vs year of introduction (on a log scale plot of course), starting about 1900, with the Voltaic cell. In the early years it took about 20 years for each new technology to come on stream, and over the previous century the doubling time stayed about 10 years. It has increased, so that now capacities triple in 10 years or less. I suspect the acceleration comes in bursts as manufacturing scales up with much greater total delivered capacity. It has still taken over 20 years for lithium battery systems to dominate, so the total amount of the technology delivered has to be a factor in the faster recent growth. If you run across estimates of the total delivered capacity, it may give a more accurate characterization of the potential growth in batteries going forward, since we can estimate the battery content implied by various product adoption rates.
Incidentally, handheld power tools, I suspect, have driven a big part of this increase. They come into wide use in the 1990s, and at first were the last big use of the previous, Ni-metal hydride batteries, which could provide higher currents. And then digital cameras exploded, but are now seeing their niche filled by the smartphone.
That's really interesting that it sped up!!
100% agree that batteries and autonomous machines go together and love to see it highlighted. Electrification not only makes industrial machines clean and quiet but also dramatically more robust without all the moving parts required by ICE power. It also gives far better self-diagnostics with all the sensors that can be trivially implanted.
You mention mowers and robots separately, so I thought I’d shamelessly plug Scythe https://ScytheRobotics.com where we’re leveraging all those battery-electric advantages to build the next generation of off-road power equipment to be clean, quiet and autonomous. Strong believer in the decade of the battery _and_ AI since they’re so symbiotic.
That's really neat.
One thing I realized the other day is how well batteries deal with the intermittency of renewable generation. We could well get a situation where electricity has substantially different prices over the course of a day and a season, maybe with difficult to predict fluctuations over the week. But if all of your home appliances have batteries, then (for instance) your air conditioner could fill up its battery on a cheap electricity day rather than contributing to peak demand while the wind has stopped blowing and the sun is covered. I’ve gotten used to putting my dishwasher on delay to run in the middle of the night when electricity demand is low, but with batteries, you can run whenever you want, and just charge when demand is low, particularly with devices that are either stationary, or plugged in whenever you’re not using them.
Yep. Now imagine if everything in your house had a battery in it!
It made me realize how useful a home-scale battery could be! My family in India had those when we visited in 2003, because at that time, power generation was intermittent even in big cities. But now, the battery is much cheaper, and could thus be used just to arbitrage against daily fluctuation in electricity prices (due to some combination of daily fluctuation in supply and demand) rather than just dealing with blackouts.
And actually, it seems quite plausible that a car that is plugged in most of the day could serve a significant fraction of the role of a home battery.
They can also serve as generators. Hybrid trucks have power outlets for workers to charge tools. Rivian also boasted of this feature for its electric trucks, which are out on the roads as of this year.
Check this out. Not for cars or devices but for multi GW power grids (and will free up more resources for the other categories. First project is underway and opens in NV in October. Www.aresnorthamerica.com
Very cool!!
Great to see e-bikes get a shout out. Also, ramping up manufacturing and supply chain will be a big lift. It's a great opportunity for SE Asia and India.
Hard tech, deep tech are really the future. It’s pieces like these that help get people focused on the big problems we face, and start the process of building solutions towards it.
I’d like to see the innovation and development state side.
But we will need to really focus (re-focus?), be patient and invest (over invest?).
Govt will have to support with more CHIPS like programs.
Electronic devices have been pretty pervasive as is, but I'm sure having higher capacity smaller batteries will make them even more widespread. AR glasses? Health monitoring clothes? Personal drone? So many potential uses.
Yeah. I'll try to get a battery engineer to do a guest post on the specific performance characteristics that batteries need in order to enable all these things.
Back when computing was still young, they used to produce analyses like that all the time. How fast would a computer have to be to solve various problems or work in various niches? How small and light would it have to be? How much long term storage would be necessary to serve as a particular kind of archive?
It would be nice to see something like this for battery storage.
Nice post. Finally got me to be a paid subscriber. I'm curious if Noah has any thoughts on iron flow batteries, like the ones ESS is said to be making? It's essentially made of iron, salt and water, so it isn't dependent on mining the ocean floor or mining in unfriendly countries for lithium, cobalt, nickel, etc. The materials are readily available and the technology would be workable for energy scale storage.
I followed the links at the end of the post, but didn't see any mention of it. Peter Coy mentioned it in his newsletter a few months back. Maybe is the future is the Iron Age?
Yeah! Iron-flow batteries look like a great cheap, scalable option for grid storage. But that storage will only work for maybe 1 day, so it's not a solution to the seasonal storage problem...
I wonder if Berkeley will once again offer classes on mining in the Hearst Mining Building.
Hoping the vast resources of the american west get put to good use
Reminds me of this very cool thread on the energy density of batteries vs gasoline https://twitter.com/DrSimEvans/status/1540332129745342465
Yeah. But this still doesn't take into account the weight and size of the extraction machinery, does it?
I'm surprised e-bikes get more mention than electric scooters, like Yadea
https://yadea.com/electric-scooter/yadea-m6l-electric-scooter
Once you stop trying to shoehorn the bicycle components and styling it just seems much more useful.
But outside of Asia I never really see them discussed, only e-bikes.
They're great, but a bit bulkier and harder to park than bikes.
https://www.nationalgrideso.com/news/electricity-explained-battery-storage
Some notes on battery storage in the UK.
I think they mention 30-90 minutes of storage capacity.
Great and interesting article! Thank you!
Thank you!!
This is the kind of unique content I subscribe to you for, Noah. Well done, kiddo.
Thanks!!