Anyone who has been around tech for the last couple of decades will be aware of the liberating and industry disruptive effect of batteries. A piece that showed up in my Medium feed yesterday caught my attention – “Tesla May Have Invented a Million-Mile Electric Car Battery“. It was well worth the read.
Medium does have a paywall if you don’t have a subscription but it allows a free full read of a certain number of articles per month.
Since Edison a century ago, the world of batteries has been riven with exaggeration and fibs. U.S. university and private labs routinely announce purportedly important new leaps, only to go quiet when the cameras are gone.
Tesla CEO Elon Musk himself has been roundhoused for what a lot of people, including U.S. regulators and Wall Street, suggest is an addiction to making claims he cannot fulfill, and sometimes outright violate the law. But now the carmaker has filed patents for a battery system that could last one million miles, enough for 76 years of driving for the average American motorist. Has anyone ever asked for such a car? It’s not obvious they have. But if you are in the mind of Musk, you will see past that, to a future fleet of cars that can transform into driverless, automated taxis when not being used by their owners. A fleet that he hopes will turn into a big new profit center for Tesla.
Leading battery researchers in the United States and Europe, while uncertain about the cost of the Tesla system, say a new academic paper describing the million-mile battery is rigorous and convincing. “The results are spectacular,” said Gerbrand Ceder, a professor of materials science at University of California, Berkeley.
The paper, co-authored by Jeff Dahn, a professor at Dalhousie University in Canada, who is on contract with Tesla, suggests a substantial advance for driverless taxis, buses, and semi-trucks that can recharge in roughly 20 minutes, along with electric grid batteries boasting two-decade lifespans. These are among the greatest ambitions of the new electric age, and a new lithium-ion battery that does what Dahn describes would go far in reviving Musk’s reputation for mastery of applied cutting-edge technology.
If you think about the effect of battery technology recently, it is pretty clear that the shift into Lithium base batteries of various kinds has substantially shifted our technological base. Their characteristic over older chargeable batteries is that they charge faster (at least up to ~80% capacity), have few ‘memory’ effects after repeated charges, are significantly lighter than other battery forms for the same charge, and last longer.
For me, they allow my phone that last for days even when I use it intensively, provide up to nearly two weeks worth of power for my e-bike that I use to commute to work, and power the laptops that I often use in weird locations.
On my last return trip from work in Singapore, I wrote code for most of 11 hour trip on a single battery pack. That would have been unheard of in any of my previous laptops, including those I has only two years before. I was also sending my updates back on the servers.
All of my planned future purchases show the slow flip to lithium. The next upgrade of the UPSes (uninterruptible power supply) that back the server for this site will shift from lead-acid to lithium. It lasts longer both for the site if the power goes out, and for my power cycles. The last car that I’m likely to buy will be an electric vehicle. Our current car is a 1992 Toyota Corona with 250k on the clock and starting to get some rust.
As the article points out, battery technology hasn’t followed anything like Moore’s law for silicon chips. But it has significantly improved over time (costs in US dollars).
A decade ago, a lithium-ion battery cost more than $1,100 kWh, the measure for energy density. At the time, the U.S. Department of Energy set a goal of $100 kWh, a milestone that, if reached, would elevate electrics into a head-to-head battle for primacy with combustion cars. To those like me hearing the goal at industry conferences year after year, it seemed all but absurd. Never did I hear a researcher suggest it was possible.
Yet, according to a recent study by BloombergNEF, a renewable energy research firm, we are almost there. Last year, the cost declined to an average of $176 kWh. Within five years, it will drop to under $100, BNEF says.
What is interesting about the claims in this paper is that they’re not looking so much at innovations as simply being able to consolidate all the known best practice. Even the innovations are incremental rather than revolutionary. That helps a lot for pushing these into production.
Yet there are innovations. Dahn’s primary advance — the “secret sauce,” according to Venkat Viswanathan, a professor at Carnegie Mellon — is the electrolyte, the crucial liquid that facilitates the movement of ions between the two electrodes. It is there that Dahn, adding chemicals such as methyl acetate, gains the ability to charge fast without damaging the battery.
And, to achieve the leapfrog in life, Dahn, among other things, fundamentally changed the battery structure. Current batteries tend to fracture during the charge-and-discharge cycle. But when you enlarge the crystals that make up the cathode — swapping out relatively small polycrystalline particles for larger, “single crystals” — the cracks diminish and even vanish. “Single crystals are more robust,” says Allan Paterson, head of program management at the Faraday Institution in the U.K.
The more general faster charge would revolutionise the operation of electric vehicles. Especially if the faster charging also didn’t damage the battery. That would change the design of how they are used.
I have dealt with batteries a lot directly or indirectly for most of my work over the last decade. Even the most modern lithium batteries die, mostly in response to repeated charges. Charging is hard on batteries. This isn’t hard to detect. My e-bike battery would make a good foot warmer when it is charging up for that last 20% of capacity. I haven’t looked into the hardware design of the batteries. But I’d take a bet that spilling heat isn’t a desired design feature for a long life.
But this means that a lot of the design and weight of modern lithium batteries is directly related to the requirement to have to dealing with charging heat and how to replace damaged batteries.
The battery usually doesn’t last as long as the rest of the gear. That means you have packing at the individual battery level, standardised shapes, and connector systems that are designed to removal. This is all overhead and extra weight.
There are obvious exceptions to this. If you open up modern cellphones you will see contorted battery shapes required to pack everything into a small slim shape with a large screen. But cellphones are still designed for limited life cycles and rapid change.
It gets different if you’re looking at electric car or a battery pack for a solar array. These usually currently consist of discrete batteries able to be removed individually. If you start having batteries that last longer than the rest of the gear, then there are some obvious design changes that will come. Lighter fixed installations will also make the design use of batteries a whole lot easier.
All of which means that the next decade of batteries is likely to be very interesting. As the article points out Dann is not into ‘BS’ whereas Musk has been pulled up on it repeatedly. So if this lives up to its potential in production, then it could be pretty transformational.