GM has made the much-overlooked claim that by mid-decade, it expects to commercialize the most exotic of the stretch futuristic batteries currently on electric vehicle drawing boards.
In remarks yesterday at Barclay’s Automotive Conference, GM CEO Mary Barra announced an explicit campaign to capture the lead in EV development from Tesla and everyone else in the industry. That included a 35% bump in EV development investment to $27 billion by 2025 and the deployment of EVs that will go up to a whopping 450 miles on a charge, farther than any of its major competitors have discussed.
Most coverage of the presentation focused on those announcements plus the accelerated release of the electric Cadillac Lyriq to the first quarter of 2022, nine months early. But the reports that focused on these developments missed the big news. Doug Parks, one of Barra’s executive vice presidents, said that GM’s next-generation EV battery, contained in its mid-decade EVs, will have a metallic lithium anode.
This is a huge deal: Because of the gargantuan energy lift that metallic lithium can bring, it has been a holy grail of the battery world for almost a half-century. But no one has managed to figure out how to get past an unfortunate side reaction, which is that metallic lithium catches fire, grows dangerous spikes that short circuit batteries, and is generally a difficult beast to tame.
In September, the high-spending Silicon Valley startup QuantumScape went public in a reverse merger solely on the strength of its claim to have solved metallic lithium. QuantumScape’s partner is VW, which has suggested it, too, will have metallic lithium anodes in its batteries by 2025.
But when QuantumScape failed to publicly release any data, leading independent battery experts withheld any cheering until they could see the goods. Likewise now with GM.
I asked for thoughts on GM’s announcement from a few of the battery experts. One emailed back to say, “That’s bullshit.” A second, on WhatsApp, said, “That’s just crazy.”
Parks said the anode would reduce the cost of GM’s batteries well below $100 per kWh, which has been another battery industry holy grail threshold.
The tenor of the battery race is driving companies to make big moves and big claims. Ford has said it is spending $11.5 billion on electric vehicles over a five-year period ending in 2022. VW says it will spend $54.5 billion on EV and hybrid technology by mid-decade. Tesla says it’s entirely reengineering its battery and cars.
“Tesla’s got a good jump, and they’ve done great things,” Parks told reporters in a pre-briefing Wednesday evening. “They’re formidable competitors … and there’s a lot of startups and everyone else invading this space. We’re not going to cede leadership there.”
What GM is saying specifically is that by the middle of the decade, it will produce a second generation of its new lithium-ion EV battery, which it calls the Ultium. This battery will include a lithium-metal anode. It has already taken these pretty far, it says — achieving an impressive 500 charge-and-discharge cycles. Usually, EV batteries are thought to be ready for showtime when they can reliably do 1,000 charge-and-discharge cycles while retaining more or less the same capacity. Getting hundreds means it’s not quite there but is seriously robust.
To questions yesterday from Barclays managing director Brian Johnson, Parks said the anode would reduce the cost of GM’s batteries well below $100 per kWh, which has been another battery industry holy grail threshold. Parks seemed in fact to say that the cost of the entire battery pack, including all the weight of the various filler and junk that surrounds the individual cells, would be well below $100 per kWh, which, if true, would be stunning.
In response to a question I asked in the pre-briefing about the battery’s composition, Parks said the battery’s energy density would be 1,200 watt-hours per liter, a feat that experts I later spoke with had a hard time believing.
The GM battery would differ from QuantumScape’s in that it is not solid state — it uses a liquid electrolyte. Parks said there is much work left to do scaling up the battery but that “we think we are at the leading edge of this curve.”
Parks did not give a precise year in the mid-decade when all this will happen nor how many vehicles will feature the anode. In a statement, a company spokesman said there would be a “phase-in of many new technology areas related to EVs, including battery chemistry.”
When I queried Jeff Sakamoto, a lithium-metal researcher at the University of Michigan, he said that using a liquid electrolyte with metallic lithium would make it “tough” to achieve any more than a few hundred charge-discharge cycles — “insufficient for EVs.” He said the rough 2025 time frame seemed optimistic. “To me, most of the underlying physics works out on paper, but there are still some knowledge gaps that must be bridged before any lithium solid-state battery can be considered a viable solution, especially for EVs.”
Paul Albertus, a professor at the University of Maryland, said that GM would need to start production slowly because it will probably have to change the process it’s using to manufacture its current lithium-ion chemistry. “I think there’s also an aspect of starting small to see if any surprise field failures or challenges arise — put out a small vehicle run and wait a year or two and make sure no surprises happen, that type of thing,” he said.
Albertus said he wouldn’t be surprised if GM totally misses the deadline and never produces a metallic lithium battery at all. But he added that he also wouldn’t be astonished if GM succeeds. “What makes me a little more optimistic is that we know there are some very large efforts working on this now,” he said. “… With automotive batteries growing so much in the coming 10 years, there’s now motivation to tackle something of the complexity of [metallic lithium]. There’s also been a lot of really good science finally getting done in the past few years, and that does help people innovate.”