Where Was the Battery at Tesla’s Battery Day?
ON TUESDAY AFTERNOON, Elon Musk greeted several hundred investors sitting in their Teslas from a makeshift stage in the parking lot of the Tesla factory in Fremont, California. After months of Covid-induced delays, it seemed like an appropriate setting for the company’s much-hyped Battery Day event. Details about what the outspoken CEO had in store were scarce leading up to the day, but Musk had promised to show the world something “very insane” that would result in a “step change in accelerating sustainable energy.”
This turned out to be a fat lithium-ion battery called a 4680—a reference to its diameter, 46 millimeters, and its length, 80 millimeters—that is being produced in-house at Tesla. To be sure, Tesla’s new battery appears to offer large performance gains in a few key areas, but it was unclear whether Tesla has actually achieved these upgrades or whether this is the projected performance for the finalized battery.
Neither Musk nor Drew Baglino, the senior vice president of powertrain and engineering at Tesla, who shared the stage, offered specific numbers about the new 4680 cell’s actual performance achieved during tests, only relative percentages of improvement compared to existing Tesla batteries. They claimed to be “unlocking” up to a 54 percent increase in the range for Tesla vehicles and energy density for its energy products like the Powerwall. (A battery’s energy density describes its capacity to do work given its weight or volume.)
Baglino and Musk also claimed a 56 percent reduction in the dollars per kilowatt hour compared with Tesla’s existing cells but did not name a dollar figure. Battery experts say that $100/kWh is necessary to make electric vehicles cost competitive with gas guzzlers, but it is unclear if Tesla hit this mark, since the company doesn’t publish the dollars per kilowatt hours for its existing batteries. Analysts estimate that Tesla achieved somewhere around $150 per kilowatt hour last year, which means that its new battery may have broken through that barrier.
Baglino said that the company had already produced tens of thousands of batteries at its new production facility down the street—but not a single physical cell made an appearance at the event. “I want to stress that this is not just a concept or a rendering. We’re starting to ramp up manufacturing of these cells at our pilot production facility just around the corner,” said Baglino. Both Musk and Baglino acknowledged that Tesla engineers are still in the process of refining the manufacturing process, so it’s possible that the cells currently coming off the line don’t quite meet this mark. “We’re still ironing out the kinks,” Baglino said. “It’s super demanding, because every atom has its place, if you want to deliver the energy density and the cycle life. We’re confident we can get there, but it will be a lot of work along the way.”
Based on the digital mockups shown at the event, the new 4680 cell is a lot different from the lithium-ion cells currently used in Teslas. For starters, it’s big. The diameter of the cylindrical cell is twice as wide, and it is 14 percent longer than the largest batteries that power Musk’s electric empire today. Altogether, that makes its volume about six times greater than that of the Panasonic 2170 cells used in newer-model Teslas. The upshot of its size is that it increases capacity while reducing the number of cells needed to provide a given amount of power in a battery pack. According to Baglino, the larger form factor alone was enough to boost the energy by five times, the power by six times, and the range of a car using these batteries by 16 percent. Baglino didn’t elaborate, but presumably this is relative to the current batteries used by Tesla.
“Large format reduces all the ‘inactive’ materials, like packaging. So pack-level energy density will improve and cost will come down,” says Shirley Meng, a materials scientist who runs the Laboratory for Energy Storage & Conversion at the UC San Diego. It’s exactly the sort of beefy power supply that Tesla will need for its planned heavy-duty vehicles like the Cybertruck and its electric semi. But the real innovation in Tesla’s battery is what you can’t see.
In all EV batteries, a thin layer of copper foil serves as a current collector for the anode and a layer of aluminum foil for the cathode. A tab is joined to each of these current collectors and serves as the battery’s connection to the outside world. But these tabs hobble the performance of the cells and make them more difficult to produce. Manufacturers must use a specialized welding technique to connect them to the foil current collectors, which results in wasted time and material. Even worse, the tabs reduce the battery’s efficiency, because electric current must travel the full length of the electrode to reach each tab. One of the major innovations in Tesla’s new battery is that it is tabless.
“Taking the tabs out of the equation will allow you to have more coated area on your electrode, increasing the capacity of the cell without changing anything else about the design,” says Greg Less, the technical director of the University of Michigan’s Battery Lab. “It's not a new idea, but there are a lot of engineering challenges to making something like this work reliably and reproducibly.”
The key to Tesla’s battery breakthrough was an image emblazoned on the black graphic T-shirts worn by Baglino and Musk: What looked like a bunch of random white lines was, according to Musk, a “very esoteric” representation of a tabless battery’s structure. According to Baglino, the company laser-patterned existing foil current collectors so that they can have dozens of connections to the electrode materials. The result is a current collector that Baglino described as a “shingled spiral” that looks a bit like a curled-up, copper-plated armadillo.
“The distance the electron has to travel is much less,” Musk said. “So you actually have a shorter path length in a large, tabless cell, then you have in a smaller cell with tabs. So even though the cell is bigger, it actually has a better power-to-weight ratio.”
Researchers at Tesla also significantly altered the chemistry of their electrodes to boost performance. Tesla is now one of several manufacturers producing silicon-rich anodes, which are meant to supplant the more common graphite anodes used in lithium-ion batteries today. When a lithium-ion battery is charging, lithium ions flow to the anode, displacing electrons and creating an electric charge. Compared to graphite, silicon can absorb a lot more ions.
“Silicon can store roughly 10 times the number of lithium atoms as graphite, which gives silicon-containing batteries 20 to 40 percent higher energy density,” says Francis Wang, the CEO of NanoGraf, a company that has developed a silicon-graphene battery anode. But silicon can also cause the battery to swell like a balloon. Over time, wear and tear on the components will make performance plummet. Safely incorporating silicon into an anode typically requires nanoengineering the cathode components in ways that contain the swelling. Nanograf, for instance, wraps its silicon-based anodes in graphene, which acts like a flexible blanket when the anode expands, protecting it from corrosion.
Tesla has been using small amounts of silicon oxide in its anodes for years. This silicon blend effectively comes prepuffed to reduce damage from swelling, but this means it also enables only modest performance gains. To get a bigger boost typically requires nanoengineering the particles of silicon to retain their benefits as lithium ion sponges while reducing the risk of destruction. But Baglino and his team opted to take a simpler route. Their silicon anodes aren’t nanoengineered at all; they’re raw silicon stabilized with some elastic ions. According to Baglino, this can increase the range of Tesla’s vehicles by 20 percent.
“The first silicon anode companies were started as far back as 2006, so folks were thinking about it even back in the Roadster days,” says Gene Berdichevsky, who was Tesla’s seventh employee until he left the company to found SilaNano, which is developing pure silicon anodes. “It’s taken them longer to get there than anyone thought, but it was pretty obvious even back then that that was where the biggest gains can be had.”
Tesla is also working on improvements to its cathode chemistry. Most lithium-ion batteries found in electric vehicles today—including Tesla’s—use a cathode made from lithium nickel manganese cobalt oxide, otherwise known as NMC. But Musk and other manufacturers have indicated their desire to ditch NMC for cathodes that don’t use cobalt, which is one of the most expensive materials in an electrode and is also linked to unethical mining practices in the Democratic Republic of Congo. The idea has been to use a nickel-rich cathode instead, which is cheaper, lighter, and, depending on how it's sourced, better for the planet.
Although cobalt-free cathodes have been around for a while, they typically come with large performance trade-offs, like shorter battery life and reduced energy density. “Increasing nickel is a goal of ours and really of everybody in the battery industry,” Baglino said. “But one of the reasons cobalt is used at all is because it's very stable.” He indicated that Tesla is working on developing a nickel-rich cathode with zero cobalt by using “novel coatings” but didn’t say how close the company was to achieving that goal or detail what those coatings would be.
Baglino also said the company is exploring different cathode options for different applications. For instance, some batteries would use iron cathodes for vehicles that have shorter range, nickel manganese cathodes for medium-range vehicles, and then high nickel cathodes for long-range applications like an electric semi. Musk suggested that the reason for this is supply chain uncertainty. “We really need to make sure that we’re not constrained by total nickel availability,” Musk said. “I spoke with the CEOs of the biggest mining companies in the world and said, ‘Please make more nickel.’ So I think they are going to make more nickel. But I think we need to have a three-tiered approach to making batteries.”
Tesla’s entry into the world of battery production is also notable. For the company’s entire history, it has purchased the batteries at the heart of its cars and Powerwalls from other companies like Panasonic and LG. At Tuesday’s event, Musk confirmed the existence of Tesla’s fabled Roadrunner project—previously reported by the EV blog Electrek—which has spent the last few months standing up a pilot production facility at the Fremont factory. Although Musk acknowledged that Tesla would continue to purchase batteries from Panasonic and other suppliers, he said that on their own these companies wouldn’t be able to meet Tesla’s growing appetite.
According to Musk, the projected annual output of Tesla’s Roadrunner project production facility, in terms of the energy capacity of the batteries, is about 10 gigawatt-hours per year. But the big question, says Berdichevsky, is how fast Tesla can scale production to that level. “You can show a widget that's high-performing, but that manufacturing scale is really, really hard,” he says.
Musk himself acknowledged, as he often does, the difficulties that come with “building the machine that builds the machine,” a nod to the fact that Tesla will have to create a bunch of battery bots to make it happen. He says he expects it to be “about a year” before Tesla’s new battery factory is producing cells at its full capacity—but even that is a remarkably fast turnaround by the industry’s standards.
Musk is known for making big promises on aggressive timelines and then blowing right past them. The battery he hyped at the Tesla event is full of promise, but without an actual cell to show off or published performance metrics, it still felt like just a promise. Now he has to deliver.
