By Kyle Proffitt
January 16, 2025 | AABC 2025 was held last month in Las Vegas, Nevada. Over four days, we heard from auto OEMs, battery startups, academics, and other industry professionals, with a central focus on the electrification of automotives. The keynote speakers this year included Kurt Kelty of GM, Colin Campbell from Redwood Materials, Jeong Hun Seo representing Hyundai, Rachid Yazami from KVI, and Emeritus MIT professor Donald Sadoway. In part two of our coverage (see part one here), we hear from Yazami and Sadoway about better charging protocols and batteries that don’t use lithium.
Rachid Yazami, KVI: Non-Linear Voltammetry for Better Charging; Lithium Dimers and Trimers
Rachid Yazami, inventor of the graphite anode, Founding Director and CTO of KVI PTE LTD, based in Singapore, came to AABC 2025 with a heavy focus on non-linear voltammetry (NLV) charging, which he described as a “disruptive technology”. It’s a topic he has been passionate about for years. Unlike traditional charging, which uses constant current, with NLV, you control the voltage. “We apply a series of constant voltage steps, and we follow the currents—the current is never constant,” he explained. “The current that is flowing in the battery is what the battery can afford at that level, at that voltage.” Another fundamental feature of the NLV protocol is inclusion of 2-second rests, during which no current flows, and Yazami says this allows battery cooling and prevents dendrite formation. Determining the voltage for a subsequent step is a function of the current; “when the current drops too much, we go to the next step.” Yazami says this protocol works well with both NCA (nickel-cobalt-aluminum) and LFP chemistries.” He showed the ability to charge a Samsung 30T, 3000-mAh NCA cell 0-100% within 10 minutes using this protocol, and a 2500-mAh LFP variant could do the same in 6 minutes. The battery temperature stayed under 45 °C.
There was a surprise, though, when they tested different charging speeds. “The energy density is much higher when I charge faster in 10 or 15 minutes than when I charge in 60 minutes,” he said. That was also true for discharge; you release more energy with a faster discharge. Yazami says this led to collaboration with colleagues at Caltech, particularly Bill Goddard, to run theoretical calculations and better understand the behavior. The result, he says, “is really a breakthrough in science.” At the anode side, with these fast charge conditions, “lithium forms dimers and trimers inside the lithium layers,” Yazami said. Additionally, within the graphite, between the graphene sheets, he said the interlayer spacing is greater than 3.71 Angstroms. “When the dimers and trimers are formed, there is a stress on the graphene layer, so they expand … the door to extract lithium out is much wider, so we have more energy when we discharge the battery faster,” Yazami explained. He says this work has been submitted to “one of the prestigious journals” and should be published soon.
Yazami said their NLV protocol is also beneficial for estimating state of charge and state of health. They can measure a pseudo-open circuit voltage (pOCV) during the zero-current rest periods, and with some mathematical transformation, achieve 99.7% accuracy in measuring charge level. The is especially useful for LFP, which has a pretty flat open circuit voltage profile that makes charge estimation during operation difficult. One more advantage: Yazami says the pOCV also changes as a function of temperature, and by measuring over weeks and months, a pseudo-entropy value can be obtained. Modeling indicates that this measurement can be used to detect and predict internal short circuits, which would be very useful in preventing thermal runaway events.
Donald Sadoway, MIT Emeritus: Outside the (Lithium) Box
“I’m going to invite you to think about something besides lithium,” Sadoway began his presentation. He reminded the audience that there are many additional elements in the periodic table. While he promised that his work was relevant to transportation, his talk focused on grid energy storage, a necessary component to address the intermittency of renewable energy options.
He poured some cold water on humanity’s collective efforts toward climate change mitigation, however, saying that “2050 is not going to be net zero … we’re going to be burning hydrocarbons well beyond mid-century.” Still, to move it forward, he said, “the key enabler is massive energy storage.” More batteries, and for that, we’re going to need new technologies, not just lithium. He added that fires associated with batteries, such as those that have occurred at Moss Landing, are a “misapplication of lithium” that we should avoid going forward.
However, “if we try to get to the new technology, we’re going to have to get involved in things that are resource intensive, going into mining and smelting … these are topics that nobody wants to talk about [at a dinner party],” he said. Sadoway focused on two primary flavors of energy storage technologies he had worked on: liquid metal batteries based on magnesium-antimony and a similar aluminum-sulfur battery. He cautioned, as with over-reliance on lithium, that we need to think differently. “Everybody else starts with the right circular cylinder that’s big enough that I can hold it between my index finger and my thumb and tries to figure out how to make something that will be the size of this room. To which I say that’s kind of stupid, isn’t it?”
Inspiration for his ideas came from looking at an aluminum smelter, a giant facility (think 100 foot wide by half-mile long building) that uses electricity to cheaply convert aluminum oxide to aluminum. Could some of the same principles be applied to store and retrieve electricity, he wondered, and his liquid metal batteries, which “work at 10 MWh and up,” were born.
Sadoway explained the principles upon which these batteries function. They use electropositive magnesium on top, electronegative antimony on bottom, and a molten salt in between (the battery must be kept above 700 °C to maintain operation). These materials “stratify, like salad oil and vinegar, by density,” he said, and they don’t mix. Magnesium would like to alloy with antimony, but it can’t get through the molten, ionic salt. However, “magnesium is smart,” he says. It loses two electrons, which go through the external circuit to power the grid (or whatever else), and then it can transit the molten salt and alloy with antimony. Over time, the magnesium layer shrinks and the alloy layer thickens, but passing current back into the cell is an electrometallurgical step, like the aluminum smelter, that pulls the magnesium away. There’s a big benefit compared to lithium-ion, in that inefficiencies that generate heat are welcome. “In this case, it’s an advantage,” he said. “We trap that heat … And then with appropriate insulation … by using it, discharging, charging every day, you keep the cell operating.”
The idea worked well. The company Ambri was formed and gathered several billion dollars of orders, but “early this year, there was a spike in the antimony price from $10,500 a ton to $50-60,000 a ton, at the same time as LFP prices are crashing,” Sadoway said. The company folded.
Undaunted, Sadoway pursued another route, this time chasing the cheapest materials. “If you want to make something dirt cheap, you make it out of dirt,” he said, and aluminum-sulfur batteries were created.
In this case, aluminum metal is the anode. “I chose aluminum because it’s the most abundant metal. And what’s the most abundant non-metal? It’s sulfur, so I made those my bookends and then searched for an electrolyte.” The principle is the same, aluminum would like to bond with sulfur to form aluminum sulfide, but a molten alkali chloride/aluminum trichloride salt layer prevents access. However, like magnesium, aluminum can give up its three valence electrons to an external circuit, and Al3+ can transit the molten salt to bond with sulfur. In this case, it is not a liquid metal battery. “We got this down to below the boiling point of water,” Sadoway said. He explained that this was possible because AlCl3 in combination with alkali chlorides forms chloroaluminates, such as AlCl4–, Al2Cl7–, etc., which create Al-Cl-Al networks, greatly reducing the melting point while being really good at passing aluminum ions.
The end result is very cheap energy storage. Sadoway says the “cell costs $9 a kilowatt hour … that’s a little bit less than lithium ion.” At best, LFP batteries can be obtained in the $50/kWh range. Performance is pretty impressive, too. “This thing you can run over a 40 hour discharge if you want, or we found it could run at very high C rates, 20C, 20D.” The original report showed 500 completed cycles at 10C charge and D/2 discharge.
Sadoway brought the conversation back to automotives by divulging how they were able to “sneak” a little comment past the editors in that Nature article. “It has not escaped our notice that its immunity to thermal runaway and fire makes this battery chemistry especially attractive for electric vehicles,” it reads. Nobody has accomplished this feat yet, however. According to Sadoway, these batteries are expected to be useful in the 100-kWh range, for operation in individual family homes and similar. Another company, Avanti, was formed from this concept. “But within 2 years they walked away,” Sadoway said, indicating that while the idea is sound, external forces and bad management can derail an operation. Sadoway is looking beyond the setback, however, and closed with encouragement for others, saying that, “batteries are going to be a thing for a long time”.
