QuantumScape says its technology is ready to move from the lab to VW’s dealerships. But this secretive startup is very familiar with failure.
April 14, 2021
The future of batteries, sought for decades by academics, startups, and corporate R&D armies is—quite possibly—just a slender sheet of ceramic material that’s supple enough to bend between two fingers. But no one outside of a Silicon Valley startup is allowed to know what it’s made of. Even the color of this man-made substance is a closely held secret, so of course it’s never been independently analyzed.
And this battery material has already made Jagdeep Singh a billionaire, long before it makes its way into a single electric vehicle. His company, QuantumScape Corp., came out of stealth mode late last year for a public stock listing backed by promising data but no commercial product and zero revenue. In no time, the company briefly surpassed the valuation of Ford Motor Co., which sold more than half a million cars and trucks in the U.S. in the last quarter of 2020.
The battery leap claimed by Singh and his team promises to extend the range of electric cars by as much as 50% over today’s lithium-ion technology, while reducing charge time for a long drive to just 15 minutes. Investors swept up by Wall Street’s mania for special purpose acquisition companies, or SPACs, seem to particularly prize battery-related startups without profits. QuantumScape’s claims are now worth about $20 billion. To understand why, you need to look at the composition of the battery itself. QuantumScape’s story is one in which hundreds of engineers and scientists have spent round-the-clock laboratory shifts reimagining every atom in the modern lithium-ion battery. Without showing anything concrete to the public until recently. Or even telling who did what.
“I don’t want to give you names,” says Singh, a co-founder who serves as chief executive officer, “because I don’t want to risk other people poaching them.” The company even required the use of a color-shifting filter before a photographer could capture images, lest the material’s true hue give away hints.
Secrecy is standard in battery development. That’s because QuantumScape is hardly alone in its quest: Startups such as Solid Power, ProLogium Technology, and Ilika, as well as large companies like Toyota Motor and Samsung Electronics, are gunning for the big, difficult prize of next-generation energy storage.
Consider that lithium-ion batteries, with modest updates, have taken over the world since their first commercial use in 1991, powering everything from consumer devices to electrical grids. As costs have fallen more than 90% in the past 10 years alone, battery performance has improved only a little bit every year. These incremental gains have come from slowly reducing the amount of materials needed to store the same amount of energy or slightly tweaking chemical composition.
QuantumScape claims to have totally swapped out two of the four main materials in lithium-ion batteries. “It’s not impossible to visualize another few decades with this chemistry,” says Singh, who aims to radically upgrade electric vehicles and extend the reach of batteries into other technologies such as flying taxis.
Perhaps the biggest obstacle to glimpsing the battery’s future, other than all the secrecy, is that few of us know much if anything about how batteries have evolved. Floor the accelerator pedal inside an EV, and what happens outwardly is pretty much the same as before. Off goes the car, minus the vroom. What’s going on within? That’s something the average Tesla superfan might struggle to describe. Which is odd, since billions of adults carry the same lithium-ion technology with us all day and sleep beside it every night.
The invisible inner life of the battery involves trillions of charged lithium atoms rushing between two electrodes—cathode and anode—in a liquid electrolyte that makes the quick movement possible. Put an electric car or smartphone on to charge, and the same thing happens in reverse. And just as tires wear down with use, a battery suffers degradation with every back-and-forth swirl of a big chunk of its constituent atoms.
The breakthrough promised by QuantumScape is backed only by limited preliminary data released since December. It’s been achieved in large part by replacing the liquid electrolyte with something solid—a long-sought accomplishment that would earn a new name: the solid-state battery. But there’s a staggering risk that comes with believing in someone else’s unseen invention. Few battery startups subject their technology to independent verification, so comparisons are virtually impossible. Cautionary tales of overhyped battery advances are easy to find. British appliance maker Dyson Ltd. had to write down its $90 million acquisition of a startup called Sakti3 Inc. that had promised a solid-state battery. Pellion Technologies Inc. supposedly created a next-generation lithium battery only to have investors pull the plug once the cost of manufacturing skyrocketed to hundreds of millions of dollars.
QuantumScape’s biggest shareholder is Volkswagen AG, which means the world’s largest automaker is lined up to become its initial customer. In March, at a splashy PR event dubbed “Power Day,” VW executives laid out a plan to electrify almost its entire fleet. The transformation will likely cost hundreds of billions of dollars. The news sent VW’s share price up 23% that week; QuantumScape’s shares also surged. Solid-state batteries are “the endgame for lithium-ion battery cells,” said VW’s battery chief Frank Blome.
A battery revolution could not come at a better time. Governments are strengthening regulations to cut emissions, and every automaker is rushing to offer electric models. “It’s very rare to take one of the largest industries and literally swap out the heart of that industry from underneath it,” Singh says. “That’s what we’re doing.”
But a solid-state advance won’t help reduce emissions unless it can enter the mass market. And that presents a second enormous chasm of innovation that QuantumScape and its investors will have to cross. Not just secretly creating of a new battery material, but also perfecting industrial-scale production that can supply hundreds of thousands of vehicles within just a few years. Botching either step would mean inventing yet another failed battery.
“Five years ago, given the number of problems associated with solid-state batteries, there was skepticism if they could ever work,” says James Frith, head of energy storage for BloombergNEF. “That’s no longer the case.”
Fundamental battery innovations take more than a generation, and the process developed by Singh’s team promises to shave off at least a few years—and possibly many more. The world will need more advances to tackle the climate challenge, meaning many technologies will have to travel the arc from laboratory to the market as fast as possible. That’s why QuantumScape’s attempt to reinvent the battery may hold real lessons, even if there’s a danger its technology won’t work out.
Singh is a slim-built former marathoner who wears a turban as a practicing Sikh. Born in New Delhi, he came to the U.S. when his father worked as a statistician for the World Health Organization. He started college at age 15, studying computer science at the University of Maryland, and began working at Hewlett-Packard’s data communications division by the time he was 19. He founded his first company six years later.
His obsession with batteries started behind the wheel of a Tesla Roadster. It was 2009, not long after the debut of the first car sold by future battery-powered billionaire Elon Musk. The number of plug-in vehicles worldwide was still in the thousands, and the only way to achieve a driving range of about 200 miles with the latest technology was to drop more than $100,000 on the car. “I kept thinking that this could be the future of the automotive industry,” says Singh, 53, remembering his drives around Northern California. “But someone has to build a better battery.”
At that point, Singh was a serial founder who’d led his fourth startup, Infinera Corp., a manufacturer of equipment used for optical data transmission, through a successful initial public offering. He decided to resign and take up the role of “entrepreneur in residence” with a venture fund run by investor Vinod Khosla, a legendary Silicon Valley figure.
To get a better grounding in batteries, Singh started attending lectures at Stanford. That’s where in 2009 he met the physicist Friedrich Prinz, whose research group works on energy issues at the atomic scale. Prinz and a graduate student named Tim Holme were working on an all-electron battery, regarded as an idea far ahead of its time. Since they were first invented in 1799, batteries have remained chemical devices that shuttle charged particles known as ions back and forth as a means of moving massless electrons outside to power devices. The Stanford duo wanted to do away with bulky ions, in a shift that could theoretically allow the storage of more energy in a smaller space. The idea was appealing enough that the U.S. government gave the lab $1.5 million in what Holme describes as a “high-risk, high-reward” research program. The money came from the Obama-era stimulus package passed in the middle of the financial crisis. Tesla Inc., then struggling to introduce its slightly less-expensive Model S sedan, would get a far larger loan from the same stimulus.
The material that was to underpin the all-electron battery was a type of quantum dot with distinctive electrical properties that result from oddities at the atomic scale. The material served as the inspiration for the name of the company Holme and Singh founded in 2010, with Prinz joining the board. QuantumScape was born with early investments from an all-star group that included Bill Gates, Khosla Ventures, and Kleiner Perkins.
The startup plowed a bit of the initial investment into a three-year lease for office space on North First Street in San Jose. “What if we go out of business in less than three years?” Singh remembers thinking. It wasn’t an unreasonable anxiety. Over the next several years, his battery startup ran into failure after failure.
To understand QuantumScape’s trajectory, we have to open the black box that is the battery.
The contents can be divided into four components: anode, cathode, separator (to stop the anode and cathode from touching, which causes a dangerous short circuit), and liquid electrolyte (through which the ions flow). This standard configuration hasn’t changed in more than two centuries; nor has the chemical composition of lithium-ion technology evolved much in three decades of tinkering. The cathode typically contains cobalt. The anode is almost always made of graphite, a form of carbon. The liquid electrolyte is often a lithium salt in a carbon-based chemical. The separator is a thin sheet of porous plastic. “Battery chemistry doesn’t change very quickly,” Singh says.
When QuantumScape began working on an all-electron battery, it set out as if from scratch. What would happen if engineers replaced all four main ingredients? This was the great promise of quantum dots that had convinced everyone from federal bureaucrats to the co-founder of Microsoft Corp. to put millions of dollars into a startup—and the theory that led Singh to Stanford in the first place.
But the big idea didn’t work, and QuantumScape discarded quantum dots within a year. Singh and Holme realized early on that trying to reinvent the battery was just too difficult. “Our mission to start the company was not based on a specific technology,” Singh says. The goal, he insists, was to build a denser, safer, faster-charging battery, no matter what it was made of.
Because the first failure came so fast, there was plenty of money left. The company had about 30 employees at that point, and the founders did what anyone in Silicon Valley would do in the face of a failed technology: have an awkward meeting with the money people, then pivot to something else. There was a less radical but still quite challenging idea they wanted to pursue.
“We told our investors, ‘Look, guys, we don’t know if this approach can be successful. But what we do know is that if it is successful, it can change the world,’” Singh says. “Luckily the core investors were visionary enough.”
By this time, in 2012, the lack of any tangible success beyond mollifying investors had the uncanny consequence of attracting even bigger investors. Prinz had brought QuantumScape to the attention of Volkswagen executives with whom he had a relationship. VW hadn’t yet been caught up in the 2015 Dieselgate scandal, in which the company admitted to installing software that let it cheat on laboratory emissions tests. The fallout led to more than $30 billion in financial penalties and would eventually force the German automaker to accelerate its transition to electric cars—so fast, in fact, that VW is set to sell more EVs than Tesla by next year.
A decade ago, however, VW had no fully electric car on offer. It was at this time that Tesla looked like it might be more than an annoyance, perhaps justifying a small bet on the battery-powered future. VW would end up investing $300 million in QuantumScape.
The next battery idea from Singh’s team was inspired by the past. The very first lithium-ion battery was invented in the 1970s by researchers at Exxon Corp., of all places. This was the era of the OPEC oil embargo, when even one of the biggest oil companies entertained doubts about the future of gasoline. Exxon’s original prototype had lithium metal on its anode. In a battery that depends on shuttling around charged lithium particles, there’s no more energy-dense anode than pure metallic lithium.
But there was an intractable problem that came with an all-lithium design. Peer into the battery through an electron microscope, inspecting the materials at the atomic level, and an all-lithium anode will likely be covered with hairlike structures called dendrites—the bane of battery engineers. Such nanostructures have the strength to puncture a battery’s thin plastic separator and reach the cathode, causing short circuits and fires. Because batteries are densely packed energetic materials, dousing a battery fire takes a lot more water and care than a comparable situation with an internal combustion engine. That’s one reason Exxon abandoned battery research and stuck to oil and gas drilling.
Researchers in the ’80s found that graphite made for a more stable anode, and it became the mainstay of the lithium-ion batteries that started appearing in consumer products such as bulky camcorders. Still, the dream of defeating dendrites and returning to lithium metal never died in academic labs. By the time QuantumScape began considering switching things up, scientists had an idea that could be the solution to the problem. That led to the next big question: Could the liquid electrolyte be replaced with something solid without adversely affecting battery performance?
Every company making a new type of batteries is, in effect, focused on novel materials. Without finding new materials, there was no way to make the battery Singh wanted. The failure of the first year hurt, but it wasn’t a surprise. Finding materials that can perform at extreme conditions is a really difficult problem.
As QuantumScape embarked on the search for a solid electrolyte, Singh and Holme went back to the drawing board. The team listed the properties the material would need to have: 1) resist dendrites and 2) enable lithium ions to flow freely. “We did not know if a material existed in nature that could meet the requirements,” Singh says. “Much less that we would be capable of finding it.”
To tackle such a challenge, scientists have one ultimate weapon: brute force. Conduct as many experiments as possible, learn from each iteration, and tweak the tests to perform yet more experiments. Straightforward, but expensive and slow. Luckily, QuantumScape had a surplus of cash. Thus began the build-out of what Singh calls one of the best materials labs any startup owns, powered with computers that could handle vast amounts of data. For a battery nerd, QuantumScape’s lab would be a coveted workplace regardless of what type of battery you wanted to research. Its suite of scientific instruments, Singh boasts, could be beaten only by what was on offer at the world’s top universities.
To cut down the time needed to make a breakthrough, if not the cost, the company turned the scientific work into a 24/7 operation. That practice continues today, almost 10 years on and with 300 employees. The process of creating a new material starts with shaping virtual prototypes inside supercomputers and testing their theoretical capabilities to a rough approximation. It’s almost like having a version of the replicator from Star Trek, capable of creating any material from a vast database—just digitally. This kind of theorizing backed by powerful computing has accelerated innovation across the physical sciences.
The computer research narrowed the list of materials that QuantumScape had to actually make, though round-the-clock lab teams still made many dozens of different materials. Singh wouldn’t give examples, but the company’s patent portfolio is littered with exotic-sounding substances such as lithium lanthanum zirconium oxide.
From 2010 to 2015, Holme says, the company ran “millions of tests” on these materials. None worked. The nanoscale hairs haunted each one. “After you see every flavor of material you’ve tried still form dendrites, it kind of affects you,” Singh says. “Honestly, there was a time when I was getting depressed. I was like, ‘Dendrites may be just one of those problems that you cannot solve.’ ”
The lab eventually got lucky. It found not one but two materials that seemed resistant to dendrites. Two teams fast-tracked work on their respective materials, and friendly competition ensued to test and refine. By 2015, QuantumScape had settled on the winner.
Singh won’t give much detail about the material apart from saying it’s a dendrite-resistant ceramic that lets lithium ions pass through “like it’s a highway.” He’s also willing to reveal that some liquid remains in the cathode, meaning his prototype is actually more of a semi-solid-state battery. Doing away with liquid altogether, if possible, would be a goal of a future iteration that might boost storage capacity even further.
Finding a material is only half the job. QuantumScape spent the next five years, from 2015 to 2020, trying to perfect it. “That may seem like a long time,” says Venkat Viswanathan, an associate professor in the department of mechanical engineering at Carnegie Mellon University, who serves as an adviser, “but it’s how long it takes to solve the ‘and’ problem.”
The solid electrolyte QuantumScape found needed to meet several “and” criteria: It had to allow lithium ions to flow and stop the dendrites; it also had to be flexible enough that it wouldn’t break inside a battery and be easy enough to produce at scale. Singh puts it another way: “You cannot have a wing of an airplane tested to the exact stress it’s going to see in real flying conditions. You’ve got to test it to much more strenuous conditions.” If you want the EV battery to be charged in 15 minutes, there better be headroom for the material to handle even faster charging without failure.
With the scientific hurdles potentially cleared, there’s still no evidence QuantumScape has figured out the challenge of manufacturing a new type of material at the scale required by Volkswagen—a company that sold more than 6 million cars in 2019—while keeping all those additional qualities. QuantumScape says it’s nearly three years away from putting its battery in test cars. Don’t expect to see it at your local VW dealership until 2026, if everything goes right.
Even in manufacturing, QuantumScape is taking on a challenge few others have dared. No U.S. startup in the past decade has built its own gigafactory, a term popularized by Musk for a plant that produces gigawatt-hours’ worth of batteries each year, at least enough for 200,000 cars. All existing large battery factories in the U.S. are either owned by Asian companies, such as LG Chem Ltd. or SK Innovation Co., or were built in partnership with those overseas giants, like the original Tesla-Panasonic gigafactory outside Reno, Nev. Prior battery startups haven’t been able to raise the billions of dollars needed to build gigafactories, says BNEF’s Frith, and instead license technology to existing producers.
But that’s not the end goal for QuantumScape now that it’s sitting on almost $1.5 billion, after having raised $450 million in March. By 2023 it plans to build a pilot manufacturing plant in San Jose, where test cells can be made for other automakers. The facility will also develop applications to reflect what Singh describes as growing interest from consumer-electronics companies and the nascent industry of flying cars. By 2024 the company plans to build a large-scale factory at an undecided location as a joint venture with Volkswagen, to start supplying batteries to some of VW’s brands, including Audi and Porsche.
“That’s what one company can do with a lot of money,” Frith says. “With the money pouring across the industry, imagine what all the other companies, whether startups or giants, will be able to achieve.”
That competition is starting to creep up. In March, General Motors Co. announced that it’s working with Massachusetts-based SolidEnergy Systems Corp. to build a solid-state battery. The same month, European battery maker Northvolt AB acquired California-based startup Cuberg Inc. Previously, Ford and BMW AG invested in Colorado-based startup Solid Power Inc. Daimler AG is working with Blue Solutions SA on solid-state tech.
And some of these rivals are making next-generation batteries without having to face the challenge QuantumScape set for itself. Cuberg, for example, has built a lithium-metal battery that may have overcome the dendrite problem for its liquid electrolyte.
There’s pressure, too, in being the largest publicly traded U.S. battery company. Secrecy becomes harder to maintain when you have to report progress every quarter and offer investors a modicum of transparency—if only about the wait until revenue materializes. After a Sanford C. Bernstein analyst warned in December that QuantumScape’s manufacturing risk is high, the company’s share price fell 24%. In February, when it announced a technical triumph earlier than expected, the price jumped 17%.
“We do not set expectations that we can’t meet,” Singh says. “Because as a public company if you miss expectations, the reaction is swift and merciless.”