Section: Archived News Stories

Amprius: Building a Better Battery, from the Anode Up

September 15th, 2010 ›

By Josie Garthwaite,

Thinner electronics that can last for days without recharging and electric cars that can go hundreds of miles between fill ups: These are some of the benefits that could result if lithium-ion battery startup Amprius delivers on its promise to enable batteries with four times more energy density (the amount of energy that can be stored in a battery of a given size) compared to today's state of the art technology. The key, according to Amprius, is a silicon nanostructured anode, or a material that draws in the lithium ions when a battery recharges.

Amprius, whose investors include Google CEO Eric Schmidt, VantagePoint Venture Partners, and Stanford University, remains at an early stage of development. But it has just hit an important milestone in its journey toward commercialization. The company tells us that it has demonstrated in the lab the basic recipe, or platform, for a much higher energy density battery that can tolerate the hundreds of charge cycles needed for long-lasting electronic devices.

Amprius's Secret Sauce

Consumer electronics companies typically demand at least 500 cycles for a battery, explained Kang Sun, PhD, CEO of Amprius and former president and chief operating officer for JA Solar. Amprius has recently demonstrated 250 cycles (charge capacity drops below 80 percent after that), and plans to use this platform to achieve 3,000 cycles by 2012, Kang said in an interview at the company's Menlo Park, Calif. headquarters. That level of durability would help make Amprius's technology viable for use in vehicles.

With this platform, Amprius expects to be able to "put a protoype product on the table" by the first quarter of 2011 for device makers to consider for use in phones and other electronics. It's also a step toward batteries for electric vehicles, and potentially grid storage or military applications. Amprius has produced a total of 15,000 electrodes and batteries in its Menlo Park lab to date, and the company claims this latest design delivers four times as much energy per kilogram and twice the energy per liter compared to state of the art lithium-ion batteries.

Founded May 2008, Amprius closed a Series A round 10 months later. VantagePoint and Trident led the company's round, while Schmidt and Stanford invested smaller amounts, according to Amprius business development chief Ryan Kottenstette. (founder Yi Cui, a professor of materials science and engineering, developed Amprius's core technology at Stanford.) In addition to its venture capital investors, Amprius also received a $3 million grant earlier this year under the National Institute of Standards and Technology's Technology Innovation Program, or TIP.

Today's lithium-ion batteries, in which lithium ions shuttle back and forth between the anode and cathode during charging and discharging, mostly have graphite-based anodes. As Paul Mutolo, PhD, the director of external partnerships for Cornell University's Energy Materials Center, explained to me, graphite has been a good anode because its layers are held together by loose bonds, and lithium ions can push easily between them. Those loose graphite bonds are also "why pencils write. They're shedding layers," says Mutolo.

But improvements in energy density have slowed during the last decade - an indication, said Kottenstette, that "we've squeezed all we could" out of the conventional materials. While "hundreds of companies" are focusing on cathode technology for lithium-ion batteries, says Kottenstette, Amprius is focusing on the anode. As a result, he said, "We hold a much rarer key" for next-generation batteries.

According to Mutolo, anodes in today's lithium-ion batteries already have triple the energy density of the cathodes. In other words, currently available cathodes can only supply a third of the charge that anodes are equipped to accept - so advances in cathode technology are needed to realize the full potential of even existing anode materials, let alone the next generation of materials that Amprius and others have in the works.

Increasing energy density in batteries for electric vehicles, said Mutolo, can be thought of as increasing the size of the gas tank, in that it extends the distance you can travel between charges. Increasing power density, meanwhile, is more like adding horse power, although it also allows for faster charge times. According to Amprius, its anode alone could enable batteries with double the energy density of today's state of the art batteries. And pairing that anode with more advanced cathodes, said Kottenstette, could enable a battery with four times the energy density of today's batteries.

Nano Anode

Instead of graphite, Amprius uses nanostructured silicon for the anode material. This is not a new concept, but attempts to utilize silicon in batteries have so far been unsuccessful because silicon structures swell and crack when lithium ions move from the cathode to the anode during charging. That puts a serious damper on battery life.

Structured as nanowires, however, Cui's team found that silicon can swell without breaking. Kang compared the structure to rebar and concrete, or a fur coat with strands that are connected at the root, but still flexible to move.

Cui led research published back in 2007 that showed the potential for this nanowire structure to provide a high-capacity electrode. But that was just a starting point for Amprius. As Kottenstette explained, Amprius then had five more steps: First, it had to optimize the material by adjusting variables like porosity (spacing between the nanowires) and processing of the silicon. Second, it had to design the battery cell, which includes the anode, cathode and electrolyte. Third, the company had to optimize all of those pieces. Amprius is now about 80 percent of the way through this step, said Kottenstette.

The final two steps are designing and optimizing commercial products, including batteries for phones, and later, electric vehicles. Around 2012, said Kang, potential customers could include global automakers such as General Motors, Toyota or Nissan.

These are ambitious goals, considering that in order to dramatically reduce cost and boost efficiency, Amprius has said it needs to produce its silicon nanowires "by the mile" at a scale 1,000 times greater than the company's current operations. (The three-year NIST grant is meant to support development of a continuous manufacturing process for this purpose.)

Mutolo noted that nanostructuring is one of the "primary techniques" for manipulating materials in pursuit of greater energy and power density. But while Amprius is sold on silicon, Mutolo said it remains an "open question," whether silicon is the right material. Other companies working on advanced battery materials include Envia Systems, which is developing manganese-rich cathodes for higher energy-density batteries. With the award of a $4 million ARPA-E (Advanced Research Projects Agency-Energy) grant this spring, Envia expanded its focus to also work on a non-graphite anode for vehicle batteries.

Amprius could be a $300 million to $400 million company if it sold only to the consumer electronics market, Kang claims, but battery driven trains, fork lifts, trucks and military applications, he said also look promising. Within 4-5 years, he believes Amprius could be a $100 million business.

Indeed, Amprius has kicked into high gear since Kang came on as CEO six months ago. The company currently employs 20 people, and according to Kottenstette, most of them have been working around the clock in Amprius's 6,000-square-foot lab in recent months in order to reach the 250-cycle milestone before summer's end.

Still, a long road lies ahead. Even if Amprius hits the next milestones for proving out the potential of a battery with its technology, and overcoming the significant hurdles of setting up large scale manufacturing, actually integrating that device into a vehicle, commented Mutolo, is another lengthy and complex task.