Why This Ultra Dense Battery Breakthrough Matters

It might surprise you to learn that the basic chemistry of the lithium ion battery at the heart of a brand new Tesla or iPhone hasn’t changed all that much in the last 30 years.¹ So, when several of you left comments pointing me in the direction of a new company that’s replacing a key part of the battery with silicon and some nanowires, my curiosity was piqued. To add to that, one of my science advisory team members brought them up, too, which only added fuel to the curiosity fire. Now, we’ve covered a lot of batteries on this channel, so what makes the company Amprius, and other similar companies going after silicon, stand out for the future of battery tech?

Silicon batteries really piqued my curiosity to try and learn more. So to get right into it, why is Amprius’s approach with simple silicon and some nanowires standing out in the world of batteries?

Amprius, which is a Stanford University startup based out of Fremont, California, is building on the tried and true lithium ion battery idea by replacing one small but important battery component with a very common material — and it’s resulting in massive improvements. What component is that? Well, the graphite or carbon anode of a lithium ion battery is critical, practically the “heart” of the battery. While it allows the whole thing to function, it’s not very efficient.²

In a lithium-ion battery, an electrolyte carries positively-charged lithium ions from the carbon or graphite anode to the cathode. Important to note: both the anode and the cathode store lithium.³ It takes six graphite atoms to store just one lithium ion during charging, through a process called intercalation.⁴ This is a pretty major bottleneck that limits the final energy density that a lithium cell can achieve.

For smaller applications, such as wearable tech or your phone, this might not be an issue. But when it comes to something like an electric vehicle (EV), limited lithium storage starts to dredge up bogeymen like “range anxiety” and long charge times. These problems are often exaggerated by EV critics, but they’re still real problems. When we try to go for even bigger applications, like rechargeable semi trucks or commercial planes, these setbacks only become more pronounced. Plus, the increased size and weight of these larger vehicles generally makes the battery’s job much harder.

Amprius and others have identified the graphite or carbon anode as the squeaky-but-easily-fixable wheel here, and settled on silicon as the perfect fix. Silicon is much, much better at storing lithium than graphite, with a single silicon atom capable of storing four lithium ions.⁵ That alone means that silicon anodes are up to 24 times more efficient than a typical graphite anode. It also allows silicon anode batteries to be nearly 10 times more energy dense than lithium ion batteries with conventional graphite anodes. More density equals less battery. In other words, less battery weight to achieve the same performance.

But that of course raises the question: why isn’t everyone using silicon?⁶ I know I was asking that question. Well first off, silicon is maybe a little too good at storing lithium. It has a tendency to grab so much lithium that it can increase in volume three to four times during charging. After just a few charge/discharge cycles, this swelling can cause the anode to crack and eventually destroy the battery. As a result, we usually only see a little bit of silicon mixed into our standard carbon anodes, just enough to up the efficiency without creating those swelling issues.⁷ We’re seeing this across the industry right now. Tesla spoke at length back in 2020 at their battery day event on how they’re working to integrate small amounts of silicon into their newest batteries.

We’re talking a lot about Amprius, but I was really curious who else out there is going down a similar path with silicon. That’s what got me curious about this topic in the first place. If this really is a good path for batteries, then you’d expect to see a lot of people and organizations investigating the same thing. Well, Amprius is definitely not alone.

Coming out of California, there’s OneD from Palo Alto and Sila Nanotechnologies in Alameda. Sila’s actually building a factory in Moses Lake, Washington, which puts them not too far from Group 14 Technologies over in Woodinville, who are also using silicon anode technology. These groups all have slightly different spins on the silicon anode, and they’re already seeing commercial successes. We’ll talk about them more in a bit, but suffice to say, there’s a growing silicon consensus.⁸

How are Amprius and the others getting around the swelling issue?

Nanotech.⁹ Amprius grows silicon nanowires directly from the current collector, then coats them in silicon to form an anode that almost looks like a forest full of trees. The result is a geometric structure that’s nearly 100% silicon. This makes it much easier for ions and electrons to move as they can travel a shorter distance, which improves charge and discharge times. The nanostructure also provides more surface area than graphite, which further improves the energy density. And of course, it helps deal with the hungry-hungry-silicon issue, too. By using a nanowire form factor, there’s enough space in between those wires to accommodate volume expansion, making the anode much more durable. Not to mention the thinner profile makes it flexible enough to bend with the substrate.¹⁰ It’s really clever.

Normally, these nanotech solutions are difficult to scale. After all, we are talking about silicon nanowires the size of a few microns — several times smaller than the width of a hair. Plus, if you want to reap all the amazing benefits, you have to find a way to make these “small” structures at a “large” capacity. Amprius hopes to solve this issue by partnering with the Dutch company, Centrotherm. They have decades of experience scaling materials for the semiconductor and solar industries.¹⁰ Their pilot-scale process is trying to achieve lower production costs and higher throughput with roll-to-roll manufacturing.

So Amprius’ battery is energy dense, reliable, and potentially cheaper, but how do the numbers work out? It looks really good on paper. Amprius’ battery has an Energy Density of 450 Wh/kg, or 1,150 Wh/L. That’s 50% better than some of the best lithium ion batteries on the market.⁹ Amprius also claims their power density is high, up to 10C versus lithium’s usual 3C.

The C-rating is a measure of how quickly a battery can be safely charged and discharged. The higher the C-rating, the faster the charging and discharging rates. This not only affects the time spent charging the battery but how quickly power can be released for intense applications. For example, a regular flashlight can operate with a low C rating, as it operates at a low discharge rate over a long time period. A performance or racing drone, which needs quick bursts of energy needs batteries with a much higher C-rating.

And, best of all for the average consumer, Amprius’ battery can hit an 80% charge in a little under six minutes. One of the most common (and realistic) complaints about EVs is that it takes much more time to charge a battery than to fill a gas tank. If the Amprius battery can charge in just six minutes, that delta between refueling a gas car and recharging an EV is getting pretty small!⁹⁹

Engineering aside, silicon has some economic benefits too. Unlike some of the other proposed battery materials, silicon isn’t an expensive or difficult-to-produce nextgen alloy. It’s actually the second most common element on earth. This should help bring down battery costs and ensure there aren’t supply chain issues.¹¹ For comparison, 90% of the world’s graphite comes from just one region in China, which could present a huge supply chain vulnerability.¹²

Silicon may have an environmental edge too, as recent studies have shown that graphite mining and smelting aren’t the best for the environment. Smelting graphite is a very energy intensive process that still relies on some of the dirtiest fossil fuels. The studies suggested that the emissions from these processes are 4 to 10 times worse than we previously thought.¹³¹⁴ So yeah, maybe we should be bringing on the silicon anode batteries, and quick!

Applications, Implications, Drawbacks

But ultimately, why does any of this matter? What can we do with radically more energy dense batteries? Well, as we’ve hinted at a few times, this could be a great leap forward for EVs. In terms of vehicle performance, an Amprius’ EV could charge in minutes and have potentially faster 0-60 performance times.¹⁵ It might even open the door to larger EVs like semi trucks and commercial flight, but that’s speculative for now.

And while it’s not as exciting as high performance EVs or some other possible applications, the silicon anode battery could be very convenient for wearables and consumer electronics. Just imagine a phone or laptop that charges to full in minutes and a charge that lasts far longer.⁹

Even with all of these possibilities, Amprius’ current focus is on unmanned aerial vehicles. Most consumer drones have a 20 to 30 minute battery life, and Amprius claims it can double that. This is great not just for drone hobbyists, but it could really make a difference for search and rescue or research applications.⁹ Then there’s Airbus’s Zephyr S, a High-Altitude Pseudo Satellite (HAPS), sort of a drone-satellite hybrid. With Amprius’ batteries onboard, the Zephyr S flew more than 25 days, setting a new endurance and altitude record for stratospheric flight.¹⁶ Amprius is also touting some successful uses of their battery in electric vertical takeoff and landing (eVTOL) aircraft.¹⁷ I have my doubts that eVTOL air taxis will ever really be viable, but it’s a cool demo of Amprius’ tech, and could have implications for electrifying commercial flight way down the road.

This begs the question, “If Amprius can electrify some cutting edge aircraft, then why isn’t your laptop or car currently powered by a silicon anode battery?” Well, the nanotech that allows Amprius’ batteries to bypass the normal silicon swelling issues could ironically be its greatest weakness. Silicon anodes are currently more expensive to produce than lithium anodes because they require specialized equipment and techniques.¹⁸

And as it turns out, the creation of nanowires is much more complicated than making standard electrical wire. Standard electrical wire is drawn through dies to achieve smaller gauges. Nanowires must be grown directionally from vapor, sol-gel or electrochemical processes, not drawn like regular wire.

So, while silicon may be plentiful, the manufacturing process is the limiting factor that could make these batteries too expensive to see use outside of the most extreme situations … for now. And while Amprius’ batteries have performed well in lab and field tests so far, nanotech can be difficult to scale up from lab to meaningful real-world applications. As we noted earlier, Amprius is actively trying to address this with Centrotherm’s help and their pilot-line roll-to-roll process. But can they do it affordably over time? That remains to be seen.

Furthermore, Amprius’ tech requires proprietary anodes that are currently not compatible with existing large EV cell factories.⁸ This is partly why Amprius’ current strategy is to focus on niche markets like unmanned aerial vehicles. If that’s successful enough, they can ramp up production, bring down costs, and start to expand into the consumer EV market. Even if everything works perfectly, it will be a while before you can purchase a long-range, fast charging EV with an Amprius battery inside.¹⁹

Outlook

Still, Amprius’ battery sounds very promising… too promising. Is this a “too good to be true” situation? For the concept of the battery, probably not. Their tech has been vetted by Mobile Power Solutions, a leading battery testing house, but it’s going to take a few more verifications from national labs or universities before I can fully move it out of the “wait and see” category.¹⁹

Even with further testing required, things are looking good for Amprius so far. The company is gearing up for mass production by building a 103 acre campus, including an existing 1,278,455 square foot facility in Brighton, Colorado. This factory is expected to greatly increase Amprius’ manufacturing capacity, with an initial 500 megawatt-hours (MWh) and the potential of up to 5 gigawatt-hours (GWh) available within just the existing footprint. Their goal is to be up and running within a couple of years.²⁰

As I mentioned earlier, Amprius isn’t the only company seeing success with silicon and nanotech. OneD has partnered with GM, and are also looking at nanotech solutions for silicon anode batteries. Together they’re creating GM Ultium silicon anode battery cells. Sila Nanotechnologies have had their batteries inside Whoop’s fitness tracker wearable since 2021. Their more powerful battery will come standard in the Mercedes G-Class SUV starting in 2026. Also, Group 14 Technologies should have its silicon battery powering Porsches sometime next year. All these companies are in the process of preparing for mass production, which is always a good sign.⁸

Will silicon anode batteries change the world? It looks like we won’t have to wait very long to find out.