Articles are making the rounds on social media claiming a new battery technology that can make batteries that will last 28,000 years on a single charge. There is some truth to these claims, but they are mostly misleading. They make some unwarranted claims and leave out some critical context. This is likely mostly corporate self-promotion and fishing for investors, but what is the real science behind the claims?

Here is a typical quote form the popular press:

While they would be undeniably useful in EVs, their long life also makes them perfect for devices like pacemakers.

The company claims that its technology can be scaled up, and it could be used to make battery packs suitable for an electric car that lasts up to 90 years.

It is completely deniable that they would be useful in EVs, and I am always skeptical of claims that a technology can be “scaled up”. That should never be taken as a given, and is often the deal-killer with new technology. But let me review what these batteries actually are and then put the claims into context.

The concept was first introduced in the UK in 2016 – the idea is to encase nuclear waste in diamond so that the beta-decay of the waste interacts with the carbon in the diamond to generate a small electric current. The version being presented as a nuclear diamond battery (NDB) uses carbon-14 from graphite reactor rods as the source of beta decay. This represents a small percentage of the radioactivity of nuclear waste, 95% of which is the spent fuel itself, but the graphite waste is now also radioactive and there is a lot of it, about 250,000 tonnes world-wide.

The idea behind NDBs is that you make artificial diamond out of the carbon-14 from waste graphite from nuclear reactors. This would have the side benefit of taking care of some of the nuclear waste stream. Beta decay from the carbon-14 within the diamond would interact with other molecules causing the release of electrons and the generation of a small current. The entire thing could be encased in non-radioactive artificial diamond, which would prevent the escape of radiation and also protect it from damage.

This concept appears to be legitimate. Carbon-14 has a half-life of 5,700 years, which means that these NDBs will be generating electricity for a very long time. This makes them perfect for applications that requires a long life without maintenance or recharging. Satellites and deep-space probes, for example – once they are out there we are not going to be replacing the batteries. But also, pacemakers and other implantable devices need a very long battery life. However, the engineers report that they use a variety of isotopes and the life of the batteries will vary based on applications, from 9 years for cell phones, to 90 years for vehicles, to 28,000 years for pacemakers.

This all sounds great – but there is a critical factor left out of much of the reporting I have seen. What is the power density (electricity produced per volume) of these devices? I know from previous reporting back in 2016 that the power density is extremely low, much lower than chemical batteries like lithium-ion. The engineers from NDB admit their power density is about the same as other nuclear diamond technology.

For example, one prototype recently reported:

The battery power density of 10 μW/cm³ and specific energy of 3300 mWh/g were achieved due to cell thickness decreasing.

Ten microwatts per cubic centimeter is not a lot of electricity, but it’s not nothing either. Clearly, you won’t be powering a cell phone, let alone a car, with such a power density. So what is this company talking about? While I have yet to see an interview or report that says so explicitly – the nuclear diamond battery must be incorporated into a regular chemical battery, like a lithium-ion battery. This actually makes perfect sense, and is a great idea. So the chemical battery provides the power density and the output to power the device, and the embedded NDB slowly recharges the battery. The company claims – “With the same size battery, it would charge your battery from zero to full, five times an hour.” This sounds like a claim that needs to be verified, and seems to be out of proportion to the typical power density of such devices.

But even if recharge is much slower, that could still extend the life of a single charge of your cell phone significantly, and also self-recharge over night. You would never have to plug your phone in or worry about it going dead while traveling. The same concept could apply to electric vehicles – the batteries could be self-charging, extending their range and self-charging over night or during the day while you’re at work. But again – it all comes down to how much electricity with the NDB actually provide. Right now we are just getting a “proof of concept” report from the company, and promise of a commercial prototype 6-9 months after the pandemic shutdown ends. I am also not seeing any hard power density numbers.

Even if, however, the power density turns out to be too low to be useful for some applications, it might still be a good idea to mass produce NDBs in one form or another. First, this is a safe way to deal with nuclear waste. Right now we have nuclear waste sitting in pools or deep storage, just wasting their significant energy. If we could recycle that waste into safe diamond batteries that also produce electricity, that seems like a win-win.

Even with low power density, we could theoretically fill a warehouse-sized building with millions of NDBs and hook them up to the electrical grid. This would provide steady power for thousands of years. Even if this is a small percentage of our power needs, it’s something. It can also be useful for remote areas that are difficult to get energy to, or where the power lines would be vulnerable. A station in Antarctica, for example (or on the Moon), could have a building packed with NDBs to provide some of its power, and also serve as reliable emergency power.

Probably it will all come down to cost-effectiveness. The NDB company is claiming that their batteries are cost competitive to lithium-ion. We’ll see – we don’t even have a commercial prototype yet. But it seems likely that there will be applications where this technology is cost effective (such as use in space). The question is – which applications will be cost effective, and where will the cutoff be? Will it be cost effective to use this technology for grid power?

I would love to see some calculations by someone with more technical expertise. How large would an NDB have to be to power a typical home? If we converted all nuclear waste we can into some form of NDB, how much power would that generate? How much would the electricity cost? How do we factor in the saved cost of storing and dealing with nuclear waste? What is the net carbon footprint? How close are we to needing to answer these questions?

In any case, this seems like a technology to watch. It may only every have niche applications, but there does seem to be some potential here. But the hard numbers will tell the tale.