Oklo launches Aurora advanced fission clean energy plant in US | Hacker News Reader

Oklo launches Aurora advanced fission clean energy plant in US

nsenergybusiness.com

122 pointstosh39 comments

Very much an aside, but I love the name "Oklo" for a nuclear power company. Oklo is a location in Gabon where they found proof of a "natural" nuclear reaction.

https://en.m.wikipedia.org/wiki/Natural_nuclear_fission_reac...

You smarty pants may have already known about Oklo, but as a kid, and even today, this information absolutely delights me.

They were originally called UPower and changed the name. I really like the name Oklo too. When they first changed it, I thought for sure it was a reference to "little tiny reactors in Africa." So far their target market seems to be rural Alaska. Also, one of the co-founders is from Oklahoma, so I think that factored in as well.

Fun fact: Oklo isn't so far, globally speaking, from lat/lon 0,0 -- a point many of us are familiar with :)

Agree about the delight of weird information :) Lookup "argyle" socks on wikipedia, it stems from the TarTar of the "Clan CampBells"

I'm not able to find much on this reactor.

It's quite small (which is a good thing for expediency and cost of getting a new design running).

It looks like they just recently got a site permit: https://morningconsult.com/2019/12/10/advanced-reactor-devel...

The OP's press release indicates their use of metal fuel (as opposed to oxide) and heat pipes.

The 1.5MW design, if it uses heat pipes, likely won't scale to a larger, higher output design without changes in the heat transfer system (at the least). But, since they're testing out new fuel, that may just be part of their long term plans...

Instead, it can scale out, by mass-producing small modules and running batteries of them.

1.5 MW is not a lot, of course, you need two orders of magnitude more to feed a large city. Still could work great for small cities, especially remote and cold, if cost of production, construction, and service would be reasonably low. That's where the economy of scale could kick in.

1.5 MW is the output of a medium sized wind turbine. Of course, the "battery" has controllable, rather than variable, output.

> The Aurora advanced fission power plant, also known as fission battery, produces approximately 1.5MW of electric power, as well as possesses an ability to generate usable heat.

> The plant utilises metal fuel to generate heat, a superior fuel type which is established with decades of experimental data.

Understandable PR move, but still bizarre that the word "nuclear" appears nowhere in the article.

> Furthermore, it demonstrated the natural shutdown behavior of the fuel, the use of a fission spectrum that can recycle fuel and ultimately convert nuclear waste to clean energy.

It does have the word once.

Also conspicuously absent: "radiation", "radioactive". I consider myself pro-nuclear, but this sort of omission tends to backfire when someone publicly draws attention to to it.

I'm impressed they went with the word battery. I would keep any hint of 'unclear' as far away as I could too.

They specifically mention metal fuel which is an industry term for a type of nuclear fuel.

Also fission. But there’s no way to get a quote that will go nookular from this.

"Oklo wanted “to channel a little bit of Apple-type vibe” and build up excitement for the plant concept, said DeWitte, Oklo’s chief executive, in an interview ahead of the permit announcement."

What are they thinking? Something like a radioisotope thermoelectric generator? Those are possible, but there's a Pu-238 shortage.

A similar announcement from General Electric in the 1950s.[1] Except that actually worked.

[1] https://youtu.be/nn1zESA0V8k

That movie is amazing, makes you want to build your own reactor.

I'm not seeing the stories on this mention the risk of widely distributing radioactive materials in these mini plants. How hard will it be for black hats to extract it for a dirty bomb? How vulnerable is it to being bombed directly? Is it cost effective to surround them with enough concrete and rebar to keep the nasty stuff inside?

They'll likely have to answer the airplane impact question directly during licensing. A combination of steel mesh structures and below-grade can help, as can the fact that this reactor is very low power and low burnup (meaning the radiological source term will be fairly low for worst-case). But it will still be important and non-trivial to physically protect the plant. This is a fundamental challenge for small reactors, and is part of the reason reactors got big in the first place.

Another article mentions their site permit is for INL: https://morningconsult.com/2019/12/10/advanced-reactor-devel...

They'll be using all the existing (and likely expanded) security and other infrastructure at that site.

If you've ever been to a running plant you would notice the high security, including heavily armed guards. You can't just walk into these places and take stuff.

The problem for a 1.5 MW commercial reactor is that paying security guards and building physical barriers very quickly becomes cost prohibitive. If you have a 1000 MW plant, it's doable. But at 1.5 MW it's really hard to run 3 shifts of guards and operators cheaply. Some have considered significantly undergrounding nuclear plants to help deal with this, but then cost of excavation gets way up there.

> The problem for a 1.5 MW commercial reactor is that paying security guards and building physical barriers very quickly becomes cost prohibitive.

It's not clear that we won't have to do this anyhow as shooting transformers with rifles is a problem, too:

https://www.stgeorgeutah.com/news/archive/2016/09/26/jla-tho...

https://www.latimes.com/nation/la-na-grid-attack-20140211-st...

there is a substantial difference in severity between a transformer being compromised and a fission reactor being compromised

Probably less than you think.

A lot of transformers used polychlorinated biphenyls (PCBs) in the insulating oil. Those tend to contaminate groundwater and degrade on the same timescales as radioactivity.

We have, fortunately, banned the usage of such things for quite a while. But, if a transformer hasn't been replaced in the last 20 years, it's probably hazardous.

When I was a teen I was working at Burger King and the power went out because somebody plinked a transformer on the West Side of Manchester, NH. We got to eat everything on the heat chute, and watched a drunk driver crash in an intersection with a non-working traffic light.

Which is why, as mechhacker pointed out, they are using a site with existing security arrangements. That’s a common approach for plans to use mini reactors like these.

Makes all the sense in the world for reactor development and testing.

For commercial deployment it'd be nice to understand the basic plan.

Could you power a datacenter with a few of these? They'd typically have security already.

Absolutley. There are a lot of good reasons to look into this. Most tech companies operating data centers want to be low carbon and need large scale, long term reliable backup power.

Good branding taking the nuclear out of fission

And adding battery haha. This is how you know wind and solar are scaring them.

“Imitation is the sincerest form of flattery.”

Rather limited information in the article. Not really useful.

It seems to me the biggest hurdle in nuclear energy is the (lack of) speed with which you can build a new plant (and therefore cost). It seems to me that any new type of reactor must be built much more efficiently in order to gain traction. I bet a new plant with "old" design which can be built in 2yrs would be much more popular than a new plant with state of the art reactors which can be built in 5 years.

That's why so many companies, like Oklo, are building small reactors that can be mass-produced in factories.

"cheap nuclear reactor" isn't the look they are going for.

I couldn’t find anything I’m the article about cost. Typically nuclear is a bit more expensive than natural gas. Is that the case here?

They probably haven't worked out all the kinks yet, so I'd expect the first few reactors to be more expensive than typical nuclear plants. They wouldn't want to advertise that.

No word "nuclear" in the article ?

There is no verified fission reaction that is used to produce energy on this planet yet, is this a scam ?

Oklo said the Aurora plant can produce power for decades without any need to refuel. It can also recycle fuel and ultimately convert nuclear waste to clean energy.

Oklo has been engaged in pre-application activities with the US Nuclear Regulatory Commission since 2016 for the Aurora design, and says it is preparing to submit its first licence application. The company has previously received federal cost-shared funding for technology development through the US Department of Energy's cost-shared Gateway for Accelerated Innovation in Nuclear (GAIN) initiative.

https://www.neimagazine.com/news/newsoklo-reveals-design-for...

Oklo faces significant hurdles on its road to regulatory approval, though. For one thing, Aurora is a liquid metal-cooled fast reactor, a design that has been used almost exclusively on submarines. “Frankly, the regulatory paradigm is built for large reactors,” DeWitte says. https://www.wired.com/story/the-next-nuclear-plants-will-be-...

> For one thing, Aurora is a liquid metal-cooled fast reactor, a design that has been used almost exclusively on submarines.

That is incorrect. We have about 450 reactor-years experience as a world in terrestrial sodium-cooled fast reactors [1]. The Seawolf was a sodium-cooled submarine for a while, but Rickover converted it to water since it was a pain to deal with sodium in an ocean environment. The Russians ran some Alfa-class lead-cooled submarines, but these are a tiny fraction of the world's liquid metal experience.

[1] https://www.iaea.org/publications/7581/fast-reactor-database...

In fact, the first nuclear electricity was produced by a liquid-metal cooled reactor called EBR-I in 1952. It's a museum now near Arco, ID, and is totally worth visiting. In the US we ran EBR-1, EBR-2, and FFTF across decades, as well as some smaller experimental reactors. France ran Phénix well and SuperPhénix poorly. The Russians ran BOR-60, BN-350, BN-600, and now BN-800. The Indians have FBTR. China is operating the CEFR. All of these are sodium-cooled.

The Oklo design is highly de-rated. If you have a fuel loading that lasts in a normal plant for 3 years at 300 MW, you can run it for 9 years at 100 MW and 90 years at 10 MW. From a nuclear perspective, the fuel wouldn't know. Except for corrosion. Corrosion will get ya with time-at-temperature. To beat corrosion, you need good chemistry control and low temperatures to operate a long-lived nuclear core.

Do you mean fusion isn't working yet? All existing nuclear power is based on fission.

>There is no verified fission reaction that is used to produce energy on this planet yet, is this a scam ?

Fission, not fusion.

Reply on news.ycombinator.com