Datacenters Line Up For 750MW of Oklo's Nuclear-Waste-Powered Small Reactors (theregister.com) 62
Datacenter operators are increasingly turning to small modular reactors (SMRs) like those developed by Oklo to meet growing energy demands. According to The Register, Oklo has secured commitments from two major datacenter providers for 750 MW of power, pending regulatory approvals. It brings the firm's planned nuclear build-out to 2.1 gigawatts. From the report: Oklo's designs are, from what we understand, inspired by the Experimental Breeder Reactor II (EBR-II) and utilize liquid-metal cooling. They are capable of producing between 15MW and 50MW of power, depending on the configuration. That means Oklo's datacenter customers plan to deploy somewhere between 15 and 50 of the reactors to satisfy their thirst for electricity. However, they may be waiting a while.
According to Oklo's website, the nuclear startup hopes to bring its first plant online before the end of the decade. Before that can happen, though, Oklo will need to obtain approval from the Nuclear Regulatory Commission -- something for which it says it's already submitted applications. In 2022, the watchdog rejected an Oklo plan to build a small atomic reactor in Idaho, citing "significant information gaps" on safety-related measures.
That said, Oklo has lately received support from US government agencies including the Department of Energy (DoE), which has awarded a site use permit, while Idaho National Laboratory -- home of EBR-II -- has provided fuel material to support the efforts. Speaking of fuel, Oklo's designs may not suffer from the challenges other SMR startups, like Terrapower, have encountered. Oklo's designs are intended to run on recycled nuclear waste products from traditional reactors. In fact, the startup is currently working with DoE national labs to develop new fuel recycling technologies. Oklo hopes to bring a commercial-scale recycling plan online by the early 2030s.
According to Oklo's website, the nuclear startup hopes to bring its first plant online before the end of the decade. Before that can happen, though, Oklo will need to obtain approval from the Nuclear Regulatory Commission -- something for which it says it's already submitted applications. In 2022, the watchdog rejected an Oklo plan to build a small atomic reactor in Idaho, citing "significant information gaps" on safety-related measures.
That said, Oklo has lately received support from US government agencies including the Department of Energy (DoE), which has awarded a site use permit, while Idaho National Laboratory -- home of EBR-II -- has provided fuel material to support the efforts. Speaking of fuel, Oklo's designs may not suffer from the challenges other SMR startups, like Terrapower, have encountered. Oklo's designs are intended to run on recycled nuclear waste products from traditional reactors. In fact, the startup is currently working with DoE national labs to develop new fuel recycling technologies. Oklo hopes to bring a commercial-scale recycling plan online by the early 2030s.
Waste transport (Score:3)
How are they getting the waste on-site for use as a fuel? It's been impossible to move waste to Yucca Mountain or elsewhere due to NIMBYs blocking transport efforts.
Re: (Score:3)
Re: (Score:2)
In this case, nuclear waste is the fuel for SMRs.
SMRs are a scam. (Score:1)
Nope, it is absolutely not.
1) Nuclear waste is the raw material for new fuel in this hypothetical scam scheme. It needs very extensive chemical processing to separate all kinds of isotopes.
2) This processing is extremely messy. It means passing extremely radioactive stuff through a chemical plant, which will be completely contaminated, and thus needs to be built to be serviced completely remotely.
3) Making fuel with new minerals is much less problematic, because it's 10 000x less radioactively dangerous. Yo
Re:SMRs are a scam. (Score:4, Insightful)
None of what you said actually refutes my statement.
Re: SMRs are a scam. (Score:3)
Waste reprocessing will be easier with advanced robotics. Once it gets automated, you'll be able to reprocess and use the same nuclear fuel for power over and over again for centuries. This is a hard engineering problem, like landing rocket boosters, but it will be worth it to solve.
Re: (Score:3)
Nuclear waste is radioactive glass. They just place in a silo and it can stay at the same location forever. If they need to move it, that move is safe and not a problem. Those silos that they use are designed to withstand almost everything. So moving it to a new type of reactor is not a problem. Half life of the isotopes is at most few years. The real dangerous stuff has half life of few hundred thousands of years. There are few things that half life of 700 million years*. Nuclear energy is really energy de
Re: (Score:2, Insightful)
You need to learn how to read.
Re: (Score:1)
Read about what?
You basically claim the "waste" is a Greenpeace made up problem.
Which it is not.
Re: (Score:1)
Nuclear waste is radioactive glass.
Ah, no? Not even remotely?
Re: (Score:2)
If you know what it is. Tell me. I've already told you that radio activate waste is melted into glass and other solid material and stored as such. I even have sources for my claims. Now its your turn. Show me your sources on the claim that it is not glass and solid material and stored as such in large silos.
Re: (Score:2)
I work for a company that did work on the Waste to Glass project. While you're correct that this is a thing, its purpose was to get rid of waste resulting from nuclear weapons production, and not civilian power reactors--in other words, vitrification is NOT the primary way of storing nuclear waste, much less waste from commercial power. The vast majority of waste from nuclear power sits in cooling pools until cool enough to be moved to dry casks, which are currently stored on site--the casks were supposed
Re: (Score:2)
A simple web-search and then some reading (warning: hard to do!) will tell you:
- https://en.wikipedia.org/wiki/... [wikipedia.org]
- https://en.wikipedia.org/wiki/... [wikipedia.org]
But let me make it even easier (can be done for free and without registration):
Question: "ChatGPT, is nuclear waste glass?"
Answer: "Nuclear waste itself is not inherently glass, but it is sometimes immobilized in glass to safely store and contain radioactive materials. This process is known as vitrification. [...]"
Apparently you have zero fact-checking skills.
Re: (Score:2)
I raise you properly research United States government article on nuclear waste.
https://www.nrc.gov/reading-rm... [nrc.gov]
https://www.gao.gov/nuclear-wa... [gao.gov]
The yellow barrel also do not contain any liquid. Like, at all. Here's a simulation of what's inside of those barrels. I was unable to find a proper source of this image.
https://www.reddit.com/r/nucle... [reddit.com]
Popular culture has lied to people on nuclear waste and how this process works. This has also made global warming much worse, since to replace nuclear power. A coa
Re: (Score:2)
Funny thing: Your first reference does not even contain the word "glass" and the second one says "contained _in_ glass" in the only place the word "glass" shows up. So, no, nuclear waste is not "just radioactive glass".
Seriously, you cannot fact-check for shit.
Re: (Score:2)
Checking your other comments here, shows that you are just a boring troll looking for a fight. Your failure is also that you trust A.I for correct answers, the problem is that A.I can't and won't give you correct answers. Even if they run it on nuclear energy.
I have given you sources, with actual data. You have arrived with nothing, therefore you leave with nothing but your same and continued ignorance. It is you that can't "cannot fact-check for shit", because, from the looks of it. You're just a trash hum
Re: (Score:2)
I looked at your "sources". They do not contain what you claim they contain. And no, I am no "troll", that is just you trying to hide your non-understanding. FYI, the first time I looked at vitrification of nuclear waste was about 40 years ago. I _know_ what it does. You do not.
I think you may just not be smart enough to understand the difference between "A is contained/encased in B" and "A is B". What a pathetic failure. But hey, you like nuclear energy, so that you are dumb and uninformed is a given. If y
Re: (Score:2)
If you're not troll. Why are you behaving like one. Now you're suddenly a nuclear expert. I've heard that one before. By a troll that was losing the argument.
Re: (Score:2)
The real dangerous stuff has half life of few hundred thousands of years.
You have that bit backwards.
The dangerous stuff has a very short half-life. It emits a lot of radiation as it decays, so it decays quickly... into successively less dangerous materials. The stuff with a super long half-life is no more dangerous than dirt. Maybe don't powder it and snort it, but outside the body it is nothing to worry about.
Ah, liquid metal cooled reactors... (Score:4, Funny)
Seriously... just about every attempt at building a liquid metal cooled reactor has ended in either abject failure or outright disaster, from the Santa Susana field lab meltdown through the Monju coolant leak. It's nature's way of saying, "this is a really bad idea."
Re: (Score:2)
Yep, True. Also Superphenix produced some electricity, but at a cost over 50x higher than regular nuclear, decommissioning included.
A measly 7700 GWh for 25 billion euros and counting (decommission is still costing EDF today.)
Re: (Score:2)
Last I checked Oklo's application for a combined licence was denied due to them not supplying enough information to the regulator. They seem to be using liquid sodium for the coolant, which catches fire on contact with air, as well as corroding everything. Their plan is to bury the reactor, which is probably a terrible idea when it comes to decommissioning.
Overall they don't seem to have any revolutionary ideas that are going to make it work this time, when it has failed catastrophically every time it has b
Re: (Score:2)
Those reactors use a special alloy that does not corode at all.
Re: (Score:2)
Ah! Magic metal.
Re: (Score:1)
Well,
You can sent me $100 and I google it for you.
Or you google it yourself.
Or stay stupid.
Up to you.
Pipe dream (Score:4, Informative)
Nobody has a proven SMR, let along an even working one. That means at the very least this is 20 years away. But since nuclear tech is a _lot_ harder and the hardware fails in surprising places, probably more like 30-50 years, if it materializes at all. No, military reactors do _not_ count as they have an entirely different safety and efficiency profile and are massively more expensive.
This is just the basically "criminal enterprise" nuclear industry trying to get the cash flowing their way again.
Re:Pipe dream (Score:5, Insightful)
actually there are lots... they operate on subs, ships, carriers, but since you said no military-
Ice breakers have been operating SMRs for decades pretty safely.
As for safety and efficiency... the safety record of the US Navy operating nuclear reactors is much better than any countries commercial nuclear program.
So maybe from a safety margin, one WOULD want to look at those designs.
Plus newer designs are rated for 25+yr operations without a need to refuel.
so, no... not a pipe dream... just that those existing advanced designs are classified, and private enterprises are trying to re-invent that wheel on a shoe string budget in comparison.
Re: (Score:2)
Nuclear ice-breakers use military reactors. There are exactly zero non-military design SMRs. Military designs can be a _lot_ more expensive in creation and operation than civilian ones and still make sense. Seriously. Civilian SMRs that are economically viable are a pure hallucination at this time.
Re: (Score:3)
Rolls-Royce SMR is slated to build a number of SMRs in the UK, and the EU.
The Russians currently running KLT-40S SMRs... an design borne from the ship based OK 150/900 models.. non military.
and China is operating a demo SMR... the HTR-PM.. since 2021
So outside of the dozens of designs waiting regulatory review and approval... there are a host of military ones... and 2 long running designs... and another one slated for multiple locations.
"20 years away" was the case 20 years ago...
They're more expensive than
Re: (Score:2)
https://www.power-eng.com/nucl... [power-eng.com]
Re: (Score:2)
Funny thing: No working instance of the reactors planned to be installed there exists.
Re: (Score:2)
Funny thing: No working instance of the reactors planned to be installed there exists.
Funny thing: that is how it is with everything new. Good thing that is irrelevant or we would all still be living in caves. History will be leaving you behind soon enough.
Re: (Score:2)
The KLT-40S is essentially a military design. If you refuse to understand that these are very different, you can of course hallucinate that SMRs are a solved problem.
Re: (Score:2)
None of those naval reactors are commercially viable though. They only exist because the nations that built them want to project power. Even the civilian ones are heavily subsidized and only really civilian for political reasons, i.e. not sailing military vessels into sensitive areas that the USSR and now Russia wants to stake a claim for.
There are good reasons why commercial cargo ships don't use nuclear power. It's a shame because it could remove a big source of emissions, but the reality is that it's jus
Re: (Score:2)
Indeed.
I don't think the problem is they're classified (Score:2)
As for the Ice Breakers, the risk there is pretty low since they're not near by a city. Also it helps that if they start to overheat they're sitting in freezing temperatures.
When we get right down to it what worries us NIMBYs about nuclear power is always corner cutting by CEOs.
Re: (Score:2)
actually there are lots... they operate on subs, ships, carriers, but since you said no military-
Errr no. SMRs being talked about here for power generation share virtually nothing in common with those on ships other than some atoms smash together.
As for safety and efficiency... the safety record of the US Navy operating nuclear reactors is much better than any countries commercial nuclear program.
The safety of me driving my car is better than that of the US Navy operating reactors, it's equally as relevant here to the discussion of commercial SMRs. The designs are different, operating parameters are different, risk is different, and risk acceptance is different.
Plus newer designs are rated for 25+yr operations without a need to refuel.
I'll wait for an SMR to actually function even as a pilot project before we'll settle on pro
Economic Suicide. (Score:2)
Producing electricity with a SMR is Economic Suicide.
It's sole purpose is just to get subventions from the Govt.
Re: (Score:2)
Yep, pretty much. The only exception is powering large ships and submarines in a military or very long-endurance context (ice-breaker). It is still more expensive than other forms of propulsion, but in these circumstances, long endurance can pay off in other ways.
Re: (Score:2)
> That means at the very least this is 20 years away.
First experimental nuclear reactor: 1942.
First nuclear reactor connected to a power grid: 1954.
That was only 12 years with '40s and '50s technology.
Where are you getting 20 years with 21st Century technology?
International waters (Score:3)
Idea: Build nuculear or other power generation system on a ship, anchor in international waters far far away from NIMBY regulatory reach, and then run a cable to power the mainland.
https://en.wikipedia.org/wiki/... [wikipedia.org]
The Australia–Asia Power Link (AAPowerLink) is a proposed electricity infrastructure project that is planned to include the world's largest solar plant, the world's largest battery, and the world's longest submarine power cable.
Initial plans forecast that a new solar farm in the Northern Territory of Australia would produce up to 20 gigawatts of electricity, most of which would be exported to Singapore, and at a later point Indonesia, by a 4,300 km (2,700 mi) 3 GW HVDC transmission line.
Re:International waters (Score:5, Insightful)
Hilarious.
1) Take the most expensive and dangerous power generation option.
2) Make it more expensive using a huge ship and an underwater cable.
3) Make it more dangerous with a danger of sinking and contaminating an entire ocean
4) Make it more expensive to decommission by putting it onto a ship that has to be disposed of
5) Make it more dangerous with the real possibility of leaks of molten metal into water or water into molten metal (Hint : Big BadaBoom)
Nope.
Re: (Score:2)
> Take the most expensive and dangerous power generation option.
This is nonsense.
The nuclear power industry has been operating in the United States for going on 70 years without a single fatality to a member of the general public. Not one.
That is a better safety record than ANY other power source, including solar and wind.
Re: (Score:2)
There's a reason nuclear submarines are extremely finicky machines (the form of SMR that exist today)
Water and nuclear fuel on a very enclosed space are inerently dangerous.
Only an extreme amount of care prevents them from exploding.
Sometimes they fail anyway.
https://www.theguardian.com/wo... [theguardian.com]
Re: (Score:2)
> Build nuclear or other power generation system on a ship
The point is it has been common practice for 70 years, shifting production (steel, clothing, chemicals, etc.) from highly regulated economies to much lower regulated economies is an option for power generation.
There probably is a tipping point of 'soft-nationalization' via regulations where a government increasing regulatory burden on an industry forces the industry out of business; yet for critical industries - electricity generation, going out
Re: (Score:3)
So to be clear your idea is to take the most expensive form of power, scale it down to make it more expensive, put it on a floating ship to make it more expensive, and then run an expensive submarine cable to it, all the while being ignorant of the fact that international waters aren't the lawless lands you think they are to say nothing of the fact you still need to get your uranium from somewhere and transmit your power to somewhere (presumably a place with more laws and regulations than the place you thin
Re: (Score:2)
International waters aren't some magical place where laws do not apply. The ship would need to be registered to a country, which would be responsible for any nuclear accidents it experienced.
Besides, putting it on a ship won't make it any more affordable. It will make it less affordable, because now you need a large ship and crew of experienced sailors, as well as the staff to run the reactors. Maintenance costs are multiplied by the need to move all parts, equipment, and personnel from land to the ship, an
The interesting thing... (Score:5, Insightful)
The interesting thing about this story is that people whose business is based on totally reliable low cost power are not buying wind, solar or some combination plus batteries. Instead they are going for relatively untested mini nukes.
These are not stupid people. If untested mini nukes, with all their risks, seem to them like a better alternative than renewables, that is telling us something important: its telling anyone who wants to listen that when you are betting your company, wind+solar+storage is not operationally viable.
Something the UK is going to find out, with regard to national power supplies, in the next few years.
Re:The interesting thing... (Score:4, Insightful)
The interesting thing about this story is that people whose business is based on totally reliable low cost power are not buying wind, solar or some combination plus batteries. Instead they are going for relatively untested mini nukes.
These are not stupid people. If untested mini nukes, with all their risks, seem to them like a better alternative than renewables, that is telling us something important: its telling anyone who wants to listen that when you are betting your company, wind+solar+storage is not operationally viable.
Something the UK is going to find out, with regard to national power supplies, in the next few years.
Alternative theory: Datacenter management is currently obsessed with pipe dream bullshit like current gen AI, and thinking they need the biggest, bestest, grandest, largest, most power-sucking datacenter possible, so it's entirely possible they love the idea of this pipe dream as well, and buy fully into the hype machine surrounding it rather than looking at the history, or the obvious "will it actually happen" question that should be addressed before tossing money at possible grifters selling a non-viable tech. Hmm. Grifters buying from grifters. Maybe I've been wrong in thinking we're developing an advertising based economy. Maybe we're really shifting to a grifter based economy. Why not? I mean, we're not at all interested in developing or dealing with reality. Maybe daydream bullshit is the way forward?
Re:The interesting thing... (Score:5, Insightful)
Re: (Score:2)
If you have a constant load 24/7 that can't be time-shifted, renewables are a non-starter.
The entire Cooper Basin disagrees with you where major operators including Santos has moved their constant load oil and gas processing operations to renewables + storage, incidentally this is something that they said improved the reliability over traditional oil and gas power plants they ran locally.
Re: (Score:2)
Instead they are going for relatively untested mini nukes.
That is actually a "completely untested". If they did their research, they know they will not get these anytime soon, so this is essentially just political bullshit. On the other hand, it may just be abject management stupidity.
Plutonium Reactors (Score:2)
If untested mini nukes, with all their risks, seem to them like a better alternative than renewables, that is telling us something important:
Yes, it is telling us that batteries are still really expensive and that nuclear power offers a possible cheaper alternative so these companies are willing to take a risk that it will work out. I'm not sure that it is a good risk when dealing with liquid metal cooled fast breeder reactors since there has been a _very_ long history of these not working and, even when they do, nuclear proliferation is usually an issue since they use the neutrons from plutonium fission to breed more plutonium using depleted u
Re: (Score:2)
These are not stupid people.
Sorry but there's no evidence of that. The people who manage datacentres are no different than the management of any other company. They are driven by buzzwords, hype and promises. What we're getting here is a bunch of promises with a huge amount of unknown variables. When these variables get analysed the boom bust cycle finishes and management moves on to the next big thing the industry is promising to help them run their datacentres better.
that is telling us something important: its telling anyone who wants to listen that when you are betting your company, wind+solar+storage is not operationally viable.
Just so we're clear you job is about to be taken by AI? AI is goin
Re: (Score:3)
If you are a company in AI, totally dependent on reliable consistent power, what you can see coming down the road at you is intermittent and unreliable. Blackouts and load shedding, in short. This will result from the national attempt to get to net zero (in generation first) by moving to wind and solar.
So what you would really like to do is build a gas generation plant. But you know this is politically impossible, so you look for another solution, and there is only one readily available, and that is the
Outatime (Score:2)
How small are these reactors? Can one generating 1.21 jiggawatts be installed in a DeLorean?