Can Bill Gate’s Small Nukes Be Competitive?

November 13, 2022

Slightly weird video trying a little too hard to be cool, while it seems whoever wrote the script had no idea of how the science they were describing actually works.

It’s not that I’m “against” nuclear power, but we have a generation of climate activists who seem to think all you have to do to get a nuclear plant is go pick one up at the nuclear plant store. As Arjun Makhijani shows, below, it’s a little more complicated than that.
And I’m reposting Mark Jacobson’s recent assessment of where the nuclear industry is to further underline the high hurdles nuclear has to get over to be viable.

Below, GE executive hopes that small modular reactors can, by 2050, deliver electricity at about double the price of where solar and wind are today.

Utility Dive:

  • Small modular nuclear reactors can be developed with a levelized cost of electricity of about $60/MWh, Jon Ball, executive vice president of market development at GE Hitachi Nuclear Energy, said Friday on a panel hosted by the U.S. Energy Association. The company is developing the 300-MW BWRX-300 small modular reactor.
  • A recent poll by the Nuclear Energy Institute predicted that at $60/MWh, there could be about 90 GW of SMR capacity on the U.S. grid by 2050.
  • NuScale has developed the first SMR design to receive approval from the Nuclear Regulatory Commission and has projected a $58/MWh LCOE for a project with Utah Associated Municipal Power Systems, though some critics doubt the company can achieve that.

Cost is a key component in the development of new nuclear resources, and critics points to recent overruns at traditional reactor projects in South Carolina and Georgia as cautionary tales amid falling renewables prices.

Santee Cooper and SCANA Corp. eventually abandoned their South Carolina project when cost estimates topped $25 billion. In Georgia, with Southern Co. subsidiary Georgia Power as the primary owner, Vogtle units 3 and 4 are years behind schedule and billions of dollars over their initial budget.

Meanwhile, the costs of renewables and storage are falling. Utility-scale solar-plus-storage costs are about $45/MWh; wind power costs are $30/MWh; and stand-alone utility-scale solar costs are at $32/MWh, according to the Institute for Energy Economics and Financial Analysis. And that group has doubts that SMRs can be developed as cheaply as their backers claim.

SMRs are a new technology, and “if you look at the cost of the first one that rolls off, it’s going to be fairly high,” U.S. Nuclear Industry Council President and CEO Bud Albright said at the USEA event Friday. But, he added, “we’re counting on volume, and volume is predicted. And so that should bring costs down.”

GE Hitachi’s Ball called the $60/MWh cost “achievable.”

“We believe that the future of nuclear is in small modular reactors,” he said. “The industry’s focus for decades has been on large reactors, but you’ve seen what has occurred in terms of these projects going over budget, way past schedule. … The reason we pivoted was to create a design that could be cost-competitive with all forms of generation.”

SMRs will have a longevity advantage over renewables, Albright added. “A small modular reactor should last a minimum of 60 years. Probably more, up to 100, frankly, if maintained properly. Wind and solar, after about 20 years you have to replace everything.”

Even at a higher price, there is a market for SMRs, said NEI Vice President and Chief Nuclear Officer Doug True.

The group surveyed members about a $60/MWh resource and found that members would install 90 GW by 2050. 

“When we raised our price to $90 per megawatt hour, the number drops a little bit but is still interesting because of the value nuclear brings to the overall system in terms of reliability and versatility,” True said.

Stanford University:

Nuclear reactors generate reliable supplies of electricity with limited greenhouse gas emissions. But a nuclear power plant that generates 1,000 megawatts of electric power also produces radioactive waste that must be isolated from the environment for hundreds of thousands of years. Furthermore, the cost of building a large nuclear power plant can be tens of billions of dollars.

o address these challenges, the nuclear industry is developing small modular reactors that generate less than 300 megawatts of electric power and can be assembled in factories. Industry analysts say these advanced modular designs will be cheaper and produce fewer radioactive byproducts than conventional large-scale reactors.

But a study published May 31 in Proceedings of the National Academy of Sciences has reached the opposite conclusion.

“Our results show that most small modular reactor designs will actually increase the volume of nuclear waste in need of management and disposal, by factors of 2 to 30 for the reactors in our case study,” said study lead author Lindsay Krall, a former MacArthur Postdoctoral Fellow at Stanford University’s Center for International Security and Cooperation (CISAC). “These findings stand in sharp contrast to the cost and waste reduction benefits that advocates have claimed for advanced nuclear technologies.”


9 Responses to “Can Bill Gate’s Small Nukes Be Competitive?”

  1. John Oneill Says:

    Couple of points from Mark Jacobson’s interview –
    ‘.. another reactor shut down in Belgium’ – only because the Green Party there made it a condition of support for the government, and arranged for state-subsidised gas turbines to replace it, at a time when European gas prices are through the roof. Belgian emissions, lower than any of their neighbours apart from France, will undoubtedly rise markedly if and when they close the rest of the nuclear plants that have historically provided more than half their power.
    ‘..harvest plutonium, which is even easier to produce weapons from..’ Nobody has ever made a weapon using spent fuel from a light water reactor. The US tried a test explosion using fuel from a British Magnox gas cooled reactor, but only because the fuel had only been irradiated for a short period. More than a month or so, and too much Pu239 becomes Pu240, which is too unstable to make a practical weapon out of. The same would be true for a fast reactor.
    ”..radon risk from underground uranium mines..’ More than half the uranium mined these days is with in situ leach mining, where two rows of holes are drilled, and an acidic or basic solution pumped between to leach out the uranium, while leaving the bedrock in place. No dust, no crushing needed, hardly any mine waste, and no radon. Modern underground mines have much better ventilation than back in the fifties, anyway, so the life expectancy of a uranium miner should be rather better than his coal-mining counterpart, who had to deal with black lung and methane explosions. They’d only have to mine just over half a tonne of uranium to match the energy output output of 100,000 tonnes of coal – the amount needed each day by a one gigawatt power plant.
    ‘So we’re talking about fifteen to seventeen years longer, and seven to eight times the unit cost to get the same power..’ Power which can be relied on is not the same as power which can disappear for a day, or a week. The Chinese manage to build reactors in four to six years – including in Pakistan, and including American-designed reactors. Nor do they saddle them with 9% finance rates, enough to triple the cost of a slow build. The Europeans, Americans, Canadians, Japanese, and Koreans used to be able to build reactors expediently, and will do so again.
    The technologies Jacobson calls proven have not reduced power emissions anywhere to the consistently low levels of a nuclear grid, except where there’s a preponderance of hydro to smooth the bumps. Nor have they even made a start on industrial heat demand, a much larger part of the word’s emissions.

    • On the subject of industrial heat, B.F. Randall is a rather new energy commenter who has a lot to say about it in this video:

      he’s been attracting a lot of attention with long, well done, detailed Twitter threads and now a free Substack:

    • rhymeswithgoalie Says:

      “Couple of points from Mark Jacobson’s interview –
      ‘.. another reactor shut down in Belgium’ – only because the Green Party there made it a condition of support for the government, and arranged for state-subsidised gas turbines to replace it, at a time when European gas prices are through the roof.”

      Yes, that sucks. While it may have been an extemporaneous error, someone in Jacobson’s position should know not to conflate losing nuclear to political causes vs. losing nuclear to age (retirement), loss of cooling, operating cost, or some form of expensive disability. He should have stayed with Hinkley and Flamanville as examples of long project times that showed how new nuclear power plants won’t be part of the GHG Emergency Stop we need now even if they are found to be cost-effective in the long term.

      Per Wikipedia (citation is in Dutch or German):
      “[The Belgian phaseout law (2003)] stipulates that no new commercial reactors are to be built and that Belgium’s seven reactors would be shut down when they reach an operational lifetime of 40 years. All reactors reach this age in the period 2015–2025.”

      Doel 4 and Tihange 3 had their closure date extended to 2035, ten years past what the 2003 legislature had scheduled.

      • John Oneill Says:

        The Belgian nuclear phase-out was pushed through by the Greens, who were afterwards voted out of parliament for ten years. The other parties effectively ignored it, but were too fractious to rescind it. So when the Greens finally got back into parliament, the legislation was already in place. Belgium is split four ways, between left and right, French and Dutch speakers, so forming a coalition can take years. A party whose only real raison d’etre is anti-nukery used that impasse to wreck a well-functioning, low emission power source – and the country’s largest, by far.

  2. John Oneill Says:

    As regards demand for cooling water, the Natrium should provide steam at 500-550 C, which would allow significantly more thermally efficient operation than a light water reactor, which is restricted to 320 C. That means less cooling is needed for the same power output. If water restrictions are likely, a hyperboloid cooling tower would also greatly reduce demand – the Kemmerer plant will use one. In the last few years, a new technique of zapping the air rising through the cooling tower with charged ions has been developed. This attracts water droplets to a sloping wire grid at the top of the tower, so that much of the cooling water can be recycled at minimal energy cost.
    Arjun Makhijani’s criticism that reactors not running all the time have more difficulty paying off their capital costs is also less true for the Natrium, which is specifically designed to run at 100% 24/7, while dumping unused heat into a molten salt storage system, for use when demand climbs again. This would suit the changes of grid demand quite well, since diurnal variation is usually only on the order of about 2x between minimum and maximum – probably less if car charging becomes common. It’s true that it’s not enough to remedy the vagaries of renewables, where wind can drop for a week or two, and solar for the whole winter. That’s more a fault of wind and solar than of the Natrium, though – only cheap gas can match that degree of randomness.

    • rhymeswithgoalie Says:

      The Natrium model is, in effect, a hybrid of nuclear power and storage, though it seems molten salt storage is much more demanding tech than on-slab batteries and even pumped hydro. The key value might be in the duration of energy storage.

      We’ll see where they are in 2028.

      • John Oneill Says:

        If you need a heat engine to make electricity anyway, it makes sense to let the plant load-follow by storing heat upstream of the generator, in a tub of salt, rather than downstream, in a battery. The battery is much more complicated, and only lasts about ten years. (The Musk mega-battery at Mills Landing, currently largest in the world, caught fire long before that, but survived.) Most existing pumped storage was built to complement always-on nukes, putting excess from night production into the evening demand peak, and giving faster response.

        • rhymeswithgoalie Says:

          The battery is much more complicated, and only lasts about ten years.

          Engineering-wise, scale batteries for the grid should be considered simple arrays of a plug-in commodity, the ultimate in mass scalability

          The Australian battery was the first major grid implementation of old-school Li+ batteries (and I whined at the time about that tech being wasted on slabs). Five years is a long time for battery tech considering the intense competition among labs and companies, and the expanding base of expertise, and the relatively short turn-around for prototyping (compared to other industrial projects).

          Even now we already have LFP batteries, which are less energy dense, but that’s a perfectly fine trade-off for something sitting on a slab. They’re safer, use less problematic materials, and last at least four times longer.

          We should also consider the eager investment money and the newly-minted hoards of nerds working on the battery chemistry coming down the pike. As with the tech boom of the nineties, I’d expect most battery storage technologies won’t pan out (or be competitive), but the rest will start to compete or fill in specific niches (e.g., the military pays big money for custom kit).

          Let’s compare tech accomplishments again in 2028, shall we?

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: