Debating the DOE Report on Pathways to New Advanced Nuclear - Episode 101

The DOE recently released a report outlining a pathway for the US to meet recent pledges of tripling of domestic nuclear capacity by 2050. Nuclear needs to play a role in a clean energy economy, but the industry may be fooling itself in thinking that learning curve cost decreases will help to make these advanced nuclear reactors more economically viable.

In this episode, Paul discusses the challenges facing the nuclear industry in reducing costs, including the need to customize between facilities, the challenge in translating workforce experience across construction firms, and the ongoing costs associated with the high regulatory hurdles associated with this technology.

How can the US support new nuclear, but in a way that ensures affordability for the energy transition?

For more research:

"Pathways to Commercial Liftoff: Advanced Nuclear" - Department of Energy

"To meet US nuclear goals, big reactors need to get built today, DOE says" - Canary Media

"Historical construction costs of global nuclear power reactors" - Jessica Lovering, et al. (2016)

"The economics of nuclear power: Further evidence on learning, economies of scale, and regulatory effects" - Cantor, et al. (1988)

"EDF’s UK Hinkley Nuclear Costs Balloon as Plant Delayed Anew" - Bloomberg

Follow Paul on LinkedIn.

Transcript

This is Eight Minute Climate Fix – a podcast helping you understand the energy and climate challenge in just a few minutes – I’m your host, Paul Schuster.

Welcome to our third season! 

Before I launch into this week’s topic – a couple of quick show notes to share with everyone as we start off our NEXT 100 episodes. First – yep, we’ve tweaked the title of our podcast a bit, going from just “Eight Minutes” to … now … “Eight Minute Climate Fix”. Figured it made more sense as to the intent of the show.

Let us know how you like the new title – or, for that matter, what you’re enjoying about the show – what topics you’d like to see in future episodes – there’s now a link at the top of the episode notes that directly texts us with your thoughts. We’d love to hear from you.

Okay – housekeeping done. Let’s shift to today’s topic which is all about a recent DOE report outlining the pathways toward deploying a massive increase in nuclear generating stations by 2050. Nuclear – is having a moment, right now. Between the bipartisan ADVANCE act that sailed through congress a few months ago, to the recent announcements by some tech giants on relicensing and reactivating reactors at Palisades and Three Mile Island to support data center growth. 

But is the industry, and the DOE, being a bit too bullish on this technology? Advanced reactors are still new technologies and there are a LOT of assumptions that need to go right for these projects to make sense. Let’s unpack the details and figure it out together. 

Eight Minutes – it’s how long it takes the sun’s rays to hit earth – or about how long Aaron Rodgers has left in his career as an NFL QB. Guy’s looking old after my Patriots knocked off the Jets this weekend.

Let’s get it on.

Nuclear needs to play a big role in the future clean energy system. It doesn’t emit emissions, it’s stable, baseloaded, reliable power. Lots of benefits of having nuclear as part of a portfolio of generation.

But – it’s expensive. Like, REALLY, expensive. The new Vogtle Units 3 and 4 that were just completed and brought online in Georgia cost over $36 billion dollars and took over 15 years to complete.

Despite those huge costs, nuclear remains a really hot topic. Just a couple of months ago a rare, bipartisan bill reducing regulatory issues and supporting the growth of advanced nuclear sailed through Congress. The ADVANCE act showed that, even in our current, hyperpartisan environment, nuclear is one of those few issues that both sides of the aisle can agree upon. 

And, then, recent announcements from Microsoft and Amazon show just how important nuclear could be to meet growing datacenter demand needs. Nuclear provides clean, around the clock power that helps those tech giants meet their energy needs while also holding to their climate commitments.

Microsoft announced plans to reopen a reactor at Three Mile Island – while Amazon announced intentions to work with Dominion Energy in Virgina in exploring small modular reactors.

And then the DOE released a report earlier this month on the commercial pathways for liftoff of new, advanced nuclear in the US. At COP28, in Dubai, the US announced an ambition to triple nuclear capacity by 2050. That’s – an EXTRAORDINARY target – as it would require somewhere on the order of 13 new gigawatts of capacity every single year from 2030 on. As Canary Media points out – even in our nuclear construction heyday of the 1960s and 70s – we were only pumping out 6 GWs per year. This would more than double it.

But – let’s face it, none of that is going to happen if costs remain as high as they were for the Vogtle units. And it’s not just Georgia that saw those types of runaway costs. Flamanville in France cost north of $14 billion dollars for its reactor and the Okiluoto facility in Finland cost about $12 billion. And then there’s the Hinkley generating station in the UK which cost around $33 billion dollars to construct.

The industry recognizes that these costs are unsustainable – but point out that learning efficiencies from these first-of-a-kind units will continue to improve. The DOE points out that the costs for Vogtle Unit 4 dropped by 30% from that of Vogtle Unit 3.

The DOE report suggests that the current costs of around $11,000 per kilowatt COULD drop down to about $4700 per kilowatt once future projects are built. Now that these initial units have built a domestic supply chain and have trained a new generation of nuclear workforce on this - *maybe* that makes sense?

A lot is riding on that projected experience curve dropping that far, very, very quickly. But there’s a LOT of research suggesting that the industry and policy makers are looking at this far too optimistically. 

For one thing, the improvements we see along a typical learning curve come from experience, process improvements, scaling and standardization opportunities that drive down costs over time. But as Lovering, et al point out in a 2016 paper on nuclear construction – there’s not a TON that replicates between one nuclear project and another. They’re EACH bespoke, customized projects that may use similar reactor technology, but the balance of plant and surrounding construction is all very individualized to the project. PLUS, while they ARE improvements that can be seen WITHIN a construction firm’s workforce as they gain experience with this type of construction – the experience ACROSS an industry, or even a consortium such as what the DOE suggests as a potential solution for this problem – is minimal.

That means that the workforce to build 13 GWs a year of new nuclear - is still decades away from being strong enough to do so.

But building nuclear generation is also very different than your normal construction project, because it’s SO heavily regulated. Those regulations, safety requirements, permitting, approvals – those don’t go away from one project to another … and they add TONS of additional costs to a project. For GOOD reason – but thinking that regulatory costs somehow follow the same learning curve as other construction costs is a bit silly.

These new construction costs are the primary reason as to why existing power plants are getting their operating lives extended – and why Microsoft and others are so interested in repowering existing units instead of building new ones. 

And then there’s the small modular reactor contingent. These SMRs are supposed to be more modular, more scalable – more standardized in order to absolutely take advantage of experience curves to drive down costs. The DOE’s report – kind of dismisses this type of technology, calling it super expensive and, instead, advocating for cost reductions on the economies of scale of massive advanced nuclear facilities.

So – who’s right? New nuclear is super expensive, right now – let’s agree on that. But the DOE’s stance that size and scale somehow will drive down costs contradicts the basic premise of the SMR industry that standardization and modularization is what is needed.

I’ll provide my two cents into the debate. I think the DOE missed the mark here. In an effort to outline A pathway to meet those ambitious nuclear goals as laid out at COP28, the agency HAS to make the assumptions that big nuclear is going to drop in cost and drop quickly. That’s the only way that goal could possibly be met. But the cold reality is that that’s not going to happen.

But that’s not to say that SMRs have this figured out, either. They face the same regulatory hurdles as larger plants – but spread those costs over far FEWER kilowatts of capacity. Not EVERYTHING is subject to an experience curve.

Instead, we should be focused on regulatory reform, on investing in repowering existing facilities, on safely extending the operating life of our current fleet of reactors. And, on a macro scale, we should be investing in new build of other forms of clean energy such as offshore wind, long duration storage, solar and others that can generate the same benefits of nuclear – without the cost, risk and uncertainty that this technology poses.

I’m Paul Schuster – and this has been your eight minutes.