UK moves a step closer to nuclear waste solution

Something potentially rather interesting is happening deep within the Nuclear Decommissioning Authority (NDA), the branch of the UK government responsible for dealing with nuclear waste overall and the country’s 100-tonne plutonium stockpile in particular.

For several years now the wheels of policy been grinding towards a ‘solution’ to the plutonium issue apparently designed to please Areva, the French state-owned nuclear conglomerate which is responsible for the EPR reactor design and is also heavily invested in MOX – ‘mixed oxide fuel’, where plutonium and uranium are blended together into fuel rods for re-use in civil nuclear plants.

Areva has put tens of billions into MOX capability. The problem is, however, as the NDA is clearly now beginning to realise, that MOX is the worst of all possible options for plutonium management. Why? Because it is diabolically expensive, creating a difficult-to-handle fuel that you have to pay utilities to burn. And perhaps worse, it doesn’t even get rid of plutonium – after a full MOX cycle, you in fact end up with more plutonium than you started with because new plutonium isotopes are created from irradiated uranium-238.

So why should the UK government fork out countless tens of billions for this white elephant? It is instructive to note that Areva’s MOX plant under construction in the US at Savannah River, South Caroline is already nine times over budget whilst still being little more than a giant hole in the ground (the cost has gone from $0.5 billion to $5 billion and counting). No-one wants MOX in their reactors, because it is far more expensive than virgin uranium fuel and has to be handled with massively increased security due to the presence of plutonium.

The fact that there are no customers for MOX is unintentionally highlighted in the UK’s latest missive from the NDA which suggests that Japan might be both a customer for MOX fuel, and a co-investor in any fabrication plant. At the moment, it looks like a better business bet would be to try to sell MOX to the Martians – Japan is in the process of (ill-advisedly, in my opinion) shutting down its entire nuclear programme due to the political backlash from the Fukushima accident, and is hardly likely to cough up for a huge offshore investment like this.

The idea of renewed Japanese largesse is especially ludicrous given that the existing Sellafield MOX Plant was closed last year largely because Japan cancelled all its orders following the tsunami disaster. This recent history also begs the rather obvious question: why, having just closed a billion-pound, loss-making MOX plant – following the exit of its one remaining customer – would you want to go and straightaway build a new one?

Luckily the civil servants at the NDA, and their cousins over at DECC, the Department for Energy and Climate Change, are not stupid. They know a MOX plant will never actually be built, and that Areva has wasted billions already on what is essentially obsolete technology. That’s bad for Areva, and therefore the French taxpayer, but isn’t our problem. Hence the very intriguing spin put on the NDA statement, which is that the authority is “seeking proposals on potential alternative approaches for managing the UK’s plutonium stocks” even as it simultaneously “progresses its preferred policy of converting the material into mixed oxide fuel”.

In other words, the NDA is looking for a way out of the MOX dead-end before the policy process grinds so far forward that they find themselves too politically committed. No doubt Areva is knocking loudly on the NDA’s door and vociferously lobbying anyone who will listen, but the fact this they have backed the wrong horse and will lose big-time – because there is a far better option already on the table. That – as this World Nuclear News piece suggests – is GE’s PRISM fast reactor concept.

Last month I had a meeting with some of GE’s top people, including the chief engineer for the PRISM – and I have to say I found their strategy highly compelling. The PRISM is a metal-fuelled fast reactor cooled by molten sodium. It operates at atmospheric pressure – meaning no expensive pressure vessel – and the characteristics of the fuel and the molten-sodium coolant (which conducts heat away from the core 90 times more effectively than water) make it fully passively-safe. (An early version, called the Experimental Breeder Reactor, was subjected to a loss-of-coolant flow experiment in 1986 – it duly shut itself down with no outside intervention.)

Perhaps more importantly, fast reactors can burn up all the energy in the uranium and plutonium fuel, whilst utilising MOX only increases the energy use from 0.6% to 0.8%. Because of this, as the Guardian recently reported, if all the UK’s spent fuel, depleted uranium and plutonium stockpiles are combined, they include enough energy to run the country for 500 years at current electricity use rates – without the need to mine another scrap of uranium, and without the emission of a single tonne of CO2. (Greenpeace is vociferously opposed, despite its supposed great concern for global warming, no doubt because a solution to nuclear waste leaves it high and dry after 35 years of misguided anti-nuclear activism.)

Areva’s submission to the NDA’s original plutonium consultation asserted that fast reactors are not an option until 2040 or later. However GE’s executives told me that they could get one up and running in 5 years – the PRISM is fully proven in engineering terms and basically ready to go. Nor will the UK taxpayer be asked to fund upfront capital costs – instead the proposal is for the NDA to pay a fixed price per kg of plutonium dealt with, whilst the PRISM plant will also generate a return by selling commercial electricity. (1 PRISM block has a rated capacity of 600 megawatts.) The risk of going over-budget – always a concern for first-of-a-kind projects – remains with GE.

Making plutonium unusable for bombs is only part of the deal, and not even the best part. (It is actually already unusable for bomb, because of different plutonium isotopes like pu-240, which make it non-weapons grade, so this entire thing is a political rather than a technical concern anyway.) For me, the most compelling reason to look seriously at the PRISM is that it can burn all the long-lived actinides in spent nuclear fuel, leaving only fission products with a roughly 300-year radioactive lifetime. This puts a very different spin on the eventual need for a geological repository – instead of something that will be designed to safeguard radioactive material for a million years (technically a very improbable idea), safeguarding waste for 300 years is a very different, and much less challenging, proposition.

The PRISM is not going to be a shoo-in, however. Politically the term ‘fast reactor’ reminds British people – and the government – of Dounreay, a fast reactor built in Scotland that operated as a prototype for many years but has had problems with waste management and decommissioning. Dounreay, however, whilst it was sodium-cooled, did not share many of the PRISM’s key engineering advances – in particular the pool versus loop coolant design, and metal versus oxide fuel. (Metal fuel expands if it overheats, shutting off the fission reaction and making a meltdown physically implausible.)

So don’t expect the PRISM to be selected as a preferred technology just yet. GE has a lot of work to do to convince the NDA and other stakeholders that the metal-fuelled fast reactor is a serious, and commercially-viable proposition. But I think the NDA’s latest statement is another nail in the coffin of MOX – and that has to be a good thing.

© Mark Lynas
« · »