GE-Hitachi’s proposed PRISM fourth-generation fast reactor moved a step closer to reality today with the conclusion of an independent feasibility study showing that the design is credible for licensing and could provide a cost-effective way of quickly dealing with the UK’s 100-tonne plutonium stockpile. In a phone interview for this blog (published in full below), GE’s chief engineer Eric Loewen told me that the reactor could be up and running in as little as five years after the conclusion of the licensing process – as long as the Nuclear Decommissioning Authority abandons its current support for the much more expensive and complicated process proposed by the French company Areva for a MOX (mixed-oxide)-type management of plutonium.
Update: Duncan Clark over at the Guardian has written this news piece, similarly titled ‘Nuclear waste-burning reactor moves a step closer to reality’
According to Loewen it would then take only another five years after startup to ‘disposition’ the UK’s whole plutonium stockpile, and in the process the PRISM reactor would potentially deliver an ongoing 600MW of clean energy to the UK electricity grid. If government policy eventually changes to encourage reprocessing, the reactor could later move on to consuming spent nuclear fuel and depleted uranium, of which the UK has tens of thousands of tonnes in surface storage awaiting final disposal. A recent article in the Guardian revealed that these stockpiles combined already contain enough energy to run the entire country at current electricity consumption for five centuries – without the need to emit any CO2 or mine any new uranium.
However, the current proposal for PRISM is much more narrowly-focused. The UK government’s Nuclear Decommissioning Authority is running a consultation on how best to ‘disposition’ its plutonium liability – a process for which GE-Hitachi was a late entrant, hence the ongoing preference for MOX. (MOX involves mixing small amounts of plutonium into conventional uranium oxide fuel and burning it in existing light-water reactors such as Sizewell B.) The idea is to ‘spike’ the plutonium with fission products which would make it highly radioactive and therefore eliminate it as a potential proliferation risk. Current government policy is for the used fuel then to eventually be placed in a deep repository – even though the PRISM technology would allow all the plutonium and uranium stockpiles to be burned, so no burnable fuel ever has to go into geological disposal (indeed the very idea of million-year ‘geological disposal’ begins to seem outmoded when the radioactive lifetime of the waste is a few hundred years at most).
GE-Hitachi also provided further clarification during my interview on the financial model they are proposing to the NDA. With potential financing from the US Export-Import Bank, the UK taxpayer would not shoulder the risk of building the plant or even operating it, I am told. Instead, the NDA would pay the operator of the PRISM reactor per tonne of plutonium dealt with – a so-called ‘pay-for-performance’ model. This insulates the taxpayer from the risk that a plant might go over budget, as has been the case at Areva’s EPR reactors being built in France and Finland, and its MOX plant under construction at Savannah River in the US. (The latter was proposed at $550 million, whereas costs are reported to now be close to $5 billion – nearly ten times over budget, and the plant is still a long way from completion.)
Today’s announcement essentially means that the NDA is now under pressure in future to consider PRISM at least on an equal footing with MOX, and to abandon its long-standing policy bias in favour of Areva’s proposals. The PRISM already has prominent supporters both inside and out of government, and with 1000 pages of documentation now being delivered to the NDA’s doorstep no-one can claim that GE-Hitachi has not come up with a serious offer. Within as little as two months a further announcement should be forthcoming from the NDA about this – all eyes will be on them to ensure that all potential technologies are allowed to compete on a level playing field.
Below follows an extended interview with Eric Loewen, Chief Consulting Engineer Advanced Plants, GE Nuclear Energy (with occasional inputs where stated from David Powell, head of European new nuclear sales for GE-Hitachi). I have only done minor editing for clarity.
What’s new about the announcement today and how have things advanced with PRISM over the last year or so?
The significance of today is that we completed our feasibility study with the UK government’s Nuclear Decommissioning Authority. For us this is a significant milestone that we have completed this body of work to provide the answers to many questions that the NDA had, and at this point the NDA is going to decide in the next couple of months whether this technology is credible, and if it is deemed credible by the NDA then we will enter the next process, which is the justification process.
Right, so what have the milestones been over the last year or so since this process began?
Well the first milestone happened about a year ago where we formally put in our response to the public consultation on the management of plutonium. Then we did some outreach with different stakeholders during the summer of last year, and into the fall the NDA at the end of their public consultation agreed that MOX was their preferred option but they are interested in other ones that provide value to the UK taxpayer. From then we entered into a contract with the NDA to do a feasibility study, which concluded on Friday with our deliverable of about 1,000 pages of documentation to the many questions that they had.
Is this actually physical documentation… did you have to deliver boxes, or is it done electronically these days?
Both. So it was done in boxes and also we provided electronic versions.
Can you just explain to the uninitiated what is new about the PRISM technology and what its benefits are from your perspective?
The PRISM uses different fuels for its energy to make electricity, and its efficiency in operations is significantly better than a water-cooled reactor. For example, a water-cooled reactor can use about 1% of the available energy in the uranium that comes out of the ground, whereas this technology can use close to 99% of the available energy from fission.
So what that means in this case for the UK plutonium disposition, this technology allows a very simple process to fabricate the fuel that goes into the reactor, where we take that very abrasive plutonium oxide and put that into a metal form, put that into an alloy that then goes into the reactor, and we can extract as much energy as the policy would dictate, and then the reactor used fuel is planned currently to go into the UK’s geological disposal.
OK. So just talk us round the issues here with regard to the UK’s plutonium management. The other option on the table which the NDA still prefers, the so-called MOX, mixed oxide fuel, which is mixing plutonium with uranium – what are the issues here?
Well it’s not fair for me to talk about other technologies because I’m not familiar with MOX, but I can tell you the advantages of what we have. So the advantages of this technology is that we would take the plutonium oxide that is currently stored by the UK government. As soon as we would receive one of those cans we would mix that with the same amount of uranium oxide, so we would have a 50:50 mixture, and then through chemistry that has been proven since the 1800s we do what is called electro-reduction, where the oxide form goes away with the release of oxygen, and you’re left with a reduced metal – in this case it would be 50% plutonium and 50% uranium. That then is remelted, and we get it to the alloy composition that works best with the reactor that’s been proven at Experimental Breeder Reactor no.2, and at the Fast Flux Test Facility in the United States, and that’s by adding some different constituents such as zirconium.
Then that is easily fabricated into fuel and then put into the reactor. So it’s fundamentally very easy to make a metallic fuel than it is an oxide fuel. We as a reactor vendor make oxide fuel in four different countries around the world, so we do it well, but we know that it’s a complex, very exacting business whereas making PRISM fuel from a manufacturing ability standpoint is very easy.
And can the PRISM as proposed also be used to recycle waste and spent fuel, which the UK also has a lot of, as well as plutonium?
The PRISM reactor if it was built as what we put into the feasibility study, that kit has the ability to recycle fuel. However the UK government would have to buy a different kit, if you will, to do that recycle. So you would need a different building to do the recycling. That isn’t currently part of our offer, however by taking this first step with PRISM it allows the UK government – if policy changes – to use this recycling option. Because right now the UK government’s position on reprocessing or recycling is not to do either one. So that’s why we as a technology vendor haven’t offered that, but your country would be positioned to use the reactor for recycling if you bought the building and the technology to do that recycling.
So the reactor is the same, it’s just a matter of the reprocessing stage which needs to be carried out beforehand?
Yes, the reactor is the same. So that is the unique thing about PRISM is that it can run in a mode for this feasibility study, which is just to disposition plutonium, put it into the reactor, generate fission products so that it is radioactive, and then put it in the ground – that same reactor could be used for recycling, as an Integral Fast Reactor, which provides a huge, a bountiful amount of energy for this planet.
It also strikes me that wanting to spike plutonium with fission products and put it in the ground isn’t necessarily the most sensible use of potential fuel anyway, but I guess you’ve been asked to do that, and that’s what you’ve come up with, right?
That’s what we’ve come up with, and you have to realise that the PRISM reactor has an operational lifetime of many, many decades, as does the used fuel that would be stored there on site, so it’s not as if the reactor’s going to run and then they’re immediately going to put it in geological disposal. My vision is that once the technology is operating and people see carbon-free electricity coming out of it – and your question is, why don’t we build the recycling portion, and then it’ll come. I think it’s just too much right now. What your country has, we can solve the UK plutonium problem with PRISM, because our children deserve better than an old MOX plant.
What are the safety concerns though? People often are scared by the idea of having molten sodium used as a coolant because it’s reactive with air and water – what are some of the engineering issues on safety?
Sodium is actually very beneficial for the safety of the reactor – it provides greater heat transfer capability and that allows us to be able to cool the reactor with a complete loss of all external or onsite power. Because sodium as you know is a metal it’s very conductive of heat, and when we turn off a nuclear power plant it still generates heat. So our fuel generating heat, which is metal, is located in metal cladding, which is in metal coolant, which is in a metal vessel, so it’s very easy to pass air outside that metal vessel and remove the heat.
So that’s what makes it safe. Now unfortunately we make our high school students take chemistry. Most chemistry teachers like to put metallic sodium into water, and it does come up with an interesting reaction – but it’s a chemical reaction, and engineers are very good at containing chemical reactions, like chlorine gas that we use to purify our water, to ammonium used for fertiliser. Those chemicals we know how to control very well. So that’s where I think we need to overlook that chemistry experiment that we get to see with sodium. Sodium is a great reactor coolant.
It does presumably when it’s in the reactor core get radioactive itself. Does that present any concerns?
Sodium, just like water, becomes radioactive in the neutron flux around a reactor. In the case of water it turns into nitrogen-16, which has a half-life of 7 seconds. So if you turn off the reactor and you wait 70 seconds pretty much the radioactivity’s gone. In the case of sodium it produces sodium-24, and sodium-24 has a half-life measured in days – so it’s radioactive a little bit longer, but it’s not insurmountable. If you look at lead-bismuth cooled reactors, bismuth turns into polonium, and polonium has a lot longer half-life and a more significant release of alpha particles, so from a coolant standpoint it’s a little bit longer than water, but it’s something that’s definitely manageable and taken care of in the design of PRISM.
The green groups who’ve talked about this in the UK are quite sceptical that it will ever happen. Do you have any response or interaction with environmentalists about the PRISM?
At this point we’ve been focusing on the technical stakeholders at the NDA, and we realise that we have to overcome the pessimism about this third-generation technology for plutonium disposition rather than the second-generation technology which is called MOX. And we think that with continuing to inform stakeholders to continue to have that sort of dialogue that people will see the advantages of this technology for plutonium disposition for a variety of reasons, and then I think that people can see the vision of how this technology allows the UK to be a leader when it comes to the Integral Fast Reactor concept. That they’re the ones that unlocked this key to a different way of generating electricity, and also they’ve become the leaders on how to manage plutonium.
If you look at the United States and Russia, we’re still looking at each other working on the disposition of plutonium, but so far not one tonne has been dispositioned. The UK with this technology has the opportunity to disposition 100 tonnes in 5 years. And imagine the impact in a world discussion having the UK lead those sorts of efforts, and look at the United States and Russia and say ‘why can’t you do better?’. So I think this puts the UK in a leadership role in a variety of areas, not only in energy generation and how to control waste but also as far as the management of plutonium.
But perhaps one of the reasons why there is some scepticism in the UK is bad memories from our own fast reactor programme at Dounreay. How is the PRISM significantly different from that?
You all have an interesting history – the first reactor was metal-fuelled – good – with a loop – not as good. Then in the second reactor you had oxide fuel – bad – with a pool-type design – good. What PRISM is, is metallic fuel in a pool reactor. And that’s the best combination. So that’s where it’s significantly different. And what I mean by a pool, is that all of the coolant, the sodium, is located in one vessel. We surround that by another vessel, called a guard vessel. And we put that underground in a concrete silo lined with stainless steel. And from a fundamental standpoint that design prevents a loss of coolant accident. And that’s fairly significant that we’ve designed that out – that’s like saying that we’ve designed an airplane so that it cannot crash. So in this case we’ve designed it so that we don’t have a loss of coolant accident, and that has been substantiated by the US Nuclear Regulatory Commission in 1994 where they agreed with us that we’ve designed that out.
So from a safety standpoint, saying that we’ve designed out a loss of coolant accident –that’s significant. Look at Three Mile Island – they had a loss of coolant accident which led to that core damage.
Can we just talk quickly about the financing issues? Obviously central to a lot of the opposition to nuclear is the perception that it’s more expensive than everything else – what’s the proposal here? What’s the taxpayer liability likely to be for the PRISM proposal?
[David Powell replies:] What we’re doing is that we’re developing that technology such that we’ve got a strong private sector element involved in the project, in which we are going to be a service provider in processing the plutonium. So that in exchange for that we get a fee arrangement for the amount of plutonium we process. So this is effectively providing a strong insulation for government, NDA and ultimately the UK taxpayer from a plant construction, operation, ownership and other associated risks that you have with those kinds of projects. So effectively the model shifts from the UK taxpayer to a private funding-type model based on a pay per performance approach via an operator of the plant that we would use to process the plutonium and then sell the electricity that we generate.
So the electricity sold offsets the cost of spiking the plutonium essentially?
[David Powell replies:] To some extent yes. And recognising that it is low-carbon electricity as well, which is all beneficial from a UK perspective.
Do you have any idea what the capital cost of both the reactor and the fuel fabrication facilities might be?
[David Powell replies:] Yes we do but I’m not allowed to tell you actually! Because we’re not into any commercial negotiations at this point in time and it’s probably not right for us to say what those kinds of figures would be at this point.
But in terms of the incentive to the UK taxpayer to consider this, in what way is this better than some of the competing proposals? – I’m not asking you to discuss them, but the model here is different, right?
[David Powell replies:] Yes, the model’s different, and as we’ve concluded in the study report to the NDA, we’ve put a lot of information in there about the model and how that would work, and trying to explain the roles that each player would take in that. So for example GE-Hitachi would take a technology-provider role, and the operator would implement that strategy. And what we’ve been able to say, as Eric mentioned is that the fuel fabrication process is a very straightforward and simple process, and when coupled with PRISM, because we can generate that low-carbon electricity, we can do this at less cost than a new MOX fuel fabrication plant. You know, that [MOX plant] would consume power from the grid, and we’re actually generating power. So we’re able to say that we can provide better value than MOX.
Okay. And how quickly might this be able to be up and running in the best-case scenario? Is it going to take decades or years or what?
[Eric Loewen:] OK, so best case is – I always try to think in years – the process right now is that the NDA has 2 months to evaluate whether the technology is credible, and if they view that it is credible then we enter into the consultation or the justification process that starts this fall. It is my understanding that this process takes over 2 years to do that. And during that 2-year time, one way to shorten things up is to start the process of licensing with the Office of Nuclear Regulation, that would shorten it up some. If that isn’t done, then after that 2-year justification, if that technology was then chosen through the justification, then you’d set up a contract, and then that contract would enable the formal submission of a licence.
We think the licensing of PRISM and the two associated facilities – the fuel fabrication and the fuel storage – would take about 5 years, similar to a water-cooled reactor. And then construction would be about 5 years. So there’s your first decade, and then the operation would depend on how fast the disposition of the plutonium would occur. The life of the plant is about 60 years, so it would operate in that mode to disposition plutonium, or as we talked about earlier in this call, you could choose with a policy change that you could start recycling the PRISM fuel to extract more energy, to the point where you put no plutonium into the ground.
So that’s clearly a game-changer sort of technology – if you look at MOX, you will put more transuranics into the ground than when you started – it’s just a fact of how water-cooled reactors work. With PRISM in the Integral Fast Reactor mode you can essentially put no plutonium back into the ground in a geological repository – all that plutonium would be used for energy input into our planet.
I can’t remember what the power of each block is, is it 600 each, or 300 each and you’re proposing two?
A block is 600 megawatts, and a block consists of two reactors that each produce 300.
Why is it so small? Why not do a bigger one?
Because sodium-cooled reactors are better small. And the reason for that is that if they’re small then the reactor cores are small, and if the reactor core is small that means you get neutron leakage. And if you get neutron leakage you have the ability for the reactor to shut down. And those are the experiments that were done at the Experimental Breeder Reactor no.2 when the control rods were held out. The plant was at 100% power, 100% flow, and then the reactor coolant pump was turned off. And the reactor shut itself down, at a lower power, without inserting any control rods. That same scenario for any water-cooled reactor would be significantly different… and then that test was done that afternoon with losing the heat sink, where you turn off its ability to reject heat. And it’s because having that small core that allows you to do that. So from a fundamental standpoint sodium-cooled reactors are good small. From a fundamental standpoint water-cooled reactors are better bigger. And that’s where PRISM has found that ideal spot of about 300 MW electric in a reactor vessel that’s about 30 feet in diameter, it’s about 60 feet tall.
It sounds more expensive to build lots of little things than one big thing, but isn’t the proposal for it to be modular?
When we started PRISM in 1981, we started it as a technology company that built things in the factory like jet engines, like gas turbines. The reason why we like building things in a factory is that we can control quality, we can control cost, and we can make sure that we put a product out that is going to work. That’s kind of the genesis of how PRISM started in 1981. Then we invited Mike Lineberry from Argonne National Laboratory out to our Sunnyvale site later in 1981, and that’s where the PRISM concept kind of got formalised. It wasn’t until 1983 that it started getting more insight, and in ’84 is when it actually got funded. But when we started in 1981 the concept was ‘how do you make something in factories that you can ship down the road?’ – so your limiting factor is how big that reactor vessel is. And that’s where this limit’s coming from, that’s where you get the modularity. We know from the other products that we make in the factory that the delivery times get shorter and costs start going down, when we start manufacturing things that are small and modular in a factory.
Thanks to you all very much.