Fuel review: economics

A catalyst for change?

1 October 2012

Switching to nuclear energy, a powerful source of low-carbon power, at a global level could help reduce climate change. But at that scale, current LWRs could burn through economically-extractable uranium relatively quickly, raising the price of fuel and ultimately of the power itself, as well as security of supply issues. These risks could be major drivers of development of fast neutron reactors. By Gregg Butler

Over the last half-century, light water reactors (that is, boiling water reactors, BWRs, and pressurised water reactors, PWRs), fuelled with low enriched uranium oxide, have become the mainstay of world nuclear programmes.

Most international projections for future nuclear power generation [1] are extrapolations from current plans (that is, business as usual—‘BAU’) rather than being aimed at ameliorating climate change, and have the effect of reducing energy carbon emissions by around 50%. Such programmes envisage around 1200 GWe of nuclear by 2050, largely using developments from existing LWR technology with used fuel being disposed of in deep geological facilities: the ‘once-through’ fuel cycle.

While moderately ambitious from a ‘business as usual’ perspective, these programmes are nowhere near rapid enough to support decarbonisation sufficient to satisfy global warming criteria even approaching the 2°C aim of international actions enshrined in the Copenhagen Accord [2] to restrict the harmful effects of climate change. In fact, current emissions are on track for a 6°C rise by 2100, with potentially catastrophic results [3-6]. Most effective decarbonisation programmes envisage much higher nuclear generation. As an example, the World Nuclear Association Nuclear Century (2012) envisages around 2400 GWe being required by 2050 [7].

Uranium resources

‘Once-through’ LWRs are relatively profligate users of uranium, with around 85-90% discarded as tails at the enrichment stage, and only some 0.6% of the mined uranium actually being consumed by the fission reaction. LWR uranium usage varies with the price of uranium and enrichment, but a 1 GWe reactor operating at 90% load factor will typically use around 180 teU per annum. With a prospective life of around 60 years, this 1 GWe reactor would commit around 11,000 teU for its lifetime.

This means that the ‘BAU’ programmes would use around 5.7 million teU from 2010 to 2050, while the WNA programme (which is not remotely extreme if 2°C is even slightly likely to be reached) would give LWR uranium usage of nearly 11 million teU in the period 2010-2050, and commits (assuming a 60-year reactor life) to some 24 million teU for new reactors built by 2050. The amounts of uranium mentioned can be compared with supply estimates of around 6.3MteU at a price of $100/lb U3O8 [8]. Thus the cost, and security of supply, of uranium over a reactor’s life could become a crucial factor both in the investment decision on individual reactors and in the delivery of LWR programmes.

One well-proven method of reducing uranium usage is to reprocess the LWR fuel and recycle the plutonium and uranium into the LWRs as mixed uranium plutonium oxide (MOX) and reprocessed uranium (RepU) fuel. However, at equilibrium this ‘thermal recycle’ typically only reduces uranium usage by around 20% (increasing the 0.6% uranium usage to around 0.72%). While further recycles can increase this, the overall improvement remains fairly trivial in global resource terms.

A 2012 Smith School study ‘Towards a low carbon pathway for the UK’ [9] gives an LWR scenario that accounts for around 60% of the world’s electricity supplies by 2060. This would commit 25 million teU by 2050, and 98 million teU by 2100; that is between four and 15 times the currently-estimated reserves at $100/lb U3O8. The ability of LWRs to effectively contribute to global decarbonisation is therefore inextricably linked to the availability of economically extractable uranium.

System options

A large variety of nuclear reactors and fuel cycles were developed to at least pilot plant scale in the 1950s to the 1980s: thermal and fast reactors, uranium, uranium oxide, MOX and thorium fuels, light water, heavy water, and graphite moderation, light water, heavy water, liquid metal, carbon dioxide and helium cooling—the list is not endless, but very substantial. Virtually all of these systems used either direct or indirect state funding, at least to the stage of the first full-scale demonstration reactor. Some of them, like Magnox and AGR in the UK, and CANDU in Canada, were developed to the scale of a commercial fleet.

Of this extensive range of systems, LWRs have become the nuclear reactors of choice, with over 80% of current world nuclear generation. They have benefitted from five decades of incremental development, with longer reactor life, increased fuel burn-up, and fewer fuel failures—a process which has mirrored the functionality and reliability developments that have occurred over the same period in cars, aircraft, and home appliances.

As well as operational developments, safety standards and regulations have tightened, so that the entry point for a new system would entail passing very stringent tests even to begin development. The days of first-of-a-kind systems being tried out on remote sites to ‘learn by doing’ has, certainly in the Western world, well and truly ended.

This means that a ‘new kid on the block’ faces very considerable barriers to entry, and would need to be funded to well beyond the pilot plant stage before it could hope to present a reasonable investment opportunity in the market. A new system must therefore have at least one major driver (be it economic, environmental, or whatever) to have an expectation of success. These drivers could be envisaged as:

  • Sustainability. This relatively nebulous term could be taken to mean ‘capable of being operated within likely material supply constraints, while contributing to reducing carbon emissions’. Uranium supply is discussed later in this paper as the most plausible ‘game changer’.
  • Reduced radiotoxicity of waste. The less harmful the waste, the easier it is assumed to be to dispose. Provided that a robust safety case can be made for geological disposal of LWR spent fuel, then the amount of LWR fuel to be disposed is unlikely to be limiting. Any advantage from reduced radiotoxicity of waste would then flow from an increased likelihood of geological disposal being societally, rather than scientifically, achievable.
  • Enhanced proliferation resistance and inherent physical protection. Security is always an important part of fuel recycling programmes, but few such systems are likely to offer significant advantages when compared with a ‘once-through LWR’ base case.
  • Improved economics. It is unlikely that any alternative system could undercut the economics of the ‘once-through LWR’ until either the new systems had been operating fully for perhaps some tens of years, or until some change in circumstances made LWRs very significantly less economic.
  • Plutonium disposition. Superior performance for plutonium disposition has also been mooted for some countries, like the UK, though how this might be measured is less than clear.

Most ‘new’ systems claim one advantage or another, but all cases need to be stringently examined to assess whether these advantages are significant enough to generate funding in the amounts, and over the timescales, required to overcome the entry barriers.

Also, many of the advantages are only realised when the system is at equilibrium. For example, although some recycling systems may output relatively short-lived waste, the operating system also contains a significant inventory of spent fuel, contaminated and activated reactors, and fuel cycle plant. Thus though the advantages may be there, many decades may pass before they are fully realised.

The effectiveness of these growing barriers to entry is illustrated by the fact that many ‘new’ systems are not in fact new at all, having been developed to prototype reactor stage in the 1960s-1980s, only to be re-developed in the 1990s-2000s, and again fail to become accepted into the mainstream. The South African PBMR project is a relatively recent case in point. Recent proposals have involved metal-fuelled sodium-cooled fast reactors, molten salt thorium reactors, and accelerator-driven systems, and while the flow of novel or revamped schemes will doubtless continue, the challenge facing ‘new’ systems should not be underestimated.

There will, of course, be exceptions. It is not difficult to see small and medium reactors (SMRs) succeeding in the niche requiring relatively small generators to serve small unintegrated grids. Considerations of national resource and energy security may trump economics, such as plans for thorium power in India. Also, it is not difficult to see game-changers significant enough to change the whole paradigm of regional power generation. Cheap superconducting power lines and/or TWh-scale energy storage would rewrite the rulebook, but it would take a brave person, organisation or state to bet serious outcomes on these developments coming to early fruition—and very early fruition would be required for new reactors to have any meaningful effect on climate change.

Perhaps the most clear-cut example of a potential entry driver is the fast reactor system, whose outstanding advantage is that it uses far less uranium per unit power generated than current LWRs. By essentially burning uranium-238, a fast reactor uses 50-60 times less uranium than an LWR, allowing the generation of 50-60 times more energy from any given amount of uranium.

However, fast reactors rely on fuel reprocessing and fuel re-fabrication plants, and the systems involved are a long way from commercial optimisation. They face challenges on the technical process, materials and licensing fronts, and the problems of restricting the risk of diversion and/or proliferation. There are many proposals for ‘proliferation resistant’ fuel cycles, but these are even more novel and complex, making the economics more uncertain and the barriers to market entry even higher.

Uranium resources

From the above, the most easily-assessed limitation in the use of the ‘once-through LWR’ system would be whether the deployment of these reactors is likely to be limited by restrictions in uranium supply. Put simply, if nuclear is to make significant inroads into the world energy market, to the extent that it materially assists in decarbonising energy supply, is there enough uranium to support this using LWR reactors as currently understood?

There is absolutely no problem with the overall amount of uranium on the planet, with 80x1012 teU estimated to be in the crust of the earth to a depth of 25km [10]. It is the amount that is recoverable economically, both from a financial and carbon standpoint, which is crucial.

A recent study by MIT, ‘The future of the nuclear fuel cycle,’ examined a world LWR programme which postulated building 3050 GWe of LWRs and operating them for 100 years, giving a cumulative uranium usage of around 61 million teU, or ten times currently-estimated reserves [11].

The study concluded: “Our analysis of uranium mining costs versus cumulative production in a world with ten times as many LWRs and each LWR operating for 100 years indicates a probable 50% increase in uranium costs. Such a modest increase in uranium costs would not significantly impact nuclear power economics.” Certainly, if this does describe the economics of uranium, it is difficult to imagine what could halt the pre-eminence of LWRs.

However, recent work by the University of Manchester [12], initially funded by the UK’s EPSRC’s SPRing Project [13], has identified issues with the MIT methodology (see box). This work examined LWR programmes similar to those in the Smith School study (and therefore also similar to MIT), and concluded that that separation cost (for example, milling) would rise more quickly that predicted by MIT, and would eventually grow very rapidly. The modelling undertaken indicated that uranium cost rises would begin to significantly affect ‘once-through LWR’ economics at 40 MteU total consumption, and would become prohibitive at 95 MteU. If this were so, then the MIT programme, or any of the nuclear programmes aimed at giving major decarbonisation, would run the risk of being severely affected by uranium cost.

Another sensitive factor, not modelled in the University of Manchester study, is that any increase in uranium cost would be largely driven by the energy input into the mining and milling process, which would itself lead to increased carbon emissions to the extent that this energy was supplied by fossil fuel, as is currently the case for most uranium extraction. Studies of the use of very low-grade ores (0.001%) indicate that carbon emissions could rise to over 20% of unabated natural gas generation [14], making nuclear around four times as carbon-intensive as wind power, and significantly less effective as a decarbonisation energy source.

So: the most clear-cut driver for a new system, improved uranium usage, is either a valid driver (University of Manchester study) or an irrelevance (MIT study). This surely indicates the need for a broad-based study on uranium resources, economics and carbon-intensity (as indeed advocated by MIT). More fundamentally, it suggests the need for a fundamental retargeting of research into nuclear power reactors and fuel cycles, away from ‘roadmaps’ which concentrate on the ‘what’ and ‘how’, and towards studies which concentrate on the fundamental drivers for adopting new systems, the ‘why’. In particular, future work by the University of Manchester will, hopefully with wider involvement, examine the economics and kinetics of the handover from LWRs to recycling reactor systems over a range of programme, economic and uranium resource assumptions.


Professor Gregg Butler (Dalton Nuclear Institute, University of Manchester and Director, IDM Ltd).

Acknowledgements: The author would like to thank Prof Kevin Anderson (Deputy Director of the Tyndall Centre for Climate Change Research) for clarifying the Climate Change aspects, Prof Sydney Howells (Manchester Business School) for making the modelling available, and Grace McGlynn (Director, IDM) for involvement in developing the concepts, together with ensuring transparency and readability. This article was published in the September 2012 issue of Nuclear Engineering International.



1. e.g. Energy Electricity and Nuclear Power Estimates to 2050, IAEA 2010

2. Copenhagen Accord, 2009. FCCC/CP/2009/L.7. United Nations Climate Change Conference 2009, Copenhagen. Most nations (~150) are signatories to the clear statement that the international community must "hold the increase in global temperature below 2 degrees Celsius, and take action to meet this objective consistent with science and on the basis of equity"

3. E.g. Fatih Birol, chief economist at the IEA commented "When I look at this data, the trend is perfectly in line with a temperature increase of 6°C by or before 2100], which would have devastating consequences for the planet. http://uk.reuters.com/article/2012/05/24/co2-iea-idUKL5E8GO6B520120524 - May 2012

4. Anderson, K., and Bows., A., 2011, Beyond dangerous climate change: emission pathways for a new world, Philosophical Transactions of the Royal Society A, 369, 20-44, DOI:10.1098/rsta.2010.0290

5. New, M., Liverman, D., Schroder, H., Anderson, K., 2011, Four degrees and beyond: the potential for a global temperature increase of four degrees and it implications Phil. Trans. R. Soc. A 2011 369, 6-19

6. Anderson, K. and Bows, A., 2008, Reframing the climate change challenge in light of post-2000 emission trends, Philosophical Transactions A, 366, 3863-3882.

7. World Nuclear Association Nuclear Century Outlook, see http://www.world-nuclear.org/outlook/nuclear_century_outlook.html

8. Uranium 2009: Resources, Production and Demand, OECD Nuclear Energy Agency and the International Atomic Energy Agency, 2010

9. Towards a low carbon pathway for the UK, Sir David King, Gregg Butler, Michael Evans, Oliver Inderwildi, Grace McGlynn; Smith School for Enterprise and the Environment, March 2012

10. K S Deffeyes and I D MacGregor, "World Uranium Resources", Sci. Am., Vol 242, I, 66 (1980), reported in ‘Long-Term Uranium Supply Estmates, E A Schneider and W C Sailor, Nuclear Technology, Vol 162, June 2008

11. The Future of the nuclear fuel cycle, various authors, MIT, 2011

12 Challenges for worldwide nuclear programmes: some major technical and economic constraints presented by various fuel cycles, G. G. Butler, S. D. Howell, P. V. Johnson, S. J. Hall, D.W. Liu, P.W. Duck, I Chem E Nuclear fuel cycle conference 23-25 April 2012, Manchester, UK, Paper NFCC042, see http://www.icheme.org/events/conferences/nuclear%20fuel%20cycle%20conference/programme.aspx

13. Sustainability Assessment of Nuclear Power: An Integrated Approach: http://www.springsustainability.org/ See, for example, Joint economic and physical constraints on nuclear power: How much Uranium would be needed to decarbonize? David Liu, Gregg Butler, Stuart Hall, Paul Johnson, Peter Duck, Geoff Evatt1, and Sydney Howell, Journal of Power and Energy Volume 226 No 3, May 2012

14. Wiberg, 2009, reported in The Role of Nuclear Energy a Low Carbon Future, NEA/OECD, 2012


Yellowcake Yellowcake
The disparity in efficiency between nuclear and wind power depends on the cost of uranium extraction The disparity in efficiency between nuclear and wind power depends on the cost of uranium extraction

Privacy Policy
We have updated our privacy policy. In the latest update it explains what cookies are and how we use them on our site. To learn more about cookies and their benefits, please view our privacy policy. Please be aware that parts of this site will not function correctly if you disable cookies. By continuing to use this site, you consent to our use of cookies in accordance with our privacy policy unless you have disabled them.