Nuclear power will feed on itself

I hope that Joshua Pearce’s Nuclear cannibals is wrong. His conclusions are very different from the MIT study “The future of Nuclear Power“. A full life-cycle analysis is required to correctly assess any proposed energy supply chain – that is evidently what Pearce has undertaken, though I don’t know the details.

I’ve found the peer-reviewed paper upon which this short summary is based: “Thermodynamic limitations to nuclear energy deployment as a greenhouse gas mitigation technology“. But I’m not willing [yet] to pay EU 30.00 to purchase the paper. I’ll be looking for reviews and commentary on the paper.

Nuclear energy production must increase by more than 10 percent each year from 2010 to 2050 to meet all future energy demands and replace fossil fuels, but this is an unsustainable prospect. According to a report published in Inderscience’s International Journal of Nuclear Governance, Economy and Ecology such a large growth rate will require a major improvement in nuclear power efficiency otherwise each new power plant will simply cannibalize the energy produced by earlier nuclear power plants.

Physicist Joshua Pearce of Clarion University of Pennsylvania has attempted to balance the nuclear books and finds the bottom line simply does not add up. There are several problems that he says cannot be overcome if the nuclear power option is taken in preference to renewable energy sources.

For example, the energy input required from mining and processing uranium ore to its use in a power plant that costs huge amounts of energy to build and operate cannot be offset by power production in a high growth scenario. There are also growth limits set by the grade of uranium ore. “The limit of uranium ore grade to offset greenhouse gas emissions is significantly higher than the purely thermodynamic limit set by the energy payback time,” he explains.

In addition, nuclear power produces a lot of heat as a byproduct and this directly heats the Earth. This is only a relatively small effect, but as energy consumption grows it must be taken into consideration when balancing the energy equation.

However, it is the whole-of-life cycle analysis that Pearce has investigated that shows nuclear power is far from the “emission-free panacea” claimed by many of its proponents. Each stage of the nuclear-fuel cycle including power plant construction, mining/milling uranium ores, fuel conversion, enrichment (or de-enrichment of nuclear weapons), fabrication, operation, decommissioning, and for short- and long-term waste disposal contribute to greenhouse gas emissions, he explains.

Nuclear may stack up against the rampant fossil-fuel combustion we see today, but only by a factor of 12. This means that if nuclear power were taken as the major option over the next forty years or so, we would be in no better a position in terms of emissions and reliance on a single major source of energy than we are today given the enormous growth nuclear required over that timescale.

Pearce’s analysis is based on current practice in the United States with regard to the mining and enrichment of ore. He suggests that rather than abandoning nuclear power, efforts should be made to improve its efficiency considerably. First, we could start utilizing only the highest-concentration ores and switch to fuel enrichment based on gas centrifuge technology, which is much more energy-efficient than current gaseous diffusion methods.

Nuclear plants might be used as combined heat and power systems so the “waste” heat is used, rather than allowing them to vent huge quantities of heat to the environment at the end of the electricity generation cycle. Pearce also suggests that we could “down-blend” nuclear weapons stockpiles to produce nuclear power plant fuel.

If in fact there are real thermodynamic limitations on nuclear power, then good policy options are few. There is no other proven, economic and well-understood “carbon-free” power generation option that can be deployed at the required scale.

Technorati Tags:

10 thoughts on “Nuclear power will feed on itself

  1. He’s an idiot.

    He’s worried about the waste heat from nuclear power when all energy utilization in human civilization from nuclear to coal amounts to less than .1% of the solar insolation. Its something to be concerned about in a thousand years when civilization is a thousand times as hungry I suppose.

    If he used publicly avaliable data to measure the energy inputs of uranium mining he would find the Rossing mine in Nambia which mines ore of 300ppm (the lowest ore grade mined today) consumes 1/500th the energy that is produced in light water reactors today from the fuel produced in that mine. Using the standard industry assumption that energy costs are inversly proportional to the ore grade, ore with an energy payback of 15-30 fold burnable in light water reactors numbers some 1 trillion tons, enough to power all of civilization for some 250000 years.

    His analysis is innumerate nonsense with an agenda.

  2. Joshua Pearce is an assistant professor at Clarion and environmental activist whose primary focus is solar power. It’s not surprising to find him down on nuclear power and indeed, his conclusions are ludicrous. Nuclear power is hundreds of millions of times more energetic than chemical power; the idea that there’s some net “energy imbalance” from mining ore is laughable at best.

    Hundreds of physicists with actual experience in nuclear power have long since pointed out that the technology is sustainable for many thousands of years of power generation.

    Let’s look at his “heating the earth” claim. I find it shocking anyone with even a basic understanding of science could make such an alarmist claim.

    The earth receives a constant 174,000 terawatts of heat energy from the sun (plus an unknown amount more from geothermal sources).

    Total world electricity production is under 2 terawatts. Assuming 40% efficiency (most reactors run well above this), that works out to a grand total of 5 TW of heat energy.

    Add that to the solar constant, and it works out to to an increase factor of 0.00002. According to the Stefan-Boltzmann law, temperature rises by the fourth root, meaning a temperature differential of 1.000004. Good luck measuring that!

    Furthermore, the true value is even lower, as any nuclear plant replacing a coal or gas-fired one replaces its waste heat as well, resulting in essentially zero increase for those.

  3. Dez and Michael,

    It is clear that neither of you have actually read my study. First concerning the direct thermal heating — I very clearly said this was not currently a problem and would not be a problem until very significant increases in global energy demand are seen..especially when compared to GHG thermal forcing in the paper.

    Second, concerning motivation – I was asked to write the article applying the same LCA work I normally do with solar PV to nuclear. I used the same guidelines that I would in any of my studies – all my sources for the energy input at each stage of the nuclear life cycle are very clearly cited and in the peer reviewed literature. My conclusions are a list of solid recommendations to improve the nuclear energy plant efficiency – which currently is unacceptably low from a climate change perspective.

    I never said that a given nuclear reactor does not produce far more energy than it consumes to construct and operate — it does. What the study did show is that the growth rate that the nuclear industry would need to undertake to replace fossil fuels was high enough that the energy created by the existing nuclear power plants would not be enough to cover the emissions caused by the construction of new ones. Stated another way – The nuclear power plant ensemble will not produce any net energy if the cannibalistic energy is equivalent to the total energy produced. You can easily follow my math that this occurs when the reciprocal of the growth rate is equal to the energy payback time. This is a subtle relationship – that to the best of my knowledge was never really considered before. The ramifications, however, are huge. Although nuclear energy production has relatively minor emissions — as compared to say a coal plant — you can not ignore them. Nuclear is NOT an emission free source.

    Past studies on the nuclear life cycle showed a range of energy payback times and it was clearly a function of the type of analysis used. In the paper I included both types (those funded by the industry itself came up with much lower numbers and tended to ignore energy costs like decommissioning) and other that were more thorough. To be correct you must include the emissions from the entire life cycle – mining is not enough. For the actual analysis I used the most realistic numbers for the U.S. (which is actually considerably less efficient than say the French nuclear industry).

    Please read the article – if you do find any errors or ways that I could improve the analysis I would be grateful for your feedback.


  4. There are perhaps dozens of life cycle analysis reports on nuclear power. They nearly all (with the exception of the (in)famous Storm van Leeuwen & Smith analysis) conclude that nuclear power stacks up favorably. It isn’t clear what this one brings to the table that is new or what his assumptions are (e.g. does he assume some fuel enrichment will continue to be done using old inefficient gas diffusion technology or newer centrifuge and laser separation). Moreover, without putting these life cycle emmissions in context with those of other sources such a coal, natural gas, PV, and wind no meaningful conclusion can be drawn. If you started to scale up ANY power source, such as PV, of course some of the power produced by it would be cannabalized to support the construction of new factories and production lines for yet more PV cells (or wind turbines or…).

  5. Thanks to all of you for the comments.

    I neglected to reference one of the more useful life cycle analysis sources that I value, a November 2006 meta-analysis by Sydney University Nuclear energy: life cycle analysis of energy and greenhouse gas intensities. The study, commissioned by the Australian Government, aggregates a large body of prior LCA work on the full range of operational power generation options. The authors’ methodology is hybrid input-output-based life-cycle assessment (IO-LCA).

    The LCA conclusions for light water reactors are energy intensities of 0.18 kWh_th/kWh_el compared to e.g. solar PV of .33, or hydro of 0.05. The corresponding energy payback time of nuclear energy is around 6½ years for light water reactors.

  6. Since the process is already in hand to switch to centrifuge enrichment, and downblending weapons material for reactor use has been going on for years now, two at least of the conclusions described in the article are not that remarkable.

    This appears to be a “cash-flow” type question, and the point of displacing other activities is well made by gumsh0e. The break-even for construction of plants is much shorter than given; the Sydney 6.5 years includes lifetimes worth of fuel enrichment and decommissioning, neither of which are a cash-flow type issue in infrastructure build-up.

    All of which is based on the summary (and comments here) as I don’t have the spare cash to buy the study and read it in full.

    The caution I’d offer on the Sydney study is that, despite Storm van Leeuwen & Smith (SLS)’s very poor record in every real-world comparison, the Sydney group still gave SLS some weight and used their figures when real-world data was sparse.

  7. Joffan,

    Thanks for your informed comments. Your point on the Sydney energy payback time is correct. I’ve studied the Sydney report attempting to determine the time of energy breakeven on a proper “cash flow” basis. As far as I can tell the necessary data isn’t disclosed in the report. It may be in their associated spreadsheets, but I don’t think so. A breakeven estimate must be in the Pearce paper, but like you I’ve not been willing to pay the EU30 to acquire the full paper.

    Pearce raises an important question — i.e., what is the net energy contribution over time of a projected large scale nuclear deployment? AFAIK that projection has not been published. If anyone can link to an objective paper on that please comment.

  8. Fundamentally, nuclear power is outdated, failed technology. It is not the solution to global warming–which requires major changes within the next 10 years. Whereas nuclear power would take at least 40 to produce a meaningful impact on climate change.

    This is without mentioning that the last nuclear reactor yet to be built in Finland has taken over 10 years, with endless delays and technical/construction problems, and BILLIONS of dollars over budget.

    Nuclear power is a money suck, a time suck, and apparently an energy suck. I find it ironic that people are willing to buy the nuclear industry’s endless propaganda on their failed technology. And also buying the propaganda from nuclear industry paid scabs like Patrick Moore who takes thousands (if not millions) from the industry to push their sub-standard product.

    There is no independent research that supports nuclear power as a viable solution, and yet you people remain totally brainwashed. Amazing.

  9. Thanks for your comments. There are some problems with your analysis. Starting with France, where 80% of electrical generation is nuclear, and residents around nuclear sites are keen to have more nuclear construction.

    Second, nuclear has not yet transitioned to mass-production of pre-certified modules. That is a totally new world — read the MIT “Future of Nuclear Power” study. Then see what you think.

  10. Sorry I’m late to this party. I think Joshua Pearce should make his paper web-accessible, as I have done with my ‘How fire can be tamed’.

    It is principally about a zero-local-emission internal combustion scheme that I predict motorists will demand as soon as they can, but also outlines a fission-powered production method for the zero-local-emission combustibles, and this method does not depend on any isotopic enrichment.

    Thus it avoids completely, as the CANDU scheme avoids almost completely, the energy input of isotope separation, and — more important, in my view — avoids the logistical difficulty that separation poses.

Comments are closed.