You are here

Can nuclear power slow down climate change? An analysis of nuclear greenhouse gas emissions

Nuclear Monitor Issue: 
Jan Willem Storm van Leeuwen − Ceedata Consultancy

This is a summary of a November 2015 report commissioned by the World Information Service on Energy (WISE). The full report is posted at

Nuclear power is claimed to be nearly carbon-free and indispensable for mitigating climate change as a result of anthropogenic emissions of greenhouse gases. Assuming that nuclear power really does not emit carbon dioxide CO2 nor other greenhouse gases (GHGs), how large is the present nuclear mitigation share and how large could it become in the future? Could the term 'indispensable' in this context be quantified? These issues are assessed from a physical point of view, economic aspects are left outside the scope of this assessment.

How large is the present nuclear mitigation share?

The global GHG emissions comprise a number of different gases and sources. Weighted by the global warming potential of the various GHGs 61% of the emissions were caused by CO2 from burning of fossil fuels for energy generation. Nuclear power could displace fossil-fuelled electricity generation, so hypothetically the maximum nuclear mitigation share would be 61% if the global energy supply were to be fully electric and fully nuclear.

In 2014 the nuclear contribution to the global usable energy supply was 1.6% and consequently the nuclear mitigation share was 1.0%.

The International Atomic Energy Agency (IAEA) asserts that the nuclear contribution to the global energy supply was 4.6% in 2014. However, this figure turns out to be based on a thermodynamically inaccurate statistical trick using virtual energy quantities.

How large could the nuclear mitigation to climate change become in the future according to the nuclear industry?

We found no hard figures on this issue, for that reason this study analyses the mitigation consequences of the envisioned developments of global nuclear generating capacity. During the past years the International Atomic Energy Agency and the nuclear industry, represented by the World Nuclear Association (WNA), published numerous scenarios of global nuclear generating capacity in the future, measured in gigawatt-electric GWe. Four recent scenarios are assessed in this study, as these can be considered to be typical of the views within the nuclear industry:

  • IAEA low: the global nuclear capacity remains flat at the current level until 2050.
  • IAEA high: the global nuclear capacity grows to 964 GWe by 2050, nearly three times the current global capacity of 333 GWe.
  • WNA low: the global nuclear capacity grows to 1140 GWe by 2060 and to 2062 GWe by 2100.
  • WNA high: the global nuclear capacity grows to 3688 GWe by 2060 and to 11046 GWe by 2100.

The nuclear mitigation share in the four scenarios depends not only on the nuclear generation capacity, but also on the growth rate of the global GHG emissions. The IAEA expects a growth rate of the global energy consumption of 2.0−3.5% per year until 2050. This study assumes that the global GHG emissions will grow during the next decades proportionally to the global energy consumption: also at 2.0−3.5% per year. Based on this assumption – and still assuming nuclear power is free of CO2 and other GHG emissions (which it is not) – the mitigation shares would be as follows, the high figure at a global growth of 2.0%/yr, the low figure at 3.5%/yr:

  • IAEA low: 0.5−0.3% by 2050.
  • IAEA high: 1.4−0.9% by 2050.
  • WNA low: 1.4−0.7% by 2060 and 1.1-0.3% by 2100.
  • WNA high: 4.5−2.4% by 2060 and 6.2-1.8% by 2100.

What next after 2050?

The IAEA scenarios are provided through 2050. Evidently the nuclear future does not end in 2050. On the contrary it is highly unlikely that the nuclear industry would build 964 GWe of new nuclear capacity by the year 2050 without solid prospects of operating these units for 40-50 years after 2050.

How does the nuclear industry imagine development after reaching their milestone in 2050? Further growth, leveling off to a constant capacity, or phase-out? Or: let tomorrow take care of itself?

What global construction rates would be required?
By 2060 nearly all currently operating nuclear power plants (NPPs) will be closed down because they will reach the end of their operational lifetime within that timeframe. The current rate of 3−4 GWe per year is too low to keep the global nuclear capacity flat and consequently the global nuclear capacity is declining. To keep the global nuclear capacity at the current level the construction rate would have to be doubled. The average global construction rates that would be required in the industry scenarios are:

  • IAEA low: 7-8 GWe per year until 2050.
  • IAEA high: 27 GWe/yr until 2050.
  • WNA low: 25 GWe/yr until 2060 and 23 GWe/yr from 2060 until 2100.
  • WNA high: 82 GWe/yr until 2060 and 184 GWe/yr from 2060 until 2100.

In view of the massive cost overruns and construction delays of new NPPs that have plagued the nuclear industry for decades it is not clear how the required high construction rates could be achieved.

How are the prospects of new advanced nuclear technology?

The nuclear industry promises the application within a few decades of advanced nuclear systems that would enable mankind to use nuclear power for hundreds to thousands of years. This promise concerns two main classes of closed-cycle reactor systems: uranium-based systems and thorium-based systems. For reasons discussed in the detailed version of this report, Uranium-Plutonium as well as Thorium-Uranium breeder concepts turn out to be based on inherently unfeasible assumptions. From this observation it follows that nuclear power in the future would have to rely solely on conventional once-through reactor technology based on natural uranium. As a consequence the size of the uranium resources will be a restricting factor.

How much uranium would be needed to sustain the various scenarios?

As pointed out above the nuclear generating capacity in the scenarios will not fall to zero at their end date.

The minimum amounts of uranium that would be required in the IAEA scenario's are estimated here by assuming no new NPPs would be build after 2050 and consequently the nuclear power plants operational in 2050 would be phased out by 2100. In case of the WNA scenario's extension after 2100 seemed too speculative. The masses of uranium are given in teragram (Tg): 1 Tg is 1 million metric tonnes.

  • IAEA low: 2.3 Tg until 2050 plus 1.7 Tg during phase-out by 2100, total 4.0 Tg uranium
  • IAEA high: 4.5 Tg until 2050 plus 4.8 Tg during phase-out by 2100, total 9.3 Tg uranium
  • WNA low: 6.6 Tg until 2060 plus 12.7 Tg from 2060 until 2100, total 19.3 Tg uranium
  • WNA high: 17.5 Tg until 2060 plus 58.4 Tg from 2060 until 2100, total 75.9 Tg uranium.

Obviously the uranium demand in the IAEA scenarios would be higher if the nuclear capacity were to remain flat after 2050, as opposed to phasing out after 2050 as assumed above; in case of a constant capacity after 2050 the total demand would be about 5.7 Tg in IAEA low and 14.1 Tg in IAEA high.

The known recoverable uranium resources of the world in the cost category of up to 130 USD/kg U amounted to 5.9 Tg in 2013 according to the IAEA; the market price in September 2015 was about 82 USD/kg U. An additional amount of 1.7 Tg of uranium is known to exist in the higher cost category 130-260 USD/kg U.

How are the prospects of the global uranium supply?

Uranium in the earth's crust is unevenly distributed among the rocks comprising the crust. The grade distribution of uranium in uranium-bearing rocks in the earth's crust show a geologic pattern common to other metals: the lower the grade of uranium the larger the amounts of uranium present in the crust. The size distribution of uranium deposits show a similar pattern as a result of the geologic ore-forming mechanisms: the larger the size, the more rare the deposits. From this observation it follows that the chance of discovering new resources increases with lower grades and smaller sizes of the deposits. One may assume that the most easily discoverable resources have been found already and that most easily minable deposits are already being mined. The chances of discovering new large high-grade resources seem low; in reality no such discoveries have been reported during the past two decades.

Based on a simple economic model the nuclear industry states that the global uranium resources are practically inexhaustable, apparently suggesting that any scenario could be materialized. However, the generation of nuclear energy from uranium resources is a physical phenomenon governed by the laws of nature, not by economic notions. The economic model does not include physical and chemical realities with regard to uranium deposits in the earth's crust. Thermodynamics sets the boundaries for the resources that fit the conditions of uranium-for-energy resources.

What are the thermodynamic boundaries of uranium-for-energy resources?

Energy cliff

The energy content of natural uranium that is in any sense extractable is limited: the nuclear power stations that would form the backbone of future nuclear capacity could not fission more than about 0.6% of the nuclei in natural uranium.

The thermodynamic boundaries of the uranium-for-energy resources are determined by the energy required to extract uranium from the resources as found in nature. Analysis of the physical and chemical processes needed to recover uranium from the earth's crust and all the processes needed to release the potential energy in uranium and convert it to useful energy proves that the amount of energy consumed per kg recovered natural uranium rises exponentially with declining ore grades. Below a grade of 200−100 ppm (0.2−0.1 grams U per kg rock) no net energy can be generated by the nuclear system as a whole from a uranium resource, this relationship is called the energy cliff. From this conclusion it follows that only uranium resources at grades higher than 200 ppm (0.2 g U/kg rock) are actually energy sources.

The ore grades of the known uranium resources which are by definition economically recoverable varies widely: from about 200 down to 0.1 gram uranium per kg rock. A part of the resources classified by the IAEA as 'recoverable' falls beyond the thermodynamic boundaries of uranium-for-energy resources.

Unconventional uranium resources

The nuclear industry classifies the global uranium resources into two categories: conventional and unconventional resources. Phosphates are the main constituent of unconventional uranium resources, other types of uranium-bearing resources (e.g. black shales) are insignificant on global scale.

Phosphates are irreplaceable for agricultural use, so mining of these minerals should be tailored exclusively to agricultural needs. Moreover, the thermodynamic quality of phosphates as a uranium-for-energy source lies beyond the energy cliff: no net energy generation is possible by exploitation of phosphate rock; this holds true also for other unconventional uranium resources, including uranium from seawater.

How much CO2 does nuclear power emit?

Nuclear CO2 emission originates from burning fossil fuels in all processes and factories needed to extract uranium from the ground, prepare nuclear fuel from the recovered uranium, construct the nuclear power plant and to safely manage the radioactive wastes. The fission process in the nuclear reactor is the only process of the nuclear system that has (virtually) no CO2 emission. In addition CO2 is generated by chemical reactions during the production of necessary materials and chemicals, for example cement (concrete) and steel. A generic NPP contains some 150 000 tonnes of steel and 850 000 tonnes of concrete, in addition to several thousands of tonnes of other materials. The sum of all materials consumed by an NPP during its operational lifetime is about 76 grams per kilowatt-hour delivered to the grid, excluding the mass of rock displaced for mining and final sequestration of the radioactive wastes.

By means of the same thermodynamic analysis that revealed the energy cliff, see above, the sum of the CO2 emissions of all processes constituting the nuclear energy system could be estimated at 88-146 gram CO2 per kilowatt-hour. This figure is based on the assumption that all electric inputs of the nuclear process chain are provided by the nuclear power plant itself, to avoid discussions of the local fuel mix of electricity generation.

The large uncertainty range is chiefly caused by uncertainties regarding the processes of the back end of the process chain, these are the processes needed to safely isolate the inevitable radioactive wastes from the biosphere, including the dismantling of the NPP after its service life. The emission figure will rise with time, as will be explained below.

CO2 trap

The energy consumption and consequently the CO2 emission of the recovery of uranium from the earth's crust strongly depend on the ore grade, and several other physical and chemical factors that are not discussed here. In practice the most easily recoverable and richest resources are exploited first, a common practice in mining, because these offer the highest return on investment. As a result of this practice the remaining resources have lower grades and uranium recovery becomes more energy-intensive and more CO2 intensive. Consequently the specific CO2 emission of nuclear power will rise with time; when the average ore grade approaches 200 ppm, the specific CO2 emission of the nuclear energy system will surpass that of fossil-fuelled electricity generation. This phenomenon is called the CO2 trap.

If no new major high-grade uranium resources are found in the future, nuclear power will run aground in the CO2 trap within the lifetime of new nuclear build.

Does nuclear power also emit other greenhouse gases?

No data are found in the open literature on the emission of greenhouse gases other than CO2 by the nuclear system, likely such data never have been published. Assessment of the chemical processes required to produce enriched uranium and to fabricate fuel elements for the reactor indicates that substantial emissions of fluorinated and chlorinated gases are unavoidable; some of these gases may be potent greenhouse gases, with global warming potentials thousands of times greater than CO2.

Unknown are the GHG emissions of the construction of a nuclear power plant, with its large mass of high-quality and often exotic materials. Unknown are the GHG emissions of the operation, maintenance and refurbishment of nuclear power plants. Unknown are the GHG emissions of the backend of the nuclear process chain: the handling and storage of spent fuel and other radioactive waste.

It is inconceivable that nuclear power does not emit other greenhouse gases, this matter is still a well-kept secret. Absence of published data does not mean absence of emissions.

Nuclear power stations and reprocessing plants discharge substantial amounts of a number of fission products, one of them is krypton-85, a radioactive noble gas. Krypton-85 is a beta emitter and is capable of ionizing the atmosphere, leading to the formation of ozone in the troposphere. Tropospheric ozone is a greenhouse gas, it damages plants, it causes smog and health problems. Due to the ionization of air krypton-85 affects the atmospheric electric properties, which gives rise to unforeseeable effects for weather and climate; the Earth's heat balance and precipitation patterns could be disturbed. Would nuclear power exchange alleged mitigation of CO2 emissions for enhanced emissions of climate changer krypton-85?

Are the published nuclear GHG emission figures comparable to renewables?

Scientifically sound comparison of nuclear power with renewables is not possible as long as many physical and chemical processes of the nuclear process chain are inaccessible in the open literature, and their unavoidable emissions cannot be assessed.

When the nuclear industry is speaking about its GHG emissions, only its CO2 emissions are involved.

Erroneously the nuclear industry uses the unit gCO2eq/kWh (gram CO2-equivalent per kilowatt-hour), this unit implies that other greenhouse gases also are included in the emission figures, instead the unit gCO2/kWh (gram CO2 per kilowatt-hour) should be used. The published emission figures of renewables do include all greenhouse gases. In this way the nuclear industry gives a false and misleading impression of things, comparing apples and oranges.

A second reason why the published emission figures of the nuclear industry are not scientifically comparable to those of renewables is the fact that the nuclear emission figures are based on a very incomplete analysis of the nuclear process chain, for instance the emissions of construction, operation, maintenance, refurbishment and dismantling, jointly responsible for 70% of nuclear CO2 emissions, are either not taken into account, or use unrealistically low figures. It is these exact components that are the only contributions to the published GHG emissions of renewables. Solar power and wind power do not consume materials for conversion into electricity, as nuclear power does.

What is the energy debt and what are the delayed CO2 emissions of nuclear power?

Only a minor fraction of the back end processes of the nuclear chain are operational, after more than 60 years of civil nuclear power. The fulfillment of the back end processes involve large-scale industrial activities, requiring massive amounts of energy and high-grade materials. The energy investments of the yet-to-be fulfilled activities can be reliably estimated by a physical analysis of the processes needed to safely handle the radioactive materials generated during the operational lifetime of the nuclear power plant. No advanced technology is required for these processes.

The energy investments for construction of the nuclear power plant and those for running the front end processes are offset against the electricity production during the operational lifetime. The future energy investments required to finish the back end are called the energy debt.

The CO2 emissions coupled to those processes in the future have to be added to the emissions generated during the construction and operation of the NPP if the CO2 intensity of nuclear power were to be compared to that of other energy systems; effectively this is the delayed CO2 emission of nuclear power. Whether the back end processes would emit also other greenhouse gases is unknown.

Claiming that nuclear power is a low-carbon energy system, even lower than renewables such as wind power and solar photovoltaics, seems strange in view of the fact that the CO2 debt built up during the past six decades of nuclear power is still to be paid off.


Assuming nuclear power emits no greenhouse gases (which is not true), the nuclear mitigation share would grow from the present level of less than 1% to at most 1.4% of the global greenhouse gas emissions by 2050-2060, if the global nuclear capacity were to grow according to scenarios projected by the nuclear industry.

Materialization of the nuclear capacity scenarios proposed by the nuclear industry are doubtful because of the unrealistically high construction rates of new nuclear power plants that would be required.

Nuclear generating capacity in the future will have to rely completely on reactors in the once-through mode, because closed-cycle systems, including the thorium cycle, are inherently unfeasible. As a consequence future nuclear power depends exclusively on the availability of natural uranium resources.

Net energy contribution to the global energy supply by nuclear power is limited by the availability of uranium-for-energy resources. Exploiting resources at ore grades below 0.02-0.01% uranium the nuclear system becomes an energy sink instead of an energy source: nuclear power falls off the energy cliff.

The average ore grade and other qualities of the yet-to-be exploited global uranium resources decline with time, because the highest quality resources available are always mined first.

The chances of discovering new major uranium-for-energy resources are bleak.

Mining of phosphates should be tailored exclusively to agricultural needs, for phosphorus is irreplaceable in agriculture.

Uranium from seawater is no option. If feasible at commercial scale at all, this resource lies far beyond the energy cliff: no net energy generation is possible.

From a practical viewpoint only the low IAEA scenario seems feasible, resulting in a mitigation share of 0.5-0.3% of the global GHG emissions by 2050, provided nuclear power is GHG free. The mitigation share would become negligible if the nuclear GHG emissions are taken into account.

At present nuclear power emits 88-146 gCO2/kWh. Likely the nuclear CO2 emissions will grow from the current level to values approaching fossil fuel generation within the lifetime of new nuclear builds in the scenarios of both the IAEA and WNA.

Emissions of GHGs other than CO2 by nuclear power are not reported, but are almost certain from a technical point of view.

Krypton-85, discharged by all nuclear power plants and reprocessing plants, generates greenhouse gases in the troposphere, in addition it causes other weather and climate changing effects.

The published figures of nuclear GHG emissions are not comparable to the figures of renewables, because different quantities and estimation methods are applied.

Due to the après nous le déluge culture of the nuclear industry the health hazards posed by radioactive materials in the human environment will increase with time, in addition to risks of Chernobyl-like disasters and of nuclear terrorism.