Uranium occurs naturally in many rocks, and even in seawater. However, like other metals, it is seldom sufficiently concentrated to be economically recoverable. There are around 40 trillion tons of uranium in Earth's crust, but most is distributed at trace concentration over its mass. Estimates of the amount concentrated into ores affordable to extract for under $130 per kg can be less than a millionth of that total. Due to the currently low price of uranium, the majority of commercial
light water reactors operate on a "once through fuel cycle" which leaves virtually all the energy contained in the original , which makes up over 99% of natural uranium, unused.
Nuclear reprocessing can recover part of that energy by producing
MOX fuel or
Remix Fuel for use in conventional power generating light water reactors. This technology is currently used at industrial scale in France, Russia and Japan. However, at current uranium prices, this is widely deemed uneconomical if only the "input" side is considered.
Breeder reactor technology could allow the current reserves of uranium to provide power for humanity for billions of years, thus making
nuclear power a sustainable energy.
Reserves Reserves are the most readily available resources. About 96% of the global uranium reserves are found in these ten countries: Australia, Canada, Kazakhstan, South Africa, Brazil, Namibia, Uzbekistan, the United States, Niger, and Russia. The known uranium resources represent a higher level of assured resources than is normal for most minerals. Further exploration and higher prices will certainly, on the basis of present geological knowledge, yield further resources as present ones are used up. There was very little uranium exploration between 1985 and 2005, so the significant increase in exploration effort that we are now seeing could readily double the known economic resources. On the basis of analogies with other metal minerals, a doubling of price from price levels in 2007 could be expected to create about a tenfold increase in measured resources, over time.
Known conventional resources Known conventional resources are resources that are known to exist and easy to mine. In 2011, this increased to 7 million tonnes. Exploration for uranium has increased: from 1981 to 2007, annual exploration expenditures grew modestly, from US$4 million to US$7 million. This increased to US$11 million in 2011. The OECD Redbook cites areas still open to exploration throughout the world. Many countries are conducting complete aeromagnetic gradiometer radiometric surveys to get an estimate the size of their undiscovered mineral resources. Combined with a gamma-ray survey, these methods can locate undiscovered uranium and thorium deposits. The U.S. Department of Energy conducted the first and only national uranium assessment in 1980 – the National Uranium Resource Evaluation (NURE) program.
Secondary resources Secondary uranium resources are recovered from other sources such as nuclear weapons, inventories, reprocessing and re-enrichment. Since secondary resources have exceedingly low discovery costs and very low production costs, they have displaced a significant portion of primary production. In 2017, about 7% of uranium demand was met from secondary resources. But as the supply of former weapons uranium has been used up, mining has increased, so that in 2012, mining provided 95 percent of reactor requirements, and the OCED Nuclear Energy Agency and the International Atomic Energy Agency projected that the gap in supply would be completely erased in 2013.
Inventories Inventories are kept by a variety of organizations – government, commercial and others. The US
DOE keeps inventories for security of supply to cover for emergencies where uranium is not available at any price.
Decommissioning nuclear weapons Both the US and Russia have committed to recycle their nuclear weapons into fuel for electricity production. This program is known as the
Megatons to Megawatts Program. Down blending of Russian weapons high enriched uranium (HEU) will result in about of low enriched uranium (LEU) over 20 years. This is equivalent to about of natural U, or just over twice annual world demand. Since 2000, of military HEU is displacing about of uranium oxide mine production per year which represents some 13% of world reactor requirements. The Megatons to Megawatts program came to an end in 2013. The U.S. also has commitments to dispose of of non-waste HEU.
Reprocessing and recycling Nuclear reprocessing (or recycling) can increase the supply of uranium by separating the uranium from
spent nuclear fuel. Spent nuclear fuel is primarily composed of uranium, with a typical concentration of around 96% by mass. The composition of reprocessed uranium depends on the time the fuel has been in the reactor, but it is mostly
uranium-238, with about 1%
uranium-235, 1%
uranium-236 and smaller amounts of other isotopes including
uranium-232. Currently, there are eleven reprocessing plants in the world. Of these, two are large-scale commercially operated plants for the reprocessing of spent fuel elements from light water reactors with throughputs of more than of uranium per year. These are La Hague, France with a capacity of per year and
Sellafield, England at uranium per year. The rest are small experimental plants. The two large-scale commercial reprocessing plants together can reprocess 2,800 tonnes of uranium waste annually. The United States had reprocessing plants in the past but banned reprocessing in the late 1970s due to the high costs and the risk of
nuclear proliferation via plutonium. The main problems with uranium reprocessing are the cost of mined uranium compared to the cost of reprocessing, At present, reprocessing and the use of plutonium as reactor fuel is far more expensive than using uranium fuel and disposing of the spent fuel directly – even if the fuel is only reprocessed once. Reprocessing is most useful as part of a
nuclear fuel cycle using
fast-neutron reactors since
reprocessed uranium and
reactor-grade plutonium both have isotopic compositions not optimal for use in today's
thermal-neutron reactors.
Unconventional resources Unconventional resources are occurrences that require novel technologies for their exploitation and/or use. Often unconventional resources occur in low-concentration. The exploitation of unconventional uranium requires additional research and development efforts for which there is no imminent economic need, given the large conventional resource base and the option of
reprocessing spent fuel. Phosphates, seawater, uraniferous coal ash, and some type of
oil shales are examples of unconventional uranium resources.
Phosphates Uranium occurs at concentrations of 50 to 200 parts per million (ppm) in phosphate-laden earth or
phosphate rock. As uranium prices increase, there has been interest in extraction of uranium from phosphate rock, which is normally used as the basis of phosphate fertilizers. There are 22 million tons of uranium in phosphate deposits. Recovery of uranium from phosphates is a
mature technology; Historical operating costs for the uranium recovery from phosphoric acid range from $48–$119/kg U3O8. In 2011, the average price paid for U3O8 in the United States was $122.66/kg. Worldwide, approximately 400 wet-process
phosphoric acid plants were in operation. Assuming an average recoverable content of 100 ppm of uranium, and that uranium prices do not increase so that the main use of the phosphates are for
fertilizers, this scenario would result in a maximum theoretical annual output of U3O8.
Seawater Unconventional uranium resources include up to of uranium contained in sea water. Several technologies to extract uranium from sea water have been demonstrated at the laboratory scale. According to the OECD, uranium may be extracted from seawater for about US$300/kgU. In 2025, researchers introduced a technique that uses electrodeposition on a
covalent organic framework (COF), a porous crystalline
polymer PEDOT that infused the pores of an
amidoxime-functionalised, fully π-conjugated sp2c-COF-A. The electrode can be repeatedly regenerated and reused.
Uraniferous coal ash ,
uranium and
thorium radioisotopes naturally found in coal and concentrated in heavy/bottom
coal ash and airborne
fly ash. As predicted by
ORNL to cumulatively amount to 2.9 million tons over the 1937–2040 period, from the combustion of an estimated 637 billion tons of coal worldwide. According to a study by
Oak Ridge National Laboratory, the theoretical maximum energy potential (when used in
breeder reactors) of trace uranium and thorium in coal actually exceeds the energy released by burning the coal itself. An international consortium has set out to explore the commercial extraction of uranium from uraniferous coal ash from coal power stations located in Yunnan province, China. The three coal power stations at Xiaolongtang, Dalongtang and Kaiyuan have piled up their waste ash. Initial tests from the Xiaolongtang ash pile indicate that the material contains (160–180 parts per million uranium), suggesting a total of U3O8 could be recovered from that ash pile alone.
Breeding A breeder reactor produces more nuclear fuel than it consumes and thus can extend the uranium supply. It typically turns the dominant isotope in natural uranium, uranium-238, into fissile plutonium-239. This results in a hundredfold increase in the amount of energy to be produced per mass unit of uranium, because uranium-238, which comprises 99.3% of natural uranium, is not used in conventional reactors, which instead use uranium-235 (comprising 0.7% of natural uranium). In 1983, physicist
Bernard Cohen proposed that the world supply of uranium is effectively inexhaustible, and could therefore be considered a form of
renewable energy. He claims that
fast breeder reactors, fueled by naturally-replenished uranium-238 extracted from seawater, could supply energy at least as long as the sun's expected remaining lifespan of five billion years. A few commercial breeder reactors exist. In 2016, the Russian
BN-800 fast-neutron breeder reactor started producing commercially at full power (800 MWe), joining the previous
BN-600. , the Chinese
CFR-600 is under construction after the success of the
China Experimental Fast Reactor, based on the BN-800. These reactors are currently generating mostly electricity rather than new fuel because the abundance and low price of mined and reprocessed uranium oxide makes breeding uneconomical, but they can switch to breed new fuel and
close the cycle as needed. The
CANDU reactor, which was designed to be fueled with natural uranium, is capable of using
spent fuel from Light Water Reactors as fuel, since it contains more
fissile material than natural uranium. Research into "DUPIC" – direct use of PWR spent fuel in CANDU type reactors – is ongoing and could increase the usability of fuel without the need for reprocessing.
Fast breeder A fast breeder, in addition to consuming uranium-235, converts
fertile uranium-238 into
plutonium-239, a
fissile fuel. Fast breeder reactors are more expensive to build and operate, including the reprocessing, and could only be justified economically if uranium prices were to rise to pre-1980 values in real terms. In addition to considerably extending the exploitable fuel supply, these reactors have an advantage in that they produce less long-lived
transuranic wastes, and can consume nuclear waste from current
light water reactors, generating energy in the process. Uranium turned out to be far more plentiful than anticipated, and the price of uranium declined rapidly (with an upward blip in the 1970s). This is why the United States halted their use in 1977, and the UK abandoned the idea in 1994. Significant technical and materials problems were encountered with FBRs, and geological exploration showed that scarcity of uranium was not going to be a concern for some time. By the 1980s, due to both factors, it was clear that FBRs would not be commercially competitive with existing light water reactors. The economics of FBRs still depend on the value of the plutonium fuel which is bred, relative to the cost of fresh uranium. At higher uranium prices
breeder reactors may be economically justified. Many nations have ongoing breeder research programs. China, India, and Japan plan large scale use of breeder reactors during the coming decades. 300 reactor-years experience has been gained in operating them.
Thermal breeder Fissile uranium can be produced from
thorium in thermal breeder reactors. Thorium is three times more plentiful than uranium. Thorium-232 is in itself not fissile, but it can be made into fissile
uranium-233 in a breeder reactor. In turn, the uranium-233 can be fissioned, with the advantage that smaller amounts of
transuranics are produced by
neutron capture, compared to
uranium-235 and especially compared to
plutonium-239. Despite the
thorium fuel cycle having a number of attractive features, development on a large scale can run into difficulties, mainly due to the complexity of fuel separation and reprocessing. Advocates for liquid core and
molten salt reactors such as
LFTR claim that these technologies negate the above-mentioned thorium's disadvantages present in solid-fueled reactors. The first successful commercial reactor at the
Indian Point Energy Center in
Buchanan, New York, (Indian Point Unit 1) ran on thorium. The first core did not live up to expectations. == Production ==