Related Pages
(September 2008)
All mineral commodity markets tend to be cyclical, ie, prices rise and fall substantially over the years, but with these fluctuations superimposed on long-term decline in real prices. In the uranium market, very high prices in the late 1970s gave way to very low prices in the early 1990s, the spot prices being below the cost of production for most mines. In 1996 spot prices recovered to the point where many mines could produce profitably, though they then declined again and only started to recover strongly late in 2003.
However, "Spot prices" apply to marginal trading from day to day and usually represent less than 20% of supply. Most trade is 3-7 year term contracts with producers selling direct to utilities. The contacted price in these is often related to the spot price.
The reasons for fluctuation in mineral prices relate to demand, and perceptions of scarcity. The price cannot indefinitely stay below the cost of production (see below), nor will it remain at very high levels for longer than it takes for new producers to enter the market and anxiety about supply to subside.
Note that the Euratom long-term price is the average price of uranium delivered into the EU that year under long term contracts. It is not the price at which long-term contracts are being written in that year.
Demand
About 435 reactors with combined capacity of over 370 GWe, require 77,000 tonnes of uranium oxide concentrate containing 65,500 tonnes of uranium (tU) from mines (or the equivalent from stockpiles or secondary sources) each year. The capacity is growing slowly, and at the same time the reactors are being run more productively, with higher capacity factors, and reactor power levels. However, these factors increasing fuel demand are offset by a trend for increased efficiencies, so demand is dampened - over the 20 years from 1970 there was a 25% reduction in uranium demand per kWh output in Europe due to such improvements, which continue.
Each GWe of increased capacity will require about 195 tU/yr of extra mine production routinely, and three times this for the first fuel load.
Fuel burnup is measured in MW days per tonne U, and many utilities are increasing the initial enrichment of their fuel (eg from 3.3 to more than 4.0% U-235) and then burning it longer or harder to leave only 0.5% U-235 in it.
The graph from Sweden's Oskarsamn-3 reactor shows that with increasing fuel burn-up from 35,000 to 55,000 MWd/t a constant amount of uranium is required per unit of electrical output, and energy used (indicated by SWU) for increased levels of enrichment increases slightly. However, the amount of fabricated fuel used in the reactor drops significantly due to its higher enrichment and burn-up.
In the USA, Exelon has also pursued higher enrichment and burnup, but in addition has reduced the tails assay from enrichment so that significantly less natural uranium feed is required. However, more energy input to enrichment is then needed. There is a clear trade-off between enrichment energy and uranium input.
Because of the cost structure of nuclear power generation, with high capital and low fuel costs, the demand for uranium fuel is much more predictable than with probably any other mineral commodity. Once reactors are built, it is very cost-effective to keep them running at high capacity and for utilities to make any adjustments to load trends by cutting back on fossil fuel use. Demand forecasts for uranium thus depend largely on installed and operable capacity, regardless of economic fluctuations. For instance, when South Korea's overall energy use decreased in 1997, nuclear energy output actually rose, to replace imported fossil fuels.
Looking ten years ahead, the market is expected to grow significantly. The WNA reference scenario shows a 33% increase in uranium demand over 2010-20 (for a 27% increase in reactor capacity - many new cores will be required). Demand thereafter will depend on new plant being built and the rate at which older plant is retired - the reference scenario has a 16% increase in uranium demand for the decade to 2030. Licensing of plant lifetime extensions and the economic attractiveness of continued operation of older reactors are critical factors in the medium-term uranium market. However, with electricity demand by 2030 expected (by the OECD's International Energy Agency, 2008) to double from that of 2004, there is plenty of scope for growth in nuclear capacity in a greenhouse-conscious world.
Supply
Mines in 2008 supplied some 51,600 tonnes of uranium oxide concentrate (U3O8) containing 43,760 tU, about 68% of utilities' annual requirements. (See also paper World Uranium Mining). The balance is made up from secondary sources including stockpiled uranium held by utilities, but those civil stockpiles are now largely depleted.
The perception of imminent scarcity drove the "spot price" for uncontracted sales to over US$ 100 per pound U3O8 in 2007 but it has settled back to $40-55 over the twelve months to September 2009. Most uranium however is supplied under long term contracts and the prices in new contracts has in the past reflected a premium above the spot market.
Note that at the prices which utilities are likely to be paying for current delivery, only one quarter of the cost of the fuel loaded into a nuclear reactor is the actual ex-mine (or other) supply. The balance is mostly the cost of enrichment and fuel fabrication.
The above graph, from International Nuclear Inc. as of end of 2007, shows a cost curve for world uranium producers, and suggests that for 45,000 tU/yr production from mines (approximately the present level) and up to 60,000 tU/yr, US$25-28/lb plus profit margin is a plausible price.
Supply from elsewhere
As well as existing and likely new mines, nuclear fuel supply may be from secondary sources including:
Major commercial reprocessing plants are operating in France and UK, with capacity of over 4000 tonnes of used fuel per year. The product from these re-enters the fuel cycle and is fabricated into fresh mixed oxide (MOX) fuel elements. About 200 tonnes of MOX is used each year, equivalent to less than 2000 tonnes of U3O8 from mines.
Military uranium for weapons is enriched to much higher levels than that for the civil fuel cycle. Weapons-grade is about 97% U-235, and this can be diluted about 25:1 with depleted uranium (or 30:1 with enriched depleted uranium) to reduce it to about 4%, suitable for use in a reactor. From 1999 to 2013 the dilution of 30 tonnes such material is displacing about 10,600 tonnes per year of mine production. (see also paper on Military Warheads as a source of Nuclear Fuel).
The following graph gives an historical perspective, showing how early production went first into military inventories and then, in the early 1980s, into civil stockpiles. It is this early production which has made up the shortfall in supply from mines since the mid 1980s.
As a result of the 1994 "Megatons to Megawatts" agreement between USA and Russia, Russia now owns a considerable amount of natural uranium which corresponds with the diluted high-enriched uranium it has supplied as described above since January 1997. In 1999 an agreement was signed which restrains this material from entering the market in the short term. Some other supply from Russian and other CIS stockpiles is possible in the short term.
The USA and Russia have agreed to dispose of 34 tonnes each of military plutonium by 2014. Most of it is likely to be used as feed for MOX plants, to make about 1500 tonnes of MOX fuel which will progressively be burned in civil reactors.
The following graph (WNA 2009 World reference scenario) suggests how these various sources of supply might look in the decades ahead:
Table of the World's Nuclear Power Reactors and Uranium Requirements
Sources:WNA 2009 Market Report (also earlier reports).IEA 2008 World Energy Outlook.