As was noted in my article on nuclear fuel demand in the January 2005 edition of NEI, enrichment services can be at least a partial substitute for uranium. However, the extent to which it is possible for buyers to save money by using more enrichment is worthy of close examination, as it is not as simple a matter as it may at first seem.
Reactor operators require enriched fuel to supply to their fuel fabricator and can choose a variable mixture of uranium and enrichment services (ie. separative work units, SWUs) to achieve this. This choice is made through deciding on the contractual tails assay to be used at the enrichment plant. Natural uranium has a U235 assay of 0.71% and the amount of SWUs applied to the feed material determines the assay of the waste stream (ie. the tails assay). If the buyer chooses a low tails assay (perhaps 0.25% in present conditions), he can supply a smaller amount of uranium as feed to the enrichment company, which will then apply more SWUs to obtain the required quantity of enriched uranium. Alternatively, in choosing a high tails assay (say 0.35%), the fuel buyer will supply a larger quantity of feed uranium, which will then receive fewer SWUs to produce an identical amount of enriched uranium.
The decision on the contractual tails assay is motivated by the relative prices of uranium and enrichment services and there is an optimum rate for each price relationship. As the price of uranium has risen (with the price of SWUs remaining relatively stable), the optimum tails assay has fallen from around 0.33% to 0.26%. In theory, a reduction of the tails assay of this magnitude could reduce uranium requirements by up to 15%, while increasing the demand for SWUs by a similar amount. There are, however, good reasons to expect the effect to be rather smaller than this, as the rather theoretical basic calculation contains a number of underlying assumptions.
The most obvious of these assumptions is that there is sufficient spare enrichment capacity available to supply the additional SWUs that will be required. For the Western enrichment companies, this is certainly doubtful. The part of current capacity represented by the old gaseous diffusion plants is gradually being closed down by both USEC and Eurodif. Urenco, which has been using centrifuges for many years, has been steadily increasing capacity to match the contracts it has received but does not maintain any spare capacity to cope with sudden fluctuations in demand. It will be some years before its incremental capacity in its new US plant becomes available.
On the other hand, it has always been assumed that the Russian enrichment plants, with rated capacity of over 20 million SWUs per annum, have a lot of spare capacity. These plants are subject to trade restrictions, limiting the extent to which they can contract with Western buyers, but, more importantly, the magnitude of their spare capacity is now also somewhat in doubt. It is possible to account for all of the 20 million SWU/y capacity if one makes realistic assumptions about the very low tails assays (down to 0.10%) that they use, their needs to fulfil the requirements of Russian-designed reactors worldwide, contracts with Western utilities, the needs for down-blending Russian HEU and also for re-enriching old tails material in stockpiles.
If capacity is tight, the obvious impact of buyers requesting lower contractual tails assays will be for the price of SWUs to rise. Once this happens, the incentive for lower tails assays will diminish – indeed, if the enrichment price rises substantially, it may be completely eliminated. Enrichment companies are unlikely to go about installing new capacity solely to cope with a reduction in the optimum tails assay, which may well be only temporary. Or if they do, they will want to receive substantially higher prices for the additional SWUs, therefore diminishing the attraction of switching.
Given that there are already substantial new investments taking place in centrifuge plants, the future enrichment price is also very hard to assess. Centrifuge enrichment uses much less energy than diffusion, but billions of dollars of new capital are required to build the plants. The new capacity (apart from that of Urenco in the USA) is replacing facilities where the investments were amortised years ago and so must earn a good return. So enrichment prices may well have to rise anyway, in addition to any effect from the pressure on capacity caused by the move towards lower tails assays. Indeed, a future scenario is conceivable where uranium prices have fallen back from current levels as additional low-cost production facilities come on-line, but enrichment prices are substantially higher owing to the need to finance the new facilities. The impact of this, of course, would be to push optimal tails assays back up again.
In theory, a reduction of the tails assay from 0.33% to 0.26% could reduce uranium requirements by 15%
In addition to the question of adequate capacity, enrichment companies have also become accustomed, in the recent past, to using a lower operating tails assay than that implied in contracts, a process known as underfeeding. This generates additional uranium for their use and works as follows. The customer supplies a quantity of uranium as feed to the enrichment company, based on the contractual tails assay (which in turn is optimised from the uranium-SWU price relationship faced by the buyer). The relevant optimal tails assay for the enrichment company may well be different, as the marginal cost of using some more SWUs is probably substantially lower than their market price. This implies that the enrichment company may have a lower optimum tails assay and should therefore use more SWUs (and less uranium) than is implied in the contract with the customer. This allows the enrichment company to hold back some of the uranium supplied as feed and then sell it on the market as either natural uranium or enriched uranium product.
The impact of past ‘underfeeding’ has been to soak up spare enrichment capacity. This makes it unlikely that there will be much spare capacity available to satisfy increased SWU demand flowing from higher uranium prices, at least in the short term.
The process can work in reverse, of course (overfeeding), where the enrichment company faces high marginal SWU production costs for some reason (maybe needing to install additional capacity just to cover one contract) and chooses to use additional uranium to that supplied by customers in order to produce the enriched uranium. This is rather unlikely in current market conditions, but demonstrates the point that enrichment plants and their customers do not necessarily face the same optimal tails assay – indeed, it is most likely that they won’t.
A final consideration concerns the recent practice of Western enrichment companies sending the tails material produced at their plants each year to Russia to be re-enriched. The economics of this are crucially dependent on the assay of the tails. If this is substantially lower than the assumed 0.30% the financial advantages will rapidly diminish, cutting off a potential source of revenue for enrichment companies. Again, the impact is to provide less incentive for enrichment companies to employ lower tails assays.
The pure arithmetic would suggest that the recent upsurge in uranium prices may cause a 15% reduction in annual uranium requirements (about 7000t/y), with a similar percentage increase in SWU requirements (about 5 million SWU/y). What, however, will be the likely impact in the real world? The factors already mentioned suggest that it will be substantially lower than this, perhaps of the order of a reduction of uranium demand of 3000t/y and an increase in SWU demand of 2 million SWUs.
There are definitely indications from world utilities that tails assays are beginning to fall (towards 0.25% in some cases) with the sharply higher uranium price, but it will also take time for the higher prices to feed through into current contracts. In 2004, around 40% of uranium requirements were still met by secondary supplies of one kind or another. These are influenced by rather different economic factors to those of primary supply sources. Yet once primary uranium production begins again to take a higher share of reactor requirements, the impact of varying tails assays should once again become much more transparent.
Steve Kidd is Head of Strategy & Research at the World Nuclear Association, where he has worked since 1995 (when it was the Uranium Institute). Any views expressed are not necessarily those of the World Nuclear Association and/or its members.