Supply excess for a decade8 October 2014
A year after the end of the ‘megatons to megawatts’ programme, URENCO and Rosatom dominate a market characterized by excess supply; China could be a major player in the future. By Thomas Meade and Eileen Supko
All of the front-end nuclear fuel markets are in state of over-supply, and the enrichment services market is no exception. When it was characterized by significant capacity utilizing gaseous diffusion technology, some supply flexibility existed, as the large variable costs for electric power allowed gaseous diffusion plants to adjust output in response to demand. Now that 100% of primary enrichment supply uses centrifuge technology, that flexibility has been lost. The modular nature of centrifuge technology allows it to be efficient when supply capacity growth is called for, but the technology is relatively inflexible when capacity exceeds demand. Centrifuges operate at full capacity until they are permanently retired, so output from existing supply is fixed regardless of market demand. Since the technology has high capital costs, but low cash operating costs, competitive prices in a market characterized by over- supply are much lower than prices needed to support the capital investments required in a market where supply growth is needed. Over the past few years, the enrichment market has devolved from a market characterized by new supply (both for GDP replacement and for anticipated demand growth) to a market where plans for expansion or new plants have been shelved and suppliers compete for the scraps available from highly-committed customers with stagnant demand. The ability to underfeed and refeed existing tails material does provide a relief valve of sorts for excess enrichment supply, but to the detriment of the uranium and conversion markets. The resulting downward pressure on uranium prices limits the financial benefits of redirecting enrichment supply to uranium production as a result. Furthermore, the current excess in enrichment supply cannot be fully absorbed by redirection to uranium production.
Picture: Urenco USA facility in New Mexico
Nuclear power plant requirements increased 2% to 41.6 million SWU in 2013, but remained depressed due to the prolonged outages at Japanese nuclear power plants. A strong increase is projected to finally occur in 2016 as enrichment services jump 17% to 51 million SWU. Requirements are forecast to average 53 million SWU per year between 2016 and 2020, rising to 60 million SWU per year between 2021 and 2025, 67 million SWU per year between 2026 and 2030 and 72 million SWU per year between 2031 and 2035. The average annual growth rate is 2.7% per year. Led by China, growth is strong in East Asia, but also takes place in the Other region and the Commonwealth of Independent States (CIS)/ Eastern Europe. The ERI requirements outlook is more conservative than the World Nuclear Association, particularly over the next two years.
Primary enrichment services supply is essentially provided by three active primary suppliers - Urenco, Rosatom and AREVA. USEC Inc. no longer has primary capacity and utilizes supply from inventory and bought from Russia.
Urenco's supply capability includes three European enrichment plant sites, which are located in Gronau, Germany, Almelo, Netherlands and Capenhurst, United Kingdom. These plants had a combined annual production capability of 14.4 million SWU at the end of 2013. No further expansion is planned at present. The Urenco USA enrichment facility in Lea County, New Mexico began operating in 2010. At the end of 2013, it had an annual enrichment capacity of 3.2 million SWU. Urenco continues to add enrichment cascades and expansion to 4.7 million SWU by the end of 2015 and 5.7 million SWU by 2022 has been approved by Urenco's board. The decision to slow introduction of the last 1.0 million SWU at Urenco USA was in response to deteriorating market conditions.
Rosatom's utilization of its substantial enrichment capacity and how it interacts in the world market is more complex than is the case for any of the other enrichers. Production during 2013 is estimated to have been 26 million SWU. Rosatom production is reduced approximately 7% from nameplate capacity due to the low operating tails assay employed. Rosatom exports include direct and indirect sales to Western customers and were 5.6 million SWU during 2013 and are expected to increase in 2014 and thereafter. Since Russian enrichment capacity is constrained due to US and European trade policies, a substantial amount will be used to underfeed in meeting its commercial enrichment contracts and to re-enrich depleted uranium tails. An operating tails assay of 0.13 percent by weight (w/o) U-235 tails is assumed.
AREVA's new plant, Georges Besse II (GB II), became operational in 2011, reached 5.5 million SWU in 2013, and is continuing to ramp up to its nameplate capacity of 7.5 million SWU per year by 2016. AREVA indefinitely suspended the US Eagle Rock Enrichment facility, and has stated it has no plans to extend the GB II capacity now.
Any new supply added by AREVA and Urenco would use ETC centrifuge technology. ETC is in the midst of a restructuring that is severely reducing its centrifuge manufacturing capacity. ETC plans to retain the skills and expertise necessary to ramp manufacture back up again in the future if appropriate based on customer demand, but that could take many years.
China is now thought to have 2.5 million SWU of annual centrifuge enrichment capability through China Nuclear Energy Industry Corporation (CNEIC). The majority of this capability is used internally, although China exports modest amounts to the US and Europe. The current Chinese enrichment capability primarily uses centrifuges that are imported from Russia (1.5 million SWU). A plant using Chinese centrifuge technology started pilot operations in 2010, and is believed to have a capacity of 1 million SWU as of the end of 2013, although there is uncertainty regarding the plant's exact capacity. Total Chinese centrifuge enrichment capacity is expected to expand dramatically as China ensures its ability to fill its own requirements with domestic supply.
Other capability is primarily in Japan, where JNFL put into operation the second "half" cascade of a next-generation centrifuge in 2013. JNFL capacity is expected to slowly rise to 1.5 million SWU by 2026. Brazil has started operation of a small uranium enrichment facility at Resende, with a capacity of about 18,000 SWU at the end of 2013. Resende is expected to slowly ramp capacity up to 200,000 SWU by 2025 and will be devoted to internal requirements. Despite international efforts, Iran continues to install centrifuges and could have 0.1 million SWU of capacity in operation.
The Russian HEU-derived LEU supplied 5.5 million SWU in 2013, the last year of the US-Russian megatons-to-megawatts agreement for down-blending HEU. The SWU figure is based on the contractually-agreed tails assay of 0.30 w/o U-235. However, it was equivalent to 6.0 million SWU when evaluated at a 0.24 w/o U-235 tails assay. An additional small quantity of SWU is derived from Russian HEU directly blended with European RepU. The programme provided an estimated 0.5 million SWU per year and will decline in the future, as supplies of non-weapons-grade HEU are limited.
Total releases to the commercial enrichment market from the down-blending of surplus US HEU are expected to decline over the next six years. No plans have been announced for the small amount of unallocated HEU that may exist after 2022. HEU down-blending capability is unlikely to be maintained for the longer term. Recycle materials contributed about 1.2 million SWU-equivalent to supply in 2013 and are expected to supply the equivalent of 1.6 million SWU by 2020 and later. This includes both commercial MOX fuel and MOX utilizing government-excess weapons-grade plutonium.
Early closure of reactors has left some utilities with excess inventory of EUP or excess enrichment services which will be marketed. LEU previously created from DOE tails in 2012 and 2013 will be loaded in TVA and Energy Northwest reactors over the next ten years. In addition, the reactor outages in Japan have caused a buildup of EUP inventory, as some enrichment deliveries are still taking place under fixed quantity contracts.
USEC hopes to construct the 3.8 million SWU per year American Centrifuge Plant (ACP). While USEC has received a license from the NRC to build and operate the ACP, it has been unable to secure the $4 billion in financing needed to complete construction. USEC has stated that based on the current market situation, commercialization of the ACP is not economically viable. In early April 2014, DOE announced that it was examining its alternatives for preservation of the centrifuge technology and was exercising its right to take over management of the ACP R&D programme that was managed by USEC. DOE's interest in the ACP technology revolves around national security concerns, the availability of US enrichment technology and preserving intellectual property.
GLE received a NRC construction license for a commercial plant using laser enrichment technology in 2012 and has continued to develop and test the technology for the last few years. A commercial plant would have an initial enrichment capacity of one million SWU which would then increase by one million SWU per year to a planned maximum target of six million SWU per year. In July 2014 GLE decided to scale back its activities and place any commercialization decision on hold in light of current and expected adverse market conditions.
The supply of enrichment services is significantly greater than requirements at present, as demonstrated in Figure 1, based on ERI's forecast of supply and reference nuclear power growth requirements through 2035. Only existing capacity and firmly planned new capacity are assumed and the supply shown is for all enrichment capacity, prior to any redirection for uranium production via underfeeding and refeeding of tails stockpiles.
It has long been recognized that the supplier with the largest enrichment capacity, Rosatom, devotes a significant portion of its enrichment supply to uranium production. Rosatom utilizes enrichment capacity for uranium production by operating at low tails assays (underfeeding) and by refeeding existing tails inventory. Rosatom's uranium production strategy is in part based on the desire to sell LEU, containing natural uranium equivalent and enrichment services, rather than just enrichment services. Rosatom also remains under trade restrictions in both the US and the Euratom countries.
In addition, many buyers limit the share of enrichment services sourced from Russia for supply diversity and supply security reasons. As a result, Rosatom has more enrichment supply than it can market.
Rosatom's long-term market share is projected to be limited to about 27% of world requirements outside the CIS and Eastern Europe and about 34% if the Chinese market is excluded as well. Rosatom has announced a market share goal of 33% of Western-designed reactors, which is consistent with ERI's assumption if the goal does not count Chinese reactors as Western-designed. Fig. 2 provides the base supply adequacy after redirection of Rosatom supply to uranium production. Excess supply is still apparent on a world basis, particularly over the next few years, but is dramatically reduced when the Rosatom supply directed to uranium production is removed.
The long-term supply excess will exist without any additional supply added by Western enrichers other than the completion of LES Phase 3 (to reach 5.7 million SWU) and the completion of GB II (to reach 7.5 million SWU). The redirection of enrichment to uranium production by Rosatom, Urenco and others results in the supply of 24 million pounds U3O8/yr (9240 tU) through 2022 and 14 million pounds U3O8/yr after that.
The long-term supply adequacy shown in Figure 2 includes the assumption that Urenco will replace cascades at the European sites as they retire after 25 years of operation, keeping installed capacity constant. However, during 2013 Urenco retired 0.1 million SWU at each of the European sites, for a total of 0.3 million SWU. The decision regarding whether or not to replace retiring centrifuges is of course based on economics and must be supported by market prices and contract commitments. If Urenco does not replace retiring centrifuges, the supply excess would be eliminated around the year 2027, and a small amount of new capacity would then be needed (4 million SWU by 2031 and 7 million SWU by 2035).
The long-term supply excess shown in Figure 2 is also based on the assumption that CNEIC will increase indigenous supply to keep pace with increases in Chinese requirements. This assumption is widely held by the industry but such a rapid build-out is not a foregone conclusion. If the rate of increase in CNEIC supply is reduced by 50% from the nearly 1 million SWU per year assumed in Figure 2, a need for new supply would open up around the year 2025, growing to about 6 million SWU between 2030 and 2035. If CNEIC's ability to ramp up indigenous supply is in fact limited, CNEIC might make up at least some of the difference by using Russian centrifuge technology, adding to the 1.5 million SWU already in operation. If CNEIC's technology proves successful it may also become a more significant supply source for foreign customers.
The long-term price indicator reported by TradeTech was essentially flat between mid- 2009 and mid-2011, but has declined nearly continuously over the last three years. The impact of reduced Japanese demand and resulting open supply, as well as increasing inventory availability caused the price to steadily decline to $92 per SWU by the end of July 2014. The current long-term price represents a $22 per SWU (19%) decline over the last 12 months and a $66 per SWU (42%) decline since August 2011. The spot price has demonstrated similar behaviour, as the current spot price of $90 per SWU represents a $20 per SWU (18%) decline over the last 12 months and a $65 per SWU (42%) price decline since February 2011. Enrichment market activity declined considerably in 2013, and for the fifth year in a row. ERI's estimate of new commitments (both term and spot) executed by end-users during 2013 was 15 million SWU, a 35% decline from 2012. None of the primary suppliers were able to maintain backlog, as new sales were less than deliveries during the year for all.
Each supplier's world market share is provided in Table 1. The market shares are given as a percentage of total deliveries by suppliers and therefore exclude recycle (MOX) and utility inventory use. Note that market shares may include purchases by customers to build inventory, intentionally or otherwise. The sales of Russian HEU-derived EUP under the US-Russian HEU Agreement, which accounted for 14% of the world market in 2013, are included in USEC's market share. European and US supplier market shares include 2.5% from wholesale purchases of Rosatom supply, which are then delivered to their customer base. The market shares of Urenco and Rosatom are expected to remain high as USEC continues its decline and AREVA's increases modestly as GB II reaches full capacity. CNEIC's market share will rise dramatically to keep pace with China's rapidly increasing domestic requirements. It remains to be seen if CNEIC will become a major player outside the Chinese market.
Trade restrictions continue to influence the enrichment market, but their impact is declining. In November 2013, the US Department of Commerce decided to continue the 19.95% duties on imports of French low-enriched uranium by AREVA, ruling that revocation of the antidumping order would lead to continuation or recurrence of material injury to the US industry. Rosatom's direct market share is growing, and approximately 75-80% of the 2014 to 2020 US quota under the Russian Suspension Agreement has been filled. Russia will also deliver 21.5 million SWU to USEC from 2013 through 2022 under a wholesale arrangement, but USEC may have difficulty marketing the full amount. To date, economic sanctions imposed by the US and Europe over Russia's role in the Ukraine have not impacted nuclear fuel supplies, but some customers are growing concerned. Current inventories and supply excess appear to be more than adequate to cover any short term disruption, although affected customers would certainly be inconvenienced if forced to scramble to replace Russian EUP supply.