Cementing change in nuclear waste
Adrian Covill an EngD student with AMEC Nuclear UK Ltd, based in Warrington, UK, describes the development of cement mixtures to safely contain nuclear waste.
Current UK energy policy is to overhaul Britain’s ageing energy infrastructure and lower the UK’s carbon footprint. This follows the 2005 decision by government to create the Nuclear Decommissioning Authority (NDA) as the independent body responsible for the £70bln programme to clean up complex facilities such as Sellafield, UK, and decommission obsolete Magnox reactor stations. Older stations will be replaced with new pressurised water reactor (PWR) stations which are more efficient and provide greater power outputs.
The successful delivery of these programmes is highly dependant on a skilled workforce. The UK nuclear workforce is ageing and stripping the nuclear industry of highly skilled personnel, so new scientists are needed to take over these roles. The transfer of knowledge is crucial to successfully deliver both decommissioning and new build projects.
When seeking individuals to develop cost effective solutions to nuclear waste management problems, training such as an Engineering Doctorate (EngD) degree in Nuclear Engineering is preferred. The EngD is an industrial qualification requiring the majority of research to be carried out at the sponsoring company’s premises, investigating issues of relevance to the nuclear industry.
Much of the UK’s intermediate level waste (ILW) will eventually be placed in a geological disposal facility (GDF). Most ILW is encapsulated using conventional cementing systems based on partial replacements of ordinary Portland cement with either pulverised fly ash (PFA) or blast furnace slag. The encapsulated nuclear waste is packaged in stainless steel containers that comply with the requirements for disposal defined by the NDA’s Radioactive Waste Management Directorate (NDA RWMD). This is the body overseeing the GDF and it assesses waste to ensure it will be compatible with geological disposal.
Some nuclear waste is not compatible with conventional cements for example, wastes containing metals such as aluminium and uranium. The problem comes from the highly alkaline environment of the cement hydroxyl ions and the high free water content of conventional systems. When aluminium or uranium are encapsulated using these systems, excessive corrosion occurs, leading to volumetric changes and generation of significant amounts of hydrogen gas. Alternative matrices are required that are capable of encapsulating those metals, while also satisfying the NDA RWMD requirements.
A novel encapsulant for problematic ILW is magnesium phosphate cement. One aim of the EngD qualification is to build confidence that this material would meet the requirements for disposal.
Magnesium phosphate cement is a form of acid-based cement that is produced by reacting basic magnesia with an acid-phosphate. In this project a potassium di-hydrogen phosphate was used as the acid-phosphate.
Theoretically, one molar weight of magnesia to one molar weight of potassium di-hydrogen phosphate to five molar weights of water is required to produce cement paste saturation, with magnesium potassium phosphate hexahydrate being the principal reaction product. This theoretical formulation has a quick set of around two hours and has a high associated peak exotherm. This fast setting means magnesium phosphate cements have been used over the last 40 years for rapid repair of roads and runways. Although this cement has been used industrially for many years, it has not been fully considered in the UK for radioactive waste encapsulation.
Rapid cement set is not desirable in a waste encapsulation plant, so modifications have been made to increase the setting time. This has been achieved by heat treating the magnesia to reduce the overall surface area for reaction and adding boric acid, which is effective as a chemical retardant.
The high exotherm has been controlled by adding a significant amount of PFA into the system, which acts as a filler and dilutes the reagents. Using PFA is also advantageous because magnesium phosphate cement is relatively expensive to produce, typically costing £400 per 50 litres. Adding PFA helps reduce the cost. When producing conventional systems, a relatively wide envelope of water content ratios can be used. In contrast, magnesium phosphate cement is more sensitive to small changes in water content. Testing has shown that significant changes in the mechanical properties occur due to subtle changes in water content. This value needs to be carefully controlled for a reliable product. Two formulations have been investigated where all variables, apart from the water-to-solid ratios (w/s), were constant. The first formulation investigated has a ratio of 0.26w/s and the second 0.28w/s.
The right mix
Workability studies have shown that formulation 0.28w/s is more fluid than 0.26w/s, for up to two and a half hours after the initial mix due to the increased water content. When formulating the two cements, the target value for fluidity is at least 400mm displacement down a horizontal channel at two and a half hours (standard method to quantify cement fluidity) is achieved. Both meet this requirement, even at water contents close to the theoretical confines required for paste saturation.
The workability results suggest flow characteristics of the two formulations pose no issues for a cement packaging plant and should enable infiltration of any nuclear waste items.
The pH of the wet pastes has been measured up to two and a half hours after the initial mix and is to be near neutral. This was expected as the cement sets via an acid-base reaction. The pH is well below that of conventional systems (≈12), resulting in a more favourable environment for encapsulating amphoteric metals such as aluminium.
Compressive strength measurements have shown that, overall, the formulations with 0.26w/s is stronger than 0.28w/s for up to 360 days. The NDA RWMD guideline for cement strength is that at least seven MPa is attained by 90 days. Both met this requirement.
It is also important to understand the cement expansion over time as the waste form is contained within a stainless steel container. Significant expansion could cause issues with strain placed on the container. Cement expansion measurements taken up to 360 days show a plateau is reached for both formulations below the NDA RWMD limit, 2,000µstrain at 90 days.
The main reaction product of the cement acid-base reaction is magnesium potassium phosphate hexahydrate or a potassium analogue of the mineral struvite. Two other compounds have been identified by X-ray diffraction, quartz deriving from the PFA loading and unreacted magnesia. Results show that the systems are relatively phase pure and show phase stability up to 360 days.
The microstructures of both formulations have been shown to be dense and exhibit little porosity up to 360 days. This has been confirmed by mercury intrusion porosimetry. Scanning electron micrographs show that the microstructure is dominated by platy crystals of magnesium potassium phosphate hexahydrate. It is believed that the strong network of interlocking crystals contributes significantly to the strength.
These results confirm compliance of the material to NDA RWMD guidelines for strength and expansion. Ongoing investigations will involve encapsulating uranium samples using both formulations. The aim is to understand if any corrosion occurs and, if so, what the products are and the effect on the chemistry of the cement. In addition, the amount of hydrogen gas escaping from the samples is being measured to show the rate of corrosion.
Magnesium phosphate cement as an encapsulant of ILW may eventually provide a workable solution to an existing problem.