Nuclear activities worldwide generate around 200,000 m3 of radioactive waste every year. Of this about 10,000 m3, less than 5% by volume but containing most of the radioactivity, needs deep, long-term geological storage in the form of geological disposal facilities (GDFs). These caverns purpose-built hundreds of metres underground are packed with waste containers, cement backfill, and a suitable host rock such as clay.
Cement plays several roles in these facilities, including holding waste in place, supporting tunnels and vaults, and helping slow the movement of radioactive elements if groundwater ever gets in.
Conventional Portland cement has a very high pH, above 12, which is good for strength and for trapping many radionuclides, but it also creates problems. When water is exposed to this cement it becomes very alkaline itself, and thus able to corrode steel and aluminium, produce hydrogen gas, and damage clay barriers like bentonite by weakening their ability to absorb fluids.
To avoid creating such problems in the repository, engineers have developed low-pH cements with porewater pH closer to 10–11. These mixes reduce the amount of ordinary Portland cement and add materials such as silica fume and blast furnace slag, producing calcium-silicate-hydrate gels.
Double-edged sword
One such formulation, known as the CEBAMA mix, has emerged as a strong candidate for use in European disposal concepts. It has a lower pH with a favourable mechanical performance and chemical compatibility with clays and host rocks. However, there are still important questions about how CEBAMA will evolve over many centuries — due to chemical influences as well as in the presence of microbes that live in alkaline and low oxygen environments.
Microbial activity in cements is a famously double-edged sword: microbes can corrode concrete in structures like sewers even as certain bacteria can spur a process called microbially induced carbonate precipitation (MICP). In ordinary concrete, MICP has been shown to seal micro-cracks and pores by filling them with calcium carbonate, improving their strength and durability. These studies have however mostly been conducted in aerobic conditions and with standard cement, which has higher pH.
The conditions in a GDF are markedly different: they’re anoxic, alkaline, and rich in dissolved ions. While abiotic carbonate formation is limited, alkaliphilic, anaerobic microbes are still expected to be able to colonise niches where they can find water and nutrients. The authors of a new study, from the University of Manchester, wanted to know whether such microbes could drive MICP in low-pH cement in a repository’s conditions and whether the net effect would be beneficial or harmful.
Six-month study
To these ends, the researchers set up a long period experiment. They cast small tablets of CEBAMA low-pH cement. Then they placed four tablets each into a sealed bottle containing some synthetic groundwater that mimicked Oxford clay porewater and a small quantity of sediments from the high-pH soil form Harpur Hill in the UK (which contained alkaliphilic microbes). The researchers sealed the bottles, flushed them to remove oxygen, and incubated them in the dark at 20º C.
Next, they split these bottles into three groups. Each group had a distinct carbon group. In the high-carbon group, each bottle also contained lactate ions (C3H5O3—) as an organic carbon source while nitrogen filled the headspace. In the low-carbon group, each bottle contained a small amount of yeast extract and had a hydrogen headspace, which represented the hydrogen that could be produced when steel is corroded. And in the no-carbon group, each bottle had no added organic carbon and a nitrogen headspace.
For the main experiments, the team added some nitrate ions to all bottles.
Over six months, the researchers periodically sampled the liquid in each bottle. They also removed one cement tablet from each bottle for analysis using spectroscopy and microscopy. And they sequenced genes in the slurry to follow changes in the microbial community.
Metabolising carbon
Based on these studies, the researchers found that in low-pH cements, MICP doesn’t happen automatically, instead it depends strongly on the availability of organic carbon and a suitable electron acceptor (like nitrate ions). When these conditions are met, the microbes produce carbonates that ‘heal’ cracks and close pores over many months. But if carbon is scarce, the cement tends to leach calcium and, to a lower extent, magnesium ions into the water while the rate of MICP remains low.
“Although the bulk groundwater entering a repository is expected to be oligotrophic, localised zones with elevated concentrations of organic carbon are considered likely,” Ananya Singh, the study’s first author and a scholar at the University of Manchester, told The Hindu in an email. “Multiple forms of organic nuclear waste degrade over time, releasing dissolved organic carbon in heterogeneous ‘pockets’ within the engineered barrier system, including cement, thereby forming organic-rich niches that can support microbial growth.”
“Nitrate (a by-product of nitric acid used in reprocessing of spent nuclear fuel) was used in our experiments as a model electron acceptor, but it is not expected to be the sole oxidant available in a GDF. Additional electron acceptors such as sulphate, ferric iron, or even trace oxygen during early post-closure phases may also be present,” Dr. Singh added. “Thus, the overall supply of both electron donors and electron acceptors is not strictly limited.”
Experts who design GDFs have been apprehensive before that the microbes in a facility might simply attack the cement and weaken barriers, including in newer low-pH formulations. But the new study has revealed the opposite: in high-carbon niches, the microbes’ metabolism leads to carbonate deposits that thicken the outer rim of the cement and seal cracks.
The findings were published in ACS Omega on November 19.
Cement integrity
The self-healing property comes with tradeoffs, however. In a sealed repository, when MICP clogs pores and seals cracks, gases such as hydrogen and methane — produced by corroding metal and organic degradation — could build up and affect the mechanical stability along alternative routes. The researchers acknowledged that future models will have to examine this balance between being watertight and constraining gas flow.
“Our … experiments indicate that in low-pH cement, microbial activity is more likely to produce a relatively thin, surface-confined carbonate zone, a few hundreds of micrometres thick, which provides a protective layer and improves local sealing rather than completely blocking the internal pore network,” Dr. Singh explained. “Therefore, the results suggest that during the early stages, when the cement is still structurally coherent, microbially mediated processes don’t compromise cement integrity.”
But over the timescales relevant for a GDF, i.e. lakhs of years, “the cement is not going to remain intact indefinitely; it will gradually alter and contribute to an alkaline plume that maintains high-pH conditions in the near field… This alteration will create cracks beyond the capacity of either abiotic or biotic healing. The good news is that these cracks provide plenty of escape routes for any gases generated.”
“To fully assess the balance between self-healing benefits and potential impacts on gas transport in the early stages, further work is needed. This will require targeted gas-flow experiments and coupled reactive-transport modelling to extrapolate our centimetre- and month-scale observations to repository-scale behaviour over centuries, in a way that can be meaningful for safety case assessments.”
Not simply ‘good’ or ‘bad’
The study also highlighted that hydrogen alone, even though it’s an electron donor, may not suffice to drive strong microbial activity and MICP in the alkaline cements over practical timescales, going by the conditions used in the study. The microbes that responded most vigorously in these conditions were heterotrophic nitrate reducers fed by organic carbon rather than hydrogenotrophs. This finding imposes limits on which microbial processes are likely to matter more for the cement’s evolution in a GDF.
Finally, the results are a reminder that microbes in extreme environments aren’t automatically ‘good’ or ‘bad’ for engineered systems. Alkaliphilic communities in a pH 10-11 could be like repair crews when given the right resources. But these findings were made in controlled laboratory conditions and over only six months. In a real repository, groundwater flow can replenish substrates and processes unfold over decades to centuries, and the net effects of microbial activity on cement performance will depend on how these mechanisms scale up.
That is to say the new study doesn’t close the chapter on microbes in low-pH cements but instead points to a more complex and potentially useful role. It suggests any future safety assessments and engineering designs for radioactive waste disposal should treat microbial metabolism as active variables to be understood and possibly harnessed rather than as background noise to be ignored.
