Uranium from the sea

C&I Issue 8, 2016

Although there are some 4bn t of dissolved uranium in the world’s oceans, it is present at an extremely low concentrations – of the order of 3.3 parts per billion – in the highly stable uranyl tricarbonate complex, [UO2(CO3)3]-4, which is surrounded by other ions at much higher concentrations. Any material used to adsorb uranium must therefore have a very high affinity and selectivity for the element at the pH of around 8 of seawater.

Efforts to extract uranium from seawater began in the UK in the 1960s, led by scientists from the Atomic Energy Research Establishment (AERE) in Oxfordshire. The researchers screened hundreds of potential adsorbents for uranium and identified hydrous titanium oxide as the best candidate.1 By 1981, the Japanese had built a plant to collect uranium from seawater using this adsorbent but it was too inefficient to be of any real use: the plant extracted about 0.1g of uranium per kg of TiO2.

A more efficient adsorbent was then identified by scientists at the Nuclear Research Centre in Julich, Germany, in the form of an organic polymer. They screened over 200 polymers and found one – poly(acrylamidoxime), a copolymer of carboxylate and amidoxime monomers – that was stable and selective enough to extract uranium at pH 8.3, although it also has a high affinity for other ions, in particular vanadium, which is present at much higher concentrations.2 The amidoxime monomer acts as the ligand, binding uranyl ions and, although it is not clear why, the researchers found that the carboxylate monomer also seems to play a vital role in enabling the reaction.

In the early 2000s, researchers at the Japan Atomic Energy Research Institute, led by Noriaki Seko and Masao Tamada, put together stacks of 52,000 sheets of poly(acrylamidoxime) fibres in an attempt to capture uranium from seawater. High density polyethylene (HDPE) was used as the ‘trunk’ polymer on which the amidoxime groups were attached using a process known as radiation-induced graft polymerisation. In brief, the researchers blasted the HDPE with radiation to create free radical defects on its surface, and poly(acrylonitrile) was attached onto these reactive sites; the cyano groups were then converted to amidoxime via reaction with hydroxylamine.

The stacks, which weighed nearly 800lb, were lowered into the Pacific Ocean and, after 30 days, washed in a solution of HCl to recover the uranium. The scientists were able to extract 0.5g of uranium per kg of adsorbent, but it was a very expensive process.3 By replacing the stacks with long braids of polymer similar in appearance to seaweed, they increased the average uranium uptake to 1.5g of uranium/kg-adsorbent.4

The success of the Japanese spurred US scientists to try to do even better. In 2011, the US Department of Energy’s Office of Nuclear Energy brought together a multidisciplinary team of scientists from three national laboratories, seven universities and several research institutes to tackle the problem.

Using amidoxime as the adsorbent, two US laboratories – the Oak Ridge National Laboratory (ORNL) and Pacific Northwest National Laboratory (PNNL), led by Sheng Dai at ORNL – decided to modify the shape of the HDPE scaffold in an attempt to increase its surface area and thus the amount of uranium that could be adsorbed. Within a year, they announced they had increased the surface area of polyethylene fibres by a factor of 10 by making them hollow with a flower-like cross section.5

Tests confirmed the design was not only more effective at adsorbing uranium, with a capacity of around 3g U/kg adsorbent, but it also cut the cost of extracting a kg of uranium from seawater from $1232 to $660. The adsorbent could be reused several times, and could draw between three and four mg of uranium per gram of plastic each time it was used. ‘The high surface area fibres allowed more amidoxime functional groups to be grafted onto the fibres per gram of polymer material, thus increasing the adsorption capacity,’ explains Gary Gill, who leads the testing programme at PNNL.

Meanwhile, other members of the consortium looked at ways to make the adsorbent mats biodegradable – the idea of putting plastic into the oceans, which would take thousands of years to degrade, is not ideal. So Robin Rogers and his group at the University of Alabama, US, used the chitin from discarded shrimp shells to replace the HDPE scaffold in the mats.6 To take the idea further Rogers and fellow University of Alabama researchers have formed the spin off company, 525 Solutions.

To date, they have tested their product in the lab using simulated seawater and managed to extract 94mg of uranium uptake/kg of adsorbent. According to Ronnie Hanes, CEO at 525 Solutions, work over the past year has been mainly to increase the efficiency of the chitin support and the cost of adsorbent production. ‘We have refined our electro-spinning step to produce adsorbent mats to the point that the diameter of the individual fibres has been reduced from a nominal micron to the nm range,’ says Hanes. ‘The resulting increase in surface area provides increased loading of the adsorbent functional group in the bed as well as more efficient use of the chitin backbone.’

Not all the US researchers, however, are focusing their efforts on using polyethylene as the central scaffold. In 2013, a US group, led by Wenbin Lin, professor of chemistry at the University of North Carolina at Chapel Hill, and currently at the University of Chicago, replaced HDPE with a metal-organic framework (MOF) that comprised a phosphoryl urea group as part of the bridging ligand.7 In lab tests, and using water that contained only uranium ions, Lin’s material adsorbed over 200mg of uranium per gram of adsorbent, which is higher than the plastic adsorbents developed by the ORNL team. However, according to Costas Tsouris, professor of chemical engineering at ORNL, although the MOF adsorbent works well for low pHs, it does not work well in seawater.

More recently, in 2014, Chuan He at the University of Chicago, US, and Luhua Lai at Peking University, China, have combined their efforts to engineer a protein that can bind to uranium from simulated seawater.8 They started by thinking about what functional groups would bind to uranyl ions and reasoned the best binding motifs would contain the amino acids Glu17, Asp68 and Arg71 arranged so as to surround the uranyl ion, bind to it and hold it in place. Negatively charged amino acids Asp13 and Glu64 would stabilise the positively charged uranyl ions.

With these targets in mind, the researchers searched the Protein Data Bank for a suitable candidate. In the end, their chosen protein had to be modified to increase its affinity, and the resulting engineered protein was called Super uranyl-binding protein.

To test how effectively the protein binds uranium in the ocean, the scientists first needed to attach it onto a solid support. They experimented with two options – the surface of E coli bacteria, and a sulfhydryl resin. When the protein was attached to the resin, it selectively bound uranium ions, almost ignoring the other competing metal ions. When incorporated onto the surface of E. coli cells, the protein extracted over 60% of uranium ions from the simulated seawater. However, whether either material is sturdy enough to be used in the oceans remains to be seen.

For now, it seems like braids of polymer fibres are still the most promising uranium adsorbing technology, and the most likely to be deployed in the next decade. In 2016, the US Department of Energy consortium announced they had developed a range of adsorbents with an even higher uranium adsorption capacity.9

They are still using the same polyethylene backbone and the amidoxime as the adsorbent, but they have grafted different organic acids onto the plastic fibres to increase the hydrophilicity of the adsorbent. By making the fibres hydrophilic, they increase the amount of seawater that is in contact with the amidoxime, thus increasing the amount of uranium that is adsorbed.

‘ORNL has developed a series of adsorbents that contain a variety of co-monomers that indirectly aid in the extraction of uranium from seawater,’ explains Gill. ‘These include mono and dicarboxylic acids, as well as phosphonic acid. Identifying suitable monomers and the optimum amidoxime:monomer ratio has resulted in significantly increasing adsorbent capacity over previous attempts,’ he says.

Other changes include treating the adsorbent with a base solution before it is deployed in the sea. This also increases the hydrophilicity of the adsorbent, so increasing its accessibility to uranyl ions. Using these techniques, the team has recovered over 5g of uranium per kg of adsorbent used, the most by any team so far.

For the US Department of Energy consortium, the next step is to develop adsorbents of even higher uranium capacity and selectivity that can be recycled at least six times without significant loss of capacity. ‘Within a period of five years, we have developed adsorbents that have four to five times higher uranium capacity than in the past 50 years,’ says Tsouris. ‘Working at the same pace in the next five years, we may have an adsorbent that can make the uranium-from-seawater process economically competitive.’

Charles Forsberg, principal research scientist of Nuclear Science & Engineering at MIT, Boston, US, comments: ‘Seawater uranium is not currently competitive with uranium from other sources, particularly because advances in mining and in drilling technologies have reduced the cost. However, the importance of seawater uranium is that it creates a second option for meeting long-term needs for nuclear fuel and so puts an upper limit on the cost of nuclear fuel for 1000 years or more.’

‘The ORNL researchers have made major advances in reducing the cost of uranium from seawater by improving the adsorbent. Costs have been reduced by a factor or two or three. There is the potential for further large reductions in the cost of seawater uranium,’ says Forsberg.

‘The challenge with the resin is that it also likes to adsorb vanadium—a metal that is worth a lot less than uranium. The next step in reducing costs is to modify the adsorbent so it more selectively removes uranium from seawater, relative to vanadium. That will require several more years of research and testing to see if that can be accomplished.’


1 R. V. Davies et al, Nature, 1964, 203, 1110.

2 H. J. Schenk et al, Sci. Technol., 1982, 17, 1293.

3 K. Saito et al, Nuclear Technology, 2003, 144, 274.

4 N. Seko et al, Proc. of Civil Eng. in the Ocean, 2002, 18, 737.

5 DOE Office of Nuclear Energy, Uranium from Seawater Programme Review 2013, www.osti.gov/servlets/ purl/1154652/

6 P. S. Barber et al, Green Chem., 2014: doi:10.1039/ c4c00092

7 W. Lin et al, Chem. Sci., 2013, 4, 2396.

8 Y. Lu, Nature Chemistry, 2014, 6, 175.

9 Industrial & Engineering Chemistry Research, 2016, 55(15), 4101.

Become an SCI Member to receive events discounts

Join SCI