Extracting Uranium From Seawater

If, like most scientists, you haven’t thought much about the composition of seawater, it may come as a surprise that the oceans contain uranium. In fact, they hold a staggering quantity of the heavy element.

“It’s estimated that more than 4 billion metric tons of uranium are dissolved in the Earth’s oceans,” according to Benjamin P. Hay, a distinguished scientist at Oak Ridge National Laboratory (ORNL). That quantity is roughly 1,000 times greater than all known terrestrial sources combined—enough to fuel the world’s nuclear power industry for centuries, even if the industry grows aggressively. World production of uranium in the past decade has ranged from 40,000 to 50,000 tons per year. But as nuclear-fuel scientists point out, tapping the ocean’s uranium for industrial use means coming up with an economically viable, long-term, and large-scale method of extracting an extremely dilute supply of uranium under challenging marine conditions.

A number of researchers working to surmount those challenges gathered at last month’s American Chemical Society national meeting in Philadelphia at a symposium sponsored by the Division of Industrial & Engineering Chemistry to review the field’s status and share their latest findings. Those researchers, along with others not present at the meeting, are drawn by the potential payoff of meeting those challenges. They’re working to develop chemically selective and durable adsorbents—typically functionalized polymeric fibrous materials—that can be dropped into the ocean to soak up uranium, withdrawn periodically, stripped of the metal, and reused. These scientists are evaluating various types of polymers and functional groups and analyzing the economics of extracting uranium from the oceans.

The idea of pulling uranium from seawater isn’t new. Scientists at the United Kingdom Atomic Energy Authority (UKAEA) addressed the topic in detail in a 1964 paper in Nature that is cited frequently by researchers in this specialized field (DOI: 10.1038/2031110a0). In that paper, which refers to work published as early as 1953, the British team cited energy security benefits as the motivation for looking to the sea for an abundant supply of uranium.

The UKAEA group went on to point out that uranium’s concentration in seawater is “remarkably constant” but incredibly low—only 3.3 μg per L, or 3.3 ppb, which makes extracting the metal a major challenge. In addition to its low concentration, uranium in seawater is bound up as a chemically stable carbonate complex of the uranyl cation, UO₂²⁺. Further complicating potential extraction strategies is the presence of alkali and many other metal ions, some of which are present at overwhelmingly greater concentrations.

It was clear to scientists some 60 years ago, just as it is today, that to exploit the ocean’s reserves of uranium, researchers would have to come up with a high-performance extractant. The material would need to do its job effectively and selectively at the slightly basic pH and high ionic strength of seawater. And it would need to be extremely insoluble.

The British team reviewed published studies, including some based on solvent extraction techniques, and evaluated several solid sorbents. They judged adsorption on solids to be the most sound extraction method and found titanium hydroxide to be among the most promising sorbents. Yet with an uptake of only about 0.1 g of uranium per kg of sorbent, that method came to be regarded as too inefficient for industrial application.

At this point in the story, the uranium trail goes cold, at least for a while. Little progress was reported in the field for roughly the next 20 years, according to Sheng Dai, an ORNL group leader and University of Tennessee, Knoxville, chemistry professor. Then, in the early 1980s, scientists at the Nuclear Research Center in Jülich, Germany, conducted a systematic evaluation of 200 ion-exchange resin materials. The team found that among all materials tested, only amidoxime-based compounds, specifically poly(acrylamidoximes), met the requirements for chemical stability and selective uptake of uranium under typical marine conditions.

Although a number of other uranium extraction methods have been studied, their various shortcomings have kept the focus on amidoxime-based adsorbents. Those shortcomings include solvent losses sustained in solvent extraction processes and the high cost of pumping seawater through columns of inorganic sorbents. Motivated by the Jülich team’s findings, nearly all researchers in the past dozen years or so have sought ways to avoid those limitations and have focused their efforts on the somewhat obscure amidoximes. Those compounds are oximes, R₁R₂C=NOH, in which one of the Rs is an NH₂ group.

By far the largest field test was a massive two-year-long effort between 1999 and 2001 led by Noriaki Seko and Masao Tamada of Japan Atomic Energy Research Institute, now known as Japan Atomic Energy Agency. The team prepared nonwoven sheets of amidoxime-functionalized polymer and loaded stacks of the sheets—separated by spacer nets—into three large connected cages that were lowered from a floating rig into the Pacific Ocean several miles from Japan’s coastline.

To make the adsorbent, the researchers exposed sheets of a polyethylene-polypropylene blend to an electron beam, which generated surface radicals. Then they grafted acrylonitrile together with a hydrophilic monomer onto the fabric at the radical sites and treated the graft with hydroxylamine to convert the acrylonitrile-cyano groups to amidoxime units.

The truck-sized rig contained a total of 52,000 adsorbent sheets that weighed nearly 800 lb when dry. The team withdrew the contraption every few weeks to analyze the sorbent for uranium uptake. In total, the group submerged the rig for 240 days and recovered just more than 1 kg of uranium from ocean currents flowing through the cages, thereby avoiding the need for pumps. With this setup, the Japanese team extracted, on average, 0.5 g of uranium per kilogram of sorbent in a 30-day period. They described the work in detail in a 2003 paper in Nuclear Technology (144, 274).

Having firmly established that uranium can be extracted from the oceans in appreciable quantities, Seko, Tamada, and coworkers turned to lowering the cost of the operation and boosting the sorbent’s uptake capacity. They determined that 40% of the cost of retrieving metal from the sea was associated with the rig, adsorption cages, and other mechanical components of the gear. So the group took a different tack. They fashioned long seaweedlike braids of amidoxime-functionalized polyethylene fiber and attached the braids via remotely controllable fasteners to anchors and lowered them to the ocean floor. The artificial seaweed configuration raised the 30-day uranium uptake to roughly 1.5 g per kg of sorbent.

“The jump in uranium uptake was a major breakthrough,” Dai told C&EN, but to make sorption technology practical and economically competitive with conventional mining, “we need to double that capacity.” Researchers also need to maximize the number of times a sorbent can be stripped of accumulated uranium and recycled, in order to cut costs. Those goals are among the aims of a Department of Energy program in nuclear fuel resources. The program, for which Dai serves as technical lead, brings together researchers from several national labs and universities.

The basic strategy for raising the uranium extraction capacity of sorbents is two-pronged: design highly active and chemically selective uranium chelating agents and append the ligands comprising those molecular units to a support material (a polymer core, for example) in a manner that leads to an enhanced number of uranium binding sites. Executing that strategy requires filling in several basic-science blanks.

Many studies have been carried out on amidoxime-based sorbents, Dai remarks, “but we still don’t really understand why this functional group binds uranium so well.” His ORNL colleague, Hay, adds that even the exact nature of the uranyl-amidoxime binding mode is unclear. Researchers have proposed a number of possibilities. Knowing which of the proposed binding configurations is the most stable, and hence the most prevalent, is sure to be one of the keys to designing enhanced chelating agents, which may consist of multiple amidoxime groups.

To weigh in on that issue, the ORNL team—Hay, Sinisa Vukovic, and coworkers—applied quantum mechanical methods to assess the stability of complexes formed via three distinct uranyl-amidoxime binding motifs. For simplicity, the study addressed free amid­oximes, as opposed to ones bound up in polymeric sorbents.

In a paper published earlier this year, the team reported that the calculations show that edgewise binding between amidoxime’s N–O bond and uranium, the so-called η² binding mode, is more stable than the other proposed arrangements—monodentate and bidentate structures. The group confirmed the computational results by conducting X-ray diffraction studies on single crystals of uranyl complexes with acetamidoxime and, separately, benzamid­oxime (Inorg. Chem., DOI: 10.1021/ic300062s).

The ORNL team took a similar approach—studying simple analogs of polymer-bound species—to evaluate the chemistry of cyclic amidoximes. Several researchers have proposed that cyclic imide dioximes are the molecular entities that actually bind uranium in amidoxime-functionalized polymer sorbents.

Prevailing wisdom holds that these structures are formed during treatment of the sorbent with strong base, a step that is known to enhance uranium adsorption. Conventional wisdom also says that the gradual loss of capacity that occurs each time the sorbent is stripped of accumulated uranium by treatment with concentrated hydrochloric acid is a result of acid degradation of the cyclic dioximes.

To better understand those processes, Hay, ORNL’s Sung Ok Kang, and coworkers treated a dinitrile compound, glutaronitrile, with hydroxylamine and used nuclear magnetic resonance spectroscopy to monitor the influence of heat, base, and acid treatments on the products. The reaction yields two products at room temperature—an open-chain bisamidoxime and a cyclic imide dioxime. The group found that at elevated temperature, however, only the cyclic compound forms. The team also found that heating the open-chain product converts it to the cyclic one even in the absence of base. And in the same study, Hay’s group demonstrated that aromatic cyclic imide dioximes are orders of magnitude more resistant to acid degradation than nonaromatic ones (Ind. Eng. Chem. Res., DOI: 10.1021/ie300492z).

These findings suggest that simply heating amidoxime-based polymers may boost their uranium capacity, Hay said. The results also suggest that new adsorbents that incorporate phthalimide dioxime-type moieties should display improved resistance to acid stripping.

The NMR study sheds light on some of the basic properties of select cyclic imide dioximes. But do they actually bind uranium? According to Linfeng Rao, Xiaoqi Sun, and coworkers at Lawrence Berkeley National Laboratory, the cyclic glutarimidedioxime binds UO₂²⁺ ions tightly. The LBNL team just published their microcalorimetry and potentiometry findings in Dalton Transactions (DOI: 10.1039/c2dt30978e). Now the group is evaluating uranium uptake in phthalimide dioxime.

Several other research groups are working on various pieces of the uranium extraction challenge. For example, University of Idaho chemistry professor Chien M. Wai is exploring options for using supercritical CO₂ to leach uranium from amidoxime-based sorbents in place of hydrochloric acid, the standard leaching agent. Avoiding strong acids should extend the sorbents’ lifetime and recyclability.

And at Scripps Research Institute, Julius Rebek Jr. and coworkers are studying alternatives to amidoxime-based chelating agents. Specifically, the Scripps group observed that when UO₂²⁺ ions coordinate to three 2,6-terphenyl carboxylic acid ligands, the ligands’ bulky phenyl groups sequester the ion in a self-assembling cage (Chem. Sci., DOI: 10.1039/c0sc00116c).

Others are looking to boost the surface area of the materials to which the chelating agents are attached. At ORNL, Christopher J. Janke and coworkers have teamed up with engineers at West Melbourne, Fla.-based Hills Inc., a company specializing in advanced fiber extrusion technologies, to produce high-surface-area polyethylene fibers. The material, which features a record-setting diameter of just 0.24 μm, earned the team R&D Magazine’s 2012 R&D 100 Award.

By using amidoxime functionalization methods similar to the ones developed by the Japanese team, Janke and coworkers have made a variety of braided sorbents from these fibers that they say outperform state-of-the-art adsorbents in terms of uranium capacity, selectivity, and adsorption rate. At the ACS meeting, the team reported that in seawater tests conducted with Gary Gill and coworkers at Pacific Northwest National Laboratory’s (PNNL) marine science center in Sequim, Wash., the new material extracted 3.4 g of uranium per kg of sorbent, a record-setting value.

Meanwhile, at the University of Alabama, chemistry professor Robin D. Rogers’ research group recently developed an alternative high-surface-area material. They use an ionic-liquid-based method to extract the biopolymer chitin from shrimp shells and then spin the material into nanofiber mats. Just recently, the team conducted tests showing that they could attach amidoxime groups to the chitin fibers and use the biobased adsorbent to selectively extract uranium from seawater-like samples. Using a chitin-based adsorbent in place of a synthetic one is attractive, Rogers told C&EN, because when the material eventually breaks down, “we’d be putting back into the ocean something that comes from the ocean and is biodegradable.”

Pulling uranium from the seas is technically possible, but is it economically feasible? Erich A. Schneider, a mechanical engineering professor at the University of Texas, Austin, recently completed a top-to-bottom cost analysis and concluded that the technology could generate uranium at an approximate cost of $1,200 per kg, which is roughly 10 times today’s spot price for the metal. The newly reported uranium uptake value by the ORNL-PNNL group lowers the projected price to roughly $600 per kg. Schneider noted that the factors most strongly affecting cost are the sorbent’s extraction capacity, the intense use of chemicals in sorbent preparation, and sorbent recyclability. Improvements in any of those areas will further lower costs.

It’s been 50 years since British scientists pointed out that “the ocean is a virtually limitless reservoir of uranium.” Tapping that source has proven to be very challenging, Dai told C&EN, “but I’m passionate about this project. If we can figure out how to overcome these hurdles, this technology will be a real game changer.”