Feasibility of uranium extraction from seawater

[Shinova]

I've read about this method of extracting uranium from seawater. How feasible and efficient is it?

[Sikon]

════════════════════════════════

Uranium Reserves & Land Mining

Right now, the spot price of uranium oxide (U3O8) is around $72 per pound, $33 per kilogram.

While total uranium in earth's crust is trillions of tons, that is just the total spread over millions of trillions of tons of rock down to a depth of a number of kilometers, with the amount actually accessible and affordable to mine being a very tiny portion of the total. For example, while the average rock has uranium at a concentration of 1 to 3 ppm or 0.0001% to 0.0003%, uranium is mined in the few locations where it is present in ore at concentrations upwards of 0.1%, typically at several tenths of a percent concentration where it is practical and economically competitive to mine.

With regular mining on land, world uranium reserves available for under $130/kg are estimated as 4.7 million tons. That is sufficient for about 85 years at the current nuclear electricity generation rate consuming around 0.06 million tons annually. However, unless breeder reactors were used for dozens of times less uranium consumption rate relative to electricity generation, those uranium reserves would last much less long in a hypothetical scenario of much expanded nuclear generation.

Nuclear power is currently only 25% of the world total, and total electricity consumption is increasing every year and decade. Nuclear generation costs vary only very little with a change in the price of uranium, due to it being a small component of total costs. If more than $130/kg uranium was acceptable, more than 4.7 million tons of uranium could be mined on land. But at some point uranium in seawater becomes more affordable to collect than to mine increasingly lower-quality resources on land.

────────────────────────────────

... Versus Uranium From Seawater

The billions of tons of uranium in seawater could last practically forever as implied here. The lowest cost now obtainable by one group is "25,000 yen [per kg of uranium] with 4g-U/kg-adsorbent used in the sea area of Okinawa, with 18 repetitions." That would be $210/kg for the cost of uranium from seawater, rather impressive performance compared to what I read about earlier a few years ago. In comparison, the $33/kg current U3O8 price from land mining is $39/kg of uranium, about 5.5 times less.

But uranium from seawater could become economically competitive if the currently mined uranium ore on land becomes more depleted eventually, and/or if seawater extraction cost decreases enough in the future. As suggested by the plot at the top of this post from that group's website, uranium from seawater might eventually cost less than the current $210/kg. The group states "when 6g-U/kg-adsorbent and 20 repetitions or more becomes possible, the uranium cost reduces to 15,000 yen," and that would be $129 per kilogram, only 3.3 times current uranium cost. Besides, the preceding is for a relatively small enterprise, and there would have to be economies of scale.

The technique of the one group using polyethylene fibers for absorption of uranium from seawater is not the only technique possible, as there have been other proposals, and even better methods may be found in the future. But the preceding is enough to get an idea of the overall picture.

────────────────────────────────

... And The Effect on Electricity Generation Cost

How does total electricity generation cost varies with uranium cost? There are different types of reactors, but illustrate here for a relatively typical light-water nuclear reactor:

Looking at one table, 7.5 kilograms of natural uranium produce 1 kilogram of reactor fuel. That would be from around 0.035 kg of U-235 ending up in the reactor fuel enriched to around 3.5% U-235 for light-water reactors. Another 0.018 kg of the 0.053 kg total original U-235 in the 7.5 kg of natural uranium (0.711% concentration) would end up in the tailings, e.g. 6.5 kg of depleted uranium with ~ 0.28% U-235 content left over. At 45,000 MWd/t burn-up, that kilogram of reactor fuel would give around 49.6 megawatt-days of thermal power, which would correspond to 360,000 kWh electricity at the 30% thermal --> electrical efficiency apparently assumed in the original source of the table.

The fuel cost relative to electricity generated would usually be calculated as the total reactor fuel cost including conversion, enrichment, and fabrication. However, such gives a figure higher than the cost of the original uranium alone. For this discussion, one wants to know how much of the generation cost was the original natural uranium cost. At current spot prices, the 8.9 kg of U3O8 with 7.5 kg of natural uranium used to produce 1 kg of reactor fuel generating 360,000 kWh would cost about $291. So the uranium cost would be around 0.081 cents per kilowatt-hour of electricity generated, for a relatively typical reactor.

This gives the following data for approximately how nuclear electricity generation cost would change with uranium cost, neglecting countermeasures that might be taken like more efficient power plants or reprocessing and breeder reactors:

2x uranium cost, $66/kg U3O8 = 0.081 cent/kw-hr increase
5x uranium cost, $170/kg U3O8 = 0.32 cent/kw-hr increase
10x uranium cost, $330/kg U3O8 = 0.73 cent/kw-hr increase

For general perspective, the average selling price of electricity sold in the U.S. (nuclear and non-nuclear average) is 6 cents per kilowatt-hour for industrial consumption.

The $210/kg-uranium current cost for seawater extraction mentioned earlier corresponds to 5.5 times the cost for land mining today, approximately a 0.36 cent per kilowatt-hour generation cost increase. The 15000 yen per kilogram or $129/kg that they expect to obtain would be 3.3 times current uranium cost and approximately a 0.19 cent per kilowatt-hour generation cost increase. Compared to the 6 cents per kilowatt-hour average sales price of U.S. industrial electricity, that is rather small indeed.

Possibly there may be use of reprocessing and breeder reactors in the future, making use of much of the U-238 that comprises more than 99% of natural uranium, converting it to Pu-239 so it could be fissioned like the U-235 present in so much smaller quantity. An International Atomic Energy Agency article implies fast reactors could get about 30 times more electricity generation per unit mass of natural uranium mined.

In any case, the preceding suggests that uranium extraction from seawater is a solution to any concerns about limited uranium reserves, provided that a relatively slight potential rise in nuclear electricity generation costs is acceptable, perhaps on the order of $0.0019/kw-hr extra to an industrial customer paying around $0.06/kw-hr.