The other day a conversation about seawater recovery of lithium got me curious on what the latest progress in seawater mineral extraction has been. The world's oceans contain vast, nearly inexhaustible reserves of many different dissolved metals, some very valuable... but most, unfortunately, in parts per billion quantities. This usually prices their recovery at several times higher to orders of magnitude higher than the costs of mining on land. Don't get me wrong - it actually does work. For example, there have been various pilot projects mining uranium with long plastic ropes lowered into ocean currents, and the concept of using desalination brine for resource extraction is a new hot topic. But so far, extraction of only a few common oceanic minerals have reached commercial viability.
Many common materials, such as plastics, tend to accumulate dissolved ions over time, acting as "chelators". Reversing the process is usually no more complex than a soak in strong acid. While all plastics can chelate to some extent, certain families, such as PAN-based plastics, are more effective than others (a commonly used one for uranium extraction is polyacrylamidoxime). Unfortunately, no matter how good or selective your plastic, the concentrations of the target minerals in the ocean are so low that you need a high surface area spread out over a large area interacting with a lot of water over a long period of time. In the process of thinking about how one could achieve that, I converged on something.
"What if you released a bunch of floating plastic absorbers into the ocean, then collected them later?"
"No, that would be known as 'polluting the ocean'."
"Well, what if they were released somewhere that they'd naturally collect themselves, like the North Pacific Gyre?"
"And what, make the Great Pacific Garbage Patch even worse? Isn't there too much plastic there already?"
"Wait a minute... there is, isn't there?"
Could it be? Estimates for the total amount of plastic in the North Pacific Gyre vary wildly, but a common figure cited is 150 million tonnes, largely broken down into tiny (high surface area) bits. Could there really already be 150 million tons of metal-collecting plastic out in the Pacific that's been floating around there for years with a massive surface area amidst vast quantities of water? Plastic that most people would pay to get rid of?
Random plastic trash certainly isn't going to be a good absorber, and it definitely won't be selective absorber - we can't pick and choose what it's going to pick up. Yet it does absorb metals. Quick searches reveal many claims that oceanic plastic has heavy metal concentrations "millions" of times higher than that of sea water. But what does the research say? Digging, I found this graph in a paper based on floating common plastics in a bay for up to a year in three different locations (labelled CC, SI, and NMF). Each type of plastic is given a color code, but their absorption rates for different metals are surprisingly similar. Using these graphs, we can estimate that the Great Pacific Garbage Patch has absorbed somewhere on the order of 7500 tonnes of aluminum, 15000 tonnes of iron, 12000 tonnes of manganese, 900 tonnes of zinc, and 75 tonnes of cobalt. Rather than being locked up in tough-to break-down oxides like most land-mined minerals, these will be present as free chelated metals, and thus relatively cheap to refine.
(Above: Plastic consumption has been connected to population declines in seabird species like Laysan Albatross. Plastic trash in the oceans poses a choking hazard, a risk of digestive blockage, and accumulates toxic organic chemicals and heavy metals.)
At current market rates, these metals in the trash would sell for $15m, $6m, $26m, $2m, and $2m, respectively. Now, this doesn't mean that you earn that much; you have to economically justify your refining and especially the cleanup, for which price estimates for the cleanup range from many tens of millions of dollars up into the billions. But it should be pointed out that nobody is proposing mining common, rather cheap minerals like aluminum, iron, manganese, and zinc from seawater, nor moderately expensive but quite rare minerals like cobalt. The focus is on things like uranium and lithium that are unusually common relative to their value.
Unfortunately, concentrations the most desirable elements weren't tested for in this research paper, nor any other paper I've been able to find. But what might we expect from them? Uranium is of roughly equal abundance to iron in seawater (0.0033 ppm), 60% that of aluminum, 165% that of manganese, and 4100% more than cobalt. It is known to be rather good at absorbing into many plastics. If we take a pessimistic approach and use the ratio for cobalt, we come up with an estimate of 3000 tonnes of recoverable uranium, which at a current market price of $70000 a tonne is worth $210 million. If we instead use the ratio for manganese, we come up with an estimate of about 20000 tonnes of uranium, worth $1.4B.
Now we're getting somewhere. Lithium is a very abundant element in seawater, at 0.18ppm. If we extrapolate from a similar light reactive metal like zinc, we come up with an estimate of 32400 tonnes recoverable. If we extrapolate from something like manganese, we get over a million tonnes. But I don't think it's realistic to expect that this much of the plastic's weight will end up composed of lithium. If we assume that it caps off at say 15000 tonnes like iron, then at a market rate for lithium metal of $60000/tonne, it'd be worth $900m. For lithium carbonate, it'd be worth half that much. On the other hand, if it truly can collect those higher concentrations of lithium, then the value could easily be in the billions.
The list goes on. Strontium is ridiculously abundant in seawater, at 8.5ppm, and currently sells for about $5k/kg in its metallic form. Rubidium is only slightly less common than lithium at 0.12ppm, and goes for $1.2 million per tonne, although it's a low-demand metal. Molybdenum, a high demand metal, is found at 0.01ppm - double that of aluminum and at $30k/tonne sells for 15 times higher a price. Copper, at 0.03ppm, is as common as iron and at $6k/tonne sells for 20 times its price. Vanadium, slightly less common than manganese at 0,0015ppm, sells for $14k/tonne, over 6 times the price of manganese. At 0.005ppm, high purity cesium metal sells for over $10m/tonne, although it's more commonly found as low purity cesium formate selling for (a still very significant) $100k-200k/tonne. And even gold - yes, gold - is found in earth's oceans. Remember that 75 tonnes of cobalt recoverable from the plastic? There's 60% as much gold in the Earth's oceans as there is cobalt, so if that ratio held and 45 tonnes of gold was recovered, at $50m/tonne it'd add another $2B to the value. If gold was recovered proportional to concentration at the same rate as manganese, the value would be $15B.
Now, to reiterate, just because X value of a metal is present in an available source material doesn't mean you earn X dollars from it. At the very least recovery will cost you something, and recovery might prove in many cases to be more expensive than the value of the metal within. And in this case, we can only speculate at what sort of concentrations of the more valuable elements are available to recover. But with so many minerals that are common in the oceans but rare in the crust, and with the trash accumulating these metals (making them a hazard for ocean life but a value for us), perhaps it's time we take a closer look at cleaning up our oceans... for a profit.