Selenium Remediation

Where selenium comes from and why it is hard to remove

Selenium reaches water from coal mining and coal ash, phosphate and copper mining, oil refining, and irrigation drainage through seleniferous soils, as well as from naturally selenium-rich shales. It is an essential nutrient at trace levels and toxic to fish and birds at levels not far above them, which is why the regulatory pressure comes less from the 50 µg/L federal drinking water standard than from aquatic life protection: EPA’s 2016 freshwater criterion is tissue-based with water column values of 1.5 µg/L for lakes and reservoirs and 3.1 µg/L for flowing water, and discharge permits at mines and power plants are often written in single-digit micrograms per liter.

The treatment problem is speciation. In oxygenated water selenium is present as selenate, Se(VI), the SeO₄²⁻ anion, which is highly soluble, weakly sorbed and chemically similar to sulfate. Selenite, Se(IV), binds well to iron oxides but is a minority form in most mine and ash-pond waters. Elemental selenium, Se(0), and iron selenide are insoluble. Conventional coagulation removes selenite but very little selenate, so a selenate-dominated water has to be chemically or biologically reduced before anything will hold it.

How zero valent iron treats selenium

Ferox zero valent iron reduces selenate and selenite to insoluble elemental selenium and iron selenides, and the iron oxides that form as the iron corrodes adsorb what remains. That combination is why zero valent iron appears in so much selenium treatment research. In anoxic batch tests on synthetic mine water, 90% of selenate was gone in 1.5 hours with no competing ions, while 1,800 mg/L of sulfate stretched that to 33 hours; a Québec study took a real mine effluent below the 0.5 µg/L detection limit at a 1.25% solids-to-liquid ratio and eight hours of contact. On paper, reducing selenate to elemental selenium takes six electrons, two to three moles of iron per mole of selenium, or about 1.4 to 2.1 lb of iron per pound of selenium; in practice the iron loading is set by the water’s sulfate, nitrate and dissolved oxygen, which consume the same electrons, and by contact time.

Two field cautions come with the chemistry. Iron released from the media can carry selenium out with it unless a downstream filter catches the iron, and zero valent iron reduces nitrate mostly to ammonium, which can trade one discharge limit for another at a mine with blasting residue in its water. Both are checked on the bench before a reactor is sized.

Granular Ferox PRB, with more than 95% iron, 96% of particles in the 20/60 mesh range and no rust, dust or oil, is the media for above-ground reactors and for permeable reactive barriers that intercept seepage from waste rock, tailings and ash ponds. The finer Ferox Flow powder carries the same chemistry into slurry injection where a selenium plume has to be treated in place. Our zero valent iron remediation guide covers how particle size and delivery method are chosen.

Biological reduction

Selenate-reducing bacteria are common in soil and sediment, and given an organic electron donor they reduce selenate through selenite to elemental selenium, which accumulates as fine red particles inside and around the cells. That is the basis of the passive bioreactors and engineered biological systems used for mine drainage, coal ash water and refinery effluent. Emulsified vegetable oil is the slow-release donor that sustains a reducing zone for years; sodium lactate is the fast-acting donor used to establish reducing conditions quickly, and most programs use both. Nitrate is consumed before selenate, so donor demand is set by the nitrate load as much as by the selenium, and temperature matters: biological rates fall sharply in cold mine water, which is one reason zero valent iron is often chosen for year-round discharge compliance.

Ferrous sulfate for selenite

Where selenium is present mainly as selenite, coagulation with ferrous sulfate, oxidized in the water to hydrous ferric oxide, removes it by co-precipitation, and jar testing sets the iron dose. The limitation is that this route does not remove selenate, so a speciation analysis comes first; a water that is mostly selenate needs iron reduction or biological treatment instead.

What decides success at a selenium site

Speciation. Selenate and selenite behave differently in every process above. Measure both before choosing a technology.

Sulfate and nitrate. Sulfate slows selenate reduction by iron and competes for sorption sites; nitrate consumes iron and biological electron donor first. Mine water routinely carries hundreds to thousands of milligrams per liter of both.

Contact time and iron carryover. Reactor sizing follows from bench kinetics on the actual water, and a filtration step for fine iron is part of the design, not an afterthought.

Temperature. Chemical reduction by iron is less temperature-sensitive than biological reduction, which matters for winter discharge limits.

What Hepure provides

We supply the chemistry and help you size it: Ferox zero valent iron from 1-lb and 5-lb samples through 100-lb bags, 1,000-lb barrels and 2,205-lb super sacks, with full containers for large programs; emulsified vegetable oil as a 60% solution in 55-gallon drums, 265-gallon totes or by tanker; sodium lactate; and ferrous sulfate in 50-lb bags. Send Pat or Bob the selenium concentration and speciation, sulfate and nitrate, pH, temperature and flow, and whether the work is a reactor, a barrier or an injection, and you will get a product recommendation and a quantity estimate. Lab treatability testing and column studies for barrier design are available. We do not design the treatment system or operate it; that stays with your engineer.

Common questions

Why is selenium harder to treat than most metals? Because selenate is an anion that behaves like sulfate: it does not precipitate with hydroxide or sulfide and barely sorbs. It has to be reduced first, and the water it comes in usually contains far more sulfate and nitrate than selenium.

How low can zero valent iron get selenium? Bench and pilot work has reached sub-microgram-per-liter levels on real mine effluent with sufficient contact time and iron loading. Whether that holds on your water, and at what cost in iron and residence time, is what the bench test answers.

Is the elemental selenium stable? Elemental selenium and iron selenide are stable under reducing conditions. Exposure to oxygen over long periods can slowly re-oxidize them, so iron barriers and bioreactors are kept anoxic and spent media is handled accordingly.

Can zero valent iron and biological treatment be combined? Yes. Iron corrosion produces hydrogen and a reducing environment that supports selenate-reducing bacteria, which is why iron and an organic electron donor are sometimes placed together. The split between the two is a design decision for your engineer; we supply either or both.