Final Report: Highly Efficient Removal of Mercury From Industrial Flue Gas

EPA Contract Number: EPD04036
Title: Highly Efficient Removal of Mercury From Industrial Flue Gas
Investigators: Hensman, Carl E.
Small Business: Frontier Geosciences Inc.
EPA Contact: Richards, April
Phase: I
Project Period: March 1, 2004 through August 31, 2004
Project Amount: $70,000
RFA: Small Business Innovation Research (SBIR) - Phase I (2004) RFA Text |  Recipients Lists
Research Category: SBIR - Waste , Hazardous Waste/Remediation , Small Business Innovation Research (SBIR)


It is estimated that 144-189 megagrams (158-207 tons) of mercury are emitted annually into the atmosphere by anthropogenic sources in the United States (Keating, 1997; National Atmospheric Deposition Program Mercury Deposition Network). Approximately 87 percent of the mercury comes from combustion-point sources, and 10 percent is from manufacturing-point sources. The combustion-point sources can be broken down further into four classes: (1) coal-fired utility boilers, (2) municipal waste combustion, (3) commercial/industrial boilers, and (4) medical waste incinerators. All of these are high-temperature waste combustion or fossil fuel processes. In each case, the mercury is an impurity in the fuel or feedstock and is volatilized as a result of the low mercury boiling point and discharged to the atmosphere with the flue gas. Even though mercury is a minor impurity, the large quantity of fuel or feedstock used results in massive mercury discharges.

Future trends in mercury emissions are dependent on how utility industries address the Clean Air Act, the rules that the U.S. Environmental Protection Agency (EPA) will introduce, and state- implemented regulations requiring discharge levels lower than current federally mandated levels. In an ideal situation, mercury would not be included in the raw materials used in the processes described above, thus negating the concern over mercury emissions. Unfortunately, it is not currently feasible to remove the trace mercury before it enters the process. Industry has started to contemplate removing the impurities during the manufacturing cycle; however, the easiest location for mercury capture is still flue gas discharges.

Wet scrubbing is used in 20-30 percent of U.S. coal-fired plants. The systems are designed mainly for solid particle or SOx removal. A typical system sprays water counter-current into the flue gas, particulates or SOx are captured by the scrubber water, which goes to a separation system to remove the particulates or SOx. As a serendipitous side benefit, organic and inorganic materials easily dissolved in water (e.g., HgCl2) also are partitioned into the scrubber water and removed from the flue gas. A large fraction of the mercury in flue gas, however, is elemental mercury (Hg0) and will not be removed by a simple wet scrubbing system, ultimately ending up in the environment. Although inorganic mercury itself is not bioaccumulative, it is readily converted to methyl mercury in the ambient environment.

Summary/Accomplishments (Outputs/Outcomes):

To address these and other concerns, Frontier Geosciences, Inc., has discovered a method in which Hg, no matter what the form, will be removed by chemically modified scrubber water. The “trapped” Hg is easily separated from the scrubber water in a form that passes all required Toxicity Characteristic Leaching Procedure (TCLP) control limits. Additionally, the technology is easily adaptable to existing plant equipment, thereby reducing capital and implementation costs.

As mentioned, scrubber water naturally removes Hg species that are easily dissolved (e.g., monomethyl Hg and HgCl2). Therefore, this project focused on Hg0 removal. Hg0 is extracted into the scrubber water with a powdered solid amendment or a flocculating polymer. The combined waste solids then can be removed by an industry-standard filter system and the scrubber water can be recycled back into the treatment system.

During this project, the solid and polymeric amendments were tested in a bench scrubber system under a variety of conditions using both nitrogen and synthetic flue gas containing between 10-50 μg/m3 of Hg0, and two wet scrubber designs. The studies addressed optimal dosing amounts of the amendments, the effect of pH, the physical optimization, and the ability to recycle materials.

When compared to the solid amendment, analogous materials with an inert surface did not remove Hg0 from the gas phase. A range of materials from the family of the solid amendment materials had a range of responses. The best solid amendment had a removal of 91 percent Hg0.

The polymer needs an activator, in this case Fe3+, to start the Hg0 removal process. This is not a concern for Frontier Geosciences, because Fe will be ubiquitous in the scrubber water in an industrial application. Of the three polymers studied, two performed in an outstanding manner. One specific polymer was chosen for the remainder of this research project because it is commercially available.

The polymer had its greatest Hg0 removal efficiency of 92 percent for the lowest doping concentration, 10 ppm. This was expected because this type of polymer needs to have a low concentration to force the formation of a good flock structure (thus good complexation). Unfortunately, the formulation preservative reduces the separation ability of the solid from the scrubber water. This is a formulation issue, and will be addressed further in Phase II.

There is a relationship between the volume of scrubber water and the Hg0 removal efficiency. In both the solid and polymeric amended scrubbers, the best result was achieved with a scrubber water volume greater than 600 mL for the gas flow and amendment concentrations. This relationship will need to be further pursued for a venturi nozzle designed scrubber in Phase II to truly assess the requirements for scale-up.

The Hg0 removal efficiency is relatively independent of pH for the solid material amendment. The polymeric amendment is optimal at pH 5, which is consistent with its known operating values. The mass balance indicates that in the highly oxidizing environment of pH 1, the solid amendment is not able to retain the Hg on the material. As the pH rises above pH 7, the environment becomes reducing and there is a decrease in the amount of Hg bound to the solid amendment. This decrease, however, is not significant.

When using the synthetic flue gas and fly ash in the place of nitrogen gas, the polymeric system requires a longer period of time to reach a steady state. There also is an approximate 14 percent reduction in the polymeric efficiency to remove Hg0 at a steady state, and a faster consumption of the removal capacity. The solid amendment system operates the same when using the synthetic flue gas and fly ash as when using nitrogen carrier gas. Fortunately, the fly ash and synthetic flue gas appear to stabilize the system better than when using only nitrogen gas as a carrier. It also appears that the solid amendment performs slightly better than the polymer under industry-simulated conditions.

A TCLP on the solid amendment saturated with Hg0 results in 1.3 ± 0.6 ng/L Hg being leached. A TCLP on the polymer precipitated complex results in 35.64 ± 7.2 ng/L Hg being leached. Both of these concentrations are significantly below the EPA-recommended 200,000 ng/L Hg limit. Therefore, both materials could be deposited in a nonhazardous landfill.

For both amendments, there is a reduction in the amount of Hg0 removed from the gas phase as they become poisoned. The polymeric amendment demonstrates a better response than the solid amendment, but both show a linear decrease in removal efficiency over the 13-hour period. These results suggest that only 3.84 g of solid amendment or 0.19 g of polymer per gallon of scrubber water would give a constant greater than a 90 percent reduction in Hg0 over a 13-hour period. It also has been shown, for the solid amendment, that the scrubber water can be directly reintroduced to the system without further treatment after separation of the contaminated solid. Finally, it has been shown that the solid precipitate of the polymer/Hg complex can be crushed to powder and reintroduced to the system as a solid amendment. The waste polymer solid continues to remove Hg0 with an efficiency of 85 percent.


Currently, the technology has been presented to the Senate Committee reviewing mercury emissions reduction technology. Several negotiations are underway with the U.S. Department of Energy, and coal-fired utilities are expected to begin beta testing. The petroleum industry has shown significant interest in the ability of the system to remove Hg from natural gas.

Supplemental Keywords:

small business, SBIR, EPA, mercury, industrial flue gas, hazardous waste management, combustion, coal-fired utility boilers, medical waste incinerators, feedstock, emissions, scrubbing, SOx, Toxicity Characteristic Leaching Procedure, TCLP, polymer, fly ash, Hg0,, RFA, Scientific Discipline, Air, Toxics, Waste, INDUSTRY, National Recommended Water Quality, Chemical Engineering, air toxics, Environmental Chemistry, HAPS, Industrial Processes, Incineration/Combustion, 33/50, Engineering, Chemistry, & Physics, Environmental Engineering, combustion byproducts, mercury, medical waste incinerator, air pollution control, flue gas, coal, combustion technology, industrial boilers, Mercury Compounds, mercury & mercury compounds, flue gas emissions, aqueous scrubbing, combustion, combustion exhaust gases, coal combustion, coal fired power plants, combustion flue gases, toxicity characteristic leaching procedure, removal, flue gases