4d-4: Technologies and approaches for mitigating mercury emissions

Mercury emissions from coal - fired power and Cement producing plants have been recently regulated in the U.S. by a set of rules set forth in the Mercury and Air Toxics Standards (MATS) and National Emission Standards for Hazardous Air Pollutant (NESHAP). These rules are anticipated to significantly reduce mercury emissions and consequently lower the concentration of mercury in the flue gas effluent from these facilities.

This presentation will give an overview and discuss the current status of mercury measurements and compliance methods available for Plant Operators.

Accumulated field experience on continuous monitoring at very low levels using sorbent trap technology will be presented. As sorbent traps pre-concentrate the sample prior to analysis, they have the ability to accurately measure very low mercury levels. In fact, sorbent trap-based mercury monitors (STMM) might be the only approach that can accurately measure at the levels mandated by MATS.

A number of STMM systems have been operating in states that required mercury emission monitoring prior to MATS. As a result, electric utilities have become familiar with the technology and maintenance requirements and costs are well characterized. Although operating costs are comparable to those of a traditional continuous emission monitoring system (CEMS), the annualized total cost for STMM systems is considerably less.

The disadvantage of sorbent - trap based mercury monitoring is the lack of temporal resolution in the data as would be required for process control. However, STMM systems could be coupled with a CEMS if feedback for process control is needed. The CEMS could be configured just for process intelligence without the constraints when accumulating data for compliance demonstration. Experience from different Power Plants using portable CEMS for process control along with STMM will be presented.

Mercury emissions from coal - fired power and Cement producing plants have been recently regulated in the U.S. by a set of rules set forth in the Mercury and Air Toxics Standards (MATS) and National Emission Standards for Hazardous Air Pollutant (NESHAP). These rules are anticipated to significantly reduce mercury emissions and consequently lower the concentration of mercury in the flue gas effluent from these facilities.

This presentation will give an overview and discuss the current status of mercury measurements and compliance methods available for Plant Operators.

Accumulated field experience on continuous monitoring at very low levels using sorbent trap technology will be presented. As sorbent traps pre-concentrate the sample prior to analysis, they have the ability to accurately measure very low mercury levels. In fact, sorbent trap-based mercury monitors (STMM) might be the only approach that can accurately measure at the levels mandated by MATS.

A number of STMM systems have been operating in states that required mercury emission monitoring prior to MATS. As a result, electric utilities have become familiar with the technology and maintenance requirements and costs are well characterized. Although operating costs are comparable to those of a traditional continuous emission monitoring system (CEMS), the annualized total cost for STMM systems is considerably less.

The disadvantage of sorbent - trap based mercury monitoring is the lack of temporal resolution in the data as would be required for process control. However, STMM systems could be coupled with a CEMS if feedback for process control is needed. The CEMS could be configured just for process intelligence without the constraints when accumulating data for compliance demonstration. Experience from different Power Plants using portable CEMS for process control along with STMM will be presented.

Gypsum is a byproduct of the wet flue gas desulphurization (WFGD) system, which is most commonly used clean technology in the coal burning industry. In addition to gypsum, being the main product of desulphurization process, several other compounds are formed, among which ferro-ferry oxy-hydroxides are very important due to their abundances in WFGD gypsum and their adsorptive properties. The thermal stability of mercury compounds in gypsum has become an important issue recently, due to a potential use of gypsum in the context of circular economy.

In this work, quadrupole mass spectrometer (QMS), with direct sampling into the closed ion source with cross beam configuration was applied as a detector, instead of commonly used CV-AAS. The arrangement enabled us to measure samples with Hg concentration bellow 50 ng g-1 using at least 5 mg of sample. The QMS allows the detection of several ions simultaneously with a dynamic range of 107. In addition to Hg, other ionic species such as HCl+, H2S+, SO2+, SO3+ can also be measured. Unfortunately no correlation could be found with Hg+ ions in desorption spectra, probably due to high difference in Hg+ versus HCl+ etc. concentrations.

Three different gypsum slurries of block 6 were taken from lignite burning power plant. All gypsum slurries were additionally separated into two phases by gravitation (particle size of finer ~0.1-50 µm and coarse ~10-150 µm fraction). Thermograms (desorption spectra) obtained from all analysed samples exhibit one single desorption peak at about 230±20°C. Considering iron as a crucial factor in understanding adsorption/desorption process we performed a number of simulation experiments with and without iron, in the form of FeOOH, using saturated solution of CaSO4·2H2O (~2g/l) and different Hg-compounds such as HgCl2, Hg2Cl2, HgO, HgSO4 and Hg2SO4. In all cases the resulting thermograms (that contain FeOOH) exhibit single peak at similar temperature (230±30°C) irrespective to the form of Hg that was added to gypsum solution showing similar picture to the WFGD samples.

Thermograms obtained in this study differ from those in previous work, where gypsum slurry was taken from different blocks. This reflects importance of WFGD chemistry and APCDs (only block 6 has installed SCR). It is assumed that in a specific gypsum slurry adsorption/binding process of Hg is similar in all cases no matter to Hg-species present. Further studies are needed to confirm this.

The development of effective sorbents for Hg0 removal during coal gasification has attracted increasing concern in recent years. H2S presents in the flue gas of coal gasification and plays an important role in Hg0 adsorption. However, the effect of H2S on Hg0 adsorption and the stable forms of mercury on carbon surface during coal gasification are still unclear. Furthermore, there is even controversy on the effect of H2S on Hg0 adsorption. Understanding the effect mechanism of H2S on Hg0 adsorption is important to the design of carbons with faster kinetics and greater capacities for Hg0 removal during coal gasification. The present study addressed the effect mechanism of H2S on Hg0 adsorption at the molecular level by performing density functional calculations. The theoretical results indicate that the Eley-Rideal mechanism with Hg0 adsorption on sulfur adsorbed carbon surface is the most possible reaction process for the formation of HgS. The theoretical results are further verified by experiment and a group of adsorption experiments are conducted. The Hg0 removal efficiency of carbon increases significantly in the presence of H2S. The Hg0 removal efficiency of H2S pre-adsorbed carbon is higher than that of the adsorption on original carbon. This suggests that H2S adsorption can form active sites on carbon surface and then improve the removal of Hg0, which is in good agreement with the DFT calculation results. Temperature programmed desorption results indicate that Hg0 is more likely to be adsorbed by active sites which formed by H2S adsorption. XPS analysis suggests that Hg0 can react with sulfur species on carbon surface. Combining experimental and computational results together, the EleyRideal mechanism with H2S pre-adsorption can be determined.

To remove elemental mercury (Hg0) in chlorine-free coal combustion flue gas efficiently, a series of sorbents based on CuCl2 loaded Ce0.67Zr0.33O2 solid solution (denoted as CuCl2/CZ) were developed. In this way, the abundant Cl and affluent active chemical adsorbed oxygen (O*) on the sorbent would be combined and made full use for Hg0 removal. The XRD, XPS, FSEM-EDX, Raman, TG, and N2 adsorption/desorption were employed to characterize the catalysts. The Hg0 removal behaviors over CuCl2/CZ were studied using a laboratory-scale fixed-bed reactor. The catalyst with optimal loading 6% CuCl2 exhibited better performances (89.6%-97.1%) within a much wider applicable temperature range (100-300 °C) in the absence of HCl with a much higher gas hourly space velocity (380,000 h-1). The results indicated that the mercury removal efficiency over CuCl2/CZ was higher than those over the reported CuCl2 supported on Al2O3, Fe2O3, TiO2, activated carbon, zeolites, etc. The interaction between Cl and chemical adsorbed oxygen was probably responsible for its superior performance. The XPS results of the fresh, thermally treated, and spent 6% CuCl2/CZ might explain the Hg0 oxidation process. Hg0 was firstly oxidized by consuming lattice Cl in CuCl2, and reducing it to CuCl. Then, the CuCl will be oxidized to an intermediate product (Cu2OCl2) by O2. As no Cu+ was observed and oxygen enrichment was detected on the spent CuCl2/CZ sorbent, the Cu2OCl2 then could be decomposed and generated CuCl2. In this way, some lattice Cl was consumed and there was no more Cl species in the chlorine-free flue gas which could replenish the consumed lattice Cl. Fortunately, for the 6% CuCl2/CZ, the Hg0 removal efficiency still kept higher than 80% after 24 hour tests with 380,000 h-1 space velocity. The spent catalyst could be regenerated under HCl and O2 stream after thermally treated at 350-450 °C. SO2 (1000 ppmv) just slightly inhibited the Hg0 removal efficiency even though there was some steam (5% v/v) in the flue gas.

Air pollutant control devices in coal-fired power plant were mainly consisted of a selective catalyst reactor (SCR), an electrostatic precipitator (ESP) and a flue gas desulfurization (FGD). In this configuration, elemental mercury was oxidized by temperature decrease, existence of complexity, and reaction with halogens. Then the oxidized mercury was usually removed by adsorption in fly ash or wet type control device. As a result, the elemental mercury was dominant in flue gas from coal-fired power plant. Adsorption by activated carbon was introduced as one of the best available technologies in removal of element mercury in flue gas. In this study, mercury removal efficiency was investigated for the Hybrid filter (HF) with activated carbon coating. HF was combined with bag filter and ESP in a single chamber and originally designed to remove fine particulates and retrofitted to the flue gas control device for mercury simultaneously. In lab-scale experiment, pressure drop and mercury removal efficiency were analyzed for determining optimal activated carbon coating rate at the operating facility conditions at a coal-fired power plant. After then, the activated carbon optimally coated filter was applied to the pilot-scale HF which was installed between ESP and FGD in a commercial coal-fired power plant. The effect of activated carbon filter in HF on mercury removal efficiency was investigated by measuring and comparing mercury concentrations between inlet and outlet of HF.

Black carbon sorbents, such as activated carbon and biochar, are used for mercury vapor capture for flue gas treatment and other applications. In our previous research, we developed easy-to-manufacture sulfurized activated carbon and biochar sorbents for Hg(0) vapor capture for low-resource end-users such as artisanal mining operations. These sulfurized carbons sorbents substantially reduced Hg vapors in gas streams relative to unsulfurized sorbents. However, the stability and bioavailability of the sorbed Hg, particularly in landfills and other waste disposal impoundments, has not been closely studied. In this research, we examine methylmercury (MeHg) production potential of spent carbon sorbents after incubating in sediment slurries. Five types of carbon sorbents (including sulfurized and unsulfurized activated carbon, biochar, and a commercially produced activated carbon for flue gas application) were each loaded with approximately 10 mg/kg of elemental Hg. The sorbents were added to anaerobic sediment slurries to a mass content of 5% relative to sediment dry mass. A replicate slurry received dissolved Hg as a control to simulate atmospheric deposition or highly reactive Hg. After a 5 day incubation at room temperature, net MeHg production was ten times greater in slurries amended with low-technology sulfurized sorbents as compared to unsulfurized activated carbon or biochar alone. Sulfurized sorbents leached significantly more Hg and sulfate than their non-sulfurized counterparts in desorption experiments performed in parallel to the sediment slurry experiments. Analysis of the sulfurized sorbents via X-ray spectroscopic methods revealed that the sorbents contained a mixture of sulfur species, including sulfate, thiosulfate and reduced sulfur (oxidation state equal to 0 or less). Sorbed Hg tended to co-locate with reduced sulfur on the surface of the carbon particles. The substantial presence of sulfate on the sulfurized sorbents likely contributed to Hg methylation by stimulating the growth of methylating sulfate reducing bacteria in the slurries. This research shows that spent sorbents for Hg vapor capture can have unintended consequences if they are inappropriately disposed.