There is strong evidence that the combination of wet scrubbers and selective catalytic reduction can prove a viable and formidable combination for knocking out mercury.
By Constance Senior, PhD and Bradley Adams, PhD Reaction Engineering International
Many utilities have added or will add selective catalytic reduction (SCR) systems and wet flue gas desulfurization (FGD) scrubbers to their plants for NOx and SO2 emissions control, respectively. Having addressed these airborne pollutant emissions, utilities are now focusing on mercury (Hg) emissions control. Fortunately, the combination of an SCR and wet FGD scrubber can also provide significant mercury emissions reduction for utility boilers. In fact, under the right conditions, this combination has the potential to be an effective compliance option for mercury emissions under the Clean Air Mercury Rule. This article analyzes the capabilities and limitations of the SCR-FGD combination for mercury compliance, including applicability to different types of coals and issues with scrubber by-products.
Mercury Removal Across Wet FGDs
Removing mercury across scrubbers varies significantly from boiler to boiler, but this variation can be explained by considering the species of mercury that enter the scrubber. Wet scrubbers do a good job of removing oxidized mercury with removal efficiency greater than 75 percent in most of the wet scrubbers tested to date.
However, wet scrubbers do not remove elemental mercury. This is mainly because elemental mercury is not very water-soluble. Figure 1 shows data from the Environmental Protection Agency’s Information Collection Request (ICR) in which speciated mercury measurements were made at the inlet and outlet of wet FGDs.
Figure 1 shows high removal of oxidized mercury across wet FGDs, but also negative removal of elemental mercury in some cases. Although this negative removal may seem unusual, or even perhaps unrealistic, there is evidence that some wet scrubbers can actually emit more elemental mercury than what comes in.
The causes of this re-emission are not fully understood but may be due to chemical reactions between oxidized mercury and dissolved sulfur compounds in the scrubbing solution that reduce some absorbed oxidized mercury back to elemental form. This re-emission means that a wet scrubber is not as efficient as it could be in capturing elemental mercury. There are several companies investigating additives to the scrubbing solution that will reduce or eliminate this re-emission.
Mercury removal efficiency of wet FGD scrubbers depends largely on the fraction of oxidized mercury entering the scrubber. Figure 2 shows that as the fraction of oxidized mercury at the inlet to the scrubber increases, the amount of mercury removed across the scrubber also increases. Thus, a wet scrubber is most effective for mercury removal when the amount of oxidized mercury entering the scrubber is high.
Several factors affect mercury oxidation, including:
- Chlorine content of the coal: The chlorine content of the coal can explain much of the variation in the removal of mercury across scrubbers. As shown in Figure 3, coals with high levels of chlorine tend to have increased mercury removal across a scrubber. Previous work has shown that coals with less than 100 ppmw (parts per million by weight) of chlorine (Cl) have predominantly elemental mercury at the inlet of their respective pollution control devices (PCD) while coals with greater than 500 ppmw Cl have less than 20 percent elemental mercury.
- Unburned carbon in fly ash: The amount and chemical content of fly ash in the flue gas can affect mercury speciation. There is evidence that some fly ash catalyzes the oxidation of mercury. Unburned carbon is the primary oxidation catalyst, but iron oxides in ash may also catalyze the oxidation of mercury. The presence of acid gases such as HCl, NO and NO2 have also been shown to cause mercury oxidation in flue gases in the presence of fly ash.
- Temperature upstream of the particulate control device: Though not yet thoroughly understood, experimental work in the lab and at full scale has demonstrated the effect of time and temperature on the oxidation of elemental mercury. As it cools, elemental mercury is oxidized in the temperature range from 850 F down to 300 F. The characteristics of the air preheater, therefore, impact mercury oxidation.
Since mercury removal depends on the extent of oxidized mercury entering the scrubber, it is natural to wonder if there are ways to increase the amount of oxidized mercury in the flue gas upstream of a scrubber. Conveniently, SCRs have been observed to convert elemental mercury to oxidized mercury. SCRs use special alloy metal surfaces to catalyze the conversion of NO to N2 using ammonia. Testing has shown that these same catalysts effectively oxidize elemental mercury for some coals. This process appears to key off of the presence of HCl as the oxidant. Thus, coal chlorine content is one of the principal factors in the oxidation of mercury across SCRs.
There have been a number of campaigns recently to measure the speciation of mercury at the inlet and outlet of SCRs on coal-fired boilers. Figure 3 is derived from published measurement campaigns sponsored by DOE’s National Energy Technology Laboratory (NETL) and EPRI. It shows that if chlorine in the coal was greater than about 500 ppmw, more than 70 percent of the mercury leaving the SCR was in the oxidized form. The data in the figure are scattered, demonstrating that other SCR design and operating factors influence mercury oxidation. These factors include the space velocity and catalyst formulation, the catalyst temperature and the sulfur content of the fuel. In spite of these factors, it is clear that SCRs help increase the amount of oxidized mercury in the flue gas, leading to improved mercury removal across wet FGDs.
For plants burning sub-bituminous or lignite coals, the combination of an SCR and a scrubber will not increase mercury removal across the scrubber to the same extent expected with bituminous coals. This is due to the difference in mercury speciation among the different coals. Most of the difference in mercury speciation among the different ranks of coal can be attributed to two variables previously discussed:
- The level of unburned carbon in the fly ash
- The coal’s chlorine content
Bituminous coal flyash often has more unburned carbon than fly ash from low-rank coals, particularly on boilers with low-NOx combustion systems. Unburned carbon can absorb mercury (Hg0 and Hg2+) to form particulate-bound mercury. Unburned carbon can also act as a catalyst of oxidation and convert elemental mercury to oxidized mercury, if the level of chlorine compounds in the flue gas is sufficient.
The chlorine content of bituminous coals varies from several hundred to several thousand ppmw. Conversely, the chlorine content of low-rank coals has a much lower and more narrow range: from 5 to 100 ppmw. In both cases, chlorine compounds in the flue gas decrease the amount of gaseous elemental mercury present at the inlet to PCDs and increase the amount of oxidized mercury there.
When mercury is removed from the flue gas in a scrubber, it’s important to know where it goes and whether or not it poses an environmental risk. In many FGD systems, captured mercury ends up in the solid byproduct. Research sponsored by DOE-NETL and EPRI has demonstrated that mercury in the solid byproduct is not leachable to any large extent (using accepted test methods for leaching of heavy metals). Furthermore, the mercury in the solid byproduct remains there when heated to 300 F. This is significant because some types of scrubbers produce gypsum, which is used in the manufacture of wallboard. The manufacturing process contains a step in which the gypsum is heated to about 300 F, therefore any mercury present in the gypsum will not be released during wallboard manufacturing.
Due to the high level of interest in mercury control and the relatively limited understanding of mercury behavior in power plants, a number of research, development and demonstration programs are focused on improving the understanding of mercury emissions and control in plants with SCRs and wet FGDs. These programs include:
- Measurement campaigns to better characterize mercury speciation for different coals, boiler operating conditions, and SCR and FGD designs
- Investigation of additives that can be added to scrubbing solutions to eliminate re-emission of elemental mercury from wet FGDs
- Investigation of halogen-containing compounds that can be added to the fuel or the boiler in order to increase the amount of oxidized mercury in the flue gas. Low-temperature catalysts specific to mercury oxidation are also being field-tested
- Development of models for predicting mercury oxidation, reduction and capture across SCRs and FGDs. These models include empirical correlations, kinetic-based process models and computational fluid dynamic (CFD) models. One such kinetic model predicts mercury oxidation across SCR catalysts, based on analysis of laboratory and slipstream data on mercury. The model incorporates the effects of diffusion within the porous SCR catalyst and the competition between ammonia and mercury for active sites.
If successful, these new technologies could further improve the ability of wet scrubbers in combination with an SCR to remove mercury in coal-fired boilers.
There is ample evidence that SCRs in combination with wet FGD scrubbers can remove significant amounts of mercury. The combination of an SCR and an enhanced wet FGD (one that is equipped with elemental mercury re-emission control) has the potential for removal of more than 80 percent of the mercury across the scrubber. Parameters that can affect the extent of mercury removal across SCR-FGD systems include coal chlorine and sulfur content, unburned carbon in fly ash, and SCR design.
Utilities will benefit from understanding the potential contributions and limitations of SCR and wet FGD scrubbers on mercury behavior for their specific units. Specialized assessment models and analytical tools can help utilities navigate the complexities of mercury emissions and control by providing:
- Planning, oversight and review of measurement campaigns to characterize current mercury behavior at a plant to determine what mercury emissions a plant presently has;
- Mercury and chlorine content analysis and predictions for specific coal sources to determine if the plant can maintain compliance if fuels are changed;
- Assessment of current and future compliance status to determine if current equipment be tuned to improve mercury reduction;
- Evaluation of pollution control device impacts on plant mercury emissions to determine how planned PCD changes will impact current emissions;
- Screening and recommendation of appropriate mercury control technologies to determine what additional mercury control options are available, if needed;
- Predictive modeling of potential operating changes or technology additions to assess how well new technology will control emissions;
- Identification of monitoring options to ensure on-going compliance in order to determine how well the mercury control is working.
These types of assessments and analytical tools will allow utilities to devise technically sound and cost-effective solutions when formulating their mercury compliance strategies.
Brad Adams, PhD is president of Reaction Engineering International (REI), a Utah-based energy and environmental consulting firm. He has 15 years of experience in modeling and evaluating combustion systems performance and air pollution control technologies. Connie Senior, PhD, is manager of engineering R&D at REI. She has more than 20 years of power industry experience with coal combustion and formation of air pollutants and is an expert on mercury chemistry in combustion systems.