Brominated powdered activated carbon has been demonstrated to capture over 90 per cent of mercury emissions from coal fired plants, operating at high temperatures and capturing the metal in concrete-friendly, non-leaching forms to help tackle a rising global health threat.
Wolfgang Hardtke, Albemarle Europe, Germany
Mercury emissions are a global problem. Since the beginning of industrialization, more than 200 000 tonnes of mercury1 are estimated to have been released into the atmosphere from anthropogenic sources, with coal combustion contributing about half of current global emissions.
It is well known that mercury is toxic, persistent, and bio-accumulative. Mercury can travel around the globe and can cause severe health effects at very low concentrations. Studies by the United States Environmental Protection Agency found about 8 per cent of American women of child-bearing age have blood mercury levels sufficient to increase the risk of health defects of the unborn child.2
The desired method of mercury capture is one in which the mercury is fixed and not allowed to re-enter the environment. It has been discovered that brominated powdered activated carbon (PAC) can remove upwards of 90 per cent of the flue gas mercury and is effective in removing both elemental and oxidized mercury.
The value of a brominated sorbent was first demonstrated by Albemarle Mercury Control in a full-scale trial at a utility boiler firing subbituminous coal or a blend of subbituminous and bituminous coal. The mercury removal results from this trial conducted at the Detroit Edison St Clair Plant are presented in Figure 1. The results are compared with those of plain PAC.
Figure 1: Results from a mercury removal trial at the Detroit Edison St Clair plant showing the greater effectiveness of brominated powdered activated carbon (B-PAC) over plain powdered activated carbon (PAC)
Successful mercury removal requires effective sorbent distribution in the flue gas. And in systems with effective sorbent distribution, sorbent injection rates can be expressed in terms of pounds of sorbent injected per million actual cubic feet of flue gas treated or in milligrams of sorbent injected per Normal cubic meter of flue gas treated. Even at high injection rates, the plain PAC can only achieve a maximum of 60 per cent or 70 per cent of mercury removal, whereas the brominated PAC easily achieves greater than 90 per cent mercury removal with lower injection rates.
After years of extensive testing in the United States, permanent PAC injection systems are now being installed at power plants, cement kilns and other industrial applications to provide compliance with mercury reduction regulations. Furthermore the United Nations’ Environmental Programme is currently developing a globally binding instrument concerning anthropogenic mercury.
Why bromine for mercury abatement?
Bromine is highly effective for mercury control. As demonstrated in the example above, the beneficial and effective use of brominated PAC has resulted in increasing application of bromine and bromide compounds for mercury mitigation.
A small amount of bromine can oxidize elemental mercury (Hg0) to oxidized mercury (Hg+2), which is more easily removed from the combustion stack by adsorption onto fly ash or in a scrubber. Note that while oxidized mercury is soluble in water, elemental mercury is not.
How bromine or bromides can be used for mercury control is demonstrated in Figure 2. A bromide additive sprayed on coal before it is combusted generates bromine in the boiler to oxidize the mercury. The oxidized mercury is then suitable for capture on an activated carbon sorbent or in a scrubber. An additive such as HBr applied to the exit flue gas performs the same function. Brominated sorbents injected in several different locations will aid in mercury capture.
|Figure 2: Bromine or bromides can be used for mercury control in various ways, including in sprays on coal before combustion, in flue gas additives such as HBr, and in sorbents injected at several points in the generation process|
There are several challenging situations in which mercury sorbents must perform. These include applications where the concrete properties of the ash or dust must be preserved, high-temperature applications and use in applications other than boilers, such as cement kilns. In addition, the mercury once captured must not leach from the sorbent. Each of these issues will be addressed separately.
High temperature applications
The term “temperature insensitive” may have different meanings in different applications. In regard to mercury capture, the term means that the sorbent has the same capture rate for that pollutant over the temperature ranges of the application. A sorbent with this property is said to possess thermal stability. A sorbent with stable capture ability up to or beyond 350 °F (about 176°C) is required for most applications, since the flue gas in the particulate control routinely reaches this temperature.
It has been discovered that gas-phase brominated sorbents perform better with regard to temperature stability for mercury capture than either non-halogenated PAC sorbents or salt-impregnated bromide PAC sorbents. Salt-impregnated sorbents merely spray a bromide salt on the PAC while a gas-phase brominated sorbent is created by reacting bromine with PAC. The mercury capacity of a gas-phase brominated sorbent is nearly invariant up to a temperature of about 550 °F (about 288 °C).
The temperature insensitivity of a gas-phase brominated PAC, as it translates into stable mercury removal, has been demonstrated in numerous full-scale field trials. The results from one such trial are presented in Figure 3. Its data points are from a test conducted by URS Corporation at a Midwestern power plant,3 where the target mercury removal rate was 80 per cent. This was achieved at an injection rate of 1.6 lb/MMacf (about 26 mg/Nm3) in the case of the gas-phase brominated B-PAC (shown in blue in the figure) and the removal rate was stable with temperature. In contrast, the bromide salt-impregnated sorbent (shown in pink in the figure) did not achieve 80 per cent mercury removal, even at the higher injection rate of 1.8 lb/MMacf (about 29 mg/Nm3) and the removal rate declined rapidly with temperature.
|Figure 3: Results from a test by URS Corporation at a Midwestern power plant, where gas-phased brominated B-PAC (shown in blue) hit the 80 per cent mercury removal goal, while bromide salt-impregnated sorbent (in pink) failed|
“Concrete-friendly” describes the characteristic of a sorbent that has little or no impact upon the cementitious properties of the material into which it is mixed. Alone, a concrete-friendly sorbent will not be cementitious, but will not impact the properties of the material into which it is mixed.
One main requirement of a concrete-friendly sorbent is that it should not absorb the air entraining admixtures (AEA) that generate the air bubbles in concrete that provide the freeze/thaw capability of a concrete. Because the sorbent base is PAC, providing a sorbent that minimizes the capture of the AEA while capturing a high percentage of the mercury from the flue is not a simple feat.
Albemarle has developed a gas-phase brominated PAC-based concrete-friendly sorbent which has been widely tested in full-scale field trials in the utility industry and is now in commercial use.
The United States Department of Energy through the National Energy Technology Laboratory partly funded a trial4 that found that a mercury removal rate of over 80 per cent could be achieved with an injection rate of less than 5 lb/MMacf (about 80 mg/Nm3). Since the time of this trial, the properties of the gas-phase brominated PAC have been further improved and optimized to allow greater than 90 per cent mercury removal at injection rates as low as 2 lb/MMacf (about 32 mg/Nm3).
The mercury removal performance of the gas-phase brominated concrete-friendly PAC sorbent was demonstrated in these tests and in several commercial applications. But the question remained: “What impact did the sorbent have on the cementitious properties of the material in which it is mixed – in this case, fly ash?”
Several tests were used to evaluate the impact upon the fly ash’s cementitious properties. The simplest and fastest is a Foam Index (FI) test, under which a known amount of a sample is mixed into a known amount of water, with the exact amounts depending on the techniques selected. An air entrainment admixture (AEA) is then added drop-wise to the mixture. Finally, the sample is shaken to see if stable foam is created. If not, the procedure continues until the number of drops required to generate stable foam is determined.
A wide variety of AEA materials are now used to create the air bubbles in the final concrete that provide freeze/thaw capabilities. Plain PAC absorbs large quantities of AEAs making these sorbents inappropriate for use in concrete.5 A long-term test at the Midwestern plant found that the FI of baseline samples of fly ash were fairly low, but increased modestly when the concrete-friendly sorbent was introduced into the ash. These levels of FI are acceptable for use in concrete. It should be noted that the FI is very stable during sorbent injection since the FI of the sorbent is constant and the sorbent is injected on a flue gas flow basis (MMacf).
Many other tests were performed after the foam index test. However, the ultimate test is the concrete compressive strength. The compressive strength data for concrete samples made with no fly ash, baseline fly ash and fly ash containing the sorbent are presented in Figure 4.
|Figure 4: Concrete compressive strength data for samples made with no fly ash, baseline ash and fly ash containing concrete-friendly B-PAC sorbent found no significant difference between the three categories|
The compressive strengths were measured at 3, 7, 28, 56, and 91 days. The findings suggest that there was no significant difference in strength between the concretes that contained no ash, baseline fly ash or fly ash containing the sorbent. These results continue to be repeated in numerous tests and the gas-phase brominated concrete-friendly sorbent is now in commercial use.
Mercury leaching – the unintended release of mercury from a substrate – can be a consideration when choosing mercury mitigation strategies. Many studies have been performed on this subject and the basic finding is that the mercury will not leach out of a brominated PAC.6
Leaching data from one of the Albemarle full-scale trials using a gas-phase brominated sorbent (Figure 5) found that, although the concentration of mercury in the ash was high, the concentration of mercury in the leachate was below that of the blank. The explanation for this finding is that the sorbent continued to have mercury capacity and captured any mercury with which it cames into contact.
Figure 5: An Albemarle trial into mercury leaching from a gas-phase brominated solvent found that, while the concentration of mercury in the ash is high, the concentration in the leachate is below that of the blank
Fixing a global health threat
Bromine and bromide materials are highly effective in controlling emissions of mercury – a neurotoxin that poses a worldwide health issue. Brominated sorbents have shown an ability to capture more than 90 per cent of the mercury in a flue gas stream.
In addition, gas-phase brominated PACs have displayed an ability to operate at elevated temperatures. Bromide chemicals have also been used to oxidize elemental mercury so that the oxidized mercury formed can be more easily captured. A concrete-friendly brominated mercury sorbent – developed, tested and commercially available – allows for mercury capture with minimal impact of the cementitious properties of the dust in which it is captured.
Mercury captured on a brominated sorbent has further been found not to leach back into water, thus providing a permanent sink for this potentially harmful pollutant.
This article was jointly authored by Wolfgang Hardtke with Ron Landreth, Will Pickrell, Florian Kohl and Jiang Yong of Albemarle Mercury Control Division, Albemarle Corporation, Baton Rouge, Louisiana, USA.
1. Krabbenhoft, D., et al. “Atmospheric Mercury Deposition During the Last 270 Years”, Environmental Science & Technology, 36:11, 2002.
2. Schrober, S.E., et al. “Blood Levels in US Children and Women of Childbearing Age, 1999–2000”, Journal of the American Medical Association, 2003.
3. Richardson, et al., URS Corporation, “Evaluation of Novel Mercury Sorbents and Balance of Plant Impacts at Stanton Unit 1”, 2008 Mega Symposium, Baltimore, MD.
4. National Energy Technology Laboratory, “Brominated Sorbents for Small Cold-Side ESPs, Hot-Side ESPs and Fly Ash Use in Concrete”, Award DE-FC26-05NT42308 to Sorbent Technologies Corporation, September 16, 2005.
5. Feeley, T., “Overview of DOE/NETL’s Mercury and CUB R&D Program”, DOE/NETL’s Mercury Control R&D Program Review, Pittsburgh, PA July 2005.
6. Sanchez, F., et al., “Characterization of Mercury-Enriched Coal Combustion Residues from Electric Utilities Using Enhanced Sorbents for Mercury Control”, United States Environmental Protection Agency Report EPA/600/R-06/008, January 2006.