1f: Quantifying mercury emissions from energy generation

The key point in the research of mercury emission control from new coal combustion type, such as coal oxy-combustion (O2/CO2), is full knowing of mercury speciation emission from coal combustion in homogeneous O2/CO2 atmosphere, especially definite understanding of the influence of high H2O concentration on mercury oxidization in homogeneous O2/CO2 atmosphere. Based on the established bench-scale experimental apparatus, it studied the influence of high H2O concentration on mercury oxidation in the simulated flue gas of homogeneous O2/CO2 atmosphere. The results showed that the high H2O concentration in simulated O2/CO2 atmosphere inhibited the oxidization of mercury through three ways. The high H2O concentration not only inhibited the generation of oxidizing Cl but enhanced the consumption of oxidizing Cl, and it inhibited directly the mercury oxidization in the simulated flue gas of homogeneous O2/CO2 atmosphere. The oxidation rate of elemental mercury (Hg0) with the addition of 5% water was smaller average 30% than that without additional water. The existence or addition of SO2 also inhibited the transformation of Hg0 to Hg2+. SO2 enhanced the consumption of oxidizing Cl to inhibit the mercury oxidization. Conversely, the inhibitory effect of high H2O on mercury oxidization was overcame at the extent by the addition of HCl in the simulated flue gas of homogeneous O2/CO2 atmosphere. The high HCl concentration promoted the conversion of HCl to oxidizing Cl to improve the mercury oxidization at last.

Sorbent trap monitoring systems have been successfully used in the USA to measure mercury at low levels mandated by the MATS (Mercury and Air Toxics Standards) and the NESHAP (National Emission Standards for Hazardous Air Pollutants). The mass of mercury collected on sorbent traps can vary in range across six orders of magnitude, depending on the Hg concentration in stack gas and sampling duration. During several days of sampling stack gas according to the monitoring procedure prescribed by EPA Performance Specification 12B, the collected mass of mercury may reach hundreds of micrograms. Most often, the sorbent traps are analyzed via atomic absorption spectroscopy (AAS) using a thermal technique (desorption or combustion), which does not require any chemical treatment. However, there is a risk to lose the analytical data if the mass of collected mercury exceeds the calibration range of the analytical instrument. Often while working with conventional gold-trap analyzers, the sample is homogenized and divided into several independently analyzed subsamples to avoid saturation of the analytical signal. Such an approach increases measurement error.

Here we present a novel approach to sorbent trap analysis which eliminates the risk of losing data while analyzing samples with unknown amounts of accumulated mercury. The technique is based on the commercially available Zeeman atomic absorption analyzer RA-915M. Direct analysis without intermediate mercury pre-concentration on a gold trap enables automated real-time control of the atomizer heating based on dynamics of the mercury released from a sample. Accumulated laboratory experience of sorbent trap analysis across a range of 1 1,000,000 ng Hg has shown excellent reliability of this rapid method for routine analysis of sorbent traps and other samples with unknown Hg content.

The re-emission of mercury (Hg), as a consequence of the formation and dissociation of the unstable complex HgSO3, is a problem encountered in flue gas desulphurization (FGD) treatment in coal-fired power plants. A model following a pseudo-second-order rate law for Hg2+ reduction was derived as a function of [SO3 2-], [H+] and temperature and fitted with experimentally obtained data to generate kinetic rate values of (0.120 ± 0.04, 0.847 ± 0.07, 1.35 ± 0.4) mM-1 for 40°, 60°, and 75°C, respectively. The reduction of Hg2+ increases with a temperature increase but shows an inverse relationship with proton concentration. Plotting the model-fitted kinetic rate constants yields ΔH = 61.7 ± 1.82 kJ mol-1, which is in good agreement with literature values for the formation of Hg0 by SO32. The model could be used to better understand the overall Hg2+ re-emission by SO32- happening in aquatic systems such as FGD wastewaters.

In every thermal combustion plant you can measure and detect mercury from the scrubber through the flue gas to the chimney up to the environment. As a standard treatment, activated carbon/coke is used to adsorb the mercury. The loaded adsorbent is disposed on a landfill. In terms of gas limit values, this system works well. But not for a sustainable, holistic treatment. Mercury leach out and forms by microbiology, organic mercury. Mercury becomes even more toxic. For this reason, mercury must be removed as insoluble HgS from this cycle. HgS is a natural mineral, called cinnabar. The only non-toxic mercury compound. Scrubber systems have an opportunity to buffer and store mercury salts (e.g. HgCl2/HgCl4) in high saturation. Under normal conditions the systems are running stable, with no risk of Hg0-gas blow out. But the concentration of dissolved mercury is the magnitude of influence for a Hg0-blow out event. Of course in an event of high input of reducing agent (e.g.SO2), the ionic(dissolved)Hg compounds transformed to elemental Hg. In this case a Hg-gas blow out is happened. In case of a mercury event, the mercury breaks through the scrubber and contaminates the subsequent purification stages. Hg deposits can be bleed out all the time and generate a baseline of Hg-emission, only a few g below the current limits. But the environment is continuously contaminated and poisoned with tons of Hg/year. The use of a special inorganic polymeric sulfur compound in an acidic scrubber or FGD plant produces only insoluble non-toxic HgS. HgS is chemically and thermally very stable. Therefore, HgS can be treated both in a classic sewage treatment and in a spray dryer. Here is the only point for Hg, where its allowed to leave the plant. Only as HgS. With this technology, it is made impossible to form an internal dissolved Hg cycle in the scrubber system. Unexpected events can be buffered. There are no Hg-blow outs possible. This is an effective way to separate mercury from a scrubber system. The special inorganic polymeric sulfur liquid (NETfloc SMF1) is easy to dose into any scrubber systems. Regular analyzes of the mercury balance (Hg dissolved/Hg insoluble) in the scrubber help to determine the optimum dosing level also for best cost efficiency. The German Umwelt Bundesamt (UBA) has tested HgS as the best choice for deposition of mercury compound in underground dumps.

The sorbent trap method (STM) is popularly used for Hg sampling and monitoring from coal power plant.

US EPA Method 30B is reference method for RATA (Relative Accuracy Test Audit) test for Hg CEM and sorbent trap monitoring system, and short-term snapshot of Hg emission. PS-12B is for long-term Hg monitoring method up to 10 days sampling.

Iodine treated activated carbons (I-ACs) is mostly employed for STM, but iodine causes matrix interference for wet analysis as well as it can be easily decomposed to iodine vapors (I2) for thermal analysis that results in quick destruction of catalyst/gold amalgam, which restricts various applications of Hg analysis device and method for STM.

The iodine contaminated Hg analyzer may lead analytical bias issues such as discrepancy of elemental Hg spiked and liquid Hg standard sample, low recovery of Hg, and poor calibration, which result in invalidation of field sampling and significant failure of QA/QC criteria.

Newly developed brominated ACs (Br-ACs) sorbent media showed enough Hg adsorption capacity as much as I-ACs and observed good thermal stability at desorption analysis temperature (650 oC).

In this study, a successful calibration could be achieved with the alternative thermal desorption Hg analysis system by using new Br-ACs.

The new Br-ACs sorbent can be consider as a replacement of I-ACs, which allows various type of Hg analysis devices (either catalyst or non-catalyst type) and analysis methods (e.g. acid digestion) for STM.