Fate of Mercury in Synthetic Gypsum Used for Wallboard Production

Abstract

This report presents and discusses results from the project 'Fate of Mercury in Synthetic Gypsum Used for Wallboard Production', performed at five different full-scale commercial wallboard plants. Synthetic gypsum produced by wet flue gas desulfurization (FGD) systems on coal-fired power plants is commonly used in the manufacture of wallboard. This practice has long benefited the environment by recycling the FGD gypsum byproduct, which is becoming available in increasing quantities, decreasing the need to landfill this material, and increasing the sustainable design of the wallboard product. However, new concerns have arisen as recent mercury control strategies involve the capture of mercury in FGD systems. The objective of this study has been to determine whether any mercury is released into the atmosphere at wallboard manufacturing plants when the synthetic gypsum material is used as a feedstock for wallboard production. The project has been co-funded by the U.S. DOE National Energy Technology Laboratory (Cooperative Agreement DE-FC26-04NT42080), USG Corporation, and EPRI. USG Corporation is the prime contractor, and URS Group is a subcontractor. The project scope included seven discrete tasks, each including a test conducted at various USG wallboard plants using synthetic gypsum from different wet FGD systems. The project was originally composed ofmore » five tasks, which were to include (1) a base-case test, then variations representing differing power plant: (2) emissions control configurations, (3) treatment of fine gypsum particles, (4) coal types, and (5) FGD reagent types. However, Task 5,could not be conducted as planned and instead was conducted at conditions similar to Task 3. Subsequently an opportunity arose to test gypsum produced from the Task 5 FGD system, but with an additive expected to impact the stability of mercury, so Task 6 was added to the project. Finally, Task 7 was added to evaluate synthetic gypsum produced at a power plant from an additional coal type. In the project, process stacks in the wallboard plant were sampled using the Ontario Hydro method. In every task, the stack locations sampled included a gypsum dryer and a gypsum calciner. In Tasks 1 and 4 through 7, the stack of the dryer for the wet wallboard product was also tested. Also at each site, in-stream process samples were collected and analyzed for mercury concentration before and after each significant step in wallboard production. These results and process data were used to construct mercury mass balances across the wallboard plants. The results from the project showed a wide range of percentage mercury losses from the synthetic gypsum feedstocks as measured by the Ontario Hydro method at the process stacks, ranging from 2% to 55% of the mercury in the gypsum feedstock. For the tasks exceeding 10% mercury loss across the wallboard plant, most of the loss occurred across the gypsum calciner. When total wallboard emissions remained below 10%, the primary emission location varied with a much less pronounced difference in emission between the gypsum dryer, calciner and board dryer. For all seven tasks, the majority of the mercury emissions were measured to be in the elemental form (Hg{sup 0}). Overall, the measured mercury loss mass rates ranged from 0.01 to 0.17 grams of mercury per dry ton of synthetic gypsum processed, or 0.01 to 0.4 pounds of mercury released per million square feet of wallboard produced from synthetic gypsum. The Coal Combustion Product Production and Use Survey from the American Coal Ash Association (ACAA) indicate that 7,579,187 short tons of synthetic gypsum were used for wallboard production in 2006. Extrapolating the results of this study to the ACAA industry usage rate, we estimate that mercury releases from wallboard production plants in 2006 ranged between 150 to 3000 pounds for the entire U.S. wallboard industry. With only seven sets of wallboard plant measurements, it is difficult to draw firm conclusions about what variables impact the mercury loss percentages across the wallboard plants. One significant observation from this study was that higher purge rates of chlorides and fine solid particles from the wet FGD systems appear to produce gypsum with lower mercury concentrations. Any chemical interaction between mercury and chlorides is not well understood; however, based on the information available the lower mercury content in the gypsum product is likely due to the blow down of fine, mercury-rich particles as opposed to a decreased chloride concentration. One possible explanation is that a decrease of fine particles in the FGD slurry allows for less adsorption of mercury onto those particles, thus the mercury remains with the FGD liquor rather than the gypsum product. A more detailed discussion on synthetic gypsum sources and FGD chemistry data can be found in the Experimental section of this report and Table 4.« less