Roasting of Nickel Concentrates
- PDF / 984,908 Bytes
- 10 Pages / 593.972 x 792 pts Page_size
- 20 Downloads / 284 Views
INTRODUCTION
VALE Inco’s smelter in Thompson, Manitoba, Canada, and Xstrata Nickel’s smelter in Sudbury, Ontario, Canada, process nickel sulfide concentrates by fluid bed roasting, electric furnace smelting, and Peirce-Smith converting to produce a Bessemer matte. The Bessemer matte produced at Thompson is cast into nickel sulfide anodes and processed in the Thompson refinery, whereas the converter matte produced at Xstrata is granulated and shipped to the Kristiansand nickel refinery in Norway. Currently, nickel concentrate is mixed with a silica flux and fed to the roasters where it is partially oxidized. At Thompson, 40 pct of the sulfur in the concentrate is removed in the roasters, whereas 70 pct is removed at Xstrata.[1] As part of an effort to decrease SO2 emissions from the smelters, one option currently being examined involves increasing the amount of concentrate sulfur removed in the fluid bed roasters, which is known as the degree of roast. This would allow for a greater amount of sulfur to be oxidized to SO2 and captured in the acid plant. Increasing the degree of roast is known to lead to problems related to the formation of metal sulfates in the roaster and in the off-gas handling system.[2] These sulfates can build up on plates of the electrostatic precipitator and are difficult to remove. In addition, after reaching the calcine banks in the electric furnace, they may decompose as they are heated up beyond their range of thermodynamic stability, potentially releasing SO2. R. PANDHER, Graduate Student, and T. UTIGARD, Professor, are with the Department of Materials Science and Engineering, University of Toronto, 184 College St., Toronto, Ontario, Canada M5S 3E4. Contact e-mail: [email protected] Manuscript submitted December 14, 2009. Article published online May 4, 2010. 780—VOLUME 41B, AUGUST 2010
In an effort to increase the understanding of the sulfation of oxidized concentrates, concentrate samples were heated and oxidized in a thermogravimetric analysis (TGA) unit under various conditions. The results of these experiments are discussed in this article.
II.
BACKGROUND
Several investigators have studied the reactions that occur during the roasting of nickel concentrates at the laboratory scale. Thornhill and Pidgeon[3] studied the oxidation mechanism of individual grains of pentlandite, pyrrhotite, chalcopyrite, and other metal sulfides by placing individual grains in a single layer on stainless steel plates in a vertical tube furnace. By X-ray diffraction (XRD) and micrographic analysis of particles roasted at temperatures from 823 K (550 C) to 998 K (725 C) in 1 L/min air, they found that pyrrhotite (Fe1–xS) oxidizes to form an inner layer of magnetite and an outer layer of hematite. They also reported the formation of iron oxide columns in roasted particles as opposed to dense oxide layers. In the case of chalcopyrite (CuFeS2), their suggested mechanism was the initial formation of digenite (Cu9S5) by the selective oxidation of iron and sulfur, followed by diffusion of copper toward the center of
Data Loading...
