Petroleum Coke and Coal Start-Up Testing in Bubbling Fluidized Bed Combustors

Abstract

Bench-scale and 160 MWe demonstration tests were conducted for petroleum coke and high volatile bituminous coal blends. The bench-scale apparatus was a 100-mm-dia reactor located at the Canada Centre for Mineral and Energy Technology (CANMET), Energy Research Laboratories. The demonstration tests were conducted on the Tennessee Valley Authority’s (TVA) 160 MWe Shawnee Atmospheric Fluidized Bed Combustion (AFBC) Unit located at Paducah, Kentucky. Five and ten percent nominal volatile petroleum cokes were tested in the bench-scale unit. In addition, for the five-percent petroleum coke blends of 25, 50, and 75-percent petroleum coke, with the balance coal, were also examined at the bench scale. Eight start-up tests have been conducted with 50 percent blend of green delayed petroleum coke at the Shawnee AFBC unit. The bench-scale tests revealed that the volatile content in the petroleum coke was the primary factor affecting start-up. The tests showed that the volatile content from the coke and coal ignited at similar times; the char required longer to ignite. Bench-scale tests showed adequate start-up performance with blends up to 75 percent petroleum coke. Cold start-ups were conducted at the Shawnee AFBC Unit with 7 to 10 percent volatile green delayed petroleum coke. In all the start-ups, the operating temperature of 816°C was reached within 15 min of introducing the petroleum coke blend; this is similar to when high volatile bituminous coal was used. One start-up required a longer time because limestone had to be used to generate the bed. Local hot spots (982°C) were noticed in several start-ups for short periods, but subsided when additional air was supplied. Although more difficult to control, TVA routinely starts the Shawnee AFBC Unit with 50 percent shot petroleum coke and 50 percent high volatile bituminous coal.