CARBON DIOXIDE CAPTURE FROM FLUE GAS USING DRY REGENERABLE SORBENTS

Abstract

The objective of this project is to develop a simple, inexpensive process to separate CO{sub 2} as an essentially pure stream from a fossil fuel combustion system using a regenerable, sodium-based sorbent. The sorbent being used in this project is sodium carbonate which is converted to sodium bicarbonate, or ''baking soda,'' through reaction with carbon dioxide and water vapor. Sodium bicarbonate is regenerated to sodium carbonate when heated, producing a nearly pure CO{sub 2} stream after condensation of water vapor. This quarter, five cycle thermogravimetric tests were conducted at the Louisiana State University (LSU) with sodium bicarbonate Grade 3 (SBC{number_sign}3) which showed that carbonation activity declined slightly over 5 cycles following severe calcination conditions of 200 C in pure CO{sub 2}. Three different sets of calcination conditions were tested. Initial carbonation activity (as measured by extent of reaction in the first 25 minutes) was greatest subsequent to calcination at 120 C in He, slightly less subsequent to calcination in 80% CO{sub 2}/20% H{sub 2}O, and lowest subsequent to calcination in pure CO{sub 2} at 200 C. Differences in the extent of reaction after 150 minutes of carbonation, subsequent to calcination under the same conditions followed the same trend but were less significant. The differences between fractional carbonation under the three calcination conditions declined with increasing cycles. A preliminary fixed bed reactor test was also conducted at LSU. Following calcination, the sorbent removed approximately 19% of the CO{sub 2} in the simulated flue gas. CO{sub 2} evolved during subsequent calcination was consistent with an extent of carbonation of approximately 49%. Following successful testing of SBC{number_sign}3 sorbent at RTI reported in the last quarter, a two cycle fluidized bed reactor test was conducted with trona as the sorbent precursor, which was calcined to sodium carbonate. In the first carbonation cycle, CO{sub 2} removal rates declined from 20% to about 8% over the course of three hours. Following calcination, a second carbonation cycle was conducted, at a lower temperature with a lower water vapor content. CO{sub 2} removal and sorbent capacity utilization declined under these conditions. Modifications were made to the reactor to permit addition of extra water for testing in the next quarter. Thermodynamic analysis of the carbonation reaction suggested the importance of other phases, intermediate between sodium carbonate and sodium bicarbonate, and the potential for misapplication of thermodynamic data from the literature. An analysis of initial rate data from TGA experiments suggested that the data may fit a model controlled by the heat transfer from the sorbent particle surface to the bulk gas.