Field Test Program for Long-Term Operation of a COHPAC System for Removing Mercury from Coal-Fired Flue Gas

Bustard, C. Jean ; Lindsey, Charles ; Brignac, Paul

ADA-ES, Inc.

关键词: Removal;    Epri;    Particulates;    Polishing;    Activated Carbon;   

DOI  :  10.2172/887092
RP-ID  :  none
RP-ID  :  FC26-02NT41591
RP-ID  :  887092

美国|英语

来源: UNT Digital Library

【 摘 要 】

This document provides a summary of the full-scale demonstration efforts involved in the project ''Field Test Program for Long-Term Operation of a COHPAC{reg_sign} System for Removing Mercury from Coal-Fired Flue Gas''. The project took place at Alabama Power's Plant Gaston Unit 3 and involved the injection of sorbent between an existing particulate collector (hot-side electrostatic precipitators) and a COHPAC{reg_sign} fabric filter (baghouse) downstream. Although the COHPAC{reg_sign} baghouse was designed originally for polishing the flue gas, when activated carbon injection was added, the test was actually evaluating the EPRI TOXECON{reg_sign} configuration. The results from the baseline tests with no carbon injection showed that the cleaning frequency in the COHPAC{reg_sign} unit was much higher than expected, and was above the target maximum cleaning frequency of 1.5 pulses/bag/hour (p/b/h), which was used during the Phase I test in 2001. There were times when the baghouse was cleaning continuously at 4.4 p/b/h. In the 2001 tests, there was virtually no mercury removal at baseline conditions. In this second round of tests, mercury removal varied between 0 and 90%, and was dependent on inlet mass loading. There was a much higher amount of ash exiting the electrostatic precipitators (ESP), creating an inlet loading greater than the design conditions for the COHPAC{reg_sign} baghouse. Tests were performed to try to determine the cause of the high ash loading. The LOI of the ash in the 2001 baseline tests was 11%, while the second baseline tests showed an LOI of 17.4%. The LOI is an indication of the carbon content in the ash, which can affect the native mercury uptake, and can also adversely affect the performance of ESPs, allowing more ash particles to escape the unit. To overcome this, an injection scheme was implemented that balanced the need to decrease carbon injection during times when inlet loading to the baghouse was high and increase carbon injection when inlet loading and mercury removal were low. The resulting mercury removal varied between 50 and 98%, with an overall average of 85.6%, showing that the process was successful at removing high percentages of vapor-phase mercury even with a widely varying mass loading. In an effort to improve baghouse performance, high-permeability bags were tested. The new bags made a significant difference in the cleaning frequency of the baghouse. Before changing the bags, the baghouse was often in a continuous clean of 4.4 p/b/h, but with the new bags the cleaning frequency was very low, at less than 1 p/b/h. Alternative sorbent tests were also performed using these high-permeability bags. The results of these tests showed that most standard, high-quality activated carbon performed similarly at this site; low-cost sorbent and ash-based sorbents were not very effective at removing mercury; and chemically enhanced sorbents did not appear to offer any benefits over standard activated carbons at this site.