(608d) ON-Site PRODUCTION of Activated CARBON for Removal of Mercury FROM Coal Flue Gas | AIChE

(608d) ON-Site PRODUCTION of Activated CARBON for Removal of Mercury FROM Coal Flue Gas

Authors 

Rostam-Abadi, M. - Presenter, University of Illinois at Urbana-Champaign

Many coal-fired power plants are considering activated carbon injection (ACI) as an option for controlling mercury emissions. A major operating cost of the ACI is the cost of the activated carbon (AC), estimated to be in the range of $0.75 to $1.5/lb.  For a 500 MW base-loaded plant with a 90% capacity factor, this would result in an annual cost in the range of $0.5 to 5 million/year. If AC could be produced on site for less than half the cost of the purchased commercially available ACs, a 500 MW plant could save $0.25 to 2.5 million/year, making ACI for mercury compliance a more cost-effective technology.

Electric Power Research Institute (EPRI) and the University of Illinois at Urbana-Champaign (UIUC) have developed a patented technology (U.S. Patents 6,451,094 and 6,558,454 B1 issued in 2002 and 2003, respectively) for on-site production of AC to help reduce the operating cost of AC injection processes for mercury control. This technology, Sorbent Activation Process (SAP), will help reduce the operating cost of activated carbon injection processes for removal of mercury emissions from coal combustion flue gas.  SAP involves processing coal to AC in an entrained-bed reactor (Figure 1). The feedstock coal is the same coal as that burned by the utility.  SAP can be used to prepare ACs with various surface areas, pore structures, and surface chemistry (halogenated AC) from bituminous, subbitumious, and lignite coals.  The product gas from SAP unit containing the AC is then directly injected into the flue gas duct upstream of a particulate control device, such as an electrostatic precipitator (ESP), or downstream of an ESP and collected in a baghouse (not shown) to avoid affecting fly ash properties. Alternatively, although not shown in Figure 1, a cyclone could be use to separate the AC from activation product gas before injection to the flue gas duct. The product gas from the activation unit is then recycled to the boiler.

 

 
Figure 1. On-site activated carbon (AC) production process.

A study was initiated to evaluate the proof-of-concept of the SAP using both a bench-scale and a pilot-scale unit. The bench SAP, Figure 2, was used to produce 1 to 5 pounds AC per hour from different types of coals.  More than 50 tests were performed with the bench SAP using an Illinois coal, a Powder River Basin (PRB) coal, and a lignite coal. The purpose of these tests was to compare quality of the AC products, as measured by their surface area and mercury adsorption capacity, and to obtain process scale up data. Equilibrium mercury capacities of the ACs were measured at URS Corporation (Austin, TX) using a fixed-bed reactor. The relationship between the mercury capacity and BET surface area of bench SAP ACs, with both values normalized with the carbon content in the AC (Hg/C and BET/C), revealed that: 1) PRB-derived ACs had higher surface areas and higher mercury adsorption capacities than their Illinois coal- and lignite-derived ACs, 2) mercury adsorption capacity increases linearly with carbon surface area in the range of 100 to 600 m2/g-C, 3) data for Illinois coal-, PRB-, and lignite-derived carbons follow a linear plot indicating that mercury adsorption performances of activated carbons derived from the three coals are comparable at a given value of surface area, and 4) capacities of SAP ACs were >80% of the capacity of Darco Hg, one of the most commonly used commercial AC for mercury removal.

Figure 2. The bench scale SAP reactor (front view and back view).

Carbon monoxide (CO) gas is generated during pyrolysis and oxidation of coal as well as gasification of coal char with steam under SAP process conditions. A CO concentration >15 vol% was observed in SAP flue gas in some of the bench tests. It is very likely that the concentration of CO in the SAP flue gas must be reduced in order to meet the required power plant's air permit limits on CO.  It should be noted that the volume of the SAP flue gas generated for an injection rate of 5 lb AC/Macf is about 500 times lower than the volume of flue gas of a 500 MW boiler.  Therefore, a 500/1 volume dilution will be achieved when SAP flue gas containing AC particles is injected into the boiler's flue gas.

A theoretical kinetic study was initiated to explore the conditions under which CO in the SAP flue gas could be oxidized without burning the AC particles. The results from this study revealed that at a certain temperature range, CO in SAP flue gas can be oxidized to CO2 with a very small amount of carbon being oxidized. Bench-scale tests were performed to investigate the competitive oxidation of CO and coal char when carbon particles existed in the CO containing flue gas, Figure 3. In the presence of the carbon particles, an immediate increase in CO concentration (>4 vol%) in the flue gas was observed, indicating that the CO was produced by either the thermal release of oxygen functional groups from the carbon surface and/or partial oxidation of the carbon particles. Air injection at a location downstream of the coal injection port reduced the CO concentration in the flue gas from about 3.8% to below 1%. The CO concentration decreased with increasing the air injection rate. The results from these tests show the effectiveness of air addition as an option to reduce CO concentration in SAP flue gas.

Figure  3. CO concentration profiles with air and carbon/coal injection (O2 measurement was drifted to ~0.9% during these tests).

The process engineering data collected during the bench SAP tests were used to design and fabricate a pilot SAP unit capable of producing 50 pounds AC per hour.  Pilot SAP was integrated with a coal-fired boiler rated at approximately 325,000 acfm (half of 220 MW) at a power plant. Fourteen days of start up and shake down tests have been performed to date to evaluate the performance of the pilot SAP, which is the first-of-its-kind in the world, and its deviation from the design specifications. 

A cost analysis study was performed for a 600 lb/hr AC production scale. This amount of AC is equivalent to injecting 5 lb/Macf into the flue gas generated from a 500 MW (net) boiler burning an Illinois coal. The results from the cost study revealed that, for an Illinois coal-to-AC SAP plant, the total capital requirement is estimated $1.2M, and the annual O&M cost amounts to about $1.0M. The cost of AC production and injection using the SAP unit at the 600 lb/hr scale is estimated at $0.28/lb carbon produced.