General Background on Activated Carbon Sorbents

Sorbent-based Mercury Control

Mercury in plant flue gas exists in 3 forms:
Elemental (gas phase) Oxidized (gas phase) Particulate (captured on particulate matter)

The basic concept of sorbent-based control is to enhance the conversion from gas phase to particulate phase and then remove the particulate bound mercury with the fly ash

Mercury and Carbon

Carbon is useful for capturing (adsorbing) gas phase mercury due to the many active sites on it surface. Therefore, to remove mercury from the flue gas the mercury must be brought into physical contact with an adsorbent (carbon) where it becomes attached and then collected in the plant particulate control system (ESP or FF).

Background to Carbon Capture

The initial DOE sponsored work looked at how well the carbon already in the fly ash (LOI) would work in capturing mercury. It was found that it had some positive effect, but that its ability to capture, and then hold, the mercury diminished with increasing temperature and became almost ineffective above 300F. It was also found that plants with low LOI (most PRB sites) obviously had very low resident capture rates as there was very little carbon in the gas stream.

Stepping up to Activated Carbon

The next logical extension of the DOE work was to look at injecting activated carbon to enhance the capacity of the resident carbon to a meaningful level. Activated carbon is, as is implied, carbon material that has been modified in such a way as to enhance the number of active sites available for combination with other compounds (mercury in this case). By controlling the amount of activated carbon to be injected, the level of mercury removal should be controlled.

Early Activated Carbon Results

Testimony of Dr. Steven A. Benson, Univ. of North Dakota Energy & Environment Research Center, to the U.S. Senate, Committee on Environment and Public Works, Subcommittee on Clean Air, Climate Change, and Nuclear Safety, June 5, 2003.

Activated Carbon Limitations

While introducing activated carbon to the flue gas improved the overall mercury collection efficiency, it was not a complete solution. It turned out that:
Activated carbon had a similar temperature threshold issue that made it generally ineffective above 350F, When used in PRB fueled systems it seemed to reach a capture threshold of roughly 70%, regardless of the amount injected. While only a small amount of activated carbon was used (generally < 12% by weight of the total ash) the enhanced activity of the carbon made the final ash unusable for concrete sales purposes.

The Effects of Halogen

Study of the activated carbon injection results revealed some interesting findings:
There seemed to be an interaction between the mercury and the halogens in the flue gas (specifically Chlorine) that converted the elemental mercury to oxidized mercury, making it easier for the carbon to capture, Because PRB fuel is low in chlorine, when the chlorine was fully utilized in oxidizing mercury, collection essentially stopped. Thus the plateau at roughly 50% collection.

Halogens lead to Bromine

Following the thread of halogen enhancement to activated carbon mercury collection, several elements were tried including chlorine, iodine, and bromine. What has been discovered to date is that bromine, when chemically bonded to the activated carbon delivers:
Much wider operating temperature capability. Current evaluations show effective collection up to 700F, When bromine is introduced with the activated carbon, the amount of resident chlorine become less important. Brominated activated carbons have shown mercury capture rates approaching 95% in PRB fuel systems. With special handling and bonding processes, the bromine can remain active for mercury collection yet the carbon can be passivated against interaction with concrete additives. This allows brominated activated carbon to be used in plants where fly ash sales are actively pursued.

2004-5: Brominated PAC (B-PAC) Results

Lignite SD/FF - B-PAC Lignite in-flight B-PAC PRB ESP - B-PAC Bituminous - B-PAC PRB ESP - B-PAC

Full-Scale - PRB & Lignite: Great River Energy Stanton 1 & 10 (EERC/URS) Brominated PAC - PRB ESP: Detroit Ed. St. Clair (Sorbent Technologies) in Injection Demos - Bituminous: Dukes Allen Plant (Apogee)

What is Toxecon?
When the effects of activated carbon on fly ash were recognized, one approach (patented by EPRI) was to install a new fabric filter baghouse, downstream of the existing particulate collection device. The activated carbon was then injected following the first collector but before the baghouse. This allowed the first device to remove the salable ash without the activated carbon, and then allowed the activated carbon to work in the secondary baghouse. The system is very effective at mercury removal without hurting fly ash sales, but does require the capital and operating cost penalty of a new fabric filter baghouse.

Sorbent Injection for Mercury Control

Implementation Considerations

Where is the carbon injected?

The location of the carbon injection nozzles depends on the configuration of the particulate collection device.
Cold-side ESP Fabric Filter Hot-side ESP Toxecon Ahead of the ESP & after the APH Ahead of the FF & after the APH Ahead of the ESP which is ahead of the APH After the exiting collection system & ahead of the new baghouse

Where is B-PAC injected?

Cold-side ESP and Fabric Filter Applications:

A P H

ESP/FF

Where is B-PAC injected?

Hot-side ESP Applications:

ESP/FF

A P H

Where is B-PAC injected?

Toxecon Applications:

A P H

ESP/FF

New FF

Extremely Low Equipment Capital Costs

$/kW SO2 Scrubbers NOx SCR ToxeconTM Baghouse PAC Injection alone $200 $120 $60 <$2

With PAC Injection alone: Almost no installation time needed Little trade labor needed No losses if scrubbers installed later Take advantage of future sorbents Costs are incurred only when operating

How much is used?

Brominated activated carbon (B-PAC) usage is traditionally defined in terms of pounds per million actual cubic feet per minute of gas flow. (#/MMACFM) Depending on fuel type, particulate collection device type, and desired mercury removal rates this amount can vary from 0.5#/MMACFM to in excess of 10#/MMACFM. For reference purposes, these rates generally result in a mass loading increase to the collection system of 0.5% to 2+%.

Mercury Sorbent Costs

Plain powdered activated carbon (PAC) ~ $0.50/lb e.g. Norits Darco Hg (formerly FGD) Brominated PAC e.g. Sorbent Technologies B-PACTM ~ $0.75/lb ~ $0.85/lb Norits Darco Hg-LHTM Non-carbon sorbents (still experimental)

Sorbent Technologies B-PAC Family

B-PAC for standard use H-PAC for Hot-side ESPs C-PAC for Concrete sales
All are plain PACs treated with a small amount of Bromine.
(Unlike Chlorine, HCl, & HF, bromine & HBr are not considered by the EPA to be air toxics. Unlike the others, HBr is not included in Toxic Release Inventory reporting.)

Bromine is the 3rd-most-common anion in the ocean. Seawater contains ~80 ppm dissolved bromine.