PREPARATION AND EVALUATION OF MODIFIED LIME AND SILICA-LIME
SORBENTS FOR MERCURY VAPOR EMISSIONS CONTROL
S. Behrooz Ghorishi
and
Carl F. Singer
ARCADIS Geraghty and Miller Inc.
Post Office Box 13109
Research Triangle Park, North Carolina 27709
and
Charles B. Sedman
National Risk Management Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
Abstract
Previous work has shown that mercury chloride vapor is readily absorbed by calcium-based
sorbents as an acid gas in environments typical of coal-fired boiler flue gas, while elemental
mercury vapor is absorbed by calcium-bascd sorbents only when sulfur oxides are also present.
Current efforts are attempting to improve uptake of mercury species by increasing active sites
and adding oxidative species to the sorbent. Preparation of modified lime and silica-lime
sorbents and their behavior toward mercury species are compared to those of commercially
available lime on a fixed-bed, bench reactor. The implications of findings toward development
of multipollutant control technologies and planned field pilot evaluations of more promising
multipollutant control concepts are discussed.

-------
Introduction
Title HI of the 1990 Clean Air Act Amendments (CAAA) requires the U.S. Environmental
Protection Agency (EPA) to submit a study on 189 hazardous air pollutants (HAPs). This study
would include emissions and a risk (to public health) assessment of the 189 HAPs. Of these
pollutants, mercury has drawn special attention due to its increased levels in the environment
and the well documented food chain transport and bioaccumulation of this element and its
compounds such as methyl mercury [1,2]. An EPA report to Congress cites the largest emitters
of mercury as coal-fired utilities, medical waste incinerators (MWIs), municipal waste
combustors (MWCs), chlor-alkali plants, copper and lead smelters, and cement manufacturers
[3]. These sources are estimated to account for over 90% of all anthropogenic mercury
emissions. Utility boilers account for nearly 25% of the total anthropogenic emissions, of which
more than 90% are attributed to coal-fired utility boilers.
Mercury, a trace constituent of coal and waste, is readily volatilized during combustion
processes. Mercury is the most volatile trace metal, and major portions of it can pass through
existing particulate matter (PM) control devices [4]. A sorbent reacting with this metallic vapor
can effectively convert it to a sorbed liquid or solid phase, facilitating its removal with sorbent
particles in a PM control device. Emission control processes which use dry sorbents do not pose
the problem of the treatment and stabilization of a waste liquid stream and, therefore, seem very
attractive for coal combustors and waste incinerators.
Several methods of controlling mercury emissions are in either commercial use or development
for MWCs and MWIs [5], Dry sorbent injection (DSI) of activated carbon, followed by fabric
filtration (FF), has shown high (>90%) mercury removal in MWC applications. Spray drying
(SD) followed by FF, and wet scrubbing (WS) have been both successfully applied for acid gas
control, and found to remove substantial (60-90%) amounts of mercuiy in MWCs. However, all
three technologies have been less successful in removing mercury from coal-fired flue gases [6].
Previous investigations conducted in EPA laboratories have indicated that mercury control
strategies are dependent upon the mercury species that exist in the coal/waste combustion flue
gases [7-9], These studies have shown the relative ease of controlling oxidized mercury,
specifically mercuric chloride (HgCI2), as opposed to elemental mercury (Fig0) in fixed-bed
applications. Reagent grade hydrated lime exhibited as high as 85% removal of HgCl2 sustained
for at least 24 hours [9]. The same sorbent under identical conditions showed no Hg° capture,
indicating the degree of difficulty in the capture and control of Hg°. Since Hg° is expected to be
present in coal combustion flue gases, it is therefore not surprising that the MWC mercury
control technologies have been less successful in removing mercury from coal-fired flue gases.
Furthermore the operators of some waste incinerators also need to be concerned about Hg°
emission control. While the mercury in most incinerator streams is predominantly HgCl2, this is
not always the case. Prior compliance measurements from the Ft. Dix municipal Waste-To-
Energy Plant in New Jersey, for example, indicated a high degree of Hg° in flue gases [10].
Thus, the focus of this study was to optimize the development of novel calcium (Ca)-based
sorbents so that maximum capture of Hg° could be achieved. As discussed previously, HgCl2 can
be controlled using any alkaline sorbent such as unmodified reagent grade lime [7-9].
2

-------
Development of novel sorbents for the control of mercury is the subject of a number of current
research programs [11-14]. Our previous studies [8,15,16] have shown that a potential method of
cost reduction in controlling mercury emissions in coal-fired utilities (applicable to waste
incinerators as well) would be to utilize cheaper Ca-based sorbents. These results also showed
that further modifications and improvements in these Ca-based sorbents are needed in order to
augment their Hg° sorption capabilities.
Summary of Previous Studies
Two novel Ca-based sorbents and a reagent grade hydrated lime were evaluated with regard to
their Hg° capture capabilities in our previous investigations (fixed-bed applications). Hydrated
lime with a total surface area of 13 m2/g exhibited no Iig° capture. The two novel Ca-based
sorbents exhibited relatively good Hg° capture in the presence of sulfur dioxide (S02), removing
35 and 18% of gas-phase Hg°, respectively, during a 30-minute test. Under identical conditions,
a commercially available activated carbon (Darco FGD, Norit Americas Inc.) removed about
70% of inlet Hg°. It was hypothesized that activated carbon, was effective (as compared to Ca-
based sorbents) in adsorption of Hg° in large part due to its large surface area (about 550 m2/g)
and large population of small pores as was illustrated in its pore size distribution results [8,15].
The two novel Ca-based sorbents that exhibited some reactivity toward adsorption of Hg° were
also found to have significantly higher numbers of small pores and relatively high surface areas
(compared to hydrated lime) [8,15]. This previous work was conducted using a different
approach than the work reported herein, and results are not directly comparable.
From this previous work, the conclusion was that Hg° capture requires small pores (and
subsequently higher surface area) in the Ca-based sorbents normally used for S02 control. It
should be noted that the two novel Ca-based sorbents exhibited measurable Hg° capture only in
the presence of S02 in the gas phase. It was concluded that S02 reacts with Ca-based sorbents
and creates active sites inside the fine pore structure and facilitates the capture of Hg° molecules.
It was determined that the extent of Hg° capture by modified Ca-based sorbents, upon further
improvements, could provide an economic alternative to commercially available activated
carbons.
As reported at the 1997 Mega Symposium, subsequent testing of two higher-surface area (and
pore volume) calcium silicates exhibited high affinities for Hg° [16] Sorbent 43 had a surface
area of 205 m2/g, compared to 74.5 m2/g for sorbent 40. For sorbent 43, Hg° uptake continued
beyond 200 minutes of exposure. On the other hand and despite a higher initial rate of uptake,
sorbent 40 showed a relatively small uptake capacity, reached after 30-40 minutes of exposure
to the Hg°-laden simulated flue gas. Hg° sorption directly correlated with the total surface area of
the two sorbents.
It has been hypothesized that fine pores (< 100A) are instrumental in capturing Hg° The number
or volume of pores smaller than 100A was noticeably higher in sorbent 43. Activated carbon,
FGD, was also tested under similar conditions, and continued to remove Hg° beyond 200
minutes. Capacities of Sorbents 40,43, and FGD after 200 minutes of exposure are plotted in
Figure 1. The initial rate of Hg° uptake (in jag Hg°/g sorbent-min) for silicate sorbents in
3

-------
comparison to FGD is shown in Figure 2. Unlike slaked limes, modified silicates did not exhibit
the same trends as capacities (Figure 1). Sorbent 43, despite a far superior Hg° capacity, showed
a lower initial rate of uptake. At that time the reasons for these observations were not
understood, or proposed. As expected, activated carbon exhibited a better initial rate of Hg°
uptake.
Present Work
Since it has been established that Hg° is more difficult than HgCl2 to control, and that Iig°
control by alkaline sorbents requires fine pore structure and an S02 presence, this paper focuses
on developing calcium-based sorbents for Hg° control. The key assumptions are that (a)
alkaline sorbents will sorb HgCl2, and (b) alkaline sorbents, having both fine pore structure and
and an oxidizing species in the pore structure, can oxidize and sequester Hg° from flue gas.
Apparatus and Methodology
Bench-scale Hg° removal tests were performed on the vertical fixed-bed reactor apparatus
illustrated in Figure 3. Operation and construction details of this apparatus have been previously
described[8,15]. A simulated flue gas was generated containing 40 ppb Hg°, 4% oxygen, 10%
carbon dioxide (C02), and 1 % water vapor. Although some tests were performed with no S02
present, the majority of tests were on simulated flue gas containing 500 ppm S02. Simulated
flue gas was then passed through the sorbent bed, a Lindberg furnace, a NAFION™ diyer, and to
serial Hg° and S02 analyzers. A total of 250 mg of sorbent was exposed to 300 cm3/min
(dry@STP) simulated flue gas for 2 hours. Sorbents were tested at either 70 or 100°C reactor
temperature. The Lindberg furnace was maintained at 100°C to prevent condensation and avoid
undesirable side reactions. Breakthrough curves from the Hg° and S02 analyzers were evaluated
to determine initial and total Hg° uptake rates; initial removal rates were calculated as the
average removal rate during the first 5 minutes of sorbent exposure to the simulated flue gas.
Hydrated lime sorbents were prepared from commercial powdered quicklime (Marblehead Lime
Co.) in a Parr reactor. The solution used for hydration was prepared from 30 % oxidant solution
and deionized water as required. Quicklime was added to the reactor at ambient temperature
prior to assembly. The hydrating solution at ambient temperature was injected into the reactor,
and the reactor was sealed. Hydrations were performed with a substoichiometric amount of
solution, estimated at 0.85, to ensure that liquid water was eliminated from the sorbent in an
attempt to minimize oxidant decomposition. Thirty minutes after hydrating solution injection,
the reactor was vented and sorbent was removed from the reactor. Due in part to the large
thermal mass of the reactor, reaction temperatures did not exceed 100°C. No further drying was
performed prior to testing. Sorbents were tested on the mercury apparatus within 48 hours of
preparation.
The physical properties of the hydrated lime sorbents are shown in Table 1 with respective test
conditions. The surface area and pore diameter of hydrated lime prepared with 6 and 30 %
4

-------
oxidant were similar. Thermogravimetric Analysis (TGA) coupled with Residual Gas Analysis
(RGA) confirmed the presence of oxidant on the solid sorbent in the form of a low
decomposition temperature oxygen peak. Each sorbent preparation was tested once in the
mercury test stand.
Table 1. Modified Lime Hydrates — Preparation, Physical Properties, and Test Conditions
Ilydrating
BET
Average Pore
SOz Concentration,
Reactor
Solution,
Surface
Diameter, A
ppmv
Temperature,
% Ox
Area, m2/g


°C
30
15.49
204.3
500
70
6
14.61
188.1
0
100
30
15.64
182.9
0
100
30
16.09
199.4
500
100
6
15.33
195.2
500
100
6
14.02
188.4
0
70
30
15.88
199.0
0
70
6
13.55
185.0
500
70
Calcium silicate sorbents were prepared in a Parr reactor from reagent grade hydrated lime and
ground amber post-consumer glass. An inorganic additive was dissolved in de-ionized water
prior to addition of hydrated lime or ground glass. Glass, hydrated lime, and water were added
to the Parr reactor at room temperature, and the reactor was sealed. Reagents were heated to
150°C and maintained at this temperature for 1 hour. Pressure was then released from the reactor
and the slurry was vacuum filtered through Whatman #42 paper. The filter cake was dried
overnight in a vacuum oven at 100°C and was removed and allowed to cool. A freshly regenerated
molecular sieve scrubber was installed on the inlet air side of the vacuum oven each day to
minimize C02 contamination. Dried sorbents were ground with a mortar and pestle. Recovered
solids were then slurried at ambient temperature by adding solids to a wetting solution. The wetting
solution was either de-ionized water or 30 % aqueous oxidant used in previous lime hydrations.
After mixing for 5 minutes, the slurry was vacuum filtered through Whatman U42 paper and the
filter cake was dried overnight in a vacuum oven at 100°C. Sorbents were tested for on the mercury
test stand within 48 hours of preparation.
Sorbents were tested in the same manner as the hydrated limes on the bench-scale apparatus. All
were evaluated with 500 ppmv SG2 and at a 100°C reactor temperature While each preparation
was tested once on the mercury test apparatus, each preparation was replicated. The average
physical properties of the calcium silicate sorbents are summarized in Table 2. The additive
significantly enhanced the surface area of the silicate sorbent. In addition, wetting the sorbent in
aqueous oxidant appears to both increase the surface area and decrease the average pore diameter
compared to wetting the sorbent in water.
5

-------
Table 2. Silicate Sorbent Preparation and Physical Parameters
Silicate Preparation
Wetting Solution
Average BET
Surface Area,
m2/g
Average Pore
Diameter, A
3:1 glass/Ca(OH)? with additive
water
167.8
164.8
1:1 glass/Ca(OH)? with additive
water
130.0
231.2
1:1 glass/Ca(OH)? no additive
water
50.48
233.7
1:1 £lass/Ca(OH)? no additive
30 % Oxidant
76.51
176.4
1:1 elass/Ca(OH|> with additive
30 % Oxidant
156.8
202.6
Results
Figure 4 illustrates the effects of oxidant-enriched lime on the initial rate of mercury removal.
With 500 ppmv of S02 present in the simulated flue gas, calcium oxide (CaO) with 6% oxidant
hydration removes Hg° at an initial rate of 0.06 ng/g-min at 70°C and 0.13 |.ig/g-min at 100"C.
Without SO? present, the effects of oxidant hydration arc more pronounced, and initial mercury
uptake rates arc higher, indicating an interference between the oxidant and S02. Also the impact
of adding the more concentrated oxidant solution, 30 vs 6%, is to further increase the rate of
uptake in the absence of S02. With S02 present, the effect of increased oxidant concentration
above 6% is negligible. For comparison, lime hydrated with water only has an initial Hg° uptake
of less than 0.01 jig/g-min.
Figure 5 illustrates the effects of oxidant hydration upon long-term (2 hour) Hg° capacities of
lime. With 500 ppmv S02 present, the Hg° capacities average about 2.5 jag/g sorbent for
oxidant hydration, compared to a base hydrated lime (no oxidant) Hg° uptake of about 1.0 ng/g
sorbent. In the presence of 500 ppmv S02, the strength of oxidant solution is not important.
With no S02 present, the capacities are increased several fold over that with S02 present; the
strength of oxidant solution is also important, with 30% solution nearly doubling (15.5 vs 8.8
jig/g) the capacity over 6% solution at 70°C and increasing capacity over threefold (11.7 vs 3.1
jig/g) at 100°C. ~
Figure 6 illustrates the effects upon S02 sorption capacity by oxidant hydration, and shows
slight, but perceptible, improvement with increasing oxidant addition, more so at 70 than 100°C.
Figure 7 illustrates the impact of inorganic additive and oxidant on two sorbents prepared from
brown glass and lime at 1:1 and 3:1 glass-to-lime weight ratios. The effects of inorganic additive
(added to increase porosity) and oxidant (added to increase the rate of mercury oxidation) are
similar, resulting in nearly 100% increases in the initial IIg° uptake rate, when added separately. The
effect is confounded when both are added, since the apparent improvement is the same as adding
either separately. Limited data suggest that the initial uptake rate may be intrinsically higher for
1:1 glass-lime sorbent than for 3:1. For comparable flue gas temperatures and conditions (500 ppmv
6

-------
S02 and 100°C), the initial Hg° uptake rates for unmodified silicates are comparable (-0.12-0.13
Hg/g-min) to that of oxidant hydrated lime and the modified silicates nearly twice (-0.22-0.23 |ig/g-
min) that of oxidant hydrated lime.
Similar impacts are shown in Figure 8 for the total capacity of Hg° uptake, where about 75%
increases in capacity are noted with either additive, but again no synergism between additives is
evident. However, the Hg° capacity of one 3:1 glass:lime sorbent with inorganic additive is greater
(6.0 vs 4.5 |ig/g) than for the comparablel: 1 sorbent, despite having a lower initial uptake rate.
Further, the base (no additives) sorbent at 1:1 glass:lime showed Hg° capacities of 2.5 fig ig sorbent,
while the enhanced glass-lime sorbents averaged about 4.4 ^g /g sorbent. These may be compared
to similar tests of lime sorbents, where Hg° capacities of 2.5 jig/g sorbent for oxidant-enriched lime
were observed, and 1.0 |ag/g sorbent for base lime hydrate
Figure 9 illustrates the impact of additives on S02 uptake by glass-lime silicates; inorganic additive
slightly improves S02 uptake capacity, while little effect of oxidant addition is observed. The
combined inorganic additive and oxidant effects appear to lower S02 capacity significantly, but the
mechanism is not yet understood.
Discussion of Results
An explanation of the above results is that the oxidant hydration forms a solid calcium oxidant,
CaOx, which dissociates to oxidize Hg°. The oxidized Hg vapor is sorbed onto the alkaline sorbent
matrix. The S02 present also will promote Hg° oxidation by reacting with the lime to form solid
calcium sulfite (CaSO^or calcium sulfate (CaSO,,) but, in doing so, blocks the pores, restricting
further Hg° diffusion into the reactive zone. The proposed reactions are:
R>0 + OX + 2CaO -> CaOOX + Ca(OH);
(1)
2Hg° + CaO OX -> Hg/J + CaO
(2)
The overall reaction for (1) and (2) is:
H20 + OX + 2Hg° + CaO -> Hg?0 + Ca(OH):
(3)
S02 interferes with reaction (2) by consuming the solid oxidant phase :
CaOOX + S02 >CaS04
(4)
S02 promotes a similar oxidation of Hg°:
7

-------
Ca(OH)2 + S02 + 02 -> CaS04 +H2Q +0
(5)
2Hg°-f 0 ->Hg20	(6)
The overall reaction for (5) and (6) becomes:
Ca(OH)2 + S02 + 02 + 2IIg° -> CaS04 i-H20 4- Hg20	(7)
S02 also inhibits reactions (3) and (7) by forming CaS03 and closing off pores:
Ca(OH)2 + S02 -> CaS03 + II20	(8)
The oxidant-enriched lime tests in Figures 4-6 support the theory that the uptake of Hg° by calcium
sorbents is a two-step reaction, where Hg° is oxidized at the surface (preferably in < 100A pores) by
reactions (3) and (7) and then farther interacts with the alkaline sorbent. While S02 likely promotes
Hg° oxidation to Hg20 by reactions (5) and (6), it also forms solid reaction products by equations
(4), (5), (7), and (8)r gradually closing off pores with CaS03 and CaSO,, layers. Therefore, S02 can
actually inhibit Hg° uptake capacities by closing off the pore reaction sites for Hg20, while
simultaneously promoting the formation of Hg20. S02 apparently intercepts the CaO OX formed
by oxidant hydration of lime by reaction (4), thus further inhibiting Hg° sequestering.
The temperature effect upon Hg° uptake is similar for oxidant-enriched and base lime hydrates in
that decreasing the temperature below 100°C tends to slow down the necessary reactions; this is
further indication that the observed reactions do not involve an Hg° condensation mechanism, which
would increase with decreasing temperature. The significant findings from this work are that the
initial rate of Hg° uptake can be dramatically increased by oxidant-enrichment of lime hydrates, and
the overall capacity for Hg° may be significantly increased if a reduced S02 environment can be
established. Although hydrogen chloride (I IC1) was not present in any of the above work, it is
expected that the role of HCI in sorbent pore pluggage would be similar to that for S02. Other prior
studies with combined lime and activated carbon, show dramatically improved mercury uptake,
presumably due to the lime's sequestering of acid gases before product chlorides and
sulfites/sulfates can deactivate the highly porous activated carbon (17,18).
Silicate sorbents prepared from glass and lime are intrinsically higher in surface area and porosity
than hydrated lime (Table 2 vs. Table 1) and, therefore, would be expected to have higher Hg°
uptake rates and capacities than lime. The only data in Figures 7 and 8 are for 500 ppmv S02
present, so the capacities of silicates at low or no S02 conditions are yet to be determined.
Attempting to increase surface area and oxidation potential by dual additives to silicate sorbents
simultaneously appears to have been unsuccessful, although either individual additive shows
positive benefits.
8

-------
CONCLUSION
The practical significance of these results is that it has been shown that for sorbents to be effective
for total mercury control, where an appreciable portion of mercury is Hg°, pore structure and
oxidants are necessary. This implies that sorbents added for acid gas control will not be as effective
for mercury control unless steps are taken to reduce the pore closure by solid chlorides, nitrates, and
sulfites/sulfates. This further suggests that staging of sorbent addition will be more effective with
less expensive alkaline sorbents added upstream for bulk acid gas removal, followed by downstream
addition of premium, higher porosity, oxidant-enriched sorbents for mercury control.
ACKNOWLEDGMENTS
The authors acknowledge the technical support of Wojciech Koslowski and Jaroslaw Karwowski of
ARCAD1S Geraghty & Miller in the conduct of bench-scale experiments and physical
characterization of sorbents.
REFERENCES
1.	D.G. I,angley, "Mercury Methylation in an Aquatic Environment," J. Water Pollut. Control
Fed., 45: 44-51 (1973).
2.	G. Westoo, "Methyl Mercury; a Percentage of Total Mercury in Flesh and Viscera of
Salmon and Sea Trout of Various Ages," Science, 181: 567-568,(1973).
3.	M.H. Keating, K.R. Mahaffey, R. Schoeny, G.E. Rice, and O.K. Bullock, Mercury Study
Report to Congress, EPA 452/R-97-003, U.S. EPA, Office of Air Quality Planning and
Standards, Research Triangle Park, NC ( December1997).
4.	D.H. Klein, et al., "Pathways of 37 Trace Elements Through Coal-Fired Power Plant,"
Environ. Sci. & Technol., 9:10,973(1975).
5.	T.G. Brna, and J.D. Kilgroe, "The Impact of Particulate Emissions Control on the Control of
Other MWC Air Emissions," J. Air & Waste Mgt. Assoc., 40(9): 1324 (1990)
6.	R. Chang, and G.R. Offen, "Mercury Emission Control Technologies: An EPRI Synopsis,"
Power Engineering (November 1995).
7.	S.V. Krishnan, B.K. Gullett, and W. Jozewicz,"Sorption of Elemental Mercury by
Activated Carbons," Environ. Sci. & Technol., 28:8, 1506-1512 (1994).
8.	S.V. Krishnan, H. Bakhteyar, and C.B. Sedman, "Mercury Sorption Mechanisms and
Control by Calcium-Based Sorbents," in: Proceedings of the 89th Air & Waste Management
Association Annual Meeting, Paper No. 96-WP64B.05, Nashville, TN (1996).
9

-------
9.	S.B. Ghorishi, and B.K. Gullett, "An Experimental Study on Mercury Sorption by Activated
Carbons and Calcium Hydroxide." in: Proceedings of the Fifth Annual North American
Waste-To-Energy Conference, Research Triangle Park, NC (April 1997).
10.	S. Nelson Jr., and J. Miller, "New Mercury Control Technologies for the Ft. Dix Waste-To-
Energy Plant," in: Proceedings of the Fifth Annual North American Waste-To-Energy
Conference, Research Triangle Park, NC (April 1997).
11.	R. Chang, C.J. Bustard, G. Schott, T. Hunt H. Noble, and J.Cooper, "Pilot-Scale Evaluation
of Carbon Compound Additives for the Removal of Trace Metals at Coal-Fired Utility
Power Plants," presented at the Second international Conference on Managing Hazardous
Air Pollutants, Washington, DC (July 1993).
12.	S.J. Miller, D.L Laudal, R. Chang, and P.D. Bergman, "Laboratory Scale Investigation of
Sorbents for Mercury Control," Paper No. 142, presented at the AWMA Annual Meeting,
Cincinnati, OH (June 1994).
13.	C. Wu, T.G. Lee, E. Arar, and P. Biswas, "Novel In-Situ Generated Sorbent Methodology
and UV Irradiation for Capture of Mercury in Combustion Environments," in: Proceedings
of the First EPRI-DOE-EPA Combined Utility Air Pollutant Control Symposium (The Mega
Symposium), EPRI TR-108683-V3, Washington, DC (August 1997).
14.	D. Roberts, R.M. Stewart, and T.E. Broderick, "Capturing and Recycling Par Per Billion
Levels of Mercury Found in Flue Gases," in: Proceedings of the First EPRI-DOE-EPA
Combined Utility Air Pollutant Control Symposium (The Mega Symposium), EPRI TR-
108683-V3, Washington, DC, August 1997.
15.	S.B. Ghorishi, and C.B. Sedman, "Low Concentration Mercury Sorption Mechanisms and
Control by Calcium-Based Sorbents; Application in Coal-Fired Processes," Journal of Air &.
Waste Management Association, 48: 1191-1198 (1998).
16.	S.B. Ghorishi, and C.B. Sedman, "Combined Mercury and Sulfur Oxides Control Using
Calcium-Based Sorbents," in: Proceedings of the First EPRI-DOE-EPA Combined
Utility Air Pollutant Control Symposium (The Mega Symposium), EPRI TR-108683-V3,
Washington, DC (August 1997).
17.	A. Licata, E. Goetz, and L.P. Nethe,"Lime Enhanced Moving Bed Filters for Mercury and
Dioxin Control," in:Proceedings of the Fifth Annual North American Waste-To Energy
Conference, Research Triangle Park, NC (April 1997).
18.	E.G. Waugh, et al., "Mercury Control on Coal-Fired Flue Gas Using Dry Carbon-Based
Sorbent Injection Pilot-Scale Demonstration, presented at the AWMA Annual Meeting,
San Diego, CA (June 1998).
10

-------
100
50
30
20
10
lilf
~ fgd
Figure 1- Hg° uptake by calcium silicates, 100 °C,
2 hours of exposure to 40 ppb Hg°, 500 ppm SO2,
1% H2O, 4% O2, 10% CO2, and balance N2
11

-------
~ fgd
Figure 2. Initial rate of Hg° uptake by calcium silicates, 100 °C,
2 hours of exposure to 40 ppb Hg°, 500 ppm SO2,1% H2O, 4% O2,
10% CO2, and balance N2
12

-------
Mercury Source
N
Carbon Trap
Manifold
On-off Valves
By-Pass
"iii" i
3-Way Valve
3-Way Valves
CO
Heater
Water Batfl
Water Generator
Carbon
Trap
Analyzer
Rotameter
Data Acquisition
	 -		1


Mercury
Analyzer



CD

.O

TJ

c

-J

NAFION
Dryer
Figure 3. Schematic of the bench-scale fixed-bed reactor system
13

-------
3 °-2
-------
Temperature, C
500 ppm SO2
no SO2
Figure 5. Hg° sorption capacity (2-hour exposure)
for modified lime sorbents; effect of temperature,
SO2, and oxidant
15
-------
Temperature, C
Figure 6. SO2 sorption capacity (2-hour exposure)
for modified lime sorbents; effect of temperature
and oxidant
16
-------
c
£
i
-5*
*S)
3
B
CO
0
CO
a
2
U)
X
no A, no Ox A, no Ox no A, Ox
A: additive, Ox: oxidant
A, Ox
Figure 7. Initial rate of Hg° uptake by calcium silicates;
500 ppm SO2,100°C
17
-------
no A, no OX A, no OX no A, OX
A: additive, OX: oxident
A, OX
Figure 8. Hg° sorption capacity of calcium silicates:
500 ppm SO2,100°C
18
-------
cn 30
A, OX
no A, no OX A, no OX
A: additive, OX: oxident
Figure 9. SO2 sorption capacity of calcium silicates;
500 ppm SO2,100°C
19
-------
NRMRL-RTP-P-431 (Plea far?,completing)
1. REPORT NO. 2.
600/A-99/066
3. RECIPIENT'S ACCESSION NO.
4. TITLE ANO SUBTITLE
Preparation and Evaluation of Modified Lime and
Silica-Lime Sorbents for Mercury Vapor Emissions
Control
5. REPORT DATE
6. PERFORMING ORGANIZATION COOE
7. author(s)s# Ghorishi and C. F. Singer (ARCADIS),
and C. B. Sedman (EPA)
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND AODRESS
ARCADIS Geraghty and Miller, Inc.
P.O. Box 13109
Research Triangle Park, North Carolina 27709
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68C-99-201
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Air Pollution Prevention and Control Division
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Published paper; 10/98-7/W
14. SPONSORING AGENCY CODE
EPA/600/13
is.supplementary notes ^ppQ£) project officer is Charles B. Sedman, Mail Drop 04, 919/
541-7700. For presentation at Mega Symposium, Atlanta, GA, 8/16-20/99.
16. abstract ^he paper discusses current efforts to improve the uptake of mercury spe-
cies by increasing active sites and adding oxidative species to the sorbent. (NCTE:
Previous work showed that mercury chloride vapor is readily absorbed by calcium-
based sorbents as an acid gas in environments typical of coal-fired boiler flue gas,
while elemental mercury vapor is absorbed by calcium-based sorbents only when
sulfur oxides are also present.) Preparation of modified lime and silica-lime sor-
bents and their behavior toward mercury species are compared to those of commer-
cially available lime on a fixed-bed bench reactor. The implications of findings to-
ward development of multipollutant control technologies and planned field pilot eval-
uations of more promising multipollutant control concepts are discussed.
17. KEY WORDS AND DOCUMENT ANALYSIS
a. DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TCRMS
c. COSATl Field/Group
Pollution Flue Gases
Mercury (Metal) Coal
Emission Combustion
Sorbents Boilers
Calcium Cxides
Silicon Dioxide
Pollution Control
Stationary Sources
13	B 21B
07B 21D
14	C
11G 13 A
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report/
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (This page)
Unclassified
22. PRICE
EPA Form 2220-1 (S-73)
-------