Journal of Minerals & Materials Characterization & Engineering, Vol. 3, No.1, pp 13-22, 2004
jmmce.org Printed in the USA. All rights reserved
13
Adsorption of Vapor Phase Mercury on Various Carbons
Jinjing Luo
1
, A. M. Hein
2
, and J.Y. Hwang
3*
,
1
Dept. of Civil & Environmental Engineering,
2
Institutue of Materials Processing, and
3
Dept. of Materials Science & Engineering,
Michigan Technological University, Houghton, MI 49931, U. S. A.
Unburned carbon was found to be a component of fly ash resulting from
incomplete combustion in a pulverized-coal based power plant. Previous
investigations found that unburned carbon separated from fly ash exhibited good
mercury adsorption property. It would offer an opportunity to substitute activated
carbon with low cost unburned carbon for mercury adsorption from power plant
emission gases. This study provides a comparison of mercury adsorption by
carbon from various sources, including activated carbon and unburned carbon
from two different power plants. The experiments were conducted under various
temperatures and mercury concentrations to determine whether good mercury
adsorption properties can be obtained from various carbon sources. This study
revealed that mercury adsorption depended on the carbon sources and conditions.
Activated carbon (F400) demonstrated the best mercury absorbability among the
three tested carbons, followed by AEP unburned carbon. Pepco unburned carbon
showed very little mercury absorbability. Increasing the temperature generally
resulted in the decrease of mercury adsorption. Adsorption rate could be
effectively increased with increasing gaseous Hg concentration. Desorption
treatment before adsorption test could improve unburned carbon’s adsorption
capacity, especially for Pepco carbon.
KEY WORDS: Mercury, Adsorption, Activated Carbon, Unburned Carbon
INTRODUCTION
In 1995, around 5500 tons of mercury was emitted globally into the atmosphere
1
. Once in
the atmosphere, elemental mercury can float for over a year while oxidized mercury compounds
could drift several days before precipitating to the earth (soil or water). Mercury in water can be
converted by microorganisms to methylmercury, a highly toxic form, which may bio-accumulate
in the food chain. With consumption of contaminated fish, humans are exposed to the mercury
compounds. Research indicates that mercury compounds cause ecological and human health
impacts.
Based on the ICR (Information Collection Request) from U.S. EPA (Environmental
Protection Agency), coal-combustion power plants emitted around 48 tons of mercury in 1999,
which was estimated to be one third of the total nationwide mercury emissions from human
*
Author to whom correspondence should be addressed, Tel: 1-906-487-2600, Email: jhwang@mtu.edu
14 Jinjing Luo, A. M. Hein, and Jim Hwang Vol. 3, No. 1
activities
2
. C.O. Bauer believed the estimate was 39.8% based on EPA’s 1999 NEI (National
Emission Inventory)
3
. As a matter of fact, coal-fueled electric utilities have become the largest
anthropogenic source of atmospheric mercury in the United States.
EPA is developing a MACT (maximum achievable control technology) regulation to
limit mercury emissions from power plants. The final MACT standard will be issued by
December, 2004 and compliance is to be required by December 2007. In addition to EPA’s
MACT process, several multi-pollutant legislative proposals have been introduced in the 108
th
Congress, including the Clear Skies Act, which would cap mercury emissions at 26 tons in 2010
and 15 tons in 2018 down from 1999 baseline level (48 tons)
4
.
The three technologies being evaluated recently for flue gas mercury control are activated
carbon injection, FGD (flue gas desulfurization) spray dryers and wet scrubbers (wet FGD).
Among them carbon injection technology is the closest to commercialization. Its average
mercury removal efficiency is around 80-98%
5
. But the high cost (several billions of dollars as
estimated in EPAs report to Congress) hinders its application in coal-fueled electric utilities.
Finding a cost-effective sorbent with high mercury capture ability has generated great interest.
Unburned carbon must be extracted from fly ash in the fly ash recycling and cleaning
processes before fly ash can be used in the cement industry
6
. To be used as cement additive, fly
ash should not contain over 6% carbon due to ASTM C618 specification (American Society for
Testing and Materials). Under specific conditions, the market also forces carbon content in fly
ash to be no more than 2% for concrete industry
7,8,9
. Meanwhile with the low NOx burners being
employed in power plants to meet Clean Air Act Amendments in 1990s, the current carbon
content in fly ash increased to as much as 20%. Unburned carbon content is required to be
reduced in fly ash to meet cement industry demands. In 2002, over 76.5 million tons of fly ash
was produced from coal-fired power plants. The cement, concrete and grout industry utilized
14.5 million short tons of the available fly ash, amounting to 19% of total
10
. In 1991, this ratio
was 9% of the total amount of 51 million tons of fly ash
11,12
. Thus, a larger supply of unburned
carbon separated from fly ash will be available in the future.
Based on the data from the American Coal Ash Association, the typical 2003 price for
“concrete quality fly ash” was $20-45/ton, which is $0.01 to $0.0225/lb. The prices for other
usages of fly ash are much lower. The average price of activated carbon is over $0.5/lb, showing
that unburned carbon can beat activated carbon on price.
Unburned carbon possesses the property to capture mercury during the combustion
process. Due to the short residence time in boilers, unburned carbon from fly ash may not reach
its adsorption equilibrium and may still possess mercury capturing properties. Previous
research
13
has proven the Hg adsorption capacity of unburned carbon at low temperature (20
o
C
& 40
o
C). In this study, simulated flue gas temperature (150
o
C) was used to investigate the Hg
adsorption properties of carbon from various sources.
Vol. 3, No. 1 Adsorption of Vapor Phase Mercury on Various Carbons 15
EXPERIMENTAL DESIGN
In this study, two unburned carbon samples and one activated carbon sample were tested.
AEP unburned carbon and Pepco unburned carbon were obtained from AEP (American Electric
Power) fly ash, and Pepco (Potomac Electric Power) fly ash respectively. Both unburned carbon
samples were extracted from class F fly ash using the froth flotation method under the same
experimental conditions
14,15,16
. Activated carbon, F400, was obtained from Calgon Carbon
Corporation, Pittsburgh, PA, USA. F400 is one of the most studied and widely used activated
carbon products in vapor phase applications
17
.
Figure 1 illustrates the schematic diagram of the mercury vapor adsorption apparatus. Hg
source was a 0.5 cm long mercury permeation tube (VICI Metronics. Inc., CA). The tube was
placed at the bottom of a U-shaped glass tube and covered with glass beads to maintain uniform
mercury vapor. A temperature-adjustable water bath maintained a required stable temperature.
The carried gas was P.P. grade nitrogen gas. The concentrated mercury vapor was diluted with a
bypass line of nitrogen gas before being introduced into the carbon reactor, which was a 1cm
I.D. (inside diameter), 22 cm long glass column. The mixture of the carbon sample and short
glass fiber was placed in the middle of the column. Table 1 lists the related parameters. The
carbon bed temperature was regulated by a tube furnace. Tygon tubing from Saint-Gobain
Performance Plastics was selected for the connecting materials.
Figure1. Schematic Diagram of Mercury Vapor Adsorption Apparatus
Mercury
Analyzer
Carbon Bed
Tubular
Furnace
Mercury
Source
N
2
1.5%KMnO
4
+ 10%H
2
SO
4
16 Jinjing Luo, A. M. Hein, and Jim Hwang Vol. 3, No. 1
Table1. Parameters in Carbon Reactor
Carbon bed inside diameter 1 cm
Carbon loading 1.5g
Mercury source temperature 45
o
C
Carbon bed temperature 150
o
C
Flow rate of carrier gas (N
2
) 50 ml/min
Pressure of carrier gas 20 psi (1.36 atm)
Mercury vapor was colleted using the one-liter Tedlar sampling bag at the site upstream
and downstream of the carbon bed respectively. The collection time of the sample was set to
seven minutes. Effluent mercury was recorded every 20 minutes in the first hour and every hour
afterwards. Mercury vapor concentration was determined by a gold film mercury vapor analyzer
(JEROME 431-X, Arizona Instrument Corp)
18
.
Exhaust vapor was introduced to the impinger solution before being expelled into the air.
The impinger solutions were prepared daily by adding 1.5% potassium permanganate in 10%
sulfuric acid
19
. The blank test was performed before each new adsorption experiment and lasted
for 8 hours. After each test, the entire system was purged with pure nitrogen gas to expel Hg
residues and this process lasted for 6-8 hours.
Calculation Method:
Total mercury adsorption was calculated using following integration equation:
0
0'()*
tQ
tt
qCCdQ
=
(1)
where q’ is total amount of adsorbed mercury, C
0
and C
t
are influent and effluent Hg
concentrations at time t, and Q
t
is gas volume flowing into carbon bed at time t.
The quantity of mercury adsorption per unit carbon was calculated by following equation:
0
0()*
tQ
tt
m
CCdQ
q=
(2)
where q is total captured mercury, and m is the mass of carbon sample.
RESULTS & DISCUSSION
Mercury concentration in flue gas ranges from around 1ppbv to over several hundred
ppbv
20,21
. For bituminous coal, this range is between 0.01ppmv and 3.3ppmv
22
. The nominal flue
gas temperature is 149
o
C (300
o
F). In this study, carbon bed was set at 150
o
C, and mercury
influent concentration was adjusted at around 0.05mg/m
3
.
Vol. 3, No. 1 Adsorption of Vapor Phase Mercury on Various Carbons 17
Effect of Carbon Sources:
The origin of carbon affects its adsorption capacity. Carbon from various sources shows
entirely different performance. Figure 2 illustrates the adsorption curves of AEP unburned
carbon, Pepco unburned carbon and F400 activated carbon at 150
o
C and with influent Hg of
0.05mg/m
3
. F400 possesses much better adsorption behavior, it captured around twice the
amount of mercury that AEP adsorbed within the same time frame of testing. During the entire
sorption process, Pepco carbon did not show any positive adsorption ability. Actually, it emitted
over 0.066µgHg/gCarbon, which is assumed to be the result of the fleeing of preloaded mercury
from unburned carbon surface.
Effect of Temperature:
The influence of temperature on the adsorption capacity of the carbon samples was
studied at one typical low temperature, 20
o
C and one typical high temperature, 150
o
C, and with
flowing Hg content as 0.05mg/m
3
. As shown in Figure 3, with temperature increasing,
adsorption capacity of two unburned carbon samples decreased dramatically. The capacity of
AEP carbon dropped 65% with temperature increasing from 20
o
C to 150
o
C. It adsorbed 2.6µg
Hg/gCarbon at 20
o
C, and 0.9µgHg/gCarbon at 150
o
C. Pepco carbon captured 0.24µg Hg per
gram carbon at 20
o
C, and it did not capture any mercury at 150
o
C. High temperature caused a
reduction in the adsorption capacity of the carbon samples, which is consistent with physical
sorption theory.
Effect of Influent Mercury Concentration:
Gaseous Hg content in flowing vapor affected the adsorption behavior of carbon samples.
Figure 4 presents the adsorption curves of the three carbon samples at a temperature of 150
o
C
and for the concentrations of 0.05mg/m
3
and 0.1mg/m
3
. All carbon samples, including unburned
carbon and activated carbon, indicate faster capturing rates at high inlet Hg concentration.
Especially, the Pepco carbon, which adsorption capacity increased from negative to over
0.09µgHg/gCarbon with increasing of Hg concentration from 0.05mg/m
3
to 0.1mg/m
3
. With high
content in flowing vapor, vapor-phase Hg atoms have more chances to hit on carbon surface and
be attached, which results in better adsorption behavior.
Effect of Preloaded Mercury:
Unburned carbon contains preloaded mercury due to its origin from fly ash. The emission
of preloaded Hg from unburned carbons at the temperature of 150
o
C with Hg-free vapor passing
through was examined and the influence of this emission on the adsorption behavior of carbons
was also studied. Figure 5 displays desorption curves of Pepco and AEP carbons at 150
o
C. AEP
carbon was recorded to emit around 0.019µgHg/gCarbon in around 30 minutes testing, and the
total amount of Hg preloaded on the AEP unburned carbon should be higher, since it still can
emit 0.007 mgHg/m
3
at the end of the experiment. The Pepco carbon emitted about 0.14
µgHg/gCarbon in around 90 minutes of testing, and the preloaded Hg it contained should be
higher than this since it did not emit all the Hg it held till the end of testing.
The phenomenon that the preloaded mercury could emit from the carbon surface under
experimental conditions would affect mercury adsorption performance of the unburned carbon.
In this study, the adsorption capacities of unburned carbons were modified respectively by
adding the amount of Hg captured in the adsorption test and the amount of Hg emitted in the
18 Jinjing Luo, A. M. Hein, and Jim Hwang Vol. 3, No. 1
desorption tests, and the results are shown in Figure 6 with those without modification. As
shown in Figure 6, unburned carbons with modification indicated better adsorption behaviors,
especially the Pepco carbon, with modification its adsorption capacity changed from negative to
close to AEP carbon. The actual adsorption performance shows poor performance than the
modified one, which demonstrates the previous assumption that the fleeing of the preloaded Hg
from carbon surface reduced the further Hg capturing ability of carbon sample. Meanwhile, it
supplied the explanation why Pepco carbon showed negative adsorption capacity during testing.
To diminish the influence of preloaded mercury emission from carbon surface at a
temperature around 150
o
C, a desorption process could be conducted before the adsorption test.
Previous researches demonstrated that AEP carbon increased its adsorption capacity at a low
temperature (20
o
C and 40
o
C) after a desorption at 400
o
C with the presence of oxygen. The
desorption test for unburned carbon at a high temperature (over 400
o
C) will be conducted in the
future.
CONCLUSION
The adsorption capacity of carbon samples was source-dependent. At the temperature of
150
o
C and the influent Hg concentration of 0.05mg/m
3
,
commercial grade activated carbon
demonstrated the best adsorption ability among the three tested carbons. AEP unburned carbon
was better than Pepco unburned carbon. Pepco unburned carbon had a range of negative to very
little mercury adsorption capability.
The adsorption capacities of the three tested carbons were temperature sensitive. The mercury
capturing capacity of all carbon samples decreased when carbon bed temperature increased. This
is consistent with the physisorption theory.
The adsorption rates of all three carbon samples increased with influent Hg content. When
gaseous Hg concentration increased, Pepco unburned carbon changed its capacity from negative
to positive. AEP unburned carbon indicated larger increasing rate than F400 activated carbon
when the gaseous Hg content shifted from low to high.
The specific property of unburned carbon caused its adsorption process a combination of
adsorbing mercury from flowing vapor and desorbing the preloaded Hg to the flowing vapor.
The emission of preloaded Hg weakened the further mercury capturing ability of the carbon. For
improvement, a desorption pretreatment before adsorption process for unburned carbon is
required, especially for Pepco carbon. The desorption test at the temperature of 400
o
C with the
presence of oxygen was performed for AEP carbon, which enhanced its adsorption capacity at a
low temperature (20
o
C & 40
o
C).
Vol. 3, No. 1 Adsorption of Vapor Phase Mercury on Various Carbons 19
Fig. 2 Hg Adsorption Capacities of 3 Different Carbons at 150
o
C, 0.05 mg/m
3
-0.1
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1.1
1.2
1.3
1.4
1.5
1.6
0200400600800100012001400160018002000220024002600
Time (min)
Adsorption (ug/gCarbon)
AEP
Pepco
F400
Fig. 3 Influence of Temperature on adsorption for Two Unburned
Carbons
20
o
C
20
o
C
150
o
C
150
o
C
0
0.5
1
1.5
2
2.5
3
AEPPepco
Adsorption (ug/gCarbon)
20 Jinjing Luo, A. M. Hein, and Jim Hwang Vol. 3, No. 1
Fig. 4 Effect of Hg Concentration on the Adsorption for Different Carbons at 150
o
C
-0.2
0
0.2
0.4
0.6
0.8
1
1.2
1.4
1.6
0200400600800100012001400160018002000220024002600
Time (min)
Adsorption (ug/gCarbon)
AEP-High
Content
AEP-Low
Content
Pepco-High
Content
Pepco-Low
Content
F400-High
Content
F400-Low
Content
Fig. 5 Hg Contained in Unburned Carbons
-0.16
-0.14
-0.12
-0.1
-0.08
-0.06
-0.04
-0.02
0
0102030405060708090100110120130140150160170180190200
Time (min)
Adsorption (ug/gCarbon)
0
0.005
0.01
0.015
0.02
0.025
C (mg/m
3
)
AEP
Pepco
C-AEP
C-Pepco
Vol. 3, No. 1 Adsorption of Vapor Phase Mercury on Various Carbons 21
Fig. 6 Comparison Hg Adsorption Capacities Between Unburned Carbons With
or Without Modificatoin at 150
o
C, 0.055mg/m
3
-0.02
0
0.02
0.04
0.06
0.08
0.1
0.12
0.14
020406080100120140160180200
Time (min)
Adsorption (ug/gCarbon)
AEP-
Modified
AEP
Pepco-
Modified
Pepco
REFERENCE:
1
U.S. EPA Mercury Study Report to Congress, EPA-452/R-97-003, December 1997
2
Thomas J. Feeley,III, James Murphy, Jeffrey Hoffmann, and Scott A. Renninger, A Review of
DOE/NETL’s Mercury Control Technology R&D Program for Coal-Fired Power Plants, April
2003
3
Carl O. Bauer, DOE-NETL's Mercury R&D Program, 2003
4
http://www.epa.gov/air/mercuryrule/
5
Mercury Control Options for Coal-Fired Power Plants, Clean Air Network Fact Sheet, August
1999
6
J.Y. Hwang, “Unburned Carbon from Fly Ash, A Hidden Treasure,” 3
rd
Annual Conference on
Unburned Carbon on Utility Fly Ash, U.S. Department of Energy, May 1997, Pittsburgh,
Pennsylvania
7
Use of High-Carbon Fly Ash in Cement Manufacture
8
J.Y. Hwang, “Powder Technology in Concrete,” Powder Metallurgy, V. 40, No. 3, p. 160
9
X.M. Song, J.Y. Hwang, and X. Liu, “Utilization of Beneficiated Low NO
X
Fly Ash in
Concrete and Concrete Block,” Proceedings: 12
th
International Symposium on Coal Combustion
By-Product (CCB) Management and Use, Vol. 1, EPRI TR-107055-V1, January 1997, pp.30-1-
30-16
10
2002 Coal Combustion Product Production and Use Survey, American Coal Ash Association
11
American Coal Ash Association
22 Jinjing Luo, A. M. Hein, and Jim Hwang Vol. 3, No. 1
12
R. S. Kramer, J. Y. Hwang, X. Huang, and T. Hozeska, "Characterization of Recyclable
Components in Fly Ash to Produce Marketable Products," 1994 TMS Annual Meeting &
Exhibition, San Francisco, California, February 27-March 3, 1994
13
Z.Li, X.Sun, J.Luo, J.Y.Hwang, J.C.Crittenden, “Unburned Carbon from Fly Ash for Mercury
Adsorption: II. Adsorption Isotherms and Mechanisms”, Journal of Minerals & Materials
Characterization & Engineering, Vol1. No.2, p79-96
14
J.Y.Hwang, "Wet Process for Fly Ash Beneficiation," U.S. Patent 5,047,145 (1991).
15
J.Y.Hwang, X.Sun, Z.Li, “Unburned Carbon from Fly Ash for Mercury Adsorption: I.
Separation and Characterization of Unburned Carbon”, Journal of Minerals & Materials
Characterization & Engineering, Vol1. No.1, p39-60
16
J.Y. Hwang, X. Huang, J. Gillis, A. Hein, D. Popko, R. Tieder, and M. McKimpson,
“Separation and Utilization Technologies of Low NOx Ash,” Proceedings: 13
th
International
Symposium on Use and Management of Coal Combustion By-Products, Vol. 1, pp. 19-1-19-22.
17
Product brochure, Calgon Carbon Corporation
18
JEROME 431-X Mercury Vapor Analyzer Manue
19
Shendrikar, A.D.; Damle, A.; Gutknect, W.F. Collection Efficiency Evaluation of Mercury
Trapping Media for the SASS Train Impinger System, U.S. Environmental Protection Agency.
U.S. Government Printing Office: Washing, DC, 1984, EPA-600/7-84-089
20
V.K. Mathur and Z.Y. Chen, Mercury Oxidization in Non-Thermal Barrier Discharge System
21
U.S. EPA Mercury Study Report to Congress, Volume VIII: An Evaluation of Mercury
Control Technologies and Cost, December 1997
22
Livengood,C.D.; Huang,H.S.; Wu,J.M. “Proceedings of the 87
th
Annual Meeting of the Air
and Waste Management Association, 1994, reprint, p14