Journal of Minerals & Materials Characterization & Engineering, Vol. 1, No.1, pp39-60, 2002
Printed in the USA. All rights reserved
Unburned Carbon from Fly Ash for Mercury Adsorption: I. Separation
and Characterization of Unburned Carbon
J. Y. Hwang, X. Sun, Z. Li
Institute of Materials Processing,
Michigan Technological University, Houghton, Michigan 49931
In searching for a low cost adsorbent for mercury removal from flue gas, this
study focuses on the utilization of unburned carbons from fly ash as the substitute
material for the costly activated carbons. In this first paper of the series, various
separation technologies are introduced for the extraction of unburned carbon from
different sources of fly ash. The unburned carbons have been efficiently separated from
clean ash, which is a value-added product for the concrete industry, with the separation
technologies such as gravity separation, electrostatic separation, and froth flotation.
Carbon concentrate with a LOI (Loss On Ignition) value of 67~80% has been generated
from the processes. Characterization of the carbon products has been performed to
determine the physical and chemical properties of the material. It has been found that the
unburned carbon particles had a porous structure, which is similar to the activated carbon.
The BET surface area of these materials was in a range of 25~58m
/g. The majority of
the pores are in the range of macropore, and some parts of the surface were embedded
with glass spheres. There is a linear relationship between the LOI value and the carbon
and sulfur content in the carbon concentrate. Chemical analysis indicated that the
mercury content in unburned carbon was much higher than the other separation products,
which suggests that the carbon has certain ability to capture mercury from flue gas.
Mercury has long been known as a potential health and environmental hazard. It
is identified as one of the 189 toxic air pollutants by the Clean Air Act Amendments of
1990. Mercury emissions are of particular concern because mercury accumulates in the
biosystem, is very difficult to monitor and capture, and is high in the public
Coal combustion accounts for 80% of the total energy consumption by the utility
industry and all coals contain some level of mercury. Since mercury is a highly volatile
element, when the coal is burned, most of the naturally occurring mercury will be
released as a vapor phase into the atmosphere. According to the EPA Mercury Study
the annual mercury emission rate during years of 1994-1995 is estimated to be
158 tons, of which 33% is from coal-fired utility boilers, the largest among the four major
mercury emission sources. (The other three are municipal waste combustion (19%),
commercial/industrial boilers (18%) and medical waste incinerators (10%).)
40 J. Y. Hwang, X. Sun and Z. Li Vol. 1, No. 1
To cope with the mercury emission problem, efforts have been made to remove
various mercury species from the flue gas of utility boilers. However, due to technical
and economic limitations, no process has been commercially utilized beyond pilot scale
tests. Among the current technologies being evaluated, activated carbon injection is the
most promising process for removing mercury from flue gas due to its high removal
In this process, activated carbon powder is injected into the flue gas stream
and collected, after adsorption, with a particulate matter control device. However, the
high cost of activated carbon hinders the large-scale applications in utility boilers.
Therefore, there is a desire to find an alternative carbon..
Fly ash is a by-product generated from the burning of pulverized coal in coal-fired
facilities such as utility boilers. Because fly ash is composed of very fine solid particles,
it must be captured from flue gas before being introduced into atmosphere. Every year,
according to a survey of American Coal Ash Association (ACAA) in 1996, there are
about 54 million tons of fly ash generated in USA from all coal-fired utilities. Therefore,
beneficial use of fly ash becomes a major issue in both environmental and economic
Currently, the biggest market for fly ash is in concrete applications as an additive
for cement. One of the criteria for such application is that the carbon content or the LOI
(loss-on-ignition) limit must be lower than 6% according to ASTM standard 618 and
lower than 3% in market practice.
Usually the carbon content in fly ash is in the range
of 2-12%. However, with the introduction of the 1990 Clean Air Act Amendments, caps
have been established on the emission of nitrogen oxides (NO
). Many coal-fired utilities
have started to retrofit with low NO
burners to meet the emission requirements. As a
result of such transition, the carbon content of fly ash increases significantly, up to 20%
in some cases, due to the low oxygen and/or low temperature combustion conditions that
are required by those low NO
combustion units. Therefore, the carbon content in fly ash
needs to be reduced before the low NO
fly ash can be utilized in concrete applications.
Since the unburned carbon separated from fly ash is a byproduct, any practical
application of such material would be economically and environmentally advantageous to
the overall fly ash beneficiation process. Preliminary study has shown that some
unburned carbon from fly ash has certain capabilities of adsorbing elemental mercury.
Such findings triggered the idea of using fly ash carbon as a low cost adsorbent in
removing elemental mercury from gas phases such as utility flue gas, in replacement of
costly activated carbons.
This series of papers presents the research efforts that have been conducted to
separate unburned carbon from fly ash, the characterization of the carbon, the mercury
adsorption tests of the carbon at various conditions, as well as the elucidation of the
relations between adsorption behavior of unburned carbons and their surface physical and
chemical properties.
Vol. 1, No. 1 Unburned Carbon from Fly Ash for Mercury Adsorption
Separation technologies
As mentioned earlier, the main purpose of separating unburned carbon from fly
ash is to obtain high quality fly ash for concrete applications. Another incentive for a
well-established separation technology is that, to efficiently utilize fly ash, high value
products must be generated from raw ashes to counter balance the transportation cost of
the material. According to the characterization study on different fly ash samples,
has been found that the major mineral components in fly ash are silicates, iron oxides,
low-density silicates (cenospheres) and unburned carbons. These particulate mineral
materials, after various treatments, can be used as fillers in plastics, reinforcement
material in metal matrix composites, as well as refractory materials.
according to this study, the unburned carbon can be used as an adsorbent in removing
hazardous substances such as mercury in flue gas. By separation of individually
functional components, fly ash is converted from a completely waste material into
various high value-added products.
Usually, there are four major processing techniques for the physical separation of
various components from fly ash, i.e. gravity separation, electrostatic separation,
magnetic separation, and froth flotation.
The gravity separation process can be used to separate dense particles with light
particles, here in this special case, to extract unburned carbon from the coarse fraction of
fly ash. This coarse fraction of fly ash has a high LOI value and is difficult to process
with other techniques such as froth flotation. Also, the process is useful in separating
hollow cenospheres, which have a specific gravity of about 0.64.
The electrostatic separation is used to separate particulate materials with different
electrical conductivity. Under a high electrostatic field, the less conductive particles are
polarized while the conductive particles are not, so that a high-tension drum can pick up
the polarized particles to make the separation. This process is suitable for the separation
of relatively coarse materials, such as particles of +100 mesh.
Froth flotation is a widely used separation technique in mineral processing
industry. In the slurry of solid and water, physico-chemical processes take place at the
interface of solid, liquid and gas phases, during the process of froth flotation. A collector
selectively coats the surfaces of certain mineral particles causing the surface of one or
more of the components in the slurry to become hydrophobic and responsive to the
attachment of air bubbles introduced into the slurry. Separation is accomplished as the
mineral-laden air bubbles rise to the surface in a froth (or a concentrate) which flows over
a weir, leaving in the slurry particles that are not coated with the collector. These pass out
the bottom of the cell. Under proper conditions, almost all materials can be made to float.
Success depends on the capability to control the surface chemistry to yield selective
adsorption of collectors. A second reagent, known as a frother, is used to stabilize the air
bubbles upon which the floatable minerals become attached.
42 J. Y. Hwang, X. Sun and Z. Li Vol. 1, No. 1
The froth flotation process to separate unburned carbon from fly ash has been
developed by Michigan Technological University. A target of less than 1% carbon in
clean ash with 90% or more recovery, or 80% of LOI in the carbon concentrate with 70%
or more recovery of total carbon, can be achieved with this process.
In this study the
froth flotation was used to separate unburned carbon from the 100 mesh portions of fly
ash samples. Froth flotation is the most common physical separation technology utilized
in the mining industry because of its reliability, low maintenance, and economics.
The iron oxide component in fly ash can be extracted with the magnetic
separation process. Because iron oxides are strong paramagnetic materials, the separation
efficiency is well established. In this paper, as the first step of the series, various
separation techniques for unburned carbon will be introduced for the fly ashes from
different sources, and results of characterization for the separation products will also be
presented in detail.
Three representative fly ash samples from different power plants were used in the
separation tests. Since froth flotation is suitable for separation of particulate materials at
sizes of 100 mesh, the fly ash samples were screened in order to obtain the coarse
portion +100 mesh. The separation of unburned carbon from the coarse portion was
carried out with gravity separation followed by electrostatic separation. The fine portion
was fed into flotation circuit for carbon separation.
A diagonal-deck concentration table (Wilfley table), Super Duty RH15S SD, was
used for the gravity separation. The tilt angles, stroke and length as well as the flow rate
of wash water were adjusted to the optimum levels. For each experiment run, the material
was fed to the table through a vibrating feeder. The feed rate was determined by the dry
weight of the samples taken at an interval of 15 to 30 seconds. Three separation products
were obtained at the end of the process, i.e. carbon concentrate, clean ash product, and
The electrostatic separation was carried out with an Eriez electrodynamic
electrostatic separator. The test conditions are listed as following,
Exiting voltage: 18 kV,
Drum speed: 60 rpm,
Angle of the ionizing electrode: 31
Angle of the static electrode: 17
The froth flotation tests were carried out in a pilot plant operation running at 200
lb/hr. Fly ash was mixed with water at 20% solid content, and the slurry was fed into a
tank where the cenospheres were skimmed off from the top since they have a density
lower than that of water. Then the slurry was fed into a magnetic drum separator to
recover the magnetic spheres. After magnetic separation, the slurry was conditioned with
an oil (fuel oil #2) collector at a dosage of 2 lbs./ton (Fuel oil at this dosage was proved
Vol. 1, No. 1 Unburned Carbon from Fly Ash for Mercury Adsorption
not damaging to mercury adsorbability in a separate study). The oil has an affinity for
carbon and is preferentially adsorbed on the carbon particles. The slurry was then fed into
a flotation machine where air was bubbled through it. During flotation, the rising air
bubbles collided with the oil coated carbon particles and attached themselves to these
particles due to a hydrophobic interaction. This caused the carbon particles to float to the
top of the flotation cell, where they were skimmed off. This flotation operation left the
clean ash in the cell and the carbon fraction was transferred to another flotation cell and
re-floated in order to upgrade the carbon content in the carbon concentrate. The reject
from the carbon-refloat operation was then returned to the first flotation cell.
The effectiveness of separation was evaluated with LOI value analyzed at 950
in the presence of air. Densities of the separation products were measured with a
pycnometer, Accupyc 1330 (Micromeritics). Helium was used as the measurement
medium. Five data points were taken for each sample, and the average number was
reported with a standard deviation of less than 0.0010 g/cm
, for a density range of 0.6
to 2.5 g/cm
. The particle size distribution of the unburned carbon was analyzed
with a particle size analyzer, Microtrac (Micromeritics), in combination with the sieving
The surface morphology was investigated with a Scanning Electron Microscopy,
JEOL 850. The surface area and the pore structure were analyzed with FlowSorb II 2300,
The chemical composition of ashes and their components were determined with
AA and ICP spectroscopy. Mercury content of ash components was analyzed with a
Manual Cold Vapor AAS technique, according to the EPA Method 7471.
Separation Results
Gravity separation. Three fly ash samples representing different classes of ashes
were tested in the experiment. FA1 fly ash is a Class F ash, FA2 a mixture of Class F and
Class C ash, and FA3 a Class C ash (based on ASTM classification). In general, Class F
ash is the combustion residue of the bituminous coal and Class C ash is the residue from
sub-bituminous coal and lignite. Figure 1 shows the particle size distributions associated
with the LOI values of as-received ash sample. For ash samples FA2 and FA3, it can be
seen from the figure that the LOI value increases as the particle size increases. For the
FA1 sample, however, the LOI value increases with the particle size until it reaches 65
mesh (175 µm), and the LOI value becomes lower when the particle size is at 465 µm.
Such difference may be attributed to the rank of the coal sources and the initial sizes of
the coal as well as the combustion flue gas and particulate collector conditions (For
example, high flue gas velocity may carry more coarse particles into electrostatic
Since the gravity separation is suitable for the coarse particles, only +100 mesh fractions
of the fly ash samples were tested with this method. Tables 1-3 list the separation results
for each fly ash sample. The head +100 mesh ashes had LOI values from 36% to 42%,
44 J. Y. Hwang, X. Sun and Z. Li Vol. 1, No. 1
whereas the carbon concentrate for each fly ash sample, as a separation product, has a
much higher LOI value, from 63% to 66%. The LOI recovery in the carbon concentrate
ranged from 84% to 95%. The clean ash, as the other product, had a very low LOI value
for the FA1 sample, at 0.03%. Reasonably low LOI values were also obtained for FA2
and FA3 samples.
Electrostatic separation. Although separation of unburned carbon with clean ash
had been achieved with the gravity separation method, the LOI values in carbon
concentrates were still relatively low. Microscopic examinations indicated that there were
some siliceous materials with porous structure in the carbon concentrates. Hence, to
further concentrate the unburned carbon, the electrostatic separation method was used,
based on the fact that the unburned carbon particles are conductive while the siliceous
particles are not. Carbon concentrates derived from the gravity separation tests were
subjected to two stages of electrostatic separation, i.e. a rough and a scavenging stage
respectively. Results of bench scale tests were shown in Figure 2. The FA2 sample had
the best separation, the LOI value increased from 66% to 80% with a yield of 76%. The
FA1 sample had an upgrade in LOI from 63% to 79% with a yield of 73%, and the FA3
sample from 65% to 70% with a yield of 79%.
Froth flotation. Table 4 shows the results of separation with froth flotation of
FA1 sample (-100 mesh fraction). It can be seen that the LOI value in the carbon
concentrate increased to 67.7% from 21.7% of the head sample, and the LOI value was
greatly reduced to 0.40% in the clean ash, which indicates that the clean ash is suitable
for the concrete applications. These results demonstrate that the froth flotation is effective
in extracting the unburned carbon from fly ash.
Characterization Results and Discussion
Density measurement of the fly ash components. The densities of fly ash
separation products were measured with a pycnometer so as to set up a guideline for the
separation process. As mentioned before, a fly ash can be separated into components such
as clean ash, unburned carbon and cenospheres, according to their differences in density.
The criterion for separation ability as based on the differences in density (i.e. gravity
separation) is a parameter called the separation index (SI), which is defined as
where D
is the density of dense particles, D
is the density of light particles, and D
density of the fluid medium. Usually, the gravity separation is effective when SI is
greater than 2.5, and is difficult when SI is lower than 1.25. However, the size and the
shape of the particles also play an important role in the gravity separation.
Table 5 lists the density measurement results for each component of the fly ash
samples. Also listed are the densities of ash components from electrostatic separations,
Vol. 1, No. 1 Unburned Carbon from Fly Ash for Mercury Adsorption
specified as ES-ash. The ash component, also known as the siliceous particles, had the
highest density (2.3g/cm
to 2.4 g/cm
), and the carbon component had a density of 2.0
, whereas the ES-ashes had a lower density of 1.7 to 2.0
. Cenospheres had the
lowest density among these components, with a density of 0.68 g/cm3, which is lower
than water. The SI between the ash and unburned carbon ranges from 1.46 to 1.06, which
explains the difficulty of gravity separation in some cases.
Correlation of LOI values with carbon and sulfur content. The LOI value has
been used extensively in this study to express the carbon content of the ashes and their
components, due to the fact that the determination of LOI is much easier than the carbon
analysis. Because the LOI value includes primarily the carbon and sulfur contents, it is
possible to derive a relationship between these three parameters, provided that the later
two follow a certain pattern with the LOI values. Figures 3-5 shows the relationship
between the carbon/sulfur contents and LOI values for the three fly ash samples. The data
points represent the results from clean ash, middlings (gravity), electrostatic rejects,
carbon concentrate (gravity), and electrostatic conductive, respectively. A linear trend
was observed for all three samples.
Particle size distribution of carbon concentrates. Carbon particle size is an
important factor in the adsorption applications. For a fix bed adsorber, the carbon
particles must be large to keep the bed sufficiently packed; while for a carbon injection
method, the particles should be relatively fine to be spread into the gas stream.
Figure 6 shows the size distribution of the carbon concentrate from various ash
sources. From the figure it can be seen that most of the particles are in the range of 200 to
400 µm. Figure 7 shows the size distribution of the carbon concentrate from FA1 samples
separated with froth flotation. In this case, most of the particles are under 140 µm.
Therefore, for fixed bed adsorption, pelletization of the carbon particles will be needed.
Morphologies of unburned Carbon. Figure 8 shows an SEM image of unburned
carbon from the FA1 sample. The porous structure of the particle is clearly seen from the
photo, indicating that the unburned carbon is a good candidate for the adsorbing material.
Within the resolution range of SEM, macropores of 1 to 10 µm are observed and also
visible are micro-sized glassy spheres embedded in the carbon particle. These glassy
spheres may be formed during combustion and they may have some influence on the
adsorption properties of the unburned carbon, which will be discussed in details in the
following papers of the series.
Surface area and pore structure of unburned carbon. Surface area, pore
structure, and surface chemistry are the critical parameters of a sorbent. In general, the
higher the surface area, the greater the adsorption capabilities. For example, the most
commonly used adsorbent activated carbon usually has a surface area of 800 to 1200
per gram of the material. The pore structure of the material includes the porosity, the
pore size and the pore size distribution. Pores of different sizes serve different functions
in the adsorption process. Macropores (>50 nm) act as a conduit for the bulk medium;
46 J. Y. Hwang, X. Sun and Z. Li Vol. 1, No. 1
mesopores (2~50 nm) and micropores are the primary source of the active adsorption
sites. Surface chemistry will be discussed in the following paper.
Surface areas of the carbon concentrates and activated carbons (F-400, BPL, from
Calgon Co.) as references were measured with BET method and results are presented in
Table 6. From the data listed in the table, it can be seen that the specific surface area of
unburned carbon is in a range of 25 to 58 m
/g, which is much lower than the activated
carbon (945 m
/g for F-400). This can be explained by the fact that the unburned carbon
has not experienced an activation process during the combustion, collection and
separation processes as activated carbon experiences, although the steam in the
combustion gas may have caused a limited extent of activation.
Not surprisingly, the micro pore volume of the unburned carbon is very low in
comparison with the activated carbon. The volume of pores between 17-3000 Å was in a
range of 0.033 to 0.059 ml/g, according to the nitrogen adsorption tests. Thus, most of the
pore volume is taken by the macropores as observed with SEM.
Chemical analysis of unburned carbon. Since the focus of this study is to use
the unburned carbon for mercury removals, only the chemical analysis for unburned
carbon is reported in this paper. Unburned carbon samples from the above mentioned
separation processes were analyzed for their chemical compositions, and Table 7 lists the
results of the analysis. Due to the fact that fly ash came out from a high temperature
environment, the major elements were reported as oxides, except carbon and sulfur. It can
be seen that all the unburned carbons have certain ash contents, which are mainly
composed of siliceous materials. The results of trace element analysis are presented in
Table 8. Apparently the mercury content in unburned carbon is much higher than that in
clean ashes.
To further illustrate how mercury is distributed, the mercury contents of fly ashes
as well as their components are analyzed, and the results are shown in Table 9 (The data
was based on the products of whole beneficiation process). It is obvious that the mercury
concentration in the carbon concentrates is almost 100 times higher than the other
separation products. In other words, the unburned carbon in fly ash captured certain
amount of mercury during the process, which provides two important findings. First,
since the majority of the mercury carried by fly ash is in the carbon concentrate, the
detoxification of fly ash in terms of mercury can be achieved by removing carbon from
the fly ash and carbon cleansing thereafter. Second, the abnormally high content in the
carbon concentrate indicates that the unburned carbon has a high affinity to mercury that
is released during coal combustion. Therefore, it is possible to use unburned carbon
separated from fly ash as the adsorbent for the mercury in the flue gas. Since the
unburned carbon is almost a no-cost material, the cost of flue gas cleansing for mercury
can be greatly lowered, so that the implementation of carbon adsorption technology for
industrial facilities can be accelerated.
It has also been found that the mercury content in carbon concentrates is
proportional to the LOI value for each sample. Figure 9 shows the correlation between
Vol. 1, No. 1 Unburned Carbon from Fly Ash for Mercury Adsorption
the mercury content of various separation products derived from the three fly ashes, and
it can be seen that there is an almost linear relationship between them. This finding
further indicates that the mercury is associated with the unburned carbon in fly ash.
1. EPA, Mercury Study Report to Congress, 1997, EPA-452/R-97-10
2. Clarke, L. B.; Sloss, L. L. IEA Coal Research, IEACR/49, 1992, July.
3. White, D. M.; Kelly, W. E.; Stucky, M. J.; Swift, J. L.; Palazzolo, M. A. Field test
of carbon injection for mercury control: Canden County municipal waste
combustor. Prepared for Office of Research and Development, U.S. EPA,
Washington, D.C. EPA-600/R-93-181 (PB94-101540)
4. Miller, S.; Laudal, D.; Dunham, G. Proceedings of Eleventh Annual Coal
Preparation, Utilization, and Environmental Control Contractors Conference,
Pittsburgh, Pennsylvania, 1995, July.
5. Ramsay Chang ; Preparation and Evaluation of Coal-Derived Activated Carbons for
Removal of Mercury Vapor from Simulated Coal Combustion Flue Gases, Energy
& Fuels; 1998; 12(6); 1061-1070.
6. Paul F. Sanders ; Kinetics of Mercury(II) Adsorption and Desorption on Soil,
Environmental Science & Technology; 31(2); 496-503.
7. Hwang, J. Y.; Huang, X.; Gillis, J. M. Proceedings: 13
International Symposium
on Use and Management of Coal Combustion Products (CCPs), 1999, 1, 19-1-19-
8. Shannon D. Serre* and Geoffrey D. Silcox ; Adsorption of Elemental Mercury on
the Residual Carbon in Coal Fly Ash, Industrial & Engineering Chemistry
Research; 2000; 39(6); 1723-1730.
9. Tanaporn Sakulpitakphon, James C. Hower,* Alan S. Trimble, William H. Schram,
and Gerald A. Thomas; Mercury by Fly Ash: Study of the Combustion of a High-
Mercury Coal at a Utility Boiler, Energy & Fuels; 2000; (3); 727-733.
10. James C. Hower,* M. Mercedes Maroto-Valer, Darrell N. Taulbee, and Tanaporn
Sakulpitakphon ; Mercury Capture by Distinct Fly Ash Carbon Forms, Energy &
Fuels; 2000; 14(1); 224-226.
11. James C. Hower,* Robert B. Finkelman, Robert F. Rathbone, and Jennifer
Goodman; Intra- and Inter-unit Variation in Fly Ash Petrography and Mercury
Adsorption: Examples from a Western Kentucky Power Station, Energy & ; 2000;
14(1); 212-216.
12. P. Fermo, F. Cariati, S. Santacesaria, S. Bruni, M. Lasagni,# M. Tettamanti,# E.
Collina,# and D. Pitea*# ; MSWI Fly Ash Native Carbon Thermal Degradation: A
TG-FTIR Study, Environmental Science & Technology; 2000;
13. Hwang, J.Y. and Li, Z., “Control of Mercury Emissions Using Unburned Carbon
From Combustion By-Products,” U.S. Patent 6,027,551 (2000).
14. Hwang, J.Y., Hein, A.M., Kramer, R.S., Liu, J., Hozeska, T.J., Zhang, Q., and
Scott, T.E., “Application of Characterization on Fly Ash Beneficiation to Produce
Quality Controlled Products,” Process Mineralogy XI: Characterization of
Metallurgical and Recyclable Products, 1991, pp. 167-180.
48 J. Y. Hwang, X. Sun and Z. Li Vol. 1, No. 1
15. Hwang, J.Y., Liu, X., Zimmer, F.V., Thiruvengadam, T.R., and Patzias, T.,
“Beneficiation Process for Fly Ash and Utilization of Cleaned Fly Ash for Concrete
Application,” Proceedings: 11
International Symposium on Use and Management
of Coal Combustion By-Products, Vol. 1, Paper 11, EPRI Publication, 1995.
16. Hwang, J.Y., “Wet Process for Fly Ash Beneficiation,” U.S. Patent 5,047,145
(1991), and 5,227,047 (1993).
Vol. 1, No. 1 Unburned Carbon from Fly Ash for Mercury Adsorption
Table 1. Results of gravity separation for +100 mesh FA1 sample
Products Weight percent LOI/carbon % LOI/carbon recovery %
Carbon concentrate
54.36 63.44 95.01
Middlings 26.47 6.81 4.97
Clean ash 19.17 0.03 0.02
Head 100.00 36.30 100.00
Table 2. Results of gravity separation for +100 mesh FA2 sample
Products Weight percent LOI/carbon % LOI/carbon recovery %
Carbon concentrate
54.42 65.93 83.96
Middlings 23.70 25.41 14.63
Clean ash 23.88 2.44 1.41
Head 100.00 41.17 100.00
Table 3. Results of gravity separation for +100 mesh FA3 sample
Products Weight percent LOI/carbon % LOI/carbon recovery %
Carbon concentrate
52.33 65.03 90.93
Middlings 25.10 11.50 7.73
Clean ash 22.57 2.22 1.34
Head 100.00 37.42 100.00
Table 4. Separation results from froth flotation for FA1 sample
Head sample
Carbon conc. Clean ash Cenosphere Magnetics
LOI 21.7 67.7 0.40 2.4 N/A
44.0 19.3 58.6 57.6 14.3
22.4 9.92 29.2 29.57 8.20
5.30 0.04 5.20 3.71 77.18
O 2.35 0.80 3.16 4.23 0.43
1.11 0.70 1.33 0.91 0.31
50 J. Y. Hwang, X. Sun and Z. Li Vol. 1, No. 1
Table 5. Densities of fly ash components (g/cm
Fly ashes Ash Carbon ES-carbon Cenospheres
FA1 2.3183 1.9827 1.9271 0.6804
FA2 2.3542 2.0099 1.8313 N/A
FA3 2.3945 2.0144 1.6930 N/A
Table 6. Surface area and pore structure data for the carbon concentrates
Properties Unit FA1 FA2 FA3
BET surface area m
/g 25.60 18.71 58.33
Pore surface area (17-3000 A) m
/g 20.26 12.83 38.43
Micropore area m
/g 3.87 4.78 16.94
Pore volume (17-3000 A) ml/g 0.0325 0.0227 0.0586
Micropore volume ml/g 0.0015 0.0021 0.0074
Average pore diameter (BET) Å 49.97 48.87 43.74
Average diameter (sorption) Å 64.23 71.15 61.01
Table 7. Major components of the carbon concentrates (%wt)
Elements FA1 FA1 (flotation) FA2 FA3
12.94 19.26 17.15 19.44
3.80 9.92 5.27 3.99
0.93 0.04 0.83 0.94
MgO 0.11 0.5 0.44 0.57
CaO 0.09 0.5 N/A N/A
O 0 0.05 0 0.78
O 0.13 0.8 0.12 0
0.20 0.7 0.30 0.22
LOI 80.90 67.70 73.40 70.10
C 77.25 N/A 66.55 67.35
S 0.81 0.69 0.59 0.48
Vol. 1, No. 1 Unburned Carbon from Fly Ash for Mercury Adsorption
Table 8. Trace elements in the carbon concentrates (ppm)
Elements FA1 FA1 (flotation) FA2 FA3
B 186.5 N/A 201.9 391.7
Ba 97.20 96 47.30 37.93
Be 7.80 1 10.40 7.40
Ce 1040 N/A 2984 3545
Co 9.8 3 19.8 13.1
Cu 13.2 11 16.2 13.7
Hg 645 ppb 620 ppb 388 ppb 924 ppb
Sr 1018 72 3048 3752
Y 137.0 5 181.0 133.1
Zn 9.30 7 10.70 10.70
Table 9. Mercury content (ppb) in the components of fly ashes
Fly ash As received Carbon
Cenospheres Magnetic
Clean ash
FA1 145 620 7 <5 18
FA2 27 496 3 6 10
FA3 122 800 49 16 32
52 J. Y. Hwang, X. Sun and Z. Li Vol. 1, No. 1
Figure 1. LOI as a function of particle size for fly ashes.
Vol. 1, No. 1 Unburned Carbon from Fly Ash for Mercury Adsorption
Figure 2. Results of electrostatic separation for the +100 mesh fly ashes.
54 J. Y. Hwang, X. Sun and Z. Li Vol. 1, No. 1
Figure 3. LOI versus carbon and sulfur content of FA1 sample.
Vol. 1, No. 1 Unburned Carbon from Fly Ash for Mercury Adsorption
Figure 4. LOI versus carbon and sulfur content of FA2 sample.
56 J. Y. Hwang, X. Sun and Z. Li Vol. 1, No. 1
Figure 5. LOI versus carbon and sulfur content of FA3 sample.
Vol. 1, No. 1 Unburned Carbon from Fly Ash for Mercury Adsorption
Figure 6. Particle size distribution of fly ash carbons.
58 J. Y. Hwang, X. Sun and Z. Li Vol. 1, No. 1
Figure 7. Size distribution of FA1 carbon separated with flotation.
Vol. 1, No. 1 Unburned Carbon from Fly Ash for Mercury Adsorption
Figure 8. SEM image of FA1 carbon particle
60 J. Y. Hwang, X. Sun and Z. Li Vol. 1, No. 1
Figure 9. Mercury content at different LOI values.