The Influence of Carbon Burnout on Submicron Particle Formation
from Emulsified Fuel Oil Combustion
C. Andrew Miller*
William P. Linak
Air Pollution Prevention and Control Division, MD-65
National Risk Management Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711 USA
Prepared for presentation at
28th International Symposium on Combustion
University of Edinburgh
Edinburgh, Scotland
July 30 - August 4, 2000
Oral Presentation Preference
Spray and Droplet Combustion Colloquium
Word Count:
Text:
3337
364
400
1000
References: 52 lines x7
Tables: 1x400
Figures: 1x200. 2x400
Total:
5101
^corresponding author
tel: (919) 541-2920
fax: (919) 541-0554
e-mail: miller.andy@epa.gov
-------
Abstract
A series of experiments have examined particle behavior and particle size distributions from the
combustion of different fuel oils and emulsified fuels in three experimental combustors. Results
indicate that improved carbon burnout from fuel oil combustion, either by decreasing the
temperature quench rate or by forming smaller fuel droplets through the secondary atomization
characteristic of oil/water emulsions, increases the volume of the submicron particle fraction. In
addition, the use of water-in-oil emulsions can increase the submicron particle volume compared
to a non-emulsified oil burned in the same combuslor. In contrast to larger coarse mode panicles
'which are composed largely of carbon char and inherently bound metals and sulfur, these
submicron particles appear to be composed of metal sulfates that arc more water-soluble than the
larger coarse mode particles.
For fuel oils, submicron particle volume varies directly with carbon burnout, and inversely with
total particle mass. These metal-sulfate-enriched submicron particles are formed by vaporization
and subsequent nucleation, coagulation, and condensation mechanisms. Where normal
atomization, high quench rates, or other obstacles to complete combustion exist, substantial
amounts of inorganic material remain bound with the unbumed carbon char. This material is,
therefore, unavailable to form the metal-sulfate-enriched submicron particle fraction.
These results may have significance relative to combustion modifications designed to reduce
unburned carbon levels. Health effects studies indicate that water-soluble transition metals, such
as those found in the submicron fraction of heavy fuel oil fly ash, may play a key role in
explaining possible mechanisms of pulmonary damage due to inhalation exposure to fine
particles. Thus, reducing emissions of total fuel oil particulate mass by means of combustion
modifications may have the unexpected result of increasing emissions of fine particles,
particularly those suspected of being closely associated with increased health risks.
-------
Introduction
Fine panicle matter (PM) in the atmosphere has been of considerable en\ ironmental interest in
recent years because of epidemiological studies which correlated short term exposure to ambient
levels of fine PM with acute adverse health effects [1, 2], These studies concluded that adverse
health effects were occurring at current ambient fine PM concentrations and became the basis for
revision of the National Ambient Air Quality Standards for PM that included a standard for PM
less than 2.5 /im in diameter (PM;5) [3],
In addition to the epidemiological studies associating increased levels of fine PM with adverse
health effects, research is also being conducted to identify the specific mechanisms that cause
these effects. Among the numerous hypothesized causal mechanisms are exposure to water-
soluble transition metals or to ultrafine particles (<0.1 /xm in diameter) [4-7], These metals may
preferentially partition to particles smaller than 1 in diameter and can even contribute to an
ultrafine mode [8]. Therefore, understanding the mechanisms controlling the formation and
speciation of submicron PM is important in the context of the current fine PM issue.
Background
Miller et al. [9] showed that PM25 samples from heavy oil burned in a fire-tube boiler contained
both a submicron mode formed by nucleation. coagulation, and condensation of vaporized metals
(the accumulation mode) and a coarse mode consisting primarily of low density, porous,
carbonaceous cenospheres, resulting in weakly bimodal behavior. The smallest particles,
captured using a cascade impactor (0.4 jum and smaller), showed that 50 to 90% of the mass was
composed of metals and sulfates, while particles larger than 2.5 /xm were composed of 50-80%
carbon [based on loss on ignition (LOI)]. The enrichment caused by metal
vaporization/condensation processes was diluted by variations in PM carbon content, which is
the major constituent of the PM25 by mass. Thus, in this case, metal enrichment depended on
carbon oxidation processes rather than on variations in metal volatility.
-------
Subsequent studies [10], burning the same fuel in a much hotter refractory-lined combustor,
produced a nearly unimodal particle size distribution (PSD) with a mean diameter of
approximately 0.1 pim and very few large particles. These ultrafine particles were composed
primarily of trace element species containing copper (Cu), iron (Fe), nickel (Ni), vanadium (V),
zinc (Zn), and sulfur (S). Additionally, these particles contained very little carbon (C) (based on
LOI), and the particulate bound S was speciated almost exclusively as sulfates.
This characteristic behavior of residual oil combustion is very different from the behavior of
pulverized coal combustion, in which improved C burnout does not necessarily significantly
impact the resulting PSD or the distribution of metals within the PSD. Coal contains
significantly more ash than does fuel oil, resulting in substantially lower C fractions for similarly
efficient combustion processes. Additionally, while essentially all the inorganic elements
associated with residual fuel oils are inherently bound within the oil's organic structure,
substantial quantities of the inorganic elements identified with coal ash are associated with
excluded materials and mineral inclusions within the coal particles. While mechanisms of
particle formation during the combustion of coal have been widely studied [11, 12], information
regarding particle formation mechanisms during fuel oil combustion, particularly for emulsified
fuels, is much more limited.
Emulsified Fuels
Interest in emulsified fuels, particularly water-in-oil emulsions, increased as emission limitations
led to development and application of cost-effective emissions controls for oil-fired combustion
equipment [13]. Water-in-oil emulsions are characterized by a continuous phase of oil, with
water uniformly dispersed throughout the oil in discrete droplets. Dryer [13] discussed the work
of Ivanov and Nefedov [14], which postulated that, when heated, the small droplets of water
(surrounded by a fuel oil of higher boiling point) would rapidly and disruptively vaporize and
expand, shattering the original emulsion droplet into many smaller droplets. Further work by
Dryer et al. [15] demonstrated that this secondary atomization resulted in very small fuel droplets
that devolatilize and burn out more quickly and completely than the larger fuel droplets produced
-------
by mechanical atomization. The secondary atomization and the presence of water allow heavy
fuels to be combusted more completely at lower peak temperatures and excess air levels than
would be possible with non-emulsified or "neat" fuels [13]. In addition to decreased C mass
emissions, the benefits of secondary atomization may also include decreases in nitric oxide (NO)
and carbon monoxide (CO) emissions as well as improved operability at lower excess air levels.
•Orimulsion® 400 differs from water-in-oil emulsions in that the hydrocarbon (bitumen) is the
dispersed phase and the water is the continuous phase. This 70% bitumen in 30% water
emulsion does not exhibit the classical microexplosion behavior, but does behave similarly due
to the small bitumen droplet sizes (roughly 8-24 ju.m) in the emulsion. When the water is rapidly
vaporized in the combustion zone, bitumen droplets remain that are much smaller than the
original Orimulsion® 400 droplets generated by the atomizing nozzle. While Orimulsion® 400
is not typically used for the specific purpose of reducing NO, CO. or PM emissions, its
combustion and emissions behavior are similar to those for non-emulsified heavy fuel oils [16,
17]. Furthermore, the small bitumen structures which remain after evaporation of the water can
result in very low unburned C levels compared to non-emulsified heavy fuel oils burned under
the same conditions. However, as has been shown for non-emulsified fuels [10], improved C
burnout in oil-fired systems has implications beyond simple reduction of PM mass emissions.
This paper examines the effect of emulsified fuels on the fine PM behavior and may have
implications for the relative toxicity of emitted particles.
Experimental Approach and Methods
A series of projects over the past several years at EPA's National Risk Management Research
Laboratory in North Carolina have examined particle behavior and PSDs from different types of
fuel oils and emulsified fuels in three experimental combustors. Test series A and B (see Table
1) were designed to compare air pollutant emissions from two emulsified water-in-oil fuels (a
No. 6 residua] oil and a No. 2 distillate oil, respectively) with the same non-emulsified oils [18,
19]. Test series B also included an emulsified naphtha. Test series C was designed to quantify
and compare the potential air emissions of the No. 6 residual oil with a bitumen-in-water fuel
-------
(Orimulsion® 400) [16]. Finally, test series D and E were part of an on-going project at EPA to
provide in-depth characterization of PM2? generated from combustion of fuel oil [9, 10],
These fuels were burned in three in-house experimental combustors including laboratory-scale,
pilot-scale, and small commercial-scale units. These combustion systems varied in design and
time-temperature characteristics. A small commercial-scale fire-tube package boiler, rated at
• 732 kW, was used for several of these tests. Due to its design and volumetric heat release rate,
this unit exhibited the highest heat transfer and gas quenching rates as is evident from the high
amounts of unbumed carbon (based on LOI) in the emitted fly ash. A pilot-scale 585 kW water-
wall package boiler simulator was used in the evaluations of the Orimulsion® 400. This unit had
a larger combustor volume and a smaller furnace heat transfer surface area and gas quenching
rates compared to the fire-tube boiler. Finally, a refractory-lined combustor rated at 73 kW was
used in series E. This furnace exhibited the highest operating temperatures and lowest
quenching rates, and was more characteristic of the temperature profiles of large utility and
industrial furnaces. Sampling was performed at stack locations at temperatures of 175-275, 370-
400, and 400 °C for the fire-tube boiler, water-wall boiler, and refractory-lined combustor,
respectively. Details of the designs of the three combustion systems are presented elsewhere [9,
20,21],
PSDs were measured by several methods. PSDs of sampled particles were determined by
electrical mobility using a Thermo-Systems Inc. scanning mobility particle sizer (SMPS), which
classifies and counts particles within a working range of 0.01 to 1.0 /im diameter. PM was also
sampled using an atmospheric-pressure in-stack cascade impactor designed to collect samples
smaller than about 15 fim in diameter on nine stages (including the afterfilter) for subsequent
gravimetric and chemical analysis. Sampling and analytical procedure details are described
elsewhere [8, 9, 22]. During series E experiments, an atmospheric-pressure 10-stage Micro-
Orifice Uniform Deposit Impactor (MOUDI) from MSP Inc. was used to provide enhanced PSD
resolution of smaller particles. Table 1 summarizes the combustors, fuels, and PM emissions for
each test series.
-------
Results and Discussion
Figure 1 presents cascade impactor data for three fuels (No. 6 oil, emulsified No. 6 oil, and
Orimulsion® 400) burned in the fire-tube and water-wall boilers. The top two panels are for a
No. 6 oil and an emulsified No. 6 oil bumed in the fire-tube boiler in test series D and A,
respectively. The bottom two panels are for a No. 6 oil and Orimulsion® 400 bumed in the
• water-wall boiler simulator during test series C. The non-emulsified No. 6 oils (series C and D)
both show PSDs with relatively large particle mass fractions in larger particle sizes (>10 /im in
diameter), while both the emulsified No. 6 oil and the Orimulsion® 400 (series A and C) have
very low mass fractions in this size range. Previous work examining No. 6 oil combustion PM
indicates that these larger panicles are composed primarily of carbonaceous material [9, 10],
This is confirmed by the LOI values (see Table 1) of 86 and 38% for the No. 6 oils, compared to
13% for the Orimulsion® 400. Note also, that the No. 6 oil burned in the water-wall boiler
produces ash with significantly lower LOI than from the fire-tube boiler. LOI for the emulsified
No. 6 fuel (series A) was not measured. Table 1 also includes corresponding PM mass emission
concentration for comparison.
Figure 2 presents submicron PSDs taken by the SNIPS for these four tests (series A, C, and D),
as well as for a No. 6 fuel oil bumed in the refractory-lined furnace (series E) and a No. 2 oil, an
emulsified No. 2 oil, and an emulsified fuel naphtha (series B). Note that the three panels in Fig.
2 compare submicron PSDs over three levels of volume concentrations. In the top panel, the No.
6 oil bumed in the refractory-lined combustor (series E) and in the water-wall boiler (series C)
show the greatest submicron volume concentrations and distinct unimodal submicron
accumulation modes. In comparison, the accumulation mode of the No. 6 oil bumed in the fire-
tube boiler is much smaller, and can be seen only when magnified in the bottom panel of Fig. 2.
In the center panel, results for the No. 6 fuel oil bumed in the fire-tube boiler are repeated, along
with an emulsified No. 6 oil bumed in the same fire-tube boiler (series A) and Orimulsion® 400
bumed in water-wall boiler (series C). Interestingly, the emulsified fuels all show distinct
accumulation modes between 0.01 and 0.1 /im, and the emulsified fuel oils (No. 6 and 2)
indicate a second submicron mode between 0.1 and 1.0 /im. In contrast, PSDs for these same
non-emulsified fuels indicate the beginnings of large coarse modes above 0.5 /im. The modes
-------
between 0.1 and 1 appear to be unique to the emulsified oils, suggesting that some aspect of
emulsified fuel combustion behavior is responsible for particle formation in this size range. The
most likely explanation is that of secondary atomization, which produces a range of secondary
fuel droplets smaller than the 40-70 /xm diameter particles typically generated by the mechanical
atomization of the nozzle [10]. This second submicron mode for the two water-in-oil emulsified
fuel oils are believed to be residual char particles which are the consequence of enhanced carbon
bumout of the smaller fractured fuel droplets. Interestingly, the bitumen-in-water Orimulsion
and the emulsified naphtha do not produce the same distinct bimodal submicron PSDs. This is
likely due to lack of secondary atomization for the Orimulsion® 400 and to increased fuel
volatility for the emulsified naphtha.
From these results, we can gain a better understanding of the processes governing particle
formation in oil-fired systems and how those processes impact particle size and composition.
We postulate that two principal processes govern the formation of particles in oil-fired systems,
and that these processes largely determine particle size and composition. In oil-fired systems,
larger particles are formed largely from unburned carbonaceous material such as cenospheres or
cenosphere fragments that are of the same order of magnitude in size as the original fuel droplets
[9]. For non-emulsified fuels, the size of the fuel droplets will be governed by nozzle and
atomization parameters, while for emulsified fuels, droplet size will also depend on the
secondary atomization caused by the microexplosions characteristic of these fuels.
In general, the impactor data (Fig. 1), the SMPS data (Fig. 2), and the PM mass emissions and
LOI values (Table 1) all support the findings that emulsification enhances carbon bumout and
increased submicron particle formation. Additionally, these data (in particular data from test
series C, D, and E) also show that combustor design and temperature quench rate directly affect
total PM emissions and LOI. However, it may be counter-intuitive to note that submicron
particle mass emissions are increased for emulsified fuels and for combustor designs and
temperature profiles that promote more complete fuel burnout.
Particles formed by nucleation, coagulation, and condensation of vaporized inorganic material
produce the accumulation mode seen near 0.1 /xm diameter or smaller [9, 10], Linak et al. [10]
-------
concluded that, in oil-fired systems, the magnitude of the accumulation mode depends strongly
upon the extent of carbon burnout. For heavy fuel oil combustion in a highly quenched
environment, the resultant unburnned carbon cenospheres retain considerable amounts of
organically bound metals and sulfur, substantially reducing the amount of vaporized material
available to form the submicron accumulation mode. In addition, the significant concentrations
of larger diameter particles provide condensation sites for inorganic elements that do vaporize,
'further reducing the mass available for nucleation. However, vapor-phase material that does
nucleate to form the accumulation mode is composed largely of metals and sulfur. This behavior
is significantly different than that for coal, for which the large amounts of included and excluded
inorganic material, much of which are alumino-silicates, act to tie up the trace elements that
might otherwise vaporize. For coal combustion, the release of metal does not depend upon the
degree of carbon burnout to nearly the same extent as for residual oil combustion. Therefore,
processes which enhance carbon burnout in coal systems will have a much smaller effect on the
submicron particle mass than for residual oil combustion.
The relationship between carbon burnout and submicron particle mass can be seen most clearly
in Fig. 3, which presents submicron panicle volume concentration as a function of total PM mass
and LOI. Because the level of unbumed carbon has such a strong influence on total PM mass,
the relationship between submicron particle mass and unbumed carbon is qualitatively the same
as between submicron particle mass and total PM. Submicron particle volume concentration in
Fig. 3 is calculated by integrating the areas under the dV/dlog(Dp) curves measured using the
SMPS (see Fig. 2) to obtain submicron volume concentration in /im3/cm3.
The emulsified No. 6 oil burned in the fire-tube boiler had a total PM mass approximately equal
to the No. 6 oil burned in the hotter water-wall boiler, but a slightly higher submicron volume.
Both measurements are consistent with improved carbon burnout. The improvement in carbon
burnout between the No. 6 and the emulsified No. 6 oils burned in the same fire-tube boiler is
similar to that seen between the No. 6 oil burned in the fire-tube boiler and the water-wall boiler.
The secondary atomization of the emulsified No. 6 oil results in significantly smaller initial fuel
droplets, enhanced carbon burnout, more extensive metal and sulfur vaporization, and, therefore,
-------
larger quantities of nucleated materials (see Fig, 2). A second result of secondary atomization is
that any unburned carbon particles from the emulsified oils are, in general, smaller than those
from the non-emulsified oils, and also contribute to the submicron mass (Fig. 2), This second
characteristic submicron mode between 0.1 and 1 /xm diameter for the emulsified oils (Fig. 1)
may also help explain the increase in submicron volume concentration for these fuels from the
fire-tube boiler compared to the non-emulsified oils bumed in the water-wall boiler (see Fig. 3).
For Orimulsion® 400, the submicron volume concentration is lower than that for the No. 6 oil
when both are burned in the water-wall boiler, While these two fuels are not directly
comparable, the lower submicron volume concentration may be due to differences in densities of
the submicron particles from the two fuels. In Fig. I, we note that the submicron mass
generated by Orimulsion® 400 is substantially higher than that of the No. 6 oil bumed in the
same boiler, while in Fig. 2 we see that the submicron volume concentration is slightly lower.
Table 1 also indicates a difference in particle composition; the No. 6 oil had a higher LOI value
than the Orimulsion® 400 for a similar total PM mass.
Finally, in addition to particle size, the speciation of the metals and sulfur in the accumulation
mode of PM from the combustion of these fuels is likely to affect their solubility and possibly
their bioavailability. Previous work [10] showed that the preferred equilibrium forms of the
transition metals of interest at flue gas conditions are sulfates which are, in general, more water-
soluble than the oxide forms of the same metals. Initial investigations into the speciation of
these accumulation mode and coarse mode particles by x-ray absorption fine structure
spectroscopy [23] indicate that the accumulation mode particles are composed of metal sulfates
while the coarse mode particles contain larger quantities of thiophenic sulfur. Particle speciation
and solubility may be significant in terms of the potential particle toxicity, based on the work of
Dreher et al. [4-6].
Conclusions
-------
Improved carbon burnout in fuel oil combustion, either by decreasing the temperature quench
rate or by forming smaller fuel droplets through the secondary atomization characteristic of oil-
water emulsions, increases the volume of the submicron particle fraction. In contrast to larger
coarse mode particles which are composed largely of carbon char and inherently bound metals
and sulfur, these submicron particles appear to be composed of metal sulfates that are more
water-soluble than the larger coarse mode particles. A series of studies using different fuel oils,
• including oil-water emulsions, burned in different combustor designs show that the use of water-
in-oil emulsions can increase the submicron particle volume compared to a non-emulsified oil
burned in the same combustor. The degree of difference is similar to that exhibited by a non-
emulsified fuel burned in a high-quench-rate fire-tube boiler compared to the same fuel burned in
a significantly lower quench-rate water-wall boiler.
For fuel oils, submicron particle volume concentration varies directly with carbon burnout, and
inversely with total particle mass. These submicron particles are formed by vaporization and
subsequent nucleation, coagulation, and condensation mechanisms. Where normal atomization,
high quench rates, or other obstacles to complete combustion exist, substantial amounts of
inorganic material remain bound with the unburned carbon char. This material is, therefore,
unavailable to form the metal-sulfate-enriched submicron particle fraction.
These results may have significance relative to combustion or fuel modifications designed to
reduce unburned carbon levels. Health effects studies indicate that water-soluble transition
metals, such as those found in the submicron fraction of heavy fuel oil fly ash, may play a key
role in explaining possible mechanisms of pulmonary damage due to inhalation exposure to fine
particles. Thus, reducing emissions of total fuel oil particulate mass by means of combustion
modifications may have the unexpected result of increasing emissions of fine particles,
particularly those suspected of being closely associated with increased health risks.
Acknowledgements/Disclaimer
Portions of this work were conducted under EPA Contract 68-C-99-201 with ARCADIS
Geraghty & Miller. The research described in this article has been reviewed by the Air Pollution
Prevention and Control Division, U.S. EPA, and approved for publication. The contents of this
-------
presentation should not be construed to represent Agency policy nor does mention of trade
names or commercial products constitute endorsement or recommendation for use.
References
1. Wilson, R., and Spengler, J.D., eds., Particles in Our Air: Concentrations and Health
Effects, Harvard Univ. Press, Cambridge, MA (1996).
2. Bachmann, J.D., Damberg, R.J., Caldwell, J.C., Edwards, C, and Koman, P.D., "Review
of the National Ambient Air Quality Standards for Particulate Matter: Policy Assessment of
Scientific and Technical Information," EPA-452/R-96-013 (NTIS PB97-115406), Office of Air
Quality Planning and Standards, Research Triangle Park, NC, July 1996.
3. Federal Register, 62 FR 38652, July 18, 1997.
4. Dreher, K., Costa, D., Hoffman, A., Bonner, J., and Osornio-Vargas, A., "Pulmonary
Toxicity of Urban Air Particulate Matter (PM)," Air & Waste Management Association Meeting
on Measurement of Toxic and Related Air Pollutants, Research Triangle Park, NC, 1996.
5. Dreher, K., Jaskot, R., Richards, J.H., Lehmann. J.R., Winsett, D., Hoffman, A., and
Costa, D., Amer. J. Resp. Crit. Care Med153, (4 :A15) (1996).
6. Dreher, K., Jaskot, R., Lehmann, J.R., Richards, J.H., McGee, J.K., Ghio, A.J., and
Costa, D.L., J. Toxicol and Environ. Health, 50:285-305 (1997).
7. U.S. Environmental Protection Agency, "Air Quality Criteria for Particulate Matter,"
EPA-600/P-95/001 (NTIS PB96-168224), National Center for Environmental Assessment,
Research Triangle Park, NC, 1996.
8. Linak, W.P., and Wendt, J.O.L., Fuel Process. Technol., 39, 173-198 (1994).
-------
9. Miller, C.A., Linak, W.P., King, C., and Wendt, J.O.L., Combust. Sci. TechnoL, 134:477
(1998).
10. Linak, W.P., Miller, C.A., and Wendt, J.O.L., "Fine Particle Emissions from Residual
Fuel Oil Combustion: Characterization and Mechanisms of Formation," submitted to the Twenty-
Eighth Symposium (International) on Combustion, The Combustion Institute, Pittsburgh, PA,
• 1999.
11. Damle, A.S., Ensor, D.S., and Taylor, D.D., Aerosol Sci. TechnoL, 1:119-133 (1982).
12. Linak, W.P., and Peterson, T.W., Twenty-First Symposium (International) on
Combustion, The Combustion Institute, Pittsburgh, PA, pp. 39-410, 1984.
13. Dryer, F.L., Sixteenth Symposium (International) on Combustion, The Combustion
Institute, Pittsburgh, PA, pp. 279-295, 1976.
14. Ivanov, V.M., and Nefedov, P.I., Trudy Instituta Goryachikh Iskopayemykh, Vol. 19
(1962 - in Russian), Translation: "Experimental Investigation of the Combustion Process of
Natural and Emulsified Liquid Fuels," NASA TTF-258, 1965.
15. Dryer, F.L.. Rambach, G.D., and Glassman, I., "'Some Preliminary Observations on the
Combustion of Heavy Fuels and Water-in-fuel Emulsions," presented at the Central Section
Meeting of the Combustion Institute, Columbus, OH, 1976.
16. Miller, C.A., Srivastava, R.K., and Hall, R.E., ' Comparison of Air Pollutant Emissions
from the Combustion of Orimulsion® and Other Fossil Fuels," poster presented at the EPRI-
DOE-EPI Combined Utility Air Pollutant Control Symposium, Atlanta, GA , 1999.
17. Marcano, N„ Pourkashanian, M., and Williams, A., Fuel, 70:917-923 (1991).
-------
18. Miller, C.A., "Hazardous Air Pollutants from the Combustion of an Emulsified Heavy
Fuel Oil in a Firetube Boiler," EPA-600/R-96-019 (NTIS PB96-168281), National Risk
Management Research Laboratory, Research Triangle Park, NC, February 1996.
19. Miller, C.A., "Verification Testing of Emissions from the Combustion of A-55® Clean
Fuels in a Firetube Boiler," EPA-600/R-98-035 (NTIS PB98-142169), National Risk
•Management Research Laboratory, Research Triangle Park, NC, April 1998.
20. Linak. W.P., McSorley, J,A., Hall, R.E., Srivastava, R.K., Ryan, J.V., Mulholland, J.A.,
Nishioka, M.G., Lewtas. J., and DeMarini, D.M., Haz. Waste & Haz. Marls., 8:1-15 (1991).
21. Linak, W.P., Srivastava, R.K., and Wendt, J.O.L., Combust. Sci. Technol., 101:7-27
(1994).
22. Scotto, M.A.. Peterson, T.W., and Wendt. J.O.L., Twenty-Fourth Symposium
(International) on Combustion, The Combustion Institute. Pittsburgh, PA, pp. 1109-1118, 1992.
23. Huggins, F.E., and Huffman, G.P., J. Hazardous Materials, in press (1999).
-------
List of Figures
Fig. 1 Particle mass distributions determined by cascade impactor.
Fig. 2 Submicron particle volume distributions determined by SMPS,
Fig. 3 Comparison of Total PM mass emissions and LOI with total submicron volume
emissions. The line represents a linear fit of the No. 6 oil data.
-------
Tabic I. Summary of tests. fuel analyses, and PM emissions.
Test Series
A
B
C
D
E
Combustor
Fire-tube boiler
Fire-tube boiler
Water-wall boiler
Fire-tube boiler
Refractory-lined
combustor
Fuels
No. 6 Oil
Emulsified
No. 6 Oil
No. 2 Oil
Emulsified
No. 2 Oil
Emulsified
Naphtha
No. 6 Oil
Orimulsion
400®
No. 6 Oil
No. 6 Oil
C
85.49
77.83
86.92
57.40
53.36
86.45
58.12
85.49
85.49
H
10.36
10.16
13.01
8.77
9.16
10.23
7.14
10.36
10.36
0
0.92
1.10
0.42
2.42
1.10
0.90
3.35
0.92
0.92
N
0.35
0.24
0.49
0.48
0.32
0.26
0.17
0.35
0.35
S
2.33
1.70
0.03
0.01
0.002
2.07
2.23
2.33
2.33
h2o
0.50
9.00
0.05
30.90
36.07
0.70
28.92
0.50
0.50
Ash
0.10
0.07
0.001
0.003
-0
0.08
0.07
0.10
0.10
PM Emissions
niK/m1
230
160
12
5
3
156
155
184
93
LOI'
%
NM2
NM
~0
~0
~0
38
13
86
-0
'Loss on ignition.
2Not measured.
-------
w
w
CO
E
"5
c
o
'¦*—>
o
-------
100000-
80000-
70000-
60000
50000-
40000
30000-
20000-
10000-
0
25000-
;
No. 6 Fuel Oil (D-Fire-tube
Boiler)
Orimulsion 400 (C-Water-wall
Boiler)
No. 6 Fuel Oil (C-Water-wall
Boiler)
A No. 6 Oil (E-Refractory Furnace)
;
A
A
~
A
~
A
~
A
^
A
i
k
A
A
i ¦
Ml M
Ik _ _JV
a 15000-
O)
o
T3
> 10000
T2
5000-
0-
1000-
900-
¦ No. 6 Fuel Oil (D-Fire-tube Boiler)
; + Orimulsion 400 (C-Water-wall Boiler)
¦ • No. 6 Fuel Oil (C-Water-wall Boiler)
- ~ Emulsified No. 6 Oil (A-Fire-tube Boiler)
¦
v ¦
1 i w
- m. 1 4t
%
•
~ X ¦
£
£ *• J ***
+-h-'V
m No, 6 Fuel Oil (D-Fire-iube
Boiler)
O #2 Oil (B-Fire-tube Boiler)
Emulsified #2 Oil (B-Fire-tube
Boiler)
Emulsified Naphtha (B-Fire-tube
Boiler)
O
O
O
idi * -
0.01
0.1
Particle size, pm
Figure 2 ,
-------
100000
LOI
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
' ) ¦ \ * * i *,' ¦ ' i
10000
1000
Refractory
Furnace
Water-wall
Boiler
Fire-tube
Boiler
o (No. 2 Oil)
~
100
Boiler
¦ No. 6 Oil
• Emuls 6 Oil -
+ Ori 400
~ No. 2 Oil
o Emuls 2 Oil
x LOI, No. 6 Oil
0 20 40
60 80 100 120 140 160 180 200
Total PM, mg/dscm
Figure 3.
-------
*-»rr-t-» f. >10 TECHNICAL REPORT DATA
N RM xiL- K1 r- p- 4 to (Please read Instructions on the reverse before completing)
1. REPORT NO. 2.
EPA/600/A-00/070
3. RECIPI ENT*S ACCESSION NO.
4. TITLE AND SUBTITLE
The Influence of Carbon Burnout on Sub micron
Particle Formation from Emulsified Fuel Cil
Combustion
S, REPORT DATE
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
C. Andrew Miller and William P. Linak
8, PERFORMING ORGANIZATION REPORT NO,
9. PERFORMING ORGANIZATION NAME AND ADDRESS
See Block 12
10. PROGRAM ELEMENT NO.
11. CONTRACT/GRANT NO.
68-C-99-201 (ARCADIS
Geraghty and Miller)
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; 8/97-12/99
14. SPONSORING AGENCY CODE
EPA/600/13
16.supplementary notes APPCD project officer is C. Andrew Miller. Mail Drop 65, 919 /
541-2920. Presented at 28th Int. Symp. on Combustion, Edinburgh, Scotland, 7/30-
8/4/00.
16. abstract The paper giyes results of an examination of particle behavior and particle
size distributions from the combustion of different fuel oils and emulsified fuels in
three experimental combustors. Results indicate that improved carbon (C) burnout
from fuel oil combustion, either by decreasing the temperature quench rate or by
forming smaller fuel droplets through the secondary atomization characteristic of
oil/water emulsions, increases the volume of the sub micron particle fraction. Also,
the use of oil/water emulsions can increase the submicron particle volume comparec
to a non-emulsified oil burned in the same combustor. In contrast to larger coarse-
mode particles which are composed largely of C char and inherently bound metals
and sulfur, these submicron particles appear to be composed on metal sulfates that
are more water-soluble than the larger coarse-mode particles. For fuel oils, sub-
micron particle volume varies directly with C burnout, and inversely with total par-
ticle mass. These metal-sulfate-enriched submicron particles are formed by vapor-
ization and subsequent nucleation, coagulation, and condensation mechanisms. Whert
normal atomization, high quench rates, or other obstacles to complete combustion
exist, substantial amounts of inorganic material remain bound with the unburned C
char.
17. key words and document analysis
a. DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. cosati Field/Group
Pollution Burnout
Fuel Gil
Combustion
Emulsions
Particles
Carbon
Pollution Control
Stationary Sources
Particulate
13 B
21D
2 IB
Q7D
14G
07B
18. DISTRIBUTION STATEMENT
Release to Public
19, SECURITY CLASS (ThisReport}
Unclassified
21. NO. OF PAGES
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |