xk"'
United States
Environmental Protection
Agency
Hazardous Waste Engineering
Research Laboratory
Cincinnati OH 45268
Research and Development
EPA/600/S2-85/115 Dec. 1985
&ERA Project Summary
Assessment of Organic
Contaminants in Emissions from
Refuse-Derived Fuel
Combustion
James Chrostowski, Dallas Wait, Eleanor Kwong, Ann Jefferies, Cheryl
Rodgers, and Neil Mosesman
Organic contaminants in emissions
from refuse-derived fuel combustion
were investigated in a 20-inch-diameter
atmospheric fluidized-bed combustor.
Combinations of coal/Eco- Fuel/ MSW/
toluene were burned in the combustor
with temperatures ranging from 1250
to 1550°F. A Source Assessment
Sampling System (SASS) was used to
sample the stack gas; Level 1 method-
ology was used to analyze the organic
contaminant levels. Combustion effi-
ciences of 93 to 98 percent were
achieved in the test burns. Combustion
of the Eco-Fuel generated fewer organic
emissions than combustion of coal at
similar combustion temperatures. The
fine particulate collected by the SASS
train filter contained higher concentra-
tions of extractable organics than the
reactor fly ash and the SASS cyclone
samples. Combustion of a toluene/Eco-
Fuel mix generated a large number of
benzene derivatives not seen in the
combustion of pure Eco-Fuel. Polycy-
clic aromatic hydrocarbons were the
dominant organic compounds con-
tained in the XAO-2 resin extract from
coal combustion. A number of different
priority pollutants were identified in the
samples collected.
This Project Summary was developed
by EPA's Hazardous Waste Engineering
Research Laboratory, Cincinnati, OH,
to announce key findings of the research
project that is fully documented in a
separate report of the same title (see
Project Report ordering information at
back).
Program Objectives
The primary objectives of this study
were to assess organic and trace metal
emissions from the combustion of refuse-
derived fuels (RDF) in an atmospheric
fluidized-bed combustor (AFBC). A total of
four fuels were used in the test program,
namely coal, Eco-Fuel, Municipal Solid
Waste (MSW), and reagent grade toluene.
A Source Assessment Sampling (SASS)
was used to collect the samples (solid and
liquid) for organic and trace metal anal-
ysis.
Because of funding limitations for the
program, the SASS samples were anal-
yzed for organic content only and no trace
metal analyses were performed. A Level
1 approach was used to characterize the
organic content of the RDF combustion
effluents.
Description of Equipment
A schematic of the 20-inch diameter
AFBC is shown in Figure 1. The solid fuel
is shredded to the desired size in the
William's hammermill. The size feedstock
is then loaded into a feed hopper. The
material is then fed through a rotary valve
into an Acrison metering screw which
directly discharges into the lower section
of the fluidized bed. In the run in which
liquid toluene was added to the reactor, a
tank was pressurized with N2 and the
liquid toluene was metered through a
rotameter and then injected directly into
the bed through a small orifice. Combus-
tion air is provided by a positive displace-
-------
ment blower and is metered by an Erdco
flowmeter which was previously cali-
brated with SO2. A start-up burner is used
to preheat the bed sand to 800-900°F
before the introduction of the solid fuel.
Thefluidized-bed reactor is a refractory-
lined vessel 20 inches in inside diameter
and approximately 20 feet high. A per-
forated Inconel plate is used as the com-
bustion air distributor. Refractory grog, a
high AI203 material, is employed as the
bed material.
Description of Feedstocks for
Combustion Testing
The three solid feedstocks employed in
the AFBC testing were Battelle Coal, Eco-
Fuel provided by Combustion Equipment
Associates, and MSW provided by Ameri-
cology. Reagent grade toluene was also
used to simulate combustion of a haz-
ardous waste. The properties of the solid
fuels (size, HHV, moisture, ultimate anal-
ysis, and fusion temperature) are sum-
marized in Table 1.
The test runs were performed with
feedstock mixes shown in Table 2.
Table 1. Coal RDF Suppliers and Fuel Properties
Supplier
Crushed Coal
Battelle
Feedstock
Eco-Fuel
Combustion
Equipment
Associates
Americology-MSW
American Can
4
Size
Heating Value
(HHV.Btu/lbm.f.)
Moisture (% wet basis)
Ultimate Analysis
(wt%. m.fj
Sulfur
Carbon
Hydrogen
Nitrogen
Oxygen
Chloride
Ash
Fusion Temperature f°F)
23%>1.7 urn
59% mid-range
18%>300 urn
12953
8.11
1.4
74.2
4.9
1.3
9.0
9.2
27%<170mesh
30% 170/230 mesh
42%230/325 mesh
7827
4.7
0.61
40.0
4.6
<0.1
45.5
0.26
9.0
-1800
6430
26.6
0.3
51.0
6.8
0.7
26.0
15.2
-1300
Freeboard Upper Temperature
/
Freeboard Cooling i
Freeboard Lower Temperature
>v
Gas Analysis
fOz. SOa
/VO* CO. GC)
Rotary Valve
l\£_^t__2I—^BBM
Acrison Screw
Preheat Burner —W"
• Air
Fluid/zing Air Blower
Figure 1. Pilot-plant schematic.
Flyash
In-Bed Cooling Tubes
M^M.
-«.—__
— Fluid Bed Temperature
Inconel Distributor
-------
Table 2. Feedstock Mix Composition
Run ID Coal Eco-Fuel
MSW
Toluene
Total
17
18
22
23
24
25
50
100
100
50
—
--
50
..
..
50
100
82
100
100
100
100
100
18 100
Organic Analysis—Level 1
Procedure
Analysis of the SASS samples and the
reactor fly ash sample was performed by
the Level 1 methodology developed by
EPA-IERL/RTP (1978). The samples
which were analyzed included the (1)
XAD-2 resin, (2) combined cyclones, (3)
particulate filter, (4) organic trap conden-
sate, (5) reactor fly ash, and (6) the
feedstocks. A summary of these methods
is shown in Figure 2.
Extracts were first analyzed by flame
ionization gas chromatography to deter-
mine the total chromatographable organ-
ics (TCO) present.
After the initial TCO screen, the extract
is concentrated and a gravimetric deter-
mination is done to quantitate organics
with boiling points higher than 300°C. If
the sum of the gravimetric data and TCO
is less than 15 mg, indicating minimal
organic contamination, then infrared
spectroscopy (IR) is performed on the total
extract, thus completing the analysis. The
IR spectrum indicates the types of func-
tional moieties present in the sample, as
well as providing a control for subsequent
sampling. The IR analyses were per-
formed on a double-beam infrared spec-
trophotometer equipped with a microproc-
essor and NaCI plates. The scans were
performed from 4000 cm'1 to 200 cm"1 at
a scan rate of 12 minutes. A polystyrene
film was used to calibrate the instru-
ments.
If the sum of the gravimetric data and
TCO is greater than 15 mg, the extract
must be fractionated into a similar chem-
ical class according to polarity, and each
fraction is subjected to TCO and grav-
imetric analysis and depending on those
results, is analyzed by IR and capillary
GC/MS. It should be noted that IR
interpretation of complex samples is
prone to errors. For this reason the
GC/MS analysis is useful in providing
corroborative information for IR, as well
as supplemental data for additional iden-
tification.
Organic Analysis—Level 1
Results
A preliminary examination of the total
extractable content indicates that a ma-
jority of the RDF combustion samples
contains relatively low amounts of char-
acterizable organics. The values for the
sum of the gravimetric data and TCO
were used to determine whether there
was an adequate quantity of organics
present to employ the column chroma-
tography fractionation scheme and sub-
sequent GC/MS analysis. A summary of
the relative quantities of organics present
in the RDF samples is given in Table 3.
Regardless of organic content, all feed-
stock extracts were fractionated and
analyzed to provide background informa-
tion for data interpretation.
Substantial quantities of organics were
generally found only in the XAD-2 resin,
which was used to filter the combustion
vapor. Interestingly, quite a few of the
organics identified here have previously
not been detected in combustion efflu-
ents. A review of the organics found in
each RDF combustion experiment follows:
Pure Coal Burned at 1500°F
(Run #18)
In terms of concentration, most of the
organics detected in the combustion ef-
fluent of this burn resided on the XAD-2
resin, and to a lesser extent on the partic-
ulate filter. The fractionation scheme
indicates organics to be present through-
out the full polarity range. Polycyclic
aromatic hydrocarbons (PAH) were the
dominant organic compounds contained
in the XAD-2 resin extract, which is not
surprising, since PAH are recognized
constituents of coal, as well as the by-
product of combustion processes. This is
of special interest since many PAH are
known chemical carcinogens and many
are considered by EPA to be priority
pollutants. Other chemical classes de-
tected include some heterocyclic com-
pounds, cyanoaromatics, phenols, and
oxygenated aromatics (e.g., benzaldehyde
and anthraquinone). A GC/MS screen
was also performed on the total extract
derived from the particulate filter. This
sample contained PAH compounds of
higher molecular weight than those seen
in the XAD extract, such as benzofluor-
anthenes, benzopyrenes, and perylene.
Pure Coal Burned at 1250°F
(Run #22)
The combustion of coa I at lower temper-
atures than the preceding burn produced
effluents with lower organic content. This
may be attributable to the reduced fluid-
izing velocity for this run compared to run
#18. The only GC/MS analysis performed
on this sample was the total extract from
the organic trap condensate. PAH were
again quite evident. A very unusual com-
pound, dibenzoheptafulvne, was also
detected. This compound has not yet
been recorded in the literature as being
present in combustion emissions.
Pure Eco-Fuel Burned at 1500°F
(Run #24)
The organic composition of the Eco-
Fuel refuse material was primarily ali-
phatic hydrocarbons, as well as trace
quantities of organic acids and esters.
After combustion, GC/MS analysis of the
cyclone material and the organic trap
condensate indicated that significant
quantities of phenol and benzoic acid
were present. Both compounds are likely
oxidative degradation products of lignin
(polymer of phenylpropanoids). Lignin is
the main constituent of wood, and thus is
present in the paper waste contained in
the Eco-Fuel refuse. Other compounds
that may be indirectly generated from
lignin include chlorophenol, bromophe-
nol, phthalide, cumene, and chromone.
PAH compounds and a few nitrogen
heterocyclic compounds were also pres-
ent. Nitrogen heterocyclics, which were
detected in the trap condensate, are also
known to be found in combustion pro-
cesses and in some cases are suspected
carcinogens.
50 Percent Eco-Fuel/'50
Percent Coal Burned at 1500°F
(Run #17)
A majority of the organic emittants
produced from the combustion of this
feedstock combination were contained
on the XAD-2 resin. A variety of com-
pounds were detected in this extract.
Compound classes included aliphatic
hydrocarbons, PAH, heterocyclics, oxy-
-------
Organic Extract
Concentrate
Extract
Infrared A nalysis
TCO Analysis
Gravimetric
Analysis
Repeat TCO
Analysis if
Necessary
Aliquot Containing
15-100 mg
Solvent Exchange
Liquid
Chromatographic
Separation
Seven Fractions:
Infrared A nalysis
SE-52 Capillary
GC/MS Analysis
TCO and
Gravimetric
Analysis
Figure 2. Level 1 organic analysis methodology (EPA, 1978).
4
genated aromatics, and aromatic nitrites.
Most of these compounds were originally
present in the combustion effluent of
either the pure coal or pure Eco-Fuel.
£'co-FuelV Toluene Burned at
1550°F(Run#25)
The addition of toluene to the Eco-Fuel
refuse generated a large number of
benzene derivatives not seen in the
combustion of pure Eco-Fuel. The list of
new compounds includes bi-, tri-, and
tetrachloro isomers of benzene, diphenyl
methane, terphenyl isomers, benzoic acid
esters, and thiobenzoic acid. Trichloro-
benzene is currently included on EPA's
priority pollutant list. Some alkylated PAH
compounds were also unique to this
combustion effluent, as well as some
keto-aromatics (fluorene, benzophene,
and acetophenone).
50 Percent MSW/50 Percent
Coal Burned at 1250°F
(Run #23)
The combination of MSW and coal
seems to have generated lower quanti-
ties of organics in the SASS than the
combination of Eco-Fuel and coal. Quali-
tatively the MSW feedstock contained
many of the same organic acids and acid
derivatives present in the Eco-Fuel; how-
ever, the large qua ntity of aliphatic hydro-
carbons present in the Eco-Fuel refuse
was not present in the MSW. In the
combustion of MSW coal, all organics
identified in the cyclones were similar to
those identified in the Eco-Fuel coal burn.
However, in the condensate there were
many unique compounds identified, in-
cluding isophorone, 2, 4-dimethylfuran,
2-formylimidazole, 2-methyl hydroqui-
none, nitrobenzene, and 1,3-dlcyanoben-
zene. Isophorone is currently listed as a
priority pollutant.
Conclusions
• Combustion efficiencies of 93 to 98
percent were achieved with the var-
ious coal/Eco-Fuel/MSW/toluene
feedstock mixes in the 20-inch-diam-
eter fluidized-bed reactor. Reaction
temperatures ranged from 1250° to
1550°F and fluidization velocities
ranged from 2.7 to 4.4 fps.
• Most of the extractable organic emis-
sions collected by the SASS train were
collected on the XAD-2 resin and in
the organic module condensate.
• The extractable organic concentration
of the particulate from the SASS train
-------
filter is higher than that of the reactor
fly ash and the SASS combined cy-
clones.
There is no correlation between the
total extractable organics emitted per
pound of feedstock compared with the
extractable organic content of the
feedstock. However, the extractable
organics collected by the XAD-2 resin
and the organic module condensate
(per pound of feedstock) are clearly
less for the pure Eco-Fuel run (#24)
compared to the runs with either pure
coal (#18) or 50/50:Eco-Fuel Coal
(#17).
Reducing the coal combustion temper-
ature from 1500°F (Run #18) to
1250°F (Run #22) increased the con-
centration of extractable organics in
the reactor fly ash.
Adding toluene to the Eco-Fuel did not
increase the total organic emissions
(per pound of feedstock), but did result
in significantly more emissions asso-
ciated with the XAD-2 resin and the
organic module condensate samples.
Polycyclic aromatic hydrocarbons
(PAH) were the dominant organic
compounds contained in the XAD-2
resin extract from the coal combustion
runat1500°F(Run#18).
i In the combustion of pure Eco-Fuel at
1500°F (Run #24), reduced amounts
of PAH compounds were found. Signif-
icant amounts of phenol and benzole
acid were identified.
i Adding toluene to the Eco-Fuel refuse
generated a large number of benzene
derivatives not seen in the combustion
of pure Eco-Fuet. The list of new
compounds includes bi-, tri-, andtetra-
chloro isomers of benzene.
Table 3. Relative Amounts of Organics in SASS
Run ID
18
22
24
17
25
23
Experiment
Pure Coal at 1500°F
Pure Coal at 1250°F
Pure Eco-Fuel at 1500°F
50% Eco-Fuel/
50% Coal at 1 500° F
Eco-Fuel/
Toluene at 1550°F
50% MSW/
50% Coal at 1250°F
XAD-2
Resin
A
B
B
A
A
B
Combined
Cyclones
B
B
B
B
B
B
Paniculate
Filters
B
B
B
B
B
B
Trap
Condensate
B
B
B
B
A
B
Fly Ash
B
B
B
B
B
B
Key
A = Organics present in substantial quantities. Full Level 1 protocol used.
B - Organics present in relatively low concentration.
•&U. S. GOVERNMENT PRINTING OFFICE:1986/646-l 16/20733
-------
James Chrostowski, Dallas Wait, Eleanor Kwong. Ann Jefferies, CherylRodgers,
and Neil Mosesman are with Energy Resources Co., Inc., Cambridge. MA
02138.
Michael Black is the EPA Project Officer (see below).
The complete report, entitled "A ssessment of Organic Contaminants in Emissions
from Refuse-Derived Fuel Combustion," (Order No. PB 86-110 921 /AS; Cost:
$16.95, subject to change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield, VA 22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Hazardous Waste Engineering Research Laboratory
U.S. Environmental Protection Agency
Cincinnati, OH 45268
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Official Business
Penalty for Private Use $300
EPA/600/S2-85/115
«»
"10N
CHICAGO
------- |