v>EPA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S7-81-100 Aug. 1981
Project Summary
Evaluation of Relative
Environmental Hazards from a
Coal Gasifier
S. K. Gangwal, J. G. Cleland, and R. S. Truesdale
During the past 4 years, a laboratory-
scale coal gasification facility was
developed and used to study the gen-
eration and environmental assessment
of pollutants from coal gasification
operations. Detailed chemical analy-
ses of the four effluent streams, namely
gas, aqueous condensate, tar, and
ash, were performed for more than 30
runs in which a variety of coals ranging
from lignite to bituminous were gasi-
fied.
Brief descriptions are given for the
gasification reactor and the associ-
ated sampling and analysis system.
Problems encountered with analysis
and special techniques for analysis of
complex samples are described. The
relative environmental hazards of the
various effluent streams are deter-
mined, using multimedia environmen-
tal goals (MEG) methodology. Toxic-
ity and mutagenicity of the streams
are assessed using bioassays.
More than 400 constituents are
identified in the various effluent streams.
Environmentally significant and non-
significant constituents are ranked
according to discharge severity. The
nonsignificant constituents are candi-
dates for elimination in future research.
Finally, data from the laboratory gasi-
fier are discussed in relation to those
reported on large-scale processes.
This Project Summary was devel-
oped by EPA's Industrial Environ-
mental Research Laboratory. Research
Triangle Park. NC. 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.
Introduction
Under the sponsorship of the Fuel
Process Branch of the U.S. Environmen-
tal Protection Agency's Industrial Envi-
ronmental Research Laboratory, Re-
search Triangle Park, N.C., Research
Triangle Institute (RTI) conducted an ex-
perimental research program to study
pollution problems associated with coal
gasification. A laboratory-scale coal
gasification reactor and sampling system
was installed and more than 60 tests
were conducted using a variety of U.S.
coals. The gasifier was operated with
coal, steam, and air at about 1000°C
and 200 psig to yield a low-Btu gas as a
primary product and aqueous conden-
sate, tar, and ash (reactor residue) as by-
products. Each of these four product and
by-product streams was characterized
in detail,"using modern chemical analysis
and bioassay techniques. More than
400 constituents were identified in the
effluent streams and some 100 con-
stituents were quantitated for more
than 30 gasification tests.
Using the Environmental Protection
Agency's multimediaenvironmental
goals (MEG) methodology, the quantita-
ted constituents were grouped into
insignificant and significant constituents
from an environmental (health) hazard
point of view. The significant constitu-
-------
ents were than ranked according to
their severity of discharge. Ranges, 95
percent confidence intervals, arithmetic
means, and geometric means for the
stream concentrations and production
factors (amount produced per unit
amount of coal gasified) of the significant
constituents were plotted as bar charts.
Data for the stream production factors
from the RTI gasifier were compared to
those available in the literature for
large-scale gasifiers. Also, data for the
significant stream constituent concen-
trations and production factors were
compared to those available in the
literature for larger-scale gasifiers, coke
plants, coal liquefaction units, combus-
tors and incinerators, ambient back-
ground levels, and regulated levels.
Results and Discussion
The complete report which this pub-
lication summarizes contains data for
significant stream constituents from
more than 30 gasification tests. These
results are presented as bar graphs
(including minimum, maximum, 95
percent confidence interval, arithmetic
mean, and geometric mean) of concen-
trations and production factors. Typical
figures appearing in the report for
concentration and production factors
are shown here as Figures 1 and 2 for
the product gas stream.
In Figure 1, concentrations of most
significant product gas constituents are
plotted versus their DMEGs. Here DMEG
is a health-based stream concentration
and is defined as a concentration of a
pollutant in an undiluted effluent stream
which will not adversely affect people
exposed for short periods of time. DMEG
is generally derived using existing
toxicity and biological data; a complete
explanation is available in the report.
Here it is sufficient to say that the lower
the DMEG value is for a certain pollutant,
the more hazardous is the pollutant. In
Figure 1, the 45° line (dotted) corre-
sponds to concentration - DMEG. The
concentration bars include the mini-
mum, maximum, arithmetic mean, geo-
metric mean, and 95 percent confidence
interval for the product gas constituents
of greatest significance. The signific-
ance of the 45° line is that it allows easy
visual inspection of the relative degree
of significance and relative hazard of the
various constituents. The higher the
bars or means are above the 45° line the
greater are the hazards. Also constitu-
ents plotted on the left of Figure 1 have
lower DMEGs and are consequently
5
I
o
O
o
10'
10s
107
706
W5
10*
JO3
w2
CT"
Q
»*
"o
o J
r <*>
u
< o _ _
00 Uj c*T ' 4 '
I
'
i 4
^
T -S-
a x " -s T i-
(D ;;.
5 S n 7 § T ± T §
c^
5 C " II ^ « -Q
V m S ^ "^ I , '
O 0 13,
- -e $<
_ W3
II <-> ">
"
CO
L ^J
5 ^
Q -S '
--. "S t
.0 £ j
C ^
"* QJ -C
: 1 i
Uj T
-r-
-
i k
- " u
4>
-
f
*
/
/
a
c
o
N V,
k Cj
^
4
1
f
b
111:: i/x :
OS G ^
^ m """" -C
;l 1 1 /x|
/* "i
*
^
"j
r
/
-»»;
S
2
^
i
i \
"\ f .
/
/
s
'
.a_
05 s
1
, 1
s
'
-5'
>s
/ '"c
/
k
'% '
r/ §
3
- -1- Maximum Values
<£ T 95% Confidence Limits
\\
C" ^J
10*
10"
105
10°
W7
DMEG (fjg/m3)
Figure 1. Ranges of concentration for most significant product gas constituents.
more hazardous at the same concentra-
tion levels.
In Figure 2, the production factors
(expressed as ug produced per g of coal
gasified) of the significant product gas
constituents are shown. Again minimum,
maximum, arithmetic mean, geometric
mean, and 95 percent confidence in-
terval are shown on production factor
bar graphs. The constituents in Figure 2
are listed in order of increasing discharge
severity from left to right. Here discharge
severity is defined as the arithmetic
mean concentration divided by DMEG
and represents a hazard factor associ-
ated with each constituent. Thus, CO
(the product that one is trying to maxi-
mize) represents the highest product gas
hazard. Figure 2 gives an immediate
account of how much of a constituent to
expect from the coal gasifier, i.e., how
much of a by-product can be produced
(e.g., benzene, toluene, and xylenes) or
how much of a pollutant needs to be
controlled (e.g., H2S and carbonyl
sulfide).
-------
W7
106
10s
<0
o
tJ
O)
103
Vl
w
o
c
o
10
10'
-
i ,
Q) Q) < l
\
§ 1
i , 1 i * " s 1 1
» i i if .1 'Ui *?
' ! 1 1 11 111 *l| !lJ
- 5 !. lift sihiMsi?8!
_ T . 3 S ^ 5 STTSfeTS* «
:: « Ml " i
' -; g^?l .
-c £ S '
c 9- ' x -
k, CO i L (* 1
1. C
-C ^ -5
T Q
-L
*
i
" ^
.
\
, t
J
-
_L ^ k
» o
* ?*. *- o °j c
r"T"5.o 0 T '« c^:^,. G|
TT *
*' 'i' i
J. " jk
-L ^ k
"-[{ [[ 1J
r >
r ^
1 Carbonyl S
: H
K
Geometric
A Arithmetic
]
.!:
s^ -
"T" ^
C
Q
Q
i : '
" 5
|
-L
^5
"3
S)
Carbon Monox.
i . . 1 , ,
-
~
z.
01
s,
o
w
5
Mean
Mean
Maximum &
. Minimum Values
*/Vo D/WfG ex/'sfs. T 95% Confidence
Limits
-------
lends credence to the RTI gasifier's
ability to simulate large-scale processes
with respect to both pollutant production
and production of major gas constitu-
ents. The only gas constituent for which
the RTI concentration appears low in
comparison to other gasif iers is ammonia.
This may be caused by low scrubber
efficiency for ammonia capture. Fortu-
nately, the majority of the ammonia
condenses out as aqueous ammonia in
the condensate and thus the low gas
value introduces little error to overall
ammonia production. Also adequate
comparison with other gasifiers is
difficult because of differences in con-
densation or quench temperatures and
procedures. Tables similar to Table 2 are
available in the report for aqueous
condensate, tar, and ash streams.
Comparisons of pollutant levels in the
RTI gasifier condensate with pollutant
levels in the aqueous effluents from
other coal conversion processes, ambi-
ent levels in U.S. waters, and regulated
levels are given in Table 3. Comparisons
with most stringent effluent standards
for the U.S. and Canada (and with end-
of-pipe effluent standards set by the
State of Illinois) suggest that, in certain
locations, aqueous effluents will have to
be processed to reduce the levels of all
the pollutants listed. Considering the
variation in the chemical nature of these
pollutants, a diverse approach may be
necessary when deciding on wastewater
treatment options; several techniques
will have to be combined for proper
treatment in areas, such as Illinois, with
strong regulation of effluent composition.
Comparison of the levels of the pol-
lutants (ammonia, sulfides, thio-
cyanates, phenol, and cyanides) in the
RTI condensate with coke plant liquors
suggests that these streams are very
similar in gross chemical composition.
Point source effluent limitations have
been established by the EPA for aqueous
effluents from by-product coking plants
under the Clean Water Act. Cyanide,
phenol, ammonia, and sulfide are regu-
lated under these source-based effluent
limitations. As a result, control tech-
niques have been developed for these
streams and are presently well-estab-
lished technology. Because of the simi-
larity between coke plant liquor and
gasifier cpndensate, thistechnology
from the coking industry can probably
be easily applied to gasifier aqueous
condensate.
Similarly, point source effluent limita-
tions also have been established for
certain aqueous effluents from petroleum
refining; techniques developed for pol-
lution control in these streams may also
be applicable to gasification wastewater.
Thus, the basic technology for the
control of major pollutants in gasifier
wastewater streams is already available,
although certain modifications may be
necessary for the proper application of
these technologies. Tables similar to
Table 3 are available in the complete
report for product gas, tar, and ash
streams.
Conclusions
Significant conclusions from this
study are:
1. Efficient fixed-bed gasification of
caking coals presents problems be-
cause of the tendency of these coals
to agglomerate in the gasifier, lead-
ing to poor gasification rates and
heat transfer characteristics.
Table 2. Comparison of Concentrations [/jg/m3] and Production Factors [fjg/g coalY of Significant Product Gas Constituents
from Different Gasifiers
Carbon Monoxide
Benzene
Hydrogen Sulfide
Carbonyl Sulfide
Hydrogen
Carbon Dioxide
Thiophene
Methanethiol
Methane
Ammonia
Toluene
Xylenes
95% Confidence
Ranges for
This Study
2.0E8-2.8E8
1.6E6-2.6E6
(4.2E3-7.0E3)
3.0E6-6.0E6
(8.4E3-1 .8E4)
1.4E5-2.8E5
(3.8E2-8.8E2)
1.2E7-1.8E7
2.2E8-2.8E8
2.9E4-1.3E5
(7.0E1-4.1E2)
1.0E4-1.8E4
(2.6E1-4.4E1)
2. 1E7-3. 1E7
4.9E4-1.3E5
(1.2E2-4.0E2)
3.9E5-2.5E5
(9.4E2-1 .7E3)
1.3E5-6.9E5
(3.4E2-6.2E2)
METC Range (3 Coals)
(Average for Air-Blown We/lman-Galusha Koppers-Totzek
7 Coals) Synthane (Penn.-Anthr. Coal) (Oxygen-Blown)
2.6E8 1.3E8-2.0E8
2.4E6-3.4E6
(4.8E3-7.4E3)
5.5E6 8.8E5-7.6E6 8.9E5
(2.0E3-1.3E4) (4.9E3)
5.4E4-3.8E5 2.2E5
(1.2E2-7.6E2) (1.2E3)
1.3E7 2.0E7-2.7E7
1.9E8 3.7E8-3.8E8
<1.9E4-2.6E5
(<3.8E1 -4.4E2)
1.7E4-8.6E4
(3.4E1-1.5E2)
1.8E7 3. 1E7-4.0E7
1.2E5
(6.9E2)
2.9E5-9.0E5
(5.8E2-1 .8E3)
9.5E4-2.8E5
(1 .9E2-6.2E2)
7.3E8
_
4.3E6-2.6E7
(6.6E3-4.6E4)
5.5E5-1.8E6
(8.0E2-3.2E3)
2.9E7
2.1E8
0-8.3E5
Dry Ash
Lurgi
6.7E6
(9.6E3)
continued
-------
Table 2 (continued).
95% Confidence
Range for
This Study
Carbon Monoxide
Benzene
Hydrogen Sulfide
Carbonyl Sulfide
Hydrogen
Carbon Dioxide
Thiophene
Methanethiol
2.0E8-2.8E8
1.6E6-2.6E6
(4.2E3-7.0E3)
3.0E6-6.0E6
(8.4E3-1.8E4)
1.4E5-2.8E5
(3.8E2-8.8E2)
1.2E7-1.8E7
2.2E8-2.8E8
2.9E4-1.3E5
(7.0E1-4.1E2)
1.OE4-1.8E4
(2.6E1-4.4E1)
GFETC
Bigas (Oxygen Blown)
2.8E8
8.5E6
(3.9E4)
3.0E6*
(1.3E4)*
1.4E7
1.5E8
7.5E8
8.7 E 5
(1.5.E3)
7.6E4*
(1.3E2)*
2.8E7
1.6E8
4.9E4**
(8.5E1)**
Lurgi
(Oxygen Blown-
COz Chapman Woodall Range for
Acceptor Wilputte Duckham W inkier 5 Coals)
2.6E8 3.757 2.9E8 2.0E8-2.7E8
1.6E6 4.0E5 7.0E6-1.6E7
(2. 1E3)
7.4E4
(9.7E1)
1.4E7 1.6E7 1.3E7 3.7E7-3.9E7
1.8E8 3.7E8 1.5E8 6.0E8-6.5E8
Methane
2.1E7-3.1E7 3.2E7
4.0E7***
1.1E7 2.057
7.5£6
6.7E7-8.4E7
Ammonia
Toluene
Xylenes
4.9E4-1.3E5
(1 .2E2-4.0E2)
3.9E5-6.9E5
(9.4E2-1 .7E3)
1.3E5-2.5E5
(3.4E2-6.2E2)
3.3E6
(1.6E4)
4.2E6
(5.5E3)
^Production factors in parenthesis.
Blanks in table indicate values not available.
*lncludes carbon disulfide.
** Includes other thiols.
***lncludes
2. Glass capillary gas chromatography
(GC2) with specific element detection
is a complementary analytical tech-
nique to gas chromatography/mass
spectrometry (GC/MS) for charact-
erization of complex coal gasifier tar
samples. Identification and quantita-
tion of high molecular weight sulfur
hetecocyclics and primary aromatic
amines is difficult because they are
present at very low concentrations
(~ ppm) in an extremely complex tar
matrix.
3. The most significant coal gasifica-
tion effluent stream from an environ-
mental standpoint is aqueous con-
densate, followed by tar, product
gas, and ash. However, on an equiv-
alent weight basis, the tar stream is
more toxic and mutagenic than
aqueous condensate based on cyto-
toxicity and bioassy tests. Polycyclic
aromatic hydrocarbons (PAHs) and
tar bases are the most mutagenic of
the various tar fractions. Coal pyro-
lysis and gasification at higher tem-
peratures (as in continuous coal feed
versus batch feed) leads to reduced
tar mutagenicity. The ash stream is
very likely nonhazardous under the
Resource Conservation and Recovery
Act (RCRA) extraction procedure.
4. The most environmentally significant
product gas constituents are CO,
benzene, H2S, and carbon sulfide.
The product gas represents a "well-
known" hazard in the chemical
industry, namely that the product
(CO) that one is trying to maximize
contributes the greatest hazards.
The environmentally significant
aqueous condensate constituents
include phenol/cresols/xylenols
(PCX), ammonia, sulfides, thio-
cyanates, cyanide, arsenic, and chlo-
rides. The environmentally signifi-
cant tar constituents include various
PAHs (especially dibenzo(a.h)-
anthracene, benzo(a)pyrene, and
benzo(a)anthracene), PCX, and ar-
senic. The environmentally signifi-
cant ash elements include arsenic,
nickel, beryllium, and selenium.
The RTI laboratory gasifier produced
pollutant and total stream data which
compared very favorably to those re-
ported in the literature for larger-
scale gasifiers. Thus, a laboratory
gasifier can simulate large-scale
gasifiers and serve as a model for
studying problems associated with
large-scale gasifiers in a cost-effective
manner.
Coal liquefaction processes produce
liquid and solid effluent streams
which are qualitatively similar to
gasification aqueous condensate,
tar, and ash. Combustion of coal pro-
duces, in general, far lower quantities
of PAHs compared to gasification of
coal on a unit coal basis. Coke plant
wastewater is qualitatively similar to
coal gasification aqueous condensate:
pollutant concentrations in the two
are within an order of magnitude.
-------
Table 3. Comparisons of Concentrations[ug/L] of Significant Aqueous Condensate,Constituents with Concentrations in Effluents
from Other Coal Conversion Processes, Ambient Levels, and Regulated Levels
Cresols
Xylenols
Ammonia
Sulfides
Arsenic
Thiocyanates
Phenol
Cadmium
Selenium
Chromium
Cyanide
Iron
Chlorides
Silver
Lead
RTI
95% C.I.
4.8E5-8.8E5
1.4E5-3.2E5
4.8E6-7.8E6
0-4.7E5
1.6E2-1.2E3
1.5E5-3.1E5
7.9E5-1.6E6
0-2J £2
2.5E2-1.0E3
3.6E2-3.0E3
1.5E3-7.3E3
4.7E3-5.1E3
9.4E5-2.7E6
0-1. 5E2
1.1E1-2.9E2
Coke Plant
Waste Ammonia
Liquor Range
1.8E6-4.3E6
0-5.0E4
1.0E5-1.5E6
4. 1E5-2.4E6
1.0E4-3.7E4
Coal Ash
Pond Effluent
5.0E1
2.0E1
<2.5E1
<1.0E2
<1.0E1
5.0E2
<5.0E1
Coke Plant
Liquor
5.0E6
1.3E6
1.0E6
1.6E6
5.0E4
6.0E6
SRC-I
Wastewater
5.6E6
4.0E6
4.5E6
H-Coal Foul
Process Water
8.0E2
1.0E2
1.0E4
1.2E3
2.9E3
SRC
Wastewater
9.4 E 5
3.8E5
3.9E5
8.0E2
9.0E4
5.6E5
Rivers & Most Stringent Most Stringent
Cresols
Xylenols
Ammonia
Sulfides
Arsenic
Thiocyanates
Phenol
Cadmium
Selenium
Chromium
Cyanide
Iron
Chlorides
Silver
Lead
RTI
95% C.I.
4.8E5-8.8E5
1.4E5-3.2E5
4.8E6-7.8E6
0-4.7E5
1.6E2-1.2E3
1.5E5-3.1E5
7.9E5-1.6E6
0-2. 1E2
2.5E2-1.0E3
3.6E2-3.0E3
1.5E3-7.3E3
4.7E3-5.1E3
9.4E5-2.7E6
0-1. 5E2
1.1E1-2.9E2
Wabash
River
Indiana
8.0E1
2.0EO
1.0EO+
<1.0EO
1.0EO
<1.0E1
O.OEO
1.8E3
2.3E4
O.OEO
4.0E1
Fresh
Water
4.0E-4
<2.0E-2
1.8E-4
6.7 E-1
7.8EO
1.3E-4
2.2E1
Lakes
Mean of 8
Regions
8.1 E1
1.2E1
4.3E1
2.2EO
2.2E1
Water Quality
Standards
(U.S.)
2. Of/
1.0E1
1.0EO+
2.0EO
5.0EO
5.0E1
5.0EO
3.0E2
1.0E5
1.0E-1
1.0E2
Effluent
Limitations
(U.S. & Canada}
5.0E2
1.1 £1
5.0E1
5.0EO+
5.0EO
1.0E1
5.0E1*
2.0E1
3.0E2
2.5E5
5.0E1
5.0E1
EPA
NIPDWS°
5.0E1
1.0E1
1.0E1
5.0E1*
5.0E1
5.0E1
Illinois
Effluent
Standards
(all point
sources)
2.5E2
3.0E2
1.5E2
3.0E2*
2.5E1
2.053
1.0E2
1.0E2
*Hexavalent Cr
+Total Phenols
oNational Interim Primary Drinking Water Standards
7. The 95 percent confidence intervals
for concentration and production of
significant pollutants under a variety
of gasification conditions (including
coal type, temperature, pressure,
and coal particle size) are well within
an order of magnitude except for
sulfides and some trace elements. A
major factor probably is the coal type
for sulfur compound production since
as much as an order of magnitude or
greater variation can occur in the
amount of sulfur present in the coal
itself.
S. K. Gangwal, J. G. Cleland, and R. S. Truesdale are with Research Triangle
Institute, P.O. Box 12194, Research Triangle Park, NC 27709.
N. Dean Smith is the EPA Project Officer (see below).
The complete report, entitled "Evaluation of Relative Environmental Hazards
from a Coal Gasifier," {Order No. PB 81-217 648; Cost: $11.00. subject to
change) will be available only from:
National Technical Information Service
5285 Port Royal Road
Springfield. VA22161
Telephone: 703-487-4650
The EPA Project Officer can be contacted at:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park. NC 27711
0 U.S GOVERNMENT PRINTING OFFICE 1961 -757-012/7272
-------
United States
Environmental Protection
Agency
Center for Environmental Research
Information
Cincinnati OH 45268
Postage and
Fees Paid
Environmental
Protection
Agency
EPA 335
Official Business
Penalty for Private Use $300
PS U000329
U S EMVIR PROTECTION
REGION 5 LiBHAKY
230 S DEARBORN STREET
CHICAGO JLL 606U4
------- |