dEPA
United States
Environmental Protection
Agency
Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
Research and Development
EPA-600/S7-81-142 Dec. 1981
Project Summary
Environmental Assessment:
Source Test and Evaluation
Report—Lurgi (Kosovo)
Medium-Btu Gasification,
Final Report
K. W. Lee, W. S. Seames, R. V. Collins, K. J. Bombaugh, and G. C. Page
An environmental data acquisition
program, sponsored jointly by the
U.S. EPA and the government of
Yugoslavia, has been conducted in the
Kosovo Region of Yugoslavia as a
cooperative effort between Yugoslav
and American scientists. The aim of
the program was to gather data which
could contribute to the characteriza-
tion of potential environmental prob-
lems associated with coal gasifiers
using Lurgi gasification technology.
Processes involved in Lurgi gasifica-
tion technology are of particular
interest to the U.S. EPA because they
have a significant potential for use in
the U.S. The test plant, though not
representative of state-of-the-art
pollution control practice, provided
process discharge streams that are
typical for Lurgi gasification tech-
nology.
An extensive sampling and analysis
program was conducted at Kosovo.
The main thrust of the study was to
characterize the discharges of the
plant's key processes and to prioritize
pollutant discharges in terms of their
potential for causing adverse health
and/or environmental effects. Prior-
itization was accomplished using the
Source Analysis Model/IA (SAM/IA)
developed by EPA's Industrial Envi-
ronmental Research Laboratory (IERL)
at RTP. The prioritized data provides a
basis for making engineering judg-
ments on control technology.
The major conclusions drawn from
this study are that, without environ-
mental controls, a Lurgi type process
exhibits a significant potential for
polluting the environment; that virtually
all discharge streams, whether gaseous,
aqueous, or solid, present a significant
potential for transferring pollutants
from the process to the environment;
and that particulates in gaseous dis-
charges are carriers of potentially
hazardous PNAs. Based on the SAM/
IA prioritization, the Kosovo discharge
streams of highest concern are: H2S-
rich waste gas, extracted phenolic
water, and heavy tar. Gasifier ash was
found to be of low concern. Discharge
stream pollutants of major concern
are sulfur species and aromatic hydro-
carbons, including polycyclic aromatics.
Trace elements were found to be of
much lower significance than trace
organics.
This Project Summary was devel-
oped by EPA's Industrial Environmen-
tal 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
The purpose of this program was to
characterize the discharge streams
from a commercial Lurgi-type coal
gasification facility as a means of
determining the pollution potential of
Lurgi type processes. The program was
designed in response to a need for
representative data on potential envi-
ronmental problems associated with
the commercial application of Lurgi coal
gasification technology. The Lurgi coal
gasification process was selected for
study because it has a significant
potential for use in the U.S. and is being
considered as a commercially feasible
technology for the production of substi-
tute natural gas and for the indirect
liquefaction of coal. The Lurgi type plant
chosen for this study is in the Kosovo
region of southern Yugoslavia. It uses
3.5 m diameter gasifiers. The opportunity
to make a comprehensive environmental
characterization of an operational,
commercial-scale, Lurgi type coal
gasification plant was considered
valuable since a number of U.S. compa-
nies have announced plans to construct
such plants. Thus, characterization of
selected process and discharge streams
for the Kosovo plant provided a valid
insight into problems that must be
considered by U.S. designers in devel-
oping process modifications and/or
control schemes necessary to meet U.S.
environmental standards.
The program, conducted over 3 years,
was a joint effort among scientists from
the U.S. and Yugoslavia. The participating
organizations, and their roles, are
shown in Table 1.
The program consisted of four phases:
Phase
Objective
1 Identify and measure major and
minor pollutants in discharge
streams.
2 Identify and measure trace
pollutants in discharge streams.
3 Characterize ambient air pol-
lutants in the vicinity of the
plant.
4 Measure fugitive emission rates
from leak sources in the plant
(cofunded by DOE).
This report presents information on
major, minor, and trace pollutants in the
discharges of the Kosovo Lurgi type
plant. Results from testing Phases 1 and
2 were combined to provide a "best
value" for use in evaluating the dis-
charges. This report includes an assess-
ment of specific discharges and also an
evaluation of discharge severity as
determined by EPA/IERL's SAM/IA
model for prioritizing pollutants on the
basis of their estimated potential for
causing adverse health and ecological
effects.
Plant Description
The Kosovo gasification plant is a part
of a large mine-mouth industrial complex
that includes a coal mine, a coal
preparation plant, a coal gasification
plant; an air separation plant, an
ammonia plant, a power generating
plant, and a steam generating plant. The
gasification plant consists of six opera-
tional units as illustrated in Figure 1.
The plant consumes dried lignite and
Table 1. Functions of Various Organizations Participating in the Kosovo Test
Program
Organization
Location
Function
EPA/IERL
Radian Corporation
Rudarski Institut
Kombinat Kosovo
Kosovo Institut
Institut za Primenu
Nuklearne Egergije
Research Triangle Park
North Carolina
Austin. Texas
Belgrade, Yugoslavia
Obilic, Yugoslavia
Obilic, Yugoslavia
Belgrade, Yugoslavia
Funding Agency, Project
Director
Prime Contractor and
Coordinator
Sampling/A nalyses/Data
Analyses /Overseas Coor-
dinator
Plant Operation/Sampling
Sampling/Trace Element
Analyses
On-Site GC Analyses/
Organic Analyses
produces two primary products (E
medium-Btu fuel gas and hydrogen) anc
four by-products (tar, medium oil,
naphtha, and crude phenol). The plant's
operation is as follows: run of the mine
coal, brought from a nearby open-pit
mine, is dried by the Fleissner process
and sized to select particles 6-60 mm in
diameter. The dried coal is fed to the
Lurgi type gasifiers where it is reacted
with steam at 2.5 MPa (25 atm) pres-
sure to produce a crude gas which is
quenched, cooled, and then cleaned by
the Rectisol process before being
transported to the utilization site. As the
crude gas is quenched and cooled, tars,
oils, and phenols are removed in a
stream with the gas liquor (phenolic
water). In the gas-cleaning operation,
condensible organics are removed by
refrigeration, after which the acid-
gases (H2S and COz) are removed by
sorption in cold methanol. The acid-gas-
rich methanol is regenerated by depres-
surization and heating, releasing a gas
rich in H2S (which is flared) and a gas
rich in C02 (which is vented to the
atmosphere).
Tars and oil are separated from the
phenolic water by decantation after
which the water soluble organics (crude
phenols) are extracted with diisopropyl
ether. Ammonia, removed from the
phenolic water by steam-stripping, is
vented to the atmosphere.
The Kosovo gasification plant is a
commercial-scale facility. Figure 2
shows the design flow rates of the
plant's major inlet and outlet streams.
These data indicate that the plant is
designed to produce 25 Mg (65,000 m3
at 25°C) of product gas for every 80 Mg
of dried coal consumed. Although this
plant is smaller than proposed first-
generation U.S. Lurgi gasification
facilities, it contains many of the
process units which are likely to be
employed in future U.S. Lurgi plants. For
this reason, the plant is felt to be
representative of many aspects of the
Lurgi gasification facilities which are
being considered for commercialization
in the U.S. However, the environmental
control practices followed at the Kosovo
plant are not representative of proposed
U.S. Lurgi plant designs. Many of the
plant's waste streams are controlled but
none of the controls would be charac-
terized as best available by current U.S.
standards. Thus, while the discharges
that enter the environment at Kosovo
are not representative of those that
would be encountered in similar U.S.
facilities, the types of control problems
-------
Fines to
Steam and Power
Generation
Steam
Steam
Run-of-Mine
Coal
Wastewater
By-Products
to Steam and
Power
Generation
Wastewater
Figure 1. Simplified flow diagram of the Kosovo coal preparation/gasification plant operations.
facing U.S. Lurgi plant operators will be
similar to those found at Kosovo. A
study of the waste streams data
generated at Kosovo, therefore, should
aid U.S. plant designers in developing
the process modification and control
schemes necessary to achieve U.S.
control standards.
Stream Selection and Test
Methods
The Kosovo test program was con-
ducted over 3 years as a phased effort to
characterize the plant's major, minor,
and trace pollutant discharges. Initially,
screening studies were conducted to
select the most significant streams for
more detailed characterization. Stream
selection was based on: high discharge
rate, high pollutant concentrations,
trace pollutants of high concern, and
important process information. Streams
exhibiting a high discharge rate were
selected for study because at high flow
rates, even very low concentrations of
moderately toxic pollutants could result
in a significant environmental burden.
Streams exhibiting a moderate or low
discharge rate were selected for study if
they contained high pollutant concen-
trations of pollutants of high concern.
Some streams were selected to
provide information on the fate of trace
elements and trace organics throughout
Rectisol
Acid Gases
(HzS-Rich and COz-Rich)
(45}
\
Dried Coal (80)
Steam (65)
02(14)
Kosovo
Gasification
Plant
•
•
•
V
Gasifier Ash
(14)
T
1
Waste -
waters
(68)
Heavy Tar
(0.5)
Figure 2.
Light Tar (2.2)
Medium Oil (1.3)
Naphtha (0.7)
Phenols (0.4)
Ammonia (1)
Design flow rates of key streams in the Kosovo gasification plant (all
values in Mg/hr based on five gasifiers in service).
the process; and some streams were
selected to provide a better understand-
ing of process operations that affect the
distribution of pollutants in those
processes.
The streams considered in the program
are shown in Table 2; of these streams.
50 were tested (35 gaseous, 5 aqueous,
6 solid phase, and 4 organic liquid) and
31 (17 gaseous, 5 aqueous, 6 solid
phase, and 3 organic liquid) received
detailed characterization.
The detailed characterization program
addressed the following parameters:
-------
Gaseous Streams
• Flow rate.
• Paniculate concentration.
• Gas composition.
• Condensible organics.
• Trace elements content.
Aqueous Streams
• Water quality parameters.
• Trace elements content.
• Organic constituents.
Solids
• Proximate analysis.
• Ultimate analysis.
• Trace elements content.
• Leachate composition.
Liquid By-Products
• Bulk composition.
• Trace elements content.
• Polynuclear aromatics content.
Table 2 . Number and Type of Significant Streams in Each Section of the
Kosovo Plant
Type of Stream
Plant Section
Coal Preparation
Gas Production
Rectisol
Tar/ Oil Separation
Phenosolvan
By-Product Storage
Gaseous
3
8
5
7
12
_j6
41
Aqueous Solid
1 2
1 3
1 0
1 1
3 0
0 0
7 6
Organic
Liquid
0
0
1
2
2
J>_
5
Table 3. Measurement Methods for Flow Rate
Measurement
Parameter Method
Velocity
Temperature
Pilot
Thermocouple
Total
6
12
7
11
17
_6_
59
Analytical
Method
—
Methods used to study these parameters
are summarized in Tables 3 through 7.
Bioassays were performed on samples
from seven streams: dry gasifier ash,
heavy tar, light tar, medium oil, naphtha,
Phenosolvan inlet water, and Pheno-
solvan wastewater. These tests provided
information about potential biological
effects from the Kosovo plant's solid,
liquid by-product, and wastewater
streams.
Sampling Methods
Sampling methods used in this study
are listed in Tables 3 through 7. Gas
stream sampling was the most complex
and consumed an estimated 90 percent
of the sampling effort. Although pub-
lished methods were followed in prin-
ciple, a considerable amount of adapta-
tion was needed to fit the collection
requirements. Problems encountered in
sampling included: high stream pres-
sure, widely pulsating stream flow
rates, entrained mist, high levels of
condensable organics, and reactive
sample components.
Particulate sampling was considered
the most difficult because the entrained
mist and condensables present in many
streams obstructed the filter in a
conventional collection train. For such
streams, particulates were collected in
a train of water-filled impingers. The
collected matter was isolated from the
impinger liquid (by a combination of
extraction, filtration, and evaporation)
and determined gravimetrically as tars
and oils, filterable solids, and dissolved
solids.
Pressure
Molecular Weight
Manometer/Gauge
Gas Collection &
Conditioning system
TC-GC
Table 4. Measurement Methods for Particulates
Component Collection Method
Analytical Method
Suspended
Particulates
Suspended
Particulates &
Condensables
(tar-laden streams)
Filtration
- in stack at duct
temperature
- out of stack at 250°F(121°C)
Cold Impinger
Gravimetric
Filtration/Extrac-
tion/ Gra vimetric
Table 5. Measurement
Component
Methods for Gas Composition
Collection Method
Analytical Method
Moisture
Fixed Gases*
Hydrocarbons*'
Sulfur Species*
HZS
HCN
Phenols
ImpingerVice bath
Gas Collection and
Conditioning System/
Glass Bombs
Same as Fixed Gases
Same as Fixed Gases
Impinger/CdOA c
Impinger/NaOH
Impinger/HzSO*
Impinger/NaOH
Gravimetric
TC-GC
FID-GC
FPD-GC
Titration
Distillation/Titration
Distillation/Titration
Color/metric
*CO, C02, Oz, H2. CHA,
**Ci-C6, C6+, BTX
***H2S, COS. RSH, SOZ
-------
Table 6. Measurement Methods for Condensable Organics
Component Collection Method Analytical Method
Condensable
Hydrocarbons
(xylene and up)
Condensation/XAD-2 Resin
Benzene, Toluene,
and Xylene
Charcoal Tube
Extraction: TCO
GRAV
IFt
LC Fractionation: TCO
GRAV
IR
GC-MS
Extraction/FID-GC
Table 7. Measurement Methods for Trace Elements
Group Collection Method Analytical Method
Nonvolatile
Elements*
Volatile
Elements**
Acid/ Base Impinge r
Series
Acid/Base Impinger
Series
Dissolution/ AA,
Graphite Furnace
Dissolution/ AA,
Hydride Generation
Carbonyls*
Filtration/HCI
Impingers
AA, Graphite Furnace
*Be, Cd, Ce, Cr, Cu, Mo, Ni. Pb. Sr, Tl, V, In
**Hg, As, Sb, Se
***Ni, Fe
Trace elements, volatile phenols, and
reactive species (including NH3, H2S,
and HCN) were also collected from gas
streams by impinger train. Gas samples
for gas chromatographic analysis were
collected in 0.5-L silanized glass bombs
at 0.5 atm pressure with the aid of a
specially built apparatus that included a
heated inlet, a permeation drier, and a
diaphragm pump.
Analysis Methods
Analysis methods used in the study
are also listed in Tables 3 through 7. All
gas analyses were performed on-site.
Fixed gas analyses were run on all GC
samples to detect air leakage and to
obtain the compositional information
provided by the analyses. Any sample
showing an abnormal level of air was
discarded and the sampling repeated
until a valid sample was obtained from
the collection point.
Test Results
Composition data were obtained on
49 streams which included discharge,
feed, product, and process streams.
Detailed composition and flow rate data
were obtained in 19 gaseous, 2 aqueous,
and 2 solid discharge streams. Results
were used to rank both discharge
streams and pollutants in terms of their
potential for causing 'adverse health or
ecological effects according to the EPA-
IERL SAM/IA model. Of the three
discharge media, gas streams were
found to be the major carrier of
pollutants in terms of both pollutant
mass and discharge severity (DS).
Discharge severity was computed in
terms of both DSH (potential for adverse
health effects) and DSE (potential for
adverse ecological effects).
Gaseous Discharge
The most significant gaseous dis-
charge stream in the Kosovo gasification
plant, in terms of DSH, is the H2S-rich
waste gas from the two-stage Rectisol
plant (at Kosovo, this stream is flared).
The next most significant stream is the
ammonia stripper vent discharge which
contains 42 percent ammonia and high
concentrations of H2S, phenols, HCN,
and mercaptans (this stream has the
highest DSE). The C02-rich waste gas,
which is the largest atmospheric
discharge stream at Kosovo, releases
significant quantities of hydrocarbons
and reduced sulfur species. Although
the DS of this stream is 2.5 percent of
the HzS-rich stream, it is environmen-
tally very significant because of its high
flow rate.
Tank vents were found to be signifi-
cant sources of pollution. Although their
flow rates were small (<5 mVhr), their
pollutant concentrations were high (up
to 4 percent benzene).
The major pollutant class, based on
discharge mass, was light hydrocarbons.
The most significant class in terms of
Weighted Discharge Severity was sulfur
species.
Particulates in Gaseous
Discharges
Particulate loadings in the discharge
from both the start-up vent and the low
pressure (LP) coal lock vent were about
10 times higher than in the discharges
from the other vents that were tested.
About 90 percent of the particulate
loading in these discharges consisted of
condensable organics (tar and oils)
which contained appreciable amounts
of PNAS. It is estimated that 500-1500
ug/m3 benzo(a)pyrene is contained in
the LP coal lock discharge, resulting in
an estimated discharge rate of 3-9
//g/sec.
Aqueous Discharge
Aqueous discharges from both
Phenosolvan and gasification had high
values for solids, COD, and permanga-
nate value. Gasification wastewater
exceeded effluent guidelines for sus-
pended solids by 87/1. Phenosolvan
effluent, the plant's largest waste
stream, exceeds effluent guidelines in
both Total Suspended Solids and pH.
The major pollutant of this discharge
was phenol. Although the Phenosolvan
process removed 90 percent of the
incoming phenol, it removed only 70
percent of the total organic carbon.
So fid Discharges—
Gasifier ash, the plant's major solid
waste, presents no unusual environ-
mental problems. Although it has a
positive heating value, it would not be
considered ignitable by RCRA standards.
Trace element levels in ash leachate
were below the concentration levels
that would cause gasifier ash to be
considered toxic according to RCRA
limits.
Heavy tar, which has a higher heating
value than the feed coal, contains
polycyclic aromatic hydrocarbons in-
cluding a significant concentration of
benzo(a)pyrene (0.024 percent).
Comparison of All Discharge
Media—
A comparison of 18 discharge streams
in all media (gas, aqueous, and solid),
shown in Figure 3, indicates that the
-------
Gaseous Streams
Aqueous Streams
Hz-Ftich Waste Gas (7.1)
Ammonia Stripper Vent (14.5)
COz-Rich Waste Gas (7.2)
High Pressure Coal Lock Vent (3.6)
Autoclave Vent(1.2)
Tar Separation Vent (13.6)
Gasoline Storage Tank Vent (15.3)
Low Pressure Coal Lock Vent (3.2)
Phenolic Water Tank Vent (13.7)
Medium Oil Tank Vent (13.3)
Gas Liquor Tank Vent (3.4)
Condensate Tank Vent (13.5)
Cooler Vent (14.6)
Tar Tank Vent (13.1)
Phenosolvan Wastewater (14.11)
Ash Quench (12.3)
Heavy Tar (13.8)
Dry Gasifier Ash
(TDS Based on RCRA Leachate)
234567
Log,0 TDS+Log\o Flow (g/sec)
8
O Log,0 TDS
P Log-io flow
0 Logw Flow is Negative
TDS = I DS,
/ = all components
Figure 3. A comparison of the total weighted discharge severity values (health)
key Kosovo gaseous, aqueous, and solid streams.
streams of highest environmental con-
cern, in terms of health, are the H2S-rich
waste gas, the heavy tar, and the
ammonia stripper vent discharge,
respectively.
Bioassay Results—
Of the nine samples tested, the liquid
by-products were the most toxic in
rodent toxicity tests. By-product tar and
phenolic water each gave a positive
response to the Ames test for carcino-
genicity. However, the Phenosolvan
effluent water was negative, indicating
a reduction in bioactivity by the process.
Gasifier ash and ash leachate showed
little or no biological activity.
Mass Balances—
Mass balance calculations on the test
data show that most of the major
elements in the feed coal aredischarged
in the gaseous streams. The major
transporter of sulfur is the h^S-rich
waste gas stream; of nitrogen, the
ammonia stripper vent; and of carbon,
the product gas and a combination of
the HzS- and COa-rich waste gas
streams. Accountability of the sulfur,
nitrogen, and carbon was 180, 51, and
92 percent, respectively.
-------
K. Lee. I/I/. Seames, R. Collins, K. Bombaugh, and G. Page are with Radian
Corporation, P.O. Box 9948, Austin, TX 78766
William J. Rhodes is the EPA Project Officer (see below].
The complete report, entitled "Environmental Assessment: Source Test and
Evaluation Report—Lurgi (Kosovo) Medium-Btu Gasification, Final Report,"
(Order No. PB 82-114 075; Cost: $24.00, 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:
Industrial Environmental Research Laboratory
U.S. Environmental Protection Agency
Research Triangle Park, NC 27711
7
it U.S. GOVERNMENT PRINTING OFFICE.1981--559-092/3363
-------
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
------- |