Environmental Assessment Source Test And Evaluation Report Koppers-Totzek Process

United States
Environmental Prot
-------
DISCLAIMER

This report has been reviewed by the Industrial Environ-
mental Research Laboratory (RTF), U.S. Environmental
Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the views
and policies of the Agency, nor does mention of trade names
or commercial products constitute endorsement or recommen-
dation for use.
-------
U.S. ENVIRONMENTAL PROTECTION AGENCY

Research Triangle Park, North Carolina 27711
3/11/81
The attached report presents data
that were obtained from data acquisition
performed at the AECI facility at
Modderfontein in the Republic of South
Africa. This facility utilizes Koppers-Totzek
gasifiers which have potential use in this
country for indirect coal liquefaction and
other synthetic fuels from coal systems.
W. J. Rhodes 541-2853
-------
United States
Environmental Protection .
Agency EPA-600/7-81-OOS
&EPA Research and January1981
Development
ENVIRONMENTAL ASSESSMENT:

SOURCE TEST AND EVALUATION REPORT

KOPPERS-TOTZEK PROCESS
Prepared for

EFFLUENT GUIDELINES DIVISION
Prepared by

Industrial Environmental Research
Laboratory
Research Triangle Park NC 27711
-------
DISCLAIMER

SOURCE TEST AND EVALUATION

REPORT KOPPERS-TOTZEK PROCESS

January, 1981
Prepared by

C. A. Zee, J. F. Clausen, K. W. Crawford

TRW Inc.
One Space Park
Redondo Beach, CA 90278
Prepared for

William J. Rhodes
U.S. Environmental Protection Agency
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
-------
ABSTRACT

A source test program was conducted at a Koppers-Totzek (K-T) coal
gasification facility operated by AECI Limited at Modderfontein, Republic
of South Africa. The EPA's interest in the K-T process stems from the
fact that the process economics and demonstrated commercial reliability
make it a viable prospect for U.S. applications. The responsibilities
for sampling, analysis, and engineering descriptions of the Modderfontein
plant were shared between TRW and Krupp-Koppers GmbH of Essen, Federal
Republic of Germany. EPA's phased approach for environmental assessments
was followed. Level 1 and Level 2 data were collected along with priority
pollutant screening data. Much of the effort was focused on wastewater
streams. The wastewater treatment, consisting of a clarifier and settling
pond, was adequate to produce a final discharge that had lower pollutant
levels than the fresh input waters supplied to the plant. The complete
data are presented in this report along with descriptions of the K-T
process and the Modderfontein plant. The purpose of the Source Test
Evaluation (STE) was intended as an initial effort and was somewhat limited
in scope. Thus recommendations for future STE programs are also provided.
11
-------
TABLE OF CONTENTS

Page
Abstract ii
Figures iv
Tables v
Acknowledgments vi
1. Introduction and Summary 1
1.1 Program Summary 2
1.2 Conclusions and Recommendations 8
2. Plant Description 11
2.1 Plant Description 11
2.2 Plant Operation 14
2.3 Sampling Points 14
3. Sampling and Analysis 17
3.1 Krupp-Koppers Methods 17
3.2 TRW Methods 22
4. Results 29
4.1 Coal Feed Stream 29
4.2 Gas Streams 31
4.3 Aqueous Streams 35
5. References 53
Appendix
A. Krupp-Koppers Report 55
m
-------
FIGURES

Number Page
1 Schematic of Modderfontein Koppers-Totzek Coal
Gasification Facility 3
2 Summary of SAM/IA Results for Koppers-Totzi;'. > -nlity. . . 7
3 Koppers-Totzek Gasifier 12
4 HPLC Chromatogram of November 12, 1979 Rectisol Unit Sample. 43
5 HPLC Chromatogram of November 19, 1979 Rectisol Unit Sample. 44
IV
-------
TABLES
Number Page
1 Process Streams Requested for STE Program 4
2 Schedule for Sample Acquisition 5
3 Additional Data Needs for Koppers-Totzek Process 9
4 Distribution of Sampling and Analysis Responsibilities . . 19
5 Summary of K-K Gas Analysis Methods 20
6 Summary of K-K Water Analysis Methods 21
7 Priority Pollutant Screening Analytical Parameters .... 23
8 Volatile Organic Compounds Analyzed 24
9 Acid Semi-Volatile Compounds Analyzed 24
10 Base/Neutral Semi-Volatile Compounds Analyzed 25
11 PAH Compounds Used as Standards 28
12 Proximate and Ultimate Results from Coal Analysis 29
13 SSMS Results from Coal Analysis 30
14 Results from Gas Analyses 32
15 Summary of SAM/IA DS Results for Gas Streams 34
16 Summary of SAM/IA IDS and TWOS Results for Gas Streams . . 34
17 Results from K-K Wastewater Analyses 36
18 Results from TRW and M&L Wastewater Analyses 37
19 Results from Level 1 Organic Survey 38
20 Distribution of Compound Classes in Aqueous Samples. ... 39
21 Results from Level 1 SSMS Analysis 40
22 Results from Level 2 Analysis of Aromatic Hydrocarbons . . 45
23 Results from Level 2 Quantisation of Fe and Mn 46
24 Results from Organic Priority Pollutant Screening 47
25 Results from Inorganic Priority Pollutant Screening. ... 49
26 Summary of SAM/IA TDS and TWOS Results for Aqueous Streams 50
-------
ACKNOWLEDGMENTS

TRW wishes to acknowledge Krupp-Koppers GmbH (K-K) of Essen, Federal
Republic of Germany for their interest and willingness to participate in
this effort thus making the field tests possible. Some of the key indivi-
duals from K-K were Dr. Bruno Beck, Mr. Herbert Stempelmann, Dr. Gerhard
Preusser, and Dr. B. Firnhaber. The cooperation of AECI Limited in
allowing their plant to be tested is also gratefully acknowledged. Fur-
ther acknowledgment goes to Mr. Robin Lazar and the staff of McLachlan &
Lazar (pty) LTD, for their assistance in the analysis, preservation, and
shipment of samples for TRW.
vi
-------
1. INTRODUCTION AND SUMMARY

TRW, under contract EPA 68-02-2635 to the Environmental Protection
Agency (EPA), is performing a comprehensive environmental assessment of
high-Btu gasification and indirect liquefaction technologies. A major
portion of this environmental assessment project is to obtain data on
operating facilities through Source Test and Evaluation (STE) programs.
The objective of each STE program is to obtain the data necessary to:
1) evaluate environmental (ecological and health) effects of waste streams
or streams that may potentially be discharged in plants designed for
U.S. sites, and 2) allow subsequent evaluation of the equipment avail-
able or required for controlling these streams.
An STE program was conducted by TRW on a Koppers-Totzek (K-T) coal
gasifier. The EPA's interest in the K-T process stems from two principal
factors: first, in the national drive to supplement liquid and gaseous
fossil fuels through coal conversion, process economics dictate that the
more viable conversion products will be those having the highest unit
retail value. The K-T process represents one of the prime candidates
for converting raw coal into the intermediate synthesis gas needed to
produce these hiyh-value products. Secondly, the K-T process has a lengthy
history of successful application to a variety of foreign coals and
promises to be equally adaptable over the range of American coals. This
factor is particularly important in view of the contrasting lack of
demonstrated commercial reliability on the part of the developmental U.S.
gasifiers, and is viewed in a very positive light by both conversion pro-
ject financiers and program managers.
The K-T process operates on an entrained bed principle. It utilizes
a high temperature (1400° - 1600°C), atmospheric pressure reaction fueled
by a continuous co-current input stream of coal, oxygen and steam. The
licensor-developer of the Koppers-Totzek gasification process is Krupp-
Koppers GmbH (K-K) of Essen, Federal Republic of Germany. As of 1978, there
were 54 K-T gasification modules operating in the world of which 47 were

1
-------
using coal as a feed stock. All of the K-T gasifiers in operation as of
1978 were used to make synthesis gas as an input stream for the produc-
tion of ammonia. The facility selected for testing was the Number 4
Ammonia Plant at Modderfontein, Republic of South Africa. The plant is
owned and operated by AECI Limited and has a design production rate of
1000 tonnes per day of ammonia. The plant was commissioned in 1974.

1.1 PROGRAM SUMMARY
The Source Test Evaluation (STE) program was carried out as a joint
effort between TRW and K-K. TRW's initial review of the Modderfontein
plant, shown schematically in Figure 1, resulted in the selection of 25
streams as necessary to the comprehensive STE goals. Of these 25 streams,
as summarized in Table 1, nine were actually tested (i.e., Streams 7, 15,
16, 32, 33, 38, 40, 46, and 50). The selection of streams for testing
resulted from discussions between K-K and TRW in which streams considered
proprietary, not applicable to STE goals, or otherwise restricted were
eliminated from the list. The STE thus became limited in scope and
focused on the remaining available streams. Later developments indicated
that several of the 25 streams were not considered feasible.
The on-site sampling and analysis were performed by K-K. Samples
were taken according to the schedule shown in Table 2. Their overall
effort spanned a three-week period in November, 1979. The gas samples
were analyzed for the species H20, H2, CO, C02, N2, CH4, H2S, COS, CS2,
R-SH, S00, NH_, HCN, and NOY. Aqueous samples were analyzed by K-K for
L. J A
the standard wastewater tests (e.g., pH, alkalinity, conductivity, BOD,
COD, anions, etc.) with a few supplemental wastewater tests also being
performed by a local commercial laboratory, McLachlan & Lazar (pty) LTD.
Wastewater samples were shipped to TRW for comprehensive organic and
inorganic analyses per the EPA procedures for Level 1, Level 2, and Priority
Pollutants (references 1, 2, 3). The Level 1 methods provide a broad
semi-quantitative survey from which constituents found to be present at
levels of potential concern are selected for further quantitative exami-
nation (Level 2). The Priority Pollutant screening consists of analyses
for a specific list of 129 pollutants of concern to the EPA.
-------
RAW COAL
DRV MILLED COAI
COAL FINES
RFCYCLEDCOAL
CONVEYING GAS
WASTE GAS
PURGE GAS
COAL DUST
19 SULFUR FRE(
14 STEAM CONDENSATE, 70COMPRESSED SULFUR
BLOWER 13 CO; FfltE PRODUCT GAS 21 NITROGEN WASH TAII
I INPUT WATER (CW 23 NH3 SYNTHESIS FEED GAS 79 METHANOL
1 COMPRESSED HAW GAS 74 NH, SYNTHESIS FEED 30 RECYCLE Mf THANOl
— (COMMIEBEIt) 31 CO; RICH METHANOL
1BHCNFRFE RAW GAS
31 DILUTED RECTISOL CONDENSAIl
33 TAIL GAS
34 CO, RICH ACID GAS
3S H-fS RICH ACID GAS
M HjS RICH ME THANOL
37 HjS RICH MFTHANOl
3t TAILG.AS
» HCN WASH CONDENSA1
40 CtMIPRESSOflS CONDFK
41 tLECTBOSTATtC
COOLFR P
4bClARIf 1ER FffO
4KINPLIT WA1FR (PSE)
4B SETTLE DLLARIF IE R SOI IDS
SOSFtTLING POND DISCMAR{,F
WATER
51 SETTLING POND SIIJDGI
Figure 1. Schematic of Modderfontein Koppers-Totzek Coal Gasification Facility
-------
Table 1. PROCESS STREAMS REQUESTED FOR STE PROGRAM
' "• ' ' '" ' "' ""' • "" ' .11 ...——.—— . I. -m, .1 <
-------
Table 2. SCHEDULE FOR SAMPLE ACQUISITION
Streams Sampled/Stream Numbers*
Gas Streams:
Raw Gas after Raw Gas Blower/15
Tail Gas from H2S Absorber/38
Tall Gas from CO,, Absorber/33
Aqueous Streams:
Input Water-Treated Sewage/46
Process Water-Cooling Water/16
Compressor Condensates/40
Settling Pond Effluent/50
Diluted Rectisol Condensate/32
Solid Streams:
Sized Coal Feed/ 7
November 1979
11

X
X
X

-
13

X
X

X
X
X
X

X
20

X
* Stream numbers correspond to those shown on Figure 1.
-------
All of the data obtained from this STE were used in the EPA's Source
Analysis Model/IA which compares the measured concentrations on the con-
stituents analyzed to the EPA's Discharge Multimedia Environmental Goals
(references 4, 5). This model calculates discharge severities based on
the constituent concentrations alone and on the concentrations combined
with the stream flowrate (weighted discharge severity). This approach
provides a consistent basis for evaluating STE data.
The results of utilizing the SAM/IA approach with the data from the
Modderfontein Koppers-Totzek facility are summarized in Figure 2. The two
tail gas streams are direct emissions at Modderfontein. The discharge
water is the settling pond effluent. Results from the input waters (puri-
fied sewage .effluent and cooling water-) supplied to the gasification
facility are also provided for comparison. The data from Modderfontein
indicate that the streams tested do not appear to be of particular concern.
The discharge severity values obtained are similar to or lower than those
obtained on similar streams from other gasifiers (references 6, 7).
The discharge severities presented should not be construed as the result
of optimised control of pollutants from this unit. Depending upon design
of each plant and auxilliary processes, the number and location of effluent
streams could vary widely. A more conclusive determination of health and
ecological effects or lack thereof requires complementary biological tests.
Such tests were not included in this STE.
It should be noted that the magnitude of the discharge severity
values result from relatively few constituents. The TDS and WDS values
for the two tail gas streams are due primarily to the CO and NH3 concen-
trations. The TDS and WDS for the aqueous streams are due mainly to P and
Mn and to a lesser extent the metals Fe, Cd, Cu, Ni, Pb, and Zn. The
lowest DMEG value in the phosphorous class of compounds was used because
individual phosphorous species were not determined. The reduction in both
TDS and WDS values for the discharge versus the input waters appears to
be due to a decrease in concentrations of P, Cu, Pb, and Zn. These appear
to be lost to the settling pond sludge.
-------
TOTAL WEIGHTED
DISCHARGE SEVERITY (TWOS)
TOTAL STREAM
DISCHARGE SEVERITY (TDS)
ft)
ro
to
o
•D
•u
n>
-HO
O -t,
r-t-
N 00
(D :>
ft)
O
— -m
_i. in
O
-S
D
m
O
O
O
O
O
-------
1.2 CONCLUSIONS AND RECOMMENDATIONS
The limited source test program conducted at the Modderfontein
facility has provided some of the key data needed for the environmental
assessment of Koppers-Totzek based synthetic fuels plants which may be
built in the United States. The data obtained do not indicate that any
special problems should be encountered in controlling the process effluents
to environmentally acceptable levels for plants built in the U.S. Rela-
tively steady state conditions were realized during the test period thus
most of the samples taken were generally representative of typical plant
operation. This in turn indicates that the data can reliably be used as
intended. One exception was the Rectisol unit which apparently was not
operating properly at the time and hence data on Rectisol tail gas char-
acteristics are not believed to be typical.
Except for the Rectisol tail gases, additional sampling of the streams
which were the subject of the initial test program is not expected to
yield information other than of a confirmatory nature. Hence only limited
additional sampling of these streams is suggested in conjunction with
aqueous stream sampling as outlined below. In the case of the Rectisol
tail gases, no additional gas stream sampling is recommended since the
specific Rectisol design for ammonia production featuring "cold" shift
between H2S and COg removal would not be employed for synfuels production
and data on this type of design would not be especially useful for eval-
uation of synfuels discharge streams.
Several aqueous and solid waste streams were not subject to testing
in the initial program, however, data relating to the characteristics of
these streams would be helpful in the evaluation of pollution control needs
for U.S. facilities. Table 3 identifies these streams along with the type
of data of interest for each stream. As indicated in the table, data on
the characteristics of aqueous streams resulting from raw gas cooling and
particulate removal, from slag quenching and from the cold water wash
unit (HCN removal) are needed. Of major concern are constituents in the
aqueous wastes (e.g., NH3, HCN, H2S) which may become volatilized in the
clarifier or cooling tower systems resulting in atmospheric discharges.
In addition, characteristics of the gas cooling/washing wastewaters would
provide an indication of some of the original constituents in the crude

8
-------
Table 3. ADDITIONAL DATA NEEDS FOR KOPPERS-TOTZEK PROCESS
Stream Name
Stream
Number
Constituents/Parameters
of Interest
Uses of/Justification for
Additional Data
Coal Feed to Gasi-
fier

Input Water {Puri-
fied Sewage
Effluent

Input Water (Cool-
ing Water)

Washer Cooler
Slowdown
Disintegrator
Slowdown

ESP Wash Water
Raw Gas Compressors
Condensate
HCN Removal Wash
Slag Quench Slowdown
Clarifier Influent
Clarifier Effluent
Cooling Tower
Recycle Water

Quenched Gasifier
Slag
Settled Clarifier
Solids/Clarifier
Underflow
Raw Gas after Blower
Raw Gas prior to
Acid Gas Cleanup
and Shift
7

18
Proximate/Ultimate, Trace elemental survey.

Standard wastewater tests , Trace element
survey. Organic compounds survey, priority
pollutants, Level 2 as needed (POM's).

Standard wastewater tests , Trace element
survey. Organic compounds survey. Level 2
as needed (POM's).

Standard wastewater tests , Trace element
survey. Organic compounds survey. Level 2
as needed (POM's).
*
Standard wastewater tests , Trace element
survey, Organic compounds survey, Level 2
as needed (POM'sj.

Standard wastewater tests*, Trace element
survey, Organic compounds survey, Level 2
as needed (POM's).
*
Standard wastewater tests , Trace element
survey, Organic compounds survey, Level 2
as needed (POM's).
*
Standard wastewater tests , Trace element
survey, Organic compounds survey, Level 2
as needed (POM's).

*
Standard wastewater tests , Trace element
survey. Organic compounds survey, priority
pollutants, Level 2 as needed (POM's).
Standard wastewater tests , Trace element
survey, Organic compounds survey, priority
pollutants. Level 2 as needed (POM's).
Standard wastewater tests , Trace element
survey, Organic compound survey, priority
pollutants, Level 2 as needed (POM's).

RCRA leach test for soluble elements/
substances which may be potentially toxic
(POM's).

RCRA leach test for soluble elements/
substances which may be potentially
toxic (POM's).
Flow rate, temperature, H2
COS, CS?, mercaptans, NHj, ._ ,
higher hydrocarbons, POM s, particulate
matter, H20.

Flow rate, temperature,
COS, CS2, mercaptans, NH,
higher hydrocarbons """
matter, H20.
POM9
CO, COo, H2S,
HCN, methane,
CO, C02, H2S,
HCN methane,
particulate
To corroborate data collected from initial
STE.

To corroborate data collected from initial
STE and to provide background comparisons
for the aqueous process streams.t

To indicate those constituents of crude K-T
gas which are likely to be removed/condensed
with water in this or alternate quench
designs.t

Same as for Washer Cooler Slowdown.!
Same as for Washer Cooler Blowdown.t
To corroborate data collected from initial
STE and to allow constituent material balances
around gasification operations.t

To allow constituent material balances around
gasification operations.t

To indicate solids buildup and consequent blow-
down requirements in the slag cooling circuit
and to allow constituent material balances
around gasification operations.t

To allow constituent material balances around
gasification operations.t

To compare to Clarifier influent in order to
indicate degree of removal of both dissolved
and suspended materials expected during clari-
fication and the possible atmospheric emissions
of volatile substances.t

To indicate possible atmospheric emissions of
volatile substances in Clarifier effluent and
to allow constituent material balances.

To provide an indication of the likely disposal
requirements for K-T solid wastes for facilities
constructed in the U.S. and to be able to
relate data to U.S. coals.t

To provide an indicatior of the likely disposal
requirements for K-T solid wastes for facilities
constructed in the U.S.t
To corroborate initial STE data and to allow
constituent material balances around gasifica-
tion operations.

To allow constituent material balances around
gasification operations.
Standard wastewater tests include: Flow rate, temperature, hardness, conductivity, dissolved oxygen, pH, alkalinity,
total suspended solids, total dissolved solids, BOD, COD, TOC, NH3, SCN", CN~, Cl", sulfur species, phosphorus species.
Bioassay Tests
me future data base may have to include bioassay data to fully determine the requirements for meeting U.S. environ-
mental standards. Such tests would focus on final discharges such as Stream 10 above and any final aqueous effluents.
However bioassay tests on selected in-process streams would have value because the resultant larger data base would
aid in correlating biological toxicity with chemical composition.
-------
gas, which would be helpful in evaluation of potential wastes generated
by K-T designs featuring other gas cool ing/particulate removal systems.
In order to complete constituent mass balances around the gasifier/gas
cooling systems, repeat sampling of the raw gas (after blower) would be
desirable so that a consistent set of data is available.
Also indicated in Table 3 are solid wastes/sludges generated by the
slag quenching operating and by the clarifier unit. The primary concern
with these wastes is the Teachability of specific trace elements and
other potentially toxic substances. Such data are specific to each coal.
Samples can be generated in a test gasifier. The leach test referred to
in the table is that specified in regulations promulgated by the EPA
under the Resource Conservation and Recovery Act of 1976 (RCRA). This
type of data would be used as an indication of disposal requirements/
methods for solid wastes generated by facilities built in the U.S. Con-
ceivably it could also become pertinent to perform bioassay tests in
conjunction with future STE efforts if these data should also be necessary
to understanding the requirements of U.S. facilities.
It should be commented that additional sampling/testing activities
at the Modderfontein facility would have as the primary goal that of
providing basic characterization data on K-T generated wastes so that
control technology requirements for facilities built in the U.S. can be
identified early in the planning stages. It is not intended that any
data resulting from tests of a commercial operating facility at Modder-
fontein be used for the purpose of either promoting or criticizing specific
process designs or operating practices of that facility. The Modderfontein
plant was designed in 1972 to meet the specific environmental requirements
in force at that time.
10
-------
2. PLANT DESCRIPTION

The testing of a Koppers-Totzek (K-T) coal gasification facility was
conducted at the Number 4 Ammonia Plant at Modderfontein, Republic of
South Africa. The plant is owned and operated by AECI Limited and has a
design production rate of 1000 tonnes per day of ammonia. The plant was
commissioned in 1974 and operated in 1978 with an on-stream time of 81%.
This plant utilizes six K-T two-burner coal gasification reactors. A
process schematic for the Number 4 Ammonia Plant showing the various
process modules is presented in Figure 1.
Descriptions of the K-T process in general, operating conditions
specific to the Modderfontein plant, and sampling point locations are
provided in this section. Further detailed discussions of the Modder-
fontein plant can be found in the appended K-K report.

2.1 PROCESS DESCRIPTION
The licensor and developer of the Koppers-Totzek (K-T) gasification
process is K-K of Essen, Federal Republic of Germany. As of 1978, there
were 47 K-T coal gasification modules operating in fifteen plants
throughout the world. All of the K-T gasifiers in operation as of 1978
are used to make synthesis gas as an input stream for the production of
ammonia.
The K-T process operates on an entrained bed principle. It utilizes
a high temperature, atmospheric pressure reaction fueled by a continuous
co-current input stream of coal, oxygen, and steam. The gasification
reactor vessel is a horizontal, ellipsoidal, double-walled steel chamber
with a refractory lining. Two gasifier designs are available. The two-
burner gasifier design has a burner head located on each end of the ellip-
soid. The four-burner gasifier resembles two of the two-burner gasifiers
which intersect one another at a 90° angle. A burner head is located at
each of the ends of the two intersecting ellipsoids. Figure 3 schemati-
cally depicts a two-burner gasifier design of the type employed at
11
-------
RAW PRODUCT
GAS TO WASTE
HEAT BOILER
LOW
PRESSURE
STEAM

y
— a
\ _

COAL
STEAM
OXYGEN
BURNER
COOLING
WATER
BOILER FEED WATER
LOW PRESSURE STEAM
FRESH
INPUT WATER
WASTE WATER
FROM SLAG
QUENCHING
- QUENCHED SLAG
*•
CONVEYOR REMOVAL
FROM PLANT
Figure 3. Koppers-Totzek Gasifier

12
-------
Modderfontein. The reaction gases exit the gasifier vertically from a
port located on top of the gasifier in the center of the ellipsoid.
The process reactants are fed to the gasifier in the following
manner: a continuous screw conveyer feeds the pulverized coal to mixing
nozzles which are located at the ends of the gasifier but which are not
part of the burner head. The coal is then entrained in a premixed
stream of steam and oxygen. The mixture is then injected into the gasi-
fier through sets of two adjacent nozzles comprising each burner head.
Coal ash residue from the gasification process is removed from the
reactor by two mechanisms. Approximately 50 percent of the ash flows
down the gasifier walls as a molten slag and drains into a slag quench
tank where circulating water causes it to shatter into a granular form.
A conveyer lifts the granules out of the quench tank and transports them
out of the plant area. The remainder of the coal ash leaves the gasifier
as a "soot" entrained in the raw product gas. The entrained soot is
largely removed from the gas stream in a water spray tower.
The gasifier operates with a flame temperature of 2000 C (3650 F)
or more, and a gas outlet temperature of 1400° to 1600°C (2550° to 2900°F).
The pressure inside the gasifier is essentially one atmosphere. The
coal is gasified within about 1 second. Opposing burner heads in the
reactor provide for high turbulence and efficient mixing of reactants.
The heterogeneous reactions between carbon, oxygen and steam in the input
stream are generally characteristic of coal gasification. The major con-
stituents of the gasifier output stream are carbon monoxide and hydrogen.
Most of the organic and inorganic sulfur contained in the coal is
converted to HoS and COS at a ratio of about 9 to 1. Smaller amounts of
CSp and S0? are also formed. A portion of the feed coal sulfur is retained
with the ash, with the retention ranging from 5% to over 30% depending on
the coal. Organic nitrogen contained in the coal is converted mainly to
elemental nitrogen, although small amounts of NH3> HCN and NO are also
generated.
Other auxiliary processes and operations of the K-T process include
coal preparation, oxygen production, particulate removal from the raw gas,
and treatment (and recycle) of process water used for gas cooling and

13
-------
cleaning and slag quenching. These auxiliary processes operate more or
less independently of the actual gasification process and do not repre-
sent specialized development or adaption to the K-T process. Consequently
they are not described in detail in this report.

2.2 PLANT OPERATION DURING THE TEST PERIOD
The joint sampling and analysis activities involving K-K, AECI, and
TRW were conducted during the period November 7, through November 29, 1979.
During this period one of the six gasifiers in the plant was not operating.
However, nearly full design capacity was obtained throughout this period
with the remaining 5 gasifiers. All collection of samples and associated
operating data occurred at production rates of between 102,000 and 104,000
normal cubic meters per hour (Nm /h) of dry raw gas. K-K personnel repor-
ted that during the test period the gasification plant operated in a very
stable manner with no process upsets. However, problems were encountered
with the operation of the downstream Rectisol unit for H2S removal which
prevented the collection of sulfur species data on the tail gas stream.

2.3 SUMMARY OF SAMPLE POINT LOCATIONS
Sampling locations for each of the nine streams tested are provided
below. The stream numbers given in parentheses correspond to those shown
in Figure 1. The coal dust sample (7) was taken at the exit of the coal
dust bunker in the coal preparation operation. The raw gas (15) was
sampled from the common duct leading to the raw gas holder. Thus the
raw gas sample represents the average gas composition from all five oper-
ating gasifiers. The Rectisol tail gases (33, 38) were sampled from tap
lines fitted to the respective vent lines. The purified treated sewage
input water (45) was sampled from the main line entering the plant. The
cooling water input (16) was sampled from the pressure line entering the
plant. The cooling water input (16) was sampled from the pressure line
entering the plant. Both of these input water streams originate from
facilities in the Modderfontein complex other than the coal gasification
facility. Condensate from the raw gas compressor was taken from the line
leading to the wash water system which collects the various wastewaters
and conveys them to the clarifier. The hot condensate effluent from the
methanol/water separation column of the Rectisol unit is diluted with
14
-------
cooling water. This diluted condensate (32) was sampled from the line
leading to the wash water system. The settling pond effluent (49) was
sampled at the exit of the channel'which collects the overflow from numer-
ous drain pipes in the pond.
15
-------
-------
3. SAMPLING AND ANALYSIS METHODS

The sampling and analysis responsibilities for the K-T facility test
were divided between TRW and K-K. K-K performed all of the sampling and
most of the on-site analyses. TRW arranged to have the remaining time-
critical analyses performed by a local South African laboratory and to have
portions of the coal feed and aqueous process stream samples shipped back
to TRW for analysis. Table 4 summarizes responsibilities of the parti-
cipants and the following sections describe the methods used by K-K and
TRW for their respective activities.

3.1 K-K METHODS
K-K's responsibilities were for:
t All sampling,
• All gas stream analysis, and
• Much of the coal and aqueous stream analyses.
The methods used in each of these areas are described in the following
paragraphs.
3.1.1 Coal Feed Sampling and Analysis
The coal feed dust was sampled over half-hour periods on November 19
and 23. The November 19 sample was supplied to TRW and the November 23
sample was analyzed by K-K using Deutsche Industrie Norm (DIN) procedures
for the following parameters:
t Particle Size Distribution - DIN 51 704
• Moisture Content - DIN 51 718
• Ash Content - DIN 51 719
• C and H Content - DIN 51 221
• N Content - DIN 51 722
t Total and Combustible S - DIN 51 724
• 0 Content - by difference
• Ash Composition - DIN 51 729.

17
-------
3.1.2 Gas Stream Sampling and Analysis
The sampling procedure for all three gas streams consisted of flow-
ing the gas through a manifold to the various absorption trains over a
two-hour period. Each of these trains contained two to three wash bottles
in a series and a gas meter. In the case of the Draeger tubes used for
CS2 and NOX, these were placed in the stream of H2S free gas eluting from
the cadmium acetate gas scrubber bottles. Each of the gas streams was
sampled only once. The Rectisol tail gases were sampled and analyzed on
November 16 and the raw gas was sampled and analyzed on November 23.
A summary of the gas analysis methods is given in Table 5. General-
ly these methods are of acceptable specificity and accuracy for source
evaluations of this type except for the Draeger tube measurements which
can be subject to interferences from other species present, and the Orsat
method for hydrocarbons which failed to provide adequate detection limits.
Gas chromatography (GC) techniques are preferable for hydrocarbons, how-
ever, problems with the GC equipment available on-site prevented its use
in this source test.
3.1.3 Aqueous Stream Sampling and Analysis
The sampling procedure for all five aqueous streams consisted of
collecting and preserving six samples within a one-hour period on each
of two days a week apart, November 12 and 19. These samples were used
for the determination of suspended solids, one sample was acidified immed-
iately for NH3 analysis, one sample was filtered into an alkali/cadmium
carbonate solution for the analysis of H2S and other acidic species, and
one sample was filtered and used for the remaining analyses.
A summary of the analytical methods used by K-K is given in Table 6.
These methods are essentially equivalent to standard test procedures used
in the U.S. and are acceptable for this type of source evaluation.
18
-------
Table 4. DISTRIBUTION OF SAMPLING AND ANALYSIS RESPONSIBILITIES
K-K
TRW
SAMPLING
ANALYSIS
Level 1
Coal Feed
Gases
Liquids
Priority Pollutants
Liquids
Level 2
Liquids
All
Fixed Gases (CO, C02, 02, NZ, HZ)
Sulfur Species (H2S, COS, CS2, Mercaptans)
Hydrocarbons (C, to Cy)
Wastewater Tests (pH, TSS, TDS, hardness,
alkalinity, conductivity, COD, NH.,, CN~.
SCN
S0
SO,
3 ' JU4
methanol, dissolved oxygen)
PO.
-3
Cl
Trace Element Survey (SSMS*)
Proximate/Ultimate*
,,-t
Wastewater Tests (Nitrates, CN ,
SCN'
S=)
SCN~f, BOD1", CODf, TOC, total phenols,
Organic Screening (volatiles and
base/neutral and acid non-volatiles)
Inorganic Screening (Ag, As, Be, Cd,
Cr, Cu, Hg, Pb, Ni, Sb, Se, Tl, Zn)
PAH compounds
Additional Inorganic quantitation of
Fe and Mn
Subcontracted to Commercial Test and Engineering, Inc.
Subcontracted to McLachlan and Lazar (pty) LTD
-------
Table 5. SUMMARY OF K-K GAS ANALYSIS METHODS
Parameter(s)
Method
, CO, C0?, N£, hydrocarbons
cs2, so2, cos
NOX

Mercaptans

HCN

NhL
Orsat analyzer

Dew point hygrometer

Absorbed in cadmium acetate solution.
CdS precipitate is acidified in pres
ence of iodine and determined iodo-
metrically.
gas is first obtained using
copper acetate. CS? is then determined
by Draeger tube. SO;? iodometrically
after absorption in iodine solution.
COS by difference after determining
total non-H2S sulfur compounds by ab-
sorption in KOH, oxidation with
and precipitation with BaCl2.

Measured with Draeger tubes in H2S-
free gas obtained as above.

GC, Tracer 270HA Sulfur Analyzer using
Tracer "Special" silica gel column.

Absorbed in KOH. The KCN is reacted
with Br2 to yield CNBr. The CNBr is
determined iodometrically. (Ref.
Ruhrgas. A.G.)

Absorbed in H2S04 and determined per
DEV Standard Method. (1)
(1) Deutsch Einheit Vorschriften (a compilation of standard methods)
20
-------
Table 6. SUMMARY OF K-K WATER ANALYSIS METHODS
Parameter
K-K Method
Comparable
American Method
PH

Conductance

Dissolved and Sus-
pended solids

Hardness

Acidity/Alkalinity

Chloride

Sulfide
Sulfite and
Thiosulfate
Total Phosphate
Ammonia in water
Sulfate

Cyanide and
Thiocyanate
Methanol

Ammonium Ion

COD
DEV STD. Method (1)

DEV STD. Method

DEV Method HI and H2

DEV STD. Method

DEV Method Dl, No. 2

Precipitation with CdCOs. CdS pre-
cipitate is determined iodometri-
cally.

lodometric titration of filtrate
from CdS separation determines the
total.S03= is complexed with for-
maldehyde and the SoC^ is titrated
with iodine. 803= is determined
by difference.

DEV D11-1E (Molybdenum Blue)

DEV STD. Method (Make water sample
alkaline. Sparge into std. H2S04,
and back titrate excess H2S04.)

DEV Method D5
(Barium precipitation).

Boyer Method (2) Purge into KOH and
titrate with AgNOj. SCN" remaining
in solution is titrated by Brom-
cyanide method. (DEV method for
HCN in gases.)

GC/Thermal conductivity or GC plus
hydrogenation to methane and FID.

Analyze as NH3 (DEV STD. Method)

DEV Method H4-1A or IB (Chemishe
Saurstoff Bedarf)
APHA 424 (3)

APHA 205

APHA 208-B or C and
APHA 208-D

APHA 309-B

APHA 402/APHA 403

APHA 408

APHA 427

APHA 429
APHA 425

APHA 418

APHA 427

APHA 413
No routine standard
method

APHA 418

APHA 508
(1) Deutsch Einheit Vorschriften (a compilation of standard methods)

(2) This is not a standard method. The procedure is adapted from Gas und
Wasserfach. Vol. 105, Heft. 13, p. 334ff.

(3) American Public Health Association, Standard Methods for the Examination
of Water and Wastewater, 14th Edition, 1976.
21
-------
3.2 TRW ANALYSIS METHODS
TRW's responsibilities were for
• Any Level 1 analyses not included in Krupp-Koppers effort,
t Priority pollutant screening, and
• Level 2 analyses.
The methods used in each of these areas are described in the following
paragraphs.
3.2.1 Level 1 Analysis
Most of the Level 1 analyses that are time critical were performed
by K-K (i.e., all gas analyses and most wastewater quality tests). The
only wastewater quality tests remaining were nitrates and BOD, which were
then handled by a local commercial laboratory in Johannesburg. Replicate
analysis of a few of the species measured by K-K were also performed by
the local laboratory. The methods used by the commercial laboratory were
comparable to U.S. methods and were acceptable for source evaluations. The
analysis of organic materials and trace metals was performed by TRW on
preserved aliquots of the eight aqueous stream samples that were shipped
back to the U.S. The methods used for the Level 1 analyses were taken
from the EPA-IERL/RTP procedures manual (reference 1).
3.2.2 Priority Pollutant Screening Analysis
The analyses for organic priority pollutants were done in three phases.
Volatile, acid extractable non-volatile and base-neutral extractable non-
volatile organics were tested in accordance with the EPA procedures manual
(reference 2). The samples were analyzed by a gas chromatography-mass
spectrometry (GC/MS) system equipped with an INCOS data system. A computer
program was used to screen the data and the final reports were manually
examined and if necessary, modified. The specific parameters utilized in
each of the three phases of the organic priority pollutant screening are
delineated in Table 7. The compounds analyzed for are specifically man-
dated by the EPA procedures and are listed in Tables 8, 9, and 10.
22
-------
Table 7. PRIORITY POLLUTANT SCREENING ANALYTICAL PARAMETERS

METHOD
Organic Extract
Sample Size
GC Conditions:
Column
Temperature
Program

Injector
Jet separator
Ion source
Helium flow
Mass Spec Conditions:
Mass range
Scan up
Scan down
Hold top
Hold bottom
Scan time
Internal Standard (I.S.)

I.S. Amount

VOLATILE SPECIES
PURGE AND TRAP

5 mL

8 '-0.2% Carbowax 1500
60°C-hold 4 min
60°C + 170°C at 8 /min
170 C - hold 12 min
75°C
295°C
240°C
30 mL/min

40 - 540 AMU
1.90 sec
0.00 sec
0.00 sec
0.10 sec
2.00 sec
Bromochl oromethane
1,4-Dichlorobutane
20.0 yg/L
SEMI-VOLATILE
ACIDS
SEMI-VOLATILE
BASE/NEUTRALS
DIRECT INJECTION OF CONCENTRATED ORGANIC EXTRACT

1 yL

6'-l% SP12400A
30°C + 190°C at 8°/min

190°C
250°C
220°C
30 mL/min

40 - 450 AMU
1.90 sec
0.00 sec
0.00 sec
0.10 sec
2.00 sec
D,n-Anthracene
10
10.0 yg/mL Extract

1 yL

6' -3% SP2250
50°C - hold 4 min
50° + 260° at 8°/min

275°C
275°C
250°C
30 mL/min

40 - 450 AMU
1.90 sec
0.00 sec
0.00 sec
0.10 sec
2.00 sec
Dnn-Anthracene
10
10.0 yg/mL Extract
(V)
OJ
-------
Table 8. VOLATILE ORGANIC COMPOUNDS ANALYZED
Compounds Mass used to quantitate
Bromochloromethane (internal standard) 128
Chloromethane SO
D1ch1orod1fluoromethane 101
Bromomethane 94
Vinyl Chloride 62
Chloroethane 64
Hethylene Chloride 88
Tr1chlorofluoromethane 101
1,1-Dichloroethylene 96
1 ,1-Dichloroethane 63
Trans-1,2-D1chloroethylene 61
Chloroform 83
1 ,2-Dichloroethane 98
1,1,1-Trlchloroethane 97
Carbon Tetrachloride 117
BromodlChloromethane 127
1,4-D1chlorobutane (internal standard) 55
1,2-D1chloropropane 112
Trans-1,3-D1chloropropene 75
Trichloroethylene 130
Benzene 79
C1s-l ,3-01 chloropropene 75
1,1,2-Trichloroethane 97
Dibromochloromethane 127
Bromoform 173
Tetrachloroethylene 164
1,1,2,2-Tetrachloroethane 83
To!uene 92
Chlorobenzene 112
Ethyl benzene 106
Table 9. ACIDIC SEMI-VOLATILE COMPOUNDS ANALYZED

Compound Mass used to quantitate
D.p-Anthracene (internal standard) 188
2-Chlorophenol 128
2-Nitro phenol 139
Phenol 94
2,4-Dimethylphenol 107
2,4-Dichlorophenol 162
2,4,6-Trichlorophenol 196
4-Chloro-m-cresol 142
2,4-D1nitrophenol 184
4,6-D1nitro-o-cresol 196
Pentachlorophenol 266
4-Nitrophenol 65
24
T
-------
Table 10. BASE/NEUTRAL SEMI-VOLATILE COMPOUNDS ANALYZED
Compound Mass used to quantuaie
D-p-Anthracene (Internal standard) 188
1,3-Dlchlorobenzene 146
1 ,4-D1chlorobenzene 146
B1i(2-Ch1oroethyl)*ther 93
1,2-Dichlorobenzene 146
Hexachloroethane 117
B1s(2-Chloro1sopropyl)Ether 77
N-N1trosod1-n-Proplyamine 70
Nitrobenzene 123
1,2,4-THchlorobenzene 180
Hexachlorobutadlene 225
Naphthalene 128
B1s(2-Chloroethoxy)Methane 93
Isophorone 82
Hexachlorocyclopentadiene 237
2-Chloronaphthylene 162
Acenaphthylene 162
Acenaphthene 154
Dimethyl phthalate 163
2,6-D1n1trotoluene 63
Fluorene 166
4-Chi orophenylphenylether 204
2,4-D1n1trotoluene 89
1,2-01 phenyl hydrazine 77
Dlethylphthalate 149
N-N1trosodi phenyl ami ne 169
Hexachlorobenzene 284
4-Bromophenoxybenzene 248
Anthracene/Phenanthrene 178
D1-n-Butylphthalate 149
Fluoranthene 202
Pyrene 202
Benzldine 184
Butyl benzylphthalate 149
Bis(2-ethylhexyl)phtha1»te 167
Benzo(a)Anthracene 228
Chrysene 228
3,3'-D1ch1orobenzidine 252
01-N-Octylphthalate 149
Benzo(b)Fluoranthene 252
Benzo(k)F1uoranthene 252
Benzo(a)Pyrene 252
D1benzo(a,h)Anthracene 278
Indeno-1,2,3-(c,d)-Pyrene 276
Benzo(g,h,i)Perylene 276
25
-------
The components eluting from the GC column are monitored by a contin-
uously scanning mass spectrometer. The mass spectra are then stored on
computer disk to be examined at a later date. A computer program which
mimics the manual procedure for qualitative and quantitative analysis of
samples for priority pollutants is used as a first pass analysis of the
data. Before any analyses are attempted, a standard or series of stan-
dards are run by GC/MS. This operation provides the program with three
pieces of information: a reference mass spectrum, a relative retention
time and a relative response factor for each compound. Once these fac-
tors are generated, the samples are analyzed. The standards are also run
on a routine basis during the sample analysis to allow for adjustment of
the relative retention times and relative response factors.
The program tests for each compound in sequence until the list of
compounds is exhausted. The computer outputs the results which are then
manually checked for consistency, completeness and correctness. The in-
ternal standard results are manually examined to assure that the retention
time and peak area are within acceptable limits. The chromatogram is
examined to assure that all components are identified. That is, if a
chromatographic peak is present but is not identified as a priority pol-
lutant, its spectrum is manually examined to assure that it is not a
pollutant. And finally, a general comparison of the program results and
the GC/MS data is made to assure that no inconsistencies exist.
The analysis for the required 13 priority pollutant metals (i.e., Ag,
As, Be, Cd, Cr, Cu, Hg Pb, Mn, Sb, Se, Tl, and Zn) were also performed
in accordance with the EPA procedures manual (reference 2).
3.2.3 Level 2 Analysis
The level 2 analysis of the Modderfontein samples consisted of atomic
absorption techniques (AAS) for Fe and Mn, and a high performance liquid
chromatography (HPLC) technique for polynuclear aromatic hydrocarbon (PAH)
compounds. These two metals and the PAH compounds were selected on the
basis of comparing the Level 1 data to the EPA's discharge multimedia
environmental goal (DMEG) values, thus determining the constituents of
potential environmental concern which warrant further investigation. The
AAS techniques were standard methods (reference 8). The HPLC technique
was developed by TRW and is described here briefly.

26
r
-------
The HPLC technique utilizes a reverse phase, quaternary solvent system
for separation of three-ring and larger PAH compounds. Both UV and fluor-
escence detectors are used in tandem in order to yield corroborative data
for the identification and quantisation of the compounds present. A syn-
opsis of the HPLC parameters is given below.
Apparatus: A DuPont model 850 high pressure liquid chromatograph equipped
with the DuPont variable wavelength UV spectrophotometer and a fluorescence
detector in tandem was used.
Reagents: PAH standards were purchased from several sources. The sources
used included Aldrich, Inc. (San Leandro, CA 94577); Analabs, Inc.
(80 Republic Drive, North Haven, CT 06473), and Chemicals Procurement
Laboratories, Inc. (18-17 130th St., College Point, NY 11356). Chromato-
graphic solvents: methanol, acetonitrile and tetrahydrofuran; were pur-
chased from Burdick and Jackson (Muskegon, Mich.). Water was J.T. Baker
brand HPLC water.
Instrument Parameters:
HPLC Columns 2 DuPont Zorbax® ODS, 4.6 mm ID x 25 cm
(total column length was 50 cm)
Mobile Phase Solvent A - 15% water/85% methanol
Solvent B - Tetrahydrofuran/70% Acetonitrile
Gradient Hold at 5% B for 90 minutes then a linear
gradient to 100% in 20 minutes
Temperature 45 C
Flow 1.0 mL/min.
A number of PAH compounds have been obtained which are used to: 1)
spike the samples in order to determine compounds present by retention
time and relative response to the two detectors, and 2) prepare standard
mixtures for quantitation of the PAH compounds. A list of the compounds
used to identify and quantitate PAHs in the Modderfontein samples is given
in Table 11. Calibration mixtures of the compounds identified in the
samples were prepared and run at four different quantitative levels in
order to bracket the sample concentrations and provide accurate quantitation.
In addition to using the relative response data from the two detectors,
further qualitative data were obtained by collection fractions off the
HPLC and analyzing these by GC/MS. The eluent from the HPLC column was
collected over five minute intervals, which resulted in eleven fractions

27
-------
for each sample. These fractions were evaporated to near dryness at
ambient conditions under a steady flow of argon. Once the concentration
step was completed, the samples were analyzed by GC/MS. A 6 ft (1.9m)
3% Dexsil 300 GC column programmed at 4°C from 100° to 300°C was used to
separate the components in the collected fractions. These data were then
used to confirm identifications and the selection of compounds for the
calibration mixture.
Table 11. PAH COMPOUNDS USED AS STANDARDS
Phenanthrene
Fluoranthene
Pyrene
Chrysene
9-Phenylanthracene
Benzo(b)fluorene
Benzo(k)f1uoranthene
1,2-Benzofluorene
1,2-Benzanthracene
Perylene
Benzo(a)pyrene
Dibenzo(a,c)anthracene
Dibenzo(a,h)anthracene
Picene
o-Phenylenepyrene
Benzo(g,h,i)perylene
9,10-Diphenylanthracene
Dibenzo(a,i)pyrene
5,6,11,12-Tetraphenylnaphthacene
Coronene
Decacyclene
28
-------
4. RESULTS

The data presented in this section are the combined results from

the efforts of both K-K and TRW. The methods used were described in

Section 3, and the division of responsibilities was summarized in
Table 4.

4.1 COAL FEED STREAM

The coal dust sample may be characterized as Bituminous, High Volatile
B coal based on the results of proximate and ultimate analyses, as shown
in Table 12. The coal is very high in ash content and low in sulfur

compared to most U.S. coals. A trace element survey was also performed

on the coal sample, yielding the results shown in Table 13. More pre-
cise determination of the major minerals in the ash, along with particle
size distributions and other measurements, were performed by K-K and can

be found in their report (Appendix A). The differences between the South

African coal and American coals and the effect this has on the composition
of the product, by-product, and waste streams must be kept in mind by
anyone trying to use the data in this STER to assess the characteristics

of K-T facilities that might be built in the U.S.
Table 12. PROXIMATE AND ULTIMATE RESULTS
FROM COAL ANALYSIS
PROXIMATE ANALYSIS
ULTIMATE ANALYSIS
As Rec'd Dry Basis
% Moisture
% Ash
% Volatile
% Fixed Carbon
1.49
19.60
27.52
51.39
100.00
xxxxx
19.90
27.94
52.16
100.00
Btu/lb.(kcal/kg) 10853(6028) 11017(6119)
% Sulfur 0.99 1.01
As Rec'd. Dry Basis
Moisture 1.49
Carbon 64.41
Hydrogen 3.72
Nitrogen 1.12
Chlorine 0.01
Sulfur 0.99
Ash 19.60
Oxygen (diff)
8.66
xxxxx
65.38
78
14
0.01
1.01
19.90
8.78
100.00 100.00
29
-------
Table 13. SSMS RESULTS FROM COAL ANALYSIS
Element
Lithium
Sodium
Potassium
Rubidium
Cesium
Beryllium
Magnesium
Calcium
Strontium
Barium
Boron
Aluminum
Gallium
Silicon
Germanium
Tin
Lead
Phosphorus
Arsenic
Antimony
Bismuth
Sulfur
Selenium
Tellurium
Fluorine
Chlorine
Concentration
(ppm)
71
>500
>500
4
3
0.8
>500
>500
320
>500
3
>500
17
>500
0.5
2
7
>500
4
0.4
ND
>500
1
0.3
310
15
Element
Bromine
Iodine
Scandium
Yttrium
Titanium
Zirconium
Vanadium
Niobium
Tantalum
Chromium
Molybdenum
Tungsten
Manganese
Iron
Cobalt
Nickel
Copper
Silver
Zinc
Cadmium
Lanthanum
Cerium
Praseodymium
Neodymium
Thorium
Uranium
Concentration
(ppm)
0.8
0.8
1
26
>500
100
15
10
ND
51
2
ND
26
>500
2
3
5
ND
1
ND
21
52
5
10
9
6
ND = Not detected (average detection limit is 0.2ppm)
30
-------
4.2 GAS STREAMS
The three gas streams studied were sampled per the schedule shown in
Table 1. The engineering data obtained were as follows:
Stream/Stream Number (from Figure 1) Flow Rate Temperature
Raw gas after raw gas blower/15 28.3 Nm /sec 46 C
Tail gas from H2S absorber/38 3.8 Nm3/sec 27°C
Tail gas from C02 absorber/33 13.6 Nm3/sec 29°C
4.2.1 Analysis Results
All gas analyses were performed by K-K and the data obtained are
shown in Table 14. The raw gas results reflect the average composition
from all five operating gasifiers (the stream was sampled at a common line
leading to the gas holder) after the gas has been water-washed for parti-
culate removal. A description of the reactions that take place in the raw
gas washing stages is as follows:
• NHo» HCN, SCL, and to a small degree H^S and CCL, are dissolved
in the wash water.
t HpS is eventually converted to S^O^", S0.~, and insoluble iron
sulfides due to the pH, temperature and flyash content of the water.
t HCN reacts with the sulfur compounds to form SCN~ and with the
iron content of the flyash to form insoluble complexes.
• Additional oxidation reactions occur which are catalyzed by the
flyash involving NH , SO =, S203~, CN", and SON".
The main components in the water-washed gas are then H^O, CO, CQ2, H2, and
N2> Data on hydrocarbons contained in the raw gas stream were not obtained,
but low concentrations would be expected due to the high temperature of the
K-T gasification reaction. Hydrocarbon data from previous tests of the
Modderfontein plant under comparable conditions are provided by K-K in
Appendix A.
The two tail gas streams from the Rectisol module consist primarily of
C02 and the nitrogen used for stripping along with some CO and H20, and traces
of NH3 and HCN. During the test period the H2S absorber was not operating
properly and thus sulfur species data on this tail gas stream were not made
available. A design value of <2 ppm total sulfur is quoted by K-K but this
cannot be confirmed.
31
-------
Table 14. RESULTS FROM GAS ANALYSES
Species
H20
H2, (dry)
CO, (dry)
C02, (dry)
N2/Ar*. (dry)
CH4, (dry)
H2S, (dry)
COS, (dry)
CS2, (dry)
Mercaptans, (dry)
S02, (dry)
NH3, (dry)
HCN, (dry)
NOX (as N02), (dry)
Raw Gas after
Raw Gas Blower
mg/Nm3
5.4 X 104
2.3 X 104
6.9 X 105
2.0 X 105
2.1 X 104
<7 X 102
6.3 X 103
7.4 X 10
4.5 X 102
<1
14
57
76
28
Tail Gas from
H2S Absorber
mg/Nm^
5.0 X 103
<10
2.2 X 104
9.6 X 105
5.3 X 105
<7 X 102
t
t
t
<1
<3
39 39
62
<1
Tail Gas from
C02 Absorber
mg/Nm^
5.0 X 103
<10
3.0 X 103
1.6 X 106
1.8 X 105
<7 X 102
<1
<3
<10
<1
<3
3.0
8.0
<1
* By difference
t Not determined
32
-------
4.2.2 §ource Analysis Model Results
The analytical data were used to perform Source Analysis Model/IA
(SAM/IA) calculations (4). This model, developed by the EPA as part of
their standardized methodology for interpreting STE results, assesses the
potential health and ecological effects of discharge streams based on
chemical constituents. In performing SAM/IA evaluation, different eval-
uation indices may be calculated:
• Discharge Severity (DS)
• Total Discharge Severity (TDS)
• Weighted Discharge Severity (WDS)
• Total Weighted Discharge Severity (TWOS)
The DS is calculated by dividing the measured concentration of a com-
pound or class of compounds by the Discharge Multimedia Environmental Goal
(DMEG) values (5). There are two DMEG values for each compound or class.
One is based on health effects while the second is based upon ecological
effects. When a concentration is known only for a class of compounds, then
the evaluation is made using the lowest DMEG value in the class. A DS
value greater than one indicates a level of potential concern, while a
value less than one indicates little or no potential concern. A total
stream discharge severity (TDS) is calculated by summing the OS's for all
constituents found in the stream sample. A Total Weighted Discharge
Severity is then calculated by multiplying the TDS by the stream flow rate.
Because TWDS's incorporate stream flow rate data, they are useful indices
for ranking the waste stream from a facility in terms of their potential
environmental concern.
The results of calculating DS values for the two waste gas streams,
are summarized in Table 15. In the tail gas stream from the H2S absorber,
CO, HCN and NH_ are present at levels of potential concern; and in the tail
gas from the COp absorber, CO and NH3 are of potential concern. The TDS
values for the H^S absorber and C02 absorber tail gases are listed in
Table 16 along with TWOS values. Unfortunately the lack of sulfur species
data for the H2S absorber tail gas and the lack of stated detection limits
for many other species limits the usefulness of these calculations.
33
-------
Table 15. SUMMARY OF SAM/IA DS RESULTS
FOR GAS STREAMS
Gas
Species
CH4
CO
COS
cs2
RSH
H2S
so2
HCN
NH3
NOX
Discharge Severity (DS)
Health-Based
Tail Gas
from H?S
Removal
NM
5.5 E + 02
NM
NM
ND
NM
ND
5.6 E + 00
2.2 E + 00
ND
Tail Gas
from C02
Removal
NM
7.5 E + 01
ND
ND
ND
ND
ND
7.3 E - 01
1.7 E - 01
ND
Ecology-Based
Tail Gas
from H2S
Removal
NM
1.8 E + 02
NA
NA
NA
NA
NA
1.8 E + 00
1.1 E + 02
NA
Tail Gas
from C02
Removal
NM
2.5 E + 01
NA
NA
NA
NA
NA
2.5 E - 01
8.6 E + 00
NA
NM - Not Measured, no data was collected on these species

ND - Not Detected, no measurable amount of this species was found
NA - Not Applicable, no ecology DMEG value for this species.

Table 16. SUMMARY OF SAM/IA TDS AND TWOS RESULTS FOR GAS STREAMS
TDS and TWOS Values
Total Discharge Severity (TDS)
Health-Based
Ecology-Based
Total Weighted Discharge Severity
(TWOS)
Health-Based
Ecology- Based
Tail Gas from
H_S Removal

5.6 E + 01
2.9 E + 02

2.1 E + 03
1.1 E + 03
Tail Gas from
C0? Removal

7.6 E + 01
3.4 E + 01

1.3 E + 03
4.6 E + 02
34
-------
There are several assumptions implicit in the use of the SAM/IA eval-
uation technique. The major assumptions include:
t Transport of the components in the waste stream to the external
environment occurs without chemical or physical transformation
of those components.
t Actual dispersion of a pollutant from a source to a receptor will
be equal to, or greater than, the safety factors normally applied.
t The DMEG values developed for each substance are adequate for
estimating acute toxicity.
• No synergistic effects occur among the waste stream components.
Because of the uncertainties introduced by these assumptions, the SAM/IA
results should be used only as a very qualitative assessment. To more fully
determine the potential concern of any stream requires that biological
tests as well as chemical tests be evaluated.

4.3 AQUEOUS STREAMS
The five aqueous streams studied were sampled per the schedule shown
in Table 1. The engineering data obtained were as follows:
Stream/Stream Number (from Figure 1) Flowrate Temperature
Input Water (PSE)/46 215 m3/hr 23°C
Input Water (CW)/16 54 m3/hr 30°C
Compressor Condensates/40 9.2 m /hr 33°C
Settling Pond Effluent/50 230 m3/hr 23°C
Condensate from Rectisol Unit/32 3.9 m3/hr 52°C
4.3.1 Analysis Results
The analyses performed on the aqueous process streams can be cate-
gorized as follows:
• Level 1
• Wastewater Tests
• Organic Survey
• Inorganic Survey
• Level 2
• Detailed Organic Characterization
t Detailed Inorganic Characterization
35
-------
• Priority Pollutant Screening
• Volatile Organics
• Base/Neutral and Acid Organics
t Trace Elements
Results from each of these categories are presented in the following
paragraphs.
4.3.1.1 Level 1 Analysis Results
The wastewater analysis results from K-K, TRW, and McLachlan & Lazar
(M&L) are summarized in Tables 17 and 18. The relationship of these
streams to each other is best illustrated in Figure 1. The compressor
condensate and Rectisol unit water streams are process streams, they are
in effect diluted with fresh (PSE) input water before being recycled through
the cooling tower. The only waste stream discharged from the plant is the
settling pond effluent which appears from the data to be quite similar in
composition to the input waters (purified treated sewage and cooling water).

Table 17. RESULTS FROM K-K WASTEWATER ANALYSES
Hastewater Tests/Units of Measure
pH
Total suspended solids, mg/L
Total dissolved solids, mg/L
Hardness, mg/L as CaC03
Alkal1n1ty:p-Value, mg/L as CaCOj
m- Value, mg/L as CaCOj
Conductivity, umhos/cm
COO. mg/L
NH3, mg/L
CN', mg/L
SCN", mg/L
H2S, mg/L
S203", mg/L
SOj". mg/L
S04~~. mg/L
PO -3, ng/L
Cl", mg/L
Methane 1 , mg/L
Dissolved oxygen, mg/L
ll/l.1
Input.
Water
(I'M.)
6.R
<1
1580
452
0
26
2300
38
73
0.2
2.1
<1
<1
<1
584
10
135
ND
ND
11/19
Input
Water
(CW)
8.5
8
1460
621
0
167
1900
118
2.4
1.2
2.1
traces
<1
<1
853
2.4
172
ND
ND
11/12
Compressor
Condensate
R.2
<1
260
60
0
2990
6000
644
973
7.3
10.9
43.9
4.8
<1
56
2
23
ND
ND
11/19
Compressor
^ondensate
8.0
12
170
46
0
2690
5500
569
900
10.5
17.1
53.5
7.8
<\
49
3
13
ND
NO
11/12
Settl inq
Pond
8.0
<1
1580
420
0
126
2100
353
38
0.2
1.3
<1
<1
3.6
752
5
145
ND
ND
11/10
Settling
Pond
9.4
<1
1530
664
44
79
2100
43
28
0.2
2.2
<1
<1
<1
706
0.4
163
ND
ND
11/12
Rectisol
Unit
9.1
70
1390
691
,0
78
1800
28
26
2.8
110
1.1
18.5
<1
461
0.1
153
<0.1
ND
^^s=sss
11/1')
Rert iiol
Unit
8.1
20
1640
554
0
144
2000
1000
49
ND
137
4.5
16.4
<1
541
2.8
158
<0.1
ND
ND * Not Determined
36
-------
Table 18. RESULTS FROM TRW AND M&L WASTEWATER ANALYSES
S-plin,
Day
Nov. 12
Hov. 19
Nov. 12
Nov. 19
Nov. 12
Nov. 19
Nov. 1?
Nov 19
Stre*» Descnpt ion/btre*n. NuflAei •
Input Uiter--Purtfted T ruled Sewge/46
Input Utter— Cool In) Hater/16
Combined Condensitts fro» I1--4 Compressors/40
Co«>tned Condenutes fro* n--t Compressors/ 40
Settling Pond Effluent/60
Settling Pond Effluent/so
Condensate fro» Recttsol Unit/32
Conden&4te fro* RecMsol Un1t/32
BUO"
(•9/U
5
4
620
480
5
3
120
COD"
(•g/U
16
24
670
540
4
4
3iO
1490 ] 2830
Phenols
(«9/U
0.005
0.020
0.006
0.012
<0 001
0 014
0 010
0 034
IOC
(•9A. )
31
16
130
140
5 2
!> 4
16
600
Sutfide
100°C) materials. As the data in
Table 19 show, the total organic loading (volatiles plus nonvolatiles) is
low, and what is there is primarily nonvolatile. Examination of the
nonvolatile material by infrared (IR) spectroscopy indicates that the
classes of compounds present in all of the samples are primarily saturated
hydrocarbons along with some esters. There is also some IR evidence of low
levels of aromatic hydrocarbons present in the compressor condensate and
Rectisol unit samples.
Examination of the nonvolatile portion of the samples by mass spectro-
scopy yielded additional information regarding the types of compounds present.
The intensity of the mass spectra peaks were used to assign relative con-
centration levels (100 - major, 10 - minor, 1 - trace) to the compound classes
identified. These concentration factors were then applied to the total
37
-------
Table 19. RESULTS FROM LEVEL 1 ORGANIC SURVEY
Sampl i ng
Day
Nov. 12
Nov. 19
Nov. 12
Nov. 19
Nov. 12
Nov. 19
Nov. 12
Nov. 19
Stream Description/Stream Number*
Input Water—Purified Treated Sewage/ 46
Input Water—Cooling Water/16
Combined Condensates from #1—4 Compressors/ 40
Combined Condensates from #1 — 4 Compressors/ 40
Settling Pond Effluent/50
Settling Pond Effluent/ 50
Condensate from Rectisol Unit/32
ICondensate from Rectisol Unit/32
Volatiles
(nq/L)
0.04
<0.01
-------
Table 20. DISTRIBUTION OF COMPOUND CLASSES IN AQUEOUS SAMPLES
Sampling
Day
Nov. 12

Nov. 19

NOV. 12

Nov. 19

Nov. 12

Nov. 19

Nov. 12

Nov. 19

Stream Description/Stream Number
Input Water--Purified Treated
Sewage/46

Input Water--Cooling Water
/16
Compressor Condensates/40

Compressor Condensates/40

Settling Pond Effluent/50

Condensate from Rectisol Unit/32

1
Compound Category
Primary Alcohols
Esters (phthalates)
Nitro Aromatic Hydrocarbons
Esters (phthalates)

Chlorinated Cresols
Esters (phthalates)
Fused Polycyclic
Hydrocarbons, 2-3 Rings
Esthers
Sulfur (Sg)
Phenols
Chlorinated Phenols
Esters (phthaUtes)
Fused Polycyclic
Hydrocarbons, 2-3 Rings
Carboxylic Acids
Sulfur (S.)
0
Primary Alcohols
Aliphatic Hydrocarbons,
Alkynes
Unsaturated Alkyl Halides
Secondary Alcohols
Esters (phthalates)
Aliphatic Hydrocarbons,
Alkenes
Unsaturated Alkyl Halides
Secondary Alcohols
Ketones
Esters (phthalates)
Aliphatic Hydrocarbons,
Alkanes
Fused Polycyclic
Hydrocarbons, 2-3 Rings
Fused Polycyclic
Hydrocarbons , 4 Rings
Sulfur (Sfi)
o
Aliphatic Hydrocarbons,
Alkanes
Phenols
F"Sed Polycyclic
Hydrocarbons, 2-3 Rings
Fused Polycyclic
Hydrocarbons, 4 Rings
Fused Non-Alternate
Polycyclics
Esters (phthalates)
Sulfur (Sg)
MEG
Category
5A
8D
17A
8D

19B
8C

21A
3A
--
ISA
19A
8D

21A
8A
--
5A

1C
2B
5B
8D

IB
2B
5B
7B
8D
1A

21A

218
--

1A
ISA

21A

21B

Z2B
8D
"~
Approximate
Concentration, gg/L
60
560
60
880

20
190

20
1,960
1,960
30
30
310

30
30
3,080
55

5
0.5
55
5

0.09
0.9
9
0.09
0.09
10,190

100

100
1,020

48,130
480

480

480
480
4,810
wastewater analyses results, the settling pond effluent is quite similar
to the input water. The only trace elements that shown an increase in
concentration are Cs, Sr, Ba, Ga, and Mo. Other elements (i.e., Al, Fe
and Mn) actually show a significant decrease in the pond effluent versus
the input waters.
39
-------
Table 21. RESULTS FROM LEVEL 1 SSMS ANALYSIS

Lithium
Sodium
Potassium
Rubidium
Cesium
Beryl 1 ium
Magnesium
Calcium
Strontium
Bari urn
Boron
Aluminum
Gall lutn
Silicon
Germanium
Tin
Lead
Phosphorus
Arsenic
Antimony
Bismuth
Sulfur
Selenium
Tellurium
Fluorine
Chlorine
Bromi ne
Iodine
Scandium
Yttrium
Titanium
Zirconium
Vanadium
Niobium
Tantalum
Chromium
'lolybdenum
Tungsten
Manganese
Iron
Cobalt
Nickel
Copper
Silver
Zinc
Cadmium
Lanthanum
Ceri urn
Praseodymium
Neodymium
Thorium
Concentration in Process Water Samples (ug/L)
11/12
Input Water
(PSE)
100
>2.000
'9,000
30
1
<1
> 10,000
> 10, 000
1,000
80
30
> 700
6
6,000
7
12
30
8,000
70
10
ND
> 4,000
20
ND
> 10. 000
300
100
30
<1
6
400
5
5
1
ND
7
30
ND
900
200
20
100
100
ND
2
ND
8
20
2
2
<6
Uranium j 6.000
40
ND
ND
> 10, 000
> 10. 000
200
400
<1
500
2
2,000
3
ND
400
2,000
9
ND
ND
>3,000
3
ND
= 700
300
40
9
< 1
O
100
1
20
2
ND
5
5
ND
20
1,000
<1
8
300
ND
*
ND
1
<1
ND
ND
<4
20
11/12
Compressor
Condensates
3
2.000
> 10. 000
8
<1
ND
1,000
> 10, 000
70
100
<1
5
<1
300
5
2
20
70
2
12
2
>6.300
500
3
"30
100
80
4
<_!
<1
30
3
2
2
2
5
30
ND
10
500
3
4
10
<2
600
3
ND
ND
ND
ND
<8
<7
11/19
Compressor
ondensates
10
> 1.000
> 10, 000
3
ND
ND
2,000
> 10, 000
30
40
<1
9
3
100
7
<1
30
70
4
10
ND
> 2.000
1,000
3
= 3,000
60
300
8
< 1
1
200
10
<1
5
ND
5
20
ND
9
1,000
1,000
>6,000
100
6
ND
> 10, 000
> 10. 000
8.000
200
2
20
40
400
1
1
4
80
8
4
ND
> 3.000
2
ND
»7on
70
100
20
< 1
<1
60
1
9
6
ND
<_!
50
10
200
50
• 4
8
7
ND
30
8
ND
ND
ND
ND
11/19
Settling
Pond
<1
> 1,000
>6,000
40
7
ND
> 10, 000
'10,000
7.000
200
<1
200
40
2,000
'1
1
2
80
9
2
ND
>3,000
6
8
"2,00.1
311
300
20
<1
<1
500
1
10
1
ND
2
100
10
200
100 '
3
20
6
ND
30
8
ND
<1
ND
ND
<4 | <4
<3 ' <3
11/12
Rectisol Unit
6
'1,000
>5.000
9
' 1
ND
> 10,000
> 10, 000
300
200
' 1
100
'1
1,000
'1
4
20
700
20
< 1
ND
>2.000
50
ND
• 400
200
30
8
<1
<1
100
<1
3
2
ND
3
40
ND
50
> 10, 000
1
200
50
ND
6,000
1
ND
<1
ND
ND
<3
6
11/19
Rectisol Unit
3
>2,000
> 10,000
6
ND
ND
> 10, 000
> 10, 000
500
200
ND
100
<1
2,000
2
ND
10
1,000
10
ND
ND
>4,000
40
ND
= 3.0cn
300
60
9
< 1
<1
200
<_!
5
-------
As with the organic data, the data from the inorganic survey were
compared to the DMEG values for each species. This resulted in finding
that Cd, Cu, Fe, Mn, Ni, P (as P04=), Pb, S, Se, and Zn exceeded their DMEG
values in most of the samples. These elements thus became subject to
further investigation as is described in the section discussing Level 2
analysis (Section 4.3.1.2).
4.3.1.2 Level 2 Analysis Results
As is mentioned in the preceeding paragraphs, the Level 1 data were
compared to the EPA's Discharge Multimedia Environmental Goals (DMEGs) in
order to determine which species were present at potential levels of con-
cern and were thus candidates for further investigation. From the organic
survey, phenols, cresols, chlorinated phenols and cresols, phthalate esters,
and aromatic hydrocarbons were determined to be of concern (i.e., present
at concentrations greater than their DMEG values). The Level 2 analytical
needs for these materials were thus to identify the specific compounds pre-
sent and more accurately quantify the concentrations. From the inorganic
survey the elements Cd, Cu, Fe, Mn, Ni, P, Pb, S, Se and Zn were determined
to be present at levels of concern. The Level 2 needs for trace elements
required more accurate quantitation, and for major constituents such as S
and P included speciation of the various anions. The best approach to
satisfying these additional investigation needs, within the overall con-
straints of the project, was evaluated with the following results:
• The identification and quantitation of phenols, cresols, chlorinated
phenols and cresols, and phthalate esters would be accomplished
as part of the priority pollutant screening.
t The identification and quantitation of aromatic hydrocarbons
would be addressed as a separate, specific analysis.
• The quantitation of Cd, Cu, Hg, Ni, Pb, Se and Zn would be
accomplished as part of the priority pollutant screening.
• The quantitation of Fe and Mn would be addressed as a separate,
specific analysis.
• Sulfur speciation (i.e., SCN", H2S, S^", S04~) had been adequately
addressed as part of the wastewater analysis.
t Phosphorus speciation could not be addressed because adequate
samples had not been collected and stabilized for that purpose.

41
-------
Thus, of the Level 2 data needs identified, most are addressed and
reported in the priority pollutant screening results (Section 4.3.1.3) and
wastewater analysis results (Section 4.3.1.1). Aromatic hydrocarbons,
iron, and manganese required additional specific analyses, the results of
which are reported in the following paragraphs. Determination of phosphorus
species could not be accomplished due to the lack of appropriately stabi-
lized samples. Hopefully the need for phosphorus species data can be
addressed in a future source test effort.
The high performance liquid chromatograph (HPLC) is a very useful
analytical tool.for polynuclear aromatic hydrocarbon (PAH) compounds. The
technique separates by functionality thus allowing the aromatics to be
separated from the large quantities of aliphatics present. HPLC also is
not limited, as gas chromatography is, by the volatility of the compounds
to be analyzed. Even large PAH compounds such as decacyclene (MW 450) can
be determined.
The two Rectisol unit water samples were analyzed by HPLC, yielding the
chromatograms shown in Figures 4 and 5. The composition of the two samples,
even though they were obtained a week apart, is essentially the same. The
trace at the bottom of each figure is the response to a fluorescence detec-
tor (which is very specific for PAHs) and the trace at the top is the res-
ponse to an ultra-violet detector. The ratio of the response to the two
detectors along with the retention times was the means for determining the
identity of the compounds present. Those compounds which were positively
identified are indicated in the legends of Figures 4 and 5. The unknowns
did not correspond to any of the standards available (see Table 11) and
thus could not be positively identified. In order to obtain some indica-
tion of what these compounds might be, the HPLC column eluent was collected
and analyzed by gas chromatography/mass spectroscopy (GC/MS). The GC/MS
cannot identify different isomers but did indicate that the following types
of compounds are possible present:
• Compound "F" - a methylbenzofluorene
• Compounds "G" and "H" - methylbenzanthracenes
• Compounds "J" and "K" - unknown.
It should be noted that the very toxic compound, benzo(a)pyrene, is one of
the standards used thus compounds "J" and "K" are clearly some other isomer.

42
-------
Fluoranthcne
Pyrene
1,2-Benzofluorene
1,2-Benzanthracene
Unknown (MW 230)
Unknown MW 242
Unknown (MW 242)
Benzo(k)fluoranthene
Unknown (MW 252)
Unknown (MW 252)
Sulfur
120
i
100
I
90
80
70
Figure 4. HPLC Chromatogram of November 12, 1979 Rectisol Unit Sample
-------
B Fluoranthene
C Pyrene
D 1,2-Benzof! uorene
E 1,2-Benzanthracene
F Unknown (MW 230^
G Unknown (MW 242)
H Unknown (MW 242)
I Benzo(k)fluoranthene
J Unknown (MW 252)
K Unknown (MW 252)
L Sulfur
Figure 5. HPLC Chromatogram of November 19, 1979 Rectisol Unit Sample
-------
The compounds that were positively identified were quantitated,
yielding the results shown in Table 22. Those compounds which overlap
with the priority pollutant screening (i.e., fluoranthene and pyrene) are
more accurately quantitated by the HPLC technique. The priority pollutant
screening also identified a four-ringed compound as chrysene which in the
HPLC analysis was determined to be 1,2-benzanthracene (also four-ringed).
The DMEGs for the compounds identified range from 670 yg/L to
24,000 yg/L for the health-based values and an ecology-based value of
100 yg/L. Thus the levels measured would not be considered potentially
toxic. However, direct biological tests could be performed to confirm this,
Table 22. RESULTS FROM LEVEL 2 ANALYSIS OF AROMATIC HYDROCARBONS
Compounds Identified
Fluoranthene
Pyrene
1,2-Benzofluorene
1,2-Benzanthracene
Benzo ( k ) f 1 uoranthene
Nov. 12, 1979
Rectisol Unit
Sample
24 yg/L
32 yg/L
15 yg/L
23 yg/L
2 yg/L
Nov. 19, 1979
Rectisol Unit
Sample
17 yg/L
25 yg/L
15 yg/L
16 yg/L
2 yg/L
The Level 2 inorganic analyses to quantitate the elements Fe and Mn
were quite straightforward compared to the HPLC analysis. Routine atomic
adsorption techniques were used, yielding the results shown in Table 23.
These data, along with the priority pollutant screening data for Ag, Tl,
Sb, As, Se, Zn, Pb, Cd, Ni, Be, Cu, Cr and Hg; were used instead of the
less accurate Level 1 SSMS survey data in computing the SAM/IA results
discussed in Section 4.3.2.
45
-------
Table 23. RESULTS FROM LEVEL 2 QUANTITATION OF Fe AND Mn
Sampling
Day
Nov. 12
Nov. 19
Nov. 12
Nov. 19
Nov. 12
Nov. 19
Nov. 12
Nov. 19
Stream Description/Stream Number*
Input Water - Purified Sewage Effluent/46
Input Water - Cooling Water/16
Combined Condensates from #1 - #4 Compressors/40
Combined Condensates from #1 - 14 Compressors/40
Settling Pond Effluent/50
Settling Pond Effluent/50
Condensate from Rectisol Unit/ 32
Condensate from Rectisol Unit/32
Concentrations, ppb
Fe
<100
700
500
1800
175
100
4600
3400
Mn
1250
<50
<25
<25
850
580
50
50
* Stream numbers correspond to those shown in Figure 1.
4.3.1.3 Priority Pollutant Screening Analysis
Very few of the 116 organic priority pollutant compounds were found,
as shown in Table 24. Those that were present were mostly at very low
concentrations. The level of concern determined by the EPA's Effluent
Guidelines is 10 ug/L. The fact that few compounds were detected and that
those which are present are mostly below this level of concern is evidence
of the relatively acceptable composition of the streams tested, particular-
ly the settling pond effluent. The quality of the priority pollutant
screening data is believed to be quite satisfactory. The only exceptions
to this are the two Rectisol Unit samples. These samples both contain
large amounts of normal and branched saturated hydrocarbons in a molecular
weight distribution ranging from C... up through and exceeding C-Q. These
compounds have been quantified as part of the Level 1 analysis previously
described. However, because of the very large amounts of aliphatic hydro-
carbons compared to the total organic content of the Rectisol unit samples,
it is possible that other nonvolatile priority pollutants may be present
at low yg/L levels in these two samples but are completely masked.
46
-------
Table 24. RESULTS FROM ORGANIC PRIORITY POLLUTANT SCREENING
Sampling
Day
Nov. 12

Nov. 19
Nov. 12
Nov. 19

Nov. 12
Nov. 19
Nov. 12

Nov. 19

Stream Description/Stream Number*
Input Water - Purified Sewage Effluent/46

Input Water - Cooling Water/16
Combined Condensates from #1-4 Compressors/40
Combined Condensates from #1-4 Compressors/40

Settling Pond Effluent/50
Settling Pond Effluent/50
Condensate from Rectisol Unit/32

Condensate from Rectisol Unit/32

Priority Pollutant Compounds Found
Base/Neutral Fraction
Compound
Nitrobenzene
1 ,2 ,4-Tri chl orobenzene
Isophorone
Bis (2-Ethylhexyl)phthalate
Di-n-octylph thai ate
Butyl benzyl phthal ate
Naphthalene
Naphthalene
Diethylphthalate
Di-n-butylphthalate
Butyl benzyl phthal ate
None Detected
Butyl benzyl phthal ate
Naphthalene
Fluorene
Anthracene plus phenanthrene
Fluoranthene
Pyrene
Butyl benzyl phthal ate
Acenaphthalene
Dimethylphthalate
Fluorene
Diethylphthalate
Anthracene plus phenanthrene
Fluoranthene
Pyrene
Chrysene

ug/L
T
T
T
T
T
T
T
T
T
6.0
T

T
T
T
T
6.3
25
T
T
T
1.0
T
4.6
19
97
34-
Acid Fraction
Compound
None Detected

None Detected
4-Chloro-m-Cresol
Phenol

None Detected
None Detected
None Detected

Phenol
2,4-Dimethylpbenol

ug/L

2.3
T

T
T

Volatiles
Compound
None Detected

Chloroform
None Detected
Chloromethane

None Detected
Chloroform
Chloroform

Chloroform

ug/L

7.8

T
T

T = Trace (
-------
The priority pollutant metals screening involves the analysis of 13
elements each of which has its own level of concern. These elements and
the corresponding levels of concern which have been defined by the EPA are:
Ag - 5 ppb, Tl - 50 ppb, Sb - 100 ppb, As - 25 ppb, Se - 10 ppb, Zn -
1000 ppb, Pb - 25 ppb, Cd - 5 ppb, Ni - 500 ppb, Be - 50 ppb, Cu - 20 ppb,
Cr - 25 ppb, and Hg - 1 ppb. The results obtained from atomic adsorption
and emission spectroscopy analyses for these 13 elements are presented in
Table 25. The data show that the process waters (compressor condensate and
Rectisol unit samples) frequently exceed the levels of concern particularly
for Se, Zn, Cu and Hg. However, as was noticed in the Level 1 SSMS in-
organic survey, the discharged stream (settling pond effluent) is relatively
clean compared to both the process streams and the input water (purified
sewage effluent).
4.3.2 Source Analysis Model Results
The analytical data were used to perform Source Analysis Model/IA
(reference 4) calculations. This model, developed by the EPA as part of
their standardized methodology for interpreting STE results, assesses the
potential health and ecological effects of discharge streams. It uses
concentrations of chemical constituents to calculate a Discharge Severity
(DS), Total Discharge Severity (TDS), and Total Weighted Discharge
Severity (TWOS). The method for calculating these indices and the assump-
tions contained in the model are described in Section 4.2.2 and will not
be repeated here.
The results of calculating TDS and TWOS values for the aqueous streams
are summarized in Table 26. It should be noted that the only true dis-
charge stream is the settling pond effluent. The input water is provided
as a background value. The process streams (compressor condensate and
diluted Rectisol condensate) were also evaluated as an indication of the
relative potential concern of the streams produced.
The fact that the health based values for the aqueous input and dis-
charge streams reflect a potential concern is due mainly to Mn and Fe and
to a lesser extent P. The ecology-based values are almost entirely due
to P. The ecology DMEG value for P and its various anions as a class of
compounds is extremely low (0.5 yg/L) and thus easily becomes the most
significant value obtained in SAM/IA calculations. However ecology-based
48
-------
Table 25. RESULTS FROM INORGANIC PRIORITY POLLUTANT SCREENING
Samples
11/12 Input Water (PSE)
11/19 Input Water (CW)
11/12 Compressor Condensates
11/19 Compressor Condensates
11/12 Settling Pond Effluent
11/19 Settling Pond Effluent
11/12 Rectlsol Unit Condensate
11/19 Rectisol Unit Condensate
Trace Element Concentrations, in parts per billion
Silver
<1
<1
<1
<1
<1
<1
2
<1
Thallium
<5
<5
<5
<5
<5
<5
<5
<5
Antimony
10
<3
<3
<3
5
<3
4
<3
Arsenic
33
-5
-5
<5
12
6
10
12
Selenium
<2
<2
3500
3500
2
3
15
36
Zinc
660
3500
310
230
<100
<100
2400
2700
Lead
50
28
32
5
<5
<5
19
7
Cadmium
1.3
<0.5
0.6
-0.5
<0.5
2.2
-0.5
<0.5
Nickel
180
'10
<10
<10
20
'10
220
160
Beryl 1 ium
0.6
<0.5
<0.5
<0.5
3.4
<0.5
<0.5
<0.5
Copper
78
43
10
52
<5
10
110
71
Chromium
<5
7
5
6
7
<5
7
6
Mercury
0.5
<0.2
360
140
<0.2
<0.2
33
13
VO
-------
Table 26. SUMMARY OF SAM/IA TDS AND TWOS RESULTS FOR AQUEOUS STREAMS
en
o
Sampling
Day
Nov. 12
Nov. 19
Nov. 12'
Nov. 19
Nov. 12
Nov. 19
Nov. 12
Nov. 19
Stream Description/Stream Number *
Input Water - Purified Sewage Effluent*/^
Input Water - Cooling Water /16
Combined Condensates from #1-4 Compressors/40
Combined Condensates from #1-4 Compressors/40
Settling Pond Effluent/50
Settling Pond Effluent/50
Condensate from Rectisol Unit/32
Condensate from Rectisol Unit/32
Total Discharge Severity (TDS)
Health-Based
9.8 E + 00
6.7 E + 00
1.1 E + 02
6.3 E + 01
6.9 E + 00
5.2 E + 00
3.6 E + 01
3.6 E + 01
Ecology-Based
1.6 E + 04
4.2 E + 03
4.3 E + 02
5.1 E + 02
1.9 E + 02
1.8 E + 02
1.6 E + 03
2.4 E + 03
Total Weighted Discharge severity
(TWOS)
Health-Based
5.9 E + 02
1.0 E + 02
3.7 E + 02
7.0 E + 01
4.4 E + 02
3.3 E + 02
4.0 E + 01
3.6 E + 01
Ecology-Based
9.6 E + 05
6.3 E + 04
1.1 E + 03
5.2 E + 02
1.2 E + 04
1.2 E + 04
1.8 E + 03
2.4 E + 03
* Stream numbers correspond to those shown in Figure 1.

These streams are included as background values for comparison to the other streams.
-------
Discharge Severity values >1 were also obtained for Cd, Cu, Mn, Ni, Pb, S,
Zn and phthalate esters in the input water streams and Cd, Mn, Ni, and S
in the settling pond discharge stream. The reduction in both IDS and TWOS
values for the effluent versus the input water appears to be due to a
decrease in the concentrations of the phthalate esters, P, Cu, Pb, an Zn.
These and other constituents as well appear to be lost to the settling
pond sludge.
For the other streams evaluated, their TDS values resulted primarily
from the following constituents:
• Compressor Condensates - phthalate esters, phenols, cresols,
Cd, Fe, Hg, P, S, Se, and Zn.
• Rectisol Unit Water - phthalate esters, phenols, aromatic
hydrocarbons, Cd, Cu, Fe, Hg, Ni, P, S, Se, and Zn.
The TWOS values for the compressor condensates and Rectisol unit samples
turn out to be relatively low because of the small flow rates for these
two streams, approximately 9 and 4 m3/ hr, respectively. Whereas the
flow rates for the input and effluent waters is over 200 m3/hr.
51
-------
-------
5. REFERENCES
1. IERL-RTP Procedures Manual: Level 1 Environmental Assessment
(Second Edition), EPA-600/7-78-201, October 1978.

2. Sampling and Analysis Procedures for Screening of Industrial
Effluents for Priority Pollutants, EPA-EMSL, Cincinnati, Ohio,
Revised April 1977.

3. EPA/IERL-RTP Procedures for Level 2 Sampling and Analysis of
Organic Materials, EPA-600/7-79-033, February 1979.

4. SAM/IA: A Rapid Screening Method for Environmental Assessment,
of Fossil Energy Process Effluents, EPA-600-7-78-015,
February, 1978.

5. Multimedia Environmental Goals for Environmental Assessment,
Volumes I—IV, EPA-600/7-7-136 and EPA-600/7-79-176, November 1977
and August 1979.

6. Environmental Assessment: Source Test and Evaluation Report--
Chapman Low-Btu Gasification, EPA-600/7-78-202 (NTIS - PB 289940),
October 1978.

7. Environmental Assessment: Source Test and Evaluation Report—
Wellman-Galusha (Glen Gery) Low Btu Gasification, EPA-600/7-79-185
(NTIS-PB 80-102551), August 1979.

8. Standard Methods for the Examination of Water and Wastewater,
Fourteenth Edition; APHA, AWWA, WPCF; Washington, DC.
53
-------
-------
APPENDIX A

KRUPP-KOPPERS REPORT
55
-------
KRUPP-KOPPERS
Environmental Assessment
of the Koppers-Totzek Process

for

Defense and Space Systems Group
of TRW Inc.,
Redondo Beach, California, USA
Essen, February 1980
Subcontract No.: J 01440 DE 9-M
Project No. :4640
Krupp-Koppers GmbH. Mottkestr.29, D-4300 Essen 1
56
-------
KRUPP-KOPPERS
REPORT
No. P 3/8o
PAGE OF
ISSUE
Environmental Assessment of the
Koppers-Totzek Process

Investigations Performed for TRW Inc. at the
Coal-Based Ammonia Plant of AECI Limited,
Modderfontein, South Africa
Client: Defense and Space Systems Group of TRW Inc.»
Redondo Beach, California, USA
TRW Subcontract No. J 01UUO DE9 - M
KK Project No. U5HO
Project Manager: Mr. Kress
Reported by Dr. B. Firnhaber
Contents:
13 pages
3 tables
5 figures
PEP.-. re-I n*ME Dr.rirnhabeATE Febr.l9o
Sch Kfupp-Koppcr* GmDH D-4300 ElMfl 1
57
-------
KRUPP-KOPPERS
REPORT
No. P 3/8o
PAGE OF
ISSUE
Summary

According to the Professional Services Agreement of
October 2, 1979 between TRW Inc. and Krupp-Koppers
GmbH, Krupp-Koppers carried out investigations, com-
prising measurements and analytical work on waste and
by-product streams^on the coal-based ammonia plant of
AECI Limited in Modderfontein, South Africa. The plant
incorporates a Koppers-Totzek coal gasification plant
and a Rectisol gas purification unit. The aim of the
investigations was the assessment of the environmental
impact of the KT process.

The investigations on the plant site were carried out
in the period of November 7 to 29, 1979. The plant ope-
rated during the measurements at almost 100 % design
capacity of about Io3 ooo mn /h dry raw synthesis gas.

The analyses of the waste streams document the low
environmental impact of the KT process. Alternate pro-
cessing feasibilities for further reduction of environ-
mental pollution are discussed.
Note
The data and information reported hereafter shall only
be used in accordance with the terms and conditions of
the Professional Services Agreement of October 2, 1979
between TRW Inc. and Krupp-Koppers GmbH and the Letter
Secrecy Agreement referred to in Clause 16 of aforemen-
tioned agreement.
PEP.: [NAME [DATE \
Krupp Koppors GmbH D-4300 Esttn 1
58
-------
KRUPP-KOPPERS
R E P 9
No. P 3/8o
PAGE
ISSUE
Contents

1. Introduction
2. The Koppers-Totzek Process
3. The Ammonia Plant of AECI in Modderfontein
•». Experimental Procedure
5. Results
6. Discussion of the Results

Table 1: Feed Coal Analysis
Table 2: Gas Analyses
Table 3: Water Analyses

Fig. 1: No. U Ammonia Plant - Process Scheme
Fig. 2: Coal Preparation
Fig. 3: Koppers-Totzek Gasification
Fig. U: Gas Treatment
Fig. 5: Wash Water System - Flow Scheme
Fig. 6: View of the Gasification Plant
Fig. 7: View of the Gas Treatment Unit
DEP.:
NAME
PATE.
59
-------
KRUPP-KOPPERS

REPORT
Mr. P 3/8o

PAGE 1
OF
ISSUE
1. Introduction

In 1977 the Environmental Engineering Division o.f TRW Inc.
signed an agreement with the United States Environmental
Protection Agency, EPA, (Contract No. 68-O2-263S) concerning
the environmental assessment of high BTU coal gasification.
It is the aim of this investigation program to quantify
effluent streams obtained in the operation of commercial-
scale coal gasification plants, to identify possible treat-
ment or control technology, and to assess the environmental
impact on future construction sites to be evaluated for
large-scale application of the coal gasification technology.

The Koppers-Totzek process is one of nine coal gasification
processes to be investigated in this program.

In Subcontract No. T ol 4Uo DE9-M signed on Oct. 2, 1979,
it was agreed between TRW Defense and Space Systems Group
of TRW Inc., Redondo Beach, California, USA, and Krupp-
Koppers GmbH, Essen, Germany, that Krupp-Koppers carried
out measurements in the coal gasification plant in Modder-
fontein, South Africa, which were to supply data for asses-
sing the environmental impact of a commercial-scale Koppers-
Totzek plant.
The investigations to be performed in the program have been
carried out by Krupp-Koppers personnel in the No. t "Ammonia
Plant of AECI Limited in Modderfontein, South Africa in the
period of November 7 to 29, 1979. An employee of TRW Inc.
was present in South Africa during the investigation period
to receive the agreed on coal and water samples and for the
necessary liaison between the partners.

After a description of the Koppers-Totzek technology in
general and the ammonia plant Modderfontein in particular,
DEP.: NAME: DATE
Krupp-Kopoart GmbH D-4300 EB
60
-------
KRUPP-KOPPERS

REPORT
No. P 3/8o

PAGE 2 OF
ISSUE
the investigations and their results are reported herein.
Finally, discussion of alternate processing steps and
their effect on the environment is added.

2. The Koppers-Totzek Process

In 1936/19U2 Friedrich Totzek and his coworkers at Heinrich
Koppers GmbH in Germany,now Krupp-Koppers GmbH,developed a
new coal gasification principle where pulverized coal is gasi-
fied in an entrained-bed reactor,using oxygen and steam as
gasification media. The target of the development was a pro-
cess with virtually no restrictions to coal properties, a
resulting synthesis gas with CO and H, as main components,
and practically no environmental pollution.

The principle of entrained bed gasification according to
the Koppers-Totzek process operates autothermally, i.e.
without supplying outside heat. The reactants, coal, oxygen,
and steam, enter the reactor in certain proportions via
opposite burners located at the heads of the gasifiers cones.

The coal dust has a particle size, that is predominantly
smaller than o.l mm. The permitted portion of larger particles
in case of bituminous coal amounts to about 10 *, in case of
lignite to 15 % to 20 %. In a preceding coal preparation unit
the moisture content depending on the type of coal is reduced
to approximately 1 to 2 % in the case of bituminous coal and
8 to 10 % in the case of lignite. Generally, oxygen has a
purity of about 9'; %. Coal and gasification agents enter
the gasifier in cocurrent flow. The coal is gasified within
about 1 second. The temperatures in the core of the flame
amounts to approximately 2000 °C. Under these conditions
the heterogenous reactions between carbon, oxygen and steam
occur which are characteristic of coal gasification.
DEP.
I NAME
(DATE
o1-! 0 430C £•*«" I
61
-------
KRUPP-KOPPERS

REPORT
No. P 3/8o

PAGE
3 OF
ISSUE
The raw gas leaves the gasifier with a composition which is
determined by the homogeneous water gas equilibrium. Reac-
tions as the formation of methane are of minor importance
because of the high reactor temperatures of 1400 to 1600 C.

The sulfur contained in the coal as organic and inorganic
compounds for the most part is converted to H2S and COS,
at a ratio of about 9/1, and appears as such in the raw
gas. Other reactions are possible which result in trace
•amounts of CS2 and SOj.

The nitrogen contained in the coal and in the oxygen used
as gasification medium will react under the prevailing con-
ditions to form traces of NH-, HCK and NO.

The raw gas of the XT, process does not contain higher mole-
cular weight organic compounds. They are gasified completely
under the prevailing thermal conditions.

The raw gas produced from different solid fuels is charac -
terized by 80 to 88 Vol. % of CO and HZ and a CO/Hj-ratio
of 2/1 to 2.5/1. Carbon gasification degrees exceeding 98 %
have been reached, depending on the typ of coal used in the
process. The XT raw gas has a calorific value between lo.8
MJ/m and 11,8 MJ/m and based on the heating value is
between blast furnace and coke oven gas.
The high temperatures prevailing in the gasification reactor
requires suitable refractory lining to protect the reactor
walls, since at these temperatures the coal ash is liquified.
The wall structure must be designed in a way that the liquid
slag does not attack the lining. The liquid slag running down
the gasifier walls is cooled in a water bath and granulated.

XT plants are built with gasifiers with 2 or U burner heads,
their capacities amount up to 25 ooo and So ooo m^ /h raw gas,
respectively. A further increase in the output is basically
possible .
DEP.. |NAME
Sef.
DATE |

62
-------
KRUPP-KOPPERS

R r P a R T
No. P 3/8o

PAGE U
OF
ISSUE
Indepently of the eventual utilisation of the produced raw
gas, the KT process includes the stages of coal preparation
and mechanical cleaning of the raw gas, the treatment of the
wash water used for gas cooling and cleaning, as well as
package units for the production of pure oxygen by air frac-
tionation.

The KT process operating under normal pressure has shown
its efficiency in numerous large-scale plants totalling
more than 50 gasifiers.

3. The Ammonia Plant of AECI in Modderfontein
In 1972 AECI Limited, South Africa, ordered a grass-roots
ammonia plant, based on the Koppers-Totzek coal gasification
process, to produce 1000 t/day ammonia. The plant was commis-
sioned in 197U.

The highest daily production achieved so far was 1060 metric
tons ammonia. The plant operated in 1978 with an on-stream
time of 81 % and is expected to reach the figure of 86 % -
quoted for typical gas and naphtha based ammonia plants -
in the following years.

The basic plant layout is described with the help of the
block diagram shown in Fig. l.x) A single stream air separation
plant supplies oxygen at 98 % purity to six two-headed Koppers-
Totzek gasifiers. Twin ring-and-ball type mills are used to
pulverise the sab-bituminous coal feed to a nominal size of
90 % less than 90/(um. The oxygen is pre-mixed with steam and
the mixture entrains coal dust from screw feeders into the
gasifiers. The gasifiers operate essentially at atmospheric
pressure and a gas outlet temperature of about 1600 C. A
major part of the coal ash is entrained in the gas leaving
the gasifiers, and is subsequently removed by scrubbing with
water and passing through electrostatic precipitators.
/ x) Ref.: A.D. Engelbrecht, L.J. Partridge (AECI Limited], Paper
presented at the Ammonia from Coal Symposium, May 8-lo,1979
Muscle Shoals, USA
PEP.: [NAME [DATE [
-a, Ktup»«app« GmbH. D-4300 Euwi 1
63
-------
KRUPP-KOPPERS

REPORT
No. P 3/80

PAGE 5
OF
ISSUE
The gas is compressed to 3o bar in twin-stream raw gas com-
pressors and desulphurised (to less than 1 ppm HjS and COS)
in a methanol scrubbing column at about -38°C. A final stage
of compression raises the gas pressure to 5o bar after which
it is subjected to a water-gas shift reaction in a converter
unit with a conventional promoted iron oxide catalyst. Steam
for the shift reaction is supplied from waste-heat-boilers
on the gasifiers. The carbon monoxide and steam are conver-
ted to carbon dioxide and hydrogen and the residual (dry
basis) CO content is about 3 Vol.%.

Carbon dioxide is removed (to less than lo ppm) from the gas
by absorption in methanol at about-S8°C. CO- is recovered
from the methanol in a stripping column and a proportion
thereof is used in urea manufacture in another plant. Sul-
phur compounds absorbed from the gas (H,S and COS) are remo-
ved from the circulating methanol stream in a stripping
column and produce a byproduct stream containing about 60 %
HjS and COS. The gas purification process using methanol
is termed the Rectisol process.

The final traces of CO. are removed by adsorption on mole-
cular sieves and the gas then passes to a column for scrub-
bing with liquid nitrogen at -190°C. The gas purification
process results in a synthesis gas of high purity, such that
no voluntary purge of the synthesis loop is required to avoid
buildup of inerts.

A conventional ammonia synthesis loop, operating at 220 bar,
is employed.

The synthesis gas compressor, refrigeration compressor and
nitrogen compressor are single-stream centrifugal units while
there are two each centrifugal air compressors and raw gas
compressors in parallel. All the major machines are driven
by steam turbines (except one motor-driven air compressor)
and the motive steam is supplied from two large spreader-
stoker fired boilers.
DEP.- [NAME I DATE
Krupp-Koppers GmbH D-4300 E
64
-------
KRUPP-KOPPERS

REPORT
No. P 3/80

PAGE 6
OF
ISSUE
The individual process stages identifying the plant effluent
streams are shown diagrammatically in Figs. 2-5.

In the coal preparation unit, Fig. 2, raw coal is milled
and simultaneously dried using flue gas from the steam
boilers. The coal dust is conveyed pneumatically to the feed
bunkers of the gasification unit using impure nitrogen.
Both, the flue gas used for coal drying and the coal conveying
gas returned to the coal preparation unit are dedusted in
an electrostatic precipitator before venting via a chimney.

The gasification unit, Fig. 3, comprises the gasifiers proper
as well as raw gas cooling and dedusting. Part of the coal
ash leaves the gasifier after quenching in water as granulated
slag which can be deposited or used as a road construction
material. Unconverted coal and a major part of the coal ash
is entrained with the raw gas which passes after partial
quench through a waste heat boiler to final cooling and
dedusting. Cooling and coarse dedusting is reached in the
cooling washer. The following disintegrator brings the dust
content of the raw gas down to a level which allows the use
of blowers for conveying the gas to the gas holder. Compres-
sor-grade dedusting is obtained in electrostatic precipita-
tors.

Fig. 3 also shows the recycle of the wash water via settling
tank (clarifier) and cooling tower. The water purge containing
the flyash Cslurry) is pumped to a settling pond.

The gas treatment for the production of ammonia synthesis
gas is shown in Fig. 4. The cooled and dedusted raw gas is
compressed to 3o bar, water-washed for HCN removal, desulfu-
rized, and compressed in a final stage to So bar. After CO
shift conversion, CO. is removed in the second Rectisol stage.
Final purification of the hydrogen and admixture of the
Etoichiometric amount of nitrogen is obtained in the liquid
nitrogen wash.
DEP.. (NAME DATE
Krupp-Kopp*r« GmbH D-4300 E«Mn I
65
-------
KRUPP-KOPPERS

REPORT
No. P 3/8o

PAGE 7 OF
ISSUE
The water streams, compressor condensate, water from the
HCN wash, and condensate from the Rectisol unit, not iden-
tified in Fig. M, are fed to the common wash water system
of the gasification plant. A water purge from the CO shift
conversion is used as quench water in gasification.

In the Rectisol unit, a concentrated H2S stream for further
processing and a pure CO. stream are obtained. Tail gases
from desulfurization and C02 wash are vented. Tail gas from
the liquid nitrogen wash is burned in the steam boilers.

Fig. 5, finally, shows a block diagram of the entire wash
water system. It is fed with approximately 55 m /h of coo-
ling water and 145 m /h of an AECI produced water, called
"PSE". An additional 70 m /h PSE-water is used for conveying
ash from the boiler houses to the settling pond. The only
effluent water stream is the run-off from the settling pond,
comprising approximately 230 m /h.
**. Experimental Procedure

To accomplish the agreed investigation program, four Krupp-
Koppers employees travelled to South Africa and carried out
the necessary data recording,sampling,and analytical work
in the time period between November 7 and 29, 1979. They
were actively supported by AECI laboratory and operating
personnel whose friendly cooperation is greatly appreciated.

The specified work agreement necessitated the compilation
of complete material balances over gasification and gas
purification units the results of which - due to the pro-
prietory nature - cannot be included in this report. For
those measurements and analyses which are not performed
during normal operation of the plant, additional installa-
tions and analytical equipment had to be provided.
DEP..
NAME.
DATE
Sch
Krupp-Kojwt Gmbrt D-4300 Emr I
66
-------
KRUPP-KOPPERS

R F P n R T
No. P 3/8o

PAGE g
OF
ISSUE
Although the entire work was performed on several days spread
over a two-week period, particular care has been taken that
all data compilation, sampling, and analyses were carried out
under identical plant performance conditions. Thus, the data
obtained on different days for different stages of the plant
can be correlated for complete material balance.

During the investigation period one gasifiers was out of
commission. Practically full design capacity was obtained
with the remaining five gasifiers. All sampling and data
collecting was done at production rates of between Io2 ooo
and loU ooo m /h dry raw gas.

The sampling of the process waters for TRW was carried out
on Nov. 12 and on Nov. 19, 1979, between 9oo and l«*oo hours.
For each specified water stream, 11 sample containers supp-
lied by TRW were filled and then handed over to Mr. John F.
Clausen, the TRW respresentative present in Johannesburg.
Parallel samples were analyzed by Krupp-Koppers personnel.

The following water samples were taken and handed over to TRW:
on Nov. 12, 1979

Fresh Water Input (PSE-water)
Condensate from Raw Gas Compressor
(stages 1 to 4)
Condensate from Rectisol unit (effluent
from methanol/water separation dilut?d
with cooling water
Effluent from the ash settling pond
(clear water run-off)
TRW-Designation

- PW -
- RU -
- C5 -
DEP.:
I NAME
| DATE
Seh
Kiupp-Kopp*r> GmbH D-43OO £•••« I
67
-------
KRUPP-KOPPERS

R r P n R T
No. P 3/8o

PAGE
9 OF
ISSUE
on Nov. 19, 1979 TRW-Designation

Fresh water input (cooling water) - PW -
- Condensate from raw gas compressor
(as above) - CU -
Condensate from Rectisol unit
(as above) - RU -
Effluent from the ash settling pond
(as above) - C5 -
Since two different fresh water inputs to the wash water
system were used but only 2 x U sets of sample containers
were supplied, on one day PSE water on the other cooling
water was sampled.

In addition, one average sample (2 kg) of pulverized feed
coal, taken at the exit of the coal dust bunker in coal
preparation, was supplied to TRW.
5. Results
The essential results specified in the agreement between
TRW and Krupp-Koppers are summarized in Tables 1 to 3.

Table 1 shows the analysis of a feed coal dust sample
determined by Krupp-Koppers. It also contains analyses of
the average raw coal used in gasification fo~ the weeks
ending Nov. iu and 21, 1979, respectively, as determined
by the AECI laboratory. In Table 2 the analyses of the
agreed-on gas streams are compiled. The raw gas sample
was taken after the raw gas blower in the common line for
all gasifier trains leading to the gas holder. Thus, an
average sample of the total raw gas production was obtained.

Methane and higher hydrocarbon content of the raw gas is
extremely low. A previous study by AECI resulted in appro-
ximately llo ppm v/v of CH4 and 20 to 2S ppm of C2 plus C3
DEP..
Sch
|NAME

(DATE [

Krupp-Koppers GmbH D-430O E»Mn t
68
-------
KRUPP-KOPPERS

REPORT
No. P 3/Bo

PAGE j
•* Of
ISSUE
hydrocarbons in the raw gas before CO shift conversion. Some
methanation occurs during CO shift. Mercaptans and aromatics
have never been found in KT raw gas.

The tail gases from the Rectisol unit consist primarily of
COj and nitrogen used for stripping.

The results of the water analyses are summarized in Table 3.
The samples were taken in parallel to those supplied to TRW.

Two different water streams are used as make-up water for the
wash water system, PSE water and cooling water. PSE water is
a mixture of Johannesburg purified-sewage water and different
process waters from the Modderfontein plant, including water
effluent from the settling pond of No. U ammonia plant. The
only effluent from the wash water system is the run-off of
the ash settling pond, sampled and analysed on two different
days. Also two analyses each are included of the raw gas
condensates obtained in the raw gas compressor and in the
Rectisol unit.

Tables 2 and 3 list besides the analyses the determined
flow rates and temperatures of the process streams.

6. Discussions of the Results
Composition and properties of the feed coal have a strong
effect on the gasification results, but also on the side
reactions which lead to formation of the trace by-products
contained in the raw gas. In addition, in the plant layout
many alternatives are feasible for processing the product
gas as well as the sidestreams. Therefore, general applica-
tion of the reported results are limited.

The product raw synthesis gas leaving the gasifier/waste
heat system is further cooled and intensely washed for
flyash removal in the cooling washer and the disintegrator
stages.
DEP.: NAME DATE.
Kruop Kocp.il GflDH D-4303 ilttn t
69
-------
i > . V •«
:?P KOPPERS

REPORT
No. P 3/8o

PAGE
11 OF
ISSUE
The composition of the water-washed raw gas is listed in
Table 2. The main components of the raw gas are CO, H,, CO,
and Nj. The sulfur contained in the coal feed is present
in the gas as H2S, COS and CS2 which are removed from the gas
in the following Rectisol unit. Trace components of the raw
gas are NH3, HCN , S02 and NO.

A big advantage of the Koppers-Totzek process with regard
to its effect on the environment lies in the fact that the
produced raw gas contains no coal distillation products
because of their spontaneous gasification at the extremely
high temperatures. Aromatics , phenols, and mercaptans have
never been detected in KT raw gas .

Waste gas streams in the Modderfontein plant are the tail
gases from the Rectisol unit, the tail gas of the liquid
nitrogen wash and the combined stream of flue gas used for
drying in the coal preparation unit and of the conveying
gas for coal dust. The tail gases of the Rectisol unit (Table 2)
consist primarily of C02 and the nitrogen used for stripping.
The tail gas of the liquid nitrogen wash is burned in the
boiler station. If flue gas cannot be used for drying the
coal because of environmental reasons it can be replaced
by hot gas produced by burning the tail gas from the liquid
nitrogen wash and/or desulfurized raw gas. In this case the
flue gas can be vented through a relatively low stack of 25 m
after the dust content has been lowered to less than loo
In the raw gas washing stages NH3 , HCN , S02 and to a small
degree HjS and C02 are dissolved in the. water. At the pH of
the wash water, its temperature level and especially because
of its flyash content, there is a rapid conversion of H_S
2-2-
to Sj03 and SO^ . The HCN reacts with the sulfur compounds
to form SCN~ and with the iron content of the flyash
to form insoluble complexes.
PEP.: 1 NAME (DATE:
Krupp Kopp«rl GmbH D-430& E
70
-------
KRUPP-KOPPERS

REPORT
No. P 3/8o

PAGE
ISSUE
1? OF

All water streams, wash waters and condensates, which have been
in contact with the raw gas are fed to the common wash water
system (Fig. 5). The water streams are conducted through covered
gutters to the clarifier where the solids are removed. Clarified
water is recycled via a cooling tower to the main wash stages.
The flyash leaves the clarifier as a slurry and is pumped to
settling ponds for deposit.

The make-up water to the wash water recycle consists of PSE
water fed to the clarifier and cooling water which is used
in the HCN wash and for dilution of the Rectisol condensate.
Both make-up waters have a relatively high salt content. The
PSE water contains also a certain amount of NH,. A water
purge stream from the CO shift conversion unit is used as quench
water for partial quenching of the hot raw gas exiting the
gasifier.

The only water effluent leaving the No. 4 ammonia plant is
the run-off of the ash settling pond. As the analyses in
Table 3 show, is its composition very similar to that of the
make-up water. Its content of toxic substances is very low
indeed. This is due to the long residence time in contact
with the flyash in the settling pond. In one commercial KT-
coal gasification plant this waste water has for years been
used to water the fields and as drinking water for animals.

If there is not sufficient space for settling ponds the fly-
ash slurry can be filtered and the purge water stream which may
be significantly smaller than that used in Modderfontein can
be cleaned by an oxidative chemical treatment. I* is possible
to eliminate in a relatively simple way the traces of toxic
material virtually completely by oxidation so that the im-
positions by the authorities, for instance, those applicable
in the German Federal Republic, can be met. Should regulations
reduce the limits for the ammonia content and the COD value
in waste water, a biological treatment after chemical oxidation
can ensure further decomposition of detrimental ingredients.
DEP NAME DATE
71
-------
KRUPP-KOPPERS

REPORT
No. P 3/8o

PAGE 13 OF
ISSUE
While meeting stricter regulations concerning pollution is
relatively easy for KT gasification, one must take into account
that some measures necessary for extreme environmental protec-
tion show up in increased investment costs.
DEP,:
NAME
DATE
Krupf,-Kopp«r» GmDH. D-4300 £»««r\
72
-------
KRUPP-KOPPERS T.M. i
Feed Coal Analysi
Sample
Date
Determination by
Particle Size wt %
>100 um
> 61
> 29
> 19
> 11
> 6,6
< 6,6
Moisture wt %
Elemental
Composition wt %(mf)
H
C
Scombustible
N
Ash
0
Ash Composition wt %
Fe203
Si02
A1203
CaO
MgO
Na20
K20
Ti02
P2°5
so3
Total Sulfur wt % (mf)
Coal Dust
23-11-79
KK

26,9
15,1
20,5
SEITE VON
S AUSGABE
Raw Coal
11-11-79
AECI

7,2
9,3
«,1 !
16,9 '•
1,5

3,7
65,1
3,7

3,5
62,6
0,5 i
1,6 1,7
20,7 19,1
8,"

5,0
**••
( 27,8

9,6 !
21-11-79
AECI

3,6

3,5
69,7

1,6
18,3
1

1
2,2 ;
0,2 ; ;
0,6
!.«•
1,3 I
6,6
1.1

i
1.2
1.0
AST.: FE-I iNAMEDr.Firnhabe^DATUM 12.2.8^
^ rimiiiTTupiiin Omt^l D-UOO f«Mn l
73
-------
KRUPP-KOPPERS T.M. 7
Gas Analyses

Sample
Date
Flow Rate Nm3/h
Temperature °C
Composition
H20 g/Nm3
H2 Vol.% (dry)
CO
co2 "
N2/Ar "
CHU
H2S mg/Nm3(dry)
COS
cs2 "
so2
NH3 "
HCN "
NOX ppm v/v
Mercaptans
Methanol Vol.% (dry)
Raw Gas after
Raw Gas Blower
23-11-79
103 600
46

54
28.2
59.1
lo.9
1.8
^o.l ^
6333
74o
450
14
57
76
15
nil
nil
PAGE OF
ISSUE
Tail Gas from
H2S absorber
16-11-79
13 700
27

5
nil
1.9
52.6
45.5
^0.1
2)
2)
2)
nil
39
62
nil
nil
3)
Tail Gas from
C02 absorber
16-11-79
48 800
29

5
nil
0.3
84.3
15.4
•*. o.l
nil
nil
nil
nil
3
8
nil
nil
4)
Methane and higher hydrocarbons :
In January through March, 1978, a study was made by AECI on the
hydrocarbon content of the synthesis gas after desulfurization
and after CO shift and C02 wash. The average results of six ana-
lyses each were in ppm v/v.
Gas ex H_S Gas ex C02
Absorber Adsorber
CHU l^o 382
CjH^ 1.4 0.6
C2H6 15 13
C3H8 5 nil
C2H2 nil nil
2) Not determined; design number is •< 2 ppm total sulfur
3) Not determined; design number is O.12 Vol.%
4) Not determined; design number is o.oS Vol.%
DEP.: [NAME (DATE: |

74
-------
en
r

X
i
*
S
?
t
o
f
IT
i
»-*
2
S
?
H
3
0
1
I
i

•i
r*
*-

Sample
Date
flow Rate
Temperature
PH
total suspended *olld*
total dixolved lolidl
Hardness 1
Alkalinity, p-Value|
•-Value'
Conductivity
COD

HH3
CN"
SCN"
H2S
s20j"
so3~
S0l"
•V"
Cl"
Hethanol

•3/h
°c

•g/i
mg/l
mg/1
a*
CaCOj
fmhol/cm
•g/1

•g/1
n
"
n
-
»
M
n
*
mg/1

•g/1
Treih Wash
Water Input
(PSE)
12-11-79
215
23
6.1
nil
1580
1S2
o
26
23oo
31

73
0.2
2.1
nil
nil
nil
S.I
lo
US

Cooling
Water Input
19-11-79
SI
3o
I.S
1
116o
621
o
167
19oo
111

2.1
1.2
2.1
trace*
nil
nil
• 53
2.1
172

Efflue
Settli
12-11-79
23o
23
l.o
nil
ISIo
12o
o
126
21oo
3S3

31
0.2
1.3
nil
nil
3.6
7S?
S
IIS

it from
tl Pond
19-11-79
23o
23
9.1
nil
153o
661
11
79
21oo
13

21
0.2
2.2
nil
nil
nil
7 06
0.1
163

Condeniat
Raw Ga* C
12-11-79
9.3
32
1.2
nil
26o
60
0
2990
6000
611

973
7.2
lo.»
13.9
1.1
nil
56
2
23

> from
ompreaior
19-11-79
9.0
31
l.o
12
17o
16
0
169o
55oo
S69

9oo
lo.S
17.1
S3.S
7.1
nil
09
3
13

Diluted C
from Rect
12-11-79
1.1
So
9.1
To
139o
691
o
78
tloo
2t

26
2.1
llo
1.1
11. S
nil
161
0.1
153

~. 0.1
ondeniat*
i.ol Unit
19-11-79
3.6
SI
1.1
20
1610
SSI
o
111
2ooo
looo

19
ND
137
I.S
16.1
nil
SU1
2.1
151

•C.0.1

Note; The term "nil" indicate* • concentration below the limit of detection which ii for the
applied analytical method* below 1 mg/1
* Dissolved oxygen was measured in January, 1980 while the plant was at full production rates by AECI
personnel using a T.O.A. dissolved oxygen meter. The results obtained were:
Fresh Wash Water Input - 3.0 mg/L
Effluent from Settling Pond - 6.8 mg/L
Diluted Condensate from Rectisol Unit - 0.4 mg/L

C
I
B

IB
m

|
^
in
70
\/\

S
rt
1
r
*
•

H
»;
*
i*

-------
KRUPP-KOPPERS
No. U Ammonia Plant
Process Scheme
PEP.: FE-I [NAME: Dr.Firnhabft»TE:12 .2.80
76
-------
Coal Conveying Gas

Waste Gas
to Chimney

E-Precipitator]

c
Flue Gas

from Boilers
KRUPP-KOPPERS
Coal Preparation
Fig.2
77
-------
Waste Heat
Boiler
1 Cooling Washer
2 Disintegrator
3 Gas Blower
4 E-Precipitator
i i
Fly Ash Slurry
KRUPP-KOPPERS
KOPPERS-TOTZEK Gasification
Fig.3
78

-------
Steam
Nitrogen
NH3 Synthesis ^
Gas
Tail Gas ^
>-Fraction
..Nitrogen
Cold Hethanol Cycle
2 CO-Conversion
3 CC^-Removal
4 Liquid N2-Wash
5 H2S-Stripper
6 C02-Stripper
KRUPP-KOPPERS
Gas Treatment
79
-------
Seft Krupp-Kopp*r» GmbH D-43OO E»**n 1
80
'

Water (PSE
Evaporation 1
rL^ i
) Input
"45 m3/h
r ^— ,
54 m3/h (56 m3/hl . . (162 m3/hL
^ Cooling Tower **\" J Clanfier V /^
t
Slap — f-
^ cooling

^ Cooling ta.
Washer

Disinte- 0-
grator

^ Electrost.. s-
'recipitator
Evaporation Water (PSE) Input
(2 m3/hi .I
Settling Pond
s~ ^\
73 m3/h)
9 m3/h

4 m3/h
u^ — —
Wash Water System - Flow Schema

Gas
Holder

Compressor

HCN
Wash

Rectisol
Wash

Water
Effluent
23o m3/h
C.U.
C.W.
5o m3/h
^ C.W.

"1
en
o
n
3
55 J
c "
S ""
KRUPP-KOPPERS
E
u
n
r
E
D
O
n
D
"1
h1-
DO
01
-------
Fig. 6: Gasification plant with 6 Koppers-Totzek
gasifiers at the Nitrogen Fertilizer Works
Modderfontein, South Africa.
Capacity: l.ooo tons/day ammonia
81
-------
Fig. 7: Gas treatment unit and compressor house at the
Nitrogen Fertilizer Works Modderfontein, South Africa,
82
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
REPORT NO.
EPA-600/7-81-009
2.
3. RECIPIENT'S ACCESSION'NO.
TITLE AND SUBTITLE
Environmental Assessment: Source Test and
Evaluation Report, Koppers-Totzek Process
S. REPORT DATE
January 1981
6. PERFORMING ORGANIZATION CODE
. AUTHOR(S)
8. PERFORMING ORGANIZATION REPORT NO.
C.A. Zee, J.F.Clausen, and K.W.Crawford
i. PERFORMING ORGANIZATION NAME AND ADDRESS
TRW, Inc.
One Space Park
Redondo Beach, California 90278
10. PROGRAM ELEMENT NO.
INE825
11. CONTRACT/GRANT NO.

68-02-2635
12. SPONSORING AGENCY NAME AND ADDRESS
EPA, Office of Research and Development
Industrial Environmental Research Laboratory
Research Triangle Park, NC 27711
13. TYPE OF REPORT AND PERIOD COVERED
Final; 8/79-12/80
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES T£RL-RTP project officer is William J. Rhodes, Mail Drop 61,
919/541-2853. EPA-650/2-74-009a is an earlier report relating to this process.
16. ABSTRACT
The report gives results of a source test program at a Koppers-Totzek
(K-T)^ coal gasification plant operated by AECI, Ltd. at Modderfontein, Republic of
Soutfi"Africa. EPA is interested in the K-T process because process economics and
demonstrated commercial reliability make it a viable prospect for U.S. applications.
Responsibilities for sampling, analysis, and engineering descriptions of the plant
were shared by TRW and Krupp-Koppers GmbH of Essen, Federal Republic of Ger-
many. EPA's phased approach for environmental assessments was followed. Level
1 and Level 2 data were collected along with priority pollutant screening data. Much
of the effort was focused on wastewater streams. Wastewater treatment, consisting
of a clarifier and settling pond, was adequate to produce a final discharge that had
lower pollutant levels than the fresh input waters supplied to the plant. The report
contains complete data and describes the K-T process and the Modderfontein plant.
The Source Test Evaluation (STE), intended as an initial effort, was somewhat
limited in scope. Recommendations for future STE programs are provided.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.IDENTIFIERS/OPEN ENDED TERMS
c. COS AT I Field/Group
Pollution
Coal Gasification
Assessments
Waste Water
Water Treatment
Pollution Control
Stationary Sources
Koppers-Totzek Process
Source Testing
13B
13H
14B
18. DISTRIBUTION STATEMENT
Release to Public
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
89
20 SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
83
-------