EPA-650/2-73-004
August 1973
ENVIRONMENTAL PROTECTION TECHNOLOGY SERIES
FATE OF TRACE
-------
EPA-650/2-73-004
FATE OF TRACE CONSTITUENTS
OF
COAL DURING GASIFICATION
by
A. Attari
Institute of Gas Technology
IIT Center, 3424 South State Street
Chicago, Illinois 60616
Contract No. 68-02-0277
Program Element No. 1A2013
EPA Project Officer: William J. Rhodes
Control Systems Laboratory
National Environmental Research Center
Research Triangle Park, North Carolina 27711
Prepared for
OFFICE OF RESEARCH AND DEVELOPMENT
U.S. ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON, D.C. 20460
August 1973
-------
This report has been reviewed by the 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 recommendation for use.
11
-------
5/73 8937
ABSTRACT
This study was initiated by the Environmental Protection Agency to in-
vestigate the fate of trace elements of coal during the coal gasification
process for the production of pipeline-quality gas. The ultimate goal of the
investigation is to provide the necessary information that will enable environ-
mentally sound operation of future commercial-scale coal gasification plants.
The present report is the result of a 6-month preliminary effort to set
up laboratory analytical methods and to analyze for antimony, arsenic,
beryllium, cadmium', chromium, lead, mercury, nickel, selenium, tellurium,
and vanadium in a bituminous raw coal and its solid residues from a bench-
scale coal gasification unit at the Institute of Gas Technology. The samples
selected are representative of the solid effluents from the pretreatment,
hydrogasification, and electrothermal stages of the IGT HYGAS pilot plant.
The data obtained thus far indicate substantial removal of arsenic,
cadmium, lead, mercury, selenium, and tellurium from coal during the gasi-
fication process. However, since these results are based on a limited number
of samples, further analysis of a larger number of samples is required before
any firm conclusions can be drawn.
This report is submitted by the Institute of Gas Technology in fulfillment
of Contract Number 68-02-0277 under the partial sponsorship of the Environ-
mental Protection Agency. Work was completed as of November 1972.
111
INSTITUTE OF GAS TECHNOLOGY
-------
5/73 8937
TABLE OF CONTENTS
Page
INTRODUCTION 1
CHAPTER I. EXPERIMENTAL 3
A. IDENTIFICATION OF INSTRUMENTATION 3
B. SAMPLE SELECTION 3
C. PROXIMATE AND ULTIMATE ANALYSES 3
D. ASHING OF COAL 3
E. ACID DISSOLUTION OF COAL ASH 4
1. General Considerations 4
2. Acid Dissolution 4
• F. SAMPLE PREPARATION FOR MERCURY 5
G. DETERMINATION OF INDIVIDUAL ELEMENTS 6
1. Chromium 6
2. Nickel 6
3. Vanadium 6
4. Beryllium 7
5. Lead 8
6. Cadmium 9
7. Arsenic 9
8. Antimony 10
9. Selenium 10
10. Tellurium 10
11. Mercury 11
CHAPTER II. RESULTS AND CONCLUSIONS 12
A. PROXIMATE AND ULTIMATE ANALYSES 12
B. ASHING OF COAL 12
C. DETERMINATION OF INDIVIDUAL ELEMENTS 21
1. Chromium 21
2. Nickel 21
3. Vanadium 21
4. Beryllium 22
5. Lead 22
IV
INSTITUTE OF GAS TECHNOLOGY
-------
5/73 8937
TABLE OF CONTENTS (Continued)
Page
6. Cadmium 23
7. Arsenic, Antimony, Selenium, and Tellurium 23
8. Mercury 23
CHAPTER III. DISCUSSION 25
A. THE CHANGE IN CONCENTRATION OF TRACE
ELEMENTS AT VARIOUS STAGES OF GASIFICATION 25
B. ANALYTICAL METHODS 25
1. Flame Atomic Absorption 25
2. Microsampling Assembly 25
3. Graphite Furnace 28
4. Mercury Analysis System 28
REFERENCES CITED 30
INSTITUTE OF GAS TECHNOLOGY
-------
5/73 8937
LIST OF FIGURES
Figure No. Page
1 IGT HYGAS Process for Electrothermal
Gasification, Showing Pretreatment,
Hydrogasification, and Electrothermal Stages 2
2 LTA Ashing of -80 Mesh Sample of F-FP-137 15
3 LTA Ashing of -80 Mesh Sample of F-HTEG-4 16
4 LTA Ashing of-80 Mesh Sample of F-EG-10 17
5 LTA Ashing of -80 Mesh Sample of R-EG-10 18
6 LTA Ashing of -80 Mesh Sample of F-EG-11 19
7 LTA Ashing of -80 Mesh Sample of R-EG-11 20
8 Concentration of Trace Elements Tested in Pittsburgh
No. 8 Bituminous Coal (F-FP-147) and Coal
Residues at Various Stages of Gasification 26
VI
INSTITUTE OF GAS TECHNOLOGY
-------
5/73 893?
LIST OF TABLES
Table No. Page
1 Graphite Furnace Conditions for Beryllium 8
2 Graphite Furnace Conditions for Lead 9
3 Graphite Furnace Conditions for Cadmium 9
4 Graphite Furnace Conditions for Arsenic 9
5 Graphite Furnace Conditions for Antimony 10
6 Graphite Furnace Conditions for Selenium 10
7 Graphite Furnace Conditions for Tellurium 11
8 Proximate and Ultimate Analyses 13
9 Sample Temperature Versus Power Input 14
10 Trace Element Concentration (in ppm) of Pittsburgh
No. 8 Bituminous Coal at Various Stages of
Gasification, Calculated on the Raw Coal
(F-FP-147) Basis 21
11 Comparison Results of Beryllium by Direct
Aspiration Flame and by Graphite AAS 22
12 Comparison Results of Lead by APCD-MIBK
Extraction, Micro sampling Boat and Graphite
Furnace 23
VII
INSTITUTE OF GAS TECHNOLOGY
-------
8937
5/73
INTRODUCTION
With the increasing demand for energy and the growing shortage of
petroleum oil and natural gas in the United States, a great impetus has been
given to the study of conversion of our large coal deposits to pipeline-quality
gas via coal gasification processes. The Institute of Gas Technology estimates
that by 1990 we will witness the construction and operation of some 40 large-
scale coal gasification plants, each capable of processing approximately 6
million tons of coal annually, with an estimated total yearly production of
2. 7 trillion cubic feet of gas. This projected large-scale consumption of
coal necessitates a thorough investigation of its impact on the environment,
since trace quantities of toxic elements such as antimony, beryllium,
mercury, and selenium are known to exist in some U. S. coal deposits. A
knowledge of the level of these potentially hazardous pollutants and their fate
during the. coal gasification process is necessary for such an investigation.
The Institute of Gas Technology (IGT) has been engaged in gasification
research since 1942. In 1964 it was awarded the HYGAS Process Research
contract under the joint sponsorship of the U. S. Office of Coal Research
(OCR) and American Gas Association (A. G. A.). The project consisted of
the design, construction, and operation of a large pilot facility with a daily
capacity of 75 tons of coal and the production of 1. 5 million cubic feet of
pipeline-quality gas. This plant was completed in May 1971 at a cost of
$10 million and has been intermittently operational since October 1971.
In May of 1972 IGT was awarded a 6-month research contract by the
Environmental Protection Agency (EPA Contract No. 68-02-0277) to set up
and measure the concentration of 11 trace elements in a Pittsburgh No. 8
coal sample and its 5 solid residues, representing the coal input and output
of each of the three stages of the HYGAS pilot plant, as shown in Figure 1.
Since the HYGAS plant was not fully operational at that time, it was
decided to draw upon IGT1 s large collection of coal and char samples
accumulated over several years of gasification research during bench-
scale development of the HYGAS plant.
The initial 2-month period was devoted to study of the effect of ashing
temperature on the recovery of some of the more volatile trace elements.
The low-temperature plasma ashing machine was studied intensively for
INSTITUTE OF GAS TECHNOLOGY
-------
5/73
8937
PIPELINE GAS
68 aim
METHANATION
rawhydrooen-ricngas
residual char
rto power generation
GAS PURIFICATION
farbon dioxide, liquid aromatics,
sulfur, ammonia
i>«ELECTROTHERMAL GASIFIER
spent char to
power generation
1000 °c
74atm
Figure 1. IGT HYGAS PROCESS FOR ELECTROTHERMAL
GASIFICATION, SHOWING PRETREATMENT,
HYDROGASIFICATION, AND ELECTROTHERMAL STAGES
its application to the ashing of coal samples because some more volatile
elements were lost during the high-temperature ashing process — even at
550°C.
The following 4-month period consisted of the laboratory setup and
analysis of beryllium, chromium, lead, nickel, vanadium, antimony, arsenic,
cadmium, mercury, selenium, and tellurium in the No. 8 coal sample and
residues. An effort was made to use the atomic absorption spectrophoto-
meter for these analyses. The microsampling kit, the graphite furnace, and
the mercury analyzer assemblies on the conventional atomic absorption
spectrophotometer (AAS) made it possible to determine these 11 elements
with a precision comparable to other analytical methods.
2
INSTITUTE OF GAS TECHNOLOGY
-------
5/73 8937
CHAPTER I. EXPERIMENTAL
A. IDENTIFICATION OF INSTRUMENTATION
The International Plasma low-temperature asher (LTA) IPC Model
1005B was used throughout this work for ashing of coal.
The Perkin-Elmer Model 303 double-beam atomic absorption spectro-
photometer with a background compensator installed internally, a nebulizer
burner assembly, and a Sargeant TR 10-inch strip-chart recorder was used
in most of the work. The Perkin-Elmer microsampling assembly was used
for part of the work, and the Perkin-Elmer mercury analysis system 303-
0832 was used for mercury determination.
B. SAMPLE SELECTION
The six coal samples selected for this work were taken from the following
series of IGT small-scale gasification runs.
1. Pittsburgh No. 8 Seam bituminous coal used as feed in the
pretreatment Run No. FP-147 plus the partially gasified pre-
treated coal residues of this run designated as F-FP-147 and
R-FP-147, respectively.
2. Pretreated coal from Run No. FP-147, which was used as feed
in the hydrogasification Run No. OH-9, together with the residues
of this run, which are designated as F-OH-9 and R-OH-9,
respectively.
3. Hydrogasified char from Run No. OH-9, which was used as feed
in the electrothermal gasification Run No. EGO-37, and the spent
chars from this run, which are, respectively, designated as
F-EGO-37 and R-EGO-37.
C. PROXIMATE AND ULTIMATE ANALYSES
Proximate and ultimate analyses were performed on these six samples
in order to present some qualitative properties which would aid in a more
intelligent choice of samples for further analyses.
D. ASHING OF COAL
In low-temperature ashing of coal, the raw coal was utilized to show
the relationship between the radio frequency (RF) power input and the
sample temperature. The ashing rate was studied as a function of coal
particle size using duplicates of the six samples. The coal samples were
INSTITUTE OF GAS TECHNOLOGY
-------
5/73
8937
ashed for 30 hours in the low-temperature asher, using raking and mixing
to achieve a new ashing surface every 2 hours.
In high-temperature ashing, the coal samples were ashed overnight in
a muffle furnace, keeping the temperature under 550 °C.
To measure loss of volatile elements in high-temperature ashing, six
samples were ground to 80-mesh size, and each was taken in 5-gram quad-
ruplicates for ashing. The first pair of samples were ashed at 120°C (in the
LTA) and the second pair were ashed at 550°C in a muffle furnace. The ashed
materials were then reacted with mineral acids, and the resulting solution was
analyzed by atomic absorption spectroscopy.
E. ACID DISSOLUTION OF COAL ASH
1. General Considerations
The sample preparations were different for different elements. The
widely used perchloric-hydrofluoric acid dissolution1'2'3 for coal ash is suitable,
with minor modifications, for the determination of beryllium, cadmium,
chromium, nickel, lead, and vanadium. With low-temperature ashing,the final
products still contain about 1$ carbon. It is, therefore, necessary to treat
the sample with a small amount of concentrated nitric acid before adding
hydrofluoric acid. Arsenic, antimony, selenium, and tellurium might be
volatilized under perchloric-hydrofluoric acid dissolution. Therefore,
sulfuric acid and nitric acid treatments were used for these elements. 4
For mercury determination, even the low-temperature ashing process
results in loss of mercury. 5'6 It is easier to combust the coal sample and
collect the mercury from the vapor phase7"10 either by gold amalgamation or
by absorbent solution than it is to prevent loss of mercury during ashing.
2. Acid Dissolution
Methods for acid dissolution of coal ash differ according to the trace
elements being investigated.
a. For determination of beryllium, cadmium, nickel, lead, and vanadium,
0. 500 gram of coal ash is placed in a platinum crucible with 5 ml con-
centrated nitric acid and heated on a hot plate until the brown fumes
disappear. The residue is cooled, 3 ml of concentrated perchloric
acid added, and heated again until white fumes appear. When the
INSTITUTE OF GAS TECHNOLOGY
-------
5/73 8937
mixture has cooled, 10 ml 49^ hydrofluoric acid is added,
and heat applied until white fumes appear. After cooling and the
addition of 10 ml deionized water, the solution is heated to a boil,
cooled, and filtered through Whatman No. 42 paper into a 50-ml
volumetric flask. The paper and residue are washed thoroughly, then
50 mg of potassium (as potassium chloride) is added before making up
the solution to volume.
b. For determination of arsenic, antimony, selenium, and tellurium.,
0. 500 gram of coal ash is placed in a beaker with 4 ml of concentrated
sulfuric acid and 10 ml of concentrated nitric acid and heated on a hot
plate until white fumes appear. After the residue has cooled, 5 ml of
deionized water is added and the mixture heated until white fumes
again appear. After cooling and addition of 30 ml of deionized water,
the solution is heated to a boil, cooled, and filtered through Whatman
No. 42 paper into a 50-ml volumetric flask. The paper and residue
are washed thoroughly and the solution made to volume.
F. SAMPLE PREPARATION FOR MERCURY ANALYSIS
This procedure consists of the combustion of the coal sample in a stream
of oxygen and the recovery of mercury vapors from the combustion products
of the sample by the use of a suitable absorbing solution for the subsequent
analysis of the resulting solution by a nonflame atomic absorption method.
The combustion apparatus is assembled according to the specifications of
ASTM D271 (Ultimate Analysis of Coal). The equipment consists of a 97-cm-
long Vycor tube with a 1. 9-cm inside diameter having one reduced end for rub-
ber tubing connection. The Vycor tube is packed with a 20-cm section of silver
wire followed by a 30-cm section of copper oxide. Copper gauze plugs are
conveniently used to separate the packings and are also placed at each end of
the packings to prevent their displacement.
Three separate tube furnaces are used to heat the different sections of the
combustion tube at different temperatures. The silver wire section is covered
by a 23-cm-long furnace maintained at 450°-500°C, while the copper oxide sec-
tion is heated by a 33-cm furnace kept at 800°-850 °C. Finally, the sample
section is heated with a 13-cm movable furnace which can rapidly attain a tem-
perature of 900°C during the run.
Two 125-ml gas-washing bottles with fritted cylinders are connected in
series to the exit end of the combustion tube. Each bottle is charged with 50 ml
5
INSTITUTE OF GAS TECHNOLOGY
-------
5/73 8937
of an absorbing solution containing 1% potassium permanganate in 0. IN sulfuric
acid solution.
o
To make a run, a 0. 5-gram sample of coal IB weighed in a porcelain boat
and placed into the combustion tube. The oxygen flow rate is set at 75 ml/min,
and the sample furnace is fully turned on and allowed to cover the boat in 20
minutes. After an additional 20 minutes combustion time, the sample furnace
is turned off and pulled back to its original position. At the end of this period
the absorbing solutions are quantitatively transferred to a 200-ml volumetric
flask and the volume brought up to mark; the solutions are saved for further
analysis.
G. DETERMINATION OF INDIVIDUAL ELEMENTS
1. Chromium
The chromium line 357.8 nm was used in this work. The determination
of chromium was done by direct aspiration of the sample solution into an
air-acetylene flame. The sensitivity of the chromium in the flame is highly
dependent on the richness of the fuel. A rather fuel-rich flame was used
because it gave best sensitivity.
A series of standard solutions containing 0-5 ppm of chromium was
prepared. The concentrations of chromium in the sample solutions were
calculated from the standard curve.
2. Nickel
The nickel line 232.0 nm was used in this work. The determination of
nickel by direct aspiration of the sample solution into the air-acetylene flame
was reasonably good with scale expansion.
When 10X scale expansion is used, it is necessary to set the slit width
to 4 (1 mm). The sensitivity would be reduced to about half if the slit width
were set on 3 (0. 3 mm). The width of 4 will, stabilize the background better.
A series of standard solutions was prepared containing 0-5 ppm of nickel.
3. Vanadium
Vanadium line 318. 4 nm was used in this work, with a nitrous oxide-
acetylene flame. Direct aspiration of the sample solution into the flame
required 10X scale expansion.
6
INSTITUTE OF GAS TECHNOLOGY
-------
5/73 8937
An attempt was made to extract vanadium into the organic solvent by
use of ammonium pyrrolidine carbodithioate (APCD) in order to give a larger
signal size. The extraction of vanadium by chelating it with APCD followed
by methyl isobutyl ketone (MIBK) extraction has been found quite satisfactory
with the vanadium standard solutions. However, it did not work for the
sample solutions due to interference from the large amounts of the iron
existing in the sample. Separation of iron by ammonium thiocyanade
extraction or hydroxide-ion precipitation would remove vanadium at the
same time.
Therefore, the vanadium was determined by direct aspiration of the
sample solution into the nitrous oxide-acetylene flame. The sensitivity of the
measurement is strongly dependent on flame condition. The highest sensitivity
is attained when the red-feather zone is about 1 cm high. The relative standard
deviation was about 9$ at the level of 2 ppm vanadium in the sample solution
with scale expansion at 10X.
4. Beryllium
Beryllium line 234. 9 nm was used. In the first half of the work beryllium
was determined by direct aspiration of the sample solution into the nitrous
oxide flame, and in the second half of the work it was determined by graphite
furnace.
The determination of beryllium by a nitrous oxide-acetylene flame is not
very good at its concentration in this coal. A rather fuel-rich flame is re-
quired to get the highest sensitivity. One percent oxine was added to the
sample solution to offset the interference caused by the high aluminum in
the sample. The 10X scale expansion was used during the measurement.
In the second half of the work, the beryllium was again determined by
the graphite furnace technique. The sample solution was diluted tenfold and
a series of standard solutions containing 0 to 0.005 ppm of beryllium was
prepared. Operating conditions of the graphite furnace are given in
Table 1.
7
INSTITUTE OF GAS TECHNOLOGY
-------
5/73 8937
Table 1. GRAPHITE FURNACE CONDITIONS
FOR BERYLLIUM
Time, s Temperature, °C
Drying 50 150
Ashing 50 400
Atomizing 4 2600
5. Lead
Lead line 283. 3 nm was used. Lead was first determined by APCD-
MIBK extraction followed by air-acetylene flame atomic absorption. The
iron was separated prior to APCD-MIBK extraction by adding ammonium
thiocyanade and shaking with isoamyl alcohol or tributyl phosphate (TBP).
In APCD-MIBK extraction of lead, the pH of the sample solution was
adjusted to 3. 5 ± 0.1 before adding 1 ml of 4<* APCD solution. APCD was
freshly synthesized in this laboratory because it decomposes gradually
upon exposure to light. Sample solution was transferred into a 50-ml
volumetric flask, 5 ml of MIBK was added, and the mixture was shaken
vigorously for 1 min. The MIBK layer was lifted to the neck of the flask
by addition of deionized water. A series of standard solutions was prepared
in the same way. Atomic absorption is measured by aspirating MIBK solution
into the air-acetylene flame.
A sampling boat was also used for lead determination. Up to 1 ml of
sample solution was introduced into the boat and dried by being placed close
to the flame. The boat and the residue were pushed into the flame after the
drying and the signal was recorded as a peak. The peak areas of all
samples were measured as an estimation of the amounts of lead present.
The graphite furnace was used for determination of lead in the second
half of the work. The sample solution was diluted from 1 to 10. A series
of standards was prepared ranging from 0 to 0.05 ppm. Operating conditions
of the furnace are given in Table 2.
8
INSTITUTE OF GAS TECHNOLOGY
-------
5/73
8937
Table 2. GRAPHITE FURNACE CONDITIONS
FOR LEAD
Drying
Ashing
Atomizing
Time, s Temperature, "C
50 150
50 400
3 2500
6. Cadmium
Cadmium line 228. 8 nm was used in this work. The graphite furnace was
employed for cadmium determination. The sample solution was diluted from
1 to 100. A series of standard solutions was prepared containing from 0.001
to 0. 005 ppm of cadmium. Operating conditions are given in Table 3.
Table 3. GRAPHITE FURNACE CONDITIONS
FOR CADMIUM
Drying
Ashing
Atomizing
Time, s Temperature, °C
50
50
150
400
2400
7. Arsenic
Arsenic line 193. 7 nm was used in this work. The sample solution was
diluted from 1 to 10, and a series of standard solutions was prepared con-
taining 0 to 0. 15 ppm of arsenic. Operating conditions for the graphite
furnace are given in Table 4.
Table 4. GRAPHITE FURNACE CONDITIONS
FOR ARSENIC
Drying
Ashing
Atomizing
Time, s
50
50
3
Temperature, °C
150
400
2300
INSTITUTE
0 F
GAS
TECHNOLOGY
-------
5/73 8937
8. Antimony
Antimony line 231. 2 nm was used. The sample solutions were diluted a
hundredfold and a series of standards ranging from 0 to 0.003 ppm of antimony
was prepared. Operating conditions for the graphite furnace are given in
Table 5.
Table 5. GRAPHITE FURNACE CONDITIONS
FOR ANTIMONY
Time, s Temperature, "C
Drying 50 150
Ashing 50 400
Atomizing 3 2400
9. Selenium
Selenium line 196.0 nm was used. Standard addition methods were used
to offset the matrix effect, which becomes pronounced when sample solution
is used without substantial dilution. The sample solutions were mixed with
equal portions of standards containing 0, 0. 05, 0. 10, and 0. 15 ppm of
selenium. Operating conditions of the graphite furnace are given in Table 6.
Table 6. GRAPHITE FURNACE CONDITIONS
FOR SELENIUM
Time, s Temperature, °C
Drying 60 180
Ashing 70 450
Atomizing 4 2400
10. Tellurium
Tellurium line 214.3 nm was used. For the same reason as for
selenium, standard addition methods.were used. A series of standard
solutions containing 0 to 0. 15 ppm of tellurium was added to the samples.
Operating conditions of the graphite furnace are given in Table 7.
10
INSTITUTE OF GAS TECHNOLOGY
-------
5/73
8937
Table 7. GRAPHITE FURNACE CONDITIONS
FOR TELLURIUM
Drying
Ashing
Atomizing
Time, a
60
70
5
Temperature, "C
180
450
2400
11. Mercury
Mercury line 253. 7 nm was used. A Perkin-Elmer mercury analysis
system was assembled with the Perkin-Elmer atomic absorption spectro-
photometer Model 303 for mercury determination. The reducing agents used
were 5^ hydroxylamine and 10$ stannous chloride. The method was checked
according to reference materials of the National Bureau of Standards-and found
to be reliable. A 100-ml aliquot of each sample solution was taken for each
determination. A standard curve was prepared by use of 50 and 100 ng of
mercury.
N S T I T U T E
11
0 F
GAS
TECHNOLOGY
-------
5/73 8937
CHAPTER II. RESULTS AND CONCLUSIONS
A. PROXIMATE AND ULTIMATE ANALYSES
Table 8 gives the proximate and ultimate analyses of the samples.
It was expected that the volatile materials decrease as the percentage of
ash increases throughout the gasification process. However, the reverse
happened at Runs R-FP-147 and F-OH-9, and R-OH-9 and F-EGO-37, which
should have identical values on the table. This indicates some difficulties
with the homogeneity of the samples, and resulted in an increase in con-
centration of beryllium, vanadium, and chromium in the samples between
the feed and residue stages of pretreatment. This increase is illustrated
by the values of beryllium in Figure 8 and for chromium and vanadium in
Figure 9, referenced later in the paper.
B. ASHING OF COAL
The relationship between RF power input and sample temperatures
is shown in Table 9.
To discover the effect of the sample particle size on the rate of
oxidation, samples were used in a series of new experiments under the same
ashing conditions as before except that the samples were unground and
contained on the average of 90$ by weight of —8 to +80 mesh size fractions.
A comparison of the results indicates that for partially gasified coal
and char residues, the particle size of crushed coal below 8 mesh size has
little or no effect on the rate of low-temperature ashing, but —8 to 480 raw
coal samples require up to 60 hours for ashing as compared with 30 hours
for a similar raw coal passing an 80 mesh screen. The ashing times versus
the percentages of ash for various —80 mesh samples are shown in Figures
2 through 7.
Based on the values of Table 9 and the desirability of keeping the LTA
ashing temperature as low as possible while still keeping the ashing time to
a minimum, we have arrived at the following schedule.
Oven-dried coal samples are initially placed in the 6-cell LTA and oxidized
12
INSTITUTE OF GAS TECHNOLOGY
-------
-J
OJ
Table 8. PROXIMATE AND ULTIMATE ANALYSES, wt
Proximate, As Rec'd F-FP-147 R-FP-147 F-OH-9 R-OH-9 F-EGO-37 R-EGO-37
H
m
O
•n
u>
O
en
H
m
X
z
O
O
0
Moisture
Volatile Matter
Ash
Fixed Carbon
Total
Ultimate, Dry Basis
Ash
Carbon
Hydrogen
Sulfur
Oxygen
Nitrogen
Total
1.4
38.6
11.6
48.4
100.0
11.73
71.0
5.06
4.75
6.33
1.13
100.00
3.7
22.4
12.3
61.6
100.0
12.76
68.4
3.01
3.36
11.10
1.37
100.00
•urh 4
3.1
22.4
11. 1
63.4
100.0
11.48
69.3
3.32
3.88
10.75
1.27
100.00
0.8
2.2
22. 1
74.9
100.0
22.24
73.8
1.20
2.64
—
0.60
100.48
1.6
2.6
20.9
74.9
100.0
21.23
75.0
1. 17
2.35
—
0.59
100/34
0.2
2. V
45.0
52.7
100.0
45.04
53.2
0.61
1.56
—
0.25
100.66
00
vO
OO
-------
5/73 8937
Table 9. SAMPLE TEMPERATURE
VERSUS POWER INPUT
Power Input, W/cell Sample Temperature, °C
10 95
15 110
20 125
25 150
30 190
at 10 W/cell for the first 12 hours. The wattage is then increased to 15 W/cell
for the next 10 hours, followed by a final oxidation at 20 W/cell until the sample
reaches a constant weight. This accounts for a total elapsed time of 30 hours
in the LTA.
Our experience with LTA indicates that the ashing is speeded up if the
sample is taken out and well mixed once every 2-3 hours during the 30-hour
ashing operation; otherwise, the ashing time may be twice as long as when
periodic sample mixing is used.
Figures 2 through 7 were prepared from the data based on LTA ashing
of six analysis samples, each in duplicate. They clearly indicate that the
bulk of the organic matter in coal is oxidized during the first 20 hours and
that the oxidation of the remaining 2 or 3$of carbonaceous matter then
proceeds very slowly, approaching the complete oxidation line asymptotically.
The residual organic matter from this method of ashing does not seem to
interfere in the acid dissolution of the samples because.in practice, the
sample solution is started with nitric acid followed by perchloric acid,
which completely oxidizes any residual carbonaceous matter in the ash
samples.
14
NSTITUTE OF GAS TECHNOLOGY
-------
o
o
X
z
o
(-
o
o
ORGANIC MATTER,wt %
WATTS (RF POWER INPUT)
UJ
3O
TIME,hours
Figure 2. LTA ASHING OF -80 MESH SAMPLES OF F-FP-137
oo
vO
OJ
A-82-718
-------
o
>
o
I
z
o
r
o
o
10
L
ORGANIC MATTER,wt %
WATTS (RF POWER INPUT )
90-*— - 120
I
20 30
TIME,hours
40
50
60
A-82-719
CX3
OJ
-J
Figure 3. LTA ASHING OF -80 MESH SAMPLE OF F-HTEG-4
-------
(Ji
z
(/I
m
n
x
z
o
r
o
o
90
80
70
J-? 60
I-
i
UJ
so
fe 40
CO
CO
3 30
20
10
0
10
ORGANIC MATTER,wt %
WATTS (RF POWER INPUT)
90-*— 120 ~-
20 30
TIME, hours
40
50
60
A-82-717
oo
>o
oo
Figure 4. LTA ASHING OF -80 MESH SAMPLE OF F-EG-10
-------
-0
OJ
o
>
CO
m
O
I
z
O
I-
o
o
ORGANIC MATTER,wt%
WATTS (RF POWER INPUT)
«—90—•—« —120
10
20 30
TIME,hours
40
50
60
A-8Z-7I6
00
x£>
W
-J
Figure 5. LTA ASHING OF -SO MESH SAMPLE OFR-EG-10
-------
o
-n
O
>
m
O
X
z
o
r
o
o
ORGANIC MATTER,wt%
WATTS (RF POWER INPUT)
-120 ^
10
20 30
TIME, hours
40
50
60
A-82-714
oo
vO
OJ
Figure 6. LTA ASHING OF -80 MESH SAMPLE OF F-EG-11
-------
UJ
o
o
X
z
o
r
o
o
0
10
ORGANIC MATTER,wt %
90-
WATTS (RF POWER INPUT)
120
1
20 30 40
TIME,hours
50
60
A-82-715
oo
vO
OJ
Figure 7. LTA ASHING OF -80 MESH SAMPLE OF R-EG-11
-------
5/73 8937
C. DETERMINATION OF INDIVIDUAL ELEMENTS
1. Chromium
Chromium was a relatively easy element to determine in coal samples.
Precision of the overall analytical methods, based on six separate analyses
of the Ireland mine coal, indicates a relative standard deviation of 1.1$
at the concentration of approximately 30 ppm in coal. The analytical results
are summarized in Table 10 along with other trace elements.
Table 10. TRACE ELEMENT CONCENTRATION
OF PITTSBURGH NO. 8 BITUMINOUS COAL AT VARIOUS
STAGES OF GASIFICATION, CALCULATED ON THE RAW
COAL (F-FP-147) BASIS
Total Loss,
96
74
65
64
63
62
33
30
24
18
0
With a little variation in its level, chromium is essentially totally retained
in the residue throughout the gasification process.
2. Nickel
Elements
Hg
Se
As
Te
Pb
Cd
Sb
V
Ni
Be
Cr
F-FP-147
0.27
1. 7
9.6
0.11
5.9
0.78
0.15
33
12
0.92
15
R-FP-147,
F-OH-9
0. 19
1.0
7. 5
0.07
4.4
0.59
0.13
36
11
1.0
17
R-OH-9,
F-EGO-37
ppm
0.06
0.65
5.1
0.05
3.3
0.41
0. 12
30
10
0.94
16
R-EGO-37
0.01
0.44
3.4
0.04
2.2
0.30
0. 10
23
9.1
0. 75
15
Determination of nickel at the level of approximately 20 ppm in the
Ireland mine coal gave a relative standard deviation of 3. 6%, which is
reasonably good for this purpose. The analytical results are summarized
in Table 10.
During the gasification process about one-fourth of the nickel is lost.
3. Vanadium
Vanadium was determined by nitrous oxide-acetylene flame on the AAS.
Five-tenths gram of sample-ash dissolved in a final solution made up to
21
INSTITUTE OF GAS TECHNOLOGY
-------
5/73 8937
50 ml gave a vanadium concentration of approximately 2 ppm. The detection
limit of the vanadium under the experimental conditions was about 0. 7 ppm,
with a relative standard deviation of 9$ at the 2-ppm level. The results of
the analyses are summarized in Table 10 and discussed in Chapter III.
There was a vanadium loss during the gasification process of 30$
4. Beryllium
Beryllium is a very toxic element. The detection limit of beryllium by
the AAS is very low; however, its level in coal is very low also. The
beryllium in these six samples was determined both by direct aspiration
and by use of a graphite furnace. Both results are shown in Table 11.
The standard deviation in the determination of beryllium by the flame method
was about 9$. The graphite furnace was used only to check the flame
method; therefore, no standard deviation measurement was made. The
values reported in Table 10 came from the flame method.
Beryllium showed a loss of 18$ during gasification.
Table 11. COMPARISON RESULTS OF BERYLLIUM BY
DIRECT ASPIRATION FLAME AND BY GRAPHITE AAS
Flame Graphite Furnace
Be, ppm, calculated on sample basis
F-FP-147 0.9 0.9
R-FP-147 1.1 0.8
F-OH-9 1.1 0.8
R-OH-9 1.9 1.0
F-EGO-37 1.6 1.0
R-EGO-37 2.9 1.7
5. Lead
Lead was determined by the APCD-MIBK extraction method, the micro-
sampling boat technique, and the graphite furnace. A comparison of the
results by these three methods is shown in Table 12 and discussed in
Chapter III.
The standard deviation for this method was about 4#. The microsampling
boat technique and the graphite furnace were used only to check if the results
given by extraction were reasonable. The results from the extraction-flame
method are reported in Table 10.
INSTITUTE OF GAS TECHNOLOGY
-------
5.9
4.7
4.4
6.6
5.7
8.5
--
4.9
--
7.4
6.1
6.6
5.5
4.6
4.4
5.5
5.5
7.3
5/73 8937
Table 12. COMPARISON RESULTS OF LEAD BY
APCD-MIBK EXTRACTION, MICROSAMPLING
BOAT, AND GRAPHITE FURNACE
APCD-MIBK Sampling Graphite
Extraction Boat Furnace
—Pb, ppm, calculated on sample basis —
F-FP-147
R-FP-147
F-OH-9
R-OH-9
F-EGO-37
R-EGO-37
Approximately 60% of the lead was lost during the gasification process.
6. Cadmium
The levels of cadmium in the samples were too low to be determined by
flame AAS; therefore, the graphite furnace was used. The results are shown
in Table 10. The standard deviation for the determination of cadmium was
about 6%.
Sixty-two percent of the cadmium was lost during the gasification process.
7. Arsenic, Antimony, Selenium, and Tellurium
These four elements were difficult to determine in the samples, even
by use of the graphite furnace. Certain matrix effects, which suppress the
signals given by these elements, were shown in the determination. Standard
addition methods were used; the results are summarized in Table 10.
These four elements are considered as highly volatile. In this study,
loss ranged from 33$ for antimony to 74$ for selenium.
8. Mercury
Mercury was determined by the cool vapor AAS. A NBS coal standard
reference material (SRM) was analyzed along with the samples. The certified
value for the SRM is 0. 132 ppm. The mercury found by this method was
23
INSTITUTE OF GAS TECHNOLOGY
-------
5/73 8937
0. 133 ppm, averaged from six runs with a relative standard deviation of
11$ . The analytical results are shown in Table 10.
Mercury, a highly volatile element, has a loss of 96$ during the
gasification process.
24
INSTITUTE OF GAS TECHNOLOGY
-------
5/73 8937
CHAPTER III. DISCUSSION
A. THE CHANGE IN CONCENTRATION OF TRACE ELEMENTS AT
VARIOUS STAGES OF GASIFICATION
The changes in concentration of trace elements in the coal samples at
various stages of gasification are shown in Figure 8. The figure clearly
shows the loss of certain kinds of trace constituents from the residue
during gasification. However, since there are many other effluents and
streams in the gasification process besides the solid residue and the synthetic
natural gas, we cannot conclude where these trace elements have gone.
More statistical data on the solid residue and also more work on the other
kinds of effluents are needed to come to a conclusion as to the fate of these
potentially hazardous trace constituents during coal gasification.
B. ANALYTICAL METHODS
1. Flame Atomic Absorption
The flame method is, in general, simple, fasi^ and relatively free from
interferences. However, its sensitivity did not always meet the require-
mentsof this work. Chromium, nickel, and lead were determined by the
nitrous oxide-acetylene flame method. The relative standard deviation
estimated for chromium was 1.1$, which is very good; nickel was 3. 6$,
which is fair. As for lead, vanadium, and beryllium, the relative standard
deviations were approximately 9$. For lead, the presence of iron made it
necessary to use standard addition methods to check the incomplete
recovery of lead, which gave the major error in measurements. The major
error in measurements for beryllium and vanadium was drifting of the
background signal caused by large-scale expansion.
2. Microsampling Assembly
In direct aspirating methods only approximately 10$ of the sample can
reach the flame. The microsampling assembly is an alternative way to
introduce the entire sample into the flame. The sample solution is put in
a boat or cup and the solvent is volatilized, then pushed into the flame. By
this technique, all the sample elements are volatilized, atomized, and give
peak signals at the same time. The nonvolatile residue in the sampling cup
and boat and the geometrical configuration of the sampling boat make the
25
INSTITUTE OF GAS TECHNOLOGY
-------
5/73
8937
-Se(
1.5
E
Q.
Q.
*•
z
O 1.0
te
o:
LJ
O
O
(J
0.5
Be
Cd
\
\
\
\
PRETREATMENT
HYDRO-
GASIFICATION
ELECTRO-
THERMAL-
GASIFICATION
Be
F-FP-147
R-FP-147
AND/OR
F-OH-9
R-OH-9
AND/OR
F-EGO-37
R-EGO-37
SAMPLE
A-63-983
Figure 8. CONCENTRATION OF TRACE ELEMENTS TESTED
IN PITTSBURGH NO. 8 BITUMINOUS COAL (F-FPT147)
AND COAL RESIDUES AT VARIOUS STAGES OF GASIFICATION
(Part 1. Showing Elements in Low Range of Concentration)
26
INSTITUTE
0 F
GAS
TECHNOLOGY
-------
5/73
8937
PRETREATMENT1
HYDRO-
GASIFICATION
ELECTRO-
THERMAL-
GASIFICATION
F-FP-147
R-FP-147
AND/OR
F-OH-9
R-OH-9
AND/OR
F-EGO-37
R-EGO-37
SAMPLE
A-63-982
Figure 8. CONCENTRATION OF TRACE ELEMENTS TESTED
IN PITTSBURGH NO. 8 BITUMINOUS COAL (F-FP-147)
AND COAL RESIDUES AT VARIOUS STAGES OF GASIFICATION
(Part 2. Showing Elements in 10-40 ppm Concentration Range)
27
INSTITUTE
0 F
GAS
TECHNOLOGY
-------
5/73 8937
shape of the peak change from one run to another. Usually, the volatilities
of the sample solutions decrease with the increasing amounts of nonvolatile
material left in the cup or boat. The peak areas instead of peak heights
were measured to eliminate the differences in volatilities of the different
samples.
Lead was determined by the sampling boat technique and found to be
similar to the extraction method.
3. Graphite Furnace
The graphite furnace has increasingly attracted more and more attention
from analytical chemists n"15 . It increases the residence time of free
atoms in optical paths and also creates a very reducing environment which
makes possible a higher efficiency in free atom formation processes. Most
of the elements give on the order of 104 to 106 times higher absolute
sensitivities when compared with the flame method, and some elements give
even higher sensitivities. However, generally speaking, the graphite furnace
also has more severe interferences than the flame methods. Very careful
work has to be done before one method can be checked out for routine use.
The graphite furnace was used for lead and beryllium for checking pur-
poses. It also was used for those elements which are below the detection
limit of flame methods, namely, arsenic, cadmium, antimony, selenium, and
tellurium. The determinations of arsenic and tellurium were not so good as
for the other elements. The precision was decreased because of a high
blank for arsenic and a small signal, even at 3X scale expanion, for
tellurium.
The precision was fairly good for beryllium, lead, antimony, and selenium.
The relative standard deviations for these elements at the levels measured
were all about 3-4$.
4. Mercury Analysis System
The mercury analysis system includes a plastic absorption cell which
is placed in the optical path on a Perkin-Elmer 303 atomic absorption
spectrophotometer, an aerator fitted into a biochemical oxygen demand
bottle, and a pump that circulates the air in the system.
28
INSTITUTE OF GAS TECHNOLOGY
-------
5/71 H'KW
The major error in the mercury determinations is found in the sample
preparation step, which was carried out by burning the samples in a C-H
train,then collecting them by use of 1$ potassium permanganate in 0.1N
sulfuric acid.
29
NSTITUTE OF GAS TECHNOLOGY
-------
8937
REFERENCES CITED
1. Belcher, C. B. and Brooks, K. A. , Anal. Chim. Acta 29, 202 (1963).
2. Dinnin, J., Anal. Chem. 32, 1475 (I960).
3. Obermiller, E. L. and Freedman, R. W., Fuel 44, 199 (1965).
4. Abernethy, R. F. , and Gibson, F. H. , Report of Investigation 7184.
U. S. Bureau of Mines, 1968.
5. Hillebrand, W. F. , Lundell, G. E. F. , Bright, H. A. and Hoffman, J. I.,
Applied Inorganic Analysis, 2nd Ed. New York: John Wiley, 1955.
6. O1 Gorman, J. V., Suhr, N. H. and Walker, P. L. , Jr. , Appl. Spec. 26,
44 (1972).
7. Bailey, B. W. and Lo, F. C., JAOAC 54, 1447 (1971).
8. South-worth, B.C., Hodecker, J. H. and Fleischer, K. D. , Anal. Chem.
3£, 1153 (1958).
9. Joensuu, O. I., Appl. Spec.25, 526 (1971).
10. Marineko, J. , May, I. and Dinnin, J. I. , U. S. Geol. Survey Prof.
Paper 800-B, B151 (1972).
11. L'Vov, B. V. , Spectrochim. Acta 17, 761 (1961).
12. L'Vov, B. V., Spectrochim. Acta 24B, 53 (1969).
13. Massmann, H. , Z. Anal. Chem. 225, 203 (1967).
14. Massmann, H. , Spectrochim. Acta 23B, 665 (1968).
15. Woodriff, R. and Ramelow, G. , Spectrochim. Acta 23B. 665 (1968).
30
INSTITUTE OF GAS TECHNOLOGY
-------
5/73
8937
BIBLIOGRAPHIC DATA
SHEET
. Report No.
EPA-650/2-73-004
3. Recipient's Accession No.
4. Title and Subtitle
Fate of Trace Constituents of Coal During
Gasification
5. Report Dale
August.1973
7. Author(s)
A. Attari
&• Performing Organization Kept.
No.
9. Performing Organization Name and Address
Institute of Gas Technology
IIT Center, 3424 South State Street
Chicago, Illinois 60616
10. Project/Task/Wotk Unit No.
11. Contract/Cram No.
68-02-0277
12. Sponsoring Organization Name and Address
EPA, Office of Research and Development
NERC-RTP, Control Systems Laboratory
Research Triangle Park, North Carolina 27711
13. Type of Report & Period
Covered
Final
14.
15. Supplementary Notes
u. Abstracts The report gives results of an investigation of the fate of
trace elements of coal during the Hygas coal gasification process.
Ultimately, the investigation should provide information that will enable
environmentally sound operation of future commercial-scale coal gasifi-
cation plants. The report tells of a 6-month effort to set up laboratory
analytical methods and to analyze for sb, As, Be, Cd, Cr, Pb, Hg, Ni, Se,
Te, and V. The samples were representative of the coal input and the
solid effluents of the pretreatment, hydrogasification, and electrother-
mal stages of the IGT pilot plant. Data to date indicate substantial
removal of As, Cd, Pb, Hg, Se, and Te from coal during the gasification
process. However, since there were a limited number of samples, analysis
of additional samples is required before any firm conclusions can be dram,
17. Key Words and Document Analysis. 17o. Descriptors
Air Pollution
Coal
Coal Gasification
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Lead (metal)
17b. Ideniif icrs/Open-Hnded Terms
Air Pollution Control
Stationary Sources
Trace Constituents
Coal Constituents
Hygas
Analytical Techniques
17c. COSATI F.eld/Group ?
Mercury (metal)
Nickel
Selenium
Tellurium
vanadium
18. Availability Statement
Unlimited
FORM NTI5-39 IRBV. 3-72)
19. Security Class (This
Report)
UNCLASSIFIED
20. Security Class (This
Page-
UNCLASSIFIED
21. No. of Pages
38
22. Price
USCOMM-DC 14902-PT2
31
INSTITUTE
0 F
GAS
TECHNOLOGY
------- |