EPA-600/2-77-107k
October 1977
Environmental Protection Technology Series
SOURCE ASSESSMENT:
CARBON BLACK MANUFACTURE
Industrial Environmental Research Laboratory
Office of Research and Development
U.S. Environmental Protection Agency
Research Triangle Park, North Carolina 27711
-------
RESEARCH REPORTING SERIES
Research reports of the Office of Research and Development. U.S Environmental
Protection Agency, have been grouped into five series. These five broad
categories were established to facilitate further development and application of
environmental technology. Elimination of traditional grouping was consciously
planned to foster technology transfer and a maximum interface in related fields.
The five series are:
1. Environmental Health Effects Research
2. Environmental Protection Technology
3. Ecological Research
4. Environmental Monitoring
5. Socioeconomic Environmental Studies
This report has been assigned to the ENVIRONMENTAL PROTECTION
TECHNOLOGY series. This series describes research performed to develop and
demonstrate instrumentation, equipment, and methodology to repair or prevent
environmental degradation from point and non-point sources of pollution. This
work provides the new or improved technology required for the control and
treatment of pollution sources to meet environmental quality standards.
EPA REVIEW NOTICE
This report has been reviewed by the U.S. Environmental
Protection Agency, and approved for publication. Approval
does not signify that the contents necessarily reflect the
views and policy of the Agency, nor does mention of trade
names or commercial products constitute endorsement or
recommendation for use.
This document is available to the public through the National Technical Informa-
tion Service. Springfield. Virginia 22161.
-------
EPA-600/2-77-107k
October 1977
SOURCE ASSESSMENT:
CARBON BLACK MANUFACTURE
by
R. W. Serth and T. W. Hughes
Monsanto Research Corporation
1515 Nicholas Road
Dayton, Ohio 45407
Contract No. 68-02-1874
ROAP No 21AXM-071
Program Element No. 1AB015
EPA Task Officers: E. J. Wooldridge and I. A. Jefcoat
Industrial Environmental Research Laboratory
Office of Energy, Minerals, and Industry
Research Triangle Park, NC 27711
Prepared for
U.S. ENVIRONMENTAL PROTECTION AGENCY
Office of Research and Development
Washington, DC 20460
-------
PREFACE
The Industrial Environmental Research Laboratory (IERL) of
EPA has the responsibility for insuring that pollution con-
trol technology is available for stationary sources to meet
the requirements of the Clean Air Act, the Federal Water
Pollution Control Act, and solid waste legislation. If con-
trol technology is unavailable, inadequate, uneconomical, or
socially unacceptable, then financial support is provided
for the development of the needed control techniques for
industrial processes and extractive process industries.
Approaches considered include: process modifications, feed-
stock modifications, add-on control devices, and complete
process substitution. The scale of the control technology
programs ranges from bench- to full-scale demonstration
plants.
The Chemical Processes Branch of the Industrial Processes
Division of IERL has the responsibility for investing tax
dollars in programs to develop control technology for a large
number (>500) of operations in the chemical industries. As
in any technical program, the first question to answer is,
"Where are the unsolved problems?" This is a determination
which should not be made on the basis of superficial infor-
mation; consequently, each of the industries is being evalu-
ated in detail to determine if there is, in EPA's judgment,
sufficient environmental risk associated with the process to
invest in the development of control technology. This report
111
-------
contains the data necessary to make that decision for the
air emissions from carbon black manufacture.
Monsanto Research Corporation has contracted with EPA to
investigate the environmental impact of various industries
which represent sources of pollution in accordance with EPA's
responsibility as outlined above. Dr. Robert C. Binning
serves as Program Manager in this overall program entitled,
"Source Assessment," which includes the investigation of
sources! in each of four categories: combustion, organic
materials, inorganic materials, and open sources. Dr. Dale A.
Denny of the Industrial Processes Division at Research
Triangle Park serves as EPA Project-Officer. In this study
of carbon black manufacture, Mr. Edward J. Wooldridge and
Dr. 1= Atly Jefcoat served as EPA Task Officers.
Data on the combustion efficiency of carbon monoxide in the
flare were supplied by Dr. William F. Herget, Environmental
Sciences Research Laboratory, U.S. Environmental Protection
Agency, Research Triangle Park, N.C. These data were
obtained in an effort independent of the present study using
the ROSE (Remote Optical Sensing of Emissions) System.
IV
-------
CONTENTS
Preface iii
Figures viii
Tables xii
Symbols xv
I Introduction 1
II Summary 5
III Source Description 10
A. Product Description 10
1. Physical Properties 10
2. Chemical Properties 17
B. Process Description 27
1. Feed Materials 27
2. Reactor 32
3. Product Recovery 41
4. Product Treatment and Storage 44
5. Vacuum Cleanup System 46
6. Storage Tanks 46
7. Plant Shutdown, Turnaround, and Startup 47
8. Reactor System Heat Balance 48
C. Materials Flow 48
D. Geographical Distribution 49
IV Emissions 58
A. Selected Pollutants 58
B. Location and Description 58
v
-------
CONTENTS (continued)
Section IV.B. (continued)
1. Main Process Vent Gas 58
2. Dryer Vent 74
3. Pneumatic System Vent 76
4. Oil Storage Tank Vents 77
5. Vacuum Cleanup System Vent 79
6. Fugitive Emissions 80
C. Environmental Effects 81
1. Definition of a Representative Source 81
2. Emission Factors 81
3. Source Severity 83
4. Source Severity Distributions 94
5. National and State Emissions Burdens 98
6. Affected Population 99
7. Growth Factor 101
V Control Technology 103
A. State of the Art 103
1. Primary Recovery Device 103
2. Combustion Devices 104
3. Dryer Vent Gas Control 110
4. Other Control Devices 111
B. Future Considerations 112
1. Combustion Devices for Existing Plants 112
2. Combustion Devices for New Plants 115
3. Carbon Monoxide Recovery 118
4. Current Research Studies 120
VI Growth and Nature of the Industry 123
A. Present Technology 123
B. Emerging Technology 124
C. Marketing Strengths and Weaknesses 126
1. Rubber Applications 128
2. Printing Inks 130
vi
-------
CONTENTS (continued)
Section VI.C. (continued)
3. Paint and Coatings 133
4. Other Applications 135
5. Industry Forecast 136
VII Appendixes 139
A. Storage Tank Calculations 140
B. Description of Thermal Process 144
C. Plume Rise Calculations 149
D. ASTM Nomenclature for Carbon Black '153
E. Sampling and Analytical Methods 156
F. Sampling Results 185
G. Simulated Source Severity Distribution 196
H. Remote Monitoring of Carbon.Monoxide
in Flare Off-Gas 209
VIII Glossary of Terms 219
IX Conversion Factors and Metric Prefixes 221
X References 223
VII
-------
FIGURES
Number Page
1 Schematic breakdown of U.S. carbon
black production 3
2 Carbon black particle size distributions
for various industry classifications 13
3 The oil furnace carbon black process 28
4 Five oil furnace process reactor designs 39
5 Details of the Cabot reactor 40
6 Bag filter unit for carbon black recovery 42
7 Bag filter cleaning operation 42
8 Vacuum cleanup system in modern furnace
black plant 46
9 Carbon black plant locations 55
10 Deterministic source severity distribution
for carbon monoxide emissions from main
process vent 95
11 Deterministic source severity distribution
for hydrogen sulfide emissions from main
process vent 95
12 Deterministic source severity distribution
for sulfur oxide emissions from main
process vent 96
13 Deterministic source severity distribution
for particulate emissions from main
process vent 96
14 CO boiler-thermal incinerator system for
carbon black plant 106
15 Catalytic incinerator for carbon black
plant 115
16 Thermal incinerator with waste heat boiler 116
17 Cosorb process for carbon monoxide recovery
from industrial gases 119
Vlll
-------
FIGURES (continued)
Number Page
18 Block diagram for production of carbon
black and synthesis gas 122
19 Carbon black production by process 125
20 Carbon black production by raw material 127
21 Domestic carbon black consumption history 129
22 Production history of oil furnace blacks 138
B-l The thermal process for the manufacture
of carbon black 145
E-l Process schematic showing disposition of
main process vent gas and location of
E-2 Schematic of sample pretreatment for POM
analysis 166
E-3 Flow diagram for POM analysis 167
E-4 Typical GC/MS chromatogram of POM hydro-
carbon standards 171
E-5 Chromatogram for carbon black run POM-1 172
G-l Simulated source severity distribution for
particulate emissions from main process
vent 198
G-2 Simulated source severity distribution for
carbon monoxide emissions from main
process vent 198
G-3 Simulated source severity distribution for
nitrogen oxide emissions from main
process vent 199
G-4 Simulated source severity distribution for
hydrocarbon emissions from main
process vent 199
G-5 Simulated source severity distribution for
hydrogen sulfide emissions from main
process vent 200
G-6 Simulated source severity distribution for
carbon disulfide emissions from main
process vent 200
G-7 Simulated source severity distribution for
carbonyl sulfide emissions from main
process vent 201
G-8 Simulated source severity distribution for
carbon black emissions from main
process vent 201
ix
-------
FIGURES (continued)
Number Page
G-9 Simulated source severity distribution for
benzofluoranthene emissions from main
process vent 202
G-10 Simulated source severity distribution for
benzopyrene emissions from main process
vent 202
G-ll Simulated source severity distribution for
7,12-Dimethylbenz[a]anthracene emissions
from main process vent 203
G-12 Simulated source severity distribution for
copper emissions from main process vent 203
G-13 Simulated source severity distribution for
lead emissions from main process vent 204
G-14 Simulated source severity distribution for
nickel emissions from main process vent 204
G-15 Simulated source severity distribution for
phosphorus emissions from main process
vent 205
G-16 Simulated source severity distribution for
nitrogen oxide emissions from dryer vent 205
G-17 Simulated source severity distribution for
carbon black emissions from dryer vent 206
G-18 Simulated source severity distribution for
carbon black emissions from pneumatic
system vent 206
G-19 Simulated source severity distribution for
carbon black emissions from vacuum
cleanup system vent 207
G-20 Simulated source severity distribution for
hydrocarbon emissions from oil storage
tank vent 207
G-21 Simulated source severity distribution for
fugitive particulate emissions 208
G-22 Simulated source .severity distribution for
fugitive carbon black emissions 208
H-l Block diagram of ROSE System 210
H-2 Geometry of ROSE System measurements 210
H-3 Comparison of field spectra and laboratory
spectra obtained with FTS system 213
-------
FIGURES (continued)
Number Paqe
H-4 Laboratory spectra at resolution of
0.25 cm"1 216
H-5 Spectra used to determine reduction in
CO concentrations 218
XI
-------
TABLES
Number Page
1 Source Severities and Industry Contribu-
tion to Total Emissions for Carbon Black
Production 8
2 Typical Properties of Carbon Blacks 12
3 Internal Measurements of a Typical Carbon
Black Particle 16
4 Composition of Typical Carbon Blacks 18
5 Functional Group Analyses of Carbon Blacks 19
6 Polycyclic Organic Matter Contents of
Carbon Blacks 21
7 Trace Substance Analyses of Carbon Blacks 22
8 Stream Code for the Oil Furnace Process
Illustrated in Figure 3 29
9 Analyses of Typical Natural Gases used in
Carbon Black Manufacture 30
10 Analyses of Typical Liquid Hydrocarbon
Feedstocks used in Carbon Black Manu-
facture 31
11 Hydrocarbon Content of a Typical Carbon
Black Feedstock Oil 31
12 Trace Element Content of a Carbon Black
Feedstock 33
13 Vapor Pressure Data for Typical Liquid
Hydrocarbon Feedstock 35
14 Typical Feed Ratios in the Oil Furnace
Process 35
15 Trace Element Content of a Carbon Black
Quench Water 36
16 Typical Reactor Operating Conditions 41
17 Tankage Requirements for a 4.3 x 101* metric
tons/yr Carbon Black Plant 47
XII
-------
TABLES (continued)
Number
18 Reactor System Heat Balance for the Oil
Furnace Process
19 Material Balance for a 4.0 x 104 metric
tons/yr Carbon Black Plant 50
20 Carbon Black Plants 56
21 Carbon Black Production in 1974 57
22 Selected Pollutants and Their Threshold
Limit Values 59
23 Emission Data for Main Process Vent Gas 65
24 Emission Factors for Polycyclic Organic
Compounds in Main Process Vent Gas 68
25 Emission Factors for Trace Elements in
Main Process Vent Gas 69
26 Main Process Vent Gas Composition for a
Typical Carbon Black Plant 75
27 Emissions Data for Combined Dryer Vent 76
28 Emissions Data for Pneumatic System Vent 77
29 Oil Storage Tank Working and Breathing
Losses from a 4.3 x 10^ metric tons/yr
Carbon Black Plant 78
30 Emissions Data for Product Storage and
Handling 79
31 Emission Factors for a Representative
Carbon Black Plant 82
32 Source Severity Factor Equations 86
33 Emission Heights for Representative Source 86
34 Maximum Time-Averaged Ground Level Concen-
tration for a Representative Carbon
Black Plant 87
35 Source Severities for a Representative
Carbon Black Plant 88
36 Source Severities for Polycyclic Organic
Compounds in Main Process Vent Gas 89
37 Source Severities for Trace Elements in
Main Process Vent Gas 90
38 Concentrations of High Severity Trace Ele-
ments in Feedstock and Quench Water 94
Xlll
-------
TABLES (continued)
Number Page
39 Contribution to Total Emissions of
Criteria Pollutants by Carbon Black
Industry 99
40 Affected Population 102
41 Summary of Emission Control Devices in
Furnace Black Plants 104
42 Material Balance for CO Boiler-Thermal
Incinerator Control System 107
43 Estimated Emission Factors for Solid Waste
Incinerators in Carbon Black Plants 112
44 Material Balance for Thermal Incinerator 114
45 Material Balance for Catalytic Incinerator 114
46 Material Balance for Thermal Incinerator
with Waste Heat Boiler 117
47 1974 Carbon Black Consumption 128
48 Applications of Carbon Black in the
Rubber Industry 131
49 Applications of Carbon Black in the Ink,
Paint, Paper, and Plastics Industries 134
A-l Storage Tank Input Data for Carbon Black
Plant 142
A-2 Storage Tank Calculation Summary for
Carbon Black Plant 143
D-l ASTM Carbon Black Particle Size Code 154
D-2 ASTM Carbon Black Nomenclature System 155
E-l Reproducibility Study Using POM Standards 173
E-2 Processing Recovery Study Using POM's 173
F-l Emissions Data From Run 1-TM 178
F-2 Emissions Data From Run 3-TM 181
F-3 Emissions Data From Run 4-TM 184
F-4 Emissions Data From Run POM-1 187
F-5 Emissions Data From Run POM-2 189
F-6 Emissions Data From Run POM-3 191
F-7 Comparison of Trace Element Runs with
• Blank Run: SSMS Analysis 202
F-8 Comparison of Trace Element Runs with
Blank Run: ICAP Analysis 204
xiv
-------
SYMBOLS
Symbol Definition
A Area, km2
C Diameter factor
GI Concentration of trace element in sample
from front half (probe and filter) of
sampling train, g/m3
C2 Concentration of trace element in sample
from back half (impingers) of sampling
train, g/m3
Cap Production capacity, metric tons/yr
C. Production capacity of plant i,
metric tons/yr
C Heat capacity at constant pressure of
P stack gas, kcal/g-°K
D Tank diameter, ft
D1 Stack diameter, m
Dp Mean population density, persons/km2
Dp County population density for plant i,
i persons/km2
e Euler's constant =2.72
E Emission factor, g/kg
E1 Emission factor, Ib/ton
F Hazard factor, g/m3
F Equivalent gasoline working loss, bbl/yr
F Paint factor
H Effective emission height, m
H1 Tank outage, ft
H Physical stack height, m
o
KT Turnover factor
K(A, T) Spectral absorption coefficient of CO
at wave length X and temperature T,
s2/kg
L Total petrochemical loss, bbl/yr
A
L Optical path length in plume from
flare, m
LI Total petrochemical loss, Ib/yr
xv
-------
SYMBOLS (continued)
Symbol Definition
L Total equivalent gasoline loss, bbl/yr
L Equivalent gasoline breathing loss,
y bbl/yr
M Molecular weight of material stored,
g/g mole
M1 Molecular weight of stack gas, g/g mole
m Value of right-hand side of equation 4
N Number of turnovers per year
Nbb^X' T^ Spectral radiance of a blackbody at wave-
length X and temperature T, W/cm2-sr-y
N (X, T) Spectral radiance of CO at wavelength X
s and temperature T, W/cm2-sr-y
P Vapor pressure, psia
P1 Atmospheric pressure, dyne/m2
A
P Partial pressure of CO, Pa
Q Mass emission rate, g/s
Qu Heat emission rate, kcal/s
rl
R Universal gas constant, dyne-m/g mole-°K
S Source severity
t, t0 Averaging times, min
T Ambient air temperature, °K
Cl
T Stack gas temperature, °K
u Average wind speed, m/s
v Tank capacity, bbl
V Stack gas exit velocity, m/s
»3
W Liquid density, Ib/gal
x Distance downwind from source, m
AH Plume rise, m
AT Average daily ambient temperature change,
oF
e^ Emissivity of CO at wavelength X
A
a Vertical dispersion coefficient, m
XVI
-------
SYMBOLS (continued)
Symbol Definition
X Maximum mean ground level concentration
max / i , , \ / Q
(short-term average), g/m^
x" Maximum mean ground level concentration
(long-term average), g/m3
x"(x) Annual mean ground level concentration
as a function of distance, g/m3
xvn
-------
SECTION I
INTRODUCTION
Carbon black is a generic name for the class of finely
divided carbons produced by the thermal decomposition of
hydrocarbons. The principal uses of carbon black are as a
reinforcing agent in rubber compounds (especially tires) and
as a black pigment in inks, coatings, paper, and plastics.
Two major processes are presently used in the United States
to manufacture carbon black: the oil furnace process and
the thermal process.
Two other processes, the lamp process for the production of
lampblack and the cracking of acetylene to produce acetylene
black, are used at one plant each in the United States.
However, these are small volume, specialty black operations
which constitute less than 1% of total U.S. carbon black pro-
duction. (Personal communication with H. J. Collyer, Cabot
Corporation, Billerica, Massachusetts, 11 May 1977.) Two
additional processes, the channel process and the gas furnace
process, were formerly operated in the country. The gas
furnace process was phased out in the 1960*s, and the last
channel black plant was closed in 1976. (Personal communi-
cation with H. J. Collyer, Cabot Corporation, Billerica,
Massachusetts, 11 May 1977.)
In the thermal process, natural gas is injected into a heated
refractory chamber where it decomposes to form carbon black.
Since the decomposition reaction is endothermic, the chamber
1
-------
is alternately heated (by burning natural gas) and injected
with feed gas in a cyclic process. By contrast, in the oil
furnace process, a liquid hydrocarbon feedstock is injected
continuously into the combustion zone of a natural gas fired
furnace where it is decomposed to form carbon black. In
the channel process, which is still used in some foreign
countries, thousands of small natural gas flames impinge on
channel irons and deposit carbon black which is subsequently
removed by mechanical scraping. Acetylene black is made in
a continuous process by the exothermic decomposition of
acetylene in water-cooled retorts lined with firebrick.
Lampblack is made in a semicontinuous or batch operation by
burning liquid hydrocarbons in shallow, open pans in a
restricted area with a limited air supply. The gas furnace
process was similar to the oil furnace process except that
natural gas was used instead of the liquid hydrocarbon
feedstock.
Although the manufacture of lampblack dates from antiquity,
modern carbon black production originated in the 1870*s with
the development of the channel process. Production was con-
fined to the United States, where large quantities of inexpen-
sive natural gas were available. In 1912, the serendipitous
discovery of carbon black's unique reinforcement properties
in rubber tires led to a transformation of the industry from
small-scale pigment production to bulk chemical manufacture
which today is carried out on a worldwide basis. The gas
furnace process was introduced in 1922, and the oil furnace
process in 1943. The versatility and high yields obtainable
with the latter process have resulted in its predominance
in the industry today. As illustrated in Figure I,1 the oil
1Carbon Black. In: Minerals Yearbook, Vol. 1. Metals,
Minerals, and Fuels. U.S. Government Printing Office,
Washington, D.C. 1940-1974.
-------
LO
LAMP PROCESS
( LESS THAN
1 % OF TOTAL
PRODUCTION)
LIQUID
HYDROCARBONS
a
b
OIL FURNACE
PROCESS
(90% OF TOTAL
PRODUCTION )
CARBON BLACK PRODUCTION
NATURAL GAS
A /TTYI PMT - .
T 1
THERMAL
PROCESS
( 10% OF TOTAL
PRODUCTION )
CHANNEL GAS FURNACE ACETYLENE PROCESS
PROCESS PROCESS ( LESS THAN
( PHASED OUT ( PHASED OUT 1 % OF TOTAL
IN 1976 ) IN 1960's ) PRODUCTION )
a. LIQUID HYDROCARBONS ARE USED AS THE FEEDSTOCK FOR CARBON BLACK PRODUCTION
b. NATURAL GAS IS USED AS A FUEL IN THE PROCESS
Figure 1. Schematic breakdown of U.S. carbon black production
-------
furnace process accounts for approximately 90% of total car-
bon black production in the United States. For this reason,
the oil furnace process is the primary subject of the present
study.
This document presents a detailed study of the carbon black
industry from the standpoint of atmospheric emissions and
their potential environmental impact. The major results of
the study, summarized in Section II, include emission factors
for each species emitted to the atmosphere from each emis-
sion point within a representative carbon black plant. Also
tabulated are several factors designed to measure the environ-
mental hazard potential of carbon black operations. These
consist of source severities, the industry contribution to
total atmospheric emissions of criteria pollutants, and the
number of persons exposed to high contaminant levels from a
representative plant.
The report includes detailed descriptions of the oil furnace
process (Section III), the emission points within a plant
and the materials emitted to the atmosphere (Section IV),
and present and future aspects of pollution control technology
in the industry (Section V). Economic and production trends
in carbon black manufacture, as well as in those industries
that are major consumers of carbon black, are analyzed in
Section VI. On the basis of this analysis, an estimate of
carbon black production through the remainder of the present
decade is given.
-------
SECTION II
SUMMARY
Carbon black is currently manufactured in the U.S. by two
major processes: the thermal process and the oil furnace
process. The oil furnace process is the predominant method
of production, accounting for about 90% of the carbon black
produced domestically. The 30 carbon black plants currently
in operation in the U.S. have a combined capacity of
1.88 x 106 metric tons/yr (2.1 x 106 tons/yr).a Carbon black
plant capacities range in size from 22.6 x 103 to 174.2 x 103
metric tons/yr and typically operate at ^80% of capacity.
Sources of atmospheric emissions within oil furnace carbon
black plants include the main process vent, the dryer vent,
the pneumatic system vent, the oil feedstock storage tanks,
the vacuum cleanup system vent, and fugitive sources.
In the oil furnace process, carbon black is produced by the
thermal decomposition of a liquid hydrocarbon feedstock in a
refractory-lined steel furnace. The heat required to carry
out the decomposition reaction is supplied by burning natural
gas. The carbon black is recovered from the reactor exhaust
gases by means of bag filters, and the gaseous reaction
products, together with residual carbon black, are vented to
the atmosphere through the main process vent.
1 metric ton = 10 grams = 2205 pounds = 1.1 short tons;
(short tons are designated "tons" in this document); other
conversion factors and metric system prefixes are presented
in Section IX.
-------
The recovered carbon black is mixed with water and pelletized.
The pellets are subsequently dried in a gas-fired dryer. A
part of the combustion gases is fed directly to the interior
of the dryer, and the carbon black thus entrained in the gas
stream is subsequently removed using either a bag filter or
wet scrubber. The combustion gases are then vented to the
atmosphere through the dryer vent.
Carbon black recovered from the reactor effluent stream is
transported from the bag filters to the product finishing
(pelleting, drying, packaging, and storage) area of the plant
by a pneumatic conveying system. The black is recovered
from the entraining air stream by a bag filter, and the
filtered air is exhausted through the pneumatic system vent.
The oil feedstock storage tanks are subject to working and
breathing losses which are uncontrolled in all carbon black
plants. However, these storage tanks meet EPA New Source
Performance Standards for this type of emission point.
Plantwide vacuum cleanup systems are used to pick up fugitive
carbon black from leaks and spills. They are also used in
some plants to control emissions from carbon black storage
bins and the packaging operation. The exhaust stream from
the vacuum system is passed through a bag filter to remove
the carbon black and then vented to the atmosphere.
The small particle size and light, fluffy character of carbon
black give rise to numerous sources of fugitive emissions
within the plant. These are: (1) leaks in the pneumatic
system, storage bins, bagging system, and other process
equipment; (2) cleaning of plugged process equipment;
(3) spillage from torn and broken bags; (4) spillage during
the loading of hopper cars and Sealdbins® (collapsible rubber
containers); (5) cleaning and sweeping the bottoms of hopper
cars; and (6) dismantling of process equipment for repairs.
-------
Emission factors for a carbon black plant employing the
oil furnace process were computed from field sampling data
obtained in this study for the main process vent, and from
emissions data available in the literature for the other
emission points. The emission factors were used to generate
a number of other factors designed to quantify the potential
hazard from carbon black production. The source severity
was defined as the ratio of the time-averaged maximum ground
level concentration to a potentially hazardous concentration
of a given pollutant from a given source. Using Gaussian
plume dispersion theory, source severities were calculated
for a representative carbon black plant which was defined
as a plant using the oil furnace .process, and having the
industry mean production rate of 5.1 x 104 metric tons/yr
(5.6 x 104 tons/yr). Results are summarized in Table 1.
Also listed are the annual mass emissions from the entire
U.S. carbon black industry, and the percentage contribution
of the industry to the total mass emissions of criteria
pollutants (particulate, NO ., SO , carbon monoxide, and
X X
hydrocarbons) from all stationary sources.
The average number of persons exposed to high contaminant
levels from carbon black production was estimated and desig-
nated as the "affected population." The calculation was
made for each species emitted and for each emission point
within a representative plant for which the source severity
exceeds 0.1. The largest value obtained was 11,000 persons,
due to hydrogen sulfide emissions from the main process vent.
Carbon black production in 1974 totaled 1.54 x 106 metric
tons (1.69 x 106 tons). Production in 1980 is expected
to total 1.76 x 106 metric tons (1.94 x 106 tons). Thus
-------
Table 1. SOURCE SEVERITIES AND INDUSTRY CONTRIBUTION TO TOTAL EMISSIONS FOR CARBON BLACK PRODUCTION-
Material emitted
Criteria pollutr.nts
Particulate nuitter'
Main proces;i vent
Dryer vent
Pneumatic system vent
Vacuum cleanup vent
Fugitive emissions
Sulfur oxides
Main procesn vent
Dryer vent
Nitrogen oxidtis
Main procesii vent
Dryer vent
Hydrocarbons*'
Main process vent
Oil storage tanKs
Carbon monoxide
Main process vent
Hazardous materials
Beryllium
Main process vent
Mercury
Main process vent
Chemical substances
Hydrogen sulfide
Main process vent
Carbon disulfide
Main process vent
Carbonyl sulfide
Main process vent
Carbon black
Main process vent
Dryer vent
Pneumatic system vent
Vacuum clecnup vent
Fugitive emissions
Methane6
Main proceis vent
Acetylene6
Main procer.s vent
Ethylene6
Main procens vent
Ethane6
Main procens vent
Propylene
Main procens vent
Propane6
Main proceus vent
Isobutane
Main procens vent
n-Butane
Main procens vent
n-Pentane
Main process vent
POM
Main process vent
Trace elemen'ts9
Main proceis vent
gAg
0.11 ± 70%
0.12 t 89%
0.3 ± 93%
0.03 i 58%
0.1
0.0T
0.23 t 60%
0.28 i 15%
0.6 i 200%
- 100%
50 ± 48%
0.72 ± 10%
1,400 t 19%
0.0000022 1 16%
0.00015 + 386%
- 100%
30 i 82%
30 ± 76%
10 t 99%
0.11 i 70%
0.12 i 89%
0.3 ± 93%
0.03 i 58%
0.1
25 t 47%
45 i 48%
1.6 1 85%
0.0f
f
0.0'
o.of
0.10 ± 80%
0.27 ± 57%
0.0f
0.002 ± 52%
<0.25 ± 43%
Emissions from
all plants
etric tons/yr
1,011
170
185
460
46
150
353.
Of
353
1,352
430
922
78,007
76,900
1,107
2,153,000
2,153,000
<0.0034
<0.0034
0.23
0.23
46,140
46,140
46,140
46,140
15,380
15,380
1,011
170
185
460
46
150
38,450
38,450
69,200
69,200
2,460
2,460
of
0
0
ftf
0~
307
307
153
153
414
414
of
3.1
3.1
<400
<400
tons/yr
1,116
185
203
508
50
170
390,
Of
390
1,490
474
1,016
85,909
84,690
1,219
2,371,000
2,371,000
cO.0037
<0.0037
0.25
0.25
50,814
50,814
50,814
50,814
16,938
16,938
1,116
185
203
508
50
170
42,350
42,350
76,200
76,200
2,712
2,712
of
0 ,
0
340
340
169
169
457
457
t
of
3.4
3.4
<400
<440
Source
severity"
0.02
0.034
0.053
0.008
1.2
0.0f
0.046
0.16
0.58
21
0.84
1.8
<0.028
0.42
20
7.6
2.1
0.45
0.76
1.2
0 .34
27
_e
-C
_e
_e,f
_e,f
_e f
0.001
0.003
0.0f
2.1
<2.7
Industry contribution to
total emissions," %
Nationwide
0.01
0.001
0.008
0.31
2.3
exas
0.08
0.02
0.04
1.5
3
uisiana
0.09
0.07
0.12
1.4
13
alndustry contribution calculated only for criteria pollutants.
Calculated assuming no plume rise. If plume rise is included in the calculations, severities of emissions from the
main process vent are reduced by 80%; severities of emissions from the dryer vent are reduced by 60%; severities of
other emissions are unchanged.
CThe particulate matter is carbon black. Source severities are based on the primary ambient air quality standard for
particulate matter.
dTotal nonmethane hydrocarbons. Source severities are based on the primary ambient air quality standard for hydro-
carbons. The individual hydrocarbon species are listed under chemical substances.
Classified as "inert" gas, i.e., simple asphyxiant.
fNot detected in measurements of main process vent gas composition. Detection limit was 1 ppm.
'includes beryllium and mercury.
-------
assuming that the same level of control exists in 1980 as
existed in 1974, emissions from the carbon black industry
will increase by 14% over that period, i.e.,
Emissions in 1980 _ 1.76 x 106 , ,.
Emissions in 1974 ~ 1.54 x 106
-------
SECTION III
SOURCE DESCRIPTION
A. PRODUCT DESCRIPTION
Carbon blacks are essentially elemental carbon in the form
of nearly spherical particles of colloidal dimensions. All
carbon blacks possess similar properties, and the distinction
between the various grades is one of degree rather than kind.
In determining the utility of carbon blacks for commercial
applications, the most important properties are: (1) particle
size; (2) surface area; (3) extent of particle-to-particle
association (structure); and (4) surface condition. The
basic physical and chemical properties of carbon blacks are
described below.
1. Physical Properties
a. Particle Size - The most important physical property of
carbon black from the standpoint of commercial applications
is particle size. The average particle size of unagglomer-
ated oil furnace blacks ranges from 18 nm to 55 nm, as can
be seen in Table 2.2 For comparison, the properties of
carbon blacks produced by the gas furnace, thermal, and
2Smith, W. R. , and D. C. Bean. Carbon Black. In: Kirk-
Othmer Encyclopedia of Chemical Technology, Second Edition,
Vol. 4. John Wiley & Sons, Inc., New York, New York, 1964.
pp. 243-282.
10
-------
Table 2. TYPICAL PROPERTIES OF CARBON BLACKS2
Grade
Oil furnace blacks
Super abrasion furnace
Intermediate super abrasion
furnace
Intermediate super abrasion
furnace - low structure
Intermediate super abrasion
furnace - high structure
High abrasion furnace
High abrasion furnace - low
structure
High abrasion furnace -
high structure
Fast extruding furnace
General purpose furnace
Conductive furnace
b
Gas furnace blacks
Fine furnace
High modulus furnace
Semireinforcing furnace
Thermal blacks
Fine thermal
Medium thermal
Channel blacks
High color channel
Medium color channel .
Regular color channel
Easy processing channel
Medium processing channel
Medium flow channel
Long flow channel
a
Symbol
SAF
ISAF
ISAF-LS
ISAF-HS
HAF
HAF-LS
HAF-HS
FEF
GPF
CF
FF
HMF
SRF
FT
MT
HCC
MCC
RCC
EPC
MPC
MFC
LFC
Mean particle
diameter,
nm
18 to 21
22 to 25
20 to 23
,22.5
26 to 30
25 to 26.5
22 to 25
40 to 45
50 to 55
21 to 29
40 to 50
60
60 to 80
180
470
9 to 14
15 to 17
22 to 29
29 to 30
25 to 28
23 to 25
22 to 28
Surface area
N2 adsorption,
m2/g
90 to 125
115
110 to 130
110 to 120
74 to 100
85 to 110
80
40 to 45
25 to 30
125 to 200
40 to 50
30 to 40
25 to 30
13
7
400 to 1,000
320 to 550
100 to 140
100
110 to 120
200 to 210
300 to 360
Oil
absorption ,
cm3/g
1.5.
1.3
0.8 to 0.9
1.4 to 1.6
1.15
0.7 to 0.8
1.4 to 1.6
1.3 to 1.4
0.9
1.3
0.9 to 1.1
0.85
0.7 to 0.8
0.3 to 0.5
0.3 to 0.5
2 to 4
1.5
1.1
1.0
1.0
1.1
1.2
Nigrometer
reading
(unitless)
86
88
87
88
90
87
90
95
97
86 to 93
90
95
97
107
110
58 to 69
70 to 78
80 to 85
85
83
80 to 83
80 to 84
Volatile
matter,
%
1.0
1.0
1.5
1.5
1.5
1.5
1.5
1.0
1.0
1.5 to 2
1.0
1.0
1.0
0.5
0.5
5 to 16
5 to 10
5
5
5
7 to 8
12
pH
8 to 9
8 to 9
8 to 9
8 to 9
8 to 9
8 to 9
8 to 9
9
9
8 to 9
8 to 9
8 to 9
8 to 9
9
8
3 to 4
4 to 5
5
5
5
4
3.5
Benzene
extract,
%
0.05
0.05
0.05 to O.J.O
0.05
0.05
0.05
0.05
0.05
0.05
0.06
0.05
0.10
0.15
1.75
0.3
None
None
None
None
None
None
None
a
ASTM numbers corresponding to the industry classification symbols are shown in Appendix D.
Gas furnace blacks are no longer available. Similar blacks are now made by the oil furnace process (personal communication,
H. J. Collyer, Cabot Corporation, Billerica, Massachusetts, 11 May 1977).
CChannel blacks are no longer produced domestically; however, they are still available on the international market (personal
communication, H. J. Collyer, Cabot Corporation, Billerica, Massachusetts, 11 May 1977).
-------
channel processes are also included in the table. The nomen-
clature used in this table is that of the industry descriptive
system, which is based on the manufacturing process and perfor-
mance characteristics of the black. For example, semirein-
forcing furnace black (SRF) denotes a black with intermediate
reinforcing properties in rubber that is produced by the
furnace process. The American Society for Testing and Mate-
rials (ASTM) has also established a comprehensive nomenclature
system for carbon blacks which is given in Appendix D.
Particle size is usually measured with an electron microscope,
and the arithmetic mean diameter is reported. The particle
sizes tend to be log-normally distributed, and the geometric
standard deviation increases with mean particle size.3/4
Typical particle size distributions are shown in Figure 2.
Particle size is of primary importance in determining the
reinforcement properties of carbon blacks in rubber compounds.
Small particle size blacks impart high tensile strength and
abrasion resistance to rubber, but they are difficult to mix
and process. The fully reinforcing blacks (SAF, ISAF, HAF),
which provide maximum abrasion resistance (for example, in
tire tread), range in particle size from about 18 nm to
30 nm.
Gas furnace blacks and, to a large extent, channel blacks
have been replaced by similar blacks made by the oil furnace
process. However, channel blacks are still used in some
applications. For example, federal regulations specify the
use of channel blacks in certain food processing operations.
3Davidson, H. W. , P. K. C. Wiggs, A. N. Churchouse,
F. A. P. Maggs, and R. S. Bradley. Manufactured Carbon.
Pergamon Press, New York, New York, 1968. pp. 1-55.
4
Matsubayashi, E. Carbon Black. Sekiyu Gakkai Shi,
16(5):381-386, 1973.
12
-------
30
20
o
10
0
20 40 60
PARTICLE DIAMETER, nm
80
100
Figure 2. Carbon black particle size distributions
for various industry classifications'4
b. Surface Area - The external surface area of carbon black
particles can be calculated from the particle diameter. The
total area (internal plus external) is usually measured by
gas-adsorption techniques, such as that of Brunauer, Emmett,
and Teller (BET). The difference in these two values pro-
vides a measure of the internal (porous) surface area.
Low total surface area is desirable in rubber grade blacks
since it results in low viscosity and low heat buildup during
rubber processing. The high-color and long-flow ink blacks,
on the other hand, are highly porous, having total surface
areas two to three times greater than their external areas.
Typical specific total surface areas measured by nitrogen
adsorption are given in Table 2 for the various grades of
black. Some of the newer "improved" carbon blacks have
13
-------
surface, areas lower than those of the blacks which they
replace: and are classified in a different category as to
particle size.
c. Structure - The term "structure" refers to the tendency
of carbon black particles to form chain- or grape-like aggre-
gates. The forces responsible for structure development vary
from weak physical attractions to chemical bonds between
particles. The degree of structure of a carbon black has a
major influence upon the rheology of the carbon black/rubber
system formed when the black is added to an elastomer. High
structure blacks tend to disperse more readily in elastomers
than do blacks having lower structure, which results in faster
mixing. It also results in smooth, low die-swell extrusions
of the elastomer-carbon black mixture. High structure is of
paramount importance in producing rubber compounds with high
elastic modulus and high electrical conductivity.
A quantitative measure of structure is obtained from a volume
compressibility measurement or from an oil adsorption experi-
ment. These methods measure the void volume in a sample of
carbon black which, in turn, is a measure of structure. An
example of the oil adsorption method is the soft-ball oil
adsorption test. The minimum amount of linseed oil required
to make one gram of black cohere into a soft ball when care-
fully worked with a spatula is measured. Typical values for
the various grades of blacks are given in Table 2.
The soft-ball test has now been largely replaced by the
dibutyl phthalate (DBF) test (ASTM test number D2414). The
DBF test is a fully automated procedure which is carried out
with a Cabot Adsorptometer, an electric kneader-mixer
equipped with a torque-measuring device.5 Dibutyl phthalate
5Donnet, J. B., and A. Voet. Carbon Black. Marcel Dekker,
Inc., New York, New York, 1976. 351 pp.
14
-------
is added dropwise to a sample of carbon black. At the point
of maximum adsorption (the point at which the voids are
filled), there is a transition from a free-flowing powder to
a semiplastic material which is accompanied by a rapid in-
crease in torque on the mixer. This end point is sensed
automatically and the amount of dibutyl phthalate added to
the sample is recorded. The resulting values are referred
to as DBF numbers.
d. Crystallinity - X-ray diffraction studies have revealed
that carbon black is composed of condensed polycyclic aro-
matic layers. These layers are in the form of crystallites
consisting of three to four parallel layers which are orien-
ted randomly about their normal. A carbon black particle
with a diameter of 20 nm contains approximately 1,500 crys-
tallites, which are themselves oriented randomly with respect
to each other. Thus, it may be said that carbon black par-
ticles exhibit two-dimensional crystallinity. They differ
in this way from the three-dimensional structure of graphite
particles.
Table 3 gives the internal measurements of a typical carbon
black particle.6
e. Electrical Conductivity - Electrical conductivity is an
important property of carbon blacks since they are used in
both insulating and antistatic, or conductive, rubber com-
pounds. The resistivity of carbon black at a pressure of
1.034 MPa ranges from a few tenths to 1 ohm-cm, depending
principally upon the amount of oxygen complexed on the
surface of the black.2 However, the resistivity of a rubber
stock depends primarily upon the number of conducting paths
6Mantell, C. L. Carbon and Graphite Handbook. Interscience
Publishers, New York, New York, 1968. pp. 76-105.
15
-------
provided by the black particles, and to a lesser extent upon
the resistance of the carbon particles in each path. Thus,
the resistivity of a rubber compound depends on the loading,
particle size, and structure of the carbon black. Channel
blacks provide high electrical resistance in rubber while
high structure oil furnace blacks provide a high degree of
conductivity.
Table 3. INTERNAL MEASUREMENTS OF A TYPICAL
CARBON BLACK PARTICLE6
The Carbon Black Particle
Diameter
Molecular weight
Number of crystallites
20 nm
5,000,000
1,500
The Crystallite
Diameter
Thickness
Distance between planes
Average number of planes
Molecular weight
1.7 nm
1.2 nm
0.35 nm
3.5
3,000
The Crystallite Plane
Molecular weight
Carbon atoms
Hexagonal groups
1,000
90
35
f. Density - The bulk density of carbon blacks ranges from
24 to 59 kg/m3 before pelletizing and from 97 to 171 kg/m3
after pelletizing.7 Since carbon black is not a true crys-
talline solid, its intrinsic density is not precisely defined.
7Drogi:n, I. Carbon Black. Journal of the Air Pollution
Control Association, 18:216-228, 1968.
16
-------
X-ray diffraction measurements yield a value of 2,180 kg/m3
for the average density of the crystallites comprising the
particle.2 Helium displacement measurements give values of
1,840 to 2,130 kg/m3. The value 1,860 kg/m3 is accepted for
most applications, but the value 1,800 kg/m3 is commonly
employed in the rubber industry.
When physical data for carbon black are unavailable, the cor-
responding values for graphite can be used as an approximation.
g. Jetness - The jetness (darkness) of carbon black, asso-
ciated with its particle size, increases with decreasing
particle size. Quantitative measurements of jetness are made
with a nigrometer, which measures the intensity of light
reflected by a black-oil dispersion in terms of a standard.
The nigrometer reading decreases with increasing jetness.
Typical values are given in Table 2.
2. Chemical Properties
Carbon blacks consist of 90% to 99% elemental carbon, with
oxygen, hydrogen and sulfur comprising the other major con-
stituents. Analyses of several grades of carbon black are
given in Table 4. Residual hydrogen from the hydrocarbon
raw material is distributed throughout the particle.2 It is
believed to be bonded to the carbon atoms by true valence
bonds, resulting in an unsaturated state.8 The oxygen is
chemisorbed on the surface of the carbon black.8 Since little
oxygen is present in the region of the furnace where the
carbon black originates, it follows that the presence of
chemisorbed oxygen arises from subsequent adsorption on the
8Deviney, M. L. Surface Chemistry of Carbon Black and Its
Implications in Rubber Chemistry. A Current Review.
Advances in Colloid Interface Science, 2 (3):237-259, 1969,
17
-------
surface of the black during processing. The principal oxygen-
containing functional groups on the carbon black surface are
phenol, quinone, carboxyl, and lactone. Functional group
analyses of several blacks are given in Table 5.8
Table 4. COMPOSITION OF TYPICAL CARBON BLACKS6
Type
Oil furnace blacks
Super abrasion furnace
Intermediate super
abrasion furnace
High abrasion furnace
Fast extruding furnace
General purpose furnace
Conductive furnace
a
Gas furnace blacks
Fine furnace
High modulus furnace
Semireinforcing furnace
Thermal blacks
Fine thermal
Medium thermal
Channel blacks
High color channel
Low color channel
Easy processing channel
Medium flow channel
Long flow channel
Composition, %
Carbon
97.6
97.6
97.9
98.4
98.6
97.4
98.2
98.8
99.2
99.3
99.4
88.0
95.2
95.6
94.5
91.6
Oxygen
1.01
1.17
0.79
0.58
0.21
1.18
0.40
0.23
0.22
0.10
0.0
10.5
3.6
3.5
4.9
7.7
Hydrogen
0.47
0.32
0.34
0.38
0.37
0.21
0.4
0.4
0.4
0.49
0.36
0.9
0.6
0.6
0.5
0.6
Sulfur
0.64
0.51
0.59
0.67
0.55
0.60
0.01
0.23
0.00
0.01
0.01
0.57
0.50
0.19
0.10
0.12
Gas furnace blacks are no longer available. Similar blacks
are now made by the oil furnace process.
}Channel blacks are no longer produced domestically; however,
they are still available on the international market.
When carbon black is heated progressively from 600°C to
1,500°C, the surface complexes are decomposed and carbon
oxides and hydrogen are evolved.2 The percent weight loss
18
-------
on heating to 927°C is designated the "volatile content."
Typical volatile content values are listed in Table 2.
Table 5. FUNCTIONAL GROUP ANALYSES OF CARBON BLACKS8
Black type
ISAF
HAF-HS
LFC
Functional group content,
milliequivalents/g black
Phenol
0.54
0.11
1.30
Quinone
0.02
0.29
2.29
Carboxyl
0.02
0.00
0.28
Lactone
0.11
N.A.a
0.24
Not available.
Carbon black may also contain from 0.01% to 0.7% combined
and free sulfur, depending on the amount of sulfur-containing
compounds present in the raw material. The combined sulfur
is apparently inert, but the free sulfur may contribute to
crosslinking during the vulcanization of rubber compounds.2
Carbon blacks normally have ash contents of a few tenths of
a percent. The ash content of channel blacks is approximately
0.01% and consists principally of oxides of iron and silica.2
The ash content of furnace blacks varies from 0.1% to 1.0%,
and arises mainly from the water used for quenching the
reaction (see Section III.B). This ash consists of soluble
salts of calcium, magnesium, and sodium, and accounts for the
basic pH of furnace blacks.2
Two different grades of carbon black, N-351 and N-330, were
analyzed for total sulfur in the present study. The N-351
sample contained 1.0% sulfur, while the N-330 sample con-
tained 1.5% sulfur. The high sulfur values are due to the
high (3.7%) sulfur content of the oil feedstock.
19
-------
Furnace blacks also contain an adsorbed component which can
be removed by extraction with hot benzene. The following
polycyclic organic materials (POM's) have been identified in
the benzene extract from a semireinforcing furnace black:9
Anthracene 1,2-Benzopyrene
Phenanthrene 3,4-Benzopyrene
Fluoranthene Perylene
Pyrene o-Phenylenepyrene
Benzo[mno]fluoranthene 1,12-Benzoperylene
Chrysene Anthanthrene
Benz[a]anthracene Coronene
9,10-Dimethylbenz[a]anthracene
An analysis of a high abrasion furnace black (type N339)
revealed a 3,4-benzopyrene content of 23 ppm and a coronene
content of 92 ppm.10 A medium thermal black (type N990)
contained 192 ppm 3,4-benzopyrene and 472 ppm coronene.10
The amounts of benzene extractable material contained in the
various grades of carbon black are listed in Table 2.
The PO,M contents of two oil furnace blacks were measured in
the present study as described in Appendix E. The results
are given in Table 6.
Other substances found in carbon black in trace amounts are
amines, phenols, cyanides, and heavy metals. Analyses of a
number of blacks for these substances are presented in Table 7.
9Nau, C. A. Health Studies Relating to Carbon Black. Pre-
sented at the Conference on Environmental Aspects of
Chemical Use in Rubber Processing Operations, Akron, Ohio,
March 12-14, 1975. 13 pp.
10Collyer, H. J. Carbon Black and Ecology. Presented at
the Conference on Environmental Aspects of Chemical Use in
Rubber Processing Operations, Akron, Ohio, March 12-14,
1975. 10 pp.
20
-------
Table 6. POLYCYCLIC ORGANIC MATTER CON-
TENTS OF CARBON BLACKS
(composition, ppm by weight)
Compound
ASTM no.
N330
N351
Acenaphthylene
Anthracene/phenanthrene
Benzo[c]phenanthrene
Benzofluoranthenes
Benzo[ghi]fluoranthene
Benzo[ghi]perylene/anthanthrene
Benzopyrenes and perylene
Chrysene/benz[a]anthracene
Dibenzanthracenes
Dibenzo[c,g]carbazole
Dibenzopyrenes
Dibenzothiophene
Dimethylanthracenes/phenanthrenes
7,12-Dimethylbenz[a]anthracene
Fluoranthene
Indeno[1,2,3-cd]pyrene
Methylanthracenes/phenanthrenes
Methylcholanthrene
Methy1fluoranthene/pyrene
Pyrene
1.3
_b
_a
_a
_a
0.3
1.0
_a
a
_a
0.2
3.8
_a
0.9
_a
4.9
_a
0.9
27
1.1
2.3
0.7
_a
_a
_a
5.1
9.5
_a
a
_a
0.3
4.5
_b
6.6
_a
1.1
_a
2.8
79
b
Not detected; less than 0.1 ppm.
Not detected; less than 0.2 ppm.
The chemical properties of the carbon black skeleton are
similar to those of large polynuclear aromatic molecules.
The aromatic basicity of carbon blacks is demonstrated by
+ +3
their ability to reduce weak Lewis acids (e.g., Ag , Au
and I2) and the fact that they undergo such aromatic
21
-------
Table 7. TRACE SUBSTANCE ANALYSES OF CARBON BLACKS
(composition, ppm by weight)
Material
Amines
Phenols
Cyanides
Aluminum
Antimony
Arsenic
Barium
Beryllium
Bismuth
Boron
Bromine
Cadmium
Calcium
Cerium
Cesium
Chlorine
Chromium
Cobalt
A OTIKA -K> ii-mVxj-t •**•
**.*-* J-i'A A A, Ctil.i»»"— J-
N110a
380
0.00
0.52
<0.5
<0.5
<0.1
<0.01
<0.25
<0.05
0.75
<0.05
N3263
280
0.11
0.65
<1.0
<0.5
<0.1
<0.01
<0.25
<0.05
0.69
<0.05
N3303
170
0.00
0.14
<0.5
<0.5
<0.1
<0.01
<0.25
<0.05
0.41
<0.05
N4723
460
0.04
0.80
<1.0
<0.5
<0.5
<0.05
<0.4
<0.1
1.04
<0.1
N6603
260
0.62
0.03
<0.5
<0.5
<0.5
<0.05
<0.25
<0.05
0.45
<0.05
N7743
310
0.58
0.05
<0.5
<0.5
<0.5
<0.03
<0.25
<0.05
0.3
<0.05
N330b
225
0.15
0.97
285
<0.05
<0.05
8.1
0.10
<0.05
360
1.2
<0.05
4.4
16
0.30
N351b
30
<0.05
0.34
0.55
<0.05
<0.05
0.07
<0.05
<0.05
26
1.5
<0.05
0.30
4.0
0.10
NJ
-------
Table 7 (continued). TRACE SUBSTANCE ANALYSES OF CARBON BLACKS
(composition, ppm by weight)
Material
Copper
Dysprosium
Erbium
Europium
Fluorine
Gadolinium
Gallium
Germanium
Gold
Hafnium
Holmium
Indium
Iodine
Iridium
Iron
Lanthanum
Lead
Lithium
ASTM number
NllO3
<0.15
<0.65
N3263
0.30
2.5
N3303
0.24
0.80
N4723
0.02
0.35
N6603
0.51
3.3
N7743
0.53
2.7
N330b
110
<0.05
<0.10
0.05
205
<0.05
0.05
<0.05
<0.05
<0.15
<0.05
<0.05
<0.05
400
5.0
2.2
0.70
N351b
13
<0.05
<0.10
<0.05
9.0
<0.05
<0.05
<0.05
<0.05
<0.15
<0.05
<0.05
<0.05
31
4.9
0.45
0.30
-------
Table 7 (continued). TRACE SUBSTANCE ANALYSES OF CARBON BLACKS
(composition, ppm by weight)
Material
Lutecium
Magnesium
Manganese
Mercury
Molybdenum
Neodymium
Nickel
Niobium
Osmium
Palladium
Phosphorus
Platinum
Potassium
Praseodymium
Rhenium
Rhodium
Rubidium
Ruthenium
ASTM number
NllO3
<0.50
<0.05
<0.20
2.5
N3269
0.18
<0.05
<0.20
4.3
N3303
0.58
<0.05
<0.20
11.1
N4723
0.47
<0.05
<0.20
0.49
N6603
1.5
<0.05
<0.20
3.8
N7743
1.8
<0.05
<0.20
4.1
N330b
<0.05
16
2.0
0.80
0.20
1.1
43
<0.05
<0.05
<0.05
14
<0.05
700
1.2
<0.05
0.85
<0.05
N351b
<0.05
15
0.40
0.10
0.75
2.7
2.3
<0.05
<0.05
<0.05
3.5
<0.05
12
1.1
<0.05
<0.05
< 0 . 0 5
-------
Table 7 (continued). TRACE SUBSTANCE ANALYSES OF CARBON BLACKS
(composition, ppm by weight)
Material
Samarium
Scandium
Selenium
Silicon
Silver
Sodium
Strontium
Tantalum
Tellurium
Terbium
Thallium
Thorium
Thulium
Tin
Titanium
Tungsten
Uranium
Vanadium
ASTM number
NllO3
<0.20
<0.25
<0.02
<0.30
N3263
<0.20
<0.25
<0.02
<0.30
N330a
<0.20
<0.25
<0.02
<0.30
N4729
<0.20
<0.5
<0.15
<0.30
N6603
<0.20
<0.25
<0.02
<0.30
N7743
<0.20
<0.25
<0.02
<0.30
N330b
<0.05
<0.05
<0.05
205
13
550
8.0
<0.05
<0.05
<0.05
0.05
<0.10
<0.05
46
12
<0.10
<0.05
1.7
N351b
0.05
<0.05
<0.05
95
2.4
500
0.35
<0.05
<0.05
<0.05
<0.05
<0.10
<0.05
9.0
15
<0.05
<0.05
0.25
Ul
-------
Table 7 (continued). TRACE SUBSTANCE ANALYSES OF CARBON BLACKS
(composition, ppm by weight)
Material
Ytterbium
Yttrium
Zinc
Zirconium
ASTM number
NllO3
<0.42
N3263
0.15
N3303
0.50
N4723
0.31
N6603
4.8
N774a
3.9
N330b
<0.10
0.05
600
0.40
N351b
<0.05
<0.05
8.5
<0.05
Data obtained from personal communications with H. J. Collyer, Cabot Corporation,
13 March 1975 and 2 September 1975.
Data obtained in present study. Values for arsenic, mercury, and selenium were
obtained by atomic absorption; boron was determined by inductively coupled argon
plasma excitation (ICAP); all other values were obtained by spark source mass
spectrometry.
NOTE: Blanks indicate value was not determined.
-------
electrophilic substitution reactions as sulfonation and
Friedel-Crafts alkylation.2
B. PROCESS DESCRIPTION
A flow diagram for the oil furnace carbon black process is
presented in Figure 3. The process streams are identified
in Table 8, and the process itself is described in the fol-
lowing subsections. A description of the thermal process is
given in Appendix B; descriptions of the processes for pro-
duction of acetylene black and lamp black can be found in
Reference 5.
1. Feed Materials
Feed materials used in the oil furnace process consist of
petroleum oil (stream 1), natural gas (stream 2), and air
(stream 3). In addition, small quantities of alkali metal
salts may be added to the oil feed in order to control the
degree of structure of the black. In one example, 0.012 g of
potassium nitrate per kilogram of carbon black was used.
(Personal communication with L. B. Evans, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina,
12 June 1975.) Analyses of typical natural gases used in
carbon black manufacture are given in Table 9.11
The main petroleum stocks used in carbon black manufacture
are:7 (1) cracked fuel oil from thermal cracking of cycle
stocks or the vacuum flash distillate from such cracked fuel
oil; (2) thermal or catalytic cycle stocks; and (3) aromatic
extracts from catalytic cycle stocks. The ideal raw material
1 filler, S. E., and R. E. Barrett. Sampling and Analysis of
Source Emission Samples from a Carbon Black Plant, Final
Report. Contract 68-02-1409, Task Order No. 9, U.S. Environ-
mental Protection Agency, Research Triangle Park, North
Carolina, April 1975. 24 pp. plus appendixes.
27
-------
ATMOSPHERIC EMISSIONS
CO
OIL STORAGE TANK
VENT GAS
NATURAL (2
GAS
BAG
FILTER
1
t
V-^ QUENCH
WATER
QUENCH
TOWER
I
i
(?)
t t
BAG
:ILTER
VY
TO DRYER
MAIN PROCESS VENT GAS
| i^-*. INCINERATOR STACK CAS
INCINERATOR
\
WATER
TO STORAGE
FUGITIVE EMISSIONS
PNEUMATIC SYSTEM
VENT GAS
•^ DRYER VENT GAS
»_ VACUUM CLEAN UP
SYSTEM VENT GAS
OPTIONAL STREAM
Figure 3. The oil furnace carbon black process
-------
Table 8. STREAM CODE FOR THE OIL FURNACE PROCESS
ILLUSTRATED IN FIGURE 3
Stream
Identification
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
Oil feed
Natural gas feed
Air to reactor
Quench water
Reactor effluent
Gas to oil preheater
Water to quench tower
Quench tower effluent
Bag filter effluent
Vent gas purge for dryer fuel
Main process vent gas
Vent gas to incinerator
Incinerator stack gas
Recovered carbon black
Carbon black to micropulverizer
Pneumatic conveyor system
Cyclone vent gas recycle
Cyclone vent gas
Pneumatic system vent gas
Carbon black from bag filter
Carbon black from cyclone
Surge bin vent
Carbon black to pelletizer
Water to pelletizer
Pelletizer effluent
Dryer direct heat source vent
Dryer bag filter vent
Carbon black from dryer bag filter
Dryer indirect heat source vent
Hot gases to dryer
Dried carbon black
Screened carbon black
Carbon black recycle
Storage bin vent gas
Bagging system vent gas
Vacuum cleanup system vent gas
Dryer vent gas
Fugitive emissions
Oil storage tank vent gas
29
-------
Table 9. ANALYSES OF TYPICAL NATURAL GASES
USED IN CARBON BLACK MANUFACTURE7'11
Component
Methane
Ethane
Propane
Iso-butane
n-Butane
Iso-pentane
n-Pentane
Hexanes
Helium
Carbon dioxide
Nitrogen
Oxygen
Composition, mole %
1
87.70
3.94
2.29
0.41
0.76
0.23
0.22
0.26
0.09
0.04
4.06
0.00
2
91.58
4.96
1.68
0.23
0.44
0.10
0.11
0.02
N.A.a
0.32
0.55
0.01
Not available
for the: production of modern, high structure blacks is an oil
which is highly aromatic, low in sulfur, asphaltenes and high
molecuJ.ar weight resins, and substantially free of suspended
ash, carbon, and water. Analyses of several typical liquid
hydrocarbon feedstocks are presented in Table 10. A break-
down of the hydrocarbon content of a typical feedstock is
presented in Table II.12
In addition to carbon, hydrogen, and sulfur, feedstocks
typically contain from 0.002% to 0.1% ash.13 An analysis of
12Whitsel, T. S. Theoretical Aspects of Carbon Black Manu-
facture. Rubber Plastics Age. pp. 1209-1212. October 1964
13Gerstle, R. W. Carbon Black Industry, Final Report. Con-
tract 68-02-1321, Task No. 21, U.S. Environmental Protec-
tion Agency, Research Triangle Park, North Carolina, May
1975. 25 pp.
30
-------
Table 10. ANALYSES OF TYPICAL LIQUID HYDROCARBON FEEDSTOCKS
USED IN CARBON BLACK MANUFACTURE7' l l
Component
Percent carbon
Percent hydrogen
Percent sulfur
Specific gravity @16°C
Molecular weight
Aromatics, % by weight
Feedstock
1
90.63
8.27
1.11
1.060
270
72
2
90.07
8.30
1.20
1.054
275
78.7
3
90.37
7.11
1.77
1.116
265
85.9
4
88.99
7.48
2.73
1.080
240
77
5
90.0
8.2
2.2
1.087
N.A.b
N.A.
The increasing scarcity of low sulfur fuels is resulting in
the use of higher sulfur-containing feedstocks by the carbon
black industry. A sample of a liquid feedstock was analyzed
and found to contain 3.7% sulfur.
Not available.
Table 11. HYDROCARBON CONTENT OF A TYPICAL
CARBON BLACK FEEDSTOCK OIL12
Component
Composition, weight %
Paraffins
Monocyclics
Saturated
Aromatic
Dicyclics
Saturated
Aromatic
Tricyclics
Saturated
Aromatic
Polycyclics
Aromatic
TOTAL
1.4
12.6
6.3
6.6
6.1
4.5
18.8
31
-------
the ash content in a typical carbon black feedstock oil was
made ir. the present study and is presented in Table 12. All
concentrations were obtained by spark source mass spectrometry
(SSMS), with the exception of those for arsenic, mercury, and
selenium, which were obtained by atomic absorption (AA), and
boron, which was obtained by inductively coupled argon plasma
excitation (ICAP).
Vapor pressure data for a typical liquid feedstock are given
in Table 13.
Typical oil, air, and natural gas feed ratios used in the
oil furnace process are listed in Table 14 for the various
grades of black.
In addition to the feed materials, quench water (stream 4)
is sprayed into the reactor near its outlet to cool the reac-
tion products. A sample of quench water obtained in the
present study was analyzed for trace elements in the same
manner as described above for the carbon black feedstock oil.
The results of the analysis are given in Table 15.
2. Reactor
The reactor for the oil furnace process consists of a
refractory-lined steel furnace which is from 1.5 m to 9 m in
length and 0.15 m to 0.76 m in internal diameter.2'7 To
provide maximum efficiency, the furnace and burner are
designed to separate, insofar as possible, the heat genera-
ting reaction from the carbon forming reaction. Thus, the
natural gas feed (stream 2 in Figure 3) is burned to comple-
tion with preheated air (stream 3) to produce a temperature
of 1320°C to 1540°C.11 The reactor is designed so that this
zone of complete combustion attains a swirling motion, and
the oil feed (stream 1), preheated to 200°C to 370°C, 2'7 is
32
-------
Table 12. TRACE ELEMENT CONTENT OF
A CARBON BLACK FEEDSTOCK
Element
Aluminum
Antimony
Arsenic
Barium
Beryllium
Bismuth
Boron
Bromine
Cadmium
Calcium
Cerium
Cesium
Chlorine
Chromium
Cobalt
Copper
Dysprosium
Erbium
Europium
Fluorine
Gadolinium
Gallium
Germanium
Gold
Hafnium
Holmium
Indium
Iodine
Iridium
Iron
Lanthanum
Lead
Lithium
Lutecium
Magnesium
Manganese
Mercury
Molybdenum
Neodymium
Nickel
Niobium
Osmium
Concentration in
feedstock oil,
ppm by weight
12
<0.05
0.50
0.25
<0.05
<0.05
0.10
<0.05
<0.05
5.0
0.70
<0.05
0.30
1.8
<0.05
1.6
<0.05
<0.05
<0.05
4.6
<0.3
<0.05
<0.05
<0.05
<0.10
<0.05
not determined
<0.05
<0.05
5.5
2.5
0.30
<0.05
<0.05
11
<0.05
0.16
0.10
0.40
5.5
<0.05.
<0.05
33
-------
Table 12 (continued). TRACE ELEMENT
CONTENT OF A CARBON BLACK FEEDSTOCK
Element
Palladium
Phosphorus
Platinum
Potassium
Praseodymium
Rhenium
Rhodium
Rubidium
Ruthenium
Samarium
Scandium
Selenium
Silicon
Silver
Sodium
Strontium
Tantalum
Tellurium
Terbium
Thallium
Thorium
Thulium
Tin
Titanium
Tungsten
Uranium
Vanadium
Ytterbium
Yttrium
Zinc
Zirconium
Concentration in
feedstock oil,
ppm by weight
<0.05
4.3
<0.05
8.0
0.15
not determined
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
360
0.15
20
0.10
<0.05
<0.05
<0.05
<0.05
<0.05
<0.05
29
4.2
<0.05
<0.05
0.10
<0.05
<0.05
5.5
0.05
34
-------
Table 13. VAPOR PRESSURE DATA FOR TYPICAL
LIQUID HYDROCARBON FEEDSTOCK3
Temperature, °C
23
66
79
93
107
121
135
Vapor pressure, kPa
0.0
1.4
2.8
3.5
6.2
8.3
10.3
Personal communication with L. B. Evans,
U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina,
12 June 1975.
Table 14. TYPICAL FEED RATIOS IN THE OIL FURNACE PROCESS
Grade of black
SAF
ISAF
HAF
FEF
Air/gas/oil
volumetric ratios
7,466/482/1
6,481/416/1
5,734/358/1
2,990/ 20/1
sprayed into the center of the zone. Preheating is accom-
plished by heat exchange with the reactor effluent and/or by
means of a gas-fired heater. The oil is cracked to carbon
and hydrogen with side reactions producing carbon oxides,
water, methane, acetylene and other hydrocarbon products.
The heat transfer from the hot combustion gases to the atom-
ized oil is enhanced by highly turbulent flow in the reactor.
35
-------
Table 15. TRACE ELEMENT CONTENT OF
A CARBON BLACK QUENCH WATER
Element
Aluminum
Antimony
Arsenic
Barium
Beryllium
Bismuth
Boron
Bromine
Cadmium
Calcium
Cerium
Cesium
Chlorine
Chromium
Cobalt
Copper
Dysprosium
Erbium
Europium
Fluorine
Gadolinium
Gallium
Germanium
Gold
Hafnium
Holmium
Indium
Iodine
Iridium
Iron
Lanthanum
Lead
Lithium
Lutecium
Magnesium
Manganese
Mercury
Molybdenum
Neodymium
Nickel
Niobium
Concentration in
quench water,
ppm by weight
0.005
<0.001
0.008
0.026
<0.001
<0.001
0.10
o'.io
0.007
2.8
<0.001
<0.001
33
<0.001
<0.001
0.017
<0.001
<0.001
<0.001
0.4
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
not determined
0.003
<0.001
0.56
<0.001
0.001
not determined
<0.001
1.07
0.11
<0.001
0.003
<0.001
<0.001
<0.001
36
-------
Table 15 (continued). TRACE ELEMENT
CONTENT OF CARBON BLACK QUENCH WATER
Element
Osmium
Palladium
Phosphorus
Platinum
Potassium
Praseodymium
Rhenium
Rhodium
Rubidium
Ruthenium
Samarium
Scandium
Selenium
Silicon
Silver
Sodium
Strontium
Sulfur
Tantalum
Tellurium
Terbium
Thallium
Thorium
Thulium
Tin
Titanium
Tungsten
Uranium
Vanadium
Ytterbium
Yttrium
Zinc
Zirconium
Concentration in
quench water,
ppm by weight
<0.001
<0.001
3.0
<0.001
1.6
<0.001
not determined
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
10.5
<0.001
87
0.005
32
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
<0.001
0.004
<0.001
<0.001
0.001
<0.001
<0.001
0.003
<0.001
37
-------
The design of the furnace and burner for carrying out this
process in an optimum manner constitutes an important part
of proprietary carbon black technology. Each manufacturer
has developed and patented its own furnace design, a number
of which are shown schematically in Figures 4llt'18 and 5.19
The reactor converts 35% to 65% of the feedstock carbon con-
tent to carbon black, depending on the feed composition and
the grade of black being produced. The yields are lower for
the smellier particle size grades of black. Variables that
can be adjusted to produce a given grade of black include
operating temperature, fuel concentration, space velocity
in the reaction zone, and reactor geometry (which influences
the degree of turbulence in the reactor). In addition, the
degree of structure of the black can be controlled by the
use of modifiers, usually salts of the alkali metals, which
are injected into the furnace with the oil feed.20 A typical
set of reactor operating conditions is given in Table 16.3
luKrejci, J. C. Carbon Black Process. U.S. Patent 2,632,713
(to Phillips Petroleum Company), March 24, 1953.
15Improvements in the Manufacture of Carbon Black. British
Patent 669,968 (to Columbian Carbon Company), April 9,
1952.
16Braeridle, H. A. Manufacture of Carbon Black. U.S. Patent ,
2,735,753 (to Columbian Carbon Company), February 21, 1956.
17Stokes, C. A. Process of Producing Carbon Black and Syn-
thesis Gas. U.S. Patent 2,672,402 (to Cabot'Corporation),
March 16, 1954.
18Williams, I. Process and Apparatus for Making Carbon
Black. U.S. Patent 2,625,466 (to J. M. Huber Corporation),
January 13, 1953.
19Process and Apparatus for the Production of Carbon Black.
British Patent 699,406 (to Godfrey L. Cabot Corporation),
November 4, 1953.
20Friauf, G. F., and B. Thorley. Carbon Black Process.
U.S. Patents 3,010,794 and 3,010,795 (to Cabot Corporation),
November 28, 1961.
38
-------
A B
9
0
2.1m
TRANSVERSE
SECTION AT A
TRANSVERSE
SECTION AT B
lOIL"
-AIR& GAS MIXTURE.
GAS-AIR, COUNTERCLOCKWISE,
dd
CITIES SERVICE CO.
( REF. 15)
f 0 18 rn DIAMETER, 4.6m LENGTH"
OIL, CLOCKWISE
CITIES SERVICE CO.
( REF. 16)
GAS REFRACTORY.
AIR
] 0.15m DIAMETER, 3 m LENGTH
ATOMIZING
AIR
OIL
GAS
PHILLIPS PETROLEUM CO.
( REF. 14)
AIR
OIL
CABOT CORP.
( REF. 17 )
rf
-GAS
HUBER CORP.
( REF. 18)
Figure 4. Five oil furnace process reactor designs
39
-------
AIR
DUCTS
FURNACE ASSEMBLY
AIR FOR
ATOMIZATION
(VIIXING CHAMBER
FOR OIL AND
ATOMIZING FLUID-
GAS -GAS JETS = B
I OIL NOZZLE = C
ATOMIZED
OIL
OIL, GAS AND AIR DUCTS
BURNER HEAD BLOCK
A = AIR PIPES
VIEW OF AIR DUCTS
FACING FLOW
ATOMIZING TIP
BURNER HEAD
Figure 5. Details of the Cabot reactor
19
40
-------
Table 16. TYPICAL REACTOR OPERATING CONDITIONS3
Parameter
Type of black produced
Rate of oil feed
Preheat temperature of oil
Rate of air feed
Rate of natural gas feed
Furnace temperature in reaction zone
Rate of carbon black production
Yield of black (based on carbon in
oil feed)
Value
HAF
0.76 m3/hr
288°C
6,653 rnVnr
623 m3/hr
1,400°C
390 kg/hr
60%
The hot combustion gases and suspended carbon black are
cooled to about 540°C by a direct water spray in the quench
area, located near the reactor outlet. The reactor effluent
(stream 5 in Figure 3) is further cooled by heat exchange in
the air and oil preheaters. It is then sent to a quench
tower where direct water sprays finally reduce the stream
temperature to 230°C.7
3. Product Recovery
Carbon black is recovered from the reactor effluent stream
by means of a bag filter unit. Figure 6 is a diagram of a
typical unit.21 The carbon black laden gases from the quench
tower (stream 8 in Figure 3) enter the hopper trough below
the bag cell plates at 200°C to 230°C. The gases flow into
the individual bags of each compartment through the cell
21Schwartz, W. A., F. B. Higgins, Jr., J. A. Lee, R. Newirth,
and J. W. Pervier. Engineering and Cost Study of Air Pollu-
tion Control for the Petrochemical Industry. Volume 1:
Carbon Black Manufacture by the Furnace Process. EPA-450/
3-73-006-a, U.S. Environmental Protection Agency, Research
Triangle Park, North Carolina, June 1974. 116 pp.
41
-------
REPRESSING FAN
ACCESS DOORS-
TROUGH
- REPRESSURING HEADER
]>-ACCESS DOOR
COMPARTMENT
PARTITION
CROSS SECTION
Figure 6. Bag filter unit for carbon black recovery
21
plates. The carbon black collects on the inside of the bags
and the filtered gas flows through the bags and out the bag
filter stacks. During the cleaning cycle of each compartment
(see Figure 7), the black is removed from the bag fabric and
drops beick into the hopper trough. It passes from the hopper
through a hammer mill (micropulverizer) that breaks up the
lumps and is then transported to the product treatment section
of the plant via a pneumatic conveyor system (stream 16).
REPRESSURING
FAN
N
STACK VALVE
REPRESSURING
VALVE
REPRESSING
INTAKE VALVE
CARBON
BLACK
CARBON BLACK
& GASES
Figure 7. Bag filter cleaning operation21
42
-------
The bag filter unit consists of from six to 18 compartments,
each of which contains 300 to 400 bags. The bags, made of
fiber glass coated with a graphite-silicon film, are approx-
imately 14 cm in diameter and 3.2 m long.21 Each one hangs
from the roof of the filter compartment on a metal cap which
is tapered on the sides and slightly larger in diameter than
the hem around the top of the bag. The bag is wedged around
the perimeter of the cap to provide a seal and support for
the bag. The bottom of each bag is secured to the cell plate
with a snap ring. The entire unit shown in Figure 6 has a
single stack. However, some units have a separate stack for
each compartment to aid in the location of leaking bags.
The bags are cleaned automatically at 15-minute intervals,
with only one compartment on the cleaning cycle at any given
time. The cleaning cycle requires 15 to 40 seconds and is
accomplished by means of reverse flow and mechanical shaking.
Figure 7 shows a three-compartment unit with the third com-
partment on the cleaning cycle. The gas used for the cleaning
operation is taken from the compartments that are on the
filtering cycle. The repressuring fan generates sufficient
pressure to reverse the gas flow in compartment three.
The exhaust gas from the bag filter unit (stream 9 in Figure
3) is vented directly to the atmosphere in most carbon black
plants. Alternatively, it may be sent to a flare or
incinerator to reduce the contaminant loading (stream 12).
In addition, 13% to 15%ll of the effluent (stream 10) may be
diverted to provide auxiliary fuel for the drying operation,
which is discussed below.
43
-------
4. Product Treatment and Storage
The raw carbon black collected in the bag filter unit must
be further processed to become a marketable product. After
passing through the pulverizer, the black has a bulk density
of 24 to 59 kg/m3, depending on the grade being produced,7
and it is too fluffy and dusty to be transported. It is
therefore converted into pellets or beads with a bulk density
of 97 to 171 kg/m3.7 In this form it is dust-free and suf-
ficiently compacted for shipment.
The carbon black is collected from the pneumatic system
(stream 16) by means of a cyclone, a bag filter, or a cyclone
and bag filter in combination. When a cyclone is used, the
exhaust gas (stream 17) is recycled to the primary bag filter
unit. When a bag filter is used, the exhaust gas (stream 19)
is vented to the atmosphere. The recovered carbon black is
collected in a covered surge bin. (In some plants the micro-
pulverizer is located between the pneumatic system recovery
device[s] and the surge bin.) The cyclone separator typically
has a diameter of 1.4 m and a height of.3 m to 4 m.7 The bag
filter is designed to operate at 1.2 m3/min per square meter
of fabric area.7 Since the gas stream is at approximately
ambient temperature, fiber glass bags are not required on
this unit, and wool, cotton, or Orion® bags are used.
The carbon black is fed from the surge bin via a screw con-
veyor to the pelletizer, where it is mixed with one part of
water to two parts of black. A binding agent such as molasses,
sugar, dextrin, or starch, may be added to the pelletizing
water.22'23 Typical amounts are 0.001 kg to 0.004 kg per kg
22Day, J. V. Carbon Black Pellets and a Process for Their
Manufacture. U.S. Patent 2,850,403 (to Cabot Corporation),
September 2, 1958.
23Alleman> C. E. Carbon Black Pelletizing by Controlling
Power to the Pelletizer Motor. U.S. Patent 3,266,873
(to Phillips Petroleum), August 16, 1966.
44
-------
of carbon black. The pelletizer is a horizontal housing
that contains a revolving axial shaft with pins or spikes
mounted on its periphery. Agitation by the pins causes the
mixture of carbon black and water to form nearly spherical
particles I.16 mm to 3.2 mm in diameter.7 The pellets are
then conveyed to a dryer for removal of the water. The mois-
ture content of the dried pellets (stream 31) is 1% or less
by weight (personal communication, H. J. Collyer, Cabot
Corporation, Billerica, Massachusetts, 11 May 1977).
Typical dryers are of the rotating horizontal drum type and
operate at 190°C to 230°C depending on the product load.7
The dryers are fueled by natural gas, which may be augmented
by a portion of the main process vent gas. From 35% to 70%
of the combustion gas is charged directly to the interior of
the dryer.21 After passing through the dryer, this stream
(stream 26) is sent to a bag filter for removal of entrained
carbon black before being vented to the atmosphere. The
remaining 30% to 65% of the combustion gas (stream 29) acts
as an indirect heat source for the dryer and is vented
directly to the atmosphere.
The dried, pelletized carbon black (stream 31) is screened
and sent to a covered storage bin via a bucket elevator.
Oversize pellets are removed in the screener and recycled
(stream 33) to the pulverizer. From the product storage bin,
the carbon black can be loaded into railroad hopper cars for
bulk shipment or sent to a vacuum bagging system which is
hermetically sealed to prevent emission of black. The black
is packaged in 22.7-kg paper sacks made of 3-ply kraft paper.
The bags are then sent to storage to await shipment. Some
black is also shipped in collapsible rubber containers
(Sealdbins®) and in metal Tote® bins.
45
-------
5. Vacuum Cleanup System
Most (two-thirds21) carbon black plants employ a plantwide
vacuum cleanup system to minimize emissions of carbon black
resulting from spills and leaks. The storage bins and bag-
ging operation can also be vented to this system. A typical
system is shown schematically in Figure 8. Multiple hose
connections on a vacuum line throughout the plant allow most
carbon black spills to be recovered. The recovered black is
passed through a grit separator where dirt and other foreign
matter is removed. The black is then separated from the air
stream by means of a cyclone and/or filter. After passing
through a grinder, the black is recycled to the micro-pulver-
izer in the main process sequence. The filtered air stream
from the vacuum system is vented to the atmosphere.
GRINDER
CARBON BLACK
TO BE REPROCESSED
Figure 8. Vacuum cleanup system in modern
furnace black plant21
6. Storage Tanks
The feedstock and carbon black storage tank requirements for
a 4.3 x 101* metric tons/yr carbon black plant are summarized
in Table 17.
46
-------
The oil storage tanks are maintained at temperatures ranging
from 16°C to 82°C, a typical value being 49°C (personal
communication with L. B. Evans, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, 12 June 1975).
Vapor pressures at these temperatures are typically 0.7 kPa
or less, with a maximum value of about 2.8 kPa (see Table 13).
The oil storage tanks are vented directly to the atmosphere.
Table 17. TANKAGE REQUIREMENTS FOR A 4.3 x 104 METRIC TONS/YR
CARBON BLACK PLANT9
Tank
no.
1
2
3
4
5
6
7
Material
stored
Oil feedstock
Oil feedstock
Oil feedstock
Carbon black
Carbon black
Carbon black
Carbon black
Temperature,
°C
49
49
49
Ambient
Ambient
Ambient
Ambient
Capacity,
m3
3,974
3,974
3,974
1,135
1,135
1,135
1,135
Turnovers
per year
6
6
6
23
23
23
23
Personal communication with L. B. Evans, U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina,
12 June 1975.
Carbon black is stored in covered bins or hoppers at tempera-
tures ranging from ambient to 100°C. Emissions from the vents
on these tanks can be controlled by means of the vacuum clean-
up system or with bag filters. In some plants these tanks
are vented directly to the atmosphere.
7. Plant Shutdown, Turnaround, and Startup
The product recovery bag filters and other emission control
equipment in carbon black plants remain in service during
startup and shutdown operations, and emissions do not in-
crease during these periods. In some plants the reactors
are vented directly to the atmosphere during warmup. Reactor
warmup is accomplished by burning natural gas, and the
47
-------
baghouse is put on line before the oil feed is added.21
Therefore, the only emissions that result from the direct
venting are natural gas combustion products.
The repair of process equipment results in the release of
carbon black to the atmosphere. These emissions can be min-
imized by the use of a vacuum cleanup system to clean the
equipment immediately after opening for repair.
Carbon black plants typically consist of several parallel
equipment trains, each with multiple reactors. Hence,
variations in plantwide atmospheric emissions due to indi-
vidual processing upsets are minimal. The increased particu-
late emissions resulting from leaks in the product recovery
bag filters is an exception.
8. Reactor System Heat Balance
Carbon black production involves two distinct chemical
reactions: an exothermic combustion reaction and an endo-
thermic cracking reaction. The net result is an exothermic
process which evolves an estimated 24.2 MJ/kg of carbon
black produced.21 Using this value, gross energy balance
for the oil furnace reactor system has been estimated, and
may be found in Table 18.
C. MATERIALS FLOW
Table 19 presents a material balance for a 4.0 x 10k metric
tons/yr oil furnace plant based on data contained in Refer-
ence 21 and field sampling data obtained in the present study
(see Section IV.B.I). The stream numbers correspond to
those given in Figure 3; the optional streams 10, 12, 13, and
17, indicated by dashed lines in Figure 3, are not included.
As previously noted, carbon black plants typically consist
48
-------
Table 18. REACTOR SYSTEM HEAT BALANCE FOR THE
OIL FURNACE PROCESS3
Heat sources
MJ/kg black
Heat in
Net heat of reaction
Oil preheat
Heat out
Quench to 230°C
Losses (radiation, preheaters, etc.)
Incremental heat content0
24.2
1.4
Total 25.6
17.9
3.5
4.2
Total 25.6
Basis:
1. Material balance presented in Table 19.
2. Reaction zone temperature of 1370°C.
bl MJ = 1 x 106 joules = 948 Btu (see Section X) .
c
Difference in heat, content between effluent stream at
230°C and feed streams at 27°C.
of a number of parallel process trains, each containing
multiple reactors. The flow rates given in Table 19 are
totals of all the process trains.
The sources of atmospheric emissions from the oil furnace
process are: the main process vent gas (stream 11), the
pneumatic system vent gas (stream 19), the vacuum cleanup
system vent gas (stream 36), the dryer vent gas (stream 37),
fugitive emissions (stream 38), and the oil storage tank
vent gas (steam 39).
D. GEOGRAPHICAL DISTRIBUTION
Carbon black is currently manufactured by eight companies at
30 locations in the United States with a combined capacity
49
-------
Table 19. MATERIAL BALANCE FOR A 4.0 x Id4 METRIC TONS/YR CARBON BLACK PLANT
Stream ro= ;
Description:
a
Temperature, °C: _
Gage pressure, kPa:
1
Oil
feed
300
2
Gas
feed
27
3
Air to
reactor
4
Reactor
quench
16
5
Reactor
effluent
540°
6
Gas to oil
preheater
27
7
Water to
quench tower
16
Component
Flow rates, kg/hr
Hydrogen
Carbon dioxide
Carbon monoxide
Hydrogen sulfide
Sulfur oxides
Carbon disulfide
Carbonyl sulfide
Methane
Acetylene
Ethane
Aromatic oil
Nitrogen
Oxygen
Nitrogen oxides
Carbon black
Water
TOTALS
8,173
8,173
2,042
450
211
2,703
41,010
12,294
53,304
9,255
9,255
563
3,518
7,029
66
0.6
66
23
145
174
41,221
739
8.2
5,014
14,868
73,435
e
™
e
_e
21,597
21,597
a
Blanks indicate data not available.
At reactor inlet.
d
Blanks indicate no mass flow of component.
2
"Data not available.
"At reactor outlet.
-------
Table 19 (continued). MATERIAL BALANCE FOR A 4.0 x 10^ METRIC TONS/YR CARBON BLACK PLANT
Stream no . :
Description:
Temperature, °C:
Gage pressure, kPa:
8
Quench
tower
effluent
230
3
9
Bag
filter
effluent
220
0
11
Main
process
vent gas
220
0
14
Recovered
carbon
black
15
Micro-
pulverizer
feed
16
Pneumatic
system
18
Cyclone
vent gas
Component
Flow rates, kg/hr
Hydrogen
Carbon dioxide
Carbon monoxide
Hydrogen sulfide
Sulfur oxides
Carbon disulfide
Carbonyl sulfide
Methane
Acetylene
Ethane
Aromatic oil
Nitrogen
Oxygen
Nitrogen oxides
Carbon black
Water
TOTALS
563
3,518
7,029
66
0.6
66
23
145
174
41,221
739
8.2
5,014
36,465
95,032
563
3,518
7,029
66
0.6
66
23
145
174
41,221
739
8.2
11
36,465
90,029
563
3,518
7,029
66
0.6
66
23
145
174
41,221
739
8.2
11
36,465
90,029
5,003
5,003
5,503
5,503
21,560
6,440
5,503
33,503
21,560
6,440
550
28,550
Blanks indicate data not available.
Blanks indicate no mass flow of component.
-------
Table 19 (continued). MATERIAL BALANCE FOR A 4.0 x 104 METRIC TONS/YR CARBON BLACK PLANT
-X
o L.I.CCUU 110. ;
Description:
a
Temperature, °C:
Gage pressure, kPa:
19
Pneumatic
system
vent gas
20
Solids
from bag
filter
?1
Solids
from
cyclone
22
Surge
bin
vent
23
Carbon
black to
pelletizer
24
Water to
pelletizer
16
25
Pelletizer
effluent
Component
Flow rates, kg/hr
Hydrogen
Carbon dioxide
Carbon monoxide
Hydrogen sulfide
Sulfur oxides
Carbon disulfide
Carbonyl sulfide
Methane
Acetylene
Ethane
Aromatic oil
Nitrogen
Oxygen
Nitrogen oxides
Carbon black
Water
TOTALS
21,560
6,440
1.0
28,551
549
549
4,953
4,953
e
~e
e
5,502
5,502
3,000
3,000
a
Blanks indicate data not available.
5,502
3,000
8,502
to
Blanks indicate no mass flow of component.
"Data not available.
-------
Table 19 (continued). MATERIAL BALANCE FOR A 4.0 x 104 METRIC TONS/YR CARBON BLACK PLANT
Stream no . :
Description :
a
Temperature , °C :
Gage pressure, kPa:
26
Dryer
direct heat
source vent
200
27
Dryer bag
filter
vent
190
28
Solids
from dryer
filter
190
29
Dryer
indirect heat
source vent
200
30
Hot
gases to
dryer
220
31
Dried
carbon
black
200
32
Screened
carbon
black
Component
Flow rates, kg/hr
Hydrogen
Carbon dioxide
Carbon monoxide
Hydrogen sulfide
Sulfur oxides
Carbon di sulfide
Carbonyl sulfide
Methane
Acetylene
Ethane
Aromatic oil
Nitrogen
Oxygen
Nitrogen oxides
Carbon black
Water
TOTALS
558
8,621
1,822
455
3,435
14,891
558
8,621
1,822
1.0
3,435
14,437
454
454
559
8,622
1,822
435
11,438
1,117
17,243
3,644
870
22,874
5,501
5,501
5,001
5,001
Ul
co
Blanks indicate data not available.
Blanks indicate no mass flow of component.
-------
Table 19 (continued). MATERIAL BALANCE FOR A 4.0 x
METRIC TONS/YR CARBON BLACK PLANT
Stream no. :
Description:
Temperature, °C:
Gage pressure, kPa:
33
Carbon
black
recycle
T A
-;•*
Storage
bin
vent gas
35
Bagging
systems
vent gas
35
Vacuum cleanup
system
vent gas
37
Dryer
vent gas
190
0
38
Fugitive
emissions
33
Oil
storage
tank vent
49
0
Component
Flow rates, kg/hr
Hydrogen
Carbon dioxide
Carbon monoxide
Hydrogen sulfide
Sulfur oxides
Carbon disulfide
Carbonyl sulfide
Methane
Acetylene
Ethane
Aromatic oil
Nitrogen
Oxygen
Nitrogen oxides
Carbon black
Water
TOTALS
500
500
Q
e
3
9
£
0.2
1,117
17,243
3,644
1.0
3,870
25,875
0.5
0.5
3.6
3.6
Ul
Blanks indicate data not available.
Blanks indicate no mass flow of component.
"Data not available.
-------
of 1.88 x 106 metric tons/yr. Table 2013'21"25 lists the
manufacturers and plant capacities, and the plant locations
are shown in Figure 9.2lt'25 The population densities of the
counties in which the plants are located range from 3 to 539
persons per square kilometer.
Figure 9. Carbon black plant locations2k' 25
It can be seen that carbon black production is highly local-
ized near sources of hydrocarbon raw materials. Thus, 20 of
the 30 plants are located in Texas and Louisiana. These two
states accounted for 78% of the total production in 1974.26
The total production for 1974 is given in Table 21.26
Monsanto's lampblack plant at Camden, New Jersey and Union
Carbide's acetylene black plant at Ashtabula, Ohio are not
included in these figures. These plants produce small vol-
umes of specialty blacks which are generally not included
in carbon black industry statistics by the U.S. International
Trade Commission and the Bureau of Mines.
24Chemical Profiles. Carbon Black. Chemical Marketing
Reporter, 210(24):9, 1976.
25Tire, Rubber, and Carbon Black Plant Locations (Map).
Rubber World. 172(2) :42-44, 1975.
26Carbon Black in 1974. Mineral Industry Surveys, U.S.
Department of the Interior, Bureau of Mines, Washington,
D.C., 1975.
55
-------
Table 20. CARBON BLACK PLANTS13'24'25
Company
Ashland Oil
Cabot Corporation
Cities Service
(Columbian
Division)
Commercial Solvents
Continental Carbon
J. M. Huber
Phillips Petroleum
Sid Richardson
Carljon Co.
TOTAL
Location
Arkansas Pass, TX
Belpre, OH
Ivanhoe, LA
Mojave, CA
Shamrock , TX
Big Spring, TX
Franklin/ LA
Pampa , TX
Ville Platte, LA
Waver ly, W.V.
Conroe , TX
El Dorado, AR
Eola, LA
Franklin, LA
Mojave, CA
Moundsville, W.V.
Seagraves , TX
Ulysses, KS
Sterlington, LA
Baker sfield, CA
Ponca City, OK
Sunray, TX
Westlake, LA
Bay ton, TX
Borger, TX
Borger, TX
Orange, TX
Toledo, OH
Addis, LA
Big Spring, TX
Nominal
capacity,3
103 metric
tons/yr
69.8
30.4
108.9
30.8
49.4
90.7
174.2
27.2
88.9
50.8
49.9
44.4
29.9
97.1
22.6
68.0
43.1
27.2
56.7
32.2
55.3
44.0
49.4
120.6
59.9
144.2
43.1
36.3
36.3
54.4
1,880.7
County
population
density,
persons/km2
25
34
36
15
3
16
37
11
18
89
17
16
17
37
15
47
3
4
8
15
19
6
50
386
10
10
76
539
233
16
Process
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Thermal
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Furnace
Nominal capacities are approximate since they depend on the type of
black being produced.
56
-------
Table 21. CARBON BLACK PRODUCTION IN 197426
State
Texas
Louisiana
All others3
TOTAL
Production,
10 3 metric tons
651
541
346
1,538
Percent of
total
42.3
35.2
22.5
100.0
Includes Alabama, Arkansas, California,
Kansas, Ohio, Oklahoma, and West Virginia.
57
-------
SECTION IV
EMISSIONS
A. SELECTED POLLUTANTS
The species considered during this study are listed in Table
22, along with the corresponding threshold limit values
obtained from Reference 27. For criteria pollutants, the
primary ambient air quality standards are also listed since
the latter values are used in the calculation of source
severity (see Section IV.C). In addition to emitted species
identified in previous studies of the carbon black indus-
try,11'21 the present study also considered the polycyclic
organic materials (POM's) and trace elements listed in Table 22,
B. LOCATION AND DESCRIPTION
1. Main Process Vent Gas
The main process vent gas (stream 11 in Figure 3) is the
principal source of atmospheric emissions from carbon black
plants. The emissions consist of gaseous reaction products
together with residual trace elements from the oil feedstock
27TLVs® Threshold Limit Values for Chemical Substances and
Physical Agents in the Workroom Environment with Intended
Changes for 1976. American Conference of Governmental
Industrial Hygienists, Cincinnati, Ohio, 1976. 94 pp.
58
-------
Table 22. SELECTED POLLUTANTS AND THEIR
THRESHOLD LIMIT VALUES
Material emitted
,27 g/m3
Criteria pollutants
Particulate matter
Sulfur oxides
Nitrogen oxides
Hydrocarbons
Carbon monoxide
Hazardous pollutants
Beryllium
Mercury
Chemical substances
Hydrogen sulfide
Carbon disulfide
Carbonyl sulfide
Carbon black
Methane
Acetylene
Ethylene
Ethane
Propylene
Propane
Butanes
n-Pentane
Polycyclic organic matter
Acenaphthylene
Anthracene/phenanthrene
Benzo[c]phenanthrene
0.013 (0.00026b)
0.013 (0.000365b)
0.009 (0.0001b)
0.00016b
0.055 (0.00004b)
0.000002
0.00005
0.015
0.060
0.060C
0.0035
_d
_d
_d
_d
_d
_d
1.4
1.8
0.0002C
0.00026
0.0000011
59
-------
Table 22 (continued). SELECTED POLLUTANTS
AND THEIR THRESHOLD LIMIT VALUES
Material emitted
TLV,27 g/m3
Polycyclic organic matter (cont'd)
Benzofluoroanthenes
Benzo[ghi]fluoranthene
Benzo[ghi]perylene/anthanthrene
Benzopyrenes & perylene
Chrysene/benz[a]anthracene
Dibenzanthracenes
Dibenzo[c,g]carbazole
Dibenzopyrenes
Dibenzothiophene
Dimethylanthracenes/phenanthrenes
7,12~Dimethybenz[a]anthracene
Fluoranthene
Inde:no [1,2, 3-c, d] pyrene
Methylanthracenes/phenanthrenes
Methylcholanthrene
Methylfluoranthene/pyrene
Pyrene
Trace elements
Aluminum
Antimony
Arsenic
Barium
Bismuth
Boron
Bromine
Cadmium
Calcium
0.000001
0.0002s
0.00026
0.0000011
0.00026
o.oooooi1
o.oooooi1
o.oooooi1
0.0002s
0.00026
O.OOOOOI1
0.0002s
o.oooooi1
0.00026
o.oooooi1
0.0002S
0.0002s
_g
0.0005
0.0005
0.0005
_g
_g
0.0007
0.00005
.g
60
-------
Table 22 (continued). SELECTED POLLUTANTS
AND THEIR THRESHOLD LIMIT VALUES
Material emitted
TLV,27 g/m;
Trace elements (cont'd)
Cerium
Cesium
Chlorine
Chromium
Cobalt
Copper
Dysprosium
Erbium
Europium
Fluorine
Gadolinium
Gallium
Germanium
Gold
Hafnium
Holmium
Iodine
Iridium
Iron
Lanthanum
Lead
Lithium
Lutecium
Magnesium
Manganese
Molybdenum
Neodynium
Nickel
Niobium
_9
_9
0.003
0.0005
0.0001
0.0002
_g
_g
_g
0.002
_g
_g
_g
_g
0.0005
_g
o.ooi
_g
_g
_g
0.00015
_g
_g
_g
0.005
0.005
_g
o.oooi
g
61
-------
Table 22 (continued). SELECTED POLLUTANTS
AND THEIR THRESHOLD LIMIT VALUES
Material emitted
g/m;
Trace elements (con't)
Osmium
Palladium
Phosphorus
Platinum
Potassium
Praseodymium
Rhodium
Rubidium
Samarium
Scandium
Selenium
Silicon
Silver
Sodium
Strontium
Tantalum
Terbium
Tellurium
Thallium
Thorium
Thulium
Tin
Titanium
Tungsten
Uranium
Vanadium
Ytterbium
_9
0.0001
0.000002
_g
_g
o.oooi
_g
_g
_g
0.0002
o.oi
o.ooooi
_g
_g
_g
_g
o.oooi
o.oooi
_g
_g
0.002
_g
o.ooi
0.0002
0.0005
_g
62
-------
Table 22 (continued). SELECTED POLLUTANTS
AND THEIR THRESHOLD LIMIT VALUES
Material emitted
Trace elements (con't)
Yttrium
Zinc
Zirconium
TLV,27 g/m'
0.001
0.005
aValue for "nuisance" particles.
Primary ambient air quality standard.
CAssumed equal to value for carbon disulfide.
Classified as "inert" gas, i.e., simple asphyxiant,
No TLV established.
i
f
Value for coal tar pitch volatiles.
Value for carcinogenic compounds adopted for this
program. It corresponds approximately to the
minimum detectable limit.
has not been established.
and quench water, and the carbon black fraction which is not
recovered in the bag filter units. Emissions data for the
main process vent are summarized in Tables 23, 24, and 25.
Table 23 contains the results from a survey of operating
plants conducted by the Houdry Division of Air Products
and Chemicals,21 field sampling data obtained by Battelle
Columbus Laboratories,11 and field sampling data obtained by
MRC in the present study. In the Battelle study, three sets
63
-------
of measurements were performed on the vent gas from each of
the two production trains, one of which was producing N375
grade black while the other was producing N351 grade black.
The emission factors given in Table 23 represent the means
of these three data sets for each unit, except for the POM
value which represents a single measurement. The MRC values
represent averages of data obtained in six separate runs;
the original data are tabulated in Appendix F. During one
run, both N375 and N351 grade blacks were being produced;
grade N330 black was being produced during all other runs.
The error bounds represent 95% confidence intervals about
the mean values.
The emission factors obtained by MRC are based on the assump-
tion that none of the process off-gas is diverted to the
dryers. The corresponding emission factors for plants which
use the off-gas to fuel the dryers would be approximately
15% lower than those listed in Table 23.
Calculations by plant personnel indicate that the flow rate
measurements made in the Battelle study11 were low by about
35% (personal communication with C. B. Beck, Cabot Corpora-
tion, Pampa, Texas, 26 February 1976). If this is indeed
the casa, the corresponding emission factors in Table 21 are
low by approximately 35%. In addition, the emission factors
for the unit that was producing N351 grade black represent
only 85% of the process off-gas since 15% of the off-gas
from that unit was diverted to the dryers.1:
Emission factors for POM's and trace elements emitted from
the main process vent are presented in Tables 24 and 25,
respectively. These values were obtained from the field
sampling program conducted by MRC as described in Appendix E.
64
-------
Table 23. EMISSION DATA FOR MAIN PROCESS VENT GAS11'21
(emission factor, g/kg)
Material
Hydrogen
Carbon dioxide
Carbon monoxide
Hydrogen sulfide
Sulfur oxides
Methane
Acetylene
Nitrogen
Argon
Nitrogen + Argon
Nitrogen oxides
Carbon black
Water
Oxygen
Carbon disulfide
Carbonyl sulfide
Ethylene
Ethane
Propylene
Propane
iso-Butane
n-Butane
POM
9 1
Plant code number
50-1
1,500
9.5
3.0
4.0
50-2
1,694
6.5
2.0
4.5
50-4
1,713
10.5
3.0
5.0
50-5
1,890
10.5
3.0
5.0
50-6
162
1,31.0
2,192
19
5.0
35
12,413
3.0
8,337
50-9
69
559
936
8.0
2.4
15
5,295
1.3
3,557
50-10
52
419
701
6.0
1.6
11
3,967
1.0
2,665
CTi
U1
Blanks indicate no data reported.
-------
Table 23 (continued). EMISSIONS DATA FOR MAIN PROCESS VENT GAS11'21
(emission factor, g/kg)
Material
Hydrogen
Carbon dioxide
Carbon monoxide
Hydrogen sulfide
Sulfur oxides
Methane
Acetylene
Nitrogen
Argon
Nitrogen + Argon
Nitrogen oxides
Carbon black
Water
Oxygen
Carbon di sulfide
Carbonyl sulfide
Ethylene
Ethane
Propylene
Propane
iso-Butane
n-Butane
POM
Plant code number21
50-11
101
736
1,003
19
7.0
45
42
5,444
3.0
4,563
50-19
141
1,238
1,663
17
0.3
34
129
10,193
1.2
50-24
145
1,362
1,710
16
6
45
69
12,308
2.8
0.7
359
50-25
183
1,218
1,747
8.9
60
32
13,404
1.0
0.5
638
50-27
111
1,077
1,369
Trace
Trace
21
6
7,163
6,872
50-28
139
1,718
1,864
22
12
36
80
13,148
1.2
5.1
12,963
50-29
59
857
1,600
20
7.0
21
57
11,494
0.1
en
Blanks indicate no data reported.
-------
Table 23 (continued).
EMISSIONS DATA FOR MAIN PROCESS VENT GAS11'21
(emission factor,9 g/kg)
Material
Hydrogen
Carbon dioxide
Carbon monoxide
Hydrogen sulfide
Sulfur oxides
Methane
Acetylene
Nitrogen
Argon
Nitrogen + Argon
Nitrogen oxides
Carbon black
Water
Oxygen
Carbon disulfide
Carbonyl sulfide
Ethylene
Ethane
Propylene
Propane
Iso-butane
n-Butane
n-Pentane
POM
Trace elements
Plant
code
number21
50-30
100
750
1,625
10
3.0
25
32
9,583
0.1
Battelle data12
N375 black
79.6
504
935
5.4
1.44
20.4
91.5
5,222
85.6
5,308
4,440
78.7
10.7
1.46
1.27
<0.03
0.11
0.41
0.20
0.93
N351 black
85.1
428
791
6.1
0.12
27.4
93.2
1,577
74.1
4,725
4,000
29.5
9.67
1.09
1.29
<0.03
0.11
0.41
0.20
0.95
0.000091
Mean
110
937
1,467
11.5
3.6
30
63
8,877
79.8
8,066
1.7
2.5
5,924
276
10.2
1.27
1.28
<0.03
0.11
0.41
0.20
0.94
0.000091
Range
52 to 183
419 to 1,718
701 to 2,192
trace to 22
trace to 12
11 to 60
6 to 129
4,577 to 13,404
3,967 to 13,148
1.0 to 2.8
0.1 to 5.1
2,665 to 12,963
29.5 to 638
MRC
120 ± 39%
700 ± 21%
1,400 ± 19%
30 L ± 82%
r\
0.0°
25 ± 47%
45 ± 48%
9,000 ± 19%
0.28 ± 15%
0.11 ± 70%
6,000 ± 21%
240 ± 99%
30 ± 76%
10 ± 99%
1.6 ± 85%
„.„
0.0b
0.23 + 109%
- 100%
0.10 ± 80%
0.27 ± 57%
0.0b
0.002 ± 52%
<0.25 ± 43%
Blanks indicate no data reported.
Not detected at detection limit of 1 ppm.
-------
Table 24. EMISSION FACTORS FOR POLYCYCLIC
ORGANIC COMPOUNDS IN MAIN PROCESS VENT GAS
Compound
Acenaphthylene
Anthracene/phenanthrene
Benzo [c ] phenanthrene
Benzofluoroanthenes
Benzo [ghi] f luoranthene
Benzo [ghi ] perylene/anthanthrene
Benzopyrenes & perylene
Chrysene/benz [a] anthracene
Dibenzanthracenes
Dibenzo [c , g] carbazole
Dibenzopyrenes
Dibenzothiophene
Dimethyl-anthracenes/phenanthrenes
7 , 12-Dd.methylbenz [a] anthracene
Fluoranthene
Indeno 1 1 , 2 , 3-cd] pyrene
Methyl-anthracenes/phenanthrenes
Methylcholanthrene
Methyl fluoranthene/pyrene
Pyrene
TOTAL
Emission factor
8
7
<2
3
4
2
3
9
<2
<2
<2
1
1
7
6
<2
1
<2
2
5
1
.0
.0
.0
.0
.0
.3
.0
.0
.0
.0
.0
.4
.4
.0
.0
.0
.0
.0
.3
.0
.9
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
10-
10"
10~
10"
10-
10-
10-
10-
10-
10"
10~
10-
10~
10-
10"
10-
10-
10-
10-
10~
10-
u
5
6
5
5
5
5
6
6
6
6
5
k
5
5
6
k
6
5
i*
3
±
±
+
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
, g/kg
142%
170%
41%
125%
187%
130%
123%
45%
50%
50%
50%
183%
128%
112%
166%
50%
100%
50%
86%
120%
52%
68
-------
Table 25. EMISSION FACTORS FOR TRACE
ELEMENTS IN MAIN PROCESS VENT GAS
Element
Aluminum
Antimony
Arsenic
Barium
Beryllium
Bismuth
Boron
Bromine
Cadmium
Calcium
Cerium
Cesium
Chlorine
Chromium
Cobalt
Copper
Dysprosium
Erbium
Europium
Fluorine
Gadolinium
Gallium
Germanium
Gold
Hafnium
Holmium
Indium
Iodine
Iridium
Iron
Emission factor
<2.1 x 10~3 +
2.5 x 10~5 ±
4.8 x 10~5 ±
<2.2 x 10~3 ±
<4.0 x 10~6 ±
<1.4 x 10~5 ±
<5.9 x 10~3 ±
8.8 x 10~6 ±
1.4 x 10~5 ±
<4.0 x 10~2 ±
5.4 x 10~6 ±
<3.6 x 10"6 ±
<6.0 x 10~3 ±
<3.9 x 10~k ±
1.1 x lQ~k ±
<1.0 x 10~2 ±
<2.6 x 10~6 ±
<1.0 x 10~5 +
<6.5 x 10~6 ±
<2.0 x 10~3 +
<2.1 x 10~5 +
<2.2 x 10~6 ±
<2.6 x 10~6 ±
<2.6 x 10~6 ±
<1.6 x 10~5 ±
<2.6 x 10~6 ±
_b
<3.0 x 10~6 ±
<5.8 x 10~6 ±
<4.5 x 10~3 ±
/ g/kg
20%3
78%
36%
84%a
8%
32%
34%3
26%
40%
32%a
28%
28%
102%a
66%3
57%
113%a
33%
5%
28%
52%a
47%
8%
33%
33%
8%
33%
128%
15%
48%3
69
-------
Table 25 (continued). EMISSION FACTORS FOR TRACE
ELEMENTS IN MAIN PROCESS VENT GAS
Element
Lanthanum
Lead
Lithium
Lutecium
Magnesium
Manganese
Mercury
Molybdenum
Neodymium
Nickel
Niobium
Osmium
Palladium
Phosphorus
Platinum
Potassium
Praseodymium
Rhodium
Rubidium
Samarium
Scandium
Selenium
Silicon
Silver
Sodium
Strontium
Tantalum
Terbium
Tellurium
Thallium
Emission factor
3.
<1.
<1.
<4.
<2.
<1.
1.
<1.
<5.
1.
<5.
<7.
<6.
<1.
<7.
<3.
<4.
<2.
2.
<1.
<5.
<1.
<2.
<4.
<3.
<1.
<2.
<3.
<6.
9
5
8
7
07
30
50
38
8
3
2
5
0
00
5
0
5
0
0
4
8
7
0
1
0
0
2
0
8
x
x
X
X
X
X
10-
10~
10"
10-
x 10
x 10
x 10
x 10
10"
10"
x 10"
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
10"
10"
x 10
10~
10"
10"
10-
10"
lo-
10"
10-
10-
10"
10"
10-
10"
10"
10"
e
3
5
6
mm O
-u
-u
-u
6
3
6
6
6
-2
6
2
6
6
5
5
5
6
2
_b
2
«*
5
6
6
6
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
, g/kg
8%
46%
76%
78%
139%
138%
386%
38%
37%
90%
158%
55%
36%
220%
55%
137%
119%
64%
191%
191%
37%
66%
95%
42%
262%
135%
19%
71%
59%
a
a
a
a
a
a
a
a
a
a
70
-------
Table 25 (continued). EMISSION FACTOR FOR TRACE
ELEMENTS IN MAIN PROCESS VENT GAS
Element
Thorium
Thulium
Tin
Titanium
Tungsten
Uranium
Vanadium
Ytterbium
Yttrium
Zinc
Zirconium
TOTAL
Emission factor
<1.4 x 10~5 ±
<4.4 x 10~6 ±
<1.3 x 10~3 ±
2.0 x 1Q'1* ±
<2.5 x 10~5 ±
<8.0 x 10~6 ±
<1.3 x 10~5 ±
<9.6 x 10~6 ±
<8.0 x 10~6 ±
<5.0 x 10~2 ±
<6.0 x 10~5 ±
<0.25 ± 40%
/ g/kg
81%
120%
42%a
122%
45%
175%
138%a
32%
49%a
135%a
155%a
Value reported as upper bound since
significant amount of material was
found in blank run (see Appendix F),
Not determined.
71
-------
Each value in these tables represents the mean of three
individual runs, and the error bounds represent 95% confi-
dence limits for the mean values. Data for individual
sampling runs are tabulated in Appendix F. The values listed
in Table 25 are those obtained by spark source mass spectros-
copy with the following exceptions: arsenic, mercury, and
selenium were determined by atomic absorption (AA) and boron
was determined by inductively coupled argon plasma excita-
tion (ICAP).
As noted in Table 25, significant (see Appendix F) amounts
of a number of trace elements were found in a blank run which
was made as a check on the sampling and analytical procedures.
The emission factors for these elements are reported as upper
bounds since the actual emission factors must be less than
the apparent values, which contain a contribution due to the
sampling and/or analytical methods. Those elements listed
as upper bounds in Table 25, but which are not footnoted,
were present in concentrations below the detection limits
of the cinalytical methods.
The trace element concentrations found in the blank run indi-
cate sample contamination through either the sampling pro-
cedure or the analytical work-up procedures. (These procedures
are summarized in Appendix E.) One possible source of con-
tamination is the glassware used to collect and process the
samples., Silicon, sodium, calcium, magnesium, and aluminum
are major constituents of glass. Silicon was also present in
the sampling train as silica gel. Thus, the presence of
these elements in the blank samples is not surprising. Sources
which would account for the anomalous results for other ele-
ments, such as phosphorous, potassium, and zinc, have not
been identified.
72
-------
A blank POM run was also made and analyzed. No POM's were
detected in the blank samples.
The MRC data are generally consistent with the data from the
other two studies in that the MRC measurements fall within
the ranges of the values from the Battelle and Houdry studies
as shown in Table 23. The only exceptions are NO , for
J^
which only three values are available for comparison, and
hydrogen sulfide. The latter value was verified by means of
a sulfur material balance performed by plant personnel using
the MRC sampling data. Closure of the material balance was
obtained to within less than 5%.
Emissions of sulfur-containing compounds obviously depend
on the sulfur content of the feedstock. The increasing
scarcity of low-sulfur fuels is forcing the carbon black
industry to utilize higher sulfur feedstocks, even though a
high sulfur content creates a problem in the maintenance of
plant equipment.
The composition of the vent gas also depends upon the type
of carbon black produced. As previously noted, the two sets
of data obtained by Battelle correspond to two different
grades of black, N375 and N351. Differences in the oxygen
and sulfur oxides emissions between the two data sets are
apparent. However, the information necessary for a more
detailed description of the variation in emissions as a
function of carbon black type is considered proprietary by
the operating company since it would disclose specific
details of process design. (Personal communications with
H. J. Collyer, Cabot Corporation, Billerica, Massachusetts,
13 March 1975 and 2 September 1975.)
The ranges given for emission factors in Table 23 are indic-
ative of the degree of variability in emissions throughout
73
-------
the carbon black industry. Another measure of variability,
the coefficient of variation (standard deviation divided by
mean), was also computed from the data in Table 23 for five
emitted, species, with the following results: hydrogen, 38%;
carbon dioxide, 38%; carbon monoxide, 26%; hydrogen sulfide,
53%; methane, 46%.
The composition of a typical vent gas is presented in Table 26,
based on the MRC emissions data in Tables 23, 24, and 25.
2. Dryer Vent
The combined dryer vent (stream 37 in Figure 3) is a contin-
uous source of emissions of carbon black that is entrained
by the gas stream fed directly to the interior of the dryer.
Carbon black emissions are controlled by means of bag filters
or scrubbers. Other contaminants in the dryer vent gas
result from use of the main process vent gas as a supple-
mentary fuel and from the combustion of impurities in the
natural gas fuel for the dryer. These contaminants include
sulfur oxides, nitrogen oxides, and the unburned portion of
each of the species present in the main process vent gas
(see Table 26).
Table 27 summarizes the emissions data obtained from the
survey of operating plants given in Reference 21 and from
the public files of the Louisiana Air Control Commission.
The mean emission factors are 0.17 g/kg for carbon black,
0.23 g/kg for sulfur oxides, and 0.61 g/kg for nitrogen
oxides. A mean emission factor for carbon black was also
computed considering only those units employing bag filters;
it was found to be 0.12 g/kg. The error bounds given for the
mean values in Table 27 represent 95% confidence limits for
the means.
74
-------
Table 26. MAIN PROCESS VENT GAS COMPOSITION FOR A TYPICAL
CARBON BLACK PLANT
Production rate = 5.1 x 101* metric tons/yr
Temperature = 220°C
Gage pressure = 0 kPa
Component
Nitrogen and Argon
Water
Carbon monoxide
Carbon dioxide
Oxygen
Hydrogen
Acetylene
Carbon disulfide
Hydrogen sulfide
Methane
Carbonyl sulfide
Ethylene
n-Butane
Nitrogen oxides
Trace elements
Propane
Carbon black
POM
Iso-Butane
Ethane
Propylene
n-Pentane
Sulfur oxides
TOTAL
Concentration,
weight %
51.15
34.10
7.95
3.97
1.35
0.67
0.25
0.17
0.17
0.14
0.05
0.0096
0.0015
0.0015
<0.0014
0.0012
0.0006
0.00001
0.0006
o.oa
o.oa
o.oa
o.oa
100.0
Emission rate ,
kg/hr
52,700
35,100
8,200
4,100
1,400
700
260
175
175
145
60
9
1.6
1.6
<1.5
1.2
0.6
0.012
0.6
o.oa
o.oa
o.oa
o.oa
103,000
Emission
factor, g/kg
9,000
6,000
1,400
700
240
120
45
30
30
25
10
1.6
0.27
0.28
<0.25
0.23
0.11
0.002
0.10
o.oa
o.oa
o.oa
o.oa
17,670
Not detected at detection limit of 1 ppm.
75
-------
Table 27. EMISSIONS DATA FOR COMBINED DRYER VENT21
Plant code
number21
50-1
50-4
50-6
50-7
50-8
50-9
50-10
50-11
50-19
50-24
50-25
50-26
50-27
50-28
_b
Mean
Range
Mean for units
bag filters
Control
device
Bag filter
None
Bag filter
Bag filter
Bag filter
Bag filter
None
Scrubber
Bag filter
Bag filter
Bag filter
Bag filter
Bag filter
Scrubber
Bag filter
Bag filter
with
Carbon black
0.06
0.05
0.02
0.40
0.03
0.30
0.40
0.01
0.01
0.10
N.A.
0.02
N.A.
0.70
0.10
N.A.
0.17 ± 77%
0.01 to 0.70
0.12 ± 89%
Emission factors,
SO
X
N.A.3
N.A.
0.04
0.03
0.10
N.A.
0.05
N.A.
N.A.
0.20
0.20
0.50
N.A.
0.20
0.54
0.45
0.23 ± 60%
0.03 to 0.54
g/kg
NO
X
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
Trace
1.10
0.12
0.61
0.61 + 200%
- 100%
trace to 1.10
Not available.
Data obtained from public files of Louisiana Air Control Commission.
3. Pneumatic System Vent
The pneumatic system vent gas (stream 19 in Figure 3) is a
continuous source of carbon black emissions. The pneumatic
system conveys the carbon black from the product recovery
bag filters to the product treatment area of the plant. The
76
-------
transported carbon black is removed from the entraining air
by means of a cyclone and/or a bag filter. In the latter
case the filtered air stream is vented to the atmosphere.
Residual carbon black is the only contaminant present in
this stream.
Table 28 summarizes the emissions data contained in the sur-
vey of carbon black plants reported in Reference 21. The
mean carbon black emission factor for the six plants supply-
ing data is 0.29 g/kg. The values range from 0.06 g/kg to
0.70 g/kg. The 95% confidence limits about the mean are
±93%. , ,
Table 28. EMISSIONS DATA FOR PNEUMATIC SYSTEM VENT21
Plant code
number
50-6
50-7
50-9
50-10
50-29
50-30
Control
device
Bag filter
Bag filter
Bag filter
Bag filter
Bag filter
Bag filter
Carbon black
emission factor, g/kg
0.70
0.24
0.17
0.50
0.06
0.07
Mean
Range
0.29 ± 93%
0.06 to 0.70
4. Oil Storage Tank Vents
Estimates of hydrocarbon emissions from the oil feedstock
storage tank vents (stream 39 in Figure 3) for a 4.3 x 104
metric tons/yr carbon black plant are presented in Table 29.
These estimates were obtained using the data in Tables 13 and
17 together with the empirical correlations for petrochemical
77
-------
Table 29. OIL STORAGE TANK WORKING AND BREATHING LOSSES
FROM A 4.3 X 10^ METRIC TONS/YR CARBON BLACK PLANT
Vapor pressure,
!kPa
0.7
1.4
2.8
Emission rate,
kg/hr
3.6
6oO
10.1
Emission factor,
g/kg
0.72
1.21
2.06
losses from storage tanks which have been developed by the
American Petroleum Institute.28"32 Values were calculated
for three different oil vapor pressures in order to cover
the ra.nge of values expected for different feedstocks and
storage temperatures. The estimated accuracy of the empiri-
cal correlations is ±10%.28~32
The oil storage tanks are vented directly to the atmosphere
in all carbon black plants. Control could be achieved by
means of conservation vents or floating roof tanks. However,
all carbon black feedstock storage tanks meet EPA New Source
Performance Standards for these emissions points (personal
communication, C. B. Beck, Cabot Corporation, Pampa, Texas,
26 February 1976).
28Evaporation Loss from Fixed Roof Tanks. Bulletin 2518,
American Petroleum Institute, New York, New York, 1962. 38 pp,
29Use of Variable Vapor Space Systems to Reduce Evaporation
Loss. Bulletin 2520, American Petroleum Institute, New York,
New York, 1964. 14 pp.
30Petrochemical Evaporation Loss from Storage Tanks. Bulletin
2523, American Petroleum Institute, New York, New York,
1969. 14 pp.
3Evaporation Loss from Floating Roof Tanks. Bulletin 2517,
American Petroleum Institute, New York, New York, 1962.
13 pp.
3 Evaporation Loss in the Petroleum Industry - Causes and
Control. Bulletin 2513, American Petroleum Institute,
New York, New York, 1959. 57 pp.
78
-------
5. Vacuum Cleanup System Vent
The vacuum cleanup system vent (stream 36 in Figure 3) is a
semicontinuous source of carbon black emissions. This system
is used to pick up loose carbon black resulting from spills
and leaks, and may also be used to control emissions from
storage bins and the bagging operation. Before being vented
to the atmosphere, the exhaust air from the vacuum system
passes through a cyclone and/or bag filter that removes en-
trained carbon black (as shown in Figure 8).
Table 30 summarizes the emissions data for product storage
and handling operations obtained from the survey of carbon
black plants given in Reference 21. Under the assumption
that the vacuum cleanup system is employed to control these
operations, the data can be equated with emission factors
for the vacuum cleanup system vent. The mean emission
Table 30. EMISSIONS DATA FOR PRODUCT STORAGE AND HANDLING21.
Plant code
number
50-1
50-2
50-5
50-8
50-9
50-10
50-28
Control
device
Bag filter
Bag filter
Bag filter
Bag filter
Bag filter
Bag filter
Bag filter
Carbon black
emission factor, g/kg
0.01
0.035
0.02
0.01
0.045
0.02
0.05
Mean
Range
0.03 ± 58%
0.01 to 0.05
It is assumed that the additional emissions from this vent
as a result of intermittent vacuum cleanup spills and leaks
are small compared with those resulting from control of pro-
duct storage and handling operations. No data are available
for the former emissions.
79
-------
factor for carbon black, the only contaminant in this stream,
is 0.03 g/kg. The seven reported values range from 0.01 g/kg
to 0.05 g/kg. The 95% confidence limits about the mean are
±58%.
6. Fugitive Emissions
The extremely small particle size and fluffiness of loose
carbon black, especially prior to pelletizing, gives rise to
numerous sources of fugitive emissions. These include the
following:7
• Cleaning of clogged screens, production lines, and
other process equipment.
• Leaks in the pneumatic system, storage bins, bagging
system, and other process equipment. The salty
atmosphere of the Gulf Coast area results in severe
corrosion problems for the carbon black plants
located there. Leaks are also caused by the erosion
of conveying equipment by carbon black. They may
also result from defective seals on conveying equip-
ment and defective buckets in the elevator.
• Drawing of samples from various points in the
production line for testing and quality control.
• Spillage from torn and broken bags during stacking
in the warehouse or during loading in box cars.
• Spillage during the loading of hopper cars when the
flexible loading pipe is shifted from one compartment
to another. Dusting of black also occurs when the
loading pipe is inserted into the top of a Sealdbin®*
• Cleaning and sweeping the bottoms of hopper cars.
• Cleaning of process equipment prior to repair work.
• Opening of the rotary valve at the base of the filter,
in the event of a fire in a bag filter, to permit re-
moval of the burning and contaminated black. This oper-
ation results in atmospheric emissions of carbon black.
Fugitive losses were estimated in Reference 21 to be 0.5 kg/hr
for a 4.0 x 10"* metric tons/yr carbon black plant. The cor-
responding emission factor is 0.1 g/kg.
80
-------
C. ENVIRONMENTAL EFFECTS
1. Definition of a Representative Source
For the purpose of assessing the source severity, a repre-
sentative carbon black plant is defined to be one producing
5.1 x 10** metric tons/yr of an HAF grade black by the oil
furnace process. Although carbon black plants normally pro-
duce several different grades of black, the HAF blacks are
predominant, accounting for nearly 50% of oil furnace black
production (see Figure 22 in Section VI.C.5). In addition,
the HAF blacks are intermediate in properties between the
smaller particle blacks (SAF, ISAF) and the larger particle
blacks (FEF, GPF).
The production rate of 5.1 x 104 metric tons/yr is the arith-
metic mean value obtained by dividing the total 1974 production
(1.54 x 106 metric tons) by the number of plants (30). In
addition, it is assumed that the representative plant is
equipped with bag filters on the main process vent, the
dryer vent, the pneumatic system vent, and the vacuum clean-
up system vent, since this mode of operation predominates in
the industry (see Section V.A.). The oil storage tank vents,
as well as fugitive emissions, are uncontrolled in all plants.
2. Emission Factors
Emission factors for a representative carbon black plant are
presented in Table 31. The values for the main process vent
are the measurements made by MRC in the present study. The
values for POM and trace elements are totals of the values
listed in Tables 24 and 25 for individual POM's and trace
elements, respectively. As previously noted, these emission
factors correspond to a plant which does not use the process
off-gas as fuel in the dryers.
81
-------
Table 31. EMISSION FACTORS FOR A REPRESENTATIVE CARBON BLACK PLANT
CO
Emission factor,3 g/kg
Material emitted
Criteria pollutants
Particulate matter
Sulfur oxides
Nitrogen oxides
J
Hydrocarbons
Carbon monoxide
Hazardous materials
Beryllium
Mercury
Inert gases
Nitrogen & argon
Oxygen
Carbon dioxide
Chemical substances
Hydrogen
Hydrogen sulfide
Carbon disulfide
Carbonyl sulfide
Carbon black
Methane
Acetylene
Ethane
Ethylene
Propylene
Propane
Isobutane
n-Butane
n-Pentane f
POM (total) T
Trace elements
(total) 9
Main
process
vent
0.11
_c
0.28
50
1,400
2.2 x 10-6
1.5 X lO"1*
9,000
240
700
120
30
30
10
0.11
25
45
N.A.
1-6 _c
0.23
0.10
0.27
N.A.
0.002
<0.25
Dryer
vent
± 70% 0.-12 i 39%
0.23 ± 60%
± 15% 0.60 + 200%
- 100%
± 48%
± 19%
16%
+ 386%
- 100%
± 19% 2,200 ± 21%
+ 99% 400 ± 65%
± 21%
± 39%
* 82
± 76
i 99%
± 70% 0.12 ± 89%
± 47%
± 48%
± 85%
+ 109%
- 100%
± 80%
* 57%
4 52%
± 43
Pneumatic Oil Vacuum
system storage cleanup Fugitive
vent tanks system vent emissions
0.3 ± 9^ 0.03 ± 58% 0.1
0.72 ± 10%
N.A.6
N.A.
0.3 ± 93% 0.03 * 58% 0.1
aBlanks indicate no emissions.
The particulate matter is carbon black.
°Not detected at detection limit of 1 ppm.
Total nonmethane hhdrocarbons. The individual hydrocarbon species are listed under chemical substances.
eNot available.
Emission factors for individual POM species are given in Table 24, Section IV.B.
"includes beryllium and mercury. Emission factors for individual trace elements are given in Table 25,
Section IV.B.
-------
Emission factors given in Table 31 for the dryer vent, the
pneumatic system vent, and the vacuum cleanup system vent are
mean values of the data given in Tables 27, 28, and 30. The
mean value for units with bag filters was used for carbon
black emissions from the dryer vent. The emission factor
for fugitive emissions is based on the estimate of fugitive
emissions given in Reference 21.
The hydrocarbon emission factor for oil storage tanks is
that calculated assuming an oil vapor pressure of 0.7 kPa
(see Section IV.B.4).
The error bounds given in Table 31 represent 95% confidence
limits for the mean emission factors. In the case of fugitive
emissions, insufficient information was available to deter-
mine either the accuracy or precision of the estimated emis-
sion factor. Hence, no error bounds have been assigned to
this value.
Emission factors for carbon black plants are independent of
plant size since these plants are made up of separate pro-
duction units, each having multiple reactors which do not
vary greatly in size. The emission factors for all sources
except the main process vent should also be nearly independent
of the type of black produced, since the operating conditions
involved do not change with black type. Although carbon black
emissions would be expected to increase to some extent with
decreasing particle size, such differences are probably small
compared with the scatter in emissions data.
3. Source Severity
In order to obtain a quantitative measure of the hazard
potential of carbon black production, the source severity, S,
is defined as:
83
-------
S =
where v is the maximum time-averaged ground level concen-
max
tration of each pollutant emitted from a representative
plant, and F is defined as a primary ambient air quality
standard for criteria pollutants (particulate, SO , NO , CO,
X X
and hydrocarbons), while for noncriteria pollutants,
F = TLV • 8/24 • 0.01
The factor 8/24 adjusts the TLV for continuous rather than
workday exposure, and the factor of 0.01 accounts for the
fact that the general population is a higher risk group than
healthy workers.
Thus, the source severity represents the ratio of the maxi-
mum mean ground level concentration of a given pollutant to
the concentration which constitutes an incipient health hazard.
The maximum ground level concentration, x i is calculated
max
according to Gaussian plume dispersion theory:33
= -2-2- (2)
where Q = mass emission rate, g/s
TT = 3.14
H = effective emission height, m
e = 2.72
u = average wind speed, m/s
33Turner, D. B. Workbook of Atmospheric Dispersion Estimates
Public Health Service Publication No. 99-AP-26, U.S. Depart
ment of Health, Education, and Welfare, Cincinnati, Ohio,
1969. 84 pp.
84
-------
Equation 2 yields a value corresponding to an averaging
(i.e., sampling) time of approximately 3 minutes.33' 3lt For
a continuously emitting source, the maximum time-averaged
ground level concentration for averaging times between
3 min and 24 hrs can be estimated from the relation:33'31*
Amax Amax \t /
where t = averaging time
t = short-term averaging time (3 min)
In this study, the averaging time, t, for noncriteria pollu-
tants is 24 hours. For criteria pollutants, the averaging
times are those used in the definition of the primary ambient
air quality standards. The only exception is NO , for which
the primary standard averaging time is 1 year. Since Equa-
tion 3 is not valid for averaging times of this magnitude,
the calculation of x for NO is based on Equation 5.13
max x
of Reference 33 which estimates the annual average ground
level concentration.
Insertion of the national average wind speed of 4.5 m/s into
the above equations leads to the source severity equations
listed in Table 32. The emission heights used in the calcu-
lations are given in Table 33. The heights for the oil
storage tanks and fugitive emissions were assumed. The
others were obtained by averaging the operating plant data
given in Reference 21. The emission rates, Q, were calcu-
lated using the emission factors listed in Tables 24, 25, and
3t*Nonhebel, G. Recommendations on Heights for New Indus-
trial Chimneys. Journal of the Institute of Fuel.
33:479-511, 1960.
85
-------
Table 32. SOURCE SEVERITY FACTOR EQUATIONS
Pollutant
Particulate matter
S0x
N0x
Hydrocarbons
Carbon monoxide
All others
S, dimensionless
70 QH~2
50 QH~2
315 OH"2-1
162.5 OH"2
0.78 QH-2
5.5 OH~2 (TLV)"1
Q = emission rate, g/s
H = effective emission height, m
TLV = threshold limit value, g/m3
Table 33. EMISSION HEIGHTS FOR REPRESENTATIVE SOURCE
Source of emissions
Main process vent
Dryer vent
Pneumatic system vent
Oil storage tanks
Vacuum cleanup system vent
Fugitive emissions
Emission height, m
25
20
25
15
20
3.1
31. The resulting values of the maximum time-averaged
ground level concentration, x" , are presented in Table 34,
max
while the source severities are given in Tables 35, 36, and
37. In arriving at these results, it was assumed that
Gaussian plume dispersion theory was equally valid for all
emissions, irrespective of their chemical, physical, or
topological characteristics. In addition, plume rise was
86
-------
Table 34.
MAXIMUM TIME-AVERAGED GROUND LEVEL CONCENTRATION FOR A
REPRESENTATIVE CARBON BLACK PLANT
Material emitted
Criteria pollutants
Particulate matter''
Sulfur oxides
Nitrogen oxides
Hydrocarbons"
Carbon monoxide
Hazardous materials
Beryllium
Mercury
Chemical substances
Hydrogen sulfide
Carbon disulfide
Carbonyl sulfide
Carbon black
Methane
Acetylene
Ethylene
Ethane
Propylene
Propane
Isobutane
n-Butane
n-Pentane
POM (total)
Trace elements6
(total)
"max'** "g/m3
Main
process
vent
5.2
0.0C
13.2
2,360
66,110
1.7 x 10-1*
7.1 x ID'3
990
1,510
410
5.2
1,180
2,220
76
o.oc
0.0C
9.4
4.7
12.7
o.oc
0.094
<11.8
Dryer
vent
8.8
16.9
44.9
8.8
Pneumatic
system
vent
13.7
13.7
Oil
storage
tanks
9.2
Vacuum
cleanup
system vent
2.2
2.2
•
Fugitive
emissions
307
307
Blanks indicate no emissions. Averaging times for criteria pollutants are those specified in the
primary ambient air quality standards. For noncriteria pollutants, the averaging time is 24 hours.
b
The particulate matter is carbon black.
Not detected in process vent gas at detection limit of 1 ppm.
Total nonmethane hydrocarbons; individual hydrocarbon species are listed under chemical substances.
P
Includes beryllium and mercury.
87
-------
Table 35. SOURCE SEVERITIES FOR A REPRESENTATIVE CARBON BLACK PLANT
Material emitted
Criteria pollutants
Particulate matter"
Sulfur oxides
Nitrogen oxides
Hydrocarbons^
Carbon monoxide
Hazardous materials
Beryllium
Mercury
Chemical substances
Hydrogen imlfide
Carbon dinulfide
Carbonyl siulfide
Methane
Acetylene
Ethylene
Ethane
Propylene
Propane
Isobutane
n- Butane
n-Pentane -
Carbon ble.ck
POM (total) 9
Trace elen.entsQ''1
(total)
Source severity"
Main
process
vent
0.02
0.0C
0.16
21
1.8
0.028
0.042
20
7.6
2.1
_e
. _e
_e
_e
o.oc
_e
0.001
0.003
o.oc
0.45
2.1
<2.7
Dryer
vent
0.034
0.046
0.58
0.76
Pneumatic
system
vent
0.053
1.2
Oil
storage
tanks
0.84
Vacuum
cleanup
system vent
0.008
0.34
Fugitive
emissions
1.2
27
Calculated assuming no plume rise. If plume rise is included in the calculations, severities of
emissions from the main process vent are reduced by a factor of 0.2; severities of emissions from
the dryer vent are reduced by a factor of 0.6; severities of other emissions are unchanged. Blanks
indicate no emissions.
The particulate matter is carbon black. Source severities were computed using the severity equation
for particulate matter.
Not detected at detection limit of 1 ppm.
Total nonmethane hydrocarbon severities were calculated using the severity equation for hydrocarbons.
@
Classified as "inert" gas, i.e., simple asphyxiant. TLV has not been established.
Severities calculated using TLV of 0.0035 g/m3 for carbon black.
"source severities for individual POM's and trace elements are given in Tables 36 and 37.
Includes beryllium and mercury.
88
-------
Table 36. SOURCE SEVERITIES FOR POLYCYCLIC
ORGANIC COMPOUNDS IN MAIN PROCESS VENT GAS
Compound
Acenaphthylene
Anthracene/phenanthrene
Benzo [c] phenanthrene
Benzof luoroanthenes
Benzo [ghi] f luoranthene
Benzo [ghi] perylene/anthanthrene
Benzopyrenes & perylene
Chrysene/benzo [a] anthracene
Dibenzanthracenes
Dibenzo [c,g] carbazole
Dibenzopyrenes
Dibenzothiophene
Dimethylanthracenes/phenanthrenes
7 , 12-Dimethylbenz [a] anthracene
Fluoranthene
Indeno [1,2, 3-cd] pyrene
Methylanthracenes/phenanthrenes
Methylcholanthrene
Methylf luoranthene/pyrene
Pyrene
Carcinogenicity
indicator3
-
-
+++
++
-
-
+++
±
+
+++
+++
-
-
++++
-
+
-
++++
-
—
Source
severity^
0.057
0.005
<0.028
0.43
0.0028
0.0016
0.42
0.00062
<0.028
<0.028
<0.028
0.001
0.01
0.99
0.0042
<0.028
0.0071
<0.028
0.0016
0.035
aThe carcinogenicity indicators are defined as follows: 35
- not carcinogenic
± uncertain or weakly carcinogenic
+ carcinogenic
++, +++, ++++ strongly carcinogenic
Calculated assuming no plume rise. If plume rise is in-
cluded in the calculation, the severities are reduced by
a factor of 0.2.
35Particulate Polycyclic Organic Matter. National Academy
of Sciences, Washington, B.C., 1972. 361 pp.
89
-------
Table 37. SOURCE SEVERITIES FOR TRACE
ELEMENTS IN MAIN PROCESS VENT GAS
Element
Source Severity3
Aluminum
Antimony
Arsenic
Barium
Beryllium
Bismuth
Boron
Bromine
Cadmium
Calcium
Cerium
Cesium
Chlorine
Chromium
Cobalt
Copper
Dysprosium
Erbium
Europium
Fluorine
Gadolinium
Gallium
Germanium
Gold
Hafnium
Holmium
Iodine
Iridium
Iron
Lanthanum
Lead
Lithium
Lutecium
Magnesium
Manganese
Mercury
Molybdenum
Neodymium
Nickel
Niobium
Osmium
Palladium
Phosphorus
Platinum
Indeterminate^
0.00071
0.0010
<0.063C
<0.028
Indeterminate"
Indeterminate'3
0.00018
0.0040
Indeterminate'5
Indeterminate'5
Indeterminate'3
<0.028C
<0.01lC
0.016
<0.7lC
Indeterminate^3
Indeterminate^
Indeterminate"
<0.014C
Indeterminate^
Indeterminate^
Indeterminate^
Indeterminate"
<0.00045
Indeterminate b
<0.000071
Indeterminate!3
Indeterminate"
Indeterminate*3
<0.14C
Indeterminate^
Indeterminate.
Indeterminate
<0.00037C
0.042
<0.00039 b
Indeterminate
0.18 b
Indeterminate.
Indeterminate.
Indeterminate
<1.4C
<0.053
90
-------
Table 37 (continued). SOURCE SEVERITIES
FOR TRACE ELEMENTS IN MAIN PROCESS VENT GAS
Element
Potassium
Praseodymium
Rhodium
Rubidium
Samarium
Scandium
Selenium
Silicon
Sodium
Strontium
Tantalum
Terbium
Tellurium
Thallium
Thorium
Thulium
Tin
Titanium
Tungsten
Uranium
Vanadium
Ytterbium
Yttrium
Zinc
Zirconium
Source Severity
Indeterminate!*
Indeterminate
<0. 00028 b
Indeterminate.
Indeterminateb
Indeterminate
<0. 00011
<0.028C b
Indeterminate.
Indeterminateb
Indeterminate.
Indeterminate
<0. 00042
-------
not taken into account in the calculations, i.e., the effec-
tive emission height was equated with the physical stack
height. It is shown in Appendix C that if plume rise is
taken into account, the values of xm= and source severity
lUclX •
for the main process vent gas are decreased by a multiplica-
tive factor of 0.2, while those for the dryer vent are de-
creased by a multiplicative factor of 0.6. The values for
the other emission points are unchanged.
As indicated in Table 35, source severities for carbon black
were computed using both the primary ambient air quality
standard for particulate matter and the TLV of 0.0035 g/m3
for carbon black. Since carbon black is not classified as
a toxic substance10, it should properly be assigned the TLV
of 0.0.1 g/m3 for "nuisance" particles. However, the lower
TLV is being retained for "housekeeping" purposes by the
American Conference of Governmental Industrial Hygienists. 1 °
If the TLV for nuisance particles were used, the source
severities listed in Table 35 under "carbon black" would be
decreased by a multiplicative factor of 0.35. From a human
health standpoint, the severities listed in Table 35 for
particulate matter are more appropriate than those based on
the TLV for either carbon black or nuisance particles. The
latter two values are derived from considerations other than
health effects.
Extensive small animal and laboratory tests with carbon
black have demonstrated that polycyclic organic material
absorbed on the black is not released in vivo and does not
represent a health hazard. 9' 36~it0 Hence, it is important
36Neal, J., M. Thornton, and C. A. Nau. Polycyclic Hydro-
carbon Elution from Carbon Black or Rubber Products.
Archives of Environmental Health, 4:46-54, 1962.
(continued)
92
-------
to determine how much of the total POM emitted from the main
process vent is associated with the carbon black in the vent
gas. Using the data obtained in this study on the POM con-
tent of carbon black (see Table 6, Section III.A), it is
found that less than 1% of the POM in the vent gas is adsorbed
on the carbon black.3 Thus, essentially all of the POM is
emitted in vapor form which is capable of being absorbed by
living tissue. A similar calculation yields an identical
result for trace elements.
From Table 36, it is seen that only three POM's have severi-
ties exceeding 0.1: 7,12-dimethylbenz[a]anthracene, benzo-
pyrenes/perylene, and benzofluoranthenes, all of which are
For example, consider pyrene. From Table 24, the total emis-
sion factor for pyrene is 5.0 x IQ~k g/kg. From Table 6, the
concentration in N-330 black is 27 ppm, or 27 x 10~6 g pyrene/
g carbon black. The emission factor for carbon black is
0.11 g/kg from Table 31. Hence, the pyrene emitted with the
carbon black is given by
0.11 g black emitted 6 g pyrene = -6 /
kg product g black y/ *
The percentage of total pyrene emissions which is adsorbed on
carbon black is thus
3.0 x 10-6 _ fi
5.0 x ID'1* - °'6%
37Nau, C. A., J. Neal, and V. Stembridge. A Study of the
Physiological Effects of Carbon Black. I: Ingestion.
Archives of Industrial Hygiene, 17:21-28, 1958.
38Nau, C. A., J. Neal, and V. Stembridge. A Study of the
Physiological Effects of Carbon Black. II: Skin Contact.
Archives of Industrial Hygiene, 18:511-520, 1958.
39Nau, C. A., J. Neal, and V. Stembridge. A Study of the
Physiological Effects of Carbon Black. Ill: Absorption
and Elution Potentials. Subcutaneous Injections. Archives
of Industrial Hygiene, 20:512-533, 1960.
lt°Nau, C. A., J. Neal, V. Stembridge, and R. N. Cooley.
Physiological Effects of Carbon Black. IV: Inhalation.
Archives of Environmental Health, 44:15-431, 1962.
93
-------
strongly carcinogenic. However, if plume rise were taken
into account, the severities of the latter two species would
be less than 0.1.
From Table 37, it can be seen that for most trace elements
the source severity is either indeterminant (because no TLV
has bean established) or is determined only to within an
upper bound due to limitations of the sampling and analytical
procedures. The severity of only one element, nickel, defi-
nitely exceeds 0.1. The severities of copper, lead, and
phosphorus may exceed 0.1 since the upper bounds determined
for these elements are greater than 0.1. If plume rise were
taken into account, the severities of both nickel and lead
would be less than 0.1.
All four of the above elements were detected in significant
amounts in either the oil feedstock or the process quench
water, as shown in Table 38. Clearly, copper, lead, and
nickel originate from the feedstock, while phosphorus origi-
nates from both the feedstock and the quench water.
Table 38. CONCENTRATIONS OF HIGH SEVERITY TRACE ELEMENTS
IN FEEDSTOCK AND QUENCH WATER
Element
Copper
Lead
Nickel
Phosphorus
Source
severity
<0.71
<0.14
0.18
<1.4
Concentration in
feedstock, ppm
1.6
0.30
5.5
4.3
Concentration in
quench water, ppm
0.017
0.001
<0.001
3.0
4. Source Severity Distributions
Source severity distributions are presented in Figures 10
through 13 for carbon monoxide, hydrogen sulfide, sulfur
94
-------
CO BOILER AND
INCINERATOR
SflMPLE SIZE = 30
MIN. VflLUE '= 0-005
MflX. VflLUE = 10.044
MERN = 2.764
STD. DEV. = 2.101
0.100 1.000
SOURCE SEVERITY FOR CO
Figure 10. Deterministic source severity distribution
for carbon monoxide emissions from main process vent
9
a?
, CO BOILER AND
( INCINERATOR
SflMPLE SIZE = 30
MIN. VflLUE = 0.022
MFIX. VflLUE = 55.475
MEflN = 13.817
STD. DEV. = 11.249
1.000 10.000
SOURCE SEVERITY FOR H2S
Figure 11. Deterministic source severity distribution
for hydrogen sulfide emissions from main process vent
95
-------
SflMPLE SIZE = 30
HIM. VflLUE = 0.028
MflX. VflLUE - 3-972
HERN = 0.742
STD. OEV. = 0.966
0.100 1.000
SOURCE SEVERITY FOR S02
Figure 12.. Deterministic source severity distribution
for sulfur oxide emissions from main process vent
SflMPLE SIZE = 30
HIM. VflLUE = 0.003
MflX. VflLUE = 2.466
MEflN = 0.438
STD. OEV. = 0.480
0.010 0.100 1.000
SOURCE SEVERITY FOR PflRTICULRTE
Figure 13. Deterministic source severity distribution
for particulate emissions from main process vent
96
-------
oxides and particulate emissions from the main process vent.
These distributions are based on data from the survey of
operating plants given in Reference 21. Simulated source
severity distributions for other emissions (for which data
were not given in Reference 21) are presented in Appendix G.
The ordinate in Figures 10 through 13, labeled "cumulative
frequency," should be interpreted as the percentage of carbon
black plants having source severities less than the corres-
ponding value on the abscissa. For example, from Figure 10,
approximately 20% of carbon black plants have carbon monoxide
severities less than 1.0. The solid curves in the figures
were drawn by fitting a Weibull frequency distribution to the
data points.**1
The points corresponding to the three plants which burn the
main process vent gas are indicated on the graphs. The
following assumptions were made in calculating these points:
For the carbon monoxide boiler-thermal incinerator
combination, combustion efficiencies of 99% for
carbon monoxide and hydrogen sulfide, and 95% for
particulate matter, were assumed.
For the flares, 90% combustion efficiency for
carbon monoxide and hydrogen sulfide, and 20%
for particulate matter were assumed as in
Reference 21.
The particulate severities were computed using
the primary ambient air quality standard for
particulate matter (as opposed to the TLV for
carbon black or for nuisance particles).
^Eimutis, E. C., B. J. Holmes, and L. B. Mote. Source
Assessment: Severity of Stationary Air Pollution Sources
A Simulation Approach. EPA-600/2-76-032e. U.S. Environ-
mental Protection Agency, Research Triangle Park, North
Carolina, July 1976. 119 pp.
97
-------
In all other cases, missing data were equated with the mean
values of the existing data.
It should be noted that field measurements made in the present
study indicate a carbon monoxide combustion efficiency in
excess of 99.9% in the flare. This value would yield carbon
monoxide severities two orders of magnitude lower than those
given in Figure 10 for the plants which flare the vent gas.
It will be noted that the mean severities given in Figures
10 through 13 are not identical to the severities for a
representative plant listed in Table 35. There are two
reasons for the differences: (1) the mean emission factors
for the data given in Reference 21 are not identical to the
emission factors for the representative plant (Table 31),
and (2) source severity is a nonlinear function of the two
variables/ emission rate and emission height, both of which
vary from plant to plant. Therefore, the mean severity is
not equal to the severity calculated using mean emission
rate and mean emission height.
5. National and State Emissions Burdens
The mass emissions of criteria pollutants (particulate, SO ,
X
NO , carbon monoxide, and hydrocarbons) resulting from
J\.
carbon black production were calculated using the emission
factors from Table 31 and the production data from Table 21.
The appropriate emission factor was multiplied by the annual
production nationwide and by that for the states of Texas
and Louisiana, in which the carbon black plants are concen-
trated. The total annual mass emissions of criteria pollu-
tants from all sources nationwide and from sources within
the states of Texas and Louisiana were obtained from
98
-------
Reference 42. The percentages of the total emissions re-
sulting from carbon black manufacture were computed using
these values and are presented in Table 39.
Table 39. CONTRIBUTION TO TOTAL EMISSIONS OF
CRITERIA POLLUTANTS BY CARBON BLACK INDUSTRY
Material emitted
Particulate matter
Sulfur oxides
Nitrogen oxides
Hydrocarbons
Carbon monoxide
Industry contribution, %
Nationwide
0.01
0.001
0.008
0.31
2.3
Texas
0.08
0.02
0.04
1.5
13
Louisiana
0.09
0.07
0.12
1.4
13
6. Affected Population
A measure of the population which is exposed to a high con-
taminant concentration due to a representative carbon black
plant can be obtained as follows. The values of x for which
X(x) _
F ~
(4)
for
m = 0.1 and 1.0
are determined by iteration. The value of xM , the annual
mean ground level concentration, is computed from the
equation:3 3
421972 National Emissions Report. EPA-450/2-74-012, U.S.
Environmental Protection Agency, Research Triangle Park,
North Carolina, June 1974. 422 pp.
99
-------
X(x) = ^^exp -iff- (5)
a ux
z
where Q = emission rate, g/s
H = effective emission height, m
x = downwind distance from source, m
u = average wind speed, 4.5 m/s
j = vertical dispersion coefficient, m
Z
For atmospheric stability class C (neutral conditions), a
Z
is give:n by: ^
az = 0.113 (x°-911) (6)
The affected area, A(km2), is then computed as
A = TT(X22 - Xl2) (7)
where xi and x2 are the two roots of Equation 4 for a given
value of m.
The capacity-weighted mean population density, D , is
calculated as follows:
Dp = ^£—- , persons/km2 (8)
. i
where C. = production capacity of plant i
Dp = county population density for plant i
The product A*Dp is designated the "affected population."
The affected population was computed for each compound and
each emission point for which the source severity for a
^Eimutis, E. C., and M. G. Konicek. Derivations of Continu-
ous Functions for the Lateral and Vertical Atmospheric
Dispersion Coefficients. Atmospheric Environment, 6:859-
863, 1972.
100
-------
representative plant exceeds 0.1. The results are presented
in Table 40.
The mean population density (from Equation 8) for carbon black
plants is 57 persons per square kilometer. The largest value
of affected population results from hydrogen sulfide emis-
sions from the main process vent: 11,000 persons where
x"/F > 0.1 and 1,000 persons where \/F Z 1.0.
7. Growth Factor
In 1974, 1.54 x 106 metric tons of carbon black was produced
in the United States.26 As discussed in Section VI, pro-
duction in 1980 is expected to total from 1.73 x 106 to
1.79 x 10 6 metric tons. Thus, taking an average value of
1.76 x 106 metric tons for 1980 production, and assuming
that the same level of control technology exists in 1980 as
existed in 1974, the emissions from the carbon black industry
will increase by 14% over that period, i.e.:
Emissions in 1980 _ 1.76 x = ! 14
Emissions in 1974 1.54 x 10b
101
-------
Table 40. AFFECTED POPULATION
Emitted species
Emission point
No. of persons
where x/F ^ 1.0
No. of persons
whe re )T/F £ 0.1
Particuiate matter
Nitrogen oxides
Nitrogen oxides
Hydrocarbons
Hydrocarbons
Carbon monoxide
Hydrogen sulfide
Carbon disulfide
Carbonyl sulfide
Carbon black
Carbon black
Carbon black
Carbon black
Carbon black
Lead
Nickel
Copper
Phosphorous
Benzofluoroanthenes
Benzopyrenes
7,12-Dimethylbenz[a]anthracene
Fugitive emissions
Main process vent
Dryer vent
Main process vent
Storage tanks
Main process vent
Main process vent
Main process vent
Main process vent
Main process vent
Dryer vent
Pneumatic system vent
Vacuum cleanup vent
Fugitive emissions
Main process vent
Main process vent
Main process vent
Main process vent
Main process vent
Main process vent
Main process vent
0
0
0
700
0
50
1,000
300
70
0
0
20
0
20
0
0
0
40
0
0
20
10
40
100
8,000
80
800
11,000
4,000
1,000
200
200
500
40
170
40
60
300
700
200
200
500
-------
SECTION V
CONTROL TECHNOLOGY
A. STATE OF THE ART
1. Primary Recovery Device
The application of emissions control technology in furnace
black plants is summarized in Table 41. It can be seen that
all but four of the 30 plants covered in the Houdry survey21
employ bag filters on all process trains to recover carbon
black from the reactor effluent. Electrostatic precipitators
and cyclones, which predated the use of bag filters, are
still in operation in some plants as supplementary recovery
devices. However, the trend is toward the use of bag filters
as the sole means of product recovery.
Bag filters in carbon black plants reportedly have recovery
efficiencies of up to 99.95%;21 99.99% recovery was measured
in the field sampling program conducted in the present study.
Since their principal function is product recovery rather
than emissions control, bag filters were discussed in detail
in Section III.
Four plants still employ scrubbers on at least one process
train. These units have efficiencies of only 90% to 95%, and
for this reason they are being phased out of product recovery
service.2
103
-------
Table 41. SUMMARY OF EMISSION CONTROL DEVICES IN
FURNACE BLACK PLANTS21
Control device type
Number of
plants
one or
Primary recovery device
Bag filter only - all process trains
Cyclone & bag filter
Electrostatic precipitator & cyclone)
& bag filter \ D?ocess
Electrostatic precipitator & cyclone) £
_ i_ *_
& scrubber
Combustion device for process vent gas
None
Flare
CO boiler & incinerator
CO boiler only
Incinerator only
Dryer vent control device
Bag filter
Scrubber
None
Vacuum cleanup system
16
4
6
27
2
1
0
0
19
6
5
20
2. Combustion Devices
It can be seen from Table 41 that most carbon black plants
vent the main process off-gas directly to the atmosphere after
product recovery. Two plants flare the off-gas and report
"about 90% combustion."21 However, the degree of combustion
attainable by this method depends on the heating value of the
vent c;as. Thus, the removal efficiency for combustible gases
is estimated to range from 75% to 80% for the production of
low yield blacks and up to 95% for high yield blacks.21 It
is estimated that 20% to 30% of the vented carbon black is
burned.2l
Field sampling data obtained in the present study indicate a
much higher flare removal efficiency for carbon monoxide than
the above estimates. The concentration of carbon monoxide in
104
-------
the flare off-gas was measured at 50 ppm, corresponding to a
combustion efficiency greater than 99.9% in the flare.
(Details of the experimental procedure are given in Appendix H)
One other plant in the United States burns the main process
vent gas, utilizing a carbon monoxide boiler and thermal
incinerator in combination. About 60% of the vent gas is
burned with supplemental fuel in the carbon monoxide boiler
to provide the process steam requirements for the plant.
Another 10% of the vent gas is used in the dryers, and the
remaining 30% is disposed of in the incinerator.
A typical carbon monoxide boiler-thermal incinerator control
system is shown schematically in Figure 14. This system
differs from the one actually in use in that provision is
made for energy recovery in the incinerator by preheating
the feed streams. The supplemental fuel requirement of the
incinerator is thereby minimized. Also, in Figure 14 it is
assumed that none of the vent gas is utilized in the process
dryers. A material balance for this system, based on the
process vent gas (stream 11) of Table 19, is presented in
Table 42. Complete combustion of carbon monoxide, sulfur
compounds, hydrocarbons, and particulate matter is assumed
in the material balance.
Both the carbon monoxide boiler and the thermal incinerator
are designed to operate at a temperature of 980°C in the com-
bustion zone with excess air equivalent to 4 mole percent
oxygen in the stack gas. Based on a heating value of
1.49 MJ/m3 for the process vent gas, the energy recovery
in the incinerator is sufficient to maintain the required
operating temperature without any additional fuel except for
the pilots.
105
-------
PROCESS
VENT
GAS
NATURAL GAS (4
FUEL
AIR
(27°C)
(40%)
Figure 14. CO boiler-thermal incinerator
system for carbon black plant21
The following problems are associated with the use of carbon
monoxj.de boilers in carbon black plants:21
«• Process vent gas is available only at low
pressure and is corrosive. Hence, a large
investment is required for blowers, large
diameter pipes and valves, control equip-
ment, and burning equipment.
" Because of the low heating value of the
process vent gas (1.12 to 1.86 MJ/m3), up
to 35% of the total heating value must be
added as supplemental fuel in order to
achieve complete combustion.
106
-------
Table 42. MATERIAL BALANCE FOR CO BOILER-THERMAL INCINERATOR CONTROL SYSTEM
2 1
Stream no. :
Description:
Temperature, °C:
Gage pressure, kPa:
Component
Hydrogen
Carbon dioxide
Carbon monoxide
Hydrogen sulfide
Sulfur oxides
Methane
Acetylene
Ethane
Carbon black
Nitrogen
Oxygen
Water
Nitrogen oxides
Carbon disulfide
Carbonyl sulfide
TOTALS
1
Process
vent
gas
232
2
Vent gas
to CO
boiler
232
3
Air
to CO
boiler
27
4
Natural
gas to CO
boiler
27
5
Flue gas
from
boiler
400
6
Boiler
feed
water
116
Flow rates, kg/hr
563
3,518
7,029
66
0.6
145
174
11
41,221
739
36,465
8.2
66
23
90,029
338
2,111
4,217
40
0.4
87
104
6.6
24,733
443
21,879
4.9
40
14
54,018
30,428
9,224
432
40,084
185
40
19
244
10,002
158
55,182
2,919
26,074
11
94,346
25,973
25,973
Except as indicated, all streams are at low pressure, i.e., atmospheric or slightly
above.
5Blanks indicate no mass flow of component.
-------
Table 42 (continued). MATERIAL BALANCE FOR CO BOILER-THERMAL
INCINERATOR CONTROL SYSTEM21
Stream no. :
Description:
Temperature, °C:
Gage pressure, kPa :
Component
Hydrogen
Carbon dioxide
Carbon monoxide
Hydrogen sulfide
Sulfur oxides
Methane
Acetylene
Ethane
Carbon black
Nitrogen
Oxygen
Water
Nitrogen oxides
Carbon disulfide
Carbonyl sulfide
TOTALS
7
Steam
400
3,103
8
Boiler
blowdown
238
9
Vent gas to
incinerator
232
i
10
Air to
incinerator
27
11
Incinerator
stack gas
870
Flow rates, kg/hr
24,727
24,727
1,246
1,246
225
1,407
2,812
26
0.2
58
70
4.4
16,488
296
14,586
3.2
26
9
36,011
17,919
5,428
253
23,600
6,251
105
34,407
1,819
17,023
6
59,611
o
00
Except as indicated, all streams are at low pressure, i.e., atmospheric or slightly
above.
'Blanks indicate no mass flow of component.
-------
In addition to the above problems, combustion of the process
vent gas results in increased emissions of sulfur oxides and
nitrogen oxides. Removal of these contaminants would be
difficult and expensive due to their low concentrations (0.02%
to 0.17% by volume, dry basis).21 Based on the material
balance in Table 42, a stack height of 41 m,21 and no plume
rise, the source severities for the carbon monoxide boiler-
incinerator system in a 5.1 x 10** metric tons/yr plant (the
representative plant size) would be 2.8 for SO and 0.76 for
X
NO . The SO severity which results from combustion of the
X X
vent gas depends on the amount of sulfur-containing compounds
in the vent gas, as well as on the emission height of the
combusted gas. The emission height is affected because com-
bustion increases the temperature of the vent gas, which
results in greater plume rise. The SO severity resulting
X
from combustion was also calculated for a vent gas having the
composition listed in Table 26 (Section IV.B.I). This com-
position was measured by MRC in the present study, and was
used as the basis for the source severities for the repre-
sentative plant listed in Table 35 (Section IV.C.3). This
vent gas corresponds to a feedstock having a relatively high
(3.7%) sulfur content. The following assumptions were made:
• Complete combustion of all sulfur compounds in
the vent gas
• A stack height of 25 m, the same as that for the
main process vent in the representative plant
• Production rate of 5.1 x 104 metric tons/yr, the
same as that for the representative plant
• No plume rise
The resulting SO severity was 15. Inclusion of plume rise
H
in the calculation gave a severity of 2.3. Thus, control
of hydrogen sulfide, carbonyl sulfide, and carbon disulfide
(uncontrolled severities of 20, 2.1 and 7.6, respectively,
109
-------
neglecting plume rise; uncontrolled severities of 4.0, 0.4
and 1.5, respectively, including plume rise) is achieved at
the expense of a relatively high SO severity.
X
3. Dryer Vent Gas Control
Nearly two-thirds of the furnace black plants employ bag
filters to remove entrained carbon black from the direct-
fired dryer vent gas stream. The control efficiency of these
devices is 99.8% to 99.95%.21
One manufacturer uses a three-stage water scrubbing system
in its plants, in which the dryer vent gas is sucked into a
tangential entry, vertical, cylindrical scrubber by means of
a blower located downstream of the scrubber. The discharge
from the1, blower is fed to a venturi scrubber which vents
through a cyclone separator to the atmosphere. Water is
sprayed into the blower inlet as well as the two scrubbers
to provide three scrubbing stages. Slurry collected from the
cylindrical scrubber and cyclone separator is recycled to the
scrubber sprays and reactor quench streams. All of the
carbon black recovered from the dryer vent gas is thus
return to the main process sequence.
Impinger-reactor scrubbers, which are described in Reference
44, are in use in at least one carbon black plant. The
particulate removal efficiency attained by the units in-this
installation is 92.8%, with an estimated availability of
80%. ^
^Hesketh, H. E. Scrubbing with an Impinger Reactor.
In: Proceedings of the Second National Conference on
Energy and the Environment, college Corner, Ohio,
November 13-15, 1974. pp. 174-176.
110
-------
4. Other Control Devices
Carbon- black is recovered from the pneumatic transport
system by means of a bag filter and/or a cyclone. When only
a cyclone is used, the exhaust gas is recycled to the product
recovery bag filters to form a closed system. When a bag
filter is used with the cyclone, the exhaust gas is vented
to the atmosphere.
Two-thirds of the furnace black plants incorporate a plant-
wide vacuum cleanup system such as the one described in
Section III.B.5. This system is used to pick up loose carbon
black resulting from leaks and spills. It can also be used
to control emissions from the carbon black storage bins and
from the bagging system. Some plants employ separate bag
filters on the storage bin vents to control carbon black
emissions.
The oil feedstock storage tanks are not equipped with control
devices. Although conservation vents or floating roof tanks
could be used for control, all feedstock storage tanks meet
EPA New Source Performance Standards for this type of
emission point (personal communication, C. B. Beck, Cabot
Corporation, Pampa, Texas, 26 February 1976).
One carbon black plant reportedly employs a fluid bed in-
cinerator to dispose of waste sludge.45 Solid waste
incinerators in two plants are reported in the National
Emission Data System point source listing. The sludge
burned in these units originates from scrubbers and/or
washdown of carbon black spills. Liquid waste from these
45Wall, C. J-, J. T. Graves, and E. J. Roberts. How to Burn
Salty Sludges. Chemical Engineering, 82:77-82, 1975.
Ill
-------
operations is usually recycled via the quench water streams
or allowed to evaporate in settling ponds. The sludge
from the settling ponds, which ranges from 0 to 20 g/kg of
product, is normally used as landfill rather than being
incinerated.21 Other solid waste from carbon black opera-
tions consists of used filter bags (0.15 to 0.25 g/kg
product) and used furnace refractory bricks (0.35 to 0.75
g/kg product).21
Estimated emission factors for solid waste incineration in
carbon black plants, based on emission rates obtained from
the National Emission Data System, are given in Table 43.
Table 43. ESTIMATED EMISSION FACTORS FOR SOLID WASTE
INCINERATORS IN CARBON BLACK PLANTS
Material
emitted
Particulate matter
so}
NO
x
X
Hydrocarbons
Carbon monoxide
Emission factor,
0.12
0.01
0.04
0.01
0.01
B. FUTURE CONSIDERATIONS
1. Combustion Devices for Existing Plants
a. Thermal Incinerator - The principal source of atmospheric
emissions from carbon black plants is the main process vent
gas. Hence, the implementation of combustion devices to
control these emissions will play a central role in future
control strategy. Since most existing carbon black plants
are not designed for steam-driven operation, the most effec-
tive combustion device for these plants would probably be a
112
-------
flare or a thermal incinerator of the type discussed pre-
viously in Section V.A.2 and illustrated in Figure 14. This
incinerator minimizes supplementary fuel consumption by
recovering the heat of combustion via heat exchange with the
air and vent gas feed streams.
Table 44 presents a material balance for a thermal incinera-
tor treating the vent gas stream (stream 11) of Table 19.
Complete combustion of all combustible material is assumed.
Based on this material balance, a stack height of 41 m,21
and no plume rise, the source severities for SO and NO for
H J\.
a 5.1 x 101* metric tons/yr plant (the representative plant
size) would be 2.6 and 0.64, respectively.
b. Catalytic Incinerator - Catalytic incineration is an
alternative to thermal incineration with feed preheat. A
catalytic incinerator is shown schematically in Figure 15,
and the corresponding material balance is given in Table 45.
The material balance assumes an inlet temperature to the
catalyst bed of 480°C and a maximum bed temperature of 650°C
in order to prevent loss of catalyst surface area. Complete
combustion of carbon monoxide, sulfur compounds and hydro-
carbons is also assumed. Based on this material balance, a
stack height of 41 m,21 and no plume rise, the source
severities for a 5.1 x 10^ metric tons/yr plant would be
2.6 for SO , 0.47 for NO , and 1.0 for carbon black.
X X
Catalytic incinerators have not been used previously in
carbon black plants, and their applicability to these opera-
tions is uncertain. Some possible problem areas are:21
• Loss of incinerator efficiency as the catalyst ages.
• Catalyst poisoning by sulfur compounds.
• Formation of sulfur trioxide on the catalyst.
• Problems associated with periods of high carbon
black loading due to leaks in bag filters.
113
-------
Table 44. MATERIAL BALANCE FOR THERMAL INCINERATOR21
Stream:
Temperature, °C:
Vent gas to
incinerator
232
Air to
incinerator
27
Incinerator
stack gas
870
Components Flow rates/ kg/hr
Hydrogen.
Carbon dioxide
Carbon monoxide
Hydrogen sulfide
Sulfur oxides
Methane
Acetylene
Carbon black
Nitrogen
Oxygen
Water
Nitrogen oxides
Carbon disulfide
Carbonyl sulfide
TOTAL
563
3,518
7,029
66
0.6
145
174
11
41,221
739
36,465
8.2
66
23
90,029
44,797
13,567
635
58,999
15,627
263
86,014
4,549
42,560
15
149,028
3Blanks indicate no mass flow of component.
Table 45. MATERIAL BALANCE FOR CATALYTIC INCINERATOR21
Stream:
Temperature, °C:
Vent gas to
incinerator
232
Air to
incinerator
27
Incinerator
stack gas
650
Component
Flow rates, kg/hr
Hydrogen
Carbon dioxide
Carbon monoxide
Hydrogen sulfide
Sulfur oxides
Methane
Acetylene
Carbon black
Nitrogen
Oxygen
Water
Nitrogen oxides
Carbon disulfide
Carbonyl sulfide
TOTAL
563
3,518
7,029
66
0.6
145
174
11
41,221
739
36,465
8.2
66
23
90,029
114,119
34,652
2,033
150,804
15,624
263
4
155,339
25,633
43,959
11
240,833
Blanks indicate no mass flow of component.
A large amount of excess air is used to prevent the tempera-
ture in the catalyst bed from exceeding 650°C.
114
-------
STACK GAS
PROCESS VENT
GAS
CATALYST BED
AIR
NATURAL GAS
(PILOTS)
Figure 15. Catalytic incinerator for carbon black plant.21
2. Combustion Devices for New Plants
New carbon black plants can be designed to operate with
steam-driven equipment. However, more steam can be produced
by burning the process vent gas than can be utilized by the
plant. If a market exists for exported steam, then a carbon
monoxide boiler such as the one illustrated in Figure 14
might be the most economical combustion device. In the
absence of a market for steam, the most economical device
would probably be a flare or a thermal incinerator equipped
with a waste heat boiler to supply the steam requirements
of the plant. The latter unit is illustrated in Figure 16.
115
-------
A material balance based on the process vent gas (stream 11)
of Table 19 is given in Table 46. Complete combustion of
all combustible species is assumed. Data from a similar
unit in operation in a Canadian carbon black plant give
combustion efficiencies of 99.6% for hydrogen sulfie and
99.8?i for carbon monoxide.46
STACK GAS
BLOWDOWN
NATURAL GAS
(PILOTS)
PROCESS
VENT GAS
L_L
AIR(27°C)
BOILER FEED
WATER
STEAM
Figure 16. Thermal incinerator with waste heat boiler.
21
46Source Testing of a Waste Heat Boiler. George D. Clayton
Associates. EPA 75-CBK-3, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, 1975. 29 pp.
116
-------
Table 46. MATERIAL BALANCE FOR THERMAL INCINERATOR WITH WASTE HEAT BOILER21
Stream no. :
Description:
Temperature , °C :
Gage pressure, kPa:
Component
Hydrogen
Carbon dioxide
Carbon monoxide
Hydrogen sulfide
Sulfur oxides
Methane
Acetylene
Carbon black
Nitrogen
Oxygen
Water
Nitrogen oxides
Carbon disulfide
Carbonyl sulfide
TOTALS
1
Process
vent gas
232
2
Preheated
air
413
3
Stack
gas
496
4
Boiler feed
water
116
5
Boiler
blowdown
238
6
Steam
400
3,103
b
Flow rates, kg/hr
563
3,518
7,029
66
0.6
145
174
11
41,221
739
36,465
8.2
66
23
90,029
44,797
13,567
635
58,999 .
15,627
263
86,014
4,549
42,560
15
149,028
25,973
25,973
1,246
1,246
24,727
24,727
Blanks indicate data not available.
Blanks indicate no mass flow of component.
-------
The process vent gas is burned with preheated air at 980°C.
The only supplemental fuel required is natural gas for the
incinerator pilots. Hence, this device is more efficient
than the carbon monoxide boiler-thermal incinerator combin-
ation discussed in Section V.A.2. In addition to the
Canadian plant mentioned above, at least one European plant
employs a device of this type.21
Based on the material balance given in Table 46, a stack
height of 41 m,21 and no plume rise, the source severities
for a 5.1 x 10^ metric tons/yr plant would be 2.6 for sulfur
oxides and 0.64 for nitrogen oxides.
3. Carbon Monoxide Recovery
A new process for the recovery of carbon monoxide from
process gas streams has recently been described in the
literature.47 The process, termed Cosorb by its developer,
Tenneco Chemicals, recovers carbon monoxide by absorption
in a solvent consisting of copper(I) aluminum chloride
(CuAlClif) in a toluene base. This process represents a
possible alternative to combustion devices for control of
carbon monoxide emissions.
A flow diagram for the Cosorb process is shown in Figure 17.
The carbon monoxide containing feed gas first passes through
a drying unit that removes water which would otherwise react
with the copper(I) aluminum chloride. The feed gas is then
contacted with the Cosorb solvent in a countercurrent
absorber at ambient temperature and any practical pressure.
The carbon monoxide is removed from the gas stream by for-
mation of a chemical complex with the solvent. Removal
efficiencies of up to 99% by volume are feasible.
l|7Haase, D. J. New Solvent Cuts Costs of Carbon Monoxide
Recovery. Chemical Engineering, 82:52-54, 1975.
118
-------
. H2-RICH PRODUCT
[AROMATIC]
^RECOVERY]
FEED GAS
ABSORBER
, RICH SOLVENT
| SOLVENT I*
[MAINTENANCE^
I I
COOLING WATER
CO PRODUCT
STEAM
LEAN SOLVENT
x^
COOLING
WATER
Figure 17. Cosorb process for carbon monoxide recovery from industrial gases47
-------
The carbon monoxide-rich solvent from the absorber is sent
to a flash unit where those compounds which were physically
(as opposed to chemically) absorbed in the toluene are sepa-
rated from the solvent. The vapor stream from the flash
unit is compressed and recycled to the absorber.
Carbon monoxide is recovered from the solvent in a stripping
column. The stripping action of the toluene solvent base,
together with heat supplied by low-pressure steam, results in
an overhead stream which is typically greater than 99% carbon
monoxd.de by volume. As shown on the flowsheet, toluene
recovery systems may be required on both the carbon monoxide
product stream and the absorber tail gas. These systems may
employ compression, refrigeration, activated carbon, or non-
volatile-liquid scrubbing.
Lean solvent from the stripping column is recycled to the
absorber after heat exchange with the stripper feed. A
solvent maintenance section may be needed to remove compounds
other than carbon monoxide which are complexed by the solvent.
For example, hydrogen sulfide and sulfur dioxide both react
with the Cosorb solvent to some extent. Hence, the possi-
bility exists for removing these compounds, as well as the
carbon monoxide, from the process vent gas. The carbon
monoxide recovered from the stripping column is of suffi-
cient purity to be used for chemical synthesis.
4. Current Research Studies
Industry research projects in air pollution control are con-
cerned with reducing the volume of vent gas which must be
processed. The methods being studied include the following:21
• Operation of the reactors at pressures up to several
atmospheres to reduce the gas volume and provide more
efficient operation of combustion control equipment.
120
-------
However, the operating pressure range is limited in
practice since reactor pressure is one of the variables
that determine the type and quality of black produced.
Use of alternate heat exchange systems to minimize the
amount of quench water required. A reduction in the
amount of quench water would improve the heating value
of the vent gas and reduce its volume.
Substitution of oxygen for air in the reactor feed.
Pilot plant tests have been unsuccessful because the
use of pure oxygen results in a flame temperature which
is too high for the furnace refractory. However, this
modification merits further study since, in addition
to reduced emissions, it could result in higher product
yields and increased reactor capacity. In addition,
the vent gas could be used as a synthesis gas for the
Fischer-Tropsch process or for ammonia production, thus
providing a valuable energy resource (see the following
paragraph).
Recycle of the vent gas to the reactor. This method of
control has also been unsuccessful because the low
heating value of the vent gas results in a flame tem-
perature which is too low for high-quality blacks. If
the water and nitrogen content of the vent gas could
be reduced as indicated above, vent gas recycle might
prove to be feasible. A process employing oxygen feed
and vent gas recycle to produce carbon black and syn-
thesis gas was patented in 1954.17 However, no
experimental results were presented to support the
feasibility of the proposed process. A block flow
diagram of the patented process is presented in Figure
18.
Use of desulfurized feedstocks to reduce sulfur and
trace metal emissions. One problem with this procedure
is that hydrotreating reduces the aromatic content of
the oil. This reduction results in lower carbon black
yields and reduced reactor capacities.
121
-------
NITROGEN
AIR
1
MM
0>
OXYGEN
SEPARATION UNIT
f
CYGEN
MAKE MATERIAL
HYDROCARBON FEEDSTOCK
RECYCLE
CARBON BLACK
FURNACE
QUENCH WATER
CARBON BLACK
1 vutiNun \
L I
CARBON BLACK
COLLECTION SYSTEM
WATER
NJ
CARBON DIOXIDE
CARBON DIOXIDE
REMOVAL
WATER
I
3YE
I
SYNTHESIS GAS FOR
FISCHER-TROPSCH PROCESS
OR AMMONIA PRODUCTION
Figure 18. Block diagram for production of carbon black
and synthesis gas17
-------
SECTION VI
GROWTH AND NATURE OF THE INDUSTRY
A. PRESENT TECHNOLOGY
Carbon black is presently produced in the United States by two
major processes: the furnace process and the thermal process.
The furnace process accounts for more than 90% of total pro-
duction, while the thermal process accounts for about 10%.
The furnace process consists of three basic steps: feedstock
cracking, product recovery, and product modification. The
first step involves the partial oxidation and thermal decom-
position of the hydrocarbon feedstock in a refractory-lined
furnace to produce finely divided carbon particles. The
reactor effluent is water quenched to terminate the cracking
reaction and prepare the effluent for the product recovery
step.
Prior to 1965, most plants recovered the carbon black from
the reactor effluent stream by means of electrostatic precip-
itators and cyclones, with or without wet gas scrubbers.
Nearly all furnace black plants now use bag filters, which
result in reduced particulate emissions as well as improved
product yield. The bag filters either have been added on to
existing recovery devices or have replaced them entirely.
The exhaust gas from the bag filter units is normally vented
to the atmosphere, although a few plants employ combustion
devices to reduce carbon monoxide, hydrogen sulfide, and
hydrocarbon emissions.
123
-------
Product modification involves forming the raw carbon black
into pallets to increase its bulk density for ease of trans-
portation and handling. Pelletizing is accomplished by
working the carbon black with water which is subsequently
removed in a drying operation. Emissions from the dryer
exhaust are controlled with bag filters or water scrubbers.
Most (two-thirds) plants also employ a plantwide vacuum
cleanup system to minimize emissions from carbon black spills.
This system may also be used to control emissions from car-
bon black storage tanks and packaging operations.
B. EMERGING TECHNOLOGY
Since about 1950, the trend in the carbon black industry
has been toward the oil furnace process, due to its greater
efficiency and versatility. This trend is illustrated in
Figures 19 and 20, which depict the history of carbon black
production by process and by raw material. The rising cost
and limited supplies of natural gas in the United States
are expected to result in the continuation of this trend.
The channel process is no longer operated domestically; the
last U.S. plant ceased operations in 1976.21t
Recent technological emphasis in the industry has been on
obtaining a greater degree of control over product properties.
This has resulted in the advent of many new, "improved" carbon
blacks.1*8 The present state of technology is such that blacks
having virtually any combination of particle size and struc-
ture level can be produced to meet new or existing processing
requirements in the consuming industries.
"*8Hael, R. The Improved Blacks: What They Are and Where to
Use Them. Rubber World, 172:42-44, 1975.
124
-------
TOTAL
FURNACE PROCESS
1950
1980
I960 1970
YEAR
Figure 19. Carbon black production by process
125
-------
In the area of pollution control, the emphasis has been on
the development and implementation of bag filters for product
recovery, on dryer vents, pneumatic system vents, and, in
some cases, on carbon black storage bin vents. Future devel-
opments will most likely center on combustion devices for
controlling emissions of carbon monoxide, hydrogen sulfide,
and hydrocarbons in the main process vent gas. Two plants
presently flare the off-gas, and one plant employs a carbon
monoxide boiler and thermal incinerator in combination.
Although incineration provides effective emission control,
most existing carbon black plants can neither use nor sell
the steam which can be generated by incineration. Since
the economics of this control method are unfavorable, new
plants can be designed for steam-driven equipment.
C. MARKETING STRENGTHS AND WEAKNESSES
The consumption pattern of carbon black in the United States,
presented in Table 47, shows that the rubber industry accounts
for nearly 88% of total consumption and 93% of the carbon
black consumed domestically. The ink, paint, and paper in-
dustries account for an additional 3% of the total, while
exports and miscellaneous uses make up the remaining 9%. The
recent consumption history of carbon black by the three major
industrial users is depicted in Figure 21. The major cate-
gories are discussed in the following subsections.
126
-------
2,000
1.000
c
o 500
£ 400
300
CO
s 200
z"
o
100
o
o
50
40
30
20
10
5 -
TOTAL PRODUCTION
FROM LIQUID
HYDROCARBONS
FROM NATURAL GAS
1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 20. Carbon black production by raw material1'26
127
-------
Table 47. 1974 CARBON BLACK CONSUMPTION26
Industry
Rubber ..
Ink
Paint
Paper
Miscellaneous
Exports
TOTALS
Carbon black
consumption,
10 3 metric tons
1,326.8
37.7
8.6
1.6
52.6
87.5
1,514.8
Percent of
total sales
87.6
2.5
0.6
0.1
3.5
5.7
100.0
1. Rubber Applications
It is estimated that 60% of the carbon black consumed by the
rubber industry is used in tires of all types.10 The remainder
is used in a wide variety of other rubber products such as
inner bubes, conveyor belts, wire and cable covering, motor
mounts, fan belts, hose, gaskets, rubber heels, etc. The
type and loading of carbon black vary with the application.
For example, highly reinforcing blacks are required in tire
tread, where resistance to abrasive wear is of major impor-
tance,, In tire sidewalls and ply compounds, a minimum of
hysteresis and low heat generation are required, and semi-
reinforcing blacks are used. Carbon black constitutes an
average of about 24% of the total weight of rubber in tires,49
30% to 40% of the total in wire and cable applications,
and about 50% of the total in mechanical rubber goods
Richardson, H. M. Forecasting in the Rubber Industry. In:
Hydrocarbons. The Dilemma in Forecasting. Presented at
the joint meeting of the Chemical Marketing Research
Association and the Commercial Development Association.
New York, New York. May 1974. pp. 77-84.
128
-------
2,000
1,000
500
400
£ 300
o
5 200
CD
e
-. 100
~2.
O
I 50
I? 40
I 30
20
10
TOTAL
ELASTOMERS
PRINTING INK
X
PAINT
/~\
1940 1945 1950 1955 1960 1965 1970 1975 1980 1985 1990
YEAR
Figure 21. Domestic carbon black consumption history1'26
129
-------
(belting, floor mats, molded products, etc.).50 However,
the loading varies widely depending on the product applica-
tion. Usage of the various grades of carbon black in rubber
products is summarized in Table 48.
From 1960 to 1973, total new rubber consumption in the United
States increased at an average annual rate of 5.4%. In 1974,
consumption decreased approximately 1% from the 1973 level.51
Through 1980, consumption is expected to increase at a more
moderate rate of from 2.4% to 3.8%,'t9'52 primarily due to the
effects of energy conservation programs and socio-economic
trends on the transportation industry. This rate of growth
will result in total new rubber consumption from 3.52 x 106
to 3.82 x 106 metric tons in 1980, compared to an estimated
3.05 x 106 metric tons in 1974. However, a growth rate of
only 1.6% per year is forecast for rubber consumption for
automotive tires.49 This rate of growth will result in the
use of 2.14 x 106 metric tons of rubber for tires in 1980,
compared to 1.95 x 106 metric tons in 1974.
2. Printing Inks
Carbon black provides a pigment with extremely high opacity
for use in printing inks. Since well over half of all inks
are black, carbon black is the most widely used pigment in
the industry. Loadings range from 15% to 22% in offset inks,
15% to 20% "in most letterpress inks, 10% to 12% in letterpress
news inks, and 8% to 12% in gravure inks (personal communica-
tion, C. B. Beck, Cabot Corporation, Pampa, Texas, 2/26/76).
50Burgess, K. A., F. Lyon, and W. S. Stoy. Carbon. In:
Encyclopedia of Polymer Science and Technology, Vol. 2.
Wiley-Interscience Publications, New York, New York, 1965.
pp. 820-836.
5Rubber Demand Faces Lower Growth Rate. Chemical and
Engineering News, 52:12, 1974.
52Rubber Consumption to Increase. Rubber World, 172:83, 1975.
130
-------
Table 48. APPLICATIONS OF CARBON BLACK IN THE RUBBER INDUSTRY2
Type
General properties in rubber
Typical uses
U)
Oil furnace blacks
Super abrasion
Intermediate super
abrasion
Intermediate super abra-
sion - low structure
Intermediate super abra-
sion - high structure
High abrasion
High abrasion - low
structure
High abrasion - high
structure
Fast extruding
General purpose
Conductive furnace
Maximum reinforcement; hard proces-
sing; high heat buildup
Fully reinforcing; good processing
smoothout; medium heat buildup;
good flex resistance
Good reinforcement; low heat build-
up; low modulus; softer compounds
Excellent reinforcement; excellent
processing and extrusion; high
modulus
Fully reinforcing; good processing
smoothout; medium heat buildup;
good- flex resistance
Channel black (EPC) properties; low
heat buildup; low modulus
Fully reinforcing; superior proces-
sing and extrusion properties;
high modulus
Good reinforcement and resilience;
good processing smoothout; low
compression set
Moderate reinforcement; high resil-
ience; low compression set; good
processing smoothout
Excellent electrical conductivity;
good retention after overmilling
or flexing
Tire treads and camelback; heels and
soles of shoes; mechanical goods
Tire treads and camelback; heels and
soles of shoes; mechanical goods
Tire treads and other abrasive ser-
vice applications
Tire treads and high abrasive service
applications; difficult processing
elastomers
Tire treads and camelback; heels and
soles of shoes; mechanical goods
Channel black replacement in natural
rubber and off-road tire treads
Tire treads and other high abrasive
services, particularly in difficult
processing elastomers
Tire carcass, tread base, and side-
wall; butyl innertubes; hose;
extruded stripping
Tire carcass, tread base, and side-
wall; sealing rings; cable jackets;
hose; soling; extruded stripping
Antistatic and conductive rubber
goods; belts; hose; flooring
-------
Table 48 (continued). APPLICATIONS OF CARBON BLACK IN THE RUBBER INDUSTRY2
Type
General properties in rubber
Typical uses
(jJ
Gas furnace blacks3
Fine furnace
High modulus
Semireinforcing
Thermal blacks
Fine thermal
Medium thermal
Channel blacks"
Medium processing
Easy processing
Good reinforcement; moderate heat
buildup
Moderate reinforcement and resil-
ience; easy processing
Semireinforcing; high loading ca-
pacity; easy processing; good
resilience and flex resistance
Low reinforcement; low hardness;
high elongation; good tear resis-
tance; high resilience; good flex
resistance
Lowest reinforcement; high loading
capacity; soft stocks; high
resilience; low stress-strain
properties
Fully reinforcing; medium in proc-
essing and in properties
Fully reinforcing; easy processing;
low hysteresis
Truck tire carcass, breaker, and
cushion
Tire carcass and sidewall; footwear;
mechanical goods
Tire carcass and bead insulation;
mechanical goods; footwear and
soling; wire jackets; belts; hose;
packings
Natural rubber innertubes; inflations;
footwear uppers; mechanical goods
Wire insulation and jackets; mechan-
ical goods; footwear; belts; hose;
packings; stripping mats; proofed
goods
Tire treads; heels and soles of shoes;
mechanical goods
Tire treads; heels and soles of shoes;
mechanical goods
Gas furnace blacks are no longer available. Similar blacks are now produced by the oil furnace process.
Channel blacks are no longer produced domestically; however, they are still available on the international
market. Similar blacks are now produced by the oil furnace process.
-------
News inks, which account for more than 50% of production,
contain carbon black dispersed in a petroleum oil. Although
channel blacks were used in these inks for many years, they
have been largely replaced by HAF grade oil furnace blacks.
Furnace blacks in the particle size range from 25 nm to 75 nm
are the most widely used in inks, often in combination with
medium particle size channel blacks. Furnace blacks alone
tend to develop a blue undertone, whereas channel blacks
develop a browner undertone. Coarse furnace blacks are often
used in carbon-paper inks, which contain mixtures of hard and
soft waxes and mineral oils. Conductive inks used in printed
circuitry are prepared with high concentrations of conductive
carbon blacks. Applications of carbon blacks in the ink and
related industries are summarized in Table 49.
The consumption of printing ink increased at an average
annual rate of 4% from 1960-1974, as did the consumption of
carbon black for printing inks.1 While this rate of growth
can be expected to continue through 1980, the effect on total
carbon black consumption will be minor due to the small per-
centage of total consumption involved.
3. Paint and Coatings
The paint industry employs a wider range of carbon blacks
(as pigments) than any other industry. The finest particle,
high-color blacks (10 nm to 13 nm) are used in thermosetting
acrylic finishes for automobiles and appliances. Medium-
color channel and furnace blacks (14.5 nm to 25 nm) are used
in general purpose maintenance and utility paints, while
coarser blacks (25 nm to 40 nm) are used in drum enamels,
chassis paints, and other inexpensive coatings. Carbon black
usage in the paint industry is summarized in Table 49.
133
-------
Table 49. APPLICATIONS OF CARBON BLACK IN THE INK, PAINT,
PAPER, AND PLASTICS INDUSTRIES2
Type
Applications
oo
Oil furnace blacks
High abrasion
Fast extruding
General purpose
Conductive furnace
;
Gas furnace blacks9
Fine furnace
High modulus
Semireinforcing
Thermal blacks
Fine thermal
Medium thermal
Channel blacks^
High color channel
Medium color channel
Regular color channel
Medium flow channel
Long flow channel
Newspaper and publication inks; black paper; carbon paper;
conductive paints
Blue tone and high color strength applications
Tinting and shading paints and plastics
Conductive paints, plastics, and antistatic compounds
Gravure and news inks; black paper; one-time carbons;
paints; plastics
Carbon paper inks; printing inks; maintenance paints;
tinting dispersions
Blue tone and low oil absorption; printing and gravure
inks; one-time carbon paper
Tinting in many applications
Pigment in plastics and utility paints
Jettest carbon blacks; used in lacquers, enamels, and
coloring plastics
Enamels; plastics coloring and light stabilization
Plastics; news and book inks; carbon paper; utility
paints; black paper
Good flow properties in inks; used in paints, coloring
nylon and other fibers
Lithographic inks; letterpress ink; carbon paper; type-
writer ribbon inks
aGas furnace blacks are no longer available. Similar blacks are now produced by the
oil furnace process.
"Channel blacks are no longer produced domestically; however, they are still available
on the international market. Similar blacks are now produced by the oil furnace process.
-------
Carbon black consumption by the paint industry increased at
an average annual rate of 4% from 1960 to 1973.* Consumption
in 1974 declined about 15% from the 1973 level, reflecting
the recession in the construction and automotive industries.
The annual growth rate is expected to average between 4% and
5% over the period 1975 - 1980. Again, this will have little
impact on the overall demand for carbon black due to the
relatively small amount involved.
4. Other Applications
The paper industry employs carbon black in the production of
a variety of black papers, including album paper, leather-
board, wrapping and bag papers, opaque backing paper for
photographic film, highly conducting and electrosensitive
paper, and black tape for wrapping high voltage transmission
cables. Channel blacks were formerly used for these appli-
cations, but have now been largely replaced by furnace blacks
of the appropriate jetness and wetting characteristics.
Typical loadings are in the range of 2% to 8% by weight of
pulp.2
Carbon black is used in plastics as a pigment and, in some
instances, for filling and reinforcement. For example,
crosslinked polyethylene compounds for cable coatings or
pipe may contain up to 50% furnace or thermal black. Carbon
black also functions as a stabilizer, preventing degradation
of the polymer by ultraviolet radiation. In addition to
absorbing ultraviolet radiation, it is effective in termin-
ating the free radical chains produced in the photochemical
reaction. Conductive carbon black is used to produce con-
ductive plastics, which are of importance in the wire and
cable industry for semiconductive outer sheathing and strand
shielding in high voltage cable.
135
-------
Carbon black is also used for insulation for high temperature
operations, for the fabrication of carbon brushes and elec-
trodes, and in the production of antistatic phonograph
records,, and as a source of elemental carbon for ore reduc-
tion and carburizing, a fuel and adsorbent in liquid oxygen
explosives, a grinding agent in the cement industry, and a
pigment in linoleum, leather coatings, polishes, plastic
tile, and concrete.
Exports have declined from a peak of 246 x 103 metric tons
in 1960 to 88 x 103 metric tons in 1974, but have remained
nearly static since 1969 (89 x 103 metric tons).1
5. Industry Forecast
It is obvious from Figure 21 that the total demand for carbon
black is determined by the demand in the rubber industry.
As previously noted, new rubber consumption is expected to
increase at an annual rate of 2.4% to 3.8% through 1980.
However, consumption for tires, which account for an esti-
mated 60% of the carbon black used in the rubber industry, is
expected to increase at a yearly rate of only 1.6%. Based
on these figures, a growth rate of 2% to 2.5% per year through
1980 can be anticipated for carbon black. This rate of
growth will result in the production of 1.73 x 106 to
1.79 x 106 metric tons in 1980 compared to 1.54 x 106 metric
tons in 1974. By comparison, the average growth rate for
carbon black from 1964 through 1974 was 4.3%.1
The oil furnace process will continue to dominate the industry
in the foreseeable future. With the decline of the channel
process, the oil furnace process has become the only source
of the fine-particle blacks required for tire tread and other
high reinforcement applications. In addition, the rising
136
-------
cost and dwindling supply of natural gas will prevent any
major expansion in the thermal process, although it is
possible to operate the latter process on liquid feedstocks,
The same factors may also force the substitution of liquid
hydrocarbons for natural gas as fuel in the oil furnace
process.
The recent production history of the five major grades of
oil furnace blacks is shown in Figure 22. It can be seen
that the HAF grade blacks predominate, presently accounting
for 40% of all carbon black produced in the U.S., and for
nearly one-half of all furnace black production.
137
-------
2,000 -
1940 1950 1960 1970 1980
YEAR
Figure 22. Production history of oil furnace blacks 1
138
-------
SECTION VII
APPENDIXES
A. Storage Tank Calculations
B. Description of Thermal Process
C. Plume Rise Calculations
D. ASTM Nomenclature for Carbon Black
E. Summary of Sampling and Analytical Methods
F. Results of Field Sampling Program
G. Simulated Source Severity Distributions
H. Remote Monitoring of Carbon Monoxide in Flare Off-Gas
139
-------
APPENDIX A
STORAGE TANK CALCULATIONS
The procedure for calculating the emissions from storage
tanks is outlined in this section. The equations given below
were derived in references 28 through .32.
Step 1. Calculate the equivalent gasoline breathing loss:
P *
r)1.73H'0.51 (AT)0*50 F C (A-l
U ' FC (
Y 1000 14.7-P P
where Ly = equivalent gasoline breathing loss, bbl/yr
P = vapor pressure of material stored at bulk
temperature, psia
D = tank diameter, ft
H' = average tank outage, ft
AT = average daily ambient temperature change, °F
F = paint factor
C = diameter factor
The calculations were performed for three different vapor
pressures, 0.1, 0.2, and 0.4 psia. The tank diameter was
calculated from the data in Table 14 by assuming a tank
height of 50 ft. The average tank outage, i.e., freeboard,
was taken as one-half the tank height, or 25 ft.
The average daily ambient temperature change, AT, was assumed
to be 20°F, the national average value. The paint factor,
F , was assumed equal to unity, the value for white paint in
140
-------
good condition. This factor can be as high as 14.6 for gray
surfaces. The diameter factor, C, is equal to unity for
tanks 30 ft or larger in diameter.
Step 2. Calculate the equivalent gasoline working loss:
PVNKT
where F = equivalent gasoline working loss, bbl/yr
V = tank capacity, bbl
N = number of turnovers per year
1.0 for
180 + N
KT = turnover factor =1.0 for N £ 36
-cio for N > 36
6N
Step 3. Compute total equivalent gasoline loss, L
Lg ' Ly + Fg
-------
where L! = petrochemical loss, Ib/yr
Cap = production capacity, ton/yr
E1 = emission factor, Ib/ton
E = emission factor, g/kg
The necessary input data for the above calculations are
tabulated in Table A-l, while the results are summarized in
Table A-2. The tank numbers in these tables correspond to
those in Table 14.
Table A-l. STORAGE TANK INPUT DATA FOR CARBON BLACK PLANT
Tank number
1, 2, & 3
Input data
Average ambient temperature
change, °F
Material molecular weight
Liquid density, Ib/gal
Vapor pressure,9 psia
Tank diameter, ft
Tank outage, ft
Paint factor
Diameter factor
Turnover factor
Number turnovers per yr
Tank capacity, bbl
20
255
9.0
0 . 1 ; 0 . 2 ; 0 . 4
60
25
1.0
1.0
1.0
6
25,000
Calculations were performed for each of the three vapor
pressures shown; all other input data were unchanged for
each vapor pressure level.
142
-------
Table A-2. STORAGE TANK CALCULATION SUMMARY FOR CARBON BLACK PLANT
Tank no.
1,2,3
1,2,3
1,2,3
Vapor pressure, psia
0.1
0.2
0.4
Losses
gal/yr
2,550
4,273
7,238
lb/yr
22,954
38,453
65,142
Emission factor,
g/kg
0.2416
0.4048
0.6857
143
-------
APPENDIX B
DESCRIPTION OF THERMAL PROCESS
The thermal process for manufacture of carbon black is shown
schematically in Figure B-l. Carbon black is produced by the
thermal decomposition of natural gas in the absence of oxygen.
The decomposition takes place in the reactor, which consists of
two reirractory-lined cylindrical furnaces, known as generators.
The dimensions of the furnaces are 3.7 m (diameter) x 10.7 m
(height). They are lined with an open checkerwork of silica
brick.'- The generators are heated to 1300°C by firing with
a stoichiometric mixture of air and fuel. Natural gas is
then admitted and decomposed to carbon and hydrogen as it
passes through the hot checkerbrick.
Each generator is automatically alternated from the produc-
tion cycle to the reheat cycle every five minutes, resulting
in a continuous flow of product from the reaction area. The
carbon black product is entrained in the effluent gas from
the production cycle, which consists of 90% hydrogen, 6%
methane, and 4% higher hydrocarbons.2 After leaving the
generator, the effluent gas passes through a cooling tower
where its temperature is reduced to 125°C by water sprays.
The ca.rbon black product is separated from the exhaust gas
stream by means of cyclones and bag filters. The recovered
black is then processed in the same manner as in the oil
furnace process.
144
-------
U1
EXCESS H2
TO FIRE
BOILER \ JYDROGEN
BLOWER
/-v^-' BUCKET ELEVATOR
, BAG-FILTER EXHAUST (H?)
BUCKET ELEVATOR
DUST AND REJECT
SEPARATOR
VIBRATING SCREEN
MAGNETIC SEPARATOR
'RECYCLE CONVEYOR BELT
\ LJ IX
VT\VALVE N»J'
^ / \ \\ 7?
£-T_SO
^* H«^. . , *^ I
SCREW CONVEYOR
AGITATOR TANK
SEMIAUTOMATIC
PELLET BAG PACKER
Figure B-l. The thermal process for the manufacture of carbon black2
-------
The hydrogen-rich vent gas from the bag filters is cooled,
dehumidified, compressed, and used as fuel to reheat the
generators. Since more hydrogen is produced than is required
for reheating, the excess is used to generate steam and
electricity to operate the plant.2 Alternatively, the excess
vent geis may be flared.
Atmospheric emissions from the thermal process consist of the
following: (1) vent gas from the furnace which is on the
reheat cycle; (2) boiler stack gas; (3) dryer vent gas;
(4) pneumatic system (if used) vent gas; (5) vacuum cleanup
system (if used) vent gas; and (6) fugitive emissions. The
latter four emissions are similar to those from the oil
furnace process since the operations which give rise to these
emissions are similar in the two processes. Emissions (1) and
(2) correspond to the main process vent gas in the oil furnace
process. There is no emission point in the thermal process
which corresponds to the oil storage tank vents in the oil
furnace process.
The vent gas from the furnace reheat cycle consists primarily
of combustion products of the fuel, which, as noted above, is
90% hydrogen. In addition, the vent gas contains carbon
oxides formed by the combustion of carbon deposited on the
checkecbrick during the production cycle. Approximately 50%
of the carbon black formed in each cycle is deposited on the
checkerbrick.5
The boiler stack gas consists of combustion products of the
fuel gas (90% hydrogen) and natural gas that was used as
supplementary fuel.
146
-------
Comparison of emissions from the main process vent in the oil
furnace process with the corresponding emissions (1) and (2)
above from the thermal process shows the latter to be less
severe. In particular:
• Emissions of hydrogen sulfide, carbon disulfide, carbonyl
sulfide, and sulfur dioxide are much lower in the
thermal process due to the very low sulfur content of the
natural gas feedstock. (The hydrogen sulfide concen-
tration in pipeline natural gas ranges from 0.2 ppm to
0.7 ppm, with an average value of 0.3 ppm.53)
• Hydrogen and hydrocarbon emissions are much lower in the
thermal process because the process off-gas is burned
as fuel.
• Carbon monoxide emissions are lower in the thermal pro-
cess because the decomposition of natural gas to form
carbon black is carried out in the absence of oxygen.
Any carbon monoxide formed by reaction of carbon black
with water in the cooling tower is subsequently con-
verted to carbon dioxide since the process off-gas is
burned as fuel. Formation of carbon monoxide during the
heating cycle is limited by the fuel composition, which
is 90% hydrogen and 10% hydrocarbons. Thus, oxidation
of carbon deposited on the checkbrick during the pro-
duction cycle represents the main source of carbon
monoxide emissions.
• POM and trace element emissions are lower in the thermal
process because the feedstock is natural gas rather than
aromatic fuel oil. Exceptions are emissions of trace
elements which originate in the quench water; these
emissions should be comparable in the two processes.
53Finneran, J. A., L. J. Buividas, and N. Walen. Advanced
Ammonia Technology. Hydrocarbon Processing, 51:127-130,
April 1972.
147
-------
Nitrogen oxide emissions from the two processes should
be comparable since the reactors are maintained at
approximately the same temperature in both processes.
Particulate emissions associated with the process off-
gas are lowered for the thermal process since: (1) the
same collection devices (cyclones, bag filters, scrub-
bers) are used in both processes; and (2) combustion of
the thermal process off-gas further reduces the amount
of carbon black emitted to the atmosphere. However, an
additional emission of carbon black occurs in the ther-
mal process when the furnaces are switched from the
production to the reheating cycle. Loose carbon black
deposited on the checkerbrick during the production
cycle is emitted to the atmosphere in the form of a puff,
There emissions occur at five minute intervals.2'5'13
148
-------
APPENDIX C
PLUME RISE CALCULATIONS
The source severities presented in Section IV were calculated
assuming no plume rise, i.e., the effective emission height
was equated with the physical stack height. However, both
the main process vent gas and the dryer vent gas have signi-
ficant buoyancy and momentum, which result in plume rise. In
this section, the effect of plume rise on the source severity
is estimated.
Plume rise is estimated using Holland's equation:54
AH = 4 (1.5 VsD' + 0.04 QH) (C-l)
where AH = plume rise, m
u = wind speed, m/s
V = stack gas exit velocity, m/s
o
D" = stack diameter, m
Q = heat emission rate, kcal/sec
H
The heat emission rate is calculated as follows:
0 - "D'2 V P'M' C fTs " **} (C-2)
QH - -3 Vs -g— Cpl f / (C 2)
5/+Briggs, G. A. Plume Rise. In: Critical Review Series.
U.S. Atomic Energy Commission, Technical Information
Center> Oak Ridge, Tennessee, 1969. 81 pp.
149
-------
where P' = atmospheric pressure, dyne/m2
R = gas constant = 8.314 x 105 dyne - m/g mole - °K
M1 = molecular weight of stack gas, g/g mole
C = heat capacity at constant pressure of stack
gas, kcal/g - °K
T = stack gas temperature, °K
T = ambient air temperature, °K
3,
The values calculated from equation C-l are generally too
low. Based on extensive field data given in Reference 55,
an average correction factor of 3 should be applied to the
calculated plume rise. This value is in agreement with the
correction factor of 2.92 recommended by Stumke.51* Thus,
the plume rise is estimated as three times the value com-
puted from equation C-l.
1. MAIN PROCESS VENT
Based on data obtained at the sampling site in the present
study and on the survey of operating plants given in Refer-
ence 21, the following values are used in the calculation
for the main process vent:
u = 4.5 m/s (definition of source severity)
Vs = 21 m/s
D1 = 0.9 m
P1 = 1.01 x 1010 dyne/m2
M1 = 25
T = 492°K
o
Ta = 292°K
C = ^ 4 x TO"1* kcal
Lp J.4 X ±U g_OR
Moses, H. and M. R. Kraimer. Plume Rise Determination -
A New Technique Without Equations. Journal of the Air
Pollution Control Association, 22:621, 1972.
150
-------
The value for C is the mean heat capacity between ambient
temperature and T for a gas composed of water vapor, nitro-
s
gen, carbon dioxide, carbon monoxide, hydrogen sulfide,
methane, acetylene, and oxygen in the proportions listed
in Table 26 (Section IV.B). That is, only the major com-
ponents of the stack gas were used to compute Cp.
Substituting the above values into equations C-l and C-2
yields:
AH = 4^5- (28.67 + 0.04 x 573)
= (28.67 x 22.91)
= 11.46 meters
Multiplying this value by 3 gives the corrected estimate of
plume rise:
AH . , = 34.4 meters
corrected
Thus, if plume rise is taken into account, the source severi-
ities corresponding to the main process vent for the repre-
sentative source (Tables 35, 36 and 37) should be corrected
(i.e., multiplied) by the following factor:
TTAHJ =(25 f34.4)2 = °'18SS °'2
where H denotes the physical stack height.
S
2. DRYER VENT
Based on data obtained from the survey of operating plants
given in Reference 21, the following values are used for
the calculation of plume rise from the dryer vent:
151
-------
u = 4.5 m/s (definition of source severity)
V = 6 m/s
s
D' = 0.6 m
P' = 1.01 x 1010 dyne/m2
M1 = 28
T = 642°K
s
T = 292°K
cl
Cp = 2.8 x 10"1* kcal/g-°K
The values for M1 and C_ correspond to a gas resulting from
complete, stoichiometric combustion of methane in air.
Substituting the above values into equations C-l and C-2
yields:
AH = -~ (5.4 + 0.04 x 88.0)
(5.4 + 3.5)
4.5
AH = 2.0 m
Multiplying this value by 3 gives the corrected estimate of
plume rise:
^corrected = 6'° m
Thus, if plume rise is taken into account, the source severi-
ties corresponding to the dryer vent for the representative
source; (Table 35) should be corrected (i.e., multiplied) by
the following factor:
H
S ' ' -w ' = 0.59 = 0.6.
Hs + AH
152
-------
APPENDIX D
ASTM NOMENCLATURE FOR CARBON BLACK
The carbon black classification system adopted in 1967 by the
American Society for Testing and Materials is presented in
Tables D-l and D-2. The first digit of the ASTM number desig-
nates the particle size of the black, as indicated in Table
D-l. The next two numbers are arbitrarily assigned by an ASTM
committee. However, this convention is not strictly adhered
to, especially in the case of the newer blacks. For example,
the particle size of grade N375 is approximately 23.5 nm,1*8
so that it should be assigned to the 200 series.
Table D-2 is a partial listing of ASTM carbon black numbers
together with the corresponding industry classifications.56
The letters N and S denote normal rubber cure rate and slow
rubber cure rate, respectively.
56 Standard Classification System for Carbon Blacks Used in
Rubber Products. 1974 Annual Book of ASTM Standards.
Part 38. American Society for Testing""and Materials,
Philadelphia, Pennsylvania, 1974. pp. 408-409.
153
-------
Table D-l. ASTM CARBON BLACK PARTICLE SIZE CODE56
First
ASTM
Digit of
number
0
1
2
3
4
5
6
7
8
9
Typical Particle
Size, nm
1 to
11 to
20 to
26 to
31 to
40 to
49 to
61 to
101 to
201 to
10
19
25
30
39
48
60
100
200
500
154
-------
Table D-2. ASTM CARBON BLACK NOMENCLATURE SYSTEM a,b,56
ASTM
Number
N110
N166
N195
S212
N219
N220
N231
N234
N242
N270
N285
N293
N294
S300
S301
S315
N326
N327
N330
N332
N339
N347
N351
N356
N358
Industry
Classification
SAP
SAF-HS
SCF
ISAF-LS-SC
ISAF-LS
ISAF
ISAF-LM
_C
ISAF-HS
ISAF-HS
ISAF-HS
CF
SCF
EPC
MFC
HAF-LS-SC
HAF-LS
HAF-LS
HAF
HAF
HAF
HAF-HS
HAF-HS
C
SPF
ASTM
Number
N363
N375
N440
N472
N539
N542
N550
N568
N601
N650
N660
N683
N741
N754
N762
N765
N774
N785
N787
N880
N881
N907
N908
N990
N991
Industry
Classification
HAF-LS
HAF
FF
XCF
FEF-LS
FEF-LS
FEF
FEF-HS
HMF
GPF-HS
GPF
APF
_C
SRF-LS
SRF-LM-NS
SRF-HS
SRF-HM-NS
MPF
SRF-I-HS
FT
FT
MT-NS
MT-NS
MT
MT
Personal communication, C. B. Beck, Cabot Corporation, Pampa,
Texas, 26 February 1976.
Personal communication, H. J. Collyer, Cabot Corporation,
Billerica, Massachusetts, 14 January 1976.
No corresponding industry classification.
155
-------
APPENDIX E
SAMPLING AND ANALYTICAL METHODS
1. SAMPLING METHODS
a. Location of Sampling Ports
The plant selected for sampling utilizes the processing
scheme shown in Figure E-l for handling the main process
vent gas. Gases from the carbon black furnace and quencher
enter the baghouse where carbon black product is recovered.
The main process vent gas leaves the baghouse at 220°C
(425°:?) and 75 mm to 100 mm H20 (3 in. to 4 in. H20) posi-
tive pressure in a 1.07 m (42 in.) circular duct. The main
process vent gas is split into two streams—the flare stack
gas and the fuel-user trunk line. The flare stack is 0.76 m
(30 in.) in diameter. A backpressure control valve located
at the base of the flare stack maintains a 75 mm to 100 mm
H20 (3 in. to 4 in. H20) positive pressure on the fuel-user
trunk line which is also a 0.76 m (30 in.) duct that transfers
^50% of the main process vent gas through a splitter to a
fuel-user manifold. Six 0.3 m (12 in.) ducts transfer the
main process vent gas (fuel gas) to dryers used in the
plant.
The main process vent gas was sampled at two locations shown
in Figure E-l: (1) in the flare stack; and (2) in the feed
duct to one of the fuel-users (dryers). At the first loca-
tion r only velocity traverses were made in order to determine
156
-------
the portion of the vent gas being flared. The second loca-
tion was sampled for particulate matter and trace elements;
POM's, hydrocarbons, and organosulfur compounds; CO, C02,
02, H2S, COS, CS2, and S02; H2, N2, H20, and NO .
J^
The flare stack sampling ports are 35 mm (1.4 in.) in dia-
meter (1 in. standard pipe) and are located 0.9 m (3 ft)
above the backpressure valve and 2.7 m (9 ft) below the top
of the flare stack. These ports are spaced 90° from each
other. They are 1.2 pipe diameters downstream of the back-
pressure valve and 3.6 pipe diameters upstream of the tip of
the flare. The fuel-user sampling ports are located 3 m
(10 ft) downstream of the mainfold and 0.9m (3 ft) upstream
of the water seal.
BACK PRESSURE
MAIN PROCESS VALVE —
VENT GAS
GASES FROM
CARBON BLACK
FURNACE AND
QUENCHER
— 0.76m (30 in. )
FEED DUCTS ( 6 I TO
FUEL USERS I DRYERS I
FUaUSER
SAMPLING PORTS
0.3m [12 in.]
Figure E-l. Process schematic showing disposition of
main process vent gas and location of sampling ports
157
-------
b. In-Stack Measurements
The methods used for sampling the various species in the
main process vent gas are described in the following subsections,
(1) Particulate Matter and Trace Elements - A modification
of the EPA Method 5 particulate sampling system was employed
for these materials. The basic procedures and equipment
used are specified in the Federal Register;57 however, the
sampling train was modified in order to collect both particu-
late and inorganic materials as specified in the Level I
environmental assessment sampling procedure.58
The sampling system consisted of the Method 5 probe tip and
probe assembly, with the probe containing a quartz liner.
The probe was connected to a heated filter. For this study,
the filter used was a Millipore Fluoropore® type FA filter,
60 mm in diameter with a nominal pore size of 1,000 nm. The
exit side of the filter was connected to a series of four
impingers designed to collect vaporous inorganic species.
The first impinger contained 1 x 10"1* m3 (100 ml) of 6M ^.2^21
while the second and third both contained 1 x lO"4 m3
(100 ml) of 0.2M (NHit)2S208 and 0.02M AgN03. The fourth
impinger contained 200 g of silica gel. The sampling system
was operated as specified in the Federal Register by travers-
ing and maintaining isokinetic conditions at each traverse
point for sampling periods of 2 hours per run, for a total
of three runs.
57Environmental Protection Agency - Part II - Standards of
Performance for New Stationary Sources - Proposed Amend-
ments to Reference Methods. Method 5 - Determination of
Particulate Emission from Stationary Sources. Federal
Register, 41 (111):23076-23083, 1976.
58Hamersma, J. W., S. G. Reynolds, and R. F. Maddalone.
IERL-RTP Procedures Manual: Level I Environmental Assess-
ment. EPA-600/2-76-160a, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, 1976. 131 pp.
158
-------
Typical particle sizes for the reinforcing grade blacks
(N330 and N351) produced during the sampling runs are 25
nm to 30 nm. Based on the work of Liu and Lee,59 the col-
lection efficiencies of Millipore Fluoropore® filters are
greater than 99% for particles in this size range. In the
work reported in Reference 59, monodispersed dioctyl
phthalate (DOP) aerosols ranging in size from 30 nm to
1,000 nm were generated, and the efficiencies of filtration
of Millipore Fluoropore® type FA filters with nominal 1,000 nm
pores were measured as a function of particle size and
pressure drop, the latter ranging from 10 mm Hg t,o 300 mm Hg.
The results show that the efficiencies of the filters exceeded
99.99% over the entire operating range.
The material collected in the probe and filter was used to
obtain the mass emission rate. After completion of the
sampling run, the probe and probe tip were worked and brushed
using high purity distilled water and a nylon brush. The
liquid content of each impinger was measured, and the con-
tents were then transferred to individual bottles. Each
impinger was rinsed with three portions of distilled water,
and this water was poured into a labelled bottle. The
silica gel impinger contents were weighed and the material
was placed in a bottle for shipment to the laboratory.
Prior to shipment, all liquid containing samples were acidi-
fied with high purity concentrated nitric acid to a pH of
1 to 2 to prevent formation of precipitates and adsorption
of trace elements on the glass bottle surfaces.
59Liu, B. Y. H. and K. W. Lee. Efficiency of Membrane and
Nucleopore Filters for Submicrometer Aerosols. Environ-
mental Science and Technology, 10:345-350, 1976.
159
-------
(2) PoLycyclic Organic Materials, C-6 and Higher Hydro-
carbons, and Organic Sulfur Compounds - The system employed
for sampling these materials was a modification of the EPA
Method 5 particulate sampling procedure.57 For sampling,
the standard fiberglass filter was replaced with a Millipore
Fluoropore type FA filter and an organic trap 30 mm in
diameter and 100 mm long. The organic trap contained XAD-2
resin and was placed between the filter and impingers. In
addition, a gas cooler was employed between the filter and
the resin trap in order to cool the gas and condense some of
the water in the gas stream. This cooling is essential for
carbon black sampling due to the 230°C (450°F) stack
temperature and the high moisture content (45% to 50%) of
the stack gas. The impinger system consisted of four impingers
The first impinger contained 1 x 10"^ m3 (100 ml) of 10% KOH,
the second and third impinger each contained 1 x 10"k m3
(100 ml) of toluene, and the fourth contained 200 g of 6-mesh
to 16-mesh silica gel.
The sampling system was operated for three sampling runs by
normal particulate sampling procedures, i.e., traversing and
maintaining isokinetic conditions at each sampling point.
At the completion of each run, the impinger contents were
measured and transferred to amber bottles. The impingers
were rinsed with the liquid they contained (10% KOH or
toluene), and these rinsings were added to the proper bottle,
which was then maintained at 0°C during shipment and holding
for analysis. The filter was removed, placed in a petri
dish, sealed, and held over ice at 0°C. The resin trap was
sealed, placed in a plastic bag, and held over ice at 0°C.
The entire train was rinsed with 1:1 methylene chloride-
methanol, and these rinsings were stored in an amber bottle
maintained at 0°C.
160
-------
(3) Lower Molecular Weight Hydrocarbons - The hydrocarbon
content of the gas stream was determined in two parts.
Higher molecular weight compounds (above Cg) were collected
by the XAD-2 trap employed in the POM collection system.
Lower molecular weight compounds were analyzed on site by a
portable gas chromatograph equipped with a flame ionization
detector.
Samples for low molecular weight hydrocarbon analysis were
obtained from the integrated gas samples collected during
the Method 5 particulate-trace metal sampling runs and the
POM sampling runs after portions of these integrated samples
had been analyzed for gas composition as directed under
Method 3, "Gas Analysis for Carbon Dioxide, Excess Air and
Dry Molecular Weight," in the Federal Register.60 The
samples were collected over the entire particulate sampling
period using Tedlar bags. Prior to use, the bags were
evacuated and purged with nitrogen. A portion of the nitro-
gen was analyzed for hydrocarbons as a background check.
(4) H2S, COS, CS2 • an(^ S02 ~ Sampling for these materials
was performed using EPA Method 15 as described in the Federal
Register.6 l
.1
(5) H2, N2, CO, C02, 02/ and H20 - During the modified
Method 5 particulate, trace element, and POM sampling,
6°Environmental Protection Agency - Part II - Standards of
Performance for New Stationary Sources - Proposed Amend-
ments to Reference Methods. Method 3 - Gas Analysis for
Carbon Dioxide, Oxygen, Excess Air, and Dry Molecular
Weight. Federal Register, 41 (111):23069-23072,1976.
6 Environmental Protection Agency - Standards of Performance
for New Stationary Sources - Petroleum Refinery Sulfur
Recovery Plants. Test Methods and Procedures. Federal
Register, 41 (193):43870-43874, 1976.
161
-------
gaseous samples were collected using EPA Method 3, "Gas
Analysis for Carbon Dioxide, Oxygen, Excess Air, and Dry
Molecular Weight" integrated gas procedure.60 The samples
collected in the Tedlar bags were analyzed on site by the
Orsat technique for CO, C02, 02, and by difference for N2
and H2. For this type of industry, the Method 10 NDIR-CO
procedure62 is not appropriate since the concentration of CO
may exceed the 5,000 ppm capability of the NDIR instrument.
In addition to the Tedlar bag samples, grab samples were
collected in evacuated glass sampling bulbs for later analy-
sis of H2 and N2, and for confirming the CO, C02, and 02
levels.
The moisture content of the gas stream was obtained from the
Method 5 particulate-trace element sampling train impinger
contents. As noted above, the increase in liquid level in
the impinger was measured prior to rinsing the impinger with
solvent and placing the liquid and rinses into a bottle.
(6) Nitrogen Oxides - Sampling for nitrogen oxides was
performed using Method 7 as described in the Federal
Register.6 3'6U
62Chapter 1 - Environmental Protection Agency - Part 60 -
Standards of Performance for New Stationary Sources. Sub-
chapter C - Air Programs. Subpart 0 - Standards of Per-
formance for Sewage Treatment Plants. Federal Register,
39:9319-9321, 1974.
Chapter 1 - Environmental Protection Agency - Part 60 -
Standards of Performance for New Stationary Sources. Sub-
chapter C - Air Programs. Method 7 - Determination of
Nitrogen Oxide Emissions from Stationary Sources. Federal
Register, 36:24891-24893, 1971.
6^Environmental Protection Agency - Part II - Standards of
Performance for New Stationary Sources - Proposed Amend-
ments to Reference Methods. Method 7 - Determination of
Nitrogen Oxide Emissions from Stationary Sources. Federal
Register, 41 (111):23085-23087, 1976.
162
-------
c. Remote Monitoring
Remote monitoring of carbon monoxide in the flare off-gas
is discussed in Appendix H.
d. Solid and Liquid Sampling
One sample of each of the two carbon black product types
(N330 and N351) was collected for analysis of POM and trace
element content.
One sample of liquid feedstock was obtained during run POM-1.
The same shipment of feedstock was in use throughout the
test period.
A sample of the quench feedwater was also obtained for trace
element analysis. Prior to shipment to the laboratory, the
sample was acidified with pure concentrated nitric acid to a
pH between one and two.
2. ANALYTICAL METHODS
a. Particulate Matter and Trace Elements
The weight of particulate matter collected in the sampling
train probe and filter was determined for each of the three
trace metal runs as specified in EPA Method 5.57 After the
mass-loading determination, the particulate matter was
prepared for trace element analyses. The particulate matter
was ashed overnight in a Model 505 Low Temperature Asher
manufactured by LTE Process Company of Waltham, Massachusetts,
The ash was then digested in aqua regia and diluted to a
known volume with distilled water. This volume was split
into three fractions. -One fraction was submitted to Accu-
Labs Research, Inc., for spark source mass spectroscopy
163
-------
(SSMS) analysis; a second fraction was diluted with distilled
water to the proper acidity and salt concentration for
analysis by the Jarrel Ash Atomcomp technique using induc-
tively coupled argon plasma excitation (ICAP) at Monsanto
Company's St. Louis Laboratory; the third fraction was
analyzed for mercury, arsenic, and selenium by atomic absorp-
tion at MRC's Dayton Laboratory.
The material collected in the impingers was ashed, digested
and submitted for trace element analysis as above. Thus,
for each sampling run, characterization of the trace element
contents was conducted on two separate samples, i.e., probe
and filter combined, and the remainder of the sampling train
after the filter.
b. Polycyclic Organic Materials and Other Organic Compounds
(1) Sample Pretreatment - The POM and hydrocarbon samples
collected from the three POM runs were received from the
field in the following forms:
• Particulate samples from the probe washes
• Millipore Fluoropore type FA filter
• XAD-2 cartridge
• Contents of 10% KOH impinger including washings
• Contents of toluene impingers including washings
The XAD-2 trap was subjected to Soxhlet extraction with
pentane for 24 hours. The pentane was then reduced to
5 x 10"6 m3 (^5 ml) in volume by means of rotary evaporation.
During rotary evaporation, a pressure >10 mm of N£ and a
water 'oath temperature of <45°C were maintained.
164
-------
The 10% KOH impinger contents were extracted with methylene
chloride, and the extract volume was reduced by rotary evap-
oration to about 5 ml. In a similar manner, the toluene
impinger contents were reduced in volume to about 5 ml.
Following the volume reduction, the above three samples were
combined and separated into eight fractions on a silica gel
column using the solvent systems shown in Figures E-2 and E-3
Each fraction was then reduced in volume with a Kuderna-Danish
evaporator and transferred to a tare-weighed microweighing
pan, and the remaining solvent was evaporated in air. Each
dried fraction was weighed and then redissolved in a minimum
quantity of methylene chloride.
All fractions were then examined employing a Digilab Model
FTS-15B/D Fourier Transform Infrared (FTIR) Spectrometer to
characterize the full range of organic constituents present,
from the nonpolar aliphatic hydrocarbons to the polar sulfonic
acids. This analysis was either qualitative or semiquanti-
tative depending upon the complexity of the composition of
each fraction.
Following infrared analysis, the second, third, and fourth
fractions (containing the POM components) were diluted to
2 x 10"6 m3 (2 ml) with methylene chloride and transferred to
a Viton-septum sealed vial which was covered with aluminum
foil and refrigerated until required for analysis. Just prior
to analysis, the sample underwent one more volume reduction
via the Kuderna-Danish method. The final volume was approxi-
mately 5 x 10~7 m3 (500 yl) . This volume size has been found
to be optimum for detecting the POM peaks without obscuring
contamination peaks.
165
-------
•1:1
MeOH
Peritane
GC/MS Analysis
f
Fraction Elution Solvent Composition
1 Pentane
2 20% Methylene Chloride in Pentane
3 50* Methylene Chloride in Pentane
4 Methylene Chloride
5 5 * Methanol in Methylene Chloride
6 20% Methanol in Methylene Chloride
7 50% Methanol in Methylene Chloride
8 Cone. HCI / Methanol / Methylene
Chloride (5/70/30)
Volume Collected
25ml
10ml
10ml
10ml
10ml
10ml
10ml
10ml
nn
GC / MS Analysis
NTT IT II
Figure E-2. Schematic of sample pretreatment for POM analysis
-------
Figure E-3. Flow diagram for POM analysis1
167
-------
An identical procedure was followed for pretreatment of the FA
filter and the washings from the probe. Both particulate
samples were extracted with pentane, combined, reduced in
volume to ^5 x 10~6 m3 (^5 ml), and subjected to liquid
chromatographic separation on a silica gel column. After
volume reduction via Kuderna-Danish evaporator and micro-
weighing, Fourier Transform Infrared examination was conducted
prior to specific POM analysis of the second, third, and
fourth fractions. Thus, for each sampling run, characteriza-
tion of the organic and POM contents was conducted on two
separate samples, i.e., probe and filter combined, and the
remainder of the train after the filter.
(2) Analytical Procedure for POM - In 1975, MRC developed its
present analytical method for measuring POM content in envi-
ronmental samples. The method employs the peak-area quanti-
tation technique with computer reconstructed chromatograms
from the (HP 5982-A) GC-MS. All data are collected in the
electron impact (El) mode because of the abundance of avail-
able El-mass spectra. The chemical ionization (CI) mode of
operation, which may provide greater sensitivity, is also
available. MRC has found that GC-MS with a computer-con-
trolled data system is the optimum of analytical accuracy,
sensitivity, and speed.
The gas chromatographic separation is achieved using a 1.8-m
(6-ft) Dexsil 400 glass column with temperature programming
from 160°C for 2 min, rising to 280°C at 8°C/min, and becoming
isothermal at 280°C. The carrier gas is helium at a flow rate
of 0.5 x 10~9 m3/s (30 yl/min).
168
-------
The mass spectrometer, operating in the electron impact mode,
is programmed to scan the 75-350 AMU range as the POM com-
ponents elute from the gas chromatograph. The data system
will reconstruct the chromatogram using the total ion mode.
POM's are located by their molecular mass ions which are
displayed using the selected ion mode (SIM). Their identity
is confirmed by examining their mass spectra and retention
times. Samples and standards are run in SIM for quantitation.
SIM provides good stability, reproducibility, and sensitivity.
Calibration curves are prepared for each POM of interest using
varying concentrations of the POM standards in methylene
chloride, plotting mass ion peak area versus concentration,
and determining response factors. POM peaks in samples are
compared with standard curves that have been obtained under
the same conditions: attenuation, injection volume [2 x 10~9 m3
(2 yl)], and tuning condition. The tuning condition is
important since it controls the relative sensitivity of the
ions. Calibrations were made on the same day that the samples
were analyzed. Our present (Jan. 1977) detection limit for
POM's is ^0.5 g/m3 (^0.5 ng/yl) for a typical sample volume of
5 x 10~7 m3 (500 yl). The following POM's are routinely
analyzed at MRC:
Dibenzothiophene
Anthracene
Phenanthrene
Methylanthracenes
Methylphenanthrenes
9-Methylanthracene
Fluoranthene
Pyrene
Benzo[c]phenanthrene
Chrysene
Benz[a]anthracene
NR = Not resolved.
NR
NR
7,12-Dimethylbenz[a]anthracene
Benzo[b]fluoranthene
Benzo[a]pyrene (and isomers) - NR
3-Methylcholanthrene
Dibenz[a,b]anthracene
Indeno[1,2,3-cd]pyrene
Dibenzo[d,g]carbazole
Dibenzo[a,h]pyrene \
Dibenzota,i]pyrene /
NR
169
-------
Figuro E-4 shows a typical chromatographic separation of the
POM hydrocarbon standards. The top trace is a total ion (TI)
reconstructed chromatogram of the standards. The traces below
represent single ion chromatograms using the molecular mass
ion of each individual POM. Each peak represents a specific
POM hydrocarbon. The chromatogram for carbon black run POM-1
is shown in Figure E-5.
Table E-l shows the results of reproducibility studies which
have been conducted with POM standards at our laboratory.
The standard deviations are listed for four replicate injec-
tions at the ^0.2 g/m3 (^0.2 ng/yl) level for each POM sought.
As noted, the variability becomes greater as the detection
limits are approached. Table E-2 gives the results of a
recovery study made at our laboratory using six POM's at the
100 g/m3 (100 ng/yl) and 5 g/m3 (5 ng/yl) concentration
levels. Anthracene at high concentrations appears to pre-
sent minor analysis problems.
(3) Organic Compound Analysis - The higher molecular weight
hydrocarbons (above C6) and organo-sulfur compounds collected
by the sampling train were separated into eight fractions by
the column chromatographic separation shown in Figure E-5.
Each of the fractions was examined using the Digitab FTS-15B/D
Fourier Transform Infrared Spectrometer (FTIR). The results
of the FTIR analysis were used to determine whether the
individual fractions should be examined by the GC/MS system
to further identify the compounds present.
Organic compounds in the GI to Cg molecular weight range
were analyzed on site in the field, and no further labora-
tory analysis was done on these materials.
170
-------
SflMPLE 10318
178 < M < 302
8
o
CD
•v
09
I
UJ «0
rtne
SPECT
i'J
40
TI.
184.
178 J
ise.
202 _!_
SPECT
SO lOOlJOZOQT5OTOIJ35a4Cia430SOCB5()BOa65cnOa7Sa80l»5a900S3ai3UBajO;aDl5KaCeSI13003SIH3WJOSac6S08JOB;i'7JU750i3IBIJ4JI»
SO 100l50eOCES03003S040Q4SOSOUSS060a6Sinoa7S08038SCSOO»SaiOOBOSDinDlSIKtKE5D3nD3S040I»iasaiBS080IBSircn'5nBOIBSD93nSS
Figure E-4. Typical GC/MS chromatogram of POM hydrocarbon standards
-------
O
o
Tl
I I I I I I I
i i i i i i i i i i
10 15
TIME, Minutes
Figure E-5. Chromatogram for carbon black run POM-1
172
-------
Table E-l. REPRODUCIBILITY STUDY USING POM STANDARDS
POM
Dibenzothiophene
Anthracene/Phenanthrene
Methylanthracenes/Methylphenanthrenes
9-Methylanthracene
Fluoranthene
Pyrene
Benzo [ c ] phenanthrene
Chrysene/Benz [a] anthracene
7 , 12-Dimethylbenz [a] anthracene
Benzo [b] f luoranthene
Benzo [a] pyrene
3-Methylcholanthrene
Dibenz [a , b ] anthracene
Indeno [1,2, 3-cd] pyrene
Dibenzo[c,g] carbazole
Dibenzo [ a , h/a , i] pyrenes
Overall POM % Standard Deviation
% Standard
^5 g/m3
(^5 ng/yl)
cone.
3.4
3.7
5.9
7.7
5.8
5.3
3.0
2.9
3.3
2.5
9.2
5.2
1.7
0.5
3.4
3.2
4.2
deviation
^2 g/m3
(^2 ng/yl)
cone.
11.8
11.5
11.9
10.1
5.4
9.0
9.1
12.5
10.5
8.9
10.1
9.2
6.4
6.9
8.7
7.8
9.4
Table E-2. PROCESSING RECOVERY STUDY USING POM'S
POM
Anthracene
Chrysene
7 , 12-Dimethylbenz [a] anthracene
Benzo [a] pyrene
3-Methylcholanthrene
Dibenz [ a , b ] anthracene
% Recovered
^100 g/m3 ^5 g/m3
(^100 ng/yl) (^5 ng/yl)
cone. cone.
80.1
95.5
94.4
94.9
95.8
95.4
93.9
98.0
98.3
98.3
96.1
96.5
173
-------
c. H2S/ COS, CS2, and SO?
These compounds were analyzed in the field using gas chromato-
graphic separation and flame photometric detection as speci-
fied in EPA Method 15. 61
d. H2, N2, CC>2, and 02
The grab samples collected in the gas sampling flasks were
analyzed for H2, N2, C02, and 02 using mass spectroscopy.
e. Mitrogen Oxides
Nitrogen oxide emissions were analyzed following the pro-
cedure outlined in EPA Method 7.63'61*
f . Analyses of Solid and Liquid Samples
Solid and liquid samples collected for analysis consisted
of:
• Pelletized carbon black product (N330 and N351)
• Liquid carbon black feedstock
• Quench feedwater
Each of these samples was analyzed for trace elements. The
two carbon black product samples were also analyzed for POM
content.
The samples were prepared for trace element analysis as
follows: five grams of each carbon black product and five
grams of liquid feedstock were ashed in the LTE Low Tempera
ture Asher described previously. The ash from each sample
174
-------
was digested in aqua regia and diluted to a known volume
(250 ml) with distilled water. The quench water sample was
boiled to remove any organic material and then diluted to
the same volume (250 ml) with distilled water. Each of the
samples was then split into three fractions. One fraction
of each sample was sent to Accu-Labs, Inc., for analysis by
SSMS. A second fraction of each sample was sent to Monsanto
Company's St. Louis Laboratory for analysis by ICAP. The
third fraction of each sample was analyzed for mercury,
arsenic, and selenium by atomic absorption at the Dayton
Laboratory.
The POM content of the carbon black samples was analyzed by
the GC/MS approach as described in E.2.b. The sample prep-
aration prior to analysis was similar to the technique used
for particulate samples. The carbon black was ground and
then extracted (Soxhlet) with pentane for 24 hours. The
extracted material was separated into fractions as shown in
Figure E-3. Fractions of 2 to 4 were reduced in volume as
previously described, and analyzed for POM by the GC/MS
method.
g. Carbon Monoxide in Flare Off-Gas
The measurement of carbon monoxide in the flare off-gas is
discussed in Appendix H.
175
-------
APPENDIX F
SAMPLING RESULTS
This appendix presents the results of a field sampling program
conducted at a representative carbon black plant. Sampling
was conducted on the main process vent gas. A total of six
sampling runs was made, consisting of three trace element runs
(1-TM, 3-TM and 4-TM) and three polycyclic organic material
(POM) runs (POM-1, POM-2 and POM-3). Two blank runs (5-TM and
POM-4) were also made to check the consistency of the sampling
and analytical procedures. Run 2-TM was discarded during the
test program due to a mechanical difficulty with the sampling
apparatus which arose during testing. Samples of oil feed-
stock, quench water, and carbon black product were also
analyzed.
Tables F-l, F-2 and F-3 present the emissions data for trace
metal runs 1-TM, 3-TM and 4-TM, respectively. Tables F-4,
F-5 and F-6 present the emissions data for polycyclic organic
material runs POM-1, POM-2 and POM-3.
Tables F-7 and F-8 present a comparison of trace element runs
with blank run 5-TM for the spark source mass spectroscopic
(SSMS) and inductively coupled argon plasma (ICAP) analyses.
For a number of elements, the concentrations in the blank
sample were as great or greater than the concentrations in
the run samples. For these elements, the emission factors
listed in Tables F-l, F-2 and F-3 are not the actual emission
176
-------
factors, but represent upper bounds on the actual emission
factors for these materials.
One possible source of anomalous values in the blank run is
the glass containers used to store the samples. Silicon,
sodium, calcium, magnesium, and aluminum are main constituents
of glass. The reason for the high values of other elements,
such as phosphorus, potassium, zinc, and iron is not known.
These elements may have been present as impurities in reagents
used in the sampling train or in those used to clean the glass-
ware. They might also have been introduced during preparation
of the samples for analysis.
A blank POM run (POM-4) was made and analyzed for POM content.
No POM's were detected in the blank.
177
-------
Table F-l. EMISSIONS DATA FROM RUN 1-TM
I-1
-j
oo
Emission
Inert gases
Nitrogen
Oxygen
Carbon dioxide
Water
Criteria pollutants
"^Particulate matter
"^'Sulfur oxides
" Nitrogen oxides
Hydrocarbons
Carbon monoxide
Chemical substances
Hydrogen
.-Hydrogen sulfide
Carbon disulfide
Carbonyl sulfide
Methane
Acetylene
Ethylene
Propylene
Propane
Isobutane
n-Butane
Trace elements
Aluminum
Antimony
Arsenic
Barium
Beryllium
Bismuth
Boron
Bromine
Concentration ,
mn/m 3
837,710
15,700
70,185
477,150
, 7.4
_a
21
4,480
112,800
9,735
2,560
4,510
1,010
2,450
4,480
_a .
a
~a
~a
0.17 {0.19}
3.6 x 10- 3 {7.9 x 10~ 3>
<1.2 x ID"3 {3.5 x 10"3}
0.273 {8.1 x 10"2>
<1.8 x 10~5
<9.3 x ID"11
{0.29}
7.5 x 1Q-1*
Emission factor,
g/kg
10,770
200
900
6,135
, 0.
a
0.
55
1,470
127
33
58
13
32
_a55
a
a
_a
2.
4.
1.
3.
<2.
<1.
9.
09
27
1 x 10"3 {2.4 x 10~3}
6 x 1Q-5 {1.0 x 10-M
7 x 10"5 {4.5 X 10~5}
5 x 10~3 {1.0 x ID'3}
4 x 10~6
2 x 10~ 5
{3.8 x 10"3}
7 x 10~6
(continued)
-------
Table F-l (continued). EMISSIONS DATA FROM RUN 1-TM
VD
Emission
Cadmium
Calcium
Cerium
Cesium
Chlorine
Chromium
Cobalt
Copper
Dysprosium
Erbium
Europium
Fluorine
Gadolinium
Gallium
Germanium
Gold
Hafnium
Holmium
Iodine
Iridium
Iron
Lanthanum
Lead
Lithium
Lutecium
Magnesium
Manganese
Mercury
Molybdenum
Neodymium
Nickel
Niobium
Osmium
Palladium
Phosphorus
Concentration ,
mg/m3
1.5 x 10~3
3.9
1.8 x 10~4
<9.2 x 10" 5
0.47
5.4 x 10~2
1.9 x 10"2
1.8
<1.9 x 10"4
7.5 x 10"4
<2.8 x 10~4
6.4 x 10~2
<6.5 x 10 4
<9.4 x 10"5
<1.9 x 10~4
<1.9 x 10~4
<1.2 x 10" 3
<1.9 x 10"4
<2.8 x 10"4
<2.8 x 10~4
0.47
2.8 x 10 4
5.7 x 10~2
2.1 x 10~3
<1.9 x 10~4
2.5
1.7 x 10~2
7.7 x 10" 3
<3.7 x 10"4
0.13
<2.8 x 10~4
<3.7 x 10~4
<2.8 x 10~4
1.3
{4.0 x 10"3}
{5.0}
{1.2 x 10~2}
{3.7 x 10~3}
{0.41}
{ 0. 31}
{5.4 x 10~2}
*j _
{4.5 x 10~3}
{1.4 x 10-M
{7.7 x 10" 3}
{4.4 x 10~2}
{0.13}
Emission factor,
g/kg
2,0 x 10~5
5.0 x 10~2
2.4 x 10~6
1.2 x 10~6
6.1 x 10~3
7.0 x 10"4
2.4 x 10~5
2.4 x 10"2
<2.4 x 10"6
9.7 x 10~6
<3.7 x 10"6
8.3 x 10"4
<8.6 x 10~6
<1.2 x 10~6
<2.4 x 10~6
<2.4 x 10~6
<1.6 x 10~5
<2.4 x 10~6
<3.7 x 10"6
<3.7 x 10~6
6.1 x 10" 3
3.7 x 10~6
7.4 x ID'4
2.8 x 10~5
<2.4 x 10~6
3.3 x 10~2
2.2 x 10~4
1.0 x 10~3
<4.8 x 10~6
1.6 x 10" 3
<3.7 x 10"6
<4.9 x 10=6
<3.7 x 10~6
1.7 x 10~2
{5.2 x 10~5}
{6.5 x 10~2}
{1.6 x 10"4}
{5.0 x 10"5}
{5.3 x 10" 3}
{4.0 x 10"3}
{7.0 x 10~4}
{1.8 x 10~ 3}
e C. -\
{5.8 x 10 6}
f •• C •<
{1.8 x 10 6}
{1.0 x 10"4}
{5.7 x ID'4}
{1.7 x 10"3}
(continued)
-------
Table F-l (continued). EMISSIONS DATA FROM RUN 1-TM
CO
o
Emission
Platinum
Potassium
Praseodymium
Rhodium
Rubidium
Samarium
Scandium
Selenium
Silicon
Sodium
Strontium
Tantalum
Terbium
Tellurium
Thallium
Thorium
Thullium
Tin
Titanium
Tungsten
Uranium
Vanadium
Ytterbium
Yttrium
Zinc
Zirconium
TOTAL Trace Elements
Concentration ,
mg/m3
<4.7 x 1Q-1*
3.6
<2.8 x lO'4
9.4 x 10-5
8.4 x 10-1*
<1.3 x 10~3
<2.8 x 10"1*
<1.8 x 10~4
0.77
2.7
7.2 x 10~3
<6.5 x 10~4
<9.4 x 10~5
<1.9 x 1Q-4
<3.7 x 10""*
9.4 x 10~4
<2.8 x ID"4
8.1 x 10~2
6.5 x 10" 3
1.7 x 10~3
<5.6 x 10"1*
1.6 x 10" 3
8.4 x lO"4
7.5 x 10-1*
5.9
6.2 x 10~ 3
22.6 {
(9.4 x 10~5}
{0.34}
(1.4>
(5.7 x 10~ 3}
{4.6 x 10" 2>
{4.2 x 10" 3>
{5.1 x 10"3}
{0.81}
14.2}
Emission f actor ,-
g/kg
<5.6 x 10~6
4.6 x 10"2
<3.7 x 10"6
1.2 x 10-6
1.1 x 10"5
<1.7 x 10"5
<3.7 x 10"6
<2.4 x 10~6
1.0 x 10 2
3.5 x 10"2
9.4 x 10~ 5
<8.6 x 10"6
<1.2 x 10-e
<2.4 x 10"6
<4.8 x 10"6
1.2 x 10"5
<3.7 x 10~6
1.0 x 10" 3
8.5 x 10~5
2.2 x 10"5
<7.3 x 10"6
2.1 x 10"5
1.1 x 10~5
9.7 x 10~6
7.7 x 10~2
8.1 x 10"5
0.3
{1.2 x
{4.4 x
{1.8 x
{7.3 x
{5.9 x
(5.4 x
{6.6 x
{1.0 x
{0.16}
10~6>
10~3}
ID'2}
10~5}
10-" >
io~5}
10~5}
10~2}
Not detected at 1 ppm. ~
^Values in braces (•{}) were determined by ICAP with the exception of those for arsenic,
mercury, and selenium, which were determined by atomic absorption. Values not in
braces were determined by SSMS.
CTotals do not include numbers recorded as upper bounds.
-------
Table F-2. EMISSIONS DATA FROM RUN 3-TM
00
Emission
Inert gases
Nitrogen
Oxygen
Carbon dioxide
Water
Criteria pollutants
Particulate matter
Sulfur oxides
Nitrogen oxides
Hydrocarbons
Carbon monoxide
Chemical substances
Hydrogen
Hydrogen sulfide
Carbon disulfide
Carbonyl sulfide
Methane
Acetylene
Ethylene
Propylene
Propane
Isobutane
n-Butane
Trace elements
Aluminum
Antimony
Arsenic
Barium
Beryllium
Bismuth
Boron
Bromine
Concentration ,
mg/m3
808,500
19,630
71,980
473,760
15.2
a
29.9
3,840
140,860
9,490
1,220
1,775
600
2,025
3,655
138
_3
22
10
26
0.28 {0.38}
1.6 x 10~3 {2.6 x 10~2}
<1.3 x 10~2 {4.7 x 10~3}
0.35 {0.14}
<2.3 x 1Q-1*
<4.6 x 10"1*
{0.68}
6.8 x 10" "»
Emission factor,
g/kg
7,283
175
650
4,270
. 0.14
Q
0.27
37
1,270
85
11.2
16
5.05
20
35
_a 1'2
0.20
0.09
0.24
2.5 x
1.4 x
<1.2 x
3.1 x
<2. 1 x
<4.1 x
6.2 x
10~3 {3.4 x 10"3}
10~5 {2.3 x ID""*}
10"14 {4.3 x 10~5}
10~3 {1.2 x 10~3}
10~6
10~5
{6.1 x 10~ 3}
10~6
(continued)
-------
Table F-2 (continued). EMISSIONS DATA FROM RUN 3-TM
CO
Emission
Cadmium
Calcium
Cerium
Cesium
Chlorine
Chromium
Cobalt
Copper
Dysprosium
Erbium
Europium
Fluorine
Gadolinium
Gallium
Germanium
Gold
Hafnium
Holmium
Iodine
Iridium
Iron
Lanthanum
Lead
Lithium
Lutecium
Magnesium
Manganese
Mercury
Molybdenum
Neodymium
Nickel
Niobium
Osmium
Palladium
Phosphorus
Concentra
mg/m3
1
5
4
<2
1
3
1
0
<2
1
<4
0
3
<2
<2
<2
<1
<2
<2
<4
0
2
0
6
<2
1
7
2
<4
0
<2
<6
<4
6
.4 x
.1
.6 x
.3 x
.9 x
.3 x
.8 x
.3
.3 x
.1 x
.6 x
.27
.0 x
.3 x
.3 x
.3 x
.6 x
.3 x
.3 x
.6 x
.61
. 3 x
.17
. 8 x
.3 x
.0
.8 x
.2 x
.6 x
.16
.3 x
.9 x
.6 x
.7 x
10"3
ID'"
10-"
io-2
10- 2
ID' 2
10~"
10" 3
10-"
ID"3
10-"
10-"
10-"
10-"
10-"
10-"
10-"
10-"
10-"
10-"
io-3
10"2
10-"
10-"
10-"
10-"
lO-2
tior.
{6.
{9.
{1.
{1.
{0.
{0.
{9.
{0.
{7.
{4.
{1.
{0.
{0.
t
9 x 10"3}
8}
4 x 10"2}
1 x 10"2}
1}
33}
4 x 10"2}
54}
1 x 10"3}
7 x 10"2}
6 x 10~2}
17}
16}
Emission factor
g/kg
1.2
4.6
4.1
<2 . l
1.8
3.0
1.7
2.7
<2 .1
1.0
4.1
2.4
2.7
<2. 1
<2. 1
<2. 1
<1. 4
<2.1
<2.1
<4. 1
5.5
2.1
1.6
6.2
<2. 1
9.4
7.0
2.0
1.5
<2. 1
<6.2
<4. 1
6.1
x
x
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
X
io-5
10"2
ID'6
io-6
10-"
10-"
10-"
lO'3
10"6
io-5
io-6
10~3
10-"
io-6
ID'6
io-6
io-5
10~6
io-6
io-6
lO'3
io-6
10- 3
10~6
io-6
io:3
10 5
10""
_ ^ — c
10_6
10 3
10~6
10"6
10- |j
10
{6.2
{8.9
{1.2
{1.0
{9.5
{3.0
{8.5
{4.9
{6.4
{4.2
{1.4
{1.5
{1.4
1
X
X
X
X
X
X
X
X
X
X
X
X
X
io-5}
ID"2}
10-"}
10-"}
10-"}
10~3}
10-"}
10- 3}
10" }
10""}
10""}
10"3}
10" 3}
(continued)
-------
Table F-2 (continued). EMISSIONS DATA FROM RUN 3-TM
oo
u>
Emission
Platinum
Potassium
Praseodymium
Rhodium
Rubidium
Samarium
Scandium
Selenium
Silicon
Sodium
Strontium
Tantalum
Terbium
Tellurium
Thallium
Thorium
Thullium
Tin
Titanium
Tungsten
Uranium
Vanadium
Ytterbium
Yttrium
Zinc
Zirconium
TOTAL Trace Elements0
Concentration ,
-mg/m3
<6.9 x icr4
1.4
<2.3 x lO'4
<2.3 x 10"1*
1.1 x 10~3
<2.3 x 10"1*
<4.6 x 1Q-1*
<2.3 x 10"1*
1.4
4.4
9.8 x 10- 3
<2.3 x 1Q-1*
<2.3 x 10"1*
<2.3 x 10"1*
<4.6 x 1CT1*
1.1 x 10" 3
<2.3 x 10"1*
0.15
3.1 x 10~2
6.8 x 10~3
<2.3 x 10"1*
4.6 x 10-"
6.9 x 10-4
6.9 x 10""
2.5
1.3 x 10~2
18.4 {
{2.3 x 10"1*}
{0.68}
{2.4}
{9.6 x 10~3}
{6.9 x ID"2}
{9.8 x 10~3}
{1.4 x ID"2}
{0.16}
17.5}
Emission factor,
q/kcr
6.2 x 10~6
1.3 x ID"2
<2.1 x 10"6
<2.1 x 10"6
1.0 x 10"5
<2.1 x 10"6
<4.1 x 10~6
<2.1 x 10~6
1.3 x 10~2
3.9 x 10~2
8.9 x 10~5
<2.1 x 10~6
<2.1 x 10~6
<2.1 x 10 6
<4.1 x 10~6
1.0 x 10~5
<2.1 x 10~6
1.4 x 10~3
2.9 x 10"1*
6.2 x 10~6
<2.1 x 10~6
4.1 x 10~6
6.2 x 10~6
6.2 x 10"6
2.3 x 10~2
1.2 x lO"4
0.17
{2.0 x
{6.2 x
{2.1 x
{8.7 x
{6.2 x
{8.9 x
{1.2 x
{1.4 x
{0.16}
10~6}
10~3}
ID"2}
10~5}
10~6}
10~5}
ID"1*}
10~3}
Not detected at 1 ppm.
Values in braces ({}) were determined by ICAP with the exception of those for arsenic,
mercury, and selenium, which were determined by atomic absorption. Values not in
braces were determined by SSMS.
Totals do not include numbers recorded as upper bounds.
-------
Table F-3. EMISSIONS DATA FROM RUN 4-TM
Concentration,
Emission factor,
oo
4_*Iitj.33 j.OIi
Inert gases
Nitrogen
Oxygen
Carbon dioxide
Water
Criteria pollutants
Particulate matter
Sulfur oxides
Nitrogen oxides
Hydrocarbons
Carbon monoxide
Chemical substances
Hydrogen
Hydrogen sulfide
Carbon disulfide
Carbonyl sulfide
Methane
Acetylene
Ethylene
Propylene
Propane
Isobutane
n-Butane
Trace elements
Aluminum
Antimony
Arsenic
Barium
Beryllium
Bismuth
Boron
Bromine
mg/m 3
930,580
22,250
62,985
528,260
=, 7.5
_a
24
3,190
138,570
b
2,085
3,390
950
1,785
3,180
_a63
10
5
17
0.15 {0.27}
9.5 x lO""" {5.5 x 10 2>
<2.3 x 10~3 {5.0 x 10~3}
6.1 x 10" 3 {0.10}
<1.9 x 10~1*
<1.5 x 10~3
{0.7}
7.6 x 10~*
g/kg
10,460
250
710
5,940
, 0.083
_a
0.27
35
1,560
b
23
30
9
20
35
_a °'71
0.12
0.06
0.19
1.7 x
1.1 x
<2.5 x
6.8 x
<2.2 x
<1.8 x
8.9 x
10~3 {3.0 x 10~3}
10~5 {6.2 x 10-"}
10~5 {5.6 x 10"6}
10~5 {1.2 x 10~3}
10~6
10" 5
{7.8 x 10~ 3}
10~5
(continued)
-------
Table F-3 (continued). EMISSIONS DATA FROM RUN 4-TM
CO
Emission
Cadmium
Calcium
Cerium
Cesium
Chlorine
Chromium
Cobalt
Copper
Dysprosium
Erbium
Europium
Fluorine
Gadolinium
Gallium
Germanium
Gold
Hafnium
Holmium
Iodine
Iridium
Iron
Lanthanum
Lead
Lithium
Lutecium
Magnesium
Manganese
Mercury
Molybdenum
Neodymium
Nickel
Niobium
Osmium
Palladium
Phosphorus
Concentration ,
mg/m3
5.7 x lO"1*
2.3
<3.8 x ICT1*
<1.9 x 10"1*
0.72
1.7 x 10~2
3.4 x 10~3
0.4
<1.9 x ID"1*
7.6 x IQ-1*
<5.7 x 10-1*
0.25
2.5 x 10~3
<1.9 x 10"1*
<1.9 x 10"1*
<1.9 x lO"1*
1.5 x ID"3
<1.9 x IQ-*
<5.7 x 10"1*
<3.8 x 10-"1
0.18
3.8 x 10"1*
0.19
<1.9 x 10" **
<3.8 x ID"1*
1.8
1.0 x 10~2
1.0 x 10~2
<3.8 x 10"1*
6.6 x 10~2
<5.7 x 10~"
<5.7 x lO"4
<3.8 x lO"4
0.19
{2.8 x 10~3}
{3.5}
{2.3 x 10~3>
{7.2 x 10~3)
{0.23}
(0.43}
(5.5 x 10~2}
{0.38}
{7.5 x 10~3}
{2.7 x 10~3}
{9.4 x 10~3}
{0.11}
{0.13}
Emission factor,
g/kg
8.5 x 10~6
2.7 x 10~2
<4.5 x 10~6
<2.2 x 10~6
1.2 x 10~3
2.0 x lO"4
4.0 x 10~5
4.7 x 10~3
<2.2 x 10~6
8.9 x 10~6
<6.7 x 10 6
2.9 x 10~3
2.9 x 10~5
<2.2 x 10~6
<2.2 x 10~6
<2.2 x 10~6
1.8 x 10~5
<2.2 x 10~6
<6.7 x 10~6
<4.5 x 10~6
2.1 x 10 3
4.5 x 10 6
2.2 x 10~3
<2.2 x 10~6
<4.5 x 10~6
2.1 x 10~2
5.3 x ID"4
5.3 x lO"4
<4.5 x 10~6
7.7 x 10~"*
<6.7 x 10~6
<6.7 x 10~6
<4.5 x 10~6
2.3 x 10~3
{3.2 x 10~5}
{4.0 x 10~2}
{2.6 x 10"5}
{8.1 x 10"f>
t2.6 x 10~3}
{4.7 x 10~3}
{6.2 x ID'1*}
{4.3 x 10~3}
{8.5 x 10~5}
{3.1 x 10~5}
{7.0 x 10"")
{1.2 x 10~3}
{1.5 x 10~2}
(continued)
-------
Table F-3 (continued). EMISSIONS DATA FROM RUN 4-TM
CO
Emission
Platinum
Potassium
Praseodymium
Rhodium
Rubidium
Samarium
Scandium
Selenium
Silicon
Sodium
Strontium
Tantalum
Terbium
Tellurium
Thallium
Thorium
Thullium
Tin
Titanium
Tungsten
Uranium
Vanadium
Ytterbium
Yttrium
Zinc
Zirconium
TOTAL Trace elements
Concentration ,
mg/m°
<5.7 x 10"1*
2.3
<3.8 x 10"1*
<1.9 x 10-"
3.2 x 10~ 3
<1.9 x 10"3
<3.8 x 1Q-1*
<3.8 x lO"1*
2.9
4.3
5.7 x 10" 2
<9.5 x 10"1*
<1.9 x lO"1*
<1.9 x 10"1*
<5.7 x ICT1*
1.5 x 10"3
<3.8 x 10"1*
0.13
1.3 x 10" 2
2.3 x 10" 3
<9.5 x 10""
7.6 x 10"1*
7.6 x 10"*
3.8 x 10" "*
3.4
6.6 x 10" 3
19.4
{1.9 x 10"1*}
{1.2}
{2.3}
{1.3 x 10"2}
{0.13}
{8.7 x 10" 3}
{9.4 x 10"3}
{0.37}
{14.2}
Emission factor,
q/kq
<6.7 x 10~6
2.7 x 10~2
<4.5 x 10~6
<2.2 x 10"6
3.8 x 10" 5
<2.2 x 10~6
<4.5 x 10~6
<4.5 x 10"6
3.5 x 10"2
5.1 x 10"2
6.7 x 10"1*
1.1 x 10" 5
<2.2 x 10"6
<2.2 x 10~6
<6.7 x 10~6
1.8 x 10" 5
<4.5 x 10~6
1.5 x 10~3
1.6 x 10"1*
2.7 x 10" 5
<1.1 x 10~5
8.9 x 10~6
8.9 x 10~6
<4.5 x 10~6
4.0 x 10~2
7.8 x 10~5
0.22
{2.0 x
{1.4 x
{2.6 x
{1.5 x
{1.5 x
{1.0 x
{1.0 x
{4.1 x
{0.15}
10~6}
10~2}
10"2}
10~M
10~3}
10~M
10"5}
10"3}
Not detected at 1 ppm.
Sample thrown out; too low.
cValues in braces ({}) were determined by ICAP with the exception of those for arsenic,
mercury, and selenium, which were determined by atomic absorption. Values not in
braces were determined by SSMS.
Totals do not include numbers recorded as upper bounds.
-------
Table F-4. EMISSIONS DATA FROM RUN POM-1
oo
Emission
Inert gases
Nitrogen
Oxygen
Carbon dioxide
Water
Criteria pollutants
Particulate matter
Sulfur oxides
Nitrogen oxides
Hydrocarbons
Carbon monoxide
Chemical substances
Hydrogen
Hydrogen sulfide
Carbon disulfide
Carbonyl sulfide
Methane
Acetylene
Ethylene
Propylene
Propane
Isobutane
n-Butane
POM
Acenaphthylene
Anthracene/phenanthrene
Benzo [c] phenanthrene
Benzof luoranthenes
Benzo [ghi] f luoranthene
Concentration /
mg/m3
778,730
70,675
52,190
581,035
~K
U
25
3,762
104,210
11,940
2,055
2,690
785
1,825
3,575
146 .
U
15
7
26
4.7 x 10~2
3.5 x 10 3
2.2 x ID""
4.4 x 10~3
2.6 x 10~3
Emission factor,
q/kg
7,630
690
510
5,695
~h
U
0.25
37
1,020
117
20
26
8
18
35
^ M
_D
0.15
0.07
0.25
4.7 x 10"1*
3.5 x 10~5
2.2 x 10~6
4.3 x 10~5
2.6 x 10~5
(continued)
-------
Table F-4 (continued). EMISSIONS DATA FROM RUN POM-1
oo
00
Emission
Benzo [ghijperylene/anthanthrene
Benzopyrenes/perylene
Chrysene/benz [a] anthracene
Dibenz anthracenes
Dibenzo [c,g]carbazole
Dibenzopyrenes
Dibenzothiophene
Dimethylanthracene/phenanthrene
7 , 12-Dimethylbenz [a] anthracene
Fluoranthene
Indeno [ 1 , 2 , 3-cd] pyrene
Methylanthracene
Methylcholanthrene
Methylf luoranthene
Pyrene
TOTAL POMC
Concentration ,
mg/m3
1.3 x 10~3
3.1 x 10~ 3
8.7 x 10"^
<1.1 x 10"1*
<1.1 x lO"4
<1.1 x 10"4
3.3 x lO"4
8.4 x 10" 3
1.1 x 10~2
3.1 x ID" 3
<1.1 x 10~4
7.5 x 10"3
<1.1 x lO""
2.3 x 10~3
2.6 x 10"2
1.2 x 10"1
Emission factor,
q/kq
1.3 x 10~5
3.0 x 10"5
8.5 x 10"6
<1.0 x 10"6
<1.0 x 10"6
<1.0 x 10"6
3.3 x 10~5
8.2 x 10"5
1.1 x ID'1*
3.0 x 10~ 5
<1.0 x 10~6
7.4 x 10~5
<1.0 x 10~6
2.2 x 10" 5
2.6 x lO'1*
1.2 x 10~3
QNot determined.
Not detected at 1 ppm.
CTotal does not include numbers recorded as upper bounds.
-------
Table F-5. EMISSIONS DATA FROM RUN POM-2
oo
VD
Emission
Inert gases
Nitrogen
Oxygen
Carbon dioxide
Water
Criteria pollutants
Particulate matter
Sulfur oxides
Nitrogen oxides
Hydrocarbons
Carbon monoxide
Chemical substances
Hydrogen
Hydrogen sulfide
Carbon disulfide
Carbonyl sulfide
Methane
Acetylene
Ethylene
Propylene
Propane
Isobutane
n-Butane
POM
Acenaphthylene
Anthracene/phenanthrene
Benzo [c] phenanthrene
Benzof luoranthenes
Benzo [ghi] f luoranthene
Concentration ,
mg/m3
794,800
6,020
64,785
605,970
a.
^ •
_b
27
5,997
135,130
11,950
4,775
12,420
2,375
3,028
5,705
237 b
U
24
13
31
1.0 x 10"1
9.4 x 10~3
1.7 x 10"^
3.1 x 10~3
5.2 x 10 3
Emission factor,
g/kg
10,505
80
855
8,010
_
~ r
b
0.35
79
1,785
160
63
164
31
40
75
3-jL
_D
0.47
0.18
0.41
1.4 x 10~3
1.2 x ID"1*
2.2 x 10~6
4.1 x 10~5
6.8 x 10~5
(continued)
-------
Table F-5 (continued). EMISSIONS DATA FROM RUN POM-2
Emission
Benzo [ ghi ] pery lene/anthanthrene
Benzopyrenes/perylene
Ghrysene/benz [a] anthracene
Dibenzanthracenes
Dibenzo[c,g] carbazole
Dibenzopyrenes
Dibenzothiophene
Dime thy lanthracene/phenanthrene
7,12-Dimethylbenz [a] anthracene
Fluoranthene
Indeno [1,2, 3-cd]pyrene
Methylanthracene
Methylcholanthrene
Methyl f luoranthene
Pyrene
TOTAL POMC
Concentration ,
mg/m3
2.8 x 10~3
2.8 x 10~3
7.9 x 10"1*
<1.6 x 10"1*
<1.6 x 10"1*
<1.6 x iQ-*
1.8 x 10~3
1.7 x 10~2
4.2 x 10~3
7.9 x 10~3
<1.6 x 10~lt
1.1 x 10~2
<1.6 x 10~4
2.5 x 10~3
5.6 x ID'2
2.2 x ID"1
Emission factor,
g/kg
3.8 x 10~5
3.8 x 10~5
1.0 x 10~5
<2.1 x 10~6
<2.1 x "10~6
<2.1 x 10~5
2.4 x 10"5
2.2 x 10"1*
5.6 x 10~5
1.1 x lO"1*
<2.1 x 10~6
1.4 x 10"4
<2.1 x 10~6
3.3 x 10~5
7.4 x 10"1*
3.0 x 10~3
Not determined.
Not detected at 1 ppm.
•>
"Total does not include numbers recorded as upper bounds.
-------
Table F-6. EMISSIONS DATA FROM RUN POM-3
Emission
Inert gases
Nitrogen
Oxygen
Carbon dioxide
Water
Criteria pollutants
Particulate matter
Sulfur oxides
Nitrogen oxides
Hydrocarbons
Carbon monoxide
Chemical substances
Hydrogen
Hydrogen sulfide
Carbon disulfide
Carbonyl sulfide
Methane
Acetylene
Ethylene
Propylene
Propane
Isobutane
n-Butane
POM
Acenaphthylene
Anthracene/phenanthrene
Benzo [c] phenanthrene
Benzof luoranthenes
Benzo [ghi] f luoranthene
Concentration ,
mg/m3
794,800
6,020
64,785
589,690
a
~b
24_a
135,130
a
a
a
a
a
a
a
~a
6.7 x 10~2
4.6 x 10~3
1.6 x 10""
1.2 x 10~3
9.9 x 10"1*
Emission factor,
g/kg
7,880
60
665
6,050
a
~b
°*§5
_ D
1,385
_a
~a
~a
a
_a
~a
_a
a
_a
a
~a
6.9 x 10"1*
4.8 x 10 5
1.6 x 10~6
1.3 x 10~5
1.0 x 10~5
(continued)
-------
Table F-6 (continued). EMISSIONS DATA FROM RUN POM-3
10
NJ
Emission
Benzo [ghilperylene/anthanthrene
Benzopyrenes/perylene
Chrysene/benz [a] anthracene
Dibenzanthracenes
Dibenz [c,g] carbazole
Dibenzopyrenes
Dibenzothiophene
Dime thy Ian thracene/phenanthrene
7 , 12-Dimethylbenz [a] anthracene
Fluoranthene
Indeno [1,2, 3-cd]pyrene
Methylanthracene
Methylchlolanthrene
Me thy If luoranthene
Pyrene
TOTAL POMC
Concentration ,
mg/m3
1.9 x 10~3
8.6 x 10""
6.9 x 10""
<1.5 x 10~*
<1.5 x 10~k
<1.5 x 10~4
1.5 x 10~3
1.1 x 10~2
4.7 x 10~3
3.7 x 10~3
<1.5 x 10~4
6.7 x 10~3
<1.5 x 10~1+
1\ 4 x 10" 3
3.7 x 10~2
1.4 x 10""1
Emission factor,
q/ka
1.9 x 10~5
8.8 x 10~6
7.1 x 10~6
<1.6 x 10~6
<1.6 x 10~6
<1.6 x 10~6
1.5 x 10~5
1.1 x ICT*
4.8 x 10~5
3.8 x 10~5
<1.6 x 10~6
6.9 x 10 5
<1.6 x 10~6
1.4 x 10~5
3.7 x 10 *
1.5 x 10~3
Not determined.
Not detected at 1 ppm.
*
'Total does not include numbers recorded as upper bounds.
-------
Table F-7. COMPARISON OF TRACE ELEMENT RUNS WITH
BLANK RUN: SSMS ANALYSIS
+ C2, mg/£)a
Element
Aluminum
Antimony
Arsenic**
Barium
Beryllium
Bismuth
Bromine
Cadmium
Calcium
Cerium
Cesium
Chlorine
Chromium
Cobalt
Copper
Dysprosium
Erbium
Europium
Fluorine
Gadolinium
Gallium
Germanium
Gold
Hafnium
Holmium
Iodine
Iridium
Iron
Lanthanum
Lead
Lithium
Lutecium
Magnesium
Manganese
Mercury"
Molybdenum
Neodymium
Nickel
Niobium
Osmium
Palladium
Phosphorus
Platinum
Run
1-TM
1.8
0.039
0.038
2.9
<0.003
<0.011
0.008
0.016
42
0.003
<0.002
5.0
0.57
0.11
20
<0.003
<0.008
<0.004
0.68
<0.007
<0.002
<0.003
<0.003
<0.013
<0.003
<0.004
<0.004
5.0
0.003
0.61
0.023
<0.003
27
0.18
0.0015
0.083
<0.004
1.4
<0.004
<0.005
<0.004
14
<0.005
3-TM
1.2
0.008
0.021
1.5
<0.002
<0.003
0.003
0.007
22
0.003
<0.002
0.09
0.15
0.081
1.3
<0.002
<0.005
<0.003
1.2
<0.013
<0.002
<0.002
<0.002
<0.007
<0.002
<0.002
<0.003
2.7
0.002
0.76
0.004
<0.002
4.6
0.034
0.205
0.097
<0.003
0.72
<0.002
<0.004
<0.003
0.30
<0.004
4-TM
0.8
0.005
0.027
0.03
<0.002
<0.009
<0.005
0.004
12
0.003
<0.002
3.9
0.091
0.019
2.1
<0.002
<0.005
<0.004
1.3
<0.013
<0.002
<0.002
<0.002
<0.008
<0.002
<0.004
<0.003
0.95
<0.003
0.98
NRC
<0.003
9.4
0.055
0.011
0.054
<0.003
0.35
<0.004
<0.003
<0.003
1.0
<0.004
5-TM
(blank)
2.5
<0.005
0.015
1.4
<0.003
<0.011
<0.006
<0.003
79
<0.003
0.005
14.2
0.076
0.005
0.51
<0.002
<0.007
<0.004
0.90
<0.018
<0.002
<0.003
<0.003
<0.010
<0.003
<0.004
<0.004
2.2
< 0.003
0.29
0.004
<0.003
2.7
0.073
<0.0004
0.04
<0.003
0.09
<0.004
<0.005
<0.004
13
<0.005
(continued)
193
-------
Table F-7 (continued). COMPARISON OF TRACE ELEMENT RUNS
WITH BLANK RUN: SSMS ANALYSIS
Element
Potassium
Praseodymium
Rhodium
Rubidium
Samarium
Scandium.
Selenium
Silicon
Sodium
Strontium
Tantalum
Terbium
Tellurium
Thallium
Thorium
Thullium
Tin
Titanium
Tungsten
Uranium
Vanadium
Ytterbium
Yttrium
Zinc
Zirconium
Run
1-TM
38.2
<0.004
0.002
0.009
0.014
<0.004
<0.002
8.3
29
0.077
<0.017
<0.002
<0.003
<0.005
<0.011
<0.004
0.87
0.070
<0.018
<0.007
0.017
<0.009
0.008
63
0.066
3-TM
6.3
<0.002
<0.002
0.006
<0.002
<0.003
<0.002
6.1
19
0.043
<0.002
<0.002
<0.002
<0.003
<0.005
<0.002
0.68
0.135
<0.004
<0.002
0.003
<0.004
0.004
11
0.058
4-TM
12
< 0.003
<0.002
0.018
<0.011
<0.003
<0.002
13.3
23
0.30
<0.006
<0.002
<0.002
<0.004
<0.009
<0.003
0.67
0.070
<0.013
<0.006
0.005
<0.005
0.003
18
0.035
5-TM
(blank)
36
<0.004
<0.002
0.002
<0.013
<0.004
<0.002
9.0
9.2
0.042
<0.007
<0.002
<0.003
<0.005
<0.012
<0.004
0.24
0.013
<0.017
<0.007
0.017
<0.006
0.005
130
0.018
GI = concentration in sample from front half of
sampling train; C^ - concentration in sample from
back half of sampling train. All sample volumes
were 250 m£.
Determined by atomic absorption.
•%
'Not reported.
194
-------
Table F-8.
COMPARISON OF TRACE ELEMENT RUNS
WITH BLANK RUN: ICAP ANALYSIS
(G! + C2, mgA)a
Element
Aluminum
Antimony
Barium
Boron
Calcium
Cadmium
Cobalt
Chromium
Copper
Iron
Lead
Magnesium
Manganese
Molybdenum
Nickel
Phosphorus
Silicon
Sodium
Strontium
Tin
Titanium
Vanadium
Zinc
Run
1-TM
2.0
0.084
0.87
3.2
54
0.043
0.040
0.13
4.4
3.3
0.58
1.5
0.048
0.083
0.47
1.4
3.6
15
0.061
0.49
0.045
0.055
8.7
3-TM
1.7
0.11
0.59
3.0
43
0.030
0.050
0.060
0.46
1.5
0.41
2 . 4
0.031
0.070
0.74
0.68
3.0
10
0.042
0.30
0.043
0.060
7.7
4-TM
1.4
0.29
0.57
3.7
19
0.015
0.038
0.12
1.2
2.3
0.28
2.0
0.040
0.050
0.56
0.69
6.5
12
0.071
0.69
0.046
0.050
1.9
5-TM
(blank)
1.4
0.12
0.42
1.2
60
0.013
0.047
0.034
0.20
0.59
0.22
0.82
0.013
0.043
0.72
0.60
6.4
5.7
0.042
0
0.024
0.053
11
GI = concentration in sample from front half
of sampling train; C2 = concentration in
sample from back half of sampling train. All
sample volumes were 250 m£.
195
-------
APPENDIX G
SIMULATED SOURCE SEVERITY DISTRIBUTIONS
Simulated source severity distributions are presented in
Figures G-l through G-22. These distributions were com-
puted according to the methodology set forth in Reference 41.
Due to the large number of emission species associated with
this source type, distributions are shown only for those
emissions for which the representative plant source severity
is 0.1 or greater.
Input data to the simulation routine consisted of stack
heights and production rates taken from the survey of oper-
ating plants in Reference 21, and emission factors from
Table 31. For the purpose of this simulation, the carbon
black industry is considered to be homogeneous with respect
to emissions. That is, variability in emission factors is
due only to process variability and variability in sampling
and analytical procedures; no plant-to-plant variability is
assumed. This assumption is consistent with the sampling
strategy employed in this study, i.e., only one plant was
sampled. Thus, emission factors were assumed to be normally
distributed with parameters determined by the means and 95%
confidence limits listed in Table 31. Furthermore, the
simulated source severities were computed assuming no plume
rise and no combustion devices used to burn the main process
vent gas.
196
-------
In Figures G-l through G-22, the ordinate, labelled "cumu-
lative frequency," should be interpreted as the percentage
of plants which have a severity less than the corresponding
value on the abscissa.
197
-------
»s
3
£8
MIN. VRLUE = 0.00180
MRX. VflLUE = 0.09800
MEflN = 0.02700
STD. DEV. = 0.02000
t'.ootoo
0.01000
SOURCE SEVERITY
Figure G-l. Simulated source severity distribution for
particulate emissions from main process vent
8
8
I
8
f
s?
V
8.
ft
HIM. VflLUE - O.Z370
MflX- VflLUE = 14.2000
MEflN = 3.6674
STD. DEV. = 2.4676
I'.ooo ib.ooo
SOURCE SEVERITY
1DO.OM
Figure G-2. Simulated source severity distribution for
carbon monoxide emissions from main process vent
198
-------
3
e
5,8
I
u
8
8'
8
8
8.
a
MIN. VflLUE = 0.0192
MflX. VflLUE = 1.1500
MEON = 0.2962
STO. OEV. = 0.1993
O.IM 1-000
SOURCE SEVERITY
Figure G-3. Simulated source severity distribution for
nitrogen oxide emissions from main process vent
9
u
8
8
I
B
•Y.ooooo
MIN. VRLUE = 1.73000
HflX. VflLUE = 105-00000
MEflN = 27.24000
STO. OEV. = 18.95000
it)- ooooo i bo. ooooo
SOURCE SEVERITY
Figure G-4. Simulated source severity distribution for
hydrocarbon emissions from main process vent
199
-------
8
i
8
8
I
S
P
f.8
5$
MIN. VflLUE = 2.3100
MflX. VflLUE = 159.0000
MEflN = 37.0556
STD. DEV. = 27.9251
Tooo
IDO-OCO
SEVERITY
Figure G-5. Simulated source severity distribution for
hydrogen sulfide emissions from main process vent
MIN. VflLUE = 0.58000
MflX. VflLUE = 38.40000
MEflN = 9.25000
STD. DEV. = 6.84000
ib.00000
SOURCE SEVERITY
Figure G-6. Simulated source severity distribution for
carbon disulfide emissions from main process vent
200
-------
MIN. VflLUE = 0.1630
MflX. VflLUE = 14.6000
MEflN = 3.0901
STO. DEV. = 2.4789
I'.OOD ib.oon
SOURCE SEVERITY
Figure G-7. Simulated source severity distribution for
carbonyl sulfide emissions from main process vent
L
5?
I*1
U
8
S
MIN. VflLUE = 0.0365
MflX. VflLUE = 2.3700
HEflN = 0.5818
STO. OEV. = 0.4259
Von
0.100 1.000
SOURCE SEVERITY
Figure G-8. Simulated source severity distribution for
carbon black emissions from main process vent
201
-------
5?
5?
S*
M1N. VflLUE = 0.0183
MflX. VflLUE = 2-8100
MEflN - 0.5762
STO. DEV. = 0.5313
0.100 1.000
SOURCE SEVERITY
Figure G-9. Simulated source severity distribution for
benzofluoranthene emissions from main process vent
M1N. VflLUE = 0.0025
MflX. VflLUE = 2.2500
NEflN = 0.5104
STO. OEV. = 0.4579
t'ooi
SOURCE SEVER 1 TV
Figuro G-10. Simulated source severity distribution for
benzopyrene emissions from main process vent
202
-------
8
|J
ft
£8
U
8
MIN. VflLUE = 0.0002
MRX. VRLUE = 7.0800
MERN = 1.2620
STD. DEV. = 1.1830
Van o'.ooi o'.oio o'.ira I'.ooi ID.MII
SOURCE SEVERITY
Figure G-ll. Simulated source severity distribution for
7,12-Dimethylbenz[a]anthracene emissions
from main process vent
0.100 I'.ODO
SOURCE SEVERITY
MIN. VRLUE = 0.0417
ttflX. VRLUE = 4.1500
MEflN = 0.9624
STD. DEV. = 0.7458
ll.OOO
Figure G-12. Simulated source severity distribution for
copper emissions from main process vent
203
-------
MIN. VflLUE = 0-0118
MRX. VflLUE = 0.7100
MEHN = 0.1850
STO. OEV. = 0.1281
O'.IOO
SOURCE SEVERITY
l.ooo
Figure; G-13. Simulated source severity distribution for
lead emissions from main process vent
s|
e?
MIN. VRLUE = 0.0149
MflX. VflLUE i 1.0900
hEflN = 0-2410
STO- DEV. = 0.1875
tf.100 I'.OM
SOURCE SEVERITY
10.000
Figure G-14. Simulated source severity distribution for
nickel emissions from main process vent
204
-------
HIN. VflLUE = 0.0047
MHX. VflLUE = 14.2000
MEflN = 2.3057
STO. DEV. = 2.2069
•tf.ooio.oio o'.ioo I'.on
SOURCE SEVERITY
ib.ooo ibo.ooo
Figure G-15. Simulated source severity distribution for
phosphorus emissions from main process vent
u
8
a
8
s
Too i
HIM. VflLUE = 0.0012
MflX. VflLUE = 2SJBO
MEflN = 1.2730
STD. OEV. - 2.3390
o'.oio o'.in r.coo
SOURCE SEVERITY
Figure G-16. Simulated source severity distribution for
nitrogen oxide emissions from dryer vent
205
-------
0.100 1-0
SOURCE SEVERITY
M1N. VflLUE = 0.0282
MflX. VflLUE = 10.2000
MEflN = 1.3310
STO. OEV. = 1 .6550
Vooo
Figure G-17. Simulated source severity distribution for
carbon black emissions from dryer vent
HIM. VflLUE = 0.0637
MflX. VflLUE = 9.1500
MEflN = 1 .6710
STD. OEV. = 1 .5170
Voio
SOURCE SEVERITV"
Figure G-18. Simulated source severity distribution for
carbon black emissions from pneumatic system vent
206
-------
A
M1N. VflLUE = 0.0141
MflX. VALUE = 1.9900
MEflN = 0-2699
STO. OEV. = 0.2724
SOURCE SEVERITY'
Figure G-19. Simulated source severity distribution for.
carbon black emissions from vacuum cleanup system vent
HIM. VHLUE = 0.07500
MflX. VBLUE = 2.16000
MEflN = 0.81600
5TD. DEV. = 0.44900
t'01000
'SOURCE SEVER ITY'OM™
Figure G-20. Simulated source severity distribution for
hydrocarbon emissions from oil storage tank vent
207
-------
Voio
O'.IM 1.000
SOURCE SEVERITY
HIM. VflLUE = 0.0217
ttflX. VflLUE = 4.4100
MEflN = 1.1860
STO. OEV. = 0.8519
Yooo
Figure G-21. Simulated source severity distribution for
fugitive particulate emissions
8
t.ioo
HIN. VflLUE = 0.4870
tlRX. VflLUE = 98.9000
MEflN = 26.6100
STD. OEV. = 19-1200
i.on it.
aOURCC SEVERITY
Figure G-22. Simulated source severity distribution for
fugitive carbon black emissions
208
-------
APPENDIX H
REMOTE MONITORING OF CARBON
MONOXIDE IN FLARE OFF-GASa
1. MONITORING EQUIPMENT AND PROCEDURES
The combustion efficiency of carbon monoxide in the flare
was determined by measuring the CO concentration in the gas
emitted from the flare stack using the ROSE (Remote Optical
Sensing of Emissions) System.
Infrared energy radiated by the off-gases from the flare
stack was collected by the ROSE system telescope and pro-
cessed to display signal intensity as a function of wave-
length (microns) or wave number (citT1). A block diagram of
the ROSE System is shown in Figure H-l.11 The field of view
of the ROSE System through the flare was a rectangular
section approximately 30 mm wide and 100 mm high. This
rectangular section intersected the stack axis at about
0.5 m above the stack exit plane at an elevation angle of
30° as indicated in figure H-2. For analysis purposes, the
assumptions were made that the off-gases were confined to a
vertical cylinder with a diameter equal to that of the stack
and that they were uniformly distributed along the line of
sight within the cylinder.
This work was performed by EPA personnel under the direction
of Dr. William F. Herget.
209
-------
TRACKING
MIRROR
TELESCOPE
GRATING
MONOCHROMATOR
STRIP CHART
RECORDER
DETECTOR
PREAMPLIFIER
AMPLIFIER
Figure H-l. Block diagram of ROSE System
11
= 1.45m
x STACK /
0.76m
HEIGHT: 41m
TO ROSE SYSTEM, 90 m
ELEVATION ANGLE, 30°
Figure H-2. Geometry of ROSE System measurements
11
210
-------
The basic equation for the calculation of species concentra-
tions from the emission spectra are:
= Ng(X, T)[Nbb(X, T)]'1
' T'PL (H-l)
where e, = emissivity of the species at
wavelength, X
N (X, T) = spectral radiance of the species at
wavelength, X, and temperature, T
N,. (X, T) = spectral radiance of a blackbody at
wavelength, X, and temperature, T
K(X, T) = spectral absorption coefficient of
the species at X and T
/\
P = partial pressure of the species
s\
L = optical path length in the plume
/\
The two unknowns in these equations are T and P. The quan-
tity K(X, T) is determined in the laboratory from spectra
obtained at known concentrations and temperatures or from
theoretical calculations. The plume temperature can be
calculated from an approximate knowledge of the C02 concen-
tration and measurement of the CC>2 emission spectra at 2,275
cm"1 using the known spectral absorption coefficient for
C02. The emissivity was calculated at 2,190 cm"1 by ratio-
ing the spectral radiance of CO to that of a blackbody.
Using the CO spectral absorption coefficient from the litera
ture, Equation H-l was used to calculate the CO partial
pressure, and hence, its concentration.
2. ANALYTICAL PROCEDURE
The above . equipment (EPA Rose System) and procedures used in
the field measurements were identical to those used in a
previous study.11 To assist in the analysis of the field
211
-------
data, laboratory studies were carried out in a bunsen burner
flame utilizing the "new" ROSE System. In the new system,
the grating monochromator has been replaced by a Fourier
Transform Spectrometer (FTS) capable of 0.06 cm"1 resolution.
(The meiximum resolution of the grating system was about
1.0 cm"1.) It was thus possible to study the bunsen burner
flame at high resolution and at a resolution comparable to
that used in the field studies, but with a signal to noise
ratio greater than that attainable with the grating system.
Figure H-3 shows spectra obtained with the FTX System and
also field spectra. The uppermost spectrum (A) shows absorp-
tion due to atmospheric species and CO (combined in a cell).
A composite of four spectra of the off-gas emission (with
flare ignited) is shown also (B). Next is shown an emission
spectrum of a bunsen burner flame (C) obtained with the FTS
system; the lowermost spectrum (D) is from the same flame
but with the field of view above the visible portion of the
flame.
The spectral detail in (B) from approximately 2,156 cm'1
to 2,220 cm""1 was attributed to CO in the previous work.11
Spectrum (C) shows, however, that if this detail were due to
CO, it would also occur with similar intensities in the
2,004 cnT1 to 2,132 cm"1 region. Spectrum (B) actually
compares most favorably with spectrum (D), and the
detail in the 2,156 cm"1 to 2,220 cnT1 region is actually
due to C02. Figure H-4 shows this spectral region at 0.25 cm"1
it is seen that the detail in (D) is actually composed of
groups of spectral lines, and therefore is not due to CO
emissions.
212
-------
to
M
co
(A)-
EPR ROSE SYSTEM
'BUNSEN BURNER FLPME7rRES-=-II~CM-If
RTM05PHERIC RND CO RBSORPTION.. .;.
1 CM RBOVE BURNER (LYE = 0.051
(C)
) FIELD SPECTRA (COMPOSITE OF FOUR SCANS)
RBOVE FLRME (LYE = 0.03).
i
ISuO 1908 1916 I924~1932..., 1940
1996 2004 20'l2 2020 2o!
Figure H-3. Comparison of field spectra and laboratory spectra
obtained with FTS System
-------
... .-Al- ,.._,, _._-
28 2036 20!»» 20*52 • 20^0 ...20887, 20*76
20b2 2lbo 2108 21 'l6 212^" Z.1^2 2l¥5~
UQVFNl IMBFRS
Figure H-3 (continued). Comparison of field spectra and laboratory
spectra obtained with FTS System
-------
U1
3t 21722180 21B8 '. 2196... 2204,. 2212 2220 2228 ,2236 22li4 2252 2260 22B8 2276 2284 2292 23i
Figure H-3 (continued). Comparison of field spectra and laboratory
spectra with FTS System
-------
N)
M
CTi
RESCLUTIOf! = 0.25 CM
.-1
IN FLAME
I , !
! K ill
I ,1
'I i
lll.i
j | I j
f ; 3
Mil!
ill
L
:i
l,.iM
h.J1
I!
! \ti
i :! i » r;
II
5 ' I!1
ill T ] '!!: I ; >\l | !ii: • '•!'.:!
|i:j^jj|.!|:| llliiiihi'i
ABOVE FLAME .
aO t'::ba 2:l"u3 211?1* ildF
2123 3S:U6 2314 222Z 2230 2233 22li6 ZZSi 2232 22/0 Z'^Jti
HflVEMUNBEfiS
Figure H-4. Laboratory spectra at resolution of 0.25 cm~
-------
The concentration reduction in the bunsen burner flame was
determined from spectra shown in Figure H-5. Spectrum (A)
shows the atmospheric absorption between the bunsen burner
flame and the FTS dector; (B) and (C) are flame absorption
spectra, and the lower spectrum is ambient temperature CO
absorption. The CO reduction from (C) to (B) was calculated
to be at least 25. This reduction, coupled with the factor
of 100 determined as in the previous study11 (flare off-gas
spectra with respect to incorrectly identified CO lines)
gives an overall reduction of at least 2,500 in CO
concentration.
3. SUMMARY OF RESULTS
The bunsen burner studies showed that the field spectra were
comparable to bunsen burner spectra obtained from just above
the visible portion of the flame, and that the CO content was
considerably lower than those calculated previously.11 The
previous calculation was incorrect because spectral lines
due to C02 emissions were incorrectly interpreted as being
due to CO. The bunsen burner spectra obtained with the FTS
System conclusively identified the error. It was determined
that the CO concentration in the flare off-gas was lower by
at least a factor of 25 that the concentrations reported in
the previous work. The present results indicate that the
flare reduces the CO concentration by at least a factor of
2,500.
The concentration of CO measured in the main provess vent
gas was 11% by volume. A reduction in concentration by a
factor of 2,500 corresponds to a concentration of approxi-
mately 50 ppm in the flare off-gas. The CO control
efficiency of the flare is therefore greater than 99.9%.
217
-------
EPA ROSE SYSTEM
RESOLUTION = 0.125 CM
,-1
i
.._. ( A )ATMOSPHERIC BACKGROUND
( B )FLAME ABSORPTION (ABOVE FLAME)
.. — I. ... — I -.-i I
.--.-- ( C ;FLAME ABSORPTION (IN FLAME)
i
j.TTZT CO ABSORPTION (1.3 TORR IN 10 CM CELL) H 1
2120 2122 21211 2126 2128 2130 2132
Figure H-5.
Spectra used to determine reduction
in CO concentrations
218
-------
SECTION VIII
GLOSSARY OF TERMS
ATMOSPHERIC STABILITY CLASS - Class used to designate degree
of turbulent mixing in the atmosphere.
CRITERIA POLLUTANT - Emission species for which an ambient
air quality standard has been established.
EMISSION FACTOR - Weight of material emitted to the environ-
ment per unit weight of carbon black produced.
FEF CARBON BLACK - Fast extruding furnace grade black made
by oil furnace process (see Table 2 for properties; see
Tables 48 and 49 for applications).
GPF CARBON BLACK - General purpose furnace grade black made
by oil furnace process (see Table 2 for properties; see
Tables 48 and 49 for applications).
HAF CARBON BLACK - High abrasion furnace grade black made by
oil furnace process (see Table 2 for properties; see
Tables 48 and 49 for applications).
ISAF CARBON BLACK - Intermediate super abrasion furnace
grade black made by oil furnace process (see Table 2 for
properties; see Tables 48 and 49 for applications).
NONCRITERIA POLLUTANT - Emission species for which no ambient
air quality standard has been established.
OIL FURNACE PROCESS - Continuous process for production of
carbon black from a liquid hydrocarbon feedstock.
PELLETIZER - Machine which agglomerates raw carbon black
into pellets for ease of handling and transporting.
PNEUMATIC SYSTEM - Pneumatic conveyor used to transport raw
carbon black from product recovery area of plant to product
treatment area of plant.
219
-------
POLYCYCLIC ORGANIC MATERIAL (POM) - Organic compounds con-
taining two or more benzene ring structures.
SAP CARBON BLACK - Super abrasion furnace grade black made
by the oil furnace process (see Table 2 for properties;
Tables 48 and 49 for applications).
STRUCTURE - The formation of chain-like aggregates of carbon
black particles; it affects the rheology of carbon black-
elastoner systems.
TANK OUTAGE - Distance from liquid surface to top of storage
tank.
THERMAL PROCESS - Cyclical process for production of carbon
black from natural gas.
VACUUM CLEANUP SYSTEM - Plant-wide vacuum system used to
pick up loose carbon black.
220
-------
SECTION IX
CONVERSION FACTORS AND METRIC PREFIXES65
To convert from
degree Celsius
joule (J)
kelvin (K)
kilogram (kg)
kilogram (kg)
kilogram/meter3 (kg/m3)
kilogram/meter (kg/m3)
kilometer2 (km2)
meter (m)
meter (m)
meter2 (m2)
meter3 (m3)
meter3 (m3)
meter3 (m3)
metric ton
ohm-meter (Q-m)
pascal (Pa)
pascal (Pa)
CONVERSION FACTORS
to
degree Fahrenheit
British thermal unit
degree Celsius
pound-mass (Ib
mass avoirdupois)
ton (short, 2000 Ib
mass
Q
gram/cm3
pound-mass/foot3
mile2
feet
mile
feet2
barrels (42 gal)
feet3
gallon (U.S. liquid)
pound-mass
ohm-centimeter
inch of mercury (60°F)
o
pound-force/inch (psi)
Multiply by
t° = 1.8 t« + 32
9.479 x lO-4
t» = t» - 273.15
2.204
1.102 x 10~3
1.000 x 10~3
6.243 x 10~3
2.591
3.281
6.215 x ID"1*
1.076 x 101
6.293
3.531 x 101
2.642 x 102
2.205 x 103
1.000 x 102
2.961 x I0~k
1.450 x 10-4
65Standard Metric Practice Guide. ASTM Designation E380-74,
American Society for Testing and Materials, Philadelphia,
Pennsylvania, 1974. 34 pp.
221
-------
METRIC PREFIXES
Prefix
mega
kilo
milli
micro
nano
Symbol
M
k
m
W
n
Multiplication
factor
106
103
10
10
10
-3
-6
-9
Example
1 MJ = 1 x 106 joules
1 kPa = 1 x 103 pascals
1 mg = 1 x 10~3 gram
1 ym = 1 x 10~6 meter
1 ran = 1 x 10"^ meter
222
-------
SECTION X
REFERENCES
1. Carbon Black. In: Minerals Yearbook, Vol. 1. Metals,
Minerals, and Fuels. U.S. Government Printing Office,
Washington, D.C. 1940-1974.
2. Smith, W. R., and D. C. Bean. Carbon Black. In:
Kirk-Othmer Encyclopedia of Chemical Technology, Second
Edition, Vol. 4. John Wiley & Sons, Inc., New York,
New York, 1964. pp. 243-282.
3. Davidson, H. W., P. K. C. Wiggs, A. N. Churchouse,
F. A. P. Maggs, and R. S. Bradley. Manufactured Carbon.
Pergamon Press, New York, New York, 1968. pp. 1-55.
4. Matsubayashi, E. Carbon Black. Sekiyu Gakkai Shi,
16(5) :381-386, 1973.
5. Bonnet, J. B. and A. Voet. Carbon Black. Marcel Dekker,
Inc., New York, New York, 1976. 351 pp.
6. Mantell, C. L. Carbon and Graphite Handbook. Inter-
science Publishers, New York, New York, 1968. pp. 76-105.
7. Drogin, I. Carbon Black. Journal of the Air Pollution
Control Association, 18:216-228, 1968.
8. Deviney, M. L. Surface Chemistry of Carbon Black and
Its Implications in Rubber Chemistry. A Current Review.
Advances in Colloid Interface Science, 2(3):237-259, 1969.
9. Nau, C. A. Health Studies Relating to Carbon Black.
Presented at the Conference on Environmental Aspects of
Chemical Use in Rubber Processing Operation, Akron,
Ohio, March 12-14, 1975. 13 pp.
10. Collyer, H. J. Carbon Black and Ecology. Presented at
the Conference on Environmental Aspects of Chemical Use
in Rubber Processing Operations, Akron, Ohio, March 12-14,
1975. 10 pp.
223
-------
11. Miller, S. E., and R. E. Barrett. Sampling and Analysis
of Source Emission Samples from a Carbon Black Plant,
Final Report. Contract 68-02-1409, Task Order No. 9,
U. S. Environmental Protection Agency, Research Triangle
Park, North Carolina, April 1975. 24 pp. plus appendices.
12. Whitsel, T. S. Theoretical Aspects of Carbon Black Manu-
facture. Rubber Plastics Age. October 1964, .pp. 1209-1212,
13. Gerstle, R. W. Carbon Black Industry, Final Report.
Contract 68-02-1321, Task No. 21, U. S. Environmental
Protection Agency, Research Triangle Park, North Caro-
lina, May 1975. 25 pp.
14. Krejci, J. C. Carbon Black Process. U.S. Patent 2,632,713
(to Phillips Petroleum Company), March 24, 1953.
15. Improvements in the Manufacture of Carbon Black. British
Patent 669,968 (to Columbian Carbon Company), April 9,
1952.
16. Braendle, H. A. Manufacture of Carbon Black. U.S. Patent
2,735,753 (to Columbian Carbon Company), February 21, 1956.
17. Stokes, C. A. Process of Producing Carbon Black and
Synthesis Gas. U.S. Patent 2,672,402 (to Cabot Corpora-
tion) , March 16, 1954.
18. Williams, I. Process and Apparatus for Making Carbon
Black. U.S. Patent 2,625,466 (to J. M. Huber Corporation),
January 13, 1953.
19. Process and Apparatus for the Production of Carbon Black.
British Patent 699,406 (to Godfrey L. Cabot Corporation),
November 4, 1953.
20. Friauf, G. F., and B. Thorley. Carbon Black Process.
U.S. Patents 3,010,794 and 3,010, 795 (to Cabot Corpora-
tion), November 28, 1961.
21. Schwartz. W. A., F. B. Higgins, Jr., J. A. Lee, R. Newirth,
and J. W. Pervier. Engineering and Cost Study of Air
Pollution Control for the Petrochemical Industry.
Volume 1: Carbon Black Manufacture by the Furnace
Process. EPA-450/3-73-006-a, U.S. Environmental Protec-
tion Agency, Research Triangle Park, North Carolina,
June 1974. 116 pp.
22. Day, J. V. Carbon Black Pellets and a Process for Their
Manufacture. U.S. Patent 2,850,403 (to Cabot Corporation)
September 2, 1958.
224
-------
23. Alleman, C. E. Carbon Black Pelleting by Controlling
Power to the Pelletizer Motor. U.S. Patent 3,266,873
(to Phillips Petroleum), August 16, 1966.
24. Chemical Profiles. Carbon Black. Chemical Marketing
Reporter, 210(24):9, 1976.
25. Tire, Rubber, and Carbon Black Plant Locations (Map).
Rubber World. 172:42-44, May 1975.
26. Carbon Black in 1974. Mineral Industry Surveys, U.S.
Department of the Interior, Bureau.of Mines, Washington,
D.C., 1975.
27. TLVs® Threshold Limit Values for Chemical Substances
and Physical Agents in the Workroom Environment with
Intended Changes for 1976. American Conference of
Governmental Industrial Hygienists, Cincinnati, Ohio,
1976. 94 pp.
28. Evaporation Loss from Fixed Roof Tanks. Bulletin 2518,
American Petroleum Institute, New York, New York, 1962.
38 pp.
29. Use of Variable Vapor Space Systems to Reduce Evapora-
tion Loss. Bulletin 2520, American Petroleum Institute,
New York, New York, 1964. 14 pp.
30. Petrochemical Evaporation Loss from Storage Tanks.
Bulletin 2523, American Petroleum Institute, New York,
New York, 1969. 14 pp.
31. Evaporation Loss from Floating Roof Tanks. Bulletin
2517, American Petroleum Institute, New York, New York,
1962. 13 pp.
32. Evaporation Loss in the Petroleum Industry - Causes and
Control. Bulletin 2513, American Petroleum Institute,
New York, New York, 1959. 57 pp.
33. Turner, Dt B. Workbook of Atmospheric Dispersion
Estimates. Public Health Service Publication No.
999-AP-26, U.S. Department of Health, Education, and
Welfare. Cincinnati, Ohio, 1969. 84 pp.
34. Nonhebel, G. Recommendations on Heights for New Indus-
trial Chimneys. Journal of the Institute of Fuel,
33:479-511, 1960.
225
-------
35. Particulate Polycyclic Organic Matter. National
Academy of Sciences, Washington, D.C., 1972. 361 pp.
36. Neal, J., M. Thornton, and C. A. Nau. Polycyclic Hydro-
carbon Elution from Carbon Black or Rubber Products.
Archives of Environmental Health, 4:46-54, 1962.
37. Nau, C. A., J. Neal and V. Stembridge. A Study of the
Physiological Effects of Carbon Black. I: Ingestion.
Archives of Industrial Hygiene, 17:21-28, 1958.
38. Nau, C. A., J. Neal, and V. Stembridge. A Study of the
Physiological Effects of Carbon Black. II: Skin Con-
tact. Archives of Industrial Hygiene, 18:511-520, 1958.
39. Nau, C. A., J. Neal, and V. Stembridge. A Study of the
Physiological Effects of Carbon Black. Ill: Absorption
and Elution Potentials. Subcutaneous Injections.
Archives of Industrial Hygiene, 20:512-533, 1960.
40. Kau, C. A., J. Neal, V. Stembridge, and R. N. Cooley.
Physiological Effects of Carbon Black. IV: Inhalation.
Archives of Environmental Health, 44:15-431, 1962.
41. Eimutis, E. C., B. J. Holmes, and L. B. Mote. Source
Assessment: Severity of Stationary Air Pollution
Sources - A Simulation Approach. EPA-600/2-76-032e.
U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, July 1976. 119 pp.
42. 1972 National Emissions Report. EPA-450/2-74-012,
U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina, June 1974. 422 pp.
43. Eimutis, E. C., and M. G. Konicek. Derivations of
Continuous Functions for the Lateral and Vertical Atmos-
pheric Dispersion Coefficients. Atmospheric Environ-
ment, 6:859-863, 1972.
44. Hesketh, H. E. Scrubbing with an Impinger Reactor.
In: Proceedings of the Second National Conference on
Energy and the Environment, College Corner, Ohio,
November 13-15, 1974. pp. 174-176.
45. Wall, C. J., J. T. Graves, and E. J. Roberts. How to
Eturn Salty Sludges. Chemical Engineering, 82:77-82,
1975.
46. Source Testing of a Waste Heat Boiler. George D. Clayton
Associates. EPA 75-CBK-3, U.S. Environmental Protection
Agency, Research Triangle Park, North Carolina, 1975.
29 pp.
226
-------
47. Haase, D. J. New Solvent Cuts Costs of Carbon Monoxide
Recovery. Chemical Engineering, 82:52-54, 1975.
48. Hael, R. The Improved Blacks: What They Are and Where
to Use Them. Rubber World, 172:42-44, 1975.
49. Richardson, H. M. Forecasting in the Rubber Industry.
In: Hydrocarbons. The Dilemma in Forecasting.
Presented at the joint meeting of the Chemical Marketing
Research Association and the Commercial Development
Association. New York, New York, 1974. pp. 77-84.
50. Burgess, K. A., F. Lyon, and W. S. Stoy. Carbon. In:
Encyclopedia of Polymer Science and Technology, Vol. 2.
Wiley-Interscience Publications, New York, New York,
1965. pp. 820-836.
51. Rubber Demand Faces Lower Growth Rate. Chemical and
Engineering News, 52:12, 1974.
52. Rubber Consumption to Increase. Rubber World, 172:83,
1975.
53. Finneran, J. A., L. J. Buividas, and N. Whalen. Advanced
Ammonia Technology. Hydrocarbon Processing, 51:127-130,
1972.
54. Briggs, G. A. Plume Rise. In: Critical Review Series.
U.S. Atomic Energy Commission Technical Information
Center, Oak Ridge, Tennessee, 1969. 81 pp.
55. Moses, H. and M. R. Kraimer. Plume Rise Determination
A New Technique Without Equations. Journal of the Air
Pollution Control Association, 22:621, 1972.
56. Standard Classification System for Carbon Blacks Used
in Rubber Products. 1974 Annual Book of ASTM Standards.
Part 38. American Society for Testing and Materials,
Philadelphia, Pennsylvania, 1974. pp. 408-409.
57. Environmental Protection Agency - Part II - Standards of
Performance for New Stationary Sources - Proposed Amend-
ments to Reference Methods. Method 5 - Determination
of Particulate Emission from Stationary Sources.
Federal Register, 41 (111):23076-23083, 1976.
58. Hamersma, J. W., S. G. Reynolds, and R. F. Maddalone.
IERL-RTP Procedures Manual: Level I Environmental
Assessment. EPA-600/2-76-160a, U.S. Environmental
Protection Agency, Research Triangle Park, North Caro-
lina, 1976. 131 pp.
227
-------
59. Liu, B. Y. H. and K. W. Lee. Efficiency of Membrane
and Nucleopore Filters for Submicrometer Aerosols.
Environmental Science and Technology, 10:345-350, 1976.
60. Environmental Protection Agency - Part II - Standards of
Performance for New Stationary Sources - Proposed Amend-
ments to Reference Methods. Method 3 - Gas Analysis for
Carbon Dioxide, Oxygen, Excess Air, and Dry Molecular
Weight. Federal Register, 41 (111):23069-23072, 1976.
61. Environmental Protection Agency - Standards of Perform-
ance for New Stationary Sources - Petroleum Refinery
Sulfur Recovery Plants. Test Methods and Procedures.
Federal Register, 41(193):43870-43874, 1976.
62. Chapter 1 - Environmental Protection Agency - Part 60 -
Standards of Performance for New Stationary Sources.
Subchapter C - Air Programs. Subpart O - Standards of
Performance for Sewage Treatment Plants. Federal
Register, 39:9319-9321, 1974.
63. Chapter 1 - Environmental Protection Agency - Part 60 -
Standards of Performance for New Stationary Sources.
Subchapter C - Air Programs. Method 7 - Determination
of Nitrogen Oxide Emissions from Stationary Sources.
Federal Register, 36:24891-24893, 1971.
64. Environmental Protection Agency - Part n - Standards
of Performance for New Stationary Sources - Proposed
Amendments to Reference Methods. Method 7 - Determina-
tion of Nitrogen Oxide Emissions from Stationary Sources.
Federal Register, 41(111):23085-23087, 1976.
65. Standard Metric Practice Guide. ASTM Designation E380-74,
American Society for Testing and Materials, Philadelphia,
Pennsylvania, 1974. 34 pp.
228
-------
TECHNICAL REPORT DATA
(Please read Inunictions on the reverse before completing)
1. REPORT NO.
EPA-600/2-77-107k
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Source Assessment: Carbon Black Manufacture
5. REPORT DATE
October 1977
6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
:.W. SerthandT.W. Hughes
8. PERFORMING ORGANIZATION REPORT NO.
MRC-DA-720
9. PERFORMING ORGANIZATION NAME AND ADDRESS
Monsanto Research Corporation
1515 Nicholas Road
Ohio 45407
10. PROGRAM ELEMENT NO.
1AB015; ROAP 21AXM-071
11. CONTRACT/GRANT NO.
68-02-1874
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
Task Final; 6
I" AND PERIOD COVERED
I/75-7/77
14. SPONSORING AGENCY CODE
EPA/600/13
15. SUPPLEMENTARY NOTES ffiRIj.RTP task officer for this report is LA. Jefcoat, Mail Drop
62, 919/541=2547, Similar previous reports are also in the EPA-600/2-76-032 series.
16. ABSTRACT
repOrf- summarizes the assessment of air emissions from the manufac-
ture of carbon black, currently manufactured in the U.S. by two major processes:
thermal and oil furnace. Sources of atmospheric emissions within oil furnace plants
(about 90% of the 30 U.S. carbon black plants) include the main process vent, dryer
vent, pneumatic system vent, oil feedstock storage tanks, vacuum cleanup system
vent, and fugitive sources. To assess the severity of emissions from this industry,
a representative plant was defined as using the oil furnace process and with a mean
production rate of 50,000 metric tons/yr. For a representative plant, calculated
source severities were: 0.02 for particulates emitted from the main process vent;
0.046 and 0. 58 for SOx and NOx, respectively, from the dryer vent; 21 for HC emitted
from the main process vent; and 27 for carbon black fugituve emissions. The average
number of persons exposed to high contaminant levels from carbon black manufacture
was estimated and designated as the 'affected population. ' The calculation was made
for each species emitted and for each emission point within a representative plant for
which the source severity exceeds 0.1. The largest value obtained was 11,000 persons,
due to H2S emissions from the main process vent. Assuming the same control levels
in 1974 and 1980, emissions from the industry will increase by 14% by 1980.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS
C. COSATI Held/Group
Air Pollution
Assessments
Carbon Black
Dust
Sulfur Oxides
Nitrogen Oxides
Hydrocarbons
Hydrogen Sulfide
Air Pollution Control
Stationary Sources
Oil Furnace Process
Particulate
Fugitive Emissions
13B
14B
11G
07B
07C
18. DISTRIBU1 ION STATEMENT
Unlimited
19. SECURITY CLASS (This Report)
Unclassified
21. NO. OF PAGES
244
20. SECURITY CLASS (Thispage)
Unclasj
22. PRICE
EPA Form 2220-1 (9-73)
229
------- |