&EPA
United States
Environmental Protection
Agency
Office of
Toxic Substances
Washington DC 20460
April 1980
EPA-560/13-80-014
Toxic Substances
Materials
Balance for
Benzene
Review
Copy
Level 1 - Preliminary
-------
FINAL REPORT
LEVEL I MATERIALS BALANCE
BENZENE
Prepared for:
U.S. ENVIRONMENTAL PROTECTION AGENCY
OFFICE OF PESTICIDES AND TOXIC SUBSTANCES
SURVEY AND ANALYSIS DIVISION
Task No. 16
Contract No. 68-01-5793
Michael Callahan - Project Officer
Elizabeth F. Bryan - Task Manager
Prepared by:
JRB ASSOCIATES, INC.
8400 Westpark Drive
McLean, Virginia 22102
Project Manager: Karen Slimak
Task Leader: Robert L. Hall
Contributing Writers: Carlos Buitrago
Frank Montecalvo
Tom Yatsko
Submitted: May 9, 1980
-------
THIS IS A DRAFT REPORT ON A LEVEL I MATERIALS BALANCE ON BENZENE.
AS SUCH, IT IS PRESENTED AS A FOCUS OF DISCUSSION AND AS A BASIS
FOR FUTURE MATERIALS BALANCE STUDIES; IT IS NOT MEANT TO BE A
DEFINITIVE STUDY. THE RESULTS REPORTED WERE BASED ON A 700-HOUR,
SEVEN-WEEK ANALYSIS OF A SET OF READILY AVAILABLE LITERATURE
SUPPLIED BY EPA. ALTHOUGH SUPPLEMENTARY INFORMATION UNDOUBTEDLY
EXISTS, OBTAINING IT WAS OUTSIDE THE SCOPE OF THIS TASK.
-------
MATERIALS BALANCE LEVELS
Materials balance studies are performed at three levels or depths
of study and effort. In general the study of a chemical proceeds
sequentially through these three levels. Particular chemicals are
assigned to be studied at one of the levels on the basis of availability
of information. The three levels are described below.
Level I:
A LEVEL I MATERIALS BALANCE requires the lowest level of effort
and involves a survey of readily available information for constructing
the materials balance. Ordinarily, many assumptions must be made in
accounting for gaps in information; however, all are substantiated
to the greatest degree possible. Where possible the uncertainties in
numerical values are given, otherwise they are estimated. Data gaps
are identified and recommendations are made for filling them. A
Level I materials balance relies heavily on the EPA's Chemical Information
Division as a source of data and references involving readily available
information. Most Level I MB's are completed within a 3-6 week period;
CID literature searches generally require a 2 week period to complete.
Thus the total time required for completion of a Level I materials
balance ranges from 5-7 weeks.
Level II;
A LEVEL II MATERIALS BALANCE involves a greater level of effort,
including an in-depth search for all information relevant to the
materials balance. The search includes all literature (concentrating
on primary references), contacts with trade associations, other agencies,
and industry to try to uncover unpublished information, and possibly
site investigations. Uncertainties and further data needs are identified
in the Level II report. Recommendations for site sampling needs for
Level III are also identified.
Level III:
A LEVEL III study requires generation of new data through monitoring
and other means. It builds on the Level II literature searches and
reviews of industrial production data by filling in data gaps through
site visits and necessary monitoring. The data generated in this type
of study are intended to be statistically valid and have known confidence
values. The goal is a study upon which regulations or legal proceedings
may be based.
-------
ABSTRACT
A Level I materials balance was performed on benzene. Data are
reported for benzene production from petroleum by four processes
(catalytic reformation, toluene dealkylation, toluene disproportion-
ation, and isolation from pyrolysis gasoline) and for production
from coal during coking. Amounts of benzene consumed for the synthesis
of nine direct derivatives (ethylbenzene, cumene, cy.clohexane,
nitrobenzene, maleic anhydride, mono- and dichlorobenzenes, alkyl-
benzenes, anthraquinone, and biphenyl) and exports are presented.
These uses constitute approximately 93 percent of total benzene usage.
Nonconsumptive uses (solvents and inventory changes) are also tabulated.
Emissions due to each of the above processes are reported or estimated
where possible. In addition, emissions due to indirect production
(refinery operation, coke oven operations, oil spills, non-ferrous
metals manufacturing, ore mining, wood processing, coal mining, and
two phases of the textile industry) are presented. Production of
benzene as a component of gasoline and emissions due to gasoline use
are estimated. Locations of sites with high densities of benzene
producers and users are tabulated: the major "hotspots" are Houston/
Galveston, Texas; Corpus Christi, Texas; Beaumont/Port Arthur, Texas;
and Puerto Rico. The uncertainty ranges of all numbers used or derived
in this report are evaluated when possible and tabulated. Data gaps
are evaluated and general recommendations are presented. The results
of the report are summarized in two figures: the Environmental Flow
Diagram for benzene in Appendix A, and the Materials Balance Diagram
in the Executive Summary.
-------
TABLE OF CONTENTS
Page
EXECUTIVE SUMMARY vil
1.0 INTRODUCTION 1-1
1.1 PROPERTIES OF BENZENE 1-1
1.2 ENVIRONMENTAL FLOW DIAGRAM FOR BENZENE 1-2
2.0 DIRECT PRODUCTION OF BENZENE 2-1
2.1 PRODUCTION FROM PETROLEUM 2-1
2.1.1 Producers and Locations 2-1
2.1.2 Amounts Produced 2-5
2.1.3 Production by Catalytic Reformation 2-5
2.1.4 Production by Dealkylation of Toluene 2-7
2.1.5 Production by Toluene Disproportionation 2-9
2.1.6 Production from Pyrolysis Gasoline 2-10
2.1.7 General Emissions Due to Petroleum-Derived 2-11
Benzene Production
2.1.8 Emissions Due to Transportation, Loading and 2-13
Storage Associated With Production of Benzene
from Petroleum
2.2 BENZENE PRODUCTION FROM COAL 2-20
2.2.1 Summary 2-20
2.2.2 Benzene Production from Coal - The Process 2-22
2.3 SUMMARY 2-24
3.0 INDIRECT PRODUCTION OF BENZENE 3-1
3.1 INDIRECT PRODUCTION OF BENZENE FROM REFINING 3-1
OPERATIONS
3.2 BENZENE EMISSIONS FROM COAL COKING OPERATIONS 3-3
3.3 INDIRECT PRODUCTION OF BENZENE FROM OIL SPILLS 3-9
3.4 INDIRECT PRODUCTION OF BENZENE FROM VARIOUS SOURCES 3-10
4.0 IMPORTS OF BENZENE 4-1
4.1 AMOUNT IMPORTED 4-1
4.2 EMISSIONS DUE TO IMPORTS 4-1
4.2.1 Emissions to Air 4-2
4.2.2 Emissions to Water 4"2
ii
-------
5.0 CONSUMPTIVE USES OF BENZENE 5-1
5.1 CONSUMPTIVE USES - TOTALS 5-1
5.2 CATEGORIES OF USE 5-1
5.3 EMISSIONS BY CATEGORY USE 5-1
5.3.1 Consumption of Benzene by Ethylbenzene 5-1
Synthesis
5.3.2 Consumption of Benzene by Cumene Synthesis 5-8
5.3.3 Consumption of Benzene by Cyclohexane 5-12
Synthesis
5.3.4 Consumption of Benzene by Maleic Anhydride 5-17
Synthesis
5.3.5 Consumption of Benzene by Nitrobenzene 5-20
Synthesis
5.3.6 Consumption of Benzene by Chlorobenzene 5-26
Synthesis
5.3.7 Consumption of Benzene for Alkylbenzenes 5-3A
Synthesis
5.3.8 "Consumption of Benzene by Synthesis of 5-37
Anthraquinone
5.3.9 Consumption of Benzene by Synthesis of 5-38
Biphenyl
5.3.10 Benzenesulfonic Acid! 5-40
5.3.11 .Export of Benzene 5-41
6.0 NONCONSUMPTIVE USES OF BENZENE 6-1
6.1 TOTAL NONCONSUMPTIVE USE 6-1
6.2 CATEGORIES OF USE 6-1
6.3 EMISSIONS BY CATEGORY OF USE 6-1
6.3.1 Use of Benzene as a Solvent 6-1
6.3.2 Changes in Benzene Inventory 6-5
7.0 USE OF BENZENE AS A FUEL COMPONENT 7-1
7.1 BENZENE IN GASOLINE 7-1
7.1.1 Concentration of Benzene in Gasoline 7-1
7.1.2 Amount of Benzene in Gasoline 7-1
7.1.3 Benzene Emissions from Gasoline Use 7-2
8.0 SUMMARY OF DISPOSAL/DESTRUCTION AS END-PRODUCTS 8-1
9.0 LOCATIONS OF BENZENE EMISSION SITES 9-1
10.0 SUMMARY OF UNCERTAINTIES 10-1
iii
-------
11.0 DATA GAPS AND RECOMMENDATIONS 11-1
11.1 EMISSIONS DUE TO BENZENE PRODUCTION BY COKE H-l
OVEN PLANTS
11.2 BREAKDOWN OF EMISSIONS DUE TO PETROLEUM H-J.
REFINING BY METHOD
11.3 BENZENE EMISSIONS TO WATER J.1,-2
11.4 TREATMENT OF SOLID RESIDUES 1.1-2
11.5 EMISSIONS DUE TO MINOR CONSUMPTIVE USES 11-2
REFERENCES
APPENDIX A
APPENDIX B
iv
-------
TABLE OF FIGURES
Figure Page
2.1 Petroleum Based Benzene Producers, 1978 2-2
2.2 Coal-Derived Benzene Producers-Locations 2-21
2.3 Flow Diagram for Coal Processing 2-23
3.1 Benzene: Refinery Emissions by State (kkg) 3-4
3.2 Benzene: Coke Oven Emissions by State (kkg) 3-8
5.1 The Manufacture of Ethylbenzene Employing 5-4
Aluminum Chloride as Catalyst
5.2 Production of Ethylbenzene from Benzene 5-6
Locations (1976)
5.3 Solid Phosphoric Acid Process for Cumene Production 5-10
5.4 Cumene Production: Producers and Locations 5-11
5.5 Hydrogenation of Benzene 5-14
5.6 Production of Cyclohexane from Benzene, 1978 5-15
5.7 Flow Chart for Maleic Anhydride Synthesis 5-18
5.8 Maleic Anhydride Producers and Locations 5-21
5.9 Process Flow for Nitrobenzene Synthesis 5-22
5.10 Production of Nitrobenzene from Benzene 5-23
Locations (1976)
5.11 Schematic Diagram for the Production of Chlorobenzene 5-26
and Dichlorobenzenes
5.12 Production of Chlorobenzenes from Benzene: 5-28
Locations
5.13 Production of Alkylbenzenes 5-35
5.14 Biphenyl from Thermal Dehydrogenation of Benzene 5-44
7.1 Gasoline Distribution System 7-3
A-l Environmental Flow Diagram for Benzene A-l
B-l Platforming Method of Catalytic Reformation B-l
B-2 Separation of Benzene from Catalytic Reformate B~3
by the Sulfolane Process
B-3 Toluene Disproportionation by the Tatoray Process B~^
B-4 Recovery of Benzene from Pyrolysis Gasoline by the B-5
Pyrotol Process
B-5 Recovery of Benzene from Pyrolysis Gasoline B-6
(Dripolene) by the IFP ( Institute Francaise de
Petrole) Process
-------
TABLE OF TABLES
Page
General Benzene Emission Factors for Benzene 2-12
Production from Petroleum
2.2 Benzene Emission Factors for Storage 2-15
2.3 Benzene Emission Factors for Loading Operations 2-17
2.4 Benzene Emission Factors for Transportation 2-19
3.1 Benzene Emission Factors - Petroleum Refineries 3-2
3.2 Benzene Emission Factors - Coal Coking Operations 3-7
4.1 Benzene Imports, 1974-1979 (kkg) 4-1
5.1 Consumptive Uses of Benzene (1978) 5-2
5.2 Consumptive Uses of Benzene, 1975-1979 5-45
5.3 Summary of Estimated Benzene Emissions to Air Due 5-7
to Ethylbenzene
5.4 Summary of Estimated Benzene Emissions to Air Due 5-24
to Nitrobenzene Synthesis te
5.5 Estimated Chlorobenzenes Production and Benzene 5-31
Requirement by Companies
5.6 Benzene Emission Factors (kkg of Benzene Emitted 5-33
per kkg of Chlorobenzenes Produced)
5.7 Benzene Emissions Due to Chlorobenzenes Manufacture 5-33
5.8 Emission Factors and Emissions Due to Benzene 5-42
Export, 1978
6.1 Nonconsumptive Uses of Benzene 6-2
6.2 Estimated Emissions of Benzene as a Solvent 6-4
9.1 Locations of Benzene Emission Sites, 1978 9-2
10.1 Summary of Uncertainties 10-2
vi
-------
EXECUTIVE SUMMARY
This Level I materials balance on benzene was performed in response
to Task Order No. 16 under Contract No. 68-01-5793 with the Office of
Pesticides and Toxic Substances, EPA.
The Materials Balance Summary Diagram for benzene is included at the
end of this Executive Summary. It shows only those processes for which
data were obtainable. The more comprehensive Environmental Flow Diagram
for benzene appears as Appendix A of the report. All data refer to the
year 1978 unless otherwise stated.
DIRECT PRODUCTION
Benzene is obtained by the fractionation and enrichment of fossil
fuels. It is produced from petroleum by four processes: 1) catalytic
reformation of the naphtha petroleum fraction, 2) dealkylation of
toluene, 3) disproportionation of toluene, and 4) as a byproduct of
ethylene manufacture (isolation from pyrolysis gasoline). Benzene is
obtained from coal by extracting it from the light oil formed during
coking. This extraction may be done either at the coking plant or after
sale of the light oil to refineries. The contribution by each process
and the total production of benzene from petroleum are shown below.
Process kkg Benzene Produced, 1978 % of Total
Catalytic reformation 2,360,000 48
Toluene dealkylation 1,300,000 26
Toluene disproportionation 121,000 2
Pyrolysis gasoline 925,000 J.9
PETROLEUM SUBTOTAL 4,710,000 95
Extraction of light oil
by coking plants 178,000 4
Extraction of light oil
by refineries or process
unknown 74,000 1
PETROLEUM AND COAL TOTAL 4,960,000 100
vii
-------
Total benzene emissions due to the four production processes (includ-
ing storage and transport) were estimated to be 2,400-7,900 kkg. These
were estimated to be distributed 1,800-7,300 kkg to air, and 630 kkg to
water. No estimate of releases to landfills was possible. The ab.ove
emissions include benzene released during gasoline production by refiner-
ies.
Benzene emissions due to benzene-producing coke oven operations
could not be estimated.
INDIRECT PRODUCTION
Indirect sources of petroleum were analyzed. Their total estimated
benzene emissions were: refinery operations (20,000 kkg to air, 1 kkg
to water), coke oven operations (3,000-59,000 kkg to air), oil spills
(30 kkg to water), and miscellaneous operations listed on the Materials
Balance Summary Diagram.
IMPORTS
225,000 kkg of benzene were imported in 1978. The estimated emis-
sions due to importing were 13 kkg to air and 13 kkg to water.
CONSUMPTIVE USES
The predominant use of benzene is as a starting material for the
synthesis of other organic compounds. The nine major direct derivatives
of benzene and their contributions to total benzene consumption are
listed in Table ES-1.
A summary of the emissions due to synthesis of each product is
shown in Table ES-2.
EXPORTS
Exports accounted for 151,000 kkg of benzene in 1978. Estimated
emissions due to exportation (dockside loading) were 15 kkg to air,
2 kkg to water, and a negligible amount to land.
viii
-------
Table ES-1 Summary of Consumptive Uses of Benzene
Product
Ethylbenzene
Cumene
Cyclohexane
Nitrobenzene
Maleic Anhydride
Chlorobenzenes
Alkylbenzenes
An t hr a qu inon e
Biphenyl
Formed by Reacting
Benzene with -
Ethylene
Propylene
Hydrogen
Nitric Acid
Oxidant
Chlorine
10- to 14- Carbon
alkyl chlorides
Phthalic Anhydride
Itself (with heat)
Secondary Products
or Uses
Styrene; Polystyrene
Phenol
Cyclohexanone ; Nylon 66
Aniline
Chemical Intermediates
Chemical Intermediates
Detergents
Dyes
PCBs; Dyes
kkg Benzene
Used, 1978
2,890,000
1,058,000
811,000
170,000
161,000
133,000
132,000
23,000
11,000
5,389,000
Percentage of
Consumptive Uses
54
20
15
3
3
2
2
0.4
0.2
99.6
-------
Table ES-2 Emissions of Benzene Due to Consumptive Uses (1978)
Product
Synthesized
Ethylbenzene
Cumene
Cyclohexane
Nitrobenzene
Maleic Anhydride
Chlorobenzenes
Alkylbenzenes
Anthraquinone
Biphenyl
TOTALS
Estimated Emissions (kkg) to-
Air Water Land
2,600
14,000
Uk
2,970
2,200
34,000
2,060
170
Uk
65
44,100-
55,500
720
Uk
30
Uk
Uk
Uk
Uk
Uk
0.3
750
Uk*
Uk
Uk
Uk
Uk
Uk
Uk
Uk
Uk
Uk
Total
3,300-
15,000
380
3,000
2,200
34,000
2,060
170
Uk
64
45,100-
56,800
Percent of
Total
7-26
0.7 - 0.8
5.3 - 6.7
3.9 - 4.9
60 - 75
3.6 - 4.6
0.3 - 0.4
Uk
0.1
*Uk = Unknown
The range of values is not a statistical range or set of error limits, but results from the range
of total values used.
-------
NONCONSUMPTIVE USES
Approximately 5.1 percent of benzene production was used non-
consumptively in 1977, i.e., benzene was not converted to another com-
pound before use. The categories of nonconsumptive use and the amounts
used are:
Use kkg Benzene Used
Solvent 27,000 (higher estimate)
Inventory Increase -272,000
Solvent use of benzene has decreased since the 1977 OSHA Emergency
Benzene Standard and the 1977 Consumer Products Safety Commission ban on
benzene in consumer goods. 1978 emissions due to solvent use were esti-
mated to be 1,450 kkg to air, 1,450 kkg to water, and an unknown amount
to land.
The inventory of benzene as of January 1, 1979, was 272,000 kkg
less than on January 1, 1978. This decrease in inventory is treated as
increased supply when balancing benzene sources with benzene sinks.
BENZENE AS A FUEL CONSTITUENT
The largest single source of benzene emissions is due to its pre-
sence as a minor component of gasoline. The production of gasoline by
refineries is a major indirect source of benzene, generating an esti-
mated 7 x 10 kkg of benzene in 1979. Emissions due to refining are
included under petroleum refining operations above. Other sources of
emissions are listed below:
Process Emissions to air, kkg, 1979
Gasoline storage 4,000
Gasoline transport 3,000
Gasoline vending 7,000
Gasoline combustion 30,000 - 70,000
Sum: 40,000 - 80,000 kkg
xi
-------
The difference between the amount of benzene consumed in gasoline
6 4
(7 x 10 kkg) and total benzene emissions due to fuel use (4-8 x 10 kkg)
was 6.92-6.96 x 10 kkg. This value represents the difference: (benzene
destroyed by combustion minus benzene generated by combustion). Since
the amount generated is not known, this difference is reported as ben-
zene destroyed.
Geographic distribution of benzene emissions due to gasoline use
had not yet been analyzed at the time of submission of this revised
draft.
DISPOSAL/DESTRUCTION OF END-PRODUCTS
No information was readily available to permit estimation of benzene
emissions due to landfilling or incineration of benzene-containing solid
residues or of benzene-containing end-products.
LOCATIONS OF MAJOR BENZENE EMISSION SITES
Tabulation of sites and production/consumption volumes at those
sites indicated that the top five counties for potential benzene emis-
sions were in Texas: Harris, Galveston and Brazoria Counties (Houston/
Galveston area), Nueces County (Corpus Christi), and Jefferson County
(Beaumont/Port Arthur). The sixth most likely emission region is Puerto
Rico. This tabulation does not include emissions due to the presence of
benzene in gasoline.
UNCERTAINTIES AND DATA GAPS
Whenever a basis for evaluation existed, an attempt was made to
quantify the uncertainties of estimates. These uncertainties were ex-
pressed as a range of values.
The following data gaps were encountered in performing this study:
1) Lack of information on benzene emissions at coke ovens. A literature
search and industry inquiries are recommended. 2) Lack of breakdown of
petroleum-based benzene emissions by process (catalytic reformation,
xii
-------
toluene disproportionation, etc.)- A literature search and possible
monitoring of air emissions are recommended. 3) Lack of information on
benzene emissions to water. These emissions are generally assumed to be
small, but the assumption is important and needs to be confirmed by a
literature search and industry inquiries. 4) Lack of information on
generation and disposal of benzene-containing solid residues during
production and consumptive uses. Industry inquiries are recommended.
5) Lack of information on emissions due to minor consumptive uses:
synthesis of anthraquinone and biphenyl. A search of specialized lit-
erature is recommended. For instance, dye industry literature may be a
source of information on anthraquinone, and literature on PCBs may yield
data on biphenyl emissions.
SUMMARY OF EMISSIONS BY CATEGORY
The table below summarizes estimated emissions by major category of
production and use.
Estimated kkg Emitted to-
Category Air Water Land Total
Production from 1,800- ,,- ? 2,400-
Petroleum 7,300 7,900
Production from coal ? ? ? ?
Imports 13 13 ? 25
Indirect 23,000- 200- , 23,000-
Production 79,000 11,000 ' 90,000
Consumptive 44,000- nnn „ 45,000-
Uses - 56,000 i>uuu ' 57,000
Nonconsumptive ' Q
Uses ,
Component or 40,000- t ? ? 40,000-
Gasoline 80,000 ' ' 80,000
SUMS: 110,000- 3,000- , 110,000-
220,000 14,000 ' 240,000
1. A range of values is not meant to be a statistical range or set of
error limits, but results from differing independent estimates of
the emissions.
xiii
-------
SUMMARY; MATERIALS BALANCE FOR BENZENE
The diagram summarizes the data and estimates available at present
for the materials balance for benzene. It shows a breakdown of the
categories in the table above Into individual processes. Operations
for which no data were available were not included. In particular,
there are no entries under Carryover to Product and Destruction.
These will be important topics for a Level II materials balance on
benz ene.
The "balance" equation for the materials balance is shown below.
It was formulated as follows:
Benzene available = Benzene accounted for by uses + losses +
storage
Production + Imports = Consumptive uses -f Exports + Inventory
changes + Destruction + Emissions
12,300,000 kkg = 5,480,000 kkg + (-272,000 kkg) + 6,920,000
+ 240,000 kkg
12,300,000 kkg - 12,400,000 kkg
The apparent imbalance of 100,000 kkg may be explained by the
fact that consumptive use amounts were calculated independently from
direct production amounts. The imbalance also obscures the fact
that 21,000 kkg used as a solvent were unaccounted for as emissions,
and thus can be considered as having been "stored". The exces-s of
121,000 kkg on the "uses" side of the balance represents one percent
of the benzene available. We feel this is within the uncertainty
range of the respective totals.
xiv
-------
EXECUTIVE SITARY - ntt:3|.MS BALA'tCE FOR -E'tZESE, 1978 <«c>
COHSO&TIVI UJT5
1100»-•* (*etor of :-LO
(;)-*• decor of 10-tOOO
-------
1.0 INTRODUCTION
This Level I materials balance on benzene was prepared in response
to a task order from Survey and Analysis Division, U. S. Environmental
Protection Agency. The report is organized according to the processes
during which benzene might be emitted to the environment (direct pro-
duction, consumptive use, etc.). In order to present the large amount
of available information most clearly, current (usually 1978) data are
presented in tables interspersed in the text of each chapter and five
year trends (1979-1975) are presented in tables at the end of the
chapter.
In the tables and figures reporting emissions data and estimates,
the term "releases to land" was defined as meaning "material applied
to the soil," as opposed to material applied to a landfill. A landfill
has been interpreted as a form of waste storage rather than as a sink.
The rationale was that benzene-containing solids in a landfill would
readily lose benzene by leaching or evaporation to yield water or air
releases. Therefore, benzene-containing solids applied to a landfill
have been apportioned to air and water to reflect long-term reality.
On the other hand, actual "releases to land" would include use of
pesticides, fertilizers, or other chemicals that would yield benzene
due to microbial or photochemical degradation. Such releases would
be tabulated as indirect production of benzene.
1.1 PROPERTIES OF BENZENE
Benzene is a clear, flammable liquid with a pungent odor. Its
physical properties are summarized below (Weast, 1977-78; Strecher,
1968):
Melting Point 5.5°C
Boiling Point 80.1°C
Density 0.87865 g/cc
Solubility in HO 0.70 g/1 at 25°C
Temperature at which
vapor pressure =
1 Torr -36.7°C
1-1
-------
Benzene's physical properties indicate a compound likely to be
emitted to air or to evaporate to air from the surface of a biphasic
water mixture. Benzene's density value has been used to convert
volumes to kkg. No temperature correction was applied to the density
value.
1.2 ENVIRONMENTAL FLOW DIAGRAM FOR BENZENE
Appendix A shows the environmental flow diagram for benzene.
This diagram attempts to demonstrate as many as possible of the
sources and uses of benzene, as well as potential points of release
to the environment. It is not our purpose to complete the diagram—or
even to discuss it in detail—in this report. It serves as a guide,
however, in formulating the questions to be asked in this and other
benzene materials balances.
1-2
-------
2.0 DIRECT PRODUCTION OF BENZENE
Benzene is directly produced from two sources: from coal during
coke production, and from petroleum by various methods. While coal
was the original source of benzene (Kirk-Othmer, 1976), petroleum is
the primary source of benzene today. In 1978 according to the U.S.
International Trade Commission, 178,000 kkg of benzene were obtained
from coal while 4,780,000 kkg were derived from petroleum.
2.1 PRODUCTION FROM PETROLEUM
It has been known for many years that benzene is present in
small amounts in crude oil. Recovery of this benzene fraction is
uneconomical unless the benzene concentration is increased by some
catalytic or thermal reaction. There are four basic methods by
which benzene is produced from petroleum: (1) catalytic reforma-
tion, (2) dealkylation of toluene, (3) disproportionation of
toluene, and (4) from pyrolysis gasoline (Kirk-Othmer, 1976).
Some refineries also recover benzene from purchased light oil (coal
derived). This benzene is included in total production from coal.
2.1.1 Producers and Locations
Figure 2.1 shows the locations of U.S. plants producing benzene
from petroleum. The majority of these are located on the Texas Gulf
Coast, specifically the Houston and Corpus Christi areas.
An additional piece of important information on the figure is
the type of process used: catalytic reformation, dealkylation of
toluene, disproportionation of toluene, or isolation from pyrolysis
gasoline. The data are from SRI (1977) and A.D. Little, Inc. (1977),
Some plants were listed in the literature as using more than one
process without indication of relative capacities. In these cases,
plant capacity was assumed to be evenly distributed among the
2-1
-------
Locations
NJ
i
to
Unplotced - Channelview, TX; St. Crolx, VI; Penuelas, PR; Guayama, PR; Alliance, LA; Taft, LA
Figure 2.1 Petroleum-Based Benzene Producers, 1978
-------
NJ
I
UJ
Estimated Production
Company
Allied Chemical
Amerada Hess Corp.
American Petrofina, Inc.
(Cosden Oil & Chemical Co.)
Ashland Oil, Inc.
Atlantic Richfield Co.
Charter International Oil Co.
Cities Service Co., Inc.
Coastal States Gas Prod. Co.
Commonwealth Oil Refining Co.
(Commonwealth Petrochemicals)
Crown Central Petroleum Corp.
Dow Chemical Co.
Eastman-Kodak Co.
(Texas Eastman Dlv.)
Exxon Corp.
Getty Oil
Gulf Oil Corporation
Kerr-McGee Corp.
(Southwestern Oil & Ref. Co.)
Marathon Oil Co.
Mobil Oil Corp.
Monsanto Co.
Pennzoll United, Inc.
(Atlas Processing)
Phillips Petrol. Co.
Quint ana-Howell
Location
Winnie, IX
St. Croix, Virgin Islands
Port Arthur, TX
Big Spring, TX
Ashland, KY
North Tonawanda, NY
Houston, TX
Wilmington. CA
Channelviev, TX
Houston, TX
Lake Charles, LA
Corpus Chrlstl, TX
Penuelas, Puerto Rico
Pasadena, TX
Bay City, MI
Freeport, TX
Plaquemine, LA
Longview, TX
Baton Rouge, LA
Bay town, TX
El Dorado, KS
Alliance, LA
Philadelphia, PA
Port Arthur, TX
Corpus Christ!, TX
Texas City, TX
Beaumont, TX
Chocolate Bayou, TX
Shreveport, LA
Sweeny, TX
Guayama, Puerto Rico
Corpus Christ!, TX
j.979
162
144
159
57
139
29
162
12
62
174
460
57
74
124
149
174
149
32
167
92
186
39
17
149
211
87
24
273
224
,000
,000
,000
,200
,000
,800
,000
,400
,100
,000
,000
,200
,600
,000
,000
?
,000
,000
,300
,000
,000
,000
,800
,400
,000
.000
,000
,900
,000
.000
1978
146
130
144
51
126
27
146
11
56
157
416
51
67
112
157
135
29
151
83
169
36
15
135
191
78
22
247
202
.000
.000
.000
,700
,000
,000
,000
,200
,200
.000
,000
,700
,500
,000
i
,000
,000
,200
,000
,200
,000
,000
,700
,000
,000
.700
,500
,000
,000
1977
159.000
142,000
156,000
56.200
103,000
29,300
78,200
12,200
61,100
171,000
452,000
56,200
73,300
122.000
?
171.000
147,000
31,800
164,000
90,400
97,700
17,100
147,000
208.000
36,600
24,400
269,000
17,100
(kkg)
1976
8
71
42
128
142
42
125
34
14
71
199
525
65
85
142
185
176
"37
199
93
108
17
170
213
42
62
312
.520
.000
.600
,000
,000
,600
,000
,100
,200
,000
,000
,000
,300
,200
,000
?
,000
,000
.000
,000
,700
,000
,000
,000
,000
,600
,500
.000
Production Processes <•
1975 and Use
5,680
47,400
28,400
85,300
95,000
28,400
83,400
22,700
9,470
47,400
133.000
350,000
43,600
56,900
94,700
?
123,000
117,000
24,600
133,000
62,500
72 ,000
11,400
114 ,000
142,000
28,400
41,700
208,000
CR
C
CR. TD
CR, TD
CR, LO
CR, TP
CR
PG
CR
CR
25Z C,
PC, CR
CR. TD
C. TD,
C, TD,
CR, PG
67Z C.
C, CR
C. CR,
C, CR,
C, CR,
C, CR
CR, PG
C, CR,
PC
C, CR
PC, CR
CR
. LO
CR, TD
, TD, PG
PG, LO
PG
CR
TD
TD
PG
TD, PG
, TD
Figure 2.1 Producers and Production, 1975-1979 (Continued)
-------
NJ
I
-p-
Estimated Production (kkg)
Company
Shell Oil Co.
Standard Oil Co. of Calif.
Standard Oil Co. (Ind.) (AMOCO)
Standard Oil Co. (Ohio)
(B.P. Oil Co.)
Sun Oil Co.
Tenneco, Inc.
Texaco, Inc.
Union Carbide Corp.
Union Oil Co. of Calif.
Union Oil-American Petrofina
Union Pacific Corp.
(Champlln Petroleum Co.)
TOTAL USITC
Location
Deer Park, TX
Odessa, TX
Wood River, IL
El Segundo, CA
Texas City, TX
Marcus Hook, PA
Marcus Hook, PA
Corpus Chris ti, TX
Tulsa, OK
Toledo, OH
Chalmette, LA
Port Arthur, TX
Westvllle, NJ
Taft, LA
Lemon t , IL
Beaumont, TX
Corpus Christ!, TX
PRODUCTION
1979
298
29
112
57
211
72
94
59
ISA
24
112
87
174
42
54
24
5,430
,000
,800
,000
,200
,000
,100
,500
,700
,000
,900
,000
,000
.000
,300
,700
,900
,0003
1978
270
27
101
51
191
65
85
54
166
22
101
78
157
38
49
22
,000
,000
,000
,700
,000
,200
,400
,000
,000
,500
,000
,700
,000
,200
,500
,500
4,780,000
1977
220,
29,
110,
56,
208,
70,
92,
58,
120,
24,
110,
85,
171,
41,
53,
24,
4,570
000
300
000
200
000
800
800
600
000
400
000
500
000
500
700
400
,000
1976
213,000
17 ,000
114,000
65,300
22,700
42,600
99,400
68,200
28,400
128,000
99,400
199,000
54,000
54,000
28,400
4,540,000
1975
142.000
11,400
76,000
43,600
161,000
15,200
28,400
66,300
45,500
18,900
85,300
66.300
133,000
36,000
36.000
18.900
3,190,000
Production Proccesses
and Use
C, CR,
CR, TD
CR, LO
C, CR
C, CR
CR, TP
C, CR,
CR, TP
CR, TD
CR, LO
73Z C,
PC, CR
C, PC
CR, LO
50Z C,
C, CR
PC
TD
CR
TD
1. Sources: A.D. Little, Inc., 1977; SRI, 1977; Versar, 1979; Neufeld ei al., 1978.
2. Key. C- captive use, PC" partially captive, CR- catalytic reformation, TD- toluene dealkylation,
TP- toluene dlsproportlonation, PC- pyrolysle gasoline, LO- light oil.
3. Derived from plant capacities and USITC production totals as described in Section 2.1.2.
4. Estimated by extrapolating USITC data for the months January through July, 1979.
Table 2.2 Producers and Production, 1975-1979 (Continued)
-------
processes listed. No allocation was made to the process of recovering
benzene from light oil unless specifically indicated in the literature.
The final column on this figure indicates if the production was
captively used (A.D. Little, Inc., 1977; SRI, 1977).
2.1.2 Amounts Produced
Figure 2.1 lists U.S. producers of benzene from petroleum and the
estimated production by each plant. The years 1975-1979 are included.
Company names and locations were compiled from information contained
in USITC publications, Versar (1979), Neufeld et_ al. (1978), and SRI
(1977).
Estimated production (kkg) for a given plant in a given year was
A
calculated using the following formula: Estimated production = — x C,
o
where A = total production of benzene by petroleum refineries for that
year as per USITC information, B = total U.S. plant capacity for that
year as per SRI information minus the capacity of plants not listed by
USITC as having produced benzene that year, and C = the capacity of
the individual plant named (from SRI). Since USITC production is in
gallons, a conversion factor of 3.33 x 10 kkg = 1 gal has been
applied. The other bases for calculation of production figures were:
(.1) It was assumed that an individual plant's production was propor-
tional to its capacity. (2) Estimates for 1979 were based on USITC
data from January - July which were extrapolated to cover the whole
year. (3) Plant capacities for 1975-1976 were assumed to be equal
to the plant capacities for 1976 as listed in Versar (1979).
(A) Plant capacities for 1977-1979 were assumed to be equal to the
plant capacities indicated in SRI (1977).
2.1.3 Production by Catalytic Reformation
Of the four methods of producing benzene from petroleum, catalytic
reformation is the most commonly used. Of the 43 refineries believed
to have been producing in 1978 and for which we also had data on pro-
duction methods, 37 were reported to have the capacity to produce
2-5
-------
benzene by catalytic reformation (See Figure 2.1). The aromatics
fraction of petroleum is enriched by the process, which basically
converts non-aromatics to aromatics (including benzene) with such
reactions as dehydrogenation, isomerization, cyclization, dealkyla-
tion, and combinations thereof. The reaction conditions and the
catalyst determine which reactions predominate and these reactions'
kinetics. Benzene is extracted from the aromatic-rich reformate in
a subsequent separation process.
2.1.3.1 Amounts Produced by Catalytic Reformation
Total capacity for benzene production by petroleum refineries
was estimated to have been 7,080,000 kkg during 1978. Of this amount
3,500,000 kkg were estimated to be capacity available to the catalytic
reformation process (SRI, 1977; Neufeld _et '_al., 1978). Based on a
total petroleum-derived benzene production (as opposed to capacity)
of 4,780,000 kkg (USITC), 2,360,000 kkg were attributed to production
by catalytic reformation. Assuming production was proportional to
capacity, for 1978:
Production (kkg) by catalytic reformation = (total production, kkg)
(capacity of catalytic reformation plants) .
(total 1978 benzene capacity)
Information on plants having catalytic reformation capacity is
presented in Figure 2.1.
2.1.3.2 Emissions Due to Catalytic Reformation
Possible sources of emission during catalytic reformation are
listed below. These are based on the method called Platforming, a
process flow chart of which is presented in Appendix B.
1) Catalyst regeneration by controlled combustion to remove coke.
2) Separator gas.
One method of extracting benzene from the aromatic-rich reformate
is the Shell-UOP Sulfolane extraction process. Possible sources of
2-6
-------
emissions from this method are listed below. A process flow chart is
presented in Appendix B.
1) Raffinate wash water
2) Raffinate
It has been estimated that less than 1% of the benzene produced
was lost through reforming operations (Walker, 1976). Thus, a maximum
_2
emission factor would be 1 x 10 kkg/kkg benzene produced by catalytic
reformation. Application of this emission factor to benzene produc-
tion by this method yielded: (2,360,000 kkg) (1 x 10~2 kkg/kkg) =
20,000 kkg benzene released during production by catalytic reformation
of petroleum. This is a maximum value.
No factors for emissions specifically to air, water or land by
the catalytic reformation process were found in the literature, thus
no emissions to these compartments could be calculated. See also
Section 2.1.7.
2.1.4 Production by Dealkylation of Toluene
A second method of benzene production from petroleum is by
dealkylation of toluene. Here the petroleum has already been refined
to produce toluene, thus raw material costs make benzene produced by
this method more expensive than from other sources (Kirk-Othmer, 1976).
Dealkylation performed in the presence of hydrogen is known as hydro-
dealkylation. Toluene hydrodealkylation is performed by a catalytic
or thermal process. These processes are equally split in terms of
frequency of use, and are very similar to each other. In both pro-
cesses, toluene is heated and then charged to a reactor in the
presence of excess hydrogen. Toluene reacts with the hydrogen to
produce benzene and methane gas. The reaction product is separated,
stabilized, and distilled to recover benzene. The only difference
in the two processes is that the toluene is heated to a higher tempera-
ture in the thermal process, and the reaction occurs in the presence
of a catalyst with the catalytic process (Kirk-Othmer, 1976).
2-7
-------
A newer method of dealkylation, steam dealkylation, reacts toluene
with steam to produce benzene, carbon monoxide, carbon dioxide and
hydrogen. The advantage is that hydrogen is produced rather than con-
sumed (Kirk-Othmer, 1976).
Of the 43 refineries believed to have been producing benzene in
1978 and for whom we have data on production methods, 15 were reported
to have the capacity to produce benzene by toluene dealkylation. (See
Figure 2.1.)
2.1.4.1 Amounts Produced by Toluene Dealkylation
Estimated capacity for benzene production by toluene dealkylation
was 1,920,000 kkg for 1978 (SRI, 1977; Neufeld et al., 1978). The
actual production is estimated to have been 1,300,000 kkg based on
USITC data, and calculated using the formula presented in Section
2.1.3.1 above.
Information on plants having toluene dealkylation capacity is
presented in Figure 2.1.
2.1.4.2 Emissions Due to Toluene Dealkylation
Possible sources of emission during toluene dealkylation are:
1) Separation of reaction product (benzene, methane)
2) Distillation
3) Regeneration of catalyst (.catalytic method only)
4) Separation of benzene from CO, C02, + H~ (steam dealkylation)
No factors for emissions specifically to air, water or land by
the toluene dealkylation process were found in the literature, thus no
emissions to these media could be calculated. Total emissions due to
petroleum-based production are discussed in Section 2.1.7.
2-8
-------
2.1.5 Production by Toluene Disproportionation
Toluene disproportionation, also called transalkylation, is the
least used of the four methods of producing benzene from petroleum.
As with dealkylation, petroleum has already been refined to produce
the toluene feedstock (Kirk-Othmer, 1976). Since xylene is simul-
taneously produced with benzene in this process, the xylene may contain
benzene as a contaminant.
One method for toluene disproportionation is the Tatoray Process.
An industrial process flow chart for this process is described in
Appendix B.
Of the 43 refineries believed to have been producing benzene in
1978 and for which we also had data on production methods, only 3 had
the capacity to produce benzene by toluene disproportionation. (See
Figure 2.1.)
2.1.5.1 Amounts Produced by Toluene Disproportionation
Estimated capacity for benzene production by toluene dispropor-
tionation was 180,000 kkg for 1978 (SRI, 1977; Neufeld, et_ al., 1978).
The actual production was estimated to have been 121,000 kkg as
calculated from USITC data and capacity information as described in
Section 2.1.3.1.
Information on plants having toluene disproportionation capacity
is presented in Figure 2.1.
2.1.5.2 Emissions Attributable to Toluene Disproportionation
Possible sources of emissions due to the Tatoray process pre-
sented in Appendix B were:
1) Regeneration/disposal of catalyst
2) Separation process
No factors for emissions specifically to air, water or land by the
toluene disproportionation process were found in the literature; no
2-9
-------
specific emissions were estimated. Total emissions due to petroleum-
based production are discussed in Section 2.1.7.
2.1.6 Production from Pyrolysis Gasoline
In this production method, heavy naphthas or gas oils are steam
cracked to produce ethylene and a liquid by-product high in unsaturated
aliphatic and aromatic hydrocarbons called dripolene or pyrolysis
gasoline. Benzene, toluene, and xylene may be removed from the
pyrolysis gasoline by extractive distillation (Kirk-Othmer, 1976).
Industrial process flow charts for two extraction processes (Pyrotol
and IFF) are presented in Appendix B.
Of the 43 refineries believed to have been producing benzene in
1978 and for whom we also had data on production methods, 11 were
reported to have the capacity to produce benzene from pyrolysis
gasoline.
2.1.6.1 Amounts Produced from Pyrolysis Gasoline
Estimated capacity for benzene production from pyrolysis gaso-
line was 1,320,000 kkg for 1978 (SRI, 1977; Neufeld £t al., 1978).
The actual production was estimated to have been 925,000 kkg based
on USITC data and capacity information. The estimation method is
described in Section 2.1.3.1.
Information on plants having capacity for production from pyroly-
sis gasoline is presented in Figure 2.1.
2.1.6.2 Emissions Due to Production from Pyrolysis Gasoline
Possible sources of emissions are:
1) Catalyst regeneration/disposal (IFP process) (.gum removal
and coke burnoff)
2) Distillation process
3) Separation of benzene from toluene and xylene
2-10
-------
4) Pyrotol process steps
a) gum removal
b) desulfurization
c) hydrodealkylation of toluene (.see above)
d) removal of light gases in stabilizer
e) clay treatment to remove contaminants from stabilized
product
f) distillation
No factors for emissions specifically to air, water or land by
the pyrolysis gasoline process were found in the literature; thus, no
emissions could be estimated to environmental compartments. Total
emissions due to petroleum-based production are discussed in Section
2.1.7.
2.1.7 General Emissions Due to Petroleum-Derived Benzene Production
With the exception of catalytic reformation, no emission factors
for the various individual processes by which petroleum-derived benzene
may be produced were found in the literature.
Table 2.1 presents two emission factors—one for releases to air
and one for releases to water—which may be applied to petroleum-
derived benzene production generally (as opposed to production by a
specific process).
Applying these factors to the 1978 petroleum-derived benzene pro-
duction yields approximately 85 kkg benzene emitted to air and 610 kkg
benzene emitted to water.
The emission factor for releases to air was based on a Union
Carbide estimate for a plant utilizing a crude olefins byproduct stream
and is therefore not necessarily representative of other producers.
The emission factor for releases to water was based on an average
figure for benzene loss to the environment as reported in an industrial
questionnaire. The actual values reported in the questionnaire ranged
from 0.0006% ti 0.44%. Taking a simple average was probably not valid.
Also this emission factor required the assumption that 6% of this loss
was to water. No basis for this 6% value was given.
2-11
-------
Table 2.1 General Benzene Emission Factors for Benzene Production from
Petroleum
Source Factor Remarks
— Releases to Air —
Based on 1.8 x 10 kkg/kkg Based on a Union Carbide
PEDCo, 1977 produced estimate of 0.035 Ib/ton
produced and metric conver-
sion factor 454 x 10~6 kkg/lb
and 0.907 kkg/ton. This
estimate assumed an aromatics
plant utilizing a crude
olefins byproduct stream.
— Releases to Water
Based on 1.3 x 10~4 kkg/kkg (X) x (0.22%) x (6%)
Versar, 1977 produced Where (X) = annual benzene
production, (0.22%) = aver-
age annual production lost
to the environment (derived
from an industrial ques-
tionnaire) , (6%) = amount of
above allocated to water.
2-12
-------
The relative sizes of the two emission estimates are also incon-
sistent with previous results and engineering judgements. The value
for emissions to water was greater than that to air. Nothing in the
literature indicates that water plays an important part in benzene
production. Benzene is volatile and one would expect greater emissions
to air. Looking at refinery emissions in general (Section 3.1),
emissions to air were considerably greater than emissions to water.
This inconsistency may be resolved by more detailed analysis of the
primary data in a Level II study.
In estimating the uncertainty of the estimate of emissions to
air during benzene production, the main consideration was the validity
of extrapolating the Union Carbide single-plant estimate to the entire
benzene-from-petroleum industry. While a quantitative estimate of the
uncertainty range was not possible, it is suggested that extreme cau-
tion be exercised when basing conclusions on the air emissions
estimated by this method.
In estimating the uncertainty of the emissions to water during
benzene production, two factors were taken into account: (1) the 730-
fold range for industrial estimates of percentage of benzene lost to
the environment, and (2) the allocation to water of 6% of total losses.
Based on these two factors, an uncertainty of + a factor of 1,000 was
assigned to the estimate of benzene emissions to water.
2.1.8 Emissions Due to Transportation, Loading and Storage Associated
with Production of Benzene from Petroleum
2.1.8.1 Sources of Emissions During Storage
Benzene is typically stored in a floating roof storage tank of
which there are three types: pan, pontoon, and double-deck floating
roof.
With the pan type, extreme tilting may cause the roof to buckle
or sink causing a high loss of benzene vapor to the atmosphere. If
the roof is in direct contact with the liquid, heating caused by the
sun shining on the roof may cause the liquid to boil resulting in an
2-13
-------
emission to the air from around the roof's perimeter. The pontoon
type and double-deck type were designed to overcome these problems.
All tanks lose vapors at gauging hatches, sample hatches and
relief vents unless these points are designed and maintained for
proper closure. Also, liquid surface may be exposed to the atmosphere
resulting in emissions if the seal between the roof and vessel is
improperly fitted. (PEDCo, 1977.)
Benzene loss due to storage can be characterized in two ways,
storage standing loss and storage withdrawal loss.
The storage standing loss is caused by the factors mentioned
above and increases with the length of storage time.
The act of withdrawing benzene from the tank increases the amount
lost, usually from the evaporation of benzene retained on the sides
of the tank as the roof sinks (PEDCo, 1977).
Emission factors for these losses are presented in Table 2.2.
The first column lists the source of the factor; the second column
lists the factor; and the "remarks" column describes how the factor
was derived either directly in the reference or from reference data,
the assumptions to be used when applying the factor, and other infor-
mation.
Applying the first two factors listed in the table to the total
1978 benzene consumption of 5,389,000 kkg as reported in Section 5.1,
_3
and using a conversion factor of 3.33 x 10 kkg/gallon, it was
estimated that 100 kkg of benzene were lost due to standing storage,
and 5.4 kkg were lost due to storage withdrawal. The sum of these was
105 kkg lost to air due to storage.
Applying the SRI emission factor to total benzene consumption in
1978, assuming uncontrolled storage, and assuming that all benzene
consumed has been stored, 4,900 kkg of benzene were lost due to storage.
In the absence of information that would permit choosing between
these two disparate emission estimates, the emissions to air due to
storage are presented as 105-4,900 kkg.
2-14
-------
Table 2.2 Benzene Emission Factors for Storage
Source
Factors
Remarks
ro
I
Based on
PEDCo, 1977
PEDCo, 1977
— 8
8 x 10 kkg/gal benzene
consumed
Storage standing loss
(X) (30) (4.6 x 10 J)
(0.75) (2.3 x
...
= kkg benzene
Where X = benzene consumption in gallons
30 = retention time in days (assumed)
4.6 x 10 = tank emission factor kkg/tank/day
0.75 = tanks are 75% full (assumed)
2.3 x 10 = size average storage tank in gallons
3.4 x 10 9 kkg/gal ben-
zene consumed
Storage withdrawal loss
(X) (7.4 x 10"6)
2.2 x
kkg benzene emitted
Where X = benzene consumption in gallons
7.4 x 10 = withdrawal loss in Ibs/gal
2.2 x 103 = Ibs/kkg
SRI, 1978
3.0 x 10 6 kkg/gal ben-
zene stored
Uncontrolled Storage
0.3 x 10~6 kkg/gal
Internal floating roof,
vapor recovery
SRI included these estimates when calculating emissions
from certain refineries as presented in Figure 3.1.
-------
2.1.8.2 Sources of Emissions During Loading
Benzene is transported by rail tank cars, tank trucks, barges on
inland waterways, and pipelines. Before benzene is transported, it is
first collected and temporarily stored in a tank called a rundown
tank, where it is inspected for product quality. Then it is trans-
ferred to two sets of shipping tanks, one for railcar and truck loading
and the other for barge loading. The railcar and truck loading tank
also issues to the pipelines. Benzene losses from these tanks may be
characterized as standing losses (caused by evaporation around peri-
meter roof seals) and withdrawal losses (occurring as the tank is
emptied). Emission factors for these processes are presented in
Table 2.3. (Dunavent, 1978.)
Loading losses are produced as liquid benzene is pumped into the
carrier and vapors present (.either from previous loa'ds or currently
generated) are displaced. Emission factors for railcars, trucks, and
barges are also presented in Table 2.3. (Dunavent, 1978.)
Before applying these emission factors, it was necessary to
determine the amount of benzene that was not captively used in produc-
tion of other chemicals. Of a total production capacity of 2,125 x 10
gallons for companies actually producing benzene in 1978, 987 x 10
gallons are allocated to captive use, thus, 1,138 x 10 gallons may be
allocated to non-captive production. These figures are based on data
from USITC, A.D. Little (1977), and SRI (1977). It was assumed that
any refinery listed as "partly captive" without further breakdown is
actually 50% captive. Applying these data to total production by
refineries (including extraction of light oil), as shown below, the
total benzene production not captively used was determined:
1 138
2\25 x 4>780>000 kkS = 2,560,000 kkg benzene not captively used
Applying the emission factor to this number with a conversion factor
of 3.33 x 10~ kkg/gallon, and assuming that all the benzene passes
through the rundown tank, and 50% of the benzene is loaded on railcars
2-16
-------
Table 2.3 Benzene Emission Factors for Loading Operations
to
i
Source
Based on
Dunavent, 1978
Based on
Dunavent, 1978
Based on
Dunavent, 1978
Factor
Rundown Tank:
.-8
Standing: 6.4 x 10 kkg/gal
Withdrawal: 5.8 x 10~8 kkg/gal
Rail car/Truck loading tank
Standing: 1.7 x 10~7 kkg/gal
loaded
Withdrawal: 3.6 x 10~8 kkg/
gal loaded
Barge loading tank
Standing: 4> x 10 kkg/gal
loaded
Withdrawal: 3.0 x 10"8 kkg/
gal loaded
Loading Losses
Railcar: 1.3 x 10~6 kkg/gal
loaded
Truck: 1.3 x 10"6 kkg/gal
loaded
Barge: 1.1 x 10~6 kkg/gal
loaded
Remarks
Converted from estimated emissions for a 40 x 10 gal capacity
petroleum-derived benzene plant working at capacity.
5,600 Ibs emitted
2.2 x 103 ibs/kkg
5.100 Ibs
2.2 x 103 Ibs/kkg
40 x 10 gal = emissions benzene/gal produced
40 x 10 gal = emissions benzene/gal produced
Converted from estimated emissions for a 40 x 10 gal capacity
petroleum-derived benzene plant working at capacity; 28 x 10*> gal of
this are handled through rallcar/truck loading tanks (14 x 10& gal of
this later placed In pipeline) ; the remaining 12 x 106 gal are handled
through barge loading tanks.
10.200 Iba • 28 x 106 gals loaded rail/truck
2.2 x 10-« Ibs/kkg ~
3 T
10
12.100 . 12 x 10
2.2 x 10J ~
800
2.2 x 10
3 - 12 x 10
Assume 10 x 10 gals loaded In rallcar and 4 x 10 gals loaded on
truck tankers; 12 x 106 gals loaded on barges.
3 , 10 x 10* gals
" * ">
-------
or trucks, and 50% is loaded on barges, the following emissions may
be determined:
From rundown tanks -
standing: 49 kkg
withdrawal: 45 kkg
From railcar/truck loading tanks -
standing: 64 kkg
withdrawal: 14 kkg
From barge loading tanks -
standing: 176 kkg
withdrawal: 12 kkg
From loading losses to:
railcars/trucks: 500 kkg
barges: 423 kkg
TOTAL BENZENE EMISSIONS DUE
TO LOADING OPERATIONS: 1,300 kkg
No independent criteria were available with which to judge the
uncertainty ranges of these estimates.
2.1.8.3 Emissions During Transit
Emission factors for emissions during transit are summarized in
Table 2.4. The first factor listed assumes equal distribution between
rail/truck and marine transit, and does not include emissions due to
loading and unloading. Applying this factor to the amount of benzene
not captively used in 1978 yields 270 kkg of benzene lost due to
transit.
The SRI factors listed are for specific modes of conveyance. It
is not known whether these factors include emissions due to loading/
unloading. Applying these factors to the amount of benzene not cap-
tively used and assuming 50% is transported by rail-truck and 50% by
barge, yields:
losses from transit by rail-truck: 690 kkg
losses from transit by barge: 290 kkg
TOTAL TRANSIT LOSS: 980 kkg
2-18
-------
Table 2.4 Benzene Emission Factors for Transportation
Source
Factors
Remarks
to
i
Based on
PEDCo, 1977
3.6 x 10 7 kkg/gal
For Transit
(1.000)
3
Where (0.78) = assumed transit loss lbs/wk/10 gals
Where (2.2 x 103) = Ibs/kkg
(1,000) = gals/103 gals
Where (X) = gallons transported
Assume transport for 1 week
Here it also is assumed that half the benzene is trans-
ported by truck or tank car, half by marine operations.
Emissions due to loading/unloading are not included in
this factor.
SRI, 1978
0.76 x 10~6 kkg/gal
inland barge
1.8 x 10~6 kkg/gal
tank truck
1.8 x 10~6 kkg/gal
rail car
All above uncontrolled
It is unknown if these factors include losses from
loading/unloading
-------
No independent criteria were available with which to judge the uncer-
tainty ranges of these estimates.
2.1.8.4 Total Emissions Due to Transportation, Loading, and Storage
The sum of estimated emissions for transport, loading, and storage
operations for petroleum-derived benzene was:
105-4,900 kkg (storage) + 1,300 kkg (loading) + 980 kkg (transport)
= 2,400-7,200 kkg benzene emitted.
Due to the nature of the losses (evaporation), this total emission
was judged 'to be a release to air. The range of values presented is
neither a statistical range nor a set of error limits. It was not pos-
sible to estimate the uncertainties of the emission factors used;
therefore, the uncertainties of the emissions estimates could not be
determined.
2.2 BENZENE PRODUCTION FROM COAL
\
2.2.1 Summary
Figure 2.2 shows the production and location of producers of
benzene from coal. Company names and locations were compiled from
information contained in Versar (1979) and Neufeld iet_ al. (1978).
Estimated production (kkg) for the years 1975-1979 was calculated using
the following formula:
^
— x C = Estimated production
o
where. A = total U.S. production of coal-derived benzene for the.
year under consideration as per USITC information, B = total U.S. plant
capacity for that year, and C = the capacity of the individual plant
named (capacity information as per Versar, 1979; SRI, 1977; and Neufeld
et al., 1978). It was assumed that all plants with listed capacity
were producing benzene in proportion to that capacity. Since USITC
production is listed in 1,000 gallons, a conversion factor of
2-20
-------
LOCATIONS
PRODUCTION OF BENZENE FROM COM.
Coal-derived Benzene Producers
Company Location
Armco Steel Corp. Mlddletown, OH
Bethlehem Steel Corp. Bethlehem, PA
Lackawanna, NT
Sparrows Point,
Mead Corporation Chattanooga, TN
Woodward, AL
C.P. & I. Steel Corp. Pueblo, CO
Inter lake, Inc. Toledo, OH
Jones 4 Laughlln Aliqulppa, PA
Steel Corp. (LTV Corp)
Northwest Industries, Lone Star, TX
Inc. (Lone Star Steel Corp)
U. S. Steel Corp. Clairton, PA
Geneva, UT
Actual Total Production by U. S. Coke
Oven Operators (kkg) (DSITC)
1. Sources: Versar, 1979; Neufeld et
Estimated Production (kkg)
1979
6,500
8,700
0
MD 33,000
0
4.400
6,500
2,200
22,000
2,200
109,000
8,705
202,397
al. , 1978.
2. Estimates for Individual plants based on capacity
3, Sources: Versar, 1979; SM, 1977;
Neufeld et_ al;
1978
5,700
7,700
0
29,000
0
3,800
5,700
1,900
19,000
1,900
96,000
7,700
178,062
estimate!
,1978.
1977
7,300
9,800
0
37,000
0
4,900
7,300
2,400
24,000
2,400
110,000
9,800
215,021
i. See !
1976
6,400
8,600
17,000
32,000
0
0
6,400
2,100
21,000
2,100
96,000
8,600
201,169
ectlon 2
1975
6,900
9,200
18,000
35,000
0
0
6,900
2,300
23,000
2,300
104,000
9,200
216,617
.2.2.
Plant Capacities Used
In Estimations: 10 gal
1978-79
3
4
0
15
0
2
3
1
10
1
50
4
93
1977
3
4
0
15
0
2
3
1
10
1 '
45
4
88
1975-76
3
4
8
15
0
0
3
1
10
1
45
4
94
Figure 2.2 Coal-Derived Benzene Producers
2-21
-------
_3
3.33 x 10 kkg = 1 gallon has been applied. Estimates for 1979 were
based on USITC data from January-July which were extrapolated to
cover the whole year. Plant capacities for 1975-1976 are assumed to
be equal to capacity for 1976 as listed in Versar (1979). Plant
capacities for 1977-1979 are assumed to be equal to the 1977 plant
capacities indicated in SRI (1977) and Neufeld e± al. (1978).
2.2.2 Benzene Production from Coal - the Process
Benzene is produced as a byproduct of the carbonization of coal
to coke. Coal is heated in a byproduct oven in the absence of air,
driving off the volatile gases. These hot gases are collected over-
head and shocked-cooled with a flushing liquor, which results in the
removal of a large portion of the tars and inorganic salts. The
gases are further processed to remove additional tars and ammonia.
The gas is cooled once more with water, then scrubbed with a high-
boiling absorbant petroleum oil in a tall column. The wash oil
removes the light oil containing benzene, toluene, xylene, etc.,
from the gas, and the wash oil/light oil mixture is separated by
steam distillation. The crude light oil consists of 55-70% by volume
benzene. (A.D. Little, 1977.) The yield of light oil from coke
ovens producing blast furnace coke is 3-4 gals/ton coal carbonized
(.PEDCo, 1977).
Another source of light oil is coal tar. Coal tar may be frac-
tionally distilled to yield a light oil fraction which is usually
combined with the light oil from coal gas before further refining
for benzene (PEDCo, 1977). Light oil produced is either refined on
site or is sold. Several petroleum refiners refine this coal-derived
light oil (SRI, 1977; A.D. Little, 1977).
The light oil is refined by various processes resulting in its
separation into benzene, toluene, xylene, and residue fractions.
Benzene recovered from coke oven gas amounts to typically 1.85 gals/ton
coal carbonized .(PEDCo, 1977).
A schematic diagram summarizing benzene production from coal is
presented in Figure 2.3. This diagram is based upon information pre-
sented in Weissermel e£ al. (1978) and A.D. Little (1977).
2-22
-------
I
ho
U>
Sources: Weissermal & Arpe, 1978; A.D. Little, 1977
Figure 2.1 Flow Diagram for Coal Processing
-------
2.2.2.1 Amount Produced
In 1978, 178,000 kkg of coal-derived benzene were produced accord-
ing to the USITC. This was 4% of total benzene production.
2.2.2.2 Emission During Coal-derived Benzene Production
Possible sources of emissions during production of benzene from
coal are:
1) Flushing liquor shock cooling (removing tars and inorganic
salts)
2) Tar and ammonia removal
3) Second water cooling
4) Scrubbing process Cwashing with higher-boiling oils)
5) Wash oil/light oil separation and steam distillation
6) Light oil separation
No emission factors specifically related to the production of
benzene from coal were found in the literature, thus no emissions
attributable to this process were calculated. Emission factors for
the coal coking process in general were found, and are presented in
the section related to indirect production of benzene, Section 3.3.
No emission factors for transportation, loading and storage of coal-
derived benzene were found in the literature; therefore, no emissions
were calculated.
2.3 SUMMARY
The total amount of benzene produced in the U.S. according to the
USITC for 1978 was 4,960,000 kkg, of which 4,710,000 kkg were petroleum-
derived and 252,000 kkg were coal-derived. The petroleum-derived
portion may be further broken down by the following estimates:
2,360,000 kkg from catalytic reformation; 1,300,000 kkg from toluene
dealkylation; 121,000 kkg from toluene disproportionation; and 925,000
from pyrolysis gasoline. Coal-derived benzene was produced as follows:
178,000 kkg by coke-oven operators, and 74,000 kkg by extraction of
purchased light oil at refineries and by unnamed methods. Total
2-24
-------
estimated emissions from petroleum-derived benzene production
(exclusive of storage and transportation) were 85 kkg to air and
610 kkg to water. Emissions to land were unknown. Maximum total
emissions due to catalytic reformation were estimated to be 20,000
kkg. Losses to air from petroleum-derived benzene loading, storage
and transportation ranged from 2,400 to 7,200 kkg. Emissions specifi-
cally attributable to coal-derived benzene were unknown, but would
be included in any emissions from coke ovens (See Section 3.2).
2-25
-------
3.0 INDIRECT PRODUCTION OF BENZENE
3.1 INDIRECT PRODUCTION OF BENZENE FROM REFINING OPERATIONS
Since crude oil contains an average of 0.2 percent] benzene
(Walker, 1976), it can be expected that petroleum refining operations
in general would be a source of benzene emissions. The literature
revealed several emission factors for benzene to air and to water.
These are presented in Table 3.1.
The Mara and Lee (1978) factors were based on average hydrocarbon
emissions and percent of the total hydrocarbon emissions attributed to
benzene. It was estimated by Mara and Lee that emissions from refiner-
ies with catalytic reforming capacity would be double those of other
refineries. The extra emissions result from storage losses for gasoline
(approximately 50 percent), which is apparently produced from catalytic
reformate exclusively. Other sources of emissions are leaks and stacks.
Using a list of petroleum refineries and their capacities, Mara and Lee
applied these factors along with metric conversion factors to determine
total benzene emissions to air. Emissions due to storage and loading
of pure benzene were estimated and added for those refineries producing
pure benzene not captively used. Controlled emissions were assumed only
for those refineries where information on emission control technology
was available. Full-capacity production was assumed. These emissions
were totaled state by state and appear on the map presented in Figure 3.1.
Total U.S. benzene emissions to air from petroleum refineries operating
at 1977 capacity were estimated by this method to be 20,000 kkg.
The PEDCo emission factors, which were not applied, were consider-
ably less than those of Mara and Lee. The largest PEDCo factor, that
for uncontrolled refineries, was one half that of the smallest Mara
and Lee factor.
One factor for emissions to water from petroleum refining was cal-
culated based on data presented in Versar (1977). That source examined
3-1
-------
Table 3.1 Benzene Release Factors - Petroleum Refineries
Source :
Factor
Remarks
Mara and Lee
(1978)
PEDCo
(1977)
Based on
Versar
(1977)
Releases to Air from Petroleum Refineries
*4.6 Ibs Benzene
1,000 bbl crude
refined
*9.2 Ibs Benzene
1,000 bbl crude
refined
0.415 lbs/1,000 bbl
2.27 lbs/1,000 bbl
0.759 lbs/1,000 bbl
Calculated from
920 Ibs
(0.5%) where 920 Ibs = estimated hydrocar-
1,000 bbl
bon releases per 1,000 bbl and (0.5%) = estimated % of hydrocarbons
attributed to benzene. Applies to refineries without catalytic
reformation benzene production
Calculated from
920 Ibs
1 000 bbl
carbons attributed to benzene.
reformation benzene production
For controlled refineries
For uncontrolled refineries
Weighted industry average
where 1.0% = estimated % of hydro-
Applies to refineries with catalytic
Releases to Water from Petroleum Refineries
1.64 x 10 10 kkg/bbl
,-12
(1 ug/D(43.36 gals water used per bbl) (3.785 1/gal) x (10 kkg/ug
From Effluent Guidelines Division screening sampling data, benzene
was delected in 1 of 6 refineries and the concuiiLration was 6 JH/J.,
therefore average concentration is 1 ug/1.
*Factor used in Figure 3.1
-------
EPA Effluent Guidelines Division screening sampling data for six refin-
eries and found that one refinery had a benzene concentration of 6 yg/1
in its effluent while the other five had none. The source took the
average concentration (1 yg/1), the average amount of water used per
barrel of crude refined, information on the number of refineries dis-
charging directly, and metric conversion factors and applied these to
the number of barrels refined to estimate the amount of benzene emitted.
Using this same method, we calculated the factor presented in Table 3.1.
The reliability of this factor is low due to the small amount of data
upon which it was based. Since it was the only emission factor avail-
able for benzene emissions to water from petroleum refining, it was
applied to the petroleum refining capacities listed in Mara and Lee in
order to have an :order of magnitude estimate of such emissions in 1978.
We assumed capacity production and direct discharge of effluent from
all refineries. The results of our calculations appear on the map in
Figure 3.1 as state totals. Based on the assumptions stated above,
total U.S. generated benzene emissions to water from petroleum refin-
eries was approximately 1 kkg in 1978.
Some possible sources of benzene emissions from petroleum refiner-
ies indicated by PEDCo (1977) are:
1. Process emissions from light and heavy naphtha streams
from the crude unit, fluid catalytic cracking units,
hydrocracking units, gasoline treating units, and pumps,
flanges, and other sources of fugitive emissions, and
2. Non-process emissions from wastewater treatment facilities,
heaters and boilers.
3.2 BENZENE EMISSIONS FROM COAL COKING OPERATIONS
Coke was produced by 65 plants in the United States which had an
annual coal capacity of 88,000,000 kkg (Mara and Lee, 1978). Of these
65, 10 (PEDCo, 1977) with an annual coal capacity of 25,600,000 kkg
(Mara and Lee, 1978) produced benzene as a byproduct from light oil.
According to A. D. Little (1977), half of the light oil produced was
processed on-site (at 10 coke plants) to extract benzene. The "other
half" was sold to petroleum refiners who extract benzene. Assuming
3-3
-------
U)
I
I HI
T5
(0.006)
Key:
- State
-f-jr - Emissions
(0.0) - Emissions to Water
Figure 3.1 Benzene: Refinery Emissions by State (kkg)
Sources: Mara and Lee, 1978; Versar, 1977
-------
that the 10 plants referred to by A. D. Little were the same 10 plants
identified by PEDCo, and assuming that these 10 plants used all their
light oil to produce benzene, it appears that the remaining 55 plants
with an annual coal capacity of 62,700,000 kkg (Mara and Lee, 1978)
produced the light oil sold to petroleum refiners. The yield of light
oil is 3-4 gallons per kkg of coal carbonized (PEDCo, 1977). Applying
this factor, the 10 benzene-producing coke plants had the capacity for
producing and using approximately 100,000,000 gallons of benzene-
containing light oil. The remaining 55 plants theoretically had the
capacity for producing about 200,000,000 gallons of light oil. This is
in sharp contrast to the 100,000,000 gallons (the "other half" of light
oil production) that A. D. Little claims was sold to petroleum refiner-
ies. Some of the difference may be light oil which was used on-site in
various processes (A. D. Little, 1977). It is unknown where the remain-
der of this difference went. If the 55 plants did not produce this un-
accounted-for light oil, then the materials which could have gone into
the light oil must have been used elsewhere or emitted to the environ-
ment. More data are needed to evaluate this potentially large source
of emissions.
Possible sources of benzene emissions from coke oven operators, as
indicated by PEDCo (1977), are:
1. Uncontrolled charging (i.e., placing the coal into the oven).
Evaporation and coking of volatile components in the coal
occurs upon the coal's contact with the hot oven floor.
This is perhaps the greatest potential source of hydrocarbon
(including benzene) emissions.
2. Topside emissions of fugitive benzene occurring from many
sources including leakage from weakened refractory materials,
ascension-pipe-elbow covers, leveling apertures, badly-fitted
charging-hole covers, and collecting main pipe valves.
3. Coke pushing (the discharge of hot coke from the oven). This
can be a major emission source unless the coal is completely
carbonized.
4. Door leaks, on both the push and coke sides of the oven.
5. Waste-gas stacks if coal or coke escapes through leaks into
areas between the oven and the heating flue.
3-5
-------
6. Quenching (the flooding of pushed hot coke with a huge quantity
of water to cool it rapidly) •. Much steam is produced, and
emissions would depend upon how completely the coal was car-
bonized.
The literature revealed three factors for benzene emissions to air
from coke oven operators. These are presented in Table 3.2.
The amount of benzene emitted to the air during coking operations
was estimated using Walker's (1976) emission factor based on the fol-
lowing :
1. Full-capacity production requiring consumption of 88,000,000 kkg
of coal (derived from Table C-l in Mara and Lee, 1978).
2. The yield of coke from coal is 68.4 percent. The benzene
emissions to air were estimated to be:
(88,000,000 kkg coal)(68.4%)(9.80 x 10~4 kkg benzene/kkg coke
produced)= 59,000 kkg benzene emitted.
Applying PEDCo's factor, the result would be:
(88,000,000 kkg coal)(7.8 x 10~ kkg benzene/kkg coal used)
=-6,900 kkg benzene emitted.
Applying Mara and Lee's factor in the same way, the result would
be:
3,000 kkg benzene emitted
The Mara and Lee factor was also applied to capacities state-by-
state to derive the distribution map of the 3,000 kkg benzene emissions
from coke ovens. This map is presented in Figure 3.2.
No emission factors for benzene to water from coke ovens were re-
vealed by the literature, therefore no emissions could be calculated.
The precision of emissions estimates based on Walker's emission
factor was estimated to be + a factor of 6, based on evaluation of his
assumptions. The uncertainty is probably due almost entirely to the
emission factor.
3-6
-------
Table 3.2 Benzene Emission Factors - Coal Coking Operations
Source
Walker
(1976)
PEDCo
(1977)
Mara and Lee
(1978)
Factor
o fin TT in~4 kkg benzene
J . OU X -LU . , ,
kkg coke
produced
7 R -s in"5 kkg benzene
/ . o X -LU . , ,
kkg coal
used
o _. in~5 kkg benzene
kkg coal
used
Remarks
C\ i -IT i n~ "^^8 hydrocarbon. ,_ ^09-1 f ^s f }-.or.-70T.0 -t,-. «-,-,*
kkg coal coked , , ,
hydrocarbor
L n rt^^.', ,0.776% benzene in coke over year .
34.8% hydrocarbons in coke over year
Estimated by multiplying the hydrocarbon emission factor
(4.2 Ibs/ton coal) by the fraction of benzene in the total
hydrocarbon emission (0.0132). Based on EPA Document AP-42.
The hydrocarbon emissions factor used here is different from
one used above. The factor for benzene content used here is
different from that used above.
.al
is)
the
U)
*Used here in deriving Figure 3.2.
-------
U)
I
CD
100
0.00 - Releases to Air
Unknown - Releases to water and lardfill.
Figure 3.2 Benzene: Coke Oven Emissions by State (kkg)
Source: Mara and Lee, 1978
-------
The uncertainty of the PEDCo emission factor was estimated by them
to be + a factor of 10. This would also be the uncertainty of the
emissions calculated from the emission factor.
The uncertainty of Mara and Lee's emission factor could not be
estimated. The factor had been obtained from EPA (1976).
3.3 INDIRECT PRODUCTION OF BENZENE FROM OIL SPILLS
Walker (1976) estimated benzene emissions to the oceans due to oil
q
spills of all types using the following assumptions: 1) 11-12 x 10 Ibs/
year were discharged to the oceans. This included both natural and man-
caused events. 2) Crude oil contains an average benzene concentration
of 0.2 percent by weight (the range is reportedly 0.001 - 0.4 percent).
Total benzene emissions to oceans were thus:
(11-12 x 109 lbs/year)(4.54 x 10~4 kkg/lb)(0.002 fraction benzene
in oil) = 10-11 x 10 kkg benzene released to oceans
The uncertainty of the estimate is very large, however, The basis
for the annual oil discharge value was not available, but the range of
benzene contents for oils introduces a 400-fold uncertainty range by
itself.
Versar (1977) estimated the gross annual discharge of benzene to
U.S. waters, based on U.S. Coast Guard information on crude oil spills
in 1976. Their estimate was based on the following: 1) The amount
of crude oil lost to U.S. waters through spills in 1976 was reported
to be 4,990,691 gallons. 2) Crude oil contains an average of 0.2 weight
percent benzene (range: 0.001-0.4 percent). Benzene emissions were
estimated to be:
(4,990,691 gallons)(7.211 Ibs/gal benzene)(4.54 x 10~4 kkg/lb) x
(0.002 fraction of benzene in oil) = 30 kkg benzene released to
U.S. waters due to oil spills in 1976.
The uncertainty of this estimate is at least + a factor of 400
(the.,range of benzene contents in crude oil).
3-9
-------
This source of benzene emissions should be the subject of more
detailed analysis if a Level II study is performed.
3.4 INDIRECT PRODUCTION OF BENZENE FROM VARIOUS SOURCES
Versar (1977) reported estimates of gross annual discharges of
benzene to water in 1976. The results of these estimates are listed
below:
Estimated Gross Annual Discharge
Process To Water, kkg
Nonferrous metals manufacturing 2.85
(Al, Cu)
Ore mining (Pb, Zn) 1.1
Wood processing 0.4
Coal mining 141.1
Textile industry (subcategories 2.51
40 and 60)
3.5 SUMMARY
Benzene emissions from petroleum refineries were estimated to be
20,000 kkg to air and 1 kkg to water, the amount to land unknown.
Benzene emissions from coke oven operations were estimated to be
3,000 to 59,500 kkg to air with the amounts to water and land unknown.
Benzene emissions from oil spills were estimated to be 30 to
Q
10-11 x 10 kkg to water (depending on the area being considered) with
the amounts to air and land unknown.
Benzene emissions to water from other sources were estimated to
be 2.85 kkg from non-ferrous metals manufacturing, 1.1 kkg from ore
mining, 0.4 kkg from wood processing, 141.1 from coal mining and 2.51 kkg
from the textile industry. Amounts emitted by these processes to air
and land were unknown.
3 -10
-------
4.0 IMPORTS OF BENZENE
4.1 AMOUNT IMPORTED
Table 4.1 summarizes the amounts of benzene imported during the past
five years:
TABLE 4.1 BENZENE IMPORTS, 1974-1979 (kkg)1
1979 1978 1977 1976 1975
232,0002 225,000 204,000 175,000 234,OOo'
1. Source: Bureau of the Census, Department of Commerce
2. Based on extrapolation of January through October data for 1979.
4.2 EMISSIONS DUE TO IMPORTS
Emissions due to imports were interpreted as emissions due to unloading
plus transport to the point of consumption. PEDCo (1977) estimated emission
factors for these processes:
Process Emission Factor
Marine loading 2.0 x 10~4 T-: f—-3—r (uncontrolled)
6 kkg unloaded
kkg transported/week
In applying these emission factors due to 1978 imports, the following
additional bases were used: (1) Emissions due to loading were 95 percent
controlled at dockside. (2) The average time for imported benzene to be
in transit was one week.
Emissions were then calculated to be:
[(225,000 kkg) (2.0 x 10~4 kkg/kkg) (0.05)] +[(225,000 kkg)(l week)
(1 x 10~^ kkg/kkg/week)J = 25 kkg benzene released per year due to
importation
We estimate the uncertainty of this emission to be +_ a factor of 10, based
on the difficulty of estimating emission factors for processes of this type.
4-1
-------
4.2.1 Emissions to Air
Of the total emissions, it was judged that 50 percent would go to
air and 50 percent to water. This estimate was based on the following:
(1) Most "non-accident" emissions would be to air because of benzene's
volatility. (2) Most emissions due to accidents would ultimately be to
water, due to the fact that transfer operations are dockside and water
would be used to hose down the area.
Therefore, air emissions = (25 kkg)(0.5) = 13 kkg benzene released
to air due to importation,
4.2.2 Emissions to Water
By the reasoning presented in 4.2.1, the emissions to water due to
importation of benzene would also be 13 kkg.
There would be no emissions of benzene to landfills or due to disposal
of solid residues during importation.
4-2
-------
5.0 CONSUMPTIVE USES AND EXPORTS OF BENZENE
The utilization of benzene can be broadly divided into consumptive
uses and nonconsumptive uses. This chapter reports on emissions due to
consumptive uses: processes in which benzene is chemically converted
to another compound. Nonconsumptive uses, in which benzene is used as
an end-product rather than as an intermediate, are discussed in Chapter
6.0. For the purposes of a materials balance, exports are considered
consumptive uses because the chemical is permanently removed from the
U.S. "inventory" by this process. Therefore, exports are also discussed
in this chapter.
5.1 CONSUMPTIVE USES—TOTALS
5,389,000 kkg were estimated to have been consumptively used in
1978. This total was derived in Table 5.1. The main use of benzene
was as a chemical intermediate in the synthesis of other organics.
5.2 CATEGORIES OF USE
The major compounds derived from benzene in 1978 are listed in
Table 5.1. In addition, minor uses are included in the benzene
Environmental Flow Diagram, Appendix A. Ethylbenzene synthesis was by
far the largest consumer of benzene, as it has been historically (See
Table 5.2 at end of chapter for data on the years 1975-1979). The
data on these two tables were based on USITC reports. Their uncer-
tainty is not known, but it is estimated to be small (less than +20%).
5.3 EMISSIONS BY CATEGORY OF USE
5.3.1 Consumption of Benzene by Ethylbenzene Synthesis
5.3.1.1 Processes, Producers, and Locations
Approximately 90-95% of ethylbenzene production is carried out by
alkylation of benzene with ethylene (Versar, 1979; Weissermel and Arp,
1978; A.D. Little, Inc., 1977):
5-1
-------
TABLE 5.1 CONSUMPTIVE USES OF BENZENE (1978)
1 3
Use kkg Produced kkg Benzene Required
Ethylbenzene 3,803,000 2,890,000
Cumene 1,533,000 1,058,000
Cyclohexane 1,057,000 811,000
Nitrobenzene2 261,000 170,000
Maleic anhydride 155,000 161,OOO4
Chlorobenzenes
(mono- plus di-) 171,000 133,000
Detergent Alkylate s
(linear plus branched) 330,000 132,000
Anthraquinone 17,000 23,000
Biphenyl 29,000 11,000
5,389,000
1. USITC figures except as noted.
2. Includes nitrobenzene destined for aniline synthesis (96%), plus
non-aniline usage (4%).
3. Conversion factors from Neufeld £t_ ail^., 1978.
4. This assumes that 81% of maleic anhydride was made from benzene
(PEDCo, 1978).
5. Derived from 1978 USITC production figure for straight-chain
alkylbenzenes (239,000 kkg) by assuming a ratio linear/branched =2.64
(Neufeld et_ al., 1978) .
6. Total Vat Dye production (USITC, 1978, Neufeld et al., 1978).
5-2
-------
CH CH
Ethylbenzene
b.p. 136°C
The process is carried out in either liquid- or vapor-phase, using an
excess of benzene to minimize diethylbenzene byproducts. It is not
known whether the liquid- or vapor-phase process predominates. The
isolation of ethylbenzene as a byproduct of xylenes and cumene produc-
tion accounts for the rest of ethylbenzene production. Almost all
ethylbenzene is used for the subsequent production of styrene monomer
(Versar, 1979). Figure 5.1 is a process flow diagram for ethylbenzene
synthesis.
Figure 5.2 lists ethylbenzene producers, production capacities,
and plant locations. Of the 3,547,000 kkg of 1976 benzene-consuming
capacity listed, at least 75% is located along the Gulf Coast
(Channelview, TX, and Welcome, LA, could not be located on the maps
used). The plant capacity data were from SRI (1977). * It was diffi-
cult to assess the reliability of these,data. Comparison of 1976
benzene consumption due to ethylbenzene synthesis (Table 5.2) with the
amount of benzene required for full-capacity ethylbenzene synthesis
(Figure 5.1) showed that ethylbenzene synthesis was at 67% of capacity
in 1976.
5.3.1.2 Amount of Benzene Consumed
2,890,000 kkg of benzene were used for the synthesis of ethyl-
benzene in 1978 (Table 5.1). This value was obtained by the operation:
(kkg ethylbenzene produced) (conversion factor). The conversion factor
used was 0.76. This factor is equal to the ratio of molecular weights
(benzene/ethylbenzene) 7 the percent theoretical yield. The latter
was estimated to be 97% by A.D. Little, Inc. (1977). The uncertainty
in the estimate of the amount of benzene consumed during ethylbenzene
synthesis was +20%.
5-3
-------
Off-gat Benze
i scrubbing lo
| system
Rea
Ethylene | .
't
Benzene. .
recycle
and fresh
f— f
1 t
> (j 0 conden
if- ' s~^
^7^ Settler
ctor T~^
Jt ^_
4 J 1
Aluminum ^
f
aer B
c
r^ n
! t
1(1 )
chloride Water wash Caustic wash
complex and settler and settler
ie recycle
reactor
T Ethylbenzene
i A r~!
snzene I ' 1 (polyethyll-
olumn Ethyl- benzene
benzene column
column 1
JjVjlJ
Heavy
(polyethyl)benzenes
and tar
Recycle (polyethyl)benzenes
Figure 5.1 The Manufacture of Ethylbenzene Employing Aluminum Chloride as
Catalyst
5-4
-------
5.3.1.3 Benzene Emissions Due to Ethylbenzene Synthesis
5.3.1.3.1 Generated Emissions to Air
Previous studies have proposed emission factors for release of
benzene to the air due to consumption processes. One approach was to
set an upper limit on emissions by relegating all unaccounted-for ben-
zene to this category (Howard et_''§_!., 1974). An average yield of 97%
of theoretical has been estimated for benzene alkylation (A. D. Little,
1977). Applying this information to 1978 ethylbenzene synthesis would
give:
(3,803,000 kkg ethylbenzene producedX0.03 fraction of ethylben-
zene lost)(78 g benzene used) _ 84,000 kkg benzene
(106 g ethylbenzene used) unaccounted for
A second approach was to estimate an overall emission factor for
total benzene consumption. Patterson et^ al_. (1976) used the value
0.005 kkg/kkg benzene consumed, although the origin of the value is
unclear. Application of this emission factor to 1978 consumption for
ethylbenzene production gives:
(2,890,000 kkg benzene consumed)(0.005 kkg/kkg) = 14,000 kkg benzene
emissions generated for release to air during ethylbenzene synthesis.
A third emission factor was obtained from Versar (1979), based on
PEDCo Environmental (1977). In Versar (1979), airborn emissions were
estimated to be 1200 kkg for production of 1,790,000 kkg ethylbenzene
in 1976. The emission factor calculated from these values was:
1 790? QQO kk = ^"^ x ^" kkg/kkg produced. This was equivalent
to (6.7 x 10~ kkg/kkg ethylbenzene produced)(106 kkg ethylbenzene produced
_A 78 kkg benzene used
= 9.1 x 10 kkg/kkg benzene used.
Application of this emission factor to 1978 benzene consumption for
ethylbenzene synthesis gave:
(2,890,000 kkg benzene consumed)(9,1 x 10~4 kkg/kkg benzene
consumed) = 2600 kkg benzene emissions generated for release to air.
These emissions are summarized in Table 5.3.
Of the three values, the 84,000 kkg estimate is an absolute maxi-
mum and is undoubtedly too high. By assuming either recycling or de-
struction of a large proportion of the unaccounted-for benzene, lower
emissions are estimated. Pending evaluation of further information,
emissions of benzene to air during ethylbenzene synthesis will be
5-5
-------
Locations (1976)
COMPANIES AND OPACITIES (1976)
Hap
No.
1.
2.
3.
4.
3.
6.
7.
8.
9.
10.
11.
12.
13.
14.
13.
16.
1.
2.
Conpany
American Petroflna
ARCO/Polreen. Inc.
Coe-Mar, Inc.
Dov Ounlcel
Dow Chalcal
U Peeo Nat. Caa
Poeter Grant
Culf Oil Che..
Joe Oil Co.
Monaanto Co.
Monaanto Co.
Ox Irene Qm
Phillip. Petrolevm
Standard Oil (Ind)
Sun Oil Co.
Union Carbide
Arthur D. Little, Inc., 1977
Conversion factor uaed wee 0
la the ratio of aoleculer we
the reference cited above).
Plant Location
Dig Spring, TX
Port Arthur, TX
Carvllle, U
Preeport. TX
Midland, m
Odesea, TX
Baton Rouge, LA
Uelcos*. L»
Houston, TX
Alvia, T>
Texas City, Tx
Chennelview, TX
Phllllpe. TX
Texas City, TX
Corpus Christ!, TX
Seadrlft, TX
SUM
Ethylbeniene
Capacity, kkn/yr
20,000
200,000
689,000
846,000
249,000
125,000
440.000
249,000
HA
23,000
658,000
544,000
DA
429,000
43,000
154,000
4,669,000
.76 kkg beniene/kkg ethylbeniene (Arthur D. Little, Inc
ighta (benaene/ethylbenicne) T the Z theoretical yield
Pull Capacity
Requirenent .
15,000
151,000
524.000
643.000
189,000
95,000
334,000
189,000
IU
17,000
500,000
413.000
HA
326,000
33,000
117,000
3,547,000
Benzen
kkn/yr
. , 1977). This value
(aesuved to be 97Z in
Figure 5.2 Production of Ethylbenzene from Benzene
5-6
-------
Table 5.3 Summary of Estimated Benzene Emissions to Air Due to Ethylbenzene Synthesis
Emission Factor Emissions to
Method Used (kkg/kkg benzene consumed) Air, 1978, kkg Reference
Sum of Unaccounted-for
Benzene ~ 84,000 Howard et^ al., 1974
Estimate Emission
Factor (basis not ,
stated) 5 x 10 14,000 Patterson e± al., 1976
Estimate Emission
Factor (from emissions _,
and consumption data) 9.1 x 10 2,600 Versar, 1979
-------
expressed as the range 2,600 - 14,000 kkg.
5.3.1.3.2 Generated Emissions to Water
The amount of benzene generated for release to water was estimated
to be 345 kkg in 1976 (Versar, 1977). Ethylbenzene production that year
was 1,790,000 kkg; the emission factor was therefore:
345 1' 1,790,000 = 1.9 x 10"4 kkg benzene released/kkg ethylbenzene
produced.
Application of this emission factor to 1978 ethylbenzene production yields:
(3,803,000 kkg ethylbenzene produced)(1.9 x 10"4 kkg/kkg) = 720 kkg
benzene released to water due to ethylbenzene production.
This value has an estimated uncertainty of + a factor of 10, since it was
based on estimates of both the emission factor and the fraction of gener-
ated emissions entering water. The main source of these water emissions
was probably scrubber effluents.
5.3.1.3.3 Emissions Due to Disposal of Solid Residues
It was not possible to estimate emissions of benzene in solid
residues. In order to estimate these emissions, it would be necessary
to know: (1) the rate of production of benzene-containing sludges, (2) the
weight % benzene in the sludge, and (3) the method of disposal of the
sludge (landfill, incineration).
5.3.1.3.4 Carry-Over of Benzene during Ethylbenzene Synthesis
Data were not available to permit estimation of benzene contamina-
tion in ethylbenzene. This would appear to be an important topic for
a Level II study.
5.3.2 Consumption of Benzene by Cumene Synthesis
5.3.2.1 Processes, Producers, Locations
All chemical-grade cumene produced in the U.S. at present is made
by the alkylation of benzene with propylene (Peterson, 1979). The basic
reaction in this process is:
5-8
-------
+ CH2=CH-CH2
CH3
CH-
CH3
cumene
Most cumene is produced by the Solid Phosphoric Acid Process, as
shown in Figure 5.3. Benzene and propylene are reacted at elevated
temperatures and pressures in the presence of an acidic catalyst (Peter-
son, 1979).
Figure 5.4 shows the producers and locations for 1978 production.
5.3.2.2 Amounts Manufactured
According to USITC (1979), 1,533,000 kkg of cumene were produced
in 1978. The amount of benzene required to synthesize this amount of
cumene was estimated by using a conversion factor. The factor for
calculating benzene consumption was 0.69 kkg benzene required/kkg cumene
produced (SRI, 1977). This value is equal to the ratio (molecular
weight benzene/molecular weight cumene) -f- theoretical yield. A 94%
yield was estimated by SRI for cumene synthesis. Applying this conver-
sion factor to the ITC production data for cumene production yielded:
(1,533,000 kkg)(0.69) = 1,058,000 kkg benzene required for cumene
production in 1978.
5.3.2.3 Emissions of Benzene Due to Production of Cumene
Data on benzene emissions during cumene synthesis for 1978 were
not found in any available sources. Based on 1978 USITC production data
and an emission factor from PEDCo (1977) on benzene emissions from the
production of cumene in 1976, emissions were calculated to be:
(1,533,000 kkg) x (0.25 x 10~3 kkg/kkg) = 380 kkg benzene released
during 1978.
_o
This 0.25 x 10 kkg/kkg emission factor was for fugitive emissions.
There was no information to permit estimation of the accuracy of this
factor.
5-9
-------
- Water injection
Rerun column
Con
feed drum
Fresh propylene^
propane "-
Propane
Cumene
bottoms
Figure 5.3 Solid Phosphoric Acid Process for Cumene Pro-
duction
Source: A.D. Little, Inc., 1977
5-10:
-------
(Puerto Rico)
1.
2.
}.
4.
5.
6.
7.
a.
9.
10.
11.
12.
1).
14.
15.
16.
taoco OU Co.. Taxaa Clt», TX
Aahland Oil Co., CatleCteburf . R
ChenTon Oil Co., El Segimdo, CA
Cl.ck OU Co., Blue talaod. It
Coaetal Statee Cae Co.. Corpua
Do» dialed. Midland, HI
Georgia Pacific Corp., Houeton
catt? ou Co.. EI Doredo. TX
Gulf OU. PhUadelohla, PA
Gulf OU. Fort Arthur, n
Karethon Oil Co.. Tesel City, TX
Honeanto Cheaical Co., Chocolat
Sh.ll Oil Co.. Dur Park. TX
Sun Patrolaua Proeucta Co., Co;
Tauco. Inc., Uaaulll, IU
Union Carblda Core., Ponca, PR
PRODUCERS AND BENZENE REQUIREMENTS
Cuaene Production kks
9,000
R 118,000
CA 27. 000
35,000
I Chriatl, TX 44.000
9.000
,. TX . 221.000
40,000
13), 000
130,000
TX 56.000
it. Baron, TX 206,000
206.000
irpu. Chriatl. TX 68,000
•»,000
[ 189.000
Benzene Uaed kkg
6,000
61,000
19.000
24,000
30,000
2.000
152,000
28,000
92,000
90,000
39.000
142.000
142,000
47,000
30,000
130:000
1.329.000
1.034.000
Soure*: Nusbcr* were ext»poUt«d from ITC 1978 Production data by ucing 1978 Cua«n« O«p«cltr figure* fro*
P««T»on 1979 for til plant • except 1976 Cunene Capacity figure* for Dow Oitaical and Marathon Oil (fc.D. Llttla.
Inc.. 1977). The*e numbera were aultiplied by 0.6$ which waa derived by dividing 1978 ITC production by total
ci»ene capacity (Peteraon. 1979):
1.333. OOP V
2,352,000
The cumcne production waa directly proportional to capacity of each plant and waa not actual production.
Bentene uae waa calculated by applying the converaion factor for beniene coniunption in the production of
cuaenei 0.69 kkg benzene per kltg of cuaene produced. Thia converaion factor waa taken fron SRI (1977)
and la alao uaed in other aourcea.
i rounding.
Figure 5.4 Cumene Production: Producers and Locations
5-11
-------
5.3.2.3.1 Emissions to Air
There were no data on exact amounts of the emissions of benzene into
the air from the production of cumene. It was assumed that almost all
emissions are to the air. This assumption was based on the volatility
of benzene, and the fact that the reaction is carried out at elevated
temperature or in the vapor phase (A.D. Little, Inc., 1977). More infor-
mation is needed to properly determine the quantities of benzene emitted
to the air from cumene production.
5.3.2.3.2 Emissions to Water
While no data on the exact amounts of emissions of benzene to
water from the production of cumene are available, it can be assumed
that only a small percentage of the total emissions are to water (see
above). More information is needed to properly determine the quantities
of benzene emitted to the water from cumene production.
5.3.2.3.3 Emissions Due to Disposal of Solid Residues
No information on the disposal of solid residues was provided.
In order to estimate these emissions, it would be necessary to know;
(1) the rate of formation of benzene-containing residues, (2) the
weight % benzene in the residues, and (3) the method of residue disposal
(incineration, landfill).
5.3.2.3.4 Carry-Over of Benzene into Product
Carry-over of benzene as a contaminant of cumene may be a signifi-
cant source of emissions, and should be addressed in a Level II study.
5.3.3 Consumption of Benzene by Cyclohexane Synthesis
5.3.3.1 Processes, Producers, Locations
Three processes are used in the synthesis of cyclohexane:
1) Hydrogenation of benzene,
2) Isolation from crude gasoline by fractional distillation, and
3) Isolation from crude gasoline by fractional distillation with
5-12
-------
simultaneous isomerization (Weissermel and Arpe, 1979).
Of these methods, hydrogenation of benzene accounted for 80-85% of
cyclohexane production, while the other two accounted for the remaining
15-20%.
Hydrogenation of benzene produces cyclohexane of high purity, over
99%, as compared to 85% for the fractionating process and 98% for the
fractionation combined with isomerization (A.D. Little, Inc., 1977).
The reaction equation for the catalytic hydrogenation of benzene
is shown below.
Catalyst
+ 3H,
2
Benzene Cyclohexane
b.p. 80°C
The process is carried out at elevated temperature in either
liquid- or vapor-phase, at a hydrogen pressure of 20-40 atm (Weissermel
and Arpe, 1978). Figure 5.5 shows the industrial process flow chart
for liquid-phase benzene hydrogenation.
Several reactors are used to progressively improve the conversion
to cyclohexane until residual amounts of benzene and methylcyclopentane
are reduced to less than 100 ppm.
Figure 5.6 shows the location of U.S. cyclohexane producers.
5.3.3.2 Amounts Produced
1978 production data are also shown in Figure 5.6. The benzene
requirement was calculated by using the conversion factor 0.93 kkg
benzene required per kkg cyclohexane produced (Neufeld et al., 1978).
Cyclohexane production by each plant in 1978 was estimated by assuming
that the quantity produced was proportional to the fraction of the total
U.S. production capacity that it contributed in 1976:
1978 cyclohexane production by a plant = ( Total U, S. production
of cyclohexane in 1978) (% of total production capacity contributed
by.plant in 1976)
Cyclohexane production figures were taken from USITC.
5-13
-------
Cyclohexane
Figure B-ll Hydrogenation of Benzene
5-14
-------
(5) (Puerto Rico)
(j) (Puerto Rico)
PRODUCERS^ AND BENZENE REQUIREMENTS
Map
No.
1
2
3
5
6
7
I
9
10
CoananT
Aaer. Patrofina
Covonvaalth Oil
Exxon Choa.
Golf Oil Che,.
Phillip. Che,.
Phillip. Patrolaia
Sim Oil Co.
Taxaco, Inc.
Onion Oil Calif.
Union Pacific
Location
Bl| Springa. TX.
Panualaa, PR
Baytovn, TX
Port Arthur. TX
Swaney, TX
Cuayaaa, Pi.
Tulaa, OX
Port Arthur. TX
luuaont, TX
Corpua Chrlatl. TX
Total
kkj Cyclohexana Produead2
31.000
101,000
101,000
84.000
216.000
226.000
51,000
101.000
87.000
56.000
1.056.000
kkfc Bcntena Uaed
29.000
94.000
94,000
78,000
106,000*
212,000
47,000
94.000
81,000
52.000
887, OOO
1 Score.: SRI, 1976.
2Deriv*d from Individual plant ctp.cltlt. (A.D. Little, Inc.. 1977) u»lng the eesunptlon that a plant'*
percent contribution to total production In 1978 equal* the plant1! percent contribution to total capacity
In 1976. Total 1978 production vai froa USXTC.
Eatlaated ualn| the convert loo factor 0.93 kkg benzene uaed/kkg cyclohexane produced (Neufeld ct^. al.. 1978).
Bentcne feeditock wa« estimated to account for 47-391 of plant production (A.D. Little. Inc., 1977). The
average value (SIX) uaa uaed ae the percent of production derived froa benzene.
Figure 5.6 Production of Cyclohexane from Benzene, 1978
5-15
-------
5.3.3.3 Benzene Emissions Due to Cyclohexane Synthesis
Benzene emissions would be expected to be negligible during cyclo-
hexane production from simple fractionation of natural gasoline (A.D.
Little, Inc., 1977). This is because benzene is present only as a minor
constituent of the crude gasoline and its concentration would only be
measured at trace levels. The same analysis applies to the fractionation/
isomerization method of cyclohexane synthesis.
An emission factor of 0.0028 kkg of benzene emitted per kkg of
cyclohexane produced was cited by Mara and Lee (1978). They regarded
it as an order of magnitude estimate. In the absence of specific infor-
mation on the applicability of this emission factor, it was assumed
that it describes an average of the three manufacturing processes with
no attempt made to distinguish among these.
To estimate atmospheric benzene emissions, total U.S. cyclohexane
production for 1978 (Figure 5.6) was multiplied by 0.0028 kkg/kkg:
(1,056,000 kkg cyclohexane produced)(0.0028 kkg/kkg product) =
3,000 kkg benzene emissions due to cyclohexane synthesis.
5.3.3.3.1 Emissions to Air
Figure 5.5 indicated that the only vent in the otherwise closed
reactor sequence was the gas vent for hydrogen. Any other losses would
be due to leaks, spills, and catalyst regeneration. We estimated that
99% of emissions were to the air. Application of this percentage to
total emissions estimated in 5.3.3.3 yielded:
(3,000 kkg benzene emitted) x (0.99 fraction to air) = 2970 kkg
benzene released to air during cyclohexane synthesis.
The uncertainty of this value was estimated to be the same as that of
the emission factor: + a factor of 10.
5.3.3.3.2 Emissions to Water
As discussed above, emissions to water would be:
(3,000 kkg benzene emitted) x (0.01 fraction to air) = 30 kkg benzene
released to water during cyclohexane synthesis.
5.3.3.3.3 Benzene Emissions to Land
5-16
-------
Neither qualitative nor quantitative data concerning benzene emission
to land as a result of cyclohexane manufacture were available.
5.3.3.3.4 Carry-Over of Benzene into Product
Hydrogenation of benzene produces a cyclohexane with over 99%
purity. Weissermel and Arpe (1978) indicated that where a finishing
reactor was used in conjunction with a series of reactors, hydrogenation
of benzene could be complete enough so as to leave the cyclohexane end
product with less than 100 ppm (0.01% by mass) of benzene and methylcy-
clopentane contamination. Also, A.D. Little, Inc. (1977) estimated that
from 0-0.5% benzene by volume may be detected in cyclohexane. The amount
of this carry-over that is ultimately released could not be predicted
due to the lack of information on cyclohexane uses.
While cyclohexane production from fractionation of crude gasoline
and simultaneous isomerization has a 98% yield, neither the quantity
of benzene residue nor its rate of release could be ascertained.
According to A.D. Little, Inc. (1977) (who cited R.M. Stephenson,
Introduction to the Chemical Process Industries), the 15% impurities in
cyclohexane extracted from gasoline (Section 5.3.3.1) contained benzene,
2,2-dimethylpentane, and 2,4-dimethylpentane. However, A.D. Little,
Inc. (1977) also reported that the plant manager at Borger, Texas stated
to them in a personal communication that no benzene existed in the 85%
cyclohexane end-product. No information was available to resolve this
conflict nor has the possible release of these contaminants been studied.
5.3.4 Consumption of Benzene by Maleic Anhydride Synthesis
5.3.4.1 Producers, Processes, Locations
5-17
-------
Maleic anhydride is produced by the catalytic oxidation of benzene
according to the equation:
V2°5
Figure 5.7 shows the industrial flow chart for this process.
AIR.
1
1
MIXER
[-«•
1
L
FUSED
SALT
COOLER
—\
1
*-J VAPOR
* COOLER
TO ABSORBERS
il
?S
1
KASTE
MALEIC
ANHYDRIDE
Figure 5.7 Flow Chart for Maleic Anhydride Synthesis
Figure 5.8 shows the locations and benzene requirements of maleic
anhydride producers.
5.3.4.2 Amounts Produced
USITC reported a production of 155,000 kkg maleic anhydride in
1978. The corresponding benzene requirement was estimated based on
the following information: 1) Theoretically, 78 g benzene yields 110 g
maleic anhydride. 2) In practice, a 55% yield is achieved (A.D. Little,
Inc., 1977). 3) 81% of maleic anhydride is currently derived from ben-
zene (PEDCo, 1978). The factor for conversion of maleic anhydride
production to benzene consumption was
78 g benzene
-i
110 g maleic anhydride
1.55 yield I
(0.81 fraction from benzene)
= 1.04 kkg benzene used/kkg maleic anhydride produced.
5-18
-------
This conversion factor was used in Figure 5.8. The total estimated
benzene consumption for maleic anhydride synthesis in 1978 was 161,000 kkg.
5.3.4.3 Emissions Due to Maleic Anhydride Synthesis
5.3.4.3.1 Emissions to Air
The main source of benzene emissions during maleic anhydride syn-
thesis was reported to be the condenser vent (labeled "to absorbers"
in Figure 5.7) (PEDCo, 1977). PEDCo (1977) cites the results of a 1972
emissions inventory showing emission factors for controlled plus uncon-
trolled emissions from this vent to be 0.06 - 0.20 kkg/kkg maleic anhy-
dride produced. Also cited was a report by Monsanto Research containing
an emission factor of 0.0967 kkg/kkg product for controlled emissions
from a maleic anhydride plant. In order to present a worst-case estimate,
the emission factor 0.20 kkg/kkg product was employed to estimate 1978
benzene emissions. Conversion of this value to kkg benzene released/kkg
benzene used yields:
(0.20 kkg/kkg product) x ( en'zene^sed } = °'21 kkg/kkf
enzene used benzene used.
Application of this emission factor to 1978 benzene consumption (Figure
5.8) yields:
(161,000 kkg) (€.21 kkg/kkg) = 34,000 kkg benzene released due
to maleic anhydride synthesis in 1978.
Based on the 3.7 - fold range of the literature emission factors
and the judgment that emissions have probably decreased since 1972 due
to added controls, the uncertainty of the 1978 emissions to air was esti-
mated to be + 10%, - 90%.
5.3.4.3.2 Emissions to Water
The synthesis of maleic anhydride is of necessity an anhydrous process,
since the anhydride is destroyed by hydrolysis. Thus, any loss of
benzene to water would be indirect (scrubber effluents, leaks). Versar
(1979) cited Versar (1977) as the source of an estimated aqueous emission
of 8 kkg in 1976 due to maleic anhydride synthesis. Consulting the latter
5-19
-------
source, however, failed to reveal any mention of benzene emissions due
to maleic anhydride synthesis. Thus, although the low estimated aqueous
emission was qualitatively.reasonable, its origin was unclear. No un-
certainty range was assigned to this value.
5.3.4.3.3 Emissions Due to Disposal of Solid Residue
No data were available on benzene emissions due to disposal of
solid residues. The vacuum column waste stream would be expected to
contain no benzene, because benzene has a lower boiling point than
maleic anhydride and would exit the column with the product. Based on
this brief analysis, it was estimated that emissions of benzene due to
disposal of solid residues from maleic anhydride synthesis were negligible.
5.3.4.3.4 Carry-Over of Benzene into Product
Information was not readily available on benzene contamination of
maleic anhydride product. It would be necessary to obtain analyses of
maleic anhydride batches in order to evaluate this possibility.
5.3.5 Consumption of Benzene by Nitrobenzene Synthesis
5.3.5.1 Processes, Producers, and Locations
Synthesis of nitrobenzene is carried out by the nitration of benzene:
HNO
3 \ J] + H00
The process is carried out in liquid phase. Almost all nitrobenzene
is used in the subsequent production of aniline (Versar, 1979). Figure
5.9 is a process flow diagram for nitrobenzene synthesis.
Figure 5.10 lists the nitrobenzene producers, production capacities,
and plant locations. The data are for 1976. Comparison of the full
capacity benzene requirement (356,000 kkg; Figure 5.10) with the 1976
consumption of benzene for nitrobenzene synthesis (169,000 kkg; Table 5.2)
indicates that nitrobenzene plants were operating at 48% of capacity
that year.
5-20
-------
COMPANIES AND BENZENE CONSUMPTION
p No. Company
1. Allied Chemical
2. Koppera Co.
3. Monsanto Co.
4. Petro-Tex Chem.
5. Reichold Chem.
6. Reichold Chem.
7. Tenneco, Inc.
8. U. S. Steel
9. Ashland Oil
1. Source: A. D. Little, Inc., 1977.
SUM
1978 Benzene Consumption, kkg
20,000
11,300
27,500
16,700
10,000
20,000
8,700
26,700
20,000
161,000
2. The conversion factor used was 1.04 kkg benzene used/kkg maleic anhydride produced (see section 5.3.4.2).
Total benzene consumption for 1978 was allocated to individual plants in proportion to their 1976
capacities (A. D. Little, Inc., 1977).
Figure 5.8 Maleic Anhydride Producers and Locations
5-21
-------
AMBIENT
EMISSIONS
MIXEDt
ACID
1
BENZENE
o
tir
\
£
1
CD
QC
O
s
1
1
* —
•4
NITRATOR
-H20
-AIR
^TO ANILINE
CRUDE
NITROBENZENE
»,
ae
o
i
Q
ll
I/
SPEND
k.
J
1
PRODUCTION
WATER Oil UTE
nn i 1 1\ 9 L/ i uu 1 1
SODIUM CARBONATE
i
.fc WASHER
1
WASH-WATER
WASTE
ACIO
3
C
i
>
j
)
U
1
I
NITROBENZENE
(REFINED)
I
WASTE
TO RECOVERY
Figure 5.9 Process Flow for Nitrobenzene Synthesis
-------
LOCATIONS (1976)
COKPAXUS ACT) CAPACITIES (1976)
Nap
Ito.
1.
2.
3.
4.
3.
6.
7.
a.
Co»paoy
Alllad Conical
Anariean Cyonaald
E. I. duPont
I. I. ouPont
ririt Mlaaiaalppl Corp.
Itorbay Oienlcal
Honaanto CD.
Rubicon Chenicala
Plant Location
Mouadnllla, WV
Bound Brook, NJ
iMtomc. IX
ClbbKmn. RJ
Puusoula, MS
•» NirtlnnUl*. VV
S«|
-------
5.3.5.2 Amount of Benzene Consumed
170,000 kkg of benzene.were used for the synthesis of nitrobenzene
in 1978. This value was obtained by the operation:
(kkg nitrobenzene produced)(conversion factor)
The conversion factor used was 0.65 kkg benzene used/kkg nitrobenzene
produced. This factor was equal to the ratio of molecular weights
(benzene/nitrobenzene) 7 the % theoretical yield. The latter was estimated
to be 97% by Neufeld et^ a!L., 1978. The uncertainty in the estimate
of the amount of benzene consumed during nitrobenzene synthesis was + 20%.
5.3.5.3 Benzene Emissions Due to Nitrobenzene Synthesis
5.3.5.3.1 Emissions to Air
Previous studies have estimated values for release of benzene due
to nitrobenzene synthesis. The results are summarized in Table 5.4.
_3
Patterson et al. (1976) apply an emission factor of 5 x 10 kkg/kkg
(0.5%) to benzene-consuming processes. This value, in turn, is based
on "available data in (Compilation of Air Pollutant Emission Factors,
U.S. EPA Report AP-42, April, 1973) and emission data for other similar
processes." Applying this emission factor to 1978 benzene use for nitro-
benzene synthesis yields:
(170,000 kkg) (5 x 10~3 kkg/kkg) = 900 kkg benzene released to
air during nitrobenzene synthesis .
Table 5.4 Summary of Estimated Benzene Emissions to Air Due to
Nitrobenzene Synthesis
1978 Emissions
Method Used Emission Factor to Air, kkg Reference
General Emission 5 x 10 kkg/kkg 900 Patterson, et_
Factor for Ben- " benzene used al., 1976
zene Consumption
. _3
Measured Emissions 8.3 x 10 kkg/kkg 2200 Process Research,
at Reactor Vent product Inc., 1972
_2
Estimate Emission 1.1 x 10 kkg/kkg 1900 Versar, 1979
Factor from Emis- benzene used
. sions and Consump-
tion Data
5-24
-------
Because of the lack of background information, it was not possible to
evaluate the uncertainty of this emission.
_3
A second emission estimate was based on an emission factor of 8.3 x 10
kkg benzene released to air/kkg of nitrobenzene produced (Process Re-
search, Inc., 1972). This emission factor was measured at the reactor
absorber vent of a benzene nitration plant, and is characterized by the
authors as being based on little or no census or experimental data.
No uncertainty range was associated with the number. Application of the
emission factor to 1978 nitrobenzene production yields:
(261,000 kkg) (8.3 x 10~3 kkg/kkg product) = 2200 kkg benzene
released to air during nitrobenzene synthesis.
A third estimate was that of Versar, 1979. Based on PEDCo Environ-
mental (1977), they calculated airborne emissions of 3390 kkg for
consumption of 302,000 kkg benzene. This yields an emission factor of
(3390 kkg/302,000 kkg = 1.1 x 10~2 kkg/kkg).
Application of this emission factor to 1978 benzene consumption yields:
(170,000 kkg) (1.1 x 10~2 kkg/kkg used) = 1900 kkg benzene
released to air during nitrobenzene synthesis.
No uncertainty range could be assigned to this value.
The three values tabulated varied over a range of a factor of 2.6.
Since no value was clearly superior, the highest emission (2200 kkg)
was selected in order to present a "worst case" estimate.
5.3.5.3.2 Generated Emissions to Water
No information was readily available on benzene emissions to water
during nitrobenzene synthesis. Because of benzene's volatility, this
category would be expected to have less release than air emissions. Also,
the nitration reaction is carried out at temperatures between 55° and 90°C
(Process Research, Inc., 1972).
5.3.5.3.3 Emissions Due to Disposal of Solid Residues
No data were available that permitted estimation of benzene emis-
sions due to disposal of benzene-containing solid residues. In order
to estimate these emissions, it would be necessary to know: 1) the rate
5-25
-------
of production of benzene-containing sludges; 2) the weight % benzene
in the sludge; and 3) the method of disposal of the sludge (landfill,
incineration).
5.3.5.3.4 Carry-Over of Benzene into Nitrobenzene
Data were not available to permit estimation of benzene contamination
in nitrobenzene. Analytical data on nitrobenzene batches would be
necessary. A Level II study could pursue this area more completely.
5.3.6 Consumption of Benzene by Chlorobenzene Synthesis
5.3.6.1 Processes, Production, Locations
The 1978 chlorobenzenes production accounted for 0.3% of the total '
U.S. benzene consumption (USITC; see also Appendix A). The three main
commercial products obtained through the chlorination of benzene are
monochlorobenzene,or_tho_-dichlorobenzene, and para-dichlorobenzene. Ad-
ditional chlorination of these produces other polychlorobenzenes, e.g.,
1,2,4-trichlorobenzene. Also, meta-dichlorobenzene is manufactured by
the isomerization of ortho or para-dichlorobenzene.
Commercial benzene chlorination may take two forms: 1) batch
process, or 2) Continuous process (Arthur D. Little, Inc., 1977). These
may be executed in the vapor phase for the production of monochlorobenzene
as is done by Dow, Midland, Michigan; but more generally the chlorination
is carried out in the liquid phase (Arthur D. Little, Inc., 1977).
Figure 5.11 schematically depicts the manufacture of chlorobenzene and
the representative reaction formulae.
crtloroDeruene
Chlorine
a\ \ 2
Hydrochloric 2 | | | 2 1
«c.d * jjl "hiM
HP T-1 Hvfl'~-t'lc,i-
Chlormilor Sodium «"d
|
Neutr*iinnt '
link
.enienr »r»d w«ler "N
Ifnitne jnd cMorobenjentJ
Chlorobeniini
Oichloro- ind
polychiorobentenei
lo ditliiuuon
Dich'orobrnrcne
sludge
to recovery
Figure 5.11 Schematic Diagram for the '"reduction of Chlorobenrene
and Dichlorobenzenes
5-26
-------
Cl
catalyst
HC1
Cl,
catalyst
or
HC1
ortho
In the batch process the benzene chlorination takes place at 40°C
to 60°C in the presence of a catalyst. After the proper density is at-
tained, the solution is neutralized, whereupon a sludge with a high
dichlorobenzene content settles out and is made ready for recovery of
the ortho and para structures. The balance of the neutralized solution
is distilled to obtain several fractions, one of which contains 75% of
the distillate and is pure monochlorobenzene (PEDCo, Inc., 1977). The
residue from this distillation is the principle source of para and ortho-
dichlorobenzene.
The alternate method for chlorobenzenes manufacture employs a
continuous chlorination and fractionation procedure so that the mono-
chlorobenzene is isolated as quickly as it is formed (PEDCo, Inc., 1977).
At this point neutralization and distillation of the monochlorobenzene
is executed in the same manner as in the batch process (PEDCo, Inc., 1977)
Figure 5.12 shows the producers of monochlorobenzene, ortho-, meta-,
and para-dichlorobenzene and their locations together with their respec-
tive 1976 capacities and the corresponding benzene requirements. Inde-
pendent capacity data that would assist in an assessment of the accuracy
of the figures given in this table were unavailable. Furthermore, the
listed capacities are subject to substantial change since the chemical
processes that are involved are inherently easy to control in meeting
demands for specific chlorobenzene isomers (Arthur D. Little, Inc., 1977).
The full capacity benzene requirement figures were based on the follow-
ing (SRI, 1977): 1) 85% yield for synthesis of the respective chloro-
benzenes, 2) a conversion factor of 0.82 kkg of benzene required per
5-27
-------
Ho CoajaoT
1 Allied Clnlcal Cor;
1976 CBLOROgENZENCS PRODUCTION CAPACITT ADD JEKZENt IIEQUIREMENT 81 COM?AMI£5
Monochlorobantana Dichlorobenxana Cap (kkg/yr) Full Capacity Bcnaena
Location
InduatriAl Chcalcala Divlalon Syractiaa, NT
Capacity (kVa/yr) Ortho
2 Chanical Producta Corp.
} to. Chaaical D.S.A.
» Eaauan Kodak Co., laanan
Organic Cheaicala
• 3 Koppara Co., In.
6 Guardian Qieaical Co.,
Eaatam Chcnical Dlviaion
Cartaravllla, GA
Midland, KI
Bochaatar, HI
rollaaabaa, WV
8auppau(a, NT
11,000
136.000
3,600
10,000
11,300
— 5.100
— 17,200
7 Dovar Chaailcal Corp.,
Subaldiary of I0C Induatrlaa Dovar, OR
8 (tonaanto Co., Monaanto
Induatrial Chanicala Co. Sau|at, IL
52,000
10
11
12
13
1*
13
U
California
Occidantal Patrolauai Corp.
PTC tnduatriaa. Inc.,
Chamical DlTlaion
Solvant OieBieal Co., Inc.
Solvent Chaaiical Co., Inc.
Spacialty Orianlca, Inc.2
Standard Chlorina Chaaical
Co., lac.
Standard Chlorlaa Ownical
Co., Inc.
Randaraon, HE
Hiaiara ralla, NT
Katrtua, HV
Maiden, MA
Niagara Palla, NT
Inindala, CA
Dalavara City, DC
Kaarny. IU
32,000
7,000
41.000
M
«A
-
34,000
__
7,300
9,100
MOO6
9.1007
1,000*
la.ioo9
7,300
BA HA
— 3.»00
DA
— 13.600
— HA
— 9.1007
— l.OOO8
— 27,200
— HA
14,600
6,200
129,200
HA
50.500
26,200*
1,700
«7,700
.OO5
J.6005
600
16,000
3.0005
Figure 5,12 Chlorobenzenes from Benzene (1976)
5-28
-------
Notes: on Figure 5.12
NA = Not Available
— = Not Manufactured
1. Exercise care in interpreting capacities since these chemical processes can be easily altered to meet
changing capacity demands.
2. From Mara and Lee, 1978, who cite SRI 1976 Directory fo Chemical Producers.
3. From Arthur D. Little, Inc., 1977 who cite SRI 1976 Directory of Chemical Producers.
4. Based on SRI, 1977 who state that 0.82 pounds of benzene are required per pound of monochlorobenzene produced,
and that 0.62 pounds of benzene are required per pound of ortho- or para-dichlorobenzene produced.
5. Based on known chlorobenzenes capacities.
6. Production to be phased out.
7. Includes capacity for ortho- and para-dichlorobenzena as well as 1,2,4-trichlorobenzene.
8. Includes capacity for ortho- and para-dichlorobenzene.
9. From Arthur D. Little, Inc., 1977, who cite Chemical Profiles (o_-dichlorobenzene, April 1, 1974), Schnell
Publishing Company, Inc., New York.
i
NJ
-------
kkg of monochlorobenzene produced, 3) a conversion factor of 0.62 kkg
of benzene required per kkg of ortho- or para-dichlorobenzene produced.
Although information concerning processes and capacities of commercial
meta-dichlorobenzene production was limited, the companies producing
this isomer are listed to indicate production locations. Benzene release
would only occur if benzene chlorination took place at. these facilities.
The amount of the meta isomer produced was not reported and was pre-
sumably small (Arthur D. Little, 1977).
Table 5.5 lists estimated 1978, 1979 chlorobenzehes production and
the corresponding benzene requirement. These data have been extrapolated
from the 1976 capacity figures of Figure 5.12. It was assumed that
each company produced the derivatives in a quantity that was proportional
to the fraction of the total U.S. chlorobenzenes production capacity
that the company supplied in 1976. Where plant capacity data were not
available then the capacity was assumed to have been zero. In equation
form the above is described by the following:
chlorobenzenes production of company in 1978 or 1979 = (% of total
U.S. capacity for chlorobenzenes production that a company contributed
in 1976) x (total U.S. chlorobenzenes production in 1978 or 1979).
Total monochlorobenzene production was taken from the U.S. Inter-
national Trade Commission January to August, 1979 supplement. A total
1979 monochlorobenzene production was calculated by assuming that the
production from January through August equals two-thirds of that which
will be complete by the end of December. Total 1979 monochlorobenzene
production was projected to be 137,000 kkg. Since 1979 ortho- and para-
dichlorobenzene production data were unavailable, 1978 figures were
taken from USITC: 19,000 kkg for both a- and £-dichlorobenzene. Accord-
ing to USITC information, several plants that were listed as having
capacity for production of chlorobenzenes in 1976 did not produce these
in recent years and therefore are not included in the production table
nor in the calculations involved for the generation of this table.
5.3.6.2 Benzene Emissions
5.3.6.2.1 Emissions to Air
The chlorination stage in the chlorobenzenes manufacture was
5-30
-------
Table 5.5 Estimated Chlorobenzenes Production and Benzene Requirement by Companies
Ui
I
OJ
No
Company
location
1 Allied Chemical Corp.
2 Dow Chemical U.S.A.
3 Dover Chemical Corp.,
Subsidiary of ICC Industries Dover, OH
4 Monsanto Co., Monsanto
Industrial Chemicals Co.
5 Montrose Chemical Corp. of
California
6 PPG Industries, Inc.,
Chemical Division
7 Standard Chlorine Chemical
Co., Inc.
8 Standard Chlorine Chemical
Co., Inc.
1979 1978
Monochlorobenzene Dichlorobenzene Prod(kkg/yr) Benzene
Production (kkg/yr) Ortho Meta Para Requirement (kkg/yr)
Syracuse, NY 4,900
Midland, MI 61,000
Dover, OH
Sauget, IL 23,300
Henderson, NE 14,300
Natrium, WV 18,400
Delaware City, DE 15,200
Kearny, NJ
1,200 — 1.600 5,800
3.700 ~ 5,100 55,500
NA NA
2,400 — — 20,600
NA — NA 11.8002
3,000 — 4,000 19,400
6,000 — 8,012 21,200
2
2,400 — NA 1,500
Notes:
NA = Not Available
— - Not Manufactured
1. Exercise care in interpreting chlorobenzenes production alnce these chemical processes can be
easily altered to meet changing demands. Figure 5.12 and USITC data have been used to calculate
production by plant.
2. Based on known chlorobenzenes production.
-------
presumably the principle source of atmospheric benzene emissions (PEDCo,
1977). Benzene emissions from the production of monochlorobenzene,
o_- and _£-dichlorobenzene will be quantitatively discussed in this section.
While m-dichlorobenzene is also a benzene derivative, the isomerization
of £- or £-dichlorobenzene required for its production is not a source
of benzene emissions unless the £- and £- isomers contain benzene as
an impurity. However, benzene emissions were associated with m-dichloro-
benzene production when one considers the emissions occurring in the
synthesis of the o_- or j)-dichlorobenzehe, the required intermediate.
These emissions could not be quantified since neither the amount of £- or
]3-dichlorobenzene used as the intermediate nor the actual production
volume of the meta structure was known. PEDCo (1977) listed the total
1975 atmospheric benzene emissions attributable to monochlorobenzene,
para-dichlorobenzene and ortho-dichlorobenzene production to have been
1065, 692, and 529 kkg respectively.
The U.S. International Trade Commission lists monochlorobenzene,
para-dichlorobenzene, and ortho-dichlorobenzene 1975 production to have
been 138,789, 24,798 and 20,751 kkg, respectively. The emission/production
quotient for each chlorobenzene isomer yielded the emission factor.
These are listed in Table 5.6. PEDCo (1977) also calculated emission
factors by computing the emission/production quotient. These are included
in Table 5.6. They employed the same 1975 emission quantities as above.
However, in order to estimate actual chlorobenzenes production they
cited 1977 plant capacity data and assumed that the facilities operated
at 80% of capacity.
The emission factors were applied to total U.S. production obtained
from USITC to derive total atmospheric benzene emissions due to the
production of each of the chlorobenzenes. USITC total production data
for monochlorobenzene from January to August, 1979 were available and
extrapolated to represent the entire year's production. The extrapolation
assumed that production from January through August equals two-thirds
of that which will be complete by December. Total 1979 monochlorobenzene
production was projected to be 137,000 kkg. Since 1979 production
figures for dichlorobenzehes were not accessible, total production of
5-32
-------
Table 5.6 Benzene Emission Factors (kkg of Benzene Emitted per kkg of
chlorobenzene produced)
Benzene Derivative
Monochlorobenzene
ortho-dichlorobenzene
para-dichlorobenzene
Emission Factors, kkg/kkg
1
0.0077
0.0279
0.0255
.0035J
.0100]
.0072]
1. Source: PEDCo (1977)
Table 5.7 Benzene Emissions Due to Chlorobenzenes Manufacture
Benzene Derivative
2
Monochlorobenzene
•3
ortho-dichlorobenzene"
para-dichlorobenzene
Benzene Emissions, kkg
1050
530
480
480J
1903
140]
1. Based on PEDCo (1977) emission factor.
2. Based on 1979 production.
3. Based on 1978 production.
5-33
-------
these was taken from USITC.1978, 19,000.kkg for both jo- and £-dichloro-
benzene.
When presenting benzene emissions due to chlorobehzenes synthesis
the higher of the two estimates will be quoted in order to represent a
worst-case estimate. See Table 5.7.
5.3.6.2.2 Benzene Emission to Water
Information needed to quantitatively assess benzene emissions was
not available. Versar (1979) estimated that discharges to POTW from
the manufacture of monochlorobenzene may be significant.
5.3.6.2.3 Benzene Emission to Land
Neither qualitative nor quantitative data concerning benzene
emissions to land as a result of chlorobenzenes manufacture were available.
5.3.6.2.4 Benzene Residue in Chlorobenzenes
Inherent in the processes used to manufacture chlorobenzenes is
the possibility that these are contaminated with benzene residues.
Benzene emissions as a result of this phenomenon currently cannot be
quantified due to the lack of relevant information.
5.3.7 Consumption of Benzene for Alkylbenzenes Synthesis
5.3.7.1 Processes, Producers, and Locations
Alkylbenzenes (also called detergent alkylates and dodecylbenzenes)
are formed by the alkylation of benzene with long-chain hydrocarbons.
Use of dodecene as the alkylating agent yields branched chain dodecylben-
zenes, while Friedel-Crafts alkylation with n-alkylchlorides yields
linear alkylbenzenes with 10-14 carbon chains (A.D. Little, Inc., 1977).
A1C1
3->
CH CH CH CH CITCH
3 Z. Z. 2. £-
5-34
-------
LOCATIONS
PRODUCERS AND BENZENE REQUIREMENTS
Map
No..
1
1
2
3
4
5
Company
Continental Oil
Continental Oil
Monsanto C.
Standard Oil
(Chevron)
Union Carbide
Ultco Chan.
Location
Baltimore, MD
Baltimore, MD
Chocolate Bayou, TX
Richmond, CA
South Charleston, WV
Carson, CA
Sum
Benzene Requirement, 1978
kkg
29,000 (straight chain)
3,800 (branched chain)
36,000 (straight chain)
34,000 (branched chain)
23,000 (straight chain)
8,400 (straight chain)
132,000
4Source: PEDCo, 1977.
Derived from 1976 plant capacities by the procedure described in Section 5.3.7.2.
Figure 5.13 Production of Alkylbenzenes
5-35
-------
I1 A1C1
5.3.7.2. Amount Consumed
Figure 5.13 includes the benzene requirement for each alkylbenzene
producer. This amount of benzene consumed was calculated as follows:
1) 1976 full capacity benzene requirements for each plant were
calculated from 1976 production capacities (PEDCo, 1977) by using the
conversion factor 0.40 kkg benzene used/kkg alkylbenzene produced for
both straight- and branched- chain alkylbenzenes (Neuf eld et al_., 1978).
2) The percent contribution made by each plant to the total
1976 full-capacity benzene requirement was calculated.
3) It was assumed that this percent contribution also described
the percent contribution to true production (as opposed to capacity)
in 1978 (as opposed to 1976). The percent contribution of each plant
was then multiplied by the estimated 1978 benzene consumption for
alkylbenzene synthesis (Figure 5.13) to yield the 1978 benzene con-
sumption by each plant.
5.3.7.3 Emissions Due to Alkylbenzene Synthesis
PEDCo (1977) cited a report of a personal communication between
EPA and Union Carbide as the source of an emission factor for benzene
release from a controlled linear alkylate process: 5 x 10 kkg/kkg
product. This was converted to units of kkg/kkg benzene consumed
using the conversion factor 0.40 kkg benzene/kkg alkylbenzene:
(5 x 10~4 kkg/kkg) 4- (0.40 kkg/kkg) = 1.3 x 10~3kkg
benzene released per kkg benzene consumed.
No data were available on synthesis of branched alkylbenzenes.
Therefore, it was assumed that the emission factor for the linear
alkylbenzene also applied to the branched alkylbenzene. This was
based on the chemical similarity between the two synthesis reactions.
5-36
-------
5.3.7.3.1 Emissions to Air
Emissions to air were judged to constitute 100% of total
emissions due to alkylbenzene synthesis. This judgement was based
on the following: 1) The reactions are carried out at 40-60 C,
where benzene would have a significant vapor pressure (A.D. Little,
1977). 2) Versar (1979) estimated water emissions to be sig-
nificant. 3) PEDCo (1977) mentions the possibility of airborne
emissions from incineration of light and heavy organic ends.
Application of the overall emission factor to benzene con-
sumption for alkylbenzene synthesis yielded: (132,000 kkg) x
_3
(1.3 x 10 kkg/kkg benzene used) = 170 kkg benzene released to
air in 1978 due to alkylbenzene synthesis. In the absence of
background information, no estimate of the uncertainty of the
emission factor was possible.
5.3.7.3.2 Emissions to Water
Emissions to water were judged to be negligible. See 5.3.7.3.1.
5.3.7.3.3 Emissions Due to Disposal of Solid Residues
Emissions due to solid residue disposal were included in the
estimate of releases to air (5.3.7.3.1).
5.3.7.3.4 Carry-Over of Benzene into Product
No information was available to permit estimation of benzene
contamination of alkylbenzenes. This question would be addressed
in a Level II materials balance by obtaining analytical data on
alkylbenzene batches.
5-37
-------
5.3.8 Consumption o,f Betizene fry SynthesjLjs_of Anthraquinone
5.3.8.1 Processes, Producers, Locations
According to (Merck, 1968) anthraquinone is produced from
phthalic anhydride and benzene in the presence of aluminum chloride
by a Friedel-Crafts reaction. The basic reaction in this process is:
0
II
-C
\ sZ-—^
AlCli
Anthraquinone
The only plant listed in any literature (Fentiman et al., 1979)
as producing anthraquinone is Toms River Chemical Corp., Toms River,
New Jersey.
5.3.8.2 Amounts Manufactured
Neufeld et al. (1978) estimated anthraquinone production by
using the USITC value for total Vat Due production. The USITC states
17,000 kkg of Vat Dye was produced in 1978. Neufeld et^ al. (1978)
used a conversion factor of 1.37 and Vat Dye production to calculate
benzene consumed for the production of anthraquinone:
(1.37) (17,000 kkg) = 23,000 kkg benzene
required for the production of anthraquinone in 1978. The un-
certainty of this value was unknown.
5-38
-------
5.3.8.3 Emissions of Benzene Due to Production of Anthraquinone
No emissions data were available with which to estimate benzene
emissions due to anthraquinone synthesis. A more detailed literature
search in this area might be carried out in a Level II study.
5.3.8.4 Carry-Over of Benzene into Product
Carry-over of benzene as a contaminant of anthraquinone may be
a source of emissions, and should be addressed in Level II study.
5.3.9 Consump»tion__qf BenzeneLby^Synthesis^qf^ Biphenyl
5.3.9.1 Processes, Producers, Locations
30% of the total biphenyl production is directly from benzene
by the process of thermally reacting benzene vapors. (Meylan and
Howard, 1976). The rest of the yearly biphenyl production does not
use benzene feedstock. The basic reaction in this process is:
H2
Biphenyl is produced from a thermal dehydrogenation of benzene
shown in Figure 5.14. Benzene and the recycled benzene are vaporized,
heated to about 600 C, and injected into a thermal reactor at 1-2 atm
pressure. The reactor raises the temperature to 700-850 C; the time
of exposure to the higher temperatures is about one second (Meylan
and Howard, 1976). Biphenyl was produced by this process only at
the Anniston, Alabama plant of Monsanto Industrial Chemicals Co.
5-39
-------
5.3.9.2 Amounts Manufactured
According to USITC 29,000 kkg of biphenyl were produced in 1978.
Based on the 30% figure given in Meylan and Howard (1976) for the
amount of biphenyl produced from benzene:
(29,000 kkg) x (0.3) = 8,700 kkg biphenyl produced
from benzene in 1978.
5.3.9.3 Emissions of Benzene Due to Production of Biphenyl
Data on emissions for 1978 were not found in any available
sources. Based on 1976 benzene consumption and emission data for
other non-fuel uses of benzene (Versar, 1979) an emission factor
was calculated to be:
200 kkg benzene released _ ,^,-0 .. . ... , .,
_. nnrr. , r • T = 0.0059 kkg benzene emitted per kkg
34,000 kkg benzene consumed e , 6 , re,
of benzene consumed
Using the benzene conversion factor given in Meylan and
Howard (1976) of 0.8 kkg benzene used/kkg biphenyl produced,
benzene consumption in the production of biphenyl was:
' Q = 11,000 kkg of benzene consumed
U. o
in the production of biphenyl in 1978. Based on these calculations,
benzene emissions were estimated to be:
(11,000 kkg)(0.0059 kkg/kkg) = 65 kkg benzene
released due to biphenyl synthesis during 1978.
Considering the uncertainties of the 1976 benzene emission
value (200 kkg + an estimated factor of 2) and the conversion
factor (0.8 + an estimated 20%) we estimated the cumulative
uncertainty of the benzene emissions in 1978 to be 65 kkg + a
factor of 3.
5-40
-------
5.3.9.3.1 Emissions to Air
Based on the calculations for emissions from Versar (1979), it
was judged that 99.5% of the emissions of benzene from the production
of biphenyl were to the air. Application of this percentage to total
estimated emissions yielded!
(65 kkg)(0.995) = 64 kkg of benzene released to
air during biphenyl synthesis.
5.3.9.3.2 Emissions to Water
As discussed in 5.3.9.3.1, it was estimated that 0.5% of total
emissions were to water. Application of this value to total estimated
emissions yielded:
(64 kkg)(0.005) = 0.3 kkg benzene released to
water during biphenyl synthesis in 1978.
5.3.9.3.3. Emissions Due to Disposal of Solid Residues
According to Meylan and Howard (1976), the only wastes from
the biphenyl process are the crude still pot residues. It is likely
that a small amount of benzene will be present in these residues,
however no data were available to permit estimation.
5.3.9.3.4 Carry-Over of Benzene into Product
Carry-over of benzene as a contaminant of biphenyl may be a
source of emissions, and should be addressed in a Level II study.
5.3.10 Benzenesulfonic Acid
Benzenesulfonic acid is a direct consumptive use of benzene,
used only in the production of phenol. According to Neufeld et_ al.
(1978), Reichold Chemicals, the only company producing phenol via
5-41
-------
the benzenesulfonic acid method, closed its plant in 1978. It was
not possible to estimate emissions for previous years.
5.3.11 Exports of Benzene
5.3.11.1 Amount Exported
Data provided by the Bureau of the Census (personal com-
howed the following amounts exp
1978 19791
munication) showed the following amounts exported:
,1
..151,000 68,000
1. Extrapolated from January - October preliminary figures.
5.3.11.2 Emissions Due to Benzene Export
Emissions due to export were considered to be attributable to
transport plus dockside loading. Emission factors for these processes
have been estimated by PEDCo (1977), and these factors were applied
to the export value (above). The results are summarized in Table 6.4.
Total estimated emissions were 17 kkg.
Table 5.8 Emission Factors and Emissions Due to Benzene Export, 1978
Process Emission Factor, kkg/kkg Emissjxms^^kkg
--4 2
Benzene transport 1.1 x 10" /week 17
—8
Benzene loading 6.9 x 10 0.01
Sum: 17 kkg
1. Source: PEDCo, 1977.
2. The average time in transit was estimated to be one week.
5.3.11.2.1 Emissions to Air
Of the total emissions estimated in 5.3.11.2, it was estimated
that 90% were released to air and 10% to water. This was based on
5-42
-------
the following: 1) The only significant emissions were those due to
transport. 2) Most spill losses would be over land, and the benzene
would evaporate before reaching water. 3) Fugitive losses would be
mostly evaporation during transfers.
Application of this breakdown of total emissions yielded:
(17 kkg) (0.9 fraction to air) = 15 kkg benzene
released to air due to exportation.
5.3.11.2.2 Emissions to Water
As discussed in -5.3.11.2.1, emissions to water were estimated
to be:
(17 kkg)(0.1 fraction to water) = 2 kkg benzene
released to water during exportation.
It was estimated that emissions to landfills or to solid
residues due to export of benzene would be negligible.
5-43
-------
Bwran*
Comprmor
or
dura*
O»«V«m A.
Bmnn*
Distal
Column
Biph«nvl;Po»vp»>.nvl
12-1SXBip»Mnyl
1-5%PD»yptttnvli
80-86% BOTIMM
8% o-Tvplwnyt
«9%m-Ttt|ilwny«
. etc.
Figure 5.14 Biphenyl from Thermal Dehydrogenation of Benzene
Source: Meylan and Howard, 1976
-------
Table 5.2 Consumptive Uses of Benzene, 1975-1979
Use
Ethylbenzene
Cumene
Cyclohexane
Nitrobenzene
Maleic Anhydride
Chlorobenzenes
Detergent Alkylates
Anthraquinone
Biphenyl
Benzenesulfonic Acid
Total
Amount of Benzene Consumed, Thousands of kkg
1975
kkg %
1,760 48
619 17
596 16
228 6
115 3
140 4
122 3
26 0.7
9 0.2
44 1
3,660 98.9
I-, - J
1976
kkg %
2,380 51
840 18
752 16
169 4
137 3
151 3
131 3
33 0.7
9 0.2
54 1
4,660 99.9
1977
kkg %
2,580 53
816 17
771 16
228 5
120 2
151 3
131 3
33 0.7
9 0.2
54 1
4,890 100.9
1978
kkg %
2,890 54
1,058 20
811 15
170 3
161 3
133 2
132 2
23 0.4
11 0.2
0 0
5,389 99.6
1979^
kkg %
2,910 52
1,240 22
839 15
1873 3
172 3
145 3
157 3
n.a.
n.a.
0 0
5,650 101
1. Neufeld ^t _al., 1978.
2. Partial data from USITC were used to estimate these values.
3. Estimated by the operation: 1979 benzene consumption
an:ne production
1978 aniline production
x 1978 benzene con-
sumption.
-------
6.0 NONCONSUMPTIVE USES OF BENZENE
Nonconsumptive uses are uses in which benzene is not removed from
the materials balance by chemical destruction or export. Examples are
given and discussed below.
6.1 TOTAL NONCONSUMPTIVE USE
Nonconsumptive use is a minor category of benzene use. An economic
input/output analysis for benzene (Neufeld &t^ al_., 1978) estimated that
the categories Miscellaneous Chemical Conversions + Solvent Uses +
Inventory Changes accounted for 5.1 percent of production in 1977. In
the absence of independent data, the 5.1 percent contribution was applied
to 1978 data:
(5,389,000 kkg benzene consumed)(0.051) = 270,000 kkg benzene
used for miscellaneous chemical syntheses* + solvent use + inven-
tory change in 1978.
Assuming the 1977 percent contribution is applicable, this value
represents an upper bound for nonconsumptive uses, since it also
contains an unknown amount of minor consumptive uses.
6.2 CATEGORIES OF USE
There are two categories of nonconsumptive uses for benzene, as
shown in Table 6.1. The amounts attributed to each category are estimates.
6.3 EMISSIONS BY CATEGORY OF USE
6.3.1 Use of Benzene as a Solvent
A small fraction of benzene production is used as a solvent in various
processes. Its use in this way has been dropping since the 1977 OSHA
Emergency Benzene Standard and the issuance of a regulation by the Consumer
Product Safety Commission banning benzene in consumer products, according
6-1
-------
Table 6.1 Nonconsumptive Uses of Benzene
Use
Amount (kkg), 1978
Solvent
Inventory Change
Total
9.5001; 27,0002
-272,000
270,000
1. Source: Estimate by Neufeld et_ ad., 1978.
2. Source: Mara and Lee, 1978.
3. Maximum value, estimated independently in section 6.1; not the sum
of uses.
6-2
-------
to Neufeld et_ al^. (1978). This statement was based on interviews with
405 companies that used benzene as a solvent in either chemical manu-
facturing processes or in consumer goods.
6.3.1.1 Locations of Use
Analysis of benzene sales for solvent use in 1976 revealed the
following geographical distribution (Neufeld et_ al., 1978):
Percent of Total
Region States Benzene Solvent Use
West South Central TX, LA, AR, OK 51
Middle Atlantic NY, PA, NJ 31
East North Central WI, IL, IN, MI, OH 10
East South Central KY, TN, MS, AL 5
West North Central ND, SD, NB, KS, MN, 1
IA, MO
6.3.1.2 Amounts of Use
Two estimates of the amount of benzene used as a solvent are
presented in Table 6.1. The value 9,500 kkg was estimated by Neufeld
^t jil. (1978) by summing projections of 1978 usage. These projections
would have been complicated by the OSHA and CPSC actions of 1977, and
the uncertainty of the estimate is approximately +10 percent, -50 percent,
The second estimate (Mara and Lee, 1978) is based on estimates of
the fraction of total benzene production that is used nonconsumptively.
Their estimate of 27,000 kkg used as solvents would have an estimated
uncertainty of +80 percent.
In the absence of a sound method to select between the values, the
higher estimate was selected in order to give a worst-case estimate of
emissions.
6.3.1.3 Emissions Due to Solvent Use
Neufeld _et^ £l- (1978) have recently reported on use of benzene as
a solvent and the emissions associated with this use. Table 6.2
summarizes their estimates. These authors also document the effect
6-3
-------
Table 6.2 Estimated Emissions of Benzene as a Solvent
2
Type of Solvent Use
General Organic Synthesis
Pharmaceutical Synthesis
Small Volume Chemicals
Aluminum Alky Is
Alcohols
Paint Removers
Estimated Emissions, kkg, 1978
1,000
200
1,000
200
5003
Sum: 2,900
Source: Neufeld et_ al., 1978. Estimates of emissions were
based on emission control information obtained during interviews
with companies using benzene as a solvent.
Other types of use that either no longer employed benzene or
produced an estimated zero emissions of benzene were: Bisphenol-A,
ethyl cellulose, adhesives, tire manufacture, tire retreading,
industrial rubber products, tire patch repair kits, automotive
adhesives, shoe adhesives, paints and coatings.
Due to product manufactured before May, 1977, and sold in 1978.
6-4
-------
of the 1977 OSHA and CPSC actions on benzene use: Estimated losses of
benzene due to solvent use were 6,000 - 7,000 kkg in 1976 and 2,900 kkg
in 1978. Cyclohexane is replacing benzene in many solvent uses.
6.3.1.3.1 Emissions to Air
Distribution of the total emissions among air, landfill, and water
required estimation. On the average, 50 percent of emissions were
estimated to be airborne and 50 percent waterborne. This estimate was
based on the observations that: 1) Chemical manufacturing is more
likely to cause releases to water, due to the nature of the synthesis
processes, and 2) Formulation uses are more likely to yield uncontrolled
evaporation.
Based on this reasoning, the estimated emissions to air were:
(2,900 kkg) x (0.5) = 1,450 kkg benzene released to air during use as
a solvent in 1978.
6.3.1.3.2 Emissions to Water
Based on the reasoning in 6.3.1.3.1, the emissions to water would
be 1,450 kkg benzene released during solvent use in 1978.
6.3.1.3.3 Emission Due to Disposal of Solid Residues
Releases of benzene due to disposal of solid residues were not
quantifiable but were estimated to be small. In order to evaluate these
emissions, it would be necessary to know: 1) the rate of production
of benzene-containing residues, 2) the weight % benzene in the residues,
and 3) the method of residue disposal.
6.3.1.3.4 Carryover of Benzene into Products
Benzene may be carried over into products during use as a solvent in
chemical synthesis. This area should be addressed in Level II study.
6.3.2 Changes in Benzene Inventory
Changes in the inventory of benzene during a given year may be
considered "negative" or"positive" nonconsumptive use. Emissions due
6-5
-------
to inventory have been considered under storage emissions. The purpose
of the present estimate of inventory change is to aid in balancing sup-
plies with uses.
The December 31 inventories of benzene for 1977 and 1978 are shown
below. During 1978, inventories dropped by 272,000 kkg.
Year Benzene Inventory as of 21/31, kkg Change During 1978
1977 643,OOO1
-272,000 kkg
1978 371,OOO2
1. Source: National Petroleum Refiners Association, as cited in
Neufeld et al., 1978.
2. Source: National Petroleum Refiners Association (personal
communication with Linda Dziuban).
6-6
-------
7.0 USE OF BENZENE AS A.FUEL COMPONENT
7.1 BENZENE IN GASOLINE
7.1.1 Concentration of Benzene in Gasoline
The benzene concentration in gasoline depends on several factors
including the source of the crude oil from which the gasoline was made,
the geographical location of the source and the refiner, the grade of
gasoline, refinery operations, and the seasonal blends produced by each
refinery (PEDCo, 1977).
Prior to 1974, the average benzene content in U.S. gasoline was less
than 1 percent by volume. More recent data indicate that the average
benzene content has been increased to maintain octane levels as lead con-
tent has been reduced (Mara and Lee, 1978). A NIOSH study using 1976
data found an average benzene concentration of 1.24 percent by liquid
volume (a weighted average based on fuel type and amounts produced). A
nationwide survey during February and March 1977 produced a range of con-
centration from 1.25 to 5.0 percent, averaging out to 2.5 percent benzene
by volume. This value might be high since benzene concentrations are
greater in winter blends. The investigators in this survey estimated an
average of 2 percent throughout the year (PEDCo, 1977). In the present
report, we will assume a value of 2 percent by liquid volume for benzene
concentration in gasoline during 1979.
7.1.2 Amount of Benzene in Gasoline
The 1979 consumption of gasoline was estimated to be 1.08 x 10 gal-
lons (American Petroleum Institute, 1980). We judge the uncertainty of
this value to be +10 percent. The 1979 benzene content of gasoline
totaled:
(1.08 x 1011 gal;) (2%) = 2 x 109 gal benzene = 7 x 10 kkg benzene.
Since this is larger than the annual production of benzene, it is prob-
ably more accurate to consider benzene as a native constituent of
7-1
-------
gasoline rather than an additive. It is unknown how much of this benzene
originates in the 4,960,000 kkg of direct benzene production, but is un-
likely that a significant amount does. The basis for this statement is
the fact that aromatics (including benzene) for gasoline are contained
in reformate which makes up 20 percent of the "pool" of materials going
into gasoline. BTX (a mixture of benzene, toluene, and xylene) is often
separated from reformate and then is blended back into the gasoline pool
along with other octane-raising compounds (USITC, 1978). Thus, if ben-
zene is used as an octane-raising additive, it is mixed in the gasoline
pool as a component of BTX and not as pure benzene. In this situation
the benzene would not be counted in total benzene production reports
since it was not separated from the BTX. Gasoline production, therefore,
significantly contributes to the amount of benzene made available to the
environment, and gasoline usage thus becomes a. potentially great source
of benzene emissions.
7.1.3 Benzene Emissions from Gasoline Use
7.1.3.1 Benzene Emissions Due to Gasoline Production
No factors for benzene emissions to air, land or water specifically
attributable to the production of gasoline were found in the literature.
Since gasoline is produced at refineries where presumably other processes
are also taking place, any emissions from gasoline production specifically
would be included in emissions from petroleum refineries generally.
See Section 3.1 for emissions from petroleum refineries in general.
7.1.3.2 Benzene Emissions Due to Gasoline Storage
Figure 7.1 is a diagram representing the flow of gasoline from
production center to its ultimate combustion in a motor vehicle engine.
The gasoline distribution system involving transport from the petroleum
refineries to the consumer, with intermediate storage stops, may be a
significant source of atmospheric benzene. Gasoline is shipped from
refinery storage areas to bulk terminals (regional distribution centers)
by ship, barge, railcar, and pipeline. It is then transported from the
7-2
-------
REFINERY STORAGE
_L
SHIP, RAIL, BARGE
BULK TERMINALS
TANK TRUCK
PIPELINE
\
f
SERVICE STATION
i
)
r
AUTOMOBILES,
TRUCKS
V
BULK PLANTS
i
1 r
TRUCKS
^
COMMERCIAL,
RURAL USERS
1
(PEDCo,
Figure 7.1 Gasoline Distribution System
7-3
-------
terminal by tank truck to service stations and commercial and rural users,
either directly or via bulk plants (local distribution centers) (PEDCo,
1977; Mara and Lee, 1978).
Gasoline may be stored in either floating-roof or fixed-roof tanks,
each with its own emission characteristics. Floating-roof tanks are more
common to refineries and very large storage sites. They do not emit the
"breathing" emissions associated with fixed-roof tanks (JRB, 1980). We
have assumed that all the gasoline produced is stored in floating-roof
tanks at bulk terminals, since it is not known exactly how many terminals
have floating-roof tanks. Emissions from refinery storage are covered
in the section on refinery emissions.
Hydrocarbon emissions from floating-roof tanks occur primarily from
standing and withdrawal losses. Standing losses are caused by the im-
proper fit of the seal and shoe to the vessel shell and when vapor escapes
between the flexible membrane seal and roof. Withdrawal losses are caused
by evaporation of gasoline from the tank walls as the roof descends during
emptying operations (Mara and Lee, 1978). In this report, we have used
PEDCo's methods of calculating emissions from floating-roof tanks since
their explanation of the process appeared more complete.
Method of calculation (from PEDCo, 1977):
1. Assume that all gasoline is stored in floating-roof tanks
with a typical capacity of 2.3 x 106 gal (8,700 m3).
2. Assume that the average tank retention time is 30 days.
3. Assume that the average tank is 75 percent full.
A. Use the following factors for hydrocarbon losses:
a. 132 Ib/day/tank for standing storage losses.
3
b. 0.025 lb/10 gallons for losses during withdrawal.
5. Apply these factors to the Annual Domestic Gasoline Consump-
tion (ADGC) in gallons to get total hydrocarbon emissions.
7 -4
-------
6. To convert hydrocarbon emissions to benzene emissions assume
benzene to be equal to 2 percent by liquid volume of gasoline
and vapor-phase concentration at 40 percent of liquid concen-
tration (converted to vapor weight basis, this is 45 percent).
We have used 1.08 x 10 gallons from Section 7.1.2 as the 1979 ADGC.
Benzene emissions to air from standing storage of gasoline in floating-
roof tanks equal:
(1.08 x 101:Lgal)(132 -= — ^— r)(30 days) (0.02) (0.45) (454 x 10~6 •£
__ ctcty
(2.3 x 106 ff^MO.75 full)
- 1,000 kkg benzene emitted to air during standing storage in 1979.
i
Because the uncertainty of the emission factor was not discussed in the
literature available, an uncertainty could not be assigned to this emis-
sion estimate. Benzene emissions to air from withdrawal of gasoline from
floating- roof tanks equal :
(1.08 x 10UgaD (0-025 -) (0.02) (0.45) (454 x 10~6
10 kkg benzene emitted to air during withdrawal in 1979.
It was not possible to evaluate the uncertainty range of this value.
There was no indication in the literature of emissions due to the
filling of these tanks, thus none were calculated.
Gasoline bulk plants (centers for local distribution of gasoline)
contain a large number of fixed-roof gasoline tanks (JRB, 1980). These
plants handle a low volume of gasoline --- approximately 4,000 gallons
per day (uncertainty: + 25 percent) --- as opposed to the 250,000 gallons
per day for bulk terminals (Mara and Lee, 1978).
The 1972 Census of Business indicated that there were 23,367 bulk
plants in the United States. The number is believed to be decreasing.
The number of plants dropped 11 percent between 1967 and 1972. Assuming
that the number of plants continued to decrease at approximately the same
rate (2.2 percent per year, or about 15 percent between 1972 and 1979),
there were about 20,000 (+ 20 percent) such plants in operation in 1979
7-5
-------
(JRB, 1980). Using these values, the volume of gasoline passing through
bulk plants in 1979 was estimated to be:
(4,000 gal/plant/day)(365 days)(20,000 plants) = 3 x 1010gallons.
This was about one quarter of the total U.S. annual consumption. The
uncertainty of this estimate was judged to be + 50 percent.
Thirty percent of the bulk terminals are underground storage facili-
ties, and thus do not use fixed-roof tanks. The typical gasoline bulk
plant contains three tanks. An above-ground fixed-roof tank is estimated
to lose 3 kg/day of gasoline vapors to breathing losses (caused by
diurnal temperature changes causing expansion of the material in the
tank with resultant vapor escape through vents to air)(JRB, 1980). Based
on the above, we estimated the hydrocarbon emissions to air from tank
breathing to be:
po.ooo"]
[jilantsj
70% above-"
ground
facilities
[3 tanks /I
plant
3 x 10
-3
day/tank
loss
kkg/
365
days/
year
5 x 10 kkg
= hydrocarbon
emissions
This is converted to benzene emissions by using the 2 percent benzene in
gasoline by liquid volume factor and 0.45 vapor-phase conversion factor
(PEDCo, 1977):
(5 x 10 .kkg/1 yr)(0.02)(0.45) = 500 kkg/yr benzene lost to air
from tank breathing
The uncertainty of this value could not be estimated because the origin
and reliability of the emission factor were unknown.
Fixed-roof tanks also emit hydrocarbon vapors while they are being
filled. The liquid gasoline will displace an equal volume of vapor and
force it into the atmosphere. Filling operations emissions can be con-
trolled by vapor balance systems, hydrocarbon oxidation systems, and
refrigerated vent systems. These emission control systems typically
reduce tank filling losses and emissions by approximately 80 percent.
We assumed that 25 .percent of fixed-roof tanks have emission controls
(JRB, 1980).
7-6
-------
Benzene emissions due to filling. of fixed-roof tanks were estimated
using the following assumptions and steps:
1. (3 x 10 gal stored/yr)(70% in above-ground fixed-roof tanks) =
2 x 10 gal stored in above-ground fixed-roof tanks.
2. There will be 2 x 10^ gal of vapor vented in 1 year
(or 8 x IQlO £).
3.
4.
5.
Assume this gasoline vapor is at 1 atm pressure.
The volume of benzene in the
(8 x 1010 £) 0-02 fraction~
of benzene in
gasoline
vapor =
0.40 conversion
of benzene liquid
_to benzene vapor
Assume standard temperature and pressure (25 C,
22.4 L 6 x 108£ ,,
i($9\ ( 78 N 9
6 x 108£
= benzene vapor
displaced.
1 atm) ; then-
1 n" -rr =
71
2,000 kkg displaced (uncontrolled).
6. Assuming 25 percent of these emissions are controlled (500 kkg),
and that the controlled emissions are reduced by 80 percent
(or 400 kkg), the total benzene emissions from filling above-
ground fixed-roof tanks = 2,000 - 400 = 1,600 kkg. We estimate
the uncertainty of this value as + a factor of 4, based on the
values entering into the total.
No information on emissions to air from fixed-roof tanks during gaso-
line withdrawal was found in the literature, therefore none were calcu-
lated. We believe such emissions to be unlikely, since fresh air would
be drawn into the tanks during gasoline withdrawal, rather than hydro-
carbon vapor being pushed out.
As mentioned earlier, 30 percent of the bulk plants are underground
facilities (JRB, 1980). We assumed, therefore, that (0.30)(3 x 10 or
1 x 10 gallons of gasoline were stored in such facilities. No infor-
mation on such facilities was found in the literature, though it would
be reasonable to assume that underground storage tanks are of fixed
volume and would be subject to the same type of filling "displacement"
emissions as above-ground fixed-roof tanks are. Assuming no controls,
and making the same assumptions and calculations as done above for
above-ground fixed-roof tanks (but for half the volume), we estimated
7-7
-------
that 1,000 kkg of benzene were emitted from underground tanks during fil-
ling. This estimate has a probable uncertainty of 4- a factor of A.
As with above-ground fixed-roof tanks, it would be reasonable to
assume that gasoline withdrawal from underground tanks causes negligible
emissions. There is, however, the possibility of standing storage loss
from underground tanks either "breathing" losses or losses specific to
that type tank. We encountered no information in the literature on such
losses and, therefore, made no estimates. This is an area which should
be considered in a Level II or III report.
No factors for emissions of benzene to water or land during gasoline
storage operations were found in the literature, thus no emissions were
calculated. Due to the high volatility of benzene, it is believed that
such emissions, if any, would be negligible.
Total benzene emissions to air from gasoline storage at bulk termi-
nals and plants were estimated to be 1,000 kkg (floating-roof) + 10 kkg
(withdrawal from floating-roof) + 500 kkg (fixed-roof, breathing) +
1,600 kkg (fixed-roof, filling) + 1,000 kkg (underground filling) =
4,000 kkg plus underground storage losses. It was not possible to assign
an overall uncertainty to this estimate.
7.1.3.3 Benzene Emissions Due to Gasoline Transportation
As can be seen from Figure 7.1, presented earlier, there are several
points on the gasoline flow diagram at which the product is loaded into a
conveyance of some type and shipped to a distribution center. No data
were available on amounts transported by pipeline, ship, rail, or barge.
It appears that all gasoline is transported by truck eventually in order
to get to the service stations and commercial and rural users. Transit
times, however, are unknown, and probably vary depending upon each region's
and locality's transportation and distribution system. Several assumptions
were made in order to estimate emissions from transportation of gasoline.
These assumptions were unsupported by direct data. The data should be
developed as part of a Level II or Level III materials balance.
7-8
-------
We assumed the following:
1. All the gasoline produced leaves the refinery storage area by
railcar or marine conveyance (ship or barge), since we have
no data whatsoever on pipeline transport.
2. Fifty percent of the gasoline is transported by railcar and
50 percent by marine conveyance between the refinery and bulk
terminal; and all transport between terminal, bulk plant and
service stations is by truck.
3. Seventy-five percent of the tank cars are filled by the submerged-
filling or bottom-loading methods (gasoline being charged to the
vehicle below the liquid surface) and 25 percent by the splash-
filling method (involving a "freefall" of gasoline through the
air)(PEDCo, 1977).
4. An overall efficiency of 95 percent was assumed for vapor
recovery/containment for loading and unloading of rail tank-cars
and tank-trucks, and unloading of barges and ships (PEDCo, 1977).
5. Gasoline has an average benzene content of 2 percent by liquid
volume; the conversion factor for benzene liquid volume to
benzene vapor weight is 0.45 (PEDCo, 1977).
6. The following emission factors were used (PEDCo, 1977):
Uncontrolled Hydrocarbon
Emission Source Emission Factor
Loading:
3
Tank car/Truck-Splash 12 lb/10 gal
3
Tank car/Truck-Submerged 5 lb/10 gal
Marine 1.8 lb/10 gal
Transit 3 lb/wk/103 gal
Unloading 2.1 lb/103 gal
7. Transit time between the refinery and bulk terminal was assumed
to be one week; bulk terminal to service station or bulk plant
was one day; and bulk plant to service station and/or commercial-
rural users was one day.
8. The annual domestic gasoline consumption was 1.08 x 10 gallons
(see Section 7.1.2).
9. The amount of gasoline handled by bulk plants was 3 x 10 gallons
(see Section 7.1.3.2).
7 -9
-------
Benzene emissions due to transport of gasoline from refinery stor-
age to bulk terminal were estimated as follows;
Loading tank cars, splash method:
(1.08 x 1011 gal)(0.50 to tank cars)(0.25 splash method) x
(12 lb/10 gal hydrocarbon emission factor)(0.05 uncontrolled
emissions)(0.02 benzene in gasoline)(0.45 vapor phase conver-
sion) (454 x 10~6 kkg/lb) = 30 kkg.
Loading tank cars, submerged method:
(1.08 x 10 gal)(0.50 to tank cars)(0.75 submerged method) x
(5 lb/10 gal hydrocarbon emission factor)(0.05 uncontrolled
emissions)(0.02 benzene in gasoline)(0.45 vapor phase conver-
sion) (454 x 10~6 kkg/lb) = 40 kkg.
Loading marine conveyances:
(1.08 x 1011 gal)(0.50 to marine)(1.8 lb/103 gal hydrocarbon
emission factor)(0.02 benzene in gasoline)(0.45 vapor phase
conversion)(454 x 10~6 kkg/lb) = 400 kkg.
Transit (all modes):
(1.08 x 1011 gal)(3 lb/wk/103 gal)(l week transit time) x
(0.02 benzene in gasoline)(0.45 vapor phase conversion) x
(454 x 10~6 kkg/lb) = 1,000 kkg.
Unloading (all modes):
(1.08 x 1011 gal)(2.1 lb/103 gal)(0.05 uncontrolled emissions)
(0.02 benzene in gasoline)(0.45 vapor phase conversion)
(454 x 10~6 kkg/lb) = 50 kkg.
Total benzene emissions, transport of gasoline from refinery to
bulk terminal = 30 kkg + 40 kkg + 400 kkg + 1,000 kkg + 50 kkg =
2,000 kkg.
Emissions due to transport of gasoline by tank truck from bulk
terminal to service stations and bulk plants were calculated as follows;
7-10
-------
Loading tank truck, splash method:
(1.08 x 1011 gal)(0.25 splash method)(12 lb/103 gal hydrocarbon
emission factor)(0.05 uncontrolled)(0.02 benzene in gasoline)
(0.45 vapor phase conversion)(454 x 10~ kkg/lb) = 70 kkg.
Loading tank truck, submerged method:
(1.08 x 1011 gal)(0.75 submerged method)(5 lb/103 gal hydro-
carbon emission factor)(0.05 uncontrolled)(0.02 benzene in
gasoline)(0.45 vapor phase conversion)(454 x 10~ kkg/lb) = 80 kkg.
Transit:
(1.08 x 1011 gal)(3 lb/wk/103 gal)(1/7 week in transit) x
(0.02 benzene in gasoline)(0.45 vapor phase conversion) x
(454 x 10~6 kkg/lb) = 200 kkg.
Unloading:
(1.08 x 1011 gal)(2.1 lb/103 gal)(0.05 uncontrolled emissions)
(0.02 benzene in gasoline)(0.45 vapor phase conversion)
(454 x 10~6 kkg/lb) = 50 kkg.
Total benzene emissions, transport of gasoline from bulk terminals
to service stations and bulk plants = 70 kkg + 80 kkg + 200 kkg + 50 kkg
400 kkg.
Benzene emissions due to transport of gasoline from bulk plants to
commercial-rural users and service stations were calculated as follows:
Loading tank truck, splash method:
(3 x 1010 gal thru plants)(0.25 splash method)(12 lb/103 gal
hydrocarbon emission factor)(0.05 uncontrolled)(0.02 benzene
in gasoline)(0.45 vapor phase conversion)(454 x 10 kkg/lb) =
20 kkg.
7-11
-------
Loading tank truck, submerged method:
10 "?
(3 x 10 gal thru plants)(0.75 submerged method)(5 lb/10 gal
hydrocarbon emission factor)(0.05 uncontrolled) (0.02 benzene
' kkg/lV
20 kkg.
in gasoline)(0.45 vapor phase conversion)(454 x 10~ kkg/lb) =
Transit:
10 "?
(3 x 10 gal thru plants)(3 lb/wk/10 gal)(1/7 week in transit)
(0.02 benzene in gasoline)(0.45 vapor phase conversion)
(454 x 10~6 kkg/lb) = 50 kkg.
Unloading:
(3 x 1010 gal thru plants)(2.1 lb/103 gal)(0.05 uncontrolled
emissions)(0.02 benzene in gasoline)(0.45 vapor phase conver-
sion) (454 x 10~6 kkg/lb) = 10 kkg.
Total benzene emission from transport of gasoline from bulk plants to
commercial-rural users and service stations = 20 kkg + 20 kkg + 50 kkg +
10 kkg = 100 kkg. Total benzene emissions due to transportation of
gasoline = 2,000 kkg (refinery to bulk terminal) + 400 kkg (bulk terminal
to bulk plants and service stations) + 100 kkg (bulk plants to service
stations, etc.) = 3,000 kkg. It was not possible to assign an uncertainty
to this sum due to a lack of information on the emission factors contained
in the individual terms.
7.1.3.4 Benzene Emissions Due to Gasoline Vending
We assumed that commercial and rural users have the same types of
storage and dispensing facilities, and therefore the same emission char-
acteristics, as service stations. This assumption allowed us to consider
the national gasoline consumption volume as having passed through service
stations.
There are several sources of benzene emissions at service stations.
Emissions occur from the displacement of hydrocarbon vapors to the at-
mosphere while gasoline is being loaded into storage tanks,from under-
ground tanks breathing, and from vehicle refueling operations (PEDCo, 1977)
7-12
-------
The estimates of benzene emissions from service stations were based
on the rate of gasoline consumption (from Section 7.1.2), a 2 percent
benzene content by liquid volume, and a 0.45 factor used to convert
liquid concentration (volume percent) to vapor concentration (weight
percent). The emission factors and estimated frequencies of usage from
PEDCo (1977) are summarized below.
Hydrocarbon Relative
Emission Rate, '"'Frequency
Emission Source lb/103 gal of Use, %
Filling Storage Tank:
Submerged 7.3 50
Splash 11.5 25
Balanced 0.3 25
Underground Tank Breathing 1.0 100
Vehicle Refueling Displacement Losses
Uncontrolled 9.0 75
Controlled 0.9 25
Spillage (all assumed to evaporate) 0.7 100
Emissions due to the individual processes listed in the table are
estimated below.
Filling storage tanks, submerged:
(1.08 x 1011 gal)(0.50 usage)
rate)(0.02 benzene) x (0.45 vapor phase conversion)
(454 x 10~6 kkg/lb) =2,00
Filling storage tanks, splash:
(1.08 x 1011 gal)(0.25 usa
rate)(0.02 benzene)(0.45 vapor phase conversion)
(454 x 10"6 kkg/lb) = 1,000
Filling storage tanks, balanced:
(1.08 x 1011 gal)(0.25 usage
rate)(0.02 benzene)(0.45 vapor phase conversion)
ckg from balanced filli
Total filling emissions: 3,000 kkg.
(1.08 x 1011 gal)(0.50 usage)(7.3 lb/10 gal submerged emission
32 I
(454 x 10~6 kkg/lb) = 2,000 kkg from submerged filling.
(1.08 x 1011 gal)(0.25 usage)(11.5 lb/103 gal splash emission
32 t
(454 x 10"6 kkg/lb) = 1,000 kkg from splash filling.
(1.08 x 1011 gal)(0.25 usage)(0.3 lb/103 gal balanced emission
32 I
(454 x 10~6 kkg/lb) = 30 kkg from balanced filling.
7-13
-------
Underground tank breathing:
(1.08 x 1011 gal)(1.0 lb/103 gal emission factor)(0.02 benzene)
(0.45 vapor phase conversion)(454 x 10~ kkg/lb) = 400 kkg from
underground tank breathing.
Vehicle refueling displacement losses, uncontrolled:
(1.08 x 1011 gal)(0.75 usage)(9 lb/103 gal uncontrolled emission
factor) x (0.02 benzene)(0.45 vapor phase conversion)
(454 x 10~ kkg/lb) = 3,000 kkg from uncontrolled refueling.
Vehicle refueling displacement losses, controlled:
(1.08 x 1011 gal)(0.25 usage)(0.9 lb/103 gal controlled emission
factor)(0.02 benzene)(0.45 vapor phase conversion)
(454 x 10~ kkg/lb) = 100 kkg from controlled refueling.
Total vehicle refueling emissions: 3,000 kkg.
Spillage emissions:
(1.08 x 1011 gal)(0.7 lb/103 gal emission factor)(0.02 benzene)
(1 total evaporation) x (454 x 10~ kkg/lb) = 700 kkg from
spillage.
Total benzene emissions due to gasoline vending operations: 7,000 kkg.
7.1.3.5 Benzene Emissions Due to Gasoline Combustion
Hydrocarbon emissions from gasoline-powered vehicles without emission
controls may be divided into two types: evaporative and exhaust emissions.
The evaporative emissions originate from a) the carburetor (evaporation
of fuel after a hot engine is turned off); b) the fuel tank (from vents,
with emissions increasing as tank temperature increases); and c) the
crankcase (from "blowby" past the piston rings). The exhaust emissions
result from incomplete combustion of the fuel (PEDCo, 1977).
Benzene emission levels depend not only on benzene levels in gaso-
line, but also on particular characteristics of the gasoline belend.
7-14
-------
For example, if the fuel blend contains ethylbenzene, incomplete combus-
tion may result in the ethylbenzene being converted into benzene; thus,
it is possible that more benzene may be exhausted than is present origi-
nally in the fuel (PEDCo, 1977). In a material balance scheme, therefore,
combustion of fuel in a vehicle engine should be considered both as a use
and as a source of benzene. It would be difficult to determine how much
of the exhaust benzene is attributable to non-combustion of the benzene
originally present in the fuel, and how much is attributable to the
breakdown of more complex substances.
Emission control systems in use on current-model motor vehicles
include PCV (positive crankcase ventilation), EGR (exhaust gas recircu-
lation), evaporative controls and catalytic oxidation systems (PEDCo,
1977).
Using an estimate of 96 million cars in the United States (PEDCo,
1977), and an emission factor for evaporative emissions of 0.148 g
benzene per trip with an average vehicle making 3.3 trips per day
(Mara and Lee, 1978), 17,000 kkg of benzene are estimated to be released
to the atmosphere through evaporation. It is unknown whether or not
control technology was taken into account in the development of the
above emission factor.
EPA-ORD and General Motors determined benzene tailpipe emission
factors of 0.005 g/mile - 0.020 g/mile for automobiles with catalytic
converters, and 0.05 g/mile - 0.15 g/mile for those without catalytic
converters (Gray, 1979).
Using an estimate of 96 x 10 vehicles (PEDCo, 1977) (assuming
that the number of vehicles hasn't changed substantially since 1976,
the year for which the estimate was made), an average of 9,494 miles
travelled per car (PEDCo, 1977)(assuming that driving habits haven't
changed since 1974, the year for which this estimate was made), an
estimate that 73 percent of these vehicles have exhaust controls
(PEDCo, 1977) (assuming that this 1976 estimate is still applicable),
and the above emission factors, we estimated tailpipe emissions of
7-15
-------
benzene to be:
(96 x 106 vehicles)(9,494 mi/vehicle)(73% emission-controlled)
(5-20 x 10~9 kkg/mi) = 3,000 - 10,000 kkg.
(96 x 10 vehicles)(9,494 mi/vehicle)(27% not controlled)
(5-15 x 10~8 kkg/mi) = 10,000 - 40,000 kkg.
*
Total benzene exhaust emission: 10,000 - 50,000 kkg.
17,000 kkg + (10,000-50,000 kkg) = 30,000 - 70,000 kkg total ben-
zene emissions attributable to automobile use.
7.1.3.6 Total Benzene Emissions Due to Benzene in Gasoline
Process Emissions (to air), kkg
Gasoline production Tabulated in Section 3.1
Gasoline storage 4,000
Gasoline transport 3,000
Gasoline vending 7,000
Gasoline combustion 30,000 - 70,000
SUM: 40,000 - 80,000 kkg
7-16
-------
8.0 SUMMARY OF DISPOSAL/DESTRUCTION AS END-PRODUCTS
No data were presented in the readily available literature that
permitted evaluation of the disposal of benzene-containing residues
or the destruction of benzene-containing end-products. It was judged,
however, that benzene would readily volatilize from landfilled solid
residues. The question of benzene re-formation during destruction of
end-products remains open. A deeper literature search—such as that
accompanying a Level II materials balance—would be necessary to
obtain information on these points.
8-1
-------
9.0 LOCATIONS OF BENZENE EMISSION SITES
Table 9.1 summarizes the relative benzene "activity" at each site
containing at least one production or use facility. The "activity"
factor is in units of 10 kkg benzene produced plus consumed. Since
this report assumes throughout that emissions of benzene at a site are
proportional to the amount produced or consumed at that site, it is
this "activity" factor that is used to rank the sites as "hotspots."
The table also shows the number of plants contributing to benzene
processes in each area. Also, the locations are grouped by county,
so that a group of individual plants in adjacent small towns would
not go unremarked.
The results show that the most likely emission sites, in decreas-
ing order, are: Harris County, TX; Jefferson County, TX; Nueces
County, TX; Brazoria County, TX; Galveston County, TX; and Puerto Rico.
9-1
-------
Table 9.1 Locations of Benzene Emission Sites, 1978
County, State
Number of Plants
"Activity" Factor:
Total Production and
Consumption in units
of 105 kkg
Harris Co, TX
Jefferson Co., TX
Calveston Co., TX
Brazoria Co., TX
Nueces Co., TX
Puerto Rico
East Baton Rouge Co.,
Iberville Co., LA
Howard Co., TX
Los Angeles Co., CA
Lucas Co., OH
Caddo Co., LA
Ector Co., TX
Tulsa Co., OK
Gloucester Co., NJ
Allegheny Co., PA
Marshall Co., WV
Saint James Co., LA
Midland Co., MI
Baltimore Co., MD
Virgin Islands
Boyd Co., KY
Erie Co., NY
Calcasieu Co., LA
Bay Co., MI
Butler Co., KS
Philadelphia, PA
Madison Co., IL
Delaware Co., PA
St. Bernard Co., LA
Cook Co., IL
Butler Co., OH
Northampton Co., PA
Pueblo Co., CO
Beaver Co., PA
Somerset Co., NJ
Jackson Co., MS
Wetzel Co., WV
Ascension Co., LA
Calhoun Co., TX.
St. Louis Co., MO
Union Co., NJ
Middlesex Co., NJ
Grundy Co., IL
Contra Costa Co., CA
Kanawha Co., WV
Calhoun Co., AL
LA
9
9
4
5
7
4
2
1
3
3
2
2
2
2
2
2
2
1
1
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
14
11
11
10
9
9
5
5
3
3
3
2
2
2
2
2
2
2
2
2
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
1
9-2
-------
10.0 SUMMARY OF UNCERTAINTIES
Table 10.1 summarizes all significant values used in this report
and the estimated range or precision for each. As used in the table,
N.B. means No Basis for Evaluation.
10-1
-------
Table 10.1 Summary of Uncertainties
Section
2.0
2.0
2.1.2,
2.1.3.1
2.1.2
2.1.2
2.1.2
2.1.3.1
2.1.3.1
2.1.3.2
2.1.3.2
Description of Value
1978 direct benzene production from petroleum
1978 direct benzene production from coal
Total 1975-1979 capacity for benzene production
from petroleum
1975-1979 capacity for benzene production from
petroleum for a given plant
Total 1975-1979 benzene production from petroleum
1975-1979 benzene production from petroleum for
a given plant
Total 1978 capacity for benzene production from
petroleum through catalytic reformation
Total 1978 production of benzene from petroleum
through catalytic reformation
Benzene emission factor due to production
through catalytic reformation
Benzene emissions due to production through
catalytic reformation
Value
4,780,000 kkg
178,000 kkg
various
various
See Figure 2.1
See Figure 2.1
3,500,000 kkg
2,360,000 kkg
0.01 kkg/kkg
20,000 kkg
Estimated Range
or Precision
+ 2%
+ 2%
+ 20%
+ 20%
+ 10%
+ 30%
+ 20%
+ 30%
N.B.
N.B.
o
I
Isi
-------
Table 10.1 Summary of Uncertainties (continued)
Section
2.1.4.1
2.1.4.1
2.1.5.1
2.1.5.1
2.1.6.1
2.1.6.1
2.1.7
2.1.7
2.1.7
Description of Value
Total 1978 capacity for benzene production from
petroleum through toluene dealkylation
Total 1978 production of benzene from petroleum
through toluene dealkylation
Total 1978 capacity for benzene production from
petroleum through toluene disproportionation
Total 1978 production of benzene from petroleum
through toluene disproportionation
Total 1978 capacity for benzene production from
petroleum from pyrolysis gasoline
Total 1978 production of benzene from petroleum
from pyrolysis gasoline
Benzene emission factor to air due to benzene
production from petroleum (all benzene
production methods that use petroleum)
Benzene emissions to air due to benzene
production from petroleum (all benzene
production methods that use petroleum)
Benzene emission factor to water due to benzene
production from petroleum (all benzene methods
that use petroleum)
Value
1,920,000 kkg
1,300,000 kkg
180,000 kkg
121,000 kkg
1,370,000
925,000 kkg
1.8 x 10~5 kkg/kkg
(See Table 2.1)
85 kkg
1.3 x 10~A kkg/kkg
(See Table 2.1)
Estimated Range
or Precision
+ 20%
+ 30%
+ 20%
+ 30%
+ 20%
+ 30%
low
reliability
low
reliability
+ a factor
~~of 1,000
o
I
UJ
-------
Table 10.1 Summary of Uncertainties (continued)
Section
2.1.7
2.1.8.1
2.1.8.1
2.1.8.2
2.1.8.2
2.1.8.2
2.1.8.3
2.1.8.3
2.1.8.4
2.2.1
2.2.1
Description of Value
Benzene emissions to water due to benzene
production from petroleum (all benzene
production methods that use petroleum)
Benzene emission factors to air due to benzene
storage
Benzene emissions to air due to benzene storage
Benzene production that was not captively used
Benzene emission factors to air due to benzene
loading
Benzene emissions to air due to benzene loading
Benzene emission factors to air due to benzene
transport
Benzene emissions to air due to transport
Total benzene emissions to air due to transpor-
tation, loading, and storage of benzene
Total 1975-1979 capacity for benzene production
from coal
1975-1979 capacity for benzene production from
coal for a given plant
Value
610 kkg
See Table 2.2
105-4,900 kkg
2,560,000 kkg
See Table 2.3
various
See Table 2.4
various
2,400-7,200 kkg
See Figure 2.2
See Figure 2.2
Estimated Range
.or Precision
+ a factor
of 1,000
N.B.
N.B.
+ 30%
N.B.
N.B.
N.B.
N.B.
N.B.
+ 20%
+ 20%
o
I
-p-
-------
Table 10.1 Summary of Uncertainties (continued)
Section
Description of Value
.Value
Estimated Range
or Precision
2.2.1
2.2.1
3.1
3.1
3.1
3.1
3.1
3.2
3.2
Total 1975-1979 benzene from coal
1975-1979 benzene production from coal for a
given plant
Benzene emission factors to air due to indirect
production of benzene from petroleum refineries
Benzene emissions to air due to indirect produc-
tion of benzene from petroleum refineries
operating at 1977 capacity
Benzene emission factor to water due to indirect
production of benzene from petroleum refineries
Total 1978 benzene emissions to water due to
indirect production of benzene from petroleum
1978 benzene emissions by state to water due to
indirect production of benzene from petroleum
refineries
Annual coal consumption capacity of coke
producing plants
Walker's (1976) benzene emission factor to air
due to coke oven operations
See Figure 2.2
See Figure 2.2
See Table 3.1
20,000 kkg
e64 x 10
-10
1 kkg
See Figure 3.1
88,000,000 kkg
+ 10%
+ 30%
N.B.
N.B.
+ a factor of 10
+ a factor of 10
+_ a factor of 10
+ 20%
9.80 x 10
-4 kkg benzei
ie
kkg coke
produced
a factor
of 6
-------
Table 10.1 Summary of Uncertainties (continued)
Section
3.2
3.2
3.2
3.2
3.2
3.2
3.3
3.3
3.3
Description of Value
Total benzene emissions, based on Walker's (1976)
emission factor, to air due to coke oven opera-
tions
PEDCo's (.1977) benzene emission factor to air
due to coke oven operations
Total benzene emissions based on PEDCo's (1977)
emission factor due to coke oven operations
Mara and Lee's (1978) benzene emission factor
to air due to coke oven operations
Total benzene emissions based on Mara and Lee,'s
(1978) emission factor due to coke oven operations
Benzene emissions by state based on Mara and
Lee's (1978) emission factor due to coke oven
operations
Annual oil discharge to oceans according to
Walker (1976)
Benzene content of oils
Annual benzene discharge to oceans according to
Walker (1976)
Value
59,200 kkg
7 o „ in-5kkg benzen
/ . O X J.U , , T~
kkg coal
used
6,600 kkg
- in-5kkg benzene
kkg coal
used
3,000 kkg
See Figure 3.2
11-12 x 109 Ibs/year
0.001 - 0.4%
10-11 x 10 3 kkg
Estimated Range
or Precision
+ a factor of 6
k
'- + a factor
of 10
+ a factor of 10
N.B.
N.B.
N.B.
N.B.
+ a factor of 400
+ a factor
of 1400
o
I
-------
Table 10.1 Summary of Uncertainties (continued)
Section
3.3
3.3
3.4
4.1
4.2
4.2
4.2
4.2.1
4.2.2
5.1
Description of Value
Annual oil discharge to U.S. waters according
to Versar (1977)
Annual benzene discharge to U.S. waters
according to Versar (1977)
Gross annual discharge of benzene to water from
various indirect sources
1975-1979 benzene imports
Benzene emission factor for marine loading
during importation
Benzene emission factor for transport of
imported benzene
Total 1978 benzene emissions due to benzene
imports
1978 benzene emissions to air due to benzene
imports
1978 benzene emissions to water due to benzene
imports
Total 1978 consumptive use of benzene
Value
4,990,691 gal
30 kkg
various
See Table 4.1
- n TT in~4kkg benzen
i. , U X J.U , , n ,
kkg unload
i -r in~4kkS benzene
kkg trans-
ported/wk
25 kkg
13 kkg
13 kkg
5,389,000 kkg
Estimated Range
or Precision
N.B.
+ a factor
of >400
N.B.
+ 20%
'-. + a factor
:d ~ of 10
+ a factor
of 10
+ a factor of 10
+ a factor of 15
+ a factor of 15
+ 20%
-------
Table 10.1 Summary of Uncertainties (continued)
Section
5.2
5.2
5
5
5
5.3.8.2
5.3.9.3
5.3.1.3.1
5.3.1.3.2
5.3.1.3.2
Description of Value
1975-1979 consumption of benzene for derivative
synthesis
1978 production of each benzene derivative
Capacities for production of each benzene
derivative
Production of each benzene derivative by a
given plant
Consumption of benzene for the synthesis of
each benzene derivative (except anthraquinone
and biphenyl) by a given plant
Benzene consumption for anthraquinone synthesis
Benzene consumption for biphenyl synthesis
Benzene emission factors to air and corresponding
1978 emissions due to ethylbenzene synthesis
Benzene emission factor to water due to
ethylbenzene synthesis
1978 benzene emissions to water due to
ethylbenzene synthesis
Value
See Table 5.2
See Table 5.1
various
various
various
23,000 kkg
11,000 kkg
See Table 5.3
1.9 x 10~4 kkg/kkg
720 kkg
Estimated Range
or Precision
+ 20%
+ 10%
+ 20%
+ 30%
+ 40%.
N.B.
+ 50%
N.B.
+ a factor of 5
+ a factor of 10
-------
Table 10.1 Summary of Uncertainties (continued)
Section
5.3.2.3
5.3.2.3
5.3.3.3
5.3.3.3.1
5.3.3.3.2
5.3.4.3.1
5.3.4.3.1
5.3.4.3.2
5.3.5.3.1
5.3.6.2.1
Description of Value
Benzene emission factor due to cumene synthesis
1978 benzene emissions to air due to cumene
synthes"is
Benzene emission factor due to cyclohexane
synthesis
Benzene emissions to air due to cyclohexane
synthesis
Benzene emissions to water due to cyclohexane
synthesis
Benzene emission factor to air due to maleic
anhydride synthesis
1978 benzene emissions to air due to maleic
anhydride synthesis
1976 benzene emissions to water due to maleic
anhydride synthesis
Benzene emission factors and corresponding 1978
benzene emissions to air due to nitrobenzene
synthesis
Benzene emission factors to air due to ....
chlorobenzene synthesis
Value
0.25 x 10"3 kkg/kkg
380 kkg
2.8 x 10~3 kkg/kkg
3,000 kkg
30 kkg
0.20 kkg/kkg
34,000 kkg
8 kkg
See Table 5.3
See Table 5.6
Estimated Range
or Precision
N.B.
N.B.
+ a factor of 10
+ a factor of 10
+ a factor of 10
+10%, -90%
+10%, -90%
N.B.
N.B.
N.B.
o
-------
Table 10.1 Summary of Uncertainties (continued)
Section
Description of Value
.Value
Estimated Range
.or Precision
o
5.3.6.2.1
5.3.7.3
5.3.7.3.1
5.3.9.3
5.3.9.3.1
5.3.9.3.2
6.1
6.3.1.2
6.3.1.2
1978, 1979 benzene emissions to air due to
chlorobenzenes synthesis
Benzene' emission factor to air due to
alkylbenzene synthesis
1978 benzene emissions to air due to
alkylbenzene synthesis
Benzene emission factor due to biphenyl
synthesis
1978 benzene emissions to air due to
biphenyl synthesis
1978 benzene emissions to water due to
biphenyl synthesis
Total 1978 nonconsumptive use of benzene
1978 solvent nonconsumptive use of benzene
according to Neufeld et_ al. (1978)
1978 solvent nonconsumptive use of benzene
according to Mara and Lee (1978)
See Table 5.7
5 x 1(T4 kkg/kkg
170 kkg
5.9 x 10~3 kkg
benzene emitted pel
kkg benzene con-
sumed
64 kkg
0.3 kkg
270,000 kkg
9,500 kkg
27,000 kkg
N.B.
N.B.
N.B.
+ a factor of 2
+_ a factor of 3
+ a factor of 3
+ 60%
+10%, -50%
+ 80%
-------
Table 10.1 Summary of Uncertainties (continued)
Section
6.3.1.3
6.3.1.3.1
6.3.1.3.2
7.1.1
7.1.2
7.1.2
7.1.3.2
7.1.3.2
7.1.3.2
7.1.3.2
7.1.3.2
Description of Value
1978 benzene emissions due to solvent non-
consumptive use of benzene
1978 benzene emissions to air due to solvent
nonconsumptive use of benzene
1978 benzene emissions to water due to solvent
nonconsumptive use of benzene
Volume percent benzene in gasoline, 1979
1979 Annual Domestic Gasoline Consumption (ADGC)
Amount of benzene in gasoline, 1979
Capacity of floating-roof storage tanks
Average tank retention time
Average percent of tank filled
Emission factor for hydrocarbon standing
storage losses
Emission factor for hydrocarbon withdrawal
losses
Value
2,900 kkg
1,450 kkg
1,450 kkg
2% (V/v)
1.08 x 1011 gal
2 x lOJ? gal;
7 x 10 kkg
2.3 x 106 gal
30 days
75%
132 Ib/day/tank
0.025 lb/103 gal
Estimated Range
or Precision
N.B.
N.B.
N.B.
+ 50%
+ 10%
+ 60%
N.B.
N.B.
N.B.
N.B.
N.B.
o
-------
Table 10.1 Summary of Uncertainties (continued)
Section
Description of Value
Value
Estimated Range
or Precision
o
7.1.3.2
7.1.3.2
7.1.3.2
7.1.3.2
7.1.3.2
7.1.3.2
7.1.3.2
7.1.3.2
7.1.3.2
7.1.3.2
Conversion factor, benzene liquid volume to
benzene vapor weight
Emissions of benzene to air during standing
storage, 1979
Emissions of benzene to air during withdrawal,
1979
Daily volume of gasoline handled by bulk
plants
Number of bulk plants, 1979
Annual volume of gasoline handled by bulk
plants
Emissions of benzene to air due to tank
breathing, 1979
Percent above-ground fixed-roof canks at bulk
plants
Annual volume stored in above-ground fixed-roof
tanks
Emissions of benzene to air from filling above-
ground fixed-roof tanks
0.45
1,000 kkg
10 kkg
4,000 gal
20,000
3 x 1010 gal
500 kkg
70%
2 x 1010 gal
1,600 kkg
+ 5%
N.B.
N.B.
+ 25 %
+ 20%
+ 50 %
N.B.
+ 10%
+ 60%
+ a factor of 4
-------
Table 10.1 Summary of Uncertainties (continued)
Section
7.1.3.2
7.1.3.2
7.1.3.3
7.1.3.3
7.1.3.3
7.1.3.3
7.1.3.3
7.1.3.3
7.1.3.4
Description of Value
Emissions of benzene from filling underground
tanks
Emissions of benzene due to all gasoline
storage
Emission factor for hydrocarbon losses during
tank car/truck splash loading
Emission factor for hydrocarbon losses during
tank car/truck submerged loading
Emission factor for hydrocarbon losses during
marine loading
Emission factor for hydrocarbon losses during
transit
Emission factor for hydrocarbon losses during
unloading
Emissions of benzene to air due to transport
of gasoline
Hydrocarbon emission rate, filling storage
tank - submerged
.Value
1,000 kkg
4,000 xkg
12 lb/103 gai
5 lb/103 gal
1.8 lb/103 gal
3 lb/wk/103 gal
2.1 lb/103 gal
3,000 kkg
7.3 lb/103 gal
Estimated Range
or Precision
+ a factor of 4
minimum: does not
consider all
possible losses
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
o
I
-------
Table 10.1 Summary of Uncertainties (continued)
Section
7.1.3.4
7.1.3.4
7.1.3.4
7.1.3.4
7.1.3.4
7.1.3.4
7.1.3.4
7.1.3.4
7.1.3.4
7.1.3.4
Description of Value
Frequency of use: filling storage
tank - submerged
Hydrocarbon emission rate, filling storage
tank - splash
Frequency of use: filling storage
tank - splash
Hydrocarbon emission rate, filling storage
tank - balanced
Frequency of use: filling storage
tank - balanced
Hydrocarbon emission rate, underground task
breathing
Frequency of use: underground task breathing
Hydrocarbon emission rate, uncontrolled vehicle
refueling
Frequency of use: uncontrolled vehicle
refueling
Hydrocarbon emission rate, controlled
vehicle refueling
Value
50%
11.5 lb/103 gal
25%
0.3 lb/103 gal
25%
1 lb/103 gal
100%
9.0 lb/103 gal
75%
0.9 lb/103 gal
Estimated Range
or Precision
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
o
I
-------
Table 10.1 Summary of Uncertainties (continued)
Section
7.1.3.4
7.1.3.4
7.1.3.4
7.1.3.4
7.1.3.4
7.1.3.4
7.1.3.4
7.1.3.4
7.1.3.4
7.1.3.4
7.1.3.4
7.1.3.4
Description of Value
Frequency of use: controlled vehicle refueling
Hydrocarbon emission rate, spillage
Frequency of use, spillage
Benzene emissions, filling storage
tanks - submerged
Benzene emissions, filling storage
tanks - splash
Benzene emissions, filling storage
tanks - balanced
Total filling emissions
Benzene emissions, underground task breathing
Benzene emissions, uncontrolled vehicle
refueling losses
Benzene emissions, controlled vehicle
refueling losses.
Total vehicle refueling benzene emissions
Spillage benzene emissions
Value
25%
0.7 lb/103 gal
100%
2,000 kkg
1,000 kkg
30 kkg
3,000 kkg
400 kkg
3,000 kkg
100 kkg
3,000 kkg
700 kkg
Estimated Range
or Precision
N.B.
N.B.
No range
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
o
-------
Table 10.1 Summary of Uncertainties (continued)
Section
Description of Value
Value
Estimated Range
or Precision
7.1.3.4
7.1.3.5
7.1.3.5
7.1.3.5
7.1.3.5
7.1.3.5
7.1.3.5
7.1.3.5
7.1.3.5
7.1.3.5
7.1.3.5
7.1.3.5
7.1.3.5
Total benzene emissions, gasoline
vending operations
Estimate of vehicles in U.S.
Benzene emission factor for evaporative
emissions
Average number of trips per vehicle per day
Benzene emissions due to evaporation
Benzene tailpipe emission factors - non-
catalytic
Benzene tailpipe emission factors - catalytic
Average number of miles driven per car
Percent of vehicles with exhaust controls
Benzene exhaust emissions, controlled
Benzene exhaust emissions, uncontrolled
Total benzene exhaust emissions
Total benzene emissions, auto use
7,000 kkg
96 million
0.148 g/trip
3.3 trips/day
1'7,000 kkg
0.05 - 0.15 g/mile
0.005 - 0.020 g/mile
9,494 miles
73%
3,000 - 10,000 kkg
10,000 - 40,000 kkg
10,000 - 50,000 kkg
30,000 - 70,000 kkg
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
N.B.
-------
Table 10.1 Summary of Uncertainties (continued)
Section
Description of Value
Value
Estimated Range
or Precision
7.1.3.6
Sum of benzene emissions due to benzene in
gasoline
40,000 - 80,000 kkg
N.B.
o
-------
11.0 DATA GAPS AND RECOMMENDATIONS
In the course of performing this Level I materials balance, the
following significant data gaps were encountered:
11.1 EMISSIONS DUE TO BENZENE PRODUCTION BY COKE OVEN PLANTS
Light oil containing benzene is produced as a byproduct of coal
coking. In Section 3.2, a question arose concerning the fate of
472,000 kkg of benzene-containing light oil produced by coke ovens.
This amount was the difference between the light oil production capa-
cities of non-benzene-producing plants (798,000 kkg) and the amount
of light oil accounted for by sales to other plants for benzene
extraction (326,000 kkg). A more detailed literature search and
inquiries to the appropriate companies should be made to determine
what fraction of this unaccounted-for light oil is a source of benzene
emissions.
This problem is one aspect of the larger question of benzene
emissions due to coal coking. The readily available literature did
not distinguish between benzene-producing and non-benzene-producing
coking operations. A Level II survey of primary literature and, if
necessary, monitoring to determine relative emissions for the two
types of coking plants should be performed.
11.2 BREAKDOWN OF EMISSIONS DUE TO PETROLEUM REFINING BY METHOD
No data were available to distinguish., for instance, between
emissions due to catalytic reformation and those due to toluene
dealkylation. The emission factors for these processes may be quite
different. A more detailed literature search would reveal whether or
not monitoring would be necessary. If monitoring were recommended,
it might be satisfactory to determine an emission factor for one pro-
cess plus relative emission concentrations for the other three. This
would permit estimation of emission factors for the latter three.
11-1
-------
11.3 BENZENE EMISSIONS TO WATER
In the absence of data, it is generally assumed (as it was in
this report) that the majority of benzene emissions were to the air.
A closer search for wastewater monitoring data from benzene producing
and consuming plants (including coke ovens) should be performed. If
no data exist, waste stream volumes should be ascertained from indus-
try and maximum benzene emissions should be calculated from the
solubility limit (0.7 g/1). This would give an estimate of the rela-
tive importance of water emissions. Monitoring resources could be
allocated accordingly.
11.4 TREATMENT OF SOLID RESIDUES
Data on production and treatment of benzene-containing solid
residues (tars, gums, sludges) were unavailable for the Level I study.
Industry sources should be interviewed for information on both refin-
ery solids and solid residues formed during consumptive use. Informa-
tion on rates of solid generation, compositions of residues, and
disposal methods and efficiencies should be sought.
11.5 EMISSIONS DUE TO MINOR CONSUMPTIVE USES
Neither process information nor emission factors were available
for the minor consumptive uses of benzene: anthraquinone and biphenyl.
These data are probably available in the literature (.perhaps dye
industry or PCB information), and it should be obtained in order to
form a more complete materials balance.
11-2
-------
REFERENCES
Dunavent, S.W., Gee, D. and Talbert, W. M. 1978. Evaluation of Control
Technology for Benzene Transfer Operations. EPA-450/3-78-018, U.S.
Environmental Protection Agency, Research Triangle Park, NC.
•
Fentiman, A. F., Neher, M. B., Kinzer, G. W., Sticksel, P. R., Coutant,
R. W., Jungclaus, G. A., Edie, N. A., McNulty, J., and Townley, C. W.
1979. Environmental Monitoring Benzene. EPA-560/6-79-006, U.S.
Environmental Protection Agency, Research Triangle Park, NC.
Gray, Charles L. 1979. U.S. Environmental Protection Agency internal
memorandum to David Hawkins re benzene tailpipe emissions. January 7,
1979.
Howard, P. H. and Durkin, P. R. 1974. Benzene: Environmental Sources
of Contamination, Ambient Levels, and Fate. EPA 560/5-75-005 for U.S.
Environmental Protection Agency, Washington, D.C.
Howard, P. H., Santodonato, J., Saxena, J., Mailing, J. E., and Greninger,
D. 1976. Investigation of Selected Potential Environmental Contami-
nants, Nitroaromatics. EPA 560/2-76-010, U.S. Environmental Protection
Agency, Washington, D.C.
JRB Associates, Inc. 1980. Materials Balance: 1,2-Dichlbroethane;
Level I - Preliminary. EPA-560/13-80-002. U.S. Environmental Protec-
tion Agency, Washington, D.C.
Kirk-Othmer. 1976. Encyclopedia of Chemical Technology, 3rd edition,
Vol. 3, "Benzene", Interscience Publishers, NY, pp. 744-771.
Little, A. D., Inc. 1977. Technology Assessment and Economic Impact
Study of an OSHA Regulation for Benzene, Volume I, prepared for
Occupational Safety and Health Administration, Washington, D.C.
Little, A. D., Inc. 1977. Economic Impact Statement, Benzene, Vol. II,
prepared for U.S. Department of Labor, Occupational Safety and Health
Administration, Washington, D.C.
Mara, S. J. and Lee, S. S. 1978. Assessment of Human Exposure to
Atmospheric Benzene. EPA-450/3-78-031, U.S. Environmental Protection
Agency, Research Triangle Park, NC.
Meylan, W. M. and Howard, P. H. 1976. Chemical Market Input/Output
Analysis of Selected Chemical Substances to Assess Sources of
Environmental Contamination: Task II Biphenyl and Diphenyl Oxide.
EPA 560/6-77-003, U.S. Environmental Protection Agency, Washington, D.C.
-------
References (Continued)
Neufeld, L. M., Sittenfield, M., Henry, R., and Hunsicker, S. 1978.
Market Input/Output Studies Task Benzene Consumption as a Solvent.
EPA-560/6-77-034, U.S. Environmental Protection Agency, Washington, D.C.
Patterson, R. M. , Bornstein, M. I., and Garshick, E. 1976. Assessment
of Benzene as a Potential Air Pollution Problem. EPA Contract No. 68-
02-1337, U.S. Environmental Protection Agency, Research Triangle Park,
NC.
PEDCo Environmental, Inc. 1977. Atmospheric Benzene Emissions. EPA-
450/3-77-029, U.S. Environmental Protection Agency, Research Triangle
Park, NC.
Perry, D. L., Chuang, C. C., Jungclaus, G. A., and Warner, J. S. 1978.
Identification of Organic Compounds in Industrial Effluent Discharges.
EPA 560-78-009, U.S. Environmental Protection Agency, Athens, Georgia.
Peterson, C. A. 1979. Emissions Control for the Synthetic Organic
Chemicals Manufacturing Industry. EPA Contract No. 68-02-2577,
Cumene Product Report, U.S. Environmental Protection Agency, Research
Triangle Park, NC.
Process Research, Inc. 1972. Air Pollution from Nitration Processes.
U.S. Environmental Protection Agency, Washinton, D.C.
S.R.I. 1977. The Chemical Economics Handbook. "Benzene". Stanford
Research Institute, Menlo Park, CA.
Stecher, P. G., ed. 1968. The Merck Index, 8th ed., Merck & Co., Inc.,
Rahway, NJ.
U.S. International Trade Commission. 1974-1979. Synthetic Organic
Chemicals, U.S. Production and Sales.
Versar, Inc. 1977. Gross Annual Discharge to the Waters in 1976:
Benzene. EPA Contract No. 68-01-3852, U.S. Environmental Protection
Agency, Washington, D.C.
Versar, Inc. 1979. Production and Use of Benzene. U.S. Environmental
Protection Agency, Washington, D.C.
Walker, P. 1976. Air Pollution Assessment of Benzene. EPA Contract
No. 68-02-1495, U.S. Environmental Protection Agency, Washington, D.C.
Weast, R. C., ed.1972. Handbook of Chemistry and Physics, 53rd ed. The
Chemical Rubber Co., Cleveland, OH.
Weissermel, K. and Arpe, H., trans by A. Mullen. 1978. Industrial
Organic Chemistry, Chapter 12: "Aromatics Production and Conversion"
and Chapter 13: "Benzene Derivatives". Verlag Chemie, NY.
-------
References (Continued)
PERSONAL COMMUNICATIONS
American Petroleum Institute (API): telephone communication between
Donna McMillian and Phillip Spooner, J.R.B. Associates, January 3, 1980.
U.S. Bureau of the Census, Department of Commerce: personal communication
to R. L. Hall, J.R.B., December,1979.
National Petroleum Refiners Association: personal communication between
Linda Dziuban, Washington, D.C. and R. L. Hall, J.R.B., December 1979.
-------
APPENDIX A
ENVIRONMENTAL FLOW DIAGRAM FOR BENZENE
-------
APPENDIX A BEIZE'lE EWIWWEKTAL FLOW DIAGW1
-------
APPENDIX B
PROCESS FLOW DIAGRAMS FOR PRODUCTION
OF BENZENE FROM PETROLEUM
-------
APPENDIX B
Reactors
.atalyst ',
i
Catalyst
Regeneratoi
'
Separator gas
(CH, and other)
desulf.erized
naptha
aromatic-rich
reformate
Figure B-l Platforming Method of Catalytic Reformation (Kirk-Othmer, 1976)
Desulfurized naphtha is mixed with recycled hydrogen, heated, then
fed through a series of moving catalyst bed reactors with intermediate
heating. The platinum chloride-rhenium chloride catalyst is regenerated
by controlled combustion coke removal, and is recycled. The reformate
from the reactors passes through the separator where hydrogen and other
gases are removed. The aromatic-rich reformate then goes to a stabilizer.
Aromatics are extracted from the stabilized reformate.
B-l
-------
APPENDIX B
An example of the separation process is the Sulfolane process which
is presented schematically in Figure B-2 on the next page. An aliphatic-
aromatic stream is charged to the extractor through which the solvent is
flowing in a counter-current. The solvent will dissolve the hydrocarbons
and carry them to the extractive stripper where most of the hydrocarbons
are separated from the solvent. The hydrocarbons are recycled through.
this system again, then are issued to the extract recovery column where
additional solvent contamination is removed. The resulting BTX stream
is then fractionated into individual products.
B-2
-------
extractor
t»
OJ
charge
(aliphatic - aromatic
stream)
raffinate
recycled hydrocarbons
solvent
Figure B-2 Separation of Benzene from Catalytic Reformate by the Sulfolane Process (Kirk-Othmer, 1976) £
-------
APPENDIX B
Additional
H
_H
Reactors
Catalytic
Dispropor-
t ionation
tbble-meta:
or rare
earth
catalyst
Heating
Catalyst
Regenerat ion
NX
Heavy
Ends
Recycled Toluene
(and/or C)
Figure B-3 Toluene Disproportionation by the Tatoray Process
(adaptation from Kirk-Othmer, 1976)
B-4
-------
distillation
Pyrolysis
Gasoline
i
Ln
cut
Evaporator
"*-
T
remainder
Pretr-eat
Reactor
hydrogenat ion
with catalyst
heater
Unconverted
Alkyl Aromatics
Polymers
Gums
Pyrotol Reactor
•Hydrodealkylation
of Toluene and other
Alkyl Benzenes
•Desulfurization and
hydrocracking of
n on —a romaticj
Clay Treatment
$tabil-J
izer *—4
light
gases
Contaminants
Benzene
distillation
Figure B-4 Recovery of Benzene from Pyrolysis Gasoline by the Pyrotol Process (Kirk-Othmer, 1976)
§
i
o
n
-------
DRIPOUENE
i
CTv
1ST REACTOR
2ND REACTOR
CATALYTIC
HYDROGENATION
(NICKEL OR
PALLADIUM
CATALYST)
HYDRODESULFUR -
IZATION
(CO-MO CATALYST)
AROMATICS
>AVAILABLE
FOR SOL -
VENT EX r.
TRACTION
I
U
>
CATALYST
REGENERATION
T
GUM REMOVAL
COKE BURNOFF
Figure B-5 Recovery of Benzene from Pyrolysis Gasoline (Dripolene) by the IFF
(Institute Francaise de Petrole) Process: After Kirk-Othmer, 1976
§
O
M
X
O3
-------
TECHNICAL REPORT DATA
(Please read Instructions on the reverse before completing)
1. REPORT NO.
EPA-560/13-80-014
2.
3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
Level I Materials Balance: Benzene
5. REPORT DATE
May 9, 1980
6. PERFORMING ORGANIZATION CODE
2-800-03-379-51
7. AUTHOR(S)
Robert L. Hall, Carlos Buitrago, Frank Montecalvo,
Tom Yatsko, Karen Slimak
8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
10. PROGRAM ELEMENT NO.
JRB Associates, Inc.
8400 Westpark Drive
McLean, Virginia 22102
11. CONTRACT/GRANT NO.
68-01-5793
12. SPONSORING AGENCY NAME AND ADDRESS
Survey and Analysis Division (TS-793)
Office of Pesticides and Toxic Substances
U.S. Environmental Protection Agency
Washington, DC 20460
13. TYPE OF REPORT AND PERIOD COVERED
Final Report
14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
Project Officer - Michael Callahan
16. ABSTRACT
A Level I materials balance was performed on benzene. Data are reported for benzene
production from petroleum by four processes (catalytic reformation, toluene dealkyla-
tion, toluene disproportionation, and isolation from.pyrolysis gasoline) and for pro-
duction from coal during coking. Amounts of benzene consumed for the synthesis of nine
direct derivatives (ethylbenzene, cumene, cylcohexane, nitrobenzene, maleic anhydride,
mono- and dichlorobenzenes, alkylbenzenes, anthraquinone, and biphenyl) and exports are
presented. These .uses constitute approximately 93 percent of total benzene usage. Non-
consumptive uses (solvents and inventory changes) are also tabulated. Emissions due to
each of the above processes are reported or estimated where possible. In addition,
emissions due to indirect production (refinery operation, coke oven operations, oil
spills, non-ferrous metals manufacturing, ore mining, wood processing, coal mining, and
two phases of the textile industry) are presented. Production of benzene as a compo-
nent of gasoline and emissions due to gasoline use are estimated. Locations of sites
with high densities of benzene producers and users are tabulated: the major "hotspots"
are Houston/Galv.eston, Texas; Corpus Christi, Texas; Beaumont/Port Arthur, Texas; and
Puerto Rico. The uncertainty ranges of all numbers used or derived in this report are
evaluated and general recommendations are presented. The results of the report are
summarized in two figures: the Environmental Flow Diagram for benzene in Appendix A,
and the Materials Balance Diagram in the Executive Summary.
17.
KEY WORDS AND DOCUMENT ANALYSIS
DESCRIPTORS
b.lDENTIFIERS/OPEN ENDED TERMS C. COSATI Field/Group
18. DISTRIBUTION STATEMENT
Release Unlimited
19. SECURITY CLASS (ThisReport)
Unclassified
21. NO. OF PAGES
155
20. SECURITY CLASS (Thispage)
Unclassified
22. PRICE
EPA Form 2220-1 (9-73)
------- |