-------
                                 CO
                                 .c
                                 *••
                                 .c
                                 Q.
                                 CD
                                 E
                                 o
                                 .2 3
                                 ~ S
                                 o £
                                 a c
                                 ^ 3
                                 o "
                                 
                                 L.
31

-------
          TABLE 4.  STREAM DESIGNATIONS FOR FIGURE 6, PRODUCTION
                    OF BENZENE FROM PYROLYSIS GASOLINE
Stream
Number                                 Stream Description
  1                          Naphtha or gas oil feed
  2                          Fuel gas and oil
  3                          Ethane/propane recycle stream
  4                          Cracked gas
  5                          Cracked gas
  6                          Gasoline fractionator
  7                          Furnace exhaust
  8                          Slurry of collected particles
  9                          Quenched cracked gas
 10                          Surplus fuel oil
 11                          Light fractions
 12                          Overheads from gasoline fractionator
 13                          Condensed organic phase
 14                          Raw pyrolysis gasoline intermediate storage
 15                          Quench tower
 16                          Heat exchangers
 17                          Surplus water
 18                          Recycle steam generator blowdown
 19                          Overheads from quench tower
 20                          Water
 21                          Organic fractions.
 22                          Acid gas
 23                          Diethanolamine (DEA)
 24                          Liquid waste stream
 25                          Liquid waste stream
 26                          Process gas stream
 27                          Solid waste stream
                                     32

-------
          TABLE 4.  STREAM DESIGNATIONS FOR FIGURE 6,  PRODUCTION
                    OF BENZENE FROM PYROLYSIS GASOLINE (Continued)
Stream
Number                                 Stream Description
 28                          Process gas
 29                          Hydrogen rich stream from demethanizer
 30                          Methane rich stream from demethanizer
 31                          €„ components from de-ethanizer
 32                          C, and heavier components from de-ethanizer
 33                          Acetylene converter
 34                          Overheads from ethylene fractionator
 35                          Ethane recycle pyrolysis furnace
 36                          Overheads from depropanizer
 37                          Propylene (purified)
 38                          Propane recycle pyrolysis furnace
 39                          C. and heavier components to debutanizer
 40                          Overheads from debutanizer
 41                          Cf. and heavier components from debutanizer
 42                          Combined C,. components and gasoline stripper
                             bottoms fraction
 43                          Light ends to cracked gas compressor
 44                          Cj. and heavier components
 45                          Superheated stream
 46                          Stream and hydrocarbons
 47                          Compressor
 48                          Refrigeration compressor
 49                          Refrigeration compressor
                                      33

-------
operation, coke accumulates on the inside walls of the reactor coils, and
each furnace must be periodically taken out of service for removal of the
accumulated coke.  Normally, one furnace is out of service for decoking at
all times.  Decoking is accomplished by passing steam and air through the
coil while the furnace is maintained at an elevated temperature, effectively
burning the carbon out of the coil.  While a furnace is being decoked, the
exhaust is diverted (Stream 7) to an emission control device (Vent B) whose
main function is to reduce particulate emissions.  The collected particles
are removed as a slurry (Stream 8).  The cracked gas (Stream 4) leaving the
pyrolysis furnaces is rapidly cooled (quenched) to 250-300 C by passing it
through transfer-line exchangers, which end pyrolysis and simultaneously
generate steam.  The streams from the transfer-line exchangers (Stream 5)
are combined and further quenched by the injection of recycled pyrolysis
fuel oil from the gasoline fractionator (Stream 6).

     The quenched cracked gas (Stream 9) passes to the, gasoline
fractionator, where pyrolysis fuel oil is separated.  Most of the fuel oil
passes through water-cooled heat exchangers and is recycled (Stream 6) to
the preceding oil-quenching operation.  The surplus fuel oil (Stream 10),
equivalent to the quantity initially present in the cracked gas, passes
first to the fuel oil stripper, where light fractions are removed, and then
to fuel oil storage.  The light fractions (Stream 11) removed in the fuel
oil stripper are recycled to the gasoline fractionator.  The gasoline
fractionator temperatures are well above the vaporization temperature of
water, and the contained water remains as superheated steam, with the
overhead stream containing the lighter cracked-gas components.

     The overhead stream from the gasoline fractionator (Stream 12) passes
to the quench tower, where the temperature is further reduced, condensing
most of the water and part of the C_ and heavier compounds.  The condensed
organic phase (Stream 13) is stripped of the lighter components in the
gasoline stripper and is passed to raw pyrolysis gasoline intermediate
storage (Stream 14).  Most of the water phase, which is saturated with
organics, is separated in the quench tower (Stream 15), passed through
                                      34

-------
water-cooled heat exchangers (Stream 16),  and then recycled to the quench
tower to provide the necessary cooling.  The surplus water (Stream 17),
approximately equivalent to the quantity of steam injected with the
pyrolysis furnace feed, passes to the dilution steam generator, where it is
vaporized and recycled as steam to the pyrolysis furnaces.  Slowdown from
the recycle steam generator is removed as a wastewater stream (Stream 18).

     The overhead stream from the quench tower (Stream 19) passes to a
centrifugal charge-gas compressor (first three stages), where it is
compressed.  Water (Stream 20) and organic fractions (Stream 21) condensed
during compression and cooling are recycled to the quench tower and gasoline
stripper.

     Lubricating oil (seal oil) discharged from the charge-gas compressor is
stripped of volatile organics in a separator pot before the oil is
recirculated.  The organic vapor i:s vented tp the atmosphere (Vent G) .
Similar separator pots separate volatile organics from lubricating oil from
both the ethylene and propylene refrigeration compressors (Streams 48 and
49).

     Following compression, acid gas (H9S, C0_) is removed by absorption in
diethanolamine (DEA) or other similar solvents in the amine wash tower
followed by a caustic wash step.  The amine stripper strips the acid gas
(Stream 22) from the saturated DEA and the DEA (Stream 23) is recycled to
the amine wash tower.  Very little blowdown from the DEA cycle is required.

     The waste caustic solution, blowdown from the DEA cycle, and wastewater
from the caustic wash  tower are neutralized, stripped of acid gas, and
removed as liquid waste streams (Streams 24 and 25).  The acid gas stripped
from the DEA and caustic waste  (Stream 22) passes to an emission control
device (Vent D), primarily to control H_S emissions.

     Following acid  gas removal, the remaining process gas stream
(Stream 26)  is further compressed and  is then passed through drying  traps
containing a desiccant, where the water content is  reduced to  the low  level
                                       35

-------
necessary to prevent ice or hydrate formation in the low-temperature
distillation operations.  The drying traps are operated on a cyclic basis,
with periodic regeneration necessary to remove accumulated water from the
desiccant.  The desiccant is regenerated with heated fuel gas and the
effluent gas is routed to the fuel system.  Fouling of the desiccant by
polymer formation requires periodic replacement, which results in the
generation of a solid waste (Stream 27).  However, with a normal desiccant
service life of possibly several years, this waste source is relatively
minor.

     The demethanizer separates a mixture of hydrogen and methane from the
C^ and heavier components of the process gas (Stream 28).  The demethanizer
overhead stream (hydrogen and methane) is further separated into
hydrogen-rich and methane-rich streams (Streams 29 and 30) in the
low-temperature chilling section.  The methane-rich stream is used primarily
for furnace fuel.  Hydrogen is required in the catalytic hydrogenation
operations.

     The de-ethanizer separates the C. components (ethylene, ethane, and
acetylene) (Stream 31) from the C. and heavier components (Stream 32).
Following catalytic hydrogenation of acetylene to ethylene by the acetylene
converter (Stream 33), the ethylene-ethane split is made by the ethylene
fractionator.  The overhead from the ethylene fractionator (Stream 34) is
removed as the purified ethylene product, and the ethane fraction
(Stream 35) is recycled to the ethane/propane cracking furnace.

     The de-ethanizer bottoms (C~ and heavier compounds) (Stream 32) pass to
the depropanizer, where a C--C, split is made.  The depropanizer overhead
stream (primarily propylene and propane)  (Stream 36) passes to a catalytic
hydrogenation reactor (C. converter), where traces of propadiene and methyl
acetylene are hydrogenated.  Following hydrogenation, the C, fraction passes
to the propylene fractionator, where propylene is removed overhead as a
purified product (Stream 37).  The propane (Stream 38) is recycled to the
ethane/propane pyrolysis furnace.
                                      36

-------
     The C, and heavier components (Stream 39) from the depropanizer pass to
the debutanizer, where a C.-C,. split is made.  The overhead C,  stream
(Stream 40) is removed as feed to a separate butadiene process.
     The stream containing C,. and heavier compounds from the debutanizer
(Stream 41) is combined with the bottoms fraction from the gasoline stripper
as raw pyrolysis gasoline.  The combined stream (Stream 42) is hydrogenated
in the gasoline treatment section.  Following the stripping of lights
(Stream 43), which are recycled to the cracked-gas compressor, the C^ and
heavier compounds (Stream 44) are transferred to storage as treated
pyrolysis gasoline.  This stream contains benzene and other aromatics formed
by pyrolysis.

     The three catalytic hydrogenation reactors for acetylene, C_ compounds,
and pyrolysis gasoline all require periodic regeneration of the catalyst to
remove contaminants.  The catalyst is generally regenerated every four to
six months.  At the start of regeneration, as superheated steam (Stream 45)
is passed through a reactor, a mixture of steam and hydrocarbons leaving the
reactor (Stream 46) is passed to the quench tower.  After sufficient time
has elapsed for stripping of organics (approximately 48 hours), the exhaust
is directed to an atmospheric vent (Vent F) and a steam-air mixture is
passed through the catalyst to remove residual carbon.  This operation
continues for an additional 24 to 48 hours.  The presence of air during this
phase of the regeneration prevents the vented vapor from being returned to
the process.

     Because the olefins and di-olefins present in pyrolysis gasoline are
unstable in motor gasoline and interfere with extraction of aromatics, they
                                                  3
are hydrogenated prior to extraction of aromatics.   The actual
hydrogenation process routes used are determined by the final product that
is desired.    For example, Figure 7 shows one method of producing benzene,
toluene, and "xylene.  Pyrolysis gasoline is fed with make-up hydrogen into
the first stage hydrogenation reactor (Stream 1) where olefins are
hydrogenated.  The reaction conditions are mild (40-95 C and
10-40 atmospheres pressure).
                                       37

-------
                                 (0
                                 0)

                                 o
                                _w

                                TJ
                                 0)
                                 o
                                 a.
                                 co
                                 o
                                 o
                                               
 o
                                   co
                                   CO
                                   CD
                                   O
                                 3 CO
                                 = z
                                 CO O
                                 £ ^
                                CO CD
                                   *-
                                 A ®

                                If
X

CD
»^.
 O
 C
 o
 o

£
  •

 
38

-------
     The catalyst in the first stage reactor (nickel or palladium) requires
more frequent regeneration than most refinery catalysts due to formation of
                               aerate
                                3,11
gums. '     Catalyst may be regenerated about every 4 months and coke is
burned off every 9 to 12 months.

     From the first reactor, the hydrogenated di-olefins and olefins are
sent to a second reactor (Stream 2).   Reactor effluent is then cooled and
discharged into a separator (Stream 3).  Part of the gas stream from the
separator is recycled back to the reactor (Stream 4) after being scrubbed
with caustic solution.    The liquid phase from the separator is sent to a
coalescer (Stream 5) where water is used to trap particles of coke formed in
the reactor.    Next, the light hydrocarbons are removed from the liquid in
the stabilizer (Stream 6).  At this; point, the process becomes similar to
the solvent extraction of reformate; in the catalytic reforming of naphtha.
The stabilized liquid is extracted with a solvent, usually Sulfolane or
tetraethylene glycol (Stream 7).

     The raffinate (Stream 8) contains paraffins and may be sent to a
cracking furnace to produce olefins.     The solvent may be regenerated
(Streams 9 and 10).  Dissolved aromatics (benzene, toluene, and xylene) are
separated from the solvent by distillation (Stream 11) and are then sent
through clay towers (Stream 12).   Individual components (benzene, toluene,
and xylene) are finally separated (Stream 13) and sent to storage.

Benzene Emissions from Ethvlene Plants and Benzene Recovery from
Pyrolysis Gasoline

     Production of ethylene from naphtha/gas oil does not emit large
quantities of total VOC or benzene from process vents during normal
          9
operation.   Emission factors for benzene from sources at ethylene plants
are shown in Table 5.  The chief source of benzene emissions during normal
operations is the charge gas compressor lubricating oil vent (Stream 47,
Vent G in Figure 6).  The emission factors in Table 5 were developed from
data supplied by ethylene manufacturers.  The derivation of the emission
                                      39

-------






















z
o
h^
CO
CO
1— 1

w

z
N
Z

cfl

a
j

o
0$
Z
o
CJ
Q
5

Q
,-J
0
H
Z
O
u
z
*~







03
.^^
H








































(fl
-
H

^?

P-<

Z
S
X
H


<
U
M
H
H
O
fri
>>


Oi
0
on
as
O
H
U
<
k*4




























X
*j C
C 4)
41 -4
U U


ft! <+4
W


c
0
w
*

__ 1
0
Ll
u
g
u




o
u

-< Ib
>-l g
O C 00
U o S
C 1 00








V4
D O
^ fc,o
2 a S
S3-£
o n
u «
w





g
V -4

I I

VI
V













U
u


0
w






CO
Ov
III Ot 1

III CO 01
O CO




c
1 5
*J U
m LI v>
v) O Li J4
S 41 I 41 0 O « 41
C 1 ^ *J U  1
III CM CO 1
CM i-t CM
r*» r-i







§s s
NX 0
C 41 -••
01 iJ V)
^ VI ^
V -j 4J
O O 4J 4) O*<-l S X
oodvi .-r»4»«OA r*
C-4 -» 3 ^ rH ^-t -» +J -
u (^ o 4H C 41 V
-HO N V)
-J VI fO
n w
0 § Z




Cb
fc
Q
O W O OB ^
CQ

,







« C C
4J ^ -4
c « c <* c
41 Ll -4 V) U -4
> 4J (9 2 4J *
W v) O Ll
,-4 Ll 4-1 W -4 U 4-1
-4 0 4) V) O
0 V) Li U tf) Vt Ll
V) 0 O -4 vi o
V 4) vi Ll 6 » «
,Q Ll tfl Q* 41 Ll ID
3 a v o. « «
r-t g M .-t 4J 6 Ll JSJ
LiUgE 41UE4 *O T3
O OCv)4-i 0 4-» 41 (A X
V)VUO^4J4IO 6 4t U
« -1 C O -^ -1 VO > *
agt—f -< £ ofl .-« ao *o -^ "O
urimutflCcrtrt 4J c
f-43VV)0)-^3M4J -4 O
«Q^-<4-t«QOrH 00 U
4J 6 C 4J 4 3 4t



X




CM
^
»n

o
>">
-4
O
«
a

1
0
a
M
4)
Ll
0
U


g


u
o

V
2
3
O

JD









































•o
01
U
3
•O
O
u
a
£
^4
4J
4)

2?

Ll
a
*d
V
u
E


4)
C
41
N
C
41
4-t
0

00
V)
«

•d
41
V)
VI
41
U
g-
4)

VI
0
4J
O
tfl

0




Ll
4)
4J

i
tl

*
0
•4-1
41
V)
Q

4J

•d
3
*y
s
V)
c
o

VI
V)
8
4>
X
Ll
T3
C
0
y
41
en


<
X
"O
4)
a
5-
C
—4
g
5
VI
4)
Ll
4
V)
§
in
—4

U
41
•*4
4-1
00
a



o

w
Ll
O
4J
y

4H

1
-------
factors is shown in Appendix A.   Most benzene emissions from ethylene plants
are intermittent emissions that occur during plant startup and shutdown,
                                         9
process upsets, and emergencies (Vent E).    For example,  benzene may be
emitted from pressure relief devices, intentional venting of
off-specification materials or depressurizing and purging of equipment for
            9
maintenance.   Charge gas compressor and refrigeration compressor outages
are also potential sources of benzene.  Emissions from these compressors are
generally short-term in duration, but the pollutants may be emitted at a
high rate.  In general, intermittent emissions, all pressure relief devices,
and emergency vents are routed through the main process vent (Vent E in
Figure 6).  The vent usually is controlled.  The relief valve from the
demethanizer is usually not routed to the main vent.  But the valve is
                                                            9
operated infrequently and emits mainly hydrogen and methane.

     Potential sources of benzene such as flue gas from the cracking furnace
(Vent A), pyrolysis furnace decoking (Vent B), acid gas removal (Vent D),
and hydrogenation catalyst regeneration (Vent F) generally are not
                               9
significant sources of benzene.   Flue gas normally contains hydrogen,
products of methane combustion, and emissions from pyrolysis furnace
decoking consist of air, steam, carbon dioxide, carbon monoxide, and
                             9
particles of unburned carbon.   Emissions from acid gas removal are hydrogen
sulfide, sulfur dioxide, and carbon dioxide; these emissions are generally
controlled to recover hydrogen sulfide as sulfur or convert hydrogen sulfide
to sulfur dioxide.  As discussed earlier, catalyst regeneration is
infrequent and no significant concentrations of benzene have been reported
                            9
as present in the emissions.

     Fugitive benzene emissions at ethylene plants may originate from pumps,
valves, process sampling, and continuous process analysis.  The derivation
of the emission factors is shown in Appendix A.  Storage of ethylene in  salt
domes is not a potential source of benzene emissions because the ethylene
generally does not contain benzene.
                                      41
-------
     The emission factor for benzene from storage tanks shown in Table 5 was
                             9
derived from AP-42 equations.   No supporting data showing how the equations
were applied were provided by the emission factor reference.     /
     Secondary emissions include those associated with handling and disposal
of process wastewater.  The emission factor in Table 5 was derived from
estimates of wastewater produced and the estimated percent of the total VOC
emitted from the wastewater that is benzene (see Appendix A).

     No data were available concerning benzene emissions from recovering
benzene from pyrolysis gasoline.  Likely sources include reactor vents,
compressors, and any vents on the benzene column (Figure 7).

     The primary control technique available for intermittent emissions of
benzene (pressure relief valves, emergency vents) is flaring.  Other control
methods are not as attractive as flaring because the emissions are
infrequent and of short duration.  The estimated efficiency of flares is
              9
90-98 percent.   Flares can be operated to achieve efficiencies of
                                                                13
98 percent, if not greater, by controlling operating conditions.    EPA
research efforts have documented the efficiencies of flares.  The reader may
wish to consult Reference 13 for more details on flares.

     Fugitive emissions may be controlled by inspection/maintenance plans or
                                                    9
use of types of equipment such as tandem seal pumps.   Emissions from
sampling lines can be controlled by piping sample line purge gas to the
                                              jr.
                                               9
                                                 9
charge gas compressor or to a combustion chamber.    Streams from process
analyzers may be controlled in the same manner.
     The primary means of controlling emissions from pyrolysis gasoline or
naphtha feedstock storage is the use of floating roof tanks.  Emissions can
                                                                      9
be reduced by 85 percent when internal floating roof devices are used.
                                      42
-------
Process Description:  Coke Oven Light Oil Distillation

     Although most benzene is obtained from petroleum,/ some is recovered
through distillation of coke oven light oil at coke by-product plants.
Table 2 lists facilities which produce benzene from light oil.  Light oil is
a clear yellow-brown oil which contains coal gas components with boiling
points between 0 and 200 C.   Most by-product plants recover light oil,  but
not all plants refine it.  About 13-18 L of light oil can be produced from
coke ovens producing 1 Mg of furnace coke (3-4 gal/ton).   Light oil itself
contains from 60 to 85 percent benzene.

     The coke by-product industry recovers various components of coke oven
gas including coal tar, pitch, ammonium sulfate, naphthalene, and light oil.
Because benzene is contained in the coke oven gas, benzene may be emitted
from coke by-product plants which do not specifically recover or refine
                                                                    14
benzene.  Table 6 lists coke by-product plants in the United States.
Figure 8 shows a process flow diagram for a representative by-product
recovery plant.  The figure does not necessarily reflect any given plant,
nor does the figure include all possible operations that could be found at a
given coke by-product facility.  The number of units and the types of
processes used at specific plants are available.  For example, naphthalene
recovery is not practiced at all plants and some plants do not separate
benzene from the light oil.  Therefore, it is advisable to contact a
specific facility to determine which processes are used befo.re estimating
emissions based on data in this report.

     Coke by-product plants convert coal to coke in coke ovens.  About
99 percent of the United States production of coke uses the slot oven
process; the remainder is produced with beehive ovens.  The coking time
affects the type of coke produced.  Blast furnace coke results when coal is
coked for about 18 hours.  Foundry coke, which is less common and is of
higher quality, results when coal is coked for about 30 hours.
                                      43
-------
TABLE 6.   COKE OVEN BATTERIES CURRENTLY OPERATING
          IN THE UNITED STATES14

Plant, Location
Dramon Company , Tarrant , AL


Empire Coke, Holt, AL

Koppers, Woodward, AL




Gulf States, Gadsden, AL

Jim Walters, Birmingham, AL


U. S. Steel, Fairfield, AL



National Steel, Granite City, IL

Interlake , Chicago , IL

LTV Steel, South Chicago, IL
Bethlehem Steel, Burns Harbor, IN

Citizens Gas, Indianapolis, IN


Battery
Identification
Number
A
5
6
1
2
1
2A
2B
4
5
2
3
3
4
5
2
5
6
9
A
B
1
2
2
1
2
E
H
1
Status3
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
2
2
2
0
0
0
0
0
0
0
0
0
0
                        44
-------
TABLE 6.   COKE OVEN BATTERIES CURRENTLY OPERATING
          IN THE UNITED STATES14 (Continued)

Plant, Location
IN Gas, Terre Haute, IN

Inland Steel, East Chicago, IN




U. S. Steel, Gary, IN





LTV Steel, East Chicago, IN


Armco, Inc., Ashland, KY

Bethlehem Steel, Sparrows Point, MI)








Rouge Steel, Dearborn, MI

National Steel, Detroit, MI

Battery
Identification
Number
1
2
6
7
8
9
10
1
5
7
13
15
16
4
9
3
3
4
1
2
3
. 4
5
6
11
12
A
A
C
4
5
Status3
0
0
0
0
0
0
0
0
0
1
1
1
1
2
2
2
0
0
2
2
2
2
2
2
0
0
0
1
1
0
2
                        45
-------
TABLE 6.  COKE OVEN BATTERIES CURRENTLY OPERATING
          IN THE UNITED STATES   (Continued)

Plant, Location
Carondolet, St. Louis, MO


Bethlehem Steel, Lackawanna, NY


LTV Steel, Warren, OH
Armco, Inc. , Middletown, OH


New Boston, Portsmouth, OH
Koppers, Toledo, OH
LTV Steel, Cleveland, OH





U. S. Steel, Lorain, OH




,

Dramon Company, Keystone, PA

Bethlehem Steel, Bethlehem, PA



Battery
Identification
Number
1
2
3
7
8
9
4
1
2
4
1
C
1
2
3
4
6
7
D
G
H
I
J
K
L
3
4
A
2
3
5
Status
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
2
0
0
2
2
2
2
2
2
2
2
2
0
0
0
2
                        46
-------
             TABLE 6.  COKE OVEN BATTERIES CURRENTLY OPERATING
                       IN THE UNITED STATES14  (Continued)

Plant, Location
LTV Steel, Aliquippa, PA

LTV Steel, Pittsburgh, PA •




Koppers , Erie , PA

Shenango, Pittsburgh, PA

U. S. Steel, Clairton, PA











U. S. Steel, Fairless Hills, PA

Wheieling-Pitt, Monessen, PA


Southern, Chattanooga, TN

Battery
Identification
Number
Al
A5
PI
P2
P3N
P3S
P4
A
B
1
4
1
2
3
7
8
9
15
19
20
21
22
B
1
2
1A
IB
2
1
2
Status3
2
2
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
0
0
2
2
0
2
2
2
1
1
0
0
Lone Star Steel, Lone Star, TX
                                      47
-------
              TABLE 6.  COKE OVEN BATTERIES CURRENTLY OPERATING

                        IN THE UNITED STATES14 (Continued)
                                                  Battery
                                              Identification
    Plant, Location                               Number              Status
U. S. Steel, Provo, UT



Wheeling- Pitt, East Steubenville, WV



1
2
3
4
1
2
3
v 8
1
1
1
1
0
0
0
0
 Status:  0 - on-line or operating.
          1 - hot idle, removed from service but likely to be returned to
              production in short time.
          2 - cold idle, removed from service and not likely to be returned
              to production.

Note:  This listing is subject to change as market conditions change,
       facility ownership changes,  plants are closed, etc.  The reader
       should verify the existence of particulate facilities by consulting
       current listings and/or the plants themselves.  The level of benzene
       emissions from any given facility is a function of variables such as
       capacity, throughput and control measures, and should be determined
       through direct contacts with plant personnel.   These operating
       plants and locations are current as of December 1987.
                                      48
-------
                                     co
                                     a
                                     o



                                   I!
                                   co c
                                   o _o
                                   w 0)
                                   O CO
                                   ™
                                   5?
                                   *-

                                   8

                                   c
                                   « 5 "5
                                   i 75

                                     § ®
                                     co
                                       CO
                                     J2 o
49
-------
     The coking process is actually thermal distillation of coal to separate
volatile and nonvolatile components.   Pulverized coal is charged into the
top of an empty, but hot, coke oven.   Peaks of coal form under the charging
ports and a leveling bar smooths them out.  After the leveling bar is
withdrawn, the topside charging ports are closed and the coking process
begins.

     Heat for the coke ovens is supplied by a combustion system under the
coke oven.  The coking process requires about 15-18 hours to produce blast
furnace coke and 25-30 hours for foundry coke.    The gases that are evolved
during the thermal distillation are removed through the offtake main
(Figure 8) and sent to the by-product plant for recovery.

     As shown in Figure 8, coke oven gas leaves the oven at about 700 C and
immediately is contacted with a cooling spray (Stream 1).  The spray reduces
the temperature of the gas and acts as a collecting medium for condensed
tar.  The gas then passes into the suction main.  About 80 percent of the
tar is separated from the gas in the mains and is flushed to the tar
decanter (Stream 2).     Another 20 percent of the tar is condensed and
collected in the primary cooler (Stream 3).    Smaller amounts of tar are
removed from the gas by collectors (electrostatic precipitators or gas
scrubbers) (Stream 4).    The tar is then separated from the flushing liquor
by gravity and sent to the tar dewatering tank (Stream 5).   The tar decanter
generally separates "heavy" tar from flushing liquor and the primary cooler
decanter (or primary cooler intercepting sump) accepts "light" tar which is
cleaner and less viscous (Stream 6).     Depending on plant design, these two
tar streams (Streams 5 and 6) may be merged or separated.  Tar from the
decanters is further dewatered in the tar dehydrator (Stream 5), and the
separated water is recirculated (Stream 7).  Tar may be sold to coal tar
refiners or it may be refined on-site.  Tar and tar products are stored
on-site in tanks.  Wastewater processing can recover phenol (Stream 8) and
ammonia, with the ammonia routinely being reinjected into the gas stream
(Stream 9).  Ammonia salts or ammonia can be recovered by several processes.
Traditionally, the ammonia-containing coke oven gas is contacted with
sulfuric acid (Stream 10) and ammonium sulfate crystals are recovered
                                      50
-------
(Stream 11).   The coke oven gas from which tar and ammonia have been
recovered is sent to the final cooler (Stream 12).   The final cooler is
generally a spray tower, with water serving as the cooling medium.    Three
types of final coolers and naphthalene recovery technologies are currently
used:  (1) direct cooling with water and naphthalene recovery by physical
separation, (2) direct cooling with water and naphthalene recovery in the
tar bottom of the final cooler, and (3)  direct cooling with wash oil and
naphthalene recovery in the wash oil.    Most plants use direct water final
coolers and recover naphthalene by physical separation.    In this method,
naphthalene in the coke oven gas is condensed in the cooling medium and is
separated by gravity (Stream 13).  After the naphthalene is separated, the
water is sent to a cooling tower (Stream 14) and recirculated to the final
cooler (Stream 15).   The coke oven gas which leaves the final cooler is sent
to the light oil processing segment: of the plant (Stream 16) .

     As shown in Figure 8, light oil is primarily recovered from coke oven
gas by continuous countercurrent absorption in a high boiling liquid from
                                           3
which it is stripped by steam distillation.   Coke oven gas is introduced
into a light oil scrubber (Stream 16).  Packed or tray towers have been used
in this phase of the process, but spray towers are now commonly used.   Wash
oil is introduced into the top of the tower (Stream 17) and is circulated
through the contacting stages of the tower at around 1.5-2.5 L/m  of coke
oven gas.    At a temperature of about 30 C, a light oil scrubber will
remove 95 percent of the light oil from the coke oven gas.  The
benzene-containing wash oil is steam stripped (Stream 18) to recover the
                                                                   1
                                                                   16,17
light oil.     Steam and stripped vapors are condensed and separated
(Streams 19 and 20).'  The light oil is sent to storage (Stream 21).

     To recover the benzene present in the light oil, processes such as
Litol (licensed by Houdry) or Hydeal (licensed by UOP) are used.  Figure 9
shows a process diagram of the Litol process.  The following discussion of
                                                                       1 $
the Litol process is.drawn from a published description of the process.
The light oil is prefractionated (Stream 1) to remove the C  and lighter
fractions, and the Cq and heavier fractions (Stream 2).  The remaining
"heart cut" is sent to a vaporizer where it contacts gas with a

                                      51
-------
CO  ,

80  W
                         ID »•
                         C  °
                         O  3
                         3-0

                         O  S
                        HO.


                         CD —
                         C O
                         CD 3
                         N 13
                         = O

                        Set
                           O
o
X ,
111*







1
*^ 1
"c -
« c i
1 Hydrog
Generatio
»n^i«, kj«i
J3

i







a.c
3 °
en
5S
Gas Compr




^3
C
CD
O

O





1





OB
C
»i
•S
£






r













i
T-T


©u
T\

„/

^ ^y
^^^^

J

r*oj
k- *- 1.
O CO C
« 2 1
11 I a

CB
•^^

CO
J

CO

o
     co
     co
     3
     O
     m

    •o
     CD

    *-
     o
    .**
    •o

     o
     Q.
     CO

     2

     o
     o

    2
    3
    O>
                                                                                  § «
                                                                                  0 CO
                                                                                  CO CD
                                                                                  E  a.
                                                                                  3  CO

                                                                                  <=  Z


                                                                                  CO  O
                                                                                  CD  ••-
                                                                                                   00
                                                                                                   r-l


                                                                                                     2
                                                                                                     CT
                                                                                                     03

                                                                                                    Q
                                                                                                     CO
                                                                                                     CO
                                                                                                     o
                                                                                                     o
                                                                                                     o

                                                                                                    a!

                                                                                                    'o
                                                                                                    O)
                                                                                                    O)
                                                                                 CO
       X
       CD
                                          52
-------
high-hydrogen content (Stream 3).   The light oil and hydrogen then flow to a
pretreat reactor (Stream 4) where  styrene,  di-olefins,  and some sulfur
compounds are hydrogenated.  The partially hydrogenated stream is heated
(Stream 5) and sent through a set  of fixed bed reactors (Streams 6 and 7),
where all remaining sulfur compounds are converted to hydrogen sulfide and
organics are dehydrogenated or dealkylated.  The reactor effluent is cooled
(Streams 8 and 9).   The product is then stabilized (Stream 10) and benzene
                                                      18
is recovered by conventional distillation (Stream 11).     Benzene product
may then be sent to storage (Stream 12).

Benzene Emissions from Coke By-product Plants

     Benzene may be emitted from many sources within a coke by-product
plant; emissions are not limited to the benzene recovery section of the
process.  The coke ovens themselves are potential sources of benzene
emissions.  Sources of emissions at coke ovens include charging, leaking
coke oven doors, and leaking coke  oven lids and offtakes.  Emissions from
lids and offtakes are often collectively called topside emissions.

     During charging, moist coal contacts the hot oven floor and walls, and,
as a result, the release of volatile components begins immediately.  Control
of charging emissions is more dependent on operating procedures than on
equipment.  Control options include staged charging, sequential charging and
use of wet scrubbers on larry cars (the mobile hoppers that discharge the
coal).  Staged charging involves pouring coal into the coke ovens so that an
exit space for the gases generated is constantly maintained.    The hoppers
delivering the coal are discharged such that emissions are contained in the
ovens and collecting mains by steam aspiration.  Generally, a maximum of two
hoppers discharging at the same time.  In sequential charging, the first
hoppers are still discharging when subsequent hoppers begin discharging
coal.  As with staged charging, the coke ovens are under aspiration in
sequential charging.  The  sequential charging procedure is designed to
shorten the charging time.  The final control method for charging emissions
is the use of wet scrubbers on larry cars.  With this control technique, the
scrubber emissions are contained by hoods or shrouds that are lowered over
the charging ports.
                                      53
-------
     Another potential source of benzene emissions at coke ovens is leaking
doors.  The doors are sealed before the coking process begins.  Some doors
have a flexible metal band or rigid knife edge as a seal.  The seal is
formed by condensation of escaping tars on the door's metal edge.  Other
doors are sealed by hand by troweling a mixture into the opening between the
coke oven door and door frame.   After the coking process is complete,  the
doors are opened to push the coked coal out into special railroad cars
called quench cars for transport to the quench tower.  Quenched coke is then
discharged onto a "coke wharf"  to allow quench water to drain and to let the
coke cool.  Emissions occur during the coking process from leaks in door
seals.  Control techniques for leaking doors include oven door seal
technology, pressure differential devices, hoods/shrouds over the doors, and
use of more efficient operating/maintenance procedures.

     Oven door seal technology relies on the principle of producing a
resistance to the flow of gases out of the coke oven.  This resistance may
be produced by a metal-to-metal seal, a resilient soft seal, or a luted
seal.  Small cracks and defects in the seal permit pollutants to escape from
the coke oven early in the cycle.  The magnitude of the leak is determined
by the size of the opening, the pressure drop between the oven and the
atmosphere, and the composition of the emission.

     The effectiveness of a pressure differential control device depends on
the ability of the device to reduce or reverse the pressure differential
across any defects in the door seal.  These systems either provide a channel
to permit gases that evolve at the bottom of the oven to escape to the
collecting main or the systems provide external pressure on the seal through
the use of steam or inert gases.

     Oven door emissions also can be reduced by collection of the leaking
gases and particulates and subsequent removal of these pollutants from the
air stream.  A suction hood above each door with a wet electrostatic
precipitator for fume removal is an example of this type of system.
                                      54
-------
     Other control techniques rely on operating and maintenance procedures
rather than only hardware.  Operating procedures for emission reduction could
include changes in the oven cycle times and temperatures,  the amount and
placement of each charge, and any adjustments of the end-door while the oven
is on-line.  Maintenance procedures include routine inspection, replacement,
and repair of control devices and doors.

     Topside leaks are those occurring from rims of charging ports and
standpipe leaks on the top of the coke oven.   These leaks  are primarily
controlled by proper maintenance and operating procedures  which include:

          replacement of warped lids;
          cleaning carbon deposits or other obstructions from the mating
          surfaces of lids or their seats;
          patching or replacing cracked standpipes;
          sealing lids after a charge or whenever necessary with a slurry
          mixture of clay, coal, and other materials (commonly called lute);
          and
          sealing cracks at the base of a standpipe with the same slurry
          mixture.

     Luting mixtures are generally prepared by plant personnel according to
formulas developed by each plant.  The consistency (thickness) of the
mixture is adjusted to suit different applications.

     Emission factors were often not available specifically for benzene
emissions at coke ovens.  One test, examining emissions of door leaks,
detected benzene in the emissions.  The coke oven doors being tested were
controlled with a collecting device, which then fed the collected emissions
to a wet electrostatic precipitator.  Tests at both the precipitator inlet
and outlet showed benzene concentrations of 1 to 3 ppm (or about
         3
3-10 mg/m  ).  These data translated into an estimated benzene emission
factor of 0.6 to 2.4 kg benzene per hour of operation for coke oven doors.
No further emission factors for benzene and coke ovens were found in the
literature.  However, an analysis of coke oven gas indicated a benzene
content of 21.4 to 35.8 g benzene/m  .
                                      55
-------
     Other potential sources of benzene emissions associated with the
by-product plant are given in Table 7 along with emission factors.
Equipment leaks may also contribute to benzene emissions.  Emission factors
for pumps, valves, etc., are shown in Table 8.  This section of the report
describes the potential sources of benzene emissions listed in Tables 7 and
8.  Emission sources and control technologies are described in groups of
related processes, beginning with the final cooling unit.

     The final cooler unit itself is not a source of benzene because coolers
are closed systems.  However, the induced draft cooling towers used in
conjunction with direct water and tar bottom final coolers are potential
sources of benzene.  Benzene can be condensed in the direct-contact cooling
water, and, in the cooling tower lighter components (such as benzene) will
be stripped from the recirculating cooling water.  The emission factor of
270 g/Mg coke (0.54 Ib/ton) shown in Table 7 was based on actual
measurements of benzene concentrations and volumetric gas flowrates taken
from source testing reports.    Control technologies available for cooling
tower emissions are conversions to different types of cooling towers.  For
example, a facility with a direct water final cooler could insert a
one-stage mixer-settler into the final cooling process and thus obtain the
benefits of a tar bottom cooler.  Although a tar bottom cooler does not
eliminate benzene emissions from the cooling tower, it does eliminate
benzene emissions associated with the physical separation of naphthalene and
water.

     Use of a wash oil final cooler effectively eliminates the benzene
emissions associated with direct water or tar bottom coolers because the
wash oil is cooled by an indirect heat exchanger thereby eliminating the
need for a cooling tower.    Wash oil is separated after it leaves the heat
exchanger and recirculates back through the circulation tank to the final
cooler.

     As discussed earlier in this section, coke by-product plants may recover
naphthalene by condensing it from the coke oven gas and separating it from
the cooling water by flotation.  Benzene may be emitted from most
                                      56
-------

























Q
W
o

0 >
l-i 03
C/2
CO U
M ^
W C_>

H JH
z erf
to o
N Z
Z 5
U O
CQ r*i


r-
W
_!
CQ
H














0
-i C
O V
U -4 *-*
4J O X
C -1 —*
O *W
0 •«
U








g
-4
4J
O
u

I



4)
O 0
.0"
Of >t
-^ T3
* " i
v u o
-4 0 En
.-i u
0 0
U 4
U Eb V
g c-S
O 0 O
in «
in u
•g a
H u
U,


V

o o

u C*
•OS'S
J5 M I
-1 14 (t,
0 0
M 4^

C «
O [14 »
0 -*
ceo
o o u
vf tt
to u
tU k4
3
[I.


g
V 4 U
U U V
u -4 ja
ga I
Ul U Z
C
V
M







V

U
3
. 0
(A



1+4 1
rH O O 00 1
CO O O O* 1
rt -H


U
«
-4 14 U
0 « U
0-4-1
U O 0
0 O
-t o o
(4
C -J -I
-4 « «
•* 5 5
I ^ ^ ?
±J -4 -4 U
0 O 0 «l
f i i "§
tt M in 4
5 S S 3
0) V 4) «]
tft «l vt 4
3 » 3 O «


-* 1 1 O> 1
r^ i i o i
CO





1 1 00 >
i < • i
rH 1 1 fH 1







r*- rH 00 O«
rH






O o ON r*-
pv P*. CO O
CM





*< ^00



. 1
C
11 C
> 5
U 4J
4) 4
2 «
« a
M 4) C tfl

34 0 O 4)
0 J w C
HI u ^H 4)
4J 0 -4 -H
g1 a « ? j
*4 14 U U U
_i -l « .C -C
0 Q H M a
0 ~4 tt
0 J Z




cocoo*ncoocooooco













U M U
4) 4) 4) 
M4)tfi4)4)W)cJ4)tf)lU
% -X •% % C -? -^
SS-'SQ^1 4)^-4^
44-^44-<-l 4-^4
J3 f
C14U C 4 ^ — 4 4
-44)4 S U ,4*4 U
« 4-1 5 CJ 0 00 O
n C v at u ii u
4I-4Q QCO ^ti "*
u r J- •£
OUUl MU 0000 X
U44 44 ^-4 H
Ou t-i t-» HE-4 »J .J A




O CO



G
V

to

V) V)
4
00 00
C
4 U
14 U OJ
V 3 -X
J3 U C
•§ S J!
l-« J3
O U
wi 0 op
-4 C! "4
~* 4» U
0 bO QJ
l 0 -^

3 S5 3

^
CO O
0 0





CO CM
lO r4
O O





rH

en






00
O







4)
-J










4

Benzene Sto
57
-------



























•o
a |
< 4J

Cd 0
U U
•
M eq
CO
CO Cd
2 U
Cd ><
Z Cti
Cd Q
N Z
Z D
§ £



•

Cd
-I
CQ






















>,
u
^ C
O V
U -4
u u
C -4

o m
u




















*i

j!

T, "
« M
-1 O
— 1 4J
O O
id
0 0
n
(A
•g
w


o
£
trolled
§4
h.
C C
3 O
— t
n
(A
1





§
M
« 4
o u
M -4
O -H
01 u
d

T)





















M












d
0
jj
<§•

a
M

C
0
u




s
s

£

1
£

V
o
V
2
H
m

1
1
tt
"o
U
8
a
3





4
u
1
2













U
u
3
o
W







00 00 O 00 CO













M
Jg
M) 00 JO 00 M}
C C 3 C C
-4 ^4 n -4 ^4
jj JJ U J-l JJ
V V «) V V
•* •? _, -i i
C u *-^ C C
rt 4-444
-t ^ o -^ -^
JO J3 1 JO jS

in in in n «i
4 <« 4 <« <*
O O 3 O O


en co 
0)
•8
U
i
d)
00
g
8.
"S
JJ
jj
•H
§

c
tt
N
C
0



2
00

0
V)
i

d


«
W)
s
s-
tt
V
4
V)
O

j{
»H


-4
1
U



























GO
S
-4

d
•H

JJ
-4
§

V
§
V)
It
g
J2
4D
S
4
n
M
«
r-4
8
u
It
a
*
^
«
|
•*4
4


*W

jj
4
,
JJ
a
«

•d

V
3

n
S
4
r- t
a.
M
«

1


Jj
g

^
^
3
n
3
t)




















00
u
M
3
U
~4



-1
agram
^

JS
4-1
d
o

o
5
5
•d
V
5
~4
•o
a
~4
tl
QA
ct stora
^
o
o.
-^
It
«
14


V
14

tt


i
«
d
V
J3

O

g



.

*l
U
a
a

— i
—4
4
I
0) V







M

§
JJ

V
O
o
u

1
£H
V4
V
jj
3
i
u
u

•A

v
U
o

a
0
*-!
4J

•o
V
u
ission :
§
JJ
d
V
u
u
a
«
«4
VI
V
8


S
*
A



a

o
•+4
d
5
jJ
S
•6
ti
u
d

5
•g
•
4J
d
8
M
D.

O
O
•H
V)
«
1
1
M
M
58
-------
























^
CO
C/5 H


^c ^c

J CM

EH H
Z C.3
w g
PL, O
M OJ

s> ><
« *
O Cd
Pn s/;
O
to O
O SH
EH Od
U Q
< Z

Z C*4
M a
to 5
2* W
<;



[ij HD
N fa
Z
U EH
cQ <


.
00
W
hJ
03
H



•









































•°.

n
a

V
u

3 3
— 1 M
CL, —
£"*
R U
3 -4
0 H
U
4J
C
O
u


.n
it
«
Q
41
U U

ELI W
V tt
C ai



4J
0
U
§



g
01
V)
-

c
>

"8
U
C
V
£














2
4J
C
o
0





















~4
o

£
U)
-r4
J











_l
g
M
3














^
o
4J
u




n
V
o
u
1













u
11
»4
U
~4
HH
U

















t


X

»

o
£



t

t-^
-4
O
4J
H

|

fc
X
cove
^


<



u
00
3




o
u
kl
3
0
to
5



I


0
_1













^













c
_J
u
<§•
g
4J
o
u



e
v4
c
~4
»*4
V

V
d
V
N
£
a
i

u
0
u
X
£
)
I
c
>H
3.
V
C
V
N
c
V
CQ
)



O
OS
^

CQ



x
4
Q



X
^H
— 1
d

4J
-4
h-|














coe^o r-tcno m-a-a -*oio o o
*O r«* o r^ooo to\oo "ffirio o o




V) V)
0) -1
> rt M E
— 1 ft) ,4 v
<•  C ££Cw)C
O O^O & O-X-J
~4d« -^C« -4fi« -^C« -^
o -< o 0-^-4 o -* a» u-^jrf e
a) u -t H«)«-< ->^V)V 00 CL
C -0 CE Cn CMkiA
X-^« X-4 X-^W X-^3 3 O
_4 ^4^4 -4^ (1 ^ 4Ja
«<-<« Vi-l3 «r-t« UHpTJ O
U J= « 4J J= T3 U J3 T) « 4J.GM «
Lt 4-1 UI4-) UIW4JU4J n&O
«C« rt c at rt C ti C rt C W o 7
3Qoi 3Q«I 3 O « 4) 3 O •» ^-( -4
<^aS3 O'SS O'XS> 0*23 O S




CO
O *O CO •*
N «-l M «-» WO
0 W O to 0 O






»n
2 ^. s tn as
O f-t O (M O O




o r.
OJ 00 rH -»
f>i CO N -» CO O
O N O (O O O







U CO
1-4 9 C4 I** C4 O
O i-* O i (^ to o
O N •-< t*> O O






t-l CJ «



g
*J «
'H U V
« v c
~* s •**
'H 2 -J
U 0
n 04 u -o
M V
« V 00 13
w Mm £ C
(rt v) 3  f «
(t 3 x u v S Qi
> CU U] Ok Q Ui O

4J
00
— i


^
S
0)
(4
C
4)
-Q


0
w
C
V
o
a
v
4-1
1

0)
•H
I


11
u
g

«
-4
O
k4
i
G

y-t
3
o
*4
01
M
'e
^
>
g|
0
J
w
a
o
TJ
11
5

41
U
rt
X
X
H


T-*
^
g
V
M
*H
^
1
Li
*H
«l
U
(«

J5
s

0
u
(I
6
5
u
Q

b
a

u
q
PL,
it












01


U
»
4-t
o

V

<*4
U

4J
C
V
w
V
a
v
u

4J
g
V
fl»
u
i
vt
4J

O.
jj
1
u
a
i
J3
J
O
U



V)
«
X
u
4J
C
s
u
4-1
^4
"°
i
4J
U
o

"S
s
11

£
a
«

*
•i
o
u
3


g
W)
w>

I
u







































11
S
it
u
«i
w
5
e
>
»
Ji
x
d
0)
u
—4

U-l

iH
O
U
u
d
o
u

Jl
4J
•§
*
ul
0
4J
O


g
«
«
•1^
i
-o
11

o

g



g
•d
11

it
59
-------
naphthalene separation and processing operations.     Vapors from naphthalene
separation tanks have been reported to contain benzene,  benzene homologs,
and other aromatic hydrocarbons.    The emission factors for naphthalene
separation and processing shown in Table 7 are based on source testing data
from a flotation unit, drying tank, and melt pit at a coke by-product
recovery plant.
     Benzene may also be emitted from the light oil plant which includes the
light oil condenser vent, decanter, storage tank, intercepting sumps, the
wash oil decanter, circulation tank, and BTX storage.  An available control
technique is the use of gas blanketing with clean coke oven gas from the gas
holder (or battery underfire system).  With this technology, a positive
pressure blanket of clean coke oven gas is piped to the light oil plant and
the enclosed sources are connected to the blanketing line.  Vapor emissions
from the sources flow back into the clean gas system.  Ultimate control of
the vapors is accomplished by the combustion of the coke oven gas.    Such
systems are currently in use at some by-product recovery plants and
reportedly have operated without difficulty.  The control efficiency is

                                                                          16
estimated to be about 98 percent.    The emission factors for benzene
sources in the light oil plant shown in Table 7 are based on source tests.

     Sources of benzene emissions from tar processing include the tar
decanter, tar intercepting sump, tar dewatering and storage, and the
flushing liquor circulation tank.  Emission factors for these sources are
shown in Table 7.

     Benzene emissions from the tar decanter are sensitive to two operating
practices:  residence time in the separator and optimal heating of the
decanter.    These two variables should be kept in mind when using the
emission factors presented in Table 7.  Benzene is emitted from tar
decanters through vents.  Coke oven gas can be mechanically entrained with
the tar and liquor that are fed into the decanter.  Because tar is fed into
the decanter at a slightly higher pressure, the coke oven gas will build up
                                      60
-------
in the decanter if it is not vented.     Emissions were measured at tar
decanters at several locations in the United States and the emission factor
shown in Table 7 is the average of the test values.

     The water that separates from the tar in the decanter is flushing
liquor and it is used to cool the coke oven gas leaving the coke oven.
Excess flushing liquor is stored in the excess ammonia liquor tank.  Benzene
may be emitted from the flushing liquor circulation tank and the excess
ammonia liquor tank.  The emission factor of 9 g benzene/Mg coke was derived
from a source test of fugitive emissions from a primary cooler condensate
tank.  It was assumed that the condensate tank contained liquids similar to
the two sources of concern and that the tank was of a similar design.    The
actual benzene emission rate from the flushing liquor circulation tank and
excess ammonia liquor depends on the number of tanks, the number of vents,
and the geometry of the tanks.

     The tar intercepting sump is a type of decanter which accepts light tar
and condensate from the primary cooler.  Some of this condensate may be used
to make up flushing liquor and some may be forwarded to ammonia recovery.
No significant benzene emissions have been identified from the recovery of
ammonia, but benzene can be emitted from the intercepting sump.  An emission
factor of 95 g/Mg coke was reported in the literature.

     Tar dewatering may be accomplished by steam heating or centrifugal
separation or a combination of the two methods.  Use of centrifugal
separation will probably not be a source of benzene emissions directly, but
benzene may be emitted as a fugitive emission if storage vessels are used.
In steam heating, benzene could be driven off in the vapors.  The emission
factor for tar dewatering in Table 7 was derived by averaging three factors
based on source tests at tar dewatering tanks.  The emission factors were
41, 9.5, and 12.9 g/Mg coke.16

     The control technology applicable to sources of benzene from tar
processing is gas blanketing.  As described earlier in this section, gas
blanketing has a control efficiency of 95 to 98 percent.
                                      61
-------
     The final group of sources of benzene emissions at coke by-product
plants includes emissions from equipment leaks, such as pumps,  valves,
exhausters, pressure relief devices, sampling connection systems,  and open
ended lines.  Emission factors for these sources are shown in Table 8.  The
factors in Table 8 were based on emission factors from a comprehensive
survey of petroleum refineries and the percent of benzene in the liquid
associated with each type of equipment.    Two different sets of emission
factors are presented, one set for a plant practicing light oil and BTX
                                                              4
recovery and one set for a plant producing refined benzene in addition to
light oil.  Emission factors for exhausters were derived by multiplying the
VOC emission factor for compressors in hydrogen service and refineries by
0.235, the measured ratio of benzene to nonmethane hydrocarbons present in
the coke oven gas at the exhausters.

     The control options available for equipment leaks involve quarterly or
monthly inspections and maintenance, replacing equipment, or enclosing or
plugging the leaking sections.  Inspection/maintenance includes tightening
seals, etc., as they are found during the inspection.  The control
efficiency for quarterly inspection and maintenance ranges from
40-70 percent and 50-80 percent for monthly inspections and maintenance.
Replacing equipment or enclosing/plugging the sources eliminates the
emissions.

Process Description:  Petroleum Refineries

     Crude oil contains small amounts of naturally occurring benzene.  One
estimate indicates that crude oil consists of 0.15 percent benzene by
       19
volume.    Therefore, some of the operations at petroleum refineries may
emit benzene independent of specific benzene recovery processes.  Table 9
lists the locations and capacities of petroleum refineries in the U. S.
Most refineries are located along the Texas-Louisiana coastline of the Gulf
of Mexico.  A flow diagram of processes likely to be found at a model
refinery is shown in Figure 10.  (Note:  In Table 9, calendar day means the
time period midnight to midnight.  Stream day means any 24-hour actual
operation of a processing unit, regardless of calendar day.)
                                      62
-------





o
CSI
U
t— 1
H
O
&4
(J
Q
«j
W
H
H
W
Z
O
H
U
3

w
Ed
M
06
Cd
Z
M
fa
S
CO
Ed
H
^*
H
V)
a
H
z
p
o\
H





























































J3

>>
U
4
Q
4
U
ction
1
M
a,







u
1
a
3
J




. §
n ~4
o u
-4 4
*J N
0 «
•< i
«4



II
*



p
•o
^
>• A
^
o
2
3
•o
2
u>
•o
u
JO







4
C
O
u
o
o
•g
X
t"
u
              o
              o
              o
              o
              o
              o
                    o o o
                    o o o
                    O tA O
              o
              o
              CO
• o
• o
• CM
• r^
o
o
-»
«
o
0
0
o
• o
• 0
• o
• CO
o
o •
M •
o% >
• 0
• 0
• U1
• i-4
o
o
 I
  tl
U -4 0
C 3 O
         1
          O
          u
     £


    i5J5
     M O
     O 04
     Z
    • o o
   o c z
   B HI I
          U 4
          -4 0)
2.2

§1
-H a.
      ^   rs
Ul O


= £2
B 0. «
o
MOO
^ U U
  a u
  4 v
  X 04
2  §
O  N
«  M
           3
           
4
U
&
i— I
H
1
1
o
u

— 1
•r*
o
c
o
•*4
_)
f ^
£ •"
1 O
1 O
^4 IO
W
m
^4 fH
*4
O
«
«
u
Ib
?
•r4
o:

c
rt
^4
^-1
^
1
G
O
•~4
*J
S
u
4J
X
u

4J
B

_4
5






<
M



b
M
J
O
"o
^
o

'2
*-4
C

*H
V



Ji
u
-Carson
i
cS
T)

"H

U

Bi

U


B
^

•O
h
i
33

i
O
O

•-4
••4
o

c
0
o
4
M
-Wilmington
i
6
O
§
u
o
u
4-1
V
Oi

B


1"
o
kersfield
PQ
1
I
HI



V)

3

B
0

O






o

B TJ
3 «
tt 0
v 3
co £
u
u oi

u
1
4
B
u

1

o
c
Ht

0
u
o
g
u
1
«
I

u

^4

o


o

f
•o
U]


4
U
B
V
a

i

o
o

§
X
Cd
0
in
u
4
U
1
O
u
8
eflni:

—

^

O




u
4J
01
b.
                            63
-------



c
c
0
u
V^"

o
CN
Cd
l_(
EH
M
O
CD
0
H
^
H
03
Z
O
H
U
3



Cfl


M
0.
u
z
H^

u
oi

Cd
H
H
C/2
a
H
M

pr


Cd
D3
2



•o
.0
>
^
c
4
O
ft
0
c

4J
0
s
t
t-
ou






































4
1
o>
m
0*
1






B
I 0
o u
4J N
4 H
§ S
a i
«i
••-4

IS F*"!
S-o1
< 5.




•o
VI
>• 2
4J
U
4
O
4
U

V
1
o

o
A












G
O
Company and Locati
O OOO* • • O O • O O •
o ooo< ' • o o • o o •
CO O if\ tf\ * » ->OO >OO
n *T ^ *& . , , ^ rt > IS) r-t •
o .... 	 o •
o .... 	 • o •









. • • . 	 o
. * • • - 	 o



Q... 4 o • • • * O O
o-*. . o • • • • o o
o • * • -.*»•••• o ... • iN • • • • 00 0
1-1



O OOOO OOOO OOO
0 OOOO OOOO OOO
m cnooo mo • m erf o o o

»H •* H e ^ ^f — ' " Mm'i^-j
O 0 3 «l 8 O«M44^
u u n MX xzopwvin



















































|
Superior Processing Co
o
o
o
«-l










0
0
o



o
o
•T
CM



O
0
o
*o
•*










pig














Santa Fe Springs














































i



IH
-4
0
Texaco Refining fc Mark























o
o
o
CM









0
0
o
00
rt












Bakerafield

















0
0
"*



o
o
00









o
o
o
I/I
f*.












Wilmington

















o o
o o
VI O
o o
tH f-i



OOO
000
CM *H m
en rH CM








OOO
OOO
OOO
iO 00 00
CM O rt
*H i-i fH











VI
V
Toxco Corp . — Mart ine z
Unocal Corp. — Los Ange
Rodeo
















































VI

•°
o
o
o
o"
m
Solvent extraction:
            o o
            o o
            o m
                     o
                     o
                     o
   o
   o
   m
o o
o o
o o
                    V
                    M
                    » ><
                    B u
                 2  " 3
                 I  S^
                 3  x a
                 Q  H
   5

   9
                                 I
                                 c
2
     o
     g
     w
o

0
64
-------
                                                                                       o
                                                                                       o
                                                                                       o
                                                                        oo  en
                                                                        CM
  4J

   C

   O
  U
                    .a
O
CM
  V)
  U

  a
                                                                  o   •  o  o
                                                                  o   •  o  o
                                                                  o   •  oo  tn
   Ed

   H


   H
   CO
*  *
   o
o o o o o o
o o o o o o
O O to O O O

to oo r> in e>T to
         N (M tH
o    o
o    o
r»    m
o o
o o
o «n
   Z
   O
   M

   H


   U
   to
   Ed
   M

   ai
   Ed
                                                            o  o  o  o  o  o
                                                            o  o  o  o  o  o
                                                            o  o  o  o  o  o

                                                            oo  in  »n  o  to  r*
                                                            to  to  o  o  r^  in
                                                                  Ct|  eN  C4  tH
                                                                           1
                                       o o
                                       a o
                                       o o
                           o    en
                           o    CM
                           o    r-
                                          o
                                          o
                                          m
   CO
   Ed

   H


   H
   CO


   a
   Ed
   H

   M
   CT\
   eg
                                          o o o o o o
                                          o o o o o o
                                          « o o o o o

                                          •T o m o 
-------




0)
3
c


•u
C
O
o

o
CM
to
w
H



«<•<
Oi
0
a
H
H
CO
tj

CM
Z
o
H

0
3


00
w
HH
oi
w
Z
1— 1


&4

F/^
w
H
H
to

a
[d
H
Z



ON
TABLE
p
1
P*
<




" S
2 "3

•
u
g
a

U
§ 0
-4 1 O
u n -H
O 0 4J
3 -1 <»

O4*^
a, § v
< g
^4

X X
A "a
*







T3
n
*j
W
O
It
0
o

•o
2
o

o
_£














Company and Location
0
0







m
00









• O 0
• O 0

• «  O
tH .-* 00 O Og

f-t CM
*n






n
v n
<-i it
M -i
<• *J U
o gs a g
i £ S .8

Calcasleu Refining Co. — La
Calumet Refining Co. — Prin
Canal Refining Co. — Church
Citgo Petroleum Corp. — Lak
Clairborne Gasoline Co. — L
0
* o
. Ol
. 00
CM





. . ^
• - r*»
H











. .

V.

o o o
o o o
•ON O>
* M






O O O
O O O
o n o








o oo
o o o
m oo o
to f» m
in m
*-4 -f







»
«
O X
O V
H 5
Conoco Inc. — Lake Charles
Cotton Valley Refinery (Ke
Refining Corp.) — Cotton
Exxon Co. — Baton Rouge
. 0
. 0 • • •
.in ...
CM
















. o « - O
. o . - - o
. ,H . . O
• in ••• r»



O 'OOOOOO O
o * oooooo o
oo -ooocncnooi CM
* CM *H «H |H f-l
*





OOOOO O O
OOOOO O O
in o -» o o o o
r-  a u
-^ O M -H
•° 9 °
g i ^
Solvent extraction: *7,10
MICHIGAN
Crystal Refining Co. — Cars
Lakeside Refining CO.— Kal
Marathon Petroleum Co. — De
Total Petrol eum . — Alma
66
-------





X™*.
•o
0)
3


•H

C
o
Jf,

o
cs
CO

M
H
M
U
^
PL,
U
Q
H

H
ft*
Vl
OQ
z
o
H
o
3
HH


co
M
Q*
w
'Z
M
fo
oj


CO
H
H
CO

a
u
H

z
D

.
3
H


w

J3
V)
<






3

w
o
(t
Q
el

B
0 B
-I I 0
4J W ~4
U U U
3 -4 It
-OWN
0 <» -4
tt, O «

n


>» X
3£
^ PM





2
> 43
H
0
o
•3
E
u
•o
o
J3














B
O
H
V
It
o
2
1

S
u
O 0
0 0
0 0
*T m
H en
















o
o
o
a

»..


ooooo
o m o o o
e> en m H rv
m oo H en




O 0
eN 0
CM o
e* o"

""*



en o
•» o
•H O
r-~ ^
*6 m








M
el
a.
3
a, w
Ul |
i n
0 Ǥ
i
3 0
« U
"g 00

5 * *
8 •§ 3
u el
123
£ 5 3
• o o o o ooo •
• o o o o ooo •
• O O *n «H O ut O •
• O (N CO iT) V0 \£ <*1 -
N fH






*









. 0 ' • - 0 O
. o • • - 00
• in • • • o •*
• in • M «n

V.


0 0 - • • OOO O 0
o o • • • ooo o ui
«o P4 * • • oor*- *ft I NO-V %O*O
M IH •* >n -»




ooooo ooo oo
ooooo ooo oo
otnomr-i o co  il 0 « Jj U
& « -4 jo q o
3 0. > S N 3 1 ,
a, i i 3 n to i
1 1 1 J 00
i • • i B a w o
• o o i -4 d 3 u
&fi 6 • H 6 0
M M O -1 -J 1 00
0 0 -1 -4 1 B
U • » V iH ID -4 • -4
4 Ufl

a> ii • H > 3 o i «
H CU 03 3 *H T) » It C • - Oi
OM 1 DBM .JHO1-)
« M it e o; S « < i o *
M in -o o -4-0 z 10 « c
JH » M C -C C < XUB BO4Jel Ei VOO -^ u
Ul ««e*3(/l Z BBX X a
O M SXMO Q VOX HO
X ^UWV3 £ UOU PhX
O O
o o - -
O • 0 • •
in oo
en en



. • o .

. . . m •
' ' ' r»T '
.








* O O O • *
•ooo
* o o o
* o 10 m * *
H es

*

O O O ' O O O O
O O O ' O O O O
m m o • o m m o
CN •* m • en eN to m




o o o o o
o o o o o
r~ o O 0 o gg
.» m m o -»
00 ej> HO
H H



O OOOOO
o ooooo
m ooooo
< 4- o o o o m
Q 00 e> o O -T
CU H H
•"
"2
*
o
o
m
OS H
** H

T) 00 X B U
«1 B 0 V et

J3 -OBI 0
-C it ^ i u
O- « V O
o o. cti £ o o -C
O O *J O M H
B u U 01
^ o K i> H -a i
H 6-1 0 B. -4 M •
1 04 1 O HO
1 II 3 B
1 • V ft H
B O o B CK
o u a S H B i E
H 14 M 0 V 1 3
«j u o eu TJ ' «
O S O Ban
It H < V H M O
u B "l • H J O u
4J *4 n tfl eo i O *J
X IH >4 V * el I V
v vbiaaw -HO,
QEt eyi i OH
u o! it C H O O S
C •< « 3 -0 0 < el U >•! OH>
H > > 3 V Vel XJJit
O U V W fi A O XOtl
Ul Z Z Z < UU UZU1
67
-------





•v


M
H
M
O

p-l
a
1
Cfl

CQ
0
H

^4
O
o
<«
u
c
0 B
•H 1 O
4J 4A 
Be
•< §

X X
3 £
X 0.





•e

4J
U
a '
^
u
•5
S
u

u
^3










g
~4
«J
«
u
2
1
X

"

o o o c
0 00 C
m oo c

en CM r* *
fH

O O
0 0
•H O
CM" CM"









o • o o o • o o o
o -000*00 o
o • -* -n *o • in ** m
•49tNfH en*
fH fH fH fH




iH
•e
~V T)
•H a
w -H —
§ a e « a
o 3 « o c -3 o :
•H fH f4 W g J O
to >H « e -H i o
1 44J • 3 0 1
1 O 4J 13 O C 0 fH
• 1 U 0 t -H -H fH X
0 1 < 4J 1 U X -H 4J
u • i 4 " • i O -H (
O 1 1 -4 O 1 U
» C • -o-i u • IH
C M O S -1 U 0 C *
•HO 5-* 9s ° o-
c « x 3 3 M • -H e
*4UU II V O fH 4J 0-
•H-413 II .-1 • 0 O 0 O.'J
1} M -H - O •< -^ (4J 1
a* «j e •< o o M . -i « M ic
«-4SoeS 4JHC
>HOfH>C 0 > £ o XHji «3 HC -^ H c
-i-i«oS-ia3«i£u 3 o 2 o
zoauzz O 0 0 0
900 0 0
> m oo o m

1 CN -» HJ fH
r-i

O
O
m
r-









O 0 0 0
O 0 0 O
in o in > « J3 -O
t VI C
Q O (I
BO r-4
• CD *Q O U
3. 01 U . u
L4 -4 •< HI 0
93 1  O >
-4 B e 4J -H U V
II •< 3 0 O fi J3
.. 
-------



V

*J
o
<*
0
4
U

g
•H
P
§
1
04








































*J
X
1


VI
«
-D
3





B
i S
o u
•H a
U N
s u
§ *
s i
VI


X X
j? •?
2 0
^ p*
*







n
> *j£
u
4
a
4
O
«
2
u

u
^










and Location
x


o
o
o
HI
CM













• • O O
• • O O
• • o <*>
» « •* r*




0 O
O O
CM O
00 O
CM




OOO
OOO
OOO
O !•* O
«-4 rH (^




OOO
OOO
O O H
m" oo •*"
CM >0 r*
rH rH r*






1
t
ining & Marketing Corp.
ia
-Marcus Hook
A. Inc . —Philadelphia
•< "M A 1 •
M «l Q, • 03
1 Hfg
i 3I*>
w w o. .a
04 •< pa O
. . . . o
. . . . o
. . . . o
. . . . oo


o o o o
»n in O O
9> f* O O
en ^ CM o *







. o •
. . . o
. . . (*1
• • • in •




• • • o o o
• • • 0 O 0
• • • O O CM
• CM en N
!-< *




O O OOO
O O OOO
O  00 O O
o n to  o

f-t








ie Division,
leal Co. --Bradford
ducts Co. — Rouseville
Oil Refining Corp. —
lley
us Hook
ing Co. — Warren
JS U 4J > kl *4
J3 04 0 » n CM
•» CM



O
0
o
c«C
CM




O
o
o
m"
*




o
0

0

.»








v4
U
VI
*
X
«
K
1
§
.H
V4"
O
Amoco
0 0
0 O
m m
in in


O 0
o o
-* CM
10 »T







oooooininoo
000000*000
tHCMrHCMCl^CMCMr^
rt



OOOOOO
OOOOOO
o 10 10 o in 9>
9i r* tH in m 10
<1 « rt




o o o o
o o o o
o o o g o
o in o oo
CM -. -.




00 00
O 0 00

00* o* *O *Q
r* f* P^ o
CM rH «





4H CM
-4 B
*J O
VI 4J
hfield Co. — Houston
roleum Co. — Corpus Chri
rnational Oil Co. — Hous
A. Inc. — El Paso
r
U u « • 3
x £, e " w
1-1 S U
us <
** rf "
BO* 4J M U
II S 55
*iS 66
o
o
o
o
d













OOO
OOO
OOO
•H 1^ CM
rH



O O
O O
00 O
CM" ^
H




o
o
1 °.

2




0 O
0 O
o o
m" o"
o\ o
r^






C
0
•J
es Petroleum Co. —
isti
1 Petroleum Corp. — Hous
* JS H
in ° B
VI V
-13 U
n~
00



















•»
VI
u
_>
06
V
0)
g
69
-------




•o
0)
c


c
o
u


o
CO
cd
M
H
H
U
 -o
o
4
Q
O
•O
2

"8
U
ja












pany and Location
§
u

o o
o o
o *rt
"^ *


o
en

en










-






0 0
0 O
o o
o\ \o
* O Q. O
fS M tTu
'H ao *j o
-4 fc V fj *O
0 II ft, VO
• A «3 . x
T:,^0^ g
O -J U -J -J V
-4 3 4 •* ~l 9
O O1 U J3 -J UJ
.5.13££
o
o

*


o
o

•H








00 • O
o o • o
fH 00 • fH
o\ r* . 0*
fH




• o
• r^





o o
o o
o o
10 0
* M





O O
RS
01 00
H -^
« -i
06 O
« _<
•o ^
CL, M


















O

en






o
o
o
en





o
o
oT






o
o
«
00
CM











4
«1
S


















o
o
m
«





g§
O CM
HT HT
*




O
O
O
CO*
fH





O
o
o
HT
O
*H









|
stem Refining Cl
s Christ!
1 &
JJ 0
3 U
0





















































I
1

i
M
S
Refining I Market
0
o
4
X
o o
o o
o "-ft

«fffT (^
tH
o o
0 0
00 r-t

r-f








0 00
O O 0
in O 00
*r f-4





o o o o o o
o o o o o o
in in O O| C4 •<*
r-l OV  at ue
OH
t> 1
Z 1 *"%
SO
Arthur and Port 1
ity Refining Inc.
Corp . — (Beaumnont
4 U
O. 4J -1
U « 4
-1 O 4 U
ti] O, X 0
• a
l-> 3
O
O

f^
fH






















O O
O 0
en in
oo eg





o
o
o
fH







i
SB










U
5
n
3
Refining Co.-Corpi
0
M

                               o
                               o
                               o
                                        «t  o as in
                                         »   » Z  -
                                                  O O

                                               O O O

                                                  o o
                               o
                               §
                                             O pi  O <
                               5   g
70
-------





•5"
s
C

4J

O



o
CM
00


rH
H
M
u

4-1
U


u

c
^
4-1
O
|
PL,




















































U
.-t
4
•§.
1

^
3









1 0

U 4J
•H 4
4J N
O V
•< 1
n






-1 O
< a.








J
•o
n
5s
u
4
O
4
U
g>
•O
2
U 1

o
•»















«c
o
•M
4J
4
U
5
1
i!
!


* o • • o o o o o
• o • • o o o o o
• o • • o o 10 r*>. o
• u"» • • vo oo en en in
fH



	 • en t-t i-l •











. . . o • . o . - -
. . . o • . o • • •
. . • ov . _ . m • - •
• • • CM • " . r* . - .



• -




o 00*00 o oooooo
o oo-oo o omoomo
-&r o* r-t • *o 1-1 r* r-. r^ •« o r* ao
N HI  c 3 -^ ixr oc • BcjitHji
i 4 ^-t « o w i- E a Bio*
S • M -a u H j< O.H B w xco-ti
O O>6r>lrl CJ •••> t rIV-rlUII
4J UIUOI4  3 O*-t*Hf-l*-4
O -4COOOC^« U Z -^X VI U O*l *4-*4 QA
Z Z u 0 0 0 0 C > H« Z X 2 « e 4 4 S
M 0 r-t Crlr-t.-4'r*U4 V U 3 • O J2 r-t O3r-4r>^ *4
U U S3 4 > -1 — : C 4 C • > H JtOU U P- S U " S y §
S o vi --t«,oii3K<< v> -i M 4zw » ri o o « u 3 E S
r-l B ^1 JJ^OJ1>OV • 0 W 3 I-* 3 >* S**4-TtQTt X
> •< 3 -t
•1 ^ S -H
4 _, 4 -J
b] (Q - >
O OS)
Z - - -4 O
• • -4 E
O U — 1 V
O fi U rl
E M 4 S
O QA E 4
•H 3 M < i-I
U -1C
4 B -4 - -
N -1 B • •
•rl *H -4 O U
M «l <44 B B
tO! « 1-4 M
M O O
r-l t> « U U
O U rl 4 4
PU V) ^ K X
•X V 4 U 9)
PL, 01 H H

X ....
4 00 91 0 rH
•O r-l r-l
§

M
W
ffl

£
Cl*
^z
VI

4
a
t

u

•o
X 0
•o - -o -i? -o
•o o o
H — . — . 0 —.
4 O A O .O
T-) .a .0
B O TJ - O
« o o a o o
•-I O M — (N r*
4 00 - JO 1
O - -T 1  -4 0 Z 1
U 1 1 O I-l 1
CD 1 1 (N H
Q. < Q - - Z
o* 2 m »
•A r4 B
-4 . . I X U
41 S 41 1 4 E
rl 8 rl O 3 -4
U T) 0 0 0
4 41 E -a,
ta it -4 - u
U U 4 Db — t *
tO O •• U
00 S (3 • X E
a> 3 M o u M
1-4 W 41 U 01
U r-4 • E ao
X i-l O < » -4 E
rl pi U • B -) bl -O B 41 3 »
S B O pi  X 13 0
> J3 « rl 4 E
n t-t n .fi 4 -4 v
< H •< U O O ^
'1 ....
4 H-t i-l «N *O vfl P-.
71
-------
                                              1!
                                             .2
                                             JQ
                                             CO
                                             TJ
                                             CD
                                             0)
                                             c
                                             (0
                                             

Li.
72
-------
     The specific processes which are potential sources of benzene emissions
at petroleum refineries are listed in Table 10.

     Complete description of the various operations at petroleum refineries
is beyond the scope of this report.  The reader is referred to
References 21, 22, and 23 for descriptions of refining technology for
detailed information.

Emissions of Benzene from Petroleum Refineries

     Emissions of benzene from the processes listed in Table 10 are fugitive
                                                   21
emissions, or emissions from catalyst regeneration.    No emission factors
                          *
were found specifically for benzene emissions from these potential sources.
Table 11 presents estimated concentrations of benzene in nonmethane
hydrocarbon fugitive emissions for some refinery processes.  Such data were
not available for each of the processes listed in Table 10.  The data in the
table show that, for example, from the crude distillation unit, about
74 percent of the fugitive emissions from the unit may be attributed to the
crude oil and that benzene constitutes 46 ppm (weight) of that stream.

     The nonmethane hydrocarbon fugitive emissions from refineries occur
from valves, pumps, drains, flanges, relief valves, and compressors.
Emission factors for total nonmethane hydrocarbons from these sources are
shown in Table 12.  No emission factors were found for benzene or
nonmethane hydrocarbon emissions from catalyst regeneration.

     Some older literature has presented estimated benzene emission factors
                              26
for entire refinery complexes.    These factors were based on an assumed
benzene content of 0.1 to 1 percent, in the various fugitive emissions and  an
assumed "very low" benzene content in catalytic cracker regenerator
emissions.  For an uncontrolled refinery, the estimated emission factor is
1 kg benzene/1000 barrel crude feed.  For controlled refineries, the
                                                                    f\ £
estimated emission factor is 0.19 kg benzene/1000 barrel crude feed.
                                       73
-------
TABLE 10.  POTENTIAL SOURCES OF BENZENE EMISSIONS
           AND PETROLEUM REFINERIES21
    A  Crude storage
    B  Desalting
    C  Atmospheric distillation
    D  Vacuum distillation
    E  Naphtha hydrodesulfurization
    F  Catalytic reforming
    G  Light hydrocarbon storage and blending
    H  Kerosene hydrodesulfurization
    I  Gas oil hydrodesulfurization
    J  Fluid bed catalytic cracking
    K  Moving bed catalytic cracking
    L  Catalytic hydrocracking
    M  Middle distillate storage and blending
    N  Lube oil hydrodesulfurization
    0  Deasphalting
    P  Residual oil hydrodesulfurization
    Q  Visbreaking
    R  Coking
    S  Lube oil processing
    T  Asphalt blowing
    U  Heavy hydrocarbon storage and blending
    V  Wastewater treating
                        74
-------
  TABLE 11.  ESTIMATED CONCENTRATIONS OF BENZENE IN NONMETHANE HYDROCARBON

             FUGITIVE EMISSIONS FOR SELECTED REFINING PROCESSES21
   Process/Unit
Process Streams (Percent of
 Total Fugitive Emissions
   Attributed to Stream)
     Benzene
  Concentration
of Stream (ppmw)'
Crude distillation
Aromatics extraction
Delayed coking

Fluid catalytic
cracking

Hydrotreating
Catalytic reforming
Hydrogen production
 crude oil (74)
 straight run naphtha (24)

 reformate (12)
 aromatic extract (44)
 raffinate (44)

 cracked naphtha (57)

 cracked naphtha (45)
 straight run naphtha (47)
 desulfurized naphtha (47)

 desulfurized naphtha (47)
 reformate (47)

 straight run naphtha (19)
        46
        59

       648
     7,850
     1,642

     1,296
       119
       119

       119
     2,538

        48
 Parts per million, by weight.
                                       75
-------
   TABLE 12.  EMISSION FACTORS FOR NONMETHANE HYDROCARBONS AND ESTIMATED

              BENZENE FRACTION AT PETROLEUM REFINERIES21'24'25

Emission Source
Valves



Open end valves
Flanges
Pump seals

Compressor seals

Drains
Pressure relief valves
Process
Stream Type
gas/vapor
light liquid
heavy liquid
hydrogen
all
all
light liquid
heavy liquid
gas/vapor
hydrogen
all
gas/vapor
Uncontrolled
Emission
Factors
( Kg/day - s our c e )
. 0.64
0.26
0.005
0.20
0.05
0.0061
2.7
0.50
15
1.2
0.76
3.9
Stream
Composition
(Weight
Percent Benzene)
b
b
b
b
b
0.10
0.50
0.50
b
b
2.4
b
These data were not available for all streams.   The data that were available
were included in the interest of presenting as much data as possible.

Species data for oil/gas production indicate possible benzene content of
0.10 weight percent in "unclassified fugitive emissions."  These species
data could be used to approximate benzene factor emissions.
                                     76
-------
REFERENCES FOR SECTION 4
 1.  Chemical Products Synopsis.  Benzene.   Mannsville Chemical Products.
     Cortland, NY.  1985.

 2.  Chemical and Engineering News.   Key Chemicals:   Benzene.
     October 24, 1983.

 3.  Purcell, W. P.  Benzene (In).  Kirk Othmer Encyclopedia of Chemical
     Technology.  Volume 3.  John Wiley and Sons.   NY.  1978.

 4.  U. S. Environmental Protection Agency.  Materials Balance for Benzene
     Level II.  EPA-560/13-80-009.  Washington, D.C.  1980.  pp. 2-6 - 2-34.

 *.  U. S. Environmental Protection Agency.  The Environmental Catalog of
     Industrial Processes.  Volume 1 - Oil/Gas Production, Petroleum
     Refining, Carbon Black and Basic Petrochemicals.  EPA-600/2-76-051a.
     Research Triangle Park, NC.  1976.

 6.  U. S. Environmental Protection Agency.  Evaluation of Benzene - Related
     Petroleum Process Operations.  EPA-450/3-79-022.  Research Triangle
     Park, NC.  1978.

 7.  SRI International.  1987 Directory of Chemical Producers.  Menlo Park,
     CA.  1987.

 8.  Otani, S.  Benzene, Xylene Bonanza from Less-Priced Aromatics.
     Chemical Engineering.  77(16):118-120.  1970.

 9.  U. S. Environmental Protection Agency.  Organic Chemical Manufacturing.
     Volume 9:  Selected Processes.   Report 3.  Ethylene.
     EPA-450/3-80-028d.  Research Triangle Park, NC.  1980.

10.  Sangal, M. L., K. M. Murad, R.  K. Niyogi, and K. K. Bhattachanyya.
     Production of Aromatics from Petroleum Sources.  Journal of Scientific
     Industrial Research.  31(5):260-264.  1972.

11.  Sittig, M.  Aromatic Hydrocarbon Manufacture and Technology.  Noyes
     Data Company.  1976.

12.  Telecon.  P. Cruse, Radian Corporation with L. Evans, U. S.
     Environmental Protection Agency.  Research Triangle Park, NC.
     January 22, 1988.

13.  U. S. Environmental Protection Agency.  Evaluation of the Efficiency of
     Industrial Flares:  Flare Head Design and Gas Composition.
     EPA-600/2-85-106.  Research Triangle Park, NC.   1985.

14.  American  Iron and  Steel Institute.  Letter and attachments from
     E. F. Young, Jr.,  AISI, to W. B. Kuykendal, U. S. Environmental
     Protection Agency.  December 16, 1987.
                                      77
-------
15.   U.  S. Environmental Protection Agency.   Coke Oven Emissions from
     Wet-Coal Charged By-Product Coke Oven Batteries - Background
     Information for Proposed Standards.   Draft EIS.  EPA-450/3-85-028a.
     December 1985.

16.   U.  S. Environmental Protection Agency.   Benzene Emissions from Coke
     By-Product Recovery Plants - Background Information for Proposed
     Standards.  EPA-450/3-83-016a.  Research Triangle Park, NC.  1984.

17.   U.  S. Environmental Protection Agency.   Environmental Assessment of
     Coke By-Product Recovery Plants.  EPA-600/2-79-016.  Research Triangle
     Park, NC.  1979.

18.   Dufallo, J. M., D. C. Spence, and W. A. Schwartz.  Modified Litol
     Process for Benzene Production.  Chemical Engineering Progress.
     77(l):56-62.  1981.

19;   U.  S. Environmental Protection Agency.   Assessment of Human Exposures
     to Atmospheric Benzene.  EPA-450/3-78-031.  Research Triangle Park,  NC.
     1978.

20.   Cantrell, A.  Annual Refining Survey.  Oil and Gas Journal.
     84(12):100-115.  1986.

21.   U.  S. Environmental Protection Agency.   Assessment of Atmospheric
     Emissions from Petroleum Refineries.  Volume 1.  Final Report.
     EPA-600/2-80-075a.  Research Triangle Park, NC.  1980.

22.   U.  S. Environmental Protection Agency.   Assessment of Atmospheric
     Emissions from Petroleum Refineries.  Volume 5.  Appendix F.  Refinery
     Technology Characterization.  EPA-600/2-80-075e.  Research Triangle
     Park, NC.  1980.

23.   U.  S. Environmental Protection Agency.   Assessment of Atmospheric
     Emissions from Petroleum Refineries.  Volume 3.  Appendix B.
     EPA-600/2-80-075C.  Research Triangle Park, NC.  1980.

24.   U.  S. Environmental Protection Agency.   Compilation of Air Pollutant
     Emission Factors.  Third Edition (Including Supplements 1-15 and
     Updates).  Publication No. AP-42.  Research Triangle Park, NC.  1977.

25.   U.  S. Environmental Protection Agency.   Air Emissions Species Manual,
     Volume I.  Volatile Organic Compound (VOC) Species Profiles.  Draft
     Report.  October 1987.

26.   U.  S. Environmental Protection Agency.   Atmospheric Benzene Emissions.
     EPA-450/3-77-029.  OAQPS.  Research Triangle Park, NC.  October 1977.

27.   U.  S. Environmental Protection Agency.   Docket No. A-79-16.
     Item IV B-7.  Letter to Ms. Gail Lacy,  U. S. Environmental Protection
     Agency, from Mr. David Coy, RTI.  Subject:  Benzene Emissions from
     Foundry Coke Plants.  March 11, 1985.
                                      78
-------
-------
                                  SECTION 5
                   EMISSIONS FROM INDUSTRIES USING BENZENE

     Emission sources related to benzene's use as a feedstock and as a
solvent are described in this section.   Descriptions of benzene emissions
from gasoline marketing operations and mobile sources are also included.
Emission sources are identified and emission factors are presented as
available.  The reader is advised to contact the specific source in question
to verify the nature of the process, production volume, and control
techniques used before applying any of the emission factors presented in
this report.

     The largest portion of the total amount of benzene produced is used in
production of ethylbenzene/styrene.  Other major chemicals for which benzene
is used as a feedstock include cyclohexane, cumene, phenol, nitrobenzene,
and linear alkylbenzene.

     Although benzene has been used in the production of maleic anhydride,
the process is no longer used.  All capacity for producing maleic anhydride
in the United States is now n-butane based.  The last facility using the
benzene process has been placed on standby, with no stated intent of
resuming operations.  However, a brief description of the benzene process
for maleic anhydride production is included in this section for reference.

ETHYLBENZENE PRODUCTION

     Ethylbenzene is a liquid at standard conditions, having a boiling point
of 136 C and a vapor pressure of 1284 Pa.   About 50 percent of the United
States' production of benzene is used to produce ethylbenzene.  The
ethylbenzene industry is closely tied to the styrene industry since styrene
is produced only from ethylbenzene.  There can be approximately a
                                                                          2
0.3 percent by weight carry-over of benzene into ethylbenzene and styrene.
                                      79
-------
Additionally, some benzene is reformed in the production of styrene.
Ethylbenzene production processes and uses thereby constitute a major
potential source of benzene emissions, particularly since styrene is
anticipated to show good growth through the 1980's.  Due to increased
process efficiency, ethylbenzene growth is anticipated to be slightly less
(about 3 percent per year).

     Ethylbenzene is used almost exclusively to produce styrene.  Some
ethylbenzene is used as a solvent, often replacing xylene,  and in the
                        4
production of some dyes.   A total ethylbenzene production capacity of
4308 million kg/year currently exists.   Approximately 95 percent of this is
based on benzene alkylation with the remainder based on extraction from
mixed xylene streams.  Most styrene is produced by two methods:
hydrogenation of ethylbenzene (89 percent) and peroxidation of ethylbenzene
with subsequent hydration (11 percent).  The latter process can also
co-produce propylene oxide.   A third process, converting ethylbenzene
isothermally to styrene, has been recently developed in Europe.  To date, no
United States facilities report using this method.

     Another method which co-produces both ethylbenzene and styrene has been
patented.   In this process, toluene and light alkanes other than ethane are
reacted at 1000 to 1200 C and then gradually cooled to produce an 80 percent
ethylbenzene/12 percent styrene product with a mass of about 25 percent by
weight of the toluene reactant.  These products can be separated by
distillation and the ethylbenzene either recycled, sold, or converted to
styrene by another process--dehydrogenation or peroxidation.  This process
is not reported to be in use at this time.

     Table 13 lists United States producers of ethylbenzene and styrene.
Most facilities produce both ethylbenzene and styrene on-site, thus reducing
shipping and storage.  Only two styrene production sites do not have an
ethylbenzene production capacity.  Ethylbenzene from mixed xylene separation
is generally shipped or supplemented with another ethylbenzene source for
styrene production.  Only one site uses the peroxidation process to produce
styrene.  Table 13 also gives the latest  facility capacity.
                                      80
-------









3
1
H
H
i— i
cj
^
OH
^
U

Q
z"
O
HH
H
^
o
Q
,_3

..
CO
erf
w
u
P
Q
O
03
P_|

U
z
u

^
H
CO
a
^
u
N
Z
U
3
^"*
n:
H
w
CO
u
£— I
<^
H
CO

Q
Cd
H
M
Z
P
CO
r-l
U
h4
03
^
H








































































































5


Ul
Ul
0)
O
O
1-1
P-I








CU
c
(U
N
C
Oi
XI
1— 1
^
x:
4-1
w









w
XI
4J
•rl C
0 0

O, r-l
cd r-i

s









Ul
Ul
(U
o
0
V-l
OH










fi
0
•r4
4J
cd
O

J












^
c
cd
ex
a
o
CJ





td o
i O ' ' O
i O i " O
i  i i CM
1-1






i i i
i CJ i i Q
i i i





cd xi xi o
me o o o
CO I r-l O O
O i r-l in CM
V r-l r-l














I
<; i pq <; •<
i










< XX
J H H
X 0) X H 3e~
H 00 H 3 0)
3 X! -n
O - 4J >
4J Oi C rl r-l
H 0 <  -n
rl C
M O- O
0) CO -U
> 01
cd oo 3
0) T-I O
03 03 3B












j_j
C  cd cd o o xi cd
r-l O O ' O O O O
so co in i in oo in o
i— i p*« co i co **o r^
1-1 r-l r-l
^












CQ
I
o3 
tn cu 4J - H cu H -ri
Cd r-l )-l T3 E CJ
rH i— 1 O C cd
CU-rl &,Cd Cd>-3 Ct/1
3 > CU r-l W T-I CO
C )-i cu T3 cu > X
Cd Cd !4 -rl "O 4J 1— 1 0)
CUUtSOco Ul
S w 3 cu C
o o o r-i jr o
U U Q Cd U S
81
-------




y~ s
T3
CU
3
C
•r-l
4J
C
o
CJ
^
vO
•>
CO
••
l-l
^H
£H
1— 1
CJ
0

Q
<;

»
Z
0
r-l
EH
^J
CJ
3

00
cd
Cd
o
D
a
o
OS
CM

Cd
Z
Cd
EH
CO
a
Cd
z
Cd
N
Z
Cd
3
£M
3C
EH
Cd

co
Cd
EH

H
oo
a
Cd
H
J~ H
z
^
.
ro
r-H
Cd
J
CQ
^J
H













a>
C
CU

^.
4J
OO








01
XI
4-1

O cj
•H cd
rH CX
rH Cd
•rl CJ
as






01
01
CU
0
O
!H
CH









* CU
C
cu
N
C
cu
XI
^
r!
4-1
Cd









^
01
XI
4J """'
•n C
0 O

CXrH
Cd rH
CJ -iH
5^
**^s






Ul
U)
CU
u
o
^
CM











C
o
•rl
4-1
cd
CJ

^J












j^
C
cd
CX
E
O







n
00
ON













U







X)
in
00
a\














^












X
£.4
fc
>.
4-1
•iH
CJ

0]

l/^
VO






X
H

_
•H
jj
w
•r-t
^4
x:


01
3
CX
1-1
o







rH
•H
O

x:
0
o

s^x

c
3
CO
        x>
          CTi
                   O
                   CN
   O
   Ul
  CQ
  Cd
          c_)
XI
  U"!
  CO
ON
CO
CO
   CU
   4J
   4-1
   CU
   S
  rH
   Cd


  O
           X
           H
 4-1
UH
•rl
 1-1
•a
 cd
 co
CO
                   z
 c
 o
4J
 Ul

 o
   o
   cj
   CU
   C
   c
   cu
   H
 CO
•o
•rl
XI
 ±t
 id
CJ

 C
 o

'c
D
 C
 o
•H
 4J
 id
 M
 o
 (X
 IH
 O
O

X
CO
D
                              C
                              o
                             •H
                             4J
                              CJ


                             T3
                              O
                              W
                             CM


                              CO

                              CO

                              >
                             4-1
                             OO
                            O
                           •rl
                           4-1
                            cd
                            c
                            cu
                            bO
                            O
                           CQ
                           Cd

                            n

                           u
                                              c
                                              o
                                              cd

                                              •o
                                              CO
                                              Q
                           C
                           cd


                           o
                           •H
                           4-1
                           cd
                                    O
                                    IH
                                    co
                           ca
                           Cd

                            1

                           Q
                              c
                              o

                              4-1
                              O

                              T3
                              O
                              cu

                              CU
                              N

                              cu
                              ca
 c
 o
•H
 4J
 cd
 s
 CO
 N
 C
 cu
09
 C
 O
•H
 4J
 Cd

 cd
 cx

oo

 H
<4H

«
C
o
•r-l
4-1
0
3
•0
O

a

r-
r-
ON
rH
XI































































































.
co

cu
CJ
C
CU
J-t
cu
<4H
CU
B*
CJ
CU
)H U)
cd 4J cd cd
C
01 CU 01 tJ 4J
4J J-l -rl CU C
C IH C cu
cd 3 ^-H n
rH CJ 4J E !H
CX -r4 ^ 3
bOi-H cu O
- C -iH 4J
01 -rl fj 0) CU
CO U Cd T) S-4
bO rH (4H Id
C 3 CU
cd Ul C X) m
X C cu C
0 0 > T3 0
U w4 rH -iH
CX bO 3 4J
•r* X O Cd
x: xi >ix: o
01 C M O
>H 01 Id i— 1
CO OJ -0
§-r< E C T3
4J o cd c
O -iH (-1 Cd
rH <4H -
>-, -iH Ul Ul
4-1 O M 0) l-l
•r) Id C ^ 0)
rH 14H O 3 0
••H -rl Ul 3
CJ J-l 01 Cd T)
id id ui cu o
 C 4-1 CU
•iH CU rH 3 X:
TJ 4J 0> CX EH
c ui > x:
O >rl CU bO
0 X rH 3 •
CU O i-H
4_> CU !^ CU
co cu x: x: c
,M XI EH 4J C
S-l 4J O
cd - ui
E >!•>-, IH
MH ui 4J co
01 -iH (U •,-! CU
cd n > o
CU rH Cd 4->
CU > cu CX C
bO ui id cd
C TJ E 0 r-l
cd I-H cu cx
x: 3 x: ui
fj o ^> <9 j:
x: w
o ui 01 x: --H
4J 4J CJ 3
SH C 3
4J  0 0
bO • "O CU rH
C U C <4H )H
••H 4J Id O -r4 >-
4-1 CU -O >H
01 01 C Id
•rH - bO O x: 3
I-H t) C -ri bO C
CU -iH 4J 3 Id
Ul 01 4J U O •-)
•H O 01 C IH
X! rH -H 3 X! "4H
H a •— I 4H 4J O


. •
CU
4-1
o
z
                         82
-------
Process Description:  Ethvlbenzene and Stvrene Production Using Benzene
Alkylation and Ethylbenzene Dehydrogenation

     Since most ethylbenzene production is integrated with the
dehydrogenation process to produce styrene, these processes are described
together.  The primary reactions are (1) catalytic alkylation of benzene
with ethylene to produce ethylbenzene, and (2) catalytic dehydrogenation of
ethylbenzene to produce styrene.  A detailed process flow diagram is shown
in Figure 11.  The process steps are described briefly as follows.

     First, both fresh feed and recycled benzene are dried to remove water.
Dry benzene  (Stream 1) and ethylene are fed continuously into an alkylation
reactor.  A granular aluminum chloride catalyst is also fed at a constant
rate.  A temperature of 95 C is maintained by cooling water.  The reactor
effluent, which contains benzene, ethylbenzene, and an insoluble catalyst
complex, goes to a settler where the crude ethylbenzene is decanted.  The
heavy catalyst complex is recycled to the reactor.  Reactor vent gas (A) is
routed to a condenser and scrubbers to recover any aromatics and to remove
hydrogen chloride.

     Pressure in the reactor can range from near ambient to about 2700 kPa
and temperature may range from 80 to 400 C.  One process uses a high
pressure reactor with high-purity ethylene.  In this process, inert gases
are removed  in a degassing step when the pressure on the alkylate effluent
is removed.  Another process (UOP or Mobil/Badger) uses a solid support
catalyst with low-purity (5-10 percent) ethylene.

     Next, crude ethylbenzene from the settler (Stream 2) is washed with
water and neutralized with a caustic solution.  The crude ethylbenzene
contains 40  to 55 percent benzene, 10 to 20 percent polyethylbenzene (PEB),
and high-boiling point materials.  Benzene is removed in the benzene
recovery column and recycled (Stream 3).  Next, the product ethylbenzene
(Stream 4) is separated from the remaining polyethylbenzene and high-boiling
point materials.  These are then distilled and the PEB recycled from the
resultant residue oil.
                                      83
-------
                                    X
                                    CD
                                   (0
                                   CO
                                   CD
                                   O
                                   — .2
                                   £ CO
                                   ** ^2
                                   o £
                                   C CD
                                   O o»
                                   CO C
 ,

                                         l«
                                         CT>,

                                         O (D

                                         II
O
O
                                            C
                                            CO
                                         3
                                         O)
84
-------
     The purified, ethylbenzene is preheated in a heat exchanger.   The
resultant vapor (Stream 7) is then mixed continuously with steam at 710 C in
the dehydrogenation reactor which contains one of several catalysts.  The
reaction product (Stream 8) then exits through the heat exchanger.  It is
then further cooled in a condenser where water and crude styrene vapors are
condensed.  The hydrogen-rich process gas is recovered and used as a fuel
(Stream 12) and the process water is purified in a stripper and recycled to
the boiler.  The remaining crude styrene liquid (Stream 11) goes to a
storage tank.  Benzene and toluene (Stream 13) are removed from the crude
styrene in the benzene/toluene column.  They are then typically separated by
distillation.  The toluene is sold and the benzene is returned to
ethylbenzene production section (Stream 15), or it may also be sold.  Next,
the ethylbenzene column removes ethylbenzene which is directly recycled
(Stream 6).  Tars are removed and the product styrene emerges from the
styrene finishing column.  In some facilities, an ethylbenzene/benzene/
toluene stream is separated from the crude styrene initially and then is
processed separately.

Process Description:  Ethylbenzene from Mixed Xylenes

     Ethylbenzene can also be extracted from mixed xylene streams.
Proportionately, however, very little ethylbenzene is produced in this
fashion.  The two major sources of ethylbenzene containing xylenes are
(1) catalytic reformats from refineries, and (2) pyrolysis gasoline from
ethylene production.  The amount of ethylbenzene available is dependent on
upstream production variables.  The ethylene separation occurs downstream of
the benzene production.  For this reason, the ethylbenzene produced by this
process is not considered a source of benzene emissions.  Instead, benzene
emissions from the entire process train are considered as emissions from
benzene production and are included elsewhere in this report (Section 4).

     When combined with the dehydrogenation process previously described to
produce styrene (Figure 11), the process is similar except that the benzene
recycling  (Stream 15 in Figure 11) cannot be reused directly.
                                      85
-------
Process Description:  Styrene from Ethylbenzene Hydroperoxidation

     Presently, only one United States facility uses the hydroperoxidation
process to produce styrene.  Figure 12 gives a process flow diagram.  The
four major steps are described below.

     Ethylbenzene (Stream 1) is oxidized with air to produce ethylene
hydroperoxide (Stream 2) and small amounts of a-methyl-benzyl alcohol and
acetophenone.  The exit gas (principally nitrogen) is cooled and scrubbed to
recover aromatics before venting.  Unreacted ethylbenzene and low-boiling
contaminants are removed in an evaporator.  Ethylbenzene is then sent to the
recovery section to be treated before reuse.

     Ethylbenzene hydroperoxide (Stream 3) is combined with propylene over a
catalyst mixture and high pressures to produce propylene oxide and
acetophenone.  Pressure is then reduced and residual propylene and other
low-boiling compounds (Stream 4) are separated by distillation.  The vent
stream containing propane and some propylene can be used as a fuel.
Propylene is recycled to the epoxidation reactor.  The crude epoxidate
(Stream 5) is treated to remove acidic impurities and residual catalyst
material and the resultant epoxidate stream is distilled to separate the
propylene oxide product for storage.  Residual water and propylene are
recycled to the process train and liquid distillate is recovered as a fuel.
The organic layer is routed (Stream 6) to the ethylbenzene and
a-methyl-benzyl alcohol recovery section.  Distillation removes any
remaining ethylbenzene and then separates organic waste streams from the
a-methyl-benzyl alcohol and acetophenone organic waste liquids are used as
fuel.

     The mixed stream of a-methyl-benzyl alcohol and acetophenone  (Stream 7)
is then dehydrated over a solid catalyst to produce styrene.  Residual
catalyst solids and high-boiling impurities are separated and collected for
disposal.  The crude styrene goes to a series of distillation columns where
the pure styrene monomer product is recovered.  The residual organic stream
                                      86
-------
                                     UZ900/8C
                                        CO
                                        CO
                                        0)
                                        o

                                        2
                                        Q.

                                        co
                                        X

                                       2
                                        CD  .
                                       c. co
                                       ** c

                                       £ 5
                                        f O
                                        o"
                                        n E
                                        CO CD
                                        ^ ID
                                        O c
                                        « i
                                       T3 N
                                       O ^

                                       « S
                                       w CD
                                       ii
                                          ECO
                                          CD
                                       co  £
                                       CD  O.
                                       52
                                       co  „
                                         5
                                                       05

                                                       O>
                                                       CO
                                                       o
CD

 CO
 CO
 
-------
contains crude acetophenone,  catalyst residue,  and various impurities.  This
mixture is treated under pressure with hydrogen gas to convert the
acetophenone to a-methyl-benzyl alcohol.   Catalyst waste is separated from
the a-methyl-benzyl alcohol which is returned to the recovery section for
processing and reuse.  Hydrogen and organic vapors are recovered for use as
fuel.

Process Description:  Styrene Production by an Isothermal Method

     Ethylbenzene may also be converted to styrene by an isothermal process
(Figure 13).  Liquid ethylbenzene is vaporized by condensing steam in a heat
exchanger (Stream 1).  Process steam (Stream 2) is then introduced into the
ethylbenzene stream and the feed mixture is superheated (Stream 3) before it
                                                          Q
enters the molten-salt reactor (Stream 4) (see Figure 13).

     In the reactor, the ethylbenzesne/steara mixture passes through the tubes
where it comes into contact with the catalyst and is dehydrogenated.  Heat
for the dehydrogenation reaction is supplied by molten salt (preferably a
mixture of sodium carbonate,  lithium carbonate, and potassium carbonate)
surrounding the tubes (Stream 5).  The reactor is maintained at a uniform
wall temperature by circulating the molten-salt mixture through the heat
                                       Q
exchanger of a fired heater (Stream 6).

     The reaction products are cooled and condensed in a separator
(Stream 7).  The liquid phase is a mixture of organic products:  styrene,
unreacted ethylbenzene, and small quantities of benzene, toluene, and
high-boiling compounds.  Styrene (Stream 8) is separated from the other
                                                          a
liquid constituents which then are recovered and recycled.

     The gas phase  from the condensation step in the separator consists
mainly of hydrogen, with small quantities of carbon dioxide, carbon
monoxide, and methane.  After these gases are compressed, they are cooled.
Condensable products from this final cooling stage are then recovered and
                                      88
-------
ueoxuar
                    CO
                             co
                             CO
                             03
                             o

                             2
                             Q.

                             co
                             X
                             CD
                             c
                             o

                             CO
                             co

                             Q

                            .2
                            •o
                             CD
                             O
                             Q.
                             CO

                             2

                             o
                             o
                             co


                            I
CO

2
**
co
• •


o
             0)
             c
             
             o
             o
             (0

             E
                                         o
                                         CO
            CO


             2
             5
             g>

            LL
89
-------
recycled to the separator.  When hydrogen-rich offgas is used as fuel for
the heater of the molten-salt reactor,  the fuel requirement for this stage
                       Q
of the process is zero.
Benzene Emissions from Ethvlbenzene and Styrene Production

Emission Estimates from Ethylbenzene Production and Dehydrogenation to
Styrene--

     Estimated emission factors have been developed based on an uncontrolled
300 million kg/year capacity integrated ethylbenzene/styrene production
plant.  Major process emission sources are:  the alkylation reactor area
         •
vents (A in Figure 11), atmospheric and pressure column vents (B-l), vacuum
column vents (B-2), and the hydrogen separation vent (Stream 12).  Emission
rates from these sources are given in Table 14.  The first four process vent
streams in Table 14 are low-flow, high-concentration streams.  The hydrogen
separation stream (Stream 12 in Figure 11) is high-flow, low-concentration.
Other emission sources listed in Table 14 include storage losses (D) ,  (E) ,
(F),  and shipment losses (G).   Fugitive emissions from valves and other
equipment leaks are not indicated in Figure 11.

     Reactor area vents remove various inerts plus entrained aromatics
(benzene).  Inerts include nitrogen or methane used in pressure control,
unreacted ethylene, reaction by-products, and ethylene feed impurities.  In
typical plants using liquid phase Aid, catalyst with high-purity ethylene,
vent streams are usually cooled and scrubbed to recover aromatics.  In
plants using the newer solid support catalysts of the UOP or Mobil/Badger
process, reactor vent flows are very large due to the low-purity ethylene
feed.  Process economics require that these vent gases be burned as fuel.

     Atmospheric and column vents remove non-combustibles in the column
feeds, light aliphatic hydrocarbons, and any entrained aromatics.  The
benzene drying column also removes impurities  in the benzene feed.  Most
emissions occur in the first column of-the distillation train (benzene
recovery column in Figure 11).
                                      90
-------














<



z
o

H


Q
O
(6


Cd
Z
Cd

F

00

Cd
Z _
u at
N Z


3 5

i 1
w o
Cu
cd a
2 £
Cd
CO O
B!
o a
H Z
O ^J

fa Z
o
Z M
0 H
lo £
M ^

w <;



,j.'
i-i

Cd

(O


H


























































































TJ
U

-i Ml
O X
U -~
U M
B
O
O




X
u
B
1)
~4
O

IH
IH
U

x











41
U

^
a

o
M
4J
B
O
o





•o
V
2 jf
4J ~—
B 00
0
1

B
V O
O -4
M U
3 It
O 0
Wl O









U
u
g
VI

g
•1







o
o
o

o









*


a>
















u

u
It
Q)

in
n

u
0
a.





u
o






x










it
S
14
O
U
a
it
1 at ion Re
s
X U
Jt S
^ l>

00

O r-l

1 1

CM en
.H O
0
0
o




•O TJ

0. 0.
| |

U O
0 O






















V V
« «
^-i ^-1





o u
CM 0
IH en






iH CS
1 1







i s

1— ( >
0
1
41 3
3 it
n >


04 V
O -1
•a &
£ ;
a B
•) W V

S S u
^ J> rt

o
o
o

1


o
0
o

o




at
o»
i

u
0






en -d r*
o r* r-i vo
o m IH o
0 O 00






0 ?
Q








4J
B
^
0 -1

it B
M It
1 "
w § x
S , ? S
00 M -4 TJ
0 it U B
U U -1 O
•do oo o
X 4J 3 II
93 W b, u]


•D T3
B II
el u
u

3 ~B
0. II
J3
00 II
3 B
0 II
M N
A B
vi «
.B
> «
it

H T)
It V
V)
X II
>&
n V

O ll
«J LI
U el
it
IH n
B 0
0 u
•H O
 V
B
U H
g g^
— t « V
a. 4-1 o
« 3
•S13"?
•O 41 U
O J3 O.
a
T3 II
« -1 C
3 II
U 0 U
O £ X
IH m 4-1
41 -O — .
U B II
el 3 E
II
HI « N
W I) B
OUD
*J 3 J>
U •> — 1
II el X
B8H
1 0 I
« U Jl
•1 4-1
-4 C U
B 0 41
w o a
it


























































rH

II

3
00
te.
B

p


X
HI
n
it

«l
B
0
v4
rce locat
erence 7.
erence 1.

0 II II
J3 O *O
91
-------
     Vacuum column vents remove air that leaks into the column,  light
hydrocarbons and hydrogen formed in dehydrogenation,  non-combustibles in the
column feed, and entrained aromatics.   Most emissions occur on the
benzene/toluene (B/T) column (B-2 in Figure 11).   Uncontrolled distillation
                   .3
vents emit 4.2 x 10   kg HC per kg styrene in one plant where HC is benzene
and toluene.  Another condenser controlled vent emits 0.4 x 10"   kg
                   2
benzene/kg styrene.

     Following dehydrogenation, a hydrogen-rich gas (Stream 9 in Figure 11)
containing methane, ethane, ethylene,  carbon dioxide, carbon monoxide, and
aromatics is normally cooled and compressed to recover aromatics.   The
stream should be vented to the atmosphere (Vent C) only during startup,
shutdown, and during recovery section compressor outages.  Some  plants may
also vent this stream to a flare.  Flares are an efficient (99 percent)
emission control only when flair diameter and gas flow are closely matched
for optimum turbulence and mixing.  They can be better controlled when the
stream is routed to a manifold and burned with other fuels.

     Stripper vents have been reported to emit 32 g ethylbenzene per kg
        2                              -3
styrene.   This corresponds to 9.6 x 10   g benzene per kg styrene.  Benzene
in shipping and storage (D-l in Figure 11) must also be considered as a
source if benzene is not produced on-site (in which case these emissions
would be considered part of the benzene production process).   Test data
indicate an uncontrolled tank car/truck loading emission factor of
0.11 g/liter benzene, and an uncontrolled marine loading factor of
                     2
0.18 g/liter benzene.   Storage tank losses for floating-roof tanks can be
                     -5                              -4
estimated at 1.3 x 10   g/liter standing and 8.9 x 10   g/liter during
           2
withdrawal.   Other storage tanks  (D-2, E, and F in Figure 11) are typically
fixed-roof and have higher total emission rates but lesser benzene
concentrations.

Emissions from Styrene Production Using Ethylbenzene Hydroperoxidation--

     Only one United States facility currently reports using this method.
Therefore, emission estimates  are based on its capacity of 1200 million
Ibs/year styrene.
                                       92
-------
     There are three main process emission sources:   the ethylbenzene
oxidation reactor vent (A in Figure 12),  the propylene recycle purge vent
(B),  and the vacuum column vents (C) and (D).   Propane vapor (B) is
considered a fuel if it is not vented to the atmosphere.  Of these sources,
only the vacuum vents are large benzene emitters.  This benzene results from
benzene impurities in the ethylbenzene feed and minor side reactions in the
process train.

     The ethylbenzene oxidation reactor vent (A) releases carbon monoxide,
light organics, entrained aromatics with nitrogen, oxygen, and carbon
dioxide.  The vent gas is scrubbed with oil and water for a 99 percent
removal efficiency for organics.  The resulting vent stream contains
approximately 35 ppm benzene or 7.2 kg/hour.

     The propylene recycle vent (B) releases propane, propylene, ethane, and
other impurities.  No flow volume data are available but, based on a similar
procedure in high-grade propylene production,  this stream is a high-Btu gas
and would be used as a fuel.  No significant benzene emission is expected.

     The ethylbenzene hydroperoxidation process contains numerous vacuum
columns and evaporators.  Vents on these operations (C-l - C-3) release
inerts and light organics dissolved in the column feeds, nitrogen used for
process pressure control, and entrained aromatics.  A combined vent flow is
reported to be 1.0 x 10  liters/hour containing about 27 kg/hour benzene.

Other Sources of Emissions--

     No specific information is available on storage, transport, or fugitive
emissions for this process.  The dehydrogenation vent (D) may be an
emergency pressure vent similar to the separation vent  (C in Figure 11).

     Control methods for the two ethylbenzene/styrene processes in use in
the United States include condensation, adsorption,  flaring, and combustion
in boilers or other process heaters.  Controls for fugitive emissions from
                                      93
-------
storage tanks, equipment leaks, and others include the use of floating-roof
tanks and leak detection/correction programs.   No information is available
on control methods specific to the two processes mentioned in this report
but not in use in the United States.

     Condensers may be used to control benzene emissions associated with
ethylbenzene/styrene production.   The control  efficiency of a condenser is
determined by the temperature and pressure at  which the condenser operates
and by the concentration and vapor pressure of the organics in the vent
stream.  At typical pressures of 1 to 3 atmospheres and coil temperatures of
2 to 5 C, condensers can achieve 80 to 90 percent benzene reduction when
used on vent streams at 70 to 100 percent saturation in benzene at 40 to
50 C.   Higher efficiencies become prohibitively expensive.

     Condensers have limited use in handling high volume streams, short
duration emergency releases, or cyclic releases such as the hydrogen
separation vent.  Furthermore, condensers are  inefficient at low saturations
such as with the alkylation reactor vents and  the column vents of Figure 11.

     In an ethylbenzene/styrene plant, a packed tower can be used to remove
benzene.  Polyethylbenzene (PEB) and various ethylbenzene produced during
benzene alkylation are good absorbers of benzene and are normally recycled.
This system is unsuitable, however, for handling high volume or intermittent
releases of gases beyond the tower design capabilities.  Absorption systems
can maintain 80-99 percent benzene removal efficiencies for both saturated
and unsaturated benzene streams depending on the tower design and operating
variables.

     Flare systems can control some streams for which condensation or
absorption are not suitable.  Flares can efficiently handle high saturated
streams such as the alkylation vents.  They can also control upset releases
and other irregular releases, although the efficiency can be variable.  The
major difficulty here occurs in manifolding.  High nitrogen or other low- or
non-combustible gases may also be problematic.  Consequently, there are no
                                      94
-------
conclusive data on flare efficiency.  Limited data show benzene destruction
efficiencies ranging from 60 to 99 percent.  A properly designed flare
system must account for a range of flow and gas composition as well as the
potential for explosion.

     Use of vent gases as a fuel combined with regular process fuel is
advantageous because vent flow variations can be better accounted for.
Also, better gas/air mixing occurs along the entire flare front.  As with
flares, however, manifolding to ensure optimal combustion characteristics is
the major technical problem.  Process pressure variations and the
possibility of emergency releases are complicating factors.

CYCLOHEXANE PRODUCTION

     About 15 percent of the United States supply of benzene is used to
                    9
produce cyclohexane. '  Table 15 lists the location and current capacity for
the United States cyclohexane producers.  Of the nine plants, seven are
located in Texas.  One cyclohexane plant is located in Oklahoma and one in
Puerto Rico.  Two basic methods are used to produce cyclohexane:
hydrogenation of benzene and petroleum liquid separation.  About 85 percent
of the cyclohexane produced domestically is produced through hydrogenation
of benzene and the remainder is produced through separation of petroleum
liquids.  The following discussions of these two processes are taken from
Reference 10.

Process Description:  Benzene Hvdrogenation

     Figure 14 shows a model flow diagram for the manufacture of cyclohexane
by benzene hydrogenation.  High-purity benzene (Stream 1) is fed to the
catalytic reactors in parallel and hydrogen (Stream 5) is fed into the
reactors in series.  Part of the cyclohexane separated in the flash
separator is recycled (Stream 3) and fed to the reactors in series.
Recycling helps to control the reactor temperature, since the reaction is
                                      95
-------
           TABLE 15.   UNITED STATES PRODUCERS OF CYCLOHEXANE
                                                            11,12
             Company
    Location
Company,  Inc.

Phillips Petroleum Company

  Phillips 66 Company


  Phillips Puerto Rico Core, Inc.

Sun Company, Inc.

  Sun Refining and Marketing
  Company

Texaco, Inc.

  Texaco Chemical Company

Union Pacific Corporation

  Champlin Petroleum Company

Unocal Corporation

  Union Oil Company

TOTAL
Borger,
Sweeny, TX'

Guayama,  PR
Tulsa, OK
Port Arthur, TX
Corpus Christi, TX
Beaumont,  TX
                       Annual Capacity
                        (106 gallons)
Chevron
E.I. duPont de Nemours and
Port Arthur, TX
Corpus Christi, TXa
38
50
                              45
                             111

                              89
                             >30
                              65



                              22



                              30

                            >480
 Plant closed July 1986.

bPlant closed mid-1986.
Note:  This listing is subject to change as market conditions change,
       facility ownership changes, plants are closed, etc.  The reader
       should verify the existence of particular facilities by consulting
       current listings and/or the plants themselves.  The level of benzene
       emissions from any given facility is a function of variables such as
       capacity, throughput and control measures, and should be determined
       through direct contacts with plant personnel.  These plant locations
       and capacities are current as of January 1, 1987.
                                      96
-------
                                   CD
                                   O>
                                   C

                                  "(0
                                   c
                                   o
                                  *->
                                   o
                                   3
                                   (D CO
                                   C 0)
                                   CO O
                                   x
                                   O C
                                   0.2
                                   >*-s
                                  O  ®
                                  > N
                                  O C

                                  t&
                                  CO
                                  CO
                                  CD
                                  o
                                  o
                                  O)
97
-------
highly exothermic.  The temperature: is also controlled by generating steam,
which is used elsewhere in the petrochemical complex.  Both platinum and
nickel catalysts are presently used to produce cyclohexane.

     After leaving the flash separator, the cyclohexane (Stream 9) is sent
to a distillation column (stabilizes) for removal of methane, ethane, other
light hydrocarbons, and soluble hydrogen gas from the cyclohexane product.
These impurities (Stream 10) are routed to the fuel-gas storage system for
the facility and used as fuel in process heaters.  Cyclohexane (Stream 11)
purified in the stabilizer may be greater than 99.9 percent pure.  The
residual benzene content is typically less than 500 mg/liter.  This pure
product is stored in large tanks prior to shipment.

     Gas from the flash separator, largely hydrogen, is not pure enough for
direct reuse.  Therefore, the stream (6) is purified before being recycled
to (Stream 5) the reactor.  Typical processes used for hydrogen purification
are absorption and stripping of the; hydrogen gas and cryogenic separation.
Some plants use a combination of the two processes.  Organic liquids
(Stream 12) that are separated from the hydrogen in the hydrogen
purification unit are sent to other petroleum processing units in the
petrochemical complex.  The separated gases (Stream 13) are used as fuel
gas.

     Depending on the type of hydrogen purification used,  inert impurities
present in the gas from the flash separator can-be purged  from the system
before the gas enters the hydrogen purification equipment.  This stream (8)
is sent to the fuel gas system.

Benzene Emissions from Cvclohexane Production via Benzene  Hydrogenation

     There are no process emissions during normal operation.    During
shutdowns, individual equipment vents are opened as required during final
depressurization of equipment.  Except for the  feed streams, the
concentration of benzene  in the process, equipment is low;  therefore, few or
no benzene emissions would be expected during a  shutdown.
                                       98
-------
     Fugitive leaks can emit benzene or other hydrocarbons.  Fugitive
emissions from process pumps, valves, and compressors may also contain
benzene.  Storage of benzene (A in Figure 14) may also contribute to benzene
emissions.

     Other potential sources of emissions are catalyst handling (B) and
absorber wastewater (C) (when an aqueous solution is used to purify the
recycled hydrogen).  Plants comprising at least 16 percent of the total
cyclohexane capacity use an aqueous solution to purify hydrogen.  Caution is
taken to remove the organic from the spent catalyst before it is replaced.
The spent catalyst is sold for metal recovery.

                                                              v
     Table 16 presents estimated emission factors for benzene emissions due
to storage and fugitive emissions.  The control technique applicable to
storage facilities is the use of internal floating roof tanks.    The
estimated efficiency of emission reduction is about 85 percent.  For
fugitive emissions, the control technique is an inspection/maintenance
program, which can reduce emissions about 80 percent.

Process Description:  Separation of Petroleum Fractions

     Cyclohexane may also be produced by separation of select petroleum
fractions.  The process used to recovery cyclohexane in this manner is shown
in Figure 15.  A petroleum fraction rich in cyclohexane (Stream 1) is fed to
a distillation column, in which benzene and methylcyclopentane are removed
(Stream 2) and routed to a hydrogenation unit.  The bottoms (Stream 3) from
the column containing cyclohexane and other hydrocarbons are combined with
another petroleum stream (4) and sent to a catalytic reformer, where the
cyclohexane is converted to benzene.  The hydrogen generated in this step
may be used in the hydrogenation step or used elsewhere in the petrochemical
complex.

     The benzene-rich stream (5) leaving the catalytic reformer is sent to a
distillation column, where compounds that have vapor pressure higher than
benzene (pentanes, etc.) are moved (Stream 6) and used as by-products.  The
                                      99
-------





o
1— 1
S!
O
^^
H


2
O
M
EH
U
§
O
OS
PH

W
z
•<
X
u
cs
0
f]
CJ
>H
CJ

g
OS
fa

Cd
55
PL)
N


«

M
0
fa

CO
OS
o
B
-,
u
4J B
B 0)
4) -H
U U
l-l •*
01 >4H
CM 14-1
U




01
U"
•H
B
43
u
01
H
r-l
O
M
4-1
B
O
CJ






13
0) CO
r-t B
i-i oa
O -H 60
i- co s
4-1 CO •*•-.
B 'i-l 60
8%








13
0>
i-H CO
rH B
O 00
l-i -i-l 60
4J CO S3
B a ~-
O -H 60
0 s
B W
O





CO
B
O
ca -i-i
CO 4-1
Ol CO
U B
O 00
M •<-!
PM co
0)
O













01
u
l-i
O
CO







^-e 6^e
m o
oo ao







60
B
•H
4-1
cd
O
•-! -^. tt>
MH co B U
^ OB
ft B -H eg
CO CO 4J B
B 4J U 0)
l-l 01 4J
oi u-i a. B
4-1 O CO -H
BO B CO
l-l M M @










CN r-«
• •
m — <
VO -H












•*
•
•-H f(
es m
«*















<5 «










CO
01
0) >
60 -H
CO 4J
W4 -i-l
O 60
4-1 3
co fa

41 co
B ca
0) 01
N U
B 0
01 l-i
pq Pn




O B
u O
• H
"O CO
0) ca
60 •!-!
I-i B
S S
ca 0)
•H CO
01
M A
01 4J
•a
cd 60
01 B
l-l -i-l
>,
01 ft
fi a,
H a.
cd
. a>
i-i i-t
>> o
~^~ 4-1
60 01
U 43

O CO
m 01
i-H 3
-. U
4-1 O)
•i-l 4J
y
CO i— *
a. o
cd M
O 4J
B
i— I O
g »
B T3
C B
CO cd
B > M
3 43 0
60 U-I
•iH M
fa O CO
U-I 01
O -i-l
4-1 0)4-1
U -H
M CO i-i
0) -.-1
14-1 ca o
01 M cd
M O <41
44
co y y
B co -H
O M-l 14-1
•H -i-l
4-1 B U
CO O 01
o .H a.
O ca ca
i— i ca •
'•H 44 CO
ai S o i-i
y 01 eg O
l-l 4-14-1
3 0> B U
O 43 O cd
co H cj VM
CO 43
100
-------
                                                       didl90CU8e
   > c
                                   £«
                                   co
   CD
   <*^


   O)
                     CO

                     CO
                     CD


                    2



                    I
                  CO
                  CD

                  CO
                  *-



                  I
                  >s
                                                                           O
                    ©        t
CD

CO
X
CD
                                           O

                                           (0
2 o
CO CD
otr
                    0
                                                             ©
                                          CD
                                          _c

                                          "o
                                          co
                                          CO
                                          o
  CO CO
                                                                                 co
                                                                                 co
                                                                                 CD
                                                                                 O

                                                                                 2
                                                                                 Q.
                                                                                  O
                                                                                 X
                                                                                 CD
                                                                                 _c

                                                                                 c
                                                                                 CO
                                                                                 co
                                                       TD

                                                       CD
                                      •o
                                      C
                                      O
                                      a
                                      CO

                                      S
                                      o
                                      o
                                      co

                                      CD
                                      £)

                                      E
                                      3
                                      C


                                      CO

                                      S

                                      53

                                      (B
                                      «rf
                                      o
                                      z
                                                                                           CO
                                                                                           c
 o
 cd
 u.
LL

 E
                                                a?
                                               a.



                                                2
                                               LL
                                                O
                                                C
                                                08
                                                X
                                                a>
                                               £1
                                                O
                                                o
                                                >s
                                               O
 to

 O)
 CO

b


_0
LL

 CO
 CO
 CD
 O
 O
                                 ID


                                  CD

                                  3
                                  g>

                                 LL
                                     101
-------
benzene-rich stream (7) that is left is sent to another distillation column,
where the benzene and methylcyclopentane (Stream 8) are removed.  The
remaining hydrocarbons (largely dimethylpentanes) are used elsewhere in the
petrochemical complex as by-products (Stream 9).

     Stream 8 (benzene and methylcyclopentane) is combined with stream 2 and
sent to a hydrogenation unit (Stream 10).  Hydrogen is fed to this unit and
the benzene is converted to cyclohexane.  Isomers of cyclohexane such as
methylcyclopentane are converted to cyclohexane in an isomerization unit
(Stream 11) and the effluent from this equipment (Stream 12) is separated in
a final distillation step.  Pure cyclohexane (Stream 14) is separated from
isomers of cyclohexane and compounds with lower vapor pressures.

Benzene Emissions from Cyclohexane Production via Separation of Petroleum
Fractions

     There are no process emissions during normal operation.    During
emergency shutdowns, individual equipment vents are opened as required.

     Equipment leaks can be sources of benzene, cyclohexane, methane, or
other petroleum compounds emissions.  Leaks from heat exchangers into
cooling water or steam production can be a. potential fugitive loss.
Fugitive losses have special significance because of the high diffusivity of
hydrogen at elevated temperatures s.nd pressures and the extremely flammable
nature of the liquid and gas processing streams.    No specific emission
factors were found for benzene associated with fugitive emissions at these
plants.

     A potential source of benzene emissions is catalyst handling.  Special
efforts are made to remove the organics from the spent catalyst before it is
replaced.  The spent catalyst is sold for metal recovery.    No emission
factors were found for benzene as related to catalyst handling.
                                      102
-------
CUMENE PRODUCTION

     In the Unites States, cumene is produced by alkylating benzene with
propylene.  The location and capacities of United States producers of cumene
are shown in Table 17.  Benzene and propylene react at high temperatures and
                                                13
pressures in the presence of an acidic catalyst.    By far, the most common
catalyst used is solid phosphoric acid.  Figure 16 shows a process flow
diagram for production of cumene using solid phosphoric acid as the
catalyst.  As shown in the figure, propylene and benzene are introduced into
a pressurized combined feed drum (Streams 1 and 2).   The feed ratio is
usually 4 moles of benzene per mole of propylene.    From the feed drum, the
benzene and propylene are sent to the reactor (Stream 3).  From the reactor,
the by-products, unreacted material, and product are separated by
distillation.  The depropanizer (Stream 4) removes propane, which is then
sent through a condenser (Stream 5).  Unpurified product is sent from the
depropanizer to the benzene distillation column (Stream 6), where unreacted
benzene is recovered and sent back to the combined feed drum (Stream 7).
From the benzene column, the cumene product is sent to a cumene column
(Stream 8), where the final product is separated (Stream 9) from residual
bottoms (Stream 10).   Residual bottoms may be burned as fuel or returned to
a refinery for reforming.

     The production of cumene using an aluminum chloride process is similar
to that using a solid phosphoric acid, catalyst.  The aluminum chloride
method requires additional equipment to dry recycled streams and to
neutralize reaction products.  A simplified process flow diagram is shown in
Figure 17.  Aluminum chloride is a more reactive and less selective catalyst
than solid phosphoric acid.    As shown in Figure 17, the propylene feed
stock used in this process must be dried and treated to remove organic
sulfur compounds.  The benzene must also be dried.  Benzene and propylene
are sent to a catalyst mix tank (Streams 1 and 2), where aluminum chloride
powder is added to form the catalyst complex (Stream 4).  Hydrogen chloride
gas is added to activate the catalyst (Stream 5).  The resulting catalyst
suspension (Stream 6) and fresh benzene (Stream 7) are fed into the
                                      103
-------
       TABLE 17.  UNITED STATES PRODUCERS OF CUMENE AND ANNUAL CAPACITY
                                                                       11
         Plant
    Location
 Annual
Capacity

(106 Ib)
    Notes
Amoco Chemicals Company
Ashland Chemical Company

BTL Specialty Resins
Corporation

Champlin Petroleum
Company

Chevron Chemical Company
Coastal States Marketing

Georgia Gulf Corporation




Koch Refining Company

Shell Chemical Company


Texaco Chemical Company
Texas City, TX


Catlettsburg, KY

Blue Island, IL


Corpus Christi, TX


Philadelphia, PA

Port Arthur, TX

Westville, NJ

Pasadena,  TX
Corpus Christi, TX

Deer Park, TX


El Dorado, KS
    30


   400

   110


   450


   450

   450

   140

   750
   550

   900


   135
Captive for
methylstyrene

Cumene is sold

Captive for
phenol
Cumene is sold

Cumene is sold

Cumene is sold

Some cumene
transferred to
company's phenol/
acetone plant

Cumene is sold

Captive for
phenol/acetone

Captive for
phenol/acetone
Note:  This  listing  is subject to change as market conditions change, facility
       ownership changes, plants are closed, etc.  The reader should verify the
       existence of  particular facilities by consulting current listings and/or
       the plants  themselves.  The level of benzene emissions from any given
       facility is a function of variables such as capacity, throughput and
       control measures, and should be determined through direct contacts with
       plant personnel.  These locations, producers, and capacities are as of
       January 1987.
                                         104
-------
              USZOOZ8C
                                     CO
                                     co
                                     CD
                                     0.
                                     co
                                     X
                                     CD
                                     co
                                     co

                                     O
                                     n
                                     O
                                     ^  "

                                     Is
                                     o
                                     o o
                                     CD CO
                                     CD °
                                     ® CO
                                     CO o

                                     CD o

                                     C °-
                                     §2
                                     c -o

                                     E§
                                     2 o-
                                     CD 09
                                     ib »
                                     13
£>
 k_
 o
.c
 Q.
 CO
 O
CO
 O)
^
 (0

 (D
 c
 (D
 E
H_  (0

 (D "cO



,2J2

 c<
 CO
s
 (D
 (0
 (0
 O
 O
 O
CO


2
3
O)
105
-------
                                                            US90CU9C
    CO
    CO
    CD  •
    O 2

    u
    «s
    SI
    o £
    c c
    O CD
y

CD
|



^i^
•o
5






0
e
CD
>.
O.
O
ct
(E

—
(D
5

e
                             «a>
    2§
||

ii
82
£ (D
                                            § 2
                                            i®
                                            E «
                                            a -J
                                            £  .
                                                              l
uujnioo eueuuno
   J
                                        uuun|OQ euezueg
                                                                          a>
                                                                          c
                                                                          CO
                                                                         •o
                                                                          2
                                                                         a.
                                                                          a>
                                                                          c
                                                                          CD
                                     ECO
                                  *°
                                  2  CD

                                                                          CO .=
                                                                          CD C
                                                                          o E
                                                                          0.2
                                                                         •o
                                                                          CD
                                                                          Q.
                                                                          E
                                                                         co
                                                                          CD
                                                                          ^
                                                                          •3
                                                                          a>
                                 106
-------
alkylation reactor and propylene is introduced into the bottom of the
reactor (Stream 8).  The crude reaction mixture is then fed to a degassing
vessel (Stream 9),  where hydrocarbons such as propane are released from
solution.  This vapor stream is scrubbed with a weak caustic solution.  The
washed hydrocarbon vapor is recontacted with diisopropylbenzene (DIPB) in
the DIPB scrubber (Stream 10) to extract residual unreacted propylene.
Then, the stream containing the propylene is sent back to the catalyst mix
tank (Stream 11).

     The product from the alkylation reactor is forwarded to an acid wash
tank where the catalyst complex is broken down.  A resulting
hydrocarbon-water mixture is sent through a series of decanters (Stream 12)
and the decanted hydrocarbon layer is stored in a washed product receiver
tank (Stream 13).  The wash/decanter system is blanketed with nitrogen.

     The crude product from the washed product receiver (Stream 14) tank is
sent to the benzene recovery column to strip excess benzene.  Any benzene
recovered is returned to the benzene feed tank (Stream 15).   Crude cumene
product is then distilled (Stream 16).

     The cumene distillation column and receiver tank are blanketed with
nitrogen to protect the cumene from air.  Cumene product is then sent to
storage (Stream 17).  The bottoms from the cumene distillation column
(Stream 18) contain crude DIPB, some cumene, alkylbenzenes,  and
                                                      i Dl
                                                      13
miscellaneous tars.    The bottoms stream is sent to a DIPB stripper and
DIPB is eventually returned to the alkylation reactor.

Benzene Emissions from Cumene Production

     In the solid phosphoric acid process, the cumene column vent  (A in
Figure 16) is a potential source of benzene emissions.  The system operates
at a pressure slightly higher than atmospheric pressure to make sure than no
air contacts the cumene product.    Methane is used to pressurize the
system.  This methane is eventually vented carrying with it other
                                      107
-------
hydrocarbon vapors.    No specific emission factors were found for benzene
emissions from the cumene column.  One factor for total VOC emissions
indicated that 0.03 g VOC were emitted per kg cumene produced and that
benzene constituted a "trace amount" of the hydrocarbons in the stream.

     Emissions of benzene from the production of cumene using an aluminum
chloride catalyst are associated with the benzene azeotropic drying column
(A in Figure 17),  the scrubber or the catalyst mix tank (B),  the wash
                                                         13
decanter system (C),  and the benzene recovery column (D).     No specific
emission factors were found for benzene emissions from these sources.
However, one reference provided VOC emission factoring and estimated percent
                             13
composition of the emissions.    Table 18 presents these data.  The percent
(weight) of benzene may be used along with a cumene production volume to
estimate benzene emissions from these sources.  The control technique most
applicable to these sources is flaring, with an estimated efficiency of
           13
95 percent.
PHENOL PRODUCTION

     Over 90 percent of the phenol produced in the United States is based on
                       14
peroxidation of cumene.    Table 19 shows the locations, capabilities, and
production methods of the phenol producers in the United States.  Because
benzene may be present in the cumene feedstock, benzene may be emitted
during production of phenol.

     A flow diagram for the production of phenol via cumene peroxidation is
shown in Figure 18.  The process essentially involves two steps.  First, air
is introduced into an emulsion of purified cumene to produce cumene
hydroperoxide (Stream 1).  Then, dilute sulfuric acid is added to cleave the
compound directly into phenol and acetone.  The acetone are separated
(Stream 3) by distillation and phenol is recovered from the finishing column
           14
(Stream 4).
                                      108
-------



















H
S5 m
 OH


•
OO
r-l

Cd
i-J
g
H



















C dp
O ^— '"
•i-l
co C
CO O
1-1 1-1
3 4J
Cd O
3
cj -0
O cu
> tti



r-l
O CU
u u
4J i-l
C >
o cu
o o






,0

o o
> 4J
u
•O cfl
cu b oo
rH S
0 r^

O O










< CQ





.





cu
Cu
o
i-i
4-> C l-l
§iX 4)
*^4 jQ
N ^ 32 ^3
< 0 3
CJ 4J U
cu ca cj
C oo >-, co
CU C i-H
N i-l Cfl Jrf
C ^ we
0) l-l CQ CO
CQ a cj H






in in
ON ON









CU CU

cfl cd
r-l r-l











m
in oo
O O
o o
o o
o o











*^
•
00 O
r^ o
i-H









^
r-l r-l
O O

CD 0










CJ Q












^^
^
cu
^ ^
CU O
U 0
C cu
CO Crf
CJ
cu cu
as c c
i cu cu E

ca ca C i— i
cfl >i CU O
3 co ca cj



































































oo
i-H

CU

•S
oo
•H
b

O
4-1

^1
cu
<4H
CU
iH

CO
C
O
iH
4J
Cfl
C
00
•H
CO
cu
73
E
cfl
cu
l-l
4J
CO
cfl .

*-M
0) O

EH N «
l-l CO
cu C
Cu 0
0) 4J
C cfl
(U l-i
§0)
a
u o

00 00 •
U C co
iH !-l
r- 4J o
CM 4J 4J
CM >r4 O
S cd
OO CU <4-4
c
i-4 («H C
cj o o
•O cu co
o cj ca
U G -n
Cu 

CO 4J OO
cj C
1-1 E "-i
CU iH r-l
• -C <4H CU
T3 4J C Cu
cu o o co
U Cu U
3 ^ CU
•a js o IH
O 4-1 O
IH cd <4-l
CU i-H CU
C CU ft
CU O C
C C >>
cu "d o 4J
§CU CO iH
ca )H I-H
CJ Cfl CU -r4
J3 Cu CJ
oo cd
ii - 4J <4H
ca C
IH cu co i-i
CU 4J i— 1 Cfl
CU Cfl CurH
CJ iH 4J O
O 4J CJ i-l
> Cfl Cd 4J
CU 4J l-l
<4H C CO
O CU O CU
IH U
ca to CQ
B O
Cfl CO 4J 4J
)-i l-i eg
00 O T3
4J 0) T)
ca o oo cu
cfl cfl co ca
« "S3
cu c o cu
ca o u u
CO -rl C fl
cu ca cu >
u ca cu
CU iH CO T3
X S -^
CU CU rH
b 0
cu cu cu in
IH w -a 4J
cfl cu cfl C
X CU 0
ca H IH U
j-i
o '
4J
u cu
Cfl 4J
b O
Q 2
109
-------






^
H
to
M
H
tH
O

(X,
^J

a


CO
g
H

q


a;
M
w
H
•k
1
S
PL,
O
C/3
CEi
U
3
a
o
£
CO
03


g^
CO

a
w
H
M




i-(
**3
H

























































l|
n w
0 »
U 4
ft, os



_
n

w 3
-4 O
o 04
4
O*

•o
^
B
O
-4
4
O
fr
O
u

p_l

li

Stimson Lumber Co
Northwest Petro



B
0
-4
U
•3
I
a
g
u








SI S!
1 tn
1 V









a

o
u
a
H











h^
4
•o
•H
«
jQ
3
•






Texaco, Inc.
Texaco Chemical
TOTAL
r*.
<*4 CO
MO O»
• ,H
•fl -4 "fl
« e C >•
S > 3 u
he 4
—I • 3
*J I "
« 44
4 0
O • V
U II g II
• -1 "O «
•O « U U
« « W 4
UBS
0 « 0 «
-4 J! 0 B
O 4J O
« « C 4J
U u 3 4
4 B 0
4 - O
8*1^


°" w 5 «
« t* M M
ffxl
£ 44 e
jj o a a
o
-4 « o a
52 sf
B -H -4
0 -4 4 »
00
4 « «
*J *t 1 —4 4J
i U »S
i-SS g
a -^ o
•1 *4 (I
« w <* u

««'**«
f0* § i
*M ^ "9
O «• J3
S S &f
e 4 u
4J 4J X
O • 8 W
• -4 5
-1 M U •»
ji e <« e


• 4J O H
K 11 U
?<*4 « •
^ .4 -e
Note: This listl
should ver
benzene em
should be
110
-------
o

CD
)L
                               CD


                               3
                                                                  C
                                                                  o


                                                                  3
                                                                  O





o
8
y eueoinj
P
Q.
E















k.
2
CO
CD
O)
8
>,
X
.'T
II
CD C
C CD
CD O>
E 2
°£



















CD
O
«
O








i
•1
J
u
1




-*-







©
0
3
5
j
j
i




^•^i^
.-.

O
«
CO
Q.
CD
CO
1 ,


S
•o
6
"ts
uiunioQ
tunnoBA
t

uiunioo
U0|l*|||18|a "
1 1





o
CD
- &
5 8
^\ w
® 0
<
) i -"
                              §

                              oo
               o
                                                    X
                                                    CD
                                                   eo
                                                   co
                                                   CD
                                                   u
                                                   O
                                                    «- CO
                                                    o c
                                                    c5
                                                    O CO
                                                    •m*

                                                    S|
8                                                      CD
                                                      CD

                                                    ^i
                                                    O N
                                                    JZ C

                                                    oS
                                                     <
                                                   O CO
                                                   Q. O
                                                   CO g
                                                   CD 3
                                                   S= O
                                                   O «°
                                                  c o
                                                  A  CD
                                                  <-• *^
                                                                   o
                                                                   c
                                                                   (D
                                                                      CO
                                                                      CO
                                                                  c  c
                                                                  (Q  O



                                                                 H
                                                                 5  X
                                                                  -  O
                                                                  CO

                                                                  O)
                                                                  CO

                                                                 b


                                                                  o
                                                                 LL


                                                                 06

                                                                  a

                                                                  3
                                                                  O)
               CO
                   111
-------
Benzene Emissions from Phenol Production
     "Spent air" from the oxidizer (A in Figure 18) is the largest source of
                                                  14
benzene emissions at cumene-process phenol plants.     An emission factor of
2 x 10   kg benzene/kg phenol was reported for one source from a post
                                      14
oxidizer concentration condenser vent.    An order of magnitude estimate of
1.0 x 10   kg benzene/kg phenol has been reported in the literature for
plants using the cumene peroxidation method of phenol production.    No
details were given in the literature concerning the derivation of any of
these emission factors.  The reader is urged to contact specific plants to
obtain information on emissions, control techniques, and processes used
before applying these emission factors.
NITROBENZENE PRODUCTION

     Benzene is a major feedstock in commercial processes used to produce
nitrobenzene.  In these processes, benzene is directly nitrated with a
mixture of nitric acid, sulfuric acid, and water.  As of January 1986, four
companies were producing nitrobenzene in the United States.   Their names and
plant locations are shown in Table 20.

     A discussion of the nitrobenzene production process, potential sources
of benzene emissions, and control techniques is presented in this section.
Unless otherwise referenced, the information that follows has been taken
directly from Reference 13.

Process Descriptions

     Nitrobenzene is produced by a highly exothermic reaction in which
benzene is reacted with nitric acid in the presence of sulfuric acid.  Most
commercial plants use a continuous nitration process, where benzene and the
acids are mixed in a series of continuous stirred-tank reactors.    A flow
diagram of the basic continuous process is shown in Figure 19.
                                       112
-------
              TABLE 20.  UNITED STATES NITROBENZENE PRODUCERS,

                         LOCATIONS, AND CAPACITY11
          Company
     Location
 Annual
Capacity

(106 Ibs)
E.I. duPont de Nemours and
Company , Inc .
Beaumont , TX
Gibbstown, NJa
350
240
First Chemical Corporation
Pascagoula, MS
   300
Rubicon Chemicals,  Inc.
Geismar, LA
   427
Mobay Chemical Corporation
New Martinsville,  WV
   190
 Plant is on standby.

Note:  This listing is subject to change as market conditions change,
       facility ownership changes, plants are closed down, etc.  The reader
       should verify the existence of particular facilities by consulting
       current listings and/or the plants themselves.  The level of benzene
       emissions from any given facility is a function of variables such as
       capacity, throughput, and control measures, and should be determined
       through direct contacts with plant personnel.  These data on producers
       and location are as of January 1987.
                                      113
-------
                                                 0)
                                                 c
                                                 0)
                                                 N
                                                 C
                                                2
                                                +-•
                                                z
                                                «*—
                                                o
                                                
                                                O
                                                O
                                               O)


                                                
-------
     As shown In the figure, nitric acid (Stream 1) and sulfuric acid
(Stream 2) are mixed before flowing into the reactor.  Benzene extract
(Stream 6), two recovered and recycled benzene streams (Streams 7 and 8),
and as much additional benzene (Stream 9) as is required make up the benzene
charge to the reactor.

     For the process depicted here, nitration occurs at 55 C under
atmospheric pressure.  Cooling coils are used to remove the heat generated
by the reaction.  There is evidence to suggest that other methods/conditions
may also be used in the continuous nitration i
whether these processes are currently in use.
may also be used in the continuous nitration process.    It is not known
     Following nitration, the crude reaction mixture (Stream 3) flows to the
decanter,  where the organic phase of crude nitrobenzene is separated from
the aqueous waste acid.  The crude nitrobenzene (Stream 12) subsequently
flows to the washer and neutralizer, where mineral (inorganic) and organic
acids are removed.  The washer and neutralizer effluents are discharged to
wastewater treatment.  The organic layer (Stream 13) is fed to the
nitrobenzene stripper, where water and most of the benzene and other low
boilers are carried overhead.  The organic phase carried overhead is
primarily benzene and is recycled (Stream 7) to the reactor.  The aqueous
phase (carried overhead) is sent to the washer.  Stripped nitrobenzene
(Stream 14) is cooled and then transferred to nitrobenzene storage.

     The treatment, recycling, or discharge of process streams are also
shown in the flow diagram.  Aqueous waste acid (Stream 4) from the decanter
flows to the extractor, where it is denitrated.  There, the acid is treated
with fresh benzene from storage (Stream 5) to extract most of the dissolved
nitrobenzene and nitric acid.  The benzene extract (Stream 6) flows back to
the nitrating reactor, whereas the denitrated acid is stored in the waste
acid tank.

     Benzene is commonly recovered from the waste acid by distillation in
the acid stripper.  The benzene recovered is recycled (Stream 8) and water
carried overhead with the benzene is forwarded (Stream 11) to the washer.
                                      115
-------
The stripped acid (Stream 10) is usually reconcentrated on-site but may be
  1.1 13
sold.
     Typically, many of the process steps are padded with nitrogen gas to
reduce the chances of fire or explosion.   This nitrogen padding gas and
other inert gases are purged from vents associated with the reactor and
separator (Vent A in Figure 19),  the condenser on the acid stripper
(Vent B),  the washer and neutralizer (Vent C), and the condenser on the
nitrobenzene stripper (Vent D).

Benzene Emissions from Nitrobenzene Production

     Benzene emissions may occur at numerous points during the manufacture
of nitrobenzene.  These emissions may be divided into four types:  process
emissions, storage emissions, fugitive emissions, and secondary emissions.

     Process emissions occur at the following four gas-purge vents:  (1) the
reactor and separator vent, (2) the acid stripper vent, (3) the washer and
neutralizer vent, (3) the washer and neutralizer vent, and (4) the
nitrobenzene stripper vent.  The bulk of benzene emissions occur from the
reactor and separator vent.  This vent releases about 3 times the level of
benzene released from Vents B and D (Figure 19), and about 120 times that
released from Vent C.  For all these vents, the majority of VOC emissions is
in the form of benzene.  Benzene accounts for 99, 100, 76, and 99 percent of
total VOC emissions from Vents A, B, C, and D, respectively.  Table 21 shows
estimated emission factors for benzene from these sources.

     Other emissions include storage, fugitive, and secondary emissions.
Storage emissions (G) occur from tanks storing benzene, waste acid, and
nitrobenzene.  Fugitive emissions of benzene can occur when leaks develop in
valves, pump seals, and other equipment.  Leaks can also occur from
corrosion by the sulfuric and nitric acids and hinder control of fugitive
emissions.
                                      116
-------













M

O
H
U
^J
fa
Z
0
1— 1
V)
w
M
^F^
H
BENZENE

a
w
.J

o

H
Z
O
o

a


a

»J
^J
§
H
z
0
u
z
i-H
CM
Cd

03
H


























m
r-t
V)
H
5
j
CLl

Z
o
M
H
U
I
OL,
Cd
z
Cd
N
Z
Cd

O
C*
H
M
Z

J

H- 1
H
Cd
X
H
2
^H
•C
as
£















u
0
u
jj^n
cS
0 <.
-t »
n
I
*t
3 g C
_| _4 Q
O n -H
M n u
4J -H 0
C S 3
O [3 T3
u v
**








t)
* 1
4) -4
:i
8fl
4J
C M
3 °





O
•O 4-1
V U
^ (•
IS0*
u O ~-
83*
C "*
= 1




"c
S.2
*j
S I
V3 -H
i











1






»0 •* »rt
l l r* •» i l C*J tr>
1 l ^> t-l 1 1 -* »O O O» d
i i o o i t o »-t ^4 «n c*i

o o o o o o o



1 1 *6 O 1 1 ^ III
1 1 • • 1 1 • III
1 1 -» & 1 1 IA f» III
O* O* 00 *O





U
>« c
8-g -j ^
6 « "
u u

u c* 8 S
U — ' tj r*
N M fcl O
M -1 4 O «
« -a oos
S> ~t -t
U X *H "O W 4^
U 00 C 3 C
c « -< 3 -^ 3
it j -4 « a -i
f^ 44 C *J Ou
o. E Corn

>l I C « 1 IUW44CI IU
wl i V.CI I C « • « 0 i 1-1
- H 1 1 I-H Q -1 «1 Z 1 1 U
2 S
u

£b ^
^. «
£ z
rl O
oooooooo*-icn o
o«or^or*fvioaB« oioino
OOtt-IOrHOOC4O tH-»-*
o
OOOOOOOOO OMCMIO
h M
0 0
(•< b«







<;»OQOOOX n


u
«
N
C K X
o -o o o
u -4 u ii n
O 4J II 00 .£ -4
w M « a « -o «4
• « N a. u « e
u a -i -i o t* ue
« ^o. _i u u j -o e_S
«UU« OK > *J

"n3«e«2 Si £
T) II C II O U U
C T) C « N -O U _l _l
3-iNC-i«io x j 3
o -a c « o • t*
U4g«J94«> « _4 _4
OI4J) IC t 1
I i > H ! !





^
O O 00 O f^ CO
at ^ o «H o o

d o o o o o









•< « u a o o





c
0 T3
U -1 VI w
0 4-1 41 00
u b n a n -a
« «i N a. u D
u a. -^ —i o «o
S, 5 * « ti 2
« U U M 0
«W*J « JJ
^ " S S £ -
S 2 c S S 2
O T) C V 0
U 4 C V A «
O l « -Q 1
u « o — * a;
o u j: n -4 u
V 4 4 -4 E 
-------












^^
•o

3
q


u
c
O
W U
04 >-'
O
H f>
^J ^^
< co
fa H

Z <
O J
l-l CL,
w Z
M O

W HI
u
U 5
Z O
u o
N aS
CQ Cd
o w
Ex] N
d §
O oQ
fVt O
H OH
Z H
O M
CJ Z
Q J
5 o
Q H
U Cd
J X
J H
0 0
Z S
o
CJ p^
Z O


.
*"^

u
ca
H

























T)



0
14
IJ
J















^
"o

B
O
u
=


























u
o
u
u
4


0 *-

W
•4
i§




JJ
B
0 B
** O
« *4
•2 o
i






V
o «

« •§"
Q tf
2V
H
C U







-4 0*

•* O fi en C4
o ...
0 O 0 O
o









r» til
i i i
to r*. iii


•3 .
v «

<•
2 "
U — t
b (|
•4-t 0 U
g ° -s
U T) 4
ao So
B |

4 II n V
O u 3 B > I
-H II -4 0 1 1
b. Q a Z 1 1

00 00 O ON 9
1 n. *"!""! I
O O O W N






OS n






u

N
U -4
« -o
ja -4
b X
0 O
J3 -1
• n

•a B «
« II X


u XX
O *J 4J
So x » s
« ? S -, _
6 -4 -044
V U BUM
N -4 0 O O
B 80 O t"4 f"i
« 3 II
ah. in
—~ H
in o
111 II W
14 > 0
l< -4 fl
0.1*4  * •

8ti
X
Si • u
O X -4
« T>
TI a «
S 1 1
S& «-4


•! <
fi •>
- xl

i.i -* C W
•^ C ^4 v4
S 3 ?
•0 "c K **
o n • «
UK X
a SH
B ii S • 4J
S S S S 9
S t "•& £
J3 tl 4 L
0 - U «l -1
u e 4 «i ii
u i u -e
B . B S S £
> O II II
ec -. N u n
j: u -0 C « >•
B i> ' 1 *
a *< j* e •
•4 a. o
•o <« o
V O -Q M 0
t! . -4 S 1 e
•g s =» w *
HUM • U
3 V4 ? tl 0
SO -4 « h.
tl -4 II X
» s - -1 -°
r4 B 4 « B
II -4 u O
S "° B * -0
3 "M II « II
<0 O U « 1
£ „ 0 j> J«
o § • a > «
u u u u
It 4 O T) 4
U IN C
« n « 4 n
1*4 « II Tl b
« > S a 0
U T> -H 4 ft, W
II 4 6-0
3 it > - 3 < 4
5 s i i .3 B
U fl1 "-3 :-8 S
B « « > X II «
3> a w a •
•5 S I 5^5- e
ii 4 a « o «
T) OB
n « o W -4 •
B u n b B »
3 0 « a V B 4
« U O 0 « H
U O O O U > 0

4 J3 0 T)



































^0°
1 s
S3
J 4 O O
K CM tM
2 S S
H CL
«
Q
£
*- LI
S -i
V U ^
r^ <• ^ ^O X
o a CM c*t
C > CM O
C o
^ °
o
to
•H
M
£
S
^ o -»
 a




































u
•0°
3 S
O1 3
-4 W
Bulk L
Tempera



M
U
II 14
g **
3
t*

II
N

U}
^
S
H















                                 -s   -s
                                      c
                                      V
                                     a
118
-------
     Secondary emissions can result from the handling and disposal of
process waste liquid.  Three potential sources of benzene storage emissions
(J) are the wastewater from the nitrobenzene washer, waste caustic from the
nitrobenzene neutralizer, and waste acid from the acid stripper.  Where
waste acid is not stripped before its sale or reconcentration, secondary
emissions would be significantly affected (increased), unless the
reconcentration process is adequately controlled.

     Table 21 gives benzene emission factors before and after the
application of possible controls for two hypothetical plants using the
continuous nitration process.  The two plants differ in their plant
capacities; one plant produces 90,000 Mg/yr and the other produces
150,000 Mg/yr or nitrobenzene.  Both plants use a vent absorber or thermal
oxidizer to control process emissions in conjunction with waste-acid storage
and small benzene storage emissions.  The values presented for the main
benzene storage emissions were calculated by assuming that a contact-type
internal floating roof with secondary seals will reduce fixed-roof-tank
emissions by 85 percent.  The values presented for controlled fugitive
emissions are based on the assumption that leaks from valves and pumps,
resulting in concentrations greater than 10,000 ppm on a volume basis, are
detected and that appropriate measures are taken to correct the leaks.
Secondary emissions and nitrobenzene storage emissions are assumed to be
uncontrolled.  Uncontrolled emission factors are based on the assumptions
given in footnotes to Table 21.  The controlled emission rates from these
hypothetical plants range from 0.22 kg/Mg to 0.39 kg/Mg.  Actual emissions
from nitrobenzene plants would be expected to vary, depending on process
variations, operating conditions, and control methods.

     A variety of control devices may be used to reduce emissions during
nitrobenzene production, but insufficient information is available to
determine which devices nitrobenzene producers are using currently.  Process
emissions may be reduced by vent absorbers, water scrubbers, condensers,
incinerators, and/or thermal oxidizers.  Storage emissions from the
waste-acid storage tank and the small benzene storage tank can be readily
                                      119
-------
controlled in conjunction with the process emissions.  (A small storage tank
contains approximately one day's supply of benzene; the larger tank is the
main benzene storage tank.)  In contrast, emissions from the main benzene
storage tanks are controlled by using floating-roof storage tanks.  Fugitive
emissions are generally controlled by detecting and correcting leaks,
whereas, secondary emissions are generally uncontrolled.

ANILINE PRODUCTION

     Almost 97 percent of the nitrobenzene produced in the United States is
converted to aniline.    Because of its presence as an impurity in
nitrobenzene, emissions of benzene may occur during aniline production.
Therefore, a brief discussion of the production of aniline from nitrobenzene
and its associated benzene emissions is included in this report.  Table 22
lists the United States producers of aniline and the production method.  A
process flow diagram is shown in Figure 20.  As shown in the figure,
nitrobenzene (Stream 1) is vaporized and fed with excess hydrogen (Stream 2)
to a fluidized bed reactor.  The product gases (Stream 3) are passed through
a condenser.  The condensed materials; are decanted (Stream 4) and
non-condensible materials are recycled to the reactor (Stream 5).   In the
decanter, one phase  (Stream 6) is crude aniline and the other is an aqueous
phase (Stream 7).  The crude aniline phase is sent to a dehydration column
that operates under vacuum.  Aniline is recovered from the aqueous phase by
stripping or extraction with nitrobenzene.  Overheads from the dehydration
column  (Stream 8) are condensed and recycled to the decanter.  The bottoms
from the dehydration column, which contain aniline, are sent to the
purification column.  Overheads (Stream 10) from the purification column
contain the aniline product, while the bottoms (Stream 11) contain tars.

Benzene Emissions from Aniline Production

     Process emissions of benzene typically originate from the purging of
non-condensibles during recycle to the reactor and purging of inert  gases
                                                             14
from separation and purification equipment (A in Figure 20).
                                      120
-------






4
4
C/3

H"4
H
1— i




^
O

a


V)
Z
o
M
£j!
o
3

at!
1—4

X
H
W
Z
3
M
O

Crt
a!
3
^
p
a
o
£

en

^

co
a
w
H
M

CM
CM
S
CQ
<:
H




/

» n
II


11








n
X B

O O<
«
a 
«
n
















Ihemlcal Corporation
u
•8
*
«J
VI
*4
U













































.



8

»
1
a
it de Nemours and Con
B
1
•O
HI
hi


i i
N M
S S
M A
O O
t4 U
u w
~4 »4
Z 9E













4
O O

111?
." .-S


• H 3 -3
- o • •
u «i 6 «
e e «
- -4 O •
•a • M n
• « w 4
• SB
o S o «
-1 £ O C
g *• o
• • *B «<
» u 3 «
« c u
3 • o


O.U 3 *
O M
• W M 14
« o x e
fl ' '
« 8
o «- a
a B • B
-4 -4 a o
J3 W «
• • o «
S •* 2 5
c • ^
3 4J
0 B -C e
• on
>. M 3 •
4J M « J3
-< 3 H
^ O •!
•^ •
0 M -1 •
» S jo -H
>H -4 « •
M -4 B
- -i u H
• 3 • O
ia:s
.e o o a
o

§" "° 5 3
« «j ^
-4 « « a
" w 3 J3
•§ ^ •** "
0 -> » »
U 0
4J 4*4 «4 4J
• O
J< U IK •
U 4 4J *J
8-1 -4 6
3-40
0-40
tl -4 O
» u » a
b IH a
&8.B S
I'sl*
o <* js
O u X 3
4J B B O
e 3 b
s listing Is subject
uld verify the exist
zene emissions from
uld be determined th

Slil
4J
O
z
121
-------
                                o>
                                o>
                                CD
                                o
                                o
                                0>
                                X
                                0
                                O 
-------
     Only one emission factor was found for benzene emissions from aniline
production.  For process vents (A),  an uncontrolled emission factor of
0.0057 kg benzene/Mg aniline produced was reported in the literature.    No
details of its derivation were provided, other than it was based on data
provided by an aniline producer.

     Control techniques available for emissions associated with the purging
                                                                 13
of equipment vents include water scrubbing and thermal oxidation.    No data
were found to indicate the efficiencies of these control devices for benzene
emissions.  The reader is urged to contact specific production facilities
before applying the emission factor given in this report to determine exact
process conditions and control techniques.
CHLOROBENZENE PRODUCTION

     Of the chlorobenzenes, mono-, di-, and trichlorobenzene have important
                        18
industrial applications.    Therefore, this section of the report focuses on
benzene emissions associated with production of these three types of
chlorobenzenes.  Table 23 lists the producers of mono-, di-, and
trichlorobenzene in the United States.  The producing companies'
capabilities are flexible such that different chlorobenzenes may be
isolated, depending on market demand.  Di- and trichlorobenzenes are
produced in connection with monochlorobenzene.   The relative amounts of the
                                          19
products can be varied by process control.    Figure 21 shows a process flo
diagram for the manufacture of mono-, di-, and trichlorobenzenes.
     The benzene is dried and then transferred to the dry benzene feed tank
(Stream 1).  Chlorine is also moved from storage to the feed tank
(Stream 2).  The dry benzene is reacted with the gaseous chlorine in the
presence of an iron turning catalyst (Stream 3).  The quantity of chlorine
added to the reactors determines the product distribution.  Hydrogen
chloride is generated through the chlorination reaction.  This material is
sent to the hydrogen chloride scrubber (Stream 4), where it is scrubbed with
                                      123
-------


















•4
H
Cd
55
W
N
55
[aj
CQ
§
S
33
O
M
cd
£-4


4^

-
1
M
Q

»
1
0
z
o
£

pLj
O

CO
Oi
W
u
JID
a
o
85
CO
Cd
H

g_i
CO
a
Cd
£-4
M
53
p3

.
PO
CN

Cd
rJ
<
















£
•r-l
0
cd ^
ex 45

u
VO
r-4 O
cd i-i
3 ^"
C
C
^















4J
O
3
•d

PH





















o
1-1
4J
cd
o
o
^




















s*.
C?

ex
o
u











VO O U"»
r*. 1-1 CN
I-l


















000
CCC
000
N N N
CCC
000
45 45 45
O O O
O O O
r-4 r-4 r-4
£ £ rC
O O CJ
O iH ~H
G a o
O • i
£ o p.
















H- 1

M
•U
0
60
5
CO









{*,
C
cd
fr
o
u

r-i
cd
CJ
1-1
B
0
CJ
O
4J
c
S
en
i






in o o
*tf CN CO



















000
CCC
000
N N M
CCC
000
45 45 45
O O O
O O O
r-l r-l r-4
4: £ X
O U U
O i-l -^
coo
O i i
£00.
















£
•.
S
3
•H
M
4J
(0














•
a
C
M

n
' en ex

•^ o
4J CJ)
en
3 r-l
•o cd
C 0
M 1-4
S
S£
PL, CJ
en r-4
* 0 cd 0
§4: c
4J 4: C
0 O U O

o o m < < CM CM <£ 00 o en n
mifir^zz Z NUTS 0
r-i C cd 0 en - ex
0 en 00 en
45 en O C 0 •*•>
O 4-> r-l -H t-4 C
)-i en o 4-> 45 cd
O 1-1 en cd r-i
I— 1 r-4 0 i-l i-l CX
4: r-l r-l r-l
U 0 Cd Cd JS
T* tit 4-> > W
•O 0 CQ C i-l
i 4: 4-) 0 CM s
a EH C u O
cd ^i ca
T3 r-l 3 C -U
C • (X u O cj
00 0 cd 0 >H cd
CC C C >-i e>0 -u 4J
00 0 I0OCOC
NN ta o N t-i c o
CC C C - 4J 3 0
00000 00 0 00cnr-i(4-i
CCJC4545 CC 45 4J4503 4J
00000 00 o cdocvocncdcj
NMNMV4 N N M r-4 l-l C C 0
CCCoo CC o oocdocnvj
0 4> 0 r-l r-l 00 r-4 Cnr-I^CJi-l-H
454:i454:rC 4545 4: -H 4: CJ "O
OOOOO OO 0 0 >-,>^
O Ci O >-i )-i OO )-i 4J >-i -r-l i-lbO
rHr-ir-iHE-i 1-1 i-i H 4Jj:enr-i3
X 4:', 4S i i 4:4: i 0 M0-HO
ucicjcn^ ocj in C -o n -r-i o >-< •
Oi-i^-- ^1-1 - 0 c a» 4-> cd ,c r-.
C O Q CN CN Q Q en NCdCi-l<4-l4-)ca
QII--II - C ?r-i o\
£ 0 CXr-l r-4 OCXi-4 0 -O-r^C'Oi-l
45 i O 0 0
0 -H >, cd > C >,
(^ -O JJ (4-1 i-l i-l in
O *rl ClO S Cd
r-l -r-4 M H 3
4: • 1-1 cd >, 0 c
CJ O CJ iH C 4-1 Cd
C «d 3 cd 0 i-j
(4-1 O <4-4 CJ -O
OB 1-1 S M-i
. 4J 0 0 0

O Z a> e c>o co IM w
>-i i-i c o« *o cd
3 U cd en i— i
^% ^ en 4J 3 -C *4-4 C 3 0
•U U 3 X T3 O O O 0 l-i
u - •o a s-i M 0 tn M
0 cd ex c u en 01
0 r-l 0 Ol O C "-I T> C
I-i cd £ 0coi-i0BCO
cd T3 en04J4J0cdiH
S C J-J cdi-ii-ien-u
cd in cd 4: 4-i *d IH M-I - cd
r-i S 0 cji-icxocnu
0 1-1 (-4 1-1 r-l O 0 0) O
Q r-l O 3-^OrHl^i-l
(X CJ 003
cd 4J 4: ^ en *d
•O <4-t 0 4J 0 Cd C
C -*! r-l 0 Cfl
cd )-i >-i >-. S
O cd (4-1 0 tn
V4 1-1 B 1-1 4: r-i )H
O ca rj H 0 0
en S en 0 i-t o
en cd > 4J 3
a) IM • C "O
U O 0 T3 0) O O
U O bOr-l 0 O >-l
C U en C 3 > ex
M C CX4JcdOi-4T)
O • en 45 4! 0 C C
1-1 0cdi-4cjencncdo
en 4J r-i r-i B
0 cj cd C 45 01 o)-i04Jcd
CX *-i V4O cdi-i0iJ04:34J
•HC C O-H rH WT34JCXcd
j-i cd cd ex 01 T-I >i cd 4J ca j; "O
o a oo V4-H cdiJ cj0cnbo
r-l 8 ^ O > >i-400V4-U30
jSO O OIH ,cd i— i en i-> C O en
x -o 4Jcj4:34:r-ij3j:
•O r-l 4J C r-l OCflHOlHCXWH
U Cd r-4 Cd Cd C <4-4
cd u cd i-i cj
•d 1-1 -^ 43 1-1 1 en
CB 0 4JB ^•^a)
cfl 
-------
  s
  CD
Jo.
oS
*•£
  CD
           5
          2o

        c
mz
  o
                                                      U/ZOOZ8C
                                                           m
                                                           m
                                                           o
                                                           o

                                                           2
                                                           Q.
                                                           CO
                                                      x co
                                                      (D C

                                                      • •2

                                                     Is
                                                      = 6
                                                      c «
                                                     ^ CD
                                                       g-
                                                     11
                                                     o o
                                                     
                                                                     o
                                                                     (D
                                                               0)
                                                               1=
                                                               (D
                                                               C
                                                               
-------
                                                                 20
benzene or chlorobenzene (Stream 5) to remove entrained organics.    The HC1
is then absorbed in water (Stream 6) co produce commercial grade HC1
(Stream 7).
     The main chlorobenzene product is sent to the neutralizer (Stream 8),
where a 10 percent sodium hydroxide solution is flushed through the crude
chlorobenzene to neutralize any remaining acidity.  The product is then sent
to the settler (Stream 9).  Much of the dichlorobenzene is retained in the
sludge which is sent to recovery by distillation (Stream 10).  The liquid
portion is sent to fractionation columns (Stream 11).  The first step is to
remove the unreacted benzene, which -is recycled to the dry benzene feed tank
(Stream 12).  Monochlorobenzene is separated from the bottoms of the benzene
                                       •
column (Stream 13) and transferred to storage (Stream 14).  The remaining
product is then fractionated (Stream 15) by distillation.   The separated
products are run to dichlorobenzene scorage (Stream 16) and trichlorobenzene
and polychlorinated aromatic resinous material storage (Stream 17).

     The primary source of benzene emissions during chlorobenzene production
is the tail gas treatment vent of the tail gas scrubber (A in Figure 21) .
Usually, this vent does not have a control device.    Other potential
sources of benzene emissions are atmospheric distillation vents from benzene
drying column and benzene recovery columns (B and C), process fugitive
emissions, and emissions from benzene storage.

     Table 24 presents estimated controlled and uncontrolled emission
factors for benzene emissions from the tail gas treatment vent, atmospheric
distillation vents, fugitive emissions, and benzene storage.  The point
source factors are based on emissions reported to EPA in response to
information requests and trip reports.    The fugitive emission factors  are
based on factors for petroleum refineries.    Storage emission factors are
based on AP-42 factors for storage tanks.    The Appendix provides more
detailed descriptions of the derivation of the fugitive and storage emission
factors.  As noted in Table 24, carbon adsorption is an appropriate control
technology for control of emissions from tail gas treatment and distillation
                                      126
-------











o
r-l
00
H
0

O
O
OS
Qd

Cd
Z
Cd
N
Z
Cd
03
q
a*
3
EC
0

o
z
r-l
OS
£3
^
o
^J
r*4
e?
2j
2
5E

H
**5
^5
t_j
&4

rJ
^4
O
M
H
Cd
X
o
fc
SH
X

<;

£
O
OS
f&4

C/i
Z
o
h-l
C/3
CA
rH
£
CM

Cd
z
Cd
CN
Z
Cd
CO


•
N^
CM

a
eo
H









rl
O
4-1
•o y
cu cd
r-l Cb
o c: &o
rl O X
^ \ T^| ^v
C en oo
O en
O -rl
a
Cd




C
s o
O i-l
i-l 4J ^x
en y dp
en 3 v^
i-l T)
E 0)
Cd OS







cu
y cu
•H 3
> cr
CU i-l
0 C
X
r-l U
o cu
rl H
4J
C l-i
o o
o






rl
o
*O 4J
cu y
i-l cd
rH PuQ
O 00
>-i B r^<
4J O v^
E -rl W>
o en
y en
B 1-1
D B
Cd




^
B
O

rt u
cu cd
V4 C
4J 00
C/3 »f"4
en
0)
o










CU
y
u
o
CO
B
o
1-1
en
en
•i-i
&









f**
VO
O
o
•
o










r^*
•
00
CT\









c
o
•fl
4J
a
W
0
en
•o
cd

C
o
^
v^
CQ
o










CM
m
o














^

















4J
C
CU
B
4J
cd
cu
i-i
4J

en
cd
00
rH
•H
(0
H





CM
O
O

o










r*^
•
CO
ON









B
O
*»H
4_>
a

o
en
•o
cd

B
o
,0

cd
o










CM
CO
O













u

_
03









B
o
•H
4J
cd
r-l
r-l
i-l
4J
en
•H
T3
y
•r^
1-1
0)
a
en en
O 4J
E C
4J CU
<3 r*






vO
r-l
•
O












f^
p^







•I-l
cd
a
0)
^4
en
•O Ji

cd cu
r-l
C
O >-i
•H O

u cd
CU E
4J
CU 4-1
a o










OO
m
o













i
i
1
























cu
1-1
4J
i-l
00
3
b






vO
O

O












in
00









00
B
1-1
4j
cd
0
i— i
4-1

rH •
cd
B
rl
CU 4-4
4-1 O
B O
r-l Wl










r-l
^
O













i
i
i


















0)
00
CO
l-i
0
4J
en

0)
B
0)
N
B
cu
00









































































.
r-l
CN

CU

3
oo
•rl
U.

0
4J

}^
cu
4-1
cu


en
B
O
•rl
cd
B
00
1-1
en
cu
TJ

E
id
cu

4J
CO
cd .
cu
B
•O cu
CU N CU
en B ri
cd cu o
,0 .0 ««-i
o cu
CU 1-1 -O
rJ O
CO r-l -O
rC CU
en y en
1-1 *r* 3
0 -0
4J i en
y O. cu

4-1 oo cr
C3 -H
cu B
en vO £.
cu • y
X m cu
H r-l 4J

•O r-l
• B 0
T) co U
a) 4-i
y - B
3 cu O
T3 E y
O cu

a B B
cu CD
u ,a
y o -
3 l-i cu
TJ O T)
O rH Cd
1-1 rfi S
tx y
•H en
cu -O 4-i
B i y
cu o 3
N T)
C M) O
CU O l-i
JD a.
o -*
>-i • -
O CM en
rH rH en
rB CU
y - u
cu O
00 B rl
r* « O,
N
l-i B O
CU CU 4J
CuX)
O en
T) w cd
cu o
4J rH 4J
4-> rB B
•rl U  -rl
cu r-i en
en cx TJ en
en cu -H
CU rH 00 E
l-i CO l-i CU
a y 3
X -r* CU
cu 4.) en en
eu 1-1 cu
cu 43 £
rl 4J 1-1 4-1
CO O CU
Q. T) 00
en >-, id B
1-1 rB CU -H
O u >»,
4J Cd i— 1
y a) ex
co c j= a
fa O H cfl
Q
127
-------
column vents.  The control technique applicable to process fugitive
emissions is an inspection/maintenance program for pumps,  valves,  and
flanges.  Internal floating roof tanks may be used to control benzene
emissions resulting from benzene storage.

LINEAR ALKYLBENZENE AND BRANCHED ALKYLBENZENE PRODUCTION

     Approximately 2.5 percent of benzene  production in the United States is
used in the production of linear alkylbenzene (LAB).   Linear alkylbenzene,
or linear alkylate,  improves the surfactant performance of detergents.   The
current demand (including exports) is 565  million pounds (256 million kg);
projected 1990 demand is 635 million pounds (288 million kg).  Thus, there
is a 3 percent projected annual growth rate.  The locations of the linear
alkylbenzene producers in the United States are shown in Table 25.

     Monsanto's Carson plant, Vista's Baltimore plant,  and Union Carbide's
Institute plant use a monochloroparaffin LAB production process; however,
Union Carbide is currently not in operation.  Vista's Lake Charles division
and Monsanto's Chocolate Bayou division use an olefin process, wherein
hydrogen fluoride serves as a catalyst.

     The paraffin chlorination process accounts for about 64 percent of the
LAB production in the United States.  Approximately 36 percent of linear
alkylbenzene is produced by the olefin process.

Production of LAB Using the Chlorination Process

     The linear alkylbenzene chlorination process takes place with two
reactions.   In the first step, n-paraffins are chlorinated to, monochlorinated
n-paraffins.  In the second reaction, benzene and crude chloroparaffin is
blended with an aluminum chloride catalyst forming crude linear
alkylbenzene.  The following discussion of LAB production using the
chlorination process is taken from Reference 13.
                                      128
-------












H
-i
V)
Ed
H
i
|2
w
C5
cd
Cd
H
Q


Cd
Z
Cd
N
Z
Cd
1
£bi
o
CO
u
o
D
Q
O
Oi
CM
CO
Ed
EH
<
CO
a
Cd
H
z
in
CM

Cd
i^
PQ
^
H























CO
CO
cu
CJ
o
Vl
cu






4J
O x-s
cd xi
CX i-l
cd
CJ V0
O
^J ™J
^^ ^^
3
B
B












B
O

4-1
cd
u
o
^

















^
B
cd
f^,
g
o
u









CM
CM
cd
H
cd
cx
o
VI
o
1-1
•g
§
> 0
£







cd o
<; m
Z
















^ ^
rj u

B" B"
O O
ca co
Vl Vl
cd cd
0 0





£ B"
cd cd
cx cx
s a
o o
0) 0 CJ
c
CU r-l CU r-l
N cd B cd
B U CU O
CU >-ii-l N >>i-l
X) C B B B B
r-i cd cu cu cd cu
^ Pi X! XI CX X!
_V B C3 rH B CJ
r-l O > O
 Vl
B CU
M S







m
CM
CM














§

A
cu

|

4-1
r-l
cd
BO









S*.
B
cd
cx
S
o
u

r-4
cd
u
•rl
a
cu
X
CJ

cd
4J
CO
1-1
>





















C^l
inl
r-l|












J

CO
cu
r-4
Vl
cd
6

cu

3































































in
r^
SO































































Jg
H
O
H
O
4J

4J
U
cu
*^^
Xi
3
CO
CO
1-1

4-1
CO
1-1
r-l

CO
i-l
£


•
cu
C
cu
N
B
CU

r-l

*S-
r-l
cd
•o
cu
JS
o
B
cd
VI


•o
C
cd
Vl
cd
cu
G
•rl
r-4

60
a
1-4
O
3

O
Vl
o.

co
cu
•ri
4J
T-4
f^
•^4
U
cd
tM

Vl
o

Vl
cu
•o
cd
cu
V4

cu
X






CO
4J
C
cd
	 i








CO
cd
F^ ^^
cxx

•
g
o
•o

^
cu
CO
o
i-i
CJ

cu

cd

CO
4J
B

r-l
0.

Vl
o
*
CO
cu
60
C
cd

u


cu
fj

Vl
o

CO
60
i-l
4-1
CO
1-1
TH

4J
fj
cu
Vl
Vl
3
u

60
B
1-1
4J
r-4
3
CO
B
O
U

>,
CXX>
Xi
co
Vl
cu

^
o

^^
4-1
•H
r-l
1-4
CJ
cd


•»
cu
60
C
cd
«-j
CJ

CO
C
O
1-1
4-1
i-l
CO
cu
1-1
4-1
•r4
r-l
i-l
CJ
cd


Vl
cd
r-i
3
U

4-1
Vl

cx

IM
O

cu
0
B
cu
4J
CO
i-lTJ .rl
cd
B

CM
O

4J
CO
i-l
,—4

«

cu co

Xi
cd co
r-l i-l
•rl JS
a H
cd
4-1 CU
O 4J
Z 0
cd z
B
O
CJ

4J
CU

Vl
cd
5

CO
cd

cu
60
c
cd
o





p{
cu

cu
Xj


£>>^
CM
•rl
Vt
CU


•t)
r-l
3
O
X
CO





o
3
CO
*
CO
cu
r-i
Xi
1-1
^_)
cd


tM
o

B
O
i-l
4-1
0
B
3
IM
cd

CO


^^
4J

i— 1
i-l
U
ed
(4-1

rj
cu
>
1-1
60

^"*
B
cd

B
O
M

CU
I-l

0)
p*l
£•(


.
CO

^
i-4
cu
CO
B
-------
     First, n-paraffins are transferred to the n-paraffin feed tank
(Stream 1) (Figure 22).  The n-paraffins are dried (Stream 2) and routed to
the n-paraffin feed tank (Stream 3).  The paraffins (Ln excess) are then
sent to the UV catalyzed chlorination reactors (Stream 4) where they are
blended with liquid chlorine (Stream 5).  The reaction forms
n-chloroparaffins (Stream 6) and HC1:
     Rn - CH. - R0 + Cl ---> Rn - CH - R0 + HC1 + heat
      1     f.    f.            1    I     2.
                                  Cl
     Benzene is moved from the bulk storage tank to the feed tank  (Stream  7)
and is then dried in the benzene azeotropic column (Stream 8).  Some benzene
emissions can escape from the vent in the column (A?).  The quantity is
dependent on the dryness of the benzene and design of the column condenser.
The dried benzene is then transferred, to the dry benzene feed tank
(Stream 9).

     The crude chloroparaffin (Stream 10) and an excess of dry benzene
(Stream 11) are mixed in an alkylation reactor with an aluminum chloride
catalyst.  The subsequent reaction produces linear alkylbenzene, illustrated
below:

R, - CH - R0 + C-H, 	> R- - CH - R0 + HC1 + heat, possible olefins,
 I    |     i.    o 6       1         2.
      '                                        short-chained paraffins, etc.
     OJL

     The HC1 gas and some fugitive volatile organics are treated in  a VOC
absorber (Stream 12) and then an acid absorber (Stream 13).  Most  of the
product goes to hydrochloric acid storage (Stream 14), but some is vented
off (A.).  The amount of benzene emissions given off here  is dependent on
the fluid temperature in the absorber and the vapor pressure of the  mixed
absorber fluid.

     The main product from alkylation are transferred to the catalyst slurry
settling tank (Stream 15) to separate crude LAB and catalyst sludge.  The
sludge is hydrolyzed (Stream 16) to separate the aluminum  chloride from

                                      130
-------
                                                        £
                                                        •«_>
                                                        c
                                                        T3
                                                        O
                                                        CO
                                                        CO
                                                        (I)
                                                        O
                                                        O
                                                       O>
                                                       (0
                                                       5
                                                       CN
                                                       CD
                                                       a
                                                       g>
                                                       LL
131
-------
organics (benzene, LAB, tar, etc.).  The wastewater (Stream 17) is sent to
the treatment system; the organics are sent to a storage tank (Stream 18) to
be used as fuel or sold.

     Alkaline water neutralizes .the LAB (Stream 19) and is then decanted
(Stream 20).  Wastewater again goes to the treatment system (Stream 21),
while the product undergoes a second wash (Stream 22) which is again
decanted (Stream 23).  Wastewater goes to the treatment system (Stream 24),
while the washed LAB is sent to a series of stripping columns (Stream 25).

     The benzene is removed by the benzene stripping column.  The removed
material is sent to the benzene feed tank (Stream 26),  Residual inert gases
and benzene can be vented (A,); the amount of benzene in the stream depends
on the quantity of inert gases and temperature and design of the reflux
condenser.  n-Paraffins are then stripped off (Stream 27) under vacuum
pressure and sent to the n-paraffin feed tank (Stream 28).

     Alkylbenzene and the by-products are removed under vacuum conditions
(Stream 29) and are transferred to the by-product storage tank (Stream 30)
to be either used as fuel or sold.

     The LAB distillation column (Stream 31) separates the "overhead" LAB
from the "bottoms" or residual high-boiling materials.  The overhead LAB  is
sent to the LAB receiver (Stream 32), to the LAB "polish" column
(Stream 33), then to the finished storage tanks (Stream 34).  The "bottoms"
are then transferred to storage tanks (Stream 35) prior to sale.

Benzene Emissions from LAB Production Using the Chlorination Process

     Benzene emissions using the LAB chlorination process are shown in
Table 26.  The source designation listed in the table refers to Figure  22.
The four major sources of benzene emissions are the benzene azeotropic
column vent, the hydrochloric  acid absorber vent,  the atmospheric wash
                                       132
-------









n
•H
CO
co
w
u
o
&

z
o
I-H
H
^3
z
M
as
0
^J



u
e

o
z
1— 1
co
^

H
^J
S3

pQ
^
hJ

J
d
HH
H
W
e
o
CM
33

^
OH
""^S.
60
C3

O
ON

^3

06!
£

CO
Z
0
1-1
co
CO
HH
S
Ed
W
Z
W
z
H
A


•
vO
«N
U
I-H!
O3

Ol
B B
Ol S
N 9
Ol O
to u





00
ON
•
ON
ON















in
o
•
o









i-H
CU
UH
CO
CO

"O
Ol
CO
9









O
in
CN














PH












•a
• iH
O
4-1
O B
•H 01
IH >
O
i~^ V^
42 Ol
O 43
O M
M 0
•O CO
sT<





ao
ON
•
ON
ON













m
CM
O
o
•
o









i-H
01
9
<4H
CO
CO

•o
Ol
CO
D







m
•
cs
p-H















U













«•-
CO
CO
S 4J
B
O Ol
•H >
)4
Ol V^
j3 Ol
Gb jj
CO C
O CO
g O
4J 01
< a





ao
ON
•
ON
ON












«^
r-
O
O
O
•
O









1— *
01
MH
m
CO

*o
01
CO
a







r~.
•
en
















O












60
B
• H
O.
o.
.^4
M ^
4J B
co v
^>
Ol
B B
Ol 5
N S
B ^
Ol O
CQ O

•^
B
CO

CO
B
O

1— 1 4-1
CO CO
0 |H
•H 01
4-1 O.
CU 0
42
4-1 60
O B
Q.-*
^» 4J
42 4-1
••H
CO g
•H S
42
4J UH
O

Ol
IH O
UH 0
Ol •
l-< 4J B
O to o
IH -H IH
4J X UH
e 01 o>
0 tH
O 0) 0)
42 42
UH 4-1 4-i
O
eco
S
O> -H O
> UH -i-l
Ol B co
I-H O CO
O T-I
*Q fl
COO)
CO 4-1
60
S r-4 B
O 0) -pH
•H S 4-1
4- S CO
Cfl O S
^ CO -H
3 U 4-1
60 0) CO
•H O. O>
B 4-1 O
O S 4-1
0 CO
I-H |H
B O. O
•H -,H
4-1 IH
>, 0 G.
W QQ
« 4-1 >,
> B 4-1
O -H
X 0 i-H
CO -fH
BOO
4-1 CO
4J UH
C T3
CO Ol IH
I-H 60 CO
O. CO I-H
IH 9
60 9 0
B O -H
•H O 4-1
CJ B IH
3 Ol CO
T) S.
O CO
IH -H CO
CX
B 01 to
•H -O
^•4 CQ ^»
Ol 01 60
•-i M O
0 fH
0) O
ea 42 B
^ H 42
hJ O
0)
B . 4-1
0> >.
^ *J rH
•H -H O
60 I-H IH
•fH 4-1
>i o B
B co O
<; UH o
CO




























































«
CN
CM

Ol
^
9
60
• r-l
to
B
• iH

•a
Ol
4-1
CO
B
60
.fH
Ol
•o
ca

UH
Ol


CO
Oi
4-1

Ol
43
B
0
• i-l
4-1
CO
B
• H
IH
0
i-H
42
U

Ol
42
4-1
^
43

•o
Ol
o
9
•a
o

a.

CP
^
j
U-l
0



IH
60
CO
60
01

01
a.

•o
01
44
4-1
•H
8
01

Ol
B
Ol
N
B
01
43

UH
O



IH
60

O
4-1

IH
01
UH
01
IH

60
X

60

UH
O
- co
g
01
^J
B
•i-l

ca
IH
O
4-1
0
CO
UH

B •
O co
•i-i ca
ca 01
ca o
•rH O
t§ a.

133
-------
decanter vents, and the benzene stripping column vent.  The most frequently
applied control option for all these sources is to use the emissions for-
fuel.

Production of LAB Using the Olefin Process

     Production of LAB using the olefin process consists of two steps:  a
dehydrogenation reaction and an alkylation reaction.  The n-paraffins are
dehydrogenated to n-olefins which are reacted with benzene under the
influence of a hydrogen fluoride catalyst to form linear alkylbenzene.  The
discussion of LAB production using the olefin process is taken from
Reference 13.

     First, n-paraffins are transferred from bulk storage to the n-paraffin
feed tank in Stream 1 (Figure 23).   The paraffins are heated to the point of
vaporization and passed through a catalyst bed to form n-olefins (Stream 2)
by the following reaction:

     RX - CH2 - CH2 - R2 ---> RX CH -• CH - R2 + HZ

The resulting olefins contain from 10-30 percent alpha olefins, and a
mixture of internal olefins, unreacted paraffins, some diolefins, and lower
molecular weight "cracked materials."  The gas mixture is quickly quenched
with a cold liquid stream as it exits: to process thermally-promoted side
reactions (Stream 3).  The gases (e.g;., hydrogen, methane, ethane, etc.) are
then separated from the olefins (Stream 4), which are stored (Stream 5)
while the gas is used as process fuel (Stream 6) or vented to a flare stack.

     Benzene is fed from the bulk storage tank to the feed tank (Stream 7)
and is dried by azeotropic distillation (Stream 8) to remove all traces of
water.  An excess of benzene and olefins are transferred and mixed to the
alkylation reactor (Stream 9) to be blended with a hydrogen fluoride
catalyst (Stream 10).  The blend is held at reaction conditions long  enough
for the alkylation reaction to go to completion as follows:
                                      134
-------
135
-------
     R..CH - CHR0 + C,H,  --->  R,CH0 - CHR.
      1        /bo         1  i      i

The liquid hydrogen fluoride is then separated from the hydrocarbon in the
settler (Stream 11).  Remaining hydrogen fluoride is recycled (Stream 12) to
the alkylation vessel to be mixed with fresh hydrogen fluoride.

     A series of stripper columns extract the essentials from the
hydrocarbons.  The first column (Stream 13) removes benzene which is
recycled to the benzene feed tank (Stream 14).

     A lime water solution is then fed into the hydrogen fluoride scrubber
column (Stream 15) to neutralize the hydrogen fluoride.  The solution is
filtered (Stream 16); the wastewater is routed to the treatment facility and
the solids are transferred to a landfill.  Some benzene can be emitted
through the hydrogen fluoride'scrubber column vent (A.).  Inert gases and
air venting from the unit, temperatuie, and purge rate of the scrubber can
influence the amount of volatiles emitted.  These gases are usually sent to
a flare.  Unreacted paraffin is then stripped off (Stream 17) and returned
to the paraffin feed tank (Stream 18).  Any remaining by-product is removed
by the next stripping column (Stream 19) and is sent to the by-product
storage tanks (Stream 20).  The last distillation column purifies the main
linear alkylbenzene (Stream 21).  "Heavies" by-products are stored
(Stream 22) and the pure LAB is transferred to storage tanks (Stream 23)
awaiting sale.

Process Emissions from the LAB Olefin Process

     Benzene emissions from the LAB clefin process are shown in Table 27.
The two major sources of emissions axe the benzene azeotropic column vent
(A) and the hydrogen fluoride scrubber column vent (B).  The control for
both of these emissions is use as fuel.  Emissions from the hydrogen
fluoride scrubber column vent can also be flared.
                                       136
-------











CO
i-H
CO
CO
w
O
85

z
I— 1
fa
w

o
w
B
o
z
1— 1
CO
Ca

H
2
^
i_j
CM

03
^
J

^
cj
M
H
W
B
2
^4
BC

£
**^»
00
0

o
Q\

^J

OS
O
fa

CO
Z
o
f—l
CO
CO
1— I
as
w

u
Z
N
§
oa


•
f^
CN

W
hJ
PQ
^
H










Q
SMI'
B
0 B
•H O
CO T4
CO 4J
•H U
9 3
CM "O
CU






o
TJ 4J
01 U
i-i CO
i-l facj
O 00
v* e 5
C -H 00
O CO
CJ CO
•H
(1










1—4
0
Id
4-1
B
o





l-l
•O 0
CU 4J
^ 0
r-l CO
O faU
U 00
4-1 B £
B 0 -.
O -H 00
O CO
B co
a -H
s
w




^3
B
O
O) -H
U 4-1
I-i CO
3 B
0 00
CO -H
CO
cu
°








<0
cu
u
l-l
3
o
CO

B
o
•H
CO
co
• H
fl





co
ON
•
ON
ON











^f
l«^
O
O
O
o









r— 1
01
3
M-l

co
CO

01
co
^








p^
•
en
















^Q












u
••-i
(X
o
t-l
4-1
0
01 4-1
N B

T3
•H B

O 3
3 i— i
-H O
fa CJ

B M
01 01
00X1
0 JO
U 3
>. 0
» CO
T)
B
<0

CO
S
O
•H
r-4 4-1
CO CO
CJ Id
•H CU
4J O.
01 O
1"!
4-1 00
ac
•H
>» •«•>
XI 4-1
•H
co 3
• H vD
43
4J ^J
o
5 cu
Id CJ
01 •
l-l 4-1 g
O co O
U -H l-l
4-1 X M-l
B 01 01
O W
o cu oi

14-1 4J 4-1
0
g co
-4 t4 B
CU -H O
> •M-l -H
O) B co
•— 1 O CO
CJ -H
•o B
BOO)
CO 4-1
00
B i-i B
O 0) -H
•H C 4J
4J B CO
CO O g
V4 CO •!-!
31-14-1
OO 01 CO
•H CX CU
C 4-1 O
O B 4J
U CO
i-i Id
B (X O
••-1 -i-l
4-1 Id
>» CJ (X
Id CO
CO 4-1 >,
> B 4J
0 -H
SN CJ ^d
CO -H
g o u
4J CO
4J Ud
B -0
CO OI Id
i-4 00 CO
(X CO i-i

00 3 U
B O •*
•H CJ 4-1
O C V-l
3 CU CO
•o o.
o to
Id -H co
(X
Id 4-1
B 01 CO
•-I -o
14-1 CO >,
cu cu oo
1-4 Id O
O f-i
01 O
pq xj B
«? t"4 X!
hJ CJ
01
B • 4J
01 >.
> 4J ^H
•H -H O
00 ^d M
•id 4J
>> cj B
B CO O
•< 14-1 O
CO .


























































•
en

o

(X
pQ
2
IH
o

S
c3
Ij
00
CO
00
01
g

M
CU
(X

•o
01
4-1
4-1
•H

0)
cu
cu
N
B
cu
Xt

>4d
O




00

o
4-1

l-l
OI

01
Id

00

«*«^
oo

W-l
o
co
g
cu
4-1

B
•H

CO

o
4J
CJ
CO


B •
O CO
•H CO
co 0)
CO CJ
• H O
W CX
U
137
-------
MALEIC ANHYDRIDE PRODUCTION
     Maleic anhydride is now produced in the United States using n-butane.
In the past, a benzene based process had been used.    For reference, a
brief description of the benzene based maleic anhydride process is provided
here.
Process Description
     With the benzene process, maleic anhydride was produced by oxidizing
benzene in the presence of a catalyst.    A mixture of vaporized benzene and
air were sent to a fixed bed reactor,   The catalyst for the oxidation
reaction was vanadium or molybdenum oxide on inert carriers.  The stream was
then sent to a cooler, a condenser, and a separator, in which the maleic
anhydride was condensed and separated.  The maleic anhydride was then
dehydrated by azeotropic distillation using xylene.  The crude product is
then aged and sent to a fractionation column that yielded molten maleic
anhydride as the product.
     Sources of benzene emissions from the benzene oxidation process
included main process vents, refining vacuum vents, fugitive emissions, and
emissions from waste handling and disposal.

     Again, maleic anhydride is now produced in the United States
completely through a n-butane process.   This description of the benzene
process was provided for reference.
GASOLINE MARKETING

     Gasoline storage and distribution activities represent potential
sources of benzene emissions.  The benzene content of gasoline ranges from
less than 1 to almost 2 percent by liquid volume, but typical concentrations
                                 14
are around 1.2 percent by volume.    Therefore, total hydrocarbon emissions
resulting from storage tanks, material transfer, and vehicle fueling include
emissions of benzene.  This section describes sources of benzene emissions
                                      138
-------
from gasoline marketing operations.  Because the sources of these emissions
are so widespread, individual locations are not identified in this section.
Instead, emission factors are presented, along with a general/ discussion of
the sources of these emissions.  The discussion is taken from Reference 21.

     The flow of the gasoline marketing system in the United States is shown
in Figure 24.  From the refinery, gasoline may be transported by pipeline,
rail, ship, or barge to intermediate stations known as bulk terminals.
Terminals are bulk wholesale marketing outlets, of which there are about
                 22
1,500 nationwide.    From the bulk terminal, tank trucks deliver gasoline to
                                                 22
the national network of about 15,000 bulk plants.    Daily throughput at a
terminal averages about 950,000 liters (250,000 gallons) and is greater than
the average 19,000 liters (5,000 gallons) daily throughput of a bulk
      22
plant.    Both bulk terminals and
commercial, and retail customers.
      22
plant.    Both bulk terminals and bulk plants deliver gasoline to private,
     The transport of gasoline with marine vessels, distribution at bulk
plants, and distribution at service stations and associated benzene
emissions are discussed below.

Benzene Emissions from Loading Marine Vessels

                                                                *
     Volatile organic compounds (VOC) can be emitted as crude oil and
refinery products (gasoline, distillate oil, etc.) and are loaded and
transported by marine tankers and barges.  Loading losses are the primary
                                                              23
source of evaporative emissions from marine vessel operations.    These
emissions occur as vapors in "empty" cargo tanks are expelled into the
atmosphere as liquid is added to the cargo tank.  The vapors may be composed
of residual material left in the "empty" cargo tank and/or the material
being added to the tank.  Therefore, the exact composition of the vapors
emitted during the loading process are difficult to determine.

     Emission factors for volatile organic compounds from marine vessel
                                    23
loading were found in EPA documents.    Assuming a benzene/VOC ratio of
0.006 (Reference 21), emission factors for benzene from marine vessel
loading are given in Table 28.  Factors are available for crude oil,
                                      139
-------
                                 Refinery Storage
                Ship, Rail, Barge
                                  Bulk Terminals
Pipeline
\

Tank Trucks
i

p
Service Stations

I



\ '
Bulk Plants
\ <
Trucks

1 '
Commercial,
Rural Users
                                Automobiles, Trucks
Figure 24. The Gasoline Marketing Distribution System in the United States
                                                                               21
                                       140
-------
     TABLE 28.  UNCONTROLLED VOLATILE ORGANIC COMPOUND AND BENZENE EMISSIONS

                FROM LOADING GASOLINE IN MARINE VESSELS21'23
          Emission Source
     Volatile
 Organic Compound
 Emission Factor
(mg/1 Transferred)
      Benzene
  Emission Factor
(mg/1 Transferred)
Tanker Ballasting

Transit

Ship/Ocean Barge:

  Uncleaned; volatile previous cargo

  Ballasted; volatile previous cargo

  Cleaned; volatile previous cargo

  Gas-freed; volatile previous cargo

  Any condition; nonvolatile previous
  cargo

  Typical situation; any cargo

Barge:

  Uncleaned; volatile previous cargo

  Gas-freed; any cargo

  Typical situation; any cargo
        100

        320°



        315

        205

        180

         85

         85


        215



        465

        245

        410
        0.6

        1.9C



        1.9

        1.2

        1.1

        0.5

        0.5


        1.3



        2.8

        1.5

        2.5
^Factors are for nonmethane-nonethane VOC emissions.

DBased on benzene/VOC ratio of 0.006 (Reference 21).
•^
"Units for this factor are mg/week-liter transported.

 Ocean barge is a vessel with compartment depth of 40 feet; barge is a
 vessel with compartment depth of 10-12 feet.
                                        141
-------
                                23
distillate oil, and other fuels.    However,  reliable estimates of the
benzene content of these fuels were not found.   Therefore,  it was not
possible to provide benzene emission factors for marine vessel loading of
fuels other than gasoline.

Benzene Emissions from Bulk Gasoline Plants.  Bulk Gasoline  Terminals and
Service Stations

     Each operation in which gasoline is transferred or stored is a
potential source of benzene emissions.  At bulk terminals and bulk plants,
loading and unloading gasoline and storing gasoline are sources of benzene
emissions.  The gasoline that is stored in above ground tanks is pumped
through loading racks that measure the amount of product.  The loading racks
consist of pumps, meters, and piping to transfer the gasoline or other
liquid petroleum products.  Loading of gasoline into tank trucks can be
accomplished by one of three methods:  splash,  top submerged, or bottom
loading.  In splash loading, gasoline is introduced into the tank truck
                                                               21
directly through a compartment located on the top of the truck.    Top
submerged loading is done by attaching a downspout to the fill pipe so that
gasoline is added to the tank truck near the bottom of the tank.  Bottom
loading is the loading of product into the truck tank from the bottom.
Because emissions occur when the product being loaded displaces vapors in
the tank being filled, top submerged loading and bottom loading reduce the
                                                       21
amount of material (including benzene) that is emitted.

     Vapor balancing systems, consisting of a pipeline between the vapor
spaces of the truck and the storage tanks, are closed systems.  These
systems allow the transfer of vapor displaced by liquid in the storage tank
                                                                 21
into the transfer truck as gasoline is put into the storage  tank.
Table 29 lists emission factors for gasoline vapor and benzene from gasoline
loading racks at bulk terminals and bulk plants.  The gasoline vapor
emission factors were taken from AP-42 (Reference 23).  The  benzene factors
were obtained by multiplying the gasoline vapor factor by the average
                                             21
benzene content of the vapor (0.006 percent).
                                      142
-------
          TABLE 29.  BENZENE EMISSION FACTORS FOR GASOLINE LOADING

                     AND BULK TERMINALS AND BULK PLANTS21
Loading Method
 Gasoline Vapor
Emission Factor'
      mg/1
    Benzene
Emission Factor
      mg/1
Splash
      1430
      8.6
Submerged
     '  590
      3.5
Balance Service
       980
      5.9
 From AP-42 (Reference 23).   Gasoline factors represent emissions of
 nonmethane-nonethane VOC.  Factors are expressed as mg gasoline vapor per
 mg gasoline transferred.

 Based on a benzene/VOC ratio of 0.006 (Reference 21).
Q
 Submerged loading is either top or bottom submerged.
                                      143
-------
      A more recent study has provided another  estimate  of the  benzene
                            22
 content of gasoline vapors-.     An EPA test  program was  designed to
 characterize refueling emissions.   Results  of  the  study showed 0.0079 g
 benzene/g hydrocarbon in .the refueling emissions.  The  author  of the  EPA
 report summarizing the test results indicated  that an estimate of 0.0079 g
 benzene/g hydrocarbon was quite similar to  the 0.006  g/g hydrocarbon
 estimate developed earlier.   The difference in the two  estimates is likely
 attributed to differences in the benzene content of gasoline and the  fact
 that the original reference (Reference 21)  did not take dynamic effects such
                                         22
 as liquid/vapor turbulence into account.

      Storage emissions of benzene at bulk terminals and bulk plants depend
 on the type of storage tank used.   A typical bulk  terminal may have four or
 five above ground storage tanks with capacities ranging from
               3 21
 1,500-15,000 m .     Most tanks in gasoline  service have an external floating
 roof to prevent the loss of product through evaporation and working losses.
 Fixed-roof tanks, still used in some areas  to  store gasoline,  use
 pressure-vacuum vents to control breathing  losses. Some tanks may use vapor
 balancing or processing equipment to control working  losses.   A breather
 valve (pressure-vacuum valve), which is commonly installed on  many
 fixed-roof tanks, allows the tank to operate at a  slight internal pressure
 or vacuum.

      The major types of emissions from fixed-roof  tanks are breathing and
 working losses.  Breathing loss is the expulsion of vapor from a tank vapor
 space that has expanded or contracted because  of daily  changes in
 temperature and barometric pressure.  The emissions occur in  the absence of
-any liquid level change in the tank.  Combined filling  and emptying  losses
 are called "working losses."  Emptying losses  occur when the  air that is
 drawn into the tank during liquid removal saturates with hydrocarbon vapor
 and expands, thus exceeding the fixed capacity of  the vapor space and
 overflowing through the pressure vacuum valve.

      A typical external floating-roof tank consists  of  a cylindrical steel
 shell equipped with a deck or roof that floats on  the surface of the stored
 liquid, rising and falling with the liquid level.   The  liquid surface is
                                       144
-------
completely covered by the floating roof except in the small annular space
between the roof and the shell.  A seal attached to the roof touches the
tank wall (except for small gaps in some cases) and covers the remaining
area.  The seal slides against the tank wall as the roof is raised or
lowered.  The floating roof and the seal system serve to reduce the
evaporative loss of the stored liquid.

     An internal floating-roof tank has both a permanently affixed roof and
a roof that floats inside the tank on the liquid surface (contact roof) , or
supported on pontoons several inches above the liquid surface (noncontact
roof).  The internal floating-roof rises and falls with the liquid level,
and helps to restrict the evaporation of organic liquids.

     Losses from floating-roof tanks include standing-storage losses and
withdrawal losses.  Standing-storage losses, which result from causes other
than a change in the liquid level, constitute the major source of emissions
from external floating-roof tanks.  The largest potential source of these
losses is an improper fit between the seal and the tank shell (seal losses).
As a result, some liquid surface is exposed to the atmosphere.  Air flowing
over the tank creates pressure differentials around the floating roof.  Air
flows into the annular vapor space on the leeward side and an air-vapor
mixture flows out on the windward side.

     Withdrawal loss is another source of emissions from floating-roof
tanks.  When liquid is withdrawn from a tank, the floating roof is lowered
and a wet portion of the tank wall is exposed.  Withdrawal loss is the
vaporization of liquid from the wet tank wall.

     Table 30 shows emission factors for storage tanks at a typical bulk
terminal.  The emission factors were based on AP-42 factors and the weight
                                          21
fraction of benzene in the vapor of 0.006.

     The two basic types of gasoline loading into tank trucks at bulk plants
are the same as those used at terminals.  The first is the splash filling
method, which usually results in high levels of vapor generation and loss.
                                      145
-------
           TABLE 30.   BENZENE EMISSION FACTORS FOR STORAGE LOSSES

                      AT A TYPICAL GASOLINE BULK TERMINAL21
                                                      Benzene Emission Factor
     Storage Method                                         Mg/yr/Tank


Fixed Roof

  Working Loss                                                  0.2

  Breathing Loss                                                0.05


External Floating Roof

  Working Loss   '                                                d

  Storage Loss

  -  Primary Seal                                               0.05

  -  Secondary Metallic Shoe Seal                               0.02

3Terminal with 950,000 liters/day (250,000 gallons/day) with four storage
 tanks for gasoline.  See Appendix A for derivation.
v                                                      2
 Typical fixed-roof tank based upon capacity of 2,680 m  (16,750 bbls.).
c                                                         3
 Typical floating-roof tank based upon capacity of 5,760 m  (36,000 bbls.).
 Emission factor -  (0.46 x 10"  Q) Mg/yr, where Q is the throughput through
 the tanks in barrels (Reference 21).
                                       146
-------
The second method is submerged filling with either a submerged fill pipe or
bottom filling, which significantly reduces liquid turbulence and
vapor-liquid contact resulting in much lower emissions.  Table 31 shows the
uncontrolled emission factors for benzene from a typical bulk plant.

     Gasoline tank trucks have been demonstrated to be major sources of
vapor leakage.  Some vapors may leak uncontrolled to the atmosphere from
dome cover assemblies, pressure-vacuum (P-V) vents, and vapor collection
piping and vents.  Other sources of vapor leakage on tank trucks that occur
less frequently include tank shell flaws, liquid and vapor transfer hoses,
improperly installed or loosened overfill protection sensors, and vapor
couplers.  Since terminal controls are usually found in areas where trucks
are required to collect vapors after delivery of product to bulk plants or
service stations (balance service), the gasoline vapor emission factor
associated with uncontrolled truck leakage was assumed to be 30 percent of
the balance service truck loading factor (960 mg/liter x 0.30 -
              21
288 mg/liter).    Thus the emission factor for benzene emissions from
uncontrolled truck leakage is 1.7 mg/liter, based on a benzene/vapor ratio
of 0.006.

     The discussion on service station operations is divided into two areas:
the filling of the underground storage tank and automobile refueling.
Although terminals and bulk plants also have two distinct operations (tank
filling and truck loading), the filling of the underground tank at the
service station ends the wholesale gasoline marketing chain.  The automobile
refueling operations interact directly with the public and control of these
operations can be performed by putting control equipment on either the
service station or the automobile.

Benzene Emissions from Service Stations--

     Normally, gasoline is .delivered to service stations in large tank
trucks from bulk terminals or smaller account trucks from bulk plants.
Emissions are generated when hydrocarbon vapors in the underground storage
                                      147
-------
             TABLE 31.  UNCONTROLLED GASOLINE VAPOR AND BENZENE
                                                           21
                        EMISSIONS FROM A TYPICAL BULK PLANT
                                   Gasoline Vapor                Benzene
                                   Emission Factor   .        Emission Factor
    Emission Source                   mg/liter                   mg/liter
Storage Tanks - Fixed Roof

  Breathing Loss                         600                       3.6

  Filling Loss                          1150                       6.9

  Draining Loss                          460                       2.7


Gasoline Loading Racks

  Splash Loading                        1430                       8.6

  Submerged Loading                      590                       3.5

  Submerged Loading                      980                       5.9
  (Balance Service)

aTypical bulk plant with a gasoline throughput of 19,000 liters/day
 (5,000 gallons/day).
 Based on gasoline emission factor and benzene/vapor ratio of 0.006
 (Reference 21).
                                      148
-------
tank are displaced to the atmosphere by the gasoline being loaded into the
tank.  As with other loading losses, the quantity of the service station
tank loading loss depends on several variables, including the quantity of
liquid transferred, size and length of the fill pipe, the method of filling,
the tank configuration and gasoline temperature, vapor pressure, and
composition.  A second source of emissions from service station tankage is
underground tank breathing.  Breathing losses occur daily and are attributed
to temperature changes, barometric pressure changes, and gasoline
evaporation.

     In addition to service station tank loading losses, vehicle refueling
operations are considered to be a major source of emissions.   Vehicle
refueling emissions are attributable to vapor displaced from the automobile
tank by dispensed gasoline and to spillage.  The major factors affecting the
quantity of emissions are gasoline temperature, auto tank temperature,
gasoline Reid vapor pressure (RVP), and dispensing rates.  Table 32 lists
the uncontrolled emissions from a typical gasoline service station.  The
emission factors presented in Table 32 are from EPA's AP-42 document
(Reference 23).

Control Technology for Gasoline Transfer

     At bulk terminals and bulk plants, benzene emissions from gasoline
transfer may be controlled by a vapor processing system in conjunction, with
                          21
a vapor collection system.    Figure 25 shows a Stage I control vapor
balance system at a bulk plant.  These systems collect and recover gasoline
vapors from empty, returning tank trucks as they are filled with gasoline
                   23
from storage tanks.

     At service stations, vapor balance systems contain the gasoline vapors
within the station's underground storage tanks for transfer to empty
gasoline tank trucks returning to the bulk terminal or bulk plant.
Figure 26 shows a diagram of a service station vapor balance system.
                                      149
-------
             TABLE 32.  UNCONTROLLED GASOLINE VAPOR AND BENZENE

                        EMISSIONS FROM A TYPICAL SERVICE STATION21'24
                                   Gasoline Vapor              Benzene    ,
                                  Emission Factors        Emission Factors
   Emission Source                    mg/liter                 mg/liter
Underground Storage Tanks

  Tank Filling Losses

  -  Submerged Fill                      880                     5.3

  -  Splash Fill                        1380                     8.3

  -  Balanced Submerged                   40                     0.2
     Filling

  Breathing Losses                       120                     0.7


Automobile Refueling

  Displacement Losses

  -  Uncontrolled                       1320                     7.9

  -  Controlled                          132                     0.8

  Spillage                                84                     0.5

STypical service station has a gasoline throughput of 190,000 liters/month
 (50,000 gallons/month).
 Based on gasoline vapor emission factor and benzene/vapor ratio of 0.006
 (Reference 21).
                                      150
-------
                                         a>
                                         ••-•
                                         (0
                                         >s
                                         CO
                                         
                                        \L
151
-------
                                                 H2/00/8C
  Si
  2
  •*•*
  c
  o
  O
  S"
 w
 s
  CO
  
  c
  o
 *-•
  (0
 4^
 CO
  
-------
Control Technology for Gasoline Storage

     The control technologies for controlling benzene emissions from
gasoline storage involve upgrading the type of storage tank used or addition
of a vapor control system.  For fixed-roof tanks, emissions are most readily
                                                      21
controlled by installation of internal floating roofs.    An internal
floating roof reduces the area of exposed liquid surface on the tank and,
therefore, decreases evaporative loss.  Installing an internal floating roof
                                                                        21
in a fixed-roof tank can reduce total emissions by 68.5 to 97.8 percent.
     For external floating-roof tanks, no control measures have been
                                                           21
identified for controlling withdrawal losses and emissions. -  These
                                                                           V
emissions are functions of the turnover rate of the tank and the
characteristics of the tank shell.  Rim seal losses in external floating
roof tanks depend on the type of seal.  Liquid-mounted seals are more
                                                               21
effective than vapor-mounted seals in reducing rim seal losses.    Metallic
shoe seals are more effective than vapor-mounted seals but less effective
                          21
than liquid-mounted seals.
Control Technology for Vehicle Refueling Emissions

     Vehicle refueling emissions are attributable to vapor displaced from
the automobile tank by dispensed gasoline and to spillage.  The quantity of
displaced vapors is dependent on gasoline temperature, vehicle tank size and
                                                            21
temperature, fuel level, gasoline RVP, and dispensing rates.

     The two basic refueling vapor control alternatives are:  control
systems on service station equipment (Stage II controls),  and control
systems on vehicles (onboard controls).   Onboard controls are basically
limited to the carbon canister.

     There are currently three types of Stage II systems in limited use in
the United States:  the vapor balance, the hybrid, and the vacuum assist
systems.  In the vapor balance system, gasoline vapor in the automobile fuel
                                      153
-------
tank is displaced by the incoming liquid gasoline and is prevented from
escaping to the atmosphere at the fillneck/nozzle interface by a flexible
rubber "boot."  This boot is fitted over the standard nozzle and is attached
to a hose similar to the liquid hose.  The hose is connected to piping which
vents to the underground tank.  An exchange is made (vapor for liquid) as
the liquid displaces vapor to the underground storage tank.  The underground
storage tank assists this transaction by drawing in a volume of vapor equal
                                21
to the volume of liquid removed.

     The vacuum assist system differs from the balance system in that a
"blower" (a vacuum pump) is used to provide an extra pull at the
nozzle/fillneck interface.  Assist systems can recover vapors effectively
without a tight seal at the nozzle/fillpipe interface because only a close
fit is necessary.  A slight vacuum is maintained at. the nozzle/fillneck
interface allowing air to be drawn into the system and not allowing the
vapors to escape.  Because of this assist, the interface "boot" need not be
as tight fitting as with balance systems.  Further, the vast majority of
assist nozzles do not require interlock mechanisms.  Assist systems
generally have vapor passage valves; located in the vapor passage somewhere
other than in the nozzles, resulting in a nozzle which is less bulky and
                                                          21
cumbersome than nozzles employed by vapor balance systems.

     The hybrid system borrows from the concepts of both the balance and
vacuum assist systems.  It is designed to enhance vapor recovery at the
nozzle/fillneck interface by vacuum, while keeping the vacuum low enough so
that a minimum level of excess vapor/air is returned to the underground
storage tank.

     With the hybrid system, a small amount of the liquid gasoline (less
than 10 percent) pumped from the storage tank is routed (before metering)  to
a restricting nozzle called an aspirator.  As the gasoline goes through this
restricting nozzle, a small vacuum is generated.  This vacuum is used to
draw vapors into the rubber boot at the  interface.  Because the vacuum is  so
                                      154
-------
small, very little excess air, if any, is drawn into the boot, hose and
underground storage tank, and thus there is no need for a secondary
                                                   21
processor, such as the vacuum assist's incinerator.
     Onboard vapor control systems consist of carbon canisters installed on
the vehicle to control refueling emissions.  The carbon canister system
adsorbs, on activated carbon, the vapors which are displaced from the
vehicle fuel tank by the incoming gasoline.  Such a system first absorbs the
emissions released during refueling and subsequently purges these vapors
from the carbon to the engine carburetor when it is operating.  This system
is essentially an expansion of the present evaporative emissions control
system used in all new cars to minimize breathing losses from the fuel tank
and to control carburetor evaporative emissions.  However, unlike the
present system, a refueling vapor recovery system will require a tight seal
at the nozzle/fillneck interface during refueling operations to ensure
                                                                        21
vapors flow into the carbon canister and are not lost to the atmosphere.

BENZENE EMISSIONS FROM MOTOR VEHICLES

     Hydrocarbons may be emitted from gasoline-powered vehicles from the
carburetor (evaporation),  fuel tank (vents),  crankcase (blow-by past piston
                           25
rings), and engine exhaust.    The level of benzene emissions from these
sources depends on the gasoline blend.  If the fuel contains ethylbenzene,
for example, incomplete combustion may lead to conversion of ethylbenzene to
        25
benzene.    Furthermore, there is some evidence that toluene dealkylation
                                                            p £
occurs, producing small amounts of benzene that are emitted.    In an
experiment using a fuel mixture containing hydrocarbons and toluene but no
                                                                     o (•
benzene, benzene was found in the exhaust of single cylinder engines.
According to another author, more than 50 percent of the benzene emitted
from motor vehicles is formed during the combustion of gasoline and is not
                                     27
from the benzene present in the fuel.     Therefore, more benzene may be
found in vehicle exhaust than was originally present in the fuel.
                                      155
-------
     Emissions of benzene fron vehicle exhaust vary by vehicle type, fuel
type, and control technology in use.  Table 33 lists the fractions of
vehicle exhaust and evaporative emissions for several different vehicle
types.  The data in Table 33 were derived from tests on low mileage, well
maintained vehicles.  Then, total hydrocarbons emission factors from the
computer program MOBIL3 were speciated for benzene, based on the percentages
in Table 33.  The resulting 1985 emission factors for benzene from
composites of Federal Test Procedures ranged from 0.128 to 0.135 g/mile.
The ranges result from the ranges of benzene content in the evaporative
emissions (see Table 33).

     Control techniques are available and in use for both evaporative and
exhaust emissions of benzene.  For example, positive crankcase ventilation
                                                                       14
(PCV) and evaporative controls reduce evaporative emissions of benzene.
PCV systems circulate air through the crankcase to pick up blow-by gases and
                                                                         14
take them to the intake manifold where they enter the combustion chamber.
                                                              14
Fuel evaporative controls were installed on all 1971 vehicles.    An
absorption/regeneration system, one of the most common evaporative control
techniques, virtually eliminates evaporative hydrocarbon emissions including
        14
benzene.    With such a system, a canister of activated carbon traps vapors.
The vapofs are ultimately fed back to the combustion chamber.

     A recent EPA study considered the difference in benzene emissions in
vehicle exhaust for vehicles tested under different roadway conditions
(urban expressway and stop/go city driving) and the Federal Test Procedure
and with different control technologies.  The results of the tests are shown
in Appendix A, Table A-2.  No significant trends were noted when comparing
benzene emissions as percentages of the total hydrocarbon emissions for
                               27
different control technologies.    For vehicles under the crowded urban
expressway test, benzene emissions ranged from 31.1 mg/mile (no
catalyst/air) to 9.4 mg/mile for a three-way catalyst/air (Table A-2).  For
the simulated "New York City driving" and a vehicle operated with no
                                                     27
catalyst/air, the benzene emissions were 150 mg/mile.    For the same test
situation and a vehicle operating with an oxidation catalyst/no air, the
                                                        27
reported benzene emissions were 170 mg/mile (Table A-2).
                                      156
-------
     TABLE 33.  BENZENE EMISSIONS EXPRESSED AS PERCENTAGE OF EXHAUST AND
                                                                   28
                EVAPORATIVE EMISSIONS FOR DIFFERENT VEHICLE CLASSES
        Vehicle Class
Benzene Content
   of Exhaust
Hydrocarbons (%)
 Benzene Content
 of Evaporative
Hydrocarbons (%)
Light Duty Gasoline Vehicle

  3-way Catalyst

  3-way and Oxidation Catalyst

  Noncatalyst or Oxidation Catalyst

Light Duty Gasoline Truck

Light Duty Diesel Vehicle

Light Duty Diesel Truck

Heavy Duty Gasoline Vehicle

Heavy Duty Diesel Vehicle
      5.12

      2.78

      3.95

      3.24

      2.40

      2.40

      3.48

      1.10
                      0.35 - 1.53'
       1.1
       1.1
 For fuel-injected vehicles, the range is 0.35 - 0.46 percent.   For
 carbureted light duty gasoline vehicles, the range is 1.11 - 1.53 percent.
                                      157
-------
BENZENE USE AS A SOLVENT

     More than 95 percent of all benzene used as a raw material is used in
producing the chemicals d'       .in this section.  However,  benzene has
also been used in a wi*.     ..scy of operations as a solvent.   Table 34 lists
industries and products in which benzene has reportedly been used as a
solvent.  However, for a number of reasons including cost and potential
adverse health effects, the use of benzene as a solvent has been decreasing.
In 1971, about 8 percent of benzene produced was used in solvent
             29
applications.    In 1982, benzene use as a solvent accounted for only
                                       29
1 percent of the total amount produced.     Of the source categories or
products listed in Table 34, ten have reportedly discontinued use of benzene
as a solvent.

     Telephone contacts with industry sources and trade associations in some
of the categories listed in Table 34 shows that benzene use as a solvent was.
declining or had ceased in those industries.  For example, as of 1983, a
trade association contact reported that benzene was generally no longer used
in textile manufacture.  However, benzene has been found in wastewaters from
                                                     29
textile finishing and non-woven manufacturing plants.    Similar reports of
decreased benzene use as a solvent were received from other industry
contacts, including representatives, of the degreasing source category,
                                                           29 30
Pharmaceuticals manufacture, and general organic synthesis.  '

     However, benzene is used as a denaturant for ethyl alcohol and as an
azeotropic agent for dehydration of 95 percent ethanol and 91 percent
isopropanol.    Companies producing, these alcohols are shown in Table 35.
No specific data were found concerr.ing emissions of benzene from these
facilities.
     Benzene is also used in aluminum alkyls production.  Producers of
aluminum alkyls include Ethyl Corporation in St. Louis, Missouri and Texas
                           29
Alkyls in Deer Park, Texas.    Both of these facilities were using benzene
           29
as of 1984.    However, no specific emission factors were found for these
                                      158
-------
             TABLE 34.  INDUSTRIES AND PRODUCTS POSSIBLY USING

                        BENZENE AS A SOLVENT13'14'29
               Rubber tires

                                            a
               Miscellaneous rubber products

                                    a
               Adhesives manufacture


               Gravure printing inks

                                      £*
               Printing and publishing


               Trade and industrial paints


               Paint removers


               Synthetic rubber

                              Q
               Floor coverings


               Laboratories


               Degreasing of metal furniture, primary metals


               Pharmaceuticals manufacture


               Alcohols production


               Miscellaneous small volume chemicals


               General organic synthesis


               Textiles


A previous report, cited in Reference 29, indicated that benzene use in
these categories has been discontinued.
                                     159
-------
         TABLE 35.  UNITED STATES PRODUCERS OF ETHANOL OR ISOPROPANOL
                                                                     11
         Facility*1
  Location
Annual Capacity
 (106 Gallons)
ETHANOL
American Development Corporation

American Fuel Technologies, Inc.
Archer Daniels Midland Company
  ADM Processing Division

Dawn Enterprises
Eastman Kodak Company
  Eastman Chemical Products, Inc.
  subsidiary
    Texas Eastman Company
Energy Fuels Development
Corporation
Georgia-Pacific Corporation
  Chemical Division
Grain Processing Corporation
High Plains Corporation
Kentucky Agricultural Energy
Corporation
Midwest Grain Products, Inc.

National Distillers and Chemical
Corporation
  Chemicals Division
    U.S. Industrial Chemicals
    Company, division
New Energy Company of Indiana
Pekin Energy Company
Shepherd Oil, Inc.
South Point Ethanol
Hamburg,  IA
Hastings, NE
Federalsburg, MD
Cedar Rapids, IA
Decatur, IL
Peoria, IL
Valhalla, ND
Longview, TX
Portales, NM

Bellingham, WA
Muscatine, IA
Colwich, KS
Franklin, KY

Atchison, KS
Pekin, IL
Tuscola, IL

South Bend, IN
Pekin,  IL
Jennings, LA
South Point, OH
       15

       45

      500

       10
       25
       10

        6
       60
       10
       21

       22
       19
       66

       52
       60
       35
       60
                                        160
-------
   TABLE 35.  UNITED STATES PRODUCERS OF ETHANOL OR ISOPROPANOL   (Continued)
                                                                Annual Capacity

         Facility3                        Location               (10  Gallons)


A. E. Staley Manufacturing Company
  Special Products Group
    Ethanol Division                    Loudon, TN                      40

Tennoll Energy, Inc.                    Jasper, TN                      26

Union Carbide                           Texas City, TX                 120

Universal Foods Corporation             Juneau, WI                      10

TOTAL ETHANOL                                                         1214

ISOPROPANOL

Atlantic Richfield Company              Channelview, TX                 50

Exxon Corporation                       Baton Rouge, LA                835

Shell Oil Company                       Deer Park, TX                  600
                                        Wood River, IL                 340

Union Carbide                           Texas City, TX                 700

TOTAL ISOPROPANOL                                                     2525


 The following companies have the facilities to produce ethyl alcohol, primarily
 for fuel use, in quantities of 10 million gal/year or less (except as noted):
 Alcohol Energy Corporation - Staley, NC; Baca Food and Fuel Cooperative -
 Campo, CO; Bornhoft, Paul - Merino, CO; A. Smith Bowman Distillery Corporation -
 Reston, VA; Channel Energy Company - Muleshoe, TX; Coburn Enterprises, Inc. -
 Sherman, SD; Colorado Gasohol, Inc. - Walsh, CO; Crystal Fuel, Inc. - Bonaparte,
 IA; Ecological Energy, Inc. - Roca, NE; Food and Energy, Inc. - Litchfield, MA;
 Lenox Grain Fuels, Inc. -' Lenox, IA; Marlin Car Care - Marlin, TX; A. E. Montana,
 Inc. - Amsterdam, MT; Raven Alcohol Distillery - Selma, CA; Southern Distilleries
 Company - Ashford, AL; Spudcohol - Pingree, ID; Syncorp, Inc. - Roberta, GA; U.S.
 Gasohol Corporation - Lockeford, CA; White Flame Fuels, Inc. - Van Buren, AR

Note:  This listing is subject to change as market conditions change, facility
       ownership changes, plants are closed, etc.  The reader should verify the
       existence of particular facilities and processes used by consulting current
       listings and/or the plants themselves.  The level of benzene emissions from
       any given facility is a function of variables such as capacity, throughput,
       and control measures, and should be determined through direct contacts with
       plant personnel.  These data on producers and locations are as of
       January 1987.
                                       161
-------
sources.  A report from Texas Alkyls indicated that -of the 490,000 pounds of
benzene purchased, about 100 pounds would be lost to the atmosphere as
                                                                         29
fugitive emissions from tank truck off-loading and laboratory activities.

BENZENE EMISSIONS FOR TREATMENT, STORAGE, AND DISPOSAL FACILITIES (TSDF)

     Benzene may be emitted from treatment storage and disposal facilities
(TSDF) and domestic refuse sites, depending on the types of waste handled or
disposed.  For example, in one recent study of landfill gas from municipal
and industrial disposal sites in England, benzene was detected in all eight
                   31
sites investigated.    This study sampled the landfill gas in situ and
ambient measurements were not taken.  However,  it seems likely that benzene
                                                                     *
may be emitted from landfills.

     No emission factor for benzene may be estimated for landfills or TSDF
because the quantity of emissions depends on waste type and disposal
techniques.  The reader is urged to investigate specific sites to determine
the potential for benzene emissions from these sources.  Reference 32
documents some analytical results: showing benzene concentrations of
municipal landfill gas.

BENZENE EMISSIONS FROM STATIONARY COMBUSTION SOURCES

     Benzene may be emitted from stationary combustion sources such as
external combustion boilers, hazardous waste incinerators, hospital waste
                                                 T'O T*3 T / *3^
incinerators, and ferrous metallurgical furnaces.   '   '  '    Few data are
available quantifying these benzene emissions.   The data located during this
research effort are discussed below.

     Small amounts of hydrocarbons, including benzene, are produced and
emitted during coal, oil, and gas combustion.  More hydrocarbons are
produced and emitted if a combustion unit is not properly operated and
maintained.  No specific emission  factors were discovered for benzene
emissions from external combustion boilers.  However,  one reference reported
                                      162
-------
that the exhaust stream from a natural gas-fired boiler contained
                                                        32
1.18 percent by volume (or 4 percent by weight) benzene.    The same
reference reported benzene content of 0.42 percent by volume (1.90 percent
                                                           32
by weight) for emission stream from a coke oven gas boiler.    (See also
Appendix B.)
     Benzene has also been identified in air emissions from hazardous waste
             34
incinerators.    In a series of comprehensive tests of hazardous waste
incinerators, benzene was frequently observed in the emission/exhaust
streams (along with toluene, chloroform, naphthalene, and
tetrachloroethylene).   Several reasons were suggested to explain the
prevalence of -benzene (and the other compounds) in the emissions.  First,
                             v
the destruction efficiency for the compounds could be poor, given the low  •
concentrations of the compounds in the waste feed.  Second, the compounds
could be input from sources other than the waste feed.  Finally, the
                                                                34
compounds could be actual products of some combustion reactions.    None of
the likely reasons was especially favored over the others in the referenced
report.  Also, no actual emission factors were given in the report.  The
                                                                       34
reported benzene content of the stack effluent ranged from 12-670 ng/1.

     Few data are available describing the emissions of benzene (or other
compounds) from hospital refuse incinerators.  One test report, conducted by
the California Air Resources Board, showed mass emission rates of
0.1-0.4 Ib/hour benzene.    The dual chamber incinerator had an average  feed
rate of 783 Ib refuse/hour during the test.  Feed consisted of 10 percent
moisture, 30 percent plastic, 65 percent paper, and 5 percent other
material.
     Metallurgical furnaces are another type of stationary combustion source
                                      •3 f\ O £
that are sources of benzene emissions.  '     For example,  organic compounds
have been identified as potential in emissions from iron and steel
manufacturing sources.  Possible sources of the organics include fuel
combustion, the presence of organics on scrap metals,  and use of organic
binders for pellets and agglomerates.  At foundries, organics may be emitted
                                      163
-------
because they are present in core binders and additives,  coating oils, lubes,
           O £                                                        -30
and paints.    The organic emissions are reported to contain benzene.
While no emission factors specifically for benzene emissions from
metallurgical operations were found, one reference contained emission
factors for organics from the iron and steel industry,  iron foundry
industry, and ferroalloy industry.  These emission factors are summarized in
Table 36.  Some speciation data are available to help determine what
fraction of the organic emissions may be benzene.  For example, one
reference (see Appendix B) indicates that for a gray iron foundry, exhaust
gases from 12 binder systems and sand formulations during pouring/casting
                                                    33
operations contained 34.7 percent benzene by weight.    No other more
specific emission factors for benzene were found for metallurgical furnaces.
     Sewage sludge incineration is another potential source of benzene
emissions.  In a series of tests using the volatile organic sampling train
(VOST),  benzene was detected in stack gases at concentrations of 5 to
31 ug/dscm.    No specific emission factors were found for benzene emissions
from sewage sludge incineration.
                                      164
-------
       TABLE 36.  SUMMARY OF ORGANIC EMISSION FACTORS FOR THE IRON AND

                  STEEL, IRON FOUNDRY, AND FERROALLOY INDUSTRIES36
                                                               Normalized
                                                                 Organic
                                                             Emission Factor
                                                                (lb/ton)a
Iron and Steel Industry

By-product Cokemaking
Blast Furnace Ironmaking
Sinter Production
Basic Oxygen Process Steelmaking
Electric Arc Furnace Steelmaking
Open Hearth Furnace Steelmaking
Hot Forming and Finishing Operations

Iron Foundry Industry

Cupola Furnace Melting
Electric Arc Furnace Melting
Inoculation
Metal Pouring and Cooling
Casting Shakeout

Ferroalloy Industry

Open Submerged Arc Furnace Smelting
  Silicon Metal Alloys
  Ferrosilicon Alloys
  Ferromanganese Alloys
  Ferrochromium Alloys

Covered Submerged Arc Furnace Smelting
  Ferrosilicon Alloys
  Ferromanganese Alloys
  Ferrochromium Alloys
 3.90
 1.17
 3.20
 0.0018
 0.0995
 0.0188
 0.184
 0.195
 0.160
 0.0013
 0.209
 1.89
10.2
 1.86
 0.340
 0.645
 0.225
 0.0479
 0.0614
 Emission factors normalized to Ib/ton finished product (i.e., total steel
 products, net castings, total ferroalloys).
                                      165
-------
REFERENCES FOR SECTION 5
 1.  U.S. Environmental Protection Agency.   Organic Chemical Manufacturing.
     Volume 6:  Selected Processes.  Research Triangle Park,  NC.   Office of
     Air Quality Planning and Standards.   EPA-450/3-80-028a.   1980.

 2.  U. S. Environmental Protection Agency.   Atmospheric Benzene Emissions.
     Research Triangle Park, NC.   Office  of Air Quality Planning and
     Standards.  EPA-450/3-77-029.  1977.

 3.  Chemical Marketing Reporter.   Chemical Profile:  Ethylbenzene.
     July 28, 1986.  50.

 4.  Austin, G. T.   Industrially Significant Organic Chemicals.   Chemical
     Engineering.   81(9):145.  1974.

 5.  Scott, E. Y.  D.   Inventory.   Mobil Oil Corporation, Assignee.  High
     Temperature Method for Producing Styrene and Ethylbenzene.   U.S.  Patent
     No. 3,396,206.  August 6,  1968.

 6.  SRI International.  1986 Directory of Chemical Producers.  Menlo  Park,
     CA.  1986.

 7.  U. S. Environmental Protection Agency.   Benzene Emissions from
     Ethylbenzene/Styrene Industry - Background Information for Proposed
     Standards and Draft Environmental Impact Statement.  Research Triangle
     Park, NC.  Office of Air Quality Planning and Standards.
     EPA-450/3-79-035a.  1980.

 8.  Short, H. C and L. Bolton.  New Styrene Process Pares Production  Costs.
     Chemical Engineering.  92(17):30-31.   1985.

 9.  Purcell, W. P.  Benzene (In).  Kirk  Othmer Encyclopedia of Chemical
     Technology.  Volume 3.  John Wiley and Sons.  NY.  1978.

10.  U. S. Environmental Protection Agency.   Organic Chemical Manufacturing.
     Volume 6:  Selected Processes!.  Report 1.  EPA-450/3-80-028a.  Research
     Triangle Park, NC.  1980..

11.  SRI International.  1987 Directory of Chemical Producers.  Menlo  Park,
     CA.  1987.

12.  Chemical Marketing Reports.   Chemical Profile.  Cyclohexane.
     October 20, 1986.

13.  U. S. Environmental Protection Agency.   Organic Chemical Manufacture.
     Volume 7:  Selected Processes.  EPA-450/3-80-028b.  Research Triangle
     Park, NC.  1980.  Reports 1,  2, 3, 6, and 7.
                                      166
-------
14.  U. S. Environmental Protection Agency.  Atmospheric Benzene Emissions.
     EPA-450/3-77-029.  Research Triangle Park, NC.  1977.   pp. 4-19 - 4-25.

15.  U. S. Environmental Protection Agency.  Assessment of Human Exposures
     to Atmospheric Benzene.  EPA-450/3-78-031.  Research Triangle Park, NC.
     1978.

16.  Dunlap, K. L.  Nitrobenzene and Nitrotoluenes.  (In) Kirk Othmer
     Concise Encyclopedia of Chemical Technology.  John Wiley and Sons.  NY.
     1981.  pp. 790-791.

17.  Chemical Products Synopsis:  Nitrobenzene.  Mannsville Chemical
     Products Corporation.  Cortland, NY.  1984.

18.  U. S. Environmental Protection Agency.  Locating and Estimating Air
     Emissions from Sources of Chlorobenzenes.   EPA-450/4-84-007m.  Research
     Triangle Park, NC.  1986.

19.  U. S. Occupational Safety and Health Administration.  Technology
     Assessment and Economic Impact Study of an OSHA Regulation for Benzene.
     Volume II.  OSHA-EIS-77-500-II.  1977.

20.  U. S. Environmental Protection Agency.  Source Assessment:  Chlorinated
     Hydrocarbons Manufacture.  EPA-600/2-79-019g.  Research Triangle Park,
     NC.  1979.

21.  U. S. Environmental Protection Agency.  Evaluation of Air Pollution
     Regulatory Strategies for Gasoline Marketing Industry.
     EPA-450/3-84-012a.  Washington, D.C.  1984.

22.  Laing, P. M.  Factors Influencing Benzene  Emissions from Passenger Car
     Refueling.  SAE Technical Paper Series. Paper No. 861559.  Presented
     at the International Fuels and Lubricants  Meeting and Exposition.
     Philadelphia, PA.  October 6-9, 1986.

23. ' U. S. Environmental Protection Agency.  Estimation of the Public Health
     Risk from Exposure to Gasoline Vapor Via the Gasoline Marketing System.
     A Staff Paper Submitted to the Science Advisory Board.  Office of
     Health and Environmental Assessment.  June 1984.

24.  U. S. Environmental Protection Agency.  Compilation of Air Pollutant
     Emission Factors.  Fourth Edition (Including Supplements 1-15 and
     Updates).  Publication No. AP-42.  Research Triangle Park, NC.  1985.

25.  U. S. Environmental Protection Agency.  Materials Balance for Benzene
     Level II.  EPA-560/13-80-009.  Washington, D.C.  1980.  pp. 2-6 - 2-34.

26.  Bradow, R. L. and F. Black.  Benzene, Toluene and Xylene Concentrations
     In Car Exhausts and In City Air.  Atmos. Env. 18(2):479-480.   1984.
                                      167
-------
27.  Klausmeier, R.,  P. Anderson, and K.  Wert.   Evaluation of Mobile Source
     Control Strategies for Benzene.   Prepared for the Western Oil and Gas
     Association.  Radian DON No. 85-245-024-01.   Austin,  TX.   1985.

28.  Carey, P. M.  Technical Report.   Air Toxics Emissions from Motor
     Vehicles.  EPA-AA-TSS-PA-86-5.   Emission Control Technology Division.
     Office of Mobile Sources.   Ann Arbor,  MI,   pp.  43-45.  September 1987.

29.  Forrest, A. S. and G. E. Wilkins.  Benzene:   Solvent Usage and Waste
     Disposal.  Technical Memorandum.   Prepared for SASD,  OAQPS, Research
     Triangle Park, NC.  EPA Contract No.  68-02-3513.   July 21, 1983.

30.  Ahmed, S.  Benzene:  Technical Assessment of Benzene Solvent Usage.
     Technical Memorandum.  Prepared  for PAB,  OAQPS,  Research Triangle Park,
     NC.  EPA Contract No. 68-02-3818.  March 30,  1984.

31.  Young, P. and A. Parker.  Vapors, Odors and Toxic Gases from Landfills.
     Hazardous and Industrial Waste Management and Testing:   Third
     Symposium.  ASTM STP 851.   L.  P.  Jackson,  A.  R.  Rohlik, and
     R. A. Conway, editors.  American Society for Testing and Materials.
     Philadelphia, PA.  1984.  pp.  24-41.

32.  Wood, J. A. and M. C. Porter.  Hazardous Pollutants in Class II
     Landfills.  A Research Report.   Laboratory Services Branch.  Technical
     Services Division.  South Coast  Air Quality Management District.
     El Monte, CA.  December 1985.

33.  U. S. Environmental Protection Agency.  Air Emissions Species Manual.
     Volume I.  Volatile Organic Compounds  (VOC)  Species Profiles.  Draft
     Report.  Prepared for the Noncriteria Pollutants Program Branch.  U. S.
     Environmental Protection Agency.   Research Triangle Park, NC.  1987.

34.  U. S. Environmental Protection Agency.  Performance Evaluation of Full
     Scale Hazardous Waste Incinerators.   EPA-600/2-84-181e.  Cincinnati,
     OH.  1984.

35.  California Air Resources Board.   Evaluation Test on a Hospital Refuse
     Incinerator at St. Agnes Medical Center.   Fresno, CA.  ARB Stationary
     Sources Division.  Sacramento, CA.  January 1987.

36.  U. S. Environmental Protection Agency.  Organic Emissions from Ferrous
     Metallurgical Industries:  Compilation of Emission Factors and Control
     Technologies.  EPA-600/2-84-003.   Research Triangle Park, NC.  1984.

37.  Telecon.  P. A.  Cruse, Radian with Gary T. Hunt, ERT, Concord, MA.
     January 26, 1988.
                                      168
-------
                                  SECTION 6
                       SAMPLING AND ANALYTICAL METHODS

     The procedures presented in this section are methods that have been
published by the U. S. Environmental Protection Agency (EPA) as viable
methods for the sampling and analysis of volatile organic compounds (VOCs)
from fugitive and industrial process sources.  The analytical schemes have
been modified to allow the detection of benzene at low levels (1 part per
million, ppm, or less) with good precision and reproducibility.   The
sampling and analytical procedures for each recommended method are
summarized together in this section.  These methods are EPA Reference
Method 21, EPA Reference Method 18, and the "Protocol for the Collection and
Analysis of Volatile Principal Organic Hazardous Constituents (POHCs) Using
the Volatile Organic Sampling Train (VOST)."

     EPA Reference Method 21 - Determination of Volatile Organic Compound
Leaks  is the method to use to detect the presence of fugitive emissions of
benzene and other volatile organic compounds.  This method applies to leaks
from process equipment such as valves, flanges, connections, pumps,
compressors, pressure relief valves, pumps and compressor seal system
degassing vents, accumulator vessel vents, agitator seals, and access door
seals.  Method 21 cannot be used to quantitate volatile organic compound
leaks from these sources.

     EPA Reference Method 18 - Measurement of Gaseous Organic Compound
Emissions by Gas Chromatography describes the sampling and analytical
procedures that are recommended for quantifying benzene in gaseous process
        2
streams.   The procedures published in this method have been used to permit
quantification of benzene from fugitive sources.   They can also be used to
sample and analyze benzene in ambient air.
                                      169
-------
     The detection limit for the procedures in Method 18 is determined by
the detector and column utilized for analysis but is about 1 ppm.   The upper
limit may be extended past the detector and column limits by sample dilution
or by using a smaller gas sampling loop.

     In general, direct interface sampling, dilution interface sampling,
integrated sampling using evacuated containers with Tedlar or aluminized
Mylar bags, or sampling onto adsorption tubes filled with activated charcoal
are the methods that are recommended in Method 18 for sampling benzene
emissions.  The components of the sample are analyzed with a gas
chromatograph and measured using a suitable detector.   The eluted compounds
are identified and quantified by comparing elution times and peak heights
with those of known standards.
     An alternative to monitoring for benzene emissions by Method 18 is the
"Protocol for the Collection and Analysis of Volatile POHCs Using VOST."4
This method was recently developed by EPA's Office of Research and
Development to determine the destruction and removal efficiency of volatile
POHCs from the stack gas effluents; of hazardous waste incinerators.
Volatile POHCs are those POHCs with boiling points between 30° and 100°C.
The analytical portion of this procedure is very similar to the procedure
described in EPA Reference Method 602 - Purgeable Aromatics which requires
thermal desorption of the sample onto an analytical trap.

     The EPA Urban Air Toxics Monitoring Program is a national ambient air
screening study, sponsored by EPA, to help State and local agencies address
the magnitude of the air toxics problem in urban areas.  Three types of
samples are collected under the program.  One type is collected using
stainless steel canisters.  This method is appropriate for VOC, including
benzene.  The canister method is described in this section.
                                      170
-------
SAMPLING AND ANALYTICAL PROCEDURES

Reference Method 21 - Determination of Volatile Organic Compound Leaks

     A portable volatile organic compound (VOC) instrument detector capable
of measuring benzene concentrations less than 10,000 ppmv is used to
determine whether there are fugitive emissions or no detectable emissions.
Some detection methods that are acceptable are catalytic oxidation, flame
ionization, infrared absorption, and photoionization.

     The precision and accuracy of this" method are dependent on the
compounds that are to be identified and the detection instrument that is
chosen.  Due to the nonlinear response of benzene around 10,000 ppmv, this
technique cannot be used to quantitatively determine fugitive VOC emissions
or to predict mass emission rates from fugitive sources.

     An approach used in the past to predict mass emissions due to
industrial fugitive sources has been equipment inventories.  This involved
physically enclosing all fugitive sources within a plant with an inert bag
and measuring the emissions to determine the levels of and frequency of
           6
occurrence.

     As an alternative to an equipment inventory,  a fenceline measurement
technique can be used to quantitate fugitive emission concentrations at
ground level outside of a plant.   A study was conducted outside of a maleic
anhydride plant utilizing the Method 18 integrated bag sampling technique
described later in this section.

     Ambient air samples were collected from strategic locations around the
plant.  Care was taken when choosing the sampling locations so that the
background concentration of benzene in the air could be differentiated from
the concentrations attributed to fugitive and source emissions.  Benzene
analysis was conducted using a gas chromatograph (GC) equipped with a flame
ionization detector and a 10 ft. x 1/8 in. column of 10 percent
SP-2100/0.1 percent Carbowax on 100/120 mesh Supelcoport.
                                      171
-------
     Modeling predictions were compared to the test data results to confirm
the sources of emissions.  This ambient monitoring study concluded that the
approach utilized in this technique may be accurate enough to predict actual
benzene fugitive emissions from industrial sources.

Reference Method 18 - Measurement of Gaseous Organic Compound Emissions by
Gas Chromatographv

     A field site presurvey is initially conducted to obtain all the
information necessary to design an emission test.  Presurvey data include,
but are not limited .to, the average stack temperature and range, static
pressure, and the particulate and moisture concentrations in the stack
effluent.  In addition,"presurvey samples are collected to identify and
approximate the concentration of benzene and other components in the
effluent.  When this information is known, the most appropriate sampling and
analytical scheme to use can be determined.

     The prescribed sampling protocol requires that the entire sampling
train be leak free throughout the entire source testing process.
Precautions to eliminate interferences from particulate matter, moisture
condensation, and contamination need to be included in the sampling
protocol.

     There are basically four procedures that may be utilized to sample and
analyze benzene from stack effluents.  The correct procedure to use is
dependent on the source conditions.

Direct Interface Sampling and Analytical Procedure--

     This procedure can be utilized for determining benzene concentrations
from sources with temperatures less than 100 C.  This procedure requires
that the analyte concentration is not so great as  to saturate the detector
and that the moisture  content of the gas does not  interfere with the
analysis.  A calibration standard or sample is heated to 3 C above the
                                      172
-------
source temperature and pumped through a sample line to a GC.  The instrument
should include a Flame lonization Detector (FID), column, temperature
controlled loop and valve assembly, and temperature programmable oven.  A
schematic diagram of the direct interface sampling system is shown in
Figure 27.

     The appropriate column to use would be one that yields rapid and
complete elution of benzene.  The National Institute for Occupational Safety
and Health (NIOSH) recommends using a glass,  3.0 m x 2 mm column, 10 percent
OV-275 on 100/120 mesh Chromosorb W-AW, or equivalent.   This column will
prevent interference by alkanes, but if interference by other volatile
organic solvents is suspected, a less polar column is recommended.   Optimum
chromatographic conditions can be determined with calibration gas standards
by varying the temperature program and/or the flow rate through the column.

     Once a calibration curve has been established and optimum operating
conditions are known, the samples can be analyzed using the same conditions
as those used when analyzing the calibration gas mixture.  Thorough flushing
of the sample loop with the sample prior to analysis is necessary to
eliminate crossover between standards and samples.  Analysis is repeated
until two consecutive analysis peak areas agree to within 5 percent of their
mean value.

Dilution Interface Sampling and Analytical Procedure--

     Some source samples may contain relatively high concentrations of
benzene or other volatile organics which may overload the detector during
analysis.  A dilution interface sampling and analysis procedure allows for
either a 10:1 or 100:1 dilution of the stack gas prior to detection.

     The dilution interface sampling procedure is similar to the direct
interface procedure.   It requires a dilution system consisting of three
pumps, three-way valves and connections contained in a box capable of
heating to 120 C.  The heated box is placed in line between the heated probe
                                      173
-------
                          U81KU8C
                                                 E
                                                 N
                                                CO

                                                o
                                                  .


                                                CO
                                               CO
                                                CD
                                                O
                                                03
                                               «4—

                                               CD
                                               o
                                               CD

                                              5
                                               CD
                                               O


                                              2
                                              a>
                                              to
                                              CO


                                              CD
                                             -C
                                              o
                                             CO
                                             CM

                                             CD
174
-------
and the heated gas sampling valve in the gas chromatograph.  The temperature
of the system is set to 3 C above the source temperature.  A schematic
diagram of the heated box required for the dilution interface sampling
procedure is shown in Figure 28.

     The analytical procedure for samples taken by using the dilution
interface procedure is similar to that when using the direct interface
procedure.  The proper operation of the dilution system is proven by
analyzing high level standards that are split through the 10:1 and 100:1
dilution stages.  Data from the diluted standards must agree within
10 percent of the known values.  Samples must be analyzed using the same
chromatographic conditions and dilution ratios as the calibration mixture.
Analysis is repeated until two consecutive analysis peak areas agree to
within 5 percent of their mean value.

The Integrated Bag Sampling and Analytical Procedure

     The integrated bag sampling technique collects samples by pulling stack
gas into a prepurged evacuated bag, which is encased in a rigid container,
at a rate proportional to the stack velocity.  A diaphram-type pump, capable
of delivering at least 1.0 L/minute, and a needle valve connected by a
vacuum line to the bag container are in line prior to a flow meter.  If
samples contain high concentrations of organic materials, a dilution
interface system used in conjunction with the integrated bag sampling
procedure will prevent instrument detector saturation.

     Variations of the integrated bag sampling procedure are:

     1.   the direct pump bag sampling procedure, and
     2.   the explosion risk area bag sampling procedure.

The first procedure places the pump and needle valve between the probe and
bag, while the latter replaces the pump with another evacuated can.
Figures 29 and 30 are schematic diagrams that show these modifications to
                                      175
-------




o
i
»»
i ,
*
*»
-





W
4.
|
C
.k
V
^
«
rii'J
"j
if
!
•





> «
!§•
I*.
V>
IS''
W








8!
••
a


i
r








8S
«

9
«.« Sc
III If 5
Is5 II I

!!!T3~^ s" -
OL i_
r vi § 5 - cQ ^
t «!*»«»• « X
*| ?ie I ?I
s' ;si I go,
-m s«u ? x
^f* "U s S-e
1 .^^ ^o *J ^~
| > «-4 C \ /i\
«O-«- \ O CO
/*vi *^^ 1 S co co
_/vn >» M / ~* »- 1
fv^U ^££ J - §>»
j|l 	 M-^X « S =
°- T ^ « o c
«t- s ?^
barfs i |i
^S10. o
! co
A ^t; oo
f o « CM
11 £
. J " 2.
                                  O)
•90.

-s
se
       176
-------
                                          O>
                                          oa c
                                          coco
                                             ?
                                         COT?
                                            
-------
                                                      •*r 03

                                                       I?
                                                       co -=
                                                      Q CO
                                                      o -co
                                                      '£Z  CO
                                                      co O

                                                      § co
                                                      O (I)
                                                       . CO
                                                      o  c
                                                         X
                                                        UJ
178
-------
the integrated bag sampling procedure.  Other modified bag sampling
procedures may be considered when condensation is formed in the bag during
sampling and the direct method of sampling can not be used.  By heating the
rigid container and sample bag to the stack temperature during sampling,
condensation in the sample will be eliminated provided that the sample
remains at the source temperature until analysis and that the components of
the bag and rigid container can withstand the heat.   Tedlar bag seals
decompose at temperatures greater than 105 C.   An alternate procedure is to
partially fill the sample bag with a known amount of inert gas prior to
sample introduction.  A measured amount of sample is then added into the
partially filled bag and appropriate dilution factors are developed.

     The analytical procedure for samples taken by using any of the
variations of the integrated bag sampling procedure  is similar to that when
using the direct interface procedure.  The difference is that samples have
up to two hours after sampling before analysis is required.  This is an
advantage which allows the analyst time for other tasks.

     Another advantage to using the integrated bag sampling procedure is
that it does not restrict the analytical instrumentation location to the
vicinity of the source location.

Adsorption Tube Sampling and Analytical Procedure--

     Another Method 18 procedure that is approved by EPA for the sampling
and analysis of benzene is the NIOSH Method No. 1501, issued on
February 15, 1984.   This method is applicable for sampling benzene
concentrations between 0.09-0.35 mg benzene per 1.0  mL desorption volume.
The major components of this train include absorption tubes, a personal
sampling pump (0.01 to 1.0 L/min),  and a gas chromatograph equipped with a
FID, column, and integrator.

     For the purposes of source sampling, the adsorption tubes contain
800 mg coconut charcoal in the primary tube and 200  mg coconut charcoal in
the backup tube.  An alternate adsorption medium is  silica gel.   Adsorption
                                      179
-------
tubes containing silica gel have 1040 mg in the primary tube and 260 mg in
the backup tube.  Tenax GC resin has been used in the past but can yield
                                     9
false positive responses for benzene.   Desorption efficiency of the
adsorption medium must be determined by using an internal standard in the
expected sample concentration range.

     A sample pump and If .    orifice flow rate are calibrated prior to
sampling with an adsorption tube in line by using a bubble flow meter.
During sampling, a rotometer verifies that the sample rate remains constant
at <0.20 L/minute.

     Sample analysis requires transferring the adsorption medium into front
and back half vials, adding 1 mL of carbon disulfide eluent into each vial
and allowing the sample to stand for 30 minutes with occasional agitation.
Then, 5 uL of this solution are injected into a precalibrated gas
chromatograph.  Instrument calibration is conducted daily by the least
squares method over the appropriate range.  Sample collection efficiency is
determined by analyzing the front and back tubes separately.  If the backup
portion contains more than 10 percent of the total loading, sample
breakthrough has occurred and a larger sampling portion of adsorption
material is required.  If sample*; contain high concentrations of benzene, a
dilution interface system used in conjunction with the adsorption tubes will
reduce instrument overloading and sample breakthrough.

     The principal interference in this method is water vapor. - This can be
eliminated by the use of silica gel in line prior to the adsorption medium.
If more than one compound is, detected in the gaseous emissions, relative
adsorptive capacity information must be determined.  If present, water vapor
may reduce this capacity further.

Protocol for the Collection and Analysis of Volatile POHCs Using VOST

     Effluent gas is sampled with & Volatile Organic Sampling Train (VOST).
The gas is cooled to 20°C and pulled through sorbent cartridge pairs
containing approximately 1.6 grams of Tenax GC resin in the first trap, and
                                      180
-------
Tenax GC resin followed by petroleum based charcoal (weight ratio 3:1) in
the second trap.  Twenty liters of sample are collected during the sampling
period.  This may require up to six pairs of cartridges for each sample.   A
schematic diagram of the VOST is shown in Figure 31.

     Due to high levels of moisture associated with the types of effluent
that this method was designed to sample,  the recommended analysis protocol
is by thermal desorption, purge and trap by gas chromatography/mass
spectrometry (P-T-0, GC/MS).     The sorbent cartridges are spiked with an
internal standard and thermally desorbed with nitrogen through 5 mL of
organic-free water onto an analytical absorbant trap.   The analytical trap
is rapidly heated to 180 C into a temperature programmed gas chromatograph
that separates the effluent components.  Low resolution ma.ss spectrometry
detects the compounds and quantitates them based on the internal standard.
     A laboratory audit was conducted at four laboratories to evaluate the
combined accuracy and precision of the VOST method.    Three of the
laboratories analyzed their samples using thermal desorption purge and trap
and gas chromatography/spectrometry (GC/MS).   The other laboratory analyzed
their samples by photoionization detector/Hall detector.  The results of
this audit indicate that based on an 18 ppb benzene standard, the VOST
procedure can yield precision between 16.3 + 1.2 ppb and 23.7 + 1.0 ppb
(mean + standard deviation).   These results indicate an accuracy range of
the VOST method of between -9.7 percent to 32.0 percent.  The VOST method is
not officially designated as an approved method by EPA.

                                                          12
Stainless Steel Canister Method - Urban Air Toxics Program
     Urban air toxics samples are collected in evacuated 6-liter
polished canisters.  SUMMA-'polishing involves treating the canisters such
that a pure chrome/nickel oxide is formed on the interior surface.  Canister
sampling for VOCs is an alternative to VOC collection on solid sorbents
(e.g., Tenax).  Many VOCs which cannot be collected on Tenax (due to high
                                      181
-------
Ul
3
                                         CD
                                         H-* "
                                         H-.E
                                         O (0
                                         EH
                                         2 en
                                         o>.E

                                         Si
                                         O <0
                                         •^:co
                                         2 o
                                          o
                                             >
                                          qj CO
                                          =
                                          SP
                                         LL
        182
-------
volatility, polarity, or reactivity) can be efficiently collected and stored
in canisters.  Also, the canister contents may be analyzed repeatedly.  The
method is appropriate for benzene and is described in more detail below.

     The evacuated canister method does not require a pump to collect a
sample.  Since the canister is at a pressure lower than atmospheric, the air
sample flows through the heated manifold and mass flow controller and into
the canister.  The sampling assembly is shown in Figure 32.  The electronic
flow controller regulates and maintains a constant flow of approximately
3 cc/min to the canister.  Prior to sample collection, a vacuum pump is
activated by a timer to flush the heated manifold and particulate filter
assembly with sample air.  After a predetermined flushing period, a second
timer opens the magnelatch solenoid valve to allow sample gas to enter the
evacuated canister.  After sample collection, the canisters are returned to
a central laboratory for analysis of the ambient air samples.  Collection of
ambient air samples in canisters provides a number of advantages, including:

          convenient integration of ambient samples over a specific period,
          ability to ship and store samples,
          remote sampling capability with subsequent central laboratory
          analysis,
          unattended sample collection,
          analysis of samples from multiple sites with one analytical
          system, and
          collection of duplicate samples for assessment of sampling
          precision.

Sample Analysis

     The air toxics samples are analyzed by a gas chromatograph equipped
with photoionization, electron capture, and flame ionization detectors.  For
separation and subsequent identification by the selective 'detectors, the GC
multi-detector technique relies on the analytes' retention times and
selective detector peak response ratios for specific compounds such as
benzene.
                                      183
-------
HEATED INLET LINE
*&&&&''
*
1 ^
1
1
I
1
1
1
PROG.
TIMER
^M
1
1
l

PUMP
"x
-------
     The analytical system consists of a cryogenic trap, a GC column, and
selective detectors.  In the cryogenic trap, the air toxics are concentrated
prior to GC separation.  The trap is then heated and the previously
collected organic compounds volatilize and are carried to the GC column by
the carrier gas.  The GC column is temperature programmed for subambient
cooling to provide improved resolution for the compounds of interest.  After
the column separates the organics, the effluent gas stream containing the
separated organics enters a series of detectors consisting of electron
capture, photoionization and flame ionization.  The individual detectors
respond to the sample components as a succession of peaks above a baseline
on the chromatogram.

     The area under the peak approximately quantifies the component, while
the time lapse between injection and emergence of the peak serves as a
preliminary identification.  Detector peak response ratios are also used to
perform peak identification.  Further confirmation and identification will
be performed on 20 percent of the total samples by gas chromatography/mass
spectroscopy (GC/MS).  Concentrations of toxic species are reported in units
of ppb by volume (ppbv).
                                      185
-------
REFERENCES FOR SECTION 6
 1.  Addition of Reference Method 2L to Appendix A.  Federal Register.
     48(161) :37600-37602.  August 1:3, 1983.

 2.  Standards of Performance for New Stationary Sources:  Synthetic Organic
     Chemical Manufacturing Industry; Equipment Leaks of VOC, Reference
     Methods 18 and.22.  Federal Register.  48(202):48344-48360.
     October 18, 1983.

 3.  Cheney, J. L. and D. Bullard.  Measurements and Estimations of Local
     Impacts from Fugitive Emissions.  Journal of Environmental Science and
     Health.  Al7(6):819-835.   1982.

 4.  Hansen, E. M.  Protocol for the Collection and Analysis of Volatile
     POHCs Using VOST.  Prepared for U. S. Environmental Protection Agency.
     Research Triangle Park, NC.  Publication No. EPA-600/58-84-007.
     August 1983.

 5.  Guidelines Establishing Test Procedures for the Analysis of Pollutants.
     Federal Register.  44(233):69474-69578.  Decembers, 1979.

 6.  State of California Air Resources Board.  Emissions from Leaking
     Valves, Flanges, Pump, and Compressor Seals, and Other Equipment at Oil
     Refineries.  LE-78-001.  April 24, 1978.  86 p.

 7.  NIOSH Manual of Analytical Methods, U. S. Department of Health and
     Human Services.  National Institute for Occupational Safety and Health.
     Center for Disease Control.  Cincinnati, OH.  February 1984.  Third
     Edition.

 8.  Knoll, J. E., W. H. Penny, and M. R. Midgett.  The Use of Tedlar Bags
     to Contain Gaseous Benzene Samples at Source Level Concentrations.
     U. S. Environmental Protection Agency.  Research Triangle Park,  NC.
     Publication No. EPA-600/4-78-057.  September 1978.  38 p.

 9.  Telephone communication.  Syke:s, A., Radian Corporation with
     Holliman, D.  November 19, 1986.  Advantages and limitations of using
     Tenax adsorption medium for benzene sampling.

10.  Guidelines Establishing Test Procedures for the Analysis of Pollutants.
     Federal Register.  44(233) :69f>32-69539.  Decembers, 1979.

11.  Jayanty, R. K. M.  Performance Audit Results for POHC  (Principle
     Organic Hazardous Constituents;) :  VOST  (Volatile Organic Sampling
     Train) and Bag Measurement Methods.  Research Triangle Institute.
     Research Triangle Park, NC.  Publication No. EPA-600/4-84-036.
     May  1984.  41 p.

12.  U. S. Environmental Protection Agency.  Urban Air Toxics Monitoring
     Program.  Office of Air Quality  Planning and Standards.  Research
     Triangle Park, NC.  EPA-450/4-87-022.   September 1987.

                                       186
-------
          APPENDIX A




DERIVATION OF EMISSION FACTORS
-------
1
9878
1224
2
6857
857
3
5469
735
4
4653
612
5
3837
490
Years (Me/Year) -
6138
783
.7
.6
                                 APPENDIX A
                       DERIVATION OF EMISSION FACTORS

A.I  BENZENE EMISSIONS FROM ETHYLENE PLANTS

     The emission factor for other intermittent emissions (Vent E) in
Table 23 were derived as follows.

     Data from Reference 1 showed that, for a 453.5 Gg/year capacity
ethylene plant with 50:50 naphthalene/gas oil feed, material lost as a
                •
result of compressor outages over the first 5 years of plant operation were:
                              Total Material
                      Lost (Mg) by Year of Operation        Average Over Five
Single train
Dual train
On a per hour basis, the material lost from compressor outages in
700.8 kg/hr for single trains and 89.5 kg/hr for dual trains (assuming
8760 hr/year operation).  Then, assuming that there is 5.85 percent benzene
in the charge gas (Reference 1):

     For single trains - 700.8 kg/hr x 0.0585 - 40.99 kg benzene/hr
     For dual trains   -  89.5 kg/hr x 0.0585 -  5.24 kg benzene/hr

Then the ratio of 453.5:544.2 (plant capacity) was applied to the emission
rates for benzene to account for differences in material use and the
emission rate in kg/hr was converted to g/Mg ethylene:

Single train - 40.99 x 1.2 (size ratio) x 16.1 (conversion to g/Mg) - 791.9
Dual train   -  5.24 x 1.2 (size ratio) x 16.1 (conversion to g/Mg) - 101.1
                                     A-l
-------
     For secondary emissions (emissions from wastewater handling and
treatment),  the emission factor was derived from estimates of the amount of
wastewater produced per Gg ethylene,  the average concentration of total
organics in the wastewater, the estimated kg VOC emitted/kg wastewater, and
the estimated benzene concentration in the emitted VOC.  All data were taken
from Reference 1.

For a 544 Gg/year plant:

     2166 m  wastewater generated/Gg ethylene x 544 Gg ethylene -
   .  1,178,304 m  wastewater or 1,178,304,000 kg wastewater per year or
     134,510 kg/hr (based on 8,760 hr/year operation)
      •
A VOC concentration in the wastewater of 104 ppm was used:

     104 kg/10  kg - X kg/134,510 kg or 14 kg VOC/hr in wastewater

Then, assuming 25 percent of the VOC is emitted and, of that, 38.5 percent
is benzene,

     14 x 0.25 x 0.385 - 1.35 kg/hr or 21.7 g/Mg

A.2  DERIVATION OF EMISSION FACTORS FOR FUGITIVE EMISSIONS FROM
     CHLOROBENZENE PRODUCTION

     Emission factors for fugitive benzene emissions from chlorobenzene
manufacture were based on factors established for fugitive emissions at
           2
refineries.    According to the source of these data, preliminary test
results suggest that fugitive emissions from refineries are comparable to
                         2
those at chemical plants.   Thus, Che factors for benzene in Table 24 are
based on summation of the factors in Table A-l and production capacity of
96 Gg/year.   The hypothetical plant is estimated to have 102 pumps handling
VOC, 22 of which handle benzene.  The estimated number of valves is 792, 45
of which service benzene vapor and 173 service benzene liquid.  The
                                     A-2
-------
    TABLE A-l.  FUGITIVE EMISSION FACTORS FOR PETROLEUM REFINERIES USED
                TO DEVELOP FACTORS FOR CHLOROBENZENE PRODUCTION
       Source
 Uncontrolled
Emission Factor
    (kg/hr)
   Controlled
Emission Factor'
     (kg/hr)
Pump Seals
Light- liquid service
Heavy- liquid service
Pipeline Valves
Gas/vapor .service
Light- liquid service
Heavy- liquid service
Safety /Relief Valves
Gas/vapor service
Light- liquid service
Heavy- liquid service
Compressor Seals
Flanges
Drains

0.12
0.02

0.021
0.010
0.0003

0.16
0.006
0.009
0.44
0.00026
0.032

0.03
0.02

0.002
0.003
0.0003
"
0.061
0.006
0.009
0.11
0.00026
0.019
a
Based on monthly inspection of selected equipment; no inspection of
heavy-liquid equipment, flanges, or light-liquid relief valves; 10,000 ppmv
VOC concentration at source defines a leak; and 15 days allowed for
correction of leaks.
Light-liquid means any liquid more volatile than kerosene.
                                    A-3
-------
estimated number of pressure relief valves is 12, 6 of which service
benzene.  The estimated number of flanges is 1848, of which 486 service
benzene.  The fugitive emission factors from Table A-l were applied to this
valve, pump, and flange count to determine the fugitive emissions shown in
Table 24.2

     The storage- and handling-emission calculations in Table 24 were based
on fixed-roof tanks, half full, and a diurnal temperature variation of 11°C
                                                      2
and with the use of the emission equations from AP-42.   However, breathing
losses were divided by four to account for recent evidence indicating that
                                                          2
the AP-42 breathing loss equation overestimates emissions.

     Breathing and working losses can be estimated from emission equations
developed for EPA Publication No. AP-42 (Reference 3).  The equations used
in estimating emission rates from fixed-roof tanks are:

     Ltotal " LB + ^J
     L-1.02xlO-5M   __P	  0.681.730.510.5
      B                v  14.7 - P                         P  C
     L.. - 1.09 x 10'8 M PVNK K                                        (2-2)
      w                v    n c
Where:
     1^, - total loss per tank  (Mg/yr)
     Lg - breathing loss per tank (Mg/yr)
     L., » working loss per tank (M§;/yr)
     M  - molecular weight of  product vapor (Ib/lb mole)
     P  - true vapor pressure  of product (psia)
     D  - tank diameter  (ft)
     H  - average vapor  space  height (ft); use an assumed value  of  one-half
          the tank height
     T  - average diurnal temperature change in  F; assume  20  F  as  typical
          value
     F  - paint factor  (dimensionless)
     C  - tank diameter  factor (dimensionless) for diameter equal to  or
          >30 feet, C -  I
                                     A-4
-------
     K  - product factor (dimensionless) - 1.0 for volatile organic liquids
      °   (VOL)
     V  — tank capacity (bbl)
     N  — number of turnovers per year (dimensionless)
     K  - 1, for turnovers <36
      n

Several assumptions were made in order to calculate emission factors on a
per tank basis, for both breathing and working losses, from a typical
fixed-roof tank storing gasoline.  The following assumptions were used:

     M  - 66 Ib/lb mole (for gasoline)
     P  - 5.2 psia (for gasoline)
     D  - 50 feet
     H  - 48/2 - 24 feet
     T  - 20°F
     F  -1.0 (for a white tank)
      P
     C  - 1.0
     K  - 1.0 (for VOL)

Therefore, after substituting into equation 2-1,

Ln - 1.02 x 10'5 (66)      5.2     °-68  (50)1'73 (24)°'51 (20)°'5(1)(1)(1)
 B                     14.7  - 5.2
   - 8.8 Mg/yr

Using equation 2-2 and these additional  assumptions:

     N  - 13 turnovers per year
     V  - 16,750 bbl tank capacity
     LW - 1.09 x 10"8 (66) (5.2) (16,750) (13) (1)  (1)
        - 34.2 Mg/yr

In summary, the VOC emission factors for a typical  fixed-roof tank storing
gasoline are 8.8 Mg/yr from breathing losses and 34.2 Mg/yr from working
losses.
                                     A-5
-------
     Standing-storage losses and withdrawal losses are the major sources  of
emissions from external floating-roof storage tanks.  From the equations
presented below, it is possible to estimate both the withdrawal l,oss and  the
standing-storage or roof seal loss from an external floating-roof  tank.
These equations are taken from Reference 3.
     Lv.  - 4.28 x 10"4 QCW /D (1 + N F /D)                        (2-3)
      W                   Lt         C C
     LSE " Ks Vn P*DMVKC/2205                                     (2'4)
Where:
     Lw.  - total loss (Mg/yr)
      w  - withdrawal loss (Mg/yr)
     LS_, - standing-storage or seal loss  (Mg/yr)
     Q   - product average throughput (bbl/yr)
                                                           3   2
     C   - product withdrawal shell clingage factor  (bbl/10  ft )
     D   - zero for self-supporting floating-roof tanks
      C
     F   - effective column diameter
      c
     WT  - density of product (Ib/gal)
     D   - tank diameter  (ft)
     K   - seal factor (dimensionless)
      s
     V   - average windspeed (mph); 10 mph assumed average windspeed
     N   - seal windspeed exponent (dimensionless)
     p*  - vapor pressure function (dimensionless)
     M   - molecular weight of product vapor (Ib/lb  mole)
     K   - product factor (dimensionless) - 1.0 for  VOL

For the purposes of calculating the external floating-roof emission factors,
several additional assumptions were made  as follows:

     Q   - the value for  product throughput varied from  State  to  State
           (bbl/yr)
     C   - 0.0015 (for light rust)
     WT  — 5.6 Ib/gal (for gasoline)
     K,,  - 1.2 (for a metallic shoe with  primary seal, assume  welded tank)
           and 0.8 (for a metallic shoe with secondary shoe  s,eal) ;  0.2  (for
           metallic shoe  with rim mount secondary shoe seal)
                                     A-6
-------
     V   =10 mph
     N   -1.5 (for a metallic shoe with primary seal) and 1.2 (for a
           metallic shoe with secondary seal)
     P*  - (P/P)/[1 + (1 - P/P)°'5]2
           Where:        P  - average atmospheric pressure - 14.7 psia
                          A
                         P  - true vapor pressure at average actual organic
                              liquid storage temperature =5.6 psia
           Therefore:    P* - (5.6/14.7)/[l + (1 - 5.6/14.7)°' 5]2 = 0.10871
Therefore, after substituting into equation 2-3 and 2-4:
         = 4.28 x 10"4 Q (0.0015) (5.6)/78
         = 36 x 10"7 Q/78 Mg/yr
         - (1.2) (10)1'5 (0.10871) (78) (66) (1)/2205
           9.6 Mg/yr for a metallic shoe with primary seal
           (0.8) (10)1'2 (0.10871) (78) (66) (D/2205
           3.2 Mg/yr for a metallic shoe with primary seal
     In summary, the VOC emission factors for a typical external
floating-roof tank storing gasoline are 36 x 10   Q/78 Mg/yr from withdrawal
losses, 9.6 Mg/yr from storage or seal losses on a tank with a metallic shoe
primary seal and 3.2 Mg/yr from storage or seal losses on a tank with a
metallic shoe secondary seal.
A.3  DERIVATION OF BENZENE EMISSION FACTOR FOR EXHAUST EMISSION
     OF LIGHT DUTY AND HEAVY DUTY TRUCKS

     Reference 4 provided the following derivation of emission factors for
benzene from truck exhaust:
Light duty trucks:  3.4 g hydrocarbon x 1.17% benzene x 0.45 - 0.02 g/mile
                          mile          in gasoline
Heavy duty trucks:  13.6 g hydrocarbon x 1.17% benzene x 0.45 - 0.07 g/mile
                          mile           in gasoline
                                     A-7
-------
The factor of 0.45 converts from v/v percent of benzene in the gasoline
liquid to the w/w percent benzene in the vapor.

A.4  BENZENE EMISSIONS FOR VARIOUS VEHICLE TEST TYPES AND CONTROL
     TECHNOLOGIES

     Table A-2 lists observed benzene emissions and hydrocarbon emissions
for a series of tests on vehicles equipped with different control
technologies.
                                     A-8
-------
            TABLE A-2.   BENZENE EMISSIONS FOR DIFFERENT EXHAUST

                        EMISSION CONTROL TECHNOLOGIES5


CUE (Crowded Urban Expressway)
No Cat/Air
Oxid. Cat/No Air
Oxid. Cat/Air
Dual -Bed Cat/Air
*3-Way Cat/No Air
Federal Test Procedures
No Cat/Air
Oxid. Cat/No Air
Oxid. Cat/Air
Dual -Bed Cat/Air
*3-Way Cat/No Air
NYCC (New York Citv Cvcle)
No Cat/Air
Oxid. Cat/No Air
Oxid. Cat/Air
Dual -Bed Cat/Air
*3-Way Cat/No Air
N
3
12
10
12
3

3
12
10
12
3
3
12
10
12
3
HC
(g/mile)
1.173
1.360
0.892
0.414
0.160

2.233
1.960
1.626
0.603
0.350
7.897
6.092
6.013
2 . 447
1.390
% Benzene
2.42
4.11
3.42
3.83
4.88

2.33
3.26
2.81
2.63
4.16
1.81
3.03
2.23
2.14
2.92
Benzene
Emissions
(rag/mile)
31.1
54.7
30.0
16.4
9.4
-
52.3
61.9
45.1
16.6
14.7
150.0
170.4
148.3
56.3
40.6
Single-bed
Cat - catalyst
N - sample size
                                    A-9
-------
REFERENCES FOR APPENDIX A
1.   U.S. Environmental Protection Agency.   Organic Chemical Manufacturing.
     Volume 9:  Selected Processes.  Report 3.   Ethylene.
     EPA-450/3-80-028d.  Research Triangle Park,  NC.  1980.

2.   U. S. Environmental Protection Agency.   Organic Chemical Manufacturing.
     Volume 6:  Selected Processes.  Report 1.   EPA-450/3-80-028a.  Research
     Triangle Park, NC.  1980.

3.   U. S. Environmental Protection Agency.   Compilation of Air Pollutant
     Emission Factors.  Fourth Edition (Including Supplements 1-15 and
     Updates).  Publication No. AP-42/  Research Triangle Park, NC.  1985.

4.   U. S. Environmental Protection Agency.   Materials Balance for Benzene
     Level II.  EPA-560/13-80-009.  Washington, D.C.  1980.   pp. 2-6 - 2-34.

5.   Klausmeir, R., P. Anderson, and K. Wert.  Evaluation of Mobile Source
     Control Strategies for Benzene.  Prepared for the Western Oil and Gas
     Association.  Radian Corporation DCN No. 85-245-024-01.  Austin, TX.
     1985.
                                      A-10
-------
                 APPENDIX B "

DATA ON POTENTIAL SOURCES OF BENZENE FROM THE
 AIR EMISSIONS SPECIES MANUAL (DRAFT REPORT)
-------
-------
                                 APPENDIX B
                DATA ON POTENTIAL SOURCES OF BENZENE FROM THE
                 AIR EMISSIONS SPECIES MANUAL (DRAFT REPORT)
     The Pollutant Characterization Section, Noncriteria Pollutant Programs
Branch of the U. S. Environmental Protection Agency is preparing a document
that will assist in determining the composition of many emission streams
from a variety of sources.    The data in that document are arranged in
profiles containing the source type, control device,  pollutant,  CAS number,
and weight percents of emission stream components.  To provide the most data
available about potential sources of benzene, the most recent species
profiles from the draft report were reviewed and summarized.

     Table B-l shows the source descriptions, source classification code
(SCC), indication of data quality, and percent (weight) of the emission that
is benzene.  The actual species profiles (and appropriate references cited
in the profiles) should be checked before applying any of the profile data.

     These profiles are presented to provide a means to estimate benzene
emissions when a benzene-specific emission factor has not been established.
For example, if the total VOC emissions for aircraft landings/takeoff are
known, then the species data showing 1.79 weight percent benzene would be
applied to the total VOC emissions to obtain an estimate of benzene
emissions.

     The species profile data are provided to supplement the emission
factors in the main body of the report.  The emission factors in the text
were obtained from published literature and, in some cases, may be more
process-specific than the factors obtained using the VOC species data in
Table B-l.   Before applying the emission factors on the species profiles,
the reader is encouraged to review the references for both types of data to
discern any peculiarities concerning the derivation of the data.

     Contact the Pollutant Characterization Section,  Noncriteria Pollutant
Programs Branch, U. S. Environmental Protection Agency, Research Triangle
Park, North Carolina, for more information about the benzene data summarized
here, or for more information on the VOC species profiles.
                                     B-l
-------
     TABLE B-l.   POTENTIAL SOURCES OF BENZENE EMISSIONS AND APPROXIMATE
                 BENZENE COMPOSITION OF THE EMISSIONS3
    SCC
        Source Description
Weight
Percent
Benzene
Profile
 Data
Quality
3-03-009-01
4-02-999-99
1-02-007-07
3-90-007-02

2-02-004-01
2-04-004-02

2-01-001-02
2-02-001-02
2-03-001-01

3-03-003-02
3-03-003-03
3-03-003-05
3-03-003-06

3-05-001-04
3-05-002-02
3-06-005-03
3-06-005-04
5-01-001-01
5-01-001-02

3-90-007-01
 3-01-026-20
Open Hearth Furnace with Oxygen
Lance - ESP Control

Flavorseal Citrus Coating Wax -
Uncontrolled Surface Coating

External Combustion Boiler Burning
Coke Oven Gas - Uncontrolled

Reciprocating Diesel Fuel Engine -
Uncontrolled

Reciprocating Distillate Oil Engine
Uncontrolled
By-product Coke Oven Stack Gas -
Uncontrolled
 13.60
  3.11
  1.90
  7.90
  7.90
 14.10
Asphalt Roofing Tar Kettle -               0.80
Uncontrolled

Asphaltic Concrete - In Place Road         9.50
Asphalt - Uncontrolled

Refinery Fugitives - Covered Drainage/     2.40
Separation Pits - Uncontrolled

Solvent Use - General Pesticides -        12.30
Uncontrolled

Bar Screen Waste Incinerator -             7.70
Uncontrolled

Coke Oven Blast Furnace Gas -             43.00
Uncontrolled

Automotive Tires - Tuber Adhesive  -        2.80
Uncontrolled
                                     B-2
-------
TABLE B-l.  POTENTIAL SOURCES OF BENZENE EMISSIONS AND APPROXIMATE
            BENZENE COMPOSITION OF THE EMISSIONS3 (Continued)

sec
3-01-026-20
4-03-01-010
4-03-01-011
4-03-01-012
4-03-01-109
4-04-003-01
4-04-003-02
3-06-008-01
3-06-008-03
	
—
3-10-88-801
3-10-001-01
3-10-002-03
4-02-002-01
4-02-002-10
4-03-010-03
4-03-010-01
3-01-133-03
Source Description
Automotive Tires - Tuber Adhesive,
White Sidewall - Uncontrolled
Fixed Roof Tank - Crude Oil Production
Fixed Roof Tank - Crude Oil Refinery -
Uncontrolled
Pipe/Valve Flanges - Uncontrolled
Pump Seals - Composite - Uncontrolled
Off -road Gasoline Motor Vehicles
Vessels - Gasoline
Oil and Gas Production - Fugitives
(Unclassified)
Oil and Gas Production - Fugitives -
Valves/Fittings in Liquid Service
Oil and Gas Production - Fugitives -
Valves/Fittings in Gas Service
Surface Coating Application -
Water-based Paint - Uncontrolled
Evaporative Emissions - Summer-blend
Gasoline - Uncontrolled
Evaporative Emissions - Winter-blend
Gasoline - Uncontrolled
Acetic Anhydride Production,
Weight
Percent
Benzene
7.70
0.10
2.40
0.10
0.50
2.66
2.66
0.10
0.10
0.10
0.36
0.77
1.56
2.00
Profile
Data
Quality
C
C
C
C
C
E
E
D
D
D
C
B
B
D
           Distillation Column Vent - Uncontrolled
                                B-3
-------
TABLE B-l.  POTENTIAL SOURCES OF BENZENE EMISSIONS AND APPROXIMATE
            BENZENE COMPOSITION OF THE EMISSIONSa (Continued)

sec
3-01-156-06
3-01-158-22
3-01-195-02
3-01-195-04
3-01-197-45
3-01-206-01
3-01-206-02
3-01-206-03
3-01-301-01

3-01-301-02
3-01-301-06
3-01-301-08

—
4-06-004-01
4-06-004-03
Source Description
Cumene Production, Cumene Distillation
System Vent - Uncontrolled
Cyclohexanone/Cyclohexanol - Phenol
Hydrogenation Process, Distillation
Vent - Condenser
Nitrobenzene Production - Reactor
Vent, Washer /Neutral izer Vent
Ethylen^ .-ion, Compressor Lube
Oil Venc
Styrene Production - General
Styrene Production - Benzene Recycle -
Condenser
Styrene Production, Styrene
Purification - Condenser
Chlorobenzene Tail Gas Scrubber -
Scrubber
Chlorobenzene - Benzene Drying
Distillation
Chlorobenzene - Vacuum System Vent -
Steam Jet
Chlorobenzene - Dichlorobenzene
Crystal Handling - Uncontrolled
Residential Wood Combustion
Gasoline Marketing (Refueling Cars) -
Uncontrolled
Weight
•Percent
Benzene
38.49
28.07
99.71

2.07
23.28
34.38
41.46
56.96

95.04
19.62
0.11

18.91
1.00
Profile
Data
Quality
C
E
E

C
D
D
E
C

D
C
E

D
C
            Commercial Aircraft: Landing/Takeoff  -
            Uncontrolled
1.93
                                 B-4
-------
     TABLE B-l.   POTENTIAL SOURCES OF BENZENE EMISSIONS AND APPROXIMATE
                 BENZENE COMPOSITION OF THE EMISSIONS3 (Continued)
    SCC
        Source Description
Weight      Profile
Percent      Data
Benzene     Quality
3-04-003-20
2-01-002-02
2-02-002-02
2-02-002-04
2-03-002-01
3-01-900-99

3-05-002-01


1-01-006-01
General Aviation - Aircraft Landing/       1.79
Takeoff - Uncontrolled

Military Aircraft Landing/Takeoff         .2.01

Light-duty Gasoline Vehicles -             2.80
Exhaust/Evaporative - Catalyst

Secondary Metal Production, Gray Iron     34.70
Foundries - Pouring/Casting (Exhaust
Gases from Binder Systems and Sand
Formulations)

Internal Combustion Engine -               0.11
Reciprocating Engine
Industrial Incineration                    1.49

Chemical Manufacturing - Flares           10.00

Asphaltic Concrete - Natural Gas-fired     4.00
Rotary Dryer

External Combustion Boiler - Natural       4.00
Gas-fired - Uncontrolled

Diesel Fuel - Light-duty, Heavy-duty,      1.90
and Off-highway Vehicles
                                                                        B

                                                                        B
               E

               D

               D
 Based on data gathered as part of the effort to revise the Air Emissions
 Species Manual.  Species profiles that showed benzene as a component of the
 total exhaust stream are summarized in this table.   The speciation data
 profiles used were dated August 4, 1987.   Profile data quality are defined
 as follows:
 Data Quality A:
    Data set based on a composite of several tests using
    analytical techniques such as GC/MS and can be
    considered representative of the total population.
                                     B-5
-------
     TABLE B-l.   POTENTIAL SOURCES OF BENZENE EMISSIONS  AND APPROXIMATE
                 BENZENE COMPOSITION OF THE EMISSIONS3 (Continued)
Data Quality B:
Data Quality C:
Data Quality D:
Data Quality E:
Data set based on a composite of several tests using
analytical techniques such as GC/MS and can be
considered representative of a large percentage of the
total population.

Data set based on a small number of tests using
analytical techniques such as GC/MS and can be
considered reasonably representative of the total
population.

Data set based on a single source using analytical
techniques such as GC/MS; or data set from number of
sources where data are based on engineering
calculations.

Data set(s) based on engineering judgement; data sets
with no documentation provided; may not be considered
representative of the total population.
Emission control devices are given in Table B-l if they were included in the
species profile data.
                                     B-6
-------
REFERENCES FOR APPENDIX B
1.   U. S. Environmental Protection Agency.  Air Emissions Species Manual.
     Volume I.  Volatile Organic Compound (VOC) Species Profiles.  Draft
     Report.  Noncriteria Pollutant Programs Branch.  Research Triangle
     Park, North Carolina.  October 1987.
                                     B-7
-------
-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing]
1. REPORT NO.
        EPA-450/4-84-007q
                              2.
                                  3. RECIPIENT'S ACCESSION NO.
A. TITLE AND SUBTITLE
  Locating And Estimating Air Emissions From  Sources
  Benzene
                                  5. REPORT DATE
                                         March 1988
                                                            6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)

   Patricia A. Cruse
                                  8. PERFORMING ORGANIZATION REPORT NO.

                                         87-203-061-03-15
9. PERFORMING ORGANIZATION NAME AND ADDRESS
   Radian Corporation
   Post Office Box 13000
   Research Triangle Park,  NC 27709
                                                            10. PROGRAM ELEMENT NO.
                                  11. CONTRACT/GRANT NO.

                                         68-02-4464, W/A //:   18
12. SPONSORING AGENCY NAME AND ADDRESS
  U.S.  Environmental Protection Agency
  OAR,  OAQPS, AQMD, NPPB,  PCS (MD-15)
  Research Triangle Park,  NC 27711
                                  13. TYPE OF REPORT AND PERIOD COVERED
                                          Final 8/86 - 3/88
                                  14. SPONSORING AGENCY CODE
15. SUPPLEMENTARY NOTES
  EPA Project Officer:
William B. Kuykendal
16. ABSTRACT
       To assist groups  interested in inventorying  air emissions of various  potentially
  toxic substances, EPA is  preparing a series  of documents such as this  to  compile
  available information on  sources and emissions of  these substances.  This document
  deals specifically with benzene.  Its intended audience includes Federal, State and
  local air pollution personnel and others interested in locating potential emitters of
  benzene making gross  estimates of air emissions  therefrom.

        This document presents  information on (1) the types of sources  that  may emit
  benzene, (2) process  variations and release  points that may be expected within these
  sources, and (3) available emissions information indicating the potential for benzene
  release into the air  from each operation.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                              b.lDENTIFIERS/OPEN ENDED TERMS C.  COSATI Field/Group
  Benzene
  Emissions Sources
  Locating Air Emission Sources
  Toxic  Substances
18. DISTRIBUTION STATEMENT

   Unlimited
                     19. SECURITY CLASS (This Reponj
                            Unclassified
                                                                          21. NO. OF PAGES
                     20. SECURITY CLASS (This page)

                     	Unclassified
                                                22. PRICE
22Q
EPA Form 2220-1 (R«v. 4-77)    PREVIOUS EDITION is OBSOLETE
-------
-------