-------
SyBCATEGQEY_B
Product
Formaldehyde Oxidation of Methanol
In the plant visited, formaldehyde is manufactured by oxidation of
methanol. The process is a gas-phase reaction, operated with an iron-
molybdenum oxide catalyst and a lean methanol-air mixture. The chemical
reaction is given below:
CH3OH + 1/2 02 —*• HCHO + H20
Methanol Oxygen Formaldehyde Water
A flow sheet for the methanol oxidation process is shown in Figure IV-
14. A mixture of methanol and water is vaporized by a closed steam
loop, which circulates between the reactor and feed vaporizer. The
reactants, mixed with air, flow through a thin layer of catalyst
crystals in the reactor. The product gases are cooled by water, and
product formaldehyde is recovered as a 50-52 percent aqueous solution by
two-stage absorption. Product concentration is adjusted by controlling
the amount of water supplied to the second stage absorber. The
remaining unabsorbed gases from the absorber are disposed of by
incineration.
A portion of the formaldehyde product may be passed through an anion ex-
changer to produce high purity formaldehyde by removing formic acid and
sodium formate.
Waste water streams generated in this process are intermittent. For
example, waste water from the washing of the absorber occurs at most
twice per year. The contaminants in this stream are formic acid,
methanol, formaldehyde, and ammonia. Wastewater created by regenerating
the ion exchange unirs occurs three times per month at the plant
visited. Another possible waste stream is withdrawn as an aqueous slip
stream from the bottom of the feed vaporizer whenever heavy impurities
(such as acetone and oxygenated organics) occur in the methanol feed;
the total flow of this waste stream, estimated by plant personnel, is
about 131 gallons per 1,000 pounds of formaldehyde. A sample was not
taken for analysis, since a continuous and representative sample is not
available.
The alternate approach for formaldehyde manufacture from methanol
involves a combined dehydrogenation and oxidation reaction over a silver
or copper catalyst. This process operates with a rich methanol-air
mixture.
127
-------
ffi
i
eo
tC:
a
i—
CJ
Lul
QC
S3
UJ
CO
CO
z
o
p
o
x
O
SO
8*
<
I
25
1
S3
g
S
S
x\
CO
I
B
"!
a
i;
t
TOO
-------
About 90 percent of the formaldehyde produced in the U.S. is based on
methanol as a raw material. The balance of the formaldehyde production
is as a co-product of butane oxidation. The basic chemical reaction is
summarized as follows:
2C3H8 + 2CUH1.0 + 902 —*• 14HCHO + 4H2O
Propane Butane Oxygen Formaldehyde Water
The U.S. formaldehyde capacity and the estimated economics for
formaldehyde production of a 100 million pounds per year (100 percent)
unit based on iron-molybdenum catalyst process are shown in Tables IV-20
and IV-21,
129
-------
Producer
Allied
American Petrofina
Borden
Celanese
Commercial Solvents
DuPont
GAP
Georgia Pacific
Gulf
Hercules
Monsanto
Table IV-20
U. S. Formaldehyde Capacity
Plant Location
Ironton, Ohio
Calumet City, 111.
Bainbridge, N. Y.
Demopolis, Ala.
Dibol1, Texas
Fayettevi1le, N. C.
Fremont, Calif.
Kent, Washington
La Grande, Oregon
Louisville, Ky.
Missoula, Mont.
Shebpygan.Wi sc.
Springfield, Oregon
Bishop, Texas
Newark, N. J.
Rock Hill, S. C.
Agnew, Calif.
Seiple, Pa.
Sterling, La.
Belle, W. Va.
LaPorte, Texas
Perth Amboy, N. J.
Toledo, Ohio
Calvert City, Ky.
Coos Bay, Ore.
Columbus, Ohio
Crosett, Ark.
Vicksburg, Miss.
Hercules, Calif.
Louisiana, Mo.
Addyston, Ohio
Eugene, Ore.
Springfield, Mass.
Estimated Capacity"1'
(MM Ibs. 37% Soln./Yr.)
310
75
80
70
200
80
70
M)
70
80
120
250
1,170
115
115
30
65
30
•^90
200
150
150
100
80
100
160
95
170
100
100
280
130
-------
Producer
Occi dental
Reichhold
Rohm and Haas
Skelly
Tenneco
U.C.C.
Wright
Table IV-20
(con't)
Plant Location
N. Tonawanda, N. Y,
Charlotte, N. C.
Hampton, S. C.
Kansas City, Kan.
Moncure, N. C.
Racoma, Wash.
Tuscaloosa, Ala.
White City, Ore.
Bristol, Pa.
Phi ladelphia, Pa.
Springfield, Ore.
Fords, N. J.
Garfield, N. J.
Boundbrook, N. J.
Acme, N. C.
Malvern, Ark.
TOTAL
Estimated Capaci ty-'--
(MM Ibs. 37% Soln. /Yr,
135
10
100
40
70
50
25
25
70
160
175
150
150
100
6.570
"Capacity data are as reported by Stanford Research Institute,
C.E.H. for late 1970
131
-------
Table IV-21
Estimated Economics for Formaldehyde Production
(100 MM Ib. 100% Formaldehyde Plant)
Total Fixed Capital=$0.45 MM
Estimated Operation Cost
Methanol
Catalyst and Chemicals
Utilities (including demineralized
process water)
Labor and overhead
Capital charges
TOTAL
Captive
methanol
3.5
0.3
0.4
0.8
6.5
Merchant
methanol
5.2
0.3
0.4
0.8
1.5
8.2
132
-------
SUBCATEGORY B
Process
Ethylene Dichloride Direct Chlorination of Ethylene
The direct Chlorination of ethylene is carried out in the presence of a
ferric chloride catalyst suspended in liquid ethylene dichloride,
C2H4 + C12 —+• C1CH2 CHC1
Ethylene Chlorine Ethylene Dichloride
The gas stream from the reactor is passed through a caustic scrubber,
where the unreacted gases and a trace amount of hydrogen chloride are
removed by a caustic solution. The liquid stream from the reactor is
first sent to a distillation column to remove heavy ends and then to a
wash tower, where a caustic solution is used to remove some impurities.
The crude product is finally discharged to a distillation column for
purification. A process flow sheet is shown in Figure IV-15.
There are two waste streams in this process. One is liquid effluent
from the scrubber and the other is the waste water from the wash tower.
The results of a survey at one plant are shown in the following
tabulation:
Flow 96 gallons/1,000 Ib
COD 6,050 mg/1
U.8U lb/1,000 Ib
BOD5 Inhibitory
TOC 1,106 mg/1
0.89 lb/1,000 Ib
A surface heat exchanger can be used to condense water vapor in the off-
gas to the scrubber, while the remaining uncondensed gas from the
reactor (which contains primarily unreacted ethylene and chlorine) can
be totally recycled to the reactor. The scrubber and its waste water
can then be eliminated, with this modification, RWL of BADCT and BATEA
for this process can be expected to have low values of 0.072 pounds of
COD and 0.106 pounds of TOC per 1,000 pounds of ethylene dichloride.
Total process water usage of this process is 0.82 pound of water per
pound of ethylene dichloride, and cooling water usage is 93 pounds of
water per pound of product.
133
-------
z
g
u
G
UJ
a
z
z
O
m fi
> Z
* 0
* 9
3 =f
o 5
E >-
x
O
u
o
UJ
o
X
UJ
Z
nus
mis
tosantns
uaaanuas
HOiOVBH
134
-------
An alternate route in manufacturing EDC is oxychlorination of ethylene
with hydrochloric acid and air over a supported copper chloride
catalyst. The characteristic waste water stream from this process will
contain most of the same impurities found in the direct chlorination
process.
Ethylene dichloride has moved from fifth to third place in consumption
of ethylene in the last decade. This growth has been at the expense of
acetylene. The common point of intersection is vinyl chloride, which
accounts for 75% of ethylene dichloride usage. Ethylene dichloride
production has grown more than four-fold since 1961 with a concomitant
decline in price to about 3t per pound. The U.S. ethylene dichloride
capacity and estimated economics of EDC are presented in Tables IV-22
and IV-23, respectively.
135
-------
Table IV-22
U.S. Ethylene Dichloride Capacity (1972)
Company
Allied
American Chemical
Continental Oil
Diamond Shamrock
Dow
Ethyl Corp.
B.F. Goodrich
PPG
Shell
Union Carbide
Vulcan
(Baton Rouge, La.)
(Long Beach, Calif.)
(Lake Charles, La.)
(Deer Park, Texas)
(Freeport, Texas)
(Plaquemine, La.)
(Baton Rouge, La.)
(Houston, Texas)
(Calvert City, Ky.)
(Lake Charles, La.)
(Guayanilla, P.R.)
(Deer Park, Texas)
(S. Charleston, W. Va.)
(Texa s City, Texa s)
(Geismar, La.)
MM Ib.
Total
Source: Oil. Paint, and Drug Reporter. Sept. 20, 1971.
Table IV-23
Estimated Economics for Ethylene Dichloride
(100. MM Ib. plant)
Total Fixed Capita1=$1.0 MM
Estimated Operation Cost
Cost.
Ethylene
Chlorine
Utilities
Labor and overhead
Capital charges
Total
-------
CATEGORY.-!
Product .Process
Vinyl Chloride Cracking of Ethylene Dichloride
Recent developments in ethylene technology, coupled with the low cost
and ready availability of ethylene, dictate ethylene as feedstock in all
new vinyl chloride plants. Vinyl chloride monomer is produced by
cracking purified Ethylene Dichloride (EDC) in a pyrolysis furnace as
follows:
C2H4C12 —* C2H3C1 + HCl
EDC Vinyl Chloride Hydrochloric Acid
After quenching by direct contact cooling, the furnace products are
separated into HCl and high-purity vinyl chloride monomer. The liquid
streams from the quencher are fractionated to separate the vinyl
chloride product from unreacted EDC, which is then recycled. A flow
sheet for this process is shown in Figure IV-16.
The major waste water sources are the effluents from scrubbing systems
required for hydrogen chloride removal, recycle purification of EDC, and
the effluent from associated aqueous acid by-product production units.
The survey data for one plant are presented in the following tabulation.
FLOW 336 gallons/1,000 Ib
COD 2,733 mg/1
7.661 lb/1,000 Ib
BOD5 Not available
TOC 120 mg/1
0.33 lb/1,000 Ib
A large fraction of the RWL shown above is contributed by the aqueous
acid production unit. If the by-product were left in an anhydrous form,
the anhydrous acid by-product could actually replace the aqueous acid
by-product. The RWL of this process will be reduced to 85 gallons of
flow per 1,000 Ib of product, 0.203 Ib COD/1,000 Ib, and 0.054 Ib
TOC/1,000 Ib; this level of RWL will be considered as the standard of
BADCT and BATEA control technology for vinyl chloride manufactured by
EDC cracking.
Total process water usage in existing processes is 2.80 pounds per pound
of vinyl chloride, and cooling water usage amounts to 3,464 pounds per
pound of product.
137
-------
•o
I
5
o
9
O
_j
x
u
UJ
Z
UJ
O
Z
*
u
2
u
_J
D
s
o
X
u
C-5
>-
CJ
CO CJ
CO i
138
-------
An alternate route in manufacturing of vinyl chloride is the classical
acetylene addition reaction. This has been covered under the discussion
in Category A.
Table IV-24 presents the U.S. vinyl chloride capacity, and Table IV-25
estimated economics for various processes.
Table IV-24
U.S. Vinyl chloride capacity
(MM Ib)
Company
Allied Chemical (Moundsvi1le, W. Va.)
(Geismar, La.)
American Chemical (Long Beach, Calif.)
Continental Oil (Lake Charles, La.)
Cumberland Chemical (Calvert City, Ky.)
Diamond Shamrock (Deer Park, Tx.)
Dow Chemical (Freeport, Tx.)
(Plaquemine, La.)
Ethyl Corp. (Baton Rouge, La.)
(Houston, Tx.)
General Tire (Ashtabula, Ohio)
B. F. Goodrich (Calvert City, Ky.)
(Niagara Falls, N.Y.)
Goodyear (Niagara Falls, N.Y.)
Monochem (Geismar, La.)
PPG (Lake Charles, La.)
PPG-Corco (Puerto Rico)
Shell (Deer Park, Tx.)
Tenneco (Houston, Tx.)
Union Carbide (S. Charleston, W. Va.)
(Texas City, Tx.)
Totals
1967 1969 1972
Process
100
-
170
-
60
100
200
250
270
150
75
koo
kQ
70
250
-
-
-
200
120
230
2,685
_
300
170
600
-
100
200
300
270
150
-
400
-
-
250
300
-
-
200
120
230
3,590
_
550
170
600
-
-
525
575
270
150
-
koo
-
-
250
300
500
700
200
120
-
5,310
Acetylene
Ethyl ene
Ethylene
Ethylene
Acetyl ene
Acetylene
Ethylene
Ethylene
Ethylene
Ethylene
Acetylene
Ethyl ene
Acetyl ene
Acetyl ene
Acetylene
Ethylene
Ethyl ene
Ethylene
Acetylene
Ethylene &
Acetylene
1
Based on Oil. Paint & Drug Reporter. March 17, 19&9.
139
-------
Table IV-25
Estimated vinyl chloride economics
(500-MM-lb plant; 1972 construction)
Total fixed capital
Process
Ethylene oxychlorination
Acetylene
Ethane (transcat)
Production cost
Process: Ethylene
Raw materials
Ethane (0.59 lb/lb at 0.9C/lb)
Ethylene (0.^9 lb/lb at 3.0e/lb)
Chlorine (0.67 lb/lb at 2.5
-------
SUBCATEGORY B
Product __________ Process ____________
Styrene Dehydrogenation of Ethyl Benzene
Styrene is produced by vapor-phase dehydrogenation of ethyl benzene over
supported zinc oxide, magnesium oxide, and iron oxide catalysts. Steam
is used as the diluent.
Cj6H5 C2H5— *C6H5 C2H3 + H2
Ethyl Benzene Styrene Hydrogen
A flow sheet for styrene via the dehydrogenation of ethyl benzene is
shown in Figure IV-17. Feedstock ethyl benzene and superheated steam
are mixed in a dehydrogenation reactor. After being condensed, the
reactor effluent goes to a separator, where three phases are formed.
The uncondensed gases are passed through a scrubber where organic vapors
are removed by the scrubbing water. The water phase is removed from the
separator and discharged from the system, and the organic dehydrogenated
mixture passes to the distillation section.
Since the dehydrogenation reaction operates at about 60% ethyl
benzeneconversion, it is necessary to fractionate the process unreacted
ethyl benzene for recycle. Styrene will polymerize at temperatures
approaching its normal boiling point; therefore, it is necessary to
operate the styrene ethyl benzene distillation under vacuum to prevent
styrene loss due to polymerization.
The draw-offs from separator and scrubber are two of the three major
waste water pollution sources in the process. The other source is a
steam-ejector system used to produce vacuums for distillation columns.
The survey data derived from plant visits are summarized as follows:
Flow 2,810 gallons/1,000 Ib 657 gallons/1,000 Ib
COD 219 mg/1 426 mg/1
5.13 lb/1,000 Ib 2.34 lb/1,000 Ib
BODS 69 mg/1 70 mg/1
1.62 lb/1,000 Ib 0.381 lb/1,000 Ib
TOC 22 mg/1 22 mg/1
0.53 lb/1,000 Ib 0.12 lb/1,000 Ib
141
-------
N
z
LU
CO
7
o o
= 2
O
X
142
-------
The smaller amount of waste water in Plant 2 is attributed to its use of
steam jets with surface heat exchangers in contrast to the steam jets
with barometric condensers used in Plant 1, and also to effective
operation of the scrubber system. Use of untreated river water as
quenching water for the barometric condensers at Plant 1 introduces some
contaminants into the waste water stream. Plant 2 discharges unccnden-
sible vapors (consisting of some organic contaminants) from surface heat
exchangers into the atmosphere.
To achieve BADCT and BATEA control technology, the steam jets (with
eirher surface or barometric condensers) should be replaced by vacuum
pumps. RWL for BAECT and BATEA can then be expected to be lower than
that represented by Plant 2.
An example based on a 5 x 108 Ib per year styrene plant has been devised
for illustrating the advantages of vacuum pumps over steam jets. A
description is given in the following paragraphs.
A two-stage steam ejector system is currently used to obtain the vacuum
in the distillation section. The ejector system illustrated uses
surface exchangers for both inter and after condensers. A schematic
flow sheet, depicting steam and effluent flow rates and effluent
composition, is presented in Figure IV-18. The effluent steam from the
ejectors contains a fair amount of organics and represents a source of
pollution. The cost of operating the two-stage ejector system is
presented in Table IV-26. Some producers reportedly fractionate the
ejector effluent srream and recycle the organics back to the process.
However, it is not known if this technique is widespread or successful.
Note that the use of barometric condensers will result in an excessively
large effluent stream.
The vacuum pump most suitable for this application is a two-stage unit
which uses a rotating mass of liquid to draw the vacuum. In this case,
the compressant liquid would be essentially ethyl benzene. Most of the
organics in the inlet vapor stream from the tower condense in the
compressant fluid and can be recycled back to the process. Process flow
sheets showing the use of vacuum pumps are presented in Figure IV-19.
The amount of organic substances actually leaving the vacuum system in
the exhaust air is extremely small and is itemized in Table IV-"27. The
amount shown in this table as recycled is actually discharged from the
system via the steam ejector system. The operating costs of using
vacuum pumps are summarized in Table IV-28.
It is evident that a two-stage, liquid-sealed vacuum pump is more
economical than a two-stage steam ejector using surface condensers. The
economic advantage is due to the extremely low loss of ethyl benzene and
styrene in the exhaust stream from the vacuum pumps. In other words,
this modification not only has an economic advantage, but also reduces
143
-------
144
-------
t/3
Q.
Z)
a.
D
U
D
Z)
UJ I-
»Z
u. )
Q
UJ
T.
UJ
M
"Z.
UJ
CO
X
t—
UJ
Z
UJ
145
-------
Table IV-26
Operating Cost of Two-Stage Steam Ejectors
Styrene-Ethyl Benzene Distillation
500 MM Ibs/yr Styrene, 8,200 hrs/yr
Two-Tower System
Investment, $ (for ejectors etc.)
Utilities
Steam, x 55$/M Ib
Cooling Water, 2.5C/M gal
AT-20°F
Total Utilities, $/Yr
Investment Related
Maintenance Material and Labor,
2% of Investment
Plant Overhead, 65% of Maintenance
Insurance and Taxes, 1.5% of
Investment
Depreciation, 10% of Investment
Total Investment Related Expenses, $/Yr.
Prduct Losses
Styrena, 7.0
Ethy1b*nzene, 3.5
Total Preset Losses, $/Yr
Total Operating Costs, $/Yr
Tower No. 1
10,000
Lb/Hr $/Yr
1,330 6,000
GPM $/Yr
150 T73oo
7,800
$/Yr
200
130
150
1.000
1.480
Lb/Hr $/Yr
13 7,500
340 97f600
105,100
m.380
Tower No. 2
77*00
Lb/Hr $/Yr
790 37560
GPM $/Yr
5§ 1.100
4,700
$/Yr
150
100
110
740
1,100
Ub/Hr $/Yr
95 54,500
41 ll
66,300
72.100
186,480
Total Operating Costs, t/lb styrene produced
0.037
146
-------
Table IV-27
Organ!cs in Exhaust Air From Vacuum Pumps
500 MM Lbs/yr Styrene-8,200 hrs/yr
(Ibs/hr)
Two-Tower System
Tower No. 1 Tower No. 2
Styrene
In 13 95
Out In Exhaust _jB _k
Amount Recycled 5 91
Ethyl benzene
In 3^0 k]
Out In Exhaust 11
Amount Recycled 329
147
-------
Table IV-28
Operating Costs For Vacuum Pumps*
Styrene-Ethyl Benzene Fractionation
500 MM Lbs/yr Styrene, 8,200 hrs/yr.
Investment, $ (vacuum pumps etc.)
Utilities
Power, 0.800 e/kwh
Cooling Water, 2.5
AT-20 F.
Total Utilities, $/yr
Investment Related Expenses
Maintenance Materials and Labor,
4% of Investment
Plant Overhead, 65% of Maintenance
Insurance, Taxes, 1.5% of Investment
Depreciation, 10% of Investment
Total Investment Related Expenses, $/yr
Product Losses
Two Tower System
Tower No. 1
30,000
kwh $/yr
51 3,350
GPM $/yr
13 160
3,510
$/vr
1,200
780
450
3.000
5,430
Tower No. 2
23,000
kwh $/yr
29 1,900
GPM $/vr
7 90
1,990
$/vr
920
690
920
2.300
4,830
Lbs/yr $/vr Lbs/vr $/vr
Styrene, 7C/lb.
Ethylbenzene, 3.5$/lb.
Total Product Losses, $/yr
Total Operating Costs, $/yr
Total Operating Cost, $/lb styrene product
*Per letter from Nash Engineering Co. of 5-29-73 to Chem Systems
8 4,600
11 3.160
7,760
16.700
4
5
2,300
1.440
3,740
10.560
27,260
0.005
148
-------
the RWL of the process. Styrene is used exclusively for homo-, co-, and
terpolymers and is produced on the Gulf Coast. Production capacity has
grown rapidly to accommodate demand. Installed styrene capacity is
presented in Table IV-29, and estimated economics for a competitive 5 x
108 Ib plant are shown in Table IV-30.
Amoco
Cosden
Cos-Mar
Dow
El Paso
Foster-Grant
Gulf Oil
Ma rbon
Monsanto
Shell
Sinclai r-Koppers
Sun Oil
Union Carbide
Table IV-29
U.S. Styrene Capacity
(MM Ib)
Company
(Texas City, Texas)
(Big Spring, Texas)
(Carville, La.)
(Freeport, Texas)
(Midland, Mich.)
(Odessa, Texas)
(Baton Rouge, La.)
(Dojialdsvi lie, La.)
(Baytown, Texas)
(Texas City, Texas)
(Torrance, Calif.)
(Houston, Texas)
(Kobuta, Pa.)
(Corpus Christi, Texas)
(Sea Drift, Texas)
(Institute, W. Va.)
1
1
1967'
1970
1972
Total
300
100
-
500
300
85
200
-
125
650
210
70
200
60
300
110
3,210
800
100
500
650
350
120
250
-
135
800
240
110
430
80
300
shut down
4,865
800
100
500
650
350
120
250
500
shut down
1,300Z
240
110
430
80
300
shut down
5,730
1
-Oil. Paint & Drug Reoortgr. July 7, 19&9 and earlier profiles.
New plant that replaced 800 MM-lb unit.
149
-------
Table IV-30
Estimated Economics For Styrene
(500 MM-lb plant; 1972 construction period)
A. Total fixed capital=$35.0 MM
B. Production costs
C/lb styrene
2
Raw materials 3.95
Labor 0.13
Utilities 0.91
Maintenance 0.34
(6% ISBL + 3% OSBL)
Overhead 0.56
(kS% maint + labor)
Taxes 0.10
(1.5% of invest)
Depreciation (10 yr) O.JO
Total 6.69
.Denydrogenation process.
1.10 Ib ethybenzene at 3.50«j/lb + catalyst and chemicals.
150
-------
SUBCATEGORY B
Product Process
Methyl Amines Synthesis of Methanol and Ammonia
Methyl amines are synthesized by methanol and ammonia in the presence of
catalyst to form a mixture of mono-, di-, and trimethylamine.
CH30H + NH3 _*. CH3NH2 + H20
Methanol Ammonia Monomethylamine Water
2C30H + NH3 _ » (CH3) 2NH + H2O
Dimethylamine
3CH30H + NH3 — *• (CH3) 3N + H2O
Trimethylamine
Reactants are first preheated by the converter effluent, thereby
recovering some of the exothermic reaction heat. The product stream is
then flashed to remove the noncondensibles and is sent to the recovery
system. First, ammonia is taken overhead and recycled, together with
some trimethylamine. Next, water is added to break the TMA- Ammonia
azeotrope, and pure TMA is taken overhead from a distillation column.
The mixture of mono- and dimethylamine is first dehydrated and then
fractionated to separate DMA and MMA. The ratios of three amines can be
varied by changing reaction conditions. The process flow diagram is
shown in Figure IV-20.
This process uses water to scrub ammonia from all off-gases. The liquid
effluent from the absorber is then flashed to recover ammonia. The
major waste water source, containing a significant amount of
unrecoverable ammonia, is the bottoms from the flash column. The other
two waste water streams are the bottoms from the separation
fractionators. The characteristics of the waste water are summarized in
the following tabulation.
NOj^l Sample No. _ 2
Flow 429 gallons/1,000 Ib 429 gallons/1,000 Ib
COD 6,303 mg/1 1,178 mg/1
22.56 lb/1,000 Ib 4.21 lb/1,000 Ib
BOD5 99 mg/1 174 mg/1
151
-------
152
-------
0.351 lb/1,000 lb 0.62 lb/1,000 Ib
TOC 11,634 mg/1 3,808 mg/1
41.65 lb/1,000 lb 13.63 lb/1,000 lb
The above data show significant, variation. The extraordinarily high
ratio of COD/BOD5 is due to the ammonia contaminant which contributes to
the measurement of COD but not to that of BOD5. It is believed that
Sample I was taken under the upset operating condition of the ammonia
flash column.
Total process water usage, including steam directly supplied to the
process, is 3.1 pounds water per pound of methylamines, while cooling
water usage amounts to 16,700 pounds water per pound of product.
Minor process modifications such as reusing waste waters from
fractionators as ammonia absorption water can reduce the amount of waste
water. The ammonia content in the waste water can be treated only by
end-of-pipe treatment.
Investment for a methylamines plant depends somewhat on the intended
product mixture; a unit to produce 10 million pounds per year costs
around $1.5 million. A summary of U.S. production capacity and
estimated production costs for dimethylamine are presented in Tables IV-
31 and IV-32.
153
-------
Table IV-31
U.S. Methyl Amines Capacity (1970)
Company Location Capacity
MM Ibs.
Commercial Solvents Terre Haute, Ind. 18
DuPont Belle, W. Ma. 75
Strang, Texas 26
Escambia Pace, Fla. 50
GAP Calvert City, Ky. 10
Pennwalt Wyandotte, Mich. 10
TOTAL 189
Table IV-32
Estimated Economics for Methylamines
(10 MM Ib. Plant)
Total Fixed Capital =$1.5 MM
Estimated Production Cost
Cost
. DMA
Methanol (captive, 3.0Vlb.) k.6
Ammonia (merchant, 4.0fc/lb.) 1.6
Utilities 1.5
Labor and Overhead 1.2
Capital charges 5.0
Total 13.9
154
-------
SUBCATEGORY_B
Product
Vinyl Acetate Synthesis with Ethylene and Acetic Acid
Fresh ethylene, oxygen, and acetic acid are combined with their respect-
tive recycle streams, and then are vaporized and fed to a fixed-bed
reactor. Typical operating conditions are 5 psig and 250°C. Conversion
per pass is about 5 percent, with very high (99 percent) selectivity.
The catalyst is usually a mixture of palladium, copper, and iron
chloride on alummina. The acetic acid-to-water mole ratio in the
reactor is kept at about 40:1 to suppress acetaldehyde formation.
Reactor effluent vapor is partially condensed to recover some of the
acetic acid for recycle. Further cooling and fractionation separate a
crude product stream from ethylene, which is recycled to the reactor.
The crude product stream is then fed to a series of fractionators for
further removal of acetic acid and light ends. Hydroquinone is usually
added as a polymerization inhibitor before vinyl acetate is sent, to
storage. The process flow diagram is shown in Figure IV-21.
Since the process is a vapor-phase reaction, waste water is minimal.
The major waste water stream is generated as bottoms from one of the
fractionators. The light ends and heavy ends separated out are either
recycled, sold, or disposed of by incineration.
Results of survey data are summarized in the following tabulation:
Flow 28 gallons/1,000 Ib
COD 516 mg/1
0.13 lb/1,000 Ib
BOD5 150 mg/1
0.04 lb/1,000 Ib
TOG 220 mg/1
0.25 lb/1,000 Ib
This level of RWL can be considered as standards for BADCT and BATEA
control technology for this process.
The classical alternate route in manufacturing of vinyl acetate is the
simple vapor-phase reaction of acetylene and acetic acid in the presence
of a zinc acetate catalyst on a carbon support. Acetylene conversion is
about 60 percent per pass at high (96 percent) selectivity.
A third route is by liquid-phase synthesis if ethylene and acetic acid.
The reaction is carried out in a palladium chloride solution at 450 psig
and 250°C. Conversion per pass is about 5 percent with 97-98 percent
slectivity. Acetaldehyde co-product yield is controlled by suitable
155
-------
o
u
u
»—
01
U
O
Z
CN UJ
i
O »"
NN0103
NO Him us io
NNR103
NOI1V11I1SIQ
01
<
z
cj
CJ
I
156
-------
adjustment of the water content, and this co-product is oxidized in-situ
to form acetic acid, which is used for the main reaction. The
literature indicates that this route produces the best economics.
The U.S. vinyl acetate capacity and comparative economics of the
acetylene and ethylene processes are presented in Tables IV-33 and IV-
34.
Table IV-33
U.S. Vinyl Acetate Capacity
Producer
Air Products
Border Chemical
Celanese Chemical
OuPont Company
Monsanto Company
National Starch
Union Carbide
U.S. Industrial Chemical
Total
% acetylene
Locati on
Calvert City, Texas
Gei smar , La .
Geismar, La.
Bay City, Texas
Pampa , Texas
Clear Lake, Texas
Niagara Fal Is, N.Y.
La Porte, Texas
Texas City, Texas
Long Mott, Texas
S. Charleston, W.Va.
Texas City, Texas
La Porte, Texas
196?
MM Ib
95
90
100
65
75
65
50
55
Tt5
_
7**0
78
1969
MM Ib
95
115
100
65
75
65
50
55
195
_
815
80
1970
MM Ib
95
115
75
100
200
75
80
60
55
300
300
1,^55
59
1972
MM Ib
115
75
200
kOQ
-
60
300
300
M50
38
Process
Acety lene
Acetylene
Acety lene
Ethylene
Acetaldehyde-
acetic anhydride
Ethylene
Acetylene
Ethylene
Acetylene
Acetylene
Acetylene
Acetylene
Ethylene
Source: Qi1. Paint & Drug Reporter Profile, Jan. 1, 1970 and other trade publication*
157
-------
Table
Comparative Vinyl Acetate Economics
(300-MM-lb plants; 1972 construction period)
Estimated Total Investment Cost
Acetylene process = $12.6 MM
Ethylene process (gas phase) = $17.3 MM
Raw materials
Acetic acid (6.0e/lb)
Ethylene (3.0$/lb)
Acetylene (8.0^/lb)
Catalyst and chemicals
Total materials
Labor
Utilities
Ma Int. (6% ISBL + 3% OSBL)
Overhead (1*5% of ma Int. + labor)
Taxes and ins. (1.5% of investment)
Depreciation (10 years)
Total
Estimated Production Costs
C/lb vinyl acetate
Acetylene Route Ethylene Route
*K31 k.23
1.02
2.56
0.32
7.19
0.17
0.29
0.20
0.17
0.07
QM
8.51
-
0*22
5.5**
0.19
0.70
0.27
0.21
0.09
0^52
7.57
158
-------
SUBCATEGORY_C
Product
Phenol 1. Cumene Oxidation and Cleavage
2. Chlorobenzene Process
1. Cumene Oxidation and Cleavage
The cumene process is currently the most popular route and the one upon
which most expansions will be based. The manufacture of phenol from
cumene is carried out by a process involving the following basic steps:
a. Oxidation of cumene with air to form cumene hydroperoxide.
C6H5CH (CH3) 2O2 —*• C6H5C (CH3)200H
Cumene Oxygen Cumene Hydroperoxide
b. Cleavage of cumene hydroperoxide to form phenol and acetone.
C6H5C (CH3) 200H —* C6H5OH + CH3 COCH3
Cumene Hydroperoxide Phenol Acetone
A process flow sheet is shown in Figure IV-22. Cumene and air are fed
to a liquid-phase reactor, operating at 25-50 psig and 130-140°C, in the
presence of a small amount of alkali, to produce the hydroperoxide
intermediate. Reactor liquid effluent is fed to a fractionating tower,
where unreacted cumene is recovered and recycled to the reactor.
Cumene hydroperoxide from the fractionator is fed to a hydrolysis
reactor where the cumene hydroperoxide is cleaved to phenol and acetone
with the aid of a sulfuric acid catalyst. Typical operating conditions
are 5 psig and 150-200°F, and conversion is essentially complete, with
minimal formation of undesired by-products. The crude phenol-acetone
mixture if passed through an ion exchange system and then fed to a
series of tower fractionation trains, where pure phenol and co-produced
acetone are separated from light and heavy ends and other by-products.
2. Chlorobenzene Process
The process flow diagram of Chlorobenzene process is shown in Figure IV-
23, and the basic reactions are summarized below:
C6 H5C1 + 2NaOH (Excess) —*• C6H5 ONa + NaCl + H2O
Chlorobenzene Sodium Hydroxide Sodium sodium Water
Phenate Chloride
159
-------
CN
III <
s >
o x
Z O
z
LU
X
Q.
160
-------
LU
z
UJ
M
Z
LU
CO
o
ex.
o
CO -J
ON I
>6
uj Z
§i
o *
O
Z
NNH10G
Hoiivnusio
1
t
^
MM
-vzn
i
'
i
vt
i
10N3Hd
'
1NV33Q
NNV1
NOI1
t
k.
11 '^
L
1
F
CJ
^
1
4
wuimnn iji
nnn i u u i»i
-iiisio ;
^ / / j
/, /
(31VN3Hd S
ON3Hd 313
\ r
NOI1V1
CJ
CO
H010V3H
I
N08H(fO 031U 119V
MOIiV
-NiaOIHG
I
1
n3aanuGs
* ^
C3
LU C3
Z oc
Os!
CLD
CJ3
\
O
t**l
LU
CO
CV
CO
161
-------
C6H 50Na +HC1 "* C6H50H + NaCl
Sodium Hydrochloric Phenol Sodium
Phenate Acid Chloride
The feed materials (chlorobenzene and excess caustic solution) are fed
into a liquid- phase reactor, and the effluent is discharged into a de-
canter. The upper layer of unreacted chlorobenzene is recycled back to
the reactor. The bottom layer of sodium phenate is neutralized to pro-
duce a mixture of phenol and brine; this mixture is then decanted. The
upper layer is sent to a fractionator , where pure phenol is obtained,
and the bottom brine stream is passed through an activated carbon bed to
remove the reamining phenol, which is eventually recycled back to the
reactor.
The chlorobenzene process is used by only one company in the U.S. The
major waste water source in this process is the brine solution from the
second decanter, which is contaminated with phenol and acetic acid.
However, an activated carbon system and chlorination reactor, both being
considered as parts of an integral system of the process, are used to
remove phenol by adsorption and to destroy the acetic acid component.
The effluent from the system is totally recycled for chlorine
production. The adsorbed phenol is desorbed with caustic solution to
form sodium phenate, which is recycled back to the reactor. Therefore,
the process is free of discharge and can be considered as a standard for
BADCT and BATEA.
The cumene oxidation process recycles the water present in the hydro-
peroxide reactor. Water from the dilute sulfuric acid in the cleavage
reactor is also recycled. The only significant waste water stream is
generated by water scrubbing the vapor effluent from the cleavage
reactor; this stream contains dissolved sulfuric acid, sulfates, and
oxygenated organic compounds.
The major paramters of surveyed RWL data from two cumerie oxidation
plants are summarized in the following tabulation. The results of the
analyses also show that phenol and oil contaminants in waste waters from
both plants are in excess of general discharge criteria for biological
treatment processes and would interfere with the normal functioning of
such processes.
Flow 279.6 gallons/1,000 Ib 164 gallons/1,000 Ib
COD 4,770 mg/1 84,304 mg/1
11.1 lb/1,000 Ib 11.5 lb/1,000 Ib
162
-------
BOD5 2,410 mg/1 17,575 mg/1
5.6 lb/1,000 Ib 24 lb/1,000 Ib
TOC 194 mg/1 77,406 mg/1
0.45 lb/1,000 Ib 105.6 lb/1,000 Ib
The survey data show a significant difference in RWL between two plants.
The lower RWL of Plant 1 is attributed to the installation of
dnphenolizer facilities (steam stripper). These facilities are
considered as part of the process rather than end-of-pipe treatment,
since phenol is recovered at this unit and recycled back to the
oxidation reactor. The higher RWL of Plant 2 is attributed mainly to
tie disposal of concentrated light ends and heavy ends from acetone an3
phenol fractionators into the sewers, instead of by incineration as
commonly practiced. RWL represented by Plant 1 can be logically
considered as standard for BPCTCA control technology.
The activated carbon system mentioned in the chlorobenzene process has
been claimed to be effective in reducing phenol concentration from about
100 mg/1 down to 1 mg/1. The saturated activated carbon beds can be re-
generated with caustic solution by desorbing phenol into phenate salt.
The salt is then recycled to the oxidation reactor. With this system,
phenol is recovered for reuse, and the RWL of the process is reduced as
well. Consequently, BADCT and BATEA should require a steam
stripper/dephenolizer with an activated carbon system to achieve a low
RWL standard.
Gross cooling water usages for the two processes discussed above differ
greatly: 3.85 and 463 pounds of water per pound of phenol,
respectively, for the clorobenzene and cumene processes.
Several other process routes in manufacturing of phenol are currently
practiced. These include the Hooker-Raschig process, toluene oxidation,
and sulfonation. Again, the cumene route is by far the most important,
and it is predicted that all phenol capacity installed over the next ten
years will be based on this process. The current U.S. phenol production
capacity and its estimated economics are presented in Tables IV-35 and
IV-36.
The Hooker-Raschig and sulfonation processes are briefly described in
the following paragraphs.
The Hooker-Raschig process is a two-step, vapor-phase reaction. A
benzene chlorination reaction is carried out at 400°F with air, over a
copper and iron chloride catalyst. The copper-iron catalyst oxidizes
the hydrogen chloride to chlorine and water. The chlorine attacks the
benzene ring to yield chlorobenzene and additional hydrogen chloride.
The chlorobenzene is then hydrolyzed over silica at 900°F to yield
phenol and hydrogen chloride. There is no net production of hydrogen
163
-------
chloride since it is continually convereted to usable chlorine. The net
products are, therefore, phenol and water.
The sulfonation process is a liquid-phase reaction. Benzene is first
reacted with sulfuric acid to produce benzenesulfonic acid, which is
then converted to phenol by caustic fusion. The sulfuric acid employed
in this process is totally lost.
Table IV-35
U.S. Phenol Capacity*
Producer
Plant Location Estimated Capacity Process Route
Allied
Chevron
Clark Oil
Dow
Hercules
Hooker
Monsanto
Reichold
Shell
Skelly
Union Carbide
Frankford, Pa.
Richmond, Cal.
Blue Island, 111.
Kalama, Wash.
Midland, Mich.
Gibbstown, N,J.
N. Tonawanda, N.Y.
S. Shore, Ky.
Alvin, Texas
Monsanto, Ml.**
Tuscaloosa, Ala.
Houston, Texas
El Dorado, Kansas
Bound Brook, N.J.
Marietta, Ohio
MM Ibs/yr
k20
50
70
40
230
100
65
65
375
115
90
50
50
150
125
Cumene
Cumene
Cumene
Toluene oxidation
Chlorobenzene
Cumene
Raschi g
Raschi g
Cumene
Sulfonation
Sulfonation
Cumene
Cumene
Cumene
Raschig
Natural phenol produced
TOTAL
2,08
*As of mid-1970. Estimated based on trade literature.
**Reported shut down.
164
-------
Table IV-36
Estimated Economics for Phenol Production
(400-MM-lb plant; 1972 construction)
Raw materials
Labor
Utilities
Mai ntenance
(6% ISBL + 3% OSBL)
Overhead
Taxes and insurance
(1.5% of investment)
Depreciation (10%)
4
By-product credit
FIXED INVESTMENT COSTS
Process
Cumene
Tol uene
Raschig
PRODUCTI
BL)
labor)
lent)
TOTAL
NET
$MH
26.6
30.0
36.1
ON COSTS
Cumene Toluene
C/lb C/lb
5.811 3.452
0.29 0.29
0.92 0.71
0.32 0.36
0.27 0.30
0.10 0.11
0.67 0.76
8.38 5.98
2.7** -
5.64 5.98
Raschig
C/lb
3.673
0.29
0.78
0.43
0.32
0.13
0.91
6.53
- _ - -..
6.53
1.45 lb cumene/lb at 3.7^/lb + catalyst and chemicals.
Includes 1.3 lb toluene at 2.5C/1b.
0.94 lb benzene/lb at 3.4
-------
SUBCATEGORY C
Product Process
Oxo Chemicals Carbonylation and condensation
The oxo process is a broadly applicable technology which is used to pro-
duce aldehydes which are usually converted to the corresponding
alcohols. The process is used on a number of feedstocks, the two most
important being propylene and alpha olefins, to produce linear alcohols
for plasticizers and surfactant usage.
2-ethylhexanol, produced primarily from propylene via n-butyraldehyde,
is the most important oxo chemical in terms of volume. A process
flowsheet describing the manufacture of 2-ethylhexanol is shown in
Figure IV-24 and the basic chemical reactions are given below:
CH3
C3H6 + CO + H2 •"*• CH3CH2CH2CHO + CH3 CHCHO
Propylene Carbon Hydrogen n-butyraldehyde iso-butyraldehyde
Monoxide
2CH3 CH2 CH2 CHO —^ CH3 CH2 CH2 CH=C-CHO + H2O
CH2H5
n-butyraldehyde 2-ethylhexanal Water
H
i
CH3 CH2 CH2 CH=C-CHO + H2 —+• CH3 CH2 CH2-CH2-C-CH20H
r u - - -CH2H5~
C2H5
2-ethylhexenal Hydrogen 2-ethylhenanol
Carbon dioxide, natural gas, and steam are passed into a synthesis gas
reactor to produce water gas (1:1 ratio of H20 and CO) which is then
mixed with propylene in a liquid-phase reactor in the presence of a
cobalt solution. The reaction is carried out under pressure and the
reactor is maintained approximately isothermal, A liquid-gas mixture of
aldehydes and unreacted materials is taken overhead from the reactor,
cooled, and then separated in successive high- and low-pressure flashing
stages, whence unreacted synthesis gas is recycled to the oxo reactor.
The catalyst cobalt is then removed continuously from the liquid phase.
The liquid product, containing n-butyraldehyde, iso-butyrcildehyde, and
solvent, is separated in two distillation columns.
N-butyraldehyde is then sent to a condensation reactor, where the subse-
quent reaction is carried out at moderate temperature and atmospheric
pressure in the presence of strong base such as sodium or potassium hy-
166
-------
— I CD
fc O3
o or »
U- U_J A
0 :*• T
a: CD I
CD- CD 1
>r- |
*
O^l
to
^r <
CN (J
> ^
114 -L.
=> u
"" X
O
CXI
CO
CD
a
or
T
4
1 d ^
CO ^t
u. i
4k
d
H-
co
4 k
^
^
t— >.
CO QC "
— > ^
^ R
«X UJ
cj a:
-k
ac
CD
4 ff u *
* 2 *
— I LU
^^
Z =" *-
t ^ 2.
ct 3
C3 Q=
U_ C3
S CJ —1
c= *r c_5
>- LU >-
U-l
«. rr-
^
^k
CO
co cac
UJ C3
^ S
^ -I U_J
co CD ae
n ^ 4 >
C-SJ -^T -«C LU
0 3= LU 3E
CJ CO 1— LU
C/0 —1
a.
r
i
LU
>-
5
5
-tr
•^3-
CD
— 1
CJ
CD
UJ
OC
£
1 —
=3
C3
CO
CD
I —
CO
CD
-J co
«= C=
tf 3=-
3 ^
LU
3 5
^ *
LU
*- S -t
j fc
UJ
1 —
g
*• J -»l
-1 £ £
-•a 5
§
i —
3
£3
CD
1 —
CO
5 „ _ ^
k 5 _ . ,
5 LU
-'S S
— i co S
CO
*
k
1 g 0.
L 1 — e=
% S5S
c= Z -I
>- LU U-l
31 CE* QC
t,
r
CD
s
>-
1 2»-
i — or
_, S ^ k. 2
^^ ,! 0 cj ^^ LU
1 O O Ll_ CO
-I CJ — IX -I
=
4 k
-------
droxide. Continuous removal of the water produced during reaction
drives the aldol condensation to completion. The unreacted c°4°
aldehyde is separated from the product 2-^ethylhexenal by distillation
and is recycled to the condensation reactor.
The 2-ethylhexenal produced is then hydrogenated to 2-ethylhexanol in
the presence of a solid nickel catalyst in a pressurized reactor at 50
to 100 atmospheres. After being washed with caustic solution and water,
the reactor effluent is sent to a fractionator to recover the product 2-
ethylhexanol.
The major waste water streams in oxo^chemical manufacturing are the
water removed from the aldol condensation and the water used in washing
the crude product before fractionation into final product. The waste
water may contain some intermediates, product, and by-product losses.
No significant catalyst loss from the reactor is expected. Heavy ends
from various stills are disposed of by incineration.
The characteristics of the waste water obtained from the plant survey
are summarized in the following tabulation. It should also be noted
from the results of analyses that the oil concentration in the waste
stream is beyond the limits of the general discharge criterion for
biological treatment processes.
Flow 420 gallons/1,000 Ib
COD 1,212 mg/1
4.25 lb/1,000 Ib
BODS 900 mg/1
3.15 lb/1,000 Ib
TOC 549 mg/1
1.92 lb/1,000 Ib
Other than reusing the aldol condensation water as wash water, it is
deemed unfeasible to further reduce RWL of the process by any in-process
modification. consequently, RWL presented can be considered as
standards for BADCT and BATEA control technology for this manufacturing
process.
An alternate route in oxo chemical manufacturing is based on a new
catalyst system. By carrying out the hydroformation reaction in an
alkaline medium using phosphine-promoted cobalt carbonyl processes, 2-
ethylhexanol and butanol can be produced directly in one step. Olefin
feed and the recycled catalyst stream are charged to the first of a
series of packed reactors at control rates. Synthesis gas (E2/CO molar
ratio = 2/1) is fed separately to each reactor. The stream taken
overhead from the final reactor is directly sent to the product recovery
column. The bottoms from the product recovery column will contain the
168
-------
catalyst complex dissolved in a mixture of alcohols and heavy ends.
This stream is recycled to the first reactor with periodical purging to
remove the built-up heavy ends.
The U.S. capacity for production of oxo chemicals is presented in Table
IV-37 and the estimated economics for a 40 million pounds-per-year plant
to produce 2-ethylhexanol from propylene is shown in Table IV-1R-
Table IV-37
The U.S. Oxo-Chemicals Capacity
(Mill ions of pounds)
Company Locat ion Capaci ty
Dow Badische Freeport, Texas 200
Eastman Longview, Texas 275
Enjay Baton Rouge, La. 200
Getty-Air Products Delaware City, Del. kO
Oxochem Penuelas, P.R. 250
Shell Geismar, La. 150
Houston, Texas 200
Union Carbide Ponce, P.R. 140
Seadrift, Texas 120
Texas City, Texas 200
USS Chemicals Haverhill, Ohio 70
TOTAL 1,845
Source: Oil, Paint and Drug Reporter, Chemical Profile,
April 1, 1971
Table IV-38
The Estimated Economics for Oxo-Chemicals
(40. MM Ib. 2-ethyIhexanol-from-propylene plant)
Total Fixed Capital=$5.7 MM
Estimated Operation Cost
Cost
C/lb. 2-ethylhexanol
Propylene 2 ]
Synthesis gas j c
Catalyst and chemicals 2.4
Utilities j
Labor and overhead
Capital charges
Total '3-5
169
-------
SUBCATEGORY_C
Product Process
Acetaldehyde Oxidation of"Ethylene (Wacker Process)
The Wacker process employs an aqueous catalyst solution of palladium
chloride, promoted (for metal oxidation) by copper chloride. The chem-
istry involved in the process can be summarized as follows:
C2H4 + 1/2 02 —* CH3CHO + Heat
Ethylene Oxygen or Air Acetaldehyde
The catalyst acts as the oxygen carrier and causes selective conversion
of ethylene to acetaldehyde. The reaction steps essentially are:
Reaction:
C2H4 + 2CUC12 + H20 PdCl CH3CHO •»• 2HC1 * 2CUC1
Ethylene Cupric Water .—+• Acetaldehyde Hydrochloric Cuprous
Chloride Acid Chloride
Regeneration:
2CUC1 * 2HC1 + 1/2 02 —* 2CuC12_ + H2O
Cuprous Hydrochloric Oxygen Cupric Water
Chloride Acid or Air Chloride
There are two basic process variations, and choice depends upon such
factors as oxygen cost, utilities prices, and available ethylene purity.
In the single-stage process, pure oxygen is employed as the oxidant.
The reactor effluent is condensed and water-scrubbed. Unreacted gas is
recycled into the reactor. By-products and water are separated from the
acetaldehyde product by distillation, Both the reaction and
regeneration are effected at the same time.
In the two-stage process, the oxidant is air. The reaction is carried
out with catalyst solution and ethylene in one reactor, and the re-
generation is carried out with air in a separate reactor. Lowerpurity
ethylene can be used with this version of the process. However, this
process forms more by-products and requires high operating pressures.
The process flow sheet for two-stage Wacker process is shown in Figure
IV-25. The major waste water sources in this process are the effluents
from the scrubber that is required for removal of unreacted ethylene and
uncondensed acetaldehyde vapor, and from the aqueous bottoms of the
170
-------
a.
of.
x
IVJ
O
__i
<
114
cs
>-
X
171
-------
acetaldehyde still. The characteristics of the wastewater are shown in
the following tabulation.
Plant_I Eiant_2 Plant_3
Flow 90 gallons/1,000 Ib 61 gallons/1,000 Ib 35 gallons/1,000 Ib
COD 58,718 mg/1 11,400 mg/1 20,240 mg/1
BOD5 3,700 mg/1 11,500 mg/1
TOC 7,000 mg/1 12,500 mg/1
The foregoing data show the same order of magnitude of raw waste loads
in Plants 2 and 3, and this level of RWL can be considered as standard
for BPCTCA. The high RWL of Plant I is mainly due to sloppy operation
of the acetaldehyde still. TO define BADCT and BATEA control
technology, it is required that a steam stripper be installed to recover
and reuse the organic contaminants in the waste water. A description of
the steam stripper, as well as its estimated economics has been given in
the section on aniline.
Because of the aqueous- phase reaction, catalyst metals are present in
the waste water from the acetaldehyde still bottoms as a result of
carry-over from the reactor. The aqueous catalyst solution is also
quite acidic and corrosive. Survey data also shows that, in addition to
metallic contaminants in waste water stream, sulfate and oil
contaminants are found at concentrations in excess of general criteria
for biological treatment processes. Pretreatment or dilution to reduce
their concentrations is required.
Average process water usage for this process, including steam directly
supplied to the process, is 0.92 pounds per pounds of acetaldehyde,
while cooling water usage amounts to 330 pounds per pound of product.
Alternate routes for manufacturing of acetaldehyde as well as U.S.
production capacity have been discussed under Acetaldehyde in
Subcategory B. Estimated economics for production of acetaldehyde by
the ethylene route are shown in Table IV-39.
172
-------
Table IV-39
Estimated Economics for Acetaldehyde
(200 MM-lb plant; 1972 construction)
Fixed Capital Investment
Process $ MM
Ethylene 14.80
Estimated Operation Cost
Raw materials
Utilities
Labor
Maintenance (6% ISBL + 3% OSBL)
Overhead (45% labor and maintenance)
Taxes and insurance (1.5% of investment)
Depreciation (10 years)
TOTAL
Cost
C/lb ethylene
2.^5
0.82
0.24
0.35
0.27
0.11
0.75
4.99
1
Includes 0.68 Ib ethylene/lb at 3.3
-------
SUBCATEGQRY C
Product
Acetic Acid Oxidation of Acetaldehyde
Acetic acid is produced by the liquid-phase oxidation of acetaldehyde,
using either air or oxygen according to the reaction given below:
CH3CHO + 1/2 02 _^ CH3COOH
Acetaldehyde Oxygen Acetic Acid
The reaction is carried out in the liquid phase at 150°F and 60 psig,
with manganese acetate dissolved in aqueous solvent as catalyst.
The process flow sheet is shown in Figure IV-26. Acetaldehyde and
solvent are fed to the oxidation reactors with a mangamese acetate
catalyst solution. The reactor effluent (containing unreacted oxygen,
nitrogen, acetaldehyde, and solvent) is cooled, and the acetaldehyde and
solvent are condensed and recycled back to the reactor. The non-
condensibles are water-washed before being discharged into the
atmosphere. The degassed liquid stream as well as water from the
scrubber are sent to a light-ends column, where the light ends are
distilled overhead. The bottoms from these distillation columns are
sent to a dehydration column in which water is removed overhead using
benzene as the azeotropic agent. The aqueous phase in the distillate
stream is sent to a solvent stripping column, where acetic acid is
removed as distillate while the bottoms are sent to a weiste disposal
unit.
The major waste water source in this process is the water taken overhead
from the dehydration column. The possible contaminants are unrecovered
formic acid and acetic acid. The characteristics of the waste water
obtained from plant surveys is presented in the following tabulation:
Plant_l Plant_2
Flow 500 gallons/1,000 Ib 10.2 gallons/1,000 Ib
COD 186 mg/1 306,100 mg/1
0.78 lb/1,000 Ib 26.18 lb/1,000 Ib
BODS 8U mg/1 64,000 mg/1
0.35 lb/1,000 Ib 5.U4 lb/1,000 Ib
The foregoing data show a significant variation in RWL between two
plants. Examination of each process shows that the concentrated light
174
-------
sO
c*
at
O
z
g
9
x
O
5
UJ
U
<
U
UJ
O
«-Q
175
-------
ends and heavy ends from distillation columns are discharged into sewer
lines by Plant 2 instead of being disposed of by incineration as
commonly practiced. If these concentrated streams are excluded, the RWL
of Plant 2 as shown below is comparable to RWL of Plant 1.
Flow 1.48 gallons/1,000 Ib
COD 7,500 mg/1
0.925 lb/1,000 Ib
BODS 26,700 mg/1
0.33 lb/1,000 Ib
There is a slight difference in manufacturing process between Plants 1
and 2. Plant 1 utilizes ethanol as part of its feedstock and generates
at most 35 gallons of reaction water per 1,000 pounds of product, based
on 100% ethanol feedstock. Also, instead of combining scrubber water
with aqueous reactor effluent, Plant 1 sends scrubber water directly to
an acetaldehyde recovery still and disposes of the bottom stream of the
distillation column. This modification allows Plant 1 to use more
scrubbing water in the scrubber and results in a high amount of waste-
flow.
Based on the foregoing analysis, the RWL of Plant 1 can be considered
the standard of BADCT and BATEA of this process. The standards of BADCT
and BATEA should require recycling of scrubber water in Plant 1. This
modification when implemented will reduce the flow of BPCTCA to one-
tenth its current level, and the RWL by one half.
Total process water usage of this process is directly proportional to
the amounts of waste water generated. The survey data show a variation
from 4.2 pounds of process water per pound of acetic acid at Plant 1 to
0.024 at Plant 2. The gross cooling water usages are 54 and 185 pounds
per pound of product for Plants 1 and 2, respectively.
Several other process routes to acetic acid are also practiced comm-
ercially. The specific processes utilized by each firm with their
respective capacities are presented in Table IV-40. The CO-Methanol and
Petroleum Gases (n-butane) processes are discussed briefly in the
following paragraphs.
Direct liquid-'phase oxidation of n-butane in petroleum gases is normally
carried out at 300-350°F under a pressure of 700-800 psig, and the
chemical reactions taking place are extremely involved. The reactor
effluent is sent to a vapor-liquid separator, the gaseous products from
this separator are scrubbed with a heavy hydrocarbon to recover un-
reacted n-butane, and the liquid product from the separator is split
into an organic and aqueous phase. The organic phase is recycled, while
the aqueous phase is fractionated to remove intermediate by-products.
176
-------
The CO-Methanol process is the most recent commercial route.
monoxide and a liquid stream containing the catalyst system of cobalt
iodide and cobalt carbonyl hydride are fed to a sparged reactor oper-
ating at 500°F and 10,000 psig. Product acetic acid is recovered by
fractionation. The methanol feedstock is normally not introduced
directly to the oxidizer, but rather is used to scrub the reactor off-
gases, which contain catalyst in the form of methyl iodide vapor.
Table IV-40
Acetic Acid Capacity (1972)
Producer
Borden
Celanese
Eastman
FMC
Hercules
Monsanto
Publicker
Union Carbide
Others
Loca t ? on MM 1b
Geismar, La. 100
Bishop, Texas 200
Pampa, Texas 600
Clear Lake, Texas 300
Kingsport, Tenn. 325
Bayport, Texas 45
Parlin, N.J. 20
Texas City, Texas 300
Philadelphia, Pa. 80
Brownsville, Texas 400
Texas City, Texas 100
S. Charleston, W.Va. 140
Taft, La. 90
100
TOTAL 2,800
Process
CO-methanol
Petroleum gases
Petroleum gases
Acetaldehyde
AcetaIdehyde-ethanol
Acetaldehyde
Acetaldehyde
CO-methanol
AcetaIdehyde-ethanol
Petroleum gases
Petroleum gases
Petroleum gases
Acetaldehyde
177
-------
SUBCATgGQRY C
Product Process
Methyl Methacrylate Acetone Cyanohydrin Process
Methyl Methacrylate is produced by the acetone cyanohydrin process. The
overall chemical reactions are given below:
CH3COCH3 + HCN —•* (CH3) 2OHC (CN)
acetone Hydrogen Cyanide Acetone Cyanohydrin
H2S04
(CH3) 20HC (CN) ^ CH_CH2CONH3HSOU
Acetone Cyanohydrin Methacrylamide Sulfate
CH3OH
CH3CH2CONH3HSO4 -I» CH3CH2CCOOCH3 + NH4HSOU
Methacrylamide sulfate Methyl Methacrylate Ammonium Bisulfate
A process flow diagram is shown in Figure IV-27. Acetone cyanohydrin is
produced by the reaction of hydrogen cyanide and acetone with an
alkaline catalyst in a cooled reaction kettle. The excess catalyst is
neutralized, and crude acetone cyanohydrin passes to holding tanks. The
salt formed by neutralization of the catalyst is removed in a filter
press before the crude acetone cyanohydrin is fed to a two-stage
distillation unit. Most of the water and acetone are removed and
recycled overhead from the first column, and the remainder of the water
is removed at high vacuum from the second column.
Acetone cyanohydrin and concentrated sulfuric acid are pumped into a
cooled hydrolysis kettle to make the intermediate, methacrylamide
sulfate, which is then sent to an esterification kettle to react with
methanol continuously. To prevent polymerization, inhibitors are added
at various points in the process. The esterified stream is pumped to
the acid stripping column, from which the acid residue, made up of
sulfuric acid (40X by weight), ammonium sulfate (28%), water (20%) and
organic substances (10%) is sent to a Spent Acid Recovery unit (SAR).
The recovered sulfuric acid is recycled back to the hydrolyssis reactor.
The overhead stream from the acid-stripping column is then distilled to
remove methyl metacrylate and unreacted methanol, which is recycled.
The last traces of methanol in the methyl metacrylate are removed by
water extraction, after which the monomer is finally purified in a rerun
tower.
The acid residue from the acid-stripping column is the major waste
stream generated in the process, and this waste stream is either sent to
the SAR unit previously mentioned or is discharged into sewers. The
waste streams generated as bottoms from various stills are combined with
the acid residue for spent acid recovery. Water samples from streams
leading to and leaving the SAR unit were taken for analysis, and the
results are shown in the following tabulation:
178
-------
FIGURE IV-27
METHYL METHACRYLATE - ACETONE CYANOHYDRIN PROCESS
TO VAC
HYDROGEN CYANIDE
METHANOL SOLUTION RECTIFIER
179
-------
Into_SAR From_SAR
Flow 260 gallons/1,000 Ib 213 gallons/1,000 Ib
COD 178,000 mg/1 110 mg/1
BOD5 20,700 mg/1 15 mg/1
TOC 69,998 mg/1 18 mg/1
A high concentration of floating solids was observed in the stream lead-
ing to the SAR, and it was impossible to obtain a well-mixed sample.
Therefore, samples from the stream were actually taken from the aqueous
phase beneath the floating solids. The floating solids removed in the
SAR were disposed of by incineration. High concentrations of metal
contaminants such as copper and iron are indicated by the results of the
analysis. Although a large portion of these metals are removed along
with floating solids in the SAR unit, the metal concentrations in the
streams discharged to sewers are still beyond the general discharge
criteria for biological processes. Although sulfuric acid concentration
had been reduced from 40% by weight in the influent to the SAR to 1% by
weight in the effluent, the sulfate concentration in the discharge
stream was still high enough to inhibit the normal functioning of the
biological treatment process.
Because of the highly exothermic reactions involved, the process
requires a large amount of cooling water. The survey data show that
gross cooling water usage amounts to 366 pounds per pound of methyl
metacrylate. Process water is introduced into the system in the form of
direct steam stripping in the amount of 0.56 pounds per pound of
product.
To define BADCT and BATEA, this process should have a Spent Acid
Recovery unit. Two types of SAR units have been devised, and
descriptions of the equipment processing required, and estimated
economics are presented in the following paragraphs.
1• Spent Acid RecQyerY_bY_Neutralization
As shown in Figure IV-28, spent acid is neutralized with ammonia gas to
form ammonium sulfate. The effluent from the neutralization tank is
sent to crystallization and filtration units to separate ammonium
sulfate from the aqueous solution. The economics of this unit are shown
in Table IV-U1.
2- Sp.en^Acid_Re£p^er^_bY_Comp_lete_Combustion The spent acid solution
(see Fig.IV-28) is heated to~such a high temperature (about 1,000°C)
that sulfuric acid decomposes into SO2, 02_, and water vapor.
Simultaneously, the organic substances are oxidized, and the contained
180
-------
CO 03
Oi CV1
^ •
m c=>
m en
•X CO
ui
o
u
CD a
ae x
a o
T
CO
CD
CO
OJ
181
-------
Table IV-A1
~<~
Economics of Spent Acid Recovery by Neutralization"
Investment
Battery Limits = $2,200,000
Off-site 800.000
Total Investment $3,000,000
Operating Costs
Utilities $/yr
Steam: 720,000 M Ib @ 60 0.8e/Kwh = 80,000
Cooling Water: 2,000,000 M gal & 3C/M gal = 60.000
$ 570,000
Chemicals
NH-: 68,000,000 Ib 2c/lb =$1,360,000
Amortization = M+0,QOO
Labor = 200.000
Sub-total $2,000,000
Return on Total Investment @ 20% = $ 600,000
Total Annual Cost = $3.170.000
Net Revenue from Recovered Ammonium Sulfate
390 MM Ibs/yr 0 0.70$/lb = $2,730,000
.
Based on ^85,000,000-lbs/yr Spent Acid Recovery plant.
182
-------
ammonia converted to N2 and water. The SO 2 gas stream is passed over a
catalytic converter to oxidize the SO2 to SO3, which is then absorbed to
form concentrated acid for recycle. The economics of this unit are
shown in Table IV-42.
The economic analyses are based on the following flow rate and
composition of spent acid.
H2S04 = 245,000 Ib/hr
(NH4)2S04 = 16,500 Ib/hr
H20 = 13,500 Ib/hr
Organic substances = 6,150 Ib/hr
~6o7650~lb/hr~"
The acetone cyanohydrin process is the only methacrylate process used
commercially in the U.S. An alternate route used in Japan is nitric
acid oxidation of isobutylene to metacrylic acid, followed by esteri-
fication with methanol.
Producers of methyl methacrylate in the U.S. are shown in Table IV-43.
The estimated economics of production, based on a unit that produces 40
million pounds per year, are presented in Table IV** 4 4
183
-------
Table IV-42
"r\
Economics of Spent Acid Recovery by Complete Combustion
Investment
Battery Limits = $3,000,000
Off-site - 1.000.000
Total Investment $4,000,000
Operating Costs
Utilities $/yr
Fuel: 800,000 MM BTU/yr @> 50^/MM BTU = $ 400,000
Power: 3,000,000 Kwh <5> 0.8<;/Kwh = 24,000
Cooling Water: 750,000 M gal & 3C/M gal = 22.500
$ 446,500
Amortization = $ 600,000
Labor = 100.000
$ 700,000
Return on Total Investment <® 20% - $ 800,000
Total Annual Cost - $1.946.500
Net Revenue on Recovered HjSOr
144,000 tons/vr $20/ton - $2,880,000
?
Based on 485,000,000-lbs/yr Spent Acid Recovery plant.
184
-------
Table IV-43
U.S. Methyl Methacrylate Capacity
Producer Locat ion Capac ity Route
MM Ibs/yr.
Rohm and Haas Houston, Texas
Louisville, Ky. 2kQ.O Acetone-HCN
Bristol, Pa.
DuPont Belle, W. Va. 80.0 Acetone-HCN
American Cyanamid Fortier, La. 40.0 Acetone-HCN
Escambria * Pensacola, Fla. 20.0 Isobutylene oxidation
TOTAL 380.0
* Shut Down
Source: Oil, Paint and Drug Reporter, March 6, 1967
Table IV-M
Estimated Economics for Methyl Methacrylate Production
kO. MM Ib. plant
Total fixed capital=$3.2 MM
Acetone Cyanohydrin Process
Estimated Operation Cost
Cost
. methyl methacrylate
Acetone 5.7
HCN 2.9
Methanol 2.6
Catalyst and chemicals (net) 1.2
Utilities 0.6
Labor and overhead 1.0
Capital charges 2.6
TOTAL 16.6
II. Isobutylene Process
Cost
C/lb. methyl methacrylate
Raw materials 9.3
Utilities 1.8
Labor and overhead 1.0
Total 12.1
185
-------
SUBCATEGORY_C
Product
Ethylene Glycol Hydration ofEthylene Oxide
Ethylene glycol is produced from ethylene oxide by liquid-phase, acid-
catalyzed hy drat ion.
H2COCH2 + H2O —* HOCH2CH2OH
Ethylene Oxide Water Ethylene Glycol
Ethylene oxide and water are reacted at about 300 psig and 180°C in the
presence of sulfuric acid solution. By selection of the oxide-to-water
ratio, it is possible to control the production of the mono-, di-, and
higher glycols produced. Excess water is required for temperature
control and to prevent the formation of undesirable by-products.
Reactor effluent is dehydrated in a multiple-effect evaporator system.
The effluent from the dehydration section is fed to a series of
fractionators. The first tower removes water and traces of the light-
ends, the second produces fiber-grade mono-ethylene glycol, and the
subsequent towers produce diethylene and higher glycols.
A flow sheet for this process is shown in Figure IV-29.
The condensate from the dehydrator is partially recycled, and the
remainder of this stream is the only source of water pollution in the
process. The characteristics of this waste stream obtained from survey
data is shown in the following tabulation:
Flow 584 gallons/1,000 Ib
COD 1,800 mg/1
8.77 lb/1,000 Ib
BODS 69 mg/1
0.34 lb/1,000 Ib
TOC 929 mg/1
4.53 lb/1,000 Ib
The high flow of the waste stream is caused by steam jets with baro-
metric condensers which are utilized to produce vacuum for the multiple-
effect evaporator system. If vacuum pumps with surface heat exchangers
were to replace steam jets and barometric condensers, the flow of this
waste stream could be significantly reduced. The condensate fr,om the
dehydrator could then be totally recycled back to the reactor, con-
sequently, BADCT and BATEA standards should require zero discharge from
this process.
186
-------
<=>
CJ
CO
1
C=>
a
CO
CN
LU
o
X
O
Ul
>-
X
h-
UJ
o 2
!E O
U
O
LU
z
LU
—I
>-
187
-------
The manufacture of ethylene glycol is invariably associated with
ethylene oxide production, and glycol growth rates are moderate. The
U.S. ethylene glycol capacity is presented in Table IV-45.
Estimated economics for ethylene glycol, based on ethylene oxide
availability at 8.50 per pound, are presented in Table IV-46.
Table IV-45
U.S. Ethylene Glycol Capacity
Producer
Location
Mid-1970
Estimated Capacity
Allied
Calcasieu
Celanese
Dow
Eastman
GAP
Houston-PPG
Jefferson
Matador
Olin
Shell
Union Carbide
Wyandotte
Orange, Texas
Lake Charles, La.
Clear Lake, Texas
Freeport, Texas
Plaquemine, La.
Longview, Texas
Linden, N.J.
Beaumont, Texas
Port Neches, Texas
Orange, Texas
Brandenburg, Ky.
Giesmar, La.
Institute, W.Va,
Ponce, P.R.
S. Charleston, W.Va.
Texas City, Texas
Tor ranee, Calif.
Seadrift and Taft, Texas
Giesmar, La.
TOTAL
MM Ib/yr
60
180
300
500
175
75
35
85
360
35
110
100
230
130
120
220
50
130
150
3.0^5
188
-------
Table IV-^6
Estimated Economics for Ethylene Glycol
(80 MM Ib plant)
Total Fixed Capital = $0.8 MM
Estimated Production Cost
C/lb ethylene glycol
Ethylene oxide 6.3
Utilities 0.2
Labor and overhead 0.2
Capital charges 0.3
TOTAL 7.0
189
-------
SUBCATEGORY_C
Product Process
Acrylic Acid Carbon Monoxide Synthesis with Acetylene
Acrylic acid is synthesized from acetylene anc carbon monoxide in a
catalytic solution. The chemistry can be represented by the following
reaction:
C2H2 + H2O + CO —* C2H3COOH
Acetylene Water Carbon Monoxide Acrylic Acid
The acetylene feedstock is first dissolved in THF (tetrahydrofuran) in
an absorption tower. This solution and carbon monoxide are then mixed
in a reactor, and the reaction is carried out at approximately 450°F and
1,500 psig in the presence of a nickel bromide and cupric bromide
solution. The off-gas from the reactor is passed through a THF absorber
to remove acrylic acid vapor and unreacted acetylene, and is then
scrubbed by caustic water for further removal of THF and carbon monoxide
from the gas stream. The liquid reactor effluent, a mixture of acrylic
acid, byproduct acetaldehyde, and catalyst solution, is fed to a
separtion column. The overhead is extracted with water to recover THF
and is distilled to yield purified acetaldehyde. The raffinate from the
separation column is sent to a series of vacuum distillation and
extraction columns. The THF and catalyst solution are recovered and
recycled to the acid reactor. Technical grade glacial acrylic acid is
produced in final distillation columns.
The process flow diagram is shown in Figure IV-30.
The major waste water source is the caustic scrubber water. The
contaminants are THF and Na2C03. The characteristics of waste water
samples obtained during recent plant surveys are summarized in the
following tabulation:
Flow 475 gallons/1,000 Ib
COD 414 mg/1
1.64 lb/1,000 Ib
BODS 186 mg/1
0.737 lb/1,000 Ib
TOC 387 mg/1
1.53 lb/1,000 Ib
190
-------
<
o ^
CO ^
^ o
O u
>-
oc
U
L
NOiivniisio
CO
oe co
CD
CJ 0=
LLJ CJ
NOI1V11I1SIQ
CD
UJ
CO i
•OL I
tV I ij
NOIiVllllSIQ
n
NOIlVllliSIQ
aaauosav
NOIlVllliSIQ
J
LJ
J t
t_J
- CO LU
"^» LU CO
^ r>- -.-r
u NOiivniisio
JH1I
NOiivniisio
]
aaaanaas
H38HOSgV
1
u yoiov3a
— =3
I—I OO
191
-------
Historical data over a period of two months show that TOG ranges from
1.73 to 6.92 pounds per 1,000 pounds of acrylic acid and probability
analysis indicates that 50 percent occurrence is equivalent to 3.08.
The high waste water flew rate is attributed to the utilization of steam
jets used to produce a vacuum in the distillation columns. Converting
steam jets to vacuum pumps can certainly reduce the amount of waste
water generated, although the RWL in terms of COD, BODI5, etc. will
remain the same. Other than reducing waste water flow rate, in-process
modification is deemed unfeasible to further reduce RWL, and
consequently, the data presented can be considered as standards for
BADCT and BATEA.
A wide range of technology is used to produce acrylic acid. The other
important route is based on propylene technology. A mixture of propy-
lene, air, and steam is fed to two tubular catalytic reactors in series
and cooled by circulation of molten salt. Most of the acrylic acid is
condensed and separated from the gaseous stream by quenching. The re-
sulting aqueous solution is then subjected to an extraction with
solvent, followed by distillation for purifying the product and
recovering the solvent.
U.S. manufacturing capacity of acrylic acid and the individual specific
processes used are presented in Table lv-47, and an estimated economic
comparison of the acetylene- and propylene-based technologies is shown
in Table IV-48.
192
-------
Table
Producer
Celanese
Dow Badische
Dow Chemical
Goodrich
Rohm and Haas
Union Carbide
TOTAL
U.S. Acrylic Acid and Acrylates Capacity
Plant Location
Pampa, Texas
Freeport, Texas
Freeport, Texas
Calvert City, Ky.
Bristol, Pa.
Houston, Texas
Institute, W. Va.
Taft, La.
Est. Capacity
(MM Ibs./yr.)
80
10
10
250
70
200
660
Process
Used
b-propiolactone
Acetylene-CO
Propylene
b-Propiolactone
Acetylene-CO
Ethylene oxide-HCN
Propylene
-'-Capacities as of mid-1970 estimated by Stanford Research Institute, CEH . CEH
comments that the Dow facility is not due for start-up until late 1970 and the
Carbide cyanohydrin plant will be shut down when the propylene plant is up to
full capacity by early 1971.
Table
Estimated Acrylic Acid Economics
(150-MM lb.' plant; 1972 Construction )
Total Investment Cost
Process
Acetylene
Propylene
$MM
10.0
16.9
Production Cost
Route:
Acetylene
.0.42 Ib./lb. at 8.0c/1b.
20.88 Ib./lb. at 3.0^/lb.
Propylene
Raw materials
Util ities
Labor
Maintenance (6% ISBL
Overhead (kS% maint.
Taxes and insurance
Depreciation
i TOTAL
+ 3% OSBL)
* labor)
(1 .5% of invest.)
6.85
0.80
0.27
0.32
0.27
0.10
0.67
9.28
3.24
1.12
0.33
0.54
0.39
0.17
1 . ]k
6.93
193
-------
SUBCATEGORY C
Product
Acrylates Esterification of Acrylic Acid
Acrylates are manufactured by esterification of acrylic acid. There are
four main acrylates plus a large number of specialty, smaller-volume
derivatives. The main four are ethyl, 2-ethylhexyl, methyl, and n-butyl
in decreasing order of market share. The 2-ethylhexyl and butyl
acrylates are produced in a separate facility from the methyl and ethyl
esters due to their differences in volatility and solubility.
In the manufacture of methyl or ethyl acrylates, acrylic acid is reacted
with an excess amount of methanol or ethanol in a concentrated sulfuric
acid solution. The effluent from the reactor goes to an extraction
column, where caustic removes the excess alcohol. The effluent water
stream is sent to a distillation column; alcohol is recovered overhead
and recycles, while the aerylate stream is purfied in two distillation
columns by removal of light and heavy ends.
In the manufacture of butyl, 2-ethylhexyl, and higher acrylates, the
esterification is conducted in the presence of cyclohexane, which is
used to remove the water of reaction. The reactor effluent is first
neutralized with caustic and then sent to a series of distillation
columns. Acrylate is purified, while the excess alcohol is recovered
and recycled.
The major process units of the first process are shown in Figure IV-31,
and the chemical reaction can be expressed by the following formula:
C2H3COOH + R-OH H2SO4 C2H3COOR H20
Acrylic Acid Alcohol Acrylates Water
The two main sources of water pollution in acrylate manufacture are the
bottoms of the alcohol recovery still and the effluent of the saponifi-
cation kettle. The possible contaminants in the waste stream are
acrylic acids, alcohols, and sodium salts of various acids. The results
of the plant survey are presented in the following tabulation:
Flow 2,856 gallons/1,000 Ib
COD 4,870 mg/1
117.5 lb/1,000 Ib
BODS 1,942 mg/1
47.1 lb/1,000 Ib
194
-------
£
CO
o "
u. ^
LU CJ>
CO LiJ
LO
3=
CJ
195
-------
TOC 3,290 mg/1
79.5 lb/1,000 Ib
Historical data over a period of two months show that total carbon in
the waste stream ranges from 15.50 to 46.36 pounds per 1,000 pounds of
acrylate produced. Probability analysis of the data indicates that 50
percent occurrence is equivalent to 30.8 pounds per 100 pounds of pro-
duct.
From the data presented in the preceding paragraphs, it is known that
inefficient operation of distillation columns causes significant losses
of organics such as alcohol, acrylic acid, and acrylates into the waste
stream. Recovery of these organics can be achieved by modification of
the distillation columns or by installation of a steam stripper. The
amount of waste flow can also be reduced by recycling the waste water to
an extraction column.
BADCT and BATEA in-process controls should require a steam stripper to
recover organic contaminants in the waste stream and thus achieve a low
RWL.
The U.S. acrylate capacity is presented in the same table used for
acrylic acid (Table IV-47).
196
-------
SUBCATEGORY C^
Product Processes
Terephthalic Acid 1.Nitric Acid Oxidation of Para-Xylene
2. Catalytic Oxidation of Para-Xylene
Terephthalic acid (TPA) constitutes virtually the sole use for p-Xylene.
Based on the mode of oxidation, manufacturing processes can be divided
into the following two classifications:
1. Oxidation of p-Xylene with nitric acid,
2. Catalytic oxidation of p-Xylene.
Only one company is using the nitric acid oxidation of p-xylene in the
United States. This process is a liquid-phase reaction at approximately
300°F and 125-200 psig in dilute HNO3 (about 30-40 weight percent). Ox-
ygen or air is passed into the reactor, where oxidation of p-xylene and
lower oxides of nitrogen takes place simultaneously. The nitric oxides
can be used for nitric acid regeneration.
The second reaction, represented by at least three commercial processes,
utilizes acetic acid as a reaction medium and also involves a heavy met-
al oxidation catalyst. The most widely used commercial process is the
Mid-Century process, in which the oxidation is reported to be based upon
a bromine-promoted heavy metal catalyst, such as cobalt-manganese. Re-
action conditions are 350-450°F and 200-400 psig. The second process
utilizes acetaldehyde as a promoter in place of bromine compounds, and
the reaction is carried out at 250-350°F and 100-200 psig. The third
process uses methyl ethyl ketone' as the catalyst activator and operates
at 200-300°F and 50-150 psig.
A typical flow sheet for the catalytic oxidation process is shown in
Figure IV-32. Preheated acetic acid, p-xylene and bromine catalyst,
together with high-pressure air are charged to a well-agitated reactor
operating at moderate temperature and pressure. The reactor contents
are continuously discharged from the bottom of the reactor as a hot
slurry into a crystallizer vessel, where cooling takes place by flashing
off part of the acetic acid, unreacted xylene, and some water of reac-
tion. The terephthalic acid slurry is passed to a centrifuge for re-
moval of acetic acid and xylene. The filter cake is washed to remove
the remaining reactants and then is dried to give the terephthalic acid
product. The spent reaction liquor and condensate from the crystallizer
vessel are distilled to remove water, recover unreacted Xylene and ace-
tic acid, and remove any other by-products. The acetic acid is
recycled. The off-gas from the reactor is scrubbed with water before
being discharged into the atmosphere. TPA obtained from this process is
considerably purer than that produced by nitric acid oxidation, usually
197
-------
r
Q.
i_
UJ
O
Of.
jY
UJ
2£
^>
_l
O
CN °"
CO Q
r~
114 TF u
* m E
J -J
u? >?
Q_ .
. J
v *— J
Q *1-
< =
X
CD
u
<
X
I-
X
UJ
tt
L
a
:
0
UJ
K
u
e
a
U
«
3
1
c
j
i
B
JJ
CC
UJ
1—
*
t/
E
J
J
>
u
c_
UJ
ce
°
UJ
UJ
CJ
LLJ
CC
(
t
z
UJ
1
<
L
0U
CJ
CJ
—
«x
UJ
*
a;
CJ
-»
— 9
1
E
^
_5
4
4J
x:
^
J
u -
JJ C.
E =
J C
LJ a
-i a
—
^
cv.
t—
UJ
'
J
^
4
*c
cc
CJ
4
JJ
z
UJ
_J
i a
3 U.
3 >
: a
r—
a
«
UJ
k
k
1
^
0
UJ
k
J
1
3
•4
>
Q
a
c/
h
1
|
1
L
C
: «
c c
: *
j «
i h
n
i-
_t
Q.
1—
CO
t-1
UJ
»_
1
1
u
J
c
c
c
c
*l
]
t— a
UJ 3
O U
(O Q
uJ
CJ
CJ
UJ
UJ
' n.--
«£
O
UJ
—1
CJ
CJ
UJ
A
^'
r
c_ £
ii i
*
a.
UJ
r o
0
Of
c_
«3
h-
e/
c:
i
^" '
ac
t- x
t—
CO
•
i
i
k
.
v,
/
s
o- cc
\ — — 1 «s -
1 !•«:—>-
1 • il CJ U. _1 >- ^
1 CJO OC
I j^ 0 ^
1 3 ce
A uj UJ
t . i -
1 _ UJ Ct
r- -^. i o
>> i:
Ii -^
3 * N390HQAH
CO
b~Ny
/
C3
ID UJ
a. uj
u, u.
1
-------
more than 99 weight percent TPA in contrast to 93 weight percent TPA of
the other process.
At some plants, the TPA product is further purified to produce fiber-
grade material. The TPA is washed with hot water to remove traces of
catalyst and acetic acid. The hot water slurry is then heated and
pumped into fixed-bed reactors and hydrogenated . This is followed by
crystallization and drying to recover the fiber-grade TPA.
The major waste water streams in the oxidation process are the bottoms
from the solvent recovery unit and the effluent of the off -gas scrubber,
and the major waste source in the purification process is the discharged
mother liquid from the centrifuge. The characteristics of the waste-
water obtained from plant visits are summarized in the following tabu-
lation.
p£2£§§§ ___ ^22 __ COD __ BOD 5 ___ _TOC
gal/TToOO Ib Ib/l7oOO~lb~
(mg/1)
1 Catalytic 43.4 1.95 1.30 1.52
(5,400) (3,600) (4,200)
1 Purification 715 8.22 5.15 3.53
(1,380) (865) (510)
2 Catalytic
10% Occurrence 186 0.915 0.51 0.55
50% Occurrence 186 1.72 0.82 0.86
90% Occurrence 186 2.52 1.18 1.16
3 Catalytic 1,090 227 68.3 34
(24,950) (7,500) (3,730)
4 Nitric Acid 659 104 58.7 44.9
(18,900) (10,700) (8,180)
Plant 2 has five indentical modules operating in parallel. Data obtain-
ed at this plant over a two day period were analyzed for probability of
occurrence.
Historical RWL data on process waste water flow and COD were obtained
for the catalytic oxidation process at Plant 1. At this plant, there
are actually two oxidation process modules, which operate in parallel.
The data from these two units were subjected to analysis for probability
of occurrence. The following tabulation summarizes the results of this
analysis:
199
-------
Probability of Occurrence
10X 50% ~ ~~90X
90/50
Flow RWL,
gallons/1,000 Ib
Oxidation Unit A
Oxidation Unit B
Purification Unit
COD RWL,
lb/1,000 Ib
Oxidation Unit A
Oxidation Unit B
Purification Unit
132
95
754
174
137
969
217
181
1, 185
1.25
1.32
1.22
8.5
4.9
12.8
12.5
11.2
27.4
16.5
25.5
58.5
1.33
2.28
2.14
The probability analysis was conducted on monthly average data taken by
the manufacturer over a period of twenty-four months. Comparison of the
sampling results and historical results for Plant 1 shows that both the
measured process waste water flow and the COD RWL were significantly
lower at the time of sampling. This is attributed to the fact that the
historical data include surface runoff from the battery limits area.
This amounts to approximately 85 gallons/1,000 Ib of product, with an
associated COD loading of 3.5 Ib per 1,000 Ib of product.
The differences in RWL among the plants can be explained. The nitric
acid oxidation process produces nitric oxides which are supposed to be
used in producing nitric acid. However, it is likely that these nitric
oxides are discharged into sewer at the plant which was visited during
the survey. This results in a "high organic loading in the waste water.
The high RWL of Plant 3 is due to poor process performance, since both
Plant 3 and Plant 4 are scheduled to be phased out in the very near fu-
ture, further investigation of possible in-process modifications to re-
duce RWL is not warranted.
Both Plant 1 and Plant 2 utilize steam ejector systems to obtain vacuum
for process needs. In contrast to discharging the exhaust stream into
the atmosphere, as at Plant 1, Plant 2 employs barometric condensers to
condense the exhaust stream. This causes a significant difference in
the amounts of waste water generated.
To define BADCT and BATEA of the oxidation process, vacuum pumps with
surface condensers should take the place of steam ejectors and
barometric condensers, to reduce the amount of waste flow as well as to
preserve the ambient air quality. If a steam stripper like that
described in the discussion of aniline should be installed to recover
organic contaminants in the waste water of the purification process, RWL
can be reduced approximately by about three-fourths.
200
-------
Process water usages as well as gross cooling water usages are varied
among plants and processes. Information obtained from the plant survey
is shown in the following tabulation. Plants are identified with the
same identification as that used for RWL.
Plant
1 (Oxidation)
1 (Purification)
2
3
4
Process^Water_ysage
Ib/lb product
N.A.
N.A.
N.A.
N.A.
4
S22liS2_ Water^Usage
Ib/lb product
N.A.
N.A.
188
N.A.
20,000
several approaches to manufacture of TPA are under investigation, but
none of them has been commercialized in the United states. The current
U.S. capacity for TPA is presented in Table IV-49. The estimated eco-
nomics for TPA manufacture bythe oxidation process are shown in Table
IV-50.
Producer
Amoco
DuPont
Eastman
Mobil
Total
Table IV-49
U.S. Terephthalic Acid Capacity
Plant Location
Decatur, Ala.
Joliet, 111.*
Gibbstown, N. J.
Old Hickory, Tenn.
Kingsport, Tenn.
Beaumont, Texas
Est. Crude
TPA Capacity
(MM Lbs./Yr.)
•''May be shut down or switched to isophthalic acid produc-
tion.
Source: Chem Systems' estimates as of mid-1970.
201
-------
Table IV-50
Estimated Economics for Terephthalic Acid
(400-MM Ib plant--1972 construction)
Investment cost
Process $ MM
Oxidation (Bromine compound) 52.9
Oxidation (Methylethyl Ketone) 58.6
Production costs
tf/lb
1 2
Amoco Mobil
Raw materials 6.62
Utilities 0.65
Labor 0.09
Ma int. (6% ISBL + 3% OSBL) 0.61+
Overhead (45% ma int. + labor) 0.33
Taxes & insurance (1.5% of invest.) 0.20
Depreciation (10 yr) 1.32
Total 9.85
By-product credit
Net
^Includes 0.67 Ib p-xylene at 6.5
-------
SUBCATEGgRY_C
Product. _______ ______
Dimethyl Terephthalate Esterification of TPA
The high-purity monomer required for the development of polyester fibers
and films is produced by converting terephthalic acid (TPA) to dimethyl
terephthalate (DMT) . However, with improved technology for the manu-
facture of fiber-grade TPA, it is expected that most of the new fiber
and film capacity installed will be based on purified TPA.
In the process for the esterification of TPA to DMT, preheated TPA and
methanol are fed to a reactor in the presence of sulfuric acid as a
catalyst. DMT in the reactor effluent is recovered and purified by con-
ventional methods such as crystallization and distillation.
A flow sheet for this process is shown in Figure IV-33.
The water separated after condensation and the benzene used in the reac-
tor to prevent the methanol from vaporizing too rapidly are the major
water pollution sources. The waste water may contain some alcohol, ben-
zene, and proproduct or by-product losses. Another water pollution
source is the waste stream resulted from cleaning up scattered product
resulting from leaks in various portions of the equipment. The
characteristics of the waste water obtained from plant surveys are shown
in the following tabulation:
_1. £lant_2 _
Flow, gal/1,000 Ib 68.8 388 1,070
COD,
lb/1,000 Ib 8«93 55.2 0.91
mg/1 15,000 17,000 102
BODS,
lb/1,000 Ib a. 81 31.0 0.19
mg/1 8,400 9,580 21
TOG,
lb/1,000 Ib 3.88 22.5 0.62
mg/1 6,800 6,950 69
Historical RWL data on process waste water flow and COD were obtained at
Plant 1. At this plant, there are actually two modules, with different
production capacities, operating in parallel. The results of the anal-
ysis for probability of occurrence are summarized in the following tabu*
lation:
203
-------
a
u
<
u
a.
iu
ac
co Z
co o
> P
O
UJ
a.
UJ
X
K
UJ
NOIlVllllSIQ
Noiiniiisia
NOIlVllllSIQ
H013V3H
AUBA033X
10NVH13N
204
-------
___ Flow RWL____ __ _CQD_RWL
__ I2al/I.t000_lbl_
Unit A Unit_B
10% Occurrence 167 150 13.5 16.1
50% Occurrence 313 248 34. 33.7
90% Occurrence 461 344 86.5 70.5
Ratio 90X/50K 1.47 1.39 2.54 2.06
The analysis was based on consecutive 30-day average data collected by
the manufacturer over a period of 24 months. The data show that there
is only a slight variation between two units of different sizes at the
same plant. However, the measured RWL is significantly lower than that
from historical data. Again, the difference is due to the fact that
historical data includes surface runoff caused by rainfall and
housekeeping.
The survey data also reveal significant variations among plants, The
high waste water flow of Plant 3 is caused by steam jets with barometric
condensers, while the low flow of Plant 1 is due to discharging steam
jets directly into the atmosphere. The variation in organic loadings
between Plant 1 and Plant 3 is due mainly to different performance ef-
ficiencies of the solvent recovery units and to varying effectiveness of
preventive measures for process leakages. The high RWL presented by
Plant 2 is attributed to the low- purity TPA manufactured by nitric acid
oxidation. Plant 2 is scheduled to be phased out in the very near
future.
To define BADCT and BATEA, it is certain that vacuum pumps with surface
heat exchangers should be utilized in producing vacuum for process needs
and that good performance of solvent recovery units should be required.
Also, excellent preventive maintenance should be emphasized to reduce
RWL.
Process water usage and gross cooling usage are presented in the follow-
ing tabulation:
Plant £rocess_Water_Usac[e Cooling Water Usage
Ib/lb product Ib/lb product
Plant 1 N.A. N.A.
Plant 2 2 23,000
Plant 3 N.A. 150
An alternate route in the manufacture of DMT is the Hercules process.
This synthesis involves liquid-phase oxidation of p-Xylene in acetic
acid with a cobalt acetate or naphthenate as a catalyst to produce p-
toluic acid. This is subsequently esterified with methanol to produce
diethyl hydrogen terephthalate, which is finally esterified to form DMT.
205
-------
The U.S. capacity for DMT is shown in Table IV-51.
Table IV-51
U.S. Dimethyl Terephthalate Capacity
(Million Ibs./yr.)
Estimated Capacity
Producer Plant Location p-Xylene
Amoco Joliet, 111.
Decatur, Ala.
DuPont Gibbstown, N.J.
Old Hickory, Tenn. --
Eastman Kingsport, Tenn.
Hercules Burlington, N.J. 100
Spartenburg, S.C. 100
Wilmington, N.C. j+00
Total 600
Crude TPA
150
150
250
250
300
1100
Iota 1
150
150
250
250
300
100
100
koo
1.700
206
-------
SUBCATEGORYj:
Product_
Para-cresol Sulfonation of Toluene
As in the case with other coal-tar derivatives, the supply of coke-oven
by-product cresylics has failed to keep up with demand. P-cresol was
the first isomer to be synthesized commercially and is produced by
sulfonation of toluene. The basic chemical equations are given below:
C6H5CH3 + H2SO4 — •*• (SO3H)c6fWCH3
Toluene Sulfuric Acid
(S03H) C6H4CH3 + NaOH _^ (OH) C6H4CH3 «• Na2SO3
P-Cresol
A process flow sheet is shown in Figure IV-34. Toluene and a gas mix
ture of sulfur dioxide and sulfur trioxide are fed into a sulfonation
reactor. The reactor effluent gas is passed through a caustic scrubber
to remove unreacted sulfur dioxide. The liquid effluent from the reac-
tor is first diluted with steam and then sent to a caustic fusion col-
umn, where crude p-cresol is produced. The crude product is then sent
to a washing-separation column, where excess caustic solution is
neutralized and two phases are formed. The aqueous phase is discharged
from the system, and the organic phase is fractionated to obtain pure p-
cresol.
Since the sulfonation reaction approaches 100 percent conversion of sul-
fur dioxide and trioxide, the vent gas scrubber water does not present a
significant water pollution source. The major waste water stream is the
aqueous phase discharged from the sulfuric washing/separation column.
The average composition of this stream is 77 percent water, 15.2 percent
sodium sulfite, 5.1 percent sodium sulfate, 0.4 percent cresylic
compounds, and 1.7 percent other organic substances such as cresols,
phenols, etc. The data obtained from Plant 1 are shown in the following
tabulation:
Flow 1,291 gallons/1,000 Ib
COD 23,800 mg/1
256 lb/1,000 Ib
BODS 11,400 mg/1
123 lb/1,000 Ib
TOC 5,020 mg/1
54 lb/1,000 Ib
207
-------
208
-------
The suifite and organic contaminants cause the high oxygen demand in the
waste water, while the cresol contaminant (10 mg/1) constitutes an odor-
ous nuisance in the atmosphere.
According to the literature, the organic contaminants in the waste water
exhibit very strong anti-oxidant properties and present a difficulty to
ordinary biological treatment processes. several possible methods of
controlling this waste water discharge have been investigated. The most
promising scheme appears to be activated carbon adsorption of organic
contaminants prior to oxidation, followed by chemical regeneration of
cresylic compounds adsorbed on the carbon, to return a valuable product
to the process, eliminate the odor problem, and reduce the discharge of
pollutants,
A demonstration plant and its economics are briefly described in the
following paragraphs. The system consists of two 4ft-diameter by 30ft
high columns of 304 L stainless steel. Each column is loaded ro a
height of 18.5 ft. with approximately 6,000 pounds of activated carbon.
The system was designed to have sufficient capacity for <.i one-day
operational cycle, requiring one column to be regenerated each 2u hours.
Ten percent sodium hydroxide solution is used to regent:" -i^e spent
activated carbon, and the desorbed cresylic compounds are rec^',.'lPd back
to the process. During a seven-month period, the columns were operated
at an average superficial velocity of 3.2 gpm/ft. Influent
concentrations during the period were 3,500 to 6,500 mg/1 cresylic
compounds, and effluent concentrations were between 0 and 700 mg/1
cresylic compounds. During this time, 271,600 pounds of p-cresol were
returned to the process. This amount of p-cresol represents a value of
$114,000.
As demonstrated, the activated carbon system not only can recover p-
cresol from the waste water and turn it into profit, but also can
decrease the RWL of the system. Furthermore, it improves the
treatability of the waste water. Consequently, to define BATEA and
BADCT control technologies, an activated carbon system should be
incorporated into the process.
Two other process routes for the manufacture of p-cresol are currently
practiced: vapor-phase methylation of phenol over alumina catalysts,
and liquid-phase oxidation of meta- and para-cumene.
Producers of p-cresol in the U.S. and the economic of production are
presented in Tables IV-52 and IV-53
209
-------
Table IV-52
U.S. Cresol Capacity (1972)
Company MM 1b Process
Hercules, Inc. (Gibbstown, N.J.) 6 p-cymene oxidation
Koppers (Follansbee, W. Va.) 10 phenol and methanol
Pitt-Consol (Newark, N.J.) 80 phenol and methanol
Sherwin Wil1iams (Chicago, 111.) 10 toluene sulfonation
Total 106
Table IV-53
Economic Evaluation of Activated Carbon System
for Wastewater from p-Cresol*
1. Annual Operational Cost
Depreciation (10 year straight line) $ ]k,
Maintenance (5% of installed cost) 7,000
Utilities 1,050
Raw Materials (NaOH and Filter Aid) 17,250
Labor (using existing manpower) 0
Carbon Make-Up ' k . OOP
$ 43,700
11. Annual Net Revenue $210,320
(500,770 pounds p-cresol recovered/year,
sale price= $0.42/pound)
111. Analysis
Gross Profit= $210,320 -- $43,700= $166,620
Tax (50%) = 83.310
After Tax Profit = $ 83,310
After Tax Cash Flow= $83,310 + $14,400= $97,710
After Tax RDl-$' x 100%- 67.9%
Payout Time = = ] 'k7 yrs*
-'•"Recovery of P-Cresol from Process Effluent," Baber, C.D., Clark,
E.W., Jesernig, W.V., and Huether, C.H., Presented at the 74th
AlChE, New Orleans, La., March 1973.
210
-------
SUBCATEGORY C
Product
Aniline Nitration and Hydrogenation of Benzene
Benzene is first converted to nitrobenzene in a mixture of nitric and sul-
furic acids:
H2S04
C6H6 + HNO3 ~ t ~ C6H5NO2 + H20
Benzene Nitric Acid Nitrobenzene Water
The reactor effluent is decanted into a liquid/liquid separator, where
crude nitrobenzene is separated from the acid solution. The acid
solution is concentrated by steam stripping and recycled back to the
reactor. Crude nitrobenzene is washed, vaporized, and fed to a
fluidized-bed reactor containing a copper-silica hydrogenation catalyst,
where the following hydrogenation reaction occurs:
C6H5NO2 + 3H2 _*, C6H5NH2 + 2H2O
Nitrobenzene Hydrogen Aniline Water
The unreacted hydrogen is recycled to the reactor. Reactor effluent
goes to a separator, where two phases are formed. The organic phase
contains water, and is fractionated in a two-tower system to remove
heavy residue and water from the aniline product. The aqueous layer,
formed by the water of reaction, contains some aniline and is discharged
into sewers.
The process flow diagram is shown in Figure IV-35.
The major waste water sources in this process are the crude nitrobenzene
wash water and aniline water formed in the final separator. RWL survey
data of this process are shown in the following tabulation:
Flow 190 gallons/1,000 Ib
COD 13,400 mg/1
21.2 lb/1,000 Ib
BOD5 15 mg/1
0.02 lb/1,000 Ib
TOC 12,150 mg/1
19.2 lb/1,000 Ib
211
-------
> z
oi =!
* Z
^ ^
o <
.UJ^l
SL
s h
UJ I :
UJ
CO
CJ
e
212
-------
Results of analyses indicate that, in addition to the parameters shown
above, sulfate concentrations in waste water streams are at levels
inhibitory to biological treatment processes. The high RWL of this
process is attributed to the high aniline concentration (3 percent) in
aniline water from the final separator. It is a common practice to
recover aniline by extraction either with incoming nitrobenzene or with
benzene. However, such recovery was not practiced at the plant visited
during the survey.
BADCT and BATEA in-process controls are defined by implementing an
aniline recovery system to reduce process RWL. Instead of using a
nitrobenzene extraction scheme, an effective steam-stripping system has
been devised, and the following is a description of the equipment and
processing required.
Water from a 108 Ib/yr aniline plant is steam stripped in a 2.5' x 40'
tower. The feed to the stripper is 17 gpm containing 3.1 percent
aniline by weight. The bottoms from the stripper will contain about 0.2
percent aniline. The overhead, essentially a 50/50 mixture of aniline
and water is sent to incineration. Figure IV*-36 is a process flowsheet
of the proposed aniline stripper system.
With this modification, RWL can be expected to achieve the following
values:
Flow 184 gallons/1,000 Ib
COD 1,390 mg/1
2.13 lb/1,000 Ib
TOC 1,490 mg/1
2.29 lb/1,000 Ib
The totally installed cost for the stripper, including heat exchange,
pumps, instrumentation, piping, foundations, electrical wiring,
structures, etc. is $115,000. The total annual operating cost,
including depreciation, is about $45,000. For the 108 Ib/yr aniline
plant, this adds about .052/lb to the cost of the aniline. Table IV-54
presents the economics of the proposed aniline stripper.
The alternate routes in manufacturing aniline are the traditional
technique of nitrobenzene liquid-phase reduction with iron filings and
the liquid-phase nitrobenzene hydrogenation technique. U.S. aniline
capacity from these processes is presented in Table IV-55. Assuming
that nitric acid and sulfuric acids are available at $30 per ton, esti-
mated production costs for a 40.0 million pounds per year aniline plant,
including benzene nitration facilities, are shown in Table IV-56.
213
-------
FIGURE IV-36
ANILINE STRIPPER
100 °F
8236 #/HR.
WATER
264#/HR.
ANILINE
248# ANILINE
230# WATER
TO INCINERATOR
20 PS G
150° F
•2-1/2 fl x
18 TRAYS
2500 #/ HR.
40 PSIR
STEAM
c.m.
<&
95
°
8006# WATER
-* IB* ANILINE
-------
Table IV-5^
Aniline Stripper Economics
Investment
Tower Cost, including trays, pumps, exchanges, = $]] 000 Totally Inst
instruments, piping, foundations, etc. ' y
Operating Costs
Uti1ities
$/Yr
Steam: 2500 #/hr. x $.55/M# x 8000 = $11,000
Power: 800,000 kwh x $,01/kwh = 8,000
Cooling Water: 20 X 106 Gals, x $.25/M Gals = 5.000
$2^,000
Investment Related
Maintenance Material & Labor k% = $ 4,600
Plant Overhead 65% of Maintenance = 3,000
Insurance, Taxes 1.5% = 1,700
Depreciation 10% on BLCC - 11.5QO
Total Expenses
C/Gal Handled 0.55 C/Gal.
C/lb Aniline removed 2.36
-------
Table IV-55
U.S. Aniline Capacity (1972)
Company
Allied
American Cyanamid
DuPont
First Chemical
Mobay
Rubicon
Total
Location
Moundsvilie, W. Va
Bound Brook, N.J.
Willow Island, W.Va.
Gibbstown, N.J.
Beaumont, Texas
Pascagoula, Miss
Hew Martinsvi1le, W. Va.
Geismar, La.
MM Ib
60
60
kO
130
200
35
70
Jt9_
585
Table IV-56
Estimated Economics for Aniline
( kO. MM Ib. plant)
Total*Fixed Capital=$3.2 MM
Estimated Operation Cost
Cost
Benzene
Nitric Ac i d
Hydrogen
Catalyst and chemicals
Utilities
Labor and overhead
Capital charges
C/lb. aniline
3.1
2.k
0.8
0.3
Q.k
0.6
2.6
TO
216
-------
SUBCATEGORY C
Product
Bisphenol-A
Process
Condensation of~Phenol and Acetone
Diphenyl propane, also known as bisphenol-A, is produced by reacting
phenol with acetone in the presence of acid catalyst, and the chemical
reaction is given below:
2C6H50H + CH3COCH3 _* CH3C(C6H<£OH) 2CH. + H2O
Phenol Acetone Bisphenol-A * Water
A number of by-products are formed in conjunction with the main
reaction. The earlier processes eliminated these impurities by
batchwise crystallization, while the new process, the Hooker process,
employs a continuous distillation and extractive crystallization under
pressure to purify the product.
A process flow diagram of the Hooker process is shown in Figure IV-37.
Phenol and acetone at a molar ratio of approximately three to one are
mixed, saturated with hydrogen chloride gas, and sent to the reaction
vessel. Reaction conditions are about 40°C, close to atmospheric pres-
sure, with a mercaptan used as a catalyst. The crude product is
stripped of HCl and water of reaction. The overhead is decanted into an
organic phase (consisting mainly of phenol which is recycled) and an
aqueous phase. The latter goes on to an HCl-recovery unit, and water is
sent to disposal.
Bottoms from the stripper are sent to a series of purification distilla-
tion chambers, where excess phenol, isomers, and heavy ends are removed
from the system for either recycle or disposal. Distillate from the
last chamber is sent to the extraction operation, which produces a
slurry of pure crystals. The filtrate from the centrifuge is partially
recycled to the crystallizer, and the remainder is concentrated in an
evaporator to produce liquid bisphenol-A.
The water separated from the HCl recovery unit, the extracted aqueous
phase from the crystallizer, and the condensate from the final
evaporator are the major waste water sources. The characteristics of
the waste water obtained from survey data are presented in the following
tabulation:
Flow 66.8 gallons/1,000 Ib
COD 30,699 mg/1
17.11 lb/1,000 Ib
TOG 9,216 mg/1
5.13 lb/1,000 Ib
217
-------
UJ
PS <
> O
01 Z
«v i| i
O 1
s
£2
-------
Phenol 12,713 mg/1
7.1 lb/1,000 Ib
The high concentration of phenol produces an inhibitory effect and
interferes with the BOD5 measurement. The organic contaminants in the
waste water are mainly phenol, bisphenol, and organic solvent.
Incomplete separation of the aqueous and organic phases in the decanter
causes the high loss of organics into the waste water. Organic vapor
escaping from the final evaporator also contributes a significant amount
of contaminants.
To define BADCT and BATEA, a steam stripper should be required to
recover and recycle these organic contaminants in the two major waste
streams. The specification and the estimated economics of a steam
stripper have been presented in the discussion of Aniline.
The total process water usage of this process is approximately 0.25
pounds per pound of bisphenol-A, while the gross cooling water usage is
about 197 pounds per pound of product.
The U.S. Bisphenol-A capacity and estimated economics are presented in
Tables IV-57 and IV-58.
219
-------
Producer
Dow
General Electric
Monsanto
Shell
Union Carbide
Table IV-57
U.S. Bisphenol-A Capacity
Location
Midland, Mich.
Mt. Vernon, |nd.
St. Louis, Mo.
Houston, Texas
Marietta, Ohio
Estimated Capacity*
MM Ib/yr
58
25
30
100
TOTAL
238
*As of mid-1969. Reported by Chemical Profiles 7/1/69.
Shell is reportedly expanding to 100 MM Ib/yr by 1/1/71,
and Dow is reportedly planning a new 100 MM Ib/yr plant
for Freeport, Texas due in 1972.
220
-------
Table IV-58
Estimated Economics for Bisphenol-A
(20 MM Ib plant)
Total Fixed Capital = $1.9 MM
Estimated Operation Cost
C/lb bisphenol-A
Phenol 7.2
Acetone 1 .k
Catalyst and chemicals 0.1
Utilities 1.0
Labor and overhead 0.9
Capital charges 3.1
TOTAL 13.7
221
-------
SUBCATEGQRY C
Product_ Process
Caprolactam Oxidation of Cyclohexane
Caprolactam is produced in the Beckman process by the addition of
hydroxylamine sulfate to cyclohexanone, which is derived from
cyclohexane. The basic chemical equations are given below:
H3BO3 H2NOH' HSOU
C6H12 + 02 . ^ ~ C6H1.1.0 «^ ~ C6HJJNOH
Cyclohexane Oxygen Cyclohexanone Cyclohexanone Oxime
or Air
H2SOU
CH(CH2)5CONH + (NHU) 2S04
••*• Caprolactam Ammonium Sulfate
A process flowsheet is shown in Figure IV-38. Feed and recyled
cyclohexane are mixed with air in an oxidation reactor in the presence
of boric acid, which minimizes adipic acid production. The oxidation is
carried out at approximately 150 psig and 160°C. The gaseous effluent
is scrubbed to separate unreacted cyclohexane from what is essentially
nitrogen. The liquid effluent is flashed with water and separated into
an organic phase and an aqueous catalyst phase, which is then sent to a
catalyst recovery unit. The organic phase is essentially a mixture of
unreacted cyclohexane, cyclohexanone, and cyclohexanol. This mixture is
first distilled to recover unreacted cyclohexane and followed by
saponification and fractionation to separate cyclohexanone from
cyclohexanol, which is then converted to cyclohexanone by
dehydrogenation.
The hydroxylamine sulfate is obtained from ammonium nitrates and sulfur
dioxide. Ammonia gas and air are fed to a converter where ammonia is
burned at about 700°C in the presence of a catalytist and converted to
disulphonate by contacting with ammonium carbonate and sulfur dioxide in
series. The disulphonate is then hydrolyzed to hydroxylamine.
By addition of cyclohexanone to hydroxylamine sulfate, cyclohexanone
oxime is first produced and rearranged in nearly quantitative yield to
caprolactam in the presence of concentrated sulfuric acid. The product
is neutralized, and the ammonium sulfate solution is extracted with
benzene to recover the lactam product and discharged to a concentration
and recovery step.
The major water pollution sources in this process are the draw-offs from
catalyst recovery unit, saponification and wash tower, and the final
product purification step. The contaminants in the waste stream are
small amounts of diacids formed during the oxidation step, sodium salts,
222
-------
> u
O a-
"- 5
-------
and unrecovered interirediate products. The characteristics of the waste
water obtained from the plant survey are summarized in the following
tabulation:
Plant 1 Plant 2
Flow 1,334 gallons/1,000 Ib 2,500 gallons/1,000 Ib
COD 358 mg/1
4.0 lb/1,000 Ib N.A.
BOD5 147 mg/1
1.64 lb/1,000 Ib 11.2 lb/1,000 Ib
TOC 109 mg/1
1.22 lb/1,000 Ib N.A.
Since it is deemed unfeasible to reduce RWL of this process by any in-
process modification, the RWL presented in the preceding tabulation can
be considered as standard for 3ADCT and BATEA.
Several other commercial routes to caprolactam are available, and pro-
cess highlights of each route are summarized in the following para-
graphs.
In the Toyo Rayon process, nitrosylchloride is first manufactured by
reacting ammonia gas with air at 700°C and atmospheric pressure using
platinum-rhodium gauze as a catalyst, then with concentrated sulfuric
acid, and finally with hydrogen chloride. The nitrosylchloride gas
mixture is then reacted with cyclohexane to give the cyclohexane oxime
hydrochloride. The reaction is carried out in the liguid phase, using
the visible light emitted by mercury lamps to induce the
photonitrosation. Subsequently, cyclohexanone oxime hydrochloride is
treated with oleum to produce a sulfuric acid solution of caprolactam,
which is then purified by a series of purification steps.
The Snia Viscosa process is based on the nitrosation of hexahydrobenzoic
acid with sulfuric acid in oleum. The feed toluene is oxidized with air
and then hydrogenated over a palladium catalyst to form hexahydrobenzoic
acid. Caprolactam is then formed by reacting hexahydrobenzoic acid with
nitrosylsulfuric acid, which is prepared by bubbling N2O3_ into the
cyclohexane carboxylic acid dissolved in oleum.
The other route (referred to as the Caprolactone Process) produces
caprolactam without any ammonium sulfate by-product. Caprolactone is
first produced by oxidation of cyclohexanone with peracetic acid, which
is produced by acetaldehyde oxidation. The resulting Caprolactone is
distilled under vacuum and reacted with ammonia at high pressure to form
224
-------
caprolactam, which is
techniques.
purified using conventional distillation
Although many processes exist for caprolactam production, the only pro-
cess used commercially in the U.S. as shown in Table IV-59 is the Beck-
mann process. The relative economics for the Beckmann, Caprolactone and
Toyo Rayon processes are summarized in Table IV-60 which shows that the
Beckmann has the lowest investment cost.
Table IV-59
Caprolactam Capacity
(MM Ib.)
Company
Al1ied Chemical
Columbia N1PRO
Dow Badische
DuPont
Union Carbide
TOTAL
Locat ion
Hopewel1, Va.
Augusta, Ga.
Freeport, Texas
Beaumont, Texas
Taft, La.
1967
300
kk
90
50
50
53^
1972
300
150
176
shut down
shut down
626
Process
Beckmann
Beckmann
Beckmann
N i t rocyc1ohexane
Caprolactone
225
-------
Table IV-60
Estimated Economics for Caprolactam
(150-MM-lb. plant; 1972 construction)
TOTAL FIXED CAPITAL
S MM
Beckmann 37.4
Caprolactone 39.8
Toyo Rayon 40.0
Investment includes cyclohexanone
and oximat ion.
Investment includes peracetic acid
and caprolactone units.
PRODUCTION COST
C/lb. caprolactam
Raw materi als
Ut i 1 it les
Labor
Maintenance
(6% ISBL + 3% OSBL)
Overhead
C*57o of maint, £• labor)
Taxes and insurance
(1.57= of inv.)
Depreciation (10 yr. )
TOTAL
By-product credit
NET
Becknjgnr;
11.431
1.60
0.58
1.20
0.80
0.38
18.49
fr.W»
14.05
Cflpro 1 aqtong
10. 712
1.9)
0.40
1.28
0.76
0.41
2.66
18.13
6.22 '
11.91
Toyo Ravon
9. 14*
2.25
0.36
1.28
• 0.74
0.41
2.66
16.81+
1.58
15.26
Includes cyclohexane (0.88 Ib. at 3.3C/lb.), NH3 (1 Ib. at 2 c/lb.) and
oleum (1.7 Ib. at $36/ton). Ammonium sulfate credit at $23/ton.
Includes cyclohexane (1.0 Ib. at 3.3c/)b.) and acetaldehyde (0.62 Ib. at
S.Oc/lb.). Acetic acid credit at 6c/lb.
Includes cyclohexane (0.95 Ib. at 3.3c/lb.) ammonia (I Ib. at $*tO/ton and
oleum (1.7'lb. at $36/ton). Ammonia sulfate credit at $23/ton.
226
-------
SUBCATEGORY C
Product P£2£§§§
Long Chain Alcohols Ethylene Polymerization
Long-chain alcohols are manufactured from ethylene in the presence of
Ziegler catalysts. The process begins by reacting aluminum metal with
ethylene and hydrogen to form triethyl aluminum (TEA). Ethylene is
added to this compound at high pressures to give trialkyl aluminum
compounds, which are then oxidized with dry air to aluminum tri-
alkoxides. These are hydrolyzed by sulfuric acid to primary alcohols
having an even number of carbon atoms. The basic chemical equations are
summarized as follows:
3C2H4 + 1 1/2H2 + Al _^ (C2H5) 3Al
Ethylene Hydrogen Aluminum Triethyl Aluminum
(C2H5)3A1 + nC2H4 —* R,^
~ ~ R - Al
Triethyl Ehtylene Triethyl Aluminum
Aluminum
R.-O ^
+°2 —* R\-0 - Al
Aluminum Trialkoxides
H2SO4
~ R.10H + R20H + R30H + A12 (SOU) 3
H20 Long-Chain Alcohols Alum
A simplified flow diagram is shown in Figure IV-39. An atomized
aluminum powder is first activated in a non-aqueous slurry media and
next hydrogenated with dry hydrogen gas under pressure to give diethyl
aluminum hydride. The hydride is then contacted with ethylene to
produce TEA. Approximately two moles of TEA are recycled to the
hydrogenator and one mole goes to the polymerization step. Recycle TEA
solvent and aluminum are separated by means of a centrifuge.
In the polymerization section, TEA is reacted with ethlyene under
pressure to make trialkyl aluminum, which is then oxidized to produce
alkoxides. A non-aqueous solvent such as toluene is circulated and
227
-------
r
o
x
S
£ <
13 £
Z
O
s
I
n
UJ LU
x>
£
!§
228
LT
LT
*" 85
ii
is
en
T
§
85
-------
recycled in this section. In the hydrolysis section, the alkoxides are
hydrolyzed with sulfuric acid and water to yield alcohols and a solution
of alum and water. The alum solution is separated from the alcohols in
a decanter. The sulfuric acid residue is first neutralized with dilute
caustic solution and next washed with hot water to remove sodium
sulfate. In both the neutralization and wash steps, the alcohols are
separated from the aqueous phase in decanters.
The crude alcohols are then dehydrated and fractionated in a series of
distillation columns to obtain pure alcohol products. Steam jets are
used to produce vacuum in the stills.
The major water pollution sources in this process are the draw-offs from
decanters and the condensate of the steam jets. Depending upon the
desired concentration of the alum solution recovered, the cycle of
decanter draw-off waters, and the modes of condensing ejected steam, the
volume of waste water per unit production will vary.
Straight-chain alcohols are also obtained by the oxo reaction starting
from straight-chain - olefins and by direct oxidation of normal
paraffins. Producers of long-chain synthetic alcohols in the U.S. are
presented in Table IV-61.
Table IV-61
U.S. Long-Chain Alcohol Capacity
Producer
Continental
Ethyl
Shell
Location
Lake Charles, La.
Houston, Tex.
Houston, Tex.
Shell* Geismar, La.
Union Carbide Texas City, Tex.
1965
Capacity
MM Ibs/yr.
100.00
50.00
50.00
100.00
JfO.OO
Type of
Alcohol
Primary
P r i ma ry
80% Primary
20% Secondary
80% Primary
20% Secondary
Secondary
Process
Ziegler
Ziegler
Oxo
Oxo
Oxidation
Raw Material
Ethylene
Ethylene
Cracked wax
Cracked wax
n-paraffins
"Due on stream in 1966.
Source: Oil. Paint, and Drug Reporter. August 26, 1965.
229
-------
SUBCATEGORY C
Product Process
Tetraethyl Lead Addition~of Ethyl chloride to
Lead in Sodium - Lead Alloy
Over 90 percent of all tetraethyl lead is produced by some version of a
conventional forty-year-old batch process in which an alkyl halide reacts
with sodium-lead alloy. The reaction, occuring in a horizontal autoclave
provided with a reflux condenser to recover any vaporized alkyl halide,
yeilds a mixture of TEL, salt, and lead. The reaction, carried out at
60 psig and 70°C, is given below:
UPbNa + 4C2H51 —* (Q2H5) 4Pb + 3Pb
Sodium Lead Ethyl~"chloride "~ TEL Lead
Alloy
The product mixture is fed batchwise to a still, where the tetraethyl
lead is separated from the by-product lead and sodium chloride by direct
steam stripping. The tetraethyl lead and stripping steam are condensed
and sent to a decanter, where tetraethyl lead is drawn off as a bottoms
stream. The upper aqueous layer in the decanter, containing unreacted
ethyl chloride and dissolved organic by-products, is discharged into a
process ditch.
The salty sludge bottoms from the still are sent to a lead recovery
unit, and the centrate is combined with the supernatant from the TEL
decanter before being discharged into a settling basin for final
recovery of solid lead.
The process flow sheet is shown -in Figure IV-40.
since recovery of by-product lead is considered an integral part of the
TEL manufacturing process, the effluent from the settling basin is con-
sidered as the waste water source of the process. The waste water
characteristics obtained from the plant visit are shown in the following
tabulation:
Flow 12,000 gallons/1,000 Ib
COD 1,100 mg/1
110 lb/1,000 Ib
BODS 40 mg/1
4 lb/1,000 Ib
TOC 56 mg/1
5.6 lb/1,000 Ib
230
-------
<=>
Q_
O
CJ
Q
UJ
uu
I—
CO
231
-------
The high amount of waste water is due mainly to the nature of batch pro-
cesses, which require a large quantity of water in cleaning up the
reactor between reaction batches. Another cause of high water use is
the vent-gas scrubber at the "lead" recovery unit. The intermittent
dosage of "still-aids" such as soap or iron to control the plating out
of lead on the still walls, as well as unrecovered ethyl chloride, TEL,
and metallic lead, all contribute to the high chemical oxygen demand.
In defining levels of control technology, it is suggested that recycling
of the aqueous layer in the decanter to reduce fresh water usage, and
consequently the amount of waste water discharged, can be considered for
BPCTCA. BADCT and BATEA should have a steam stripper for effective
recovery of unreacted ethyl chloride and product TEL from the stream
prior to their discharge into the settling basin.
An alternate process, which is based on the electrolysis of an alkyl
Grignard reagent, is used by only one company in the world. This
involves a totally different approach and offers at least three
advantages: 1) it gives higher product yields; 2) it does not make by-
product lead, hence eliminating the inefficient recovery and recycle of
metallic lead; and 3) it can produce TEL as well as alkyl lead
compounds. The first processing step is the preparation of the Grignard
reagent. Agitated propane-cooled reactors receive metallic magnesium
that reacts exothermically with fresh and recycled alkyl halide in the
presence of an electrolytic solvent consisting of a mixture of ethers
such as tetrahydrofuran and diethylenegylcol dibutyl ether. The yield
of alkylmagnesium halide is over 98X. The effluent of the electrolysis
cell is sent to a stripper, where a separation of alkyl halide and alkyl
lead is performed.
The U.S. tetraethyl lead capacity and the estimated economics for tetra-
ethyl lead production are presented in Tables IV-62 and Iv-63.
232
-------
DuPont
Ethyl
Houston Chem.
Nalco Chem.
Total
Table IV-62
U.S. Tetraethyl Lead Capacity
Plant Location
Antioch, Calif.
Deepwater, N.J.
Baton Rouge, La.
Houston, Texas
Beaumont, Texas
Houston, Texas
Est. 1970 Capacity
(Million Pounds/Year)
390
100
895
Table IV-63
Estimated Economics for Tetraethyl Lead
(kQ. MM Ib. plant)
Total Fixed Captial=$10.0 MM
Estimated Operation Cost
Cost,
. TEL
Ethyl chloride
Sodium
Lead (17$/Ib.)
Utilities
Labor and overhead
Capital charges
Total
k.S
3.8
U.k
1.5
1.6
31.5
233
-------
SUBCATEGORY_C
Product P.£2£§§§
Coal Tar Products Coal Tar~Distillation
Coal tar is a mixture of many chemical compounds (mostly aromatic) which
vary widely in composition. The process of coal tar distillation
separates these fractions into commercially valuable products.
In the plant visited, crude coke-oven tar is fractionally distilled into
solvent, carbolic oil, road tar, creosote, and pitch fractions. These
products are then purified or further fractionated into fine products.
The processes of coal tar distillation, anthracene refining, pitch
forming, and naphthalene refining, together with their associated waste
water sources, are briefly described in the following paragraphs;
simplified process diagrams are presented in Ficrures IV-41 through IV-
44.
Crude coke-oven tar and dilute caustic solution are fed into a
dehydration column. The vapor stream taken overhead from the column is
condensed, and water is removed from the solvent and discharged to a
sewer line. The liquid stream is sent then to a vacuum still and to a
series of fractionators where crude carbolic oil, road tar, creosote,
and pitch fraction are generated. There are two steam jets associated
with the distillation columns; the condensates of these jets contain
organic contaminants and are the major water pollution sources.
In the anthracene refining process, creosote is first washed with water
in a crystallizer, and the creosote anthracene slurry is passed through
filters and centrifuges to produce crude anthracene. The crude product
is then sent to a crystallizer, where furfural is used to purify the
product. Refined solid anthracene is obtained after solid separation
and drying steps. The liquid streams from the second-stage purification
units are collected for furfural recovery. The acqueous stream
discharged from the first-stage purification unit and the condensate of
the steam jet associated with the furfural recovery unit are major waste
water sources. The liquid pitch from tar distillation is cooled by
direct contact with water and then dried to form the final product. The
contact cooling water is another major waste water source.
The first step in naphthalene refining is extraction of topped carbolic
oil with a caustic solution. The bottom layer in the extractors is the
by-product of carbolate. The upper aqueous layer in the extractors is
sent to a series of stills where naphthalene and intermediate products
are generated. The only water pollution source is the condensate of the
steam jets which are used to produce vacuum in the naphthalene stills.
End-of-pipe treatment and in-plant abatement have been achieved: segre-
gation of clean water from process waste water, replacement of
234
-------
IU
«
O
z
-------
o
z
CN -j
i SI
> UJ
UJ LU
55
O u
•*
OS CJ CJ
UJ UJ ^
t— ec t—
-------
o
z
z
U_
UJ
at
UJ
Z
00
UJ X
C* Q.
3 <
O Z
"- Q
Z
Z
X
UJ
-------
L
ss
O 5
g
238
-------
barometric condensers with indirect condensers, installation of phenol
recovery units, etc. These modification have resulted in a low RWL.
The characteristics of waste water obtained from the plant survey are
shown in the following tabulation:
Coal Tar Pitch
Distillation £201123
405.3 gallons/1,000 gallons 126.1 gallons/1,000 Ib
COD 2,570 mg/1 61 mg/1
8.68 lb/1,000 gallons 0.064 lb/1,000 Ib
BODS 833 mg/1
2.81 lb/1,000 gallons N.A.
TOC 3,010 mg/1
10.16 lb/1,000 gallons N.A.
The historical data provided by the plant indicate that pitch forming
has a waste flow of 200 gal/1,000 Ib of product, with 0.13 pounds of COD
while the naphthalene refining has a waste flow of 408 gal/1,000 Ib of
product, with 0.86 pounds of COD.
Although there is a variation between the survey and the historical
data, the raw waste loads derived from the above-mentioned abatements
can be considered as representative of BPCTCA control technology of each
individual process. However, standards for BATEA and BADCT should
require that the remaining barometric condensers be converted to
indirect condensers. Thus, the quantities of waste water from the
processes of coal tar distillation and naphthalene refining can be
reduced, although RWL may not be correspondingly reduced.
239
-------
SUBCATEGORY_D
Product ££ocess_
Dyes and Pigments Batch~Manuf acture"^
The manufacture and use of dyes and pigments constitute an important
part of modern chemical technology. Because of the variety of products
that require a particular material to give maximum coverage, economy,
opacity, color, durability, and desired refluctance, manufacturers now
offer many hundreds of distinctly different dyes and pigments. Usually
dyes are classified according to both the chemical makeup and the method
of application. The manufacturers look at dyes from the chemical
aspect, and arrange and manufacture them in groups, usually of like
chemical conversions, while the users of dyes group them according to
the methods of application. Table IV-64 lists the principal types of
dyes by application classification, and Table IV-65 by chemical
arrangement. The selected pigments and their corresponding production
figures are presented in Table IV-66.
The raw materials for the manufacture of dyes are mainly aromatic hydro-
carbons, such as benzene, toluene, naphthalene, anthracene, pyrene, and
others. These raw materials are almost never directly useful in dye
synthesis. It is necessary to convert them to a variety of derivatives,
which are in turn made into dyes. These derivatives are called
intermediates. However, the industries which utilize either raw
materials or intermediates to produce final-product dyes are all
subcategorized as the dye industry.
Because of the large number of compounds that are required, often in
limited amounts, most dyes, -if not all, are manufactured in batches.
Since the purpose of this project is to investigate process-related
waste water generation sources rather than to examine detailed unit pro-
cesses/operations of manufacturing processes for each class of dyes/
pigments, a typical manufacturing process for dyes is given to
illustrate the waste water sources.
A typical process flow sheet for manufacture of azo dyes is presented in
Figure IV-45. Raw materials (which include aromatic hydrocarbons,
intermediates, various acids and alkalies, and solvents) are
simultaneously or separately fed into the reactor, where the reaction is
carried out ordinarily at atmospheric pressure. Because the reactions
are exothermic, adequate temperature control is required to avoid side
reactions. Temperature control is accomplished primarily by direct
additon of ice to the reaction tank. When the reaction is complete, the
dye particles salt out from the reaction mixture. The vent gases taken
overheads from the reactor are continuously passed through a water
scrubber before being discharged into the atmosphere. The liquid
effluent from the reactor is then sent to a plate-and-frame filter press
where the dye particles are separated from the mother liquor. The
240
-------
uj
i
GO
CO
s
I
O
55
i
i—
GO
GO
GO
Tin
S
ce
^
e_D
G
Sea
—
-------
Table IV-64
U. S. Production of Dyes,
by Classes of Application, 1965
Sales
Production,
Class of application 1
Total
Acid
Azoic dyes and components:
Azoic compositions
Azoic diazo components, bases
(fast color bases)
Azoic diazo components, salts
(fast color salts)
Azoic coupling components
(naphthol AS and derivatives
Basic
Direct
Di sperse
Fiber-reactive
Fluorescent brightening agents
Food, drug, and cosmetic colors
Mordant
Solvent
Sulfur
Vat
All other
,000 Ib.
207,193
20,395
2,100
1,558
2,835
) 3,172
10,573
36,080
15,514
1,586
19,420
2,923
4,745
9,837
18,648
57,511
296
Quantity,
1,000 Ib.
189,965
18,666
2,043
1,310
2,646
2,429
9,553
33,663
13,522
1,558
18,284
2,736
4,246
8,930
17,471
52,439
469
Value,
$1,000
292,284
39,025
3,968
2,057
2,683
4,669
23,907
50,970
32,878
6,744
34,516
10,238
5,706
15,351
9,960
48,728
884
Unit
value,
Per lb.$
1.54
2.09
1.94
1.57
1.01
1.92
2.50
1.51
2.43
4.33
1.89
3.74
1.34
1.72
0.57
0.93
1.88
Source: Synthetic Organic Chemicals. U. S. Tariff Commission
242
-------
Chemical class
Total
Anthraquinone
Azo, total
Azoic
Cyanine
Indigoid
Ketone imine
Methine
Nitro
Oxazine
Phthalocyani ne
Quinoline
StiIbene
Sulfur
Thiazole
TriaryImethane
Xanthene
All other
Table IV-65
U.S. Production and Sales of Dyes,
by Chemical Classification, 1964
Sales
Production ,
1,000 Ib.
184,387
41,661
57,897
8,787
373
5,729
731
1,074
720
172
1,987
637
18,488
17,776
462
5,607
1,312
20,974-
Quantity ,
1,000 Ib.
178,273
40,675
57,367
7,399
362
6,144
782
974
679
144
1,868
519
17,640
17,268
480
5,312
: 737
19,923
Value,
$1,000
264,023
66,889
96,579
12,149
1,113
3,302
1,614
3,367
1,258
601
4,800
1,658
29,166
9,798
1,043
12,682
3,473
14,531
Unit value
per lb.$
1.48
1.64
1.68
1.64
3.07
0.54
2.06
3.46
1.85
4.17
2.57
3.19
1.65
0.57
2.17
2.39
4.71
0.73
Source: Synthetic Organic Chemicals, U.S. Tariff Commission
In 1965 total dye production increased 12.5% to
207 million Ib.
243
-------
Table IV-66
Production or Shipment of Selected Pigments in the United States, 1958 and 1963
_ Short tons _
Pigments 1958 1963
1
J
Titanium pigments, composite and pure (100%) ^03,86? 555,211*
White lead, except white lead in oil:
Basic lead carbonate
Basic lead sulfate
Zinc oxide pigments:
Lead-free zinc oxide 130,075 162,281*
Leaded zinc oxide 23,127 12,281*
Lithopone
White extender pigments:
Barites, etc. (excluding whiting) 823,625
Whiting (calcium carbonate) 28,393 158,773
Color pigments and toners (except lakes), chrome colors:
Chrome green 3,907 2,867
Chromium oxide green ^,820 6,^73*
Chrome yellow and orange 22,365 26,620*
Molybdate chrome orange 5,675 \ Q Unrv,
Zinc yellow (zinc chromate) 6,005 / yt-+uu->
Iron oxide pigments 62,923 73,251
Colored lead pigments:
Red lead 23,311 25,780
Litharge - 121,698 93,958
Iron blues (Prussian blue, Milori blue, etc.) ^,265 5,030
Blacks:
Bone black 11,^71
Other blacks (carbon black) 1,138,500*
Source: Chemical Statistics Handbook. 5th ed., Statistical Summary k,
Manufacturing Chemists'Association, Washington, D.C., August, 1961.
244
-------
mother liquor is either directly discharged into sewers or sent to
another filter press to recover some of the metal salts. The filter
cake is first washed with compressed air while still in the press. The
moist cake is discharged into shallow trays which are placed in a
circulating air drier, wherein the moisture is removed at temperatures
between 50 and 120°C. Vacuum driers and drum driers may also be used.
The dried dye is ground and mixed with a diluent, such as salt, to make
it equal in color strength to a predetermined standard. Dilution is
necessary because batches differ in their content of pure dye.
Uniformity is assured by dilution to a standard strength.
The great majority of dyes and pigments are manufactued by the typical
process flow diagram described. However, the manufacture of some
special dyes or pigments may require more or fewer processing steps.
For example, in the manufacture of alkali-blue pigment, the process
requires a steam ejector to produce vacuum for the batch reactors. The
barometric condenser is then used to condense the exhaust steam. In the
manufacture of Direct Blue 6 dye, the filter cake is not washed but
merely freed from the adhering liquid by air drying.
The major water pollution sources of this process are the mother liquor
from the filter press, the intermittent reactor clean-up waters, the
draw-off from the vent gas scrubber, and the housekeeping cleaning
waters. The data obtained from the plant survey are summarized in the
following tabulation. Multiple data were collected at one of the
plants, and these data were subjected to the analysis for probability of
occurrence. The results of probability analysis are also shown in the
tabulation.
Summary of Survey Wastewater Data
Flow COD BOD5TOC
gal/1,000 Ib lb/1,000 Ib lb/1,000 Ib lb/1,000 Ib
(mg/1) (mg/1) (mg/1)
Dye Sample 1 13,700 1,075 220 450
(9,400) (1,920) (3,945)
Dye Sample 2 13,700 652 126 269
(5,700) (1,100) (2,350)
Dye 21,050 175 59 60
(997) (337) (360)
Dye 1056 95,069 50 5 40
Occurrence (63) (6) (51)
245
-------
50% 95,069 1,850 79 790
Occurrence (2,331) (100) (995)
90X 95,069 3,700 156 1,580
Occurrence (4,662) (197) (1,991)
4; Pigment 124,000 4,925 1,470 819
(4,764) (1,422) (792)
Because of frequent changing of feed materials and desired products,
dyemaking requires large amounts of water and of cleaning aids (such as
detergent and bleach) to clean up reactors and filter presses on each
reaction cycle. Chemical reactions involved are often exothermic and
require strict temperature control. Due to the necessity of rapid
cooling in order to avoid side reactions, direct cooling with ice, in
addition to jacket cooling, is commonly practiced, and this also
contributes a significant amount of waste water. While the high organic
loading in the waste water is primarily the result of incomplete
crystallization and separation of dye products from the mother liquor,
organic losses and cleaning aids from clean-up operations also
contribute. Different from other organic chemical industries, jacket
cooling water is required to be discharged into sewers to dilute the
waste water to be treated.
Reuse or recycle of waste water from this type of process is deemed un-
feasible, because the waste waters are contaminated with many different
salts, metal ions, and a high intensity of color, which will in turn
contaminate the product.
246
-------
SECTION V
WASTE CHARACTERIZATION
In order to develop production based effluent limitations and
performance standards (expressed as unit weight of pollutant per unit
weight of product), it is first necessary to define a raw waste load
(RWL) for the process. Appropriate reduction factors can then be
applied to the RWL to establish the desired production based
restrictions.
The choice of the specific pollution parameters for which restrictions
are to be recommended is to a large extent governed by existing
conventions which have been established within the water pollution
control field. Although it would be desirable to identify the specific
chemicals which are present in the waste water streams associated with
the organic chemicals industry, many of these would be present in the
waste water from only a few processes so that the development of
generalized restrictions which are applicable to large categories would
not be possible. For this reason conventional general parameters
related to oxygen demand, toxicity, turbidity, color, and taste were
examined during the course of this study.
The waste water associated with each process was differentiated
according to whether it was considered as contact waste water or non-
contact waste water. It is impossible to equitably define production
based RWL for the noncontact water streams This is caused by the fact
that these streams are always associated with a number of different
processes with no equitable means available for allocating the
pollutants which are present.
In a typical chemical process plant, utility functions such as the
supply of steam and cooling water are set up to service several
processes. Boiler feed water is prepared, and steam is generated in a
single boiler house. Noncontact steam used for surface heating is
circulated through a closed loop whereby varying quantities are made
available for the specific requirements of the different processes. The
condensate is nearly always recycled to the boiler house, where a
certain portion is discharged as blowdown.
Noncontact cooling waters are also supplied to several processes. The
system generally is either a closed loop utilizing one or more
evaporative cooling towers, or a once-through system with direct
discharge.
The amounts of blowdown from boilers and cooling towers are not directly
related to individual processes but depend rather on the design of the
particular plant utility system. Although noncontact steam and cooling
water requirements were presented for the processes which have been
247
-------
examined, the quantities of blowdown associated with utility recycle
loops cannot be correlated back to individual processes. Similarly, the
amounts of waste brine and sludge produced by ion exchange and water
treatment systems cannot be allocated among the individual processes
within a plant.
The quantities of pollutants such as dissolved solids, suspended solids,
alkalinity, and other parameters which are associated with the
noncontact streams and water treatment equipment were not included in
the calcualtion of the production based RWL for each process.
Subsequently, no production based limitations or standards are
recommended for these parameters at this time. Studies currently
underway will establish bases for development of effluent limitations
for noncontact waste waters at a future date. Instead/ contact process
waste water streams formed the basis for all RWL calculations included
in this study.
The RWL data to be presented in this section was based on past
historical data supplied by some of the manufacturers surveyed as well
as actual data obtained by sampling.
The RWL for each process was Calculated by taking 24 hour composite
samples of the contact process waste water streams. The pollutant
concentrations obtained from the analysis of these samples were
multiplied by the associated waste water flow during the same 2H hour
period to give pollutant generation rate as Ib per day. These
generation rates were divided by the corresponding production to provide
a series of production based RWL's.
It should be noted that many of the processes examined generate
nonaqueous wastes. These may be liquid or semi-liquid materials, such
as tars, or gaseous materials, such as by-product hydrocarbon vapors.
As such, these wastes are normally burned as auxiliary fuel or are
disposed of in some way that is unrelated to the contact process waste
water. These materials were not included as part of the RWL calculated
for the processes examined.
The RWL for a specific process module is based on the actual production
rate of the principal product and the measured contact process waste
water flow. Co-products are not included in the RWL calculation unless
they have specific waste waters asscoiated with their own purification
or processing. An example of this situation is the RWL associated with
butadiene as a co-product of ethylene manufacture. In this case,
butadiene purification has a specific waste water flow and loading;
therefore, a separate RWL has been defined.
Dissolved oxygen demanding material was found to be the major pollutant
associated with production operations in this industry. Standard Raw
Waste Loads (SWRL) , expressed as average or median valuesf have been
248
-------
developed fcr the industrial subcategories. Four major parameters were
considered:
1. Process Wastewater Flow Loading
(expressed as liters/kkg and
gal/1,000 Ibs of product)
2. BOD5 Raw Waste Loading
(expressed as kg BOD5/kkg and
Ib BO 5/1,000 Ib of product)
3. COD Raw Waste Loading
(expressed as kg COD/kkg and
Ib COD/1,000 Ib of product)
4. TOC Raw Waste Loading
(expressed as kg TOC/kkg and
Ib TOC/1,000 Ib of product)
The RWL data relating to individual manufacturing processes were first
grouped according to the subcategory in which the process is assigned.
The data for the processes within each subcategory were then plotted as
pollutant raw waste loading versus contact process waste water flow
loading. These plots are shown in the following figures:
Subcategory A
BOD5 vs. Flow (Figure V-1)
COD vs. Flow (Figure V-2)
TOC vs. Flow (Figure V-3)
Subcategory B
BOD5 vs. Flow (Figure V-4)
COD vs. Flow (Figure V-5)
TOC vs. Flow (Figure V-6)
Subcategory C
BOD5 vs. Flow (Figure V-7)
COD vs. Flow (Figure V-8)
TOC vs. Flow (Figure V-r9)
Subcategory D
BOD5 vs. Flow (Figure V-10)
COD vs. Flow (Figure V-11)
Since both the loading (ordinate) and flow (abscissa) are expressed on a
production basis, dividing the loading by the flow gives a slope which
may be expressed as a concentration. For orientation, reference lines
of constant concentration have been drawn diagonally across each of the
plots. Relating a specific data point to one of these lines provides a
convenient estimate as to the raw waste concentration. Although
249
-------
insufficient RWL data were obtained to establish definitive increasing
relation between loading and flow the additonal RWL data may provide
confirmation of such a relationship.
The five manufacturing processes examined in Subcategory A were
described in the previous section. No clear range or SRWL can be
defined for this category. This may partially be caused by the fact
that external runoff, washings, and contaminated spray cooling water
amount to a significant portion of the waste water flow in each case.
One of the major difficulties in obtaining meaningful RWL data for
Subcategory A processes is the fact that a large portion of the waste
water comes from sources which are difficult to sample or where
pollutant loadings result from contact with chemicals on the ground.
Unlike other process subcategories where specific process pipes or
sewers can be used to sample and measure all process flows, Subcategory
A waste waters are intermittently dumped directly into open ditches or
common sewers within the process area. In some cases, Subcategory A
waste waters flow by gravity to holding tanks where batch treatment is
provided; in other cases, they are discharged directly into the overall
plant treatment system.
There is also a question as to whether the continuous water washes are
truly representative of the process or are necessitated by a specific
feed impurity (ethyl benzene) or nonaqueous absorbent (Benzene, Toluene,
Xylene recovered by solvent extraction) used by the particular
manufacturers sampled.
When compared with the range of pollutant loadings presented for the
other subcategories, it is apparent that those from Subcategory A are
generally lower. The RWL for Subcategory A products-processes are
summarized in Table V-1.
During Phase II, an additional effort will be made to supplement the
date for this category by sampling numerous processes over long periods
of time. This will eliminate some of the difficulties associated with
sampling and measuring the sporadic flows.
The individual process RWL data for Subcategory B are plotted in Figures
V-4 through V-6. General increasing trends between pollutant RWL and
flow RWL appear to exist within the category.
The BOD5 RWL for 13 Subcategory B processes generally falls in a
concentration range of 100 to 500 mg/1. Loadings vary from 0.09 to 7.0
Ib COD/1,000 Ib of product. The corresponding range of flows increases
from 50 to 3,000 gal/1,000 Ib of product. It should be noted that two
of the processes in Subcategory B ethylene dichloride (EDC) manufactured
by the chlorination of ethylene, and vinyl chloride monomer (VCM)
manufactured by the purolysis of EDC product contact process waste
waters which are not amenable to the BODS test. This was caused by
250
-------
co
LU
>
CO
LU
O
O
CC
0.
o
LU
_0
TO
OC
o
(U
ro
<
c
O
4-1
C
O
ro
O
O
ro
O.
-a
ro
O
d)
(D
ro
a:
o
CD
1—
Q
O
O
Q
0
CO
3
0
,—
LL.
c
o
4-J
Q
•—
1_
O
1/5
0>
CD
(ft
>
~
X
CD
c
ro
X
(U
-C
o
^~
u
o
LA
v£>
O
vO
00
•
•—
CvJ
ro
^~
•
O
r^
•
i^
ro
O
•
LA
V-
m
jr
4J
2
(U
c
0)
N
C
C
ro (U
>^ >~
— ^ -f~
1 — 4-*
^
-C
LU
vO
0
o
CM
, —
•
0
1
1
f — ^
o
_^
CM
O
•
_^
O
o
CM
0)
C
(D
—
4-1
(U
O
<^
o
4-1
1 —
(_J
3:
<4-
o
c
o
4->
._
"Q
-Q
"*
(U
• —
1_
o
JC
CJ
*_
^
C
>
OA
o
o
r_
OO
•
o
_.J.
0
r—
«
O
vO
*
v+\
*—
O
•
J-
r—
Q)
C
...
P_
0
t/i
ro
CJ3
1/1
•—
i/i
>,
~-
O
L.
>^
Q-
cn
c
. —
4->
ro
i_
•i->
O
s_
-D
X
(/)
O
4-1
ro
L.
^
X
H~
CD
-*
o
i
i
i
i
i
^?
•
o
\D
o
•
-3"
0
LA
0)
4-J
ro
e
u
0
y-
0)
c£
E
O
u
14-
c
o
• _
4-1
O
ro
4-J
X
LU
4J
C
(U
t_—
0
CO
t/)
o
4-1
ro
£1
Q
L.
^
X
1
p™*
CO
CM
OJ
O
VO
e
O
CM
«~-
•
0
^— s
o
r^
co
oo
LA
LU
o
^^
QC
LU
>
<
^O
fv-l
0
r—
f^
•
0
CM
,—
•
O
s — ^
0
MD
_^-
O
LA
Z
^
—
Q
LU
31
251
-------
FIGURE V-1
RELATIONSHIP BETWEEN BOD RWL AND ROW RWL FOR CATEGORY A
nl r-
10" -
101
m
«
- KH
/
/
LEGEND
1 8TX MOMATICS
2. ETHYL BENZENE
• SURVEY SMFLING DATA
an ML (GN.. / I03 IBS PRODUCT)
t«s io4
I 1
HOI OH (LITER/103 KILOSRW PRODUCT)
252
-------
FIGURE V-2
RELATIONSHIP BETWEEN COD RWL AND FLOW RWL FOR CATEGORY A
10°
g
cs
3
10-1
10-2
LEGEND
1 BTX AROHATICS
2 ETHYL BENZENE
3 VINYL CHLORIDE
• SURVEY SAMPLING DATA
ID2 103
FLOI ML (EH./ I03 LBS. PRODUCT)
I I
I
FLOW RM. (LITER/103 KILOGRM PRODUCT)
253
"OD,:T,
-------
FIGURE V-3
RELATIONSHIP BETWEEN TOC RWL AND FLOW RWL FOR CATEGORY A
ID1 i- 101
(- 100
10"
>|_ 10-'
L_ 10-2
«>
$s
y
/
<&
f
/
/
• \
<&
®s
/
/
/
/
/
\- /
/
4
y
/
/
/
/
i
/
y
y
/
/
/
/
•
i
/
/
/
f
V
s
S
/
/
/
/
/
r
/
/
*3
/
/
/
/
/
^
/
/
/
/
/
ft
A
'
/
/
/
/
/
/
/
/
LEGEND
\ m momms
1 ETHYL BENZENE
3 VINYL CHLORIDE
/
• SURVEY SWdPLINS DUTA
/
/
/
O1 102 103 1
FLOI RIL (G»LS/103 LBS PRODUCT)
1 I 1 1
10 2 10 3 10 *
FLOI RIL ( LITER/103 KILOGRAH I-RJDUI.,)
254
-------
objectionable conditions related to the high concentrations of wastes.
In such cases, the wastes may still be degraded biologically, but
require dilution with other less concentrated wastes or non-contact
cooling water.
The COD RWL concentrations for 16 Subcategory B processes are bewteen
100 and 5,000 mg/1. Loadings vary from 0.5 to 21.5 lb/COD/1,000 Ib of
product within the same range of flows as presented for the BOD5 RWL.
The TOC RWL concentrations for 16 Subcategory B processes are generally
between 100 and 2,000 mg/1. TOC loadings vary from a minimum of 0.2 to
a maximum of 40 Ibs TOC/1,000 Ib of product.
There is no definite correlation between the BOD5 and COD RWL within
Subcategory B. COD/BOD5 ratios generally vary between 2/1 and 10/1.
This is understandable since there is still a wide variety of specific
chemicals which may be present in the waste waters from this process
category.
The wide spread in RWL data obtained for Subcategory B has led to the
establishmnet of two subcategories designated as B1. and B2_. The
individual products, processes, and associated RWL asllocated to each
subcategory are indicated in Table V-2. It can be seen that the average
flows and RWL for the two subcategories conform to the general
relationship of increased loadings being associated with increased
flows.
The individual process RWL data for Subcategory C are plotted in Figures
V-7 through V-9.
As with Subcategory B, there appears to be an increasing trend between
BOD5 RWL and flow RWL. This relation is not nearly so definitive for
the COD and TOC parameters.
The BOD5 RWL for the Subcategory C processes generally fall in a
concentration range of 3,000 to 10,000 mg/1. Loadings vary from 1.3 to
125 Ib BOD5/1,000 Ib of product. The corresponding range of flows
increases from 30 to 3,000 gal/1,000 Ib of product.
The COD RWL data for the subcategory C processes are between 10,000 and
50,000 mg/1. Loadings vary from 5.5 to 385 Ib COD/1,000 Ib of product
within the same range of flows as presented for the BOD5 RWL.
The TOC RWL concentrations for the subcategory C processes are generally
between 3,000 and 15,000 mg/1. TOC loadings vary between 1.5 and 150
lb/1,000 Ib of product. An envelope drawn around the TOC data
commensurate with the BPCTCA technology is shown in Figure 1-7.
As with Subcategory B, there is no definite correlation between the BOD5
and COD RWL within this subcategory. COD/BOD5 ratios generally vary
255
-------
o
00
^ LA LA oo
CO CO ^— rf\ • O •— LA
• • • a \^ • • •
O O O O — O CA O
I OO n~\ I O^ ro cv) rA
I T— OO O~\ I OO on LA •
I OA . . i . . . vO OO
ICM.— r>JlOOOCN rA
.— O
PA LA OO I -3" v£> rACXD
CM
|
^>
0)
-Q
1-
Z
LU
CO
C£
O
CO
CO
**
Q£
o
[ 1
/"*\
l-
z
LU
I-
^
_J
Q
S
<
LU
1-
co
CO
<
1—
<
o
0
cc
LU
1—
•~*
CO
CO
LU
O
Q-
1
CO
^.
(£.
0
CD
LU
|—
<
CJ
(U
-Q
TO
—
TO
>
>^
4-J
C
- TO
Ol O
O
— -a
O TO
C O
-C _J
U
a) 3
C TO
O a:
o
**-s
Ol
_*!
o
o
o
^«
v^
_J
in
0)
in
in
(U
o
o
a.
^^
4-J
O
3
T3
0
Q-
1"
ca
PA O O~\ •— 1 •— PA —
CM CM O •— 1 O i— —
LA PA CTVVO 1 J" -3" LA
PAVO -4" CM 1 O PA PA
• •«•!• • *
O O O O 1 O O O
LAPAOLAOOO O~V OO
LAOLAr-^.-4-CM LA LA
PA CM •— — r- r-
-4-LAOOr-cri-d- oo cn
vjOOA»— vOvOrA CM •—
O^i VD -4" -4" — CM PA PA
CM — t- — — —
LU
in — o z
TO O <£ «f
C3 C CC —
TO LU O
— Q.-— > LU
TO O O T3 < Z
U t- U C C
O- 3 Q. TO TO
— 1 0) 4-i O -C
\ C TO ifl 4-J (U
TO a> z — LU c
-C — 0)
4J >* 4- 4- 4- —
-C -C O O O >«
a 4-i j=
TO LU Ol C C 4-> T3
z c o O LU —
4- — — — 0
4- O £ 4J 4-1 4- <
O f- TO TO O
4-> O C C U
in O it- 0) 0) in ~
.— 3 0) Ol O1-— 4J
in T3 oC O O in a)
>- O I- U 0) U
— »- E T) -a .c <
o a. ra >- >^ 4->
1- 1 , o 4-1 D •
Q- o co a a co
in
Q)
in
t/>
(U
0
0
0) l-
0) 4-1 Q_
•a TO \
» TO -E 4-i 4-> >- Q_
^; 4-1 4-< 0) 0) C
-4->3 CO
CM LA tTl-d- 1
PA ^1 LA v£) !
vQ CM O 1
cr\ O"\ r^^ i
• CM • • 1
CM 1 — t— O 1
O i— i— -4" «—
^O LA vJD t*^* PA
— .4- LA .-
*~s
v£) VD -4" OO -*
oo — oo — a\
cr\ --d" •—
CM
^~
0)
c
TO
4-1
3
CO
4-
--C
k- 4-1 £ 4-i
C "D IU 4-i (U
o >*z: LU z
— -C
4-1 c c c
01 > O O 0
o — — — —
1_ 4-) 4-) 4J 4-1
"O TO TO TO TO
-C — — — —
•
0) (U - • — TO
TO TO 4-> >• E
4-J 4-> (D -C L.
M 3 3 O 4-J O
CQ CO < LU U-
00
J-
o
X
o
I—
VJD
O*\
CM
O
0)
C
.
JC
4-J
LU
4-
o
C
O
•—
4-1
TO
C
• ~.
U
0
f
0
4_i
O
u
•«•
Q
(D
T)
>^
l_
0
^~
£
u
• —
o
(U
c
(U
*—
>
-C
4-1
LU
vD
1^-
O
X
o
vD
PA
PA
vD
O
OO
CM
0)
XI
U
_o
x:
o
0
N
c
•—
>
«— •
CM OO
vD -^
• •
— o
o o^\
— PA
OO -3"
CM * — '
-4" CM
VO OO
-4" i-f\
• PA
PA
CM
.
x: O
4J 4J
LU
TO
II ,_„
O C
C E
O E
*"~ ^
4-J
TO 4-
C O
0)
01 C
O 0
•O 4->
-E T)
(U T3
Q <
in
IU
C
•—
1
0)
c —
Q) >-
V- -C
>~ 4J
4-1
CM
vO
p*^
p-^
•
, —
^
CM
-d"
CM
oo
LA
PA
X
<
a
LU
^
256
-------
FIGURE V-4
RELATIONSHIP BETWEEN BOD RWL AND FLOW RWL FOR CATEGORY B
8 HT1
10-2
10«
§
IB-'
^x
t^yw
«$X
/
N^/
y
^f
.&J/L-
y
/
f
/
/
y^
y
/
/
y
/
/
/
11
/
/
^
/
y
u
V
/
x
^
'
f
/
,
/
^
/
I/1
'
^
/"
» . /
/
/
/
s
S
/
\$s
s
,/
f \
<
>3
t*7
• 9
3
/
/
s
/
/
/
s
2/
4
/
/
12
6.
I
/
/
/
/
«
M
/
/
/
13
•
/
y
^/
y
5
8
(
1 1
y
.
/
'
/
/
'
I/1
/
2
3
4
5
6
7
B
9
ID
12
13
1
/
r 4 y
/
i
/
y
- S
/
I
*
/
/
/
f
5
f
L E G E N D
\ \
1
i
ZTI ix
IA
/
A
/\ •
i
] j
,_ t__i_
1
\
t
t
_^
i
H-
»CETALDEHY OE FBOK ETHANOL
OXIDATIVE-DEHYOROCENATION
ACETONE FBDM IP* DEHYDROPENAT I ON
BUTADIENE FROM C2H4
EXTRACTIVE DISTILLATION
BUTADIENE FROM N-BUTANE
OEKYOROGENATION (HOUDBY)
STYRENE FROM E B
ETHYLENE FROM C2H5
ETHYLENE FROM LPG '
ETHYLENE FROM NAPHTHA
ETHYLENE FRO* CjHj
ETHYLENE OXIDE
METHANOL
METHYL AMINES
ACETYLENE
SURVEY SAMPLING DATA
90% OCCURRENCE
HISTORICAL PLANT DATS
50% OCCURRENCE
10% OCCURRENCE
102
FLOI ML (6AL/I03 LB PMDUCT)
103
10 J
FLOI RIL (LITER/103 KILOGRAM PRODUCT)
257
-------
FIGURE V-5
RELATIONSHIP BETWEEN COD RWL AND FLOW RWL FOR CATEGORY B
LEGEND
HCETAUEin DE FRO* ETHkNOL
OXIDATIVE-DEHYDROGENATION
9. ETHYLENE FROM C3H8
10. ETHTLENE OXIDE
SURVEf SAMPLING DATA
90% OCCURRENCE
nl -
nO -
S
10-' L
2. ACETOHE FROM IPA DEHYOROGEHATI OH
II METHANUL
3. BUTADIENE FROM C2H4 |? |ETm ,„,,,„
EXTRACTIVE DISTILLATION ,,_ ulmiK
4. BUTADIENE FROM N-8UTANE
DEHYDROGENATION (HDUDRY)
5 STYRENE FROM E. B.
6. ETHYLENE FROM C2H5 15' EDC FR1)M C2H4
J. ETHYLENE FRO* LP G 16 Yc* fm w
8. ETHYLEIE FROM NAPHTHA
to2
- 10'
»_
o
o
S
O-
m
tn
O
|
u
oa
_i
_j
i
_ 10-1
10 •
.
(
HISTORICAL PLANT DATA
50% OCCURRENCE
10% OCCURRENCE
^
/
/
/
/
/
/
/
/
1
T^
y
y
/
y
yf
/i
10' 102 103
FLO» RIL (GAL/103 LBS PRODUCT)
II 1 1
/
/
f
^
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
/
71
/
>; -
10
i I
10° 10*
FLO! RIL (LITER/103 KILOGRAM PRODUCT)
258
-------
FIGURE V-6
RELATIONSHIP BETWEEN TOC RWL AND FLOW RWL FOR CATEGORY B
102
10'
10°
n-2
10-3
10-2 -
10-3
FLOW RWL (GAL/'IO3 IBS PROOICT)
102
103
FLO* R*L (LITER/103 KILOGRAM PRODUCT)
104
259
-------
Product/Process
Table V-3
Category C - Aqueous Liquid Phase Reaction Systems
Best Practicable Control Technology Currently Available
Process Description
Flow
r. Prodiirt /Proc£«i«l
1
Acetaldehyde
Acetaldehyde
Acetic Acid
Acrylic Acid
Aniline
Bis Phenol A
Caprolactam
Coal Tar
Dimethyl Terephthalate
Ethylene Glycol
Oxo Chemicals
Phenol
Terephthallc Acid
C. Product Processes
~~ £
Acrylates
p Cresol
Methyl Methacrylate
Terephthallc Acid
Tetra Ethyl Lead
Oxidation of Ethylene with Air
Oxidation of Ethylene with Oxygen
Oxidation of Acetaldehyde
Synthesis with Carbon Monoxide
and Acetylene
Nitration and Hydrogenat ion of
Benzene
Condensation of Phenol and
Acetone
Oxidation of Cyclohexane
Pitch Forming
Distillation
Esterification of TPA
Hydrogenat ion of Ethylene Oxide
Carbonylation and Condensation
Oxidation of Cumene
Ox 1 da t lop of P-xylene
Average
Median
Esterification of Acrylic Acid
Salfonation of Toluene
Acetone Cyanolrydrin Process
Nitric Acid Process
Addition of Ethyl Chloride to
Lead Ama 1 gam
Average
Median
1 iters/1000 kg
752
509
4,175
3,966
1,586
559
10,855
1,01*1*
3.340
2,254
4,876
3,507
2,338
1,553
2,973
2,338
15,280
10,780
2,171
5,503
100,000
28,1*99
10,280
(gal/1000 #)
(90)
(61)
(500)
(475)
(190)
(67)
(1,300)
(125)
(400)
U70)
(584)
(420)
(280)
(186)
(356) -
(280)
(2,856)
(1,291)
(200)
(659)
(12,000)
(3.413)
(1,291)
y
kg/ 1000
26.6
1.9
0.35
0.74
—
—
1.64
2.8
24.45
0.34
3.2
5.6
.82
5.83
• 1.9
47
123
45
59
—
68
53
fig or lb/1000
44
5.8
0.78
1,64
21.2
17.1
4.0
0.06
8.7
38.2
8.76
4.25
II. 0
1.72
11.3
6.5
118
256
386
104
110
195
118
Ib
2.8
5.5
—
1.53
19.2
5.13
—
10.2
13.8
4.52
1.92
0.45
0.86
6.38
3.6
79
54
152
45
5.6
67
54
260
-------
FIGURE V-7
RELATIONSHIP BETWEEN BOD RWL AND FLOW RWL FOR CATEGORY C
io* r 10"
1Q3
102
10'
10°
10-'
r~~ 1 1 i i i T i i i 1
LEGEN
1
2
3
4
5
6
7
9
10
12
0
METHYL METHACRYLATE
TEREPHTHALIC ACID "POLYMER GRADE"
DIMETHYL TEREPHTHALATE
TEREPHTHALIC ACID INDUSTRIAL GRADE
ACRYLATES
TETRAETHYL LEAD
PHENOL/ACETONE
ACETALOEHYDE
ACETIC ACID
ANILINE
ETHYLENE GLYCOL
SURVEY SAMPLING DATA
rS^i/
~ ^5
S
/
9/
*/
y
/ t
i^^X
j^!/^
/^
/
^
X
/
/
y
^
/
'
/
*\
/
/
f
/
'
/
4
/
/
/
/
y
i
/\
>
*3
/
y
j
£
/
,/
/
s
/
/
\
/
/
/ V
' 7 1
/
/ /
/ /
f
/
10
/
s
1*
/
/I
/
r
^
f
/
/
/
3j
3*
/
f
x
x
^
x
K'
•^
/
1
/
/
/
/
/\
.
/
I
J
1
/
/
^
2
^
2
/
^
/
/
f
f
? f
* 2 j/
/
/ /
' _/
/
/
/
/
/
/
E
f
f
f
/
/
/
X
/
/
/
/
^
/
/
/
/
'
£
/
_/
/
S
/
/
/
/
6
'
I
i
-4-
~\ "
i
_|. _ ___
i
±4 " ~
J -L-
P^r
r-
- — — -
r
|
10'
102
103
FLO* R*L (GAL/103 LBS PRODUCT)
10"
105
104 10s
FLO* R*L (LITEPv 103 KILOGRAM PRODUCT)
106
1C1
-------
FIGURE V-8
RELATIONSHIP BETWEEN COD RWL AND FLOW RWL FOR CATEGORY C
10*
H LEGEND
1 METHYL METHACRYLATE
2 TEREPHTHALIC ACID ''POLYMER GRADE '
3 DIMETHYL TEREPHTHAUTE
4 TEREPHTHALIC ACID INDUSTRIAL GRADE"
5 ACRYLATES
6 TETRAETHYL LEAD
7 PHENOL/ACETONE
8 ACETALDEHYDE
9 ACETIC ACID
10 ANILINE
11 SIS-PHENOL
103
1Q3
1D2
102
in i
10° .
10°
ID2 I03
FLO* RWL (GAL/103 LBS PRODUCT)
10 3 10 «
FLOW RWL (LITER/103 KILOGDAM PRODUCT)
10 s
262
-------
FIGURE V-9
RELATIONSHIP BETWEEN TOC RWL AND FLOW RWL FOR CATEGORY C
10"
-1—I—1—I I I
- LEGEND
103
103
1 METHYL METHACRYLATE
2 TEREPHTHALIC ACID POLYMER GRADE1
3 DIMETHYL TEREPHTHALATE
4 TEREPHTHALIC ACID 'INDUSTRIAL GRADE'
5 ACRYLATES
7 PHENOL/ACETONE
8 ACETALDEHYDE
10 ANILINE
II BIS-PHENOL
12 ETHYLENE GLYCOL
14 DXO-CHEMICALS
15 ACRYLIC ACID
16 CAPROLATAM !
17 PARA-CRESOL
• SURVEY SAILING DATA
Wi OCCURRENCE
HISTORICAL PLANT DATA
1' 50', XCURRENCE
10', OCCURRENCE
102
102
1
*=
•a
101
10'
ioc
10°
IO-2L
y
y
-/
f,
y
8~ii
y
ft
X
z
^
T
~7
12 '•
• i?
a
/
i
fl^-
y
!_
10°
10'
102 103
FLO* R*L (GAL/103 LBS PRODUCT)
10-1
10'
10 3 10 4
FLO* RWL (LITER/103 KILOGRAM PRODUCT)
10 5
-------
between 3/1 and 5/1. However, some specific processes vary widely
outside this range.
There is quite a wide spread in the RWL obtained for the processes
surveyed within Subcategory C. For this reason, two subcategories
designated as Cl and C2 have been established. The specific products,,
processes, and associated RWL assigned to each subcategory are indicated
in Table V-3. As with Category B, it can be seen, that the average flows
and RWL for the two subcategories conform to the general relationship of
increased loadings being associated with increased flows.
The individual process RWL data for the batch plants in Subcategory I)
are plotted in Figures V-10 and V-11. As with Subcategory A, the data
are insufficient to establish any clear relationships between pollutant
loading and flow. The ranges of loadings and flows are quite wide.
This is caused mainly by the highly variable product mix and the
inclusions of contact cooling and cleaning waters.
It should be noted that the loadings shown for Subcategory D are based
on the entire production from the batch plant. The RWL for Subcategory
D were subjected to analysis for probability of occurrence and are
summarized in Table V-4.
264
-------
^
>-
4-1
C
,
O
^—
O
c
^z
u
V
^—
o
L.
4-1
c
o
0)
•~
TO
U
U
TO
CL
4J
in
0)
oo
TO
4J
TO
O
t)
4-"
in
TO
S
^
TO
o:
Ol J3
o| -
O
O
o
U
o
en
o
o
o
O
IA
j-
O LA LA
LA r--- r->
oo o —
O
LTv
.
A A «
LA — LA
.c
U
TO
OQ
Q|
U
0)
L.
3
U
U
o
0)
en c
TO TO
s. —
0) T3
> 0)
< z
265
-------
FIGURE V-10
RELATIONSHIP BETWEEN BOD RWL AND FLOW RWL FOR CATEGORY D
ID* r in41
|
^ io2
I- ° .02 |
9
i
10'
10' I
LEGEND
I PLASTIC 12ERS
2 DYES-PIGMENTS
• SURVEY SAMPLING DATA
JF
10° L io» I
10'
V
I02
103 104
FLOW R*L (GAL/103 LBS PRODUCT)
105
ii
10" 10"
FLO* RVfL (LITEP/103 KILOGRAM PRODUCT)
266
-------
FIGURE V-ll
RELATIONSHIP BETWEEN COD RWL AND FLOW RWL FOR CATEGORY D
ioai-
I
a
a
3
in?
in1
LEGEND
2 DYES - PIGMENTS
• SURVEY SAMPLING DATA
2 it
1Q3
105 10s
FLOU R»L (GAL ' ID3 LBS PBDDUCT)
10'
I |
FLOW RBL (LlltR/103 KILOGRAM PHODUCT)
267
-------
SECTION VI
SELECTION OF POLLUTANT PARAMETERS
An extensive literature review resulted in the selection of twenty-five
parameters which were examined during the field data collection program.
These parameters are listed in Table VI-1, and all field data are
summarized in Supplement B.
Based on the degree of impact on the overall environment, the pollutants
are divided into subcategories as follows:
Pollutants of Significance
Pollutants of Minimal Significance
The rationale and justification for pollutant subcategorization within
the above groupings will be explored. This discussion will provide the
basis for selection of parameters upon which the actual effluent
limitations were postulated and prepared. In addition, particular
parameters were selected for discussion in the light of current
knowledge as to their limitations from an analytical as well as from an
environmental standpoint.
Pollutants observed from the field data as present in sufficient concen-
trations to interfere with, be incompatible with, or pass thru
inadequately treated in a publicly owned works are discussed in Section
XII.
Pollutants^g|_Significance
Parameters of pollutional significance for which effluent limitations
were developed in the organic chemicals industry are the major organic
parameters of BOD5, COD and TOC.
SQD5
Biochemical oxygen demand (BOD5) refers to the amount of oxygen required
to stabilize biodegradable organic matter under aerobic conditions. The
BOD5 test has been used to gauge the pollutional strength of a waste
water in terms of the oxygen it would demand if discharged into a
watercourse. Historically, the BOD test has also been used to evaluate
the performance of biological waste water treatment plants and to
establish effluent limitation values. However, objections to the use of
the BOD5 test have been raised.
268
-------
Table VI-1
List of Pollutants Surveyed for the Organic Chemicals Industry
Chemical Oxygen Demand (COD)
Biochemical Oxygen Demand (BOD5)
Total Organic Carbon (TOC)
Total Suspended (Nonfilterable)
Solids (TSS)
Oil and Grease
Ammonia Nitrogen
Total Kjeldahl Nitrogen (TKN)
Phenols
Cyanide, Total
Color
Sulfate
PH
Acidity
Alkalinity
Total Dissolved (Filterable)
Solids
Chloride
Hardness - Total
Total Phosphorus
Calcium - Total
Magnesium - Total
Zinc - Total
Copper - Total
Iron - Total
Chromium - Total
Cadmium - Total
Cobalt - Total
Lead - Total
269
-------
The major objections are as follows:
1. The standard BOD5 test takes five days before the results
are available," thereby negating its use as a day-to-day
treatment plant operational indicator.
2. At the start of the BOD5 test, seed culture
(microorganisms) is added to the BOD5 bottle. If the seed
culture was not acclimated, i.e., exposed to a similar
waste water in the past, then it may not readily be able
to biologically degrade the waste. This results in the
reporting of a low BOD5 value. This situation is very
likely to occur when dealing with complex industrial
wastes, for which acclimation is required in most cases.
The necessity of using "acclimated bacteria" makes it very
time-consuming for regulatory agencies to duplicate
industrial BOD5 values unless great care is taken in seed
preparation.
3. The BOD5 test is sensitive to toxic materials, as are all
biological processes. Therefore, if toxic materials are
present in a particular waste water, the reported BOD5
value may very well be erroneous. This situation can be
remedied by running a toxicity test, i.e., subsequently
diluting the sample until the BOD value reaches a plateau
indicating that the material is at a concentration which
no longer inhibits biological oxidation.
There has been much controversy concerning the use of BOD5 a measure of
pollution, and there have been recommendations to substitute some other
parameter, e.g., COD or TOG. EPA has recently pointed out that some or
all of the previously cited reasons make the BOD5_ test a non-standard
test, and ASTM's Subcommittee D-19 has also recommended withdrawal of
the BOD test as a standard test.
However, some of the previously cited weaknesses of the BOD test also
make it uniquely applicable. It is the only parameter now available
which measures the amount of oxygen used by selected microorganisms in
metabolizing a waste water. The use of COD or TOC to monitor the
efficiency of BOD5 removal in biological treatment is possible only if
there is a good correlation between COD or TOC and BOD5. During the
field data analysis, varying ratios within each subcategory and between
subcategories were evident. This is particularly true of subcategory D
batch chemical production. After consideration of the advantages,
disadvantages and constraints, it is felt that BOD5 should continue to
be used as a pollutional indicator for the organic chemicals industry.
The BOD5 data acquired during the sampling program for Subcategories, A,
B, C, and D are presented in Figures V-l, V-4, V-7, and V-10, which
indicate the relationship between BOD RWL and flow RWL for each
270
-------
previously described subcategory. Typical RWL concentration ranges for
each subcategory are presented below:
Subcatecjorjj BOD 5 RWL Range
____ mg/1 ___
A 500-1,000
B 100-500
C 3,000-10,000
D 100-3,000
As a matter of reference, typical BOD5 values for minicipal waste waters
range between 100 and 300 mg/1.
COD
Chemical oxygen demand (COD) provides a measure of the equivalent oxygen
required to oxidize the organic material present in a waste water
sample, under acid conditions with the aid of a strong chemical oxidant,
such as potassium dichromate, and a catalyst (silver sulfate) . One
major advantage of the COD test is that the results are available
normally in less than three hours. However, one major disadvantage is
that the COD test does not differentiate between biodegradable and
nonbiodegradable organic material. In addition, the presence of
inorganic reducing chemicals (sulfides, etc.) and chlorides may
interfere with the COD test.
S5Sstandards Methods for the Examination of water and Wastewater, the
principal £§f§£§nce for anlaytical work in this field ^ cautions that
aromatic £22!E2U11^§ 1S^ straight-chain alghatic comiDOundsx fe2£h prevalent
in £he organic chemicals industry^ are not completely Qxidize_d during
the" COD test^ The addition of silver sulfatex a catalyst^ aids in the
oxidation of the straight- chain alcohols and acids but does not affect
§£omatic hy.<|E2£a.r.E2!is_i The exact extent of this partial oxidation has
b Documented "the literature.
COD RWL data for the four subcategories is presented in Figures V-2, V-
5, V-8, and V-ll. A summary of the concentration range is presented
below:
Subcategory. COD RWL^Range
mg/1
A 100-10,000
B 200-5,000
C 10,000-50,000
D 1,000-10,000
Typical COD values for municipal waste waters are between 200 mg/1 and
400 mg/1.
271
-------
TOC
Total organic carbon (TOC) is a measure of the amount of carbon in the
organic material in a waste water sample. The TOC analyzer withdraws a
small volume of sample and thermally oxidizes it a 150C. The water
vapor and carbon dioxide is monitored. This carbon dioxide value
corresponds to the total inorganic value. Another portion of the same
sample is thermally oxidized at 950°C, which converts all the
carbonaceous material; this value corresponds to the total carbon
(carbonates and water vapor) from the total carbon value.
The TOC value is affected by any one or more of the following:
1. One possible interference in the measurement occurs when
the water vapor is only partially condensed. Water vapor
overlaps the infrared absorption band of carbon dioxide
and can therefore inflate the reported value.
2. The sample volume involved in the TOC analyzer is so small
(approximately 40 microliters) that it can easily become
contaminated, with dust, for example.
3. Industrial wastes from the organic chemicals industry with
low vaporization points may vaproize before 150C and
therefore be reported as inorganic carbon.
TOC RWL data for Subcategories A, B, and c are shown in Figures V~3, V-
6, and v-9. A summary of the concentration ranges are presented below:
Subcategory TOC_RWL Ranqe
"" mg/1
A 100-3,000
B 100-2,000
C 3,000-5,000
Typical values for municipal waste waters range between 50 and 250 mg/1.
Effluent limitations were not established for the TOC parameter,
although its use is not precluded if a suitable correlation with BODJ or
COD is established.
Other Significant Pollutants
Suspended solids, oil, ammonia nitrogen, total Kjeldahl nitrogen,
phenols, dissolved solids, cyanide, sulfate, and color, in general were
present in smaller concentrations. Effluent limitations are specified
for TSS and phenols in all subcategories since these are generally
present in all subcategories. other pollutant parameters which are
discussed in this section but no effluent limitations established are
272
-------
not present, in all Subcategories, and are generally controled at the
source. These may, however, present environmental problems where water
quality standards dictate and may ultimately be limited.
TSS
Total Suspended (nonfilterable) Solids in the form of RWL are plotted on
Figures VI-1 and VI-2 for Subcategories B and C respectively. In
general, most of the data points are below 50 mg/1. There are, however,
particular processes and certain plants which ahve very high suspended
solids loadings, on the order of 500 mg/1. In some cases (e.g.,
terephthalic and production), dry housekeeping with minimal use of
washdown water would drastically reduce the discharge of Total Suspended
(nonfilterable) Solids.
Total Suspended (nonfilterable) Solids concentrations for typical
municipal waste waters range from 100 to 300 mg/1.
Oil and Grease
Oil (extractables) is a measure of the insoluble hydrocarbons and the
free-floating and emulsified oil in a particular waste water sample.
One particular problem of importance is the obtaining of a
representative waste water sample when free-floating oil is present.
Representative samples may generally be obtained if there is a freefall
in a sewer line, e.g., a drop manhole. Sample collection from a sump
where there is an oil accumulation attributable to the sump's inherent
detention time should be avoided.
Oil and grease RWL's for Subcategories B and C are presented in Figure
Vl-3, and tabulated in Tables VI-2 and VI-3. Most of the oil
extractables are within the range of 5 to 50 mg/1 (by carbon
tetrachloride solvent). Specific processes involving high oil
concentrations are acetaldehyde, acetic acid, phenol via cumene, oxo-
chemicals, and ethylene. Only the ethylene production wastes have free-
floating and emulsified oils. The oil and grease data for the remaining
processes merely reflect the amount of insoluble (in water) hydrocarbons
which are soluble in the solvent. Based on the previous qualifications,
no effluent limitation values were established for the discharge of oil
and grease from the organic chemicals industry.
Ammonia nitrogen (NH3-N) and total Kjeldahl nitrogen (TKN-N) are two
parameters which have received a substantial amount of interest in the
last decade. TKN-N is the sum of the NH3-N and organic nitrogen present
in the sample. Both NH3 and TKN are expressed in terms of equivalent
nitrogen values in mg/1, to facilitate mathematical manipulations of the
values.
273
-------
Organic nitrogen may be converted in the environment to ammonia by
saprophytic bacteria under either aerobic or anerobic conditions. The
ammonia nitrogen then becomes the nitrogen and energy source for au+-o~
trophic organisms (nitrifiers). The oxidation of ammonia to nitrite and
then nitrate has a stoichiometric oxygen requirement of approximately
4.6 times the concentration of NH3-N. The nitrification reaction is
much slower than the carbonaceous reaction, and, therefore, the
dissolved oxygen utilization is generally observed over a much longer
period.
Ammonia and TKN RWL data for Subcategories B and C are presented in
Figures VI-4 and VI-5 and tabulated in Tables VI-2 and VI-3. Most of
the NH3 and TKN data points are below 10 mg/1. This is low compared to
the concentrations typical of municipal waste waters, 15 to 30 mg/1.
However, Tables VI-2 and VI-3 show that some processes (caprolactam,
aniline, butadiene) have extremely high nitrogen values.
Phenols
Phenols in waste water present two major problems; (1) at high
concentrations phenols act as bactericides; and (2) at very low
concentrations, when disinfected with chlorine, chlorophenols are
formed, producing taste and odor. Past experiences has indicated that
biological treatment systems may be acclimated to phenol concentrations
of 300 mg/1 or more. However, protection of the biological treatment
system against slug loads of phenol must be given careful consideration
in the design. Slug loadings, depending on concentration, could be
inhibitary to the biological population.
The phenol RWL data are presented in Figure VI-6 and tabulated in Tables
VI-2 and VI"3. The concentrations are generally below 1 mg/1. Specific
processes (Bisphenol and phenol via cumene) have concentrations in the
5,000-10,000 mg/1 range. In both these processes, phenol is amenable to
in-plant recovery and therefore would probably not be discharged in
their waste water.
Total Dissolved (Filterable) Solids
Dissolved solids in organic chemicals waste waters consist mainly of
carbonates, bicarbonates, chlorides, sulfates, and phosphates. Sulfate
RWL data for subcategories B and C are presented in Figure VI-7 and
tabulated in Tables VI-2 and VI-3. It is interesting to note that most
of the data above 300 mg/1 are from Subcategory C, while most of the
data below 10 mg/1 are from Subcategory B. This is an interesting
commentary on the process differences between Subcategories B and C, and
is applicable also to dissolved solids concentration. The extensive
amount of process water recycle and reuse is primarily responsible for
these high concentrations.
274
-------
The high dissolved solids and sulfate concentrations in Subcategory
D(unlike the other subcategories) are the direct result of inorganic
chemical additions due to intimate contact with the batch production
chemicals. Chemicals compounds introduced in the other subcategories
are organic in nature and do not contribute to the overall magnitude of
dissolved solids.
Because dissolved solids and sulfate concentrations are intimately tied
to process recycle and the quality of the process raw water source, it
is recommended that these parameters be dictated by local water quality
requirements.
Cy.anide.t.jrotal
Cyanide was analyzed using the distillation procedure in Standard
Methods and the Orion specific ion probe. The cyanide values are
reported in terms of CN-ion. The cyanide ion is in equilibrium with
hydrogen cyanide as follows:
[H+] + [CN-] [HCN]
At a pH of 8 or less, the HCN is largely undissociated; then as the pH
increases, the equilibrium shifts toward CN-.
CN RWL data for Subcategories B and C are presented in Figure VI-8.
Much of the data is below 0.1 mg/lf and practically all the data points
are below 1.0 mg/1. At these concentrations, the values are such that
specific limitations are not required.
Color
Color is objectionable from an aesthetic standpoint and also because it
interferes with the transmission of sunlight into streams, thereby
lessening photosynthetic action. Color is measured against a platinum
cobalt standard which is basically a yellow-brown hue. This color
shading was developed to simulate domestic waste waters. The use of the
procedure on highly colored industrial waste waters is subject to
question. During Phase II of this study, a more intensive investigation
will be made as to the most appropriate procedure for reporting color.
Color RWL data for Subcategories B and C are generally not a major
consideration. However, in Subcategory D color is as high as 50,000 Pt-
Co-units for pigment and dye waste waters. There were two major reasons
for not trying to set limitations for Subcategory D:
1. Sufficient RWL data were not collected during the sampling
program. (This will be remedied during Phase II of this
project) .
2. Scarcity of treatment data on color removal presented major
275
-------
m
>*
L.
8,
0)
m
£
' -S
> g
OJ -1
-Q -J
H 2
C
ZI
(U
u
ui
o
Sulfate
i.
o
2
z
X
2
1
1
U.
4->
U
3
1.
O.
H
-O
~J
1
H
«J
6
J3
Q
H
^~
_1
Cn
e
^
g
H
£
_j
oi
E
£
3
H
£
_j
oi
.Q
8
H
.a
_j
E
-Q
i
i
i
i
1
1
VD
OJ
O
O*
\O
OJ
H
O
3
H
H
OJ
H
H
3
LA
£
^
LO
H
d
a
H
Q_
LO
CVI
d
a
CVJ
H
o
NO
H
H
O
V
CO
°
o
o
LO
-*
3
o
CO
cvl
ON
0°
LA
00
H
£
H
*
O
(0
rO
rO LA H
H O VQ
0 O OJ
d d o
SO -st- ro
* * ON
CO t~-
LAH
H
-d- lTN_d-
O O O
333
q q cvi
ro ro H
V
r^'oo
o o o
VV
ro
I
O CVI
CM • -=fr
O O ro
•*
• ro Cvl
CVI CVJ -
CVJ
H
o t-
• O
H
t-CVJ t-
°>ro.c?,
O -=t LA
LA LA LA
1 1 1
O O O
vjD CO fO
H HVO
gi5s
°'d °"
roco ON
coco to
i-t rO
^1
(Q
V
S S2i
CM X !>•
O
ONO
CO -^" NiJ
LO LOCO
S 8s)
° do
o
V
^-i roco
0 t-
V
v-i
CVJ O O
8 33
O ro
° Hro
V
^t ^J3 OJ
O ^ r-l
d -d
V
-.
ro o t-
^ ^ H
o t — o
O ro O
O ONU3
r^- oj ,4-
LA rO
OJ X X
- vD
O t-- •
OJ LA C~
-^ OH
LA LA
I 1
H O O
O* • •
OJ H
O ro
OJ « O
o o
O
c— ^j- ^o
\O LA rO
H
§
4-1
C
o
. 5
(O
UJ (J
OJ LA LAJ- t— _J
OJ KN. r^- H CO H
LAO O O • O
d o* d d ° o
ro t-OJ HCO H
-cf H H HO
H H
lO OJ IO (O
H fO IAI 1
O MD O rOO O
^- fO • ' H H
• • O O X X
O O LA rO
• H
ON O M) O CO LA
-3- OJ -CO ON *
MD H ro OJ • H
t- O
"f-tVY^
O 0 O O O
H H H rH H 1
X. X X K ti
OJ VD O O \Q
H rO OJ LA OJ
ro LA
H OJ^t -^- LA
r-l H O O O
d d d d o*
V
-* ro
_) LA O CVJ H O
o o S^ o o x
d o . d d •
OJ rA
V
LA OJ LTN b— O M3
LA LAO ON O\ oj
•O OJ
V
ro -tf root ro
i HIII
O CO O O O O
ONO - • O "ON
OJ-cJ- • •
vD H H
vv
-d- so LA ON OJ -3-
• ON O fO OJ H
SO H
V
ro ro ro
1 i i
O O O C--CO H
H H H O ro O
X X X O O •
rocb jo - • o
rO O O O O
t- H r-l
CO \O V£> ON LA-3-
t~ TO KMA €VJ OJ
MD O OJ lACO t^-
O -3: LAO ^f t~-
H ro OJ H LA H
LA
I/)
Ul
e
i~
(D
*
Q>
C
0)
Q.
O
v-t
o o
ro ro
roco
ro H
NO H
LACO
d d
o o
LO
O O
CO O
-d- N£)
V
_-J- _-J-
-^ °.
0 0
V
roro
1 I
0 0
*8 H.
rO
co-d-
CVl
J-
32
IR?-
LO fO
V
LON£)
O LA
V
LA LA
1 1
O O
roro
to Cvl
-^ IA.
d d
HCO
KV •
H t^
H
0
—
fO
t-i
o
d
ND
H
V
H-
O
C^
.
V
_-(.
o
o
V
H
C?,
°.
O
ro
NO
CVJ
CVI
o
CO
Cvi
ON
t-^-
O
•H
rH
rH
0
d
CM
U
m
i
o
LO
NO'
rH
ON
O
ro
H
s
CVJ
V
o
V?
.
H
OJ
H
to
O
t—
ON
to
O
-=f-
co
OJ
s
d
\o
LA
ro
i
O
IA
fO
^.
o
o
c — .-
E 91
0) < C
C 01
O — _
276
-------
O
H
.O C
(0 O
-t
o
1
t—
V
LA
O
0
V
o
1
t—
V
LA
O
V
LA
LA
O
H
LA
"^
O
1
(A
o7
o
o
1
rA
LA
t-
rA
O
LA
1
d
ON
H
*
(O
<0
LA LA-d-
I 1 1
O O O
333
NO O LA
IN--3- H
V V V
IA tA LA
O O O
O 0 O
V V V
LA LA^t
1 1 1
O O O
333
\£> O LA
t— -3- r-)
V V
IA LA LA
O O O
o o o
V V V
ro
i i O
O O r-1
333
IACO LA
CO -4- H
\Q LA LA
LA • O
O
-T^OH
ill
OJ H
5S*
O O O
-t
3"
CO
rH
OJ
H 1 '
O
CO 1 CO
O'NO •
• O
IA
LA t-- O
1 — 1 • •
F\0 rA
^
o
0}
LA
1
0
I
CM
V
LA
o
o'
V
LA
o
1
CM
V
LA
O
0
V
rA
1
o
1
CM
LA
O
1
O
H
H
OJ
O
1
o
1
t~
rH
O
cf
o
»
o
od
UJ
LA'i
1 0
3
3 ON
rH H
V V
tA LA
O O
O O
vv
1 O
3^
rH H
V V
o o
0 0
VV
IA
-=f i
1 O
0 rH.
H ON
"8-*.
H
LA 1
1 O
^ N^l
fA /A
-=f -a
OJ rH
0 O
LA
1
O
co'
OJ
NO 1
0
LA-d-
1 1
O 0
u
t— OJ
3,°.
• o
o
8
(0
C
'l.
0
i
0
LA LA
'f^'xVo^o
LA IA O X X X
• • • co -3- t-
LA H H • • •
v vv H w ^
U~\ LT\ (M tf} r-l \Q
O O O-* LAO
o d o H. . d
O O
VV V V
i LA IA
O -4 1 -=*• 1
M f.
"o
fD
>
0)
o.
_
o o
o o
NO CO
V V
IA LA
0 0
d d
V V
"f LA
S'o
<§3
• H
V
LA t—
0 0
o o
V
rA^T
O O
33
LAO
• OJ
LA NN.
rH •
1 LA
01
33
CO 3
IA O
•H NO
LAH
OJ-=t
H 0
-TV
o o
(AOJ
IAH
O O
O O
33
NHco
rH rH
-jj CM
rH H
O*
Q)
•D
,
OJ
O
"
tr\
^
o
ON
IA
OJ
O
-t
o
OJ
cf
.
o
VI
-^f
o
3
rA
OJ
V
LA
O
o
V
o
3
fA
OJ
V
°
d
V
rA
o
1
H
-*
d
-t
0
OJ
V
LA
O
V
1-
O
3
LA
KN
ON
OJ
rA
o
H
H
LA
OJ
O
I £
277
-------
t
8,
0)
4J
£
U)
1
3
in
< >^ 1
X-d- KN
• H OJ
t-
V V
&** ,
d°°
vv
ooVo^^
d o 6 "^
o
00 CO OD
Jt OJ OJ ©
^1 tACO
OJ O H H
33°.°.
• 0 O O
o •
LA
H H OJ •
KN
KNLA
O O £H (
H S-0
04-*
OJ H t~ 1
O O rH
ON J- LA KN
O • H ON
•• KN t— LA
H -3"
•o
"y
U
•g
j:
|
^
3 i
d
O 1
KN
t—
1
O
t- I
LA
1
O
3 1
KN
OJ
V
•B t
a
V
"T LA
i*
• o
00
• LA
^tVo
^
o ,4-
•* ^
O -*
V
o
H
CO
H
O 1
0
co LA
• OJ
CO KN
2
10
"(D
.C
j;
£
H
>•
I
a
^
d
H
LA
H
LA
KN
LA
d
0
KN
H
O
1
rH
LA
o
o
OJ
d
H
ON
Ej
O
^
o
KN
KN
°
d
S
»
u
1
1
o
g
H
LA
€
d
-=t
rH
H
-*.
^
d
o
ON
O
LA
H
•3
d
V
OJ
o
ON
KN
H
LA
KN
KN
I
0
OJ
O
1
0
H
d
£
•a
'o
u
u
u
KN
H
LA
LA
S
LA
OJ
OJ
O
LA
ON
V
3
d
V
LA
rH
OJ
§
O
d
KN
0
KN
C-
H
o"
LA
ON
CO
Oj"
0)
+J
U
«J
e
Q.
m
o
S
d
H
H
H
H
8
o
H
^
O
OJ
3
d
LA
KN
C—
C—
S
KN
^
LA
rH
o
o
5
rH
g
"
'i
i
h_
o
0*
CO
-1
o
H
OJ
KN
O
KN
1
O
LA
CO
LA
LA
H
KN
O
H
H
O
a
CO
f
1
.
c
H
H
^
rH
H
^
o
OJ
d
a
d
H
OJ
O
rH
OJ*
C^
o
d
d
1
o
H
8
KN
0
a"
•a
.
-5
•
U
te
j=
*
x:
4-1
0)
OJ
LA
H
KN
t
(
LA
1
O
CO*
•B
d
V
1
,
'
'1
O
co
KN
oj
s
OJ
a>
>
L.
U
m
4J
5-
m
">>
u
4-*
0)
4-1
0)
X
278
-------
-0
O
_Q
_,
D>
E
J3
o
H
J3"
_J
1
J3
§
H
^
-.
E
_Q
D
H
.0
•^
£
5
H
;£
Ol
E
^
H
s-
Ol
E
UN LA LA
I \ I
000
333
t- o o
KN rA KN
V V V
LfN UN IA
000
o o o
V V V
KN
1 UN l/N
Ol 1
HOC
g\ 3 3
ON o o
H fA OJ
V V
UN tA
OJ O O
VV
^r-t
$33
H O4 04*
KN UN_3-
OJ O* •
0
^•f
S'o'o
o' H O
oj in
^"^ rH
O
J- L/N LfN
1 1 1
0 0 O
04 KN rA
\0 IA r-J
do"
KN
l KN rA
Ol 1
33 °
ON ON 04
HCO 04
CO -d- VJD
ITN . .
OJ rH
^
0
V
£
o
V
XO
O
LA
V
LA
o
0
V
UN
1
O
3
H
KN
*
o'
^
0
ON
H
o"
V
O
g
UN
S
o*
UN
1
0
-*
UN
LTS
O*
^
o
-3-
OJ
V
IA
O
o
V
-t
0
04
V
IA
0
O
V
i
^
•o
H
vo
1
H
•
o
1
H
8j
0*
O
m
3
0*
1
0
3
V
IA
O
o
V
LA
1
O
V
UN
O
o
V
o
ON
t-
o
V
o
IA
ON
t^
o'
^
o
to
81
0*
O
OJ
VO
o*
-t vf
000 1
• • • CO
-=T rH KN
04
V V V v
o o o o
o o o o
VV V V
KN 1 "1 KN
100 1
O rH H O
• • rA •
LA H OJ
V
UN
^?R°. , uA .
o • o
-* v "
fA KN
OJ O -=fr O
H.3°. ' 3 '
O ON O O
H H
'V. T , t ,
H H
rA-d1 -d"
i 1 -d- rA I H
O O H r-t O CO
KN— rH
3^^
o KN 04 • oj •
O O OJ O* OJ
V"^" rA "t
O O H O
338 ' 3 '
HVD • KN
. • O
CO L/N \O
O t- ON H
OS IA H 1 H '
0* r-I 0* H
-d- -d:
CO O O4 O
H H, O 1 r-f I
• X • X
o ^t a co
ON -^
od-0 , §,
04* ""* O*
-T
O
^
rH
LTN
°.
0
KN
1
O
n
KN
H
_=}•
O
04
04
O
•0
rA
KN
O
1
CO
H
LA
O
1
1
KN
'o
1
OJ
IA
\D
0*
t
O
3
&
rH
UN
°.
O
V
o
•a
GO
rH
OJ
V
UN
tA
O
V
V
o
1
CT\
H
ITS
o
-t
o
H
KN
CO
O
O
1
1
KN
i
o
1
H
OJ
fA
o"
KN
O
a
H
H
LfN
,
O
V
0
KN
rA
o"
o
o"
p
o*
KN
1
O
H
O
'
,
IA
1
O
1
H
§
o"
O
u-i
IA
l/N
0
o
V
o
UN
V
UN
O
o
V
OJ
o
OJ
8)
OJ
KN
O
-s
rH
ON
O
1
1
1
0
H
H
H
O*
-Y
O
V
g
o
V
KN
i
0
t—
O4
•t
O
"*
CO
OJ
o*
tr\
i
0
3
KN
£—
l/N
O
O
^
O
LfN
UN
KN
O
O
o
KN
IT\
1
o
04
V
UN
O
o
V
l/N
1
O
OJ
V
in
O
0
V
V
o
ITN
O
•H
^
0
-*
KN
O
O
•a
H
KN
*
O*
'
(
UN
1
O
04
rH
V
LT\
o
o
V
UN
O
rH
t—
0*
^
O
04
H
O
UN
O
in
O
H
t—
O
O
^
o
OJ
OJ
H
O
•t
0
CO
LA
51
OJ
8. ,
O
V
IA
O
0
V
LfN
8
O*
UN
O
0*
V
V
o
H. H
O 04
0
04
O -3-
H •
H
KN O4
H ON
° °-
O O4
rH LP\
o"
CO
ON 1
O
1 '
O
1 ,
rH
'
|
1
|
L/N
°
H
O
O
LfN
KN
1
O
X
L/N
OJ
\0
CO
CO
'
1
•
1
8 2 *
^ S .-
O U "-
O n
~ -- o
I«
?
I
S f
£
'5
I
*J
I
279
-------
FIGURE VM
RELATIONSHIP BETWEEN SS RWL AND FLOW RWL FOR CATEGORY B
1Q-3 L_ ir:
FLO* R*L (GAL/103 LB PRODUCT)
I i
10
FLO* R«L (LITER/103 KILOGRAM PRODUCT)
280
-------
FIGURE VI-2
RELATIONSHIP BETWEEN SS RWL AND FLOW RWL FOR CATEGORY C
102
101
10°
10-'
10-3
1C2
10'
10°
10-'
10-2
10-3
LEGEND
1 ME
2 T
3 D
4 T
3 A
6 T
7 ft
8 A
9 A
10 ft
11 B
- 12 E
~~ 13 V
i— 14 0
- 15 ft
>- IB C
i_ • s
A
qyS
$**/
/
/
*?
/
^
^*x
/
'/
/
f
,
tfc/
<&x
y
'
f
1 1 — i — r — i i i i 1
THYL MET HACRY LATE
REPHTHALIC ACID ' 'POLYMER GRADE
METHYL TEREPHT HAUTE
REPHTHALIC ACID ' ' INDUSTRIAL GRADE'
TWLATES
TRAETHYL LEAD
ENOL/ ACETONE
;ETALDEHYDE
,ETIC ACID
ILINE
S-PHENOL
HYLENt GLYCOL
NYL ACETATE
0-CHEMICALS
RYLIC ACIO
WROLATAM
JRVEY SAMPLING DATA
/
'
/
/
/
/
/
'
/
/
/
/
/
'
^
/
•^
X
/
/
/
/
/
/
/
/
X
L/
X
^
/
(
/
{
'
£
/
f
/
/
/
/
/
/
/
/
/
/
A
/
/
/
^3
/
/
'
/
/
/
/
|/
/
/
13
/
X
/
,
/
/
X
_^
8
7
/
^
,9
/
/
/
/
/
/
/
7
^
X
8
/
x
1
i
/
i
/
f
^
2 ^
/
/
/
./
/
'
/
/
/
/
'
/
/
x*
/
/
10
/
/ i
/"
/
/
'
/
; /
s
4
/i
^
14
x
/
^
^
/
3
,
/
/
/
/
•
L/n
1
/'
/
/
»
Fs
> —1
/
I*4
15
2 ,'
/
^
/
/
/
/ /
/
/
/
'
/
/
7 /
/
/
'
•IB
/I
/
'
T3
9
/
/
^
^ 5<
/
/
X
/
/
/
/
^
/
/
^
^
/
-
^c
i
i
/
"t
IT
; 1
7
/
/ y
/
/
/*
/
S
f
X
yf' *
f 1
t "-I
i
—
10"
10'
102 103
FLOK R»L (GAL/1000 LB PRODUCT)
I03 104
FLOf ML (IIITER/IO3 KlincRMI PRODUCT)
I05
281
-------
FIGURE VI-3
RELATIONSHIP BETWEEN OIL RWL AND FLOW RWL FOR CATEGORIES B AND C
10'
Id"
g ,r'
5
O
10-2
10-3
IC^
103 LBS PROOUCT)
o
C3
10 2
m 3
10"*
— LEGEND
A CATEGORY B
• CATEGOR
S^
^>X
/
/
'?
C
/
f/
r
/
,
x
'
too
•
1 1
/
/
Jf
/
/
.'
/
/
/
/
/
/
/
"
/
y11]
•
/
/
>
X
•
x
'
A
/
/
/
m
*»
/
t /
A
•
v
y
/
1
A
/'
.
A
/
•
^
-
/
A
/
y
10' 102
FLOK R*L (GAL/103 LBS PRODUCT)
i 11 i
I
/
/
, '
/
A
B
/
;'
X
/"
'
X
/
s
'
'
/
/
/
7
Xj
X
/
/
103 I04
1 1 - .___ .
/
L,
--
/
1
J
10 ' 102 |03 1(j4 105
FLO* RW. (LITER/103 KILOGRAM PRODUCT)
282
-------
FIGURE VI-4
RELATIONSHIP BETWEEN NH,-N RWL AND FLOW RWL FOR CATEGORIES B AND C
3
5
£
I
10-2
102
10'
ItP
u t
- i 10-'
3
10-2
10-3
i 1—
LEGEND
A CATEGORY B
• MTEGORV C
10"
/
10'
FLO» RWL (GAL/103 LBS PRODUCT)
103
10'
103 104
FLO* R»L (LITER/103 KILOGRAM PRODUCT)
105
283
-------
FIGURE VI-5
RELATIONSHIP BETWEEN TKN-N RWL AND FLOW RWL FOR CATEGORIES B AND C
102
10'
10°
10°
10-3
ir4
10'
ID4
FLO! RWL (GAL/103 L8S PRODUCT i
-J L, L-
I0
10s
FLO* (Wl 'LITER/103 KILOGRAK PROOUCT)
284
-------
FIGURE VI-6
RELATIONSHIP BETWEEN PHENOL RWL AND FLOW FOR CATEGORIES B AND C
101
ID1
LEGEND
* CATEGORV B
• CATEGORY C
10°
iir2
\- §10-2
10-3
10-3
10-4
llH I
ID'S
10-51
V
—\—
I/I
10°
10'
|02 ID3
FLO* RM. (GAL/103 IBS PRODUCT)
104
10r
lo3
FUN MIL (LITER/103 KILOGRW PRODUCT)
104
I05
285
-------
FIGURE VI-7
RELATIONSHIP BETWEEN SULFATE RWL AND FLOW RWL FOR CATEGORIES B AND C
10*
LEGEND
A CATEGORY B
• CATEGORY C
10'
10'
10°
10-3
KHL
10°
10'
_! 1_
102
FU» R«L (ML/103 LBS PRODUCT)
103
102
103
FLOW RiL (LlTER-'tO^ KILOGRAM PROOICT)
-\~
—4
~|-
105
286
-------
FIGURE VI-8
RELATIONSHIP BETWEEN CN RWL AND FLOW RWL FOR CATEGORIES B AND C
10-1
10-2
1 '°"3
Z
*
S
,n-6
~ iH
1 t
O3 IBS PRODUCT)
i"
3
i ]
~ LEGEND
_ AC
• C
&
UTEGOR
ATEGOR
/
\ 1
B
r r
,
/
'
/
f
.
/
f
f
/
/
y
^
f
/
y
/
•
/
/
^
•
/
<,
^ '
^
/
/
'
/
^
/
r
•
x
•
m
/
A*
^
A
/
^
/
/
11
/
/
A
/
f
/
\
/
i
/
i
<
/
/
r •
/
/
m
/
/
^
x
^
/
|/
/
/
/
^
'
.
,
f
/
/
'
/
10"
10'
102
FLO) fflHL (GAL/103 IBS PHOOUCT)
103
102 103
FLO» MIL (LITER/103 KILOGRAM PRODUCT)
I05
287
-------
technological questions concerning levels of color removal
for various types of dyes and pigments. This situation can be
be remedied during Phase II of our study by a concentrated
study of the color removal of various waste water unit
processes. However, there is recent evidence that carbon
filters can be a satisfactory treatment agent for many
color problems.
Pollutants_qf Minimal Significance
The remaining parameters which were examined were calcium, magnesium
total hardness, chlorides, total phosphorus, pH, alkalinity, acidity,
and various heavy metals. These pollutants are generally not considered
significant in comparison to the oxygen demand pollutant parameters.
Effluent limitations for the pH are specified for all discharges within
the range of 6.0 to 9.0. Heavy metals concentrations are not limited
since most processes recover metals catalysts as an in-process control.
This does not preclude the possibility of such limitation being required
for specific processes or situations where such in-process controls are
not applied or properly functioning. Hardness is an indication of the
soap-neutralizing power of water. Any substance which will form an
insolubel curd with soap causes hardness. Waters are commonly
classified in terms of the degree of hardness as follows:
Concentration.
mg7T~as~CaC03
0-75
75-150
150-300
300 and up
Terminology
soft
Moderately Hard
Hard
Very Hard
The major detrimental effects of hardness include excessive soap
consumption, problems when used for particular process waters (e.g., in
the textile industry), and the formation of scale in boilers and water
heaters.
Many of the specific comments made previously regarding dissolved solids
are directly applicable to these parameters of minimal significance.
Concentrations of calcium, magnesium, chorides and hardness are
generally higher for Subcategory C because of extensive recycling. In
addition, particular processes in Subcategory C product NaCl as a
product of reaction, e.g. tetraethyl lead production. Subcategory D
waste waters likewise have high concentrations as a result of inorganic
chemical additions.
Phosphorus occurs in organic chemical waste waters as orthophosphate
(H2P04, - HP4=,P04=,) or as polyphosphate. All polyphosphates gradually
hydrolyze in an aqueous solution and revert to the ortho from.
288
-------
Phosphorus concentrations in the organic chemicals industry are
relatively low (less than 10 mg/1) and reflect the quality of the intake
water and the amount of recycle employed with the process. Phosphates
and polyphosphates are used for corrosion control and boiler water
conditioning.
Specific processes utilize phosphoric acid as a catalyst (by
impregnating silica alumina media) for polymerization, alkylation, and
isomerization processes. The spent acid catalyst may be found in
process effluents if it is not segregated for separate disposal. In
these cases, total phcphorus values over 500 mg/1 have been observed,
The acidity of a waste is a measure of the quantity of compounds
contained there in which will dissociate in an aqueous solution to
produce hydrogen ions. Acidity in organic chemicals wastewaters can be
contributed by both organic and inorganic compound dissociation. Most
mineral acids found in waste waters (sulfuric acid, hydrochloric acid,
nitric acid, phosphoric acid) are typically strong acids. The most
common weaker acids found include the organic acids such carborylic and
carbonic.
Compounds which contribute to alkalinity in waste waters are those which
dissociate in aqueous solutions to produce hydroxyl ions. Alkalinity is
often defined as the acid-consuming ability of the waste water and is
measured by titrating a given volume of waste with standard acid until
all of the alkaline material has reacted to form salts. In effect,
alkalinity is the exact opposite of acidity.
Both inorganic and organic compounds can contribute to alkalinity, but
the most important alkaline wastes in the organic chemicals industry are
the spent caustics, which contain sodium, calcium, and potassium salts
of weak organic acids, and carbonates. These compounds tend to raise
the pH to values over 10.
The hydrogen ion concentration in a aqueous solution is represented by
the pH of that solution. The pH is defined as the negative logarithm of
the hydorgen ion concentration in a solution in gram equivalents per
liter. The pH scale ranges from below zero to fourteen, with a pH of
seven representing neutral conditions i.e., equal concentrations of
hydrogen and hydroxyl ions. Values of pH less than seven indicate
increasing hydrogen ion concentration or acidity; pH values greater than
seven indicate increasing alkaline conditions. The pH value is an
effective parameter for predicting chemical and biological properties of
aqueous solutions. It should be emphasized that pH cannot be used to
predict the quantities of alkaline or acidic materials in a water
sample. However, most effluent and stream standards are based on
maximum and mimimum allowable pH values rather than on alkalinity and
acidity. Typical pH values recommended for stream standards are 6.5 to
8.5.
289
-------
Since pH RWL values are not additive, it is not always possible to
predict the final pH of a process waste water made up of multiple
dischargers. in addition the individual plant's production mix will
dictate final pH ranges, which may be kept within the acceptable range
merely by equalization, or which may require more sophisticated
neutralization facilities.
Minimal concentrations of heavy metals were observed in most of the RWL
data. Particular processes in Subcategory C (e.g. terephthalic acid)
had higher concentrations of cobalt due to the loss of the catalyst.
Particular waste waters from Subcategory D (e.g. metallic dye
production) had very high concentrations of Cu, 10 mg/1. The presence
of heavy metals is contingent on batch metallic dye production, which
may occur one day per week or five days/week depending on the demand in
the market place.
290
-------
SECTION VII
CONTROL AND TREATMENT TECHNOLOGIES
It is the aim of this section to describe and present available data on
the different pollution control and treatment technologies which are
applicable to the organic chemicals industry. Based on that data
avilable, conclusions have been drawn relative to the reduction of
various pollutants which is commensurate with three distinct levels of
technology. These levels are defined as:
BEST PRACTICABLE CONTROL TECHNOLOGY
CURRENTLY AVAILABLE (BPCTCA)
BEST AVAILABLE TECHNOLOGY ECONOMICALLY
ACHIEVABLE (BATEA)
BEST AVAILABLE DEMONSTRATED CONTROL
TECHNOLOGY (BADCT)
The conclusions relative to what combination of control and treatment
technologies are consistent with these definitions are embodied in the
reduction or removal of pollutants specified for each level. In later
sections of this report specific reduction factors are applied to the
process RWL developed for each industrial category to obtain numerical
values for effluent limitations and new source performance standards.
These reductions are general and are considered to be attainable by all
of the rpocesses considered within the category.
The costs associated with these effluent limitations and new source
performance standards have been estimated based on model systems which
are considered capable of attaining the reduction factors associated
with each technology. It should be noted and understood that these
particular systems chosen for use in the economic models are not the
only systems which are capable of attaining the specified pollutant
reductions. There exist many alternate systems which either taken
singly or in combination are capable of attaining the effluent
limitations and standards recommended in this report. These alternate
choices include:
1. different types of end-of-pipe waste water
treatment,
2. different in-process modifications and pollution
control equipment,
3. different integrated combinations of end-of-pipe
and in-process technologies.
291
-------
It is the intent of this study to allow the individual manufacturers
within the organic chemicals industry to make the ultimate choice of
what specific combination of pollution control measures is best suited
to his situations in complying with the limitations and standards
presented in this report.
ess systems
It is not possible to recommend a general list of process modifications
or control measures which are applicable to all of these processes
within the organic chemicals industry or even to the processes within
one industrial subcategory. The following discussions deal with
individual techniques which may be applicable to groups of processes or
to single processes. The techniques described are based on both the
practices observed during the sampling visits as well as those which
have been described in the literature. In most cases, they can both be
implemented with existing processes or designed into new ones.
The general effect of these techniques is to reduce both the pollutant
RWL and the volume of contact process water discharged for end-of-pipe
treatment. This corresponds to moving the data shown in Figures V-1
through V-11 toward the lower left side of the RWL envelopes.
The control technology described in the following paragraphs starting on
page VTI-1 to page VII- 4 comes from:
Thompson, S.J., "Techniques for Reducing Refinery Wastewater, "oil_and
Gas_Journalf Vol. 68, No. 10, 1970, pp. 93-98.
cooling Water yjed in Barometric_Cgndensers
Figure VII- 1 illustrates the classic barometric condenser. In the
typical example shown, the volume of water being contaminated can be
decreased from 260,000 Ib/hr to 10,000 Ib/hr for a condensing duty of
10,000,000 BTU/hr. This can be accomplished by substituting an air
exchanger for water sprays. This type of process modification can be
sized to cover almost an infinate number of specific process cooling
duties.
It should be noted that water cooled surface condensers can also be used
in this application. However, these require the use of non-contact
cooling water.
Regeneration of Contact Process Steam f rom ^Contamina ted Condensate
Figures VII- 2 illustrates the trade-off between contaminated contact
process steam condensate and non-contact steam blowdown. The contact
process waste water is reduced to a small amount of condensate. This
scheme can be used to regenerate stripping steam in distillation towers
292
-------
FIGURE Vll-l
BAROMETRIC CONDENSER
CUSTOMARY
WATER VAPOR IN
COOLING WATER
FOR 10-MILLION-BTU/HR DUTY,
COOLING WATER AT 85°,
OUTLET TEMPERATURE AT 125°
PROCESS WATER 10,000 LB/HR
COOLING WATER 250,000 LB/HR
TOTAL 260,000 LB/HR
CONTAMINATED WATER
SUBSTITUTION OF AN AIR FAN
WATER VAPOR IN
PROCESS WATER
COOLING WATER
TOTAL
10,000 LB/HR
0
10,000 LB/HR CONTAMINATED WATER
U
293
-------
FIGURE VII-2
PROCESS STEAM COMPENSATE
CONTAMINAIEDJROCESS.
REGENERATED
PROCESS STEAM
STEAM CONDENSATE
NON-CONTACT STEAM
BLOWDOWN
(CONTACT PROCESS
«ASTE»ATER)
NON-CONTACT
CONDENSATE
294
-------
or dilution steam in pyrolysis furnaces. Heat exchange is through a
surface shell-and-tube heat exchanger, which can be sized for a wide
variety of heat transfer duties. A system similar to this was described
in detail for ethylene manufacture in Section IV.
Substitution_of_Vaciium Pumpg^f or_Steam_Jet Ejegtors
The use of vacuum pumps in place of steam jet ejectors is shown in
Figure VII-3, This practice can be used to eliminate process RWL from
the condensed steam used to draw a vacuum on the process. A specific
vacuum pump system has been sized and priced for application in the
process for manufacturing styrene by the dehydrogenation of ethyl
benzene (Section IV) . This same type of system is applicable to many
other processes where operation under bacuum is necessary.
It should be noted that in many cases the steam jet ejector system may
be coupled with a barometric condenser instead of the surface cooler
shown in Figure VII-3. In this case, the volume reduction of contact
process waste water will be quite substantial. It may also be possible
to use the hydrocarbon vapors from the vacuum pump in the plcmt fuel-gas
system (because of the reduced moisture content) rather than venting
them to a flare.
The liquid compressant in a vacuum pump can protect it from corrosion.
The manufacturers have accumulated operating data on performance of many
liquids with different gas mixtures. It has been concluded that
ordinary cast iron will often stand up well in resisting corrosive
gases. More expensive materials for pump construction, such as monel or
hastelloy c are available for particularly corrosive gases such as
halogens.
Recycle of Scrvabber^Water
Figure VII-4 illustrates a method of concentrating contaminants in
scrubber bottoms nearly to their saturation point. This is accomplished
by recirculation of the scrubbing or wash water. Theoretically, the
tower would require more trays or contacts, as dictated by the specific
vapor-liquid equilibrium of the system. However, in many cases,
existing towers can be modified to work in the manner illustrated.
RecoyerY-Qf^Insoluble Hydrocarbons
Two methods for improving the separation of insoluble hydrocarbons from
water are shown in Figures VII-5 and 6. This type of separation is
usually done by gravity in tanks which are similar to the oil/water
separators used in refineries. The first technique involves the mixing
of lighter oils to make the total hydrocarbon stream lighter and easier
to separate. The second is the use of fuel gas to create an upward
current in the separator. These techniques are widely used in ethylene
plants to separate insoluble hydrocarbon by-products from the cracked
295
-------
FIGURE VII-3
NON-CONDENSIBLE REMOVAL
CUSTOMARY - VACUUM JETS
COOLING
HATER
FLARE
POLLUTED WATER
ALTERNATE - VACUUM PUMP
TO FUEL-GAS HEADER OR FLARE
296
-------
FIGURE VII-4
WATER SCRUBBING
CUSTOMARY
CONTAMINATED
MATERIAL IN
ALTERNATE
CLEAN
MATERIAL
OUT
FRESH WATER
CONTAMINATED
MATERIAL IN
CONTAM
INATED
WATER
CLEAN
MATERIAL
OUT
FRESH WATER
LESS WATER, MORE
CONTAMINANTS
PER POUND
297
-------
FIGURE VII-5
OIL AND WATER SEPARATION
LIGHT-OIL ADDITION
LIGHT OIL
OIL AND WATER
MIXTURE
RELATIVELY LIGHT OIL
OIL TO
PROCESSING
WASTE WATER
298
-------
FIGURE VII-6
OIL AND WATER SEPARATION
FUEL-GAS ADDITION
OIL AND
WATER IN
PROCESS
GAS IN
FUEL GAS OUT
GAS
OIL
yfiSMfMSS^mZM^
GAS SPARGER
I
OIL OUT
WATER OUT
299
-------
gas quench water. Other systems such as filters and coalescers are also
used for this type of separation.
The separation of oil by gravity is a common unit process in the cleanup
of any oily waste water. The primary method of separation is to provide
holding time so that the flow can be maintained in a quiescent
condition. Typical efficiencies of oil separation units are presented
in Table VII-1.
Sjoen t_Ca us t i c_and_Oi ly Sludge Incinerator
The final disposal of spent caustic and oily sludges has been
successfully accomplished by using a fluid-bed incinerator. As the
sludge is burned, the solids remain in the bed, while the gaseous
products of combustion and water vapor discharge through the gas-
cleaning system. When the operation on oily sludge has been stabilized,
spent caustic is introduced. Water in the caustic solution is vaporized
and the combustible material is oxidized; the solids accumulate in the
fluid bed. The bed level is maintained by withdrawing ash as it
accumulates from the deposition of solids. Solids removed from the
process consist of iron oxide, sodium sulfate, sodium carbonate, and
other inert solids, and have been used for landfill. Stack gases from
the incinerator consists of water vapor, nitrogen, oxygen, carbon
dioxide, and a few tenths of a part per million of sulfur dioxide.
Various phenol recovery systems using solvent extraction, carbon
adsorption, and caustic precipitation are also described in Section IV.
These recovery processes are all associated with phenol manufacturing
processes.
Phenol Removal
Solvent Extraction
Solvent extraction has been used very effectively by the petroleum
industry to remove phenols from various streams. Some of these solvents
which have been used to extract phenols are aliphatic estesrs, benzene,
light cycle oil, light oil, and tri-cresyl phosphates. Among those
solvents, tri-cresyl phosphates are excellent solvents due to their low
solubility in water and their high distribution coefficients for phenol
but they are expensive and deteriorate at high distillation
temperatures. However, it might be used when high phenol recoveries are
desired for economic reasons. Most of the other solvents are
consierably cheaper to use in waste treatment operations. Several types
of extraction equipment such as centrifugal extractors, electrostatic
extractors, etc., are available and the type of extraction equipment
required for the use of a particular solvent is an important economic
consideration. Reported efficiencies of some solvent extraction for
phenol removal are given in the following tabulation.
300
-------
TABLE Vll-l
Typical Efficiencies of Oil Separation Units*
Oil
1 nf luent
(mq/L)
7000-8000
3200
400-200
220
108
108
90-98
50-100
42
Content
Effluent
(mq/L)
125
10-50
10-40
49
20
50
40-44
20-40
20
Oi 1 Removed
%
98-99+
98-99+
90-95
78
81.5
54
55
60
52
Type of
Separator
Ci rcular
Impounding
Parallel Plate
API
Ci rcular
Ci rcular
API
API
API
*''Petrochemical Effluents Treatment Practices," Federal Water Pollution
Control Administration, U.S. Department of the Interior, Program No. 12020—2/70.
301
-------
Typical Efficiencies for Phenol Removal by Solvent Extraction*
Phenol Phenol
Solvent Influent^ mg/1 Effluent^mcj/l Removal.
Aromatics, 15% 200 0.2 99.9
Paraffins, 25*
Aliphatic Esters 4,000 60 98.5
Benzene 750 34 95.5
Light Cycle Oil 7,300 30 90
Light Oil 3,000 35 99
Tri-cresyl Phosphates 3,000 300-150 90-95
*"Petrochemical Effluents Treatment Practices", Federal Water Pollution
Control Administration, U.S. Department of the Interior, Program No.
12020, February 1970.
Steam Stripping
Steam stripping method has also been successfully used in removing
phenol from waste streams. The method involves the continuous downward
flow of the waste water through a packed or trayed tower while the
stripping steam flows upward removing the desired constituent. The
removed phenols are recycled back to the appropriate process. This
stripping method can achieve at least a phenol reduction of 90 percent.
Chlorine Oxidation
Chlorine has been applied in oxidizing phenol in waste waters. The
oxidation of phenol must be carried to completion to prevent the release
of chlorophenols. An excess of chlorine is usually required because of
the reaction with various other chemical compounds such as ammonia,
sulfides, and various organics which can interfere with the chlorination
process. Despite the potential for formation of chlorophenolics,
chlorine can be used to completely (100%) oxidize phenolics under proper
conditions.
Ammonia and Sulfide Stripper
Removal of hydrogen sulfide and ammonia from sour water can also be
accomplished by stripping methods. Most of these stripping methods also
involve the continuous downward flow of the waste water through a packed
or trayed tower while the stripping gas or steam flows upward removing
the desired constituent. steam is considered to be the preferred
heating and stripping agent, since hydrogen sulfide, which is
concentrated in the steam condensate, may be further treated. Flue
gases are frequently used because carbon dioxide produces a slightly
stronger acid than hydrogen sulfide thus releasing hydrogen sulfide from
the solution. The typical removal efficiencies are:
302
-------
H2S removal 98-99+%
NH3 removal 95-97X
In many cases steam stripping may also remove as much as 20-40 percent
of any phenols present.
Cyan ide_ Removal
Cyanide can be oxidized to carbon dioxide and nitrogen by chlorination.
The waste water must be kept at a pH value greater than 8.5 during
treatment to prevent the release of toxic cyanogen chloride. The
reaction time usually is one to two hours and the process is subject to
the interference of various compounds such as ammonia, sulfides, and
various organic substances.
Qzone^Treatment
Ozone has been proposed as an oxidizing agent for phenols, cyanides, and
unsaturated organic substances, since it is a considerably stronger
oxidizing agent than chlorine. The chief disadvantages are the high
initial cost of the equipment for energy needs and cooling water
requirements for ozone generation. Ozone has several advantages, the
most important being its ability to rapidly react with phenol and
cyanide. The optimum pH for phenol destruction is 11 to 12.
Thiocyanates, sulfates, sulfides, and unsaturated organic compounds will
also exert an ozone demand which must be satisfied. This demand serves
as the basis of design for an ozonation unit treating a waste water
containing these compounds. Sulfides also can be removed from a waste
water which is to be ozonated by air stripping them at low pH values,
thus economically reducing the ozone demand. The pH of the waste water
can then be raised to the appropriate level required for optimum
ozonation.
Recent investigations have indicated the applicability of ozonating
wastes from the manufacture of chlorinated hydrocarbons. The optimum pH
for ozonation of this waste water was found to be 12.6, and as much as
90 percent of the waste COD was removed. This waste contains large
quantities of unsaturated hydrocarbons, which are readily amenable to
ozonation. Ozonation of a waste water can be either a batch or
continuous operation, depending on the characteristics of the waste and
the waste flow rate.
It must be appreciated that these systems are useful only for specific
processes and may not be recommended on a general basis. This is
definitely true when evaluating the possible use of activated carbon
adsorption as an in-process control measure. Table VII-»2 provides
orientation as to its widely varying effectiveness for specific
chemicals. This table illustrates the limited amenability of many
303
-------
Table VII-2
Relative Amenability To Adsorption Of
Typical Petrochemical Wastewater Const!tutents*
Percentage Removal
of Compound
Compound at 1000 mg/L at a 5 mg/L.
Initial Concentration powdered carbon dosage
Ethanol 10
Isopropanol 13
Acetaldehyde 12
Butyraldehyde 53
Di-N-propylamine 80
Monoethanolamine 7
Pyridine ^7
2-Methyl 5-ethyl pyridine 89
Benzene1 95
Phenol 8l
Nitrobenzene 96
Ethyl acetate 50
Vinyl acetate 64
Ethyl acrylate 78
Ethylene glycol 7
Propylene glycol 12
Propylene oxide 26
Acetone 22
Methyl ethyl ketone ^7
Methyl isobutyl ketone 85
Acetic acid 2k
Prop ionic acid 33
Benzoic acid 91
^Benzene test at near saturation level, 420 mg/L
*Conway, P.A. etc., "Treatabi1ity of Wastewater from Organic
Chemical and Plastics Manufacturing - Experience and Concepts,"
Research and Development Department, Union Carbide Corporation,
South Charleston, West Virginia, Feb. 1973.
304
-------
common low molecular-weight, oxygenated chemicals to adsorption on
activated carbon.
Incineration of Chlorinated Hydrocarbon^
There are a limited number of devices currently available for burning
waste chlorinated hydrocarbons with the recovery of by-product HCL. In
the past, the traditional disposal routes for these waste materials have
been ocean discharge, open-pit burning, drum burial, and deep-well
injection. Recently, more stringent regulations have disallowed many of
these methods. Subsequently, there has been an increase in activity by
industry aimed at the development of systems for these hard-to-treat
wastes. The weight of these materials is estimated at 350,000 tons/year
of chlorinated hydrocarbon residues generated during production of
almost 10 million tons/year of chlorinated hydrocarbons by chemical
companies.
It should be noted that there are still serious drawbacks associated
with most incineration systems. These relate to both the emissions from
the systems as well as corrosion and other operating difficulties. The
following paragraphs describe the systems currently utilized. It. is not
clear whether or not systems such as these truly represent a viable
alternative for the disposal of hard-to-treat wastes. However,
incineration is an alternative which will receive additional consider-
ation by manufacturers whose processes generate concentrated reduced
volume waste streams.
More chemical companies now incinerate wastes that cannot be treated.
For example, one chemical company uses a high-temperature incincerator
to dispose of polychlorinated biphenyls. Another chemical company has
developed an efficient tar-burning unit, and is selling know-how. A
system based on this technology was recently completed.
Some plants have also added scrubbers to clean emissions from
incinerators. But for highly chlorinated hydrocarbon wastes—i.e.,
those containing more than 50% chlorine--the emission of gaseous
hydrogen chloride is more than ordinary incinerator-scrubber units can
cope with.
For example, a neoprene plant at one time operated a horizontal
incinerator and vertical scrubber with a packed column in the stack.
Maintenance costs were excessive (about $40,000/year) and hydrogen
chloride emissions were too high.
This plant has since turned to the only system for chlorinated
hydrocarbon disposal and by-product recovery now operated in this
country, the system.
305
-------
Four units are now operating at different chemical plants. In addition,
another unit is scheduled to go on stream shortly. There is only one
company which is not recovering by-product hydrogen chloride. The
company decided against recovery because high-pressure operating
conditions at the plant would have required the addition of equipment to
compress the gas stream before stripping hydrochloric acid.
The system incincerates chlorinated liquid waste, cools the combustion
gases, strips the aqueous product and turns out anhydrous HC1.
Hydrogen chloride gas is soluble in water. But, absorption is
complicated by the heat generated in large quantities during combustion.
For example, 36.4 million BTUs/hour must be removed from a 4,000 Ibs/hr
unit.
In the system, the sticky chlorinated hydrocarbon residue is atomized
and incinerated in a combustion chamber that has a vortex-type burner
supplied by Thermal Research. The incinerated material is cooled from
2,500°F to 800°F in a graphite cooling chamber, where it is sprayed with
27% HC1.
The cooled gas passes through three falling-film acid absorbers made of
impervious graphite. Stripped liquid is recycled through the absorbers
in reverse order, removing heat of absorption and HC1 from the gas
stream.
Gas from the last absorber enters a final scrubber to reduce HC1
emissions about 5 ppm. This scrubber is 5 ft. in diameter, contains 3
ft. of 1-inch-diameter plastic packing and includes a spray header and a
demister made of polypropylene.
At some plants, the gas is released to the atmosphere through a stack
designed for silencing the exhaust. It is a packed centrifugal unit
with a diameter enlargement before the stack outlet to reduce gas
velocity and permit entrained liquid particles larger than 100 microns
to settle out.
The major problem with units has been the junction between the
combustion and cooling chambers. The carbon blocks of the cooling
chamber oxidize at 75O°F and all parts of the chamber must be covered
with liquid. If the spray is not properly adjusted, liquid HC1 backs up
into the combustion chamber and attacks the mortar joints and steel
outer shell. A ceramic sleeve is now used to protect the furnace
refractory at the joint from the HC1 spray.
One company has also switched from field-erected to preassembled cooling
chambers. Field-erected units were made of dense (100 Ibs./cu.ft.)
carbon blocks, keyed together by graphite rods, cemented with a special
carbonaceous cement and reinforced by rubber-covered steel bands. The
306
-------
preassembled chambers have graphite wall units, eliminating the
possibility of leaky joints.
From a pollution-control standpoint, the most significant change that
can be made in process chemistry is from a "wet" process to a "dry"
process, that is the substitution of some other solvent for water in
which to carry out the reaction or to purify the product.
If any organic solvent can be used, the process can probably be worked
out to produce an organic concentrate that will contain all the
undesirable impurities and by-products. Their disposal in an organic
concentrate is much simpler and cheaper than coping with them in an
aqueous medium. Incinceration costs for descruction of organic
concentrates by contractors usually run between $0.01/lb. and $0.03/lb.,
depending on the halogen content and the presence of other inorganic
compounds.
If water must be used in the process, its use should be restricted, and
every opportunity for the replacement of fresh water with recycle water
should be explored and implemented. (This is especially important in
the inorganic chemical processes.) Use of water can be restricted by
countercurrent washing techniques. Discarding of waste water used for
pruifying a reaction product when fresh water is used for the reaction
medium is also uncalled for. Similarly, another useful water-
conservation practice is collection of vacuum-jet condensate, rain
water, and floor water for reuse.
Another process change that can yield significant pollution-control
benefits is the elimination of troublesome by-products by a change in
the reactants, or a change in the catalyst. An example of the former is
the emergency of oxychlorination processes (that generate by-product
hydrochloric acid).
From these discussion it is apparent that significant reductions in the
quantity of pollutants generated by a process are possible.
Quantitative estimates for specific processes indicate that in some
cases waste water flows can be reduced to approximately 10 gallons/1,000
Ib of product, and corresponding COD loadings of 0.1 Ib of COD/1,000 Ibs
of product. In some specific cases the discharge of pollutants can be
reduced to near zero through the use of by-product recovery processes
such as adsorption. Such systems generally take advantage of the
specific characteristics of the chemicals in questions. It is not
possible to specify a uniform restriction based on such systems that
could be applied throughout the indistry, or even one category.
End-of-PiQg_Treatment System
Gengral^Considerations
One of the initial criteria used to screen organic chemicals plants for
the Phase I field survey was the degree of treatment provided by their
307
-------
waste water treatment facilities. Therefore, the selection of plants
was not based on a cross-section of the entire industry, but rather was
biased in favor of those segments of the industry that had the more
efficient waste water treatment facilities. A summary of the types of
treatment technology which were observed during the survey are listed in
Table VII-3. Of the plants surveyed in Subcategories A, B, and C over
90 percent provided their own waste treatment facilities while 50 to 60
percent of the Subcategory D plants discharged to municipally owned
treatment facilities.
During the survey program, waste water treatment plant performance
history was obtained when possible. The sampling data obtained during
the survey were then used to verify each plant's analytical procedures.
The historical data were analyzed statistically, and the individual
plant's performance evaluated in comparison to the origianl design
basis. After this evaluation, a group of plants were selected as being
exemplary, and these plants were presented in Tables VII-3 and VII-4.
The treatment data in Table VII-4 represent the average historic
treatment plant performance (50% probability of occurrence) based on a
thirty-day average reporting period, and the data in Table VII-5
represent the sampling data obtained during the plant survey. It is
true that the treatment data in Tables VII-4 and VII-5 were obtained
from plants producing multi-products from more than one category.
However, based on the great majority of the products produced in the
plant, most of the plants could be directly associated with a particular
category.
In preparing the economic data base, all the waste water treatment plant
data were analyzed to develop a basis for subsequent capital and
operating costs.
The treatment plant data presented in Tables VII-4 and VII-5 were
evaluated on the basis of similar categories and this resulted in the
generation of treatment efficiencies for BPCTCA and BATEA affluent
levels. These required treatment efficiencies will be presented in the
succeeding sections.
A review of the exemplary waste water treatment systems indicated that
the biological treatment system is typified by a variety of treatment
systems such as physical, chemical or biological waste treatment
systems. In order to measure the economic impact of the proposed
effluent standards, a series of model treatment systems were developed
for each subcategory and were sized to remove 95, 90 and 85 percent of
the influent BOD.
The end-of-pipe treatment models were designed to cover the range of
actual contact process waste water flows which were encountered within
308
-------
Table VII-3
Organic Chemicals Study
Treatment Technology Survey
Number of
Plants Observed
7 Activated Sludge
1 Activated Sludge-aerated lagoon
1 Activated Sludge-polishing pond
1 Trickling Filter-Activated sludge
J> Aerated lagoon-settling pond
2 Aerated lagoon-no solids separation
k Facultative Anaerobic lagoon
1 Stripping Tower
3 No current treatment -
system in planning stage
5 To Municipal Treatment Plant
2 Deep-we11 disposal
k_ Physical Treatment, e.g. API Separator
4 Total
309
-------
c
.
-a-
*"^
0)
H
fH
H
O
1-
« CM
1 1 1 1 1 1 vO 1 •
i
a-v i cr\ i i i oo (Tv en i
0 1-
— Q-
U
o &s
4J CO
(fl LA
X
8
4->
C
0) -1
3 \
— O
£ E
IU
o o o o i ii
cs r^ o CN PA LA i o i i
PA .* CM T- CO r^ 1 CO 1 1
(Q
Q
i
DC
^ CM LA LA
• • • 1 1 • 1 1
LA vO PA .* ff\ 1 1 -* 1 1
p~< cr\ vo vo r*- i i r** i i
«". o
U. 0-
en
c:
o
L.
H-
O)
c
o
a.
a.
*
a.
ai
T3
3
o
cr
a)
4J
(D
o
QOQOQOOQQOOQO
CQ
m
in
>.
to
a.
i
)
•
<
i
310
to
LU
-------
CO
CO
«
3 '
U-
LU
ad
O
0
CM
0
LTV
LA
0
CO
00
fO
OO
T
— cn -3-
co cn co
CO I
cn
oo
o
CM
vO
i
o
I-
c
- CM
cn
CM
vO
4
0)
H
JO
H
(D
O
i-
3
CO
C
(D
0)
E
re
o>
o
CO
c
-------
each of the process subcategories. The tabulation below summarizes
these ranges:
Process wastewater^Flow fgpd)
Minimum Average Maximum
Continuous Process
Subcategories A, B, and C 7,200 360,000 2,160,000
Batch Process
Subcategory D 72,000 360,000 720,000
There is an approximate correlation between the actual waste water flow
and flow RWL as expressed in production units. Generally, the highest
actual flow rates are generated by those processes which produce the
highest flows per unit of product.
BPCTCA Treatment Systems
The single stage activated sludge system was chosen as a model system
for BPCTCA and a general flow diagram for the waste water treatment
facilities is shown in Figure VII-7. A summary of the general design
basis is presented in Table VII-6.
312
-------
Q
2
I
M
H
M
H
313
-------
Table Vll-6
BPCTCA Model Treatment System Design Summary
Treatment System Hydraulic Loading
(capacities covered, in gpd)
7,200 360,000
43,200 720,000
72,000 1,440.000
216,000 2,160,000
Pump_ Station
Capacity to handle 200X of the average hydraulic flow
Equalization
One day detention time is provided for Subcategories A, B, and
C, and three days for Subcategory D. Floating mixers and
provided to keep the content completely mixed.
Neutral izgt4on
The two-stage neutralization basin is sized on the basis of an
average detention time of twenty minutes. The lime-handling
facilities are sized to add 2,000 Ib of hydrated lime per mgd
of wastewater, to adjust the pH. Bulk-storage facilities
(based on 15 days usage) or bag storage is provided, depending
on plant size. Lime addition is controlled by two pH probes,
on in each basin. The lime slurry is added to the
neutralization basin from a lime slurry recirculation loop.
The lime handling gacilities are enclosed in a building.
Facilities are provided for the addition of phosphoric acid and
aqua ammonia to the biological system in order to maninain the
ratio of BOD:N: at 100: 5: 1/
Platform- mounted mechnaical aerators are provided in the
aeration basin. In addition, concrete walkways are provided to
all aerators for access and maintenance. The following data
were used in sizing the aerators:
314
-------
Oxygen utilization 1.5 Ib 02/lb BOD removed
alpha factor 0.9
beta factor 0.9
Wastewater Temperature 20 C
Oxygen transfer 3.5 Ib 02/hr/shaft hp
at 20 C and zero DO in tap water
Motor Efficiency 85%
Minimum Basin DO 1 mg/1
Oxygen is monitored in the basins using D.O. probes.
Secondary, Clarifjers
All secondary clarifiers are rectangular units with a length-
to-width ratio of 3 to 4. The side water depth is 10 ft. and
the overfolw rate varies between 100 and 500 gpd/sq ft
depending on plant size. Sludge recycle pumps are sized to
diliver 100* of the average flow.
^i£ Flotation
The air flotation units recommended for Subcategory C plants
are sized on a solids loading of 20 Ibs/sq/ft/day. In addition, liquid
polymer facilities are provided to add up to 50 mg/1 of polymer to
enhance solids separation.
Sludge Holding Tank-Thickener
For the smaller plants, a sludge~holding tank is provided, with
sufficient capacity to hold 5 days flow from the aerobic digester. The
thickener provided for the large plants was designed on the basis of 6
Ib/sq/ft/day and a side water depth of 10 ft.
Aerobic Digester
The aerobic digester is sized on the basis of a hydraulic
detention time of 20 days. The sizing of the aerator-mixers
was based on 1.25 hp/1,000 cu ft of digester volume.
Vacuum Filtration
The vacuum filters were sized on a cake yields of 2
Ib/sq/ft/hr, and a maximum running time of 18 hr/day. The
polymer system was sized to deliver up to 10 Ib of polymer/ton
dry solids.
§i Sludge Disposal
Sludge is disposed of at a sanitary landfill assumed to be 5
miles from the wastewater treatment facility.
315
-------
2§§i3D Philosophy
The plant's forward flow units are designed for parallel flow,
i.e., either half of the plant can be operated independently,
thus providing reliablility as well as flexibility in
operation. The sludge facilities are designed on the basis of
series flow. All outside tankage is reinforced concrete. The
tops of all outside tanks are assumed to be 12" above grade.
The following is a brief discussion of the treatment technology
availagle and the rationale for the selection of the previous unit
processes to be included in the BPCTCA treatment system. An optional
API oil-water separator was sized for an average flow of 720,000
gal/day. This was done in oder to indicate the percentage increase in
the total BPCTCA cost which would be involved as a result of excess
floating oil.
The pump stations are generally located after the API separator so as to
avoid emulsifying the floating oil. Topography of a particular plant
site will dicate whether pumping equipment is required., Equalization
facilities are provided in order to minimize short-interval (e.g.,,
hourly) fluctuations in the organic loading to the tieatrnent plant, a:.;
tf.jll as 'co absorb slug loads from reactor cl,*:arouts,<. accidental spijls,
ojt. and to minimize the usage of nf-ntrrjl i;:ati.cm "henteal",
*;hrfe&-day detention time bas€<3 on ^-.•^raw. flow in p.i<~~/ld
'•;).•'., rugoT. y £• flow?, in con tr
i „ a-?-cit;; ht B, aari C, This Ine.f!-
tor f. r, * -!.
r
4- \~,f-
"",£ -• - S »• — V,- -.- a- £-* -J~ ^ -VX « . -- ,-,*,-. ^ I *
/..Is ~ ;. biological treatment. Alkal j ;rr: -^ouh
1:. r,.uj irrsi: of hydrated lime sto-i.-ia.-; 115,3
c." arii'lcatiori facilitiec wex-e rot included !•-. t.he '- ^r ill 1: ies for th'-.
cost; estimate, because the TSS RWL data indicated that it would not be-.
necessary to remove TSS before biological treatment. It should be note1
that a plant's particular product mix should be evaluated before a
decision is made to omit this process.
An activated sludge process was selected for the biological portion of
the system. However, single stage activated sludge is not the only
system which may be applicable nor should it be construed as being
totally applicable to all process wastewaters which may be generated in
the industry. In addition, aerated lagoons are also applicable to meet
316
-------
the proposed effluent requirements. The following is the rationale for
selection of activated sludge and its inclusion in the cost estimates.
Both aerated lagoons and activated sludge processes involve aeration
basins, the major difference being that aerated lagoons normally
discharge directly without a clarification step. This results in a much
lower concentration of microorganisms in the aerated lagoon than in an
activated sludge basin because the microorganism mass is not
recirculated back to the aeration basin. Therefore, for camparable
organic loadings, a much larger aerated lagoon would be required to
provide treatment equivalent to that of an activated sludge plant. In
actual practice, the aerated lagoon process has found wide application
where it is not necessary to provide a mentod of sealing the lagoon or
where the soil characteristics are such that they form a natural seal.
The activated sludge process was selected for cost estimating purposes
in order to provide an estimate of the economic impact of the proposed
effluents limits on any treatment facility including the proposed
effluent slimis on any treatment facility including those for which it
is necessary to make the basin linings impervious.
Beside activated sludge and aerated lagoons, various other combinations
of biological treatment processes may be utilized. The combination of
available unit processes are numerous and the treatment scheme should be
selected only after a thorough engineering as well as economic analysis.
In the biological process, for every pound of BOD removed from a waste
water, approximately 0.6 pound of biological solids is produced, and
this must be removed from the system. In many areas where aerated
lagoons are applicable, settling lagoons are used to separate these
biological solids.
These settling lagoons are periodically dewatered and dredged, and the
dredgings are pumped to sludge-holding lagoons. (In the activated
sludge process sludge wasting is done daily to dewatering facilities).
These sludge lagoons may act as solar evaporation ponds or as drying
lagoons if the water and precipitation are decanted off. The sludge
will generally dewater naturally if applied at depths less than 15
inches. The dewatered sludge may then be landfilled, or the existing
lagoon can be covered with earth and a new lagoon constructed. Which
alternative is used depends upon individual state1 sanitary landfill
requirements and the possiblity of ground water pollution through
contamination with leachate.
Many plants in the United States are so geographically situated that
aerated lagoons provide a viable alernative. However, in order to make
the subsequent cost estimates more meaningful and universally
applicable, activate sludge was selected. In addition, because the
subcategory C wastes are so concentrated and the TSS mixed liquor levels
are so high (as previously discussed), an air flotation unit is included
to facilitate solids separation after the secondary clarifier.
317
-------
Activated sludge facilities pose a distinct sludge disposal problem.
Because biological sludge must be wasted on a daily basis from most
larger plants, it creates a handling problem which often cannot be
solved as expeditiously as with an aerated lagoon.
The relatively small amounts of sludge generated by the BPCTCA plants
dictated the selection of the most viable sludge disposal alternative.
The quantities of sludge are such that sludge incineration cannot be
justified because of the very small equipment sizes that would be
involved. Therefore, it was decided, for smaller plants, to aerobically
digest the sludge, decant the supernatant, and then vacuum filter the
sludge. The sludge cake (which would be at least 2Q% solids) would then
be acceptable at a sanitary landfill.
For the larger plants, sludge thickening is provided to concentrate the
waste activated sludge from 0.8* to 2.0% before digestion. Thickening
is not applicable for the smaller paints because the sludge quantities
are such that odors could develop because the small equipment would
result in long detention times. For these cases, a sludge holding tank
is provided in order to add flexibility to the operation. During vacuum
filter down time, digested sludge may also be held without upsetting the
solid handling facilities.
BATEA Treatment Systems
As previously discussed in in-process recovery systems, activated carbon
has possible applications in the organic chemicals industry for in-plant
recovery of specific chemicals. In addition, activated carbon has also
been demonstrated in many cases that it can be used as an end-of-pipe
waste water treatment technology.
During the plant survey program, seven samples of the individual
industrial treatment plant effluents from industrial Subcategories B, C,
and D were obtained and carbon isotherms were performed using powered
activated carbon. The raw carbon isotherm data are presented in
Supplement B, and a summary of the analytical results are presented in
Table VTI-8. The complexity of the organic chemical industry is
evidenced by the results of the carbon isotherms. The initial COD
varied tenfold, while the variation in carbon exhaustion rate was over
one hundred fold.
The carbon adsorption isotherm is widely used to screen various
activated carbons and to quantify overall removal efficiencies.
However, the exhaustion capacity, generated by a carbon isotherm is not
sufficient to be used for design purposes. In pilot scale column
operations, the following factors should be recognized: the limitations
inherent in extrapolating laboratory data to multi-column systems, and
the problems of channeling and wall-effects that limit the utility of
data taken in small diameter laboratory columns. Ideally, pilot-plant
318
-------
continuous column studies should be run to generate reliable design
data.
The carbon isotherm data shown in Table VII-8 indicated that, except in
the particular case of Plant 6, COD removal efficiencies by carbon
adsorption of biological treatment plant effluents are well above 70
percent. In addition, it is known that additional organic substances
can be degraded through biological activity which occurs in the carbon
bed.
In order to develop BATEA effluent criteria, the activated carbon system
in addition to BPCTCA treatment systems was chosen as a model system for
BATEA for Subcategories A, B, C, and D.
BATEA effluent criteria may be attained in actual practice via a number
of possible routes. In order to quantify the impact of BATEA criteria
on the individual categories, two treatment systems were designed and
subsequent cost data reported in section VIII. A general flow diagram
for the BPCTCA treatment system for Subcategories A and B is shown in
Figure VII-8 and for Subcategories C and D in Figure VII-9. A summary
of the general design basis is presented in Table VII-7.
Dual-media gravity filtration is provided since activated carbon
typically requires that the concentration of TSS be 50 mg/1 or lower in
order to maximize carbon adsorption and minimize the filtration
function. High TSS would involve shortened filter runs and increased
amounts of backwash water usage.
319
-------
Table VII-7
BATEA^End of Pipe Treatment^System
Design Summary
Filtration
The filters are sized on the basis of an average hydraulic loading of 3
gpm/sq ft Backwash facilities are sized to provide rates up to 20 gpm/sq
ft and for a total backwash cycle of up to 20 minutes in duration. The
filter media are 24" of (No, 1 1/2) and 12" of sand (0.4-0.5 mm sand).
Granular Carbon Columns
The carbon columns are sized on a hydraulic loading of 4 gpm/sq ft and a
column detention time of 40 minutes. A backwash rate of "20 gpm/sq ft
was assumed for 40* bed expansion at 70°F.
Design Comments:
Subcategory A and B are fixed-bed downflow units, while the
Subcategories C and D systems are pulsed-bed upflow unit, with the
carbon being wasted over a prescribed time sequence, e,g, wasted for
15 minutes every two hours.
Filter.-SSlHSHl Bezant SuffiE
Tanks are provided to hold the backwash water and decant it back to the
treatment plant over a 24 -hour period. This will eliminate hydraulic
surging of the treatment units.
Regeneration £U£nace
The following exhaustion rates were used for the sizing of the
regeneration facilities:
lafiasat C.QB jj&baa^ioj
Subcategory mg/1 lb COD/lb carbon
A 100 4.5
B 120 4.5
C and D 1200 0.35
These exhaustion capacities were selected, based on the carbon isotherm
data previously presented in Table VII-8.
A multiple-hearth furnace is employed for regeneration of the carbon
only for Subcategory D. The quantities of carbon exhausted based on the
320
-------
previous exhaustion capacities for Subcategories A and B are not
sufficiently large to warrant the investment in a regeneration furnace.
Regenerated Exhausted Carbon Storage
Tanks are provided to handle the regenerated and exhausted carbon both
before and after regeneration.
Table VII-8
Carbon Isotherm Data Performed on Individual Bfologica
Treatment Plant Effluents
.Plant ID
J
2
3
j
4
5
_
Industrial
Category
B
D
C
C
B
B
B
Initial
Soluble
COD cone
mq/L
146
304
525
573
774
972
1297
Final
Soluble
COD cone
mq/L
19
21
150
146
97
758
397
Exhaustion Capacity
% Removed
87
93
71
75
87
22
70
Ibs COD Removed
Ib. Carbon
4.5
1.35
0.35
0.36
0.50
0.035
0.34
Ibs Carbon
1000 Gallons
0 27
v « ff /
1.87
1 • V /
12.2
13 3
1 J * J
18.7
232
32.2
The treatment plant effluents were filtered to insure the removal of all insoluble COD.
321
-------
u
,KL
Ii
322
-------
ei
Q
CTi
I
U
s
u w
,w a
a
323
-------
Section VIII
COST, ENERGY, AND NONWATER QUALITY ASPECTS
This section provides quantitative information relative to the suggested
end-of-pipe treatment models.
The cost, energy, and nonwater quality aspects of in-plant controls are
intimately related to the specific processes for which they are develop-
ed. Although there are general cost and energy requirements for equip-
ment items (e.g. surface air coolers), these correlations are usually
expressed in terms of specific design parameters, such as the required
heat transfer area. Such parameters are related to the production rate
and specific situations that exist at a particular production site.
Reference to the Tables in Section IV, which show plant sizes for spe-
cific process modules, indicates that even in the manufacture of a sin-
gle product there is a wide variation between process plant sizes. When
these production ranges are superimposed on the large number of
processes within each subcategory, it is apparent that many detailed
designs would be required to develop a meaningful understanding of the
economic impact of process modifications. Such a development is really
not necessary, because the end-of-pipe models are capable of attaining
the recommended effluent limitations at even the highest RWL within any
subcategory. The decision to attain the limitations through in-plant
controls or by end-of-pipe treatment should be left up to individual
manufacturers. Therefore, a series of designs for the end-of-pipe
treatment models are provided. These can be related directly to the
range of influent hydraulic and organic loadings within each
subcategory.
The range of costs associated with these systems can then be divided by
the range of production rates for any single process within any
category. This will show the maximum range of impact on the required
realization of any single product (i.e. the range of impact in terms of
$/lb of product). Total industry coat for BPCTCA is estimated at $1,030
billion ("Economic Impact of Water Pollution control an this Organic
chemicals Industry, "Arthur D. Little, Inc., Cambridge, Mass., 1973).
It is estimated that this cost includes a substantial portion of capital
investment as of 1973.
The major nonwater quality consideration which may be associated with
in-process control measures is the use of alternative means of ultimate
disposal. As the process RWL is reduced in volume, alternate disposal
techniques such as incineration, ocean discharge, eind deep-well
injection may become feasible. Recent regulations are tending to limit
the applicability of ocean discharge and deep-well injection because of
the potential long-term detrimental effects associated with these
disposal procedures. Incineration is a viable alternative for
324
-------
concentrated waste streams, particularly those associated with
Subcategory C. Associated air pollution and the need for auxiliary
fuel, depending on the heating value of the waste, are considerations
which must be evaluated on an individual basis for each use.
Other nonwater quality aspects, such as noise levels, will not be per-
ceptibly affected. Most chemical plants generate fairly high noise lev-
els (85-95 dB(A)) within the battery limits because of equipment such as
pumps, compressors, steam jets, flare stacks, etc. Equipment associated
with in-process or end-of-pipe control systems would not add
significantly to these levels, in some cases, substituting vacuum pumps
for steam jets would in fact reduce plant noise levels.
As discussed previously, design for the model treatment systems proposed
in Section VII were costs estimated in order to evaluate the economic
impact of the proposed effluent limitations. The design consideration
(namely, the influent RWL) was selected so that it represented the
highest expected RWL within each category. This resulted in the
generation of cost data for each level of technology
Activated sludge was proposed in Section VTI as the BPCTA model
treatment system. The plant designs were varied to generate cost
effectiveness data within each subcategory. Dual-media filtration and
activated carbon adsorption were proposed in Section VII as best
available technology economically achievable (BATEA) treatment for Cate-
gories A, B, C, and D. New source end- of- process treatment involves the
addition of dual media filtration to biological waste treatment model
processes.
Capital and annual cost data were prepared for each of the proposed
treatment systems previously discussed in Chapter VII.
The capital costs were generated on a unit process basis, e.g. equali-
zation, neutralization, etc. The following "percent add on" figures
were applied to the total unit process costs in order to develop the
total capital cost requirements:
Percent of ^Unit
Process Capjtal~Cost
Electrical 12
Piping 15
Instrumentation 8
Site work 3
Engineering design and
Construction supervision fees 15
Construction contiguency 15
Land costs were computed independently and added directly to the total
capital costs.
325
-------
Annual costs were computed using the following cost basis:
Item
Amortization
Operations and
Maintenance
Power
Cost_Allocation
20 years for capital recovery at 8 percent
(10.2% of capital costs)
Includes labor and supervision, chemicals, sludge
hauling and disposal, insurance and taxes (com-
puted at 2 percent of the capital cost) r and
maintenance (computed at 4 percent of the capi-
tal cost).
Based on $0.02/kw hr for electrical power. Only
BATEA Subcategory D (activated carbon regeneration)
has a fuel oil allocation.
The following is a qualitative as well as a quantitative discussion of
the possible effects that variations in treatment technology or design
criteria could have on the total capital costs and annual costs:
Techno j.ogYQr-Pgsiqn Criteria
1. Use aerated lagoons and sludge de- 1.
watering lagoons in place of the
proposed treatment system.
2. Use earthen basins with a plastic 2.
liner in place of reinforced con-
crete construction, and floating
aerators versus platform-mounted
aerators with permanent-access
walkways.
3. Place all treatment tanks above
grade to minimize excavation, es-
pecially if a pumping station is
required. Use all-
steel tanks to minimize capital
cost.
4. Minimize flow and maximize concen-
trations through extensive in-plant
recovery and water conservation, so
that other treatment technologies
(e.g. incineration) may be economi-
cally competitive.
3.
The cost reduction could be
60 to 70 percent of the pro-
posed figures.
Cost reduction could be 10
to 15 percent of the total
cost.
cost savings would depend
on the individual situation,
Cost differential would de-
pend on a number of items,
e.g. age of pleint, accessibil-
ity to process piping,
local air pollution
standards, etc.
326
-------
The recommendation of a level of treatment for BPCTCA comparable to
biological treatment fixes the minimum organic removal (expressed as
BOD5) at approximately 90 percent.
The total cost requirements for implementing BPCTCA effluent standards
are presented in Table VIII- 1. Annual cost adjustment factors are also
shown for 95, 90, and 85 percent removal BODJ. These factors are shown
below:
Percent
Removal BOD 5
A B C D
95 1.19 1.0 1.0 1.00
90 1.00 0.84 0.88 0.88
85 0.86 0.72 0.87 0.87
All cost data were computed in terms of August, 1971 dollars, which cor-
responds to an Engineering News Records Index (ENR) value of 1580. The
model treatment system is activated sludge.
The following costs data were abstracted from the preceeding table for a
flow of 720,000 gpd and the treatment system required to meet the re-
commended BPCTCA effluent criteria:
Subcateqory Capital Cost ______________ AnS!3ai_Co§ts _____________
$ I/ill* ~ i/iooo_aii $/ib BODS iircent
BOD 5 Removed
A 1,410,000 284,300 1.08 0.78 90
B 2,538,000 487,900 1.86 0.27 95
C 8,144,000 1,657,000 6.31 0.17 95
D 1,878,000 341,900 1.30 0.25 95
The following production capacities were selected for calculating
the $/lb BOD5 removed: Subcategory A- 10 million Ib/day,
Subcategory B-5 million Ib/day, Category C-1 million Ib/day,
Subcategory D-0.05 million Ib/day.
Higher annual costs for Subcategory C reflect present technology in the
industry toward water reuse, which tends to generate very concentrated
waste waters. These waste waters require relatively longer aeration
times and more extensive sludge handling facilities. As indicated
above, any criterion (such as flow) which does not take into consider-
ation the amount of organic removal (e.g. Ib BOD5_ removed/day) , will not
be meaningful in describing the treatment system. The preceeding data
on decreasing annual unit cost illustrate treatment system economies of
scale.
Total costs as $/year, $/1000 gallons and $/lb BOD5 Annual costs and
effectiveness data for EPCTCA are shown in Table VIII-2 for 95, 90, and
327
-------
r- i
IB
Ej Q
«rt i |
w o
OS S5
IH B
o
w w
fr^
*l
9»
H >
&
8
H
1
4J
•r(
U
cd
ft
c?
u
a
o
•H
4J
0
3
o
cu
LA)
H«i CT>CO vOr>-rH lAf^tO «— (ONU'N
gp csr^ Oi
W- OOO HOP^(- OJLAH LAvS^t M
HCVJ (-(^t— LAVOON Hcvjrn a
f« »• 4.
H cvj R
•s
OOO W
8O OOO OOO OOO
O OOQ OOO OOO
QQ OOO OOO • K d
(9- «>»> r>«» fi « r OO 00 OO PO
OOO ONOO-* 1A-*O CVlLAt—
CO H CVJ 00 LA QN-^ O\ CO ^t CO >•
LA-sJ1 \OLAt~ OOHCVJ •>•> H
f> •>•< rf-o H H Q
H CVJ 00 CVJ OO 00 5
I—I fli
n *w
!
5
j*( pO (^ £^
t>cv| fc'cvj fc*cvj hcvio •<
H t1™ CV( K t^ O4 VO X t~* CVJ \Q C t^" VO CVJ ••
•o 5 o t— E ol>—H MI oi^-H H. ooot— n >•
VM «t i ... Jn* ... TT ... u r
S oo S oocvj »{ oocvj 2J ooo d
» 8 ' 1 1 1 1
5 % | •§ •§ ft
H ^ M CO CO +> J
fi ^
S
9 Ma
5> «
'D O
j»j !•
* • •
H O O
9\ o \e
i-l O 00
H H O
1
1
LA O LA
ON O\OO
-------
<"
p.
N
1
<
E-
^
CO
CO
7 §
w e
n &
> S
a S
•3 w
C_j
^
*
CO
H
CO
o
o
H
3
0
H
g
en
£]
H
§
M
u
M
4-1
M
4-1
Pi
.
d
T3
If
CM
1^.
O
d
C
o
•H
4J
n\
iS
4) t_.
PQ
<< :g
cd
O
H
PJ-I
P
J
H
O
5
'000 S
OOO OOOOP1 OOOOrH H
en en in^ CM o o> oo rS it co
00-*-* r-5r-HO OOO CO
ro co sr <
fO CM CM
H
CJ
<
g
c
p:
Fj"
E-
<
3
OOO ._
f ^& .— \ V.
r^voin r-O\cM cMtom CO
,«^> ooom P-ICMCM W
SoS -*-*M ooo Q
rH i-l £
p
g
W
£
[•/
a1 I
o r
M 8
p^
M ^
ir>oin ioom
CT»O>OO OCTiOO CT\O>OO
* * *
OOO
OOO srt^.vo SIJ^
•»•»* ••• »..
invovo ooo ooo
•* CM CO
r^ vo if)
ooo
ooo
O\ 00 CO vo vo *^" r-^ rs^ fs.1
««« 0010CO CMtMCM
r^ er> H ... ...
COOlO ^^.^ ooo
OOO o rts n\
ooo OOVOCM voS-a-
r^o>r^ -»r^cM odo
r^oo-^" ^coco c> a f~>
i-H ON 00
I-l
ooo ooo ooo «>
rOvOO» covocn cOvOCTi c!
o
•H
^J
fl)
4J
•H
5
H
4-1
S
2{
p
t— 1
in
4-1
ioom iooio ir>oio
CTtONOO vTvCT^OO CTNCT^OO T3
* * * (1)
•o
c
o
m
CO
1 t
CO
o
o
Annual
CO
g
rH
i-H
>
to-
o
o
o
H
«>-
CO
C
0
I-l
1
S1
o
CO
XI
H
vt-
cd
M
Removed
S-l
«d
>
>
O
O
O
i-H
•co-
1-1
1-1
O
S'
o
m
xi
r-l
>
M-l
CO
*4
Removed
CO
m
M
*
329
-------
w
E-i
co
H
O
H
8
CO
OOO O v£) CM
OOO t^ CM CM
CO CM O ...
f> * * co co co
P*> t^ 00
rH vo CO
O\ IO IO
CM CM CM
o co CA
co CM CM
"O O O
OOO
OOO
O O vO
1/1 10
H rH H
OOO
a)
« H
4-1 J3
id id
Q rH
ooo
ooo
o oo io
CS\0»0
10-*-*
(S
O vO CO ffi
IO •* -* CJ
OOO E-
C> O O K
Q
JH
1
4J -H
8^
a H
id i-l
4-1 J3
10 0)
°3
o at
ooo
ooo
r-
O
co
o
CO
ooo
ooo
§SS gs3S
SK5 cirA<-;
CM CM CM
C7> GO ON
rH r-l r-l
OOO
ooo
000
rH
ml
Q
o
PQ
J3
rH
Qd
id
cu
-------
85 percent removal BOD5. Effluent concentration BOD5 is also shown for
each removal efficiency and subcategory.
Depending on the particular production mix of the individual plant,
floating oil could be a treatment consideration. For that reason, an
API separator was sized for 720,000 gpd. The capital cost of the sepa-
rator was then compared with the previously reported capital cost for
the 720,000 gpd treatment system designed for each category. The fol-
lowing tabulation represents the percentage increase in capital costs if
a separator were required:
Percentage Increase
Subcategory
A 9
B 5
C 2
D 7
Sludge cake quantities from vacuum filtration corresponding to each
treatment system design are presented in Supplement A. The following
table summarizes the general ranges of sludge quantities generated by
plants in each subcategory:
Subcategory Cu^yd/year*
A 30 - 200
B 30 - 2,000
C ' 1,500 - 44,000
D 300
*1X net~weight basis
Particular plants within Subcategory C may be amenable to sludge
incineration because of the large quantities of sludge involved. For
example, sludge incineration would reduce the previous quantities by
about 90 percent. Sludge cake is 80 percent water, which is evaporated
during incineration, and more than half of the remaining (20 percent)
solids are thermally oxidized during incineration. Sludge incineration
costs were not evaluated for those specific cases in Subcategory C,
because the particular economics depend to a large degree on the
accessibility of a sanitary landfill and the relative associated haul
costs.
Before discussing the actual variations in costs within each cateogry,
the following discussion is presented to help visualize the complexities
involved in evaluating cost effectiveness data. Every treatment system
is composed of units whose design basis is primarily hydraulically de-
pendent, organically dependent, or a combination of the two. The fol-
lowing is a list of the unit processes employed, and a breakdown of the
design basis:
331
-------
Hydraulically Organically Hydraulically and
_ _Dej>endent_ Organically^Dependent
Pump station Thickener Aeration basin
API separator Aerobic digester Oxygen transfer eqpt.
Equalization Vacuum filter Air flotation unit
Neutrali zation
Nutrient addition
sludge recycle pump
Clarifier
The annual cost associated with the hydraulically dependent unit pro-
cesses is not a function of effluent level. On the other hand, the
sizing of the organically dependent units should theoretically vary in
direct proportion to the effluent level: e.g. reducing the BOD5 removal
from 95 to 85 percent should reduced the sizes of the sludge'handling
equipment by approximately 10 percent. However, there are two compli-
cating factors: 1) only a relatively few sizes of commercially available
equipment; and 2) broad capacity ranges. These two factors, especially
in regard to vacuum filters, tend to negate differentials in capital
cost with decreasing treatment levels.
The relationship between design varying contaminant levels and the de-
sign of aeration basins and oxygen transfer equipment is somewhat more
complex. The levels are dependent on the hydraulic flow, organic con-
centration, sludge settleability, and the relationship between mixing
and oxygen requirements. For example, to reach a particular effluent
level, the waste water's organic removal kinetics will require a
particular detention time at a given mixed-liquor concentration. The
oxygen transfer capacity of the aerators may or may not be sufficient to
keep the mixed liquor suspended solids in suspension within the aeration
basin. Therefore, the required horsepower would be increased merely to
fulfill a solids mixing requirement. Alternatively, the oxygen
requirements may be such that the manufacturer's recommended minimum
spacing and water depth requirments would require that the basin volume
be increase^ to accommodate oxygen transfer requirements.
capital an4 annual costs for new sources are presented in TaJ4e vni-3.
The treatment model used, in developing the costs is activated sludg
followed by dual media filtration. The same annual cost adjustment
factors applicable to BPCTC& are also relevant to new sources due to th
similarity of these systems. As expected, the end-of-pipe costs are not
appreciably higher than those for BPCTCA. The following information wa
extracted from Table VI1-3
332
-------
tf
en oo
d o
r-~ oo en
o CN
i-H CM CM
• • •
OOO
g
li1
!& M
o o
LPi LPv 1T>
ON O\ O\
LA LA IT\
C7\ O\ ON
in
CO
?
sol
•H
«
-P
M
8
S
00 H
-=f H
O\
o\ o\
* *
H O
vo H OJ
VD l/N,co
o
OJ
on
ON OJ t-
OJ H 00
MD OJ H
to
o
XI
3
co
O O tv.
O <7\ 00
-H O* O
tfl
EH
CQ O
O O
O O
00 0\
**
J- CJ
H O
H 00
OOO
oo o oo
ft ft ft
-=t H H
OJ H CO
H U\ t~
OOO
OOO
o t- o
ft ft ft
oo o oo
-tf H H
LPl t— O
ft ft
H 00
OOO
OOO
OJ -=1- O
ft ft ft
LTNOO H
vo t-vo
H OJ 00
o
O CO t-
O O oo
H O O
ill
4J
§
r-l
PH
88
O O
CM
OOO
ooo
OOO
•^ n ft
m M3 O\
•^ A
OJ OO
OOO
OOO
OOO
A ft A
ON co o
oo LT\ t-
ON
oo oo
I—I
000
OOO
OOO
ft ft ft
t--oo oj
VO 00 ON
CD LTN ON
M
!>^
M
i,
sr
4-1
CO
0
^ 1S-
t— OJ VO .r*>
0 t-H ^
... O
0 0 OJ N
4J
<0
a
XI
CO
Q
^ ' !».
t- oj vo r1
0 t-H ^
• • • ^
0 0 OJ gf
4-1
tO
CJ
X)
CO
CXJ <
t-VO CM •• .
O OO t-- to >i
.... h M
000 0 ?„
H-3 OC
o cu
*rt 4J
00 *r
fc 3
* •§
v co
ON O vO
H O 00
H rH o
o
*
•H
o
rH
PH
OO
rH r-(
333
If) l/i Lfi
OOO
OOO
•O
3
-p
to
o
o
•d
g
1
s
s
c
o
^H
-p
o
3
-------
Q£2i£sll_S2§£§
~$~ ~ $Zy.§a£ ~|/1QQO gal ig/lb_BOp5
"" Rgmoyal
A 1,524,000 302,900 1.15 0.83
B 2,652,000 511,000 1.94 0.28
C 8,258,000 1,710,700 6.51 0.17
D 1,992,000 361,000 1.37 0.26
The following production capacities were selected for calculation
of the $/lb BOD5 removed: Subcategory A-10 million Ib/day,
Subcategory B-5 million Ib/day, Subcategory C-1 million Ib/day and
Subcategory D-0.05 million Ib/day.
Capital and annual costs are calculated for the best available
technology economically achievable model treatment systems. These
systems are discribed as follows: two stage biological treatment plus
dual media filtration and activated carbon. Activated carbon treatment
for Subcategories A and B consists of fixed bed columns. For
Subcategories C and D pulsed bed columns with a carbon regeneration
system are recommended. Costs are presented in Table VIII-4 for the
BATEA model treatment system. The following information is extracted
from this table for a 720,000 gallon per day facility.
Annual Costs
B§ffl22§i
A 2,498,000 477,100 1.82 0.47
B 3,626,000 682,500 2.60 0.11
0. 10
0.08
2,498,000 477,100 1.82
a 3,626,000 682,500 2.60
C 10,410,000 2,110,500 8.03
D 3,529,700 1,496,100 5.69
The following production capacities were selected for calculation
of the $/lb COD removal: Subcategory A-10 million Ib/day,
Subcategory B-5 million Ib/day, Subcategory C-1 million Ib/day
and Subcategory D-0.05 million Ib/day.
334
-------
o O
in
CM i-H OV
O rH tH
O O O
0\
O rH iH
• • •
O O O
\o r~ CM
rH -* 00
O O O
o o o
o
M
•H
8
n
-P
ca
O
O
^ §
u
H
1
rl
t)
£
bO
O
O
O
H
4)
^
•ee-
o
4J
c
dn
•U
a
M
H
§
0)
N
CO
P.
a
o
•H O
H fn
O O
Ch ON
O\ O\ ON
-* -3- -=t
ON ON ON
000
ON ON ON
<3
>
cat ego r
1
vo oj
_=t CO
LTN H
o o
0 0
in H
oo t—
H -*
O O
o o
o o
H CO
VO ON
CO ,3-
f\
OJ
M
.
t— oj vo L
0 t- rH 0
o o oj a)
n)
CO
rH rH rH
-=f O ON
onvo vo
• • .
H\O LfN
H
OOO
OOO
0 t- H
*% n «%
CO t— VO
ONM3 ON
OJ CO -=f
f\
H
OOO
OOO
o on t—
CO O ON
PO-3- OJ
OJ LA LA
•\ n n
H OJ 00
OJ O O
i— vo oj
o oo ^—
ooo
m m in
ooo
• • •
ooo
l>
o >d
a v
•H -P
in id
ft
• — * «H
Lf\| t)
P> -H
O -P
*>%
1 •
•P -P
c
^ S
U H
a -P
HJ «J
ri v
h Vi
s-p_x 13
ft) Dl
H W
-8 g
o o
•H Jn
H ft
ft i
ft > w UN|
K O Q
O O
-d pq
d u
w -P H
O «3 H
•H l> ji
S -H r<
4) -P D
XJ 0 f>
O Cfl O
^
LfN
ON ON
ON ON
1 1
P
1
A
pq
n
^*
w °
fl)
y* t^^
u fr
o O
M 60
At
(11 ^
* «
rt "
fj 0
^ Xl
CO W
335
-------
SECTION IX
BEST PRACTICABLE CONTROL TECHNOLOGY CURRENTLY
AVAILABLE - EFFLUENT LIMITATIONS
Best practicable control technology currently available (BPCTCA) for the
organic chemical industry is based on the utilizations of both in
process controls and end-of-process treatment technologies.
Alternative in-process controls commensurate with PBCTCA include the
implementation of process observation and sampling to determine the
quantity, compositions, concentration, and flow of the process waste
streams. Such waste characterization studies logically lead to the
selection of various process waste sources for segregation. Exemplary
plants within the industry segregate contaminated contact process water
streams from non-contaminated streams such as cooling water. This
practice appreciably reduces the waste volume to be treated in a
centralized waste treatment plant. In addition process water streams
are segregated on the basis of the ease with which certain constituents
can be recovered as well as the ease with which the wastes can
ultimately be treated.
Process modification consistant with BPCTCA include the substitution of
nonaqueous media in which to carry out the reaction or to purify the
products. In some cases aqueous waste by-products are eliminated by
changes in the reactants, reactant purity, or catalyst system. where
waste is used in the process, its use should be restricted and the
possibility of using recycled or reused water should be investigated.
Examples of this practice include recycle between an absorber and a
steam stripper, countercurrent washing techniques, and the collection of
vacuum-jet condensate, rain water and floor water for reuse.
Equipment associated with the separation of an organic phase from an
aqueous phase, such as decanters, are provided with backup coalescers or
polishing filters for the aqueous phase. Direct vacuum-jet condensers
are replaced with indirect condensers or vacuum pumps.
in addition to waste reductions obtained through segregations and
process change, exemplary plants using BPCTCA combine recovery of
products and by-products with waste water purifications. The recovery
of chemicals from the waste waters includes both the physical separation
of chemicals from the waste water as well as subjecting the waste water
to additional chemical reactions that will render them moire aminable to
recovery and purification.
Physical separation processes utilized by exemplary plants include ad-
sorption, solvent extration, and distillation. Adsorbents in use
include activated carbon, zeolites, and synthetic resins. The adsorbed
chemicals are recovered by desorption which also serves to regenerate
the saturated adsorbent. One system for the non-destructive, inplace,
336
-------
regeneration of activated carbon is the use of pH change to cause the
adsorbed chemicals to desorb. Such a system has been used successfully
to recover phenol and acetic acid by the addition of caustic.
Solvent extraction is used for the recovery of phenol from the waste
water of the cumene process for phenol manufacture. Solvent extration
is practiced when the chemical can be extracted into a solvent already
in use in the process. Excess solvent is steam stripped from the
effluent. Effluent phenol concentration is expected to average 0.1 to
0.5 mg/liter from the treatment, system.
Distillation is used to recover by-products from reduced volume waste
water streams by steam stripping. This concentration step rpoduces an
overhead condensate containing the strippable organic substances and
water. This condensate is then reused in the process. Exemplary plants
utilizing either solvent extraction or steam distillation of waste
waters usually apply additional polishing treatment to the effluent to
removed the small remaining quantities of organic substances.
Chemical reactions such as chlorination, hydrolysis, cracking,
dechlorination and dealkylation have been used to convert impurities
into forms suitable for subsequent physical separations. A typical
example is the hydrolysis of aromatic tars with caustic with subsequent
acidification and physical separation of the organic and aqueous phases.
It is not possible to delineate a specific sequence or combination of
in-process controls which could be considered as an across the board
definition of BPCTCA. However, methods taken from those previously
described should enable all processes within each category to attain the
following standard raw waste loads. These values are listed in Table
IX-1.
End-of-pipe treatment technologies commensurate with BPCTCA are based on
the ulitization of biological systems including the activated sludge
process, extended aeration, aerated lagoons, trickling filters, and an-
aerobic and faculative lagoons. These systems include additional
treatment operations such as equalization, neutralization, primary
clarification with oil removal, and nutrient addition. Because the
removal of certain organic materials may require the utilization of high
concentrations of biological solids, effluent polishing steps such as
coagulation, sedimentation, and filteration are considered as
commensurate with BPCTCA. Effluent suspenced solids are expected to be
maintained below 30 mg/liter average concentration.
337
-------
The following waste reductions are considered consistent with BPCTCA:
Subcategory
A
Bl
B2
Cl
C2
D
Percent Reduction of
BPCTCA Raw Waste Load
Median Values
BODS COD*
90X
90X
98%
95*
99*
95X
75»
75*
15%
15%
15%
15%
* COD effluent limitations are not specified for BPCTCA
These reductions have been applied to the standard raw waste loads for
each subcategory to give a set of effluent limitations for each
subcategory. The effluent limitations for BPCTCA are listed in Table
IX-2.
It should be noted that because biological systems have been proposed as
the mode of treatment consistant with BPCTCA, the BOD5 parameter is
controlling and is the only one for which the effluent limitations are
to be applied. It may be desirable in certain cases to establish
limitations for COD or TOC instead of the BOD5 parameter. The
feasibility of such a substitution can only de determined on an
individual basis after adequate correlation has been established.
Effluent limitations are specified on the bases of the maximum for any
one day and the maximum average of daily values for any period of 30
consective days. The rationale and basis for determining the daily amd
monthly maximum variations are presented in Section XIII.
Table IX-1
Summary of Median Raw Waste Iioad Data as the
Basis for calculating Effluent Limitations
oces s
, .pw
or .b/lQOO Ib
A
Bl
B2
Cl
C2
D
500
1320
3580
2340
10,800
175,800
(60)
(158)
(429)
(280)
(1300)
(21,050)
0.12
0.35
1.77
1.90
53.0
79.0
0.31
1.1
6.2
6.5
118.0
1075.0
338
-------
Table ix-r2
BPCTCA Effluent Limitations
Subcategory. A
"BODS
TSS
Phenols
Subcateqory B
Byproduct-Proce sses
BOD5 ~
TSS
Phenols
B2 Product-Processes
BOD5
TSS
Phenol
Maximum Average of Daily
Values for Any Period
of Thirty Consecutive ^Days
"" kg/kkg Production*"
0.025
0.023
0.00025
0.06
0.06
0.00066
0.17
0. 16
0.0017
Maximum for Any
One_Day, _________
kg/kkg Production
0.045
0.038
0.0005
0.10
0.10
0.0013
0.30
0.27
0.0034
cl_Product- Process e s
BOD5
TSS
Phenols
C2 Product-Processes
~ BOD5 "
TSS"
Phenols
0.17
0.16
0.0017
0.9
0.49
0.005
0.30
0.27
0.0034
1.5
0.8
0.011
BOD5
TSS
Phenols
9.0
7.88
0.088
15.0
13.0
0.17
*kg/kkg production is equivalent to lb/1000 Ib production.
339
-------
SECTION X
Best Available Technology Economically
Achievable (BATEA)
The best available technology economically achievable is based upon the
most exemplary combination of in-process and end-of-process treatment
and control technologies.
The full range of treatment and control technologies which are
applicable to the majcr organic chemicals segment of the organic
chemicals manufacturing industry has been described in Section VII.
This level of technology is primarily based upon significant reductions
in the chemical oxygen demand (COD), as well as the biochemical oxygen
demand pollutant parameters.
End-of-process treatment has been determined to be biological plus
additional activated carbon treatment. It must be noted that this does
not preclude the use of activated carbon as an in-process treatment in
lieu of its use at the end-of-process. This may be desirable when
product can be recovered or when harmful pollutants must be removed
prior to treatment.
Two model systems are presented for cost estimation purposes:
1. Activated sludge treatment followed by filtation and
activated carbon adsorption in fixed-bed columns
(applied to Subcategories A and B)
2. Activated sludge treatment followed by carbon adsorption
in pulsed bed columns (applied to Subcategories C and D).
These systems or equivalent combinations can provide the reduction in
BOD5 and COD pollutant parameters as listed below:
Percent Removal Basj.s
subcategory BOD5 coj>
Percent Percent Percent
BPCTCA BPCTCA BADCT
RWL RWL Effluent
A 99 90 70
Bl 99 94 70
B2 99.5 94 70
Cl 99.5 94 70
C2 99.7 94 70
D 99 90 70
The applicable reductions were used as a basis for determining effluent
limitations. Low concentrations for TSS and phenols are also
attainable via BATEA treatment and control technologies. The maximum
average for any 30 consecutive day period, and daily maximum effluent
340
-------
concentrations for TSS are 15 mg/1 and 25 mg/1 respectively. For
phenols these values are 0.1 mg/1 and 0.2 mg/1 for the 30-day and daily
values respectively.
BATEA treatment and control technologies are expected to provide maximum
control of effluent variability by process controls and end-of-process
investment.
Effleunt limitations for BATEA are presented in Table X-1.
341
-------
Table X-1
BATEA Effluent Limitations
Effluent Characteristics
Subcategory A
COD ~
BOD5
TSS
Phenols
Subcateqory B
Bl Product Processes
COD
BOD5
TSS
Phenols
B2 Product^Processes
COD
BODS
TSS~
Phenols
Subcateqory C
Cl Product Processes
COD
BOD5
TSS~
Phenols
C2 Product Processes
COD
BOD5
TSS~
Phenols
Subcategoryp
COD
BOD5
TSS"
Phenols
Maximum Average of
Daily Values for Any
Period of Thirty Maximum for Any
Consecutiye^Days Qne^Day.
kg/kkg Production* kg/kkg Production*
0.02
0.002
0.004
0.000025
0.065
0.004
0.01
0.000065
0.37
0.01
0.0025
0.00017
0.39
0.01
0.005
0.00034
7.2
0.2
0.16
0.0011
65.0
0.4
1.30
0.0085
0.04
0.004
0.006
0.00005
0.13
0.008
0.017
0.00013
0.74
0.02
0.0042
0.00034
0.78
0.02
0.0083
0.00068
14.4
0.4
0.27
0.0022
130.0
0.8
2.19
0.017
* kg/kkg production is equivalent to lb/100 Ib production
342
-------
SECTION XI
New source Performance Standards
Determination of the best available demonstrated control technology
(BADCT) for new major organic sources involves the evaluation of the
most exemplary in-process control measures with exemplary end of process
treatment. Some major in-process controls which were fully desicribed
in section VII are applicable to new sources as follows:
(1) The substitution of non-contact heat exchangers
using air, water or refrigerants for direct
contact water cooling equipment (barometric condensers);
(2) The use of nonaqueous quench media, e.g. hydrocarbons
such as furnace oil, as a substitute for water,
where direct contact quench is required;
(3) The recycle of process water, such as between absorber
and stripper;
(H) The reuse of process water (after treatment) as make-up
to evaporative cooling towers through which
noncontact cooling water is circulated;
(5) The reuse of process water to produce low
pressure steam by non-contact heat exchangers in reflex
condensers or distillation columns;
(6) The recovery or spent acid of caustic solutions for reuse;
(7) The recovery and reuse of spent catalyst solutions;
(8) The use of nonaqueous solvents for extraction of products.
Although these control measures are generally applicable, no attempt was
made to identify all of these or any single one as universally
applicable.
The end of process treatment model has been determined to be biological
treatment with the additional suspended solids removal by clarification,
sedimentation, sand and/or dual medai filtration. The following system
is proposed for cost estimating purposes and does not limit the use of
equivalent systems: two stage activated sludge plus dual medium
filtration. These costs are presented in Section VIII.
Although biological treatment has been described as the basis for the
BADCT, it is recognized that chemical-physical systems such as activated
carbon may also be employed as an end-of-process technology or as an in-
process or by-product recovery system. It may also be necessary to
remove certain wastes which are toxic to or interfere with biological
waste treatment systems by in-process chemical-physical control
processes.
The reduction in major pollutant parameters as defined by BADCT is
listed by Category in the following tabulation:
343
-------
Percent Reduction of
BPCTCA Raw Waste Load
§ubcategory. Median Values
BOD5 COD
A 95 80
Bl 95 80
B2 97 80
Cl 97 80
C2 99.5 80
D 97 80
Total suspended solids and phenol effluent concentration with BADCT
technology are equivalent to those for the BATEA. Daily and any 30
consecutive day period maximum concentrations for suspended solids are
25 mg/liter and 15 mg/liter respectively. Phenol concentration on a
daily and any 30 consecutive day period maximum are at 0.2 mg/1 and 0.1
mg/liter respectively.
Effluent limitations are presented in Table XI-1 for new sources for
major organic sources.
344
-------
Table XI"1
New Sources Performance standards
(BADCT)
Effluent Character!stics
Maximum Average of
Daily Values for Any
Period of Thirty
Consecutive Days
kg/kkg Production*
Maximum for Any
One Dav
kg/kkg Production*
BOD5
COD~
TSS
Phenols
Subcategory. §
Bl_Prgduct_Prpcesses
~BOD5
COD
TSS
Phenols
11 Product Processes
BOD5
COD~
TSS
Phenols
Subcategory. C
C J rProduct ^.grocegses
BODJ~"
COD
TSS
Phenols
C2_Product Processes
BODJ ""
COD"
TSS
Phenols
Subcategory, D
BOD5
COD""
TSS
Phenols
0.012
0.10
0.0075
0.00005
0.035
0.40
0.02
0.00013
0.085
2.2
0.05
0.00034
0.085
2.3
0.05
0.00034
0.40
40.0
0.16
0.0011
0.85
390.0
2.60
0.017
0.020
0.15
0.012
0.00010
0.06
0.55
0.033
0.00026
0.15
3.0
0.083
0.00068
0.15
3.3
0.083
0.00068
0.75
60.0
0.27
0.0022
1.5
540.0
4.38
0.034
*kg/kkg production is equivalent to lb/1000 Ib production
345
-------
SECTION XII
PRETREATMENT GUIDELINES
Pollutants from specific processes within the organic chemicals industry
may interfere with, pass through, or otherwise be incompatible with a
publically owned treatment works. The following section examines the
general waste water characteristics of the industry and the pretreatment
unit operations which may be applicable.
A review of the waste water characteristics indicated that certain
products can be grouped together on the basis of pollutants requiring
pretreatment. Accordingly, the previously determined subcategories were
divided into two Sub-groups as follows:
-Subgroup j. Subgroup g
subcategory A Subcategory C
Subcategory B Subcategory D
The principal difference in the general characteristics of the process
waste waters from the manufacture of chemicals in these two Sub-Groups
is that the waste waters of subgroup 1 are more likely to include
significant amounts of free and emulsified oils, whereas the wastewaters
of subgroup 2 are more likely to include significant amounts of heavy
metals.
Detailed analyses for specific products in the industry are presented in
Supplement B.
The types and amounts of heavy metals in the waste water depend
primarily on the manufacturing process and on the amounts and types of
catalysts lost from the process. Most catalysts are expensive and,
therefore, recovered for reuse. Only unrecoverable catalysts (metals),
generally in small concentrations, appear in the waste water. The
products and processes in Subgroup 2 are most likely to have metals in
their waste water, and waste waters associated with dye/pigment
production (Subcategory D) also may have high metal concentrations due
to the presence of metallic dyes.
The manufacture of acrylonitrile (Subcategory C) produces a harmful
waste water which is difficult to treat biologically. The harmful
characteristics have been attributed to the presence of hydrogen cyanide
in excessive quantities (500 to 1,800 mg/1). In addition, the waste
water is generally acidic (pH U to 6) and contains high concentrations
of organic carbon. These waste waters are generally segregated from
other process wastes and disposed of by other means (e.g. incineration),
and they are not generally discharged to municipal collection systems.
346
-------
For these reasons, the pretreatment unit operations developed in the
following section do not include the process waste waters from the
manufacture of acrylonitrile.
Table XII-I shows the pretreatment unit operations which may be
necessary to protect joint waste water treatment processes.
Oil separation may be required when the oil content of the waste water
exceeds 10 to 15 mg/1.
The heavy metals present in organic chemical wastes are in many cases so
low in concentration that metals removal is not required from the
standpoint of treatability characteristics. However the effluent
limitations for metals and harmful pollutants may require additional
pretreatment (chemical precipitation) for removal of these materials.
The pretreatment unit operations generally consist of equalization,
neutralization, and oil separation. In addition, phenol recovery (to
reduce the phenol concentration) and spill protection for spent acids
and spent caustics may be required in some cases.
Biological _Trgatmen_t^Inhibition
The survey data collected during the sampling program were examined from
the standpoint of the occurrence of specific pollutants which may
inhibit biological treatment. This review indicated agreement with the
results of the comprehensive study of biological treatment in EPA's
£§£§!<*! Guidelines-Pretreatment of Discharges to Publicly, Owned
Treatment Works, and no changes in the lists of inhibitory pollutants
are warranted.
The following is a brief discussion of the reference material used to
determine the phenol and iron values.
347
-------
Phenol is biologically degradeable in an acclimated system. McKinney,
for example, reports that concentrations as high as 2,000 to 3,000 mg/1
of mixed phenolic substances are degradable in a properly designed
system. However, concentrations as low as 50 mg/1 can inhibit
biological treatment if the organisms are not properly acclimated.
Nemerow has reported in his literature review that concentrations of
iron on the order of 5 mg/1 can be inhibitory to anaerobic sludge
digestion.
Concentrations of iron on the order of 5 mg/1 have been reported by
Nemerow to be inhibitory to anaerobic sludge.
348
-------
SECTION XIII
ALLOWANCE FOR VARIABILITY IN TREATMENT PLANT PERFORMANCE
As previously discussed in End-of-Pipe Treatment, in section VII, the
historic treatment plant data were analyzed on the basis of monthly
averages. Subsequent effluent limitations for BPCTCA, BADCT, and BATEA
were based on both the maximum for any one day (daily maximum) and
maximum average of daily values for any period of thirty consecutive
days.
Daily historic data from two biological treatment plants treating
Subcategory C waste waters were reviewed; weekly and consecutive thirty
day averages were calculated, and then the data were analyzed
statistically. The results of these analyses are summarized in Table
XIII-1.
The significance of the data is that a biological treatment plant on the
average (50% of the time) is producing an effluent with a BODJ3
concentration of 20 mg/1, will also produce an effluent with 90 mg/1 of
BOD5 556 of the time.
Variations in the performance of a treatment plant are attributable to
one or more of the following:
1. Seasonal variations in waste water temperature
which either accelerate or depress the biological
kinetics.
2. Variations in the sampling technique or in the
analytical procedures.
3. Variations in one or more operating parameters, e.g.,
amount of sludge recycle, dissolved oxygen in the
aeration basin, etc., which can affect performance.
4. The relationship of the plant's hydraulic and organic
loading to the plant's design values. The degree
of underloading or overloading could be reflected
in performance.
5. In-plant process bottle necking which can be responsible
for degrading the effluent when seasonal loadings
strain these particular facilities. For example,
inadequate sludge handling facilities during peak
periods of sludge production may require modified
wasting of the sludges. The overall effect would
manifest itself in an increase in TSS and BOD5 in
the plant effluent.
349
-------
••
_J_
E
X
4>
p
10
CJ
O
1-
"fli
U
'5>4J
O C
— 0)
O 3
n £ o
H- 0
<4- UJ U
o
^J
86
4-> O.
.- 4-1
L. C
fl
tl
u-
IU
o
o
CO
>
-C —i
+J ~~^
C CO
0 t
5;
•-- -i
\/ *^.
"? en
> E
^
— cn
s|e
>.
!!E -4
+-> ~~^.
C cn
£ E
>.
— -i
.* "^^
D D1
E
5
>J-I
~ cn
s|E
>
JC -1
4J ^*.
c cn
a E
X
>.
•- -1
1]f
»
•^K
- cn
JS E
t~VOOJONVOKNHCV|OO
VOt— COCOONOi— IOJ-3-IA
H H H H H
IA_*CVJOCOIAIAOOO
VOt-COONONOH^1^^-
H r-l H H H
lAt— COCOOOOOOO
IAVO t— CO O rHKNlACOCVI
H H H H H 04
OOOOOOOOOO
O ONVO KNOCOCOO ^AK^
^j- UN IA VO C — t*~ CO O CVJ ^t~
H H H
OOlAiAlAOOOOO
t~-lAHCO lAJ-^tCO tT\ O
r*i r~4
coiAScr»i?-RcAi>-8o
H CVJ K^fA-^- lAVDCOOJvO
H H
CVJ J- VO CO O CM IA ON IA O
Hr-|HHCV|CvJCVJCV|fA-=*-
O fA vO t— O vO vO O O O
HHHHCvlCMCVlfA-^-lA
VO ON KN VO O IA O O O O
HHCVICXIrA^l-VOON
CO
f H
O
R
t
H
§
CVJ
8
IA
CVJ
IA
IA
£
vS
H
>-
4j g
•—
•—
n
10
JD
(1)
U
C
(U
t-
3
U
u
C
ID
JZ
^
i/1
t/>
-------
These variations are purely a function of the treatment plant design and
performance. They will still occur even if the treatment plant has
provisions for equalization of variations in the influent raw waste load
which it receives.
Selected statistical data in Table XIII-1 were examined to compare the
ratios of the 99% probability of occurrence to the 50% probability of
occurrence, the 9551 to the SOX value, and the 90* to the SOX value, as
shown below:
Ratio of
Probability
99/50
95/50
90/50
BODS
Daily.
8.0
a.5
3.0
Thirty consecutive
Weekly Day Period
3.5~ ~ ~2.7
2.5 2.0
2.0 1.7
99/50
95/50
90/50
5.3
3.4
2.5
COD
3.6
2.5
2.0
2.8
2.0
1.8
TOC
99/50
95/50
90/50
3.0
2.2
1.8
2.2
1.7
1.5
1.9
1.6
The daily 90/50 BOD ratio is 3.0, while the corresponding monthly ratio
is 1.7. This indicates that a substantial day-to-day variation
witnessed in plant performance is tempered when the variation is based
on monthly data. For this reason, it is recommended that a 30
consecutive day period average be used as the time basis for the
effluent guidelines. In addition, a 90% confidence limit should be
used, in that the 90/50 values should be within a range typically
observed in the past as being reasonable when treatment plant data were
anlyzed statistically.
The following effluent variability factors are proposed for
following pollutant parameters and time intervals:
the
351
-------
Average Thirty
Consecutive
Day Effluent weekly Effluent Daily Effluent
Adjustment Factor1 Adjustment Factor* Ad justment_Fac tor *
BODS 1.7 2.0 3.0
COD~ 1.8 2.0 2.5
190/50 ratio of confidence limits
Both of these treatment plants utilize activated sludge and were
designed based on the criteria presented in Table XIII-2. Plants A and
B have primary settling and nutrient addition. In Plant A, there are
four parallel trains of 3 aeration basins each for a total of 12 basins.
Flow from each of the parallel trains goes to a clarifier. Additional
organic and solids removal is accomplished by using an aerated polishing
lagoon.
Plant B has two parallel trains of 3 aeration basins each for a total of
6 basins. Clarification and air flotation are provided in order to
reduce the aeration basin mixed liquor (MLSS) which average about 7-
8,000 mg/1 of organics components and solids. Plant A is located in the
southern United states and not subject to extreme seasonal temperature
fluctuations. Plant B is in the Midwest and it has been found necessary
to add steam to the aeration basin during the winter to maintain the
basin temperature above H5°F,
Daily analyses of TOC and BOD were available from Plant A and only COD
data were available from Plant B. Weekly and thirty consecutive day
period averages were calculated and then the data were analyzed
statistically. The results of the analyses were summarized in Table
XIII-2.
352
-------
TABLE XI I 1-2
Summary of Plant Design Criteria
Description Plant A Pi ant B
Flow - mgd l.O 0.55
Pr imary Settling
Detention Time - days 2.5 9.1
Aeration Basin
Sludge Recycle - percent forward flow 50 100
Detention Time - hours including recycle 20 36
Aeration Equipment - Hp/M.G. ^50 5^0
Final Clarifier
OFR - gpd/sq.ft. 425 150
SWO - ft. 10 10
Diameter - ft. kQ 40
Flotation Unit-
Solids - Ibs/sq.ft./day - 7.5
Detention Time - hours - 2.5
Polymer Dosage - mg/1 - 100
Poli shing Pond
Detention Time - days 0.6 118
.Aeration Equipment - Hp/M.G. 10 1.5
353
-------
SECTION XIV
ACKNOWLEDGEMENTS
This report was prepared for the Environmental Protection Agency by the
staff of Roy F. Weston Co. under the direction of Mr. James Dougherty,
Project Director. The following individuals of the staff of Roy F.
Weston Co. made significant contributions to this effort:
Mr. David Smallwood, Project Manager
Mr. Charles Mangan, Project Engineer
Mr. Kent Patterson, Project Engineer
Mr. James weaver, Project Engineer
Dr. Sun-nan Hong, Project Engineer
The technical assistance provided by Chem Systems Inc. is also
acknowledged.
Mr. John A. Nardella, Project Officer, Effluent Guidelines Division,,
contributed to the overall supervision of this study and preparation of
the draft report.
Mr. Allen Cywin, Director, Effluent Guidelines Division, and Mr. Walter
J. Hunt, Chief, Effluent Guidelines Development Branch, offered guidance
and helpful suggestions. Members of the Working Group/steering
Committee who coordinated the internal EPA review are acknowledged:
Mr. Walter J. Hunt, Effluent Guidelines Division
Mr. John Nardella, Effluent Guidelines Division
Mr. George Rey, Office of Research and Development
Dr. Thomas Short, Ada Laboratory, Office of Research and Development
Mr. John Savage, Office of Planning and Evaluation
Mr. Alan Eckert, Office of General counsel
Mr. Wayne Smith, NFIC, Denver
Mr. John Lank, Region IV, Atlanta
Mr. Joseph Davis, Region III, Philadelphia
Mr. Ray George, Region III, Philadelphia
Mr. Albert Hayes, Office of Solid Waste Management
Mr. Frank Kover, Office of Toxic substances
Acknowledgement and appreciation is also given to the secretarial staffs
of both Effluent Guidelines Division and Roy F. Weston Co. for their
efforts in the typing of drafts, necessary revisions, and final
preparation of the effluent guidelines document. Appreciation is
especially given to the following:
Ms. Kay Starr, Effluent Guidelines Division
Ms. Chris Miller, Effluent Guidelines Division
354
-------
Ms. Brenda Holmone, Effluent Guidelines Division
Ms. Jane Mitchell, Effluent Guidelines Division
Ms. Janet Gilbert, Roy F. Weston Co.
Ms. Kit Krickenberger, Effluent Guidelines Division
Ms. Sharon Ashe, Effluent Guidelines Division
Ms. Nancy Zrubelc, Effluent Guidelines Division
Appreciation is also extended to both the Manufacturing Chemists'
Association and the Synthetic Organic Chemical Manufacturers'
Association for the valuable assistance and cooperation given to this
program. Appreciation is also extended to those companies which
participated in this study:
Allied chemical Corp.
American Cyanamid Corp.
Amoco Chemical Corp.
Atlantic Chemical Corp.
Celanese Corp.
Chemplex Corp.
Crompton-Knowles Co.
Dow Corp.
Dow Badische Corp.
E.I. duPont de Nemours Co.
Eastman Kodak Corp.
Tennessee Eastman Div.
Texas Eastman Div.
Ethyl Corp.
Gulf Oil Corp.
Kay Fries Chemical Co.
Mobil Corp.
Monochem Corp.
Sherwin-Williams Corp.
Sinclair Koppers Corp.
Southern Dyestuffs Co.
Tenneco Corp.
Phillips Petroleum Corp.
Union Carbide Corp.
355
-------
SECTION XV
BIBLIOGRAPHY
Albright, P.N. , "The Present Status of Phenol Waste Treatment." Public
Works, Vol. 98, No. 6 (June 1967), 124-127.
"Are You Drinking Biorefractories Too?" Environmental_Science and_Tech-
Vol. 7, No. 1 (January 1973), 14-15.
Bengly, M. , "The Disposal of Liquid and Solid Effluents from Oil Re-
fineries." Proceedings of 21st industrial Waste conference, Purdue Uni-
versity (May 1966) , 759-767.
Beychok, M.R., "Wastewater Treatment." Syjltocajrbon, ££ocjessing, Vol. 50,
No. 12 (December 1971), 109-112.
Black, G.M. , and schocnman, W. , "Save Water: Air condense Steam." Hydro-
carbon^ recessing, Vol. 49, No. 10 (October 1970), 101-103.
Borkowski, B., "The Catalytic Oxidation of Phenols and other Impurities
in Evaporated Effluents." Water gesearcji, vol. 1 (Pergamon Press, 1967) ,
367-385.
Browning, J.E., ed., "Activated Carbon Bids for Wastewater Treatment Jobs
Vol« 77» No- 9 (September 1970) , 32-34.
Games, A., Eller, J.M. , and Martin, J.C., "Reuse of Refinery and Petro-
chemical Wastewaters." industriai HSteI-SiaillSer iS3 > Vol-> 9/ No« 3 (June/
July 1972), 25-29.
Dorris, T.C., Patterson, D., and copeland, B.J. , "Oil Refinery Effluent
Treatment in Ponds." Journal of the Water Pollution control Federation,
Vol. 35, No. 7 (July 1963), 932-939.
Easthagen, J.H., Skrylov, V., and Purvis, A.L, , "Development of Refinery
Wastewater Control at Pascagoula, Mississippi." Jouf|jai^f_Water_£ollutic
Control Federation, Vol. 37, No. 12 (December 1965) , 1671-1678.
Eisenhauer, H. R. , "Increased Rate and Efficiency of Phenolic Waste
Ozonization." Jgugna j. o| JfotgE Pol j.utj.on Cgnt^oj. ^fe^eration , Vol. 43,
No. 2 (February 1971) , 200->202.
Elkin, H.F. , "Activated Sludge Process Applications to Refinery Effluent
Waters." sewage and Industrial Wastes, Vol. 28, No. 9 (September 1956) ,
1122-11297
356
-------
Emery, R.M., Welch, E.B., and Christman, R.F., "The Total Organic Carbon
Analyzer and Its Application to Water Research." Journal_of_Water_Pol-
lation_Cgntrol_Federation, Vol. 43, No. 9 (September 1971)7 1834-1844.
Ewinq, R.C., "Modern Waste Treatment Plant." Oil_and_Gas_Journal,
Vol. "68, No. 9 (September 1970), 66-69.
Figueroa, L.O., "Water Pollution Control at Phillips Puerto Rico Petro-
chemical Complex." Ch_emica l_En2iS§§£iB3 » Vol. 67, No. 6 (June 1960),
377-379.
Franzen, A.E., Shogan, V.G., and Grutsch, J.F., "Successful Tertiary
Treatment at American." Oil_and_Gas_Journal, Vol. 70, No. 4 (April 1972) ,
48-49.
Gilliam, A.S., and Anderegy, F.C., "Biological Disposal of Refinery
Wastes." Proceedings of 14th Industrial Waste Conference, Purdue Univer-
sity (May 1959), 145-154.
Gloyna, E.F., Brady, S.O., and Lyles, H, , "Use of Aerated Lagoons and
Ponds in Refinery and Chemical Waste Treatment." Journal_of_Water_Pol-
lution_Control_Federation, Vol. 41, No. 3 (March 1969f7~«2'S~438.
Gloyna, E.F., Ford, D.L., and Eller, J., "Water Reuse in Industry."
Journal_of_Water Pol lut ign_Contr ol_Feder ation , Vol. 42, No. 2 (February
1970) ,~237- 242.
Gazzi, L., and Pasero, R., "Selection." Hydrocarbon Processing, Vol. 49,
No. 10 (October 1970) , 83-90.
Harris, A.J., "Water Pollution Control Activities of the Central Ontario
Lakeshore Refineries." Journal qf Water Pollut ion Control Federation ,
Vol. 35, No. 9 (September 1963) ,~1 154- 1U5.
Hart, J.A. , "Air Flotation Treatment and Reuse of Refinery Waste Water."
Proceedings of 25th Industrial Waste Conference, Purdue University (May
1970) , 1-14.
Hart, J.A., "On Improving Wastewater Quality." Industrial ^Water^Engi-
neering, Vol. 7, No. 8 (September/October 1970) , 20-26?
Henshaw, T.B., "Adsorption/Filtration Plant Cuts Phenols from Effluent."
Chemical Engineering, Vol. 78, No. 5 (May 1971), 47-55.
Hovious, J.C., Conway, R.A., and Garze, C.W., "Anaerobic Lagoon Pretreat-
ment of Petrochemical Wastes." Journal_of__Water PQllution^Control^Federa-
tion. Vol. 45, No. 1 (January 1973) , 71-84?
357
-------
Jaeschke, L., and Trobisch, K. , "Treat HPI Wastes Biologically." Hydro-
carbon^ Pro cess ing, Vol. 46, No. 7 (July 1967) , 111-115.
Kent, J.A. , Industrial_cheinistry.; Reinhold Publishing Corp., New York
(1962).
Kumke, G.W., Conway, R.A., and Creagh, J.P., "Compact Activated Sludge
Treatment of Combined Petrochemical Municipal Waste." Water_and_Wastes
» Vol. 9, No. 11 (November 1972), 342-351.
Lewis, W.L. , "New Process TO Remove Phenols from Wastewater." Journal of
Wa£er_Pollutign_Control_Fede£ation, Vol. 40, No. 5 (May 1968) ,~869-8727"
Lund, H.F., ed., Industrial Pollution Control gandbgojc; McGraw-Hill, Inc.
New York (1971) .
Mapstone, G.E. , "control Cooling Tower Blowdown." gv^ocgrfeon^Processing,
Vol. 46, No. 1 (January 1967), 155-160.
McKinney, R.E., "Biological Treatment systems for Refinery Wastes."
Journal_o|_Water_Pollution_Control_£eder§tion, vol. 39, No. 3 (March
T967) , 346-359.
McPhee, W.T., and Smith, A.R., "From Refinery Wastes to Pure Water."
Proceedings of 16th Industrial Waste Conference, Purdue University (May
1961), 31V326.
"Methods for chemical Analysis of Water and Wastes. "Environmental Pro-
tection Agency, National Environmental Research Center, Analytical Qual-
ity Control Laboratory, Cincinnati, Ohio (1971) .
Morissey, A.J., and LaRocca, S.A. , "Wastewater Load Evaluated at a Multi-
Product Organic Chemical Plant." Industr^aL, Water Engineer 40.3' Vol. 7,
No. 5 (May 1970), 173-178.
Mytelka, A. I., and Manganelli, R., "Energy-Induced Changes in an Azo
Dyestuff Waste." Journal of Water gQJ.j.u-{:4.Qp Controj. Federation, Vol. 40,
No. 2 (February 1968) , 260-268.
1972-73 OPJD Chemical Buyers pirectQEv, chemical Marketing Reporter.
Schnell'publishing Co., Inc., New York (1973).
Parmelley, C.S., and Fox, R.D. , "Reuse comes out Ahead." Water and
Hastes_Eagin.ee£ing, Vol. 9, No. 11 (November 1972).
Paulson, E.G., "The War on Pollution," Oj,l_and Gas Journal, Vol. 68,
No. 6 (June 1970), 85-92.
358
-------
Perry, J.H., Chemical^Engineer's^Handbook; McGraw-Hill, Inc., New York
(1963).
Polss, P., "What Additives Do for Gasoline." Hydrocarbon_Processing,
Vol. 52, No. 2 (February 1973) , 61-68.
Prather, V., "Advanced Treatment of Petroleum Refinery Wastewater by
Autoxidation." Journal of Water Pollution_Control_Federation, Vol. 42,
No. 4 (April 1970), 596-603.
Pursell, W.L., and Miller, R.B., "Waste Treatment of Shelly Oil Company's
ElDorado, Kansas Refinery." Proceedings of 16th Industrial Waste Con-
ference, Purdue University (May 1961), 292-303.
Rambow, C.A., "Industrial Wastewater Reclamation." Proceedings of 23rd
Industrial Waste Conference, Purdue University (May 1968), 1-9.
Rey, G., Lacy, W.I., and Cywin, A., "Industrial Water Reuse: Future
Pollution Solution." Environinental_Science^and^Technglogyg Vol. 5, No. 9
(September 1971), 842-845.
Rose, B.A., "Water Conservation Reduces Load." Industrial^ivater^Engi-
neering. Vol. 6, No. 9 (September 1969), 4-8.
Rose, W.L., Gorringe, G.E., "Activated Sludge Plant Handles Loading Vari-
ations." Oil and Gas Journal, Vol. 70, No. 10 (October 1972), 62-65.
Ross, W.K., and Sheppard, A.A., "Biological Oxidation of Petroleum
Phenolic Waste Waters." Proceedings of 10th Industrial Waste Conference,
Purdue University (May 1955), 106-119.
Santoleri, J.J., "Chlorinated Hydrocarbon Waste." Chemical Engineering
Process, Vol. 69, No. 1 (January 1973), 68-74.
Sawyer, G.N., "Fertilization of Lakes by Agriculture and Urban Drainage."
Journal of, New^EncLland Water Works Association, (1949) .
"Sequential Gasification" Qil_and_Gas_Journal, Vol. 70, No. 10 (October
1972), 116-117.
Shreve, R.N., Chemical Process Industries, third edition; McGraw-Hill,
Inc., New York" (1956) .
Smallwood, D.S., Ramanathan, M., and Dougherty, J.H., "Reference Effluent
Guidelines for Organic Chemical Industries, Process Narratives." Roy F.
Weston, Inc. report to Environmental Protection Agency, Washington, D.C.,
Contract No. 14-12-963, (Unpublished).
359
-------
Snoeyink, V.L. , Weber, W.J., and Mark, H.B., "Sorption of Phenol and
Nitrophenol by Active Carbon." Enyironmental_Science and Technology,
Vol. 3, No. 10 (October 1969), 918-926.
"Sohio Uses Sewage-Plant Effluent as Stop Feedwater." Oil_and__Gas
Journal/ Vol. 70, No. 2 (February 1972), 80-82.
"Standard Methods for the Examination of Water and Wastewater. " American
Public Health Association, Inc., New York (1971).
Steck, W. , "The Treatment of Refinery Wastewater with Particular Con-
sideration of Phenolic streams." Proceedings of 21st Industrial Waste
Conference, Purdue University (May 1966) , 783-790.
Stroud, P.W. , Sorg, L.V., Lamkin, J.C., "The First Large Industrial
Waste Treatment Plant on the Missouri River." Proceedings of 18th In-
dustrial Waste Conference, Purdue University (May 1963) , 460-475.
Taras, M.J. , ed., standard Metfcod.8 for the Bxaminat^on^of ^Water_and
Wastewater . American Public Health Association, WashingtonT D?c.,~(1971)
Thompson, C. S. , stock, J., and Mehta, D.L. , "cost and Operating Factors
for Treatment of Oily Waste Water." oil and gas Journal, Vol. 70, NO. 11
(November 1972), 53-56.
Thompson, S.J. , "Techniques for Reducing Refinery Waste Water." oil_and
Gas_Journal, Vol. 68, No. 10 (October 1970), 93-98.
Vania, G.B., Bhatla, M.N., Thompson, A. F., and Bralston, C.W., "Process
Development, Design, and Full-scale Operational Experience at a Petro-
chemical Manufacturing Wastewater Treatment Plant." Proceedings of 44th
Conference of the Water Pollution Control conference (October 1971) ,
1-25.
Water \___ Atmospheric Analysi s , Part 23, "Standard Method of Test for Bio-
chemical"5xygen Demand of Industrial Water and Industrial Waste Water."
1970 Annual Book of A STM Standards, American Society for Testing and
Materials, Philadelphia, Pennsylvania (1970) .
"Water Quality Criteria." state Water Quality Control Board, Publication
No. 3-A, Sacramento, California (1963).
Wigren, A. A., and Burton, F.L, "Refinery Wastewater Control." Journal of
Water_Poilution control Federation. Vol. 44, No. 1 (Janueiry 1972) 117-
"128."
360
-------
SECTION XVI
GLOSSARY
The terms defined here relate to common chemical conversions utilized
extensively in the organic chemicals industry.
Acylation Subcategory^A
The acylation reaction introduces an acyl group, RCO-, into an aromatic
ring. The product is an aryl ketone. The arylating reagents commonly
used are acid halides, ROCOCl, or anhydrides, (RCO)2O. The catlyst is
aluminum chloride. The reaction is usually carried out in an organic
solvent, commonly carbon disulfide or nitrobenzene.
Acylation is utilized in the manufacture of dye intermediates such as
acetanilide, and acetyl-p-toluidine. The reaction for acetanilide is
shown below:
AICU
C6H5NH2 + (CH3CO)20 _+ C6lfrNHCOCH3 + CH,COOH
Catalyst •*
Aniline Acetic AcetanlUde Acetic Acid
Anhydride
Although the reaction itself is nonaqueous (Subcategory A) , water may be
used in the subsequent separation of the reaction products. When
carried out batchwise the reaction may fall within the context of an
overall Subcategory D system.
Alcohglysis (Transesterification) Subqategory c
Alcoholysis is the cleavage of an ester by an alcohol. It is also
called transesterification. The reaction is usually catalyzed by
aqueous sulfuric acid. A generalized equation for the reactions is
shown below:
H2SO,
RCOOR1 + R"OH T+* RCOOR" + R'OH
Transesterification is an equilibrium reaction. To shift the
equilibrium to the right it is necessary to use a large excess of the
alcohol whose ester is desired, or else remove one of the products from
the reaction mixture. The second approach is used in most industrial
applications, since in this way the reaction can be driven to
completion.
361
-------
An excellent example of the application of transesterif ication is found
in the synthesis of the polymer, polyvinyl alcohol.
H-SO,
-CH-CH- + CH,OH _*M CHoCOOCH, + -CH0CH-
2, 3 332,
0 Met Hanoi Methyl OH
I Acetate
C » 0 Polyvinyl
I Alcohol
CH3
Polyvlnyl
Acetate
Although there are hundreds of acetate groups in every modecule of poly-
vinyl acetate, each of them undergoes the reactions typical of any
ester. In the presence of aqueous sulfuric acid, polyvinyl acetate and
methyl alcohol can exist in equilibrium with methyl acetate and
polyvinyl alcohol. The reaction mixture is heated so that the lowest
boiling compound, methyl acetate, distills out and the reaction proceeds
to completion.
Ammonoly.sis Subcateggry C
The reaction is classified within Subcategory C as it is conducted with
an aqueous catalyst system.
Alkylation Subcategories^A and^B
Alkylation refers to the addition of an aliphatic group to another
molecule. The media in which this reaction is accomplished can be vapor
or liquid phase, as well as aqueous or non-aqueous.
Benzene is alkylated in the vapor phase over a solid catalyst (silica-
alumina impregnated with phosphoric acid) with propylene to produce
cumene .
Benzene Propylene Cumene
This reaction is nonaqueous and is considered within Subcategory A.
Tetraethyl lead (the principal antiknock compound for gasolines) is alsc
a very important alkylated product. It is prepared by the action of
ethyl chloride on a lead-sodium alloy.
1| PbUa + A C2H5Cl — +• PbCCiHj),, + 3 Pb + *» NaCl
Alloy Ethyl Tetra Lead Sodium
Chloride Ethyl Chloride
Lead
362
-------
The reaction is carried out in an autoclave equipped with a heating
jacket, a stirrer to agitate the lead alloy, and a reflux condenser.
The mixture is heated at the start and then cooled. After 6 hours, the
excess ethyl chloride is distilled off, and the tetraethyl lead is steam
stripped from the reaction mixture. This type of staged batch reaction
with direct contact steam is considered typical of Subcategory D.
The alkylation reaction is also utilized in the manufacture of dyestuffs
and intermediates. Dimethylaniline is employed intensively in the
manufacture of triarylmethane dyes. It is prepared according to the
following reaction:
2 + 2 CH3OH ^-* C6H5N(CH3)2 + 2 H20
Aniline Methanol D Imethylanil Jne Water
Aniline, with an excess of methanol and aqueous sulfuric acid, is heated
in an autoclave. The dealkylated product is discharged through a
cooling coil, neutralized, and vacuum distilled. This is again typical
of the chemical conversions with Subcategory D.
Amination_bY_Reduction §ubcategories_B_and_p
Amination by reduction involves the formation of an amino group (-NH2)
through the reduction of a nitro group (-NO2) . The reaction can be
carried out batchwise in an aqueous liquid phase (Subcategory D) or
continuously in the vapor phase (Subcategory B) .
The reducing agents in the batch conversion are iron and an aqueous acid
catalyst (such as hydrochloric acid) . Aniline is produced by the
reaction as follows:
HC1
1» C6H5N02 + 9 Fe + A H20 — •*• 4 CgHjNHj + 3 FejOj,
Nitrobenzene Iron Water Anfltne Iron Oxide
This batch reaction for reducing nitrobenzene with iron to aniline is
being replaced by the continuous vapor phase reduction shown below:
3 H2 — *• C6H5NH2 + 2 H2°
Nitrobenzene Hydrogen Aniline Water
363
-------
The reaction is conducted with a very short contact time in a tube
packed with copper on SiO2 as the catalyst. The hydrogen is adsorbed to
the catalyst surface. Molecules of nitrobenzene are next adsorbed on
the hydrogenated surface. The reaction products, aniline and water
vapor, then desorb from the catalyst. This type of vapor phase reaction
is typical of Subcategory B.
Ammonolysis Subcategory C
Amination by ammonolysis relates to those reactions in which an amino
compound is formed using aqueous ammonia. Industrial applications in-
clude the production of ethanolamines and methylamines.
A mixture of mono-, di-, and triethanolamine is obtained when ethylene
oxide is bubbled through aqueous ammonia as shown by the following
equation:
JHOCH2CH2NH2 Honoethanolamlne
n(C2Hi,0) + MH3 — »• < (HOCH2CH2)2NH Dlethanolamf ne
^(HOCH2CH2)3N Trlethanolamfne
Methylamines are formed similarly by the ammonolysis of methanol. These
continuous reactions are also considered within Subcategory C.
Aromatization is the conversion of saturated cyclic compounds to
aromatic compounds. The reaction is illustrated by the following
equation:
Heat and
C6HnCH3 _* C6H5CH.. + 3 H2
Catalyst J
Methylcyclohexane Toluene Hydrogen
The reaction is carried out in the vapor phase with or without
catalysts. It is nonaqueous and considered within Subcategory A.
Condensation SubcategoryD
condensation reactions involve the closure of structural rings in
aromatic compounds. They are carried out batchwise in aqueous acid
solutions and are of great importance in the manufacture of dye
intermediates.
2§£y.3ration Subcategories B and^C
364
-------
Ethers are commonly produced by the dehydration of alcohols. When
carried out in the liquid phase using sulfuric acid as a catalyst, the
reaction is considered within Subcategory C. However, it can also be
accomplished in the vapor phase over solid alumina catalysts within
Subcategory B.
The following reaction for the production of ethyl ether from ethanol
can be accomplished by either route:
2 C,H,.OH — * (C2H5)20 + H20
Ethanol Ethyl Ether Water
Ester if i cat ion Subcategory__C
Esterif ication generally involves the combination of an alcohol and an
organic acid to produce an ester and water. The reaction is carried out
in the liquid phase with aqueous sulfuric acid as the catalyst. The use
of sulfuric acid has in the past caused this type of reaction to be
called sulfation. The equation for producing ethyl acetate from acetic
acid and ethanol is shown below:
CH3CH2OH + CH3COOH — * C
Ethanol Acetic Acid Ethyl Acetate Water
Continuous esterif ication reactions are considered within Subcategory c.
Friedel-Crafts^Reactions Subcategory _A
Friedel-Crafts reactions involve the alkylation or acylation of an
aromatic ring in the presence of such catalysts as AICI3, BF3, SnCI<£,
12. These addition reactions are sensitive to trace quantities of
moisture and must be carried out under anhydrous conditions.
Halogenation and Hydrohalogenation Subcategory A
These reactions refer to the addition of a halogen (CI2, Br2, 12, F2) to
an organic molecule. The various products are obtained through ""both
liquid and vapor phase reactions with or without catalysts. Aliphatic
compounds such as methane and ethane can both be chlorinated in the
vapor phase with the cocurrent production of HCI gas.
CH3CH3 + CI2 — +• CH3CH2CI + HCI
Ethane Chlorine Ethyl Hydrogen
Chloride Chloride
365
-------
The by-product HCI can also be reacted with ethylene to form ethyl
chloride by hydrohalogenation. This later reaction is carried out over
an anhydrous aluminum chloride catalyst.
The addition of halogens to unsaturates (alkenes) serves to give many
other derivatives such as ethylene dichloride, ethylene dibromide,
dichloroethylene, trichloroethylene, and tetrachloroethane. The
preparation of ethylene dichloride is typical:
C2H3Br2
C2H4 + CI2 *J+
Ethylene Chlorine Ethylene Dichloride
The chlorine gas is bubbled through a tank of liquid ethylene dibromide
(catalyst) , and the mixed vapors are sent to a chlorinating tower where
they meet a stream of ethylene. The products from the tower pass
through a partial condenser, followed by a separator, with the crude
ethylene dichloride passing off as a gas and the liquid ethylene
dibromide being returned to the systems.
These reactions are all non-aqueous and are within Subcategory A.
However, it should be noted that some of these reactions may also be
carried out batchwise in dye manufacture and as such may fall within the
context of a Subcategory D system.
HydroformyRation fOXO Process) Subcategory C
The oxo process is a method of converting olefins to aldehydes
containing one additional carbon atom. The olefin is reacted in the
liquid phase with a mixture of hydrogen and carbon monoxide in the
presence of a soluble cobalt catalyst to produce the aldehyde. A
typical reaction follows, in which propylene is converted to n-
butyraldehyde:
HCo(CO)j,
C3H6 * CO + H2 ^ C/,H80
Propylene Carbon Hydrogen n-Butyraldehvde
Monoxide
The reaction itself is nonaqueous. However, the regeneration of the
cobalt carbonyl catalyst complex requires extensive usage of aqueous
solutions of sodium carbonate and sulfuric acid. This aqueous catalyst
regeneration causes the hydroformylation reaction to be classified in
Subcategory C.
Hydrogenation and Dehydroqenation Subcategory B
366
-------
The hydrogenation reaction involves the addition, while dehydrogenation
involves the removal of hydrogen from an organic molecule. Both types
of reaction are carried out in the vapor phase, at elevated
temperatures, over solid catalysts such as platinum, palladium, nickel,
copper, or iron oxides. Steam is added in many cases as a diluent to
reduce the partial pressure of hydrocarbons in the reactor and prevent
the formation of coke on the catalyst. These reactions are considered
within subcategory B.
Typical hydrogenation products include methanol produced from carbon
monoxide and hydrogen as well as other alcohols produced from aldehydes.
Dehydrogenation products include ketones, such as acetone, produced from
alcohols, such as isopropanol.
Hydration_4Hydroylsis ) Subcat egg r ie s_B_a nd_C
These reactions can be either liquid or vapor phase. Liquid phase
systems include the production of ethanol from ethylene with aqueous
sulfuric acid or isopropanol from propylene. The corresponsing vapor
phase routes are carried out over solid H3P04 catalysts. The equation
shown for ethanol can be done either way:
Ethylene Water Ethanol
Ethylene glycol and ethylene oxide can also be produced by either a
liquid or vapor phase route. The liquid reaction involves the formation
of ethylene chlorohydrin, which is produced by the reaction of aqueous
chlorine with ethylene,
CH2CH2 + CI2 + H20—*CH2OH-CH2CI + HCI
Ethylene Chlorine Water Ethylene Hydrogen
ChlorohydrIn Chloride
The ethylene chlorohydrin is treated with aqueous sodium bicarbonate
solution to produce ehtylene glycol.
CH2OH-CH2CI + NaHC03 —*• CH2OH-CH2OH + NaCI + C02
Ethylene Sodium Ethylene Sodium Carbon
Chlorohydrin Bicarbonate Glycol Chloride Dioxide
More recently the chlorohydrin route to ethylene oxide and glycol has
been replaced by the reaction of ethylene with oxygen and water:
C2H4 + 1/2 02 -» C2H<,0
Ethylene Oxygen Ethylene Oxide Ethylene Water Ethylene Glycol
Oxide
367
-------
Ethylene and oxygen are charged to a tubular reactor which is filled
with silver catalyst (vapor phase) or sulfuric acid (liquid phase).
Ethylene oxide is recovered from the gaseous reactor effluent by
absorption in water. The wet ethylene oxide is then reacted with water
in the presence of sulfuric acid to produce ethylene glycol.
Depending on whether these reactions are aqueous liquid phase or vapor
phase they may be considered in either Subcategory B or C.
Subcategory C
The treatment of reactor effluents with either caustic or acid is a
necessary part of many reaction systems. The neutralizing reagents
normally used are sulfuric acid or sodium hydroxide. Gaseous effluents
are normally treated in an absorber while liquid effluents are treated
in a liquidliquid contactor. Both types of treatment are considered
within Subcategory C.
Nitration §ubcatec[ories_C_and_D
This reaction involves the introduction of nitrogen onto a hydrocarbon
by the use of nitric acid. It is usually carried out in the liquid
phase and may be either continuous or batch. Nitrobenzene is produced
as a dye intermediate by the direct nitration of benzene, using a
mixture of nitric and sulfuric acids according to the following
equation:
C6H6 * HNO + HjSO^ — * C6H5N02 + HjSO,, * HjO
Benzene Nitric Sulfuric Nitrobenzene Sulfuric Weter
Acid Acid Acid
This type of reaction is considered either in Subcategory C or D.
Oxidation Subcategories B and C
This family of reactions may be carried out either in aqueous solutions
or in the vapor phase. The oxidant may be either air or oxygen.
The liquid phase systems all utilize dissolved mineral salts such as
cobalt acetate. A typical reaction is the oxidation of acetaldehyde to
give acetic acid in an aqueous mixture of cobalt acetate and acetic
acid.
CH3CHO + 1/2 02 — * CH3COOH
Acetaldehyde Oxygen Acetic Acid
368
-------
Alternatively, acetadehyde can be produced by the vapor phase oxidation
of ethanol over a silver gauze catalyst.
C2H5OH + 1/2 02 —+• HjCHO + H20
Ethanol Oxygen Acetaldehyde Water
Depending on whether the reaction is vapor or liquid phase it may be
considered within Subcategory B or C.
Pyrolyjsis (Cracking^ Subcategory^ B
These reactions involve the breaking of carbon chains in alkanes with
the subsequent formation of alkanes and alkenes of lower molecular
weight. The equation below illustrates the cracking reaction by which
ethylene is produced:
CH3CH2CH3 —* CH2CH2 + CH^.
Propylene Ethylene Methane
The reactions are all carried out in the vapor phase at very high
temperature. steam is usually added as a diluent to prevent the
formation of coke. For this reason, the reactions are considered within
Subcategory B.
369
-------