DRAFT
DEVELOPMENT DOCUMENT FOR
EFFLUENT LIMITATIONS GUIDELINES
AND STANDARDS OF PERFORMANCE
ORGANIC CHEMICALS INDUSTRY
PHASE II
PREPARED BY ROY F.WESTON, INC.
FOR UNITED STATES
ENVIRONMENTAL PROTECTION AGENCY
UNDER CONTRACT NUMBER 68-01-1509
DATED: February, 1974
-------
DRAFT
NOTICE
The attached document is a DRAFT CONTRACTOR'S REPORT. It includes
technical information and recommendations submitted by the Contractor
to the United States Environmental Protection Agency ("EPA") regarding
the subject industry. It is being distributed for review and comment
only. The report is not an official EPA publication and it has not
been reviewed by the Agency.
The report, including the recommendations, will be undergoing extensive
review by EPA, Federal and State agencies, public interest organizations
and other interested groups and persons during the coming weeks. The
report and in particular the contractor's recommended effluent limitations
guidelines and standards of performance are subject to change in any and
all respects.
The regulations to be published by EPA under Sections 3Qk(b) and 306 of
the Federal Water Pollution Control Act, as amended, will be based to a
large extent on the report and the comments received on it. However,
pursuant to Sections 304(b) and 306 of the Act, EPA will also consider
additional pertinent technical and economic information which is developed
in the course of review of this report by the public and within EPA. EPA
is currently performing an economic impact analysis regarding the subject
industry, which will be taken into account as part of the review of the
report. Upon completion of the review process, and prior to final promul-
gation of regulations, an EPA report will be issued setting forth EPA's
conclusions concerning the subject industry, effluent limitations guide-
lines and standards of performance applicable to such industry. Judgments
necessary to promulgation of regulations under Sections 3Qk(b) and 306 of
the Act, of course, remain the responsibility of EPA. Subject to these
limitations, EPA is making this draft contractor's report available in
order to encourage the widest possible participation of interested persons
in the decision making process at the earliest possible time.
The report shall have standing in any EPA proceeding or court proceeding
only to the extent that it represents the views of the Contractor who
studied the subject industry and prepared the information and recommenda-
tions. It cannot be cited, referenced, or represented in any respect in
any such proceedings as a statement of EPA's views regarding the subject
i ndust ry.
U.S. Environmental Protection Agency
Office of Air and Water Programs
Effluent Guidelines Division
Washington, D.C. 20k6Q
-------
DRAFT
ABSTRACT
A__s_t_udy of the seconds ry__products . &£gmenjL...of the -QX^3Djj^cj]^m|caj s_ man_u -
factjJ_rio^L-i!ldjJStry_._was_ conducted by Roy F_t Wes.LQ.tlj ___ LctC-. for the United
States Environmental Protection Agency. The purpose of this study was
JXD_j2SjLab-LLs.b__a£ flue nt 1 i m i t at i_o n s g uj^dej __Lne_s_fp^r_ex^[ s_tj_ n £_poj n t^-source_
discharges and standards of performance and pretreatment standards for
new sources. This study and its proposed regulations were undertaken
in fulfillment of Sections 30A, 30jS, aruJ _301_P f the Federal Pollution
Control Act Amendments of 1972^
For the purposes of this study, 5^ product/process segments of the in-
dustry were investigated. S^jfjFjj: lent data on__p_r o ce s s r aw__was_t e j_oa d_
we re ob it aj ned __ _f ^q r. Jt k_ of __ t h,es, e. , j^_ tjna L.,^f.fJjieo t . JJjjLUajtJonj. ,suldejjjie,s_,
c^ould"j,ubseq,uently ,be deiiejjapfid, , This study was the second part of a
two phase effort. In the first pjT§J^x_PJ!O£gs_s.^ravij!iaste . lp_ads__anjd__e_f^-
fluent limitations guj defines were.,est^blJ5Jieji_jFor_Jj^^rodjj£^pro^e_^^
groups, total coverage of the industry has now been extended to include
"85 groups.
i n both 5tudjes_, the prodjjct/pj^£cejj^ groups _ we re jgut in to__fpu£ _majo£,
subcateqo r ijes , based on process technology as related to contact wastes
usage. These industry segments were further subcategor i zed on the
b5S_Ls_-Qf__the..q,uaQiLt Les_p_f _2oll_ujtjrr^s_, ^measured as t_h e._firxicfi5 s.. raw__was_te
JkxgcL The first phase of this study focused on major organic products
and used six subcategories . This second phase focused on secondary
products and used thirteen subcategories.
Sega r_a_tg. ^ifluent. . J jmLtaJ: ions ..aujjdejj nes_have been developed for each
subcategory on__the basis of treatment aj1 d_jcont£p 1_ technologies. Sup-
portive data and the rationale for development of the proposed effluent
limitations guidelines and standards of performance are contained in
this report.
NOTICE: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
i i i
-------
DRAFT
CONTENTS
Sect ion Page
ABSTRACT i j \
FIGURES vi
TABLES ix
I CONCLUSIONS 1
I I RECOMMENDATIONS 5
III INTRODUCTION 11
IV INDUSTRY CATEGORIZATION 21
Discussion of the Rationale of Categorization 21
Description of Subcategories 22
Basis for Assignment of Subcategories 23
Process Descriptions 2k
V WASTE CHARACTERIZATION 253
VI SELECTION OF POLLUTANT PARAMETERS 261
Pollutants of Significance 261
BOD 263
COD 264
TOC 264
TSS 265
Pollutants of Limited Significance 274
Oil 274
Nitrogen 275
Phenol 275
Total Dissolved Solids, Chlorides, Sulfates 276
Cyanide 276
Heavy Metals . 277
VII CONTROL AND TREATMENT TECHNOLOGIES 279
In-Plant Pollution Abatement 280
End-of-Pipe Treatment 281
Single-Stage Biological Treatment 281
Multiple-Stage Biological Treatment 284
Filtration 284
Carbon Adsorption 286
i v
-------
DRAFT
CONTENTS
(cont i nued)
Sect ion Page
BPCTCA Treatment Systems 289
BATEA Treatment Systems 29^
BADCT Treatment Systems 29^
VII I COST, ENERGY AND NON-WATER QUALITY ASPECTS 299
Cost 299
Energy 301
Non-Water Quality Aspects 302
IX BEST PRACTICABLE CONTROL TECHNOLOGY CURRENTLY
AVAILABLE (BPCTCA) 335
X BEST AVAILABLE TECHNOLOGY ECONOMICALLY ACHIEVABLE
(BATEA) 357
XI NEW SOURCE PERFORMANCE STANDARDS 361
XII PRETREATMENT GUIDELINES 365
XIII ALLOWANCE FOR VARIABILITY IN TREATMENT PLANT
PERFORMANCE 369
Biological Wastewater Treatment 369
Activated Carbon Wastewater Treatment 369
XIV ACKNOWLEDGMENTS 375
XV BIBLIOGRAPHY 377
-------
DRAFT
FIGURES
Figure No. T i11e Page No.
Flow Diagrams
4-1 Cumene 27
4-2 p-xylene 31
4-3 Adiponitrile 40
4-4 Benzofc Acid and Benzaldehyde 44
4-5 Chlorinated Methanes 48
4-6 Chlorobenzene 50
4-7 Chlorobenzene (incl. dichlorobenzene) 51
4-8 Chlorotoluene 54
4-9 Diphenylamine 58
4-10 Hexamethylenediamine (Adiponitrile) 62
4-11 Hexamethylenediamine (Hexanediol) 66
4-12 Maleic Anhydride 70
4-13 Methyl Chloride 74
4-14 Methyl Ethyl Ketone 77
4-15 Perchloroethylene 81
4-16 Phthalic Anhydride (o-Xylene) 85
4-17 Phthalic Anhydride (Naphthalene) 88
4-18 Tricresyl Phosphate 92
4-19 Acetic Esters 102
4-20 Acrylonitrile 105
4-21 p-Aminophenol 109
4-22 Calcium Stearate 112
v?
-------
DRAFT
FIGURES (continued)
Figure No. Title Page No.
4-23 Caprolactam (UBE-lnventa) 120
4-24 Caprolactam (DSM) 123
4-25 Cyclohexanone Oxime 128
4-26 Cresol 132
4-27 Formic Acid 135
4-28 Hexamethylene Tetramlne (Plant 1) 139
4-29 Hexamethylene Tetramine (Plant 2) 140
4-30 Hydrazine 143
4-31 Isobutylene 147
4-32 Isopropanal 151
4-33 Oxalic Acid 156
4-34 Pentaerythritol 160
4-35 Propylene Glycol 164
4-36 Propylene Oxide 168
4-37 Saccharin 171
4-38 Secondary Butyl Alcohol 175
4-39 Citric Acid 179
4-40 Citronellol and Geraniol 181
4-41 Fatty Acids and Primary Derivatives 209
4-42 lonone and Methylionone 212
4-43 Methyl Salicylate 215
4-44 Monosodium Glutamate 226
4-45 Naphthenic Acid 229
VI I
-------
DRAFT
FIGURES (continued)
Figure No. Title Page No.
4-46 o-Nitroaniline 232
4-47 p-Nitroaniline 235
4-48 Pentachlorophenol 238
4-49 Pigments 241
4-50 Plasticizers 243
4-51 Tannic Acid 247
4-52 Vanillin 251
7-1 BCPTCA Waste Treatment Model 295
7-2 BATEA Waste Treatment Model 296
vi i i
-------
DRAFT
TABLES
Tablj No. Title Pa^e No.
2-1 Subcategories of the Organic Chemicals
Manufacturing Industry 6
2-2 Effluent Limitations for the Best
Practicable Control Technology
Currently Available (BPCTCA) 8
2-3 Effluent Limitations for the Best
Available Technology Economically
Achievable (BATEA) 9
2-4 Standards of Performance for New
Organic Chemicals Manufacturing
Sources 10
3-1 Chemicals Listed under SIC Code
2815 12
3-2 Chemicals Listed under SIC Code
2818 14
4-1 Raw Waste Load Based on UBE-lnventa
Caprolactam Process 119
4-2 Process Raw Waste Load Based on DSM
Process 126
4-3 Process Raw Waste Load Based on DSM
Process 127
4-4 Historical RWL Data for Organic
Solvent Complex 153
4-5 Usage Classification of Dyes 188
4-6 Chemical Classification of Dyes 189
4-7 U.S. Production of Dyes by Classes
of Application, 1965 190
4-8 U.S. Production and Sales of Dyes
by Chemical Classification, 1964 191
i x
-------
DRAFT
TABLES
(conti nued)
Table No. Title Page No.
4-9 Some Examples of Commercial Fatty
Acids Showing Typical Percentage
of Constituent Acids 194
4-10 Chemical Conversions and Unit Opera-
tions Contributing to Process Raw
Waste Loads for the Manufacture of
Fatty Acids and Primary Derivatives 206
4-11 Process RWL Associated with Manu-
facture of Fatty Acids 207
4-12 Process RWL Associated with Manu-
facture of Primary Derivatives from
Fatty Acids 208
4-13 RWL Data for Batch Chemical Complex 219
4-14 Treatment Plant Performance Data for
Batch Chemical Complex Treatment
Plants Utilizing the Activated Sludge
Process 220
4-15 Effluent Discharged from Batch Chem-
ical Complex after Biological Treatment 221
5-1 Major Subcategory A - BPCTCA Process
Raw Waste Loads 254
5-2 Major Subcategory B - BPCTCA Process
Raw Waste Loads 254
5-3 Major Subcategory C - BPCTCA Process
Raw Waste Loads 255
5-4 Major Subcategory D - BPCTCA Process
Raw Waste Loads 256
5-5 Subcategory A-2 -- BPCTCA Process
Raw Waste Loads 258
5-6 Subcategories B-3, B-4, B-5 --
BPCTCA Process Raw Waste Loads 258
-------
DRAFT
Table No.
5-7
5-8
6-1
6-2
6-3
6-4
6-5
7-1
7-2
7-3
7-4
7-5
7-6
7-7
7-8
TABLES
(cont i nued)
Title
Subcategories C-3, C-4, C-5,
C-6, C-7 -- BPCTCA Process Raw-
Waste Loads
Subcategories D-1, D-2, D-3, D-4 --
BPCTCA Process Raw Waste Loads
List of Pollutants and Indication of
Pollution Examined for the Organic
Chemicals Industry
Miscellaneous RWL for Category A
Miscellaneous RWL for Category B
Miscellaneous RWL for Category C
Miscellaneous RWL for Category D
Organic Chemicals Study Treatment
Technology Survey
Historic Treatment Plant Performance:
50 Percent Probability of Occurrence
Treatment Plant Survey Data
Removal by Filtration
Activated Carbon Plants Treating Raw
Wastewaters
Summary COD Carbon Isotherm Data
Summary BOD Carbon Isotherm Data
Summary TOC Carbon Isotherm Data
Page No.
259
260
262
266
267
270
272
282
283
285
287
288
290
291
292
XI
-------
DRAFT
Table No.
8-1
8-2
8-3
8-k
8-5
8-6
8-7
8-8
8-9
8-10
8-11
8-12
8-13
8--\k
8-15
8-16
8-17
8-18
8-19
8-20
TABLES
(cont i nued)
WASTEWATER TREATMENT COSTS
BADCT, AND BATEA EFFLUENT
Pol lutant
BTX
Para-Xylene
Acetone
Vinyl Chloride
Sty rene
Phthal ic Anhydride
Chloromethanes
Methyl Chloride
Ad i poni t ri 1 e
Hexamethy 1 enedi ami ne
Benzoic Acid
Methyl Ethyl Ketone
Mai ei c Anhydri de
1 sopropanol
Formic Acid
Propylene Oxide
Secondary Butyl Alcohol
Hydrazi ne
Calcium Stearate
Caprolactam
FOR BPCTCA,
LIMITATIONS
Category
A-1
A-2
B-1
B-2
B-2
B-3
B-3
B-k
B-U
B-4
B-4
B-5
B-5
C-3
C-k
C-5
C-5
C-5
C-5
C-5
Page No.
303
30k
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
XI I
-------
DRAFT
Table No.
8-21
8-22
8-23
8-24
8-25
8-26
8-27
8-28
8-29
8-30
8-31
9-1
9-2
9-3
9-4
9-5
9-6
TABLES
(cont i nued)
WASTEWATER TREATMENT COSTS FOR BPCTCA,
BADCT, AND BATEA EFFLUENT LIMITATIONS
(cont i nued)
Pollutant Category
Isobutylene C-5
Hexamethy 1 ene Tetramine C-6
Aery Ion i t ri 1 e C-6
Pentaerythri tol C-7
Fatty Acid D-3
Citronellol and Geraniol D-3
Dye D-4
Sodium Glutamate D-4
Tannic Acid D-4
Citric Acid D-4
Naphthenic Acid D-4
Major Process Subcategory A (Non-Aqueous Pro-
cesses) RWL for Each Product/Process Grouping
Considered as BPCTCA
Major Process Subcategory B (Vapor-Phase Pro-
cesses) RWL for Each Product/Process Grouping
Considered as BPCTCA
Major Process Subcategory C (Aqueous Reaction
Systems) RWL for Each Product/Process Grouping
Considered as BPCTCA
Major Process Subcategory D (Batch Processes)
RWL for Each Product/Process Grouping Considered
as BPCTCA
Range of RWL Variation Within Major Process
Subcategories
Division of Major Process Subcategories
Page No.
323
32k
325
326
327
328
329
330
331
332
333
340
3^1
343
345
347
349
XI I I
-------
DRAFT
TABLES
(cont i nued)
Table No. Title Page No.
9-7 Division of Major Process Subcategories 350
9-8 Division of Major Process Subcategories 351
9-9 Division of Major Process Subcategories 352
9-10 Summary of Mean RWL for Each Secondary
Organic Product Subcategory 353
9-11 Effluent Limitations Guidelines for the
Secondary Organic Segment of the Organic
Chemicals Point Source Category Commen-
surate with Best Practicable Control Tech-
nology Currently Available (BPCTCA). 356
10-1 Effluent Limitations Guidelines for the
Secondary Products of the Organic Chemicals
Point Source Category Commensurate with Best
Available. Technology Economically Achievable
(BATEA). 359
11-1 Effluent Limitations Guidelines for the Secon-
dary Organic Products Segment of the Organic
Chemicals Point Source Category Commensurate
with Best Available Demonstrated Control
Technology. 362
12-1 Pretreatment Unit Operations for the Organic
Chemicals Industry 3&7
^
13-1 Summary of Plant Design Criteria 370
13-2 Effluent Variation of Biological Treatment
Plant Effluent 372
x i v
-------
DRAFT
SECTION 1
CONCLUSIONS
The complex task of establishing effluent guidelines limitations for the
Organic Chemicals Industryrequired that the industry be divided into a
two-phase study. A draft of the Phase 1 development document was issued
in June 1973- The document recommended the use of process-oriented sub-
categories which were developed as follows:
Subcategory A: Nonaqueous Processes
Contact between water and reactants or products is minimal. Water
is not required as reactant or diluent, and is not formed as a
reaction product. TJie_j&p-l-y_water usage stems J_rom_P ejjjod?c washe_s_
or catalyst hydration.
Subcategory B: Processes with Process Water Contact only as Steam
Diluent or Absorbent
Process water is in the form of dilution steam, direct product quench,
or absorbent for effluent gases. Reactions are all vapor-phase over
solid ca_LaJjysts_. Most processes have an absorber, coupled with steam
stripping of chemicals for purification and recycle.
Subcategory C: Aqueous Liquid-Phase Reaction Systems
ReactLQlis^aj^^JJLaui^-Rjl&iej with, the ca_tajj£sjt__i n__aji. .aguequs me_dj_um.
Continuous regeneration of the cafaTyst requires extensive water
usage, and substantial removal of spent inorganic by-products may be
required. AdAULJ.2D_a 1 , Pro_ce§? water, is involved in final puri f i cati,pju
or neutralization of products.
Subcategory D: Batch and Semicontinuous Processes
Many reactants are liquid-phase, with aqueous catalyst systems. Re-
quirements for very rapid process cooling necessitate provisions for
the direct addition of contact quench water or ice. Reactants and
products are transferred from one piece of equipment to another by
gravity flow, pumping, or pressurization. Much of the materials
handling is manual, and there is only limited use of automatic process
control. Filter presses and centrifuges are commonly used for solid-
liquid separations, and air or vacuum ovens are used for drying.
Cleaning of non-continuous prndjjction^eqj^pment^r^sfjnjtpq a major
source of wastewater. ~ ~-
NOTICE: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-------
DRAFT
The recommended effluent limitations were computed on a sliding scale
employing a flow/pollutant relationship. During the EPA review procedure,
it was decided that the flow parameter was too cumbersome for use in
large chemical complexes. The Contractor's approach was replaced with a
procedure utilizing 6 subcategories, namely A, B1, B2, C1, C2 and D.
The BPCTCA-RWL data for each subcategory was based on median RWL calcula-
tions. The previously recommended effluent guidelines were recomputed
on the basis of the new subcategories, and the new limitations were pub-
lished recently in the Federal Register. The supporting development document
was published by EPA in December 1973-
The Contractor has been directed by EPA Effluent Guidelines Division to
handle the Phase 2 data independently of the Phase 1 data; data
handling should complement the previous EPA development document of
December, 1973-
With this background, the data from kb product/processes were grouped
into 13 subcategories, e.g. B3, B4, B5, etc. The product mix within each
major subcategory is based on order-of-magnitude differences between each
subcategory's mean RWL.
Individual effluent limitations were recommended for all 13 subcategories
for BOD and COD. TSS effluent concentrations were recommended for each
technology level, i.e. BPCTCA, BADCT, and BATEA. In order that exces-
sively stringent effluent limitations are not implemented, minimal pol-
lutant concentrations are recommended as being indicative of a technology
level as follows:
BPCTCA 20 mg/1 BOD
BATEA 10 mg/1 BOD
50 mg/1 COD
Other RWL parameters were considered during the study, and specific
products/pollutants which might be inhibitory or incompatible with BPCTCA
treatment technology were cited in Section VI,
End-of-process treatment for the 1977 standard is defined as biological
treatment as typified by current exemplary processes: activated sludge,
trickling filters, aerated lagoons, and anaerobic lagoons. These systems
may require pH control and equalization in order to control variable waste
loads, and also may require clarification with the addition of chemicals
to aid in removing suspended solids. These systems do not preclude the
use of equivalent chemical-physical systems such as activated carbon in
situations where necessary land area is not available. Additionally,
suitable in-process controls are also applicable for the control of those
pollutants which may be inhibitory to the biological waste treatment
system.
NOTICE: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-2-
-------
DRAFT
econ om i ca 1 1 y ach i e vajbj^_ JJATJL
, of actvated
_
This technology is based upon substantial reductions of dissolved or-
ganics which are b ioref ractory as well as those which are biodegradable.
Exemplary in-process systems are also applicable to this technology.
End-of-process activated carbon treatment does not preclude the use of
such treatment as an in-process technology.
End-of-process technology for new sources utilizing the best available
demonstrated control technology (BADCT) is defined as biological treat-
ment with suspended solids removal via clarification, sedimentation, sand,
or dual-media filtration. In addition, exemplary in-process controls
are also assumed to be applicable, particularly where biologically inhibi-
tory pollutants must be controlled. This technology does not preclude
the use of equivalent chemical -physi cal systems such as activated carbon
as either an in-process or end-of-process treatment. This may be advan-
tageous in areas where land availability is limited.
In conclusion, effluent limitations were derived on the basis of the
maximum for any one day and the maximum average of daily values for any
period of thirty consecutive days. The factors used in deriving these
time-based limitations were determined from long-term performance (i.e.
daily, weekly, monthly) from the best treatment systems evaluated. Time-
based limitations consider the normal variations of exemplary designed
and operated waste treatment systems.
NOTICE: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-3-
-------
DRAFT
SECTION I I
RECOMMENDATIONS
^ with the_
technology .curi:ently,,ava i lable_jju;e .presented for each industrial
category of the organic chemicals manufacturing "industry. Major product/
process segments of the industry which are applicable to these limita-
tions are listed in Table 2-1, and effluent limitations for the 1977
standard (BPCTCA) are presented in Table 2-2. It should be noted that
process wastewaters subject to these limitations include all contact
process water but do not include non-contact sources such as boiler and
cooling water blowdown, laboratories, and other similar sources.
Implicit in BPCTCA RWL data is the segregation of non-contact wastewaters
from process wastewaters and the maximum utilization of applicable in-
plant pollution abatement technology in order to minimize capital expen-
ditures for end-of-pipe wastewater treatment facilities.
End-of-process technology for BPCTCA involves the application of biologi-
cal treatment as typ i f i ed j3y_ajcjJ\^a_£££L-S-l udge^_,tjj_ck]Jn g f l.Ltejrj.j_^aj5£a_tgd
lagoons^ or ^jajiaerobic J^ajjopjTs_. Equalization with pH control and oil
separation may be~Teq"ui red in order to provide optimal as well as a uni-
form level of treatment. Chemical flocculation aids, when necessary,
should be added to the clarification system in order to control suspended
sol ids levels .
Effluent limitations to be attained by the application of the best
available technology economically achievable are presented in Table 2-3
for the major product/process segments listed in Table 2-1. End-of-
process treatment for BATEA includes the addition of activated carbon
systems to biological waste treatment processes. Exemplary in-process
controls, as discussed in the later sections of this document, are also
applicable to this technology. It is emphasized that the model treatment
system does not preclude the use of activated carbon within the plant.
Such systems are frequently employed for recovery of products, by-products,
and catalysts.
The best available demonstrated control technology for new sources includes
the most exemplary process controls, as previously enumerated, with bio-
logical waste treatment and systems for removal of suspended solids.
Effluent limitations for the major product/process segments are presented
in Table 2-k.
NOJ_I_CE_: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-5-
-------
DRAFT
Table 2-1
Subcategories of the Organic Chemicals Manufacturing
Industry (Phase Il-Major Product-Processes)
Subcateqory A
Cumene
p-Xylene
Subcategory B
B-3 Products
Chloromethanes
Chlorotoluene
Di phenylami ne
Pe rch1o roet hy1ene
Phthalic Anhydride
Tricresyl Phosphate
-4 Products
Adiponi trile
Benzoic acid and Benzaldehyde
HMDA
Methyl Chloride
B-5 Products
HMDA
Maleic anhydride
Methyl Ethyl Ketone
Subcategory C
C-3 Products
Cyclohexane Oxime
Isopropanol
C-k Products
Formic acid
Oxalic acid
Process Description
Alkylation of Benzene by Propylene
Isomerization, Crystallization, and
Filtration of Mixed Xylenes
Chlorination of methyl chloride and
Methane mixture
Chlorination of Toluene
Deamination of aniline
Chlorination of chlorinated hydrocarbons
Oxidation of naphthalene
Condensation of Cresol and Phosphorus
Oxychloride
Chlorination of Butadine
Catalytic oxidation of Toluene with Air
Hydrogenation of Adiponitrile
Esterification of Methanol with Hydro-
chloric Acid
Ammonolysis of 1,6-Hexanediol
Oxidation of Benzene
Dehydrogenation of Sec. Butyl Alcohol
Hydroxylamine Process
Hydrolysis of propylene
Hydrolysis of Formamide
Nitric acid oxidation of carbohydrates
NOTICE: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-6-
-------
Table 2-1
(cont i nued)
DRAFT
C-5 Products
Calcium Stearate
Caprolactam
Hydrazine Solutions
Isobutylene
Propylene Oxide
Sec. Butyl Alcohol
C-6 Products
Acrylon i trile
Hexamethylene Tetramine
C-7 Products
p-Aminophenol
Cresol, Synthetic
Pentaerythri tol
Saccha rin
Subcategory D
D-1 Products
o-Ni troani1i ne
p-Ni troani1i ne
D-2 Products
Citronellol and Geraniol
Fatty acids
Fatty acid derivatives
lonone and Methyl lonone
D-3 Products
PIast i ci zers
D-k Products
Citric acid
Dyes and Intermediates
Naphthenic acid
Pigments
Monosodium Glutamate
Tannic acid
Neutralization of Stearic acid
DSM Caprolactam Process
Raschig Process and Formaldehyde
Extraction from a mixture of C^ Hydro
carbons
Chlorohydrin Process
Sulfonatfon and Hydrolysis of Mixed
Butylenes
Ammoxidation of Propylene
Synthesis with Ammonia
Catalytic reduction of Nitrobenzene
Methylation of Phenol
Aldehyde Condensation
Synthesis from Phthalic Anhydride
Deri vat ives
Ammonolysis of o-Nitrochlorobenzene
Ammonolysis of p-Nitrochlorobenzene
Citronella Oil Distillation
Hydrolysis of Natural Fats
Esterification, Amination, etc.
Condensation and Cyclization of Citral
Condensation of Phthaljc Anhydride
Fermentation of Molasses
Batch Manufacture
Extraction and acidification of Caustic
sludge from Petroleum refinery
Diazotization and coupling of amine,
sulfuric, etc.
Fermentation of Beet Sugar and Molasses
Extraction of Natural Vegetable Matter
NOTICE: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-7-
-------
DRAFT
Effluent Characteristics
Subcateqory A2
~BODZ
COD
Subcategory B
b3 Jroducts
BOD"2"
COD
64 Products
COD
B5 Products
COD
Subcategory C
6~:r
COD
C4 Products
Bo'52*
COD
_C5 Products
BOD"2
COD
_C6 Products
BOD7 ~
COD
C7 Products
COO
Subcategory D
01 Products
BOD2"
COD
D2 Products_
BOD2
COD
D3 Products
BOO2
COD
Table 2-2
Effluent Limitations for the Best Practicable
Control Technology Currently Available (BPCTCA)
Organic Chemical Manufacturing Industry
Major Product-Processes By Subcategory
Effluent Limitations - kq/kkq Product ion
Maximum Average of Daily
Values for Any Period of
Average Dai ly Thi rty Consecutive Day_s_
0.00089*2 O".0017"t
0.00388 0.00775
D4 Products
B~OD2
COD
0.04i7
0.927
1.33
20.2
9.6
97.9
1 .44
1.18
4.43
1.79
29.7
5.74
63.2
35.6
3o5.
3.03
23.5
2.30
21.0
4.30
25.6
19.8
403.
0.0934
1 .9'+
2.66
40.14
19.2
135.8
0.159
2.88
2.36
8.85
3.58
50.4
11.5
126.
75.2
730.
6. 16
57.0
4.60
42.0
8.60
51.2
39.6
806.
Maxi mj.T \/a 1 ue
For Any One
Day
o fooTo i
0.0132
0.210
08.7
43.2
23L
0.35S
4.90
5.31
15.1
101.
25.S
215.
165.
1 ,240.
13.9
96.9
10.4
71.4
19.4
87.0
89.1
1,370.
Average Monthly Effluent Limitations Guidelines for TSS = ^5 nig'!
Ikg/kkg production is equivalent to lb/1000 Ib production.
Controlling design Para-neter.
NOTICE: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-------
Effluent Characteristics
Table 2-3
Effluent Limitations for the Best Available
Technology Economically Achievable (BATEA)
Organic Chemical Manufacturing Industry
Major Product-Processes By Subcategory
Effluent Limitations - kg/kkg Production
Subcateqory A2
BOD
COD
Subcategory B
63 Products
BOD
COD
B4 Products
BOD
COD
B5 Products
BOD
COD
Subcategory C
C3 Products
BOD
COD
C4 Products
BOD
COD
C5 Products
BOD
COD
C6 Products
BOD
COD
C7 Products
BOD
COD
Subcacegory D
01 Products
BOD
COD
D2 Products
BOD
COD
D3 Products
BOD
COD
D4 Products
BOD
COD
Average Dai ly
0.000443
0.0012
0.0234
0.287
0.133
6.26
0.960
21 .1
0.00796
0.446
1.18
4.43
0.179
9.21
0.574
19.6
3.66
113.
1.54
8.84
0.230
6.51
0.430
7.94
1.98
125.
Maximum Average of Daily
Values for Any Period of
Thirty Consecutive Days
0.000886
0.0024
0.0468
0.574
0.266
12.5
1.92
42.2
0.0159
0.892
2.36
8.86
0.358
18.4
1.15
39.2
7.32
226.
3.08
17.7
0.460
13.0
0.860
15.9
3.96
250.
Maximum Value
For Any One
Day
0.00199
0.00408
0.105
0.976
0.599
21.3
4.32
71.7
0.0358
1.52
5.31
15.1
0.806
31.3
2.58
66.6
16.5
384.
6.93
30.1
1 .04
22.1
1.94
27.0
5.91
425.
Average Monthly Effluent Limitations Guidelines for TSS = 15 mg/1
'kg/kkg production is equivalent to lb/1000 Ib production.
NOTI_CE_: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-9-
-------
DRAFT
Table 2-4
Standards of Performance for New Organic Chemicals Manufacturing Sources
Major Product-Processes By Subcategor/
Effluent Characteristics
Effluent limitations - kq/kkg_Production1
Subcateqory A2
BOD
COD
Subcategory B
B3 Products
BOD
COD
B4 Products
BOD
COD
B5 Products
BOD
COD
Subcategory C
C3 Products
BOD
COD
C4 Products
BOD
COD
C5 Products
BOD
COD
C6 Products
BOD
COD
C7 Products
BOD
COD
Subcategory D
D1 Products
BOD
COD
D2 Products
BOD
COD
D3 Products
BOD
COD
Ok Products
BOD
COD
Average Monthly Effluent L
kg/kkg production is equi
NCTICE
Average Dai ly
0.000443
0.00310
0.0234
0.742
1.11
16.2
7.97
54.3
0.066
1.15
1.18
4.43
1.49
23.8
4.76
50.6
30.4
292.
1.54
22.8
1.91
16.8
3.57
20.5
16.4
322.
imitations Guidelines
valent to lb/1000 Ib p
: THESE ARE TENTATIVE
Maximum Average of Daily
Values for Any Period of
Thirty Consecutive Days
0.000886
0.0062
0.0468
1.48
2.22
32.4
15.9
108.
0.132
2.30
2.3b
8.86
2.98
47.6
9.52
101.
60.8
584.
3. 08
45.6
3.62
33.6
7.14
41.
32.8
644.
for TSS --- 15 mg/1
roduct ion.
RECOMMENDATIONS BASED UPON INFORMATION
Maximum Value
For Any One
Day
0.00199
0.014
0. 105
2.52
5.0
55.1
35.9
185.
0.297
3.91
5.31
15.1
6.71
80.9
21 .4
172.
137.
993.
6.93
77.5
8.60
57.1
16.1
69.7
73.8
1 ,095.
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-10-
-------
DRAFT
SECTION I I I
INTRODUCTION
The Organic Chemicals Industry was originally defined to include only
those commodities listed under SIC 2815 (Cyclic Crudes and Intermediates)
and SIC 2818 (industrial Organic Chemicals Not Elsewhere Classified),
and those manufacturers included in industry group 281. Although these
boundaries may seem practical for purposes of dividing U.S. private in-
dustry into manageable point-source categories for the development of
effluent limitations guidelines, serious problems arise when one attempts
to "force fit" an industry as diverse as this one into such constraints
based on SIC code.
Tables 3-1 and 3-2 show 260 materials listed under SIC 2815 and 2818.
The lists are somewhat ambiguous in that many specific commodities (e.g.
ethylene, adiponitrile, hydrazine, synthetic vanillin, sodium sulfoxalate
formaldehyde, etc.) are in the same list with references to very large
families of products (e.g. synthetic organic dyes, coal tar distillates,
organic pigments, alcohols, flavors, enzymes, etc.).
It must be understood that these lists are developed by the United States
Department of Commerce and are oriented toward the collection of economic
data related to gross production, sales, and unit costs. They are not
related to the true nature of this industry in terms of actual plant
operations, production, or considerations associated with water pollution
control. As such, they do not provide a realistic or definitive set of
boundaries for this study. It should also be noted that all the major
producers of organic chemicals are not included in the 281 group. Major
companies not in group 281 are covered in such diverse classifications
as petroleum refining, meat and dairy products, and photographic and op-
tical equipment.
Some appreciation of the number of distinct commodities which are manu-
factured and subsequently offered for sale can be gained by examining
the OPD CHEMICAL BUYERS DIRECTORY, which lists over 5,000 organic chemi-
cals.
It must also be understood that the exact nature of the manufacturing
operations at any specific facility is unique and characteristic only of
that facility. There are very few, if any, organic chemicals plants
which manufacture one product by a single process. Instead, almost all
plants are multi-product/process facilities where the final mix of pro-
ducts shipped from each plant is unique. In some cases, the actual number
of commodities produced can be in the thousands (such as a batch chemi-
cals complex), while other facilities manufacture only two or three high-
volume products.
-11-
-------
DRAFT
Table 3-1
Chemicals Listed Under SIC Code 2815
Cyclic Intermediates, Dyes, Organic Pigments (Lakes and
Toners), and Cyclic (Coal Tar) Crudes
Acid dyes, synthetic
Acids, coal tar: derived from
coal tar distillation
Alkylated diphenylamines, mixed
Alkylated phenol, mixed
Ami noanthraquinone
Ami noazobenzene
Ami noazotoluene
Aminophenol
Ani1ine
Ani1ine oil
Anth racene
Anthraquinone dyes
Azine dyes
Azobenzene
Azo dyes
Azoic dyes
Benzaldehyde
Benzene, product of coal tar
di st i1lation
Benzoic acid
Benzol, product of coal tar
di sti1lation
Biological stains
Chemical indicators
Chips and flakes, naphthalene
Chlorobenzene
Chloronaphthalene
Chlorophenol
Chlorotoluene
Coal tar acids, derived from
coal tar distillation
Coal tar crudes, derived from
coal tar di sti1lat ion
Coal tar distillates
Coal tar intermediates
Color lakes and toners
Color pigments, organic: ex-
cept animal black and bone
black
Colors, dry: lakes, toners, or
full strength organic colors
Colors, extended (color lakes)
Cosmetic dyes, synthetic
Cresols, product of coal tar
disti 1 1 at ion
Creosote oil, product of coal tar
di sti 1 1 at ion
Cresylic acid, product of coal tar
disti 1 lation
Cyclic crudes, coal tar: product
of coal tar distillation
Cyclic intermediates
Cyclohexane
Diphenyl amine
Drug dyes, synthetic
Dyes, synthetic organic
Eos ine toners
Ethyl benzene
Food dyes and colors, synthetic
Hydroqui none
I socyanates
Lake red C toners
Lithol rubine lakes and toners
Male? c anhydri de
Methyl violet toners
Naphtha, solvent: product of
coal tar distillation
Naphthalene, product of coal tar
di st i 1 lat ion
Naphthol , alpha and beta
Naphthol sul fon ic acids
Nit roan i 1 ine
Ni t robenzene
Nitro dyes
Nit rophenol
Nitroso dyes
Oils: light, medium,
product of coal tar
Orthodi chlo robenzene
Paint pigments, organic
Peacock blue lake
Pen tachlo rophenol
Persian orange lake
Phenol
Phloxine toners
Phosphomolybdic acid lakes and
toners
and heavy
distillation
-12-
-------
DRAFT
Table 3-1
(continued)
Phosphotungstic acid lakes and
toners
Phthalic anhydride
Phthalocyanine toners
Pigment scarlet lake
Pigments, organic: except
animal black and bone black
Pitch, product of coal tar
d i sti11 at ion
Pulp colors, organic
Quinoline dyes
Resorcinol
Scarlet 2 R lake
StiIbene dyes
Styrene
Styrene monomer
Tar, product of coal tar dis-
tillation
Toluene, product of coal tar
d i st i1lat ion
Toluol, product of coal tar
di sti1lation
Tolui di nes
Toners (reduced or full strength
organic colors)
Vat dyes, synthetic
Xylene, product of coal tar
di sti1lation
Xylol, product of coal tar
di st i Hat ion
-13-
-------
DRAFT
Table j-/
Chemicals Listed Under SIC Code 2818
Industrial Organic Chemicals, Not Elsewnere ..
a s s
i e d
Accelerators, rubber processing:
cyclic and acycli c
AcetaIdehyde
Acetates, except natural acetate
of clime
Acetic acid, synthetic
Acetic anhydride
Acet in
Acetone, synthetic
Acids, organic
Acrolei n
Aery Ion i trile
Adi pi c acid
Ad i pon i trile
Alcohol, aromat ic
Alcohol, fatty: powdered
Alcohols, industrial: de-
natured (nonbeverage)
Algin products
Amines of polyhydric alcohols,
and of fatty and other acids
Amyl acetate .and alcohol
Antioxidants, rubber processing:
cyclic and acyclic
Bromochloromethane
Butadiene, from alcohol
Butyl acetate,, alcohol, and
prop i onate
Butyl ester solution of 2, k-D
Calcium oxalate
Camphor, synthetic
Carbon bisulfide (disulfide)
Carbon tetrachloride
Casing fluids, for curing fruits,
spices, tobacco, etc.
Cellulose acetate, unplasticized
Chemical warfare gases
Chloral
Chlorinated solvents
Chloroacetic acid and metallic
sa 1 ts
Chloroform
Chlorop ter i n
Citral
C I trot.-5
Citric ac i d
C i t rone 1lol
r.OU!Tlcj r i P
C ream of 1j r M r
Cyc I opi-opane
DDT, ',,'chnicdl
Decahyd ronaph !:ha 1 ene
D i ch 1 oroJ i f i ourc'iethane
Dietryicyclodexane (mixed isomers)
Diethviene g ycol ether
D inif:thy 1 di vinyl acetylene (di-
isoprojenv'i acetylene)
D i i,iet hy 1 ^ ydr;iz i n .>, unsymmet r i ca 1
En/>iiie =
Esters of phtbalic anhydride: and
: f phcsphoric, adipic, lauric,
c ' e : j_ , s f=4 --, c i c . and s t e a r i c
cu; i c <
Lster-_ o! poiv''V>Jric alcohols
Ft nan ,] i cd>. .: i ,'a 1
Lt.it' -,-
Lthyl .icet-itu, synthetic
fi'iyl fll.olio!, industrial (non-
t r-Vi'i'a ;.
tth>I bu L>; ite
fcthy! cellulose, unplasticized
E t hyI chloride
Ethyl ether
Ethyl formate
Ethyl n i tri te
Ethyl pprhydrophenanthrene
Ethyleno
Ethylrne glyco!
Ethy I ene glycol etf'er
Ethyifc-ne glycol, inhibited
r t h y I e n e o x i d t:
Fe rr i c H mmoni urn oxalate
i" i a * i r s a n d f ' a jo'. I n g materials,
byntheti c
F!uormated hydrocarbon gases
-or niti 1 defiyde ,for'ral in)
Formic acid and pictallic salts
F'.ron
-------
DRAFT
Table 3-2
(conti nued)
Fuel propellants, solid organic
Fuels, high energy, organic
Gases, fluorinated hydrocarbon
Geraniol, synthetic
Glycerin, except from fats
(synthetic)
Grain alcohol, industrial
Hexamethylenediami ne
Hexamethylenetetramine
High purity grade chemicals,
organic: refined from
technical grades
Hydraulic fluids, synthetic base
Hydraz ine
Industrial organic cyclic compounds
lonone
I sop ropy 1 alcohol
Ketone, methyl ethyl
Ketone, methyl isobutyl
Laboratory chemicals, organic
Laurie acid esters
Lime citrate
Malononitrile, technical grade
Metallic salts of acyclic organic
chemi cals
Metal 1i c stearate
Methanol, synthetic (methyl alco-
hol)
Methyl chloride
Methyl perhydrof1uorine
Methyl salicylate
Methyl ami ne
Methylene chloride
Monoch1orodi f1uoromethane
Monornethy 1 pa ram! nophenol sul fate
Monosodium giutamcte
Muscard gas
Nitrous ether
Normal hexyl decalin
Nuclear fuels, organic
Oleic acid esters
Organic acids, except cyclic
Organic chemicals, acyclic
Oxalates
Oxalic acid and metallic salts
Pentaerythri tol
Perchloroethylene
Perfume materials, synthetic
Phosgene
Phthalates
Plasticizers, organic: cyclic
and acycli c
Polyhydric alcohols
Potassium bitartrate
Propellents for missiles, solid,
organ i c
Propy 1ene
Propylene glycol
Q.ui nucl id inol ester of benzyl ic
aci d
Reagent grade chemicals, organic:
refined from technical grades
Rocket engine fuel, organic
Rubber processing chemicals, or-
ganic: accelerators and anti-
oxidants--cyclic and acyclic
Saccha ri n
Sebacic acid
Si 1icones
Soaps, naphthenic acid
Sodium acetate
Sodi urn algi nate
Sodium benzoate
Sodium glutamate
Sodium pentachlorophenate
Sodium sulfoxalate formaldehyde
Sol vents , organ i c
Sorb i tol
Stearic acid esters
Stearic acid salts
Sulfonated naphthalene
Tackifiers, organic
Tannic acid
Tanning agents, synthetic organic
Tartaric acid and metallic salts
Tart rates
Tear gas
Terpi neol
Tert-butylated bis (p-phenoxy-
phenyl) ether fluid
-15-
-------
DRAFT
Table 3-2
(conti nued)
Tetrachloroethylene
Tetraethyl lead
Thioglycolic acid, for permanent
wave lotions
Trichloroethylene
Trichloroethylene stabilized,
degreasing
Trichlorophenoxyacetic acid
Trichlorotri f1uoroethane tetrachloro-
difluoroethane isopropyl alcohol
Tricresyl phosphate
Tridecyl alcohol
Trimethyltrithiophosphite (rocket
propellants)
Triphenyl phosphate
Urea
Vani11 in, Synthet ic
Vinyl acetate
-16-
-------
DRAFT
Furthermore, the production quantities associated with the product mix
shipped from a plant are not necessarily a true indication of the extent
or type of manufacturing activities carried out within that plant. In
many cases, several products are produced captively within the plant and
subsequently utilized as feedstocks in the production of those products
ultimately shipped from the plant.
These factors are worthy of consideration in that the water usage and
subsequent water pollution caused by the Organic Chemicals industry are
directly related to the specific nature of its diverse manufacturing pro-
cesses. Analysis of these manufacturing processes is, therefore, the
logical starting point for any study whose objective is the development
of production-based effluent limitations guidelines.
In order to develop such production-based effluent limitations guidelines,
it was necessary to utilize some "common denominator" which would relate
diverse production activities (waste generating activities) with water
pollution control technologies (waste treatment activities). The process
raw waste load (RWL) was considered as the best tool for accomplishing
this objective.
For purposes of this study, the_._p_rocess RWL J_s__def i ned as the qjjant j.ty .of.,
wa^p__amj^ pnl Lufan f_s.._gjpng_rgi-pH by jj_jriarHjfacturJ nc] p,r.Q£e,ss , d i vjji§dMby_the
quantity of ^chemical product derived f,rom the pxQ.ces^. In this context,
the process represents a unique set of chemical conversions and unit
operations by which a specific feedstock is transformed into a specific
set of products, co-products, and by-products. Thj?__aua,at i tj_e_a cf.w^ter
and pollutants are measured prior to any treatment for removal of pollu-
tants. These quantities include all water which contacts chemicals within
the process battery limits and excludes non-contact water associated with
heating and cooling surface heat exchangers. This differentiation was
drawn on the basis that oxygen-demanding hydrocarbon parameters (for
which effluent limitations guidelines are subsequently developed) are
associated primarily with such contact wastes. It is appreciated that
surface run-off, tank drainage, and other sources outside the battery
limits may also contribute to this type of pollution, but very little data
are available to indicate the significance of these sources compared to
actual process wastes.
A much more detailed discussion of the process RWL, contact and non-contact
water usage, and the interactions of feedstocks, products, and associated
chemical conversions and unit operations within the manufacturing plant
context is given in the Development Document for Phase 1 of this study
(EPA4W1 -73/009).
During Phase 1, process RWL's were established for 40 product/process
groupings considered to be the Major Organic Products in terms of the
magnitude of their gross production. It was anticipated that data on
many more product/process groups could be obtained by utilizing historic
-17-
-------
DRAFT
monitoring data provided by manufacturers. However, industry could not
provide useful data of this type because most previous sampling efforts
considered only the unique combination of total plant wastes for each
facility and could not be allocated on an individual product basis.
Subsequently, the Phase 1 RWL data were obtained by field sampling in-
dividual process facilities for periods of 3 to 5 days. The sampling
periods were so limited because of the necessity to collect data on as
wide a variety of product/processes as possible.
During Phase 2, 5^ additional product/process groupings were sampled,
with the data from kk considered as adequate for establishment of mean-
ingful process RWL's. Thjj_^ejcojid_acc up J^s__d £f _LQecLas__ Jiejccmd axy.
__ __ ___ . __
e _commod i t j es .JjejTjLraJJbLJTBrmf ac- tuxed J n smajjjer
~.~- _ .
vol Ujpes^hajx__t_hjoje surveyed in the Phase 1 study^
In order to complement the EPA development document, 13 additional sub-
categories were established for the A4 groups for which data were ob-
tained during the Phase 2 study. Effluent limitations were then calcu-
lated for each of these sub-categories.
The development of these waste reduction Factors is another item requiring
preliminary comment. It was originally anticipated that these factors
(relating to demonstrated treatment technologies commensurate with BPCTCA,
BATEA, and BADCT) could be obtained from performance data on many operat-
ing treatment plants in the Organic Chemicals Industry.
It was known that this industry did not have numerous wastewater treat-
ment facilities; however, the original assumptions proved to be overly
optimistic. Relatively few organic chemicals manufacturers provide
substantial treatment of their wastewaters.
Information from the plant survey visits in both phases of this study,
from previous industry surveys, and from Refuse Act permit applications
was used to define the extent of treatment facilities in the organic
chemicals industry. Ajaji^oxjjnaJ^! 1 _y 8p_£ercent of the industry's 600 .pro-
du,c±lon__f_a_c i 1 i t i es p^rpvMje_np_qn -s|te t reatmervt of R'erTfiafiY neut ra 1 i zat Lon
of their wastewatejis_. (it should be understood that many of these pres-
ently dTscfiarge to municipal treatment systems.) Of the remai nder, ap-
proximate l_y 10 percent provide miscellaneous physical treatment such as
sedimentation, while approximately 10 percent pro._vTde biological treat-
ment of some tyjpe.
Although not widely practiced, bj^3 1 jaal c_aj treatment as def ined^.b,y._tJie-
actlvated s Ijudge ^p_race-S.s was con s i de red as BPCTCA. Jhe addi t ipn.jof ac-
tiV£ted__ carbon or .a second biological stage~~w"1th suspende_d_sol i d.s.,. removaJL
by f i 1 t_ra LJOTL _wa s con s i dered _a s BAT EA . _wh M_e the addijt ion_p,f a, filtration^
sjtjep_._ tP_the__B.P£J£A--ac±ivat_ed sludge process was considered as BADCT. .
These choices appear to be the most reasonable considering the lack of
available data and diversity of wastes involved.
-18-
-------
DRAFT
Reduction factors based upon the performance of existing biological
systems and information available in the literature were the basis used
to develop each technology level. These factors were uniformly applied
to each of the mean RWL's for each of the 13 subcategories subsequently
presented.
-19-
-------
DRAFT
SECTION IV
INDUSTRY CATEGORIZATION
Discussion of the Rationale of Categorization
The goal of this study is to broaden the RWL data base and to further
substantiate the Phase 1 subcategorization. The following is a synopsis
of the subcategorization rationale which was thoroughly discussed in
the Phase 1 study.
The diverse range of products and manufacturing processes to be covered
suggested that separate effluent limitations be designated for different
segments within the industry. To this end, a categorization of the Or-
ganic Chemicals Industry was developed. The categorization is process-
oriented. Chemical commodities have been grouped according to the RWL
associated with their specific manufacturing process.
Manufacturing processes have been examined for the type of contact process
water usage associated with each. Contact process water is defined to be
all water which comes in contact with chemicals within the process and
includes:
1. Water required or produced (in stoichiometric quantities)
in the chemical reaction.
2. Water used as a solvent or as an aqueous medium for the
react ions.
3. Water which enters the process with any of the reactants
or which is used as a diluent (including steam).
*t. Water associated with mechanical devices such as steam-jet
ejectors for drawing a vacuum on the process.
5. Water used as a quench or direct-contact coolant such as in
a barometric condenser.
Non-contact flows not included in the RWL data include the following:
1. Sanitary wastewaters.
2. Boiler and cooling tower blowdowns or once-through cooling
water.
3. Chemical regenerants from boiler feed water preparation.
k. Storm water runoff from nonprocess plant areas, e.g., tank
farms.
-21-
-------
DRAFT
The type and quantity of contact process water usage are related to the
specific unit operations and chemical conversions within a process. The
term "unit operations" is defined to mean specific physical separations
such as distillation, solvent extraction, crystallization, adsorption,
etc. The term "chemical conversion" is defined to mean specific reactions
such as oxidation, halogenation, neutralization, etc.
Description of Subcategories
Four process subcategories have been established and are discussed in the
following text. Subcategories A, B, and C relate to continuous processes,
while Subcategory D relates to batch processes.
Subcategory A: Continuous Non-Aqueous Processes
This group involves minimal contact between water and reactants or
products within the process. Water is not required as a reactant
or diluent, and is not formed as a reaction product. The only water
usage stems from periodic washes of working fluids or catalyst hy-
dration. Heating and cooling are done indirectly or through non-
aqueous (hydrocarbon) working fluids. Process raw waste loads
should approach zero, with variations caused only by spills or pro-
cess upsets.
Subcategory B: Continuous Vapor-Phase Processes Where Process
Water is Used as Diluent or Absorbent
Process water usage is in the form of dilution steam, a direct con-
tact quench, or as an absorbent for reactor effluent gases. Reac-
tions are all vapor-phase and are carried out over solid catalysts.
Most processes have an absorber coupled with steam stripping of
chemicals for purification and recycle. Steam is also used for de-
coking of catalyst. It appears feasible to reduce some process raw
waste loads almost to zero through increased recycle and/or reuse
of contact water in this subcategory.
Subcategory C: Continuous Liquid-Phase Reaction Systems
Liquid-phase reactions involve a catalyst in an aqueous medium such
as dissolved or emulsified mineral salt, or acid/caustic solution.
Continuous regeneration of the catalyst system requires extensive
water usage. Substantial removal of spent inorganic salt by-products
may also be required. Working aqueous catalyst solution is normally
corrosive. Additional water may be required for final purification
or neutralization of products. Requirements for purging waste mate-
rials from the system may prevent process raw waste load from ap-
proaching zero.
-22-
-------
DRAFT
Subcategory D: Batch Processes
Processes are carried out in reaction kettles equipped with agitators,
scrappers, reflux condensers, etc., depending on the nature of the
operation. Many reactions are liquid-phase with aqueous catalyst
systems. Reactants and products are transferred from one piece of
equipment to another by gravity flow, pumping, or pressurization
with air or inert gas. Much of the material handling is manual, with
limited use of automatic process control. Filter presses and centri-
fuges are commonly used to separate solid products from liquid.
Where drying is required, air or vacuum ovens are used. Cleaning
of non-continuous production equipment constitutes a major source of
wastewater.
Basis for Assignment to Subcategories
The categorization assigns specific products to specific subcategories
according to the manufacturing process by which they are produced. Where
more than one process is commercially used to produce a specific chemical,
it is possible that the chemical may be listed in more than one subcate-
gory since the unit operations and chemical conversions associated with
different feedstocks may differ drastically in regard to process water
usage and associated RWL.
A comprehensive listing of chemicals and manufacturing processes which
have been assigned to each of the four subcategories is provided in the
Phase 1 study.
It should be pointed out that the field sampling in Subcategory D for
fatty acids, dyes, pigments and plasticizers were based on end-of-pipe
sampling. Therefore, the associated RWL flow data contains minimal
amounts of non-contact waters. This is in contrast to other products
where non-contact waters were able to be excluded.
-23-
-------
DRAFT
Product ; Cumeme
Process : Alkylation of Benzene by Propylene
Process RWL Category: A
Chemical Reactions:
CH3CH = CH2
Benzene Propylene Cumene
(Isopropyl Benzene)
Typical Material Requirements:
1000 Kg Cumene
Benzene 722 Kg
Propylene 389 Kg
Phosphoric Acid (SPA) 666 Kg
Cumene is an intermediate in the production of phenol. (Acetone and
Acetophenone are co-products.) Cumene is also an excellent blending
fuel in high-octane gasoline.
Although cumene is a naturally-occurring chemical, present in many crude
oils, the conmercial product is synthesized by the catalytic alkylation
of benzene by propylene. The principal side reactions, depending on the
catalyst system employed, include polya Ikylat ion to form di- and tri-
isopropyl -benzene, polymerization of a portion of the propylene, and
the production of n-propyl benzene by i somerizat ion.
Catalyst systems that have been used to produce cumene include such mate-
rials as su If uric acid, hydrofluoric acid, boron trifluoride, silica,
zinc chloride on alumina, and others. However, the two catalyst systems
most commonly used for commercial production are aluminum chloride and
solid phosphoric acid (SPA). The catalyst system used at the process
plant visited during the field data collection program employed the SPA
catalyst .
A typical flow diagram for the solid phosphoric acid process of producing
cumene is illustrated in Figure *t-1 . As shown, this process requires a
reactor and three distillation columns - one for rejection of propane,
one for the recycle of unreacted benzene, and the third to rerun the
cumene product so as to reject a minor quantity of polyal ky lated mate-
rials (mainly di isopropyl benzene) . Yields obtained by this process are
over 90% of stoichiometric values on both benzene and olefin.
-2k-
-------
DRAFT
The aluminum chloride process is similar to that illustrated for solid
phosphoric acid but requires additional equipment for the drying of
recycle streams and neutralization of the reactor products.
A mixture of propylene and propane is blended with both fresh and recycled
benzene in a raw materials feed tank. Water (as steam condensate) at 100-
150 ppm is injected into the reaction mixture, and the reaction mixture
is feed to the top of a fixed-bed reactor, where the liquid trickles down
through the catalyst bed. Steam (non-contact) is used to preheat the
reaction mixture. The process is carried out in a continuous manner.
The reaction product (effluent from the reactor) is filtered; the water
phase (<1.0 liter/day) is removed to a water drain. A depropanizer still
receives the organic phase, and propane is separated out; the propane
can be recycled to the reactor. Wastewater from the propane accumulator
amounts to about 0.3 kg/1000 kg of product.
Unreacted benzene is removed in a benzene distillation column. The
benzene is recycled with fresh benzene to the raw materials mix or feed
tank. Bottoms from the "benzene column" contain cumene and higher alkyl-
aromatics.
Cumene is removed as a product in the finishing still. Diisopropylbenzene
is the major by-product removed as still bottoms.
The only continuous wastewater streams are from the benzene storage area,
the wastewater following the propane accumulator (~0.3 liter/1,000 cumene
produced), and the water phase from the filter following the reactor.
This latter stream is an intermittent flow (^1.0 liter/day).
In newer processing plants, a combination air/water cooling system can
greatly reduce the non-contact water requirement. This is illustrated as
fo11ows:
Cooling Water Circulation Requirement
With air cooling 2^,200 liters/kkg product
Without air cooling 82,600 liters/kkg product
Contributions from storm water runoff, housekeeping, maintenance, water
leaks, and small non-process streams in the cumene area amount to about
65,000 liters/day. Steam required to heat the distillation towers is
7,590 kg per 1000 kg of product.
-25-
-------
DRAFT
Process RWL based on contact wastewater flows are indicated in the
tabulation below:
PROCESS FLOW
liters/kkg 0.331*
gal/M Ibs 0.0*»
BOD,. RWL
mg/liter1 180
kg/kkg2 0.0001
COD RWL
mg/Uter 490
kg/kkg2 0.0001
TOC RWL
mg/liter1 180
kg/kkg2 0.0001
Raw waste concentrations are based on unit weight of pollutants
per unit volume of contact process wastewaters.
2
Raw waste loadings are based on unit weight of pollutants per
1000 unit weights of product.
The raw waste load from cumene is minimal when expressed on a production
basis. The very small contact process wastewater flows make it impractical
to assign any specific effluent limitation to this process. However, it is
recommended that an allowance be made for runoff from the process area. This
should be done on an individual plant basis. Wastes of this type should be
treated such that the BOD concentration is 20 mg/L or less. It should be
noted that this plant (like all cumene plants) is only part of a large
mult!-process facility. As such, it would be more practical to calculate
an allowance for runoff based on the entire plant.
-26-
-------
CO
CO
DRAFT
1
a_
O
CQ
N
- Z
^t LU
CO
is
\J jj^
3NI HSlf
JU 3N3WHG
J
t
d3ddldiS 3N3ZN39
LLJ
Z
LU
U
dU13V3d
-27-
-------
DRAFT
Product: p-Xylene
Process: Isomerization, Crystallization, and Filtration of
mixed Xylenes.
Process RWL Category: A
Because of the rapidly increasing demand for aromatic di-functional acids,
interest in pure xylene isomers have been growing. The Co aromatics found
in catalytic reformate consist roughly of '»5 percent m-xylene, 20 percent
each o- and p-xylene, and 15 percent ethyl benzene. It so happens that
there is much less demand for m-xylene than for either of the other two
xylene isomers. Therefore, an isomerization unit, which is used to shift
methyl groups, converting n- and o-xylene to additional p-xylene, is
frequently built onto existing xylene-isomer separation facilities.
All p-xylene processes currently in operation use a combination of crystal-
lization and centrifugation to separate and purify p-xylene. The crystal-
lization step is usually in concert with o-xylene and/or ethylbenzene
removal and isomerization.
Figure k-2 illustrates a typical two-stage crystallization process with
an isomerization unit.
Crystallization processes generally have the following steps in common,
although the techniques may vary:
Feedstock drying
First-stage crystallization (to about -80°F to -90°F)
Recovery and melting of crystals from first stage
Second-stage crystallization (to about 0°F to -25°F)
Recovery and melting of crystals from the second stage
It is necessary to lower the water content in the feedstock to about 10
ppm, because water introduced into the process will freeze and cause
plugging of the centrifuges and rotary filters. All processes utilize
similar drying techniques, which usually consist of passing the feed
through alumina or silica-gel beds. One bed is on-strearn while the
other bed is being regenerated either by electric heaters or with
jacketed steam.
The major differences in the processes are the mechanics of the crystal-
lization and separation facilities, Most processes use direct refrigera-
tion. The feed is precooled to about -40°F using propane and ethylene,
and the chilled feed is then sent to the first-stage crystallizer at an
operating temperature of -80°F to -90°F. The first-stage crystal 1izers
are usually scraped-surface tubular exchangers or tank crystal 1izers.
In each of these devices, an agitator with spring-loaded blades is used
to scrape the p-xylene crystals from the walls.
-28-
-------
DRAFT
The crystals formed in the first stage are relatively small. Therefore,
strict control of their size is necessary to insure that the centrifuges
or filters used in their recovery will be of adequate size. Increasing
the residence time in the first stage at a relatively low chilling rate
enhances crystal growth.
Considerable advances have been made in the last several years in develop-
ment of more efficient solid-liquid separation devices. Most modern
domestic plants utilize continuous solid-bowl centrifuges in the first
stage. Two bowls rotating horizontally at different speeds cause a
helical screw motion on the outer surface of the inner bowl. The helical
motion moves the solids from the settling slurry pool through a draining
section and then emits a near-dry cake. The centrifuges can be regulated
for bowl revolution speeds, bowl differential, and slurry pool depth.
This can achieve p-xylene first-stage purity of 85%. The centrifuges
can be fitted for backwash, but its benefit to p-xylene is doubtful.
Crystals grown in the first stage tend to be long and thin monoclinic
needles that drain with difficulty. As a result, a sizable portion of
mother liquor tends to remain occluded in the interstices between the
p-xylene crystals.
Crystals from the first stage are melted (or partially melted) and re-
crystallized at about -25°F. The second-stage crystal 1izers are similar
to the first-stage units. The second-stage crystals tend to be cylin-
drical in shape, approximately 200 X 360 microns, and drain more easily.
Furthermore, the viscosity of the mother liquor in the second stage is
about 1 cp (compared to 5 cp in the first stage); this difference in
viscosity enhances the drainage rate. Also, a pusher-plate mechanism
is available, and this tends to increase the drainage rate in the exit
section of the crystal 1izer. About 99.5% pure p-xylene is obtained from
this type of operation.
The filtrates from the first- and the second-stage liquid-solid separation
are sent to the isomerization reactor after mixing with hydrogen gas.
In the reactor, m-xylene is isomerized over a platinum catalyst into a
equilibrium mixture of the three isomers, which is then recycled through
the p-xylene crystallization steps. The isomerization reaction is run
to extinction, that is, only the para isomer of xylene is withdrawn from
the manufacturing operation. The hydrogen gas is required to provide
the hydrogen atmosphere necessary for conducting the xylene isomerization
process and is generally produced on the site. Consequently, the isomeri-
zation and the hydrogen facilities are an integral part of the p-xylene
process.
Other than spills and pump leakages, the major sources of wastewater are
the regeneration water used during decoking of catalyst beds and the
steam drum washdown from the hydrogen plant. These streams are inter-
mittent, and the amounts are insignificant. In addition, the monoethanol-
amine (MEA) contaminated wastewater from the hydrogen plant is difficult
to treat biologically, and is usually hauled off-site for deep-well dis-
posa1 ,
-29-
-------
DRAFT
During the sampling visits, only one (out of three) plant was accessible
for sampling, and the averages of four sets of composite samples are
presented in the tabulation below:
PROCESS FLOW
liters/kkg M.3
gals/M Ib 5.3
BODC RWL
mg/liter1 238
kg/kkg2 0.01
COD RWL
mg/liter1 580
kg/kkg2 0.025
TOC RWL
mg/liter 159
kg/kkg2 0.00?
Raw waste concentrations are based on unit weight of pollutants
per unit volume of contact process wastewaters.
2Raw waste loadings are based on unit weight of pollutants per
1000 unit weights of product.
As with the cumene processes, it is not considered practical to establish
a single-value effluent limitation for the production of p-xylene as
previously described. Instead, it is recommended that waste allocation
assigned to this process be related to runoff or drainage from the process
area. Such an allocation should be based on treating these wastewaters
to a BOD concentration of 20 mg/L. It should also be noted that p-xylene
plants do not exist as separate entities but rather are part of a facility
manufacturing other aromatic chemicals or oxidized product acids, such as
terephthalic acid.
-30-
-------
DRAFT
X
I
Q
LLJ
X
z
o
Q
Z
UJ <
co
>-
oe.
(J
*»*.
Z
O
M
at.
O
to
X
I
<
OL
<
Q_
-31-
-------
DRAFT
Product: BTX (Benzene, Toluene and Xylene)
Process: Extraction with DIethylene Glycol (Licensed UOP Process)
Process RWL Category: A
Chemical Reactjpns: None (Liquid - Liquid Extraction)
This process is a licensed (UOP) extraction process that handles both
reformate charges as well as pyrolysis gasoline. The reformate charge
(naphtha feed) is first fractionated, and only the intermediate fraction
is charged to the extraction unit. The heavy and light ends from the
fractionator are sent to motor gasoline production. The two sources of
feed (reformate and pyrolysis gasoline) are charged into the extraction
column, along with the extraction solvent (diethylene glycol). The raf-
finate is stripped of any carryover solvent and sent back to the refinery.
The extract is treated to remove the solvent and then fractionated to
obtain (primarily) benzene and toiuene. No attempt is made to recover
any xylenes, much less separate then in this unit. The main sources of
water pollution are the final wash waters of the sulfiner unit, the glycol
recovery unit, and pump cooling water.
The product/process has been investigated in Phase I of this study, and
the detailed process description and process flow diagram can be found
on Pages 86-88 of EPA's document, EPA 440/1-73/009 (Development Document
for Proposed Effluent Limitations Guidelines and New Source Performance
Standards for the Major Organic Products, Segment of the Organic Chemicals
Manufacturing Point Source Category) published December, 1973.
The process RWL calculated from flow measurements and analyses of waste-
water samples obtained in the Phase II survey period are shown in the
tabulation below:
Plant 2
PROCESS FLOW
1iters/kkg
gals/M Ibs
BOD5 RWL
mg/1iter
kg/kkg2
COD RWL
mg/1iter
kg/kkg2
TOC RWL
mg/1i ter'
kg/kkg2
Plant 1
Sample
Period #1
933
112
34
0.032
121
0.113
37
0.035
Sample
Period #2
933
112
26
0.025
135
0.126
34
0.032
Sample
Period #3
933
112
34
0.032
120
0.112
29
0.027
Sample
Period #1
312
37.4
2990
0.92
37500
11.6
64430
19.9
'Raw waste concentrations are based on unit weight of pollutant per unit
volume of contact process wastewaters.
2Raw waste loadings are based on unit weight of pollutant per 1000 unit
weight of product.
-32-
-------
DRAFT
These data are reported to supplement the information on RWL for this
type of process obtained In Phase I. However, since effluent limitations
for the process have already been established, they are not included in
the subsequent development of Phase II limitations.
-33-
-------
DRAFT
Product; BTX Aroma tics
Process: Fractional Distillation
Process RWL Category; A
Chemical Reactions: None
The product obtained here is a mixed one consisting of benzene, toluene,
and xylenes which are separated from parafinic, olifinics, and mixed high-
boiling aromatic organics (over 150°C). Some ethyl benzene is also recovered,
The process involves a series of fractionating columns. There is no direct
contact water; consequently raw waste loads are very low.
PROCESS FLOW
liters/kkg 1*6.7
gals/M Ibs 5.6
BODj. RWL
mg/liter1 320
kg/kkgz 0.015
COD RWL
mg/liter1 1150
kg/kkg2 0.053
TOC RWL
mg/liter1 328
kg/kkg2 0.015
1Raw waste concentrations are based on unit weight of pollutant
per unit volume of contact process wastewaters.
2
Raw waste loadings are based on unit weight of pollutant per
1000 unit weights of product.
These data are reported to supplement the process RWL data obtained in
Phase I. Since effluent limitations for this type of process have
already been established, these data are not included in the subsequent
development of Phase II limitations.
-------
DRAFT
Product: Ethyl Benzene
Process: Alkylation of Benzene with Ethylene
Process RWL Category: A
Chemical Reactions:
Al C13
Benzene Ethylene Ethyl Benzene
This particular product process and reaction were covered in Phase I of
this study and can be found on Pages 76 and 77 of EPA's document, EPA
440/1-73/009 (Development Document for Proposed Effluent Limitations
Guidelines and New Source Performance Standards for the Major Organic
Products Segment of the Organic Chemicals Manufacturing Point Source
Category) published December 1973.
However, this product/process has also been surveyed in Phase II of this
study, and process RWL calculated from the flow measurements and analyses
of wastewater samples are shown in the tabulation below:
Plant 1 Plant 2
Sample Sample Sample
Period #1 Period #2 Period
PROCESS FLOW
liters/kkg 339 315 717
gals/M Ibs 40.6 38.7 86
BODr RWL
mg/liter1 6 34 302
kg/kkg2 0.002 0.011 0.217
COD RWL
mg/liter1 1,700 1,270 1,400
kg/kkg2 0.576 0.411 1.01
TOC RWL
mg/liter1 139 120 443
kg/kkg2 0.047 0.039 0.318
1Raw waste concentrations are based on unit weight of pollutant
per unit volume of contact process wastewaters.
2Raw waste loadings are based on unit weight of pollutant per
1000 unit weights of product.
These data are reported to supplement the process RWL data obtained in
Phase I. Since effluent limitations for this type of process have
already been established, these data are not included in the subsequent
development of Phase II limitations.
-35-
-------
DRAFT
Product: Adiponitrile
Process: Chlorination of Butadiene
Process RWL Category: B
Chemi ca1 React ions:
CH CH CH CH9 + C19 > Mixture of isomeric d ichlorobutenes
Butadiene + Chlorine
H9
NCCH2CH = CH CH2CN 5> NC (CH^CN
Dicyariobutene Adiponitrile
Adiponitrile (ADN) is commonly used during the manufacture of hexamethylene
diamine (HMDA). Hexamethylene diamine is in turn used as a raw material
during the production of Nylon 6/6.
A schematic flow diagram of the process is given in Figure ^-3, and the
general process information is described in the following paragraphs:
The first step in the manufacture of ADN via butadiene is the vapor-phase
chlorination of butadiene. The reaction is carried out at 250°C and at
pressures slightly above atmospheric. The inlet section of the reactor
is designed to insure complete and rapid mixing of the reactants, and
the rest of the vessel should be constructed to guarantee plug-flow condi-
tions, which favor the high degree of chlorine conversion desired. By-
product formation is minimized if the molar ratio of butadiene to chlorine
is kept around 6.5 to 1 and if the vapor phase is diluted with nitrogen
gas to keep a 10% (volume) of inerts in the reactor feed stream. The
ratio of fresh to recycle butadiene is approximately 0.2 to 1. When the
chlorine conversion is close to 100%, the molar yield of chlorine to
chlorinated butenes is 95%. The ratio of the 1,4 and 3,^ dichlorobutene
isomers is of no importance since in subsequent reaction steps both
isomers are converted to the same final product. The composition of the
chlorinated butenes is as follows: 93.*» wt % dichlorobutenes , 1.6 wt %
low boilers and k.3 wt % high boilers. The vapor-phase reaction is carried
out in an empty reactor. The reactants enter at 120°C and are heated up
to the reaction temperature (250°) by the heat evolved from the reaction,
which is 39 kilocalories per mole.
The effluent stream from the reactor is cooled down to room temperature,
and condensed. The unreacted butadiene is separated from the liquid phase,
which contains the chlorinated products, and recycled back to the reactor.
The liquid phase is sent to the cyanization section of the plant. This
liquid is composed of a mixture of dichlorobutenes, chlorobutenes, chlori-
nated butanes, and heavy materials such as tar and polymeric compounds.
-36-
-------
DRAFT
The cyanization reaction is carried out in the liquid phase at 100°C
and under atmospheric pressure in the presence of an inert gas such as
nitrogen. The reaction is catalyzed by an aqueous mixture of cuprous
chloride, copper powder, and cuprous cyanide. The molar ratio of hydro-
gen cyanide to dichlorobutenes is close to 2.3 to 1. The composition
of the liquid feed to the reactor is given below on a weight percent
basis.
cuprous cyanide 0.93
calcium carbonate 15.6
water 5^.6
hydrogen cyanide 9.6
dichlorobutenes 19-5
The reaction time to achieve a 92.5 mole % conversion to 1,A-dicyano-
butenes is approximately 'iO minutes.
The next step in the production of ADN is the isomerization and purifica-
tion of cyanobutenes. The purpose of the isomerization is the conversion
of some of the 1,4-dicyanobutene-2 to its isomer, 1,A-dicyanobutene-1.
The purpose of the purification is to render the crude dicyanobutene
mixture non-corrosive.
The liquid reaction product stream leaving the cyanization reactor is
cooled to 25°C. The cyanobutenes and organic by-products are extracted
with benzene and diluted until the benzene concentration is approximately
66% by weight. This organic mixture is sent to an agitated tank, where
the temperature is raised to 60°C and the pH is adjusted, with 10% sodium
hydroxide, to 11.5. The two liquid phases that result are separated in
a liquid-liquid separator; the aqueous phase is sent to the waste treat-
ment plant, while the organic phase, after being heated to 70°C, is sent
to another agitated tank where dilute (10%) sodium hydroxide is added
until its weight percent in the aqueous phase is around 15. The two
liquid phases are agitated vigorously in the tank, which has an average
residence time of 25 minutes. The two-phase effluent stream is sent to
a decanter where the organic and aqueous phases are separated; the aqueous
phase is sent to the waste treatment plant, and the organic phase (con-
taining the isomeric mixture of 1,^-cyanobutenes) is sent to the adiponi-
trIle plant.
The benzene-cyanobutene mixture is diluted with benzene until the cyano-
butene concentration is 20% by weight, and then is fed to the hydrogena-
tion reactor. This reaction vessel contains a packed bed of activated
charcoal promoted with 2% (by weight) palladium. The hydrogenation
reaction is carried out under a hydrogen pressure of AOO psig and at a
temperature between 100 and 120°C. The hydrogen is bubbled cocurrently
with the liquid at a ratio of 35 moles of hydrogen per mole of dicyano-
butene. The liquid space velocity in the reactor is approximately O.A
-37-
-------
DRAFT
volumes of liquid per hour per volume of catalyst. The molar conversion
to adiponitri1e is 35%, with a selectivity of 99%. Since the reaction
is quite exothermic, the reactor must be equipped with an efficient heat-
removal system. For example, the reaction should be carried out inside
of tubes packed with the catalyst while cooling water runs through the
shell side of the reactor. The catalyst has an active life of about
500 hours, and it can be easily regenerated by passing hydrogen at 500°C
over the catalyst until no more sulfur is evolved from the reactor tubes.
The reactor effluent stream is cooled down to room temperature, and the
hydrogen is separated from the liquid phase and recycled back to the
reactor. The crude adiponitrile organic phase is sent to the adiponi-
trile purification plant.
The first step in the purification consists of flashing the liquid in
order to vaporize the hydrogen dissolved in the liquid. The flashed
vapors contain some benzene; after cooling to 50°C this benzene is con-
densed and returned to the benzene tank, while the non-condensable gas
is flared. The liquid stream leaving the flash drum (containing most
of the benzene, adiponitrile, unreacted byanobutenes, and organic im-
purities) is sent to the benzene evaporator. The benzene recovered in
the evaporation step is condensed and sent to the benzene storage units.
The bottoms from the evaporator are sent to a distillation column, where
the unreacted cyanobutenes and the benzene remaining in the system are
distilled off and recycled back to the hydrogenation reactor. The crude
adiponitrile leaving at the bottom of this column is usually further
purified prior to hydrogenation to HMDA. The purification begins by
removing water with a dehydration agent such as HMDA. This is achieved
by sending the crude adiponitrile to a dehydration column where a water-
HMDA azeotropic mixture is collected as distillate. The water phase is
discharged, and the HMDA-rich phase is used as the column recycle. The
bottoms from this column are sent to a second distillation unit, where
low boilers are distilled off. In a subsequent distillation step, re-
fined adiponitrile is collected as the distillate.
The wastewater pollution sources of this process include various scrubber
effluents and stream jet condensates from each reaction step as indicated
in the process flow diagram. Although it is apparent from the previous
process description that the sections of the process associated with the
manufacture of dicyanobutene and adiponitrile involve aqueous liquid-phase
reaction mixtures, the wastewaters discharged to the sewer are primarily
spent scrubber water and vacuum jet condensates. These wastes are typical
of a Category B vapor-phase manufacturing operation. Process raw waste
loads calculated from flow measurements and analyses of these streams are
shown in the following tabulation:
-38-
-------
DRAFT
PROCESS FLOW
Liter/kkg
(gal/M Ib)
RWL
mg/1 iter1
kg/kkg2
Ad i pon i t r i 1 e
Sample Period #1 Sample Period #2 Sample Period #3 Average
COD RWL
mg/1iter
kg/kkg2
9766
1170
1250
12.2
15,000
TOC RWL
mg/1iter
kg/kkg2
1
4600
9766
1170
2850
27.8
14,400
141
4500
43.8
9766
1170
1800
17.6
12,000
118
4450
43.4
9766
1170
1970
19.2
13,800
135
4500
44.0
1
Raw waste concentrations are based on unit weight of pollutant per
unit volume of contact process wastewaters.
2Raw waste loadings are based on unit weight of pollutant per 1000
unit weights of product.
The analytical results indicate that the pollutants in the wastewaters
(such as ammonia nitrogen, sulfate, cyanide, chloride, and copper) are
at levels hazardous to biological treatment processes. The low BODt;
values shown in the tabulation are due to the interference of the
logical-inhibiting pollutants.
bio-
It should be noted and clearly understood that although biological treat-
ment was chosen as the model treatment system most generally applicable
for BPCTCA, several of the process plants surveyed utilize other means
of disposal, such as deep-well injection. This is the case with this
adiponitrile plant. The raw waste loads shown previously are all treated
by filtration prior to disposal by deep-well injection. The effluent
limitations developed subsequently for adiponitrile would be applicable
only if deep-well injection were no longer feasible. In this regard,
it should be noted that this adiponitrile plant uses non-contact cooling
water on a once-through basis. The volume of cooling water amounts to
424,000 liters/kkg of adiponitrile. If it becomes necessary to treat
the wastes from this adiponitrile plant in a biological system, dilution
with some of the once-through cooling water might lessen the inhibitory
characteristics of the waste in its present concentrated state.
-39-
-------
DRAFT
UJ
Z
UJ
O
<
UJ <
§ I
I
U
Z
o
a.
O
<
«r CD cj
-40-
-------
Product: Benzoic Acid and Benzaldehyde
Process: Catalytic Oxidation of Toluene with Air
Process RWL Ca tegpry: B
Cnem1 ca 1 Reac t i ons :
2C,HCH + 300 > 2C,HCOOH + 2H00
D 5 3 ^ "5
Toluene Benzoic acid
C,H.-CH, + 0. >C,H_CHO + H 0
b t> 3 L 65 2
Toluene Benzaldehyde
A simplified flow diagram for this oxidation process is shown in Figure k-k.
Toluene and air, with a mixture of recycle gas and toluene, are combined
in the reactor. Reaction temperatures may range from 150°C to 500°C., with
a corresponding pressure range of 10 to 1 atmospheres. Specific manufac-
turers operate at different conditions. Depending upon the operating
conditions, the material in the reactor may be in the liquid or vapor
phase, and reaction conditions may be varied to give different selectiv-
ities to benzoic acid or benzaldehyde. The plant surveyed produces 6-7
kg of product benzoic acid for every kg of benzaldehyde co-product. The
process RWL's have been calculated on the basis of the total production
of benzoic acid and benzaldehyde during the sampling period. Waste
streams associated with either the benzoic acid or the benzaldehyde recovery
sections can be allocated by using the product/co-product ratio mentioned
above.
The catalysts most frequently used consist of oxides of metals belonging to the
fifth or sixth groups of the periodic system. A mixture of uranium oxide
(93 percent) and molybdenum oxide (7 percent), impregnated on a pumice or
asbestos carrier, is claimed to give relatively high yields of benzaldehyde,
with low percentages of toluene going to complete combustion. The addi-
tion of small amounts of copper oxide to the catalyst mixture reduces by-
product maleic anhydride formation. A cobalt acetate catalyst in aqueous
solution has also been successfully used; in this case, the metal catalyst
concentration is kept below 100 mg/1 in the reactor.
As shown in Figure *»-A, the reactor effluent is sent to a decant tank.
A three-phase mixture of gas, organic liquid, and aqueous liquid usually
exists in this vessel. The gas (mainly inert nitrogen) is vented under
pressure control, with a portion recycled to the reactor to maintain
proper dilution. The aqueous liquid phase (containing water of reaction)
is decanted off using an interfacial level controller.
-41-
-------
The organic layer is sent to the main stripper, where toluene is taken
overhead and recycled to the reactor, which normally operates at 30-35%
toluene conversion per pass. The bottoms from the main stripper are
crude benzoic acid and by-products such as maleic anhydride, anthraquinone,
and complex high-boiling aromatics of unknown composition. The crude
acid is sent to the product (benzoic acid) purification section of the
process.
A side stream containing crude benzaldehyde co-product is also taken
off the main stripper as shown in Figure 4-4. The crude benzaldehyde is
sent to the benzaldehyde stripper, where benzaldehyde is taken overhead
and the bottoms recycled to the main stripper.
The benzaldehyde is washed with an aqueous solution of sodium carbonate
in the wash tank. This step is necessary to neutralize organic acid by-
products (and some benzoic acid) present with the benzaldehyde. The
aqeous layer from the wash tank is drained and discharged. The organic
layer from the wash tank is sent to the benzaldehyde still, where puri-
fied benzaldehyde is taken overhead. The benzaldehyde still operates
under vacuum drawn by a vacuum pump. Seal water from the pump is con-
tinuously discharged. An organic residue, which must be removed periodi-
cally by water washing, forms in the bottom of the benzaldehyde still.
This material is discharged.
The benzoic acid purification section consists of two rectifying columns
and a batch tar stripper. Each of these distillation columns operates
under vacuum drawn by steam jet ejectors. The steam is not condensed,
but rather is discharged directly into the atmosphere; consequently,
this steam was not considered in the process RWL calculations. It should
be noted that the gases drawn into the jets have been scrubbed; conse-
quently, little carryover of organic material is anticipated.
The crude benzoic acid from the main stripper is sent to the middles
column, where partially rectified benzoic acid is drawn off as bottoms.
The overhead gases are drawn by vacuum through the middles column scrubber,
where they are scrubbed with an aqueous caustic solution prior to dis-
charge through the steam jets. The caustic solution is recycled by
means of the scrubber recycle tank; fresh caustic is added periodically
to the circulating solution, with a portion drawn off to maintain a high
pH.
The bottoms from the middles column are sent to the product column, where
product benzoic acid is drawn off. The overhead vapors from the product
column are scrubbed with caustic in a recirculating system similar to
that used with the middles column (see Figure 4-4). Bottoms from the
product column are sent to the tar stripper, where additional benzoic
acid is stripped batchwise from by-product tar.
-42-
-------
The major RWS's for this process are summarized in the following tabula-
tion:
Waste Stream Flow BODg COD TOC
TTkkg kgTkkg kgTkkg kgTkkg
Water of Reaction 2k7 8.7A 15.0 6.52
Aqueous Drain from
Benzaldehyde Wash Tank 52.2 11.9 16.1 8.3^
Benzaldehyde Still Residue 5.0
Vacuum Pump Seal Water 2,200 0.91 1.61 0.73
Middles Column Scrubber
Slowdown 209
Product Column Scrubber
Slowdown 122
TOTAL 2,840
The major source of waste loadings in the process is the aqueous drain from
the benzaldehyde wash tank, where organic concentrations are greater than
100,000 mg/1. Materials present include sodium benzoate and other aroma-
tics such as biphenyls. It is questionable whether this stream could be
incinerated, because of its high alkalinity. However, some removal of
organics may be possible by acidification of the wastewater followed by
gravity separation. The organics removed could then be burned.
The raw waste load as presented in the foregoing tabulation was considered
as BPCTCA. It is discharged, without pretreatment, to the municipal sewer
system.
-------
UJ CJ>
1 O
U 0=
DRAFT
n
a. r^i
A | Z
To UJ
IcjCO
mis
3QAH3Q1VZN38
CO
LU
a
i
LU
O
_J
M
Z.
«t uj
uj O
at "7.
=> <
go
U
<
u
O
N
-z.
LU
CO
_
0= C=
U- CO
H3ddlHlS
3QAH301VZN38
-»
H38
ce ^ oo
anass
-3'd
-*
LU LU
CO _J
CO CJ
t!398n)J3S
NiVN
NWII103
isnuotid
.. . .,
J W
UJ LU
CO I
CO CJ
O- CJ C3
CO CO CO
S31QO N
J
-44-
-------
DRAFT
Product: Chlorinated Methanes (Methylene Dichloride, Chloroform, and
Carbon Tetrachloride)
Process: Chlorination of Methyl Chloride and Methane Mixture
Process RWL Category: B
Chemical Reactions:
CH^ + C12 > CH3C1 + HC1
Methane Methyl Hydrogen
Chloride Chloride
CH3C1 + 2C12 > CH2C12 + HC1
Methyl Chloride Methylene
Dichloride
CH3C1 + 2C12 > CHC1- + 2HC1
Methyl Chloride Chloroform
> C Cl. + 3HC1
Methyl Chloride Carbon
Tetrachloride
CH2C12 + C12 > CHC13 + HC1
Methylene Dichloride Chloroform
CHC13 + C12 > c C1A + HC1
Chloroform Carbon
Tetrachloride
Typica' Material Requirements:
The material requirements depend upon which of the chlorinated products is
desired. Ir. general, the chlorine consumption is 1% beyond theoretical
needs, but it ffiay be lowered depending on the chloromethanes products mix.
The methyl chloride requirement also is approximately 7% beyond theoretical
needs.
CMoromethanes find their widest application as solvents. Methylene di-
chloride 's used in the plastics field as a solvent for polycarbonates,
isocyar-:res and cellulouse diacetate, a urethane-foam blowing agent; in
-------
DRAFT
Europe, it is important as the spinning solvent for cellulose acetate.
Chloroform is used as a solvent for textile degreasing and an extractant
for food flavors, steroids, and antibiotics. Carbon tetrachloride is
used as a solvent in non-flammable cleaning agents.
In addition to their wide use as solvents, there are numerous other appli-
cations of chloromethanes. Methylene dichloride is used as a non-flammable
paint remover, and also in purification of steroids as a reaction medium,
e.g., in the manufacture of phosphates, insecticides, and vegetable oil
extracts. Two developments of great potential importance are the use of
methylene dichloride as a polyurethane-foam blowing agent and in aerosal
hair sprays.
Chloroform finds wide use as an intermediate in the production of other
materials. In addition, chloroform is used in certain pharmaceutical
formulations such as cough medicines, rubbing liniments, and anesthetics.
Carbon tetrachloride is used primarily in the manufacture of methylene
chloride and chloroform. However, it is also used as a fluid in certain
fire extinguishers.
Chlorinated methanes can be produced from methane, from methyl chloride,
or from a mixture of both materials. The reactions may be carried out by
either thermal or photo activation. The former, which requires a tempera-
ture of approximately 700°F, is preferred commercially, because it requires
lower investment and maintenance, and allows more complete conversion of
chlorine.
A typical process flow diagram for producing chloromethanes starting with
a mixture of methane and methyl chloride is shown in Figure A-5.
The feed methane is first purified and dried in a purification unit, and
then is fed to the chlorination reactor along with fresh and recycled
methyl chloride and chlorine gas. Following the reaction step, the
chlorinated products are quenched and absorbed in a refrigerated mixture
of recycle carbon tetrachloride and chloroform. Methane and HC1 are taken
overhead from the quench column and are absorbed in weak hydrochloric acid,
which removes the HC1. The remaining methane is then passed through a
caustic scrubber before being returned to the chlorination reactor.
The bottoms stream from the quench tower is stripped of its light consti-
tuents. These are absorbed in water to remove the remaining HC1, and are
then neutralized with caustic solution. This light-product stream is
then passed through a series of distillation towers from which methyl
chloride, methylene dichloride, chloroform, and carbon tetrachloride are
in turn obtained as final products for sale or recycled back to the process
as raw material or as absorbent.
-k6-
-------
DRAFT
The major water pollution sources of the process are the waste streams
discharged from the HC1 absorber and the caustic scrubber. Process RWL
calculated from the flow measurements and analyses of water samples ob-
tained in the survey period are presented in the following tabulation.
The analytical results also indicate that, in addition to the parameters
shown in the tabulation, parameters such as pH and chloride are at levels
hazardous to biological treatment processes.
Plant 1
PROCESS FLOW
1iter/kkg
gal/M Ib
BOD5 RWL
mg/1i ter^
kg/kkg2
COD RWL
mg/15 ter'
kg/kkg2
TOC RWL
mg/1iter
kg/kkg2
Sample
Period #1
598
72
113
0.0?
A20
0.25
Sample
Period #2
598
72
18
0.011
385
0.23
0.25
Plant 2
Sample
Period #1
2800
335
77
0.22
335
0.3k
132
0.37
Raw waste concentrations are based on unit weight of pollutant
per unit volume of contact process wastewater.
2
Raw waste loadings are based on unit weight of pollutant per
1,000 unit weights of product.
The RWL shown for Plant 2 was considered commensurate with BPCTCA. These
wastewaters are neutralized and discharged to surface waters. The wastes
shown for Plant 1 are not considered as representative of the process, in
that the chloromethanes process was integrated as part of a chemical complex
so that both of the waste streams could be utilized elsewhere in plant.
The values shown for Plant 1 represent only surface runoff.
-if?-
-------
DRAFT
CD
CO
Z uj
* Z
o <
-i X
u ^
uj Z 5
c^ < <
=> X UJ
O I- Q
_ LLJ
< y
Z >
-48-
-------
DRAFT
Product: Chlorobenzene
Process: Chlorination of Benzene
Process RWL Category: B
Chemical Reaction:
C6H6 + C12 > C6Hccl + HC1
Benzene Chlorobenzene
Chlorobenzene, an important intermediate in the manufacture of dyes and
insecticides, is manufactured by the chlorination of benzene. Two faciil-
ities were visited during the field-data collection program, one which
manufactured Chlorobenzene exclusively, and one which also produced di-
chlorobenzene. Figures ^-6 and k-J are process flow diagrams of these
two facilities, respectively.
At the first facility (chlorobenzene only), benzene and chlorine are fed
to the reactor, and the hydrochloric acid produced by the reaction is
absorbed in water to make aqueous hydrochloric acid; the chlorobenzene
formed by the reaction is neutralized by sodium hydroxide and then puri-
fied by distillation. The brine formed during the neutralization step
goes to the sewer, and a tar-like residue from the chlorobenzene distil-
lation is incinerated.
The second facility (chlorobenzene-dichlorobenzene) uses a very similar
processing strategy, with a few additions. The hydrochloric acid coming
from the absorption step is purified by carbon adsorption, and the tail
gas which passes through the absorber is scrubbed with water prior to
exhaust to the atmosphere. Another difference in the processing strategy
at this facility is in the purification of the chlorobenzene reaction
products. The products proceed to a distillation column, where any un-
reacted benzene is taken overhead and recycled to the chlorinator. The
monochlorobenzene proceeds from the bottom of this distillation column to
a second distillation column, where chlorobenzene product is taken over-
head while the bottoms from the distillation column proceed to dichloro-
benzene refining.
In the survey period, wastewater samples from each facility were collected
for analysis. Unfortunately, the flow rates (in terms of gal/1000 Ibs of
product) were not provided by the manufacturers, and no RWL can be calcu-
lated.
-------
DRAFT
CD
CD
as
N
z
O
z
O
u
N
Z
UJ
00
O
c*
O
Nwmoo
NOIimilSIQ
-50-
-------
DRAt'T
z
LJJ
CO
LJ_
O
o o
iZ x
U
M
Z
LJJ
CQ
o
u
Q
NOIldHOSIJtf
CJJ Z.
I CQ err
"
NU 1 l\m US 1 U
1
NWII103
NOIlsnilSIQ ,
-51-
-------
DRAFT
Product; Chlorotoluene
Process : Chlorination of Toluene
Process RWL Category: B
Chemical Reaction;
C6H5CH3 * C12 > C6H4 (CIV C1 + HC1
Toluene Chlorine Chlorotoluene Hydrogen Chloride
A simplified process flow diagram for the manufacture of Chlorotoluene is
shown in Figure 4-8. The chlorine gas is reacted with liquid toluene in
the presence of a catalyst in the reactor. The unreacted chlorine gas
is absorbed by water in an absorption tower, and the resulting aqueous
hydrochloride is sent to the muriatic acid plant for reprocessing. The
crude product mixture is purified by passing through a series of distil-
lation stills. The light ends and residues are disposed of by incineration,
while the unreacted toluene is recycled back to the reactor.
Vacuum distillation is employed in purifying the crude product. The
steam jets (with barometric condensers) used to pull the vacuum consti-
tute the only wastewater pollution source of the process. The process
RWL calculated from flow determinations and the analyses of the waste
stream are indicated in the tabulation below:
PROCESS FLOW
liter/kkg 121,000
gal/M Ib 14,500
BOD5 RWL
mg/liter1 2
kg/kkg2 0.24
COD RWL
mg/liter1 15
kg/kkg2 1.82
TOC RWL
mg/liter 2
kg/kkgz 0.2k
'Raw waste concentrations are based on unit weight of pollutant
per unit volume of contact process wastewaters.
2Raw waste loadings are based on unit weight of pollutant per
one thousand unit weights of product.
-52-
-------
DRAFT
The only wastewater from this process is the large quantity of steam jet
condensate from the vacuum still. The contaminant concentrations in
these streams are, however, quite low. In the near future, the process
will be modified from vacuum distillation to atmospheric distillation.
Plant personnel anticipate that this modification will reduce the waste-
water volume and pollutants by 25%. At the present time, all process
wastewater is discharged to the local municipal wastewater treatment
plant.
All cooling water is non-contact. Steam which is used in the process
is also non-contact, and the condensate is returned to the boiler plant.
-53-
-------
DRAFT
FIGURE 4-8
CHLOROTOLUENE - CHLORINATION OF TOLUENE
WATER
TOLUENE
CHLORINE
ABSORBER
AQUEOUS HCL TO
MURIATIC ACID
PLANT
OTF.H
STE*M
.CHLORO-
TOLUENE
STEAM JET
CONDENSATE
RESIDUE TO
INCINERATION
-------
DRAFT
Product : Diphenylamine
Process : Deamination of aniline
Process RWL Category : B
Chemical R e a c t ? on :
Aniline Diphenylamine Ammonia
Typical Material Requirements:
1000 kg Diphenylamine
Anil ine 800 kg
By-Products:
Ammonia 80 kg
Tars 120 kg
Diphenylamine (DPA) is used extensively in the rubber chemicals field,
generally as a retarder, and its derivatives are employed as ant ioxidants.
Production of diphenylamine may proceed by various routes; however, the
plant visited during the field survey utilized a vapor-phase catalytic
reaction involving the deamination of aniline. A simplified process flow
diagram is shown in Figure A-9-
As shown in the diagram, liquid aniline is pumped at a uniform rate from
storage tanks into the externally-heated vessel which serves as preheater
and vaporizer. In this vessel, aniline is vaporized, and the vapors are
heated to a temperature of approximately 400°C to 500°C. Hot aniline
vapors pass through the catalyst chamber, which is maintained at approxi-
mately AOO°C to 550°C. The exit gases of the DPA converter then pass
into an aniline stripper, in which aniline and other volatile constituents
(such as ammonia, small amounts of water, and other volatile by-products)
remain in the vapor state, while DPA condenses with some aniline and is
drawn off as crude DPA. The crude product is then purified in a series
of distillation columns.
The gases leaving the aniline stripper are cooled in an aniline condenser.
A portion of the liquid aniline is refluxed, and the remainder is recycled
to the aniline feed line. The vapors (containing mostly ammonia, small
amounts of water, and other volatiles) pass to an ammonia scrubber.
-55-
-------
DRAFT
DPA converters will not operate indefinitely, because the reaction causes
a certain amount of decomposition of aniline, which deposits carbonaceous
residues in the catalyst. These residues reduce the efficiency of the
catalyst, and it is necessary to regenerate. As a result, in a normal
plant there is more than one DPA converter. Some converters are on a re-
generation cycle while the remaining ones are on a production cycle.
On the average, each production cycle lasts 50 hours.
Regeneration is effected by the following procedures:
1. Steam is introduced to vaporize the aniline and DPA.
2. Steam and air are introduced to burn the carbonaceous impuri-
ties deposited on the catalyst surface.
3. Steam is then introduced to purge the reactor.
k. An aniline purge Is used to remove water vapor,.
The off-gases resulting from the burning of tars are exhausted, and ani-
line and DPA are recycled.
The major wastewater source in this process is the effluent from the am-
monia scrubbing tower. The wastewaters from the catalyst regeneration
process are usually disposed of by incineration and are not included in
the raw waste load calculations. The total process RWL from flow measure-
ments and the analyses of the contact wastestreams during the sampling
survey are presented in the following tabulation:
PROCESS FLOW Sample Period #] Sample Period if2
liter/kkg 52552^
gal/M Ib 63.0 63.0
BOD RWL
mg/Hter1 220 108
kg/kkg2 0.116 0.057
COD RWL
mg/liter1 550 650
kg/kkg2 0.287 0.339
TOC RWL
mg/liter1 **50 *»20
kg/kkg2 0.237 0.218
!Raw waste concentrations are based on unit weight of pollutant per unit
volume of contact process wastewaters.
2Raw waste loadings are based on unit weight of pollutant per 1000 unit
weights of product.
-56-
-------
DRAFT
The arithmetic average of the values presented In the foregoing tabulation
for BPCTCA. Process wastes are discharged to surface waters.
All cooling water used during the production of diphenylamine is indirect;
tube and shell exchangers are employed. Steam usage is 1.15 1bs per Ib
product (including the steam employed during catalyst regeneration), and
approximately 75% of the condensates are collected. Steam used in the
regeneration cycle is live steam, and the rest is reboiler steam.
Diphenylamine can also be produced by catalytic liquid-phase process,
in which a mixture of primary arylamine and catalyst is mixed with re-
actant aniline in a stainless steel reaction chamber. The temperature of
the reaction mixture is raised to 175°C to 450°C, and the pressure is per-
mitted to build up to keep the reaction mixture in a liquid state. During
the course of the reaction, ammonia is split and vaporized. The ammonia
vapor pressure can be utilized to maintain the reaction mixture in a
liquid state. After holding the reaction mixture for the requisite time
at the desired temperature and pressure, the hot mixture is permitted
to flow into the receiver. On cooling the mixture to about 275°C, the
catalyst is substantially crustallized from the mixture. The crude
reaction mixture can be filtered and then purified to the final product.
-57-
-------
DRAFT
CO
UJ
CO
C£J<
Z
_J
z
z
o
51
UJ
O
UJ
Z
DIPHENY
NoiidHosav
VINONNV
c:,
NoiiniiisiQ
-58-
-------
DRAFT
Product: Hexanethylenediamine (HMDA)
Process: Hydrogenation of Adiponitrile (ADM)
Process RWL Category: B
Chemical Reactions:
NH,
NC(CH2)j,CN + 4H2 J » H2N(CH2)6NH
Adipon i tri1e Hexamethylenediamine
Hexamethylenedfamine (HMDA)is used in the production of nylon, where
it is combined with adipic acid to form polyhexamethylene adipamide,
commonly known as nylon. The demand for HMDA is estimated to be 900
million pounds by 1975, and 1,250 million pounds by 1980.
Figure 4-10 presents a process flow diagram of hexamethySenediamine
production via the hydrogenation of adiponitrile.
The hydrogenation of adiponitrile is carried out by contacting hydrogen
with an ammonia-solvent-ADN liquid mixture in a fixed-bed reactor contain-
ing 8-14 mesh cobalt-oxide catalyst at 100-250°C and ?00-700 atmospheres.
This transformation is characterized by its high selectivity to HMDA at
almost complete conversion of ADN.
Fresh adiponitrile and the recycle solvent (toluene) are nixed at 90°F
before being pumped to about 4,400 psia pressure and heated to 190°F.
This stream is then nixed with fresh and recycled hydrogen and ammonia.
The gaseous effluent is recycled, while the liquid effluent is depressur-
ized to 3,000 psia and heated to 400°F. This stream goes to a separator
followed by a flasher for ammonia removal. A portion of the ammonia is
compressed and recycled to the reactor, and the remainder proceeds to
an absorber; water is used to absorb the ammonia, which then proceeds to
an ammonia stripper. The stripped ammonia is recycled to the reactor,
while the bottoms from the column qo to the sewer.
The liquid product stream from the flasher is sent to a toluene stripper.
Overhead toluene is recycled to the reactor, and the bottoms are recovered
for further purification.
The first step in refining consists of removal of HMDA (water azeotrope)
as bottoms from the first column operated at 18 psia. The overhead water-
imine mixture is discarded for fuel, and the bottoms are sent to the
medium boiler stills (concentrators) to recover HMDA. The bottoms stream
from the medium boiler column is then discharged to the HMDA refining
column. Refined HMDA is taken overhead under vacuum.
-59-
-------
DRAFT
The major water pollution sources of this process are the bottoms from
the ammonia recovery column, water withdrawn as overhead from the medium
boiler stills, and the steam jet condensate from the HMDA refining column.
Process RWL calculated from flow rate data and the analyses of wastewater
samples collected in the survey period are presented in the following
tabulat ion.
Plant 1 Plant 2
Sampling Period Sampling Period
£i £! ILB.H.
PROCESS FLOW
Liter/kkg 1,695 1,695 1,010 1,010 1,010
gal/M Ibs 203 203 121 121 121
BOD RWL
rig/liter1 58,850 12,800 5,500 4,500 1 ,800
kg/kkg2 99.8 21.7 5-55 k.^ 1.82
COD RWL
mg/Uter 71,850 62,1*00 20,200 22,200 20,300
kg/kkg2 122 106 20k 22. k 20.5
TOC RWL
mg/liter1 17,800 31,^00 4,600 5,700 5,000
kg/kkg2 30.2 53.2 4.65 5.76 5.05
Raw waste concentrations are based on unit weight of pollutant per
unit volume of contact process wastewaters.
2Raw waste loadings are based on unit weight of pollutant per 1000
unit weights of product.
The data shown in the above tabulation reveal that there are significant
differences in RWL from the plants surveyed. There are three major
reasons for these differences. Plant 2 partially recycles the bottoms
from the ammonia recovery column to the HMDA refining section and incin-
erates all the light-end organics from the refining portion of the
operation, while Plant 1 totally discharges such streams into the sewer
line. Another reason is the effective separation by the ammonia absorber
in Plant 2. Plant 1 is equipped with a stack which uses water to quench
the unabsorbed vapors, and discharges the quench water into the sewer.
If the above-mentioned reasons are taken into account, the data from
both plants become comparable.
An average RWL value calculated from the data obtained at Plant 1 is con-
sidered commensurate with BPCTCA. Again, it should be noted that the
process wastes from both Plant 1 and Plant 2 are actually disposed of by
deep-we11 injection. Therefore, the effluent limitations presented would
only be applicable if deep-wel1 injection were no longer feasible.
-60-
-------
DRAFT
The following paragraphs describe the pretreatment system used at Plant 2.
It should be noted that this system also handles wastes from adiponitrile
manufacture.
Aqueous waste from the hexamethylene diamine (HMDA) process is fed along
with by-product brine from other plant operations into a settling tank;
HMDA-waste comprises less than 10% of the total flow. Sludge which
settles out is pumped to a pit for further settling, while the main stream
goes through primary filtration in precoated, horizontal, leaf filters.
After a second filtration, the waste stream is pumped into injection
wells for disposal. Backwash from both filtration steps is pumped to
the sludge pits.
The primary purpose of the waste treatment area is to remove all solids
which might cause plugging of the injection wells.
Geological conditions in this area are favorable for deep-well injection,
and experience has been good for the past 20 years. These wastes are
being injected into a porous sandstone stratum at 3,700 - ^,200 feet
capped both top and bottom by practically impervious shale formations.
Non-contact cooling water usage for both plants is tabulated below:
Plant 1 - ^53,000 liters/kkg of product
Plant 2 - 31,500 liters/kkg of product
HMDA can also be produced by the ammonolysis of 1,6 hexanediol, a process
which wi11 be described in the following sub-section.
-61-
-------
DRAFT
z
o
a.
O
Z
o
Nwniuo
Q
LU
Z
II
t
sa3aaosav 1
EHN
t
EHN
Q^
LU
aX.
^
J
cr.
LLJ
CO
-a:
LJ_
1
H01VdVd3
i
k
,«1SVM
1
^ "
r 3
i i
1
s J
X
LU
X
H3ddiais
3N3mUl
_r
J
LU
C_3
e_3
LU
0=
U_l
LU
O
1
CO
%
4
LU NO
=>
CJ
I
CO
C_D
CD
or
i VH1
Q±
1
CC
^
d]d
i
NOIlVNBDOdQAH
3NIWI
-62-
-------
DRAFT
Product: Hexamethylenediamine (HMDA)
Process: Ammonolysis of 1,6 - Mexanediol
Process RWL Category: B
Chemical Reaction:
H2
HO(CH2)6OH + 2 NH3 > H2N (CH?)6MH? + 2 H20
Hexanediol Hexamethylenediamine
A typical process flow diagram of hexamethylenediamine production via
the ammonolysis of 1,6 - Hexanediol is shown in Figure 4-11.
The ammonolysis reactor is a flooded, fixed-bed, catalytic reactor. The
fresh and recycled hydrogen and liquid ammonia are pumped to the reactor
at a pressure of 3,000 psig. The liquid ammonia is preheated to a temper-
ature of 180 - 220 C and fed to the reactor bottom. At this temperature,
the ammonia is above its critical temperature (T = 132.6°C). The organic
feed stream, consisting of recycle hexamethyleneimine (HMI) (30 wt %)
plus fresh and recycle hexanediol is also pumped as a liquid into the
reactor. By means of two heat exchangers (one a product-feed exchange,
and the second a preheater), the organic liquid phase is raised to the
reaction temperature of 180 - 220°C (356-*»28°F) . Positive displacement
pumps are necessary to transfer both liquid NH- and the liquid organic
feed at the 3,000-psig pressure level.
l/ithin the reactor, the liquid organic phase floods the fixed-bed catalyst,
and the liquid product stream overflows through a discharge line at the
top of the bed. Both H_ and NH, enter the bottom of the fixed bed as
gases, but because of the high pressure and the solvent action of the HMI,
the NH- gas is absorbed in the HMI and subsequently reacts in the liquid
phase with the 1,6-hexanediol to produce HMDA. Since the concentration
of HMI in the liquid phase is already at its equilibrium value, further
conversion of the diol and NH, to the imine is negligible. A reflux
condenser is situated at the top of the reactor to condense DIOL-HMI-HMDA
carryover. The liquid condensate is returned to the top of the fixed
bed. The vapor stream leaving this condenser is used to pre-cool the
hydrogen gas prior to recycle to the compressor, and to condense NH? vapor
carryover from the reactor. This ammonia condensate is returned to the
reactor ammonia feed.
The ammonolysis reactor operates at a pressure of 2,800 to 3,300 psig
and at a temperature of 180 to 220°C. Catalysts suitable for the ammono-
lysis step include Raney cobalt, Raney nickel, reduced copper, Raney
copper, and nickel on kieselguhr. The preferred catalyst is pelletized
Raney nickel. At reaction conditions, an ammonia feed stream of 19 moles
-63-
-------
DRAFT
NH? per mole of diol (2.73 kg MH^/kg diol) and a hydrogen supply of
0.125 moles H2 per mole of MH3 (0.0147 kg H2/kg HH^) have been found
to give high yields. A yield of 93 mole °A HMDA on a diol-reacted basis
is practical, and conversions of 70 mole % per reactor pass are used
for design. A residence time of 1 hour is employed. The reactor mate-
rial of construction is normally 316 Stainless Steel and is fabricated
in multi-layer construction to meet the high-pressure rating. Catalyst
life amounts to about 100 kg HMDA produced per kg of catalyst.
The liquid reactor product consisting of HMDA, HMI , dissolved NH-^ and H2,
unreacted diol, by-product H20, and high boilers is heat, exchanged v/ith
the incoming HMI-diol feed stream and then fed to a series of three flash
drums. The flashing operation provides a stagewise reduction in pressure
and recovers the dissolved hydrogen as a vapor for refeecling to the com-
pressor. A small portion of the dissolved MH, is also vaporized, and
this is recovered in the knockout drums of the compressor facility. The
major part of the dissolved ammonia is recovered in the ammonia stripper
column. Ammonia is liquefied overhead by a water-cooled condenser, from
which it is pumped back to the HH3 feed tank.
The ammonia stripper column operates at about 200 psig, to condense the
ammonia overhead at 100°F. The bottoms temperature averages about 390°F.
The bottoms product (containing HMDA, HMI, diol, H20 and high boilers)
is fed to a drying column and then to the Water Stripper column. The
stripper column operates as an azeotropic column using cyclohexane as the
entraining agent to promote separation of the water. The use of cyclo-
hexane counteracts the formation of the H20/HMI azeotrope. (This azeo-
trope has been reported as a means of separating the HM1 from the HMDA.)
The overhead stream from this tower consists of the heterogeneous cyclo-
hexane-water azeotrope, which condenses as a two-phase liquid in the over-
head decanter. The cyclohexane upper layer is returned to the column as
a reflux, and the lower layer, mostly water, is sent to waste. This
column normally operates at atmospheric pressure, with an overhead tempera-
ture of about 165°F and a bottoms temperature of 212°-225°F.
The bottoms from this tower, essentially anhydrous, are fed to the HMI
stripper Column to remove HMI for recycling. This stripper column is
operated under vacuum of about 200 mm Hg. such that reboiler temperatures
in the range of 310~330°F are not exceeded. Temperatures of 390°F and
higher tend to promote reactions between the HMDA and the unreacted diol
to produce high boilers. The HMI Stripper bottoms are then fed to the
HMDA product column, where the HMDA is taken overhead at 99.9^ purity.
The bottoms, consisting of a small percentage of HMDA and the unreacted
diol plus high boilers, are recycled to the diol feed tank. Periodic
buildup of the high-boiler concentration will necessitate an intermittent
purge of the bottoms. Alternatively, a 1 ,6-hexanediol polishing still may
be added to provide for continuous separation of the diol from the high
boilers. In actual operation, the reactor conditions may possibly be
adjusted so that the high-boiler content is minimized to some equilibrium
value.
-6k-
-------
DRAFT
The major wastewater pollution sources for this process are wastewaters
from the drying column, the decanter associated with the azeotropic
column, and the steam jet condensate. The analytical results obtained
from the sampling survey are shown in the following tabulation.
PROCESS FLOV/
liter/kkg 1,100
gal/M Ib 132
BODr RWL
3,630
kg/kkg2 1».0
COD RWL
mg/liter1 10,600
kg/kkg2 11.7
TOC RWL
mg/liter1 2,260
kg/kkg2 2.5
1Raw waste concentrations are based on unit weight of pollutant
per unit volume of contact process wastewaters.
n
Raw waste loadings are based on unit weight of pollutant per
1000 unit weights of product.
It should be noted that most of the wastewater upon which the previous
RV/L calculations are based is disposed of via deep-well injection. Only
wastes such as slab washdown water and storm runoff are treated in an
aerated lagoon.
-65-
-------
DRAFT
Q
LLJ
z
<
x
LLJ
X
CO
co
- o
O
o
LLJ
Z
x
LU
X
-66-
-------
DRAFT
Product : Hal etc Anhydride
Process : Oxidation of Benzene
P roce s s RWL_Cat ego ry : B
Chemical Reactions:
V205
02 - > (CH)2-(CO)2-0 + 2 C02 +
2
Malei c
Benzene Anhydride
15 percent per year after having remained static for several years; the
growing use of polyester resins is the main reason for this rapid increase
in demand. Additional uses of maleic anhydride include pesticides, sur-
face coatings, plast icizers , and lubricants. Also, maleic anhydride is
used as a raw material for the production of fumeric and naleic acid.
Fumeric acid finds wide use as a food acidulant; additional applications
of fumeric acid resemble those of maleic anhydride. Maleic acid is
used solely as a food acidulant.
Essentially all maleic anhydride in the U.S. is based on the oxidation
of benzene. A process flow diagram of this oxidation process is shown
in Figure ^-12. As shown in the process flow diagram, the production of
maleic anhydride is achieved by a process consisting of:
1. A reaction section, where benzene is oxidized with air to form
maleic anhydride.
2. A recovery section, in which maleic anhydride is separated from
noncondens ibles.
3. A dehydration section, in which maleic acid is dehydrated to
maleic anhydride.
4 . A f ract ionat ion section, in which pure maleic anhydride is
produced .
Benzene is vaporized and fed to a fixed-bed reactor, to which compressed
air is also fed. Typical operating conditions are 25 psig and 750°F.
Conversion of benzene is essentially complete on a once-through basis.
Temperature control is achieved by circulation of a heat-transfer salt
through the shell side of the reactor, with indirect steam generation.
Vanadium pentoxide is used as the catalyst.
-67-
-------
DRAFT
Reactor off-gas Is cooled at a precondenser to condense as much of the
malelc anhydride as possible from the vapor. Condensed maleic anhydride
is piped to crude storage tanks prior to f ract ionat ion. The remainder
of maleic anhydride is recovered by scrubbing with water, forming maleic
acid. The acid is then dehydrated, thus forming maleic anhydride.
Xylene is added to the dehydrator to form an azeotropic mixture, thus
facilitating dehydration. The xylene is recovered from the excess water
by decantation.
The recovered maleic anhydride in the separator and dehydrator is fed to
a fract ionator , where pure maleic anhydride is produced. The light and
heavy ends are withdrawn from the dehydration and fract ionation columns
respectively, and are disposed of by incineration.
The major water pollution sources of this process are the excess water
withdrawn from the decanter and the periodic washings from the dehydrator,
fract ionator , and storage tanks. Process RWL calculated from the flow
measurements and analyses of wastewater samples obtained in the survey
periods are presented in the following tabulation:
Plant 1 Plant 2
PROCESS FLOW
liter/kkg 6,600 2,300
(gal/M lb) 788
BODc RWL
mg/liter1 63,500 ^7,000
kg/kkg2 lH8 108
COD RWL
mg/liter1 90,000 126,000
kg/kkg2 592 287
TOC RWL
mg/liter1 23,500 52,500
kg/kkg2 155 120
1
Raw waste concentration are based on unit weight of pollutant
per unit volume of contact process wastewaters.
2
Raw waste loadings are based on unit weight of pollutant per
1000 unit weight of product.
Based on the previous process description, the flow and raw waste loadings
for Plant 1 appear extremely high. This may be partially explained by
the fact that wastewaters from the dehydrator are presently discharged
rather than recycled to the scrubber. This flow is approximately 2/3 of
-68-
-------
DRAFT
the total flow. These wastewaters contain high concentrations of
fumeric acid, thus contributing to a high BOD, COD, and TOC. It is
anticipated that this stream will be recycled in the near future. At
the Plant 2 facility, the dehydrator water is recycled, and both the flow
and pollutant level are lower than that of Plant 1. An arithmetic average
of the RWL from Plant 1 and 2 was used for BPCTCA.
Wastewater from Plant 1 is combined with other process wastes from the
facility and is treated by the activated sludge process. The wastewater
is fed to the biological system at a slow, controlled rate because of its
high concentrations. The wastewater from Plant 2 is hauled away by a
contract disposal service.
-69-
-------
DRAFT
LJJ
LLJ
N
LU
CO
U-
o
z
CN r: ^
- ^
^ Q SS
_ i-^
2! 0 = =
J V^ =» 1
O LU
iE Q
0£
>-
T
Z
<
u
LU
1
<
^
1
LU
QC Q
UJ LLJ ! !
t ' LJ CD UJ C_D Q^
^ a ^* **" ^ r ,
«c u_i cc LU >-
C_D 1 CD LJ- I ZC
LU >_ j LU «r 3K >
CD xoo cc^«x:3
T - ' t =
-Ops > ' 1
p i ^ »m VNIII I HVH
1 "
0=.
=1
1
LU cr ^
t
Li_l
C3
, ._.. __ C..3 Q=
LLJ CT
. H u IV a V d3 S fc = ^ = ^ ^
CJ ^ *n I
&
T ^^
oo
LU
^F**^ CJ
1 l_l_i
1 CC
^ 3N3ZN38
4
CH
LU
CO
UJ CO
Z LU
UJ CC
r-g Q_
21 gr Q^
LU C3
* c
J =
- Ll
r
i
\
-70-
-------
DRAFT
Product: Methyl Chloride
Process: Esterification of Methanol with Hydrochloric Acid
Process RWL Category: B
Chemical Reactions:
CH3OH + HC1 Metal Catal^ ^ + ^
Methanol Hydrochloric Acid Methyl Chloride
Methyl Chloride is used primarily as an intermediate during the production
of other chemicals. Primary end-products include silicones, tetra methyl
lead, and cellulose ethers. Chlorinated solvents such as carbon tetra-
chloride can also be manufactured, using methyl chloride as a raw feed-
stock. Methyl chloride is also used as a catalyst solvent during the
production of butyl rubber.
A process flow diagram for the production of methyl chloride by esterifi-
cation of methanol with hydrochloric acid is shown in Figure ^-13-
Methanol and hydrochloric acid are heated and then combined in the presence
of ZnCl£ in the reactor. The crude product is discharged into a frac-
tlonator from which the catalyst stream is recycled back to the reactor.
The vapor phase from the fractionator is then passed through a series of
scrubbing units. Water and caustic solution are used to scrub the product
vapor, which is then sent to a condenser. Finally, the product is treated
with highly concentrated sulfuric acid to remove the water from the product
stream.
The major pollution sources of this process are the waters discharged
from various scrubbing units and the concentrated sulfuric acid stream
from the drying unit. Process RWL calculated from the flow measurements
and the analyses of wastewater samples obtained in the survey periods are
presented in the following tabulation:
-71-
-------
DRAFT
Sample
Period #1
Plant 1 Plant 2*
Sample Sample
Period #2 Period #1
PROCESS FLOW
liter/kkg
(gal/M Ib)
BOD5 RWL 1
mg/1iter
kg/kkg2
COD RWL
mg/1iter1
kg/kkg2
TOC RWL
mg/1iter1
kg/kkg
1
583
69-9
1,210
0.703
119,700
69.8
29,000
16.9
583
69.9
1.13
112,800
65.8
31,600
18.4
12,000
1,430
1,480
17-7
5,240
62.7
1,070
12.8
Plant 3
Sample
Period #1
842
101
371
0.314
4,090
3.45
1,080
0.908
Raw waste concentrations are based on unit weight of pollutant per
unit volume of contact process wastewater.
2
Raw waste loadings are based on unit weight of pollutant per 1000
unit weights of product.
*BPCTCA
In the foregoing examples, a considerable difference in flow is evident.
This may be explained partially by the fact that Plants 1 and 2 utilize
scrubbers for removal of hydrochloric acid while Plant 3 employs a freez-
ing step. Wastewaters common to both installations include pump leakages
and washdowns.
The analytical results also indicate that, in addition to the parameters
shown in the tabulation, water pollution parameters such as pH, sulfate,
chloride, and zinc are at levels hazardous to biological treatment pro-
cesses; the low BOD^ values shown in the tabulation for Plant 1 are the
results of these biological inhibiting factors. The differences in pol-
lutant loadings among three facilities visited in the survey period are
attributed to differences in operating efficiencies in the scrubbing and
drying units. The values represented by Plant 3 can be considered as
BATEA control technology. The RWL in Plant 2 is considered BPCTCA.
The final disposition of process wastewaters for the three plants sur-
veyed is indicated below:
Plant 1 - deep-well injection
Plant 2 - municipal treatment
Plant 3 ~ neutralization and discharge to
surface waters.
-72-
-------
DRAFT
An alternative method for producing methyl chloride is by direct chlorina-
tion of methane. Despite the ample availability of cheaper methane,
approximately 65% of all methyl chloride is produced in the U.S. from
methanol. Part of the reason lies in the economics of chlorine utiliza-
tion .
-73-
-------
DRAFT
Q
U
y
of.
O
_j
U
O
at
O
O
Z
<
^ u_
o o
U- H-
u
uu
I
to
UJ
I
UJ
g
a:
O
_i
X
U
X
I
^
L
I
GO LLJ
=3 I
OC 00
" »-
00 CO
J
^
LlJ
CC LU
0. 31
-Ik-
-------
DRAFT
Product: Methyl Ethyl Ketone (MEK)
Process: Dehydrogenation of Secondary Butyl Alcohol (SBOH)
Process RWL Category: B
Chen-ncal Reactions:
CH3CH(OH)CH2CH > CH^COC^CH^ + HZ
Sec-butyl Alcohol methyl ethyl ketone
A flow diagram for this process is shown in Figure 4-14. Secondary butyl
alcohol is passed through a reactor containing a catalyst bed of zinc
oxide or brass (zinc-copper alloy), maintained at 400 to 550°C. The
vapor-phase dehydrogenation reaction takes place at near atmospheric
pressure, with MEK, by-product hydrogen, and unreacted secondary butyl
alcohol subsequently separated. This process is analogous to the produc-
tion of acetone via the dehydrogenation of isopropyl alcohol . Both are
considered typical of Category B vapor-phase processes.
The MEK process, however, is complicated by the fact that the secondary
butyl alcohol feed is a minimum boiling point azeotropic binary mixture
containing approximately 70 wt. % alcohol and 30 wt. % water. The
presence of this water in the feed complicates the process because of
the azeotropes it forms with both the feed alcohol and product MEK.
As shown in Figure 4-14, the crude secondary butyl alcohol is fed to the
azeotrope column, where light hydrocarbons are taken overhead and heavy
organics are discharged as bottoms. These heavy hydrocarbons are usable
by-product material and are used in other processing. Relatively concentrated
streams of secondary butyl alcohol and water are taken off as side streams
in this distillation.
The alcohol stream is used first to scrub the hydrogen gas produced in the
reactor to remove unreacted alcohol , and then is sent from the alcohol
scrubber to the conversion furnaces and reactors for dehydrogenation.
A condensed product stream containing MEK and unreacted alcohol is taken
from the reactor and sent to two distillation columns. The first dehydrates
the product mix by taking water and light hydrocarbons overhead. Product
MEK is separated from recycle alcohol in the second column (MEK column
in Figure 4-14).
Figure 14-4 illustrates the extensive reuse of water observed at one of
the two process plants surveyed. Fresh water (in addition to that present
in the SBOH feed) is added in the azeotrope column to facilitate separation,
and is also added to the hydrogen scrubber. The hydrogen scrubber recovers
-75-
-------
DRAFT
additional SBOH from the gas leaving the alcohol scrubber. The hydrogen
gas leaving the hydrogen gas scrubber is burned as fuel.
The aqueous bottoms from the hydrogen scrubber are combined with water
cuts taken from the azeotrope and dehydration columns, and the combined
aqueous stream is sent to the butyl water column. Light hydrocarbons
are taken overhead in this column, and part of the aqueous bottoms is
discharged as wastewater while the remainder is reused in the light-ends
scrubber which recovers additional SBOH from the light material distilled
in the azeotrope column. The aqueous bottoms from the light-ends scrubber
are sent to the dehydration column for recovery of SBOH.
The following tabulation summarizes the raw waste load data obtained from
the two plants sampled. The RWL shown for Plant 2 is much higher because
some of the light hydrocarbons purged (used elsewhere or burned) in
Plant 1 are discharged in the wastewater from Plant 2.
Plant 1 Plant 2
PROCESS FLOW
liters/kkg 1,310 795
gal/M Jb 157 95
BODr RWL
mg/literl 3,000 91,000
kg/kkg2 3.92 72.1
COD RWL
mg/liter1 1,627 260,000
kg/kkg2 2.13 206
TOC RWL
mg/liter1 521 102,000
kg/kkg2 0.68 80.9
Raw waste concentrations are based on unit weight of pollutant
per unit volume of contact processed wastewaters.
2Raw waste loadings are based on unit weight of pollutants per
1000 unit weights of product.
An average of the RWL values presented above was considered as BPCTCA.
In both plants, the MEK process is part of a large chemical complex.
Process wastes from MEK are combined with large volumes of other wastes
prior to treatment and discharge to surface waters.
-76-
-------
DRAFT
kJ -
CO
CO
0
X
O
u
_i
<
>-
H-
Z>
oo
U
ii i
CO
U.
O
7.
, __
^7 t
^r <
UJ Z
£ £
» t "I
J w
O O
= a
>
I
UJ
O
sz
UJ
^
UJ
Z
s~~
UJ
^
i
>~
L».
METHYL E'
4
1
= S H3fl8n«3S z
- = ' S1H3 1 -
_._ :=
< t C3
O
or
Q_
4 i
£2... ^ NWniOG .
S" NWP10G ^* >I3W
_, = H3ivi TAina ^ y k
£
LU
^
T s
jj °°
s ^2^ Nwmoo
- i^4 NUIll/dQAHBO
35 ^ .. ,
I . JO. ,
<= - H3aanaos
oj =j NdUOdUAH *
T
I
cc
^ ^
LU
Q±
^1 _..-. ^^
^ , . C^
QQ
OO
UJ
1 1
, aqannMn?; ^
10H031V ~* =
T,
. * >
' t I -
__^__ Mwnim n
J Nnll IUJ _w ^J
3dOdl03ZV *£ =
J 4
cc
UJ 3C
i o
cc en
^ CO
-77-
-------
DRAFT
Product: Perchloroethylene
Process; Chlorinatfon of chlorinated hydrocarbons
Process RWL Category: B
Chemical Reactions:
Chlorinated + ri n r r n
... . + L I o :> CI0 C = C C10
Hydrocarbons z 7 L <-
Chlorine Perchloroethylene
Perchloroethylene is used largely in dry-cleaning and vapor degreasing.
Dry-cleaning consumes approximately 85 percent of the total; the rest
going into general solvent services and as an Intermediate for fluoro-
carbons.
A process flow diagram for the manufacture of perchloroethylene is shown
in Figure 4-15.
The chlorinated hydrocarbons and chlorine gas are reacted thermally
(rather than catalytically) at temperatures in the range of 500~700°C.
The reactions are highly exothermic, and control of heat transfer is a
key to smooth and efficient reactor performance.
The reaction mechanism may be visualized in terms of an initial, very
rapid, thermally activated, free-radical chlorination to transient com-
pounds having high chlorine/carbon ratios. These transient compounds
decay quickly, primarily to the low free-energy forms. The perchloro-
ethylene/carbon tetrachlorJde mixture then shifts toward equilibrium in
the rate-controlling step.
The reactor effluent is quenched, and the gas stream leaving the quench
vessel is absorbed with recycled HC1 solution in an absorption tower.
The remaining unabsorbed chlorine gas is then dehydrated and recycled
back to the reactor.
The liquid reaction product stream goes from the quench vessel to the
separator, where the aqueous stream is removed as HC1 solution. The
product mixture (carbon tetrachlorlde and perchloroethylene) is separated
by distillation. The perchloroethylene/carbon tetrachloride mixture is
controlled by suitable recycle and by provision of appropriate residence
time In the reactor.
The wastewater pollution sources of this process are pump-seal leakages
and miscellaneous reactor washdowns. Process RWL calculated from flow
measurements and analyses of the wastewater streams are shown in the
following tabulation.
-78-
-------
DRAFT
Sample Period #1 Sample Period #2
PROCESS FLOW
liter/kkg 5,*»00 5,^00
gal/M 1b 6*»3 ^J>
BODc RWL
mg/1iter1 83 79
kg/kkg2 0.^9 0.^27
COD RWL
mg/Uter1 357 695
kg/kkg2 1.92 3.73
TOC RWL
mg/liter1 30 31
kg/kkg2 0.16^4 0.169
waste concentrations are based on unit weight of pollutant
per unit volume of contact processed wastewaters.
2Raw waste loadings are based on unit weight of pollutant per
1000 unit weights of product.
An average of the foregoing values was considered as BPCTCA. The wastes
from this process are disposed of by deep-well injection.
An alternative method for the manufacture of perchloroethylene is by the
Detrex-SD process, in which the feedstocks (ethylene and chlorine) enter
a liquid-phase chlorination reactor. Temperature and pressure are moder-
ate, and concentrations are carefully controlled by recycle quantity and
composition to give the desired product distribution at optimum economics.
In the classical route, perchloroethylene is prepared from acetylene, via
trichloroethylene. The chemical reaction is shown as follows:
-HC 1
C2H2 + 2C12 - >C12CHCHC12 - >C1CH - CC12
Acetylene Tetrachloroethane Trichloroethylene
C 1 o -u
>C12C = CC12
Pen tachloroe thane Perchloroethylene
The chlorine and acetylene are brought into contact with each other in a
reactor at a temperature of 250-300°C in the presence of barium chloride
deposited on carbon as catalyst. The product, tetrachloroethane, is then
-79-
-------
DRAFT
dehydrochlorinated in a catalytic reactor to produce trichloroethylene,
which is chlorinated at 80-90°C over a catalyst containing 0.2-0.3% FeCl
to yield pentachloroethene. The perchloroethylene is then obtained by
the dehydrochlorination of pentachloroethane by milk of lime at 110°C and
200 mm Hg.
-80-
-------
DRAFT
*o
1
-t
O
CO
<
U
O
Of
O
Q
LU
<
Z
o
X
o
UJ 21
^ o
O i=
x
u
UJ
z
LLJ
O
I
u
-81-
-------
DRAFT
Product: Phthal ic Anhydride
Process; Oxidation of o-Xylene
Process RWL Category: B
Chemical Reactions:
2 + 3°2 - >C6HA (c°)2° + 3H2°
o-xylene phthal ic
anhydride
Phthal ic anhydride is commonly produced by either of two methods: oxida-
tion of o-xylene, or oxidation of naphthalene. In the U.S., about 80%
of the capacity is still based on napthalene, although there is a definite
trend in favor of the vapor-phase oxidation of o-xylene. This trend is
based primarily on economics (o-xylene feedstock is cheaper than naph-
thalene). A description of the production of phthal ic anhydride using a
naphthalene feedstock is found in the following section.
Phthal ic anhydride has become one of our most important intermediates.
It is commonly used during the production of plast icizers. The less vola-
tile phthalates are used principally in wire- and cable-coatings which
are subject to higher temperatures. Phthal ic anhydride derivatives are
also used in the production of alkyl resins. These resins are in turn
used in coatings, such as latex paints, thermosetting acrylic finishes,
and epoxy coatings. Phthal ic anhydride is usad directly in making a num-
ber of dyes such as eosin, quinoline yellow, phenolphthalein , and copper
phthalocyanine. It is also used during the product 'on of anthraquinone
and anthraquinone derivatives by condensation (Friedel Crafts) procedures
Production of phthal ic anhydride by oxidation of o-xylene is based on
vapor-phase, fixed-bed technology. Typical operating conditions are 5
psig and 700°F. The process uses a carrier-supported vanadium pentoxide
catalyst, which normally lasts from 3 to 5 years. The crude product ob-
tained is 99-99.5$ phthal ic anhydride, with some maleic, benzoic, and
other acids.
An ortho-xylene oxidation process flow diagram is shown in Figure A-16.
Filtered air is first compressed and preheated in a heat exchanger. The
o-xylene feedstock is also preheated and vaporized by injection into the
hot-air stream; unevaporated o-xylene is trapped before the stream enters
the reactor.
During the reaction step, a considerable quantity of heat is generated.
The heat is removed by molten salt circulating on the shell side of the
reactor. The molten-salt solution is passed from the top of the reactor
to a heat-exchange system, where process steam is produced. Gases leav-
ing the reactor at 375°C are passed through, a waste-heat boiler for addi-
tional steam generation. Cooled gases enter a bank of automatically con-
trolled switch condensers.
-82-
-------
DRAFT
When a condenser is on the crystallization cycle, cold mineral oil is
circulated through the coils to cool and crystallize phthalic anhydride
from the gaseous phase. When the condenser is switched to phthalic an-
hydride recovery, hot oil is circulated through the coils to remelt the
crude product. The crude product is then drawn off to an intermediate
storage tank. Residual gases from the condensers are scrubbed to reduce
the volatile organic content. The scrubbed gas passes through a demister
and is vented to the atmosphere.
Crude product from the storage tank is then passed through a heater and
into the continuous pretreatment section. Dissolved phthalic acid is
dehydrated under a slight vacuum to the anhydride. Additives may be
introduced at this point to remove impurities produced by polycondensa-
tion of heat-sensitive compounds. The crude phthalic anhydride is then
pumped through a pre-cooler to a continuous distillation section. Two
columns are employed, both operating under vacuums created by two-stage
steam ejectors; wastewaters from these ejectors are sent to an incinerator.
In the first-stage stripper column, maleic anhydride and benzoic acid are
separated as overheads. Bottoms from the first-stage column are passed
to a second-stage rectifier column. Final product (99.993 phthalic an-
hydride) is withdrawn as distillate overhead.
Since waters from stream ejectors are disposed of by incineration, the
only water pollution source of the process is the waste stream withdrawn
from the vent gas scrubber. Process RWL calculated from the flow measure-
ments and analyses of water samples obtained during the survey period are
shown in the following tabulation:
PROCESS FLOW
liter/kkg 593
(gal/M Ib) 71.2
BODr RWL
mg/liter1 215
kg/kkg2 0.128
COD RWL
mg/liter1 1,080
kg/kkg2 0.6^2
TOC RWL
mg/lfter1 34
kg/kkg2 0.02
^Raw waste concentrations are based on unit weight of pollutant
per unit volume of contact processed wastewaters.
o
^Raw waste loadings are based on unit weight of pollutant per
1000 unit weights of product.
-83-
-------
DRAFT
These waste loads are considered as BPCTCA. They are combined with other
wastes In this plant and treated in a biological system prior to discharge.
-------
DRAFT
X
I
O
X
O
o 9
^ s
z
u
I
X
X
o.
n
L__
|__^
n
I
J.
I «r co
LU LU CO
T
X
I
-------
DRAFT
Product: Phthalic Anhydride
Process: Oxidation of Naphthalene
Process RWL Category: B
Chemical Reactions;
C1QH8 + H02 - - >C6Hj,(CO)20 + 2C02 + 2H20
Naphthalene Phthalic Anhydride
In the United States, approximately 80% of the present phthal ic anhydride
capacity is based upon the oxidation of naphthalene. However, there is
a trend toward the vapor-phase oxidation of o-xylene, which has been dis-
cussed in the previous section. Phthalic anhydride may be produced from
naphthalene using either a fixed or fluid catalyst bed. In addition to
process variations, the purity of the final product is a function of the
reactor type, in that maleic anhydride is formed as a by-product in the
fixed-bed reactor but it is not formed in the f luidized-bed reactor.
During the sampling period, an installation employing a f luidized-bed
reactor was visited; the fluid catalyst consisted of a finely powdered
vanadium catalyst.
A process flow diagram for the naphthalene oxidation process is shown in
Figure 4-17. The reactor containing the catalyst is heated to an operat-
ing temperature of approximately 900°F. Molten naphthalene is then intro-
duced into the reactor and vaporized by direct contact with the catalyst
charge. The vapors become admixed immediately because of the agitated
nature of the catalyst bed. The air-naphthalene vapor mixture passes up-
ward through the bed, and the naphthalene is converted to phthal ic anhy-
dride, carbon dioxide, carbon monoxide, and water vapor. The product
gases, after leaving the dense catalyst phase, pass through a settling
zone and into a cyclone system for removal of the catalyst. Recovery of
the catalyst is reported to be 100%; thus, make-up catalyst is not re-
quired. Following removal of the catalyst, the product gases pass through
a condensing system. Aqueous products are then purified, using a series
of distillation columns.
There are two major discharges from the phthal ic anhydride process area.
The most significant of these discharges is a gaseous waste stream vented
during the reaction step. The gas contains a high concentration of or-
ganics and is discharged to an incinerator.
The second discharge is primarily steam vapor from a vacuum jet in the
distillation section. The stream is contaminated with a low concentration
of phthal tc anhydride. This stream is also discharged to the incinerator.
-86-
-------
DRAFT
Since the only source of wastewater is periodic process washings, the
raw waste load approaches zero. Although it is recognized that pollu-
tant loadings from the washings contribute to the raw waste load, it is
not possible to obtain representative samples of the wastewaters. Thus,
equipment washings have not been Included in the raw waste evaluations.
Non-contact wastewaters associated with phthalic anhydride include an
involuntary blowdown from the internal tempered water system in the
crude-product condensing step. In the condensing step, there is a swing
between the use of steam and tempered water. During the swing, the
tempered water condenses the trapped steam and is discharged. The only
pollutants are corrosion inhibitors which are in the initial water.
-87-
-------
DRAFT
CO
LLJ
<
z
K 2
o
o
at
O
X
z
u
X
_i o-
LA
suns
NQI1VGN IdHd
-------
DRAFT
Product; Tricresyl Phosphate (TCP)
Process: Condensation of Cresol and Phosphorus Oxychlorlde
Process RWL Category: B
Chemical Reaction;
3 C6HIt(CHjOH + POC13 > (C6Hi,(CH3)0)3PO + 3HC1
cresol (mixture of o-, phosphorus tricresyl
m-, and p-lsomers) oxychloride phosphate
TCP has the property of reducing the flammabi1ity of films. This property
has led to its use as a plasticizer for nitrocelluous and vinyl chloride
plastics. This flammabi1ity-reducing property qualifies it for use as an
additive in hydraulic fluids and lubricants. TCP is also used as a gaso-
1ine addi tive.
Figure 4-18 presents a simplified process flow diagram of tricresyl phos-
phate production. Basically, the processing involves purification and
then condensation of the raw materials, preliminary purification, and
final purification of the product.
Cresol supply has been a problem to manufacturers of tricresyl phosphate
for a number of years, both as to availability and quality. Until the
post-war period almost all cresol for tricresyl phosphate manufacture was
derived from coal tar acids. Most recently, improved processing of carbo-
late liquids from petroleum sources has resulted in cresols essentially
equivalent to those produced from coal tar acids. Furthermore, the short-
age of cresols which has long plagued tricresyl phosphate producers and
which, probably more than any other factor, had led to commercialization
of cresyl diphenyl phosphate in the post-war period, appears to have been
eliminated. Since the ortho form of tri-cresyl phosphate is considerably
more toxic than those derived from m- and p-cresols, the cresol used for
the production of plasticizer-grade tricresyl phosphate is primarily a
mixture of m- and p-cresol. Although the xylenol content may be allowed
to reach as high as 30 or kO%, the o-cresol content is held below 3%.
Because of the random distribution of the three isomers in the resulting
phosphate ester, the tri o-cresyl phosphate content of the product is
quite low. The cresols used for production of tricresyl phosphate for
addition to gasoline may contain a slightly higher o-cresol content and
a higher percentage of xylenols.
Cresols react rather readily with phosphorus oxychloride at a temperature
of approximately 100°C. to form a mixture of aryl and diaryl phosphoryl
chlorides, with only small amounts of triaryl phosphate. The diaryl
phosphoryl chloride being the least reactive, more drastic conditions are
-89-
-------
DRAFT
required to obtain complete reaction. The presence of significant quanti-
ties of aryl phosphoryl chlorides not only reduces yield but leads to
difficulty in subsequent refining operations. Therefore, the condensation
of cresol and the oxychloride is carried out at elevated temperatures
(150° to 300°C.) depending upon catalyst, and the purification scheme
employed. A slight excess of cresol favors complete esterification. The
time required for the condensation will vary with the catalyst and tempera-
ture of reaction. Loss of oxychloride in the hydrogen chloride off-gas
is minimized by operating under moderate pressure and/or venting through
a condenser. Many catalysts have been reported, but the metal halides
appear to be preferred, becuase they permit condensation times of 6-9
hours at temperatures of approximately 200°C. Because the reaction mix-
ture is highly corrosive, glass-lined or alloy kettles are used. The
condensation may be operated continuously, by permitting the reaction
mixture to pass through a series of reactors at successively higher temp-
eratures.
The purification techniques employed appear to have become rather well
standardized. Variation lies rather in the sequence of application, the
use of classical batch washing in lieu of the use of columns, and the ex-
tent of purification required for product end-use. Preliminary purifica-
tion may involve direct flash distillation of the crude reaction mixture,
or the crude reaction product may first be washed with dilute caustic to
neutralize any hydrogen chloride and to hydrolyze and extract traces of
partial esterification products and unreacted cresylic compounds. The
addition of lime to the reaction mixture prior to distillation, to mini-
mize corrosion, has been reported. Final purification of plasticizer-
grade products employs washing with dilute caustic and water (to remove
traces of organic acidity), treatment with dilute permanganate solution
(to improve color and oxidation stability of the product, a widely accepted
quality factor), dehydration by heating under reduced pressure, bleaching
with activated carbon and/or earth, and finally filtration. The use of
an amphoteric metal in conjunction with an alkaline wash has also been
claimed as a col or-improvement refining step. In production of tricresyl
phosphate as a gasoline additive, some or all of the final purification
steps may be omitted; low acidity is required, but color and oxidation
stability are not critical.
As indicated in the process flow diagram, each step of the process is per-
formed under vacuum conditions, and each step is equipped with steam jets
and barometric condensers. The major pollution sources are the waste
streams from those barometric condensers. The analytical results from
the sampling program are presented in the following tabulation.
-90-
-------
DRAFT
PROCESS FLOW
liter/kkg 28,000
gal/M Ibs 3,355
BOD
mg/1iter1 kO
kg/kkg2 1.12
COD
mg/liter1 *»08
kg/kkg2 11. ^
TOC
mg/liter1 70
kg/kkg2 1.96
Raw waste concentrations are based on unit weight of pollutant
per unit volume of contact process wastewaters.
2
Raw waste loadings are based on unit weight of pollutant per
one thousand unit weights of product.
By using surface condensers it is possible to reduce the process flow
requirements, although the process RWL will remain at the same level.
Non-contact wastewaters include cooling water flows and steam condensate;
the cooling water usage is approximately 7^0 kg per kg of product, while
condensate flow to the sewer is at the rate of 0.96 kg per kg of product.
The process RWL shown above are considered as BPCTCA. All wastes from
the plant are discharged to the municipal sewer system.
-91-
-------
DRAFT
U_ LLJ
o 9
_ Of.
Z o
o ::
$5
/o >-
z *
£ O
/N
5 >
_. ^r -^N
00 ^- -J
"TOO
"* 0 C2
j LU cc -^
*S C=J
LU GQ C_D
t t
~-E t.^
«X LU
LU h
CO ^
1"
; CD
Z C_D
*t or
Q^ 1 1 t
ae i co
C3 LU ry
CJ C3 CD
1 i or: -7-
« c=
LU OQ C_3
T t
£ CC-
«X UJ
UJ 1
I -r
z
^
^ cc
rr ^i i
or » co
iCS LU ^
1 K LU
CJ C3 CZ
LU or ^
> -a: CD
tl
S cc
«r LU
LU 1
1 *c
co =e
^ CO
^ cc
0= LU
C= LU 3;
| £ LU
CJ CD CD
LU eg ^
. -I CD
LU CO CO
CC
LU
1
£
LU
1
CO
«r
, f
M
CC
LU
1
CE
^
LJ_I
CO
-er
^
CC
LU
-a:
u_
I
CO
«c
%
A
uf
CC
LU
LU
CO
%
c:
cz
=
cz
3
}
f
CC
C
1
«a
CC
-a
a.
u_
e^
J
j
^^
J
LJ>
!
LU zr:
cc ex.
CO CO
CD
or D=
1 CL.
:s
«^
LU
1
CO
1
. I
j 1
NWII103 j
UNIdd IdlS '
j
1
NWniOG
3 N 1 N 1 J ] a '
a Jl i j
j
T
CD
1
CJ
«=C
LU
CC.
f 1 '
J I
^D CO 1
0= C=J 0-.
Q_ CD -*
C^ Q_ t
CD J£ «,;
^= CD CO
1
i
Noiivniisia
1
n
-92-
-------
DRAFT
Product: Acetone
Process : OehydrogenatIon of Isopropyl Alcohol
Process RWL Category: B
Chemical Reaction:
CH-CHOHCH, > CH3COCH3 + H2
j j
Isopropanol Acetone Hydrogen
This product/process was investigated in Phase 1 of this study, and the
detailed process description and process flow diagram can be found in EPA's
document, EPA 440/1-73/009 (Development Document for Proposed Effluent
Limitations Guidelines and New Source Performance Standards for the Major
Organic Products, Segment of the Organic Chemicals Manufacturing Point
Source Category) published December, 1973.
Process RWL calculated from flow measurements and analyses of wastewater
samples obtained in Phase 2 survey period are shown in the tabulation
be1ow:
Plant 1 Plant 2
PROCESS FLOW
liter/kkg 2,300
gal/M Ib 276 5^0
BODc RWL
mg/liter1 500 2,530
kg/kkg2 1.15 11.A
COD RWL
mg/liter1 8,820 5,150
kg/kkg2 20.3 23.2
TOC RWL
mg/liter1 2,400 3,150
kg/kkg2 5.53 14.2
Raw waste concentrations are based on unit weight of pollutant
per unit volume of contact process wastewaters.
2
Raw waste loadings are based on unit weight of pollutant per
1000 unit weights of product.
-93-
-------
DRAFT
Product : Ethylene Dlchloride
Process: Combination of Oxychlor (nation and Direct Chlorination of
Ethylene
Process RWL Category: B
Chemical Reactions;
Direct chlorination: C^ + C12 - >
Ethylene Ethylene Dichloride
Oxych 1 or i nation: C2Hj, + 2HC1 + i02 - >C2H^C124- FLO
Ethylene Ethylene
dichloride
This product/process was investigated in Phase 1 of this study, and the
detailed process description and process flow diagram can be found in
EPA's document, EPA 440/1-73/009 (Development Document for Proposed Ef-
fluent Limitations Guidelines and New Source Performance Standards for the
Major Organic Products, Segment of the Organic Chemicals Manufacturing
Point Source Category) published December, 1973-
Process RWL calculated from flow measurements and analyses of wastewater
samples obtained in Phase 2 survey period are shown In the tabulation
below:
Sample Sample
Period #1 Period #2
PROCESS FLOW
liter/kkg
gal/M Ib 28.2 28.2
BODc RWL
mg/liter1 4,200 6,200
kg/kkg2 0.988 1.46
COD RWL
mg/liter1 14,100 12,900
kg/kkg2 3-32 3.05
TOC RWL
mg/liter1 6,300 6,200
kg/kkg2 1.48 1.46
1
Raw waste concentrations are based on unit weight of pollutant
per unit volume of contact process wastewaters.
1Raw waste loadings are based on unit weight of pollutant per
1000 unit weights of product.
-94-
-------
DRAFT
Product; Styrene
Process: Dehydrogenation of Ethyl Benzene
P rpee s s RV/L Cat ego r y; B
Chemical Reactions:
C H C H
6525
ethyl benzene
C H C H +
6523
Styrene
This product/process was investigated in Phase I of this, study and the
detailed process description and process flow diagram can be found in
EPA's document, EPA MO/1-73/009 (Development Document for Proposed Ef-
fluent Limitations Guidelines and New Source Performance Standards for
the Major Organic Products, Segment of the Organic Chemicals Manufacturing
Point Source Category) published December, 1973.
Process RWL calculated from flow measurements and analyses of wastewater
samples obtained in Phase 2 survey period are shown in the following
tabulation:
PROCESS FLOW
Hter/kkg
gal/M Ib
BOD5 RWL
mg/1iter1
kg/kkg2
COD RWL
mg/1iter1
kg/kkg2
TOC RWL
mg/Uteri
kg/kkg2
Plant 1
Plant 2
Sample
Period #1
k, 100
85
0.35
380
1 .56
80
0.31
Sample
Period #1
5,970
715
0.29
325
1.93
56
0.33
Sample
Period #2
5,970
715
0.25
370
2.19
20
0.12
Sample
Period #3
5,970
715
220
1.31
525
100
0.60
'Raw waste concentrations are based on unit weight of pollutant per unit
volume of contact process wastewaters.
2Raw waste loadings are based on unit weight of pollutant per 1000 unit
weights of product.
-95-
-------
DRAFT
Both of the production facilities visited during the sampling period have
initiated various pollution abatement programs aimed at achieving BATEA
technology. Process modifications Include:
1. Condensing the steam used to pull the vacuums in the distil-
lation section and piping the wastewater to a separator.
2. Steam Stripping of wastewaters from the separator to remove
organics, followed by filtration and recycle as boiler feed
water.
Following these modifications, the resulting raw waste loads are as follows:
Plant 1 Plant 2
PROCESS FLOW
liter/kkg
gal/M Ib
BOD5 RWL
mg/1iter1
kg/kkg2
COD RWL
mg/1iter1
kg/kkg2
TOC RWL
mg/1Iter1
kg/kkg2
iRaw waste concentrations are based on unit weight of pollutant per unit
volume of contact process wastewaters.
Raw waste loadings are based on unit weight of pollutant per 1000 unit
weights of product.
Sample
Period #1
7*»
8.9
7
0.001
78
0.006
4
0.0005
Sample
Period #1
6,260
750
6
O.OA
59
0.2k
13
0.08
Sample
Period #2
6,260
750
16
0.10
50
0.31
16
0.1
Sample
Period #3
6,260
750
10
0.07
30
0.19
11
0.0?
-96-
-------
DRAFT
Product: Vinyl Chloride
Process: Thermal Cracking of Ethylene Dichloride
Process RWL Category; B
Chemical Reaction:
CjfyCl > C2H3C1 + HC1
Ethylene Vinyl
Dichloride Chloride
This product/process was investigated in Phase 1 of this study, and the
detailed process description and process flow diagram can be found in
EPA's document, EPA MO/1-73/009 (Development Document for Proposed Ef-
fluent Limitations Guidelines and New Source Performance Standards for
the Major Organic Products, Segment of the Organic Chemicals Manufacturing
Point Source Category) published December, 1973-
Process RWL calculated from flow measurements and analyses of wastewater
samples obtained in Phase 2 survey period are shown in the following
tabulation:
PROCESS FLOW
Uter/kkg
gal/M Ib 17
BODc RWL
mg/liter1 105
kg/kkg2 0.015
COD RWL
mg/liter1 8H»
kg/kkg2 0.116
TOC RWL
mg/liter1 50
kg/kkg2 0.007
'Raw waste concentrations are based on unit weight of pollutant per unit
volume of contact process wastewaters.
2Raw waste loadings are based on unit weight of pollutant per 1000 unit
weights of product.
-97-
-------
DRAFT
Product; Acetic Esters (Ethyl Acetate, Propyl Acetate, Butyl Acetate)
Process: Batch or continuous esterifIcation of appropriate alcohol with
Acetic Acid catalyzed by Aqueous Sulfuric Acid
Process RWL Category: C (if continuous), D (if batch).
Chemical Reactions;
H2SO/,
RON + CH^COOH > CHjCOOR + H20
alcohol acetic acid acetic ester
where
R is Ethyl (CH3CH2-),
Propyl (CH3CH2CH2-),
or Butyl (CH3CH2CH2CH2-)
Typjca1 Haterial Requ i rements:
Basis 1000 kg Ethyl Acetate 1000 kg Butyl Acetate
Ethyl Alcohol (95$) 620 kg
Butyl Alcohol -- 713 kg
Acetic Acid (100%) 688 kg 550 kg
Sulfuric Acid (66°Be') 15 to 150 kg 2,708 kg
Acetic acid esters are used mainly as solvents; the shorter the alcohol,
the faster-boiling the solvent. Ethyl acetate is a low-boiling solvent
for lacquers, and the various propyl and butyl acetates are popular medium-
boiling solvents used mainly in surface coatings. In all cases, acetic
acid is esterified with ethyl, propyl, or butyl alcohol in the presence
of aqueous sulfuric acid to form the respective acetate. When the reaction
is carried out continuously in acqueous solution, it is considered to be
in Category C, while batch type processes are considered to be in Category
D.
The process plant visited during the field data collection program included
manufacturing facilities in ethyl, propyl, and butyl acetates. The acetates
were produced in a semi-continuous manner in two independent systems. Since
there were only two systems, only two different acetates could be manu-
factured simultaneously. At the time of the visit, ethyl acetate was being
produced In one system and propyl acetate In the second.
The facilities normally are operated continuously for a period of from 1
to 6 weeks. When it is desired to change production (turnaround, e.g. from
ethyl acetate to butyl acetate), the reaction stills are washed. Production
may then be resumed. Thus, all three acetates are manufactured in two in-
dependent facilities by alternating production within each facility.
-98-
-------
DRAFT
Ethyl acetate is produced by the esterification of acetic acid and ethyl
alcohol in the presence of a catalyst such as sulfuric acid. A flow diagram
for the process is shown in Figure *t-19« The reaction is reversible and
eventually reaches an equilibrium at about a 67 percent conversion to ethyl
acetate. In order to obtain high yields, the reaction must be forced to
completion by removing the water formed and employing one reactant in excess.
There are many modifications of the process, but all operate on the same
general principle; that is, acetic acid is reacted with an excess of ethyl
alcohol in the presence of catalytic amounts of sulfuric acid.
The process is carried out either batch by batch or in a continuous manner,
depending on the nature of the raw materials and the size of operation.
The main variable is acetic acid concentration, which may range from very
dilute (about 8 percent) to concentrated (about 100 percent). Generally,
95 percent ethyl alcohol and 50 to 66 Be' sulfuric acid are used. The
ratio of reactants varies according to the process used and the type of
equipment available. For a batch process, the reactants may be mixed in
the following proportions: 10 parts by weight of 8 percent acetic acid,
10 parts by weight of 95 percent ethyl alcohol , and 0.33 part by weight
of 50 to 66°Be' sulfuric acid.
The continuous process may be used for any acid concentration, but it is
particularly applicable to the utilization of dilute acetic acid, such
as that obtained from ethyl alcohol by fermentation. Acetic acid, excess
95 percent ethyl alcohol, and about 1 percent of 66°Be' sulfuric acid are
mixed and continuously passed through a preheater to an esterifying column.
The column and other equipment are generally constructed of copper. The
mixture is allowed to reflux, and a suitable amount of distillate is with-
drawn from the top of the column, which is held at about 80°C.
The distillate, containing about 70 percent alcohol, 20 percent ester,
and 10 percent water (the acetic acid is consumed in the esterifying
column) is run to a separating column. Here the mixture is refluxed,
and a ternary azeotrope (83 percent ethyl acetate, 9 percent ethyl
alcohol, and 8 percent water) is removed from the top of the separating
column at approximately 70°C.
This homogeneous mixture is run to a proportional mixer, where it is
blended with (approximately) an equal volume of water. The mixture is
allowed to settle in a decanter, where the two layers that form are
separated. The bottom aqueous layer, containing small amounts of
alcohol and ester, is recycled to the lower part of the separating
column, where the ester and alcohol are removed in the constant-boiling
ternary mixture taken overhead. A wastewater stream is continuously
drawn off as bottoms from the separating column. This stream is neces-
sary to provide a route for removal of water from the process. It in-
cludes stoichiometric water from the ester ification reactions, dilution
water which may be present in the feedstocks, decanter water, and condensed
stripping stream used in all three distillation columns.
-99-
-------
DRAFT
The top layer in the decanter contains about 93 percent ethyl acetate,
5 percent water, and 2 percent ethyl alcohol. This layer overflows to
a drying column, where a sufficient amount of the ester is distilled to
carry over all the water and alcohol present. This overhead material
is returned to the separating column for recovery of the ester and reuse
of the alcohol.
Ethyl acetate is withdrawn as bottoms from the drying column and either
run to storage or redistilled. The latter is generally necessary to
remove copper salts formed in the copper columns. Other impurities such
as higher-boiling esters may also be present, depending on the purity
of the raw materials. The yield of ethyl acetate is 90 to 100 percent
based on acetic acid. Batch process yields are about 95 percent.
During the sampling visit, propyl acetate was being produced in a manner
analogous to that described in detail for ethyl acetate. During the
actual production runs, the only continuous contact process wastewater
stream from either unit was the aqueous bottoms from the separating
column. Process RWL calculated from flow measurements and the analysis
of these streams are indicated in the following tabulation:
Ethyl Acetate Propyl Acetate
PROCESS FLOW
liter/kkg 1,290 1,190
gal/M 1b 155 1^2
BODc RWL
mg/liter1 38 7
kg/kkg2 0.0*49 0.008
COD RWL
mg/liter1 79 10
kg/kkg2 0.102 0.012
TOC RWL
mg/liter1 26 k
kg/kkg2 0.03^ 0.005
'Raw waste concentrations are based on unit weight of pollutant per unit
volume of contact process wastewaters.
2Raw waste loadings are based on unit weight of pollutant per 1000 unit
weights of product.
Based on the previous process description, the raw waste concentrations
and subsequent loadings appear quite low for Category C processes. This
may be partially explained by the fact that each of the three distillation
-100-
-------
DRAFT
columns in the process utilizes direct steam sparging to volatilize
organic materials and drive them overhead. It may be that the samples
were taken during a period when nearly all of the organics in the
decanter water were being effectively steam stripped in the separating
co1umn.
Although the recycle of decanter water and overheads from the drying
column for recovery of unreacted alcohol and product esters is con-
sidered good operating practice, it is questionable whether such low
organics concentrations in the separating column bottoms can be maintained
on a steady-state basis. It should be also noted that both the ethyl and
propyl units were operating respectively at 8?.5% and 90.5% of original
design capacity. Under such conditions, each unit may be able to provide
sufficient hold-up so that a bottoms stream which is relatively free of
organics may be withdrawn for limited periods of time.
There are also non-continuous waste streams associated with process turn-
arounds when production shifts from one ester to another. Organic residues
from the esterifying column may amount to approximately 3 kg/kkg of ester
product on a cumulative basis. These organic residues contain no ap-
preciable water and are burned in an incinerator.
In addition, the esterifying columns are normally washed with a detergent
or cleaning agent during production turnarounds. This wastewater is highly
concentrated in organics and amounts to an additional 3 liter/kkg of esters
on a cumulative basis.
Although it is clear that these concentrated organic streams would add to
the process RWL, it is not possible to specify quantitative values for
pollution parameters such as BODc, COD, etc. It is also questionable
that the RWL's calculated are truly representative of the process.
Non-contact wastewaters include cooling water flows. Cooling water is
circulated throughout the reaction, refining, and stripping stills. A
loop system is employed. Make-up water for the entire plant is approxi-
mately 78,000 liters/kkg of esters (20,000 gal/M Ib). Most of this loss
results from evaporation in the cooling towers. A small flow is bled
from the esters units to prevent a chloride build-up. This wastewater
flows to the dilute stream and eventually to the wastewater treatment
faciliites.
Condensate from the entire plant is piped back to a condensate return
system for boiler feed water. The total quantity of condensate is ap-
proximately 2,000 kg/kkg of ester. Condensate from the various processes
cannot be segregated, but it is estimated that condensate from ester pro-
duction is less than 1 percent of the pollution abatement systems.
The process RWL presented for these two acetic acid esters are not con-
sidered to be representative of the process. Therefore, no effluent
limitations are proposed.
-101-
-------
DRAFT
o
z
O -i
5 °
< z
u <
I- Q
^K {/) -r^
» ii i *-
* <*
CL
3
o
u
<
u
LU
102
-------
DRAFT
Product: Acrylonitrile
Process : Ammonoxidation of Propylene
Process RWL Category: C
Chemical Reaction:
2 CH2CHCH3 + 2 NH3 + 3 02 > 2 CH2 CHCN + 6 H20
propylene Ammonia acrylonitrile
Acrylonitrile 5s used in the manufacture of acrylic fibers, Acrylonitrile-
Butadine-Styrene (ABS) and Stryene-Acrylonitrile (SAN) resins, and nitrile
rubber,
A typical process flow diagram for the manufacture of acrylonitrile via
ammonoxidation of propylene is shown in Figure 4-20.
Approximately stoichiometric proportions of air, ammonia, and propylene
are introduced Into a fluid-bed reactor at 15 psig and 750-925°F. One
attractive feature of the process is that it is not necessary to use
polymerization-grade propylene. The contact time is several seconds,
and once-through flow is used because conversion of propylene is practically
complete after a single pass.
The catalysts originally developed were oxides of bismuth, cobalt and molyb-
denum, and molybdates of bismuth and cobalt. However, the use of a newly
developed depleted-uranSum catalyst is described as considerably enhancing
acrylonitrile yield at the expense of the by-products.
The reactor effluent is neutralized with sulfuric acid to remove unconverted
ammonia. The quenched liquid stream from the neutralizer proceeds to a
steam stripper to recover some of the reacted products. The overhead from
the neutralizer goes to an absorption column, in which the stream is washed
with wafer to produce an unabsorbed stream of inert gases and a solution
of acetonitr5le, acrylonitrile, and HCN.
The solution is then stripped of the dissolved products, which are fractionated
to rfcmove pure HCN and then sent to the main purification section. First,
d main fracttonator produces an overhead consisting of wet acrylonitri le,
and a bottoms stream of wet acetonitrile. The overhead is treated first
by extractive distillation, and then by conventional fractionation to re-
move undesirable light and heavy ends, to produce pure acrylonitrile. The
bottoms from the main fractionator can be fractionated in a two-column
system to produce pure acetonitrile.
Although the actual ammonoxidation reactions is vapor-phase, the process
was considered within Category C because of the aqueous separation and
purification train.
-103.
-------
DRAFT
The major water pollution sources of this process are the wastewater dis-
charged from the steam stripping columns as shown in the process flow
diagram. Process RWL calculated from flow measurements and analyses
of the waste streams are indicated in the following tabulation:
Plant 1 Plant 2 Plant 3 Plant 4
Sampling Periods
#1 #2 #3 #1 #1 #1
PROCESS FLOW
liter/kkg 3,920 3,920 3,920 6,590 4,010 2,820
gal/M Ib 471 471 471 790 480 338
BODc RWL
mg/liter1 18,000 18,700 19,300 3,330 13,900
kg/kkg2 71-7 73-3 75.8 21.9 55.5
COD RWL
mg/liter1 57,400 60,300 60,700 21,100 36,200 41,100
kg/kkg2 229 237 238 139 145 116
TOC RWL
mg/liter1 25,600 24,000 25,100 8,700 15,300 19,100
kg/kkg2 102 94.2 98.6 57-3 61.2 53-9
waste concentrations are based on unit weight of pollutant per unit
volume of contact process wastewaters.
2Raw waste loadings are based on unit weight of pollutant per 1000 unit
weight of product.
The foregoing data indicate that the RWL of Plant 1 is higher than that of
Plants 2, 3, or 4. This difference is attributed to the discharges by
Plant 1 of light hydrocarbons removed as the overhead from the acetonitrile
purification column into the sewer lines. The analytical results from the
sampling survey also reveal that, in addition to the pollution parameters
shown in the above tabulation, the concentration of parameters such as
nitrogen, sulfate, and cyanide are at levels hazardous to the biological
pretreatment process.
An average of all the RWL data was considered as BPCTCA. It should be
noted that all of the plants surveyed use deep-well injections to dis-
pose of wastes from the manufacture of acrylonitri le. The effluent
limitations developed for this process would be applicable only if deep-
well injection becomes unfeasible.
The alternative routes for the manufacture of acryloni tri le include:
1) catalytic dehydration of ethylene cyanohydrin; 2) catalytic reaction
of acetylene and hydrogen; and 3) catalytic reaction of propylene with
nitric oxide. However, present practice concentrates exclusively on the
ammonoxidat ion of propylene.
-104-
-------
DRAFT
LU
Q_
O
z
o
0 p
^ <
^ 9
LU X
S °
o I
" <
o
u
=*
1 LU 1
=n "1
cs -a: 1
t Nouvomund
31iailN0133V
, i
H^ U
f£
*1
z P Noinmusia
S° 3All3tfaiX3
ssl T
-- L
\ Nwmoo
NOH
i
^
.^^^i 1 1 ^^I^H ' ' '
1
L
a3ddiais
T
L
\ >
asaaosav
t
i
^3-
\~
[=
I
a3znvain3N
^.
p"
L »i.n.
IT
LU «a:
i
"It
J
1
-»
1
J
I
1
_
r
i
-i
r
LU
1
0=
h- C
cu c.
\ >- 3
^ oc. -
1!f « ;
t
1
NOIIVG may
31iailN01Aa3V
t
T
i nai inu >Aauv
f
T
-a:
LU
CO
^
hi ill n 1 no i
^ li Inl 1 1 1 U J i^^^»^^^^
3NlddiaiS
t
1
105
-------
DRAFT
Product: p-Aminophenol
Process: Catalytic Reduction of Nitrobenzene
Process RWL Category: C
Chemical Reactions:
H-SOj,
C£H5N02 + 2 H2 > C6Hj,(OH)NH2 + H20
metal
nitrobenzene catalyst p-aminophenol
Typical Material Requirements:
Basis 1000 kg p-Aminophenol
Nitrobenzene 1800 kg
Sulfuric Acid 1550 kg
Surfactant 20 kg
Anhydrous Ammonia 500 kg
Toluene 80 kg
Hydrogen Gas
Nitrogen Gas
Antioxidants
By-Products
Aniline 220 kg
p-Aminophenol is versatile in its use as a dye intermediate, as it can
be used in the preparation of disperse, nitro, acid, mordant, direct,
sulfur, and oxidation dyes. It Is also widely used as a photographic
developer. The by-product aniline produced during the reaction also
has wide uses in the dye, drug, rubber, plastics, and animal-feed in-
dustries.
Production of p-aminophenol is based upon the reduction of nitrobenzene
with hydrogen in the presence of aqueous sulfuric acid and a metal-
containing catalyst (platinum, palladium, or mixtures of the two), as
shown in Figure 4-21. The raw materials (nitrobenzene, deionized water,
hydrogen gas, sulfuric acid, surfactants, and the metal catalyst) are
fed into a reactor at a temperature ranging from about 60° to 120°C.
Prior to completion of the reaction, the reduction of the nitrobenzene
is interrupted for catalyst recovery and recycle. During the catalyst
recovery step, a distinct interface forms between the reaction products
and the catalyst-containing nitrobenzene. The nitrobenzene layer is
then separated and can be employed in a subsequent reduction step.
-106-
-------
DRAFT
The reaction products proceed to a purification and isolation step. Ad-
ditional materials (anhydrous ammonia, toluene) are added to facilitate
separation of the by-product aniline from product p-aminophenol. After
isolation of the aniline, solvent is recovered and recycled to the puri-
fication and isolation step.
Following purification, the p-aminophenol is dried and packaged for sale.
The major pollution sources of the process are the wastewaters generated
during the aniline recovery step and the p-aminophenol drying step. These
wastewaters are collected and passed through two evaporators connected
in series. This evaporation process separates the very concentrated am-
monium sulfate wastes for landfill disposal. It was claimed by the plant
which was visited during the sampling survey that the evaporation process
should be considered as pollution abatement facilities rather than as
part of the manufacturing process. However, separate samples as influent
to and effluent from the evaporation process were sampled, and the RWL's
are presented in the following tabulation. Also included in the RWL
calculations are the wastewater resulted from floor, tank washes, and
pump seal leakages.
Influent to Evaporators
FLOW
1iter/kkg
gal/M Ib.
1
6005 RWL
mg/5
kg/kkg'
COD RWL
1
mg/1
kg/kkg'
TOC RWL
mg/1
kg/kkg
Sample
Period #\
15,000
1,800
59,300
890
115,000
1,725
33,600
505
Sample
Period #2
15,000
1 ,800
59,300
886
101,000
1,508
27,100
^07
Effluent from Evaporators
Samp1e
Period #1
12,600
1,510
3,300
5,850
73.7
1,730
21.7
^Raw waste concentrations are based on unit weight of pollutant per unit
volume of contact process wastewaters.
Raw waste loadings are based on unit weight of pollutant per 1000 unit
weights of product.
The process RWL based on the influent to the evaporators was considered
as BPCTCA. The effluent from the evaporators (aqueous condensate) is
-107~
-------
DRAFT
combined with wastewaters from other processes and treated In an acti-
vated sludge plant prior to discharge to surface waters.
Non-contact wastewaters Include cooling tower blowdown and boiler con-
densate. Each of these wastestreams is bled continuously for the control
of dissolved solids. The streams are combined with the treated process
wastewaters and eventually discharged to a receiving stream.
p-Aminophenol can also be prepared conveniently by the Beehamp method of
reducing p-nitrophenol in the presence of iron filings in an acid medium.
Alternatively, the product may be prepared by catalytic hydrogenation of
p-nitrophenol, by treatment of phenylhydroxylamine with an acid catalyst,
(such as sulfuric acid), by reduction of azoxybenzene in acid solution,
and by treatment of p-chlorophenol with ammonia.
-108-
-------
DRAFT
UJ t
N
Z
UJ
CO
0
* °
t- 25
Z =
£
., z
O -
z 1
2 - >
H- - -
« = s 1 r
Z5 -i = _j_^j_
Q = -2
"J *. °c= g
OSJ i- i- - z
cj « ^:
11 = _l CO >- D CJ
U « -.. -52
w^ n- M I
? 1 e=
«J >- C= 03
i t t l _
*T ..J, L z
^ -J" '^ =,
U, i ?r ~j __* *~ :z
>- ^" LU »" CJ SB 1
' _1 9- >C=UJ Ll_«t«I
(-V K- ua e= uj * ^ c^
VX -r =» coccoa £_
C_3 d >
ISJ 1 UJ
T "
« o- =
ffll. (_
04
^ t .. " n
1 ^,..... a.
^ *" S i «
< s H- >. -_^ s s
. LU co o= LU - 3 it;
I 00 9- LU _1 " 0=
^, o _i =- u ^. J , *
a» °= -* e= >- T T
« t- I- CJ CJ 1 I
rff -I LU LU 1 1
2 -»« 1
, _ _.,. * CO
LU LU
z ,_, oo
LU *~ «r
r^i c=
3E
LU "^ C
m r, z
=
NOIlVaOdtfA]
39V1S pl)Z
t
- NOIlVaOdVA3
39V1S lsl
CX
3C
co «a
ZD C
o= :z x
CD 3E UJ C
>- C3 ZD
ac Z i h-
= =E c=3 a
-^ ^ t ^
j ,
i
M «C
ea
\- °z
-------
DRAFT
i
Product: Calcium Stearate
Process: Neutralization of Stearic Acid
Process RWL Category: C
Chemical Reactions;
2 C17 H35 COOH + 2 NaOH + CaCl2 XC^H^COO^ Ca + 2 MaCl + 2 H20
sodium calcium
stearic acid hydroxide chloride calcium stearate
Calcium stearate belongs to a group of water-insoluble metallic soaps.
Their water insolubility differentiates them from ordinary soap, and
their solubility or solvation in organic solvents accounts for their
manifold uses. The precipitation process is the classical method of
preparing metallic soaps, including calcium stearate. It is now used
primarily in making high melting-point soaps which precipitate as light,
fluffy powders and can be recovered by filtration.
Raw materials for the production of calcium stearate including stearic
acid, sodium hydroxide, and calcium chloride. A flow diagram of the
process is presented in Figure *»-22. The feedstocks are fed continu-
ously into a reactor, where a two-stage precipitation process takes
place. In the first reaction, an alkali soap is formed by reacting the
stearic acid with caustic soda. In the second step, the alkali soap is
treated with a water solution of the calcium chloride to precipitate the
calcium stearate. Sodium chloride remains dissolved in the mother liquor.
The resulting slurry, containing about 10 percent of calcium stearate,
is continuously fed to a filtration step for solids concentration. Vacuum
filters are employed, and the filter cake is subsequently washed. The
partially dried filter cake is then placed in tray dryers, where the re-
maining water is removed. The dried soap may then be ground and sepa-
rated to produce a powder of uniform particle size.
The major water pollution sources of the process are wastewater discharged
from the filtration step and from the "Battery-Limit" clean-up water.
Multiple wastewater samples were obtained during the plant visit in the
survey period. Process RWL calculated from flow measurements and the
analyses of these streams (subjected to probability analyses of occur-
rence) are presented in the following tabulation:
-110-
-------
DRAFT
10% $0% 30%
Occurrence Occurrence Occurrence
PROCESS FLOW
liter/kkg 5^,100 5MOO 5l»,100
gal/M Ib 6,1*60 6,1*60 6,1*60
BOD5 RWL
kg/kkg1 13.2 13-8 }k.k
COD RWL
kg/kkg1 30.2 32.8 35.5
TOC RWL
kg/kkg1 22.1* 23.1 23-9
1Raw waste loadings are based on unit weight of pollutant per
1,000 unit weights of product.
The high flow of the waste stream can be explained by the fact that a
considerable quantity of water is utilized during the filtration step to
wash the solid cake. The analytical results also indicate high concen-
trations of calcium and chloride in the waste streams; these are attri-
bu'ed to excess raw material required by the reaction and to the reaction
by-product, salt.
The 50 percent occurrence shown in the RWL data above was considered as
BPCTCA, The wastes from calcium stearate manufacture are discharged to
the local municipal sewer systems.
-Ill-
-------
DRAFT
Q
U
<
U
0£.
<
P
CO
u_
o
Z
CN O
^ <
^ U
LU <
§ I
O
I
CO
U
3
CJ ^ C_D
-112-
-------
DRAFT
Product: Caprolactam
Process: UBE-lnventa Caprolactam Process
Process RWL Category: C
A schematic flow diagram illustrating the UBE-lnventa process is pre-
sented in Figure 4-23. The overall process plant can conveniently be
divided into three sections:
1. Cyclohexanone
2. Hydroxylamine
3. Caprolactam
Cyclohexanone Section
The reaction to produce Cyclohexanone and cyclohexanol by the oxidation of
cyclohexane is represented by the following equations:
C6H12 + i 02 -> C6HnOH
cyclohexane cyclohexanol
C6Hi2 + °2 * C6H10° + H2°
cyclohexane Cyclohexanone
Approximately equal quantities of the hexanol and hexanone are formed.
This process is characterized by use of air as the oxidant. As shown in
Figure 4-23, fresh feed and recycle cyclohexane are first delivered to
the cyclohexane mix tank and then passed in a fixed quantity through a
surface pre-heater to the oxidation reactor. The reaction takes place
at a pressure of 10 atmospheres and a temperature of 160°C. The reaction
does not require a catalyst and is usually carried out in a series of
continuously-stirred tank reactors rather than the single oxidizer shown
In Figure 4-23.
The oxidant air is compressed and sparged into the oxidizer by a circu-
lation gas blower. The gases leaving the oxidizer are cooled (by pre-
heating the cyclohexane feed) and sent to a condenser. The gaseous
phase from the condenser is vented through an off-gas scrubber used to
recover unreacted cyclohexane vapor. The liquid phase from the condenser
is recycled and combined with cyclohexane in the mix tank. An aqueous
layer is drawn from the mix tank to remove stoichiometric water of re-
action .
The liquid effluent from the oxidizer is combined with the bottoms from
the off-gas scrubber. Cyclohexanone recovered later in the process is
used to absorb cyclohexane in this scrubber.
-113-
-------
DRAFT
Condensed steam-jet water (from vacuum distillations later in the pro-
cess) is added to the reaction mixture prior to entering the acid wash
column. The overflow from the acid wash column is separated into an
aqueous and organic phase in a liquid/liquid separator. The aqueous
layer drawn from the separator is called "Acid Water" because it con-
tains significant concentrations of adipic and hydroxy caproic acid.
Aqueous caustic soda is added to the organic phase from the separator '
prior to entering the saponification column. The process to saponify
the reaction liquor from cyclohexane oxidation is represented by the
following equations:
RCO - OC/-HH + NaOH > RCOONa + C,-H, OH
oil b 11
RCOOH + NaOH > RCOONa + H20
Saponification is necessary to remove by product esters and acids formed
in the relatively non-selective oxidation reaction.
The saponified liquor is separated into heavy and light phases in the
saponification separator. The heavy phase is sent to a steam distil-
lation column, where contact steam is used to recover cyclohexanol by
stripping. The column overhead is combined with the light phase from
the saponification separator. The bottoms from the column contains high
concentrations of sodium salts of organic acids as well as excess free
caustic. This material, called "Caustic Water", is combined with the
"Acid Water" for neutralization prior to disposal by deep-well injection.
The light spaonification liquor is fed to a distillation column, where
cyclohexane is distilled overhead for recycle. The bottoms from the
cyclohexane recovery column are fed to the light-ends column, where low-
boiling components are distilled overhead. These by-products are burned
as fuel. The feed to the light-ends column also contains crude cyclo-
hexanone produced by the dehydrogenation of cyclohexanol later in the
process.
The bottoms obtained by removal of the low-boiling components are fed to
the cyclohexanone column, where cyclohexanone is fractionated. The dis-
tillate from the top of this column is 99 percent refined cyclohexanone,
which is stored in the refined cyclohexanone tank.
The bottoms from the cyclohexanone column are fed to the cyclohexanol
column. In this column, cyclohexanol is separated from high-boiling
components. These high boilers are taken off as bottoms and burned as
fuel.
The refined cyclohexanol is taken overhead and sent to the dehydrogenation
reactor. The process to obtain cyclohexanone by dehydrogenation of the
refined cyclohexanol is represented by the following reaction:
-m-
-------
DRAFT
C6HnOH -> C6H1Q0 + H2
cyclohexanol cyclohexanone
This reaction takes place at about 380°C by passing cyclohexanol vapor
over a Zn-Ca catalyst. The cyclohexanone produced is recycled to the
light-ends column, and the hydrogen is vented.
Hydroxylamine Section
Ammonia gas is mixed with air and fed to the ammonia oxidizer, where
oxides of nitrogen are formed by the following reactions:
k NH3 + 5 02 - > k NO + 6 H20
nitric oxide
2 NO + 02 - > 2 N02
nitrogen dioxide
The oxides of nitrogen are fed to the nitroso absorption column, where
they are absorbed by ammonium carbonate and converted to ammonium nitrite.
NO + N02 + (NH/,)2 CO, - ^ 2 NH/, N02 + C02
Off gases (mainly nitrogen) are cooled and vented from the nitroso ab-
sorption column.
Aqueous ammonia is added to the ammonium nitrite and mixed in the nitrite
tank. The mixed solution then enters the S02 absorption column, where
it is converted to disulfonate by contact with S02 gas flowing in paral-
lel. This reaction proceeds as follows:
NHjjN02 + 2 S02 + NH3 + H20 - > NOH (SO^NH/,^
The disulfonate is fed to the hydrolysis tank, where it is hydrolized to
the amine.
NOH (S03NHil)2 + 2 H20 - > NH2 OH i ^SOj, + i H
The aqueous mixture of hydroxyl amine, sulfuric acid, and ammonium sulfate
is stored in the amine tank prior to the reaction with the refined cyclo-
hexanone in the oxime reactor.
Caprolactam Section
The oximation reaction to produce cyclohexanone oxime by adding cyclohexonone
to the amine is represented by the following equation:
C6H1QNOH + i (NHjj)2 SO^ + H20
-115-
-------
DRAFT
An aqueous ammonia solution is added to the oxime reactor, along with
the hydroxylamine and cyclohexanone, to satisfy the requirements of this
reaction.
The effluent from the oxime reactor is sent to a liquid/liquid separator,
where the aqueous phase (containing ammonium sulfate) is drawn off from
the organic phase (containing cyclohexanone oxime). The ammonium sulfate
solution is sent to a crystallization and drying process to recover amw
monium sulfate crystals for sale as fertilizer. The oxime is sent to
the rearrangement reactor.
The reaction to produce caprolactam by Beckmann's rearrangement and to
neutralize the oleum used in this process, is represented according to
the following equation:
CgH^NOH + H2SOi, + 2 NH3 > (^2)5 NHCO + (NHlt)2 SO/^
The oxime undergoes Beckmann's rearrangement in the presence of oleum
catalyst in the rearrangement reactor. The effluent mixture of caprolactam
and H2SO/J is fed to the neutralization tank, where the acid is neutral-
ized with an aqueous ammonia solution. The caprolactam solution and am-
monium sulphate solution obtained by neutralization are cooled and sepa-
rated in the crude lactarn vessel ; both of these solutions are aqueous.
The crude caprolactam solution is sent to the crude lactam tank. Part
of the aqueous ammonium sulfate is recycled back to the neutralization
vessel to prevent a violet rise in temperature from the heat of neutral-
ization. The remainder of this solution is sent to a series of solvent
extraction columns to recover additional caprolactam.
In order to remove light impurities, the caprolactam solution from the
crude lactam tank is steam stripped in the purification column. The
overhead condensate from this distillation column is discharged as waste-
water. The bottoms from the column, called "caprolactam oil", are sent
to a three-stage countercurrent solvent extraction unit for further puri-
fication.
The "caprolactam oil" is fed to the third column, where it countercurrently
contacts the bezene solvent fed to the first column. The aqueous ammonium
sulfate solution formed in the neutralization reaction is fed to the top
of the first extraction column. Condensed steam-jet water from vacuum
evaporators used later in the process is combined with the ammonium sulfate
solution prior to entering the first extraction column. Aqueous raffinate
solution from the second column also enters the top of the first column.
The benzene solvent entering the bottoms absorbs caprolactam from both
aqueous streams entering the top. The benzene and caprolactam extract
phase is taken overhead while the aqueous ammonium sulfate phase is drawn
off as bottoms. This ammonium sulfate, freed from lactam, is delivered
to the ammonium sulfate recovery unit. The benzene-caprolactam extract
leaving the third column is cooled and sent to a liquid-liquid separator.
The heavy liquid is recycled to the third extractor, while the light liquid
(containing caprolactam and benzene) is sent forward to the solvent re-
covery column.
-116-
-------
DRAFT
The next processing step is to remove impurities in the lactam by reacti-
fication. The distillation of the lactam must be done minimizing the
heating time at a low temperature. Consequently, thin-layer, falling-
film type vacuum evaporators and fractionators, which feature a short
residue time and a small pressure drop, are used.
The lactam solution is fed to the Stage I evaporator, where low-boiling
materials are distilled off. The lactam flowing from the bottom is fed
to the Stage II evaporator. Forerun materials, near the lactam boiling
point, are evaporated in this stage, and the vapors obtained are frac-
tionated in a distillation column (Stage II fractionator). The distil-
late foreruns are combined with the light ends and burned as fuel. A
portion of this material is also used as a solvent in the final batch
stills used to strip additional caprolactam from process residues.
The bottom product from the Stage M evaporator, is fed to the Stage III
evaporator, where pure lactam is distilled off. The bottoms are dis-
charged to the Stage IV evaporator and fractionator, where pure lactam
and high boiling materials are separated. The pure lactam distilled
from Stages III and IV is condensed and sent to the Pure Lactam Tank.
The bottoms product of the Stage IV evaporator is mixed with a portion
of the intermediate fraction and then charged to a batch still, where
any lactam present is recovered. The distilled lactam is fed back to
the Stage II fractionator. The batch still residue is removed by water
washing and either incinerated or discharged as wastewater.
One manufacturer utilizing the UBE-lnventa cyclohexane process was sur-
veyed during the field data collection program. The manufacturer actu-
ally operates two process plants which run in parallel. Process raw
waste loads (RWL/ were calculated using both historical data provided by
the manufacturers and the analytical results from field sampling. This
information is shown in Table 4-1. A separate tabulation is provided
for each of the two plants.
It should be noted and clearly understood that the RWL's shown in Table 4-1
are based only on wastewater discharged to the process plant sewers. It
Is apparent that significant quantities of waste materials (i.e., ACID
WATER, CAUSTIC WATER, light and heavy ends, etc.) shown in Table 4-1 are
disposed of by deep-well injection or by burning. It should also be noted
that approximately 4.5 kg. of by-product ammonium sulfate are formed for
each kg. of finished caprolactom. The ammonium sulfate is sold to fertil-
izer manufacturers and is not shown in Table 4-1.
The ACID WATER and CAUSTIC WATER discussed previously and shown in Table 4-1
would drastically increase the process RWL if the injection well could
no longer be used for this disposal. Typical compositions provided by
the manufacturer for these waste streams are listed in the following
tabulat ions.
-117-
-------
DRAFT
Caustic Water Acid Water
Free NaOH k.5% Adipic Acid 11%
Organic Carbon 1^.5% Hydroxy Caproic Acid 9%
Total Na 10.0% Other Acids and Diols 15%
Acid Salts 24.8% Water 65%
Anone, Anol 0.5%
Water 63%
One alternative disposal method would be incineration of the ACID WATER.
The CAUSTIC WATER, however, can not be burned because of its alkaline
composition.
The previous discussions also indicated that other waste streams in the
Cyclohexanone and Caprolactam Sections are concentrated and incinerated.
Steam-jet condensates from the vacuum evaporators are recycled to the
process. It was, therefore, concluded during Phase I of this study that
the RWL shown for Plants I and 2 in Table A-1 should be considered as
minimum values commensurate with BADCT for this process.
-118-
-------
DRAFT
CM
4-J
C
~-
Q_
in
t/)
<1J
O
O
Q_
E
4-J
U
ro
r
0
l_
Ct.
ro
o
ro
4-J
c
0)
i >
4" c
_
0) 1
i UJ
XI CO
ro =>
I-
c
O
XI
o>
iyi
(0
CO
§
|
0)
4-1
ro
3
^;
ro
a:
(_J
O
Q
O
O
O
CO
3
O
.
U.
1
4-J
C
ro
0.
~S
l/l
ro
CO
o i o i
ro o >. c c
4J O 3 4J U
O o ,_ -^_ o c£ o a:
i- d)' * o E QJ
CLCO t-o Ot/iu^in
ro a) o ' u D u
OQ) 4-» O4-OHO
c s-ro me a) c a)
x> O a)3 xiro «c c
o)C4-> cxxiroEra
x: ro rau ujtu c nj
w x3 x: O
CD 4-J4->OOQ)>vOJ
cx:x>inx o x o
O 3 cnij (/) o i/i
u. o ro >- cr i-
u >
o o| :r
CM
cn
o
CM
CM
~
XI
C Q) ^O
u c
3 1-
CQ ^ ro
O
o
J-
CO
cn
_y cn
V v/
V
C \ O
O C J-
XI O
S- XI
ro
.
.
ro <-
LJ ra
o
cn
-^ cn o
J t-J
0
cn cn 1
LA '
O LA
ro
i
cn
_v
v^
t/1
I- t T
OJ cn cn
,^_ my _y
*~~ ^^ ^^-
cn cn
o _y ^
o
CO CM ro
CO O
3 i=
0 Q 0
o xi i-
U_ CO 1- CD
ro i-
O 3
o
ro ro
o u
^^ f- c
4-
0) 1- O ^
4-1 01 1 . XI
ro 4-1 c L.
' > f \ M- ro 3 en xi
>- x: 3 or: 3 a)
L. u 3 a)
CD 4-1 O1 4J l/l CO '
> ro zi -o i/i aj
M *D E o -M a.
C_CO» O wi 21 t/1 fDE
Ocnc/itn i_c fD Ofu
. a) y- (U " 15 co
M _J £ cc. ~a ' * i-
u O QJC-OC O -a
QJ "O L- **b -M O QJ O -t ' »""
CACM- tnc_j' i/io)
ro i- vi CD o "O
E . - QJ XU_
(TJ 10 "XJ 4-J ^~" >^ U Ul
W CC (D U1L.QJ rD C C
(DUO CO O >
QJ -C QJO-M ~aT5
O U ^ DUO CUQJ
u o cr 03 O" QJ ^* m 1/1 t/i
D. U_
-------
DRAFT
U
o
ee
a.
u
<
_j
O
OL
a-
<-> <
^ U
*
it, <
O
IX
-------
DRAFT
Product: Caprolactam
Process: DSM Caprolactam Process
Process RWL Category: C
A second Caprolactam manufacturer surveyed utilizes the Dutch States
Mines (DSM) process. A schematic flow diagram for this process is shown
in Figure k-2^. Cyclohexanone is prepared in two stages: 1) oxidation
of cyclohexane to cyclohexanol; 2) catalytic dehydrogenation of cyclo-
hexanol to cyclohexanone.
Cyclohexane is first oxidized with air in the presence of a cobalt cat-
alyst. The major oxidation product is cyclohexanol, with some cyclo-
hexanone produced as a co-product. Oxidation by-products such as acetic
acid, adipic acid, and cyclohexanol esters are also formed.
Off gases from the oxidizers are scrubbed with kerosene prior to venting.
Unreacted cyclohexane is subsequently recovered for recycle in a strip-
per. The process operates at a cyclohexane conversion of less than
10 percent, so that recovery of unreacted cyclohexane is an important
part of the process.
The liquid effluent from the reactor is sent to a decanter where the
aqueous phase is removed as wastewater. The organic phase is combined
with the cyclohexane stripper overhead.
This mixture is then combined with aqueous saponified liquor (containing
free caustic) and sent to a neutralization tank. The effluent from this
tank is decanted. A heavy layer containing saponified acids, and esters
are drawn off and sold. This stream is similar in nature to the com-
bined "ACID WATER" and "CAUSTIC WATER" from the Inventa process. Its
only value is for recovery of dibasic acids such as adipic acid. This
can be done by lowering the pH and extracting the organic acids with a
solvent such as methanol.
The light layer from the decanter is sent to the cyclohexane recovery
column. The column overhead is condensed and separated into aqueous and
organic layers. The organic layer is cyclohexane, which is recycled to
the reactor; the aqueous layer is discharged.
The bottoms from the cyclohexane column are sent to the saponification
column, where an excess of caustic and water are needed. The effluent
from the saponification column is decanted into heavy saponified liquor,
which is sent back and mixed with the reactor effluent, and a light or-
ganic layer. The light layer is sent to a series of distillation columns,
which remove light ends, cyclohexane, and cylcohexanol respectively. The
cyclohexanol is sent to a catalytic dehydrogenation reactor to produce
cyclohexanone. The reactor effluent is sent back to the light-ends column,
-121-
-------
DRAFT
The overhead from the light-ends column is burned in a flare. Hydrogen
from the dehydrogenation reactor is passed through a seal pot containing
water and is vented. Water from the seal pot is discharged to the sewer.
Refined cyclohexanone is combined with ammonium hydroxide and hydro-
xylamine sulfate in the oxime reactor to produce cyclohexanone oxime.
The hydroxylamine sulfate is produced by ammonia oxidation similar to
the Inventa process.
It should be noted that DSM has also developed an alternative process to
produce cyclohexanone oxime which minimizes by-product ammonium sulfate.
This process utilizes hydroxylamine phosphate oxime (HPO) and produces
an amine which may be fed directly to the rearrangement reactor. The
process has been integrated into the total manufacturing scheme so that
the rearrangement reaction can utilize oxime feed from either the con-
ventional oxime reactor or HPO.
Excess sulfuric acid in the effluent from the rearrangement reactor is
neutralized with ammonium hydroxide in the neutralization reactor. The
reaction mixture is then sent to a benzene solvent reactor, in which the
crude lactam is absorbed in benzene and separated from aqueous ammonium
sulfate. The aqueous salt solution is combined with that from the oxime
reactor and sent to a recovery plant to produce crystals for use in fertilizer.
The lactam/benzene extract is mixed with water and sent to a solvent re-
covery column. The overhead mixture of benzene and water is decanted,
with benzene recycled to the extractor and extract water discharged.
An aqueous lactam solution is drawn off as bottoms from the solvent re-
covery column. The lactum is then purified by cation exchange followed
by hydrogenation. The ion exchange resin beds are regenerated by back-
washing with water and acid. The backwash wastewater is discharged.
The purified lactam is dried in a vacuum evaporator, with the water re-
cycled to the neutralization reactor. Product caprolactam is withdrawn
from the bottom of the evaporator.
Process RWL's for the DSM caprolactam process are presented and discussed
at the conclusion of the following section which describes the DSM pro-
cess for manufacturing cyclohexanone oxime.
-122-
-------
CO
CO
m
U
O
of.
a.
U
a.
U
CO
X
U
l
13
O
U
<
«J
O
ae
a.
<
U
-123-
-------
DRAFT
Product: Cyclohexanone Oxime
Process: DSM Cyclohexanone Oxime Process
Process RWL Category: C
Dutch States Mines has developed a new process for manufacturing cyclo-
hexanone oxime without by-product ammonium sulfate. This process has
been integrated into the overall caprolactam production scheme to pro-
vide Cyclohexanone oxime feed directly to the rearrangement reactor as
shown by the dashed line in Figure k-2^.
The basic chemistry for the new process Is shown in the reactions listed
below:
1. 4NH3 + 5 (
2. 2 NO + 02
3. 3 NO, + H,
> k NO +
2 N02
» 2 HNO
6 H20
+ NO
A simplified flow sheet for the process is shown in Figure ^-25. The
first processing step is the oxidation of ammonia with air to nitrogen
dioxide (equations 1 and 2). The gases from the oxidizer are then ab-
sorbed in the recirculated aqueous process solution in the nitrogen gas
absorber. Nitrogen is vented as off-gas from the absorber.
The liquid effluent from the absorber is a buffered aqueous solution of
nitric acid (equation 3), phosphoric acid, and ammonium nitrate. This
mixture is passed to the HPO reactor, where hydroxylamine is produced
(equation A).
The HPO reactor is a column sparged with compressed hydrogen gas. Un-
used hydrogen gas is separated from the catalyst suspension (palladium
metal on carbon), and recycled by means of a compressor.
The aqueous hydroxylamine solution from the HPO reactor is sent to a
series of stirred tank oximation reactors, where it countercurrently
contacts Cyclohexanone in the presence of toluene. Cyclohexanone oxime
is produced as shown in equation 5. Hydrogen ions liberated by the
oximation reaction are accepted by the phosphoric acid buffer system
(phosphate ions in equation 6). The oximation occurs at a pH of 1 to 2.
Due to the countercurrent flow, nearly quantitative conversion of cyclo-
hexanone is obtained.
-121.-
-------
DRAFT
Product cyclohexanone oxime is drawn from the oximation cascade (left-
hand side of Figure A-25), as a mixture of oxime, water, and toluene.
This product mixture is contacted with additional toluene in the oxime
extractor. The aqueous bottoms from the oxime extractor is drawn off as
wastewater.
The oxime-toluene solution from the oxime extractor is sent to the oxime
rectifier, where refined cyclohexanone oxime product is drawn off as
bottoms. This material can be used directly in the rearrangement re-
action to produce caprolactarn.
The overhead from the oxime rectifier is sent to a decanter, where re-
flux water is drawn off. The separated toluene layer is split, and part
is recycled to the oxime extractor; the remainder is sent to the process
1iquid extractor.
In this extractor, the toluene contacts aqueous process liquid leaving
the oximation cascade (right hand side of Figure k-25) The organic
phase from the extractor is mixed with fresh cyclohexanone and sent to
the oximation reactors. The process liquid leaving the oximation section
must be purified thoroughly to protect the catalyst in the hydroxylamine
reactor. To achieve this, the liquid is extracted with toluene and sub-
sequently stripped with steam in the process liquid stripper. Passage
through the stripping column also removes water formed during the prepa-
ration of hydroxylamine and oxime (equations A and 5).
The aqueous process solution is then recirculated to the nitrous gas ab-
sorber. Hydrogen and nitrate ions are restored in the process liquid by
production of nitric acid (equation 3)
Calculated process RWL's for Oxanone, Caprolactam, and HPO Sections of
the DSM plant are shown on Tables h-2 and ^-3. The data shown here have
also been combined to provide a total RWL for the integrated DSM capro-
lactam process (bottom of Table A-3).
Examination of the data indicates that the backwash from the cation ex-
change resin beds in the Caprolactam Section of the DSM process is a
major waste source not found with the Inventa process. The DSM route
uses chemical treatment by ion exchange and hydrogenation to purify the
product lactam, while the Inventa process uses an extensive vacuum dis-
tillation train for purification. It is not clear as to whether the
physical purification used with the Inventa process could be substituted
for the chemical method used by DSM. The impurities to be removed may
be significantly different.
The new HPO process incorporated in the DSM scheme also produces a signi-
ficantly higher RWL than the hydroxylamine section of the Inventa pro-
cess. However, this must be evaluated considering the *nct that by-
product ammoniu'Ti sulfate formation is minimized.
It is therefore recommended that the DSM waste loads shown in Tables k-2
and 4-3 be considered as commensurate with BPCTCA for this process.
-125.
-------
DRAFT
Table 4-2
Process Raw Waste Load Based on DSM Process
Oxanone Section
Saponified Acids & Esters
Light Ends Column Overhead
Cyclohexanol Col. Btms.
Decant Water from Oxidation
Reactor
Reflux Water from
Cyclohexane Column
Hydrogen Seal Water
Flare Condensables
Total (Based on Wastewater
to Sewer)
Caprolactam Section
Extraction Wastewater
Backwash from Cation Exchange
Resin Beds
Slowdown from Recycle
Process Water
Total (Based on Wastewater
to Sewer)
RWL Based on Cyclohexanone
F low
COD
(liters/kkg) (kg/kkg)
Sold for Product Recovery
Burned in Flare or Steam Boilers
358
478
5.37
8.35
239 6.78
No Data Ava ilable
1075
20,5
RWL Based on Caprolactam
Flow
COD
']i ters/kkg,
394
1,730
(kg/kkg)
0.39
37.1
No Data Ava liable
2,120
37.5
-126-
-------
DRAFT
Table 4-3
Process Raw Waste Load Based on DSM Process
Hyam Phosphate Oxime (HPO) Section RWL Based on Cyclohexanone Oxime
Flow COD
(liters/kkg) (kg/kkg)
Aqueous Reflux from 25.5 0.011
Cyclohexanone Oxime Rectifier
Stripper Overhead Condensate 1146 3.43
Wastewater from Oxime Extractor 509 1 .02
Miscellaneous Run-off 229 1.83
Total (Based on Waste 1910 6.29
Water to Sewer)
Average Raw Waste Load for Total Plant
Based on Finished Caprolactam Product
Flow BOD COD
Plant Section (liters/kkg) (kg/kkg) (kg/kkg)
Oxanone, HPO, and Caprolactam 7057 39.1 78.1
Process Wastewaters
Hydrogen Seal Water, Drainage 6113 3.2 6.6
From Oxanone, HPO, and Capro.
Ground Drainage Product Loading 630 0.5 0.9
Utilities Slowdown, Salt Drainage 15,268 4.3 7.4
Total (Based on Water 29,100 47.1 93.0
to Sewer)
-127-
-------
DRAFT
CO
CO
LLJ
U
O
ce.
0.
LJJ
3
X <=
-25
slES CYCLOHEXANONE O
RCULATED AQUEOUS PROCESS SOLUTI
FIGURE 4
IME-DUTCH STATE Mlt
RED
X
0
GO
* o^ ]
LLjt=- Z '
.. =z «
*=« 1 '"
] "
HBddidlS
UIIIUI 1 S'SJUUdd *^
T
aU13VdlX3
, o nu 1 1 ssnaoad ~*
'
t
i
^ ... p
- o
^ ' ^ x
00 s; => cc
<-> < T S >: * V->>^ ^ ^_
°= 0 IE LU i Z z;
_« U=> "=*=,=-
E;« ^' " = =
o, i ' H3 j i m a 1
Ol ^liJJQ 1 ^
cs> >- K ^ JW i Xu I
LU « 1__ pU 1
0= - ' 1 LU f
LU X X z
0 = '! >*
0 -i
H :=
- o i ,
J X CD 1 «c
J "~|^_ aoiovaix] i »*
^S ^wuo . ^
LLJ CC Z f ^
= = " , . J
*r 3= «c
CO O= Lt_!
«r c; . c^
1 ^ CJ K "" l GZ
U- Q_ LU CJ >-
<=; ^ cc jz
t
5 ". H3adosav J
^ LL.* sva snoaiiN *
Z
4 ^ LU
x - -~
UJ g g . S£
0 §°I^Z
IL d3zio xo * g- = =
f 1 LU C_D 31
>
u
o
-a:
- s *- ^ z
JV =! CC CO CD
IT 1 1 CD Z
= <= nr K
CO CO 1C Ck. «X
0=.
«x
-123-
-------
DRAFT
Product: Cresol (Synthetic)
Process: Methylation of Phenol
Process RWL Category: C
Chemical Reaction:
C6H5OH + CH3OH > CH.C6H/,OH + H20
phenol methanol cresol
Cresol (cresylic acid) is an isomeric mixture (o-, m-, and p-cresol) ob-
tained by refining the phenolic constituents present in coal tar, re-
fining the petroleum acids formed during the thermal and catalytic crack-
ing of petroleum, or by producing cresols syntheticly. Cresols are
used primarily as raw materials for the production of phenolic resins,
tricresylphosphate, disinfectants, and solvents.
The facility visited during the field survey produced an iosmeric mixture
of cresols by the methylation of pheonol. Figure A-26 is a process flow
diagram of this production technique.
The first step in the process is the preparation of the reactor feed.
Fresh and recycled methanol and phenol are mixed in a weight ratio close
to 2.5:1f. The mixed stream is then vaporized, and condensing Dowtherm
vapors are used to vaporize and heat the feed to 300°C. The reaction is
carried out at atmospheric pressure in a mixed-bed reactor which contains
activated alumina catalyst. The heat of reaction is removed by passing
liquid Dowtherm through the tubes imbedded in the fixed bed of the re-
actor. (The heat of reaction is approximately 20 K cal/mole.) Some of
the heat evolved by the reaction will be taken up as sensible heat by
the reactants, which are heated from 300°C to the reaction temperature
of 350°C. The contact time in the reactor to the point when 52 percent
of the phenol is alkylated is less than 30 seconds. A typical distri-
bution on a water-free basis is given below.
Weight %
Phenol l»8
o-Cresol 30
m- and p-Cresol 12
2, 6-Xylenol 15
Anisoles and Hexamethylbenzene 5
The vapors leaving the reactor are condensed and cooled to about 25°C,
and then sent to a phase separator. The vapor space is provided with a
vent in order to periodically remove any noncondensables (C02» CHr)
which might be produced in the reactor. The weight percent of water in
-129-
-------
DRAFT
the liquid phase is below 10 percent. Consequently, an aqueous phase
will not appear, because the water will stay in the organic phase. The
solubility level of water in phenol is close to 3k gms/100 gms phenol at
25°C. The presence of cresols and xylenol will undoubtedly decrease the
solubility limit, but the decrease is not expected to be sufficient to
form a separate aqueous phase.
The liquid phase is sent to an extraction column in which benzene (con-'
taining some methanol) is circulated countercurrent to the liquid product
stream. The phenol, cresols, xylenols, and by-products are transferred
to the benzene phase, while the water and the methano'l form an aqueous
phase.
The aqueous phase is then sent to a methanol recovery column, in which
methanol is removed as the distillate, while water (with small amounts
of phenol) remains as bottoms. The methanol is recycled to the reactor
feed-preparation section, while the water is discharged. The organic
phase, containing benzene and the reactor products, is sent to the ben-
zene recovery column, where this solvent is distilled overhead and re-
cycled to the extraction unit. This column is operated at atmospheric
pressure, with an overhead temperature close to 75°C. Cooling water is
used in the condenser, and condensing Dowtherm vapors are used in the
reboiler.
The bottoms stream leaving the solvent recovery still is sent to the
distillation column, in which anisole and light by-products are removed
in the overhead. Phenol is recovered in a subsequent distillation step
and recycled to the reactor section. The bottoms from the phenol re-
covery column are then sent to the product recovery section. In the
first column, o-cresol is recovered as distillate; in the second, a m-
and p-cresol isomer mixture is distilled overhead. The bottoms are sent
to the 2, 6-xylenol column, in which this co-product is distilled off,
while the bottoms containing the heavier by-products (such as hexamethyl-
benzene) are sent to organics disposal.
The major water pollution source is the water stream withdrawn as the
bottom of methanol recovery column. Process RWL's calculated from flow
measurements and analyses of the wastewater samples are indicated in the
following tabulation:
-130-
-------
DRAFT
PROCESS FLOW
liter/kkg 2,090
gal/M Ib. 250
BOD5 RWL
mg/liter1 1*1,300
kg/kkg2 297
COD RWL
mg/liter1 303,000
kg/kkg2 631
TOC RWL
mg/liter1 10,1*00
kg/kkg2 21?
iRaw waste concentrations are based on unit
weight of pollutant per unit volume of contact
process wastewaters.
2Raw waste loadings are based on unit weight
of pollutant per 1,000 unit weights of product.
The analytical results also indicate that phenol concentration in the
waste stream is at a level hazardous to biological treamtent processes.
The waste stream can either be pretreated with lime to form calcium
phenolate before being discharged into biological treatment processes
or can be steam stripped to reduce the phenol concentrations.
The process RWL shown above is considered as BPCTCA. The waste from the
plant is discharged to the municipal treatment plant.
-131-
-------
DRAFT
o
z
LLJ
Q.
0
Z
o
1
CN >
-4 jE
UJ ^
a: <
-3
O u
"- h-
LU
I
1
Z
CO
co
1
co
UJ
c_
c=
1
"1
3
)
t_
-
OQ
ac
CO
3=
rc
t_
NWI1103
Noiiniusia
J
NWPIOG
Nouvniis o
10N3Hd
i t
Nwnioo
Nouvn is a
T
NW0103
Noiivniis o
3N3ZN39
i i
3N3 7N39
t ,
H0i3tf3d * i
1
r> s
< *
A5
i
j
__j
»
I^^H^M
J
LLJ
cn
CO
^ C_D
X rj-i
or
T
T NUimOD. _J
HOIlvniiSIQ
T
c= o |
«C UJ
1
^^^^M ij w n n n n "^"^^
Noiimiisia
r
cc
'.LI
1
LU
CO
Nwmoa
______ Nni iwi°ii i"in ^
10NVH13W
,
z
I
«=c
or
UJ
z:
C_3
-z.
0
-132-
-------
DRAFT
Product^ Formic Acid
Process : Hydrolysis of Formamide
Process RWL Category: C
Chemical Reaction:
2 HCONH2 + H2S01} + 2 H20 >2 HCOOH +
suIfuric formic ammonium
formamide acid acid sulfate
The main application for formic acid outside the United States is as a
coagulant for natural rubber latex; domestically, over half of the total
is used instead of sulfuric acid in high-temperature acid textile dyeing,
and to some extent in leather tanning.
A process flow diagram for the manufacture of formic acid via hydrolysis
of formamide is shown in Figure k-27.
A feed stock, formamide, is stoichiometrically mixed with water and con-
centrated sulfuric acid in a pre-mixer, where the hydrolysis reaction
occurs. The mixture of reaction products is then sent to a series of
vacuum stills where ammonium sulfate in the form of a solid powder is
withdrawn as the bottom of a still. The crude formic acid vapor taken
as the overhead from the still is condensed by passing through a heat
exchanger and led to a storage tank before being discharged into a
purification still. The product formic acid is then taken as the over-
head from the purification still, while the residue in the still is
periodically flushed.
The major water pollution source of the process is the contact water
utilized for the water eductor which is used to pull vacuum for the dis-
tillation columns. Although the residue withdrawn as the bottom of the
formic acid purification still is periodically flushed into sewer lines,
this stream can be disposed of through incineration and is therefore not
Included in the raw waste load calculation. Process RWL's calculated
from the waste flow measurements and analytical results of the sampling
program were subject to analysis for probability of occurrence. The
following tabulation presents the results of the analysis.
-133-
-------
DRAFT
Probability Occurrence
ToT 50% Tof
PROCESS FLOW
liter/kkg 135,000
gal/M Ib. 16,000
BOD5 RWL
mg/liter1 0 7-8 20
kg/kkg2 0.0 1.05 2.75
COD RWL
mg/liter1 6.7 33 60
kg/kkg2 0.9 *».5 8.1
TOC RWL
mg/liter1 0 10 24
kg/kkg2 0.0 1.A 3.2
1Raw waste concentrations are based on unit weight of pol-
lutant per unit volume of contact process wastewaters.
2Raw waste loadings are based on unit weight of pollutant
per one thousand unit weight of product.
The high flows and low concentrations shown are from the water educator
used in the process. The 50 percent occurrence from the data above was
considered as BPCTCA. It should be noted that the ammonium sulfate by-
product is not considered in the RWL calculations. If this product could
not be disposed of by fertilizer manufacturers (or other non-aqueous
means), the process RWL would increase tremendously. The process wastes
considered are discharged directly to surface receiving waters.
Most of the formic acid produced in the U.S. is obtained as a by-product
from the manufacture of acetic acid via butane oxidation. Some of it,
however, is made by absorbing CO either in caustic soda, followed by
neutralization or in methanol, at high pressure, followed by conversion
to formamide and hydrolysis.
-134-
-------
DRAFT
4
o
O
Li.
to
>]
O
I
I
Q
U
O
-135-
-------
DRAFT
Product: Hexamethylene Tetramine
Process: Synthesis with Ammonia and Formaldehyde
Process RWL Category: C
Chemical Reactions:
6HCHO + km. > (CH,),N. + H00
5 /bH i
hexamethylene
formaldehyde ammonia tetramine
Two manufacturing plants were surveyed during the field data collection
program. Flow diagrams designated Plant 1 and Plant 2 are shown in
Figures 4-28 and 4-29.
Both plants combine aqueous formaldehyde and ammonia in a liquid phase
reactor to produce an aqueous solution of hexamethylene tetramine. The
reaction shown above produces about 0.8 kg of water per kg of hexamethylene,
and can be run with either an excess of ammonia or formaldehyde in the
reaction mixture. The ammonia is used preferably in a substantially an-
hydrous state, in either gaseous or liquid form initially. In view of
the high exothermic nature of the synthesis reaction, the desired control
and stabilization of the reaction temperature is somewhat more readily
achieved if the ammonia is introduced in liquid form so as to take ad-
vantage of the cooling resulting from its vaporization.
Reaction temperatures of 20-75°C constitute the range specified for manu-
facture of hexamethylene tetramine. The reaction is conducted at es-
sentially atmospheric pressure, although slightly superatmospheric pres-
sures may be employed for operating convenience. The production of
hexamethylene tetramine is carried out in the liquid phase in the presence
of an aqueous solution of the product as a rule. No catalyst Is employed
in the reaction of ammonia with formaldehyde to give hexamethylene tetramine,
In both plants the liquid effluent from the reactor is concentrated in a
vacuum evaporator, where water from the reactor and formaldehyde feed-
stock are drawn off. In Plant 1, the overhead from the evaporator is
sent to a vapor-liquid separator with the liquid phase recycled to the
evaporator; the water vapor from the separator is passed through a non-
contact surface cooler and then sent to a second vapor-liquid separator;
and condensate is drawn off from this second separator and discharged to
the sewer as wastewater.
A steam jet is used to draw vaccum on the evaporator. Water vapor from
the second separator is drawn off into the steam line as shown in
Figure 4-28. The steam and water vapor then pass through a jet and are
subsequently vented to the atmosphere.
-136-
-------
DRAFT
The concentrated liquid effluent from the evaporator Is sent to a centri
fuge where solid crystals of hexanethylene »>re separated. The centrate
Is 'ecycled to the evaporator
Hexamethylene crystals from the centrifuge are dried by hot air and
pulverized before being snipped off In bags or drums. Air from this
pa-'t of the process s -.".".^-.d u~,''^ dry dust collectors.
A sampling ..ironrarn covering sever-ii week,;1 operation at Plant 1 was con-
ducted. The riata on pro-..t.«:s RWL are summarized below on a probability
Process RWL Plant 1
^Q£ ur.currence
Sn^ C5r.cu; rencc
y«2: Occurr
I ite"
'.i t*
ow
<-,/kkg)
j = j
00
5%,
f i1* n ^ <£. \f (i i
^'«
COD
(kg/Fkg)
13.8
29.4
40.0
TOC
(kgTBTg)
4.5
9.3
15.1
sSn.-ijld be noted that. *'»> RV/L. data presented from Plant 1 do not in-
'i<* the ado * l - AnM , ioi.s :.-,K' pollutant "'>ads from carry-over into the
.! !-- "uojup. s 7 1 *-'' » ; apprccier 'on for the magnitude of these
- : ,>«'. - .'ami 'i.iq the process used in Plant 2.
,, ,w-j. (*,»- ;-.)r M- s orf<, ft-'.':. ; sno'./n , i: '"Igure 4-29. The liquid ef-
. ' ' ;,n i mi .ea.:t;j" :\u.r\. to a f , P conditioner and evaporator oper
'.- %«;'.:: ---jr.i" !« i (.Te.; at in-j ;^ a simiitir fashion to a baro-
coiic*>,v. i'.- -a: -irf "-j.-.uum or, Sotn unit'.. Water from both jets is
> " ;vf. ted i "i a .;,,p 'aval of dissolved solids
ir'nr? r^""<;et. f.Q" i ; 'u-; water ,
era-" * dischar-jeti from a wet scrubber used with
'^e -/'f>-' rrr^tri"' The fol!ow«r-g tabulations sum-
;r.~i r/,th th«; scrubber discharge and cooling
37-
-------
DRAFT
Process RWL Plant 2
Coo1 Ing Towe r B1 owdown
Scrubber Slowdown
Total RWL
Flow
(liter/kkg)
550
490
BODs
(kg/kkg)
83.2
COD
(kg/kkg)
116
112
228
TOC
(kgTkkg)
71.0
These data, from Plant 2, are significantly higher, both with respect to
flow and loading, than those obtained from Plant 1. The flow and loadings
attributed to the wet scrubber in Plant 2 are not sufficient to explain
the difference between the two plants. Instead, the difference relates
back to the fact that Plant 1 does not condense the steam used to draw
vacuum on the evaporator. Although Plant 1 uses a surface condenser
ahead of the vapor-liquid separator, there may be significant carryover
of contaminants to the steam jet. These contaminants are then vented to
the atmosphere with the uncondensed steam. The data shown for Plant 2
also may not fully show the magnitude of this difference in that the
flow from the cooling tower represents a blowdown and does not include
evaporation losses.
The manufacturer at Plant 2 indicated that the water circulation rate
between the cooling tower and barometric condensers is approximately
73,900 liters/kkg of hexamethylene. Flows greater than this would be
required if a once-through system were employed.
The RWL data from Plant 2 were considered as BPCTCA. These wastewaters
are discharged to a municipal treatment plant, as are the wastes from
Plant 1 .
-138-
-------
DRAFT
LU
Q
I
LU
Q
C£
O
u_
Q
Z
Z
o
00 »-
CN tl .
CO
LU
Z
LU
X
I
LU
X
LU
X
-139-
-------
DRAFT
-140-
-------
DRAFT
Product; Hydrazine Solutions
Process; The Raschfg Process
Process RWL Category; C
Chemical Reactions;
NaOH + C12> NaOCl + HC1
sod!urn
hypochloride
NaOCl + NH3 > NH2 Cl + NaOH
chloramine
NH2C1 + NH3 > N2Hj, + HC1
hydrazine
The principal use for hydrazine is as a military missile fuel where an-
hydrous hydrazine Is required. Hydrazine also has a number of non-military
applications. The most important of these is maleic hydrazine, a plant-
growth regulator used for tabacco suckering and tree pruning. Other uses
Include wash-and-wear finishes, Pharmaceuticals, anti-oxidants used in
foam, hydrazine monobromide, and soldering flux.
A process flow diagram for the manufacture of hydrazine hydrate is shown
in Figure A-30.
Sodium hydroxide and chlorine are mixed in a reactor system to produce
sodium hypochloride. Glue is added to the solution as an inhibitor until
the mix is viscous; a dilute solution of ammonia (5 to 15 percent) is ad-
ded until a molar ratio of 3NH? to 1 hypochlorite is obtained. This mix-
ture forms chloramine which, when reacted with anhydrous ammonia in a
ratio of 20:1 to 30:1, produces hydrazine. The temperature reaches 130°C.
The effluent from the hydrazine reactor is fed to an ammonia removal
still where excess ammonia is taken off overhead and recycled back to the
reactors. The tails are fed to an evaporator where concentrated sodium
chloride is removed. The vapors from the evaporator are fractionated to
yield (as bottoms) a commercial grade of hydrazine hydrate.
Anhydrous hydrazine may be produced from hydrazine hydrate by extractive
distillation. Hydrazine salts are produced by neutralizing hydrazine
hydrate with the appropriate acid and then dehydrating the resulting
slurry. The salts produced are hydrazine hydrobromide, hydrazine hy-
drochloride and hydrazine sulfate.
-------
DRAFT
The wastewater stream from this process is a sodium chloride solution
from the crystallizing evaporator. Process RWL calculated from the flow
measurements and analyses of the wastewater obtained in the sampling
survey are indicated in the following tabulation. The extreme high
chloride concentration is the wastewater results in an Inhibitory effect
on the BODt; test and, consequently, the analytical results show a high
COD/BODj ratio. The low TOC is due to the lack of organic carbon in-
volved in the process.
PROCESS FLOW
liter/kkg 30,300
gal/M Ib. 3,630
BOD5 RWL
mg/liter1 300
kg/kkg2 9.09
COD RWL
mg/liter1 3,800
kg/kkg2 115
TOC RWL
mg/liter1
kg/kkg2
'Raw waste concentrations are based on unit
weight of pollutant per unit volume of con-
tact process wastewaters.
^Raw waste loadings are based on unit weight
of pollutant per 1000 unit weight of product.
The process RWL shown above was considered as BPCTCA. The salt stream
is combined with other plant wastewaters in a settling pond to remove
suspended solids. The effluent from the pond is discharged to surface
waters.
-142-
-------
DRAFT
to
>
LLJ
U
O
ex.
a.
O
U
O }
*
3
o
LU
z
N
<
at
O
>
I
VINONNV
Nwmoo
NOIimiiSIQ
BNIZVilOAH
3NiHVd01HO
UOIOVHU
CO C^
<=> z
oc
o
>- «:
a: ra
3E <=>
x
CD o= o:
-143-
-------
DRAFT
Product: Isobutylene
Process: Extraction with Sulfuric Acid from a Mixture of C, Hydrocarbons
Process RWL Category: C
High-purity isobutylene is required for applications such as the pro-
duction of butyl rubber and the alkylation of aromatics. In the U.S.,
the most common process for manufacturing pure isobutylene is extraction
with 65 percent sulfuric acid. The feedstock is usually a mixed C^ cut
from a refinery. The process is continuous and is characterized by ex-
tensive contact between water (as H2SO^) and hydrocarbons. As such, it
is assigned to process Category C (Aqueous Phase Reactions). It should
be noted that regeneration of the sulfuric acid absorbent is considered
as an integral part of the process.
A flow diagram for the extraction process is shown in Figure 4-31. A
mixed C^ hydrocarbon feedstock is drawn from the feed tank and passed
through two-stage countercurrent absorbers using 65 percent H2SOi|. The
extract from the rich-stage absorber contains sulfuric acid and iso-
butylene. Steam is injected into the extract to flash off light hydro-
carbons in the extract vent drum. The liquid phase is sent from the vent
stream to the acid regenerator, where isobutylene is separated from sul-
furic acid by heating. The heat is supplied by contact steam injected
into the bottom of the tower.
Dilute sulfuric acid (45 percent) is withdrawn from the bottom of the
regenerator and recycled after reconcentration. The regenerator off-gas,
containing isobutylene and light polymer plus t-butyl alcohol, is scrub-
bed with caustic and water washed prior to being stored in the crude
isobutylene storage tank.
Alcohol by-product is withdrawn from an accumulator between the caustic
scrubber and water scrubber. The bottoms from the water scrubber are
reused as make-up on the caustic scrubber. The aqueous bottoms from the
caustic scrubber are discharged through an alcohol stripper, which re-
moves alcohol by injection of contact steam. The stripper alcohol is
passed back through the caustic scrubber.
Final purification of isobutylene is accomplished in a drying tower,
where water is discharged overhead, and a rerun tower, which takes iso-
butylene at greater than 99 percent purity overhead.
The only continuous contact wastewater stream and significant source of
water pollution associated with the isobutylene flow in the process
(shown by darkened lines in the flow diagram) is the alcohol stripper
bottoms. Aqueous batch dumps are also taken periodically from:
1 . The C, hydrocarbon feed tank.
2. Knockout drums following the water scrubber.
3. The crude isobutylene tank.
-------
DRAFT
The overhead water from the drying column amounts to a very small inter-
mittent discharge.
The remaining process contact wastewater streams are associated with by-
product and recycle streams and sulfuric acid regeneration. Unabsorbed
normal C/j hydrocarbons from the lean-stage absorber may be scrubbed with
caustic before being used in other processes, or merely stored if sub- ,
sequent processing also involves the use of sulfuric acid (as is done in
the process plant sampled). The aqueous bottoms from the normal Ci^ holding
tank contain residual acid and are drained periodically.
The hydrocarbon vapors from the rich extract vent drum may contain some
isobutylene along with normal Cl» hydrocarbons absorbed by the sulfuric
acid. These vapors are passed through a vent gas scrubber where they are
first scrubbed with caustic and then with water. The overhead vapors from
the vent gas scrubber are passed through a knock-out drum where entrained
water from the scrubber is periodically drained. The vapors are then re-
cycled to the Clj feed tank. The aqueous bottoms from the vent gas scrub-
ber are discharged through the alcohol stripper.
The dilute sulfuric acid from the regenerator is reconcentrated from
kS percent to 65 percent by vacuum evaporation. The evaporators are
equipped with steam heated calandrias (shell and tube surface heat ex-
changers) which evaporate excess water. This water is actually con-
densed contact stripping steam used in the acid regenerator to separate
the absorbed isobutylene from sulfuric acid.
The overhead vapors from the vacuum evaporator are drawn through a surface
condenser, which utilizes non-contact cooling water. The surface con-
densate is discharged through a pipe which has its discharge end sub-
merged in a hot well which is divided by internal baffles. The remain-
ing vapors are condensed in a series of barometric condensers associated
with the steam jet vacuum system.
Reconcentrated 65 percent sulfuric acid is drawn from the vacuum evapo-
rator and recycled to the process. Aside from unabsorbed normal butylenes,
the principal by-products of this process are di-isobutylene and tertiary
butyl alcohol. The former is formed by six polymerization reactions and
the later from the hydrolysis of t-butyl sulfate.
Process RWL for isobutylene extraction are summarized in the following
tabulat ion:
-------
DRAFT
PROCESS FLOW
liter/kkg 20,400
gal/M Ibs. 2,440
BOD5 RWL
mg/liter1 669
kg/kkg2 13.6
COD RWL
mg/liter1 3,150
kg/kkg2 64.1
TOC RWL
mg/liter1 633
kg/kkg2 12.9
'Raw waste concentrations are based on unit
weight of pollutant per unit volume of con-
tact process wastewaters.
2Raw waste loadings are based on unit weight
of pollutant per 1,000 unit weights of product.
This process RWL data was considered as BPCTCA. The wastes from this
process are combined with those from other processing areas and treated
by the activated sludge process before discharge.
-------
u
OL
5
to
O
ca
BL
Q
>:
X
O
Z
X
UJ
I
UJ
Z
CO
o
-lit?-
-------
DRAFT
Product: Isopropanol
Process: Continuous hydrolysis of propylene
Process RV/L Category: C
Chemical Reactions:
s* it *. f* it f* LJ /"* LJ _i_ 7 C Q* LJ Crt ^ ^ LJ f* Ltf* LJ _i_ LJ /") v~ ^ L
Cn2^n2^n ~ trio T / p? nnoUr ' '> truLnLno + n^U > Or
£ H j J £
propylene HSO^ OH
iospropyl
hydrogen iospropyl
sulfate alcohol
Typical Material Requirements
100 kg isopropanol (87% isopropanol azeotrope)
Propylene 820 kg
75% sulfuric acid <1.0 kg
25% caustic 6.5 kg
Benzene < 1 .0 kg
Isopropanol is widely used in two major areas: feedstocks for the pro-
duction of other organic chemicals, and solvents. Organic chemicals which
employ isopropanol as a raw material include acetone, glycerine, isopropyl
acetate, amines, and hydrogen peroxide. The largest single use for iso-
propanol, which accounts for nearly one-half the total production, is
the manufacture of acetone by catalytic dehydrogenation.
As a solvent, isopropanol is used mainly for gums, shellac, and synthetic
resins, competing on a price-performance basis with ethanol. It is also
used widely as a rubbing alcohol.
The process plant visited during the sampling program was operated con-
tinuously except for occasional equipment washings. Since there is con-
tact with aqueous waters the process belongs to Category C. The flow diagran
for the process is shown in Figure 4-32.
The liquid propylene feedstock (65 percent) combined with recycled hydro-
carbons, is absorbed in 75 percent sulfuric acid to form a solution of
diisopropyl sulfate and isopropyl acid sulfate. The reaction takes place
at approximately 400 psig and 140°F. The sulfated hydrocarbon solution
Is converted to an acid solution of isopropyl alcohol, ether, and polymer
by hydrolysis reactions with the addition of dilution water in the
hydrolyzer-stripper. Hydrolyzed reaction products are steam-stripped
from the acid and the vapors are condensed following neutralization with
a caustic solution.
-148-
-------
DRAFT
Dilute acid is returned for reconcentration. The liquid products are
then charged to a distillation column, where isopropyl alcohol is sep-
arated from isopropyl ether. The ether goes overhead while remaining
isopropanol is extracted with recycled water and returned to the crude
storage tank; the ether is then distilled and stored for subsequent sale.
The isopropanol from the other distillation column is charged to the
isopropanol (8? percent azeotrope) distillation column and then to
storage. If a dry isopropanol is required, the azeotropic mixture is '
broken with benzene.
During the sampling visit, isopropanol was being produced at the normal
rate; thus the wastewaters are considered to be typical of the everyday
operation. The major sources of wastewater include pump seals and mis-
cellaneous drips, the crude isopropanol waterwash scrubber, the crude
isopropanol caustic scrubber, the isopropanol azeotrope distillation
step, and the organic (benzene) recovery step. Process raw waste loads
calculated from flow measurements and the total pollutant loadings are
indicated in the tabulation below:
Sample Period #1 Sample Period #2
PROCESS FLOW
Hter/kkg 2,537 2,537
gal/M Ib 30^
BOD5
mg/liter1 kOO 386
Ib/M Ib2 1.01 .979
COD
mg/liter1 1,123 1,132
Ib/M Ib2 2.85 3.13
TOC
mg/liter1 508 530
Ib/M Ib2 1.29 1.31*
Raw waste concentrations are based on unit weight of pol-
lutant per unit volume of contact process wastewaters.
2Raw waste loadings are based on unit weight of pollutant
per 1000 unit weights of product.
Waste streams included in the above calculations are continuous. However,
there are occasional washings which were not sampled during the plant
visit. The organic stripper is cleaned with water once per month. The
-149-
-------
DRAFT
cleaning procedure requires approximately 100,000 gallons of water, equi-
valent to only A gallons per M Ibs of product. Although it is not pos-
sible to specify quantitative values for pollution parameters such as
BODc, COD and TOC for this stream, it is believed that this stream can
either be ignored for the RWL calculations or disposed of by incinera-
tion. Therefore, the RWL presented in the above tabulation can be con-
sidered as representative of the process. Another discontinuous source
of waste is the yearly cleaning of the absorption tower which yields ap-
proximately 1000 Ibs of carbon tar to landfill disposal.
Non-contact wastewaters include cooling water flows. The process uses
approximately TOO kg of once-through cooling water per kg of product. The
treatment of the intake cool ing water consists of bar screening and
chlorination. Boiler blowdown is an additional non-contact flow. The
blowdown is added to the cooling water return system. Hot phosphate
softening is employed to treat the boiler feedwater.
The process RWL data presented were considered as BPCTCA. The wastes
from this plant are neutralized and discharged to the local municipal
treatment plant.
-150-
-------
DRAFT
LL.
O
CN CO
«? £
LJJ Q
o g
"- X
I
O
"9^
Q.
O
a.
O
to
N im m 11 s i o;
M3H1J
U3NIJ3H
-151-
-------
DRAFT
Products ; Organic Solvents Complex
Processes:
Isobutylene - Sulfuric Acid Extraction (Category C)
Sec-Butyl Alcohol - Sulfonation 6 Hydrolysis (Category C)
Methyl Ethyl Ketone - Dehydrogenation of Sec-Butyl Alcohol (Category B)
Acetone - Dehydrogenation of Isopropyl Alcohol (Category B)
Methyl Isobutyl Ketone - Acetone Co-Product (Category B)
Isophorone - Acetone Co-Product (Category B)
Mesityl Oxide - Acetone Co-Product (Category B)
The historical data on wastewater flows and total carbon loss, shown in
Table k-k, were provided by a manufacturer producing the chemicals listed
above. The data are based on weekly composite samples and were normalized
to reflect total weekly production for the complex.
Based on 27 weeks operation, the arithmetic average (mean) wastewater flow
was 3,900 liters/kkg total production (k67 gallons/1,000 Ibs) with a
standard deviation of 1,201 liters/kkg (\kk gallons/1,000 Ibs). The
average total carbon loss was 23-9 kg/kkg with a standard deviation of
10.5 kg/kkg. These data are presented for orientation with no intention
to develop effluent limitations relating to this type of chemical complex.
The manufacturing personnel at this facility were very cooperative during
the sampling survey and each individual process plant appeared to be well
operated with respect to both housekeeping and control measures relating
to water pollution. The spread in the historical data provided can there-
fore be interpreted to be typical of the spread in RWL which will occur
in a well run process plant.
-152-
-------
DRAFT
Table 4-4
Historical RWL Data for Organic Solvent Complex
Flow % Occurrence Total Carbon % Occurrence
gal/1000 Ibs lb/1000 Ibs
322.300 3.570 8.190 3.570
337.200 7.140 9.630 7.140
342.200 10.710 10.830 10.710
348.400 14.290 10.840 14.290
356.000 17.860 12.770 17.860
362.600 21.430 15.240 21.430
372.900 25.000 17.400 25.000
381.000 28.570 17.700 28.570
386.000 32.140 18.140 32.140
415.600 35.710 19.140 35.710
416.400 39.290 21.240 39.290
432.600 42.860 21.280 42.860
432.900 46.430 23.240 46.430
435.700 50.000 23.310 50.000
446.100 53.570 23.340 53.570
453.000 57.140 23.380 57.140
466.500 60.710 25.520 60.710
485.600 64.290 26.560 64.290
490.800 67.860 27.180 67.860
502.000 7L430 27.570 71.430
504.900 75.000 27.810 75.000
508.200 78.570 30.110 78.570
513.400 82.140 34.350 82.140
533.700 85.710 37.870 85.710
535.000 89.290 38.690 89.290
893.500 92.860 45.360 92.860
933.900 96.430 49.670 96.430
-153-
-------
DRAFT
Product: Oxalic Acid
Process; Nitric Acid Oxidation of Carbohydrates
Process RWL Category; C
Chemical Reaction;
C6H12°6 + 6HN03 * 3HOOCCOOH + 6NO + 6H20
nitric nitric
glucose acid acid
Typical Material Requirements;
Basis: 1 ton oxalic acid d?hydrate
Glucose (60 percent) 2,112 Ibs
Nitric Acid (90 percent) 5,135 Ibs
Sulfuric Acid (100 percent) 116 Ibs
The uses for oxalic acid center around its calcium-ion removal and re-
ducing properties; it is used as a laundry "sour", as a bleach for re-
moving iron stains from a variety of materials, and in cleaning compounds.
It also finds use in automobile radiator cleaners, leather tanning and
manufacture, chemical processing, photography, medicinals, dyes, and
inks.
Figure ^-33 is a process flow diagram of oxalic acid production via
nitric acid oxidation of carbohydrates. Carbohydrates (in the form of
corn starch), vanadium catalyst, steam, sulfuric acid, and nitric acid
are added to the reactor. The reaction time is twelve hours and the re-
actors are sequenced so as to continuously provide products for the re-
mainder of the processing equipment. Spent nitric acid is withdrawn from
the reactor and recovered. The reaction products go to a vacuum crystal-
lizer followed by a wringer. The liquid effluent from the wringer goes
to a liquid-solid separation step, in which the solids are recycled to
the crystallizer while the liquid proceeds to an evaporator; the dis-
charge from the evaporator is recycled to the reactor. The crude oxalic
acid crystals from the wringer are redissolved and then filtered. The
solids from the filter are disposed by landfill while the oxalic acid is
recrystal1ized under vacuum conditions; the crystals proceed to a second
wringer; and the liquid from this wringer undergoes a liquid-liquid
separation step, along with liquid from the previous crystallization
step. Some liquid is recycled to the crystal 1izer, while the remaining
portion proceeds to a second liquid-liquid separation step. Some of the
liquid is recycled to the reactor and the remainder goes to an evaporator.
The discharge from the evaporator is recycled to the liquid-liquid sepa-
rator. The oxalic acid crystals from the wringer are dried and then
packaged for sale.
-------
DRAFT
The wastewaters from this process consist exclusively of barometric con-
denser effluents. This effluent was sampled and its flow measured. The
process RWL calculated from the flow measurements and the analyses of the
samples are indicated in the tabulation below:
PROCESS FLOW
liter/kkg 436,000
gal/M Ib 52,300
BOD5 RWL
mg/liter1 3
kg/kkg2 1.31
COD RWL
mg/liter1 10
kg/kkg2 4.36
TOC RWL
mg/liter1 3
kg/kkg2 1.31
Raw waste concentrations are based on unit
weight of pollutant per unit volume of con-
tact process wastewaters.
Raw waste loadings are based on unit weight
of pollutant per 1000 unit weights of product.
The high flows and low concentrations seen above are caused by the vacu-
um system associated with the process. These data were considered as
BPCTCA. The wastes from the process are discharged to a municipal treat-
ment plant.
The alternate route in the manufacture of oxalic acid is via sodium
formate. Sodium formate is produced by the reaction of solid sodium
hydroxide and carbon moxoxide at 200°C and 150 psi in an autoclave.
After the reaction is completed, the pressure is reduced and the tempera-
ture is raised to AOO°C. The sodium formate is converted into sodium
oxalate, which is then precipitated by calcium hydroxide to form calcium
oxalate. Calcium oxalate is further acidified by sulfuric acid to form
oxalie ac ? d.
-155-
-------
DRAFT
CO
II 1
h-
^
C*
O
>-
I
O
rrt
<
U
O
z
co O
CO p
^ <
m 0
o* X
D O
O
£ 9
u
u
Q£
1
Q
U
<
| 1 I
y i
^
S - -
O t ^
W -C H
CJ «a
OE
CD c:
2 >
-a u
^r c
CD Ct
co a
5
J
C
3
3
3
1
«C
>~ z
0= *C
Si
CO 1
z co ae
LU >- z
CO CD 1
^ ,
CD
CD
LU
DC
>J ^ J ,
CJD CC IDT
CD ^ =J =«=
1 CD ^
co ye il
t
PORATOR
LU
t
or
LU
*>: t
LU CD
3: :E
t
CD
z:
1 2
co ^
A
1
LU
C3
-> z
CC.
f
I 1
CD
a= 2
| C_
*S
LU C
1 O
CO «a
^ -
CD
CJ
a:
CJ
or
i
QC
CD
=D
CD
I
«C
£
1
3
3
C
4
-»
3=
LU
> CO
^
3C
LU
CO
CO
CD
1
LU
DC
CD
DC
«X
GO
^
1^
k
^
-
C
<
<
L»
p^^
-|
1
C
C
h
C
L
L
1
CD
cvl
i
c-
2
c^
=:
_J
X3
=3
L
C
t.
^^
C
3
f
1£
1_
«
U
DC
i > i
-
S
'
.LU
a-r
:i
JCD
h
DC
jj
^ -
CD C
D_ 3
KC C
> C
1
o=
UJ
LJ-
I.
Jj
O
I Si
JJ 3
J3 t-
4
0=
1 1
a
*>-
CD
CD
_1
X
O
J
£
f 1
DC
C_3
iC
=:
i
J
-^
v
DC
LU
-a:
^c^
h-
CD
CD
,_^
CD
z. u
CD
CD
QC.
^
^
U_
CD
*t
DC
i i i
C3
DC
%
J
CD
CD
«r
C.D
1
«I
cc
oc
CD
t
1
L
-^
3
C
C
.
CQ CO
CD LJ-I 1
co 3:
DD CD
-K I
"~ ^
r: ^
* i
o t
l-T-
LU
CD CD
Z CD
DC CD
=T£
z -=c
DC 1
I i
CD
C_D
i i
0= iz
1 -el
^Z Q_
-156-
-------
DRAFT
Product: Pentaerythritol
Process: Aldehyde Condensation
Process RWL Category: C
Chemical Reactions:
4HCHO + CH3CHO + NaOH > CtC^OH)^ + HCOONa
sodium sodium
formaldehyde acetaldehyde hydroxide pentaerythritol formate
The most important end use of pentaerythritol is in the manufacture of
alkyl resins, in competition with glycerol. Next in importance are
pentaerythritol resin esters, which are used in floor polish and in
flexographic inks. Other applications of pentaerythritol esters are in
the manufacture of fire-retardant paints, use as high pressure lubricants,
and production of PVC plasticizers of low volatility suitable for use in
wire insulation.
A typical process flow diagram for producing pentaerythritol by aldehyde
condensation is shown in Figure 4-3^.
To a solution of formaldehyde (20 to 30 percent by weight) is added either
50 percent sodium hydroxide or 50 to 80 percent calcium hydroxide slurry
while the temperature is maintained at 15 to 20°C. With suitable agi-
tation, 99 percent liquid acetaldehyde is slowly added under the surface
of the formaldehyde-alkali solution. Since the reaction is exothermic,
external cooling is used to maintain the reaction temperature between 20
and 25°C. The mole ratio of formaldehyde to acetaldehyde generally used
is between 4.5:1 and 5-0:1. A ratio of 1.0 to 1.5 moles of hydroxyl ion
per mole of acetaldehyde appears to be the optimum amount of condensing
agent. The temperature of the reaction mixture is held at 25° to 30°C
for several hours and is then raised to about 60°C until the free aldehyde
content is less than 0.1 percent. Numerous side reactions occur simultane-
ously, mostly other condensations and autocondensations.
The crude reaction mixture is distilled, with tops (mainly formaldehyde)
going back to the reactor and bottoms transferred to the neutralizing
tank, where an acid is added to neutralize the excess alkali and to ef-
fect removal of the metallic ion of the condensing agent. If sodium hy-
droxide is used, formic acid may be added to reduce the pH of the solution
to 7-8 to 8.0 and subsequently to remove the sodium ion present as sodium
formate. If calcium hydroxide is the condensing agent, sulfuric acid or
oxalic acid, either alone or conjunctively, may be employed to precipitate
the calcium ion as calcium sulfate or calcium oxalate. The calcium salt
is removed by filtration.
-157-
-------
DRAFT
The solution is then evaporated to a specific gravity of about 1.27. It
is chilled to crystallize pentaerythritol, and the resulting slurry is
filtered. The mother liquor goes to a recovery system, where it is re-
worked. If sodium hydroxide acted as the condensing agent, the sodium
ion is removed as sodium formate during this operation. If the calcium
process is used, the calcium ion is removed in the previous step by
filtration.
The filter cake contains pentaerythritol and polypentaerythritols. The
latter materials are formed by side reactions and are a mixture of ether-
linked polymers such as dipentaerythritol and tripentaerythritol. Other
by-products found in the reaction liquors include both linear and cyclic
formals of the various pentaerythritols. The amounts of these poly-
pentaerythr i tols and formals formed vary with reaction procedures and
may be kept at a minimum under proper reaction conditions. The filter
cake is then redissolved in hot water and passed through an ion-exchange
purification unit which removes the last traces of formic acid. The con-
centration of this solution is sufficient; thus another evaporation step
is unnecessary. Deionization is followed by vacuum crystallization,
filtration, and drying. The yield is a technical product containing
about 11 percent dipentaerythritol. Some producers treat part of the
product further to obtain both products in pure form.
The major pollution sources of the process are residue from the distil-
lation column and mother liquors withdrawn from filtration units. The
condensates from the steam jets which are used to pull vacuum are re-
cycled back to the process, and are not considered wastewaters. The
process RWL calculated from the flow measurements and analyses of waste-
water samples obtained in the sampling period are presented in the fol-
lowing tabulation:
-158-
-------
DRAFT
PROCESS FLOW
liters/kkg 10,200
gal/H Ibs 1,220
BOD5 RWL
mg/liter1 38,100
kg/kkg2 390
COD RWL
mg/liter1 155,000
kg/kkg2 1 ,580
TOC RWL
mg/liter1 81,200
kg/kkg2 830
'Raw waste concentrations are based on
unit weight of pollutant per unit volume
of contact process wastewaters.
2Raw waste loadings are based on unit
weight of pollutant per 1000 unit weights
of product.
The above process RWL are considered as BPCTCA. The wastewater from
this process is combined with other wastes for treatment in an activated
sludge plant. The plant effluent is discharged to solar evaporation
ponds.
-159-
-------
DRAFT
o
LU
0=
£ *«-
o s
i «
io °
z
UJ
O
Oc
c
u ~z
«
III "
"J «sc c_
J CD tt
13= L!
II 1 <=> =
JJ* Ll_ c/-
<
1
1
0
l_
h-
r*
UJ
<
1
Z
LU
Q_
CD
J
^^^H
CD
9 -a:
c
CO
= «c
J X
_ CD
3 0=
) CD
4
C_3
C
S
_*
^
r
car
UJ
2
C
1-
«
r-
<
c-
>
Q
C_
1
doiva
:
i
i
u_ .
1
z
2 C=
_ 1-
1 -a
-I i>-
- cc
CO H-
LU ^
CC Ll.
0- 2
*^.
NOll
4 < A
LU
c:
>-
-U z
3
C
J
J
J
C
>
3
OdVI
'
=E
=>
X
3=
f
3
1
)
J
s
'
NW
VII
,
O
CO
«a:
LJ_I
cc
j
^
^^^M
m
1!
^
n
*
^^«
c
f-
h-
:
C-
1
U
M
1
1
J
D
_1
C
0
^
H
1
LJ_
C3
z:
«c
=3
c:
^c<
LJJ
cr:
E
=3
CO
«et
CD
|"
|_ aaz
E
^
CO
<=>
t
1
CD
t
cc
LlJ
LU
LU
Q=
cn
4
or
UJ
o= *r
^ ^^
11 i
00 CD
U- *E |_
1 > or
DC
0^
TltfiSAHO «
LU| 1 en
"*l 1 ^
, . ^ ,,.... UJJNVllJAJ .J
uu NO, i
CD ^ Z
C/O B^^^^^rf ^^
d h
- t
. _ 1
Q=
111
ce TO « a^
LU ^ >" LU
« I c= t=s ^t
1 ^^' CD ^ ^^^^CO
co u_i -a:
u- 3: or 3:
' '
-160-
-------
DRAFT
Product: Propylene glycol
Process: Hydrolysis of propylene oxide
Process RWL Category: C
Chemical Reactions:
Propy1ene
oxi de
H20
water
CH CHOHCH?OH
propylene glycol
Propylene glycol is by far the most widely used difunctional alcohol for
the manufacture of unsaturated polyesters. This end use consumes nearly
one-half of the propylene glycol produced in the United States.
Propylene glycol is also used as a cellophane plasticizer. This market is
relatively new, since propylene glycol is a substitute for ethylene glycol,
which has recently been deemed toxic. Toxicity regulations also deter-
mine the use of propylene glycol in tobacco. Miscellaneous outlets for
propylene glycol include uses as a binder for cork bottle caps, as a
humectant in cosmetics, and as an intermediate for propylene carbonate.
Production of propylene glycol is based on the liquid-phase hydrolysis
of propylene oxide, and is shown as a flow diagram in Figure ^-35. The
hydrolysis reaction occurs at an elevated temperature and pressure in the
presence of a sulfuric acid catalyst.
By selection of the ratio of feedstock propylene oxide to water, 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
formation of undesirable by-products.
Reaction products are passed from the reactor to a series of evaporation
and drying towers to remove excess water. This excess vapor is often dis-
charged to the atmosphere, although it could be recycled to the reactor.
After passing through the drying operation, the crude propylene glycol is
sent to a series of fractionators. The first tower removes water and traces
of the light ends. Mono-propylene glycol is then separated, condensed, and
stored as an industrial grade product. The bottoms from the mono-propylene
glycol tower are sent to the crude di-propylene glycol storage facilities.
-161-
-------
DRAFT
Crude di-propylene glycol is then vacuum-distilled. The tops are recycled
to the reactor hydrolyzer to dry the materials further and to recover
glycol product. Bottoms leaving this first column are further distilled.
Food-grade di-propylene glycol is obtained as bottoms from the second
distillation step; the tops are then further distilled to separate
additional mono-propylene glycol. Both light and heavy ends are waste
products.
Wastewaters from the production of propylene glycol are generated during
the evaporation-drying operation and the distillation processes. Steam
ejectors and barometric condensers are employed in most of the distilla-
tion columns. During the plant visit, samples of these contact process
wastewaters were obtained. Process raw waste loads calculated from flow
measurements and analyses of these streams are indicated in the tabulation
below:
PROCESS FLOW
liter/kkg 5,500
gal/M Ib. 660
BOD RWL 1
^ mg/liter 3
kg/kkg .016
COD RWL 1
mg/1i ter 10
kg/kkg2 .055
TOC RWL 1
mg/1i ter 1
kg/kkg2 .006
Raw waste concentrations are based on unit
weight of pollutant per unit volume of
contact process wastewaters.
2
Raw waste loadings are based on unit weight
of pollutant per 1000 unit weights of product.
These RWL values are not considered to be truly indicative of the process
because all of the steam from the ejectors is discharged to the air and
is not condensed. For this reason, it is not possible to recommend
effluent limitations for this process.
-162-
-------
DRAFT
RWL data obtained from a plant survey are summarized in the following
tabulation. Since the plant surveyed is designed strictly for manufac-
turing of one plasticizer (diethyl phthalate), it requires less frequent
reactor clean-up. Consequently, both flow and RWL of this process are
not as high as expected.
PROCESS FLOW
liter/kkg 653
gal/Mlbs 78.3
BOD- RWL
^ mg/1 82,600
kg/kkg^ 53.9
COD RWL 1
mg/1 127,000
kg/kkg2 82.6
TOC RWL 1
mg/1 51,200
kg/kkg 33.^
Raw waste concentration are based on unit
weight of pollutant per unit volume of
contact process wastewaters.
2
Raw waste loadings are based on unit weight
of pollutant per 1000 unit weights of product.
These data were considered as BPCTCA.
-163-
-------
DRAFT
Q
X
O
o
u_
o
O Q
o
(J
o
UJ
z
O-
o
H3M01
3NIHSINN
T03A13
-164-
-------
DRAFT
Product: Propylene oxide
Process: Chlorohydrin Process
Process RWL Category: C
Chemical Reactions:
HOCl
propylene hypochloric acid propyl
chlorohydrin
2 HOCH2CH2CH2C1 + Ca(OH)2 - ^ 2 CH3CH - CH2 +
^0'
propyl lime propylene oxide calcium water
chlorohydrin chloride
The most important outlet for propylene oxide is in the manufacture of
propylene glycol. It is also used during the production of polyethers,
or which are in turn used in the manufacture of urethane forms and elast-
omers,
The manufacture of propylene oxide from propylene is similar to that of
ethylene oxide by the chlorohydrin route. In fact, many propylene oxide
manufacturing facilities are converted ethylene oxide plants which were
rendered obsolete by the advent of direct oxidation. In recent years,
considerable research has been undertaken to discover a viable process
for making propylene oxide without using chlorine. A number of alter-
natives have been developed, including epoxidation of propylene by means
of a hydroperoxide and direct oxidation of propylene.
.'reduction of propylene oxide by the chlorhydrin process involves a
reaction between propylene and chlorine, and the process flow sheet is
shown in Figure 4-36. The raw materials are fed into a reactor and
water is added. The reaction products are primarily chlorohydrin and
dischlorohydrin. Vent gases from the reactor are passed through a
water scrubber. A second stage caustic scrubber is also employed to
neutralize potential acid carry-over in the off-gases.
Following the caustic scrubbing, the gases are passed to an oil absorption
unit where propylene dichloride is selectively concentrated and recovered.
(An activated carbon adsorption system may be substituted for the oil
absorption unit). Other gases, such as propylene and propane are often
vented to a fuel gas supply. At some installations, pure propane may
be recovered in a dehydrator (activated alimina).
-165-
-------
DRAFT
Reaction products, containing mainly propyl-chlorohydrin, are sent to
a saponification reactor. A lime slurry is fed into the reactor along
with live steam. In the reactor, propyl-chlorohydrin is converted to
propylene oxide. Dichlorohydrin is also converted to propylene dichloride,
which may be recovered as a by-product. Following the lime addition,
the products are passed through a stripper. The products (propylene oxide
and propylene dichloride) are separated from unreacted lime and calcium
chloride. The unreacted lime solution is passed through a clarifier
from which the underflow, containing unreacted lime, is recycled back
to the saponitrication reactor; the overflow is discharged as wastewater.
The product lime is then passed through a recovery tower, where propylene
oxide is separated. The underflow, containing propylene dichloride, is
sent to a stream stripper where propylene dichloride is separated from
other impurities. The propylene dichloride is combined with the propylene
dichloride separated from the off-gases. Bottoms from the stripper con-
stitute a second major waste stream.
Production facilities for propylene oxide were visited during the
sampling period, and samples of the contact process wastewaters were
obtained. Process raw waste loads calculated from flow measurements
analyses of these wastewaters are indicated in the tabulation below;
and
Plant 1
Sample
Period #1
Sample
Period #2
Plant 2
Sample
Period #1
Plant 3
Sample
Period #1
PROCESS FLOW
1iters/kkg
gal/M Ib.
BOD,- RWL
(fig/1it«
kg/kkg2
COD RWL
mg/1iter
kg/kkg
TOC RWL
mg/1iter
kg/kkg
1
60,000
7,150
290
17.2
2,480
310
18.6
50,200
6,020
575
28.9
2,7^0
138
320
16.0
69,300
8,300
480
33.2
1 ,680
116
365
25.3
66,000
7,910
578
38.1
2,580
170
385
25.4
NOTE: 1
Raw waste concentrations are based on unit weight of pollutant
per unit volume of contact process wastewaters
"Raw waste loadings are based on unit weight of pollutant per
1000 unit weights of product.
-166-
-------
DRAFT
A plant average of the values presented above was considered as BPCTCA.
The wastewaters from all three plants are combined with wastes from
other processes and pumped to settling basins from which they are dis-
charged to surface receiving waters. The high flows may be partially
explained by the fact that a large quantity of water is required in the
reactor and also in the many scrubbers. Furthermore, a large quantity
of stichiometric water is formed during the reaction.
It is possible that both pollutant loadings and flows may vary at other
production facilities. The pollutant loadings may be greatly increased
if propylene dichloride is not recovered following separation of the
propylene oxide. At such an installation, the flows would probably be
lower since the propylene dichloride stripper would not be required.
An additional feature of the process plant visited during the sampling
period which may differ at other installations relates to the recycling
of unreacted lime. A clarifier had been installed to remove lime and
calcium chloride from the wastewater. These materials were then re-
cycled to the reaction products, thus reducing the total flow and the
pollutant loadings. This was considered as part of the process since
required materials were recycled.
-167-
-------
DRAFT
to
c/)
LU
U
O
Of.
Q.
Of.
a
o
CO
0* X
D U
O -
"- x
O
a.
O
of.
Q.
O- CD LU
-168-
-------
DRAFT
Product: Saccharin
Process: Synthesis from Phthalic Anhydride Derivatives
Process RWL Category: C
Saccharin is the imide of the mixed anhydride of o-carboxyIbenzenesu Ifonic
acid. It is a powerful sweetening agent, having a sweetness from 550 tp
750 times cane sugar. Saccharin has no food value and is used only when
it is desirable to reduce the consumption of carbohydrates.
A continuous process for the manufacture of saccharin from phthalic anhy-
dride derivatives is shown in Figure 4-37.
The phthalic anhydride derivatives along with water and a 50 percent sodium
hydroxide solution enter the reactor. A metal catalyst is used during the
reaction. The fumes leaving this reactor are hydrochloric acid vapor,
nitrogen, and sulfur dioxide. These fumes are scrubbed with water and
sodium hydroxide. The resultant wastewater flows to a neutralization tank,
where caustic is added to raise the pH to approximately 6.5
The product mixture from the first reactor is then chlorinated in the second
reactor; and the fumes from this reaction include chlorine, hydrochloric
acid, and water vapor. These fumes are also piped to the aforementioned
caustic scrubber. A portion of the metal catalyst is removed from the
second reactor for regeneration. The catalyst is treated with sodium hy-
droxide and then filtered. The filtrate goes to the neutralizer, while
the catalyst is recycled to the first reactor.
The products from the second reactor are then reacted with NH3 in the third
reactor, with any unreacted ammonia vented to an absorber where the ammonia
is absorbed with water and returned to the third reactor. An organic sol-
vent is used to extract the product saccharin from the reaction mixture.
The raffinate stream from the extraction operation is first neutralized
and then distilled to recover the solvent. The extractive phase is steam
stripped to recover solvent, which is condensed at a scrubber and recycled
back to the third reactor. The aqueous solution leaving the stripper
is precipitated by the addition of hydrochloric acid. The precipitate is
further concentrated by filtration and crystallized under vacuum conditions
and then dried to produce saccharin. The filtrate is neutralized by a caustic
solution prior to its discharge to the sewer.
The major water pollution sources of the process are the wastewaters from
caustic scrubber, catalyst filtration unit, solvent recovery still, pro-
duct filtration unit, and barometric condenser. Process RWL's calculated
from flow measurements and analyses of the wastewater streams are presented
in the following tabulation. The analytical results also indicate that,
in addition to the parameters shown in the tabulation, pollution parameters
such as nitrogen, sulfate, chloride, and metals are at levels hazardous to
the biological treatment process.
-169-
-------
DRAFT
PROCESS FLOW
liter/kkg 269,000
gal/M lb 32,200
BOD5 RWL
nig/liter 945
kg/kkg2 254
COD RWL
mg/liter1 3,270
kg/kkg2 879
TOC RWL
mg/liter1 1,430
kg/kkg2 384
Raw waste concentrations are based on unit weight of pollutant
per unit volume of contact process wastewaters.
2
Raw waste loadings are based on unit weight of pollutant per
1,000 unit weights of product.
-170-
-------
DRAFT
to
LLJ
O
LU
O
ac
O
>-
I
z
<
u
W)
LU
I
I
to
X
u
u
<
CO
-------
DRAFT
Product : Secondary Butyl Alcohol
Process : Sulfonation and Hydrolysis of Mixed Butylenes
Process RWL Category: C
Chemical Reactions;
CH-CH = CHChL + H2SO,4 - > CH-CH9CHCH_
Butene-2 Sulfuric Sec-Butyl
Acid Sulfate
CH CH2CHCH + H20 - > CH CH2CH(OH)CH3 +
Sec-Butyl Water Sec-Butyl Sulfuric
Sulfate Alcohol Acid
Secondary butyl alcohol is made from mixed butylenes. However, because
of its alternate uses, isobutylene is normally extracted from the C,
feedstock prior to the manufacture of secondary butyl alcohol. This is
also done to prevent the formation of excessive quantities of tertiary
butyl alcohol from isobutylene.
Normal butylenes are first absorbed by concentrated sulfuric acid (75
percent) to form isobutyl sulfate. This is subsequently hydrolyzed with
water to secondary butyl alcohol and dilute sulfuric acid. A flow dia-
gram for the process is shown in Figure 4-38. The extensive requirements
for contact water usage make this continuous process typical of Category
C. The reconcentrat i on of sulfuric acid for recycle is considered an
integral part of the process. It should also be noted that this process
for secondary butyl alcohol is analogous to that used to produce isopropyl
alcohol from propylene.
As shown in Figure 4-38, the normal C, hydrocarbon feed is passed into
multiple countercurrent absorbers using 75 percent sulfuric acid. The
absorption and the i sobuty 1 -sul fate-f ormi ng reaction are strongly
exothermic and require large volumes of non-contact cooling water.
The rich extract from the first-stage absorber is passed through an olefin
soaker, which acts as an absorption stage for the recycle Ci olefins
that have not been converted to secondary butyl alcohol.
The mixture of sulfuric acid and isobutyl sulfate is combined with hydro-
lysis water and enters the hydrolysis generator, where crude secondary
butyl alcohol and regenerated C, olefins go overhead and dilute acid is
taken off as bottoms. The acid is sent to regeneration by vacuum evap-
oration, while the alcohol and olefins are separated in a combination
-172-
-------
DRAFT
scrubber-stripper. The hydrolysis reaction is also highly exothermic
and requires large quantities of non-contact cooling water.
Direct-contact steam is injected into the bottom of the stripping section
of the scrubber-stripper to remove the more volatile olefins and alcohol
from the excess hydrolysis water. This aqueous waste stream is withdrawn
as bottoms and combined with other aqueous wastes as shown in Figure 4-38.
Product secondary butyl alcohol as a 70 percent alcohol/30 percent water
mixture is drawn off as a side stream and C^ olefins are taken overhead.
These olefins are scrubbed with water to recover any entrained alcohol
in the olefin scrubber, with the aqueous bottoms returned to the scrubber-
stripper. The regenerated C, olefins are then returned to the process
via the olefin soaker described previously. The olefin recycle line is
equipped with a knockout drum to remove water entrained from the olefin
scrubber.
The unextracted Cif raffinate is taken from the fourth-stage absorber,
combined with aqueous caustic, and sent to a scrubber-stripper. Caustic
addition is necessary to neutralize excess acid from the absorbers.
Contact steam is injected into the lower section of the raffinate scrubber-
stripper to drive saturated C/^ hydrocarbons overhead. The aqueous
bottoms contains a di-isobutylene by-product, which is decanted from
the water layer in a settler.
The saturated C^ hydrocarbons are then sent to a rerun tower to remove
additional dimer by-product. The rerun tower bottoms contains both water
and dimer, which are separated by decantation in a second settler. The
organic dimer layer from both settlers is combined, as are the aqueous
layers.
The major wastewater streams associated with the flow of hydrocarbons
within the process (shown by darkened lines in Figure 4-38) are:
1. Crude secondary butyl alcohol scrubber-stripper bottoms
2. Water layer from regenerated olefin knockout drum
3. Water layer from raffinate scrubber-stripper bottoms
k. Water layer from rerun tower bottoms
5. Water layer from rerun tower overhead
Except for the second stream, all of these wastewaters are combined and
sent to a holding drum. This tank discharges to the wastewater treatment
plant, with a small drawoff to provide a seal for the process flare stock.
The weak acid bottoms from the hydrolysis generated are reconcentrated
by multi-stage vacuum evaporation. Normally, three evaporators in series
are used to bring the acid concentration back to 75 percent strength.
However, during the sampling program, only two were in service, with the
final stage down for cleaning. This type of operation is possible by
pulling a higher vacuum on the first two evaporators.
-173-
-------
DRAFT
The overhead vapors from the first evaporator are drawn through a surface
condenser which utilizes non-contact cooling water. The non-condensible
vapors are entrained in two barometric condensers operating with steam
jet vacuum pumps in series. The second- and third-stage evaporators are
also equipped with this condenser-steam jet arrangement.
Apart from dehydrogenation to butadiene, secondary butyl alcohol pro-
duction is the main end use for normal butylenes. Secondary butyl alcohol
itself is used mainly as a solvent or to make secondary butyl acetate or
methyl ethyl ketone. It should be noted that the alcohol product is ob-
tained as an aqueous solution in this process. The product is really an
azeotropic binary mixture drawn as a side stream from the crude secondary
butyl alcohol scrubber-stripper. It contains approximately 70 wt percent
(36 mole percent) alcohol and 30 wt percent (64 mole percent) water, and
is withdrawn at a minimum boiling point of approximately 90°C. Dehydra-
tion of the alcohol is accomplished in subsequent processes, and this
wastewater is not included in the wasteload computed for the alcohol
process.
During the field data collection program, two plants utilizing the
previously described process were sampled. The following brief tabula-
tion summarizes the RWL for each plant:
Plant 1 Plant 2
PROCESS FLOW
liters/kkg 64,900 626
gal/M Ib
BOD5 RWL
mg/liter1 374 22,800
kg/kkg2 24.3 14.2
COD RWL
mg/liter1 3,280 62,000
kg/kkg2 213 38.8
TOC RWL
mg/liter1 665 38,300
kg/kkg2 43.2 23.9
Raw waste concentrations are based on unit weight of pollutant
per unit volume of contact process wastewaters.
2
Raw waste loadings are based on unit weight of pollutant per
1,000 unit weights of product.
-174-
-------
Z
LLJ
Q
LJJ
X
u_
o
c/5
to
_J
O
Of
Q
TO X
Q
^ Z
LJJ <
O Z
o o
z
o
O
x
O
I
U
y^
«3a»osa»
33»1S 1SI
3
C
:
t
*^s
1
5
^
*
I
>
cc
iz!
aoiva3N3^
S ISA10HUAH
1 ' co
= £
±
83KVDS
NN310
i
*
-------
DRAFT
Product: Citric Acid
Process : Fermentation of Molasses
Process RWL Category: D
Reaction :
CHO or CH0 + 0
sucrose or dextrose water _ . citric carbon water
acid dioxide
Typical Material Requirements
1000 kg citric acid
Molasses ^000 kg
Nutrients 5 to 15 kg
Sulfuric Acid (95%) 7000 kg
Lime 5000 kg
Citric acid is one of the widely employed organic acids. Its major use is
as an acidulant in carbonated beverages, jams, jellies, and other foodstuffs.
Another important use of citric acid is in the medicinal field, including
the manufacture of citrates and effervescent salts. There are a few in-
dustrial uses, including citric acid as a sequestering agent, and acetyl
tributyl citrate as a vinyl resin plasticizer.
Except for small amounts produced from citrus fruit wastes, citric acid is
manufactured by aerobic fermentation of crude sugar or molasses. The fer-
mentation changes sugar, a straight-chain compound, into a branched chain.
Production of citric acid may proceed in either of two methods: fermentation
in shallow pans, or fermentation in aerated tanks. Both processes may be used
simultaneously as shown in Figure 4-39. In the tray process, a sugar solution
is placed in the trays, and air is circulated for 9 to 12 days over hundreds
of shallow, pure-aluminum trays. The trays are placed in a closed cabinet
provided with facilities for sterilization, ventilation, and temperature control.
After the fermentation, the mycelium is removed from the broth by screens
or filters, and the spent mycelium is washed and disposed. Mycelium wash-
water containing citric acid is combined with acid separted in the filtration
step. The citric acid then passes through an oxalic acid recovery stage.
(Oxalic acid is available for recovery only from the tray process.) The
addition of calcium sulfate precipitates calcium oxalate. Addition of sul-
furic acid recrystal 1 izes the calcium sulfate, and the solids are removed by
filtration. Hydrous oxalic acid is then crystallized under a vacuum pulled by
a barometric condenser.
-176-
-------
DRAFT
After removal of the oxalic acid, the broth is combined with the liquor pro-
duced in the deep tanks. The essential steps of the deep tank process are
similar to the tray process, but a different type of mycelium culture is
grown and oxalic acid is not generated as a by-product. Molasses and water
are first introduced into the deep fermentation tanks. Air is introduced to
maintain aerobic conditions and to keep the tank contents well mixed. The
fermentation process is complete after approximately k to 9 days at 30 to
32° C. The tanks are then emptied and partially filled with boiling water
to sterilize them prior to the next batch. The crude liquor is filtered to
remove the mycelium, and the broth is combined with the broth produced ifi the
tray processes.
Citric acid is recovered from the broth by precipitation with calcium hydrox-
ide. The solution is then passed through a series of filters for removal of
the crystalized calcium citrate. Both filtrate and washwater used during the
filtration process are discharged as waste. The calcium citrate is then chem-
ically reacted with sulfuric acid to form calcium sulfate and citric acid.
The calcium sulfate is removed by filtration to purify the citric acid. Some
of the filter cake is recycled to the oxalic acid precipitation process, while
the remainder is wasted. The crude citric acid is then concentrated from
30 to 60 percent in a double-effect evaporator equipped with two-stage steam
jet to pull the vacuum. Barometric condensers are employed following the
steam jets. Since the citric does not vaporize, the only loss in the evaporator
is by entrainment; thus, entrainment separators are used prior to the con-
densers.
The crude crystals are redissolved in water during the finishing step. Treat-
ment processes, including granular activated carbon, are employed to remove
trace heavy metals and color. The white liquid is then crystallized
(utilizing a barometric condenser), dried, and packaged for sale.
To determine the raw waste load, samples of the various wastestreams were
obtained. These wastestreams included tray and deep tank wash waters, filtrate
from the calcium citrate filtration step, and barometric wastewaters from the
purification steps. A wastewater stream (barometric condenser wastewater) re-
sulting from the purification of oxalic acid was included in the raw waste load;
thus, production is expressed as the combined total of oxalic and citric acids
in their anhydrous forms. Process raw waste loads calculated from flow
measurements and the analyses of these streams are indicated in the following
tabulation:
Oxalic plus Citric Acids (Anhydrous form)
Process Flow
liter/kkg 477,000
gal/M Ib. 57,200
BOD5 RWL
mg/liter 690
kg/kkg2
-177-
-------
DRAFT
Oxalic plus Citric Acids (Anhydrous form)
COD RWL 1
mg/liter 1,380
kg/kkg2 657
TOD RWL 1
mg/liter 507
kg/kkg2 242
Raw waste concentrations are based on unit weight of pollutant per
unit volume of contact process wastewaters.
2
Raw waste loadings are based on unit weight of pollutant per 1000
unit weights of product.
The foregoing data were considered as BPCTCA. The wastes from this
plant are discharged to surface water without treatment.
The analytical results also indicate that, in addition to the parameters
shown in the tabulation, pollution parameters such as sulfate, nitrogen,
chloride, calcium, magnesium, and zinc are at levels hazardous to bio-
logical treatment processes. Proper pretreatment to reduce the afore-
mentioned parameters is necessary before the wastestreams can be dis-
charged into any biological treatment unit.
Non-contact wastewaters include cooling water and stream condensate. The
total cooling water usage is approximately 107,000 gallons per 1000 1bs of
product. A large quantity of the cooling water is employed in tube-and-shel1
heat exchangers. The total steam usage (live plus reboiler) is estimated to
be 15,500 Ibs. per 1000 1bs. of product.
-178-
-------
DRAFT
LU
I
0
o
-z.
O
^ LU
LU oL
Q£ tu
§8
U_ <£
U
<
x
n
Q
u
u
QL
iz
u
,.
\i
1 LJ
ce on
I
:*:
*
Q_
LU
t O
3t CC
1 UJ
CO £
* ce
s
t "
L__
r
O
ri
UJ
H>
1
1
o
CO
_J
1
^ LI.
1 0
C S^J
- w *
^ Mil. IIP Ul
UJ t
ce *x
u.
"Z.
D
U ^
ce
LU
PVJ
UJ 1
C3 CO
ce cc.
\
z.
0
t "Z.
o
Z CO
t-3 » Q. ca
CJ
tr:
Zi.
Q
0 "
u! ce |
UJ ^ "
co *t T
s° 1
CC
r-4
3= <£
CO t
ce
A .
£ ]
z.
Z -I LU ^
U- CD I U-
^
tf
Tt
179
-------
DRAFT
Products: Citronellol and Geraniol
Process: Distillation of Citronella Oil
Process RWL Category: D
Chemical Structure:
(CHJ2 C = CH (CH2)2- CH CH2CH2OH
c i tronellol
(CH )2~C = CH(CH2)2-C = CHCH2OH
Geran iol
Citronellol and Geraniol are used as odorants in perfumery. Natural
geraniol is produced by the distillation of citronella oil; citronellol is
made by limited hydrogenation of geraniol. Figure ^-40 is a process flow
diagram of geraniol and citronellol production. Citronella oil, the raw
material, is vacuum distilled to separate citronellol. The remaining
citronella oil components undergo a three-stage potassium hydroxide
washing prior to entering into the vacuum steam distillation step,
followed by a vacuum distillation. In this step, geraniol heavies and
citronellol are recovered; the geraniol fraction is washed with a caustic
solution and then water, and the geraniol terpene fraction undergoes boration
and washing prior to being rerun through the vacuum distillation step.
As shown in Figure k-kO, the major wastestreams from the process are vacuum
jet condensates and wastewater from product washes. The following tabulation
summarizes the RWL calculated for the process:
Flow 10,000 1 iters/kkg
BOD. 58.1 kg/kkg
COD 111 kg/kkg
TOC 37.7 kg/kkg
The foregoing values were considered as BPCTCA. These wastes are dis-
charged to a sewage treatment plant.
-180-
-------
z
O
z
§ °
T «
^r t
LU U
e* i
O O
ul z
o
Q
z
Z
O
£*
t
U
DRAFT
js r
LL
r
*
A
T
1
_J C
1 OK
UJ
CO C= I 1
II
Nwmoo
Noiiniiisio
U-l UJ
oe co
cc
J
Nwmoo
NOIHmilSIQ
J
Lid
Noiivniisio
I
181
-------
DRAFT
Product: Dyes and Dye Intermediates
Process: Batch Chemical Reactions
Process RWL Category: D
Dyes may be defined as intensely colored substances which, when applied to
a substrate, impart color to this substrate by a process which, at least
temporarily, destroys any crystal structure of the colored substances.
The dyes are retained in the substrate by adsorption, solution, and mechanical
retention, or by ionic or covalent chemical bonds. Pigments, on the other
hand, are usually applied in vehicles (although the substrates themselves
may serve as vehicles, e.g., in the mass coloration of polymeric materials),
and retain, to some degree, their crystal or particulate structure.
The color of a dye is due to electronic transitions between molecular
orbitals of the molecule, the probability of these transitions determining
the intensity of the color. The energy differences between the orbitals
determine whether the "color" falls in the visible range of the electro-
magnetic spectrum and, if it does, the precise shade or hue. Only organic
molecules of considerable complexity, containing extensive conjugated
systems and polar or semipolar substituents, are useful as dyes.
Much of the complexity of present-day dye technology arises from the great
diversity of materials to be dyed, such as foods, drugs, cosmetics, waxes,
greases, solvents, plastics, rubber, photographic film, leather, fur, paper,
and, primarily, textiles. In the textile field, the introduction of syn-
thetic fibers, such as cellulose acetate, nylon, polyester, acrylics, cellulose
triacetate, and polypropylene, as well as the more stringent fastness re-
quirements and the continuous-dyeing techniques, have presented the dye manu-
facturer and dyer with new problems. These problems cannot be separated
from one another; for example, the new polyester fibers are sometimes dyed
by a continuous heat-treatment process requiring new dyes with a new fastness
property, sublimation fastness; and a special physical form. Dispersed dyes,
for instance, are more hydrophobic than those used previously for cellulose
acetate; they must be nonvolatile (i.e. must not sublime off the fiber during
the heat treatment) and must have a physical form, either powder or paste, which
gives rapid and stable dispersions in the dye bath.
Dye intermediates are derived from a wide variety of aromatic organic com-
pounds, such as benzene, naphthline, anthracene, higher polycyclic derivatives,
and hetrocyclics. The United States Tariff Commission lists some 230 com-
pounds under the heading "cyclic intermediates", of which more than 210
are used in the dye industry. Many of the large-volume intermediates have
a principal use outside the dye industry. For example, about 60% of the
aniline produced is used in the rubber industry, and practically the entire
phenol and phthalic anhydride production is consumed by the plastics industry.
Originally, however, all three of these materials were dye intermediates ex-
clusively.
-182-
-------
DRAFT
The dyes themselves are usually much more complicated than the inter-
mediates from which they are derived. Some dyes are mixtures, while
others (such as aniline and sulfur colors) are still of unknown structures.
Therefore, systematic chemical names are rarely used. Almost all dyes
also have a multiplicity of trade names in addition to their common
names. Despite the diversity of trade names, a certain amount of rationality
can be found in these names. Thus, most may be considered to be divided
into three parts, as in the example of Cibacete/Bri11iant Blue/BG. The
first part usually gives the particular dyeing class,and, from this trade
name, we learn that the dye is a disperse dye intended for application
on acetate. Cibanone would be the same manufacture's designation for a
vat dye of high fastness. The second part of the name is obviously the
color, and the third part (BG in this case) refers to the shade. Often the
letters subdivide the numerous reds, oranges, blues, and greens into bluish
(B), greenish (G), yellowish (Y) (Y or G for the German "gelb"), and
reddish (R) shades. A number in front of these letters indicates their depth,
as in the name "Crystal Violet 6B". Other letters show other properties;
thus, K stands for cold dyeing (the German "kalt"), L for 1 ightfastness,
N for new, CF for copper free (as in the case of goods to be vulcanized),
A for acetate left unstained, W for washfastness, and S for sublimation fast-
ness. Strength and physical form are designated at times by such terms as
Cone, (concentrated), Dbl. Pst. (double-strength paste), Pdr. (powder), etc.
Dyes have been classified by a wide variety of schemes. Classes based on
usuage and application are useful to the dyer and also to the dye manufacturer
who must supply the demands of the dye user. However, this type of classi-
fication results in groups containing a great diversity of chemical structures.
Table 4-5 is arranged according to a usage classification and indicates
briefly the major substrates, method of application, and representative
chemical types. A second type of classification based on chemical structure
has also been used in the industry. Table 4-6 is arranged according to
chemical classification, giving typical examples of the characteristic
structural units and the application dying classes which fall in each chemical
group. Tables 4-7 and 4-8 show U.S. production and sales of dyes of usage
and chemical classifications.
The primary source of organic raw materials for the dye industry has traditionally
been products recovered from the fractional distillation of coal tar; hence
the name "coal-tar dyes" is frequently used in place of the more correct
"synthetic dyes" Coal tar is a by-product of the gasification (or car-
bonization) of coal, the primary purpose of which is the production of coke
for steel manufacture and coal gas for industrial and domestic heating. The
coal tar is refined by distillation, and from over 300 products which have been
isolated and characterized the most important for the dye industry are benzene.
toluene, xylene. naphthalene, anthracene, acenaphthene. pyrene, pyridine,
carbazole, phenol, and cresols.
The petroleum industry is now supplying an increasing proportion of the primary
raw materials, notably benzene, toluene, xylene, and more recently, naphthalene.
-183-
-------
DRAFT
In addition ot the organic materials above, a great variety of inorganic
chemicals are used in the dye industry. These include sulfuric acid
and oleum for sulfonation. nitric acid for nitration, chlorine and
bromine for halogenation, caustic soda potash for fusion and neutralization,
and sodium nitrate for diazotization, as wel1 as hydrochloric acid,
sodium carbonate, sodium sulfate, sodium sulfite, sodium sulfide, aluminum
chloride, sodium dichromate, manganese dioxide, iron powder, and many
others.
The great number of intermediates used to manufacture dyes and, where their
use is confined to dyes, the comparatively small tonnages involved make
manufacture by continuous process uneconomical. Anthraquinone, produced
from anthracene by catalytic air oxidation, is one of the very few
intermediates used solely by the dye industry that is made by a cont i nuous
process. Other large-volume intermediates (e.g., aniline, phthalic
anhydride, and phenol) are also manufactured by continuous process but,
as mentioned previously, the bulk of production is used in other industries.
The batch processes for the production of dyes and intermediates are
carried out in reaction kettles made from cast iron, stainless steel, or
steel, often lined with rubber, glass (enamel), brick, carbon blocks, or
stainless steel. These kettles have capacities ranging from 500 to 10,000
gallons and are equipped with mechanical agitators, thermometers, con-
densers, etc, depending upon the nature of the operation. Products are
transferred from one piece of equipment to another by gravity flow, pumping,
or by blowing with air or inert gas. Plate and frame filter-presses, filter
boxes, and centrifuges are used for the separation of solid products from
1iqu ids. Where possible, the intermediates are used for the subsequent
manufacture of other intermediates or dyes without drying. Where drying
is required, air or vacuum ovens (in which the product Is spread on trays)
and rotary dryers are used. Less frequently used are drum dryers (flakers)
and spray dryers, although the latter are becoming increasingly important.
Most of the material handling is manual, and labor costs represent a signi-
ficant part of the final dye price. Automatic process control, based on
feedback from temperature, redox potential, and pH measurement, is finding
increasing use in the industry.
The actual manufacture of intermediates and dyes proceeds by a series of
chemical reaction steps. The major steps and associated chemical reactions
typical of each are listed as follows:
Step 1 - Addition of Functional Groups to Raw Materials
The attachment of one or more chemical groups
onto the aromatic hydrocarbon raw material.
Typical reactions include sulfonation, nitration,
halogenation, and oxidations. Normally, the
starting raw material (such as benzene, toluene
anthracene, etc.) is reacted in the presence of
aqueous sulfuric or nitric acid.
-------
DRAFT
Step 2 - Replacement of Functional Groups on
Intermediates Produced in Step 1
The replacement of the functional groups intro-
duced in Step 1 by other groups of higher reactivity
which cannot be introduced directly. The starting
materials for this step may be the intermediates
produced in Step 1 or intermediates purchased from
another manufacture. Typical reactions include:
caustic fusion, to replace a sulfonic acid group
by a hydroxyl group; replacement of a sulfonic
acid group by an amino group by reaction with
ammonia and replacement of halogen atoms by
hydroxyl or amino groups.
Step 3 - Further Modification of Functional Groups
On Intermediates from Step 2
This step involves the further modification or
development of functional groups on the inter-
mediates produced in Step 2 ( or purchased inter-
mediates). There can be few generalizations about
these reactions, as each particular case depends
on the specified end product required. Examples
of these reactions include alkylation of functional
groups by reactions of the intermediate with an
alcohol, acylation with organic acid chlorides
or anhydrides, and other rearrangements.
Step 4 - Combination of Two or More Intermediates to
Form a Dye
In this step, two or more intermediates are com-
bined to form a product having a skeletal, if not
complete, dye structure. Typical reactions include
diazotization and coupling, condensations, and di-
mirization. It should be noted that many of these
products may themselves also be intermediates in the
synthesis of dyes of greater complexity.
When defining the production of dyes and intermediates at a specific plant,
it must be understood that some manufactures purchase the intermediates
associated with Steps 1, 2, and 3, so that Step k may be the only chemical
reaction processing done at the plant. Other manufacturers carry out Steps
1,2,3, and k in the synthesis of a dye product. Although the final quantities
of dye product sold at each plant might be the same, the actual production
activity (as measured by intermediates production) would be drastically
di fferent.
It should also be noted that production schedules vary drastically in dye plants
-185-
-------
DRAFT
Both the types and specific quantities of dyes and intermediates which are
manufactured change on a week-to-week basis.
Because of the nature of the batch operations, it was not possible to isolate
dye products according to chemical or usage classification for the develop-
ment of production based raw waste load data. Instead, the entire
plant was sampled for periods up to one month. Wastewater flows and
analyses were developed on a 24-hour composite basis. Production for the
corresponding period was defined to include all dyes and intermediates
which were separated within the establishment during the sampling period.
A commodity was considered separated when it was isolated from the reaction
mixture and/or when it was weighed, analyzed, or otherwise measured.
It should be noted that in some cases it was not possible to differentiate
between production as 100% active ingredients and as standardized material.
The difference between these two relates to the quantity of inert diluent which
is added to some dyes prior to shipment to the user. The raw waste load
data to be presented subsequently are based primarily on standardized
materialo The ratio of standardized material to 100% active ingredients
can be as high as 10 to 1. This means that if the raw waste loads were
based on 100% active ingredients, the data reported should be increased by
a factor of ten (depending on the specific plant.)
The following tabulation summarizes the raw waste load data obtained from six
plants sampled during both phases of this study.
Process RWL for Dyes and Interned Jare_
Plant
Flow
BO DC;
COD
TOC
1 . (10% occurrence)
(50% occurrence)
(90%* occurrence)
2. (10% occurrence)
(50% occurrence)
(90% occurrence)
3. (10% occurrence)
(50% occurrence)
(90% occurrence)
4. ( sampled 3 days)
(1 it./kkg)
795,000
205,000
185,000
(kg/kkg)
5
79
156
17
62
106
278
602
930
(kg/kkg) (kg/kkg)
32,800
32,800
32,800
1.18
50
1,850
3,700
104
212
318
1 .060
1 ,595
2,155
195
19.5
189
40
790
1,580
25.0
57.0
89.5
350
502
656
205
5.24
49.2
-186-
-------
DRAFT
Process RWL for Dyes and Intermediates (continued)
Plant
5. ( sampled 2 days)
6. ( sampled 1 day)
Mean (based on 90%
Plants 1 ,2,3)
days)
day)
90%
Flow
(lit./kkg)
114,000
114,000
175,000
395,000
BOD5
(kg/kkg)
220
126
397
COD
(kg/kkg)
1,075
652
175
2,060
TOC
(kg/kkg)
450
2.69
60
775
Only Plants 1, 2, and 3 were used to compute the mean RWL shown, as each
plant was sampled for 1 month (30 days.) In such case, the 90% occurrence
value was used. This is a value greater than 27 of the 30 RWL values com-
puted during the sampling period for each plant. The use of this value is
justified because of the extremely high variability shown by the dye plants,
-187-
-------
0. E
E O/
- E
- >
fO 0
- .
i- re u u
* x E w
§
>--C
fO
fD -0
1 D
13 -o
"O T3
~ fD
Q.
CL O
5
impregnatec
_o
c c
fD 0 E
3
1) 0
CON
CL
E -c -o
3 0)
en N
O L- -M
c
-a
i_
E
c
c
c
c
-a
v
x
E
0)
-C
O
c
O
OJ
1_
in
rrj -Q
3 C
H
fO
E -^
O
TJ
CL -
CL
fO
C
O
-a
-a
U
tn +j
c >-
'c 0
0
._ a)
in CL
i- E
a u
- -C
u
j .a
c a.
- Q_
-- 0
(U
O Cf'
L, 0)
O >
T3
L. T)
(1 O
m
J2
JD
>~
E ^
-O ill
T3 -C
.a ^
c
c
r o
O o
C CJ
O C
X -2
O 2- -0
- <£ 'TO
j-i E
3 j- O
0 - - C
c 5
-a "3
^^ ;
3
^
c
"D
>
- 3
O =
rr
.f _3
"O 0
--
£
D -
y 3
G_ ID
"j 0
0
.X 5
6
- u
> 3
> i_
(IL _TJ
3
3j
3
-c
fD
>;
(U
c
OJ
.,
tl C
O CL
- i_
*J O
C
O -
1 ° *
fO O ru
O s- a;
-------
DRAFT
Table 4-6
Chemical Classification of Dyes
Class
Nitro
monoazo
d i sazo
trisazo
polyzao
StiIbene
Dyei ng C 1 asses
Acid , d i sperse, mordant
Acid, disperse, mordant
Ac i d , d i rect, mordant,
disperse, basic, reactive
D i rect, react Ive
Remarks
Dyed as a metal che 1 ate
A large and varied class
produced almost without
exception by the coupling
of a d i azotized aromat i c
anine to a pheno1, ami ne,
pyrazolone, or other cou-
pling component
Insoluble dye formed di-
rectly on fiber from sol-
uble components by d i-
azotization and coup 1 ing
Class also includes mix-
tures of indeterminate
constitution made for
exa-np ! e by condensat i on
of n11 ro st i1bene compounds
and aromat i c ami nes
DiphenyImethane
(ketone tmine)
Tr i aryImethane
Xanthene
Acr i d i ne
Qu i noli ne
Methine and
Polymeth i ne
Th i azote
Indami ne and
indophenol
Basic, acid, mordant
Basic acid, mordant
Basic (sulfur)
Ac id, basic, direct,
di sperse
DIrect, bas *c, react ive
(sulfur)
Ac id , bas ic , ox idat ion
(sulfur)
Brilliantly colored dyes of
only moderate 1ightfastness
Pure , bright hues
Basic dyes used chiefly on
leather, also for anti-
sept ics
Used for cotton, paper, and
more recently in disperse
dyeing
Important in photography
Intermedlates for photo-
graphic and sulfur dyes
The ffrst commercially im-
portant synthetic dye,
Perkin's Mauve belongs
to thf s class
Aminoketone and
hydroxyketone
Anthraquinone
Indigo i d
Phthalocyanine
Oxidation bases
Basic, mordant di rect
(sulfur)
Basic, mordant vat
(sulfur)
Sulfur, vat
Ac id, mordant, vat,
d i spersed, bas i c,
di rect , reactive
Vat, acid
Acid, di rect, azoic, vat,
sulfur, basic react!ve
Incompletely characterized
oxidation products from
ami nes, diami nes, and
ami nophenols
Obtained by heating a vari-
ety of organic compounds
with sulfur or polysuifides
to give disulfide or sulf-
oxide bridges
The natural dye logwood is
included in this class
Condensed po iycyc f ic qu f n-
onoi d dyes of great im-
portance
Derivatives of indigo and
th ioi ndi go
Only b1ue or green dyes and
pigments (of high light-
fastness) are found in th i s
c 1 ass
Aniline Black is a member
of this class
-189-
-------
DRAFT
Table 4-7
U. S. Production of Dyes
by Classes of Application, 1965
Sales
Class of Application
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)
Bas ic
Oi rect
Di sperse
Fi ber-react i ve
Fluorescent brightening agents
Food, drug, and cosmetic colors
Mordant
Solvent
Sulfur
Vat
All Other
Product i on
in 1 ,000 Ibs.
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
Quant i ty
in 1 ,000 Ibs.
189,965
18,666
2,043
1 ,310
2,646
2,429
9,553
33,663
13,522
1,558
18,2.84
2,736
4,246
8,930
17,471
52,439
469
Value
in $! ,000's
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
Uni t
Val ue
S/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
-190-
-------
DRAFT
Table 4-8
U. S. Production and Sales of Dyes
by Chemical Classification 1964
Sales
Production Quantity Value
Chemical Class _i n 1 ,000 Ibs. in 1 ,000 Ibs. in $1 ,000's
Total 184,387 178,273 264,023
Anthraquinone 41,661 40,675 66,889
Azo, total 57,897 57,367 96,579
Azojc 8,787 7,399 12,149
Cyanine 373 362 1,113
Indigoid 5,729 6,144 3,302
Ketone I mine 731 782 1,614
Methine 1,074 974 3,367
Nitro 720 679 1,258
Oxazine 172 144 601
Phthalocyanine 1,987 1,868 4,800
Quinoline 637 519 1,658
Stilbene 18,488 17,640 29,166
Sulfur 17,776 17,268 9,798
Thiazole 462 480 1,043
Triarylmethane 5,607 5,312 12,682
Xanthene 1,312 737 3,473
All Other 20,974 19,923 14,531
Source: Synthetic Organic Chemicals, U. S. Tariff Commission
in 1965 total dye production increased 12.5% to
207 mi 11 ion Ib.
-191-
-------
DRAFT
Products : Fatty Acids and Primary Derivatives
Process RWL Category: D
Fatty acids are organic acids characterized by a straight chain carbon
structure terminating in a single carboxyl group. The carbon structure
may contain only saturated carbon atoms or may contain one or more
double bonds.
In general, the saturated acids having 12 to 18 carbon atoms are of
major commercial concern. Their uses tend to reflect their stability
or lack of easy reactivity at any point along the chain except at the
carboxyl position. Saturated acids, including those saturated by
hydrogenation, comprise about one-third of all fatty acids commercially
produced.,
Acids of reduced hydrogen content, having one or more double bonds
between carbon atoms along the chain, constitute the other main fatty
acid family of high commercial importance. The unsaturated acids
of the ethylenic family are made up of a number of separate; series of
the following compositions:
C H0 000 (monoethenoic)
n Zn-2 L
C H0 ,00 (diethenoic)
n Zn-4 L
C H_ ,-00 (t r iethenoi c)
n zn-o z
The chemical reactivity and utility of unsaturated fatty acids in various
polymeric or "drying" reactions is determined in part by the number of
double bonds along the carbon chain. These double bonds also introduce
different properties arising from the location along the chain, i.e.,
conjugated and nonconjugated fatty acids. A different structure also
results from a reversed positioning of those parts of the molecule on
either side of the double bond, known as cis and trans isomerism.
To avoid confusion of terms, it is necessary to distinguish between
fatty acids as defined by the organic chemist and fatty acids as com-
mercially produced and sold. The term "stearic acid" is an example of
this confusion.
Here the commercial product name has priority, going back to the split-
ting of tallow for the manufacture of hard, high-melting candles, before
individual acids of definite composition have been isolated or defined.
-192-
-------
DRAFT
Historically, commercial stearic acid has been a crystalline combin-
ation of the chemist's palmitic and stearic acids in a 55 to 45% ratio,
respectively, with some small percentage of unsaturated acids. This
ratio is that which naturally occurred in the mixed acid derived from
splitting tallow, after most of the liquid acids were removed by the
original pressing method. The liquid acids so removed are known as
oleic acid, or red oil.
Other commercial acids have been identified by origin - for instance,
coconut fatty acid or cottonseed fatty acid - in terms reflecting the
characteristic ratio in which the component acids exist when released
from the glycerides of these oils. Fractionation to separate or to
enrich the wanted acids, however, has made identification by origin
obsolete as a means of characterizing many of today's acids. Blends
and "cuts" are offered to meet particular market demands, such as
drying characteristics in paint firms. Composition expressed in
terms of component acids as the chemist defines them is now finding
its way more and more into product descriptions and specifications.
Table k-9 lists 15 of the most important commercial fatty acids, along
with their composition in terms of specific constituent acids.
In addition to the saturated and unsaturated acids, primary derivitives
based upon acids modified by amination, esterification, ozonation, and
polymirization are considered within the context of this study. Also,
the processing assiciated with the recovery and purification of by-
product glycerine is considered. The relation between glycerine and
fatty acid production will become apparent during subsequent discussions.
However, at this point it should be clearly noted and understood that
the production of other products (such as soaps) from fatty acids is
not considered within the context of this study,,
Raw materials for the production of fatty acids fall into three groups:
1, Tall oil derived from Kraft papermaking.
2. Animal tallow and grease.
3. Vegetable oils and soap stocks.
As a raw material for fatty acid production, crude tall oil ranks first
in volume. Crude tall oil is a mixture of rosin acids and free fatty
acids (in addition to a large percentage of water and impurities) and
is a by-product of kraft papermaking. Most of the crude tall oil
is fractionated, and the separated rosin acids and tall oil fatty acids
are sold directly to consumers after purification. The fatty acids
derived directly from tall oil can be used in the manufacture of primary
derivitives such as dimers, polyamides, nitriles, primary amines, di-
fatty tertiary amines, and fatty quatenary ammonium chloride. The
major constituents of tall oil fatty acid are oleic acid and linolenic
-193-
-------
DRAFT
_
L_
1 1
o
<
c
o
D
4-»
4->
1/1
C
O
o
4-
o
0)
en
ra
4J
c
0)
o
0)
Q-
ra
o
CL
^
1
cp en
-* 3
0
_c
CO
LU
_l in
co xi
<.
h- 0
<
>-
4-1
4-1
ra
l_l_
i
ra
u
s_
0)
o
o
14-
0
t/>
(U
CL
E
ro
X
LU
0)
Q
CO
D|U8[OU I")
3 1 8 [ OU l~|
^1310
3 1 U3U8g
3 1 D 1 U~) P J W
!r'H\7
o | jeaag
o i i i u [ ed
01 }s i jAw
*
3 ijne"i
oudeQ
3 1 [ AdBQ
D louden
'
OO
CO
~
CO
CM
CM
0
CM
CO
v£>
j-
CM
o
CO
v£>
I/I
0
ro
CJ
r^\
CM
o
o
o
o
o
o
o
o
o
t/1
XI
c
o
_o
0)
_Q
D
o
o
1/1
XI
ra
._ o
ra o ra
o u
~
CL £ 4-1
^ E 4J
h- O ra
0 M-
CM T LA , 1
-J-
CM o^oj- crii o r-^j-
* rv» LP\ ^D -^- CM
vD LTv-j- CM CM^OCTiLrvOroCflO^' CT>
r^co LA m CM on LP\,
o
LA
COr«~>OCr\COCOCM CMi LAfM-d- c^-\
J- J" CM LA MD CM T
Of LA LA O^ LA C7> r^- LA LA-J- T CM^D
LALAP^rOCNro CMCMi
t CM i o r^. CNI m -J" i
(T\
J-
i
-
1)
Q-
^-x >. XJ
OJ 4-1 -
D- --- 0
>- 0) 4> ra
^ v - * 4-1 i CL - v .- v
4) (U _Q >- "U C >-
D- CL 2 ra +-> ' ^ (U 034-"
>~>-O4-j - - x> a) 0)4-1
4J4J ^- (U -C 4-1. . 0) "D tn T3 C "D J3 TO
m 1/1 D ._ c ro >* xi M-
ra i) c ' OOUOJOO-
Q-Q_ 4-1 > u- OO -XI ra*-|ro.arac/iuxi
XJX»x)~uo 4-j 4-1 >- rocu
H- oo ._ xi. -xi xi ox) mo xo5^a>-x> D
" ' uua>oa)uoa> ra 4-JO4-ii/i4-ia)Xui
ura4~ira4-ira4-iro u - . xs - ^ 4-j 4-1 4-j 4-* 4-1 c
XJXI- ra ra ro x>- xi >-xiro"oroxiroro4-i
4-i>-c>.c>-C(ja) M-ra4-i 4-1 _a 4-1 _Q 4-j _Q u
CL xi xi ~O' t/i 1/1 ' t/iroi/irai/iraroO1
ui_ xi>^x)>^xi>- o * o ^ ra -u 4-1 4-1 u .
raro-n 3: n: :r _c o Q 2; EQ a) a a) a
m D en - - en ' a) i> ' - en ' en en m
4J4J.-O o o c: a) a> a; ro
OOOOOICO CO 00 I O O < > > > 1 Q
-19/,-
-------
DRAFT
acid which contain two and three double bonds respectively (see Table ^-9),
It should be noted that the production of tall oil fatty acid involves
only distillation of free acids and is simpler than the processes used to
manufacture fatty acids from animal and vegetable sources.
There are four main types of animal fats: edible tallow, inedible tallow,
lard and inedible grease. Inedible tallow and grease are important raw
materials as a primary source of stearic and oleic acids. These materials
are supplied to fatty acid manufacturers from meat packing plants and
rendering plants. It should be noted that fatty acid production is a
relatively small part of the tallow market, with the major end use of
tallow being in the production of soap.
Coconut oil (obtained by crushing coconuts) is an important vegetable
raw material for fatty acid and fatty ester production, because of its
high Cg, C,Q, and C-2 content. Although soybean oil and cottonseed
oil are produced in nuge quantities, they are not often used for fatty
acid manufacture because of their high costs. However, by-products
from the refining of these oils for other end uses are used in the manu-
facture of fatty acids. These by-products are called "soap stocks" or
"foots" and vary in composition, depending on the refining process by
which they are produced.
It should be noted that fatty acids are present in animal and vegetable
sources as glycerides. Tall oil, which is not a glyceride, is the only
raw material which contains high concentrations of free fatty acids.
The three-pronged glycerol linkage, typical of all fats, must be severed
in the production of all fatty acids derived from animal and vegetable
sources. The chemical reaction for this cleavage involves the following
hydrolys i s:
0
11
R-C-0-ChL
0
11
R-C-O-CH
0
11
R2-C-0-CH2
Fat (triglyceride)
CH-OH
CH2--OH
Glyceri ne
R--COOH
4- F^-COOH
RZ-COOH
Fatty acids
The products of the reaction are crude glycerine and fatty acids,
-195-
-------
DRAFT
Depending upon the nature of the feedstock and the desired end-product
fatty acids, different combinations of the following processing steps may
be found in a fatty acid plant:
1. Pretreatment For Purification Of Feedstock
2. Saponification/Acidulation Of Feed Or Products
3. Hydrogenation Of Feed Or Product Acids
4. Hydrolysis Of Feed (Fat Splitting)
5. Distillation Of Product Acids
6. Distillation For Glycerine Recovery
7. Separation Of Saturated And Unsaturated Acids
The following paragraphs briefly describe each of these processing steps.
A generalized process flow diagram showing their interrelation is pre-
sented in Figure 4-41.
Pretreatment
Many of the impurities present in the feedstock will decompose, volatilize,
or contaminate the product acids and must be removed. Water-soluble salts
will deposit on heating surfaces. Mineral acid will attack the fatty
acid upon heating and attack the equipment as well. Water-washing and
drying of feedstock containing these impurities are essential.
Feed stock containing gums, proteinaceous material, calcium, and iron
soaps should also be acid-washed. Acid-washing is generally performed
in lead-or monel-lined tanks. A common method is to treat a charge of
melted fat at about 140 F with 2 to 4% sulfuric acid in a fairly con-
centrated (30 to 50°/0 solution, with good agitation for about 1 hr., after
which the charge is heated to 200 F with open steam. After settling,
the acid water is drawn off, and the charge is washed with water to remove
the mineral acid. Acid washing breaks up calcium and iron soaps that act
as catalysts, forming ketones, or that remain as soap in still bottoms,
Sulfuric acid removes proteins and other organic inpurities.
Acid-washing also removes impurities that hinder hydrolysis. Bleaching
with acid clays is effective in removing oxidized acids and impurities
that hinder splitting and/or hydrogenation.
Saponi fi cat ion/Aci dulat ion
In special cases where the feedstock is by-product "soap stock" or "foots"
produced in the caustic refining of vegetable oils, the saponification of
the feed followed by acidulation is a convenient means of manufacturing
fatty acids. It has an advantage in the handling of raw refinery foots
because very little additional caustic soda is required to complete the
saponification before the acid treatment. Completely saponified foots
-196-
-------
DRAFT
upon acidulation yield raw material satisfactory for distillation. Although
more expensive than the other methods using water, it gives practically
100% conversion and avoids the use of high temperatures. The splitting of
fatty acids from waxes and sperm oil is done by complete saponificat ion,
using anhydrous alkali, followed by steam distillation to remove the
high molecular weight alcohols and by subsequent splitting of the soap
with sul furi c acid.
Hydrogenation
Hydrogenation is the process of adding hydrogen to materials that are
deficient in hydrogen, e.g., oleic acid requiring one molecule of hydrogen
to form the C^ saturated acid. Either the glyceride or the fatty acid
can be hydrogenated. The process is essentially the same, with the ex-
ception that the acid must be processed in a corrosion-resistant apparatus,
usually stainless steel.
Refined oils and fats and fatty acids that have been properly distilled
are hydrogenated with little difficulty; but impure materials or mixtures
of fatty acids and neutral oil, such as acidulated foots, are hydrogenated
only after they have been specially treated, and then usually with dif-
ficulty.
Hydrogen addition to the unsaturated groups occurs only when it is acti-
vated by means of a catalyst. Reduced nickel, the usual hydrogenation
catalyst, is made by reducing nickel salts, i.e., formate, carbonates,
or the aluminum nickel alloy.
Hydrogenation is done in both batch and continuous processes. The batch
process, which is older and not as generally used today, uses a vessel
that is equipped with a top-entering agitator and a means for heating and
cooling and that is built to operate at pressures of from 25 to 250 psi
(in some instances up to 500 psi and higher). The nickel catalyst is
added to the pre-dried charge in amounts varying from about 0.05 to 0.5%.
The vessel is purged several times with hydrogen and then brought up to
reaction temperature with steam. Once reaction starts, however, it is
exothermic, and cooling is required to prevent overheating of the catalyst
and the stock. The exothermic heat is significant, and the absorption
of 1 Ib-mole of hydrogen releases about 40,000 Btu, requiring generous
amounts of cooling surface as well as good agitation. Good agitation
is also required to keep replenishing the reaction with hydrogen gas.
Reactors in use today vary in size from those holding a few pounds to
large ones holding 50,000 Ibs.
Continuous methods for the hydrogenation of all classes of chemicals have
been developed in recent years, especially in connection with petro-
chemicals and for petroleum processing in general. The fat and fatty
acid industries have been much slower in developing and using continuous
-197-
-------
DRAFT
methods, owing perhaps to the wide variety of products produced, the
relatively smaller volumes, and the need for modifying the product by
partial and selective hydrogen addition.
Hydrolys i s
Hydrolysis or splitting is the process of reacting the fat with water to
form glycerine and fatty acid. A series of three hydrolysis steps is
required to obtain free acid and free glycerine, and it is believed that
they are homogeneous reactions taking place mainly in the fat phase, as
the solubility of water in a fat is greater than the solubility of fat
i n water.
It is necessary only to mix fat with water to cause some hydrolysis, but
the reaction is very slow. When sulfuric acid is added to a fat, the sul-
fonated products formed increase the solubility of water in fat, and
hydrolysis takes place at a faster rate. Solubility is further increased
by the use of an alkaline catalyst, i.e., zinc oxide, magnesium hydroxide,
or calcium oxide.
A major development in hydrolysis techniques was made by Twitchell, who
developed the sulfonic acid catalyst bearing his name. As the solubility
of water in fat increases very rapidly at high temperatures, high-pressure
autoclaves, both of the batch and continuous type have been developed.
Commercial fat splitting is carried out by several processes, each of
which has advantages and disadvantages; the selection depends upon the
type of raw material, the size of the operation, and the number of kinds
of raw materials handled.
A. Twi tchel1 Process
The Twitchell process, developed in 1890, is still used to
some extent in the United States. It is carried out in a
lead- or Monel-lined tank at atmospheric pressure; the
charge consists of the fat mixed with about 50 wt.% of
water, 1% of the sulfonic acid catalyst, and 0.5% sulfuric
acid. The misture is boiled with open steam for 16 to 2k
hr, allowed to settle, and the sweet water is drawn off
and replaced with fresh water. The boiling is continued
until a split of about 95% is reached. The overall split
depends upon the glycerol concentration of the sweet water.
By countercurrent use of water, it is possible to obtain
high splits and high concentrations of sweet water, usually
10 to 15%. The advantages of the Twitchell operation are
in the relatively simple equipment and the use of low
-198-
-------
DRAFT
temperatures. Low-temperature operation is highly desirable
for splitting stocks containing multiple unsaturation. Dis-
advantages are the length of time required, high steam con-
sumption, and poisoning of the catalyst, which necessitate
the pretreatment of most stocks, especially the lower-
grade variety.
B. Batch Autoclave
Batch autoclaving is a very old method, which, as originally
developed, used a closed cylindrical pressure vessel with
agitation provided by open-steam injection with a continuous
vent. The usual temperatures are about 365 F, equal to
about 150 psi. About 2% lime, zinc, oxide, and so forth,
are used; the amount of water is about 50 wt.% of fat. The
degree of split reaches an equilibrium at about 90%; to
obtain a higher split, the glycerine water must be replaced
with fresh water.
This catalytic autoclave process is suitable for all types
of fats and has the advantage over the Twitchell process
of a shorter reaction period and the production of lighter
colored fatty acid because of the absence of air. A more
recent development in autoclave splitting is the use of
higher temperatures and higher pressures requiring no
catalyst. This process is usually carried out using
mechanical agitation, with the temperature held at about
450 F under a pressure of 450 psi. Splitting is rapid,
and there is no catalyst to remove. Its disadvantage
is the effect of high temperature on stocks having a
high degree of unsaturation.
C. Continuous Countercurrent Splitting
Over the past 30 years, several companies have designed
and built successful continuous hydrolyzers, which have
replaced the older methods. Fat enters near the base of
the hydrolyzer and passes upward through a sparger pipe
near the top. Steam from the high-pressure steam boiler,
at 750 psi, is injected at two locations, the top ad-
dition into a water-distributing tray and the lower
addition through a sparger pipe near the fat-water inter-
face.
The fat rises slowly through the column in a continuous
phase while the water drips through it, constantly re-
placing the glycerine-laden water that is in the solution
in the fat. At the bottom of the tower, the glycerine
water collects and is removed continuously to maintain
a constant interface level near the base of the tower.
-199-
-------
DRAFT
The fatty acid leaves the top of the tower through a
pressure control valve set to maintain about 720 psi ,
sufficiently high to prevent vaporization of water at
the required operating temperature, usually about 250
to 260 C.
The fat must be under pressure for about 2 hr to reach
a split of 96 to 99%, although the time required may
vary, depending upon the type of fat and quantity of
water used. It is usual practice to remove sweet
water with a concentration between 12 to 20% glycerine,,
Continuous splitting has the advantage of yielding
high splits and more concentrated glycerine solutions
than any other process. It has a high capacity and
short reaction time and requires less room, less
process inventory, and less labor. It does, however,
restrict flexibility in changing stocks. The high
temperatures, as mentioned above, have the advantage
of speeding up the hydrolysis reaction; however, it
has a real disadvantage when handling highly un-
saturated fat, such as fish oil, because these oils
tend to polymerize.
Recovery and Purification of Glycerine
The dilute (10-20 wt.%) glycerine solution (" sweet water) may be concen-
trated by stripping water overhead in an evaporator or distillation column.
In most operations, non-contact steam is used to drive off the water to an
approximate 80% glycerine concentration. The non-contact steam may be
condensed and used in the hydrolysis reactor. The separated water vapor
and some glycerine are normally condensed in a barometric condenser and
discharged. Additional treatment steps for purification of glycerine
water may include filtration, ion exchange, and activated carbon adsorption
Disti1lation
Distillation as a means of purifying fatty acid has been in use for the
better part of a century. It is an economical and successful method of
producting high-purity fatty acids, but it is beset with many problems
because fatty acids have high boiling points and decompose when held at
elevated temperatures. The early distillers soon discovered that pro-
cessing temperatures must be held at about 250 C maximum to limit decom-
position. At higher temperatures, fatty acids first lose water, forming
anhydrides, and later, under continued heating, break down into ketones
and hydrocarbons. Unsaturated fatty acids polymerize, forming dimers,
trimers, and so forth.
-200-
-------
DRAFT
The early stills were operated at atmospheric pressure and large quan-
tities of injected steam were used to maintain the temperature at 250 C.
As better equipment became available, the stills were operated at lower
and lower pressures until today most fatty acid stills operate at pres-
sures of between 5 and 50 mm Hg abs. At low pressures, fatty acids
will distill without injected steam, but it is usually desirable to
use some injected steam because even a small amount will aid in pre-
venting the formation of anhydrides. When stills are operated at
pressures lower than the vapor pressure of the available cooling water,
it is necessary to provide a steam compressor in addition to the ordinary
air pumps to maintain desired operating pressures.
Fatty acid stills now in general use may be classified into three general
types: 1) those for semicontinuous batch operation; 2) those for con-
tinuous simple distillation; and 3) those for continuous fractional
distillation. If only a decolorizing step is necessary (one that removes
the unhydrolyzed oil, polymers, and high-boiling color bodies), a simple
distillation will produce satisfactory distillate. If considerable amounts
of odor bodies (low-boiling unsaponifiable matter and compounds that cause
color reversion) are present, some means of fractionally concentrating
these "low boilers", either by fractional distillation or fractional con-
densation, is required. If the component acids must be separated, an ef-
ficient fractionating still is necessary.
Separation Processes
Separation of tallow fatty acids into solid (saturated) and liquid (un-
saturated) components cannot be accomplished by the fractionation methods
previously described because major components have the same chain length
and an insignificant difference in molecular weight. Two major separation
processes have achieved commercial significance in the separation of solid
saturated acids from liquid unsaturated acid, namely, the panning and press-
ing method and solvent crystallization.
The panning and pressing process at one time accounted for a major portion
of the total production of stearic and oleic acids. The idea of pressing
liquid fatty acids from solid fatty acids probably occurred from observing
the production of lard oil. In fact, equipment used for the pressing of
grease to yield lard oil was initially used for the separation of fatty
acids. This method is used for the separation of animal fatty acids to
produce commercial stearic and oleic acids. Simply, this method involves
the crystallization of solid or saturated fatty acids in a solvent of
liquid or unsaturated fatty acids. In consequence, the solid (saturated)
fatty acids must exhibit a reasonably good crystal formation so that the
liquid acids may be easily and efficiently expressed.
-201-
-------
DRAFT
In the production of commercial stearic acid, the optimum crystal
structure is attained when the saturated acids of the fatty acid
mixture have a composition of 55% palmitic acid and k$% stearic acid.
Limited variation in this ratio may be tolerated in actual practice,
but the variation in normal animal fats is suffient so that blending
is necessary to secure a proper crystalline structure. Fatty acid
mixtures of a non-crystalline structure may be partially separated
by the pressing method.
In the panning and pressing operation properly prepared melted fatty
acids are cascaded into rectangular aluminum trays, which are stacked
in racks in cold-storage rooms. Cooling takes place very slowly to
assure the formation of large and well-defined crystals. Cooling by
mechanical refrigeration to a final temperature of 35°F is attained
in a cold-storage room in about 6 to 8 hr. The solidified cakes of
fatty acids are removed from the trays, wrapped in burlap or cotton
cloths, and stacked in vertical hydraulic presses. Metal sheet
separators are placed between every two layers of wrapped cakes to
serve as stabilizers to conduct the expressed oleic acid to a trough
at the edge of the press. Pressure is very slowly applied to the
stack of cakes until a maximum of approximately 3000 psi is attained.
The expressed oleic acid amounts to about 50 to 60% of the original
fatty acids. Depending upon the temperature of the cakes of fatty
acid during pressing, the titer of the oleic acid will range from
6 to 10°C.
The pressed cakes, now referred to as "cold pressed cake", are melted,
again cascaded into racks of aluminum trays in an open room, and al-
lowed to solidify at room temperature. The solidified cakes are placed
in hair mat slings suspended between steam-heated hollow metal plates
in a horizontal ram press. Application of pressure and heat removes
most of the oleic acid along with a small portion of the solid acids,
resulting in a mixture of fatty acids known as "hot press oil". The
pressed cake from this "hot pressing operation" is called "double-
pressed" stearic acid. It contains about 5 to 8% oleic acid and has
a titer of approximately 5^ to 55 C. In much the same manner "triple-
pressed" acid is produced with the additional production of hot press
oil. Triple-pressed stearic acid contains from 1 to 3% oleic acid
and has a titer of approximately 55 to 56 C. The hot press oil has
a composition of saturated and unsaturated acids similar to the feed-
stock and is therefore mixed with the incoming fatty acid feed. As
a result, about 40% of the fatty acids in this process is being recycled,
The panning and pressing method is gradually being replaced by the
newer solvent methods, as has countercurrent extraction by immiscible
solvents.
-202-
-------
DRAFT
Solvent processes utilizing a 1iquid-to-1iquid extraction method, i.e.,
depending on the selective action of a solvent on the liquid fatty acid,
have been proposed, but the mutual solubility of mixed fatty acids in
solvents results in an inefficient separation. This has limited the
commercial use of these methods.
Separation methods involving the crystallization of one-component fatty
acids from a solvent solution of a mixture of fatty acids eliminate, to'
a large extent, the effect of mutual solubility. Solvent crystallization
methods may be applied to the separation of most fatty acid mixtures,
provided the component acids can be selectively removed in a solid state,
from a solvent solution of the mixture.
Of the many solvent crystallization separation methods proposed, the
Emersol Process has the most commercial installations. This process
involves the controlled crystallization of fatty acids from a polar
solvent to achieve a separation of solid fatty acids from liquid fatty
acids. Separation of saturated acids or of triglycerides may also be
accomplished. One of the best illustrations of the application of this
process is in the separation of animal fatty acids to yield commercial
stearic acid and oleic acid.
Animal fatty acids are dissolved in 90% methanol to yield a 25 to 30%
concentration. The methanol solution of the fatty acids is pumped
continuously to a multitubular crystallizer fitted with agitator scraper
blades and cooled to about -15 C. Cooling is accomplished by circulating
refrigerated methanol through the jackets of the crystallizer tubes.
During chilling the solid fatty acids crystallize from the solvent solution
to form a slurry that is fed to a rotary vacuum filter. The solid acids
filter to form a cake that is continuously washed with fresh 90% methanol
and then discharged from the filter. This filter cake containing ap-
proximately 40 to 60% methanol is melted and pumped to a solvent recovery
still in which the solvent is removed from the fatty acids and returned
to the system. The solid acids are discharged from the still ready for
finishing and packaging operations into commercial stearic acid.
The filtration containing the liquid acids is passed through a heat
exhanger to a solvent recovery still. The discharged liquid acids
from the still are ready for finishing and packaging into commercial
oleic acid. Reported capacities of the Emersol units range from 2000
to 5000 Ib of fatty acids per hour. At present, there are a total of
seven units in operation in the United States, Great Britain, Holland,
and Austra 1 ia.
Certain other separation methods, although not yet commercial, have in-
teresting possibilities. A solvent crystallization process for separating
tallow fatty acids using hexane as a solvent has been employed. Much has
-203-
-------
DRAFT
been published on the separation of fatty acid mixtures by use of urea
or thiourea complexes. Urea or thiourea adducts or complexes of fatty
acids have varying degrees of stability, depending on the carbon chain
length and configuration of the molecule. This difference in adduct
stability is the basis for obtaining a separation of the fatty acids.
Recovery of Fatty Acids From Tall Oil
When non-glyceride feedstocks such as tall oil are used, a different
process from that shown in Figure 4-41 must be used.
In the production of tall oil fatty acids and rosin by fractional dis-
tillation, crude tall oil is passed through a vaporizer into a fraction-
ating column where the volatile rosin and fatty acids are separated
from higher-boiling and nonvolatile impurities, which are drawn off as
pitch. The vapor enters the fractionation column and is separated into
three fractions: (1) rosin that is taken off at the bottom of the
column, (2) a fatty acid containing from 1 to 5% rosin near the top of
the column, and (3) a heads fraction overhead, containing a high per-
centage of low-boiling fatty acids (mainly palmitic), unsaponifiables,
and color bodies. The fatty acid fraction is then passed into a third
column where it is further stripped of rosin, unsaponifiables, and
color and odor bodies to yield a product with a rosin content of 0.3
to 2.0% and unsaponifiables content of 0.3 to 1.5%. The fraction con-
taining 25 to 35% rosin, taken off at the bottom of this column, is
either sold as such or refractionated. By taking fatty acids off at
the second column or at various points on the third column, the rosin
content of the fatty acids can be varied.
The fractionation of tall oil presents a number of problems. Both the
fatty acids and the rosin have low vapor pressures and are subject to
decomposition on heating. Either very high vacuum or vacuum with the
addition of superheated steam is needed to keep the temperature low
enough to prevent damage to the products. Overheating must be pre-
vented during vaporization by creating high flow velocities in heaters
and vaporizers, and entrainment must also be prevented. The equipment
is, therefore, spec! fical ly designed for the purpose. Various types of
fractionating trays, grid trays, sieve trays, and bubble-cap trays
are used; all of them have certain advantages and disadvantages and
unless they are carefully chosen and designed for the purpose, will
fail to produce the desired results.
Different operating schemes are used by the various producers, employing
single and multiple tower system. A two-tower system can produce excel-
lent results by refractionating the fatty acids. This raises the operat-
ing cost but the investment cost for the plant is lower.
-204-
-------
DRAFT
For the purposes of this study, the previously described unit operations
and chemical conversions are defined to be the basic steps necessary
to produce fatty acids. Field sampling was used to establish raw waste
loads for plants encompassing these steps. Because of the batch or
semi-continuous nature of the operations, it was not possible to break
out the total RWL according to individual step sources. Rather, com-
bined waste streams from the acid production areas were sampled in
most cases. A total of 6 plants were sampled. Table 4-10 indicates
the specific operations relating to acid and derivatives production.
Table 4-11 summarizes the RWL data related to acid production for 5
plants, and presents an arithmetic average for these operations. It
should be noted that the RWL data presented in Table 4-11 are based
on samples taken after gravity separation of fatty acids from the
wastewater. It is common practice to recycle the skimmed material
back to the acid pretreatment section of the plant for recovery.
Table 4-12 summarizes the raw waste loads calculated from production
of primary derivatives. As with acid production, it was not possible
to develop separate RWL's for specific derivatives because of the
batch nature of the operations. The data shown relate to groups of
processes in operation during the sampling program. Associated pri-
mary derivative products include esters, amines, nitriles, dimers
and trimers, polyamides, and fatty Quaternary ammonium chloride. An
average RWL value for these processing operations is provided in Table
4-120 Again, it should be noted that the values are based on samples
taken after gravity separation of free oil and in-process survey steps,
except as noted,,
The average values shown in Tables 4-11 and 4-12 were considered as
BPCTCA. The plants surveyed discharge both to surface waters and to
municipal treatment plants.
-205-
-------
DRAFT
e
o
L.
U-
C
C
o
4-J
o
"O
o
Q_
"D
0
-M
03
0
.
4-J '
l/l
03
Q j-
C
in
QJ
>
.
03
>
1_
QJ
Q
O
4-J
03
._
U
QJ
E
O
C
O
4_l
03
N
.
1_
QJ
E
i_
c
0
J~
03
C
^
<
C
o
U
~o
o
!_
CL
>.
c
O
in
QJ
>
4_i
03
>
1_
Q)
U
O
~o
0-
E
03
QJ
!_
QJ
' CXI
c
o
4-J
(J
"D
O
Q_
-0
"_!_,
03
C
en
0
!_
>^
X
"^
Q-
co
03
U.
QJ
"D
S_
QJ
U
_
0
!_
0
U
or
o
ro
,~_
-w
._!
Q
QJ
.
03
>
QJ
Q
03
U
4-
^_
QJ
tn
LU
c:
C)
c:
EE
<
o -o
O 0>
-O S- L- 4J CL
o
o
01 O
U U
O
o
a;
QJ
C
L-
QJ
O
>.
CD
_Q
>~
S_
QJ
>
O
L)
QJ
C
O
4-J
03
.
4-J
in
O
0
QJ
rc
3
QJ
4-J
(/)
03
1
0)
QJ
C
!_
OJ
U
,
LD
0)
O_
E
03
C/l
>-
!_
0)
0
o
>-
4-J
03
>
i_
OJ
O
1
to
U
. _
'_
4-1
m
LU
~
c
0
4-J
03
C
E
U,
QJ
LO
UJ
CL
OJ
CO
~o
QJ
CL
E
ro
CO
o
CO
a:
U.
>.
c
o
o
0
^
_Q
03
0)
L.
4-J
QJ
1_
"5
03
]5
O
<
TD
to
C
0
03 O
C
OJ
CT) 03
O h-
TD U-
>- O
1C
>-
LO
in
>,
O
>-
O
Q_
OJ
O
4-J
,_
3
1
O
LJ
OJ
c£
c
._
0)
o
s
o
c
o
4-J
03
4-J
in
s
^
~o
o
<
4-J
03
'4-
O
C
o
4-J
03
03
O-
GJ
in
u
Q_
_Q
in
~D
U
"*
^- CM
CT
C
O
4-J
O
-Q
O
U
Q_
-o
0
C
Jd
U1
03
"TD
U
^
QJ
3
Ol
in
QJ
i_
CL
_C
en
T
CN!
01
C
.
j-J
4->
.
Q.
LO
4-J
LJ_
GJ
>
O
o
0)
rV
OJ
C
1_
QJ
U
>-
C3
p^
C
fC
._
"4-
1_
=i
a
"D
C
03
4-J
03
U
03
QJ
CO
4-1
C
QJ
O
J-'
in
-D
o
e£
>-
4-J
03
LJ_
O
0
c
o
.
c
OJ
en
o
i_
X
LA
_
^ C
O
01 C -
-a i/i
<
>- 03
4-J >
4-J .
03 1-
LL (U
Q
O 4-J
.- 03
4-1 N
03 -
N S-
.- QJ
i- E
QJ .-
E i-
O
r-
O -
03 CL
3
-206-
-------
DRAFT
TABLE 4-11
PROCESS RWL ASSOCIATED WITH MANUFACTURE OF FATTY ACIDS (ALL
RWL BASED ON SAMPLES TAKEN AFTER GRAVITY SKIMMING FREE FATTY
ACIDS FROM WASTEWATER)
Flow BOD COD TOC
(lit./kkg) (kg/kkg) (kg/kkg) (kg/kkg)
Plant
Plant
Plant
Plant
Plant
1
2
3
4
5
(1
(1
(1
(1
(1
(1
(1
(1
day)
day)
day)
day)
day)
day)
day)
day)
10
10
63
3
45
65
12
12
,300
,300
,900
,700
,500
,900
,100
,100
(a)
(a)
(b)
(c)
(d)
(d)
12
18
12
11
14
21
15
37
.7
.5
.8
.7
.6
.5
.6
.5
44
52
23
37
41
5
42
53
.3
.5
.6
.3
.6
.8
.4
.2
4.89
10.3
4.41
10.6
14.3
1.2
13.6
18.9
Average 28,000 18.1 37.6 9.8
(a) RWL includes wastewater from dimerization
(b) RWL does not include wastewater from glycerine recovery
(c) RWL based only on glycerine recovery (i.e., gased on glycerine)
(d) RWL includes small contribution from derivatives
-207-
-------
DRAFT
TABLE VI 2
PROCESS RWL ASSOCIATED WITH MANUFACTURE OF PRIMARY DERIVATIVES
FROM FATTY ACIDS (NOTE RWL BASED ON SAMPLES TAKEN AFTER GRAVITY
SKIMMING FREE OIL INDICATED BY ASTERISK)
Flow BOD COD TOC
(lit./kkg) (kg/kkg) (kg/kkg) kg/kkg)
Plant 1* (1 day)
(1 day)
Plant 3
Esters (1 day)
Amines (1 day)
Plant 4* (1 day)
Plant 6* (1 day)
Average 6,400 18.0 27.9 8.47
(Exclude Plant 3)
4,590
4,590
127.0001
18,8002
8,920
5,700
16.5
17.0
26,500
495
28.6
9.82
32.6
41.3
54,700
1,070
14.7
23.0
9.75
14.7
13,600
245
5.7
3.73
Sample taken prior to methanol recovery
2
Sample from blowdown on recirculating system
-208-
-------
DRAFT
co
0
o
z
<
OZ
i
CO
- U
< O
C£
- O 3
^f < O
^52:
UJ > CO
OL > Q
ID o n
o i ^
"- CO ^
CO ^
UJ I-
^ <
O "-
Q
LU
M
CC C_D
UJ 3
CD CD
s
0
209
-------
DRAFT
Product: lonone and Methylionone
Process: Condensation and Cyclization of Citral
Process RWL Category: D
Chemical Reactions:
Citral + Acetone > pseudo- lonone
H SO
pseudo-lonone 2 k <. - lonone +{3-lonone
lonones and methylionones are used in perfumery and flavors; the beta
ionone isomer is an intermediate in the manufacture of vitamin A.
The production of ionones and methylionones involves two steps. First,
the pseudo-ionone is prepared by the condensation of citral obtained
from lemongrass oil; the condensation reaction uses either acetone or
methyl ethyl ketone to produce pseudo-ionone or pseudo-methylionone.
Second, these pseudo-ionones are cyclized with an acid catalyst. Com-
mercial ionones are generally mixtures of the alpha and beta isomer,
with one form predominating, although separations are sometimes made
through bisulfite compounds.
A typical process flow diagram for producing ionone/methyl-ionone is
shown in Figure k-k2. Citral, acetone/methyl ethyl ketone, sodium
hydroxide, and organic solvent are put into the first batch reactor,,
The solvent from this reaction step is recycled, and the product
vapors are condensed and stored in a receiver. The crude product
from the receiver is then distilled to remove the heavy-end residues
and washed with caustic solution to obtain pseudo ionone (or pseudo-
methyl ionone).
These materials are fed into the second reactor, where cyclization
is accomplished via a carbonium ion reaction with H_PO., HLSO, , or
BF, as catalyst. The reaction products proceed through a series of
washing tanks, where the products are quenched with sulfuric acid
and then washed by water, caustic solution, and acid solution.
After the washing steps, the product mixture is discharged into a
series of distillation columns, where the organic solvent and heavy-
end residue are recovered and withdrawn from the main product stream.
The product vapor leaving the last distillation column is condensed
and sent to a liquid-liquid separator for removal of water.
-210-
-------
DRAFT
The major water pollution sources of this process are wastewaters from
the various washing steps and from periodic reactor washings. During
the sampling visit, pseudo-ionone was being produced; hence the waste-
water samples obtained at the facility reflect ionone production RWL.
However, the RWL of methyl-ionone is believed to be of the same order
of magnitude as that of ionone. Process RWL's calculated from flow
measurements and the analyses of the wastewater streams are indicated
in the following tabulation:
PROCESS FLOW
liter/kkg 9,370
gal/Mlb 1,120
BOD RWL
mg/11 2,450
kg/kkg 23
COD RWL
mg/11 _ 10,000
kg/kkg 3k
TOC RWL
mg/11 3,520
kg/kkg 33
Raw waste concentrations are based on unit weight of
pollutant per unit volume of contact process waste-
waters .
2
Raw waste loadings are based on unit weight of pol-
lutant per 1000 unit weights of product.
The analytical results also indicate that, in addition to the parameters
shown in the tabulation, pollution parameters such as pH, sulfate, oil,
and chloride are at levels hazardous to biological treatment processes.
These data, shown above, were considered as commensurate with BPCTCA.
The wastes from this plant are discharged to a municipal treatment plant,
-211-
-------
DRAFT
>- co cc
< It
Q£ i l__
i o
^"^ UJ
c/3
LL. a. CD
§ f
^~ cr C_D ^
N LU =1
-i 3
5 t !_
Z
^ o l~*~-
^ P 3=^
LU ""C 'I
a: to T
V T 1
-J Z I
O "J !
LU ^
1 LU
LU -
z H r
O £
^C. C^ LU
O " »
__ 1 LU
1 CJ
5 t ^"-
-^
LU
Z
1 I
11 IS
A din 033 a
I hi i A T n c1
J ,
»
cc.
LU CO
i » ~ ^
-a: co ^
^ -a: -a:
^1
' 1
Ct=
CD
C-D
1 1 1
^^ CC.
. 1
I C_3
i » % cc:
LU =>
1 Ll_ C
CO 1 -
*c r^ c
^ CO
n is
t
cc.
LU
CJ3
CC
CD
CJ
U
CC
LU CD
CD
CC
LU
1
LU
1
CO
^
CC CO
LU «t
1 ^
CO C_3
CC
JJ
X
j (
rr
DZ:
CO
a: cc.
"E^
( - C_5
CO -I
M=C LU
1
LU
00 uu
C3 ^
CD -gr;
CD CC
1 1
1 Q_ 1
1 CO -I
_I
1
LU
CO
LU
II
CD
~^ ,
z
CD
CD
r h
"" 1 1 1 i S ' -* = uu
CD Ty
1 CO C_J
LU -y
CC
212
-------
DRAFT
Product: Methyl Salicylate
Process: Esterif ication of Salicylic Acid with Methanol
Process RWL Category: D
Chemical Reactions:
COOH
salicylic acid methanol methyl salicylate
Methyl salicylate, the methyl ester of salicylic acid, also known as
"oil of wintergreen" is used as a flavoring compound in mouthwash
and certain food products.
The esterification of salicylic acid is carried out in the presence of
a catalyst (such as sulfuric acid) and a solvent (such as methylene
dichloride or ethylene dichloride). Ethylene dichloride is generally
preferred, but the choice of the solvent depends to some extent upon
the boiling point of the desired ester. A flow diagram for the methyl
salicylate manufacturing process is shown in Figure k-kj,, and the
general procedure is described in the following paragraphs.
For each mole of salicylic acid, 3 moles of commercial methanol, 300
ml ethylene dichloride and 15 ml of concentrated sulfuric acid are
used. The mixture is refluxed in the reactor for 6 to 15 hours.
Progress of esterification is usually, but not invariably, indicated
by the development of cloudiness and separation of an upper layer
containing water, methanol, and sulfuric acid. When the reaction is
complete, the mixture is cooled, and the bottom ethylene dichloride
layer is washed successively with water, sodium bicarbonate solution,
and again with water. The ethylene dichloride layer, which contains
the product stream, is then distilled in vacuum or at atmospheric
pressure, and the residue methyl ester is further purified by dis-
tillation or crystallization before being packaged for subsequent
sale.
The aqueous upper layer formed in the reactor is recycled to the
distillation operation for recovery of methanol. The water generated
in the washing steps is discharged into a 1izuid/1iquid separator,
from which the organic phase is recycled back to washing towers.
-213-
-------
DRAFT
The major water pollution sources of the process are the waste streams
resulting from the washing steps. The steam jets used in pulling the
vacuum for distillation are discharged directly into the atmosphere
in the facilities visited and do not contribute to any pollution load-
ing in the contact wastewaters. Samples were obtained in the survey
period for analyses, but unfortunately, flow rates were not provided
by the manufacturer, and no RWL's are listed.
-------
DRAFT
o
u
<
u
_J
>-
CO
^4
LLJ
Of.
O
to
LI-
O
z
o
1
u
LL.
iii
LLJ
K-
to
U
to
X
»
LLJ
^
e_3 ^ 'LLJ
CO Z C_D GO
215
-------
DRAFT
Products: Miscellaneous Batch Chemicals
Intermediates
Dyes
Rubber Chemicals
Pi gments
Pharmaceut ica1s
Process: Numerous Batch Processes (Batch Chemicals Complex)
Process RWL Category: D
The RWL data presented and discussed in the following paragraphs were
obtained by sampling the total effluent from a large chemical plant.
This plant manufactures thousands of chemicals within the five general
classifications indicated above. In such a case, it is impossible to
sample each of the process operations on an individual basis. It should
also be noted that it will be impossible to effectively develop effluent
limitations for such a facility using specific limitations relating to
each specific batch process. A facility of this nature can be addressed
only in terms of its total aggregate production,,
Daily 24-hour composite samples were taken over a period of 9 days, fol-
lowed by three, 5-day composite samples over the next 15 days. The measured
flows and concentrations were put on a production basis by means of a
weekly production activity report supplied by the manufacturer. The RWL
data obtained in this manner are summarized in Table 4-13.
The RWL data shown in Table 4-13 represent the total raw waste from the
plant. Examination of the data based on the nine consecutive 2k-hour
composite samples indicates significant variation. For example, the
calculated mean for BOD is 20.4 kg/kkg of product, with high and low
values of 27.4 and 11.9 during the nine-day period.
It is significant to note that the three 5-day composite samples show
approximately the same range of variability. The calculated average
BOD raw waste load for the three periods is 24.1 kg/kkg, with high and
low values of 35.8 and 14.1 respectively.
This type of variation may be caused principally by the fact that in
this type of chemical plant it is impossible to relate production data
to the exact sampling periods, and, consequently, calculated ratios
will always have this type of error present.
Production at a batch chemical complex involves the startup and shut-
down of thousands of discrete batch processing operations on a day-to-
day basis. Although the plant maintains a materials inventory, this
-216-
-------
DRAFT
is usually updated only on a monthly basis. In this particular case,
the manufacturers expended considerable effort to provide weekly pro-
duction figures based on changes in the materials inventory.
However, it was still necessary to divide each of the weekly total
aggregate production quantities by seven to approximate a daily pro-
duction figure. In the case of the 24-hour composite sample data, it
was this production figure which was divided into the product of the
corresponding daily flows and pollutant concentrations to calculate
the RWL. Similar difficulties occurred with the 5-day composite
samples in that sampling periods did not match the periods for which
production quantities were developed.
In light of these difficulties, it would be informative to examine
some of the possible results which may occur in the application of
effluent limitations to this type of facility. The average of the
nine 24-hour composite RWL provides a number comparable in accuracy
to many of the RWL values assigned to other processes in this study.
The BOD and COD RWL values relating to this plant (from Table 4-13)
are:
Average of Nine 24-hour Composite Samples
BOD 20.4 kg/kkg
COD 77.2 kg/kkg
In later sections of this report, reduction factors based on the per-
formance of end-of-pipe treatment systems are developed. These numbers
were uniformly applied to the RWL developed for each sub-category of
process. The reduction factors for BPCTCA relating to BOD and COD are
92% and 69% respectively. If these reductions are applied to the
average RWL for the batch chemical complex indicated above, the follow-
ing effluent limitations will result:
BOD = (20.4)(0.08) = 1.63 kg/kkg
COD = (77.2)(0.31) = 23.9 kg/kkg
It should be noted and clearly understood that adjustment factors
based on treatment plant variability must then be applied to the
above values. The adjusted effluent limitations in terms of a maxi-
mum value for any one day are indicated as follows:
BPCTCA Daily Adj. Max.for Any
Parameter Effluent Limitation Factor One Day
BOD 1.63 kg/kkg 4.5 7.34 kg/kkg
COD 23.9 kg/kkg 3.4 81.3 kg/kkg
-217-
-------
DRAFT
It is possible to compare the actual treated effluent discharged by the
plant with the maximum one-day values, utilizing the data presented in
Tables 4-14 and 4-15.
Table 4-14 presents data on the performance of the activated sludge
treatment plant associated with the batch chemical complex during the
same period covered by the RWL sampling. Table 4-15 presents the
calculated effluent discharged from the complex on a production basis.
The effluent values shown in Table 4-15 were obtained by multiplying
the BOD and COD raw waste load values from Table 4-13 by the corres-
ponding reduction factors in Table 4-14.
Examination of the data presented indicates that in no case were the
BOD and COD effluents greater than the Maximum Values for Any One-Day
presented above. An examination of the data presented in Tables 4-13
and 4-14 shows that the variability in the actual treatment plant
performance (as % reduction) was less than that taken into account
in developing the daily adjustment factor for treatment plant per-
formance. Thus, the variability correction associated with the treat-
ment plant cancelled much of the variability exhibited by the process
plant.
Obviously, this simple example does not prove that there may be gross
inequities resulting from the application of production-based effluent
limitations derived from a very limited data base. However, it does
indicate that such limitations can serve as a meaningful basis for
negotiations in the development of an NPDES permit.
It is recommended that large batch chemical complexes which produce a
great number of specific commodities be approached in the manner il-
lustrated here. This should be done on an individual basis. In these
cases, it should be incumbent upon the manufacturer to develop production-
based RWL data for presentation to EPA. The time period should preferably
be longer than was possible for sampling during this study project.
A reduction factor (such as 92% for BOD) can then be applied to the RWL
to determine a production-based limitation. This limitation should then
be adjusted for variability depending upon the sampling period.
-218-
-------
Table 4-13
RWL Data for Batch Chemical Complex
DRAFT
Sample Period
24 hr. composite
11
11
11
11
Average
5-day composite
II M
Average
Flow
L/kkg
65,300
74,800
90,100
89,300
71,300
113,000
68,800
70,400
78.700
80,100
70,600
65,500
100,000
78,700
BOD
(kg/kkg)
19.6
16.5
27.0
27.4
23.8
27.4
11.9
13.0
16.9
20.4
22.4
14.1
35.8
24TT
COD
(kg/kkg)
63.5
68.4
90.1
96.4
83.9
99.4
55.3
60.4
77.0
77.2
80.9
76.6
144.
100.5
TOC
(kg/kkg)
17.0
19.5
23.4
29.9
22.1
29.2
17.9
23.9
23.6
22.9
27.1
27.1
40.1
3T3
-219-
-------
DRAFT
Table 4-14
Treatment Plant Performance Data for Batch Chemical Complex
Treatment Plant Utilizing the Activated Sludge Process
BOD COD TOC
24-hr, comp.
24-hr, comp.
24-hr, comp.
24-hr, comp.
24-hr, comp.
24-hr, comp.
24-hr, comp.
24-hr, comp.
24-hr, comp.
5-day comp.
5-day comp.
5-day comp.
Inf.
mg/1
300.
220.
300.
307.
333.
243
173.
185.
215.
318.
215.
358.
Eff.
mg/1
29.
21.
18.
14.
14.
15.
9.
17.
35.
25.
32.
23.
% Red.
90.3
90.5
94.0
95.4
95.8
93.8
94.8
90.8
83.7
92.1
85.1
93.6
Inf.
mg/1
972.
914.
1000.
1080.
1176.
882.
804.
860.
980.
1147.
1171.
1439.
Eff.
mg/1
311.
295.
310.
270.
333.
363.
314.
350.
420.
330.
278.
272.
% Red.
68.0
67.7
69.0
75.0
71.7
58.8
60.9
59.3
57.1
71.2
76.3
81.1
Inf.
mg/1
260.
260.
260.
335.
310.
259.
260.
340.
300.
385.
415.
400.
Eff.
mg/1
75.
67.
79.
68.
104.
104.
83.
125.
109.
90.
91.
64.
% Red.
71.1
74.2
69.6
79.7
66.5
59.8
68.1
63.2
63.7
76.6
78.1
84.0
NOTE: These performance data were obtained during December from a plant located
in the Northeatern U. S. The high removal efficiencies indicate that
treatment technology is available to design a highly efficient biological
treatment facility which will operate satisfactorily during winter conditions,
Historic performance data from this plant are reported in Table 7-2 as
Plant No. 3
-220-
-------
DRAFT
Table 4-15
Effluent Discharged From Batch Chemical Complex
After Biological Treatment
Sample Period
24-hour composite
BOD
COD
TOC
(kg/kkg) (kg/kkg) (kg/kkg)
11
11
n
11
n
11
BPCTLA Effluent Limitation
Maximum for any one day
1.90
1.57
1.62
1.26
1.00
1.70
0.62
1.20
2.75
7.34
20.3
22.0
27.9
24.1
23.7
41.0
21.6
2k.6
33.0
81.3
91
03
11
07
4,
5,
7,
6,
7.40
11.7
5.71
8.80
8.57
-221-
-------
DRAFT ^
Product: Monosodium Glutamate (MSG)
Process: Batch Fermentation of Beet Sugar Molasses
Process RWL Category: D
Chemical Reaction:
Fermentation (Glutamic Acid Production)
Beet Sugar + Bacteria Culture + CL + NH >
(sucrose)
NH.-CH-COOH + H,0 + CO + Bacteria Mass
I } £ i
CH2-CH2-COOH
glutamic acid
MSG Conversion and Neutralization
NH -CH-COOH + NaOH > NH -CH-COOH + HO
2. , I } t
CH2-CH2-COOH CH2-CH2-COONa
mono sodium glutamate
Typical Raw Materials
Beet Sugar Molasses
Ammonia (NH,)
Bacteria Culture
Nutrient Salts
Compressed Ai r
Diatomeceous Earth Filter Aid
Hydrochloric Acid (HC1)
Sodium Hydroxide (NaOH)
Steam
Cooli ng Water
Bleaching Activated Carbon
Monosodium Glutamate (MSG) is an amino acid salt which is used as a
flavor enhancer. MSG is produced by the conversion of Glutamic Acid (GA)
with caustic. There are several routes to obtaining the GA, including
hydrolysis of Steffen's Waste Liquor with caustic or acid, acid hydrolysis
of wheat and corn gluten, and fermentation of a carbohydrate.
-222-
-------
DRAFT
The process visited during the field data collection program was the
fermentation of beet molasses (sucrose) to produce glutamic acid. This
fermentation reaction is a batch reaction; subsequent separation processes
and the conversion of the glutamic acid to monosodium glutamate are done
on a semi -conti nuous basis (refer to Figure k-
The aerobic fermentation of the sugar beet molasses occurs in a jacketed
vessel, which is either steam-heated or water-cooled to maintain tempera-
ture. The cooling water may be recirculated without organic pickup. The
pH of the fermenting liquor is monitored and controlled by ammonia
addition. The added ammonia also supplies the amino group in the product
formulation and supplements the nitrogen content of the raw molasses for
bacteria culture growth. The surplus air from the fermentation vessels
contain odorous compounds; therefore, these gases are sent to the boilers
for air makeup, and the odorous materials are thermally destroyed.
At the completion of the fermentation reaction, the bacteria cell mass
(cell cream) is separated via centrifuges and discarded (point No. 2);
this cell mass represents the most significant part of the raw waste load
The high pollutant loading and commensurate high substrate content of
this waste stream make it a prime candidate for by-product recovery and
sale if a market can be found.
The clarified centrate is heat treated to precipitate miscellaneous
protei naceous material. The precipitated material and any fugitive cell
mass are then removed via vacuum filtration. The resulting sludge is
cyclone classified to recover most of the filter aid material. The waste
stream (point B, appcor cyclone overflow) is characterized as heavily
contaminated with inert and organic suspended solids. Intermittent dis-
charge of filter precoat material impacts considerably on the RWL for all
parameters measured (point No. 3).
The clarified filtrate (glutamic acid solution) is concentrated via a
two-stage evaporation process. The condensate of the first effect is
recycled as water make-up to the fermentation process. The second-stage
vapors are condensed in barometric contact condensers. The barometric
condenser system is serviced with recirculated cooling-tower water. The
blowdown (point No. 7) from this cooling-tower system was found to be
highly contaminated with materials exerting a significant biochemical
oxygen demand.
The concentrated glutamic acid (GA) solution is treated with concentrated
hydrochloric acid to a pH of 3.2, the isoelectric point of GA. The GA is
crystallized out of solution, filtered, and washed. The wash water,
-223-
-------
DRAFT
which represents the major blowdown of impurities from the system, is
recovered and sold as an animal feed supplement, "Dynaferm". The
primary constituent of Dynaferm is GA, but it also contains many other
salts and proteinaceous impurities which escape the heat coagulation/
filtration treatment. This stream was not sampled during the data-
gathering survey, but its recovery and sale probably impact very favorably
in reducing the RWL.
The GA crystals are then solubilized and partially converted to MSG with
caustic. The GA/MSG solution is decolorized with activated carbon. The
spent carbon is recovered and thermally regenerated; this represents a
significant in-process pollution control measure because the color bodies
and other adsorbed impurities are oxidized in the regeneration furnace.
There are three small water discharges (points No. k, 5 and 6) which do
not significantly impact on the RWL.
Potential odors in the exhaust gases from the carbon regeneration
facilities are oxidized via a thermal afterburner. The three discharges
(point D) from the carbon regeneration are significant in volume but
contribute little of any of the pollutants monitored during the survey.
Further downstream processing includes: pH adjustment, filtration, MSG
crystallization, and separation of water and GA. These streams are
recycled in order to retain valuable product and by-product materials.
The results of combining the two grab composites of all major contributing
sources to wastewater discharges yielded values which compare to 62% and
11% BOD occurrence probability of the historical data; this indicates
that the samples are representative of the MSG production facility surveyed.
The RWL can be reduced significantly from those levels of pollutants
measured during the data-gathering survey. The approach with the greatest
potential for reducing the RWL is the recovery and sale of cell cream
(point No. 2) for animal feed. It is not possible to mark the impact of
cell cream recovery precisely, but it is estimated that approximately 20
to 50 percent of the suspended solids RWL could be removed. A somewhat
similar impact on the organic and oxygen-demanding parameters would also be
experienced. The prospects for recovery of this material depend on the
development of nearby markets and favorable economics.
The recovery of the precoat wastewater (point No. 3) is also an
attractive alternative to discharge. This material, soluble and suspended,
could be recirculated to achieve almost total recovery.
The RWL for MSG production during the survey and the estimated impact
on RWL in regard to recovery of cell cream and precoat wastes are shown
in the following tabulation:
"22k-
-------
DRAFT
Sample Sample RWL After
Process Flow Peri od #1 Period #2 Reduction
liter/kkg 67,000 67,000 62,200
gal/Mlb 8,030 8,020 7,460
BOD 1
mg/1 1,510 1,020 980
kg/kkg2 101 68.4 61
COD
mg/11 4,060 4,410 3,600
kg/kkg2 272 296 224
TOC
mg/l1 1,360 1,350 1,090
kg/kkg2 91.4 90.5 67
SS
mg/11 2,260 305
kg/kkg2 151 19
Raw waste concentrations are based on unit weight of pollutant
per unit volume of contact process wastewaters.
Raw waste loadings are based on unit weight of pollutant per
1,000 unit weights of products.
-225-
-------
DRAFT
CO
LU
CO
CO
<
o
CO
O
z
o
o £
Ll_ LU
LU
1
<
O
o
co
2L
CD
O_
HFAT+-COAGULATION
ANO FILTRATION p »
i ,
1
3=.
T L
1
CD
1 CO
S LU
il" CO
i_I_ CO
Ll_ ="
RECYCL
LU
CO
LU
cr
:z.
CD
LU
CO
t
1
3
CJ
LU
r-
OO QC CD
>-«!=.
CC CCL _
CD
CE
CO CC CD
226
-------
Product: Naphthenic Acid
Process: Extraction and Acidification of Caustic Sludge from Petroleum
Refi nery
Process RWL Category: C
Chemical Structure: Naphthenic acids are cyclo-paraffinic organic acids
and usually are mono-carboxylic
The term naphthenic acids is applied to the mixture of carboxylic acids
obtained from the alkali washes of petroleum fractions. They are com-
plex mixtures of normal and branched aliphatic acids, alkyl derivates of
cyclopentane- and cyclohexane-carboxylic acids, and cyclopentyl and cyclo-
hexyl derivates of aliphatic acids. Naphthenic acids are used chiefly in
the form of metallic salts which are soluble in oils and organic solvents.
Copper naphthenate is an excellent fungicide for wood and canvas treating;
the lead, manganese, zinc, and iron salts are used as dryers (oxidation
catalysts) for paints and varnishes.
The napthenic acids are present in the caustic sludge primarily as sodium
napthenates. Figure ^-^5 is a process flow diagram of naphthenic acid
recovery. The caustic sludge, which has been used to scrub the light
distillate crude oil fractions, is treated with water and an alcohol
soap solvent. The resulting oil and soap phases are separated; the oil
phase is stripped with the oil going to fuel and the solvent recovered
overhead; the soap phase is treated with an oil solvent, and another oil-
phase/soap-phase separation takes place. The oil phase is recycled to
the caustic sludge, while the soap phase is treated with an acid. The
naphthenic acids are then separated from the solvent.
The solvent proceeds to a stripper where it is recovered as the overhead
and sent to solvent storage; the bottoms from this column are quenched
with water and discharged to the sewer. The extracted naphthenic acids
go to a stripper, where solvent is taken overhead and sent to storage;
the bottoms from the stripper are vacuum distilled. The bottoms from
the vacuum still are sent to fuel, while the naphthenic acids are re-
covered.
Process RWL calculated from flow measurements and the analyses of waste-
water streams are indicated in the following tabulation. In addition,
the sulfate concentrations in the wastewaters are extremely high and are
hazardous to biological treatment processes.
-227-
-------
PROCESS FLOW
liter/kkg 39,800
gal/M Ib it,760
BOD5 RWL
mg/liter1 3,550
kg/kkg2
COD RWL
mg/liter1 7,500
kg/kkg2 298
TOC RWL
mg/liter1 2,630
kg/kkg2 10^
1 Raw waste concentrations are based on unit weight of pollutant
per unit volume of contact process wastewaters.
2
Raw waste loadings are based on unit weight of pollutant per
1,000 unit weights of product.
These data were considered as BPCTCA.
The process plant visited during the field data collection program used
direct-contact cooling to reduce the temperature of the bottoms stream
from the soap stripper. The substitution of non-contact cooling would
reduce the water requirements for this process by approximately 75 per-
cent. However, the raw wasteload would not be appreciably affected.
-228-
-------
DRAFT
I- C*
U Z
u_ EE
9 £
U ^
"^ Z>
O LLJ
z X
iO <
a < o
^ p u_
O >< UJ
LZ O
9 2
% y
U co
Z ^
S O
<
z
-X _1 CD O
C_D OO CO GO
229
-------
DRAFT
Product: o-NitroaniHne
Process: Ammonolysis of o-Nitrochlorobenzene
Process RWL Category: D
Chemical Reaction:
HO
N02C6h^Cl + NH3 (excess) > NO^H^Nh^ + NH^Cl
Ammonia o-Nitroani1ine + ammonium
, , . ch1 ori de
o-nitrochlorobenzene
The manufacture of o-nitroani1ine is similar to that of p-nitroani1ine,
which is described in the next section. A typical process flow diagram
is shown in Figure 4-46.
The o-nitrochlorobenzene and ammonia enter the batch reactor, and
o-nitroani1ine is formed. The excess ammonia in the reaction mixture is
distilled off the absorbers, from which ammonia is recycled back to the
reactor. The aqueous stream from the distillation step goes to a
washing operation and proceeds to dehydration and refining operations.
Both the dehydration and the refining steps are performed under vacuum
conditions, and steam jets with barometric condensers are generally used
in pulling the vacuum for the operation. The refined o-nitroani1ine may
be used directly or it may be further transformed into a flake.
The major water pollution sources of the process are wastewaters with-
drawn from the scrubber in the ammonia-recovery system, from the washing
step, and the condensates from barometric condensers. Process RWL's
calculated from flow measurements and analyses of the wastewater streams
are indicated in the following tabulation. The analytical results also
indicate that waste streams contain high concentrations of nitrogen and
chloride which are hazardous to biological treatment processes.
-230-
-------
DRAFT
PROCESS FLOW
liter/kkg 269,000
gal/M Ib 32,200
BOD RWL
mg/11 61
kg/kkg 16
COD RWL
mg/11 391
kg/kkg 105
TOC RWL
mg/11 115
kg/kkg^ 30.9
Raw waste concentrations are based on unit
weight of pollutant per unit volume of
contact process wastewaters.
2
Raw waste loadings are based on unit weight
of pollutant per 1000 unit weights of products,
-231-
-------
DRAFT
LLI
z
iu
N
Z
UJ
CO
o
ot
O
u
O
LU
« CO
1 O
"- z
o
I
UJ
Z
-j
Z
<
o
at
£
Z
6
232
-------
DRAFT
Product: p-NitroaniIine (PNA)
Process: Ammonolysis of p-Nitrochlorobenzene
Process RWL Category: D
Chemical Reactions:
HO
N02C6H4C1 + NH (excess) > NOjCgH^Nn^ + Nh^Cl
Ammonia
ammonium
p-nitrochlorobenzene p-nitroani1ine chloride
p-Nitroaniline is an important intermediate for dyes and pigments and
also for the preparation of numerous antioxidants and antiozonants of
the N-substituted p-phenylenediamine type.
A typical process flow diagram for batchwise manufacturing of
p-nitroani1ine is shown in Figure k-kj.
p-Nitroani15ne is manufactured by heating p-nitrochlorobenzene with aqua
ammonia at 175°C under pressure. A jacketed autoclave provided with
efficient stirrers should be used as the reactor. Molten p-nitrochloro-
benzene is added to aqua ammonia (28%) and heated gradually; the
temperature is increased over a period of 3 hours to 175 C at a pressure
of 530-580 psi, and these latter conditions are maintained for 16 hours
to complete the reaction.
Some of the ammonia gas is then vented to an absorption system, and the
excess ammonia in the reaction mixture is distilled off to absorbers,
from which it is recycled back to the reactor. The aqueous reaction
mixture is passed through a pressure filter and sent to wooden crystalizing
tubs, where the p-nitroani1ine separates as a finely divided, canary-
yellow crystalline mass. After cooling to 30 C, the solid product
(averaging 99% purity) is removed by centrifuging. The centrate passes
through an elaborate system of catch boxes to cool it to room temperature,
and additional product settles out. This is recovered periodically by
siphoning the water from the catch boxes.
The solid cake leaving the centrifuge is then passed through drying and
packaging steps to obtain final product. Dust collectors and scrubbers
are used in the drying steps to remove the dust in the vent gases.
-233-
-------
DRAFT
The major pollution sources of this process are wastewaters withdrawn
from scrubbers, filtration units, and the crystallization unit. Process
RWL's calculated from flow measurements and analyses of the wastewater
streams obtained in the survey period are shown in the following tabula-
tion. The analytical results also indicate that, in addition to the
parameters shown in the tabulation, pollution parameters such as nitrogen,
chloride, calcium are at levels hazardous to biological treatment processes.
PROCESS FLOW
liter/kkg 39,100
gal/M Ib 4,680
BOD RWL
mg/11 ? 65
kg/kkg 2.55
COD RWL
mg/11 2,030
kg/kkg 79.1
TOC RWL
mg/11 570
kg/kkg 22.2
Raw waste concentrations are based on unit
weight of pollutant per unit volume of
contact process wastewaters.
Raw waste loading are based on unit weight
of pollutant per 1000 unit weights of product.
These data are considered as BPCTCA.
Continuous processes can also be employed for ammonolysis of p-nitrochloro-
benzene. In such a process, preheated aqueous ammonia is forced through
an inlet pipe to the bottom of the vertical reaction cylinder, then passes
upward through an annular space, and overflows through a central outlet
pipe leading to a still, where excess ammonia is stripped off before the
product is crystallized. The space above the overflow pipe provides a
vapor space to absorb fluctuations of the internal pressure. The lower
part of the cylinder, where the major part of the reaction occurs, is
provided with packing and is constructed of stainless steel to minimize
corrosion. The operating pressure is about 1200 psi.
-23^-
-------
DRAFT
z
LLJ
CO
O
Q£
o
X
u
Q£
h-
z
<
Q£
<
n
5 -
4 0
II 1 ***
at: co
"^ S
J >-
£ o
^ Z
O
^
^
<
1
UJ
z
_J
z
<
o
0£
1
z
1
rrv
a:
£«
UJ
=
1
CC
LLJ
I
^
GO
-=c
%
1
«i cr:
C3 C3
^ CO
^ LLJ
«r or
t
c
n
c.
t
«a
.
el
Z.
Z3
E
F^
c
<:
4
C
C
- t-
3 C
- «
C LJ
1
C3
CC
1
I
DC
> 1
3D
33
_C
^)
S3
-»
E.
J)
r
LJ
OBENZFNE *
cd
u_
«t
%
U_l
1
00
-=c
^
1 J
1
e_1 LJ_
cr. czi
-IT) 31
i «r
CO 1
3K
«r
CD
CC
1
1
t
I
Dc=
C3^
ZS
>^CJ>
s-
1
cr
«a:
rxj
j
«r
i
a^
CO
t
z.
\
3:
CC
Ll_
1
L
K. ^
C
C
h
:
Lj
0-
«;
3
DZ.
j_£
a;
x^
£
c.
r
c
_i
0
c
en
LU
*a:
^
1
\
\
«c
en
i
u_
t i
>
CC
1 1 1
LU
CO
235
-------
DRAFT
Product : Pentachlorophenol
Process : Chlori nat ion of Phenol
Process RWL Category: D
Chemical Reaction: C,H,_OH+5C10 - > C Cl OH+5HC1
03^05
Pentachlorophenol is widely used as a wood preservative, especially in
such applications as residential construction where creosote would be
undesirable. It is also used as a fungicide in paint and adhesives,
and i n paper mills.
Although pentachlorophenol can be manufactured by continuous processes,
it is generally produced by batch reaction in the facility visited in
the survey period. A simplified process flow diagram for the production
of pentachlorophenol via the chlorination of phenol is shown in Figure
Phenol and chlorine are fed to the reactor, where the chlorination reaction
occurs. The product of the first step of the reaction is t ri chlorophenol ,
which is further chlorinated to form pentachlorophenol.
The product mixture leaving the second reactor is discharged into a quench
vessel, in which the gases are separated from the molten product stream.
The gases proceed to the HC1 absorber and caustic scrubber, while the
liquid stream is solidified in a flaker. The solidified pentachlorophenol
is then passed into a rotary kiln for glazing.
The major water pollution sources of this process are the waste streams
generated at the caustic scrubber, flaker, and glazing units. Process
RWL's calculated from flow measurements and analyses of the samples
obtained in the survey period are shown in the following tabulation.
Sampling Period #1 Sampling Period #2
PROCESS FLOW
liter/kkg 2,960 2,560
gal/M Ib 35^ 35^
BODc RWL
mg/11 330 306
kg/kkg2 0.975 0.906
COD RWL
mg/11 5,7^0 6,020
kg/kkg 17 17.9
TOC RWL
mg/11 768 781
kg/kkg2 2.27 2.31
'Raw waste concentrations are based on unit weight of pollutant per unit
volume of contact process wastewaters.
2Raw waste loadings are based on unit weight of pollutant per one thousand
unit weight of product.
-236-
-------
DRAFT
The analytical results also indicate that parameters such as chloride,
phenol, and sulfate are at levels hazardous to biological treatment
processes. The low BODr values shown in the foregoing tabulation are
possibly caused by the interferences of these biological-inhibitory
pol1utants.
The demand for pentachlorophenol in the U.S. is approximately ^0 million
pounds per year, of which 15 million pounds are used in mixtures with
creosote. Pentachlorophenol can also be made by caustic fusion of hex-
achlorobenzene, which is a cheaper route but gives a product contaminated
with NaCl.
-237-
-------
DRAFT
LU
O
z
LU
I
O_
Z
o
00 ±
<
Z
C£
O
u
O
Z
LLJ
Q_
O
x
u
z
LU
Ou
3
1
«=C GO
CD CO
0=
LU
aaaaosav
OH
-=y GO
LU CO
T
238
-------
DRAFT
Product: Pigments
Process RWL Category: D
Pigments are various organic and inorganic insoluble substances. They
are used in surface coatings, and also in the ink, plastic, rubber, ceramic,
paper, and linoleum industries to impart color. A large number of pigments
are produced because different products require a particular choice of
material to give maximum coverage, ^economy, opacity, color, durability,
and desired reflectance.
Organic pigments may be subclass ified into toners and lakes. Toners are
insoluble organic dyes that may be used directly as pigments because of
their durability and coloring power. Toners are used in paints, printing
inks, and wallpapers, and especially for the pigment printing method for
textiles, employing pigments of metal phthalocyanines and other types.
Lakes result from the precipitation of organic colors, usually of synthetic
origin, with salts of Ca, Ba, Cr, Al, or phosphomolybdic acid. The dye
molecule frequently contains -OH or -S03H groups. Such lakes, when ground
in oil or other media, form the pigments of many of our paints and inks.
Some basic dyes are used for the tinting of paper in a water-dispersed
form or phosphomolybdic (or tungstic) acid lakes. Wallpapers are fre-
quently colored with lakes from basic dyes containing a sulfonic group.
The use of pigments in the "dope" before spinning rayon, acetate, and
synthetic fibers is growing rapidly and is producing excellent colors
of outstanding all-around fastness.
One facility visited during the field survey produced toners for the pigment
printing of textiles. Figure k-k9 is a process flow diagram for the manu-
facture of this product. Hydrochloric acid, sodium nitrite, water, an
amine, and sulfamic acid are put into the first reactor. The coupler is
prepared in the second reactor; this entails the addition of a coupling
agent, sodium hydroxide, water, and steam. The reaction products from
both reactors then proceed to a third reactor, where they are combined
with acetic acid, sodium acetate, and water. The pigment which forms
in this reactor is filtered. The solid phase is recovered as the pigment
dye, while the mother liquor is discharged into the sewer.
Samples were obtained on two successive days to characterize the waste-
water from this process. The samples for the two days indicate that the
characteristics of the wastewater are quite variable, attributable mainly
to the batch nature of the operation. The second plant produced lakes
for use in paints.
The following tabulation of process RWL, calculated from flow measurements
and the analyses of the wastewater stream, includes data obtained for both
plants.
-239-
-------
DRAFT
Plant 1 (Toners) Plant 2 (Lakes)
1st Day 2nd Day
PROCESS FLOW
Hter/kkg 313,000 313,000
gal/M lb 37,500 37,500 i,uou,uoo
BODr RWL
mg/11 3,000 61*0 1 ,W
kg/kkg2 9^0 200
COD RWL1
mg/1 _ 11,300 1,000 4,930
kg/kkg^ 3,520 315
TOC RWL
mg/11 2 750 570
kg/kkg 235 178 ^
Raw waste concentrations are based on unit weight of pollutant per unit
-volume of contact process wastewaters.
Raw waste loadings are based on unit weight of pollutant per 1,000 unit
weight of product.
The average of the two plants was considered as BPCTCA.
-240-
-------
DRAFT
O
Z
_j
a.
ID
O
u
<
Z
o ^
= o
N
o
Q_
ac LU u_
co ^ co
-------
DRAFT
product: Plasticizers
Process: Condensation of Phthalic Anhydride
process RWL Category: D
Plasticizers are organic chemicals that are added to synthetic resins to
improve workability during fabrication, to modify the natural properties
of these resins, and to develop new improved properties not present in
the original resins. This type of chemical is manufactured by liquid-
phase batch reactions.
A generalized process flow diagram is shown in Figure 4-50, while the
overall chemical reaction is given below:
C6\C2°3 + 2ROH > C6H4C2°2(OR)2 + H2°
phthalic anhydride alcohol phthalate
Feed materials, an alcohol and an anhydride, along with acid catalyst,
are fed into a reactor. The esterification is carried out at a pressure
of about 10 psig and at temperatures ranging from 104 to 356 F, for six
to 20 hours, depending on the feed and the desired product. The reactor
effluent is either sent directly to a wash tank or passed through a
filter press to remove solids, depending on whether activated charcoal or
a catalyst neutralizer is being added to the reactor.
Caustic soda (or soda ash) is used in the wash tank to remove unreacted
acid and anhydride. Wastewater from the washing is sent to a series of
settling tanks before being discharged to the sewer. Oil removed in the
settling tanks is recycled back to the reactor.
The plasticizer stream from the wash tank flows to a stripper, where any
remaining alcohol is taken overhead and recycled to the batch reactor.
The main stream from the stripper is further polished by an activated
carbon filter to obtain 99.5+% purity plasticizer.
Because the feed materials and desired products change, a large quantity
of water for intermittent reactor clean-up is required. This clean-up
water is the major source of wastewater pollution. The other two waste-
water streams are wash water from the wash tank and backwash water from
the filters. A high RWL for this type of operation is expected because
of high loss of organics during the wash operations.
-242-
-------
DRAFT
o
CL
a
u
_l
<
Q_
O it.
LU
O
o
u
CO
a:
LJJ
N
U
t
to
-243-
-------
DRAFT
Product: Tannic Acid
Process: Extraction of Natural Vegetable Matter
Process RWL Category: D
Tannic Acid, a glucoside of gallic acid, can be obtained by extraction
of natural vegetable matter with water. The water extract is then con-
centrated and upon drying yields technical tannin, which is used as a
mordant in dyeing and as a source of gallic acid. Extraction of nutgalls
with alcohol or other bleachings, or extraction with mild reducing agents
(such as sodium bisulfite) and evaporation of the extract yield medicinal
grade tannin. Tannic acid is commonly used for burns, as an astringent,
in gargles, and to precipitate proteins in wineries and breweries.
Figure 4-51 shows a typical flow diagram for tannic acid manufacture.
The raw materials (natural nutgalls) are fed into the preparation section
and ground into small pieces. The ground material is then batchwise ex-
tracted with an organic solvent. The slurry raffinate phase is diluted
with water, put through a steam stripper for solvent recovery, and dis-
charged into the sewer. The extract phase, a mixture of organic solvent,
extracted material, and a small amount of water, is steam stripped for
removal of the solvent. The solvent is then condensed and recycled to
the extraction step.
The steam-stripped aqueous solution containing tannic acid is cooled and
passed through a filter press for removal of suspended solids. The
filtrate (containing tannic acid) is concentrated in an evaporator before
final drying and packaging. The evaporation step employs steam jets to
pull a vacuum, and the gases are subsequently condensed in barometric
condensers. Waters from the barometric leg are recycled to the cooling
towers for reuse.
The major pollution source of the process is the slurry waste stream with-
drawn as the bottom of the raffinate stripper. The intermittent reactor
washings, as well as "battery limits" clean-up, also contribute to the RWL
of the process. The amounts of contaminants were estimated to be 5 per-
cent of that of the major waste stream. The results of the sampling survey
are summarized in the following tabulation.
-------
DRAFT
Tannic Acid
Sample Sample Sample
Process Flow Period #1 Period #2 Period #3
liter/kkg 10,000 10,000 10,000
gal/M Ib 1,200 1,200 1,200
BOD5 RWL
mg/11 16,100 14,700 15,100
kg/kkg2 161 147 151
COD RWL
mg/11 109,000 99,400 112,000
kg/kkg2 1,093 995 1,120
TOC RWL
mg/11 14,000 16,800 21,000
kg/kkg2 140 168 210
Raw waste concentrations are based on unit weight of pollutant per
unit volume of contact process wastewaters.
Raw waste loadings are based on unit weight of pollutant per 1,000
unit weights of product.
The water from the barometric condensers in the evaporation step was not
included in the foregoing RWL calculations. Since the organics in the
evaporation step are not volatile, the pollutant loading in the barometric
water is low. Furthermore, this water has been totally recycled for re-
use in the process.
The analytical results indicate that the high RWL of this process is
attributable to the high suspended organic vegetable matter present in
the slurry stream. The removal of suspended solids from this stream can
substantially reduce the RWL of the process. As was indicated during the
sampling survey, the following three possible methods for removal/disposal
of suspended solids are being investigated.
1. Filtration and landfill disposal of Suspended Solids (SS).
2. Filtration and incineration of SS.
3. Filtration and recycle of SS into other industrial and/or
agricultural uses.
The RWL of the process should be based on the amounts of contaminants in
the filtrate, and is subject to further investigation.
-2k5-
-------
DRAFT
Non-contact cooling water is employed in the solvent recovery area and
prior to filtration. The total quantity of non-contact cooling water
is estimated to be approximately 6^1 Ibs per Ib of product.
-21*6-
-------
DRAFT
CO
o
UJ
c* O
O^
_ O
U- H=
U
i
UJ
Q
U
U
Z
-247-
-------
DRAFT
Product: Vani11 in
Process: Alkaline oxidation of spent sulfite liquor
Process RWL Category: D
Chemical Reactions:
Spent sulfite liquor + 02 + NaOH > C^H (OCH ) (CHO) OH
(1ignosulfonic acid) vanillin
Vanillin is one of the most widely used food flavors. It is also used
in perfumery and in the deodorizing of manufactured goods. During the
plant visit, company representatives indicated that approximately one-
third of the vanillin produced at this plant is used for flavor manu-
facture, with the remaining two-thirds used for perfumes and other
miscellaneous items.
A typical process flow diagram for the manufacture of vanillin is shown
in Figure 4-52.
Waste sulfite liquor containing about 15 percent sulfite solids (mainly
1ignosulfonic acid), supplied by paper mills, is treated with lime in a
series of three tanks. In the first stage (pH = 10.5) calcium sulfate
(principally) precipitates; in the second, the pH is increased to 12.0
to precipitate calcium 1ignosulfonate; in the third a further excess of
lime is added. Thickened liquors from the third stage are then reused
as a supplementary lime source in the first tank. The precipitated calcium
1ignosulfonate from the second tank is filtered under vacuum, and redis-
solved in a caustic soda solution to yield a solution containing 3.5 per-
cent lignin solids and 10 percent caustic soda.
The alkaline solution is then pumped to a falling-film contactor counter-
current to a carefully controlled flow of air, which oxidizes the sulfonate
to sodium vanillate. In one application of the process, the oxidation
reaction is carried out at 1,500 psi and 225 C. Residence time of the
liquid in the contactor is of the order of k minutes, with an overall
liquor-oxygen ratio of 0.01 volume of liquor per volume of air (STP).
At this rate, an 18 to 20 percent conversion of lignin to vanillin is
effected, and overoxidation is minimized. Because the reaction is exo-
thermic, the temperature of the effluent liquor rises to about 250°C.
The vanillate from the oxidizer is then extracted with organic solvents,
such as birtanol or isopropanol, by a conventional one- or two-stage
countercurrent process. Both the extracted and raffinate phases are steam
stripped to recover, respectively, organic solvents and vanillate, which
are then recycled back to the process line. The vanillate separated from
the organic solution is further purified by vacuum extractive distillation
followed by acidification. Vanillin is then recovered by vacuum crystal-
ization, centrifuging, and vacuum tray-drying,
-248-
-------
DRAFT
The air oxidation process described has replaced a former process in
which nitrobenzene was used as the oxidizing agent. Also, a Canadian
plant has installed an air oxidation process similar to that described,
but with lime replacing caustic as the alkaline agent. Other differences
are the use of carbohydrate-free waste sulfite liquor from an alcohol
plant, the use of toluene as the extracting agent, and vacuum distillation
to separate vanillin from contaminating by-products.
The major pollution sources of the process are waters discharged from
the pretreatment of sulfite solution, the raffinate stripper, the centri-
fuge filtration, and the steam jets connected with the barometric conden-
sers. Additional process wastewaters are generated during spills and
washings. Process RWL's calculated from the flow measurements and analyses
of the aforementioned waste streams are presented in the following tabu-
lation.
Sample period #1 Sample period #2
PROCESS FLOW
liter/kkg 133,000 133,000
gal/Mlb 15,900 15,900
BO05 RWL
mg/11 17,900 17,500
kg/kkg2 2,380 2,320
COD RWL
mg/]1 118,000 116,000
kg/kkg2 15,600 15,^00
TOC RWL
mg/11 30,400 29,900
ko/kkg2 4,030 3,960
^Raw waste concentrations are based on unit weight of pollutant per
unit volume of contact process wastewaters.
"~R.-!'A waste loadings are based on unit weight of pollutant per 1,000
unit weights of product.
The high raw waste loads shown in the tabulation may be explained by the
fact that spent sulfite liquor is a waste product from paper mill opera-
tions containing 15 to 20 percent 1ignosulfonic acid. The reaction re-
alizes a maximum of 15 to 20 percent yield, and the remaining unusable
spent sulfite liquor must be discharged as a waste. The analytical results
also indicate that, in addition to the parameters shown in the tabulation,
pollution parameters such as dissolved solids, sulfate, and phenol con-
centrations are at levels hazardous to biological treatment processes.
-------
DRAFT
No effluent limitations are proposed for this process because it is
utilized by only one manufacturer in the United States.
The primary non-contact wastewater flows are cooling tower and boiler
blowdown. Samples of the boiler blowdown have been taken, and the flow
has been estimated. Samples of the cooling tower blowdown were not
taken, because typical analyses were available and the flows are highly
variable. The cooling tower make-up averages approximately 120 gpm and
the cooling water makes approximately 3 cycles. A phosphate corrosion
inhabitor and a biocide are added to the make-up.
-250-
-------
DRAFT
o
o
__l
y
LL.
ID
to
h-
Z
LU
sfc
-o
LU
a: -7
Q
X
O
UJ
Z
<
I
'Z.
Z
<
>
-251-
-------
DRAFT
SECTION V
WASTE CHARACTERIZATION
The process RWL data obtained for each of the 5^ Secondary Organic Prod-
ucts were discussed previously in Section IV - Industry Categorizations.
These descriptions related the raw waste flows and loadings to specific
sources such as chemical conversions and unit operations within each
product/process grouping. The discussions in this section relate the
single net of RWL values assigned to each product/process grouping, and
compare waste loadings and concentrations between product/process group-
ings.
Tables 5-1, 2, 3, and k list the single set of BPCTCA-RWL values which
has been assigned to each product/process grouping. These values in-
clude the following parameters:
Contact Process Wastewater Flow (liters/kkg of product)
BOD Raw Waste Load (kg BOD/kkg of product)
COD Raw Waste Load (kg COD/kkg of product)
TOC Raw Waste Load (kg TOC/kkg of product)
Although the sampling data indicate that in some cases there is consider-
able variation between two manufacturers who nominally operate the same
process and between different time sampling periods for the same manu-
facturers, it was necessary to specify one set of values for each product/
process grouping. This was done in the most equitable manner possible
from the data available. Rather than arbitrarily choosing the lowest
observed value for each product/process grouping, an effort was made to
choose values which were considered consistent with BPCTCA in-process
technologi es.
The RWL data for each product/process grouping shown in Tables 5~1, 2,
3, and 4 have been inserted in the major process Subcategories A, B, C,
and D. For orientations, concentrations have been calculated for the
BOD parameters by dividing the BOD loading by the corresponding contact
process wastewater flow. Examination of these data indicates quite a
large spread in flows, loadings, and resulting concentrations.
It should be noted that the BOD concentrations shown are based on waste-
waters coming directly from the process and do not necessarily represent
the waste concentrations which a treatment plant would accept. If the
plant manufactured a single product which generated concentrated wastes,
these may be diluted with contaminated cooling wastes and steam conden-
sate or other non-contact waters prior to biological treatment depending
upon concentrations and economic considerations. In a multi-product
plant, the concentrated wastewater could be diluted with less concentrated
-253-
-------
DRAFT
O
o
1
0
o
0
L
Q
O
CD
J
u_
_J
^
Ol
E
Ol
.*
-*
\
Ol
_^
'"~~1-
_J
\
Ol
E
Ol
_y
^
^x
Ol
_*
(^^1
_J
\
01
^
\
Ol
_^
-*
\
CJ1
_^:
Ol
3
^
O '-s
O 00
ft LA
V ~
_
o
o r^
0 0
0
o
V°
-~~*-
O ^
o -d-
^yiD
r
o
O LA
O CN
O
O
v°
^_^
o ^
O v£)
ft CM
""^^
O
O
o
o
o
V °
J-
ft ro
CO
l/l
_J
CU
in
03
03
C£
Ul
m
CU
8
Q_
<
O
LJ
O_
CO
fD
O
1
CU
in
J
3
cc
^
o
I
Q
o
CJ
L
o
o
m
J
U.
^
Ol
Ol
Ol
^ «
_l
01
Ol
-^
Ol
Jl ^
v^_
Ol
^
^
Ol
ty
Ol
-^
Ol
3
^
o o
o o
K K
J- ^D
-»° Ot
* ~
O 0
o o
OO O~\
CO
LTV
LA
o o
ot o
CTv
CM ^i>
O^» LA
CM
O O
I~-CO
O - 0 ^~- O
0 O O 0 0
(v-j , , roj- .. «co -4"O
~ CMJ-CMCMro^ooro ror-
CO
CO CMCM .LACO ^ OO~1
0 OOJ- CN-- ° 0 0-
"D
0 O O -- O 4-i -
O O O O O 03 O
f^ 1 ' - LT\ « * - CN "-3" H> O r^
m LA\Q OMnOMA -*
c
-4" CN r-- r-. «rocu-4"j-
O «^O J-^O' CN03O1
- ^ ^ - , ,
0 v O 0
' - ( U~\ > ^) "CO "* - LTv - ~.
CO - VOLH »O>J~Ur\1 O
r-- co
CN -3" CO OO [""« LA J" fN CN
CNICNO -OPA -_J-
0 OOMDJ-CN PAO 0
.
O
o . . . o o
o o oooooo o
CNI --OOrAOLno »O
co LT\ r^ j- *o-d~ -d~
CN LT\ Jt LTV LACN
o
-Q t-
ro o
a) ra
C 4J
a) in
00 II)
!4- C >-
O O X
C 4-> -o
o ra 0)
NX
X
z
of Butadine
idation of Toluene With Air
c x
0 0
4J O
(U '
C V
>-
of Methyl Chloride & Methai
c
0
4->
c
of Benzene
c
0
4-t
C
of Toluene
c
o
4->
C
<1J
c
c
<
M-
o
c
.
ro
n of Ad i poni tri le
of 1,6- Hexaned i ol
o
' w
4-»
(U -
CO
1
u
0)
c
4->
0}
c
4-J C
O ro
OC_)OQX 4-1
~o ro
-C
0)
C
a) a)
c
4) >-
E M
=] I
(_> Q-
0> "D
L)
4J
U
O O
Q. tsl
C
>-
«J
ro
c
0)
a)
m
c
L_
O
jbenzen
)tol uen
L_ L_
O O
C C
. 0)
.c
>- 4-1
0) E
-c ro
Q- X
C
ID
-C
4~>
E
ro
y
O L_
>- 0
< 0
0) -C
>~
UJ
^
4-> C
0) <
o
1- 0
o
0 -C
c
<
0
.c
Q-
>-
o
~D
0)
0
ro
Q_
4-1
-254-
-------
DRAFT
^D
CXI
-d-
r"~\
0
0
CTl
[^
CN
O
0
, .
CO
f-\
CT\
J-
0
o
.
J-
CTl
CN
"
~O
._
<
u
M
0)
o
<
jz
4-J
.
3
,
o
jz
o
u
<
1
>-
JZ
4-J
UJ
4-
o
c
o
4-J
ro
t_
4-
1_
4-J
l/l
UJ
QJ
4-J
fD
1/1 4-J
U QJ
QJ U
4-> <
I/I
UJ
>.
U -C
4_J
0>
u
- QJ
CL
O >-
L. a
a. o
i_
4- 0-
o
c o
o
- c
4-J O
fD
O 4-J
. ro
4- -D
1- X
QJ O
-M E
ui E
LU <
OJ
4-J
ro
4-J
OJ QJ
u
< .-
^_
. 4_j
>, .-
CL C
O O
u
Q_ >-,
1_
U
<
O
o
-d~ * -
f r-.
O CN
ro J-
\£> .
LA CO
-3- CM
0
o
o
CO ^D
O O
vjD
O
CN CO
\D
- CN
- CO
_
0
O
CN .--
* LA
cn LA
LA CN
CO
CO
CO CO
CO
.
O O
O 0
o
LA J-
LA
OJ
C
0)
N
C
QJ -O
J3
O 0
U <
4-J
u
~Z-
1_
4- fD
O QJ
4-J
C CO
O
._ t_|_
4-J 0
O
3 rz
TD 0
QJ .-
C£ 4-J
ro
._ ._
4-J
>- fD
. i_
4J D
ro a)
o -z.
OJ
4-1
fD
. L.
O ro
C 0)
QJ 4-»
JZ CO
CL
O E
c n
.- ._
<
1 fD
ao
i i
s-~ . .
o o
o en
CN CN
ro ro
o cn
CN
ro.
en ^o
o
CN
*-£)
-
,
r^
-3- i
o
O 0
- cn
cn *
CM
1/1
1/1
aj 1/1
o ^
O QJ
L. U
a. o
E CL
ro
4-J QJ
U C
ro
L- i
a >.
fD X.
0 O
:£ -D
t/1 >-.
Q X
QJ
E
X
0
QJ
C
£ O
ro c
M nj
U X
fD QJ
_C
O O
a u
ro >-
O LJ
O
O
-3" '
o o
r^ j-
CN
O
o
o
CN - .
o j-
ro ro
CN LA
ro
^0 J-
O
0
o
CN CO
J-
r--
LA
r-- o
cn
CN
0
o
o o
en -
o _j-
* ro
CN
OJ
-D
O E
c: ro
QJ E
JZ 1-
D- O
LJ_
4-
0 4-
O
C
O (/i
4-J t/1
ro >
>- o
4-J ~O
QJ >
21 X
U
4-J
GJ
JZ
c -o
>, ._
co o
<
o
o
"i £
-
CO
QJ
C
'i
ft3
U
4-j
aj
\
OJ
c
QJ
>
jz:
4-j
aj
E
ro
X
QJ
X
CN
- - ("A
VD >JD
CN
CO CT\
o
* . f ~
0 0
0 Jj
CO
ro ro
. i
LA
< J-
Co
O CO
O vO
ro v£)
cn ^o
o
ro
cn
0 0
o o
OA_j-
0 0
ro CN
i/i
C
o
JD
1_
fD
-D I-
> T3
-C X
QJ X
~o
fD (-J
£
l_ M-
o o
LJ_
QJ
TJ L-
C 3
ro 4-J
X
i/i
m s:
QJ
U TO
o
L- E
0_ O
L-
U-
CT-
.- C
(J
t/1 4-J
fD U
a: ro
a) 4->
-C X
1 UJ
ai
c
0
-M
D
,
O
CO QJ
rz
. QJ QJ
C
>-
N 4-J
fD =>
-a 0
>. i/>
o .
CN
LA '
CN
ro
, .,
0
CO
r.
1
cn
cn
CN
CM
on
ro
LA
cn
cn
0
0
-3-
LA
n
CN
t
QJ
C
QJ
>-
CL
O
L_
CL
4-
0
1/1
tn
>-
0
TJ
>-
X
0
c
ro
. CL
0
o
I/)
o
o
d-
v£>
ro
CN
I
1
0
J-
vD
CO
X
OJ
CL
O
4-J
rz
QJ
>
0
CO
u
c
cn
l_
O
o
0
-j-
roOO
ro O
OO
o
o
o
s~~, VJD
O LA
~ ,
0
^o on
ro u~\
-3-
_
O
0
CN
CO
ro
O
cn
1 CO
o
0 0
O 0
- CN
r-- -
ro O
-d-
t/i
QJ
4-J
ro
i_
TD
>,
J3
O
JD
L.
ro
O
4-
o
c c
0 O
-M 4-J
fD (D
"O CO
- C
X
JZ
M -D
z <
i
0
-o s~
._ JZ
U ->-J
< >
i_
U QJ
.- ro
ro C
X 0)
o a.
r-^ .
- - LA
ro
vD
o r--
o
O CN
, ,
0
LA
' - CN
O -
i CM
LA
LA '
O ro
jj-
O
vO
- cn
ro^f
\D
LA
O
O ro
O
0 O
O LA
LA *
> ro
LA ^O
QJ
T3
X
0
QJ
C
OJ
I t/1
>-. l/l
CL QJ
O 0
L- O
Q_ !_
CL
4-
O C
- J=
. 0
O i-
-D
>--C
X CJ
O QJ
U "O
>
X
o o
0) - >
0 0
U !_
Q_ Q_
0 O
ro
J- o
-d-
OO ro
ro ro
' ^ ^ ^
O O
r---d-
CN CO
ro ro
CnvJD
r~- CN
oo
o
o ^~-
-3 CO
>co
cn LA
ro
ro
LA cn
ex
0
O 0
o o
-co
cn -
vO CN
CN CO
l/l I/I
4) QJ
> C
.- QJ
4-1 ,
ro X
> -M
. 3
L. CQ
QJ
a -o
QJ
QJ X
T3
51
1_
T3 4-
>- O
JZ
C
<
01
o >-
O
ro i-
JZ -0
4J >~
JZ X
Q_
-o
§c
fD
1_
4- C
O
. 4-J
(T> ro
0) C
_c 0
c
>- 3
CO CO
O
JZ
o
u
<
,
c >.
4-J
L- 3
ro CQ
JZ
u u
fD QJ
V) CO
-255-
-------
DRAFT
.0
ro
O
?
o
K
0
o
o
L
0
o
CQ
3
O
HZ
c
o
4-J
Q_
L.
O
Wl
Q
in
in
QJ
O
O
Q-
4-1
o
=
c
L.
Q_
^
\
E
^
-V.
01
, ^
1
Ol
E
01
_v£
*x.
01
,. .,
\,
Oi
E
\"
cn
y
Oi
-*
01
"X^
1
O O O O
s~* . -CT VO -~- CM CM
r [*-* o*\ o PA LA
O - - LA - -
LA PA PA PA
r-N r-v. o
CM LA CO -Jfr
CM CA r CT* CO PA 1
o o o o o o
CO O CM -^ *«O O
rA * CM PA PA *
- LA -3~
O
^O \O O^ G
r- o -
LA * P-- r^. j-
MD CM PA CSI CT* 1
o o o o
x-v Lf\ ^- LA
co r^ o s£> co j-
oo -j- * -
vD LA ^O CM cs|
1
. _ . ,~- Q
co i-*.
CM 00 O^CO CO rA
PA LA PA CSI 1
O O
O O O 0 -
O O O O O O
.~ . o o h-
r-*- * UA * J- PA
Is** o cnco -
^t * PA CM \O (T\ 1
^_
O
C
ro
_c
4_j
OJ
ro -c
L. 4J
3
"O
o o
c
c o o
O in .-
m 4-> ro ro >*
QJ ro u_ NO
in .
ro ro o ro
4J i- >- C/l
O m D o
o i_ ro -o O
M- D 2: C
o 4-* ro c
._ o M- o
c o ro O c *
O U- O -M
ro ^ - in "4-
c c * c
QJ O _C O Q) U
E 1- U 1- -DO)
QJ ro > o wi
LL. O CQ X CJ LU
QJ
"o ro Q)
.- .- m c
c ~o QJ O
ro QJ > c.
l. E 0 £
(1) I- 4-» O
rj (D ro QJ 4-»
4J > >. 4-j ro
-o c _c ro CQ
C U 4-»
ro QJ QJ >- ui
>- m *- o
OQT3T3T3 OJ
o .- .- c ro c
- >- c >. ru
L_ l_tn-M4->O-C O
4_i 4_i QJ 4-1 4_i C 4-1 l/l
. . >. Q ro o tu
(JOQLULU ^^
^.^ 0 0 O
o o o o o
PA ^O LA CO J" O * * *
* «* \i> r . co r^- o^
CM < LAr^.P'^LA' CM
o
O^ CM o~i j" ^o
-J" "CM PA PA O"\
^ O o CM -* PA r^ *>
CT\' PACMCSILArAPA
"o o"o
OO OOOOOO
-4" --csi O - - «
CMLA OOO^CSirxr^LA
* « O^\""-CS|O^-
_j-r^ PA CM u^ UA
o
CM O O
. , LA CN vD r*^ ^j-
j- co LA -3" o
oo o*1* o O"\ r** * CN * LA
CNCN ' r^- PACO- ~
^^^ ^ _^ ^ ^
^-^-^ o o o
o o o o o
^O LA - - -^"-- 0~\ PA PA
CM LA ^ x ^-^ CO LA * * "
" - PA LA CO CN LA t--
r*-~ O^ LA -3" O^ '
'LArTiLA -PACA
^t"JJ" VD »COPALA *
CO' CM O CA LA » CM
O O O
O O O O O O O
OO OOOO OO
* * &\ " oS co o * j"
r~~^ O^\ 'O O^ * LA LA O PA
MOpA CMPACMvO'^D1""1
O
4->
OJ
OJ
cn u
D i-
(/) U- O
3
o u cr
._ m u ._
4J . _Q l/l
2: c O o O XT ro
O 1- 1- C 4-J 4-J
U . OO O1^* C Ol QJ
Ol(Ol--n-C O QJ Q-
3UQJGO ^._>uo
(/I . C O O Q-i
4-. L.L-O 3 ro u-
w . M- 4-» 4J c o.c ro o
(UT3QJ QJ04JJ-
(U.-a:ar2-^: ^rsc
CO O 1 1 D- -Q Q_ 4-* O
< E o o. c ro
U- 13 4- (D 4- Z 4-»
OHDQJ^U-O O ro
C O O C M-
croo coco>-
O L. t/i i/i O O X
._ c 4-J 4-» CO
4-iO4>tn-ro N ro-- OJ
4J4J _ ^. C in 4-J C
C O E O O 4-» C U
curoOcci-OQjro
£l_l_OOONT3^-ro
i-4->'4- E E roc4-».y
QJX E E J: O X
U_ LjJ ^ «^ C_> O O LU
-256-
-------
DRAFT
wastes from other processes. The point to be kept in mind here is that
calculated concentrations for individual product/process groupings can
not be correlated with real situations in existing process plants unless
a complete case study, considering the total mix of processes which are
operated, is developed for each plant.
This same argument extends to hypothetical process effluents obtained by
applying a reduction factor to the process RWL. An effluent concentra-
tion calculated by dividing the process effluent loading by the process
contact wastewater flow is meaningless, because in almost all instances
the actual treatment plant which accomplishes the percent removal or
waste reduction will be accepting the combined wastes from multiple pro-
cesses whose overall concentrations are lower because of the inclusion
of products/processes of lesser loads and slightly contaminated wastes
such as steam condensate, pump seals, etc.
The large spread in RWL within the major process subcategories has led to
the establishment of additional subcategories. Tables 5~5, 6, 7 and 8
list the new subcategories which have been derived from the previous
major process categories. From the 5^ product/process groupings consid-
ered in Phase 2, the RWL data from kk were considered as adequate for the
ultimate determination of effluent limitation guidelines. In order to
make these guidelines as equitable as possible, 13 new subcategories were
considered necessary to subdivide the A4 product/process groupings. This
number of groupings was necessary so as not to penalize a product/process
which had reached the degree of inplant control commensurate with BPCTCA.
Examination of Tables 5~5, 6, 7, and 8 shows that the new subcategories
have been established based on similarities in waste loadings for BOD,
COD, and TOC. Effluent limitations were then calculated for each of the
new subcategories by applying reduction factors to the calculated mean
RWL.
-257-
-------
Table 5-5
Subcategory A BPCTCA - RWL
Subcateqory A-2
Cumene
p-Xy1ene
F low
L/kkg
0.334
44.3
BODr
kg/kfcg
< 0.0001
0.01
COD
kg/kkg
< 0.0001
0.025
TOC
kg/kkg
<-0.0001
0.007
Table 5-6
Subcategories B-3, B-4, B-5 BPCTCA - RWL
Sub category B-3
Ch1oromethanes
Ch1orotoluene
D i phenylami ne
Perch 1oroethylene
Phthalic Anhydride (o-Xylene)
Tricresyl Phosphate
MEAN ~
Subcategory B-4
Adi poni trile
Benzole Acid S- Benzaldehyde
HMDA (hexanediol)
MethyIch1 ori de
MEAN
Subcategory B-5
HMDA (Adiponitrile)
Ma lei c Anhydride
Methyl Ethyl Ketone
MEAN
F 1 ow
L/kkg
2,820.
121 ,000.
526.
5,400.
594.
28,000.
2,340
BOD
kg/kt
-------
Table 5-7
Subcategories C-3, C-4, C-5, C-6, C-7 BPCTCA - RWL
Subcategory C-3
Cyclohexane Oxime
Isopropanol
MEAN
F 1 ow
L/kkg.
1,910.
2,540.
2,230.
. BOD
kg/kkg
0.995
0.995
C_OD_
kg/kkg
6.29
2.99
__TOC_
kg/kkg
1.32
1.32
Subcategory C-4
Formic Acid
Oxalie Ac i d
MEAN
134,000.
437,000.
285,000.
1.05
1.31
4.36
4741"
1.4
LJ.L
1.36
Subcategory C-5_
Ca 1 c i urn Stea rate
Caprolactam (DSM)
Hydrazine Solutions
Isobuty1ene
Propylene Oxide
j> ? c . _B u t y 1 A 1 c oh oj
MEAN
54,100.
29,100.
30,300.
20,400.
63,500.
32,800.
38,400.
13.8
47.1
9.09
13.6
31.5
19.3
22.4
32.8
93.
115.
64.1
143.
126.
95.7
23.1
-
-
12.9
22.7
33.1
23.0
Subcategory C-6
At ryloni trile
Hf_xomethylene Tret ramine
MEAN
4,210.
M4o._
5,130.
60.
83J.
71.7
179.
229.
204.
78.
7.1..
74.5
Suucategory C-7
p-Ami nophenol
C r = sol, Synthet i c
Pentaerythri to)
_S^a_c£ha ri ri
MEAN
15,000.
2,090.
10,200.
26,900.
13,500.
888.
297.
390.
253.
^57.
1 ,620.
632.
1,590.
879.
1 ,180.
456.
217.
830.
384.
472.
-259-
-------
Table 5-8
Subcategories D-l, D-2, D-3, D-4 BPCTCA - RWL
Subcategory D-1
o-Ni troani1i ne
p-Ni troanl11ne
MEAN
Flow
L/kkg
269,000
39JOO
154,000
9.73
COD
kg/kkg
105.
79.1
92.
TOC
kg/kkg
30.9
22.2
26.5
Subcategory D-2
Citronellol and Geraniol
Fatty Acids
Fatty Acid Derivatives
Ionone and Methyl Ionone
MEAN
10,100.
28,000.
6,400.
9,370.
13,500.
58.1
18.1
18.
23.
29.3
111.
37.6
27.9
9^.
67.6
37.7
9.8
8.47
33.
22.2
Subcategory D-3
Plastic!zers
MEAN
53.9
53.9
82.6
82.6
Subcategory D-4
Citric Acid
Dyes and Intermediates
Naphthenic Acid
Pi gments
Sodi urn G1utamate
Tannic Acid
MEAN
477,000.
395,000.
39,700.
658,000.
67,000.
10,000.
274,000
328.
397.
141.
385.
84.7
153.
248.
657.
2,060.
298.
3,422.
284.
1,070.
1,300.
242.
775.
104.
513.
91.
173.
316.
-260-
-------
DRAFT
SECTION VI
SELECTION OF POLLUTANT PARAMETERS
Twenty-eight parameters were examined during the field data collection
program. These parameters are listed in Table 6-1, and all field sampling
data are summarized in Supplement B.
Based on the degree of impact on the overall environment, the pollutants
are divided into groups as follows:
1. Pollutants of Significance.
2. Pollutants of Limited Significance.
The rationale and justification for pollutant categorization 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 light of the current knowledge as to their
limitations from an analytical as well as from an environmental stand-
point.
Pollutants observed from the field data that were present in sufficient
concentrations so as to interfere with, be incompatible with, or pass
inadequately treated through publicly-owned works are discussed in
Section XI I.
Pollutants of Significance
Parameters of pollutional significance for which effluent limitations
were developed in the organic chemicals industry are the major organic
parameters of BOD and TSS.
The reason for expressing the effluent limitations in terms of only one
oxygen-demanding parameter is that almost all the biological treatment
plants observed during the study were designed on the basis of one con-
trolling design parameter. Survey data indicates that the parameter
most widely used was BOD. This is also the argument why multiple efflu-
ent limitations are ambiguous. Most of the biological plants surveyed
were not designed specifically to remove any parameter but BOD and TSS.
Therefore biological treatment plant effluent concentrations reported
in subsequent sections for other parameters do not reflect optimal re-
moval kinetics. The use of the BOD parameter is logical since almost
all stream assimulation capacities are assessed on the basis of ultimate
BOD. It has only been in recent years that effects of refractory organics
as measured by COD and TOC have been of concern in water quality analyses.
-261-
-------
DRAFT
Table 6-1
List of Pollutants and Indicators of Pollution
Examined for the Organic Chemicals Industry
Chemical Oxygen Demand
Biochemical Oxygen Demand
Total Organic Carbon
Total Suspended Solids
Oil (Freon extractables)
Ammonia Nitrogen
Total Kjeldahl Nitrogen
Phenol
Cyanide - Di sti1lat ion
Color
Sulfate
pH
Aci di ty
Alkalinity
Total Dissolved Solids
Chlori des
Hardness - Total
Total Phosphorus
Ca1ci urn
Maqnes i urn
Z i nc
Copper
I ron
Chromium - Total
Cadmi urn
Cobalt
Lead
Nickel
-262-
-------
DRAFT
Possible biological inhibition associated with the product/processes
surveyed will be discussed in the latter part of this Section.
BOD
Biochemical oxygen demand (BOD) refers to the amount of oxygen required
to stabilize biodegradable organic matter under aerobic conditions. The
BOD test has been used to gauge the pollutional strength of a wastewater
in terms of the oxygen it would demand if discharged into a water course.
Historically, the BOD test has also been used to evaluate the performance
of biological wastewater treatment facilities and to establish effluent
limitation values. However, objections to the use of the BOD test have
been raised.
The major objections are as follows:
1. The standard BOD 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 BOD test, seed culture (microorganisms) is
added to the BOD bottle. If the seed culture was not acclimated,
i.e., exposed to a similar wastewater in the past, then it may
not readily biologically degrade the waste. This results in
the reporting of a low BOD 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 review-
ing agencies to duplicate industrial BOD values unless great
care is taken in seed preparation.
3. The BOD test is sensitive to toxic materials, as are all biolog-
ical processes. Therefore, if toxic materials are present in a
particular wastewater, the reported BOD value may very well be
erroneous. This situation can be remedied by running a micro-
organism toxicity test, i.e., subsequently diluting the sample
until the BOD value reaches a plateau indicating that the mate-
rial is at a concentration which no longer inhibits biological
oxi dation.
There has been much controversy concerning the use of BOD as a measure of
pollution, and there have been recommendations to substitute some other
parameter, e.g., COD or TOC. EPA has recently pointed out that some or
all of the previously cited reasons make the BOD 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
-263-
-------
DRAFT
measures the amount of oxygen used by selected microorganisms in metabo-
lizing a wastewater. The use of COD or TOC to monitor the efficiency of
BOD removal in biological treatment is possible only if there is a good
correlation between COD or TOC and BOD. Under normal circumstances, two
correlations would be necessary, one for the raw wastewater and one for
the treated effluent. 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 and disadvantages, the BOD parameter
should continue to be used as the pollutional indicator for the organic
chemicals industry.
COD
Chemical oxygen demand (COD) provides a measure of the equivalent oxygen
required to oxidize the organic material present in a wastewater, under
acidic conditions, with the aid of a strong chemical oxidant, such as
potassium dichromate, and a catalyst (silver sulfate). One major ad-
vantage 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 chemi-
cals (sulfides, etc.) and chlorides may interfere with the COD test and
produce erroneous results. In the case of chlorides, this problem can
be eliminated with the addition of mercuric sulfate.
Standard Methods for the Examination of Water and Wastewater, the prin-
cipal reference for analytical work in this field, cautions that aromatic
compounds and straight-chain aliphatic compounds, both prevalent in the
organic chemicals industry, are not completely oxidized during the COD
test. The addition of silver sulfate, a catalyst, aids in the oxida-
tion of the straight-chain alcohols and acids, but does not affect aromatic
hydrocarbons. The exact extent of this partial oxidation has not been
documented in the literature.
TOC
Total organic carbon (TOC) is a measure of the amount of carbon in the
organic material in a wastewater sample. The TOC analyzer withdraws a
small volume of sample and thermally oxidizes it at 150°C. The water
vapor and carbon dioxide from the combustion chamber (where the water
vapor is removed) is condensed and sent to an infrared analyzer, where
the carbon dioxide is monitored. This carbon dioxide value corresponds
to the total inorganic value. Another portion of the same sample is ther-
mally oxidized at 950°C, which converts all the carbonaceous material;
this value corresponds to the total carbon value. TOC is determined by
subtracting the inorganic carbon (carbonates and water vapor) from the
total carbon value.
-26k-
-------
DRAFT
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 adsorption band of carbon dioxide and can there-
fore inflate the reported value.
2. The sample volume involved in the TOC analyzer is so small
(approximately 40 microliters) that it can easily become con-
taminated, with dust, for example.
3. Industrial wastes from the organic chemicals industry with low
vaporization points may vaporize before 150°C and therefore be
reported as inorganic carbon.
TSS
Total suspended solids (TSS) is a measure of the organic and inorganic
solids removed when filtered through a preformed glass fiber mat in a
Gooch crucible. As in the Phase 1 study, the TSS RWL's for the organic
chemicals industry are characteristically low. TSS RWL values for the
Phase 2 products surveyed are presented in Tables 6-2 through 6-5. The
following is a summary of specific products with significant TSS RWL's:
Subcategory Product RWL Concentration
mg/L
8 Maleic Anhydride 2,380
B Chloromethanes 1,170
B Methyl Chloride 1,380-7,100
C Hydrazine 1,180
C Acrylonitrile 630
C Propylene Oxide 4,520
D Fatty Acids and Derivatives 57-3,840
D Sodium Glutamate 2,260
D Para-nitro-analine 1,430
Since TSS RWL's are not generally significant throughout the industry,
it is not meaningful to set an effluent limitation in terms of lbs/1,000
Ibs product because the RWL TSS data are so low. The problem is that much
of the TSS eventually discharged to surface waters are biological solids
which have been produced in the end-of-pipe biological treatment facili-
ties and only a fraction of which are finally removed before discharge. To
minimize this problem, the effluent limitations will be based on a con-
centration value which will be attainable with adequate solids handling
facilities. This subject is discussed in detail in Section VII.
-265-
-------
DRAFT
Table 6-2
t see 11aneous RWL for Category A
Flow, gal/1,000 Ibs
Phenol
mg/1
kg/kkg
NH -N
3mg/l
kg/kkg
TKN
mg/1
kg/kkg
CN
mg/1
kg/kkg
Sulfate
mg/1
kg/kkg
Oil
mg/1
kg/kkg
T-P
mg/1
kg/kkg
Zn
mg/1
kg/kkg
Cu
mg/ 1
kg/kkg
Fe
mg/ ]
kg/kkg
Cr-Total
mg/ 1
kg/kkg
Cd
mg/ 1
kg/kkg
TSS
mg/1
kg/kkg
TDS
mg/1
kg/kkg
cr
mg/1
kg/kkg
Phase 1
p-Xylene
5.25
.16
.00001
1.0
.00005
3.12
.00013
2.6
.00011
7k6
.0327
4.76
.00021
.287
.00001
.1*1*
.00002
2.30
.00010
2.50
.00011
< .05
0
17.3
.00076
162
.0711
32
.0011*
1 1 Data /
*
Cumene
.01*
11*. 6
Negl igible
3.1*
Negl igible
8.2
Negl igible
.02
Neg 1 igible
12.3
Negl igible
21
Negl igible
.005
Negl igible
.13
Negl igible
.05
Negl igible
109
.00003
.01
Negl igible
.01
Negl igible
100
.00003
138
.00004
37
.00001
i
J
BTX
87.5
.155
.00009
1.27
.0010
1*.05
.0028
11*0
.0872
251
.709
.45
.00035
1.1*9
.00095
< .08
< .00006
1.1*5
.00090
5.16
.00315
< .05
< .00005
1*1*. 5
.0270
2,01*5
.31*3
168
.103
Supplementary Phase 1 RWL Data
}
BTX '
52.3
1.98
.00087
18.0
.00785
1*9.3
.0215
5,860
2.56
26
.011
1.112
.0001*9
.116
.00005
.09
.00001*
13.8
.006
.171
.00008
.127
.00006
331
.11*1*5
1*7,600
20.8
3,280
1.1*3
J
BTXV
5.56
.2
.00001
142.7
.00662
322.1*
.0151*
.818
.00001*
1,1*00
.061*9
9
.0001*
.033
.00000
.31*3
.00002
.05
.00000
1.57
.00007
.01
.00000
.01
.00000
9
.0001*
21
.00097
36
.00169
Ethyl Benzene
55.5
1 .91*
.00069
13.2
.00546
18.2
.00772
23.9
.0160
55
.0318
.079
.00004
2.46
.00087
.18
.00006
2.55
.00096
1.26
.00089
.052
.00003
61.5
.0254
20,300
6.68
4,1*10
1.48
Ethyl Benzene
39.7
1.03
.000335
11.8
.00387
105.5
.0355
12.8
.00578
9
.00298
.0178
.00006
30.5
.0102
21.5
.00721
7.9
.00149
.807
.00027
.051
.00002
5
.00172
13,200
4.44
50,800
16.9
-266-
-------
DRAFT
CO
O eg
O
O C^J
O O
CT\
LA
O
O
xO
m
O
O
CO
v£>
O
O
LA
LA
O
O
CO
r--.
co
co j-
MD J-
O
O
o
CM
O
LA
CO
LA
O
O
o
o
f^
CM
O
CM
O
O *- O
»- i cr» - -
I
Si 3
3 °.
LA
CM r- co
CM \D f"\ O i
, O MD CM
1 O O O O O O O
' J-OOOOO *
*-.§
» CM LA
co
O « CTi LA CM .
-d*
o o
- LA
r*"\ LA
-d- O O O r^
-d~ (v\ o o -d-
^- o o o o o
o o o o o o
or-. o-~oo-d---
T-O oooooo
oo oooooo
oo oooooo
J- CM LA
O CM
O O T- O
o o o
{V, r CN fv-i , fj~l CM CM
o csi r^ -J- o -d~
CM i i «i O"\ vD ^~
» «^O P-~-
OJ
c
^ 0)
QJ OJ E
tv
:ic Anhydrit
i i c Anhyd ri <
yj HJ
2: 2:
>on i t r i 1 e
TJ
)romethanes
LJ
O
>romethanes
>rotol ul ene
V. ' U
-C -C
0 CJ
4-J +-J
UJ i ' i
>< >
M -M
OJ OJ
2: 2;
QJ
C
'i
m
C
QJ
D-
O
-M
C
a
-D
0
2:
c
§.
-o
0
2;
3:
c
QJ
O
X
QJ
X
iy 1 Chi or t d«
w 1 Chi ori di
OJ OJ
2: 2:
i_
O
_c
LJ
OJ
2:
-C -C
-M C
(D <
O
1_ U
o
OJ -C
Q_ Q_
_ ^_ ^_
-------
DRAFT
J-
CM »-
j- oo
cr>
oo
O O O
T LA r^.
LA . CM
CM CM
LA J- O
-d- r- o o
J- CM'
o o
o o r--. LA
MD
CM O
CM i
T CM
O LA
O <""»
MD LA
O O O T-
-» O O O J-
LA O l^x. LA
O O CO LA
O O . .
CM f"> O\ CM
. CM \D r~~-
O O po T^
O CO «- O
-
PA v£> O CO
o r--. LA ^
O O O i
I =
O O LA
OO O CO . cTl
CM " CA LA CM
Q) .-
r 4_l
J3 C
13 O
- U3 J-
CO T- LA «-
CM O « O
000 O
O O O O
V
o o^ r^-
.cT o i
o o CM
o o o
-*
o
o
o
o
PA
O
O
O
0
LA LA
O O
O
0
O
0
o
o
o
o
LT\
CO
o
o
o
0
LA
O
O
O
0
LA LA
0 0
CM
O
O
O
PA
LA
O
O
O
O
PA CM -d"
O
O
O
O
CM
O
O
0
O
O
MO
o
o
o
CM
O
0
o
o
o
LA LA
O O
O
0
O
O
J-
LA
0
o
o
LA
CM
O
O
0
LA
C\J
LA
O
O
O
0
^
CM
O
O
O
LA
O
^
O
LA
O
_
O
O
o
o
o
CO
CM
O
o
LA
O
o
3
LA
O
0
s
O
>
cc
"i
£
>*
S
g
i
a
Q.
3,
LA
- MD * CM
O O PA O
O O O O
O O O O
o o o o
1^- O -J- CO
CM LA -^ CO LA i
O O M3 O J- O
O CO PA O
O CM O O
O O O O
O O O O
MD LA
VO LA LA f~
O LA O O
LA O~\ LA O
^o o o o
_
LA *-
O O
O O
O O
CM O
O O
O O
CM LA CT\ -3" r- PA
O CT\ LA CM CM O
O MD « O PA O
O O O O O O
-
O LA r- T-
o *- o o
o o o o
1
hydride vi<
hydride vi<
o o
a) d)
2: i:
CD
c
o
CL
-o
I/)
0)
c
ro
o
-C
o
l/l
c
(D
.c
oromet
x
c
CD
D
oroto
x
o
c
0
+J
0)
*>
X
x
<1)
(L)
C
O
4-J
0)
">*
x
UJ
x
2:
ami ne
c
CD
X
Q
poni tri le)
poni tri le)
-a TJ
:E 2:
n: x
E
(B
O
0)
c
a)
X
4-1
H)
E
(13
CD
n:
iloride
ilor i de
CJ <_)
*>">*
X X
4) CD
2: s:
-
i- X
^^
o u
O
>-C
X O
CD
. a) CO
-------
DRAFT
j- o en
.- o i^-
o o .j-
LA
vO r>"\ CM LA f~, .jj- [-- CM .3- r^
OO r*"i O O CO OA O m O CTt
OO OO O O CM O O O -
OO OO OOCsJOOO O
OO OO OOOOOO O
O f~> > T-
T CM n^ T
o o o o
o o o o
o o o o
1
CM i CO
LA
LA c
CM CO CM (v»
f~i O i CM O CO-J- -3"
OO OO O CO O O~\ CM O i-
OO OO O MO O i CM O CTl
OO OO O - CM O O O O
O CM (V\ *
O CM O O
o o o o
o o o o
ij *
E] J-
i (V-, O O t
LT\
o r--
O CM I--.
OOO OCM OOOOOO
OOO OO OOOOOO
OOO OO OOOOOO
o T cr\ LA \o i
CPi O ^ LA »- O
o o o o o o
o o o o o o
o o o o o o
MO -4-
m -4" CM
o o
CM CM r-~
CO O CM
-4- - -
CM i
-
CO CO r -3" UA
O>or--OO{-n
rv^OcMl--Oo
O LA O O O O
o o
- o co oo r-»
- CO r-~ iV\ CM
00
OOO
OOO
CO LA O r^i
O « O ^D LA wD uD
OO O r-~\ CO -4" MO O
o o CM o o r*.
\ o~\ ... . .
LA CO * * « LA LA
LA O CM O O \D
CM co o -d- o
O J" CO CO
1 CM O
LA J- i
(U . >,
CC CL,L.l_.C_c
OO (UOOO4--C
Q_Q_ w^ . , Oct
._ ._ >. jz j: j; o
T5"D -CCJOCJl-U
< < 4J O
E >->->. x: fo
C 0)
o
M >-
-------
DRAFT
-3- LA
-* O
O o
O O
r-
-------
. CO
o o
o o
o o
o o
O O
o o
-271-
-------
DRAFT
« fNI ,_
It- - CO
CTi i-
0) _l
4J *- C
-------
DRAFT
CTi
1
i VD r-
. -3" o o
CsJ J CO
0 O i- O
o o o o
CSJ LA Cnv£>
0 LA r- 0
-j" >
1_ 1_
01 0)
0 0
ub u6
VI Ul
"D T3
U U
< <
>- >
m TO
0
C
(U
_c
Q_
1/1 O
.- o
u
-------
DRAFT
Pollutants of Limited Significance
The following parameters, which were investigated in particular cases,
have significant effects on the applicability of end-of-pipe treatment
technologies.
Oil
Oil (Freon extractables) J.s_a measure of the insp 1.tibie hydjQCarbons and
the fr ee-jfJ^at_Lng_andL^rnuJs.i f i e
-------
DRAFT
N i trogen
Ammonia nitrogen (NhU-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 NI-U-N and organic nitrogen present
in the sample. Both NFU and TKN are expressed in terms of equivalent
nitrogen values in mg/L to facilitate mathematical manipulations of the
values.
Organic nitrogen may be converted in the environment to ammonia by sapro-
phytic bacteria under either aerobic or anaerobic conditions. The ammonia
nitrogen then becomes the nitrogen and energy source for autotrophic or-
ganisms (nitrofiers). The oxidation of ammonia to nitrite and then to
nitrate has a stoichiometric oxygen requirement of approximately 4.6
times the concentration of NHo-N. The nitrification reaction is much
slower than the carbonaceous reactions, and, therefore, the dissolved
oxygen utilization is observed over a much longer period.
The ammonia and TKN RWL data of major significance is summarized below
by product and subcategory:
RWL Concentration
Subcategory Product NH-^-N TKN-N
mg/L mg/L
B Acrylonitrile 13,600 22,000
Diphenylamine 15,500 16,700
Hexamethylene Diamine 7,630 9,170
C Hexamethylene Tetramine 7,040 8,260
D Fatty Acids and Derivatives 1-7,730 5-7,890
Ammonia concentrations of 3,000 mg/L have been reported inhibitory to
anaerobic sludge digestion.
Phenol /
Phenols in wastewater present two major problems: at high concentrations
phenol acts as a bactericide; and at very low concentrations, when dis-
infected with chlorine, chlorophenols are formed, producing taste and
odor. Past experience has indicated that biological treatment systems
may be acclimated to phenol concentrations of 300 mg/L or more. However,
protection of the biological treatment system against slug loads of phenol
should be given careful consideration in the design. Slug loadings as
low as 50 mg/L could be inhibitory to the biological population.
The following are phenol RWL's of significance:
Subcategory Product RWL Concentration
mg/L
C Synthetic Cresols 6,500
-275-
-------
DRAFT
Total Dissolved Solids, Chlorides, Sul fates
Total dis^lj^^^oJJ^^n or^anicchemical wa^^ej^^ers^gfi^j^_t_jiTainly of
carbon a tesj_,b^i-ca_r^on.ajtes ,._ c.h lor i des, § u 1 fa tejs, j^and _Eho^£hart>es . High
cnTorides and sulfates in Subcategories C and D are primarily from acid
scrubbing or the by-product formation of ammonium sulfate. Sul fate con-
centrations of 500 mg/L have been reported to be inhibitory to anaerobic
digestion, while NaCl concentrations of 10,000 mg/L are reported inhibi-
tory to biological treatment.
The following is a summary of the significant TDS, SO^, and Cl data from
Tables 6-2 through 6-5:
Subcategory
Product
Benzoic Acid
Maleic Anhydride
Adi pon i tri1e
Chloromethane
Methyl Chloride
Secondary Butyl Alcohol
Isobutylene
Pentaerythritol
Acrylon itrile
Hydrazine
Propylene Oxide
Naphthenic Acid
Glycerine
Citric Ac i d
lonone
Vani11 in
Dyes
Fatty Acid Derivatives
Tannic Acid
Citronellol-Geraniol
Plasticizers
RWL Con cent rat
TDS
mg/L
34,900
53,700
124,000
28,300
117,000
-
-
289,000
57,200
118,000
14,500
7,690
10,800
41,300
85,400
348,000
42,400
1,400
108,000
32,600
94,800
SOij
mg/L
36
-
156
349
5,800
1,280
1,205
2,570
2,700
-
-
3,230
4,200
2,070
3,980
5,330
548
1,640
-
40
2,030
ion
Cl
mg/L
290
18
83,400
-
-
-
-
868
858
98,700
5,050
69
17
15,500
191
-
12,100
71
-
282
160
Cyan i de
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+ + CIT
HCN
At a pH of 8 or less, the HCN is largely undissociated; as the pH increases,
the equilibrium shifts toward CN".
-276-
-------
DRAFT
Cyanide (as HCN) values of 1.0 mg/L have been reported inhibitory to
biological treatment.
The following is a summary of cyanide data which is significant in regard
to inhibition of biological treatment:
Subcategory Product RWL Concentration
mg/L
B Adiponitri le 187
C Acryloni tri le 270
Cyanide Titration Procedure
The cyanide results from the previous processes were determined using
Standard Method's Cyanide Titration Procedure with and without prelimi-
nary distillation. The following is a comparison of results from one
plant :
CN_
mg/L
Standard Method Titration Procedure ^,870
Standard Method Distillation - Titration Procedure
It has been reported that aldehydes combine with HCN to form cyanohydes
during distillation and that this complex interferes with the titration
procedure. Based on this analytical experience, it is recommended that
only the titration procedure be used with these wastewaters.
Heavy Metals
Heavy metals (such as zinc, copper and cadmium) are inhibitory to micro-
organisms because of their ability to tie up proteins in their key enzyme
systems. The following concentrations of heavy metals have been reported
as being inhibitory to biological treatment:
Pol 1 utant Inhibitory Concentration
mg/L
Copper 1 . 0
Zinc 5.0 - 10.0
Cadmium 0.021
Total Chromium 3.0
Iron 5.01
Inhibitory to anaerobic sludge digestion.
-277-
-------
DRAFT
The following is a summary of the metals data from Tables 6-2 through
Tables 6-5 which is considered significant in regard to inhibition of
biological treatment:
Subcategory
Product
Adi poni trile
Chloromethanes
Methyl Chloride
HMDA
Maleic Anhydride
Perchloroethylene
Propylene Oxide
Saccharin
Hydrazi ne
Plasticizers
Dyes
Van!11 in
Miscellaneous Dyes
Metal RWL Concentration
mg/L
Copper
I ron
Total Chromium
Copper
Cadmi urn
I ron
Zi nc
Z inc
Cadmi urn
Total Chromium
Cadmi urn
I ron
Z i nc
Copper
Cadmi urn
Copper
Copper
I ron
Total Chromium
48
7-7
26.3
0.21
18.8
33-8
74.5
0.4
9.45
0.1
23-9
26.9
27.6
0.23
97.9
61.6
17.2
12.6
-278-
-------
DRAFF
SECTION VI I
CONTROL AND TREATMENT TECHNOLOGIES
Control and treatment technologies which are available for the Organic
Chemicals Industry can encompass the entire spectrum of wastewater treat-
ment technology. The selection of a particular technology is dependent
on the technology economics and the magnitude of the final effluent con-
centration. Control and treatment technology may be divided into two
major groupings; namely,
1. In-plant pollution abatement.
2. End-of-pipe treatment.
After discussing the available performance data, conclusions will be made
relative to the reduction of various pollutants commensurate with the
following distinct technology levels:
1. Best Practicable Control Technology Currently Available (BPCTCA).
2. Best Available Technology Economically Achievable (BATEA).
3. Best Available Demonstrated Control Technology (BADCT).
To access the economic impact of these proposed effluent limitations on
the industry, model treatment systems have been proposed which are con-
sidered capable of attaining the recommended RWL reduction. It should
be noted and understood that the particular systems chosen for use in
the economic analysis are not the only systems which are capable of at-
taining 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. Various types of end-of-pipe wastewater treatment.
2. Various in-plant modifications and installation of pollution
control equipment.
3. Various combinations of end-of-pipe and in-plant technologies.
The complexity of the Organic Chemicals Industry dictated the use of only
one treatment model for each effluent level. This situation poses some
problems for specific chemicals whose wastewaters are, for example, bio-
logically inhibitory. The use of the biological treatment model for these
chemicals is done only to facilitate the economic analysis and not to be
inferred as a recommended technology.
It is the intent of this study to allow the individual manufacturer within
the Organic Chemicals Industry to make the ultimate choice of what specific
combination of pollution control measures is best-suited to his situation
in complying with the limitations and standards presented in this report.
-279-
-------
In-Plant Pollution Abatement
The complexity of the Organic Chemicals Industry precludes the pos
of providing a specific list of process modifications or control m
which are applicable to all of the Industry's processes.
The elimination or reduction of in-plant pollution depends upon any one
of the following factors:
1. New plant process selection to minimize pollution. Present
corporate environmental awareness requires that new products
and processes be evaluated with regard to their environmental
aspects.
2. The modification of process equipment to improve product re-
covery or to minimize pollution. These areas have been dis-
cussed thoroughly in a previous study.^
3o Maintenance and good housekeeping practices minimize pollution.
The competitive nature of the industry in conjunction with the
flammable nature of many of its products requires most producers
to operate their plants in the most efficient manner possible.
This necessitates good maintenance and housekeeping practices.
However, there are other segments of the industry who have
minimized maintenance expenditures and whose management does
not adequately fund their environmental control staff nor
support them in their quest to enforce rigid housekeeping
regulat ions.
4. The age of the plant and process equipment as it impacts on
pollution. Poorly maintained process equipment does not
warrant consideration under this age consideration. The real
problem is that older technology generally pollutes more than
new technology. In addition, older plant layouts do not allow
for economic modifications to the process equipment to minimize
pollution and, in many cases, prohibit segregation of storm and
process waters.
The age problem has its greatest impact on the batch chemicals
segment of the Organic Chemicals Industry. However, as of this
time, none of the approaches which have been suggested to handle
the age factor would be manageable as well as quantifiable. Since
Development Document for Proposed Effluent Limitations Guidelines and
New Source Performance Standards for the Major Organic Products. U.S.
Environmental Protection Agency; EPA 440/1-73/009; December 1973=
-280-
-------
DRAFT
many of the older plants are located in the inner cities and
discharge to municipally owned treatment plants, it is recom-
mended that age be a factor which ;nay be negotiated for plants
classified as point sources.
End-of-Pipe Treatment
As explained previously, the RWL data from the Phase 2 field survey is
being handled as a separate report. However, because of the scarcity
of treatment plant performance data, it was decided to combine the
Phase I and II data for this study. A summary of the types of treatment
technology which were observed during both phases are listed in Table 7-1.
During the Phase II study, 70 individual plants were surveyed;; however,
6 of the 70 plants were previously surveyed during the Phase I study.
Table 7-1 has been prepared taking this duplication into consideration.
Of the plants surveyed in Phase II which are subcategorized in A, B or
C, over 80 percent provide their own waste treatment facilities, while
60 percent of the Subcategory D plants discharge to municipally owned
treatment facilities.
Single-Stage Biological Treatment
During the plant survey program, historic wastewater treatment plant
performance data were obtained when possible. The data were statistically
analyzed, and, when possible, the individual plant performance was
evaluated with respect to the original design basis. Subsequent to this
evaluation, a group of plants were selected as being exemplary in per-
formance. These particular exemplary plants are indicated in Table 7-2,
which is a summary of all of the historic performance data made available
by industry for the purposes of the study. The amount of analytical
data used in the statistical analyses are indicated in the "data base
column" of Table 1-1. The following is a summary of the average re-
ductions capable of exemplary treatment plants:
COD BOD TOC Efflusnt
Remova1 Remova1 Remova1 TSS
percent percent percent mg/1
Exemplary Single- and
Multiple-Stage Plants Jk 93 79 13^
Exemplary Single-Stage Plants 69 92 60 65
The major differences observed in performance from the previous analyses
are in the TOC removals. This is because only two historic TOC data
points are available.
-281-
-------
DRAFT
Table 7-1
Organic Chemicals Study
Treatment Technology Survey
Type of Treatment or Disposal Facility
Activated Sludge
Activated Sludge-aerated lagoon
Activated Sludge-polishing pond
Activated Sludge-solar evaporation pond
Trickling Filter-activated sludge
Aerated lagoon-settling pond
Aerated lagoon-no solids separation
Facultative Anaerobic lagoon
Stripping Tower
No current treatment -
system in planning stage
To Municipal Treatment Plant
Deep-well disposal
Physical Treatment, e.g. API Separator
Activated Carbon
Incinerat ion
TOTAL
Number of Plants Observed
Phase 1
7
2
0
0
1
3
2
4
1
3
5
2
4
0
0
Phase II
9
0
1
1
0
1
1
4
1
7
23
6
3
6
1
64
-282-
-------
DRAFT
c
u
E
CO CT\Oi CM tA-J LPiVD P^CO
QJ
-283-
-------
DRAFT
During the survey program, 2^-hour composite samples were obtained in
order to verify the plants' historic performance data, as well as to
provide a more complete wastewater analytical profile. These results
are presented in Table 7-3. The following is a summary of the average
reductions capable of being attained by exemplary treatment as verified
by composite sampling:
COD BOD TOC
Removal Removal Removal
percent percent percent
Exemplary Treatment Plants 72 87 58
The TOC removal of 58 percent would seem to substantiate tha lo^/er value
of 60 percent as previously indicated for the historic values appearing
in Table 1-2. As indicated by the TSS removal data, 9 of the 17 plants
surveyed had negative TSS removal and over 75 percent of the plants had
inadequate solids handling facilities.
The impact of TDS and oil on the TSS levels for the plants surveyed is
indicated in Table 7-3. There is a trend indicating that high TDS and
oils in the plant effluent contribute to high TSS levels, e.g. note
Plants 16 and 18 /vhich treat fatty acid industry wastewaters. However
the direct effect of TDS on the TSS is not clear from the sampling data,
e.g. Plants 21 and 22 have high TDS and relatively low TSS, while
Plant 19 has a high TDS as well as TSS in its effluent. The major
problem is that biological sludge in many facilities is not wasted, there-
by increasing its TSS effluent levels. Unless one is thoroughly familiar
with a particular plant's operation, it is very difficult to interpret
TSS data, much less draw conclusions concerning variables affecting sludge
settleability. For this reason, recommendations concerning TSS for the
technology levels BATEA and BADCT will be based on the performance ex-
perience of adequately designed units functioning in other industries.
Multiple-Stage Biological Treatment
During the course of the plant surveys, three plants were observed to
have multiple-stage biological treatment. Plant 5 (see Table 7-2)
required two-stage treatment for phenol removal, while Plants 2 and 11
required it because of relatively high raw waste loads and rigid water
quality criteria. The Phase I recommendation that single-stage bio-
logical treatment be considered BPCTCA is further substantiated by the
Phase II survey data.
F i 1 trat ion
Supplemental organics and solids removal is being practiced within the
industry in one particular case using a polishing pond. One major
-28k-
-------
DRAFT
LTl .
QJ - CD
O 1-
C O -Q
£ .C -i-1 CM
i- O O
O 1i (D 4-J
QJ
0)
4-J (13
0) Q_
n]
CL in oo
£ Q_ nj ft)
O
12
o
i/> C C T3
m ra o
(D C
CO Q_ O O
-285-
-------
DRAFT
problem during summer periods is algal blooms which, if unchecked, can
drastically increase the TSS and COD of the polishing pond effluent.
In addition, the acreage requirements of this system limits its po-
tential uniform application.
In contrast, filtration has many of the advantages of polishing ponds
and few of the disadvantages. In order to quantify the effectiveness
of effluent filtration, samples of biological treatment plant effluents'
were collected and filtered using filter paper. The results are presented
in Table ~l-k. Average percent COD, BOO, and TOC removals associated with
filtration are 20, 17, and 20, respectively.
Carbon Adsorption
Granular activated carbon technology is continuously being developed
and is beiginning to compete actively with biological treatment as a
viable treatment alternative or as a biological treatment effluent
polishing process for some industrial wastes. As was indicated pre-
viously in the Phase 1 study, there exists a limited amenability of
many low molecular weight, oxygenated chemicals to adsorption on acti-
vated carbon. In addition, experience has indicated that TSS in amounts
exceeding 50 mg/1 and oils above concentrations of 10 mg/1 should not
be applied directly to carbon beds. These materials tend to clog and
coat the carbon particles, thereby reducing the adsorption effective-
ness. This is an obvious consideration in the fatty acid industry.
Plants 16, 17, and 18 treat wastewaters from this industry and are
characterized by high effluent solids and oils. (See Table 7-3.)
To place carbon adsorption technology in perspective, it must be under-
stood that one of the first full-scale industrial wastewater treatment
plants was installed in Pennsylvania in 1969. During the plant survey
program, 6 activated carbon plants treating raw wastewaters were sur-
veyed, and the results are presented in Table 7-5. The most interesting
fact is that domestic wastewater treatment experience indicates that
efficient treatment is provided with contact times between 10 and 50
minutes, while the design contact times in Table 7-5 vary between 22
and 660 minutes (calculated on an empty column basis). These higher
contact times are required because of the much higher raw waste loads
generated by industry.
The major problems encountered in trying to compare design criteria and
present performance of carbon plants are as follows:
Donald Hager, "A Survey of Industrial Wastewater Treatment by Granular
Activated Carbon" Presented at the 4th Joint Chemical Engineering Con-
ference; Vancouver, British Columbia; Sept. 10, 1973.
-286-
-------
Table 7-4
Removal by Filtration
(Performed on Biological Treatment Plant Effluent)
DRAFT
Plant
% COD
% BOD
% TOC
3
15
15
14
9
9
13
it
24
12
21
16
25
20
35
26
27
18
17
19
Average '
9
87
85
24
11
10
32
--
8
21
3
84.3
39.3
8.5
51.4
26.2
--
86.8
88.4
33.3
20
4
56
28
--
3 ,
78
82
14
5
17
36
57.8
17.2
71.4
12.5
72.1
55.6
17
20
7
8
75.9
39.4
33.0
27.7
41 .2
25.0
90.6
91.6
66.0
20
Average does not include plants 15, 16, 17, 18, and 26, since these plants
have excessively high effluent TSS and would bias the results.
-287-
-------
DRAFT
-------
DRAFT
1. In most cases, design loadings, both organic and hydraulic,
have not yet been attained. This means the new plants are
sometimes grossly under-loaded.
2. Thermal carbon regeneration is presently an art which is
acquired only with actual operating experience. For this
reason, start-up problems are often extended, and it is not
unusual for the pollutant concentrations of the activated
carbon effluent to be higher than the design value. This
situation continues until the carbon is regenerated thoroughly.
3. Plants with insufficient spill protection and/or inadequate
housekeeping practices may discharge specific low molecular
weight hydrocarbons which are not amenable to adsorption.
This situation results in an erratic plant performance.
The carbon adsorption isotherm is widely used to screen the applicability
of different activated carbons and to calculate theoretical exhaustion
rates. The comparison of isotherm and design exhaustion rates for
Plant 29 in Table 7-5 further substantiates the fact that isotherm data
is preliminary and should not be used for design purposes. However,
carbon isotherm data does indicate relative amenability of the particular
wastewater to treatment and to fairly typical removal efficiencies.
To investigate the possibility of using activated carbon technology on the
effluents from biological treatment plants treating organic chemical waste-
waters, a series of carbon isotherms were run at standard conditions using
a contact time of 30 minutes. The results of the isotherms are presented
in Tables J-6 through 7-8. Average performance values are presented as
follows:
Soluble
Pollutant
Parameter Carbon Exhaustion Rate Remova1
Ibs removed/lb carbon percent
COD 0.41 69
BOD 0.03 20
TOC 0.06 87
Inspection of the specific data in Tables 7-6 through 7-8 indicates that
carbon adsorption has varying degrees of amenability with regard to cost-
effective wastewater treatment. However, the data does indicate that
specific wastewaters are readily treatable using activated carbon.
BPCTCA Treatment Systems
The major purpose for the review of the historic treatment plant data
was to be able to quantify BPCTCA reduction factors, which would then
-289-
-------
DRAFT
Table 7-6
Summary COD Carbon Isotherm Data
(Performed on Biological Treatment Plant Effluent)
Carbon Exhaustion Rate
Plant No.
lit
15
15
3
9
9
13
13
k
2k
12
21
16
25
20
35
26
1o
23
17
Average^
Ibs COO Removed
Ib Carbon
0.035
0.8
0.2
1.35
0.30
0.36
0.1+2
0.36
0.51
0.34
4.5
0.11
.12
4.o
.1*5
.069
0.094
.41
Ibs Carbon
1 .000 qal Ions
232
8.9
28.6
1.87
13.9
13.3
10.6
12.6
2.2
32.2
0.27
21.4
29.5
.25
2.0
3.9
44.3
15.7
Max. Soluble
COD Removal (%)
22
87
87
87
74
84
79
75
70
57
69
87
3
50.2
57 8
41.6
42 4
72.8
83 4
63.6
93.9
69.0
Cateqo ry
B
D
C
B-C
B
B
B
C
D
C
B-C
A
D
B
g
D
The average does not include Plants No. 12, 14, 20 and 21.
-290-
-------
DRAF'i
c
o
ro
cn
O
O
CO
-d-
LA
LA
O
CM
1/1
D
CM
O
C
CD
ro 4-i
M a
ro fo
O
Q-
0)
ro
4-1 E
O -M
ui to
0)
L_
C I
o
-Q
s_ ro
ro O
O
en
Q o
o
co o
>- CQ
L_
ro C
E O
3 -o
CO 0)
L.
o
14-
L.
1U
Q_
ro
Q >
O O
CO E
(U
QC
O T
LA CT\
r~^ o
. o
o
LA
- A
LA
00
oo
oo
O)
00
4-1 (1)
0) _Q
Ol
E
CN
oo
,- cr\
CM
LA
T- O
CXI
LA
CM
T3
C
ro
r--
4-) CD
c
CD -D
3 U
^ O
C in
E ,-
CM -J-.
. CM
CO
CM
c
(D
o
t/>
0)
O
c
4->
c
ro
LA
CM
O
CM
LA
OO
CM , ^
CD
CD
0)
en
ro
-291-
-------
DRAFT
LA CM
CM i
LA
00
LA
CM
CM
CO
I"-.
0)
.0
ro
r~~
ro
ro
Q
E
O
E
0)
^
o
o
*~
1/1
JD
c
O VD LA
-Q ro CO
s- 1 » I ro LA O 1 J-
ro I O I i ro O I O
O I I I
r-
-C
i
Remova 1
o-4"-3-CMroOvO OvD
°rooor--coo« OVD
roOOroCOCTiCO LACO
0)
J2
3
"o
HcOLACM_j-cMCT\ roo
^ LA » T CN
E
CO
o
IX
00
-o
C
ro
LA
CM
en
E
co
ro
J-
oo
CM
J-
ro
O
CM
co
c
m
OJ
-o
3
O
C
0)
en
LA
CM
O
CM
LA
ro
vO CO
CM <
ro
CM
r--r-~.cn
CM r- >
-292-
-------
be applied to BPCTCA raw waste load figures for each subcategory in
order to generate recommended effluent limitation guidelines. Based
on the previous discussions of single-stage biological treatment, it
is recommended that the following pollutant reduction factors be con-
sistent with BPCTCA treatment technology:
Percent Reduction Factors Monthly Minimum Average
Applied to Average BPCTCA Effluent Concentration
Parameter RWL mg/L
BOD 92 20
COD 69
TSS 65 mg/L 20
Controlling Parameter
The BPCTCA effluent discharge recommendations will be made only for BOD.
TSS is expressed as a concentration limitation because BPCTCA raw waste
loads are minimal as indicated in Chapter VI. The major source of TSS in
biological treatment plant effluents are biological solids which, in many
cases, are intentionally not wasted for further sludge dewatering but
rather are permitted to pass out in the plant effluent. This situation
is further compounded in certain plants which have very high TDS, oil,
and grease concentrations which tend to hinder settling and thereby con-
tribute to the high effluent TSS.
The major justification for minimum effluent concentration is that a
number of the BPCTCA - BOD - RWL data are in the vicinity of 100 mg/L.
If BPCTCA reduction factors are applied without due consideration, the
resulting effluent concentrations will be below what is achievable with
BPCTCA technology. The recommended minimum effluent concentrations were
selected based on EPA's preliminary definition of BPCTCA municipal
secondary treatment. The minimum TSS concentration is specified for
plants attaining the minimum BOD concentration. This insures that
adequate solids handling facilities will be provided. The use of these
minimum concentration figures is illustrated as follows:
RPCTCA Fffliient I imitation Minimum BOD-BPCTCA
Category kg BOD/kkg mg/L concent ration
mg/L
C-5 1.79 12 20
-293-
-------
If a particular product is in C-5, the effluent BOD-BPCTCA effluent
limitation is 1.79 kg BOD/kkg product. When using the actual plant
flow RWL, the concentration calculation resulted in a BOD of 12 mg/L.
In this case, the minimum BOD concentration of 20 mg/L would apply
and the BPCTCA effluent limitation would be recomputed using 20 mg/L,
thereby resulting in a BOD effluent limitation of 2.98 kg/kkg.
To evaluate the economic effects of the BPCTCA effluent limitations
on the organic chemicals industry, it was necessary to formulate a
BPCTCA treatment model. The model selected was single stage activated
sludge. (See Figure 7-1.) The BPCTCA design basis and the unit treat-
ment process rationale are discussed at length in the previously de-
scribed Phase I report.
BATEA Treatment Systems
Based on the previous performance data from multiple-stage biological
treatment plants, existing carbon treatment plants and various carbon
isotherms, it has been possible to formulate waste reduction factors
commensurate as BATEA treatment technology:
Percent Reduction Factors
Applied to BPCTCA Minimum Monthly Average
parameter Effluent Limitation Effluent Concentration
BOD 90 10
COD 69 50
TSS 15 mg/L 10
The BATEA effluent discharge limitations will have two controlling para-
meters, i.e., BOD and COD. The major emphasis, however, should be on
COD removal since the major portion of the carbonaceous oxygen demanding
materials should have been removed with BPCTCA technology.
The BATEA treatment model used for economic evaluation of the proposed
limitations includes the BPCTCA treatment model followed by dual media
filtration and carbon adsorption. A typical flow diagram is shown in
Figure 7-2. The BATEA design basis and the unit sizing criteria are
discussed in the Phase I study. The carbon regeneration facilities
were sized using 0.^1 Ib COD removed/lb carbon, which is the average re-
sult as determined from the carbon isotherm data.
J3ADCT Treatment Systems
Based on the previous filtration data, it has been possible to form-
ulate waste reduction factors commensurate as BADCT treatment technology:
-23k-
-------
DRAFT
I
I--
0)
-o
o
fO
1- I-
O) (1)
o
I-
o
-295-
-------
3
£
DRAFT
I =
CM
I
-------
Percent Reduction Factors Minimum Monthly
Applied to BPCTCA Average Effluent
Parameter Effluent Limitation Concentration
mg/L
BOD 17 10
COD 20
TSS 10 mg/L 10
The BADCT treatment model used for economic evaluation of the proposed
limitations includes the BPCTCA treatment model followed by dual media
filtration. The design basis and unit sizing criteria are discussed
thoroughly in the Phase I study.
-297-
-------
DRAFT
SECTION VI I I
COST, ENERGY AND NON-WATER QUALITY ASPECTS
Cost
This section provides quantitative cost information relative to assessing
the economic impact of the proposed effluent limitations on the organic
chemicals industry. A separate economic analysis of treatment_cost im-
pact wi_lL_he. ^rejaa£ed__by_Arihur J). Little,^ \^c]^p^^'pfoduct^byj^roducr"
bajjl. In order to evaluate the economic impact on a uniform treatment
basis, end-of-pipe treatment models were proposed which will provide the
desired level of treatment as follows:
End-of-Pi pe
Technology Level Treatment Model
BPCTCA Single-Stage Activated Sludge.
BADCT Activated Sludge and Filtration.
BATEA Activated Sludge, Filtration,
and Carbon Adsorption.
The method used to attain the effluent limitations whether through in-plant
controls or by end-of-pipe treatment is left up to the individual manu-
facturer as to which is the most attractive economically.
Extensive annual and capital cost estimates were prepared for numerous
end-of-pipe treatment models, which were presented in the Phase 1 study.
As an expedient, these cost estimates were linearly extrapolated to in-
clude the similiar Phase 2 treatment models.
Of the products which were surveyed during Phase 2, Arthur D. Little,
Inc. has selected 35 products for detailed economic evaluation and has
provided typical average production figures. The following is a list
by subcategory of the products to be evaluated:
-299-
-------
DRAFT
Subcategory
A-1
A-2
B-1
B-2
B-3
B-3
B-3
B-3
C-3
C-4
C-4
C-5
C-5
C-5
D-3
D-3
D-3
Product
BTX
Para-Xylene
Acetone
Vinyl Chloride
Styrene
Chlorotoluene
Phthalic Anhydride
Chloromethane
Isopropanol
Oxalie Acid
Formic Acid
Propylene Oxide
Sec. Butyl Alcohol
Hydrazine
Fatty Acid
Citronellol-Geraniol
lonone
Dye
Subcateqory Product
A-2 Cumene
B-4
B-5
B-5
C-5
C-5
C-5
C-6
C-6
C-7
D-k
Q-k
Q-k
Methyl Chloride
Ad ipon i trile
HMD A
Benzoic Acid
MEK
Maleic Anhydride
Calcium Stearate
Caprolactam
Isobutylene
Hexamethylene Tetramine
Aery Ion i tr ile
Pentaerythritol
Sodium Glutamate
Tannic Acid
C i tric-Oxalie Aci d
(fermentation)
Naphthenic Acid
With this information, end-of-pipe treatment models were selected, and ap-
propriate capital and annual costs were provided for each technology level.
These costs are presented in Tables 8-1 through 8-31 for each of the prod-
uct or product groupings listed above. The costs presented on these tables
are incremental costs for achieving each technology level. For example
in Table 8-10, the total capital cost for single-stage activated sludge
treating wastewater from a hexamethylene diamine plant with an average pro-
duction of 5^8,000 Ibs/day is $1,500,000. Assuming an average COD removal
of 69 percent, the expected effluent COD concentration from BPCTCA is
3,280 mg/1 which is lower than both BATEA and BADCT recommended effluent
COD limitations. This situation is caused by the wide variability in the
BPCTCA-RWL data in Subcategory B-4 and the fact that the effluent limitations
are based on average values. In this particular case, BOD would be the
critical control parameter for each technology level.
The incremental capital costs for this plant to provide filtration to
achieve the recommended BADCT - BOD effluent limitation would be $79,000.
In contrast, the incremental capital costs for this plant Lo attain the
BATEA effluent limitation is $5^0,000. For this particular case, multiple-
stage biological treatment was provided as BATEA since BOD was the con-
trolling parameter. Table 8-7 indicates the use of the minimum concen-
tration values reported in Section VII. The average BPCTCA effluent
-300-
-------
DRAFT
limitation for Subcategory B-3 is reported as 0.0467 kg BOD/kkg of pro-
duction. For the chloromethanes plant examined in Table 8-7, this RWL
would reflect a concentration of 6.5 Tig/1. Since the minimum BPCTCA
BOD concentration has been set at 20 mg/1, this value would govern in
this particular case. That is why the BPCTCA - BOD effluent limitation
for the chloromethanes plant described in Table 8-7 was recomputed to
0.056 kg BOD/kkg based on 20 mg/1.
Capital cost estimates were provided for 31 of the 35 products previously
listed. Based on the available information, the treatment costs associated
with the 4 remaining products are minimal as indicated aelow:
Product
Cumene
Chlorotoluene
Oxalie Aci d
Ionone
Production
RWL Flow
Ibs/day ga 1/103 Ibs
822
30
27.4
0.6
Wastewater
F low
gpd
COD RWL
1CP Ibs
Category
0.04
14,500
52,500
1,130
33
435,000
1 ,440,000
678
0.001
1 .82
4.36
94
300
15
10
9,970
A-2
B-3
C-4
D-3
The minimal COD-RWL concentrations for chlorotoluene and oxalic acid pre-
clude the necessity for further treatment. The minimal flows in terms of
gpd associated with cumene and ionone production will have minimal impact
on the associated treatment costs. In actuality, these flows would be
treated in conjunction with other wastewaters eminating from other pro-
duction areas or hauled away and disposed of by a licensed hauler.
The preceding cost estimates were prepared for idealized plant sites with
no consideration for treatment site conditions, e.g., high ground water
table, poor soil conditions, etc. In addition, there is the problem of
segregation of process wastewater from non-contact process and stormwater
flows. In particular plants, segregation of wastewater could increase
the previously reported BPCTCA capital cost estimates from 10 to 50 per-
cent depending upon the physical plant layout and localized topography.
Energy
The BPCTCA treatment models were designed assuming sludge dewatering
using vacuum filtration with sludge cake disposal to a sanitary land-
fill. An alternative sludge disposal method is incineration, which
oxidizes the sludge organics and evaporates the water in the sludge.
The remaining inorganic ash is 90 percent of' the original volume. The
offsetting factor is the substantial amount of energy required to
realize this volume reduction.
Since the previous cost tables are computed using August 1971 dollars,
the recent energy crisis has not impacted on these figures. The future
availability and pricing of energy will play an ever increasing role
in the selection of wastewater treatment processes and sludge handling
alternat ives.
-301-
-------
DRAFT
Non-Water Quality Aspects
The major non-water 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, and deep-well in-
jection) 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 concen-
trated 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 non-water quality aspects, such as noise levels, will not be
perceptibly affected. Most chemical plants generate fairly high noise
levels [(85-95 dB(a)] within the battery limits because of equipment such
as pumps, compressors, steam jets, flare stacks, etc. Equipment as-
sociated with in-process or end-of-pipe control systems would not add
significantly to these levels.
-302-
-------
Ul
c
O
^~
1
oo
4)
_o
ro
I
u
0
14-
Ul
+J
Ul
O
o
4->
C
Q)
X 4->
H ro
CQ d)
1_
1_
(U
ro
a)
4->
Ul
ro
3
«
E
(^
:
4-*
C
0)
3
M-
H-
LU
LU
1-
<
T>
C
ro
1
o
a
^
CQ
O
t-
CJ
Q-
CQ
U)
4->
Ul
0
O
,_
r^
«- ^
<
4->
>~ Ul
1_ 3
O cn
cn 3
a> <
ro i
o
o
oo
LA
QC.
z
LU
(U
0)
^
Ol
o
"o
c
JI
u
K
<£
LU
1-
^
CQ
h-
0
O
CO
^
1-
o
Q-
CQ
_J
' ^
C£
i
1 1 1 1 1 PO
1 1 1 1 1 O
o
CT\
1 1 1 1 1 LA
1 1 1 1 1 0
O
LA
1 1 1 O CM 1
O
O
o r^«
1 O LA J- O OO
i « r^ . LA
o~i o r--. LA
o
o
o
i ~
o o
LA a^
^r
o
o
-d- o
en
CM
oo
o
o
o
1 0
o
en
0
r^*- i
CM 1
f
cn
o o o o
o o o o
o o o o
V fl f. *k
O CM -j- v£)
LA OA LA CO
i «~ ro
O O O O
o o o o
-d~ ' * ^D
K r. «s
00 co T-
O O O O
0000
O CO CM O
CM LA CM O
Cn LA LA
1 1 1 1
1 1 1 1
DRAFT
ro
T3
l/l
J3
"\
O
X
1
-3"
o
K
r
1
C
o
4_l
O
3
TJ
O
U
a.
Q)
Ol
ro
i_
0)
ro
._
E
_l
4-1
C
OJ
*
<4-
M-
LU
Q
O
CQ
1
COD/1000 Ibs product
1
^ in \
Ol -Q Ol
E E
i
c
o
.
4-1
03
«
E
'_!
4_j
C
0)
3
^~
M-
14-
LU
0
O
(_)
Ul
4-1
Ul
O
0
r
CO
_
Q.
ro
*
ro
4-J
0
1-
Ul
4_i
in
O
o
^
ro
3
C
C
^
o
CM
vs/
0*5
OO
c
o
»
4-1
03
N
4->
L.
0
E
<
^
01
1_
a)
c
LU
Ul
4-1
Ul
O
^
__
ro
3
C
C
<^
P_
ro
4-1
O
H-
01
c
'o
c
o
o
01
o
LA
14-
0
C
o
4->
03
L-
4_J
C
0)
o
c
o
0
<
LU
(
CQ
E
3
E
c
2:
-303-
-------
1/1
c
o
CM
|
oo
O
o
4-1
C
c
jjj
OJ
OJ
I
H-
l_
0)
fO
0)
4^
*E
_1
4-J
C
0)
t[
LU
^
UJ
h-
OQ
^
C
ro
o
o
CM
«3^
^^
L-
o
01
0)
4-)
ro
0
in
4-1
in
O
o
r~.
>
4-"
in
01
3
LA O
CM
O
0
\O O
CA *
T- O
CM
O
O
O
1 "
1 O
en
CM
LA 1
\O 1
LA
O O O O
O O O O
O O"1 OA CM
M *
CM CA
O O O O
O O O O
0 0 «- «-
s *i r,
CM « OA
O O O O
o o o o
O MD -3" O
s * *
O CTs O
OA CM vD
1 1 1 1
1 1 1 1
o
D-
co
DRAFT
ro
T3
in
_Q
A
O
X
oo
LA
1
o
4->
O
3
^8
u
o_
0)
01
1_
a)
<
u u
4J -^ "^3
o o o
| cl fe.
1- l/l t/1
Q^ _Q O
in
_a O O
O 0
O O
Or- T-
O *N. ~^
O Q Q
t- O 0
^^ CO O
T> m v. in \
D- 03 -D O1 -Q O1
01 01 E E
1 1
c c
O 0
4-J 4-1
ro nj
4-1 4->
Z
O
U-
u
0)
4-1
(D
0)
tn
ID
.2
E
_J
4->
C
(1)
^^
<4-
14-
UJ
i*^
O
CO
E
'_!
4->
C
0>
^_
14-
M-
LU
Q
O
o
in
4-1
in
O
0
«
ID
4-1
*O_
ro
O
i
ro
4->
0
1-
in
in
O
(__j
ro
3
C
C
4
in
O
CM
0
4-1
ro
N
t
4-"
O
E
+ Maintenance
01
c
4_J
ID
0)
Q.
O
in
i- in
0) O
5 o
O
0-
ro
+ c
C
<^
^.
01
1- ID
C O
LU H-
cn
c 01
c
"o ^
1- 0
4-1 1-
o c
0 O
O
^v
OT--
E 01
O
CM O
LA
O >*-
O
C
O C
o
ID 4-1
1- ID
4-> l-
C 4J
0) C
U 0)
C 0
O C
0 O
0
«c£
CJ
-------
CO
Q
CO
«^
0
o
Q_
CD
3
cc.
o
o
CA O
1 1 1 1 1 O O O
1 1 1 1 1 .LA LA
O
0 VJ3 0
1 OD LA CM CO « co i
i * r^-. t**^ LA i
CM t o « « r--
LA
O O O O
o o o o
o VD » r-»
vO vD 1 CTl
V\ -4"
>
O O o O
O O o O
O LA , v£)
^ * *
r^ CM cr\
o o o o
0000
O P"*-. f*\ O
f> * . «v
t- LA T- OO
vD -J- O
1 1 1 I
1 1 1 I
DRAFT
1- T3
to <
ro CD
_
a.
CO
o
oo
LA
-a
J3
CA
O
X
o
1
0
4->
u
D
g
l_
Q_
QJ
cn
ra
u
0)
<;
0 O
4-> -o -a
o o o
-o Q. a_
O
U 01 1/1
a JD _a
1/1
_0 0 0
^0 O
0 0
0 «- --
o \ \
O Q Q
i O O
^x m o
t/1 1
~D ' l/l ^^ {/I ^^
Q- ro -Q cn _Q cn
cn cn E E
I I
C C
o o
4-< 4-1
fO ro
S
O
* .
u_
u
a)
4-1
fD
3
(U
4-1
in
3
E
_J
4-1
C
0)
3
<4-
<+-
LU
Q
0
CD
E
!_i
4J
C
0)
14-
14-
LU
O
O
o
l/l
4-J
t/1
O
0
>
fD
._
Q.
fD
(_)
^~
fD
O
1-
t/i
m
o
o
^~
fD
3
C
C
«c
l/l
1_
>-
o
CM
C
o
4^
fD
N
-
4-t
O
E
+ Maintenance
cn
c
_
4-J
fD
0)
Q.
0
l^
4-1
I- l/l
a) o
0 °
Q- .
fD
+ C
C
c£
>^
cn
>- fD
0) 4-J
c 0
LU 1-
cn
c
O
O
o
cn
E
O
LA
C
O
ra
4-1
c
(U
u
c
O
o
E
c
-305-
-------
(U
TJ
« °
x:
o
-Q
(D
in
C
o
ro
4-1
o 'i
in c
O (U
o 3
c «*-
03 UJ
Q
1/1 <
Q_
00
CM
I
OQ
O
Dl
0)
in
O
o
cn
a
4>
_l
cn
O
"o
C
u
1 1 1 1 1 O
>
""oo
on
CM VO O
j- o
1 1 1 1 1 CO O
1 1 1 1 1 -
1 1 1 1 1 0
LA
OO
r~
l/V
^O
vO
O 1 1 O
o on CM
i o on i r-» i
i " t - i
1 i I CM I
vD
0
0
O
M
CM
CM
0
0
on
^
0
o
o
0
cn
|
|
|
0
o
o
t
UA
on
CM
0
o
CM
r"
0
0
0
M
on
v£>
T
1
|
1
O
o
o
^
o
*£)
CM
O
O
on
O
O
0
1
1
1
o
o
o
*
I""*-
*"
^
o
o
CO
*
J-
CM
0)
ID
U
(D
CL
o
u
4->
C
o
o
ro
o
o
0)
in
O
oo
LA
ID
TJ
on
O
X
o
on
"O
Q.
CD
4J
U
4-1 T3
o o
3 1-
-O Q-
O
u m
Q. J3
in
£i O
O
O
O «-
O ^
O Q
i- O
^ OQ
in
m
ID -O
CL
in
o
O
o
Q
o
o
(ft
.D
- .
0) ^ E
CT>
U
o
o
o
0) ID
cn 3
to D
o
ro
4-1
*i
c
a)
UJ
o
o
03
O
ID
4-1
'I
C
(U
LU
O
O
in
O
o
ro
(D
0)
in u
i- c
>- ID
O
(a, .-
ID
S^ ^
C
o
in
O
ID
C
c
ID
N
O
ID
(U
Q.
O
>- in
0) O
5 o
O
n ,_
(D
+ C
C
cn
i- ID
Q) 4->
C O
05
V.
i
C
O
J3
^
C
fD
u
cn
O
1/1
O
Q_
-306-
-------
un
co
a)
a)
c
0)
S_
>-
4-1
CO
in
C
o
ro
4-1
i_ .
O E
(ft C
O 0)
O 3
C M-
. Q
81 S
o
c
.c
u
H
BPCTCA
_j
3
a:
oo
CM
«- o
LA
1 1 1 1 1
1 1 1 1 1
1 1 1 1 1
t LA
LA
1 1 1 1 1
1 1 1 1 1
1 1 1 1 1
r-. oo
O 'o r- O
CM LA
1 1 1
1 1 1
CM ro
MD .-
O O i O^ LA OO
O CM MD <
1 O OO CM
1 O CM
co
CO
1 1 1 1 1
1 till
1 1 1 1 1
1 1 1 1 1
1 1 1 1 1
*
O OOOO
O OOOO
O OOOO
O .4- LA t O
O O CTv CM CM
O CM r- J-
CM
1 1 1 1 1
1 1 1 1 1
1 1 1 1 1
DRAFT
CTl
C
CTl
C
O
O
o
l-~
o
-o
o
L.
Q_
0)
CTl
ro
v
-a
a.
CTl
1 1
o
3
TJ
O
l_
a.
1/1
o
*
o
0
0
-v.
in
i
TO
CT)
-o
O
^
a.
m
-Q
o
0
o
f
^^
Q
0
CO
in
J3
o
3
-o
O
1-
Q-
O
O
o
a
o
LJ
in
CTl J3
CT)
in
u
0
CM
enance
o
in
C
o
TO
4-1
'i
c
0}
3
Q
O
CO
C
o
TO
'i
4-1
C
o
o
o
in
O
o
a.
TO
o
TO
O
C +
O
4-> CD
to c
N
i- in
0) O
O
o_
TO
+ C
C
in
O
o
TO
c
c
o
ro
a) a)
Q- c
o LU
CTl
TO
O
C
O
u
CT)
E
O
CM
C
o
4->
TO
C
0)
o
o
o
o
Q-
00
E
E
'c
c
o
o
CT)
E
O
LTv
C
o
c
0)
o
c
o
o
CO
E
E
C
-------
I/I
c
o
vO
CO
CD
f
-Q
03
I
QJ
-a
TD
-C
c
"-t
u
"
0!
-C
4-J
-C
CL.
L.
O
14-
l/l
l/l
0
o
4-J
C
4->
r,
0)
i_
h-
0)
'(T
'E
c
CD
J^
UJ
UJ
1
O
CD
0)
-,-J
(0
O
Costs
,_
(T\
1/1
3
cn
3
<£
o
CO
LA
0)
^*
oj
i
CD
O
"o
c
JZ
U
OJ
I-
"i
UJ
p~
CO
I
o
CD
^
O
1
Q_
CD
3
(A
<^ 00
CM OO
O CM
III- . 1
1 1 1 O O O vD 1 O
III J- OO 1 O
J- o
o
LA
CA
CM I--.
o r-.
o o o o o
III -j- O 1 O
III ("A 1 O
III "1
o
UA
*
-4" CM
o o~\
O CTv O O LA O
III 1^-. O PA O
1 1 1 ^O rA O
CD
CO
CO rsl
^ vO
O O vT> O O
i o r-» i co i i
1 J- CM O 1 1
1 - 1 1
CM
O O O O
O O O 0
o r\ oo
LA t '.O
i/y
O 0 0 0
C: CD 0 O
O ~^- r- CO
1 A r- \£>
^-f~>'
o o o o
o o o o
O LA LP, O
CTi LA in O i
J- rsj CO
1
I
i
1 1 ! 1
1 I ! 1
DRAFT
a:
CC
O
V-
n_
CD
Ibs/day
^
o
X
CM
CA
i
C
o
1
4-J
u
3
"O
o
L.
a.
OJ
01
03
U
OJ
'E
«
_i
4_>
C
OJ
i
14-
M-
LU
Q
O
CD
O
3
-o
O
!_
a.
(fi
-D
O
O
o
^
0
o
o
\ (ft
CD JD
E
t
C
o
_
4-J
03
4-J
'E
_j
4-J
C
0)
*
14-
14
UJ
Q
O
CJ
CD
E
1
<.
<_>
h-
<_J
Q_
CD
-o
^ (D
CD -M
E 0
0)
a.
X
UJ
c
o
-c
0)
in
OJ
CD
Q
O
(__5
4-f
t~_
0)
3
j
4-
4-
LU
in
-t1
IT;
O
O
*
a) in
4-J 4-J
in
a o
01 O
c_i
r
OJ
m 3
4J C
0 C
1- <
in
>*
O
CM
';"
>^
OO
C
o
4-J
CD
N
._
4-*
1_
o
E
'D
O
c
03
OJ
4-J
C
03
3*~
f
CD
C
.._
4-1
!D
CD
Q.
O
1-.
CD
^
O
a.
4
~;^
cn
!
CD
C
UJ
10
4-1
in
0
u
CD
.3
C
C
<
1
CD
4-1
O
1
CD
E
OJ
1_
OJ
D.
O
u
OJ
o
u
I
o
CD
-303-
-------
r-^
co
in
0)
c
(13
-C
O
in
C
o
u
O E
1/1 c
O 0)
o 3
M <
(D LLJ
0) H-
1- <
\- CD
!- -a
a) c
(1) O
4-J Q
l/l CTi
I -
CO
O
en
OJ
4->
flj
o
01
D
CTl
o
CO
LA
LU
H
^
CD
0)
01
o
>- o
en <
O CO
"o
C
JC
o
ni
I
CO
M3 i O OOOO
CM O O OOOO
OA T O O CO CM O
s A K >s M
O VD CM i O
LA ^D LA CNI
-OO- -ut-
OA
OA
OA 1 1 1 1 1 1
1 1 1 1 1 f
DRAFT
TO
T3
(/I
BOD
vO
LA
OA
o
o
3
T3
O
en
(TJ
0)
gpd
gals/1000 Ibs product
tation - Ibs 1000 Ibs product
I
c.
0)
i
0)
4-"
in
m
LU
Q
O
CD
bs product
000
mg/l
Ibs COD
tation
C
0)
o
o
mg
/l
ted BPCTCA
g
o
0)
Q.
X
LU
C
o
-o
D
1/1
(D
CO
Q
O
C
0)
in
O
o
Q.
en
fO C
N
Po
4-1
l/l
O
1_
o
E
en
L,
0) D
Q. C
O LU
c/1
O
c
c
(D
4-J
O
en
c:
O
L-
4->
C
O
o
en
E
O
CM
C
o
0)
o
c
o
u
_
D-
CQ
E
E
ro
c
c
-30S-
-------
°? o
CO
o
0) _
^ ^
= s
O
(D
4-1
V.
O E
in c
O tu
o 3
C 14-
1) LU
ID LLJ
0) t-
I- CO
>- -O
0) C
4-> it)
m * >
1^ Cf 0 i-
LA cn o LA
« CM
<^
O O O O
O 0 0 0
cn co « co
MM M
Cn r«~v rr,
w
o o o o
0000
o o o o
M «% <> «\
^O CM CM O
CO » rr\ CO
CA CM LA
1 1 1 1
1 1 1 1
o
CO
LA
CC
z
LU
DRAFT
ID
~~>
in
r>
O
X
CM
CM
CO
C
o
4-1
0
3
"8
i_
0-
0)
cn
(0
0)
>
<
3 3
tJ 0 "8
31- t.
-g a
u i/i i/i
a ja J3
m
J2 O O
O 0
O 0
O t- .-
0 ^ ^
O Q Q
,-0 0
^^ CO O
~o "*+* in ^^
Q_ (D -O cn ^ cn
cn cn E E
1 i
C C
0 O
4J 4-1
(D ID
1
2
Q
,_
u_
u
0)
4-1
(D
2
(1)
4-1
in
5
4-1
,__
E
'_!
4->
C
a)
p_
M-
<4-
LU
a
o
co
4-1
«
E
_i
4->
C
a)
.^
M-
l*-
LU
a
o
o
in
4-1
in
O
CJ
»
ID
4-1
^
a.
ID
o
ID
o
t-
i/>
4-1
in
O
O
*~
§
C
c
in
O
CM
c
o
4^
(D
N
U
0
E
+ Maintenance
cn
C.
4-i
(D
0)
Q.
O
Power
1 Costs
(D
3
+ C
C
*.
cn
»- ID
0) 4-J
C 0
LU 1-
-------
O
TO
4-1
l_ .
O E
l^ C
O 0)
Ul
4-J
O
cn
-4" *~
1
oo
a)
-Q
TO
!_
4J
._
C
o
a.
-a
<^
E
TO LU
0) H"
i_ ct|
t- 03
L. 13
D CI
*J TO
i i-
Q) O
4-J O
Ul
TO
o
Ul
15
cn
3
<^
i
o
00
LA
a:
2!
LU
vO
1 1 1 1 1
1 1 1 1 1
11111
oo
vO O
i CM
r
i i i i i
1 1 1 1 1 T
1 I 1 1 1
PA CM
« PA O O
PA CM Is
r- 0
III
III CM
1 1 1
CM
O CM O"^ O LA O
O l~- r v£) PA O
i o i cn i oo
| * *1 » *
1 CM < t PA
O
O
o
o
o
w
o
o
o
LA
CT\
1/V
o
o
o
o
o
LA
1
1
1
0 O
0 0
o o
PA ^
v£> LA
O O
o o
o -d-
o oo
CXI
O 0
o o
o o
0 0
CM OO
LA -3-
1 !
1 1
1 1
0 0
o o
o o
cn oo
\£> OO
o o
o o
PA r--
00
c-;
\n-
o o
o o
o o
*. *1
0 0
LA LA
o
*
1 i
1 1
1 1
DRAFT
(U
t->
a>
E
TO
V-
Q_
Oi
o
u
4-»
c
o
o
0)
-C
I/)
(0
o
o
TO
-O
A
O
*""
X
OO
LA
1
C
o
4-1
u
T3
0
L.
a_
D
cn
TO
1_
(U
TO
^
0)
Ul
TO
J2
Is/1000 Ibs product
s BOD/1000 Ibs product
TO -Q
cn
1
c
O
*
4-J
03
._
E
_l
4J
C
0)
F
4-
M-
LU
0
O
CO
4-J
o
o
o
1_
Q.
Ul
O
O
O
^
Q
O
o
\ Ul "^
cn -Q cn
E E
1
0
.
4-*
fO
*
E
'
jj
C
Q)
3
4-
4
LU
Q
O
o
in
4-1
Ul
0
O
TO Ul
4-J 4->
. Ul
Q- 0
TO CJ
O
*
TO
TO 3
4J C
0 C
1- <
l/l
0
CM
(S
o^
oo
c
o
4J
TO
N
'.U
L.
O
E
a nee
c
0>
4-J
c
TO
-4-
cn
C
TO
0)
Q.
O
U
3r
O
CL
^_
Ol
u
cu
c
LU
4-J
Ul
0
o
i
TO
^
C
C
<
,__
TO
4-1
O
h-
1/1
3
a)
_c
4-"
!/)
4-J
C
TO
C
0
4->
_Q
.C
,
TO
O
cn
O
Q
.
-Q
-------
l/l
c
o
L, .
O E
l/l
o
o
co
cu
c.
'i
ro
T3
CU
c
cu
>-
-C
4-J
cu
ro
X
cu
n
4-J
in
0
c_>
^ i
C
CU
E
4-1
ro
cu
1_
h-
!_
0)
4->
i
->
0)
4-J
l/l
3
4-1
C
CU
3
I;
4-
LU
<
LU
h-
<
CO
-o
c
ro
i
cj
Q
<
ca
J-
OQ
>-
i_
0
cn
D
4-J
ro
o
r
t^.
cn
^~
»
4-J
l/l
3
cn
3
<
1
O
co
LA
Ct
z
LU
0)
>
cu
>-
cn
O
'o
c
-C
o
CU
h-
<
LU
1-
<
CO
1
o
o
<
CO
o:
oo
.
o co
III I
III T-
1 1 1
o
i O
o
III O
III -
III «
oo
^ o
oo
III «
III **
o
1 O CN J- O
1 0 00 f
1 o t >>O
CN 00
r^-
oo
CN
VJO O O
i o
^ 1 0
" 1 "
LA o
-j-
LA
>
oo
v£l O O
« o o
00 1 O
1
r-- i cn
. l-~
w
CN
o o o o
CN O OO O
OO CN O
CO oo O
T- O
LA
>
r^
< O 1 1
i O 1 |
vO 1 i
o
o
o
o
,
LA
LA
o
o
o
K
CO
o
o
o
oo
LA
1
1
O 0 O
0 CD 0
o o o
. * «»
Csl vD OO
f"> cn
>
O CD O
0 CD 0
cn i~ o
n *i
CN >
w
O CD O
O CD O
O CD O
O Cxi LPl
r- T~ rv.
i CN
^
ro
TJ
l/l
_O
O
O
X
oo
-4-
LA
1
C
O
4->
O
H3
TJ
O
1_
0.
CD
cn
ro
i_
(t)
<
O U
3 D
4J TJ TJ
U O O
31- L.
TJ Q. Q.
O \
i- i/i 1/1 cn
D- -Q XI E
1/1 i
-O O O
0 0 <
O O 0
O « T (
o \ \ o
O Q Q Q-
« O O OQ
^ CQ O
l/l TJ
TJ l/l \ I/) -^ CU
Q- ro -Q cn _Q cn 4J
cn cn E E O
cu
1 1 Q.
X
C C LU
O O
«--. C
4-> 4-J O
ro ro
4-J 4-J T!
1 .- .- 0)
3
o
L_
1-
cu
4-J
2
CU
4->
I/)
ro
3
E
Ll
4-J
C
cu
3
,_
4-
4-
LU
a
o
CQ
E
_J
4-J
c
cu
3
_
u-
14-
UJ
a
o
0
i/i
ro
CO
Q
0
o
4->
C
0)
3
"
4-
4-
LU
l/l
4-J
I/)
C
CJ
ro
4-<
._
Q.
ro
o
r^
ro
4-J
O
H-
1/1
4-J
l/l
O
o
>
ro
D
C
c
<
l/l
l_
>-
o
CN
©
^<
oo
c:
o
4->
ro
N
4->
l_
O
E
<
+ Maintenance
cn
c
ro
i_
cu
CL
0
L.
₯
6
Q.
+
^
x^-
cn
L.
cu
c:
UJ
i/i
4-J
l/l
O
O
m
c.
c
<
,
ro
4->
O
H-
cu
ro
ro
Q_
O
O
ro
o
!-
O
Q
O
CQ
c
0)
E
4-1
ro
CU
ro
O
cn
O
"o
_Q
D
cn
ro
4-J
l/l
0)
Q_
-------
CO
o
<
o
N
C
CD
OQ
C
O
ro
4-1
\-
O E
4-
_1
in
4-J 4-J
in c
O CD
I-1 <
ro LLl
D
__J
>-
O
"o
c
.c
o
CD
1-
^
LU
1
<^
CQ
I
C_>
a
CD
o
(
CJ
Q_
CQ
|
3
a£
2
r \£) V^D O
-^~ CM
III CM
III
III CM
"""CD
T- OO
» r - vo o
OO *- 0
ro cn
iii «
III LA
III
r r^ CM o
LA O O
-4- CM i
111 r~^
i i i
i i i
MD OO
O i LA O O O
1 O -3" CM O LA O
1 CT> OO O Cn
LA CT\ 1^
LA i
CM
O O
0 0
O 0
M ft
vD oo
LA
LA
>
0 0
0 0
O CM
t M
i r^-
o o o
o o o
v^O O O
LA O O
LA OO
r^ CM
*
CXI
1 I
o o o
o o o
o o o
n *s A
cn -j" *^O
-* o
-
o o o
000
v.0 , 0^
f, M
CM cn
o o o
000
o o o
oo r^- o
1 ro cn
- -d-
l \ 1
1 1 1
DRAFT
o
CO
LA
ro
-o
O
i
X
vD
O
O
3
-o
O
V
cn
ro
u
Q)
0
3
TJ
o
1-
in
J3
2
o
O
U
3
T3
0
U
a.
0
O
o
-D
Q.
cn
O
S
0)
4-1
in
5
o
o
\ CQ
in
in
ro -Q
cn
I
c
O
ro
4-1
'E
c
0)
Q
O
CD
en
U
3
T3
0
I/I
J3
0
O
o
Q
O
cj
in
JD
C
O
ro
'i
c
0)
Q
O
o
<
o
Q_
CD
T)
\ CD
cn 4-J
E O
CD
Q.
X
LU
O
T3
CD
in
ro
CQ
Q
O
O
C
CD
O
O
Q.
ro
o
ro
O
in
O
CJ
ro
3
C
C
O
CM
C
O
o
E
CD
U
c
ro
C
CD
cn
c
a)
a.
o
o
Q_
cn
u
CD
C
LU
in
O
CJ
ro
C
c
a)
4-J
CD
ro
s_
ro
Q_
O
L.
C
O
u
ro
U
V-
o
Q
O
OQ
C
CD
E
4-"
ro
a)
ro
O
cn
O
"o
-D
CD
cn
ra
4->
in
CD
CL
4-J
3
-------
t/1
O
u .
O E
tn
O
o
Q)
O
O
o
c
(U
3
CM
oo
0)
03
LA
CO
^
4J
LU
^
>
jC
4-1
0)
2:
03
fl)
1_
H-
u
0)
4-1
03
s
0)
^^
ro
LU
^
<
CO
T3
C
03
h-
O
<
O3
0
Ol
(U
4-J
03
O
Ol
3
«^
1
o
OO
LA
^~
z
LU
0)
>
(U
^J
>.
Ol
0
"o
c
.c
u
a)
h-
<
UJ
l~
^t
00
1
u
0
GO
^
t\
I
o
O.
CO
*
O-
en o
O CO i O |
III « CM O I
111 en o i
iii
0
CM
LA
Cn CO
r^ o vo o i
LA LA O I
LA VO |
III 1-^ CO
III LA
vo cn
en o r*~ o o
CO >>O O O
i vO -j-
III - >
III Cn -j- O
III vD CO
LA
o vo r^ o co o i
1 O CM CO O O O 1
i LA « r^ » o i
| ft A »t
-j- LA OO
CO CO Cn
"^
0
O
*t
O
CO
o
0
o
o
00
LA
>
0
o
o
*»
o
LA
-3-
*
CM
;
1
0
o
o
f.
LA
0
0
o
o^\
M
LA
0
o
o
h
o
LA
CM
,
1
O
o
o
M
,_
en
o
o
,_
^
CM
0
O
O
n
_J-
LA
,
i
o
0
o
t
vO
t
O
o
»
0
0
o
M
vD
CO
1
1
1
O
o
o
Pt
CM
V
CM
0
O
^_
^
OO
-C/>
0
0
o
M
o
_^-
_J-
1
1
DRAFT
o
fl-
OO
(D
T>
4-> TJ
U O
3 U
§ °-
u in
Q. -Q
-D O
O
O
O .-
O
O
-a
O
Q.
01
03
<0
0
Q.
Ol
1
3
O
Lu
u
0)
i
0)
1 1
l/>
.£5
O Q
^^ CO
(/)
f l/l
03 -O
Ol
I
C
o
4-1
OJ
4-1
E
'_!
4-1
C
(U
3
<4-
14-
LU
O
O
Q
O
O
t
-^ in
CTI -Q
E
i
c
o
_
4-1
03
4-1
_
E
_l
4->
C
Q)
3
4-
M-
LU
O
O
Q_
CO
TJ
\ (U
CT) 4J
E U
0)
Q.
X
LU
C
o
TJ
Q)
in
03
CO
o
0
o
4J
c
I/I
0)
D.
Q
O D
oo z:
t CM
03
O
-------
C
O
+J
(0
p
L»
o e
l/l
O
ro
1
I
CO
(1)
F
_a
fD
J_
^
O
CD
0)
4_J
ft!
C_J
,_
P*v.
en
^~
*
4-J
in
3
en
3
«c
1
o
00
LA
>
0)
_J
>*
en
0
0
c
-C
o
i-
^
LU
|
CQ
|_
(_j
O
CQ
ej
<->
J"T
Q_
CO
_J
12
o:
CTi
O MD
Y
III CM
1 1 1
1 1 1
LA
CT\
r^. o
en
!\.
Ill A
III *-"
III
vD
cn o
**£>
«-^
III CM
III
1 1 1
O CM OA O
O OA v£> O
1 OO LA CM OA
|
1 CM
o
o
o
r.
o
00
O
O
O
O
O
J-
LA
1
1
1
o o o
o o o
LA O 0
» ^ M
1 O O
|\ | * Q~\
O OA
,_ ,_
o o o
o o o
CM O i
M »\
OO OA
o o o
o o o
o o o
i en o
LA VD OO
LA OA CM
1 1 1
1 1 1
1 1 1
o
o
LA
*
r^
OA
LA
CM
-co-
o
o
OA
A
«
1
>
o
o
0
o
o
CM
1
1
1
DRAFT
0) o
4-J Q
in <
.TO CQ
o
I
o
Q-
CQ
03
-o
en
_Q
OA
O
bs product
gpd
gals/1000
C
o
4-J
o
-D
O
I-
Q-
OJ
cn
(D
0)
O
i
0)
bs product
000
bs BOD
ation
c
3
Q
O
CQ
mg
bs product
000
bs COD
I
mg
on
at
C
0)
o
o
o
mg
<
o
I-
o
fl-
ea
T3
0)
-I-J
O
(L)
Q.
X
LU
C
O
T3
0)
CQ
O
o
c
a)
I/I
L.
O
CM
ntenance
in
O
a_
TO
o
(3 .-
vg 5
O^- ^
00
§ +
4-1 en
TO c
N
O
Q-
in
O
o
3
C
C
TO
in
O
o
TO
3
C
C
o
en
i- TO
a) a) 4-1
D. c o
O LU (
-------
I
00
_Q
(D
O
C
2.
o
L.
Q.
O
(fl
o
u .-
O E
w C
O 0)
O 3
C It-
(V LU
(U I-
1- <
1- TJ
(U C
0) O
4-i a
i/i <;
.ro co
CO
o
>-
O
en
a)
t/)
O
o
en
CD
3
o
CO
LA
0)
0)
>*
CTl
O
"o
C
o
h-
^c
LU
1-
GO
1
^^
o
^
CO
^
o
1
o
Q-
CQ
3
C£
O on O
P*^. C?
1 1 1 1 1 T- 1 O
1 1 1 1 1 1
1 1 1 1 1 1 0
o
o
en
T- OO
1 1 1 1 1 -3-1 I
1 1 1 1 1 1 1
1 1 1 1 1 I 1
v>
oo j-
o -j-
O t-- i vO VO CD
CO \O v^ CD
III LA m O
III O
O
-~
LA
cn cn
cn en
O LA O t- CM O
i o o cn oo i i
1 O fl CO t 1 1
| » M | |
OO i-
o o o o
o o o o
O vO J" O
S ts M A
CM i^ cn cn
o r~- LA r«->,
CM
1 1 1 1
1 1 1 1
1 1 1 1
>
O O O O
O 0 0 O
O CM OO O
CN CM LA O
t i CM
1 1 1 1
1 1 1 1
1 1 1 1
DRAFT
U
W -O
o o
3 u
Q.
"8
U
a.
u
"D
Q
Q.
JO
o
o
o
X
O
ro
O
U
3
"§
D_
0)
cn
ID
0)
O
Q.
cn
J3 O
o
O
O t-
o ^
O Q
»- O
^ co
tn
ifi ***> if>
(D -Q CTl J3
O
U
CTl
E
(U
4-1
I/I
I
O
m
4-1
c
(U
a
o
co
I
o
c
(U
a
o
o
E
l
<
o
J-
CQ
T3
(U
Q.
X
O
D
0)
tfi
to
CO
Q
O
CJ
c
a>
LU
O
CM
enance
(/I
O
O
(D
4J
'Q-
o
80 z
o
4J CTl
(D C
N
l/l
4-1
U)
O
3
C
C
U
O
to
a)
a.
o
o_
CT)
L-
(1)
C
I/I
4-J
W)
O
o
(D
C
C
(0
4-J
O
-------
r
00
-O
TO
H-
<
o
'i
i_
o
to
c
o
m
4-1
o 'i
14- ._
_i
in
o a)
o 3
4J 4-
C 4-
(D UJ
£
4J <
(D LU
a) i-
O
o
DO
1-
u
P
>*
i_
o
cn
a)
^
fD
O
cn
3
0)
cu
o")
"o
c
_c
u
4)
1-
LU
1
OQ
O
BPCTCA
CC.
-4"
i i i i i i i i i
1 1 1 1 O O r^ I Illl
1 1 1 1 r- t Illl
Illl 1 Illl
1 1 1 1 0 i- cr, 1 Illl
-GO-
CD t O
CM i
Illl 1 Illl
Illl 1 Illl
Illl 1 Illl
-
u.
L_
0)
0)
3
D
4_>
l/l
(T3
3
Is/1000 Ibs product
; BOD/1000 Ibs produc
(D -D
cn
1
c
0
4->
to
4-1
,_
E
_l
4->
c
3
r
4-
4-
LU
a
o
aa
'1
; COD/1000 Ibs produc
'1
^X VI ^*+.
cn _ _a cn
E E
i
C
o
4->
ro
4_)
,«.
E
_j
4-1
C
fl)
3
^~
4-
4-
LU
a
o
o
in
4-1
in
O
(_)
i
fD
4->
»
Q.
fD
' CJ
«
fO
4 '
o
i-
10
4-1
in
0
o
r
(^
3
C
C
«t
in
l_
*"
0
CSJ
^
eg
C
0
Ji_)
fD
N
4_»
l_
0
E
(1)
U
C
c
0)
4_l
C
(D
X
-{.
cn
C
.-
4^
ta
a)
a.
o
in
+j
1- (ft
Q) O
2 o
o
Q_
05
3
+ C
C
<^
;x
cn
I- (T!
a) 4-*
c o
UJ |-
-------
l/l
c
o
4-1
(T!
4->
o 'E.
l/l
O
O
^
1
oo
il)
J3
(D
h-
(1)
T3
X
o
0)
c
(U
^
Q_
o
1
CL
in
O
o
4-J
C
01
E
TO
0)
1_
H
i_
0)
TO
OJ
in
C
D
14-
UJ
LU
1-
<
OQ
C
TO
0
O
cn
0)
TO
o
^
cr\
M
4-1
in
cn
CM CM VD
J-
1 1 1
1 1 1
1 1 1
LA
O O i LA OA O
o OA OA cn -4" LA
O VD -4" T CM
i o r--. CM
i co
-4-"
o
o
1 O
1 "
1 O
o
CM
LA
O
O
O
>
i 0
1 CM
i J-
1
1
1
!
1
O
O
0
M
o
OA
LA
O
O
O
*\
OA
_j-
O
0
o
co
OA
cn
[
1
o
o
o
H
o
LA
-:±
'
O
o
o
n
1
CM
O
O
O
CM
CM
O
1
1
0
o
o
r,
O
O
1"*-
*
O
o
[\
o
o
o
o
J-
T
!
1
o
o
o
t
o
oo
vD
^
o
o
1
^>
VD
-">-
O
o
o
o
o
«~
CM
!
1
DRAFT
o
Q.
CQ
TO
OA
O
X
CO
J-
LA
1
c
o
4_j
u
3
T3
O
Q_
0)
cn
(TJ
(U
TO
4-J
,_
E
l
4_>
C
0)
3
t
14
14
LJJ
Q
0
CQ
T3
0
1_
Q.
in
J3
O
O
O
Q
O
0
\ in
01 _n
E
i
c.
O
_
4-J
TO
4-1
...
E
1
4_>
C
0)
3
i
14-
U-
LU
Q
O
t_)
cn
E
<
0
1-
Cu
m
-D
\ 0)
cn 4-1
E 0
(U
Q.
X
LU
^
0
~Q
O
in
TO
CQ
Q
O
O
4-J
C.
0)
3
"
4-
*4-
LU
in
4-J
in
O
o
^~
TO
4-J
Q.
TO
O
f
TO
4-J
O
h-
in
4-1
in
O
o
TO
3
C
C
<3!
in
L.
O
CM
(o^
o~5
CO
C
o
.
+_>
03
N
±J
l_
o
E
<^
ance
c
0)
4-J
C
II .
ro
z:
+
cn
c
4-1
TO
L-
D
Q_
O
L.
gj
5
o
Q.
4_-
>~
cn
i_
0)
c
u_,
in
4->
in
O
0
i
TO
3
C
C
<£
i
(D
4-J
O
t
-318-
-------
oo
-D
(D
,__
O
_f
o
^f
<
>*
4-J
r:
CO
o
0)
1/5
L.
o
Ul
4 '
tn
O
o
4-J
c
0)
£
ro
d)
u
1-
1_
(U
ro
0)
Ul
ro
3
ui
C
0
4J
ro
4'
'i
_i
c
0)
*4
i+-
UJ
<
LU
I-
<
CO
*^
c
ro
o
<
CO
0
1
p^
(J
CO
Ul
4-1
o
o
,_
r^
cn
LA -
O
4J
>- Ul
O cn
cn 3
ro i
C_)
0
co
LA
z
LU
0)
s
_J
^
cn
O
"o
C
_C
u
H
LU
1
<^
CO
h-
o
Q
CO
^
J- ,- o
CM LA °
1 1 1 1 1 f< 0
1 1 1 1 1 -
1 1 1 1 1 0
CM
CM
>
cn
r^ r^~«
i -4- cn oo o
LA CM O O
111 cn o
III O
o
LA
OO
o o cn i v£> o
o -j- < co CM oo
i o cn LA T co i
i i
1 CM OO OO 1
CO
O
O
O
t
co
vO
o
o
o
«.
CM
CM
O
O
o
o
LA
|
1
1
O O O
0 O 0
o o o
I » »s
CM vO VD
LA cn "
o o o
o o o
vQ oo Cn
oo
>
o o o
o o o
o o o
v£> -4" O
CM LA LA
LA
^
1 1 1
1 1 1
1 1 1
DRAFT
ro
Ul
oo
o
X
en
CM
1
c
o
_
4->
0
3
"§
1_
a.
0)
cn
ro
i_
ri)
^
<;
a.
cn
i
S
O
_
LJ_
1_
0)
4-1
i
0)
4->
Ul
ro
3
Is/1000 Ibs product
s BOD/1000 Ibs product
ro -Q
cn
i
c
0
4->
ro
4-1
,__
E
_J
4-1
C
0)
H-
M-
UJ
Q
O
CO
4-1
U
-a
o
a.
Ul
J3
0
o
o
\
o
Ul "-x
cn j3 cn
E E
i
C
0
4-1
ro
4-1
,_
E
_l
4-1
C
0)
1
u-
<+-
UJ
o
o
o
Ul
4-J
Ul
O
O
1
ro
4-1
_
a.
ro
0
ro
4-J
O
h-
Ul
4-1
Ul
0
O
.
ro
3j
c
c
<;
Ul
o
CM
^
eg
c
o
4-J
ro
N
._
£
0
E
a nee
c
0)
4-1
C
ro
2:
-f_
cn
C
4_)
ro
0)
Q_
0
i_
S
o
0.
-L
>^
cn
0)
C
LU
Ul
Ul
o
p
ro
c
c
<£
ro
4-1
O
1-
-319-
-------
oo
*~
1
00
0)
XI
^
u
0
^-
U)
O
0
^
0) C
C CJ
e
N 4J
CD (0
i- a)
x
i_
0)
03
3
a)
4-J
01
ro
3
1/1
c
o
4-1
m
i i
'I
i
4-1
C
0)
[^
l+-
UJ
UJ
1-
<
CO
XI
c
ro
H
O
O
c£
OQ
«
O
1-
D_
CO
in
4-J
0
,_
^ cn
o *~
4-J
>^ l/l
O cn
CT) 3
4*
ro i
o
o
oo
LA
1
a:
~^
LLJ
o
o
o
o
0
en
co-
1
1
1
o o o o
o o o o
r--. on en en
. * * *
i en CM
o o o o
o o o o
OO vO » LA
u>
0 0 O O
O O O 0
o o o o
-j- on on o
cn CNJ « on
>-
i i i i
i i i i
i i i i
DRAFT
ro
-a
ift
on
o
X
LA
LA
1
C
O
] 1
o
"O
o
a_
OJ
cn
ro
u
CD
\
Q_ fQ X3 CTI XI CTI
CTI CT) E E
C C
0 0
ro ro
i
2
0
U-
1_
I)
4->
ro
3
0)
4-1
l/l
TO
_
E
_l
4-J
C
D
D
LL-
<4-
LLJ
Q
o
CO
^
E
'
4-J
C
0)
>4_
Lt-
UJ
a
o
0
4-J
l/l
0
O
to
._
Q.
ro
' (3
~-
ro
4-J
O
1
(/)
4_J
l/l
O
0
^""
ro
3
C
C
-------
l/l
c
o
_
O E
(fi C
1/1
o
o
G\
r
|
oo
Q
cn <
O CO
O
c
-C
o
I-
BPCTCA
_j
o:
CM
Cn O
! ! !
r--.
CM LA
-4-
1 1 1
1 1 I
cn
i CM .4- r-
LA t
I I I
oo co
O O PA LA CM VJO
1 O cn i LA ro O
1 O .4- CM VO
Cn v£>
LA
1 1 1 1 1
1 1 1 1 1
1 1 1 1 1
i 1 1 1 1
1 1 1 ! 1
1 1 1 i (
>
O
-o
1_
Q.
(/)
JD
^
0
o
o
"^
01
^~-
fD
cn
4J
O
|
Q.
in
-Q
0
O
O
V
*^^
o
o
CO
I/I
o
1
C
o
_
4-1
fD
4->
_
E
1
C
CD
D
4-
4-
LU
4-1
O
3
-a
0
^
Q.
l/l
-Q
0
o
0
^~
"^^
0
0
CJ
^-~
\ l/l
cn jo
E
1
C
O
M
13
-M
^
E
_j
4->
C
0)
4-
4-
LU
^^
cn
E
I
<
CJ
0
a_
CD
T3
\ fJJ
cn 4-1
E 0
co
a.
K
UJ
c
o
~o
o
fD
CO
a
o
CJ
4-1
c
0)
o
CSI
eg.
enance
0)
Q
O
CO
a
o
CJ
4-
LLJ
O
O
Q.
ro
o
fD
o
oo
c +
o
4-1 cn
ro n
N
O
Q-
in
4-1
i/i
O
o
ro
D
C
C
fD
in
O
CJ
ro
c
c
cn
i_ !_ i_ ro
O 0) CD J->
£ a. c O
<£ O UJ h-
-321-
-------
in
C
o
(D
4-1
I- .
O E
14- ,
_J
in
4-> 4-J
in c
O (U
(_> 3
l/l
O
LA
I
O
CM
00
1
1
DRAFT
O
ui
i
2
O
p
U-
u
(U
i i
(D
2
0)
1 1
in
-2
O
o
o.
in
O
O
O
O
CO
in
JO
C
O
(D
'i
C
0)
O
o
CO
cn
E
"O
o
Q.
in
O
O
O
Q
O
O
C
o
O
CM
ntenance
in
O
a.
ro
o
c +
O
4-1 cn
to c
N
O
Q-
in
4-J
in
O
o
ro
C
c
in
O
o
Q
CD
a.
o
a
LU
(U
Q
O
CJ
(D
O
C
c
-------
CM
I
oo
_o
ro
H-
0)
c
CD
J3
O
(fl
(fl
c
O
O E
1/1 c
O 4)
o 3
C M-
0) UJ
E
4J <
re LU
0) H-
i- <
I- CO
I- t3
0) C
4-! TO
0) O
10 <
- Q
° s
"o
c
o
/n
T<
BPCTCA
_t
Q;
CM O
1 1 1 1 1 'LA
1 1 1 1 1 t -J-
1 O -31 .LA t
LA CM VO VO OO
O O O O O
O O O O O
o o o o c
o CM in i~- oo
o o co J- oo
O t i CM LT>
£.
o o o o o
o o o o o
O ro CM CM f~~~
O LA MD «
LA i CM
o o o o
o o o o
O O i O O
O O LA LA
O ro ro VJD
CM J- J-
ev
1 1 1 1 1
1 1 1 1 1
O
oo
LA
DRAFT
4-J ~O
u o
-O Q-
O
i- in
a _o
TJ
O
Q.
in
Q.
CO
in
J3
ro
O
X
ro
C
o
-a
o
0)
cn
05
0)
-D O
O
O
O i-
o \
O Q
t- O
\ CD
in
-O in
Q- 05 -Q
cn cn
i
0}
c
o
05
4-1
I ._
s J
O _J
C
05
in
JP
cn
O
o
o
Q
O
C
o
05
4-J
'i
cn
enance
Q
O
CO
Q
O
O
in
O
o
a_
03
o
o
CM
(s
co
C -f
o
4-1 cn
0! C
N
O
CL
in
O
o
c
c
l/l
o
o
o
E
05 Cn
U L 05
(1) QJ 4-1
Q- C O
O LU I
ro
05
O
C
C
-------
CM
CM
OO
_Q
m
(U
c
'i
OJ
<0
c
0)
0)
(0
X
0)
U1
C
O
L. ~
O E
l/> C
O 0)
(_J 3
in
o
o
VD .-
OA
o
o
o
*
o
,
CM
O
O
O
x
CM
J"
>-
0
O
O
o
o
J-
*«
CM
1
O O O O
O O 0 O
j- oo cn ^-
s *> w
« cn OA LA
CM vD OO p^
O O o O
O O O O
OA -4" « OO
J" « LA
0000
o o o o
o o o o
LA O~\ vO O
-3- CM vO J-
CM t- J-
-c/v
till
1 1 1 1
DRAFT
U
O
"8 "8
(D
13
in
J3
PA
O
CM
O
U
o
o
-o
Q.
I
(U
(U (D
CT1 2
<0 V
L. 4->
(U in
> <0
o
TJ Q.
o
i- in
Q. -D
in
£ §
o
O i-
o ^
O Q
t- O
^^ CO
in
in *^
(D -Q cn
cn E
O
fO
4J
'I
C
0)
LU
O
O
CO
Q.
o
o
o
o
o
o
in \
-Q O>
E
O
to
"e
§
LU
a
o
o
O
CM
ntenance
in
O
CJ
%0 «
sf 2:
00
o
4-1 cn
(D C
N
O
O-
m
o
o
c.
c.
4-»
._
Q.
tD
0
_
(TJ
'S
^j
I/)
O
o
to
3
C
C
U
O
E
i_
0)
Q.
O
u
(U
c
LU
10
4J
O
1-
-------
l/l
C
o
O E
I/)
O
CJ
ro
CM
g
oo
(U
1
_C
fO
h-
Q)
'
L.
4-J
C
O
"^7
^
u
<
O
CJ
4->
C
0)
^}
ro
0)
1-
l_
4-1
C
cu
'<
14-
UJ
(=j-
LU
<
DO
"T3
C
MD
1
CJ
u
0
CD
0)
4-1
ro
o
cn
*,
w
-^
Ol
u
1
o
oo
0)
0)
^
cn
O
"o
c
f-
o
1)
h-
<^
UJ
1
^
CO
H
o
Q
CO
"^
*J
1
Q_
CO
_J
3
an
LA V£)
o vo cn o o
ro i LA o
T- vO O
III
III J- O
III O
LA
CM
^ -d-
-j- O CM O O
ro LA O O
III r~ J- O
III r- CM O
I LA
>
-4-
r~- CM
LA O ro O O
^O v£> O O
III ro O O
111
III r LA O
. O
CM
LA
o LA o o cn o
o o vo o r*-» o
I O LA CM . LA 1
1 CM J- CM 1
oo < ^t
O
o
o
LA
LA
CM
O
O
ro
LA
O
O
O
^
O
ro
LA
1
1
1
0
o
o
0
1
cn
i--
o
o
I
^
o
o
o
>
o
-f
I
1
1
o
0
o
o
CM
LA
CM
o
o
CM
O
o
o
*
o
1
1
1
o
o
o
LA
^
o
o
vD
s
CM
>
C
o
o
s
o
LA
O
O
in
T3
O
O_
l/l
O
O
o
ro
cn
o
I
Q.
1/1
XI
o
o
o
o
in
c
o
O
i
-------
I/)
o
TO
4-1
o 'I
14- ._
_l
in
4-< 4-1
ui C
O 0)
O 3
O
O
00
-O
TO
V
-C
"^"*
t
Pentaei
c
E
TO
0)
t-
!_
0)
fO
14-
UJ
<
UJ
5
CO
a
c
<_>
>.
L.
O
CT)
0)
4-J
TO
O
n
U1
3
01
i
o
oo
LA
1)
1
O)
O
"o
c
0
1)
UJ
CO
t
o
0
«^
CO
^
1-
Q_
CO
3
o£
vD
OO O
MD
III OO
1 1 1
-4-
0 0
pr\ Q^
CPv
CM
1 1 1
1 1 1
1 1 1
vO O
OO O
VO
III CO
1 1 1
1 1 1
0 0 0 O
O CM O~^ O
1 OO CM OO oo
1 oo i- CO
OO oo
00 O
t O
£
CO O
O 0
CO CO
*
cr\
CM
LA O O
VJD O O
oo O~\ -4"
LA CO
ro -J"
LA O
CO O
LA O
** *s
^ vD
LA
O
O
O
O
O
O
^
O
o
o
oo
0>
o
0
o
o
o
r a
(D CO
o
Q_
CO
(D
-a
OO
o
X
LA
oo'
o
g
0)
CTl
TO
0)
u ^
O O
3 l-
-0 Q-
O
u in
Q. -Q
1/1
-Q O
O
O
O <^
O ~~-
O Q
- O
>* oo
Ifl
CL ro -D
Di CD ^~
O
TO
I .-
O J
LI- 4->
C
D_
m
~a
01 4-j
E O
1)
D-
X
UJ
O
a
a)
>
TO
CO
Q
O
O
4->
C
O
CM
nance
(D
in
O
o
Q_
TO
TO
O
4-1
1- 1/1
m O
5 o
o
a_ ^
TO
3
4-J O1 + C
TO C C
N <
oo
O
l/l
O
TO
C
C
o
E
TO
OJ 0)
Q- C
O LU
cn
-------
l/l
C
o
LA
CXI
I
OO
_D
TO
I
in C
O 0)
O 3
(L) LU
4-- <
TO LU
TO
(D O
4-> Q
l/l
0)
1
en
0
"o
c
^
u
a)
I-
^
UJ
1-
^C
CQ
h-
O
Q
CO
O
1
O
fl-
ea
_j
"-
oo
cx; LA
O O ^£> cxi O
i OA O
III CX! O
III
III O
O
O
CXI
01 -d-
co r--. « o
^O T CXI O
vO -O
III -
III LA
ill r--
OA
CXI CX! > O1 LA O
OO CXI J- i O
l-~ -4" O
o
III 0
III 1
III
0000^1^.0
1 O \£> i J" PA -d" 1
1 O PA ^£> PA 1
| . |
-4" PA i
O
LA
0
O
O
rxi
0
O
O
01
r--
o
o
o
ai
r--
_^-
i
i
j
o o o
O O o
^ PA oq
PA [^ CXI
OO ^D LP,
CXI
000
o o o
LA PA r-.
LA PA
C-J
O O O
o o o
o o o
v£> LA O
OA "j~ *^O
-d- o\
1 1 1
1 1 1
! 1 '
DRAFT
(0
-o
i/i
-D
PA
O
X
o
LA
1
0
4J
O
1
w
a_
a)
en
TO
i
a)
>
«i
w - gpd
gals/1000 Ibs product
imitation - Ibs BOD/1000 Ibs product
mg/1
imitation - Ibs COD/1000 Ibs product
mg/1
ased on Expected BPCTCA - mg/1
o
IZ
i-
o
4->
TO
3
(U
4-*
tn
TO
S
1
C
0)
3
i
14_
14_
LU
0
O
CQ
'
4-1
C
0)
*
14.
H-
LU
O
O
O
CQ
Q
O
O
4-1
C
(U
^~
M-
14-
LU
1/1
4-J
l/l
O
t_3
TO
4-1
CL
TO
C_5
!
TO
O
|
01
1/1
O
o
, .
re
^
c:
c:
<;
i/i
L.
O
CXI
co
c
o
4-)
TO
N
L.
o
£
ng + Maintenance
TO
0)
CL
o
+ Power
nnual Costs
<
en
1- TO
0) 4->
c o
LU |
cr>
c
O
u
en
O
4-1
fO
C
0)
o
c
o
o
<
LU
h-
OQ
E
C
o
Q
0)
E
ro
s_
(D
Q.
O
u
m
o
Q
o
CQ
-327-
-------
CM
1
co
t/1
ro
3
1/1
0
4-J
ro
._
E
_j
4-<
C
CO
3
^^
W
LJ
<
LU
5
CO
T3
H
O
<
CO
o
o
0-
co
I/)
4->
l/l
0
O
,_
r~~
m cn
i ^^
4-J
U 3
O cn
cn D
CO <
4-i
ro I
o
o
co
LTv
T
oi
z
LU
CO
CO
cn
0
"o
c
P"
u
CO
H
UJ
<;
CO
1
O
<£
CO
o
I
o
Q-
CQ
.-J
-£
CM LT\
O r«i ^O -d~ O
CM J- O
III ^D O
III LT\
CM
cn j-
»- o r-- o o
cn » CM o
r~" !***« C3
III VD
III 1
CO-
CM co « o o o
CM CM CO O O
CM O J- O
CM r<-\ O
III LT,
III
III
CO
O O CO O . O
P-- i LA LTl < O
i vo CM r~- i o i
i * «\ * * i
1 « i LA > 1
O 0 O CD
0 0 O CD
VQ -d" J- -d"
C\T (V^ ("0 Cn
O O O CD
0 CD 0 CD
VD CD i- r--
r r CM
O CD O CD
O CD O CD
O O O CD
r^ CO LT\ CD
T ro CM' OO
* *
co-
1 1 1 1
1 1 1 1
DRAFT
ro
"O
w
-O
O
*~~
X.
co
CO
r
1
C
o
4->
(J
3
"S
1_
Q.
CD
Cn
ro
i_
0)
<£
-Q
Q.
cn
i
3
O
.
Ll-
L_
CO
4->
ro
2
0)
(/)
-2
^
u
S "§
3 L-
-0 CL
O
^ 1/1
Q- _Q
(fl
.0 0
O
O
o -
0 \
o o
r- O
~«~ CO
l/l
U1
ro -Q
cn
i
c
O
4-1
ro
_
E
_l
4-1
C
co
3
^
<4_
14-
QJ
Q
O
CD
O
3
O
l_
Q.
l/l
O
0
o
r~
^
Q
O
C_J
^* l/l
cn _o
E
i
C
o
4-1
ro
4-1
_
E
_i
4-1
C
V
3
T
4-
14-
LU
Q
O
CJ
E
1
O
H
o.
CO
TJ
\ (U
cn 4-1
E 0
CO
Q.
X
LU
C
o
"Q
CO
l/l
ro
CO
O
(_J
4->
C
CO
3
i
M-
M-
UJ
l/l
4^
I/I
O
(_)
ro
t
Q.
ro
o
ro
4-1
O
h-
1/1
1/1
O
CJ
F
ro
3
C
C
.
O
CM
&
CJO
C
o
^
4-1
ro
N
4-J
O
E
^
CO
u
c
ro
c
CO
4-1
c
ro
+
cn
c
.
4-J
ro
CO
Q.
O
L_
SI
0
0_
+
>*
cn
CO
C
LU
I/I
4.J
tn
O
t
fD
13
C
C
<[
OJ
4J
o
i-
-228-
-------
in
O
i- .-
O E
Ul
o
o
CM
00
in c
O 0)
CJ 3
4J U-
C (4
0) LU
E
4-* <^
fO LU
01 (-
1- <
1- CO
1
Q
>^
1»
O
CD
l/l
3
Ol
3
<£
ej-
UJ
1-
CQ
U
0)
-1 h-
o
>~ £3
Cn <
O CO
"o
C
-C
o
(!)
h-
<
,
J "
Q.
CQ
_J
3
Q£
LA
CM
*-
1 1 1 1 1
1 1 1 1 1
1 1 1 1 1
OA
1 1 1 1 1
1 1 1 1 1
1 1 1 1 1
co
cn o OA
. LA O
-3-
1 1 1
1 1 1
1 1 1
o o r^~ o o
i o o cn o vo
i o -3" OA o o
| » ts * «\
CM 1 . t CM
cn j-
vO o
t- O
rA O
O
O
00
f.
J- o
J- o
oo o
*
0
t
CM
-CO-
o o o
CM < O
O vO O
1 T O
o
LA
-
!
0
CM 1
" 1
LA
O O O O
O O O 0
o o o o
.3- -d- CM o
OO CM LA ^D
t >>D I. LA
i
I
O O O O
o o o o
J- OO rA LA
* »l M
>- cn t-
CM OA
>
O O O O
0 0 O 0
0000
o o o o
r~^ OA o o
r- i\ oo r-~
« -
CM
1 1 1 1
1 1 1 1
1 1 1 1
DRAFT
TS
t-
0) C
4-> (TJ
SH
a> cj
.TO CQ
CL.
CO
0
O
00
LA
CC.
z
LU
u
4-J Tl
U O
3 1-
§ °-
u in
Q_ n
O
3
T3
O
cn
E
ro
-o
i/)
_Q
^^
f\
O
^y
f^^
B
v£>
^
1
C
O
_ .
4-1
O
3
"D
O
in
-Q O
O
o
0 *-
o \
O Q
«- o
\ CQ
in
~o *~~ in
Q_ m n
cn cn
1
C
O
._
4-1
ro
i
E
> t
0 -I
r
U- 4-1
C
o
o
o
*\
o
0
o
r~
*^x t/1
cn (*t
E
I
C
O
_
4-1
(0
_
E
i
4_)
C
1
c^
t_)
^-
Q.
CQ
T3
\ tt)
CT1 4-J
E U
0)
Q.
X
LU
C
0
"O
0)
I/)
(U
CQ
r">
0
Q-
cn
ro c
N
o
Q-
in
O
O
(D
C
C
L.
o
(\j cn
i_ i_
(D 0)
Q- C
O LU
in
O
C
C
ro
O
-329-
-------
CO
CM
oo
0)
ro
*->
CJ
J
o
o
CO
t/1
c
o
re
4->
S_ .
O E
M_ ._
_l
I/)
4-1 4->
in c
O (U
O 3
4-J 4-
C >4-
0) LU
re LU
u ^
I- co
i- -o
Q) c
4-1 re
o> o
4-1 a
ca
u
o
Q_
CO
-3-
Q
O
cn
0)
4-1
re
o
O
o
1/1
3
cn
3
O
CO
LA
15
0)
*
>^
cn
O
"o
C
f
JL
o
-d- J- O
r ^~ on t^
CM on cn
ill j-
i i i
i i i
CO
cn LA on o
T- cn o cn
CM -d- cn
III LA
1 1 1
1 1 1
0 0 J- O -d" 0
O LA CO v^> 00 on
1 O O CM CM CM
| « M ft *
i -df co i -d~
CM
,
O
1 O
1 O
0
on
cn
0
o
o
1 LA
i on
1 T
^n-
0 O
«- O
on o
i O
o
LA
1 1
| |
' 1
O
0
0
r^C
cn
o
0
oo
on
v
0
o
o
s
o
CM
LA
|
|
1
O
0
o
^o"
vO
O
o
LA
LA
o
o
o
o
-3-
1
1
1
o
o
o
o
LA
o
0
CM
o
0
o
t
o
*
1
1
o
o
o
on
^
"*
o
0
LA
cn
^~
>
o
o
o
o
LA
o
w
1
1
1
DRAFT
ro
TJ
in
-Q
CO
o
u
3
§
a.
a
cn
I
IT
u
-o
o o
-O Q.
O
t- t/1
Q. fi
l/l
J3 O
o
O
O i-
O \
O Q
i- O
^^ CO
I/I
I/I
ro -Q
cn
i
o
re
4-J
'I
c
a>
o
o
00
o
-a
o
u
a.
o
o
o
a
o
o
cn -Q
O
cn
c
(U
LU
E
I
o
O.
CO
U
(U
a.
x
o
re
CO
o
o
c
(U
O
CM
enance
c
0)
Q) Q)
E *-
TO 4-1
re
Q. TO
O
O cn
!- O
4-1
C O
O
O -Q
i/l
O
Q.
re
re
o
So *
o
+j cn
ro c.
N .
l/l
O
re
c
c
o
ro
Q.
1- VI
M O
o °
Q-
re
+ c
c
cn
i- TO
(1) 4-«
C O
LU |-
re
o
a)
cn
re
i- a)
o
Q_
O 3
DO s:
- CM
-330-
-------
CTi
CM
oo
a)
O
<
C
c
ro
O
O E
U1 C
O 0)
O 3
ro UJ
(U (-
l- <
c
(TJ
(A <
J3 DQ
O
_
Q_
CD
I/)
O
U
O
cn
O
4-1
(D
C_J
tn
3
CTI
o
00
LA
CU
0)
_l
>-
en
O
"o
c
^
o
t-
<
LU
K
^C
CQ
j_
CJ
O
^^
CQ
^
1
0.
CQ
_J
o;
LA O 1
CM O 1
i LA I
CM
1 1 1 1 1 t-
1 1 1 1 1
1 1 1 1 1
-d-
v£> O -d" O 1
« -4" rA O 1
vO rA -d" 1
III t- rA
III CA
l i i
co
CT\ O rA O O
« co o o o
CTi -d" rA CM^
O rA
III -d- rA
1 1 1
1 1 1
O O rA O O O I
O O LA O r^ O I
1 O CM i OA O O 1
1 *
i o i LA i r-~
vO T- O
o o o o
o o o o
O CM O O
O t o O
O vO r^- O
v£> r~^ o
^
0 O O O
o o o o
O CO LA rA
vD i CO
>
o o o o
o o o o
o o o o
0 CO CN O
O * CM rA
< -d' rA i
*
-d-
1 III
1 III
1 III
o
o
CM
J
(Y
CO'
0
o
vD
VD"
t
-ov
o
o
o
co
DRAFT
T3
U1
OA
O
X
LA
LA
1
C
O
4-"
O
3
-D
O
O_
fl)
en
ro
^
!/)-
"O l/l \ (/) ^
Q. ro xi en xi
en en £
i i
§ j
|
3
O
Lu
OJ
4-1
ro
5
ro
^
4-1
ro
4-J
_
E
_J
4_l
C
0)
f^
14-
4-
LJ
Q
0
CQ
4->
ro
4J
'i
i
4_j
C
3
i^
14-
4-
LU
O
O
o
mg/ i
n Expected BPCTCA - mg/1
o
T3
0)
If)
ro
CQ
Q
o
0
4-1
c
0)
L^_
14-
LU
CO
4-J
C/l
O
O
,
ro
O-
ro
o
,
ro
4-J
0
h-
10
4_j
LO
0
(_)
^~
ro
c
c
e<
10
0
CM
00
c
o
4-J
ro
N
4-1
U
O
e
^
Mai ntenance
+
en
c
1 1
ro
0>
CL
o
co
4-J
1- 10
cy o
-i o
o
D_
ro
+ c
C
^
en
i- ro
0) 4-J
C O
LjJ I
0)
E
ro
i_
ro
Q.
C
o
o
ro
U
L.
O
Q
O
CO
-------
0
1
00
0)
.0
TO
1
T3
o
<
o
1_
4-"
(_)
u
o
u_
)
4->
o
CJ
4-1
C
C
tt)
^
r
4-
14-
LU
<
LU
h-
<
T3
C
(D
I
r^
O
Q
<£
CO
o
1-
0_
co
1
Q
>
U
O
01
(1)
TO
0
Ul
4-1
O
CJ
cn
^
4-J
in
13
cn
3
<
1
0
oo
LA
C£
z.
UJ
«
>
OJ
P>v
cn
O
C.
o
D
1-
^
LU
|
^
CO
J_
o
o
<£
CO
2
1
o
o_
CQ
^
-2
CC
oo
T CM
-J O LT\ « 1 O
« CM ^O 1 O
r CM 1 O
III
III O
III O
o
0
vO J- J- vO 1 O
i oo oo CT1 1 O
m M) i o
III 0
ill .3-
w
oo
O^l » OO O LA O
« -3" O -j- CM O
j- oo -d- o
o
III O
III 0
III »
co
CM
o o oo -3- r-^ o i
o o CM oo LA r-^ i
O LA oo >>£> v£> oo 1
1 «\ ». *. 1
i o r^ < i
1 O LA 1
oo
I--
o
o
0
o
CM
o
"~
o
o
'-A
LA
i-
o
o
o
0
v^
oo
4
CM
1 1
1 1
1 1
o
o
0
o
o
cn
*
o
o
0
'CM
"
O
O
o
o
-3-
o
^
CM
I
1
1
O
o
0
o
cn
o
CNI
O
O
oo
,
o
o
o
o
o
oo
i
»
1
1
1
O
o
o
o
o
LA
o
o
co
oo
lp_
^S>-
o
o
o
o
o
»1
MD
1
1
1
DRAFT
cn
c
ra
-o
I/)
oo
o
X
O
00
1
c.
O
O
3
|
o.
0)
cn
(D
(U
^
4-
4-
UJ
Q
0
0
01
<
CJ
1-
o
0-
CQ
T3
E 0
Q.
X
LU
O
TJ
TO
CO.
O
O
4J
C
0)
3
*
4-
U-
LU
Ul
4-;
l/l
O
TO l/l
l/l
Q- O
TO CJ
O
~
TO
TO 3
4-> C
O C
1- <
OJ
ui U
L. C
>- TO
C
O (U
CM 4~l
C
TO
co
c +
o
_
4J Ol
H3 C-
N
4-1
1- 1.-
O 0)
E Q-
< o
ui
4-J
L. Ul
O
0
Q-
CD
3
+ C
C
e^
>^
cn ~
1- TO
C O
LU H-
^/
O
^v.
Ol
E
O
4-
o
c
o
4_l
L.
4-1
C
<1)
o
O
u
<£
LU
1
1
fy)
E
c
2:
parameter.
o
u
4J
C
0
o
'rn
O
±J
L.
U
o
o
fSJ
-------
I
00
-Q
ro
in
c
o
ro
4->
L.. .
O E
M_ .-
_J
in
4-> 4->
in c
in O 0)
T3 03
C
CD
CL.
fD
O
O
cn
c
(U
E
ro
0)
u
H-
u
ro
14.
LU
LU
(-
<
CO
T3
C
03
>
0
cn
V
4-1
ro
O
V
4->
in
cn
3
<3;
1
0
oo
LA
0)
,
cn
O
"o
c
(
o
a)
i-
^
LU
^
CO
1
o
o
<^
CO
^J.
o
o
Q_
CO
.J
^
Q£
00
cn
CXI
r- O LA O O
III LA CXI LA 1 O
III * 1 O
III * 1 *
ro c""i
O
LA
MD J" J- O O
T i oo O O
-3- oo _j- O
III OO 1 CXI
111 i cn
ill i
co-
CO
cn o ro o o o
i O O O ro O
LA Jj T CO O
III O CXI O
III i- O
III ro
-d"
O O r- O OO O
o LA -3~ *£> cn cxi
1 O r-- i LA CXI LA 1 1
1 . . . - 1 1
i -d- j- ro r^ i i
o
o o o o
o o o o
0 0 O 0
» * n M
ro -j- o r^-
LA o -4~ cn
i i CXI
*f\-
o o o o
o o o o
-S" LA i O
Cn ro co
1
>
o o o o
o o o o
o o o o
OO CXI O O
ro cxi J" o
-d- ro T cn
1 1 1 1
1 1 ! 1
1 1 1 1
DRAFT
(D
4->
in
U
3
T3
O
TJ
O
1-
CL
O-
ca
(0
TJ
O
cn
CXI
O
o
-o
o
0)
cn
ro
in
-Q O
O
0
0 <-
o \
O Q
r O
\ OQ
in
"a in
D. (D -Q
cn cn ~-
§
to
I
o !3
T3
O
L-
Q.
O
O
0
Q
O
O
in
JD
§
CD
E
L.
(L)
4-J
i
(U
c
0)
a
o
co
c
(1)
o
o
cn
E
<
o
0-
O3
4-1
o
0)
CL
X
0)
in
ro
co
Q
O
o
c
0)
O
CXI
enance
in
O
o
Q.
ro
o
©, -
c +
o
4-J cn
ro c
N
O
0_
in
O
o
O
E
0)
E
ro
!_
ro
CL
c
O
U
ro
u
c
-------
DRAFT
SECTION IX
BEST PRACTICABLE CONTROL TECHNOLOGY
CURRENTLY AVAILABLE (BPCTCA)
Best practicable control technology currently available (BPCTCA) for the
Secondary Organic Products segment of the organic chemicals manufacturing
point source category is based upon the utilization of both in-process
controls and end-of-process treatment technologies. The goal of these '
controls and technologies is the reduction and eventual elimination of
oxygen-demanding materials from all discharges. These pollutants (,as
measured by the BOD, COD, and TOC parameters) are responsible for most
of the degradation this industry presently inflicts upon the aquatic
environment. The following discussions describe the pollution control
technologies commensurate with BPCTCA and the procedure used to establish
effluent limitations guidelines.
The nature of the specific manufacturing process will determine the com-
bination of process controls and modifications which are best suited for
at-source water pollution control. However, some practices are generally
applicable to all process plants within this point source category.
The first of these is the implementation of process observation and
sampling programs to determine the identity, location, quantity, and com-
position of all aqueous streams within the plant. Monitoring should in-
clude all aqueous process streams as well as storage tank drainage, flare
and pump seal waters, storm runoff, and wastewater associated with support
activities such as laboratories, materials receiving and shipment, and
intake waters treatment. Although not considered as major sources of
oxygen-demanding pollutants, utility waters and steam condensate from non-
contact cooling and heating should also be included in any such survey.
The flows and loadings developed should be allocated to the different
processes in the plant in terms of product ion-based ratios. This provides
a clear understanding as to which specific products require high water
utilization or generate large amounts of water-borne pollutants. At the
present time, this information has been developed by only a very few manu-
facturers within this industry.
Waste characterization studies of this type logically lead to the selec-
tion of various streams for segregation or the application of at-source
controls. Exemplary chemical process plants segregate all wastewater
which has become contaminated with oxygen-demanding hydrocarbons. These
wastewaters include contact wastes which flow continuously from within
the process battery limits as well as intermittent wastewaters which have
contacted chemicals in other sections of the plant. The segregation and
collection of these contaminated wastes from non-contaminated streams
such as non-contact cooling waters appreciably reduce the volume of waste-
water to be treated in a centralized wastewater treatment plant.
NOTICE: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-335-
-------
DRAFT
In most chemical manufacturing processes, the volume of non-contact
cooling water required is between 3 and 100 times greater than the water
which contacts chemicals within the process. The exact ratio depends
upon whether the plant uses once-through cooling water or a recirculating
system with cooling towers. However, the quantity of oxygen-demanding
hydrocarbons present in these waters is less than one percent of that
present in contact waters. The major pollutants associated with non-contact
waters are inorganic anions and cations existing as dissolved solids and
other chemicals to control slimes, algae, and corrosion. These materials
do not normally affect dissolved oxygen levels in receiving waters.
In exemplary plants where contact waters are collected separately from
non-contact waters, it is possible to design and operate a treatment
system which has been sized to treat the optimum hydraulic and organic
pollutant loadings from the manufacturing operations. Fluctuations in
influent concentration can be reduced by equalization of the wastes.
This practice also serves to smooth out shock loads resulting from pro-
cess upsets. These may normally cause concentration gradients which
could be toxic or inhibitory in a biological treatment system.
One of the most common at-source controls utilized by organic chemical
plants is the separation and selective burning of hydrocarbon by-products.
These materials are invariably formed because very few chemical reactions
are 100 percent selective in the formation of the desired product. In
almost every plant surveyed for the production of secondary organic pro-
ducts, some waste organic chemicals are disposed of by burning or hauling
to landfill disposal. Although they may not be significant on a flow
volume basis, these materials if flammable should not be discharged to the
sewer and if not flammable would increase the quantities of hydrocarbons
present in wastewaters by an order of magnitude if they were combined
with aqueous process wastes.
The devices used for burning may range from simple flares (for materials
with high vapor pressures) to complex liquid waste incinerators or py-
rolysis furnaces with extensive air pollution control equipment. In
many cases, waste organic by-products are suitable as auxiliary boiler
fuel. Although the continuation of this practice should be evaluated to
insure that adequate air pollution control devices are utilized, it
represents a significant contribution to the reduction of hydrocarbons
present in the wastewaters from most plants surveyed. For this reason,
it is considered as one of the generally applicable in-process control
practices for BPCTCA.
A second practice involves the separation of insoluble hydrocarbons from
process wastewaters. In many cases, this can be accomplished by simple
gravity separation or the use of flotation systems. The fatty acid product/
process grouping uses this technique to advantage. Insoluble hydrocarbon
skimmings are collected and treated by filtration and sulfuric acid
addition to increase product yields.
NOTICE: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-336-
-------
DRAFT
Process Modifications consistent with BPCTCA include the regeneration and
reuse of aqueous working fluids in the process. This is normally done by
vacuum evaporation. Typical examples include the regeneration of sulfuric
acid in processes to produce isobutylene and secondary butyl alcohol.
Other modifications include the direct recycle of a working aqueous fluid
such as the absorbent in a gas scrubber or water used in barometric con-
densers. These techniques are utilized in the processes to manufacture-
benzoic acid and hexamethy1ene tetramine. In these cases, the volume of
contact water discharged is reduced from once-through operation to the
blowdown from a recirculating system. Because of the diverse nature of
even the limited number of processes examined in this study, it is not
possible to generalize beyond practices such as sewer segregation and non-
aqueous disposal of hydrocarbon wastes. For this reason, BPCTCA effluent
limitations guidelines were calculated based upon an end-of-process treat-
ment model.
End-of-process treatment technologies commensurate with BPCTCA are based
on the utilization of biological systems including the activated sludge
process, extended aeration, aerated lagoons, trickling filters, and an-
aerobic and facultative 1agoons. These systems include additional treat-
ment operations such as equalization, neutralization, primary clarifi-
cation with separation of insoluble hydrocarbons, and nutrient addition.
Final removal of suspended solids is accomplished by clarification.
Although the activated sludge process is considered as the single treat-
ment system most generally applicable to the wide variety of wastes gen-
erated by this industry, it must be recognized that many specific pro-
cesses generate wastes which, if directly treated, would be toxic or
inhibitory to a biological system accepting only those wastes. However,
most of these wastes become non-inhibitory if the influent concentrations
are reduced from the extremely high values existing in the raw waste.
This has been demonstrated in biological systems treating chlorinated
hydrocarbons from the manufacture of ethylene dichloride and vinyl chloride
and in systems which treat the combined wastes from a batch chemical plant
where the treatability of individual batch wastes may vary widely.
In the case of specific wastes which are biorefractory at any concentration,
BPCTCA does not preclude the use of carbon adsorption or other types of
physical/chemical treatment to achieve BPCTCA effluent limitations guide-
1i nes.
It should also be noted that, at the present time, most manufacturers
have elected to dispose of toxic, inhibitory, or difficult wastes by
means of deep-well injection or ocean dumping. The geographic location
of many plants manufacturing these materials has made this a rather common
practice. If deep-well injection or ocean dumping were considered as
viable alternatives consistent with BPCTCA, some of the process plants
NOTICE: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-337-
-------
DRAFT
surveyed would not discharge any process wastes to receivinq waters.
However, because of the potential danger of underground leakage or con-
tamination inherent in the practice of deep-well disposal, effluent
limitations guidelines have been developed as if these wastes were to
be treated for discharge to surface waters.
The effluent limitations guidelines for BPCTCA were developed using
a step-wise approach starting with the process raw waste load (RWL).
The process RWL is the production-based ratio relating specific pol-
lutants to specific products manufactured by specific manufacturing
schemes. During the field sampling program, the process RWL was
developed for different manufacturing operations by sampling contact
process wastewaters. The RWL is the necessary link between the wide
diversity of products and manufacturing operations existing in this
industry and the production-based effluent limitations guidelines.
A single set of RWL values was first established for each of the
secondary product/process groupings covered in Phase II of this study.
This set of values included the following pollution parameters:
Contact Process Wastewater Flow (liter/kkg of production)
BODc; Raw Waste Loading (kg/kkg of production)
COD Raw Waste Loading (kg/kkg of production)
TOC Raw Waste Loading (kg/kkg of production)
It was necessary to develop single numerical values for flow as well
as values for each of the three waste load parameters despite the fact
that field sampling data indicated that significant variation exists
between sampling periods for a single process and between different
manufacturers operating nominally the same process.
The single set of values assigned to each process was designated as
the RWL which can be obtained through the application of in-process
pollution control practices which are commensurate with BPCTCA0 These
practices were discussed previously in this section. The actual RWL
values were indicated with each of the separate process descriptions
presented in Section IV.
It should be noted that the lowest observed values were not arbitrarily
chosen as RWL for BPCTCA. Factors such as process plant age, size,
geographic location, and method of discharge (to surface waters or to
a municipal treatment plant) were considered when drastically differing
RWL were obtained for the same process.
There were many cases where only a single value from one sampling period
could be developed for the process, or where the RWL varied drastically
between sampling periods or between manufacturers for no apparent reason,
In such cases it was necessary to use either the single value or the
arithmetic mean as the RWL assigned to the process.
NOTICE: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-338-
-------
DRAFT
There were also instances where the data obtained were simply not considered
as representative of the process. This occurred when aqueous waste streams
existing in the vapor phase were vented to the atmosphere, or where the
analytical results were unrea1istically low for no explicable reason.
In some of the batch or semi-batch manufacturing operations associated
with major process Subcategory D, it was not possible to develop individual
RWL's for each of the hundreds of specific batch processes in operation.
For this reason, commodities such as dyes and dye intermediates, fatty
acids, primary derivitives of fatty acids, plasticizers, and pigments were
considered as groups of materials whose manufacturing operations must be
approached in terms of their aggregate production.
The RWL values assigned to each of the 5^ secondary organic product/process
groupings were then placed in the major process subcategories described
in Section IV (Industry Categorization). Consideration of differences in
the feedstocks, unit operations, and chemical conversions is implicit in
this method of subcategorization as discussed in Section IV.
Tables 9-1 through S-k list the product/process groupings within each
major process subcategory along with the RWL values commensurate with
BPCTCA for each. The type of treatment and method of ultimate disposal
of the process wastes are also indicated for each of the product/process
groupings. It should be understood that an end-of-process treatment plant
would normally not accept many of the extremely high concentrations shown
because of processes existing as part of a multi-process facility. It
should be noted that the wastes from 6 of the processes are disposed of
by deep-well injection and 23 discharge to municipal treatment systems.
Examination of the RWL data presented in Tables 9-1 through 9-^ indicate
large variations in the flows and loadings exhibited by the different
product/process groupings within each major process subcategory. The
brief tabulation provided in Table 9-5 provides orientation as to the
magnitude of these variations.
Although there is a general increasing trend when comparing the major
process Subcategories A, B, C, and D, the variation within any one sub-
category was considered too large for the practical and equitable appli-
cation of a single set of values such as the mean for each major sub-
category. It should be noted that this variation exists even aften ten
of the product/process groups were screened out because of lack of adequate
waste characterization data. These products are listed as follows:
NOTICE: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-339-
-------
DRAFT
o
o
o
O
en
-* O
-iT O
^ o
en
.* O
oo
L/\
o
o
O
O-
OQ
in
ID
o
-
>- Q.
.* O
< Q-
. <4-
ID O
4J
in c
>~ O
i_ ._
O 4J
ID
C 4-1 in
O Q)
.- .- c
W U- flj
ID
N TJ >
I. ID
0) -0
OCX
in -
x
Q.
-------
DRAFT
cn
E
O
o
LT\
O
o
-a-
CM
O
-3^
CO
o
o
J-
CM
O
r^
o
O
J-
o
Q-
00
O
O
^ o
cn oo
cn ~-
--. LTN
cn ro
O
Cr
00
O
un o
o- o
LTV
o
o
o
o
cr
cr
o
j-
o
o
CO
cr
CO
4->
in
3
ro
OL
^ o o
Cn i^ --
E CT o
cn cr
-^ CM VD
0)
in
(/i
0)
o
o
o
Q.
ID
CL
00
a)
T3
oo
o
cn
-if
O
00
o
o
o
a)
o
o
L.
a.
JC
o
£
o-
CL
CO
O
01
O
ro
ro
O-
0)
4-J
ID
cu
o
ra
^
^_
(T3
4-J
(D
O
!_
<
-C
4-J
.
2
co
c
co
3
0
h-
co
o
c
0
4_t
0)
^
.
1_
O
_c
(_3
. i-
^ 4-J
O
in
>_
in
^
,_
O
c
o
[= !
E .0
^4-
o
c
o
4_)
ID
T3
X
O
E
ID
D-
O
-C
4-t CO
0) C
1C E
X ro
0)
-C U
4_i
a i-
- s
(U "O
M O
O
-C
o
o
CD ^
cn -t-j
O =1
1- D3
T3 I
>~ .
-C O
a) cu
a)
^ c:
-C O
-------
DRAFT
^"^
0)
c
c
o
0
CM
cr
UJ
CO
^
1-
, .
in
U
in
U
u
O
l_
o.
(U
in
ra
.c
0_
1
1.
0
a.
ra
^>
CO
L
O
cn
11
ro
o
_a
CO
in
in
a)
o
o
L.
Q_
O
""1
ro
3:
o
Q_
CO
in
ro
73
(0
L.
^^-
CM
J-
O
^ ^
LA
CM
CO
CM
O
J-
cr
LA
o"
VO
f ,
1
O
* ^
,
o
^^
CM
0
o
o
oo
:§
O-
01
01
a
o
c
c
ID
L
1C
L.
ID
Q_
ID
D.
°!
1
a
c
i_
o
o
14_
O
c
o
4~l
It
C
L_
O
_c
o
CO
c
o
_Q
V.
ra
o
o
i_
"5.
"T~
73
01
11
C
0)
ra
x:
^
Q.
ra
2
14^
0
c
o
4-t
ro
73
,~
X
o
01
c
01
"J.
X
o
14-
0
c
o
4-J
ID
73
._
X
o
(1)
T3
I.
0
o
in -C
01 U
(-) X
o
O in
C L.
0 0
-C
4J O-
ro in
in o
C -C
01 Q-
T3
C 73
O C
o ra
0)
N
l- (U
o c
-C
o >
I- .C
. V
co ra
01 -C
!- D.
U CO
O
L- JC.
I- O-
cr
0)
T)
ra
c
cn
c
O
3
"o
0-
-------
DRAFT
CJ
o
1
u
1
0
Q-
CQ
^-^
VI
T3
ID o
o o
-1 0
D
in
ID
S
S
ID
or
in
i/i
4)
U
O L
V- O
0- 0
CD
^
O
u_
I
^
en
^E_
en
V
V
en
_*
__ ^
^.
en
E
Ol
_*
^
^
cn
_^
^^
-J
^
en
E
en
Jsi
A _*:
-^
Cn
_y
01
_^i
V
\
_J
^~v
vjO
CN
J-
PA
O
O
,^-^
o-
r^
CN
O
o
,
CO
'
o
IN
VD
"~
f ,
O
o
CN
cr
LA
CO
oo
CO
o
o
o
LA
^-^
\~~
CN
-£_
PA
CN
^-^
vjO
o
vjO
^-^
CO
CVJ
OO
f ^
LT\
UT\
* -
>
r^.
J-
o
o
cr
i
o
cr
CN
OO
^
CT"
CM
v^D
1
O
*
cr
_'
o
o
o
-
-J-
o
^_
r-.
.
CNJ
O
o
o
CN
o
ro
-«-^
CM
ro
v£)
O
O
O
CM
J-
* -
r--
cr
CM
O
cr
o
CM
x-^-
o
V -
J"
,
--v
-3-
ro
V '
LA
^r
, ^
CO
r^-
LA
O
*_
O
0
o
J-
o
o
CO ^~~
^o
r CN
- CO
o ~~
^ o
r^.
^^
o ^~~
o o
cr o
CO
r--
PA rA
* ' * '
cr Ln
CM *"
CM
O
o
co -^
o
(^ 0
. ro
s - * '
rA CT
O
1^
co cr
0 S
J- O
0 CA
MO 0
4)
a:
i
cr
O
cr,
4)
(Uj
4-i'
cnj
4)
U
O
§1
Ul
l_
4)
4-J
Ul
UJ
o
4-1
41
U
C
ID
D.
cn
as
~j
o
ti
>,
-C
UJ
14_
O
c
o
JO
13
(J
l4_
._
1_
0)
4-1
t/)
LU
I
4)
u
<
,-_
^.
_c
4-»
UJ
~0
.
o
o
4~>
HI
o
f~
J_l
._
2
,
O
u
*
4)
ID
C
4)
cn
ID
M-
O
Q-
Q-
o
Q_
M-
o
C
O
4-J
CD
U
U-
.
1_
UJ
t/1
UJ
1
OJ
u
,
>^
a
O
CL.
.
o
<
o
M
0)
u
<
_c
4_t
,_
3
r
o
o
<-
4)
4-f
ID
4-J
C
O
+J
U
41
C
4)
S
D_
4)
41
O
4)
C
a)
*-
>-
a.
0
i_
o_
14-
o
c
c
4-J
ID
TD
,
X
o
E
E
^
o
Evaporation
^~
o
c
o
4-*
o
13
-D
4)
or
O
4-J
>^
,_
ID
4-J
ro
o
,_
o
c
ID
N
,
ID
L
4-J ~O
_
E
ID
,
^
X
o
u
T3
f
ID
C
O
C
ro
X
4)
-C
O 4)
E
O
>~ X
0 0
Municipal Treatment
o
c
4)
_c
O.
L^-
o
c
o
>
M
"3
>^
_C
4-J
Q)
1
C
>^
on
0
t/i
0)
I_
<_)
o
4-J
0)
_c
4-J
ace
Fo
>.
"o
TO -
ID
CL
O
E
E
<
C
00
I It
-C 4J
4J 01
41 I-
E
10 4)
X C
41 4)
Wa
ge
(D
-C
-------
DRAFT
^
o
I-
cj
D_
CO
* -
01
13
ID
O
41
01
01
*-~- t 4)
01 CJ O
E 0. O
4) DO L.
40 CL
01 01
X ID
CO
13
§ £
CJ
^ u
o
o
1
o
o
o
5
01 *~*
E CM
*^ ' (V"\
O^ vD
V '« *
.* cr
*s^
cn cxi
v
,-,
"^ O
cn J-
£ *"~~
V_^- v.
cn ro
nv *»
V t~
v^
cn J-
J^ ^D
I
O^ '"^
£ OO
cn vD
LA J^ VO
O en ro
CO
2
Q
,_
U.
1 1
o
o o o
-d- -d- o
LA vo o
« »
CN m r-.
j-
^
o
_^-
c
oo
o
ro
oo
o
o
o
,
^o
LA
* -
O
CT*
LA
_'
, ,
O
O
CM
CO
0
cr
0
0
CM
O
,
i
* "
O
o
0
^^
o
,
v_--
LA
LA
0
CD
f ^
f^
vO
O
O
O
O
LA
»
LA
^
1^^
LA
V '
p^
CM
CM
,^^
o
LA
CM
r
CM
>
CO
-d-
f ,
vO
j-
LA
r
(V-,
0
0
LA
("**>
^D
, ,
O
(V-\
J-
%_
j-
co
^ ^
O
f^.
CM
(*~\
*^~-
o
4)
o-
<
ID
U
J3
2
CL
L.
o
a
g
O
O
u
CL
O
O-
JZ
o
1C
UJ
13
O
£|
tylene Extraction from a Mixture Part of Large Plant
of C(i Hydrocarbons
3
oZ)
O
Ol
opanol Hydrolysis of Propylene Municipal Treatment
L.
a
o
in
ic Solvent Numerous Oxidation Pro- Activated Sludge Treatment
ex cesses
c
ra fo-
oi F.
t- O
o o
c Acid Nitric Acid Oxidation Municipal Treatment
of Carbohydrates
._
^
ID
X
0
ery- Aldehyde Condensation Act. Sldg. Trt (Part of
ol Large Plant)
ID 40
40 .
C 1-
4) JZ
lene Hydrolyses of Pro- Discharge to Surface Water:
1 pylene Oxide
>-. o
Q. 0
0 >-
1- _
CL. o
lene Oxide Chlorohydrin Process Discharge to Surface Water:
^
CL
O
L.
CL.
arin Synthesis from Phthalic Municipal Treatment
Anhydride Derivatives
JZ
o
0
ID
CO
Butyl Sulfonation and Hydroly- Part of Large Plant
ol sis of Mixed Butylenes
-C
0
o o
41
CO <
41
41
13
3
O
-------
DRAFT
<
O
1-
o
D_
CQ
-
m w
0> 03
(/)
in T3
0) 0)
O l~
o a)
"<
o
h-
0
a.
CD
**-*
c/l
X)
03
0
_J
0)
4-J
t/)
03
^5
s
fU
CtT
U1
Ul
01
o
o
L
Q-
0
O
-
g
o
Ll
a
o
CD
>
6
U_
^ ,
_J
\.
C71
^
cn
JKf
\
cn
^
1
^
Cn
E
* '
cn
J^
^
^
^
-j
\
cn
E
> -
cn
_*
J
n* ^^
cn
-^
^r
cn
V
5
\
_1
X-,
r-.
0
is\
cv
&
C^i
s^.
O
CO
CO
r
' '
I^>
U^
VO
, s
CO
00
\D
CO
Cvl
ro
O
O
O
!-*-
r^
j-
, s
o
ro
r--
ro
r-^
h^
ro
O
0
O
-
,
1 "
.-'
r_
f -
O
LT\
r^-
LTS
,
CO
LA
O
O
o
,- ^
o
^0
0^
,__
-
LA
r-^
r-v.
,
o
CM
CM
un
o
\D
o
CM
x~-^.
O
o
« ,
r-
cr
ro
O
O
o
u^
O"1
ro
^ ^
0
UA
ro
» '
CO
CT>
O
J-
ro
>
vO
r^.
ro
f V
vO
J-
vD
.
CO
o
o
o
CO
CM
o
CM
ro
^
.
s -
r^
J-
co
o
^D
ro
'
-J^
CT
r~-
CM
s-~~.
O
,
CO
CM
CO
o
o
~J-
vO
s-^
O
CM
LA
ro
O
ro
ro
,
O
O
0
o
v
o
J-
cr
f ,
o
LA
J-
CN
O
ro
CM
O
l-v.
ro
C>
e ,
O
\£>
ro
,
-
,
cr-
s~~.
O
-J-
CM
J-
J-
co
CM
f ,
O
vO
CM
,
r--
-f
CO
o
o
o
r-^
vO
( ,
o
CM
vO
CM
-d~
o
1
o
LA
r^
CO
cr
CM
" -
O
LA
LA
"
ro
s '
'
J~
O
O
r^
CT-
ro
-3-
J
CT-
U-l
_J
CO
<
H
CD
0
>,
L,
O
cr
fT
4->
ro
O
_Q
D
in
1/1
cr-
c
CL
3
O
1_
O
CJ1
(Ti
03
O
O
1_
Q_
"s^
4~>
!
£!
O
CJ V
-O
O
o
c
O c
0) OJ
>- JJ
Q ro
fO E
-o
o
-D CD
- >
U
>~ >
4_J .
P V-
CD I)
C H-
o o
ro O
in
C J-1
(U ro
-O N
O ^-
o o
to o
.
O JZ
C 4J
O H)
o
>
ro 13
u
- o
in
in to
=3 U
o
0) E
C UJ
to -c
0 o
tfl 4->
ro
c
-------
r---
DRAFT
o
i
I!
O
O
._
c
o
4_
o
l/l
tft
>,
,
?
o
p
E
CD
C
,
o
c
o
E
E
<
(D
C
(1)
N
C
OJ
_Q
O
1_
O
_c
o
Q-
i: L.
O 3
*J E
03 ro U
N CTl-O
ro c
._ o
0) H
X 0) C.
I
O
-------
DRAFT
000
LA O O
vo o o
. o o
-t CM
CO un *
co r
o o o
o o
CTi O O
o cvj oo
ro 00
oo
» c^j o
LTV m
oo
T3
0)
0)
-a
oo
o
O O
O O
r^ O
o
CM
O CM
O ro i
QJ
-o
c
c
J-
c>~i
cr\ rr\ fi
O J- -d-
o
o
o o o
o LT\ Ln
O CM CM
000
o r-^ r^-
o o o
o o o
o o o
NX
a)
_c
o
_c
L
O
CT:
C
1- l/l
O cr
en c
a)
W Cl-
IO ^
cj o
I L
J3 O
D
oo tn
i/i
un o
un
-------
DRAFT
Product
Chlorobenzene
Phthalic Anhydride
Ethyl Acetate
Propyl Acetate
Propylene Glycol
Solvent Complex
Methyl Salicylate
Batch Chemicals Complex
Vani11i n
Pentachlorophenol
Process
Chlorination of Benzene
Oxidation of Naphthalene
Ester ification of Ethanol
Esterification of Isopropanol
Hydrolysis of Propylene Oxide
Numerous Oxidation Processes
Synthesis with Salicylic Acid
Numerous Batch Processes
Synthesis from Sulfite Liquor
Chlorination of Phenol
Effluent limitations guidelines for these operations should be developed
on an individual basis after more data has been obtained.
The development of separate effluent limitations guidelines based on the
RWL for each of the remaining kk processes was considered impractical be-
cause of the kk separate sets of guidelines this would entail.
Therefore, each major subcategory was divided into additional subcategories.
Product/process groups were assigned to each of these new subcategories
based upon the magnitude of their RWL. Tables 9-6 through 9-9 show the
division of the major process categories and list each of the product/process
groups within the new subcategories. It should be noted that this tech-
nique is an extension of that utilized in Phase 1 of this study dealing
with the Major Organic Products (EPA ^0/1-73/009). Because of this, the
new subcategories relating to the Secondary Organic Products have been
numbered sequentially to follow those developed previously.
The mean RWL for each of the new subcategories was used as the baseline
in the subsequent development of effluent limitation guidelines for that
subcategory. These mean values are summarized in Table 9-10. Separate
effluent limitation guidelines were developed for each of the 13 sub-
categories through the application of waste reduction to the mean values
shown in Table 9-10.
It should be noted that, even with the use of these 13 subcategories,
there is still some spread above and below the mean for each. Obviously
those product/process groupings which have RWL's above the mean for their
subcategory must reduce process waste discharges to a greater extent than
required by the application of reduction factors to the mean. However,
the lowest observed RWL values for many of these processes are significantly
below the mean for their subcategories. Examination of the process de-
scriptions provided in Section IV indicates that the lowest observed RWL
value for 36 of the kk product/process groupings are less than or equal
to the BPCTCA mean RWL for their respective subcategories. Some of these
reduced RWL values are directly related to specific in-process controls
or modifications. Others represent the lowest value obtained during
several sampling periods.
NOTICE: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-3*8-
-------
DRAFT
Cn
-* O
.* O
^ o
en
.* o
V
O
cn
OJ
-M
ro
o
i
J3
3
-------
DRAFT
o oo o
r-. -d- CM i~-
CA CM CM O
o
> LACO
o o o o o l
-3- CTN CM CM
J-
O r"l
j- CM
LA
j- o cr
-3- O *-
t.
o
cr
LA -4" I|v£>
CM ^1
00
o
rA r- O
vO CA CM
O O VO O J- O
CM O CM O CPi O
OO O LA-4- LA O
CM LA OO
CM CM
o o o o
1^ -3- O O
i^ oo - o
o o o o
o ro IA'CTN
r^ -* o - >- O
C C C ~
CO 0) ~
JC D E
4-> ID _
» O «-
K 4J >- O
.C -C
C D-
5 -
o
tn
O c ~- 0)
i_ u a) _c ro i-
o o .c o .c o
O- 1- 4J
j: £ v j: i-
O O O O- Q- 1|
J-
CD
>
t.
O
cn
V
4-1
ID
O
1
_Q
3
CO
.c
4-1
IU
z
^
! ... i.
C T3 ^
O X >-
Q.-C ^
C 4-1
-O < LU
<
*-- U
>-
< 0) ^
a 4J
a: 10 a)
i: s: s:
-350-
-------
DRAFT
CM I I
r^ CM o
- CMlcM
CO
i
CT-
cr
OO O v
,-CM
CO (^ O
CO * in
-a -D
o o
o o
E
1- ID
O X
U- O
LA
|
0
>,
L.
O
cr
D
10
o
1
D
CO
(U
4-J
2
.
X
0)
C
0
4_»
^
-O
o
t/1
D
^
X
o
1)
c
^
3
00
0
0)
CO
1
CJ
>.
o
cr
O
0]
o
1
_Q
13
,
o
c
a)
CL
o
c
E
1
Q-
4-1
0)
4^
C
>~
CO
"o
1/1
a)
L.
O
O
4-J
L,
jr
4-t
L.
0)
to
4-1
C
Q_
C
10
-C
o
o
ID
CO
-351-
-------
DRAFT
O^ CSJJLA
Q CM vO
-o oglCN
.CO -CT
, cnco r*~\
CM LA -CT f\ '
J~ r-^ o -- cn
LA OV
O 1^
I
o-> ko
i CM CTiko
CO CO
. o co CM -3- o o
N vo G^ CNJ co r^- 1 o
O CNI _j- csl O I
-
I r*>
H '
I
J3
0)
O
O
cn
.Y
^
o-
CM * CM
co r-^. LT\ j- m'o
CM cr -3- co co in
O O O O]O
o o o r^ o
i O J- r-"\ LA
O O
LT\ UT\
O CO \
i CM
0 0
o o
o o
i~- LP\
0
0
r-^
CT
0
o
o
CO
o
o
o
5
0
<-)
o
o
0
0
0
ID ID
O O
oj
si
rol
f!
in 01
a) c
T3
C
ID
O O
O < <
ID ro
O U- U_
°!
= z
ID <
O1
c
o
>-
L
O
s
4_il in
re 03
V s:
-§
a
(U T3
E
-d-
i
o
-
CJ O
<
o
htheni
ex
ID
^
4-1
C
a)
E
cn
.
D_
E
*J
E
13
-O
0
TD
o
e£
O
C
c
ID
|,
-352-
-------
DRAFT
Table 9-10
Summary of Mean RWL for Each Secondary
Sub-Category
A-2
B-3
B-4
B-5
C-3
c-4
C-5
C-6
C-7
D-l
D-2
D-3
D-4
Organ! c
Flow
(lit./kkg)
44.3
2335.
6430.
2390.
2225.
285,000.
38,400.
5,130.
13,500.
154,000.
13,500.
650.
274,000.
Product Sub-Category
BOD
(kg/kkg)
0.010
0.373
16.6
120.
0.995
1.18
22.4
71.7
457.
9.72
29.3
53.9
248.
COD
(kg/kkg)
0.025
2.99
65.1
219.
4.64
4.43
95.7
204.
1180.
92.0
67.6
82.6
1300.
TOC
(kg/kkg)
0.007
0.498
19.7
73.3
1.32
1.36
23.0
74.5
472.
26.5
22.2
33.4
316.
-353-
-------
DRAFT
Although the effluent limitations guidelines for BPCTCA may be obtained
by whatever combination of in-process and end-of-process means is best.
suited to the individual manufacturers, the numerical values for the
guidelines were calculated through the application of waste reduction
factors based upon the use of a biological treatment system. The waste
reduction factors used for calculating the BPCTCA effluent limitations
guidelines for the BOD and COD parameters are listed as follows:
BOD 92 percent reduction (Effluent is 8 percent of RWL.)
COD 69 percent reduction (Effluent is 31 percent of RWL.)
These factors are based upon the performance of biological treatment
systems described in Section VII - Control and Treatment Technologies.
The performance data used to develop these factors, in many cases,
were obtained over a full year's operation and are indicative of treat-
ment systems which accept a wide variety of wastes. For this reason,
these same reduction factors were applied to the BPCTCA mean RWL for
each subcategory.
It should be noted that BOD should be used as the controlling parameter
for BPCTCA effluent limitation guidelines. The values specified for
COD should only be used where it can be demonstrated through the use of
biological treatability studies that the waste (even after dilution with
other wastes from the plant) can not be effectively reduced in
-------
DRAFT
The remaining two parameters for which BPCTCA effluent limitations
guidelines are specified are total suspended solids (TSS) and pH.
It is not possible to specify production-based limitations for these
parameters. Concentration guidelines have been established as follows:
TSS 65 mg/L
PH 6-9
The basic effluent limitation guidelines for BPCTCA are presented in
Table 9-11. BPCTCA effluent values for the BOD and COD parameters were
obtained in the manner indicated previously. It should be noted and
clearly understood that these values shown as the BPCTCA effluent should
not be directly applied as effluent limitations guidelines. They must
be first multiplied by adjustment factors based on treatment plant per-
formance developed in Section XIII. The last two columns in Table 9-11
indicate the actual effluent limitations guidelines for BPCTCA which
have been adjusted for:
1. The maximum average of daily values for any period of 30
consecutive days.
2. The maximum value for any one day.
The total effluent limitation for a multi-process plant would be the sum-
mation of these values relating to individual processes. Concentration
guidelines for TSS and pH relate to the entire facility and should be
applied in the manner shown in Table 9-11.
NOTICE: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-355-
-------
DRAFT
TABLE 9-11
Effluent Limitations Guidelines for the Secondary Organic Segment of the Organic Chemicals
Point Source Category Coimensurate with Best Practicable Control Technology Currently Available (BPCTCA)
Sub-Cateqory
A-2
BOD
COD
8-3
BOD
COD
8-4
BOD
COD
B-5
BOD
COD
C-3
BOD
COD
C-4
BOD
COD
C-5
BOD
COD
C-6
BOD
COD
C-7
BOD
COD
0-1
BOD
COD
D-2
BOD
COD
D-3
BOD
COD
D-4
BOD
COD
BPCTCA Average
BPCTCA Maximum
Average
BPCTCA
RWL
(kg/kkg)
0.010
0.025
0.373
2.99
16.6
65.1
120.
219.
0.995
4.64
1 18
4.43
22.4
95.7
71.7
204.
457.
1180.
9.72
92.0
29.3
67.6
53.9
82.6
248
1300.
Monthly Effluent Limitations
Reduction
Factor
N.S.
69%
N.S.
69%
92%
69%
927,
697,
927
69%
N.S.
N.S.
927
697
927
697
927
69%
N.S.
697,
92%
69%
927
697
927
697
Guidel ines for TSS
Average
BPCTCA
Effluent
(kg/kkg)
0.0008922
0.00388
0.04673
0.927
1.33
20.2
9.60
67.9
0.0796
1.44
1 18
4.435
1.79
29.7
5.74
63.2
36 6
365
3.086
28.5
2.30
21 .0
4.30
25.6
19.8
403.
Daily Effluent Limitations Guidelines for pH
Maximum Average of Daily
Values for Any Period of Maximum Value
Thirty Consecutive Days For Any One Day
(kg/kkg)
0.00178
0.00776
0.0934
1 .94
2.66
40.4
19.2
135.8
0.159
2.88
2.36
8.86
3 58
59.4
11.5
126.
73.2
730.
6.16
57.0
4.60
42.0
8.60
51.2
39.6
806.
65mg/L
6.0-9.0
(kg/kkg)
0.00401
0.0132
0.210
3.15
5.99
68.7
43.2
231.
0.358
4.90
5.31
15 1
8.06
101 .
25 8
215.
165.
1240.
13.9
96 9
10.4
71.4
19.4
87.0
89.1
1370.
N.S. indicates Value Not Specified
Individual Product/Process Groupings assigned to each sub-category are indicated in Tables 9-6,7,8,9
Value obtained by multiplying 20 mg/CBOD by mean flow of 44.3 L/kkg for Sub-category A-2
Value obtained by multiplying 20 mg/
-------
DRAFT
SECTION X
BEST AVAILABLE TECHNOLOGY ECONOMICALLY
ACHIEVABLE (BATEA)
Best available technology economically achievable (BATEA) for the Secon-
dary Products segment of the organic chemicals point source category is
based upon the most exemplary combination of in-process and end-of-process
treatment and control technologies.
In-process practices include those mentioned previously for BPCTCA as
well as the following:
1. The reuse of aqueous waste streams from one process in another
so that the discharge is eliminated.
2. The recycle of waste streams from one unit operation within a
process to another with the subsequent recovery of a product
or co-product.
3. The concentration and disposal of wastewaters by means which
eliminate the discharge entirely.
Process modifications for BATEA go beyond those described for BPCTCA in
that they would require changes to major unit operations or chemical con-
versions within the process. Modifications for segregation of the sewer
and drainage system for a process consistent with BPCTCA might be expected
to amount to 1 to 4 percent of the battery limits capital cost. Recycle
of aqueous waste streams for product recovery might involve replacement
of existing distillation columns or reactors which could amount to more
than 10 percent of the battery limits capital cost.
Spe: if"c examples of practices consistent with BATEA include: the reuse
of -queous hydrochloric acid streams in the manufacture of different
ch I f'-; nated methanes; the recycle of aqueous waste streams for product
recovery in the manufacture of hexamethylene diamine, maleic anhydride,
and methyl ethyl ketone; and the use of evaporators or incinerators to
completely eliminate discharges in the manufacture of phthalic anhydride
iru p-amino phenol.
Unfortunately, the wide diversity of the organic chemicals industry pre-
vents prescribing a concise list of process modifications which are
applicable to the industry as a whole (or even to the small fraction of
its products covered in both phases of ,this study). This problem is
aggr
-------
DRAFT
position. However, it makes the development of effluent limitation guide-
lines based solely upon the application of in-process technologies im-
possible. Therefore, although the use of in-process techniques may repre-
sent a viable alternative for specific manufacturers, general effluent
limitations guidelines for BATEA have been developed based upon the appli-
cation of additional end-of-process treatment technologies.
Treatment commensurate with BATEA requires the application of activated-
carbon adsorption or the use of additional biological systems in series
with the treatment previously described for BPCTCA. The specific choice
of which system should be utilized depends upon the specific process or
group of processes in operation at any given facility.
The performance of these treatment systems has been discussed in Section
VII - Control and Treatment Technologies. The incremental waste reduc-
tions associated with these technologies is indicated as follows for the
BOD and COD parameters:
BOD 90 percent reduction (BATEA effluent is 10 percent of BPCTCA
effluent)
COD 69 percent reduction (BATEA effluent is 31 percent of BPCTCA
effluent)
Effluent limitations guidelines for BATEA were calculated by applying
these reduction factors to average effluent for BPCTCA shown in Table 9-11.
There are specific subcategories where the direct use of these reduction
factors will still result in effluent concentrations which are below
the capabilities of the control systems considered as BATEA. In the case
of Subcategories A-2, B-3, and D-1, effluent limitations guidelines for
BATEA were obtained by applying minimum concentrations of 10 mg/L BOD
and 50 mg/L COD to the mean wastewater flow for the subcategory. The
mean RWL was again used for Subcategory C-^ since concentrations in the
raw waste are below 10 mg/L.
It should also be noted that the BATEA requires suspended solids removal
to an average concentration of 15 mg/L through the use of filtration.
This concentration limitation should again be applied to the total effluent
from any multi-process facility.
The effluent limitation guidelines for BATEA are presented in Table 10-1.
Again it must be understood that the BOD and COD values specified as the
average effluent for BATEA should not be directly applied before adjust-
ment for variations in treatment plant performance. The factors used here
are the same as for BPCTCA and are discussed in Section XIII - Variation
in Treatment Plant Performance.
NOTICE: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-358-
-------
TABLE 10-1
Sub-Category
A -2
BOD
COO
B-3
BOD
COD
B-f*
BOD
COO
B-5
BOD
COD
C-3
BOO
COD
C-l*
BOD
COD
C-5
BOD
COD
C-6
BOD
COO
C-7
BOD
COD
D-1
BOD
COO
D-2
BOD
COD
D-3
BOD
COD
D-1*
BOO
COD
BATEA Average
SATEA Maximum
Effluent Limitations Guidelines for the Secondary
Chemicals Point Source Category Commensurate with Best Available
Average Average
BPCTCA Reduction BATEA
Effluent Factor Effluent
(kg/kkq)
000892
00388
OU67
927
1 33
20.2
9 60
67.9
0796
t.M*
1 18
1* 1*3
1.79
29.7
5 7k
63.2
36.6
365.
3 08
28 5
2.30
21.0
1* 30
25 6
19 8
U03
N.S.
697
N S.
697
90/
69V
907
697
907
697
N.S.
N.S.
907
697
907
697
907
69%
N.S.
697,
907,
697,
90%
6
-------
DRAFT
SECTION XI
NEW SOURCE PERFORMANCE STANDARDS
Determination of the best available demonstrated control technology (BADCT)
for new process plants involves the evaluation of the most exemplary water
pollution control measures. As was the case with BPCTCA and BATEA, it is
not possible to prescribe effluent limitation guidelines based upon the
direct application of a uniform set of in-process modifications. Instead,
an end-of-process treatment model utilizing a biological system followed
by additional removal of BOD, COD and TSS by filtration was utilized.
Reductions in the BOD and COD parameters were obtained through laboratory
tests of the effluents from activated sludge treatment systems sampled
during the study. The results of these tests are described in Section VII,
Control and Treatment Technologies. These reductions were applied to the
effluent obtained from BPCTCA and are listed as follows:
BOD 17% reduction (BADCT eff. is 83% of BPCTCA eff.)
COD 20% reduction (BADCT eff. is 80% of BPCTCA eff.)
As with BATEA, the suspended solids limitation is indicated as a concen-
tration value of 10 mg/L to be applied to the effluent from an entire
faci!i ty.
Tabie 11-1 indicates the BADCT effluent limitation guidelines for the
kk product/process groupings according to the 13 subcategories established
for secondary organic products. As with BPCTCA and BATEA, the values
shown for average BADCT effluent should not be directly applied until
they are adjusted for variation in treatment plant performance.
NOTICE: THESE ARE TENTATIVE RECOMMENDATIONS BASED UPON INFORMATION
IN THIS REPORT AND ARE SUBJECT TO CHANGE BASED UPON COMMENTS RECEIVED
AND FURTHER INTERNAL REVIEW BY EPA.
-361-
-------
DRAFT
i < M-
< CD LUI'
O "O
o O
O O
-d- co
r CI-\
'*D
^ ^_1
4- o
LPv
"'.6
p^ O
r~. CD
^- CM
O
O
cn
1_
o
(U C
cn < QJ
03 O D
I_ I-
0) O M-,
> D- M-|
< CD
-K- 00 CO
*^, O t>~\
cn o o
^ o o
^
0)1
o o
o o
CO CJ
o o
o o
O O
CO O
o cn
o o
CQ O
a a
o o
CO CJ)
a a
o o
CO CJ
o a
o o
-362-
-------
VD Cn
-JD
DRAFT
a
CD
a)
4-J
(D
M- i_ a)
o a) >
Q-
cn >-
1- < I
^U-
c[cr
J-
O M
-a-
UNOO
1-1
°l
h-
C
~D
CD
TO
.!£
c
CO vO CO
o~\ o~i r-. CN j-
5
f\
CTN CO LA LA -4-
vO ro O *-ii CN
CN CN
0)
<
-JO >
\ . L.
cocr\ o
E i cn
O CD
LA . 4-»
o
i
_Q
to
_C
0
LL)
O
~O
-
m
o
CD
i;
1
O
cn
- 0
CD 1
Q 3
O CO
CD
^f L.
^ O
1
0)
fD
CJ>
1
JD
3
O
Lu
cr<
^
_J
0
O
o
"
LA
i
u-
o
"f 5
0 0
i_
O C
^
1 CQ
3 Q
CO O
CD
L. ^y
_i c
c ui -
QJ
OO OO OO OO OO <
COO CQO CClO CDCJJ COO
h-
0
r~~ ' cvj ro J" o
1 [ 1 1 I <-
o o o o o m
to
E
D
ro
o
-------
DRAFT
SECTION XI I
PRETREATMENT GUIDELINES
Pollutants from specific processes within the organic chemicals industry
may interfere with, pass through inadequately treated, or otherwise be
incompatible with a pub 1icaMy-owned treatment works. The following
section examines the general wastewater characteristics of the industry
and the pretreatment unit operations which may be applicable.
A review of the wastewater 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:
Sub-Group 1 Sub-Group 2
Subcategory A Subcategory C
Subcategory B Subcategory D
The principal difference in the general characteristics of the process
wastewaters from the manufacture of chemicals in these two Sub-Groups is
rhaS: the wastewaters of Sub-Group 1 are more likely to include significant
amounts of free and emulsified oils (petroleum origin), whereas the waste-
waters of Sub-Group 2 are more likely to include significant amounts of
heavy metals. Detailed analyses for specific products in the industry are
presented in Section IV.
Trie c/pes and amounts of heavy metals in the wastewater depend primarily
or the manufacturing process and on the amounts and types of catalysts lost
from the process. Most catalysts are expensive, and therefore, are recovered
f,;r reuse. Only recoverable catalysts (heavy metals), generally in small
concentrations, appear in the wastewater. The products and processes in
5;,v-Group 2 are most likely to have heavy metals in their wastewater, and
WcJ^tewal-. rs associated with dye/pigment production (Subcategory D) also may
havc hiys heavy metal concentrations due to the production of metallic
dyeb, Fatty acid wastewaters (Subcategory D) contain free and emulsified
oi ' (c-mlffia; and vegetable origin) of significance.
The manufacture of aerylonitrile (Subcategory D) produces a highly toxic
wastewater which is difficult to treat biologically. The toxicity
characteristics have been attributed to the presence of hydrogen cyanide
in excessive quantities (200 mg/L). in addition, the wastewater is
generally acidic (pH 4 to 6) and contains high concentrations of organic
carbon (TOC = 18500 mg/L). These wastewaters are generally segregated from
other process wastes and are disposed of by other means (e.g., incineration);
-365-
-------
DRAFT
they are not generally discharged to municipal collection systems. For
these reasons, the pretreatment unit operations developed in the follow-
ing section do not include the process wastewaters from the manufacture
of aeryloni trile.
Table 12-1 shows the pretreatment unit operations which may be necessary
to protect joint wastewater treatment processes.
Oil separation may be required when the oil (petroleum origin) content
of the wastewater exceeds 50 mg/L. Animal and vegetable oils in the
fatty acid wastewaters will have to be segregated in order to minimize
solids separation problems in the wastewater treatment facility.
The heavy metals present in organic chemical wastes are in many cases so
low in concentration that heavy metals removal is not required from the
standpoint of treatability characteristics. However, the effluent
limitations for toxic pollutants may require additional pretreatment
(chemical precipitation) for removal of these materials.
The pretreatment unit operations generally consist of equalization,
neutralization, and oi'l separation. In addition, phenol recovery (to
reduce the phenol concentrations) and spill protection for spent acids
and spent caustics may be required in some cases.
Biological Treatment Inhibition
The scope of this study did not allow for a specific toxicity evaluation
of individual product wastewaters. However, the completeness of the RWL
analytical data did provide a wastewater profile which could be used to
evaluate possible biological inhibition. The list and concentrations of
inhibitory pollutants in EPA's Federal Guidelines - Pretreatment of
Discharge to Publicly Owned Treatment Works were examined, and specific
comments are presented in Section VI. This previous list was amended in
the Phase I report to include phenol and iron:
Inhibition to Inhibition to
Pa rameter Biological Treatment Anaerobic Sludge Digestion
Phenol 50 mg/L
I ron 5 mg/L
-366-
-------
DRAFT
I
CNI
ro
i _
!_
O
U1 ^
C U
O 4-1
l/l
4-J D
(D "O
1- C
'i
0)
4-J -C
C O
0)
E O
4-1
ro c
0) (T
i- CT)
Growth Fixed Growth Independent Physical -
System Biological System Chemical System
T3
o>
-Q
c
co
CL
1/1
ZJ
CO
4-1
CO
E
4-J
ro
(L>
4-J
CO
1_
D_
i
rc
u
CT
0
C
CD
Q.
^
O
S_
CO
-D
33
co
c
0
4-J
ro
4J
+ +
CL
C + C + C
O O 0 O
- c co c
4-J O 4-> S- O 4-J
ro * ro Q- ro
i- 4-> N 4-J N
ro ro *~~ ro
Q. N ro N -
0) ro U ro
CO 1- l_
ro 4-1 E ro 4-J
D D CO D 3
CT a) _c cr ri)
O UJ Z <_> LU Z
"H ro
4-J
+ + C
o o-
c + c - + c
O O 4-1 U O
C O 0) C
4-1 O 4-1 CO U O 4-J
ro to 4-> O. - ro
U 4-J N O 4-J N
ro ro i- ro
Q. N CL ro N
a) ro O C ~ ro
CO 1- O L.
ro 4-J E ro 4-J
D D .- CL) 4J U3
cr co CL _c cr co
O LU 2 CO CJ LU "Z.
+ co + ro
4-J 4J
+ + c + c
O CL O Q.
C + C - + c
O O 4-J O O 4J (J
c - o co c .- o co
4-JO4-JC01 O4-JC01-
ro ro 4-J D_ . ro 4-J Q-
1- 4-J N O 4-J N O
ro ro !- ' ro >- '
CL N Q- (C N Q- ro
CO ro uc ro oc
LT> i- O i_ . o
ro 4-J E ro 4-1 E
33 CO 4-J 33 CO 4-J
CT CD CL -d CTCOCL-C
OLUZCOCJ LUZCOCJ
CM
i CM
O
u
co
ro
O
c
CO
4-J
X
CD
c
o
in
T3
c
co
Q.
CO
T3
C
o
o
CD
fO
O
co
T3
C
u
ro
ro
M-
CO
s_
O
-o
co
13
cr
CO
co
-D
ro
O
ro
ro
CL
CO
in
TO
CO
co
z c
« CSJ
-367-
-------
DRAFT
SECTION XI I I
ALLOWANCE FOR VARIABILITY IN TREATMENT PLANT PERFORMANCE
Biological treatment, as well as activated carbon wastewater treatment,
will produce an effluent of variable quality. In order to enforce the
recommended average effluent limitations, it is necessary for EPA to set
maximum allowable limitations so as to provide a procedure for evaluating
plant performance compiled from a limited data base.
Biological Wastewater Treatment
Historically, biological treatment plants have been designed to remove
some particular percent removal, e.g., 90 percent BOD, which was defined
as a yearly average. Very little emphasis was placed on the effluent
variability during winter periods. As long as plants met their required
removal efficiencies during the summer months when stream quality was
critical, the plant performance was considered adequate. Increased under-
standing of biological kinetics and advanced instrumentation have provided
design engineers with the knowledge required to substantially minimize
effluent variability.
During the survey program, daily historic performance data were provided
for only four treatment facilities, namely Plant Nos. 2, 11, 13, and 1*t.
(See Table 7-2.) Plants 2 and 11 are multiple-stage biological plants
treating Subcategory C wastewaters. Both of these treatment plants utilize
activated sludge and were designed based on the criteria presented in
Table 13-1. Plants 11 and 2 have primary settling and nutrient addition.
In Plant 11, 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
darifier. Additional organic and solids removal is accomplished by
using an aerated polishing lagoon.
Plant 2 has two parallel trains of 3 aeration basins each for a total of
6 basins. Clarification and air flotation are provided in order to re-
duce the aeration basin mixed liquor (MLSS) which average about 7,000 to
8,000 rng/L. A polishing lagoon with an atypical detention time provides
additional organic and solids removal. Plant 2 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 ^5 F. Plant 11 is located
in the southern United States and is not subject to extreme seasonal tem-
perature fluctuations. However, Plant 11 also has provision for steam
addition to maintain basin temperature.
Plant 13 provided only effluent data so that it was not possible to
effectively evaluate the plant's performance. Lastly, the performance
data from Plant No. 14 indicated it was not considered exemplary in per-
formance because the plant's annual BOD removal was only 73 percent, and
the plant had not been designed to adequately perform during colder winter
periods.
-369-
-------
DRAFT
Table 13-1
Summary of Plant Design Criteria
Description Plant No.11 Plant No.2
Flow - mgd 1.0 0.55
Primary 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/MG ^50 5^0
Final Clarifier
Overflow Rate - gpd/sq.ft. 425 150
SWD - ft. 10 10
Diameter - ft. kO kO
Flotat i on Uni t
Solids - Ibs/sq.ft./day -- 7.5
Detention Time - hours -- 2.5
Polymer Dosage - mg/L 100
Poli shi ng Pond
Detention Time - days 0.6 118
Aeration Equipment - HP/MG 10 1.5
-370-
-------
DRAFT
Twelve months of daily historic data from Plants 2 and 11 were statisti-
cally analyzed graphically and the normalized results are summarized in
Table 13-2. Only COD data was available from Plant 2, while BOD and TOC
data were available from Plant 11.
The significance of the data is that the biological treatment plant on
the average (50 percent of the time) is producing an effluent with a 8005
concentration of 20 mg/L, but will also produce an effluent with 90 mg/L
of BODr 5 percent of the time.
Variations in the performance of a treatment plant are attributable to
one or more of the following:
1. Seasonal variations in the wastewater temperature which either
accelerate or depress the biological kinetics.
2. Variations in the sampling technique or in the analytical pro-
cedures .
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 could be
reflected in performance.
5. !n~plant process bottle necking 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 it-
self in an increase in TSS and BODc in the plant effluent.
6. The effluent COD variability is also a function of the changing
ratios between refractory and readily degradable organics in
the plant's influent wastewater. This would be more of a con-
cern in Subcategory D wastes than in the existing data from Sub-
category C wastewaters.
These variations are purely a function of the treatment plant design and
performance. They will still occur even if the treatment plant has pro-
visions for equalization of variations in the influent raw waste load
which it receives.
Selected statistical data in Table 13-2 were examined to compare various
ratios of probability of occurrence to the 50 percent probability of
occurrence. This data is as follows:
-371-
-------
o
o
COll^vO CM (T\ vO fl CM O
OEvO 1^- OO CO CT> O i CMJ-
-^^LAJ-CMOCOLALAOO
ojcnvo i co cr\ cr\ o i LA
v E »- .- .- *-
l^x LAr^oocoooooo
reloiLAvo r-^co o CV>LACO
Q| E - - r~ r-
O LA
LA CO
o o
i CM
0 0
CM O
CM
DRAFT
0)
10
re
o
en -M
O c
0)
O 3
CQ M!
4-
14- LU
O
4-1
C C
o re
i- c
5 I
4-1
4-> (0
C Q)
0) 1-
CM
O
0
0
o
4-1
c
18
>J
^
E
VD CO O
* i CM
CM
CM
LA
v o
01 T-
E
^^- C3
i CM
vD
CM
VD
CM
O
f«-l
O
-d"
O
LA
U\
CM
O
J-
O
VO
O
O'V
O
\£J
0)
>> o c o
4-i C ID 4-i
.- 0) JC
jQ 3 W 3
ID u v> a
jQ O 0> W
oo- ,_
a. M- S3 o
o
o
CO
o
cr>
in
-372-
-------
DRAFT
Ratio of
Probabi1i ty
99/50
95/50
90/50
99/50
95/50
90/50
99/50
95/50
90/50
5.3
3.4
2.5
3.0
2.2
1.8
COD
TOC
Weekly
3.5
2.5
2.0
3.6
2.5
2.0
2.2
1.7
1.5
Month 1y
2.7
2.0
1.7
2.8
2.0
1.8
1.9
1.6
1.4
The daiiy 95/50 BOD ratio is 4.5, while the corresponding monthly ratio
is 2.0. This indicates that a substantial day-to-day variation witnessed
in plant performance is tempered when the variation is based on monthly
dat'i. For this reason, it is recommended that a monthly average be used
as the time basis for the effluent guidelines. In addition, a 95-percent
confidence limit should be used, since the 95/50 values should be within
a r£r,qe typically observed in the past as being reasonable when treatment
p'ao' data were analyzed statistically.
Use of a 95-percent confidence limit suggests that the plant could be in
violation no more than one month per year. Therefore, review of the in-
dui>* ial sel f-repo rt i ng questionnaire would indicate if the plant were
permanently in violation. If the stipulated effluent limit was 20 mg/L
and a plant reported 40 mg/l. on two successive months, then the plant
wou' :! be in violation.
The following effluent adjustment factors are proposed for the following
parameters and time intervals:
Monthly Effluent. Weekly Effluent Daily Effluent
Parameter Adjustment Factor Adjustment Factor Adjustment Factor
BOD 2.0 2.5 4.5
COD 2.0 2.5 3.4
95/50 ratio of confidence limits.
-373-
-------
DRAFT
Activated Carbon Wastewater Treatment
Of the six activated carbon wastewater plants surveyed, only one had
sufficient data that could be analyzed statistically. Five months of
effluent data (May through September) were obtained from Plant No. 30.
(See Table 7-5.) The results of the statistical analyses are shown
below:
Ratio of Dai 1y Basis
Probabi1ity
99/50
95/50
90/50
The limited data does not justify the recommendation of a specific daily
adjustment factor.
BOD
2.0
1.6
1.U
COD
2.1
1.5
1.3
Phenol
11*
11
6.3
-37k-
-------
DRAFT
SECTION XIV
ACKNOWLEDGMENTS
Roy F. Weston, Inc. wishes to express appreciation to the Fatty
Acid Producers' Council, Division of the Soap and Detergent Association.
This organization provided valuable assistance on the selection of
representative process plants within the industry.
Acknowledgment is made of the cooperation of personnel in many plants in
the organic chemicals industry who provided valuable assistance in the
collection of data relating to process RWL and treatment plant performance,
Special acknowledgment is made of those plant personnel and company
officers who cooperated in providing detailed plant operating data to
support this study.
Acknowledgment is made also of the assistance and direction provided
by the Project Officer, Mr. John A. Nardella, and others associated with
the Effluent Guidelines Division: Messrs. Allen Cywin, Ernest P. Hall,
Walter Hunt, and others who provided helpful suggestions and comments.
Finally, the valuable technical assistance provided by our sub-contractor,
Chem Systems, Inc., is acknowledged.
-375-
-------
DRAFT
SECTION XV
BIBLIOGRAPHY
This bibliography supplements the main bibliography previously presented
in the Phase I study.
1. Development Document for Proposed Effluent Limitations Guide!ines and
New Source Performance Standards for the Major Organic Products.
U. S. Environmental Protection Agency; EPA 440/1-73/009; December,'1973.
2. Development Document for Effluent Limitations Guidelines and Standards
of Performance - Organic Chemicals Industry - Draft. Prepared by
Roy F. Weston, Inc. for U. S. Environmental Protection Agency;
Contract No. 68-01-1509; June, 1973.
3. Hager, Donald. "A Survey of Industrial Wastewater Treatment by
Granular Activated Carbon." Presented at the 4th Joint Chemical
Engineering Conference; Vancouver, British Columbia; September 10, 1973.
4. Kennedy, D. C., and others. "A New Adsorption/1 on Exchange Process
for Treating Dye Waste Effluents." Rohm and Haas Company; Philadelphia,
Pennsylvania.
5. Pattison, E. Scott, editor. Fatty Acids and Thei_r Industrial
Applications. New York: Marcel Dekker, I nc. , 1968.
-377-
------- |