440182007
United States
EnvironmentaTPri
Agency c
Water and Waste Management
   iuideiines Division
WH-652
Washington DC 20460
Development
Document for
Effluent Limitations
Guidelines and
Standards for the

Inorganic Chemicals
Manufacturing

Point Source Category

-------

-------
           DEVELOPMENT DOCUMENT

                   for

      EFFLUENT LIMITATIONS GUIDELINES

     NEW SOURCE PERFORMANCE STANDARDS

                   and

          PRETREATMENT STANDARDS

                 for the

     INORGANIC CHEMICALS MANUFACTURING
          POINT SOURCE CATEGORY

             Anne M. Gorsuch
              Administrator

             Steven Schatzow
                 Director
Office of Water Regulations and Standards
                m
      Jeffery Denit,  Acting Director
       Effluent  Guidelines  Division

            G. Edward Stigall
    Chief, Inorganic Chemicals Branch

           Dr. Thomas Fielding
             Dwight Hlustick
             Elwood  E. Martin
                June 1982
       Effluent  Guidelines  Division
Office of Water Regulations and Standards
  U.S.  Environmental  Protection Agency
         Washington, D.C.  20460

-------

-------
                 TABLE OF CONTENTS
LIST OF FIGURES                                        xx

LIST OF TABLES                                     xxviii

ACKNOWLEDGEMENTS                                     xliv

CONCLUSIONS AND SUMMARY                                 1

       TOXIC POLLUTANTS                                 1
       CONTROL AND TREATMENT TECHNOLOGY                 2
       COSTS OF ADDITIONAL IN-PLAtfT TREATMENT           2
       SUBCATEGORIZATION                                2
       RESTUDY OF REMANDED REGULATIONS                  3

RECOMMENDATIONS                                         5

INTRODUCTION                                           17

       AUTHORITY                                       17
                 The Federal Water Pollution           17
                  Control Act Amendments
                 Court Remand of Regulations           19
                 The Settlement Agreement              21
       GENERAL APPROACH AND METHODOLOGY                25
                 Industry Data Base Development        26
                  and Subcategorization Review
                 The Screening and Verification        26
                  Sampling Programs
                 Engineering Evaluations               26
                 Treatment System Cost Estimates       2b
                 Treatability Study                    2/
       GENERAL CRITERIA FOR EFFLUENT LIMITATIONS       27
                 BPT, Effluent Limitations              27
                 BAT Effluent Limitations              28
                 BCT Effluent Limitations.              28
                 New Source Performance                30
                  Standards
                 Pretreatment Standards for            30
                  Existing Sources
                 Pretreatment Standards for            31
                  New Sources

-------
           TABLE OF CONTENTS - Continued
SOBCATEGORIZATION REVIEW

        BASIS FOR SUBCATEGORIZATION
                  Factors Considered
                  General Conclusions
        SECONDARY SUBCATEGORIZATION
                  Chlor-Alkali
                  Titanium Dioxide
                  Hydrogen Cyanide
        REVIEW OF POSSIBLE INTEGRATION OF
         SUBCATEGORIES
                  Hydrofluoric Acid and Aluminum
                   Fluoride
        SUMMARY

SCREENING AND VERIFICATION SAMPLING PROGRAMS

        SCOPE AND METHODOLOGY
                  Selecting Plants and Making
                   Preliminary Contacts
                  Screening and Verification
                   Sampling
                  Analytical Methodology for
                   Toxic Pollutants
                  Quality Assurance Provisions
        SUMMARY OF ANALYTICAL RESULTS

PROCESS AND WASTE TREATMENT INFORMATION
 DEVELOPMENT AND EVALUATION

        INDUSTRY DATA BASE DESCRIPTION
                  Literature Review
                  Plant Visits
                  Telephone and Direct Contact
                  308-Questionnaire Responses
        PROCESS WASTE SOURCES AND CURRENT
         TREATMENT PRACTICES
                  Data Acquisition
                  Evaluation of Data
                  Model Plant and BPT Treatment
                   System Specification
                  Dissolved Solids  in Wastewater
                   Effluents
38

39

41

41
41

42

45

50
53

59
59
59
59
59
59
60

60
60
62

63

-------
                 TABLE OF CONTENTS  -  Continued
                                                            Page

7.    ASSESSMENT OF TECHNOLOGY FOR ADVANCED  TREATMENT        65
       AND CONTROL

              INTRODUCTION                                    65
              HYDROXIDE PRECIPITATION                         66
              FERRITE COPRECIPITATION                         70
              SULFIDE PRECIPITATION                           71
              THE XANTHATE PROCESS                            74
              ION EXCHANGE                                    76
              REDUCTION PROCESSES                             2B
              OXIDATION PROCESSES                             79
              MEMBRANE PROCESSES                              82
              ADSORPTION                                      83
              FLUORIDE REMOVAL                                87
              CHLORINE REMOVAL                                88

8.    TREATABILITY ESTIMATES AND LONG-TERM DATA               91
       ANALYSIS

              THE DEVELOPMENT OF TREATABILITY ESTIMATES       91
                        Preliminary Analysis                 91
                        Final Analysis                       101
              SELECTION OF TOXIC METAL CONTROL               109
               PARAMETERS
                        Control Parameters for               109
                         Hydroxide Precipitation
                        Control Parameters for               109
                         Sulfide Precipitation
              THE USE OF HISTORICAL POLLUTANT                111
               DATA
                        Determination of Limitation         111
                         Guidelines Based Upon
                         Historical Performance
                        Assumptions Concerning Daily        112
                         Pollutant Level Measurements
                        Assumptions Concerning 30-Day       117
                         Average Pollutant Level
                         Observation

9.    TREATMENT TECHNOLOGY APPLICATIONS FOR  TOXIC            125
       POLLUTANT REMOVAL

              SELECTION OF POLLUTANTS TO BE  CONTROLLED      125
              APPLICATION OF ADVANCE LEVEL TREATMENT        125
               AND CONTROL ALTERNATIVES
                        General Design Objectives            125
                        Pretreatment Technology              128
                              111

-------
           TABLE OF CONTENTS - Continued
11.
                  New Source Performance
                   Standards
        ESTIMATED ACHIEVABLE PERFORMANCE
         CHARACTERISTICS FOR ADVANCED LEVEL
         APPLICATIONS
                  Advanced Level Removal of BPT
                   Pollutants
                  Advanced Level Removal of Toxic
                   Pollutants
        POLLUTION CONTROL PARAMETERS TO BE
         REGULATED
                  Conventional Pollutants
                  Nonconventional Pollutants
                  Toxic Pollutants

COST OF TREATMENT AND CONTROL SYSTEMS

        INTRODUCTION
                  Purpose of Cost Data
                  General Approach
                  Cost References and Rationale
                  Definition of Levels of
                   Treatment and Control Cost
                   Development
                  Treatment and Disposal
                   Rationale Applied to Cost
                   Development
                  Expression of Costs
        COST ESTIMATES FOR EACH SUBCATEGORY

CHLOR-ALKALI INDUSTRY

        INDUSTRY PROFILE
                  General Description
                  Subcategorization
        MERCURY CELL PROCESS INDUSTRY
         PROFILE
                  General Description
                  General process Description
        WATER USE AND WASTEWATER SOURCE
         CHARACTERISTICS
                  Water Use
                  Waste Sources
        DESCRIPTION OF SPECIFIC PLANTS
                  Screening Program
                  Verification
 128

129

 129

 129
130
130

133

133
133
134
134
134


135
 135
142

145

145
145
145
146

146
146
149

149
151
153
153
153

                         IV

-------
   TABLE OF CONTENTS - Continued
                                             Page
                                             —^•h.^AdM,

          Descriptions of Plants Not         158
           S ampled
          Summary of the Toxic Pollutant     163
           Data
POLLUTION ABATEMENT OPTIONS                  165
          Toxic Pollutants of Concern        165
          Prevailing Control and             166
           Treatment Practices
          Process Modifications and          166
           Technology Transfer Options
          Best Management Practices          170
          Advanced Treatment Technologies    170
SELECTION OF APPROPRIATE TECHNOLOGY AND      170
 EQUIPMENT
          Technologies for Different         170
           Treatment Levels
          Equipment for Different            171
           Treatment Levels
TREATMENT COST ESTIMATES                     172
          General Discussion                 172
          Chlorine Bearing Wastes            172
          Model Plant Treatment Costs        175
BASIS FOR REGULATIONS                        175
          Basis for BPT Limitations          1/5
          Basis for Final BAT Effluent       176
           Limitations
          Basis for BCT Effluent             192
           Limitations
          Basis for New Source               192
           Performance Standards
          Basis for Pretreatment             192
           Standards
DIAPHRAGM CELL PROCESS INDUSTRY PROFILE      192
          General Description                193
          General Process Description        193
WATER USE AND WASTE WATER SOURCES            196
          Water Use                          196
          Waste Sources                      195
DESCRIPTIONS OF SPECIFIC PLANTS              199
          Screening                          199
          Verification                       200
          Descriptions of Plants Not         200
           S ampled
          Toxic Pollutant Concentrations     202.
POLLUTION ABATEMENT OPTIONS                  218
          Toxic Pollutants of Concern        218
                 v

-------
                     TABLE  OF  CONTENTS  - Continued
12.
                  Prevailing Control and
                   Treatment Practices
                  Process Modifications and
                   Technology Transfer Options
                  Best Management Practices
                  Advanced Treatment Technologies
        SELECTION OF APPROPRIATE TECHNOLOGY AND
        EQUIPMFNT
                  Technologies for Different
                   Treatment Levels
                  Equipment for Different
                   Treatment Levels
        TREATMENT COST ESTIMATES
                  General Discussion
                  Model Plant Treatment Costs
        BASn FOR REGULATIONS
                  Basis for BPT Limitations
                  Basis for BAT Effluent
                   Limitations
                  BCT Limitations
                  Basis for New Source
                   Performance Standards
                  Basis for Pretreatment
                   Standards

HYDROFLUORIC ACID INDUSTRY

        INDUSTRY PROFILE
                  General Description
                  Subcategorization
                  General Process Description and
                   Raw Materials
        WATER USE AND WASTE SOURCE
         CHARACTERISTICS
                  Water Use
                  Waste Sources
        DESCRIPTION OF PLANTS VISITED AND SAMPLED
                  Screening
                  Verification
                  Summary of the Toxic Pollutant
                   Data
        POLLUTION ABATEMENT OPTIONS
                  Toxic Pollutants of Concern
                  Process Modifications and
                   Technology Transfer Options
                  Best Management Practices
                                                               218

                                                               223

                                                               224
                                                               224
                                                               225

                                                               225

                                                               22P

                                                               227
                                                               727
                                                               233
                                                               233
                                                               233
                                                               249

                                                               254
                                                               255

                                                               256


                                                               261
                                                               261
                                                               261
                                                               261
                                                               264
                                                               267

                                                               267
                                                               267
                                                               272
                                                               272
                                                               272
                                                               278

                                                               282
                                                               282
                                                               282

                                                               285
                                  VI

-------
                    TABLE OF CONTENTS - Continued
13,
14,
                  Prevailing Control  and
                   Treatment Practices
                  Advanced Treatment  Technologies
        SELECTION OF APPROPRIATE TECHNOLOGY AND
         EQUIPMENT
                  Technologies  for Different
                   Treatment Levels
                  Equipment for Different
                   Treatment Levels
        TREATMENT COST ESTIMATES
                  General Discussion
                  Model Plant Control Costs for
                   Existing Sources
        BASIS FOR REGULATIONS
                  Evaluation of BPT Treatment
                   Practices
                  Basis for BPT Effluent
                   Limitations
                  Basis for BCT Effluent
                   Limitations
                  Basis for BAT Effluent
                   Limitations
                  Basis for the New Source
                   Performance Standards
                  Basis for the Pretreatment
                   Standards

HYDROGEN PEROXIDE INDUSTRY

        SUMMARY OF DETERMINATIONS
        ASSESSMENT OF THE WATER POLLUTION
         POTENTIAL
                  Production Processes  and
                   Effluents
                  Plants
                  Toxic Pollutants
        STATUS OF REGULATIONS

TITANIUM DIOXIDE INDUSTRY

        TITANIUM DIOXIDE, CHLORIDE PROCESS
         INDUSTRY PROFILE
                  General Description
                  Subcategorization
                  General Process Description  and
                   Raw Materials
286
287

287

288

293
293
295

295
295

303

316

31 fi

323

323


327

327
327

327

327
329
329

331

331

331
331
331
                                 VI i

-------
   TABLE OF CONTENTS - Continued
WATER USE AND WASTE SOURCE                    335
 CHARACTERISTICS
          Water Use                           335
          Waste Sources                       335
DESCRIPTION OP PLANTS VISITED AND SAMPLED     338
          Screening                           338
          Verification                        338
          Toxic Pollutant Concentrations      340
POLLUTION ABATEMENT OPTIONS                   345
          Toxic Pollutants of Concern         345
          Process Modification and            345
           Technology Transfer Options
          Best Management Practices           345
          Prevailing Control and              347
           Treatment Practices
          Advanced Treatment Technologies     348
SELECTION OF APPROPRIATE TECHNOLOGY AND       348
 EQUIPMENT
          Technologies for Different          348
           Treatment Levels
          Equipment for Different             349
           Treatment Levels
TREATMENT COST ESTIMATES                      349
          General Discussion                  349
          Model Plant Control and             354
          Treatment Costs
BASIS FOR REGULATIONS                         354
          Evaluations of BPT Treatment        354
           Practices
          Basis for BPT Effluent              36!
           Limitations
          Basis for BCT Effluent              370
           Limitations
          Basis for BAT Effluent              371
           Limitations
          Basis for New Source                374
           Performance Standards
          Basis for Pretreatment              377
           Standards
TITANIUM DIOXIDE - SULFATE PROCESS            377
 INDUSTRY PROFILE
          General Description                 377
          General Process Description and     381
           Raw Materials
WATER USE AND WASTE SOURCE                    384
 CHARACTERISTICS
          Water Use                           384
          Waste Sources                       384
                 Vlll

-------
   TABLE OF CONTENTS - Continued
DESCRIPTION OF PLANTS
          Screening
          Verification
          Other Plant Descriptions
          Toxic Pollutant Concentrations
POT"T)TION ABATEMENT OPTIONS
          Toxic Pollutants of Concern
          Process Modifications and
           Technology Transfer Options
          Best Management Practices
          Prevailing Control and
           Treatment Practices
          Advanced Treatment Technologies
SELECTION OF APPROPRIATE TECHNOLOGY AND
 EQUIPMENT
          Technologies for Different
           Treatment Levels
          Equipment for Different
           Treatment Levels
TREATMENT COST ESTIMATES
          General Discussion
          Model Plant Control and
           Treatment Costs
BASIS FOR REGULATIONS
          Evaluation of BPT Practices
          Basis for BPT Effluent.
           Limitations
          Basis for BCT Effluent
           Limitations
          Basis for BAT Effluent
           Limitations
          Basis for New Source
           Performance Standards
          Basis for Pretreatment
           Standards
TITANIUM DIOXIDE - CHLORIDE ILMENITE
 PROCESS INDUSTRY PROFILE
          General Description
          General Process Description and
           Raw Materials
WATER USE AND WASTE SOURCE
 CHARACTERISTICS
          Water Use
          Waste Sources
DESCRIPTION OF PLANTS VISTED AND SAMPLED
          Screening
          Verification Program
          Toxic Pollutant Concentration
385
385
387
387
390
395
395
396

396
396

396
396

396

400

402
402
403

404
404
404

417

417

417

424

427

427
427

4 3D

430
430
431
431
431
431

-------
                 TABLE OF CONTENTS - Continued
15.
        POLLUTION ABATEMENT OPTIONS
                  Toxic Pollutants of Concern
                  Process Modifications and
                   Technology Transfer Options
                  Best Management Practices
                  Prevailing Control and
                   Treatment Practices
                  Advanced Treatment Technology
        SELECTION OF APPROPRIATE TECHNOLOGY AND
         EQUIPMENT
                  Technologies for Different
                   Treatment Levels
                  Equipment for Different
                   Treatment Levels
        TREATMENT COST ESTIMATES
                  General Discussion
                  Model Plant Control and
                   Treatment Costs
        BASIS FOR REGULATIONS
                  Evaluation of BPT Treatment
                   Practices
                  Basis for BPT Effluent
                   Limitation
                  Basis for BCT Effluent
                   Limitations
                  Basis for BAT Effluent
                   Limitations
                  Basis for the New Source
                   Performance Standards
                  Basis for Pretreatment
                   Standards

ALUMINUM FLUORIDE INDUSTRY

        INDUSTRY PROFILE
                  General Description
                  General Process Description and
                   Raw Materials
        WATER USE AND WASTE SOURCE
         CHARACTERISTICS
                  Water Use
                  Waste Sources
        DESCRIPTION OF PLANTS VISITED AND SAMPLED
                  Screening
                  Verification
                  Summary of the Toxic Pollutant
                   Data
                                                           Page

                                                            435
                                                            435
                                                            436

                                                            436
436
437

437

440

441
441
442

452

452

452

456

456

456

460


461

461
461
461

461

461
461
465
465
467
467
                               x

-------
                   TABLE OF CONTENTS - Continued
16.
                                                     Page

        POLLUTION ABATEMENT OPTIONS                   476
                  Toxic Pollutants of Concern         476
                  Process Modifications and           476
                   Technology Transfer Options
                  Best Management Practices           477
                  Prevailing Control and              477
                   Treatment Practices
                  Advanced Treatment Technologies     477
        SELECTION OF APPROPRIATE TECHNOLOGY AND       478
         EQUIPMENT
                  Technologies for Different          478
                   Treatment Levels
                  Equipment for Different             482
                   Treatment Levels
        TREATMENT COST ESTIMATES                      484
                  General Discussion                  484
        BASIS FOR REGULATIONS                         485
                  Evaluation of BPT Treatment         485
                   Practices
                  BPT Effluent Limitations            485
                  Basis for the BCT Effluent          495
                   Limitations
                  Basis for BAT Effluent              495
                   Limitations
                  Basis for New Source                496
                   Performance Standards
                  Basis for Pretreatment              498
                   Standards

CHROME PIGMENTS INDUSTRY                              499

        INDUSTRY PROFILE                              499
                  General Description                 499
                  Subcategorization                   499
                  General Process Description and     499
                   Raw Materials
        WATER USE AND WASTE SOURCE                    508
         CHARACTERISTICS
                  Water Use                           508
                  Waste Sources                       508
        DESCRIPTION OF PLANTS                         513
                  Screening                           513
                  Verification                        513
                  Toxic Pollutant Concentrations      515
                               XI

-------
                  TABLE OP CONTENTS - Continued
17. „
        POLLUTION ABATEMENT OPTIONS
                  Toxic Pollutants of Concern
                  Process Modifications and
                   Technology Transfer Options
                  Best Management Practices
                  Prevailing Control and
                   Treatment Practices
                  Advanced Treatment Technologies
        SELECTION OF APPROPRIATE TECHNOLOGY AND
         EQUIPMENT
                  Technologies for Different
                   Treatment Levels
                  Equipment for Treatment
        TREATMENT COST ESTIMATES
                  General Discussion
                  Model Plant Costs
        BASIS FOR REGULATIONS
                  Evaluation of BPT Treatment
                   Practices
                  Basis for BPT Effluent
                   Limitations
                  Basis for BCT Limitations
                  Basis for BAT Effluent
                   Limitations
                  Basis for New Source
                   Performance Standards
                  Basis for Pretreatment
                   Standards

HYDROGEN CYANIDE INDUSTRY

        INDUSTRY PROFILE
                  General Description
                  Subcategorization
                  General Process Description and
                   Raw Materials
        WATER USE AND WASTE SOURCE
         CHARACTERISTICS
                  Water Use
                  Waste Sources
        DESCRIPTION OF PLANTS VISITED AND SAMPLED
                  Screening
                  Verification
                  Toxic Pollutant Concentrations
                                                            Page

                                                              525
                                                              525
                                                              526

                                                             526 :
                                                              527

                                                              528
                                                              528

                                                              528

                                                              530
                                                              532
                                                              532
                                                              533
                                                              53?
                                                              B33

                                                              540

                                                              548
                                                              548

                                                              549

                                                              550
551

551
551
551
551

554

554
554
557
557
559
562
                               XII

-------
                  TABLE OF CONTENTS - Continued
18.
        POLLUTION ABATEMENT OPTIONS                   568
                  Toxic Pollutants of Concern         568
                  Process Modifications and           568
                   Technology Transfer Options
                  Best Management Practices           568
                  Prevailing Control and              568
                   Treatment Practices
                  Advanced Treatment Technologies     570
        SELECTION OF APPROPRIATE TECHNOLOGY AND       571
         EQUIPMENT
                  Technologies for Different          571
                   Treatment Levels
                  Equipment for Different             571
                   Treatment Levels
        TREATMENT COST ESTIMATES                      573
                  General Discussion                  573
        BASIS FOR REGULATIONS                         575
                  Evaluation of BPT Treatment         575
                   Practices
                  Basis for BPT Limitations           576
                  Basis for BCT Limitations           586
                  Basis for BAT Limitations           586
                  Basis for New Source                588
                   Performance Standards
                  Basis for Pretreatment              590
                   Standards

SODIUM DICHROMATE INDUSTRY                            593

        INDUSTRY PROFILE                              593
                  General Description                 593
                  Subcategorization                   593
                  General Process Description and     593
                  Raw Materials
        WATER USE AND WASTE SOURCE                    596
         CHARACTERISTICS
                  Water Use                           596
                  Waste Sources                       596
        DESCRIPTION OF PLANTS VISITED AND SAMPLED     599
                  Screening                           599
                  Verification                        599
                  Toxic Pollutant Concentrations      602
                   and Loadings
                              xxn

-------
                  TABLE OF CONTENTS - Continued
               POLLUTION ABATEMENT OPTIONS
                         Toxic Pollutants of Concern
                         Process Modifications and
                          Technology Transfer Options
                         Best Management Practices
                         Prevailing Control and
                          Treatment Practices
                         Advanced Treatment Technologies
               SELECTION OF APPROPRIATE TECHNOLOGY AND
                EQUIPMENT
                         Technology for Different
                          Treatment Levels
               TREATMENT COST ESTIMATES
                         General Discussion
                         Model Plant Control Costs
               BASIS FOR REGULATIONS
                         BPT Effluent Limitations
                         BCT Effluent Limitations
                         BAT Effluent Limitations
                         NSPS Effluent Limitations
                         Pretreatment Standards

19.     CARBON DIOXIDE INDUSTRY

               SUMMARY OF DETERMINATIONS
               ASSESSMENT OF THE WATER POLLUTION
                POTENTIAL
                         Production Processes and
                          Effluents
               STATUS OF REGULATIONS

20.     CARBON MONOXIDE AND BY-PRODUCT HYDROGEN INDUSTRY

               SUMMARY OF DETERMINATIONS
               ASSESSMENT OF THE WATER POLLUTION
                POTENTIAL
                         Production Processes and
                          Effluents
               STATUS OF REGULATIONS

21 .     COPPER SULFATE INDUSTRY

               INDUSTRIAL PROFILE
                         General Description
                         Subcategorization
                         General Process Description and
                          Raw Materials
606
606
606

609
609

609
610

610

614
614
615
615
615
625
625
625
626

627

627
627

627

627

629

629
629

629

629

631

631
631
631
631
                               xiv

-------
                TABLE OF CONTENTS - Continued
             WATER USE AND WASTE SOURCE
              CHARACTERISTICS
                       Water Use
                       Waste Sources
             DESCRIPTION OF PLANTS VISITED AND SAMPLED
                       Screening
                       Verification
                       Toxic Pollutant Concentrations
             POLLUTION ABATEMENT OPTIONS
                       Toxic Pollutants of Concern
                       Process Modifications and
                        Technology Transfer Options
                       Best Management Practices
                       Prevailing Control and
                        Treatment Practices
                       Advanced Treatment Technologies
             SELECTION OF APPROPRIATE TECHNOLOGY AND
              EQUIPMENT
                       Technologies for Different
                        Treatment Levels
                       Equipment for Different
                        Treatment Levels
             TREATMENT COST ESTIMATES
                       General Discussion
                       Model Plant Cost Estimates
             BASIS FOR REGULATIONS
                       Evaluation of BPT Treatment
                        Practices
                       Basis for BPT Effluent
                        Limitations
                       Basis for BCT Effluent
                        Limitations
                       Basis for BAT Effluent
                        Limitations
                       Basis for New Source
                        Performance Standards
                       Basis for Pretreatment
                       Standards
                       Basis for Level 1 Treatment
                       Performance
634

634
634
637
637,
639
639
643
643
645

645
645

646
22.   NICKEL SULFATE INDUSTRY
             INDUSTRIAL PROFILE
                       General Description
                       Subcategorization
                       General Process Description and
                        Raw Materials
 646

 647

 ,649
 649
 651
 651
 651

 652

 663

 663

 664

664

 664


 673

 673
 673
 673
 673
                             xv

-------
                 TABLE  OF  CONTENTS  -  Continued
23.
        WATER USE AND WASTE SOURCE                    676
         CHARACTERISTICS
                  Water Use                           676
                  Waste Sources                       676
        DESCRIPTION OF PLANTS VISITED AND SAMPLED     676
                  Screening                           676
                  Verification                        679
                  Summary of Toxic Pollutant Data     679
        POLLUTION ABATEMENT OPTIONS                   685
                  Toxic Pollutants of Concern         685
                  Process Modifications and           687
                   Technology Transfer Options
                  Best Management Practices           687
                  Prevailing Control and              687
                   Treatment Practices
                  Advanced Treatment Technologies     688
        SELECTION OF APPROPRIATE TECHNOLOGY AND       688
         EQUIPMENT
                  Technologies for Different          688
                   Treatment Levels
                  Equipment for Different             689
                   Treatment Levels
        TREATMENT COST ESTIMATES                      694
                  General Discussion                  694
                  Model Plant Control Costs           694
        BASIS FOR REGULATIONS                         695
                  Evaluation of BPT Treatment         695
                   Practices
                  Basis for BPT Effluent              695
                   Limitations
                  Basis for BCT Effluent              700
                   Limitations
                  Basis for BAT Effluent              7po
                   Limitations
                  New Source Performance              708
                   Standards
                  Basis for pretreatment              708
                   Standards

SILVER NITRATE INDUSTRY                               709

        SUMMARY OF DETERMINATIONS                     709
        ASSESSMENT OF THE WATER POLLUTION             709
         POTENTIAL
                  Production Processes and            709
                   Effluents
        STATUS OF REGULATIONS                         709
                               xvi

-------
                  TABLE OF CONTENTS - Continued
24.     SODIUM BISULFITE INDUSTRY

               INDUSTRY PROFILE
                         General Description
                         Subcategorization
                         General Process Description and
                          Raw Materials
               WATER USE AND WASTE SOURCE
                CHARACTERISTICS
                         Water Use
                         Waste Sources
               DESCRIPTION OF PLANTS VISITED AND SAMPLED
                         Screening
                         Verification
                         Toxic Pollutant Analytical
                          Results
               POLLUTION ABATEMENT OPTIONS
                         Toxic Pollutants of Concern
                         Prevailing Control and
                          Treatment Practices
                         Advanced Treatment Technologies
               SELECTION OF APPROPRIATE TECHNOLOGY AND
                EQUIPMENT
                         Technologies for Different
                          Treatment Levels
                         Equipment for Different
                          Treatment Levels
               TREATMENT COST ESTIMATES
                         General Discussion
                         Cost Estimates
               BASIS FOR REGULATIONS
                         Evaluation of BPT Treatment
                          Practices
                         Basis for BPT Effluent
                          Limitations
                         Basis for BCT Effluent
                          Limitations
                         Basis for BAT Effluent
                          Limitations
                         Basis for New Source
                          Performance Standards
                         Basis for Pretreatment
                          Standards
711

711
714
714
714
714
714

724
724
728

728
728

728

729

733
733
734
734
734

734

746

746

747

747
                              xvii

-------
                   TABLE OF CONTENTS - Continued
25
26
SODIUM HYDROSULFITE (FORMATE PROCESS) Industry (Excluded)
        SUMMARY OF DETERriMATIOMS
        INDUSTRY PROFILE
                  General  Description
                  Subcategorization
                  General  Process Description and
                   Raw Materials
        WATER USE AND WASTE SOURCE
         CHARACTERISTICS
                  Water Use
                  Waste Sources
        DESCRIPTION OF PLANTS VISITED AND SAMPLED
                  Screening and Verification
                  Toxic Pollutant Concentrations
        POLLUTION ABATEMENT OPTIONS
                  Toxic Pollutants of Concern
                  Prevailing Control and
                   Treatment Practices
                  Advanced Treatment Technologies
        SELECTION OF APPROPRIATE TECHNOLOGY AND
         EQUIPMENT
                  Technologies for Different
                   Treatment Levels
                  Equipment for Different
                   Treatment Levels
        TREATMENT COST ESTIMATES
                  General  Discussion
                  Cost Estimates
        BASIS FOR GUIDANCE
                  Evaluation of Level 1
                   Treatment Practices
                  Basis for Level 1 Treatment
                   Performance
                  Basis for Level 2 Treatment
                   Performance

EXCLUDED SUBCATEGORIES

        ALUMINUM SULFATE
        AMMONIUM CHLORIDE
        AMMONIUM HYDROXIDE
        BARIUM CARBONATE
        BORAX
        BORIC ACID
        BROMINE
        CALCIUM CARBIDE
        CALCIUM CARBONATE
                                                              752

                                                              752
                                                              754
                                                              754
                                                              754
                                                              757
                                                              762
                                                              762
                                                              762

                                                              763
                                                              763

                                                              763

                                                              764

                                                              767
                                                              767
                                                              767
                                                              767
                                                              767

                                                              773

                                                              777
                                                              781

                                                              781
                                                              782
                                                              784
                                                              7R4
                                                              7P6
                                                              788
                                                              789
                                                              791
                                                              791
                               XVlll

-------
                      TABLE OF CONTENTS - Continued
                                                      •J
       CHROMIC ACID
       CUPROUS OXIDE
       FERRIC CHLORIDE
       FERROUS SULFATE
       FLUORINE
       HYDROCHLORIC ACID
       HYDROGEN. „
       IODINE
       LEAD MONOXIDE
       LITHIUM CARBONATE
       MANGANESE SULFATE
       NITRIC ACID
       OXYGEN AND NITROGEN
       POTASSIUM CHLORIDE
       POTASSIUM DI£HRPMATE
       POTASSIUM
-------
                        LIST OF FIGURES

                                                           Page

5-1    Sample flow sheet for metal analysis                  47

7-1    Theoretical solubilities of toxic metal               68
       hydroxides/oxides as a function of pH

7-2    Theoretical solubilities of toxic metal               73
       sulfides as a function of pH

7-3    Electrodialysis process                               85

8-1    Cumulative distribution of daily concentrations      115
       of mercury in treated effluent from Plant 1251

8-2    Cumulative distribution of daily concentrations      116
       of cyanide in treated effluent from Plant 1765

8-3    Statistical distribution for daily pollution         119
       measurements

8-4    Cumulative distribution of 30-day averages           121
       of total cyanide in treated effluent from Plant
       #782

8-5    Cumulative distribution of 30-day averages           122
       of ammonia in treated effluent from Plant #782

8-6    Statistical distributions for 30-day average         123
       pollution measurements

11-1   General process diagram for production of            150
       chlorine/caustic by mercury cells

11-2   General process flow diagram at Plant #299           155
       showing the sampling points.  Chlorine/caustic
       (mercury cell) manufacture

11-3   General process flow diagram at plant #747.          159
       showing the sampling points*  Chlorine/caustic
       (mercury cell) manufacture
                              XX

-------
                  LIST OF FIGURES - Continued
11-4   General process flow diagram at plant #167
       showing the sampling points.  Chlorine/caustic
       (mercury cell) manufacture

11-5   General process flow diagram at Plant #317
       showing the sampling points.  Chlorine/caustic
       (mercury cell) manufacture

11-6   Level 1 waste water treatment for
       chlorine-mercury cell subcategory

11-7   Level 2 waste water treatment for
       chlorine-mercury cell subcategory

11-8   General process flow diagram for production
       of chlorine/caustic by diaphragm cells

11-9   General process flow diagram at Plant #014
       showing the sampling points.  Chlorine/caustic
       (diaphragm cell) manufacture

11-10  General process flow diagram at Plant #261
       showing the sampling points.  Chlorine/caustic
       (diaphragm cell) manufacture

11-11  General process flowsheet at Plant #738-A
       showing the sampling points.  Chlorine/caustic
       (diaphragm cell) manufacture

11-12  General process flow diagram at Plant
       #738-B showing the sampling points.
       Chlorine/caustic (diaphragm cell)  manufacture

11-13  General process flow diagram at Plant #736
       showing the sampling points.  Chlorine/caustic
       (diaphragm cell) manufacture

11-14  General process flow diagram at Plant #967
       showing the sampling points.  Chlorine/caustic
       (diaphragm cell) manufacture

11-15  Level 1 waste water treatment for
       chlorine-diaphragm cell subcategory

11-16  Level 2 waste water treatment for
       chlorine-diaphragm cell subcategory

11-17  Level 3 waste water treatment for
       chlorine-diaphragm cell subcategory
161



173


174


198


203



206



207



208



209



210



228


229


230
                             xxi

-------
                  LIST OF FIGURES - Continued
                                                           Page

12-1   General process flow diagram for production          266
       of hydrofluoric acid

12-2   production vs. waste flow data for HF plants         270

12-3   General process flow diagram at Plant #705           276
       showing the sampling points.  Hydrofluoric acid
       manufacture

12-4   General process flow diagram at Plant #251           280
       showing the sampling points.  Hydrofluoric acid
       manufacture

12-5   Level 1 waste water treatment for hydrofluoric       289
       acid subcategory

12-6   Level 2 waste water treatment for hydrofluoric       290
       acid subcategory

12-7   Level 3 waste water treatment for hydrofluoric       291
       acid subcategory

12-8   Level 4 waste water treatment for hydrofluoric       292
       acid subcategory

12-9   Level 5 waste water treatment for hydrofluoric       294
       acid subcategory

12-10  Fluoride loads and concentrations discharged         308
       at selected hydrofluoric acid plants

14-1   General process diagram for production of            336
       titanium dioxide (chloride process)  from high
       grade ores

14-2   General flow diagram at Plant #559 showing           341
       the sampling points.  (Titanium dioxide -chloride
       process manufacture)

14-3   General flow diagram at Plant #172 showing           343
       the sampling points.  Titanium dioxide (chloride
       process) manufacture

14-4   Level 1 (BPT/BAT)  waste water treatment for          350
       titanium dioxide - chloride process

14-5   Level 2 (NSPS) waste water treatment for             351
       titanium dioxide - chloride process
                             XXI1

-------
              LIST OF FIGURES - Continued
14-6   Level 3 waste water treatment for titanium
       dioxide - chloride process

14-7   General process flow diagram for production
       of titanium dioxide by sulfate process

14-8   General flow diagram at plant #559 showing
       the sampling points.  (Titanium dioxide -
       sulfate process)

14-9a  Level 1 waste water treatment for titanium
       dioxide - sulfate process

14-9b  Level 2 waste water treatment for titanium
       dioxide - sulfate process

14-10  Level 3 waste water treatment for titanium
       dioxide - sulfate process

14-11  General process flow diagram of the titanium
       tetrachloride portion of a. titanium dioxide
       plant using the chloride-ilmenite process.

14-12  Level 1 waste water treatment for titanium
       dioxide - chloride (ilmenite ore) process

14-13  Level 2 waste water treatment for titanium
       dioxide - chloride (ilmenite ore) process

15-1   General process flow diagram for production
       of aluminum fluoride

15-2   General process flow diagram at Plant #705
       showing the sampling points.  (Aluminum
       fluoride manufacture)

15-3   General process flow diagram at Plant #251
       showing the sampling points.  (Aluminum
       fluoride manufacture)

15-4   Level 1 waste water treatment for aluminum
       fluoride subcategory

15-5   Level 2 waste water treatment for aluminum
       fluoride subcategory

15-6   Level 3 waste water treatment for aluminum
       fluoride subcategory
Page

 352


 382


 388



 398


 399


 401


 426



 438


 439


 463


 468



 470



 479


 480


 481
                            XXlll

-------
LIST OF FIGURES - Continued
15-7
16-1
16-2
16-3
16-4
16-5
16-6
16-7
i6-a
Level 4 waste water treatment for aluminum
fluoride subcategory
General process diagram for production of
anhydrous chrome oxide
General process diagram for production of
hydrated chromic oxide
General process diagram for production of
chrome yellow
General process diagram for production of
molybdenum orange
General process diagram for production of
chrome green
General process diagram for production of
zinc yellow
General process diagram for production of
chrome pigment complexes
General waste water treatment process flow
Page
483
503
504
506
507
509
510
512
516
       diagram at Plant #002 showing the sampling
       points.  (Chrome pigment manufacture)

16-9   General waste water treatment process  flow
       diagram at Plant #894 showing the sampling
       points.  (Chrome pigment manufacture)

16-10  Level 1 waste water treatment for chrome
       pigments

16-11  Level 2 waste water treatment for chrome
       pigments

17-1   General process flow diagram for production
       o£ hydrogen cyanide by the Andrussow Process

17-2   General waste water treatment process  flow
       diagram at Plant #765 showing the sampling
       points.  (Hydrogen cyanide manufacture)

17-3   General waste water treatment process  flow
       diagram at Plant #782 showing sampling points
       (Hydrogen cyanide manufacture)
                                              518



                                              529


                                              531


                                              555


                                              560



                                              564
                XXIV

-------
              LIST OF FIGURES - Continued
                                                           Page

17-4   Level 1 waste water treatment for hydrogen           572
       cyanide subcategory

17-5   Level 2 waste water treatment for hydrogen           574
       cyanide subcategory

18-1   General process diagram for production of            597
       sodium dichromate

18-2   General waste water treatment process flow           600
       diagram at Plant #493 showing the sampling
       points.  (Sodium dichromate manufacture)

18-3   General waste water treatment process flow           603
       diagram at Plant #376 showing the sampling
       points.  (Sodium dichromate manufacture)

18-4   Level 1 waste water treatment for sodium             612
       dichromate subcategory

18-5   Level 2 waste water treatment for sodium             613
       dichromate subcategory

21-1   General process flow diagram of the                  635
       manufacture of copper sulfate

21-2   General process flow diagram at Plant #034           640
       showing the sampling points. (Copper sulfate
       manufacture)

21-3   Level 1 waste water treatment for copper             648
       sulfate subcategory - batch process

21-4   Level 2 waste water treatment for copper             650
       sulfate subcategory - batch process

22-1   General process flow diagram for nickel sulfate      677
       manufacture

22-2   General waste water treatment process flow           680
       diagram showing sampling points at Plant #369.
       (Nickel sulfate subcategory.)

22-3   General process flow diagram at Plant #572           682
       showing the sampling points.  (Nickel sulfate
       manufacture.)
                             xxv

-------
              LIST OF FIGURES - Continued
                                                           Page

22-4   General waste water treatment process flow           683
       diagram at Plant #120 showing the sampling
       points.  (Nickel sulfate manufacture)

22-5   Level 1 waste water treatment for nickel             690
       sulfate subcategory - high production model -
       batch process

22-6   Level 1 waste water treatment for nickel             691
       sulfate subcategory - low and medium production
       models - batch process

22-7   Level 2 waste water treatment for nickel             692
       sulfate subcategory - high production model -
       batch process

22-8  'Level 2 waste water treatment for nickel             693
       sulfate - low and medium production models -
       batch process

24-1   General process flow diagram at plant #282           718
       showing the sampling points.  Sodium bisulfite
       manufacture

24-2   General flow diagram at Plant #586 showing           719
       the sampling points.  Sodium bisulfite
       manufacture

24-3   General process flow diagram at Plant #987           722
       showing the sampling points.  Sodium bisulfite
       manufacture

24-4   Level 1 waste water treatment for sodium             730
       bisulfite subcategory - batch process

24-5   Level 2 waste water treatment for sodium             731
       bisulfite subcategory - batch process

24-6   Level 3 waste water treatment for sodium             732
       bisulfite subcategory - batch process

25-1   General process flow diagram at Plant #672.          753
       (Sodium hydrosulfite manufacture)

25-2   General process flow diagram at Plant #672           756
       showing the sampling points.  (Sodium
       hydrosulfite manufacture)
                             xxvi

-------
              LIST OF FIGURES - Continued


25-3   Level 1 waste water treatment for sodium
       hydrosulfite subcategory

25-4   Level 2 waste water treatment for sodium
       hydrosulfite subcategory
Page

 765


 766
                             XXVI1

-------
                        LIST OF TABLES
2-1



2-2


2-3


2-4


2-5


3-1

3-2



5-1

5-2


5-3


6-1

7-1

7-2

8-1
Summary of Regulations - Best
Practicable Control Technology Currently
Available (BPT)

Summary of Regulations - Best
Available Technology (BAT)

Summary of Regulations -Pretreatment
Standards for Existing Sources (PSES)

Summary of Regulations - New
Source Performance Standards (NSPS)

Summary of Regulations -Pretreatment
Standards for New Sources (PSNS)

Recommended List of Toxic Pollutants

Scope of Industry Coverage within the
Inorganic Chemicals Manufacturing Point
Source Category

Analytical Detection Limits for Metals

Pollutant Frequency Based on Sampling
Program Results Including Raw Waste

Distribution of Pollutants According
to Subcategory

308-Questionnaire Response Data

Solubility Products of Toxic Metals

Comparison of Reverse Osmosis Concepts

Waste Water Treatment Options and
Performance Data Summary - Antimony and
Arsenic Removal
Page

   6



   8


  10


  12


  15


  23

  24



  51

  55


  56


  61

  69

  86

  92
                           XXVlll

-------
8-2



8-3


8-4



8-5


8-6


8-7


8-8


8-9



8-10


8-11


8-12



8-13



8-14


9-1


11-1
         LIST OF  TABLES  -  Continued
                                                 Page

Waste Water Treatment Options and                  93
Performance Data Summary - Beryllium and Cadmium
Removal

Waste Water Treatment Options and                  94
Performance Data Summary - Copper Removal

Waste Water Treatment Options and                  95
Performance Data Summary - Chromium III and
Chromium VI Removal

Waste Water Treatment Options and                  96
Performance Data Summary - Lead Removal

Waste Water Treatment Options and                  97
Performance Data Summary - Mercury II Removal

Waste Wat,er Treatment Options and                  98
Performance Data Summary - Nickel Removal

Waste Water Treatment Options and                  98
Performance Data Summary - Silver Removal

Waste Water Treatment Options and                  99
Performance Data Summary - Selenium and Thallium
Removal

Waste Water Treatment Options and                 100
Performance Data Summary - Zinc Removal

Estimated Achievable Maximum 30-Day Averages      102
for the Applied Technologies

Industrial Waste Water Treatment System           105
Performance — Summary of Effluent
Concentration Data on Toxic Metals

Estimated Achievable Long-Term Average            108
Concentrations for Toxic Metals with BPT or BAT
Treatment Options

Theoretical Solubilities of Toxic Metal           110
Hydroxides/Oxides at Various pH Values

Prioritization of Toxic Pollutants Found in       126
Each Subcategory

Subcategory Profile Data Summary                  147
                            xxxx

-------
                  LIST OF TABLES - Continued
                                                          Page

11-2      Status of  Regulations  - Effluent Limitation        148
          Guidelines

11-3      Summary of Waste Water Flow Data for               154
          Chlorine Mercury Cell  Plants

11-4      Pollutant  Concentrations  and Loads  at Plant        156
          #299

11-5      Pollutant  Concentrations  and Loads  at              157
          Verification Plants

11-6      Average Toxic Pollutant Raw Waste                  167
          Concentrations and Loads  at Verification
          Plants

11-7      Summary of Raw Waste Loadings  at                   168
          Verification Plants

11-8      Model Plant Treatment  Costs                       178

11-9      Model Plant Treatment  Costs                       179

11-10     Model Plant Treatment  Costs                       180

11-11     Model Plant Unit Treatment Costs                   181

11-12     Mercury Discharges from Selected Chlor-            182
          Alkali Mercury Cell Plants

11-13     Residual Chlorine Discharges at Selected           184
          Chlor-Alkali Plants

11-14     Comparison of Raw Waste Concentrations             185
          of Toxic Pollutants with  Treatability

11-15     Effluent Limitations,  BAT                         187

11-16     Effluent Concentrations of Toxic Pollutants        189
          from Verification Sampling

11-17     Subcategory Profile Data  Summary                   194

11-18     Status of  Regulations  - Effluent Limitation        195
          Guidelines

11-19     Waste Water Flows at Diaphragm Cell Chlorine       201
          Plants
                             XXX

-------
                  LIST OF TABLES - Continued
11-20     Pollutant Concentrations and Loads  at
          Screening and Verification Plants

11-21     Results of Asbestos Sampling at Diaphragm
          Cell Plants

11-22     Maximum Raw Waste Concentrations of Toxic
          Metals Observed at Diaphragm Cell Chlorine
          Plants (mg/1)

11-23     Toxic Metal Concentrations and Loads at
          Screening and Verification Plants

11-24     Summary of Raw Waste Loadings at Screening
          and Verification Metal Anode Plants

11-25     Toxic Metal Concentrations and Loads in
          Cell Room Waste Waters at Screening and
          Verification Plants

11-26     Raw Waste Toxic Metals Concentration and
          Loads in Process Streams Other Than Cell
          Room Wastes from Screening and
          Verification Plants

11-27     Raw Waste Toxic Organics at a Graphite
          Anode Plant

11-28     Raw Waste Toxic Organics by Waste Water
          Source at a Graphite Anode Plant

11-29     Model Plant Treatment Costs

11-30     Model Plant Treatment Costs

11-31     Model Plant Treatment Costs

11-32     Model Plant Unit Treatment Costs

11-33     Summary of Unit Flows at Diaphragm  Cell
          Plants

11-34     Comparison of Raw Waste Concentrations
          of Toxic Pollutants with Treatability

11-35     Effluent Limitations, BPCTCA
Page

 204


 213


 214



 216


 217


 219



 220
 221


 222


 235

 236

 237

 238

 240


 241


 243
                            XXXI

-------
                  LIST OF TABLES - Continued
11-36     Lead and TSS Discharges from Selected
          Diaphragm Cell Chlorine Plants

11-37     Comparison of Toxic Pollutants  in Diaphragm
          Cell Plant Effluents with Treatability

11-38     Effluent Limitations,  BAT

11-39     Comparison of Haw Waste Characteristics
          at a New Metal Anode Plant with Treatability
          of Toxic Metals

11-40     Effluent Limitations,  NSPS

12-1      Subcategory Profile Data Summary

12-2      Status of Regulations - Effluent  Limitation
          Guidelines

12-3      Water Usage in the Hydrofluoric Acid
          Subcategory

12-4      Waste Water Flow and Reuse Data for  the
          Hydrofluoric Acid Subcategory

12-5      Waste Flow from Hydrofluoric Acid
          Manufacturing plants

12-6      Flow and Pollution Concentration  Data of
          Spar Drying and Total Process Waste Water
          for Plants #251 and #837 Producing
          Hydrofluoric Acid

12-7      Solid Waste Generated at the Hydrofluoric
          Acid Plants Sampled

12-8      Gypsum Solids production in the Hydrofluoric
          Acid Subcategory

12-9      Flow and Pollutant Concentration  Data of
          the Sampled Waste Streams of Plant #705
          Producing Hydrofluoric Acid

12-10     Flow and Pollutant Concentration  Data of
          the Sampled Waste Streams for Plants #705,
          #251, and #167 Producing Hydrofluoric Acid

12-11     Toxic Pollutant Raw Waste Data
246


247


252

257



259

262

263


268


269


273


274
274


275


277



279



283
                            XXXll

-------
                  LIST OF TABLES - Continued
                                                           Page

12-12     Summary of Raw Waste Loadings Pound in            284
          Screening and Verification Sampling

12-13     Model Plant Treatment Costs                       297

12-14     Model Plant Treatment Costs                       298

12-15     Model Plant Treatment Costs                       299

12-16     Model Plant Unit Treatment Costs                  300

12-17     Summary of Waste Water Control and Treatment       302
          Technology Employed at Hydrofluoric
          Acid Plants

12-18     Summary of Long-Term Monitoring Data from         304
          Four Hydrofluoric Acid Plants

12-19     Toxic Pollutant Treated Effluent Data             305

12-20     Development of TSS and Fluoride Limitations        311

12-21     Effluent Limitations, BPCTCA                      312

12-22     Summary of Concentrations  and Filter              319
          Removal Efficiencies of Pollutants Selected for
          Study in the Hydrofluoric Acid Subcategory

12-23     Effluent Limitations, BAT                          321

12-24     Effluent Limitations, NSPS                         325

13-1      Subcategory Profile Data Summary                  328

14-1      Subcategory Profile Data Summary                  332
          (Chloride Process)

14-2      Status of  Regulations - Effluent Limitation        333
          Guidelines

14-3      Water Usage in Titanium Dioxide-Chloride           337
          Process/High Grade Ores Subcategory

14-4      Waste Water Flow for Titanium Dioxide-             339
          Chloride Process Subcategpry
                           XXXlll

-------
                  LIST OF TABLES - Continued
                                                          Page

14-5      Flow and Pollutant Concentration Data of          342
          the Sampled  Waste Streams of Plant #559
          Producing Titanium Dioxide  by Chloride-Rutile
          Process

14-6      Flow and Pollutant Concentration Data of          344
          the Sampled  Waste Streams for Plant #172
          Producing Titanium Dioxide  (Chloride Process)

14-7      Raw Waste Pollutant  Data Summary of the           346
          Sampled Streams

14-8      Model Plant  Treatment Costs                      355

14-9      Model Plant  Treatment Costs                      356

14-10     Model Plant  Treatment Costs                      357

14-11     Model Plant  Treatment Costs                      358

14-12     Model Plant  Treatment Costs                      359

14-13     Model Plant  Treatment Costs                      360

14-14     Model Plant  Unit Treatment  Costs                  362

14-15     Historical Effluent  Monitoring  Data Summary       363
          with Variability Factor —  Daily Measurements,
          Plant #559

14-16     Historical Effluent  Monitoring  Data Summary       364
          with Variability Factors — Daily Measurements,
          Plant #172

14-17     Treatment Performance Data  of Sampled Plants      365
          #599 and #172

14-18     Effluent Limitations, BPCTCA                      371

14-19     Effluent Limitations, BAT                         373

14-20     Effluent Limitations, NSPS                        378

14-21     Subcategory  Profile  Data Summary                  379
          (Sulfate Process)

14-22     Analysis of  Ilmenite Ores                         380
                            XXXIV

-------
                  LIST OF TABLES - Continued
                                                           Page

14-23     Water Usage in Titanium Dioxide -  Sulfate         383
          Process Subcategory

14-24     Raw Waste Characteristics  (Industry Data)          386
          for Plant #555 (Production  of TiO, by  Sulfate
          Process)

14-25     Flows and Pollutant Concentrations for            389
          the Waste Streams  Sampled for Plant #559
          Producing Titanium Dioxide

14-26     Process Waste Water Flow at Plants #555,           391
          #694 and #559 Titanium Dioxide
          (Sulfate Process)

14-27     Summary of Raw Waste Loadings Found in            393
          Screening and Verification  Sampling

14-28     Toxic Pollutants:  Average Raw Waste Loads          394
          and Concentrations

14-29     Model Plant Treatment Costs                       405

14-30     Model Plant Treatment Costs                       406

14-31     Model Plant Treatment Costs                       407

14-32     Model Plant Unit Treatment  Costs                  408

14-33a    Historical Effluent Monitoring  Data Summary—      410
          Treatment Without  Iron Removal

14-33b    Historical Effluent Monitoring  Data Summary—      411
          Treatment With Iron Removal

14-34     Verification Results from - Sulfate Process        412
          Titanium Dioxide Plant #559

14-35     Effluent Limitations, BPCTCA                      415

14-36     Effluent Limitations, BAT                          418

14-37     Effluent Limitations, NSPS                         420

14-38     Subcategory Profile Data Summary                  425
          (Chloride-llmenite)

14-39     Average Water Usage for TiO- Production            428
          by the Chloride -  Ilmenite  Process
                            xxxv

-------
                  LIST OF TABLES - Continued


14-40     Average Raw Waste  Loads  for TiO- Production
          by the Chloride -  Ilmenite Process

14-41     Summary of Raw  Waste Loadings Found  in
          Screening  and Verification Sampling

14-42     Toxic Pollutant Average  Raw Waste Loads
          and Concentrations

14-43     Model Plant Treatment Costs

14-44     Model Plant Treatment Costs

14-45     Model Plant Treatment Costs

14-46     Model Plant Treatment Costs

14-47     Model Plant Treatment Costs

14-48     Model Plant Treatment Costs

14-49     Model Plant Treatment Costs

14-50     Model Plant Treatment Costs

14-51     Model Plant Unit Treatment Costs

14-52     Effluent Limitations, BPCTCA

14-53     Effluent Limitations, NSPS

15-1      Subcategory Profile Data Summary

15-2      Status of  Regulations -  Effluent Limitation
          Guidelines

15-3      Water Usage in  the Aluminum Fluoride
          Subcategory

15-4      Waste Water Flow at Plants #837, #705 and
          #251 for Aluminum  Fluoride Subcategory

15-5      Solids Generated at Plant #705  and #251
          Producing  Aluminum Fluoride

15-6      Flow and Pollutant Concentration Data of
          the Sampled Waste  Streams for Plant  #705
          Producing  Aluminum Fluoride
Page

 429


 433


 434


 443

 444

 445

 446

 447

 448

 449

 450

 451

 454

 458

 462

 464


 464


 466


 466


 469
                            XXXVI

-------
                  LIST OF TABLES - Continued
                                                          Page

15-7      Flow and Pollutant  Concentration Data of          471
          the Sampled  Streams for Plant  #251 Producing
          Aluminum Fluoride

15-8      Toxic Pollutant Average Raw  Waste Loads           473
          and Concentrations

15-9      Toxic Pollutant Effluent  Concentrations           474
          During Sampling

15-10     Summary of Raw  Waste Loadings  Found  in            475
          Screening and Verification Sampling

15-11     Model Plant  Treatment Costs                       486

15-12     Model Plant  Treatment Costs                       487

15-13     Model Plant  Treatment Costs                       488

15-14     Model Plant  Unit Treatment Costs                  489

15-15     Effluent Limitations, BPCTCA                     497

16-1      Subcategory  Profile Data  Summary                  500

16-2      Status of Regulations - Effluent Limitation       501
          Guidelines

16-3      Water Usage  in  the  Chrome Pigments Subcategory    511

16-4      Summary of Waste Water Flow                       514

16-5      Flow, Pollutant Concentration  and Load            517
          Data of the  Sampled Waste Streams for
          Plant #002

16-6      Flow, Pollutant Concentration  and Load            519
          Data for the Sampled Streams at Plant #894

16-7      Toxic Pollutant Raw Waste Data                   522

16-8      Summary of Raw  Waste Loadings  Found  in            523
          Screening and Verification Sampling

16-9      Toxic Pollutant Treated Waste  Data                524

16-10     Model Plant  Treatment Costs                       534

16-11     Model Plant  Treatment Costs                       535
                            xxxvn

-------
                  LIST OF TABLES - Continued
                                                          Page

16-12     Model Plant Treatment  Costs                       536

16-13     Model Plant Treatment  Costs                       537

16-14     Model Plant Unit Treatment Costs                  539

16-15     Summary of Long  Term and Verification             541
          Effluent Sampling  Results at Plant #894

16-16     Effluent Limitations,  BPCTCA                      545

17-1      Subcategory Profile Data Summary                  552

17-2      Status of Regulations  - Effluent Limitation       553
          Guidelines

17-3      Water Usage in Hydrogen Cyanide - Andrussow       556
          Process Subcategory

17-4      Waste Plow Data  for HCN Production by  the         558
          Andrussow Process

17-5      Flow and Pollutant Data of the Raw and            561
          Treated Waste Streams  of Plant #765 Producing
          Hydrogen Cyanide by Andrussow Process

17-6      Flow and Pollutant Concentration Data of          563
          the Sampled Waste  Streams for Plant #765
          Producing Hydrogen Cyanide

17-7      Flow and Pollutant Concentration Data of          565
          the Sampled Waste  Streams for Plant #782
          Producing Hydrogen Cyanide

17-8      Unit Flow and Unit Pollutant Loading for          566
          Raw and Treated  Waste  Effluents at Plant 1782

17-9      Summary of Pollutant Raw Waste Loading            569
          Found in Screening and Verification Sampling

17-10     Model Plant Treatment  Costs                       577

17-11     Model Plant Treatment  Costs                       578

17-12     Model Plant Treatment  Costs                       579

17-13     Model Plant Unit Treatment Costs                  580
                          XXXVlll

-------
                  LIST OF TABLES - Continued
                                                          Page

17-14     Statistical Analysis of Historical Effluent        582
          Monitoring Data on Free Cyanide  and Ammonia
          from Plant #765

17-15     Statistical Analysis of the  28-Day Effluent        585
          Sampling Results on Total Cyanide from
          Plant #765

17-16     Effluent Limitations, BPCTCA                      587

17-17     Effluent Limitations, BAT                          589

17-18     Effluent Limitations, NSPS                         591

18-1      Subcategory Profile Data Summary                  594

18-2      Status of Regulations - Effluent Limitation        595
          Guidelines

18-3      Water Usage in Sodium Dichromate Subcategory       598

18-4      Flow and Pollutant Concentration Data of           601
          the Sampled Waste Streams for Plant #493
          Producing Sodium Dichromate

18-5      Flow and Pollutant Loading Data  of the             604
          Sampled Waste Streams for Plant  #376
          Producing Sodium Dichromate

18-6      Flow and Pollutant Loading Data  of the             605
          Sampled Waste Streams for Plant  #398
          Producing Sodium Dichromate

18-7      Toxic Pollutant Raw Waste Data                     607

18-8      Summary of Raw Waste Loadings Found in             608
          Screening and Verification Sampling

18-9      Model Plant Treatment Costs                        616

18-10     Model Plant Treatment Costs                        617

18-11     Model Plant Treatment Costs                        618

18-12     Model Plant Unit Treatment Costs                  619

18-13     Effluent Sampling Data from  Sodium                 622
          Dichromate Plants

18-14     Effluent Limitations, BPCTCA                      624
                           XXXIX

-------
                  LIST OF TABLES - Continued
                                                          Page

19-1      Subcategory Profile Data Summary                  628

20-1      Subcategory Profile Data Summary                  630

21-1      Subcategory Profile Data Summary                  632

21-2      Status of  Regulations  - Effluent Limitation       633
          Guidelines

21-3      Water Usage in Copper  Sulfate Subcategory         636

21-4      Waste Water Flow for the Copper Sulfate           638
          Subcategory

21-5      Flow and Pollutant  Concentration Data of          641
          the Sampled Waste Streams  for Plant #034
          Producing  Copper Sulfate

21-6      Raw Waste  Data                                   644

21-7      Model Plant Treatment  Costs                       653

21-8      Model Plant Unit Treatment Costs                  654

21-9      Summary of Long-Term Monitoring Data from         655
          Plant #034

21-10     Treated Effluent Data                             657

21-11     Average Pollutant Levels and Removal              658
          Efficiency for Plant #034

21-12     Effluent Limitations,  BPCTCA                      662

21-13     Effluent Limitations,  BAT                         665

21-14     Guidance for Effluent  Control                     668

22-1      Subcategory Profile Data Summary                  674

22-2      Status of  Regulations  - Effluent Limitation       675
          Guidelines

22-3      Water Use  in the Nickel Sulfate Subcategory       678

22-4      Flow and Pollutant  Concentration Data of          681
          the Sampled Waste Streams  for Plants Producing
          Nickel Sulfate

22-5      Toxic Pollutant Raw Waste  Data                    686
                             xl

-------
                  LIST OF TABLES - Continued
                                                           Page

22-6      Model Plant Treatment Costs                       696

22-7      Model Plant Treatment Costs                       697

22-8      Model Plant Treatment Costs                       698

22-9      Model Plant Unit Treatment  Costs                   699

22-10     Toxic Pollutant Treated Effluent  Data              701

22-11     Effluent Limitations, BAT                          705

23-1      Subcategory Profile Data Summary                   710

24-1      Subcategory Profile Data Summary                   712

24-2      Status of Regulations - Effluent  Limitation        713
          Guidelines

24-3      Water Usage in  the Sodium Bisulfite                715
          Subcategory

24-4      Waste Water Flow at Plants  #987 and #282           716
          for Sodium Bisulfite Subcategory

24-5      Flow and Pollutant Load Data of the Sampled        717
          Waste Streams for  Plant #282 Producing Sodium
          Bisulfite

24-6      Flow and Pollutant Load Data of the Sampled        720
          Waste Streams for  Plant #586

24-7      Flow and Pollutant Load Data of the Sampled        721
          Waste Streams for  Plant #987

24-8      Toxic Pollutant Raw Waste Loads                    725

24-9      Summary of Raw  Waste Loadings Found in             726
          Screening  and Verification  Sampling

24-10     Toxic Pollutant Concentrations Observed            727
          in Treated Effluent During  Verification
          Sampling

24-11     Model Plant Treatment Costs                       735

24-12     Model Plant Treatment Costs                       736

24-13     Model Plant Treatment Costs                       737

24-14     Model Plant Unit Treatment  Costs                   738
                             xli

-------
                  LIST OF TABLES - Continued
                                                          Page

24-15     Comparison of Maximum Raw Waste                   739
          Concentrations with Treatability

24-16     Effluent Limitations, BPCTCA                      742

25-1      Subcategory Profile Data  Summary                  750

25-2      Status of Regulations - Effluent Limitation       751
          Guidelines

25-3      Waste Source Data at Plant  #672                   755

25-4      Flow, Pollutant Concentration, and Load           758
          Data of the Sampled Waste Streams for
          Plant 1672 Producing Sodium Hydrosulfite

25-5      Sampling Results and Treatment System             760
          Performance for Toxic Pollutants — Plant #672

25-6      Summary of Raw Waste Loadings and                 761
          Concentrations Found at a Sodium Hydrosulfite
          Plant (Formate Process)

25-7      Model Plant Treatment Costs                      768

25-8      Model Plant Unit Treatment  Costs                  769

25-9      Subcategory Performance Evaluation Summary        772
          at Plant #672 for Conventional and
          Nonconventional Pollutants  in the Effluents

25-10     Guidance for Effluent Control - Level  1           776

25-11     Guidance for Effluent Control - Level  2           779

26.2-1    Subcategory Profile Data  Summary                  783

26.3-1    Subcategory Profile Data  Summary                  785

26.4-1    Subcategory Profile Data  Summary                  787

26.6-1    Subcategory Profile Data  Summary                  790

26.9-1    Subcategory Profile Data  Summary                  794

26.17-1   Subcategory Profile Data  Summary                  801
                             xlii

-------
                  LIST OF TABLES - Continued
26.23-1
26.24-1
26.27-1
26.32-1
26.35-1
26.36-1
26.37-1
26.38-1
26.40-1
26.42-1
26.43-1
26.44-1
Subcategory
Subcategory
Subcategory
Subcategory
Subcategory
Subcategory
Subcategory
Subcategory
Subcategory
Subcategory
Subcategory
Subcategory
Profile
Profile
Profile
Profile
Profile
Profile
Profile
Profile
Profile
Profile
Profile
Profile
Data Summary
Data Summary
Data Summary
Data Summary
Data Summary
Data Summary
Data Summary
Data Summary
Data Summary
Data Summary
Data Summary
Data Summary
Page
 807
 809
 814
 819
 823
 825
 827
 829
 831
 833
 836
 837
                             xliii

-------
                       ACKNOWLEDGEMENTS
     The technical study supporting the proposed regulation was
conducted by Jacobs Environmental Division of Jacobs Engineering
Group Inc. of Pasadena,  California under the direction of  Mr.
Henry Cruse, Vice  President,   and Mr.  Michael  Warner,  Project
Manager.  Major contributors  were  Ms.  Bonnie J.   Parrott,  Mr.
Chester Kaminski,  Mr.  Mahendra L.  Shah, Mr. Dale  Newkirk,  Mr.
Carl B.  Johnston,  Mr.  Dennis Merklin, Mr.  William E. Rorie,
Dr. David Ben  Hur, Dr.  Ashley Levy, Mr. Milford  Walker,  Jr.,
Ms. Garnett Gray,  Mr.  Dale Rushneck,  Dr. Ben C. Edmondson  and
Dr.  Martin D.   Schwartz.

     Cooperation and  assistance  of  the Chemical Manufacturers
Association and the Chlorine  Institute  and  numerous individual
corporations is  appreciated  for  their review  and comment  on
the  draft development document.    The numerous  company  and
plant personnel who submitted information,  opened  their plants
to   the  program   staff,    and   otherwise   cooperated    are
acknowledged and thanked for their  patience and  help.

     Mr. Steve Schatzow, Director, Office of Water Regulations
      S tandards,    is   gratefully   acknowledged   for    his
     :ibutions to the proiect.
and   S tandards,    is    grat
contributions to the project.
     Ms. Susan  Lepow, Mr.  Joseph Freedman,  and  Staff of  the
Office of General Counsel are specially  acknowledged  for  their
extensive contribution to  the  drafting of  the  regulations  and
this development document.

     Ms. Debra  Maness and  Mr.  Richard Cotz  of the Office  of
Analysis and Evaluation, Ms, Alexandra Tarnay, Monitoring  and
Data  Support  Division,  and  Mr.   Pred  Talcott,   Office   of
Planning and Evaluation are acknowledged for their  assistance.
                             xliv

-------
                              SECTION 1

                       CONCLUSIONS AND SUMMARY
                           Toxic Pollutants

The  following  35  inorganic  chemical  product  subcategories   were
screened   for   the   purpose  of  establishing  wastewater  effluent
limitations guidelines for existing sources, standards of  performance
for  new  sources,  and  pretreatment  standards  for new and existing
sources in this study:
          1.    Chlor-Alkali
          2.    Hydrofluoric Acid
          3.    Titanium Dioxide
          4.    Aluminum Fluoride
          5.    Chrome Pigments
          6.    Hydrogen Cyanide
          7.    Sodium Bichromate
          8.    Copper Sulfate
          9.    Nickel Sulfate
          10.  Sodium Bisulfite
          11.  Sodium Hydrosulfite
          12.  Hydrogen Peroxide
          13.  Hydrochloric Acid
          14.  Nitric Acid
          15.  Sodium Carbonate
          16.  Sodium Metal
          17.  Sodium Silicate
          18.  Sulfuric Acid
19.   Carbon Dioxide
20.   Carbon Monoxide and
     by-product Hydrogen
21.   Silver Nitrate
22.   Ammonium Chloride
23.   Ammonium Hydroxide
24.   Barium Carbonate
25.   Boric Acid
26.   Calcium Carbonate
27.   Cuprous Oxide
28.   Manganese Sulfate
29.   Strong Nitric Acid
30.   Oxygen and Nitrogen
31.   Potassium Iodide
32.   Sodium Hydrosulfide
33.   Sodium Silicofluoride
34.   Sodium Thiosulfate
35.   Sulfur Dioxide
The screening studies showed that only the plant  process  wastewaters
from  subcategories  1   through  11  contain treatable amounts of toxic
metals (see Table 3-1), cyanide and asbestos.   Very few of the organic
toxic pollutants were found in process waste streams  and  those  that
were identified were present at low level concentrations.

The screening results which indicated the presence of toxic pollutants
in  significant  amounts  were largely confirmed by the results of the
verification program.  Verification.sampling accounted for  50  to  75
percent  of  the  current  inorganic  chemical  production rate in the
subcategories covered.

The sources of most of the toxic pollutants found in  the  raw  wastes
and  treated  effluents  were  traced  to specific process-related raw
materials and chemicals used in the manufacturing operations.  In  the
case  of  certain  pollutants  found in widely varying amounts or with
erratic frequencies of  occurrence,   the  precise  identities  of  the

-------
sources  remain unknown at this time, but are suspected to be process-
related.

Control and Treatment Technology

A considerable amount of toxic pollutant removal is presently achieved
in the industry by  the  existing  control  and  treatment  practices.
Additional removal can be accomplished by the application of available
and  demonstrated  technologies  which would add to or modify existing
treatment systems.  Recovery of toxic metals for value or reuse  in  a
process  does  not  appear  to  be  an attractive alternative in those
industries where the product recovery practices now in effect  do  not
already accomplish this.

The  treatment  of  toxic  metal-bearing  waste streams results in the
production of sludges or residues which are potentially hazardous  and
may require special means for handling and disposal under the Resource
Conservation and Recovery Act (RCRA) regulations.

Costs of Additional In-Plant Treatment

The   estimated  incremental  costs  of  applying  the  candidate  BAT
treatment options represent  a  relatively  small  proportion  of  the
investment  and  operating  and maintenance costs already committed to
the existing BPT level treatment systems.  These costs, however,  vary
widely   from  industry  to  industry  and  are  highly  dependent  on
site-specific factors.
Subcategorization

A review of the product/process basis  for  Subcategorization  of  the
inorganic chemical product subcategories designated for study revealed
that  certain  modifications  may  be  appropriate  in the interest of
developing effective regulations.  The toxic pollutant problem per  se
impacts  sub-categorization directly only in the Chlor-Alkali Industry
where the use of graphite anodes  contributes  to  the  generation  of
chlorinated  hydrocarbons.   In  the  Titanium Dioxide Industry, major
process and  raw  material  differences  justify  the  creation  of  a
separate  segment  for  the sulfate process, the chloride process, and
for the chloride process using ilmenite ore.  Consideration was  given
to  creating a subcategory for the combined production of hydrofluoric
acid  and  aluminum  fluoride  in  view  of   their   similiar   waste
characteristics  and  the  current  practice  of combined treatment at
several plants.  However,  combining  these  products  into  a  single
subcategory does not appear to offer any regulatory advantages.

Hydrogen  cyanide  is  produced  by the Andrussow process and as a by-
product in the  manufacture  of  acrylonitrile.   By-product  hydrogen
cyanide  will  be covered under  its primary product, acrylonitrile, in
the Organic Chemicals Manufacturing Category.   The  hydrogen  cyanide
subcategory includes only manufacture by the Andrussow process.

-------
Restudy of Remanded Regulations

The   Fourth   Circuit,   U.S.  Court  of  Appeals  remanded  effluent
limitations guidelines promulgated for  11  major  inorganic  chemical
products.  E.I.  du Pont de Nemours versus Train, 541 F. 2d 1018 (4th,
Cir.  1976)  reversed  in  part,  430  U.S.  112  (1977}.   The factors
affecting the control and treatment of pollutant discharges  in  those
industries  have  been studied in response to the remanded issues.   It
has been concluded that alternative control and treatment technologies
to those originally considered for BAT and NSPS may be appropriate.

-------

-------
                              SECTION  2

                           RECOMMENDATIONS
On the basis of the toxic pollutant screening and  verification  results
and the evaluation of applicable  technologies   for discharge   control
and  treatment, it is recommended that effluent limitation  quidelines,
new source performance standards  and pretreatment  standards  for   new
and  existing  sources  be  promulgated for  the following  10  inorganic
chemicals manufacturing subcategories:
     Chlor-Alkali
     Hydrofluoric Acid
     Titanium Dioxide
     Aluminum Fluoride
     Chrome Pigments
Hydrogen Cyanide
Sodium Dichromate
Copper Sulfate
Nickel Sulfate
Sodium Bisulfite
Table  2-1 summarizes the  regulations  for  Best   Practicable   Control
Technology  Currently  Available   (BPT).  Summaries of  regulations  for
Best Available  Technology   (BAT),  Pretreatment   Standards,   and   New
Source  Performance  Standards  are given in Tables 2-2,  2-3,  2-4,  and
2-5.   These tables indicate  that Chlor-Alkali has  been  divided   into
two segments and Titanium Dioxide  in three segments before  listing  the
numerical effluent limitations.

The  Agency proposed BPT, BCT, and BAT limitations and  NSPS, PSES,  and
PSNS for the Sodium Hydrosulfite (Formate Process)  subcategory.    The
proposed  regulation  basically'1"1 added control of  selected  toxic metal
pollutants to existing  treatment  practiced  in   the   industry.    The
Agency  reviewed  the  basis for the proposed regulation  and concluded
that the total current treated discharge load of only 0.42  pounds   per
day  total  toxic  metals  from  all  plants in the subcategory is  too
insignificant   to   justify   developing   a   national    regulation.
Accordingly,   this   subcategory  has  been  excluded  from   national
'regulation development under  Paragraph  8(a)(iv)  of   the  Settlement
Agreement.

-------
            TABLE 2-1.  SUMMARY OF REGULATIONS -
                        BEST PRACTICABLE CONTROL TECHNOLOGY
                        CURRENTLY AVAILABLE (BPT)
Subcategory
Parameter
                                     Effluent Limitations
kg/kkg (or lh/1000 Ib) of product
Chlor-alkali,
Mercury Cells

Chlor-alkali,
Diaphragm Cells



Hydrofluoric
Acid



Sodium
TSS
Mercury
pH
TSS
Copper (T)
Lead (T)
Nickel (T)
pH
TSS
Fluoride (T)
Nickel (T)
Zinc (T)
pH
TSS
Dichromate Hexavalent Chromium



Titanium
Dioxide
(sulfate
process)
Titaniun
Dioxide
(chloride
process)
Titanium
Dioxide (chlor-
ide ilmenite
process)
Chromium (T)
Nickel {T)
pH
TSS
Chromium (T)
Nickel (T)
&
TSS
Chromium (T)
pH

TSS
Chromium (T)
Nickel (T)
pH
0.32
0.00014

0.51
0.0070
0.010
0.0056

5.3
2.9
0.011
0.036

0.22
0.00050
0.0044
0.0034

38
0.21
0.14

6.4
0.030


9.6
0.053
0.035

0.64
0.00028

1.1
0.018
0.026
0.014

11.0
6.1
0.036
0.12

0.44
0.00090
0.0088
0.0068

140
0.48
0.29

23
0.057


35
0.12
0.072



6.0 to 9.0




6.0 to 9.0




6.0 to 9.0




6.0 to 9.0



6.0 to 9.0


6.0 to 9.0




6.0 to 9.0
                                                                    (continued)

-------
Table  2-1.  Continued
                                     Effluent Limitations
SuDcategory

Aluminum
Fluoride
parameter

TSS
Fluoride (T)
Chromiun (T)
Nickel (T)
pH
Max
30-day Avg
kg/kkg (or lb/1000 Ib)
1.2
0.63
0.0045
0.0024
24-hr
Max
of product
2.4
1.3
0.015
0.0079
pH Range

6.0 to 9.0
Copper Sulfate
TSS
Copper  (T)
Nickel  (T)
Selenium  (T)
0.023
0.0010
0.0020
0.00050
0.069
0.0030
0.0060
0.0015
                                                                     6.0 to 9.0
Hydrogen
Cyanide
Nickel Sulfate
Chrome
Pigments
Sodium
Bisulfite
TSS
Cyanide A
Cyanide  (T)
TSS
Nickel  (T)
pH

TSS
Chromiun  (T)
Lead (T)
Zinc (T)
TSS
COD
Chromium (T)
Zinc (T)
3.2
0.021
0.23
0.032
0.0020
3.8
0.13
0.15
0.13
0.080
0.95
0.00063
0.0015
8.6
0.10
0.65
0.096
0.0060
9.1
0.31
0.36
0.31
0.32
3.8
0.0020
0.0051
                                                                     6.0 to 10.5
                                                                     6.0 to 9,0
                                                                     6.0 to 9.0
                                                                     6.0 to 9.0

-------
                TABLE 2-2.  SUMMARY OF REGULATIONS -
                            BEST AVAILABLE TECHNOLOGY  (BAT)
                                      Effluent Limitations
Subcategory
Parameter
                                    Max
                                 30-day Avg
                                      24-hr
                                       Max
                                kg/kkg (or lb/1000 ib) of product
Ghlor-alkali
Mercury Cells
Mercury (T)
Total Residual
Chlorine
0.00010
0.0019
0.00023
0.0032
Chlor-alkali     Copper (T)         0.0049
Diaphragm Cells  Lead (T)            0.0024
                 Nickel (T)         0.0037
                 Total Residual
                  Chlorine          0.0079
                                       0.012
                                       0.0059
                                       0.0097

                                       0.013
Hydrofluoric
Acid
Fluoride (T)
Nickel (T)
Zinc (T)
Sodium Chromium (T)
Dichromate Hexavalent Chromium
Nickel (T)
Titanium
Dioxide
(sulfate
process)
Titanium
Dioxide
(chloride
process)
Titanium
Dioxide
(chloride-
ilmenite
process)
Aluminum
Fluoride
Chromium (T)
Nickel (T)
Chromium (T)
Chromium (T)
Nickel (T)
Fluoride (T)
Chromium (T)
Nickel (T)
1.6
0.0060
0.022
0.0044
0.00050
0.0034
0.21
0.14
0.030
0.053
0.035
0.63
0.0045
0.0024
3.4
0.020
0.072
0.0088
0.00090
0.0068
0.48
0.29
0.057
0.12
0.072
1.3
0.013
0.0079
(continued)

-------
 Table 2-2.   Continued
                                     Effluent Limitations
Subcategory
Parameter
                                    Max
                                  30-day Avg
                                       24-hr
                                       Max
                                kg/kkg  (or  lb/1000  UD) of product
Chrome
Pigments

Chromium
Lead (T)
Zinc (T)
0.13
0.15
0.13
0.31
0.36
0.31
Copper Sulfate
Copper  {T)
Nickel  (T)
Selenium (T)
0.0010
0.0020
0.00050
0.0030
0.0060
0.0015
Hydrogen
Cyanide
Nickel Sulfate
Sodium
Bisulfite
Cyanide A
Cyanide  (T)
Total Residual
 Chlorine

Copper (T)
Nickel (T)

COD
Chromium (T)
Zinc (T)
0.021
0.23

0.051

0.00024
0.00024

0.95
0.00063
0.0015
0.10
0.65

0.086

0.00074
0.00074

3.8
0.0020
0.0051

-------
TABLE 2-3.  SUMMARY OF REGULATIONS -
            PRETREATMENT STANDARDS FOR EXISTING
            SOURCES (PSES)

Subcategory
Oiler-alkali
Parameter
Mercury (T)
Effluent Limitations
Max 30-day 24-hr
Avg Max
(rog/1) or (kg/kkg) (mg/1) or (kg/kkg)
0.048
0.00010
0.11
0.00023
Mercury Cells
Chlor-aUcali
Diaphragm
Cells
Hydrofluoric
Acid

Sodium
Diohromate

Titanium
Dioxide
(sulfate
process)
Titanium
Dioxide
(chloride
process)
Titanium
Dioxide
(chloride-
ilmenite
process)
Chrone
Pigments

Copper (T)
Lead (T)
Nickel (T)
0.80
1.1
0.64
Fluoride (T) 50
Nickel (T)
Zinc (T)
Chromium (T)
Hexavalent Chromium
Nickel (T)
Chromium (T)
Nickel (T)


Chromium (T)



Chromium (T)
Nickel (T)



Chronium (T)
Lead (T)
Zinc
0.20
0.66
0.50
0.060
0.40
0.44
0.29


0.30



0.44
0.29



1.2
1.4
1.2
0.0070
0.010
0.0056
1.6
0.0060
0.022
0.0044
0.00050
0.0034
0.21
0.14


0.030



0.053
0.035



0.13
0.15
0.13
2.1
2.9
1.6
100
0.66
2.2
1.0
0.11
0.80
1.0
0.60


0.57



1.0
0.60



2.9
3.4
2.9
0.018
0.026
0.014
3.4
0.020
0.072
0.0088
0.00090
0.0068
0.48
0.29


0.057



0.12
0.072



0.31
0.36
0.31
                                                   (continued)
                           10

-------
TABLE 2-3.  Continued
                                          Effluent Limitations
Subcategory
Parameter
                                     Max 30-day
                                         Avg
                                 (mg/1) or  (kg/kkg)
                                            24-hr
                                             Max
                                       (mg/1) or (kg/kkg)
Copper
Sulfate

Hydrogen
Cyanide
Copper (T)
Nickel (T)
Selenium (T)
Cyanide A
Cyanide (T)
1.1
2.1
0.53
0.36
4.0
0.0010
0.0020
0.00050
0.021
0.23
3.2
6,4
1.6
1.7
11
0.0030
0.0060
0.0015
0.10
0.65
Nickel
Sulfate

Sodium
Bisulfate
Copper (T)
Nickel (T)
0.36
0.36
COD          630
Chromium (T)   0.42
Zinc (T)       1.0
0.00024
0.00024

0.95
0.00063
0.0015
   1.1
   1.1

2500
   1.3
   3.4
0.00074
0.00074

3.8
0.0020
0.0051
                                     11

-------
               TABLE 2-4.  SUMMARY OF REGULATIONS -
                           NEW SOURCE PERFORMANCE STANDARDS  (NSPS)
Subcategory
Parameter
                                     Effluent Limitations
                                    Max
                                 30-day Avg
                                       24-hr
                                       Max
                                kg/kkg  (or lb/1000 Ib) of product
                                                        Range
Chlor-alkali
Mercury Cells
TSS            0.32
Mercury (T)    0.00010
Total Residual
 Chlorine      0.0019
                                       0.64
                                       0.00023

                                       0.0032
                                                                     6.0 to 9.0
Chlor-alkali
Diaphragm Cells
Hydrofluoric
Acid
Sodium
Dichromate
Titanium
Dioxide
(sulfate
process)
TSS            0.32
Lead  (T)       0.0018
Total Residual
 Chlorine      0.0079
TSS
Fluoride  (T)
Nickel  (T)
Zinc  (T)
pH

TSS
Chranium  (T)
Hexavalent
 Chromium
Nickel '(T)
pH

TSS
Iron  (T)
Chromium  (T)
Nickel  (T)
pH
               3.0
               1.6
               0.0060
               0.022
               0.22
               0.0044

               0.00050
               0.0034
              30
               1.2
               0.14
               0.095
                                      1.0
                                      0.0047

                                      0.013
  6.0
  3.4
  0.020
  0.072
  0.44
  0.0088

  0.00090
  0.0068
110
  4.0
  0.27
  0.18
                                                                      6.0  to  9.0
                                                                     6.0 to 9.0
                                                                     6.0 to 9.0
                                                                     6.0 to 9.0
Titanium
Dioxide
(chloride
process)
TSS
Iron (T)
Chromium (T)
FH
4.0
0.16
0.012

14
0.52
0.023




6.0 to



9.0
(continued)
                                      12

-------
TABLE 2-4.  Continued


Subcategory


Titanium
Dioxide
(chloride-
ilmenite
process)
Aluminum
Fluoride



Chrome
Pigments



Copper Sulfate




Hydrogen
Cyanide



Parameter


TSS
Iron (T)
Chromium (T)
Nickel (T)
£H
TSS
Fluoride (T)
Chromium (T)
Nickel (T)
pH
TSS
Chromium (T)
Lead (T)
Zinc (T)
pH
TSS
Copper (T)
Nickel (T)
Selenium (T)
£H
TSS
Cyanide A
Cyanide (T)

Efflue
Max
30-day Avg
kg/kkg (or
2.4
0.096
0.0072
0.010

1.2
0.63
0.0045
0.0024

3.8
0.13
0.15
0.13

0.023
0.0010
0.0020
0.00050

3.2
0.021
0.23

nt Limitations
24-hr
Max
lb/1000 lb) of product
8.4
0.32
0.014
0.020

2.4
1.3
0.015
0.0079

9.1
0.31
0.36
0.31

0.069
0.0030
0.0060
0.0015

8.6
0.10
0.65



pH Fange





6.0 to 9.0




6.0 to 9.0




6.0 to 9.0




6.0 to 9.0



                 Ibtal Residual
                   Chlorine
0.051
0.086
                                                                     6.0 to 10.5
Nickel Sulfate



TSS
Copper (T)
Nickel (T)
pH
0.032
0.00024
0.00024

0.096
0.00074
0.00074




6.0 to 9.0
                                                                 (continued)
                                     13

-------
TABLE 2-4.  Continued
Subcategory      Parameter
                                     Effluent Limitations
                                    Max
                                 30-day Avg
                                      24-hr
                                       Max
                                kg/kkg (or lb/1000 Ib)  of product
                                    pH Pange
Sodium
Bisulfite
TSS
COD
Chromium (T)
Zinc (T)
0.080
0.95
0.00063
0.0015
0.32
3.8
0.0020
0.0051
                                                                     6.0 to 9.0
                                      14

-------
            TABLE 2-5.  SUMMARY OF REGULATIONS -
                        PRETREATMENT STANDARDS FOR NEW SOURCES  (PSNS)
Subcategory
Parameter
                                           Effluent Limitations
                                       Max
                                   30-day Avg
                                (mg/1) or (kg/kkg)
                                           24-hr
                                            Max
                                       (mg/1) or (kg/kkg)
Chlor-alkali
Mercury Cells

Chlor-alkali
Diaphragm Cells
Mercury
Lead (T)
0.048
0.21
0.00010
0.0018
0.11
0.53
0.00023
0.0047
Hydrofluoric
Acid

Sodium
Dichromate


Titanium
Dioxide
(sulfate
process)
Titanium
Dioxide
(chloride
process)
Titanium
Dioxide
(chloride-
ilmenite
process)
Chrome
Pigments

Fluoride (T)
Nickel (T)
Zinc (T)
Chromium (T)
Hexavalent
Chroniun
Nickel (T)
Iron (T)
Chromium (T)
Nickel (T)

Iron (T)
Chromium (T)


Iron (T)
Chromium (T)
Nickel (T)


Chrcmium (T)
Lead (T)
Zinc (T)
50
0.20
0.66
0.50

0.060
0.40
2.5
0.30
0.20

1.6
0.12


1.6
0.12
0.17


1.2
1.4
1.2
1.6
0.0060
0.022
0.0044

0.00050
0.0034
1.2
0.14
0.095

0.16
0.012


0.096
0.0072
0.010


0.13
0.15
0.13
100
0.66
2.2
1.0

0.11
0.80
8.5
0.57
0.38

5.3
0.23


5.3
0.23
0.33


2.9
3.4
2.9
3.4
0.20
0.072
0.0088

0.00090
0.0068
4.0
0.27
0.18

0.52
0.023


0.32
0.014
0.020


0.31
0.36
0.31
                                                                 (continued)
                                      15

-------
TABLE 2-5.  Continued
                                           Effluent Limitations
Subcategory
Paramater
                                       Max
                                   30-day Avg
                                (mg/1) or (kg/kkg)
                                           24-hr
                                            Max
                                       (mg/1) or  (kg/kkg)
Copper
Sulfate
Hydrogen
Cyanide
Copper (T)
Nickel (T)
Selenium (T)
Cyanide A
Cyanide (T)
1.1
2.1
0.53
0.36
4.0
0.0010
0.0020
0.00050
0.021
0.23
3.2
6.4
1.6
1.7
11
0.0030
0.0060
0.0015
0.10
0.65
Nickel
Sulfate

Sodium
Bisulfite
Copper  (T)
Nickel  (T)
0.36
0.36
COD          630
Chronium (T)   0.42
Zinc (T)       1.0
0.00024
0.00024

0.95
0.00063
0.0015
   1.1
   1.1

2500
   1.3
   3.4
0.00074
0.00074

3.8
0.0020
0.0051
                                      16

-------
                              SECTION 3

                             INTRODUCTION



                              AUTHORITY

The Federal Water Pollution Control Act Amendments

The  Federal Water Pollution Control Act (the Act) Amendments of 1972,
33 USC 1251 et seq.,  stated the national goal of attaining by July  1,
1983,   a   water  quality  which  provides  for  the  protection  and
propagation of fish  and  shellfish,  for  recreation  in  or  on  the
nation's  waters,  and  the  goal  of  eliminating  the  discharge  of
pollutants into navigable waters by 1985.

Purpose and Authority

The Federal Water Pollution Control Act Amendments of 1972 established
a  comprehensive  program  to  "restore  and  maintain  the  chemical,
physical,  and  biological  integrity of the Nation's waters," Section
101(a).   By  July  1,  1977,  existing  industrial  dischargers  were
required to achieve "effluent limitations requiring the application of
the  best practicable control technology currently available" ("BPT"),
Section 301(b)(1){A); and by July  1,   1983,  these  dischargers  were
required to achieve "effluent limitations requiring the application of
the  best  available technology economically achievable...  which will
result in reasonable further progress  toward  the  national  goal  of
eliminating   the   discharge  of  all  pollutants"  ("BAT"),  Section
301(b)(2)(A).   New industrial  direct  dischargers  were  required  to
comply  with  Section  306  new source performance standards ("NSPS"),
based on best available demonstrated technology; and new and  existing
dischargers to publicly owned treatment works ("POTW") were subject.to
pretreatment  standards  under  Sections  307(b)  and  (c) of the Act.
While the requirements for direct dischargers were to be  incorporated
into  National  Pollutant Discharge Elimination System (NPDES) permits
issued under Section 402 of the Act, pretreatment standards were  made
enforceable   directly   against   dischargers   to   POTW   (indirect
dischargers).

Although Section 402(a)(l) of the 1972 Act authorized the  setting  of
requirements  for direct dischargers on a case-by-case basis, Congress
intended that for the most part control requirements would be based on
regulations promulgated by the Administrator of EPA.   Section  304(b)
of  the  Act  required  the  Administrator  to  promulgate regulations
providing guidelines for effluent limitations setting forth the degree
of effluent reduction attainable through the application  of  BPT  and
BAT.     Moreover,   Sections  304(c)  and  306  of  the  Act  required
promulgation of regulations for NSPS,  and Sections 304(f), 307(b), and
307(c)  required  promulgation   of   regulations   for   pretreatment
standards.    In  addition to these regulations for designated industry
categories, Section 307(a) of the Act required  the  Administrator  to
                                   17

-------
develop  a
applicable
501 (a)  of
additional
the Act.
            list of toxic pollutants and promulgate effluent standards
           to all dischargers of toxic pollutants.   Finally,  Section
            the  Act  authorized  the  Administrator  to prescribe any
           regulations "necessary to carry out  his  functions"  under
The  EPA  was  unable  to  promulgate many of these regulations by the
dates contained in  the  Act.   In  1 976,  EPA  was  sued  by  several
environmental  groups, and in a settlement of this lawsuit EPA and the
plaintiffs executed a "Settlement Agreement" which was approved by the
Court.  This Agreement required to EPA to develop a program and adhere
to a schedule for promulgating BAT  effluent  limitations  guidelines,
pretreatment  standards,  and  new source performance standards for 65
"priority"  pollutants  and  classes  of  pollutants  for   21   major
industries.  See Natural Resources Defense Council, Inc. versus Train,
9 ERC 2120 (D.D.C. 1976), modified March 9, 1979.

On  December  27,  1977, the President signed into law the Clean Water
Act of 1977.  Although this law makes several important changes in the
Federal water pollution control program, its most significant  feature
is  its  incorporation  of  several  of  the  basic  elements  of  the
Settlement Agreement program for toxic  pollution  control.   Sections
301 (b) {2} (A)  and  301 (b) (2) (C) of the Act now require the achievement
by July 1, 1984, of effluent limitations requiring application of  BAT
for   "toxic"  pollutants,  including  the 65 "priority" pollutants and
classes of pollutants which Congress declared  "toxic"  under  Section
307(a)   of   the  Act.    Likewise,  EPA's  programs  for  new  source
performance  standards  and  pretreatment  standards  are  now   aimed
principally  at toxic pollutant controls.  Moreover, to strengthen the
toxics control program  Section  304(e)  of  the  Act  authorizes  the
Administrator  to  prescribe  "best  management practices" ("BMPs") to
prevent the release of toxic and hazardous pollutants from plant  site
runoff, spillage or leaks, sludge or waste disposal, and drainage from
raw   material   storage   associated   with,  or  ancillary  to,  the
manufacturing or treatment process.

In keeping with its emphasis on toxic pollutants, the Clean Water  Act
of  1977  also  revises  the control program for non-toxic pollutants.
Instead of BAT for "conventional" pollutants identified under  Section
304(a-)(4)  (including  biochemical  oxygen  demand,  suspended solids,
fecal  coliform  and  pH),  the  new  Section  301 (b) (2) (E)   requires
achievement  by  July  1, 1984, of "effluent limitations requiring the
application of the best  conventional  pollutant  control  technology"
("BCT").   The  factors  considered  in  assessing BCT for an industry
include the cost  of  attaining  a  reduction  in  effluents  and  the
effluent  reduction benefits derived compared to the costs incurred by
and the effluent reduction benefits from a  publicly  owned  treatment
works   {Section   304(b)(4)(B)}.    For   non-toxic,  nonconventional
pollutants, Sections 301 (b) (2) (A) and (b)(2)(F) require achievement of
BAT effluent limitations within three years after their  establishment
or  by  July  1,  1984,  whichever is later, but not later than July 1,
1987.
                                   18

-------
The purpose of these regulations is to  provide  effluent  limitations
guidelines  for BPT, BAT, and BCT,  and to establish NSPS, pretreatment
standards for existing sources (PSES), and pretreatment standards  for
new  sources (PSNS), under Sections 301, 304, 306, 307, and 501 of the
Clean Water Act.

The United States Environmental Protection  Agency  (the  Agency)  was
entrusted with the responsibility to carry out the requirements of the
Act,  and  initiated  an  intensive  effort  to  develop the necessary
regulatory means  which  would  achieve  the  stepwise  reduction  and
elimination  of  pollutant  discharge  practices  in  all  major  U.S.
Industries.  Fgr the Inorganic Chemicals  Manufacturing  Point  Source
Category,  the  Agency  designed a comprehensive, two phase program to
identify the control parameters and establish the technological  basis
for  regulations  development.   Phase  I  covered  22 Major Inorganic
Chemical Products (1 ), and the final regulations for these  industrial
subcategories  were  published  in  the  Federal Register on March 12,
1974.   The   regulations   included   specific   numerical   effluent
limitations  and  standards  of  performance for both existing and new
sources.   Zero-discharge  requirements  specified  for  many  of  the
subcategories were to be applied either at the 1977 BPT step or later.
Phase  II  of  the Agency's effort resulted in the promulgation of BPT
based effluent limitations for an additional group of 27 subcategories
referred to as  Significant  Inorganic  Chemical  Products  (2).   The
interim  final  regulations  were  published  on  May 22, 1975.  Taken
together, the two groups of regulations cover  49  inorganic  chemical
subcategories,  many  of which include more than one specific chemical
product.  Although some toxic pollutants were covered in cases where a
direct relationship to the process was obvious (e.g.,  mercury  and/or
lead in the Chlor-Alkali Industry), the main thrust of the regulations
was  the control of the pollutant parameters which accounted, in terms
of quantity, for most of the pollution  loading  of  navigable  waters
attributable to the manufacture of inorganic chemicals.

Court Remand of Regulations

On  March  10, 1976, the United States Court of Appeals for the Fourth
Circuit decided in E.I. du Pont de Nemours & Company,  et  al.  versus
Train,  541  F.  2d  1018 (4th Cir. 1976), to set aside and remand for
reconsideration a number of general definitions and specific discharge
regulations promulgated in 1974.  These  regulations  are  all  within
Title 40, Parts 401  and 415 of the Code of Federal Regulations and are
listed below:

     General Provisions
          401.11 (i) - Definition of effluent limitations
          401.11 (q) - Definition of process wastewater
          401.11 (r) - Definition of process wastewater pollutant

     Chlor-Alkali
          415.63 - BATEA

     Hydrochloric Acid
                                   19

-------
          415.72  -  BPCTCA
          415.73  -  BATEA
          415.75  -  New  sources
     Hydrofluoric' Acid
          415.82  - BPCTCA
          415.83  - BATEA
          415.85  - New  sources
     Nitric Acid
          415.102
          415.103
          415.105
      BPCTCA
      BATEA
      New sources
     Sodium Carbonate
          415.152  - BPCTCA
          415.153  - BATEA
          415.155  - New sources

     Sodium Dichromate
          415.173  - BATEA
     Sodium Metal
          415.182
          415.183
          415.185

     Sodium Silicate
      BPCTCA
      BATEA
      New sources
          415
          415
          415

     Sulfuric
          415
          415
          415
          415
192 - BPCTCA
193 - BATEA
195 - New sources

Acid
210 - Applicability
212 - BPCTCA
213 - BATEA
215 - New sources
     Titanium Dioxide
          415.220 - Applicability
          415.222 - BPCTCA
          415.223 - BATEA
          415.225 - New sources

For  the  most  part,  the  main  target  of  the  remand was the zero
discharge regulations  from  which  the  industry  petitioners  sought
relief  on  grounds  of technological infeasibility.   During 1975,  the
Agency funded a special  study  of  the  remand  issues  (3)  and  was
prepared to propose amended regulations.
                                    20

-------
Following  the  court  remand  of  the  Phase  I final regulations, the
Agency revoked the Phase II interim  final  and  proposed  regulations
published  in  May,  1975,  for  Aluminum  Fluoride,  Chrome Pigments,
Hydrogen Cyanide, and Sodium Silicofluoride.   In  this   instance,  the
Agency's  intent  was  to reconsider the specific effluent limitations
established  for  these  industries  (1977  step)   in   the  light   of
information  made  available on process differences between plants and
additional data on the actual concentrations and treatability  of  the
regulated  discharge  constituents.   The information was presented to
the Agency in the form of various documents prepared by members of the
industries concerned.  These sources are also  cited in  the appropriate
sections of this report.

The Settlement Agreement

A consent decree was issued in a  suit  filed  by   four environmental
groups  in  Natural Resources Defense Council  versus Train, 9 ERG 2120
(June 8, 1976) modified 12 ERC 1833 (December  15, 1978).  The  consent
decree  contained  a Settlement Agreement wherein the Agency agreed to
regulate 65 toxic pollutants under Sections 301, 304, 306, and 307  of
the  Act  in  accordance  with the schedule and provisions stipulated.
The original list of 65 chemicals and classes  of chemicals attached to
the  Settlement  Agreement  was  redefined  to cover    129   chemical
substances, including specific organic compounds, pesticides and their
metabolites,  polychlorinated  biphenyls  (PCB's),  cyanide,  13 heavy
metals and  asbestos.   Table  3-1  lists  the 129  toxic  pollutants
(sometimes referred to in the literature as "priority pollutants").


The  Settlement  Agreement  also identified 21 point source categories
and specified the scope of application of  effluent  limitations,  new
source  performance  standards, and pretreatment standards within each
category in terms of the Standard Industrial Calssification (SIC) code
numbers.  For  the  Inorganic  Chemicals  Manufacturing Point  Source
Category, the major industries included are:

          SIC (2182 - Alkalies and Chlorine
          SIC 2813 - Industrial Gases
          SIC 2816 - Inorganic Pigments
          SIC 2819 - Industrial Inorganic Chemicals,
                    Not Elsewhere Classified

Within  these  industries,  the Agency has identified 63 subcategories
listed  in  Table  3-2  for  the  initial  study.    Most   of   these
subcategories,  A9  in  all,- had  already been covered by BPT and BAT
discharge regulatTotis; promulgated in 1974 and  1975.  Those regulations
established point of discharge control  levels for  the  conventional
^parameters  such  as pH, TSS, BOD, and oil and grease.   In many cases,
specific chemical parameters  were  regulated,  particularly  Arsenic,
Chromium,  Copper, Mercury, Nickel, Lead, Selenium, Zinc, and Cyanide,
which are now  included  in  the  list  of  toxic   pollutants.   Other
regulated parameters such as aluminum, barium, iron, ammonia, fluoride
and  sulfide are not presently listed as toxic chemicals but are to be
                                   21

-------
treated  as  nonconventional   pollutants
limitations and standards of performance.
under   future   discharge
Nearly  half of the  initial 63 subcategories  have been recommended for
exclusion from this  study on the basis of  specific provisions  for such
exclusion under Paragraph 8 of the Settlement Agreement.   The  bases
for these exclusions are as follows:

A.   No. 63, Ferrous Sulfate,  is  already covered  by   the   Titaniun
     Dioxide  -  Sulfate  Process  subcategory  and  does  not require
     separate consideration.

B.   No's.  60, 61 and  62  (Potassium  Permanganate,  Zinc  Oxide,  and
     Lithium  Carbonate) have only one plant  each (or one plant with  a
\     wet process discharge), and represent nonsignificant  discharges
I     of toxic pollutants.  No's. 27 and  28 (Copper Oxide  and Manganese
     Sulfate) are also single plants, but  were covered in screening.

     No's.  36  through  59  have  existing   BPT  or  BAT  regulations
     requiring zero  discharge of process wastewater to navigable water
     and    there  are  no  known  discharges  to  a  POTW.   Continued
     enforcement of  the existing  regulations will  provide   adequate
     control of toxic  pollutants.


The remaining 35 nonexcluded subcategor^gs^LTable 3-2, No's. 1 through
35)  are  covered   in  this  report.   This group also includes  the  11
subcategories whose  final regulations were remanded  for  restudy   in
E.I.   du Pont de Nemours and Company, et  al, versus Train, supra, and
the four additional  subcategories whose   interim,  final  or   proposed
regulations were revoked and reserved by  the  Agency.

It  was  anticipated by the Agency that  a  substantial number of  the  35
industries  to be screened  would  also   qualify  for  exclusion  under
Paragraph   8  on the basis of the analytical  results obtained  from the
process wastewater  toxic pollutant screening  program.   A preliminary
prioritization   indicated   that   the   initial  detailed  study  and
regulation  development would focus on the  first  15 subcategories.

This judgment has   been  substantially   supported  by  the  analytical
results   of  .the   screening  programs   and   a   number  of  additional
exclusions  are  being recommended   for   subcategories   in  which
nonsignificant  toxic  pollutant  discharges  have been determined.   A
detailed presentation  of the analytical  results  is  given  under  the
individual  subcategory  sections  of  this   report.   The  additional
recommended exclusions include the following:
                                    22

-------
                                                    TABLE  3-1.    FECOMMENEGD LIST  OF  TOXIC POLLUTANTS
to
LO
                            17.
                            1 S .
                            19.

                            20,
32.
33.
* acenaphtl^ane
• aero) e in
" acryJonitrile
• benzene
*benzidine
* carbon teirachloride
  (teir o chi of om ethane)
•chlorinated benzenes (other than
  dichiorol'pnzenes)
  chlorooer.zene
  l^^-trichiorobenzene
  hexachlorobenzene
 •chlorinated ethanes (including 1,2-
  dichloroethane, 1,1,1-trichloro-
  ethane a^id he xachioroe thane)
   1,2-dichlor oethane
   1,1,1-trichloroeihane
  hexachloroethane
   l,l-dic~.ioroethane
   l,t,:-:ncriioroethane
   1,1,2, i-teirachlor oethane
  ch!oroe:nane
 •chloroalkyl ethers (chlorornethyl,
  chloroethyl and mixed ethers
   bis(chloromethyl) et'ner^i
   bisl2-ch)oroethyl) ether
   2-chloroethyl vinyl ether (mixed)
 •chlorinated naphthalene
   2-chloronaphihaleie
 •chlorinated phenols (other than
   those listed.elsewhere; includes
   trichiorophenols and chlorinated
   cresols)
   2,ft,6-tric>ilorophenol
   p-chJoro-m-cresol
 •chioroiorui (trichioromethane)
 *2-chlorop>ienoi
 • dichiorobenzenes
   1,2-dichlorobenzene
   1,3-dichiorobenzene
   1 ,ft-dichlorobenzene
 • di chJorooenz i dine
   3,3-dichiorobenzidine
 •dichloroethylenes (1,1-dichloroeth-
   ylene and 1,2-dicnloroethylenel
   1,1-dichioroethylene
                                 *2,iliJOTamtiane( 11,12-
                                                            benzofluoranthene)
                                                          chrysene
                                                          acenaphthylene
                                                          anthracene
                                                          benzo(ghi)perylene(l,12-
                                                            benzoperylene)
                                                          fluorene
                                                          phenanthrene
                                                          dibenzo(a,h)anthracene (1,2,: 5,6-
                                                            dibenzanthracene)
                                                          indeno (l,2,3-cd)pyrene (2,3-o-
                                                            phenylenepyrene)
                                                          pyrene
•tetrachloroethylene
•toluene
•trichloroethylene
*vinyl chloride (chloroethylene)
pesticides and metabolites
  *aldrin
  *dieldrin
  "chlordane (technical mixture &
   metabolites)
•DDT and metabolites
  ft,d'-DDT
  V-DDE (p,p'-DDX)
  ft,t'-DDD (p.p'-TDE)
•endosulfan and metabolites
  a-endosulf an- Mpha
  b-endosulf an-Beta
    endosulfan sulfate
•endfin and metabolites
  endrin
  endrin aldehyde
•helptachlor and metabolites
  heptachlor
  heptachlor epoxide
•hexachlorocyclohexane (all isomers)
  a-BHC-A!pha
  b-BHC-Beta
  g-BHC (Iindane)-Gamm3
  d-BHC-Delta
•potycNorinated biphenyls (RGB's)
  PCB-12W (Arochlor J2»2>
  PCB-1251 (Arochlor 1251)
  PCB-1221 (Arochlor 1221)
  PCB-1232 (Arochlor 1232)
  PCB-12ft8 (Arochlor 12ft8)
  PCB-1260 (Arochlor 1260)
  PCB-1016 (Arochlor 1016)
•toxaphene
•antimony (total!
•arsenic (total)
"asbestos (fibrous)
•beryllium (total)
'cadmium (total)
•chromium (total)
'copper (total)
'cyanide (total)
-lead (total)
•mercury (total)
»nickel (total)
•selenium (total)
'silver (total)
'thallium (total)
 'zinc (total)
"2,3,7,8-tetrachlorodibenzo-p-
   dioxin (TCDD)
                                                                                                                                          85.
                                                                                                                                          86.
                                                                                                                                          87.
                                                                                                                                          88.

                                                                                                                                          89.
                                                                                                                                          90.
                                                                                                                                          91.
92.
93.
9ft.

95.
96 .
97.

98.
99.

100.
101 .

102.
103.
10ft.
105.

106.
107.
108.
109.
ilO.
111.
112.
113.
114.
115.
116.
117.
118.
119.
120.
121.
122.
123.
12ft.
125.
126.
127.
128.
129.
                                                                                                                             ,t;^6FR 10723.
                                                                                                              t2)oeleted CH/OS/81; t6 FR 2266
                                                                                                               •Specific compounds and chemical
                                                                                                               classes as listed in the consent
                                                                                                               decree.

-------
      TABLE 3-2.  SCOPE OF INDUSTRY COVERAGE WITHIN THE INORGANIC
             CHEMICALS MANUFACTURING POINT SOURCE CATEGORY
               Subcategories Designated for Initial Study
 1.  Chlor-Alkali
 2.  Hydrofluoric Acid
 3.  Hydrogen Peroxide
 4.  Titanium Dioxide
 5.  Aluminum Fluoride
 6.  dhrone Pigments
 7.  Hydrogen Cyanide
 8.  Sodium Dichromate
 9.  Carbon Dioxide
10.  Carbon Monoxide/Hydrogen
11.  Copper Sulfate
12.  Nickel Sulfate
13.  Silver Nitrate
14.  Sodium Bisulfite
15.  Sodium Hydrosulfite
16.  Hydrochloric Acid
17.  Nitric Acid
18.  Sodium Carbonate
19.  Sodium Metal
20.  Sodium Silicate
21.  Sulfuric Acid
22.  Ammonium Chloride
23.  Aimonium Hydroxide
24.  Barium Carbonate
25.  Boric Acid
26.  Calcium Carbonate
27.  Copper Oxide
28.  Manganese Sulfate
29.  Strong Nitric Acid
30.  Oxygen and Nitrogen
31.  Potassium Iodide
32.  Sodium Hydrosulfide
33.  Sodium Silicofluoride
34.  Sodium Thiosulfate
35.  Sulfur Dioxide
36.  Bromine
37.  Calcium Hydroxide
38.  Chromic Acid
39.  Fluorine
40.  Hydrogen
41.  Iodine
42.  Potassium Chloride
43.  Stannic Oxide
44.  Zinc Sulfate
45.  Calcium Carbide
46.  Calcium Oxide
47.  Potassium Metal
48.  Potassium Sulfate
49.  Sodium Bicarbonate
50.  Borax
51.  Ferric Chloride
52.  Lead Monoxide
53.  Sodium Fluoride
54.  Aluminum Chloride
55.  Aluminum Sulfate
56.  Potassium Dichromate
57.  Calcium Chloride
58.  Sodium Chloride
59.  Sodium Sulfite
60.  Potassium Permanganate
61.  Zinc Oxide
62.  Lithium Carbonate
63.  Ferrous Sulfate
                                   24

-------
          Number
                                   Subcateqory
          3.
          9.
          10.
          15.
          16.
          17.
          18.
          19.
          21 .
          22.
          23.
          24.
          25.
          26.
          27.
          28.
          29.
          30.
          31 .
          32.
          34.
          35.
                           f Hydrogen Peroxide       -^.,
                           f   Carbon Dioxide            "^
                           VCarbon Monoxide/Hydrogen ^
                              5o3iUifi"Hyarosul"fite
                              Hydrochloric Acid
                              Nitric Acid
                              Sodium Carbonate
                              Sodium Metal
                              Sulfuric Acid
                              Ammonium Chloride
                              Ammonium Hydroxide
                              Barium Carbonate
                              Boric Acid
                              Calcium Carbonate
                              Copper Oxide (one plant)
                              Manganese Sulfate (one plant)
                              Strong Nitric Acid
                              Oxygen and Nitrogen
                             . Potassium Iodide
                              Sodium Hydrosulfide
                              Sodium Thiosulf^te
                              Sulfur Dioxide
Silver Nitrate, No. T3, and Sodium Silicofluoride, No. 33,  are  being
deferred  for  future  study  under  Phase  H  of  the BAT regulation
development program for  Inorganic  Chemicals^   This  deferrment  was
caused  by problems with plant access during the course of the present
study.

General Approach and Methodology
Initiating
pollutant
            and  undertaking  a  comprehensive  study  of  the   toxic
           problem in the Inorganic Chemicals Industry was preceded by
                                                         of  data  and
                                                         basis for the
                                                          identity  of
                                                         and standards
                                                         by documented
an intensive evalutation by the  Agency  of  the  kinds
supporting  information  that  should  be assembled as a
development of regulations.  All major decisions on  the
pollutants and the establishment of effluent limitations
of performance for each subcategory had to be suportable
evidence  collected  from operating production facilities.  Similarly,
the  necessary  information  on  production  trates,   processes,   raw
materials,  water  use,  waste  sources, and treatment technologies in
practice has to be acquired with  sufficient  detail  and  breadth  of
coverage  to  permit  an  analysis  of  the  engineering  and economic
variables  that  are  characteristic  of  each   subcategory.    Toxic
pollutant  control  regulations  would  be based on the application of
best available  technology  for  treatment  and  reliable  performance
evaluations for the removal of specific waste substances.

The  following  paragraphs  briefly describe the major study tasks and
their results as they are presented in this report.

-------
 Industry Data Base Development and Subcategorization Review

 Information from  individual manufacturers and previous study documents
 was reviewed  in detail and an evaluation  of  the  appropriateness  of
 subcategorization  was  performed.  Section 4 presents a discussion of
 the factors considered in subcategorization and presents the rationale
 for maintaining   the  present  scheme  of  subcategorization   for  the
 industries studied.


 The Screening and Verification Sampling Programs

 The    collection   of   detailed   analytical  data  on  conventional,
 nonconventional and toxic pollutant concentrations in raw  and   treated
 process  waste  streams was completed  in a two-phase sampling  program.
 The first phase,  screening, was designed to provide a  representative,
 one-time  72-hour  sampling of a plant in each subcategory in  order to
 determine the presence of toxic and other pollutants and   to   evaluate
 their   potential   environmental   significance.   The  sampling  and
 analytical methodology is described in Section 5, along with the basis
 for making a decision on the need for  verification  sampling   in  each
 subcategory.

 Engineering Evaluations

 Section  6 describes the procedures and sources used in developing the
 industry productions and wastewater  generation  characteristics  that
 form   the  basis  of the model plant concept.  The sources of  detailed
 process and waste treatment information are also presented.  Section  7
 contains an evaluation of treatment technology  presently   applied  in
 BPT  systems  and  advanced technologies that may be recommended for BAT
 and  NSPS  applications.   Section  8  provides   estimates   of   the
 treatablilty  of  selected  toxic and  nonconventional pollutants to be
 applied in the development of achievable  performance  characteristics
 for  specific  technologies.   Section 8 also presents a discussion of
 the approach  taken  in statistical  analysis  of  long-term monitoring
 data.   The   statistically  derived  parameters, including variability
 factors for the 24-hour maximum and maximum 30-day average limitations
 are presented in  Appendix A.  Section  9 lays the  groundwork   for  the
 estimation  of  pollutant  removal  performances  for each nonexcluded
 subcategory.  The candidate toxic pollutants to be controlled  in  each
 subcategory   are   identified  on  the basis  of  the  screening  and
"verification  data and the rationale for the  application   of   advanced
 level  technologies  is presented.

 Treatment System  Cost Estimates

 Section 10 presents  the general approach to cost estimating, discusses
 the  assumptions  made,  and  gives  the  detailed   cost estimates for
 alternative levels  of treatment and  control.   For  each   subcategory
 verified,  the  total  estimated   installed  cost  of  a   typical  BPT
 treatment system  is  developed on  the basis of  the model  plant design
 specifications  and  estimated  incremental costs are given for each of
                                     26

-------
the advanced level treatment alternatives.  Costs of  RCRA  compliance
and  implementation  have  not  been included in these cost estimates.
Figures for RCRA costs  are  included  in  a  supplementary  document,
"Contractor Report on RCRA ISS Compliance coats for Selected Inorganic
Chemicals  Industries".   RCRA  costs  are  considered in the Economic
Impact Analysis of Pollution Control Technolqcuejs for Segments of_  the
Inorganic Chemicals Manufacturing Industry, EPA 440/2-81-023.

Treatability Study


Data  was  collected  through  a  treatability  study  to evaluate the
achievable performance of proposed BAT for the treatment  and  control
of  pollutant  discharges,  and  to provide empirical treatment system
performance information  applicable  to  selected  inorganic  chemical
subcategories.   The  study,  completed  in  July  1980,  specifically
concentrated  on  those  subcategories  in  the  Inorganic   Chemicals
Industry  for  which  analytical  data  on raw wastewaters and treated
effluents either did not exist or was deficient, and  for  which  data
were   needed  for  purposes  of  comparison  with  proposed  effluent
limitations.   Subcategories  for  which  treatability   was   studied
include:

     Nickel Sulfate
     Hydrofluoric Acid                          \"O
     Copper Sulfate
     Chlor-Alkali (Diaphragm Cells)
     Titanium Dioxide (Chloride Process)
     Chrome Pigments
     Sodium Dichromate
     Sodium Bisulfite
     Sodium Hydrosulfite

General Criteria for Effluent Limitations

BPT Effluent Limitations

The factors considered in defining best practicable control technology
currently  available  (BPT)  include  the total cost of applying  such
technology in relation to the effluent reductions  derived  from  such
application, the age of equipment and facilities involved, the process
employed,  non-water  quality  environmental impacts (including energy
requirements),  and  other   factors   the   Administrator   considers
appropriate  (Section 304 (b) (1) (B).   In general, the BPT technology"
level represents the average of  the  best  existing  performances  of
plants   of   various   ages,   sizes,    processes,  or  other  common
characteristics.  Where existing performance is uniformly  inadequate,
BPT  may be transferred from a different subcategory or category.  BPT
focuses on  end-of-pipe  treatment  rather  than  process  changes  or
internal  controls,  except  where  such are common industry practice.
The cost/benefit inquiry for BPT is a limited balancing, committed  to
EPA's  discretion,  which  does  not  require  the  Agency to quantify
benefits in monetary  terms.   See,  e.g.,  American  Iron  and  Steel
                                    27

-------
Institute v.  EPA, 526 F. 2d 1027 (3rd Cir. 1975).  In balancing costs
in  relation  to effluent reduction benefits,  EPA considers the volume
and nature of existing discharges, the volume and nature of discharges
expected after application of BPT, the general  environmental  effects
of  the  pollutants,  and the cost and economic impacts of the required
pollution  control  level.    The  Act  does  not  require  or   permit
consideration  of  water  quality  problems attributable to particular
point  sources  or  industries,  or  water  quality  improvements   in
particular  water  bodies.    Therefore,  EPA  has not considered these
factors.  See Weyerhaeuser Company v. Costle,  590 F.2d 1011 (D.C. Cir.
1978}.

BAT Effluent Limitations

The  factors  considered  in  assessing  best   available   technology
economically  achievable  (BAT)  include  the  age  of  equipment  and
facilities involved,  the process employed, process changes,  non-water
quality environmental impacts (including energy requirements) and such
other  factors  as  the Administrator deems appropriate.   (Section 304
(b)  (2) (B)).  At a minimum, the BAT technology level  represents  the
best  economically  achievable  performance of plants of various ages,
sizes,  processes, or  other  shared  characteristics.   As  with  BPT,
uniformly  inadequate  performace  may  require transfer of BAT from a
different subcategory or category.  BAT may include process changes or
internal  controls,  even  when  these  technologies  are  not  common
industry practice.  The statutory assessment of BAT "considers" costs,
but  does  not require a balancing of costs against effluent reduction
benefits (see Weyerhaeuser v. Costle, supra).   In developing  the  BAT
regulations,   however,   EPA  has  given  substanital  weight  to  the
reasonableness of costs.  The Agency has  considered  the  volume  and
nature  of  discharges,   the  volume and nature of discharges expected
after application of BAT, the general  environmental  effects  of  the
pollutants,  and  the  costs  and  economic  impacts  of  the required
pollution control levels.   Despite  this  expanded  consideration  of
costs,    the   primary   determinant  of  BAT  is  effluent  reduction
capability.  As a result of the Clean Water Act of 1977, 33  USC  1251
et seq. the achievement of BAT has become the principal national means
of controlling water pollution due to toxic pollutants.

BCT Effluent Limitations

The  1977  amendments  added  Section  301  (b)   (2)  (E)  to the Act,
establishing "best conventional pollutant control  technology"   (BCT)
for  discharges  of  conventional  pollutants from existing industrial
point sources.  Conventional pollutants are those defined  in  Section
304  (b)  (4)  -BOD,   TSS,  fecal coliform, and pH.  Oil and grease was
designated by the Administrator as "conventional" on July 30, 1979, 44
FR 44501.  BCT is not an additional limitation, but replaces  BAT  for
the control of conventional pollutants.

Section  304(b)(4)(B)  of  the  Act  requires  that BCT limitations be
assessed in light of a two part "cost-reasonableness" test.   American
Paper  Institute v. EPA, 660 F.2d 954  (4th Cir. 1981).  The first test
                                   28

-------
compares the cost for private  industry  to  reduce  its  conventional
pollutants  with  the  costs  to  publicly  owned  treatment works for
similar levels of reduction in their discharge  of  these  pollutants.
The   second   test  examines  the  cost-effectiveness  of  additional
industrial treatment beyond BPT,  EPA must find that  limitations  are
"reasonable"  under both tests before establishing them as BCT.  In no
case  may  BCT  be  less  stringent  than  BPT.   EPA  published   its
methodology  for  carrying out the BCT analysis on August 29, 1979 (44
FR 50732).  However, that cost test was remanded by the United  States
Court  of Appeals for the Fourth Circuit.  American Paper Institute y_._
EPA, 660 F.2d 954 (4th Cir. 1981).  The Court of Appeals  ordered  EPA
to correct data errors underlying EPA's calculation of the first test,
and to apply the second cost test.  (EPA had argued that a second cost
test  was  not  required).   The  Agency is Currently developing a new
methodology.

The proposed regulations had set BCT equal to BPT in all subcategories
except for the Chlor-Alkali subcategory (Diaphragm Cell  Segment)  and
the Hydrofluoric Acid subcategory, either because BPT was set equal to
BAT  or  because  additional  TSS removal failed the Agency's original
"cost-reasonableness" test.  In the Chlor-Alkali (Diaphragm Cell)  and
Hydrofluoric  Acid  subcategories,  additional  TSS removal passed our
original test.  Pending  formulation  by  the  Agency  of  a  new  BCT
methodology,  the  Agency is deferring promulgation of BCT limitations
for  the  Chlor-Alkali  (Diaphragm   Cell)   and   Hydrofluoric   Acid
subcategories.   BPT  is  the minimum level of BCT control required by
law.   In  all  other  subcategories,   we  h^ve  identified  no  other
economically  achievable  technologies  which  result  in  significant
additional  removal   of   conventional   pollutants.    No   possible
reassessment of BCT pursuant to the Court's remand could result in BCT
limitations different than those promulgated.  Accordingly, the Agency
is promulgating BCT equal to BPT for all oth^r subcategories.
The  cost  calculations  for  the  Chlor-Alkali  (Diaphragm
Hydrofluoric Acid subcategories are presented below.
                 Cell) and
In the diaphragm cell segment of  the  Chlor-Alkali  Subcategory,  the
cost  for  removal  of additional conventional pollutants is $0.53 per
pound.  The calculation is as follows:
               $0.22
            (0.51 kg/kkg - 0.32 kg/kkg)
$0.53 per pound
   of TSS removed,
where $0.22  is  the  increased  cost  for  SAT  filtration  over  BPT
treatment  cost  in  dollars per kkg of production, 0.51 kg/kkg is the
BPT total suspended solids  maximum  30-day  average  limitation  from
Table  11-35, and 0.32 kg/kkg is the achievable maximum 30-day average
TSS loading level with filtration.  One kilogram (kg) is 2.2 pounds.

In  the  Hydrofluoric  Acid  Subcategory,  the  cost  for  removal  of
additional   conventional   pollutants   is   $0.32  per  pound.   The
calculation is as follows:
                                    29

-------
                    $1 .46
             5.3 kg/kkg - 3.2 kg/kkg
$0.32 per pound
    of TSS removed,
where $1.46 is the increased cost for BAT treatment over BPT treatment
cost in dollars per kkg of production  from  Table  12-16,  where  5.3
kg/kkg  is the BPT total suspended solids limitation from Table 12-21,
and where 3.2  kg/kkg  is  the  effluent  TSS  loading  achievable  by
application of BAT.  See Section 12.

In  the  Hydrofluoric Acid and Chlor-Alkali subcategories where BCT is
deferred, the TSS and pH limitations are the same as BPT.

New Source Performance Standards

The basis for new source performance standards  (NSPS)  under  Section
306  of  the  Act  is the best available demonstrated technology.  New
plants have the opportunity to design  the  best  and  most  efficient
inorganic  chemicals  manufacturing processes and wastewater treatment
technologies, and Congress therefore directed EPA to consider the best
demonstrated  process  changes,  in-plant  controls,  and  end-of-pipe
treatment  technologies  which  reduce pollution to the maximum extent
feasible.

Pretreatment Standards for Existing Sources

Section 307 (b) of the Act requires  EPA  to  promulgate  pretreatment
standards  for  existing  sources {PSES} which must be achieved within
three years  of  promulgation.   PSES  are  designed  to  prevent  the
discharge  of  pollutants  which  pass through, interfere with, or are
otherwise incompatible with the operation of POTWs.  The  Clean  Water
Act  of  1977 also requires pretreatment for pollutants, such as toxic
metals, that limit POTW sludge management alternatives, including  the
beneficial  use  of  sludges  on  agricultural lands.  Pretreatment is
required for toxic pollutants  that  would  pass  through  a  POTW  in
amounts  that  would  violate  direct discharger effluent limitations.
EPA has generally determined that there is pass through of  pollutants
if the percent of pollutants removed by a well-operated POTW achieving
secondary  treatment is less than the percent removed by the BAT model
treatment system.  The legislative history of the 1977  Act  indicates
that  pretreatment  standards are to be technology-based, analogous to
the best available technology for removal of  toxic  pollutants.   The
general  pretreatment  regulations  which  served as the framework for
these pretreatment regulations can be found at 40 CFR Part 403, 43  FR
27736  (June 26, 1978).


The  Agency  is promulgating PSES for the Chlor-Alkali(Diaphragm Cell)
and Chrome Pigments Subcategories  and  the  Agency  is  amending  the
existing PSES for  the Copper Sulfate and Nickel Sulfate Subcategories.
The  Agency  is excluding the Chlor-Alkali(Mercury cell), Hydrofluoric
Acid, Sodium  Dichromate,  Titanium  Dioxide,  Hydrogen  Cyanide,  and
Sodium  Bisulfite  Subcategories  from national categorical PSES under
the provisions of  Paragraph 8(b) of the Settlement  Agreement  because
                                    30

-------
the  toxic  pollutants  in  discharges  to POTWs from sources in those
subcategories are below treatable levels or are  so  insignificant  as
not  to  justify developing pretreatment standards.  The Agency is not
promulgating PSES for the  Aluminum  Fluoride  Subcategory  because  a
well-operated  POTW with secondary treatment installed achieves better
percent removal of toxic pollutants than is provided by the BAT  model
treatment system for this subcategory.

Pretreatment Standards for New Sources


Section  307  (c)  of  the Act requires EPA to promulgate pretreatment
standards for new sources (PSNS) at the same time that it  promulgates
NSPS.  New indirect dischargers, like new direct dischargers, have the
opportunity   to   incorporate   the   best   available   demonstrated
technologies including process changes, in-plant controls, and end-of-
pipe treatment technologies, and to use plant site selection to ensure
adequate treatment system installation.
                                   31

-------

-------
                              SECTION 4

                       SUBCATEGORIZATION REVIEW
Basis for Subcateqorization

Factors Considered

The inorganic chemicals industry is very large and diversified and has
been segmented into subcategories  for  the  purpose  of  establishing
effluent   guidelines.    Factors   taken   into   consideration   for
subcategorization include:   raw  materials  used,  product  produced,
manufacturing process employed, geographical location, size and age of
equipment  and  facility  involved, non-water-quality aspects of waste
characteristics,  water pollution control technology, treatment  costs,
energy   requirements  and  solid  waste  disposal.    Following  is  a
discussion  of  each  of  the  general  factors  considered  for  this
industry.

A.   Raw Materials

     Different raw materials are used to manufacture a wide variety of
     products, and vary from raw  brines  and  ores  to  pure  reagent
     chemicals.   Some  processes use waste or by product streams from
     other plants or from other processes within the same plant.

     Because of this  diversification,  raw  material  characteristics
     generally    do    not    constitute    a   logical   basis   for
     subcategorization.  Variations in raw material quality or  purity
     are  not  normally  sufficient  to  cause  a  great difference in
     wastewater treatment needs, except in the  case  of  trace  toxic
     materials which may occur in some sources but not in others.

B.   Dominant Product

     Subcategorization by chemical  name  of  the  dominant  inorganic
     chemical  produced  involves  the  least  ambiguity  in  applying
     standards to a given point source.  This is critical  because  of
     the  great  variety  of  product  mix,  manufacturing  processes,
     wastewater consitutents, and other factors  at  existing  plants.
     Subcategorization  by  product becomes less useful as product mix
     increses in  complexity  because  multi-product  wastewater  also
     becomes  more  complex  and  less  susceptible  to simple uniform
     treatment.


     A subcategory established on the basis  of  product  manufactured
     might  have  two or more different processes but, in the majority
     of cases, the characteristics of the wastewaters are similar  and
     the  same  treatment  technology  can  be  applied  for different
     process wastewater waters.  If two or more  dissimilar  processes
                                    33

-------
D
     produce  wastewater of different quality, and different treatment
     technologies have to be used, then  the  subcategory  has  to  be
     further  classified  or  segmented, for example, the Chlor-Alkali
     Industry.

     Manufacturing Process

     Typically,  inorganic chemicals are manufactured  for  captive  or
     merchant  use in four or more steps starting from raw material to
     final product.  Two or more different products might use the same
     process but  then  the  raw  materials  used,  process  sequence,
     control,  recycle  potential,  handling, and quality control will
     vary,   producing   wastes   of   different   quality.    Primary
     subcategorization,   therefore,  by  process  is  unlikely  to  be
     useful
         However
     necessary in some
secondary
cases.
subcategorization by process has been
Geographical Location
     Inorganic chemical plants exist in all parts of the United States
     but  subcategorization  on  this  basis   is   not   appropriate.
     Geographical  location  is important in analyzing the feasibility
     of  various  treatment  alternatives.   Evaporation   ponds   are
     functional  only  in  areas  where  evaporation exceeds rainfall.
     Ocean dumping and deep well disposal are possible only in certain
     areas, and must be consistent with local, State and Federal laws.
     The possibility of ground water contamination  may  preclude  the
     use of unlined holding and settling ponds in many locations.

     In  the northern regions, climatic conditions may necessitate the
     inclusion of special provisions to prevent freezing of  treatment
     system   components,  particularly  biological  oxidation  units,
     clarifiers, ponds, and open collection  systems.   The  costs  of
     utilizing  waste  heat  sources  from  the  process  or providing
     various types of thermal protection, such as insulation or burial
     of pipes and tanks and  building  structural  shelters,  may  add
     considerably  to  the  capital  and  O&M  cost  associated with a
     treatment technology.

     Thus, the influence  of  geography,  climate,  geology,  etc.  is
     reflected,  in  waste  treatment  modifications  anql  is primarily
     manifested in the cost of treatment.  This, of itself, is  not  a
     good basis for subcategorization.

     Plant Size

     Plant size and production capacity were not found to  affect  the
     characteristics  of  the waste produced.  Although plant size can
     affect  treatment  cost,  this  variability  can   be   expressed
     graphically  or  mathematically  without  the  need  for  further
     segmentation of the category.
                                    34

-------
H
Plant Age

Plant age can have an important bearing on wastewater volume  and
quality  and  is,  therefore, a significant factor to consider in
evaluating  the  applicability  of  treatment  technologies   and
assessing  the  relative  costs of treatment for plants of widely
differing  age  producing  the  same  or  similar  products.    A
particular  problem  with  older  plants  is  that  their present
patterns of water use may have evolved over a long period of time
with little consideration for the principles of  efficient  waste
segregation,  collection,  and  treatment.   To a limited degree,
plant modernization can correct or  at  least  mitigate  some  of
these  shortcomings  in  older  facilities,  however, only a small
proportion of the cost of  revamping  collection  systems  or  of
converting  from  contact  to  noncontact  cooling systems can be
offset by the resulting lower cost  of  treatment.   In  general,
older  plants,  even  after  considerable modernization, normally
have a higher volume of wastewater flow and higher waste loadings
(although pollutant concentrations  may  be  lower  due  to  poor
segregation  from noncontact sources) in comparison to relatively
new  plants.   The  present  and  forthcoming  requirements   for
pollution  control  may impose a severe treatment cost penalty on
older plants due to the need for backfitting  and  replumbing  of
outdated  collection  systems.   Land  availability  and land use
restrictions are also factors which  may  translate  into  higher
treatment  costs  for  older  facilities  which  find  thenselves
surrounded by highly developed industrial and residential areas.

Unfortunately, plant age does  not  readily  lend  itself  to  an
unambiguous  definition where a series of plant modifications has
taken place.  The extent of  modifications  also  varies  greatly
among  plants within the same product industry.  For those plants
that have been enlarged or modified from their  original  status,
plant age is not unambiguously calculable and therefore plant age
is not a reasonable basis for subcategorization.

Non-Water-Quality Characteristics

Airborne emissions from  manufacturing  operations  can  be  kept
within  air  quality  control limits through the use of cyclones,
wet scrubbers and other methods.  The nature of the air pollution
is  related  to  the  product(s)  manufactured  and/or  the   raw
material(s)  used.   Since  both  of  these  elements vary widely
within the inorganic chemicals industry, there  is  no  logic  in
subcategorization    on    the    basis    of   non-water-quality
characteristics.

Treatment Cost

From  a  technical   viewpoint,   subcategorization   by   common
technological  requirements for treatment processes could provide
a logical basis for selecting  one  or  more  unit  processes  to
accomplish  the same treatment function, regardless of the source
                                  35

-------
     of the wastewater.   For example,   residuals  of  dissolved  heavy
     metals  will  respond  to lime precipitation and sedimentation at
     high pH without respect to the specific  origin  of  the  metals.
     This   "building  block"  concept  could  conceivably  result  in
     selecting various combinations of  unit  processes  to  meet  the
     treatment  requirements.   However,  if the treatment cost must be
     expressed in terms of dollars per unit production, this method of
     subcategorization  crosses  product   lines  and  interferes  with
     comparison  of  treatment  costs  based  on  the  production of a
     specific chemical.    Even  if  the  unit  operation  is  commonly
     applicable  for  treating  waste flows of different products, the
     cost  of  treatment  will  fluctuate  because  of  variations  in
     quality,  loading  and  flow  rates   and subcategorization on the
     basis of treatment cost is not recommended.

I.    Energy Cost

     Manufacturing  processes  in  the  Inorganic  Chemicals  Industry
     typically   have   large   energy   requirements.   In  contrast,
     wastewater treatment processes consume a small  fraction  of  the
     total   energy  used.   There  appears  to  be  no  major  energy
     requirements  for   the   wastewater   treatment   facility   and
     subcategorization on the basis of energy cost is not justified.

J.    Solid Waste

     Not all inorganic manufacturing processes produce  solid  wastes.
     Solid  waste producers practice various disposal methods, such as
     on-site landfills,  contract hauling  to  approved  dump  sites  or
     incineration.   Solid  waste  disposal'becomes very site specific
     and exhibits a wide range of  costs.   Because  of  the  lack  of
     uniformity  within  the  industry,  solid  waste  generation  and
     disposal   proctices   are   not   a   satisfactory   basis   for
     subcategorization.

General Conclusions

If  effluent  limitations  are to be tied to units of production, only
one method of primary subcategorization is broadly applicable  to  the
inorganic  chemicals  point  source  category;  viz.,  subdivision  by
dominant product.  However,  there  are  three  subcategories,  Chlor-
Alkali,  Titanium  Dioxide, and Hydrogen  Cyanide which require further
subdivision based on the difference in the quanity and quality of  the
wastewater  from  the processes, and two others, Hydrofluoric Acid and
Aluminum Fluoride, have been reviewed for  possible   integration  (see
Section 4.3).

Secondary Subcateqorization

Chlor-Alkali
                                    36

-------
Mercury and diaphragm cells are the two distinct types of electrolytic
cells  that  are  used in the production of chlorine and caustic soda.
Major process differences between  mercury  cell  and  diaphragm  cell
plants  produce  corresponding differences in the volume ,and nature of
wastewater generated.  A  principal  difference  is  the  presence  of
mercury  as  a  contaminant  in  the wastewaters from the mercury cell
process and asbestos in the diaphragm cell  plants  wastes.   The  TSS
discharges  from  diaphragm cell plants are generally larger than from
mercury cell  plants,  due  to  the  higher  volumes  of  contact  and
noncontact  water  used.    Also,  in diaphragm cells a large amount of
water is produced in the caustic evaporation process.  Such  water  is
not  produced  in  mercury  cell  plants.   The quantity of wastewater
generated from the diaphragm cell plants is several times that of  the
mercury  cell plants for the same chlorine production capacity.  Based
on  the  quantity  and  characteristics  of  the  wastewater,  further
subcategroization is justified.

Titanium Dioxide

Two  major  ores, rutile and ilmenite, are used for the manufacture of
titanium dioxide.  The ilmenite ore contains  40-70  percent  titanium
dioxide  (Ti02),  up to 35 percent ferrous oxide (FeO), and 25 percent
ferric oxide (Fe203).  Rutile ore contains more than 90 percent  Ti02.
Two  processing  techniques,  the  sulfate  process  and  the chloride
process, are used to extract titanium dioxide from the ores.

The sulfate process  uses  ilmenite  ore  and  sulfuric  acid  as  raw
materials.  The chloride process uses rutile ores or ilmenite ores and
chlorine,  with a different process and wastewater characteristics for
each ore.  The high grade rutile ore is expensive and its availability
is declining.   In  recent  years,  new  technological  advances  have
alleviated  the  raw  material  shortage  problem.   By  upgrading the
ilmenite ore quality, the chloride process  can  be  used  to  produce
titanium dioxide of high purity.  Because of the difference in quality
and  quantity  of  wastewaters generated from the sulfate and chloride
processes using the two different ores, the titanium dioxide  industry
may be further subdivided into three segments as follows:

     1.   Sulfate process
     2.   Chloride process using rutile ore
     3.   Chloride process using ilmenite ore (one step).


The sulfate  process  generates  large  amounts  of  strong  and  weak
sulfuric  acid  water-borne  wastes.  Application of pollution control
technology to the acid wastes  generates  about  five  times  as  much
gypsum  as  product.   The  chloride process generates large amount of
dissolved metal chlorides and the treatment technology   is  expensive.
Solid  wastes from both processes present difficult disposal problems.
These solids include ferrous sulfate  (FeS04) and a hydrated by-product
from the sulfate process and heavy metal  sludges  from  the  chloride
process.  Ilmenite ore has to be upgraded before it is used to extract
                                    37

-------
titanium  dioxide  by  the  chloride  process,
process step generates additional wastes.
        and  this  beneficiation
The application of the chloride process to ilmenite ore may proceed in
either one or two steps.  A  patented  one-step  process  accomplishes
both  beneficiation  and chlorination of the ore in a single fluidized
bed reactor and generates raw waste  loadings  which  are  similar  to
those  from  the  sulfate  process in terms of acidity and metals/ and
similar to wastes from the chloride-rutile process in terms  of  spent
coke  solids  and  still  redisues.   In  the  two-step  process,  ore
beneficiation resulting in either a synthetic rutile  or  an  enriched
titanium  oxide  slag  is  carried  out  separately at the mine or the
plant.  The discharge of wastewater  generated  by  the  beneficiation
step would be regulated under the Ore Mining and Dressing Point Source
Category and will not be considered in this document.  The second step
of  the  two-step  process  generates  wastes that are very similar in
quantity and quality to those from  the  chloride-rutile  process  and
will  be governed by the discharge regulations for that segment of the
Ti02 subcategory.

Therefore,  further  subclassification  based  on   the   amount   and
characteristics  of  the  wastewater  appears to be justified, and the
three process subdivisions indicated above are  appropriate  for  this
purpose.

Hydrogen Cyanide

Hydrogen  cyanide  (HCN)  is  made  from  two different processes/ the
Andrussow process and as a by-product  of  acrylonitrile  manufacture.
In the Andrussow process, air, ammonia, and natural gas are reacted to
produce the dominant product hydrogen cyanide.

Water-borne wastes from the process consist principally of ammonia and
sulfa.tes in addition to cyanide and nitriles.

The  primary  product  in the other process is acrlonitrile (CH2=CHCN)
and the hydrogen  cyanide  is  a  by-product.   Because  the  hydrogen
cyanide  is  a by-product, it will be covered in the organic chemicals
manufacturing category with the primary product.
Review of Possible  Integration of Subcateqories

Hydrofluoric Acid and Aluminum Fluoride

Aluminum fluoride (A1F3)  usually   is  produced  by   the   reaction  of
hydrated  alumina (A1203»3H20) with hydrogen fluoride (HF).  One plant
produces aluminum fluoride from  fluorosilicic  acid   (H2SiF6),  a   by-
             phophoric  acid   (H3PO,
and  is  not
product  of
regulation.  Two of the aluminum  fluoride  plants  are
integrated with hydrogen fluoride (or hydrofluoric acid
covered  by  the
   known  to  be
   production.
                                   38

-------
The  two  major  uses  of  hydrogen  fluoride  are in the fluorocarbon
industry and as raw material in the manufacture of aluminum  fluoride.
A  ban  on  the  fluorocarbon  propellants  has  curtailed  the use of
hydrogen fluoride in that industry and it was  completely  stopped  in
1978.   The  selling  of  hydrogen fluoride in the merchant market has
declined and the primary use is limited to the production of  aluminum
fluoride  and  fluorocarbon  plastics  until  some  other major use is
found.

For both products (HF and A1F3), process wastewaters are generated  by
the various gas scrubbers and by leaks and spills.  In both cases, air
pollution  control scrubber effluents contain mainly fluoride, acidity
and sulfate.  The fluoride is present as  the  free  ion  as  well  as
various  complex fluoro anions.  Calcium fluoride (CaF2), generated as
a solid waste, is  a  disposal  problem  for  both  the  subcategories
because  of  its  moderate toxicity.  Only one additional solid waste,
gypsum  (CaS04«2H20),  is  generated  from   the   hydrogen   fluoride
manufacture alone, and it can be treated and handled independently.

Combining  hydrofluoric  acid  and  aluminum  fluoride   into  a single
subcategory does not appear to offer any  regulatory  advantages  when
the  two  products  are  manufactured at the same plant  location.  The
wastewaters associated with the two products are similar and a  common
treatment  facility  is normally utillized.  In addition, the combined
manufacture of these products does not  create  a  unique  or  unusual
situation, either with regard to the wastewater treatment requirements
or  compliance  with discharge regulations.  Although the waste gypsum
produced at an HF plant supplies enough calcium for adequate  fluoride
removal from neutralized scrubber wastewaters generated  by both HF and
A1F3  production,  the applied treatment technology is essentially the
same as  that  applied  by  manufacturers  of  either  product  alone.
However,  the  effluent  water quality, wastewater flow, and the toxic
polllutant loadings are not the same.  Further, the opportunities  for
drip  acid  recycle  (or  the hydrolysis of complex fluorides prior to
treatment), and scrubber water recycle are a function of  plant  design
and age, rather than product mix.
In  view  of these considerations, a recommendation for the creaton of
an HF/A1F3 combined product subcategory is  not  being  made  at  this
time.

Summary

The  recommended  subcatetjorization  with process subdivisions include
the following:
            Subcateqory

          Chlor-Alkali

          Titanium Dioxide
   Process Subdivisions

Mercury Cell
Diaphragm Cell
Sulfate
Chloride-Rutile
                                  39

-------
Hydrogen Cyanide
Chloride-IImenite
Andrussow Process
Acrylonitrile By-Product
  {Included in the Organic
  Chemicals Regulation.)
                         40

-------
                              SECTION 5

             SCREENING AND VERIFICATION SAMPLING PROGRAMS
                        Scope and Methodology

The specific objective of the sampling programs was to  establish  the
extent of the required regulation of toxic pollutant discharges in the
inorganic  chemical  industry  in terms of factual information derived
from the chemical analysis  and  flow  measurement  of  representative
process  raw  wastewater streams and treated effluents.  Prior to this
study, most of the information available on toxic pollutants has  been
concerned  with  a  relatively  small  number of known process-related
substances contaminating a variety  of  direct  and  indirect  contact
process  waters discharged from a production facility.  There had been
no previous requirement  for  a  comprehensive  survey  of  wastewater
chemistry  addressing  the  possibility  that  a large number of other
potentially toxic substances could be present, albeit at extremely low
concentrations.

The screening phase of the sampling program was designed to  ascertain
the  presence  in  each  subcategory  of  any  of the 129 listed toxic
pollutants at raw waste concentrations or  daily  loadings  which,  if
untreated, could be environmentally significant. Screening is based on
the  sampling  of one or more typical manufacturing operations in each
subcategory.  Where significant pollutant concentrations  were  found,
additional  plants  were  sampled  during  the  verification phase for
confirmation and further quantification  of  data  on  the  particular
toxic  pollutants  in  question.   A  goal  was  set for screening and
verification sampling of a sufficient number of plants to account  for
at least 75 percent of the total U.S.  production, in each subcategory
having significant concentrations of priority pollutants.

A  detailed description of the screening and verification programs are
presented in the paragraphs below.
Selecting Plants and Making Preliminary Contacts

In each subcategory, plants were selected for screening on
of the following general criteria:
the  basis
     A.   Minimal product mix and no organic product lines which could
          increase the potential for interprocess cross  contamination
          of wastewaters.

     B.   Presence of a physical chemical  treatment  facility  rather
          than  a biological one, or no treatment system.  (Biological
          systems are neither widely used nor generally applicable  in
          the inorganic chemicals industries.)
                                   41

-------
     C.    Manufacture of industrial grade products in
          than low volume reagent grade products.
volume,   rather
     D.    Median production capacity with the subcategory.

     E.    Segregated waste streams to facilitate sampling.

     F.    NPDES discharges rather  than  discharges  to  POTWs,  since
          treatment for a NPDES discharge is usually more extensive.

     G.    Geographical clustering of  selected  plants  to  facilitate
          field  logistics,  but only to the extent that other factors
          are equal.

Preliminary phone contacts were made  with  plant  representatives  of
those  facilities which satisfied the above criteria.  If requested, a
letter was written to describe the objectives of the sampling  program
and  to  cite  the  legal  authority  of  the  Agency and its sampling
contractor under Section 308 of the Federal  Water  Pollution  Control
Act  Amendments  of  1972.  Information provided by industry for which
confidential treatment was requested has been  handled  in  accordance
with the provisions of 40 CFR Part 2.

Prior  to  the  actual  sampling of waste streams, a lead visit to the
selected plant was made to gather background information, confirm  and
update  any  308-Questionaire  responses,  and  to  obtain  additional
technical  information  regarding  processes   and   waste   treatment
practices.   Sampling  sites were selected and described relative to a
detailed waste source inventory and a flow diagram of the process  and
waste  treatment  system,   Arrangements  were made for the subsequent
sampling visit and the details of the lead visit  and  sampling  point
descriptions were documented in an interim report to the Agency.

Screening and Verification Sampling

A.   Collection of Samples for Screening

     In the screening phase of  the  sampling  program,  the  specific
     objective  was  the  detection  and  quantification of waterborne
     waste constituents included on the list of 129  toxic  pollutants
     (Table  3-1).   Each  sample of an individual raw waste stream, a
     combined waste stream, or a treated effluent was collected  where
     possible  by  an automatic, time series, compositor over a single
     72-hour sampling period.  Where  automatic  compositing  was  not
     possible,   grab  samples  were  taken  at  approximately  2-hour
     intervals  during  the  same  sampling  period   and   composited
     manually.

     Each  sample  was divided into several portions and preserved, as
     required for different types of analysis, in accordance with  the
     procedure  established  by  EPA  (4) for the measurement of toxic
     pollutants.
                                   42

-------
     Samples were also taken from the  composites,  or  as  individual
     grabs,  for  the analysis of the conventional and nonconventional
     pollutants.

     Volatile organics  were  collected  in  teflon-sealed  screw  cap
     vials.   Eight  40  ml vials were filled at each sampling site by
     grab sampling in pairs at approximately  2-hour  intervals.   The
     individual vials were cooled to 4°C and shipped to the laboratory
     where  they  were  used  to  prepare composites in duplicate just
     prior to analysis.  Three blank vials prepared and sealed in  the
     laboratory  accompanied  each  set  of samples during collection,
     shipment,  and storage.

B.   Collection of Samples for Verification

     The objective of verification sampling was to confirm  the  first
     observations    from   screening   and   further   quantify   the
     concentrations and waste loadings of  the  toxic  pollutants  and
     conventional  and  nonconventional  pollutants.   Where any toxic
     pollutant metals  were  found  during  screening  sampling  of  a
     particular  plant,  analyses  were  made  for all toxic pollutant
     metals during the verification sampling

     The established protocol for verification sampling  required  the
     collection  of  three  24-hour composites at each sampling point.
     Again,  where  composites  could  not  be  taken  with  automatic
     samplers,  grab samples were taken periodically over the same time
     period and composited manually.

C.   Sample Shipping

     All samples, individually labeled, were placed in  large  plastic
     bags,  which were then placed in a waterproof, insulated shipping
     container.  Enough ice was included to maintain a temperature  of
     approximately 4°C during shipment to the laboratory.

     Containers  were shipped by the best available route, usually air
     freight, usually arriving at the laboratory on the same day,  but
     occasionally  taking  overnight.   Upon receipt, all samples were
     immediately placed in a walk-in refrigerator maintained at 4°C.

     In order to maintain the chain of custody and to  keep  track  of
     samples,  sampling personnel kept logs of samples taken in ink in
     page-numbered, hard-bound books.   The  data  recorded  included:
     date,  time, plant code, number, sample type, and sampler.  Prior
     to their arrival at the laboratory, a list  of  samples  shipped,
     including  number,  type of samples, and analysis to be performed
     was sent to each department supervisor to alert him  of  incoming
     work.

     A  master analytical control chart was maintained which included:
     the date the sample was received, date due, number  and  type  of
     each sample, and the analysis required.
                                     43

-------
     At  the time of analysis,  the individual samples were distributed
     to the analytical chemists along with a list which included:   I.
     D.   number  of  sample,   type of sample,  analysis required, date
     samples were received,  and due dates.

     All samples were kept in a laboratory refrigerator  at  4°C  when
     not  being  handled by the analyst.   Upon completion of analysis,
     the sample was checked back into the  Sample  Control  Department
     and  kept  in  an  identified  location  in  the  Sample  Control
     refrigerator.  A report of completed samples was then sent to the
     EPA Sample Control Center.

D.    Verification Sampling Plant Selection

     After the decision was made  to  verify  the  presence  of  toxic
     pollutants  found in the screening of a subcategory, verification
     plants were selected.  The basis for  selection  was  essentially
     the same as that used in selecting screening plants.

     The  screening  program  results were evaluated to identify those
     toxic pollutants  that  were  present  at  potentially  treatable
     concentration  or  daily  loadings.    Concentrations  or loadings
     which could be reduced by the highest quality  treatment  systems
     were considered significant.  Two situations occurred:

     1.   A subcategory which had a potentially  treatable  raw  waste
          concentration  of  any  toxic  pollutant would be subject to
          verification  sampling,  and  BAT-based  regulations   would
          likely  be  proposed  by  the  Agency  for  the treatmnt and
          control of that toxic pollutant.

     2.   A subcategory which had no potentially treatable  raw  waste
          concentration of any toxic pollutant would not be subject to
          verification  sampling  and  would  likely  be excluded from
          regulatory coverage  in accordance with  the  provisions  for
          exclusion under Paragraph 8 of the Settlement Agreement.

In  analyzing  screening  data,  only those pollutants attributable to
process sources were considered.  Pollutants which result from cooling
tower  operations,  corrosion  or  'corrosion   control,   control   of
biological  growth,  or  any  other operation not directly tied to the
production process, were not used as a basis for verification.

The number of plants selected for verification in each subcategory was
roughly  proportional  to  the  number  of  existing  plants  in  that
subcategory  with  a  maximum  of  five  plants  selected.   In  small
subcategories  (relatively few production facilities),  an  effort  was
made   to  select  a  sufficient  number  of  plants to account for the
majority of the total U. S. production.


When the verifiction phase of  the program was initiated, an  important
decision  was  made with regard to metals analysis.  First, in view or
                                  44

-------
the frequent presence of metal contamination in the  wastes  screened,
and  the inability in some cases to show a direct relationship between
certain metals found and the known process chemicals or the  materials
of construction, it was decided that all 13 of the toxic metals should
be  determined  again  during verification, regardless of whether they
were found in screening.  This was intended to  provide  a  much  more
complete  data  -base  than  would  be obtained by running verification
analyses for only those  metals  found  in  screening  to  exceed  the
verification criteria levels at the time of sampling

Analytical Methodology for Toxic Pollutants
The  analytical  protocol  for the screening and verification of toxic
pollutants was established in Sampling  and  Analysis  Procedures  for
Priority   Pollllutants  by
Environmental Monitoring and
April, 1977.
U.  S.  Environmenal  Protection  Agency,
 Support  Laboratory,  Cincinnati,   Ohio,
The   specified   analytical   methodologies   were  employed  without
modification except where noted below in connection with toxic  metals
analysis during verification.

Implementation  of  the  methodology  and quality assurance provisions
required the establishment of  special  sample  handling  and  control
procedures  specifically  suited  to  each  type  of  analysis.  These
procedures, together with a discussion  of  the  achievable  detection
limits  for  each  parameter  or  group  of  similar  parameters, .are
presented in the following paragraphs.

A.   Trace Metal Analysis

     Figure 5-1 shows a data flow diagram for metals analysis.  Atomic
     absorption methods described in 40 CFR 8 136 per  Section  304(h)
     were  used.   A  set  procedure was followed in the laboratory to
     generate the analytical values and the quality control data.  The
     data  flow  diagram  shows  the  actual  sequence   employed   in
     verification analysis and the following notes, which are keyed to
     the diagram, provide additional information on the procedures.

     1.   Blanks—two for each set of analyses digested.
          Duplicates—one every seventh sample.

     2.   Quality Control at Operator Level (Atomic Absorption):

          Blanks—These were run at the beginning and the end of every
          set analyzed for each metal.  Also, air blanks were  run  on
          furnace, or heated graphite atomizer (HGA), after any sample
          with a large positive value.


          Standards—Three  different  concentrations  were run at the
          beginning and end of every  set  analyzed  for  each  metal.
                                    45

-------
3.

4.
Standards  were  also  run  every  tenth  sample  during the
analysis of a set.

Spikes—These were made according  to  the  EPA  "Method  of
Standard  Additions," by adding such a volume of standard as
to double the apparent concentration of metal present in the
sample.    Extrapolation   backwards   of   the    resultant
absorbances  allowed  correction  of  absorbance  for matrix
effects.

Duplicates—For furnace analysis, the sample was  run  twice
wherever  a  low  but  positive absorbance was obtained.  As
well as this, one sample in every seven was run in duplicate
routinely.  The average of duplicate  measurements  was  the
taken  value;  the difference between duplicate measurements
was   noted   and   recorded   on   control   charts.     If
reproducibility  was  outside the limits of ±33 percent, the
measurement was repeated.

UTD = "Unable to Determine" due to matrix interferences.

Criteria Employed in Spike Selection:  samples  were  chosen
to be spiked based upon the following criteria:

     All   samples   where  there  was  any  suspicion  that
     interference or matrix effect was present

     All samples containing a  measurable  concentration  of
     analyte

     In addition, at least one sample in every seven.

The  level  of  spike chosen was controlled by the following
factors:
          It   should   approximately   double
          concentration
                                             the
apparent
          If  this  results in an absorbance greater than that of
          the highest standard, the  spiked  sample,  is  suitably
          diluted with distilled water.

     A  reagent  blank  was  run  with each set of spiked samples
     prepared.

During the screening phase of the sampling program, the  standard
protocol followed for metals analysis was:


1.   The 13 priority pollutants metals,  with  the  exception  of
     mercury,  were  determined  by  AA  spectrophotometry in the
     furnace  (HGA) mode.
                                46

-------

t STANDARDS 1 	 J


i
FURNACE
t Cd.Pt>.SI>.Tt 1
i
i ;
UTO DETERMINATE
(3) VALUE
I
FLAME ANALYSIS
I
* f *
UTD DETERMINATE
O) VALUE
1
FLAME ANALYSIS 	 -""'^

FIELD SAMPLING
I
PRESERVATIVE ADDED
•CEO . AND AIR BHJPED
I
RECEIPT . LOG IN SAMPLES
ANO REFRIGERATE
1
QUALITY CONTROL BLANKS AND
DUPLICATES CREATED (1)
V
PREPARATION 8V
ACK) DIQEBTIOH
I
ATOMIC ABSORPTION ANALYSIS (»)

__1 	




FLAME VAPOR QENERTION
1 * , _ "*
I i ^—ii 	

"~t
OFF-SCALE DETERMINATE OFF-SCALE DETERMINATE
RESULT VALUE RESULT VALU6
t \ _
DILUTE . DILUTE
SAMPLE SAMPLE

-^ X
•^ . » . x/
CALCULATES t X/ SELECT** Of
,„ ANALYSIS OF ' ........fl TO nc
- RESULTS 	 •• 8AMM-fc" ' ,
^^- i 1 	 	 1
/ FtMAt NV
<^ INSPECTION. IS ^> N° 1. JEUISfSk
N^ DATA VALID ^X BEOWMO
1 YES
TYPING I
MPOOTttta





*
HYDRIDE GENERATION
I A*,S« >
*
* *
OFF-BCALE DETERMINATE
RESULT VALUE
1
DILUTE
SAMPLE



Figtire 5-1.  Sample flow sheet for metal analysis.

-------
     2,    If matrix  interference  was  seen,  they  were  spiked  and
          redetermined.

     3.    If difficulties due to matrix interference persisted, or  if
          metal concentrations appeared high enough, the determination
          was repeated in the flame mode.

     4.    Mercury was determined by the standard cold vapor method.

Certain  changes  in  analytical  protocol  were   instituted   during
verification   analysis   in  order  to  avoid  the  excessive  matrix
interference experienced during screening  when  the  heated  graphite
atomizer (HGA) was the primary method applied to the analysis of 12 of
the metals.   The modified protocol for metals was:
     1.    Six elements were determined by flame only, namely, Ag,
          Cu, Cr, Ni and Zn.
Be,
     2.   Four elements were determined by furnace (HGA), namely,  Cd,
          Pb,  Tl and Sb.  If interference occurred,  Cd, Pb, Tl and Sb
          were determined by flame.

     3.   Hg was still analyzed by the cold vapor method.

This modification reduced the number of preparations per  sample  from
three  to  two and achieved adequate detection limits which were still
well below the verification criteria levels.

Additional modifications were made during the verification program  to
improve  the  reproducibility  and precision for Hg,  As and Se.  These
were:

     1.   The cold vapor procedure for Hg was  modified  to  eliminate
          the  pump and allow dilution and rerun from the same sample.
          This saved time and increased reproducibility.

     2.   Selenium and arsenic were determined by  hydride  generation
          using  sodium  borohydride  (NaBH4).  This greatly minimized
          problems associated with matrix interference.  The method is
          very reproducible and the detection limits  were  at  levels
          well below the verification criteria for these two elements.

     After  the  above  modifications  were adopted,  screening samples
     which originally were  unable  to  be  analyzed,  or  which  were
     recorded  as  below  excessively high detection limits due to the
     effects  of  matrix  interferences,  were  rerun.    Satisfactory
     results were then obtained in nearly all cases due to the greatly
     improved sensitivity and reproducibility.

     It  should  be  noted  that these modifications of the analytical
     protocol  were  in  the  direction  of  improved  precision   and
     reproducibility  and  not  towards  lower  detection limits.  The
     original screening procedures generally  had  a  lower  detection
                                  48

-------
     limit  when  it  was  achievable.   However, the methods were too
     susceptible to giving no result at all  with  complex  industrial
     matrices,    and   so   the   revised  protocols  sacrificed  some
     sensitivity  for  precision  and  reproducibility.     The   final
     detection  limits  were  still in the subparts per'million range.
     See Table 5-1  for a summary of these limts, together  with  those
     of the original protocol.

B.   Organic Compound Analysis

     The organic toxic pollutants  were  determined  by  the  standard
     protocol  (40 CFR 8 136 proposed December 3, 1979}  which includes
     sample preparation,  extraction,   and  analytical  methodologies.
     Extractions were carried out using methylene chloride in the case
     of   the   acid  and  base/neutral  organic  fractions  and  with
     hexane/methylene  chloride  to  obtain  the  pesticide-containing
     fractions.    The  acid and base/neutral fractions were reduced in
     volume  and  analyzed  by  gas  chromatography-mass  spectrometry
     (GC/MS).   The  pesticides  were analyzed by electron capture gas
     chromatography  followed  by  GC/MS  confirmation   of   positive
     results.   Volatile  organics were analyzed by the purge and trap
     method of introducing the material into the GC/MS inlet system.

C.   Cyanide Analysis

     The standard methods for  the  wet  chemical  analysis  of  total
     cyanide  and  cyanide  amenable  to chlorination (Cyanide A) were
     utilized (40 CFR S 136).   Cyanide analysis is subject to  several
     sources of  interference,  including:


     1.   Metals-The presence of Fe, Cd, Ca, Ni, Ag,  and Zn may  cause
          low  results  due  to the formation of stable complexes with
          cyanide.   The iron complexes may form insoluble precipitates
          which  are particularly difficult to break  up  both  at  the
          time  of alkaline chlorination of the sampled wastewater and
          during the chemical analysis for cyanide.

     2.   Oxidizing  agents-The  presence  of  free  chlorine  in  the
          wastewater   sample  will  destroy  cyanide  and  cause  low
          analytical  results.    The  addition  of  ascorbic  acid  to
          destroy  chlorine  at  the  time  of sampling is intended to
          mitigate this  problem.    Other  oxidizing  agents  such  as
          peroxides  and chromates may also react with cyanides over a
          period of time and cause low results.

     3.   Sulfides-Sulfide or bisulfide will interfere in the analysis
          of cyanide by reacting with the colorimetric reagents.

The presence of  sulfur dioxide or bisulfite in the  wastewater  sample
should  have  no  appreciable  effect  on  cyanide results.  Detection
limits on the order of 1-4 ^g/1 can  be  achieved  by  the  analytical
                                    49

-------
method employed, but the results have to be interpreted with regard to
the possible interfering components of the sample.

D.   Hexavalent Chromium (Cr VI) Analysis

The determination of Cr VI in wastewater samples is also subject to  a
number  of  interferences which can take effect either during sampling
and storage or during analysis.

     1.    Acids-Samples taken and held at a very low pH can experience
          the conversion of other forms of chromium into Cr VI causing
          a positive interference.

     2.    Reducing   agents-Samples   containing    sulfur    dioxide,
          bisulfide,  sulfide, ferrous iron, and other reducing agents
          will result in low values of  Cr  VI  by  converting  it  to
          trivalent  chromium   (Cr  III).   Under these conditions the
          chromates originally present would be included in the  total
          chromium   determination  but  the  analytical  results  for
          hexavalent  chromium  would  be  proportionately  low   (see
          Reference 52).

The   detection   limits   for   Cr  IV  using  the  diphenylcarbazide
colorimetric method are on the order of 1-3 */g/l  in  the  absence  of
substances which interfere with color development.

E.   Asbestos Fiber Analysis

The analysis of selected samples for asbestos fiber   (chrysotile)  was
conducted  by  the  recommended method utilizing transmission electron
microscopy with selected area  electron diffraction as described by Dr.
Charles Anderson (EPA, Athens, Georgia)  at  the  Analytical  Protocol
Meeting in Denver (November, 1977)  (56).

F.   Conventional and Nonconventional Pollutants


All techniques  used  for  the  analysis  of  BPT  control  parameters
 (conventional  and  nonconventional pollutants) were  those recommended
by the Agency-  The list of approved test procedures  was published   in
the Federal Register on October 16,  1973  (38 FR 28758) and may also  be
found in Title  40 of the Code  of Federal REgulations  (40 CFR  136).

Quality .Assurance Provision

The  Agency  and  the   contractor's  analytical   laboratories maintain
consistently high standards for accuracy and quality  control.   As   an
 in-house  requirement,  a  minimum  of  ten percent of all samples are
routinely run  in duplicate.  Quantitation is based on standards  which
are  prepared   in   pure water, at  concentrations such that all sample
measurements are greater than  the absorbance of the   lowest   standard,
and   less   than  the absorbance of the highest standard.  The  standards
are also checked by participation in the EPA Reference Sample  Program
                                    50

-------
           TABLE 5-1.   ANALYTICAL DETECTION LIMITS FOR METALS
                                                             CD
Element
Antijnony, Sb
Arsenic, As
Beryllium, Be
Cadmium, Cd
Chromium, Cr
Copper, Cu
Lead, Pb
Mercury, Hg
Nickel, Ni
Selenium, Se
Silver, Ag
Thallium, Tl
Zinc, Zn
Original Screening
Protocol t21
Method (>jg/l)
HGA*
HGA
HGA
HGA
HGA
HGA
HGA
Cold Vapor
HGA
HGA
HGA
HGA
HGA
10
3
0.2
1
1
1
10
0.5
1
9
0.5
2
1
First Modification Second Modification
of Protocol ^ of Protocol (4)
Method (jag/1) Method ; (>ag/l)
HGA
HGA
Flame
HGA
Flame
Flame
HGA
Cold Vapor
Flame
HGA
Flame
HGA
Flame
10
3
15
1
25
20
10
0.5
25
9
15
2
25
HGA
Hydride
Flame
HGA
Flame
Flame
HGA
New Cold
Vapor
Flame
Hydride
Flame
HGA
Flame
10
10
15
1
25
20
10
0.5
25
10
15
2
1
    Heated Graphite Atomizer

(1)  Assuming no matrix interferences  requiring dilution of  sample.

(2)  EPA Contract ND.  68-01-4492  (September  29, 1977), Exhibit C,
    "Protocol  for  the Measurement of  Toxic  Substances", Environmental
    Monitoring and Support Laboratory, Cincinnati, Ohio

(3)  June,  1978

(4)  August, 1978
                                     51

-------
that  utilizes  a  double  blind  technique.  {EMSL, Cincinnati, Ohio,
Office of Research and Development.)

Additionally, outside laboratories are retained for checks on  quality
by  analyzing split samples and running submitted standards.  Accuracy
is also insured by analysis of a minimum of  fifteen  percent  of  all
samples  with  spikes by the method of standard additions.  The spikes
are added prior to sample preparation  and  are  carried  through  the
entire sample analysis procedure.

The   contractor's   laboratories  have  consistently  maintained  the
standards for laboratory certification which are imposed by the  State
of   California.    Certification   is  dependent  upon  the  accurate
performance of routine analyses on  check  samples  submitted  by  the
State,  as  well  as  on-site inspections by the State of California's
Sanitation and Radiation Laboratory, Department of Fish and Game,  and
the U. S. Environmental Protection Agency, NEIC, Denver, Colorado.

The  quality  assurance  provisions  outlined   in the EPA Protocol for
GC/MS Analysis of Toxic Pollutnts are rigorously adhered to  with  one
added  precaution, namely, the use of internal  standards as a means of
measuring  recovery.   Although  not  required  by  the  protocol  for
pesticide  analysis, this technique is utilized as an in-house quality
control  requirement  to  ensure  the  accuracy  of  results  in  this
analysis.

The high sensitivity of instrumentation used in trace organic chemical
analysis  dictates that contamination of the samples from any possible
source must be diligently guarded against.   Accordingly,  only  glass
sample  containers  with  Teflon-lined  lids  were used and these were
subjected to a three-step cleaning procedure prior to use, even though
only new liners and glass containers were used.   All  glassware  used
for  sample  preparation and analysis was subjected to a dual cleaning
system.


The sample extraction and preparation rooms are  dedicated  solely  to
toxic  pollutant analysis, and have their own ventilation systems that
are isolated from the other sample preparation  and  receipt  areas  of
the laboratories.

A  documented system of existing practices, including calibrations and
operational checks, is maintained to assure uniformity of  performance
and  to  serve as a basis for alteration of standardizaiton intervals.
A chemist is assigned full time to maintain this system, assure strict
record formating and controls,  and  to  direct  the  quality  control
program of the laboratories.  The primary vehicle of this sytem is the
quality  assurance  manual  containing the detailed procedures used in
sample preparation and analysis,  and  the  complete  records  of  all
quality control standards, blanks, spikes, and  duplicates.

-------
Summary of Analytical Results

The  results  obtained  during the screening and verification sampling
program are summarized in Table 5-2 and Table 5-3.  These tables  show
the frequency and distribution of the pollutants according to selected
plant  groupings, concentration ranges, and subcategories in which the
pollutants occur.

Pollutant frequencies as shown in columns 5, 6, 7, and 8 of Table  5-2
are  based on the highest individual pollutant concentration found for
each plant's raw waste during the screening and verification  sampling
program.

The  toxic  pollutant  asbestos has not been included in either of the
two tables mentioned above.   Asbestos  concentration  is  reported  in
million  fibers  per  liter   (MFL)  which  is  not compatible with the
concentration units in which the other pollutants have been  reported.
Asbestos  was  found in three plants at raw waste concentration levels
of 2.1E8, 2.0E7, and 9.4E4 MFL, respectively, where E  is  exponential
on base 10.  All three plants belong to the Chlor-Alkali subcategory.
                                    53

-------

-------
     TABLE 5-2.  POLLUTANT FREQUENCY BASED CM SAMPLING PROGRAM RESULTS
                 IWCLUDIN3 RAW WASTE *
Pollutants Detected
Antimony
Arsenic
Beryllium
Cackniun
Chroniun
Capper
Cyanide
Lead
Mercury
Nickel
Seleniun
Silver
Thallium
Zinc
Benzene
Carbon Tetrachlorida
Chlorobenzene
1, 2-Oichloroethane
1,1, l-Trichloroethane
Hexachloroethane
1, 1, 2-Trichloroethane
1 , 1 , 2 , 2-Tetrachloroethane
Chloroform
1 , 2-Dichlorobenzene
1, 1-Oichloroethylene
I, 2-Dichloropropylene
2 , 6-Dinitrotoluene
Ethylbenzene
Fluorantnene
Bis(2-Chloroiscpropyl) ether
Methylene chloride
Dichlorobrananethane
TrichlorofluoroDethane
Chlorodibronomethane
Naphthalene
Nitrophenol
Pentachlorophenol
Phenol
Bis(2-Ethylhexyl) phthalate
Butyl benzyl phthalate
Di-n-butyl phthalate
Diethyl phthalate
Dimethyl phthalate
Benzo(a) anthracene
Benzo(a) pyrene
3 , 4-Benzof luoroethane
Chrysene
Anthrac^Be
Pluoi-ene
Phenanthrene
Pyrene
Tetrachloroethylene
Toluene
Trichloroethylene
Nitrobenzene
2 , 4-Dinitrophenol
Pollutant Occurrence Based
on Plant Grouping
5 or <5
Plants






X














X

X
X
X
X

X
X

X
X
X
X
X
X
X
X
X

X
X
X
X
X
X
X
X
X
X
X

X
X
X
>5 but £10
Plants



























X
























X



>10 Plants
X
X
X
X
X
X

X
X
X
X
X
X
X








X







X









X















Pollutant Occurrence Based on
Concentration Classification (ug/1)
iSO
28
38
49
45
20
21

25
46
17
46
45
41
9
6
2
1
2
4
1
2
3
15
1
3

1
7
1
1
11
5
2
2
1
1
2
2
20
3
15
5
2
1
1
1
1
1
1
1
1
4
7
3


>50 but
£500
19
12
4
4
13
16

15
2
20
7
7
11
18
1



1



2


1

1


3



1

1
3


















>500 but
&2,500
4
3

4
9
9

7

9

1
1
14








1







1

1



1

1













1

2
2
>2,500
1



10
7
2
6
S
8



12






















1

1

















*  Blanks indicate not detected.
                                    55

-------
              TABLE 5-3.  DISTRIBUTION OP POLLUTANTS ACCORDING
                          TO SUBCATEGGRy 1
Pollutants Detected
Subcategory Numbers Where Pollutants Found
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Copper
Cyanide
Lead
Mercury
Nickel
Selenium
Silver
Thallium
Zinc
Benzene
Carbon Tetrachloride
Chlorobenzene
1,2-Dichloroethane
1,1,1-Trichloroethane
Hexachloroethane
1,1,2-Trichloroethane
1,1,2,2-Tetrachloroethane
Chloroform
1,2-Dichlorobenzene
1, l-Dichloroethylene
1,2-Dichloropropylene
2,6-DinitrotDluene
Ethylbenzene
Fluoranthene
Bis (2-Chloroisopropyl) ether
Methylene chloride
Dichlorobratutethane
Trichlorofluoromethane
Oilorodibrononethane
Naphthalene
4-Nitrophenol
Pentachlorophenol
Phenol
Bis (2-Ethylhexyl) phthalate
Butylfcenzyl phthalate
Di-n-butyl phthalate
Diethyl phthalate
Dimethyl phthalate
Benzo(a) anthracene
Benzo (a) pyrene
        7, 23, 27, 28, 33
Alf but 7, 23, 27, 28, 33
 n   n

 N   »

 n   n

 n   ti
 "   "    "    II   "    II
1, 3, 4, 10, U, 25, 32
1, 2
1, 35
1, U, 13, 22, 35
1
4,11
1, 10, 35
1, 3, 4, 10, 13, 15, 19, 21, 22, 25, 32,.35
24
1, U, 13
26
1
1
1
1, 3, 4, 9, U, 21, 25, 32
8
22
1, 4, 8, 9, 12, 13, 19, 21, 22, 25,26, 32,
1, 4, 19, 32
1, 4, 25
10  -51
                                           35
1, SZ
17
2, 3, 4, 8, 15
2, 15, 26, 31, 32
1, 4, 7f 8, 10, 11, 12,13,15,18,24,25,26,30,31,35
1, 2, 12
1, 4, 8, 11, 17, 18, 19, 21, 22, 30, 31f 34, 35
8, 10, U, 19, 31
12, 31
a
    For name of subcategory,  refer to Table 3-2.
2 "All" means subcategory numbers  1 through 35 of Table 3-2.
                              (Continued)
                                      56

-------
                            TABLE 5-3.  Continued
Pollutants Detected
Subcategory Nunbers Where Pollutants Found
3,4-Benzofluoranthane
Girysene
Anthracene
Fluorene
Phenanthrene
Pyrene
Tetrachloroethyiene
Toluene
Trichloroethylene
8
8
8
8, 12
8
8
1, 4, 10, 22
1, 3, 4, 10, 11, 15, 18, 32
1, 4, 25
                                      57

-------

-------
                              SECTION 6

         PROCESS AND WASTE TREATMENT INFORMATION DEVELOPMENT
                            AND EVALUATION
Industry Data Base Description

Information and data on the inorganic chemicals industry were obtained
from  a number of sources.  These sources included literature reviews,
plant visits, telephone contacts, industry responses  to  the  Section
308-Questionnaires   and   information   supplied  by  industry  after
proposal.   The  type  of  material  gathered  from  these  sources  is
discussed below.

Literature Review

A  review  of  the  literature  was  conducted to identify and collect
information related to manufacturing processes, raw  materials,  water
use,  wastewater  sources,  wastewater treatment technology, raw waste
characteristics,  and  economic  data.   Relevant   information   from
reports,  books, papers, conference presentations and periodicals were
identified by computer search  and  are  presented  in  the  reference
section  of  this  report.   This  information was incorporated into a
broad based assessment of process and technology  practices  aimed  at
selecting   the   best   available   treatment   technology  and  best
demonstrated technology for the various  industry  subcategories.   It
also   provided   the   background   required   for   evaluating   the
subcategorization of the industries.

Plant Visits

During the screening and verification  phase  of  this  project,  much
information was gathered from individual plants relating to production
capacity,    manufacturing   processes,   waste   flows,  water  reuse,
wastewater treatment systems  and  performance,  and  best  management
practices  (BMP).   The  lead  visits  also provided an opportunity to
update and clarify some of the information given in the 308 responses.

Telephone Direct Contact

Numerous  contacts  were  made  with  knowledgeable  persons  in  both
industry  and government to gather and exchange information concerning
all phases of this study.  These sources are  cited  in  the  text  as
personal communcations.


308-Questionnaire Responses

The  basis  for  much  of the work in this study is the responses from
industrial organic chemical firms to the 308 data requests.
                                   59

-------
Data from 284 manufacturers' responses were utilized  by  the  project
team  for  the  development of approriate guidelines for the inorganic
chemicals subcategory.  Industrial  firms,  through  their  compliance
with   the   needs  of  the  308-Questionnaire,  provided  a  valuable
industry-wide data base used extensively in this analysis.

Essential data elements from each questionnaire were extracted for the
purpose of creating a working data base  for  this  report.   Specific
elements  selected  for  this  smaller,  more-manageable data base are
given in Table 6-1.

These data provided the basis for the  subcategory  review  through  a
profile  of  each  industry.   After  compilation of the questionnaire
data, industry totals for capacity and derivative quantities  such  as
percent  utilization,  effluent  per ton of product, and conversion to
metric units were compiled.

Process Waste Sources and Current Treatment Practices

Data Acquisition

The information presented in this section was obtained from a  variety
of  published  sources  and  the  available  industry responses to the
308-Questionnaires as well as from, plant visits  and  interviews  with
industry  personnel  conducted by the Agency and its contractor during
the toxic pollutant screening and verification program.   The  results
of  visits  and  interviews are documented in field notebooks, interim
plant visit reports, and telephone  communication  records  which  are
part of the rule making record.

Plant  visits were particularly useful for confirming and updating the
detailed technical  information  contained  in  the  308-Questionnaire
responses.   The  cooperative  attitude  displayed by industry greatly
facilitated the acquisition of reliable operating data and  meaningful
sampling results.

Evaluation of Data

Each  of  the  various  industrial subcategories in which verification
sampling was conducted was the subject of an extensive  evaluation  to
provide the technical basis' for selecting candidate advanced treatment
technologies  and  developing  the  related  base and incremental cost
estimations.
In the subsections which follow,  individual  plant  descriptions
presented according to the general format for each subcategory:

     General Process Description
          Description of process reactions and unit operations.
          Inventory of raw materials used.
          Typical process flow diagram.

     Water Use and Waste Source Inventory
are
                                 60

-------
          TABLE 6-1.  308-QOESTIONNAIRE RESPONSE DATA
                         DATA ELEMENTS
              INORGANIC  CHEMICALS GUIDELINES STUDY
Datum Reference
Description
Comments
Manufacturer
Product
Plant
Process
Effluent Treatment
Name
Location
EPA Region

Name
Subcategory

Number of other
Products
Capacity
Production
Age

Name
Volume of Process
Effluent
Volume of Noncontact
Effluent

Type
Permit
Major Pollutants
Confidential
                                                 Inorganic
                                                 Chemicals
Fiscal year
1976
1976
1976
                              61

-------
          Description of individual plants visited,  sampled and plant
          information from other sources.
          Inventory of raw proces wastewater sources and
          identification of sampling points.
          Process wastewater quality and flow data.
          Solid waste generation and disposal.

     Control and Treatment Practices
          Description of specific treatment technologies and operating
          facilities.
          Description of the total input to the treatment system
          including sources attributed to other production operations
          and noncontact water (e.g.,  cooling water, etc.).

     Evaluation of Production and Waste Flow Data
          Tabular summary of plant-specific data.
          Waste flows per unit of production (unit waste flow) with
          the range and average values.
          Solid waste quantities.
          Treatment chemical requirements.

     Process Modifications and Technology Transfer Options

     Best Management Practices (BMP)
          Plant area operations and housekeeping.
          Runoff control.
          Solid waste handling (e.g.,  fugitive dust and leachate
          control, etc.).

Model Plant and BPT Treatment Sytem Specification

The  model  plant concept plays a central role in both the development
of alternative treatment system designs for priority pollutant removal
and for estimating the related internal costs  of  such  treatment  in
each  subcategory.   In  order  to be representative of a subcategory,
each set of model plant specifications was composited from  a  profile
data  summary derived from the available information on production and
waste flow.

Based on  the  typically  achievable  waste  flow  rate  per  unit  of
production,  the  model  plant  was  used  as  a  starting  point  for
appropriately  designed  and  sized  BPT  level  wastewater  treatment
system.  Certain  assumptions were made regarding the possible process
variations and the specific raw waste sources incorporated  into  each
model.   In  most  cases,   it was appropriate to assume that the waste
flow per unit of production did not vary over the particular range  of
production capacities covered.  Production rates were selected in most
subcategories  to represent the small, mid-range and large size plants
presently in  operation.   Small  subcategories  were  represented  by
single   mid-range  production  rates  for  the  model  plants.   Cost
estimates were developed for each set of base level (BPT) and advanced
level  (BAT/NSPS) treatment system design specifications.
                                  62

-------
Beginning with Section 11, the model plant  and  BPT  level  treatment
system  descriptions  and  specifications for eac:h subcategory include
the following information:

     Production rates and mode of opearation.
     Specific process type and waste sources.
     Waste flow per unit of production.
     Solid waste generation and handling.
     Treatment chemical requirements.
If applicable, the new source model plant is also  described
design specifications given for its waste treatment system.
and  the
The  model  plants  do  not  represent  exemplary or specific existing
plants, but are typical plants of adequate  design  derived  from  the
range   of  plants  and  treatment  facilities  found  in  the  entire
subcategory. For the purpose of cost estimating, it  is  necessary  to
specify  cost  rationale,   define  a  set  of initial assumptions, and
consider  the  variability  of  factors  such  as  wastewater   flows,
pollutant  load,  unit  treatment  process,  plant  age, etc.  General
assumptions have been detailed under Section 10 of this report and are
employed as  the  basis  for  developing  baseline  model  plant  cost
estimates presented in the subsequent sections dealing with individual
industries.

Dissolved Solids in Wastewater Effluents

Many waste treatment plants discharge final effluent into watercourses
which  feed  fresh  water  streams  used as sources of water supply by
downstream  agencies  or  industries.   Groundwater   aquifers   which
underlie  large  portions  of  the  country are tapped to supply fresh
water through wells serving public and industrial water needs.  Saline
wastes  discharged  into  streams  or   into   unlined   lagoons   can
significantly  alter  the salt content (total dissolved solids) of the
fresh water.  Although Federal  regulations  seldom  limit  the  total
dissolved  solids  or  the  various  ions  such  as chloride, sulfate,
bicarbonate, and nitrate,  these constituents can be of serious concern
to local water users.

To protect the mineral quality of ground and surface waters State  and
local  water  pollution control agencies typically establish limits on
the  discharge  of  substances  which  contribute  sodium,  potassium,
hardness,  chloride,  sulfate,  or conductivity, which is a measure of
total solids in solution.   This restriction can affect  the  chemicals
chosen  for  waste treatment.  For example, alkaline precipitation can
be accomplished by  using  lime,  which  forms  an  insoluble  calcium
sludge, or by adding caustic soda, forming a soluble sodium salt.

In  choosing  an  acid  for  neutralization  of alkaline wastes, it is
important to weigh the overall effects of chloride (from  hydrochloric
acid)  and  sulfate (from sulfuric acid), particularly with respect to
irrigational use of the receiving water.
                                   63

-------
Chemicals used in the model plant processes were selected on the basis
of best performance,  including  consideration  of  scaling  problems,
which can be severe when calcium and sulfate are at saturation levels.
It  may  be  necessary  to  alter  the  nature  of chemicals used at a
specific plant, in order to meet local water quality requirements.
                                   64

-------
                              SECTION 7

     ASSESSMENT OF TECHNOLOGY FOR ADVANCED TREATMENT AND CONTROL
Introduction

In the inorganic chemicals  industry,  pollution  abatement  practices
vary  and a wide range of treatment technologies can be found, ranging
from no treatment to the application of highly  advanced  technologies
for the removal of specific pollutants.

Until  the NRDC Settlement Agreement, industry attention was primarily
directed towards general pollution problems including removal of trace
metals, but not towards treatment  of  over  100  individual  specific
organic  compounds  now  listed  as  toxic  pollutants.  Even with the
classical (conventional  and  nonconventional)  pollutants,  treatment
technology  has  been directed to removal down to the part per million
level, whereas now the  thrust  is  towards  part  per  billion  level
requirements.   For  both these reasons, higher level technologies are
often not in place in the inorganic  chemicals  industry,  and  it  is
necessary  to  look  into technologies that have been applied in other
industries or  developed  at  the  laboratory  or  pilot  plant  scale
specifically for the removal of these toxic substances from industrial
wastewater,  and  determine  whether  they  can  be  adopted as viable
technological options.

A list of candidate technologies was  complied  from  the  literature,
in-house  expertise, and industry contacts.  These were evaluated with
respect to:

      1.   Treatment effectiveness

     2.   Cost

     3.   Nonwater pollution environmental effects

     4.   Applications in the inorganic chemicals industry or on other
          industrial wastes with similar wastewater characteristics.

The anticipation that few of the organic  toxic  pollutants  would  be
found  in inorganic chemical wastes was justified by the results of the
analytical   programs.    Only  one  industrial  subcategory,  namely,
Chlor-Alkali  production  using  graphite  anodes,   had   potentially
significant  levels  of  organic  toxic  pollutants.  As a result, the
initial search  for  candidate  BAT  technologies  became  limited  to
treatment technologies for the thirteen metals, cyanide, and asbestos.

The  technologies finally adopted were not new or untried technologies
since  it was found that most treatment requirements could  be  met  by
taking      conventional     techniques—for     example,     chemical
                                   65

-------
precipitation—and developing them to a higher degree  of  engineering
and  design sophistication, so that optimum removal efficiencies could
be achieved.

The following pages  describe  the  theoretical  basis  for  treatment
systems adopted for BAT application.

Hydroxide Precipitation

Hydroxide  precipitation  is  the  most  widely  used  technology  for
removing trace metals from wastewaters,  with  lime  or  caustic  soda
commonly used to supply the hydroxide ions.  Under suitable conditions
the metals form insoluble metal hydroxides which can be separated from
solution.

The chemistry of the process is not simple, and must be understood for
each   metal.    Many  metals  are  amphoteric,  the  optimum  pH  for
precipitation varies, and organic complexes can interfere.   A  simple
form of the reaction may be written as:

     M++ + 20H- = M{OH)2                                    (1 )

     Metal ion + two hydroxyl ions = insoluble metal hydroxide

If  the  pH  is  below the optimum for hydroxide precipitation soluble
complexes form:

                                                            (2a)
         + OH- = M(OH)A

     Metal ion + hydroxyl ion = soluble metal complex

If the pH is above the optimum for hydroxide precipitation, the  metal
hydroxide may redissolve by forming soluble complex hydroxides:
     M(OH)Z + OH- = M(OH)3-

     Insoluble metal hydroxide + hydroxyl ion = soluble
       metal complex
                                                               2b)
Since  most metals have the capability of coordinating with other ions
or molecules, these simple equations assume that the  hydroxonium  ion
is the coordinated species.  However, if organic radicals are present,
they can form chelates and mask the typical precipitation reactions:

     M++ + OH- + nR = [M(R)nOH]+

     Metal ion + hydroxyl ion = soluble metal
      + organic ions             chelate

Such  complexes  may  require unusual treatment to hydrolyze them, and
their presence often  explains  why  some  treatment  practices  yield
relatively poor results.
                                    66

-------
Assuming  the  absence  of  organic  complexing  agents, the treatment
levels attainable by hydroxide precipitation can be  forecast  from  a
knowledge  of  the pH of the system.  Figure 7-1 shows the theoretical
solubility of those toxic  metals  which  form  insoluble  hydroxides,
while   Table ^ 7-1  shows  the  solubility  product  constants.   For
comparison,  the values for sulfides are also given in Table 7-1.

It is clear from the range of optimum pH's illustrated that for waste-
waters containing more than one metal, no single  optimum  pH  exists,
and  problems  arise  at the threshold of the alkaline range (circa pH
10) where some metals have least solubility,  while others are  at  the
point   of   redissolving  as  an  anionic  species.   For  successful
application as a wastewater treatment technology,  careful  control  of
pH must be practiced if the best removals are to be achieved.

In  practice  the  solubility of metallic hydroxides, and the tendency
for fine insolubles to remain in suspension,  may yield effluents which
will not meet ^g/1 standards, and  hydroxide  precipitation  is  often
supplemented  by  the  use  of  coagulating  agents  to improve solids
removal, or sulfide coprecipitation to reduce ultimate solubilities.

In practice, the technology uses unit process steps which are  simple,
well-established, and well-understood by the industry.

Depending on the quantity of waste flow, the treatment can either be a
batch or continuous operation, with batch treatment being favored when
waste  flows  are  small.   In  batch  treatment the equipment usually
consists of two tanks, each with a capacity to treat the total  waste-
water  volume  expeced during the treatment period.  These systems can
be economically designed for flows up to 50,000 gallons per day (5).

The treatment tanks serve the multiple  functions  of  equalizing  the
flow,  acting  as  a  reactor  and as a settler.  During operation the
wastewater is stirred, and a homogeneous sample is taken and  analyzed
to  determine  the  chemical  dosage  requirements.  The chemicals are
added, mixed and stirred for about 10 minutes.  After the reaction  is
complete, the solids are allowed to settle for a few hours.  The clear
liquid is then decanted and discharged.  Settled sludge is retained to
serve  as  a  seed  for crystal growth for the next batch, but must be
drawn  off  periodically  and  disposed  of,   usually  in  a  chemical
landfill.

For  larger  daily  flows  a  typical continuous flow treatment scheme
consists of a flash mixer,  flocculator,  settling  unit  with  sludge
storage tank, and, in some cases, a filtration system.

The  ability  to separate the solids from the wastewater is important.
Metallic hydroxides tend to  be  gelatinous  and  separate  poorly  in
gravity separators.  Finely suspended solids tend to pass out with the
effluent  and increase the total metal content.  Thus, improvements in
precipitation applications  have  been  directed  toward  fine  solids
removal,  and  this is reflected in the addition of various filtration
                                    67

-------
 Figure 7-1.   Theoretical  solubilities of  toxic metal hydroxides/oxides as a
              function of  pH.
                                                  10    11    12    13
NOTE:  Solubilities of metal hydroxides/oxides are from data by M. Pourbaix,
       Atlas of Electrochemical Equilibria in Aqueous Solutions,
       Pergamon Press, Oxford,  1966.
                                    68

-------
              TABLE 7-1.  SOLUBILITY PRODUCTS OF TOXIC METALS
       Metal
       Solubility Product Constant (K—)

Metal Hydroxide            Metal Sulf ide
Antimony  (III)

Arsenic

Beryllium

Cadmium

Chromium  (III)

Copper

Lead

Mercury

Nickel

Selenium

Silver

Thallium  (I)

Zinc
1.6 X 10
2.5 X 10
6.3 X 10
2.2 X 10
1.2 X 10
3.0 X 10
2.0 X 10
-22 (1)

-14 (1)

-31 (1)

-20 (1)

-15 (1)

-26 (1)

-15 (1)
2.0 X 10
        -8  (1)
1.2 X 10
        -17  (1)
3.6 X 10
                                   -29  (2)
8.5 X 10
-45 (2)

-28 (2)
3.4 X 10

2.0 X 10~49 (2j

1.4 X10-24 (2>
                   1.6 X 10
                           5.0 X 10
                   1.2 X 10
        -49 (2)

        -21 (1)

        -28 (2)
NOTE:  References for above values are shown below.
(1)  Dean, J.A., Ed., Lange's Handbook of Chemistry,  12th ed., McGraw-Hill
     Book Co., New York, 1979.

(2)  Vfeast, R.C., Ed., Handbook of Chemistry and Physics, 57th ed.,  CRC Press,
     Cleveland, Ohio, 1976.
                                     69

-------
systems  and  the  use  of  flocculant  aids  as  improved  levels  of
treatment.

Soda  ash  (sodium  carbonate,   Na2C03)   is  sometimes found to be the
reagent of choice particularly  for  lead  removal.   Lead  carbonate,
PbC03, and lead hydroxide/carbonate, 2PbC03«Pb(OH)2, (basic carbonate)
are  formed  which  may  afford  improved  settling  properties  for a
particular waste.  This practice is found in  Chlor-Alkali  (Diaphragm
Cell) waste treatment.

Hydrated  lime  suspensions  are  more  commonly used than soda ash or
caustic soda  as  the  hydroxide  source  because  they  are  cheaper.
However,  if  there  is  sulfate ion present in the wastewater, gypsum
will be formed:
Ca(OH
S04>— = CaS04
                                 20H-
(4)
     Hydrated lime + sulfate ion = calcium sulfate (gypsum)
                                   + hydroxyl ions
This increases the sludge produced,  may  cause  scaling  problems  in
pipelines,  and  may  clog a dual-media filter.  Using caustic soda is
more expensive, but it generally elimiates the scaling problem.  Total
dissolved solids in the form of sodium  salts  are  increased  in  the
caustic  soda  treated  wastewaters.   Although  low concentrations of
sodium are not regarded as polluting, high levels  can  make  drinking
water unpalatable, limit the use of water for agriculture, and promote
degradation  of the structure of arable soils.  Thus, where high total
dissolved  solids  are  of  concern,  lime  would  be  the   preferred
neutralizing agent.

This  treatment  technology  is  widely applied in treating industrial
wastewaters.   Industries  that  are  using  hydroxide   precipitation
include:

     Inorganic Chemicals
     Plating and Metal Finishing
     Mining
     Textiles
     Steel and Iron
     Non-Ferrous Metal Processing and
     Electronics

Better  than   99 percent removal of trace metals have been reported in
the literature with final  concentrations  in  the  treated  effluents
ranging from sub ppm to low ppm  (see Table 8-1 through 8-10).

Ferrite Coprecipitation

An  interesting variation on the theme of hydroxide precipitation  is a
process developed  in Japan for the  removal of heavy metals from acidic
wastewater.  The process, known as  ferrite  coprecipitation,  has  the
potential  for producing a marketable residual by converting the metal
                                    70

-------
ions in solution into insoluble ferromagnetic oxides or ferrites which
can be removed magnetically or by filtration (5).   The  treatment  is
applied by adding a ferrous salt to the metal-bearing wastewater, then
neutralizing   and   oxidizing   the   complex   heavy   metal-ferrite
coprecipitate. Particle sizes are reported to be relatively large  and
sludges formed can be safely disposed of by landfilling.

Although  extensive  performance  data  have  not  been developed, the
information available indicates that very  high  removal  efficiencies
can be achieved for most of the common heavy metals, including mercury
and  hexavalent  chromium.  The method has not been considered here as
an available technology due to the lack of sufficient  information  on
chemical  dosing requirements, energy requirements, and performance in
situations similar to those found in the inorganic chemicals industry.
In connection  with  wastewater  treatment  in  the  Titanium  Dioxide
Subcategory  for  the sulfate process, the wastes contain considerable
amounts of ferrous iron from the processing of ilmenite  ore  and  the
practice of neutralization and aeration may involve the same chemistry
as the ferrite coprecipitation process.

Sulfide Precipitation

The basic principle of sulfide treatment technology is similar to that
of  hydroxide  precipitation.   Sulfide  is  added  to precipitate the
metals as metal sulfides and  the  sludge  formed  is  separated  from
solution by gravity settling or filtration.  Sodium sulfide and sodium
bisulfide are the two chemicals commonly used,  with the choice between
these   two   precipitation   agents   being   strictly   an  economic
consideration.

     Metal sulfides form according to the following equation:

     M+ + - + NazS = MS + 2Na+                                 (5)

     Metal ion + sodium sulfide = insoluble metal sulfide
                                   + sodium ions

In order to  calculate  the  theoretical  solubilities  of  the  metal
sulfides  as  a function of pH, the equilibria involved in solid metal
sulfide dissociation are taken into account:
MS =
              + S=
(6)
Metal sulfide = metal ion + sulfide ion  and,  depending  on  pH,  the
sulfide ion can react with hydrogen ions to form the bisulfide ion and
hydrogen sulfide.
     S= + H+ = HS-

     Sulfide ion + hydrogen ion = bisulfide ion

     HS- + H+ = Hj.S
                                                        7)
                                                       (8)
                                   71

-------
     Bisulfide ion + hydrogen ion = hydrogen sulfide

The   concentration   of   metal   ion  in  solution  will  equal  the
concentration of sulfide ion,  bisulfide  ion  and  hydrogen  sulfide.
Knowing  the metal sulfide solubility product (Table 7-1) and the acid
dissociation constants of hydrogen sulfide,  Kl = 9.1 x 10~8, K2 =  1.1
x 10~12 (see Reference 2 in Table 7-1) the solubility of the metal ion
can be calculated as a function of the hydrogen ion concentration and,
therefore, as a function of pH.
For a divalent metal ion the equation is:

(M++) = [Ksp (1 + (H+)/l.l x TO-12) + ((H+)*/l  x
                                                              1/2
Using the above information, the theoretical solubilities of the toxic
metal sulfides were calculated and are shown in Figure 7-1.

The  major  problem  in  applying  sulfide precipitation techniques is
associated with the toxicity of sulfides.  This warrants both care  in
application  and  post  treatment  systems  to  remove excess sulfide.
Pretreatment involves raising the pH of the waste stream  to  minimize
evolution of hydrogen sulfide gas.

A  recently  developed and patented process to eliminate the potential
hazard of excess sulfide in the effluent and the formation of  gaseous
hydrogen  sulfide uses ferrous sulfide as the sulfide source (6).  The
fresh ferrous sulfide is prepared by adding sodium sulfide to  ferrous
sulfate.   The  ferrous sulfide slurry formed is added to a wastewater
to supply sufficient sulfide ions to precipitate metal sulfides  which
have lower solubilities than ferrous sulfide.  Typical reactions are:

     FeS + Cu++ = CuS + Fe++                                (10)

     Ferrous sulfide + copper ion « insoluble copper sulfide
                                     + iron ion

     FeS + Ni(OH)2 = Fe(OH)2 + NiS                          (11)

     Ferrous sulfide + = ferrous hydroxide +
      nickel hydroxide    insoluble nickel sulfide


A  detention time of 10-15 minutes is sufficient to allow the reaction
to go to completion  (7).  Ferrous sulfide itself is also a  relatively
insoluble  compound.  Thus the sulfide ion concentration  is limited by
the solubility of ferrous sulfide, which amounts to about  0.02  mg/1,
and   the  inherent  problems  associated  with  conventional  sulfide
precipitation are minimized (8).

One other advantage of this  process  is  that  if  chromium  (VI)  is
present,  it  will also be reduced at the pH of normal operation (8 to
9) and precipitate as the trivalent hydroxide (Cr III).
                                   72

-------
Figure 7-2.  Theoretical solubilities of toxic metal sulfides as a

             function of pH.
        10
 r T

 a
i
 8
 0)
•a
 a

 I
i
                                                         12  13
                                      pH
                             73

-------
Treatment systems for sulfide precipitation are similar to those  used
for  hydroxide precipitation.  A continuous treatment scheme generally
consists of a pH adjustment tank, flash mixer,  flocculator,  settling
units with sludge storage, and a dual media filter.

Before  the  addition  of  sodium  sulfide  or bisulfide the pH of the
incoming wasteflow is adjusted to pH of 7-8 in the first reaction tank
to reduce the  formation  of  obnoxious  hydrogen  sulfide  gas.   The
chemicals  are then added to the flash mixer where they are thoroughly
mixed with the wastewater.

After the flash mix,  the  wastewater  passes  through  a  flocculating
basin  where  the  floe agglomerates and settles in the settling unit.
The overflow from the settling unit generally passes through a  filter
to  remove  any fine precipitates.  Any excess sulfide will need to be
removed before final   discharge.   This  can  be  achieved  either  by
aeration or other chemical oxidation techniques.

Sulfide  precipitation  is  being practiced in the inorganic chemicals
industry, mining industry,  textile  industry,  and  nonferrous  metal
processing  industry.   Most  of the Chlor-Alkali industry is applying
this technology to remove mercury from its waste streams.

Literature citations on the efficiency of  sulfide  precipitation  {9,
10,  11) indicate that most results are in the sub ppm range, and that
sulfide treatment is superior to hydroxide treatment for  the  removal
of  several  trace  metals.   A  recent report concluded that, with no
complexing agents in the waste, the following effluent quality can  be
achieved (11).

                    Metals Concentration
                    Cadmium
                    Copper
                    Zinc
                    Nickel
0.01  mg/1
0.01  mg/1
0.01  mg/1
0.05  mg/1
                    Chrome (total) 0.05 mg/1

Adding  ferrous  sulfide as a polishing step to remove residual metals
appears to be a promising, economical technology.  Although  there  is
no  full-scale treatment system using ferrous sulfide operating in the
inorganic chemicals industry, pilot studies on  chrome  pigment  waste
indicate  that  this  process  is superior to sulfur dioxide reduction
followed by hydroxide precipitation (12),

The Xanthate Process

The use of xanthates for the removal  of  metals  from  waste  streams
appears  to  be a new, promising technology for treating metal-bearing
wastewaters.  Xanthates contain functional groups capable  of  forming
insoluble  complexes  with  metals,  and  the  sludge so formed can be
separated by conventional means.
                                    74

-------
Xanthates can be generated by mixing starch or cellulose  with  carbon
disulfide  in  a  caustic  medium.  Three types of xanthates have been
proven in bench pilot  scale  studies  to  be  effective  in  removing
cadmium,   chromium  (III), copper, iron, lead, mercury, nickel, silver
and zinc from industrial wastewaters (13-20).  These are:

          Soluble starch xanthate with a cationic polymer,

          Insoluble starch xanthate, and

          Fibrous cellulose xanthate

     The general removal mechanism is as follows:

     2 [ROCS (=S)]Na + M++ = [ROCS (=S)J2M + 2Na+           (12)

     Xanthate + metal ion = insoluble metallic xanthate
                             + sodium ions

     where R « starch or cellulose.

Unlike  hydroxide  precipitation,  this  process  is  reported  to  be
effective  in removing metals over a wide pH range of 3 to 11, with an
optimum range between 7 and 9.

Brass mill wastewaters, lead battery  effluent,  circuit  board  rinse
waters,   electroless   copper  plating  rinse  waters,  pyrophosphate
electroplating rinse waters, and  copper  etching  rinse  waters  were
studied  in  a  pilot  plant, with  insoluble  starch  xanthate as the
complex ing  agent  (20).   This  pilot  study  demonstrated  that  the
xanthates can either be added to a reactor to mix with the wastewaters
or  be  applied  as  a  precoat on a pressure filter (20).  Results of
these pilot studies showed that metals were reduced to below  50  »g/l
(ppb).

Another  study  indicated cellulose xanthate is as effective as starch
xanthate in removing trace metals.  The following table summarizes the
result of the study with a cellulose xanthate dosage of 90 mg/1 and  a
contact time of 30 minutes (18-19).

                        Concentration,, mg/1

               Metals         Influent       Effluent
               Cadmium
               Chromium
               Copper
               Iron
               Lead
               Nickel
               Zinc
1 .35
0.30
1 .6
3.1
3.9
2.4
1 .0
0.027
0.022
0.06-0.14
0.08-0.36
0.008-0.021
0.077
0.03-0.04
                                   75

-------
This  study  also  concluded  that  cellulose  xanthate is superior to
starch  xanthate  in  terms  of   sludge   settling   characteristics,
filterability,  and handling.

Xanthate  may  also  be  used  as  a  complexing  agent to prevent the
formation  of  soluble  anions   from   insoluble   amphoteric   metal
hydroxides.

The  xanthate  process is a relatively new technology, and the reagent
compounds are  not  yet  available  in  commerical  quantities.   More
information  is  needed on dosage rates in continuous flow operations.
Potentially the metals can  be  recovered  by  leaching  the  xanthate
complex with nitric acid, but metal recovery has not been demonstrated
yet.   Sludge  disposal  problems  may  arise if the sludge complex is
unstable and, if xanthates are to be generated on site, care  will  be
needed in handling the hazardous carbon bisulfide.

Ion Exchange

Ion  exchange  is a chemical reaction between the ions in solution and
the ionic sites on an exchange  resin.   Many  natural  solids  (e.g.,
soils,  proteins, and zeolites) exhibit such exchange characteristics.
However, synthetic resins  are  the  predominant  ones  used  for  ion
exchange  applications   in modern industrial technology.  These resins
contain functional groups that can react with the  ions  in  solution.
Depending  on  these  functional  groups, the resins can be classified
into:

     Strongly acidic cation exchanger,
     Weakly acidic cation exchanger,
     Strongly basic anionic exchanger, and
     Weakly basic anionic exchanger.

Cation exchangers are capable of exchanging with cations in  solution.
Strongly  acidic  cation  exchangers contain functional groups such as
sulfonates, {-S03H and -S03Na), while weakly  acidic  exchangers  have
functional groups derived from carboxylic acids, (-COOH and -COONa).

Anionic  exchangers  are used to exchange with the anions in solution.
In general, strongly basic exchangers contain amine functional  groups
<-R3NOH  and  -R3NC1),   and  weakly  basic  exchangers contain ammonia
functional groups (-NH3OH and -NH3C1).

When the functional groups are used up in the reaction, the resins can
usually be regenerated.  Cationic resins can be regenerated by  sodium
chloride,  hydrochloric  acid,  suIfuric  acid  or  sodium  hydroxide.
Anionic  resins  are  regenerated  by   sodium   hydroxide,   ammonium
hydroxide, sodium carbonate, sodium chloride, or hydrochloric acid.

The  exchanger  can  either  be  added  to  the  wastewaters   in batch
operations or be packed  in a fixed bed or column.  Fixed bed is by far
the more effective and hence more popular.   The  operation  generally
                                     76

-------
follows    a   four-step
regeneration, and rinse.
cycle;    exchange   (service),    backwash,
During the exchange step, the reaction between the  ions  in  solution
and  the ionic sites in the resin takes place as the wastewater passes
down the bed.  The reaction is  generally  regarded  as  a  result  of
electrostatic  attraction  (20).   Therefore, the size of the hydrated
ion and the charge on the ion are  the  determining  factors  for  the
exchange  reaction.  A trivalent ion is attracted more strongly than a
bivalent  ion  which  is  in  turn  attracted  more  strongly  than  a
monovalent  ion.   For ions with the same charge, the smaller hydrated
ion is capable of moving closer to the  exchange  site,  and  is  thus
favored.

Many  synthetic resins contain functional groups that are selective to
certain metals.  For example,  a  resin  manufactured  by  a  European
company   reacts  preferentially  with  mercury  (Hg++)  and  mercuric
chloride (HgCl+) ions according to the following equations:
     2RSH + Hg++ = RSHgSR + 2H+
                                 (13)
     Resin + mercury ion = insoluble resin complex
                            + hydrogen ions

     RSH + HgCl + = RSHgCl + H+

     Resin + mercuric chloride ion = insoluble resin complex
                                      + hydrogen ions
                                  14)
The exchange reaction is governed by the law of mass  action.   During
the  reaction,  the affinity of the resin for the two ions is so great
that essentially all the mercury  or  mercury  chloride-resin  complex
formation  equilibria  are  shifted  toward  the formation of Hg++ and
HgCl+  which  are  rapidly  removed.   A  5   ppb   residual   mercury
concentration in the effluent is achieved by this process (22).

After  all the exchangeable sites in the resin are used up, the bed is
backwashed by passing clean water through to loosen up the bed and  to
remove any fine particulates that are trapped inside the bed.
After  the  backwash  cycle
appropriate regenerant.
  the  resins  can  be regenerated with the


                                 (15)
     RSHgCl + HC1 = RSH + HgCl2

     Insoluble resin complex = regenerated resin
      + hydrochloric acid       + mercuric chloride

One attractive  feature  of  the  ion  exchange  process  is  that  it
concentrates  the metals in the regeneration step, and thus provides a
potential for their recovery.  However, if recovery is  not  feasible,
this creates a secondary stream which needs to be treated.
                                    77

-------
A recent study found that sodium alumino silicates (zeolites) might be
a  low-cost exchanger that can be discarded after a one-time use (22).
This would eliminate the regeneration step.  On a batch study  with  a
five-minute contact time, cadmium and mercury were removed to below 10
ppb.   Thermodynamic considerations show this exchanger to have a high
affinity for cadmium, copper, mercury, nickel, silver,  zinc,  cesium,
and barium.

Ion   exchange   is   a   proven  technology  that  can  reduce  metal
concentrations down to low levels.  However this  technology  is  used
only in limited industrial pollution abatement applications because of
the high cost associated with the process.  Consequently, ion exchange
has not been recommended in this report for BAT technology.

Reduction Processes

Many  metals can exist in solution in several oxidation states, and it
may be necessary to convert from a higher valency state to a lower one
in order to apply a given chemical reaction.  The classic  example  is
chromium,  which  as the trivalent chromic  ion will precipitate as the
hydroxide in alkaline  solution,  while  the  hexavalent  chromate  or
dichromate   ion  will  not.    The  latter  needs  to  be  reduced  if
precipitation is to occur.

Hexavalent chromium  (e.g., Cr04= and Cr207=)  is  toxic  and  soluble.
The  most  efficient  way of removing this  from solution is a two-step
process of reduction followed by precipitation.

Chromium (III) is much less toxic than chromium   (VI),  and  forms  an
insoluble hydroxide which can be removed from solution' by settling and
filtration.

A  number  of  chemicals are used for the reduction of chromium.  Most
common are sodium bisufite, sodium metabisulf ite, sulfur  dioxide  and
ferrous  salts.   The reduction is accomplished readily at low pH with
these reagents.  Typical reduction reactions  are:
Cr207= + 2H+ * 2Cr+++ + 3S04=
                                       H0
                                                             (16)
     Sulfur dioxide + dichromate ion = trivalent chromium ion
      + hydrogen ion                    + sulfates and water
3S03= + Crz07= + 8H+ =

Sulfite ion + dichromate ion = trivalent chromium ion
 + hydrogen ion                 + sulfate + water
Crz07=
4H+ = 2Cr+++
                          6Fe+++
                                          7H0
                                                             (17)
                                                             (18)
     Ferrous ion + dichromate ion = trivalent chromium  ion
      + hydrogen ion                 + ferric ion + water
                                    78

-------
The reduced chromium  and  the  ferric  ions  produced  in  the  third
equation will exist as the soluble sulfate at acid pH's;  If the pH is
above  5,  the  reaction  rate  is  drastically  reduced, and although
dithionite will effect reduction at neutral pH's, it  is  very  costly
and its use may be contraindicated.

After  the  reduction step, lime or caustic soda is added to raise the
pH to 8.5-9.0.  Trivalent chromium will be precipitated.
     Cr+++ + 30H- = Cr(OH}
                                                       (19)
     Trivalent chromium = insoluble chromium hydroxide
      + hydroxide ion

The theoretical solubility limit of chromium hydroxide is  above  0.02
mg/1  (8).   It  is reported that applying sulfur dioxide to a pigment
waste consistently reduced Cr (VI) and Cr(T) to 0.5 mg/1 and 1.5  mg/1
respectively as 30-day averages (9) (10).  By applying ferrous sulfide
to  a  plating  waste with an initial Cr(VI) concentraiton of 128 mg/1
and Cr(T) concentration of 153 mg/1, an effluent quality of less  than
0.05 mg/1 of either species is achieved (12).

A  one-step precipitation-reduction process using sodium bisulfide was
used in a dichromate plant to remove chromium from its wastewater.  An
effluent quality with less than 1 mg/1 Cr(VI), and less  than  5  mg/1
Cr(T) was reported (3).

One  other  common  reduction  process  is  the  application of sodium
borohydride to reduce metals in waste streams.  Sodium borohydride  is
a mild but effective reducing agent (3), and is currently used in some
chlor-alkali  plants  to  reduce  the  soluble mercury ion to metallic
mercury which is removed from solution by carbon adsorption:
4Hg-
      + BH4- + 80H- = 4Hg + B(OH)4~ +
                                           4H0
20)
Mercury ion + borohydride ion = insoluble mercury metal
 + hvdroxvl ion                  + borate ion + water
        hydroxyl ion

A mercury level of 0.01 mg/1 in the final effluent has
(3).
                                                   been  reported
Sodium  borohydride  is  also  reported  to  be  effective in removing
silver,  mercury,  gold,  lead,  and  cadmium  (5).     However,   this
technology  is  only  being  applied in limited cases, the cost of the
chemical being the major drawback.  The cost of sodium borohydride was
$16.00 per pound in 1978 (23).

Oxidation Processes

The oxidation of  organic  substances  is  generally  carried  out  by
thermal  processes  such  as  wet  oxidation  and  incineration, or by
                                    79

-------
biological processes such as the activated sludge
filters, biodises, and aerated lagoons.
                                                   process,  trickling
Incineration  is  actually  a  combination of oxidation and pyrolysis.
Both involve chemical changes resulting from heat.  Oxidation involves
actual reaction with oxygen, while pyrolysis refers  to  rearrangement
or  breakdown  of  molecules  at  high  temperatures in the absence of
oxygen. There are five types of incinerators  available  commercially.
These  are  rotary  kiln, multiple hearth, liquid injection, fluidized
bed, and pyrolysis (24).  A minimum temperature of 1000 degrees C  and
a  residence  time  of  two  seconds  is  required for the reaction to
proceed.  This process has been shown to  be  successful  in  reducing
pesticides to harmless molecules (25).

Wet  oxidation  is a process in which an aqueous waste can be oxidized
in the liquid
vessel.   This
exhaust   gas)
for a variety
liquor  (26).
              phase  in  a  closed,  high-temperature,  high  pressure
              reduces some of the problems (such as air pollution from
               inherent  in incineration.  Wet oxidation has been used
              of wastes  including  pulping  waste  and  acrylonitrile
                A  percent  reduction in excess of 99.8 of some of the
toxic pollutants has been reported (27).

Thermal oxidation processes are not expected to have much  application
in the inorganic chemicals industry, mainly because of the high energy
cost  required and the low level of organic contamination found in the
wastes.

The application of chemical oxidation to  industrial  wastes  is  well
established  for cyanides, sulfite, ammonia, and other harmful species
in dilute waste streams  (phenols,  mercaptans,  polysulf ides,  etc.).
Common   chemicals   used   as  oxidizing  agents  included  chlorine,
hypochlorite, hydrogen peroxide, potassium  permanganate,  ozone,  and
chlorine dioxide.  Air and oxygen are also used.

The  most  widely used chemical oxidation technology applicable to the
inorganic  chemicals  industry  is  the  oxidation  of  cyanide.   The
oxidation reaction between chlorine and cyanide is believed to proceed
in two steps as follows:
     CN- + Cl, = CNC1 + Cl-
     Cyanide + chlorine = cyanogen chloride + chloride ion

     CNC1 + 20H- = CNO- + Cl- + H20

     Cyanogen chloride = cyanate ion + chloride
      + hydroxyl ion      ion + water
                                                           (21)
                                                           (22)
The   formation   of   cyanogen   chloride    (CNC1)   is   essentially
instantaneous.  The second reaction,  the  formation  of  cyanate,  is
accomplished  most  rapidly  and  completely  at  a pH of 10 or higher
(9,28).  A detention time of  30  minutes  to  two  hours  is  usually
allowed.
                                   80

-------
The  cyanates  can  be  further  decomposed  into  nitrogen and carbon
dioxide by excess chlorination or acid hydrolysis.
2CNO- + 40H-
             = 6C1-
 2C0
                                              2H0
(23)
     Cyanate + hydroxyl ion = chloride ion + carbon dioxide
      + chlorine               + nitrogen + water
CNO
2H0 + = C0
H0
                                                           (24)
     Cyanate + hydronium ion = carbon dioxide + ammonium ion
                                + water

The first reaction can be accomplished in about one hour if the pH  is
adjusted  to  8.0-8.5.   Acid hydrolysis usually takes place at pH 2-3
and care must be taken to avoid the liberation of the  toxic  cyanogen
chloride as a gas.  Hydrolysis is not usually the chosen option.

Other   common  chemicals  used  to  oxidize  cyanide  include  sodium
hypochlorite, ozone, and hydrogen peroxide.  The reaction  for  sodium
hypochlorite  is  essentially the same as for chlorine.  For ozone and
hydrogen peroxide, the oxidation step proceeds as follows:

     03 + CN- = 02 + CNO-                                  (25)

     Ozone + cyanide = oxygen + cyanate ion

     H202 + CN- = CNO- + H20                               (26)  .

     Hydrogen perioxide + cyanide = cyanate ion + water

The advantage of  using  these  two  oxidizing  reagents  is  that  no
dissolved  solids  are  added  to the wastewater.  In addition, excess
chlorine is not discharged.

A  patented  process  uses  hydrogen  peroxide  and  formaldehyde   to
decompose   cyanide  at  about  120°F.   This  has  the  advantage  of
precipitating cadmium and zinc simultaneously (9).

Laboratory  studies  in  one  plant  currently   practicing   alkaline
chlorination  indicated that the presence of ammonia in the wastewater
reduces the efficiency of cyanide removal.   It  is  well  known  that
ammonia reacts with chlorine or hypochlorous acid to form chloramines:
     NH, + HOC1 = NH,C1 + H»0
                          (27)
     Ammonia + hypochlorous acid = monochloramine + water, etc.

     NH2C1 + HOC1 = NHC12 + H20                            (28)

     NHC12 + HOC1 = NC13 + H20                             (29)
                                    81

-------
If  excess  chlorine  is  added,  chloromines  can  be  converted into
nitrogen oxides(s):
     2NH, + 4HOC1 = N,0 + 4HC1 + 3H,0
30)
This equation is not exact because the final form of nitrogen oxide is
believed to be a mixture of nitrous oxide, nitrogen dioxide and nitric
oxide.

The treatment of cyanide by chemical oxidation is currently  practiced
in the following industries:

     Inorganic Chemical (Hydrogen Cyanide Production)

     Mining

     Plating

The  free  cyanide  level  after treatment is generally below 0.1 mg/1
(9).

Membrane Processes

Membrane processes have emerged in the last decade as a promising  new
technology  for  the  treatment  of  saline  water and wastewaters.  A
membrane is a semi-permeable barrier which  allows  the  transport  of
some  molecules  (ions)  and  retains  others.   The driving force can
either be electropotential differences (electrodialysis)  or  pressure
difference   (reverse   osmosis   and   ultrafiltration).   The  major
application of these processes has been the desalination  of  brackish
water and sea water.  More recently, these have also found application
in a number of industries, including:

     Mining
     Electroplating
     Metal Finishing
     Printed Circuit Board Manufacturing
     Battery Manufacturing
     Pulp and Paper
     Food Processing

In  electrodialysis,  an  even  number of alternating anion and cation
selective membranes are placed between two electrodes.   When  current
is  applied  the  anions  are  attracted  to the anode, and cations are
attracted to the cathode.  In the process of  migration,  the  cations
pass  through  the  cation-permeable  membrane  and are blocked by the
anion-permeable membrane.   Likewise,  the  anions  pass  through  the
anion-permeable membrane and are blocked  by the cation membrane.  This
results in alternating paths of purified  water and concentrated reject
(Figure 7-3).

The  electrodialysis membranes are made very thin and are assembled in
stacks.   The  flow  path  is  the  active  portion  of   the   cells.
                                     82

-------
Pretreatment  to  remove  suspended materials is absolutely essential.
Other materials in the waste feed that may lead  to  membrane  fouling
include  high  organic  content,  calcium sulfate, and certain complex
ions such as ZnCl~ which can partially convert the anion  membrane  to
the cation form, with significant loss in system performance (28).

As ionic concentration decreases, the electroconductivity of the water
also decreases, making it less efficient to remove the remaining salt.
Most  operations  do not produce a product water of less than 500 mg/1
total dissolved solids.

Reverse osmosis (RO) and ultrafiltration (UF)  are  similar  in  basic
concepts.   Both  are pressure-driven separation processes that employ
high-flux  semi-permeable  membranes  operating  under  dynamic   flow
conditions  (29).    In  contrast to electrodialysis, these involve the
transport of solvent, not solute, across the membrane.

Osmosis is a process in  which  solvent  from  a  dilute  solution  is
transported  spontaneously  across  a  semi-permeable  membrane into a
concentrated solution.  By applying enough pressure to  overcome  this
osmotic pressure,  reverse osmosis, i.e., the passage of solvent from a
concentrated  solution  to  a dilute solution through a semi-permeable
membrane, occurs.   The operating pressure of reverse osmosis units  is
usually  between 350 and 600 psi.  Ultrafiltration usually operates at
a much lower pressure (5  to  100  psi).   The  predominant  transport
mechanism  is  selective  sieving through pores.  The membrane retains
high molecular weight  dissolved  solids  such  as  synthetic  resins,
colloids, and proteins.  The upper and lower molecular weight limit is
generally defined as 500,000 and 500, respectively.

Membranes are usually fabricated in flat sheets or tubular forms.  The
most  common  material is cellulose acetate but other polymers such as
polyamides  are  used.   There  are   four   basic   module   designs:
plate-and-frame,  tubular,  spiral-wound, and hollow fiber.  Table 7-2
is a comparison between the various reverse osmosis modules.  Membrane
processes are effective in  removing  (concentrating))  inorganic  and
organic substances from a wastestream.  Usually extensive pretreatment
is  required to reduce the suspended solids and control pH.  There are
uncertainties about operation efficiency, membrane lifetime, rejection
specificity, and other factors.  If  recovery  is  not  feasible,  the
concentrated reject must be disposed or treated by other methods.  The
high  operation  and capital cost limits the widespread application of
these technologies.  For  these  reasons  membrane  technique  is  not
recommended as a BAT technology for this industry.


Adsorption

Adsorption is a surface phenomenon in which a substance is accumulated
on  the surface of another substance.  Sorption of a solute on a solid
surface is widely used in pollution  abatement  practices.   The  term
"adsorbate"  refers  to the substance being concentrated, and the term
"adsorbent" refers to the material that provides the surface.
                                    83

-------
Activated carbon is the prevalent adsorbent used.  Both inorganic  and
organic  substances  are  known to be removed effectively by activated
carbon.  Certain chlor-alkali plants  are  currently  using  activated
carbon as a polishing step to remove mercury.

Activated  carbon  is made by charring basic substrates, such as wood,
coke, coal, shell,  husks, etc., at 600°C in a  controlled  atmosphere,
where  oxygen  is  kept  low  by adding carbon dioxide or steam.  This
process drives out volatiles, leaving a porous carbon  lattice  in  an
"activated" state.

Activated  carbon  can  be  obtained  in  powdered  and granular form.
Powdered carbon is about 50-70 microns in  diameter,  and  90  percent
should pass through a 300-mesh screen.  Granular carbon is about 0.1-1
mm in diameter, and because of this is three times more expensive than
powdered carbon.

The  application  involves  the  passage  of the wastewaters through a
contact bed.   When  the  bed  is  exhausted,  the  carbon  is  either
regenerated or sent to landfill.  It is economical for large plants to
regenerate   the   carbon.    This  can  be  done  either  by  thermal
regeneration in a  rotary  kiln  or  multihearth  incinerator,  or  by
chemical  regeneration  by  using  oxidizing  agents  such as hydrogen
peroxide or acids and bases.

The application of carbon adsorption has been mainly in organic  waste
treatment.   Recently,  there are studies indicating the effectiveness
of carbon adsorption in removing mercury, cadmium, cyanide,  chromium,
lead, nickel, zinc, arsenic, and copper (30, 31).

An  interesting  development in carbon technology is its use after the
wastewater is ozonized.  This combination (known as  Bacteriologically
Activate  Carbon  or  BAC)  has proved effective in treating otherwise
biologically  inactive  organic  compounds.   The   process   involves
chemical modification of the organics by the ozone.  Maintenance of an
aerobic  region  on the carbon allows a biologically activated film to
develop and the modified organics  are  further  treated  by  a  mixed
process of biological oxidation and carbon adsorption.  The system has
the advantage of being a potential add-on to existing BPT systems, and
should  be cost effective since it has been found that the carbon only
needs regeneration at infrequent intervals.
No industrial applications of
research is under way (32).
this  technology  are  known,  although
Bacteriologically  Activated Carbon is a very attractive potential BAT
technology for the removal of organic  toxic  pollutants  from   waste
streams,  although  no  application to the inorganic chemical industry
subcategories studied in this report was found.
                                   84

-------
                            i
               PRODUCT
                WATER
       CCtOMCRATE WASTE
Figtare 7-3.  Electrodialysis process,
                 85

-------
                                 TABLE 7-2.  COMPARISON OF REVERSE OSMOSIS CONCEPTS
CD




Plate-and-Frarae
Large tubes
Spiral

Polyamide hollow
fine fibers
Cellulose acetate
hollow fine
fibers
Packing
Density

-------
Fluoride Removal

The conventional method of  treating  fluoride-bearing  wastes  is  to
precipitate  the fluoride as calcium fluoride by the addition of lime.
The reaction is:
Ca(OH)
2F- = CaF
                            20H~
(31)
     Hydrated lime + fluoride ion = insoluble calcium fluoride
                                     + hydroxyl ion

Using this process alone, it is difficult to remove fluoride to  below
8 mg/1 due to the solubility of calcium fluoride (9, 33).  Adding alum
with  the  lime  generally  improves the removal efficiency.  Fluoride
ions are removed as follows*

     A1(OH)3 + F- » A1(OH)2F + OH-                         (32)

     Aluminum hydroxide = aluminum monofluorohydroxide
      + fluoride ion       + hydroxyl ion, etc.

     A1{OH)2F + F- = AKOHJF;. + OH-                        (33)

     Al(OH)Fz + F- = A1F3 = OH-                            (34)

Complexed fluorides are also adsorbed to some extent on  the  aluminum
hydroxide  surface and removed in the coagulation process (33).  Large
amounts of alum (5000  mg/1)  are  required  to  reduce  the  fluoride
concentration to below 1  ppm.

Activated  alumina has been shown to be effective in removing fluoride
and arsenic is wastewater (34) and fluoride  from  drinking  water  in
municipal  water  treatment practice (35-38).  Typically, the fluoride
content of raw water can be reduced  from  about  8  to  1  ppm  (38).
Application  of  activated  alumina to high fluoride industrial wastes
shows that a low ppm effluent can  be  achieved  (39),  although  high
capital  and  operation  costs generally limit the wide application of
this process.

Certain process  operations  used  in  the  manufacture  of  inorganic
fluoride  compounds  involve  the  use  of  sulfuric acid and starting
materials which contain silicate or borate impurities.  This may  lead
to    the    formation    of    waste    containing   fluorosulfonate,
hexafluorosilicate  or  tetrafluoroborate  complex   ions.    Although
tetrafluoroborate   is  usually  a  very  minor  constituent  and  the
hexafluorosilicate is readily  hydolyzed  in  treatment  systems,  the
fluorosulfonate  ion  is  fairly stable and presents a serious problem
where  low  levels  of  total  fluoride  are   required.    The   lime
precipitation  method is not effective in removing the fluorosulfonate
and the effectiveness of adsorption techniques is not known.
                                   87

-------
Chlorine Removal

The removal of residual chlorine (in  the  form  of  hypochlorite)  in
industrial  wastewater  is  normally  accomplished  by the addition of
sulfur dioxide or a related reducing agent such as sodium bisulfite or
sodium metabisulf ite.   Typical reactions are shown in Equations 35 and
36.
OC1~
                      = H2S04
                                                           (35
     Sulfur dioxide + hypochlorite ion = sulfuric acid
      •f water                             + chloride ion

     Na^SOj + OC1- = Na?S04 + Cl~                          (36)

     Sodium sulfite + = sodium sulfate +
      hypochlorite ion   chloride ion

Alternatively, hydrogen peroxide, although relatively  expensive,
also be used for dechlorination according to Equation 37.
                                                                   may
 OC1- = H0
                         0
                              Cl-
(37)
     Hydrogen peroxide + hypochlorite ion = water + oxygen +
                                             chloride ion

In  the  chlor-alkali  industry, certain wastewater streams may have a
sufficiently high loading of  chlorine  to  warrant  recovery  of  the
product  by  air  stripping,  steam stripping, or extraction by carbon
tetrachloride.  In some locations,  a  market  exists  for  sodium  or
calcium  hypochlorite solutions which can be generated by treating the
tail gases with caustic soda or lime.  This may serve as a  means  for
disposing  of  waste  chlorine  which cannot otherwise be economically
recovered.  As alternatives for waste chlorine disposal,  the  streams
may  be treated to form the hypochlorite and then decomposed thermally
or catalytically.  These technologies  are  discussed  in  Section  n
dealing  with the chlor-alkali  industry.  Chlorine residuals remaining
after the recovery and/or decomposition steps have been taken would be
amenable to treatment with reducing agents  such  as  sulfur  dioxide,
bisulfite, or hydrogen peroxide as described above.

-------

-------

-------
                              SECTION 8

          TREATABILITY ESTIMATES AND LONG-TERM DATA ANALYSIS


The Development of_ Treatability Estimates

Preliminary Analysis

The  review  of  technological  treatment  options  applicable  to the
removal of toxic  pollutants  has  led  to  the  conclusion  that  the
particular  contaminants  found  in the raw process wastewaters of the
subject  industries  can  be  effectively  controlled  by  the  proper
application  of  fairly  well-known  and  demonstrated techniques.  In
order  to  proceed  from  a  general  discussion  and  description  of
techniques to a detailed evaluation for each subcategory of the levels
of  removal  that  can  be  expected,  a  summary  is now presented of
selected treatability data for the 13 toxic metals.

The treated waste concentrations and removal efficiencies reported  in
the   literature   are  assumed  to  represent  the  best  performance
characteristics that can be obtained  under  the  specified  operating
conditions.    The  treatment  technologies  considered  can  thus  be
assigned a set of optimum conditions and  best  performance  estimates
for  removal  of  the  particular  toxic  metals  that are amenable to
treatment. Taking each metal in turn, Tables 8-1 through 8-10 give the
initial and final concentrations, the removal efficiencies, and the pH
conditions for different treatment technologies.  The best performance
estimates for metal removal are derived from the  tabulated  data  and
are  utilized  in  turn  as  the bases for making estimates of average
achievable performance.  The sequence of analytical steps is:

     1.   Review and analyze applicable performance data.

     2.   Estimate best performance under optimum treatment
          conditions.

     3.   Estimate average achievable performance under expected
          industrial operating conditions.

The  third  step  involves  the  consideration  of  treatment   system
variables   under   full-scale   operating  conditions  in  industrial
situations where  the  design  objective  would  be  the  simultaneous
removal of several waste load constituents.  Each industry designs for
maximum  removal  and/or  recovery  of the major process-related waste
substances and  utilizes  an  appropriate  technology  which  is  both
reliable  and  cost  effective.   Optimum treatment conditions for the
removal of a particular pollutant can rarely be achieved  consistently
and any given set of conditions will be somewhat less than optimum for
most, if not all, of the treatable constituents.  In any well-operated
production  facility  the  normal  variations in production rates, raw
material quality, the desired product mix in some cases,  and  contact
water  use  requirements may cause severe hydraulic and pollutant load
                                   91

-------
TABLE 8-1.  WASTE WATER TREATMENT OPTIONS AND PERFORMANCE DATA. SUMMARY -
            ANTIMONY AND ARSENIC REMOVAL

Treatment Technology
Antimony
Lime/Filter
Ferric chloride/Filter
Alum/Filter
Arsenic
Lime Softening
Sulfide/Filter
Lime (260 rag/1) /Filter
Lime (600 mg/1) /Filter
Ferric sulf ate
Ferric sulf ate
Lime/Ferric Chloride/
Filter
Activated alumina
(2 mg/1)
Activated carbon
(3 mgA)
Ferric Chloride
Ferric Chloride
pH

11.5
6.2
6.4
—
6-7
10.0
11.5
5-7.5
6.0
10.3
6.8
3.1-3.6
-
—
Initial
Concen-
tration
(mg/1)

0.6
0.5
0.6
0.2
-
5.0
5.0
0.05
5.0
3.0
0.4-10
0.4-10
0.3
0.6-0.9
Final
Concen-
tration
(mg/D

0.4
0.2
0.2
0.03
0.05
1.0
1.4
0.005
0.5
0.05
<0.4
<4.0
0.05
<0.13
Removal
(%)

28
65
62
85
-
80
72
90
90
98
96-99+
63-97
98
-
References

40
40
40
9,
9,
41
41
42
41
9,
43
43
9,
9,




10
10




10


10
10
                                    92

-------
TABLE 8-2.  WASTE WATER TREA3MENT OPTIONS AND PERFORMANCE DATA SUMVIARY -
            BERYLLIUM AND CAEMIUM REMOVAL

Treatment Technology pH
Beryllium
Lime/Filter 11.5
Cadmium
Line (260 mg/1) /Filter 10.0
Lime (600 mg/1) /Filter 11.5
Lime Softening 5-6.5
Lime/Sulf ide 8 . 5-11 . 3
Ferrous Sulfide (Sulfex) 8.5-9.0
Ferrite coprecipitation/ neutral
Filter
Initial
Concen-
tration
(mg/1)

0.1
5.0
5.0
0.44-1.0
0.3-10
4.0
240
Final
Concen-
tration
(mg/1)

0.006
0.25
0.10
0.008
0.006
<0.01
0.008
Removal References
(%)

99.4 40
95 41
98 41
92-98 8
98+ 44
99+ 7,8,11
99+ 5
                                    93

-------
TABLE 8->3.   WASTE WATER TREATMENT OPTIONS AND PERFORMANCE DATA SUMMARY -
             COPPER REMOVAL

Treatment Technology
Lime/Filter 8.
Lime (260 mg/1) /Filter
Lime £600 mg/1) /Filter
Ferric sulfate/Filter
Lime >
Lime
Alum 6.
Liroe/Sulfide 5.
Ferrous sulfide (Sulf ex) 8.
Ferrous sulfide (Sulf ex) 8.
Ferrite Coprecipitation/
FH
5-9.0
10.0
11.5
6.0
8.5
9.5
5-7.0
0-6.5
5-9.0
5-9.0
-
Initial
Concen-
tration
(mg/1)
3.2
5.0
5.0
5.0
10-20
3.0
3.0
50-130
3.2
4.0

Final
Concen-
tration
(rng/1)
0.07
0.4
0.5
0.3
1-2
0.2
0.2
<0.5
0.02
0.01
0.01
Removal
(%)
98
92
91
95
90
93
93
-
99
99+
99+
References
8
41
41
41
9,10
45
45
44
8
7,8,11
5
  Filter
                                     94

-------
TABLE 8-4.  WASTE WHER TSEKO^EtJS OPTICKS AND PERFOIMftNCE DRXA SOfAEQT -
            CHHCMIIM III AND GHPCMIIM VI RMDVK.
Treatment Technology
QironduD
Lima (260 mg/1) /Filter
Lima (600 rag/1) /Filter
Reduction/Lime
RBduc&icn/IjjQB
Lime Softening
Limn/Filter
Lima
Lin
Ferrita ooprecipitation/
Filter
Ferric sulfate
Ferric sulf ata/Fllter
ChranJum VI
Activated carbon
(pulverized, Pitts-
burgh type HC)
Same as above
Activated carbon
(granular)
Ferrite coprecipitatiai
Sulfur (Uorifla reduction
Bisulfite reduction
pH
10.0
U.5
7-8
7-8
10.6-11.3
7-9
9.5
9.5
	
6.5-9.3
	
3.0
2.0
6.0
—
	
	
Initial
Concen-
tration
tng/1)
5.0
5.0
140 (as
Cr VI)
1300 (as
Cr VI)
—
—
15
3.2
25
	
5.0
10
10
3
0.5
	
	
Final
Concen-
tration
ftng/1)
0.1
0.1
1.0
0.06 Crm
0.15
0.05
0.1
<0.1
0.01
—
0.05
1.5
0.4
0.05
not
detectable
0.01-0.1
0.05-1.0
Renewal
98
98
—
—
98+
	
	
	
	
98+
99
35
96
98
—
	
	
References
41
41
9,10
3,9,10
46
47
45
45
5
46
41
48
48
41
5
9,10
9,10
                        95

-------
TABLE 8-5.  WASTE WATER TREA3MEWT OPTIONS AND PERFORMANCE DATA SUKMAIOT -
            LEAD REMOVAL

Treatment Technology
Lime (260 rag/1)
Lime/filter
Lijne (260 mg/1) /Filter
Line (600 mg/1) /filter
Ferrous sulf ate/Filter
Sodium hydroxide (1 hour
settling)
Sodium hydroxide (24 hour
settling)
Sodium hydroxide/Filter
Sodium carbonate/Filter
Sodium carbonate/filter
Sodium carbonate/filter
Ferrous sulfide (Sulfex)
Ferrite coprecipitation/
pH
10.0
8.5-9.0
10.0
11.5
6.0
5.5
7.0
10.5
10.1
6.4-8.7
9.0-9.5
8.5-9.0
—
Initial
Concen-
tration
(mg/1)
5.0
189
5.0
5.0
5.0
—
—
1700
1260
10.2-70.0
5.0
189
480
Final
Concen-
tration
(mg/1)
0.25
0.1
0.075
0.10
0.075
1.6
0.04
0.60
0.60
0.2-3.6
0.01-0.03
0.1
0.01-0.05
Removal
{%)
95.0
99.9
98.5
98.0
98.5
	
	
99+
99+
82-99+
99+
99.9
99.9
References
41
5
41
41
41
10
10
49
49
10
9,10
8
5
  Filter
                                      96

-------
TABLE 8-6.  WASTE WATER TREATMENT OPTIONS AND PERFORMANCE DATA. SUMMARY -
                           MERCURY II REMOVAL
Treatment Technology pH
Sulfide
Sulfide 10.0
Sulfide/Filter 5.5
Sulfide/Filter 4 . 0
Sulfide/Filter 5.8-8.0
Ferrite coprecipitation/
Filter
Activated Carbon
Activated Carbon/Alum
Activated Carbon
Initial
Concen-
tration
-Cng/D
0.3-50.0
10.0
16.0
36.0
0.3-6.0
6.0-7.4
0.01-0.05
0.02-0.03
0.06-0.09
Final Removal
Concen- (%)
tration
(mg/D
0.01-0.12
1.8 96.4
0.04 99
0.06 99.8
0.01-0.125 87-99.2
0.001-0.005 99.9
<0.0005
0.009
0.006
Reference!
9,10
50
50
50
50
5
9,10
46
50
                                   97

-------
TftBTE 8-7.  WASTE WATER TREATMENT OPTIONS AND PERFORMANCE DATA SUMMARY -
            NICKEL REMDVAL

Treatment Technology pH
Lime 8.5-9.0
Lime (260 mg/1) /Filter 10.0
Lime (600 mg/1) /Filter 11.5
Caustic Soda/Filter 11.0
Ferrous sulfide (Sulfex) 8.5-9.0
Ferrite coprecipitation
Initial
Concen-
tration
(mg/1)
75
5.0
5.0
-
75
1000
TABLE 8-8. WASTE WATER TREATMENT OPTIONS AND
SILVER REMOVAL
Final
Concen-
tration
(mg/1)
1.5
0.3
0.15
0.3
0.05
0.20
PEKEK3BMANCE
Removal References
98 8
94 41
97 41
49
99,9 8,11
99.9 5
DATA SUMMARY -

Treatment Technology pH
Sodium hydroxide 9.0
Ferric sulfate (30 mg/1) 6-9
Lime Softening 9.0-11.5
Chloride precipitation
(alkaline chlorination
in the presence of
cyanide)
Ferric chloride/Filter 6.2
Sulfide precipitation 5-11
Initial
Concen-
tration
(mg/1)
54
0.15
0.15
105-250
0.5
-
Final
Concen-
tration
(mg/1)
15
0.03-0.04
0.01-0.03
1.0-3.5
0.04
-
Removal References
72 13
72-83 46
80-93 46
97+ 9,10
98.2 40
very high 9,10
                                    98

-------
TABLE 8-9.  WASTE WATER TREATMENT OPTIONS AND PERFORMANCE DATA SUMMARY -
            SELENIUM AND THALLIUM REMOVAL

Treatment Technology
Selenium
Ferric chloride/Filter
Ferric chloride/Filter
Alum/Filter
Ferric sulfate
Ferric sulfate
LinieAilter
Lime/Filter
Thallium
Lime/Filter
Ferric chloride/Filter
Alum/Filter
PH
6.2
6.2
6.4
5.5
7.0
11.5
11.5
11.5
6.2
6.4
Initial
Concen-
tration
(mg/1)
0.1
0.05
0.5
0.10
0.10
0.5
0.06
0.5
0.6
0.6
Final
Concen-
tration
(mg/1)
0.03
0.01
0.26
0.02
0.03
0.3
0.04
0.2
0.4
0.4
Removal
(%)
75
80
48
82
75
35
38
60
30
31
References
40
40
40
51
51
40
40
40
40
40
                                     99

-------
TABLE 8-10.  WASTE WATER TREATMENT OPTIONS AND PERFORMANCE DATA SUMMARY -
                               ZINC REMOVAL
Treatment Technology
Lime/Filter
lime (260 mg/1)
Lime (260 mg/1) /Filter
Lime (600 rng/1)
Lime (.600 mg/1) /Filter
Lime/Filter
Sodium hydroxide
Sulfide
Ferrous sulfide (Sulfex)
Perrite coprecipitation
pH
8.5-9.0
10.0
10.0
11.5
11.5
-
9,0
-
8.5-9.0
-
Initial
Concen-
tration
fmg/D
3.6
5.0
5.0
5.0
5.0
16
33
42
3.6
18
Final
Concen-
tration
(mg/1)
0.25
0.85
0.80
0.35
1.2
0.02-0.23
1.0
1.2
0.02
0.02
Removal
(%)
93
83
84
93
77
-
97
97
99+
99+
References
8
41
41
41
41
5
13
5
8,11
5
                                    100

-------
input  excursions  which  at  best  can  be  moderated  by   effective
equalization  in the treatment system.  This is considerably less of a
problem in batch treatment than with a continuously operating  system.
The  latter requires continuous feedback monitoring for pH control and
chemical dosage in order  to  maintain  the  effluent  quality  within
acceptable limits for a number of parameters.  Under these conditions,
the   30-day   averages  derived  from  the  actual  treated  effluent
monitoring data (NPDES, etc.) would equate to what has been identified
in Step 3 above as the estimated 30-day achievable  performance  using
the same general treatment technology.

The  estimated  ranges of average achievable performance are presented
in Table 8-11.  In formulating the proposed regulations, these  values
were  used  as  maximum  30-day  averages  in  cases  where there were
insufficient data from sampling or long-term monitoring of the  actual
industry discharges.

Statistical  evaluation  of  long-term monitoring data is described in
Section 8.3 and the results are presented in Appendix A where  various
derivative   quanitites   such  as  long-term  averages  and  standard
deviations are tabulated.

Final Analysis

Following publication of the proposed regulations on July 24, 1980 (45
FR 49450) additional data on performance of the BPT  and  BAT  options
for  several  subcategories were evaluated and eventually incorporated
into the basis for the final regulations.  The sources  of  additional
data include the following:

     A.   Treatability Study for the Inorganic Chemicals Manufacturing
          Point Source, EPA 440/1-80-103, July, 1980.

     B.   Industry comments on the proposed regulations - The  written
     comments  received by EPA as well as comments given orally at the
     public hearing on proposed pretreatment  standards  (October  15,
     1980)  are part of the official public record of the rule making.
     The comments are summarized and responses are given in "Responses
     to Public Comments, Proposed  Inorganic  Chemicals  Manufacturing
     Effluent Guidelines and Standards," which is a part of the Record
     for this rule.  Individual comment documents or letters are cited
     in this report where they are used as sources of information.
     C.   Treatability   Manual,   Volume   III
Technologies
for
     Control/Removal of Pollutants, EPA 600/8-80-042c, July, 1980.
Table  8-12  presents  tabular  summaries  of  the  available industry
treatment performance data for most  of  the  priority  toxic  metals.
These  include  estimated long-term averages in cases where there were
sufficient data given to utilize  the  Maximum  Likelihood  Estimation
method  for  calculating  statistical  parameters  as indicated in the
footnotes.  Overall arithmetic medians and averages are also given for
metals where five or more individual data sets were available.
                                   101

-------
                  TABLE 8-11.  ESTIMATED ACHIEVABLE MAXIMUM 30-DAY AVERAGES FOR THE APPLIED TECHNOLOGIES
o
Kl


Antimony, Sb
Arsenic V
Beryllium, Be
Cadmium, Cd
Copper, Cu
Chromium III,
Cr+3
Lead, Pb
Mercury II,
Hg
Nickel, Ni
Silver, Ag
Selenium, Se
Thallium, Tl
Zinc, Zn
Lime
Settling
0.8-1.5
0.5-1.0
0.1-0.5
0.1-0.5
0.5-1.0
0.1-0.5
0.3-1,6
Lime
Filter
0.4-0.8
0.5-1.0
0.01-0.1
0.05-0.1
.0.4-0.7
0.05-0.5
0.05-0.6
Final
Sulfide
Filter

0.05-0.1

0.01-0.1
0.05-0.5

0.05-0.4
Concentrations (mg/1)
Ferrite
Coprecip- Soda Ash Soda Ash Alum
itation Settling Filter
Filter



<0.05
<0.05
0.01
0.20 0.4-0.8 0.1-0.6
0.01-0.05 <0.01
0.2-1.5
0.4-0.8
0.2-1.0
0.2-1.0
0.5-1.5
0.1-0.5
0.2-0.4
0.1-0.5
0.1-0.5
0.4-1.2
0.05-0.5
0.05-0.2


0.02-1.2



0.2-0.5
0.02-0.5
                                                       (continued)

-------
             TABLE 8-11   continued
H
O


Arsenic V, As
Chronium VI,
Marcury II,
Hg
Final Concentrations (mg/1)
Ferric Activated SO2 Bisulfite Lime/FeCl,
Chloride Carbon Reduction Reduction Filter
0.05-0.5 0.3 0.02-0.1
0.1 0.01-0.1 0.05-0.5
0.01
Alkaline
Chlori-
nation



Silver, Ag


Selenitim, Se


Thallium, Tl


Cyanide  (Free),
 CNA
0.05-0.1


0.05-0.1


  0.7
                                                                                                    0.1-0.5

-------
An  industry  long-term  average  effluent  concentration   was   then
estimated  for  each  pollutant/treatment option combination for which
sufficient data were available.  For copper and  nickel,  the  average
values  for  lime/settling were adjusted upward from 0.32 to 0.40 mg/1
in order to show a larger decrease when filtration is added.   In  the
case  of  chromium,  the  average with filtration was adjusted to 0.16
mg/1.   Plants presently practicing filtration are generally those with
higher raw waste concentrations of  metals  in  comparison  to  plants
which  can  achieve adequate treatment without filtration.  This tends
to reduce the observed differences in  performance  with  and  without
filtration and, therefore, understates the potential benefit of adding
filtration  to  a  particular  lime/settling  system.   The  estimated
achievable long-term average concentrations, as shown in  Table  8-13,
generally fall within the estimated range of the corresponding maximum
30-day averages in Table 8-11 which were derived from literature data.
Thus,  there is substantial agreement between the two sets of estimates
and  there  is  good  reason  to conclude that the lower limits of the
treatability ranges in Table 8-11 are  actually  more  like  long-term
averages  than  maximum  30-day  averages  for the inorganic chemicals
industry.   The  final  toxic  metal  regulations  are  based  on  the
estimated achievable long-term average concentrations in Table 8-13 in
cases  where there are insufficient industry specific performance data
available.  The numerical limitation in  each  case  was  obtained  by
multiplying  the  long-term  average  concentration by the model plant
unit flow rate and an appropriate variability factor.  The variability
factors are selected to represent as accurately as possible the actual
full-scale  treatment  system's  variability  under  normal  operating
conditions.

It  is  understood  that  in  each  subcategory plant treatment system
conditions, particularly where chemical precipitation is involved, are
usually optimized for the removal of only one metal.  Other metals may
be removed incidentially under  the  same  conditions  although  their
removal  efficiencies may not be optimal.  An example is the prevalent
use of sulfide precipitation/filtration technology for the removal  of
mercury.   The  precipitation is normally carried out under neutral to
moderately-acid conditions in order to limit the  amount  of  residual
sulfide   in   the   system  and,  depending  on  specific  raw  waste
characteristics, to obtain desirable solid properties for  filtration.
Under  these  conditions, the incidental removals of other metals such
as nickel and zinc are not at  their  maximum  efficiencies,  but  are
still effective.

The industry performance data summarized in Table 8-12 for many of the
toxic  metal/treatment  combinations  express  an  observed incidental
removal rather than an optimum removal.  This  provides  an  empirical
basis  for  estimating  practical  control  levels  for  metals  under
off-optimum pH conditions in either alkaline precipitation or  sulfide
precipitation  systems.  The Agency does not regard the implementation
of more than one optimized metal removal step as necessary to meet the
final BPT/BAT regulations.
                                   104

-------
                                                               TABIE 8-12.   INDUSTRIAL WRSTE WATER TREATMENT SYSTEM PERFOiWftNCE  —
                                                                                    OF EFFLUENT OONCEHTRKTION DATA ON TOXIC METALS
(1)
O
I/I
Lime/Clarification f.i time/Pi ltration>.i
(mg/1) Source1 ' (mg/1) Source1*1
Antimony
Arsenic

Beryllium
CadmiuQ



Mercury







Seleniun
Silver
Thallitm
Cturcmium
III



















0.18
0.080


0.060
0.080
0.12












0.040
0.050
0.050
0.070
0.071
0.072
0.080
0.15
0.18

0.26

0.35
0.36

0.43

0.81
1.8
0.
0.
PM*5' ND
NSMt5) 0.30 CP(7t
0.038 CS(8>
ND ND
TDS 0.076 MF
NEM(5>
MF(11)

ND ND







ND ND
ND ND
ND ND
IS<5) 0.037 cp*15*
NEM(5) 0.046 TDCtl5>
CfiD* ' 0.072 SDC' '
TQgdO) Q>20 soc*15*
CAD(15> 0.28 MF(11)
TPC(16> 0.33 CP(7)
TMt5) 0.44 CP(21)
MF1
TDC<15)
(151
SDC113'
(171
CAD(A"
is'5'
(15)
CPU '
(151
SDC1 '
IS<5>
15 median 0.20 median
32 average 0.20 average
Sulfide/FiltrationJ.)
(mg/1) Source"1
0.23 CfiM(6>
0.17 CfiM(9)
0.096 CflM(6)
ND
ND


(Q)
0.020 CAM11"
*1 O\
0.022 CAM' '
(Qt
0.036 CAM1"'
(i ii
0.057 CflM1 '

ND
0.070 CRM<9>
ND
ND


















NA

, , . . , (2) . (2)
Lijn(5/Clarificatic«,4j larae^'iltration^;
(mg/1) Source (mg/1) Source
Copper 0.030
0.030
0.038
0.060
0.070
0.070
0.080
0.090
0.10

0.14

0.54

0.70

1.1
1.5
IS(5) 0.035 CP(7)
CAD(14) 0.13 CP<21>
OMD 0.17 CS
OMJ(5) 0.23 CS<15)
NFM15' 0.25 is'5'
SEP*5' 0.37 MF*11'
CM)'5' 0.90 MF*11'
C«D(5>
OMD(5>
(111
MF* '
(171
CAD1 '
(11
N£M
(151
CS1 '
CS<15)
O.OS5 median 0.23 median
0,32
I^ad 0.017
0.10
0.15
0.19
0.20
0.15
0.13














average 0,30 average
HF(19) 0.038 MFtU)
MF(U) 0.11 CPt7)
OMD(5) 0.41 CPt211
HEM(5)
NFM(5)
median HA
average














Sulfide Filtration!?!
(mg/1) Source1 4)
0.033 CAM(6)
0.056 CftM(9>
0.24 CAMtU1
0.43 CAM(17)














NA

0.032 cao'13'
0.12 CftM(9)
0.16 CfiM(6>
0.46 CADtl5)

NA















                                                                                                                                          (Continued)

-------
TABLE 8-12   continued
(2)
Lime/Clarification :..
Enrj/1) Source1 '
NiOfl 0.020
0.050
0.10
0.17
0.20
0.20
0.25
0.26
0.31
0.33
0.50
1.4

0.23
0.32


NOTES:
IS(5)
CAD{14)
NEM<5>
ffi(l5J
C3(15)
CAD<17J
*<5)
HEM(5)
HS(15>

median
average



Lame/FiltrationJ4J. Sulfide/Filtrationj?,
(mg/1) Source1 ' (mg/1) Source1 '
0.090 §DC118J 0.022 CAM*9'
0.11 CS(15) 0.074 CAM(6)
0.13 CS(15>
0.19 US '
0.20 CS(8)
0.59 MF(llt
0.80 NS(20)



0.13 median MA
0.30 average


(21
Lime/Clarification:,:
(ragA> Source1 J
Zinc 0.020 QMD '
0.040 OMOt5)
0.040 FI(5)
0.10 CAD(14J
0.11 TM(5)
0.15 TC6
0.20 NfM* '
0.24 IS(5)
0.25 ISt5)
0.35 CMJ(5)
0.39 MF1U)
0.54 CM)
0.55 HF(19>
0.60 NFM*5'
8.2 PMI5)
0.20 median
0.7S average

(1) Influent or raw waste concentrations of metals are at treatable levels; i.e., higher than the corresponding tn
Lime/FiltrationJ?: Sulfide/Piltrationj^j
(mg/1) Source1" (mg/1) Source14'
0.018 CS*8) 0.090 TM(5J
0.058 CP*7J 0.13 CAM(9)
0.11 SDC(1B1 0.15 CAM16'
0.25 Mf'11*
0.57 MF(11)





0.11 median NA
0.20 average

aatability ranges given in Table 8-11.
     All effluent concentrations are measured off treatment and are expressed as total (dissolved plus suspended)  for each metal.

(2)  Lime/Clarification and Lime/Filtration treatment means equalization of raw waste influent stream(s)  followed by alkaline precipitation using lime
     or caustic soda, solids removal by sedimentation or clarification,  and either discharge of the clarified effluent  directly or discharge of the
     filtrate after passage of the clarified effluent ttirough a dual media filter or its equivalent.

(3)  Sulflde/Filtration refers to a direct treatment of the equalized  raw waste influent by aulfide addition (usually in the form of sodium sulfide or
     bisulfide)  under conditions ranging from pH 5 to 11 followed by settling and/or filtration by  filter press or activated carbon column.

(4)  Source Codes:
          CflD    Oilor-Alkali,  Diaphragm Cells
          CAM    Qilor-Alkali,  Mercury Cells
          CS     Copper Sulfate
          Cp     Chrome Pigments
          Fl     foundry Industry
          HP     Hydrofluoric Acid
IS     Iron and Steel
MF     Metal Finishing (including electroplating)
HEM    Nonferrous Metals
NS     Nickel Sulfate
CM)    tee Mining and Dressing
PM     Paint Manufacturing
SEP    Steam Electric Power Generating
SEC    Sodium Dichronate
TEC    Titaniun Dioxide - Chloride Process
TDS    Titanium Dioxide - Sulfate Process
1M     Textile Mills
(5)  U.S.  Environmental Protection Agency,  Treatabllity Manual,  Vol.  ill. Technologies for Control/Removal of Pollutants, EPA  BOO  8-80 042 c, July, 1980.

(6)  This document.  Table 11-16.

(7)  This docutent,  Table 16-9.

(8)  This dcxament.  Table 21-11.
                                                                                                                      (Continued)

-------
TABLE 8-12    continued
NOTES;  continued

(9)  Olin Corporation, Chemicals Group, Charleston, TO.   Letter to Mr. Elwood E. Martin,  U.S.  EPA,  Effluent Guidelines Division,  Washington,  D.C.,
     October 20, 1980.  Maximuu likelihood estimates of the long term averages from Olin mercury treatment effluent data by Jacobs Engineering Group,  Inc.

(10) This docunent. Table 14-30.

(11) Hamilton Standard, Division of United Technologies Corp., Windsor Locks, CT.  Letter to Mr. Richard Kinch,  U.S. EPA, Effluent Guidelines Division,
     Washington, D.C., ttovaiber 25, 1980.  Tabulations of statistical parameters derived iron historical data on the metal finishing industry.

 (12) The Chlorine Institute, Inc., New York, N.Y.  Letter to Mr. G. E. Stigall, U.S. EPA, Effluent Guidelines Division, Washington, D.C., May 28, 1979.
     Attachment "C", a tabular summary of mercury treatment effluent data.

 (13) PPG Industries, Inc., Pittsburgh, PA.  Letter to Mr. Elwood E. Martin, U.S. EPA, Effluent Guidelines Division, Washington, D.C., January 2, 1981.
     Maximun likelihood estimates of the long term averages from PPG mercury and lead treatment effluent data by Jacobs Engineering Group, Inc.

 (14) This document. Table 11-37.

 (15) U.S. Environmental protection Agency, TreatabiUtY Studies for the Inorganic Chemicals tfanufacturincf Point Source Category,  EPA 440/1-80/103,  July,  1980.
     Maximum likelihood estimates of long term averages from treatability data by Jacobs Engineering Group, Inc.

 (16) This document. Table 14-12.

 (17) Diamond Shamrock Corporation, Dallas, TX.  Letter to Mr. Elwood E. Martin, U.S. EPA, Effluent Guidelines Division, Washington, D.C., October 22,1980.
     Tabular sunmary of highest values from treatment effluent during one month of monitoring.

 (18) This document. Table lfr-13.

 (19) This docunent. Table 12-22.

 (20) This document, Table 22-10.

 (21) This document. Appendix A.
     ND =  No  data available
     HA =  Not applicable

-------
TABLE 8-13.  ESTIMATED ACHIEVABLE LONG TERM AVERAGE
             CCNCEMTRATICNS FDR TOXIC METALS WITH
             EFT OR BAT TREATMENT OPTIONS

Toxic
Metal
Antimony
Arsenic
Beryllium
Cadmium
Chrcmium
Cbpper
Lead
Mercury
Nickel
Selenium
Silver
Thallium
Zinc
(1> ID:
(2) m.
Lime/Clarification
fag/1)
n>(1>
ID
ND
0.10
0.32
0.40
0.15
ND
0.40
ND
ND
ND
0.80
Lime/Filtration
(mg/1)
ND<2>
ID
ND
ID
0.16
0.30
ID
ND
0.30
ND
ND
ND
0.20
Sulfide/Piltration
(mg/1)
ID
0.15
ND
ND
ND
0.20
0.10
0.034
ID
ND
ID
ND
0.12
Insufficient data for a reliable estimate
NTn s3-i^-* 3^ia -1 1 r^Vll t*


                         108

-------
Selection of Toxic Metal Control Parameters

Control Parameters for Hydroxide Precipitation

Section 7.2 of this report describes hydroxide  precipitation  as  the
most  widely-used  technology  for  removing  trace metals from waste-
waters.   Out  of  the  thirteen  toxic  metal  pollutants,  two  have
hydroxide/oxide   solubilities   independent  of  the  -1-14  pH  range
(selenium  and  thallium)  and  two   have   minimum   hydroxide/oxide
solubilities  over  a wide pH range (antimony at pH 2-10.4 and mercury
at  pH  4-12).   Arsenic  is  removable  by  precipitation  with  lime
(probably  as calcium arsentate) in the presence of excess calcium ion
under neutral to  alkaline  conditions.   The  remaining  eight  toxic
metals  have  minimum hydroxide/oxide solubilities only over narrow pH
ranges (see Figure 7-1).  Lead may also be  effectively  treated  with
carbonate  (soda  ash,  Na2C03) to form insoluble basic lead carbonate
precipitates.

It is clear from the range of optimum pH's illustrated in  Figure  7-1
that no single pH exists which can effectively provide optimum removal
of  all  eight  of  these metals.  Relatively effective removal can be
obtained by dividing the  eight  metals  into  two  groups.   Group  A
consists  of  beryllium,  chromium,  copper,  lead, and zinc.  Group B
consists of cadmium, nickel and silver.  Because they rarely occur  at
treatable  levels  and,  therefore,  rarely require removal, one metal
from each group (beryllium and silver)  can  be  eliminated  from  the
selection  of  an optimum pH range for- each group.  The information in
Figure 7-1 was used to determine the solubility of the  six  remaining
metals  at  unit  pH  increments  from  8.5  to  11.5.  These data are
presented in Table 8-14.

Table 8-14 indicates that control of any  metal  of  Group  A  in  the
8.5-9.5  pH  range  should  control  the  other  members of the group.
Control of any metal of Group B  in  the  10.5-11.5  pH  range  should
control  the  other  members  of  the  group.   Control of metals from
different groups will depend on the details of  each  case.   Possible
approaches  to  controlling metals from different groups might involve
the use of the intermediate 9.5-10.5 pH range or the  control  of  one
metal  in  one group when the theoretical solubilities of the metal or
metals in the other group are low throughout the 8.5-11.5 pH range.

Control Parameters for Sulfide Precipitation

Section  7.2  of  this  report  describes  sulfide  precipitation   as
superior  to  hydroxide  treatment  for  the  removal of several toxic
metals.  The main application of sulfide precipitation is  in  mercury
removal  and  mercury, therefore, is the obvious choice as the control
metal for this technology.    Figure  7-2  points  out,  however,  that
mercury is the most insoluble of the toxic metal sulfides and that the
solubilities  of  the  metal  sulfides are strongly dependant upon pH.
Control of mercury in the acid  pH  range  may  result  in  less  than
optimum  control  of  the  least insoluble metal sulfides.  Therefore,
control of a second metal that is present in treatable  concentrations
                                   109

-------
          TABLE  8-14.  THEORETICAL SOLtBILITIES OF TOXIC METAL
                       HyDROXTEES/OXEDES AT VARIOUS pH VALUES

pH
Metal
Group A
Cr"^
Oa^
Pb^
zn~
Group B
Cd"
N1++
8.5 9.5 10.5 11.5
Concentration

0.030(1) 0.20
0.00010 0. 000080 (1)
8.0 0.50a>
0.60 0.070C1)

>10 1.0
1.0 0.010
(mg/1)

1.0 9.0
0.00050 0.0020
4.0 >10
0.50 3.0

0.010 0.0010(1)
o.ooioa) o.oio
(1)
    Lowest value
                                    no

-------
and  that  is  among  the  least insoluble of the toxic metal sulfides
could  give  greater  assurance  that  the  metals  without   effluent
limitations were also being removed.  However, it could also result in
higher  mercury discharges.  Operation of sulfide precipitation in the
neutral or slightly alkaline range should result in acceptable removal
of all toxic metal sulfides as  well  as  minimizing  the  problem  of
hydrogen  sulfide  evolution.   Soluble  polysulfide  formation can be
prevented by avoiding the very alkaline pH range and by close  control
of excess sulfide.

The Use Of Historical Pollutant Data
Determination
Performance
of Effluent Limitation Guidelines Based Upon Historical
In cases where there has been long-term monitoring  of  the  pollution
levels in the effluent stream discharged by a plant, it is possible to
assess  in-plant  treatment performance through analysis of historical
data that has been collected for this purpose.  The appropriateness of
standards  constructed  from  data  collected  from  a  single   plant
performance   is,   of   course,  dependent  on  the  plant's  current
performance in relation to the performance  of  other  plants  in  the
manufacturing subcategory.  As economically feasible alternative waste
treatment   technologies   become   available,   pollutant   discharge
guidelines need to be reviewed and revised to reflect these advances.

Statistical analysis of historical  monitoring  data  is  required  to
assess  a  plant's  ability  to  discharge  within set guidelines.  To
perform this analysis certain assumptions must be made  regarding  the
nature   of   applicable  statistical  or  probabilistic  models,  the
constancy of the operation of the treatment facility, and the  quality
of the monitoring methods.

The statistical analyses contained in this development document belong
to  either  of  two  principal types:  those for daily observations of
pollutant concentrations, and the others for 30-day average  pollutant
levels.

Tables  in  Appendix  A  provide  a summary of traditional descriptive
measures, i.e., number of  observations(No),  mimima(Min),  arithmetic
average(Avg),  maxima(Max),  and  coefficient  of  variation(CV).   In
addition, a descriptive statistic, the variability  factor,  pertinent
to  the development of performance standards for pollution monitoring,
is included.  These tables, prepared for both  daily  measurements  as
well  as  30-day averages, are statistical summaries derived from data
offered by industry in response  to  Section  308-Questionnaires,  and
offered in comments on the proposed regulations.  Data in these tables
are   representative   of   currently   achieved  pollutant  discharge
performance levels in the several plants presented.

Formulation of variability factors to  be  used  in  determination  of
limitation   guidelines   based   upon   historical   performance  was
accomplished by employing standard statistical analysis  of  the  data
                                   111

-------
resulting  from  long-term monitoring of effluent stream discharges of
plants in the inorganic chemical manufacturing  subcategory.   in  the
following   paragraphs   are  presented  details  of  the  theory  and
derivation of these  statistical  procedures,  and  of  the  resulting
formulae  which  relate  variability  factors  to  estimated long-term
parameter averages, standard deviations,  coefficients  of  variation,
and  "Z-values"  computed  from  the  normal probability distribution.
These details are given both for the analysis applying to daily maxima
criterion and for that applying to 30-day averages.

The term "variability factor" is used in referring to the multiple  of
the  long-term  average  which  is  used  in  formulating  performance
standards.  This  factor  allows  for  variation  in  pollution  level
measurements due to sampling error, measurement error, fluctuations in
the  amount  of  the  pollutant  in  raw  materials, and other process
variations.

In the recording of actual  data,  as  reported  by  industrial  point
sources  in their responses to 308 Questionnaires, certain data values
were entered as "less than" detectability limits.  In these cases, the
sample of monitoring data has been "censored" in the process  of  data
recording since only the threshold value has been retained (i.e., if a
pollutant  concentration  was  reported  as  <0.050 mg/1, the value of
0.050 mg/1 was used).  In the statistical analysis of monitoring data,
censored values were included with  measured  values  in  the  sample.
This  practice  provides  a  reasonable  approach,  both for assessing
industry's capability to perform and environmental concerns for  valid
pollutant limitations.

First,  since censoring was done only for "less than" bounds, any bias
from their inclusion would cause a slight increase  in  the  long-term
average,  moderately  affecting  (in  the direction of leniency toward
industry) the estimate of long-term average pollution levels.

On the other hand, the use of censored values combined  with  measured
values  tends  to reduce the variability slightly {or in the direction
of less leniency toward industrial point sources).  For  illustration,
if  the  sample  consisted  solely  of  censored values, the estimated
long-term average might be  slightly  overstated.   Nevertheless,  the
point   source   ought  have  no  difficulty  with  the  threshold  or
detectability limit as a performance  guideline,  since  none  of  the
historical data exceeded that limit.

Assumptions Concerning Daily Pollutant Level Measurements

In  the  formulation  and  calculation  of  the  following performance
standards, individual sample measurements  of  pollutant  levels  were
assumed  to  follow  the  lognormal  distribution,  a  well  known and
generally accepted statistical probability  model  used  in  pollution
analyses.   Under this assumption the logarithms of these measurements
follow  a  normal  probability  model;   It  was  also  assumed   that
monitoring  at  a  given plant was conducted responsibly and in such a
way  that  resulting  measurements  can  be  considered  statistically
                                    112

-------
independent  and amenable to standard statistical procedures.  A final
assumption was that treatment facilities and monitoring techniques had
remained substantially constant throughout the monitoring period.
As an indication  of  the  appropriateness  of  assuming  a  lognormal
distribution  for  daily  measurements,  the  plot  of  the cumulative
distribution of logarithms of daily  effluent  concentration  data  on
normal probability paper is illustrated in Figure 8-1.

The  linearity  of  the  cumulative plot indicates the degree to which
actual monitoring data are in agreement with the theoretical lognormal
model for their distribution.

In addition, Figure 8-2, reproduced here from  a  report  prepared  by
industry  for  consideration by EPA, also demonstrates the validity of
the lognormal assumption for daily data.

In the analysis of daily data, the inherent  variability  of  measured
pollutant  levels  in  the  effluent  stream  from  inorganic chemical
manufacturing processes must  be  incorporated  in  calculating  upper
limits  for  daily  pollutant discharge levels.  Even well treated and
controlled plants may experience some days  when  an  atypically  high
level  of  pollutant  discharge  is present in their waste stream.  To
allow for this variability, performance standards must necessarily  be
set   above  the  plant's  long-term  average  performance.   However,
guideline limitations must be established at a  level  low  enough  to
ensure   adequate  control.   Establishing  effluent  guidelines  that
balance these factors means that occasional, infrequent  instances  of
non-compliance  are  statistically  predictable  at  well-operated and
maintained treatment facilities.  Since pollutant discharge  is  often
expressed  in  terms  of  average  level, it is convenient to describe
standards of performance and allow variability in term of multiples of
this average.  Such a method of computing standards  as  functions  of
multiples  of average level performance is explained below.  The ratio
of the pollutant standard level to the estimated long-term average  is
commonly called the "variability factor".

This   factor   is  especially  useful  with  lognormally  distributed
pollutant levels because its value is  independent  of  the  long-term
average, depending only upon the day-to-day variability of the process
and  the  expected  number  of  excessive  discharge  periods.   For a
lognormal population, the variability factor  (P/A),  the  performance
standard P, and the long-term average A, are related by:

     ln{P/A) = S'(Z -SV2)
     where

     A.
"In"  represents  the
numerical quantity.
natural  logarithm  (base  e)  of
                                   113

-------
     B.    S'  is the estimated standard deviation of the logarithms  of
          pollutant  level  measurements.    In  the calculations which
          follow,  S1  is computed by the statistical procedure known as
          the "method of moments".

     C.    2 is a factor derived from the standard normal distribution.
          Z is chosen to give performance limitations which provide  a
          balance  between  appropriate  consideration  of  day to day
          variation in a properly operating plant and the necessity to
          ensure that a plant is functioning properly.

The value of 1 used for determining performance  standards  for  daily
measurements  of  pollutant  concentration  is chosen as Z=2.33.  This
Z-value  corresponds  to  the  99th  percentile   of   the   lognormal
distribution meaning that only 1 percent of the pollutant observations
taken from a plant with proper operation of treatment facilities would
be  greater  than the performance standard, P.  Use of this percentile
statistically predicts one incident of non-compliance  for  every  TOO
samples for a plant in normal operation.  Many plants in this industry
are required by their NPDES permits to self-monitor once per week.  At
this  frequency,  there  will  be 260 samples analyzed over the 5 year
life of the permit.  The use of the 99th percentile to establish daily
maximum limitations statistically predicts 2 to 3  incidents  of  non-
compliance per pollutant in 5 years.  This percentile has been used to
establish  daily  maximum limitations in all other guidelines proposed
or promulgated, and has been used for  daily  maximum  limitations  in
Inorganic Chemicals manufacturing.

A.   Calculation of Variability Factors

As mentioned above,  development  of  variability  factors  for  daily
pollution  level  measurements  was based on the assumption that these
data, {XI,X2,...Xn),  follow  a   lognormal  distribution.   When  this
distribution is not a precise model, lognormally based procedures tend
to  somewhat  overestimate  variability  and produce  liberal standards
which act to the benefit of permittees.

Following this assumption, if Yi=ln(Xi), where ln(Xi)  represents  the
natural logarithm or log base e of the pollution measurement, then the
Yi;  i=l, 2,...,n are each normally distributed.   If  A' and'S1 are the
mean  and  standard  deviation  of  Y=ln(X)  respectively,  then   the
probability  is k percent that an individual Y will not exceed A'+ZS',
where Z is the  k-th percentile of the  standard  normal  distribution,
e.g.   2=2.33   is   the   99th   percentile  of  the  standard  normal
distribution.   It follows that A'+ZS1  is the natural  logarithm of  the
k-th percentile of X and that the probability is k percent that X will
not exceed a performance standard P=exp(A'+S1(S'/2)).  The variability
factor VF, is obtained by dividing P by A, hence,

          VF *  P/A = exp(S'(S'/2)), and

          ln(VF) = ln(P/A) = S'(Z - SV2)
                                    114

-------
Ul
             30.0
             20.0
         H
f-4
B
              2.0
              1.0

               0.01     0.1      1   2   5   10   20  30 40  50 60  70  80    90   95   98 99
                                                                                    99.9
                                                   PERCENTAGE
                   Figure 8-1.    Cumulative distribution of daily concentrations of mercury in treated
                                 effluent from plant #251.

-------
o\
0.30
0.20







^ 0.10
S31
c 0.08
g 0.06
B 0.04
8 0.03
fi
R
§ 0.02
n






0.01











—




















-::



















=









































~ = ! :

.. .



















i:: —
", '. Zj




















*
•-








E =--:- : 	 •








bbHtttU^^KBitUHUJil
= = •• = = :! =•!: = !:; ! !i!
H 1 1 m-HrfTantTi^ltJi^^S
""ri::: i I-:.1: -..
= ;«::;:: : : : : : : • :
-«::::: :::::;::
£ - 	 	









	 	 -
_
rt J J t rl-t fi 1 1 T~l ' •.--- —

'

1

1
1 1
i
:• : ::: : : : : : : =
: ;;;•= :•:•:: = =
	 ,----








































































































fi


















































               0.01     0.1   ,512   5   10   20 30 40 50 60 70 80    90  95  98  99
99.9
                                                    PERCE2JTAGE
                  Figure 8-2.    Cumulative distribution of daily concentrations of cyanide in treated
                                 effluent from plant #765,

-------
To  estimate the VF for a particular set of monitoring data, where the
method of moments is used, S1  is calculated  as  the  square  root  of
ln(1.0  + (CV)2), where the sample coefficient of variation, CV + S/X,
is the ratio of sample standard deviation to sample average.

B.   Example Calculation of Variability Factors From Long-Term
          Data

     Given the following descriptive statistics for a particular
     parameter, as might be found for lead (mg/1) in Appendix A.
     No
Min
Avg
Max
cv
     128
0.002
0.068
0.100
0.609
     Calculate the estimated standard deviation of logarithms

     (S1)* = In (1.0 + 0.6092) = 0.315

     S * 0.56

     Then

     ln(P/A) = 0.56(2.33 - 0.56/2) = 1.148

     The variability Factor VF is,

     VF = P/A = exp(l.148) = 3.15

     The performance standard P;

     p = A(VF) = A(P/A) = (0.068)(3.15) * 0.21

That is, using the descriptive statistics for  a  pollutant  presented
above  and  the statistical approach just described, the daily maximum
limitation established for that pollutant in a guideline would be 0.21
mg/1.

The statistical distributions relevant for the anaysis of  daily  data
are shown in Figure 8-3.

The statistical interpretation of P, the performance standard, is that
one   estimates  that  99  percent  (for  the  selected  Z=2.33  value
corresponding to the 99th percentile) of  the  daily  pollution  level
measurements  will  not  exceed  P.  For large data sets, P is roughly
equivalent to an upper 99 percent confidence bound for  an  individual
daily measurement.

Assumptions Concerning 30-Day Average Pollutant Level Observation

While  individual  pollution  level  measurements  should  be  assumed
lognormally distributed,  that  assumption  is  not  appropriate  when
                                   117

-------
analyzing   30-day   averages.    These  averages  generally  are  not
distributed as lognormal quantities.  However, for averages  of  daily
(lognormal)  measurements, a statistical principle, the "Central Limit
Theorem", provides the basis for using the normal  probability  model.
Therefore,  the  methods  used  in  computing  historical  performance
characteristics for 30-day averages differ from those used  for  daily
samples.   In  this  case,  the sample coefficient of variation is the
primary determinant of the variability factor, and there is no need to
resort to logarithmic transformation.  Examples of the appropriateness
of this assumption is the cumulative distribution of  30-day  averages
shown  in  Figure  8-4  and  8-5.  A straight line plot here on normal
probability paper indicates the validity of this model.

Under these conditions, the 30-day average  values  (X,,  X2,	Xm),
for  m  months  behave  approximately  as  random  data  from a normal
distribution with mean A and standard deviation  S".   Therefore,  the
probability  is  k  percent  that  a  monthly  X  will  not exceed the
performance standard P, where

     P = A + 2(S")

     The variability factor is

     VF = P/A = 1.0 + Z(SVA) and will be estimated by

     VF = 1.0 + Z(CV)

     Where

     A.   1 is a factor derived from the standard normal distribution.
          If one wishes a performance standard based upon expecting 95
          percent of monthly averages to be   within  guidelines,  then
          Z=1.64 should be used,

     B.   CV is the estimated coefficient of  variation of  the  30-day
          averages  and   is  computed  by  Sx/X, the ratio of standard
          error of sample means to overall or grand average of monthly
          averages.

Calculation of Variability Factors

A sample calculation of 30-day average  variability  factor  is  shown
below.   The  descriptive  statistical  data  is  for zinc  (mg/1) from
Appendix A.
     No
Min
Avq
Max
CV
     30     0.010      0.151      0.815      1.03

     VF  =  1 +  Z(CV)  =  1.0  +  1.64(1.03}  =  2.7

     P = A(VF)  =  (0.151)(2.7)  =  0.41
                                     118

-------
               NORMAL DISTRIBUTION
               (MODEL DENSITY OF LOGARITHMS OF POLLUTION VALUES)
                                        ln(P) = A' +  2.33(S')

                                           Y = ln(X) = Logarithm (mg/1)
         LOGNQPMAL DISTRIBUTION
          (MODEL DENSITY OF
         POLLUTION VALUES)
                                                   X(mg/l)

                                           _ P (Performance Standard)
                       '—A (Long Term Arithmetic Average)
         SAMPLE DISTRIBUTION OF N MEASUREMENTS
             (LONG TERM MONITORING DATA.)
                f        Max^*       X(mg/l)
                X* (Sample Average)
   Note:  (a)   S'  is estimated as (S1)2 = Clnd + CV2)J
              S2=
              X= ZX/N

Figure 8-3.  Statistical distribution for daily pollution measurements.
                                 119

-------
That is, the maximum 30-day average effluent limitation  derived  from
the   descriptive  statistics  above  would  be  0.41  mg/1  for  that
pollutant.

Given the previous descriptive statistics for a particular sample, one
obtains the performance standard P, by multiplying  the  mean  of  the
30-day  averages  in  the  data set by VF.  An appropriate statistical
interpretation is that, for the selected value of Z=1.64 corresponding
to the 95th percentile of a normal distribution, one estimates that 95
percent of  the 30-day average pollution level  measurements  will  not
exceed  P,  or in other words, the statistics predict up to 3 incidents
of non-compliance with the 30-day average per  pollutant  over  the  5
year  (60  month)  life  of a permit at a well-operated and maintained
treatment facility.  This is essentially the same number of  predicted
incidents  of  non-compliance  as  was  predicted  for  daily  maximum
limitations derived using the 99th percentile  confidence  level  (see
above).

In  developing  the  statistical  derivatives for monthly averages, in
many cases, a full 30 days of daily average  determinations  were  not
available.   In the above example, the monthly average is based on four
data  points  taken  during the month.  The standard deviation is then
derived from these "monthly" averages assuming a  normal  distribution
for  the  population  of averages.  Permits are usually written on the
basis of monthly averages obtained from fewer than 30 data points  per
month.   The  use  of  "monthly"  averages rather than 30-day averages
results in a higher  variability  and,  hence,  a  higher  performance
standard than would be attained using 30-day averages based on 30 data
points per month.

Figure 8-6 shows the relationship between the normal probability model
and frequency distribution of 30-day averages.
                                   120

-------
to
             3.00
        H
             2.00
        U
        W
        a
             1.00
              0.01      0.1  0.5 12    5  10   20  30 40 50 60 70  80   90  95  98  99      99,9
                                                     PERCENTAGE
                  Figure 8-4.   Cumulative distribution of 30-day averages of total cyanide  in treated
                                effluent from plant #782.

-------
M
10
        H
you









5OO







400





















_



_



—







~"
__
— .


._







—
~_



.
"T







.




- - •











	
-

-- - -










_






• -



—














-
-

—





—
-



_.
--




-


_

—





-



-











-















- : :













-
.. .

.. . . .
-
- - •




- - -

_ -












. . —


—





. . . .






--

• --

- —
—



.

—
—



. . —
. :





•-






-"












-












-












-
•


-









-







-
-









• •- 	
r - - - -
- - - • - •t
_ . . . .
•
• -









	


- - - - • 	

- • • • 	 : -
-- - 	





..... . . j


... . . i


- . . i
'".: ('-i
- • - - ,
....
. ,



J«
. . . _
	
• - -
...
. .
• - - -

. . .
• •



t
'



:.:.... K :::::::

'
i . - ...
M - ' • " -
! ' ! • • - . - -
. , .... . . .. _





	
, . - -



• • • -
. —



















-











-


-

















^

































1— • •
.
	 !.


/
l-~








—


—









—



— . .









firn



. . _





















. . _
















-


























-









—





























-









. _ —





























—








—




— .












—









~"


—





                0.01    0.1
12    5   10  20  30 40 50 60 70 80   90S 95   98 99
99.9
                                                  PERCENTACE
                Figure 8-5.    Cumulative distribution of 30-day averages of anmonia in treated
                              effluent from plant #782.

-------
                    NOEMAL DISTRIBUTION
     (MODEL DENSITY OF 30-CAY AVERAGE POLLUTION MEASUREMENTS)
                                       xCmg/1)

                                — P (Performance  Standard)

                    _ A  (long Term Average)
             SAMPLE DISTRIBUTION OF M MONTHLY AVERAGES
                   (LONG TERM MONITORING DATA)
         ^^ M
            Min
                     :
                           n
Max
X(mg/l)
 (Average of 30-Day Averages)
          Note:   (a)  P/A - 1+1.64 (CV)

                       CV = S/X
                     X=E  X/M
Figure 8-6.  Statistical distributions for 30-day average pollution measurements,
                                   123

-------

-------
                              SECTION 9

                  TREATMENT TECHNOLOGY APPLICATIONS
                     FOR TOXIC POLLUTANT REMOVAL
Selection of Pollutants to be Controlled

In  order  to  determine  which  toxic pollutants,  if any, may require
effluent limitations,  the pollutants observed in each subcategory were
evaluated  with   regard   to   their   treatability   and   potential
environmental   significance   on   the   basis   of   the  raw  waste
concentrations  and  mass  loadings   found   during   screening   and
verification.  In an attempt to prioritize the need for regulation the
toxic metals were divided into two groups:

     Group 1 - Those priority pollutants which appear at concentration
     levels that are readily treatable using available technology.

     Group  2  - Other treatable and/or potentially treatable priority
     pollutants observed in  the  subcategory.   These  include  toxic
     metals   which   exist   at   concentrations  below  the  minimum
     treatability limit and above the minimum  detection  level.   The
     Group  2  pollutants  would  be  controlled by the same treatment
     technology used to control the Group 1 pollutants.

Table 9-1 presents the significant toxic  pollutant  metals  found  in
each group.  In general, those metals occurring in the first group are
of  prime concern and require regulation, while those occurring in the
second group are of somewhat less concern  and  are  not  expected  to
require  regulation.   Metals in Group 2 are effectively controlled by
the technologies used to control the metals in Group 1, which are  the
two  or  three  dominant metals in the raw waste load and are directly
related to the particular product or process involved.

Application of_ Advance Level Treatment and Control Alternatives

General Design Objectives

Beginning  with  Section  11  of  this  document,  the  selection  and
application  of  toxic  pollutant treatment and control technology for
model plant systems  for  each  of  the  regulated  subcategories  are
described.   Several  levels  of  treatment  are  indicated.   Level  1
represents existing BPT treatment  systems  and  the  advanced  levels
(Level  2,  3,  etc.)  are  the  selected  technologies  for step-wise
improvements in toxic pollutant removal over that achieved by the  BPT
system.   Flow  diagrams  show  BPT components as a starting point for
advanced level treatment additions and incremental cost estimates.

For both existing and  new  sources,  the  advanced  level  technology
options  are  selected  as  candidates  for  BAT  with toxic pollutant
removal as the primary objective.  Although the advanced level systems
chosen also give improved performance over the Level 1  (BPT)  systems
                                  125

-------
   TABLE 9-1. PRIORITIZATION OF TOXIC POLLUTANTS POUND IN EACH SUBCATEGORY
SUBCATEGORY
                                   Group 1
Group
Chlorine-diaphragm cell
                                   Copper
                                   Lead
                                   Nickel
Antimony
Arsenic
Cadmium
Chromium
Mercury
Selenium
Thallium
Zinc
Chlorine-mercury cell
                                   Mercury
Antimony
Arsenic
Cadmium
Chromium
Copper
Lead
Nickel
Silver
Thallium
Zinc
Hydrofluoric Acid
                                   Nickel
                                   Zinc
Antimony
Arsenic
Cadmium
Chromium
Copper
Lead
Mercury
Selenium
Thallium
Titanium Dioxide •
 Chloride Process
                                   Chromium
Lead
Nickel
Zinc
Titanium Dioxide -
 Sulfate Process and
 Chloride Ilmenite Process
                                   Chromium
                                   Nickel
Antimony
Arsenic
Cadmium
Copper
Lead
Selenium
Thallium
Zinc
(2)
Group 1 - dominant raw waste pollutants select ed as control parameters
          for the effluent limitations or guidance.

Group 2 - secondary raw waste pollutants found less frequently and at
          lower concentrations.  These pollutants have not been selected
          as control parameters but are expected to receive adequate
          treatment as a result of controlling the Group 1 pollutants.
                                     126

-------
                             TABLE 9-1 Continued
SUBCATEGORY
Group  1
Group  2
Aluminum Fluoride
Chrome Pigments
Copper
Nickel
Chromium
Lead
Zinc
Arsenic
Cadmium
Chromium
Mercury
Zinc

Ant imony
Cadmium
Copper
Cyanide
Mercury
Nickel
Hydrogen Cyanide

Sodium Dichromate
Copper Sulfate
Cyanide

Chromium
Nickel
Copper
Nickel
Selenium
None

Copper
Selenium
Silver
Zinc

Antimony
Arsenic
Cadmium
Chromium
Lead
Zinc
Nickel Sulfate
Copper
Nickel
Antimony
Arsenic
Cadmium
Chromium
Lead
Mercury
Selenium
Thallium
Zinc
Sodium Bisulfite
Chromium
Zinc
Antimony
Cadmium
Copper
Lead
Mercury
Nickel
                                    127

-------
for  the  removal of conventional and nonconventional pollutants, this
is regarded as a secondary design objective.

Pretreatment Technology

Since  untreated  heavy  metal  ions  will  either  pass  through  the
treatment provided in a typical POTW, or will be precipitated with the
POTW  solid  residue,  pretreatment  of  wastes containing significant
amounts of heavy metals is necessary.  As  a  general  rule,  alkaline
precipitation,  followed  by  settling  and removal of the solids will
suffice.  In certain subcategories, such  as  the  chlorine  industry,
specific  treatment  will be required for highly critical constituents
{such as mercury  and  lead).   Normally  the  Level  1  and  2  model
treatment  processes  shown  in  the  following  subsections  will  be
appropriate  for  pretreatment  prior  to   discharge   to   a   POTW.
Pass-through  would  occur in the absence of pretreatment when BPT and
BAT treatment would reduce toxic metal  concentrations  by  a  greater
percent than is achieved by a POTW,

New Source Performance Standards

New  Source Performance Standards are at least equal to BAT.  In a few
cases where new plants have the  opportunity  to  design  systems  for
better  toxic  removal  performance without expensive retrofitting the
higher technology systems have been 'used as a basis.

Estimated Achievable Performance Characteristics  for  Advanced  Level
Applications

Advanced  level  control  and  treatment alternatives for reduction of
pollutant discharges and their applicability to each  subcategory  are
presented  in the sections dealing with individual products.  With few
exceptions, these alternatives were selected specifically for  removal
of priority pollutants and were designed for end-of-pipe treatment.

Treatment  technologies practiced outside the industry are recommended
when appropriate and, in most cases, apply  to,  the  removal  of  toxic
pollutant  metals.   The  estimated 30-day  average treatability  levels
(Section 8, Table 8-11), long-term data parameters, and the  screening
and  verification  results  are  all  utilized  in  the development of
estimated performance  characteristics  for  the  indicated  treatment
applications  in each subcategory.

Advanced Level Removal of BPT Pollutants


Performance  estimates for these systems, when possible, were based on
effluent  quality  achieved  at  plants  currently  practicing   these
technologies.   However,  in  most  cases,  the advanced levels are not
currently being practiced within the specific subcategory of  concern,
and   performance    information  from  other  appropriate   sources  is
necessarily utilized.
                                  128

-------
When established wastewater treatment practices, such as clarification
or filtration, form a part of  advanced  treatment  alternatives,  the
specified  achievable effuent quality has been based on concentrations
accepted as achievable through proper design and control.   The  prime
example of this is suspended solids reduction by filtration.

Advanced Level Removal of Toxic Pollutants

Performance  estimates  for  toxic  pollutants  were  also based, when
possible, on effluent quality achieved at plants currently  practicing
these  technologies.   However, in most subcategories, toxic pollutant
analyses are not conducted unless a specific  pollutant  is  regulated
and requires monitoring.  Where transfer of technology is applied as a
treatment  alternative,  performance  estimates  for  toxic  pollutant
removals  were  based  on  the  demonstrated  performances  in   other
industries  while incorporating allowances for specific differences in
process waste characteristics and operating conditions.  Statistically
derived long-term monitoring data parameters were described in Section
8 and are compiled in tabular form in Appendix A.  The  screening  and
verification  data are used to supplement the available long-term data
applied to each subcategory.  A judgment is made whether the screening
and verification data represent a well-performing system or one  which
is   not   performing   at   its   technological   potential.    For  a
well-performing system, the data are  regarded  as  representative  of
long-term  averages  and  are compared with the estimated treatability
ranges from Table 8-11, as well as the 30-day averages developed  from
the  long-term  data.   In  this manner, the performance estimates for
each  pollutant,  at  each  treatment  level   for   the   nonexcluded
subcategories,  are  developed and presented in tabular summaries.  By
starting  with  the  estimated  achievable  long-term  averages,   the
specific  variability  factor ratio derived for each pollutant is used
to estimate the maximum 30-day average and daily maximum values.

The model plant waste flow per unit of production  is  then  taken  to
calculate the estimated mass emission values of the 30-day average and
daily maximum limits for each pollutant to be controlled.

Pollution Control Parameters to be Regulated
                                        are identified as conventional
Conventional Pollutants

Wastewater  quality  parameters  which
pollutants include the following:

     pH
     Total Suspended Solids (TSS)
     Biochemical Oxygen Demand, 5-Day (BOD-5)
     Fecal Coliform
     Oil and Grease

Only the first two parameters  (pH and TSS) in  this  group  have  been
selected for regulation in the Inorganic Chemicals Manufacturing Point
Source  Category.   For direct dischargers, the pH range of 6 to 9 has
                                   129

-------
been established as the general control limitation and the permissible
frequency and duration of  excursions  beyond  this  range  is  to  be
specified  in  individual plant discharge permits.  The limitations of
TSS are specified for  both  BPCTA  and  NSPS-based  regulations,  the
former being largely a function of industry performance and the latter
stemming    from   treatability   estimates   with   the   appropriate
technologies.

Nonconventional Pollutants

The  wastewater  quality  parameters  classified  as   nonconventional
pollutants  include  the  nontoxic  metals  such  as  aluminum, boron,
barium, and iron  along  with  chemical  oxygen  demand  (COD),  total
residual  chlorine,  fluoride,  ammonia,  and nitrate, etc.  Of these,
only iron, COD, total residual chlorine,  fluoride,  and  ammonia  are
considered for regulation in the inorganic chemicals industry.  Due to
its  toxicity,  chlorine would be controlled in direct discharges, but
it is  excluded  from  control  in  pretreatment  regulations  because
influent  to  POTW's is often chlorinated.  A similar argument is made
for the control of amminia, that is,  POTW's  can  use  ammonia  as  a
source  of  essential  nutrients.  However, since many POTW's are only
capable of about 20 percent ammonia removal, both direct discharge and
pretreatment   regulations   would   specify   ammonia    limitations.
Similarly,  the  type  of  COD  found  in  inorganic chemical industry
discharges may not be amenable to biochemical oxidation in a POTW.  In
addition, compounds which contribute to the COD are likely  to  create
odor  and corrosion problems in sewer systems.  Therefore, its control
would also be retained in pretreatment regulations.  Fluoride  control
is  also  required  for  both  direct  and indirect discharges largely
because  the  most   practical   technology   for   fluoride   removal
(precipitation  as  calcium  fluoride)  must  be applied to relatively
concentrated  wastewater  sources.   This  treatment  method  achieves
removal  levels  which  at  best  are  still  unacceptable  for direct
municipal or agricultral water uses.  POTW's  are  not  effective  for
fluoride  removal  and  unless sufficient dilution occurs prior to the
reuse of the water, special techniques  (e.g., adsorption on  activated
alumina) would have to be applied for further fluoride removal.

Toxic Pollutants

The  toxic pollutants found at significant levels during screening and
verification are listed by subcategory  in Table 9-1.   Out  of  these,
toxic  pollutant control parameters were selected largely on the basis
of treatability.  Since several toxic pollutants may be controlled  by
a  common  treatment  technology, it is possible to select one or more
control parameters which will act as a surrogate for others exhibiting
the same treatability  characteristics.   Treatment  system  operating
conditions  would  normally  be  optimized  for  the  removal  of  the
specified control parameters which would be  monitored .on  a  regular
basis.   The  other  toxic  pollutants  would  be  monitored much less
frequently as a periodic  check  of  the  effectiveness  of  surrogate
control.
                                  130

-------
The  following  toxic  pollutants  have  been  designated  as  control
parameters in this point source category:

     Cadmium
     Chromium (Total)
     Copper
     Cyanide (amenable to chlorination)
     Lead
     Mercury
     Nickel
     Selenium
     Zinc

The specific control parameters  selected  for  each  subcategory  are
presented  in  the  tables  entitled  "Effluent  Limitations"  in  the
sections of this report dealing with the individual industries.   Some
general comments about them are given here.

The  most  common  technology  applied  in industry for the removal of
chromium from wastewaters involves a reduction step, whereby  Cr  (VI)
in  solution  is  converted  to the less toxic Cr  (III) form which can
then be removed by alkaline precipitation.   The  efficiency  of  this
treatment depends upon the presence of an excess reducing agent and pH
control  to  drive  the  reduction  step  to completion.  When treated
effluent samples are taken to  monitor  residual  Cr  (VI)  and  total
chromium  levels,  the  analytical  results for Cr  (VI) are subject to
several   factors   which   adversely   affect   the   accuracy   . and
reproducibility  of  the  diphenylcarbazide (DPC)  colorometric method.
The problem is not so much one of analytical interferences with the Cr
(VI) - DPC color development, but rather the actual changes in Cr (VI)
concentration that can take place during sampling,  sample preservation
•and storage, and analysis.  The major cause of  such  changes  is  the
presence  of  an  excess reducing agent in the treated effluent.  This
tends to give false low readings for Cr (VI) although  in  some  cases
the  opposite may occur as a result of sample preservation and storage
under acidic oxidizing conditions.

Thus, in  view  of  the  questionable  reliability  of  the  presently
accepted  Cr  (VI)  monitoring  procedure,  total  chromium, Cr (T), is
recommended as the control parameter  to  be  used  in  the  inorganic
chemicals  industry.  The adequacy of Cr (T) as a  control parameter is
predicated on its effectiveness as a surrogate for  Cr  (VI)  control.
Since  the  concentration  of  Cr  (T) represents  the summation of all
forms of chromium normally found in solution or  suspension  including
Cr  (VI),  the  final concentration of Cr (T) in a  treated effluent is
dependent on the effectiveness of both the reduction and the  alkaline
precipitation  steps.   In  this way, the use of Cr (T) as the control
parameter assures that adequate removal of Cr (VI)  is  being  achieved
as a direct consequence of the treatment technology required.
                                  131

-------

-------
                              SECTION 10

                COSTS OF TREATMENT AND CONTROL SYSTEMS
Introduction

Purpose of Cost Data

More  complex treatment methods and higher levels of pollutant removal
are reflected in increased  costs  of  equipment,  energy,  labor  and
chemicals.   At  some  point,  the  increasing costs of treatment will
outweigh the benefits of such treatment.  Therefore, it  is  important
that  for  each  subcategory  the  Agency  know  the base cost and the
incremental costs of each level of treatment which it might prescribe.
These "options" of internal costs, which  are  the  industry's  annual
costs  of  providing  the  necessary  waste  treatment, will result in
related increases in product costs, which are termed  external  costs.
Thus annual costs of waste treatment are expressed in terms of dollars
per unit of annual production of the principal product.

Plant  visits  revealed  very  few  treatment  plants serving a single
product manufacturing line, therefore, it was  not  feasible  to  seek
actual waste treatment facilities which could serve as real models for
estimating  purposes.   Accordingly,  the  cost  data  were taken from
similar construction projects by the contractor, and from unit process
equipment costs assembled from vendors and other  commercial  sources.
Based  on  the  level  of supporting details and scope definition, the
accuracy range of the cost estimates is expected to  be  minus  15  to
plus 25 percent.

Actual  costs  incurred  by individual plants may be more or less than
the presented model plant costs.  The major causes of variability ares

     Wastewater treatment combined with the treatment of other product
     effluents.

     Site  dependent  conditions,  as  reflected  in  piping  lengths,
     climate,  land  availability,  water  and  power  supply  and the
     location of the points of final discharge and solids disposal.

     Material (reagent) costs, due to variation  in  availability  and
     distance from the source.

     Flow rate of wastewater to be treated.

The  construction  costs  are  expressed  in  mid-1978  dollars.   The
investment costs and the annual costs given in  the  preamble  to  the
regulation  are  expressed in 1981 dollars, and were updated from 1978
dollars  using  the  Department  of  Commerce  Composite   Index   for
Construction Costs.
                                  133

-------
General Approach

Since  few  single  product  waste treatment plants were available for
detailed study, the costs presented in this section are based on model
plants which closely  resemble  the  types  and  capacities  of  waste
treatment  facilities  needed  for  each separate product subcategory.
The  model  plant  selections  are  based   on   review   of   Section
308-Questionnaire  responses,  plant  visits,  development  documents,
contacts with the industries to  verify  treatment  practices  and  to
obtain  data  on  size,  wastewater  flow,  and  solid  waste disposal
systems.  Thus, each model  is  synthesized  from  actual  data  as  a
typical  plant  in  its  subcategory  with  a level of waste treatment
equivalent  to  BPT.   Variations  in  treatment  plant  capacity  are
accounted for by selecting sets of models which represent the range of
existing  production  plant  capacities  in  the  subcategory;  large,
medium, and small.  Thus, the model plants are not set up as exemplary
plants, but as typical plants of adequate design which ' represent  the
range of plants and treatment facilities found in the subcategory.

Cost References and Rationale

Cost  information  contained in this report was obtained directly from
industry,  engineering  firms,   equipment   suppliers   and   current
experience  of  the contractor.  Costs are based on similar industrial
installations or engineering  estimates.   Cost  estimates  have  been
developed from either current costs for similar plants or from general
cost estimates.

Treatment  costs  are  based on model production plant characteristics
which determine the treatment processes selected for  each  operation.
Under   set   effluent  limitations,  treatment  costs  are  primarily
functions of the pollutant load (i.e., kg/kkg of product)  and  waste-
water  flow  rate  {i.e.,  cubic meters/day).  Available data indicate
that both pollutant loads and flow rates can vary significantly  among
plants manufacturing the same product.

Definition of Levels of Treatment and Control Cost Development

For  the  purpose of establishing the base level treatment costs, each
industry  is  assumed  to  be  practicing  Best  Practicable   Control
Technology   Currently   Available   (BPT),  for  the  EPA  pollutants
(conventional and nonconventional, as well as some of the toxic  metal
pollutants)  which are specified for each subcategory.  The investment
costs and annual costs of such BPT systems are shown in this report as
either the Base Level, Level 1, or BPT costs.  This level of treatment
may also provide incidental removal of additional toxic pollutants not
previously specified in the regulations.

The  advanced treatment level  (BAT) is aimed primarily at reduction of
toxic pollutants to levels considered  acceptable  for  July  1,  1984
performance,   utilizing   Best   Available   Technology  Economically
Achievable (BAT) at incremental investment  and  annual  costs  beyond
those  shown  for  Level  1  (BPT).   For  example,  for Level 2  (BAT)
                                   134

-------
treatment, the incremental cost as given  in  the  table  is  directly
added  to  Base  Level  1   or BPT cost to obtain the total cost of the
treatment system.  The wastewater  treatment  flow  diagrams  for  the
advanced  treatment  levels, as given in this report, also include the
flow diagram for Level 1  (BPT) treatment.


Treatment and Disposal Rationale Applied to Cost Development

The following assumptions  are employed in the cost development;

A.   Noncontact cooling water generally  is  excluded  from  treatment
     (and treatment costs) provided that no pollutants are introduced.

     Water treatment, cooling tower and boiler blowdown discharges are
     not considered  process  wastewater  unless  such  flows  contain
     significant amounts of pollutants.
B




C,

D-


E.

F,
     Sanitary sewage flow is excluded.

     The plants are assumed to operate 24-hours per day,
     year, except where otherwise noted.
                                                     350  days  a
G.
Manufacturing plants are assumed to be single product plants.

The inorganic chemical industry extensively uses in-plant control
techniques such as in-process  abatement  measures,  housekeeping
practices,  and  recycling  of  process  wastewaters  to  recover
valuable materials or use  these  materials  as  feed  for  other
by-products.   Segregation  of  uncontaminated  cooling and other
waters prior to treatment  and/or  disposal,  and  other  similar
measures  can contribute to waste load reduction.  All such costs
have not been included in the cost estimates.

Excluded  from  the  estimates  are  any  costs  associated  with
environmental permits, reports or hearings required by regulatory
agencies.
Expression of Costs

The  estimated  costs  for  all  treatment  systems  are  expressed in
mid-1978 dollars to construct appropriate facilities for  each  single
product  manufacturing subcategory at various production rates.  Total
costs are given for the BPT and NSPS systems while  Incremental  costs
are given for a BAT system.

Where  a  single  product  plant  produces  more than one waste stream
requiring treatment, the respective investment and  annual  costs  are
the combined costs of all treatment.

Total  annual  costs  per  metric  ton  of  product  are  shown in the
summaries for each product subcategory.
                                   135

-------
A.   Direct Investment Costs for Land and Facilities

Types of direct investment costs for waste  treatment  facilities  and
criteria  for  estimating  major  components  of  the model plants are
contained in the following subsections;

     1.    Site Development Costs - These include  clearing  the  site,
          all  earthwork  and site improvements.  The lagoon costs are
          based on the excavation and backfill required  to  construct
          multiple  rectangular  lagoons  with  common dikes to permit
          alternate dewatering for sludge  removal  by  the  clamshell
          method.  They are also based on reasonably level sites being
          available,  consisting of sandy loam with high clay content,
          and no large rocks or rock formations.  Lagoons are unlined,
          excepting where contents are highly acidic.  Where lining is
          required, hypalon or clay is used.

          Site  improvements  include  local  drainage,  fencing,  and
          roads.   Road costs are based on graded and graveled service
          roads only within the boundaries of the plant  and  not  for
          access.   Perimeter  fencing is supplied for the lagoons and
          for the sludge disposal site.
          Equipment
          equipment
          elsewhere
           Costs  -  This  is  the  installed  cost  of  all
            except   the   monitoring  system  (  considered
              Depending  upon  the  method   of   treatment,
equipment for wastewater treatment consists of a combination
of  items  such  as  pumps, aerators, chemical feed systems,
agitators,  flocculant  feed  systems,  tanks,   clarifiers,
thickeners,  filters,  etc.   Costs  for  these  items  were
obtained from vendors' verbal quotations and were  based  on
contractors1  experience  with procurement of similar items.
Enclosures are provided for critical equipment and controls.

Chemical storage, feeders  and  feedback  equipment  include
such  items  as probes, instruments, controls, transmitters,
valves, dust filters and accessories *  Bulk chemical storage
bins are designed to hold a standard bulk truck  load,  plus
five  day's  needs, between ordering and delivery.  Critical
pumps are furnished in duplicate  and  when  clarifiers  are
used, the flow is split between two units, permitting one to
be  bypassed  for  repairs.  Single units are used for small
flows, batch treatment and intermittent service.

Added  to  the  cost  of  the  equipment  itself,   is   the
corresponding  installation  labor,  as well as the material
and labor costs  for  concrete,  structural  steel,  piping,
instrumentation,  and  electrical  work.   The  labor  costs
include all pro-ratable elements of "indirect" costs such as
fringe benefits, payroll insurance and  taxes,
equipment,  temporary  construction facilities
etc.  The hours and unit costs for the labor
                                                           construction
                                                         ,  field  staff,
                                                         are   based   on
                                     136

-------
          Los  Angeles
          craft labor.
or  Gulf  Coast  type productivity using union
          In some  subcategories,  a  portion  of  the  wastewater  is
          returned to the process from an intermediate treatment step.
          In  such  cases,  the estimated investment cost of the reuse
          pumps are included as part of the equipment cost.   However,
          the  return  piping,  accessories, operating and maintenance
          costs are considered as water supply costs.

     3.   Monitoring Equipment - In this report, it  is  assumed  that
          flow  and  pH  monitoring equipment will be installed at the
          treated effluent discharge point.  It  will  consist  of  an
          indicating  sensor  and recorder, alarms and controls and an
          automatic sampler.

     4.   Land  -  Land  availability  and  cost  of  land  can   vary
          significantly,  depending upon geographical location, degree
          of urbanization and  the  nature  of  adjacent  development.
          Land for waste treatment, and in some cases for inert solids
          disposal,  is  assumed  to be contiguous with the production
          plant site and reasonably convenient to a waterway which can
          receive permitted discharges  of  wastewater.   Where  inert
          solids  are  retained  at  the  plant  site,  enough land is
          included in the base level model plant  investment  cost  to
          accept  residual solids for a normal operating period of ten
          years at the same production rate for  which  the  plant  is
          sized.

          For  the purpose of this report, land for lagoons, treatment
          facilities, and on-site residual waste disposal is valued at
          $30,000 per hectare ($12,000 per acre).

B. Investment Costs for Supporting Services

     1.   Contractor's  Overhead  and  Profit   -   The   construction
          contractor's  fixed  or  "overhead"  expenses and profit are
          estimated as fifteen percent of the installed plant cost.

     2.   Engineering  -  This  includes  the  design  and  inspection
          services  to  bring a project from a concept to an operating
          system.  Such services broadly include laboratory and  pilot
          plant  work  to establish design parameters, site surveys to
          fix elevations and plant layout, foundation and ground water
          investigations,-and operating instructions; in  addition  to
          design   plans,   specifications   and   inspection   during
          construction.  These costs, which vary with job  conditions,
          are  often  estimated  as  percentages of construction cost,
          with typical ranges as follows:
          Preliminary survey and construction
          surveying
                               1  to 2'
                                     137

-------
     3.
          Soils and groundwater investigation

          Laboratory and pilot process work

          Engineering design and specifications

          Inspection and engineering support
          during construction

          Operation and maintenance manual
                                             1  to 2%

                                             2 to 4%

                                             7 to 12%


                                             2 to 3%

                                             1  to 2%
From these totals of 14 percent to 25 percent, a midvalue of
20 percent of in-place construction (installed equipment and
construction) costs has been used in this study to represent
the engineering and design costs applied to model plant cost
estimates.   These  costs  include,  in  addition   to   the
professional  service  hours, the costs for expenses such as
telephone,  reproductions,  computer  services,  and  travel
fees.

Contingency _ This is an allowance of 10 percent applied  to
the  estimated  total investment cost, excluding land, based
on the status of  engineering,  design  and  specifications,
quality   of   prices  used,  and  the  anticipated  jobsite
conditions.   This  covers  design  development,  {but   not
scope),  errors and omissions, impact of late deliveries and
unusally adverse weather  conditions,   variations  in  labor
productivity   and   other  unforeseen  difficulties  during
construction.
C.   Operation and Maintenance Costs

     Annual  operation  and  maintenance  costs  are   described   and
     calculated as follows:

     1.    Labor and Supervision Costs - Plant operations  are  assumed
          to  be  conducted  24-hours  per day 350 days per year, with
          attendance for only part of each  working  day.   For  batch
          wastewater  treatment  systems, adjustments are made for the
          number of working days in a year.  Personnel costs are based
          on an hourly rate of $20.00.  This includes fringe  benefits
          and   an   allocated   portion   of  costs  for  management,
          administration, and supervision.

          Personnel are assigned for specific activities  as  required
          by  the  complexity of the system, usually 4 to 12 hours per
          day.

     2.    Engergy Costs - Energy (electricity) costs are based on  the
          cost  of  $306.00  per horsepower operating 24 hours per day
          and 350 days per year.   For  batch  processes,  appropriate
                                    138

-------
     adjustments  are  made to suit the production schedule.
     cost per horsepower year is computed as follows:
                                                         The
     Cy
          1.1 (0.7457 HP X Hr X Ckw)/(E X P)
              1)
Where:
     Cy

     HP


     E

     P
          Cost per year

          Total horsepower rating of motor (1 hp
          kw)

          Efficiency factor (0.9)

          Power factor (1.00)
             0.74557
     Hr   =    Annual operating hours (350 X 24 = 8400)

     Ckw  =    Cost per kilowatt-hour of electricity ($0.040)
3.
The 1.1  factor in  equation  (1)  represents  allowance  for
incidental energy used such as lighting, etc.  It is assumed
that  no  other  forms  of  energy  are  used  in  the waste
treatment system.

Chemicals - Prices for  the  chemicals  were  obtained  from
vendors  and the Chemical Marketing Reporter.  Unit costs of
common chemicals delivered to the plant site  are  based  on
commercial  grade  of  the  strengths  of  active ingredient
percentages as follows:
          Hydrated Lime (Calcium
          Hydroxide)
          Quicklime

          Ground limestone

          Soda Ash (58% Bulk)

          Caustic Soda (58% NaOH)

          Sodium Sulfide (60-62%)

          Sulfuric Acid

          Hydrochloric Acid (32%)
                              Bulk
                              Bag

                              Bulk
$ 80/metric ton
$ 85/metric ton

$ 70/metric ton

$13.20/metric ton

$ 85/metric ton

$200/metric ton

$435/metric ton

$ 75/metric ton

$ 70/metric ton
                               139

-------
     Aluminum Sulfate (56% Alumina)

     Flocculant (Polymer)

     Sulfur Dioxide (Ton Containers

     Chlorine (Ton Containers)

     Sodium Bisulfide (72-74%)

     Ferrous Sulfate

     Diatomaceous Earth

     Activated Carbon
$250/metric ton

$2.00/kg

$335/metric ton

$220/metric ton

$385/metric ton

$ 70/metric ton

$0.30/kg

$2.00/kg
Maintenance - The annual cost of maintenance is estimated as
10 percent of the investment cost, excluding land.

Taxes and insurance - An annual provision of  three  percent
of the total investment cost has been included for taxes and
insurance.

Residual waste disposal - Sludge  disposal  costs  can  vary
widely.  Chief cost determinants include the amount and type
of  waste,  and  the  choice  of  either on-site disposal or
contract hauling which depends on the size .of  the  disposal
operation  and  transport  distances.   Off-site hauling and
disposal costs are taken as $13.00 per cubic  meter  ($10.00
per cubic yard) for bulk hauling, with appropriate increases
for  small  quantities  in  steel  containers.   For on-site
disposal from lagoons, a clamshell at $600.00 and front  end
loader at $300.00 per disposal day are used.  For very large
sludge  quantities, lower unit costs have been assumed.  The
computed sludge quantities are  spread  on  land  valued  at
$12,000 per acre.

Monitoring,  analysis,  and   reporting   -   The   manpower
requirements  covered  by  the  annual labor and supervision
costs include those activities associated with the operation
and maintenance of monitoring  instruments,  recorders,  and
automatic  samplers  as  well as the taking of periodic grab
samples.   Additional  costs   for   analytical   laboratory
services  have  been estimated for each subcategory assuming
that sampling takes place three times a week at the point of
discharge  and  that  an  analytical  cost  of  $20.00   per
constituent  is  incurred.   Approximately  10 percent of the
total analytical cost has been added for quality control and
water supply samples.  Unless otherwise  stated,  continuous
compliance  monitoring  at  the  BPT  level are based on the
determination of four constituents.  At the  advanced   (BAT)
levels, the determination of six constituents is assumed.  A
reporting  cost  of  $1,500  per  year is added for clerical
                           140

-------
          support.  Monitoring costs for periodic batch treatments are
          reduced in proportion to the number of days  per  year  when
          discharges occur.

D.   Amortization

     Annual depreciation and capital costs are computed as follows:

          CA = B (r(1 + r)n)/((l  + r)n-»)         (2)
     Where:
          CA
          B
          n
Annual Cost

Initial amount invested excluding cost of land

Annual interest rate (assumed 10 %)

Useful life in years
     The multiplier for B in equation (2) is often referred to as  the
     capital  recovery  factor,  and is 0.1627 for the assumed overall
     useful life of  10  years.   No  residual  or  salvage  value  is
     assumed.

     Items Not Included in Cost Estimates

     Although  specific plants may encounter extremes of climate,  flood
     hazard and availability of water,  the costs of model plants  have
     been  estimated  for  average conditions of temperature,  drainage
     and natural resources.   It is -assumed  that  any  necessary  site
     drainage,    access    roads,    water   development,   security,.
     environmental studies and permit costs are  already  included  in
     production facilities costs.

     Therefore,  the model costs are only for facilities, supplies and
     services  directly  related  to  the  treatment  and  disposal  of
     waterborne  wastes,  including  land  needed  for  treatment  and
     on-site  sludge  disposal.   Air  pollution   control   equipment
     required   by  the  Clean  Air  Act  is not included.  It was also
     assumed that all required utilities are provided at the  edge  of
     the plant site and the existing plant's capacities are capable of
     supplying  the  requirements.  RCRA costs have not been included.
     RCRA costs are considered in  the  Economic  Impact  Analysis  o£
     Pollution  Control  Technologies  for  Segments  of. the Inorganic
     Chemicals Manufacturing Industry,  EPA  440/2-81-023,  which  were
     developed  in part from information in "Contractor Report on RCRA
     ISS   Compliance   Costs   for   Selected   Inorganic   Chemicals
     Industries."
                                    141

-------
     Dust  collectors  normally  associated  with  package  treatment,
     chemical transfer and feeding systems are included.   Raw  wastes
     from various sources are assumed to be delivered to the treatment
     facility  at  sufficient  head  to fill the influent equalization
     basin, and final effluent is discharged  by  gravity.   Costs  of
     pumps,  pipe  lines, etc., necessary to deliver raw wastewater to
     the treatment plant or to deliver the  treated  effluent  to  the
     point of discharge are not included in the cost estimates.

     Since  the  treatment models are designed to serve single product
     manufacturing plants, no emergency  holding  basins  or  internal
     bypasses  are  provided.   Any such necessary facilities are more
     appropriately furnished as part of  a  combined  waste  treatment
     system serving several product lines.

Cost Estimates For Each Subcateqory

Estimated  costs for the wastewater treatment plants for the different
annual productions and at various levels of treatment  are  calculated
in  terms  of  total  annual  costs.   The  total  annual  cost is the
summation of the annual amortization of the investment costs  and  the
annual operation and maintenance costs.

     The types of costs shown for each model plant are:

     1.    Investment
     2.    Annual operation and maintenance
     3.    Annual amortization of investment costs (excluding land)

The total annual costs per metric ton of product have been calculated.

For the purpose of the cost estimate, the first level (BPT or NSPS) is
expressed as the total cost of the treatment system.  The other  level
(BAT) represents the incremental cost above the base cost.  The actual
additional  costs  a  plant  would incur in implementing the described
treatment processes depend on current treatment practices, and to some
extent,  on the availability of land.

In some cases, land for economical on-site sludge disposal for  a  ten
year  period  has  been  provided in the BPT model plant costs.  Since
land cost is not amortized, its value appears in the investment  cost,
but not in the total annual costs.
For  the purpose of cost estimating, a set of generally representative
model plant specifications are given for each nonexcluded  subcategory
starting   with  the  Chlor-Alkali  industry  in  Section  11
specifications together with the basic assumptions on cost
in  this section, form the basis of the cost estimates for alternative
treatment systems.  These cost estimates are presented  in  a  tabular
format  in the cost development portion of each applicable subcategory
section.  In order to take into account more fully the wide  range  of
     These
estimating
                                   142

-------
plant  specific  variables,  additional cost elements which may add to
the baseline costs are then considered on a case-by-case basis.
                                   143

-------

-------
                              SECTION 11

                        CHLOR-ALKALI INDUSTRY
Industry Profile

General Description

Chlorine and its co-product caustic soda (alkali) are  used  in  large
quantities  in  the  production  of  plastics,  organic  and inorganic
chemicals, in the pulp and paper industry,   in  water  and  wastewater
treatment and in a number of other industries.

The  production  rate  in the United States is approximately 9 million
metric tons (10 million short tons) of chlorine per year and  over  95
percent  of  that  production  is  by  the electrolysis of a sodium or
potassium chloride solution via one of two  major  processes,  mercury
cell and diaphragm cell.

Other processes for chlorine production such as the recently developed
membrane  process  are  not  addressed  here  because only pilot-scale
production exists or  no  data  are  available  from  fully  operating
facilities.

Subcategorization

The  factor  chosen for the primary subcategorization of the inorganic
chemicals point source category was dominant product (see Section  4).
Other  factors considered for subcategorization include: raw materials
used, manufacturing process employed, geographical location, size  and
age  of  equipment and facility involved, non-water-quality aspects of
waste characteristics, water pollution control  technology,  treatment
costs, energy requirements and solid waste disposal.  The chlor-alkali
subcategory was further subdivided on the basis of differences in cell
design and in the quantity and quality of wastewater generated.

Mercury and diaphragm cells are the two distinct types of electrolytic
cells  that  are  used in the production of chlorine and caustic soda.
Major process differences between  mercury  cell  and  diaphragm  cell
plants  produce  corresponding differences in the volume and nature of
wastewater generated.  A  principal  difference  is  the  presence  of
mercury  as  a  contaminant  in  the wastewaters from the mercury cell
process and asbestos in the diaphragm  cell  plant  wastes.   The  TSS
discharges  from  diaphragm cell plants are generally larger than from
mercury cell  plants,  due  to  the  higher  volumes  of  contact  and
noncontact  water  used.    Also,  in diaphragm cells a large amount of
water is used and an appreciable quantity of wastewater is produced in
the caustic evaporation  process.   Such  water  is  not  produced  in
mercury  cell  plants.   The quantity of wastewater generated from the
diaphragm cell plants may be more than four times that of the  mercury
                                   145

-------
cell  plants  for the same chlorine production capacity.  Based on the
quantity   and   characteristics   of    the    wastewater,    further
subcategorization is justified.


Mercury Cell Process Industry Profile

General Description

Approximately  30  percent  of  the  U.S. production of chlorine is by
mercury cell plants.  In 1978  of  27  known  plants,  308  data  were
available  for  15.   Table  11-1  presents  a  summary profile of the
subcategory.  Table 11-2 presents the status of discharge  regulations
for  mercury  cell  chlorine  plants  prior  to  promulgation of these
regulations.  Control of pH in the 6.0 to 9.0 range was also  included
in those regulations.

General Process Description

A.   Brine System

     The sodium chloride solution (brine or salt dissolved  in  water)
     is   treated  with  sodium  carbonate  and  sodium  hydroxide  to
     precipitate impurities such  as  calcium,  magnesium,  and  iron.
     Precipitated  hydroxides  and carbonates are then settled usually
     in a clarifier and the underflow, known as brine mud, is sent  to
     a  lagoon  or  filtered.   Brine  muds  from  mercury cell plants
     usually contain small amounts of mercury because the spent  brine
     from  the  cells is recycled.  Consequently brine mud filtrate is
     recycled or treated before discharge and solids are  disposed  of
     in secure landfills.

     Before it is sent to the cells, the treated brine is filtered and
     then  pH  adjusted.   Spent  or  depleted brine from the cells is
     acidified and dechlorinated using  vacuum  and/or  air  stripping
     before being saturated with salt and recycled.

B.   Mercury Cell Process

     The mercury cell, in  general,  consists  of  two  sections:  the
     electrolyzer  and the decomposer or denuder.  The electrolyzer is
     an elongated steel trough that  is  inclined  slightly  from  the
     horizontal.   Mercury flows in a thin layer at the bottom forming
     the cathode of the cell, and the brine flows concurrently on  top
     of  the  mercury.   Parallel  graphite  or metal anode plates are
     suspended from the cover of the cell.  Electric  current  flowing
     through the cell decomposes the brine, liberating chlorine at the
     anode and sodium metal at the cathode.  The metallic sodium forms
     an amalgam with mercury,

     NaCKaq) + Hg - Cl, + 2 Na(Hg)
                                   146

-------
         TABLE 11-1.  SUBCA3EGORY PROFILE .DATA .SUMMARY
SUBCfllEGORY
CHLORINE MERCURY
Ibtal subcategory capacity rate
Ototal subcategory production rate
Number of plants in this subcategory
308 Data on file for
    With total capacity of
    With total production of
    Representing capacity
    Representing production
    Plant production range:
            Minimum
            Maximum
    Average production
    Median production
    Average capacity utilization
    Plant age range:
            Minimum
            Maximum
    Waste water flow range:
            Minimum
            Maximum
    Volume per unit product:
            Minimum
            Maximum
                         3,550,000 kkg/year
                         2,750,000 kkg/year
                               27
                               15
                         1,280,000 kkg/year
                         1,090,000 kkg/year
                               36 percent
                               40 percent

                            19,100 kkg/year
                           198,000 kkg/year
                            77,900 kkg/year
                            70,400 kkg/year
                               75 percent

                                2 years
                               26 years

                                4 cubic meters/day
                            2,100 cubic meters/day

                               < 1 cubic meters/kkg
                               11 cubic meters/kkg
Sources of data are Stanford Research Institute, Directory of Chemical
producers, U.S.A., 1977, U.S. Department of Ccmnerce, Current Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Eooncmic Assessment of Effluent Limitations in the
Inorganic Chemical Industry,"June, 1978, and "Economic Analysis  of  Proposed
Revised Effluent Guidelines and Standards for the Inorganic Chemicals Industry,"
March, 1980.
                                   147

-------
      TABLE 11-2 STATUS OF REGULATIONS - EFFLUENT LIMITATION GUIDELINES
SUBCATEGORY

SUBPART
CHLORINE MERCURY CELL

F (40 CFR 415.60, 3/12/74)
                                    STANDARDS
Product  Parameters
Process
      BPCTCA
Max.t1)   Avg.
(kg/kkg)  (kg/kkg)
                                    (2)
                         BATEA               NSPS
                     Max.     Avg.      Max*     Avg.
                     (kg/kkq) (kg/kkg)  (kg/kkq) (kg/kkg)
Mercury
Cell
Process  TSS
         Hg
 0.64
           0.32
No discharge
 of pwwpI3)
                       No discharge
 0.00028   0.00014 __ of pwwp
0.64     0.32
                                         0.00014  0.00007
    Section 415.63 was remanded and reserved (41 FR 51601, November 23, 1976).
(1) Max. = Maximum of any one day.
(2) Avg. = Average of daily values for thirty consecutive days shall not exceed.
(3) pwwp = Process wastewater pollutant*
                                      148

-------

     The amalgam from the elctrolyzer flows to a denuder and the spent
     brine  is  recycled  to  the  brine purification process.  In the
     denuder,  the amalgam becomes an anode to a  short-circuited  iron
     or  graphite  cathode.   Deionized  water is added to the denuder
     which reacts with the amalgam to form hydrogen and caustic  soda.
     In   modern  mercury  cells,  the  denuder  or  decomposer  is  a
     horizontally or vertically laid graphite-packed bed.   The  water
     and  the amalgam flow countercurrently.  Mercury is then returned
     to the electrolyzer.

C.    Product Purification

     Chlorine from the cell  is  cooled  to  remove  water  and  other
     impurities.    The  condensate  is  usually  steam  stripped  for
     chlorine recovery and returned to the brine system or discharged.
     After cooling,  chlorine gas is dried further  by  scrubbing  with
     sulfuric  acid.   The  diluted  acid is then usually regenerated,
     sold or used for nonbrine system pH control.  When  chlorine  gas
     is  compressed  and  liquified,  it  leaves behind noncondensible
     gases known as tail or  sniff  gas.   The  tail  gas  is  usually
     scrubbed with caustic or lime, generating a hypochlorite solution
     which  is  then decomposed, used on-site, sold or discharged with
     or without treatment.

     The sodium hydroxide or caustic product formed at the denuder has
     a concentration of 50  percent  NaOH.   Some  of  the  impurities
     present  in the caustic can be removed or reduced by the addition
     of certain chemicals, and the caustic is then filtered.  In  most
     cases  it  is  sent  to  storage  or  is  evaporated  if  a  more
     concentrated product is required.

     Hydrogen gas is cooled by refrigeration to remove water vapor and
     mercury,  and can  be  treated  further  by  molecular  sieves  or
     carbon.   Condensate  from hydrogen cooling is then discharged or
     recycled to the denuder after mercury recovery.

     Figure 11-1 presents a general process flow diagram  of  chlorine
     production by mercury cell.

Water Use And Wastewater Source Characteristics

Water Use

Water  is  used at mercury cell plants for noncontact cooling, tailgas
scrubbing, cell washing, equipment maintenance, floor washings and  in
the  decomposition of sodium-mercury amalgam in the denuder to produce
sodium hydroxide.  Because most brine systems at mercury  cell  plants
are  closed  systems,  water  use in the brine system is minimal.  The
total water usage at plants was found to range from 7.6 to  204  cubic
meters  per  metric  ton  (1800 to 49,000 gallons per short ton), with
noncontact cooling water,  which  is  not  covered  by  this  effluent
guideline, comprising approximately 70 percent of the total.
                                   149

-------
H
O
SPKHT BKINt
s »
CHLOf"*

1
BftL * ^S™* »™™ I" AMAU*H 	 •» .^SSL, K- DWIKUUW.
CE"- . 1 (OECOBPOSBWJ COOLER

8YSTE* COO^IHE^HATER HVOJKMtl


' ^-^-™».i.S" (AWP J° ATHOSPHKRE
Dt SODIUM ItOMCOWfACT F1LTKB) *oB. USE
>K SOUITJESM f?fttM.IHG ^

t
COtUUER COHDEHSfttE
1 ff fR^BiinnTiH". u*^Vp*i H*70 Jffr51^
faffetUK UA¥fiD 4n^M
4
Tn tikfiTB - BACKMASU H£CVCLE
AND HECKCLE ** '*
HOMCOWACT
(COWTACT*)
COOHWG HATER c-JIPHBIP
auuuniv. BBCYCLEO ^f- 	
1 t ACID — MEIVHiBy — ^^ —
i f
S
2 i

E 11
R f f
• ' REFRIGERATIOH
V HEAK SULFURIC SYSTEM
sn. «™tllM 1 I BKCVCLB COUL1HU
FIITEH uruHSi"^1 — |P*C«*fiH*e ) 	 •» TOMER
HCKMAS0 90LUTIOH ^
BACKhUSU A 1
piLven CAUSTIC ILIMB) | f
{ LAHOFILl. |
r i s
— 1 1 	 SOLIDS 	 hrTO HASTE B
* — HKPOCBU)RITB_-.
" SOLUTION
LIQUID _,, _,._- TO USE, SALES,
CBLORIMB ~~^{at IKE °" W*51*
PRODUCT ""

f HONCOUTACT
TO UASTgOR REUSE COOLING
HATER
      USED AT SOME PLAMTS ONLY
Figiare 11-1.   General process  diagram for production of chlorine/caustic by
               mercury cells.

-------
Waste Sources
     The  following  waste  sources  are  or  can be contaminated with
     mercury and would therefore require treatment if discharged.

A.   Brine Mud

     This is the waste  produced  during  the  purification  of  brine
     before  it  is  introduced  into  the cell for electrolysis.  The
     metals  commonly  removed  during  purification  are   magnesium,
     calcium,   iron   and   other  trace  metals  such  as  titanium,
     molybdenum, chromium, vanadium and tungsten.   Calcium  and  iron
     are removed as hydroxides. _Bjrine mud is the. majpr_pprtion of the
     waste  solids  produced  from the process.  The solids content of
     'the stream varies from 2 to 20 percent and the volume varies from
     0.04 to 1.5 cubic meters per metric  ton  of  chlorine  produced.
     The  waste  is either sent to a pond for settling or is filtered.
     The overflow from the pond or the filtrate  is  recycled  to  the
     process  as  makeup  water  for  the  brine.  In the mercury cell
     process, only  16  percent  of  the  brine  is  recycled  to  the
     purification  unit  after dechlorination.  This recycled brine is
     contaminated with mercury so the  resulting  brine  mud  contains
     small amounts of mercury.

B.   Cell Room Wastes

     The major components of this stream include leaks,  spills,  area
     washdown  and  cell wash waters.  The amount varies from plant to
     plant  and  depends  largely  on  housekeeping  practices.   Data
     indicate  a range of from 0.01 to 1.5 cubic meters per metric ton
     of chlorine produced.  Cell  room  waste  constitutes  the  major
     stream  requiring treatment,..because of the. high levels of^mercury
     present in these wastes.  If graphite  anodes  are  used  in  the
     cells,  the  wastes  may also contain lead  (used as an electrical
     contact at the anode) and  chlorinated  organics.   However  most
     mercury cell plants have converted to metal anodes.

C.   Chlorine Condensate

     Condensation from the cell gas is contaminated with chlorine.  At
     some plants, the condensates are recycled to  the  process  after
     chlorine recovery.  Both contact and noncontact water is used for
     chlorine  cooling  and  for  removal  of water vapor.  Because of
     this, the amount and type of  wastewater  varies  from  plant  to
     plant.   Data from one plant indicates a waste condensate flow of
     approximately  0.01  cubic  meter  per  metric  ton  of  chlorine
     produced.

D.   Spent Sulfuric Acid

     Concentrated sulfuric acid is used in the  dryer  to  remove  the
     residual  water  from  the  chlorine gas after the first stage of
     cooling.  In most cases,  the  acid  is  used  until  a  constant
                                  151

-------
     concentration  of  50-70 percent is reached.   The spend acids can
     be regenerated for reuse, used  for  pH  control  in  a  nonbrine
     treatment system, or sold.
                                              chlorine  gas  from  the
Tail Gas Scrubber Liquid

The tail gas containing the uncondensed 	^
liquefaction  stage,  alcmg  wTttr~~gome" air
scrubbed with sodium/calcium  hydroxide  to  form  sodium/calcium
hypochlorite   solution.    When  the  equipment  is  purged  for
maintenance, the tail gas is also absorbed in calcium  or  sodium
hydroxide,  producting  the  corresponding hypochlorite solution.
The hypochlorite can be used in another process  on  site,  sold,
discharged   to  treatment  or  decomposed  before  discharge  or
treatment.  The amount of tail gas  scrubber  water  varies  from
0.04 to 0.58 cubic meter per metric ton of chlorine.
H
Caustic Filter Washdown

The 50 percent caustic produced at the  denuder  is  filtered  to
remove  salt  and  other  impurities.  The filters are backwashed
periodically as needed, and the backwash  can  be  discharged  to
treatment  or  filtered  with  the filtrate recycled to the brine
system and the solids sent  for  disposal  or  mercury  recovery,
Wastewater volume from caustic filter backwashing is variable and
no flow data are available.

Hydrogen Condensate

Hydrogen produced at the denuder is cooled to remove mercury  and
water  carried  over in the gas. The condensate  is either sent to
treatment facilities or to merjcury recovery after which it can be
returned to the denuder.  Data on the volume of  this waste stream
are not available.

Summary of Wastewater Flow

Summing the  flow  ranges  presented  above  for  specific  waste
sources results in a maximum mercury-contamined waste flow of 2.1
cubic  meters  per  metric ton (mVkkg} for plants where specific
stream data were available.  This  does  not  include  brine  mud
flows  which  are  reused instead of discharged, and therefore do
not affect total flow.

Data available on total discharges at 13 mercury cell plants  are
presented  in Table 11-3.  The average discharge volume indicated
is also 2.1 mVkkg, although flows  as  high  as  6.3  mVkkg  do
exist.
                                    152

-------
Description of Specific Plants

The  following descriptions of specific plants  includes  those  that  were
sampled during the screening and verification  program.  The discussion
primarily  covers plant practices in wastewater control  and treatment.

Screening  Program

Plant  #299 was visited in the screening  and verification phase  of the
program.   The  mercury-contaminated  waste  streams   include   outlet
end-box  wash  water,  spills  and  cleanup water, brine mud  saturator
sludge, and pump seals wastewater.  The combined wastewater is sent to
JLSurge pond.  The effluent from the surge pond is mixed  with   sodium
%8jS$jlM&lfi and sent to a settling pond.   The overflow  from the pond is4
pfi adjusted, filtered  (in a filter press) and  passed through  activated
carbon towers before discharge.  In the   sampling  program  wastewater
influent   to  the  surge  pond and the overflow from the settling  pond
were sampled.  Figure  11-2 gives the general process diagram  and shows
all  the waste streams  sampled.  Table 11-4  presents   major   pollutant
concentrations and loads for the sampled  streams.

Verification

Four  more   plants   (#747,   #167,   #106    and   #317)    producing
chlorine/caustic by mercury cells were  visited and   sampled in   the
verification  program.   Table  11-5 presents  pollutant concentrations
for  the sampled streams and loads for TSS and  mercury.

At Plant #747, the brine dechlorination system has been converted  from
barometric condensers  to  a  steam  ejector  system.    The  conversion
resulted   in   increased   chlorine   recovery  and   reduced contact
wastewater.  By providing settling and  secondary  filter  facilities,
the  brine filter backwash has been eliminated.  The tail gas scrubber
liquid  (hypochlorite  solution)  is  offered   for  sale  and if   not
marketed,  is treated  for removal of chlorine  and discharged.  Mercury
bearing  wastewaters   are  treated  with  sodium  sulfide  (Na2S)   and
filtered.   Solids  are retorted for mercury recovery  and the filtrate
is mixed with the other process wastewaters and the pH  adjusted  before
discharge.  A flow diagram of the manufacturing process, including the
wastewater treatment facility, is given in Figure 11-3.

At Plant #167, the wastewater streams, consisting of filter   backwash,
cell  room wash, rain water runoff, and leaks  and spills, are combined
and  treated for mercury removal.  The  water   is  sent  to  a holding
lagoon  and the overflow is reduced by reaction with ferrous  chloride,
which precipitates  mercury.   The  reacted  Solution   is  sent  to a
c1arifier  and  the  underflow  from the  clarifler is  disposed of  in a
landfill.  The overflow is filtered and the filtrate is passed through
activated  carbon and an ion exchange column prior to   discharge  to a
lagoon.    The  effluent from the lagoon is pH  adjusted  and discharged.
Figure 11-4 shows the  simplified process  flow  diagram  for Plant  #167,
including  the sampling locations.
                                   153

-------
         TABLE 11-3.  SUMMARY OF WASTE WATER PLOT DATA FOR CHLORINE
                            MERCURY rar.T. PLANTS
SUBCATEQORY
CHLORINE MERCURY CKT.L
         Plant
         Nurber
            Waste Water Flow
           (m Akg Chlorine)
          907
          299
          343
          106
          131
          589
          898
          741
          553
          769
  Average of 13 plants
                  0.36
                  1.6
                  1.6
                  0.67
                  1,7
                  5.8
                  0.98
                  0,51
                  1.0
                  6,3
                  2,1
                                    154

-------
Ul
                                                         SULFIPE
                                                         AMITItM
                                                                            HI
                                                                      (121
e_ fljncE | seauaa
* KW *- FOND


(Ml £ } H |
HUMEHD
e_|_-J "W« 	 ^ «CTWMH> 	 ^
*
12
4,,
1 	 1 	 ' WHBt
1
L _».*«. IK TO
LAHOflU TOJBMJSHH
*
*"* WINE _^ BUM _ff"HE . _,. mlSE ""r^f
"" SMMUtlOU "^ BUBS-H^TIOH »" **"* '*"*
ft
tbOH Hft-p^
Hacuon -,*- \
StOMGe.
•s
V*
i+\ "2°
^^^ (wW-WHjO
	 • 	 I , , 	
>y» "SS^3 .
^ AW
UOlErKTKM
»^ rauriaiT)CN — „»» n_
i
unun
KOttGttl
i
SUCK nw —HJO ID RIVEH- — •»
                                                                                                          raw.
                                                                                                          DISQMIGE
                                                                                                           (131
                               8
                                        In I \ were aanplad in
                                         , ottw* during
                                                                                                               ^^wisrero
                                                                                                              "** HlVB*
                                                                                                           IS
                                                                                                          ___^»J «
                                                                                                             srowce
              Figure 11-2.   General process flow diagram at  plant #299 showing the sampling points.
                              Chlorine/caustic (irercury cell)  manufacture.

-------
         TABLE 11-4.  POLLUTANT OM23NTKATIONS AND LOADS AT PLANT # 299
SUBCATEGQRY
CHLORINE (MERCURY CELL)
Stream       Stream
Number    Description
                  TSS
            (rog/1)     (kg/kkg)
                 Mercury
             (mg/1)     (kgAkg)
Screening Phase:* *
   1      Cell Waste
          Mercury Treatment
          Effluent
               12       0.016


                5.0     0.0070
          Tail Gas
          Scrubber
Verification Ptese:
                  .(2)
          Mercury Treatment
          Influent

          Mercury Treatment
               NA
               91
  NA
0.13
0.15      0.0002


0.029     0.00004


0.11        NA
5,9
 NA = Not available.
  (1) - Data hased on one  72-hour composite  sample of each stream.
  (2) = Data based on three 24-hour composite samples of each stream.
0.080

3
4
5
Effluent
Cell Vfeste
Brine Mud
Tail Gas Scrubber
18
120
13,000
180
0.026
0.17
NA
0.022
0.20
11.
0.54
0.17
0.0003
0.015
NA
0.00002
                                      156

-------
     TABLE 11-5.  POLLUTANT CONCENTKVTIONS AND LOADS AT VERIFICATION PLANTS

SUBCATEGORY CHLORINE (MERCURY CELL)
Stream Stream
Number Description
Plant
1
2
3
4
5
6
7
Plant
5
6
7
8
9

Plant
1
2

3
4

5

6

7
8
Plant
1
2

4
747
Cell Waste
Treated Waste
Acid Input
Acid Output
Dechlor System
Cl2 Condensate
Tail Gas Scrubber
167
All C12 Wastes
Cell Wash
Brine Process
Treated Waste
Clarifier
Underflow 5
317
Cell Waste
Brine Mud
Filtrate
Tank Car Wash
Collection
Tank 21
Treated
Effluent
Deionizer
Effluent
N-C Cooling
Final Effluent
106
Cell Wash
Treated Cell
Wash
Final Effluent
TSS
Ung/J.)

700
60
NA
NA
9.0
2.0
NA

560
57
4.0
2.0

r900

45

520
18

,000

110

18
16
18

79

20
2,0
Uog/Jckg)

1.6 x 10~*
1.4 x 10"^
NA
NA
0.0037
2.7 x 10
NA

1,9 ,
5.7 x 10 ->
7.1 x 10,
1.3 x 10"^

4.0

NA

NA
NA

8.6
'-2
4.4 x 10

5.2 x 10"3
2.2
2.4





Mercury
(mg/i)

18
0.10
0.023
0.0030
0.035
0.27
0.039

3.8
0.72
0.0050
0.32

10.4

14

34
0.033

123

0.10

0.0010
0.0010
0.0020

3.9

0.015
<0. 00050
Ucg/Kkg)

4.3 x 10 j?
2.3 x 10 J?
3.5 x 10~I
7.2 x 10~'
1.5 x 10 J?
1.8 x 10 "2
8.0 x 10

1.3 x 10^
6.7 x 10 J?
9.0 x 10 ,
1.8 x 10
-5
8.7 x 10

NA

NA
NA
-2
5.0 x 10
-5
4.3 x 10
7
2.9 x 10"
1.4 x l(f;
3.6 x 10




NA
NA - Not available.
 (1) = Data based on the average of  three  24-hour composites.
                                     157

-------
At Plant #317, the brine purification mud is mixed with spent sulfuric
acid  and sodium hypochlorite solution.  The treatment removes mercury
from the mud and transfers  it  to  the  solution.   The  solution  is
filtered  and the solids landfilled.  The filtrate is mixed with other
mercury-contaminated wastewaters, which includes the brine purge, cell
room liquid wastes and plant area wash water.  This  is  then  reacted
with sodium hydrosulfide to precipitate the mercury as mercury sulfide
and then filtered.  The solids are sent to a mercury recovery unit and
the filtrate is sent to a holding tank.  The effluent from the holding
tank  is  mixed  with  de-ionizer  waste  and noncontact cooling water
before discharge.  The process flow diagram showing the waste  streams
sampled is given in Figure 11-5.

At  Plant  #106,  mercury-bearing  wastes  are  segregated  from other
wastewaters and combined for batch treatment.  Mercury-bearing  leaks,
spills,  and  precipitation  are  contained  and  collected by curbing
around the cell room and collecting the wastes in a common sump.  From
the sump the combined waste is pumped to treatment.  In the  treatment
system,  the pH is initially adjusted using waste sulfuric acid and 20
percent caustic solution as required.  Sodium sulfide and  filter  aid
are  added  and  the waste agitated in fiberglass reaction tanks.  The
effluent from the tanks is filtered and the filter  cake  is  retorted
for  mercury recovery.  The residual waste, after mercury recovery, is
placed in a lined solid waste disposal area.  The filtrate is sent  to
the  first  of two lined lagoons.  Primary pH adjustment is made using
waste sulfuric acid and 20 percent caustic before entry into the first
lagoon; final pH adjustment is  made  between  the  first  and  second
lagoons.

Descriptions of Plants Not Sampled

At  Plant  #589,  the wastewater going to the mercury treatment system
consists of cell room washdown, brine filter backwash, leaks,  spills,
cleanup  water,  and hydrogen cooling condensate.  The wastewaters are
reacted with hydrochloric acid and sodium bisulfide and then sent to a
settling basin where mercury sulfide precipitates.   The  overflow  is
passed through a series of effluent filters before discharge.

At  Plant  #343,  the cell room wash water, brine purification sludge,
and chlorine cooling condensate are combined and sent to a pond.   The
suspended  solids  settle in the pond and are dredged out once a year.
The dredged sludge is "Chem Fixed" and disposed of in  an  appropriate
landfill.   The  overflow  from  the pond is reacted with Na2S and the
reacted solution is sent to a  clarifier.   The  clarifier  underflow,
consisting  mainly  of  mercury sulfide, is returned to the pond.  The
clarifier overflow is discharged.

All contact wastewater at Plant #907 is treated for mercury removal in
a patented process involving reduction  of  mercury  to  the  metallic
state   using  sodium  borohydride.   Previously  contaminated  wooden
flooring  in  the  cell  room  has  been  removed  and  replaced  with
fiberglass  gratings  to  reduce the amount of mercury 'in the effluent
and for better waste control.  Molecular sieves have been installed on
                                   158

-------
£
                                                                                    HVOCM.ORI TE
                                                                                     SOLUTION
^ MINE
& BRINE
t3 SJSUH
-I!

DECHLORO.
fUtlHUV


"3ST"
lti«
IRINE
COHDS.
PROCESS UAiTE i
Att » ,

cii
1*9

A,
Cl 'I*
	 "-'t. * COMPRESSION
DRVIHG
HOUEFACTIOH
A
cow. T
tit HjSO^ UASTE
TO PONDS
PIMUF.
t


r 4
*r
SULFIOE
                                                  OTHER PRODUCT
                                                   UASTE WATER
        SWKJPG POiWS
                               SOLIDS 10.^	f
                               LAHOftLL   I
Figure 11-3.   General process  flow diagram at plant #747  showing the sampling points,
                Chlorine/caustic (mercury cell) manufacture.

-------
                                                                                  MttCOMTACT
                                                                       COOLING UATE*     U2°
                     Mint ptocesi IMIM
 10 pH AOJUSI-
 H£HI ANft FINAL
 OIUUAKCf
IMOOK

+ (T\
n
ION
iUHAHGE

ttfitHP



CA1ION




S**« f tiTEl


ClMtFtU
^
1
^
UNO
                                                                    fetl.
            M
SAMPLING POINTS
                                                                               UNOFIU
Figure 11-4.   General process  flow diagram at plant #167  showing the sanpling points.
                Chlorine/caustic (mercury cell)  manufacture.

-------
        sou
SALT 	1   I
(XV GAS
               U3GEMD

                 streams eaopled.
                                                                                        DISCIULRGB
      Figure 11-5.   General process flow diagram at plant #317 showing the sampling points
                     Chlorine/caustic  (mercury cell) manufacture.

-------
cell end boxes to reduce the mercury content in the  air  vented  from
the cells.  The treatment not only cleans the air but is also believed
to reduce mercury in the plant area runoff.

In  the  treatment  system,  the  mercury-contaminated  wastewater  is
reacted with sodium borohydride to reduce  dissolved  mercury  to  the
metallic  form.  The reacted solution is filtered prior to delivery to
one of the banks of three columns packed with anthracite coal.   After
passing  through  three  absorption  columns  in  series,  the treated
wastewater is delivered to large holding tanks, from which it  may  be
discharged or returned to treatment, depending on its mercury content.
Filter  cake,  resulting from the filtration of the waste prior to the
coal absorption step, is retorted for mercury recovery.

Waste solids at this facility, including mercury treatment sludges and
brine muds, are deposited  in  an  on-site  disposal  area.   Chlorine
discharges  are  essentially  eliminated  by  three  significant waste
management  practices:  the  chlorine  condensate  is  collected   and
returned to the brine system, tail gas scrubbing effluents are used in
the  manufacture  of  another  product,  and  spent sulfuric acid from
chlorine drying is  dechlorinated  in  an  air  stripper  and  shipped
off-site  for  the manufacture of another product.  Gases from the air
stripper are returned to the chlorine purification header.

At Plant #324, the barometric condenser on  the  brine  dechlorination
was  replaced  with  an  indirect  cooler, resulting in a reduction of
chlorinated wastewater.   The tail gas scrubber effluent  is  used  for
the  manufacture  of another product, and the brine muds are sent to a
pond.  Small amounts of mercury, when detected in the brine  mud,  are
leached   with'  water  and  treated  with  other  mercury-contaminated
wastewaters which include the cell room  wash  water,  caustic  filter
backwash,  and  brine  leaks.   The  combined wastewater is mixed with
hydrogen processing wastewater, reacted  with  sulfuric  acid,  sodium
borohydride,  and  sodium sulfide, and then filtered.  The filtrate is
adjusted for pH and recycled to process.

At Plant #385, the brine mud sludge is sent to a retention pond  where
it  accumulates.   All  process  contact wastewater is collected in an
unlined pond where it is treated and the treated effluent is  used  as
the  scrubber  liquid  for tail gases.  The spent scrubber solution is
sent to an adjacent paper plant for use.

At Plant #416, the cell room wastes are used for  bleach  manufacture.
The  wastewater streams from the chlorine/caustic plant are sent to an
acTjacent paper company.

At Plant #784, the wastewater, consisting of KC1 brine filter backwash
and area washdown and spills, is sent to a basin.  The basin equalizes
the flow and the overflow is  treated  with  sulfuric  acid  prior  to
reaction  with  NaHS and clarification.  The clarifier overflow passes
through an activated carbon filter  and  to  a  final  tank  where  it
undergoes pH adjustment before discharge.
                                   162

-------
The  wastes  are  segregated at Plant #674.  The clarification pond is
used for waste streams containing suspended solids.  The streams going
to the pond include brine  purification  muds  and  spent  chlorinated
lime.   The  mercury-contaminated  wastewaters are treated separately.
These include the brine saturation waste, brine filter backwash,  cell
room   sumps,    and  tank  car  washes.   The  combined  mercury-laden
wastewater is sent to a collection pond and the overflow from the pond
is pH adjusted before the addition of Na2S.  The reacted  solution  is
sent  to another pond and the pond overflow is passed through a carbon
adsorption column before final  discharge.   A  part  of  the  treated
effluent is reinjected into the brine well.

At  Plant  #012, the brine treatment area is paved to trap all spills,
leaks, and rain runoff from that area.  The  contaminated  wastewaters
from  the  plant  are  re-injected  into  the  brine wells to keep the
hydraulic balance and maintain pressure in the salt deposits.
Summary of the Toxic Pollutant Data

Presented below are the toxic  pollutants
during screening and verification.
found  in  the  raw  wastes
Because  several  waste  streams  usually  contribute to the total raw
waste at mercury cell plants, a calculation  was  often  necessary  to
determine the pollutant concentrations that would exist in the streams
before  they  were  mixed  prior  to  treatment.   An  example of this
calculation is the "mixing" of the following hypothetical streams:
     Stream A: 100 gallons per minute, 15 mg/1

     Stream B:  10 gallons per minute, 60 mg/1

(Flow x concentration) - (Flow x concentration)
                       Total Flow

     = concentration of mixed streams

- (100 qpm) (15 mq/1 + (10 qpm) (60 mq/1) = 19 mg/1
               110 gpm

The maximum  raw  waste  concentrations  observed  during  any  single
24-hour sampling period were:
                                  163

-------
              Maximum Raw Waste Concentrations Observed
                                Ug/1)
Pollutant
Screening
  Plant
  (#299)
    Verification
       Plants
#299,  #747,  #167,
#206,  #317)
Antimony
Arsenic
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Silver
Thallium
Zinc
250
 10
  1
  8
350
  1
150
100
  1
140
230
         770
         400
         790
         180
       2,300
       1 ,900
     180,000
       2,400
         870
         440
      34,000
Section-  5  of  this report describes the methodology of the screening
and verification sampling  program.   In  the  chlorine  mercury  cell
industry,  a  total  of  18  days of sampling were conducted at Plants
#299, #747, #167, #317 and #106.  Thirty-two different sampling points
were involved covering various  raw  waste  streams  and  the  treated
effluents  at  these plants.  The evaluation of toxic metal content of
these process related waste streams was based on 949  analytical  data
points.  The screening for toxic organic pollutants at Plants #299 and
#167  generated  an  additional 490 analytical data points.  The daily
raw waste loads were  calculatd  from  the  waste  stream  flow  rates
measured  or  estimated  at  the  time  of  sampling  and the measured
pollutant concentration.
     The daily loading is determined by:

          Daily loading  (as kg of pollutant =
          per day)
                          C)(Q)
                          1000
     where:
          C  is the  concentration  of  the pollutant  expressed
          of mg/1  (Note: kg/m3  =  1000 mg/1),  and
                                     in   units
          Q  is the waste stream  flow  rate  expressed  in  units  of m3/day
          (m3, a  cubic meter,  is equal  to  264.2  U.S.  gallons).

Similarly,   the   unit  loadings   were  calculated  from  the   reported
chlorine production  rate,  the  waste stream flow  rate, and the measured
pollutant concentration:
                                    164

-------
     Unit loading (as
     kkg of chlorine)
kg of pollutant per
(CHQ)
 1000P
where C and Q are the same as described above, and P is  the  chlorine
production  rate  expressed  in  units  of  kkg/day (kkg is 1000 kg, a
metric ton, which is equal to 2205 Ibs.)

The minimum, average, and maximum values are based on data from  those
plants  where  the  particular  pollutant was found at a concentration
greater  than  the  analytical  detection  limits  and  considered   a
"significant  concentration".   The  term  "significant concentration"
means an observed concentration in any 24  or  72-hour  composite  raw
waste  sample  that  is  above  the  analytical  detection  limit, and
potentially treatable.

In Table 11-6, the toxic pollutant raw waste data are presented as the
average daily concentrations  and  the  unit  loadings  found  at  the
individual  plants.   These  averages  were  derived  by averaging the
concentrations and loads based on three 24-hour composite samples from
each plant.

In Table 11-7 daily loadings (in kg/day)   and  unit  loadings  in  (in
kg/kkg)  are presented as minimum, average and maximum values based on
the data presented in Table 11-6.

Based on the total annual  production  of  this  subcategory  and  the
average  waste  load  generated  per unit product, the estimated total
pollutant raw waste loads generated each year by this subcategory  are
as follows:
                    Pollutant

                    Antimony
                    Arsenic
                    Cadmium
                    Chromium
                    Copper
                    Lead
                    Mercury
                    Nickel
                    Silver
                    Thallium
                    Zinc

Pollution Abatement Options

Toxic Pollutants of Concern
                            Raw Waste load
                                 (kg/year)

                                 1,400
                                 1,000
                                   210
                                   360
                                   960
                                   880
                                44,000
                                   820
                                   850
                                   770
                                 7,200
                                              of
Mercury  is  the major toxic pollutant of concern in the production
chlorine by the mercury cell process.  Other toxic metals often  found
in significant concentrations in raw wastes include arsenic, antimony,
                                   165

-------
cadmium,  chromium,  copper,  lead, nickel, silver, thallium and zinc.
Sources of these metals are assumed to be impurities in the  raw  salt
or brine and corrosion products from the reaction between chlorine and
process  equipment  materials of construction.  No toxic organics were
found at treatable levels.

Prevailing Control and Treatment Practices

Specific control and treatment practices at 14 plants  were  described
above.    All   known   mercury  cell  plants  practice  treatment  of
mercury-bearing wastes, but control practices  such  as  recycling  of
brine  mud  filtrate  or pond overflow, chlorine condensates, hydrogen
condensates and caustic filter backwash, and solids handling vary from
plant to plant.  Although all known treatment  facilities  precipitate
mercury  and  separate  the  solids  formed  by  clarification  and/or
filtration, sampling data  has  shown  that  some  treatment  systems,
including  those with more advanced technologies such as adsorption or
ion exchange, are not operating efficiently.

Process Modifications and Technology Transfer Options

The following process modifications are being practiced at one or more
mercury cell plants  and  can  significantly  reduce  pollutant  loads
discharged.

A.   Anode Material

     Nearly all mercury cell plants now use metal anodes.  Their  use,
     as  opposed  to graphite anodes,  improves the power efficiency of
     the cells and reduces the potential pollutant load.

B.   Liquefaction of Chlorine

     Utilization of  high  pressure  and  refrigeration  for  chlorine
     recovery will reduce the chlorine content of tail gases.

C.   Brine Recycling

     Although practiced at many facilities not all plants are using   a
     closed-loop  brine  system  which  eliminates a significant waste
     volume requiring mercury treatment.

D.   Mercury Emissions

     Hydrogen gas produced in the  denuder  can  be. refrigerated  and
     passed  through  treated carbon or molecular sieves to remove the
     mercury  escaping  with  gas.   This  will  reduce  the   mercury
     emissions  and  reduce atmospheric fallout in the neighborhood of
     the plant.  This in turn will reduce  mercury  concentrations  in
     storm runoff.  Two plants are practicing this control technology.
                                    166

-------
     TABLE 11-6.  AVERAGE TOXIC POLLUTANT RAW WASTE CCMCENTRATICNS AND
                  LOADS AT VERIFICATICN PLANTS  (1)
\hg/kkgy

SUBCATEGORY
Pollutant

Antimony

Arsenic

Cadmium

Chromium

Copper

Lead

Mercury

Nickel

Silver

Thallium

Zinc

CHLORINE (MERCURY CELT.)
Plant 1
299
0.48
0.00077
0.23
0.00037
0.010
0.000016
0.063
0.00010
0.30
0.00048
0.060
0.000096
5.9
0.0094
*

*

0.18
0.00029
0.27
0.00043
747
0.11
0.000076
0.030
0.000021
0.020
0.000014
0.10
0.000069
0.38
0.00026
0.16
0.00011
18
0.012
0.093
0.000064
0.047
0.000032
0.022
0.000015
0.69
0.00048
167
*

0.33
0.0018
*

0.12
0.00067
0.075
0.00042
0.072
0.00040
3,8
0.021
0.060
0.00034
*

*

0.17
0.00095
317
*

0.10
0.000051
0.46
0.00023
0.080
0.000041
1.2
0.00061
1.4
0.00071
123
0.063
1.4
0.00071
0.11
0.000056
*

20
0.010
106
0.49
0.00033—
*
- ..
0.031
0.000021 —
0.013
0. 0000087- -
0.12
0.000080
0.33
0.00022
3.9
.0.0026 (
0.17
0.00011 i
0.58
0.00039
0.38
0.00025
0.96
0.00064
                                                                                A
* - Concentration

 (1) Data based on
below treatable level.

the average of three 24-hour composites,
                                                                                     ° 7
                                     167

-------
                      11-7.   SUMMARY OF RAW V&SEE LOADINGS AT
                             VERIFICATION PLANTS
X
SUECA1EGORY
Pollutant
Antimony
Arsenic
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Silver
Thallium
Zinc

min.
0.044
0.0054
0.0062
0.0043
0.045
0.036
1.6
0.037
0.0059
0.0090
0.14
CHLORINE
Daily
loadings
(kg/day)
avg.
0.17
0.11
0.013
0.037
0.10
0.070
3-1
0.056
0.082
0.086
0.41
(rtERCURY CELL)

max.
0.30
0.27
0.025
0.098
0.18
0.12
5.1
0.075
0.22
0.14
1.1
S
,_... v._
f Unit
loadings
(kg/kkg)
min.
0,000076
0,000021
0.000014
0.0000087
0,000080
0.000096
0,0026
0,000064
0,000032
0.000015
0.00043
avg.
0.00039
0.00056
0.000070
0.00019
0.00037
0.00031
0.022
0.00015
0.00016
0.00019
0.0025


1 Number of
Plants
Averaged*
max.
0.00077
0.0018
0.00023
0.00067
0.00061
0.00071
0.063
0.00071
0.00039
.0.00029
0.010

3
4
4
5
5
5
5
4
3
3
5
* - Only those plants where the pollutant was observed at "significant
    concentrations" are included in the averaging.   "Significant
    concentrations" is defined in 11.4.4.
                                     168

-------
E.   Tail Gas Emission Control

     When chlorine gas  produced  from  the  cell  is  compressed  and
     cooled,  chlorine separates as liquid chlorine, and noncondensable
     gases  (tail or sniff gas containing residual chlorine vapor) are
     produced at the discharge end of the condenser.   The  amount  of
     chlorine  present  in  the  tail gas is significant and has to be
     removed and treated or recovered before the tail gas is-vented to
     the atmosphere.  The common industrial practice is to  scrub  the
     gas  with  caustic  soda  or  lime  solution  thus  producing the
                                  The hypochlorite solution is  either
                                  to  a wastewater treatment plant, or
                                     Treatment  of  this  waste  is  a
                                   Decomposition is a common method of
                                  thermal,  and  chemical  methods  as
corresponding hypochlorite.
sold,  used  on-site,  sent
discharged without treatment
relatively  recent practice.
treatment using  catalytic,
described below.
     Catalytic decomposition involves the addition of small quantities
     of  cobalt,  nickel,  and  iron  chloride  to  the waste streams,
     followed by retention in reaction tanks for periods up to several
     days.  Of the two plants employing this technology,  one  reports
     zero  discharge  of  chlorine,  and  the other reports respective
     average and maximum chlorine discharge rates of 0.015 and 0.14 kg
     per metric ton of chlorine produced.

     Thermal decomposition occurs when the temperature of the solution
     containing hypochlorite reaches 175 degrees F.  Lime reacts  with
     chlorine exothermically, producing heat and calcium hypochlorite.
     If the hypochlorite solution is not cooled, thermal decomposition
     occurs.   One  chlorine/caustic  plant  is  using  this treatment
     method and another is  planning  to  use  it.   The  plant  using
     thermal decomposition reports complete conversion of hypochlorite
     to chloride.

     Chemical  decomposition  takes place by reacting the hypochlorite
     solution with a chemical reactant which is usually sodium sulfite
     or hydrogen peroxide.  Chemical decomposition  is  expensive  but
     complete and rapid.

     When  chlorine is present in a dissolved form (hypochlorous acid)
     in water, a stripping technique may be  applied  to  recover  the
     chlorine.   Chlorine condensate streams and spent chlorine-drying
     acid are most commonly treated by steam or vacuum stripping, with
     the chlorine frequently returned to process for purification  and
     recovery  as  a  product.  The tail gas is not generally scrubbed
     with water because water does not effectively remove chlorine and
     the chlorine concentration in the exhaust will reach 0.1  to  4.5
     percent  by  volume  after  scrubbing  with water.  One effective
     method of chlorine recovery from the tail gas is by  the  passage
     of   the  gas  through  an  absorbing  material  such  as  carbon
     tetrachloride and  subsequent  recovery  of  the  chlorine.   The
     process is proprietary and little information is available on its
     design or performance.
                                    169

-------
Best Management Practices

A.   Area Runoff

     Provisions can be made to divert and contain  storm  runoff  from
     plant areas.  Collected runoff can then be sent to the wastewater
     treatment system.

B.   Leaks and Spills

     The brine treatment area and the cell room  areas  can  be  paved
     with fiberglass gratings, and provision should be made to collect
     the leaks and spills from the operation.

C.   Mercury Contaminated Solids

     The precipitated mercury waste should be stored in a lined  pond,
     disposed  of  in  a  secured landfill or sent to mercury recovery
     operations.  Brine mud should be discharged to a lined pond or  a
     secure  landfill  after filtration.  The brine mud contains small
     amounts of mercury which can  leach  into  the  ground  water  if
     proper safety precautions are not taken.

D.   Transportation, Handling and Abnormal Operations

     Provisions should be made to remove chlorine from  air  emissions
     resulting from abnormal operating conditions such as start up and
     shut  down,  or  from  vents  on  returned  tank cars, cylinders,
     storage tanks, and process transfer  tanks  during  handling  and
     loading of liquid chlorine.

Advanced Treatment Technologies


Methods  available  for  the  removal of elemental mercury or mercuric
salts from plant wastewaters include precipitation with sodium sulfide
to form insoluble mercuric sulfide, adsorption  by  activated  carbon,
adsorption by ion-exchange and other resins, reduction by borohydride,
hydrazine,  sulfite,  hypophosphite, or iron, and biological reduction
(57).  All of these methods are patented; many of these  methods  have
been  proven  on  a  pilot  scale  only.   Sulfide  precipitation  and
adsorption techniques will also provide for the removal of other toxic
metals.

Selection of Appropriate Technology and Equipment

Technologies for Different Treatment Levels

Following the evaluation of significant toxic pollutants found in  raw
wastewaters,  current  industry  treatment  practices  and  applicable
treatment alternatives,  two  levels  of  end-of-pipe  treatment  were
selectee5,  as alternatives for application in the mercury cell chlorine
subcategory.
                                     170

-------
A.   Level 1  (BPT)

     This   treatment   consists   of   sulfide    precipitation    of
     mercury-bearing wastewater followed by pressure filtration.  This
     level  of  treatment,  which will also reduce other heavy metals,
     includes recycle of the brine mud overflow or  filtrate  back  to
     process,   and   the   settling   and   storage  of  brine  muds.
     Mercury-bearing  solids  can  be  sent  to  mercury  recovery  or
     disposal.    The flow diagram for this treatment level is shown in
     Figure 11-6.

B.   Level 2

     The filtered Level  1  effluent  is  passed  through  a  granular
     activated  carbon  bed  where  residual  metal  suIf ides  and any
     metallic mercury will be removed.   The  flow  diagram  for  this
     treatment level is shown in Figure 11-7.  Cost estimates for this
     level  can  be  obtained  from  the proposed Development Document
     (60).

Equipment for Different Treatment Levels

A.   Equipment Functions

     In  Level   1,   typical   of   existing   treatment   facilities,
     mercury-bearing  wastes  are  equalized  in  a  surge  tank,  and
     following chemical mixing, sulfide precipitates are removed in  a
     conventional  plate  and  frame filter press followed by final pH
     adjustment of the  filtrate  before  discharge.   In  Level  2  a
     conventional  granular  activated  carbon  filter  is  added  for
     further removal of residual metals before pH adjustment.


B.   Chemical Handling

     Sodium bisulfide is used with filter aid after pH  adjustment  to
     pH  5-7.  Care is needed to prevent escape of toxic and obnoxious
     H2S fumes at neutral and acid pH level.  At Level 2 no additional
     chemicals  are  used  since  the  activated  carbon  bed  is  not
     regenerated  but  is  periodically  removed  and  replaced.   The
     handling of granular carbon may cause temporary dust problems but
     it causes no special hazards.

C.   Separation and Removal of Solids

     Conventional  settling  and  filtration  methods  are  used,  but
     because  of  the toxicity of mercury, precipitated sludges should
     be disposed of in a safe chemical waste area.

D.   Monitoring Requirements

     Both levels of treatment  include  provisions  for  sampling  and
     monitoring  of  the  wastewater  discharge.   Monitoring of heavy
                                    171

-------
     metals is done  by  atomic  absorption  methods  at  a  qualified
     commerical  laboratory.   Simple field tests for heavy metals as a
     group are available for routine process control.

Treatment Cost Estimates
General Discussion
To prepare
developed.
treatment  cost  estimates,   a  model
The model plant characteristics are:
                                                   plant  concept  was
A.   Wastewater Flow
     Data presented in Table 11-3 indicate an average wastewater  flow
     of 2.1 mVkkg for 13 plants, while the average of the five plants
     surveyed during this study averaged 1.7 mVkkg.
     For  effluent  limitation  calculations (see 11.8.2)
     estimation the more conservative unit flow from the
     base (2.1  mVkkg) has been used.

B.   Chlorine Production
                                              and for
                                              larger
                                                                  cost
                                                                  data
     Approximately 50 percent of  the  production  data  for  all  the
     chlorine/caustic plants using mercury cells is available on file.
     Production  ranges  from  19,000 to 198,000 kkg of chlorine/year.
     Three model plants with  productions  of  19,100  kkg/yr,  95,500
     kkg/yr   and  191 , 000  kkg/yr  were  selected  to  represent  the
     subcategory production range.  The flow per unit of production is
     assumed  to  be  the  same  for  each  size   of   model   plant.
     Seventy-seven  percent  of  the  plants  for  which flow data was
     available have flows per unit of production equal to or less than
     the average unit flow (Table 11-3).


C.   Solid Waste Produced

     Brine mud constitutes the major source of solid  waste  generated
     at  chlorine  plants.   Although  flows  and  solids content vary
     considerably from plant to plant, a.n average flow of 0.42  mVkkg
     at  10  percent suspended solids ga.ve an estimated solids load of
     42  kg/kkg  to  be  used  for  cost  estimating  purposes.    The
     implementation of RCRA regulations has not been included in these
     estimates,  but  RCRA costs are considered in the Economic Impact
     Analysis of Pollution Control Technologies for  Segments  of_  the
     Inorganic Chemicals Manufacturing Industry, EPA 440/2-81-023.

Chlorine Bearing Wastes

In  the selection of model plants, the following assumptions have been
made for  the  chlorine  contaminated  waste  streams.   The  chlorine
condensate  waste  stream  has  not been included in the waste streams
going  to  the  treatment  facility.    in   the   majority   of   the
                                    1/2

-------
         HJDS1BEW
                                      IMXON
                                      LMXKK
                                                                        .
                                                                    erooss
                     SUUVR1C ACID
                  lODIHS 1HK
FILTER
 AID
                                                     SCDIW
                                                    D
                                     raxin?
                                                BEIUVI LIW
                                                                              FIUlfH
                                                                           SCUDS ID
                           OR 1AHVIU.   |

                          	J
          Incluiirt pH monllorlnj(, flow monitoring and
Figure  11-6.  tevel 1 waste  water treatment  for chlorine  - mercury cell subcategory

-------
      CMNB »H_
       STIVMI
                             UQQCN
-*. BEOICtETO
   PKKXSS
            Include* pH monitoring, How monitoring and
Figure 11-7.   Level 2 waste water treatment for chlorine - mercury cell  subcategory.

-------
chlorine/caustic  plants, this stream is stripped of chlorine by steam
or vacuum and the chlorine is recycled to the purification  operation.
The  wastewater  is then returned to the process and introduced to the
brine purification unit or sent to the treatment unit.   The  quantity
of  wastewater  generated  by  this  operation  is  small and does not
significantly affect  the  flow  determination.   In  some  cases  the
chlorine  gas  from  the  cells  is  contact cooled with water and the
scrubbed liquid, after steam  stripping,  is  reused.   The  stripping
operation  in  the  recovery  of  chlorine is part of the process and,
therefore, its cost is not included in the treatment cost.  The  spent
tail  gas  scrubber  solution, which contains mainly calcium or sodium
hypochlorite, is assumed  to  be  used  or  decomposed  before  it  is
discharged  or  sent  to  treatment.   Thermal  decomposition  can  be
practiced at no additional cost  at  some  facilities,  while  another
efficient  treatment  method  is  catalytic  decomposition.   The cost
estimates for decomposition are not  included  here  because  at  many
plants   the  hypochlorite  stream  is  sold,  used  on-site  or  only
infrequently discharged depending on market demand.

However, because of  the  environmental  effects  of  high  levels  of
chlorine   in   process   wastewater  discharges,  the  cost  for  the
dechlorination of such streams using sulfur dioxide has been  included
because  this  is  the  treatment  method  on  which  control of total
residual chlorine is based.

Model Plant Treatment Costs

On the basis of the model plant  specifications  and  design  concepts
presented  earlier,  the estimated costs of treatment for three models
having different production levels are shown in Tables 11-8, 11-9  and
11-10.   The  costs  of  BAT treatment are incremental over BPT costs.
Table 11-11 presents a summary of the unit cost  distribution  between
amortization and operation and maintenance components.

Basis for Regulations


Basis for BPT Limitations

A.   Technology Basis

     Existing mercury cell chlorine plants are controlling mercury  in
     their  wastewaters  in  accordance  with existing BPT regulations
     which require a discharge of less than 0.00014 kg/kkg of  product
     as a 30-day average.  The BPT regulations presently in effect (40
     CFR  415.62  (a))  will  not  be  revised.   Pollutants regulated
     include TSS, pH and mercury.  The  technology  basis  of  sulfide
     precipitation and filtration of mercury bearing streams (Level 1}
     is  currently  being  applied  at  24 plants in this subcategory.
     Other plants in the industry use mercury control methods that are
     different in detail but with the same objective.
                                  175

-------
     The existing regulations apply at the treatment  system  effluent
     for  TSS  and  mercury  and  at the plant effluent for pH.  These
     regulations are presented in Table 11-2 and are sustained by  the
     fact  that  plants  having  properly operated BPT technology have
     demonstrated the achievability of the effluent limitations  based
     on  available  long-term  monitoring  data.  Table 11-12 presents
     data from eleven mercury cell plants, seven of which are  meeting
     the  30-day  average  limitations.   The  other  four plants have
     mercury control technology installed  but  are  not  meeting  BPT
     limits.

B.   Flow Basis

     The existing regulations contain only load  limitations,  kg/kkg,
     and  no  flow basis or concentration limit was provided.  But the
     regulations did consider  the  inclusion  of  noncontact  cooling
     water in determining discharge load limitations.

Basis for Final BAT Effluent Limitations

The  original  BAT  limitations  for  this  subcategory  required zero
discharge of process wastewater pollutants.   These  regulations  were
remanded  and  are not in effect.  The final regulations allow for the
discharge of process wastewater following treatment.

A.   Technology Basis

     Utilizing the cost estimates presented in this report, the Agency
     has analyzed the cost  effectiveness  of  Level  1  and  Level  2
     treatment  options for pollutant removal.  The economic impact on
     the mercury cell  chlorine  subcategory  has  been  evaluated  in
     considering the technology basis for final BAT limitations.

     For  BAT,  the  Agency  is  promulgating limitations based on BPT
     technology  (Level  1)  with  the  addition  of   dechlorination.
     Dechlorination  is  being included in BAT because the toxicity of
     chlorine to aquatic life is well documented   (59)  and  it  is  a
     pollutant  of  concern  to the Agency.  Dechlorination, currently
     practiced by at least two plants, may be required only  at  fewer
     than  half  of the plants in the subcategory because hypochlorite
     produced in tail gas scrubbers is often sold  or  used  in  other
     operations  while  residual  chlorine  in  condensates is usually
     stripped or recovered.  Table 11-13  presents  residual   chlorine
     discharges  at  plants  that  have  reported  the  use,   sale  or
     treatment of chlorine-bearing wastewaters.  This  data  indicates
     that  some  plants  will  be  able  to meet the residual  chlorine
     limitations without the application of additional technology.

     The Agency considered  the  addition  of   carbon  adsorption  for
     additional  mercury  removal but  rejected  its use because of high
     cost and questionable performance in this  industry.
                                    176

-------
B.   Flow Basis

     The flow basis for BAT limitations is 2.1  mVkkg  based  on  the
     average  of  discharge data of 13 plants presented in Table 11-3.
     The order of magnitude of this unit flow volume was supported  by
     data  obtained  during  sampling " visits  to five plants at which
     flows ranged from 0.5 mVkkg to 5.6 mVkkg with an average of 1.7
     mVkkg.

C.   Selection of Toxic Pollutants to be Regulated

     The  selection  of  pollutants  for   which   specific   effluent
     limitations  are  promulgated  is  based on the evaluation of raw
     waste concentrations found during the sampling program and on the
     treatability of toxic pollutants using BAT technology.

     Table 11-14  presents  the  achievable  concentrations  of  toxic
     pollutants  using  the  BAT  technology  of sulfide precipitation
     followed by filtration.  The first column  gives  the  literature
     based  treatability data presented in Section 8 and summarized in
     Table 8-11 and reflects  the  lowest  level  achievable  by  this
     technology.  The second column gives actual industrial wastewater
     treatment system performance as presented in Tables 8-12 and 8-13
     of  Section 8.  Also presented in.Table 11-14 are the maximum and
     average raw waste concentrations of -toxic pollutants found during
     the sampling program with an indication of the number  of  plants
     where   the   literature-based   treatability  concentration  was
     exceeded.

     Based on the occurrence of treatable  levels  of  specific  toxic
     metals  in raw wastes and the fact that the sulfide precipitation
     technology is already utilized as BPT  in  the  chlorine  mercury
     cell  subcategory,  all  the  metals  in  Table  11-14,  with the
     exception of chromium and thallium for  which  no  sulfide/filter
     treatment  data are available, are candidates for BAT regulation.
     Consideration of Section 8 on the control parameters for  sulfide
     precipitation,  however, leads to the selection of mercury as the
     toxic pollutant to be  regulated.   Antimony,  arsenic,  cadmium,
     chromium,  copper,  lead,  nickel,  silver, thallium and zinc are
     included for guidance but no limits are set  because  control  of
     mercury will control these other metals.


D.   Basis of Pollutant Limitations

     Limitations  are  presented  as  unit  loadings  (kg/kkg)  and/or
     concentrations  (mg/1)  for  each  pollutant.   The  relationship
     between the two is based on the unit flow rate.  Although  actual
     unit  flow  rates  at plants vary by an order of magnitude due to
     such factors as raw materials and plant  control  practices,  the
     Agency  has determined that the unit loading effluent limitations
     (kg/kkg} can be met by well-operated treatment facilities.
                                    177

-------
     TABLE  11-8. MODEL PLANT TREATMENT  COSTS
Subcategory
Production
Chlorine - Mercury cell
19,100 metric tons per year
A. INVESTMENT COST

     Site development  	
     Equipment  	
     Monitoring equipment  ..

          Subtotal  ....ft... .
     Contractor's 0  &  P .. ..

          Subtotal  	
     Engineering  	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL INVESTMENT  COST  .


B. OPERATION AND
   MAINTENANCE  COST

     Labor and  supervision  .
     Energy 	
     Chemicals  	
     Maintenance  	
     Taxes and  insurance ...
     Residual waste  disposal
     Monitoring ,  analysis
      and reporting  	
                                                     ($)
                               BPT

                            26,500
                            93,000
                            20,000

                           139,500
                            20,925

                           160,425
                            32,085

                           192,510
                            19,251

                           211,761
                            21,000

                           232,761
     TOTAL OPERATION AND
     MAINTENANCE COST
                           112,000
                              1,300
                                500
                             21,176
                              6,983
                              4,500

                             15,000
                            161,459
   BAT

     0
23,000
     0

23,000
 3,450

26,450
 5,290

31,740
 3,174

34,914
     0

34,914
14,000
   500
 1,500
 3,491
 1,047
     0

 7,500
28,039
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                             34,454

                            195,912
 5,681

33,719
a Represents  the  incremental  cost  above  that for BPT treatment
  Overhead and Profit
                              178

-------
    TABLE 11-9.  MODEL PLANT TREATMENT  COSTS
Subcategory
Production
Chlorine - Mercury cell
95,500 metric tons per year
A. INVESTMENT COST

     Site development 	
     Equipment 	
     Monitoring equipment  ..

          Subtotal 	
     Contractor's 0 & Pb....

          Subtotal 	
     Engineering  	

          Subtotal 	
     Contingencies 	

          Subtotal 	
     Land	

     TOTAL INVESTMENT COST  .


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  .
     Energy	
     Chemicals	
     Maintenance  	
     Taxes and insurance  ...
     Residual waste disposal
     Monitoring,  analysis
      and reporting  	
                                                    ($
                               BPT

                            79,500
                           163,000
                            20,000

                           262,500
                            39,375

                           301,875
                            60,375

                           362,250
                            36,225

                           398,475
                            63,000

                           461,475
     TOTAL OPERATION AND
     MAINTENANCE COST
                           112,000
                             3,700
                             2,500
                            39,848
                            13,844
                            21,500

                            15,000
                           208,392
   BAT*

     0
40,000
     0

40,000
 6,000

46,000
 9,200

55,200
 5,520

60,720
     0

60,720
21,000
   700
 7,500
 6,072
 1,822
     0

 7,500
44,594
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                            64,832

                           273,224
 9,879

54,473
a Represents the incremental cost above  that  for  BPT  treatment
k Overhead and Profit
                              179

-------
    TABLE  11-10.  MODEL  PLANT TREATMENT  COSTS
Subcategory
Production
Chlorine - Mercury cell
191,000 metric tons per year
A. INVESTMENT COST

     Site development  	,.
     Equipment  	
     Monitoring equipment  	

          oUDtOtaJ.  •• • • • » ••••••••
     Contractor' s 0  &  P b.	

          Subtotal  	
     Engineering  	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL INVESTMENT  COST  	


B. OPERATION AND
   MAINTENANCE  COST

     Labor and  supervision  	
     Energy  	
     Chemicals  	
     Maintenance  	
     Taxes and  insurance  	
     Residual waste  disposal  ....
     Monitoring,  analysis
      and reporting  	

     TOTAL OPERATION AND
     MAINTENANCE  COST
                                                     ($)
                               BPT

                           130,000
                           246,000
                            20,000

                           396,000
                            59,400

                           455,400
                            91,080

                           546,480
                            54,648

                           601,128
                           123,000

                           724,128
                            112,000
                             6,500
                             5,000
                             60,113
                             21,724
                             42,500

                             15,000
                           262,837
   BAT

     0
60,000
     0

60,000
 9,000

69,000
13,800

82,800
 8,280

91,080
     0

91,080
28,000
 1,300
15,000
 9,108
 2,732
     0

 7,500
63,640
C. AMORTIZATION OF
   INVESTMENT COST

   TOT^L ANNUAL COST
                            97,804

                            360,640
14,819

78,459
f- Represents  the  incremental  cost  above  that £or BPT treatment
  Overhead and Profit
                              180

-------
        TABLE 11-11. MODEL PLANT UNIT TREATMENT  COSTS
     Subcategory Chlorine - Mercury  cell
                                Annual  Treatment Costs ($/kkg)
                                LEVEL OF TREATMENT
COST ITEM
Annual Operation
and Maintenance


Annual
Amorti zation


Total Annual
Cost


PRODUCTION
(kkg/yr)

19
95
191

19
95
191

19
95
191

,100
,500
,000

,100
,500
,000

,100
,500
,000
BPT

8.
2.
1.

1.
0.
0.

10.
2.
1.

45
18
38

80
68
51

26
86
89
BAT*

1.
0.
0.

0.
0.
0.

1.
0.
0.

47
47
33

30
10
08

77
57
41
*Represents the incremental cost above BPT
                             181

-------
    TABLE 11-12.   MERCURY DISORRGES FROM SELECTED CHLQR-^LKALI
                  CKT.T. PLANTS*

SUBCATEGOKY
Plant
#343
#907
#898
#195
#106
#589
#299
BHffimrt •
im
dgH*£*
r***££?±':jg
#195**
#324**

Average
0.000025
0.000020
0.000060
0.000040
0.000065
0.000055
0.000040
.^?9iB@|s$f

0.000022
0.00086
CHLORINE (MERCURY CELL)
Mercury
Daily Maximum
0.00094
0.00026
0.0025
0.00073
0.00022
0.00086
0.00019
» ' uldoQa^t
:•----*•-—"- -^ ^"-,"
0.00066
0.0022

Waste Load Ocg/kkg)
Maximum 30 -day Average
0.00029
0.000030
0.00043
0.00015
0.000096
0.00049
0.000056
0.0000001

0.00010
0.0018
*    See Reference 3

**   Fran Plant Long Term Monitoring Data presented in Appendix A.
                                    182

-------
BAT limitations, which apply at the treatment system effluent for
toxic metals and at the plant effluent for residual chlorine, are
presented in Table 11-15.

1.   Chlorine - Total residual chlorine limits are  based  on  an
     evaluation  of  long-term monitoring data for total residual
     chlorine as presented in Appendix A  (Tables  A-la  and  c).
     The  -long-term  average  concentration  is  0.64  mg/1.   The
     average variability factor for daily measurements  of  total
     residual  chlorine is 2.3 and the average variability factor
     for 30-day averages is 1.4.

     The 24-hour maximum concentration is:

     (0.64 mg/1) (2.3) = 1.5 mg/1

     The maximum 30-day average concentration is:

     (0.64 mg/1) (1.4) = 0.90 mg/1

     The load limitations for total  residual  chlorine  (kg/kkg)
     are  calculated  based  on the unit flow rate of 2.1 m3/kkg,
     thus:

     (1.5 mg/1) (2.1 mVkkg) (  kq/m3  ) = 0.0032 kg/kkg
                             (1000 mg/1)

     for the 24-hour maximum limit.  The maximum 30-day average
     limit is calculated similarly, i.e.,

     (0.90 mg/1) {2.1 mVkkg) (  kg/tn3   ) = 0.0019 kg/kkg
                              (1000 mg/1)

2.   Toxic Pollutants

     The effluent  limitations and  guidelines  for  the  selected
     toxic   pollutants   are   derived  from  three  sources  of
     information:    industrial   wastewater   treatment   system
     perfornamce   data   (Table  8-12   and  8-13),  verification
     sampling data, and  literature based  treatability  estimates
     (Table 8-11).

     The  results  of analysis of treated effluent represent plant
     performance observed  during  three  days  of  screening  or
     verification   sampling.    The   effluent  data  for  toxic
     pollutants  found   above   treatable  concentrations   in  raw
     wastes  are   summarized  in Table 11-16.  Data are presented
     from  four  plants  practicing   BPT   technology   (sulfide
     precipitation followed  by  filtration).  Sampling data for
     the fifth plant, #299, reflect  effluent  quality  prior  to
     filtration.

     a.   Mercury
                                 183

-------
           TABLE 11-13.   RESIDUftL CHLORINE DISCHARGES AT SELECTED
                         CHLOR-ALKALI PLANTS*

Plant
#207
#014
#819
#747
#106
#589
1747(2)
#324 (2)
Chlorine Waste Load
Average
0.33
0.040
ND(1)
0.0020
0.0010
0.0030
0.0025
3.72
(kg/kkg)
Range
1.4 maximum
0 to 1.29
0.016 to 0.14
0 to 0.0060
0 to 0.14
0.0010 to 0.011
ND
0.38 to 12.2
*  See Reference 3

(1)  - ND = No data

(2)  - Frcm Plant Long-Term Monitoring Data
                                     184

-------
           TABLE 11-14.  COMPARISON OF RAW WASTE CONCENTRATIONS OF
                         TOXIC POLLUTANTS WITH TREKTABILITY

SUBCATEGORY


Literature-
Treatability
Pollutant (mg/1)
Antimony
Arsenic
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Silver
Thallium
Zinc
ND
0.050
0.010
ND
0.050
0.050
0.010
0.050
0.050
ND
0.020
CHLORINE
Industrial
(MERCURY CELL)
Number Plants
out of Five
Waste Water Maximum
Treatment System Plant
Performance Average
(mg/1) (mg/1)
0.23(2)
0.15(3)
0.1S(4)
ND
0.20(3)
0.10 (3)
0.034(3)
0.022*41
0.070C4)
ND
0.12<3>
0.49
0.33
0.46
0.12
1.2
1.4
123
1.4
0.58
0.38
20
Average Exceeding
of 5 Literature-based
Plants Treatability
(mg/1) Level
<0.28
0.14
0.11
0.075
0.41
0.40
30.9
0.35
0.15
0.17
4.4
ND
3
3
ND
5
3
5
2
2
ND
5
 (1)  Estimates from Section 8.1, Table 8-11, given as the lower limit of
     treatability.

 (2)  Data from Table 11-16 (average effluent concentration from verification
     sampling).

 (3)  Estimated achievable long-term average concentrations from Table 8-13.

 (4)  Data from Table 8-12 (recently submitted by Olin Corporation) .

ND = No data available on treatability witii sulfide/filter.
                                       185

-------
     The BAT limitations for mercury,  although based on  the
     same   technology,    are   more   stringent   than  BPT
     limitations.    The    estimated   achievable   long-term
     average  concentration  for  mercury from Table 8-13 is
     0.034 mg/1 and  is   obtained  from  recently  submitted
     industry data.  A daily variability factor of 3.1  and a
     30-day  average variability factor of 1.4 are estimated
     from data in  Appendix A and from  the recently submitted
     industry data.

     The 24-hour maximum concentration is:

     (0.34 mg/1) (3.1 )  = 0.11 mg/1

     The maximum 30-day  average concentration is:

     (0.034 mg/1)  (1.4)  = 0.048 mg/1

     The  load  limitations   for   mercury   (kg/kkg)    are
     calculated based  on the unit flow rate of 2.1 m*/kkg,
     thus:

     (0.11 mg/1) (2.1  mVkkg)  (  kg/in a  ) = 0.00023 kg/kkg
                               (1000 mg/1)

     for the 24-hour  maximum  limit.    The  maximum  30-day
     average limit is calculated similarly, i.e.,

     (0.048 mg/1)  (2.1  mVkkg) (  kq/m^  ) = 0.00010 kg/kkg
                               (1000 mg/1)

     Comment  from  3  companies  operating 10 plants stated
     that the plants can achieve these limitations.

     Zinc

     The zinc guidance is based on the estimated  achievable
     long-term  average   concentration of 0.12 mg/1 which is
     obtained from Table 8-13.  A daily  variability  factor
     of  7.6  and  a 30-day average variability factor of 1.5
     are obtained  from the recently submitted industry  data
     on zinc effluent concentrations.

     The 24-hour maximum concentration is:

     (0.12 mg/1) (7.6)  = 0.91 mg/1

     The maximum 30-day  average concentration is:

     (0.12 mg/1) (1.5)  = 0.18 mg/1
c.    Antimony
                           186

-------
                        TABLE 11-15.  EFFLUENT LIMITATIONS
                              Chlorine-Mercury Cell
                            Best Available Technology
                           Wastewater Plow:  2.1
SUBCATEGORY
             CHLORINE MERCURY CELL
Pollutant
Subcategory
Performance
 (mq/1)
Daily
Variability
Factor	
30-day Avg.
Variability
Factor
Concentration
 Basis
 (mq/1)	
Max.
30-day   24-hr,
Avg.	Max.
                                                                 Effluent Limit
Max.
30-day
Avg.
          24-hr
          Max.
Nonconventional
 Pollutants:
Total Residual
 Chlorine*7)

Toxic Pollutants:
               2.3/1.4
             0.90
         1.5
0.0019
          0.0032
Antimony 0.23<2) 7.6/1.5 0.35 1.7
Arsenic 0.15<3> 6.7/1.4 0.21 1.0
Cadmium Q.050<2> 7.6/1.5 0.075 0.38
Chromium 0.044<2> 7.6/1.5 0.066 0.33
Copper 0.20<3> 7.6/1.5 0.30 1-5
Lead 0.10(3) 4.1/1.3 0.13 0.41
Mercury(6J 0.034<3) 3.1/1.4 0.048 0.11
Nickel 0.10<4> 5.7/1.4 0.14 0.57
Silver 0.067(2) 7.6/1.5 0.10 0.51
Thallium 0.17<2> 7.6/1.5 0.26 1.3
Zinc 0.12(3) 7.6/1.5 0.18 0.91
— IS)
__(5)
— (5)
__(5)
-_(5)
— (5)
0.00010
— (5)
— (5)
— (5)
— (5)
— (5)
— (5)
—(5)
-_(5)
—(5)
— (5)
0.00023
— (5)
— (5)
__(5)
~(5)
( 1 ) Long-term average concentration from Appendix A.
(2) Average effluent concentration from verification sampling.
(3) Estimated achievable long-term average concentration from Table 8-13.
(4) Lower limit of literature treatability for sulf ide/filter technology according
to Table 8-11.
(5) No load limits; concentration limits are provided for guidance purposes.
(6) Limits are also applicable to PSNS, and NSPS.
(7) Limits are also applicable to NSPS.

                                          187

-------
I
                  Because no industry data  is  available,  the  antimony
                  guidance is based on the average effluent concentration
                  from the verification sampling.  The value of 0.23 mg/1
                  is  used  as  a  long-term  average.   The  variability
                  factors established for zinc are  applied  to  antimony
                  since  these  would  approximate  the situation where a
                  metal is incidentally controlled.

                  The 24-hour maximum concentration is;

                  (0.23 mg/1) (7.6) = 1.7 mg/1

                  The maximum 30-day average concentration is:

                  (Q.23 mg/1) (1.5) = 0.35 mg/1

                  Arsenic

                  The  arsenic  guidance  is  based  on   the   estimated
                  achievable long-term average concentration of 0.15 mg/1
                  which is obtained from Table 8-13.  A daily variability
                  factor  of  6.7 and a 30-day average variability factor
                  of  1.4  are  obtained  from  the  recently   submitted
                  industrial data on arsenic effluent concentrations.

                  The 24-hour maximum concentration is:

                  (0.15 mg/1) (6.7) = 1.0 mg/1

                  The maximum 30-day average concentration is:

                  (0.15 mg/1) (1.4) = 0.21 mg/1

                  Cadmium

                  Because no industry data  are  available,  the  cadmium
                  guidance is based on the average effluent concentration
                  from  the  verification  sampling.   The value of  0.050
                  mg/1 is used as a long-term average.   The  variability
                  factors  established  for  zinc  are applied to cadmium
                  since these would approximate  the  situation  where  a
                  metal is incidentally controlled.

                  The 24-hour maximum concentration is:

                  (0.050 mg/1) (7.6) = 0.38 mg/1

                  The maximum 30-day average concentration is:

                  (0.050 mg/1) (1.5) = 0.075 mg/1

                  Chromium
                                       188

-------
        TABLE 11-16.  EFFLUENT CONCENTRATIONS OF TOXIC POLLUTANTS
                      FROM VERIFICATION SAMPLING

SUBCATEGORY
Pollutant
Antimony
Arsenic
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Silver
Thallium
Zinc
CHLORINE (MERCURY
CELL)

Plant Effluent Concentrations
(rag/1)
Plant
#299
0.15
0.063
0.073
<0.060
0.038
<0.050
0.029
<0.050
<0.015
0.20
0.10
#747
<0.25
<0.010
0.12
<0.050
<0.025
0.073
0.10
<0.050
<0.015
<0.045
<0.025
#317
<0.25
0.020
<0.025
<0.050
<0.030
0.17
0.19
<0.067
<0.015
<0.25
0.51
#106
<0.45
<0.0050
0.016
<0.010
0.043
0.38
<0. 00050
0.14
0.26
0.26
0.088
#167
<0.065
0.38
0.010
<0.050
<0.025
0.12
0.32
<0.050
<0.015
0.090
<0.025


Treatability *
(mg/1)
Avg.
<0.23
<0.096
<0.050
<0.044
<0.033
<0.16
<0.13
<0.074
<0.067
<0.17
<0.15

(2)
0.050
0.010
(2)
0.050
0.10
0.010
0.10
0.050
(2)
0.020
(1)   Lover limit from literature-based treatability estimates from Section 8.1

(2)   No data available for treatability with sulfide/filter.
                                    189

-------
     Because no industry data are  available,   the  chromium
     guidance is based on the average effluent concentration
     from  the  verification  sampling.    The value of 0.044
     mg/1 is used in place  of  a  long-term  average.   The
     variability factors established for zinc are applied to
     chromium  since  these  would approximate the situation
     where a metal is incidentally controlled.

     The 24-hour maximum concentration is:
     {0.044 mg/1) (7.6) = 0.33 mg/1

     The maximum 30-day average concentration is:

     (0.044 mg/1) (1.5) = 0.066 mg/1

g.    Copper

     The  copper  guidance  is  based   on   the   estimated
     achievable long-term average concentration of 0.20 mg/1
     which is obtained from Table 8-13.  A daily variability
     factor  of  7.6 and a 30-day average variability factor
     of  1.5  are  obtained  from  the  recently   submitted
     industry data on copper ef-fluent concentrations.

     The 24-hour maximum concentration is:

     (0.20 mg/1) (7.6) = 1.5 mg/1

     The maximum 30-day average concentration is:

     (0.20 mg/1) (1.5) = 0.30 mg/1

h.    Lead

     The lead guidance is based on the estimated  achievable
     long-term  average  concentration of 0.10 mg/1 which is
     obtained from Table 8-13.  A daily  variability  factor
     of  4.1  and a 30-day average variability factor of 1.3
     are obtained from the recently submitted industry  data
     on lead effluent concentrations.

     The 24-hour maximum concentration is:

     (0.10 mg/1) (4.1) = 0.41 mg/1

     The maximum 30-day average concentration is:

     (0.10 mg/1) (1.3) = 0.13 mg/1

i.    Nickel
                           190

-------
     The  industrial  wastewater  treatment  system  performance
     data  in   Table  8-12   show values for  nickel  below the
     lower  limit  of   literature  treatability  estimated  in
     Table   8-11.    Because  of  this,  the lower  limit,  0.10
     mg/1,  is  used  in place of   a  long-term  average  as  a
     basis   for  the  nickel  guidance.   A daily  variability
     factor of 5.7  and a 30-day average   variability  factor
     of    1.4   are   obtained  from  the   recently  submitted
     industry  data  on nickel effluent concentrations.

     The  24-hour  maximum concentration is:

     (0.10  mg/1)  (5.7) = 0.57 mg/1

     The  maximum  30-day average concentration is:

     (0.10  mg/1)  (1.4) = 0.14 mg/1

j.    Silver


     Because no industry  data  are  available,  the  silver
     guidance  is  based on the average effluent concentration
     from  the  verification  sampling.    The value of 0.067
     mg/1 is used in place  of  a  long-term  average.   The
     variability  factors established for  zinc are applied to
     silver  since   these  would  approximate  the situation
     where  a metal  is incidentally controlled.

     The  24-hour  maximum concentration is:

     (0.067 mg/1) (7.6) = 0.51  mg/1

     The  maximum  30-day average concentration is:

     (0.067 mg/1) (1.5) = 0.10 mg/1

k.    Thallium:  Because no industry data are available,  the
     thallium  guidance  is  based  on  the average effluent
     concentration   from  the  verification  sampling.   The
     value  of  0.17  mg/1  is  used in place of a long-term
     average.   The  variability factors established for  zinc
     are   applied  to thallium since these would approximate
     the  situation  where a metal is incidentally controlled.

     The  24-hour  maximum concentration is:

     (0.17  mg/1)  (7.6) = 1.3 mg/1

     The  maximum 30-day average concentration is:

     (0.17  mg/1)  (1.5) = 0.26 mg/1
                            191

-------
Basis for BCT Effluent Limitations

While  EPA  has  not  yet  proposed  or  promulgated  a  revised   BCT
methodology  in  response  to  the  American  Paper  Institute  v. EPA
decision mentioned in Section 3, EPA is promulgating  BCT  limitations
for  this  subcategory.   These limits are identical to those for BPT.
EPA is not promulgating any more stringent limitations since  we  have
identified   no  technology  option  which  would  remove  significant
additional amounts of  conventional  pollutants.   The  dechlorination
technology   added   to   BPT  for  BAT  does  not  remove  additional
conventional pollutants.  As BPT  is  the  minimal  level  of  control
required  by  law, no possible application of the BCT cost tests could
result in  BCT  limitations  lower  than  those  promulgated  in  this
regulation.   Accordingly,  there is no need to wait until EPA revises
the BCT methodology before promulgating BCT limitations.

Basis for New Source Performance Standards

For NSPS, the Agency is promulgating limitations equal to BPT for  TSS
and pH and BAT for other pollutants because of the prohibitive cost of
additional technology.  Pollutants to be limited are pH, TSS, mercury,
and total residual chlorine.

Basis for Pretreatment Standards

For pretreatment standards for new sources, the Agency is promulgating
limitations   based   on   BAT  technology  excluding  dechlorination.
Dechlorination is unnecessary for discharges to POTWs because influent
to POTWs is  often  chlorinated.   The  pollutant  to  be  limited  is
mercury.   The PSNS limitations are based on BAT because this provides
better mercury removal than is achieved by a well-operated  POTW  with
secondary  treatment installed and, hence, mercury will pass through a
POTW in the absence of pretreatment.

The Agency is not promulgating PSES for  this  subcategory.   Instead,
the subcategory is excluded from categorical PSES under the provisions
of paragraph 8(b) of the Settlement Agreement because the discharge of
total  toxic metals to POTWs from the two existing sources combined is
below treatable levels and amounts to only 40 pounds per  year.   Both
existing sources have installed and are operating treatment facilities
equivalent to BPT/BAT to control mercury.

Diaphragm Cell Process Industry Profile


General Description

Approximately  65  percent  of  the  U.S. production of chlorine  is by
diaphragm cell plants.  Of 40 known plants in 1978, Section  308  data
are  available  for 19.  Table  11-17 presents a summary profile of the
subcategory.  Table 11-18 presents the status of discharge regulations
prior to promulgation of this regulation for diaphragm  cell  chlorine
plants.
                                    192

-------

General Process Description

A.   Brine System

     As in the mercury cell  process,  the  sodium  chloride  solution
     (brine  or salt dissolved in water) is purified before it is sent
     to the electrolytic cells.   Precipitation  of  major  impurities
     with   sodium   carbonate   and   sodium  hydroxide  followed  by
     clarification generates a brine mud waste which is then sent to a
     lagoon or filtered.  The settled brine is  saturated  further  by
     the addition of salt from caustic evaporators and then is sent to
     the cells.

     The  fundamental  difference  between  diaphragm and mercury cell
     brine systems is that unconverted sodium  chloride  in  diaphragm
     cell  processes  is  carried  with the sodium hydroxide (caustic)
     from the  cell  and  is  then  removed  as  a  solid  in  caustic
     evaporators.   In  mercury cells, the unconverted sodium chloride
     is discharged through the brine system.

B.   Diaphragm Cell

     The treated brine solution is electrolysed in the diaphragm  cell
     to form chlorine, hydrogen, and sodium hydroxide according to the
     reaction:

     2NaCl + 2HZ0 = C12 + 2NaOH + H2

     The   diaphragm   cell   contains  a  porous  asbestos  diaphragm
     separating the anode from the cathode.  Chlorine is liberated  at
     the  anode  and hydrogen and hydroxyl ions (caustic) are produced
     at the cathode.  In the past, the predominant material  used  for
     anodes  was graphite with some cell designs using lead to provide
     an electrical contact and support.  The lead was  joined  to  the
     graphite  anode  by  an  organic  binder.   In recent years, many
     graphite anodes have been replaced  by  stabilized  metal  anodes
     made of titanium with a platinum or ruthenium oxide coating.  (An
     industry association estimate is that approximately 49 percent of
     U.S.  diaphragm  cell  capacity  still involves graphite anodes.)
     The advantages of using metal anodes compared to graphite  anodes
     are  increased  power  efficiency of the  cells, longer anode life
     and  a  reduction  in  potential  pollutant  loads  of  lead  and
     chlorinated organics.

C.   Product Purification

     As with mercury cell plants, chlorine liberated at the anode must
     be cooled and dried to remove moisture and other impurities.  The
     cooling generates a  chlorine  condensate stream  which  can  be
     stripped  to  recover  chlorine  and  then  returned to the brine
     system or discharged.  Drying the chlorine gas is accomplished by
     scrubbing with sulfuric acid.  The  resulting  diluted  acid  can
     subsequently be regenerated, sold, or used for nonbrine system pH
                                   193

-------
         TABLE 11-17.  SUBCAIEGORY PROFILE DATA
 SUBCATEQQRY
CHLORINE (DIAPHRAQ1 CELL)
Total subcategpry capacity rate
Total subcategory production rate
Number of plants  in this subcategory
308 Data on file  for
    With total capacity of
    With total production of
    Representing  capacity
    Representing  production
    Plant production range:
            Minimum
            Maximum
    Average production
    Median  production
    Average capacity utilization
    Plant age range:
            Minimum
            Maximum
    Waste water flow range:
            Minimum
            Maxinum
    Volume  per unit product:
            Minimum
            Maximum
                         8,270,000 kkg/year
                         6,430,000 kkg/year
                               40
                               19
                         6,400,000 kkg/year
                         4,200,000 kkg/year
                               77 percent
                               66. percent

                            14,700 kkg/year
                         1,500,000 kkg/year
                          221,000 kkg/year
                          103,000 kkg/year
                               67 percent

                                4 years
                               74 years

                             1,100 cubic meters/day
                             7,100 cubic meters/day

                                1 cubic meters/kkg
                               23 cubic meters/kkg
 Sources of data are Stanford Research Institute, Directory of Chemical
 Producers, U.S.A.,  1977,  U.S, Department of Commerce, Current Industrial
 Reports, December 1977; Energy ani Environmental Analysis, Inc.; Draft
 Report, "Preliminary Economic Assessment of Effluent Limitations in the
 Inorganic Chemical  Industry*" June,  1978  and "Economic Analysis of Proposed
Revised Effluent Guidelines and Standards  for the Inorganic Chemicals Industry,"
March, 1980.
                                    194

-------
       TABLE 11-18. STATUS OF REGULATIONS - EFFLUENT LIMITATION GUIDELINES
SUBCATEGORY

SUBPART
                      CHLORINE (DIAPHRAGM CELL)

                      F  (40 CFR 415.60, 3/12/74)
                                    STANDARDS
                            BPCTCA              BATEA              NSPS
                        Max ""     Avg.      Max*     Avg.      Max.     Avg.
           Parameters   (kgAkg)  (kg/kkg)  (kq/kkg) (kgAkg)  (kg/kkg) (kg/kkg)
Product
Process
Diaphragm
Cell
Process    TSS
           Pb
                         0.64
                         0.005
0.32
0.0025
No discharge
 of pwwp(3)

No discharge
 of pwwp L£\
0.64
0.32
0.00008  0.00004
    Section 415.63 was remanded and reserved  (41 PR 51601, November 23,  1976).
(1) Max. = Maximum of any one day.
(2) Avg. = Average of daily values for thirty consecutive days.
(3) pwwp = Process wastewater pollutant.
                                   195

-------
     control.    When  the  chlorine  gas  is compressed and liquified,
     noncondensible gases known as tail or sniff gases remain.    These
     are  usually  scrubbed  with  caustic  soda  or lime generating a
     hypochlorite  solution  which  can  be  sold,  used  on-site,   or
     discharged, with or without decomposition or treatment.

     The  sodium  hydroxide  or  caustic from the diaphragm cell has a
     concentration of about 11  percent  NaOH  and  a  sodium  chloride
     content  as  high as 17 percent.   The caustic is evaporated to 50
     percent NaOH by multiple  effect  evaporators.    Sodium  chloride
     remains  as a solid salt which is then separated from the caustic
     and returned to the brine system.  Further  purification  of  the
     caustic  is  necessary  for  some  applications  (such  as  rayon
     production) and extraction or  adsorption  techniques  have  been
     used  to  remove small amounts of impurities.  The caustic can be
     evaporated further if more concentrated  products  are  required.
     The vapor evolved from the last of multiple effect evaporators is
     condensed  in  barometric  condensers  generating contact cooling
     water, or in surface condensers using noncontact cooling water.

     The hydrogen gas generated in the process can be vented or cooled
     by refrigeration to remove water vapor before sale or  use  as  a
     fuel.

     Figure  11-8  is  a  general  flow diagram for the manufacture of
     chlorine by the diaphragm cell process.

Water Use and Wastewater Sources

Water Use

Water use at diaphragm cell plants is similar to that at mercury  cell
plants  with  one  exception.  Common uses include noncontact cooling,
tail gas scrubbers, cell wash,  equipment maintenance,  floor  washings
and filter backwashing.  The exception at diaphragm cell plants is the
use of water for barometric condensers in the evaporation of caustic.

Waste Sources

A.   Brine Mud

     As  with  mercury  cells,  this  is  the  waste  produced  during
     purification  of brine before it is introduced into the cells for
     electrolysis.   It  consists  of  precipitated   hydroxides   and
     carbonates  of  calcium,  magnesium, iron, and other metals.  The
     mud can be a major source of solid waste depending on the  purity
     of  the  raw salt used.  At diaphragm cell plants, brine muds are
     filtered or settled in lagoons.  The solids  are  landfilled  and
     the  filtrate  or overflow is discharged or recycled to the brine
     system.
                                   196

-------
     Brine mud is the major source of solid waste at chlorine  plants,
     and discharges range from 0,04 to 1.5 cubic meters per metric ton
     (mVkkg), with a solids content of from 2 to 20 percent.

B.   Cell Room Wastes

     These wastes include leaks, spills,  area washdown and  cell  wash
     waters.   At  diaphragm cell plants, cell wash waters are heavily
     laden with asbestos and are  therefore  settled  and/or  filtered
     before  chemical  treatment  or  discharge.  At some plants using
     graphite anodes in the cells, the cell room wastes  also  contain
     lead.  Data from diaphragm cell plants indicate a waste flow from
     0.02 to 1.2 m3/kkg from cell room operations.

C.   Chlorine Cooling Condensate

     Condensate from the indirect cooling of cell gas is  contaminated
     with  chlorine.   The chlorine is removed  (stripped) or recovered
     from the stream before discharge or  recycle.   Condensate  flows
     from three plants range from 0.16 to 0.9 mVkkg.

D.   Spent Sulfuric Acid

     Concentrated sulfuric acid is used to dry  chlorine gas after  the
     first  stage  of  cooling.  Once diluted to 50 to 70 percent, the
     spent acid can be regenerated, sold, or used for nonbrine  system
     pH control.

E.   Tail Gas Scrubber Liquid

     The  uncondensed chlorine   gas  from  the   liquefaction  stage   is
     scrubbed   with   sodium   or  calcium  hydroxide  producing  the
     corresponding hypochlorite.  The  hypochlorite  can  be  used   in
     other  processes, sold, decomposed, or discharged.  The amount  of
     tail gas scrubber water generated at diaphragm cell plants ranges
     from 0.1 to 0.29 mVkkg.

F.   Filter Backwashes

     Backwashing of filters used to treat brine before it is  sent   to
     the  cells at one graphite anode diaphragm cell plant generated a
     wastewater flow of 0.45 mVkkg.  Backwashing of filters  used   to
     clarify  caustic product  at the same plant resulted in  an average
     flow of  5.4  mVkkg.   At some  diaphragm  cell  plants,  these
     wastewaters are partially recycled  to process.

     The  relatively high flow  of caustic filter backwash is  due to the
     need  to  remove  sodium   sulfate,  an  impurity  in the caustic.
     Sulfate  ions, if allowed  to accumulate  in  the  brine   system   at
     graphite anode plants will interfere with  cell performance.
                                    197

-------
                     BRINE

SULFATE BM>
PURGE



NONCONTACT
COOLING -P>
WATER ^.

PURIFICATION
SYSTEM
*
DIAPHRAGM
CELL

DRINL HUUS ^fM jtgyaj. HOHCOHTACT COOLING HATER
t »
— 	 HYDRC
	 CHLORINE 	
1
lit TO lit SODIUM
HYDROXIDE
SOLUTION
1 RECYCLE A






HATER
— BAROMETRIC
^ CONDENSER
COOLER OR USE
t
COOLING
HATER
COOLER •*• CHLORINATED HATER COHDENSATE
t * 1


COOLING
TOHER**
M
M) f
00 f
BLOHDOHN
TO HASTE
HATER
*
LEAKS.
SPILLS
HASIIDOHN
ETC.
r~
t
SALT
REMOVAL
1
SODIUM
HYDROXIDE
SOLUTION
1
FILTER
EUliURIC -p»
ACID


*) ILIME) f 4


uint-KEaiMNi i •- SOLUTION " SALES. OR HAS
f


AT ION
SYSTEM * 	
* * *
PACKAGING
HOHCOHTACT
COOLING
TO USB
OR SALES

LIQUEFIER f
1 DIAPHRAGM
LIQUID HASHING
CHLORINE n*imsf inw
HnTER
1 t
TO BALES U»CT»
USED SOME PIAHTS ONLY

 DEFENDS UPON PLANT DESIGN
      Figure 11-8.   General  process flow diagram for production of chlorine/caustic by
                       diaphragm cells.

-------
G.   Hydrogen Condensate

     Cooling of hydrogen gas for use or  sale  produces  a  condensate
     stream  which  can be discharged.  Although no data are available
     on the volume of this flow, it is small.

H.   Barometric Condenser Wastewater

     When  vapors  from  caustic  evaporators  are  contact-cooled,  a
     significant amount of wastewater can be generated.  Flows of from
   ^,,90^	Jha ^300  mVkkg  have^ been  reported  at  facilities  where
     barometric ^'coriHe'rtser"'"^"i'water   is   "once   through"   and   not
     recirculated.    Recirculation   of  barometric  condenser  water
     requires a cooling step and a blowdown discharge.   A  number  of
     facilities  are accomplishing this with a corresponding reduction
     in water use.  The necessary blowdown of recirculating barometric
     condenser wastewater at two plants ranges from  a  flow  of  0.82
     mVkkg to 0.89 mVkkg.

I.   Summary of Wastewater Flows

     Table 11-19 summarizes unit wastewater  flow  data  available  by
     specific sources.  A separate list of flows at one graphite anode
     plant is presented to compare wastewater generation between metal
     anode and graphite anode plants.

Descriptions of Specific Plants

The  following  description  of plants includes those plants that were
sampled during the screening and verification program.  The discussion
primarily covers plant practices in wastewater control and  treatment.
Plants  were  selected for screening and verification sampling because
they were representative of the industry in that they included a  wide
range of sizes and variation in process detail.

Screening

At  Plant  1014,  visited  during  the screening program, the chlorine
condensate is stripped with steam  to  remove  and  recover  chlorine.
Brine  precipitates  (muds) are  land disposed, while the spent sulfuric
acid and the scrubber solutions are used at an  adjacent  plant.   The
condensate  from  the  hydrogen  cooler  is used as makeup water for a
cooling  tower  system,  and  the  condensate  from  the   evaporative
concentration  of  sodium hydroxide is used to dissolve salt reclaimed
from the concentration process.  The  cell  washings  are  sent  to  a
collection pond where asbestos  and other suspended solids are removed.
In  Figure  11-9,  the  general  process flow sheet is presented.  The
waste streams sampled and their waste loadings are presented in  Table
11-20.   The  only  process  waste  stream  discharged is once-through
barometric condenser cooling water.
                                  199

-------
Verification

Four plants were visited and their waste streams  sampled  during  the
verification  program.  The results of analysis of the wastewaters are
presented in Table 11-20.

At Plant #261, the cathode wash water is passed through a  filter  and
the  asbestos  drummed  and disposed of in an off-site landfill, while
the filrate goes to  the  sewer.   Brine  purification  muds  at  this
facility  are  utilized for their alkalinity on-site and then they are
settled prior to discharge of the supernatant.  Spent sulfuric acid is
used for neutralization of wastewaters.  Dechlorination of the  drying
acid  by reaction with sodium bisulfite is planned in the near future.
Figure 11-10 shows the process flow diagram and sampling points.

Plant #738 has two production lines, 738A and 738B,  that  are  almost
identical.   At  the new plant (738B) the NaOH is not concentrated nor
is the waste from the chlorine disposal system scrubbed.  In addition,
the inert gases  from  the  liquefaction  step  are  put  through  the
chlorine  disposal  system.   The  process  flow  sheets  are shown in
Figures 11-11 and 11-12.

Plant #736 has installed demisters to control the vapors evolved  from
the  last  stage of the evaporator during the concentration of caustic
soda.  In this treatment, the steam evolved from the concentration  of
cell  liquors  passes  through  metal wool filters to reduce entrained
solids.  The cell room washings are sent to a settling chamber and the
settled asbestos is  sent  to  a  landfill.   The  other  wastewaters,
consisting  of  caustic  evaporator  washings  and  wastes  from  salt
separation, brine  purification  operations,  and  caustic  filtration
backwash  waters,  are combined and sent to one of two settling ponds.
Skimming devices on the settling ponds remove any oil that  separates,
while  the  settled solids in the ponds are dredged and disposed of in
an abandoned brine well.  Figure 11-13 shows the process flow  diagram
and sampling points.

Plant  #967  uses  graphite  anodes  in its diaphragm cells.  The cell
washings at this plant are  sent  to  an  asbestos  pond  that  has   a
continuous  cover  of  water.   Periodically,  the  settled solids are
removed, sealed in drums and disposed of in a landfill.  The  overflow
from  the  pond is treated with soda ash to precipitate lead, and then
filtered.  Sulfuric acid is used to bring the pH down to the  6  to   9
range.   Figure  11-14  is  a general process flow diagram for Plant  #
967.

Descriptions of Plants Not Sampled

At Plant #999, brine mud and other streams with high suspended  solids
are collected, and filtered with leaf filters.  The cake is disposed of
in a landfill and the filtrate returned to the brine system.

At  Plant 1326, wastewater from the diaphragm cell process is combined
with other process wastewaters.  The combined wastewater  is  sent  to
                                   200

-------
  TABIE 11-19.  WASTE WATER FLOWS AT DIAPHRAC31 rftT.T, CHIORINE PLANTS
Stream Description
              Flow (m-VWcg)
    Plants with              Plant with
   Metal Anodes            Graphite Anodes
 min.  avg.  max.
Cell room wastes
   and cell wash
Chlorine Condensate

Spent Sulfuric Acid
Tail Gas Scrubber
Caustic Filter Wash
Brine Filter Backwash
Caustic Cooling Blowdown
Brine Mud
0.020  0.38  0.67

0.16   0.49  0.90

       0.010
0.10   0.17  0.29
        NA
        NA
0.82   0.86  0.89
0.040  0.42  1.5
1.2

0.78

 NA
 —
0.
——
5.4

0.45

 NA

 NA
NA:  Not Available
                                     201

-------
two  settling tanks in series.  In one of the settling tanks, skimmers
have been installed to remove oil and the overflow from the second  is
filtered before discharge.

At Plant #589, the brine mud from the clarifier underflow is sent to a
brine  mud  settling  pond.   The  overflow, which is mostly brine, is
returned to the process.   The  cell  room  washings  are  sent  to  a
settling pit and the settled asbestos fibers are removed by the use of
a  vacuum truck, and disposed of in a landfill.  The chlorine from the
cells is  contact-cooled  with  the  tail  gas  scrubber  water.   The
resulting  wastewater  is  steam stripped for chlorine recovery before
discharge.

At Plant #741, chlorine, caustic soda,  and  potassium  hydroxide  are
produced  using  both  mercury  and  diaphragm cells.  Mercury-bearing
effluent at this facility is treated by sulfide  precipitation.   Tail
gas  absorption  wastes  are  treated  by catalytic decomposition by a
process which consists of scrubbing with  caustic  soda  solution  and
treating the resulting hydrochlorite solution with nickel chloride and
iron   chloride.    Consumption   of   iron  and  nickel  chloride  is
approximately equal and consists of 0.01 kilogram per  metric  ton  of
chlorine  produced.   The  catalytic decomposition proceeds relatively
slowly,  and  wastes  are  retained  in  the   treatment   tanks   for
approximately  three  days,  after  which time no residual chlorine is
reported to be present (3).

Toxic Pollutant Concentrations

A.   Analytical Data Base

     Section 5  of  this  report  describes  the  methodology  of  the
     screening  and  verification  sampling  program.  In the chlorine
     diaphragm cell industry, a total of  15  days  of  sampling  were
     conducted  at  Plants #014, 261, 738, 967, and 736.  Thirty-seven
     different sampling points  were  involved  covering  various  raw
     waste  streams  and  the  treated effluents at these plants.  The
     evaluation of the toxic metal content of  these  process  related
     waste  streams  was  based  on  975  analytical data points.  The
     sampling for toxic organic pollutants  at  Plants  #014  and  967
     generated  2300  analytical  data  points.  Analysis of waste for
     asbestos generated an additional 13 data points.

B.   Asbestos

     Asbestos, used as a  diaphragm  sepatating  the  cell  anode  and
     cathode,  is  the  major  toxic  pollutant  consistently found in
     process  wastewater  from  diaphragm  cell  plants.   It   occurs
     primarily  in  wastes resulting from activities such as cell room
     washdown and cell repair and cleaning.

     Table 11-21 presents the results of  asbestos  determinations  of
     supply  water  and  wastewaters  at  three diaphragm cell plants.
                                  202

-------
                                                VEOTGAS
N)
O
BRIMK
BRINE MUD
,annoric
t"

i
REIUJM
1
KM
' B/M-Ofl
OOH
1
NmH 	 ^"l BULGftS^
9»t H-Ki. * mMmffl
* 1
9
C12

i 1 1
WfffB OOOI.IHG ^ i REUSE 751 HfO^ COOLIHO WIHt
ItoCtt 14
1
II
AT1OH

HSOH Hartl ttBIOH

1- | — j— |
f cnoi.iwj 501 71t ..^,,1
HUTMEmiC HJ» NKJM MflttnflUC (aim
CXMJEH3BI „ ,j.™«. CCKDOBER
•..•Mm. ~r. TO WftMB u.«*« «n
tffvniR ID d WTEK TU
MNS1C. I i MSIC




T T ^-^
1 •• Sanplln? polnto.
TJT
                                                                                                    PWHT
                 Figure 11-9.    General process flow diagram at plant #014 showing the sampling points.
                                 Chlorine/caustic (diaphragm cell) manufacture.

-------
TABLE 11-20.  POLLUTANT OKENTRATIONS AND LOADS AT
              SCREENING AND VERIFICATION PLANTS

SUBCATEGORY CHLORINE DIAPHRAGM rrrr.T.
Plant &
Stream Stream TSS
No. Description (mg/1) (kg/kkg)
#014




#261






3
4
5
6

1
2
3
4
5
r?38Al





2
3
4
5

#738B6










7
8
9
10

11
12
13

14

C12 condensate
Cell wash
Brine mud
Bar. condenser

Brine mud
Cell wash
Asbestos filtrate
Filter cake
Bar. condenser
Cell room waste
Asbestos wash
Hypo scrubber
Cl2 cooling water
Caustic cooling
tower
Cell room waste
Asbestos wash
Hypo scrubber
Cl2 cooling water
Caustic cooling
tower
Chlorate sump
Plant effluent(B)
Final effluent
(Total)
Brine mud

.20
1600
NA
7.0

NA
4800
9.0
NA
6.0
27
57
290
35
48

95
72
160
20
4.7

32.
63
58

270

1.
2.




1.



1.
7.
2.
2.
4.

4.
a.
1.
1.
3.

7.
5.




4 x
4 x
NA
3.6

NA
8 x
NA
NA
NA
4 x
0 X
7 x
2 x
3 x

5 x
3 x
4 x
7 x
8 x

0 x
7 x
NA

NA

10
10




10



10
10
10
10
10

10
10
10

-3
-2




-1



-3
~3
-2
-1
-2

-3
~3
-2
10-2
10

10
10



-3

1-3
-1



Lead
(mg/1) (kg/kkg)

0.
0.
0.
0.

0.
2.
0.
42
< 0.
0.
0.
0.

0055
26
72
0050

36
0
075

010
077
031
18
0.28
0.

0.
0.
0.
0.
< 0.

< 0.
51

067
13
20
20
010

010
0.12
0.

0.
078

10

5.
3.
1.
1.

3.
7.



3.
3.
1.
1.
4.

3.
1.
1.
1.
< 8.

< 2.
1.




OxHT6
9xlO"6
3xlQ-5
5xlO~3

OxlO"4
6xlO~5
NA
NA
NA
9X10"6
8X10"6
7xlO"5
3X10"4
SxlO-4

2xlQ-6
5xlO~5
7xlO~5
7xlO"5
2xlO"6

3X10"6
lxlO~3
NA

NA
                                                 (Continued)
                             204

-------
                               TABLE 11-20 (continued)

Plant &
Stream No
#736
1
2
3
4
5
6
7
#967l
2
3
4
5
6
7
Stream
Description
Cell wash
Cell room drain
Brine mud
50% Bar. condenser
70% Bar. condenser
95% Bar. condenser
Chlorine condensate
Cell bldg wastes
Lead pond effluent
Caustic backwash
Brine backwash
Cell wash
Condensate and I^SO^
Scrubber waste
TSS
(mg/1) (kg/kkg)
934
284
20,000
32
21
90.3
2.4
1000
54
160
13,000
310
1 1100
270
6.0 x 10~2
4.6 x 10~3
33
NA
NA
NA
3.9 x 10"4
1.8 x 10"1
3.0 x 10~2
8.6 x 10'1
5.8
5.6 x 10"2
8.7 x 10'1
1.2 x 10 ~2
Lead
(mg/1) (kg/kkg)
0.014
0.17
0.019
0.010
0.010
0.010
0.010
680
29
0.32
0.52
48
0.92
0.67
9.1X10"6
2.8xlO~6
3.1xlO~5
NA
NA
1.6xlO"6
1.2X10"1
1.6xlO~2
1.7xlO~3
2.3xlO~4
8.6xlO~3
7.3xlO"4
2.9xlO~5
NA: Not Available
                                     205

-------
 fOOIUH
CARBONATE
                   (UOH
SALT
                                               HYDROGEN TO DOILCR
                                              STEAM
                ADJACENT PLANT
HI1CAS



CURI-
FLOCCULATORS
f
1RINE HUftS


SAND
FILURS
A
BACK
i i WASH

SETTLER
                                                                                                                                FILTRATE TO
                                                                                                                               PROCESS SEWER
                              SLUDGE TO LANDFILL

                                      COOLIHG WATER
                                                                HVPOCHLORITE
                                                                    TOWER
                                                                    1 _ fc.
                                                                                'TO VENT
                                                                          SOLUTION TO
                                                                          ADJACENT PLANT
                                                         ... STORAGE TAMK
                                               CHLORINE    ANfi RAILROAD CAR
                                              TO STORAGE      WASHDOWN
r
COOLING
WATER
STEAH
\ <
IS
i
73t CAUSTIC
EVAPORATOR




1
    PUXGE roH DISPOSAL
       IV COH1UCT
                                                                                               NaQH
                                    LEGEND

                                  SAMPLING POINTS
                                                                                 BOILER
                                                                               FEED WATER

                                                                               TO SUKE LIME
                                                                              (FO* LIKE PLANT)
                                                                                                                                       PROCESS SEWER
                                                                       TO CJECTOH
                                                                        WATER


                                                                       PROCESS
                                                                        SEWER
              Figure  11-10.   General  process flow  diagram  at  Plant #261  showing  the  sampling points.
                                 Chlorine/Caustic  (Diaphragm Cell)  manufacture

-------
 7«
         BRINE
                                                                                           'Cl,
                                                                               MtSIE WTER
                                                                             OMC1 + MaOIlt
                                                           (Ftolnt 112 - uaate MMp (ccnblnatlon of «11 waste*).I
Figure 11-11.   General  process  flowsheet  at plant 1738-A  showing the sampling points.
                 Chlorine/caustic (diaphragm cell)  manufacture.

-------
T
OQOUH3 H,O
**«*•*— «MM \
. \
'/I
BRINE ~* 	 •"
WRinCWKM BKOK Q^
ro NO son
O'» , 1
f {
BHKC * CB
HZ> *,
1 '


HJ) ***
I — • 	 ^ NaCl RBCOUBm

fj
OQOUMQ H,0 ' " '



Maiio
1
NU7K
I na »
""^ i (C^I^ M«»l -^*BSS«»l
CCWfCSSlQl H^O H.60, ^ ^^
1 1 II 4 	
^J ^ CCQUHG MO '±2^ WRtfKKOM ^2^ i i^ia'AM-IlN
OflDW OMWS3KM
i, ^i *
P " Wyl» Hftsrc ware
| H^ HVDHOCMWCNS
U. ASBESTOB *!J3?
5M MUJH H2°
»3iiE!ntO 1 A
nnwOTn 1 /^V\ ^ WttnY
KM) 1 ^*^

'" m&at HMEH f
"a30* in

e~»-« JttWAIJZA'ntMl— J "* I" 	 — ^4-
»JH*UUJ 1 ™ MUBWHI ( ^y
IU
H»SHi WMEH "J |wa;|u
Q— — •
Figure 11-12.  General process flow diagram at plant #738-B showing the sampling
               points.  Chlorine/caustic (diaphragm cell)  manufacture.

-------
N)
o
TOQQUHJwaea
i V,
DtSKUKU | ^\;J
HMHCMUM f
^•MICAU
f I
I'll II 11 UlMtMKM
MICK* h*H OMUFIBII hM HUB* r*H CH.L
1 1
1 1 '•«$
^^ , . mit* 11MXMGH f
T™ " TO UU* ICU, (Mt
BRD1B H» TO WfliOW. DOH
0»- Mil. ' «•"«
BMCH
p senuHQ
1 TNK
1 1
ancta ASUKSTDO
UOUt 1O SOLID
to Biwot Wvsre
DISPOSAL
CIO
KtWXtlt TO ATMOSHEI*
OftCWTWEUSB
I MjSOj
fc COftACT
* CCOIA
1 1 1
•M Dram M (nan !*•
t_J 	 I—I «"1W
~ """P'.i-Qit
9I»M T
CCHTAJT IMEfl
TOWVal «»»
+



" tVHOt*
comXcr
wnot
[eatjeotn
HIST -*J BKKMaMC 1
EUMIIhim 1 CQtWBOd I

utiNQ 1 HTKIHJB sxa
KSCKl M— ICGOCHPOCfn
*• *"
OMftm \ 	
udtc sitaaet
lasaccM.
••H nMtut r*-i — ocwSiw
cw. ~
NaCH LICUOH *""
wonnt mxuar t 1 i
fc fHOK
MG(

UO^TIH. U
UO1ID
auawm
TO erowae
7T »u UWA
I*
OKH1CMA
»0f J
M»» -|
1
8WT HI
utracium
HM-wna
nunwrtcH 1 	 ^,
* '
comer WMCS
loeenuK) KM>
a 4 • ' " 1M
KM """ I OMCBfl

^_
WlMft
                                           «u aUeM* Mniited.
                                                              »
                                                            oWtterwreR  ™.-«.
                                                           TO EKITLIW ram ^J2J
                                                                     «J*K3
                                                                                 ; lAnukt HUBM

                                                                                 1O BETIUJC KH>
                   Figure 11-13.  General process flow diagram at Plant #736  showing the sampling points,

                                   Chlorine/Caustic  (Diaphragm Cell)  manufacture

-------
to
M
O
SOCK*


HIKER



uernui
                Figure  11-14.  General process  flew diagram at Plant #967  shewing the sanpling points.
                               Chlorine/Caustic (Diaphragm Cell) manufacture

-------
     Results are expressed as total fibers per liter (in millions)
     well as Chrisotile and Amphibole fibers per liter.
as
     There is no standardized 304(h) analytical method for asbestos in
     water  and  because  of  this,   EPA  is excluding limitations for
     asbestos from these regulations and  deferring  regulation  to  a
     later date.

C.    Toxic Metals

     Table 11-22 presents maximum daily concentrations of toxic metals
     found in raw waste samples during the screening and  verification
     of   diaphragm  cell  chlorine  plants.   Maximum  concentrations
     observed at one graphite anode plant  are  presented  separately.
     It  is clear that except for lead, toxic metals concentrations at
     the graphite anode plant are essentially no higher  than  at  the
     metal anode plants.

     Because several waste streams usually contribute to the total raw
     waste  at  chlorine  plants, a calculation was often necessary to
     determine the pollutant concentrations that would exist when  the
     streams  were  mixed  prior  to  treatment.   An  example of this
     calculation  is  the  "mixing"  of  the  following   hypothetical
     streams:

     Stream A:  100 gallons per minute, 15 mg/1 of pollutant

     Stream B:  10 gallons per minute, 60 mg/1 of pollutant

     The weighted average for the mixed streams is given by:

     (Flow A x Concentration A) + (Flow B x Concentration B?
                         (Flow A + Flow B)

           = concentration of mixed streams

     Substituting numerical values gives:

     (100 qpm)  (15 mq/I) * (10 qpm) 60 mg/1) = 19 mg/1
                             (110 gpm)

     This  method  was  used  to calculate raw waste concentrations of
     pollutants as presented in  Table  11-22.   Barometric  condenser
     wastewater  when  "once  through" was not included because of the
     high dilution effect of these  large flows.  Brine mud flows  were
     also not included.

     The  daily  raw waste loads were calculated from the waste stream
     flow rates measured or estimated at the time of sampling and  the
     measured pollutant concentration.

     That is,
                                   211

-------
Daily loading (as kg of pollutant per day) = (C) (Q)
                                             1000

where:

C  is  the  concentration  of the pollutant expressed in units of
mg/1 (Note:  kg/m3 = 1000 mg/1) and

Q is the waste stream flow rate expressed in units of m3/day (m3,
a cubic meter, is equal to 264,2 U.S. gallons).

Similarly, the unit loadings were calculated  from  the  reported
chlorine  production  rate,  the  waste stream flow rate, and the
measured pollutant concentration.
Unit loading (as kg of pollutant per kkg =
of chlorine)
(C)  (Q).
1000P
where C and Q are as described  above,  and  P  is  the  chlorine
production  rate expressed in units of kkg/day (kkg is 1000 kg, a
metric ton, which is equal to 2205 Ibs).

The minimum, average and maximum values were calculated based  on
data  from  those plants where the particular pollutant was found
at a detectable concentration.

In Table 11-23, the toxic pollutant raw waste data are  presented
as  the  average  daily  concentrations  (based  on three 24-hour
samples) and the unit loadings found at  the  individual  plants.
Beryllium,  selenium,  and thallium are not included in the table
because average concentrations were below detectable limits.

In Table  11-24,  plant  average  daily  and  unit  loadings  are
presented  as  minimum, average, and maximum values based on data
presented in Table 11-23  for  metal  anode  plants  only.   (The
graphite   anode  plant  is  considered  separately  due  to  its
particular waste source characteristics.)

Based on the average waste loads generated per unit of product at
metal  anode  plants  and  one  graph i te  anode  plant,  and  the
estimated  total  subcategory  production,  the  estimated  total
pollutant raw waste loads generated each year by this subcategory
are as follows:
                              212

-------
  TABLE 11-21.   RESULTS OF ASBESTOS SAMPLING AT DIAPHRAGM t-FT.T. PLANTS
Plant Stream
1 261 Supply
Cell Wash
Filtered Discharge
Barometric
Condenser
# 736 Supply
Cell Wash
Cell Rxm waste
Barometric
Condenser
Barometric
Condenser
Barometric
Condenser
# 967 Supply
Cell Waste
Pond Effluent
Caustic Wash
Brine Filter
Backwash
Cathode Wash Waste •
Condensate & Spent
Acid
Neutralizer Waste
Total Asbestos
Fibers (MFL)*
8.0
2.1 X 108
1.6 X 103
0.40
0.70
2.0 X 107
2.9 X 102
1.8
5.3
1.4 X 102
9.7 X 102
2.4 X 104
2.4 X 103
7.8 X 103
8.0 X 102
3.2 X 105
2.7 X 102
2.1 X 103
Chrisotile
MFL
7.5
2.1 X 108
1.6 X 103
0.40
0.70
2.0 X 107
2.8 X 102
0
5.3
1.4 X 102
9.7 X 102
2.4 X 104
2.4 X 103
7.8 X 103
6.2 X 102
3.2 X 105
1.8 X 102
2.1 X 103
Atnphibole
MFL
0.40
0
0
0
0
0
8.0
1.8
0
0
0
8.0 X 102
0
0
1.8 X 102
0
8.9 X 10
0
*Million fibers per liter
                                   213

-------
TftBLE 11-22.  MAXIMUM RAW WASTE COTCENTRATIONS OF TOXIC METALS OBSERVED AT
              DIAPHRAGM CFT.T. CHLORINE PLANTS (mg/1)

SUBCATEGORY
Toxic
Metal
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Selenium
Silver
Thallium
Zinc
CHLORINE DIAPHRAQ4 Cftf.!,
Plants with
Metal Anodes
<0.25
0.17
<0.014
0.037
7.4
17
2.0
<0.0030
22
<0.020
0.018
<0.25
3.0

Plant with
Graphite Anode
<0.065
0.59
<0.0010
0.017
<0.048
0.27
44
0.0040
0.070
<0.030
<0.016
<0.050
0.25
                                     214

-------
                       Raw Waste Load
          Pollutant

          Antimony
          Arsenic
          Cadmium
          Chromium
          Copper
          Lead
          Mercury
          Nickel
          Silver
          Zinc
kg/year

    483
  6,300
     41
  3, 100
  4,400
470,000
     48
  3,600
      5
  5,100
     Because cell room wastes including cell or cathode  wash  wastes,
     leaks,  spills  and  washdown  are  usually treated separately at
     diaphragm cell plants and because other process  wastes  such  as
     filter backwashes, condensates and caustic evaporation wastes are
     usually  discharged  after  settling,  these two waste mixes were
     evaluated separately.  Table 11-25  presents  average  raw  waste
     concentrations  and  loads  of  toxic  metals  found in cell room
     wastes at the six diaphragm cell  plants  sampled.   Table  11-26
     presents  the  similar  data  from  the sampling of other process
     wastes at these plants.

D.   Toxic Organic Pollutants

     The use of graphite anodes at  chlorine  plants  results  in  the
     generation   of  a  variety  of  simple  chlorinated  hydrocarbon
     compounds as a result of the attack of chlorine  on  the  anodes.
     These compounds are carried out of the cell with the chlorine and
     find  their  way  into  the various- waste streams which originate
     from the chlorine cooling, drying, compression, and  liquefaction
     steps.

     Table  11-27  presents  the  toxic organics that were observed in
     measurable concentrations in the raw wastes at Plant  #967.   The
     concentrations  presented  in  the  table  were  calculated  as a
     mixture of all raw waste streams weighted  on  a  flow  basis  as
     previously described.

     Table  11-28  presents  the  concentrations  of toxic organics by
     individual raw waste streams at Plant #967.  It is clear from the
     table that the highest concentrations of organics occur in wastes
     from chlorine treatment (condensate, drying  acid  and  tail  gas
     scrubber  water)  and  they  account  for 83 percent of the total
     organic waste load.
                                    215

-------
    TABLE  11-23. TOXIC  METAL CONCENTRATIONS AND LOADS AT SCREENING AND
VERIFICATION PLANTS
                                               1
                                              (mg/1)




SUBCATEGORY
POLLUTANT
Antimony
Arsenic
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Silver
Zinc
014
*
*
*
*
0.0020
0.0000018
0.019
0.000017
0.015
0.000014
0.0060
0.0000045
0.0020
0.0000018
0.90
0.00081
*
*
*




(kg/kkg)





CHLORINE DIAPHRAGM CELL
261
*
0.17
0.0000064
0.037
0.0000014
1.9
0.000071
17
0.00064
2,0
0.000075
*
*
22
0,00081
0.018
0.00000070
1.5
0.000054
Plant #
738A
*
*
*
*
*
0.52
0.0046
0.045
0.00039
0.082
0.00060
*
*
0.21
0.0018
*
*
0.29
0.0021
738B
*
*
0.011
0.000021
*
*
0.066
0.0012
0.12
0.00023
0.11
0.000021
*
*
0.067
0.00013
*
0.093
0.00018
736
0.010
0.0000033
0.057
0.000014
0.025
0.0000061
0.18
0.000044
0.43
0.00011
0.016
0.0000039
0.0030
0.0000070
0.22
0.000054
*
*
3.0
0.00074
967**
0.011
0.00015
0.30
0.0021
*
*
0.0040
0.000032
0.16
0.0011
$/*••*
0.0020
0.000014
0.68
0.00049
*
0.19
0.0014
Below measurable concentrations.
Graphite Anode plant.
Based on one 72-hour composite  or the average of three 24-hour composites.
                                   216

-------
                            TABLE 11-24.    S»«RR3f OF RAW WASTE LOADINGS AT SCREENING AND VERIFICATION JETAI* ANOCE PLANTS
M
SUBCATBQORY
CHLORINE DIAPHRAGM CELL
Loading
(kgAkg)
Pollutant
Antinony
Arsenic
Cadmium
Chromium
Copper
Lead
Ifercury
Nickel
Silver
Zinc
min.
0.
0.
0.
0.
0.
0.
0.
0.
0.
0.
00077
0019
00041
0042
0035
00090
00016
0066
00021
016
avg.
0.
0.
0.
0.
0.
0.
0.
0.
0.
2.
00077
0084
00076
59
12
094
00030
31
00021
1
max.
0
0
0
2
0
0
0
1
0
8
.00077
.020
.0014
.8
.19
.37
.00044
.1
.00021
.0
Unit Loading
(kgAkg)
min.
0
0
0
0
0
0
0
0
0
0
.0000033
.0000064
.0000014
.000017
.000014
.0000039
.0000007
.000010
.0000017
.000054
avg.
0.0000033
0.000017
0.0000032
0.00096
0.00020
0.00016
0.0000012
0.00057
0.0000007
0.00078
*Ni
I
max. (01
0.0000033
0.000030
0.0000061
0.0046
0.00064
0.00060
0.0000018
0.0018
0.0000007
0.0021
nrber of
Plants
reraged
it of 5)
1
3
3
5
5
5
2
5
1
4
                           *  Cnly those plants where the pollutant was observed at measurable concentrations.

-------
Pollution Abatement Options

Toxic Pollutants of Concern

Lead occurs in high concentrations in the  cell  room  wastewaters  of
chlorine  plants  using  lead  anchored  graphite anodes.  Other toxic
metals often found in significant  concentrations  at  diaphragm  cell
plants  include  arsenic, cadmium, chromium/ copper, nickel, and zinc.
Antimony, mercury, and silver were also detected but at concentrations
that are not treatable.  These metals are not considered further.  The
sources of these metals may be raw material  impurities  or  corrosion
products  from  the  reaction between chlorine or acid and the process
equipment materials of construction.

Toxic organic compounds also occur in wastewaters from graphite  anode
plants  because of the attack of chlorine on the anode material.  They
appear primarily in waste streams associated with the purification  of
chlorine.

Asbestos  occurs in all wastewaters from diaphragm cell plants, and in
large quantities in cell room wastewaters when cells are  cleaned  and
repaired.

Prevailing Control and Treatment Practices

Specific  control and treatment practices at ten plants were described
above.  The prevailing practices  at  diaphragm  cell  plants  are  to
control  asbestos wastes by settling or filtering cell wash waters and
to neutralize  and  settle  all  wastewaters  before  discharge.   The
recycle  or  reuse of waste streams is practiced to varying degrees in
the industry depending on plant-specific factors such as raw  material
quality and type of anodes used.

Plants  using  lead anchored graphite anodes are treating lead-bearing
wastes by chemical precipitation and settling and/or filtration before
discharge.

The control of  toxic  organic  compounds  in  the  waste  streams  at
graphite  anode  plants  also  varies  in the industry.  At Plant #967
where the end use of the chlorine is  captive,  involving  its  direct
application  to  the manufacture of a chlorinated organic product, the
bulk of chlorinated organic impurities are not removed.

At Plant #195, where a more purified product is required, the organics
are accumulated  in  the  reboiler  of  the  chlorine  scrubber.   The
residues  are  treated  batchwise  for  separation and recovery of the
organic phase materials which are  then  sold  as  feedstock  for  the
manufacture of related products.  Prior to discharge the aqueous phase
is  vacuum  stripped  to  remove  additional organics and chlorine for
recycle.  Normally, one batch of organics is treated per week.   After
separating  each  batch of organics and stripping the residual aqueous
phase, the quantity of  wastewater  discharged  is  approximately  5.7
rnVday.   The  organic loading in this waste is not known, however, if
                                    218

-------
TABLE 11-25.    TOXIC METAL CONCENTRATIONS AND LORDS IN CELL ROOM WASTE WATERS AT SCREENING AND
                VERIFICATION  PUNTS/  mg/1 \  (1)
                                    \ Jcgi/kkg /

Pollutant
014
261
Plant t
738A
738B
736
967**

Antimony
Arsenic
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Silver
Zinc
*
0.010
0.0000001
*
0.94
0.000014
0.53
0.0000075
0.26
0.0000039
*
54
0.00081
*
*
*
0.17
0.0000064
0.037
0.0000014
1.9
0.000071
17
0.00064
2.0
0.000075
*
22
0.00081
0.018
0.0000007
1.5
0.000054
0.050
0.00000 01
*
*
*
0.24
0.000042
0.044
0.0000077
0.0030
0.0000005
*
*
0.046
0.0000080
*
*
*
0.075
0.000012
0.38
0.000061
0.11
0.000018
*
0.061
0.0000098
*
0.46
0.000074
0.038
0.0000031
0.17
0.000014
*
0.54
0.000044
1.1
0.000090
0.047
0.0000038
0.0020
0.0000002
0.67
0.000055
*
0.5B
0.000048
0.41
0.00015
0.45
0.00017
0.016
0.0000059
0.086
0.000032
2.4
0.00089
<55>
0.0010
0.0000004
0.36
0.00013
*
0.92
0.00034
 Below detection limits
** Graphite anode plant

 (1)  Based on one  72-hour oonposite or the average of three 24-hour conposites.

-------
                             TABLE 11-26.
RAW WASTE TOXIC METALS  CCNGEMTRATICN AND LOADS IN PROCESS STREAMS OTHER THAN rETiT, ROOM WASTES
FROM SCREENING AND VERIFICATION PLANTS  (1)
M
O
Pollutant
Antinomy
Arsenic
Cadmium
Chromium
Copper
Lead
ffercury
Nickel
Silver
Zinc
f 014
*
*
0.0020
0.0000018
*
0.0040
0.0000036
*
0.0020
0.0000018
0.0030
0.0000027
*
*
I738A
*
*
*
0.53
0.0046
0.041
0.00035
0.083
0.00060
*
0.21
0.0018
*
0.29
0.0021
Plant!
I738B
(mg/11
(kg/kkq)
*
0.011
0.000019
*
0.065
0.00011
0.094
0.00016
0.11
0.00019
*
0.067
0.00012
*
0.058
0.00010
1736
*
*
0.038
0.0000062
*
0.090
0.000015
*
0.0030
0.0000005

*
4.3
0.00070
1967
*
0.29
0.0020
*
*
0.030
0.00020
0.40
0.0027
0.0020
0.000014
0.052
0.00035
*
0.15
0.0010
Avg
*
0.15
0.0010
0.020
0.0000040
0.29
0.00014
0.043
0.00014

0.0020
0.0000054
0.088
0.00072
*
1.5
0.0037
                               Below detection  limits

                             (1)  Based on one 72-hour composite or the average of three 24-hour ccrrposites.

-------
TABLE 11-27.
RAW WASTE TOXIC ORGANICS AT A GRAPHITE ANODE PLANT

StBCATEGQRY CHLORINE DIAPHRAGM CRT.T.
Pollutant
benzene
carbon tetrachloride
1, 2-dichloroethane
1,1, 1-trichloroethane
hexachloroethane
1,1, 2- trichloroethane
1,1,2, 2- tetr achloroethane
chloroform
1, 1-dichloroethylene
2 , 6-dinitrotoluene
roethylene chloride
bromoform
dichlorobrcnanethane
chlorodibrcmccnethane
hexachlorcbutadiene
bis ( 2-ethylhexyl) phthalate
di-n-butyl phthalate
tetrachloroethlene
toluene
trichloroethylene
Average
Concentration*
(mg/1)
0.00040
0.023
0.079
0.00014
0.010
0.00040
0.000044
0.085
0.000026
0.000026
0.00056
0.000063
0.035
0.0020
0.0040
0.00075
0.00078
0.036
0.0030
0.020
Load
(kg/day)
0.0011
0.066
0.23
0.00040
0.029
0.0011
0.00013
0.24
0.000074
0.000074
0.0016
0.00018
0.10
0.0057
0.011
0.0022
0.0022
0.10
0.0086
0.0057
 Flcw^proportioned concentration
                                     221

-------
TABLE 11-28.   RAW WASTE TOXIC ORGANICS BY WASTE WATER SOURCE AT A GRAPHITE
               ANOCE PLANT

SUBCATEGQRY CHLORINE DIAPHRAO1 rKT.T.

Stream
Cell building wastes
Caustic filter backwash
Brine fi.1t.ftr backwash
Cell wash
Chlorine condensate and
Spent H2SO4
Scrubber waste
Totals
Total Toxic
Qrganics
(mgA)
0.126
0.057
0.0030
0.20
2.2

0.81
0.30*
Total Toxic
Qrganics
(kg/day)
0.0093
0.12
0.00050
0.014
0.70

0.015
0.86
Percent of
Total Toxic
Organics
1.1
14
0.06
1.6
81.5

1.7
100
  Flow-proportioned concentration
                                     222

-------
the assumption is made that the discharge  is  saturated  with  carbon
tetrachloride  (CC14)  (800 mg/1 & 20 degrees C),  the waste load would
be 0.5 kg/day.

Although the daily mass emissions from the two plants are likely to be
similar  and  both  would  require  additional  treatment  to  achieve
acceptable  discharge levels, the wide difference in concentrations of
the chlorinated organics as well as  the  manner  in  which  they  are
handled  would  necessitate  the  application of an advanced treatment
technology specifically suited to each case.

Where  the  flow  is  large  and  the  concentrations  are  low,   the
application   of   activated   carbon   adsorption  to  the  collected
organic-bearing waste  stream  at  Plant  #967  would  be  capable  of
reducing  a CC14 mass emission from 0.066 kg/day to approximately 0.03
kg/day, assuming an achievable treatability level of 0.10 mg/1.


In the case of Plant #195, where the volume of wastewater is small but
the concentrations of residual chlorinated  organics  can  be  on  the
order of several hundred parts per million, a more appropriate removal
technology  would  be  stream stripping with an overhead return to the
process.  Assuming a treatability level of  10 mg/1 for CC14 using this
technology, its mass emission could be reduced to approximately  0.001
kg/day.

Process Modifications and Technology Transfer Options

A.   Anode Material

     The use of metal anodes rather  than   graphite  anodes  increases
     cell  power efficiency and greatly reduces the pollutant  loads of
     toxic organics and, in many cases,  lead  in  plant  wastewaters.
     Approximately  half  of the diaphragm  cell production of  chlorine
     is now by metal anodes.

B.   Caustic Evaporation Water

     The vapors from the evaporative concentration of caustic  soda are
     either contact-cooled or cooled in  surface  condensers.   Plants
     practicing contact cooling through barometric condensers  generate
     large  amounts  of  wastewater contaminated with caustic  soda and
     salt.   By  changing  from  contact  cooling  of  the  vapors  to
     noncontact  cooling,  or  by  recirculating  barometric condenser
     water,  the  amount  of  wastewater  generated  can  be   reduced
     considerably.   If  the  change is considered too expensive or is
     not  feasible,  demisters  or  similar  control  devices  can  be
     installed to reduce the salt and caustic carryover in the vapors.

C.   Diaphragm Material

     Although not in full scale  use  at  any  U.S.  chlorine  plants,
     modified  diaphragms  have  been developed which can reduce power
                                   223

-------
     consumption  and  minimize  asbestos  discharges.   ' The   polymer
     modified   asbestos  diaphragm  consists  of  a  polymer  treated
     asbestos diaphragm baked into place on the  cathode.   Its  usage
     results in power savings and has an environmental  benefit, since,
     at  the  time of diaphragm replacement, the discarded material is
     produced in stablized pieces instead of  loose  asbestos  fibers.
     Final disposal is thus safer and easier.

D.   Liquefaction of Chlorine

     Utilization of  high  pressure  and  refrigeration  for  chlorine
     recovery will reduce the chlorine content in tail  gases.

E.   Tail Gas Emission Control

     As with mercury cell plants, chlorine in tail  gases  has  to  be
     removed   and   treated   or  recovered  before  venting  to  the
     atmosphere.  The common practice is to scrub the gas with caustic
     soda producing a hypochlorite solution.   This  hypochlorite  can
     then  be  sold,  used  on-site or discharged.  Decomposition is a
     common method of removing the chlorine in this  stream  prior  to
     discharge.    Catalytic,   thermal   and   chemical   methods  of
     decomposition, described in Section 11, are effective.

Best Management Practices

The following Best Management Practices are common industry  practices
and  are provided for guidance purposes although they may not meet the
requirements  of  the  Resource  Conservation  and  Recovery  Act  (as
amended, 42 USC 6901 et seq.).

A.   Area Runoff

     Provisions can be made  to divert and contain  storm  runoff  from
     areas   where   lead    or  asbestos  contamination  could  occur.
     Collected runoff can then be treated with other wastes.

B.   Leaks and Spills

     Provisions can be made  in cell room areas to control and  collect
     the leaks or spills contaminated with  lead or asbestos.

C.   Contaminated Solids

     Asbestos waste and precipitated metals wastes should be stored in
     a  lined pond or disposed of in a secure landfill.

Advanced Treatment Technologies

The methods available and currently  used   in  the  industry   for  the
removal  of lead and other toxic metals from plant wastewaters include
hydroxide  or  carbonate  precipitation  followed   by   settling   or
                                   224

-------
filtration.   Further  removal of metals can be effected using sulfide
precipitation, adsorption and ion exchange.

Removal of asbestos from cell wastes is improved with the addition  of
coagulating agents prior to filtration of these wastes.

Selection of Appropriate Technology and Equipment

Technologies for Different Treatment Levels

A.   Level 1 (BPT)

     Level 1 treatment addresses the waste characteristics  associated
     with  diaphragm cell plants using graphite anodes.  The data from
     graphite anode plants were used because  the  pollutant  load  is
     greater  than  for metal anode plants.  Existing plants that have
     changed from graphite anodes to metal anodes will  have  residual
     effects  for  an extended time after the change—possibly as long
     as two years.  The asbestos contaminated  cell  room  wastes  are
     treated  with  a  flocculating  agent  and  a filter aid prior to
     f i1trat ion.  The  sol ids  are  removed  to  a  landf ill  and  the
     filtrate  is sent to a holding tank where it is combined with any
     other process waste sources containing treatable levels  of  lead
     and  other  toxic  metals.   Alkaline  precipitation of the toxic
     metals is accomplished by the addition of soda ash.   The  solids
     are  removed  by settling and the filtrate is combined with other
     process waste streams  such  as  chlorine  condensate,. tail  gas
     scrubber  water, caustic filter backwash and barometric condenser
     blowdown waters found to be contaminated  with  toxic  metals  at
     levels  usually  below  the  limits  of  treatability by alkaline
     precipitation.  The combined flow is sent to a polishing pond for
     additional clarification prior to discharge.  At  all  levels  of
     treatment, the brine mud is collected in lagoons and the effluent
     recycled  to  process.  The flow diagram for Level 1 treatment is
     shown in Figure 1.1-15.

     Level 1 treatment was ultimately selected as the  basis  for  BPT
     because  it represents a typical and viable industry practice for
     the control of asbestos fiber, lead, and other  toxic  metals  in
     wastewaters  associated with diaphragm cell plants using graphite
     anodes.  Note that the industry is concerned about the  discharge
     of  asbestos  and  has  taken  steps  to  reduce  this discharge.
     Treatment level 1 has  the  effect  of  reducing  total  asbestos
     discharges  even  though  we  do not set specific limits.  Plants
     utilizing metal anodes or nonlead containing graphite anode cells
     are expected to have lower levels of toxic  metal  emissions  and
     may  not  require alkaline precipitation to meet the proposed BPT
     limitations.   All  direct  discharge  plants  in  the   industry
     presently have BPT or equivalent treatment technology or meet the
     limits without treatment.
                                   225

-------
B.   Level 2 (BAT)

     The objective of Level 2 treatment technology is to achieve, at a
     reasonable cost, a greater degree of toxic  metals  removal  than
     provided by Level 1.  Thus, Level 2 adds dual-media filtration to
     the combined effluent from Level 1 treatment excluding noncontact
     waste streams.  Dechlorination of the filtered process wastewater
     discharge  is  also  included  in  Level 2 (BAT) treatment.  This
     assumes treatment by sulfur dioxide or bisulfite to remove  total
     residual chlorine.

     Level  2 was finally selected as the basis for BAT regulations on
     the strength of technology transfer options within the  inorganic
     chemicals  industry  and  because four out of five plants sampled
     were meeting BAT limits derived from published treatability data.
     In addition, two plants are known to practice  dechlorination  of
     their  effluent.  The flow diagram for Level 2 is shown in Figure
     11-16.

C.   Level 3

     The practice of sulfide precipitation of mercury in  the  mercury
     cell   segment   of   the  chlor-alkali  industry  suggested  the
     application of this technology for achieving greater  removal  of
     toxic  metals  in  diaphragm  cell  plants.  Level 3 adds sulfide
     precipitation to Level 2 as shown in Figure IT-17.   This  option
     was  not  selected  due  to its relatively high cost per- pound of
     additional metal removal obtained.  Cost estimates for this level
     can be obtained from the Development Document (60).


Equipment for Different Treatment Levels

A.   Equipment Functions

     Conventional sludge  dewatering  by  filter  press  is  used  for
     asbestos sludge before disposal and dual-media filter backwash is
     returned  to the influent surge tank.  Level 3 treatment requires
     the addition of a  reagent  mixing  tank  and  chemical  solution
     feeder  to  introduce  ferrous sulfide ahead of the Level 2 dual-
     media  filter.   All  equipment  is  conventional   and   readily
     available.

B.   Chemical Handling

     Nonhazardous solutions of aluminum sulfate and  sodium  carbonate
     are  the only solutions used at Levels 1 and 2.  Inert filter aid
     used in the  alum  sludge  filter  process  presents  no  unusual
     hazard.   At  Level  3  the  potential  hazard of handling sodium
     sulfide is nullified by reacting it with ferrous sulfate to  form
     ferrous  sulfide.   Any  excess ferrous sulfide will oxidize to  a
     ferric sulfide precipitate.  At the point where sodium sulfide is
                                   226

-------
     reacted with ferrous sulfate,  good ventilation  is  essential  to
     avoid the hazards associated with hydrogen sulfide gas.

C,    Solids Handling

     For  all  three  levels  of  treatment,  brine  mud  solids   are
     accumulated   in  lined  lagoons  on-site.   Abestos  solids  and
     precipitated metals wastes are to be sent  to  suitable  chemical
     landfills.

Treatment Cost Estimates

General Discussion

To  prepare  treatment  cost  estimates,  a  model  plant  concept was
developed.  Because higher pollutant loads and larger unit flows exist
at graphite  anode  plants  as  opposed  to  metal  anode  plants  the
characteristics  associated  with these plants were used when possible
for the model plant characteristics as discussed below.

A.    Chlorine Production

     Approximately 60 percent of the production data for all  chlorine
     plants  using  diaphragm  cells is available on file.  Production
     ranges from 15,000 to 1,500,000 kkg of chlorine per year.   Three
     model plants with production rates of  19,000, 95,500, and 191,000
     kkg   per   year  were  selected  to  represent  the  subcategory
     production range.

B.    Wastewater Flow

     Based on industry flow data (Table 11-19), waste streams  in  the
model  plants  are  segregated into brine mud, cell wash and cell room
wastes,  and  other  process  wastes  such   as   filter   backwashes,
condensates   and  tai1  gas  scrubber  wastes.   For  treatment  cost
estimates at all levels of treatment  the   following  flow  basis  was
used.

     1.   A brine mud flow of 0.45  mVkkg  is  sent  to  lagoons  for
          solids  removal.   Solids  are  disposed  of on-site and the
          clarified effluent is recirculated to the process.

     2.   Cell room wastes  consisting  of  leaks,  spills,  and  area
          washdown   contaminated  with  lead  and  other  metals  are
          combined with treated cell wash waters for a total  flow  of
          1.2  mVkkg  to  be treated for metals removal.  Included in
          this flow is the cathode  or  cell  wash  waters  which  are
          heavily  laden with asbestos.  These wash waters are sent to
          asbestos removal at a flow rate of 0.07 mVkkg, and then  to
          treatment with the other wastewaters.

     3.   Other process wastewater sources  account for  an  additional
          7,6   mVkkg  which  is  combined  with  effluent  from  the
                                    227

-------
to

03
                                          o
                       CEU. KXH

                      HUSK wow

                    (M3AL UUmilHMlD}
                                  IIOUttNQ WNK
                                           FUmXATIMl
                                             KSXt
                                  OStBDW
                                  i i*sres
                                             I
                                                (HMOl)
                                               tMOIMS 0MC
                                                  •*•
                                 L
                                                                               /
              a
                                                                                                pcuatmo
                                                                                                 KM)
  r    '
UKCflU.    I
 _  _ J
                     --<*)



                        I
                                                                                                                      ^rv-i
                                                                                                               pll iiiuiiilurlnu, Hiiw rnuiillornm
                                                                                                         and namjilc
                             Figure 11-15.   Level 1 waste water  treatment for chlorine - diaphragm  cell  subcategory.

-------
K>
to
                  CELL BOOM
                  WiSTE Nh'
                                          mxojuriw:
                                            *CEWT
                                £*"• BOM
                                 wsi*£ ' ' '•"
                                (ASBESTOS
                               U   --
                                                                 RtraCXf TO HOCESS
                                                                                   cniiDt PHXESS
                                                                                   tROTB HMEHS
                                                                                                              30-, Oft BISOU-TIE
                                                                                                                     mi.llu.-injj, lluw .
                            Figure 11-16.   level  2 waste  water treatment for  chlorine - diaphragm cell subcategory,

-------
ro
w
o
                   BRINE
                   MO
D                                         SOU ASII
                              r
                                       nfaoiunna
                                          KSNT
                             I    (ASBtiTOfl
                             '   CGHDIHINKIU)}
                                                                                   fZ±
RKSOBW
 FCPCESB
                                             •*fl^
                                                                                                       aooiiM     miix PKJCESS
                                                                                                               MSTC MITERS
                                                             SUOX
                                                            HOIOING

                                                             ™"
                                                                                                              30j OH BISU*'m
                                                                                                            1   T
                                                                                                                             -LQ-^aHjmr
                            1
                            1

                            1
                                                             u»                            I      I
                              I                                                            UMflU.    \
                        Figxire  11-17.   Level 3  waste water treatment for chlorine  - diaphragm cell subcategory.

-------
          treatment of wastes from the  cell  room  and  cathode  wash
          areas.    This  brings  the model plant total flow rate to an
          estimated 8.8 mVkkg.  The  final,  combined  process  waste
          flow  is either clarified and discharged as in Level \ (BPT)
          treatment   or   clarified,   passed   through    dual-media
          filtration,  and dechlorinated prior to discharge as in Level
          2 (BAT) treatment.

C.    Solid Waste Produced

     Brine mud constitutes the major source of solid  waste  generated
     at chlorine plants.  Although solids content varies from plant to
     plant,  an  average  of ten percent solids was used for the model
     resulting in a solids load of 42  kg/kkg  which  is  disposed  of
     on-site.    Asbestos  from  cell  wash operations and precipitated
     solids from metal treatment generate a solid waste of 0.83 kg/kkg
     that is disposed of off-site.

D.    Chlorine Bearing Wastes

     In the selection of model plants, the following assumptions  have
     been made for the chlorine contaminated waste streams.

     The chlorine condensate waste stream has not been included in the
     waste  streams  going to the treatment facility.  In the majority
     of the  chlorine/caustic  plants,  this  stream  is  stripped  of
     chlorine  by  steam or vacuum and the chlorine is recycled to the
     purification operation.  The wastewater is then returned  to  the
     process  and introduced to the brine purification unit or sent to
     the treatment unit.  The quantity of wastewater generated by this
     operation is small.  In some cases  the  chlorine  gas  from  the
     cells is contact-cooled with water and the scrubbed liquid, after
     steam  stripping,  is  reused.   The  stripping  operation in the
     recovery of chlorine is part of the process and,  therefore,  its
     cost is not included in the treatment cost.


     The   spent   tail   gas   scrubber  solution,  which  is  mainly
     calcium/sodium hypochlorite, is assumed to be used or  decomposed
     before  it  is discharged.  Thermal decomposition can be practiced
     at no additional cost at some facilities, while another efficient
     method  is  catalytic  decomposition.   The  cost  estimates  for
     decomposition  are  not   included here because at many plants the
     hypochlorite stream is sold, used on-site  or  only  infrequently
     discharged, depending on market demand.

     However,  because  of the environmental effects of high levels of
     chlorine in process  wastewater  discharges,  the  cost  for  the
     dechlorination  of  such  streams  using  sulfur dioxide has been
     included because this is the treatment method on which control of
     total residual chlorine is based.

     1.   Chlorinated Organic Wastes
                                    231

-------
The chlorine-bearing waste streams at graphite anode  plants
are  also  those streams carrying the highest concentrations
of toxic organics as indicated in Table 11-28.   Section  11
discussed the techniques used to recover and remove organics
from  waste  streams at Plant #195 and the fact that organic
contaminated streams can exist  as  either  high  volume-low
concentration  or low volume-high concentration depending on
plant  specific  factors.   Costs  for   removing   organics
(including  implementation  of  RCRA  regulations)  are  not
included in the model plant cost estimates because  organics
are  not  limited  in the regulation.  Organics occur at low
levels at most of the plants and when they are  present  the
appropriate treatment method is site specific.  Although the
costs are not included the following information is provided
as  guidance.  The additional costs for steam stripping in a
plant (such as  Plant  #195)  which  already  has  a  vacuum
vaporizer,  would  be  under $10,000 for modification of the
existing equipment.   Steam costs could vary from  $1,000  to
$5,000  per  year.   If a vaporizer is not in place, a steam
stripper to process 5  to  30  mVweek  would  cost  roughly
$50,000  to  $100,000, depending on the input concentrations
to be handled.  The corresponding steam  costs  would  range
from $2,000 to $10,000 per year.

The capital costs of an activated carbon adsorption unit for
handling  the  relatively  high  volume  wastes  with  a low
influent organic loading (as found at Plant #967) cannot  be
reliably  estimated  in the absence of specific treatability
data on the waste streams in question.

Alternatively,  incineration  of  the  chlorinated   organic
residuals  is an effective means of destroying and disposing
of this material provided that adequate measures  are  taken
to control the release of HC1 to the atmosphere..

A  process  evaluation  should be made to determine the most
efficient means for isolating  and  collecting  the  organic
bearing waste streams prior to treatment.

Incidental  removal  of chlorinated organics will occur with
the application of model plant treatment  levels  previously
presented.  Such removal, however, is expected to be erratic
and  therefore  cannot  be  predicted.   Because  raw  waste
concentrations of these organics vary considerably depending
on plant practices and are marginally  treatable  at  times,
applicable  control  and treatment technologies will need to
be assessed on a case-by-case basis.

For these reasons, the  Agency  is  not  providing  specific
numerical  discharge limitations for organic pollutants, but
is providing guidance for evaluating  control  options  that
could be applied in the industry.
                           232

-------
Model Plant Treatment Costs

On  the  basis  of  the model plant specifications and design concepts
presented earlier, the estimated costs of treatment for  three  models
having  different  production  levels are shown in Tables 11-29, 11-30
and 11-31.  The cost of Level 2 (BAT)  is  incremental  over  Level  1
(BPT)  costs  and provides for higher effluent quality with respect to
toxic pollutants.

Table 11-32 presents a summary of the unit cost  distribution  between
amortization and operation and maintenance components.

Basis For Regulations

Basis for BPT Limitations

BPT  regulations  are  currently  in  effect  for  the  diaphragm cell
chlorine subcategory, 40 CFR 815.62(b).  The Agency   is  revising  the
limitations, however, based on an increased unit flow rate.

A.   Technology Basis

     For BPT, the Agency is setting limitations based on equalization,
     alkaline precipitation and settling of lead and  asbestos-bearing
     wastes  and neutralization and settling of all wastewaters before
     discharge.  All diaphragm cell chlorine plants are  known  to  be
     using this technology (Level 1) or its equivalent.

B.   Flow Basis

     As described in Section 11, wastewater streams at diaphragm  cell
     plants are separated into two types, those that  require treatment
     for  asbestos  and  metals  removal and those that do not require
     such treatment.  From data presented in  Table   11-19,  the  unit
     flow  rate  of  1.2 m3/kkg of cell room and cell wash wastes from
     one graphite anode plant was  selected  as  the  flow  basis  for
     wastes  to  be  treated.   Graphite anode plant  data were used in
     this instance because the flows were higher than those  of  other
     plants  and  thus  represent  a conservative estimate of flow for
     other plants in the industry.   Using  available  flow  data  the
     remaining waste streams total 7.6 m3/kkg as shown in Table 11-33.
     Thus  the  total  unit  flow discharge used in the development of
     effluent limitations is 8.8 mVkkg.

C.   Selection of Pollutants to be Regulated

     The  selection  of  pollutants  for   which   specific   effluent
     limitations  are  being  established is based on an evaluation of
     raw waste data from screening and verification   sampling  and  on
     the treatability of toxic pollutants.

     Table 11-34 presents the achievable concentrations of toxic metal
     pollutants   (found  at  detectable  levels  in raw waste streams)
                                    233

-------
     using the  available  treatment  technology  options.   Based  on
     literature   treatability   data   presented  in  Section  8  and
     summarized in Table 8-11, the concentrations reflect  the  lowest
     level  achievable  by  these technologies.   Also presented in the
     table are the maximum three-day average raw waste  concentrations
     observed during the sampling program.

     Based  on  the  occurrence  of treatable levels of specific toxic
     metals, arsenic, cadmium, chromium, copper, lead, nickel and zinc
     are selected as candidate toxic pollutants for  BPT  regulations.
     Antimony,  mercury,  and  silver  were  detected but at less than
     treatable levels.

     Consideration of (1) the raw waste concentrations in Table  M-34
     and  (2)  Section  8  on  the  control  parameters  for hydroxide
     precipitation, leads to the selection of copper, lead and  nickel
     as  toxic  pollutants  to  be  regulated.   The  operation  of an
     hydroxide precipitation system at a pH of  about  9.5  should  be
     suitable  for  the  removal  of  the  metals  that are present at
     treatable  levels.    Arsenic,  cadmium,  chromium  and  zinc  are
     included  for  guidance  but no limits are set because control of
     copper, lead and  nickel  will  adequately  control  these  other
     metals.

D.   Basis of BPT Pollutant Limitations

     Limitations are presented as both concentrations (mg/1) and loads
     (kg/kkg), and the relationship between the two is  based  on  the
     unit  flow  rate (8.8 m3/kkg).  The concentration basis therefore
     represents  the  concentration  of  the  total  plant   discharge
     including  both  treated  and untreated wastewaters.  Guidance is
     presented as concentrations  (mg/1) only.

     BPT limitations, which apply at the combined  process  wastewater
     discharge  for TSS and toxic metals and at the plant effluent for
     pH, are presented in Table 11-35.

     1.   Conventional Pollutants

          a.   pH

               The treated effluent is  to  be  contolled  within  the
               range  of  6.0 to  9.0.  This limitation is based on the
               data  presented  in  Appendix   B   of   the   proposed
               Development Document (60) and the JRB Study  (52).


          b.   TSS

               The BPT limitations for TSS  are based on a  summary  of
               monitoring  data   from  Plant  #207  (3).  The long-term
               average discharge  load  of  0.30  kg/kkg  is  used  to
               develop  discharge limitations.   Because  variability
                                    234

-------
    TABLE 11-29.  MODEL  PLANT  TREATMENT COSTS
Subcategory     Chlorine  -  Diaphragm cell
Production      19,100  metric  tons per year
A.  INVESTMENT  COST                             BPT          BATa

      Site  development 	           30,000            0
      Equipment 	          136,000       43,000
      Monitoring  equipment 	           20,000            0

           Subtotal  •• -. t	          186,000       43,000
      Contractor's 0 & P°	           27,900        6,450

           Subtotal  	          213,900       49,450
      Engineering 	»           42,780        9,890

           Subtotal  	          256,680       59,340
      Contingencies  	           25,668        5,934

           Subtotal  	          282,348       65,274
      Land	           27,000            0

      TOTAL INVESTMENT COST 	          309,348       65,274
 B.  OPERATION AND
    MAINTENANCE COST

      Labor and supervision 	         112,000       14,000
      Energy 	           2,200          800
      Chemicals 	           1,500        1,500
      Maintenance 	          28,235        6,527
      Taxes and insurance 	           9,280        1,958
      Residual  waste disposal  ....           5,800            0
      Monitoring , analysis
       and reporting 	          15,000        7,500

      TOTAL OPERATION AND
      MAINTENANCE COST                     1*74,015       32,286
 C.  AMORTIZATION OF
    INVESTMENT COST                         45,938       10,620

    TOTAL ANNUAL COST                      219,953       42,906

a  Represents the incremental cost above that for BPT treatment
  Overhead  and  Profit
                              235

-------
    TABLE  11-30.  MODEL  PLANT  TREATMENT  COSTS
Subcategory
Production
Chlorine - Diaphragm cell
95,500 metric tons per year
A. INVESTMENT COST

     Site development  	
     Equipment  	
     Monitoring equipment  	

          Subtotal  ..............
     Contractor's o &  Pb	

          Subtotal  	
     Engineering  	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL  INVESTMENT  COST 	


B. OPERATION AND
   MAINTENANCE  COST

     Labor  and  supervision ,	
     Energy 	
     Chemicals  	
     Maintenance  	
     Taxes  and  insurance  	
     Residual waste disposal  ....
     Monitoring,  analysis
       and reporting 	

     TOTAL  OPERATION  AND
     MAINTENANCE  COST
                               BPT

                            85,000
                           261,000
                            20,000

                           366,000
                            54,900

                           420,900
                            84,180

                           505,080
                            50,508

                           555,588
                            69,000

                           624,588
                                                     ($)
                           112,000
                             4,900
                             7,500
                             55,559
                             18,738
                             29,000

                             15,000
                            242,696
    BAT

      0
 70,000
      0

 70,000
 10,500

 80,500
 16,100

 96,600
  9,660

106,260
      0

106,260
 28,000
  1,300
  7,500
 10,626
  3,188
      0

  7,500
 58,115
 C. AMORTIZATION  OF
   INVESTMENT  COST

   TOTAL  ANNUAL  COST
                             90,394

                            333,091
 17,289

 75,404
a  Represents  the  incremental  cost above that for BPT treatment
k  Overhead  and Profit
                              236

-------
   TABLE 11-31.  MODEL PLANT TREATMENT COSTS
Subcategory
Production
Chlorine - Diaphragm cell
191,000 metric tons per year
A. INVESTMENT COST

     Site development	
     Equipment 	
     Monitoring equipment  ...

          Subtotal	
     Contractor's 0 &  Pb.....

          Subtotal  	
     Engineering  	

          Subtotal  	,
     Contingencies  	,

          Subtotal  	
     Land	

     TOTAL INVESTMENT  COST  .


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  .
     Energy  	
     Chemicals  	
     Maintenance  	
     Taxes and  insurance  ...
     Residual waste disposal
     Monitoring,  analysis
      and reporting  	
                                                     ($)
                               BPT

                           140,700
                           353,000
                            20,000

                           513,700
                            77,055

                           590,755
                           118,151

                           708,906
                            70,891

                           779,797
                           129,000

                           908,797
     TOTAL  OPERATION  AND
     MAINTENANCE  COST
                            112,000
                              8,000
                             15,000
                             77,980
                             27,264
                             58,000

                             15,000
                            313,244
    BAT

      0
108,500
      0

108,500
 16,275

124,775
 24,955

149,730
 14,973

164,703
      0

164,703
 28,000
  1,800
 15,000
 16,470
  4,941
      0

  7,500
 73,711
C. AMORTIZATION OF
   INVESTMENT  COST

   TOTAL ANNUAL COST
                            126,873

                            440,116
 26,797

100,509
a Represents  the  incremental  cost above that for BPT treatment
b Overhead and Profit
                              237

-------
        TABLE 11-32.MODEL PLANT UNIT TREATMENT COSTS
     Subcategory Chlorine - Diaphragm  cell
                                Annual  Treatment Costs (S/kkg)
    COST ITEM
PRODUCTION
 (kkg/yr)
                                LEVEL  OF  TREATMENT
BPT
BAT*
Annual Operation
and Maintenance


Annual
Amortization


Total Annual
Cost



19r100
95,500
191,000

19,100
95,500
191,000

19,100
95,500
191,000

9.11
2.54
1.64

2.41
0.95
0.66

11.52
3.49
2.30

1.69
0.61
0.39

0.56
0.18
0.14

2.25
0.79
0.53
*Represents the incremental cost  above  BPT
                              238

-------
          factors  for  TSS  were  not   available   for   this   plant,
          factors  obtained from the  hydrofluoric acid subcategory
          are    used.    In  that   subcatergory,   where  the  same
          technology of  alkaline  precipitation   and   settling  is
          used,    the    average    variability   factor  for   daily
          measurements of  TSS  is  3.5 and  the average   factor  for
          30-day  averages is  1.7.   Thus,  utilizing the long-term
          average  discharge load  of  0.30   kg/kkg  one  obtains  a
          maximum  30-day average  load limit of:

          (0.30  kg/kkg)  (1.7)  = 0.51  kg/kkg

          and  a  24-hour  maximum limitation of:

          (0.30  kg/kkg)  (3.5)  = 1.1  kg/kkg

          The   concentration  bases   then are  derived by applying
          the  model  plant flow rate  of 8.8 mgVkkg to  obtain  a
          maximum  30-day average  concentration of 58  mg/1  derived
          as follows:

          0.51 kq/kkq  x  1000 mq/1 =  58 mg/1
          8.8  mVkkg     kg/m^

          A  24-hour maximum concentration of  120 mg/1 is derived
          from the variability factor ratio (VFR: 3.5/1.7 =  2.1)
          as follows:
           2.1
            (58mg/l
120  mg/1 (carrying two significant
     figures)
2.
     From the above data,  the  implicit  long-term  average
     concentration is:

     120 mq/1 = 34 mg/1
     3.5

     The  average  concentration  of TSS after lime addition
     and decantation in the  Treatability  Studies  document
     (61)  was  99 mg/1.  This average was calculated from a
     limited number of experiments on a nonoptimized system.
     Despite  these  shortcomings,  the  average  falls   in
     between  the  maximum  30-day  average  and the 24-hour
     maximum concentration limit for TSS.

Toxic Pollutants

a.    Lead

     The BPT limitations for lead  are  based  on  long-term
     monitoring  data  from  one  graphite  anode  plant  as
     presented in Appendix A.   The  plant  is  achieving  a
     long-term average lead discharge of 0.0064 kg/kkg.
                                 239

-------
TABLE 11-33
OF UNIT FLOWS AT DIAPHRftCM CKT.T. PLANTS
SUBCATEGORY
CHLORINE DIAPHRAGM CKT.T.
Stream Description
         Unit Flow
         Data
        Source
Cell room and cell
    wash wastes

Chlorine condensate

Tail gas scrubber waste

Caustic filter wash

Brine filter wash

Caustic moling blowdown


Spent sulfuric acid
           1.2


           0.78

           0.11

           5.4

           0.45

           0.86


           0.010
Graphite anode plant


Graphite anode plant

Graphite anode plant

Graphite anode plant

Graphite anode plant

Metal anode plants
      average

Metal anode plants
      average
      Total Unit Flow Discharge 8.8 m3Akg
                                     240

-------
            TABLE 11-34.   COMPARISON OF RAW WASTE CONCENTRATIONS
                    OF TOXIC POLLUTANTS WITH TREATABILTTY

SUBCATEGORY
Pollutant

Arsenic
Antimony
Cadimium
Chromium
Copper
Lead
Mercury
Nickel
Silver
Zinc

CHLORINE (DIAPHRAGM CELL)
Literature-based
Treatability (D
BPT
0.50
0.80
0.10
0.10
0.50
0.30
_(2)
0.20
0.40
0.50
BAT
0.50
0.40
0.050
0.050
0.40
0.050
»(2)
0.10
0.20
0.40

Maximum Plant
Raw Waste
Average
(mg/1)

0.30
d.oii
0.037
1.9
17
21
0.0030
22
0.018
3.0
(1)   Literature-based treatability estiinates from Table 8-11.

(2)   Treatability with this technology not available.
                                     241

-------
Statistical  .analysis of monitoring data from the plant
established a 30-day average variability factor of  1.6
and  a 24-hour variability factor of 4.1.  The ratio of
the two variability factors, VFR, is 2.6.  The  maximum
30-day  average limitation for lead is then obtained by
multiplying the variability factor for 30-day  averages
by  the  Jong-term  average  load;  i.e.,  1.6 x 0,0064
kg/kkg = 0-010 kg/kkg.  Similarly the  24-hour  maximum
limitation is obtained by multiplying the daily maximum
variability factor by the long-term average load; i.e.,
4.1 x 0.0064 kg/kkg = 0.026 kg/kkg.

The  concentration  basis  for lead is derived from the
relationship between concentration (C), unit load  (L),
and unit flow (Q).
C (mg/1) *
L (kq/kkq) x 1000 mg/1
Q (mVkkg)    kg/m3
Thus the concentration basis  for  the  maximum  30-day
average for lead is:

0.010 kq/Kkg x 1QOQ mq/1 =1.1 mg/1
8.8 mVkkg        kg/m*

The   concentration   basis  for  the  24-hour  maximum
limitation is  obtained  by  applying  the  variability
factor ratio (VFR) of 2.6 to the maximum 30-day average
concention:

(1.1 mg/1)  (2.6)  =  2.9 mg/1

From  the  above  data,  the implicit long-term average
concentration is:

2.9 mq/1 * 0.70 mg/1
4. 1
The average concentration of lead after  lime  addition
and  decantation   in  the Treatability Studies document
(61) was  28 mg/1.  This value  is much higher  than   the
values  calculated above from the  long-term monitoring
data in Appendix A.  The conclusion in the  Treatability
Studies   document,  that  there was   extremely   poor
settling   of   the   metal    hydroxide  sludge,  seems
justified.

Monitoring  data   from  six    diaphragm   cell    plants
presented in   Table  11-36  indicate that  plants using
metal anodes are meeting the BPT lead limitations.   One
of  two   graphite   anode   plants   is   meeting   the
limitations.
                      242

-------
                     TABLE 11-35.  EFFLUENT LIMITATIONS
                         Chlorine - Diaphragm Cell
          Best Practicable Control Technology Currently Available
                        Waste Water Flow:  8.8

Pollutant Subcategory Daily
Performance Variability
(mg/1) Factor
30 -day avg.
Variability
Factor
Concentration
(mg/1)
Max. 24-hr
30-day max.
avg.
-Effluent Limit
(kg/kkg)
Max.
30-day
avg.
24-hr
max.
Conventional Pollutants:
TSS
Toxic Pollutants:
34
                    (1)
3.5/1.7
58
120
0.51
1.1
Arsenic
Cadmium
Chromiun
Copper (5)
Lead<5>
Nickel (5)
Zinc
0.50 (2)
0.10<3>
0.32(3)
0.50(2)
0.70(1)
0.40 (3>
0.80(3)
4.V1-6
4.1/1.6
2.2/1.1
4.1/1.6
4.1/1.6
4.V1.6
4.V1.6
0.80
0.16
0.35
0.80
1.1
0.64
1.3
2.1
0.41
0.70
2.1
2.9
1.6
3.3
__(4) _J4)
__(4) __(4)
_(4) _(4)
0.0070 0.018
0.010 0.026
0.0056 0.014
~(4) "(4)
(1)  Calculated from long-term monitoring load data.

(2)  Lower limit of literature treatability for lime/clarification technology
     from Table 8-11.

(3)  Estimated achievable long-term average concentration from Table 8-13,

(4)  No load limits; concentration limits are provided for guidance purposes.

(5)  Also applicable for PSES limitations.
                                     243

-------
     The limitations and guidelines for the additional toxic
     pollutants  are  derived  from  two sources—industrial
     wastewater treatment system  performance  data  (Tables
     8-12   and   8-13)   and  literature-based  treatability
     estimates (Table ^8-11).  A comparison of these values,
     along with effluent  data  from  the  sampling  of  two
     diaphragm  cell  plants,  is  presented in Table 11-37.
     The concentration bases are derived from the industrial
     wastewater treatment system performance data unless the
     observed  concentrations  are  below   literature-based
     treatability estimates.  In such cases, the lower limit
     of the applicable treatability level is used.
b.    Copper
     The    estimated    achievable    long-term     average
     concentration for copper (0.40 mg/1 from Table 8-13} is
     lower  than  the literature-based treatability estimate
     (0.50 mg/1 from Table  8-11).   Therefore,  the  latter
     value  is  used  in  place  of  the  long-term average.
     Because no long-term monitoring data is  available  for
     copper  in  this industry,  the variability factors used
     for lead  are  employed  to  obtain  the  concentration
     limits.

     The 24-hour maximum concentration is:

     (0.50 mg/1)  (4.1 ) = 2.1 mg/1

     The maximum 30-day average concentration is:

     (0.50 mg/1)  (1.6) = 0.80 mg/1

     The load limitations for copper (kg/kkg) are calculated
     based on the unit flow rate of 8.8 mVkkg, thus:

     (2.1 mg/1) (8.8 mVkkg) (  kq/m3  ) = 0.018 kg/kkg
                             (1000 mg/1)

     for  the  24-hour  maximum  limit.   The maximum 30-day
     average limit is calculated similarly,  i.e.,

     (0.80 mg/1) (8.8 mVkkg)   (  kg/m?   ) = 0.0070 kg/kkg
                                (1000 mg/1)

     Nickel

     The nickel limits are based on the estimated achievable
     long-term average concentration of 0.40 mg/1  which  is
     obtained   from   Table  8-13.   Because  no  long-term
     monitoring  data  is  available  for  nickel  in    this
     industry,  the  variability  factors  used for lead are
     employed again.
                             244

-------
The 24-hour maximum concentration is:

(0.40 mg/1)  (4.1)   =  1.6 mg/1

The maximum 30-day average concentration is:

{0.40 mg/1) (1,6) = 0.64 mg/1

The load limitations for nickel (kg/kkg) are calculated
based on the unit flow rate of 8.8 mVkkg, thus:
(1.6 mg/1) (8.8 mVkkg}  (  kq/m^  ) = 0.014 kg/kkg
                         (1000 mg/1)

for the 24-hour  maximum  limit.   The  maximum  30-day
average limit is calculated similarly, i.e.,

(0.64 mg/1) (8.8 mVkkg)  (  kg/m*  ) = 0.0056 kg/kkg
                          (1000 mg/1)

The average concentration of nickel after lime addition
and  decantation  in  the Treatability Studies document
(61) was 0.41 mg/1.  This is esentially  equal  to  the
estimated  achievable  long-term  average concentration
which was the basis of the nickel limits.

Arsenic

The industrial wastewater treatment system  performance
for  arsenic  (0.080 mg/1 from Table 8-12) is lower than
the literature-based treatability estimate   (0.50  mg/1
from  Table 8-11).  Therefore,  the latter value is used
in place of   a  long-term  average  to  arrive  at  the
arsenic guidance.  Because no long-term monitoring data
is   available   for  arsenic   in  this  industry,  the
variability factors used for lead are employed again.

The 24-hour maximum concentration is:

(0.50 mg/1}  (4.1)  =  2.1 mg/1

The maximum 30-day average concentration is:

(0.50 mg/1)  (1.6)  =  0.80 mg/1

Cadmium

The  cadmium  guidance   is  based  on   the   estimated
achievable long-term average concentration of 0.10 mg/1
which   is    obtained  from  Table  8-13.    Because   no
long-term monitoring data  is available for   cadmium   in
                        245

-------
        TABLE 11-36. LEAD AND TSS DISCHARGES FROM SELECTED DIAPHRAGM CELL
                                CHLORINE PLANTS*1)

SUBCATEGORY
Plant
#589*
#738*
#261*
#014*
#967(3>
#207

Plant
#207
#014*
CHLORINE - DIAPHRAGM CELL
Lead Discharge
(kgAkg)
Average
0.0020
0.0010
0.0025<2)
0.0060<2),(4)
0.0064
0.021(2)
TSS Discharge
(kg/kkg)
Average
0.30*2)
2,8(2), (4)

Maximum
0,030
0.015
0.019
NA
0.026
0.054

Maximum
0.57
NA
  )  From Reference 3

* )  Plant has "once-through" barometric condenser water.

     Long Term Data Appendix A.

     Total plant discharge including other products.  Recent information from
     plant indicates current discharge levels much diminished from data given in
     Reference 3.

     Plants with metal Anodes

NA:  Not available
                                           246

-------
              TABLE 11-37.
               COMPARISON OF TOXIC POLLUTANTS IN
               DIAPHRAGM CKT.T. PLANT EFFLUENTS
               WITH TREATABILI'IY
Pollutant
Literature-based
Treatability(1)
   (mg/1)
Industrial Pfeste   Metal Anode  Graphite Anode
                                  Water Treatment
                                      System
                                   Performance
                                       (mg/1)
                   Plant #261
                   Effluent(2)
                     (mg/1)
                    Plant #967
                     Effluent*3'
                       (mg/1)
             BPT
          BAT
BPT
BAT
Arsenic
Cadmiizn
Chromium
Copper
Lead
Nickel
Zinc
0.50
0.10
0.10
0.50
0.30
0.20
0.50
0.50
0.050
0.050
0.40
0.050
0.10
0.40
0.080(4)
0.10(5)
0.32<5>
0.40(5)
0.15<5>
0.40 (5)
0.80(5)
0.17<4>
0.076(4)
0.16(5)
0.3*™
0.19(4>
0.30(5)
0.20C5)
0.12
0.0040
< 0.050
< 0.025
NA
< 0.050
< 0.025
0.30
< 0.015
< 0.050
0.031
NA
< 0.050
0.15
(1)  Literature-based treatability estimates from Table 8-11.

(2)  Cell wash waste filtered with coagulant to remove asbestos.

(3)  Flow-proportioned average discharge, consisting of lead treatment
     discharge and untreated filter backwashes, condensates, and scrubber
     wastes.

(4)  Data from Table 8-12  (value presented or average of values presented).

(5)  Estimated achievable long-term average concentrations from Table 8-13.

NA = Not available
                                     247

-------
this  industry,  the  variability factors used for lead
are employed again.

The 24-hour maximum concentration is:

(0.10 mg/1) (4.1)  =  0.41 mg/1

The maximum 30-day average concentration is:

(0.10 mg/1) (1.6)  =  0.16 mg/1

Chromium

The  chromium  guidance  is  based  on  the   estimated
achievable long-term average concentration of 0.32 mg/1
which is obtained from Table 8-13.  A daily variability
factor  of  2.2  and a 30-day variability factor of 1.1
are obtained from  the  Treatability  Studies  document
(61).

The 24-hour maximum concentration is:


(0.32 mg/1) (2.2)  =  0.70 mg/1

The maximum 30-day average concentration is:

(0.32 mg/1) (1.1 )  =  0.35 mg/1

The   average  concentration  of  chromium  after  lime
addition and decantation in  the  Treatability  Studies
document (61) was 0.071 mg/1.  This value is much lower
than   the   estimated   achievable  long-term  average
concentration used  above.   The  large  difference  is
attributable   to   the   very   low  average  chromium
concentration of raw waste in the Treatability  Studies
document (0.089 mg/1).

Zinc

The zinc guidance is based on the estimated  achievable
long-term  average  concentration of 0.80 mg/1 which is
obtained  from  Table  8-13.   Because   no   long-term
monitoring data  is available for zinc in this industry,
the  variability  factors  used  for  lead are employed
again.

The 24-hour maximum concentration is:

(0.80) (4.1)  =  3.3 mg/1

The maximum 30-day average concentration is:
                      248

-------
               (0.80) (1.6)  -  1.3 mg/1

Basis for BAT Effluent Limitations

Previous BAT regulations called for no discharge of process wastewater
pollutants.  The regulations were remanded.  The newly promulgated BAT
regulations  provide  for  the  discharge  of   pollutants   following
appropriate treatment of process wastes.

A.   Technology Basis

     Utilizing the cost estimates presented in this report the  Agency
     has  analyzed  the  cost  effectiveness  of the base level system
     (BPT) and an advanced level option (Level 2) for conventional and
     toxic pollutant removal.  The economic impact  on  the  diaphragm
     cell  chlorine subcategory has been evaluated in consideration of
     the technology basis for the BAT limitations.

     For BAT the Agency  is  promulgating  limitations  based  on  BPT
     technology  with  the addition of dual-media filtration (Level 2)
     and dechlorination of all process wastewaters.   Filtration  will
     remove  additional toxic metals and has been used successfully in
     the mercury-cell chlorine subcategory.  Dechlorination  is  being
     included  in BAT because the toxicity of chlorine to aquatic life
     is well documented and  it is a pollutant of concern to the Agency
     (59).   Two  chlorine   plants  are   known   to   be   practicing
     dechlorination.   The   Agency  considered the addition of sulfide
     precipitation  (Level 3) to the treatment of cell room wastes  but
     rejected it because further reduction of toxic pollutants in this
     stream  only  would  not  substantially  improve  total discharge
     quality.

B.   Flow Basis

     The flow basis for BAT  limitations   is  the  model  plant  total
     discharge of 8.8 mVkkg.   (See Table  11-33).

C.   Selection of Pollutants to be Regulated

     For BAT regulations, the Agency 'has selected the same three toxic
     metals identified in  the  BPT  regulations   (copper,  lead,  and
     nickel) and total residual chlorine.  The other four toxic metals
     (arsenic,  cadmium,  chromium,  and   zinc) are again included for
     guidance.

D.   Basis of Pollutant Limitations

     For BAT regulations, the Agency is  promulgating  more  stringent
     controls  on the discharge of the first three toxic metals on the
     basis of physical removal by filtration.  Alkaline  precipitation
     converts  most  dissolved  metals into less toxic insoluble forms
     and excess alkalinity   exists  in  most  of  the  process  wastes
     generated  in  this subcategory.  Plants whose raw waste has toxic
                                   249

-------
metal  concentrations  below  literature   treatability   levels,
perhaps  as  a consequence of installing metal anodes, would only
have  to  sample  periodically  to  show  they  are  meeting  the
regulations.   Final BAT limitations, which apply at the combined
process wastewater discharge for toxic metals and  at  the  plant
effluent for residual chlorine,  are presented in Table 11-38.

1.    Nonconventional Pollutant

     Chlorine:  Total residual chlorine limits are based  on  the
     data  from Appendix A that were used to calculate BAT limits
     for the chlorine mercury  cell  subcategory.   The  chlorine
     containing  waste streams from both technologies are similar
     and the methods for chlorine  removal  are  identical.   The
     long-term   average   concentration   is   0.64  mg/1.   The
     variability factor for daily measurements  is  2.3  and  the
     variability factor for 30-day averages is 1.4.

     The 24-hour maximum concentration is:

     (0.64 mg/1) (2.3)  =  1.5 mg/1

     The maximum 30-day average concentration is:

     (0.64 mg/1) (1.4)  =  0.90 mg/1

     The  load  limitations  for total residual chlorine (kg/kkg)
     are calculated based on the unit flow rate  of  8.8  mVkkg,
     thus:
      1.5 mg/1) (8.8 mVkkg)
                           kq/m3  )  =0.13 kg/kkg
                             (1000 mg/1)
2.
     for  the  24-hour maximum limit.  The maximum 30-day average
     limit is calculated similarly,  i.e.,
(0.90 mg/1) (8.8 mVkkg)


Toxic Pollutants
                               (  kq/m3  ) = 0.0079 kg/kkg
                               (1000 mg/1)
     Dual-media filtration of  BPT  effluent  will  significantly
     reduce  suspended  metal  precipitates.  BAT limitations and
     guidelines  are   derived   from   two   sources—industrial
     wastewater  treatment  system  performance data (Tables 8-12
     and 8-13) and literature-based treatability estimates (Table
     8-11).  A comparison of these values is presented  in  Table
     11-37.    The  concentration" bases  are  derived  from  the
     industrial  wastewater  treatment  system  performance  data
     unless     the    observed    concentrations    are    below
     literature-based treatability estimates.  In such cases, the
     lower limit of the applicable treatability level is used.
                               250

-------
There are insufficient data to calculate  daily  and  30-day
variability  factors for any of the toxic pollutants in this
industry.  A daily variability factor of 3.5  and  a  30-day
average  variability factor of 1.4 will be used to determine
BAT  toxic  pollutant  limitations  and  guidelines.   These
variability   factors  are  the  result  of  evaluating  the
Treatability  Studies  document   (61)   data   from   other
industries  involving  the  removal  of toxic metals via the
lime/filter method.

a.    Copper

     The  copper  limits  are  based   on   the   literature
     treatability  level  of 0.40 mg/1 (Table 8-11} which is
     larger than the estimated achievable long-term  average
     concentration of 0.30 mg/1 (Table 8-13).

     The 24-hour maximum concentration is:

     (0.40 mg/1) (3.5) = 1.4 mg/1

     The maximum 30-day average concentration is:

     (0.40 mg/1) (1.4) = 0.56 mg/1

     The load limitations for copper  (kg/kkg) are calculated
     based on the unit flow rate of 8.8 mVkkg, thus:

     (1.4 mg/1)  (8.8 mVkkg)   (  kq/m^   )   =0.012  kg/kkg
                               (1000 mg/1)

     for  the   24-hour  maximum  limit.   The maximum  30-day
     average limit is calculated similarly, i.e.,

     (0.56 mg/1) (8.8 mVkkg)   (  kg/in*   )  = 0.0049 kg/kkg
                                (1000  mg/1)


b.   Lead

     The lead limits are based on the data  in  Table   8-12,
     The  average  of  the  lead  values  in Table  8-12,  0.19
     mg/1,  is used in place of a long-term  average.

     The 24-hour maximum concentration is:

     (0.19 mg/1) (3.5) = 0.67 mg/1

     The maximum 30-day average concentration  is:

     (0.19 mg/1) (1.4) = 0.27 mg/1
                           251

-------
                     TABLE 11-38. EFFLUENT LIMITATIONS

                          Chlorine Diaphragm Cell

                         Best Available Technology
                                               3
                       Waste Water Flow:  8.8 m /kkg

Pollutant
Treatability
CngA
Daily
Variability
Factor
30-day avg.
Variability
Factor
Concentration
Basis
(mq/1)
Max. 24-hr.
30-day max.
avg.
Ef f luent Limit
(kg/kkg)
Max. 24-hr.
30-day max.
avg.
Nonconventional Pollutant
Total Residual
Chlorine

Toxic Pollutants
0.64(1)
2.3/1.4
Arsenic
Cadimium
Chromium
Copper
Lead
Nickel
Zinc
0.50(2)
0.076(3)
0.16(4)
0.40(2)
0.19(3)
0.30(4)
0.40(2)
                               4.1/1.6
                               3.5/1.4
                                 5/1.4
                                 5/1-4
                                 5A-4
                                 5/1.4
                               3.5A-4
0.90
                           0.80
                           0.11
                           0.22
                           0.56
                           0.27
                           0.42
                           0.56
1.5   0.0079
2.1
0.27
0.56
1.4
0.67
1.1
1.4
-(5)
-(5)
-(5)
0.0049
0.0024
0.0037
-(5)
0.013
                                     -(5)
                                     -(5)
                                     -(5)
                                     0.012
                                     0.0059
                                     O.(fo97
                                     -(5)
(1)  Long-term average concentration from Appendix A.
(2)  Lower limit of literature treatability for lime/filter technology from
     Table 8-11.
(3)  Industrial waste water treatment system performance data from Table 8-12,
(4)  Estimated achievable long-term average concentration from Table  8-13.
(5)  No load limits; concentration limits are provided for guidance purposes.
                                     252

-------
The load limitations for lead (kg/kkg)   are  calculated
based on the unit flow rate of 8.8 mVkkg), thus:

(0.67 mg/1) (8.8 mVkkg)  (  kg/in*  )   = 0.0059 kg/kkg
                          (1000 mg/1)

for  the  24-hour  maximum  limit.   The maximum 30-day
average limit is calculated similarly,  i.e.,

{0.27 mg/1) (8.8 mVkkg)  (  kq/m3  )    = 0.0024 kg/kkg
                          (1000 mg/1)

Nickel

The nickel limits are based on the estimated achievable
long-term average concentration of 0.30 mg/1  which  is
obtained from Table 8-13.

The 24-hour maximum concentration is:

(0.30 mg/1) (3.5) = 1.1 mg/1

The maximum 30-day average concentration is:

(0.30)  (1.4) = 0.42 mg/1

The load limitations for nickel (kg/kkg) are calculated
based on the unit flow rate of 8.8 mVkkg, thus:

(1.1 mg/1) (8.8 mVkkg)  (  kq/m3  )   - 0.0097 kg/kkg
                         (1000 mg/1)

for  the  24-hour  maximum  limit.   The maximum 30-day
average limit is calculated similarly, i.e.,

(0.42 mg/1) (8.8 mVkkg)   (  kq/m3  )    =  0.0037 kg/kkg
                           (1000 mg/1)

Arsenic

Because the BPT and BAT  literature treatability  levels
for arsenic are identical, the BAT guidance for arsenic
is set equal to the BPT  guidance.

Cadmium

The cadmium guidance is  based  on  the  data   in  Table
8-12.   The cadmium value  in Table 8-12, 0.076 mg/1, is
used in place of a long-term average.

The 24-hour maximum concentration is:

(0.076 mg/1) (3.5) = 0.27 mg/1
                      253

-------
               The maximum 30-day average concentration is:

               (0.076 mg/1)  (1.4) = 0.11  mg/1

          f.    Chromium

               The  chromium  guidance  is  based  on  the   estimated
               achievable long-term average concentration of 0.16 mg/1
               which is obtained from Table 8-13.

               The 24-hour maximum concentration is:

               (0.16 mg/1) (3.5) = 0.56 mg/1

               The maximum 30-day average concentration is:

               (0.16 mg/1) (1.4) = 0.22 mg/1

          g.    Zinc

               The  zinc  guidance  is   based   on   the   literature
               treatability  level  of 0.40 mg/1 (Table 8-11) which is
               larger than the estimated achievable long-term  average
               concentration of 0.20 mg/1 (Table 8-13).

               The 24-hour maximum concentration is:

               (0.40 mg/1) (3.5) = 1.4 mg/1

               The maximum 30-day average concentration is:

               (0.40 mg/1) (1.4) = 0.56 mg/1

BCT Limitations

EPA  has  determined  that  the BAT technology for this subcategory is
capable of removing significant amounts  of  conventional  pollutants.
However,  EPA  has  not  yet  proposed  or  promulgated  a revised BCT
methodology in  response  to  the  American  Paper  Institute  v.  EPA
decision mentioned earlier.  Thus, it is not now possible to apply the
BCT  cost  test  to  this  technology  option.   Accordingly,  EPA  is
deferring a decision on the  appropriate  BCT  limitations  until  EPA
proposes  the  revised  BCT  methodology.   However,   the  Agency  has
calculated the expected TSS  effluent  quality  after  application  of
dual-media  filtration and the cost of the additional TSS removal.  As
described in Section 3, this cost was calculated to be  $0.53  per  Ib
TSS removed.
                                    254

-------
Basis for New Source Performance Standards

A.   Technology Basis

     The Agency is basing NSPS limitations on the  BAT  technology  of
     alkaline precipitation, filtration, and dechlorination and on the
     performance  achieved  at plants which exclude the use of lead in
     cell construction.  The conversion to metal  anodes  has  largely
     eliminated  the  source of lead in wastewaters, but residual lead
     contamination at a converted plant may exist for  as  long  as  a
     year  or more.  New metal anode plants should have relatively low
     lead concentrations in their wastewaters.

B.   Flow Basis

     The flow basis of 8.8 mVkkg used for BPT and BAT limitations  is
     conservatively being used for new sources.

C.   Selection of Pollutants to be Regulated

     For NSPS regulations, the Agency initially  considered  the  same
     BAT  pollutants (seven toxic metals and total residual chlorine),
     pH and TSS.  Table 11-39 presents  the  results  of  verification
     sampling  of  raw  wastes  at  a  new  chlorine plant using metal
     anodes.  The total raw waste concentrations of toxic metals  are,
     with the exceptions of chromium and lead, substantially below the
     estimated  treatability  of these metals using BAT technology, as
     shown in the table.  Application of the information in Section  8
     on  the  control  parameters for hydroxide precipitation leads to
     the selection of  lead as the toxic  pollutant  to  be  regulated.
     Control  of  this pollutant within the limitations being set will
     assure that other  toxic  metals  are  being  controlled.   Other
     metals  are  presented  on  a  concentration  basis  for guidance
     purposes only.

D.   Basis of Pollutant Limitations

     For NSPS regulations the Agency is  promulgating  more  stringent
     controls on the discharge of lead on the basis of lower raw waste
     loads  generated  at  plants using metal anodes.  New plants that
     have raw  waste   lead  concentrations  below  estimated  industry
     long-term averages would only have to sample periodically to show
     they  are  meeting  the regulations.  The NSPS regulations, which
     apply at the combined process wastewater discharge  for  TSS  and
     lead  and at the plant effluent for residual chlorine and pH, are
     shown in Table  11-40.
     1
Conventional and Nonconventional Parameters
          a.   pH

               The treated effluent is to  be  controlled  within  the
               range  of  6.0 to 9.0.  This limitation  is based on the
                                    255

-------
               data  presented  in  Appendix   B   of   the   proposed
               Development Document (60)  and the JRB Study (52).

          b.    TSS

               Limitations for TSS are the same as BPT regulations.

          c.    Total Residual Chlorine


               Limitations for total residual chlorine are the same as
               in BAT regulations.

     2.    Toxic Pollutants


          a.    Lead

               The concentration  bases  for  lead  is  based  on  the
               estimated  achievable  long-term  average  of 0.15 mg/1
               (Table 11-39).  Variability factors used to obtain  BAT
               limits  for  lead  are  used  for the NSPS limits.  The
               concentration bases for lead are determined as follows:

               The 24-hour maximum concentration is:

               (0.15 mg/1) (3.5) = 0.53 mg/1

               and the corresponding effluent limit is:

               (0.53 mg/1) (8.8 mVkkg)  (  kq/m3  )    = 0.0047 kg/kkg
                                         (1000 mg/1)

               The maximum 30-day average is:

               (0.15 mg/1) (1.4) = 0.21 mg/1

               and the corresponding effluent limit is

               (0.21 mg/1) (8.8 mVkkg)  (  kq/m a  )   = 0.0018 kg/kkg
                                         (1000 mg/1)

          b.    Other Toxic Pollutants

               Guidance for the other toxic metals are the same as  in
               BAT regulations.

Basis for Pretreatment Standards

A.   Existing Sources

     For Pretreatment  Standards  for  Existing  Sources   (PSES),  the
     Agency is promulgating the same limitations as for BPT  (excluding
                                     256

-------
TABLE 11-39.  COMPARISON OF RAW WASTE CHARACTERISTICS AT A NEW METAL ANODE
              PLANT WITH TKEATABILITY OF TOXIC METALS
StBCATEGORY
CHLORINE DIAPHRAOt rKT.T.

Pollutant

Arsenic
Cadmium
Chromium

Copper
Lead
Nickel
Zinc

Treatability

0.50^)
0.050(1)
oae(3)
(3)
0.301^
0.15<3>
0.30(3)
0.20 (3)
Concentration(mg/l^
Plant #738B(2)
Raw Waste
0.011
<0.025
0.066

0.12
0.11
0.067
0.093
 (1)- Literature-based treatability estimates using BAT technology
     of dual media  filtration following alkaline precipitation of
     metals  (Table 8-11)

 (2)- Verification sampling  at new metal anode facility


 (3)- Estimated achievable long-term average from Table 8-13.
                                      257

-------
     TSS  and  pH)  based  on  the identical treatment technology (see
     Table 11-35).  The pollutants to be limited are copper, lead  and
     nickel.    Pretreatment  is  necessary because BPT provides better
     removal  of toxic metals than is achieved by a well-operated  POTW
     with  secondary treatment installed.  PSES are being set equal to
     BPT, because the POTW achieves equal or greater  removal  of  the
     toxic metals remaining after application of BPT (PSES) technology
     than  is  achieved  by  the  filtration required by BAT.  Thus, a
     filter is not required for PSES because there is no pass-through.

B.   New Sources

     For Pretreatment Standards for New Sources (PSNS), the Agency  is
     promulgating  the  same  limitations  as  for  NSPS  based on the
     identical treatment technology without dechlorination being  used
     for  indirect  dischargers  (see Table 11-40).  Dechlorination is
     unnecessary because influent to publicly-owned treatment works is
     often chlorinated.  The pollutant to be  limited  is  lead.   The
     other  pollutants (copper and nickel) are not being limited based
     on the assumption that all new plants will have  effluent  levels
     equal  to the levels of a new metal anode plant.  Pretreatment is
     necessary because NSPS provides better removal  of  toxic  metals
     (when present at treatable levels) than is achieved by a POTW and
     hence  these  pollutants would pass through a POTW in the absence
     of pretreatment.
                                     258

-------
                     TABLE 11-40.  EFFLUENT LIMITATIONS

                          Chlorine Diaphragm Cell

                      New Source Performance Standards

                       Waste Water Flow:  8.8 m /kkg
Pollutant       Treatability    Daily
                   (mg/1)     Variability
                                Factor
                              30-day avg.
                              Variability
                                Factor
                            Concentration
                               Basis
                              (mg/1)
                             Max.   24-hr.
                             30-day max.
                             avg.
                              Effluent Limit
                                (kg/kkg)
                              Max.
                              30-day
                              avg.
                       24-hr.
                       max.
Conventional
and
Nonconventional
TSS

Total Residual
Chlorine
34
0.64
3.5/1.7


2.3/1.4
58
120
0.90  1.5
0.51   1.1
         0.0079 0.013
Toxic Pollutants
Arsenic
Cadmium
Chromium
Copper
Lead (1)
Nickel
Zinc
0.
0,
0,
0.
0,
0.
0.
50
076
16
40
15
30
20
4.
3.
3.
3.
3.
3.
3.
VI.
5/1.
5/1.
5/1.
5/1.
5/1.
5/1.
6
4
4
4
4
4
4
0
0
0
0
0
0
0
.80
.11
.22
.56
.21
.42
.28
2
0
0
1
0
1
0
.1
.27
.56
.4
.53
.1
.70
-(2)
-(2)
-(2)
-(2)
0.0018
-(2)
-(2)
-(2)
-(2)
-(2)
-(2)
0.0047
-(2)
-{2)
(1)  Also applicable to PSNS limitations.

(2)  NO load limits; concentration limits are provided for guidance purposes.
                                      259

-------

-------
                              SECTION 12

                      HYDROFLUORIC ACID INDUSTRY
Industry Profile

General Description

Hydrofluoric acid (hydrogen  fluoride  or  HF)  is  produced  both  as
anhydrous  and  aqueous  products.   It  is used in the manufacture of
fluorocarbons which are used as refrigerating fluids and plastics, for
pressurized packing, and as dispersants in aerosol sprays.  HF is used
in the production of  aluminum,  in  the  refining  and  enriching  of
uranium fuel, in pickling of stainless steel, in petroleum alkylation,
and  for the manufacture of fluoride salts.  The industry data profile
is  given  in  Table  12-1.   The  status  of  regulations  prior   to
promulgation of these new regulations is given in Table 12-2.

Subcategorization

Hydrogen  fluoride  is  usually  used  in  the  production of aluminum
fluoride (A1F3) by reacting with hydrated alumina  (A1203»3H20).   Two
aluminum   fluoride  plants  are  integrated  with  hydrofluoric  acid
production.

For both products (HF and A1F3), process wastewaters are generated  by
the various gas scrubbers and by leaks and spills.  In both cases, air
pollution  control scrubber effluents contain mainly fluoride, acidity
and sulfate.  The fluoride is present as  the  free  ion  as  well  as
various  complex fluoro anions.  Calcium fluoride (CaF2), generated as
a solid waste, is a disposal problem for both subcategories because of
its moderate toxicity.

However, combining hydrofluoric acid  and  aluminum  fluoride  into  a
single  subcategory does not appear to offer any regulatory advantages
when the two products are manufactured at the same plant location.

The wastewaters generated from both industries do not  have  a  common
production  basis.   In  addition,  the  combined manufacture of  these
products does not create a unique or unusual  situation,  either  with
regard  to  the  wastewater  treatment requirements or compliance with
discharge regulations.

Due to  process  differences  and  the  variations  in  the  pollutant
loadings  for  the two industries, different variability factor ratios
were  used  in  obtaining  the  effluent   limitations  of  the    toxic
pollutants.   In the case of the hydrofluoric acid subcategory, nickel
and zinc are chosen as the control toxic pollutants for  the  effluent
limitations.   On  the  other  hand, chromium and nickel are the  toxic
metals  controlled  in  the  aluminum  fluoride  subcategory.     These
                             261

-------
TABLE 12-1     -     SUBCATEGORY PROFILE DATA SUMMARY

SUBCATEGORY          HYDROFLUORIC ACID

Total subcategory capacity rate                   363,000 kkg/year

Total subcategory production rate                 261,800 kkg/year

Number of plants in this subcategory                    9

308 Data on file for                                    8
     With total capacity of                             *
     With total production of                     177,000 kkg/year
     Representing capacity                              *
     Representing production                           68 percent

     Plant production range:
          Minimum                                   7,300 kkg/year
          Maximum                                  62,000 kkg/year

     Average production                            22,100 kkg/year
     Median production                             15,800 kkg/year
     Average capacity utilization                      83 percent

     Plant age range:
          Minimum                                       7 years
          Maximum                                      58 years

     Waste water flow range:
          Minimum                                       0 cubic meters/day
          Maximum                                   4,700 cubic meters/day

     Volume per unit product:
          Minimum                                       0 cubic metersAkg
          Maximum                                      86 cubic metersAkg

Sources of data are Stanford Research Institute, Directory of Chemical
Producers, U.S.A., 1977, U.S. Department of Commerce, Current Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; and Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry/ "June, 1978 and "Economic Analysis of Proposed Revised
Effluent Guidelines and Standards for the Inorganic Chemicals Industry," March,  1380.
* Data inconplete because certain plants did not respond to this question.
                                      262

-------
        TABLE 12-2. STATUS OF REGULATIONS - EFFLUENT LIMITATION GUIDELINES
SUBCATEGORY

SUBPART
HYDROFLUORIC ACID

H (40CFR 415.80, 3/12/74)
STANDARDS
Product
Process
BPCTCA*
1 7
Max. ' Avg. *•
Parameters ( mg/1 ) ( mg/1 )
BATEA*
1 ">
Max. Avg. *•
(mg/1) (mg/1)
NSPS*
Max. ' Avg.2
(mg/1) (mg/1)
Hydrofluoric
 Acid        Fluoride     (30)
             TSS          (50)
                      No discharge
              (15)     of pwwp3

                      No discharge
               (25)   of pwwp
No discharge
 of pwwp

No discharge
 of pwwp
  Section 415.82, 415.83, and 415.85 were remanded and reserved (41 FR 51601,
  November 23, 1976).
  Max. = Maximum of any one day.
  Avg. = Average of daily values for thirty consecutive days.
  pwwp — Process wastewater pollutants.
                                       263

-------
differences  are  due  to  the  different  relative concentrations for
chromium and zinc in the raw waste loads from the two products.


Furthermore,  the  opportunities  for  drip  acid  recycle   (or   the
hydrolysis  of complex fluoride prior to treatment) and scrubber water
recycle are a function of plant design and age,  rather  than  product
mix.

An  additional solid waste, gypsum (CaS04»2H20), is generated from the
hydrofluoric acid manufacture and supplies enough calcium for adequate
fluoride removal from neutralized scrubber  wastewaters  generated  by
both   HF   and  A1F3  production.   However,  the  applied  treatment
technology is essentially the same as that applied by manufacturers of
either product alone.

In view  of  these  considerations,  hydrofluoric  acid  and  aluminum
fluoride remain separated as two distinct subcategories.

General Process Description and Raw Materials

HF  is the most important manufactured compound of the fluoride family
in volume of production.  Fluorspar (mainly CaF2)  and  sulfuric  acid
are  the  raw  materials  used  for  its  manufacture.   Fluorspar and
sulfuric acid react endothermically  at  200-250  degrees  C  and  the
reaction time is 20-60 minutes.  The reaction is given as:
     CaF2 + H2S04 + heat = CaS04 + 2HF
                                                        (1)
The  reaction  kinetics and the yield of product depends on the purity
and fineness of the fluorspar.  The concentration  of  sulfuric  acid,
the  temperature  of  the  reaction, and the ratio of sulfuric acid to
fluorspar are among important variables.

Crude fluorspar, as mined, varies  in  CaF2  content  from  35  to  90
percent.  The ore is upgraded by flotation which results in 98 percent
CaF2  being  available  for  the  production of HF.  The analysis of a
typical upgraded fluorspar is given as:
     CaF
     S
     H20
     CaC03
                  Minimum 97.5-98%
                  Maximum 1.0%
                  Maximum 0.05%
                  Maximum   0.1%
                  Principal  remainder
Silica  is  a  highly  objectionable  contaminant,  since   each  pound
consumes  2.0  pounds  of fluorspar and  3.3 pounds of  sulfuric acid  by
the following reaction:
Si0
            2CaF2 +2H2S04 = SiF4  +  2CaS04  +  2H20
                                 264

-------
Sulfuric acid having a concentration as low as 93 percent or  as  high
as 99 percent is generally used.  Dilute sulfuric acid enhances better
mixing and liberation of fluoride but has two disadvantages; viz., the
dilute  acid  is  very  corrosive  and  the  water present in the acid
evaporates and distills off with the HF  gas,  thus  reducing  product
concentration.  . Concentrated  sulfuric acid (greater than 98 percent)
offsets these disadvantages  but  creates  new  problems.   The  vapor
pressure  of  concentrated sulfuric acid is sufficiently high to cause
large amounts of sulfuric acid to be carried away by the  HF.   Excess
sulfuric  acid,  when  used, will leave with the gypsum as part of the
residue.

HF generators are, in the majority of cases,., externally  fired  rotary
kilns in which acid and fluorspar are fed continuously through a screw
conveyor  at  the forward end and gypsum is removed from the other end
through an air lock.  The product HF may discharge  from  either  end.
The theoretical amount of gypsum produced is 3.4 kg/kg of HF produced,
but  because  of  the impurities in the fluorspar the actual amount of
gypsum produced is higher and varies from 3.6 to 4.8 kg/kg of HF.

One manufacturer uses a patented process to supply  internal  heat  to
the  reactor.   The  heat  is  supplied by introducing sulfur trioxide
(S03) and water (as steam).  The  exothermic  heat  liberated  by  the
reaction  of  S03  and  water to produce sulfuric acid is used for the
heat required for HF generation.  Thus a part of the sulfuric acid  is
supplied as SO3.

The  HF  gas  leaving the reactor is cooled in a precooler to condense
high boiling compounds.  The condensables are known as drip  acid  and
largely  consist of fluorosulfonic acid (HS03F) and unreacted sulfuric
acid.   In 1978, nine plants out of a total of eleven returned the drip
acid to the reactor, while the remaining two sent the drip acid to the
waste treatment plant.  The  HF  gas  from  the  precooler-   is  cooled
further and condensed in a cooler/refrigeration unit.  The uncondensed
gas  containing  the HF is scrubbed with sulfuric acid and refigerated
to recover the product.  The scrubbed acid liquor is returned  to  the
kiln,   and  residual  vent  gases  are  scrubbed further with water to
remove  HF and other fluoride compounds before they are vented  to  the
atmosphere.   The  scrubber  water is sent to the wastewater treatment
plant.  Figure  12-1 is a  block  flow  diagram  of  the  manufacturing
process.

The  crude HF is then distilled to remove the residual impurities, and
the condehsate, which is anhydrous HF, is stored in tanks.   If aqueous
HF is desired, the crude product is then diluted with water  to form   a
70 percent HF solution as the final product.
                                265

-------
                                       OLEDH
SLOSRY NK.TER



ACID
FLOORSPAJt
CM^T™ ,..,,
SCLFATE (GYP SUM)

_
F WB-ccwacr~*
nflrt nas ^^L_
	 	 *•!"!••• IMP ^^M^






COOLIMG OR ^

SYSTEM £
KftXER
1 1—
MHDEN
COLLB
9Q

t i




COO
1

DRX1

COX
1
CONDI

:EODE :
FLD<
STOR]
SABI^S
CTIOH
F

•


r mi

LES
f

t POT
r
E BOX
'
OISEK

IYDROGE
IRXDB
USE






"ER, SQL
hCZO t
SOLFOMI






•

B
	 1






PI
f
C








s







ntzc
ao-
ACZD
RBCY





AC
scum
•UUTJRX










CLE





CD
SBER.
t
: ACID

 WATER'
 EJECTOR


 HtSSEtBOER

    TO
TREATMEHT
DISTllIATIOM-C0!IDBHS*rE^ TO STORAGE 
-------
Water Use and Waste Source Characteristics
Water Use

Water  is  used  in HF production in noncontact cooling, air pollution
control, product dilution, seals on pumps and kilns, and for equipment
and area washdown.  Although noncontact cooling constitutes the  major
use  of  water,  water  is  also  used, in a majority of cases, in the
transport of gypsum as a slurry to the wastewater treatment  facility.
The water for gypsum transport is provided by either reusing the water
from  the  treatment  facility or by using once-through cooling water.
Table 12-3 summarizes the water usage found in this study.

Waste Sources

A.   Gypsum Solids

     Gypsum solids are generated as a by-product.  The amount produced
     is in the range of 3.6 to 4.8 kg/kg of HF produced.   The  gypsum
     also  contains  small  amounts  of sulfuric acid, HF, and calcium
     fluoride.  Minor amounts of other impurities present in fluorspar
     are also removed with the gypsum.  In five out of  eleven  plants
     producing   HF,  gypsum  is  slurried  with  treated  wastewater,
     neutralized with lime or soda ash, and pumped to a gypsum storage
     pond.  In one plant the gypsum slurry is pumped  to  the  storage
     pond   without   treatment-  and   in   another   plant   partial
     neutralization is employed.  Three plants transport the gypsum as
     a dry solid-and dispose of it as a solid waste after mixing  with
     lime for neutralization.  The disposal method of one plant is not
     known.   It  should  be  noted that two of the eleven plants have
     recently discontinued HF production, one of which is in the group
     of five.

     When gypsum solids from the kiln  are  slurried  with  water  for
     treatment,  the  resulting stream constitutes the major source of
     wastewater.  When kiln residue is disposed of as a  solid  waste,
     scrubber  wastewater  is  the  major source of waste.  Table 12-4
     gives the data  for  the  direct  and  indirect  process  contact
     wastewater  going  to  treatment  facilities.  Noncontact cooling
     water has not been included in the figures given in  Table  12-4.
     Figure   12-2  is  a graphical representation of production versus
     wastewater flow to inplant treatment facilities for plants  whose
     wastewater  includes  the  gypsum slurry and for those practicing
     disposal of kiln residue as a solid waste.

B..  Drip Acid

     This is  formed in the first stage of the cooling   (i.e.,  in  the
     precooler)  of the gases emitted from the kiln.  Drip acid mostly
     contains high boiling compounds consisting of complex  fluorides,
     especially fluorosulfonic acid, and small amounts of hydrofluoric
     acid, sulfuric acid, and water.  Fluorosulfonic acid is formed by
                                 267

-------
  TABLE 12-3.  WATER USAGE IN THE HYDROFLUORIC ACID SU8CATEGORY
                               Water Usage at Plants
                                  (ra3/kkg of HF)
                      (1)    (1)                                         (1)
   Source         #987   #251   #753   #426   #120   #722   #167   #705
Non-contact       154    NA     63.5   110    NA     13.6   116     30.0
 Cooling

Gypsum Slurry     NA     64.0   NA     *      NA     22.5    41.6   30.0
 Transport

Maintenance,      NA      2.40   2.11  NA     0.1    12.2     5.00  16.9
 Equipment and
 Area Washdown

Air Pollution      7.90  14.4    4.23  NA     0.586  14.5     40.0  11.3
 Control
NA = Not Available

*  = Not Applicable

(1)  Discontinued HF production.
                                   268

-------
      TABLE 12-4.  WASTE WATER FLOW AND REUSE DATA FOR THE
                  HYDROFLUORIC ACID SUBCATEGORY
Plant
#120
#426
(3)
1987
#837
#967
(31
#251
(3)
#705
#167
#753
#928
#664
#722
Kiln Residue
(D
Handling
D
D
D
S
S
S
S
S
S
S
S
S
Reuse for
Kiln Residue
(21
Slurry
(Percent)
_„,
(4)
(41
0
0
0
30.0-35.0
47.0
65.0
83.0
94.0
92.0-100
Influent to
Treatment
Facility
(m3/kkg)
HF
9.10
0
13.6
120
125
84.7
58.2
166
31.4
55.5
96.6
120
Treated
Effluent
Di scharged
(m3Akg)
HF
9.10
Not available
13.6
120
125
84.7
39.3
38. 2
11.1
9.40
5.80
7.20
Averages: (S only)

(1) D = Dry disposal
42.8 percent     95.4 m

 S = Slurried to treatment
54.6 m
(21 Percent of waste water flow reused for residue slurry after
    treatment.
(3) Dicontinued HF production.

(4) Not Applicable.
                                   269

-------
15,000
12,500 t
  1,000
                                              Dry Kiln Waste
                                        Slurrying Kiln Waste
75     100            150
 HF Production, kkg/day
                                                                    200
      Figure 12- 2.   Production versos waste flow data for HF plants.
                                 270

-------
D
reaction  between  hydrofluoric  acid  and  suIfuric  acid in the
absence  of  water.    The  quantity  of  drip  acid  produced  is
relatively small.  In the plants which recycle the drip acid back
to  the  reactor,  it is mixed with the sulfuric acid feed stream
before it enters the kiln where it is hydrolyzed to form sulfuric
acid and hydrofluoric acid.  The critical factors for  hydrolysis
are  temperature  and retention time and enough water is normally
present in the kiln for the reaction.

Noncontact Cooling Water

Noncontact cooling water is used for precooling the product gases
emitted from the kiln.   The  possibility  of  product  or  other
process  compounds  leaking into the cooling water is very small;
however,  in  the  event  that  the  cooling  water  does  become
contaminated,  the  proposed  limitations  for  fluoride  may  be
exceeded.  Depending on the merits of the  situation,  the  upset
and  bypass  provisions  may  apply.  In some plants, the cooling
water is used to transport the waste gypsum.

Scrubber Wastewater

Scrubber water is another wastewater source, and in plants  which
practice  dry  disposal of gypsum, scrubber water constitutes the
predominant  and  major  source  of  wastewater.    It   contains
fluoride,  sulfate,  and acidity.  The fluoride is present as HF,
silicon tetrafluoride (SiF4), and hexafluosilicic  acid  H2SiF6).
Silica  present  in the ore as an impurity reacts with HF forming
silicon tetrafluoride as shown in Equation 3.
SiO
       4HF = SiF
                         2H»0
(3)
     In the  scrubber,  the  silicon  tetrafluoride  is  converted  to
     hexafiuosilicic .acid according to the following equations:
SiF4 + 2HF = H2SiF6

3SiF4 + 2H20 - 2H2SiF6

Distillation Wastes
                              Si0
                                                       (4A)

                                                       (4B)
     The distillation waste generally contains HF and water.  In  some
     cases  the  vent  gases from the distillation column are scrubbed
     before they are emitted to  the  atmosphere,  and  the  resulting
     scrubber water requires treatment.

     The   range  of  wastewater  quality  of  the  different  streams
     generated from the production of HF is summarized in Table  12-5.
     The  data  are  taken  from  the prior development documents, 308
     Questionnaire responses, and industry visits.
                                  271

-------
F.   Spar Drying Wastewater
     Wet fluorspar is  first  processed  through  a  dryer  to  remove
     essentially  all  of  the water.  The steam and dust generated is
     passed through a scrubber from which the wastewater  is  sent  to
     the  gypsum  neutralization  tank.   As  shown in Table 12-6, the
     contribution of  the  spar  drying  wastewater  is  insignificant
     compared  to the total amount of process wastewater and hence can
     be  neglected.   The  data  is   taken   from   308-Questionnaire
     responses.

G.   Other Solid Wastes

     The total solids generated from the  process  and  the  treatment
     system consist of gypsum and the fluoride precipitated as calcium
     fluoride.   Table  12-7  gives  the  amount  of  suspended solids
     generated from the process and the quantity  of  total  suspended
     solids  generated  at  the  wastewater treatment plant for the HF
     plants visited in screening and verification.  The data  indicate
     that  the  gypsum  waste  constitutes more than 95 percent of the
     total solids produced.  Table 12-8 gives  the  amount  of  gypsum
     solids  produced  at  different HF manufacturing facilities.  The
     data shows that 3.8 to 4.7 kg gypsum solids are produced  per  kg
     of product.

Description of Plants Visited and Sampled

Screening

Plant  #705  was visited and process wastewater samples were collected
and analyzed for conventional, nonconventional and  toxic  pollutants.
The  process  used  at  this  site  is  similar to the conventional HF
manufacturing process described earlier.  The drip acid is sent to the
wastewater treatment facility and the gypsum produced from the reactor
is slurried with water and also sent to the treatment  facility.   The
wastewaters from the HF production facility are combined with aluminum
fluoride  plant  wastewaters.   The combined raw wastewater is treated
with lime and sent to settling ponds before  discharge.   Figure  12-3
shows  the  general  process and the locations of the sampling points.
Table 12-9 gives the flow data and the total  suspended  solids   (TSS)
and fluoride emissions.

Verification

The  same streams at Plant #705 were sampled again in the verification
phase.  The variation in the flow of the streams in the  two  sampling
phases  was  negligible.   Table 12-10 gives the TSS and fluoride load
summary of the sampled streams.

Two more  HF  plants  (Plant  #251  and  #167)  were  sampled  in  the
verification  phase.  The drip acid at both facilities is also sent to
the waste treatment plant and the hydrofluoric  acid  wastewaters  are
                               272

-------
                                          TAME 12-5.  WASTE HOW FTCM HYDROFLUORIC ACID MAHUFACttJRING PLANTS
to
-4

Source of
Waste Water
Gypsum Slurry
Drip Acid
Scrubber
Haste Water
Other
Plow in m /Wog of Hydrofluoric Acid
Plants
1251 (1) #987 (1) |753 |426 1120 1722
64.0 Dry NA Dry Dry (Total
disposal disposal disposal Recycle)
0.0490 0 000 0
14.4 8.30 2.30 NA 0.624 (Total
Recycle)
0.530 0.530 8.40 NA 5.55 NA

1167 1705 (1) 1837
122 (Total 6.50
Recycle)
NA 0.0180 0
40.0 11.3 1.12
5.20 22.5 NA
                          (1)  Discontinued HP production


                         NA = Not Available

                         *    Other does not include wasteflows from storm water runoff.

-------
      TABLE 12-6.  FLOW AND POLLUTION CONCENTRATION DATA OF SPAR IKYING
                   AND TOTAL PROCESS WASTE WKEER FOR PLANTS #251 AND
                   #837 PRODUCING HYDROFLUORIC ACID

Stream
Plant Description
# 251 Spar drying
waste water
., Flow
(m /kkg of HF)
1.20 (1.8%) (1)
Total
Susperided Solids
(kg/kkg of HF)
70.8 (1.8%) (1)

# 837

Total process waste
water
Spar drying waste
water
Total process waste
water
66.9
1.11 (1%) (1)
114
4000
16.7 (0.5%)
3140

(1)

     Nuiber in parentheses is tiie percentage contribution of a pollutant
     parameter of the spar drying waste water to that of the total
     process waste water.
 TABLE 12-7.  SOLID WASTE (DERATED AT THE HYDROFLUORIC ACID PLANTS SAMPLED
                           Gypsum Solids Going To    Total Solids Produced
      Plant                  Treatment Facility          (kgAg of HF)
                               (kg/kg of HF)


      #705(1)                      4,73                       4.78
      #251(1)                      3.81                        NA
      #167                         3.94                        NA

(1)  Discontinued HF production.
NA = Not Available
                                     274

-------
TABLE 12-8.  G¥PSIK SOLIDS PRODUCTION IN THE HYDROFLUORIC ACID SUBCATEQQRY

Plant
*837
#705 (1)
#167
#722
#120
#426
#987 Cl)
#251t:L)
#753
#967
#928
Kiln Residue Produced
(kg/kg of HF)
3.86
4.73
3.94
NA
NA
4.00
4.13
3.81
NA
NA
NA
Kiln Residue
Disposal/Treatment Method
S
S
S
S
D
D
D
S
S
S
S
S = Slurrxed with water and sent to waste water treatment facility.
D = Dry disposal.
NA = Not Available.
(1) = Discontinued HF production.
                                     275

-------
ro
-j
            FlilQBSiftg
            'W
                                                      t       f
                                              aOUBBER
                        KIUI
                        s.
                                   FMSH

                                   IATCR
                                 MIPKID
                                 '
                                        UMB
     e.
                                   14

                                   flUWJCE DBMNS
e
                                 IJUJO



                                  BtFNMB KW|>}ad.
                              ^
                                12
                                                        Uf PBOOJCT
                     Figure 12-3.  General process  flew diagram at plant  #705  showing the sampling points.

                                               Hydrofluoric acid manufacture.

-------
 TABLE 12-9.  FI0W AND POLLUTANT CONCENTRATION DATA OF THE SAMPLED WASTE
              STREAMS OF PLANT #705- PRODUCING HYDROFLUORIC

Stream
No.
1
2
3
4
Sarrpled
Stream
Description*
Kiln Slurry
Scrubber Waste
Water
Surface Drains
Cooling Tower
Slowdown
Treated Effluent
Screening
Flow
(m3Akg of HF)
26.6
10.0
20.0
23.3<3>
Data(2)
Fluoride
(kg/kkg of HF)
15
9.6
6.9
1.6
Tbtal
Suspended
Solids
(kg/kkg of HF)
4700
0.070
3.9
1.9
(1)   This  plant has discontinued the production of HF since the time of
     sairpling.

(2)   One 72-hour conposite sanple of each waste water stream.

(3)   Ihe discharged effluent consists of the treated waste waters  from
     hydrofluoric acid and aluminum fluoride plants.
                                    277

-------
combined  with  aluminum  fluoride  plant  waste  for  treatment.   In
addition to drip acid, Plant  #251  wastewater  consists  of  scrubber
water,  gypsum  slurry,  and  plant  area  hose  down.   The treatment
consists of gypsum ponds where the suspended solids are removed.   The
overflow  from the last gypsum pond is neutralized and the pH adjusted
with wastes from other product lines.   Figure 12-4 is a block  diagram
of the process showing the sampling locations at Plant #251.

At  Plant #167, the major raw waste sources are the kiln waste slurry,
the absorber tails from the condensate (drip acid) recycle system, and
the ejector water which is used  to  quench  the  off-gases  from  the
absorber.   All three of these waste streams are collected in a common
neutralization pit where lime slurry is added.  The waste  then  flows
into  a  series  of  three  lagoons  for  solids  removal and final pH
adjustment prior to discharge.  Most of the gypsum settles out in  the
first   lagoon  and  the  overflow  enters  the  second  lagoon  where
commingling  with   wastes   from   other   processes   takes   place.
Verification  sampling  data  from  this plant were obtained from four
sampling points.  These  are:   1)  the  kiln  waste  slurry,  2)  the
absorber  tails,  3)  the  ejector water, and 4) the effluent from the
first lagoon.  The fourth sampling point is the last  point  at  which
all wastewater originating in the HF plant can be intercepted.

Summary of the Toxic Pollutant Data

Eleven  toxic  pollutants  were found in the raw waste samples from HF
Plant #705.  They were also verified at three other typical HF  plants
practicing  BPT~- .treatment.   No organic toxic pollutants were found at
detectable levels.\ The results were:

            ( Maxlmum\Raw Waste Concentrations Observed
            \       1         Ug/1)
                             Screening
                           Plant #705
    Verification
Plants #705,
  #251, #167
Copper
Lead
Selenium
Zinc
Antimony-
Arsenic
Cadmium
Chromium
Mercury
Nickel
Thallium
770
5200
25
8100
70
10
2.0
73
2.0
150
5.5
600
200
230
11000
2800
160
60
1200
43
2000
63
Section 5 of this report describes the methodology  of  the  screening
and verification sampling program.  In the Hydrofluoric Acid industry,
a  total  of 12 days of sampling were conducted at Plants  I 705,  #251,
                                  278

-------
TABLE 12-10. FLOW AND POLLUTANT CONCENTRATION DATA OF THE SAMPLED WASTE STREAMS
             FOR PLANTS #705, §251, AND #167 PRODUCING HYDROFLUORIC ACID

Plant
#705 (2)





#251 C2)





#167






Stream
No.
1
2

4

5
5

6

2
3
1
2

3

4

Verification Data
Sanpled 3 Flow
Stream (m /kkg of HF)
Description
Kiln Slurry
Scrubber Waste
Water
Surface Drains
Cooling Tower
Slowdown
Treated Effluent
AHF Plant
Hosedown
SO- Scrubber
Waste
Gypsum Pond Inlet
Gypsum Pond
Outlet
Kiln Slurry
Ejector & Absorber
Unit Wastes from
Kilns #1,#2, and
#4
Ejector & Absorber
Unit Wastes from
Kilns #5 and #6
Effluent from
First Lagoon
26.6
10.0

20. Q

23.3^
1.20

14.4

84.7
84.7
122
25.0

14.6

162

(1)
Fluoride
(kg/kkg of HF)
3.8
1.5

3.4

0.54
1.9

0.31

58
27
4.9
14

20

11

Total
Suspended
Solids
(fcg/kkg of HF)
4700
0.019

4.0

0.040
0.26

0.10

3800
0.80
170
0.36

0.41

22

 (1)  Three 24-hour composite samples of each waste water stream.
 (2)  These plants have now discontinued their HF production.
 (3)  Consists of the combined flow from hydrofluoric acid and aluminum
     fluoride plants.
                                     279

-------
00
o
                             VENT
                             DUST
                          COLLECTION
               WET
               SPAR"
      •ARI  L»-
 SPAR
DRYING
                HOSE DOWN
                  WATER
                r
                    HF KILN
                                       AIR
       HANDLING
        LOSSES
    -t
                                        DRIP
                                        AC1D

                                     WATER
                                                          SLURRY
                                                        TREATMENT
                    e
                             LEGEND
SAMPLING POINTS.
                                                         AlFj PRODUCT
                                                                                      LIQUEFACTION
                                                                       ON *J
                                                                                       AHF
                                                                                   PURIFICATION
                1
                                                                                                     AHF
                                                 DILUTION WATER
                                                                                             DOWN WATER
                                                                                           AHF PLANT
                                                              #2
                                            GYPSUM
                                             POND
^
                                                                                             NEUTRALIZATION
SYSTEM
                                                                          ALKALINE STREAMS AND
                                                                        ACID FROM OTHER PLANTS
                    Figure  12-4.  General process flow diagram at Plant #251 showing the sampling points
                                  Hydrofluoric acid manufacture.
                                                                                                                    WATER
                                                                                     EFFLUENT

-------
and #167.  Sixteen different sampling points  were  involved  covering
the  raw  waste source, the various raw waste streams,  and the treated
effluents at these plants.  The evaluation of toxic metal  content  of
these  process  related waste streams was based on 572  analytical data
points.  The screening for toxic  organic  pollutants  at  Plant  ft705
generated  an  additional  635  analytical data points.  The daily raw
waste loads were calculated from the waste stream flow  rates  measured
or  estimated  at  the  time  of  sampling  and the measured pollutant
concentration.

     That is,

     Daily loading {as kg of pollutant per day)  = (C)  (Q)
                                                    1000

     where:

     C is the concentration of the pollutant  expressed  in  units  of
     mg/1 (Note:  kg/m3 = 1000 mg/1), and

     Q   is  the  waste  stream  flow rate expressed in unit of m3/day.
     (m3, a cubic meter,  is equal to 264.2 U.S. gallons)

Similarly,  the  unit  loadings  were  calculated  from  the  reported
hydrofluoric acid production rate, the waste stream flow rate, and the
measured pollutant concentration.

     Unit loading ( as kg of pollutant -  (C)(Q)
     per kkg of hydrofluoric acid)        1000  (P)

where  C  and  Q  are  the  same  as  described  above,  and  P  is the
hydrofluoric acid production rate expressed  in units of kkg/day.   (Jckg
is 1000  kg, a metric ton, which is equal  to  2205 Ibs.)

The minimum, average,  and maximum values  are based on  data from   those
plants   where  the  particular  pollutant  was found at concentrations
greater  than the analytical detection limits and significant  in   that
it  could  conceivably be treated by an available treatment technology
regardless of economic considerations.

In Table 12-11, the toxic pollutant raw waste  data  are  presented  as
the  average  daily  concentrations  and  the unit loading found  at the
individual plants.  The overall  averages  are  also   shown  and   were
subsequently  used  in  the calculations  of  the average daily loadings
and the  average unit loadings shown in  Table  12-12   along  with the
corresponding minimum  and maximum values.

Based  on  the  total  annual  production  of  this subcategory and the
average  waste load generated per unit  product,  the   estimated   total
toxic    pollutant  raw  waste  loads  generated  each  year  for   this
subcategory are as follows:
                                   281

-------
                Pollutant	Waste Load (kg/year)

                Copper                    6600
                Lead                     10000
                Selenium                   260
                Zinc                    110000
                Antimony                  8900
                Arsenic                   1400
                Cadmium                     79
                Chromium                  4700
                Mercury                    130
                Nickel                   10000
                Thallium                   840

Pollution Abatement Options

Toxic Pollutants of Concern

Toxic pollutants in raw wastewaters and slurries  typical  of  the  HF
industry  include  the  heavy  metals  often  found  as  impurities in
fluorspar.  These metals are zinc, .lead, nickel,  antimony,  chromium,
arsenic, copper, and selenium.  Raw wastewaters from plants practicing
dry  disposal of kiln wastes may include some of the same heavy metals
in scrubber and area washdown  wastes,  but  in  considerably  smaller
amounts,  since  the spent ore is hauled as a solid waste and bypasses
the wastewater treatment  facilities.   Although  the  fluorosulfonate
anion  is  found  in HF wastes containing drip acid, organic compounds
are not anticipated in  wastewaters  from  this  industry.   No  toxic
organic pollutants were found at significant levels.


Process Modifications and Technology Transfer Options

A.   Gypsum produced in the kiln can be disposed of as a  solid  waste
     instead  of  being slurried with water and sent to the wastewater
     treatment facility.  The solids in this case are mixed with  lime
     and  stored  in  piles  on  the  land  surface  until alternative
     disposal methods are found or the site abandoned.   Although  the
     dry  disposal method is labor intensive (involving transportation
     and landfill operating cost),  it  has  been  found  to  be  less
     expensive  due  to  the  reduced  initial capital requirement and
     operating costs relative to the wet slurry method which  requires
     a   more   extensive   system   of   pipes,  pumps,  and  on-site
     impoundments.

B.   The use of soda ash in place of lime for neutralization has  some
     advantages.   It  eliminates  or  reduces  the  problem  of scale
     formation  in  the  pipelines  and  scrubbers  when  the  treated
     wastewater  is  recycled.   It  offers a faster reaction time and
     better control of pH than lime.  Even though the cost of soda ash
     is higher than lime, soda ash has been found overall to be a less
     expensive alternative at some plants.  One plant reported that  a
     combination  of  brine and soda ash has been found to present the
                                 282

-------
          TABLE 12-11.  TOXIC POLLUTANT RAW WASTS DATA.
SUBCATEGQRY:  HYDROFLUORIC ACID
                                                                   (1)
Average Daily Pollutant Concentrations and Loadings at Plants Sampled
                             (rog/l)

Antimony
Arsenic
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Selenium
Tnallium
Zinc
(i
1705 (S)
0.018
0.0010
0.051
0. 0029
0.0014
0. 000080
0.062
0.0035
0.41
0.023
2.47
0.14
0.00090
0.000050
0.062
0.0035
0.0070
0.00040
*
4.0
0.23
kgAkg of Anhydrous HP)
$705 (V) *23L(V>
0.010
0.00057
* J-
*
0,0060
0.00034
0.26
0.015
0.26
0.015
0.044
0.0025
0.0053
0.00030
0.48
0.027
*
*
*
0.21
0.012
0.12
0.010
0,11
0.0091
*
0.47
0.040
0-12
0.010
0.059
0.0050
0.018
0.0015
1.18
0.10
0.017
0.0014
0.039
0.0033
0.28
0.024
ti67(\n
0.74
0.12
0.028
0.0046
0.0030
0.00047
0.074
0.012
0.32
0.051
0.062
0.010
0.0010
0.00016
0.15
0.025
0,0074
0.0012
0.019
0.0030
8.2
1.3
Overall
Average
0.22
0.033
0.062
0.0055
0.0035
0.00030
0.22
0.018
0.23
0.025
0.66
0.039
0.0060
0.00050
0.47
0.039
0.011
0.0010
0.029
0.0032
3,2
0,41
S - Screening data from one 72-hour composite  sample of
    individual or combined raw waste streams.
V - Verification data from three  24-hour composite samples, averaged,
    from each raw waste sampling .point.
* - Concentration below significant level.
(1) The methodology of the sampling program  is described  in Section
    5.1.2, and Section 12.3.3 presents the scope of sampling  in  the
    Hydrofluoric Acid industry.
                                   283

-------
      TABLE 12-12.  SUMMARY OF RStf WASTE LOADINGS FOUND IN
               SCREENING AND VERIFICATION SAMPLING
SUBCATEGORY:
Pollutant
Toxic
Antimony
Arsenic
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Selenium
Thallium
Zinc
Conventional
HYDROFLUORIC
ACID
Daily
loadings
(kg/day)
Minimum Average Maximun

0.023
0.012
0.0031
0.15
0.60
0-10
0.0021
0.14
0.016
0.16
0.49

2.0
0.50
0.'014
1.7
1.4
1.8
0.057
4.1
0.093
0.31
21

6
1
0
5
2

.4
.2
.025
.4
.80
5.4
0
14
0.
0.
72
.21

20
45

Minimum

0.
0.
0.
0.

00057
00030
000077
0035
0.0096
0.
0.
0.
0.
0.
0.
0025
QOQ05Q
00035
00040
0030
012
Unit
Loadings No. Of
(kgAkg) Plants
Average Maximum Averaged*

0,
0,
0.
0.
0.
0.
0.
0.
0.
0.
0.

034
0055
00030
018
025
039
00050
039
0010
0032
41

0.
0.

12
0090
0.00047
0.
0.
0.
0.
0.
0.
0.
1.
040
051
14
0015
10
0014
0033
3

4
3
3
4
4
4
4
4
3
2
4
& Nonconvetxtional
TSS 190000
Fluoride
13
310000
2900
520000
7900
3800
8.
8
4200
34

4800
58

3
4
* Only those plants **iere the pollutant was observed  at treatable
  levels were included.
                                   284

-------
     best alternative for operation of  the recycle system at a minimum
     cost.   After the use of soda ash,  the treated effluent water  can
     be  totally  recycled,   either to  the scrubber or to the kiln for
     transportation water for the gypsum.

     As the pH approaches 6,  sodium  in  soda  ash  replaces  calcium
     present  in  the  gypsum waste.  This frees enough calcium ion to
     precipitate fluoride as calcium  fluoride.    Where  the  scrubber
     water  is  the predominant source  of  wastewater,  the water has to
     be treated first with enough  lime  to  precipitate  fluoride  as
     calcium  fluoride.   Soda ash can then be added to the supernatant
     to precipitate calcium followed by  neutralization  with  HC1  to
     reduce scaling problems.

C.   Two out of a total of 11 plants manufacturing  hydrofluoric  acid
     send  the  drip  acid  to the wastewater treatment facility.  The
     rest of the plants recycle it to the reactor.  When discharged to
     the waste treatment system,  the  fluorosulfonic  acid  does  not
     hydrolyze  and  leaves  with  the  treated  effluent as a complex
     fluoride in soluble form.  The total   fluoride  concentration  of
     the  effluent will be higher for the plants discharging drip acid
     compared to those which do not,  after  the  same  neutralization
     treatment.   The two plants discharging drip acid to waste looked
     into the feasibility of returning it to the kiln, but because  of
     the  unique design of the kilns, they found it to be economically
     unattractive.  Bench scale studies have shown that the drip  acid
     can be hydrolyzed to free the HF.

     HS03F + H20 + heat = HZS04 + HF                              (5)

     The  two plants not returning the drip acid to the kiln should be
     able  to  hydrolyze  the  material   in  a  separate  unit  before
     commingling  it with other wastes, thus avoiding the treatability
     problem associated with complex fluorides.


Best Management Practices

A.   Runoff can be collected from raw material  and  product  storage,
     process,  and impoundment areas.  It should be treated with other
     process waste at the wastewater treatment facility.  Leachate and
     permeate control needs  to be practiced on the solid waste  stored
     in  many  plant  premises  as  gypsum piles.  There  is a risk that
     uncontrolled stockpiling may contaminate the local ground water.

B.   Ponds designed for solids removal must be deep enough to  have   a
     minimum  of disturbance from wind and rain.  In those areas  where
     the rainfall rate exceeds yearly evaporation, the  collection  of
     runoff   from   raw   material,  product  storage,  process,  and
     impoundment areas may  lead to  serious  water  balance  problems.
     Recycle  ponds  would   have  to be designed to handle this excess
     loading.
                               285

-------
Prevailing Control and Treatment Practices

Plant #705 combines the hydrofluoric acid wastes, including the gypsum
slurry,  with aluminum fluoride waste.  The combined wastewater,  after
neutralization,  is  sent  to settling lagoons before discharge.  This
plant was visited in both the screening and verification phases of the
project and a fuller description of waste treatment practice is  given
above.

Plant  #837  combines  the  gypsum  slurry  and  plant  area  hosedown
wastewater with the equipment washings, leaks, and spills  etc.   from
the  aluminum  fluoride  plant  and  neutralizes  them with lime.  The
solids are removed in settling ponds before discharge.  The wastewater
from scrubbers of both hydrofluoric acid and aluminum fluoride  plants
is sent to an adjoining facility for use.

Plant  #251  also combines the hydrofluoric acid and aluminum fluoride
wastewater.  The suspended  solids  in  the  combined  wastewater  are
removed  in  the  gypsum ponds.  The overflow from the gypsum ponds is
neutralized and  the  pH  adjusted  with  the  wastewater  from  other
products  which  are  manufactured  on  the site.  The plant is in the
process of installing a new proprietary treatment process  to  further
reduce the fluoride in its wastewaters.

Two  plants,  #120  and  #987,  dispose of the kiln residue as a solid
waste after lime addition.  The wastewater in both  cases  is  treated
with  lime  and  the  solids are separated; in one case in a clarifier
followed by a filtration, and in the other by lagooning.

At Plant #167, the  combined  wastewater  (including  the  gypsum)  is
neutralized with lime and then settled in lagoons before discharge.

Plant  #722  practices  complete recycle.  The gypsum slurry, scrubber
water, and other wastewaters are combined and treated  with  soda  ash
for  neutralization.   The  neutralized solution is settled in lagoons
and then is recycled to the scrubbers and to the kiln  to  slurry  the
gypsum.

Plant  #426  disposes  of  the  gypsum solids from the kiln as a solid
waste after lime addition.  The scrubber water is used to make another
product.  The noncontact cooling water is  neutralized  when  required
with caustic soda and settled before discharge.

Advanced Treatment .Technologies

Although  alkaline  precipitation, sulfide precipitation, the xanthate
process, and ion exchange might be applied to clarified solutions  for
control of metal ions, only alkaline precipitation can be readily used
for  slurried  kiln  wastes from HF production.  Sulfide precipitation
from cleared solutions could be used to provide additional removal  of
zinc, lead, nickel, and copper and to a lesser extent, antimony.
                                  286

-------
Selection of Appropriate Technology and Equipment

Technologies for Different Treatment Levels

A.   Level 1 (BPT and BAT)

     Neutralization with lime, used widely  in  the  HF  industry,  is
     shown  as  the  BPT  treatment, principally to control pH and the
     nonconventional pollutant fluoride -  which  is  precipitated  as
     calcium  fluoride.   Sufficient  settled  effluent  may be reused
     (0-35 percent) to transport kiln waste to the treatment  facility
     as  a slurry, and the remainder is adjusted to a pH between 6 and
     9 before discharge.  The flow diagram is shown  in  Figure  12-5.
     For BAT, 65 percent effluent reuse is incorporated into the Level
     1 treatment system.

B.   Level 2

     Treatment is alkaline precipitation  using  additional  lime  and
     close  control  of pH in second-stage neutralization, followed by
     lagoon  settling.   Sufficient  lagoon  effluent  is  reused   to
     transport  kiln  waste  to the treatment facility as a slurry and
     the  remainder  is  filtered  to  remove  finely  divided   metal
     hydroxides.  The flow diagram is shown in Figure 12-6.

C.   Level 3

     It  is assumed that 65 percent of the Level  2 effluent  is  reused
     for  transporting spent kiln waste to the treatment facility.  For
     the  remaining 35 percent, pH adjustment and sulfide precipitation
     are  used  ahead  of the Level 2 dual-media filter, to react with
     residual lead, copper, nickel, zinc, and antimony which  may  not
     have  reached their optimum pH levels for alkaline precipitation.
     The  flow diagram for this treatment is shown in Figure 12-7.

D.   Level 4

     As  an alternative to Level 2, Level 4 employs soda ash instead of
     lime for neutralization, depending on the spent  ore  to  contain
     enough  calcium to precipitate calcium fluoride.  Use of soda ash
     permits increased effluent recycling without the scaling problems
     associated with calcium  sulfate.  To control salinity and  sodium
     alkalinity,  a  final effluent blowdown of  at least  10 percent of
     the  influent rate  is maintained.  The common heavy metals will be
     precipitated as carbonates and hydroxides with varying degrees of
     effectiveness  at  pH  levels  attainable   with  soda  ash.   The
     effluent   is filtered and adjusted to a pH  between 6 and 9 before
     dicharge or process recycling.   (Figure 12-8.)

E.   Level  5

     Level 5 treatment  is dry handling and off-site chemical  landfill
     for  the  kiln  waste  and  two-stage alkaline precipitation with
                                   287

-------
     clarification and  filtration  for  the  liquid  process  wastes.
     Heavy metal precipitation with soda ash permits partial recycling
     for uses other than slurry transport.  (Figure 12-9.)

     Incremental  cost  and  performance estimates for the alternative
     levels of treatment and control were evaluated in detail as  part
     of the rule-making process which lead to the proposed regulations
     {45  FR 49450, July 24, 1980).  This material is presented in the
     proposed Development Document (60).

Equipment for Different Treatment Levels

A.   Equipment Functions

     Level 1, typical of existing treatment facilities, utilizes  very
     little  equipment, but depends on lime neutralization in settling
     lagoons, with final pH adjustment.

     In Level 2, conventional dual-media filtration is  added  to  the
     Level 1  system.  In Level 3, standard reagent mixing and solution
     feeding units are added to the Level 2 system.  In Level 4, which
     is  an  alternate  to  Level  2,  the  same type of chemical feed
     equipment is used for soda ash as was used for lime in  Level  2.
     Conventional  lagoons and dual-media filters are used in Level 4,
     but special attention  to  selection  of  materials  is  required
     because  of  the high salinity of recycled effluent.  In Level 5,
     dry kiln waste disposal  is  recommended  with  conventional  dry
     solids  handling equipment.  Lagoons, clarifiers, and filters are
     used for scrubber, noncontact cooling,  .and  other  miscellaneous
     liquid  wastes.  In this case, equipment for storing and handling
     the dry kiln waste is not considered to be wastewater  treatment,
     and the cost is not included in the cost estimates.

B.   Chemical Handling

     Lime (as CaO) is the major chemical used in Levels  1 and 2, along
     with minor amounts of hydrochloric acid for final pH  adjustment.
     With normal precautions, these chemicals pose no .special hazards.
     In  Level 3, ferrous sulfide is prepared on-site by mixing sodium
     bisulfide and ferrous  sulfate.  Although  sodium   bisulfide  can
     release  toxic  HZS  at  pH  levels  below  7,  the hazard can be
     mitigated by avoiding acid conditions and by  providing  adequate
     ventilation.   After  mixing  its components, the ferrous sulfide
     solution is stable at the pH levels employed in the process.   In
     Level  4,  only  sodium carbonate and hydrochloric  acid are used,
     with no unusual safety hazards or special handling problems.   In
     Level  5,  only  lime,  soda  ash and hydrochloric  acid are used,
     introducing no special problems of safety or handling.

C.   Separation and Removal of Solids

     Solids are accumulated in unlined settling lagoons.  In Level  4,
     calcium  fluoride  will  still precipitate in the lagoons but the
                                    288

-------
OD
                          UME
                       RAW
                                                                 RECYCLE FOR SLURRY TRANSPORT
                                                                                      LAGOON
                                                   <*>
                   WASTE WATER
-»\   JQ    /'—"»
                                  EQUALIZATION
                                                   """"""PjCy'*™™™^'^
                                                                        U.
                                                                  MIXING

                                                                                      LAGOON
                                                                                                  ^^*****PP
                                                                                                             pH ADJUSTMENT
                                                                                                                     EFFLUENT
                 Include! flow monitoring, pH monitoring and tamplar
                             12-5.   Level 1 waste water treatment  for hydrofluoric acid subcategory.

-------
vO
o
f— nukx
LJ-*n




HAW
WA6TK WATCH





kl/
KOUAIIZATION

r~-©
i
i
I
db




	 1
1
|i

M1UHO .

RCCVCLB FOB
0U4VBV TRANSVOBl
.A t'AlUOH





                                                                                   ~1
                                                                            nr*itj
                                                                   -H_h
                                                                      BUMP LQJ
fii
1WTMKMT
                                                                                    X
                                                                              Uttw HUM!luring.
                   Figure 12-6.  Level 2 waste water treatment for hydrofluoric acid subcategory

-------
to
\O
H
                        h^Qn
                                     z
                                                      J^
fU Kousaan
  0
                                                 MI10MB
                                                                               •Lh-a^1
                              ln«luile> How moolktrlnf, pH nWnllutln) and uuu
                        Figure 12-7.  Ijevel 3 waste water treatment for hydrofluoric acid subcategory.

-------
                                 r
i
1 IttMAM
M

,




w3w
IKTU

«•)




_ 	 /
	 ^MMLuOnM 	 1



1 — ^






f

UXIW




t
Li
*





._ A tiflOM L^
n n
•
'

:

—^t UUUGM /.»




	 »j [M

















«




1

.rim

X
T "


K



1



HJ
'





g)

Wtft



VC
^V
                lsiitom ftft* munMadtt), ptl
Figiore  12-8.  Level 4 vraste water treatment for hydrofluoric acid subcategory

-------
     total sludge quantities will be less than in Levels 1,  2,   and  3
     where  lime  is  used.    Solids  from  Level  4 treatment  will be
     alkaline, very saline,  and difficult to consolidate.   Dry   solids
     from the Level 5 model  are not subjected to treatment,  except for
     nominal  application  of  lime  before  hauling in dry form to an
     approved chemical landfill.

Treatment Cost Estimates

General Discussion
To prepare
developed.
treatment  cost  estimates,   a  model  plant  concept  was
The BPT model treatment consists of:
A.   Slurry transportation of kiln solids to an equalization basin.

B.   Application of lime to precipitate  fluoride  and  toxic  metals,
     followed by lagoon settling.

C.   pH adjustment before final discharge.

D.   Scrubber, cooling, and distillation wastes enter the equalization
     basin.

It is assumed that drip acid is recycled to the  process  reactor  and
does not appear directly in the waste stream.

In the BAT treatment, 65 percent of the BPT treated effluent is reused
to transport kiln wash to the treatment facility as a slurry.

Wastewater Flow

The  data  in  Table  12-4 for plants sending the gypsum solids to the
treatment  facility  indicate  that  the   unit   flow   varies   from
approximately  31.0  mVkkg  of HF to 166 mVkkg of HF.  For the model
plants, a constant unit flow of 95.4 mVkkg of HF is assumed.

HF Production

In the HF subcategory, production  ranges  from  a  minimum  of  7,300
kkg/year  to  a  maximum  of  62,000  kkg/year  with  a mean of 22,100
kkg/year and a median of 15,800 kkg/year.   For  wastewater  treatment
cost estimates, three production levels were selected as model plants.
These are 19,100 kkg/year, 38,200 kkg/year, and 57,300 kkg/year.

Wastewater Pollutant Load

The  amount  of  kiln  residue  varies  from  3.8  to  4.1 kg/kg of HF
produced.  The wastewater going to treatment model plants  is  assumed
to  contain  3.8  kg  of  solid  kiln  residue per kg of HF.  Fluoride
emissions in wastewater have been shown to vary as indicated below:
                                293

-------
                                                                              H» ADJUSTMENT
UME
                     UXK
            WLN
           WASTE
          |*-LID
**r •

i i X 1-AGOON
J- LAGOON
                             TO LAt.TlbtLI.
               Include* On* mttMtlarlni, pit nMnilt«rln| fend
       Figure  12-9.   Level  5 waste water treatment for hydrofluoric acid subcategory

-------
          Source of Data
Fluoride, (kq/kkq)
          Reference 3
          Reference 3
          Screening and Verification
          Phase Sampling
          (Tables 12-9 and 12-10)
        20
        37
    3.8 to 58
For the model plants, the average fluoride loading from kiln wastes of
31  kg/kkg HF produced was used to establish treatment requirements and
related costs.

The costs shown at each level of treatment  correspond  to  the  model
plant  BPT  system   (Level  1)  and the BAT system which may add to or
modify the existing  BPT  system  to  meet  more  stringent  pollutant
removal requirements.

Treatment  costs  at  Levels  2, 3, 4, and 5 are given in the proposed
Development Document (60).

The estimated costs  for  three  models  having  different  production
levels are given in Tables 12-13, 12-14, and 12-15.  For these models,
both  the hydraulic and the pollution loads per unit of production are
held constant over the entire range of production.

Table 12-16 presents a summary of the unit cost  distribution  between
amortization  and operation and maintenance cost components at various
production rates and levels of treatment.

At the second level of treatment, the cost estimate  is  based  on  65
percent of the wastewater flow being recirculated.

Model Plant Control Costs for Existing Sources

For  the model plant control costs for existing sources at the BPT and
BAT levels of treatment, the disposal of the  sludge  is  on-site  and
hence  the  land  requirements  are  fairly  large.  Chemicals, sludge
hauling, and disposal costs have a  signficant  impact  on  the  total
annual costs.

Basis for Regulations

Evaluation of BPT Treatment Practices

Control  and  treatment  practices  for eleven plants producing HF are
presented in Table 12-17.  Also indicated  are  other  product-related
wastewater sources and pollutant loads discharged.
                                  295

-------
It  is  clear from the table that a wide variation in effluent quality
exists within this subcategory.  The factors believed to  cause  these
variations are the following:

A.   Dry Residue Handling

     The disposal of kiln waste by dry handling rather than  slurrying
     is  practiced currently at three plants.  This process eliminates
     the major source of wastewater generated at most plants,  greatly
     reducing  the raw waste loads to be treated.  The only sources of
     wastewater remaining are from air pollution control and washdown.

B.   Effluent Reuse

     Reuse of treated wastewater for slurry transport of  kiln  wastes
     is  commonly practiced to varying degrees and clearly has a major
     effect on pollutant loads discharged.  Although  four  plants  do
     not  practice  reuse, the fact that five plants do practice reuse
     demonstrates  that  the  practice  is  both  technologically  and
     economically feasible.

C.   Recycle of Condensables

     Recycling of drip acid or condensable cooler bottoms reduces  the
     loading  of  fluoride  in the treated effluent since the fluoride
     species (fluorosulfonic acid) in this material is not removed  by
     conventional lime treatment.  Only two plants do not recycle drip
     acid.

D.   Other Related Products

     Most hydrofluoric acid plants also discharge wastes from  related
     products    such    as    aluminum    fluoride,    fluorocarbons,
     hexafluorosilicic and tetrafluoroboric acids to treatment.  These
     other product wastes can account for higher  raw  waste  loadings
     and  increase  the  potential for complex fluorides formation and
     can also impact treatment efficiency by diluting the  raw  waste.
     In  addition,  commingling of other product wastes will limit the
     percentage of reuse of the total plant treated effluent.

     In addition to the above factors, the design and operation of the
     treatment facilities affect the effluent quality.  Solids removal
     depends on retention time and surge capacity.   Precipitation  of
     fluoride  requires  careful  pH  control  and  in  areas of heavy
     rainfall or winds,  adequate  freeboard  or  multiple  ponds  are
     necessary  to  limit the discharge of high pollutant loads due to
     unfavorable climatic conditions.


E.   Pollutant Removal with BPT Treatment

     Treatment level 1 is  BPT  in  the  Hydrofluoric  Acid  industry.
     Table  12-18  presents a summary of long term effluent monitoring
                                  296

-------
             TABLE 12-13. MODEL PLANT TREATMENT COSTS
Subcategory    Hydrofluoric acid
Production     19,100 metric tons per  year
A. INVESTMENT COST

     Site development  ......
     Equipment 	
     Monitoring equipment  ,.

          Subtotal  	
     Contractor's 0 &  P b....

          Subtotal  	
     Engineering  	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL INVESTMENT  COST .


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision .
     Energy 	
     Chemicals 	
     Maintenance  	
     Taxes and insurance  ...
     Residual waste disposal
     Monitoring ,  analysis
      and reporting 	
      BPT

  410,000
  429,500
   20,000

  859,500
  128,925

  988,425
  197,685

1,185,110
  lift, 511

1,304,721
1,020,000

2,324,721
     TOTAL OPERATION AND
     MAINTENANCE COST
   55,000
   14,000
  534,300
  130,472
   59,742
  350,000

   15,000
1,170,014
   BAT

     0
45,000
     0

45,000
 6,750

51,750
10,350

52,100
 5,210

58,310
     0

58,310
14,000
 1,500
     0
 5,831
 2,049
     0

 7,500
31,880
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
  212,273

1,382,292
11,114

42,994
  Represents the  incremental  cost  above that for BPT treatment
  Overhead and Profit
                              297

-------
             TABLE 12-14.  MODEL PLANT TREATMENT COSTS
Subcategory
Production
Hydrofluoric acid
38,200 metric tons  per  year
A. INVESTMENT COST

     Site development  	
     Equipment  	
     Monitoring equipment  ..

          Subtotal  	
     Contractor's 0  5t  P*3....

          Subtotal  	
     Snginhering  	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL INVESTMENT  COST .


B. OPERATION AND
   MAINTENANCE  COST

     Labor and  supervision .
     Energy  	
     Chemicals  	
     Maintenance  	
     Taxes and  insurance  ...
     Re s i d ua 1 wa st e  d i spo sa1
     Monitoring ,  analysis
      and reporting  	
                                BPT

                            735,000
                            551,500
                             20,000

                          1,306,500
                            195,975

                          1,502,475
                            300,495

                          1,802,970
                            180,297

                          1,991,2*7
                          1,940,000

                          3,923,2*7
                             5*,000
                             19,500
                          1,0*9,000
                            193,327
     TOTAL OPERATION  AND
     MAINTENANCE  COST
                            700,000

                             15,000


                          2,175,525
                                                     ($)
    BATa
      0
 70,000
      0

 70,000
 10,500

 80,500
 1*,100

 9*,*00
  9,**0

10*,2*0
      0

10*,2*0
 14,000
  3,100
      0
 10,*?*
      0

  7,500


 38,414
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                            322,*78

                          2,498,202
 17,2S9

 55,702
a Represents  the  incremental  cost above that for BPT  treatment
k Overhead and Profit
                              298

-------
             TABLE 12-15. MODEL PLANT TREATMENT COSTS
Subcategory
Production
Hydrofluoric acid
57,300 metric tons per year
A. INVESTMENT COST

     Site development  	
     Equipment  	
     Monitoring equipment  	

          Subtotal  .. *... .....*..
     Contractor's 0  &  Pb	

          Subtotal  	
     Engineering  	.	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL  INVESTMENT COST 	


B. OPERATION  AND
   MAINTENANCE  COST

     Labor  and  supervision 	
     Energy	,	
     Chemicals	
     Maintenance  	
     Taxes  and  insurance 	
     Residual waste  disposal  ....
     Monitoring,  analysis
      and reporting  	

     TOTAL  OPERATION AND
     MAINTENANCE  COST
                                                     ($)
                               BPT

                          1,050,000
                            928,000
                             20,000

                          1,998,000
                            299,700

                          2,297,700
                            459,540

                          2,757,240
                            275,724

                          3,032,964
                          2,890,000

                          5,922,964
                             56,000
                             28,000
                          1,604,000
                            303,296
                            177,689
                          1,050,000

                             15,000
                          3,233,985
    BAT0

      0
100,000
      0

100,000
 15,000

115,000
 23,000

138,000
 13,800

151,800
      0

151,800
 14,000
  4,600
      0
 15,180
  4,554
      0

  7,500
 45,834
C. AMORTIZATION OF
   INVESTMENT  COST

   TOTAL ANNUAL COST
                            493,463

                          3,727,449
 24,698

 70,532
a Represents  the  incremental  cost above that for BPT treatment
y.
  Overhead and Profit
                              299

-------
        TABLE 12-16.MODEL PLANT UNIT TREATMENT  COSTS
     Subcategory Hydrofluoric acid
                                Annual Treatment  Costs  ($/kkg)
    COST ITEM
PRODUCTION
 (kkg/yr)
  LEVEL OF TREATMENT

BPT       BAT*
Annual Operation
and Maintenance


Annual
Amortization


Total Annual
Cost



19,100
38,200
57,300

19,100
38,200
57,300

19,100
38,200
57,300

61.26
56.95
56.44

11.11
8.45
8.61

72.37
65.40
65.05

1.67
1.01
0.80

0.58
0.45
0.43

2.25
1.46
1.23
*Represents the incremental cost above  BPT
                             300

-------
data on total suspended  solids  (TSS)  and  fluoride  from  four
plants.   Means, standard deviations, and variability factors are
given where sufficient data  are  available.    These  performance
characteristics  are  later  utilized  for the development of the
regulations on TSS and fluoride.

The ability of BPT treatment to remove toxic  pollutants  can  be
estimated  by  comparing  the  raw  waste data presented in Table
12-11 with the corresponding treated effluent data  presented  in
Table  12-19.  The latter expresses the removal efficiency as the
calculated average percent removal observed at these plants.  The
BPT  removal  efficiency  for  some  of  the  toxic   metals   is
undoubtedly  augmented  to  some  degree by the fact that the raw
waste may carry insoluble forms of the  metals  that  were  never
completely  leached out of the ore.  Removal of these forms would
take  place  simply  by  settling  out;  however,  the   effluent
concentrations  of some metals such as chromium, nickel, and zinc
remain at concentrations higher  than  should  be  achievable  by
alkaline  precipitation.   This  suggests  that  these metals are
largely in solution coming into the treatment system and that the
optimum conditions for metal hydroxide formation were  not  being
attained at the time of sampling.

The  original BPT limitations for this subcategory shown in Table
12-2 required zero pollutant discharge except during  periods  of
excess  rainfall.   Objections  to the zero-discharge limitations
concerned the feasibility of  using  gypsum-saturated  water  for
reuse  in the air pollution control scrubbers.

The  BPT  wastewater  control and treatment technology allows for
the discharge of process wastewater after appropriate  treatment.
This   technology  is  practiced widely in the industry and should
pose no technical problems.  Implementation of BPT at  all  sites
in  the  industry  will achieve the  indicated pollutant discharge
levels.

The nine plants presently producing hydrofluoric  acid  all  have
installed  BPT  treatment  or   the  equivalent.   At  the time of
sampling, seven of the  11  plants  operating  were  meeting  the
proposed fluoride limitations and eight were meeting the proposed
TSS  limitations  according to  the data available (60).  Although
there  was practically no long term monitoring data  available  to
support  the additional proposed limitations on toxic metals, the
screening and verification data indicated that all  three  plants
sampled  were  meeting  the  proposed  limitations  on  antimony,
copper, and lead, while  two  of  the  plants  were  meeting  the
proposed  zinc  limitation and  one plant was meeting the proposed
chromium and nickel limitations.   With  the  limited  amount  of
toxic  metal  data, it was not possible to estimate compliance or
noncompliance on a statistical  basis.  The Agency  has  conducted
additional treatability studies (61) and has utilized the results
of this work in formulating the final regulations as described in
the following sections.
                             301

-------
           TABLE 12-17.  SOHMARY OF WASTE HATER CONTROL AMD TREATMENT TECHNOLOGY EMPLOYED AT HYDROFLUORIC ACID PLANTS
                                                                                                                     (1)

Plant




Product-Related
Waste Hater Sources



Control and
Technology



Treatment
Employed



Amount of
Treated
Haste Hater
Reused

Cooler Bottoms
(Condensablea)
Recycled?


Effluent Volume
in IB* /»etrie ton
(gal/ftbort ton) of
Actual Production
Long Tern
Average Pollutant
Hasteload Discharged
(kg/aetric ton)
{lb/1000 Ib)
  6«4
  167
  120
' 967
  928
! 837
* 753
                                                                                                        Fluoride
I 251
     (21
I 705
     (2)
* 722
         Hydrofluoric acid
         production
Hydrofluoric acid,
floorocarboo,
Chiorine/sodi urn
hydroxide, and
hydrochloric acid
production

Hydrofluoric acid
production
         Hydrofluoric acid,
         fluorocarbon, and
         •olfuric acid
         production
                             Residue slurry, neutra-     94%
                             Illation with •odium
                             carbonate, settling,
                             recycle

                             Residue slurry, lime        47%
                             treatment,settling,
                             recycle
                             Planned dry residue          0
                             handling, li«*
                             treatment, clarification

                             Residue slurry, settling Present:  0
                             (Recycle and pa          Planned!  70%
                             polishing facilities            bo 75%
                             under construction.)
                                                                          5.78  (1,360)
                                                                                   103  (24,200)
                                                                         Tes
         Hydrofluoric acid   Residue slurry, settling,
         and aluminum        recycle {Plocculation,
         fluoride production lime treatment, and
                             clarification facilities
                             under construction.)
                                                         83%
All hydrofluoric
acid generated as
used captively for
aluminum fluoride
production

Hydrofluoric acid
production
                             Residue slurry, lime
                             tr«et»*nt, settling
                             Residue slurry, lime
                             treatment, settling,
                             recycle, pH polishing
                                                                         Yes
                                                         65%
         HF, AIT], chlorine/ Residue slurry, settling,
         sodium hydroxide,   neutralization
         aluminum oxide, and
         fluorocarbon
         production

         Hydrofluoric acid   Residue slurry, line
         and aluminum        treatment, settling,
         fluoride production recycle, pB polishing
                                                                    1 Kiln:  Tea
                                                                    3 Kilns.-  No
                                                      30% to 35%
                                                                                   134  (32,200)
                                                                                   11.0  (2,650)
                                                                          22.2 x 10s
                                                                          (553 x 10*)
                                                                                   25.9  (6,204)
                                                                                                                      TSS
426 Hydrofluoric acid
fluosilicic acids
production
'Dry residue hauling 0
and dumping; neutra-
lization with caustic
of noncontact cooling
water and floor
drainage
Yes 465 (111,397)
includes noncon-
tact .cooling
water
1.2 SD
                                                                                                             0.10    0.27
                                                                                                            18       0.45  (Net)
                                                                                                                                (3)
                                                                          125  (30.000)   Present:   24       16
                                                                                         Expected
                                                                                         with        1.8      2.1
                                                                                         additional
                                                                                         facilities

                                                                          9.44  (2,260)   Present:    1        1.7
                                                                                         Expected
                                                                                         with        0.65     0.75
                                                                                         additional
                                                                                         facilities
                                                                                                             1.8     3.1
                                                                                                              0.64     0.38
                                                                                                            46      530
                                                                                                             3.2      0.64
         Hydrofluoric and,
         in recent peat,
         fluoboric,
         acid production
                    Residue slurry, lime
                    treatment, settling,
                    recycle, pB polishing
                                                      92% to 100%
» 967l '  Hydrofluoric Acid   Dry residue hauling
                                                                         Tes
                                                                                   0-10.3  (0-2,460)         0-0.81    0  to 0.54
                                                                                    8.8
 tl)  Adapted fros Calspen  (Reference 3).
 (2)  Hydrofluoric Acid production has been discontinued at these plants since  the  time of  sampling.
 (3)  Effluent loading less the influent loading.
HD - Not determined.
                                                                  302

-------
Basis for BPT Effluent Limitations

A.   Technology Basis

     For BPT, the Agency is promulgating  limitations  for  which  the
     technology  basis  is,  or  is  equivalent to, equalization, lime
     neutralization/alkaline precipitation,  solids removal by settling
     or thickening, final pH adjustment, and  discharge  of  clarified
     effluent.    The   in-house  process  recycling  of  the  reactor
     condensables  (drip  acid)  is  necessary  to  meet  the  fluoride
     limitations.

B.   Flow Basis

     The reuse of treated wastewater to slurry kiln  residues  to  the
     treatment system is not required for meeting the BPT limitations.
     BPT or  its equivalent is practiced by all plants in this industry
     including six which reuse, for slurrying residues, proportions of
     their   treated  wastewater  ranging from 30 to 100 percent of the
     plant flow as shown in Tables 12-4 and 12-17.

     The practice  of  reusing  wastewater  in  this  manner  has  two
     opposing effects on the plant effluent:

     1.   A  decrease in the net discharge unit flow rate (m3/kkg), and

     2.   An increase in the fluoride concentrations (mg/1).

     As a result,  the fluoride unit loading (kg/kkg) in  the  effluent
     does  not  decrease as a direct proportion to the decrease  in the
     flow rate, but  is  partially  offset  due  to  the  increase  in
     fluoride  concentration  as  a  function  of  percent reuse.  The
     relationship  of percent water reuse  to  fluoride  concentrations
     and  unit loadings is shown in Figure  12-10.  The apparent  reason
     for the increase  in  fluoride  concentration  with  reuse   is   a
     calcium deficiency  which may result  from the buildup of sulfate
     at plants where reuse is practiced.  Other pollutants such  as TSS
     and metals would not be expected to exhibit similar concentration
     offset  effects  in these systems.

     It should be  noted that while the  practice of reusing  wastewater
     for  kiln residue slurrying may be advantageous in some locations
     with respect  to alternative  water supply  costs,  there   is  no
     associated  reduction in the hydraulic load,  size, or cost  of the
     BPT treatment system  itself.

     The net result of water reuse  is  a   moderate  decrease  in  the
     effluent   fluoride   loadings  which  is  achieved  at  a   small
     additional annual cost of less than one percent of the  estimated
     BPT treatment systems cost  (Tables 12-13, 12-14, and 12-15).

     The  model  plant BPT treatment system is based on an inflow rate
     of 95.4 mVkkg derived from the  average  of  nine  plants   which
                                 303

-------
  TABLE 12-18.  StWARY OF LONG TERM "WNITORING DMA FRCM FOUR
                                           (1)
                    tnEROPLUORIC ACID PLANTS
                   Treated Waste toad
                                                  or (Ib/lOOOlb)
Plant
 No.   Etararaeter
                       Daily Data
                 Long Term
                 Average    St.Dev.
                   (X)    (S)  (SM
      30-Day Average Data
(2) Long Term            (2)
W   Average  St.Dev.  W
        (X)      (S)
1664

*753

1722


*705

Fluoride
TSS
Fluoride
TSS
Fluoride
TSS
(3)
Fluoride
TSS
0.
0.
0.
0.
0.
0.

—
—
10
29
72
38
81
54



0.090
—
0.27
—
0.52
0.37

—
—
0.77
—
0.36
—
0.59
0.62

—
—
4.5
—
2.2
—
3.3
3.5

—
—
0.
0.
0.
—
	
—

0.
0.
10
27
64




49
84
0.
—
0.
—
_
— =•

0.
0.
040

15




22
37
1.7
—
1.4
— —
_
—

1.7
1.7
(I)
   Based on Reference 3 data.
(21
In the case of daily measurements, the variability factor, VF,
Cor a lognomal distribution is found by the expression ln(VF) -
S'(Z - 0.5S'), where S' is the estimated standard deviation of
the logarithm derived from the arithmetic mean, X, and the
arithmetic standard deviation, S, according to the relationship,
(S')2 - In  l.O +(_Sj\21.When the value of Z is 2.33, the
           In fl.
              L
variability factor for the 99 percentile is obtained.
For 30-day average measurements, a noraal distribution is
obtained and the variability factor is found by the expression,
VF » 1.0 + Z  ..•  When the value of Z is 1.64, the
                (.£.)•
variability factor is for the 95 percentile.  Please refer to
Section 3.2 for a more detailed discussion of the statistical
of long term data.
                                                                 analysis
(31
   M.though Plant V705 does not recycle the drip acid, the TSS
   data is not adversely affected and is used as the basis for
   the 30-day average VF.

— Not bailable.
                             304

-------
       TABLE 12-19.  TOXIC POLLUTANT TREATED EFFLUENT DATA

SU3CATEGOTX:  HYDROFLUORIC PCID

                                                                    (1)
Average Daily Pollutant Concentrations and Loadings at Plants Sampled
                                (mg/D
                         (kg/kkg of Anhydrous HP)
                                                             (2)
                                                     Overall   Average
            *705{S)    1705 (V)    f251(V)    H67(V) Average  % Removal
Antimony
Arsenic
Cadmium
Chromium
Copper
lead
Mercury
Nickel
Selenium
Thallium
Zinc
<0.010
<0. 00021
<0.0030
<0, 000063
0.00030
0.0000060
0.014
0.00029
0.10
0.0021
0.0060
0.00012
<0. 00040
<0. 0000080
0.050
0.0010
0.033
0.00069
0.0070
0.00015
0.071
0.0015
<0.0020
<0. 000042
<0.010
<0. 00021
<0. 0017
<0. 000035
<0.046
<0. 00096
<0.020
<0. 00042
<0.022
<0. 00046
<0. 00050
<0. 000010
<0.010
<0. 00021
<0.0050
<0. 00010
<0.0012
<0. 000025
0.053
o.oon
<0.17
<0. 017
<0.020
<0.0020
<0.0020
<0. 00020
0.22
0.022
0.070
0.0069
<0.031
<0.0031
<0.0010
<0.00010
0.52
0.052
<0.071
<0.0070
<0.0070
<0. 00069
0.16
0.015
0.047
0.012
0.016
0.0040
0.0087
0.0022
0.050
0.013
0.060
0.015
0.010
0.0026
0.0065
0.0017
0.090
0.023
0.010
0.0026
0.0030
0.00069
1.9
0.49
<0.057
<0.0073
<0.012
<0.0016
<0.0032
<0. 00060
<0.033
<0.0091
<0.063
<0.0061
<0.017
<0.0015
<0.0020
<0. 00044
<0.17
<0.019
<0.030
<0.0026
<0. 0045
<0. 00039
0.55
0.13
74
81
9
9
62
77
97
67
64
Effluent
>Influent
85
«3
 (S) Screening data from one 72-hour composite sample of treated
    effluent.
 (V) Verification data  from three  24-hour  composite  samples.
 (1) The affluent data  presented here  corresponds  to tfte raw waste
    data showi  in Table 12-41.  The methodology of  the  sampling
    program  is  described  in Section 5.1.2,  and  the  scope of
    sampling in the  Hydrofluoric  Scid industry  is described  in
    Section  12.3.3.
 (2) When averaging values indicated as "less than"  (O, the
    absolute value was used and the resulting average was indicated
    as a "less  than" value.
                                 305

-------
     handle  the  kiln  residues  in a slurry system as shown in Table
     12-4.   The treated effluent flow rate is 54.6 mVkkg which is the
     average  effluent  flow  rate  for  the  same  nine  plants   and
     corresponds  to  the  reuse  of  about 43 percent of the flow for
     residue slurrying and other uses.

C.   Selection of Pollutants to be Regulated

     The selection of pollutants for which specific numerical effluent
     limitations are promulgated is based  on  an  evaluation  of  raw
     waste  data from the screening and verification sampling program.
     The following two major factors were considered:

     Raw waste pollutant concentrations - A tabular summary of maximum
     raw waste concentrations is presented in Section 12.   Data  from
     the  one  plant  sampled for screening were used to determine the
     need for verification sampling.  The maximum concentrations found
     during verification are also  shown  for  comparison.   For  each
     pollutant,  the maximum concentration observed gave a preliminary
     indication of its potential significance in the subcategory.   On
     this   basis,   the   preliminary  selection  of  candidates  for
     regulation included zinc, lead, antimony, nickel,  chromium,  and
     copper in decreasing order of their apparent pollution potential.
     These  pollutants were observed at least once during the sampling
     program at concentrations considered treatable in  this  industry
     using one of the available technology options.  The other metals,
     chromium,  thallium, and mercury exhibited maximum concentrations
     that were considerably lower.

     Total subcateqory raw waste pollutant loadings  -  Pollutant  raw
     waste loading data were used to evaluate the overall magnitude of
     the  pollution  potential  for  the  subcategory.   Data from the
     plants sampled are presented in Table 12-11  and  the  daily  and
     unit  loadings  are summarized in Table  12-12.  This information,
     coupled with the estimated  total  hydrofluoric  acid  production
     rate  of  261,800  kkg/year, yielded the approximate total  annual
     pollutant loading rates for the subcategory shown in Section  12.
     This  method  of  ranking the pollution potential of the observed
     toxic metals confirmed the maximum  concentration  based  ranking
     and  indicated  that  zinc,  nickel,  lead, antimony, copper, and
     chromium were the six dominant toxic  metals   in  terms  of  both
     total mass loadings and treatable raw waste concentrations.

     In  view  of the treatment technology already  implemented  in this
     industry, the added BPT regulation of any one  of these pollutants
     may provide assurance that all of the observed toxic metals would
     receive adequate treatment and   control.   This   includes   taking
     credit  for  incidential removal of metals which are either below
     practical treatability limits or are not particularly amenable to
     removal by alkaline precipitation methods.  The  latter   includes
     cadmium,  selenium, thallium, and mercury.  Based on the  argument
     provided  in Section 8, the control of zinc and nickel, having the
     highest loadings and  concentrations  in   the  wastewater,  would
                                   306

-------
     effectively  control  the  other  toxic  metals  in the treatment
     system.   Thus,  zinc and  nickel  were  selected  as  the  control
     parameters of toxic pollutants for BPT regulations.

D.   Basis of Pollutant Limitations

     1.    Conventional and nonconventional parameters

     a.    pH:  The treated effluent is to  be  controlled  within  the
          range  of  6.0 to 9.0.  This limitation is based on the data
          presented in Appendix B of the proposed Development Document
          (60) and the JRB Study (52).

     b.    TSS and Fluoride:  The data presented in  Tables  12-12  and
          12-18  were  used  for  the  development of TSS and fluoride
          limitations.  However, because of the wide range of  product
          mixes,   significant   differences   in   residue  handling,
          wastewater treatment, reuse  practices,  and  dilution  with
          other product waste streams, it was necessary to select only
          those  plants  where  the  effect of BPT technology could be
          clearly observed.  The plants excluded are:
          #426 and 1120 because kiln residues are
          solid.
handled  as  a  dry
          #167,  1967,  and  #251 because the combined treatment of HF
          wastes along with the wastewaters from other major  products
          generated   high  fluoride  loadings  in  the  large  volume
          discharges  with  fluoride  at  its   minimum   treatability
          concentration,

          #705 because cooler bottom condensables {drip acids) are not
          recycled  back to the process but are added to the raw waste
          contributing complex  fluorides  which  tend  to  remain  in
          solution after lime treatment.  TSS data are not affected.

          Data  from  the remaining five plants are presented in Table
          12-20 which summarizes the development  of  the  regulations
          for  total  suspended  solids  and  fluoride.  Since the BPT
          level of treatment does not require  the  reuse  of  treated
          wastewater  for  slurrying kiln residues,  the performance of
          Plant #837 was used as the long term  average  unit  loading
          basis for the TSS and fluoride limitations.  The variability
          factors  used  for  fluoride are based on the long term data
          from Plants 1664 and #753 and those used for TSS are derived
          from Plant #722 for daily measurements and  Plant  #705  for
          30-day average measurements as indicated in Table 12-20.

          The  maximum  30-day  average TSS limitation was obtained by
          multiplying the variability factor for 30-day averages  from
          Table 12-20 by the long term average waste load; i.e., 1.7 x
          3.1 kg/kkg = 5,3 kg/kkg.  Similarly, the 24-hour maximum TSS
          limitation  was  obtained  by  multiplying  the  variability
                                   307

-------
                                                                 100
                              Percent Reuse
                         IEGENP
    •   tang-term data
    O  Expected with treatment system upgrading
    D  Screening and verification sanpling results
Figure  12-10,  Fluoride loads and concentrations discharged at
               selected hydrofluoric acid plants.
                           308

-------
factor for daily measurements  by  the  long  term  average;
i.e.,   3.5  x 3.1 kg/kkg = 11 kg/kkg.   The same approach was
taken to obtain the fluoride limitations;  i.e.,  1.6  x  1.8
kg/kkg  = 2.9 kg/kkg for the maximum 30-day average, and 3.4
x  1.8  kg/kkg  = t 6.1   kg/kkg  for  the   24-hour   maximum
limitation.  These' computations are shown on Table 12-20 and
the BPT limitations are presented in Table 12-21.

The  concentration  basis  (C)  for each effluent is derived
from the relationship between concentration (C),  flow  (Q),
and unit loading,

C (as mg/1) = 1000 (L)
                   (Q)

where  L  is  the  effluent  limitation  expressed as a unit
loading in kg of pollutant per kkg of product  (kg/kkg),  and
Q  is  the  flow  rate  expressed as cubic meters per kkg of
product (mVkkg).  (Note: kg/m3 = 1000 mg/1.)
Thus, the concentration basis for the maximum 30-day average
TSS limitation is:
(5.3 kq/kkq)   x
(54.6 mVkkg)
                 (1QQQ mg/1) = 97 mg/1
                 (kg/m3)
and  the  concentrations  basis  for  the  24-hour   maximum
limitation is obtained by a similar calculation or simply by
applying the variability factor ratio, VFR, from Table 12-21
to the maximum 30-day average concentration; that is:

(VFR) (max. 30-day average concentration or loading)

= 24-hour maximum concentration or loading

In  the same manner, the concentration basis for the maximum
30-day average fluoride limitation is:

(2.9 kq/kkq)  x  (1000 mg/1) =  53 mg/1
(54.6 mVkkg)    (kg/m3)

and the 24-hour maximum fluoride concentration is 2.1  x  53
mg/1  =  111  mg/1.  {Note:  due to rounding off, this value
differs just slightly from the value that appears  in  Table
12-21  which  was  obtained by calculating the concentration
directly from the 24-hour maximum limitation; i.e.,
(6.1  kq/kkq)  x
(54.6 mVkkg)

In either case,
taken. )
                  (1000 mg/1) =112 mg/1
                   (kg/m3)

                  only  two  significant  figures  should  be
                           309

-------
     Performance  evaluation  and review of discharge quality has
     been  complicated  by  problems  associated  with   chemical
     analysis.    Prior  to  July 1976,  the methods generally used
     for the analysis of fluoride in industry were  specific  ion
     electrode   or colorimetry.   These methods did not detect the
     soluble complex fluoride species present in the  wastewater.
     The best method of total fluoride detection (free as well as
     complex)  is  distillation  followed  by  analysis using the
     specific ion electrode.  Using the distillation method,  the
     complex fluorides  are  hydrolyzed  and the resulting HF is
     carried over with the distillate along with any free  HF  in
     the  sample.   Thus,   the  method of total fluoride analysis
     used for effluent monitoring is capable  of  measuring  free
     fluoride  and  the  fluoride  present in the form of complex
     ions which are not removed by  lime  treatment.   Monitoring
     data  on  effluent  fluoride levels using the revised method
     are likely to be higher than the levels previously  reported
     under the  same treatment conditions.

2.    Toxic pollutants

     The  effluent  limitations  set  for  the   selected   toxic
     pollutant   control parameters are derived from three sources
     of information.  These are  1)  screening  and  verification
     sampling  data,  2)  literature based treatability estimates
     (Section 8), and 3) a limited amount of long-term monitoring
     data from  Plant #251.

     The sampling results represent  plant  performance  observed
     during  three  days  of sampling at each of the plants.  The
     treated effluent data  on  the  toxic  pollutants  found  at
     significant  levels  are  summarized  in  Table  12-19.  The
     average values shown for each pollutant are  interpreted  as
     being  approximately  equal  to  a  long-term average unless
     there is some reason to  believe  that  abnormal  conditions
     existed  either in the process operation or in the treatment
     system at  the time of sampling.  Abnormal  conditions  would
     dictate  that  high  values  should  either  be  excluded or
     regarded as daily maxima rather than monthly averages.   For
     this  subcategory,  the  screening and verification data are
     believed to represent normal influent and effluent values at
     the plants sampled.

     For  a  number  of  metal  pollutants,  the  sampling   data
     demonstrate  that  the  effluent quality and percent removal
     with full  scale BPT systems are considerably better than the
     literature treatability data in Section 8 would indicate for
     that  particular  technology.   For  example,  even   though
     arsenic,  cadmium,  mercury,  and  thallium average influent
     concentrations are  well  below  the  accepted  treatability
     limits  for  lime/settling shown in Table 8-11, greater than
     60 percent removals were observed for all but cadmium as  is
     shown  in   Table  12-19.   This  high  degree  of incidental
                                 310

-------
    TABLE 12-20.  DEVELOPMENT OF TSS AND FLUORIDE LIMITATIONS
Plant
1837
1753
#928
#722
#664
Reuse
(percent)
0
65
83
92
94
Long
Waste
Fluoride
(kgAkg of HF)
1.8
0.72
1.0
0.81
0.10
Term Average
Load Discharged
TSS
(kgAkg of HF)
3.1
0.38
1.7
0.54
0.29
Variability Factor for
 Daily Measurements

Variability Factor for
 30-Day Averages
                        3.5(5)
1.6
Variability Factor Ratio  (VFR) 3.4/1.6 =  2.1
   (1)
                                              (2)
              3.5/1.7 - 2.1
                           (2)
Effluent Limitations for BPT
 (from Plant #837)
 a. Daily Max       3.4 X 1.8 kgAkg ~ 6.1
 b. Max 30-Day Avg  1.6 X 1.8 kg/kkg = 2.9v

Effluent Limitations for BAT
  (fron Plant #928)
 a. Daily Max       3.4 X 1.0  kgAkg = 3.4
 b. Max 30H3ay Avg  1.6 X 1.0  kgAkg = 1.6
                                           (3)
                                            (3)
                                            (4)
                                               3.5 X 3.1 kgAkg = 11
                           (3)
                                               1.7 X 3.1 kg/kkg =  5.3
                         MA
                         NA
                                                                       (4)
NA - Not Applicable
(1)  Variability factor average of Plants #664,#722 and *753 from Table 12-18,
(2)  Ratio of the daily  (24-hr) variability factor to the 30-day
     average variability factor.  This value appears on the Proposed
     Limitations tables.
(3)  The long term average loading in kgAkg multiplied by the
     variability factor for daily measurements as shown.
(4)  The long term average loading in kgAkg multiplied by the
     variability factor for 30-day measurements as_shown.
(5)  Variability factor from Plant #722, Table 12-18.
(6)  Variability factor from Plant #705, Table 12-18.
                                   311

-------
                     Table 12-21.  Effluent Limitations
                             Hydrofluoric Acid
          Best Practicable Control Technology Currently Available
             Waste Water Flow:  54.6 m3/kkg of HF (43% Reuse)*

Subcategory
Pollutant Performance

Conventional and
Nonconventional
Pollutants:
Total Suspended
Solids
Fluoride
Toxic Pollutants:
Antimony
Arsenic
Chromium
Copper
Lead
Nickel
Selenium
Zinc
(mg/1)



(?}
57 Uj
33(2>

0.80 (3>
0.50<3)
0.32(3>
0.32(3)
0.13<3>
0.17<4>
0.20(3)
0.55<4>
Concentration Basis Effluent Limit
m (mg/1) (kg/kkg of HF)
VFET '





3.5/1.7
3.4A-6

3.9/1.2
3.9/1.2
3.9/1.2
3.9/1.2
3.9/1.2
3.9/1.2
3.9/1.2
3.9/1.2
Max
30 -day
Avg




97
53

0.96
0.60
0.38
0.38
0.16
0.20
0.24
0.66
24-hr
Max




200
110

3.1
2.0
1.3
1.3
0.51
0,66
0.78
2.2
Max
30-day 24-hr
Avg Max




5.3 11
2.9 6.1

_(5> .(5)
_<5) _<5)
..(5) _(5)
__(5) __(5)
-(5) __(5)
0.011 0.036
-(5) _<5)
0.036 0.12
(1)  - VFR: ratio of the 24-hour variability factor to the 30-day variability
      factor.
(2)  - Long term average based on loading data and variability factors selected
      from Table 12-18.
(3)  - The lower limit of the literature treatability estimate (Table 8-11)  and
      industrial waste water treatment system performance (Table 8-12)  are used
      as the basis for the long term average vAien the observed average of the
      sampling data is below tJiis level.
(4)  - Average effluent concentration from screening and verification sampling
      data.
(5)  - No effluent limitation required.
*   - From Table 12-4.
                                     312

-------
removal supports the contention that  by  applying  effluent
1 imitations   just   to  the  dominant  metal  pollutant(s),
effective control of the other metals will also be assured.

In  Table  12-21,  the  concentration  bases  for  the   BPT
limitations  are derived from the averaged effluent sampling
data unless the observed pollutant concentration is actually
below the literature treatability level.  In some cases, the
lowest applicable treatability  level  from  Table  8-11  is
used.   This  approach  results in the setting of achievable
effluent limitations for all of the  pollutants  of  concern
and  provides for the possibility of wider variations in the
influent quality.  Such variations may  be  associated  with
different   fluorspar   impurity  levels  or  other  process
variables not fully taken into account by the  limited  data
obtained.

The  basis  for  the  BPT limitations on the control metals,
zinc and nickel is given below.  For the other toxic metals,
the concentration bases are derived below and  are  intended
to  serve  as  guidance  in cases where these pollutants are
found to be of serious concern.
a.
Zinc
     The raw waste concentrations of zinc ranged as high  as
     11.3  mg/1 {Section.12, Table of Maximum Concentrations
     Observed) and averaged about 3.2 mg/1  (Table 12-11} for
     the plants sampled.  BPT treatment achieved an  average
     removal  of  better  than  80  percent  with an average
     performance concentration of about  0.55  mg/1  in  the
     treated  effluent  shown in Table 12-19.  This level of
     performance approximately equals that obtained from the
     literature  treatability  data  in  Table  8-11.    The
     average  performance value is used as  the concentration
     basis for the proposed maximum 30-day average  effluent
     limitation  of  0.036 kg/kkg using the model plant flow
     of 54.6  mVkkg  (Table  12-4).   This  limitation  was
     achieved  by  all but one of the plants sampled.  Using
     the model plant flow of 54.6 mVkkg from  Table  12-15,
     the limitation was calculated as follows:

     (0.66 mg/1) (54.6 mVkkg)  (  kq/m3  )  = 0.036 kg/kkg
                               (1000 mg/1)

     where  0.66 is the maximum 30-day average concentration
     calculated by  multiplying  the  average  concentration
     (0.55) by the 30-day variability factor, 1.2:

     (0.55 mg/1}(1.2) =0.66 mg/1
                          313

-------
     Since  the  long  term  monitoring data on zinc are not
     available from the HF industry, the variability  factor
     ratio  (VFR) of 3.3 was derived from the nickel sulfate
     industry data.  The same VFR was  used  for  the  other
     toxic  metals in the HF effluent.  This is supported by
     the fact that nickel is a dominant toxic metal  in  the
     HF  process wastewater, and is assumed to have equal or
     better performance characteristics  compared  with  the
     wastewater from the nickel sulfate industry.  Thus,

     VFR = VF of daily measurements = 3.9
           VF of 30-day averages      1,2
           = 3.3

     and the 24-hour maximum limitation for zinc is:

     (3.3) (O.D36 kg/klcg) =0.12 kg/kkg.

     The  effluent  limitations on zinc and the other metals
     of concern are given in Table 12-21.

b.   Nickel

     The sampling data indicate better than 60  percent  BPT
     removal  of  nickel  resulting  in  an average effluent
     quality  of  about  0.17   mg/1.    Using   a   monthly
     variability factor of 1.2, a value of 0.20 mg/1 is used
     as  the  concentration  basis  for  the  maximum 30-«3ay
     average effluent limitation of 0.011  kg/kkg.  A VFR  of
     3.3 was used following the same rationale described for
     zinc.  Thus, the maximum 30-day average limitation is:

     (0.20 mg/1)(54.6 m3/kkg) (  kg/m*  )  = 0.011 kg/kkg
                               (1000 mg/1)

     and the 24-hour maximum limitation is:

     (3.3) (0.011 kg/kkg) = 0.036 kg/kkg.

c.   Lead

     The observed average raw waste  concentration  of  lead
     (0.66  mg/1)  was not far above the 0.13 mg/1 estimated
     long term average treatability according to  industry's
     performance  data  in  Table  8-12.   Using  a  monthly
     variability factor of 1.2, the concentration basis  for
     the maximum 30-day average is:

     (1.2) (0.13 mg/1) =0.16 mg/1

     Based  on  a  24-hour  variability  factor  of 3.9, the
     concentration basis for the 24-hour maximum is:

                          314

-------
(3.9) (0.13 mg/1) = 0.51 mg/1

No effluent: limitation is set for lead.

Antimony

According to literature treatability data (Table 8-11),
the lower limit of treatability for  antimony  is  0.80
mg/1  as a long term average.  Based on VFR of 3.9/1.2,
the concentration basis for the maximum 30-day  average
is:

(1.2) (0.80 mg/1) =0.96 mg/1

and the concentration basis for the 24-hour maximum is:

(3.9) (0.80 mg/1) =3.1 mg/1

No effluent limitation  is set for antimong.

Copper:  With 0.32 mg/1 as the average treatability for
copper  (Table 8-12), the concentration  basis  for  the
maximum 30-day average  is:

(1.2) (0.32 mg/1) =0.38 mg/1

and the concentration basis for the 24-hour maximum is:

(3.9) (0.32. mg/1) = 1.25 mg/1

No effluent, limitation  is set for copper.



Chromium: Similar to copper, an average treatability of
0.32 mg/1 is used for chromium.  The concentration  for
the maximum 30-day average is:

(1.2) (0.32 mg/1) = 0.38 mg/1

and  the concentration basis  for the 24-hour maximum is:

(3.9) (0.32 mg/1) = 1.3 mg/1

No effluent limitation  is set for chromium.

Other Metals

The  concentration bases for  arsenic  and   selenium  are
also presented  in Table 12-21.
                     315

-------
Basis for BCT Effluent Limitation

EPA  has  determined  that  the BAT technology for this subcategory is
capable of removing significant amounts  of  conventional  pollutants.
However,  EPA  has  not  yet  proposed  or  promulgated  a revised BCT
methodology in  response  to  the  American  Paper  Institute  v.  EPA
decision mentioned earlier.  Thus, it is not now possible to apply the
BCT  cost  test  to  this  technology  option.   Accordingly,  EPA  is
deferring a decision on the  appropriate  BCT  limitations  until  EPA
proposes  the  revised  BCT  methodology.   However,  the  Agency  has
calculated the TSS loading based on the estimated performance  of  BAT
treatment,  and  the cost of the additional TSS removal.  As described
in Section 3, this cost was calculated to be $0.32 per  pound  of  TSS
removed.

By  adjusting the loading to account for the decrease in effluent flow
rate from BPT (54.6 mVkkg) to BAT  (33.4  m3/kkg),  the  TSS  maximum
30-day average effluent loading becomes:

  (5.3 kg/kkg) x (33.4 mVkkq) = 3.2 kg/kkg
                 (54.6 mVkkg)

The corresponding 24-hour maximum effluent loading is then obtained by
applying the VFR value of 2.1 (Table 12-21).  That is:

     (2.1)(3.2 kg/kkg) =6.7 kg/kkg

Basis for BAT Effluent Limitations

A.   The Application of Advanced Level Treatment

     Utilizing the cost estimates presented in this report, the Agency
     has analyzed the cost effectiveness of  the  base  level  systems
     (BPT)  and  the  various advanced level options for conventional,
     nonconventional  and  toxic  pollutant  removal.   The   economic
     impacts  on the Hydrofluoric Acid industry have been evaluated in
     detail  (53) and taken into consideration in the selection of  the
     technology basis for the BAT regulations.

     For   BAT,  the  Agency  is  promulgating  limitations  based  on
     treatment  consisting  of  Level  1   technology.    The   Agency
     considered the use of treatment Level 2  (addition of a dual-media
     filter)  but  did  not  adopt  it because the installation of the
     filter  is not cost effective in the removal of TSS, fluorides and
     toxic  metals.   In  addition,  the  use  of  treatment  Level   3
     (addition  of  sulfide precipitation) was considered but rejected
     due to  lack of performance data.  EPA also considered Level 4,   a
     variation  of Level 2, that would substitute soda ash in the lime
     precipitation step and allow  90  percent  recycle  of  effluent.
     This  option  was  rejected due to being prohibitively expensive.
     Pollutants limited by the  final  BAT  regulation  are  fluoride,
     nickel, and zinc.
                                  316

-------
B.   Technology Basis

     For BAT, the  Agency  is  setting  the  effluent  limitations  on
     fluoride  and the toxic metals based on the BPT treatment system,
     coupled with the requirement of  at  least  65  percent  effluent
     reuse for kiln residue slurrying.

     In   Section   7.0   "Hydrofluoric  Acid  Subcategory"  of  EPA1s
     Treatabllity Studies for the  Inorganic  Chemicals  Manufacturing
     Point  Source Subcateqory (EPA 440/1-80/103) (61), the conclusion
     states:

     "A major conclusion that can be drawn from this study is that the
     addition of dual-media filtration  after  alkaline  precipitation
     and  settling  is not particularly effective for the reduction of
     final TSS and total fluoride concentrations.  Further, dual-media
     filtration does not appear  to  be  justified  on  the  basis  of
     additional  toxic  metal removal judging by the results presented
     in Table 7-3."

     A summary of the filter removal efficiencies for  the  pollutants
     selected for the treatability study is given in Table 12-22.

     The  minimum  reuse rate of 65 percent was selected because it is
     typical of the five plants  (Plants #167, #753, #928, # 664, and #
     722} which presently practice reuse as is shown in Table 12-4.

C.   Flow Basis

     With the model plant inflow rate of 95.4 mVkkg and the reuse  of
     65  percent  of  the treated effluent, the quantity discharged is
     33.4 mVkkg; i.e.,  (1.00 -0.65X95.4 mVkkg) = 33.4 mVkkg.

D.   Selection of Pollutants to be Regulated

     For the BAT regulations, the Agency has selected  fluoride and the
     same two toxic metals identified in  the  BPT  regulations.   The
     rationale for their selection is discussed above.

E.   Basis of Pollutant Limitations

     1.   Nonconventional pollutants

          The only nonconventional pollutant found  in  the  wastewater
          in  this subcategory is fluoride.  The limitation for BAT  is
          based on the performance of the four plants  shown  in  Table
          12-20  that  presently  reuse  at  least  65  percent of their
          treated effluent.  The long term  average  effluent  loading
          taken  from  Table 12-22 is 1.0 kg/kkq for Plant #928, since
          the percent reuse for  this plant  is  intermediate  in  range
          for the three plants considered for BAT  (Plants  #753, #  928,
          and  #722).   Plant  #664 is excluded from consideration for
          BAT because the high degree of  recycle   practiced  requires
                                 317

-------
     the  extensive use of soda ash.   The use of  this performance
     in conjunction with the 30-day average variability factor  of
     1.6 and the model  plant net discharge rate  of   33.4   mVkkg
     results in a calculated maximum 30-day average  concentration
     of 50 mg/1 total  fluoride.  Thus,  the maximum 30-day  average
     limitation is:

     (1.6)(1.0 kg/kkg)  =1.6 kg/kkg

     and its concentration basis is:
     (1.6 kq/kkq)
     (33.4 mVkkg)
         x  1000 mq/1  = 50 mg/1
             (kg/m3)
     This   represents   a  six  percent  reduction  in  fluoride
     concentration in going from BPT (43 percent  reuse)   to  BAT
     (65  percent  reuse).  The use of a fixed loading limitation
     allows  the  permissible  concentration  to  increase  as  a
     function  of  percent reuse.   The 24-hour maximum limitation
     on fluoride is obtained by utilizing the long  term   average
     and variability factor for daily measurements,

     (3.4)(1.0 kg/kkg) = 3.4 kg/kkg

     and the concentration basis is:

     (3.4 kg/kkq) x (1000 mg/2) = IGQ-mg/l
     (33.4 mVkkg)      (kg/m3)

     The  variability  factors   used  for  the BAT limitations on
     fluoride are the same as for BPT shown in Table 12-20.    The
     BAT  limitations  for  the Hydrofluoric Acid Subcategory are
     presented in Table 12-23.

2".    Toxic pollutants

     For BAT regulations, the EPA  is  proposing  more  stringent
     controls on the discharge of the two toxic metals of concern
     on  the  basis  of  a reduced volume of discharge.  Alkaline
     precipitation converts most of  the  dissolved  metals  into
     less  toxic, insoluble forms such as hydroxides and hydrated
     oxides.    Other    mechanisms    of    removal    including
     coprecipitation  and  flocculation  are  undoubtly  involved
     during the treatment process  and  probably  account  for  a
     substantial portion of the removal of certain toxic  metals.

     The  bases for the BAT limitation on the two control metals,
     zinc and nickel, are given below.  The  concentration  bases
     of  other metals are derived below and are intended to serve
     as guidance in cases where these pollutants are found to  be
     of serious concern.
     a.
Zinc
                              318

-------
  TABLE 12-22.  SU4MAHY OF CCNCENTRATIONS AND FILTER REMOVAL EFFICIENCIES
                OF POLLUTANTS SELECTED FOR STUDY IN THE HYDROFLUORIC ACID
                SUBCATEG3KY
                      Total     Total    Total      Free      Total
                     Chromium    Zinc    Nickel   Fluoride   Fluoride   TSS


Average raw            0.30      1.09     1.23      219.8      —      3315
vraste concentration
(mg/1)

Average supernatant    0.073     0.10     0.46       45.9      116      333
concentration
(ng/1)

Average filter         0.067     0.05     0.41       42.2       91      134
effluent concentra-
tion  (mg/1)

Average filter           6        36       17         6         20      53
removal efficiency
                                     319

-------
     The same value of 0.55 mg/1 used for BPT is used as the
     treatability  limit  for   zinc.    As   in   the   BPT
     regulations,   a  variability  factor  ratio  (VFR)  of
     3.9/1.2 was used for  zinc  and  the  other  metals  in
     setting  the  BAT  limitations and concentration bases.
     Thus,  the concentration basis for  the  maximum  30-day
     average effluent limitation is:

     (1.2)  (0.55 mg/1) =0.66 mg/1

     and the maximum 30-day average effluent limitation is:

     (0.66  mg/1) (33.4 mVkg)  (kg/m*  )   = 0.022 kg/kkg
                            (1000 mg/1)

     The   concentration   basis  for  the  24-hour  maximum
     limitation is:

     (3.9)  (0.55 mg/1) = 2.2 mg/1

     and the 24-hour maximum limitation is:

     (2.15  mg/1) (33.4 mVkkg) (  kq/m3  )  = 0.022 kg/kkg
                                 (1000 mg/1)

     This represents an overall reduction of 40 percent from
     the BPT loading limitation.   The  BAT  limitations  on
     zinc and nickel are included in Table 12-23.
b.   Nickel
     The same value of 0.17 mg/1 used for BPT is used as the
     treatability limit for nickel.  The concentration basis
     for the maximum 30-day average effluent limitation is:

     (1.2) (0.17 mg/1) =0.20 mg/1

     and the maximum 30-day average effluent limitation is:

     (0.20 mg/1) (33.4 mVkkg) (  kq/m^  )  = 0.0060 kg/kkg
                               (1000 mg/1)
     The  concentration  basis  for  the
     limitation is:
24-hour   maximum
     (3.9) (0.17 mg/1) = 0.66 mg/1

     and the 24-hour maximum effluent limitation is:

     (0.66 mg/1) (33.4 mVkkg)  (  kg/m3  )  =0.20 kg/kkg
                                (1000 mg/1)
                            320

-------
                        TABLE 12-23.  EFFLUENT LIMITATIONS
                                HYDROFLUORIC ACID
                            Best Available Technology
                    Wastewater Flow: 33.4 m3/kkg of HF  {65% Reuse)
                                          Concentration Basis
                                                 (mg/1)	
Pollutant
Treatability
   (mg/1)
                                 VFR
                                     (1)
           Max.
           30-day
            Avg»
         24-hr.
          Max.
          Effluent Limit
          (kg/kkg of HF)
          Max
          30-day    24-hr,
           Ayg.      Max.
Noriconvetitional
 Pollutants:	

Fluoride/ F

Toxic Pollutants
    31
      (2)
3.4/1.6
50
100
1.6
3.4
Antimony
Arsenic
Chromium
Copper
Lead
Nickel
Selenium
zinc
0.80
0.50
0.32
0.32
0.13
0.17
0.20
0.55
3.9/1,2
3.9/1.2
3.9/1.2
3.9/1.2
3.9/1.2
3.9/1.2
3.9/1.2
3.9/1.2
0,96
0.60
0.38
0.38
0.16
0.20
0,24
0.66
3.1
2.0
1.3
1.3
0.51
0.66
0.78
2.2
— (3)
— (3)
— (3)
— (3)
0.0060
— (3)
0.022
— (3)
— (3)
— (3)
— (3)
0.020
— (3)
0.072
(1)  VFR: ratio of the 24-hour variability factor to the 30-day variability
     factor.

(2)  30-day average calculated for the model plant based on data in Table  12-20.

(3)  No effluent limitations.

*    The effluent flow rate is 35 percent of the average influent shown in
     Table 12-4 (i.e., 0.35 X 95.4 m3A^g = 33.4
                                           321

-------
This represents an overall 44 percent decrease from the
corresponding BPT level.

Lead
Using a treatability limit of 0.13 mg/1 for  lead,  the
maximum 30-day average concentration basis is:
(1.2) (0.13 mg/1) =0.16 mg/1

The 24-hour maximum concentration basis is:

(3.9) (0.13 mg/1) =0.51 mg/1

No effluent limitation is set for lead.

Antimony

Using a treatability limit of 0.80 mg/1  for  antimony,
the maximum 30-day average concentration basis is:

(1.2) (0.8 mg/1) = 0.96 mg/1

The 24-hour maximum concentration basis is:

(3.9) (0.8 mg/1) = 3.1 mg/1

No effluent limitation is set for antimony.

Copper

Using a treatability limit of 0.32 mg/1 for copper, the
maximum 30-day average concentration basis is:

(7.2) (0.32 mg/1) = 0.38 mg/1

The 24-hour maximum concentration basis is:

(3.9) (0.32 mg/1) = 1.3 mg/1

No effluent limitation is set for cooper

Chromium.

Using a treatability limit of 0.32 mg/1  for  chromium,
the maximum 30-day average concentration basis is:

(1.2) (0.32 mg/1) =0.38 mg/1

The 24-hour maximum concentration basis is:
                    322

-------
               (3.9) (0.32 mg/1)  =1.3 mg/1

               No effluent limitation is set for chromium.

          g.    Other Metals

               The concentration bases for arsenic  and  selenium  are
               also given in Table 12-23.


Basis for the New Source Performance Standards

BAT was selected as the basis for NSPS limitations.  The pollutants to
be  controlled  for  NSPS  are  pH, total suspended solids, fluorides,
nickel and zinc.  The NSPS limitations are given in Table 12-24.

In addition to the toxic pollutants controlled by  the  BAT,  a  total
suspended solids limitation was developed for NSPS.  Based on the data
in  Table  12-20,  a  long term average of 1.7 kg/kkg (Plant t928) was
used for the TSS.  The variability factor ratio (VFR) used is 2.1,

     3.5 = 2.1
     1 .7

Thus, the 24-hour maximum effluent limitation for TSS is:

     (3.5) (1.7 kg/kkg} = 6.0 kg/kkg HF

The maximum 30-day average limitation for TSS is:

     (1.7) (1.7 kg/kkg) = 3.0 kg/kkg HF

Basis for the Pretreatment Standards

A.   Existing Sources

     The Agency is not promulgating PSES because there are no indirect
     dischargers in the subcategory.  Instead, we  are  excluding  the
     subcategory   from  categorical  PSES  under  the  provisions  of
     paragraph 8(b) of the Settlement Agreement.

B.   New Sources

     For Pretreatment Standards for New Sources (PSNS), the Agency  is
     setting  limitations  equal  to NSPS because NSPS provides better
     removal of nickel and zinc than is achieved  by  a  well-operated
     POTW  with  secondary  treatment  installed  and  therefore these
     pollutants  would  pass  through  a  POTW  in  the   absence   of
     pretreatment.   The  pollutants  to  be  regulated  are fluoride,
     nickel, and zinc as indicated in Table 12-24.
                                  323

-------

-------
                    Table 12-24.  Effluent Limitations
                            Hydrofluoric Acid
                     New Source Performance Standards *
             Waste Water Flow:  33.4 n^/kkg of HF  (65% Reuse)*
   Pollutant
               Treatability
                 (mg/1)
                                Concentration Basis

                                    ,U)
VFR'
                                          Max
                                          30-day   24-hr
                                          Avg     Max
       Effluent Limit
       (kg/kkg of HF)

        Max
       30-day  24-hr
        Avg     Max
Conventional and
Nbnconventianal
Pollutants;

Total Suspended
Solids, TSS

Fluoride, F

Toxic Pollutants:
                    53
                    31
                      <3>
                              3.5/1.7

                              3.4A-6
         90

         50
180

100
3.0

1.6
6.0

3.4
Antimony

Arsenic
Chromium
Copper
Lead
Nickel (2)
Selenium
2incC2)
0.80

0.50
0.32
0.32
0.13
0.17
.0.20
0.55
3.9/1.2

3.9/1.2
3.9/1.2
3.9/1.2
3.9/1.2
3.9/1.2
3.9/1.2
3.9/1.2
0.96

0.60
0.38
0.38
0.16
0.20
0.24
0.66
3.1

2.0
1.3
1.3
0.51
0.66
0.78
2.2
_V*' 	 \"*l
(A) Ml
* ' i— ^ '
_(4) _(4)
i 4) f 4)
I A\ f4)
0.0060 0.020
_J4) _<4)
0.022 0.072
(1)


(2)

(3)


(4)
 *
   - VFR:  ratio of the 24-hour variability factor to the 30-day variability
     factor.
   - Also applicable for PSES limitations.
   - 30-day average calculated for the model plant based on the data in
     Table 12-20.
   - No effluent limitation.
                             is
                                                                   shown in
t  -
     Table 12-4 (i.e.,  0.35 X 95.4 m3/kkg = 33.4 m^/kkg) .

     This  table is also applicable to  the PSNS effluent limitations  except
     for the TSS limitation.
                                    325

-------

-------
                              SECTION 13

                      HYDROGEN PEROXIDE INDUSTRY
Summary of Determinations

It has been determined that no further effort be given  to  developing
or  revising  BAT,  NSPS, or pretreatment regulations for the Hydrogen
Peroxide Subcategory using either  the  electrolytic  process  or  the
organic process.

The  bases  for  this  exclusion  are:   1) only one plant exists that
manufactures hydrogen peroxide using the electrolytic process  and  2)
no  toxic  pollutants  were  found  in  the  wastes  using the organic
process.  Therefore, this subcategory is excluded under Paragraph 8 of
the Settlement Agreement.

Assessment of the Water Pollution Potential

Production Processes and Effluents

In the electrolytic process, ammonium {or other) bisulfate solution is
electrolyzed, yielding ammonium persulfate at the anode  and  hydrogen
gas  at  the  cathode.   The  persulfate is then reacted with water to
yield hydrogen peroxide and original bisulfate.  Hydrogen peroxide  is
separated   from   bisulfate  by  fractionation,  after  which  it  is
concentrated and filtered.  The only waste is a stream  of  condensate
from the fractionation condenser.

The  organic  process  involves the reduction of alkylanthraquinone by
hydrogen over a supported metal catalyst to produce the  corresponding
alkylhydroanthraquinone.   The  reacted  mixture  is  oxidized to form
hydrogen peroxide and original alkylanthraquinone.   The  peroxide  is
extracted  with  water  and  the  organic  material  in the solvent is
recycled to the process.  Since the manufacture of  hydrogen  peroxide
by  the  organic  process  consists of a series of exothermic chemical
reactions,- the bulk of the water usage is for process cooling (contact
and noncontact).  Noncontact cooling accounts for over 90  percent  of
the  total  water  usage  in this subcategory.  The wastewater sources
include contact  cooling  (barometric-condenser)  water,  purification
washing  of the organic working solutions, regeneration waste from the
deionizers, and leaks and spills.

Plants

Only one plant exists in the United States that manufactures  hydrogen
peroxide  using  the  electrolytic  process.   The  hydrogen  peroxide
subcategory profile data received in response to Section  308  letters
is given in Table 13-1.
                                327

-------
      13-1
SUBCATEGQRY PROFILE DATA SUMMARY
SGBCATEQORy
HYDROGEN PEROXTCE
Total subcategory capacity rate
Ttotal subcategory production rate
Number of plants in this subcategory
308 Data on file for
    With total capacity of
    With total production of
    Representing capacity
    Representing production
    Plant production range:
            Minimum
            Maximum
    Average production
    Median production
    Average capacity utilization
    Plant age range:
            Minimum
            Maximum
    Waste water flow range:
            Minimum
            Maximum
    Volume per unit product:
            Mi Tvjrmm
            Maximum
                            85,700 kkg/year
                               4
                               4
                           102,200 kkg/year
                            57,000 kkg/year

                               66 percent

                             5,560 kkg/year
                            28,730 kkg/year
                               NA
                               NA
                               NA

                               15 years
                               27 years

                               NA
                               NA

                               NA
                               NA
Sources of data are Stanford Research Institute, Directory of Chemical
Producers, U.S.A., 1977, U.S. Department of Commerce, Current Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry, " June,  1978 and "Economic Analysis of Proposed
Revised Effluent Guidelines and Standards  f
-------
Three plants product hydrogen peroxide by the organic process.

Toxic Pollutants

Data  has  been received on 100 percent of the industry as a result of
Section 308 letters.  A sampling survey for toxic pollutants was  made
for  three  plants.   At  one  plant,  pentachlorophenol  was found in
significant concentrations.   However,  it  was  determined  that  its
presence  was  due  to  its use as a weed killer at the plant site and
this use was discontinued.   Two  more  plants  were  sampled  in  the
verification  phase, and the survey indicated that no toxic pollutants
were being discharged in significant quantities.

     Toxic pollutants found during sampling were as follows:
     Pollutant

     Zinc
     Pentachlorophenol
     Bis(2-ethylhexyl)phthalate
     Chloroform
     Naphthalene

Status of Regulations
 Maximum Concentration
Observed Uq/1)

        256
       4850
         20
         11
         11
Since no toxic pollutants were found  in  significant  concentrations,
the  subcategory  is  excluded  under  Paragraph  8  of the Settlement
Agreement•
                                    329

-------

-------
                              SECTION 14
                      TITANIUM DIOXIDE INDUSTRY
INDUSTRY PROFILE

General Description

Titanium dioxide (TiO2) is  manufactured  by  a  chloride  process,  a
sulfate process, and a chloride-ilmenite process.  This subcategory is
subdivided  into  three  segments, one for each process because of the
difference in raw materials used,  wastewater  flows,  and  raw  waste
characteristics.   TiOz  is a high volume chemical, ranking within the
first fifty of all U.S. chemicals production.  Over fifty  percent  of
the  titanium  dioxide  produced  is  used   in  paints, varnishes, and
laquers.   About  one  third  is  used  in   the  paper  and   plastics
industries,.   Other  uses  are  found  in  ceramics,  ink,  and rubber
manufacturing.

The industrial profile data for the chloride segment are presented  in
Table   14-S,  while the status of regulations prior to promulgation of
this now regulation is given in Table 14-2.

Subcategorization

The Titanium JHoxide  Subcategory  is  divided   into  three  different
processes requiring separate consideration.  Two processing techniques
including  sulfate  arad  tihe  chloride  process  using  two major ores
including rutlie and ilmenite dictate  a  division  into  the  sulfate
process,  chloride  process/rutile  ore, and chloride process/ilmenite
ore  
-------
TABLE 14-1
SDBGKEEX3QRY' PROFILE DKEA
SOBCKEEGQEQ?
TTIM1IUM DIOXIDE (OEOKTDE PROCESS)
Total subcategory capacity rate
Ibtal subcategory production rate
ftnfcer of plants in this subcategory
308 Data on file for
    With total capacity of
    With total production of
                 capacity
    Representing production
    Plant production ranges
    Average production
    Median production
    Average capacity rfr'*
    Plant age range:
            Minimum
            Maximum
    Waste  water flow range:
            Minimum
            Maximum
    Volume per unit product:
            Minimum
                           610,000 kkg/year
                           389,000 kkg/year
                                5
                                5
                           184,600 kkg/year
                           142,000 kkg/year
                               30 percent
                               37 percent

                           16,900 kkg/year
                           '45,200 kkg/year
                           28,400 kkg/year
                           25,600 kkg/year
                               77 percent

                                6 years
                               15 years

                            1,140 cubic meters/day
                            4,770 cubic meters/day

                             29.3 cubic meters/kkg
                              110 cubic meters/kkg
Sources of data are Stanford Besearch Institute, Directory of Chemical
Producers, U.S.A., 1977, U.S, Department of Gcnroerce, Current Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry," June, 1978, and "Economic Analysis of
Proposed Revised Effluent Guidelines and Standards for the Inorganic Chemcals
Industry," March, 1980.
                                   .332

-------
        TABLE 14-2. STATUS OF REGULATIONS - EFFLUENT LIMITATION GUIDELINES
SUBCATEGORY

SUBPART
TITANIUM DIOXIDE

 V (40 CFR 415.220, 3/13/74)
STANDARDS
BPCTCA*
Product
Process
Chloride
Process

Sulfate
Process


Max.1
kg/kkg
Parameters (mg/1)

TSS 4,6
Iron 0.72

TSS 21.0
(100.0)**
Iron 1 . 7
(8.1)
Avg. 2
kg/kkg
(mg/1)

2,3
0.36

10.5
(50.0)
0.84
(4.0)
BATEA*
Max.
kg/kkg
(mg/1)

2,6
0.36

10,6
0.84

Avg.
kg/kkg
(mg/1)

1.3
0. 18

5.3
0.42

NSPS*
Max.
kg/kkg
(mg/1)

2.6
0.36

10.6
0.84

Avg.
kg/kkg
(mg/1)

1.3
0.18

5.3
0.42

  Sections 415.220, 415.222, 415.223 and 415.225 were  remanded and reserved
  (41 FR 51601, November 23, 1976).
  Max. = Maximum of any one day.
2 Avg. = Average of daily values for thirty consecutive  days,
  **   - Flow basis 210,000 1/kkg.
                                        333

-------
The reaction takes place at a temperature of 800-1000 degrees C  and   a
fluidized  bed  reactor  is generally used.  The product gases leaving
the reactor consist of  titanium  tetrachloride,  unreacted  chlorine,
carbon  dioxide,  carbon  monoxide,  and  minor amounts of heavy metal
chlorides.  The gases are initially cooled to 250 degrees C  to   remove
the impurities, although in some cases purification  is accomplished  by
washing   the  gases  with  liquified  titanium  tetrachloride.    Iron
chloride and small amounts of vanadium,  zirconium,  and  other  trace
metal  chlorides are removed by centrif ugation and the liquid recycled
to the absorber.  Titanium tetrachloride is liquefied from   the  gases
after  the  first  stage  of  cooling  by  further   cooling  to ambient
temperature.   Copper,  hydrogen  sulf ide,   and,    in   some    cases,
proprietary  organic  complexing  agents are added for purification  to
the condensed solution.  Copper acts as a catalyst   to  decompose  the
phosgene formed in the TiCl4 stream.  Organic complexing agents  aid  in
separation  of  the TiCl* from other chlorides such  as cupric chloride
and silicon tetrachloride.

The residual uncondensed gases generally consist of  hydrochloric acid,
chlorine, carbon monoxide, carbon dioxide, nitrogen, and some titanium
tetrachloride.  They are treated to  remove  acidic  materials   before
being vented to the atmosphere.

The  liquefied  titanium, lf^JLetrachlor ide  contains  impurities  such  as
  , ,— ; ff-%fn., .-^--^x^^i^^^^^^^-''fr"<*!^^              ,     , •  L             -i  ,
aluminum chlorrae, silicon tetrachloride, etc., which are  removed  by
distillation.   The  distillate is the purified titanium tetrachloride
and the impurities remain as a residual which becomes waste.  The  tail
gases from the distillation  column  are  scrubbed   to  remove   acidic
materials.   The  titanium  tetrachloride product is then reacted  with
oxygen, as air, forming titanium dioxide and chlorine:
TiCl4 + 02 = TiOz + 2C1?                                     (2)

The rate of reaction is negligible below  600 degrees  C   but   increases
rapidly  above  this  temperature, and  is generally maintained  between
1200-1400 degrees C for efficient reaction and  conversion.   The needed
heat is supplied by passing the reactants through  heat  exchangers,   by
electric discharges, or by use of fluidized beds.  After the oxidation
              titanium dioxide, ,f p,r.m,s_, a_ so, 114 an,d_ is,7separa±.e
-------
Water Use and Waste Source Charateristics

Water Use

Water is used in noncontact cooling, for scrubbing the tail gases from
the purification and oxidation reactors to remove contaminants, and in
the  finishing  operation  of  the product.  The total amount of water
usage varies from 45.3 to 555 m3/kkg of TiO2  produced,  as  shown  in
Table  14-3.   The table also shows that cooling water constitutes the
major use of water  and  varies  from  10.7  to  426  m3/kkg  of  TiOz
produced.

Waste Sources

A.   Wastes from Cooling Chlorinator Gas

     The waste consists of solid particles  of  unreacted  ore,  coke,
     iron,  and  small  amounts  of vanadium, zirconium, chromium, and
     other heavy metal chlorides.  They are either dissolved in  water
     and  sent  to the wastewater treatment facility or disposed of in
     landfills as a solid waste.

B.   Chlorinator Process Tail Gas Scrubber Waste

     The  uncondensed  gases,  after  the  liquefaction  of   titanium
     tetrachloride,  are  initially  wet  scrubbed  to remove hydrogen
     chloride, chlorine, phosgene, and titanium tetrachloride.   In  a
     second  stage,  they  are  scrubbed  with  caustic soda to remove
     chlorine as hypochlorite.

C.   Distillation Bottom Wastes

     These -contain copper,  sulfide,  and  organic  complexing  agents
     added  during  purification in addition to aluminum, silicon, and
     zirconium chlorides.  These are removed as waterborne wastes  and
     reaction  with  water  converts  silicon  and  anhydrous aluminum
     chlorides to their respective oxides.

D.   Oxidation Tail Gas Scrubber Wastes

     The gases from the oxidation unit are cooled by refrigeration  to
     liquefy  and  recover  chlorine.   The uncondensed off-gasses are
     scrubbed with water or caustic soda to remove residual  chlorine.
     When  caustic  soda  is  used  as  the  scrubbing  solution,  the
     resulting  solution  of  sodium  hypochlorite  is  either   sold,
     decomposed,   sent  to  the  wastewater  treatment  facility,  or
     discharged without treatment.  The  scrubber  waste  stream  also
     contains titanium dioxide particulates.
                                     335

-------
                               aoj ?XGKMr
Figure 14-1.  General process diagram for production of titanium
              (chloride process) from high grade ores.
                                  336

-------
 TABIE 14-3.  WATER USAGE IN TITANIUM DIOXCDE-CHIORlbE PROCESS/HIQI GRADE
              ORES SUBCATEGORY
      Water Use
Water Usage at Plants
    (m3/kkg of
      Water Use
Plant #102
Noncontact cooling
Direct process contact
Indirect process contact
Maintenance, equipment
cleaning and work area
washdcftm
Air pollution control
Noncontact ancillary uses
Sanitary & portable water
Total
182
10.5
NA
6.65
0.25
11.6
0.23
211
Plant #172
10.7
15.5
0.72
0.52
7.14
10.4
0.31
45.3
Plant #199
426
73.2
26.5
2.80
11.3
9.5
5.6
555
NA = Not available
                                     337

-------
E.   Finishing Operations Waste

     The liquid wastes from the finishing operation  contain  titanium
     dioxide  as  suspended solid and dissolved sodium chloride formed
     by the neutralization of residual HC1 with caustic soda.

The range of wastewater flows requiring  treatment  is  summarized  in
Table  14-4.   The  wide  range of flow occurs because some plants use
additional water to wash solid process residues to the waste treatment
system.

Description Of Plants Visited And Sampled

Screening

Plant #559 was visited and the waste effluents  were  sampled  in  the
screening  phase  of  the program.  Plant £ 559 makes titanium dioxide
using both the sulfate and the chloride  processes.   The  wastewaters
from both processes are mixed and undergo combined treatment.

The  solids  from  the  chloride  process,  called pit solids, (mainly
unreacted ore, coke, iron, and trace metal chlorides including  TiCl^)
are  separated  from  the first stage cooling of the chlorinated gases
and are  slurried  with  water.   The  slurried  pit  solids  and  the
distillation column bottom residue effluents from the chloride process
are  sent  to  a large settling pond (called the weak acid pond) where
they are mixed with the weak  acid  from  the  sulfate  process.   The
overflow  from  the  settling  pond is neutralized with ground calcium
carbonate in a reactor.  The reactor effluent is filtered, aerated  to
remove  iron, and combined with neutralized strong-acid waste effluent
from the sulfate process. The combined scrubber  and  contact  cooling
wastewaters  from  both  the  sulfate  and chloride processes are also
combined at this point.  The combined wastewater  is  neutralized  and
solids settled out in a pond prior to final discharge.  A flow diagram
of  the treatment facility, including the sampling locations, is shown
in Figure 14-2.

Problems were encountered during the sampling of the  pit  solids  and
distillation  bottoms.  The pipes carrying the wastes from the process
discharged at the bottom of the settling pond and it was not  possible
to  take  the  samples  right at the outlet of the pipe.  The combined
sample of the two streams was taken at the surface of  the  discharge.
It  is probable that some solids settled before the stream reached the
surface.  Table 14-5 gives the waste flows and pollutant loadings  for
the  streams sampled at Plant 1559.  Because of the intermixing of the
waste  effluents  from  both  chloride  and  sulfate  processes,   the
pollutant  loadings  in  Table  14-5  were calculated by proportioning
according to the relative hydraulic loadings.

Verification  Plant -JtLZi^was  sampled  in  the  verification  phase.
Titanium  dioxide  is  made  at  this facility by the chloride process
only.  The strong acid wastes and the spent coke and ore residues  are
hauled  to  a  secure  chemical landfill for disposal.  The wastewater
                                     338

-------
       TABLE 14-4.  WASTE WATER FLOW FOR TITANIUM DIOXIDE-CHLORIDE PROCESS
                    SUBCATEGORy
SUBCATEGORy
TITANIUM DIOXIDE (Chloride Process)
Plant
    Unit Waste Water Flow Going to Treatment Plant

                      (m3/kkg of Ti02>
 #102

 #172

 #559

 #199
                         29.3
                         34.7
                         91.0
                        110.0
(1)

(1)

(2)

(2)
(1)  Off-site disposal of process solid residues.

(2)  Process solid residues are slurried to waste treatment.  The average
     flow of Plants #559 and #199 was used as the model plant flew for
     cost estimating and regulation development.
                                      339

-------
from the process, mainly the scrubber water, is collected in  trenches
and  sent  to  a central reactor basin.  Other discharges, including a
part of the total rain runoff, are also collected in ditches and  sent
to  the  reactor basin.  In the reactor basin sodium hydroxide is used
for neutralization and  the  resulting  effluent  is  mixed  with  the
remaining  rain  runoff  and sent to the first of two retention basins
arranged in series.  The overflow from the second retention  basin  is
pH adjusted with sulfuric acid before discharge.  A simplified diagram
of  the  treatment  system, including the sampling points, is shown in
Figure 14-3.  Table 14-6 gives the waste flow and  pollutant  loadings
for the streams sampled.


Toxic Pollutant Concentrations

Chromium  was  identified  in  the  raw wastewater for Plant #172 at a
concentration considered treatable based on information summarized  in
Section   8.   Four  toxic  metal  pollutants  were  identified  above
treatable levels at Plant #559, although it is likely that the  levels
observed  included  some contributions from the sulfate process, since
the two raw waste effluents are intermixed before treatment.  No toxic
organic pollutants were identified above treatable levels in  the  raw
wastes  at  either  plant.   The  maximum concentration of toxic metal
pollutants observed during screening and verification are presented as
follows:

           Maximum Raw Waste Concentrations Observed (ug/1)
          Pollutant
  Screening
Plant #559
 Verification
Plant #172
Chromium
Lead
Nickel
Zinc
152,000
5,150
6,320
3,300
1800
14
85
660
The screening and verification sampling program  and  the  methodology
used  have been described in Section 5 of this report.  A total of six
days of sampling was conducted at Plants 1559 and #172.  Five effluent
waste streams were sampled  at  Plant  #559  and  three  streams  were
sampled  at  Plant  #172.   At  each  sampling  point,  three  24-hour
composite samples were collected  for  analysis.   The  evaluation  of
toxic  metal  content of these process related waste streams was based
550  analytical  data  points.   The   average   unit   loadings   and
concentrations for conventional, nonconventional, and toxic pollutants
found in the raw waste effluents for Plants #559 and #172 are given in
Table 14-7.

The  total quantities of toxic pollutants generated each year for this
Subcategory (calculated as total subcategory production times  average
unit toxic pollutant load from Table 14-7) are as follows:
                                    34C

-------
JUWf WATEft
sum* WATEN


WILL
     UAH*
^T
                              SANITARY
                          ANft TIO, fINISHINC
                           AREA IlkSfE WATER
                                                             OfHEN fHODUCT
                                                             (STKOHC ACIOl
                                                             WASTE UA!£«
iUILIHC
OIHCK roooucr
(UCAK ACIO) 	 »*
WAIFI WAtEft
,
ILUKMIED — ^
fH SOLIDS
rCHD
9"
t



»i,


1C AC ION

1

	 	 	 	 	 ft
fl


^
t
^_ PISTILLATION I01TOH
^^ UA$TE IMTCI
UCtMD
•4 SAHPiINC rOINTS

(
tTH
WA
EN 1
HE
mis
* 	 	 CHLONIOC PNOCCSS SCHUIICR
WASU WATER
floouci
WATER
                   Fioure 14-2.   General  flow diagram at Plant  #559 showing the sanpling points.
                                   (Titanium dioxide — chloride  process  manufacture.)

-------
                        TABt£ 14-5.  FUCW AND POLLUTANT CONCENTRATION DATA OF THE SAILED WASTE

                                     STREAMS OF PIANT #559 PRODUCING TITANIUM DIOXIDE BY

                                     CHLORIDE-RUTILE PROCESS
                                                                                               (1)



Pollutant


TSS
Iron
Chromium
Lead
Nickel
Zinc
STREAM #2

Pit Solids arri
Distillation Bottoms
Unit
Flow Cone.,
(m3/kkg) (mg/1)
10,9
6903
1348
112
3.53
3.46
2.12
Unit
Load
(kg/kkg)

75.2
14.7
1.2
0.04
0.04
0.02
STREAM #5
Scrubber and
Contact Cooling Water
Unit Unit
Flow Cone., load
(m3/kkg) (mg/1) (kg/kkg)
80.1
314 25.2
143 11.5
0.11 0.01
0.009 0.001
0.016 0.001
0.13 0.01
CALCULATED ESTIMATE


Total Raw Waste
Unit
Flew Cone.,
(m3/kkg) (mg/1)
91
1103
288
13.3
0.5
0.5
0.3
Unit
Load

100.4
26.2
1.21
0.041
0.041
0.03
STREAM #6

Treated Effluent
Unit Unit
Flow Cone., Load
(m3/kkg) (mg/1) (kg/kkg)
91
23 2.1
4.4 0.4
0.03 0.003
0.002 0.0002
0.005 0.0004
0.06 0.005
w
**
10
          (1)   See Figure 14-2 for location of sampling points.

-------
                                   PROCESS
                                 WASTE WATER
           HOLDING  POND
               FOR
           RETREATMENT
                                   RETENTION
                                     BASIN
                                   RETENTION
                                     BASIN
                                         '#3
                                   DISCHARGE
                                                  NaOH
                                    MIXING
                                    BASIN
                                  NEUTRALIZE
                                        RAIN  RUNOFF
                                  RAIN RUNOFF
                                                .pH ADJUSTMENT
                                           LEGEND

                                       SAMPLING POINTS

                                     *  THE TOTAL  RETENTION  TIME
                                       OF WATER  IN  THE  TWO-PONDS
                                        IS 5 DAYS.
Figure
General flow diagram at Plant #172 showing the sampling points.
Titanium dioxide (chloride process) manufacture.
                                      343

-------
  TABLE 14-6.   FLOW AND POLLUTANT CONCENTRATION DATA OF THE SAMPLED WASTE
               STREAMS FOR PLANT #172 PRODUCING TITANIUM DIOXIDE
               (CHLORIDE PROCESS)


Pollutant
TSS
Iron
Chromium
Lead
Nickel
Zinc
STREAM #1
(Raw Waste
Avg. Cone.
(mg/1)
171
2.9
0.72
0.005
0.08
0.3
Influent) (1)
Unit Load(2)
(kg/kkg)
5.93
0.10
0.03
0,0002
0.003
0.01
STREAM
#3
(Treated Effluent) (1)
Avg. Cone.
(mg/1)
6.7
0.33
0.02
0.002
0.01
0.09
Unit Load
(kg/kkg)
0.23
0.01
0.0007
0.00007
0.0003
0.003
(1)   Unit flow is 34.7 m /kkg.

(2)   Unit load equals the product of the pollutant concentration in iflg/1 and
     the unit flow in m3/]ckg divided by a conversion factor of 1000.
                                     344

-------
Waste Load (kg/year)

             241,000
               8,200
               8,500
               7,800
          Pollutant

          Chromium
          Lead
          Nickel
          Zinc

Pollution Abatement Options

Toxic Pollutants of Concern

The  dominant  toxic pollutant observed in the raw waste effluents for
the Titanium  Dioxide  (Chloride  Process)  Subcategory  is  chromium.
Chromium  was found at treatable concentrations at both plants sampled
in the screening and verification phase.   Lead, nickel, and zinc  were
found in the raw waste of Plant #559 at treatable levels, but were not
present  in  the  Plant  #172  raw waste.   At Plant #559, the chloride
process waste effluents are  mixed  with  the  sulfate  process  waste
effluents  before  treatment.  It is likely that the three major toxic
pollutants found were contributed by the sulfate process wastes  since
it  uses  a  low  purity  ore  (ilmenite).   At Plant #172, the solids
generated from the chloride process (which consist of solid  particles
of  unreacted ore, coke, iron, and other heavy metals) are hauled to a
landfill for disposal.  It is probable that most of  the  toxic  metal
pollutants  are present in this solid waste and hence do not appear in
the wastewaters.

Process Modification and Technology Transfer Options

A.   Research to develop economical techniques to recover the vanadium
     and other metal values  from the solid wastes generated  from  the
     process waste treatment system would appear to be a fruitful area
     of investment.

B.   New plants can  utilize  refrigeration  and  high  pressures  for
     chlorine  liquefaction.   This  would  reduce  or  eliminate  the
     chlorine residual problem in the tail gases.  The capital cost to
     modernize old plants  is high, but  these  plants  should  have   a
     caustic  soda  or  lime  scrubber  instead of a water scrubber to
     remove residual chlorine from the tail gases.   Caustic  or  lime
     scrubbing  removes a  significant portion  of the chlorine from the
     tail  gases   in  view  of  analagous  data  from   the   chlorine
     subcategory presented in Section 11.

C.   Best Management Practices

     Provision should be made at all plants  to  collect  storm  water
     runoff from the plant site and send  it to the treatment facility.
     Three  out  of  a  total  of  six  existing  plants are presently
     treating storm water  runoff.
     345

-------
TABLE 14-7.  RAW TOSTE POLUJTANT DATA SUM-JARY OF THE SAMPLED STREAMS

SUBCATEGORY:
Average
Pollutant
Iron
ChronLum
Lead
Nickel
Zinc
TSS
TTTANUM DICKIDE (CHLORIDE
PROCESS)
Daily Pollutant Concentration and loadings at Plants
kg/kkg of TiO.
(mg/1)
Plant
#559
26.2
(288)
1.21
(13.3)
0.041
(0.5)
0.041
(0.5)
0.03
(0,3)
100.4
(1103)

Plant v J }
#172 v '
0.10
(2.9)
0.03
(0.72)
0.0002
(0.005)
0.003
(0.08)
0.01
(0.3)
5.93
(171)

Sampled
Overall
Average
13.15
0.62
0.021
0.022
0.02
53.17
                                    346

-------
D.   Prevailing Control and Treatment Practices

     At Plant #172, the solid wastes consisting of spent ore and  coke
     are   hauled   to  an  off-site  landfill.   Process  wastewaters
     consisting of scrubber and contact-cooling effluents and  a  part
     of  the  surface runoff are sent to a mixing basin where they are
     neutralized with caustic soda.  The effluent from  the  basin  is
     then  sent  to  two  retention  ponds  in  series.  Additional or
     residual  rain  water  runoff  is  added   to   the   ponds   for
     clarification.   The overflow from the last pond is monitored and
     discharged to a surface stream.  At Plant #559,  the  wastewaters
     from  both  chloride  and sulfate processes are combined with the
     weak acid effluent from the  sulfate  process  in  a  pond.   The
     overflow from the pond is neutralized with limestone and oxidized
     with  air  for the removal of iron.  The wastewater is then mixed
     with the neutralized strong acid waste (from the sulfate process)
     and  scrubber  waters  (from  both  the  chloride   and   sulfate
     processes),  and neutralized with lime in a reactor and sent to a
     final settling pond.  The overflow from the  pond  is  the  final
     discharge.

     At   Plant  #199,  all  the  process  wastewaters  are  combined,
     including storm water  and  sanitary  wastewater.   The  combined
     wastewater is sent to a four-stage neutralization system, and the
     effluent  from  each of the four stages of neutralization is sent
     to a thickener.  The thickener overflow   is  transferred  to  the
     first  of  three  settling  ponds, also in series.  The underflow
     from  the  thickener  is  heated  to   improve    its   filtration
     characteristics  and  filtered  in four rotary drum filters.  The
     thickened solids from the filters are disposed of in  a  landfill
     and the filtrate is combined with wash water and vacuum pump seal
     water  prior  to  being  recycled  to  the  fourth  stage  of the
     neutralization train.  The overflow from  the last  settling  pond
     is discharged.

     The  process wastewater streams at Plant  #102 are received in two
     tanks, neutralized with lime, and then sent to a settling  basin.
     The  settled  solids  are  retained in the settling lagoons.  The
     plant has future plans  for  treating  boiler  blowdown,  cooling
     tower blowdown, leaks, and spills with the process wastewater.

     At  Plant  #605,  the unreacted ore and coke are disposed of as a
     solid waste in the pit.   The  wastewater  from   the  process  is
     passed  to  two tanks for flow equalization and the water  is then
     reacted with  ground limestone slurried   in  water.   The  treated
     solution  is  centrifugally treated to remove coarse solids which
     are separated and landfilled.  A flocculating agent is  added  to
     the  centrate  and  the  solution  is  sent  to a clarifier.  The
     clarifier overflow  is degassed and the pH adjusted  with  caustic
     soda (if required) before discharge.
                                    347

-------
E.   Advanced Treatment Technologies

     Neutralization and settling are practiced for  the  treatment  of
     chloride  process  raw waste effluents at all the five plants for
     which  308  data   are   available.    Air   oxidation,   sulfide
     precipitation,  xanthate precipitation, and ion exchange might be
     applied  to  the  clarified  solutions  for  control  of  metals.
     Sulfide  precipitation  or  the xanthate process could be used to
     provide additional removal of zinc, lead, and nickel.

Selection of Appropriate Technology and Equipment

Technologies for Different Treatment Levels

A.   Level 1 (BPT/BAT)

     The chloride process wastes are equalized, neutralized with  lime
     to a pH range of 6 to 9, and settled in lagoons before discharge.
     Level  1  treatment  is typical of industry practice and for this
     reason was selected as the technology basis for BPT regulations.

B.   Level 2 (NSPS)

     Alkaline precipitation as a second-stage  lime  treatment  to  an
     optimum  pH (9 to 10) is added to Level 1 to precipitate metallic
     hydroxides, which are then filtered before discharge.  Filtration
     removes traces of metallic hydroxides which do not separate in  a
     gravity system.  Level 2 technology was selected as the basis for
     the  proposed  BAT  limitations because it provides an economical
     method for the removal  of  additional  toxic  metals.   For  the
     reasons  discussed  below  under Basis for New Source Performance
     Standards, the Agency did not select this option for BAT  in  the
     promulgated regulation, but did select it for NSPS.

C.   Level 3

     Ferrous sulfide treatment is added ahead of the Level 2 filter to
     precipitate the heavy metals more effectively.

     Alkaline precipitation was chosen as Level 2 because  it  readily
     supplements  existing  lime neutralization by the simple addition
     of filtration and increasing the Level 1  lime  dosage.   Sulfide
     precipitation  was  chosen  at  Level  3  because  it  provides a
     polishing treatment for most residual heavy metals beyond Level 2
     treatment.

     Figures 14-4, 14-5, and 14-6 show  the  model  treatment  systems
     adopted for the chloride process wastes.
                                  348

-------
Equipment for Different Treatment Levels

A.   Equipment Functions

     BPT treatment is essentially lagooning with lime  neutralization,
     using no special equipment except a lime feeder and mixer.

     In  Level  2,  second stage lime treatment is followed by gravity
     clarification  and  multi-media  filtration,  with  necessary  pH
     controls.

     In Level 3, ferrous sulfide is added ahead of the Level 2 filter,
     to  react  with residual heavy metals more completely than in the
     alkaline precipitation step at Level 2.

B.   Chemicals and Handling

     Lime and hydrochloric acid are fed with conventional equipment at
     all levels, and ferrous sulfide is  prepared  on-site  by  mixing
     ferrous  sulfate with sodium bisulfide.  When normal dust control
     and good ventilation are used, there should be no adverse effects
     from handling these chemicals, although care should be taken that
     hydrogen sulfide gas is not generated.

C.   Separation and Removal of Solids

     Inert ore fractions and precipitated solids  are  accumulated  in
     clay-lined  lagoons,  which  are alternately drained.  Solids are
     mechanically removed to self-draining  18 ft. high  storage  piles
     on  land provided at the site for a 10-year operatirig period.  At
     Levels 2 and 3, small amounts of heavy metal precipitates in  the
     clarifier  underflow  are  filter  pressed and hauled to a secure
     landfill.

Treatment Cost Estimates

General Discussion

To determine  the treatment cost, a model plant concept was  developed.
A  raw  waste unit flow was selected and pollutants to be treated were
selected based on the treatment system data  available  for  the  five
Ti02  plants  and  the  screening  and  verification sampling program.
Three production  levels  were  then  selected  to  cover  the  entire
subcategory   range.   Treatment costs for BPT and NSPS were calculated
for each of the model plant production ranges using the unit flow  and
unit  pollutant loads.  The unit flow of 100 mVkkg used for the model
plant in regulation development has been selected to be representative
of the subcategory and it is assumed that the imreacted ore  and  coke
are  slurried  and  sent  to  the  treatment  system   instead  of being
disposed of  in a  landfill as a solid waste.
                                 >49

-------
                                                      LIME
                 RAW
            WASTE WATER
                               LAGOON
                               LAGOON
I
                                                  MIXING
                                                              LAGOON
                                                              LAGOON
                                                                                                          EFFLUENT
w
ui
o
                               Includes flow monitoring, pH monitoring and sampler.
                        Figure 14-4.  Level 1 (BPT/BAT) waste water treatment for titanium dioxide —

                                      chloride process.

-------
                                                                    BACKWASH
U)
UI
                 P.AW
               WASTE WATB
                                                                                             Q pH ADJUSTMENT
                                                                                                    * EFFLUENT
                                                                   SUMP
                                                      TO LANDFILL
                              Include* flow monitoring, pit monitoring and •ampler,
                   Figure 14-5.  Level 2  (NSPS)  waste water  treatment for  titanium dioxide —  chloride
                                   process.

-------
                                                                        FERROUS
                                                                        SULFATE
 SODIUM
BISULFIDE
   HAW
WASTE WATER
                                                                 SUMP
                                                 TO LANDFILL
                   Includes flow monitoring, pH monitoring and sampler.
      Figure  14-6.   Level  3 waste water treatment  for titanium dioxide  —   chloride
                     process.

-------
A.   Wastewatec Flow

     The unit waste effluent flow varies from 29.3 to 110.0 mVkkg  of
     TiOz  for  the  four  plants as shown in Table 14-4.  The primary
     reason for the variation in the flow is that some  plants  slurry
     the  spent  ore  and coke (solid waste from chloride process) and
     send it to the treatment system, and others haul the  dry  solids
     to  a  landfill.   The  flow  variation  is also dependent on the
     difference in chlorine recovery process from the tail gas and the
     amount of scrubbing liquid used.  Small variations in  flow  also
     result  from  the  finishing  operation which is dependent on the
     type of titanium dioxide end product desired.  Plants #559   (unit
     flow  of  91  mVkkg) and #199  (unit flow of 110 mVkkg) send the
     solid waste from  the  manufacturing  process  to  the  treatment
     facility.   It  is  assumed  for treatment system cost estimation
     that the solids are  included   in  the  raw  waste  flow  to  the
     treatment system.  A constant unit flow of 100 mVkkg of Ti02 has
     been  used  for the model plants, which is an average of the unit
     flows of Plants #559 and #199.

B.   Pollutant Load

     The primary pollutants occurring in the wastewater are  suspended
     solids,  acidity, and the chlorides of ferric iron, chromium, and
     other trace metals.  The suspended solids  (TSS)  loading  values
     for Plants #559 and #172 are 100.4 and 5.93 kg/kkg of Ti02 (Table
     14-7).   The low value represents-a plant that hauls ore and coke
     off-site, while the high value  is believed to be too low  due  to
     nonrepresentative  sampling.    The  amount of solids produced are
     higher  than  the  values  indicated  for  the  sampled   plants.
     Consequently,  a higher suspended solids loading of 500 kg/kkg of
     Ti02  (reported  in the 308 data  from Plant #199)  is  assumed  for
     the  model  plant.   To establish treatment chemical requirements
     and related costs, the toxic pollutant  loadings  for  the  model
     plant  are  taken  as  the  average  values of the unit pollutant
     loadings of the plants sampled  in the screening and  verification
     program  (Table  14-6) and the selected pollutant values are:

     Pollutant      Unit Loading (kq/kkq of Tip,)

     Chromium               0.63
     Lead                   0.021
     Zinc                   0.020
     Iron                  13.15
     Nickel                 0.022


C.   Production Rates

     Six plants produce titanium dioxide from rutile ore  or  upgraded
     ilmenite  ore   using  the  chloride process at a total production
     rate of  142,000 metric tons per year.  Production ranges  from   a
     minimum of  16,900 kkg/year to a maximum of 45,200 kkg/year with  a
                                  353

-------
     mean  of  28,400  kkg/year  and a median of 25,600 kkg/year.  For
     wastewater treatment cost estimates, three production levels were
     selected as model plants.   These  are  16,900  kkg/year,  25,500
     kkg/year,  and 45,200 kg/year.  This range of production includes
     all United States plants.

Model Plant Control and Treatment Costs

The estimated costs for the three models having  different  production
levels  are given in Tables 14-8, 14-9, and 14-10.  The costs shown at
each level of treatment correspond to the model plant  BPT  (Level  1}
system  and  higher level (NSPS) system which may add to or modify the
existing BPT system to meet more  stringent  toxic  pollutant  removal
requirements.   The  higher  level  (NSPS)  also  furnishes  a  better
effluent quality with respect to the conventional  and  nonconventinal
parameters.  Tables 14-11, 14-12, and 14-13 present the estimated NSPS
treatment  costs  for  the  three  models  having different production
levels.   For  model  plants  at  the   base   level   of   treatment,
amortization,  chemicals, and the residual waste disposal costs have a
significant impact on the  total  annual  costs.   At  NSPS  treatment
level,  amortization,  chemicals, and labor constitute a major portion
of the additional annual costs.  Table 14-14 presents a summary of the
unit cost distribution between  amortization  and  the  operation  and
maintenance  cost components at various production rates and levels of
treatment.

Basis For Regulations

Evaluations of BPT Treatment Practices

All the plants producing titanium  dioxide  by  the  chloride  process
using  rutile ore or upgraded ilmenite ore practice neutralization and
settling  for  control  and  treatment  of  the  waste  effluents.   A
variation in the effluent quality is expected because of the method of
handling  the  unreacted ore and coke  (generated as solid residue from
the chloride process).  Two of the six plants haul the  residue  to  a
secure  landfill  for  disposal while the remainder slurry the residue
with water and send it to the treatment  system.   No  information  is
available about recycling the treated wastewater at any of the plants.

A.   Pollutant Removal with BPT Treatment

     Treatment Level 1 is equivalent to BPT in  the  Ti02  subcategory
      (chloride process).

     Plants  #559 and #172 practice neutralization and settling of the
     raw waste.  At Plant #559, the chlride process raw wastewater  is
     mixed  with  the sulfate process wastewater for treatment.  Also,
     at Plant #559 the spent ore and coke   (solid  residues   from  the
     chloride  process)  are  slurried  with  water  and  sent   to the
     treatment facility, whereas at Plant #172  the solid residues  are
     hauled  to  a chemical landfill.  Long-term treated effluent data
     have been submitted by both plants 4559 and #172.  The derivation
                                 354

-------
    TABLE 14-8.  MODEL PLANT TREATMENT COSTS
Subcategory
Production
Titanium dioxide - Chloride process
16,900 metric tons per year
A. INVESTMENT COST

     Site development  	
     Equipment 	
     Monitoring equipment  .,

          Subtotal  .........
     Contractor's O &  Pb....

          Subtotal  	
     Engineering  	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL INVESTMENT  COST  .


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  .
     Energy  	
     Chemicals 	
     Maintenance  	
     Taxes and insurance ...
     Residual waste disposal
     Monitoring, analysis,
      and reporting 	
                                                     ($)
                               BPT

                           175,600
                           205,000
                            20,000

                           400,600
                            60,090

                           460,690
                            92,138

                           552,828
                            55,283

                           608,111
                           192,000

                           800,111
     TOTAL OPERATION AND
     MAINTENANCE COST
                             56,000
                              5,900
                            140,000
                             60,811
                             24,003
                            108,000

                             15,000
                            409,714
BAT

  0
  0
  0

  0
  0

  0
  0

  0
  0

  0
  0
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                             98,940

                            508,654
  Represents the  incremental  cost  above  that  for  BPT treatment
  Overhead and Profit
                              355

-------
    TABLE  14-9.  MODEL PLANT TREATMENT  COSTS
Subcategory
Production
Titanium dioxide - Chloride process
25,500 metric tons per year
A. INVESTMENT COST

     Site development  	
     Equipment  	
     Monitoring equipment  ..

          Subtotal  	£ • • •
     Contractor's 0 &  P 9...

          Subtotal  	
     Engineering 	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL INVESTMENT  COST  .


B. OPERATION AND
   MAINTENANCE  COST

     Labor and  supervision  .
     Energy 	
     Chemicals  	
     Maintenance 	
     Taxes and  insurance ...
     Residual waste disposal
     Monitoring, analysis,
      and reporting 	
                               BPT

                           244,100
                           215,000
                            20,000

                           479,100
                            71,8*55

                           550,965
                           110,193

                           661,158
                            66,116

                           727,274
                           276,000

                         1,003,274
     TOTAL OPERATION AND
     MAINTENANCE COST
                            56,000
                             6,800
                           211,000
                            72,727
                            30,098
                           164,000

                            15,000
                           555,626
                                                     ($)
BAT

  0
  0
  0

  0
  0

  0
  0

  0
  0

  0
  0
                                                                a
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                           118,327

                           673,953
a Represents the incremental cost  above  that  for  BPT treatment
  Overhead and Profit
                              356

-------
   TABLE 14-10.  MODEL PLANT TREATMENT COSTS
Subcategory    Titanium dioxide - Chloride process
Production     45,200 metric tons per year
A. INVESTMENT COST                            BPT          BAT

     Site development 	         395,000             0
     Equipment 	         275,000             0
     Monitoring equipment  	          20,000             0

          Subtotal 	         690,000             0
     Contractor's 0 & P b	         103,500             0

          Subtotal 	         793,500             0
     Engineering	         158,700             0

          Subtotal 	         952,200             0
     Contingencies	          95,220             0

          Subtotal 	        1,047,420             0
     Land	         504,000             0

     TOTAL INVESTMENT COST 	        1,551,420             0
B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  	           56,000             0
     Energy 	           86,000             0
     Chemicals 	         374,000             0
     Maintenance 	         104,742             0
     Taxes and insurance  	           46,543             0
     Residual waste disposal  ....         294,999             0
     Monitoring, analysis,
      and reporting	           15,000             0

     TOTAL OPERATION AND
     MAINTENANCE COST                     977,284             0
C. AMORTIZATION OF
   INVESTMENT COST                        170,415             0

   TOTAL ANNUAL COST                    1,147,699             0

a Represents the incremental cost above that for BPT treatment
  Overhead and Profit
                              357

-------
   TABLE 14-11
  MODEL PLANT TREATMENT COSTS
Subcategory
Production
Titanium dioxide - Chloride process
16,900 metric tons per year
       A. INVESTMENT COST

            Site development  	
            Equipment 	
            Monitoring equipment  ..

                 Subtotal 	
            Contractor's 0 & Pa....

                 Subtotal 	
            Engineering  	

                 Subtotal 	
            Contingencies 	

                 Subtotal 	
            La nd	

            TOTAL INVESTMENT COST  .


       B. OPERATION AND
          MAINTENANCE COST

            Labor and supervision  .
            Energy 	
            Chemicals 	
            Maintenance  	
            Taxes and insurance  ...
            Residual waste disposal
            Monitoring, analysis,
             and reporting 	
                                       (S)
                                     NSPS

                                   175,600
                                   724,000
                                    20rOOO

                                   919,600
                                   137,940

                                1,057,540
                                   211,508

                                1,269,048
                                   126,905

                                1,395,953
                                   192,000

                                1,587,953
            TOTAL OPERATION AND
            MAINTENANCE COST
                                   140,000
                                    11,500
                                   174,000
                                   139,595
                                    47,639
                                   117,000

                                    15,000
                                  644,734
          AMORTIZATION OF
          INVESTMENT COST

          TOTAL ANNUAL COST
                                   227,122

                                   871,855
  Overhead and Profit
                             358

-------
   TABLE 14-12.  MODEL PLANT TREATMENT COSTS
Subcategory
Production
Titanium dioxide - Chloride process
25,500 metric tons per year
       A. INVESTMENT COST

            Site development  	
            Equipment  	
            Monitoring equipment  ..

                 Subtotal  	
            Contractor's 0 & P3 ....

                 Subtotal  	
            Engineering  	

                 Subtotal  	
            Contingencies  	

                 Subtotal  	
            Land	

            TOTAL INVESTMENT COST  .


       B. OPERATION AND
          MAINTENANCE COST

            Labor and supervision  .
            Energy 	
            Chemicals  	
            Maintenance  	
            Taxes and insurance ...
            Residual waste disposal
            Monitoring, analysis,
             and reporting 	
                                      ($)
                                      NSPS

                                   244,100
                                   853,000
                                    20,000

                                 1,117,100
                                   167,565

                                 1,284,665
                                   256,933

                                 1,541,598
                                   154,160

                                 1,695,758
                                   276,000

                                 1,971,758
            TOTAL OPERATION AND
            MAINTENANCE COST
                                  140,000
                                   13,500
                                  262,000
                                  169,576
                                   59,153
                                  175,000

                                   15,000
                                  834,229
       C. AMORTIZATION OF
          INVESTMENT COST

          TOTAL ANNUAL COST
                                  275,900

                                1,110,128
  Overhead and Profit
                             359

-------
  TABLE 14-13.  MODEL PLANT TREATMENT COSTS
Subcategory
Production
Titanium dioxide - Chloride
45,200 metric tons per year
process
       A. INVESTMENT COST

            Site development  ......
            Equipment  	
            Monitoring eguipment  ..

                 Subtotal  	
            Contractor's 0 & Pa....

                 Subtotal  	
            Engineering  	

                 Subtotal  	
            Contingencies  	

                 Subtotal  	
            Land	

            TOTAL INVESTMENT COST  .


       B. OPERATION AND
          MAINTENANCE COST

            Labor and supervision  .
            Energy 	
            Chemicals  	
            Maintenance  	
            Taxes and insurance ...
            Residual waste disposal
            Monitoring,  analysis,
             and reporting 	
                                       (S)
                                     NSPS

                                   395,000
                                 1,215,000
                                    20,000

                                 1,630,000
                                   244,500

                                 1,874,500
                                   374,900

                                 2,249,400
                                   224,940

                                 2,474,340
                                   276,000

                                 2,750,340
            TOTAL OPERATION AND
            MAINTENANCE COST
                                   140,000
                                    19,000
                                   469,000
                                   247,434
                                    82,510
                                   314,000

                                    15,000
                                 1,286,944
       C. AMORTIZATION OF
          INVESTMENT COST

          TOTAL ANNUAL COST
                                   402,575

                                 1,689,519
  Overhead and Profit
                             360

-------
     of the variability factors for daily maximum and 30-day
     for both plants are given, in Tables 14-15 and 14-16.
averages
     The  concentration  of  the  raw waste and treated effluent along
     with the percent removal  of  the  pollutants  by  the  treatment
     system  for  Plants  #559  and  #172 sampled in the screening and
     verification program are given in Table 14-17.

     In the chloride process, most iron is in the ferric state and  is
     readily  removed  by alkaline precipitation.  Toxic metal removal
     is also improved, since ferrite coprecipitation will occur  which
     tends   to  provide  a  better  effluent  quality  than  alkaline
     precipitation alone, as observed at Plant #559.

Basis for BPT Effluent Limitations

A.   Technology Basis

     For BPT,  the  Agency  is  basing  limitations  on  equalization,
     neutralization,  and  settling  or  clarification.  All plants in
     this segment of the industry have BPT technology installed.

B.   Flow Basis

     The flow going to the treatment system at different plants varies
     and is dependent on the method of disposal of the spent  ore  and
     coke  (pit solids) and on the finishing operation.  The spent ore
     and coke are either hauled to a landfill as solid residue or sent
     to the treatment system.  For the model plant  treatment  system,
     the  solid residues from the manufacturing process are assumed to
     be slurried with water and sent to the treatment  system.   Plant
     #559  and  #199  do,  in  fact,  send  the  solid residues to the
     treatment system.  The model plant treatment system is  based  on
     an inflow rate of 100 mVkkg of Ti02 which  is an average value of
     the  effluent flow of plants #559 and #199.  The treated effluent
     flow is assumed to be the same as the influent flow.   The  water
     added  or  removed  in  the  treatment  system  through  chemical
     addition, precipitation, and evaporation have been neglected,  as
     it  varies  from plant to plant and is dependent on the selection
     of treatment chemicals as well  as  climatic  conditions  and  is
     insignificant in comparison to the total flow.

C.   Selection of Pollutants to be Regulated

     The selection  of  pollutants  for  regulation  is  based  on  an
     evaluation  of the waste data from the screening and verification
     sampling program.  The two  major  factors  considered  were  the
     individual   plants   raw  waste  concentrations  and  the  total
     subcategory pollutant loadings.
                                   361

-------
        TABLE 14-14.MODEL  PLANT  UNIT  TREATMENT COSTS
     Subcategory Titanium dioxide  -  Chloride process
                                 Annual  Treatment Costs ($/kkg)
                                      LEVEL OF TREATMENT
COST ITEM PRODUCTION
(kkg/yr)
Annual Operation
and Maintenance


Annual
Amortization


Total Annual
Cost


16,900
25,500
45,200

16,900
25,500
45,200

16,900
25,500
45,200
BPT
24.24
21.79
21.62

5.85
4.64
3.77

30.10
26.43
25.39
BAT*
NA
NA
NA

NA
NA
NA

NA
NA
NA
NSPS
38.15
32.71
28.47

13.44
10.82
8.91

51.59
43.53
37.38
*Represents the incremental cost above  BPT
                              362

-------
                1ABLE 14-15. HISTORICAL EFFLUENT 1CNITORING
                    DATA SUMMARY WETH VARIABILITY FACTOR
                        TREATMENT WTTH IPDN REMOVAL
                               Daily Measurements
                        Subcategory:  Titanium Dioxide
                         Chloride Process (Rutile Ore)
                                  Plant 1559
                         April 76 through SepteKiber 78


TSS
Daily Data(1)
No. of Points 889
Average x, ppn 21
Standard gg g
Deviation, S "
Standard , ^A
Deviation, S'
Variability 11 n
Factor •"""
30-day (1)
Averages
No. of Points 30
Standard 21 8
Deviation
Variability . 04
Factor *'™
f2)
Variability
Factor Ratio
Pollutant
Cadmium Chrcmium Iron Lead
109 128 854 128
0.058 0.072 0.620 0.068
0.044 0.054 3.46 0.041
0.68 0.67 1.86 0.56
3.85 3.81 13.7 3.2
26 30 28 30
0.042 0.038 0.94 0.04
2.4 2.04 4.0 2.1


Nickel Zinc
128 128
0.080 0.15
0.07 0.20
0.76 1.02
4.4 6.4
30 30
0.05 0.16
4.4 3.1

VFR
3.6
1.6
1.9
3.4
1.5     1.0     2.1
(!)  Section 8.2 presents a discussion of the approach and methodology employed
    in the statistical evaluation of data.
(2)  VFR is the ratio of the variability factor for daily measurements to the
    variability factor for rconthly averages.
                                    363

-------
TABLE 14- 16.  HISTORICAL EFFLUENT MONTTORJKG DATA SUMMARY WITH VARIABILITY
                                  FACTORS
                             DAILY MEASUREMENTS
SUBCATBQQRY:  TTEANIUM DIOXIDE-Chloride Process  — Plant 1172
              (Rutile/Upgraded Ilmenite Ore)
                                               Pollutant

Daily Data '
No. of Points
Average x, ppn
Standard deviation, S
Standard deviation, S'
Variability factor
30-Day Averages
No. of Points
Standard deviation, S
Variability factor
O
Variability Factor Ratio *
VFR
TSS
454
5.39
9.13
1.16
7.6
15
6.31
2.92
!)
2.6
Chromium
454
0.0080
0.016
1.27
8.6
15
0.012
3.46
2.5
Copper
454
0.020
0.030
1.08
6.9
15
0.028
3.29
2.1
Zinc
454
0.020
0.027
1.02
6.4
15
0.026
3.13
2.1
 (1) Section 8.2 presents a discussion of the approach and methodology employed
    in the statistical evaluation of data.

 (2) VFR is the ratio of the variability factor for daily measurements to the
    variability factor for monthly averages.
                                    364

-------
TABLE 14-17.  TREATMENT PERFORMANCE DATA OF SAMPLED PLANTS #559 AND #172
SOBCATEGORY:  TITANIUM DICKIDE-Chlaride Process
Pollutant
TSS
Iron
Chromium
Lead
Nickel
Zinc
Plant #559
Pollutant
Concentration
(mg/U
Paw Treated
Waste Effluent
1103 23
288 4.4
13.3 0.030
0.50 0,002
0.50 0.005
0.30 0.060
Percent
Rsnoval
97.9
98.5
99.8
99.6
99.0
80.0
Plant #172
Pollutant
Concentration
Cmg/D
Raw Treated
Waste Effluent
171 6.7
2.9 0.33
0.72 0.020
0.005 0.002
0.08 0.010
0.30 0.090
Percent
Removal
96.1
88.6
97.2
60.0
87.5
70.0
                                    365

-------
          Raw Waste Concentration

          Plant #559 was visited in the screening phase  for  sampling
          of  the raw and treated wastewater.   For each pollutant,  the
          maximum concentration observed gave a preliminary indication
          of its pollution  potential.    Five  pollutants  were  found
          above treatability levels in  the raw waste of Plant #559  and
          they  were:  chromium,  iron,  nickel,   lead,  and zinc (see
          above).   A  second  plant,   #172,   was   sampled   in   the
          verification phase and chromium was the only pollutant found
          above  treatability levels in the raw waste.  At Plant #559,
          the wastewater from the chloride process is mixed  with  the
          sulfate   process   wastewater   and  the  chloride  process
          effluents were sampled at the point of mixing.  It is highly
          probable that the sampled waste included the sulfate process
          effluent  impurities  since  the  sulfate  process  for  the
          manufacture  of  TiO,  uses
                                  an
ore  of  lesser purity.  In
D.
     addition,  control of chromium should also  provide  adequate
     removal  of  nickel, lead,  and zinc which may coincidentally
     occur at treatable levels.   For these reasons, nickel, lead,
     and  zinc  will  not  require  effluent  limitations.     The
     discussion  of  wastewater  treatment chemistry in Sections 7
     and 8 presents the  basis  for  selecting  one  toxic  metal
     control  parameter  which effectively ensures the control of
     the  other  metals  present.    The  conventional  and  toxic
     pollutants  of  concern  include  chromium and TSS.   Iron, a
     nonconventional  pollutant,  is  also  present  as  a  major
     impurity  in  the  rutile  or  upgraded ilmenite ore and was
     found at treatable levels in Plants #559 and #172.

2.    Total Subcategory Raw Waste Pollutant Loading

     Chromium was the only toxic pollutant found  in  significant
     concentrations  in  the  raw waste of both plants sampled in
     the screening and verification phase.  The average unit  raw
     waste chromium loading (Table 14-7) obtained from the plants
     sampled   was  multiplied  by  the  total  Ti02  subcategory
     production by the chloride  process to evaluate  the  overall
     magnitude  of  the  pollutant potential for the subcategory.
     The value of 241,000 kg/year of chromium discharged  by  the
     subcategory  in  the  effluent  indicated  the  necessity of
     control of this pollutant.

     The treatment technology practiced by the  industry  removes
     the  chromium  and  iron  to  low  levels  as  seen from the
     effluent quality of the plants sampled and as shown in Table
     14-17.

Basis of Pollutant Limitations

1.    Conventional and Nonconventional Parameters
                                 366

-------
PH

The treated effluent is to  be  controlled  within  the
range  of  6.0 to 9.0.   This limitation is based on the
data  presented  in  Appendix   B   of   the   proposed
Development Document (60) and the JRB Study (52).

TSS Long-term effluent data is availabe  for  TSS  from
Plants  #559  and  |172.   At  Plant  #172,  the  solid
residues from the manufacturing process are sent  to  a
landfill.   Although  the  amount of solids sent to the
treatment system at Plant #559 is  high  compared  with
that   selected   for   the  model  plant  (because  of
intermixing of sulfate waste),  the  long-term  average
concentration  of 21 mg/1 (Table 14-15) calculated from
the monitored data submitted by Plant #559 is  selected
as the treatment performance basis for the subcategory.
The  variability  factors of the daily measurements and
monthly  averages   {11  and  3.04)  derived  from   the
long-term  data  of plant #559 and given in Table 14-15
are used to calculate  the  concentration  basis.   The
unit  effluent  limitations  are  calculated  using the
model plant unit flow of  100 mVkkg.  The  calculations
are given below:

The maximum 30-day average concentration is:

(21 mg/1) (3.04) =  64 mg/1

The 24-hour maximum concentration is:

{21 mg/1) (11) = 230 mg/1

The maximum 30-day average effluent limit  is:

(64 mg/1) (100 mVkkg)  (  kg/m3  )  =6.4  kg/kkg
                        (1000 mg/1)

The 24-hour maximum effluent limit  is:

(230 mg/1)  (TOO mVkkg)  (  kq/m^  ) = 23 kg/kkg
                         (1000 mg/1)

Iron

The Agency has decided not to promulgate limitations on
iron for existing sources because of the increased cost
of treatment  when iron  is controlled  and  because  two
plants  operate  both  the  chloride  process  and  the
sulfate process and send  wastewater from both processes
to the same treatment facility.  In order  for these two
plants to treat  the  chloride  process  wastewater  to
remove   iron,  they would either also have to treat the
                     367

-------
2.
     sulfate process wastewater to remove iron or  undertake
     a  massive  reconstruction  of  the treatment facility.
     Such a reconstruction or removal of iron in the sulfate
     process  would  wipe  out  the  recycle  benefits   and
     treatment  cost  reduction  associated  with  the final
     BPT/BAT limitations for the sulfate process, and  would
     probably  result  in  the  closure  of  the two sulfate
     process lines, with attendant increase in unemployment.
     Control of toxic metal pollutants will be  adequate  at
     existing  plants even without an iron removal step.  We
     are providing guidance for use  by  permit  writers  in
     cases   where   control   of   iron   is  warranted  by
     water-quality considerations.

     The  subcategory  performance  standard  of  0.62  mg/1
     selected  for iron is based on the long-term average of
     the  effluent  data  submitted  by  Plant  #559  (Table
     14-15).

     For the model plant, it is assumed that iron is present
     in   the  ferric  state  in  the  raw  waste  from  the
     chlorination process.  Using the variability factor  of
     the daily measurement of 4.0 and the variability factor
     of  the  monthly  averages  of  13.7 estimated from the
     long-term monitored effluent data  of  Plant  #559  for
     iron  (Table  14-15),  and the model plant unit flow of
     100 m3/kkg, the concentration basis is determined.   The
     limitations can be developed in  the  event  that  iron
     becomes a concern.  The concentrations are developed as
     follows:

     The maximum 30-day average concentration basis is:
     (0.62 mg/1) (4.0) = 2.5 mg/1

     The 24-hour maximum concentration basis is:

     (0.62 mg/1) (13.7) = 8.5 mg/1

     The maximum 30-day average concentration of 2.5 mg/1 is
     higher  than  the  achievable effluent concentration of
     1.6 mg/1 reported in the literature (10).   The  latter
     concentration  is  based  on  the  performance  of lime
     neutralization  follwed  by  settling  of   acid   mine
     drainage  waste-,  and  may  not be appropriate for this
     subcategory.

Toxic Pollutants

Chromium is the only regulated toxic  pollutant  because  of
its  presence  at  treatable  levels in the raw waste of the
plants surveyed.
                            368

-------
Chromium:

The chromium limitation  is  based  primarily  on  long-term
monitoring data available for Plant 1559.  A waste treatment
system  influent concentration for chromium of 13.3 mg/1 was
observed at Plant #559 during screening.  The long-term data
summarized in Table  ]4-15  indicates  a  current  treatment
performance  of  0.070  mg/1 and a monthly average and daily
measurement variability factor of 2.0 and 3.8, respectively.
Review of long-term average treatment data for  chromium  in
Tables  8-12  and  8-13  indicates  that  0.070  mg/1  shows
excellent  removal  when  compared  to  the  same  treatment
technology  for  chromium  removal  in  other industries.  A
long-term average of 0.15 mg/1 for chromium is used  as  the
basis  for the final limitation.  This value is in agreement
with the observed performance  data  from  other  industries
summarized in Tables 8-12 and 8-13.  The monthly average and
daily  measurement  variability  factors  of  2.0  and  3.8,
respectively,   are   representative   of   BPT    treatment
performance and are used for the purpose of establishing the
limitations.

The 30-day average concentration basis  is:

(0.15 mg/1)  (2.0) = 0.30 mg/1

The daily maximum concentration basis is:


(0.15 mg/1)  (3.8) = 0.57 mg/1

The 30-day average effluent limit  is:

(0.30 mg/1)  (100 mVkkg)  (  kg/m*   ) =  0.030 kg/kkg
                          {1000 mg/1)

The daily maximum effluent limit is:

(0.57 mg/1)  (100 mVkkg)  (  kq/m^   ) *  0.057 kg/kkg
                           1000 mg/1)
Other metals
Lead, nickel, and  zinc were  found  in the  raw waste of  Plant
#559  in  the  treatability range.   They were not found  in  the
raw waste of  Plant #172.  The concentration limitations   for
the   three pollutants are given  and are intended  to  serve as
guidance  in cases  where  the  pollutants are found  to  be   of
concern.

The   long-term average and corresponding  variability factors
are based on long-term data   from  Plant  #559  and data
presented in Table 8-12 for lead, nickel, and zinc.   In  all
                       369

-------
          cases,  the long-term average observed  for  Plant  1559  was
          adjusted after evaluation of treatment performance for other
          industries summarized in Table 8-12.   The long-term averages
          for  lead,  nickel,   and  zinc are 0.15 mg/1,  0.10 mg/1,  and
          0.20 mg/1, respectively,  based  on  review  of  the  table.
          Selection  of  variability  factors  was  made directly from
          Table 14-15 for the  appropriate pollutants.  The variability
          factors corresponding to chromium (the primary pollutant  of
          concern) were selected for nickel.

               The concentration basis is determined as follows:

               The 24-hour maximum concentration for lead is:

               (0.15 mg/1) (3.2) - 0.48 mg/1

               The maximum 30-day average concentration for lead is:

               (0.15 mg/1) (2.1) = 0.32 mg/1

               The 24-hour maximum concentration for nickel is:

               (0.10 mg/1) (3.8) = 0.38 mg/1

               The maximum 30-day average concentration for nickel is:

               (0.10 mg/1) (2.0) = 0.20 mg/1

               The 24-hour maximum concentration for zinc is:

               (0.20 mg/1) (6.4) = 1.28 mg/1

               The maximum 30-day average concentration for zinc is:

               (0.20 mg/1) (3.1) = 0.62 mg/1

               The limitations for BPT presented in Table 14-18.

Basis for BCT Effluent Limitations

While   EPA  has  not  yet  proposed  or  promulgated  a  revised  BCT
methodology in  response  to  the  American  Paper   Institute  v.  EPA
decision  mentioned  earlier,   EPA is promulgating BCT limitations for
this subcategory.  These limits are identical to those for BPT because
the  only  technology  option  that  removes  significant  amounts  of
conventional  pollutants  is  not  economically  achievable.   See the
discussion  of  iron  under  BPT  limitations   above.    Removal   of
significant  additional  amounts  of  conventional  pollutants  can be
achieved in this subcategory only if iron is also removed.

As BPT is the minimal level of control required by   law,  no  possible
application  of  the  BCT  cost  tests could result  in BCT limitations
lower than those promulgated today.  Accordingly, there  is no need  to I
                                 370

-------
                     TABLE 14-18.  EFFLUENT LIMITATIONS
   TITANIUM DIOXIDE - CHLORIDE PROCESS (RUTLLE OR UPGRADED ILMENITE ORE)
        Best Practicable Control Technology Currently Available
             Waste Water Flow:  100 m3Akg of TiO0

Subcategory
Pollutant Performance 1
(mg/1)
Concentration
p. Basis Cmg/1)
Max
30-day
Avg
24-hr
Max
Effluent Limit
(kg/kkg of TiO-)
Max
30-day
Avg
24-hr
Max
Conventional and
Nonconventional 'Pollutants:
Total Suspended
 Solids

Iron

Toxic Pollutants:
 21
   (2)
0.62
    (2)
11.0/3.0

13.7/4.0
64

2.5
230

8,5
6.4
23
   (4)
   (4)

Chromium

Lead

Nickel

Zinc
m
0.15u;
(3}
0.15^'
(3)
0.10U'
(3)
0.20U'

3.8/2.0

3.2/2.1

3.8/2.0

6.4/3.1

0.30

0.32

0.20

0.62

0.57 0.030
f a\
0.48 —w
(A.}
0.38 — w
/^\
1.28 — w

0.057
(4)
—
M\
—
(4)

 (1)  VFR:  Ratio of the variability factor of the daily measurements to the
           variability factor of the monthly averages.

 (2)  Long term average based on data and variability factors of plant #559
     selected from Table 14-15.
 (3)  Limitation based on long term data at plant #559 and review of Table 8-12,

 (4)  No effluent limitation.
                                      371

-------
wait  until  EPA  revises  the BCT methodology before promulgating BCT
limitations.

Basis for BAT Effluent Limitations

The Application of Advanced Level Treatment

The advanced level technologies, viz., the use of sulfide and xanthate
as a polishing step to the base level  treatment  system  (BPT),  were
considered  for  BAT  and NSPS but were rejected on the basis of cost.
Level 1, used for BPT, is selected for BAT treatment technology.


A.   Technology Basis

     Alkaline precipitation followed by settling used for  BPT  (Level
     1) is the same as for BAT.
B.
Flow Basis
A unit wastewater flow rate of 100 m3/kkg of TiO2  used  for  the  BPT
     model plants has been selected for BAT.

C.   Selection of Pollutants to be Regulated

     Chromium is the pollutant identified for regulation.

     1.    Nonconventional Pollutants

          The concentration basis was developed for iron which can  be
          used  for  guidance  in  the  event  that  iron  becomes  an
          environmental concern.

     2.    Toxic Pollutants

          a.   Chromium

               The limitations for BAT are the same  as  selected  for
               BPT.

          b.   Other Metals

               Lead, nickel,  and  zinc  are  not  limited.   However,
               achievable concentration levels are contained
          in this document for use as guidance if these pollutants are
found to be of concern.  The values are the same as those selected for
BPT.


               Table 14-19 gives the limitations for BAT.
                                372

-------
                        TABLE 14-19. EFFLUENT LIMITATIONS
         TITANIUM DIOXIDE-CHLORIDE PROCESS (RUTILE/UPGRADED ILMENITE ORE)
                            Best Available Technology
                       Wastewater Flow:  100 m /kkg of TiO2
Pollutant
Subcategory
Performance*1
 (mq/1)
             Concentration   Effluent Limit
             Basis (mg/1)     ^kg/kkg of TiOg)
             Max.             Max.
             30-day   24-hr. 30-day      24-hr,
             Avg.      Max.   Avg.        Max.
Nonconvent ional
 Pollutants:	

Iron

Toxic Pollutants:
   0.62
13.7/4.0
2.5
8.5
Chromium
Lead
Nickel
Zinc
0.15
0.15
0.10
0.20
3.8/2.0
3.2/2.1
3.8/2.0
6.4/3.1
0.30
0.32
0.20
0.62
0.57 0.030
0,48 — (3)
0.38 -<3)
1.28 -(3)
0.057
— (3)
— (3)
— (3)
(1) See Table 14-17 for details.

(2) VFR: Ratio of the variability factor of the daily measurements to the
         variability factor of the monthly averages.

(3) No effluent limitation.
                                         373

-------
Basis for New Source Performance Standards

A.   Application of Advanced Level Treatment

     The concentration of  conventional,,  nonconventional,  and  toxic
     pollutants  can  be  reduced  by filtering the clarified effluent
     from BPT in a dual-media filter.

B.   Technology Basis

     For new plants, the recommended wastewater  treatment  technology
     is   lime   neutralization   and   precipitation,  settling,  and
     dual-media filtration (equivalent to Level 2).  All the  existing
     chloride   process  plants  using  rutile/upgraded  ilmenite  ore
     currently  practice  neutralization  and   settling,   but   only
     published  treatability  data and the results of our treatability
     study are available on the preformance of dual-media filters.

C.   Flow Basis

     The raw effluent flow rate is the same  as  that  used  for  BAT,
     namely  100 m3/kkg of TiOz.  It is assumed that the unreacted ore
     and coke are slurried  with  water  and  sent  to  the  treatment
     system.   The  selected  flow .value  is  an  average of the unit
     effluent flow rate of two plants (#559 and #199) practicing  this
     method of solids disposal.

D.   Selection of Pollutants to be Regulated

     The pollutants regulated for BPT are also regulated for  NSPS  in
     addition  to  iron.   The pollutant parameters of concern are pH,
     TSS, iron, and chromium.  Concentration limits are  provided  for
     lead,  nickel,  and zinc in cases where these pollutants become a
     concern.

     NSPS are based on the addition of an iron removal step to BPT/BAT
     treatment.  New plants  can  achieve  significant  reductions  in
     wastewater flow, thus reducing overall treatment costs, even with
     the  inclusion of the iron-removal step.  See also the discussion
     of NSPS for the Sulfate Process in this section.

     T.   Conventional Parameters

          a.   pH

               For NSPS, the BPT limitation is retained.   Control  of
               the final effluent within the range of pH  6.0 to  9.0 is
               equired on the basis of data presented in Appendix B of
               the  proposed  Development  Document   (60)  and the JRB
               Study (52).
                                   374

-------
TSS

Since there is no in-plant performance data available on the
addition of filtration to the treated  wastewater  effluent,
treatability  studies  were  performed on a chloride process
raw wastewater to estimate the effectiveness  of  dual-media
filtration  for the removal of the selected pollutants (61).
Results of these studies  support  the  use  of  filters  to
achieve  a  38  percent  additional  .removal  for TSS.  This
reduction is applied to  the  selected  BPT  maximum  30-day
average   of   64   mg/1.    The   maximum   30-day  average
concentration basis is then given by: (64 mg/1}  (1.00-0.38)
= 40 mg/1.  Likewise, the 24-hour maximum concentrations and
unit   effluent   limitations  are  obtained  from  the  BPT
limitations (Table 14-18) as shown below.

The maximum 30-day average concentration is:

(64 mg/1)(0.62) = 40 mg/1

The 24-hour maximum concentration is:

(230 mg/l)(0.62) = 140 mg/1

The maximum 30-day average effluent  limit is:

(6.4 kg/kkg)(0.62) = 4.0 kg/kkg

The 24-hour maximum effluent  limit is:

(23 kg/kkg)(0.62) = 14 kg/kkg

Nonconventional Pollutants

a.   Iron

     The    Treatability   Study    (61)   which   tests   the
     effectiveness  of   dual-media filtration on the  removal
     of  iron  from the lime treated and  clarified  wastewater
     from   BPT  treatment support a  removal efficiency of  38
     percent.   The   concentration   basis   and   effluent
     limitation  for  NSPS   are  obtained by multiplying the
     selected BAT  (or BPT) limitations  (Table  14-18)  by 0.62
     as  follows:

     The  30-day average  concentration basis  is:

     (2.5 mg/1)(0.62) =  1.6  mg/1

     The  24-hour maximum concentration  basis is:

     (8.5 mg/1)(0.62) -  5.3  mg/1
                      375

-------
          The 30-day average effluent limt is:
          (0.25 kg/kkg)(0.62) =0.16 kg/kkg

          The 24-hour maximum effluent limit is:

          (0.84 kg/kkg) (0.62) = 0.52 kg/kkg

3.    Toxic Pollutants

a.    Chromium

     Results of the Treatability  Study  (61)  indicate  that  an
     additional  60  percent  removal of chromium can be achieved
     using a dual-media filtration  polishing  step.   Therefore,
     the  limitation  is  developed  by  application  of  a  0.60
     reduction factor to the BPT/BAT limitation in Table 14-16 as
     follows:

     (BPT/BAT limitation)  (1.00-0.60) = 0.40 (BPT/BAT limitation)

     The maximum 30-day cencentration basis is:

     (0.30 mg/l)(0.40) =0.12 mg/1

     The daily maximum concentration basis is:

     (0.57 mg/1)(0.40) =0.23 mg/1

     The maximum 30-day average effluent limit is:

     (0.030 kg/kkg) (0.40) = 0.012 kg/kkg


     The maximum daily effluent limit is:

     (0.57 kg/kkg) (0.40)  = 0.23 kg/kkg

b.    Other Metals

     Treatability  studies  have  indicated  that  the  following
     increased removals of lead, nickel, and zinc can be achieved
     by filtration (40,41,61).

                     Additional Removal by
               Filtration Using Settled Effluent
     Lead
     Nickel
     Zinc
80!
     The   additional  levels  of  removal  are  applied  to  the
     corresponding BAT  (or  BPT)  concentration  for  the  above
     metals to get the NSPS concentrations.
                           376

-------
          The maximum 30-day average lead concentration basis is:

          (0.32 mg/1)(0.20)  = 0.064 mg/1

          The 24-hour lead concentration  basis is:

          (0.48 mg/1)(0.20)  = 0.096 mg/1

          The maximum 30-day average nickel concentration basis
          is:

          (0.20 mg/1)(0.86)  =0,17 mg/1

          The maximum 24-hour concentration basis for nickel is:

          (0.38 mg/l)(0.86)  = 0.33 mg/1

          The maximum 30-day average concentration basis for zinc
          is:

          (0.62 mg/1)(0.94)  =0.58 mg/1

          The 24-hour maximum concentration basis for zinc is:

          (1.28 mg/1)(0.94)  =1.20 mg/1

          The  conventional,  nonconventional,  and  toxic   pollutant
          limitations for NSPS are given in Table 14-20.

Basis for Pretreatment Standards

Existing Sources

Since  there  are  no indirect dischargers in this subcategory, we are
not promulgating PSES but are instead excluding this subcategory  from
categorical  PSES  under  the  provisions  of  paragraph  8(b)  of the
Settlement Agreement.

New Sources

For Pretreatment Standards for  New  Sources  (PSNS),  the  Agency  is
basing  limitations  on NSPS.  The pollutants to be regulated are iron
and chromium   ('See  Table  14-20).   These  pollutants  are  regulated
because  NSPS  provides  better  removal  or iron and chromium than is
achieved by a well-operated POTW with  secondary  treatment  installed
and  therefore  iron  and  chromium would pass through the POTW in the
absence of pretreatment.

Titanium Dioxide - Sulfate Process Industry Profile
                                 377

-------
                     TABLE  14-20.  EFFLUENT LIMITATIONS
                    TITANIUM DIOXIDE - CHLORIDE PROCESS
                     New Source Performance Standards
                   Waste Water Flow:  100 m3/kkg  of TiO2

Pollutant Treatability

Conventional and

VFRC1)


Concentration Effluent Limit
Basis (mg/1) (kg/kkg of TiO-)
Max
30-day
Avg

Max
24-hr 30-day
Max Avg

24-hr
Max

Nonconventional Pollutants:
Total Suspended
Solids
iron'2'
Toxic Pollutants:
(2)
Chromium

Lead
Nickel

Zinc

13
0.38

0.060(4)
{&}
0.030 w
0,086(4)
(&\
0.13t4)

n.o/3.0
13.7/4.0

3.8/2.0

3.2/2.1
3.8/2.0

6.4/3.1

40
1.6

0.12

0.064
0.17

0.58

140 4.0
5.3 0,16

0.23 0,012
CVi
0.096 — ^
0.33 -<3>
I 1\
11 \ -5/
.2 —

14
0.52

0.023
( 3)
—
-J3>
/ri\

(1)   VFR:   Ratio of variability factor of the daily measurements to the
           variability factor of the monthly averages.

(2)   Also applicable for PSNS limitations.

(3)   No effluent limitations.

(4)   Long term average from Table 14-18 multiplied by the appropriate percent
     removal factor indicated in Section  14.7.5.
                                     378

-------
             TABLE 14- 21.   SUBCATEGORY PROFILE DATA SUMMARY
SUBCATEGORY
TTTANIUM DIOXIDE
SULFATE PROCESS
Total subcategory capacity rate
Total subcateogry production rate
Number of plants in this subcategory
308 Data on file for
     With total capacity of
     With total production of
     Representing  -capacity
     Representing production
     Plant production range:
             Minimum
             Maximum
     Average production
     Median production
     Average capacity utilization
     Plant age range:
             Minimum
             Maximum
     Waste water flow range:
             Minimum
             Maximum
     Volume per unit product
             Minimum
             Maximum
                         401,000 kkg/year
                         259,000 kkg/year
                               4
                               5
                         320,000 kkg/year
                         246,000 kkg/year
                              80 percent
                              95 percent

                          31,000 kkg/year
                          74,500 kkg/year
                          49,000 kkg/year
                          43,000 kkg/year
                              76 percent

                              23 years
                              54 years

                          35,000 cubic meters/day
                         125,000 cubic meters/day

                             300 cubic meters/kkg
                             780 cubic meters/kkg
Sources of data are Stanford Research Institute, Directory of Chemical
Producers, U.S.A., 1977, U.S. Department of Commerce, Current Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry," June,  1978 and "Economic Analysis: of  Proposed Revised
Effluent Guidelines and Standards for the Inorganic Chemicals Industry,"
March, 1980.                                                         Jr
                                     379

-------
                                  TABLE 14-22.   ANALYSIS OP ILMENITE ORES1

Chemical
Constituent
Ti02
PeO
Fe203
Si02
A1203
P2°5
w
00
Zr02
MgO
MnO
CaO
V2°5
Cr203
Virginia
Piney
River Roseland
44.3
35,9
13.8
2.0
1.21
1.01
0.55
0.07
0.52
0.15
0.16
0.27
51.4
37.9
1.6
4.6
0.55
0.17
NA
2.35
0.70
0.59
0.07
NA
UNITED STATES
New York Florida
44.4
36.7
4.4
3.2
0.19
0.07
0.006
0.80
0.35
1.0
0.24
0.001
64.1
4.7
25.6
0.3
1.5
0.21
NA
0,35
1.35
0.13
0.13
0.1
California
48.2
39.1
10.4
1.4
0.2
NA
0.05
0.6
0.1
0.1
0.05
0.03
Ivry
42.5
39.1
20.7
0.88
1.05
NA
NA
2.0
0.04
0.1
0.36
0.15
CANADA
Bourget
22.4
36.9
31.2
1.0
6.01
0.93
NA
1.50
NA
0.55
NA
NA
Allard
37.3
26.3
30.0
NA
NA
0.004
NA
NA
0.10
NA
0.39
NA
 Constituents expressed as weight percent.
NA:  Not Available

-------
General Description

The industrial profile for the Sulfate Process Segment of the Titanium
Dioxide Subcategory is presented in Table  14-21  and  the  status  of
regulations prior to promulgation of this regulation is shown in Table
14-2

General Process Description and Raw Materials

A.   Sulfate Process - General Description

Among the various titanium ores, ilmenite is  availabe  in  abundance.
Ilmenite  is a low-grade titanium ore with a Ti02 content varying from
45 to  60  percent.   JJLmenite  ore  and  slag  from  iron  production
generally  comprise  the  raw  materials  used  for the preparation of
titanium dioxide by the sulfate process.  Large amounts of  water  and
sulfuric acid are used in this process, and the majority of the plants
are  co-located  with  sulfuric  acid  plants.   Table 14-22 gives the
analysis of various ilmenite ores.  The preparation  of  TiO2  by  the
sulfate process utilizes three important steps:

1.   Digestion:

     FeO.TiOz + 2HZS04 = FeS04 + TiO.SO4 + 2H20

2.   Precipitation:

     TiO.S04 + 2H20 = TiOz.H20 + H2S04

3.   Calcination:

          TiOz.H20 = TiOz + HZ0

The ore is dried, ground, and then reacted with  sulfuric  acid.   The
reaction  takes  place  at  160°  and  the reacted mixture Consists of
titanyl, ferrous, and ferric sulfates.  The product  is  dissolved  in
water.   The  total iron in the reacted product is kept in the ferrous
state by the addition of scrap iron.  After the reduction, the product
is dissolved in water and  clarified  with  the  aid  of  flocculating
agents  to remove insoluble impurities such as silicon, zirconium, and
unreacted ore.  The iron is removed from the clear solution by cooling
the solution to 10°  C  when  hydrated  ferrous  sulfate  (FeS04.7HzO)
crystallizes.   The ferrous sulfate crystals  (commercial copperas) are
mechanically  separated   from   the   solution   by   filtration   or
centrifugation.   The concentrated titanyl sulfate solution is diluted
with water and heated to form titanium dioxide hydrate.  The  solution
is  filtered, and the filtrate, known as strong acid, is separated and
either  discharged  or  recycled.   The  Ti02.HzO  filter  residue  is
slurried  with  water  and  conditioning  agents  are added to control
particle size, color,  dispersibility,  and  photochemical  stability.
The conditioning agents include potassium, zinc, antimony, and calcium
compounds,  and  phosphate  salts.   The  solution is filtered and the
filtrate is known as weak acid.  The product  is then rinsed to  remove
                                381

-------
OJ
00
to


STRONG-ACID *
HATER — »
DIGESTER

HATER , 	 n.
CLARIFIER
*
	 J EVAPORATOR
4

STEAM _*
PRECIPITATION
t
— EMISSIONS 	 »•
HATBR ^* AND VEHTURI SCRUBBERS »•»-—•-


w EMISSIONS — ,_^ ' " "- .^
HATER CONDENSERS — EFFLUENT -^ 	

COOLER P MI"0iUBB """*; CONDENSERS U WPLUBHT ~» 	

WEAK -AC ID 1 1
RBCYCLB 	 ^t 	 * 	 1_ J —
HATER— to
FIRST MOORE
rn.TBn


HATER—*
SECOND MOORE
FILTER
STEMl ^— ' I '— ' •

HATBR 	 fr>
STEAM 	 ft»
DIl



CALCIHER
i
HBT HILL
rBHlBajuna 	 » COOLING SPRAYS AND ELECfc 	 fc HH.K ncm.
HATER 	 fc FROSTATIC PRECIPITATORS |
EMISSIONS
*
HATER 	 »J MIST ELIMINATORS j~ EFFLUENT -•*• 	

*
1

JET MILLS

WATER ^^^-toi 1
1
TITANIUM
JXIDB PIGMENT
t
PACKAGING
MATER 	 » JET MILL SCRUBBERS 1— . ». EFFLUENT 	



                                      TO SALES
            Figure 14-7.  General process flow diagram for production of titanium dioxide by
                          sulfate process.

-------
TABLE 14-23.  WATER USAGE IN TITANIUM DIOXIDE - SUI*FATE PROCESS SUBCATEGOKY
Uses
Water Usage per Unit of Production
                 of Ti02)

Noncontact

cooling
Direct process contact
indirect pr
ocess contact
Plant #555
47.8
390
6
Plant #694
408
588
1.6
Plant #696
149
297
4
(pumps, seals, leaks,
spills, etc.)

Maintenance, equipment
cleaning and work area
washdown

Air pollution control

Noncontact ancillary
uses  (boilers, utilities,
etc.)
                         1.8
     258

      36
78

33
81

NA
NA:  Not Available
                                     383

-------
iron  and  unreacted  acid.   Residual  acids  in  the precipitate are
removed along  with  the  water  of  hydration  by  calcination.   The
resulting  TiOz  pigment  is  sent to finishing operations, which vary
according to the end product requirement  and  application.   The  wet
finishing operations may include some, or all, of the following steps:
pulping,   milling,  surface  treatment  with  proprietary  agents  in
solution,  washing,  and  drying.   The  alternative   dry   finishing
operations   may  include  one  or  more  milling  steps  followed  by
packaging.  A simplified block diagram of the sulfate process is shown
in Figure 14-7.

Water Use and Waste Source Characteristics

Water Use


Water is used in the preparation of titanium dioxide  by  the  sulfate
process  for noncontact cooling, air emission control, and for process
reactions.  In the process, water is used to leach the soluble sulfate
salts from the reaction mass and to convert  the  titanyl  sulfate  to
titanium  dioxide  hydrate.   Water  is also used to wash the titanium
dioxide hydrate precipitate free from residual acid and  iron.   Water
is used for air emission control during the drying of ore, on digester
units,  and  for the cleaning of the kiln gases before they are vented
to the atmosphere.  In the digester unit,  water  seals  are  used  to
maintain  a  vacuum on the digester units.  Large amounts of water are
also used in the finishing operations.  Table 14-23 is  a  summary  of
water  usage  in  the  titanium  dioxide subcategory using the sulfate
process.

Waste Sources

A.   Digester Sludge

          After the  digestion  of  the  ore  in  suIfuric  acid,  the
          resulting  sulfates are dissolved in water and the insoluble
          impurities are removed in  a  clarifier  or  filter.   These
          include  silica, alumina, sulfuric acid,.and unreacted iron.
          The quality of this waste varies and depends on the type and
          quality of ore used.  Data on the  quantity  of  this  waste
          indicates that approximately 210 kg/kkg is produced.

B.   Copperas

          The recovered ferrous sulfate is marketed or disposed of  as
          a  solid  waste.   The amount of copperas generated is about
          950 kg/kkg of Ti02.  The copperas generally  contains  small
          amounts of adsorbed sulfuric acid.

C.   Strong Acid Waste

          When water is added to the titanyl  sulfate  solution  after
          the  removal  of  copperas, sulfuric acid and the hydrate of
                                384

-------
          titanium dioxide are formed.  The acid contained in  solution
          is removed by filtration and  the filtrate is known as strong
          acid  solution.    The  concentration of sulfuric acid varies
          from 15 to 30 percent as H2S04.    In  addition  to  sulfuric
          acid,   the  waste  stream contains ferrous sulfate,  titania,
          antimony, and other heavy metal  salts.  A part of  the  acid
          is  returned  to  the  process  and  the  rest  sent  to the
          treatment facility.
D.    Weak Acid Waste Stream
          The  waste  generated  from  washing  the  titanium  dioxide
          hydrate   precipitate   is   known   as   weak   acid.   The
          concentration of sulfuric acid in this waste varies from two
          to four percent as H2S04 and  contains  various  impurities,
          including  iron  sulfate, titania, antimony, and other heavy
          metal  salts.   It  also  includes,  in  some   cases,   the
          conditioning  agents  added  to  the  precipitate  prior  to
          washing, to control and improve the  quality  of  the  final
          product.   The  weak  acid may also include the kiln exhaust
          scrubber waste.
E.   Scrubber Wastes
          Scrubber wastewater results from  the  scrubbing  of  vapors
          emitted  during the drying of the ore, during digestion, and
          during kiln drying.   The  amount  of  wastewater  generated
          depends  on  the  amount  of water used and type of emission
          controls practiced.  The scrubber  water  contains  titanium
          dioxide  particulates, acid mist, sulfur trioxide and sulfur
          dioxide.   Of  all  the   waste   produced   from   titanium
          dioxide-sulfate   process   manufacture   subcategory,   the
          scrubber wastewater constitutes the major portion.
F.   Wet Milling Waste

          These wastes are  generated  during  wet  finishing  of  the
          titanium  dioxide  pigment.   Wet milling is used to produce
          pigment particles of the desired size and surface  character
          and  requires  steam  and  water  for repulping the 'pigment.
          Caustic soda is also used to  remove  any  residual  acidity
          from  the  titanium  dioxide  pigment  during  the finishing
          operation.  The wastewater  from  wet  finishing  operation,
          therefore,  contains  titania,  sodium  sulfate,  and  other
          agents added to improve or achieve desired properties in the
          final product.

Description of Plants

Screening

Plant #555 was visited and its waste streams sampled in the  screening
phase  by  an EPA Region II team.  The pigment manufacturing operation
                                 385

-------
TABLE 14-24.  HAW WASTE CHARACTERISTICS  (INDUSTRY DATA) ^FOR PLANT 1555
              (PRODUCTION OF TiO2 BY SULFATE PROCESS)
Waste Source
 Unit
 Flow
(m3/kkg
 of TiC>2)
pH
Pollutant Waste Loads, (kg/kkg of Ti02)
  Acidity    NH3
 (asH2S04) (asN)
                                                              Fe   TSS   TDS
Digestion           115        3.0
Clarification         3.58     2.5
Evaporation         113        4.0
Cooling              20        6.1
Strong Acid fron      8.49   < 0.5
first Moore Filtration
Weak Acid from       12.2      2.0
first Moore Filtration
Weak Acid from       10.4      1.7
second Moore Filtration
Weak Acid from       12.0      2.0
first stage
Calcination
Weak Acid from       40.0      2.2
second stage
Calcination
                       20.8
                       26.7
                       18.7
                        2.49
                        2.360

                       88.3

                      148

                       20.8

                       19.2
             NA    0.042    9.3  35.7
             NA    8.42   175    40.8
             NA    1.14     3.2  20.2
             NA    0.099    0.46  3.09
             NA  139        0.959 2.815
             NA    3.8
             NA
                           0.64
NA:  Not Available
* Value in ffl units
(1) - Response to 308 Questionnaire, 1976
                                    0.23 98.8
             NA    0.29     0.13 151
             NA    0.22     2.0   7.50
4.92 33.1
Calcination Mist
Eliminators
Wet Milling Washing
and Drying
Jet-Mill Condenser
Jet-Mill Scrubbers
Boiler and Water
Plants
38.

11.

27.
18.
16.

7

1

0
0
6

3.

8.

6.
7.
9.

0

0

5
4
0

7.50

NA

NA
NA
NA

NA

8.6

NA
NA
NA

0.

0.

0.
0.
0.

02

01

01
13
66

0.

2.

1.
1.
5.

21

13

1
7
25

27.9

11.0

2.7
3.58
8.92

                                     386

-------
utilizes a titania slag for the production  of  TiOz  by  the  sulfate
process.   After  digestion  of the slag in sulfuric acid the residual
gangue material is filtered out and the clear liquor is concentrted by
evaporation.  The crude pigment is formed by hydrolysis with water and
steam and processed to form  both  anatase  and  rutile  type  pigment
products.    Table  14-24  presents  raw  waste  flows  and  pollutant
characteristics for Plant #555.

Wastewater samples were collected at five points and analyzed for  the
conventional,  nonconventional,  and toxic pollutants.  These sampling
points were  designated  as  1)  the  contact  air  scrubbers  on  the
digesters,  2)  the  black  end  flume  containing  wastes  from major
cuttings, filter sludges, acid filtrates, and evaporator and condenser
waters, 3) the white end flume carrying finishing  process  filtrates,
noncontact  cooling  water,  and sanitary wastes, 4} northside jet air
scrubbers, and 5) southside jet air scrubbers.

At present, all of the  process  waste  streams  are  collected  in  a
settling  basin  which  is  open  to  tidal  fluctuations that provide
diurnal flushing of the effluent into the receiving waters.

Verification

Plant #559 was surveyed in both the screening phase  and  verification
phase  of the study.  At this plant the strong acid is sent to a lined
holding pond for equalization.  Effluent from the pond is  neutralized
with  ground  calcium  carbonate  limestone  in  a reactor.  A minimum
amount  is added to raise the pH to a level such  that calcium  sulfate,
but not, ferrous hydroxide, is precipitated.  The COZ formed during the
reaction  is  vented  to the atmosphere and  the  calcium sulfate slurry
goes to a clarifier.  The underflow from the clarifier is filtered  to
produce pure gypsum crystals at a concentration -of 70 to 80 percent.

The  weak acid  is sent to a settling pond, where it is combined with a
small quantity of other wastes.  The effluent from the weak acid  pond
is  mixed  with the calcium sulfate clarifier overflow and neutralized
with ground  calcium carbonate  in a three-stage   reactor.   Pebble  and
slaked   lime  are  also  added  to  raise  the pH and precipitate more
calcium sulfate.  Air is also  introduced to  convert the  ferrous   iron
to  ferric   iron.   The  effluent  from  the  reactor  goes to another
clarifier, and the clarifier underflow  is filtered to concentrate  the
solids  to 70 percent.  The overflow from the second clarifier is mixed
with  the  other  process  wastewaters.   These  include the scrubber,
finishing, and cooling wastewaters.  The combined water  is neutralized
with slaked  lime before  it is sent  to  a  final settling  pond,  the
effluent from which is discharged.  Figure 14-8  gives the flow diagram
of  the  treatment  process  and shows  the sampling locations for  both
screening and verification.  Table 14-25 gives the flow data  for  the
waste   streams   and   conventional    and   nonconventional  pollutant
emissions.
                                  387

-------


WEAK ACIt - fc
UAJTI ITftlJM

OTMCR PRODUCT
WASH UATU
VCAR ACID POND





n

HI VI* ^~1
SUPPLY VATIR " '" ™'^J "
H
S
\

OJ
03


UASTC STMAH



UPPLV UATfh

UPPLY UATER





PQHO



n
8
10


n
























UACTORS




































FILTIR
SOLIDS
TO



i
















REACTORS







riu»
I


— ~ ~~* *"" FINAL ^-^
h
SOLI OS TO
STORA6E/ f






























,




	 POHO ~m*
, , "



*

j OTHER PRODUCT
^ WASTI WATCR





TIO, (SULfATC PROCESS)
1 SCRUIICR
WASTE WATCR








STflHACI/LANOF.Ul












LECENO
mm SAMPLING POINTS
Figure 14-8.   General flow diagram at plant #559 showing the sanpling points.
               (Titanium dioxide - sulfate process.)

-------
      TABLE 14-25. FLOWS AND POLLUTANT CONCENTRATIONS FOR THE WASTE STREAMS
                 SAMPLED FOR PLANT #559 PRODUCING TITANIUM DIOXIDE
          Sampled
Stream    Stream
   (4)
No
Description
Unit
Flow
(m3/kkg)
TSS               Iron
Load              Load
(kg/kkg of TiO2)  (kg/kkg of TiO2)
          Weak Acid
          Pond Overflow
                                    1.23
                                   1.23
          Strong Acid
          Pond Overflow
                      6.1
              205.9
                  106,3
          Scrubber and
          Contact Cooling
           Water
                    361.4<1>(2>    113.5
                                 51.68
          Final Treatment
           Effluent
                    436(1)(2)(3)    10.0
                                   1.92
(1) The flow is contributed by the sulfate process stream.

(2) The pollutant load was calculated by multiplying the flow contributed by the
    sulfate process stream times the concentration of pollutant.  Pollutant Load
    = (total stream flotf) x (fraction contributed by sulfate process waste) x
    stream pollutant concentration.

(3) While calculating the unit flow the contributions to the treatment process
    from precipitation, the water in the treatment chemicals, losses from
    evaporation and from solids leaving the process have not been considered.

(4) See Figure 14-8 for sampling point location.
                                       389

-------
Other Plant Descriptions

At Plant 1694, the clarification sludge which contains  the  unreacted
ore  is sent to waste disposal.  The weak acid effluent from the plant
is neutralized with slaked lime  and  the  grit  is  settled  out  for
landfill  disposal.   After the separation of grit, the aqueous stream
is discharged to a municipal  treatment  system.   The  other  wastes,
together with runoff from the plant site, are collected, and sent to a
lagoon for solids removal, and the overflow discharges to a river.

At  Plant  #696,  the raw wastes were sent to thickeners to remove the
suspended solids and the overflow was discharged.   Depending  on  the
titanium  content,  the  underflow  from  the  thickeners  was  either
recycled or disposed of in a landfill.  This  plant  has  discontinued
operations.

At  Plant #605, the process raw waste streams are combined and sent to
a reactor for neutralization with a  water  slurry  of  finely  ground
calcium  carbonate.   The  effluent  from the reactor is hydrocycloned
into three fractions.  The first fraction, which is the coarse  gypsum
slurry,  is  separated from the reactor effluent at a concentration of
85 to 90 percent, and placed in  a  self-draining  dewatering  system.
The  "dry"  solids  are  finally  trucked  to  a landfill.  The second
fraction separated in the hydrocyclone is a fine gypsum  slurry  which
is  recycled  to  the neutralization reactor.  The residual gel slurry
forms the third fraction, and this is sent to a  thickener  after  C02
degassing.  A flocculating agent is added to the flow to the thickener
to  promote  solids separation and thickening.  The underflow from the
thickener is centrifuged and the solids landfilled.  The filtrate from
the centrifuge  is  recycled  to  the  thickener,  and  the  thickener
overflow is discharged.

The  volume  and  characteristics of wastewater streams from different
sulfate process titanium dioxide plants do not differ  greatly.   Some
variations,  however,  are  noted  as  a  result of differences in ore
qualities, in location and in process details.  The  majority  of  the
dissolved  pollutants  in  wastewater  from  this  segment of the Ti02
industry consist of acidity and iron.  Segregation of  the  wastewater
is   important  for  control  and  treatment  practices  and  aids  in
developing economically feasible treatment systems.   Generally,  weak
and strong acid streams are segregated from each other as well as from
the  less  contaminated  wastewaters  which  include  contact cooling,
scrubbing, and some finishing operation wastes.  The  unit  flows  for
the  segregated raw waste streams at different facilities are shown in
Table 14-26.

The average total effluent flow rate is 475 mVkkg (Table  14-16)  for
Plants  #555, #694, and #559.  Complete flow data  is not available for
Plants #696 and #605.
                                 390

-------
TABLE 14-26.  PROCESS WASTE WATER FLOW AT PLANTS # 555, #694 and #559
                 TITANIUM DIOXIDE (SULFATE PROCESS)

Plant
#555
#694
#559
Average

Strong
8.5
16
6.1
10
Flow in (irr/kkg
acid Weak acid
78.2
67
69
72
of Ti02)
Scrubber and
contact cooling
water
362
457
361
393

Total Effluent
449
540
436
475
                                 391

-------
Toxic Pollutant Concentrations

Section 5 of this report describes the scope and  methodology  of  the
sampling  program.   In  the  Sulfate  Process segment of the Titanium
Dioxide Subcategory, 18 different sampling points  were  selected  for
studying  the  toxic  pollutant characteristics of the water supplies,
the raw process wastewaters, and  the  plant  effluent  at  two  major
manufacturing   facilities.    For   the  inorganic  constituents  575
analytical data points were generated and  an  additional  1,824  data
points were obtained for the organic toxic pollutants excluding blanks
and duplicates for quality control.

The  only  organic  toxic pollutant found during the screening program
was phenol which was observed at only one of the two  plants  sampled.
The  maximum raw waste concentration of phenol was 0.020 mg/1, however
the raw water source for the plant contained as much as  0.0070  mg/1.
This  is  well  below  the  treatability  level for phenol, therefore,
phenol is not considered a significant or process related pollutant.

Daily raw waste loads were calculated from the flow rates measured  or
estimated   at  the  time  of  sampling  and  the  measured  pollutant
concentrations.  That is,

     daily loading  (as kg of pollutant per day) - (C)(Q)
                                                  1000

     where the concentration (C) of  the  pollutant  is  expressed  in
     units  of mg/1 (Note: 1 kg/m3 = 1000 mg/1), and the flow rate  (Q)
     expressed in units of mVday  ( a cubic meter, m3, is equal to 264
     U.S.  gallons).

     Similarly, the unit loadings were calculated  from  the  reported
     Ti02 production rate  (P), the waste stream flow rate  (Q), and the
     measured pollutant concentration (C).


     Unit loading (as kg of pollutant = (C)(Q)
     per day kkg of TiOa)               1000(P>

     where C and Q are expressed in the same  units  described  above,
     and  the  production  (P) is expressed in units of kkg/day {kkg is
     1000 kg, a metric ton, which  is equal to 2205 Ib).

The maximum concentration of toxic pollutants found in the  raw  waste
at  concentrations  above  the treatability level in the screening and
verification program were:
                                   392

-------
                                       TABLE 14-27.   SIM-1ARY OF RAW HASTE LOADINGS fOCND IN SCREENING AND VERIFICATION SAMPLING
W
U>
U)
SUBCftTEGORY
Pollutant
Toxic
Antimony
Arsenic
Cadmium
Chromium
Copper
Lead
Nickel
Selenium
Thallium
Zinc
Conventional
TSS
Iron, Pe
TITANHH DIOXIDE - SULFATE PROCESS
Loading Range,
(kg/day)
Miniimm Maximum
5.0 28
1.9 4.0
.068 7.2
140 530
8.2 19
3.0 65
3.7 23
7.0 9.5
0.47 1.2
1.8 85
and Nanconventional


Minimum
0.032
0.012
0.00044
1.1
0.065
0.024
0.029
0.0020
0.0030
0.014


Unit Loading,
(kg/kkg)
Average
0.11
0.19
0.019
2.0
0.085
0,18
0.080
0.031
0.0055
0.34
320
600
Maximum
0.22
0.032
0.057
3.4
0.12
0.42
0.15
0.060
0.0080
0.55


No. of
Plants W
3
3
3
3
3
3
3
2
2
3
1
1
                             (1)  - Data are taken only from those plants where pollutants were found above detection limits, or, in the
                                   case of  TSS and Iron,  where data are available.

-------
                                           TABLE 14-28.  TOXIC POLLUTANTS; AVERAGE RAW, WASTE LOADS AND OONGEKTRATIONS
w
•£>
SUBCATEGORY
TTTANIIM DIOXIDE - SULEKTE PROCESS
Screening


Antimony
Arsenic
Cadmium
Chrcmiun
Copper
Lead
Nickel
Selenium
Thallium
Zinc

(m3/l)
0.77
0.11
0.29
3.8
0.20
0.075
0.091
NA
NA
0.088
Plant I5S5
(kg/kkq)
0.22
0.032
0.057
1.1
0.065
0.024
0.029
< 0.06
NA
0.014
Plant
(mg/1)
0.16
0.029
0.0020
7.0
0.25
0.20
0.31
NA
0.020
1.1
1559
(kg/kkg)
0.080
0.014
0.0009
3.4

0.10
0.15
NA
0.0080
0.55
verification
Plant
(irci/1)
0.074
0.028
0.0010
3.1

0.96
0.14
0.0050
0.0070
1.04
1559
(kg/kkg)
0.032
0.012
0.00041
1.4
0.070
0.42
0.061
0.0020
0.0030
0.45
Average
(nw/1)
0.34
0.06
0.10
4.63
0.20
0.41
0.18
0.005
0.014
0.74
                            NA = Not Available

-------
                Maximum Concentration Observed (ng/1)

                    Screening                Verification
Pollutant      (Plants #555 & #559)                (Plant #559)

Cadmium                      340                       12
Chromium                 124,000                    31,000
Copper                     1,500                     1,000
Lead                       3,700                     5,200
Nickel                     6,400                     1,300
Zinc                       3,800                     1,700
A summary of daily and unit (per unit of production) raw  waste  loads
for  all plants sampled can be found in Table 14-27.  Individual plant
raw waste loads and concentrations found  in  sampling  are  given  in
Table 14-28.

Based  on.the total annual production of this industry and the average
waste load generated per  unit  product,  the  estimated  total  toxic
pollutant raw waste loads generated each year for this subcategory are
as follows:

                    Pollutant      Total Annual Raw
                                   Waste Load (Kg/year)

               Cadmium                    5,000
               Chromium                 510,000
               Copper                    22,000
               Lead                      47,000
               Nickel                    21,000
               Zinc *                     88,000
               Antimony                  29,000
               Arsenic                   49,000
               Selenium                   8,000
               Thallium                   1,400

Pollution Abatement Options

Toxic Pollutants of Concern

The  toxic pollutants found above treatability levels in this industry
were evaluated on the basis of the maximum concentration  observed  in
the  process  raw  wastewaters.   These values are shown above.  Using
cadmium as an example of  a  borderline  case,  its  maximum  observed
concentration  of  0.34  mg/1   (340  ^g/1)   is  considered significant
because removal efficiencies ranging  from   70  to  97  percent  could
possibly  be achieved on the basis of the lower limits of treatability
shown in Table 8-11 for lime/settling, lime/filter, and sulfide/filter
technologies.  The BAT ultimately selected as a basis for  regulations
may  not be as effective as the most advanced technology considered at
this stage of the evaluation of alternatives.
                                 395

-------
The  sampling  data  from  this  industry  indicate  that  the   toxic
pollutants  of  concern  are  chromium,  zinc,  nickel,  lead, copper,
antimony, arsenic, and cadmium in  decreasing  order  of  the  maximum
amounts  found.  Selenium and thallium were detected at levels too low
to be treated effectively.  The relative pollutant concentrations  and
loadings in the raw waste largely reflect the amounts of impurities in
the  ilmenite ore or titania slag being processed.  The major impurity
found in the various  grades  of  raw  material  is  ferrous  iron  as
indicated  in  Table  14-22.  The toxic metal impurities would also be
expected to occur in  a  wide  range  of  concentrations  in  the  raw
materials.

The  advanced  treatment  technology options evaluated for the sulfate
process segment of the industry were selected  for  their  ability  to
remove  toxic  metals  of  concern  with  greater  efficiency than the
prevailing (BPT) practice which also removes TSS,  iron,  and  sulfate
from the wastewaters.


Process Modifications and Technology Transfer Options

Specific  process modification recommendations are not made.  However,
several areas for further investigation suggest themselves.  They are:

A.   One of the water borne wastes, the strong sulfuric acid  produced
     from  the Ti02 sulfate process, has a sulfuric acid concentration
     that varies from 15 to 30 percent as H2S04.   Currently,  only  a
     small  portion  of  it  is  recycled.  Research is needed to find
     cost-effective ways to concentrate the acid to 90 percent and  to
     eliminate  the  impurities  (especially  iron)  so that it can be
     reused in the digester.  This will eliminate much of  the  alkali
     requirements  for  neutralization  and  relieve disposal problems
     associated with solid waste gypsum.

B.   Economical methods need to be developed for the recovery of  iron
     oxide,  aluminum,  and vanadium from the waste to the extent that
     markets are available for these materials.

C.   If markets could be developed for the  sale  of  ferrous  sulfate
     (copperas),  solid  waste  disposal  problems  wouId  be reduced.
     Currently, a portion is sold and the rest disposed of as a  solid
     waste.

Best Management Practices

Storm water runoff from the plant site should be collected and sent to
the treatment facility for the removal of suspended solids.

Prevailing Control and Treatment Practices

The  treatment  practices  of the plants producing TiOz by the sulfate
process are given, above.
                                   396

-------
Advanced Treatment Technologies

Although sulfide precipitation, the xanthate process, and ion-exchange
might be applied  to  the  clarified  solution  obtained  by  alkaline
precipitation,  oxidation  and  settling  the  costs incurred are high
because of the large quantity of water (more than 400 m3/kkg of  Ti02)
that  must  treated.   The sulfate process is one of two subcategories
(the other being  the  Soda  Ash  Solvay  Process)  in  the  Inorganic
Chemicals  Industry  studied in this report that generates the largest
quantities of waste effluent.

Selection of Appropriate Technology and Equipment

Technologies for Different Treatment Levels

A.   Level 1 (BPT/BAT)

In the Level 1  treatment,  the  strong  and  weak  acid  streams  are
independently  equalized  in lined lagoons and neutralized in a reator
with ground limestone to a pH range of 5 to 5.5, precipitating calcium
sulfate and forming carbon dioxide, which remains in  solution,  along
with  large  quantities  of  ferrous  iron.  Solids are generated in a
thickener at the unit  rate  of  3  kkg  per  kkg  of  products.   The
thickener  overflow  joins the large but relatively unpolluted flow of
"other wastes", and the combined wastewaters are treated to pH 6 to  9
and  passed  through  a  polishing lagoon prior to discharge.  Level 1
treatment allows recovery of gypsum solids  without  contamination  by
iron  and  toxic metals pollutants.  The flow schematic for Level 1 is
presented in Figure 14-9a.

B.   Level 2  (NSPS)

In the Level 2 treatment, the blended strong and weak acid streams are
neutralized  with  calcium  carbonate.   Most  of  the   toxic   metal
pollutants  are  precipitated and separated along with gypsum  in first
stage thickeners.  Aeration then oxidizes any ferrous iron present and
removes C02 before mixing with miscellaneous  plant  waste  containing
minor  amounts of toxic metal pollutants.  The combined stream is then
given lime treatment to pH 9 and settled in polishing  lagoons  before
discharge.   This  three-step  system  is  patterned after an existing
system which separates the acid streams from miscellaneous  wastes  in
order to make possible the recovery of pure and impure gypsum from the
relatively  consistent  acid  streams.  Alkaline precipitation of more
toxic metals and significant removal of arsenic occur during the  last
two  stages of lime neutralization, and settling of precipitated toxic
pollutants occurs in the final polishing lagoons.  Because waste  flow
rates  are  unusually  high  in  the sulfate process, long-term lagoon
settling is more  cost  effective  than  dual-media  filtration.   The
mechanical   aeration   step  used  for  oxidizing  ferrous  iron  may
contribute an important mechanism  for  the  simultaneous  removal  of
additional   heavy   metals  present,  very  similar  to  the  ferrite
coprecipitation  method  described   in   the   Treatment   Technology
                                397

-------
lO
CO
                        MCTE HUTCH
                             WOTS
                                                                                     &
SCUDS OZSPOSALOf

SHE OR
                                                                                                                         BJUJBfT
                                      *  InclulM flat BocUterlrq, fti ncrUtarlnq
                                         •nd
                        Figure 14-9a.  Level 1 waste water treatment for  titanium dioxide - sulfate process.

-------
                                                                                                             EFFLUENT
WASTE mi'
                                                               CM SHE
                    Includes flow iiioniloi-ing, pH monitoring and sampler
                        Figure 14-9b.  Level 2  waste  water  treatment for  titanium dioxide
                                         sulfate  process.

-------
Assessment section.  The flow diagram of the treatment system is shown
in Figure 14-9b.

Although  the  Model  Plant  does  not  include  equipment  for gypsum
recovery, it is based on separation of waste streams, making  pure  or
impure    gypsum   recovery   possible   by   intercepting   thickener
underflow(s).

C.   Level 3

Level 3 for the sulfate process employs the described  NSPS  treatment
for  stong acid, weak acid, and 55 percent of the "other wastes".  The
remaining other wastes receive soda ash  treatment  and  settling,  to
permit recycling a nonscaling effluent for scrubbers and miscellaneous
uses.   Toxic  metal  pollutants  in  the separated recycle stream are
settled as hydroxides or carbonates, plus any calcium arsenate formed,
and periodically removed to secure landfill.  The flow diagram of this
treatment is shown in Figure 14-10.

Euipment for Different Treatment Levels

A.   Equipment Functions

     Treatment of wastewater from the  sulfate  process  involves  the
     mechanized handling of large quantities of chemicals and reaction
     products, primarily gypsum.

     The  BPT  model  includes rail car deliveries of ground limestone
     and lime,  bucket  elevators,  storage  bins,  multiple  feeders,
     mechanical  feeders  and  one stage thickening to recover gypsum.
     Calcium saturated  thickener  overflow  and  .miscellaneous  other
     wastes  are  subjected to pH adjustment and one-day settling in a
     polishing pond prior to discharge.

     In Level 2 or NSPS, wastewater  treatment  requires  the  use  of
     mechanical  aerators,  lime addition, and second stage thickening
     equipment in addition to equipment specified for BPT treatment to
     remove additional toxic metals and iron.

     In Level 3, to reduce the raw waste load of toxic metals, only 55
     percent of the NSPS "other waste" flow  joins  the  treated  acid
     waste stream, for Level 2 treatment as described above.  However,
     the  remaining  45  percent  of  "other  waste" is given separate
     treatment with soda ash and settled in a lagoon, for  recycle  to
     miscellaneous   scrubber   and   noncontact   cooling   purposes.
     Treatment  of  the  strong  and  weak  acid  streams,   including
     oxidation  and  settling  of ferrous iron, remains the same as in
     the NSPS model.

B.   Chemicals and Handling

     First stage neutralization employs ground limestone,  while  lime
     is   used   in  cases  where  second  stage  and  final  alkaline
                              4CO

-------
                                                __.	 BECTfCUC
                                                    (MISC. USES)
                        UUESTONE
                         C.OO.
                                                                                               OTHER
                                                                                               WABTKS
                                 m
     RAPID UIX AND 8ETTUNO

MECYCLBO BrrutENT
O
                                                                                                                     crruiFHT
                     Figure 14-10.  Level 3 waste water treatment for titaniun dioxide  - sulfate process.

-------
     precipitation are employed.  Oxygen is supplied from  atmospheric
     air,  and polymer is added to assist in the second stage settling
     of iron hydroxide when required.  Aside from the bulk handling of
     large amounts of these common chemicals,  there  are  no  special
     hazards involved in their use.

C.   Separation and Removal of Solids

     Large quantities of thickener underflow are pumped  to  spreading
     areas  for consolidation of the solids, which are later pushed to
     18 foot high piles on land provided for 10  years  of  operation.
     Solids  from  occasional draining of the polishing lagoon and the
     Level 3 recycling lagoon are returned to the aeration step of the
     waste acid streams, after which they will be settled out  in  the
     second  stage  thickener,  being handled as part of the thickener
     underflow.  Although no dewatering  equipment  is  provided,  the
     first  and second stage thickeners are sources of pure and impure
     gypsum for future byproduct recovery.

Treatment Cost Estimates

General Discussion
To prepare
developed.
treatment  cost  estimates,   a  model  plant  concept  was
For conceptual design a representative unit waste flow was
selected,  together  with  three different Ti02 production rates.  The
latter were chosen to cover most of the rates  typical  for  the  TiQ2
(Sulfate Process) subcategory.  The selected daily Ti02 production for
the model plant was multiplied by the selected unit flow to obtain the
volume  of  influent  to  the treatment system.  The selected unit raw
waste  pollutant  loads  were  also  multiplied  by  the  model  plant
production  rate  to  determine  the  pollutant  load on the treatment
system.  Capital and equipment costs were  then  calculated  based  on
developed conceptual design parameters for each model plant production
rate. The rationale used for the model plant selection is given below.

A.   Wastewater Flow

     Waste effluent data is available for three plants and is given in
     Table 14-26.  For the model the average value of the three plants
     data has been used.  The unit flow data for  strong  acid  ranges
     from  6.1  to  15  mVkkg of Ti02.   (Table 14-26).  For the model
     plant the average value of 10 mVkkg has been used.   Unit  flows
     for  the  weak  acid  stream range from 67 to 78 m3/kkg.  For the
     model plant, a unit flow of 72 m3/kkg of Ti02 is used.  The third
     segregated stream includes contact cooling water, scrubber water,
     and finishing operation  wastewater.   The  unit  flow  for  this
     stream varies from plant to plant and depends largely on the type
     and  quality  of the Ti02 pigment end product desired.  For model
     plants a total flow which consists of  strong  acid,  weak  acid,
     scrubber  effluents,  and  product   finishing  wastewaters of 475
     mVkkg of Ti02 was used.
                                402

-------
B.   Production

     Four plants produce titanium dioxide by the sulfate process at  a
     total   production   rate   of  259,000  metric  tons  per  year.
     Production ranges from a minimum of 31,000 kkg/yr to a maximum of
     74,500 kkg/yr with an average of 49,000 kkg/yr and  a  median  of
     43,000  kkg/yr.   For  treatment cost estimates, three production
     levels were selected.  These were 31,800  kkg/yr,  47,700  kg/yr,
     and 74,500 kkg/yr.

C.   Wastewater Pollutant Load

As stated before, the principal pollutants occur  in  the  strong  and
     weak  acid  streams  and  include  high  acidity (sulfuric acid),
     suspended solids, iron and other heavy metal sulfates.  The other
     wastewaters contain titanium dioxide and small amounts  of  other
     heavy  metals  on  suspended  solids.   Iron  concentrations vary
     depending on the grade of ilmenite ore used.

Model Plant Control and Treatment Costs

The average raw waste pollutant loadings given  in  Table  14-27  were
used  for the model plant.  For the model plants, a total iron loading
of 600 kg/kkg was used with the assumption that two-thirds was spended
ferric hydroxide and  one-third  (200  kg/kkg  of  TiO2)  was  soluble
ferrous  iron.   The unit sulfate and suspended solid loadings for the
different wastewater streams for the model plant were:
Stream

Weak Acid
Strong Acid
Other Wastewater

E.   Chemical Usage
 Sulfate Loading
(kg/kkg of  Ti02)

     2,300
     1,800
     Negligible
  TSS Loading
(kg/kkg of 1i02)

       300
       200
       113
In the model BPT system, powdered limestone is used  for  first  stage
neutralization  of  mixed  strong  and weak acids, at the unit rate of
3,000 kg/kkg of TiOz.  For iron removal  (NSPS), pebble lime   (CaO)  is
used for second stage neutralization of  the mixed acid streams and for
the  final  neutralization  of  the total combined flow, including the
other miscellaneous wastes.  The  unit   application  of  CaO  for  all
purposes is 0.235 kg/kkg of Ti02.
F.   Solids Produced

Existing plants have attempted  to  produce  two  grades  of  saleable
gypsum  from  the strong and weak acid streams.  At present,  one plant
has successfully identified a market that will partly offset  the  cost
of  solids  handling  and  disposal.   The plant currently employs BPT
treatment which generates gypsum solids  of  sufficient  purity  to  be
sold.   NSPS  treatment generates iron oxide laden gypsum which is, at
                            403

-------
present, unsuitable for sale.  The total solids produced in the  model
plant  are assumed to be 5,200 kg/kkg of Ti02 for BPT and 5,500 kg/kkg
of Ti02 for NSPS.

Treatment Costs

The estimated costs  for  three  models  having  different  production
levels  are  given  in  Tables  14-29,  14-30,  and  14-31.  The costs
represent the total cost required for treatment system  implementation
and assume that the plant has no treatment system in place.

Table  14-32  presents a summary of the unit cost distribution between
amortization  and  operation  and  maintenance  cost   components   at
different  production  levels  and  at  the  BPT  and  NSPS  levels of
treatment.

For existing sources at  the  first  level  of  treatment,  the  costs
reflect disposal of sludge on-site, hence land requirements are fairly
large.   However,  this cost can be offset by a revenue on the sale of
gypsum, currently practiced at one plant.  These off-setting  revenues
are not included in the costs presented in Tables 14-29, 14-30, 14-31,
and  14-32,   These  off-setting revenues are included in the Economic
Impact Analysis of_ Pollution Control Technologies for Segments of  the
Inorganic   Chemicals   Manufacturing   Industry,   EPA  440/2-81-023.
Amortization, chemicals, labor, and residual waste disposal costs have
significant impact on the annual costs.

The total annual costs presented in Table 14-32 would increase  by  20
to  40  percent  if the sludge dewatering equipment is to be installed
for gypsum recovery.

Basis For Regulations

Evaluation of BPT Practices

Out of a  total  of  four  Ti02  plants   (sulfate  process)  that  are
currently  in  operation,  only  one  plant  (#559) has a BPT treatment
system.  The other three plants practice  partial  neutralization  and
settling.   The  BPT limitations are based on available long-term data
from Plant #559 and treatment performance information in Section 8.

A.   Pollutant Removal with BPT Treatment

     Treatment Level 1 is  equivalent  to  the  BPT  in  the  Titanium
     Dioxide   (Sulfate Process) industry.  Means, standard deviations,
     and variability factors were calculated from  data  submitted  by
     Plant  #559 for final effluent quality, and the results are given
     in Tables 14-33a and 14-33b.  The performance characteristics are
     utilized for the development of the BPT regulations.

     The ability of the treatment system  to  remove  conventional  and
     toxic  pollutants was estimated by comparing the treated effluent
                            4C4

-------
   TABLE  14-29.  MODEL PLANT TREATMENT COSTS
                Titanium dioxide - Sulfate process
                31,800 metric tons per year
Subcategory
Production
    INVESTMENT COST
                                       BPT
      Site development  	      232,500
      Equipment 	      958,600
      Monitoring equipment  ....        7, 800

           Subtotal	    1,198,900
      Contractor's 0 &  P  	      179,835

           Subtotal 	    1,378,735
      Engineering 	      275,800

           Subtotal 	    1,654,535
      Contingencies 	      275,800

           Subtotal 	    1,930,335
      Land	      636,000

      TOTAL INVESTMENT  COST	    2,566,335

B.  OPERATION AND
    MAINTENANCE COST

      Labor and supervision....      252,000
      Energy	       34,000
      Chemicals	    1,760,000
      Maintenance 	      193,032
      Taxes and insurance	       76,989
      Residual waste disposal..      141,000
      Monitoring, analysis,
       and reporting	       15,000

      TOTAL OPERATION AND
      MAINTENANCE COST	    2,472,021

C.  AMORTIZATION OF
    INVESTMENT COST                  314,063

    TOTAL ANNUAL COST	    2,786,084
BAT'

 0
 0
 0

 0
 0

 0
 0

 0
 0

 0
 0
   NSPS

  316,500
2,551,000
   20,000

2,887,500
  433,125

3,320,625
  664,125

3,984,750
  398,475

4,383,225
1,372,000

5,755,225
                                                        504,000
                                                         96,000
                                                      1,589,000
                                                        438,323
                                                        172,657
                                                        210,000

                                                         15,000
                                                  0   3,024,979


                                                  0     713,151

                                                  0   3,738,130
  Represents the incremental cost above that for BPT  treatment
  Overhead and Profit
                              405

-------
   TABLE 14-30.  MODEL  PLANT  TREATMENT COSTS
 Subcategory     Titanium dioxide - Sulfate process
 Production      47,700 metric tons per year


 A.   INVESTMENT  COST                    BPT        BATa     NSPS

       Site development  	      316,200       0      439,000
       Equipment 	    1,213,200       0    3,288,500
       Monitoring equipment  .	        7,800       0       20,000

           Subtotal	    1,537,200       0    3,747,500
       Contractor's 0 & P	      230,580       0      562,125

           Subtotal 	    1,767,780       0    4,309,625
       Engineering 	      353,560       0      861,925

           Subtotal 	    2,121,340       0    5,171,550
       Contingencies 	      353,560       0      517,155

           Subtotal 	    2,474,900       0    5,688,705
       Land	      960,000       0    1,920,000

       TOTAL INVESTMENT COST	    3,434,900       0    7,608,705

B.  OPERATION AND
    MAINTENANCE COST

       Labor and supervision....      336,000       0      672,000
       Energy	       51,000       0      138,000
       Chemicals	    2,640,000       0    2,384,000
       Maintenance 	      247,492       0      568,871
       Taxes and insurance	      103,047       0      228,261
       Residual waste disposal..      209,000       0      315,000
      Monitoring,  analysis,
       and reporting	       15,000       0       15,000

      TOTAL OPERATION AND
      MAINTENANCE COST	    3,601,539       0    4,321,132

C.  AMORTIZATION OF
    INVESTMENT COST                  402,669       0      925,552

    TOTAL ANNUAL COST	    4,004,208       0    5,246,684


  Represents  the incremental cost above that for BPT treatment

  Overhead and Profit
                              406

-------
   TABLE  14-31.  MODEL PLANT  TREATMENT  COSTS
Subcategory     Titanium dioxide - Sulfate process
Production      74,500 metric  tons per year

A.  INVESTMENT COST                    B?T       BATa        NSPS

      Site development 	       418,300        0       631,000
      Equipment	    1,622,700        0     4,352,000
      Monitoring equipment  	        7,800        0        20,000

           Subtotal	    2,048,800        0     5,003,000
      Contractor's O  & P ......       307,320        0       750,450

           Subtotal 	    2,356,120        0     5,753,450
      Engineering  	       471,200        0     1,150,690

           Subtotal 	    2,827,320        0     6,904,140
      Contingencies 	       471,200        0       690,414

           Subtotal 	    3,298,520        0     7,594,554
      Land 	    1,470,000        0     2,940,000

      TOTAL INVESTMENT COST	    4,768,520        0    10,534,554

B.  OPERATION AND
    MAINTENANCE COST

      Labor and supervision....       336,000        0       672,000
      Energy	       71,000        0       199,000
      Chemicals	    4,100,000        0     3,719,000
      Maintenance	       329,854        0       759,455
      Taxes and insurance	       143,056        0       316,037
      Residual waste  disposal..       283,000        0       420,000
      Monitoring, analysis,
       and reporting	       15,000        0        15,000

      TOTAL OPERATION AND
      MAINTENANCE COST	    5,277,910        0     6,100,492

C.  AMORTIZATION OF
    INVESTMENT COST                   536,672        0     1,235,634

    TOTAL ANNUAL COST	    5,814,582        0     7,336,126


a Represents the incremental cost above that  for BPT treatment
i_
  Overhead and Profit
                              407

-------
        TABLE  14-32  MODEL PLANT UNIT TREATMENT COSTS
     Subcategory  Titanium dioxide - Sulfate process
                                 Annual Treatment Costs  (S/kkg)
    COST  ITEM
PRODUCTION
 (kkg/yr)
  LEVEL OF TREATMENT

BPT       BAT*      NSPS
Annual Operation
and Maintenance


Annual
Amortization


Total Annual
Cost



31,800
47,700
74,500

31,800
47,700
74,500

31,800
47,700
74,500

77,74
75.50
70.84

9,88
8.44
7.20

37.61
83.95
78.05

NA
MA
NA

NA
NA
NA

NA
NA
NA

95.13
90.59
81.89

22.43
19.40
16.59

117.55
109.99
98.47
*Represents the incremental  cost  above  BPT
                             408

-------
     qualities with the raw  waste  qualities  of  the  sampled  waste
     streams.

Basis for BPT Effluent Limitations

A.   Technology Basis

     For BPT,  the  Agency  is  basing  limitations  on  equalization,
     limestone   neutralization  and  clarification,  followed  by  pH
     adjustment before final polish and  discharge  of  the  effluent.
     This  technology  is  chosen  because  it  has been installed and
     operated successfully by a plant in the industry.
                                            /
B.   Flow Basis

     Waste flow data is available for three  plants  and  the  average
     value  of 475 mVkkg of Ti02 (Table 14-26) is taken as the inflow
     for the  model  plant  treatment  system.   The  treatment  plant
     effluent  flow  is  taken to be the same as the influent flow and
     the loss or addition of  water  through  chemicals,  evaporation,
     precipitation, and through solids removal have been neglected.

C.   Selection of Pollutants to be Regulated

     Selection of pollutants for limitation was based on a  number  of
     criteria  including  raw  waste  pollutant  concentrations during
     sampling, total subcategory  raw  waste  pollutant  loading,  and
     experience  related  to  the  control  of  pollutant  removal  by
     alkaline treatment technology presented in Section 8.  The Agency
     decided not to regulate the nonconventional pollutant iron  under
     BPT  and BAT because of increased cost of the treatment when  iron
     is controlled and because  the  gypsum  solids  produced  by  the
     treatment  with iron removal can be reused only if dissolved  iron
     is not  controlled.  One plant has developed a market for reuse of
     the gypsum.  The Agency estimates that requiring iron removal  at
     existing  plants  would  increase  treatment  costs  by  up to 40
     percent  and  generate  large  quantities  of  waste  solids  for
     disposal.  Based on our economic impact analysis (63), the Agency
     believes  that  two  of  the  four  existing Ti02 sulfate process
     plants  would close, with the attendant increase in  unemployment,
     if  the  plants  were required to control iron.  Control of toxic
     metal pollutants wil.l be adequate at existing plants even without
     an iron removal step.

     Review  of Tables 14-27,  14-28,  and   14-34  indicates  that  six
     pollutants    were    consistently     identified   at   treatable
     concentrations.  Total suspended solids,  iron,  chromium,  lead,
     nickel,  and  zinc were all identified at treatable levels on the
     average during sampling.

     The toxic pollutants can be divided into  two  groups  exhibiting
     similar  behavior  depending  on  the  treatment  pH.   Review of
     alkaline treatment in Section 8 shows  that  chromium,  lead,  and
                               4G9

-------
          TABLE 14-33a.   HISTORICAL EFFLUENT M3NTIORING DATA SUMMARY
                        SUBCATEGQRY - TITANIUM DIOXIDE
                SULFATE PROCESS PLANT 1559 - TREATMENT WITHOUT
                                 IRON        ^1'

(2)
Parameter v '
Daily Data
No. of Points
Average, x
Standard Deviation, S
Standard Deviation,
S1 (of logs)
Variability Factor
Monthly Averages
No. of Points
Standard Deviation, S_
Variability Factor
Variability Factor
Ratio
Pollutant (mg/1)
TSS Iron Ghromiun Nickel
30(3) 84
50 330
59 73
0.95 Q.20
5.8 1.5

20 20
44 42
1.6 1.2
3.6 1.3
81
0.23
.0.20
0.75
4.4

20
0.11
1.9
2.3
82
0.14
0.12
0.74
4.3

20
0.099
2.1
2.0
Zinc
54
0.50
0.60
0.95
5.8

13
0.21
1.7
3,4
(1)
(2)
(3)
Lead and cadmiun concentrations vrere less than 0.10 rag/1 for the 20
months.

Refer to Table 14-33b for explanation of statistical parameters.

Based on one month of data for April 1981.
                                    410

-------
            TABLE 14-33b.HISTORICAL EFFLUENT r-UNlTORINC  DATTl


                          SJUZATLODRY. - TITANIUM DIOXIDE


             SULTATE PROCESS P1/MT 1559  - TTCA"T!UJT KITH HOI
Pollutant
TSS Cadnium Chrcnuum Iron Lead Nickel
DaiJy Data
No. of Points 899 109 128 854 128 128
Average, x 21.0 0.060 0.070 0.62 0.06S 0.08
Deviation S(1J 65-93 0.044 0.054 3.46 0.041 0.071
Euindard 1.54 0.68 0.67 1.86 0.56 0.76
Deviation, S' '
Factor^ ^ 11'° 3"85 3'B1 13-65 3'16 4'39
Month lv Averages
Ito. of Points 30 26 30 28 30 3D
Standard 21.84 0.042 0.038 0.94 0.04 0.048
Variability 2U 4 39
Factor
Variabil ity
Factor Ratio
VFRE5) 3.62 1.58 1.87 3.38 !..« 1.00
Zinc
128
0.151
0.204
1.02
6.41
30
0.16
3.05

2.10
 (1)      S is the ajrit>metic stairiaj-d deviation and  is  given, by
             S  =
                       i=n  ,      -, 2
                            (x:  -  x)
                           n-1


               xi is the data value for point  i




               x  is the mean value



               n  is the number of data, points





 (2)      S*  is the estimated standard deviation  of  tl* logarithms




            S'
                S  is tlie arithmetic standard deviation





               jT  is the mean value




 (3)      Tie variability factor [VF) of daily measurements for kgnonnal  distribution

         is found by the expression



                                In  (VF) =  S' (  Z - 0.5 S1)


         where S' is defined in Note 2 above.


                                2= 2.33 for 99th percentile



(4)       The variability factor  (VF) for  30-day average measurements is found by the

         expression


                   VF = 1.0  + Z
               x  is the mean value  of the monthly averages
                m


              S  is the arithmetic stan^oi'd deviation of the monthly averages
               in

              2  =1.64 for 95th r«rctntile



(5J       VFR:  Ratio of the 24-hour variability factor to the 30-day variability factor





                                             411

-------
TABLE 14-34.  VERIFICATION RESULTS FRCM - SULFATE PROCESS
              TITANIUM DIOXIDE PLANT #559

Waste Stream

Pollutant Raw Waste Treated Ef f luent
Unit Load Concentration Unit Load Concentration Perceval
Efficiency

-------
     zinc can be removed together at pH 8.0-9.5,  while nickel requires
     a  higher pH.   Therefore,  chromium and nickel were selected based
     on their relative significance in the raw waste and due to  their
     representative  nature  regarding  overall  toxic  metal removal.
     Lead, zinc, cadmium, copper, arsenic, and antimony are  of  lower
     significance  on the basis of sampling results.  No toxic organic
     pollutants were identified at significant concentrations.

D.   Basis of Pollutant Limitations

     1.   Conventional and Nonconventional Parameters

     a.   pH

          The treated effluent is to be controlled within the range of
          pH 6.0 to  9.0.   This  limitation  is  based  on  the  data
          presented in Appendix B of the proposed Development Document
          (60) and the JRB Study (52).

     b.   TSS and Iron

          The long-term average values of 50 mg/1 for TSS and 330 mg/1
          for iron derived from the  monitoring  data  of  Plant  1559
          (Table  14-33a)  was  used  as  the  subcategory performance
          values.  The variability factors for daily measurements  and
          monthly  averages  estimated  from Plant #559 long-term data
          (Table 14-33a) were used in  calculating  the  concentration
          basis and effluent limitations as described below.

               Total Suspended Solids

               The TSS maximum 30-day average concentration  is:

                (50 mg/1)(7.6) = 80 mg/1

               The 24-hour maximum TSS is:

                (50 mg/1)(5.8) = 290 mg/1

               The  TSS  maximum  30-day  average  effluent  limit was
               obtained by using the model  plant  unit  flow  of  475
               m3/kkg as follows:

                (80 mg/1) (475 mVkkg)  (  kq/m^   ) = 38 kg/kkg
                                      (1000 mg/1)

               The 24-hour maximum TSS effluent  limit is:

                (290 mg/1)  (475 mVkkg) (  kg/m3   ) =  140 kg/kkg
                                       (1000 mg/1)
                              413

-------
          Iron

          The   maximum  30-day  average  iron  concentration  is
          determined similarly from Table 14-33a as follows:

          (330 mg/l)(l.2} = 400 mg/1

          The 24-hour maximum iron concentration is:

          (330 mg/l)(l.5) = 500 mg/1

     The concentrations are presented in Table 14-35 as guidance.

Toxic Pollutants

The effluent limitations for chromium and nickel were established
on the basis of long-term monitoring data for  Plant  #559  which
are presented in Table 14-33a.

a.   Chromium

     The chromium limitation  is  based  primarily  on  long-term
     monitoring  data  available  for  Plant  1559.   A treatment
     system raw waste influent concentration for chromium of  3.1
     mg/1  was  observed  at  Plant  #559  during  sampling.  The
     long-term  data  summarized  in  Table  14-33a  indicates  a
     current  treatment  performance  of  0.23 mg/1 and a monthly
     average and daily measurement variability factor of 1.9  and
     4.4, respectively.

     The  chromium  maximum 30-day average concentration is given
     by:

     (0.23 mg/1)(1.9) =0.44 mg/1

     The chromium 24-hour maximum concentration is given by:

     (0.23 mg/1}{4.4) =1.0 mg/1

     The  chromium  maximum  30-day  average  effluent  limit  is
     developed  using  the  model  plant  flow  of 475 m3/kkg and
     concentration limit above as follows:

     (0.44 mg/1) (475 mVkkg (  kg/in3  ) = 0.21 kg/kkg
                             (1000 mg/1)

     The chromium 24-hour maximum effluent limit is given
     by:

     (1.0 mg/1) (457 mVkkg) (  kg/m3  ) =0.48 kg/kkg
                             (1000 mg/1)

-------
                     TABLE 14-35.  EFFLUENT LIMITATIONS
                      TITANIUM DIOXIDE SULFATE PROCESS

          Best Practicable Control Technology Currently Available
                   Waste Water Flew:  475 m3/kkg of TiO2

Concentration Basis Effluent Limit
(rag/1) (kg/kkg of Ti02)
Subcategory
Performance
Pollutant (mg/1)
Mav
/T \ £*JdA*
VFRU' 30-day
Avg.
Max.
24-hr 30-day 24-hr
Max. Avg. Max
Conventional and Nonconventional Pollutants
Total Suspended
Solids
Iron
Toxic Pollutants
Chromium
Lead
Nickel
Zinc
Cadmium
Copper
Antimony
Arsenic
50(2>
330 (2)
0.23 t2>
0.15<3)
0.14<2>
0.50(2)
0.10(3)
0.40 (3'
0.80(5)
0.50(5)
5.8 80
1.6
1.5 400
1.2
4.4 0.44
1.9
3.2 0.32
2.1
4.3 0.29
5.8 0.85
1.7
3.9 0.24
2.4
3.8 0.80
2.0
3.8 1.6
2.0
3.8 1.0
2.0
290 38 140
500 -<4' -(4)
1.0 0.21 0.48
0.48 -<4> _«>
0.60 0.14 0.29
(4) (4)
2.9 — w —^'
f4} (4)
0.39 — w — w
1.5 — —
3.0 — (4) — (4)
(4) (4)
1.9 — V4' — v*'
(1)   VFR:   Ratio of the variability factor of the daily measurements  to the
           variability factor of the monthly averages.
(2)   Long-term average based on loading data and variability factors  of
     Plant #559 selected frcm Table 14-33a.
(3)   Limitation based on long-term data at Plant #559 and review of Table 8-12,
(4)   No effluent limitation.
(5)   The lower limit of the literature treatability estimate (Table 8-11)  is
     used as the basis for the  long-term average limitation.

-------
b.    Nickel

     The long-term average concentration for nickel at Plant #559
     is 0.14 mg/1 from Table 14-33a  for  the  treated  effluent.
     The  average  nickel  concentration observed during sampling
     was 0.18 mg/1 for Plant #555 and #559  (Table  14-28).   The
     long-term average and variability factors observed for Plant
     #559 were used to develop the limitations as follows:

     The maximum 30-day concentration for nickel is:

     (0.14 mg/1) (2.1) = 0.29 mg/1

     The 24-hour maximum concentration is:

     (0.14 mg/1) (4.3) = 0.60 mg/1

     The nickel maximum 30-day average effluent limit is:

     (0.29 mg/1) (475 m3/kkg) (   kq/m3  ) = 0.14 kg/kkg
                              (1000 mg/1)

     The nickel 24-hour maximum effluent limit is:

     (0.60 mg/1 (475 mVkkg) (  kg/in^  ) =0.29 kg/kkg
                              (1000 mg/1)

c.    Other Metals

     Lead, zinc, cadmium, copper, arsenic, and antimony were also
     present in the raw wastewater.   A  concentration  level  is
     developed  for  these  toxic pollutants which can be used to
     develop a  limitation  in  the . event  that  they  become  a
     concern.    Development   of   the  concentration  level  is
     primarily made on a similar basis as established  previously
     for  the  primary pollutants.  Information used to establish
     the concentration level for each pollutant is summarized  as
     follows:
                           416

-------
Summary

Pollutant

Zinc

Cadmium

Copper

Arsenic

Antimony
Lead
(1)

(2)
                Long-term
                 Average

               0.50  mg/l(5)

               0.10  mg/1(1,2)

               0.40  mg/l(3)

               0.50  mg/l(4)

               0.80  mg/1(4)

               0.15  mg/l(l,2)
Variabilityd )
Factor of the
Monthly Average

       1 /7

       2.4

       2.0

       2.0

       2.0

       2.1
Variabilityd )
 Factor of the
   Daily
Measurements

       5.8

       3.9

       3.8

       3.8

       3.8

       3.2
     Table 14-33b

     Table 8-12
(3)  Table 8-13

(4)  Table 8-11
(5)  Table 14-33a

The concentration levels are developed for all the pollutants as
follows:
The maximum 30-day = (Long-term average
concentration            concentration)
                                               (Monthly average
                                               variability factor)
The 24-hour maximum =  (Long-term average  X    (Daily measurement
                        concentration)         variability factor)
Summaries of the conventional and toxic pollutant limitations for  BPT
are presented  in Table  14-35.

Basis for BCT  Effluent  Limitations

Wh i 1 e   EPA  has  not   yet  proposed  or  pr omu 1 gated  a  r evi sed  BCT
methodology in response  to  the  American  Paper  Institute  v.  EPA
decision  mentioned  earlier,  EPA is promulgating BCT limitations for
this subcategory.  These  limits are identical  to those for BPT because
the  only  technology   option  that  removes   significant  amounts  of
conventional   pollutants  is  not  economically  achievable.   See the
discussion of  iron under  BPT above.  Removal of significant additional
amounts of conventional pollutants can be achieved in this subcategory
only if iron is also removed.

As BPT is the  minimal  level of control required by  law,  no  possible
application  of  the  BCT cost  tests could result in BCT limitations
lower than those promulgated today.  Accordingly, there is no need  to
wait  until  EPA  revises the BCT methodology before promulgating BCT
limitations.
                                  417

-------
                        TABLE 14-36.  EFFLUENT LIMITATIONS
                         TITANIUM DIOXIDE SULFATE PROCESS
                            Best Available Technology
                       Wastewater Flow: 475 m3A)cg of TiO2

Concentration Effluent Limit

Subcategory( "" '
Pollutant Performance
Nonconventional
Pollutants :
Iron 330
Toxic Pollutants:
Chromium 0.23
Lead 0.15
Nickel 0.14
Zinc 0.50
Cadmium 0.10
Copper 0.40
Antimony 0.8Q(4>
Arsenic 0.50(4)

VFR<2>

1.5/1.2
4.4/1.9
3.2/2.1
4.3/2.1
5.8/1.7
3.9/2.4
3.8/2.0
3.8/2.0
3.8/2.0
Basis
Max.
30 -day
Avg.

400
0.44
0.32
0.29
0.85
0.24
0.80
1.6
1.0
(mg/1) (kg/kkg of TiO;>)
Max. ~
24-hr . 30-day 24-hr .
Max. Avg. Max.

500 —(3) —(3)
1.0 0.21 0.48
0.48 — <3) — (3)
0.60 0.14 0.29
2Q 	 \ J ) __\-3j
• 3
0.39 —0) -0)
1-5 -0) — <3>
o » U ™*^ — ^
L9 —(3) __(3)
(1)   Limitations for BPT Table 14-35.
(2)   VFR:  Ratio of the variability factor of the daily measurements to the
          variability factor of the monthly averages.
(3)   No effluent limitation.
(4)   The lower limit of the literature treatability estimate (Table 8-11) is used
     as the basis for the long-term average limitation.
                                  418

-------
Basis for BAT Effluent Limitations


For BAT, the Agency is basing limitations on treatment  consisting  of
Level  1  technology,   and  is  the same as BPT.  Treatments requiring
aeration for iron removal (Level 2) and 55 percent recycle through use
of soda ash precipitation  (Level  3)  were  considered  but  rejected
because of costs.  The limitations for BAT are given in Table 14-36.

Basis for New Source Performance Standards

For  NSPS, the Agency is basing limitations on equalization, limestone
neutralization, clarification, aeration,  alkaline  precipitation  and
settling,  followed  by  pH adjustment before final polish in a lagoon
and discharge of the effluent (Level 2).  This  technology  is  chosen
because it has been installed and operated successfully by Plant #559.

A.   Flow Basis

     Waste flow data is available for three  plants  and  the  average
     value  of  475  mVkkg  of  Ti02   (Table   14-26)  is taken as the
     influent to the model treatment system.

B.   Selection of Pollutants to be Regulated

     The same pollutants considered  for BPT are  also  considered  for
     the  NSPS  regulations.  These  pollutants  include total suspended
     solids  (TSS), iron, chromium, lead, nickel, and zinc  which  were
     identified  at  treatable  levels on the average during sampling.
     Iron was selected in addition to TSS, chromium,  and  nickel  for
     regulation on the basis of additional removal of large quantities
     of  iron  and  additional  TSS  and toxic metal removal beyond the
     BPT/BAT level with  the  application  of  aeration  and  alkaline
     precipitation to BPT treatment.

     An  iron  removal  step  added  to BPT/BAT technology is used for
     NSPS/PSNS  because  the  additional  treatment  provides   better
     removal  of  toxic  metals,  and the more stringent standards are
     unlikely to pose a significant  barrier  to  entry.   New  sulfate
     process  plants  are unlikely because the  two alternate processes
     are-more economical.  However,  if a company did want to construct
     a  new sulfate process  plant,   a  process  change   involving   the
     recycle  and  reuse of the strong  acid wastewater would likely  be
     adopted.  This process change would reduce production costs,   and
     would   also  reduce  the amount of strong  acid wastewater treated
     and discharged by 70 to  90 percent.  Reducing  the  flow  of   the
     strong  acid wastewater  reduces treatment  costs substantially  and
     also substantially reduces the  amount of gypsum  solids  produced
     by treatment.  With the  smaller amount of  gypsum solids produced,
     disposal  of  the solids as waste  is competitive with sale of  the
     solids  for  reuse, when cost of  sales is considered.  At least one
     company using the sulfate  process   is  actively  developing   the"
     recycle/reuse  technology.   One other company using the chloride
                                  419

-------
                        TABLE 14-37. EFFLUENT LIMITATIONS
                         TITANIUM DIOXIDE SULFATE PROCESS
                         New Source Performance Standards
                       Wastewater Flow: 475 m3/kkg of TiO2
Pollutant
Subcategory
Performance
 (mg/1)
Concentration
 Basis (mg/1)
Max.
30-day    24-hr.
Avg.      Max.
                              Effluent Limit
                              (kgAkg of TjQ7)
                              Max.
                              30-day     24-hr.
                              Avg.        Max.
Conventional and
 Nonconventional
 Pollutants	

Total Suspended
 Solids
Iron
    (6)
Toxic Pollutants
0.62
    <2>
11.0/3.0  64

13.7/4.0  2.5
          230

          8.5
30

1.2
110

4.1
Chromium ' ° '
Lead
Nickel (6)
Zinc
Cadmium
Copper
Antimony
Arsenic
0.15(3)
0.15(3)
0.10<3>
0.20<3)
0.10*3)
0.40<3>
0.80<5>
0.50<5>
3.8/2.0
3.2/2. 1
3.8/2.0
6.4/3.1
3.9/2.4
3.8/2.0
3.8/2.0
3.8/2.0
0.30
0.32
0.20
0.62
0.24
0.80
1.6
1.0
0.57
0.48
0.38
1.3
0.39
1.5
3.0
1.9
0.14
— (4)
0.095
— (4)
— (4)
__(4)
— (4)
— (4)
0.27
— (4)
0.18
— (4)
— (4)
— (4)
— (4)
— (4)
(1) VFR: Ratio of the variability factor of the daily measurements to the
         variability factor of the monthly averages.

(2) Long-term average based on loading data and variability factors of Plant #559
    selected from Table 14-33b.

(3) Limitation based on long-term data at plant #559 and review of Table 8-12.

(4) No effluent limitation.

(5) The lower limit of the literature treatability estimate {Table 8-11) is used
    as the basis for the long-term average limitation.

(6) Applicable to PSNS regulations.
                                        420

-------
     ilmenite process has recently built a new plant  that  discharges
     50 percent less total wastewater than comparable plants.

C.    Basis of Pollutant Limitations

     1.   Conventional and Nonconventional Parameters

          a.   pH

               The treated effluent is to  be  controlled  within  the
               range  of  pH  6.0 to 9.0.  This limitation is based on
               the data  presented  in  Appendix  B  of  the  proposed
               Development Document (60) and the JRB Study (52).

          b.   TSS and Iron

               The long-term average values of 21  mg/1  for  TSS  and
               0.62  mg/1 for iron derived from the monitoring data of
               Plant 1559 (Table  14-33b) were used as the  subcategory
               .performance  values.  The variability factors for daily
               measurements and monthly averages estimated from  Plant
               #559   long-term   data   (Table  14-33b)  were  used  in
               calculating  the   concentration  basis   and   effluent
               limitations as determined below

               Total Suspended Solids

               The TSS maximum 30-day average concentration  is:

               (21 mg/1)  (3.04) - 64 mg/1

               The 24-hour maximum TSS  is:

               (21 mg/1)  (11.0) = 230 mg/1

               The  TSS  maximum  30-day  average  effluent  limit was
               obtained  by using  the model  plant  unit  flow  of  475
               m3/kkg as follows:

               (64 mg/1)  (475 mVkkg  (  kq/m3   ) = 30 kg/kkg
                                      (1000 mg/1)

               The 24-hour maximum TSS  efffluent  limit is:

               (230 mg/1)  (475 mVkkg)  (  kq/m3   )  =  110 kg/kkg
                                        (1000 mg/1)


               Iron

               The   maximum  30-day  average   iron   concentration   is
               determined similarly  from Table  14-33b as  follows:
                                  421

-------
          (0.62 mg/1)  (4.0)  = 2.5 mg/1

          The 24-hour  maximum iron concentration is:

          (0.62 mg/1)  (13.7)  = 8.5 mg/1

          The maximum  30-day average iron effluent limit is:

          (2.5 mg/1)  (475 mVkkg) (   kg/m^  )  =1.2 kg/kkg
                                  (1000 mg/1)

          The 24 hour  maximum iron effluent limit is:

          (8.5 mg/1)  (475 mVkkg) (   kq/m3  )  =4.1 kg/kkg
                                  (1000 mg/1)


          The   conventional   and   nonconventional   pollutants
          limitations   are presented in Table 14-37 for treatment
          with iron removal.

2.    Toxic Pollutants

     The effluent limitations for chromium with iron  removal  is
     based  primarily  on long-term monitoring data for Plant #559
     and review of treatment performance  information  in  Tables
     8--12  and  8-13  for  other  industries  utilizing  the same
     treatment technology for removal of the pollutant.

     a.   Chromium

          The chromium limitation  with  iron  removal  is  based
          primarily  on  long-term  monitoring data available for
          Plant #559.   A  treatment  system  raw  waste  influent
          concentration  for chromium of 3.2 mg/1 was observed at
          Plant  #559   during  sampling.   The   long-term   data
          summarized   in   Table   14-33b  indicates  a  current
          treatment performance  of  0.070  mg/1  and  a  monthly
          average and daily measurement variability factor of 2.0
          and 3.8, respectively.

          Review of long-term average treatment data for chromium
          in Tables 8-12 and 8-13 indicates that 0.070 mg/1 shows
          excellent  removal  when compared to the same treatment
          technology for chromium removal in other industries.  A
          long-term average of 0.15 mg/1 for chromium is used  as
          the  basis  for  the final limitation based on observed
          performance data from other  industries  summarized  on
          the  tables  in  Section  8.  The final  limitations are
          developed below for chromium.

          The chromium maximum 30-day  average  concentration  is
          given by:
                               422

-------
Basis for Pretreatment Standards

A.   Existing Sources

     The only wastewater discharged to the POTW from the one  existing
     indirect  discharger  is a small portion of the weak acid stream.
     Based on data provided in response to Section 308 request, it  is
     unlikely  that  this  wastewater  contains  any  toxic  metals at
     treatable levels.  Therefore, -we are excluding  this  subcategory
     from  categorical  PSES under the provisions of Paragraph 8(b) of
     the Settlement Agreement.

B..   New Sources

     Tor pretreatment standards for New Sources  (PSNS), the Agency  is
     basing  limitations  on NSPS.  The pollutants to be regulated are
     iron,  chromium,,  and  nickel  as  indicated  in   Table   14-37.
     Pretreatment   is required because NSPS provides better removal of
     iron and  toxic  metals  than  does  a  well-operated  POTW  with
     secondary treatment  installed and therefore chromium, nickel, and
     iron would pass through the  POTW  in  the absence of pretreatment

Titanium Dioxi.de -  Chloride - Ilmenite Process  Industry Profile

General Description

Total  subcategory  production capacity is given  in Table ,14-38 Profile
Data Summary.  Tiie  30S .data available  for the   TiOz  Subcategory  does
not  adequately cover the one-step chloride-ilmenite process; however,
supplementary information iteas been -submitted by   industry   (55).   The
status  of  regulations   prior  to  promulgation of this regulation  is
presented   -in   Table    T4-2.    Additional    information   on    the
chloride-ilmenite process industry is  given  in  this section.

General Process Description and Raw Materials

For   the   manufacture   of  titanium  dioxide  by  the  combined ore
beneficiation-chloride process, a generalized  process  flow  diagram
including the waste streams is shown in Figure  14—11.

The direct use of ilmenite ore for the manufacture of titanium dioxide
pigments requires the application of either the sulfate process or the
one-step  ore  beneficiation/chlorination process which is referred  to
in this document as the chloride-ilmenite process.   Processes   which
involve  a  separate ore  beneficiation step  (either at the plant  or at
the ore source) resulting in an upgraded  or a synthetic rutile product
to be used as feed  material  for  a  chloride  process  would  not  be
classified   as   a  chloride-ilmenite   process.    A  separate ore
beneficiation process would fall  within the Ore  Mining  and  Dressing
Category  for regulatory  purposes, and the manufacture of Ti02 from an
upgraded  ilmenite  or  synthetic rutile would  be   in   the   same
classification as a chloride process using natural rutile ore.
                                427

-------
          TABLE 14-39.   AVERAGE VftTER USAGE FOR TiO? PROTECTION
                        BY THE CHDORIDE -HMENTTE PROCESS

Use Plant #237 Plant #550
, 3 - .
Cm /kkg of TiO2>
Noncontact 73-140 330-390
Cooling
Process Contact 100-140 (1) 47- 59
and Cleanup
Noncontact 9- 11 6-7
Ancillary Uses
(Boilers,
Sanitary, etc.)
Plant #713

15-16

60 (2)

5- 6



Source of data,  (55).


(1)  Tha average total flow of 120 m /kfcg is used as  the basis for 8PI.


                            3
(2)  The average flow of 60m /kkg is used as the  basis for NSPS.
                                    428

-------
                                   TABLE 14-41.  SUM4ARY OF RAW HASTE LOADINGS FOUND IN SCREENING AND VERIFICATION SAMPLING
W

SUBCATEOORY TITAN I LW DIOXIDE - SULFATE PROCESS (Applied to Chloride Ilirenite Process)
Pollutant Loading Range
(kg/day)
Mininun Maxinun
Priority
Antimony 5.0 28
Arsenic 1.9 4.0
Cadmiun 0.068 7.2
Chromium 140 530
Copper 8.2 19
Lead 3.0 65
Nickel 3.7 23
Selenium 7.0 9.5
Thalliun 0,47 1.2
Zinc 1.8 85
Conventional
TSS
Iron
Unit Loading

Minimum Average Maximum

0.032 0.11 0.22
0.012 0.19 0.032
0.00044 0.019 0.057
1.1 2.0 3.4
0.065 0.085 0.12
0.024 0.18 0.42
0.029 0.080 0.15
0.0020 0.031 0.06
0.0030 0.0055 0.0080
0.014 0.34 0.55
320
600
No. of
Plants (1)

3
3
3
3
3
3
3
2
2
3
1
1
                             (1) - Data are taken only from those plants where pollutants were found above detection limits, or
                                  in the case of TSS and Iron, where data are available.

-------
                                          TABLE 14-42.  TOXIC POLLUTANT AVERAGE RAW HASTE LOADS AND OCNCENTRftTICHS
U>
•U

SUaCATEGORY
TITANIUM DIOXIDE - Sulfate Process
(Applied to Chloride Ilmenite
Screening
Plant 1555

-------
Daily  raw waste loads were calculated from the flow rates measured or
estimated  at  the  time  of  sampling  and  the  measured   pollutant
concentrations.  That is,

Daily loading  (as kg of pollutant per day) = (C) (Q)
                                              1000

where  the concentration (C) of the pollutant is expressed in units of
mg/1,  (Note:  1 kg/m3  =  1000  mg/1 ) ,   and  the  flow  rate  (Q)  is
expressed  in  units of mVday (m3, a cubic meter,  is equal to 264 U.S.
gallons) .
Similarly, the unit loading were calculated  from  the  reported
production  rate (P), the waste stream flow rate (Q), and the measured
pollutant concentration (C).

Unit loading (as kg of pollutant =  (C) (Q)
per kkg of Ti02)                    1000(P)

where C and Q are expressed in the same units described above, and the
production (P) is expressed in units of kkg/day (kkg  is  1000  kg,  a
metric ton, which is equal to 2205 Ibs).

A summary of daily and unit per unit of production raw waste  loads for
the  plants  sampled  is  presented  in Table 14-41 and the individual
plant averages are given in Table 14-42.

The estimated total annual raw wastewater  load  of  toxic  pollutants
generated by the Chloride-Ilmenite Process is given below.
Pollutant
Chromium
Nickel
Zinc
Lead
Copper
Cadmium
Antimony
Thallium
Arsenic
Selenium
                           Total Annual Raw Wastewater Load
                                        (kg/year)

                                        1 ,050,000
                                           42,000
                                          178,000
                                           94,000
                                           44,000
                                            9,900
                                           58,000
                                            2,900
                                           99,000
                                           16,000
Section  5  of  this report described the scope and methodology of the
sampling program and this section indicates the size of the analytical
data base on toxic pollutants for the sulfate process  segment.   This
is   the   basis   for   selecting   pollutants   of  concern  in  the
chloride-ilmenite segment of the Ti02 subcategory.
                               435

-------
Pollution Abatement Option

Toxic Pollutants of Concern

Rationale for  selection  of  the  toxic  pollutants  of  concern  are
presented  in  this  section  for  the  sulfate process industry.  The
sampling data evaluations resulted in the selection of chromium, zinc,
nickel, lead, copper, antimony, arsenic, and cadmium on the  basis  of
raw waste maximum concentrations and total annual industry loads.

The  major  impurity  found  in  the various grades of raw material is
ferrous iron as shown in Table 14-22.   In  the  sulfate  process  the
unwanted  iron  remains  largely  in  the  ferrrous  state  and may be
crystallized out of the  acid  waste  streams  and  sold  as  copperas
(ferrous  sulfate}.   In  the  chloride-ilmenite process, the same ore
impurity is oxidized to some extent to the  ferric  state  during  the
chlorination  step.   This appears in the acid waste streams as ferric
chloride (FeClj) in the amounts indicated in Table 14-41.

Iron, in either the ferrous  or  ferric  state,  is  classified  as  a
nonconventional  pollutant.   However,  when present in large amounts,
such as it is in the Ti02 industry, it can be a  considerable  aid  to
toxic metal removal in treatment systems designed to take advantage of
coprecipitation processes.

Process Modifications and Technology Transfer Options

The  comments  made in regard to the Titanium Dioxide-Chloride Process
for rutile and upgraded ores in this section are generally  applicable
to the Chloride-ilmenite Process.

Best Management Practices

Storm water runoff from the plant site should be collected and sent to
the treatment facility for the removal of suspended solids.

Prevailing Control and Treatment Practices

The  three chloride-ilmenite plants from which the water use and waste
source information  was  obtained  all  handle  the  disposal  of  the
concentrated  process  waste stream separately by either ocean dumping
or deep well injection.  The availability of either of  these  methods
of  disposal  for  a  particular  plant  is  a  matter  handled  on   a
case-by-case basis by the appropriate regulatory agencies  from  which
various   approvals   and   permits  -are  required  under  the  Marine
Protection, Research and Sanctuaries Act of 1972 for ocean disposal or
by state and local authorities  for  deep  well  injection.   For  the
purpose  of  developing  the  model  plant  concept  and  specifying  a
generally   applicable   waste   treatment    technology    for    the
chloride-ilmenite  industry, the assumption has been made that neither
the ocean dumping nor the deep well  injection  disposal  options  are
generally  available,  and  that  the  concentrated  process waste is.
                              436

-------
therefore, included in the raw  waste  influent  to  the  model  plant
wastewater treatment system.

In practice,  one plant disposes of the entire metal chloride, HC1, and
TSS  waste  by  ocean dumping.  The remainder of the plants dispose of
the concentrated waste by deep well injection  after  use  of  surface
lagoons for removal of settleable solids.

The dilute process waste streams are segregated to the extent possible
from  noncontact  sources and treated in conventional in-plant systems
utilizing equalization and spill diversion facilities followed by lime
neutralization/coagulation, solid separation in a settling  pond,  and
final  discharge of the treated effluent.  Chemical coagulating agents
such as ferric chloride and alum may be used either before or after pH
control as an aid  in  the  removal  of  metal  hydroxides  and  other
suspended solids.

Advanced Treatment Technology

Advanced  treatment  technology  options for the in-plant treatment of
process  wastes  have  been  evaluated  as  possible  polishing   step
additions  to  a conventional system for equalization, neutralization,
and clarification in ponds prior to discharge.  Such options include:

A.   Aeration  for  a)  decarbonization  if  limestone   is  used   for
     neutralization,  and  b)  ferrite  coprecipitation, assuming that
     sufficient ferrous iron is already present or  is   added  to  the
     system  as  needed (the latter may^also be accomplished by adding
     scrap iron to the aqid wastes).

B.   An alkaline precipitation step under optimum conditions for metal
     hydroxide precipitation, i.e., pH 9-10.

C.   Dual-media filtration for additional removal of suspended  solids
     including toxic metal hydroxides.

D.   Sulfide precipitation for additional toxic metal removal followed
     by filtration.

E.   Other   metal    removal    technologies    including    xanthate
     precipitation,  ion  exchange,  and membrane applications, all of
     which  were  regarded  as  inappropriate  from  a   practical  and
     economic point of view,

Selection of Appropriate Technology and Equipment

Technologies for Different Treatment Levels

A.   Level 1 (BPT)

     Figure 14-12 shows the model  treatment system for treatment  with
     and  without iron removal.  Calcium carbonate (limestone) is used
     to neutralize the concentrated acid waste  stream.   Aeration  is
                            4-27

-------
00
                  Figure 14-12.
Level 1 waste water treatment for titanium dioxide - chloride
(ilmenite ore) process.

-------
Figure 14-13.
Level 2 waste water treatment for titanium dioxide - chloride
(ilmenite ore) process.

-------
     then  used  to  primarily  remove  C02  prior to pH adjustment to
     reduce demand for treatment chemicals, and mixing with dilute and
     miscellaneous plant wastes.  The combined stream  is  given  lime
     treatment  to pH 9-10 for iron and toxic pollutant removal or 6-9
     during treatment without  iron  and  toxic  metajl  removal.   All
     wastewater  passes through a clarifier and final' polishing lagoon
     prior to discharge.

     This  technology  was  used  as  the  treatment  model  for   BPT
     regulations  because  of the similarity of wastes to those in the
     Ti02-Sulfate Process Industry.  This technology is available and,
     to some degree, already employed  in  the  Ti02-Chloride-Ilmenite
     Industry.   The proposed BPT treatment can remove greater than 95
     percent of the major pollutants of concern including toxic metals
     according to preliminary treatability estimates.  No iron removal
     is required for BPT.

B.   Level 2 (NSPS)

     Level 2 treatment adds iron removal as described in Level  1  and
     dual-media  filtration  to  the Level 1 technology for additional
     removal of iron, suspended solids,  and  toxic  metal  hydroxides
     following  the  alkaline  precipitation  and settling steps.  The
     flow diagram for Level 2 is shown in Figure 14-13.  This level of
     treatment was selected as the basis for NSPS because it  provides
     a  relatively  economical  method  for  removing additional toxic
     metals.

Equipment for Different Treatment Levels

A.   Equipment Functions

     Unlike  treatment  of  the  wastewaters  from  the  Ti02  Sulfate
     Process,   limestone   neutralization  of  the  Chloride-Ilmenite
     Process wastewaters does not generate large quantities of  solids
     (e.g.,  gypsum)  which require mechanized separation and transfer
     to sizable on-site or off-site disposal areas.  The  solids  that
     are  generated  from  TSS and metal precipitate separation can be
     collected in moderate-sized lagoons and periodically  transferred
     to  appropriate  chemical  landfill  disposal sites in accordance
     with the  Resource  Conservation  and  Recovery  Act  (RCRA)  (as
     amended,  42  USC  6901,  et  seq.).  The Level 1 treatment model
     includes rail car deliveries of ground limestone and lime, bucket
     elevators, storage bins, multiple reactors and chemical  feeders,
     mechanical  aerators  and  thickeners  for  solids  removal.  The
     clarified overflow is treated  with  lime  for  additional  toxic
     metals  removal  and settled in a one-day polishing pond prior to
     final pH adjustment, monitoring and discharge.

B.   Chemicals and Chemical Handling

     First stage neutralization utilizes ground limestone  while  lime
     is  used  for  second  stage  neutralization  and  final alkaline
                              440

-------
     precipitation.   Oxygen is supplied as air to oxidize any  ferrous
     iron   remaining  in  addition  to  COZ  removal,   and  treatment
     chemicals may be added as required for  removal  of  precipitated
     metals  and  other  suspended solids.  Aside from the large scale
     bulk chemical handling requirements for limestone and lime, there
     are no particular hazards involved.

C.    Disposal of Solids

     Periodic  removal  of  solids  from  settling  impoundments  will
     require  compliace with RCRA regulations as applicable to on-site
     or off-site chemical  disposal  site  operation.    However,  this
     subcategory  has  not  been  listed  under RCRA-ISS for hazardous
     pollutants.

Treatment Cost Estimates

General Discussion

To prepare  treatment  cost  estimates,  a  model  plant  concept  was
developed.   Cost  estimates  have  been  prepared  for  Level  1 (BPT)
treatment, and for Level 2  (NSPS).   The  model  plant  specifications
given  below  were utilized for cost estimating and for development of
the regulations.

A.    Production

     There are three  plants  at  different  locations  producing  (or
     capable  of  producing) titanium dioxide by the combined ilmenite
     ore beneficiation - chlorination  process.   Annual  capacity  of
     these  plants  varies  from  136,000 metric tons to 207,000 metric
     tons.  For treatment cost estimates, four production levels  were
     selected.   These  were 35,000 kg/year, 70,000 kkg/year, 113,7570
     kkg/year, and 157,500  kkg/year.

B.    Wastewater Flows

     Wastewater is  typically  segregated   into  two  streams;  strong
     acidic   wastewater    flow   from  beneficiation-chlorination  of
     ilmenite ore and air emission scrubbing facilities, and the other
     wastewater from process reactions, washings,  product  transport,
     cooling  tower  blowdown,  water  treatment  blowdown,  and other
     operations.  For the model plants, a unit flow  of  6  mVkkg  of
     product  for the concentrated acidic wastewater and 114 mVkkg of
     product for the dilute wastes is used.  The treatment  system  is
     designed  to  handle a total flow of  120 mVkkg of product (Table
     14-39).

     For the NSPS model plant, a  unit flow  of 6 mVkkg of product  for
     the  concentrated acidic wastewater  is used.  Because of improved
     design which  allows   for  recycle   systems  and  more  efficient
     process  water  utilization,  dilute   wastewater  is considerably
     reduced  (62).  The total combined wastewater flow of 60 mVkkg of

-------
     product is used (Table 14-39).  The treatment system for Level  2
     (NSPS) is BPT plus dual-media filtration.

C.   Pollutant Load

     The principal pollutants occurring in the  wastewaters  are  TSS,
     iron,  chromium,  nickel,  zinc,  and hydrochloric acid.  For the
     model plants,  the  following  unit  pollutant  loads  have  been
     considered:

                 TSS           175 kg/kkg of Ti02
                 HC1           230 kg/kkg of TiOz
                 Iron          375 kg/kkg of Ti02
                 Chromium      1.4 kg/kkg of TiO2
                 Nickel        0.1 kg/kkg of Ti02
                 Zinc          0.5 kg/kkg of Ti02

     The  loading  values  for  TSS,  HC1,  and iron are based on data
     submitted by industry  on  the  chloride-ilmenite  process.   The
     chromium  loading  is  an  estimated  average derived from a wide
     range of ilmenite ore qualities and the nickel and zinc  loadings
     are  taken  from  the screening and verification data on the Ti02
     sulfate process.

D.   Chemical Usage

     In the model BPT system,  powdered limestone  is  used  for  first
     stage neutralization of strong acidic waste flow at the unit rate
     of  302  kg/kkg  of  Ti02.   Pebble lime (CaO) is used for second
     stage neutralization of the total* combined flows.  Lime  is  used
     at the unit rate of 42 kg/kkg of Ti02.

E.   Solid Waste

     The solids produced in the treatment  facility  consist  of  iron
     hydroxides,  the  original  suspended  solids  introduced  in the
     influent and solids derived from the  treatment  chemicals  added
     for neutralization.  The total solids produced in the model plant
     are  assumed  to  be  890  kg/kkg  of  Ti02.  Assuming an average
     production rate and 50 percent moisture content,  the  volume  of
     sludge generated is 1.8 mVkkg.

Model Plant Control and Treatment costs

The estimated costs for four models having different production levels
are given in Tables 14-43, 14-44, 14-45, an<3 14-46.

For  existing sources at the first level of treatment, the disposal of
sludges  is  on-site,  hence  land  requirements  are  fairly   large.
Amortization, chemicals, labor, and residual waste disposal costs have
significant impact on the annual costs.
                              442

-------
   TABLE 14-43.  MODEL PLANT TREATMENT COSTS
Subcategory    Titanium dioxide - Chloride  process  (Ilmenite  ore)
Production     35,000 metric tons per year
A. INVESTMENT COST

     Site development  	
     Equipment  	
     Monitoring equipment  ..

          Subtotal  	
     Contractor's 0 &  Pb....

          Subtotal  	
     Engineering  	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL INVESTMENT  COST  .


B. OPERATION AND
   MAINTENANCE  COST

     Labor and  supervision  .
     Energy 	
     Chemicals  	
     Maintenance  	
     Taxes and  insurance  ...
     Residual waste disposal
     Monitoring,  analysis,
      and reporting 	
                                                     ($)
      BPT

  292,200
  851,500
   20,000

1,163,700
  174,555

1,338,255
  267,651

1,605,906
  160,591

1,766,497
  252,000

2,018,497
     TOTAL OPERATION AND
     MAINTENANCE COST
  336,000
   31,000
  260,000
  176,650
   60,555
  105,000

   15,000
  984,205
BAT

  0
  0
  0

  0
  0

  0
  0

  0
  0

  0
  0
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
  287,409

1,271,614
 aRepresents the  incremental  cost  above  that  for BPT treatment
   Overhead and Profit
                             443

-------
   TABLE  14-44.  MODEL PLANT TREATMENT  COSTS
Subcategory
Production
Titanium dioxide - Chloride process  (Ilmenite ore)
70,000 metric tons per year
A. INVESTMENT COST

     Site development  	
     Equipment  	
     Monitoring equipment  ..

          Subtotal  .........
     Contractor's O & Pb....

          Subtotal  	
     Eng ineer ing  	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL INVESTMENT COST  .


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  .
     Energy 	
     Chemicals  	
     Maintenance  	
     Taxes and  insurance
     Residual waste disposal
     Monitoring, analysis,
      and reporting 	
                               BPT

                           520,000
                         1,023,500
                            20,000

                         1,563,500
                           234,525

                         1,798,025
                           359,605

                         2,157,630
                           215,763

                         2,373,393
                           492,000

                         2,865,393
     TOTAL OPERATION AND
     MAINTENANCE COST
                           504,000
                            43,000
                           510,000
                           237,339
                            85,962
                           105,000

                            15,000
                         1,500,301
BAT

  0
  0
  0

  0
  0

  0
  0

  0
  0

  0
  0
C. AMORTIZATION OP
   INVESTMENT COST

   TOTAL ANNUAL COST
                           386,151

                         1,886,452
  Represents the incremental cost above  that  for  BPT  treatment

  Overhead and Profit
                             444

-------
   TABLE 14-45.  MODEL PLANT TREATMENT COSTS
Subcategory
Production
Titanium dioxide - Chloride process  (Ilmenite ore)
113,750 metric tons per year
A. INVESTMENT COST

     Site development  	
     Equipment  	
     Monitoring equipment  ..

          Subtotal	
     Contractor's 0 &  P  ....

          Subtotal  	
     Engineering  	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	.	

     TOTAL INVESTMENT  COST  .


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  .
     Energy 	
     Chemicals  	
     Maintenance  	
     Taxes and insurance  ...
     Residual waste disposal
     Monitoring, analysis,
      and reporting 	
                               BPT

                           784,000
                         1,377,000
                             20,000

                         2,181,000
                           327,150

                         2,508,150
                           501,630

                         3,009,780
                           300,978

                         3,310,758
                           780,000

                         4,090,758
     TOTAL OPERATION AND
     MAINTENANCE COST
                           588,000
                            62,000
                           823,000
                           331,076
                           122,723
                           210,000

                            15,000
                          2,151,799
BAT

  0
  0
  0

  0
  0

  0
  0

  0
  0

  0
  0
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                           538,660

                         2,690,459
  Represents the incremental cost above  that  for  BPT  treatment
  Overhead and Profit
                             445

-------
   TABLE 14-46.  MODEL PLANT TREATMENT  COSTS
Subcategory
Production
Titanium dioxide - Chloride process  (Ilmenite ore)
157,500 metric tons per year
A. INVESTMENT COST

     Site development	
     Equipment 	
     Monitoring equipment  	

          Subtotal  ....v*........
     Contractor's 0 & P  .	

          Subtotal  	
     Engineering  	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	*	

     TOTAL INVESTMENT COST  	


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  	
     Energy 	
     Chemicals 	
     Maintenance  	
     Taxes and insurance 	
     Residual waste disposal  ....
     Monitoring, analysis,
      and reporting	

     TOTAL OPERATION AND
     MAINTENANCE COST
                               BPT

                         1,053,600
                         1,590,000
                            20,000

                         2,663,600
                           399,540

                         3,063,140
                           612,628

                         3,675,768
                           367,577

                         4,043,345
                         1,080,000

                         5,123,345
                           588,000
                            71,000
                         1,141,000
                           404,334
                           153,700
                           210,000

                            15,000
                         2,583,035
                                                     ($)
BAT

  0
  0
  0

  0
  0

  0
  0

  0
  0

  0
  0
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                           657,852

                         3,240,887
  Represents the incremental cost above  that  for  BPT  treatment
  Overhead and Profit
                             446

-------
     TABLE 14-47.  MODEL PLANT TREATMENT COSTS
Subcategory    Titanium dioxide - Chloride  (Ilmenite ore)
Production     35,000 metric tons per year
                                                     _

       A. INVESTMENT COST                           NSPS

            Site development 	         280,000
            Equipment  	        1,046,500
            Monitoring equipment  	           20,000

                 Subtotal  	        1,346,500
            Contractor's 0 & P a.	         201,975

                 Subtotal  	        1,548,475
            Engineering	         309,695

                 Subtotal  	        1,858,170
            Contingencies  	         185,817

                 Subtotal  	        2,043,987
            Land	         252,000

            TOTAL  INVESTMENT COST 	        2,295,987
       B. OPERATION  AND
          MAINTENANCE  COST

             Labor  and  supervision  	          350,000
             Energy 	           34,200
             Chemicals  	          260,000
             Maintenance  	          204,399
             Taxes  and  insurance  	           68,880
             Residual waste  disposal  ....          105,000
             Monitoring,  analysis
              and  reporting  	           15,000

             TOTAL  OPERATION AND
             MAINTENANCE  COST                   1,037,478
        C.  AMORTIZATION OF
           INVESTMENT  COST                        332,557

           TOTAL  ANNUAL COST                    1,370,035

   Overhead and Profit
                              447

-------
     TABLE 14-48.  MODEL PLANT TREATMENT COSTS
Subcategory    Titanium dioxide - Chloride  (Ilmenite  ore)
Production     70,000 metric tons per year

                                                     ($)
       A. INVESTMENT COST                           NSPS

            Site development 	          496,500
            Equipment 	        1,323,500
            Monitoring equipment  	           20,000

                 Subtotal  	        1,840,000
            Contractors 0 & P a.	          276,000

                 Subtotal  	        2,116,000
            Engineering 	          423,200

                 Subtotal  	        2,539,200
            Contingencies  	          253,920

                 Subtotal	        2,793,120
            Land	          492,000

            TOTAL INVESTMENT COST  	        3,285,120
       B. OPERATION AND
          MAINTENANCE COST

            Labor and supervision  	          518,000
            Energy  	           49,500
            Chemicals 	          510,000
            Maintenance	          279,312
            Taxes and insurance  	           98,554
            Residual waste  disposal  ....          105,000
            Monitoring,  analysis
             and reporting  	           15,000

            TOTAL OPERATION AND
            MAINTENANCE  COST                    1,575,366
       C. AMORTIZATION OF
          INVESTMENT COST                         454,441

          TOTAL ANNUAL COST                     2,029,806

   Overhead  and Profit
                              448

-------
   TABLE 14-49.  MODEL PLANT TREATMENT COSTS
Subcategory    Titanium dioxide - Chloride  (Ilmenite ore)
Production     113,750 metric tons per year
                                                     —

       A. INVESTMENT COST                           NSPS

            Site development  	          752,000
            Equipment  	        1,822,000
            Monitoring equipment  	           20,000

                 Subtotal  	        2,594,000
            Contractor's 0 &  Pa	          389,100

                 Subtotal  	        2,983,100
            Engineering  	          59*5,620

                 Subtotal  	        3,579,720
            Contingencies  	          357,972

                 Subtotal  	        3,937,692
            Land	          780,000

            TOTAL  INVESTMENT  COST 	        4,717,692
       B.  OPERATION  AND
           MAINTENANCE  COST

             Labor  and  supervision  	          602,000
             Energy 	           71,700
             Chemicals  	          823,000
             Maintenance  	          393,769
             Taxes  and  insurance  	          141,531
             Residual waste  disposal  ....          210,000
             Monitoring,  analysis
              and  reporting	           15,000

             TOTAL  OPERATION AND
             MAINTENANCE  COST                   2,257,000
       C. AMORTIZATION OF
           INVESTMENT COST                         640,662

          TOTAL  ANNUAL COST                    2,897,662

  Overhead and Profit
                              449

-------
     TABLE 14-50.  MODEL PLANT TREATMENT COSTS
Subcategory    Titanium dioxide - Chloride  (Ilmenite ore)
Production     157,500 metric tons per year

                                                     ($)
       A. INVESTMENT COST                           NSPS

            Site development  	        1,015,000
            Equipment 	        2,175,000
            Monitoring equipment  	           20,000

                 Subtotal  	        3,210,000
            Contractor's 0 &  Pa	          481,500

                 Subtotal  	        3,691,500
            Engineering 	          738,300

                 Subtotal  	        4,429,800
            Contingencies  	          442,980

                 Subtotal  	        4,872,780
            Land	        1,080,000

            TOTAL INVESTMENT  COST  	        5,952,780
       B. OPERATION AND
          MAINTENANCE COST

            Labor and supervision  	          602,000
            Energy 	           85,000
            Chemicals 	        1,141,000
            Maintenance  	          487,278
            Taxes and insurance  	          178,583
            Residual waste disposal  ....          210,000
            Monitoring,  analysis
             and reporting  	           15,000

            TOTAL OPERATION AND
            MAINTENANCE  COST                    2,718,861
       C. AMORTIZATION OF
          INVESTMENT COST                         792,801

          TOTAL ANNUAL COST                     3,511,663

   Overhead and Profit
                              450

-------
        TABLE 14-51. MODEL PLANT UNIT TREATMENT  COSTS
     Subcategory Titanium dioxide - Chloride  (Ilmenite  ore)
                                Annual Treatment  Costs ($/kkg)
    COST ITEM
PRODUCTION
 (kkg/yO
    LEVEL OF TREATMENT

  BPT       BAT*      NSPS
   Annual Operation
   and Maintenance
   Annual
   Amortization
   Total Annual
  35,000
  70,000
 113,750
 157,500
  35,000
  70,000
 113,750
 157,500
28.12
21.43
18.92
16.40
 8.21
 5.52
 4.74
 4.18
NA
NA
NA
NA
NA
NA
NA
NA
29.64
22.51
19.84
17.26
 9.50
 6.49
 5.63
 5.03
Cost



35,000
70,000
113,750
157,500
36.33
26.95
23.65
20.58
NA
NA
NA
NA
39.14
29.00
25.47
22.30
*Represents the incremental cost  above  BPT
                             451

-------
The unit waste flow of 6 m3/kkg of product for the concentrated acidic
wastewater  stream  is  the  same  for BPT and NSPS systems.  The NSPS
treatment technology is the same as BPT,  but the total combined acidic
and dilute wastewater flow for the NSPS system is much less  than  for
the  BPT  model;  however,  the  reduced flow has negligible impact on
costs because the unit waste loads are the same.  Table 14-47,   14-48,
14-49,  and  14-50 present the estimated NSPS treatment costs for four
models having different production  levels.   There  is  insignificant
difference  in  the  estimated  total  annual costs per kkg of product
between BPT and NSPS levels of treatment for the model plant designs.

Table 14-51 presents a summary of the unit cost  distribution  between
amortization   and   operation  and  maintenance  cost  components  at
different productions at the BPT and NSPS level of treatment.

Basis for Regulations

Evaluation of BPT Treatment Practices

The  prevailing  control  and  treatment   practices   in   the   TiOz
Chloride-Ilmenite  industry  have  been reviewed in this section.  For
the purpose of regulations  development,   it  has  been  assumed  that
neither  ocean  dumping  nor deep well injection methods are generally
available  as  disposal  options  for  all  or  any  portion  of   the
process-related wastes.

Basis for BPT Effluent Limitation

A.   Technology Basis

     The  BPT  limitations   are   based   on   technology   involving
     equalization,  limestone neutralization, clarification, aeration,
     alkaline  precipitation,  and  settling  followed  by  final   pH
     adjustment  and  discharge.   The  rationale for the selection of
     Level 1 technology is given above.

B.   Flow Basis

     The BPT model plant flow rate is based on  the  reported  average
     process contact and clean up wastewater flow at Plant #237 of 120
     mVkkg as indicated in Table 14-39.

C.   Selection of Pollutants to be Regulated

     Specific  information  concerning  the  characteristics  of   the
     combined   strong   and   dilute   acid   waste  streams  in  the
     Chloride-Ilmenite process is not available.  In view of the  lack
     of  such  information  which  would indicate the concentration of
     pollutants in this industry, it is necessary to use data from the
     Sulfate Process as a guide in the selection of pollutants  to  be
     limited.   Considering  the many similarities in the chloride and
     sulfate processes, including use  of  the  same  ore  with  known
     impurities,  treatment  technology,  and generation of strong and
                              452

-------
dilute acid waste streams,  it  is  extremely  likely  that  TSS,
chromium, and nickel are present at treatable concentrations, and
are,  therefore,  limited  for  this industry.  It is also likely
that iron, lead, zinc, cadmium, copper, antimony, and arsenic may
also  be  present  on  occasion  at  treatable  levels  and  are,
therefore,  considered  as  potential  candidates for limitation.
Details on the selection of these pollutants for  limitation  are
presented  in  this  section  for  the  Titanium  Dioxide-Sulfate
Process.

1.    Conventional and Nonconventional Parameters

     a.   TSS and Iron

          The removal characteristics of total  suspended  solids
          (TSS)  is  influenced by the removal of iron.  For BAT,
          the long-term average concentration of 50 mg/1 for  TSS
          and  330 mg/1 for iron were achieved (Table 14-33a) and
          were used as the performance values.   The  variability
          factors  for  daily  measurements  and monthly averages
          estimated from Plant #559  (Table 4-33a)  were  used  in
          determining   the   concentration  basis  and  effluent
          limitations for TSS as follows:

          The maximum monthly average TSS concentration is:

          (50 mg/1) (1.6) = 80 mg/1

          The 24-hour maximum TSS concentration  is:

          (50 mg/1) (5.8) = 290 mg/1

          The maximum monthly average  limitation is:

          (80 mg/1) (120 mVkkg)  (   kq/m^   )  =9.6 kg/kkg
                                  (1000 mg/1)

          The 24-hour maximum  limitation is:

          (290 mg/1)  (120 mVkkg)  (  kq/rn^   ) =  35 kg/kkg
                                   (1000 mg/1)

          For iron, the concentration  basis  is determined  from
          data  in Table 14-33a as follows:

          The maximum 30-day average concentration is:

          (330 mg/1)  (1.2) = 400 mg/1

          The maximum 24-hour  concentration  is:

          (330 mg/1)  (1.5) = 500 mg/1
                         453

-------
                        TABLE 14-52.  EFFLUENT LIMITATIONS
                 TITANIUM DIOXIDE CHLORIDE PROCESS USING ILMENITE
             Best Practicable Control Technology Currently Available
                           Wastewater Flow: 120 m3/kkg
Pollutant
Estimated
Treatability
                                              Concentration
                                              Basis (mg/1)
                                              Max.
                                              30-day   24-hr.
                                              Avg.     Max.
Effluent Limit
 (kq/kkq)
Max.
30-day    24-hr.
Avg.      Max.
Conventional/
Nonconvent iona 1
Pollutants
Total Suspended
Solids SO*2)
Iron 330*2)
Toxic Pollutants
Chromium*5^ 0.23
Lead 0.15
Nickel*5* 0.14
Zinc 0.50
Cadmium 0.10
Copper 0.40
Antimony 0.80
Arsenic 0.50
5.8/1.6 80 290 9.6
1.5/1.2 400 500 —(4)
4.4/1.9 0.44 1.0 0.053
3.2/2.1 0.32 0.48 — *4)
4.3/2.1 0.29 0.60 0.035
5.8/1.7 0.85 2.9 —(4)
3.9/2.4 0.24 0.39 —(4)
3.8/2.0 0.80 1.5 —(4)
3.8/2.0 1.6 3.0 —(4)
3.8/2.0 1.0 1.9 —(4)
35
— (4)
0.12
..(4)
0.072
— (4)
__(4)
— (4)
— (4)
..(4,
(1)   VFR: The ratio of the variability factor for daily measurements to the
          variability factor for monthly averages.

(2)   Long-term average from plant #559 monitoring data (Table 14-33a)

(3)   Long-term average and variability factors are based on the same rationale as
     the TiO2 Sulfate Process Table 14-35.

(4)   No effluent limitation.

(5_)   Applicable to BAT limitations.
                               454

-------
          The concentration basis and appropriate limitations
          are summarized in Table 14-52.

2.    Toxic Pollutants

     The toxic pollutants that are limited include  chromium  and
     nickel.    Lead,  zinc,  cadmium,  copper,  antimony,  and arsenic
     are  considered  as  potential   candidates  for  limitation.
     Details  concerning  the selection and treatment performance
     of these pollutants for the  purpose  of  setting  numerical
     limitations  is identical to discussions in this section for
     the Titanium Dioxide-Sulfate Process.   Only  those  factors
     considered specific to the Chloride-Ilmenite Process for the
     purpose  of setting the limitations are considered.  The BPT
     effluent limitations are presented in Table 14-52.

     a.   Chromium

          Chromium limitations for BPT treatment  were  based  on
          Table  14-35.  The maximum 30-day average concentration
          of 0.44 mg/1 and daily  maximum  concentration  of  1.0
          mg/1 were used to determine the limitations as follows:

          The maximum 30-day average limitation is:

          (0.44 mg/1) (120 mVkkg (   kq/m3  ) = 0.053 kg/kkg
                                  (1000 mg/1)

          The 24-hour maximum limitation  is:

          (1.0 mg/1) (120 mVkkg} (   kq/m^  ) = 0.12 kg/kkg
                                  (1000 mg/1)

          The limitations are presented in Table 14-52 for BPT
          treatment.

     b.   Nickel

          Nickel limitations for  BPT  treatment  were  based  on
          Table  14-35.  The maximum 30-day average concentration
          of 0.29 mg/1 and daily maximum  concentration  of  0.60
          mg/1 were used to determine the limitations as follows:

          The maximum 30-day average limitation is:

          (0.29 mg/1) (120 mVkkg) (   kq/m3  ) = 0.033 kg/kkg
                                  (1000 mg/1)

          The maximum 24-hour limitations is:

          (0.60 mg/1) (120 mVkkg) (   kq/m3  ) = 0.72 kg/kkg
                                  (1000 mg/1)
                           455

-------
               The  limitations  are  presented in Table 14-52 for BPT
               treatment

          c.   Other Metals

               Potential candidates for limitation include lead, zinc,
               cadmium,  copper,  antimony,  and  arsenic  which   are
               included in the limitations on a concentration basis in
               the  event  they  are  identified  as  a  concern.  The
               concentration basis values are presented in Table 14-52
               for BPT and are identical to values developed in  Table
               14-35 for the Titanium Dioxide -Sulfate Process.

Basis for BCT Effluent Limitations

While   EPA  has  not  yet  proposed  or  promulgated  a  revised  BCT
methodology in  response  to  the  American  Paper  Institute  v.  EPA
decision  mentioned  earlier,  EPA is promulgating BCT limitations for
this subcategory.  These limits are identical to those for BPT because
the  only  technology  option  that  removes  significant  amounts  of
conventional  pollutants  is  not  economically  achievable.   See the
discussion of iron under BPT in the Sulfate Process above.  Removal of
significant additional  amounts  of  conventional  pollutants  can  be
achieved in this subcategory only if iron is also removed.

As  BPT  is  the minimal level of control required by law, no possible
application of the BCT cost tests  could  result  in  BCT  limitations
lower  than those promulgated today.  Accordingly, there is no need to
wait until EPA revises the BCT  methodology  before  promulgating  BCT
limitations.

Basis for BAT Effluent Limitations

For  BAT, the Agency is basing limitations on the application of Level
1 technology which is- equivalent to BPT.  The model plant  flow  basis
of  120 m3/kkg used for BPT is also used for BAT.  The BAT limitations
are presented in Table 14-52.  A more advanced technology  using  soda
ash  precipitation  and  recycle  of wastewater was considered for the
similar sulfate process but was rejected on  the  basis  of  cost  and
because its performance has not been demonstrated.

Basis for the New Source Performance Standards

A.   Technology Basis

     For NSPS the Agency is basing limitations on the  application  of
     Level 2 treatment technology which adds dual-media filtration and
     iron  removal  steps  to the BPT system for greater efficiency in
     the removal of toxic metals, iron, and  suspended  solids.   This
     treatment  greatly  increases the coprecipitation of toxic metals
     and prevents large quantities of  dissolved  and  suspended  iron
     from  entering  receiving water bodies and interfering with other
     water uses.
                             456

-------
B.   Flow Basis

     The reported data on process contact and clean-up wastewater flow
     at Plant #173 is selected as the basis of a model plant  for  new
     sources.   Process modifications resulting in a greatly increased
     efficiency of water use reduce the average flow rate to 60 mVkkg
     as shown in Table 14-39.

C.   Basis for Pollutant Limitations

     1.   Conventional Parameters

          a.   pH

               The treated effluent is to  be  controlled  within  the
               range  of  pH  6,0 to 9.0.  This limitation is based on
               data  presented  in  Appendix   B   of   the   proposed
               Development Document (60) and the JRB Study (52).

          b.   TSS

               The concentration basis  for  the  TSS  maximum  30-day
               average  limitation  is obtained by applying an average
               filtration efficiency of 38 percent removal (41,60)  to
               the  corresponding  concentration  of 64 mg/1 with iron
               removal (Table 14-37).  That is:

               (1.00-0.38) (64 mg/1) = 40 mg/1

               The maximum 30-day average limitation  is  obtained  by
               applying the NSPS model plant flow rate of 60 mVkkg:

               (40 mg/1)  (60 mVkkg) (  kq/rn^  ) =2.4 kg/kkg
                                     (1000 mg/1)

               The TSS daily maximum limitation is determined as
               follows:

               (1.00-0.38) (230 mg/1) = 140 mg/1

               The daily maximum limitation is:

               (140 mg/1) (60 mVkkg)   (  kq/m^  ) =8.4 kg/kkg
                                        (1000 mg/1)

               The NSPS limitations are presented in Table 14-53.

     2.   Nonconventional Pollutants

          The only nonconventional pollutant of concern is iron.   For
          NSPS,   the  Agency  is  basing  a  maximum  30-day  average
          limitation on an average filtration efficiency of 38 percent
          removal (41,60).  Thus the appropriate  concentration  basis
                              457

-------
                     TABLE 14-53.  EFFLUENT LIMITATIONS
             Titanium Dioxide - Chloride Process Using Ilmenite

                     New Source Performance Standards*
                        Waste Water Flow:  60 m3/kkg

Pollutant
Estimated
Treatability
(mg/D \i
Concentration Basis
(mcr/1)
Max.
m 30-day 24^ir.
rFRu' Avg. Max,
Effluent Limit
Ocq/kka)
Max.
30-day 24-hr.
Avg. flax.
Conventional and
Ndnconventional~
Pollutants

Total suspended   13
  Solids
Iron
0.38
11.0/3.0   40      140


13.7/4.0   1.6     5.3
                                          2.4
         8.4
0.096    0.32
Toxic Pollutants
Chromiurrr J
I«d<3>
Nickel (3)
Zinc (3)
Cadmium1 '
Copper1 J
Antimony
Arsenic
0.060
0.030
0.086
0.19
0.075
0.23
0.80
0.50
3.8/2.0
3.2/2.1
3.8/2.0
6.4/3.1
3.9/2.4
3.8/2.0
3.8/2.0
3.8/2.0
0.12
0.064
0.17
0.58
0.18
0.46
1.6
1.0
0.23
0.096
0.33
1.2
0.29
0.87
3.0
1.9
0.0072 0.014
_(4) _(4)
0.010 0.020
(4) (4)
(4) f4)
(4) (41
J4) _(4)
_{4) _(4)
 (1)  VFR:  Ratio of the variability factor for the daily measurements to
     the variability factor of the monthly averages.
 (2)  Based on the application of pollutant - specific removal efficiencies
     for dual-media filtration to adjust the BPT performance on treatability
     estimates shown in Table 14-52.
  *  Including pretreatment standards for new sources  (PSNS) covering iron
     and toxic metals which are expressed as concentrations.
 (3)  Applicable to PSNS limitations.
 (4)  No effluent limitations.
                                     458

-------
is derived from the corresponding concentration basis of 2.5
mg/1  of  iron  used  for  the maximum 30-day average (Table
14-37).  That is:

(1.00-0.38) (2.5 mg/1) =1.6 mg/1

The limitation for NSPS is:

(1.6 mg/1) (60 mVkkg) (  kq/m3  ) = 0.096 kg/kkg
                       (1000 mg/1)

Similarly the 24-hour maximum iron concentration is:

(1.00-0.38) (8.5 mg/1) =5.3 mg/1,

and the limitation is:

(5.3 mg/1) (60 mVkkg) (  kq/m3  ) - 0.32 kg/kkg
                       (1000 mg/1)

Toxic  Pollutants

Addition of iron removal and a dual-media filtration  system
as a polishing step in NSPS is expected to remove additional
toxic  pollutants.    This  removal  is  estimated  based  on
literature treatability  (41)  and  the  treatability  study
(61).   The  following  removal  efficiencies  for the toxic
pollutants  are  used for  the  limitations  after  careful
consideration of available information:

Toxic  Pollutant	Percent Removed     Percent Remaining
Chromium
Lead ( 1 )
Nickel
ZincO )
Cadmium( 1 )
Copper ( 1 )
Antimony( 1
Arsenic( 1 )
60
80
14
6
25
42
0
0
40
20
86
94
75
58
100
100
 (1)No   limitation   given;   concentration   basis presented  in
 Table  14-53.

 a.   Chromium

     The  long-term  average  concentration  is   determined  for
     chromium  by application  of  a  60  percent  removal  to  0.15
     mg/1  in Table  14-37  as follows:

      (0.15  mg/1)  (1.00-0.60)  = 0.060  mg/1
                      459

-------
     The maximum 30-day concentration for chromium is:

     (0.060 mg/1)  (2.0) = 0.12 mg/1,

     and the 24-hour maximum concentration is:

     (0.060 mg/1)  (3.8) = 0.23 mg/1

     Therefore,  the maximum 30-day average concentration and
     24-hour     maximum    limitation    are    determined,
     respectively:

     (0.12 mg/1) (60 m3/kkg) (  kq/m 3  )  = 0.0072 kg/kkg
                             (1000 mg/1)

     (0~. 23 mg/1) (60 m'/kkg) (  kq/m3  )  = 0.014 kg/kkg
                             (1000 mg/1)
b.   Nickel
     The long-term average concentration is  determined  for
     nickel  by  application of a 14 percent removal to 0.10
     mg/1 in Table 14-37 as follows:

     (0.10 mg/1) (1.0-0.14) = 0.086 mg/1

     The maximum 30-day concentration for nickel is:

     (0.086 mg/1) (2.0) = 0.17 mg/1,

     and the 24-hour maximum concentration is:

     (0.086 mg/1) (3.8) = 0.33 mg/1

     Therefore, the maximum 30-day average concentration and
     24-hour    maximum    limitation    are     determined,
     respectively:

     (0.17 mg/1) (60 m3/kkg) (  kq/m3  ) = 0.010 kg/kkg
                             (1000 mg/1)

     CO. 33 mg/1) (60 mVkkg) (  kq/rn^  ) = 0.020 kg/kkg
                             (1000 mg/1)

     Other Metals

     The   concentration  basis  for  lead,  zinc,  cadmium,
     copper, antimony,  and  arsenic  are  determined  in  a
     similar   manner   to   the  primary  pollutants.   The
     following equation is  used  in  conjunction  with  the
     concentration   information  in  Table  74-37  and  the
     percent removal presented in this section:
                   -460

-------
               BPT long-term  x Percent remaining = NSPS long-term
               average               100            average
               concentration                        concentration

               The maximum 30-day average  concentration  and  24-hour
               maximum  concentration  are  determined  by  using  the
               appropriate variability  factors  in  Table  14-37  and
               multiplying   by   the  NSPS  long-term  average.    For
               example,  the long-term average for zinc is:

               (0.20 mg/1) (0.94) = 0.19 mg/1

               The maximum 30-day average concentration is:

               (0.19 mg/1) (3.1) = 0.59 mg/1

               The 24-hour maximum concentration is:

               (0.19 mg/1) (6.4) = 1.2 mg/1

               Similarly, the concentration bases are  determined  for
               the  remaining  toxic  pollutants  and are presented in
               Table 14-53.

Basis for Pretreatment Standards

Pretreatment is  necessary  because  it  provides  better  removal  of
pollutants  than  is achievable by a well operated POTW with secondary
treatment installed, and  thereby  prevents  pass-through  that  would
occur in a POTW in the absence of pretreatment.

Existing Sources

Since  there  are  no  indirect  dischargers   in this subcategory, the
Agency is excluding this subcategory from categorical PSES  under  the
provisions of paragraph 8(b) of the Settlement Agreement.

New Sources

The  Agency  is promulgating PSNS that are equal to NSPS because these
standards provide better removal of iron, chromium and nickel than  is
achieved  by  a  well-operated POTW with secondary treatment installed
and therefore iron, chromium and nickel would pass through a  POTW  in
the  absence  of  pretreatment.   Pollutants  regulated under PSNS are
iron, chromium, and nickel.  See Table 14-53.
                               460-A

-------

-------
                              SECTION 15

                      ALUMINUM FLUORIDE INDUSTRY
Industry Profile

General Description

Aluminum fluoride is used as a  raw  material  in  the  production  of
cryolite  (sodium  fluoroaluminate),   which  in  turn  is  used in the
production  of  aluminum.   Aluminum  fluoride  is  used  also  as   a
metallurgical  flux (for welding rod coatings), as a ceramic flux (for
glazes and enamels), and as a brazing flux (for aluminum fabrication).

The industry profile data for this  subcategory  are  given  in  Table
15-1,  while  the  status of regulations prior to promulgation of this
new regulation is given  in Table 15-2.

General Process Description and Raw Materials

In the dry process for the manufacture of aluminum fluoride, partially
dehyrdated alumina hydrate is reacted with hydrofluoric acid gas.  The
reaction is given as:

     Al£03 + 6HF = 2A1F3 + 3H20

The product, aluminum fluoride, is formed as a solid,  and  is  cooled
with  noncontact  cooling  water before being milled and shipped.  The
gases from the reactor are scrubbed with  water  to  remove  unreacted
hydrofluoric acid before being vented to the atmosphere.  A simplified
flow diagram of the process is shown  in Figure 15-1.

Water Use and Waste Source Characteristics

Water Use

Water  is  used  in  noncontact  cooling  of the product, for seals on
vacuum pumps and for scrubbing the reacted gases before  being  vented
to  the atmosphere.  Water is also used for  leak and spill cleanup and
equipment washdown.  Table 15-3 summarizes water usage in the aluminum
fluoride industry.

Waste Sources

A.   Noncontact Cooling  Water

     Noncontact cooling  water is used to cool  the product  coming  out
     of  the  reactor.   In  some  cases  it   is  recirculated and the
     blowdown treated separately from other  process contact wastewater
     or  it  is  discharged  without  treatment.   The  water  can  be
     monitored  for  fluoride  and if process  contamination occurs, it
                                   461

-------
TRBI£ 15-1.
SOBCHBSQHJf
                                         DMA SttMHBT
                     ALUKQKM PUXMOEE
Total subcategory capacity rate
Total subcategory production rat*
Nunber of plants in this subcategory
308 Data on £11* for
    With total capacity of
    Hitfa total production of
    Representing capacity
    Representing prortrrflnn
    Plant production ranges
    Average production
            capacity
    Plant age range:
    Wasts water flow range:
            Minium
    Ttaiume per unit product;
            f*i TI 1 n« ••
                                                     HA
                                               134,700 kkg/year
                                                     5*
                                                     6
                                               204,800 Idog/year
                                               120,000 kkg/year
                                                     NA
                                                     NA

                                                  1,000 kkg/year
                                                45,600 kkg/year
                                                24,300 kkg/year
                                                35,500 kkg/year
                                                    59 percent

                                                     5 years
                                                    21 years

                                                   539 cubic meters/day
                                                 2,200 cubic meters/day

                                                     5 cubic meters/kkg
                                                    12 cubic meters/kkg
Sources of data are Stanford Research Institute, Directory of Chemical
Producers, D.SJV., 1977, U.S. Department of Connerce, CLu-muL Industrial
Reports, Deceini w 1977; Energy and Eimiiuuuental Analysis, Inc.; Draft
Report, "Preliminary Goancmic AaaeMi**"1: of Effluent T.
-------
                      HATER
                                  VENT
                                  _
                          SCRUBBER

                                HASTE HATER
                  CYCLONE
HYDRATED
ALUMINA
HYDROGEN
FLUORIDE
                 REACTOR
                                       NONCONTACT
                                       COOLING WATER
                                       J	L
            COOLER
  PRODUCT
 COLLECTION
AND STORAGE
 ALUMINUM
'FLUORIDE
 PRODUCT
              Figure  15-1,
General process  flow diagram for production
  of aluminum fluoride.

-------
  TAELE 15-2  ._   STfiTOS OF REGULATIONS - EFFLUENT IjMTIMlON GUlCOJKgS	


  SUBCATEGOR*     Aluminum Fluoride

  SUEPAKT         W  (40 CFR 415.230,  5/22/75)


                                         STA.KDARDS


                               BPCTCA*            BATEA*             NSPS *

                          Max.       Avg.      Max.     Avg.      Max.       Avg.
  Product      Fara-     Xg/l&g
  Process      rretexs    (mg/D


  A1F-          Fluoride   0.68       0.34
                          (40)3       (20)

               TSS        O.S6       0.43
                          (51)         (25)

                          0.34       0.17
                          (20)         (10)
    Sections 415.230, 415.231, and 415.232 were revoked ty the Agency
    .1*1 FR 51601, November 23, 1376),

   "Vax. = Miximin of any one day,

    Avg. - !--^:'Jjttm average of daily -values for thirty consecutive days.
         basis   17,000
               15-3.  VJ?^£R USSGE IN THE AUMTNW FLUORIDE SUBCATEGQRY
       Source                   Water use per unit of production

                                       (m /Wtg of AIFjJ
Plant
i 837
Plant
I 705 U)
Plant
i 188
Plant
1 251 (2)
Tto-contact cooling         14.5         NA (1)      6.95         HA

Indirect process            12.2       wJl'lS"  f'     NA          NA
contact (pumps, sea Is,                 J*
leaks, spills)

Maintenance, e.g.            1.13       2139         NA       V'l!.02
cleaning and work area
washdown

Scrutber                     3.45      V:92        3.46       18j7
(1)   KR. = Not Available

(2)   Currently not rr-anufocturing alvminum fluoride.
                                   464

-------
     can be diverted to the wastewater treatment facility for fluoride
     removal.

B.   Floor and Equipment Washings

     The quantity and  quality  of  wastewater  generated  from  these
     operations   varies   and  depend  largely  on  the  housekeeping
     practices at the individual plants.

C.   Scrubber Wastewater

     This is the major source of wastewater requiring treatment before
     discharge or recycle to the scrubber.  It  is  contaminated  with
     hydrofluoric  acid, aluminum fluoride and aluminum oxide, and, in
     some cases, suIfuric  acid  and  silicontetrafluoride  have  been
     detected.  These originate as impurities in the hydrofluoric acid
     used in the process.  Table 15-4 presents the wastewater flows at
     different  facilities  in  the  subcategory.   Noncontact cooling
     water is excluded from consideration since it normally  does  not
     contain pollutants.

D.   Solid Wastes

     In the aluminum fluoride production, hydrofluoric  acid  gas  and
     solids, such as aluminum trihydrate and aluminum fluoride, escape
     with  the vent gases.  During scrubbing, the solids are suspended
     in the scrubber water, while hydrofluoric acid gas is  dissolved.
     In  the  treatment  facility,  the wastewater is neutralized with
     lime and calcium fluoride precipitates out and settles with other
     suspended solids.   In the  majority  of  cases,  the  solids  are
     retained  in  a  lagoon  for periods up to ten years.  Table 15-5
     gives a summary  of  the  amounts  of  solids  generated  at  two
     aluminum fluoride plants.

     Different   wastes   from   the  aluminum  fluoride  process  are
     intermixed before   treatment.   As  mentioned  earlier,  scrubber
     water  constitutes  the major source of wastewater  in the aluminum
     fluoride subcategory.  If the production of aluminum fluoride  is
     integrated with hydrofluoric acid, then the wastewaters from both
     plants are combined and treated.

Description of Plants Visited and Sampled

Screening

Plant  #705  was  visited in the screening phase of the program.  Both
hydrofluoric acid and aluminun fluoride are produced at this  facility
by  the  general processes described earlier.  The wastewater from the
hydrofluoric acid and aluminum fluoride plants is mixed  and  sent  to
the  treatment  facility.   At  the  treatment  facility  the combined
wastewater  is neutralized with lime and sent to a series  of  settling
ponds.  The effluent from the last pond is given a final pH adjustment
before  a  portion  is discharged and the rest recycled  to the process.
                                   465

-------
        TABLE 15-4.  WASTE WRTER FLOW AT PLANTS #837, #705 AND #251
                     FOR ALUMINUM FLUORIDE SUBCATEJGORy
        Source
                                       Flow rate per unit of production

                                               ( ra3/kkg of A1F3)
                                                                       (1)
Scrubber water

Maintenance equipment
cleaning and work area
washdown

Total raw waste flew
Plant #837

   3.45

   1.13



   4.58
                                                     Plant #705(4)  Plant #251(4)
                                                        &•&
w^
                                                                      19.7
Average of above
three flows
                                                       11.9
(1)  All flow information is from 308 Questionnaires and plant visits.  Unit
     flew is calculated ty dividing waste water flow in mVday by production
     in kkg/day.

(2)  From Table 15-6 (see footnotes which describe basis of information).

(3)  From Table 15-7 (see footnotes which describe basis of information).

(4)  Currently not manufacturing aluminum fluoride.
       TABLE 15-5.  SOLIDS GENERATED AT PLANT #705 AND #251 PRODUCING
                             AUKENUM FLUORIDE
        Plant
                                    Total Solids Generated(kg/fckg of AlF3)
        #705

        #251
            (1)

            (1)
               54

               69
(1)  Currently not manufacturing aluminum fluoride.
                                    466

-------
Figure 15-2 shows a simplified block diagram of the process  including
the  wastewater treatment facility and sampling locations.   Table 15-6
presents a summary of flow data of the sampled streams, and  the  data
for important classical pollutant parameters.

Verification

Plant  #705  was  visited  again  and  the same streams sampled in the
screening phase also were sampled and  analyzed  in  the  verification
phase.   The  variations  in individual stream flows were small during
the two phases of sampling.   Table 15-6 summarizes the flow  data  and
important  conventional  and  nonconventional  pollutant emissions.  A
second plant {Plant #251) was visited and sampled in the  verification
phase.   Figure  15-3  is  a  simplified  flow diagram of the aluminum
fluoride manufacturing plant and  the  wastewater  treatment  facility
showing  the  sampling  locations.   Table  15-7 presents the flow and
pollution concentration data for the plant.  The aluminum fluoride and
hydrofluoric acid waste streams are combined and sent to a gypsum pond
for suspended solids removal.  The overflow from  the  pond  is  mixed
with  alkaline  and  acid streams from other plants for neutralization
and pH adjustment before final discharge.

Summary of the Toxic Pollutant Data

Following is a list of toxic pollutants which identifies their maximum
concentration levels as found in the raw process waste streams sampled
during screening and verification.

               Maximum Raw Waste Concentrations Observed
          Pollutant        Screening            Verification
     _ Plant#7Q5 _ Plant #705 and #251

     Arsenic                  200                   480
     Selenium                  68                    97
     Chromium                  70                  1100
     Copper                   1 20                   250
     Lead                      25                    91
     Mercury                     1.6                  11
     Nickel                   150                   290
     Zinc                     450                   450
     Cadmium                     0.70                 33
     Antimony                    0                     3.0
     Beryllium                   0.80                  0.80

Section 5 of this report describes the methodology  of  the  screening
and verification sampling program.  In the aluminum fluoride industry,
seven  days of sampling were conducted at Plants #705 and #251 .  Seven
sampling points were  identified  and studied for the subcategory.   The
evaluation  of  toxic pollutant  content of these process-related waste
streams was based on  637 analytical data points.   The  screening  for
toxic  organic  pollutants  at   Plant #705 generated an additional 645
                                   467

-------
                                                            HMO*
CO
,



FIUORSPAR
	 [»»
	 h»
H2S04 «—










I





— 	 	 — •• ' -1 ' — i vtwr
1 f
	 | 	 1




WVVQ4

DRIP ACID 'P ^ PflOOUCT
HP
i ' " 	 "

TO MHOSniERB 1 1
4 ^wmn f
_ * — Allan,


4" i
A1P3 WIDOUCT-^ 	 : OOOUS
i i
. 	 ^^ ,i SLWKB DRAIK9
[ ^^ " COCCItC TOWH* BUWOHH, BIC.
' MBffMUWnai HEPHJW HW -» fiETlUIB HID ^fc "**"* i*1 £^-^> TUF
*"* TANK ^ BBraaw. KW -• M.TIWI*. i«« » MJuaneff ^J| »* JJJ


                                                                                      Huta atroana wnupled.
               Figure 15- 2.  General process flew diagram at Plant 1705 showing the sanpling points
                                         (aluminum fluoride manufacture).

-------
                           TABLE 15-6.  FLOW AND POIlilT^^CONCEETORATiai DATA OF THE SAMPLED
                               WASTE STREAMS FOR PLAN^I^O^PODUCI^ ALUMINUM FLUORIDE
Sampling
Phase


Screening







Verifica-
tion Sanpling





Sampled
Stream
No.

3
4
-------
-J
O
      VENT
                            1
                           DUST
                        COLLECTOR
              WET
              SPAR"
   SPAR DRYING
HANDLING
 LOSSES
              HOSE  DOWN
                 WATER
                r
HF KILN
                                                 WASTE
                                    AIR
                    DRIP
                    ACID

                  •WATER
                                                    SLURRY
                                                   TRANSFER
                         LEGEND
                      SAMPLING POINTS.
                                                                             PLANT
                                                                            REACTOR
                                                         T
                                                                           AIF. PRODUCT
                                                                   C
                                                                                                 VENT
                                                                                                 _L
                                                                                              S02 SCRUBBER
                                                                                               WATER
                                                   TUFj
                                                  SCRUBBER
                                                                              LIQUEFACTION
    AHF
PURIFICATION
                JAH
               PRC
 AHF
PRODUCT
                                                 DILUTION
                                                  WATER
                                              A1F3 PLANT

                                               HOSE DOWN
                                                                               HOSE DOWN WATER
                                                                                 AHF PLANT
                                                        #2
                                                                 GYPSUM
                                                                  POND
                                                      13
                                                               NEUTRALIZATION
                                                                   SYSTEM
                                                                                    h
                                                                                    WATER
                                                                           LA    JL*6
                                                                           Lt
                                                                               EFFLUENT
                                                                               TO RIVER
                                                                      ALKALINE STREAMS
                                                                      AND ACID FROM OTHER PLANTS
                   Figure 15-3.   General process flow diagram at Plant #251 showing the sampling points.
                                (aluminum fluoride manufacture).

-------
                      TAE&E 15-7.  FI£W AND POmZTAJJT QOtOMKATION DATA OP THE SAMPLED STREAMS
                                     FOR PLANTff251 PRODUCING AUMCMJM FLUORIDE
Stream
No.
Sampled
Stream
Description
Unit
Flow
(m3Akg
of AlF3)
Verification
Sailing
4
6
4&6

2
3
AlF- scrubber
water
SO™ scrubber
water W
Total raw waste
load
Gypsum pond
influent*2'
Gypsum pond
effluent*25
12.6
6.10
18.7

25.1
25.1
•total
Suspended
Solids
(mg/1) (kg/kkg)


1200
0.0
1200

19,000
9.0


16
0.0
16

470
0.23
Fluoride
(mg/1) (kg/kkg)


470 5.90
20 0.14
320 6,0

660 17
320 8.0
Aluminum
(mg/1) (kg/kkg)


50 0.60
0.20 0.0010
50 0.60

26 0.65
22 0.55
#•>
           (1)   One half flow of SO2 scrubber water is assumed to contribute to the A1F3 process since the
                total flow is connon to the A1F3 and HP process.

           (2)   Consists of hydrofluoric acid and aluminum fluoride waste water.  Plant currently not
                manufacturing AlF.,.

-------
analytical data points.  The daily raw  waste  loads  were  calculated
from  the waste stream flow rates measured or estimated at the time of
sampling and the measured pollutant concentration.

That is,
     Daily loading {as kg of polutant per day) = (C) (Q)
                                                   1000

where:  C is the concentration  of  the  pollutant  expressed  as  mg/1
     (Note: kg/m3 = 1000 mg/1), and Q is the Aluminum Fluoride process
        waste  stream  flow  rate  expressed  as m3/day.  (m3, a cubic
     meter, is equal to 264.2 U.S.  gallons)

Similarly,  the  unit  loadings  were  calculated  from  the  reported
aluminum fluoride production rate, the waste stream flow rate, and the
measured pollutant concentration.

     Unit loading (as kg of pollutant
     per kkg of aluminum fluoride)         * (C) (Q)
                                              1000P

     where  C  and  Q  are  the  same as described above, and P is the
     aluminum fluoride production rate expressed as kkg/day.  (kkg  is
     1000 kg, a metric ton, which is equal to 2205 Ibs.)

     The  P  and  Q  factors are for the Aluminum Fluoride Process and
     thereby the Agency has segregated that portion  of  the  effluent
     attributable only to the Aluminum Fluoride Process.

Tables  15-8  and  15-9  are  tabulations of the raw waste and treated
toxic pollutant concentrations and loads determined during  the  three
plant  visits.   The loads and concentrations are based on the average
of three composite  samples  during  verification  and  one  composite
sampling  during  screening.   These unit loads were used to determine
the minimum, average, and maximum unit  loading  values  presented  in
Table 15-10.

Based  on  the  total  annual  production  of 134,700 kkg/year in this
subcategory and the average waste load generated per unit  product  in
Table  15-10,  the  estimated  total  toxic  pollutant raw waste loads
generated each year for this subcategory are as follows:
                                   472

-------
            TABLE 15-8.  TOXIC POLLUTMSTr AVERAGE RAW WASTE LOADS AND CONCENTRATIONS
SUBCATEGORY
ALUMINUM FLUORIDE
Pollutant
Arsenic
Selenium
Chromium
Copper
Lead
Mercury
Nickel
zinc
Cadmium
Antiitony
Beryllium
Screening
Plant #705
(11 (21
(mg/1) IJJ (kg/kkg) v"
0.18 0^0020
0.050 0.0010
0.030 0.00030
0.10 0.0010
0.0050 0.00010
0.00040 0.0000040
0.11 0.0010
0.16 0.0020
0,00020 0.0000020
0.00020 0.0000020
Verification
Plant
(mg/1)
0.18
— t3*'
0.44
0.070
0.020
0.00040
0.22
0.080
0,010
, 0.00040**
#705
(kg/kkg) "
0.0020
— <3>,
0.0050
0.0010
0.00020
0.0000050
0.0030
0.0010
0.00020
0.0000050
Plant
i '
(mg/1)
0.020
0.050
— <3>
0.010
0.010
0.0030
0.010
0.020
__<3)
__<3)
#251
(kg/kkg)
0.00030
0.0010
— (3)
0.00010
0.00010
0.000050
0.00020
O.OQD30
_(3)
•':",'
Average ^v^ •
Concentration i.oac(?
(mg/1) -V
0.13
0.050
0.24
0.060
0.012
0.0013
0.11
0.090
0.0050
0.00040
0.00020
(1)   Concentrations based on average raw waste loads shown and total process production and waste
     flows.

(2)   kg/kkg of product.

(3)   —  belcw analytical detection limit.

-------
    TABLE 15-9.  TOXIC POLLUTANT EFFLUENT CONCENTRATIONS DURING SAMPLING
SUBCATEGORY
ALUMINUM FLUORIDE
Pollutant

Arsenic
Selenium
Chromium
Copper
Lead
Mercury
Nickel
Zinc
Cadmium
Antimony
Beryllium
Plant and Sampling Phase
#705
Screening
(mg/1)
ND(1)
ND
0.0070
0.10
0.0020
ND
0.050
0.0020
0.0020
ND
0.0020
#705
Verification
(mg/1)
ND
ND
0.040
0.0010
0.020
ND
ND
0.0010
0.0010
ND
ND
#251
Verification
(mg/D
0.0050
0.070
0.22
0.070
0.030
ND
0.050
ND
ND
ND
ND

Average
(mg/1)
< 0.0050
< 0.070
0.090
0.060
0.020
ND
0.050
0.0020
< 0.0020
ND
< 0.0020
(1)   ND — Not Detected
                                     474

-------
  TABLE 15-10.  SUMMARY OF RAW WASTE LOADINGS FOUND IN SCREENING AND VERIFICATION SAMPLING
SUBCATEGORY
ALUMINUM FLUORIDE
Pollutant
Toxic
Arsenic
Selenium
Chromium
Copper
Lead
Mercury
Nickel
Zinc
Cadmium
Antimony
Beryllium
Conventional and
Nonconventional
TSS
Fluorine
Aluminum
loading Range,
kg/day
Minimum Maximum
0.050
0.030
.0.020
0.020
0.0030
0.026
0.026
0.040
0.00010
NA<2>
NA
600
250
100
0.080
0.16
0.22
0.050
0.020
0.0080
0,12
0.080
0.0070
0.00020
0.00010
5400
980
320
Minimum
0.00030
0.0010
0.00030
0.00010
0.00010
0.0000040
0.00020
0.00030
0.0000020
NA
NA
13
5.5
0.60
Unit loading.
Average1 '
0,0013
0.0010
0.0030
0.00070
0.00015
0.000020
0.0013
0.0010
0.000080
o.oooooso'"
0.0000020
50
8.1
3.9
kg/kkg
Maximum
0.0020
0.0010
0.0050
0.0010
0.00020
0.000050
0.0030
0.0020
0.00020
NA
NA
119.0
13.0
7.0
No. of
Plants
Averaged
3
2
2
3
3
3
3
3
2
1
1
3
3
3
(1)   Average of unit loadings from Table 15-8,
(2)   Not Applicable

-------
     Pollutant	Waste Load (kg/year)
Arsenic
Selenium
Chromium
Copper
Lead
Mercury
Nickel
Zinc
Cadmium
Antimony
Beryllium
180
140
400
94
20
3.
180
140
11
0.
0.





0



70
30
Pollution Abatement Options

Toxic Pollutants of Concern

The toxic pollutants found  in  actual  plant  wastewaters  are  lead,
mercury,  cadmium,  antimony,  beryllium,  copper,  arsenic, chromium,
nickel, zinc, and selenium.  In the case of selenium, it  is  apparent
that the source was the raw water supply and is therefore not regarded
as a process-related pollutant, but control of selenium in the treated
effluent  may be required in some cases.  The toxic pollutants of most
concern are chromium and nickel.

Copper and  chromium  may  be  present  as  trace  impurities  in  the
hydrofluoric  acid  used  to  react  with  bauxite  to  form  aluminum
fluoride.  Arsenic, zinc, and nickel may originate  as  impurities  in
the  bauxite  ore.   Waste  treatment  processes should be designed to
control TSS,  fluoride,  and  the  significant  toxic  metals.   Lead,
mercury,  cadmium,  antimony,  and  beryllium  are eliminated as toxic
pollutants of concern because  levels  observed  are  too  low  to  be
considered treatable.

Process Modification and Technology Transfer Options

A.   Total recycle of wastewater to the scrubbers is feasible if final
     neutralization is with soda ash.  The calcium   in  the  waste  is
     precipitated  in  the  treatment  system as calcium carbonate and
     therefore scaling problems in pipes and scrubbers are reduced.

B.   Passage of the vent gases from  the  reactor  through  a  cyclone
     prior  to scrubbing with water will remove the aluminum oxide and
     aluminum fluoride particulates.  The collected  material  in  the
     cyclone  can  be  recycled to the reactor.  The installation of  a
     cyclone will result in material recovery and will also reduce the
     suspended solids load going to the wastewater treatment facility.
                                    476

-------
Best Management Practices

A.   Rainfall runoff in plant areas, treatment  facilities  and  other
     places susceptible to fluoride contamination can be collected and
     sent to the wastewater treatment facility.

B.   If solid wastes containing fluoride are stored on  land,  studies
     should be conducted to ascertain the risk of contaminating ground
     water.   Where  necessary,  provisions can be made for collection
     and treatment of leachate, permeate, and runoff.

C.   Settling ponds in the wastewater  treatment  facility  should  be
     deep  enough  (or  provided  with baffles) to eliminate or reduce
     turbulence caused by wind and rainfall.   This  will  reduce  the
     incidence  of  weather-related plant upsets, and suspended solids
     limitations will be met more consistently.

Prevailing Control and Treatment Practices

Plant  #705  practices  lime  neutralization  and  settling   of   the
wastewaters.   Since  aluminum  fluoride production is  integrated with
hydrofluoric acid production, the wastewaters from the  two  processes
are  combined  before  treatment.  The plant does not treat noncontact
cooling water.

At Plant ftB37 the tail gases are scrubbed with soda ash solution,  and
the  resulting  solution  is sent to an adjacent facility  for use.  The
water  from  the wet scrubbers on the hydrated alumina  dryers   is  also
sent   to  an  adjacent  facility  for  use.  The wastewaters from area
washdown are combined with  other  product  wastewater,   treated  with
hydrated lime and sent to a settling lagoon before discharge.

Plant  #188  produces  aluminum  fluoride   in  small quantities and  in
batches.  The wastewater  from  the batch  operation is first sent  to   a
collection  pond.   It  then goes to a second  pond where  lime  and alum
are added and it finally  enters a third  pond where the  pH is   adjusted
by recarbonation.

Plant  #251  mixes  the aluminum fluoride waste with hydrofluoric acid
plant  waste.  The combined wastewater is sent  to  gypsum  ponds  for
suspended solids removal.  The supernatant  is  treated with an  effluent
from   another plant for pH control  and neutralization.  Because of the
presence of complex fluorides  (from  the HF   process)   in   the  waste
waters,  the plant is planning to use a  new proprietary process in the
near future to further reduce  fluoride levels  in the final effluent.

Advanced Treatment Technologies

Metal  ions  can be precipitated as hydroxides at  alkaline pH   levels,
and  in  clarified  solutions  they  may be exchanged  for hydrogen  or
sodium ions by ion exchange.   Metal  ions at low  levels  may   also   be
controlled  by  xanthate  precipitation,  although   the process is not
widely used.  Sulfide precipitation will reduce  copper,   nickel,  and
                                   477

-------
zinc to low levels but will not control chromium or arsenic.  Although
the mechanism is not clear, arsenic levels appear to be reduced in the
lime  neutralization  process  followed  at  most  plants,  perhaps by
entrapment or adsorption of the  oxide  during  the  precipitation  of
calcium   fluoride.    A   combination  of  lime  and  ferric  sulfate
coagulation is probably the most effective and  practical  method  for
reducing arsenic concentrations.

Selection of Appropriate Technology and Equipment

Technologies for Different Treatment Levels

A.   Level 1 (BPT, BAT and NSPS)

     Neutralization with lime is widely used in the industry to remove
     the  primary  nonconventional  pollutant  as  calcium   fluoride.
     Because  lime  neutralization  to  pH  10  results in significant
     incidental removal of toxic  pollutants,  alkaline  precipitation
     was  chosen as the Level 1 technology.  The flow diagram is shown
     in Figure 15-4.

B.   Level 2

     A higher removal of suspended metal hydroxides, TSS, and CaF2 can
     be achieved by  adding  dual-media  filtration  to  the  Level   1
     system.  The flow diagram is shown in Figure 15-5.

C.   Level 3

     Sulfide precipitation is added to the Level 2 treatment to attain
     a higher level of heavy metal  removal.   Chromium  and  selenium
     levels   are   not   appreciably  reduced  although  other  toxic
     pollutants levels are reduced.  The  flow  diagram  is  shown  in
     Figure 15-6.

D.   Level 4

     The technology is similar to Level 2, except  that  soda  ash  is
     substituted  for  part  of the lime treatment, permitting partial
     recycling  of  effluent.   Eighty  percent   recycle   has   been
     demonstrated  and is used in the development of plant performance
     estimates.  The flow diagram is shown in Figure 15-7.

     A  detailed  cost  comparison  and  performance   evaluation   of
     alternative  levels of treatment were utilized in the development
     of the proposed regulations and were presented  in  the  proposed
     Development Document  (60).
                                   478

-------
/""^L
1 LDC
U-»Q>


'


RAH



QfH JID1




a
TN 1 L.^
WfiBTP VfWFF! """ x *
BQUAUZM1CN
^«
1
1
[
,t'1 to




— **- 	 tai
^ t



I '
*
I ^
HOONS









r-A. MOOCN A»,
L J
^


.


-A "DOM JL*

L _
©
Tr
Vi




                                                                                             EEFUIBfr
         •Includes flow mnltoring, pH nnitoring «nd «n|>ler.
Figure 15- 4.   Level 1 waste water treatment for the aluminum fluoride  subcategory.

-------
                                                                          UOWflH
CD
O
^





IftSTE

1




warn
— -(g>
I
1
i J



/
-** 	 A-**
BQUftLUKnCM 1
Lk^Vi— ^
._
— fl»
r*.





1





1
1
•jj
i!
J£

1


P^X, UOOdf A^l
1

• *
LI
UGOCH /-f»

i


Bi
#


















1 ^!

•— Q*-*1







X
PXUIER

f



1


\HNxMnwwr
/w\
{&•}
IT
	 ' 	 H

                   *ZnoludM flow Monitoring, ftt ROnltorlng and Hn|)ler.
                     Figure 15-5.  Level 2 waste  water treatment for the aluminum fluoride subcategory.

-------
                                                                      Qa
CO
               KMf

              WBTB—

              mmt
^   1   Z
                                    L*o^
                                                                                             T JL
                                                                                             I  w

                                                                                               " I
                                                                                                I
                                                                                suv          rtUBi
                                                                                               -4p~
                                *lnctude« now monitoring, pH mpnllortn| end
                 Figure 15-6.   Level 3 waste water  treatment for the aluminum fluoride subcategory.

-------
Equipment for Different Treatment Levels

A.   Equipment Functions

     Level 1  consists of  flow  equalization  with  first  stage  lime
     application  followed by second stage lime application and lagoon
     settling.  The final pH is adjusted with hydrochloric acid to the
     6-9 range before discharge through an effluent monitoring system.

     In Level 2, dual-media filtration  is  added  to  provide  better
     control   of  suspended  solids, including heavy metal hydroxides,
     which are returned to the lagoons as filter backwash.

     In Level 3, ferrous sulfide is  prepared  on  site  from  ferrous
     sulfate and sodium bisulfide and is added ahead of the dual-media
     filter shown in Level 2, to reduce heavy metals (except chromium)
     to lower levels by sulfide precipitation.

     Level  4  is  a  modification  of  Level  2  which allows partial
     recycling of final effluent by substituting soda ash for part  of
     the  lime treatment, and settling the resulting calcium carbonate
     in a clarifier before filtration.  This step reduces the  calcium
     saturation  and  permits  recycling  of  effluent without serious
     scaling problems.  Although  a  small  blowdown  of  effluent  is
     maintained  for  control  of salinity the total mass discharge of
     toxic pollutants is less than that achieved in Level 2 due to the
     lower effluent flow rate.

B.   Chemicals and Handling

     In Level 1 (BPT) and in  Level  2,  two-stage  neutralization  is
     accomplished   with   lime  alone,  using  conventional  handling
     equipment to deliver milk of lime to two points  of  application.
     In  Level 3, a mixture of ferrous sulfat.e and sodium bisulfide is
     prepared  in  a  well-ventilated  space  and   applied   with    a
     conventional  solution  feeder  to  the  inlet  of  the  Level   2
     dual-media filter.   With  adequate  ventilation  and  proper  pH
     control  in  this  chemical  preparation,  there  are  no unusual
     problems in chemical handling.  In Level 4, soda ash is  used  to
     furnish  part  of  the  alkalinity,  employing  conventional  dry
     chemical feeding equipment for this nonhazardous chemical.

C.   Separation and Removal of Solids

     At all levels of treatment the precipitated  solids  are  removed
     mechanically  from the lagoons at regular intervals and are piled
     in self-draining areas near the lagoons, on land provided  for   a
     ten-year  operating period.  Fluoride and toxic pollutants are in
     the insoluble or adsorbed form and do not constitute a hazard  to
     the   local  environment  when  left  at  the  plant  site  under
     controlled conditions, i.e., with leachate and permeate control.

D.   Monitoring Requirements
                                   482

-------
00
U)
                 RNf
                 WSTE-
                 wren
                                                     r
                             HMUUnoi  1
                                       (low monllotlng, pll monltorlnc and

~l

w
                                                                                                  nun    •
                                                                                                           •mum
                                                                          -Q	'
                   Figure  15-7.  Level 4 waste water treatment  for the aluminum fluoride subcategory.

-------
     Control of fluoride and toxic pollutants in the treatment process
     can be reasonably assured by pH and fluoride  ion  field  testing
     equipment.   At  advanced  levels very low values of toxic metals
     are  detected  best  by  atomic  absorption   methods,    normally
     performed  in  commercial laboratories on carefully collected and
     composited samples,

Treatment Cost Estimates

General Discussion

A model plant concept was developed for the subcategory for   treatment
cost estimation purposes.  The BPT treatment system specifications are
outlined below.

A.   Wastewater Flow

     The range of wastewater data on file shows flow  variations  from
     4.58  mVkkg  of  A1F3  to  19.7 mVkkg of A1F3 (see Table 15-4).
     Based on these values, a unit flow of 11.9  mVkkg  of   A.1F3  was
     taken as the average for the wastewater treatment model plant for
     cost estimating purposes.

B.   Production

     Five plants manufacture aluminum fluoride at a  total  production
     rate  of 120,000 kkg/yr.  Individual plant production rates range
     from a minimum of 38 kkg/yr to a maximum of 45,600 kkg/yr with an
     average of 24,300 and a median of 35,500 kkg/yr.  For  wastewater
     treatment  cost  estimates, three production levels were selected
     as model plants.   These  three  models  reflect  the  production
     levels of the plants for which data is on file (excluding a small
     batch  operation  plant)  and are 15,900 kkg/yr, 35,600 kg/yr and
     45,800 kkg/yr.

C.   Pollutant Loadings

     Observed pollutant loadings varied from 14 to 27 kg/kkg  of  AlF3
     for  suspended  solids  and  from  5.4 to 39.5 kg/kkg of A1F3 for
     fluoride.  The data sources are as follows:

     Source of Data	TSS (kq/kkq-AlF-.)	F (kg/kkq-AlF, )

     EPA-440/1-75-037            16-20                    15-20
     Screening and
     Verification Phase
     -Plant Data                 14-27                    5.4-40

     For model plants, pollutant loadings of 20 kg of total  suspended
     solids  and   18  kg  of  fluoride  per  kkg  of A1F3 were used to
     establish treatment requirements.

D.   Treatment Chemicals
                                   484

-------
     Lime (CaO powder form)  is added to precipitate  fluoride  and  to
     raise  the  pH  to  a  six  to nine range.   For each of the model
     plants,  lime is added  at  25  percent  above  the  stoichimetric
     requirements for fluoride precipitation.  For advanced treatment,
     ferrous sulfide is added to give a concentration of 10 ppm.  This
     acts  as  a polishing step to remove additional trace metals from
     the effluent.  For a more advanced level of treatment,  soda  ash
     is  added  in  addition  to  lime (CaO).  The soda ash dosage was
     assumed to be 770 kg/kkg.

E.   Generation of Solids

     From the pollutant loadings and treatment  chemicals  above,  the
     waste treatment residue consists of 20 kg/kkg of suspended solids
     plus  46.2  kg/kkg  from added chemicals.  Thus, the total solids
     generated are 66.2 kg/kkg of product.  After  mechanical  removal
     to  self-draining  piles,  the  combined  fluoride  (as  CaF2) is
     reasonably  stable  at  the  reaction  pH  reached  during   lime
     treatment.

F.   Cost Estimates

     The  estimated  treatment  costs  (BPT)   for   three   different
     production  models  are  given in Tables 15-11, 15-12, and 15-13.
     For these models, both the hydraulic and pollution  loadings  per
     unit  of  production  are  held constant over the entire range of
     production.  A summary of the annual treatment costs  is presented
     in Table  15-14.  At this level of  treatment,  chemicals,  labor,
     and amortization have significant impact on the annual costs.

Basis For Regulations

Evaluation of  BPT Treatment Practices

EPA is setting BPT limitations based on Level 1 treatment.  All plants
in this subcategory have installed BPT technology.  Pollutants  limited
by  the  BPT   regulations are TSS, fluoride,  chromium, nickel,  and pH.
The major pollutants  previously  regulated   are  TSS,  fluoride,  and
aluminum.  Aluminum is no longer  considered a pollutant of concern due
to  its  relatively  nontoxic  nature.   The  treatment selected as the
basis for BPT  regulations will actually benefit from the  presence  of
aluminum  which  will precipitate under mildly alkaline conditions and
act as a coagulant to aid the removal of toxic  metals  and  suspended
solids.

BPT Effluent Limitations

A.   Technology Basis

     The Agency is setting BPT limitations for  which  the  technology
     basis     is,    or   is   equivalent   to,   equalization,   lime
     neutralization/alkaline precipitation, solids removal by settling
     or  thickening,  final  pH   adjustment,  and  discharge  of   the
                                     485

-------
   TABLE  15-11.  MODEL  PLANT TREATMENT  COSTS
Subcategory    Aluminum Fluoride
Production     15,900 metric  tons per  year
                                               (S)
A, INVESTMENT COST                             BPT

     Site development  	           30,300
     Equipment	          198, 000
     Monitoring equipment  	           20,000

          Subtotal 	          248/300
     Contractor's 0 & P a.	           37,245

          Subtotal 	          285,545
     Engineering 	           57,109

          Subtotal 	          342,654
     Contingencies	           34, 265

          Subtotal 	          376,919
     Land	           24,000

     TOTAL INVESTMENT COST 	          400,919
B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision 	           56,000
     Energy 	            3,500
     Chemicals 	           35,000
     Maintenance 	           37,692
     Taxes and insurance 	           12,028
     Residual waste disposal  ....            5,500
     Monitoring, analysis
      and reporting 	           15,000

     TOTAL OPERATION AND
     MAINTENANCE COST                      164,720
C. AMORTIZATION OF
   INVESTMENT COST                          61,325

   TOTAL ANNUAL COST                       226,044

  Overhead and Profit
                             486

-------
   TABLE 15-12.  MODEL PLANT TREATMENT COSTS
Subcategory    Aluminum Fluoride
Production     35,500 metric tons per year
                                              ($)
A. INVESTMENT COST                            BPT

     Site development 	          45,900
     Equipment 	         245,000
     Monitoring equipment  	          20,000

          Subtotal 	         310,900
     Contractor's 0 & Pa	          46,635

          Subtotal 	         357,535
     Engineering  	          71,507

          Subtotal 	         429,042
     Contingencies 	*...          42,904

          Subtotal 	         471,946
     Land	          42,000

     TOTAL INVESTMENT COST  	         513,946
B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  	           56,000
     Energy 	            5,500
     Chemicals 	           80,000
     Maintenance  	           47,195
     Taxes and insurance  	           15,418
     Residual waste disposal  ....           12,500
     Monitoring, analysis
      and reporting 	           15,000

     TOTAL OPERATION AND
     MAINTENANCE COST                      231,613
C. AMORTIZATION OF
   INVESTMENT COST                          76,786

   TOTAL ANNUAL COST                       308,399
 Overhead and Profit
                              487

-------
   TABLE 15-13.  MODEL PLANT TREATMENT COSTS
Subcategory    Aluminum Fluoride
Production     45,800 metric tons per year
                                               ($)
A. INVESTMENT COST                             BPT

     Site development 	           53,300
     Equipment 	          283,000
     Monitoring equipment  	           20,000

          Subtotal 	          356,300
     Contractor's 0 & P a.	           53,445

          Subtotal 	          409,745
     Engineering  	           81,949

          Subtotal 	          491,694
     Contingencies 	           49,169

          Subtotal 	          540,863
     Land	           60,000

     TOTAL INVESTMENT COST	          600,863
B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  	           56,000
     Energy 	            7,500
     Chemicals 	          100,000
     Maintenance  	           54,086
     Taxes and insurance  	           18,026
     Residual waste disposal  ....           16,000
     Monitoring ,  analysis
      and reporting 	           15,000

     TOTAL OPERATION AND
     MAINTENANCE  COST                      266,612
C. AMORTIZATION OF
   INVESTMENT COST                          87,998

   TOTAL ANNUAL COST                       354,611
  Overhead and Profit
                             4SS

-------
     TABLE 15-14.MODEL PLANT UNIT TREATMENT COSTS
  Subcategory Aluminum Fluoride
                             Annual Treatment  Costs  ($/kkg)
 COST ITEM
PRODUCTION
 (kkg/yr)
LEVEL OF TREATMENT

        BPT
Annual Operation
and Maintenance   15,900
                  35,600
                  45,800
Annual
Amortization
Total Annual
Cost
  15,900
  35,600
  45,800
  15,900
  35,600
  45,800
                         10.36
                          6.51
                          5.82
        3.86
        2.16
        1.92
       14.22
        8.66
        7.74
                          489

-------
     clarified  effluent.  This technology represents current practice
     in the Aluminum Fluoride industry and was therefore  selected  as
     the basis for the BPT effluent limitations.

B.   Flow Basis

     The basis of flow for BPT  limitations  is  estimated  from  data
     provided in the 308-Questionnaires for three of the four complete
     plant  responses received, including Plant #837, #251, and # 705.
     Plant #188 was omitted in view of the batch process utilized  for
     the manufacture of aluminum fluoride.  The other three plants are
     continuous manufacturing processes.

     The  two major raw process wastewater sources contributing to the
     total  plant  flow  estimates  include  scrubber  and  work  area
     washdown.   These wastewater sources are summarized in Table 15-4
     for the three plants considered.  The model plant  flow  for  the
     AlFj  industry  is  estimated as the average total raw wastewater
     flow for the three plants, and  is  used  to  estimate  pollutant
     discharge loadings for the purpose of regulation.  Exact measures
     of  treated  effluent from the aluminum fluoride industry are not
     available, since  aluminum  fluoride  plants  normally  integrate
     process  waste  streams  with those generated by the hydrofluoric
     acid process prior to treatment and  discharge.   The  unit  flow
     rates  varied  widely  for  the three plants in the range between
     4.58 to 19.7 mVkkg of product which is largely dependant on  the
     scrubber design and water utilization.  The A1F3 process in Plant
     # 251 shares an S02 scrubber with the anhydrous hydrofluoric acid
     process.   Wastewater  generation from this combined use scrubber
     was estimated on the basis of hydrofluoric  acid  utilization  in
     the two processes.

     The cleaning and work area washdown flow is similar for the three
     plants  considered,  ranging  between  }.02  and  2.39  mVkkg of
     product.

     The average total flow for the three plants  is  11.9  mVkkg  of
     product.   This  flow is used for the model plant in the aluminum
     fluoride subcategory.

C.   Selection of Pollutants to be RegulatecJ

     The selection of pollutants for which specific numerical effluent
     limitations are set was based on an evsiuation of raw waste  data
     from  the screening and verification sampling program.  Pollutant
     data  from  the  plant  sampled  during  screening  was  used  to
     determine  the  need  for  verification  sampling.   Verification
     sampling of Plants £705 and $251  provided  additional  pollutant
     raw  waste  concentration  data needed to assess the magnitude of
     the pollution potential.

     For conventional pollutants, the Agency has selected pH and total
     suspended solids for specific treatment  and  control.   Fluorid'
                                     490

-------
     was  selected  as   the  only   nonconventional  pollutant  parameter
     because it  is a major  constituent  in  the  process  raw  waste and is
     a pollutant of concern to the Agency.   A  limitation   on   aluminum
     is not set  because this constituent of  the process wastes will be
     effectively controlled by treatment required for  removal of toxic
     metals.

     Results  of the-screening and verification sampling are  tabulated
     in Section  15 for  the  raw process  waste  stream.   The  pollutant
     concentration  listed   under   verification  is  the highest value
     observed during  sampling  at  the  two  plants  visited.   Toxic
     pollutants   are  lifted based on their  presence,  during  sampling,
     at detectable concentration levels.   Pollutants  from  this  list
     were considered candidates for regulation if their concentrations
     appeared   at  least  once  at  approximately  the   lowest  level
     estimated as treatable using  any available technology appropriate
     for their removal.  The only  two metals which  passed  this  test
     were  chromium and nickel.  The metals  arsenic, copper,  selenium,
     and zinc were never observed  in the  raw waste  at  concentrations
     equal  to  or  above  the  lowest  level estimated as  treatable as
     presented in Table 8-11 and therefore are not regulated.

     Specific numerical  effluent   loading  limitations   are  set  for
     chromium  and  nickel   for which the  average concentration levels
     (Table 15-8) are considered treatable  for  at  least  one  plant
     visited during sampling.

     No limitation is being set for aluminum because of  its relatively
     low  toxicity and its  beneficial effects in removing  toxic metals
     by coprecipitation.  In addition,   control  of  the   major  toxic
     metal  ions  should provide   adequate  control  of   the aluminum
     concentration, since the treatment pH for BPT is  in   the  region
     considered  optimal  for  alkaline  precipitation  of  most metal
     hydroxides.

D.    Basis of Pollutant Limitations

     1.   Conventional and  Nonconventional Parameters

          a.   pH

               The treated  effluent is to  be  controlled   within  the
               range  of  6.0 to 9.0.  This limitation is  based on the
               data  presented  in  Appendix   B   of   the   proposed
               Development  Document (60)  and the JRB Study (52).

          b.   TSS and Fluoride

               Pollutant limitations for TSS and fluoride   were  based
               on  the  evaluation  of  data for the Hydrofluoric Acid
               subcategory.  This evaluation is described  in  Section
               12 under "Basis of  Pollutant Limitations."   There is no
               data  available  from  plants  where  the  BPT treatment
                                    491

-------
performance can be evaluated for the treatment  of  raw
aluminum   fluoride   process  wastewater  alone.   The
aluminum fluoride plants for which  data  is  available
integrate   raw  process  wastewater  with  wastewaters
generated from the hydrofluoric acid process.

In view of the similar wastewater characteristics,  the
effluent  concentration  from a common treatment system
would be the same  for  TSS  and  fluoride  whether  it
originates  from  the A1F3 industry or the HF industry.
Therefore, based on limitations established for the  HF
industry   (Table  12-21},  a  maximum  30-day  average
concentration of 97 mg/1 TSS and 53 mg/1  fluoride  are
set  for  the A1F3 industry.  These are relatively high
values  that  are  unique  to   this   industry.    The
variability  factor  ratio of 2.1 was selected based on
the evaluation in the  HF  subcategory  (Table  12-23).
The  unit  effluent  load  limitation  is determined as
follows:

L (as kg/kkg) = (Q? (C)
                 1000

where C is the maximum 30-day average concentration  in
mg/1,  Q  is  the  unit flow in mVkkg, and 1000 is the
conversion factor for kg to  grams.   (Note:   kg/m3  =
1000 mg/1.)

The  24-hour  maximum  is  determined  by the following
relationship:'

Maximum 30-day average X VFR = 24-hour maximum
(concentration or unit-         (concentration or
loading)                        unit loading)

In this case, the daily maximum  TSS  concentration  is
2.1 x 97 mg/1 = 200 mg/1.

The 30-day average effluent is:

(97 mg/1) (11.9 mVkkg) (kg/in3) =  1.2 kg/kkg
                       (1000 mg/1)

The 24-hour maximum effluent is:

{200 mg/1) (11.9 m3/kkg)  (kq/m3) =2.4 kg/kkg
                        (1000 mg/1)

In   the   same  manner  the  concentration  basis  for
fluorides is:

2.1 x 53 mg/1 =110 mg/1.
                      492

-------
     The 30-day average effluent is:

     (53 mg/1)  (11.9 mVkkg)  (kq/m3)  = 0.63 kg/kkg
                           (1000 mg/1)

     The 24-hour maximum effluent is:

     (110 mg/1) (11.9 mVkkq) (kq/m3)  = 1.3 kg/kkg
                           (1000 mg/1)

Toxic Pollutants

The  effluent  limitations  set  for  the   selected   toxic
pollutant  control parameters are derived from three sources
of information including 1)   literature  based  treatability
estimates   (Section   8),  2)  screening  and  verification
sampling  data,   and  3)  a  limited  amount  of  long-term
monitoring data from Plant $251 .

The  sampling results represent raw process waste pollutants
observed during three days of composite sampling at each  of
the  plants  verified.   An  assessment  of treatment system
performance  was  not  possible  in  view  of  the  lack  of
representative  effluent  data available in the subcategory.
Effluent data obtained during verification sampling  is  for
treated  wastewater from the HF and A1F3 processes combined,
since no plant is available \*hich treats &1F3 wastes  alone.
Therefore,  the  screening and  verification data may be used
to  determine  candidate  toxic  pollutants  for  regulation
without  specifying  achievable  concentration  limits which
represent the A1F3 plant performance alone.  However, review
of the combined HF and A1F3 waste  effluent  data  in  Table
15-9  reveals  that  all  toxic  pollutants  of  concern are
treatable within  the  levels  of  treatability  defined  in
Section  8  for  lime  settling  (BPT).   Removal  of  toxic
pollutants from one wastewater  or the other would not differ
in light of the  similar  nature  of  HF  and  A1F3  wastes.
Therefore,   the   literature   estimates   of  treatability
discussed in Section 8 have  been  used  as  the  basis  for
determining   specific   numerical    limitations  for  toxic
pollutants.

a.   Chromium

     The literature treatability value  of  0.32  mg/1  from
     Table 8-13 for lime settling is  considered to represent
     a  long-term average concentration value for chromium in
     view  of  plant performance data  in the HF and combined
     HF/AlFj industries.

     Since long-term monitoring  data  on  chromium  is  not
     available,  the   variability  factor  ratio  (VFR) of 3.3
     was  derived  from  the   data  in  the  nickel  sulfate
                           493

-------
     industry.   This is justified by the fact that nickel is
     one  of  the  dominant  metal  pollutants  selected for
     control in the  A1F3  subcategory  and  the  wastewater
     characteristics  from the A1F3 and NiS04 industries can
     be assumed to be similar.  Therefore,

     VFR = VF of daily measurements = 3.9
            VF of 30-day averages     1.2

     VFR =3.3

     The effluent limitations for chromium are determined as
     follows:

     The maximum 30-day average is:

     (0.32 mg/1) (1.2) = 0.38 mg/1

     The 24-hour maximum concentration is:

     {0.32 mg/1) (3.9) = 1.3 mg/1

     The maximum 30-day average effluent limit is:

     (0.38 mg/1) (11.9 mVkkq) (kg/m3) = 0.0045 kg/kkg
                            {1000 mg/1)

     The 24-hour maximum effluent limit is:

     (1.3 mg/1) (11.9 m3/kkq)(kq/m3) = 0.015 kg/kkg
                           (1000 mg/1)

     The effluent limitations on chromium are  presented  in
     Table 15-15 for BPT treatment.
b.   Nickel
     For the purpose of regulation, a value of 0.17 mg/1 was
     derived  from  the  HF  industry  as  the   subcategory
     performance for nickel.  The limitations are determined
     as follows:

     The maximum 30-day average is:

     (0.17 mg/1) (1.2) = 0.20 mg/1

     The 24-hour maximum concentration is:

     (0.20 mg/1) (3.3) = 0.66 mg/1

     where 3.3 is the VFR as discussed for chromium.

     The maximum 30-day average effluent limit is:
                         494

-------
               {0.20 mg/1)  (11.9 m3/kkq)(kq/m*)  = 0.0024 kg/kkg
                                      (1000 mg/1)
               The 24-hour maximum effluent limit is:

               (0.66 mg/1) (11.9 mVkkg)   (kq/m^)  = 0.0079 kg/kkg  .
                                        (1000 mg/1)

          c.    Other Metals

               The concentration bases for arsenic,  copper,  selenium,
               and  zinc  are  also  presented  in Table 15-15.  These
               pollutants are listed to serve  as  guidance  in  cases
               where these pollutants are found to be of water quality
               concern.   The concentration limitations are also based
               on literature treatability levels  presented  in  Table
               8-11.   However in every case these treatability levels
               were above raw waste concentrations observed  for  each
               of these metals.

Basis for the BCT Effluent Limitations

While   EPA  has  not  yet  proposed  or  promulgated  a  revised  BCT
methodology in  response  to  the  American  Paper  Institute  v.  EPA
decision  mentioned  in Section 3, EPA is promulgating BCT limitations
for this subcategory.  These limits are identical to  those  for  BPT.
EPA  is  not promulgating any more stringent limitations since we have
identified  no  technology  option  which  would  remove   significant
additional  amounts of conventional pollutants.  As BPT is the minimal
level of control required by law, no possible application of  the  BCT
cost   tests   could  result  in  BCT  limitations  lower  than  those
promulgated in this regulation.  Accordingly, there is no need to wait
until  EPA  revises  the  BCT  methodology  before  promulgating   BCT
limitations.

Basis for BAT Effluent Limitations

A.   The Application of Advanced Level Treatment

     The Agency has analyzed the cost effectiveness of the base  level
     systems   (BPT)  and  the  various  advanced  level  options  for
     conventional, nonconventional and toxic pollutant  removal  based
     on  utilizing  the  cost estimates presented in this report.  The
     economic impacts on the  aluminum  fluoride  industry  have  been
     evaluated in detail and taken into consideration in the selection
     of the technology basis for the BAT regulations.

     For  BAT,  EPA is setting limitations based on Level 1 treatment.
     Pollutants limited in the BAT regulations are fluoride, chromium,
     and nickel.
                                    495

-------
     The Agency considered the use of treatment Level 2  (addition  of
     dual-media  filter) but did not adopt it because the installation
     of the filter is not  very  effective  in  the  removal  of  TSS,
     fluorides  and  toxic  metals.    EPA also considered limitations
     based on Level 3 and 4 sulfide precipitation and use of soda  ash
     to  increase  recycle, respectively.  These options were rejected
     because they remove  only  small  incremental  amounts  of  toxic
     pollutants in this subcategory.

B.   Technology Basis

     For BAT,  the  Agency  is  setting  the  effluent  limitation  on
     fluoride  and the toxic metals based on the BPT treatment system.
     Similar to the Hydrofluoric Acid  Subcategory,  the  addition  of
     dual-media  filtration  after alkaline precipitation and settling
     is not considered cost effective based on the conclusions made in
     Section 7 "Hydrofluoric Acid Subcategory" of EPA's  "Treatability
     Studies  for  the  Inorganic Chemicals Manufacturing Point Source
     Subcategory"  {61)  as  referred  to  in  Section  12   of   this
     Development Document.

C.   Flow Basis

     The  same  flow  established  for  BPT  above  is  used  in   the
     development  of  the  BAT effluent limitations.  The flow used is
     11.9 mVkkg of product (Table 15-4).

D.   Selection of Pollutants to be Regulated

     The  Agency  has  selected  fluoride  and  the  same  two   toxic
     pollutants   identified  in  the  BPT  regulations  for  the  BAT
     regulations.  The rationale  for  their  selection  is  discussed
     above.

E.   Basis of Pollutant Limitations

     The BAT limitations were set equal to BPT and are given in  Table
     15-15.

Basis for New Source Performance Standards

A.   Technology Basis

     For NSPS, the Agency set the same  treatment  technology  as  for
     BAT.

B.   Flow Basis

     The same flow  established  for  BPT  and  BAT  is  used  in  the
     development of the NSPS effluent limitations.
                                   496

-------
                   TABLE 15-15.  EFFLUENT LIMITATIONS
                           Aluminum Fluoride
        Best Practicable Control Technology Currently Available*
                     Waste water Flow;  11,9
Subcategory
Pollutant Performance

Conventional and
(mg/1)

Concentration Basis
(1) (iKf/l)
VFR
Max.
30-day 24-hr.
avg. max.








Effluent Limit
(kg/kkg)
Max.
30-day 24 hr.
avg. max.

ttonoonventiohal Pollutants:
Total Suspended
Solids, TSS
Fluoride
Tbxic Pollutants:
Arsenic
Chromium
Copper
Nickel
Selenium
Zinc

57 u'
33 (2)

0,50
0.32<3>
0.32<3>
0.17<3)
0. 20
0.55C3>

3
3

3
3
3
3
3
3



.5/1.7
.4/1

-9/1
.9/1
.9/1
-9/1
.9/1
.9/1
.6

.2
.2
.2
.2
.2
.2


(5)
(5)
(5)
(5)
(5)
(5)

97
53

0.
0.
0.



200
110

60
38
38
0.20
0.
24
0.66

2.
1.
1.
0.
0.
2.

0
3
3
66
78
2

1.2 2.4
0.63 1.3

_(6) _(6)
0.0045 0.015
_(6) __<6)
0.0024 0.0079
_(6) _<5>
_(6) _{6)
(1)  VFR:   Ratio of the 24-hour variability factor to the 30-day variability
    factor
(2)  Long  term average construction based on the HF si&categary regulation
     (Section 12.7.2).
(3)  The lower limit of treatability estimate  (Table 8-11) and the industrial
    waste water treatment system performance  (Table 8-12) are used as the
    basis for the long term average limitation and subcategory performance,
    since no plant is available where BPT treatment can be evaluated for
    the AlF 3 waste water alone.
(4)  VFR based on  HF subcategory evaluation.
(5}  VFR based on  limited long term data.
(6)  tto effluent limitation  proposed.
* The BPT effluent limitations are also applicable  to BAT except for the TSS
  limitations.  The BPT limitations  are also applicable for NSPS.
                                497

-------
C.   Selection of Pollutants to be Regulated

     The Agency has selected TSS, fluoride  and  the  same  two  toxic
     pollutants identified for the BAT regulations.  The rationale for
     their  selection  is  discussed  above.  The NSPS limitations are
     presented in Table 15-15.

Basis for Pretreatment Standards

The Agency is not promulgating PSES and PSNs because  a  well-operated
POTW provides equal or better removal of chromium and nickel (the only
toxic  pollutants  regulated  under  BAT for this subcategory) than is
achieved by BAT for  this  subcategory,  and  therefore  there  is  no
pass-through  of  these  toxic  pollutants  in  the  Aluminum Fluoride
Subcategory.  At present, there are no  indirect  dischargers  in  the
Aluminum Fluoride subcategory.
                                    498

-------
                              SECTION 16

                       CHROME PIGMENTS INDUSTRY
Industry Profile

General Description

Chrome  pigments are a family of inorganic compounds primarily used as
colorants in a number of  industries.   These  pigments  are  used  in
paints,  ceramics,  floorcovering  products,  ink, paper, and cements.
However, certain chromium compounds  (i.e., oxides) may be used as  raw
materials  in  the  manufacture  of certain metals and alloys.  Chrome
pigments vary substantially in their  chemical  makeup.   The  various
types  include  chrome yellow, chrome green, chrome orange, molybdenum
orange, anhydrous and hydrous chromium oxide  and  zinc  yellow.   The
industry  data  profile  is  given in Table 16-1 and the status of the
regulations prior to promulgation of this new regulation is  given  in
Table  16-2.

Subcategorization

Several  factors  were  originally considered in the Subcategorization
process, such as raw materials, products, manufacturing process,  size
and  age of equipment, and water pollution control technology.  It was
concluded that if effluent limitations were to be  tied  to  units  of
production,  only  subdivision  by  dominant  product  was viable as  a
method of primary  Subcategorization.   Further  subdivision  was  not
warranted  in the chrome pigment industry.  A more detailed discussion
of the Subcategorization process may be found in Section 4.
General Process Description and Raw Materials

The general manufacturing process for each of the above
given below.
                                                    compounds  is
A.
Chromium Oxide
     This pigment consists of two compounds:  anhydrous  and  hydrated
     chrome  oxide   (Guigets  Green).   The  amount  of  the anhydrous
     chromic oxide produced is approximately ten times the  amount, of
     hydrated chromic oxide produced.  It  is offered in a narrow range
     of shades from  light yellowish  to dark bluish green.

     The  anhydrous  oxide  is  almost  pure  chromium  oxide  and the
     commercial grade consists of a  minimum of 98.5 percent Crz03.  It
     is prepared by  calcination of sodium  dichromate  with  sulfur  or
     carbon according to the reactions given below:
           S = Cr0
                      23
                            Na2SO4
(1)
                                    499

-------
             TRBLK 16-1.   SUBCAIEGOBy PROFILE .DMA SUMMARY
                                                          (1)
     SUBGAOEGORy
CHROME PIOENIS
     Total subcategory capacity rate
     Ttotal subcategory production rate
     Number of plants in this subcategory
     308  Data on file for
         With total capacity of
         With total production of
         Representing capacity
         Representing production
         Plant production range: (2)
                 Minimum
        Average production
        Median production
        Average capacity utilization
        Plant age range:
                Minimum
                Maximum
        K&stewater flow range:
                Minimum
                Maximum
        Volume per unit product:
                Minimum
                Maximum
                           73,500 kkg/year
                           73,500 kkg/year
                               12
                           39,800 kkg/year

                                62 percent

                               100 kkg/year
                           18,000 kkg/year
                             6,300 kkg/year
                             6,400 kkg/year
                                78 percent

                                38 years
                                60 years

                               800 cubic meters/day
                           11,363 cubic meters/day

                                32 cubic meters/kkg
                               170 cubic meters/kkg
(1)   Sources of data are Stanford Research Institute, Directory of Chemical
     Producers,  U.S.A.,  1977, U.S, Department of Commerce, Current Industrial
     Reports, Decanber 1977; Energy and Environmental Analysis, Inc.; Draft
     Report, "Preliminary Economic Assessment of Effluent Limitations in the
     Inorganic  Chenical  Industry," June,  1978  and "Ecoronic Analysis of Proposed
    Revised Effluent Guidelines and Standards  for the Inorganic Chemicals Industry,
    March, 1980.
(2)   Based, on production at 11 plants, all other figures are based on 308
     Ques tionnaires.
                                        500

-------
     TABLE 16-2.  STATUS OF REGOIAHCNS  -  EEFUJENT-UMZTKnO? GUIDEU2ES
SDFTflTOaOKSr


SOBPAET
Chrome Pigments


AH  (40CFR 415.340, 5/22/75)
                                        STANDARDS
Product
Process
Chrome
Pigment








Para-
meters
TSS
Cr(T)
Cr*6

Pb
Zn
CN
CN(A)
Fe
BEdCA* BSOEA NSPS
Max.1 Avg.2 Max. Avg, Max. Avg.
iCQfi£tCO Tfrr /\r\rrt l^<"r /Irlrt-i T^rr /\r\ff^r If t*r J]F\rr* T^rr /I^l/*/^
«y/ JWg rj-j/ ™s-y i^y/ "-"M K*"y/ ^'iSJ ^y/ '^Ss J^^' ^^3
Cmg/1) (ing/1) (mgA) fin^l) &ng/l) Tirag/1)
5,1
(76.1)*
0.10
(1.5)
0.010
(0.2)
0.42
(6,3)
0.72
(10.8)
0.010
(1.5)
0.10
(0.2)
0.72
(10.8)
1.7 Reserved Reserved
(25.4)
0.034
(0.5)
0.0034
(0.1)
0.14
(2.1)
0.27
(4.0)
0.0034
(0.5)
0.034
(0.1)
0.27
(4.0)
          415.340, 415.341, and 415.342 were revoked by the Agency

 £41 FR 51601, November 23, 19761.

  jMax. » Maximum of any one day,
 2
  Avg. m Average of daily values for thirty consecutive days.
 *£Low basis  67,000 1/kkg,
                                   501

-------
Na2Cr2 07 + 2C =
                              NazC03 + CO
(2)
     The use of sulfur as the reducing agent  eliminates  C02  and  CO
     emissions  but increases the sulfates in the raw waste as well as
     producing S02 and S03 in the  off-gases.   In  the  manufacturing
     process  using  sulfur,   the  raw  materials consisting of sodium
     dichromate and sulfur are mixed  with  water  and  the  resultant
     solution  is  fed  to  a  kiln.   The  material is heated and the
     reacted  materials  from  the  kiln  are  slurried  with   water,
     filtered,  washed,  dried,  ground,  screened, and packaged.  The
     effluent gases from the kiln containing sulfur dioxide and sulfur
     trioxide are wet scrubbed before venting to the atmosphere.
     A general process flow diagram of the
     chrome oxide is given in Figure 16-1.
                                       preparat ion  of  anyhdrous
     Hydrated  chromium  oxide, CrzO3»2H20 or Cr20(OH)4/ also known as
     chromium hydrate and Guigets Green, is a brilliant bluish  green.
     It  is  made  by  reacting  sodium  dichromate with boric acid as
     follows:
2Na2Cr2O7 + 8H3B03 = 2Cr203»2H20
                                                          30
  (3)
     The raw materials are blended in a mixer and then  heated  in  an
     oven  at  about  550 degrees C.   The reacted material is slurried
     with water and filtered.  The filtered  solids  are  washed  with
     water,  dried,  ground, screened, and packaged.  The filtrate and
     the wash water are treated with sulfuric acid  to  recover  boric
     acid according to the reaction given below:
     2NazB407 * H20 = 4H3BO3
                                                            (4)
     A waste stream containing some boric acid and sodium  sulfate  is
     discharge from the boric acid unit.  Figure 16-2 is a generalized
     flow diagram of the process.

B.   Chrome Yellow and Chrome Orange

     Chrome yellow is one of the more  important  synthetic  pigments.
     The  chrome  yellows  cover the range of hues from light greenish
     yellow to reddish  medium  yellow  and  consist  mainly  of  lead
     chromate.   They  are made by reacting sodium dichromate, caustic
     soda, and lead nitrate.  The reactions are given as:
     2HN03 + PbO = Pb(N03)z + H20

     Na2Cr207 + 2NaOH + 2Pb(N03)2 = 2PbCr04 + 4NaN03 + H20
                                                            (5)

                                                            (6)
     Lead chromate is formed as a precipitate during the reaction.  It
     is filtered and treated with chemicals for development of desired
     specific pigment properties, dried, milled,  and  packaged.   The
     filtrate  from the filtration operation is sent to the wastewater
                                    502

-------
                          VHff
— Of
SCHUBBER 	 *• "•
IJQUID SCW-
i
in WATER
" m . 1
>2

f
^S «?" "T
f •

SOLFTm ^ PACXflGING OF
CHROME OXIDE
raceucT
                                                               VRSH HATER
Figure 16-1.  General process diagram for production of anhydrous chrome oxide.

-------
               SCOIUM
             PICHKCHATE
  BORIC ACID
ui
o
».
[ I*"

HIXEK
^

CVIK


WiTER
SUWRTf
1ANK
WKBER
PILTTO

1 — •P'1

HtVER
HVDRA.1Q) CHRCMB
C«II« TO CRIKflNG,
SCREEN! rC AtC
                                                     VRSTE HVTER
                             Figure 1G-2.   General process diagram for production  of hydrated chromic oxide.

-------
treatment  facility.    A  flow  diagram  of  the  chrome
manufacturing process is shown in Figure 16-3.

Molybdenum Orange
                                                                yellow
     Molybdenum orange is made by the coprecipitation of  lead chromate
     (PbCr04)  and lead molybdate (PbMo04).   The resulting pigments are
     more brilliant than chrome oranges.

     The process consists of  dissolving   molybdic  oxide  in  aqueous
     sodium  hydroxide  and  adding  sodium chromate.  The solution is
     mixed  and  reacted  with  a  solution  of  lead  nitrate.     The
     precipitate  from the reaction is filtered,  washed,  dried,  milled
     and packaged.   The filtrate is sent  to the treatment facility.
The reactions are given as follows:

Mo03 + 2NaOH = Na2Mo04 + H20

PbO + 2HN03 = Pb(N03}2 + H20

NazMo04 + Pb (N03)2 = PbMo04 + 2NaN03

Na2Cr04 + Pb (N03)2 = PbCr04 + 2NaN03

PbMo04 + PbCr04 = PbCr04»PbMo04
                                                            (7)

                                                            (8)

                                                            (9)

                                                            (10)

                                                            (11)
     A simplified flow  diagram  for  the  manufacture  of  molybdenum
     orange is given in Figure 16-4.

D.   Chrome Green

     Chrome greens are a  coprecipitate  of  chrome  yellow  and  iron
     blues.   They  include  a wide variety of hues from very light to
     very dark green.   Iron blues  are  manufactured  by  reaction  of
     aqueous  solutions  of  iron  sulfate  and  ammonium sulfate with
     sodium ferrocyanide.  The precipitate  formed  is  separated  and
     oxidized  with  sodium  chlorate  or sodium chromate to form iron
     blues, Fe(NH4)«Fe(CN)6.  Chrome green is produced by mechanically
     mixing chrome yellow  and  iron  blue  pigments  in  water.   The
     coprecipitate formation of chrome green is given by:
     PbCr04 + Fe(NH4)«Fe(CN)6 = PbCr04Fe(NH4)«Fe(CN)6
                                                       (12)
     Figure  16-5  gives a process flow diagram for the manufacture of
     chrome green.

     Zinc Yellow

     Zinc yellow, also called zinc  chromate,  is  a  greenish  yellow
     pigment.   It  is  a  complex  compound  of  zinc, potassium, and
     chromium    which     has     the     approximate     composition
     4ZnO»K20«4Cr207*3H20.   It is made by the reaction of zinc oxide,
                                    505

-------
                           UftDQBOCE
l/i
O
                                                                                                     CHOB VEU£H
                                                                                                     TO DWfINQ, HIUJNC
                                                                                                     MD BAOCAGJNQ
                                                                                          WTER
                               FigiJre 16-3.   General process diagram for production of chrome yellow.

-------
                                   VENT
ui
o
-J
                   CMDE
                WMER
            CAUSTIC SOCft
                            DISSOLVE*
                                                      SODIUM
                                                     CHKMUE
11
                                                          MIXER



OXIDE __

jc ACID



VENT


DISSOSJVER








^^ PILTOftTICN
mSHINQ
fASTE HA.TER




                                                                                                                    DRYING
                                                                                                                    MILLING
                                                                                                                    AND
                                                                                                                    PACKAGING
                                                                                                                    CF
                                                                                                                    MDLYBCENIM ORANGE
                                                                                                                         J4'
                                                                                                                    PRCCUCT
                           Figure  16-4.   General process diagram for production of molybdenum orange.

-------
     hydrochloric acid, sodium  di chroma te,  and  potassium  chloride.
     Zinc  yellow  is formed as a precipitate and is filtered, washed,
     dried, milled, and packaged for sale.  The  reactions  are  given
     as:
2HC1 + 2Na2Cr207»H20 = K2Cr4013 + 4NaCl + 3H20  (13)

K2Cr4013 = 3H20 + 4ZnO«K20»4Cr04»3H20           (14)
     2KC1

     4ZnO

     A  general  flow diagram of the manufacturing process is given in
     Figure 1 6-6.

Water Use And Waste Source Characteristics

Water Use

In  the  chrome  pigments  industry,  water  is  used  primarily   for
noncontract  cooling,  washing the precipitated product, and as boiler
feed for steam generation.  In some cases, water  is  introduced  into
the reactor along with the raw materials.

In  addition,   substantial quantities of water may be used in cleaning
equipment.  This occurs during product changes at plants manufacturing
a number of pigments.  This partially accounts for the increased  unit
water  use  at larger plants, since these plants have the most complex
product mix.

In anhydrous and hydrated chrome oxide manufacture, water is used  for
slurrying  of  the  reaction product and in scrubbing the reactor vent
gases.  Table 16-3 is a summary of water usage  at  different  pigment
plants in the chrome pigments subcategory.

Waste Sources

Some  plants  produce  different  pigment products sequentially in the
same equipment.  At a few plants, the different pigment  products  are
manufactured  concurrently and the wastewaters combined and treated at
a single facility.  A  generalized  flow  diagram  applicable  to  all
chrome pigment plants is given in Figure 16-7.  The wastewater sources
are  similar  for  all  pigment  products  except that at chrome oxide
plants, an additional scrubber waste is generated.  Table  16-4  gives
the  wastewater flow data summary for several plants.  The quantity of
wastewater and the pollutants are dependent of the raw materials used.
The figures in Table 16-4 represent actual plant discharges.

The data sources for the plants used in the determination of unit flow
values presented in Table 16-4 are outlined below:
Plant fIBpl  Data based on 308-Questionnaire submission.
pigment production and flows were included.
                                              Only  chrome
Plant WifBKW Oo&a-* .based  on  308-Questionnaire  submission.
pigment and iron blue production and flows were included.
                                                    Chrome
                                    508

-------
                                                                      HATER
                                                                       1
                                              IRON BLUE
                                                                     RESLURRY
tn
O
                                                  HATER
ij^-jj W.ITKA-IT: ^
SODIUM CHfcOMATE ^
SODIUM SOIFATE _

PEACTIOH

	 	 »


f 1
FILTER
AND
HASH
1


SHADE

f
TANK









DRYER
1
	 •» GRINDING
BUDDING
AND
PACKING
aiROME GREEN
PRODUCT
                                                 HASTE HATER
HASTE HATER
                                 Figxore 16-5.  General process diagram for production of chrome green.

-------
                    uo
Ul
H
O
^
KM fc
— T 	
use-new TMK


piuwncw
ma tun


HWINQ
>NIUJNSf PACKAGDC
 OP 1KB UN? YQiCH
                              Figure  16-6.  General  process diagram for production of zinc yellow.

-------
TABLE 16-3.  WATER USAGE IN THE CHRCME PIGMENTS SUBCAGEGOPY
                                                           (1)
         USE
                                     #464
UNIT FLOW (m3/kkg)
Plant Designation
     #436          #214
Noncontact cooling
Direct process contact
Indirect process contact
Maintenance
Scrubbers
Boiler Feed
Total
9.50
18.6
7.18
12.0
3.30
2.52
53.1
6.45
147
NA<2>
1.78
9.56(3>
11.1
176
NA
32.6
NA
0.152
NA.
0.152
32.9
 (1)  Includes all chrome pigment product mixes.  Values indicated only
     for those plants that reported complete information.
 (2)  NA - Not applicable.
 (3)  Iron blue pigment process.
                                   511

-------
  HATER
HASH HATER
                              i
                           IMREACTED MATERIALS,
                           ETC.)
                                                                       WASH DOWN HATER
H RAW MATERIALS ^
to 	 ^


REttTOR




FILTER




COOfER




HILLING AID
SCREENING;

	 *» PIGMEHT
PRODUCTS
TO
1 11 1 PACKAGING
PIGMENT
WASTE WATER HCN-OOWiaCT KOfflCDlATE
(BY-PPDDUCT SAL'l-S, fflVAM WASTE
Figure  16-7.  General process  diagram for production of chrome pigment complexes.

-------
Plant 1436.   Data  based  on  308-Questionnaire  submission.   Chrome
pigment production and flows were included.

Plant #002.  Data based on three days of sampHng.  Chrome pigment and
organic pigment (20 percent) productions and flows were included.

Plant  #894.   Data  based on three, davs^of^sampling.  Chrome pigment,
iron blue, and organ i c pi gmen t (15 percent) 'IpFocTucTlTDns and flow  were
included.

As  previously  discussed, various plants make several chrome pigments
sequentially or concurrently.  Thus the unit hydraulic load  going  to
the  treatment facility will be an average of all the waste loads from
the different processes.  The raw  waste  from  a  complex  plant  may
contain nearly all of the following substances: sodium acetate, sodium
chloride,  sodium  nitrate,  sodium sulfate, potassium chloride; lead,
iron, and zinc salts; soluble chromium and pigment particulates.

Description Of Plants

Screening

Plant #894 was visited during the screening phase of the program.  The
samples for this plant were analyzed for all  toxic  and  conventional
pollutants.

This  plant produces over 100 products including organic pigments such
as copper phthalocyanine, and all the wastes are combined and  treated
together.    Treatment  consists  of  hexavalent  chromium  reduction,
equalization  and  neutralization,  followed  by   clarification   and
filtration.  Sulfur dioxide is added to reduce the hexavalent chromium
to  the  trivalent state at a low pH prior to hydroxide precipitation.
The backwash from the sand filters is cycled to the  equalization tank,
while the sludge from the clarifiers is passed through filter  presses
and  then  hauled to a landfill.  The landfill has a bottom  consisting
of two clay layers sandwiching a gravel layer to allow for   collection
of  leachate  drainage.   Any  water from  the sludge is trapped  in the
gravel layer, and  is  pumped  out  and  returned  to  the  plant  for
retreatment.

Verification

Two  plants were visited during the verification phase of the program.
The first plant, #002, has a rather large  product mix.   However,  one
of  the  larger  continuous  units  can have a major impact on the raw
waste characteristics.  This unit either   produces   lead  chromate  or
zinc  chromate.   During  the sampling period, zinc  chromate was being
produced.  All process wastewaters are treated  continuously.    First,
the  wastes  are  treated   in  an  S02  reactor   to  convert  hexavalent
chromium to the trivalent state.  The pH is then adjusted to  8.5  and
then  the  waste  is  passed  through  precoated  f i1ters, followed by
discharge  to the sewer.  Figure  16-8 shows the treatment flow  diagram
and  sampling  points.  Table 16-5 shows the waste flows and pollutant
                                   513

-------
           TSBIE 16-4.  SU&MAEY OF WBSTE WaiTER FLOW
                       SUBCftTEQORY:  CHRCME PIGMENTS
           Plant Designation                  Waste Water Flovr '
                                                (m3Akg)
                 #464                               41.1
                 #214                               32.8
                 #436                               149
                 #002                               78.4(2>
                 #894                               170(2)
                                                       (3)
                 Average Flow                      105v '
(1)   includes waste water from all pigment product mixes.
(2)   Includes organic pigments.
(3)   Weighted on the  basis of production since unit waste flew is
     directly related to   plant production:
                 Weighted average =   z [ (unit flew) (production) ]
                                        I (production)
                  .e.  = Q)  + Q2(P2)  + Q3(P3)+...-K3n(Pn)
                 Where Q = Unit flew and P = production (which is considered
                 confidential information),
                                  514

-------
loadings.  At sample point #2, half the sample was filtered through  a
glass  fiber  filter  on  a Buechner funnel to simulate the filtration
process which was being bypassed at the time  of  sampling.   Analyses
were  carried  out  on the filtered and unfiltered samples in order to
make possible a comparison of the total and dissolved concentrations.

A review of the sampling  data  indicates  that  the  waste  treatment
facility was not functioning properly during the period of sampling at
Plant #002.  The inadequacies observed in the treated effluent quality
have  been  related  to  deficiencies  in  the treatment system design
including 1) inadequate equalization and S02 contact  facilities,  and
2)  inadequate  clarification  which  in  turn  caused blinding of the
filter and the subsequent need for filter bypass.

Plant #894 was  also  visited  during  the  verification  phase.   The
treatment  system has been previously discussed.  The major problem at
this plant is the high unit water use rate.   However,  this  was  the
only  plant  found  with an adequately designed and operated treatment
system in this subcategory.

During the verification phase, only certain pollutant parameters  were
analyzed.  These were pH, cyanide, suspended solids, and toxic metals.
No organics were analyzed during verification.

Figure   16-9 shows the treatment system flow diagram with the sampling
points   indicated.   Table  16-6  gives  waste  flows  and   pollutant
loadings.

Toxic Pollutant Concentrations

The  toxic  pollutants found  above treatable concentrations in the raw
wastes during sampling are given in the table below.
                 Maximum Concentration (j/g/1)

Pollutant               Screening              Verfication
                                                 2  Plants
Cadmium
Cyanide
Chromium
Copper
Lead
Zinc
Antimony
Selenium
Silver
Nickel
Phenol*
Bis (2 ethylhexyl)
Phthalate*
79
360
55,000
7,500
36,000
4,100
7,700
> 10
7
160
73

>0. 1
1,250
8,200
349,000
4,700
69,000
273,000
1,475
28
20
740



* from organic pigment process
                                   515

-------
                   RAW WASTE  SO,
                                          ACID
                         CHROME TREATMENT
                             TANK
                            pH 3.0
      CAUSTIC
                             CAUSTIC ADDITION
                                  THROUGH
                                  pH 8.5
                                                      ~l
                           FILTER  FEED
                              TANK
           LAB FILTERED
           	1
                                 FILTER AID
                                                              BACKWASH
OUTFALL
TO SEWER
                                          I	J
                                        (FILTERS NOT WORKING SO
                                       WERE BEING BYPASSED.
                                       THIS WOULD BE THE FLOW
                                       PATTERN IF FILTERS WERE
                                       OPERATING.)
                                                       LEGEND
                                                    SAMPLING  POINTS
Figure 16-8.
         General waste water treatment process flow diagram at plant #002
         showing the sampling points.   (Chrome pigment manufacture.)
                              516

-------
    TABLE  16-5.  FLOW, POLLUTANT CONCENTRATION AND LOAD
           DATA OF THE SAMPLED WASTE STREAMS FOR PLANT # 002
                      SUBCATEGORY:  CHROME PIGMENTS
               Conventional and Nonconventional Pollutants
                                 mg/1
(kg/kkg of chrome pigments)
Stream
#
1

2-U


2-F


Stream
Description
Raw Waste

Unfiltered
Treated
Waste
Filtered
Treated
Waste
Flow
(m3/kkg) TSS
78.4 700
(55)
78.4 970
(76)

78.4 NA '


Fe
1.6
(0.13)
2.3
(0.18)

0.06
(0.0047 )

Cr(VI)
300
(24)
120
(9.4)

NA(1)


(1)  NA - Not available
                                   517

-------
                                                                      SLAKED
                                                                       LIME
                     WASTE
                     H2S04   SO.

                       LJL
WASTE /   £fr
WATER '~
               #1
           (EPA SAMPLE
           POINT AISO)
                          BLEND
                          TANK
                       PH  2.S-3.0
Ul
M
03
                EQUALIZATION
                    TANK
NEUTRALIZATION
     TANK
  pH 6.2-6.5
NEUTRALIZATION
     TANK
  pH 8.0-8.3
                                        BACKWASFT
                                         HOLDING
                                          TANK
                              BACKWASH
         (2)
        SAND
       FILTERS

              CLARIFIER
               EFFLUENT
               HOLDING
                TANK
                                                                                    (3)  CLARIFIERS
                                                                     #2  (EPA SAMPLE
                                                                          POINT ALSO)
                                                                FINAL
                                                              DISCHARGE
                                                              TO RIVER
                                   #3
                                   e
                                 SLUDGE
                               LEACHATE
                            GRAB SAMPLE
                                                                  SLUDGE
                                                                  HOLDING
                                                                  TANK
                                                    FILTRATE
                                                                                  LEGEND

                                                                                SAMPLE POINTS
Figure 16-9.
                                    General waste water treatment process flow diagram at plant  #894
                                    shewing the sampling points.  (Chrome pigment manufacture.)

-------
TABLE 16-6.  FLOW, POLLUTANT, CONCENTRATION AND LOAD DATA FOR THE SAMPLED
             WASTE STREAMS AT  PLANT # 894
                      SUBCATEGORY:  CHROME PIGMENTS
              Conventional and Nonconventional Pollutants
                                mg/1
                        (kg/kkg of chrome pigments)
Stream           Stream        Flow         TSS        Fe          Cr (VT)
  #            Description    (m3/kkg)
1

2


3

5


Raw Waste

Final
Discharge

Leachate

Sand Filter
Influent

170


170

NA*1'


170

770
(130)

3.9
(0.66)
ND<2>


11
(1-9)
48
(8.2)

0.30
(0.051)
0.04
(NA)

1.0
(0.17)
ND<2>


0.023
(0.0039)
ND(2>


ND(2)

(1)   Not Applicable

(2)   Not Detected

(3)   Verification sampling which involves three 24-hour composite samples.
                                    519

-------
Screening data was obtained at Plant 1894.  Verification was completed
at Plants #894 and #002.  The only organic pollutant not listed  above
found  in  the raw waste above the protocol detectable limit (10 *g/l)
was naphthalene at 14 i*g/l.  It should be noted,  however,  that  some
nitrobenzene  (56  »/g/l}  and phthalates at levels up to 220 ^g/1 were
found in ttte treated effluent and one raw water  intake.   Since  they
were not present in the raw wastes, it is presumed they are present as
a  result  of sample contamination; i.e., plasticizer in Tygon Tubing.
No organic pollutant sampling was made during verification.

Section 5 of this report describes the methodology  of  the  screening
and verification sampling program.  In the chrome pigments industry, 9
days  of  sampling  were  conducted  at  Plants  #894  and #002.  This
involved five different sampling points  for  raw  and  treated  waste
streams.   The  evaluation  of  toxic  metals content of these process
related waste streams was based on 195 analytical  data  points.   The
screening  at  Plant #894 for organic pollutants generated another 228
data points.  The daily raw waste loads were calculated from the waste
stream flow rates measured or estimated at the time  of  sampling  and
the measured concentration.

That is.

     Daily loading as kg of pollutant = (C) (Q)
                          day            1000

where:

     C  is  the  concentration  of the pollutant expressed in units of
     mg/1  (Note:  kg/m3 =  1000 mg/1), and

     Q  is the waste stream flow rate expressed  in  units  of  m3/day.
     (m3, a cubic meter, is equal to 264.2 U.S. gallons.)

Similarly,  the unit loadings were calculated from the reported  chrome
pigments production rate, the waste stream flow rate, and the measured
pollutant concentration.

     Unit loading (as kg of pollutant   =     (C)   (Q)
     per kKg of chrome pigments)              (1000HP)

     where C and Q are the same as  described  above,  and  P   is  the
     pigment  production  rate expressed  in units of kkg/day.   (kkg is
     1000 Kg, a metric ton, which is equal to 2205 Ibs.)

The minimum, average, and maximum values are based on data from  those
plants  where  the  particular  pollutant  was  found at  concentrations
greater than the analytical detection limits  and significant   in  that
it  could   conceivably  be  treated by an available  treatment technology
regardless  of economic  considerations.

In Table  16-7, the toxic pollutant raw waste  data  are presented as the
average daily  concentrations  and  the   unit   loading   found   at   the
                                   520

-------
individual  plants.   The  overall  averages  are  also shown and were
subsequently used in the calculations of the  average  daily  loadings
and  the  average  unit  loadings  shown  in Table 16-8 along with the
corresponding  minimum  and  maximum  values.   The  toxic   pollutant
concentrations in the treated effluent are presented in Table 16-9 for
the two plants visited during verification sampling.

Based  on  the  total  annual  production  from  Table  16-1  of  this
subcategory and the average waste load generated per unit product from
Table 16-8, the  estimated  total  toxic  pollutant  raw  waste  loads
generated each year for this subcategory are as follows:
Pollutant
Waste Load (kg/year
Antimony
Cadmium
Chromium
Copper ( 1 )
Lead
Nickel (1)
Zinc
Mercury
Cyanide ( 1 )
Phenol (1)
Phenolics ( 1 )
42,000
8,000
1,160,000
54,000
252,000
1,400
104,000
260
34,000
900
1 ,500,000
 (1) Probable Source  is organic pigment process or other processes.

 The  waste  load  for zinc and lead  in the above have been adjusted to
 reflect the  fact  that  not  all  plants  produce  a  chrome  pigment
 containing  lead, and most do not produce zinc yellow.  For  example, a
 plant that does not  produce zinc yellow will have a very  low level  of
 zinc  in the raw waste load.  See the data for plant #894  in Table  16-
 7.
                                   521

-------
                16-7.  TOXIC POLLOTRNT PAH HASTE DMA
                                     CHHCMS
Average Daily Pollutant Concentrations and loadings at Plants Saiplad (1)
                        	JB/1.	
                         (kg/kkg of Chrome Pigments)
Pollutant
Antimony
Cadnimx
Chrcjnuin
Copper
Lead
Nickel
Zinc
Mercury
Cyanide, CM
Cyanide, OJ(A)
#894 (S) (2
7.7
U.5)
0.79
Co-is)
55
(10)
7.5
(1.4)
36
(6-8)
0.16
(0.030)
4.1
(0.78)
*
*
3.6
(0.68)
*
p1ap4- nM-i^n*
!) t894
-------
                              TABLE 16-8.  SU44ARY OF RAH VRSTE LADINGS PQUM) IN SCREENING AND VESUFIOVTIGN SAMPLING
W

SUBCAIBGORY
Pollutant
Itoxic
Antimony
Cadmium
ChrcmW1'
OOpper
load
Nickel
Kino
Meccuty
Cyanide, aj
CHROME PIGMENTS
leading tenge,
(kg/day)
HinimiiQ fbxinun

6.
0.
700
6.
55
0.
48
0.
3.
Cyanide, CN(A)
Phenol
Phenol ica
Conventional
Total
Suspended
Solids, TSS
Fe
Hexavalent*1*
Chroniun
Cr+fi


and

0 98
87 10
. 1300
1 96
459
19 2.0
714
0019 0,48
1 56
9.8
0.93
8.8
Nonoonvaitional
3100 8800
7.

1 550
1300
Unit loading,
(kg/kkg)
Mininuro Average

0.11
0.016
10
0.11
0.82
0.0028
0.71
0.000034
0.056



55
0.13


0.58
0.11
16
0.74
3.9
0.019
4.8
0.0036
0.53
0,15
0.014
0,13
93
4.2
24
Maxinun

1.5
0.15
24
1.4
6.8
0.030
1313
0.0072
0*84



130
8.2

K,. of'2'
Plants
Averaged

3
3
3
3
3
3
3
2
3
1
1
1
2
2
1
                              (1)  Hexavalent chromium is only one valent form of  chromiun.



                              (2)  Oily those plants where the pollutant vas observed at significant levels were included.

-------
       TABLE 16-9. TOXIC POLLUTANT TREATED V&STE DATA


Pollutant
Cmg/1)
Antimony
Cacbuim
Chromium
Copper
Lead
Nickel
Zinc
Mercury
Cyanide, CN
Cyanide, OJ(A)
SUBCATEGOIGf:
CHEDME PIQYENTS
Plant Designation
#894. #002
0.30
0.0084
0.33
0.035
0.11
0.021
0.058
ND(3>
0.065
0.0067
0.43
0.12
130
0.077
1.5
0.083
117
ND
*
*

Overall C2)
Average Concentration
0.37
0.064
65
0.056
0.81
0.052
59
ND
0.065
0.0067
(1)   Verification sampling concentration data,  average of three 24-hour
     composite samples.

(2)   Average of two plants shown during verification sampling.

(3)   Not detected.

*    No data
                                   524

-------
Pollution Abatement Options

Toxic Pollutants of Concern

The toxic pollutants found in significant amounts are mostly the heavy
metals found in the products as well as the chromium ore and other raw
materials.  These metals are cadmium, chromium,  copper,  lead,  zinc,
antimony  and  nickel.   In  addition,  some  cyanide was found in raw
wastes and treated  effluents.   This  cyanide  is  a  result  of  the
manufacture of iron blues and, at one plant site, HCN.  However, these
guidelines  do not apply to iron blues; they will be included  in Phase
II of  the  Inorganic  Chemicals  regulation  development.   There  is
significant  removal of the cyanides in the chrome pigments treatment,
however, probably due to the precipitation of ferrocyanides.   The  HCN
manufacturing  process  is  also  regulated  by another guideline {see
Section  17).  Some organic toxic  pollutants  were  found  during  the
screening  phase.   This  was  believed to be an anomaly caused by the
sampling procedure, since they were  also  found  in  the  raw intake
water,   treated  effluent,  and  in  the  raw waste.  In addition, any
organics present are probably caused by organic  pigments  manufacture
which   is not regulated by this guideline, but will be regulated under
the Organic Chemicals Category.

All the  wastewaters  generated  in  the  Chrome  Pigments  Subcategory
contain  dissolved chromium and pigment particulates.

Additional pollutants that may be anticipated are given below  for each
major pigment group.

A.   Chrome Yellow and Chrome Orange

     The raw wastewater  contains   sodium  acetate,  sodium  chloride,
     sodium nitrate, sodium sulfate, and lead salts.

B.   Chrome Oxide

     The aqueous process effluent contains sodium sulfate.   If  boric
     acid   is  used   in the preparation of hydrated chromic oxide then
     the wastewater will contain sodium borate and  boric acid.

C.   Chrome Yellow and Chrome Orange

     Additional pollutants present  in  the raw wastewater  from chrome
     yellow  and   chrome  orange  manufacture  include  sodium  acetate,
     sodium chloride, sodium  nitrate,  sodium sulfate, and lead salts.

D.   Molybdenum Orange

     Process waste effluents  from the  manufacture of molybdenum orange
     contain sodium chloride, sodium  nitrate,  sodium  sulfate,  chromium
     hydroxide,  lead  salts, and silica.
                                      525

-------
E.   Chrome Green

     The raw wastewater contains sodium  nitrate.   If  iron  blue  is
     manufactured  on-site  as  part  of  the process for chrome green
     manufacture,  the  wastewater  also  contains  sodium   chloride,
     ammonium  sulfate,  ferrous  sulfate, sulfuric acid and iron blue
     pigment particulates.

F.   Zinc Yellow

     The  raw  wastes  contain  hydrochloric  acid,  sodium  chloride,
     potassium chloride, and soluble zinc salts.

Process Modifications and Technology Transfer Options

The  major  process  problem in the industry is the high rate of water
use in some cases.  This can be alleviated in a number of ways.

A.   Close attention to product quality in conjunction with  reduction
     of product rinses.
B
Reduction  in  equipment  cleaning  rinses
methodologies:
by   the   following
     1 .

     2.
     Recycle of rinse waters.
     Minimizing of product changes by the use of better  planning
     and increased number of units.
     Equipment  cleaning  is  known  to  contribute  approximately  20
     percent of the waste load volume at one plant (#002).

C.   Use of parallel treatment for  individual  product  lines.   This
     will allow the reuse of rinse waters and the recovery of products
     presently lost in waste sludges.


D.   The use of  ion  exchange  and/or  reverse  osmosis  on  isolated
     wastewaters.   This  will allow total recovery of product as well
     as total reuse of wastewater.  This system is in use on one  line
     at Plant 1409.

The  above  options  were  reviewed,  but except for option 1 were not
considered  for  inclusion  in  the  treatment  models  due   to   the
engineering required and their capital intensive nature.

Best Management Practices

A.   All storm water and surface  area  runoff  from  the  plant  site
     should be collected and sent to a treatment facility if the water
     is  contaminated  from process wastes.  This contamination can be
     minimized by storage of chemicals indoors, proper  air  pollution
     control, and elimination of all spills.
                                    526


-------
B.   If the solids from the treatment plant are disposed  of  on-site,
     provision  should be made to control leachates and permeates.  It
     is  possible  to  monitor  the  metal  concentrations  and   when
     concentrations approach predetermined limits, the leachate can be
     pumped back to the treatment system for further treatment.

Prevailing Control and Treatment Practices

A  description of the individual treatment facilities for those plants
visited is given above.  In addition, the  following  information  was
obtained for the remaining plants.

Plant  #214 manufactures pigments and other chemicals.  The plant does
not have a wastewater treatment facility.  After pH adjustment,  waste
is discharged to a POTW.  Part of the process waste is recycled.

Plant #593 manufactures organic and  inorganic chemicals.  The existing
combined  wastewater  treatment  plant  consists  of lagoon, aeration,
clarifiers, and filters.  The sludge disposal is on-site landfill.

Plant #464 manufactures both organic and inorganic pigments.  After pH
adjustment, wastewater is discharged to a POTW.

Plant  #101  manufactures  inorganic  ceramic  pigments,   color   and
porcelain.    The  existing  combined  wastewater  treatment  facility
consists of a series of settling basins.  Sludge  disposal  is  to  an
off-site  landfill.   After pH adjustment, the final discharge  is to a
POTW.

Plant #502 manufactures both organic and inorganic pigments, of   which
chrome pigments are a small part.  Treatment consists of pH adjustment
prior to discharge.

Plant  #436  manufactures  several   chemicals  in  addition  to chrome
pigments.   The   treatment  system   consists  of  neutralization  with
caustic  and  clarification  in  settling  lagoons prior to discharge.
Sludge is contract-hauled approximately once every three years.

Plant #409 manufactures specialty chemicals  and  inorganic  pigments.
The  existing wastewater treatment facility consists of S02 reduction,
clarification, filters and pH adjustment.  Sludge disposal  is  to  an
off-site location.

Plant  #997  manufactures chromic oxide  and sulfuric acid.  Production
data is not available.  The  existing  wastewater  treatment   facility
consists of pH adjustment, S02 reduction and lagoons.

Plant  #962  manufactures  inorganic pigments   (chrome  yellow).  The
existing   waste   treatment   plant  consists    of    flocculation,
clarification,  and  filters.   After  pH  adjustment,  the effluent is
discharged to a POTW.  Sludge is recycled  to process.
                                   527

-------
Plant  #200  manufactures  small  quantities   of   chrome   pigments.
Treatment is unknown.

In  summary,  a  review  of the existing treatment system descriptions
indicates that the prevailing treatment practices appear  insufficient
except for the system at Plant #894.  The major problems besides total
lack  of  treatment  is  lack  of  sufficient  residence time, lack of
critical treatment units, and failure to collect  all  waste  streams.
As  previously  stated,  only  Plant  #894 has a properly designed and
operated treatment system.   This system is basically the same  as  the
Level 1  treatment system shown in Figure 16-10.

Advanced Treatment Technologies

The   treatment  technologies  in  use  in  the  industry  consist  of
segregation, equalization,   S02  reduction,  alkaline  neutralization,
clarification,   and   filtration.    In   addition,   the   following
technologies were  reviewed  for  model  plant  development:   sulfide
precipitation,   ion  exchange,  reverse  osmosis,  and  the  xanthate
process.

Selection o£ Appropriate Technology and Equipment

Technologies for Different Treatment Levels

A careful review of the end-of-pipe  treatment  methods  available  to
industry  was  made.   As  a result, the following methodology was set
forth as the only treatment level for this  subcategory.   Originally,
sulfide  precipitation  was  considered  as  a level 2 addition to the
level 1   technology.   However,  an  Agency  treatability  study   (61)
indicated  no  significant  improvement  in  pollutant  reduction  was
achieved  by  this  addition.   The  result  of  eliminating  level   2
technologies  from  consideration   is  to  set NSPS standards equal to
those for BPT/BAT.  Considerations made in establishing  the  level   1
model included:

     Effective reduction of pollutants.

     Established treatment practices in the industry.

     The cost of technology.

     The adaptability of the model  to different situations.

A.   Level  1  (BPT/BAT/NSPS/PSES/PSNS)

     Consists of equalization, S02  reduction,  alkaline  precipitation,
     clarification,  and  filtration.   A  flow diagram of the  level  1
     system is shown in Figure 16-10.
                                    528

-------
in
to
                         I
                          SULFURIC
                         I   ACID
                                                      BACKWASH
                                 CAUSTIC SOD A
RAW '
1 .
WASTE WATER 1 '
L
•
j
\

SULFUR
DIOXIDE
.1 .



i
	 PO

I   HOLDING TANK      REACTION MIX
                       TANK   TANK


I               ^^

w	inmiiiK-t
                 I        SLUDGE
                 f         TANK

            TO LANDFILL

 *
  Includes flow monitoring, pH monitoring and sampler.
                                                                                             ADJUSTMENT
                                                                            D - »-
                                                                           ^EFFLU
                                                                                                      EFFLUENT
                                                                                  FILTER
                           Figure 16-10.   Level 1 waste water treatment for chrcme pigments.

-------
B.   Level 2

     Sulfide treatment addition to BPT system has  been  dropped  from
     consideration  as  a  level 2 treatment.  For a discussion of the
     proposed  system  and  the  cost  evaluation  see  the   Proposed
     Development  Document  (60).   Figure  16-11  is a diagram of the
     proposed system.

Equipment for Treatment

A.   Equipment Functions

     In the model treatment system, the incoming wastes are  acidified
     in  a  holding tank and then treated with sulfur dioxide solution
     in  a  reactor  to  convert  hexavalent  chromium  to   trivalent
     chromium.   Caustic  soda,  lime,  or other neutralizing agent is
     then added to neutralize the acid  and  precipitate  heavy  metal
     hydroxides  and a polymeric coagulant is added to help settle the
     heavy metal hydroxides in a clarifier.  The settled  effluent  is
     then  filtered  in  a  dual-media  filter and discharged after pH
     adjustment to the range 6 to 9.

B.   Chemicals and Handling

     Sulfuric acid and caustic soda solutions  are  common  industrial
     chemicals  which  are  readily  handled  with conventional liquid
     feeding equipment.  Sulfur dioxide is received  as  a  compressed
     gas  which  is dissolved in water by modified gas chlorinator and
     fed to the reactor to maintain  consistent  reducing  conditions.
     Polymer  is fed by a standard package of holding tank, mixer, and
     feeder.  With normal precautions there are no unusual hazards  in
     handling chemicals for treatment of chrome pigment wastes.

C.   Separation and Disposal of Solids

     Solids from the clarifier, including recirculated filter backwash
     solids are dewatered in a filter press and hauled to  a  chemical
     landfill.   Sludge  filtrate  is returned to the influent holding
     tank.

D.   Monitoring Requirements

     Internal process monitoring consists  of  maintaining  proper  pH
     levels in the holding tank and final effluent, using conventional
     field  equipment.   A  reducing  environment is maintained in the
     reactor, using an oxidation-reduction potential instrument and/or
     analysis for excess S02.  Periodic effluent analyses for chromium
     and heavy metals should be made on composite  samples  by  atomic
     absorption  methods,  for  official  reporting purposes.  Sulfide
     monitoring is generally unnecessary  because  dissolved  sulfides
     should  not  exist  in  the  presence  of excess ferrous iron and
     oxygen.
                                    530

-------
                                                FERROUS SULFATE  SODIUM BISULFIDE

                                                                  0
                                       BACKWASH
                                                CAUSTIC
                                                I  SODA
   RAW
i
WASTE WATE
r
                                   SULFUR
                                   DIOXIDE
             | HOLDING TANK

             I
1

1





1


                                                      POLYMER

                                                       J
                      REACTION   MIX
                        TANK     TANK
             U  -
               T
                                       SLUDGE
                           TO LANDFILL
               Includes flow monitoring, pH monitoring and sampler.
                                                                                            EFFLUENT
                   Figure 16-11.  Level 2 waste water  treatment  for chrome pigments.

-------
Treatment Cost Estimates

General Discussion

To prepare cost estimates, a model plant  concept  was  developed  and
plant criteria developed for BPT (Level 1).

A.   Wastewater Flow

     The data for five plants with usable flow data is  summarized  in
     Table  16-4.   This information was used on a production weighted
     basis to determine  the  average  flow  in  the  industry.   This
     average  was  computed  to  be 105 mVkkg (25,200 gal/ton).  This
     value was used for sizing the model plants.

B.   Chromium Pigment Production

     Production in the chrome pigment subcategory ranges from a low of
     500 kkg/year to a high of  approximately  18,000  kkg/year.   The
     mean production is approximately 7200 kkg/year.  For the purposes
     of  estimating  treatment  costs,  four  production  levels  were
     selected  as  model  plants.   These  are   1500  kkg/year,   4000
     kkg/year,  6000  kkg/year,  and  18,000 kkg/year.  These cover the
     entire range of  production  rates.   Most  plants  produce  many
     chrome  pigment products on a continuous basis so the operational
     mode selected was continuous and assumed  to  run  350  days  per
     year.    Chrome  pigments  are   usually  produced  in  integrated
     facilities with the  necessary   flexibility  to  shift  from  one
     product  or  combination of products to another.  The model plant
     was selected to reflect this type of complexity.

C.   Wastewater Pollution Load

     For the model plants, the loads  are based on  verification  plant
     data.   This  data   indicated  an  average  loading  of 16 kg/kkg
     chromates as chromium (Table 16-8).  Total  toxic metals  loadings
     ranged  from  12  kg/kkg  to  47  kg/kkg.   Total suspended solid
     loadings ranges from 55 kg/kkg to 130 kg/kkg (Table  16-8)..   The
     overall solid waste generation is expected  to be 85 kg/kkg to 150
     kg/kkg  (dry solids).  For the purpose of determining solid waste
     generation, a value of 105 kg/kkg (dry solids) was selected.

     The estimated costs of the BPT system for   the  four  models  are
     given  in  Tables   16-10,   16-11,  16-12,   and  16-13.  For these
     models, both hydraulic and pollution  loads  per unit of production
     were held constant over the entire range of production.

     In order to determine the accuracy of  the  treatment  model,  an
     attempt  was  made   to  compare  the  model  costs against actual
     industry costs.  Cost data were  received on two plants, one  with
     treatment   installed and one in  the design  stage.  No attempt was
     made to compare costs item by  item since these specific costs may
     differ for  the following reasons:
                                     532

-------
     Variations  in  land  costs.

     Variations  in  hydraulic  loading.

     Varying  costs  of  solid waste  disposal.

     Modifications  to  the existing wastewater  treatment  facility  and
     other  related  costs

     The following  overall results were obtained:
             Model  Plant
             Plant  #002
             Plant  #894
Annual Costs
   ($/kkg)

    89.75
    85.38
    91 .03
     The  above  data  indicate  a  fair correlation between the model
     plant and site specific cost estimates.

     Table 16-14 presents a summary  of  the  unit  cost  distribution
     between  amorization,   operation,  and maintenance cost components
     at various production levels.

     For the model plant, the primary sources of wastewater  are  from
     product  washings,  slurrying  of  reaction products, scrubbing of
     reactor vent gases, and  washing  of  equipment  due  to  product
     changes.

Model Plant Costs

The  major  costs  for  the  Level  1 (BPT) model plant are equipment,
labor,  and  chemical  costs.   Engineering   design   and   equipment
maintenance are also fairly large.  The majority of the annual cost is
tied  up  in  operation  and  maintenance.   This cost can approach 50
percent of the total capital cost.

Basis for Regulations

Evaluation of BPT Treatment Practices

A number of factors are anticipated to contribute to a wide  variation
in  the effluent quality at chrome pigment plant treatment facilities.
Consideration  of  these  variations  is  included   in   establishing
limitations  in that the performance of the plant on which limitations
are based is a large  complex  plant  that  encounters  all  of  these
factors.  These include the following:
                                    533

-------
   TABLE  16-10-  MODEL PLANT TREATMENT  COSTS
Subcategory
Production
Chrome Pigments
1,500 metric tons per year
A. INVESTMENT COST

     Site development  	
     Equipment  	
     Monitoring  equipment  	

          Subtotal  ..............
     Contractor's 0 &  P b.	

          Subtotal  	
     Engineering  	,	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	•

     TOTAL INVESTMENT  COST  	


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  	
     Energy	
     Chemicals  	
     Maintenance  	
     Taxes and  insurance  	
     Residual waste disposal  ....
     Monitoring, analysis
      and reporting 	

     TOTAL OPERATION AND
     MAINTENANCE  COST
                                                   ($)
                               BPT

                               600
                            317,000
                             20,000

                            337,600
                             50,640

                            388,240
                             77,648

                            465,888
                             46,589

                            512,477
                              6,000

                            518,477
                            112,000
                              7,500
                             53,000
                             51,248
                             15,554
                              5,000

                             15,000
                            259,302
BAT

  0
  0
  0

  0
  0

  0
  0

  0
  0

  0
  0
C. AMORTIZATION  OF
   INVESTMENT  COST

   TOTAL ANNUAL  COST
                             83,380

                            342,682
a Represents  the  incremental  cost above that for BPT treatment
  Overhead and Profit
                              534

-------
   TABLE  16-11.  MODEL PLANT TREATMENT  COSTS
Subcategory    Chrome Pigments
Production     4,000 metric  tons  per  year
A. INVESTMENT COST                             BPT          BAT a

     Site development  	            1,200            0
     Equipment  	          564,000            0
     Monitoring  equipment  	           20,000            0

          Subtotal  	          585,200            0
     Contractor's 0 &  p b.	           87,780            0

          Subtotal  	          672,980            0
     Engineering  	          134,596            0

          Subtotal  	          807,576            0
     Contingencies  	           80,758            0

          Subtotal  	          888,334            0
     Land	           12,000            0

     TOTAL  INVESTMENT  COST 	          900,334            0
B.  OPERATION  AND
    MAINTENANCE  COST

      Labor  and  supervision  	          112,000            0
      Energy 	           15,000            0
      Chemicals  	          141,500            0
      Maintenance  	           88,833            0
      Taxes  and  insurance  	           27,010            0
      Residual waste  disposal  ....           15,000            0
      Monitoring ,  analysis
      and  reporting  	           15,000            0

      TOTAL  OPERATION AND
      MAINTENANCE  COST                     414,343            0
C.  AMORTIZATION OF
    INVESTMENT  COST                        144,532            0

    TOTAL  ANNUAL COST                       558,875            0

f* Represents the incremental  cost above that for BPT treatment
  Overhead and Profit
                              535

-------
    TABLE 16-12.  MODEL PLANT TREATMENT COSTS
Subcategory
Production
Chrome Pigments
6,000 metric tons per year
A.  INVESTMENT  COST
                               BPT
                                                      ($)
BATa
     Site development  	
     Equipment  	
     Monitoring  equipment  	

          Subtotal	
     Contractor's  0 &  P  .	

          Subtotal  	
     Engineering  	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL  INVESTMENT  COST 	


B. OPERATION AND
   MAINTENANCE  COST

     Labor  and  supervision 	
     Energy 	
     Chemicals  	
     Maintenance  	
     Taxes  and  insurance 	
     Residual waste disposal  ....
     Monitoring,  analysis
      and reporting 	

     TOTAL  OPERATION AND
     MAINTENANCE  COST
                             1,200
                           738,000
                            20,000

                           759,200
                           113,880

                           873,080
                           174,61*5

                         1,047,696
                           104,770

                         1,152,466
                            12,000

                         1,164,466
                           112,000
                            20,000
                           211,500
                           115,247
                            34,934
                            20,000

                            15,000
                           528,681
C. AMORTIZATION  OF
   INVESTMENT COST

   TOTAL ANNUAL  COST
                           187,506

                           716,187
a Represents  the  incremental  cost  above that for BPT treatment
  Overhead and Profit
                              536

-------
    TABLE 16-13.  MODEL  PLANT  TREATMENT COSTS
Subcategory
Production
               Chrome Pigments
               18,000 metric tons per  year
A. INVESTMENT COST

     Site development  	,
     Equipment  	,
     Monitoring equipment  ..,

          Subtotal  	,
     Contractor's O & P ^...,

          Subtotal  	,
     Engineering  	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL INVESTMENT  COST .


B. OPERATION AND
   MAINTENANCE  COST

     Labor and  supervision .
     Energy  	
     Chemicals  	
     Maintenance  	
     Taxes and  insurance  ...
     Residual waste disposal
     Monitoring,  analysis
      and reporting  	
                                                     ($)
                                               BPT

                                             1,800
                                         1,700,000
                                            20,000

                                         1,721,800
                                           258,270

                                         1,980,070
                                           396,014

                                         2,376,084
                                           237,608

                                         2,613,692
                                            18,000

                                         2,631,692
      TOTAL OPERATION AND
      MAINTENANCE COST
                                           112,000
                                            28,000
                                           635,000
                                           261,369
                                            78,951
                                            60,000

                                            15,000
                                         1,190,320
BAT

  0
  0
  0

  0
  0

  0
  0

  0
  0

  0
  0
 C.  AMORTIZATION OF
    INVESTMENT COST

    TOTAL ANNUAL COST
                                           425,248

                                         1,615,568
a  Represents the incremental cost above that for BPT treatment
b
   Overhead and Profit
                              537

-------
A.   Product Changes

     Changes in products require that equipment be thoroughly  cleaned
     prior  to reuse.  Therefore, frequent product changes will result
     in higher waste flows.

B.   Product Application

     The final disposition of the  product  will  affect  the  quality
     required.   The  higher  the quality, the more water required for
     rinsing.

C.   Air Pollution Control

     Equipment will be required in  many  cases  for  control  of  the
     environment  as  well as off-site air compliance.  Scrubbers will
     add some waste flow to the treatment system.  This flow, however,
     is generally small.

D.   Other Related Products

     Many plants manufacture other types of  pigments  including  iron
     blues  and organic pigments.  These products generate significant
     quantities of wastewater which  tend  to  dilute  chrome  pigment
     wastes.    However,   these  wastewaters  were  included  in  the
     computation of the  unit  waste  flow.   Therefore,  the  use  of
     parallel   treatment  for  existing  facilities  producing  other
     pigments is not required at this time as long as chromium pigment
     production is  the  majority  of  the  overall  production.   The
     following   guidelines   should   be   used   in  applying  these
     regulations:

     1.   When determining the effluent loadings, the total production
          of a facility will be used as long  as  the  chrome  .pigment
          production is in the majority.

     2.   When the chromium production is the minority of the  overall
          production,   the   total  production  should  be  used  for
          computing  the   effluent   limits   under   the   following
          conditions:   the  remaining  production   (other than chrome
          pigments)  generates  a  wastewater  containing  significant
          amounts  of  toxic  metals which will be removed by a chrome
          pigment treatment system.

     3.   For  those  facilities   (existing  sources)  where  chromium
          production  is   in  the  minority  and the wastes from other
          sources do not   contain  metals  above  accepted  levels  of
          treatability, segregation and parallel treatment of chromium
          pigment  wastes  are  recommended,.   However, the permitting
          authority or POTW  must  consider  the  following  balancing
          factors:

          a.   The economic impact on the facility balanced against
                                     538

-------
     TABLE 16-14 MODEL PLANT UNIT TREATMENT  COSTS
  Subcategory Chrome Pigments
                             Annual  Treatment  Costs ($/kkg)
                                   LEVEL OF  TREATMENT

 COST ITEM      PRODUCTION                  BPT
                  (kkg/yr)
Annual Operation
and Maintenance    1,500                  172.87
                   4,000                  103.59
                   6,000                   88.11
                  18,000                   66.13

Annual
Amortization       1,500                   55.59
                   4,000                   36.13
                   6,000                   31.25
                  18,000                   23.62

Total Annual
Cost               1,500                  228.45
                   4,000                  139.72
                   6,000                  119.36
                  18fOOO                   89.75
                          539

-------
          b.    The environmental benefits of parallel treatment.

     In addition to the above factors,  the design and operation of the
     treatment facilities affect effluent quality.   Important  factors
     are  equalization,  S02 contact time and pH depression,  S02  dose,
     proper neutralization,  and adequate solids removal.

     Table 16-15 is a summary of verification sampling  and  long-term
     effluent  monitoring  data at Plant #894 for the major pollutants
     of concern.  Plant #002 sampling results are  excluded  from  the
     subcategory  performance  evaluation,  since the treatment system
     was not functioning properly as previously discussed.

     The long-term monitoring data in Table 16-15 is for  the  maximum
     30-day average long-term monitoring results.  Sufficient data was
     not available to estimate long-term daily maximum values.

     Plant  #894 is the only known plant with Level 1 treatment system
     installed  and  operating.   Table  A-1la   sets   forth   means,
     variability  factors, and the maximum 30-day average performance.
     Maximum daily performance was not  computed  since  the  discrete
     sampling  data  was  not available at the time of the evaluation.
     The performance evaluation in Table  16-5  is  utilized  for  the
     development of regulations for TSS and applicable toxic metals.

     As  previously  stated,  only one plant of the existing twelve is
     known to have a Level 1 treatment system installed.    This  plant
     represents  approximately 20-30 percent of total production.  One
     other  plant,  with  a  different  treatment  system,   meets  the
     promulgated  BPT/BAT  limitations.   Most  other plants have some
     type of treatment installed, but  none  of  these  appear  to  be
     adequate.  This technology is expected to remove 3,450,000 pounds
     per year of toxic metals.

Basis for BPT Effluent Limitations

A.   Technology Basis

     For  BPT,  the  Agency  is  promulgating  limitations  based   on
     equalization,   reduction  of  hexavalent  chromium  followed  by
     alkaline precipitation, and dual-media filtration.  Reduction  of
     flow  by the methods given in this section was considered but not
     used since their application is site specific.  However, they are
     quite  viable  options  in  most  cases  and  could   result   in
     substantial treatment cost savings.

B.   Flow Basis

     The basis of flow for the BPT limitations is estimated from  data
     provided  in  the  308-Questionnaires  and  plant  visits  during
     sampling.  Table  16-4 presents the plant flow data used  for  the
     purpose  of  regulation.   A weighted average flow was determined
     based on plant production.  Plants producing a  greater  quantity
                                   540

-------
TABLE 16-15.  SUMMARY OF LONG TERM AND VERIFICATION EFFLUENT SAMPLING
              RESULTS AT PLANT #894

SUBCATEGORY:
Pollutant

Total Suspended
Solids, TSS
Iron
Antimony
Arsenic
Cadmium
Chromium
Copper
Lead
Mercury
Nickel
Zinc
Cyanide (CN-A)
Cyanide (Total)
Chromium (VI)
(1) ND, Not Detected.
(2) NAr Not Available.
CHROME PIGMENTS
Verification Sampling
(mg/1) (kg/kkg)

3.9
0.30
0.30
ND(1)
0.0084
0.33
0.035
0.11
ND
0.021
0.058
0.065
0.0067
0.023



0.66
0.051
0.051
ND
0.0014
0.056
0.0060
0.019
ND
0.0036
0.0099
0.011
0.0011
0.0039


Achievable Performance ^
Max 30-day Avg
(mg/1) (kg/kkg)
23 3.8
NA NA
NA NA
0.16 0.027
0.12 0.020
0.73 0.12
0.25 0.42
0.87 0.15
0.0016 0.00027
NA NA
0.074 0.013
0.068 0.012
0.31 0.053
0.30 0.051


(3) From Tables 16-6 and 16-9.
(4) Fran Table A-lla,
"Historical
Effluent MDnitt
Dring Data Summary."
                                   541

-------
     of  chrome  pigment product have a waste flow which has a greater
     influence on the average flow calculation.   This approach for the
     determination of the average flow is substantiated  by  the  unit
     waste flow which is related to the plant production rate.

     Since  plants in the chrome pigments subcategory do not segregate
     wastewaters from the various pigment processes for treatment, the
     basis of flow for the purpose of regulation includes all  process
     related  wastewater  combined.  The flow basis is 105 mVkkg from
     Table 16-4.  This flow does not include any recycle or  reuse  of
     wastewaters other than some incidental recycle being done at five
     plants included in the data base.

C.   Selection Basis for Pollutants to be Regulated

     The selection of pollutants for which specific numerical effluent
     limitations are set was based on an evaluation of raw waste  data
     from  the screening and verification sampling program.  Pollutant
     data from  the  plants  sampled  during  screening  was  used  to
     determine  the  need  for  verification  sampling.   Verification
     sampling at Plants 1002 and #894  provided  additional  pollutant
     raw  waste  concentration  data needed to assess the magnitude of
     the pollution potential.

     Results of the screening and verification sampling are  tabulated
     in this section for the raw process waste streams.  The pollutant
     concentration  listed  under  verification  is  the highest value
     observed during sampling at the two plants visited.

     Toxic  pollutants  are  listed  based  on   their   presence   at
     significant  concentration  levels  during  sampling.  Pollutants
     from  this  list  were  considered  for   regulation   if   their
     concentrations  appeared  to  equal  or  exceed  in  at  least one
     instance the  lowest  level  estimated  as  treatable  using  any
     available  technology  appropriate  for  their  removal, ignoring
     economic considerations.

     The relative significance of the toxic pollutants  was  estimated
     based on the total annual raw waste load for each pollutant which
     appears  in  a  Table  in this section.  The total annual load is
     based on the average concentration observed during screening  and
     verification which is tabulated in Table 16-8, in addition to the
     estimated  annual  production  of  73,500  kkg of product for the
     industry.

     Specific numerical effluent loading limitations had been proposed
     for those pollutants  which  appeared  at  average  concentration
     levels   (Table  16-7) considered to be treatable for at  least one
     plant visited during sampling.

     On the basis of concentration and total annual  raw  waste   loads
     determined   during   sampling,  chromium,  zinc,  lead,  copper,
     antimony, cadmium, nickel and mercury  have  been  identified  in
                                    542

-------
     chrome  pigment  wastes at treatable levels (mercury is treatable
     by sulfide precipitation which was  rejected  as  the  basis  for
     BAT).    These metals were all considered for regulation,  however,
     final  regulations include limitations on  chrome  (total),   zinc,
     and  lead only.  Organic pollutants and cyanide are not included,
     since   they  are  considered  products  of  iron  blue, '  organic
     pigments,  or  HCN  production  as discussed above.  In addition,
     these  parameters will be covered by future regulations  in  other
     subcategories.

     The Agency's  decision  to restrict metals regulation to chrome,
     lead,   and  zinc  was  made   in   light   of   the   performance
     characteristics of the treatment technology for this subcategory,
     and  the  nature  of the waste involved.  The performance of this
     technology is such that control  of  certain  "key"  metals  will
     result  in  sufficient  control of the remaining metals.   This is
     discussed in detail in Section 8.

     The  treatment  technology  designed  for  removal  of  the  more
     prominent  metals  in  this  subcategory, namely chrome,  zinc and
     lead,  will ensure control of  all  the  toxic  metals  originally
     considered   for   regulation.    Performance   data   from  this
     subcategory substantiates this (Tables 16-15 and 16-16).

     The Agency is also aware that specific plants may have  unusually
     high  loadings  for  one  or  more of the unregulated metals.  In
     these  instances, limitations should be set  on  a  case  by  case
     basis.   Table  16-16  includes  guidance  for  copper, antimony,
     nickel, cadmium, and mercury concentrations for those cases where
     additional control is deemed necessary.

     Hexavalent chromium was excluded from consideration in the  final
     regulations.   The  complexity  and  subsequent  accuracy  of the
     analysis may cause misleading conclusions if hexavalent  chromium
     is  used  as  an  effluent  monitoring parameter.  Sulfur dioxide
     reduction  under  acidic  conditions  will   convert  . hexavalent
     chromium to its trivalent form which can be conveniently verified
     by  analysis of total chromium in the treated effluent.  Chromium
     cannot be removed by alkaline precipitation unless it is  in  the
     trivalent  form.   Therefore,  if the S02 reduction step fails to
     reduce the hexavalent chromium it will  become  apparent  in  the
     effluent total chromium concentration.

Basis of Pollutant Limitations

A.   Conventional and Nonconventional Parameters

     1 .   pH

          The treated effluent is to be controlled within the range of
          6.0 to 9.0.  This limitation is based on the data  presented
          in  Appendix B of the proposed Development Document (60) and
          the JRB Study (52).
                                     543

-------
     2.    Total Suspended Solids (TSS)

          Review of the long-term monitoring and verification sampling
          data in Table 16-15 indicates a maximum 30-day  average  TSS
          discharge  of  3.8  kg/kkg for the purpose of the limitation
          determination.   The 30-day average  concentration  basis  is
          then determined as follows:

          (3.8 kq/kkq)  (1000 mq/1)  = 36 mg/1
          (105 mVkkg)  (kg/m*)

          The  24-hour  maximum loading limitation is determined by the
          following relationship:

          Maximum 30-day   X VFR = 24-hour maximum
          average loading          loading or
          or concentration         concentration

          The variability factor ratio (VFR)  is  estimated  from  the
          Titanium Dioxide Sulfate Process subcategory based on 30-day
          average   and  daily  variability  factors  for  zinc.   The
          long-term monitoring  data  on  zinc  showed  daily  average
          concentrations  ranging  from  0.010  to  1.14 mg/1 during a
          period of more than two years (Tables A-9a-l and  A-9c-l  in
          Appendix A).   This range of values for zinc nearly spans the
          observed  range  of  toxic metal concentrations found in the
          effluent from Chrome Pigments Plant $894 (Table 16-15).  The
          VFR of 2.1 for zinc in the Ti02 Sulfate Process reflects the
          overall metal removal performance of alkaline  precipitation
          followed  by  settling and discharge without filtration.  The
          choice of VFR is also supported by a statistical  evaluation
          of  data  developed  in  the Agency Treatability Study  (61).
          For  the  chrome  pigments  subcategory,  the   study   data
          indicated a VFR of 2.3 for chrome concentrations in treated,
          but  unfiltered  wastes from a level 1 system.  Therefore, a
          VFR of 2.4 is applied to the Chrome Pigments industry  as  a
          conservative   estimate  of  the  performance  of  a  similar
          treatment technology which does include a  final  filtration
          step.  Therefore, the 24-hour maximum limitation becomes,

          (3.8 kg/kkg)  (2.-4) = 9.1  kg/kkg

     3.    Other Pollutants

          The concentration basis for iron is also presented in  Table
          16-15.   This concentration is intended to serve as guidance
          in cases where iron is found to be of serious concern.

B.   Toxic Pollutants

     The  effluent  limitations  proposed  for  the   selected   toxic
     pollutant  control  parameters  are derived from three sources of
     information including  1)  screening  and  verification  sampling
                                     544

-------
                      TABLE 16-16.  EFFLUENT LIMITATIONS
                               Chrome Pigments                    A
           Best Practicable Control Technology Currently Available
                        Waste Water Flow:  105 m3/kkg

                                        Concentration Basis   Ef f luent Limit
                                              (mg/1)            (kg/tog)
Subcategory Max.
Performance / - , 30-day 24-hr
Pollutant (mg/1) VFRU' Avg. Max.
Conventional and Nonconventional Pollutants
Total Suspended ,-.
Solids, TSS 18.1 4.8/2.0UJ 36 87
Iron 0.30(4) 4.4/1.8(6) 0.54 1.3
Max.
30-day 24-hr
Avg, Max.
3.8 9.1
Toxic Pollutants
Chromium
Zinc
Lead
Antimony
Cadmium
Copper
Nickel
0.71
0.71(5)
0.66
0.40 (7>
0.13
0.21
0.30<8>
4.1/1.7
4.1/1.7
5.0/2.1
4.4/1.8(6)
3.8/1.6
4.6/1.9
4.4/1.8(6)
1.2
1.2
1.4
0.72
0.20
0.41
0.54
2.9
2.9
3.4
1.7
0.49
0.99
1.3
0.13
0.13
0.15
—
—
—
—
0.31
0.31
0.36
—
—
—
—
(1)  Subcategory long-term average concentrations adjusted for model plant flow,
     unless otherwise indicated.
(2)  VFR:  Ratio of the 24-hour variability factor to the 30-day average
     variability factor.
(3)  VFR of 2.4 from long-term evaluation in Titanium Dioxide Subcategory,
     verified by analysis of chrome pigments subcategory results from an
     Agency Treatability Study (61).  30-day variability factors from Table
     A-11A in Appendix A unless otherwise specified.
(4)  Verification sampling results based on three 24-hour composite effluent
     samples, adjusted for model plant flow.
(5)  long-term averages for zinc, which include zinc chromate production are
     not available; therefore, limitations are set equal to those for chrome.
(6)  Variability factors based on average of those for other parameters since
     no long-term data available.
(7)  Lower limit of treatability estimate (Table 8-11).
(8)  Estimated achievable long-term average (Table 8-13).
(*)  Conventional and nonconventional pollutant limitations apply also to NSPS.
     Toxic pollutant limitations apply also to BAT, PSES, and PSNS.
                                      545

-------
data, 2} literature based treatability estimates (Section 8), and
3) a limited amount of long-term monitoring data at Plant #894.

The  sampling results represent plant performance observed during
three days of sampling.  The sampling data was used primarily  to
select the pollutants of concern, and in the case of antimony and
nickel  the  sampling  results  were  used to estimate the 30-day
average concentration in view of the lack of long-term monitoring
data for these two pollutants.

The sampling data for Plant #894 appears to demonstrate  that  in
some   cases  the  effluent  quality  for  metal  pollutants  are
considerably  better  for  BPT  treatment   than   indicated   by
literature  treatability  in  Section  8.   This  high  degree of
incidental removal supports the contention that applying effluent
limitations  just  to  the  dominant  metal  pollutants,  assures
effective control of the other metals.

The   VFR   used   to  determine  the  proposed  24-hour  maximum
limitations is based on long-term data for zinc in  the  Titanium
Dioxide   subcategory,   which   is   supported  by  analysis  of
treatability results for this subcategory (61).

1.   Chromium

     The raw waste concentration for  chromium  was  observed  as
     high  as  370  mg/1  and  averaged  150 mg/1 during sampling
     (Table 16-7).  The long-term monitoring results  indicate  a
     maximum 30-day average discharge of 0.13 kg/kkg which is the
     limitation basis.  The concentration' basis then becomes:

     (0.13 kq/kkq) (1000 mq/1) =1.2 mg/1
     (105 mVkkg)  (kg/m3)

     The 24-hour maximum is determined as follows:

     (0.13 kg/kkg) (2.4) = 0.31 kg/kkg

     where the VFR is set equal to 2.4 based on data from the
     Ti02 subcategory and the Treatability Study (61).

2.   Zinc

     Zinc limitations were set equal to  chromium.   Tables   16-7
     and 16-9 indicate that the removals of zinc and chromium are
     similar  at  Plant #002 where zinc  is found at very high raw
     waste concentrations.

3.   Lead

     The raw waste concentration for lead was observed  as high as
     69 mg/1 and averaged 32 mg/1 during sampling.  The long-term
     monitoring  results  indicate  a  maximum   30-day  average
                                  546

-------
     discharge of 0.15 kg/kkg which is used as the 30-day average
     limitation.   The concentration basis then becomes:

     (0.15 kg/kkq)(100 mq/1)  =1.4 mg/1
     (105 mVkkg) (kg/m3)

     The 24-hour maximum limitation then becomes:

     (0.15 kg/kkg) (2.4) = 0.36 kg/kkg.

4.    Copper

     The raw waste concentration for copper was observed as  high
     as  6.2  mg/1  and  averaged  4,3 mg/1 during sampling.  The
     long-term  monitoring  results  indicate  a  maximum  30-day
     average   discharge   of   0.042   kg/kkg.   Therefore,  the
     concentration basis becomes:

     (0.043 kq/kkq) (1000 mq/1) = 0.41 mg/1
     (105 mVkkg)   (kg/in3)

5.    Antimony

     The raw waste concentration for  antimony  was  observed  as
     high  as  7.7  mg/1  and  averaged 3.3 mg/1 during sampling.
     Since no long-term data is available,  the  lower  limit  of
     treatability  (Table 8-11) for antimony is used to determine
     the concentration basis.  This then becomes:

     (0.4 mg/1)  (1.8) = 0.72 mg/1

     where 1.8 is an estimated 30-day average variability  factor
     for  antimony,  obtained  by  averaging  known  factors  for
     pollutants  in the subcategory (Table 16-16).

6.    Cadmium

     The raw waste concentration for cadmium was observed as high
     as  1.3 mg/1 and averaged 0.62  mg/1  during  sampling.   The
     long-term  monitoring  results  indicate  a  maximum  30-day
     average discharge of 0.021 kg/kkg.  The concentration  basis
     then becomes:

     (0.021 kq/kkq) (1000 mq/1) =0.20 mg/1
     (105 mVkkg)       (kg/m3)

7.    Nickel

     The  raw waste concentration for  nickel was observed as high
     as  0.74 mg/1 and averaged 0.17 mg/1  during  sampling.   The
     verification   sampling   results    indicate  an  achievable
     concentration of 0.021  mg/1 which compares  to  an  industry
                                 547

-------
          long-term average of 0.30 mg/1 (Table 8-13).  Therefore, the
          concentration basis becomes:

          (0.30 mg/1) (1.8) = 0.54 mg/1

          where, again,  1.8 is an estimated 30-day average variability
          factor for nickel in this subcategory.

          The limitations are summarized in Table 16-16 for BPT.

Basis for BCT Limitations

While   EPA  has  not  yet  proposed  or  promulgated  a  revised  BCT
methodology in  response  to  the  American  Paper  Institute  v.  EPA
decision  mentioned  in Section 3, EPA is promulgating BCT limitations
for this subcategory.  These limits are identical to  those  for  BPT.
EPA  is  not promulgating any more stringent limitations since we have
identified  no  technology  option  which  would  remove   significant
additional  amounts of conventional pollutants.  As BPT is the minimal
level of control required by law, no possible application of  the  BCT
cost   tests   could  result  in  BCT  limitations  lower  than  those
promulgated in this regulation.  Accordingly, there is no need to wait
until  EPA  revises  the  BCT  methodology  before  promulgating   BCT
1 imitations.

Basis for BAT Effluent Limitations

A.    The Application of Advanced Level Treatment

     Utilizing the cost estimates presented in this report, the Agency
     has analyzed the cost effectiveness of  the  base  level  systems
     (BPT)  and  various  advanced  level  options  for  conventional,
     nonconventional,  and  toxic  pollutant  removal.   The  economic
     impacts  on  the  Chrome Pigments Industry have been evaluated in
     detail and taken into  consideration  in  the  selection  of  the
     technology basis for the BAT regulations.

     The  Agency  is  promulgating  BAT limitations based on treatment
     consisting of Level 1 technology which is equivalent to BPT.  The
     implementation of BPT/BAT will remove 2,400,000 pounds  of  toxic
     metals   annually.    The   two   plants  currently  meeting  the
     promulgated  BPT/BAT  limitations  aro  removing  an   additional
     900,000 pounds per year of toxic metals.

B.    Technology Basis

     For BAT, the Agency is utilizing the identical  technology  basis
     discussed  for  BPT  in this section.  BAT includes no additional
     treatment  because  the  Agency  has  concluded,  based  on   the
     Treatability  Study   (61)  results,  that sulfide treatment after
     alkaline precipitation does not significantly increase  treatment
     performance,  and,  therefore,  does not  justify costs associated
     with such additional treatment technology.
                                    548

-------
C.   Flow Basis

     The unit flow of 105 mVkkg is also used for BAT.

D.   Selection of Pollutants to be Regulated

     The basis of pollutant selection is discussed for BPT above.  For
     BAT, the toxic metals shown  in  Table  16-16  are  selected  for
     regulation.  These include chromium, zinc and lead.

E.   Basis of Pollutant Limitations

     The basis of the limitations are discussed in  detail  under  BPT
     above.

Basis for New Source Performance Standards

A.   Application of Advanced Level Treatment

     Chrome pigment industry  wastes  primarily  contain  toxic  metal
     pollutants which are particularly amenable to removal by alkaline
     precipitation  and  sulfide  precipitation.   Almost  all  plants
     combine  wastewater  from  the  chrome   pigment   process   with
     wastewater  from  unrelated  processes.  The Agency proposes that
     for new sources, the wastewater from the chrome pigments  process
     be  segregated  from  wastewater  from other processes unless the
     other wastewater contains toxic  metal  pollutants.   Segregation
     and  separate treatment of the wastewaters can conceivably reduce
     treatment costs, and simplify the   treatment  of  metals  without
     complications from unrelated wastewater constituents not amenable
     to metals treatment.
B,
 D
Technology Basis

For New  Source  Performance  Standards  (NSPS),  the  agency  is
setting  limitations  based  on  the  identical  technology basis
discussed for BPT above.  The Agency  also  recommends  that  all
unrelated  wastewater  sources  which  are not amenable to metals
treatment be segregated before treatment as previously discussed.
                                                               setting
Flow

The basis for the unit flow  used  for  the  purpose  of
limitations is 105 mVkkg and does not differ from BPT.

Selection of Pollutants to be Regulated
      The  same   conventional,   nonconventional,   and   toxic  pollutants
      selected    for   BPT  are  also   considered  here   for   the   NSPS
      limitations.  These include TSS,  pH,  iron,   and  the  same   eight
      toxic  metal pollutants.

      Basis  of  Pollutant Limitations
                                    549

-------
     The basis of the limitations are discussed in  detail  under  BPT
     above.

Basis for Pretreatment Standards

The  Agency  is  promulgating  PSES  and  PSNS  that  are equal to BPT
limitations because BPT provides better removal of chromium, lead, and
zinc than is achieved by a POTW and,, therefore, these toxic pollutants
would pass  through  a  POTW  in  the  absence  of  pretreatment.   In
particular,  1.4  million  pounds per year of hexavalent chromium from
existing sources would pass through in the  absence  of  pretreatment.
The  promulgated pretreatment regulations will remove over 1.3 million
pounds of hexavalent chromium per year.   Pollutants  regulated  under
PSES and PSNS are chromium, lead, and zinc.

Existing Sources

There  are currently eight indirect discharge chrome pigment plants in
the subcategory.  For  Pretreatment  Standards  for  Existing  Sources
(PSES),  the  Agency  is  setting  limitations  based on BPT described
above.  The pollutants to be limited are chromium, zinc, and  lead  as
presented in Table 16-16.

We are excluding small plants discharging less than 210,000 m3 process
wastewater  per  year  to  POTW from compliance with these PSES.  They
will be subject only to the general pretreatment standards in  40  CFR
Part 403.  The exclusion is intended to apply to plants producing less
than  2000  kkg per year, but we have established a flow basis for the
convenience of POTW's, since water use is much easier to monitor  than
production.  The annual flow basis for the exclusion was calculated by
multiplying  the  unit  flow,  105 mVkkg, by 2000 kkg per year:  (105
mVkkg) (2000 kkg/year) = 210,000 mVyear.   Plants  discharging  less
than  210,00 m3 process wastewater per year produce less than 2000 kkg
per year chrome pigments.  There would be  very  significant  economic
impacts  on  this  segment  of  the  industry if they were required to
comply with these PSES.  See the Preamble to the  Regulation  and  the
Economic   Impact  Analysis  of  Pollution  Control  Technologies  for
Segments of the Inorganic Chemicals Manufacturing Industry, EPA 440/2-
81-023.

New Sources

For Pretreatment Standards for  New  Sources  (PSNS),  the  Agency  is
setting  limitations  based  on NSPS.  The pollutants are indicated in
Table 16-16.
                                    550

-------
                              SECTION 17

                      HYDROGEN CYANIDE INDUSTRY
Industry Profile

General Description

Over 50 percent of the Hydrogen Cyanide manufactured  is  produced  by
the  Andrussow  process,  while  about 40 percent is a by-product from
acrylonitrile manufacture.  A major portion of the production is  used
in  the  manufacture  of  methyl  methacrylate,  plexiglas molding and
extrusion powders, and surface coating resins.  It is also used  as  a
fumigant for orchards and tree crops.  The industrial data profile for
this  industry is given in Table 17-1, while the status of regulations
prior to promulgation of this new regulation is given in Table 17-2.

Subcategorization

The method of  primary  Subcategorization  chosen  for  the  inorganic
chemicals  point  source category was subdivision by dominant product.
Other factors taken into consideration for Subcategorization included:
raw  materials  used,  manufacturing  process  employed,  geographical
location,   size   and   age   of  equipment  and  facility  involved,
non-water-quality aspects of waste  characteristics,  water  pollution
control  technology,  treatment  costs,  energy requirements and solid
waste disposal.  A detailed discussion of these factors  is  given  in
Section 4.

Hydrogen  cyanide  (HCN)  is  made  from  two different processes, the
Andrussow process and as a by-product  of  acrylonitrile  manufacture.
In  the Andrussow process, air, ammonia and natural gas are reacted to
produce the dominant product, hydrogen  cyanide.   Water-borne  wastes
from  the  process  consist  principally  of  ammonia  and sulfates in
addition to cyanide and nitriles.  The primary product  in  the  other
process  is  acrylonitrile  (CH2=CHCN)  and  the hydrogen cyanide is a
by-product.

The Hydrogen Cyanide Subcategory in this regulation is confined to the
Andrussow process.  Since hydrogen cyanide  is  a  by-product  of  the
acrylonitrile  manufacturing  process, this process will be covered in
the organic chemicals manufacturing category with the primary product.

General Process Description and Raw Materials

The raw  materials  are  reacted  at  elevated  temperature   {900-1000
degrees C) over a platinum catalyst.  The reaction is given as:

     2CH4 -f 2NH3 4 30Z  = 2HCN -f 6HZ0                        (1)
                                    551

-------
TABLE 17-1.              SUBCATEGORY PROFILE DATA SUMMARY

SDBCATEGORY        HYDRO2N CYANIDE*

Total subcategory capacity rate                      289,000 kkg/year
Total subcategory production rate                    166,000 kkg/year
No. of plants in this subcategory                      7
Plant age range:
              Minimum                                  5 years
              Maximum                                 30 years
308 Data** on file for                                 2
     With total capacity of                          179,000 kkg/year
     With total production of                        116,000 kkg/year
     Representing capacity                            62 percent
     Representing production                          70 percent
     Average production                               57,800 kkg/year
     Average capacity utilization                     65 percent
     Waste water flow per unit product
              Minimum                                 10 m3/kkg of HCN
              Maximum                                 57 m3/kkg of HCN
Sources of data are Stanford Research Institute, Directory of Chemical
Producers, U.S.A., 1977, U.S. Department of Conmerce, Current Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry ",June 1978 and "Economic Analysis of Proposed
Revised Effluent Guidelines and Standards for the Inorganic Chemicals Industry
March,  1980.
       * Includes data from plants using Andrussow Process and from plants
         recovering HCN as a byproduct from the manufacture of acrylonitrile.
       **Includes data from plants using Andrussow Process.
                                      552

-------
TABLE 17-2.   STATUS OF REGULflTICNS  -  EFFLUENT LIMITATION GUIDELINES
SUBPART
                    HYDROC2N CYANIDE

                    AP  (40 CFR 415.420, 5/22/75)
                                    STANDARDS
                     BPCTCA*
                 Max.
                          Avg.
Product
Process
          Parana
          eters
                BATEA

            Max.     Avg.
           kg/kkg   kg/kkg
            (mg/1)   (mg/D
     NSPS

 Max.       Avg.
kg/kkg     kg/kkg
 (mg/1)      (mj/1)
Andrussow
Process
          TSS
          CN
          CN(A)
          BGDC
                  2.4      1.2
                  (48.0)**  (24.0)
                  0.005
                  (1.0)

                  0.005
                  (0.1)

                  3.6
                  (72.0)

                  0.36
                  (7.2)
0.025
(0.5)

0.005
(0.05)

1.8
(36.0)

0.18
(3.6)
 Sections 415.420, 415.421, and 415.422 were revoked by the Agency
 ( 41 FR 10681, February 23, 1977).
 jyiax. = Maximum of any one day.
2
 Avg. = Average of daily values for thirty consecutive days shall not exceed.
**
   flow basis 50,000 1/kkg.
                                      553

-------
The source of methane is natural gas  containing  50  to   100  percent
methane by volume.  In addition to hydrogen cyanide, the reacted gases
contain  ammonia, nitrogen, carbon monoxide, carbon dioxide, hydrogen,
and small amounts of oxygen, as well as  traces  of  organic  nitriles
formed  from  non-methane  hydrocarbon components of natural gas.  The
reactor gases are cooled and then scrubbed in  one  of  two  processes
which  are  used  to  remove  the  unreacted ammonia.  In  one patented
process the gases are scrubbed with phosphate  liquor,  the  resulting
solution  is  decomposed  and  the phosphate solution is recirculated.
The recovered ammonia is recycled  to  the  reactor.   In  the  second
process  sulfuric  acid  is  used  to  absorb ammonia from the reactor
gases.  At one plant the resulting ammonium sulfate solution  is  used
for the manufacture of another product.

The  hydrogen  cyanide  is  removed from the ammonia scrubber effluent
gases by absorbtion in cold water, and the waste gases are vented  to
the  atomsphere.   The  absorbed solution containing hydrogen cyanide,
water, and other contaminants is distilled to produce HCN  gas of  over
99 percent purity.

The  water  produced  during  the initial reaction  (Equation 1) of the
"formation of hydrogen cyanide is purged" with the"" distillation  bottom
stream  and  is  either  recycled to the absorber or discharged to the
treatment facility.  In order to be recycled, the distillation  bottom
water  has  to  be  cooled  by refrigeration prior to reuse in the HCN
absorber unit.   At  plant  locations  where  cold  water  is  readily
available  in large quantities, it can be used on a once-through basis
with a significant savings in energy costs.  Figure  17-1  presents   a
general  block  diagram for the manufacture of hydrogen cyanide by the
Andrussow process.

Water Use and Waste Source Characteristics

Water Use

Water is used  in  noncontact  cooling  in  the  absorber,  pump  seal
quenches,  flare stack flushes, for washdown and cleanup of tank cars,
for absorption of the product from  reactor  gases,  and   for  washing
equipment  and  cleaning  up  leaks  and spills.  Table 17-3 gives the
detailed water consumption at one plant and also the total consumption
at two plants.  There is a pronounced difference  in  water  usage  at
these  two  plants due to the use of refrigeration  at Plant #782 which
makes possible the recycling of absorber water from  the   distillation
unit  back  to the absorber.  This practice is energy intensive but is
required in locations where an abundant supply of cool  water  is  not
available.   Plant #765 has such a supply and uses  absorber water on  a
once-through basis.  In this case, a much larger flow must be  treated
prior to discharge.

Waste Sources

The  following are sources of wastewater produced from the manufacture
of hydrogen cyanide by the Andrussow process:
                                    554

-------
                        HBIIWB
T        j
                                                                      OQU)   VQIT
                                                                     HRTOR
Ul
Ul
Ul



1__ -
REMHOR



MtmuA
ABSORDGR


I1CH ABSORPTICH


oxanuATKH
1


                                                                                                         HCN PflOOKT


                                                      USB) POM TIE
                                                      MftHUFACTlffiE Of
                                                      oniot pnxucis
                                                      OR RKVCISJ,
                                                      MENRBCKXQ),
                                                      A BLEHJ IS SBTf
                                                      TOfllE WSTB
                                                      OTEMMNT IUKT.
                       ne DisnuATMH BOITDH
                       IS EIHER RBCVCUD (A
                       PURGE 18 DISQIMOED)  OR
                       SEHT
                       FACILIW.
                      Figure 17-1.   General  process flow diagram for production  of hydrogen cyanide
                                                       by the Andrussow process.

-------
TfiBLE 17-3.  WKTER USM3 IN HYDROGEN CXKNIDE - SNDBI3SSOW PROCESS
Plant
                        Water Usage,  (m3/kkg of HCN)
            Total Consurrption                 Noncontact Cooling
#782
#765
    (1)
                    29.5
                    58.3
18.9
 8.00
(1)
Detail water usage (m3/kkg) at Plant #782 is:
    Noncontact cooling
   Direct process contact
   Indirect process contact
     (pumps, seals, leaks/
      spills, etc.)
   Maintenance, e.g. cleaning
     and work area washdown
    Noncontact ancillary uses
     (boilers, utilities, etc.)
   Exported steam
                                       18.9
                                        7.45
                                        0.71

                                        0.31

                                        0.67

                                        1.44
                                      556

-------
A.   Distillation Bottoms

     The wastewater contains  ammonia,  hydrogen  cyanide,  and  small
     amounts  of  organic  nitriles.   The water consists of the water
     produced by  the  reaction  plus  scrubber  water  used  for  the
     absorption  of  HCN.   The  absorption  water  bottoms are either
     recycled to the HCN  absorber  or  discharged  to  the  treatment
     facility.   Even if the distillation bottom stream is recycled to
     the absorber, a portion of it  is discharged to stop  the  buildup
     of impurities.

B.   Scrubber Streams

     If the ammonia scrubber liquid is recycled, a portion of  it  has
     to  be  purged  to  control  the accumulation of impurities.  The
     bleed contains the acid used from scrubbing and minor amounts  of
     organic nitriles.  The scrubber solution can also be used for the
     manufacture of other products  in which case nothing  is discharged
     to the treatment plant.

C.   Other Wastewater

     This includes leaks and spills, equipment and tank car  washings,
     noncontact  cooling water blowdown and rainfall runoff.  The tank
     cars are washed out with dilute acid  or  alkali  to  remove  any
     contaminants  present,  which,   if  allowed to remain in the tank
     car, can polymerize the hydrogen cyanide causing  safety  hazards
     due  to  possible  explosion   during  shipment.   The  noncontact
     cooling water may be contaminated with the product as a result,of
     leaks.  The recirculated cooling water is monitored  for  cyanide
     and  the  cooling  tower blowdown is discharged to the wastewater
     treatment facility.  During shutdown, the equipment  is drained to
     avoid freeze-up and the resulting wastewater is discharged  to the
     treatment facility.

     The quantity of wastewater produced and  treated  at  two   plants
     producing  hydrogen  cyanide by  the Andrussow process is given in
     Table 17-4.  The large variation in flow exists because the water
     used to absorb the hydrogen cyanide from  the  reactor  gases  in
     Plant  #765  is not recylced.   As discussed earlier,  that plant is
     situated  where  sufficient    cold   water   is   available  for
     once-through use.  Since the cold water  is readily available at  a
     low cost, the water used for absorption  is discharged.  A similar
     plant  practicing  recycling,   in  the   absence of available cold
     water, can achieve a total waste effluent of 7.1 mVkkg of  HCN.

Description of_ Plants Visited and Sampled

Screening

Plant  #765 was visited and  the wastewater sampled during  the screening
phase  of the program.  The  combined  wastes   consist  of  distillation
bottoms,  ammonia  recovery  purge   liquor,   tank car washings,  leaks,
                                   557

-------
TABLE 17-4.  WASTE FLOW DATA FOR HCN PRODUCTION BY THE ANDHUSSOW
             PROCESS
Plant                Total waste going to the treatment facility (m3/kkg)

#765                                   57
#782                                    9.9*
*
 The breakdown and flow of the different waste streams comprising the total
 is given below:
           Source                            Chit Flow(m /kkg)
Recovery and purification                           6.3
Pump seal quenches                                  0.58
Flare stack flushes                                 0.09
Sample hoods                                        0.02
NH3 stripper caustic                                0.24
Steam condensate from NH3 stripper                  0.90
Freeze protection                                   0.06
Washdowns and cleanup                               0.25
Boiler blowdown and condensate                      1.48
                                      558

-------
spills and equipment clean out,  purge  from  the  noncontact  cooling
water  system,  and  stormwater  runoff.    These  combined  wastes are
commingled with the other cyanide product wastewaters and sent to  the
alkaline  chlorination  treatment  facility.   The  first  unit of the
treatment facility is a trench where  the  pH  of  the  wastewater  is
raised  to  tne  range  of  8.5  to  11  with dilute caustic soda.  The
caustic is added under controlled mixing  conditions  with  continuous
automatic  pH  recording and caustic feed adjustment.  The pH-adjusted
wastewater is sent to two 8-hour  retention  ponds.   Chlorination  is
accomplished  by adding sodium hypochlorite at the pond entrance.  The
chlorinated  wastewater  from  the  8-hour   ponds   are   alternately
discharged  to  another  small  pond  having one hour of detention and
equipped with baffles and agitators.  Caustic and chlorine  are  added
as  required in the one-hour pond to achieve the low levels of cyanide
desired.  The effluent from the pond is discharged  to  a  POTW.   The
pond  contains  a  flow  controller/analyzer,  which  will  block  the
discharge from the pond when a high cyanide level is detected  in  the
treated  effluent.   Figure  17-2  is  a flow diagram of the treatment
process indicating the sampling location  use<5  during  the  screening
program.

Composite sampling conducted consisted of one 48-hour composite sample
for   nonvolatile  organics,  metals,  and  mercury  and  one  24-hour
composite sample  of  BOD5,  TSS,  TDS,  NH3,  Fe,  Cr,  Zn,  Cu,  and
settleable  solids.   Grab  samples  for  volatile  organics, cyanide,
phenols, temperature and pH were collected on two consecutive days  at
each   sampling   location.   Table  17-5  gives  the  flow  data  and
concentration and unit loads of ammonia-nitrogen,  total  cyanide  and
thallium,  for  the  sampled streams.  We believe that thallium is not
contributed by the hydrogen cyanide manufacturing process.

Verification

Plant  1765  was  sampled  again  in  the  verification  phase.    One
additional  stream  of  hydrogen cyanide wastewater was sampled in the
verification phase at a point upstream of mixing  with  other  cyanide
produce  wastewater.   This  stream is identified in Figure 17-2.  The
variation in the flow of the streams in the two  sampling  phases  was
small.   Table  17-6  gives the flow and pollutant data of the sampled
streams.

The second hydrogen cyanide plant sampled in  the  verification  phase
was Plant |782.  The wastewater from the hydrogen cyanide plant mainly
consists  of  blowdown  from the distillation column which is combined
with a portion of the other product wastewater and sent to an  ammonia
stripper.   Effluent  from the ammonia stripper is mixed with the rest
of the process wastewater from other products and  sent  to  a  single
stage  biological  system.  The primary treatment facility consists of
grit removal, oil skimmers, and  pH  adjustment.   The  effluent  from
primary  treatment  goes  through an oil separator and into an aerated
lagoon.  Effluent from  the  lagoon  is  flocculated  and  sent  to  a
clarifier.   The  overflow  from  the  clarifier  is  sent  to a final
settling  basin  before  final  discharge.    The   surface   drainage
                                  559

-------
                                      HYDROGEN CYANIDE
                                        WASTE WATER
                                                          OTHER CYANIDE PRODUCT
                                                             WASTE HATER

                                                          DILUTE CAUSTIC
                   ±
               POND
                                 SODIUM
                              HYPOCHLORITE
                            CAUSTIC
                           —i
                                t   ''
   e
     LEGEND

Waste streams sampled

Waste stream was
sampled in the
verification program
since it is free from
other cyanide wastes.
FINAL TREATED
  EFFLUENT
                CHLORINE
Figure 17-2.  General  waste water  treatment process flow diagram at plant  #765
                showing  the sampling points.   (Hydrogen cyanide manufacture.)

-------
      TSBLE 17-5.   FLOW AND POLLUTANT DATA OF THE PAW AND TREATED WASTE
               STREAMS OF PLANT #765 PRODUCING HYDROGEN CYANIDF BY
                              ANDRDSSOW PROCESS
   Stream
Description
                    Unit
          NH3-N
                                               Total
                                              Cyanide     .
        Thallium
                                                               V 1
#2
  Influent
     to
  Treatment
57(2)    7.8    4.4 <3) 107
                                                      6.1
                  (3)
       ,'028  0.0016
  Treatment
  (Alkaline
  Chlorination)
  Effluent
                     57
                       (2)
3.5    2.0 t°'  0.36
                                 0.020
                                                           (3)
       .010  0.00057
(1)
   Unit Load *

   in kg/kkg
                Unit Flow
(57   ,i)
\    kkg/
                     pollutant
                     concentration
                     in mg/1
x
                                ( fcg/m     \
                                I 1000 mg/y
(2)  The stream is a coraraingled waste water.  The flow given is the
     amount contributed by the HCN process.

(3)  The pollutant load was calculated by apportioning the mass emitted
     between the two waste streams on the basis of measured flews.  This
     is clearly a very approximate process and the results must be used with
     caution.
                                      561

-------
consisting  of  runoff, wash down, etc., from the hydrogen cyanide and
other process areas is collected separately.  The water is sent  first
to  a  surface  pond  where it undergoes a two-stage pH adjustment and
then is piped to a trickling filter.  It then merges with the  treated
process  wastewaters  in the clarifier.   A general flow diagram of the
treatment process including streams sampled is shown in Figure 17-3.

Table 17-7 gives flow and concentration data of the  sampled  streams.
In  Table 17-8, the unit waste flow and unit pollutant loads are given
for the raw and treated effluent.  Because of intermixing  of  various
product  wastewater  streams, the unit pollutant loads (especially for
treated effluent) were calculated based on hydraulic loadings and  the
method   used   is  only  an  approximation.   The  principal  process
wastewater from the hydrogen cyanide  plant  is  the  waste  from  the
recovery  and purification operation and represents a unit flow of 6.3
m3/kkg of HCN.  The total  wastewater  flow  going  to  the  treatment
facility  from  the hydrogen cyanide plant is approximately 9.9 m3/kkg
of HCN, consisting of both process contact and noncontact effluents.

In calculating the pollutant loads, (Table 17-8) the loss or  gain  of
water  to  the  treatment  system  such  as  evaporation, loss through
filtered solids, precipitation, and the water introduced by  treatment
chemicals   has   not   been   included   because  it  was  considered
insignificant in comparison to other factors.

Toxic Pollutant Concentrations

Total cyanide and thallium were the toxic pollutants detected  in  the
raw  waste  from  Plant #765 which was sampled in the screening phase.
We believe that thallium in the wastewater is not contributed from the
hydrogen cyanide process.

The  HCN  wastewater  at  Plant  #765  is  mixed  with  other  product
wastewater and the combined flow was sampled upstream of the treatment
system.   It is probable that thallium is contributed from these other
product wastewaters.

The raw waste stream was not analyzed  for  free  cyanide.   The  same
plant  was sampled again with another plant- in the verification phase.
In addition to total cyanide, free cyanide was  found  in  significant
concentrations  in  the  raw  process  waste  sources from the two HCN
plants.  Free cyanide in the wastewater consists of hydrogen  cyanide,
sodium or potassium cyanide and cyanogen chloride which may be present
as  a  result  of  chlorination   (especially in the treated effluent).
Total cyanide includes the free cyanide and cyanides  -found  in  metal
complexes  (such  as  sodium ferrocyanide or sodium ferricyanide).  No
toxic organic pollutants were found in significant  concentrations  in
the  HCN  plant  raw  waste  sampled.  The concentrations of the toxic
pollutants  found  in  the  raw  wastewater  in  the   screening   and
verification were:
                                   562

-------
                                 TABLE  17-6.  FLOW AND POLLUTANT  CONCENTRATION  DATA OF THE  SAMPLED WASTE
                                              STREAMS FOR  PLANT #765  PRODUCING  HYDROGEN  CYANIDE

Stream
Description
#1 Raw HCN
Unit Flow TSS Load
(m3/kkg of HCN) {kgA*g of HCN)
waste 57 1.1
#2 Influent to 57 f 1 > NA
NH3-N Load
(kg/kkg of HCN)
27
11(2)
Ctf(A)
(kg/kkg of HCN)
0.82
0.39(2)
CN(T)
(kg/kkg of
1.6
1.6
HCN)


                   the pond

                #3 Treated
                   effluent from
                   the final pond
              57(2),(3)
1.9(2)
7.1(2)
NA
0.00015
en
cn
CO
(1*  The stream is  comingled wastewater.   The flow given is the amount contributed
    by the HCN process.

(2)  The pollutant load was calculated by apportioning the mass emitted between
    the waste streams on the basis of measured flows.  This is clearly a very
    approximate process and the results must be used with caution.

(3)  The addition or loss of water from rainfall, addition of chemicals and
    evaporation has not been estimated.
                          = Not Available

-------
 DISTILLATION
 BOTTOM PURGE
                                          OTHER PRODUCT
                                         'WASTE WATERS
                                       OTHER PRODUCT WASTE WATER
                          PRIMARY
                       OIL SEPARATOR
                                  pH
                              ADJUSTMENT
                        COMPOSITING
                           POND
                         SECONDARY
                       OIL SEPARATOR
                          AERATED
                          LAGOON
                        FLOCCULATOR
                         CLARIFIER
   DISCHARGE
           SETTLING
             POND
                                                              SURFACE DRAINS
     EQUALIZATION
         POND
                                                       1
                                                                 TRICKLING
                                                                   FILTER
                                                             LEGEND
SAMPLING POINTS
Figure i?~3.
Genera! waste water treatment process flow diagram at plant #782
showing sampling points.  (Hydrogen cyanide manufacture.)
                                     564

-------
 Table 17-7.  FLOW AND POLLUTANT CONCENTRATION DATA OF THE SAMPLED WASTE
              STREAMS FOR PLANT #782 PRODUCING HYDROGEN CYANIDE

Stream
No.

Waste
Stream
Description
Flow CMC]
m3/day

C) CN(A) NH3-N
(mg/1)

TSS


1  Distillation(1>
   bottom purge

2  Ammonia stripper'3^
   influent

3  Ammonia stripper^3'
   effluent

4  Influent tot3J
   primary treatment
   facility

5  Final treated^3)
   effluent
(6.3)(2>   71
5400
5400
6400
 NA
167
 51
 31
 2.2
            62
145
 41
  7.0
  1.7
             886
 410
1380
  5.6
            24
 76
           162
110
 74
(1) - The total waste is composed of the blowdown from three distillation
      columns.  Three 24-hour composite samples were collected for each unit.
      The pollutant concentration value (given in mg/1) is an average of the
      three composited samples for the three waste stream sources.

(2) - The value given is the total unit flow in m3/kkg of HCN for the three
      purge streams.

(3) - The stream is combined wastewater-  It includes the waste effluents
      from hydrogen cyanide and other products.
                                       565

-------
      TABLE 17-8.  UNIT FLOW AND UNIT POLLUTANT LOADING FOR RAW AND
                   TREATED WASTE EFFLUENTS AT PLANT #782

Unit
Stream Unit
Flow
(m3Akg)
Pollutant Loading
Total
Cyanide
CN(T)
(kg/kkg) <1)
Free Ammonia-
Cyanide N
CN/B, NH,-N
(A; 3

Total
Suspended
Solids
TSS
Process raw
wastewater
(distillation
botton purge
                6.3
0.45
0.39
5.6
(1)   Unit pollutant load = unit flow
                           (m3/kkg)
0.15
Process 6.3^J 0.014 0.011
wastewater
treated
effluent
Total HCN 9.9<3> 0.022 0.017
wastewater
treated
effluent (2>
0.035 0.47



0.055 0.74


                                         pollutant concentration   x    kg/3
                                         (in mg/1 from Table 17-7)    1000 mg/1
(2)  The pollutant load was calculated by apportioning the mass emitted from
     the total treated effluent (which includes other product wastewater} on
     the basis of measured flow contributed by the HCN process.  This is clearly
     an approximate process and the results must be used with caution.

(3)  The wastewater flow consists of direct process contract and noncontact
     effluent from the HCN plant going to the treatment system.
                                        566

-------
               Maximum Raw Waste Concentration Observed
                                Ug/1)

                      Screening              Verification
Pollutant           Plant #765          Plants #765, #782

Thallium                   25           Not Determined
Cyanide (Total)       166,000                186,000
Cyanide (Free)      Not Determined           172,000

The   general   sampling   methodology   used  in  the  screening  and
verification program is described in Section 5.  A total of nine  days
of  sampling  was  conducted  at Plants #765 (sampled twice) and #782.
Thirteen wastewater sampling points were involved which  included  the
raw  wastewater,  combined  wastewater,  and combined treated effluent
streams.   The  evaluation  of  the  toxic  metal  and  toxic  organic
pollutant   content  of  these  process  steams  was  based  on  total
analytical data  points  from  both  the  screening  and  verification
phases.

The  daily  toxic pollutant waste load in the raw waste was calculated
from  the  effluent  waste  flow  rate  and  the  measured   pollutant
concentration of the toxic pollutant.

     This is given by:

     Daily loading (as kg of pollutant per day) = (C)(Q)
                                                  (1000)

     where:

     C  is  the  concentration  of the pollutant expressed  in units of
     mg/1 (Note:  kg/m3 = 1000 mg/1), and

     Q  is the waste stream flow rate expressed in units of  mVday  (m3,
     a  cubic meter, is equal to 264.2 U.S. gallons).

     Similarly, the unit loadings were calculated   from  the  reported
     hydrogen cyanide production rate, the waste stream flow rate, and
     the measured pollutant concentration.

     Unit loading (as kg of pollutant = (C)(Q)
     per kkg of hydrogen cyanide)       1000P

     where  C  and  Q  are  the  same as described  above, and P  is the
     hydrogen cyanide production rate expressed in  units   of  kkg/day
     (kkg is  1000 kg, a metric ton, which is equal  to 2205  Ibs).

In  the  case  of  two  or  more  process  waste  streams going  to the
treatment system the daily raw waste load of the toxic  pollutant  was
calculated   by   determining  the   combined  pollutant  load  of  the
individual streams.
                                   567

-------
Table 17-9 gives the toxic, conventional and nonconventional pollutant
loadings of the raw waste for Plants #765 and #782 which were  sampled
in  the  screening  and  verification  phases.   The  overall  average
polluant loads for the sampled plants are given in the last column  of
the table.

The  approximate  toxic  pollutant  generated  per  year by the entire
subcategory is estimated  by  multiplying  the  overall  average  unit
pollutant  loading  {Table 17-9) with the hydrogen cyanide subcategory
production from Table 17-1 (165,500 kkg/yr).

               Pollutant                Waste Load (kg/year)

               Cyanide (Free)                 100,000
               Cyanide (Total)                450,000


Toxic Pollutants of Concern

The toxic pollutants of concern in the HCN  raw  waste  are  free  (or
oxidizable)  cyanide  and total cyanide.  Free (or oxidizable) cyanide
is  cyanide  amenable  to  chlorination,  and  is  designated  in  the
regulation  as Cyanide A.  No organic toxic pollutants of significance
were found in the raw waste of the sampled plants.

Process Modifications and Technology Transfer Options

Process modifications have not been identified for the subcategory.

Best Management Practices

No best management practices have been identified for the subcategory.

Prevailing Control and Treatment Practices

Out of a total of seven plants currently producing hydrogen cyanide by
the Andrussow Process, 308 data  are  available  for  only  two.   The
production at these two plants constitutes more than 70 percent of the
total   subcategory  production.   Since  the  two  plants  produce  a
significant  amount  of  the  total  subcategory   production,   their
wastewater   treatment  technologies  are  taken  as  the  subcategory
treatment practices.  The  two  plants  were  visited  to  review  the
treatment systems and to collect waste effluent samples.

Plant #765 has a high volume effluent because the water used to absorb
the reactor gases is not recylced since low cost cold water is readily
available  at  the site.  The wastewater consisting of scrubber purge,
absorption water, and plant runoff is mixed with other cyanide product
wastewaters and sent to an alkaline chlorination system.   The  pH  of
the  wastewater  is raised to about 10 with dilute caustic in an inlet
trench and then it is discharged to one of two 8-hour ponds as  sodium
hypochlorite  is  added  to  oxidize  the  cyanide  to  cyanate.   The
chlorinated wastewater is transferred to a small  pond  equipped  with
                                 568

-------
TABLE 17-9.  SUMMARY OF POLLUTANT RAW WASTE LOADING FOUND IN SCREENING AND
             VERIFICATION SAMPLING
SUBCATEGORY
HYDROOT3 CYANIDE
Average Daily Pollutant Loading and Concentrations at Plants Sampled

                           kgAkg of HCN
                              (mg/D
Pollutant
TOXIC
Free Cyanide
Total Cyanide
#765 (s)
NA
6.1
(110)
t 765 (v)
0.82
(14)
1.6
(29)
# 782 (v)
0.39
(62)
0.45
(71)
Overall
Average
0.61
2.7
Conventional
and Nonconventional
TSS
NH3-N
(S) =
(V) =
NA
4.4
(78)
Sampled in
Sampled in
2.0
(35)
37
(480)
screening phase
verification phase
0.15 1.1
(24)
5.6 12
(890)


                                     569

-------
agitators  and  baffles  before final discharge to a POTW.  Caustic or
chlorine is added to the final pond to achieve the desired low  levels
of cyanide.  The treatment system is shown in Figure 17-2.

Plant  #782  uses  a  single-stage biological treatment system for the
treatment of effluent from the hydrogen cyanide  plant.   The  process
wastewater  from  the HCN plant consists mainly of distillation column
blowdown and is combined with other  cyanide  product  wastewater  and
sent  to an ammonia stripper.  The effluent from the stripper combines
with other product wastewaters and is  treated  by  means  of  a  grit
chamber, an oil separator, a compositing pond, a second oil separator,
an  aerated lagoon, a flocculator and a final clarifier.  The overflow
from the clarifier is sent to a final settling basin before discharge.
The runoff from the HCN plant and other product manufacturing areas is
combined and sent to a  pond  for  a  two-stage  pH  adjustment.   The
effluent from the pond is treated by a trickling filter and clarifier,
and   the  clarifier  effluent  is  mixed  with  the  treated  process
wastewater A general block diagram of the treatment system is shown in
Figure 17-3.

Advanced Treatment Technologies

The three pollutants of concern in hydrogen  cyanide  plant  effluents
are  cyanide,  ammonia,  and chlorine.  The treatment technologies for
cyanide removal include alkaline chlorination,  biological  treatment,
ozonation,    wet    air    oxidation,   electrolytic   decomposition,
acidification, activated carbon, permanganate oxidation,  lime reaction
with sulfur, radiation, evaporative recovery/ catalytic oxidation  and
ion   exchange.   Except  for  alkaline  chlorination  and  biological
treatment, the remaining treatment technologies are not   effective  or
advantageous for one or more of the following reasons:

A.   The technology has low cyanide removal efficiency.

B.   The  technology  cannot  treat  wastewater  with   high   cyanide
     concentrations.

C.   The technology has air pollution problems.

D.   The technology has high operating costs.

The free cyanide in the raw waste is readily oxidizable and  exerts   a
chlorine  demand.   Sufficient chlorine is added to react with ammonia
and to oxidize cyanide.  The presence of large amounts of ammonia will
increase the cost  of  chlorination.   If  costs  are  too  extensive,
residual  ammonia  in the raw waste effluent can be reduced by steam or
air stripping before alkaline chlorination to  reduce  the  amount  of
chlorine required.
                                  570

-------
Selection of Appropriate Technology and Equipment

Technologies for Different Treatment Levels

A.   Level 1 (BPT)

     Two-stage alkaline chlorination followed  by  pH  adjustment  was
     chosen  for  the removal of cyanide from the raw waste effluents.
     The technology is being practiced  in  the  industry.   The  flow
     diagram of the treatment system is shown in Figure 17-4.

B.   Level 2 (BAT)

     The treatment is the same as BPT {Level 1) except  that  residual
     chlorine  is  reduced  to  a lower level by treatment with sulfur
     dioxide.  Chlorine  in  adequate  amounts  is  added  to  oxidize
     cyanide.   Where practiced, steam or air stripping of ammonia has
     not been considered as a part of the treatment system  since  the
     value of the recovered ammonia is the justification for doing it.
     It  has  been  assumed  to  be process related.  The general flow
     diagram of the treatment process is given in Figure 17-5.

     Addition   of   chlorine   to   remove   ammonia    ("break-point
     chlorination")  is  not  intended  for either Level 1 or Level 2.
     Break-point  chlorination for ammonia removal  generally  is  very
     expensive.   In  this industry, ammonia control, where necessary,
     should be accomplished by steam or air stripping and recovery and
     reuse of the ammonia.  The achievable concentrations  of  ammonia
     in  the final effluent from a hydrogen cyanide plant with cyanide
     removal after ammonia recovery are presented below for guidance.

Equipment for Different Treatment Levels

A.   Equipment Functions

     In  level 17  the raw wastewater enters  a  holding  tank  equipped
     with an external pump and recirculation system.  Caustic soda and
     chlorine  are  added  and  the  tank  contents  are  mixed by the
     recirculation  pump.   Following  this   first   stage   alkaline
     chlorination,  the  wastewater is chlorinated further in a second
     tank which is equipped with  automatic  pH  control.   The  final
     effluent  is neutralized to pH 6-10.5 before discharge.  In Level
     2,  cyanide control is accomplished, using the same equipment as in
     Level  1.  To remove  excess  chlorine  before  discharge,  sulfur
     dioxide  is  fed into the chlorinated wastewater by a modified gas
     chlorinator,   with   oxidation-reduction   potential    control.
     Solutions  of sulfur dioxide in water are acidic, so the addition
     of  sulfur dioxide to react with the residual chlorine  will  tend
     to  neutralize the alkalinity of the wastewater, thus reducing the
     amount  of   suIfuric  acid  that  must  be  added  for  final  pH
     adjustment.  As in Level 1, the effluent is then adjusted  to  pH
     6-10.5 before discharge.
                                    571

-------
m
-4
to
 RSW

WASTE-

WWER
                                 CAUSTIC 9DDA
                                                               CHLORINE
                                   KJUJJNG AND 13F

                                   STAGE ALKALINE

                                    CHLQRBATIDN
                                            SBOCMD 9EAGE

                                          ALKALINE CHDORIHATICH
                                                                                                    pH




                                                                                                    ~ 1
                                   Includes How monitoring, pH monitoring and sampler,
                        Figure 17-4.   Level 1 waste water treatment for hydrogen cyanide  subcategory.

-------
B.   Chemicals and Handling

     Caustic soda solution, chlorine,  sulfur  dioxide,  and  sulfuric
     acid  are  used in the waste treatment process.  Caustic soda and
     sulfuric acid are  common  industrial  chemicals  which  pose  no
     specific hazards when handled by conventional corrosion-resistant
     feeding  equipment.   Chlorine and sulfur dioxide are received in
     one-ton containers as liquified  gases,  and  are  fed  as  water
     solutions   by   vacuum-controlled  equipment  designed  for  the
     specific chemical.   No  unusual  chemical  feeding  or  handling
     problems  are  anticipated,  provided  precautions  are  taken to
     prevent gas leaks and to guard against corrosive attack.

C.   Separation and Removal of Solids

     Since few solids are produced in the treatment process, there  is
     no significant sludge disposal problem.

D.   Monitoring Requirements

     Internal  process  monitoring  is  done  largely  with  automatic
     sensing   and   control   equipment   for   regulating   pH   and
     chlorine/sulfur  dioxide  residuals.   Field  tests  for  cyanide
     and/or  chlorine   in the effluent should be made regularly by the
     operator,  and  24-hour  composite  effluent  samples  should  be
     collected  and analyzed for cyanide as required in local or NPDES
     permits.

Treatment Cost Estimates

General Discussion

A model  plant  concept  was  developed  as  a  basis  for  estimating
treatment  costs.   For  conceptual design a representative unit waste
flow (cubic meters per  kkg of HCN) was selected, together  with  three
different  HCN production  rates.  The latter were chosen to cover most
of  the  subcategory  production  range.   The  selected   daily   HCN
production  for  the  model  plant was multiplied by the selected unit
flow to obtain the volume  of  wastewater  passing  to  the  treatment
system.    The  selected   unit  raw  waste  pollutant  load  was  also
multiplied by  the  model  plant  production  rate  to  determine  the
pollutant  load  on the treatment system.  Capital and equipment costs
were then calculated based on developed conceptual  design  parameters
for each model plant production rate.

A.   Wastewater Flow

     The unit process wastewater flows for the  two plants  visited  in
     this  study  are   6.3 mVkkg of HCN  (Plant $782) and 57 mVkkg of
     HCN (Plant 1765).  The  difference  results  from  the  different
     absorption  water  discharge  practices  at  the two plants.  The
     model plant has been  developed using the larger unit flow rate of
     57 cubic meter/kkg of HCN, since  this   is a  more  conservative
                                   573

-------
ui
-J
                             CAUSTIC
                              SODA
 HAW  	
WASTE WATEH
                                             -a
                           HOLDING AND 1ST STAGE
                          ALKALINE CHLORINATION
                                                                         SULFUR
                                                                         DIOXIDE
                                                                             -----9,
                                            SECOND STAGE
                                          ALKALINE CHLORINATION
II ADJUSTMENT
                                                                                                         IFFLUENT
                               Includes flow monitoring, pH monitoring and sampler,
                               ORP  = Oxidation Reduction Potential Control
                       Figure 17-5.  Level 2 waste water treatment for hydrogen cyanide subcategory.

-------
     approach.    The  Agency considered developing effluent limits for
     two different levels of flow but rejected it because of the cost,
     complexity,  and difficulty in implementing the approach.

B.   Production

     For wastewater treatment cost estimates, three production  levels
     were  selected for the model plant.  These are 31,800, 50,900 and
     63,600 kkg/yr.

C.   Wastewater Pollutant Load

     The three pollutants of concern in the  subcategory  are  cyanide
     {oxidizable  and  total), ammonia, and chlorine.  Chlorine is not
     present  in  the  raw  waste  but  is   added   during   alkaline
     chlorination  treatment.   The  average  value of 0.61 kg of free
     cyanide/kkg of product HCN and 12 kg of NH3/kkg  of  product  HCN
     (Table  17-9)  developed  from  the  screening  and  verification
     results were used for the model plant raw waste loads.

D.   Chemicals Used

     At the BPT level of treatment, alkaline chlorination requires  33
     kg of chlorine (most to react with ammonia) and 5.0 kg of caustic
     per kkg of product HCN.  For BAT  treatment, 9.0 kg of SOZ per kkg
     of  product  HCN  is  used  for dechlorination  in addition to the
     chemicals used for BPT treatment.

E.   Solids Generated

     Few, if any, solids  are  produced   in  treating  HCN  produciton
     wastes.

     The  estimated  costs, for  the   three  model plants at different
     production levels are given in Tables 17-10,  17-11,  and  17-12.
     As  mentioned earlier, both the hydraulic and pollutant loads per
     unit of production are held constant over   the  entire  range  of
     production.

     The  costs  shown in Tables 17-10,  17-11, and 17-12 at each  level
     of treatment correspond to BPT  (Level 1} with   incremental   costs
     to meet the more stringent BAT requirements.

Basis for Regulations

Evaluation of BPT Treatment Practices

A total  of  seven  plants  produce hydrogen cyanide by the Andrussow
Process.  At one facility the raw wastes  from   the  hydrogen   cyanide
plant  are combined with the waste from an organic cyanide process and
sent to a biological treatment system  to  reduce  organic   and   cyanide
pollutants.   Four of the other six HCN producers  (using the Andrussow
Process)  use  alkaline  chlorination   for   treatment  of  raw    waste
                                   575

-------
effluents.  There is no available information concerning the treatment
practices at the other two plants.

Basis for BPT Limitations

A.   Technology Basis

     The predominant treatment practice for raw waste effluent in  the
     HCN   subcategory   is  alkaline  chlorination.   The  Agency  is
     therefore  promulgating  BPT  effluent   limitations   based   on
     alkaliine  chlorination  to destroy cyanide amenable to treatment
     by chlorination (free cyanide or Cyanide A).

B,   Flow Basis

     The effluent limitations are based on the high flow (57 mVkkg of
     HCN) model, that is, with no recycle of absorber  water.   A  low
     flow  basis (7 m3/kkg of HCN based on tne flow of Plant #782) was
     rejected as being too  energy  intensive  due  to  the  need  for
     refrigerative  cooling of the recycled absorber water.  The water
     going to the model plant treatment system is assumed  to  consist
     of process and contact cooling waste effluents, leaks and spills,
     and  storm  water  run-off.   The  boiler blowdown and noncontact
     cooling water (once through or  blowdown  discharge  in  case  of
     closed loop) are not included in the flow basis.

C.   Selection ol Pollutants to be Regulated

     The selection of pollutants on  which  specific  limitations  are
     promulgated  is  based on the evaluation of raw waste composition
     as determined during the screening and verification programs.

     Plant #765  was  sampled  during  the  screening  phase  and  the
     presence   of  toxic  pollutants  in  significant  concentrations
     established the need for verification sampling.  Two plants  were
     sampled  in the verification phase.  Free cyanide, total cyanide,
     and ammonia were found in the raw waste  at  concentrations  high
     enough  to  be  treatable  (Table 17-9) using  available treatment
     technology  options.    These   were   therefore   selected   for
     regulation.   Chlorine  concentrations  in  the  effluent are not
     affected by BPT treatment technology and therefore no  BPT   limit
     is  promulgated  for this parameter.  Thallium  is best controlled
     by management practices developed by the permitting authority  on
     a case-by-case basis.

D.   Basis of Pollutant Limitations

     1.   Conventional and nonconventional parameters

          a.   pH

               The treated effluent is to be controlled within the  pH
               range  of 6.0 to 10.5.  This limitation is based on the
                                     576

-------
   TABLE 17-10.  MODEL PLANT TREATMENT COSTS
Subcategory
Production
Hydrogen Cyanide
31,800 metric tons per year
A. INVESTMENT COST

     Site development  	
     Equipment  	
     Monitoring equipment  ..

          Subtotal  	
     Contractors 0 & P b....

          Subtotal  	
     Engineering  	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL INVESTMENT  COST .


B. OPERATION AND
   MAINTENANCE  COST

     Labor and  supervision .
     Energy  	
     Chemicals  	
     Maintenance  	
     Taxes and  insurance
     Residual waste disposal
     Monitoring,  analysis
      and reporting  	
                                                     ($)
                               BPT

                             6,000
                           850,000
                            20,000

                           876,000
                           131,400

                         1,007,400
                           201,480

                         1,208,880
                           120,888

                         1,329,768
                             6,000

                         1,335,768
     TOTAL OPERATION AND
     MAINTENANCE COST
                             84,000
                              9,000
                            296,000
                            132,977
                             40,073
                                  0

                             15,000
                            577,050
    BAT

      0
125,000
      0

125,000
 18,750

143,750
 28,750

172,500
 17,250

189,750
      0

189,750
 14,000
  3,100
 97,000
 18,975
  5f693
      0

  7,500
146,268
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                            216,353

                            793,403
 30,872

177,140
  Represents  the  incremental  cost  above that for BPT treatment
  Overhead and Profit
                              577

-------
    TABLE 17-11.  MODEL PLANT TREATMENT  COSTS
Subcategory
Production
Hydrogen Cyanide
50,900 metric tons per year
A. INVESTMENT COST

     Site development  	
     Equipment  	
     Monitoring equipment  ..

          Subtotal  	
     Contractor's 0 & P b....

          Subtotal  	
     Engineering  	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL INVESTMENT COST  .


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  .
     Energy 	
     Chemicals  	
     Maintenance  	
     Taxes and insurance ...
     Residual waste disposal
     Monitoring,  analysis
      and reporting 	
                               BPT

                             6,000
                         1,340,000
                            20,000

                         1,366,000
                           204,900

                         1,570,900
                           314,180

                         1,885,080
                           188,508

                         2,073,588
                             6,000

                         2,079,588
     TOTAL OPERATION AND
     MAINTENANCE COST
                            84,000
                             9,800
                           476,000
                           207,359
                            62,388
                                 0

                            15,000
                           854,546
                                                     (S)
    BAT

      0
135,000
      0

135,000
 20,250

155,250
 31,050

186,300
 18,630

204,930
      0

204,930
 14,000
  3,100
154,000
 20,493
  6,148
      0

  7,500
205,241
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                           337,373

                         1,191,919
 33,342

238,583
  Represents the incremental cost above  that  for  BPT treatment
  Overhead and Profit
                             578

-------
   TABLE  17-12
  MODEL PLANT TREATMENT COSTS
Subcategory
Production
Hydrogen Cyanide
63,600 metric tons per year
A. INVESTMENT COST

     Site development  	
     Equipment  	
     Monitoring equipment  ..

          Subtotal  	
     Contractor's 0 & P b....

          Subtotal  	
     Engineering  	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL INVESTMENT COST .


B. OPERATION AND
   MAINTENANCE  COST

     Labor and  supervision .
     Energy  	
     Chemicals	
     Maintenance  	,	
     Taxes and  insurance  ...
     Residual waste disposal
     Monitoring,  analysis
      and reporting 	
                                                    ($)
                               BPT

                             9,000
                         1,600,000
                            20,000

                         1,629,000
                           244,350

                         1,873,350
                           374,670

                         2,248,020
                           224,802

                         2,472,822
                             9,000

                         2,481,822
     TOTAL OPERATION AND
     MAINTENANCE COST
                             84,000
                             11,500
                            592,000
                            247,282
                             74,455
                                  0

                             15,000
                          1,024,237
    BAT

      0
190,000
      0

190,000
 28,500

218,500
 43,700

262,200
 26,220

288,420
      0

288,420
 14,000
  4,600
191,000
 28,842
  8,653
      0

  7,500
254,595
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                            402,328

                          1,426,565
 46,926

301,521
  Represents the  incremental  cost  above  that  for BPT treatment
  Overhead and Profit
                              579

-------
        TABLE 17-13 MODEL  PLANT  UNIT  TREATMENT COSTS
     Subcategory Hydrogen Cyanide
                                Annual  Treatment Costs ($/kkg)
                                   LEVEL OF TREATMENT
COST ITEM PRODUCTION
(kkg/yr)
Annual Operation
and Maintenance


Annual
Amortization


Total Annual
Cost



31
50
63

31
50
63

31
50
63

,800
,900
,600

,800
,900
,600

,800
,900
,600
BPT

18.
16.
16.

6.
6.
6.

24.
23.
22.

15
79
10

80
63
33

95
42
43
BAT*

4.
4.
4.

0.
0.
0.

5.
4.
4.

60
03
00

97
66
74

57
69
74
*Represents the incremental cost  above  BPT
                              580

-------
data  presented  in  Appendix   B   of   the   proposed
Development Document (60) and the JRB Study (52).

TSS

The concentration  of  suspended  solids  found  during
sampling  of the raw wastewater was low.  No additional
solids are produced in the treatment technology and  no
provision  presently  exists  in  the existing or model
treatment systems  for  the  removal  of  sol ids.   The
maximum  concentration  of  35 mg/1 of TSS was found in
the  raw  waste  during  screening   and   verification
sampling   (Table  17-9).  However, 35 mg/1 was accepted
as the long term average concentration on the basis  of
historical  data shown in Table A-13a.  The variability
factors estimated for ammonia (Table 17-14) are used to
calculated the concentration bases and effluent limit.

The TSS maximum 30-day average concentration is:

(35 mg/1)  (1.6) = 56 mg/1

and the TSS maximum 30-day average effluent limit
is given by:

(56 mg/1  (57 mVkkg)  (kq/m3)  =3.2 kg/kkg
                    (1000 mg/1)

The TSS 24-hour maximum  concentration is given by:

(35 mg/1)  (4.2) = 150 mg/1

and the TSS 24-hour maximum effluent  limit  is given
by:

(150 mg/1) (57 mVkkg)  (kg/in*)   =8.6 kg/kkg
                     (1000 mg/1)

Ammon i a

The Agency is not promulgating effluent limitations for
ammonia in the  discharge  for  the   reasons  described
above.   However,  we  are providing guidance for use by
permit writers and POTW's in cases where the  discharge
of  ammonia  is  of concern.  This guidance  is based on
performance data  from   plant  #765.    Plant  #765  has
recently submitted one and one-half years of monitoring
data  on   the treated  effluent for ammonia.  Plant #765
uses a proprietary  process  for  removal   of  ammonia,
however,   the same performance can be achieved by steam
stripping.  The variability factors for the  daily  data
and  30-day averages were calculated  from the long-term
data as shown in Table  17-14.   The   long-term   average
                       581

-------
         TABLE 17-14.  STATISTICAL ANALYSIS OF HISTORICAL EFFLUENT
                MONITORING DATA ON FREE CYANIDE AND AMM3NIA
                              FROM PLANT #765


Pollutant

Free
Cyanide

Ammonia



Time
Period

9-79
to
7-80

1-79
to
7-80


Monitoring
Frequency

Daily
30-day
average
Daily
30-day
average
N
No.

512
11

540
11

X
Mean
(mg/1)

0.20
0.21

45
52

S(1)
Std
Dev
(mg/1)
0.36
0.087

42
19

cv(2)
Coeff . of
Variation

1.8
0.40

0.93
0.36

VF
Variability
Factor

8.0<3)
1.7(4)

4.2 (3)
1.6(4)

For free cyanide and ammonia, the long-term monitoring data were screened of
outliers.  In the first place, values recorded as zero were interpreted to
mean "inability to measure pollutant" and were rejected prior to the statisti-
cal analysis.  For the remaining data, the reported measurements of oxidizable
CN were screened by the use of the t-statistic

                        t = max  ({X max-x) /s, (X-3anin)/s)

for extreme values as outliers.  Screening was performed on a month-by-month
basis, and any datum with a calculated t value exceeding the 99% confidence
limits from the t distribution was concluded to be an outlier.  Given rejec-
tion of a value, recomputation of statistical measures for that month was
performed.
 (1)  Arithmetic standard deviation, S

     where S2 = Z (X-X)2/(N-1)
     For 30-day averages, this is the standard error of the mean

 (2)  CV = S/X

 (3)  For daily measurements, VF is calculated by
     In (VF) = S1 (2.33 - O.SS'l
     T-_TJ_ _ ._  /f* * \ S   T«» / n i   /firi\ &
     Where (S1)  = In (1 + (CV)
is the moments estimator of the scale
parameter of the lognormal distribution
and 2.33 is the Z value corresponding to
99th percentile
(4)  For 30-day average data, VF = (l + 1.64  (CV))
     Where 1.64 is the Z value for the 95th percentile.
                                     582

-------
     concentration  of  52  mg/1   was  used as the basis for
     guidance.   The estimated variability factors and  model
     plant   flow   rate   were   used  in  calculating  the
     concentration bases as follows:

     The maximum  30-day  average  concentration  basis  for
     ammonia is given by:

     (52mg/l) (1.6) = 83 mg/1

     The 24-hour maximum concentration is given by:

     (52 mg/1)  (4.2) = 220 n\g/l

Toxic Pollutants

The toxic pollutants promulgated  for  regulation  are  free
cyanide and total cyanide.

a.   Free Cyanide

     Plant #765  practices  alkaline  ch1orination  and  has
     recently  submitted  almost one year of monitoring data
     on the treated effluent for free cyanide.  The  samples
     were   properly   stabilized   before   analysis.   The
     variability factors  for  the  daily  data  and  30-day
     averages  were  calculated  from  the long-term data as
     shown  in   Table   17-14.    The   long-term   average
     concentration  of  0.21  mg/1 (Table 17-14) was used as
     the  basis  for   the   limitations.    The   estimated
     variability factors and model plant flow rate were used
     in  calculating  the  concentration  bases and effluent
     limitations.

     The maximum 30-day average concentration basis for free
     cyanide is given by:

     (0.21 mg/1) (1.7) = 0.36 mg/1

     The 24-hour maximum concentration is given by:

     (0.21 mg/1) (8.0) = 1.7 mg/1

     The  maximum   30-day  average  effluent  limitation  is
     calculated by:

     (0.36 mg/1  (57 mVkkg)  (kq/m3)   * 0.021 kg/kkg
                           (1000 mg/1)
                          583

-------
The 24-hour maximum effluent limitation is given by:

(1.7 mg/1)  (57 mVkkg) (kg/m3)   = 0.10 kg/kkg
                     (1000 mg/1)

Total Cyanide

The variability factors for  total  cyanide  for  daily
data  and  30-day averages were estimated from a 28-day
study conducted by Plant #765 and are  given  in  Table
17-15.   In  the case of the 30-day average variability
factor,  it  was necessary to apply a different  approach
which   requires   estimation  of  the  30-day  average
arithmetic  standard  deviation.   This  approach,  also
used  in the Treatability Study (61), is necessary when
30-day average data is not available.  The approach for
determining  the  estimated  30-day  average   standard
deviation is as follows:

Estimated 30-day Average Standard Deviation =

24-hour Measurement Standard Deviation
                   30

The  limitations for total cyanide are derived from the
average unit effluent load (0.19 kg/kkg given in  Table
17-15),   variability . factors estimated from the 28-day
test, and model plant flow of 57 m3/kkg.

The maximum 30-day average effluent for  total  cyanide
limitation  is calculated by:

(0.19 kg/kkg) (1,2) = 0.23 kg/kkg

The  total   cyanide 24-hour maximum effluent limitation
is given by:

(0.19 kg/kkg) (3.4) = 0.65 kg/kkg

The total cyanide maximum 30-day average  concentration
basis is:

(0.23 kg/kkg)      1      (1000 mg/1)  = 4.0 mg/1
              (57 mVkkg)   (kg/m3)

The  total   cyanide 24-hour maximum concentration basis
is:

(0.65 kg/kkg)      1     (1000 mg/1)  = 11 mg/1
              (57 mVkkg)  (kg/m3)

The final effluent  limitations  for  Hydrogen  Cyanide
produced  by  the  Andrussow   Process are summarized  in
                     584.

-------
             TABLE 17-15.  STATISTICAL ANALYSIS OF THE 28-DAY EFFLUENT
                   SAMPLING RESULTS CN TOTAL CYANIDE
                                FBQM PLANT #765
                                      Total Cyanide

Daily Data

  No. of points                              25
  Average Unit load                        0.192
     kg/kkg of HCN
  Std. Deviation(S)                        0.128
  Std. Deviation (Sf)                      0.61
  Variability Factor                       3,44
30-Day Average Data
  The Standard error
  of the mean (A)                           0.023
  Coefficient of
  variation for the mean(CV)               0.119
  Variability factor                       1.19
Variability Factor Ratio

  V.F.R.                                   2.9
                                  585

-------
               Table    17-16    for    toxic,    conventional,    and
               nonconventional pollutants.

Basis for BCT Limitations

While   EPA  has  not  yet  proposed  or  promulgated  a  revised  BCT
methodology in  response  to  the  American  Paper  Institute  v.  EPA
decision  mentioned  in Section 3,  EPA is promulgating BCT limitations
for this subcategory.  These limits are identical to  those  for  BPT.
EPA  is  not promulgating any more stringent limitations since we have
identified  no  technology  option  which  would  remove   significant
additional  amounts  of  conventional  pollutants.  The dechlorination
technology  added  to  BPT  for  BAT  does   not   remove   additional
conventional  pollutants.   As  BPT  is  the  minimal level of control
required by law, no possible application of the BCT cost  tests  could
result  in  BCT  limitations  lower  than  those  promulgated  in this
regulation.  Accordingly, there is no need to wait until  EPA  revises
the BCT methodology before promulgating BCT limitations.

Basis for BAT Limitations

The  Agency considered different advanced level technologies and their
cost effectiveness relative to the base level  system   (BPT)  for  the
removal  of  toxic,  conventional,  and nonconventional polluants.  For
BAT, the  Agency  is  utilizing  Level  2  technology  which  includes
dechlorination before final discharge.

The  Agency  also  considered break point chlorination  for essentially
complete destruction of cyanide and  ammonia  removal.   However,  the
operational  costs  were  too high.  The reduction of effluent load to
the  treatment  system  by  recycling  the  absorber  water  was  also
considered  and  was  found to be too energy intensive  and too costly.
Therefore the only- cost effective treatment technology  beyond BPT  was
found to be dechlorination.

A.   Technology Basis

     For BAT, the Agency is promulgating limitations based on BPT with
     the addition of dechlorination (Figure 17-5, Level  2).   Control
     of  chlorine  in  the  discharge  is uniformly inadequate in this
     industry.  Its control in  BAT  is  believed  to   be  appropriate
     because  of  its  well-documented  toxicity to aquatic life.  The
     basis for the chlorine limit is transfer of technology  from  the
     chlor-alkali industry  (Appendix A).  This transfer is appropriate
     because  the  chlorine  in  both  streams is amenable to  the same
     treatment for  removal  and  removal  is  not  inhibited  by  the
     presence of other chemicals in either of the waste streams.

B.   Flow Basis

     The BPT effluent discharge rate of 57 mVkkg of HCN has been used
     as the basis for the BAT model plant.
                                  586

-------
                        TABLE 17-16.   EFFLUENT LIMITATIONS
                       HYDROGEN CYANIDE (ANDRUSSOW PROCESS)
             Best Practicable Control Technology Currently Available
                        Wastewater Flow:  57 m3/kkg of HCN




Subcategory
Performance
Pollutant (mg/1)
Convent iona I/
Nonconventional
Pollutants
Total Suspended
Solids 35(1>
Ammonia-N 5 2 ( '
Toxic pollutants:
Free Cyanide*4' 0.2l(2)
Total Cyanide <4> 3.3<3>
Daily
Variability
Factor
30-day
Variability
Factor




4.2/1.6
4.2/1.6

8.0/1.7
3.4/1.2

Concentration
Basis (mg/1)
Max.
30-day 24-hr.
Avg. Max.




56 150
83 220

0.36 1.7
4.0 11

Effluent Limit
(kgAkg)
Max.
30-day 24-hr .
Avg. Max.




3.2 9.6
— (5) __(5)

0.021 0.10
0.23 0.65
(1)  Average effluent concentration from monitoring data (Table A-13a).

(2)  Average based on recently received long-term monitoring data submitted by
     Plant #765 (Table 17-14).

(3)  Average based on a 28-day sampling data submitted by Plant #765
     (Table 17-15).

(4)  Also applicable for PSNS limitations.

(5)  No effluent limitation has been established.
                                        587

-------
C.   Selection of Polluants to be Regulated

     For the BAT regulation,  the  Agency  has  selected  chlorine  in
     addition to the pollutants identified in BPT.

D.   Basis of Pollutant Limitations

     1,    Nonconventional Pollutants

          The nonconventional pollutant promulgated for regulation  is
          total  residual  chlorine.   For total residual chlorine the
          BAT regulation is based on long-term  monitoring  data  from
          the   chlor-alkali   industry  given  in  Appendix  A.   The
          long-term average concentration is 0.64 mg/1.   The daily and
          30-day variability factors are 2.3 and 1.4, respectively.

          The 24-hour maximum concentration is:

          {0.64 mg/1) (2.3) = 1.5 mg/1

          The maximum 30-day average concentration is:

          {0.64 mg/1) (1.4) = 0.90 mg/1

          The determination of load limitations for total residual
          chlorine (kg/kkg) was calculated based on the unit flow
          rate of 57 m3/kkg, thus the 24-hour maximum limit is given by

          (1.5 mg/1) (57 mVkkg)    (kg/in*)    = 0.086 kg/kkg
                                 (1000 mg/1)

          The maximum 30-day average was calculated similarly,

          (0.9 mg/1) (57 mVkkg)  (kg/in*)    = 0.051 kg/kkg
                                 (1000 mg/1)

     2.    Toxic Pollutants

          The Agency  has  selected  the  same  limitations  for   free
          cyanide  and  total  cyanide  as  those  promulgated for BPT
          because Level 2 technology does not affect either  of  these
          pollutant parameters.

          The  nonconventional and toxic pollutant limitations for BAT
          are summarized in Table 17-17.

Basis for New Source Performance Standards

Level 2 treatment technology (also promulgated for BAT)  was  selected
as  the  basis  for NSPS limitations.  The pollutants to be controller
for NSPS are pH, total suspended solids, total residual chlorine,  free
cyanide, and total cyanide.  The NSPS limitations are given  in  Table
17-18.  For pH, the NSPS limitation is 6.0 to  10.5
                                   588

-------
                    TABLE 17-17.   EFFLUENT LIMITATIONS
                    HYDROGEN CYANIDE (ANDRUSSOW PROCESS)
                         Best Available Technology
                    Waste Water Flow:   57 m3/kkg of HCN

Daily Concentration Effluent Limit
Variability Basis (mg/1) (kg/kkg)

^ Var^lity 3^ 2«* \°^
Factor Avg Max Avg
Nonconventional
pollutants :
24-hr
Max
(4)
Total Residual
 Chlorine(2)     Q.64      2.3/1.4       0.90     1.5       0.051       0.086

Toxic pollutants:
Free
Total
Cyanide
Cyanide
-U
(2)
0.21
3.3 <3>
8.0/1.
3.4/1.
7
2
0
4
.36
.0
1.7
11
0
0
.021
.23
0.10
0.65
(1)  Average based on recently received long-term monitoring data submitted by
     Plant #765 (Table 17-14).

(2)  Average based on long-term monitoring data from, the chlor-alkali industry
     given in Appendix A.

(3)  Average based on a 28-day sampling data submitted by Plant #765
     (Table 17-15).

(4)  No effluent limitation has been established.
                                     589

-------
Basis for Pretreatment Standards

Pretreatment  is required because NSPS provides better removal of free
and total cyanide than  is  achieved  by  a  well-operated  POTW  with
secondary  treatment  installed and therefore these polutants may pass
through a POTW in the absence of pretreatment.

A.   Existing Sources

     The  Agency  is  excluding  this  subcategory  from  Pretreatment
     Standards  for  Existing  Sources  (PSES) under the provisions of
     paragraph  8(b)(ii)   of  the  Settlement  Agreement  because  the
     concentrations  of  toxic pollutants in the effluent to POTW from
     the one existing indirect discharger are below treatable levels.


B.   New Sources

     For Pretreatment Standards for New Sources (PSNS), the Agency  is
     promulgating  limitations  based on .NSPS excluding dechlorination
     of the final plant  effluent.   Dechlorination  is  not  required
     because  influent to a POTW is often chlorinated.  The pollutants
     to be regulated are free cyanide and total cyanide as  summarized
     in Table 17-16.
                                  590

-------
                     TABLE 17-18 .  EFFLUENT LIMITATIONS
                     HXDFOGEN CYANIDE (ANDRUSSOW PROCESS)
                      New Source Performance Standards
                     Waste Water Flew:  57 m3/kkg of HCN
                              Daily       Concentration       Effluent Limit
                           Variability    Basis  £mg/l)          (kg/kkg)
   Pollutant  Subcategory     -Factor
              Performance     30-day     Max                Max
                 (mg/1)     Variability  30-day    24-hr    30-day      24-hr
                              Factor     Av9       Max      Av9         Max
Conventional and
nonconventional pollutants:

Total Suspended
 Solids, TSS       35        4.2/1-6     56       150       3.2         8.6

Total Residual
Chlorine ,
Anmonia-N
Toxic pollutants:
Free Cyanide
Total Cyanide
0.64
52

0.21
3.3
2.3/1.4
4.2/1.6

8.0/1.7
3.4/1.2
0.90
83

0.36
4.0
1.5
220

1.7
11
0.051
„ (*)

0.021
0.23
0.086
__ (*)

0.10
0.65
*  No effluent limitation has been established.
                                      591

-------

-------
                              SECTION 18

                      SODIUM DICHROMATE INDUSTRY
Industry Profile

General Description

Most of the sodium dichromate produced is used in the chromic acid and
chrome  pigment  industries.  It is used for leather tanning and metal
treatment as well as a corrosion inhibitor.

The industry profile data for this  subcategory  are  given  in  Table
18-1,  and the status of regulations prior to promulgation of this new
regulation is given in Table 18-2.

Subcategorization

Subcategoration in the sodium dichromate industry  is  established  on
the  basis  of  the  manufactured  product.   This  follows  from  the
requirement that the effluent limitations are to be tied to  units  of
production.   Furthermore,  for  the  two  plants  discharging process
wastewater in the dichromate subcategory, the characteristics  of  the
wastewater are similar and therefore the saine treatment technology can
be  applied.  Subcategorization is discussed in more detail in Section
4 of this report.

General Process Description and Raw Materials

The starting materials for the preparation of  sodium  dichromate  are
chromite  ore,  limestone  and soda ash. . When the above materials are
reacted, sodium chromate is formed which is reacted with sulfuric acid
to produce sodium dichromate.  The reactions are given as:

     4FeCr204 + 8NaaC03 + 702 = 8Na2Cr04 + 2Fe203 + 8C02    (1)

     2Na2Cr04 + H2S04 = Na2Cr207 + HZ0 + Na2S04             (2)

Chromite ore is a chromium  iron  oxide  containing  ferrous  chromite
T~FeCr204  or FeOCr203).  Small amounts of alumina, silica and magnesia
are present.  For the preparation  of  sodium  chromate  and  finally,
sodium  dichromate,  high  grade  chromite  ores  are  used containing
approximately 50 percent Cr203.  These ores are  imported  from  South
Africa.

At-the..plant site, the ore is ground to a fine powder, mixed with soda
ash  and  calcined  in  rotary  kilns  at 1100 to 1150 degrees C.  The
reacted product is leached with hot water in  a  leachate  tank.   The
thickener  underflow  is  filtered  and  the  filtrate recycled to the
leachate tank or thickener.  The solid filter cake is dried in  rotary
kilns.   The  alumina  present in the thickener overflow is hydrolyzed
                                    593

-------
TAHtZ 18-L
SDBC3OD33R5r .EHCPIIZ DMA
                     SCOIUM DKSBOMWE
Total sabcategory capacity rate
total subcategory production rate
Number of plants in this subcategory
308 Data on file for
    With total capacity of
    With total production of
    Representing capacity
    Representing production
    Plant production range:
    Average production
    Median production
    Average capacity utilization
    Plant age range:
            Maxbnin
    Haste water flow range:
    Vblune per unit product:
            Minium
            Maxinun
                          140,000 kkg/year
                          136,500 kkg/year
                                3
                                3
                               NA
                          112,000 khg/year
                               NA,
                               82 percent

                           20,700 kkg/year
                           66,800 Jdcg/year
                           37,300 kkg/year
                           24,800 kkg/year
                               77 percent

                                7 years
                               28 years

                              455 cubic maters/day
                              720 cubic meters/day

                                4 cubic metersAkg
                                8 cubic meters/Wag
Sources of data ara Stanford Besearch Institute, Directory of Chemical
Producers, U.SJU* 1977, U.S. Deportment of Qatmerce, Current Industrial
Deports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Beport, "Preliminary Economic Assessment of Effluent Limitations in the
Tpr"n3anic Oimijcal Industry," June,  1578  and "Economic Analysis of Proposed
Revised Effluent Guidelines and Standards for ttie Inorganic Chsnicals Industry,1
     ,  1980.
     Mot Available
                      594

-------
        TABLE 18-2. STATUS OF REGULATIONS - EFFLUENT LIMITATION GUIDELINES

SUBCATEGORY SODIUM DICHROMATE
SUBPART p_ (40 CFR 415.170, 3/12/74)
STANDARDS
BPCTCA
Max. f 1 )
Product kg/kkg
Process Parameters (mg/1)
NaaC^Oy TSS 0.44
(52)
Avg. ( 2 )
kg/kkg
(mg/1)
0.22
(26)
BATEA* NSPS
Max. Avg. Max.
kg/kkg kg/kkg kg/kkg
(mg/1) (mg/1) (mg/1)
No discharge 0.30
of pwwp
Avg.
kg/kkg
(mg/1)
0. 15
          Cr
            +6
 0.0090t4)  0.00050
(0.11)     (0.060)
          Cr(T)
 0.0088
(1.0)
 0.0044
(0.50)
No discharge
of pwwp3

No discharge
of pwwp-*
                               0.0090<4>   0.00050
0.0088     0.0044
  Section 415.173 was remanded and reserved (41 CFR 51606, November 23, 1976).
1 Max. = Maximum of any one day.
2 Avg. = Maximum, average of daiLy values for thirty consecutive days.
3 pwwp = Process wastewater pollutants.
4 The published value in 40 CFR 415.172 and 415.175 is incorrect and should be
  0.0009 kgAkg.
                                     595

-------
and removed  from  the  chromate  solution  as  precipitated  aluminum
hydrate  in slurry form.  The solution is centrifuged and the centrate
is evaporated, to give a concentrated  solution  of  sodium  chromite,
which  is  reacted  with  sulfuric  acid to give sodium dichromate and
sodium sulfate.   Sodium  sulfate  crystallizes  as  anhydrous  sodium
sulfate  from  the  boiling  solution, and the crystals are removed by
filtration.   The  filtrate  is  concentrated   in   multiple   effect
evaporators.  The residual sodium sulfates separate out as solids from
each  of the evaporators while the hot concentrated solution of sodium
dichromate from the  last  effect  of  the  evaporator  is  fed  to  a
water-cooled  crystallizer.  Sodium dichromate crystallizes out and is
centrifuged.  The centrate, or  mother  liquor,  is  returned  to  the
evaporator.    The   sodium   dichromate  crystals  separated  in  the
centrifuge are dried in a rotary drum dryer and then packaged for sale
or stored for use.  Figure 18-1 presents a  generalized  flow  diagram
for the production of sodium dichromate.

Water Use and Waste Source Characteristics

Water Use

Water  is used for noncontact cooling, in leaching, for scrubbing vent
gases and for  process  steam  for  heating.   Water  use  information
provided in 308-Questionnaires is given in Table 18-3.  It is possible
that  the  figures  given  in the 308-Questionnaires may be the amount
going to each unit operation and not the amount added as makeup water.
The  quantities  seem  unusually  high  for  an  industry   practicing
extensive recycling of water, as this one does.

Waste Sources

A.   Spent Ore

     The unreacted ore is removed from the process as a  sludge.   The
     solids  contain  chromium and other impurities originally present
     in the ore.  The waste is disposed as a solid waste in a suitable
     landfill or is slurried with water  and  sent  to  the  treatment
     facility.

B.   Noncontact Cooling Water and Cooling Tower Slowdown

     The noncontact cooling water is either  used  on  a  once-through
     basis  and  discharged or is recycled and the blowdown discharged
     to the treatment facility.  In addition to dissolved sulfate  and
     chloride, the blowdown may contain chromates.

C.   Boiler Blowdown

     The steam used for heating is recovered as condensate, while  the
     boiler  blowdown is discharged to the treatment facility.  It may
     become contaminated with chromium escaping from the process  area
     and hence should be sent to the wastewater treatment facility for
     treatment.
                                    596

-------
                                                        M.
                                                         .TCR
Ul
U3
-J
CRUSHER
wo
QUNDER
1

SODA ASH 	 M BLEHCR
'
•
_„___„_„ KI1DS
1 i
1
flSCKCLE-^ ITMJBBER fc LBOI
FBOM f
PROCESS 1
1
f
1 SOUBBEH
1
1
10 tftSTO
RESI
BBC*
J-WIER
( nun | 	
ai
DUB
CLE


COOCUCTOMffl
MD BOILER
BUMXM)
1
1 TO WSTC
f 4
CRUDE 80OIUH
HCHW* soumc
OOOLD



.P™™ BYVtmn
i
PH
<
CBHTl
i
M
B
CT


ID SALES
OR USB
I
NCNocrracr
A COOJMJ |
T wa
un»


i
i
** f
UJZER
'
EVAPORATION ^Jffl&M CBWWHJGB
i
i
f 	 OJBfUUC
f ACID

JCIDIFIER MWER
i
SCOIUH
PBCCUCT
x 	 saoiuH *
RDCVAL HdOGIMS
1 |
CUdUH SUJDCZ 1
GKHBRAMD 	 *H X
Hf>T|Jip f f
BH£MXW TOHV9IB TO SALES

g^WEN
HCMXK
^^- ooou
^ win

K
CKT
tNG
31 ^
LIQU1
pica
1
D
CT
•
PACKWnC
ID SALES
                       Figure 18-1.  General process diagram for production of sodion dichromate

-------
 TABIE  18-3.
WATER USAGE IN SODIUM DICHRCMATE SUBCATEX3QRY

Source Water usage at plants , fa3/kkg of Na2Qr207)
Plant #398 Plant #376 <3> Plant #493
Nonoontaet cooling 277
Nbnoantact ancillary 0.5
uses
Direct
process contact ^ 5.7^
Indirect process contact 0.9^ -^
(pumps, seals, leaks and
spills)
Maintenance, e.g. 0.5^
cleaning and work area
11.39 5.7
NA 3.12
7.83(4) 2.85
0.2
» 4.16
0.2
washdown
Air pollution control
            2.5
                                (2)
 NA
1.0
Total contact waste
water influent to
treatment
            9.6
                                (2)
11.59
4.25
NA = Not Available
 (1)  Up to 50 percent solids
 (2)  Total recovery and recycle is practiced at this plant.
 (3)  Plant is no longer in operation.
 (4)  Due to a high evaporation rate, there is no discharge from the pcdmary
     pond during 9 to 10 months of the year.  There was no primary pond
     effluent at the time of sampling and only 4,16 nrVkkg of the indirect
     contact sources were being treated and discharged.
                                    598

-------
     The majority of aqueous streams resulting from the manufacture of
     sodium   dichromate   are  recycled.    Streams  recycled  include
     condensates from product evaporation and drying; product recovery
     filtrates;  air pollution control scrubber effluents from  product
     drying,   leaching  and  roasting  kilns;  filter wash waters; and
     equipment - and  process  area  washdowns.    At  two  plants   the
     wastewater,   consisting  of  boiler  and noncontact cooling tower
     blowdown, is  used  to  slurry  the  spent  ore  residue  to  the
     wastewater treatment facility.  At one plant, the only wastewater
     resulting  from  process  operations  is  the  noncontact cooling
     water, which is used on a once-through basis.

Description of Plants Visited and Sampled

Screening

Three sodium dichromate plants were visited and the wastewater streams
sampled.  Plant I4JL3. was sampled ifn the  screening  ghase  and  Plants
#376 and #398 were sampled "in the verification phase.

At Plant #493, the wastewater going to the treatment facility includes
the  boiler  and cooling tower blowdown and a small volume of effluent
from a scrubber on a by-product sodium sulfate operation.   The  total
waste  includes  the  spent  ore  residue,  which  is also sent to the
treatment  facility.   At  the  treatment   facility,   the   alkaline
wastewaters  are reacted with imported acidic industrial waste  {pickle
liquor containing ferrous  iron)  at  an  elevated  temperature  in  a
reactor.    The  chromium  is  reduced  and  precipitated  during  the
reaction.  The reacted waste is sent to clarifiers via holding  tanks.
In  the  clarifiers,  large  quantities  of water are used to wash the
precipitated solids in a countercurrent fashion.  The final  clarifier
overflow,  which  is  the treated effluent,  is filtered and discharged
and the clarifier underflow is disposed of in a quarry.   Figure  18-2
is  a  block  diagram  of  the  treatment  process and indicates which
streams were sampled.  Table 18-4 gives the flow  data  and  pollutant
emissions of the streams sampled.

Verification

At  Plant  #376,  sodium  sulfide   is  used  for simultaneous chromate
reduction and  precipitation.   The  wastewaters  at  this  plant  are
segrated  into  two streams.  One stream consists of the cooling tower
and boiler blowdown and is used for slurrying the spent ore residue to
the  treatment  facility.   The  second  waste  stream   consists   of
storaiwater  runoff  from  both  the  solids  disposal  areas  and  the
production areas.  The first wastewater stream is  mixed  with  sodium
sulfide  during  transportation  and  sent   to a diked containment and
settling pond system.  The sulfide reduces the hexavalent chromium  to
trivalent   chromium,  which  in  turn  is  precipitated  as  chromium
hydroxide.  The solids are settled  in the pond, and  the overflow  from
the  ponds  is  mixed  with  the  second waste stream and reacted with
sufficient  alkaline  sodium  sulfide  to  reduce  the  chromate   and
precipitate  chromium  hydroxide.   The  reacted solution is sent to a
                                   599

-------
                                   RAW WASTE  WATER
                        WATER
                                      SLUDGE TO
                                    LAND DISPOSAL
                                                       IMPORTED ACID
                                                       INDUSTRIAL WASTE
«. TREATED  EFFLUENT
                                                            LEGEND

                                                         SAMPLING POINTS.
Figure 13-2.   Genera? waste water treatment process flow diagram at Plant #493 showing
              the sampling points.   (Sodium dichromate manufacture).
                                        600

-------
TABLE 18-4.   FLCW 2ND POLLUTMTT CONCEWTRMTICN DATA OF THE SAMPLED
              STREAMS FCR PIANT #493 PRODUCING  SODIUM DICHRCMATE

Stream No.

1

2

Haste Stream
Disoription
Raw Waste
Water
Treated
Effluent
Unit Flow TSS Load
CntVkkg (fcg/kkg
of 1^20:207) of ift^C^OJ
4.25 183

28.91* 0.018

Cr+6 Load
(kg/kkg
of Na2Cr2a7>
3.5

0.0001

Chromium
Load
(fcg/kkg
of m&^rflj^
}
3.30

0.072

   This value includes the flow Ii.au the sodiun dichrcrnate plant,  imported
   acid used for neutralization, and the water used for vashing the solids.
                                    601

-------
settling pond where the suspended solids are settled and the  overflow
discharged.   A  simplified  flow  diagram of the wastewater treatment
process is given in Figure 18-3.  Table 18-5 gives the flow  data  and
pollutant emissions for the streams sampled.

Plant  1376 has implemented some changes in the process technology and
treatment  system  since  the  time  of  sampling.    The   dichromate
production  was  converted  from  a "high-lime" process to a "no-lime"
process, using only chromite ore and soda ash as  the  raw  materials.
This  requires  additional  treatment  facilities  to be installed for
removal of impurities from the product solution, which is treated with
acid and lime to remove alumium and vanadium/  respectively.   At  the
time  of  sampling, the data obtained from this plant was considered a
valid part of the data base for assesssing the pollution potential  of
the  industry  and  evaluating viable treatment options.  The chromate
reduction technology being used  was  evidently  subject  to  periodic
problems  associated  with the hazard of the HZS gas production.  This
has been confirmed in treatability studies conducted  by  the  Agency.
With  proper  operation  of  the  treatment system this problem can be
avoided.

At Plant #398, the only effluent produced is  the  noncontact  cooling
water.   The  noncontact cooling water is used on a once-through basis
and is discharged without treatment through two outfalls.   The  solid
waste   residuals   from   the  leaching  process  are  trucked  to  a
state-licensed hazardous waste landfill area.   The  amount  of  solid
waste  residue  disposed  of  is  approximately 290 kg/kkg of product.
Table 18-6 gives the unit flow data and pollutant  emissions  for  the
process effluent.

Toxic Pollutant Concentrations and Loadings

Toxic  pollutants  detected  in the raw wastes during sampling were as
follows:

                Maximum Concentrations Observed (


                                              Verification
Pollutants               Screening        (2 Plants)

Chromium (Total)         250,000             310,000
Chromium (Hexavalent)      	               150,000
Nickel                   13,000                 1,300
Zinc                        580                 1,200
Copper                       35                   240
Lead                          9                   24
Silver                       <0.50                230*
Arsenic                      <10                  <5
Selenium                     <5                   !40**

* Found at one plant only
                                   602

-------
COOLING TOWER
  BLOWDOWN
                                        SODIUM SULFIDE
     WASTE
      MUD
     SLURRY
                                                          SURFACE
                                                          RUNOFF
                     TREATED
                     WATER
                     RECYCLED
                                                                      SETTLING POND
                                    TREATED EFFLUENT
                                                          LEGEND

                                                      WASTE STREAMS SAMPLED.

                                                      AT THE TIME OF SAMPLING,
                                                      ONLY SURFACE RUNOFF v
                                                      (STREAM #3) WAS BEING
                                                      TREATED 111 THE REACTOR.
       Figure  18-3.
General waste water treatment process flow diagram at Plant #376
showing the sampling points.  (Sodium dichromate manufacture)

-------
TABLE 18-5.    FLOW AND POLLUTANT LOADING DATA OF THE SAMPLED VRSTE
               STREAMS FCR PLANT #376 PRODUCING SODUM DICHPCMATE

Stream
No.

Average
Waste Stream Unit Flow TSS load


7]
0.407

NA
0.057
NA

< 0.00004
Chromium
load
(kg/kkg
of Na2Cr2O<7)
1.041

0.808
0.55
0.77

0.0034
*  Due to a high evaporation rate,  there is normally no discharge fron the
   primary pond for 9 or 10 months  of the year.

NA = Itot available
                                   604

-------
TABLE 18-6.   FLOW AND POLLUTANT LOADING DATA OF THE SAMPLED V&STE
              STREAMS FOR PLANT #398 PRODUCING SODIUM DICHRCMATE

Average Observed Loadings
Stream
No.


1

2

Waste Stream
Description


Noncontact
cooling water
Noncontact
cooling water
Unit Flow TSS load Cr*6 load

(m3/kkg (kg/kkg (kg/kkg of
of Na2Cr2O_) of Na2Cr2O7) Na2Cr20^)
71 0.426 NNI*

206 0.55 NNI*

Chronium
load
(kg/kkg of
Na2Cr207)
NNI*

NNI*

* NNI=  No net increase of the pollutant load, compared to the intake
        source.
                                   605

-------
** Noncontact cooling water at one plant only

Individual plant average raw waste loads per  unit  product  found  in
sampling  can  be  found  in  Table 18-7.  A summary of daily and unit
product raw waste loads for all plants sampled can be found  in  Table
18-8.

Based  on  the  total  annual  production  of this subcategory and the
average waste load generated per unit  product,  the  estimated  total
pollutant raw waste loads generated each year for this subcategory are
as follows:
Pollutant
Total Subcategory Raw Waste Load Generation

            Waste Load (kg/year)
Chromium (Total)                   290,000
Cr (Hexavalent)                    210,000
Nickel                               3,700
Zinc                                   330
Copper                                  5 5
Silver                                  20
Lead                                   < 8.2
Selenium                                 4.0
Arsenic                                < 5.0

Pollution Abatement Options

Toxic Pollutants of Concern

The most significant toxic pollutants found are the primary pollutant,
chromium,  and  the common heavy metals often present as impurities in
the chromium ore, notably  zinc  and  nickel.   In  controlling  these
metals  by  the  processes chosen for the treatment models, incidental
removal of other trace toxic metals also occurs.

The existing BPT regulations control  pH,  TSS,  and  chromium   (Table
18-2).   In  the  new promulgated regulations, effluent limitations on
nickel  are  added  to  the  BPT-based  regulations.   Based  on   the
discussion  in  Section 8, there is no BAT effluent limitation set for
zinc,  which  is  also  removed  in  the  hydroxide  precipitation  by
controlling  chromium.   Although  copper, silver, selenium, lead, and
arsenic were detected in trace quantities (Section 18 and Tables  18-7
and   18-8),  these  five  toxic  pollutants did not occur at treatable
concentrations and, therefore, no .regulations on them are being  set.

Process Modifications and Technology Transfer Options

Appropiate process modifications can be made where opportunities exist
for recycle of chrome-bearing wastewaters for recovery  and  reuse  in
the   process  or  for  use  in other product manufacturing operations.
Plant #398 currently practices extensive recovery of  chromium   values
                                   606

-------
TfcHLE 18-7.
TOXIC POLLUTANT .RAW WASTE DATA

SUBCATEGORY
POIiLUTANT
Chromium, Cr
Copper, Cu
Lead, Pb
Nickel, Ni
Zinc, Zn
Silver, Ag
Selenium, Se
Arsenic, As
SODIUM
DICHRCMATE
AVERAGE RAW WS
\, p
PLANT #493 ,
(mg/i) (fcg/kkg) :
250.0
0.035
0.0090
1.25
0.580
< 0.005 <
< 0.005 <
< 0.010 <
0.94
0.00013
0.000030 :
0.0047 '•-
0.0022
0,00002 :
0.00002
0.00004
A ':

TE INFLUENT
PLANT
420.0
0.085
0.011
0.64
0.318
0.036
< 0.005
< 0.005

#376 (V ^
(kg/kkg)
3.30
0.00067
0.000090
0.0050
0.0025
0.00028
< 0.00004
< 0.00004
                                   607

-------
TABLE 18-8.
       OF RAW msTE LOADINGS FOUND IN
SCREENING AND VERIFICATION SAMELIN3
SUBCATEGORY
         SODILM D3SCHFCMA1E
Ibllutant
        Unit loading,  (fcg/kkg)
                     Minimum
                Average
Maximum
ND. of
Plants
Toxic
Chromium, total
Chromium,
Hexavalent
Copper
Nickel
Silver
Zinc
Selenium
Arsenic
Conventional
TSS

0.94 2.12
0.47 1.6
0.00013 0.0004
0.0047 0.027
0.000020 0.00015
0.0022 0.0024
* < 0.00003
* < 0.00004
140 2100

3.30
2.6
0.00067
0.050
0.00028
0.0025
*
*
4000

2
3
2
2
2
2
2
2
2
*  Concentrations ware at or below the detection limits
                                   60S

-------
for  use  in  other  processes
contact wastewaters.

Best Management Practices
and has no discharge of direct process
Extensive recycle and  reuse  of  process  contact  wastewater  limits
effluent  generation  at sodium dichromate plants.  At two facilities,
cooling water blowdown streams are used to slurry spent  ore  residues
and  the resultant waste stream is treated for the removal of chromium
prior to discharge.  At the remaining plant, ore residues are  removed
as  a  solid  waste  and only once through noncontact cooling water is
discharged.

Prevailing Control and Treatment Practices

At the time of verification sampling, Plant #376  was  using  alkaline
sodium   sulfide  (or  bisulfide)  for  the  reduction  of  hexavalent
chromium, followed by precipitation of metal sulfides and  hydroxides.
Problems experienced by the plant included intermittent, low level H2S
gas  generation  and  incomplete  reduction  of  the chromates.  These
problems were mitigated by the physical layout of the treatment system
and lagoons and the long retention time afforded  by  the  evaporation
ponds during most of the year.

At  Plant  #493, the treatment technology employed is the reduction of
chromate wastes with an acidic ferrous  iron  solution  (waste  pickle
liquor),  followed by lime addition for metal hydroxide precipitation,
settling,  and  filtration.   Overall,  this  technology  is   roughly
equivalent  to  the sulfide reduction/alkaline precipitation technique
used by Plant #376 and has  the  advantage  of  not  risking  operator
exposure to hydrogen sulfide gas.

Advanced Treatment Technologies

In  addition  to  the  chromate reduction and metal removal techniques
practiced in the sodium dichromate industry, consideration  was  given
to  other advanced treatment technologies considered to be equal to or
better than the proposed BAT.  These technologies include:

     The use of sulfur dioxide for chromite reduction.

     Ferrite coprecipitation i.e., the addition of ferrous iron (e.g.,
     waste pickle liquor) and  aeration  at  about  pH  5-6  for  both
     chromate reduction and metals precitation.

     Ion exchange systems.

     Xanthate precipitation.

These  options are not considered viable at this time because there is
not sufficient information on performance and cost effectiveness.
                                   609

-------
Selection o£ Appropriate Technology and Equipment

Technology for Different Treatment Levels

Alkaline precipitation or reaction with hydroxide will separate nickel
and zinc from solution.  Hexavalent chromium must be  reduced  to  its
trivalent  form  before  it  can  be  precipitated  as  the hydroxide.
Although ion exchange or xanthates can remove  metals  from  clarified
solutions  they are inappropriate for treating raw waste slurries from
this industry.

A.   Level 1  (BPT, BAT, NSPS)

     The system utilizes pickle liquor  containing  ferrous  iron  and
     hydrochloric  acid  added  to the raw wastes to reduce hexavalent
     chromium to its trivalent form  in  a  first-stage  lagoon.   The
     lagoon  effluent  is  then subjected to alkaline precipitation of
     trivalent chromium, followed by solids separation in a  clarifier
     and  by  pH  adjustment  of the overflow before discharge.  Other
     reducing agents may be utilized instead of ferrous iron  for  the
     reduction of hexavalent chromium such as sodium sulfide or sulfur
     dioxide.   Using  either  of  these  reagents, chrornate reduction
     under acid conditions would be followed  by  pH  adjustment  with
     lime  or  caustic  to  obtain alkaline precipitation of the metal
     hydroxides.  A number of operational difficulties were associated
     with  the  treatment  with  sodium   sulfide.    Based   on   the
     Treatability  Studies  on the dichromate subcategory (61), it was
     found that at pH  levels above 8, the addition of  sodium  sulfide
     required  excessive  reaction  times  to  reduce  chromate to the
     trivalent form.  At pH levels below 8, the evolution of  H2S  gas
     poses a potential safety hazard for the plant operator.  The flow
     diagram for the ferrous iron-based option for Level 1  is shown in
     Figure  18-4.

B.   Level 2

     Dual-media filtration is added  to  achieve  a  higher  level  of
     suspended solids  removal, including metallic hydroxides which may
     have passed through the clarifier.  The effluent is adjusted to  a
     pH  range  of  6  to  9  as in Level 1.  At proposal,  Level 2 was
     selected as a possible BAT and  NSPS  treatment  because  it  was
     being practiced by one plant in the industry and it could provide
     a  method  of removing additional quantities of toxic  metals from
     the wastewater.   However,  the  incremental  cost  of  dual-media
     filtration  applied  to Level 1 is measurably more costly for the
     Sodium  Dichromate Subcategory than other subcategories  and  does
     not  significantly  improve  effluent  quality.  The Treatability
     Studies on the dichromate subcategory  (61) showed that dual-media
     filtration is only marginally effective in reducing toxic  metals
     and  TSS beyond BPT.  Therefore, this  level has not been selected
     as the  basis for  BAT or NSPS.  The flow diagram for  the  ferrous
     iron-based option for Level 2 is shown in Figure 18-5.
                                    610

-------
C.   Equipment Functions

     The raw waste flows into an equalizing lagoon, where the influent
     flows are measured  by  a  magnetic  flow  meter  which  controls
     application of pickle liquor solution into the influent pipeline.
     Hexavalent chromium is converted to the less toxic trivalent form
     and  together with inert solids passes to the first-stage lagoon.
     A second application of  ferrous  iron  is  made  in  the  lagoon
     outflow,  and lime is added to precipitate trivalent chromium and
     residual trace metals prior to clarification.   In  Level  1  the
     clarifier effluent is adjusted to pH 6 to 9 and released.  In the
     Level  2 system a dual-media filter is added to remove additional
     suspended material from the overflow.   Clarifier  underflow  and
     filter  backwash  are returned to the equalizing lagoon influent,
     to be settled in the lagoon.

D.   Chemicals and Handling

    "Ferrous chloride, hydrochloric acid and  lime  are  used  in  the
     treatment  process.   The  ferrous chloride and hydrochloric acid
     solution can be applied at the influent pipeline in proportion to
     flow.  Lime slurry is fed through conventional equipment ahead of
     the clarifier.  Hydrochloric acid is used  (instead  of  sulfuric
     acid)  to  minimize  the  formation  of  gypsum scale which could
     result from heavy use of lime followed by sulfuric acid.

E.   Separation and Disposal of Solids

     As a basis for estimating model plant costs,  influent  suspended
     solids,  metallic hydroxide precipitates, and filter backwash are
     returned to, or left in, the influent lagoon(s).  As each  lagoon
     becomes filled with solids it is decanted from each filled lagoon
     and the solid material must be periodically removed to a chemical
     landfill.

F.   Monitoring Requirements

     Internal process monitoring should include both  routine  testing
     to  maintain reducing conditions and a pH above 7 in the influent
     lagoons, and simple field determination of pH to assure that  the
     optimum  level is reached for precipitation of chromic hydroxide.
     Routine testing of the effluent should also be performed  at  the
     site to show that hexalent chromium is being consistently reduced
     to  trivalent  chromium  and  that  total  chromium  in the final
     effluent does not exceed the allowable limit.  Periodic composite
     effluent samples should be analyzed for  total  chromium  by  the
     atomic absorption method for official reporting purposes.
                                     611

-------
FERROUi
JRCH
rOl
i :
RAW |
g WASTE i i

MAGNETIC
METER

T
I

1
1
1
1
1
-«X LAGOON /-M
•

-«\ LAGOON /— •»

FEB
IK
!
•
MWHH


BOOS I
CM ^
!

W?-
MPC
TANK

1LIME
QpH ADJUSTMENT
-_ . 	 «„„
9 1 ta)
1 u' u Y

Y^LARIFIER
1
1
          Includes flow monitoring, pH monitoring and sampler.
Figure  18-4.  Level 1 waste water treatment for sodium dichrcmate subcategory.

-------
                                                                      ~1
       FERROUS
        IRON   |
FERROUS
                                                 LIMB
RAW
    MAGNETIC
     MBTER
                       LfiOOON
                •A   LAGOON     /-»
         I
                                               MIX
            i
-U-o^E
                                                              SIMP     Fn/ralR
                                                    *
                                                  BEFUJENT
                                                          OARIFIER
                                                         \
                                                       -t
                 ^Includes flow monitoring, pH monitoring and sampler.
        Figure 18-5.  Level  2 waste water treatment for sodium dichromate subcategory.

-------
Treatment Cost Estimates

General Discussion

Model  plant  specifications  were  selected  for
estimation.   The  rationale  for  the  selection
characteristics is as follows:

A.   Production
the purpose of cost
 of   model   plant
     At the time of sampling, five industrial plants  produced  sodium
     dichromate  at  a  total production rate of approximately 140,000
     kkg/year.  Two of  these  plants  have  discontinued  production.
     Production  and  wastewater  flow  data,  from  which model plant
     characteristics are derived, are on file for three  plants  which
     produce  a total of 112,000 kkg/year, or approximately 80 percent
     of the United States production.  For wastewater  treatment  cost
     estimates,  three  production  levels  were  selected.  These are
     20,000 kkg/year, 50,000 kkg/year and 70,000 kkg/year.

B.   Wastewater Flow

     Unit waste flows for three plants either  treating  or  recycling
     their  wastewaters  are approximately 9.6, 11.59, and 4.25 mVkkg
     of product.  For the model plant, 8.5 mVkkg of sodium dichromate
     was used as the wastewater flow.

C.   Pollutant Loading

     For the model plant, it is assumed that the  spent  ore  residues
     are  slurried  and  transported  to the treatment facility, since
     this is the prevalent practice at  two  plants.   The  spent  ore
     (waste-generated  residue)  at Plant #969 is 290 kg/kkg NazCr207.
     The hexavalent chromium loading  in the wastewater varies from 0.5
     to 14 kg/kkg of Na2Cr207.  Pollutant loadings used for the  model
     plants  are  suspended  solids   (spent ore residue) at 290 kg/kkg
     NazCr207 produced, and hexavalent chromium at 5.0 kg/kkg.

D.   Chemicals Required

     To reduce hexavalent chromium to  trivalent  chromium  a  ferrous
     chloride  dosage  of 4.3 g/1 is  needed, but to allow for reaction
     with other metals, a model dosage of 5.0 g/1 was used.   This   is
     equivalent  to  42.5  kg/kkg  of  product  in  a unit flow of 8.5
     mVkkg.  To raise the pH to 9.5,  1.7  g/1  of   lime  is  needed,
     equivalent  to  15  kg/kkg of product.  For final neutralization,
     HC1 is used in the amount of 10  percent of the lime dosage.

E.   Solids Generated

     Tota] dry solids produced from  treatment are 260 kg/kkg of sodium
     dichromate.
                                   614

-------
F.   Model Plant Control Costs

     The cost estimates of three models  having  different  production
     levels are presented in Tables 18-9, 18-10, and 18-11.

     Table 18-12 gives a summary of the unit cost distribution between
     amorization, and the operation and maintenance cost components at
     various production rates and levels of treatment.

     At  the  first  level  of  treatment,  investment  costs are high
     because sludge lagoons costs are provided for a ten-year  period.
     Therefore,  amortization is the major portion of the total annual
     costs.  In place of annual cost for the residual  waste  (sludge)
     disposal,   a  large  investment  in land is shown.  At the second
     level of  treatment,  labor  and  amortization  have  significant
     impact on the additional annual costs.

Basis for Regulations

BPT Effluent Limitations

A.   Technology Basis

     BPT  regulations  for  the  Sodium  Dichromate  subcategory   are
     presently in effect, 40 CFR 415.172 {Table 18-2).   The technology
     basis  for  the  existing  BPT is ferrous reduction of hexavalent
     chromium,   followed  by  alkaline  precipitation  of  metals  and
     clarification.  As an alternative to the use of ferrous iron, the
     reduction  of hexavalent chromium may be accomplished by reaction
     with sodium sulfide or sulfur dioxide  under  acidic  conditions.
     All   three   plants  in  this  subcategory  have  installed  BPT
     technology and are meeting the limits.

     The control of suspended solids is necessary to  the  achievement
     of good effluent quality after precipitation of heavy metals.  In
     the  Sodium  Dichromate  subcategory,  it  can  be  assumed  that
     chromium is a significant constituent  in  the  suspended  solids
     discharged.    For  this  reason,  only  one  advanced  treatment
     alternative, addition of a filtration unit  for  solids  control,
     has been considered.

B.   Response to Remand Issues

     The zero discharge requirements originally promulgated as BAT for
     sodium dichromate  production  were  remanded  on  the  basis  of
     inadequate   technical   and   economic   justification  for  the
     evaporative  technology  required  to  eliminate  discharge.    A
     control   and  treatment  alternative,  which  allows  wastewater
     discharge,  has  been  identified  and  the  performance   levels
     achievable have been demonstrated.
                                  615

-------
  TABLE 18-9.
  MODEL PLANT TREATMENT COSTS
Subcategory
Production
Sodium dichromate
20,000 metric tons per year
A. INVESTMENT COST

     Site development 	
     Equipment 	
     Monitoring equipment  	

          Subtotal	
     Contractor's 0 & Pb	

          Subtotal 	
     Engineering  	

          Subtotal 	
     Contingencies 	

          Subtotal 	

     Land	

     TOTAL INVESTMENT COST 	


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision 	
     Energy 	
     Chemicals 	
     Maintenance  	
     Taxes and insurance  	
     Residual waste disposal  ....
     Monitoring,  analysis
      and reporting 	

     TOTAL OPERATION AND
     MAINTENANCE  COST
                                                    ($)
                           Level 1

                           608,700
                           185,000
                            20,000

                           813,700
                           122,055

                           935,755
                           187,151

                         1,122,906
                           112,291

                         1,235,197

                           156,000

                         1,391,197
                            56,000
                             2,500
                            17,000
                           123,520
                            41,736
                                 0

                            15,000
                           255,756
Level 2'

     0
38,000
     0

38,000
 5,700

43,700
 8,740

52,440
 5,244

57,684

     0

57,684
14,000
   600
     0
 5,768
 1,731
     0

 7,500


29,599
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                           200,966

                           456,722
 9,385

38,984
 a Represents the  incremental cost  above  that  for  BPT  treatment
  Overhead and Profit
                              616

-------
   TABLE 18-10.  MODEL  PLANT TREATMENT  COSTS
Subcategory
Production
Sodium dichromate
50,000 metric tons per year
A. INVESTMENT COST

     Site development  .-	
     Equipment  	
     Monitoring equipment  	

          Subtotal	
     Contractor's 0 &  P b.	

          Subtotal  ....	
     Engineering  	,	

          Subtotal ,	
     Contingencies	

          Subtotal  	
     Land	,	

     TOTAL INVESTMENT  COST  	


B. OPERATION AMD
   MAINTENANCE COST

     Labor and supervision  ......
     Energy	
     Chemicals	
     Maintenance	..
     Taxes and  insurance 	
     Residual waste disposal  ....
     Monitoring , analysis
      and reporting 	

     TOTAL OPERATION AND
     MAINTENANCE COST
                                                     ($)
                             Level  1

                         1,300,000
                           315,000
                             20,000

                         1,635,000
                           245,250

                         1,880,250
                           376,050

                         2,256,300
                           225,630

                         2,481,930
                           252,000

                         2,733,930
                            56,000
                             2,800
                            42,000
                           248,193
                            82,018
                                 0

                            15,000
                           446,011
 Level 2

      0
 89,000
      0

 89,000
 13,350

102,350
 20,470

122,820
 12,282

135,102
      0

135,102
 14,000
  1,000
      0
 13,510
  4,053
      0

  7,500
 40,063
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                           403,810

                           849,821
 21,981

 62,044
a Represents the incremental cost above that  for  BPT  treatment
  Overhead and Profit
                             617

-------
   TABLE 18-11.  MODEL PLANT TREATMENT COSTS
Subcategory
Production
Sodium dichromate
70,000 metric tons per year
A. INVESTMENT COST

     Site development	
     Equipment 	
     Monitoring equipment 	

          Subtotal ...**	
     Contractor's 0 & Pb.	

          Subtotal 	
     Engineering  	

          Subtotal 	
     Contingencies	

          Subtotal 	
     Land	,	

     TOTAL INVESTMENT COST  	


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  	
     Energy  	
     Chemicals 	
     Maintenance  	
     Taxes and insurance	
     Residual waste disposal  ....
     Monitoring,  analysis
      and reporting  	

     TOTAL OPERATION AND
     MAINTENANCE  COST
                            Level  1

                         1,766,000
                            405,000
                            20,000

                         2,191,000
                            328,650

                         2,519,650
                            503,930

                         3,023,580
                            302,358

                         3,325,938
                            324,000

                         3,649,938
                                                    ($)
                             56,000
                              2,800
                             58,000
                            332,594
                            109,498
                                  0

                             15,000
                            573,892
                                                        Level 2'
      0
105,000
      0

105,000
 15,750

120,750
 24,150

144,900
 14,490

159,390
      0

159,390
 14,000
  1,000
      0
 15,939
  4,782
      0

  7,500
 43,221
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                            541,130

                          1,115,022
 25,933

 69,153
  Represents the  incremental  cost  above  that  for  BPT treatment
  Overhead and Profit
                              618

-------
        TABLE  18-12  MODEL  PLANT  UNIT TREATMENT COSTS
     Subcategory Sodium Dichromate
                                Annual  Treatment  Costs ($/kkg)
    COST ITEM
PRODUCTION
 (kkg/yr)
 LEVEL OF TREATMENT

Level 1    Level 2
Annual Operation
and Maintenance


Annual
Amortization


Total Annual
Cost



20,000
50,000
70,000

20,000
50,000
70,000

20,000
50,000
70,000

12.79
8.92
8.20

10.05
8.08
7.73

22.84
17.00
15.93

1.48
0.80
0.62

0.47
0.44
0.37

1.95
1.24
0.99
*Represents the incremental cost above BPT
                             619

-------
C.    Flow Basis

     The model plant waste  flow  rate  is  based  on  the  raw  waste
     influent  data obtained from three plants as shown in -Table 18-3.
     The flow rate selected, 8.5 mVkkg, is the average of  the  flows
     for  these  three  plants.    All three plants are included in the
     flow averaging because the waste sources  were  typical  for  the
     industry  at  the  time  of  sampling  and represent the range of
     inflow rates expected to be handled by a BPT treatment system.

D.    Selection of Pollutants to be Regulated

     For BPT regulations the Agency is retaining the  pollutants  that
     are  presently  limited under 40 CFR 415.172, which are pH, total
     suspended solids (TSS), hexavalent  chromium  (CrVI),  and  total
     chromium  (Cr)  and  is adding nickel.  The significance of these
     pollutants is substantiated by  the  screening  and  verification
     data presented in Section 18.

     The  available  treatment  technology for the removal of chromium
     from wastewater necessitates the reduction of hexavalent chromium
     (chromate or dichromate) to the trivalent state which can then be
     precipitated as  chromic  hydroxide,  Cr(OH)3.   Thus,  from  the
     regulatory point of view, an effluent limitation on the discharge
     of total chromium effectively limits hexavalent chromium as well.
     But,  placing limitations on both forms of chromium in the Sodium
     Dichromate subcategory is consistent with the  primary  objective
     of  controlling  specifically the highly toxic hexavalent form by
     means of a two-step treatment process.  In light of the potential
     analytical  difficulties  associated  with  the  measurement   of
     hexavalent  chromium  discussed in Section 5, monitoring both the
     hexavalent chromium and the total chromium content of the treated
     effluent provides an  additional  assurance  that  high  chromate
     levels  would not go undetected.  As treatment system performance
     data  are  accumulated,  support  may  develop  for  a  decreased
     monitoring requirement or the elimination of effluent limitations
     on hexavalent chromium.

     Nickel  is  included in the limitations since it provides a means
     of controlling the group of toxic metals represented  by  removal
     at  slightly  higher  pH  values  than  for chromium.  A detailed
     discussion of the two toxic metal groups  represented  by  nickel
     and  chromium  is  in  Section 8.  Control of nickel and chromium
     will ensure that toxic metals  that  may  occasionally  occur  at
     treatable levels will be adequately controlled.

E.    Basis for Pollutant Limitations
     1.
Conventional Parameters -
          a.
     PH
                                   620

-------
2.
     After final  pH adjustment,  the BPT treated effluent is to be
     held within  the pH range of 6 to 9.   The  pH  limitation  is
     based  on Appendix  B  of  the proposed Development Document
     (60) and a study report,  "An Assessment  of  pH  Control  of
     Process  Waters in Selected Plants"   by JRB Associates,  Inc.
     (52).
     b.
     TSS
     The present study substantiates the basis for  the  existing
     BPT  limitation  on  total   suspended  solids.    The treated
     effluent sampling data from two plants  presented  in  Table
     18-13  suggest that the TSS concentrations found in sampling
     represent achievable performance  of  a  well  operated  BPT
     system.    This is in agreement with the 26 mg/1 TSS which is
     the concentration basis  for  the  existing  maximum  30-day
     average  effluent  limitation (Table 18-2).  For comparison,
     Table A-lla summarizes the   long-term  data  available  from
     another  subcategory  where a similar BPT is applied.  Plant
     #376  discharged  an  average  TSS  of   11   mg/1   without
     filtration.   Monitoring data  from Plant #493 shown at the
     bottom of Table 18-13 indicates that 25 mg/ is an achievable
     maximum 30-day  average  for  TSS  with  filtration.   Thus,
     individual  plant performance can be seen as a function of a
     very  large  number  of  operating   variables   and   waste
     characteristics.  In general, the available performance data
     support the achievabililty  of the existing regulations.

     The  variability  factor ratio (VFR) of 2.0 is derived from
     the long-term  data  on  chromium  as  presented  in  Tables
     A-9a-l,   and  following. This VFR value is used for TSS and
     chromium because a significant  proportion  of  the  TSS  is
     composed  of  suspended  metal hydroxides resulting from BPT
     treatment.  For TSS, the maximum 30-day  average  limitation
     is  related  to  the concentration basis and the model plant
     flow as follows:
      26 mq/1)  8.5 m /kkq)
               (1000 mg/1)
                       (kg/m*)
=0.22 kg/kkg
and the daily maximum limitation is obtained by  multiplying
by the VFR;

(2.0) (0.22 kg/kkg) = 0.44 kg/kkg

Toxic Pollutants
     a.
     Chromium
     For BPT,  the Agency is retaining the existing limitations on
     total and hexavalent chromium as given  in  40  CFR  415.172
     (Table   18-2).     For  hexavalent  chromium,  the  observed
     performance level of 0.004 mg/1 was below the accepted lower
                                621

-------
         TABLE 18-13.    EFFLUENT SAMPLING DATA FPCM
                         SODHM BICHROMATE PLANTS

Pollutant

Ttotal Suspended
Solids, TSS
Hexavalent
Chrcmium, Cr (VI)
Tbtal
Chrcmium, Cr (T)
Copper, Cu
Nickel, Ni
Selenium, Se
Silver, Ag
Zinc, Zn
Flow (m3/kkg)

TSS
Cr (VT)
Cr (T)
Screening & Verification Data
Plant #376 Plant #493
(mg/1)
11
< 0.01
0.81
0.012
0.20
< 0.005
0.015
0.008

Long
(mg/1)
25
0.023
0.072
(kg/kkg) (mg/1) (to^Akg)
0.046 2.0 0.0085
< 0.00004 0.004 0.00002
0.0034 2.5 0.011
0.00005 0.016 0.00007
0.00083 0.090 0.00038
< 0.00002 0.10 0,00043
0.00006 < 0.007 < 0.00003
0.0003 0.11 0.00047
4,16 4.25
Term Monitoring Data-Maximum 30-Day Averages
Plant #493(1)(2)
(kg/kkg)
0.11
0.00010
0.00031
(1)   Filtered effluent data reported in response to SOS-Questionnaire
     (12-22-76)

(2)   The nunber of samples is unknown.
                                   622

-------
limit of treatability (0.05  mg/1)  from  Table  8-11.   The
treatability  level  was  the  basis  of  the 30-day average
concentration basis of 0.060 mg/1 used for the existing  BPT
regulations (Tables 18-2 and 18-14).

The  VFR  of 4.0/2.0 used for total chromium is confirmed by
long-term data (Tables A-9a-l, and  following)  on  alkaline
precipitation of chromium in another subcategory where a 1.8
value was determined for similar BPT technology.

The  existing  24-hour  maximum effluent limitation that was
published for hexavalent chromium is in error as it  appears
in  40  CFR  415.172.    The  correct  value is 0.0009 kg/kkg
reflecting an overall  VFR value of approximately 1.8 for the
residual hexavalent chromium remaining  after  the  two-step
treatment process.

For  total  chromium,   the maximum 30-day average limitation
is:

(0.50 mq/1) (8.5 m /kkq) (kq/m3)  = 0.0044 kg/kkg
                    (1000 mg/1)

and, by applying the VFR value of 2.0, the daily maximum is:

(2.0) (0.0044 kg/kkg)  = 0.0088 kg/kkg

For  hexavalent  chromium,   the   maximum   30-day   average
limitation is:

(0.60 mq/) (8.5 m /kkq)  (kq/m3)   = 0.00050 kg/kkg
                  (1000 mg/1)

and  the  24-hour maximum is obtained using the VFR value of
1.8, that is:

(1.8) (0.00050 kg/kkg) = 0.00090 kg/kkg

b.    Nickel

The long-term average  nickel concentration was selected from
Table 8-12 which presents the industrial  wastewater  system
performance.    The  variability  factors  are  based  on the
primary pollutant chromium.

The maximum 30-day average concentration  basis  for  nickel
is,

(0.20 mg/1) (2.0) = 0.40 mg/1

The 24-hour maximum concentration basis for nickel is,

(0.20 mg/1) (4.0) = 0.80 mg/1
                            623

-------
                       TABLE 18-14.  EFFLUENT LIMITATIONS*
                                SODIUM DICHROMATE
             Best Practicable Control Technology Currently Available
                           Wastewater Flow: 8.5 m3Akg
Pollutant
Subcategory
Performance
  (mg/l)
                                VFR(
Concentration   Effluent Limit
Basis (mg/l)     (kg/kkg)	
Max.            Max.
30-day   24-hr. 30-day    24-hr.
Avg.	Max.   Avg.	Max.
Conventional
 Pollutants:

Total Suspended
 Solids

Toxic Pollutants:
              2.0
26
52
0.22
0.44(2)
Total Chromium(8)
Hexavalent
Chromium*8)
Nickel*8*
Zinc
0.
0.
0.
0.
25U)
050(5)
20(4)
50(4)
4.
1.
4.
2.
0/2. 0(5)
8(6)
0/2.0<5)
0
0
0
0
0
.50
.060
.40
.50
1.
0.
0.
1.
0
11
80
0
0.
0.
0.
0044
00050
0034
--(7)
O.OOSS*2)
0.00090(2)
0.0068
— (7)
(1)  VFR: Ratio of the 24-hour variability factor to the 30-day variability
          factor.

(2)  Existing regulations, 40 CFR 415.72 (Table 18-2).

(3)  Long-term monitoring 30-day averages from Plant #493 (Table 18-13).

(4)  Industrial Wastewater System Performance (Table 8-12).

(5)  The VFR used in original regulation is confirmed by long-term data on alkaline
     precipitation of chromium in another subcategory (Tables A-9a-1, etc.)

(6)  VFR used in original regulation.

(7)  No limitations.

(8)  Applicable to BAT, and PSNS.

* -  Also applicable to New Source Performance Standards (NSPS).
                                       624

-------
          The maximum 30-day average limitation for nickel is,

          (0.40 mq/1) (8.5 mg/kkq) (kq/m^) = 0.0034 kg/kkg
                             (1000 mg/1)

          The 24-hour maximum limitation for nickel is,

          (0.80 mq/1) (8.5 mVkkg) (kg/m^) = 0.0068 kg/kkg
                               (1000 mg/1)
          c.
Other Metals
          The concentration basis for zinc  is  also  given  in  Table
          18-14.  This  and  other  similar metals will be effectively
          removed by the BPT  alkaline  precipitation  step.   Copper,
          silver,  selenium,  arsenic,  and  lead  did  not  occur  at
          concentrations high enough to be treatable and are therefore
          not regulated.  An  adequate  removal  of  these  metals  is
          expected  with  a BPT treatment system specifically designed
          to provide  optimum  conditions  for  the  precipitation  of
          nickel and chromium.

BCT Effluent Limitations

While   EPA  has  not  yet  proposed  or  promulgated  a  revised  BCT
methodology in  response  to  the  American  Paper  Institute  v.  EPA
decision  mentioned  in Section 3, EPA is promulgating BCT limitations
for this subcategory.  These limits are identical to  those  for  BPT.
EPA  is  not promulgating any more stringent limitations since we have
identified  no  technology  option  which  would  remove   significant
additional  amounts of conventional pollutants.  As BPT is the minimal
level of control required by law, no possible application of  the  BCT
cost   tests   could  result  in  BCT  limitations  lower  than  those
promulgated in this regulation.  Accordingly, there is no need to wait
unitl  EPA  revises  the  BCT  methodology  before  promulgating   BCT
limitations.

BAT Effluent Limitations

For  BAT,  the  Agency  is setting limitations equal to BPT, since the
cost of dual-media filtration is too high to  justify  the  relatively
small  additional toxic metal removal.  The pollutants limited include
hexavalent chromium, total chromium and nickel which are the  same  as
presented in Table 18-14.

NSPS Effluent Limitations

For  NSPS,  the  Agency is setting limitations equal to BPT, since the
cost of dual-media filtration is too high to  justify  the  relatively
small  additional toxic metal removal.  The pollutants limited include
TSS, total chromium, hexavalent chromium, and  nickel  which  are  the
same as presented in Table 18-14.
                                  625

-------
Pretreatment Standards

The  Agency is promulgating pretreatment regulations that are equal to
NSPS in order to provide  better  removal  of  toxic  metals  than  is
achieved  by  a well-operated POTW with secondary treatment installed.
These  pollutants  would  pass  through  a  POTW  in  the  absence  of
pretreatment.

A,   Existing Sources

     The  Agency  is  excluding  this  subcategory  from  pretreatment
     standard  for  existing  sources  (PSES)  under the provisions of
     paragraph 8(b) of the Settlement Agreement because there  are  no
     existing indirect dischargers.

B.   New Sources

     There is an existing pretreatment standard for new sources (PSNS)
     in effect (40 CFR 415.176) which is based on BPT treatment.   The
     Agency  is  retaining this BPT treatment for NSPS and is limiting
     total chromium, hexavalent chromium, and nickel as  presented  in
     Table 18-14.
                                   626

-------
                              SECTION 19

                       CARBON DIOXIDE INDUSTRY
Summary of Determinations

We  have determined that no further effort need be given to developing
BPT, BAT, NSPS, and Pretreatment regulations for  the  Carbon  Dioxide
Subcategory.   The  basis  for  this  determination  is  that no toxic
pollutants were found at significant levels  in  the  process  related
wastewater  during  the  screening  of  one plant.  The subcategory is
excluded under Paragraph 8 of the Settlement Agreement.

Assessment of the Water Pollution Potential

Production Processes and Effluent

Carbon dioxide is produced in gaseous, liquid, or solid form.  Most of
the carbon dioxide is produced as a by-product of ammonia  production.
A  major portion of the carbon dioxide is used captively for producing
urea and secondary recovery of oil and natural gas.  It is  also  used
for  refigeration,  in  the  food  industry  for  the  carbonation  of
beverages, in fire extinguishing equipment, and oil well stimulation.

The process wastewater is derived from gas scrubbing and condensation.
The only toxic pollutant found at a significant concentration  in  the
raw waste during screening at one plant was zinc (910 pg/1).  When the
data  was  reviewed  with  plant personnel, it was discovered that the
zinc level was due to zinc corrosion inhibitors and  was  not  process
related.  Control of zinc from this type of source is best achieved by
management  on  a case-by-case basis by the permitting authority.  The
subcategory profile data is given in Table 19-1.
Maximum concentration of toxic pollutants found in
plant were:

                         Pollutant

                         Zinc
                         Copper
                         Chromium

Status of Regulations

Subpart AF has been reserved for this subcategory.
screening  at  one
                               627

-------
TABLE  19-1-   -
SUBCKEBQOKf PROFILE DATA SUMMARY
 SUBCAIEQ3R*
CARBON DIOXIDE
Total subcategory capacity rate
Total subcategory production rate
Number of plants in this  subcategory
308 Data on file for
    With total capacity of
    With total production of
    Representing capacity
    Representing production
    Plant production range:
             Minimum
             Maximum
    Average production
    Median  production
    Average capacity utilization
    Plant age range:
             Minimum
             Maximum
    Waste water flow range:
             Minimum
             Maximum
    Volume  per unit product:
             Minimum
             Maximum
                        12,194,000 kkg/year
                         1,819,000 kkg/year
                               105
                               12
                           713,947 kkg/year
                           558,667 kkg/year
                               59 percent
                               31 percent

                             1,600 kkg/year
                           155,000 kkg/year
                               NA
                               NA
                               NA

                                6 years
                               50 years

                               NA
                               NA

                               NA
                               NA
Sources of data are Stanford Research Institute, Directory of Chemical
Producers, U.S.A., 1977, U.S. Department of commerce, Current Industrial
Reports, Decenber 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chsnical Industry," June, 1978 and "Economic Analysis  of  Proposed
Revised Effluent Guidelines and Standards for the Inorganic Chemicals Industry,"
March, 1980.
NA = Not Available
                                   628

-------
                              SECTION 20

           CARBON MONOXIDE AND BY-PRODUCT HYDROGEN INDUSTRY
Summary of Determinations

We have determined that no further effort need be given to  developing
BAT,  NSPS,  and  Pretreatment regulations for the Carbon Monoxide and
By-Product Hydrogen Subcategory.  The basis for this determination  is
that  no  toxic  pollutants  were  found  at significant levels in the
process related wastewater during the screening  of  one  plant.   The
subcategory is excluded under Paragraph 8 of the Settlement Agreement.

Assessment of the Water Pollution Potential

Production Processes and Effluents

Carbon  monoxide  is produced as a result of production of hydrogen by
refining natural gas.  It is also recovered from several  gas  sources
including  partial  combustion  of  oil or natural gas, coke oven gas,
blast furnace gas, water gas, and methane reformer gas.

The major use of carbon monoxide is for the manufacture  of  methanol.
It  is also used in the production of ammonia,  acetic acid, zinc white
pigments,  and for  reducing  oxides  for  special  steels  and  nickel
refining.

The industry profile data is given in Table 20-1.

Toxic  pollutants  detected  in  the raw waste during screening at one
plant were:

                                   Concentration (
          Pollutant

          Chromium
          Zinc
          Silver
          Mercury
                                          2590
                                           820
                                             1 .4
                                             1 .2
The only pollutants of significance in terms of waste loads are chrome
and zinc.  However, those result from the additives  used  in  cooling
water  to  inhibit corrosion, and are not process related.  Control of
zinc and chromium from this type of source is best  achieved  by  best
management   practices  on  a  case-by-case  basis  by  the  permit ing
authority,

Status gf_ Regulations

Subpart AG has been reserved for this subcategory.
                              629

-------
TAHtE 20-1
SUBCATEGOBY PRCFILE DATA
SUBCATEQORY
CARBON MONOXIDE AND B^-PRODUCT HYDROGEN
Total subcategory capacity rate
Total subcategory production rate
Number of plants in this subcategory
308 Data on file for
    With total capacity of
    With total production of
    Representing capacity
    Representing production
    Plant production range:
            Minimum.
            Maximum
    Average production
    Median production
    Average capacity *yt* 1 i
    Plant age range:
            Mininun
            Maximum
   Waste water flow range:
            Minimum
            Maximum
    Volume per unit product:
            Minimum
            Maximum
                           277,200 kkg/year
                                5
                                5

                           112,400 kkg/year

                               40 percent

                               47 kkg/year
                            63,000 kkg/year
                               NA
                               MA
                               NA

                                8 years
                               19 years

                               NA
                               NA

                               NA
                               NA


Sources of data are Stanford Research Institute, Directory of Chemical
Producers, U.S.A., 1977, U.S. Department of Oomnerce, Current Industrial
Reports/ December 1977;"Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assesanent of Effluent Limitations in the
Inorganic Chemical Industry, " June,  1978 and "Economic Analysis of Proposed
Revised Effluent Guidelines and Standards  for the Inorganic Chenicals
Industry," March, 1980.
NA = Not Available
                                   630

-------
                              SECTION 21

                       COPPER SULFATE INDUSTRY
Industrial Profile

General Description

Most of the copper sulfate produced is sold in  the  merchant  market,
consequently  captive  use  is very small.  Copper sulfate is produced
either as a  liquid  solution  or  dried  crystals.   It  is  used  in
agriculture  as  a  pesticide,  and as an additive to copper-deficient
soils.  It is also used in electroplating and petroleum refining,  and
as  a  preservative for wood.   Of the 16 plants in this industry, four
plants produce copper sulfate in significant  quantities  and  account
for  70 percent of the total U.S. production.  Two of these facilities
account for over 50 percent.

The industrial profile data for this subcategory are  given  in  Table
21-1.   The  status  of  regulations  prior  to  promulgation  of this
regulation is summarized in Table 21-2.

Subcategorization

Primary  subcategorization  originally  chosen   for   the   inorganic
chemicals  manufacturing  category  was based on the dominant chemical
produced.  Other factors that were  considered  for  subcategorization
were:   raw materials, manufacturing processes, size and age of plants
and  equipment,  geographical  location,   water   pollution   control
technology  and  solid waste handling.  A detailed discussion of these
factors is given in Section 4 of this document.

In the original study in 1974, the Agency had  further  subcategorized
the  copper  sulfate process by raw material, promulgating regulations
for pure copper and one for copper slag  and  copper  refinery  waste.
The  conclusion  made  in  this  study  was  that there was no need to
subcategorize the copper sulfate industry beyond the dominant product.
This is because pure raw materials make complete recycle possible, and
using them will allow a plant  to  comply  with  effluent  limitations
without  operation of a treatment system, if the production process is
properly operated and maintained.  Both types of raw material will  be
adequately covered under one regulation.

General Process Description and Raw Materials

Copper  sulfate is produced by reacting copper with sulfuric acid, air
and water.  The general reaction is:
Cu
1/2
                    = CuS04 + H20
(1)
                              631

-------
               TABLE 21-1.  SUBCATEGOFY PROFILE DATA SUMMARY
SUBCATEGOKf
COPPER SULFAIE
Total subcategory capacity rate
•total subcategory production rate
Number of plants in this subcategory
308 Data on file for
    With total capacity of
    With total production of
    Representing capacity
    Representing production
    Plant production ranges
            Minimum
            Maximum
    Average production
    Median production
    Average capacity utilization
    Plant age range:
            Minimum
            Maximum
    Vfeiste water flow range:
            Minimum
            Maximum
    Volume per unit product:
            Minimum
            Maximum
                           Indeterminate
                           27,300  kkg/year
                                16
                                10
                           33,850  kkg/year
                           21,420  kkg/year

                                78  percent

                                45  kkg/year
                             9,100  kkg/year
                            2,100  kkg/year
                              790  kkg/year
                               63  percent

                                 3  years
                                52  years

                                 0  cubic meters/day
                                45  cubic meters/day

                                 0  cubic  meter/kkg
                                23  cubic  meter/kkg
Sources of data are Stanford Research Institute, Directory of Chemical
Producers, U.S.A., 1977, U.S. Department of Commerce, Current Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry,"June, 1978 and "Economic Analysis of Proposed
Revised Effluent Guidelines and Standards for the Inorganic Chemicals Industry/"
March, 1980.
                                  632

-------
      TABLE 21-2. STATUS OP REGOtATICNS  -  EFTTZ3ENT LIMITATION GUIDELINES
SDBCATEGORy

SUHPART
      Copper Sulf ate

      AJ  (40 CFR 415.360, 5/22/75)
                                         STANDAHDS
                            BFCTCA
                                  BfiTER
                                          NSPS
Product
Process
Para-
meters
Max.*1)    Avg. (2)   Max.    Avg.
kg/kkg    kg,fckg    kg/Wcg  kg/kkg
 (mg/1)     ftng/1)     (mg/1)  (mg/1?
                                                            Max.
                                                         Avg,
 Pure Paw
 Materials   Cu
 Process
 Recovery
 Process
TSS


Cu


Ni


Se
          0.0006    0.0002
          0.069     0.023
                       0.003
          0.001
                       0.006      0.002
                       0.0015    0.0005
 (1)- Max. = Maximum of any one day.

 (2)- Avg. = Maximum average of daily values for thirty consecutive days.
                                    633

-------
Various forms of copper feed material are used, from  pure  copper  to
copper  slag.   The  purity of raw materials significantly effects the
quality and quantity of raw waste generated.  One plant does not start
with copper metal but uses a waste stream from a copper refinery which
consists of copper, sulfuric acid, and a small amount of nickel.   The
solution  needs • to be strengthened by the addition of more copper but
the same general equation applies.

Copper metal  and/or  copper  refinery  waste  stream,  steam,  water,
sulfuric  acid  and  air  are  treated  in  oxidizer tanks at 100°C to
produce a solution of copper  sulfate.   This  solution  is  partially
concentrated by evaporation.

If pure copper is used as a raw material, the resulting copper sulfate
solution  is  pure  enough  to be either sold, or fed to crystallizers
producing copper sulfate crystals.  If impure copper feed,  or  copper
refinery  waste  is  used, the concentrated copper sulfate solution is
filtered to remove other metal impurities.  This purified solution can
be sold as is or fed to the crystallizer.  Copper sulfate crystals are
recovered by centrifugation, dried at  approximately  110°C,  screened
and  then  packed  dry for sale.  The mother liquor is recycled to the
evaporator  or  crystallizer  with  some  being  purged   to   prevent
impurities buildup.  The purges are usually sold for metal recovery.

Figure  21-1  shows a general process flow diagram for the manufacture
of copper sulfate.

Water Use and Waste Source Characteristics

Water Use

Water is used in direct contact with copper sulfate production as  the
reaction  medium.   A  portion  of  it is evaporated to the atmosphere
during crystallization, while the remainder becomes part  of  the  dry
product  as  its  water  of  crystallization   (hydration).  Noncontact
cooling water, including steam condensate, constitutes the major water
use.  This is used to cool the reactor and  crystallizers.   Water  is
also used for pump seals and washdowns.  Table 21-3 gives a summary of
plant   water   usages  found  in  this  study  for  facilities  where
information  was  available  from  308-Questionnaire   responses   and
previous documents.

Waste Sources

A.   Noncontact Cooling Water

     Noncontant cooling water is used to cool  the  crystallizers  and
     constitutes one of the main wastes.  This waste stream should not
     be contaminated by process leaks, and therefore can be discharged
     without treatment.
                             634

-------
                                              WATER
       F1ECTROLYTE
       FRCM COPPER.
       REFINER OR~
       SHOT COPPER
      SULFURIC.
        ACID
REACTOR
FILTER
                    I
U)
                                           WASH  FILTER
                                           WATER  CAKE
EVAPORATOR
CRYSTAL-
 LIZER
                                                             1
                                                                                   CENTRIFUGE
                                                                                DRYER
                                                                                       I
                                                                                     MOTHER
                                                                                     LIQUOR
                                                                                     RT.FFD
                                                                             CuS04.5H2O
                        Figure  21-1.
              General process flow diagram of the manufacture of copper sulfate.

-------
TABLE 21-3.   WATER USAGE IN COPPER SULFATE SUBCATEGOKY

Water Usage at Plants (m3/kkg)
Source #034 #284 #313 (1) #069

#571

Process * 1.21(2) 24.8 3.30
Contact
Noncontact 19 . 6 0 37 . 3 105
Cooling
0.075

0

Maintenance       1.25
Cleaning and
Washdcwn, Pumps
Seals and Leaks

Steam            38.6

Air Pollution     0
Control
                             0.35
0.28
                             0

                             0.52
3.77
0.017
(1)  Includes uses for other processes

(2)  Maxiumum - includes groundwater infiltration

*    Utilizes feed solution from another industry
                                     636

-------
B.   Washclowns, Leaks, and Spills

     Washdown, pump seal leaks, and  spills  are  sources  of  contact
     wastewater.    These  flows,  however,  are  relatively  small and
     intermittent,  and  do  not  represent  a  major  waste   source.
     Wastewaters  emanating  from this source are either combined with
     the mother liquor, or treated and discharged.

C.   Mother Liquor Purges

     A small portion of the mother liquor is purged periodically  from
     the  process  to prevent buildup of metal impurities.  The amount
     of purge is variable and depends  on  the  purity  of  feedstock.
     These   purges   are   processed   to  separate  metallic  salts,
     particularly those of copper and  nickel,  from  the  impurities.
     These recovered metallic salts are used for other processes while
     the impurities are disposed of at an approved landfill.

D.   Steam Condensate

     A few  plants  use  evaporators  to  concentrate  the  production
     solution.    Steam   condensate   is   an  additional  noncontact
     wastewater formed in the process.  This can  also  be  discharged
     without treatment.

E.   Sludge

     Solid  waste  is  generated  in  product  purification   by   the
     filtration  step.   This  is  necessary only for plants utilizing
     impure copper, or copper refinery waste, as raw material.   These
     filter sludges contain metallic impurities or copper sulfides and
     need disposal at an approved landfill.

     Plants  that produce copper sulfate in liquid form have no direct
     contact waste streams from the process.   Plants  utilizing  pure
     copper feedstock are able to recycle most contact wastewaters and
     generally  have  no  discharge  of  contact  wastes.   Table 21-4
     summarizes the quantities of wastewater that go to the  treatment
     facility,  their  sources,  and the handling practices for plants
     which do not discharge wastewaters.   The  data  was  taken  from
     308-Questionnaire  responses, previous development documents, and
     industry visits.

Description of Plants Visited and Sampled

Screening

Plant #034 was visited and process  wastewater  and  effluent  samples
were  collected  and  analyzed  for conventional and toxic pollutants.
The process used at this plant is similar to that  described  earlier,
for  one which utilizes a waste stream from a copper refining facility
as its feedstock.  The feedstock is strengthened by  the  addition  of
copper  shot.   The  filter  cake and wash water are sent to a settler
                               637

-------
TABLE 21-4.  WASTE WATER FLOW FOR THE COPPER SULFATE SUBCATEQORY
Plant
 Avg. Waste
 Water Flow to
 Treatment
 (m3/Wcg of CuSO4)
                                                 Waste Water
                                                  Handling
                                                  Practice
#034


#284




#313



#069




#571


#885


#458


#100


#969


#050
 0.94


 0.52




23.4*



 4.01




 0


 0


 0


 0


 0


 0
Segregated treatment of
CuSOj waste  (lime treatment)
Waste streams and treatment
are combined with other
mining, milling and man-
ufacturing process wastes.

Waste streams and treatment
are combined with other
metal process wastes.

Waste streams are combined
with waste from other re-
agent grade processes and
discharged to sewer.

No discharge of waste from the
process  (recycle)
No discharge of waste from the
process  (recycle)

No discharge of waste from the
process  (recycle)

No discharge of waste from the
process  (recycle)

No discharge of waste from the
process  (recycle)

No discharge of waste from the
process  (recycle)
* Flow is for the combined waste from all process per kkg of CuSOfl.
  Actual amount of flow contributed by CuSO4 process is unavailable.
                                     638

-------
where the cake and wash water are finally separated.   The  decant  of
the  settler is recycled back to the reactor, while the settled sludge
is sent to another process for melting.  Mother liquor purges from the
centrifuge are also  sent  to  other  processes.   Leaks,  spills  and
washdown  water  flow  down  to a sump in the basement of the facility
where it collects with contaminated groundwater, and is then pumped to
holding tanks.   About  one  quarter  of  this  wastewater  volume  is
comprised  of  contaminated groundwater from the immediate area.  From
the holding tanks, the waste goes to the treatment facility  where  it
is treated with lime, filtered and discharged to a collection tank.

The   uncontaminated   steam   condensate  from  the  evaporator,  and
noncontact cooling water from the crystallizer, are combined with  the
effluent  from  the lime treatment in a collection tank.  The combined
stream passes through a  cloth  filter  for  final  polishing  and  is
discharged  to  a  sewer.  The filter residue from the filter press is
hauled to an approved landfill site.  Figure 21-2  shows  the  general
process  and  treatment flow diagram with the location of the sampling
points.  Table 21-5 presents flow data, total suspended solids   (TSS),
and  copper and nickel emissions for the various waste streams sampled
during screening.

Verification

Plant #034 was sampled again during the verification phase.  Prior  to
this,  the  system  was  changed  so  that  only the efluent from lime
treatment goes to the collection tank and through  the  cloth  filter.
This  effluent  then combines with the steam condensate and noncontact
cooling water waste streams after the cloth filter and  discharges  to
the se'wer.

Figure  21-2  also  shows  this  change, and the subsequent new sample
points for verification phase sampling.  Table 21-5  also  gives  flow
and   discharge   data   for  various  waste  streams  sampled  during
verification.

Plant  #034  was  the  only  plant  sampled  for  the  copper  sulfate
subcategory.   During the program, an attempt was made to locate other
candidates  for  sampling.   A  search   was   conducted   using   the
308-Questionnaires, published materials and the telephone.  Out of the
17  other  facilities,  11  have  no  discharge of process wastewaters
(practice recycle); four plants  were  large  multi-product  complexes
with  combined  waste  treatment  systems  where segregation of copper
sulfate process wastes was impossible; and two  plants  produced  only
reagent grade product, and are therefore low volume producers.

Toxic Pollutant Concentrations

The following toxic pollutants were found at detectable concentrations
in the raw waste samples at copper sulfate Plant #034 during screening
and verification sampling.
                                 639

-------
   aor
   Cu
    HlgMC
     JCID
cu tanaaei
, *
g.T
^ OUD1
q -
J,
















ZEB
TCB


L* 	












ID a
»1 w

















IAX
croeot
H-lflc
MSH



|
SEfflJ











ID
point*
nj «4
•Uan
MIOI



1
9












pllnn
w«>Ii


















ng



















1












4
1




"






Nan





)|3">

am*





4




i
MB







CON







M
U
4

1
IV







not



OKL1



nm
«" <
carnm

UDK

CM



cuso

SrtE



"




nxz



E9t



,-SBjO
















1







anua.

f
*
'
wunu.
nuc
12)


won ra-
nt

»

^ OUWm - OKLK
T aH. STI * OH
! i
J

^*\^
^
^
•
DISC
*
SB
UMS
laiiv

A
*»
•
I2OI
S


IQI
ESS

»«

rtoH
K
-
m


i ,
*'
•
UkHX
3
*a


LU
















Figure 21-2.  General process flow diagram at plant #034 showirig the sarrpling points.
               (Copper sulfate manufacture.)

-------
TABLE 21-5.  FLOW AND POLLUTANT CONCENTRATION DATA OF THE SAMPLED
             STREAMS FOR PLANT #034 PRODUCING COPPER SULFATE

Stream
No.
1
2
Sampled Unit Flow
Stream ( m3/kkg of Cu9C
Description
Screening (1)
CuS04 waste * 1.25
Effluent from 1.25
TSS
ty (all
0.087
0.078
Cu
in kg/kkg of
4.2
0.010
Ni
CuSO4)
0.25
0.00053
              lime treatment

              Steam Condensate 0.209      0.00021

                       Verification (2)

              CuS04 waste *    1.25       1.8

              Effluent from
0.00016
5.0
0.000025
0.20

3



lime treatment 1.25 0.030
Noncontact 14.2 0.11
Cooling Water
and Steam
Condensate
0.0042 0.00038
0.024 0.0020



  (1)   From grab samples composited during the period of batch manufacturing
       and treatment process.

  (2)   Average of three daily grab samples composited during the period of
       batch manufacturing and treatment process.

   *  Infiltration of ground water into the collection sump was suspected at
      the tine of sarrpling.
                                   641

-------
               Maximum Raw Waste Concentration Observed
                                Ug/1)
Pollutant                Screening

Antimony                       330
Arsenic                      3,500
Cadmium                        870
Chromium                       140
Copper                   1,850,000
Lead                           180
Nickel                     112,000
Zinc                        11,000
1,1,1-trichloroethane          240

NA = Not analyzed
                                       Verification

                                            1,300
                                          127,000
                                            2,500
                                              940
                                       3,940,000
                                            2,200
                                          136,000
                                           17,000
                                               NA
A  large  portion  of  the  raw  wastewater  at this plant consists of
groundwater which seeps and collects in the basement, along with leaks
and washdown water from the process.  The groundwater is  contaminated
from  the  surrounding  area  which  is  heavily  industrialized.  The
trichloroethane is presumed to be external contamination because  this
chemical is not used in the process.

No   other   organic   toxic  pollutants  were  found  at  significant
concentrations during screening sampling.   Consequently,  no  organic
toxic pollutants were analyzed for in the verification phase.

Section  5  of  this report describes the methodology of the screening
and verification sampling program.  In the copper sulfate industry,  a
total  of  six  days  of  sampling were conducted at Plant t034.  Five
different sampling points  were  involved  covering  the  various  raw
wastes,  and  the  intermediate  and  treated  effluent  streams.  The
evaluation of toxic metal  content  of  these  process  related  waste
streams  was  based  on 221 analytical data points.  The screening for
toxic organic pollutants at Plant 1034  generated  an  additional  456
analytical  data  points.   The unit loadings were calculated from the
waste stream flow rates measured or estimated at the time of sampling,
the measured pollutant concentration, and the reported copper  sulfate
production rate.
     That is.

     Unit loading  (as kg of pollutant per
     kkg of copper sulfate)
                                           (C)  (Q)
                                          1000 (P)
     where:
          is the concentration of the pollutant expressed  in units
          mg/1  (Note:  kg/m3 = mg/1),
                                                          of
     Q
is the waste stream flow rate expressed in units of  mVday.
(m3,  a cubic meter, is equal to 264.2 U.S. gallons),  and
                                 642

-------
     P    is the copper sulfate production rate expressed in units  of
          kkg/day  (kkg  is  1000  kg, a metric ton, which is equal to
          2205 Ibs).

The average values are  based  on  data  from  Plant  #034  where  the
particular  pollutant  was  found  at  concentrations greater than the
analytical detection limits and in significant concentrations since it
could be treated by an available treatment technology.

In Table 21-6, the toxic pollutant raw waste data are presented as the
average daily concentrations and the unit loading  found  during  each
sampling  at  Plant #034.  The overall averages are also shown.  It is
this overall average which is used as the average raw waste load  from
the copper sulfate process in various calculations.

Based  on the total annual production rate of this subcategory and the
average waste load generated per unit  product,  the  estimated  total
pollutant raw waste loads generated each year for this subcategory are
as follows:

                                        Waste Load (kg/year)
                                               1
     26
    400
     74
     15
124,000
     30
  6,200
    700
               Pollutant

               Antimony
               Arsenic
               Cadmium
               Chromium
               Copper
               Lead
               Nickel
               Zinc

Pollution Abatement Options

Toxic Pollutants of Concern

The  principal  pollutant  of  concern  is  copper.   The  other toxic
pollutants found in plant  wastewaters  are  closely  related  to  the
purity  of  the  copper  and  acid sources.  The heavy metals cadmium,
nickel, zinc and, to a lesser extent,  antimony,  chromium  and  lead,
which  were  found  during  field  sampling,  may  originate  as trace
impurities in copper scrap and other impure  copper  sources.   Plants
utilizing  pure  copper  shot  would not experience a buildup of these
impurities in the mother liquor, and consequently would not generate a
waste stream containing these impurities.  Arsenic was also  found  in
fairly  high  concentrations in the raw wastewater.  A possible source
of arsenic, and other copper ore trace metals, is the use of  sulfuric
acid  made  from  sulfur  dioxide  produced  in the roasting of copper
sulfide ore.  1,1rl-trichloroethane was found and several other  trace
organic  toxic  pollutants  were  found in the raw waste at Plant #034
which contains infiltrated groundwater.  The general area around Plant
#034 is heavily industrialized.  The local groundwater is known to  be
contaminated with various organic compounds.  Since there are no known
organic  compounds  used  in  the feedstock, or copper sulfate process
                                  643

-------
                        TABLE 21-6.  RAW WASTE OAIA
Subcategory:  Copper Sulfate
Average Daily Pollutant Concentrations and.loadinga found during Sampling of
                               Plant #034 U)

                                         (mcrA)
(kg/Meg of CuSC^.SH^)
Pollutant
TOXK
Antimony, Sb
Arsenic, As
Cadmium, Cd
Copper, Cu
Lead, Rs
Nickel, Ni
Zinc, Zn
Chromium, Cr
Selenium, Se
CCHVEHTIOHRL
TSS
(1) The meth
Screening {2)
0.31
0.00069
3.5
0.0078
0.87
0.0019
1900
4.2
0.18
0.00039
110
0.25
11.0
0.024
0.14
0.000030
< 0.011
< 0.000024
39.0
0.087
odology of the sanf
Verification
0.54
0.0012
44.0
0.097
1.6
0.0035
2200
5.0
0.78
0.0018
91.0
0.20
L2.0
0.027
0.36
0.000080
< 0.0050
< 0.000011
790
1.80
jling program is
t3) Overall Average (4)
0.44
0.00095
24.0
0.052
1.2
0.0027
2000
4.S
0.48
0.0011
102
0.23
12.0
0.026
0.25
0.00055
< 0.008
< 0.000018
410
0.92
described in Section 5.1,;
       and Section 21.1.2 presents the scope of sampling in the Copper
       Sulfate industry.
   (2)  Screening data from one 72-hour grab composite sample of individual or
       combined raw waste streams.

   (3)  Verification data from three 24-hour grab composite samples, averaged.
   (4)  taen averaging values indicated as "less than" (<), the absolute value
       was used and the resulting average was indicated as a "less than" value.
                       644

-------
itself, the organic toxic pollutants found at Plant #034 are  atypical
and  are  related  to  the contaminated groundwater.  Selenium was not
detected in the  raw  waste  at  Plant  #034.   However,  the  average
concentration of selenium was 0.1 mg/1 in the treated effluent for all
sampling trips.  This phenomenon was also observed  in previous studies
at  this  plant.   The  increase  in  selenium occurs in the treatment
operation and the source is presently unknown.  It  is  apparent  that
copper,  arsenic,  cadmium,  chromium, lead, antimony, nickel and zinc
are typical pollutants encountered in copper sulfate wastewaters,  and
that selenium appears only in the effluent sfter lime treatment.

Process Modifications and Technology Transfer Options

Mechanical  scrapers  could  be  installed  on filters in plants using
impure raw materials.  This would eliminate the need  for  backwashing
and the wastewater from this source would b§ eliminated.  Installation
of these scrapers would constitute a small capital cost.

Best Management Practices

The  best  technology  available  for  the treatment of copper sulfate
waste, where pure copper is used as the raw material, is total recycle
of process waste.  This would require floor dikes, plumbing and  sumps
to segregate the wastes, and pumps and piping for recycle.

The  best  technology  for  waste  treatment  where  copper sulfate is
prepared from impure copper sources or copper refinery  by-product  is
collection of waste mother liquor and process spills,  washdowns, etc.,
followed by lime precipitation of metals,  settling of suspended solids
and  filtration.   This  would  require  installing  dikes,   sewers, a
treatment tank, a settling tank, filter presses, and associated piping
and pumps (2).

Prevailing Control and Treatment Practices

Plant #034 collects leaks, spills and washdown  water  in  a  basement
sump  and  pumps  it to holding tanks having a combined volume of 6000
gallons.   The  batch  is  treated  using  lime   neutralization   and
precipitation  and is filtered by a filter press.   The filtrate, after
mixing with other streams, is polished further by  passing  through  a
cloth filter and is finally discharged to a .sewer.  The filter cake is
hauled to a landfill.

Plant  #284  sends  mother  liquor  purges and filter sludges to other
processes.  Wastewaters from maintenance an<3 dust control are combined
with  a  multitude  of  other  process  wastes  and  treated  by  lime
neutralization   with   aeration,  followed  by  clarification  before
discharge.

Plant #069,  which produces a reagent  grade  product,   sends  periodic
purges  and  washdown water to a combined collection system with waste
water  from   various   other   products.     Treatment   consists   of
neutralization and equalization of the wastes and discharge to a POTW.
                              64!

-------
Plant   #313   also  combines  its  wastewaters  from  copper  sulfate
production  with  wastes  from  various  other  metal  processes   and
presently discharges the combined waste, after settling, to a pond.  A
treatment system is being designed which uses lime precipitation at pH
10  followed  by  gravity separation and centrifugation to thicken the
sludge.  The  waste  will  then  be  neutralized  to  pH  6.5-7.5  and
discharged.
                                 , #885, and #571 have no discharge of
                                 o nrnroesc; _
Plants  #100,  #969,  #050,  #458, ffooD, m
wastewater from the copper sulfate process.

Advanced Treatment Technologies

Copper, nickel, cadmium and zinc can be  separated  from  solution  by
alkaline   precipitation  at  pH  values  from  7.2  (copper)  to  9.7
(cadmium).  Alternatively, sulfide precipitation can be  used.   These
metals  can  also be removed from clarified solutions by ion exchange,
but the metal  ions  remaining  on  the  exchange  resins  or  in  the
regenerant solutions may create additional disposal problems.  Removal
of  trace  metal  concentrations  by  the  xanthate  process, although
possible, has  not  been  used  widely.   Some  reduction  of  arsenic
concentrations  at  high  pH  levels  has  been reported, although the
removal mechanism is not clear.  More effective arsenic removal  would
require  the  addition  of  ferric chloride during alkaline or sulfide
precipitation of the process wastes.

Selection of Appropriate Technology and Equipment

Technologies for Different Treatment Levels

A.   Level 1 (BPT, BAT, NSPS)

     Alkaline precipitation using caustic soda in a batch process  was
     considered  as the most effective technology for removal of heavy
     metals and arsenic.  The Agency selected Level  1  technology  as
     the  basis for BPT because it represents the prevailing treatment
     practice in this  industry.   All  direct  dischargers  have  BPT
     installed.    To   accommodate  a  40-hour,  five-day  production
     schedule, the wastes are  received  in  daily  batches,  and  are
     raised to pH 10, mixed, and settled.  At the end of the workweek,
     the  batch  is filtered and the pH adjusted to a range of 6 to 9.
     Figure  21-3  shows  the  schematic  flow  diagram  for  Level   1
     treatment.
B.   Level 2

     In the Level 2 treatment, the Level
     the  addition of ferrous sulfide  in
     alkaline precipitation, to  increase
     metals.   Metallic  hydroxides  are
     the bottom of the reaction  vessel.
     in the reactor with residual  metals
     sulfide  precipitation,  filter aid
                                          1 system is  supplemented  by
                                          the reaction vessel following
                                          the  precipitation  of  trace
                                          allowed to form and settle in
                                          Then ferrous sulfide  is mixed
                                             Following  completion  of
                                          is added while the mixture is
                                    646

-------
     being filtered through a filter press.  As in Level 1, the pH  is
     adjusted  and  the filter effluent is discharged until the weekly
     batch is exhausted.

     This technology was not recommended,   however,  because  Level  1
     afforded  adequate  control  and the additional reduction was not
     sufficient to offset the additional cost.  Figure 21-4 shows  the
     schematic   flow   diagram   for   Level  2  treatment.   Further
     information on estimated treatment system  operation,  costs  and
     effectiveness,  may be found in the proposed Development Document
     (60).

Equipment for Different Treatment Levels

A.   Equipment Functions

     At both levels the models are designed for batch operation.   Each
     day's wastes are transferred from holding  sumps  to  a  reaction
     vessel  for  storage.  At the end of a workweek the BPT treatment
     of the accumulated waste consists of raising the pH  to  10  with
     caustic  soda, mixing, and applying filter aid while filtering in
     a filter press.  After pH adjustment to the 6  to  9  range,  the
     filter  effluent  is  discharged.   In  the  Level  2  model  the
     equipment remains the same but precipitation is  accomplished  in
     two steps.

B.   Chemicals Used and Handling Precautions

     Caustic soda solution is added manually to each batch  until  the
     proper  pH  level  is  reached.   In  Level 2, batches of ferrous
     sulfide  are  prepared  by  mixing  ferrous  sulfate  and  sodium
     bisulfide in a well-ventilated area.   Inert filter aid is applied
     as  a  filter  precoat  and  is  added  continuously  during  the
     filtering process.  With normal precautions there are no  special
     chemical  handling  problems  in  the treatment of copper sulfate
     wastes.

C.   Separation and Removal of Solids

     All solids in both levels are collected as  filter  cake  in  the
     filter  press.   At  both  levels  'the  dewatered cake containing
     metallic hydroxides, metallic sulfides, and spent filter  aid  is
     hauled to an off-site chemical landfill.

D.   Monitoring Requirements

     Alkaline precipitation of the heavy metals is assured by bringing
     the reaction vessel contents to the proper pH, as  determined  by
     the  operator,  using  field  pH  equipment.   Periodic  specific
     analyses of the final effluent for toxic pollutants can  be  made
     by atomic absorption methods through a commercial laboratory.
                              647

-------
so*




WMVS1E







KUttO
INK
tKIER















i







mcrfOH
u«


V / -
f
1
<



_~*_ Jim — '
T I* MUUBBOir
.
I

^
I
	 1_ .,_»ru..,imi)air^
CO
                                                                   uwrm.
                        Include* Oow nxinllorlnf, pH manUorlng and umpler
                             Figure 21-3.  Level 1 waste water treatment for copper sulfate  subcategory
                                            batch process.

-------
Treatment Cost Estimates

General Discussion

To  prepare  treatment  cost  estimates,  a  model  plant  concept was
developed for Level 1  technology.  The BPT  model  treatment  consists
of:

     Collection  of wastewaters in a batch according to the production
     mode.

     Hydroxide treatment to precipitate metals, followed  by  settling
     and filtration.

     pH adjustment before discharge.

A.   Production

     Copper sulfate production ranges from 45 kkg/yr to 9100 kkg/yr in
     ten plants for  which  308-Questionnaires  were  available.   The
     average  of  the  ten  plants  is  2100  kkg/yr  and  the  median
     production is 790 kkg/yr.  The operational  mode  for  all  these
     plants is assumed to be batch and to run 250 days per year.

     For wastewater treatment cost estimates, one production level was
     chosen  as  the  model  plant.  This is the average production of
     2100 kkg/year.  One production level is  sufficient  because  the
     wastewaters will  be collected in batches and treated as necessary
     when  the  batch  tanks are full.  The amount of wastewater to be
     treated at any one time is then  independent  of  the  'production
     rate, although it will determine the frequency of treatment.  All
     known  plants  with  production  rates below the model plant rate
     have no discharge of wastewaters from the copper sulfate process.

B.   Wastewater Flow

     The data on Table 21-4 for plants  with  a  wastewater  discharge
     shows  a  unit  flow  range  from 0.52 mVkkg of CuS04 to over 23
     mVkkg of CuSO^.   One plant flow is for reagent grade  CuS04  and
     so cannot be considered a normal waste flow.  Only Plant #034 has
     separate  treatment  for  CuS04  wastewater,  and the flow is the
     median  of  those  normal  processes   sending   wastewater   for
     treatment.   The wastewater unit flow used for the model plant is
     0.94 mVkkg of CuS04.  All the  other  plants  except  #034  have
     either  no  discharge  of  wastewater,  combine their wastes with
     other process wastes or produce reagent grade products.

C.   Solid Waste Generation

     Copper hydroxide from filtration is the  only  solid  waste  that
     required  disposal.  This waste must be disposed of in a chemical
     landfill since the  solids  may  contain  other  contaminants  or
     become   oxidized   and   begin  to  migrate  into  the  soil  or
                               649

-------
                     FIUBRUD —
                                             Q
now monitoring. pH monitoring Bad u
  Figure 21-4.  Level  2 waste water  treatment for copper sulfate subcategory -
                 batch  process.

-------
     groundwater.  Slimes from the mother liquor  and  copper  sulfate
     solid  wastes  are  all  recycled or sent to another facility for
     metal  recovery.   There  is  no  solid  waste  generation   from
     processes using pure copper raw material.

     Based  on  sludge  production of 5 Ibs/day for 250 days/yr in the
     model plant, the annual solids production is 558  kg,  equivalent
     to unit solids generation of 0.27 kg/kkg of product.

D.   Treatment Chemicals

     Caustic soda  is  required  to  precipitate  metals  and  for  pH
     adjustment.  For the model plant, the assumed caustic soda dosage
     was 0.33 kg/kkg of copper sulfate.

Model Plant Cost Estimates

The  cost  estimate  of  the model plant having one level of treatment
(BPT) and one level of production is presented in Table  21-7.   Table
21-.8   gives   a   summary  of  the  unit  cost  distribution  between
amortization and operation and maintenance cost components for Level 1
treatment.

Cost estimates developed for the first level of treatment  (BPT,  BAT,
NSPS)  indicate that amortization and labor constitute a major portion
of the annual costs.

Basis for Regulations

Evaluation of BPT Treatment Practices

Copper sulfate can be  manufactured  using  pure  copper  as  the  raw
material or an impure copper'raw material.  Waste loads emanating from
the two sources differ greatly in that total recycle of process wastes
can  be  accomplished  at  plants using a pure copper source, while at
plants using an impure raw material, waste streams need to be  removed
to some extent to avoid buildup of contaminants in the process.

Based  on  the  process  technology of total recycle at plants in this
study using pure raw material, the industry  practices  indicate  that
the  degree  of  waste control attainable is zero discharge of process
wastes.

A.   Pollutant Removal with BPT Treatment

     BPT technology for copper sulfate  plants  utilizing  impure  raw
     materials  is  equivalent  to  Treatment  Level  1.   Table  21-9
     presents a summary of  long-term  effluent  monitoring  data  for
     Plant 1034 -on total suspended solids (TSS), copper, nickel, zinc,
     arsenic   and   selenium.    Means,    standard   deviations,  and
     variability  factors  are  given  where   sufficient   data   are
     available.   These performance characteristics are later utilized
     for the development of the regulations.
                                651

-------
     Table 21-10 presents the toxic and  conventional  pollutant  data
     for  effluent  from  the  two samplings at Plant #034 in the same
     manner as Table 21-6 did for raw waste data.   The ability of  BPT
     treatment   to  remove  toxic  pollutants  can  be  estimated  by
     comparing the overall averages from Table 21-6 and  21-10.    This
     comparison  is  presented in Table 21-11 which also expresses the
     removal efficiency as  the  calculated  average  percent  removal
     observed at this plant.

     Table  21-11  shows  that the treatment efficiency for removal of
     copper, nickel, arsenic, cadmium and zinc is above 99.5  percent,
     while removal for antimony is just slightly over 80 percent.  The
     toxic  pollution  concentrations were at or below the lower limit
     of concentration achieved  by  alkaline  precipitation  with  the
     exception  of  copper  and  nickel.  These toxic metal pollutants
     comprised the majority of the treatment  loading  which  suggests
     that  the  optimum  conditions for metal hydroxide formation were
     not being attained at the time  of  sampling.    The  thirteenfold
     increase  in  selenium  concentrations  in  the  treated effluent
     should be  noted.   This  phenomenon  was  observed  in  previous
     studies  at  this  and other plants.  The observed concentrations
     appear to remain  at  the  lowest  achievable  concentration  for
     alkaline  precipitation.  The source is presently unknown,  but it
     is suggested that the selenium may be introduced in the treatment
     chemicals.

     Treatment system  performance  data  was  unavailable  for  other
     facilities  generating  a  waste  discharge  because they combine
     their wastes with other process wastes for treatment.

Basis for BPT Effluent Limitations

The  BPT  regulations  for  the  Copper   Sulfate   Subcategory   were
promulgated  in  40 CFR 415.363.and are presently in effect (see Table
21-2),    The  technology  basis  for  the  existing  BPT  is  alkaline
precipitation   plus   filtration   and  final  pH  adjustment  before
discharge.  Of the 16 plants in this subcategory, 10 are  reported  to
have no discharge, five are direct dischargers, and one is an indirect
discharger.  All direct dischargers have BPT technology installed.

In  the existing BPT regulations, the Agency has different limitations
for pure and impure raw materials processes.  The Agency has eliminatd
this distinction for BPT and is not establishing different limits  for
these  processes  in  the  BAT,  BCT, NSPS, PSES and PSNS regulations.
This  is  because  both  processes  are  adequately  covered  by   one
regulation,  since  the  pure raw material process should, with proper
operation and design, comply without end of pipe treatment.

A.   Flow Basis

     The model plant BPT treatment system is based on an  inflow  rate
     of  0.94  mVkkg.  This is derived from the average flow of Plant
     #034, and was the median of plants with  a  wastewater  discharge
                                652

-------
     TABLE  21-7.  MODEL PLANT TREATMENT COSTS
 Subcategory
 Production
               Copper  Sulfate
               2,100 metric  tons  per  year
A. INVESTMENT COST

     Site development  	
     Equipment  	
     Monitoring  equipment  ..

          Subtotal  	
     Contractor's 0 & Pb....

          Subtotal  	
     Engineering  	

          Subtotal  	
     Contingencies  	

          Subtotal	
     Land	

     TOTAL INVESTMENT COST  .


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  .
     Energy 	
     Chemicals 	
     Maintenance  	
     Taxes and insurance .. .
     Residual waste disposal
     Monitoring, analysis
      and reporting 	
                                               BPT

                                              1,800
                                             61,500
                                             20,000

                                             83,300
                                             12,495

                                             95,795
                                             19,159

                                           114,954
                                             11,495

                                           126,449
                                              1,800

                                           128,249
      TOTAL OPERATION AND
      MAINTENANCE COST
                                             8,000
                                                30
                                             1,000
                                            12,645
                                             3,847
                                               100

                                             5,000
                                            30,622
                                                      ($)
BAT'

  0
  0
  0

  0
  0

  0
  0

  0
  0

  0
  0
 C.  AMORTIZATION OF
    INVESTMENT COST

    TOTAL ANNUAL COST
                                            20,573

                                            51,196
a  Represents the incremental cost above that for BPT treatment
b  Overhead  and Profit
                              653

-------
         TABLE 21-8 MODEL  PLANT UNIT TREATMENT COSTS
     Subcategory Copper  Sulfate
    COST  ITEM
PRODUCTION
 (kkg/yr)
Annual Treatment Costs  ($/kkg)


     LEVEL OF TREATMENT

   BPT        BAT*      NSPS
Annual Operation
and Maintenance 2,100 14.58
Annual
Amortization 2,100 9.80
Total Annual
Cost 2,100 24.38

NA NA

NA NA

NA NA
•Represents  the incremental  cost above BPT.  All plants paresently
 meet BPT which is equal to BAT/NSPS in this subcategory.
                               654

-------
TABLE 21-9.  SUMMARY OF LONG TERM MONITORING DATA FROM PLANT #034
                                                                 (1)

Pollutant
Total Suspended
Solids (TSS)
Copper
Nickel
Zinc
Arsenic
Selenium
Lead
Number
of Months
16
16
16
16
16
15
16
Long Term
Averages
(mg/1) (kg/kkg)
25.0 0.093
4.4 0.016
0.36 0.0013
0.12 0.00044
0.0012 0.0000044
0.0073 0.000027
0.033 0.00012
w«
2.4
1.6
2.2
2.4
3.4
6.2
2.5
 (1)   Values are for monthly measurements of the treated effluent combined
      with nonoontract cooling water and steam oondensate discharges.

 (2)   For 30-day average measurements,  a normal  distribution is obtained
      and the variability factor is found by the expression, VF = 1.0 +Z / S \
      where X is the arithmetic  mean and S is  the arithmetic standard    \X/
      deviation.  Hhen the value of Z is 1.64, the variability factor
      is for -95 percentile which is used to set the proposed  maximum 30-day
      average effluent limitation.  Refer to Section 8.2 for detailed discussion.
                                    655

-------
     from  industrial grade CuS04 manufacturing processes.  Four other
     plants with wastewater discharges combine their waste with  other
     processes for treatment.   All other plants either produce reagent
     grade product,  or have no discharge.

B.    Selection Basis for Pollutants to be Regulated

     The  selection  of  pollutants  for  which   numerical   effluent
     limitations  were  established  was based on an evaluation of raw
     waste data from the screening and verification sampling  program.
     The  two major factors considered were:  1)  individual raw waste
     concentrations, and 2) the total subcategory raw waste loadings.

     1.    Raw Waste Pollutant Concentrations

          A tabular summary of maximum raw waste concentrations  found
          in   sampling  is  presented  above.   Data  from  screening
          sampling was used to determine  the  need  for  verification
          sampling.    The  maximum  concentrations  found  during both
          screening and verification are  shown  for  comparison.   As
          previously  discussed,  selenium  was  not  found in the raw
          waste although it was present in  the  effluent.   For  each
          pollutant,   the   maximum  concentration  observed  gave  a
          preliminary indication of its potential significance in  the
          subcategory.   On  this  basis, the preliminary selection of
          candidates for regulation  includes  copper,  nickel,  zinc,
          arsenic,  cadmium, antimony, 1-1-1-trichloroethane, and lead
          in decreasing order of their apparent  pollution  potential.
          These   pollutants   were  observed  at  least  once  during
          screening at  concentrations  considered  treatable  in  the
          industry  using  one  of  the available treatment technology
          options.  The source of trichloroethane is  presumed  to  be
          groundwater  contamination.   It is not process related, and
          was not considered for verification.  In  verification,  the
          same  metals  found during screening appeared along with the
          addition of chromium.   The  other  metals  found  exhibited
          maximum  concentrations  that  were  considerably lower than
          those treatable by available technologies.

     2.    Total Subcategory Raw Waste Pollutant Loadings

          Pollutant raw waste loading data were used to  evaluate  the
          overall   magnitude  of  the  pollution  potential  for  the
          subcategory.  Data from the plant sampled are summarized  in
          Table  21-6.   This  information, coupled with the estimated
          total copper sulfate production  rate  of  27,300  kkg/year,
          yielded the approximate total annual pollutant loading rates
          for the subcategory shown above.  This method of ranking the
          pollution  potential  of  the observed toxic metals confirms
          the dominance of the eight toxic metals and  ranks  them  as
          copper,  nickel,  arsenic, zinc, cadmium, lead, antimony and
          chromium in terms of both total mass loading  and  treatable
          raw  waste  concentrations.   The  existing  BPT regulations
                                 656

-------
                        21-10.  TRSKCED EFFLUENT Dfllft
Subcategory:  Copper Sulfate
Average Daily Pollutant Concentrations and Loadings Found During Sampling of
                               Plant

                                {Jeg/kkg of O.iS04 .  5H20)
Toxic
PoU-utanta
Antimony
Arsenic
Cadmium
Chromium
Copper
Lead
Nickel
Zinc
Selenium
Conventional
TSS
Screening (2)
0.036
0.00008
< 0.02
< 0.00004
0.001
0.000002
0.005
0.00001
4.6
0.010
0.005
0.00001
0.24
0.00053
0.016
0.000036
0.10
0.00022
Pollutant
35.0
0.078
Verification (3)
0.12
0.00027
0.057
0.00013
0.0042
0.0000089
0.017
0.000038
1.9
0.0042
< 0.031
< 0.000069
0.17
0.00038
0.02
0.000044
0.11
0.00024
13.7
0.03
(41
Overall Average
0.08
0.00018
0.038
0.000085
0.0026
0.0000054
0.011
0.000024
3.3
0.0072
< 0.018
0.00004
0.20
0.00046
0.018
0.00004
0.10
0.00023
24.0
0.054
 (1)  The effluent data presented here corresponds to the raw waste
      data shown in Table 21-6.  The methodology of the sampling program
      is described in Section 5.1.2, and the scope of sampling in the
      industry is described in Section 21.3.3 Copper Sulfate.
 (2)  Screening data from one 72-hour grab conposita sample of treated
      effluent.
 (3)  Verificatim data from three 24-hour grab composite samples, averaged.
 (4)  Vhen averaging values indicated as "less than" ( < ) ,the absolute
      value was used and the resulting average was indicated as a "less
      than" value.
                        657

-------
TABLE 21-11.
AVERAGE POLLUTANT LEVELS AND REMOVAL EFFICIENCY FOR PLANT #034

Subcategory: Copper Sulfate
Waste Water Flow = 1.25 m3Akg
Pollutant
TSS
Copper
Nickel
Antimony
Arsenic
Cadmium
Chromium
Lead
Selenium
Zinc
Haw
(mg/1)
410
2000
102
0.44
24.0
1.2
0.25
0.48
< 0.008
12.0
Waste
(kg/kkg)
0.92
4.5
0.23
0.00095
0.052
0.0027
0.00055
0.0011
0.000018
0.026
Treated
(mg/1)
24.0
3.3
0.20
0.08
0.038
0.0026
0.011
< 0.018
0.10
0.018
Effluent *
(kg/kkg)
0.054
0.0072
0.00046
0.00018
0.000085
0.0000054
0.000024
0.00004
0.00023
0.00004
Percent
Removal
94
99 +
99 +
81.6
99 +
99 +
96
96
Effluent
> Inf luent
99 +
*  Before cotbining with noncontact cooling and steam condensate streams.
                                     658

-------
     included selenium limitations,  although  selenium  was  not
     found  to  be a significant pollutant in raw wastes at Plant
     #034.  However, its continued presence in the effluent  from
     alkaline  treatment  in significant concentrations indicates
     that  selenium  should  continue  to  be  included  in   the
     pollutants to be regulated.

3.    Final Selection of Pollutants to be Regulated

     Originally,  1 imitations were proposed for all the previously
     mentioned significant toxic pollutants.   The Agency has  now
     decided  to  promulgate  regulations  on  copper, nickel and
     selenium.  Copper and nickel are the most predominant  toxic
     metals in the wastewater of this subcategory.  The nature of
     the  treatment technology used by this industry is such that
     the control of the dominant metals will   ensure  control  of
     all the toxic pollutants of concern.  Further elaboration of
     this subject may be found in Section 8 of this document.  It
     was  also  decided  that  selenium  should  continue  to  be
     regulated where it is shown to be present  in  the  effluent
     and posing a potential problem.   At plants where selenium is
     not found at significant concentrations  in the effluent, the
     limitations  would  not  apply.    Because selenium was found
     only in the effluent, and not in the raw waste at  treatable
     levels,  it may not be directly related  to the manufacturing
     process.

     Section 8 indicates that toxic metals can  be  divided  into
     two  groups  for  pH optimization of alkaline precipitation,
     and that control of one or more of the  metals  in  each  of
     these  groups,  will  ensure control of  others in that group
     (see Table 8-14).  By controllng both copper and nickel, the
     two different metal groups are covered.   Therefore,  because
     these  two  metals  (copper  and nickel) are the predominant
     toxic  pollutants  in  the  raw  waste,   control  of   these
     parameters  will ensure control  of all the metals previously
     considered for limitations.  The data from all the samplings
     at Plant #034 supports this.

     Optimization of treatment conditions for nickel  and  copper
     removal may cause a lack in performance  of chromium control.
     However,  because  of  the  relatively low incoming chromium
     concentrations, the control of chromium  should be  adequate.
     Although  aresenic  is  not in either of the two groups, and
     the incoming concentrations were the  highest  after  copper
     and  nickel, the results of the sampling data show excellent
     arsenic removal and very low effluent concentrations.  Table
     8-14 and Figure 7-1 indicate a pH around 10 would be optimum
     for the control of pollutants in this subcategory.
                             655

-------
C.   Basis of Pollutant Limitations

     The BPT Effluent Limitations are presented in Table 21-12.  Since
     the  existing   BPT   limitations   are   being   retained,   the
     concentration  bases for 4the maximum 30-day average and the daily
     maximum are back calculated from the pollutant  mass  limitations
     by  applying  the  model  plant  flow  rate  of 0.94 mVkkg.  The
     variability factors used are consistent with those estimated from
     the statistical analysis of treatability data (61).
1
          Conventional Pollutant Limitations
               pH

               The treated effluent is to  be  controlled  within  the
               range  of  pH  6.0 to 9.0.  This limitation is based on
               data  presented  in  Appendix   B   of   the   proposed
               Development Document (60) and the JRB study (52).

               TSS

               The existing BPT maximum 30-day average TSS  limitation
               is 0.023 kg/kkg.  The corresponding concentration basis
               is:

               (0.23 kq/kkq) (1000 mg/1 ) = 24 mg/1
               (0.94 mVkkg ) (  kg/m*  )

               The  existing BPT daily maximum TSS limitation is 0.069
               kg/kkg and the concentration basis is:

               (0.023 kq/kkq) (1000 mq/1) = 73 mg/1
          (0.94 mVkkg
                                 kg/m*  )
                                                        from  the
                                                        estimated
                                                       follows:
               The long-term average for TSS is  calculated
               daily   maximum   concentration   and   the
               variability factor for daily measurements as

               (73 mq/1) = 20 mg/1
                 (3.6)

     2.   Toxic  Pollutants
          The effluent limitations for the selected  toxic  pollutants
          are  based  on  the  existing BPT regulations for the Copper
          Sulfate Subcategory.  However, the  separate  limitation  on
          copper  for  the  pure raw material process is being dropped
          and the present limitations  on  TSS,  copper,  nickel,  and
          selenium for the recovery process (Table 21-2) will apply to
          both processes.
                                660

-------
Copper

The  existing  BPT  maximum  30-day  average   effluent
limitation  on  copper  is 0.0010 kg/kkg.  Applying the
model  plant  unit  flow  rate  of  0.94  m3/kkg/   the
concentration basis is calculated as follows:

(0.0010 kq/kkq) (1000 mq/1) =1.1 mg/1
(  0.94 mVkkg ) (   kg/m3  )

Similarly, the existing BPT daily maximum limitation is
0.0030 kg/kkg, and the concentration basis becomes:
 0.0030 kq/kkq
  0.94 mVkkg)
(1000 mq/1)
(   kg/m3   )
=3.2 mg/1
The  long-term  average  for  copper is then determined
from  the  estimated  variability  factor   for   daily
measurements:

(3.2 mg/1? = 0.89 mg/1
 (3.6)

Nickel

The  existing  BPT  maximum  30-day  average   effluent
limitation on nickel is 0.0020 kg/kkg.  Using the model
plant  unit flow rate of 0.94 mVkkg, the concentration
basis is calculated as:

(0.0020 kq/kkg) (1000 mq/1) =2.1 mg/1
( 0.94 mVkkg ) (  kg/m3)

The existing BPT daily  maximum  limitation  is  0.0060
kg/kkg and the corresponding basis is:

(0.0060 kq/kkq) (1000 mq/1) =6.3 mg/1
( 0.94 mVkkg ) (  kg/m3)

The   expected   long-term   average  concentration  is
determined from the estimated- variability  factor  for
daily measurement:

(6.4 mq/1) =1.8 mg/1
 (3.6)

Selenium

The existing BPT maximum 30/day average  limitation  on
selenium is 0.00050 kg/kkg.  The concentration basis is
calculated as:
                   €61

-------
                                         TABLE 21-12.  EFFLUENT LIMITATIONS
                                                   Copper Sulfate                    ...
                             Best Practicable Control Technology Currently Available
                                       Waste water Flow; 0.94 n?/kkg of CuSO4
en
to

Subcategory -2\ Concentration Basis
Pollutant Performance VFR J (mg/1)
(rog/1)



Conventional Pollutant:
TSS 20
Toxic Pollutants:
Copper 0.89
Nickel 1.8
Selenium 0.44
Max.
30-day
Avg.

3.6/1.2 24

3.6/1.2 1.1
3.6/1.2 2.1
3.6/1.2 0.53

24-hr.
Max.

73

3.2
6.4
1.6
Effluent Limit
(kg/kkg)
Max.
30-day
Avg.

0.023

0.0010
0.0020
0.00050

24-hr.
Max.

0.069

0.0030
0.0060
0.0015
         (1)  Based on the existing BPT limitations, 40 CFR 415.360.  Also applicable to NSPS.

         (2)  Ratio of the variability factor for daily measurements to the variablility factor for
              30-day average.

-------
               (0.00050 kq/kkq) (1000 mq/1)  = 0.53 mg/1
               (  0.94 mVkkg  ) <   kg/m*  )

               The  existing  BPT  daily  maximum limitation is 0.0015
               kg/kkg and the concentration basis is:

               (0.0015 kq/kkq) HOOP mq/1) =1.6 mg/1
               (0.94 mVkkg ) (   kg/m*  )

               The expected long-term average is then determined  from
               the    estimated    variability    factor   for   daily
               measurements:

               (1.6 mg/1)  =0.44  mg/1
                (3.6)

Basis for BCT Effluent Limitations

While  SPA  has  not  yet  proposed  or  promulgated  a  revised   BCT
methodology  in  response  to  the  American  Paper  Institute  v. EPA
decision mentioned in Section 3,  EPA is promulgating  BCT  limitations
for  this  subcategory.   These limits are identical to those for BPT.
           promulgating any more stringent limitations since  we  have
             no  technology  option  which  would  remove  significant
           amounts of conventional pollutants.  As BPT is the  minimal
           control required by law, no possible application of the BCT
             could  result  in  #CT  limitations  lower   than   those
EPA is not
identified
additional
level  of
cost  tests
promulgated in this regulation.  Accordingly, there is no need to wait
until   EPA  revises  the  BCT  methodology  before  promulgating  BCT
limitations.

Basis for BAT Effluent Limitations

The Agency is promulgating BAT  limitations  that  are  equal  to  the
existing  BPT  limitations  for the selected toxic pollutants (copper,
nickel, and selenium) as indicated in  Table  21-13.   Also  shown  in
Table  21-13  are  concentration  guidance  values for the other toxic
pollutants  (arsenic, cadmium,  zinc,  chromium,  lead,  and  antimony)
which  do not have effluent limitations.  The basis for these guidance
values  are  developed  below  where  treatability  based  performance
estimates for Level 1 treatment are presented.

When  the  BAT regulations were recently proposed (60), the Agency had
considered  requiring  more  stringent  limitations  for   the   toxic
pollutants  because the results of treatability studies (61) indicated
that Level 1 treatment system performance was better than the previous
BPT data base suggested.  However", the actual  achievability  of  this
improved  performance  is not sufficiently substantiated at present to
warrant the promulgation of more stringent limitations and, therefore,
the existing BPT limitations, with the Level  1  treatment  technology
basis, are being promulgated for BAT.
                                 663

-------
The   Agency  also  considered  control  treatment  Level  2  (sulfide
precipitation), but rejected this treatment because it removes only  a
small  additional  amount  of  toxic  metals which did not justify the
additional cost.

Basis for New Source Performance Standards

The Agency is promulgating New  Source  Performance  Standards  (NSPS)
based  on  treatment  technology  equivalent to BPT/BAT for the Copper
Sulfate Subcategory.  The conventional pollutant limitations  for  TSS
are  the  same  as  shown for the existing BPT (BPCTCA) regulations in
Tables 21-2 and 21-12.  Also, pH is limited to the range  6  to  9  as
based  on  data  presented  in  Appendix B of the proposed Development
Document (60) and the JRB study (52).  The toxic pollutant  parameters
to be regulated are those identified in the development of the BPT/BAT
regulations  as  shown  in  Table  21-12  and  the  specific numerical
limitations proposed for NSPS are identical  to  those  indicated  for
BPT/BAT.

Basis for Pretreatment Standards

There  is  an existing PSES regulation, 40 CFR 415.364, which is based
on  BPT.   The  Agency  is  amending  that  section  in  these   final
regulations to establish the same effluent concentrations as were used
for  the basis on BPT.  In the original regulation, the concentrations
for copper were 0.5 mg/1 as the maximum 30-day average and 1.0 mg/1 as
the daily maximum.  For nickel, the concentrations were  1.0  mg/1  as
the  maximum  30-day average and 2.0 mg/1 as the daily maximum.  Under
the amended PSES regulation, the concentrations for copper become  1.2
mg/1  and  3.2  mg/1 for the maximum 30-day average and daily maximum,
respectively, and for nickel, 2.1 mg/1 and 6.4 mg/1 as shown in  Table
21-13.  Selenium is also regulated under the amended PSES regulation.

For new sources, the Agency is setting PSNS equal to NSPS as indicated
in Table 21-13.  Copper, nickel, and selenium are regulated.

Pretreatment  standards  are  necessary  because  BAT  provides better
removal of toxic metals than is achieved by a well-operated POTW  with
secondary  treatment  installed and, therefore, these pollutants would
pass through a POTW in the absence of pretreatment.

Basis for Level 1 Treatment Performance

A.   Technology Basis

     The screening and  verification  data  were  collected  when  the
     filter  in the treatment system was not operating properly; hence
     the  treated  effluent  data  were  not  representative  of   the
     performance  achievable  with precipitation/filtration technology
     and, therefore, the basis for Level  1  performance  was  derived
     from  other  sources of performance data.  The treatability study
     results (61) are summarized  and  incorporated  with  other  data
     submitted  by  industry,  as  well  as  the  sampling  data,  and
                                   664

-------
Ul
                                         TABLE 21-13.  EFFLUENT LIMITATIONS
                                                   Copper Sulfate   m
                                           Best Available Technology
                                      Waste Water Flow:  0.94 m3/kkg of CuSO4

Pollutant

Copper
Nickel
Selenium
Arsenic
Cadmium
Zinc
Chromium
Lead
Antimony
Subcategory
Performance
fog/1)

0.89
1.8
0.44
0.20
0.10
0.20
0.16
0.050
0.40
,„. Concentration Basis
VFIT J (mg/1)
Max.
30-day 24-hr.

3.6/1.2
3.6/1.2
3.6/1.2
4.0/1.2
4.0A-2
3.5/1.2
3.5/1.2
3.5/1.2
4.0/1.2
Avg.
1.2
2.1
0.53
0.24
0.12
0.24
0.20
0.060
0.48
Max.
3.2
6.4
1.6
0.80
0.40
0.70
0.56
0.18
1.6
Effluent Limit
(kg/kkg)
Max.
30-day 24-hr.
Avg.
0.0010
0.0020
0.0005Qv
	 (3)
	 (3)
	 (3)
	 (3)
	 (3)
	 (3)
Max.
0.0030
0.0060
0.0015
	 (*)
	 (3)
	 (3)
	 (3)
	 (3)
	 (3)
         (1)  Also applicable to PSES and PSNS.
         (2)  Ratio  of the variablility factor for daily measurements to the variablility factor for
              30-day averages.

         (3)  No effluent limitation has been established.

-------
     presented in Tables 8-12 and 8-13.  Where there was  insufficient
     data  on a particular pollutant from these sources, the published
     literature treatability data from Table 8-11 was used to estimate
     performance.


     The flow basis and the selection of pollutant parameters  is  the
     same as that presented for the BPT limitations in this section.

B.   Basis of Pollutant Control Guidance

     1.    Toxic Pollutants

          The  effluent  control  guidance  for  the  selected   toxic
          pollutants  are  derived  from estimated industry achievable
          long-term averages (Tables 8-12 and  8-13),  and  literature
          based   treatability   estimates   (Table  8-11).    This  is
          necessary because  plant  performance  data  from  long-term
          monitoring  {Table  21-9) and the screening and verification
          sampling (Table 21-10) do not reflect optimum operation of a
          BPT system for removal of copper and nickel.   In  addition,
          these  sources  show effluent concentrations below the lower
          limit of treatability estimates  for  all  the  other  toxic
          metals.

          Estimated  effluent  mass  loadings  for  copper,  nickel and
          selenium  are  shown  in  Table   21-14   along   with   the
          corresponding  concentration  basis.  The performance levels
          estimated on the basis of treatability data are better  than
          are  presently achieved by dischargers in the Copper Sulfate
          industry and the Agency does not  have  sufficiant  evidence
          that  the  improved  performance can actually be achieved in
          practice.   The  estimated  performance  is   for   guidance
          purposes only.  Also shown are the recommended concentration
          bases  for  the other toxic metals.  These are included as a
          guidance reference, should the control of  one  or  more  of
          these metals be found necessary in specific cases.

          The  concentration  bases  for  copper,  cadmium,   zinc  and
          chromium are derived  from  estimated  achievable  long-term
          industrial  averages  (Tables 8-12 and 8-13).  Concentration
          bases for nickel and arsenic are derived from  the  observed
          concentrations  from  sampling  of several Level 1 treatment
          systems, supported by industry averages, while those for the
          remaining metals of concern  were  derived  from  the  lower
          limit of the literature treatability estimates (Table 8-11).
          This approach is used to estimate achievable performance and
          provides  for  wider variations in the influent quality that
          may be associated with different  purities  of  copper  feed
          material or other process variables.

          Variability  factors  used  to  calculate the maximum 30-day
          average and the daily maximum concentrations from  long-term
                                 666

-------
averages  were  determined  by
presented in the Treatability
Sulfate Subcategory.
                           statistical
                          Study  (61)
analysis
for  the
of data
 Copper
a.
Copper
     The estimated achievable long-term average  for  copper
     indicated  in  Table  8-13 is 0.30 mg/1.   This reflects
     industry performance using Level 1 technology.  This is
     below the effluent level found during sampling of Plant
     #034, indicating that the system was not being  run  at
     its   optimum   performance.    Therefore,  using  this
     concentration value and a 30-day variability factor  of
     1.2  from  the  copper sulfate treatability results for
     copper (61),  the maximum 30-day concentration basis is:

     (0.30 mg/1) (1.2) = 0.36 mg/1
     The 24-hour maximum variability factor derived from the
     copper sulfate Treatability Study for copper  (61)  was
     found  to  be  3.5.   Therefore, using this variability
     factor and the long-term average concentration of  0.30
     mg/1, the daily maximum concentration is:
(Q.3Q
                 U.5) = \ .\ mg/1
     The  effluent loadings for copper were determined using
     a model plant flow of 0.94  mVkkg,  with  the  maximum
     30-day average concentration calculated as follows:

     (0.36 mg/1) (0.94 mVkkg)  (kq/rn^) = 0.00034 kg/kkg
                              (1000 mg/1)

     The  variability factor ratio is the ratio of the daily
     maximum  variability  factor  and  the  30-day  average
     variability factor or:

     VFR = VF of daily measurements
            VF of 30-day averages

     For copper the VFR is equal to:

     VFR = 3.5 =2.9
           1 .2

     Therefore, the 24-hour maximum is calculated to be:

     (2.9) (0.00034 kg/kkg) = 0.0010 kg/kkg
                       667

-------
                TABLE 21-14.  GUIDANCE FOR EFFLUENT CONTROL
                              Copper Sulfate
                             Level 1 Technology
                  Waste Water Flow:  0.94 m3Akg of CuSO.

Long-Term
Pollutant Average ,.»
(mg/1) VFRW



Copper 0.30(1> 3.5/1.2
Nickel 0.20(2) 4.0/1.2
C*}
Selenium 0 . 10 u } 4 . 0/1 . 2
Arsenic 0.20(2) 4.0/1.2
Cadmium 0.10(2) 4.0/1.2
Zinc 0.20(1) 3.5/1.2
Chromium 0.16(11 3.5/1.2
Lead 0.050(3) 3.5/1.2
Antimony 0.40(3J 4.0/1.2
Concentration
Basis
Effluent Loading
( kg/kkg )

(mg/D
Max
30-day
Avg
0.36
0.24

0.12
0.24
0.12
0.24
0.20
0.060
0.48

24-hr
Max
1.1
0.80

0.40
0.80
0.40
0.70
0.56
0.18
1.6
Max
30-day
Avg
-J5)
-J5>
(5)
v-* i
_J5)
_J5)
_J5)
-J5>
-J5>
_<5>

24-hr
Max
_J5>
__<5>
f 5)
—
..(5)
-J5)
_(5>
— <5)
— <5>
-J5)
(1) Industry Long-Term Average (Table 8-13) .
(2) Industry and Treatability Data
(Table 8-12) ,
(3) Lower limit of literature treatability data
(4) Variability Factor Patio: the
the monthly variability factor
study data for copper sulf ate
ratio of the
•
(Table

8-11) .


daily variability factor to
based on statistical
(61).

analysis of

treatability

(5)   Indicates  no control specification is needed.
                                     668

-------
Nickel

The observed concentration of nickel  in  the  effluent
samples  from Level 1  treatment for both copper sulfate
and nickel sulfate was 0.20 mg/1.    This  supported  by
other  industry  data  using  Level 1  technology (Table
8-12) and is used as the estimated achievable long-term
average.   The 30-day  average  variability  factor  for
nickel  is  also 1.2.  and the daily maximum variability
was 4.0.   Therefore, the  calculation  of  the  maximum
30-day average concentration basis is:

(0.20 mg/1) (1.2) = 0.24 mg/1

and the 24-hour maximum concentration is:

(0.20 mg/1) (4.0) = 0.80 mg/1

The VFR for nickel is calculated as:

4.0 = 3.3
1 .2
Therefore,  the  maximum 30-day average based on a flow
of 0.94 mVkkg is:

(0.24 mg/1) (0.94 mVkkg)  (kq/m*) = 0.00023 kg/kkg
                        (1000 mg/1)

and the 24-hour maximum is:

(3.3) (0.00023 kg/kkg) = 0.00076 kg/kkg

Selenium

Long-term  monitoring  and   sampling   data   indicate
effluent  quality either at or below the lower limit of
the  estimated  literature  treatability  data   (Table
8-11).   This  lower  limit of 0.1 mg/1 was used as the
long-term  average  concentration  because  there   was
insufficient  industry  data  in  Tables 8-12 and 8-13.
Using  the  same  variability  factors  and   rationale
applied   to   nickel,   the   maximum  30-day  average
concentration basis is:

(0.10 mg/1) (1.2) = 0.12 mg/1

and the 24-hour maximum concentration is:

(0.10 mg/1) (4.0) = 0.40 mg/1

The effluent loading for the maximum 30-day average is:
                    669

-------
(0.12 mg/1) (0.94 mVkkg)  (kg/ri^) = 0.00011 kg/kkg
                         (1000 rng/1)

and for the 24-hour maximum:

(3.3V (0.00011 kg/kkg) = 0.00038 kg/kkg

Arsenic

From the data in Table 8-12,  the observed concentration
of arsenic in the effluent of  Level  1  treatment  for
copper  indicated  a  value  of  0.20  mg/1.  Using the
variability factors from  nickel,  the  maximum  30-day
average concentration basis is:

(0.20 mg/1) (1.2) = 0.24 mg/1

and the 24-hour maximum is:

(0.20 mg/1) (4.0) = 0.80 mg/1

Cadmium

Industry  data  from  metal  finishing   indicates   an
effluent  value  of 0.10 mg/1 using Level 1 technology.
This was used as the  long-term  average  concentration
basis   rather  than  the  lower  1imit  of  1iterature
treatability estimates to account for more variation in
influent  and  treatment   operation.    Again,   using
variability  factors  for  nickel,  the  maximum 30-day
average concentration basis is:

(0.10 mg/1) (1.2) = 0.12 mg/1

and the 24-hour maximum basis is:

(0.10 mg/1) (4.0) = 0.40 mg/1

Zinc

Industry averages from Tables 8-12 and 8-13 indicate  a
long-term  average  concentration of 0.20 mg/1.  Again,
using the variability factors for copper,  the  maximum
30-day average concentration basis is:

(0.20 mg/1) (1.2) = 0.24 mg/1

and the 24-hour maximum concentration basis is:

(0.20 mg/1) (3,5) = 0.70 mg/1
                    670

-------
Chromium

Industry averages from Tables 8-12 and 8-13 indicate  a
long-term  average  concentration of 0.16 mg/1.   Again,
using the variability factors for copper,  the  maximum
30-day average concentration basis is:

(0.16 mg/1) (1.2) = 0.20 mg/1

and the 24-hour maximum concentration is:

(0.16 mg/1) (3.5) = 0.56 mg/1

Lead

From Table 8-11, the lower limit of  treatability  from
literature  data  shows a value of 0.05 mg/1.   This was
used as the achievable long-term average  concentration
for   the   same   reasons   as  selenium.   Using  the
variability factors  for  copper,  the  maximum  30-day
average concentration basis is:

(0.05 mg/1) (1.2) = 0.060 mg/1

and the 24-hour maximum concentration basis is:

(0.05 mg/1) (3.5) = 0.18 mg/1


Antimony

The long-term average concentration value of 0.40  mg/1
was   taken   from   the   lower  limit  of  literature
treatability data (Table 8-11) for the same reasons  as
selenium.   Using  the  variability factors for nickel,
the maximum 30-day average concentration basis is:

(0.40 mg/1) (1.2) = 0.48 mg/1

and the 24-hour maximum concentration basis is:

(0.40 mg/1) (4.0) = 1.6 mg/1
                  671

-------

-------
                              SECTION 22

                       NICKEL SULFATE INDUSTRY

Industrial Profile

General Description

Most of the nickel sulfate produced is sold in  the  merchant  market.
The  major use of nickel sulfate is in the metal plating industry, but
it is also used in  the  dyeing  and  printing  of  fabrics,  and  for
producing a patina on zinc and brass.

The  industry  profile  data summary is given in Table 22-1, while the
status of regulations is summarized in Table 22-2.

Subcategorization

Several factors were originally considered  in  the  subcategorization
process,  such as raw materials, products, manufacturing process, size
and age of equipment, and water  pollution  control  technology.   The
conclusion  was  made  that if effluent limitations were to be tied to
units of production, only subdivision by dominant product  was  viable
as a method of primary subcategorization.  Option was left for further
subdivision,  if  necessary;  however,  this  was not warranted in the
nickel sulfate industry.
General Process Description and Raw Materials

Nickel sulfate is produced by reacting various forms
sulfuric acid.  The general reaction is:
                                                      of  nickel  with
NiO
            = NiS04 + H20
Two different types of  raw  materials  are  used  to  produce  nickel
sulfate.   Pure  nickel  or  nickel oxide powder may be used as a pure
material  source,  while  spent  nickel  catalysts,   nickel   plating
solutions or residues are impure sources.

The  nickel  sulfate  produced  when  pure  raw  materials are used is
filtered and sold or processed further.  This is done by  heating  the
solution  to 300°C in a crystallizer to produce a solid nickel sulfate
product.  This must be classified, dried, and screened  before  it  is
ready for sale.

The  use  of  impure  raw materials produces a nickel sulfate solution
which must be treated sequentially with oxidizers, lime, and  sulfides
to  precipitate  impurities which are then removed by filtration.  The
nickel sulfate solution can be sold or it may be crystallized, and the
crystals classified,  dried,  and  screened  to  produce  sol id  nickel
sulfate  for  sale.   Figure 22-1 shows a general process flow diagram
for the manufacture of nickel sulfate.
                                673

-------
                TABLE 22-1.   SIBCKTEGORY PROFILE DATA SUMMARY
 SUBCATEGQRY    NICKEL SULEATE
 Total subcategory capacity rate (1)
 Total subcategory production rate (1)
 Number of pi ants in this subcategory (2)
 308 Data on file for
     With total capacity of
     With total production of
     Representing capacity
     Representing production
     Plant production range:
             Minimum
             Maximum
     Average production
     Medium production
     Average capacity utilization
     Plant age range:
             Minimum
             Maximum
     Waste water flow range:
             Minimum
             Maximum
     Volume per unit product:
             Minimum
             Maximum
Indeterminant
    6,350 kkg/year
       11
        6
   17,700 kkg/year
   12,650 kkg/year
       NA
       NA
       NA
       45 kkg/year
    5,900 kkg/year
    2,100 kkg/year
    1,600 kkg/year
       71.5

        3
       48

   1.5 cubic meters/day
  17.0 cubic meters/day

   0.42 cubic meters/kkg
   0.72 cubic meters/kkg
(1)= "Economic Analysis of Proposed Revised Effluent Guidelines and
     Standards for the Inorganic Chemicals industry/' March,  1980.
(2) = Sources of data are Stanford Research institute, Directory of  Chemical
     Producers, U.S.A.,  1977
NA = Not Available
                                     674

-------
    TABLE 22-2.  STATUS OF RBGOIATICNS  -  EETXOENT UMITKnaN GUIDELEffiS
SCBCKEEGORX" Nickel Sulf ate
SUBPAKT
AU (40 CF-R 415.470,
5/22/75)

STANDARDS
Product
Process
Pure
Raw
Materials
Bnpure
Raw
Materials
BfCICA,
Max*1 Avg.2
Para- kg/kkg kg/kkg
meters (mg/1) (mg/1)
... Mo discharge
Ni of pwwp3
-,- Nb discharge
ras ofpwwp
m 0.006 0,002
.j^ 0.096^ 0.032
BAIEA
Max. Avg.
kg/kkg kg/kkg
(nq/1) Ong/1)
No discharge
of pwwp
Nb discharge
of pwwp

NSPS
Max* Aver.
kg/kkg kg/kkg
(rng/1) (mg/1)
Nb discharge
of pwwp
No discharge
of pwwp

 Max, « Maxiiman of any one day*
2
 Avg, « Average of daily valiies for thirty consecutive days.

       > Process wastewater pollutants.
                                  675

-------
Water Use and Waste Source Characteristics

Water Use

Noncontact cooling water is used for nickel sulfate production in  the
reactor and crystallizers, and constitutes the major water use.  Water
is  used in direct process contact as a reaction component.  A portion
of  this  water  goes  into  the  dry  product   as   its   water   of
crystallization  and  the remainder is evaporated into the atmosphere.
Small amounts of water are used for maintenance  purposes,  washdowns,
cleanups,  etc.,  and  several  plants use water in scrubbers for dust
control.  Table 22-3 gives a summary of water usage for  plants  where
information   was   available  from  308-Questionnaires  and  previous
documents.

Waste Sources
                                                           This stream
A.   Noncontact Cooling Water

     Noncontact cooling is the main source of wastewater.
     is usually not treated before discharge.

B.   Direct Process Contact

     Plants which use impure nickel raw materials  generate  a  filter
     sludge  which  is treated as a solid waste.  They also generate a
     small filter backwash waste  stream  with  high  impurity  levels
     which  must be treated before discharge.  The filter sludges from
     processes using pure nickel can be recycled back to the  process.
     Mother  liquor,  and wastewater streams from dust control are also
     recycled back to the process,

C.   Maintenance

     Washdowns, cleanups, spills, and pump leaks are  perodic  streams
     and  account  for the remaining wastes produced by nickel sulfate
     plants.

     Table 22-3 also shows the unit flow of total wastewater generated
     from  the  nickel  sulfate  process  at  each  plant  where  this
     information was available.

Description of_ Plants Visited and Sampled

Screening

Plant  #369  was  visited and process wastewater  and effluent samples
were collected and analyzed for  conventional  and  toxic  pollutants.
The  process  used  at Plant #369 is similar to that described earlier
and utilizes nickel oxide powder feedstock.  Mother liquor is recycled
back  to  the  reactor.   Sources  of  wastewater  consist  of   small
quantities  of  mother  liquor  from the filter press, washdown water,
leaks, and spills.  Wastewater from the process area is  collected in a
                                   676

-------
NICKEL
OIGESTOR
OXIDE
SOLUTION
PRODUCT
-^
LIOU

OIGESTOR
t - *
|— FILTER — | f—
FILTER
WENT NICKEL T 1 ...»* „„..„,
^ju, 	 ^ OIGESTOR
^^— K3IOUC3
AGO I


^K— OXIDIZES

j FILTER
i
auffmi H TREATING TANK
t
| FILTER


CONCENTRATOR |
^
1 FILTER

|
1 CRYSTALJUZER ]
j


f


	 ^ 3LU08C


^h">— •-•» SPENT PL ATI NO SOLUTION


«
ACID
^ SUIOM
tZ^—



QC LAS
UENT
VAPORATION TANKf*~ 	 STEAM
^ COOL0M
^ WATER
1
OUSTS

* :-si
( COOL, SCREEN,
I PACKAGE
SOLID MWOOCT

OUSTS
i
HOLDING TANK
t ..
SCRUBBER •• w«ra»

    Figure 22-1.  General process  flow diagram for nickel sulfate manufacture,
                                  677

-------
 TABLE 22-3-  MITER USE  IN THE NICKEL SULFATE SUBCATEGORY
 Source
#313*
   Water Uses at Plants (m3Akg)

#069       #572      #369      #120
                            #603*
Direct  Process
Contact

Noncontact
Cooling
Water

Maintenance
Cleaning and
Hashdowns,
Pumps,  Seals
and Leaks

Air Pollution
Control
Vfeste pfeter

Haste
Water
Flow to
Treatment
24.8
37.3
 0.0098     0,35
 1.67
4.98
 0.278     0
          0.751     4.01   814
0.417    13.6   2035
 0.278     0.00196    0.896     0.094     Nil
                                         Nil
            0.498     0.094     1.28     0
23.4(1)     .0196(1)  20.3(1)*   0.42<2)    0.72(2>   NA
 * = Flow data  includes uses for other products.
(1)= Data source:   308-Questionnaires
(2)= Data source:   Plant visits
 NA  = Not Available
                                      678

-------
tank and treated as a batch by adjusting the pH to  about  12.5  using
sodium  hydroxide.   The  precipitated metal hydroxides are allowed to
settle, and the supernatant is decanted to another tank,  checked  for
quality  and  discharged  to  a POTW.  The sludge is hauled away to an
approved landfill.  Figure 22-2 shows a general treatment system  flow
diagram  with  the  location of the sampling points.  Table 22-4 gives
data on flow, total suspended solids  (TSS),  and  nickel  and  copper
emissions for the waste streams sampled during screening.

Verification

Plants  #572 and #120 were visited and sampled during the verification
phase of the program.  At Plant #572, pure nickel oxide is used as the
raw feedstock.  The wastewater streams discharge on a batch basis  and
are  collected  together  in  a  floor  drain.   The wastes consist of
washdowns,  leaks, and air scrubber water which  are  collected  in  an
equalization  tank.   In  the  equalization tank, alkaline wastes from
another process are mixed in and the pH is raised to 10.   Solids  are
allowed to settle and the clear supernatant is discharged to a POTW.

Plant  #120  uses  nickel  oxide  powder  and  impure  nickel  as  raw
materials.   Wastewaters from the nickel sulfate process  emanate  from
the  filter  wash, air scrubber, washdowns, and leaks,  and are sent to
the treatment system.  The raw  wastes  are  mixed  with  other  plant
nickel  raw wastes prior to treatment.  This consists of pH adjustment
to  precipitate  nickel  and  other  trace  metals  followed  by  sand
filtration.

Figures  22-3  and 22-4 show the general treatment system flow diagram
with the waste streams sampled for Plants #572 and #120, respectively.
Table 22-4 also shows the waste stream flow and waste  characteristics
for  both  plants.   The  data  for  Plant  #572  are  presented  on a
concentration basis only, because a representative flow value for  the
sampling point was unavailable.

Summary of Toxic Pollutant Data

Seven  toxic pollutants were found at detectable concentrations in the
raw waste sample from nickel sulfate at  Plant  #369.   Six  of  these
toxic  metals  were  verified  in  the  raw  waste at two other nickel
sulfate plants. In addition, two more toxic pollutants  were  observed
at  detectable  concentrations  in  the  raw waste during verification
sampling.  No toxic organics were found at  detectable  concentrations
in   the  raw  waste  at  Plant  #369.   Consequently,   organic  toxic
pollutants were not sought in the  verification  phase.    The  results
were:
                               679

-------
                                                                 MttCU
00
o
                      Figure 22-2.  General waste water treatment process flow diagram showing
                                    sampling points at plant 1369.  (Nickel sulfate sdxategory.)

-------
 TABLE 22-4 .  FLOW AND POIIOTANT CONCENTRATION DATA OF THE SAMPLED
              STREAMS FOR PLANTS PRODUCING NICKEL SULFATE

SUBCATEGQRY NICKEL SULFATE
Stream
No.
Sampled
Stream
Description
Flow
/m3/kkg\
•°f
\NiS04/
TSS
/kg/kkg\
( -°f j
\NiS04/
Ni
/kg/kkg\
V of )
XNiS)/
Screening Data ' Plant
1
2
Raw untreated waste
Treated waste
0.42
0.42
0.093
0.045
0.073
0.00058
Cu
/kg/kkg\
( -°f )
\NiSCy
#369
0.030
0.0076
Verification Data

1
2

Raw NiSO4 waste
All Nickel raw

0.72
0.72
Plant
0.031
0.05
#120 (2)
0.035
0.0089

0.00016
<0. 0000036
        vastes*
        Treated effluent*
        Scrubber waste
0.72
0.0031
0.00014
                                              Plant #572
                                                        (3)
            3.2
                 (mg/1)

                1100
0.000031
              .04
(1) =  One grab sample of each waste water stream representing a
       composited batch  sample of that day's nickel  sulfate production.
(2) =  Average  of three  24-hour composite samples of each waste water
       stream.
(3) =  Flow data  was unavailable.  Only waste water  quality is presented
       here.
 *  =  The stream is a commingled waste water. The flow given is the amount
       contributed by the nickel sulfate plant.
                                    681

-------
                                                     "
00
10
                                                                                   MUNICIPAL
                                                                                   SUPPLY WATER
LEAKS. SPILLS. ETC..  FROM
   OTHER PROCESSES
                                                               UASTt WATER
                                                               COLLECTION
                                                                  SunP
                                                                                                           LEAKS
                                                                                                           WASHDOUM
                                                             SOLID wsu
                                                             TO HSPOSAL
             EQUALIZATIOH
                 TAHK
                pN - 10
                                                                                                              ALKALINE WASTES
                                                                                                                                                                 PCINIS.
                                                                                       DISCHARGE TO
                                                                                          SEWCR

-------
                                  OSHER SXCKEL WASTES
   NtSO. PEOCESS
       HUOT
                                                                   WXCSSS
                                                      DISCHARGE
Figure  22-4.   General waste water treatment process  flow diagram at plant #120
               showing the sampling points.  (Nickel  sulfate manufacture.)
                                 683

-------
Pollutant
                Maximum
    Concentration
        Ug/1)
Screening
Plant #369
Observed
Verification (2 Plants)
 Plants #572 and #120
Nickel
Copper
Chromium
Antimony
Lead
Mercury
Cadmium
Selenium
Zinc
175,500
73,300
1 ,300
476
55
1 .0
9.0
10
430
1 ,115,000
355
20
18
120
10
160
141
382
Section  5  of  this report describes the methodology of the screening
and verification sampling program.  In the nickel sulfate industry,  a
total  of  seven  days of sampling were conducted at Plants #369, #572
and #120,  Nine different sampling points were involved  covering  the
raw  waste  source,  the  various  raw  waste streams, and the treated
effluents at these plants.  The evaluation of toxic metal  content  of
these  process-related  waste streams was based on 195 analytical data
points.  The screening for toxic  organic  pollutants  at  Plant  #369
generated an additional 342 analytical data points.  The unit loadings
were  calculated from the reported nickel sulfate production rate, the
waste stream flow rate measured or estimated at the time of  sampling,
and the measured pollutant concentration.
     That is,

     Unit loading (as kg of pollutant per
     kkg of nickel sulfate}

     where:
                                (Q)
                           1000 (P)
     C  is  the  concentration of the pollutant expressed in limits of
     mg/1 (Note:  kg/m3, - 1000 mg/1}, and

     Q is the waste stream flow rate expressed  in  units  of  mVday,
     (m3,  a cubic meter, is equal to 264.2 U.S. gallons) and P is the
     nickel sulfate production rate expressed in units of kkg/day (kkg
     is 1000 kg, a metric ton, which is equal to 2205 pounds).
The average values are based on  data  from  those  plants  where
particular pollutant was found at detectable concentrations.
                                               the
In Table 22-5, the toxic pollutant raw waste data are presented as the
average  daily  concentrations  and  the  unit  loading  found  at th ..•
                                584

-------
individual plants, with the exception of  Plant  #572  which  presents
only  the concentrations.  The overall averages are also shown and are
calculated only for Plants #369 and #120, because they represent total
composited wastewater from the entire NiS04 process, while Plant  #572
data are for one of several sources.

Based  on  the  total  annual production rate of this subcategory (See
Table 22-1)  and the average waste load generated per unit product, the
estimated total pollutant raw waste loads generated each year for this
subcategory are as follows:
              Pollutant

              Antimony
              Arsenic
              Cadmium
              Chromium
              Copper
              Lead
              Nickel
              Selenium
              Zinc
Waste Load (kg/year

         1 .27
         0.22
         0.018
         1 .72
        95.2
         0. 19
       343
      <  0.17
         0.48
Mercury is not included in this list as it was found at  a  detectable
concentration  only  in  the one stream at Plant #572.  This stream is
shared with another nickel compound  process  using  different  source
materials.  Cross contamination is suspected.  Since mercury was found
below  detectable  concentrations  in  all  other nickel sulfate waste
water, using Plant #572 would yield an erroneously high average.  This
would subsequently show an unrepresentative yearly waste load   in  the
previous  table.    Based  on  the  reliable  data,  mercury  is  not a
pollutant of concern in the nickel sulf.ate industry.

Pollution Abatement Options

Toxic Pollutants of Concern

The toxic pollutants present in a nickel  sulfate  process  wastewater
depend  upon  the  purity  of  the  sources  and the nature of  the raw
materials being used, which vary with time.

If impure raw materials are used, most of the heavy  metal  impurities
will  be  removed in the purification process, handled and disposed of
as solid sludge.   These impurities build up in the mother  liquor  and
subsequently appear in purges, leaks, and washdowns.  The toxic metals
such  as  nickel  and copper, and to a lesser extent antimony, arsenic,
cadmium, chromium, lead, selenium, and zinc found in  the  wastewaters
during  sampling  originate  in  the  raw  material  source.  Pure raw
materials make complete recycle possible, allowing plants using  these
materials to comply with the effluent limitations without operation of
a treatment system provided the manufacturing process used is properly
designed,  operated, and maintained.
                                 685

-------
TABLE 22-5.  TOXIC POLLOTBtfr
                                                WASTE DATA
           :   Nickel Suifatft
         Average Daily Pollutant Concentrations and Loadings at
                            Plants Sampled Cl)

                                 (nn/1)
                         (KKg Of
Pollutant
Antimony
Arsenic
Cadmium
Chromium
Copper
Lead
#369 IS)
0.48
0.00020
<±>
0.009
0.0000038
1.3
0.00054
73.3
0.030
0.055
0.000022
4120 (V)
* ^- : ,
0.049
0.000035
0.0027
0.0000019
0.012
0. 00 00086
0.22
0.00016
0.052
0.000038
.,*s$%[ j
0.018
,•' *
0.16
*
0-04
0.097
.';/ Overall1^
Average
0.48
0.00020
0.049
0.0000035
0.0058
0.0000028
0.66
0.00027
36.8
0.015
0.054
0.000030
Mercury
Nickel
Selenium
Thai linn
Zinc
*
176
0.073
< 0.010
< 0.0000041
0.021
0.0000088
0,27
O.OOOU
*
49.2
0.035
0.069
0.00005
*
0.055
0.00004
0.01
1100
0.009
*
0-38
*
112
0.054
< 0.04
< 0.000027
0.021
0.0000088
0.16
0.000075
(S)  - Screening data from one grab  composite sample of  die batch process
      ccnbined raw waste streams.
(V)  - Verification data frcci three  24-hour ocnpoaite  aanples, averaged, from
      each raw waste sampling point,
 *  - Concentration  below significant level.
(1)  - The methodology of the sanpling program is described in Section 5.1,2,
      and Section 22.3.3 presents tin scope of sanpling in the 'Nickel Sulfate
      industry.
(2)  - Data for Plant #572 is presented in concentration basis only.
(3)  - Average of Plants #369 and #120 only.
 +  - Vtven averaging values indicated as "less than"  (<}, the absolute value
      •was used, and the resulting average waa indicated,  as a  "less  than" valve.
                                686

-------
Wastewater quality for an air pollution control scrubber at Plant #572
is  shown in Table 22-5.  However, this source is not used to evaluate
raw waste data, since it is only one of several sources and  does  not
represent  a  total  wastewater  stream.  This scrubber also serves in
another nickel compound manufacturing process alternately with  nickel
sulfate  so  it  cannot  be  considered  totally representative of the
process of interest.
No toxic organic pollutants were found in
streams at significant concentrations.
the  process-related  waste
Process Modifications and Technology Transfer Options

Mechanical scrapers should be installed on filters at plants which use
impure  raw  materials.   This would eliminate the backwash and reduce
the amount of wastewater produced.  Solids would need to  be  disposed
of  in  a  secure  landfill.  Installation of the scrapers would incur
only a small capital cost.

Best Management Practices

The best technology for the treatment  of  wastewater  from  processes
using  pure  raw  materials  is  recycle  of  all  process waters.  To
implement this treatment, recycle piping and pumping are needed.

The best technology available where  nickel  sulfate  is  manufactured
from  impure plating solutions is caustic soda addition to precipitate
nickel and other metallic hydroxides, followed by sand  filtration  to
remove  the  suspended  solids.   This  requires  installing treatment
tanks, filters, pH control equipment, and related piping and pumps.

Prevailing Control and Treatment Practices

Plant #369 sends filter leaks and wash water  to  a  collection  tank.
When  the batch manufacturing process is complete, the collected waste
is treated with caustic soda to pH 12.5,  The metals are  precipitated
as  hydroxides,  settled,  and  the  sludge disposed of at an approved
landfill.  The supernatant is sampled and analyzed before discharge to
a POTW.

Plant #120 wastewaters are generated  from  .leaks,  washdowns,  filter
wash, and air scrubbers.  These are combined with other nickel process
wastes and treated with caustic soda to precipitate trace metals.  The
waste is then treated by filtration followed by pH adjustment prior to
final discharge.

Plant  #572  also  combines  wastes from the air scrubbers, leaks, and
washdowns.  These wastewaters are sent to an equalization  tank  where
they  are  mixed  -with  alkaline  wastes to raise the pH to 10.  After
settling, the wastewaters are discharged to a POTW.

Plant #069, which produces a reagent  grade  product, -sends  periodic
purges  and  washdown  water  to  a  combined  collection  system with
                                  687

-------
wastewater  from  numerous  other  products.   Treatment  consists  of
neutralization  and equalization of the wastes prior to discharge to a
POTW.

Plant  #313  also  combines  its  wastewaters  from   nickel   sulfate
production   with  wastes  from  various  other  metal  processes  and
presently discharges the combined waste after a period of settling  in
a  pond.   A  treatment  system  is  being  designed  which  uses lime
precipitation at pH 10 followed by gravity separation.  Centrifugation
is to be used to thicken the sludge.  The  clarified  wastewater  will
then be neutralized to pH 6.5 - 7.5 and discharged.
Plant  #603
process.
has  no  discharge of wastewaters from the nickel sulfate
Advanced Treatment Technologies

Alkaline precipitation will remove nickel and most other heavy  metals
from  solution, allowing them to be settled and filtered in successive
steps.  Nickel and the common heavy metals (except chromium) can  also
be precipitated as metallic sulfides, for later separation by settling
and   filtration.    Sulfide   precipitation  generally  yields  lower
concentrations of the metals in the final effluent.

Selection o|_ Appropriate Technology and Equipment

Technologies for Different Treatment Levels

A.   Level 1 (BPT, BAT, NSPS)

     Level 1 consists of alkaline  precipitation  with  caustic  soda,
     followed  by  filtration.   The  system  is  operated  as a batch
     process.  This technology is generally the treatment practice  in
     place  within  this  industry.  For evaluating Level 1 treatment,
     the Agency has developed  two  models—one  for  high  production
     plants  (7,000  kkg/year), and another for low (900 kkg/year) and
     medium production (4,000 kkg/year) plants—the  difference  being
     in  the method of filtration.  Flow diagrams for these models are
     shown in Figures 22-5 and 22-6.

B.   Level 2

     In proposed Level 2 treatment, the  Level  l  system  was  to  be
     supplemented by the additional step of sulfide precipitation with
     ferrous  sulfide  after  alkaline  precipitation.   However, this
     technology was not employed, because  Level  1  affords  adequate
     control and the additional pollutant reduction was not sufficient
     to   offset  the  additional  cost.   Proposed  models  for  this
     technology  are  shown  in  Figures  22-7  and   22-8.    Further
     information  on  estimated  treatment  system operation costs and
     effectiveness may be found in the proposed  Development  Document
     (60).
                                 688

-------
Equipment for Different Production Levels

A.   Equipment Functions

     In both Level 1  models, wastes are received in a one-day  holding
     tank or wastewater collection sump which is drained each day to a
     reaction  vessel.   At the end of a normal work week, the contents
     of the reaction vessel are raised to about  pH  10  with  caustic
     soda  and  thoroughly  mixed.   In the high production model, the
     contents of the reaction tank  are  allowed  to  settle  and  the
     supernatant  is  filtered  through a dual-media filter, while the
     precipitates  are  filtered  through  a  high  pressure   filter.
     Filtrate  from the dual-media filter is adjusted to a pH from 6-9
     and is discharged as a final effluent.   Filtrate  from  the  high
     pressure filter is recycled to the holding tank.

     In  the  low and medium production models, the entire contents of
     the reaction tank are sent through the high pressure filter after
     mixing.  The filtrate is adjusted for pH  and  discharged,  while
     sludge is sent to a landfill.  Filter backwash is recycled to the
     holding tank.

B.   Chemicals and Handling

     Caustic soda in solution form is used for alkaline  precipitation
     to  form  insoluble  metallic  hydroxides.  The choice of caustic
     soda avoids precipitating calcium sulfate, as  would  occur  with
     lime   application.    Caustic   soda   solution  is  handled  in
     conventional equipment, or is  drawn  in  batches  from  shipping
     containers when small volumes are needed.

C.   Separation and Removal of Solids

     In the  low  and  midrange  production  models  at  both  levels,
     essentially  all solids are collected in a filter press, which is
     cleaned periodically.   The  dewatered  sludge  is  hauled  to  a
     chemical  landfill.   In  the  larger  model plant, backwash from
     cleaning the dual-media filter returns to  the  influent  holding
     tank,   from which the suspended solids pass via the reaction tank
     to the sludge filter press.

D.   Monitoring Requirements

     Satisfactory  separation  of  heavy  metals  can  be  assured  by
     maintaining  the  proper  reaction  pH,  which  can be determined
     manually on  each  batch,  using  simple  field  equipment.   For
     reporting purposes, occasional monitoring of nickel and copper in
     the effluent should be done by atomic absorption methods.
                                    689

-------
oajsnc
 SOCft
         OXBING
          TANK
 RAWVftSTE W.TER
                  u
                                         53WK
1
                                              SLUCK2
                                                                     AH)
                                                                                    ECftL
                                                                                   MEDIA
                                                                                   FILTER
                                                                                  FILTES PRESS
                            D
                                          Incltrles flOM mwiitoring, j« nonitoring,
                                          and sanpler.
                                                                                    IWCFILL
                    Figure 22-5.  Level 1 waste water  treatment for the nickel  sulfate subcategory —
                                    high production model — batch process.

-------
CAUSTIC
 SODA
  RAH
WASTE WATER
                                        REACTION
                                         TANK
                                                               FILTER
                                                     •fi*-*
O4-
                                                                  SLUDGE:
                                                                  HOLDING
                                                                   TANK
                                            BACKWASH
                                                                                        IANDPILL
                                                                                                        ADJUSTMBir




TER


1
1
1 <
•
PRESS


t
®
\
1
1
1

*

fTT tH V
-------
       CAUSTIC 9COV
RAH
 wist
vO
K>
             rcumc
              •BWK
                     r
                                                        BKKJASH
                         FEDBCOS
                         SUUXEE
  SODIUM
EtSOffTEE
                                           REACTHM
                                             TANK
                                                          "fl*
                             SOPEPHKTMTT
                                                                   FJlTHt MD
                                                                   HMHMGT3UK

                                                                                                             •D
                                                                                                   KQSSS
                                                      H-1

                                                          I

                                                        J
                                                                                                    * Inclutfes flow nemitorinj, f« nonitoring,
                                                                                                      and sanpler.
                                                                                        UVMFUl,
                   Figure 22-7.   Level 2 waste water treatment for nickel sulfate subcategory — high production
                                    model — batch process.

-------
     RAW WASTE
      WATER
to
                                                                                                       MXTUSTMENT
                                                BACKHftSH
                                                                                                               o
                                                                                                                    EFFLUENT
                                                                                          IflNtFILL
           LEGEND

Includes Elow monitoring, pH monitoring
and sampler.
                        Figure 22-8.   Level 2 waste water  treatanent for nickel sulfate —  low and medium
                                        production models — batch process.

-------
Treatment Cost Estimates

General Discussion

To  prepare  treatment  cost  estimates,  a  model  plant  concept was
developed for Level 1  technology as follows:

A.   Wastewater Flow

     Table 22-3 shows the wastewater discharged to treatment for  five
     plants.  The unit wastewater flow for the two single waste source
     plants  ranged from 0.42 mVkkg of NiS04 to 0.72 mVkkg of NiS04.
     For the model plant cost estimates, a production-weighted average
     of  0.68  mVkkg  for  the  two  plants  was  used.    This   was
     accomplished  by  multiplying  the unit flow of each plant by its
     daily production, adding the resultant values and dividing by the
     total production of the two plants, which results in these values
     being representative of the different production level plants.

B.   Production

     Nickel sulfate production ranges from 45 kkg/yr to  5,900  kkg/yr
     in  the  plants  for which the 308-Questionnaires were available.
     The average production for these six plants was 2,100 kkg/yr, the
     median  was  1,600  kkg/yr.   For   wastewater   treatment   cost
     estimates, three production levels were selected as model plants.
     These  are  900 kkg/yr, 4,000 kkg/yr, and 7,000 kkg/yr.  The mode
     of operation at all nickel sulfate plants is  the  batch  process
     and,  for  the  model  plant,  is  assumed  to  operate  for  250
     days/year.

C.   Solid Waste Generation

     Solid wastes are generated from the filtration  and  settling  of
     metals  from  the  nickel  sulfate  solution.   The solids can be
     recycled to the process for reuse when  pure  raw  materials  are
     used.  If the solids are not recycled, they are disposed of in an
     industrial  landfill.   The  quantity of solids generated is 0.39
     kg/kkg of nickel sulfate.

D.   Treatment Chemicals

     Caustic is required for neutralization to precipitate the  metals
     as  their  hydroxides.   Acid  is needed for pH adjustment before
     final discharge.   For  the  model  plant,  these  practices  were
     estimated to use 0.016 kg/kkg and 0.00010 kg/kkg, respectively.

Model Plant Control Costs

The cost estimates for three models having different production levels
are presented  in Tables 22-6, 22-7, and 22-8.
                                 694

-------
Table  22-9  gives  a  summary  of  the unit cost distribution between
amortization and operation and maintenance cost components at  various
production rates and levels of treatment.

Cost  estimates  developed for the first level of treatment (BPT, BAT,
NSPS) indicate that  at  low  production  levels,  labor  cost  has  a
significant  impact on the total annual costs.  At a medium production
level,  amortization,  operation,  and  maintenance  costs   are   the
important  factors  in  the annual costs.  At a high production level,
amortization cost is the significant factor in the annual costs.

Basis for Regulations

Evaluation of BPT Treatment Practices

Nickel sulfate can be  manufactured  using  pure  nickel  as  the  raw
material or an impure nickel raw material.  Waste loads emanating from
the  two sources differ in that total recycle of process wastes can be
accomplished at plants using a pure nickel  source,  while  at  plants
using  an  impure  raw  material,  waste  streams  need  to  be purged
periodically to avoid build-up of contaminants in the process.

A.   Pollutant Removal with BPT Treatment

     BPT technology for nickel sulfate  plants  utilizing  impure  raw
     materials  is  equivalent  to  treatment  Level  1.   Table 22-10
     presents the toxic  pollutant  treated  effluent  data  for  both
     Plants  #369 and #120 in a similar manner as Table 22-5 presented
     the raw waste data.  In evaluating BPT treatment  the  data  from
     Plant  #120  was used, rather than Plant #369, or overall average
     data.  This is because the treatment at Plant #120  represents  a
     typical  BPT  system,  while  Plant #369 has no filtration before
     discharge to a POTW.   Long-term  effluent  monitoring  data  for
     Plant  #120 can be found in Tables A-15a through A-15d.  The data
     is for nickel only and is presented in  concentration  and  daily
     loading units for both daily and monthly measurements.

     In  comparing raw waste and effluent data (Tables 22-4, 22-5, and
     22-10), BPT treatment gave a suspended solids removal of over  93
     percent,  while  the  toxic  metals nickel and copper had over 98
     percent removal.  All of the toxic pollutant  concentration  were
     below   the   lower   limit   of   treatability-based  achievable
     concentrations.  Many of the toxic metals from  the  effluent  of
     Plant  #369  were  below  BPT  based achievable levels, with only
     hydroxide precipitation and settling.

Basis for BPT Effluent Limitations

BPT regulations for the Nickel Sulfate Subcategory  are  presently  in
effect, 40 CFR 415.472  (See Table 22-2).  The technology basis for the
existing  BPT is alkaline precipitation plus dual-media filtration and
final pH adjustment before discharge.  Most direct dischargers in this
subcategory have installed BPT technology or equivalent.
                                  695

-------
     TABLE 22-6.  MODEL  PLANT TREATMENT  COSTS
Subcategory
Production
Nickel Sulfate
 900 metric tons per year
A. INVESTMENT COST

     Site development  	
     Equipment  	
     Monitoring equipment  	

          Subtotal  	
     Contractor's 0 & P°	

          Subtotal  	
     Engineering  	

          Subtotal	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL INVESTMENT COST 	


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision 	
     Energy 	
     Chemicals  	
     Maintenance  	
     Taxes and insurance 	
     Residual waste disposal  ....
     Monitoring, analysis
      and reporting 	

     TOTAL OPERATION AND
     MAINTENANCE COST
                                                     (S)
                               BPT

                             1,200
                            35,000
                            20,000

                            56,200
                             8,430

                            64,630
                            12,926

                            77,556
                             7,756

                            85,312
                             1,200

                            86,512
                             8,000
                                30
                               200
                             8,531
                             2,595
                               100

                             5,000


                            24,457
BATa

  0
  0
  0

  0
  0

  0
  0

  0
  0

  0
  0
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                            13,880

                            38,337
a Represents the incremental cost above that for BPT treatment
  Overhead and Profit
                             696

-------
     TABLE 22-7.  MODEL  PLANT TREATMENT  COSTS
Subcategory
Production
Nickel Sulfate
4,000 metric tons per year
A. INVESTMENT COST

     Site development  	
     Equipment  ..	
     Monitoring  equipment  	

          Subtotal  	
     Contractor's 0 &  Pb	

          Subtotal  	
     Engineering 	

          Subtotal	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL INVESTMENT  COST  	


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  ....
     Energy 	
     Chemicals  	
     Maintenance 	
     Taxes and insurance 	
     Residual waste disposal  ..
     Monitoring, analysis
      and reporting 	
                                                   {$)
                               BPT

                             1,800
                            70,500
                            20,000

                            92,300
                            13,845

                           106,145
                            21,229

                           127,374
                            12,737

                           140,111
                             1,800

                           141,911
     TOTAL OPERATION AND
     MAINTENANCE COST
                             8,000
                                40
                               900
                            14,011
                             4,257
                               100

                             5,000
                            32,308
BAT

  0
  0
  0

  0
  0

  0
  0

  0
  0

  0
  0
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                            22,796

                            55,105
  Represents the incremental cost above that  for  BPT  treatment
  Overhead and Profit
                             697

-------
    TABLE 22-8.  MODEL PLANT TREATMENT COSTS
Subcategory
Production
Nickel Sulfate
7,000 metric tons per year
A. INVESTMENT COST

     Site development 	
     Equipment 	
     Monitoring equipment  	

          Subtotal 	
     Contractor's 0 & Pb.........

          Subtotal 	
     Engineering	

          Subtotal 	
     Contingencies 	

          Subtotal	
     Land	

     TOTAL INVESTMENT COST 	


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision 	
     Energy 	
     Chemicals 	
     Maintenance  	
     Taxes and insurance  	
     Residual waste disposal  ....
     Monitoring, analysis
      and reporting	

     TOTAL OPERATION AND
     MAINTENANCE COST
                                                     ($)
                               BPT

                             3,000
                           106,500
                            20,000

                           129,500
                            19,425

                           148,925
                            29,785

                           178,710
                            17,871

                           196,581
                             3,000

                           199,581
                             8,000
                                50
                             1,500
                            19,658
                             5,987
                                200

                             5,000
                            40,396
BATa

  0
  0
  0

  0
  0

  0
  0

  0
  0

  0
  0
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                            31,984

                            72,379
a Represents the incremental cost above that  for  BPT  treatment
  Overhead and Profit
                              698

-------
        TABLE 22-9 MODEL  PLANT  UNIT  TREATMENT COSTS
     Subcategory Nickel  Sulfate
    COST ITEM
PRODUCTION
 (kkg/yr)
Annual Treatment Costs  ($/kkg)


     LEVEL OF TREATMENT

   BPT       BAT*      NSPS
Annual Operation
and Maintenance


Annual
Amortization


Total Annual
Cost



900
4,000
7,000

900
4,000
7,000

900
4,000
7,000

27.17
8.08
5.77

15.42
5.70
4.57

42.60
13.78
10.34

NA
NA
NA

NA
NA
NA

NA
NA
NA

NA
NA
NA

NA
NA
NA

NA
NA
NA
*Represents the incremental cost  above  BPT
                             699

-------
A.   Conventional Parameters

     For pH the treated effluent is to be controlled within the  range
     of 6.0 to 9.0.  This limitation is based on the data presented in
     Appendix  B of the proposed Development Document (60) and the JRB
     Study (52).

     In the existing BPT regulations, EPA  has  different  limitations
     for  pure  and  impure  raw  materials  processes.    EPA  is  not
     establishing different limits for these  processes  in  the  BPT,
     BAT, BCT, NSPS, PSES, and PSNS regulations.  This is because both
     processes  are adequately covered by the one regulation since the
     pure raw  materials  process  can,  with  proper  design,  comply
     without  end  of the pipe treatment.  Only nickel,  TSS and pH are
     regulated because these are the only three parameters limited  in
     the existing BPT regulation.

Basis for BCT Effluent Limitations

While   EPA  has  not  yet  proposed  or  promulgated  a  revised  BCT
methodology in  response  to  the  American  Paper  Institute  v.  EPA
decision  mentioned  in Section 3, EPA is promulgating BCT limitations
for this subcategory.  These limits are identical to  those  for  BPT.
EPA  is  not promulgating any more stringent limitations since we have
identified  no  technology  option  which  would  remove   significant
additional  amounts of conventional pollutants.  As BPT is the minimal
level of control required by law, no possible application of  the  BCT
cost   tests   could  result  in  BCT  limitations  lower  than  those
promulgated in this regulation.  Accordingly, there is no need to wait
until  EPA  revises  the  BCT  methodology  before  promulgating   BCT
limitations.

Basis for BAT Effluent Limitations

A.   Technology Basis

     For BAT, the Agency  is  establishing  limitations  based  on  BPT
     technology which is alkaline precipitation followed by dual-media
     filtration.   The  Agency  considered  treatment Level 2 (sulfide
     precipitation), but rejected it  because  the  treatment  removed
     only   small   additional   amounts   of  toxic  metals  in  this
     subcategory.

B.   Flow Basis

     The model plant BAT treatment system is based on an  inflow  rate
     of  0.68  mVkkg for effluent limitation purposes.  The rationale
     for the flow  is the same as that used for  the  model  plant  for
     cost estimating as described in this section.
                                   700

-------
            TABLE 22-10.  TOXIC POLLUTANT TREATED EFFLUENT DATA
StBCATEGQRY:  Nickel Sulfate
Average Daily Pollutant Concentrations and Loadings at Plants Sampled

                                  (mg/1)
                                                                     (1)
Pollutant
             #369(S)
(kkg of NiS04.7H20)

             #120 (V)
Overall
Average
                                                                   (2)
Antimony

Arsenic

Cadmium

Chromium

Copper

Lead

Nickel

Selenium

Thallium

Zinc
               0.2
               0.000083

               0.26
               0.00011

              s 0.001
              : 0.00000042

               0.45
               0.00019

              18.0
               0.0075
               0.001
               0.00000042

               1.4
               0.00058

               0.012
               0.000005

               0.029
               0.000012
               0.17
               0.000071
            <0.010
            <0.0000072

            0.00013
            0.000000094

            0.057
            0.000041

            <0.043
            <0.000031
            0.003
            0.0000022

            0.20
            0.00014

            <0.008
            <0.0000058
            0.00033
            0.00000024

            0.058
            0.000042
  0.2
  0.000083
  0.13
  0.000059
  0.00056
  0.00000026
  0.25
  0.00012
  9.02
  0.0038
  0.002
  0.0000013
  0.8
  0.00036
  0.01
  0.0000054

  0.015
  0.0000061
  0.11
  0.000056
(S)
(V)
(1)
(2) -
 *  _
Screening data from one grab composite sample of treated effluent.
Verification data from three 24-hour composite samples, averaged,
The effluent data presented here corresponds to the raw waste data shown
in Table 12-5  excluding Plant #572.  The methodology of the sampling pro-
gram is described in Section 5.1.2, and the scope of sampling in the
Nickel Sulfate industry is described in Section 22.3.3.
When averaging values indicated as "less than" (<), the absolute value
was used and the resulting average was indicated as a "less than" value.
Concentration below significant level.
                                    701

-------
C.   Selection Basis for Pollutants to be Regulated

     The  selection  of  pollutants  for  which   numerical   effluent
     limitations were proposed was based on an evaluation of raw waste
     data  from  the screening and verification sampling program.  The
     two major factors considered were:  1)  individual  sampling  raw
     waste  concentrations,  and  2)  the  total subcategory raw waste
     loadings.

     1.   Raw Waste Pollutant Concentrations

          A tabular summary of maximum raw waste concentrations  found
          in  sampling  is  presented  above.   Data  from  the plants
          sampled  in  screening  and  verification  are   shown   for
          comparison.   For  each pollutant, the maximum concentration
          observed gave a  preliminary  indication  of  its  potential
          significance   in  the  subcategory.   On  this  basis,  the
          preliminary selection of nickel, copper, chromium, and to  a
          lesser  extent,  lead,  antimony,  and zinc were included as
          candidates for regulation.  These pollutants  were  observed
          at  least once during screening at concentrations considered
          treatable in  this  industry  using  one  of  the  available
          treatment  ^technology  options.   The  other  toxic  metals
          (mercury, cadmium, and selenium) were found to have  maximum
          concentrations  that  were  lower  than  the  minimum levels
          achievable by treatment.

     2.   Total Subcategory Raw Waste Pollutant Loadings

          Pollutant raw waste loading data were used to  evaluate  the
          overall magnitude of the pollution potential.  Data from the
          plants   sampled   are   summarized  in  Table  22-5.   This
          information, coupled with the estimated total nickel sulfate
          production rate of  6,350  kkg/year  found  in  Table  22-1,
          yielded the approximate total annual pollutant loading rates
          for  the  subcategory shown in this section.  This method of
          ranking the pollution potential of the observed toxic metals
          confirms  the  maximum   concentration-based   ranking   and
          indicated that nickel, copper, chromium, antimony, zinc, and
          lead  were  the  six  dominant toxic metals in terms of both
          total mass loading and treatable raw waste concentrations.

     3.   Final Selection of Pollutants to be Regulated

          Originally,  limitations   were   proposed   for   all   the
          aforementioned  dominant  toxic  pollutants.  The Agency has
          decided, however, to promulgate regulations  on  copper  and
          nickel  only.   Copper  and  nickel  are  by  far  the  most
          prominant toxic metals in  this  subcategory's  waste.   The
          nature of the treatment technology employed in this  industry
          is  such  that  control  of  certain  key parameters ensures
          control of all the toxic metals of concern.  Elaboration  on
          this topic may be found in Section 8 of this document.
                                    702

-------
          Section  8  (See Table 8-14) indicates that toxic metals can
          be divided into two groups for pH optimization  of  alkaline
          precipitation, and that control of one or more of the metals
          in  each  of  these  groups will ensure control of others in
          that group.  In controlling copper and nickel, both of these
          groups are covered.  Therefore, since these two  metals  are
          the  major  toxic  pollutants  in  the raw waste, control of
          these  parameters  will  ensure  control   of   all   metals
          previously considered for limitations.  This is supported by
          sampling data from Plant #120.

          Optimization  of  treatment conditions for nickel and copper
          removal may cause less  efficient  performance  in  chromium
          control,  but  in  light  of  the  relatively  low  incoming
          chromium concentrations, control should be adequate.    Table
          8-14  and Figure 7-1  indicate a pH in the neighborhood of 10
          would  be  optimum  for  control  of  pollutants   in   this
          subcategory.

D.   Basis of Pollutant Limitations

     1.    Toxic Pollutants

          The effluent limitations for the  selected  toxic  pollutant
          control  parameters  are  derived  from  estimated  industry
          achievable   long-term   averages     (Table    8-13),    and
          literature-based  treatability estimates {Table 8-11).  This
          is because plant performance data  from  sampling  at  Plant
          #120  (Table  22-10)   show effluent concentrations below the
          lower limit of treatability estimates for all of  the  toxic
          metals except nickel.

          BAT  effluent limitations for copper and nickel are shown in
          Table 22-11 along with  corresponding  concentration  bases.
          Also indicated are recommended concentration bases for other
          toxic  metals  detected  at  treatable  levels.   Those  are
          included for those cases where control of one or more of the
          unregulated metals is deemed necessary.

          Concentration bases for copper,  nickel,  and  chromium  are
          derived  from estimated achievable long-term averages (Table
          8-13), while those for the  other  metals  of  concern  were
          derived from the lowest applicable treatability level (Table
          8-11).   This  approach results in the setting of achievable
          limitations, and provides for  variations  in  the  influent
          quality  that  may  be  associated  with different nickel or
          nickel solution impurity levels or other process variables.

          The basis for  proposed  BAT  limitations  on  each  of  the
          selected  metals  is  given  below.  Calculations of maximum
          30-day  average  and  daily  maximum   concentrations   from
          long-term  averages  are based on variability factors of 1.2
                                    703

-------
and 3.6, respectively.   This  gives  a  variability  factor
ratio of:

3.6 = 3,0
1 .2

These values were determined by statistical analysis of data
presented  in the Treatability Study (61) for the nickel and
copper sulfate subcategories.

a.    Nickel

     Table 8-13 indicates an estimated achievable  long-term
     average  for  nickel of 0.30 mg/1 which is supported by
     the Treatability Study  (61).   This  reflects  industry
     performance  using BAT technology and is above effluent
     levels  indicated  during  sampling  at   Plant   #120.
     Therefore,  using this value and the 30-day variability
     factor of 1.2 the 30-day concentration basis becomes:
                                        the  daily
maximum
(0.30 mg/1) (1.2) = 0.36 mg/1

Multiplying  by  a  VFR  of  3.0,
concentration is:

(0.36 mg/1) (3.0) = 1.1 mg/1

The  effluent  limitations  for  nickel  are determined
using a model flow  of  0.68  mVkkg  NiS04,  with  the
maximum   30-day   average  limitations  calculated  as
follows:

(0.36 mg/1) (0.68 mVkkg)   (kg/m*)  = 0.00024 kg/kkg
                         (1000 mg/1)
     The 24-hour maximum limitation is:

     (3.6H0.3 mg/1) (0.68 mVkkg)   (kg/m*)   = 0.00074 kg/kkg
                                  (1000 mg/1)

     These values are considerably lower than the BPT limits
     presented  in  Table  22-2.   This   is   because   the
     information  on  treatment  system performance based on
     information collected to support this regulation showed
     better performance than that expected when the existing
     BPT regulation was developed.
b.    Copper
     Because the effluent concentration of  copper  observed
     at  Plant  #120  was  below treatability estimates, the
     estimated long-term average (Table 8-13) of  0.30  mg/1
                           704

-------
                     TABLE 22-11.  EFFLUENT LIMITATIONS
                               Nickel Sulfate    ,..,
                       Best Available Technology
                   Waste Water Flow:  0.68 m3/kkg of NiSO.

Concentration

Pollutant



Antimony
Cadmiun
Chromium
Copper
Lead
Nickel
Seleniun
Zinc
Subcategory
Performance
(mg/1)


0.40(3)
0.05(3)
0.20(4)
0.30(4)
0.050(3)
0.30(6)
0.10C3)
0.20(4)
(?)
VFR^'



3.6/1.2
3.6/1.2
3.6/1.2
3.6/1.2
3.6/1.2
3.6/1.2
3.6/1.2
3.6/1.2
Basis


Max
30-day
Avg
0.48
0.060
0.24
0.36
0.060
0.36
0.12
0.24
(mg/D



24-hr
Max
1.4
0.18
0.72
1.1
0.18
1.1
0.36
0.72
Effluent Limit
(kg/kkg


Max
30-day
Avg
_J5)
— (5)
_J5)
0.00024
— <5>
0.00024
_<5>
_<5>
of NiS04)



24-hr
Max
_J5>
__(5)
__(5>
0.00074
_J5)
0.00074
-J5)
_J5>
(1)   Also applies to NSPS.

(2)   Variability factor ratio:   Ratio of the daily variability factor to monthly
     based on statistical analyses of treatability study data for nickel and
     copper sulfate (61).

(3)   Lowsr limit of treatability (Table 8-11)  used as long-term average.

(4)   Industry long-term average (Table 8-13).

(5)   Indicates no effluent  limitation established.

(6)   Treatability study estimate.
                                     705

-------
was  chosen  to  determine  the concentration bases for
copper.  Again, the 30-day average  variability  factor
of  1.2  and  the VFR of 3.0 apply-  Calculation of the
maximum 30-day average concentration basis is:

(0.30 mg/1) {1,2} = 0.36 mg/1

and the 24-hour maximum is:

(3.0)  (0.36 mg/1) =1.1 mg/1

The maximum 30-day average effluent limitation based on
a 0.68 mVkkg flow is:

{0.36 mg/1) (0.68 mVkkg)   (kq/m3)   = 0.00024 kg/kkg
                         (1000 mg/1)

and the 24-hour maximum is:

(3.6X0.30 mg/1 X0.68 mVkkg)   (kq/m3)   = 0.000074 kg/kkg
                               1000 mg/1)
Chromium
Effluent concentration for chromium at Plant  #120  was
near  the  lower  limit  of  treatability,  however, to
account  for  variations   in  raw  waste  loading   the
long-term  average  estimate (Table 8-13) is used.  The
30-day average variability factor of 1.2 and VFR of 3.0
apply.  The maximum 30Lday average concentration  basis
is:

(0.20 mg/1) (1.2) = 0.24 mg/1

and the 24-hour maximum is:

(3.0) (0.24 mg/1) =0.72 mg/1

Antimony

The lower limit of treatability  (Table  8-11)  of   0.40
mg/1  is  chosen  as  the  long-term average.  Effluent
concentrations at  Plant   #120   were  well  below   this
value.  The 30-day variability factor of  1.2 and VFR of
3.0 are used.  The maximum 30-day average concentration
basis is:

(0.40 mg/1). (1.2) = 0.48 mg/1

and^the 24-hour maximum is:

(3.0) (0.48 mg/1) =1.4 mg/1
                   706

-------
Lead

By  the  same  rationale  applied  to   antimony,   the
long-term average for lead is set to the lower limit of
treatability which is 0.050 mg/1.  The same variability
factors    apply.     The    maximum   30-day   average
concentrations basis is:

(0.050 mg/1) (1.2) * 0.60 mg/1

and the 24-hour maximum is:

(3.0) (0.060 mg/1) =0.18 mg/1

Zinc

Again, the lower limit of treatability  is  used  as  a
long-term  average  by  the  rationale mentioned above.
This value is 0.20 mg/1,  therefore,  by  applying  the
same  variability  factors,  the maximum 30-day average
concentration basis is:

(0.20 mg/1) (1.2) = 0.24 mg/1

and the 24-hour maximum is:

(3.0) {0.2.4 mg/1) =0.72 mg/1

Cadmium

Using the lower limit of treatability,  which  is  0.05
mg/1/  as  a  long-term  average  and applying the same
variability  factors,  the   maximum   30-day   average
concentration basis for cadmium is:

(0.05 mg/1) (1.2) = 0.060 mg/1

and the 24-hour maximum is:

(3.0) (0.060 mg/1) =0.18 mg/1

Selenium

The lower limit of treatability applies  for  selenium.
This  value  is  0.10  mg/1, therefore, by applying the
same variability factors, the  maximum  30-day  average
concentration basis is:

(0.10 mg/1) (1.2) = 0.12 mg/1

and the 24-hour maximum is:

(3.0 (0.12 mg/1) =0.36 mg/1
                       707

-------
New Source Performance Standards

After   examination   of   the  effectiveness  of  the  two  treatment
technologies  applicable  to  nickel  sulfate  wastes,  it  has   been
determined  that  BAT  technology be the basis for NSPS.  The effluent
limits for toxic metals are the same as for BAT shown in Table  22-11,
and  the  TSS  is  being  limited at the same effluent level as in the
existing BPT regulation presented in Table 22-2.  Also, pH is  limited
to  the  range  6 to 9 as based on data presented in Appendix B of the
proposed Development Document (60) and the JRB Study (52).

Basis for Pretreatment Standards

All five of the direct  dischargers  and  four  of  the  six  indirect
dischargers  in  the Nickel Sulfate Subcategory have BPT/BAT treatment
installed.

There is an existing PSES regulation, 40 CFR 415.474.  The  Agency  is
amending  that  section  of  these  regulations based on new treatment
system performance data and the PSES limitations are the same as those
presented  for  BAT  in  Table  22-11.   EPA  is  also  setting   PSNS
limitations  equal  to  the  BAT limits presented in Table 22-11.  The
pollutants limited by the final PSES and PSNS regulations  are  nickel
and  copper.   Pretreatment  is  necessary because BAT provides better
removal of toxic metals than is achieved by a well-operated POTW  with
secondary  treatment  installed,  and therefore these pollutants would
pass through a POTW in the absence of pretreatment.
                                708

-------
                              SECTION 23

                       SILVER NITRATE INDUSTRY
Summary of Determinations

Action on this subcategory has been deferred, and  a  new  subcategory
including  all  silver  compounds  will be reviewed under Phase II BAT
review, because the logical sequence of guideline promulgation was  to
start  the  guideline  process  with  nonferrous metals to be followed
later by a regulation on silver compounds.

Assessment of the Water Pollution Potential

Production Processes and Effluents

Most of the  silver  nitrate  produced  is  for  captive  use  in  the
photographic  industry.   It is also used in the manufacture of silver
salts, mirrors, for  silver  plating,  coloring  porcelain  and  as  a
chemical reagent.

The industry profile data is given in Table 23-1.

Toxic  pollutants  found  at significant levels during sampling at one
plant were:
          Pollutant

          Silver
          Cyanide
  Concentration (
Screening      Verification
164
580
65
470
Silver  was  not  found  at   a   significant   concentration   during
verification sampling of the same plant.  However, a significant level
of  cyanide  was  found  again.  The source of cyanide was found to be
from a soaking solution which is used to remove silver nitrate  stains
from  workers'  clothes.   The solution is sent to the silver recovery
treatment system.  When plant personnel  discontinued  this  practice,
cyanides disappeared from the effluent.

Status of Regulations

BPT  limitations for this subcategory (Subpart BA) were promulgated on
May 22, 1975, (40 FR 22421) and PSES  were  promulgated  on  July  20,
1977,  (42  FR 37301).  Both the BPT limitations and PSES are still in
effect (40 CFR Part 415.530, 415.531, and 415.534).
                                     709

-------
TABLE  23-1
SUBCKTBGORY PRCFILE :DAIA
SUBCATEGORy
SILVER NITRAOE
Tbtal subcategory capacity rate
Tbtal subcategory production rate
Number of plants in this subcategory
308 Data on file for
    With total capacity of
    With total production of
    Representing capacity
    Representing production
    Plant production range:
            Minimum
            Maxinum
    Average production
    Median production
    Average capacity utilization
    Plant age range:
            Minimum
            Maximum
    Waste water flow range:
            Minimum
            Maximum
    Volume per unit product:
            Minimum
            Maximum
                            35,000 kkg/year
                                 7
                                 2
                             6,507 kkg/year
                             3,256 kkg/year
                                 NA
                                 9 percent

                                50 kkg/year
                             3,206 kkg/year
                                NA
                                NA
                                NA

                                20 years
                                64 years

                                <1 cubic meters/day
                                38 cubic meters/day

                                 1 cubic meter/kkg
                                 4 cubic meters/kkg
Sources of data are Stanford Research Institute,  Directory of Chemical
Producers, U.S.A,, 1977, U.S. Department of Commerce, Current Industrial
Reports, December 1977; Energy and Environmental  Analysis, Inc.;. Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry," June 1978  and "Economic Analysis of Proposed
Revised Effluent Guidelines and Standards for the Inorganic Chonicals
Industry," March, 1980.
NA  =  Not Available
                                  710

-------
                              SECTION 24
                      SODIUM BISULFITE INDUSTRY
Industry Profile

General Description
Sodium bisulfite (NaHS03) is manufactured both in liquid and  powdered
form.   Captive  use  is  very small.  Sodium bisulfite is used in the
             of  photographic  chemicals,  textiles,   and   in   food
              It  is  also  used  in  the  tanning industry and in the
manufacture
processing
sulfite process for the manufacture of paper products.

The industry profile data are given in Table 24-1, while the status of
regulations prior to promulgation of this regulation is summarized  in
Table 24-2.

Subcategorization

The  method  of  primary  Subcategorization  chosen  for the inorganic
chemicals point source category was subdivision by  dominant  product.
Other factors taken into consideration for Subcategorization included:
raw  materials  used,   manufacturing  process  employed,  geographical
location, size and age of equipment and facility involved,  non-water-
quality  aspects  of  waste  characteristics,  water pollution control
technology, treatment  costs,  energy  requirements  and  solid  waste
disposal.   A detailed discussion of these factors is given in Section
4.  No further Subcategorization of  the  sodium  bisulfite  industry,
besides dominant product, was required.

General Process Description and Raw Materials

Sodium,, bisulfite  .is .produced by reacting sodium carbonate (soda ash)
with sulfur dioxide and water.  The reaction is:
NaC0
         2SO
                H20 = 2NaHS03 + C02
(1)
This reaction produces a slurry of sodium bisulfite crystals which can
be sold, but which is  usually  processed  to  form  anhydrous  sodium
metabisulfite.   This  requires  thickening, centrifuging, drying, and
packaging operations.

Water Use and Waste Source Characteristics

Water Use

Direct process contact water is used to slurry  the  sodium  carbonate
for  the  reaction.   Noncontact cooling water is another water use at
one plant.  Water  is  also  used  for  pump  seals,  maintenance  and
washdowns.   Table  24-3  gives a summary of water usage at the plants
for which 308-Questionnaires were available.
                             711

-------
               TABLE 24-1.  SUBCATEGORY PROFILE DATA SUMMARY
SUBCATEGQRY
Sodium Bisulfite
Total subcategory capacity rate
Total subcategory production rate
Number of plants in this subcategory
308 Date on file for
    With total capacity of
    With total production of
    Representing production
    Plant production range:
            Minimum
            Maximum
Average production
Median production
Average capacity utilization
Plant age range:
            Minimum
            Maximum
Waste water flow range:
            Miniinum
            Maximum
Volume per unit product:
            Minimum
            Maximum
                             98,000 kkg/year£D
                                  7
                                  2
                             46,000 kkg/year
                             28,300 kkg/year
                              4,700 kkg/year
                             23,600 kkg/year
                             17,800 kkg/year
                             16,900 kkg/year
                                 62 percent

                                  4 years
                                 19 years

                                  3 cubic meters/day
                                100 cubic meters/day

                                < 1 cubic meters/kkg
                                < 1 cubic meters/kkg
Sources of data are Stanford .Research Institute, Directory of Chemical
Producers, U.S.A., 1977, U.S. Department of ccmnerce, Current Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry."
(1)"Energy and Environmental Analysis, Inc.;
    Economic Analysis of Proposed Revised Effluent
    Guidelines And Standards for the Inorganic
    Chemicals Industry," March 1980.
                                     712

-------
TABLE 24-2.
STATUS OF REGULATIONS - EFFLUENT LIMITATION GUIDELINES
SUBCATEGORY      Sodium Bisulfite

SUBPART          BB (40 CPR 415.540, 5/22/75)
                                 STANDARDS
                      BPCTCA

                 Max.      Avg.
Product   Para-  kg/kkg   kg/kkg
Process   meters (mg/1)     (mg/1)
                       BATEA

                     Max.  Avg.
                    kg/kkg kg/kkg
                     (mg/1) Cmg/1)
    NSPS

 Max.   Avg.
kg/kkg kg/kkg
 (mg/1) (mg/1)
Sodium    Reserved    Reserved
Bisulfite
                     Reserved
 Reserved
      = Maximum of any one day.
2Avg  = Average of daily values for thirty consecutive days  shall not exceed.
                                    713

-------
Waste Sources

Noncontact cooling water for the centrifuge is a source  of  waste  at
one  plant.   However, direct process contact water is the main source
of wastewater which  must  be  treated,   together  with  miscellaneous
wastes  such  as  water  used for maintenance purposes, washdowns, and
spill cleanup.

Table 24-4 summarizes the wastewater unit flows from the  major  waste
source  for  plants #987 and #282.  Plant #987 has two facilities that
produce sodium bisulfite which are designated A and B.

There is little solid waste generation in  the  production  of  sodium
bisulfite  and  process  waste  treatment.  There are minor quantities
which are precipitated as metal hydroxides resulting in  insignificant
amounts  of filter cake requiring disposal.  Generation of solid waste
is therefore assumed negligible.

Description of. Plants Visited and Sampled

Screening

Plant #282 was visited in the screening phase  of  the  program.   The
bisulfite  waste  is treated on a batch basis every two or three days.
Sodium hypochlorite is added to  the  waste  to  oxidize  the  sulfite
waste,  which is then mixed with wastes from an organic chemical plant
and neutralized.  The combined wastes are then discharged to a  sewer.
Table  24-5 shows the flow data and pollutant discharges, while Figure
24-1 gives the process flow diagram and shows the sampling points used
in screening.

Verification

In verification, two plants were visited, Plants #586  and  #987.   At
Plant  #586  the  sodium bisulfite wastes are combined with many other
process wastes and  all  treated  together.   Figure  24-2:  shows  the
flowsheet  and  the  points  sampled.   Table 24-6 gives the pollutant
emissions and flow data for the waste streams.  The filter wash is the
main process waste at Plant #987.   This  waste  is  neutralized  with
caustic  soda  to pH 9 - 10 to convert the bisulfite waste to sulfite.
The sulfite is then oxidized with air to sulfate.  The treated  waste,
including  solids,  is  discharged  to  a river.  Table 24-7 shows the
pollutant emissions and flow  data  for  the  waste  streams  sampled.
Figure  24-i  shows  the  process  flow diagram and sampling points at
Plant #987.

Toxic Pollutant Analytical Results

The following table is a tabulation, of the toxic pollutants identified
at detectable concentrations in the raw process waste during screening
and verification.  The  concentrations  presented  under  verification
represent  the  highest  observed  in  the  raw  process  waste durinp
                                7H

-------
TABLE 24-3,
WATER USAGE IN THE SODIUM BISULFITE SUBCATEGORY
Plant          Direct Contact Process  Noncontact Cooling  Maintenance
                       ,                     .,           Washdoims, etc.
                     (in/kkg)              (mVkkg)          (nT/kkg)
# 282
I 586
# 987
        0.14
          NA
        1.15
3.85
 NA
 0
1.00
 NA
0.38
NA = Not Available
                                    715

-------
           TABLE 24-4.  WASTE WATER FLCW AT PLANTS #987 AND #282
                        FOR SODIUM BISULFITE SUBCATEQOFK
SUBCATEGQRy
                               SODIUM BISULFITE
Source
                  Flow Rate Per Unit of Production (m/kkg)

Direct Process * '
Contact
Indirect process
Contact
Miscellaneous
Washdown
Total
#987A
0.018
1.50
0.31
1.83
#987B
0.018
1.17
0.42
1.61
#282
0.14
0.030
1.00<2)
1.17
      Average
                                      1.50
       Plant #987 contains tWD separate facilities labeled A and B for
       the purpose of comparison.
(1)


(2)


(3)  - tother liquor filter wash.
       Includes steam condensate which is currently treated prior to
       discharge.
                                     716

-------
TABLE 24-5.  FLCW AND POLLUTANT LOAD DATA OF THE SAMPLED WASTE STREAMS FOR
             PLANT #282 PRODUCING SODIUM BISULFITE1)
Waste Stream
  Flow
(m3/kkg)
   TSS
(kg/kkg)
   COD
(kg/kkg)
Untreated waste
Treated waste
  2.67
  2.67
  HUT •
  0.424
  4.04
  2.61
 (1) -  Data based on screening sampling
        which involves one 72 hour composite
        sample.

 (2) - Unable to determine.
                                     717

-------
          SUBLDCD
           SULFUR
H
00
                                                                                                 TO SEWER
                                    COOLING
                                     WSTE
                           Sampling points.
                     Figure 24-1.  General process flew diagram at plant  #282 showing the sampling points.
                                    Sodium bisulfite iranufacture.

-------
         SQDXIH
         Bifiuu
         PROCESS
              II WD 12
         IJBGBO


     Ibsta •traana sampled.
                           ATR
Of If MA
Figure 24-2.  General flow diagram at plant #586 showing the sampling points.
               Sodium bisulfite manufacture.

-------
TABLE 24-6.  FLOW AND POLLUTANT LOAD DATA OF THE SAMPLED WASTE STREAMS FOR
             PLANT #586

Stream
Number
1
2
3
4
5
6
7
8

Waste Stream
Description
MBS Sump #1
MBS Sump #2
Total loads (1,2)
Amine Oxidation Pond
2nSO4 Pond Effluent
Lime Treatment Effluent
Truck Washdown
SO- Wastes
Treated Effluent
Total loads (1,2,3,4,6,7)
Flow
9.68<2>
9.68<2)
19.4
2.77(35
78.5 <3)
110 (3)
0.134(3)
85.9 (3)
188 <4>
187
TSS
(kg/kkg)
0.19
0.051
0.24
2.4
12
11
0.012
2.0
4.3
17
COD
(kg/kkg)
1.1
0.46
1.6
2.3
0.76
29
0.098
53
22
57
 (1) - Data based on verification sampling
       which involves three 24 hour composite
       samples.

 (2) - Includes noncontact process water that
       does not contribute to the pollutant
       load.
 (3) - Raw process waste flews that are not
       directly related to the sodium bisulfite
       industry, but are currently treated
       in combination with raw process waste
       that is related.
 (4) - Treated effluent from contained treatment
       of a nuntoer of different raw process
       waste streams not all related to sodium
       sulfite production.
                                    720

-------
TABLE 24-7.  FLCW AND POLLUTANT LOAD DATA OF THE SAMPLED WASTE STREAMS FOR
             PLANT #987 (D

Stream
Number
1
2
3
4

5
6
Waste Stream
Description
No. 1 Filter Wash
Floor wash, spill,
No. 2 Filter Wash
Raw Process Waste
(Streams 1+2+3)
54 Hour Aeration
Treated Effluent
Flow
(m3/kkg)
0.055
etc. 0.013
0.041

0.11
0.14
0.14
TSS
(kg/kkg)
0.11
0.046
0.0052

0.32
0.33
0.0031
COD
(kg/kkg)
1.4
0.30
0.91

3.5
1.2
1.0
 (1) - Data based on verification sampling
       which involves three 24-hour composite
       samples.
                                    721

-------
                 TO A3JCSFHERE
ALKALINE SLURRY
                                                                                                111 AW) 13
                      DRAINS, DRIPS,
                    SPILLS, SftSHDCWS
OUTFALL TO RIVER
                                            12
                                                                                |6
                            LEGQJD
                         Waste streams sailed.
                                                               t    T
                                                              NaCH    AIR
              Figure  24-3.   General process flow diagram at plant  1987 shewing  the sampling points.
                               Sodium bisulfite manufacture.

-------
sampling.
levels.
           No  organic  toxic  pollutants  were  found  at  detectable
               Maximum Raw Waste Concentration Observed
Pollutant

Arsenic
Copper
Zinc
Cadmium
Chromium
Antimony
Lead
Mercury
Nickel
Silver
Thallium
                    Screening
                    Plant #282

                        12
                       380
                      2500
                         6
                         0
                        30
                         8
                         3
                       250
                         2
                         8
    Verification
Plant #586 and #987

          67
         930
        3600
          41
        3400
         650
        1100
          17
         460
          15
           8
Section  5  of  this report describes the methodology of the screening
and  verification  sampling  program.    In   the   Sodium   Bisulfite
subcategory a total of seven days of sampling were conducted at Plants
#282,   #586,  and  #987.    Sixteen  different  sampling  points  were
identified for the various waste streams at these three  plants.   The
evaluation  of toxic metal content of these waste streams was based on
429 analytical data points and an additional 516 points for the  toxic
organic pollutants sampled during screening.

The  daily  raw waste loads were calculated from the waste stream flow
rates measured or estimated at the time of sampling and  the, measured
pollutant concentration.
     That is,

     Daily loading (as kg of pollutant per day)
                                                  (C) (Q)
                                                   1000
     where:
     C  is  the  concentration  of the pollutant expressed in units of
     .mg/1 (Note:  kg/m3 = 1000 mg/1), and
     Q is the waste steam flow rate  expressed  in  units
     (m3, a cubic meter,  is equal to 264.2 U.S.  gallons).
                                                           of  mVday.
     Similarly,  the  unit  loadings were calculated from the reported
     sodium bisulfite production rate, the waste stream flow rate, and
     the measured pollutant concentration.
                               723

-------
     Unit loading (as kg of pollutant per kkg of NaHS03
                                                            1000P
     where C
     sodium
     (kkg is
and Q are
bisulfite
1000 kg,
 the same as
  production
a metric ton
 described  above,  and
rate expressed in units
 which is equal to 2205
 P  is  the
of kkg/day.
Ibs.)
Table 24-8 presents the toxic pollutant unit loading and concentration
at the  three  plants  sampled.   Each  concentration  represents  the
average  of  three  composite  samples  for  verification and a single
composite sample during screening.

In Table 24-9, the toxic pollutant raw waste data are presented as the
minimum, average, and maximum  unit  loadings  based  on  the  results
summarized  for each plant in Table 24-9.   The average unit loading is
based on the average obtained at the three plants sampled.

Based on the total annual  production  of  this  subcategory  and  the
average  waste  load  generated  per unit product, the estimated toxic
pollutant raw waste loads generated each year for this subcategory are
as follows:
                     Total Annual Pollutant Load
          Pollutant

          Antimony
          Cadmium
          Chromium
          Copper
          Lead
          Mercury
          Nickel
          Zinc
          Silver
          Arsenic
          Thallium
                      Raw Waste Load (kg/year

                                 5.0
                                 1 .0
                              1100.0
                                45
                                 9.0
                                 0.60
                                30
                               520
                                 6.2
                                 2.3
                                22
Table  24-10  presents  the  average  toxic  pollutant   concentration
observed in the treated effluent during verification sampling.

Pollution Abatement Options

Toxic Pollutants of Concern

It  is reported that some sources of sodium carbonate contain zinc and
other  trace  metals  in  measurable  amounts.   The   screening   and
verification  sampling  program revealed zinc, chromium, copper, lead,
nickel, and antimony were at significant  concentration  levels  which
may  require regulation.  Zinc may enter the waste stream by corrosion
of galvanized metals by coproduct operations or from  nonprocess  zinc
compounds  used by the industry.  Cadmium, arsenic, thallium, mercury,
and silver though detected in the  raw  waste  are  not  at  treatable
                                  724

-------
                TABLE 24-8.  TOXIC POLLUTANT RAW WASTE LOADS
SUBCATEGORY
SODIUM BISULFITE
POLLUTANT
Arsenic
Copper
Zinc
Cadmium
Chromium
Lead
Mercury
Nickel
Antimony
Thallium
Silver
PLANT AND SAMPLING PHASE
Screening
#282
(mg/1) (kg/kkg)
0.012 0.000030
0.38 0.0010
2.5 0.0070
0.0060 0.000017

0.0025 0.000007
0.0030 0.000007
0.25 0.00070
0.030 0.000070
0.0080 0.000020
0.0060 0.000017
(2)
Verification v '
#987
(mg/1) (kg/kkg)
0.067 0.000010
0.74 0.000070
2.4 0.00020
0.040 0.000004
2.6 0.00030
0.60 0.000070
0.012 0.000001
0.46 0.000050
0.65 0.000070
<0.050 <0. 000004
<0.030 <0. 000003
#586
(mg/1)
0.0020
0.018
0.52
0.00050
1.3
0.012
0.00060
0.010
0.0050
0.025
0.010
(kg/kkg)
0.000030
0.00030
0.0088
0.000010
0.022
0.00020
0.000010
0.00017
0.000080
0.00042
0.00017
(1)  - One 72-hour  composite sample



(2)  - Average of three 24-hour composite samples
                                     725

-------
                                  TABLE 24-9.  SUMMARY OP RAH HASTE LOADINGS FOUND IN SCFUMIHG AM> VERIFICATION SAMPLING
•U
to
SUBCATBQORY
Pollutant

Priority
Antimony
Cadmium
Chromium
Copper
Lsafl
Mercury
Nickel
Zinc
Silver
Arsenic
Thallium
Conventional
Total Suspended

SODIUM BISUIfTFE
Loading Range
(kg/day)
Mini"?**

0.00045
0.00023
0.018
0.0050
0.000091
0.000091
0.0032
0.016
<0. 00020
0.00040
<0. 000052
Solids (TSS)
3.20
Maximum

0.0041
0.00041
1.1
0.015
0.0095
0.00045
0.0091
0.42
0.0080
0.0014
0.020

25.4
Unit loading
OcgAkg)
Minima

0.000007
0.000004
0.00030
0.000070
0.000007
0.000001
0.000050
0.00020
<0. 000003
0.000010

-------
TABLE 24-10.  TOXIC POLLUTANT CONCENTRATIONS OBSERVED IN TREATED EFFLUENT DURING
              VERIFICATICN SAMPLING

Pollutant
Arsenic
Copper
Zinc
Cadmium
Chromium
Lead
Mercury
Nickel
Antimony
Thallium
Silver
#987 P]
(nn/1) .
ND
0.27
0.010
ND
0.11
0.15
ND
ND
ND
ND
ND
Lant #586
(mcr/1)
ND
ND
ND
ND
ND
ND
0.010
0.050
0.020
ND
ND
ND - Not Detected
                                     727

-------
concentrations  and  thus
concern.
are  not  considered  toxic  pollutants  of
Prevailing Control and Treatment Practices

Plant #987 adds 50 percent caustic solution to the oxidation  tank  to
raise  the  pH  to  approximately  9.5  and  blows  air  through while
mechanically agitating.  The waste is discharged to a river  following
a 17-hour retention period.

Plant  #282  uses  caustic  soda  or  sodium  carbonate for pH control
followed by sodium hypochlorite addition to oxidize sulfite and  other
reduced  sulfur species.  The waste is then neutralized and discharged
to a County sewer.

Plant #586 mixes the bisulfite waste, amine  plant  waste,  and  truck
wash  waste.   Lime  is  added  to  the combined wastes which are then
passed through an aeration tank with eight-hour's retention time where
zinc sulfate production wastes are  added.   The  treated  waste  goes
through primary and secondary settling before final discharge.

Advanced Treatment Technologies

Toxic  metals  may be precipitated at alkaline pH values, when reacted
with sulfides in various forms, in some cases by ion exchange  resins,
and   the   Xanthate  process.   Sulfide  precipitation  from  cleared
solutions could be used to provide additional removal of  zinc,  lead,
nickel, copper, and to a lesser extent, antimony.

Selection of Appropriate Technology and Equipment

Technologies for Different Treatment Levels

A.   Level 1 (BPT/BAT)

     Neutralization with caustic soda to  a  pH  of  9.5  followed  by
     aeration  and  settling  was  chosen  as  the most cost-effective
     method of lowering the COD associated with the primary pollutant,
     sodium bisulfite.  Toxic metals in the waste form metal hydroxide
     precipitates which are removed from the wastewater  by  settling.
     The flow Diagram is shown in Figure 24-4.

B.   Level 2

     Aerated effluent from the BPT system is chlorinated  to  complete
     COD  removal,  and  is  then  filtered  to  remove finely divided
     suspended matter carried through or produced in the  BPT  system,
     particularly  if toxic metals are present in the incoming wastes.
     The flow diagram is shown in Figure  24-5.   Cost  estimates  for
     this  level  of  treatment  are given in the proposed Development
     Document (60).
                                  728

-------
C.   Level 3

     Ferrous sulfide is applied  ahead  of  the  Level  2  filter,  to
     precipitate  any  residual  metals  by the more effective sulfide
     process.   The  flow  diagram  is  shown  in  Figure  24-6.    Cost
     estimates  for  this level of treatment are given in the proposed
     Development Document (60).

Equipment for Different Treatment Levels

A.   Equipment Functions

     In Level 1, the raw wastes are received in  one  of  two  similar
     holding/reaction  ranks   (sized for one day's flow), recirculated
     and automatically adjusted to pH 9.5.  During the second  working
     day,  while  the  contents  of the first tank are jet aerated and
     allowed  to  settle,  the  daily   raw   wastes   are   received,
     recirculated and pH adjusted in the second tank.  At Level  1, the
     aerated  and  settled  effluent  is  discharged directly from the
     alternate holding/reaction tanks.  Sludge is drawn off the  tanks
     at  suitable  intervals  and disposed of.  At Levels 2 and 3, the
     aerated and settled effluent  is  pumped  through  a  multi-media
     filter  on  a  schedule which will leave its tank empty and ready
     for the  next  working  day's  influent.   At  Levels  2  and  3,
     supplemental  ch1orination  equipment  is  provided.  At Level 3,
     residual heavy metals which may have escaped prior treatment,  or
     may  have been released by pH change during chlorination, will be
     converted to metallic sulfides by  addition  of  ferrous  sulfide
     ahead of the Level 2 filter.  Should COD and toxic metals be very
     low  in  specific instances, the chlorination and ferrous sulfide
     steps could be eliminated.

B.   Chemicals and Handling

     Caustic soda solution, chlorine, and ferrous sulfide are used  in
     the  treatment  processes.   Caustic soda and chlorine are common
     industrial chemicals which  are  fed  by  conventional  equipment
     designed  to  minimize  leaks,  spills, and hazards to personnel.
     Ferrous sulfide is prepared by mixing ferrous sulfate with sodium
     bisulfide under  well  ventilated  conditions.   When  the  usual
     precautions  are  taken  in  the proper handling of corrosive and
     toxic chemicals, there should be no special problems in  applying
     the proposed technologies.

C.   Separation and Removal of Solids

     No solids are formed in the proposed treatment, with the possible
     exception of small amounts of metal hydroxides  and  sulfides  if
     metals  should  be present in the raw wastes.  In that event, the
     precipitated solids, which are  formed  in  the  holding/reaction
     tank  or  which  are returned to the holding/reaction tank during
     backwashing, will settle  in the hopper  bottom  of  the  reaction
                                729

-------
                        RAW_

                       WASTE

                       WVTHl
-J
Ul
o
CRUSTIC
SCOft
^
1
eta !
w 1 1 ,
\
_rV-UM 	
» 1

i
KJLDINCy
REBCTICW
TBW

                                                                          -MR
                                                       SUKXX
1— @
-L. !
! 1
+ f

U>*—

'



HctDiwy
REACTION
TBJK
]
^r^O-1
                                                                                             LEEBC


                                                                                      * Includes flow nDtiiboring, ffl

                                                                                       nonitoring, and
                                                       SLUDCZ
                              Figure 24-4.  Level 1 vraste  treatment for the soditm bisiiLfite subcategory
                                              batch process.

-------
                                                        BftCKWASH
                                                                            FILTER
r-(S)
t^ '
p^ i |
O-—
^ 1
1


•
HCtDINcy
FEflCTlCW
TANK

1
=+Q>-
f*-AIR
i
                                                 -OttORlNKTICW
                                                                                  t   EFFLUEHT
                                                                  * Includes flow mcnitaring, pH monitorinj,
                                                                    and sanpler.
                       SLUDGE
Figure 24-5,  Level 2 waste water treatment for the sodium bisulfite subcategory —
                batch process.

-------
-J
CJ
10
                 RPH

                WASTE
                HATER




CMETTC
scm
n







r^>
.L. '




<*-• 	 1


_^ 1
^ T
«3UJ»
BERCT1
TANF

1
•1

C/
X»






'
— »*Oj-
                                              SUKGB
--/S^j
i \ry

exj ,







Bou>n>
FEKTl
TAM





iq/
CN








J


'
— *-&-





)
                                                                                                              FILTEH
Q	
    EtfUUHW
                                                                                        * Includes flow monitoring, ffl monitoring,
                                                                                          and sawpler.
                               Figure 24-6.  I^wel 3 waste water treatment  for the  sodium bisulfite subcategory
                                               batch process.

-------
     tank.   As  necessary, these solids can be drawn off and disposed
     of in an appropriate manner.

D.   Monitoring Requirements

     Internal process monitoring will  be  done  with  standard  field
     equipment  measuring  pH,  dissolved  oxygen,  and  chlorine.  If
     metals are present- in the raw  materials  a  periodic  laboratory
     analysis  for  metals  should  be  made  on  the  final effluent.
     Monitoring for dissolved sulfide should not be  necessary,  since
     the ferrous sulfide applied in Level 3 is rather insoluble.

Treatment Cost Estimates
General Discussion

To  prepare  treatment  cost  estimates,   a  model  plant
developed for a single level of technology as follows:

A.   Wastewater Flow
concept was
     The sources of  wastewater  include  wet  air  scrubbers,  filter
     backwash,  floor washings,  leaks,  and spills.  The unit flow rates
     ranged  from  1.8  mVkkg  to  1.2 m3/kkg of product at the three
     facilities for which information  was available.  The average  was
     approximately  1.5  mVkkg  and this was used for the model plant
     (Table 24-4).

B.   Production

     Sodium bisulfite production ranges from  4770  kkg/yr  to  31,800
     kkg/yr  at  the  three  plants for which data was available.  The
     average production is 17,800 kkg/yr.  The production rates at the
     three plants were used as the model plant production rates.   The
     operational mode is continuous and is assumed to run 350 days per
     year.

C.   Solid Wastes

     In the production of sodium bisulfite and process waste treatment
     there is little solid waste generation, although precipitation of
     metal  hydroxides  may  result  in  small  quantities  of  solids
     requiring  disposal.   The  model  plants  assumed no significant
     solid waste production.

D.   Treatment Chemicals

     Caustic soda is needed to adjust  the pH to 9.5.  The  only  other
     requirement  is  air  to oxidize  the waste.  For the model plant,
     the caustic soda dosage was assumed to be 0.195 kg/kkg.
                                 733

-------
Cost Estimates

The cost estimates of three models having different  production
are presented in Tables 24-11, 24-12 and 24-13.
level
Table  24-14  gives  a  summary  of the unit cost distribution between
amortization, operation and maintenance.  Cost components  at  various
production levels are also shown.

Basis for Regulations

Evaluation of BPT Treatment Practices

All  seven plants in this subcategory have installed BPT or equivalent
technology.   Plant  peformance  was  estimated  on   the   basis   of
verification sampling results for Plant #987.  Plant #282 was excluded
from  the  evaluation  since  the  treatment technology applied at the
particular point of treated effluent sampling does not  represent  the
appropriate   level  of  treatment.   Plant  #586  was  excluded  from
consideration, since the combined treatment of  the  sodium  bisulfite
process  waste with wastes from other unrelated processes, has made an
evaluation of the plant performance data too speculative.

Table 24-15  is a comparison of the maximum  raw  waste  concentrations
from  screening  and verification sampling with treatability data from
Plant #987,  literature-based  estimates,  and  industrial  wastewater
treatment  system  performance.   The data used in this table are from
the screening and verification results, Table 8-11, Table 8-12,  Table
8-13 and the Treatability Study  (61).

Basis for BPT Effluent Limitations

A.   Technology Basis

     The  Agency  is  setting    limitations   based   upon   hydroxide
     precipitation  of  toxic  metals  with  caustic  soda  plus batch
     aeration and settling for BPT.  The flow  schematic  for  BPT  is
     shown   in  Figure  24-4  in Section 24 as Level 1 treatment.  The
     Agency  has selected Level   1  treatment  as  the  basis  for  BPT
     because it reflects current industry practice.

B.   Flow Basis

     The basis of flow for BPT   limitations  is  estimated  from  data
     provided  in  the Section 308-Questionnaires for two of the three
     complete plant responses, including Plant #987 and  #282.   Plant
     #586  was  omitted in view  of the lack of adequate  information to
     identify  the  wastewater   streams  contributed  by  the   sodium
     bisulfite process alone.

     The three major raw process wastewater streams include direct and
     indirect  process  contact  waste and miscellaneous  floor and tank
     washdown wastewater.
                                  734

-------
    TABLE 24-11.  MODEL PLANT TREATMENT  COSTS
Subcategory
Production
Sodium Bisulfite
4,770 metric tons per year
A. INVESTMENT COST
                               BPT
                                                     ($)
BATa
     Site development  	
     Equipment  	
     Monitoring equipment  ..

          Subtotal  	
     Contractor's 0 & Pb....

          Subtotal  	
     Engineering  	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL INVESTMENT COST  ,


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  .
     Energy 	
     Chemicals  	
     Maintenance  	
     Taxes and insurance ...
     Residual waste disposal
     Monitoring, analysis
      and reporting 	
                             3,000
                            70,000
                            20,000

                            93,000
                            13,950

                           105,950
                            21,390

                           128,340
                            12,834

                           141,174
                             3,000

                           144,174
     TOTAL OPERATION AND
     MAINTENANCE COST
                            56,000
                             2,100
                             2,900
                            14,117
                             4,325
                                 0

                            15,000
                            94,443
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                            22,969

                           117,412
a Represents the incremental cost above that  for  BPT  treatment
  Overhead and Profit
                             735

-------
   TABLE  24-12.  MODEL PLANT TREATMENT COSTS
Subcategory
Production
Sodium Bisulfite
16,900 metric tons per year
A. INVESTMENT COST

     Site development 	
     Equipment 	
     Monitoring equipment ..

          Subtotal 	
     Contractor's 0 & Pb....

          Subtotal 	
     Engineering  	

          Subtotal 	
     Contingencies 	

          Subtotal 	
     Land	

     TOTAL INVESTMENT COST  .


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  .
     Energy 	
     Chemicals 	
     Maintenance  	
     Taxes and insurance ...
     Residual waste disposal
     Monitoring, analysis
      and reporting 	
                                                     ($)
                               BPT

                             3,000
                           117,000
                            20,000

                           140,000
                            21,000

                           161,000
                            32,200

                           193,200
                            19,320

                           212,520
                             3,000

                           215,520
     TOTAL OPERATION AND
     MAINTENANCE COST
                            56,000
                             2,900
                             9,800
                            21,252
                             6,466
                                 0

                            15,000
                           111,418
BAT

  0
  0
  0

  0
  0

  0
  0

  0
  0

  0
  0
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                            34,577

                           145,995
  Represents the incremental cost above that for BPT treatment
  Overhead and Profit
                             736

-------
    TABLE 24-13.  MODEL  PLANT  TREATMENT  COSTS
Subcategory
Production
Sodium Bisulfite
31,800 metric tons per year
A. INVESTMENT COST

     Site development  	
     Equipment  	
     Monitoring equipment  ..

          Subtotal  .........
     Contractor's 0 &  Pb....

          Subtotal  	
     Engineering  	

          Subtotal  	
     Contingencies  	

          Subtotal  	
     Land	

     TOTAL INVESTMENT  COST  .


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  .
     Energy 	
     Chemicals  	
     Maintenance  	
     Taxes and insurance ...
     Residual waste disposal
     Monitoring, analysis
      and reporting 	
                                                    ($)
                               BPT

                             6,000
                           195,000
                            20,000

                           221,000
                            33,150

                           254,150
                            50,830

                           304,980
                            30,498

                           335,478
                             6,000

                           341,478
     TOTAL OPERATION AND
     MAINTENANCE COST
                            56,000
                             3,800
                            19,000
                            33,548
                            10,244
                                 0

                            15,000
                           137,592
BATa

  0
  0
  0

  0
  0

  0
  0

  0
  0

  0
  0
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                            54,582

                           192,174
  Represents the incremental cost above that for BPT treatment
  Overhead and Profit
                              737

-------
     TABLE 24-14 MODEL PLANT UNIT TREATMENT  COSTS
  Subcategory Sodium Bisulfite
                             Annual  Treatment  Costs  ($/kkg)
 COST ITEM
PRODUCTION
 (kkg/yr)
LEVEL OF TREATMENT

        BPT
Annual Operation
and Maintenance
Annual
Amortization
Total Annual
Cost
   4,770
  16,900
  31,800
   4,770
  16,900
  31,800
   4,770
  16,900
  31,800
       19.80
        6.59
        4.33
        4.82
        2.05
        1.72
       24.61
        8.64
        6.04
                          738

-------
              TABLE 24-15.
COMPARISON OF MAXIMUM RAW WASTE
CONCENTRATIONS WHH TREftTABILITY
Pollutant  Maximum Avg. Raw  Treated Effluent  Literature-Based  Industrial Waste
                Waste
            Concentration
           During Screening
           and Verification
                (mg/1)
  Concentration
 Fran Plant #987
  Verification
   Sampling
    (mg/D
 Treatability
Estimates from
 Table 8-11 for
 Lime/Settling
    (mg/1)
 Water Treatment
System Performance
 for Lime/Settling

     (mg/1)
TSS
COD
Arsenic
Copper
Zinc
Cadmium
Chromium
Lead
Mercury
Nickel
Antimony
Thallium
Silver
2,250
24,700
0.067
0.74
2.5
0.0060
2.6
0.60
0.012
0.46
0.65
0.050
0.030
22
7,300
ND
0.27
0.010
ND
0.11
0.15
ND
ND
ND
ND
ND
—
—
0.50
0.50
0.50
0.10
0.10
0.30
0.20
0.80
0.20
0.40
38(1)
450 (2)
0.080(3J
0.40(4)
Q.80(4)
0;32(4>
0.15(4)
0.40(4)
0.18(3)
—
—
ND =  Not Detected

(1)   Data from Treatability Study (61)  on lime/settling in nickel sulfate and
     sodium dichromate industries.

(2)   Average of extended aeration tests from Treatability Study (61).

(3)   Data from Table 8-12.

(4)   Estimated achievable long-term average concentrations from Table 8-13.
                                     739

-------
     Table 24-4 summarizes the unit flows reported  for  each  of  the
     three  sources  at  each facility.  Plant #987 has two facilities
     which manufacture sodium bisulfite and are designated as A and B.
     The basis of model plant flow for the sodium  bisulfite  industry
     is  estimated  as  the  average total raw wastewater flow for the
     three  plants,  and  is  used  to  estimate  pollutant  discharge
     loadings  for  the purpose of regulation.  The average total flow
     for the three facilities considered is 1  .5 mVkkg of product.

C.   Selection Basis for Pollutants to be Regulated

     The selection of pollutants for which specific numerical effluent
     limitations are promulgated was based on  an  evaluation  of  raw
     waste  data from the screening and verification sampling program.
     Pollutant data from the plant sampled during screening  was  used
     to  determine  the  need for verification sampling.  Verification
     sampling at Plants #586 and #987  provided  additional  pollutant
     raw  waste concentration data needed to assess the performance of
     treatment technology.
     Results of the screening and verification sampling are
     in Table 24-15 for the raw process waste streams.
tabulated
     Toxic  pollutants  are  listed  based  on  their presence, during
     sampling, at detectable concentration  levels.   Pollutants  from
     this  list  were considered as candidates for regulation if their
     concentration appeared at least to equal  or  exceed  the  lowest
     level  estimated  as  treatable  using  any  available technology
     appropriate for their removal (Table 8-11).

     The relative significance of the candidate toxic  pollutants  was
     estimated  based  on  the  total  annual  raw waste load for each
     pollutant which appears in a Table above.  The total annual  load
     is  based  on the average concentration observed during screening
     and verification, which is tabulated in Table 24-9,  in  addition
     to  the  estimated annual production of 98,000 kkg of product for
     the industry.

     On the basis of concentration and total annual  raw  waste  loads
     determined  during  sampling,  COD,  TSS, chromium, zinc, copper,
     nickel, lead, and antimony have been  identified  at  significant
     concentration  levels  in the raw waste stream and are candidates
     for regulation.  These pollutants are listed in  order  of  their
     relative   significance  with  regard  to  decreasing. raw  waste
     concentration.

     In view of the treatment technology currently practiced  and  the
     related nature of the candidate toxic pollutants (see Section 8),
     control  of  the  more  significant  toxic pollutants will ensure
     adequate control of those metals which may occasionally appear at
     treatable levels.  Therefore, only chromium and  zinc  will  have
     effluent  limitations.  Concentration values for the other metals
     are intended for guidance only.
                                  740

-------
     The Agency conducted treatability  studies  (61)  using  Level  1
     (BPT)  technology  (without  settling)  on typical raw wastewater
     from the sodium bisulfite industry.  In the tests the average COD
     level was reduced from 950 mg/1 to 450 mg/1.   Use of the standard
     iodide-iodate test for sulfite indicated the  tests  reduced  the
     average  iodate  demand  as oxygen from 800 mg/1 to 25 mg/1.  The
     Agency is evaluating the standard iodide-iodate test for  sulfite
     for possible application in effluent monitoring but will continue
     to set effluent limitations using the conventional COD test.

D.   Basis of Pollutant Limitations

     BPT limitations are presented in Table 24-16.

     1.    Conventional and Nonconventional Parameters

          a.   pH

               The treated effluent is to  be  controlled  within  the
               range  of  6.0 to 9.0.  This limitation is based on the
               data  presented  in  Appendix   B   of   the   proposed
               Development Document (60) and the JRB Study (52).

          b.   TSS

               The BPT limitations for TSS are based on data from  the
               Treatability  Study  (61)  using lime/settling from the
               nickel sulfate and sodium dichromate  industries.    The
               long-term  average  is  38  mg/1.  The daily and 30-day
               variability factors are 5.6 and 1.4, respectively.

               The 24-hour maximum concentration is:

               (38 mg/1) (5.6) = 210 mg/1

               The maximum 30-day average concentration is:

               (38 mg/1) (1.4) = 53 mg/1

               The load limitations for TSS  (kg/kkg)  are  calculated
               based on the unit flow rate of 1.5 mVkkg, thus:

               (210 mg/1) (1.5 mVkkg)   (  kg/m3  ) = 0.32 kg/kkg
                                         (1000 mg/1)

               for  the  24-hour  maximum  limit.    The maximum 30-day
               average is calculated similarly, i.e.:

               (53 mg/1) (1.5 mVkkg)   (  kg/m3   )  = 0.080 kg/kkg
                                        (1000 mg/1)
                                    741

-------
                        TABLE 24-16.  EFFLUENT LIMITATIONS
                                 SODIUM BISULFITE
             Best Practicable Control Technology Currently Available
                            Wastewater Flow: 1.5 m^/kkg
Pollutant
Subcategory
Performance
  (mg/1)   ___
Daily
Variability
Factor	
30-day
Variability
 Factor
Concentration   Effluent Limit
Basis (mg/1)     (kg/kkg)	
Max.            Max.
30-day   24-hr. 30-day    24-hr.
Avg.	Max.   Ayg,	Max.
Conventional/
Nonconventional
Pollutants:
Total Suspended
Solids, TSs(6>
Chemical Oxygen
Demand, COD*7)
Toxic Pollutants :
Chromium*7)'*9)
Zinc*7)
Copper
Lead
Nickel
Antimony
38(D
450<2)
0.32*3)
0.80(3)
0.40*3)
0.15(3)
0.40(3)
0.80(5)
5.6/1.4
5.6/1.4
4.2/1.3
4.2/1.3
4.2/1.3
4.2/1.3
4.2/1.3
4.2/1.3
53
630
0.42
1.0
0.52
0.20
0.52
1.0
210
2500
1.3
3.4
1,7
0.63
1.7
3.4
0.080
0.95
0.00063
0.0015
— (8)
— (8)
— (8)
~(8)
0.32
3.8
0.0020
0.0051
— (8)
-.(8)
— (8)
~(8)
(1)  Data from Treatability Study (61) on lime/settling in nickel sulfate and
     sodium dichromate industries.
(2)  Based on data from Treatability Study (61).
(3)  Based on estimated achievable long-term average concentrations from
     Table 8-11.
(4)  Maximum untreated effluent concentration from screening and verification
     sampling data.
(5)  Based on lower limit of literature-based treatability estimate from
     Table 8-11.
(6)  Also applicable to NSPS regulations.
(7)  Also applicable to NSPS/ and BAT regulations.
(9)  No effluent limitation.
(9)  Also applicable to PSNS regulations*
                                       742

-------
     c.    COD

          The BPT limitations for COD are based on data from  the
          Treatability  Study  (61)   on  the  aeration  of sodium
          bisulfite waste.    The  average  concentration  of  the
          treated effluent  from the tests,  450 mg/1, is used as a
          long-term   average.    The  daily  and  30-day  average
          variability factors used for  TSS  above  are  employed
          again.

          The 24-hour maximum concentration is:

          (450 mg/1) (5.6)  = 2500 mg/1

          The maximum 30-day average concentration is:

          (450 mg/1) (1.4)  = 630 mg/1

          The  load  limitations  for COD (kg/kkg) are calculated
          based on the unit flow rate of 1.5 m3/kkg, thus:

          (2500 mg/1) (1.5  mVkkg)   (  kq/m^  )   =3.8 kg/kkg
                                     (1000  mg/1)

          for the 24- hour  maximum  limit.    The  30-day  average
          limit is calculated similarly, i.e.:
2.
     (630 mg/1) (1 .5 mVkkg)


Toxic Pollutants
(   kg/m3   )
(1000 mg/1)
                                                 =0.95 kg/kkg
     No long-term sodium bisulfite industry data is available  to
     establish  limitations and guidelines for the selected toxic
     pollutants.   The effluent  limitations  and  guidelines  are
     based  on four information sources including:  (1) estimated
     achievable long-term average concentrations from Table 8-13,
     (2)  industrial wastewater treatment system performance  data
     from  Table   8-12,   (3) screening and verification data, and
     (4)  literature-based treatability estimates.
     a.    Chromium

          The limitations for chromium are based on the estimated
          achievable long-term average concentration of 0.32 mg/1
          from  Table  8-13.    The  daily  and   30-day   average
          variability  factors  of 4.2 and 1.3,  respectively,  are
          obtained from data in the Treatability  Study  (61)   on
          chromium   removal    using   lime/settling   from   the
          chlor-alkali, titanium dioxide, sodium  dichromate  and
          chrome pigments industries.
                          743

-------
b.
     The 24-hour maximum concentration is:

     (0.32 mg/1) (4.2)  = 1.3 mg/1

     The maximum 30-day average concentration is:

     (0.32 mg/1) (1.3)  = 0.42 mg/1

     The   load   limitations   for  chromium  (kg/kkg)   are
     calculated based on the unit flow rate of  1.5  mVkkg,
     thus:
     (1.3 mg/1)  (1.5 mVkkg)
                             kq/m3   )   - 0.0020 kg/kkg
                                1000 mg/1
     for  the  24-hour  maximum  limit.    The maximum 30-day
     average is calculated similarly,  i.e.:
     (0.42 mg/1)  (1.5 mVkkg)
Zinc
                               kq/m3  )   = 0.00063 kg/kkg
                                 (1000 mg/1)
     The limitations for zinc are  based  on  the  estimated
     achievable long-term average concentration of 0.80 mg/1
     from   Table   8-13.   The  daily  and  30-day  average
     variability factors  used  for  chromium  are  employed
     again.

     The 24-hour maximum concentration is:

     (0.80 mg/1) (4.2) = 3.4 mg/1

     The maximum 30-day average concentration is:

     (0.80 mg/1) (1.3) = 1.0 mg/1
     The  load  limitations for zinc (kg/kkg) are calculated
     based on the unit flow rate of 1.5 mVkkg, thus:

     (3.4 mg/1) (1.5 mVkkg)   (  kq/m^  )   = 0.0051 kg/kkg
                               (1000 mg/1)

     for the 24-hour  maximum  limit.   The  maximum  30-day
     average is calculated similarly, i.e.:

     (1,0 mg/1) (1.5 mVkkg)   (  kg/m3  )   = 0.0015 kg/kkg
                               (1000 mg/1)
                      744

-------
Copper

The guidance for  copper  is  based  on  the  estimated
achievable long-term average concentration of 0.40 mg/1
from   Table   8-13.   The  daily  and  30-day  average
variability factors  used  for  chromium  are  employed
again.

The 24-hour maximum concentration is:

(0.40 mg/1) (4.2) = 1.7 mg/1

The maximum 30-day average concentration is:

(0.40 mg/1) (1.3) = 0.52 mg/1

Lead

The  guidance  for  lead  is  based  on  the  estimated
achievable long-term average concentration of 0.15 mg/1
from   Table   8-13.   The  daily  and  30-day  average
variability factors  used  for  chromium  are  employed
again.

The 24-hour maximum concentration is:

(0.15 mg/1) (4.2) = 0.63 mg/1

The maximum 30-day average concentration is:

(0.15 mg/1) (1.3) = 0.20 mg/1

Nickel

The guidance for  nickel  is  based  on  the  estimated
achievable long-term average concentration of 0.40 mg/1
from  Table  8-13.   The  variability  factors used for
chromium are employed again.

The 24-hour maximum concentration is:

(0.40 mg/1) (4.2) = 1.7 mg/1

The maximum 30-day average concentration is:

(0.40 mg/1) (1.3) = 5.2 mg/1

Antimony

The   guidance   for   antimony   is   based   on   the
literature-based  treatability  estimate  in Table 8-11
which  is  higher  than   the   industrial   wastewater
treatment system performance data from Table 8-12.  The
                   745

-------
               value  of 0.80 mg/1 for the lime/settling technology is
               used as a long-term average.  The  variability  factors
               used for chromium are employed again.

               The 24-hour maximum concentration is:

               (0.80 mg/1) (4,2) = 3.4 mg/1

               The maximum 30-day average concentration is:

               (0.80 mg/1) (1.3) = 1.0 mg/1

Basis for BCT Effluent Limitations

While   EPA  has  not  yet  proposed  or  promulgated  a  revised  BCT
methodology in  response  to  the  American  Paper  Institute  v.  EPA
decision  mentioned  in Section 3, EPA is promulgating BCT limitations
for this subcategory.  These limits are identical to  those  for  BPT.
EPA  is  not promulgating any more stringent limitations since we have
identified  no  technology  option  which  would  remove   significant
additional  amounts of conventional pollutants.  As BPT is the minimal
level of control required by law, no possible application of  the  BCT
cost   tests   could  result  in  BCT  limitations  lower  than  those
promulgated in this regulation.  Accordingly, there is no need to wait
until  EPA  revises  the  BCT  methodology  before  promulgating   BCT
limitations.

Basis for BAT Effluent Limitations

A.   The Application of Advanced Level Treatment

     The Agency has analyzed the cost effectiveness of the base  level
     systems   (BPT)  and  the  various  advanced   level  options  for
     conventional and  toxic  pollutant  removal  based  on  the  cost
     estimates  presented  in this report and the proposed Development
     Document (60).  The regulations being promulgated for BAT consist
     of  Level  1  or  BPT  treatment.   The  removal  of   additional
     pollutants  by Levels 2 and 3 treatment systems  is not sufficient
     to offset the  additional  cost  to   install   and  operate  these
     advanced treatment systems.

B.   Technology Basis

     The BAT treatment system is the same as that described above  for
     BPT.

C.   Flow Basis

     The model plant flow developed for BPT treatment applies also  to
     BAT  in  the development of the regulations.   Therefore the value
     of 1.5 mVkkg of product is used for the unit  flow.
                                  746

-------
D,   Selection of Pollutants to be Regulated

     For the BAT regulations, the Agency is setting the regulation  of
     COD  and the same two toxic metals considered for BPT limitations
     listed in Table 24-16.  Concentration values for the other metals
     are intended for guidance only.

E.   Basis of Pollutant Limitations

     1.    Nonconventional Pollutants

          The only nonconventional pollutant  is  COD  in  the  sodium
          bisulfite  subcategory.  Since BAT has been set equal to BPT
          by the Agency, the limitation is then identical to  BPT  for
          COD.  Refer to Table 24-16 for the BAT regulation.

     2.    Toxic Pollutants

          The Agency is setting limitations on chromium and zinc which
          equal those for BPT.  See above for the development of these
          limitations.

Basis for New Source Performance Standards

The NSPS limitations {applicable to. pH, TSS, COD and two toxic metals)
are set equal to BAT for toxic and nonconventional pollutants and  BPT
for  conventional  pollutants.   Table  24-16  for  the  BPT  and  -BAT
limitations would be identical in all respects with NSPS  limitations.
See above for the development of the regulations.

Basis for Pretreatment Standards

Pretreatment  standards  are  necessary  because  NSPS provides better
removal of chromium than is achieved  by  a  well-operated  POTW  with
secondary  treatment  installed  and  therefore  chromium  would  pass
through a POTW in the absence of pretreatment.  Based on  the  average
raw  waste  loads  given  in  Table  24-9  and  the  long-term average
subcategory performance given in Table 24-16, NSPS  treatment  reduces
COD by 76% and zinc by 77%.  A POTW will reduce COD by 80% and zinc by
76% (See Fate of Priority Pollutants in Publicly Owned Treatment Works
-  Interim  Report, EPA-440/1-80-301, October 1980).  A POTW therefore
achieves a percent removal equal to or greater than that  achieved  by
NSPS for COD and Zinc, and therefore there is no pass through of those
two pollutants in this subcategory.

A.   Existing Sources

     The  Agency  is  excluding  this  subcategory  from  Pretreatment
     Standards  for  Existing  Sources  (PSES) under the provisions of
     paragraph 8{b)(ii)  of the Settlement Agreement.  The total  toxic
     metal discharge from the one existing indirect discharger to POTW
     is  120  pounds  per  year,   which  is so insignificant as not to
     justify developing  a national standard.
                                   747

-------
B,   New Sources

     For Pretreatment Standards for New Sources  (PSNS),  the Agency   is
     promulgating limitations based on NSPS.  The pollutant limited  is
     chromium.
                                   748

-------
                              SECTION 25
      SODIUM HYDROSULFITE (FORMATE PROCESS) INDUSTRY (Excluded)
Summary of Determinations

We proposed BPT, BCT, and BAT limitations and NSPS, PSES, and PSNS for
this  subcategory.  The proposed regulation basically added control of
selected toxic metal pollutants to existing treatment practiced in the
industry.  We have reviewed the basis for the proposed regulation  and
we  concluded  that  the  total current treated discharge load of only
0.42 pounds per  day  total  toxic  metals  from  all  plants  in  the
subcategory  is  too  insignificant  to  justify developing a national
regulation.  Accordingly,  we  have  excluded  this  subcategory  from
national  regulation  development  under  paragraph  8(a)(iv)  of  the
Settlement Agreement.

The information presented in the remainder of this Section is provided
as guidance for use by permit writers.

Industry Profile

General Description

Most of the sodium hydrosulfite produced in the U.S. is  sold  in  the
merchant  market.    Sodium  hydrosulfite is used extensively in dyeing
cotton and in the printing industry.  It is a powerful reducing  agent
and  is  used  in wood pulp bleaching, and stripping operations in the
food,  vegetable oil, and soap industries.

The industry profile data are presented in Table 25-1, while status of
regulations are summarized in Table 25-2.

Subcategorization

A detailed summary  of  factors  considered  in  Subcategorization  is
presented  in  Section  4.   Sodium  hydrosulfite is produced by three
processes including the Werbs, formate, and zinc processes.  The  zinc
and  Werbs  processes  are  deferred  to  Phase  II  of  the inorganic
chemicals regulation development effort.  This  section  concerns  the
formate process only.

General Process Description and Raw Materials

In  the  formate  process, sodium hydrosulfite is produced by reacting
sodium formate solution, sodium hydroxide solution, and liquid  sulfur
dioxide in the presence of a recycled stream of methanol solvent.   The
general reaction is:

HC02Na + 3NaOH + 3S02 = Na2S204 + NaHC03 •*- Na2S03 + CO + 2H20 (1)
                                 749

-------
         TABIE 25-1.  SOECKSECOfS PRCFILE DATA SIMIRRf
 SUBCKBSGQEV
SODIUM HQSROSUCFETE  (FQEWKEE PBCCESS)
Total subcategory capacity rate
Total subcatagory production rate
Number of plants  in this subcategory
308 Data on file  for
    With total capacity of
    With total production of
    Representing  capacity
    nepmnni hij  ptt xlucti on
    Plant production range t
            Maxinun
    Avaraga prcduction
    Median production
    Avwrag* capacity utilization
    Plant ag« ranges
            Mixiinua
            Maxinun
    Wutetater flow range:
            Maxinua
    Vblun per unit product:
            Hininun
            Maximum
                           40,340 kkg/y«ar
                           39,940 Wcg/year
                                2
                                X
                           20,450 kfcg/year
                           20,450 kkg/year
                               50 percent
                               51 percent

                               NA
                              100 percent
                               NA
                              273 cubic raters/day
                               NA
                               KA

                             4.68 cubic rosters/kkg
                               NA
                               NA
Sources of data are Stanford Research Institute, Directory of
Producers, U.S.A., 1977, U.S. Department of Qcrnnarce, Current Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry,"  June, 1978 and "Economic Analysis of Proposed
Revised Effluent Guidelines and standards for the Inorganic Chemicals Industry,'
torch, 1980.
NA  •  Not Available
                      750

-------
     TABLE 25-2.   STATUS OF REGUTAUCNS  -  EFFLUEKP TJMTTWPTrW GUIDELINES
SOPCATFXjQKg

SUBPAET
      Sodium Hydrosulf ite

      BE  (40 CFR 415.570, 5/22/75}
                                        STANDARDS
Product
Process
                             BPCTCA

                      Max.1     Avg,
                                  EATEA
                              Max.   Avg.
Para-
meters
(mg/1)     (mg/1)
kg/kkg
 (ng/1) Ong/1)
     NSPS
 Max*     Avg,
kg/kkg   kg/ldcg
  (mg/1)    (mg/1)
 Sodium
 Hydro-
 Sulfite
Reserved
    Reserved
 Reserved
 Reserved
       = Maximum of any one day.
  Avg.  » Average of daily values for thirty consecutive days,
                                    751

-------
The operation occurs in several steps:

     An  aqueous solution of sodium formate is prepared and introduced
     into the reactor.
     The recycled stream of  methanol
     introduced into the reactor.
            containing  sulfur  dioxide  is
     The  sodium hydroxide and sodium formate solutions, liquid sulfur
     dioxide, and recycled methanol are then contacted under  pressure
     at slightly elevated temperatures.

Sodium  hydrosulf ite  then  precipitates  and  forms  a  slurry in the
reactor.  The by-product, sodium sulfite, and sodium  bicarbonate  and
carbon monoxide gas are formed.

There is a small amount of methyl formate produced in the reactor as a
side reaction between the sodium formate and methanol:
HC02Na + CH3OH = HCOZCH3
NaOH
                                                             (2)
This side reaction product remains in the  recycling  methanol  during
the  entire  process.   As  a  result,  some  of  the methanol must be
periodically purged from the recycle system to avoid excessive buildup
of this impurity,

The resulting  slurry  of  sodium  hydrosulfite  in  the  solution  of
methanol,  methyl  formate,  and  by-products is sent to a pressurized
filter operation which recovers the crystals of  sodium  hydrosulfite.
The  crystals  are dried in a steam heated rotary drier, recovered and
packaged.  The filtrate and backwash liquors from the filter operation
are sent to the solvent recovery system as is the  vaporized  methanol
from  the  drying  operation.   The  drying of the sodium hydrosulfite
filter cake must be done very carefully as it is  heat  sensitive  and
tends to decompose to sulfite.

A  general  process flow diagram for Plant 1672 can be found in Figure
25-1, as it is typical for this subcategory,

Water Use and Waste Source Characteristics

Water Use

Water is used in the process as make up for the reaction solutions and
for steam generation in the rotary dryers.  Water  is  also  used  for
noncontact  cooling  in  the reactor gas vent scrubbers and dryers, as
well as pump seals  and  washdowns,  and  as  dilution  water  in  the
wastewater  treatment  system  to  assist  in  biological oxidation of
organic materials.
                                    752

-------
                                         GASEOUS
                                       BY-PRCCUCTS
                                                             VENT GAS

                                                             SCRUBBER
LIQUID
                                                                      DICMIEC
        SODIUM FORMRTE SOLUTION
-4
U1
W
        SODIUM HYDPDXIDE SCUTHON
                                                          REC*ClH>"MErmNCI,, SO. AND
                                           RECTOR
                                                                           FORMATE
                                                                      BftCKNftSH
                                                                      LIQUOR

                                                            80DHM HYDROSULFITB
                                          VtWT SVS
                                          8CRJBBER
                             DISnUJBR
                              OOUMN
                             RBCOVERY)
                                                                                  VAPORIZED
                                                                                   MEIHANX
                                                                                     DRYER
                                                OOUfW BOTTCMS
                                                                                                                 TO K*D)
                                                                                                       PHCOUCT
                                     Figure 25-1.   General process flow diagram at plant  #672.
                                                       (Sodium hydrosulfite manufacture.)

-------
Waste Sources

The  strongest  process  waste  is  the  aqueous  residue   from   the
distillation  column  bottoms  (solvent  recovery system).  This waste
contains concentrated reaction by-products  and  is  purged  from  the
system  at  a  rate of approximately 14,000 gallons per day.  At Plant
#672, this waste is sent to a by-product pond where  it  is  held  and
either  sold to the pulp and paper industry or bled into the treatment
system.

The dilute wastes from the process are contributed by  leaks,  spills,
washdowns,  and tank car washing.   At Plant #672, this is collected in
a sump and then sent to the biological treatment system.

Cooling  tower  and  boiler  blowdown  constitute  a   noncontaminated
wastewater  source.   This  is  sent  to  the final compartment of the
chlorine contact  tank  without  treatment,  for  discharge  with  the
combined effluent of the treatment plant.

The vent gas scrubbers create a wastewater source which is sent to the
methanol  recovery  stills  for  recycle.   At  Plant #672, this waste
eventually goes to the by-product pond with  the  distillation  column
bottoms.

Table  25-3  presents  the unit flows for the three primary sources of
process wastewater which contribute to the pollutant load.

Solid wastes currently are generated in  the  activated  sludge  waste
treatment system.  An estimated 2,400 gallons of biological sludge are
discharged  per  day  to  an  on-site drying bed.  Application of more
stringent waste treatment of toxic pollutants is estimated to generate
an additional 6 kg/kkg  of  product  of  solid  waste  which  must  be
disposed of at an approved site.

Description of Plants Visited and Sampled

Screening and Verification

The  only  plant  visited  during the sampling program was Plant #672,
where verification sampling procedures were used.  Plant #672  is  one
of two plants that currently utilize the formate process in the sodium
hydrosulfite  subcategory.   An  evaluation  of  plants that currently
utilize the zinc process for sodium hydrosulfite manufacture has  been
deferred to a later phase of regulation development by the Agency.

Data  from Plant #672 can be considered representative of this process
for both plants, since the other plant  in  this  subcategory  has  an
identical,  though slightly smaller, production process.  However, the
second plant has a different waste treatment system.  It also receives
large loadings of waste from several other products.  Because of  this
the  plant is considered nonrepresentative of the hydrosulfite process
and visits were limited to Plant #672 for this reason.
                                   754

-------
                 TABLE 25-3.  WASTE SOURCE DATA AT PLANT #672
WASTE SOURCE
 FLCW
(m3/kkg)
Dilute Waste  (spills, etc.)
Dilution Water (contact)
By-product Waste
 1.95
 1.75
 0.95
Ttotal                                          4.65
 (Basis of flow  for model plant and regulation development)
                                     755

-------
in
                             KXU1H
                                         mDBuurm
                                0mtM  *
                                •mum
                                                     IWMHi*!,
                                              n
                                          nun
                                          nun
                                  riHlUIU MO tNK
                          Mawa went
                        tMRAUDQB
                            •r-wntcf
J-
                                                                           CHUBB VM1
                                                                         MDmURMOBNI
mugMi

u*T

                                                                       oujam
                                                      M cuuorn*
                      Mci,oaNacr
                ^
                                                                                                        •*•»««.
                                                                                                     14
                  Figure 25-2.  General process flew diagram at plant 1672 showing the sanpling points.

                                 (Sodium hydrosulfite manufacture.)

-------
A general flow diagram of Plant # 672 showing  process  waste  sources
and  sampling  points  is  shown  in  Figure  25-2.   The  sources  of
wastewater for each sampling point are as follows:

     By-product pond.

     Dilute waste from sodium hydrosulfite process area and sumps.

     Combined influent to treatment.  This point collects  waste  from
     points 1 and 2, plus the sodium bisulfite waste stream.

     Treated effluent at the outfall.

At  the  time  screening sampling was conducted at Plant #672, none of
the by-product wastewater was being sent to the  biological  treatment
system.   As  a  result,  the  sodium hydrosulfite process waste being
treated was from the dilute waste area only.

Table  25-4   presents   the   results   of   the   conventional   and
nonconventional  pollutant  concentrations  and unit loads for each of
the streams sampled.  The results are based on three 24-hour composite
samples.  It should be noted that sampling was done during a time when
no by-product waste was entering streams  3  and  4.   The  unit  flow
indicated is the estimated flow observed during sampling.

Toxic Pollutant Concentrations

Toxic  pollutants  were  identified in the raw process waste stream at
Plant #672.  The following toxic pollutants were found  at  detectable
concentration levels.
              Maximum Raw Waste Concentrations Observed
                                d.g/1)
               Pollutant
               Arsenic
               Cadmium
               Chromium
               Copper
               Lead
               Nickel
               Silver
               Zinc
               Pentachlorophenol
               Phenol
               Cyanide
               Mercury
               Selenium
Verification
 Plant #672
      79
      43
    9300
    1500
    1300
    1700
     130
   27000
     580
     170
     100
      28
      34
Two  toxic  organic  pollutants,  pentachlorophenol  and  phenol, were
identified at low, but  detectable,  concentration  levels.   The  raw
process  materials,  including sodium formate and methanol, are likely
sources.  The sodium formate  currently  used  in  the  process  is  a
                                   757

-------
TABLE 25-4.  FLOW, POLLUTANT CONCENTRATION, AND LOAD DATA OF THE SAMPLED
             WASTE STREAMS FOR PLANT #672 PRODUCING SODIUM HYDRDSULFITE

Flow TSS COD
Stream ~
Designation Description (m /kkg) (mgA) (kgA^g) (mg/1) (kg/kkg)
1
2
3
4
By-product 0.95 61 0.058 78,000 74
Dilute Waste 1.95 260 0.51 15,000 29
Dilute Haste and /,,
SBS Waste 2.05 840 1.7 16, 000 ^ 32
Final Discharge 4.87 25 0.12 740°-^ 3.6
(1)   Value is that observed during sampling which may differ significantly
     if the by-product stream is contributing.
                                   758

-------
by-product  from  an  unrelated  organic  chemicals  process which may
contain the organic impurities.  Methanol is also a suspect source  of
organic  impurities  in  view  of  the  difficulty  involved  with its
purification and high degree  of  solubility  with  pentachlorophenol.
Also  possible  is  coincidental formation of pentachlorophenol in the
process due to the presence of specific chlorinated hydrocarbons under
conditions conducive to its development.

Section 5 of this report describes the methodology  of  the  screening
and   verification  sampling  program.   In  the  sodium  hydrosulifte
industry, a total of three days of sampling were  conducted  at  Plant
#672.   Three  24-hour  composite samples were taken at four different
sampling points.  The sampling involved 169 analytical data points for
the toxic inorganic pollutants and 387 additional points for the toxic
organic pollutants.  The daily raw waste loads  were  calculated  from
the  waste  stream  flow  rates  measured  or estimated at the time of
sampling and the measured pollutant concentration.

     That is,

     Daily loading (as kg of pollutant per day) = (C)(Q)
                                                  (1000)
     where:

     C is the concentration of the pollutant  expressed
     mg/1 (Note:  kg/m3 = 1000 mg/1),  and
in  units  of
     Q  is  the  waste  stream flow rate expressed in units of m3/day.
     (m3,  a cubic meter, is equal to 264'.2 U.S. gallons.)

Similarly, the unit loadings were calculated from the reported  sodium
hydrosulfite  production  rate,  the  waste  stream flow rate, and the
measured pollutant concentration.

     Unit loading (as kg of pollutant   = (C)(Q)
     per kkg of sodium hydrosulfite)      1000P

     where C and Q are the same as  described  above,  and  P  is  the
     sodium   hydrosulfite  production  rate  expressed  in  units  of
     kkg/day.  (kkg is 1000 kg, a metric ton, which is equal  to  2205
     Ibs.)

Table   25-5  presents  the  average  toxic  pollutant  concentrations
observed during sampling for the raw and treated wastewaters at  Plant
A;b72.  The concentration indicated is based on three'24-hour composite
samples.   Table 25-6 is a tabulation of the unit loadings for each of
the toxic pollutants found at  detectable  levels  in  the  raw  waste
water.

The  estimated  total  toxic  pollutant raw waste loads generated each
year for this subcategory were based on  the  total  estimated  annual
production of sodium hydrosulfite.  The loads are as follows:
                                    759

-------
          Pollutant
Waste Load (kg/year
          Arsenic                             4.8
          Cadmium                             1 .3
          Chromium                           22
          Copper                              7.6
          Lead '                              40
          Nickel                             64
          Silver                              6.4
          Zinc                              960
          Pentachlorophenol                  33
          Phenol                              6.0
          Cyanide                             1.56
          Mercury                             0.80
          Selenium                            1.2

Pollution Abatement Options

Toxic Pollutants of Concern

Although  sodium  hydrosulfite  is being manufactured by both the zinc
process and the formate process, the  trend  is  away  from  the  zinc
process  for environmental reasons.  This discussion concerns only the
formate process, using a sodium formate feed stock from a source which
appears to contain heavy metal  impurities  (chromium,  zinc,  nickel,
lead,  and copper) as well as trace amounts of cyanide.  A predominant
characteristic of sodium hydrosulfite wastes is  their  high  chemical
oxygen  demand  resulting  from  various forms of sulfite, from methyl
formate,, and from residual methanol after a solvent recovery  process.
Low levels of phenolic compounds are also found in the raw wastes.

Prevailing Control and Treatment Practices

Due  to  the  nature of the two primary raw waste streams, each one is
handled differently.  The dilute waste is first sent to a holding pond
where the flow is equalized and the waste mechanically aerated.   This
pond  also contains approximately  1500 gallons per day of waste from a
sodium bisulfite process.  The  pond  effluent  is  pH  adjusted  with
sulfuric  acid  and  sent  to an aeration basin.   A nitrogen-phosphate
fertilizer and urea are added  to  provide  nutrients.   Approximately
3500  gallons  per  day of sanitary waste and up to 25,900 gallons per
day of clean dilution water are also  added  to  the  aeration  basin.
This basin formerly had mechanical aerators, but now has air diffusers
which  allow  better temperature control for biological oxidtion.  The
effluent from aeration goes  to  a  clariifer.   Approximately  14,000
gallons  per  day  of settled sludge is returned to the aeration basin
and 2,400 gallons per day is sent  to  drying  piles  on  site.   More
dilution  water  is  added  to  the  clarifier  when  needed for Total
Dissolved Solids control.  The overflow from the clarifier goes  to  a
chlorine  contact  tank  because   of the sanitary waste.  The blowdown
water from the'cooling tower and boilers is added to the final chamber
of the chlorine contact tank.  The effluent from this unit is sent  to
a final polishing pond for settling and equalization before discharge.
                                    762

-------
The by-product waste from the distillation column bottoms is sent to a
lined by-product pond at a rate of 14,000 gallons per day and held for
one  of two possible disposal methods.  When there is a market for the
by-products, the waste is concentrated and sold to the pulp and  paper
industry.   At  times  when  this  is not possible, and the pond nears
capacity, the waste is bled into the treatment system described  above
through the dilute waste holding pond.

Advanced Treatment Technologies

Practical  technologies  for  controlling COD include various forms of
mechanical  and  biological  oxidation.   For  the  relatively  simple
chemical  oxidation  of hydrosulfite to sulfate, intimate contact with
atmospheric  oxygen  is  effective,  using  submerged  air  diffusers,
induced  air  in a circulating system, or mechanical surface aeration.
For biochemical oxidation of  resistant  organics  such  as  formates,
phenols,  chlorinated hydrocarbons, and methanol, the use of trickling
filtration, rotating biological discs, or variations of the  activated
sludge process can provide intimate contact between organic pollutants
and the microbiological organisms which use them as food.

Technologies   for   controlling   heavy   metals   include   alkaline
precipitation, which is effective for the  common  heavy  metals,  and
sulfide  treatment,  which  precipitates nickel, zinc, and copper, but
does not increase control of chromium.  Other less  appropriate  metal
removal techniques have been discussed in Section 8.

Selection of Appropriate Technology and Equipment

Technologies for Different Treatment Levels

A.   Level 1

     Treatment system pH adjustment, biological  oxidation,  settling,
     and chlorination are used to reduce COD and coliform organisms in
     the  combined wastes, in accordance with existing plant practice.
     The flow diagram is shown in Figure 25-3.

B.   Level 2

     The  by-product  wastes  are  treated  separately   by   alkaline
     precipitation  to  remove  the toxic metals and then are combined
     with the product wastes for biological  oxidation  treatment  and
     chlorination, as in Level 1.

     If  an  acutal  formate  process  plant employs metal-free sodium
     formate in its process there is no reason to expect heavy  metals
     in  the  process  wastes  and  Level  2  treatment  should not be
     necessary.  The flow diagram is shown in Figure 25-4.
                                   763

-------
Equipment for Different Treatment Levels

A.   Equipment Functions

     Product waste and by-product wastes are received in a  mixed  and
     aerated  equalizing  basin, adjusted to a neutral pH, and aerated
     in a four-day aeration lagoon, including  50  percent  return  of
     underflow to the influent.  Plant sewage, nutrients, and diluting
     water  are added to the lagoon to promote biological oxidation of
     organics  and  other  COD.    Lagoon   effluent   is   clarified,
     chlorinated,  and  sent  to  a  polishing  pond  before discharge
     through effluent monitoring facilities.  Cooling tower and boiler
     blowdown wastes enter the system after chlorination,  since  they
     require no treatment except settling of scale and inert debris in
     the   polishing   pond.    Floating  aerators  are  used  in  the
     equalization basin and compressed air  is diffused in the  aerated
     lagoon,  for mixing and introduction of dissolved oxygen into the
     mixed liquor.

     In Level  2  treatment,  by-product  wastes  are  received  in  a
     separate  18-hour aerated and recirculated holding tank, which is
     pumped at average daily  flow  to  a   gravity  clarifier,  adding
     sufficient  lime  to  reach a pH of 10.5.  The clarifier overflow
     joins the product waste stream in the  equalization basin  of  the
     Level  1  system.   All features of the Level 1 system remain the
     same, since it  was  originally  sized to  handle  the  combined
     wastes.

B.   Chemicals and Handling

     SuIfuric acid, lime,  filter  aid,  and  chlorine  are  chemicals
     commonly  used  in  waste  treatment.   When handled in corrosion
     resistant equipment designed for their use,  no  unusual  hazards
     are   expected.    Raw  sewage  and  10-10-10  liquid  fertilizer
     introduced into the aerated lagoon become  thoroughly  mixed  and
     are  eventually  consumed  in  the  biological oxidation process,
     constituting no threat to operating  personnel.   Chlorine,  used
     for  control  of coliform bacteria, is received in ton containers
     and applied as a chlorine water solution using standard  solution
     feed  chlorination  equipment.   There are  no  unusual chemical
     handling problems in treating these wastes,  provided  the  waste
     streams are kept at a neutral or alkaline pH.

C.   Separation and Disposal of Solids

     In the Level 1 system, waste activated sludge solids are  assumed
     to  be dried in sludge beds at the site, to be'used as fertilizer
     for  plant  landscaping.   Clarifier   underflow   from   alkaline
     precipitation  of  by-product  waste   in Level 2 is assumed to be
     sent to a sludge holding tank and dewatered at suitable intervals
     in a filter press, followed by hauling of solids  to  a  chemical
     landfill.   Filter press filtrate is returned to the holding tank
     for retreatment.
                                   764

-------
                                 a
              M-ncnicr
              tAnB WlTHI
         I      MnuMf     uuinKH Nvm


1    All   L
JL    / '  ' j—\    •**    /—i
                                                          -J"
OMtPIM
                                                                                          CMJIB MM)
                                                                                          MUMMMXMI
                                                                                         RtWDOIOD
                                                                                                              •vumr
ui
                IneludM Elow Monitoring* ffl Monitoring atd
                        Figure 25-3.  Level 1 waste water treatment for sodium hydrosulfite subcategory.

-------
-j
a*
o*
EitnuuLT



HV'llK
                                          1
                                                      ute

                                                      a
                                                           I K-
UB I     •
                                                     *1   I  '

                                                     I~?H
                                                     •    i	
                                                                                         W-HCOET WSfE MUER
                                                                              uoumenMc
                                                                 saxx

                                                                UCUtDG
                                 USIUJUTION
                          fnclutkfa flaw iDniL^ring, (41 wutoring and ia^>l^r.
"t
/ 1
rtmatwr UILJT
JjJ


CM VKHK
AJ*
AB*TS1 IMXKM
j i
T


                                                                                                OBUWMMICM
                                                                                              —a-A
                                                                                                         HiJSJIItC li»ii
y^
                                                                                               •W9W SUWGC
                             Figxnre 25-4.   Level 2 waste water treatment  for sodiun hydrosulfite subcategory.

-------
D.   Monitoring Requirements

     Internal monitoring should  include simple   field   tests   for  pH,
     chlorine  residual,  and  settleable  solids.  Maintenance of the
     by-product stream  clarifier at   a  pH  of   10.5   is   expected   to
     provide  control   of  toxic metals without  need for  routine metal
     analyses.  Periodically, effluent samples should  be  analyzed  for
     chromium,  zinc, copper, nickel, and lead by atomic  absorption  in
     addition to routine COD  tests   for  general  evaluation  of  the
     treatment.

Treatment Cost Estimates

General Discussion

A  model  plant  concept  was  prepared  for  the  purpose of  the cost
estimates.  The specifications of the waste input parameters   and  the
design  of  the  model  plant Level 1  treatment system  are based on the
information presented for Plant  #672.

In this subcategory, commercial  fertilizer  and urea are  added   to
stimulate  growth of the biomass employed in biological treatment, and
not for direct reaction with a   residual  pollutant.    Therefore,  the
chemicals  used  do  not  bear   a  fixed  relationship to the  plant
production in units of  sodium hydrosulfite.

Organic solids generated in the  model treatment  system are assumed  .to
be disposed of on land  at the site, without a separate cost for sludge
disposal.

Cost Estimates

The  model  plant cost  estimate  for two levels of treatment applied  to
•the same level of production is  presented in Table 25-7.   Table  25-8
gives  a  summary  of the unit cost distribution between  amortization,
operation and maintenance cost components at two levels of treatment.

Cost estimates developed  for  the  first  and   the  second  level   of
treatment indicate that labor and supervision costs constitute a major
portion  of  the annual cost.  This reflects the manpower requirements
for operating the treatment systems on a 24-hour basis.

Basis for Guidance

Evaluation of Level 1 Treatment  Practices

There are two plants producing   sodium  hydrosulfite   by   the  formate
process, both of which  have Level 1 equipment in place and both are  in
compliance with their NPDES permits.

Level  1  technology has been specified as the technology presently  in
use at Plant #672.  Design and cost estimates are based   on  inclusion
of by-product wastes.
                                   767

-------
    TABLE  25-7.  MODEL PLANT TREATMENT COSTS
Subcategory
Production
Sodium Hydrosulfite - Formate process
20,450 metric tons per year
A. INVESTMENT COST

     Site development  	
     Equipment  	
     Monitoring equipment  ..

          Subtotal  ,.....
     Contractor's 0 &  Pb....

          Subtotal  	
     Engineering	

          Subtotal  	
     Contingencies	

          Subtotal  	
     Land	

     TOTAL INVESTMENT  COST  .


B. OPERATION AND
   MAINTENANCE COST

     Labor and supervision  .
     Energy	
     Chemicals  	,
     Ma intenance  	
     Taxes and  insurance  ...
     Residual waste disposal
     Monitoring,  analysis
      and reporting  	
                            Level 1

                            23,000
                            138,500
                            20,000

                            181,500
                            27,225

                            208,725
                            41,745

                            250,470
                            25,047

                            275,517
                            12,000

                            287,517
     TOTAL OPERATION AND
     MAINTENANCE COST
                            168,000
                             12,000
                              3,500
                             27,552
                              8,626
                                  0

                             15,000
                            234,677
                                                      ($)
Level 2

      0
121,000
      0

121,000
 18,150

139,150
 27,830

166,980
 16,698

183,678
  3,000

186,678
 84,000
  1,200
 18,500
 18,368
  5,600
      0

  7,500
135,168
C. AMORTIZATION OF
   INVESTMENT COST

   TOTAL ANNUAL COST
                             44,827

                            279,504
 29,884

165,053
a Represents the  incremental  cost above that for Level 1 treatment
  Overhead and profit
                              768

-------
        TABLE  25-8  MODEL PLANT UNIT TREATMENT COSTS
     Subcategory Sodium Hydrosulfite -Formate process
                                 Annual Treatment Costs  ($/kkg)
    COST ITEM
PRODUCTION
 (kkg/yr)
    LEVEL OF TREATMENT

Level 1    Level 2
Annual Operation
and Maintenance 20/450
Annual
Amortization 20,450
Total Annual
Cost 20,450
11.48 6.61
2.19 1.46
13.67 8.07
*Represents the incremental  cost  above Level 1.
                              769

-------
An evaluation of treatment practices was performed at Plant #672 based
on  the  pollutant  sampling,  since long-term monitoring data was not
available for the  pollutants  of  concern.   Details  concerning  the
performance  evaluation  calculations  and  assumptions  are discussed
below for^the pollutants of concern.

A.   Conventional and Nonconventional Pollutants

     1.   Chemical Oxygen Demand (COD)

          At the time of sampling,  the by-product  waste  (stream  #1,
          Figure  25-2)  was not flowing into the wastewater treatment
          system (Stream #3, Figure 25-2).

          Review of Table 25-4 indicates that a majority  of  the  COD
          load  is  contributed  by  the by-product waste stream.  The
          other major source of COD is contributed by the dilute waste
          stream #2.  Estimates of subcategory  performance  are  made
          for COD based on the following assumptions:

          a.   Assumption 1

               The COD load for the by-product stream must be included
               in the evaluation of the treatment system  performance,
               since  its contribution to the final COD load will have
               a significant influence.  Therefore, it is assumed that
               the percent COD removed in the treatment  system  would
               be  the  same  percent  COD  removed for the by-product
               waste stream as if it  had  received  treatment.   This
               assumption   is   necessary   since  plant  performance
               information is available  only  for  the  dilute  waste
               stream.

          b.   Assumption 2

               In order to estimate the COD removed  during  treatment
               in  the dilute waste stream, two minor assumptions must
               also be made to account for COD contributions from  the
               sodium  bisulfite  (SBS) and sanitary streams which are
               not considered sodium hydrosulfite process-related.  It
               is assumed that the final  COD  concentration  for  the
               treated sodium bisulfite waste stream is 680 mg/1  (from
               Table  24-16)  and  60  mg/1  which  is  a conservative
               estimate   for   treated   sanitary   wastes.     These
               assumptions  are  not critical since the total combined
               waste flow from these two waste sources  is  only  0.30
               mVkkg  compared  with  2.9  mVkkg  of  other  process
               related wastes.

               Table 25-9 is a summary of the subcategory  performance
               evaluation  of  TSS  and  COD  for Plant #672.  The COD
               evaluation is developed in the table on  the  bases  of
                                     770

-------
the assumptions above and sampling information in Table
25-4.

A  determination  of  achievable  effluent  COD load is
shown  in  the  following  steps  beginning   with   an
estimation  of  the COD removal efficiency in the waste
treatment system.

The observed effluent COD load is 3.6 kg/kkg from Table
25-4  which  includes  contributions  from  the  sodium
bisulfite  (SBS)  and  sanitary waste streams which are
not process related.  These  loads  are  determined  as
follows:

SBS load  = (680 mg/1) (0.10 mVkkg) (kg/m3)  = 0.068 kg/kkg
                                 (1000 mg/1)

(0.10 m3/kkg from Table 25-9; 680 mg/1 from Assumption
2 above)

Sanitary  waste COD load =

(60 mg/1) (0.24 mVkkg) {  kq/m3)  = 0.014 kg/kkg
                     (1000 mg/1)

(0.24 mVkkg from Table 25-9; 60 mg/1 from Assumption 2
above)

The  effluent  COD  load  contributed  by  the  SBS and
sanitary   waste  streams  are  substracted   from   the
observed   load  of  3.6 kg/kkg to obtain the actual COD
load contributed by the process related dilute waste as
follows:

3.6 kg/kkg - (0.068 kg/kkg +. 0.014 kg/kkg) =3.5 kg/kkg

The  effluent  COD  load  is  3.5  kg/kkg  which   when
expressed  as  a  ratio  with  the  raw  COD waste load
(Assumption 1) can be used to estimate  the  additional
COD contributed by the by-product waste as follows:

Raw COD load contributed = 29 kg/kkg from Table 25-4
by dilute waste

Raw COD load contributed = 74 kg/kkg from Table 25-4
by by-product waste

Effluent  COD load of = (74 kq/kkq) (3.5 kg/kkg)
                      (29 kg/kkg)

                     =8.9 kg/kkg
                      771

-------
TABLE 25-9.  SOBCA3EGORY PERFORMANCE EVALUATICN SUM4ABY AT PLANT #672 FOR
             CONVENTIONAL AND tCNOCNVENTXQNAL PCLLUTANTS IN THE EFFLUENTS

Effluent Haste Flow TSS
Description (m3/kkg) fno/ii
A - Dilute Haste
B - Sodium Bisulfite
Haste
C - Sanitary Haste
D - Dilution Hater
E - Boiler Slowdown
F - By-product
Total load (A+O+F)
Effluent Concentration
Model Plant Flow
(A4O+-F)
Concentration At
Model Plant Flow
BASIS OF UMTTATION
(1) - NA Not applicable
1.95
0.10

0.24
1.75
0.83
0.95

4.87

4.65

4.65
4.65
260
NA(1)

NA
NA
NA
61
NA
25

NA

26
25
(2)
(kq/kkg)
0.51
NA

NA
NA
NA
0.058
NA
NA

NA

NA
NA
OOD(2>
(rog/1) (kg/kkg)
1800 3.5
680 (3* 0.068
. .
60 0.014
0 0
0 0
9600 9.2
NA 13
NA NA

NA NA

2700 NA
NA 12
to evaluation.
(2) - Data based on average of
three


 (3) -
24- hour composite sanples.
Assumed value discussed in Section 25.7.1 under Chemical Oxygen Demand,
                                    772

-------
               Total  effluent COD load contributed by both the dilute
               and by-product waste =

               3.5 kg/kkg +8.9 kg/kkg = 12 kg/kkg

               The effluent load for COD is 12  kg/kkg  based  on  the
               plant performance evaluation and sampling data.

     2.   Total Suspended Solids (TSS)

          The removal of TSS from the  raw  wastewater  is  much  more
          complex  on  a  load  basis.   TSS removal must therefore be
          estimated on a concentration basis  as  indicated  in  Table
          25-9.   A  TSS  concentration of 25 mg/1 was observed in the
          treated effluent during sampling (Table 25-4) which is  used
          as the long-term average in developing the TSS guidance.

B.   Toxic Pollutants

     The removal of toxic pollutants in the treatment system at  Plant
     #672  during  sampling is indicated in Table 25-5 for the purpose
     of evaluating plant performance.

Basis for Level 1 Treatment Performance

A.   Technology Basis

     The technology basis is, or is equivalent  to,  equalization,  pH
     adjustment,   aeration   in   a   biological   oxidation. system,
     clarification, and chlorination before discharge of  the  treated
     effluent.

B.   Flow Basis

     The basis of flow used for the cost estimates, and as a basis  to
     estimate pollutant discharge loadings for the purpose of guidance
     development, was derived from plant information received at Plant
     #672.   Table 25-3 presents the unit flows from the three primary
     waste  sources  identified  in  the  industry.   The  dilute  and
     by-product wastewaters are primarily process related, whereas the
     dilution water is required for proper operation of the biological
     waste treatment system.

     There are only two plants which currently use the formate process
     for the manufacture of sodium hydrosulfite.  The model plant flow
     is  4.7 m3/kkg of product for the sodium hydrosulfite subcategory
     as presented in Table 25-3 and  is  based  on  Plant  #672  data.
     Plant   #672   was  chosen  for  evaluation  because  it  is  not
     complicated by other unrelated manufacturing processes.
                                    773

-------
Selection of Pollutants to be Controlled

The Level 1  treatment technology  is  directed  primarily  toward
removal  of  TSS  and COD.  In addition to these conventional and
nonconventional  pollutants,  toxic   organic   pollutants   were
identified . in  small  quantities.  These toxic organic polluants
include pentachlorophenol, phenol, and other trace organics.  The
presence  of  these   toxic   pollutants   is   currently   under
investigation  by  the  plant to:  1) determine the source of the
pollutants and identify whether they are process related, and  2)
determine   whether  process  modifications  or  best  management
practices might be available to eliminate their presence if  they
are discovered to be process related.

Basis of Pollutant Control Guidance

1 .    Conventional and Nonconventional Parameters
          PH

          The treated effluent is to  be  controlled
          range  of  6.0  to  9.0.   This  is  based
          presented in Appendix B  of  the  proposed
          Document (60) and the JRB Study (52).

          TSS and COD
within  the
on the data
Development
          The data presented in  Table  25-9  was  used  for  the
          development  of TSS and COD control guidance.  The data
          presented is for Plant #672 which is the only plant   in
          the  subcategory where the treatment performance can  be
          observed clearly.

          No long-term monitoring data is currently available for
          a statistical estimation of the variability  factors   in
          the Sodium Hydrosulfite Subcategory.

          Therefore,  the  30-day  average  and daily  variability
          factors are estimated from the treatability  study  (61)
          for  COD removal.  The study indicates a 30-day average
          variability factor of 1.2 and daily variability  factor
          of 4.0.

          The  maximum 30-day average COD level is estimated from
          plant performance data in Table 25-9 to be   12  kg/kkg.
          The  maximum 30-day average concentration is determined
          as follows:

          (12 kq/kkg)  (1000 mq/1) * 2600 mg/1
          (4.7 mVkkg) (kg/m3)
                               774

-------
The long-term average COD concentration  is  determined
using  the  30-day average variability factor of 1.2 as
follows!

2600 mq/1 = 2200 mg/1
1 .2

The 24-hour maximum concentration is  determined  using
the daily variability factor as follows:

(2200 mg/1) (4.0) = 8800 mg/1

The 24-hour maximum loading for COD is then:

(8800 mg/1)) (4. 7 mVkkg)  (kg/in*)  = 41 kg/kkg
                        (1000 mg/1)

The long term average TSS concentration is 25 mg/1 from
plant  performance  information in Table 25-9.  The TSS
control guidance is based on this long-term average and
the variability  factors  used  for  COD.   The  30-day
average  variability factor of 1.2 is used to determine
the  TSS  maximum  30-day  average   concentration   as
follows:
                          concentration  is  determined
          (25 mg/1) (1.2) = 30 mg/1

          The   24-hour   max imum
          similarly as follows:

          (25 mg/1) (4.0) = 100 mg/1

          The maximum 30-day average  loading  is  determined  as
          follows:

          (30 mg/1) (4.7 mVkkg)   (kq/m3)   =0.14 kg/kkg
                                 (1000 mg/1)

          The 24-hour maximum loading is:

          (100 mg/1) (4.7 mVkkg}) (  kq/m3)    =0.47 kg/kkg
                                  (1000 mg/1)

          Guidance for effluent control is presented in Table 25-
          10.

2.    Toxic Organic Pollutants

     The verification sampling results presented  in  Table  25-5
     for pentachlorophenol and phenol indicate that both of these
     toxic  organic  pollutants  are  currently  removed  by  the
     existing treatment system to the analytical detection  limit
     and are therefore excluded from further consideration.
                     775

-------
                 T&BLE 25-10.  GUIDANCE FOR EFFLUENT CONTROL
                             Sodium Hydrosulf ite
                             Level 1 Performance
                       Waste Water Flow:  4.7 m3/kkg

Pollutant
Conventional and
Subcategory ,,*
Performance VFR
(mg/1)
nonconventional
Concentration
Basis (mg/1)
Max
30-day 24-hr
Avg Max

Effluent Loading
(kg/Kkg)
Max
30-day 24-hr
Avg Max

pollutants:

Total Suspended
 Solids, TSS

Chemical Oxygen
 Demand, COD
  25(2)   4.0/1.2(3)   30
2200     4.0/1.2(3)   2600
 100
8800
 0.14
12
           (2)
0.47
41
(1) - VFR:  ratio of the 24-hour variability factor to the 30-day average
            variability factor.

(2) - Based on subcategory performance estimates utilizing three 24-hour
      composite samples.

(3) - Based on treatability study  (61) for the removal of COD.

(4) - No effluent limitation has been established.
                                      776

-------
     3.   Toxic Metal Pollutants

          The Level 1 treatment technology  is  not  amenable  to  the
          removal  of  toxic  metal  pollutants  although, as shown in
          Table  25-5,  considerable  incidental  removal  can  occur.
          Level  1  technology can not reasonably ensure the removal of
          toxic metals on a consistent basis.

Basis for Level 2 Treatment Performance

A.   The Application of Advanced Level Treatment

     The Agency has analyzed the cost effectiveness of the base  level
     systems and the various advanced level options for the removal of
     pollutants  based on cost estimates presented in this report.  No
     plant has this additional technology installed.

B.   Technology Basis

     Level 2 is  based  on  treatment  that  provides  more  stringent
     removal  of  toxic  pollutants  in the by-product waste stream by
     introducing alkaline precipitation with lime and  settling  prior
     to  base  level  treatment.   The by-product waste stream was the
     primary  source  of  toxic  metal  pollutants   observed   during
     sampling.

C.   Flow Basis

     The unit flow used for the  treatment  performance  estimates  is
     based  on  4.7  m3/kkg  of  product.  The estimated flow does not
     change for Level 1 and Level 2 treatment.

D.   Selection of Pollutants to be Controlled

     Selection of pollutants is based on an evaluation of verification
     sampling data at Plant #672.  Results of the sampling program are
     presented in this section.  On the  basis  of  concentration  and
     total  annual  raw waste loads, zinc, nickel, lead, chromium, and
     copper  have  been  identified  at  treatable  levels  and   were
     considered as candidates for control in addition to TSS and COD.

     Review  of  Section  8  indicates  that  not all toxic pollutants
     present in raw waste effluents need to be specified for  alkaline
     precipitation.   In  fact,  two  primary  groups  of heavy metals
     appear to reach a minimum solubility  at  slightly  different  pH
     levels.   Zinc  is  the major toxic pollutant determined at Plant
     #672 which represents one of these primary groups of  metals  and
     nickel, also identified at treatable levels, represents the other
     group that achieves optimum removal at slightly higher pH values.
     Adequate  control  of zinc and nickel can be achieved in a single
     step alkaline precipitation system operated in an intermediate pH
     range of approximately 9.5  to  10.5.   The  basis  for  this  is
     presented  in  Section  8  of  this  report.  Control of zinc and
                                   777

-------
     nickel will also ensure removal of  lead,  chromium,  and  copper
     which  may  occasionally occur at treatable concentrations in the
     raw waste.

E.   Basis of Pollutant Control Guidance

     1.    Nonconventional Pollutants

          The only nonconventional pollutant selected  for  limitation
          is  COD.  In view of the proposed technology for Level 2, no
          additional removal of COD  is  anticipated  beyond  what  is
          already  estimated  for Level 1.   That section discusses the
          development of the COD control guidance.

          The maximum 30-day average COD effluent loading is 12 kg/kkg
          and the 24-hour maximum is  41  kg/kkg  presented  in  Table
          25-11.

     2.    Toxic Pollutants

          Alkaline precipitation and settling of the by-product  waste
          is  expected  to  remove  the  five  candidate  toxic  metal
          pollutants to within the limits of treatability.  Review  of
          Table  25-5  indicates  that  the existing Level 1 treatment
          system is providing incidental removal of the toxic  metals.
          Table  8-12 presents the estimated levels achievable for the
          toxic metal  .pollutants  based  on  literature  treatability
          which was used for the purpose of establishing the long-term
          average concentration used for guidance control.

          No  long-term  pollutant  monitoring  data were available to
          develop  the  variability  factors.   However,  treatability
          studies (61) were performed that estimate the 30-day average
          and   24-hour  maximum  variability  factors  for a number of
          toxic  pollutants.   Variability  factors  for   zinc   were
          estimated  as  the  average of values obtained in the nickel
          and copper sulfate industries.  The 30-day average and daily
          variability factors are 1.2 and 4.1, respectively,  and  are
          used  for guidance development.

          a.    Zinc

                Review of the zinc concentration in the raw  by-product
                waste stream indicates levels as high as 27 mg/1 and an
                average of 24 mg/1 from three 24-hour composite samples
                (Table  25-6).   Literature  treatability  presented in
                Table 8-12 indicates an  achievable  long-term  average
                concentration  of  0.80 mg/1 for alkaline precipitation
                and settling  of  zinc.   The  maximum  30-day  average
                concentration is developed as follows:

                (0.80 mg/1)  (1.2) = 0.96 mg/1
                                   778

-------
                 TABLE 25-11.  GUIDANCE FOR EFFLUENT CONTROL
                             Sodium Hydrosulf ite
                             Level 2 Performance
                        Waste Water Plow:  4.7 m3/kkg





Conventional and
pollutants ;
Chemical Oxygen
Demand, COD
Subcategory f 1 ^

(mg/1)


nonconventional

2200 4.0/1.2(5)
Concentration
Basis Cmg/1)

Max
30-day 24-hr
Avg Max


2600 8800
Effluent Loading
(kg/kkg)

Max
30-day
Avg


12(2)


24-hr
Max


41
Toxic pollutants;
Zinc
Nickel (6)
Lead(6>
^ . (6)
Chromium
Copper(6)
0.8010' 4.1/
0.40(3) 4. 1/
0.15(3) 4. I/
0.32(3) 4. I/
0.40(3) 4.1/
'1.2v:j' 0.96
1.2(5) 0.48
1.2(5) 0.18
1.2(5) 0.38
1.2(5) 0.48
3.3
1.6
0.62
1.3
1.6
0.0045
0.0023
f 4 )
f 4 1
_J4>
0.016
0.0075
— <4>
_J4)
— W)
(1) - VFR


(2)
            ratio of the 24-hour variability factor to the 30-day
            average variability factor.
      Based on subcategory performance estimates utilizing three 24-hour
      composite samples.
(3)  - The long-term average concentration estimate from Table 8-12 is used.

(4)  - No control specification is needed.
(5)  - Based on treatability study (61)  for nickel'and copper sulfate
      subcategor ies.
(6)  - Also applicable for pretreatment.
                                      779

-------
The   24-hour   maximum
developed as follows:
concentration   is  similarly
(0.80 mg/1) (4.1) = 3.3 mg/1

The maximum 30-day average effluent  loading  for  zinc
is:

(0.96 mg/1) (4.7 mVkkg) (kq/m3) = 0.0045 kg/kkg
                      {1000 mg/1)

The 24-hour maximum loading for zinc is:

(3.3 mg/l)(4.7 mVkkg) (kq/m») = 0.016 kg/kkg
                     (1000 mg/1)

Guidance  for  effluent  control  is presented in Table
25-11.

Nickel

The concentration of nickel was observed as high as 1.7
mg/1 in the raw by-product waste  stream  and  averaged
1.1  mg/1 in the three 24-hour composite samples (Table
25-6).   Literature treatability presented in Table 8-12
indicates an  estimated  achievable  long-term  average
concentration of 0.40 mg/1.  The maximum 30-day average
concentration is as follows:

(0.40 mg/1) (1.2) * 0.48 mg/1

The   24-hour   maximum   concentration   is  similarly
developed as follows:

(0.40 mg/1) (4.1) = 1.6 mg/1

The maximum 30-day average loading for nickel is:

(0.48 mg/1) (4.7 mVkkg)  (kq/m3) = 0.0075 kg/kkg
                       (1000 mg/1)

The 24-hour maximum loading for nickel is:

(1.6 mg/1) (4.7 mVkkg)  (kq/m3)  = 0.0075 kg/kkg
                       (1000 mg/1)

Guidance for effluent control  is  summarized  in  Table
25-11 for Level  2 treatment.

Other Pollutants

The concentration basis for lead, chromium, and  copper
are    also    presented    in   Table    25-11.    These
                      780

-------
concentrations are intended to  serve  as  guidance  in
cases  where  these toxic pollutants are found to be of
serious concern.
                    780-A

-------

-------
                              SECTION 26

                        EXCLUDED SUBCATEGORIES
Aluminum Sulfate

Summary of Determinations

It has been determined  that  no  further  effort  will  be  given  to
developing  revised  BAT  or NSPS for this subcategory.  The basis for
this recommendation is that there is a zero  discharge  regulation  in
effect for BAT and NSPS and it controls toxic pollutants.

Production Process and Effluents

Aluminum  sulfate  is  produced  by  the  reaction of bauxite ore with
concentrated sulfuric acid.  Ground ore and  acid  are  reacted  in  a
digester yielding aluminum sulfate in solution plus muds and insoluble
wastes.   The  aluminum sulfate is sold as a solution or evaporated to
produce a solid product.  Waste muds are ponded to allow settling  and
the  clear liquor is returned to the process.  Wastes from washing and
leaks are directed to the pond  and  also  returned  to  the  process.
Toxic  pollutants  in  the  pond  include  zinc,  copper, chromium and
cadmium.

Plants

There are 82 aluminum sulfate producing facilities in the industry.

BPT Limitations

BPT limitations were promulgated March 12, 1974 (40 CFR 415.20).   The
limitations  provide  for  zero discharge of process wastewater except
that if the pond has sufficient volume  to  hold  a  10-year,  24-hour
storm,  the  amount  of  water  equal  to  the  precipitation less the
evaporation may be discharged.  The water must have a pH of 6.0 to 9.0
and average less than 25 mg/1 of suspended solids.

BAT, Pretreatment and NSPS Limitations

BAT and NSPS limitations were promulgated on March 12,   1974  (40  CFR
415.23  and  415,25).    The  limitations provide for zero discharge of
process wastewater except that if the pond has  sufficient  volume  to
hold  a 25-year, 24-hour storm, the volume of precipitation that falls
within the  pond  in  excess  of  a  25-year,  24-hour  storm  may  be
discharged.   These  zero discharge limitations adequately control the
toxic pollutants.  Development  of  Pretreatment  Standards  has  been
deferred to Phase II.
                                 781

-------
Ammonium Chloride

Summary of Determinations

It  has  been determined that no further effort be given to developing
or revising BAT, NSPS, or Pretreatment regulations and the subcategory
is excluded under Paragraph 8 of the Settlement Agreement.  The  bases
for  this  determination  are:   1) Only one of the major producers of
ammonium chloride uses  the  Solvay
recovered  as  a  by-product.   2)
         Process.    Ammonium  chloride  is
        No  toxic  pollutants were found at
significant concentrations
ammonium chloride plant.
in  the  waste  during  screening  of  one
Production Process and Effluents

Ammonium  chloride  is  used in the manufacture of dry cell batteries,
explosives, dyes, washing powder, soldering  flux,  chemical  reagent,
and  as  a  medicinal  additive to livestock feed.  It is also used in
pharmaceutical preparations and freezing mixtures.

Ammonium chloride is produced by three methods.  A major portion is  a
by-product  in  the  manufacture  of  sodium  carbonate  by the Solvay
process.  The wastes produced are associated with the sodium carbonate
subcategory.  A second  process  prpduces  ammonium  chloride  by  the
reaction  of  hydrogen  chloride  with  ammonia.  Discharges from this
process come from crystallization and wet scrubber operations.
The industry profile data for this
26.2-1.

Toxic Pollutants
        subcategory  are  given  in  Table
Data  have  been  received  on  about  50 percent of the industry as a
result of Section 308 letters.  In  additon,  a  sampling  survey  for
toxic  pollutants  was  made  at  one  plant.   The results of the 308
letters and the sampling survey indicate that no toxic pollutants  are
being  discharged  in significant quantities.  Ammonia was found to be
the only pollutant  of  significance.   Since  ammonia  is  adequately
controlled  by  the  existing  BPT  regulation  40  CFR  415.242, this
subcategory is being excluded under  Paragraph  8  of  the  Settlement
Agreement.

Pollutants found during sampling at one plant are:
          Pollutant

          Chromium

          Nickel

          Zinc

          Ammon i a
  Concentration

  29 ng/1 (max.)

  25 cg/1 (max.)

  29 *ig/l (max.)

  104 mg/1 (avg.)
                                     782

-------
 TABLE  26.2 - 1
            PROFILE DAIA SUMMARY
 SUBCA2EGOEY
AMMONIUM CHLORIDE
 Total subcategory capacity rate
 Total subcategory production rate
 Number of plants in this subcategory
 308 Data on file for
     With total capacity of
     With total production of
     Representing capacity
     Representing production
     Plant production range:
             Minimum
             Maximum
     Average production
     Median production
     Average capacity utilization
     Plant age range:
             Minimum
             Maximum
     Waste water flow range:
             Minimum
             Maximum
     Volume per unit product:
             Minimum
             Maximum
                                NA
                                NA
                                6
                                3
                           52,400 kkg/year
                           29,800 kkg/year
                                NA
                                NA

                            4,600 kkg/year
                           13,400 kkg/year
                                NA
                                NA
                                NA

                               17 years
                               43 years

                               NA
                               NA
                               NA
 Sources of data are Stanford Research Institute,  Directory of Chemical
 Producers, U.S.A.,  1977, U.S,  Department of Ocmrerce, Current Industrial
 Reports, December 1977; Energy and Environmental  Analysis, Inc.;  Draft
 Report, "Preliminary Economic  Assessment of Effluent Limitations  in the
 Inorganic Chemical  Industry, "  June, 1978 and "Economic Analysis of Proposed
Revised Effluent Guidelines and Standards for the Inorganic Chemicals
Industry," March,  1980
     Not Available
                                   783

-------
Status of Regulations

BPT regulations (40 CFR 415.242) are in effect for this subcategory.

Ammonium Hydroxide

Summary of Determinations

It  has  been determined that no further effort be given to developing
BPT,  BAT,  NSPS,  and  Pretreatment  regulations  for  the   Ammonium
Hydroxide  Subcategory.  The bases for this determination are:  1) the
process has no toxic pollutants as reactants, and 2) no direct process
waters are discharged from manufacturing operations.  The  subcategory
is excluded under Paragraph 8 of the Settlement Agreement.

Production Processes and Effluents

Ammonium  hydroxide  is  used predominately as a chemical intermediary
and reagent.  It is also used in the dyeing and bleaching of  fabrics,
the  production of ammonium salts and aniline dyes, and the extraction
of alkaloids from plants.

The most  common  method  of  ammonium  hydroxide  production  is  the
modified  Haber-Bosch  process,  wherein  hydrogen  and  nitrogen  are
reacted directly  over  a  catalyst  surface  to  form  ammonia.   The
hydroxide  is  formed  by  adding  water.  The only process wastewater
source is derived from equipment washing.

The industry profile for this subcategory is given in Table 26.3-1.

Toxic Pollutants

Data was received on six of seven plants as a result of  308  letters.
In  addition,  a  sampling  survey  was  made at one plant which  had  a
potential for discharge.  However,  no  process  water  discharge  was
found at the facility.  There are low volume discharges as a  result of
spills  and  washdowns.   The amount discharged was such that a sample
could not be obtained for analysis.

Status of Regulations

Because no significant quantities of toxic pollutants are present,  no
further   effort   will   be  given  to  development  of  pretreatment
regulations for this subcategory.  Subpart Y.  has  been  reserved  for
this subcategory.

Barium Carbonate

Summary of Determinations

It  has  been determined that no further effort be given to developing
or  revising BPT, BAT,  NSPS,  and  Pretreatment  regulations  for   the
Barium  Carbonate  Subcategory.   The  basis for this determination is
                                  784

-------
 TABLE  26.3-1 -
SUBCMEGOH* PRCFUE DKTA SUMMARY
 SUBCATEGORY
AMVDNIUM HYDROXIDE
 Total subcategory capacity rate
 Total subcategory production rate
 Number of plants in this subcategory
 308 Data on file for
     With total capacity of
     With total production of
     Representing capacity
     Representing production
     Plant production range:
             Minimum
             Maximum
     Average production
     Median production
     Average capacity utilization
     Plant age range:
             Minimum
             Maximum
     Waste. Water flew range:
             Minimum
             Maximum
     Volume per unit product:
             Minimum
             Maximum
                              NA
                              NA
                               6

                           41,800 kkg/year
                           17,000 kkg/year
                              NA
                              NA

                              206 kkg/year
                            9,500 kkg/year
                              NA
                              NA
                              NA


                               10 years
                               26 years

                              NA
                              NA

                              NA
                              NA
 Sources of data are Stanford Research Institute, Directory of Chemical
 Producers, U.S.A., 1977, U.S. Department of Connerce. Current Industrial
 Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
 Report, "Preliminary Economic Assessment of Effluent Limitations in the
 Inorganic Chemical Industry, " June, 1978 and "Economic Analysis of Proposed
Revised Effluent Guidelines and Standards for the Inorganic Qiemicals
Industry," March, 1980
N&=  Not Available
                                   785

-------
that the small quantities of toxic pollutants found  during  screening
are far below accepted treatability levels.

Production Processes and Effluents

Barium carbonate is used in glass manufacturing, as a flux in ceramics
and  enamelling,  as an intermediate in the production of barium oxide
and hydroxide, and as a coating for photographic paper.   It  is  also
used in the synthetic dyestuff industry and for the removal of soluble
sulfate in brick manufacturing.

Barium sulfide solution is reacted with soda ash to precipitate barium
carbonate.   The  reacted  solution  is  filtered.  The filter cake is
washed, dried, and calcined.   Wastewater  results  from  filter  cake
washing,  leaks  and  spills.   The  industry  profile  data  for this
subcategory are given in Table 26,4-1.

Toxic Pollutants

Data has been received on about 70 percent of the industry as a result
of Section 308 letters.  In addition,  a  sampling  survey  for  toxic
pollutants  was made at one plant.  The results of the 308 letters and
the sampling survey  indicate  that  no  toxic  pollutants  are  being
discharged in significant quantities.

The  maximum  concentration  found  the raw waste load in sampling for
this subcategory were:
          Pollutant

            Nickel
Concentration. (»q/l

        21
            Zinc                       68

Status of Regulations

Subpart Z has been reserved for this subcategory,

Borax

Summary of Determinations

It has been determined  that  no  further  effort  will  be  given  to
developing revised BAT and NSPS regulations for the Borax Subcategory.
The  basis  for  this  determination  is that existing BPT regulations
specify zero discharge of process wastewater pollutants  to  navigable
waters.   Development of pretreatment regulations  is deferred to Phase
II.

Production Processes and Effluents

Borax  is produced by dissolving sodium borate ores in recycled  mother
liquors  and water.  The insolubles settle out  in ponds or are removed
                                 736

-------
TABLE  26.4-1 -
SUBCKTEGOEY PRCFILE DATA .SUMMARY
SUBCATEGORY
BARIUM CARBONATE
Total subcategory capacity rate
Total subcategory production rate
Number of plants in this subcategory
308 Data on file for
    With total capacity of
    With total production of
    Representing capacity
    Representing production
    Plant production range:
            Minimum
            Maximum
    Average production
    Median production
    Average capacity utilization
    Plant age range:
            Minimum
            Maximum
    Waste Water flow range:
            Minimum
            Maximum
    Voluna per unit product:
            Minimum
            Maximum
                               NA
                               NA
                                7
                                5
                           57,000 kkg/year
                           48,745 kkg/year
                               NA
                               NA

                              158 kkg/year
                           26,190 kkg/year
                               NA
                               NA
                               NA

                                9 years
                               24 years
                               NA
                               NA
                               NA
                               NA
Sources of data are Stanford Research Institute, Directory of Chemical
Producers, U.S.A., 1977, U.S. Department of Commerce, Current Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry, " j^e,  1978 and "Economic Analysis of Proposed
Revised Effluent Guidelines and Standards for the Inorganic Chemicals
Industry," March, 1980
NA=  Not Available
                                   787

-------
by thickeners, and the clarified borax solution (mother liquor) is fed
to crystallizers where a slurry of  borax  crystals  is  formed.   The
borax  is  separated from the water by centrifugation, dried, screened
and packaged.  Process effluents are recycled  with  excess  going  to
evaporation ponds or returned to source.
Plants

Three  plants  produce borax in the United States.
total recycle of wastewater.

BPT Limitations
All three practice
BPT limitations were promulgated on May 22, 1975 (40 CFR 415.272), and
require no discharge of wastewater pollutants to navigable waters.

BAT and NSPS Limitations

BAT and NSPS limitations were proposed on May  22,   1975.   They  were
never promulgated.  Since BPT already requires zero discharge, BAT and
NSPS are being excluded under Paragraph 8 of the Settlement Agreement.

Boric Acid

Summary of Determinations

It  has  been determined that no further effort be given to developing
or revising BAT, NSPS, or Pretreatment regulations for the boric  acid
industry.   The basis for this determination is that there is only one
plant which manufactures boric acid *from  mined  ore.   There  is  an
indication  that  this  plant  will  discontinue operation.  All other
plants manufacture boric acid using the Trona process  and  have  zero
discharge.   This  subcategory  is  excluded  under Paragraph 8 of the
Settlement Agreement.

Production Processes and Effluents

Boric acid is manufactured by acidification of borax.  It is  used  in
the   manufacture   of   chromic  oxide,  glazes,  enamels,  textiles,
fiberglass, and heat resistant glass.  It is also used medicinally  as
a  mild  antiseptic and in atomic power plants as a nuclear moderator.
Process wastes may conist of excess boric acid  liquor,  waste  sodium
sulfate by-product liquor and filtration impurities.

The industry profile data is given in Table 26.6-1.

Toxic Pollutants

Toxic  pollutants found at significant concentrations during screening
of one plant were:
                                  788

-------
          Pollutant

          Copper

          Thallium

          Zinc

          Bis(2-ethylhexyl
              phthalate

          Mercury
Concentration (nq/1

       340

       140

      1200


       530

         1 .6
There is an indication that this plant will discontinue manufacture of
boric acid.  All other plants have zero discharge because of  the  use
of a different process.

Status of Regulations

BPT  limitations were promulgated on May 22, 1975 (40 CFR 415-272) for
this subcategory.

Bromine

Summary of Determinations

It has been determined  that  no  further  effort  will  be  given  to
developing  or  revising  BAT  and  NSPS  regulations  for the Bromine
Subcategory.  The basis for this recommendation is that  existing  BPT
regulations  specify zero discharge of process wastewater to navigable
waters.  Development of pretreatment is deferred to Phase II.

Production Processes and Effluents

Most bromine is produced from brines pumped from brine wells.  A small
amount {1 percent) is produced from  brines  from  Searles  Lake  near
Trona,  California.    This  is  not a navigable water in that it is 35
percent solids.  The brine, after appropriate dilution  and  degassing
is  extracted by debromination with chlorine and steam.  The steam and
bromine is condensed, separated and distilled to obtain bromine.   The
raw waste load from the process is the residual brine and the chloride
salts  forced  when the chlorine replaces the bromine.  The raw wastes
are returned to the brine well or brine source.
Plants

There are nine plants producing bromine in the United States,
which return their wastes to the brine source.

BPT Limitations
                              all  of
Regulations  were  promulgated  on  May  22,  1975,  (40  CFR 415.292)
requiring zero discharge of process wastewater pollutants to navigable
                                  789

-------
TABLE  26.6-1 -
SUBCATEQORy PROFILE DATA SUMMARY
SUBCATBGOK?
BORIC ACID
Total subcatecpry capacity rate
Total subcategory production rate
Number of plants in this subcategory
308 Data on file for
    With total capacity of
    With total production of
    Representing capacity
    Representing production
    Plant production range:
            Minimum
            Maximum
    Average production
    Median production
    Average capacity utilization
    Plant age range:
            Minimum
            Maximum
    Waste water flow, range:
            Minimum
            Maximum
    Volume per unit product:
            Minimum
            Maximum
                               NA
                           122,600 kkg/year
                                3
                                2
                           97,500 kkg/year
                           93,850 kkg/year
                               NA
                               77 percent

                           30,156 kkg/year
                           63,694 kkg/year
                               NA
                               NA
                               30
                                  years
                               83 years
                              NA
                              NA

                              NA

                              NA
Sources of data are Stanford Research Institute, Directory of Chemical
Producers, U.S.A., 1977, U.S. Department of Commerce, Current Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry, " June, 1978 and "Economic Analysis of Proposed
Revised Effluent Guidelines and Standards for the Inorganic Chemicals
Industry," March, 1980
NA=  Not Available
                                   790

-------
waters except that residual brine depleted liquor may be  returned  to
the  body  of  water  from  which  the  brine  solution was originally
withdrawn.  In no case is the brine source  a  navigable  water.   The
source  is  wells  except for a small portion that comes from a "lake"
having 35 percent dissolved solids.

BAT and NSPS Limitations

BAT and NSPS were proposed on  May  22,  1975,  but  never  finalized.
Since  BPT  already  requires  zero  discharge, BAT and NSPS are being
excluded under Paragraph 8 of the Settlement Agreement.

Calcium Carbide

Summary of Recommendations

It  has  been  determined  that  no  additional  effort  be  given  to
developing revised BAT and NSPS regulations for this subcategory.   The
basis  for  this  recommendation is that BPT, BAT and NSPS regulations
specify zero discharge of process wastewater pollutants.  Pretreatment
standards will be developed under Phase II.

Production Processes and Effluents

Calcium carbide is manufactured by reaction of calcium oxide and coke.
Calcium oxide and dried coke are reacted in a furnace and the  product
is  cooled,  crushed,  screened,  packaged  and  shipped.   There  are
generally no process related wastewaters except that one plant  had  a
wet scrubber discharge.

Plant

There are five plants producing calcium carbide.

BPT, BAT and NSPS Limitations

BPT,  BAT  and NSPS regulations were promulgated on March 12, 1974 (40
CFR 415.32, 415.33 and 415.35).  All subparts require  zero  discharge
of  process  wastewater  pollutants.   It has been determined that the
calcium carbide subcategory  will  be  excluded  from  development  of
revised  BAT  and  NSPS  limitations  because  the  operations are now
subject to zero discharge regulations.

Calcium Carbonate

Summary of Determinations

It has been determined that no further effort be given  to  developing
or  revising  BAT,  NSPS, and Pretreatment regulations for the Calcium
Carbonate Subcategory.  The bases  for  this  determination  are:    1)
there are only four plants manufacturing calcium carbonate, and 2) the
small  quantities of pollutants found during screening were at or very
                                   791

-------
near detectable levels of  analysis.   This  subcategory  is  excluded
under Paragraph 8 of the Settlement Agreement.

Production Processes and Effluents

Calcium  carbonate is manufactured both in pure and impure form and is
used extensively in many industries.  In the pure form, it is used  in
the rubber, paint, cement,  paper and pharmaceutical industries.

In  one  process,  slaked  lime  is reacted in slurry form with carbon
dioxide.  The slurry is then screened  and  filtered.   The  recovered
product is dried, milled and packaged for sale.  The waste liquor from
the   filtration   step   is  recycled  or  discharged,  depending  on
requirements.  The coarse materials recovered from the screening  step
are discharged.

The  second process is based on waste streams from the Solvay Process.
A solution of sodium carbonate and sodium bicarbonate  from  the  soda
ash plant is reacted with waste calcium chloride liquor which has been
treated  through  a  settler.  The calcium carbonate produced together
with by-product sodium chloride  and  unreacted  calcium  chloride  is
pumped  to  a thickener.  The overflow from the thickener is collected
with plant drainage streams in a sump  to  which  soda  ash  finishing
wastewater  is  added,  precipitating  calcium  carbonate.  This mixed
stream then goes to waste collection.  The calcium carbonate underflow
is filtered, washed, atomized with steam,  dried  in  a  spray  dryer,
collected in a particle collector and packaged for sale.

An  ultrafine  grade  of  calcium  carbonate  is produced in a similar
manner to that described above with some additional polish  filtering,
tunnel  drying  and milling.  At each plant, the neutralized brine and
process wastewater are returned  to  the  brine  cavity.   No  process
wastewater is discharged.

The industry profile for this subcategory is given in Table 26.9-1.

Toxic Pollutants
There  are  four  plants  producing  calcium  carbonate  in the United
States.  One discharges to a POTW.  Data has been  received  on  three
plants  as  a  result of Section 308 letters.  In addition, a sampling
survey for toxic pollutants was made at  one  plant  which  represents
approximately  50  percent of total industry capacity.  The results of
the 308 letters  and  the  sampling  survey  indicate  that  no  toxic
pollutants  are  being  discharged.  The sampling survey results found
pollutant levels below treatability levels.

Maximum concentration of toxic pollutants found in raw waste were:
                                   792

-------
          Pollutant
          Nickel
Concentration
      21
          Zinc
Status of Regulations
      68
Interim final regulations (40 CFR 415.302) were promulgated on May 22,
1975.  These regulations require conrol of pH and suspended solids for
both the Solvay and lime process.  No change  in  the  regulations  is
needed.

BAT  and  NSPS  regulations  (40 CFR 415.303) were proposed on May 22,
1975.  These regulations were never finalized.  It has been determined
that  the  Calcium  Carbonate  Subcategory  be   excluded   from   the
development  of  BAT  and  NSPS  limitations  under Paragraph B of the
Settlement Agreement for the following reasons:  There are  only  four
plants  manufacturing  calcium  carbonate  and  the  308  letters  and
sampling survey indicate that no toxic pollutants are being discharged
in significant quantities.

Because no significant quantities of toxic pollutants are present,  no
further   effort   will   be  given  to  development  of  pretreatment
regulations for this subcategory.

Calcium Chloride

Summary of Determinations

It has been determined that no further effort be given  to  developing
or  revising  BAT  or  NSPS for the Calcium Chloride Subcategory.  The
bases for this determination are:  1) there are existing BAT and  NSPS
regulations  that  prohibit discharge of process wastewater pollutants
from the brine extraction process,  and 2) there  is  only  one  Solvay
process plant in the United States where calcium chloride is recovered
as a by-product.

Production Processes and Effluents

There  are • two processes for the manufacture of calcium chloride.  In
the first and major production process, calcium chloride is  extracted
from  natural  brines.  The salts are solution mined and the resulting
brines  first  are  concentrated  to   remove   sodium   chloride   by
precipitation  and then purified by the addition of other materials to
precipitate sodium, potassium,  and  magnesium  salts.   The  purified
calcium  chloride  brine is then evaporated to yield a wet solid which
is flaked and calcined to a dry solid product.  The second process  is
the  Solvay  Process,  which  is primarily used for the manufacture of
soda ash.  In the Solvay Process, calcium chloride is recovered  as  a
by-product.   All  the wastes from the process are associated with the
sodium carbonate subcategory.
                                  793

-------
 TABLE 26.9-1  -
SUBCATEGOE* PROFILE: DAIA SUMMMff
 SUBCKEEGOFY
CALCIUM CARBONATE
 Total subcategory capacity rate
 Total subcategory production rate
 Number of plants in this subcategory
 308 Data on file for
     With total capacity of
     With total production of
     Representing capacity
     Representing production
     Plant production range:
             Mininun
             Maxinum
     Average production
     Median production
     Average capacity utilization
     Plant age range:
             Minimum
             Maxiitun
     Waste water flew range:
             Minimum
             Maximum
     Volume per unit product:
             Minimum
             Maxinun
                               NA
                          129,600 kkg/year
                               NA
                                3
                           81,300 kkg/year
                           72,400 kkg/year
                               NA
                               56 percent

                              555 kkg/year
                           49,800 kkg/year
                               NA
                               NA
                               NA

                               25 years
                               50 years

                               NA
                               NA

                               NA
                               NA
 Sources of data are Stanford Research Institute,  Directory of Chemical
 Producers, U.S.A.,  1977, U.S. Department of Commerce,  Current Industrial
 Reports, December 1977;  Energy and Environmental  Analysis, Inc.;  Draft
 Report, "Preliminary Economic Assessment of Effluent Limitations  in the
 Inorganic  Chemical Industry* " June, 1978 an
-------
Plants

There are 11 plants producing calcium chloride in the  United  States,
one of which recovers it as a by-product from the Solvay Process.

Status of Regulations

Existing  regulations  for  calcium  chloride  (40  CFR 415.4) include
regulations for BAT and NSPS that  prohibit  discharge  of  wastewater
pollutants from the brine process.

Calcium Hydroxide

Summary of Determinations

It  has  been determined that no further effort be given to developing
or revising BAT or NSPS for the Calcium  Hydroxide  Subcategory.   The
basis  for  this  determination  is  that  an  existing BPT regulation
provides for zero discharge of process wastewater pollutants  (40  CFR
415.312).  Pretreatment regulations will be developed in Phase II.

Production Processes and Effluents

Calcium  hydroxide  is  produced by adding water to calcium oxide in a
pug mill premixer.  The reacted mixture goes to an  agitated  hydrator
where  more  water  is added, resulting in an exothermic reaction.  No
wastewater is produced and  therefore,  there  is  zero  discharge  to
navigable waters.

Plants

There  are approximately fifteen plants producing calcium hydroxide in
the United States.

Chromic Acid
Summary of Recommendations

It has been determined that no additional  effort  will  be  given  to
developing  revised  BAT,  and  NSPS regulations for this subcategory.
The basis for this determination is that the  existing  interim  final
BPT  regulation  is  zero  discharge.   Pretreatment standards will be
developed in Phase II.

Production Process and Effluents

Sodium dichromate liquor from the dichromate  manufacturing  operation
is  reacted  with sulfuric acid and filtered to recover impure chromic
acid as a solid.  The mother liquor  is  returned  to  the  dichromate
process  for  reuse.  The recovered chromic acid is fed to a melter in
which the sodium bisulfate liquifies and is separated from the chromic
acid.  The bisulfate is returned to  the  dichromate  operation.   The
                                  795

-------
chromic  acid  is  resolidified,  flaked and packaged for sale.  Wastes
are returned to the dichromate process for reuse.

Plants

There are five plants producing chromic acid.

BPT Limitations

Regulations for BPT were promulgated on May 22, 1975 (40 CFR 415.352).
It has been determined that this subcategory  will  be  excluded  from
development  of  BAT  and  NSPS  regulations  under Paragraph 8 of the
Settlement Agreement  because  the  operations  are  subject  to  zero
discharge regulations for BPT.

Cuprous Oxide

Summary of Determinations

It  has  been determined that no further effort be given to developing
BPT,  BAT, NSPS, and Pretreatment regulations  for  the  Cuprous  Oxide
Subcategory.   The  basis for this determination is that there is only
one plant manufacturing cuprous oxide.  The  subcategory  is  excluded
under Paragraph 8 of the Settlement Agreement.

Production Processes and Effluents

Cuprous  oxide  is  manufactured  by  reducing cupric oxide by thermal
decomposition in an oxygen-free environment.  The reaction  occurs  at
high temperature aided by a proprietary catalyst.   There is no process
related wastewater.

Cuprous  oxide  is  used in the manufacture of glass, ceramics, marine
paints, and photoelectric cells.  It  is also used in agriculture as  a
seed fungicide, as an antiseptic and  as a catalyst.

Status of Regulations

Subpart AK has been reserved for this subcategory (Table 26.13-2).

Ferric Chloride

Summary of Determinations

It  has  been  determined  that  no   further  effort  will be given to
developing revised BAT or NSPS regulations  for  this  industry.   The
basis  for  this determination is that the existing regulation for BPT
is zero discharge.  Pretreatment will be developed in Phase II.
Production Processes and Effluents

Ferric chloride is produced from
liquor  is  reacted  with  iron,
waste  pickle  liquor.   The  pickle
chlorine and hydrochloric acid.  The
                                   796

-------
solution is filtered and sold as a solution or evaporated  to  dryness
to  produce a solid product.  Wastewater from filter washes, equipment
washing and leaks and spills is returned to the process.
Plants

There are 21  plants producing ferric chloride.
to discharge to POTW.

Toxic Pollutants
                                                Two plants  are  known
The  source  of  toxic  pollutants  is  the pickle liquor feed.  Toxic
pollutants involved are chromium, copper, lead, nickel and zinc.

BPT Limitations

Regulations for BPT were promulgated on May 22, 1975 (40 CFR 415.382),
which require zero discharge of process  wastewater  pollutants.   The
regulations have not been challenged.

BAT and NSPS Limitations

Zero  discharge  regulations  were  proposed  May 22, 1975 for BAT and
NSPS.  Since BPT already requires zero discharge,  BAT  and  NSPS  are
being excluded under Paragraph 8 of the Settlement Agreement.

Ferrous Sulfate

Summary of Determinations

It  has  been determined that no further effort be given to developing
BAT,  NSPS,  or  Pretreatment  regulations  for  the  Ferrous  Sulfate
Subcategory.  The basis for this determination is that ferrous sulfate
is  recovered  as  a  by-product and in each of the two processes, the
wastes are attributable to the primary process.  Recovery  of  ferrous
sulfate   actually   reduces  the  waste  load  of  both  the  primary
operations.  This subcategory is excluded under  Paragraph  8  of  the
Settlement Agreement.

Production Processes and Effluents

Ferrous sulfate is made using two processes.  In the first case, it is
recovered  from  the  waste  su If uric  acid  pickle  liquor containing
ferrous sulfate, ferric sulfate, and  unreacted  sulfuric  acid.   The
solution  is  reacted with iron to reduce ferric ions to ferrous ions.
The process is a by-product recovery from a waste solution rather than
a direct manufacturing process.  In the second  process,  the  sulfate
process,  ferrous  sulfate  is  obtained  as  a  by-product during the
manufacture of titanium dioxide.  In  the  sulfate  process,  titanium
dioxide-bearing  ores  are  dissolved  in  sulfuric  acid  at  a  high
temperature to produce iron {ferrous sulfate)  and  titanium  sulfate.
Iron  sulfate  is  removed  by crystallization and titanium sulfate is
hydrolyzed and then calcined to produce  the  final  titanium  dioxide
                                  797

-------
product.   All  the wastes from the second process are associated with
the titanium dioxide production.

Process wastewater is derived principally from gas scrubbers.

Plants

There are  13  producers  recovering  ferrous  sulfate  from  titanium
dioxide  manufacture  as a by-product or from the waste pickle liquor.
Four of the 13 producers recover ferrous sulfate as a by-product  from
the  sulfate  process.   The  ferrous  sulfate subcategory is excluded
under Paragraph 8 of the Settlement  Agreement  because  there  is  no
direct method used for its manufacture and it is either recovered from
the  waste  pickle liquor or as a by-product from the titanium dioxide
manufacture and contributes no wastewater discharge of its own.

Fluorine

Summary of Determinations

It  has  been  determined  that  no  additional  effort  be  given  to
developing  revised BAT or NSPS regulations for this subcategory.  The
basis for this recommendation is that the existing interim  final  BPT
regulations   is  zero  discharge.   Pretreatment  standards  will  be
developed in Phase II.

Production Processes and Effluents

Flourine is produced by  electrolysis  of  liquid  hydrogen  fluoride.
Fluorine  is  formed  at one electrode and hydrogen at the other.  The
fluorine is compressed and packaged in cylinders.  There is no process
wastewater from this process.

Plants

There are 3 plants producing fluorine.

BPT Limitations

Regulations for BPT were promulgated on May 22,  1975,  (40 CFR 415.402)
and require zero discharge  of  process  wastewater  pollutants.   The
regulations have not been challenged.

BAT and NSPS Limitations

Zero  discharge  regulations  were proposed for BAT and NSPS but never
promulgated.  Since BPT already requires zero discharge, BAT and  NSPS
are being excluded under Paragraph 8 of the Settlement Agreement.
                                   798

-------
Hydrochloric Acid

Summary of Determinations

It  has  been determined that no further effort be given to developing
regulations for BPT, BAT, NSPS, or Pretreatment for  the  Hydrochloric
Acid  Subcategory.   The  basis for this determinations is:  the small
quantities of toxic pollutants found during screening  are  far  below
levels   treatable   by   demonstrated   treatment  technology.   This
subcategory is excluded under Paragraph 8 of the Settlement Agreement.

Production Processes and Effluents

Most of the hydrochloric acid is  produced  as  a  by-product  in  the
manufacture  of chlorinated organic compounds.  It is used in oil well
activations,  pickling  of  steel,  metal  cleaning,   in   monosodium
glutamate  manufacture  and  starch hydrolysis.  It is also used as an
acid reagent in several chemical manufacturing processes.

The industry profile data for  this  subcategory  is  given  in  Table
26.17-1.  The data given and coverage of this subcategory applies only
to  the manufacture of hydrochloric acid by the thermal combination of
chlorine and hydrogen.  Wastes from this process come from  combustion
chamber condensate and from a fume scrubber.

While most of the hydrochloric acid is produced as a by-product in the
manufacture   of   chlorinated   organic  compounds,  the  wastes  are
attributable to  the  organic  compounds  involved.   This  by-product
production is not covered in this subcategory.

Toxic Pollutants

Data has been received on about 25 percent of the industry as a result
of  Section  308  letters.   In  addition, a sampling survey for toxic
pollutants was made at one plant.  The results of the 308 letters  and
the  sampling  survey  showed  concentrations  close  to the limits of
detectability.

The maximum concentrations of priority pollutants found were:
          Pollutant

          Lead

          Mercury

          Nickel
Maximum Concentration
   Observed Uq/1)

         3.5

         2

         5.5
                                    799

-------
Status of Regulations

BPT, BAT, NSPS and PSNS regulations (40  CFR  415.72)  requiring  zero
discharge  were promulgated on March 12, 1974.  These regulations have
since been remanded by the court and are not in effect.

Hydrogen

Summary of Determinations

It has been determined  that  no  further  effort  will  be  given  to
developing  revised BAT or NSPS regulations for this subcategory.  The
basis for this recommendation is that the existing BPT  regulation  is
zero   discharge   of   process   wastewaters   to  navigable  waters.
Preteatment standards will be developed in Phase II.

Production Processes and Effluents

Hydrogen is made chiefly from two sources:  purification of  petroleum
refinery  by-product  gases  and as a co-product in the manufacture of
carbon monoxide.  In the latter case, the wastes are attributed to the
carbon monoxide subcategory.  Only the  production  of  hydrogen  from
refinery by-product gases will be discussed.

Crude  hydrogen as a refinery by-product is passed through a catalytic
bed to remove oxygen and a drier to remove the  water  formed  by  the
catalytic  reaction.   The  gas  is  then  cooled, purified and passed
through  a  converter  to  change  ortho-hydrogen  to  the  para-form.
Hydrogen  is  usually cooled to a liquid form for storage or shipping.
No contact process water is used during the manufacture.
                                                        None are known
Plants

There are approximately 137 plants producing hydrogen.
to have discharges.

BPT Limitations

Regulations were  promulgated  on  May  22,  1975,  (40  CFR  415.412)
requiring zero discharge of process wastewater pollutants to navigable
waters.    Only  contaminated  non-process  water  is  allowed.  "This
includes rain water, waters which  come  in  contact  with  accidental
spills  and leads, and discharges for personal safety.  All reasonable
measures must have been made to  prevent,  reduce,  and  control  each
contact and to mitigate the effects.

BAT and NSPS Limitations

BAT  and  NSPS were proposed on May 22, 1975, requiring zero discharge
of process wastewater to navigable waters.  Since BPT already requires
zero discharge, BAT and NSPS are being excluded under Paragraph  8  of
the Settlement Agreement.
                                    800

-------
 TABLE 26 ..17-1  -
SUBCAIEGORY PROFILE DATA SUMMARY
 SUBCAOEGORY
HYDROCHLORIC ACID
 Total subcategory capacity rate
 Total subcategory production rate
 Number of plants in this subcategory
 308 Data on file for
     With total capacity of
     With total production of
     Representing capacity
     Representing production
     Plant production range:
             Minimum
             Maximum
     Average production
     Median production
     Average capacity utilization
     Plant age range:
             Minimum
             Maximum
     Vfaste water flow range:
             Minimum
             Maximum
     Volume per unit product:
             Minimum
             Maximum
                               MA
                               NA
                               NA
                                6
                          163,000 kkg/year
                          119,000 kkg/year
                               NA
                               NA

                               NA
                               NA
                               NA
                               NA
                               NA
                                4 years
                               20 years
                               NA

                               NA
                               NA

                               NA
                               NA
 Sources of data are Stanford Research Institute, Directory of Chemical
 Producers,  U.S.A.,  1977,  U.S. Department of Commerce, Current Industrial
 Reports, December 1977; Energy and Environmental Analysis, Inc.;  Draft
 Report,  "Preliminary Economic Assessment of Effluent Limitations  in the
 Inorganic Chemical  Industry, " June, 1978 and "Economic Analysis of
 Proposed Revised Effluent Guidelines and Standards for the Inorganic
 Chemicals Industry,"  March, 1980
NA = Not Available
                                   801

-------
Iodine

Summary of Determinations

It  has  been  determined  that  no additional effort will be given to
developing revised BAT  or  NSPS  regulations.   The  basis  for  this
determination  is  that  the  existing  regulation  for  BPT  is  zero
discharge.  Pretreatment standards will be developed in Phase II,

Production Process and Effluents

Iodine is produced from brine solutions containing iodine.  The  brine
is  acidified and chlorinated liberating free iodine.  The free iodine
is stripped from the brine and treated again  with  chlorine  yielding
solid  iodine.  The slurry is filtered, treated with sulfuric acid and
refiltered.  The product is then crushed and packaged for  sale.   The
wastes  from this process are spent brine solutions which are returned
to the well from which the brine was initially obtained.

Plants

There are 4 plants producing iodine.

BPT Limitations

Regulations for BPT were promulgated on May 22, 1975, (40 CFR 415.432)
and require zero discharge  of  process  wastewater  pollutants.   The
regulations have not been challenged.

BAT and NSPS Limitations

Zero  discharge  regulations were proposed for BAT and NSPS on May 22,
1975.  Since BPT already requires zero discharge,  BAT  and  NSPS  are
being excluded under Paragraph 8 of the Settlement Agreement.

Lead Monoxide

Summary of Determinations

It  has  been  determined  that  no  further  effort  will be given to
developing BAT or NSPS regulations for this   subcategory.   The  basis
for  this  recommendation is that the existing BPT regulation requires
zero discharge.  Pretreatment standards will  be developed in Phase II.

Production Processes and Effluents

Lead monoxide is produced by the thermal oxidation of lead.  There are
no process wastewater streams generated by lead  monoxide  production.
Dust  control  is  the  problem  in  this  subcategory.   Use  of  dry
collection systems rather  than  a  water  collection  system   is  the
control technology for meeting the regulation.
                                    802

-------
 Plants

 There are  15 plants producing  lead monoxide  in  the  United  States.   Ten
 plants  are  known  to  use  dry  bag   collection   systems and  have no
 discharge  of  wastewater.   Others  are   subject   to   existing   zero
 discharge  regulations.

 BPT Limitations

 BPT  limitations  were  promulgated  on May  22,  1975  (40 CFR  415.442).
 The  limitations  require  zero  discharge   of   process   wastewater
 pollutants into navigable waters.

 BAT and NSPS Limitations

 On  May  22,   1975, zero discharge regulations  were proposed  but  never
 promulgated for  BAT  and  NSPS.   Since   BPT  already  requires   zero
 discharge,  BAT  and  NSPS are being excluded under Paragraph 8 of the
 Settlement Agreement.

 Lithium Carbonate

 Summary of Determinations

 It has been determined that no further  efforts  be given to developing
 or  revising   regulations  for BPT, BAT, NSPS,  or Pretreatment  for the
 Lithium Carbonate Subcategory.  The bases  for this  determination   are:
. 1) there is only one plant in this subcategory  using the spodumene ore
 process  and   discharging  process  wastewater,  and  2)   there  is an
 existing zero  discharge  regulation  for  the   brine   process.    This
 subcategory is excluded under Paragraph 8  of  the Settlement Agreement.

 Production Processes and Effluents

 Lithium  carbonate  is  produced  by  two  processes.   In  one process,
 spodumene  ore  is heated at a high  temperature   to  render it  highly
 reactive.  It  is then cooled, ball-milled, and mixed with  concentrated
 sulfuric   acid.   The  acid-roasted ore is leached  with water,  and the
 excess acid is neutralized with ground  limestone.   This  mixture  is
 filtered   and  further  treated  with   lime   and soda  ash.    Further
 processing precipitates 1ithium  carbonate.   Wet   scrubbers  are   the
 sources of wastewater.  Significant quantities of toxic pollutants are
 not found  in the wastewater.

 In the other process, lithium carbonate is produced by  the reaction of
 lime with  concentrated brine, and lithium  carbonate is  precipitated by
 filtration.    Process  wastewater  consists  of  spent brines,  which are
 sent to on-site evaporation ponds.  There  is  no   process wastewater
 discharge  from this process.
                                   803

-------
Status of Regulations

There  is  an  existing  BPT  regulation  for this subcategory (40 CFR
415.452).

Manganese Sulfate

Summary of Determinations

It has been determined that no further effort be given  to  developing
or  revising  BPT,  BAT,   NSPS,  or  Pretreatment  regulations for the
Manganese Sulfate Subcategory.  The bases for this determination  are:
1)  there  is only one plant making commercial grade manganese sulfate
that has a wastewater discharge,  and  2)  the  amount  of  wastewater
produced  by  that  plant  is  low.  The subcategory is excluded under
Paragraph 8 of the Settlement Agreement.

Production Processes and Effluents

There are two processes for the manufacture of manganese sulfate;  the
hydroquinone   process   and   the  coke  and  ore  process.   In  the
hydroquinone process, manganese ore, aniline  and  sulfuric  acid  are
reacted  to  produce  manganese sulfate, quinone and ammonium sulfate.
The reacted mixture is steam distilled  to  remove  quinone  which  is
further  processed  to  hydroquinone.   The  mixture  of manganese and
ammonium sulfate is filtered, evaporated, and crystallized.  Manganese
sulfate is recovered  as  crystals,  and  the  spent  1iquor  contains
ammonium  sulfate.   In the second process, manganese ore and coke are
reacted in a kiln and the product is leached with sulfuric acid.   The
resulting  slurry  is  evaporated  to  dryness to recover a 30 percent
product for agricultural purposes.  The amount of wastewater  produced
from  the hydroquinone process is small and the other process produces
no waterborne waste.

Plants

Four plants are manufacturing manganese sulfate.  Two of the producers
use it for fertilizer  production  and  they  generate  no  waterborne
wastes.   One  plant  produces  reagent  grade  product  and the total
production is very low.  Only one other plants manufactures  manganese
sulfate  (commercial grade) and has a significant wastewater flow.

Status of Regulations

Since  only  one manganese sulfate plant discharges waste to navigable
waters, the subcategory is excluded from federal discharge  regulation
for  BPT,  BAT,  NSPS, and Pretreatment standards under Paragraph 8 of
the Settlement Agreement.
                                   804

-------
Nitric Acid

Summary of Determinations

The existing nitric acid regulation in the fertilizer category (40 CFR
418.5) is applicable to all nitric acid plants captive to a fertilizer
production facility.  In addition, sampling has shown that  there  are
no   significant   quantities  of  toxic  pollutants  in  the  process
wastewaters from stand alone nitric acid plants.    Further  BPT,   BAT,
NSPS  or  Pretreatment  regulations  will  not  be  developed for this
subcategory.

Production Processes and Effluents
Most of the nitric  acid  produced  is  used  in  the  manufacture  of
ammonium  nitrate and other nitrogen fertilizers.  On site captive use
is extensively practiced.  It is  also  used  in  the  manufacture  of
explosives, plastics and other organic products.
acidic  and  pickling  agent.   The  source  of
equipment washing operations.
                     the
                     Other uses are as an
                    process wastewater is
The industry profile data for this  subcategory  are  given  in  Table
26.23-1.

Toxic pollutants found in raw wastes during sampling were as follows:

                                Maximum Concentration Observed
                                           Ug/1)
Pollutant

Chromium
Zinc
Lead
Mercury
Silver
2,4-Dinitrophenol
Nickel
Cyanide
Screening
(2 Plants)
no
120
29
0.
0.
215
170
< 0.



47
5


04
Verification
 (1 Plant)

   100
   791
 <  10
     4.5
 <  15
Not Analyzed
    85
 <   0.02
The 2,4-Dinitrophenol is caused by contamination from organic products
manufactured  at  the  plant  and will be addressed in that guideline.
The chromium and zinc are ingredients of  cooling  water  conditioners
present   in  the  blowdown  which  is  mixed  with  process  streams.
Appropriate control is by best  management  practice  'not  end-of-pipe
treatment  via  national regulation.   Other metals are below the limit
of treatability.
                                  805

-------
Status of Regulations

BPT, BAT, NSPS, and PSNS regulations (40 CFR 415.100)  requiring  zero
discharge  were promulgated on March 12, 1974.  These regulations have
since been remanded by the court and are not in effect.

Oxygen and Nitrogen

Summary of Determinations

It has been determined that no further effort be given  to  developing
or  revising regulations for BAT, NSPS, or Pretreatment for the Oxygen
and Nitrogen Subcategory.  The bases for this determination  are:   1)
the  wastewater  discharge mainly consists of compressor water, and 2)
the only toxic pollutant detected at or above treatability  level  was
copper  which  is  at  the level of treatability.  This subcategory is
excluded under Paragraph 8 of the Settlement Agreement.

Production Processes and Effluents

Oxygen and nitrogen are produced from air by distillation of liquified
air.  Oxygen  is  used  in  the  production  of  steel,  gas  welding,
medicine,  jet fuel, in sewage treatment plants and in the manufacture
of ethylene and acetylene.  In rocket  propulsion,  liquid  oxygen  is
often used as a cryogenic liquid oxidizer.

The  greatest  use of nitrogen is in the manufacture of ammonia by the
Haber process.  It is also used  in cryosurgery.  As an inert  gas,  it
is  used  to prevent oxidation by air.  In the liquid form, it is used
for low temperature refrigeration.

The wastewater discharge mainly  consists of compressor cooling  water.
Other  wastewaters are small quantities of boiler blowdown, intake air
scrubber waters, and compressor  condensate.

The industry profile for this subcategory is given in Table 26.24-1.

Toxic Pollutants

Data has been received on  10  plants  as  a  result  of  Section  308
letters.   There  are  at  least  230  plants  in  the  United States.
However, all operate using the same  basic  process.   One  plant  was
sampled  during  the screening program.  Toxic pollutants found in the
raw waste loading during sampling were:
                                  806

-------
 TABLE 26.23-1  -
SUBCAIEGOHf PRCFILE DATA SUMMARY
 SUBCAIBGOKy
NITRIC ACID
 Ttotal subcategory capacity rate
 Total subcategory production rate
 Number of plants  in this subcategory
 308 Data on file  for
    With total "capacity of
    With total production of
    Representing  capacity
    Representing  production
    Plant production ranges
            Minimum
            Maximum
    Average production
    Median production
    Average capacity utilization
    Plant age range:
            Minimum
            Maximum
    Waste water flow range:
            Minimum
            Maximum
    Volume per unit product:
            Minimum
            Maximum
                         9,177,000 kkg/year
                         7,171,000 kkg/year
                               87
                               11
                         1,106,000 kkg/year
                          774,400 kkg/year
                               12 percent
                               11 percent

                               NA
                               NA
                               NA
                               NA
                               NA

                                 4 years
                                83 years

                               NA
                               NA

                               NA
                               NA
Sources of data are Stanford Research Institute, Directory of Chemical
Producers, U.S.A., 1977, U.S. Department of Corrmerce, Current,Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry,1' June,1978  and "Economic Analysis of Proposed
Revised Effluent Guidelines and Standards for the Inorganic Chemicals
Industry," March,  1980
NA = Not Available
                                  807

-------
          Pollutant

          Chromium

          Copper

          Lead

          Nickel

          Zinc
Concentration Ug/1)

        26

       590

        51

        79

       170
The likely sources of copper  are  corrosion  and  bearing  wear,  and
concentration  in  boiler  and  cooling  tower  blowdowns.  The copper
levels are at the accepted  levels  of  treatment,  therefore  further
reduction is not practical.

Status of Regulations

Interim Final BPT regulations (40 CFR 415.492) were promulgated on May
22,  1975.   These  regulations  require limitations on pH and oil and
grease.  These regulations remain in effect and no change is needed.

BAT and NSPS regulations (40 CFR 415.494) were  proposed  on  May  22,
1975.  These regulations were never finalized.  It has been determined
that  the  Oxygen  and  Nitrogen  Subcategory  be  excluded  from  the
development of. BAT and NSPS  limitations  under  Paragraph  8  of  the
Settlement   Agreement  for  the  following  reasons:   The  discharge
consists of compressor water wherein the only toxic pollutant found is
copper which is at the level of treatability.

Potassium Chloride

Summary of Determinations

It has been determined  that  no  further  effort  will  be  given  to
developing  revised BAT or NSPS regulations for the potassium chloride
subcategory.  The basis for this determination is  that  existing  BPT
regulations  specify  zero  discharge of process waters.  Pretreatment
standards will be developed in Phase II.

Production Processes and Effluents

Potassium chloride is produced in the United States by  two  principal
processes:  extraction from sylvite ore and extraction from lake brine
(Trona  Process).   Sylvite ore is a combination of potassium chloride
and sodium- chloride.  The  ore is crushed, screened, and wet-ground  in
brine.   The ore is then separated from clay impurities in a desliming
process.  The clay impurities are fed to  a  gravity  separator  which
removes  some of the sodium chloride precipitated from the leach brin^
and insolubles for disposal as waste.  After  desliming,  the  ore  is
chemically  treated  and   the potassium chloride  is separated from the
sodium chloride in a flotation process.  The tailings  from  flotation
                                    808

-------
TftBLE  26.24-1  -
SUBCRTEGORy PKCFZLE DKEA .SUMMARY
SUBCAIEGORY
OX3OTN AND NITROGEN
•total subcategory capacity rate
Total subcategory production rate
Number of plants in this subcategory
308 Data on file for
    With total capacity of
    With total production of
    Representing capacity
    Representing production
    Plant production range:
            Minimum
            Maximum
    Average production
    Median production
    Average capacity utilization
    Plant age range:
            Minimum
            Maximum
    Waste water flow range:
            Minimum
            Maximum
    Volume per unit product;
            Minimum
            Maximum
                        35,526,000 kkg/year
                            NA
                           113
                             9
                         1,588,000 kkg/year
                         1,473,000 kkg/year
                              4.5 percent
                            NA

                             2,400 kkg/year
                           378,000 kkg/year
                            NA
                            NA
                            NA

                                4 years
                               36 years

                            NA

                            NA

                            NA
                            NA
Sources of data are Stanford Research Institute, Directory of Chemical
Producers, U.S.A., 1979, U.S. Department of Commerce, Current Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry,11 June,  1978  and  "Economic Analysis of
Proposed Revised Effluent Guidelines and  Standards for the Inorganic
Chemicals Industry,"  March,  1980
NA = Not Available
                                   809

-------
are   wasted   and  the  resulting  potassium  chloride  slurries  are
centrifuged to recover the potassium chloride.  The  product  is  then
dried, screened, and packaged.  The centrifuge liquors are recycled to
the flotation circuit.

The  Trona Process uses a cyclic evaporation-crystallization system in
which saline brine is evaporated to nominal dryness.   The  brine  and
recycle liquor is concentrated in triple effect evaporators to produce
a  hot  liquor high in potassium chloride and borax.  Large quantities
of sodium chloride and burkeite (Na2C03.Na2S04) are  crystallized  and
separated  during evaporation.  The sodium chloride is returned to the
brine source, and the burkeite is transported to other  processes  for
separation  and  refining.   The  hot liquor is then cooled rapidly in
vacuum crystallizers and the potassium chloride is filtered  from  the
slurry.   The  potassium chloride in then dried and packaged.  A small
portion may be refined further and/or converted to potassium  sulfate.
The cool liquor remaining is then allowed to crystallize the remaining
borax  which is then refined further using recrystallization and other
processes.   The  remaining   liquor   is   recycled   back   to   the
evaporation-crystallization step.

Plant

There  are  thirteen plants producing potassium chloride, two of which
use the Trona Process.

BPT Regulations

BPT regulations were promulgated on May  22,  1975  (40  CFR  415.502)
requiring zero discharge of process wastewater pollutants to navigable
waters, except that residual brine and depleted liquor may be returned
to  the  body  of  water  from which the brine solution was withdrawn.
There are no instances where the brine soure is a navigable water.

BAT and NSPS Limitations

BAT and NSPS were proposed on May 22, 1975, requiring  zero  discharge
of  process  wastewater  to  navigable waters.  It has been determined
that the potassium chloride subcategory  will  be  excluded  from  the
development  of  revised BAT and NSPS limitations under Paragraph 8 of
the Settlement Agreement since  a  zero  discharge  regulation  is  in
effect.   In  the absence of BAT and NSPS regulations, permits will be
based on BPT.

Potassium Bichromate

Summary of Determinations

It has been determined  that  no  further  effort  will  be  given  to
developing   revised   BAT  and  NSPS  regulation  for  the  Potassium
Dichromate Subcategory.  The basis  for  this  determination  is  that
existing  BPT,  BAT  and  NSPS  regulations  specify zero discharge of
                                   810

-------
process  wastewater  pollutants  to  navigable  waters.   Pretreatment
standards will be developed in Phase II.

Production Processes and Effluents

Potassium dichromate is made by reacting a sodium dichromate dihydrate
solution   with  potassium  chloride.   The  potassium  dichromate  is
crystallized from solution requiring only removal of  water  prior  to
sizing  and  packaging.   The  process  water  is recycled back to the
initial reaction tank.

BPT, BAT and NSPS Limitations

BPT, BAT and NSPS limitations were promulgated March 12, 1974 (40  CFR
415.122, 415.123 and 415.T25).  All subparts require zero discharge of
process wastewater pollutants to navigable waters.

It  has been determined that the potassium dichromate subcategory will
be excluded from the development of revised BAT and  NSPS  limitations
under  Paragraph  8  of  the Settlement Agreement.  The basis for this
determination is that  by  maintaining  existing  BPT,   BAT  and  NSPS
limitations, no discharge of wastewater pollutants to navigable waters
will occur.

Potassium Iodide

Summary of Determinations

It  has  been determined that no further effort be given to developing
or revising BAT, NSPS,  and Pretreatment regulations for the  Potassium
Iodide Subcategory.  The bases for this determination are:  1) because
the  wastewater  discharge  is  less  than  100  gallons  per day, the
quantity  of  pollutants  discharged  is  very   low/   and   2)   the
concentrations  of  the  toxic  pollutants  are  at  or below accepted
treatment levels.  This subcategory is excluded under Paragraph  8  of
the Settlement Agreement,

Production Processes and Effluents

Potassium  iodide  is  used  in  photographic emulsions, in animal and
poultry feeds, table salts and analytical chemistry.  It  also  has  a
number of medical uses.

One  manufacturing  process  is  known  as the iron carbonate process.
This involves the reaction of  iron  powder  with  iodine  in  aqueous
solution.   An  intermediate  compound, ferrosoferricyanide, is formed
which is  subsequently  reacted  with  potassium  carbonate  to  yield
potassium  iodide.   The  raw  product  is  purified,  concentrated by
evaporation  and  cooled  to  promote  crystallization.   Water   used
directly  in  the  process is lost by evaporation.  The only source of
process wastewater is from equipment wash down operations.

The industry profile for this subcategory is given in Table 26.27-1.
                                   811

-------
Toxic Pollutants

Data has been received for approximately 50 percent of the industry as
a result of Section 308 letters.  In addition, a sampling  survey  was
made  at one plant.  The following toxic pollutants were identified in
the plant wastes:
          Pollutant

          Antimony
          Chromium
          Copper
          Lead
          Silver
          Zinc
Concentration
        48
        22
      1040
        26
        34
        30
However, the levels of these pollutants are at or below accepted level
of treatability.  In addition, the flows are less than 100 gallons per
day.  At the one  plant  sampled,  there  was  no  process  wastewater
discharged since the wash water was sent to an evaporation pond.

Status of Regulations

A.   BPT Limitations

     BPT regulations (40 CFR 415.511)  were  promulgated  on  May  22,
     1975.  These regulations require limitations on pH, TSS, sulfide,
     iron  and barium.  These regulations are adequate for the control
     of.conventional and nonconventional pollutants.

B.   BAT and NSPS Regulations

     NSPS and BAT limitations were proposed in May 22, 1975, but never
     finalized.  It has now been determined that the Potassium  Iodide
     Subcategory  be  excluded  from  the  development of BAT and NSPS
     limitations under Paragraph 8 of the Settlement Agreement for the
     following reasons: 1) very small quantities of  toxic  pollutants
     are  discharged  from  this  industry,  and  2)  those pollutants
     discharged are at or below accepted treatability levels.

C.   Pretreatment Limitations

     Because  no  significant  quantities  of  toxic  pollutants   are
     present,  no  further  effort  will  be  given  to development of
     pretreatment regulations for this subcategory.

Potassium Metal

Summary of Determinations

It has been determined  that  no  further  effort  will  be  given  to
developing  revised  BAT  or  NSPS regulations for the Potassium Metal
Subcategory.  These bases for this recommendation  are':    1)  existing
                                    812

-------
BPT,  BAT  and  NSPS  regulations  specify  zero  discharge of process
wastewaters; and 2) there is only one plant producing potassium in the
U.S. and that plant uses a dry process.  Pretreatment  standards  will
be developed in Phase II.
Production Processes and Effluents

Potassium  metal  is  prepared  by  melting
gas-fired melt pot prior to being fed  to  an
molten potassium chloride flows down through
is  contacted  by  ascending  sodium  vapors
reboiler.
chloride,
apparatus.
product.
effluents.

Plant
            The  reaction  yields  elemental
           which  is  withdrawn  continuously
            The  elemental  potassium  is  wi
           No  process  water  is  used  so
                                             potassium  chloride  in a
                                               exchange  column.   The
                                             a packed column, where it
                                              coming  from a gas-fired
                                              potassium   and   sodium
                                               from  the  base  of the
                                             thdrawn  as  an  overhead
                                             there  are  no waterborne
Only one plant produces potassium  metal  in  the  U.S.
process water and there are no waterborne effluents.
                                                          It  uses
                                                                    no
BPT, BAT and NSPS Limitations

BPT,  BAT and NSPS limitations were promulgated March
415.112), 415.113 and 415.115).  All subparts require
of process wastewater pollutants to navigable waters.
                                                      12, 1974 (40 CFR
                                                       zero  discharge
It  has  been  determined that the potassium metal subcategory will be
excluded from the development of  revised  BAT  and  NSPS  limitations
under  Paragraph  8  of  the  Settlement  Agreement.   Maintaining the
existing regulations will eliminate the discharge of toxic pollutants.

Potassium Permanganate

Summary of Determinations

It has been determined that no further effort be given  to  developing
BPT,  BAT,  NSPS,  and  Pretreatment  regulations  for  the  Potassium
Permanganate Subcategory.  The basis for this determinations  is  that
there  is  only  one  plant manufacturing Potassium Permanganate.  The
subcategory is excluded under Paragraph 8 of the Settlement Agreement.

Production Processes and Effluents

Manganese ore  is  s lurried  with  potassium  hydroxide  solution  and
treated with oxygen to produce potassium manganate.  This intermediate
product  and  the  ore  wastes are recovered by centrif ugation and the
solids are then leached to  dissolve  the  manganate.   The  resulting
slurry  is  filtered  to  remove  the  ore  wastes  and  the manganate
converted in electrolytic cells.   The  permanganate  is  crystallized
from the solution to form the product.
                                   813

-------
TABLE  26.27-1  -
SUBCATEQOKSf PRCFUE DATA SUMMARY
SUBCATEGOR*
POTASSIUM IODIDE
Total subcatecpry capacity rate
Total subcatecpry production rate
Number of plants in this subcategory
308 Data on file for
    With total capacity of
    With total production of
    Representing capacity
    Representing production
    Plant production range:
            Minimum
            Maxinun
    Average production
    Median production
    Average capacity utilization
    Plant age range:
            Minimum
            Maximum
    Waste water flow range:
            Minimum
            Maxiitun
    Volume per unit product:
            Minimum
            Maximum
                               NA
                               NA
                                9
                                4
                            1,985 kkg/year
                            1,300 kkg/year
                               NA
                               50 percent

                               79 kkg/year
                              634 kkg/year
                               NA
                               NA
                               NA

                               27 years
                               42 years

                              NA
                              NA

                              NA
                              NA
Sources of data are Stanford Research Institute, Directory of chemical
Producers, U.S.A., 1977, U.S. Department of Commerce, Current Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry, "  June,  1978 and  "Economic Analysis of
Proposed Revised Effluent Guidelines  and Standards for the Inorganic
Chemicals Industry,"  March,  1980
NA = Not Available
                                   814

-------
Potassium Sulfate

Summary of Determinations

It  has  been determined that no further effort
or revising BPT, BAT and NSPS for the Potassium
The  bases  for .this determination are there is
for BAT and NSPS that requires zero discharge
pollutants  (40 CFR 415.133 and 415.135).  The
under Paragraph 8 of the Settlement Agreement.
will be developed in Phase II.

Production Processes and Effluents
 be given to developing
  Sulfate  Subcategory.
 an existing regulation
of  process  wastewater
subcategory is excluded
 Pretreatment standards
Potassium sulfate is produced by the reaction in solution of potassium
chloride with langbeinite ore.  Langbeinite ore is a  natural  sulfate
of  potassium  and  magnesium  (K2S04.MgS04),  usually intermixed with
sodium chloride.  When the reacted solution is  partially  evaporated,
potassium sulfate precipitates out, and is recovered by centrifugation
or  filtration  from  the brine liquor, dried, and sold. The remaining
brine liquor containing magnesium chloride is the source of raw waste.
Depending on the sodium content of the ore used, the brine  is  either
sold (low sodium content) or is ponded.  In the latter case, the brine
liquor   is   recycled  or  evaporated  and  the  mud  is  landfilled.
Therefore, no discharge  results  from  the  production  of  potassium
sulfate.

Plants

There  are approximately eight producers of commercial grade potassium
sulfate in the United States.

Sodium Bicarbonate

Summary of Determinations

It has been determined  that  no  further  effort  will  be  given  to
developing revised BAT and NSPS regulations for the Sodium Bicarbonate
Subcategory.   The  basis  for this determination is the existing BPT,
BAT and NSPS regulations specify zero discharge of process  wastewater
pollutants  to  navigable  waters.   Pretreatment  standards  will  be
developed in Phase II.

Production Processes and Effluents

Sodium bicarbonate is produced by reaction of  sodium  carbonate  with
water and carbon dioxide under pressure.  The bicarbonate precipitates
from  the  solution  and  is  filtered,  washed,  dried, and packaged.
Wastewater from the filtration is used in product scrubbers  and  then
returned to the process.
                                  815

-------
Plants

Four plants produce sodium bicarbonate in the United States.

BPT, BAT and NSPS Limitations

BPT,  BAT and NSPS limitations were promulgated March 12, 1974 (40 CFR
415.142, 415.143 and 415.145).  All subparts require zero discharge of
process wastewater pollutants to navigable waters.

It has been determined that the sodium bicarbonate subcategory will be
excluded under Paragraph 8 of the Settlement Agreement.  The basis for
the determination is that maintaining the  existing  regulations  will
eliminate the discharge of toxic pollutants.

Sodium Carbonate

Summary of Determinations

It  has  been determined that no further effort be given to developing
or revising BAT, NSPS, or  Pretreatment  regulations  for  the  Sodium
Carbonate  Subcategory.   The bases for this determination are:  1) no
wastewater is discharged to navigable waters  from  the  plants  using
natural  deposits  to  produce sodium carbonate, and 2) only one plant
exists that uses the Solvay Process to produce sodium carbonate.  This
subcategory is excluded under Paragraph 8 of the Settlement Agreement.
The Solvay Process does have a discharge but because there is only one
plant, it is inappropriate to write a regulation for the subcategory.

Production Processes and Effluents

Two methods are used for the  production  of  sodium  carbonate.   One
method  is the recovery from natural sodium carbonate deposits and the
other method is the Solvay process.  The wastewater resulting from the
use of natural deposits is sent to evaporation ponds and no  water  is
discharged  to  navigable  streams.   In  the  Solvay  process, sodium
chloride (brine) is purified  and  saturated  with  ammonia  and  then
chlorinated.   The reacted solution is filtered and sodium bicarbonate
is removed as a filter cake.  The filter cake is calcined  to  produce
sodium  carbonate,  driving  off  moisture  and  carbon  dioxide.  The
production of sodium carbonate by the Solvay process requires the  use
of  large volumes of water for non-contact cooling and process contact
purposes and generates large  loads of  suspended  solids,  alkalinity,
and ammonia.

Plants

Only  one  plant  uses the Solvay process to produce sodium carbonate.
The Solvay process is energy  intensive and generates  large  pollution
loads.   The  process  is  being  replaced  by production from natural
deposits.  It is unlikely that new Solvay process plants will be built
in the future.  The industry profile is presented in Table 26.32-1.
                                    816

-------
The other plants using natural deposits have zero discharge.

Status of Regulations

The regulation originally established has been remanded by the  court.
The Solvay* process does have a discharge but because there is only one
plant it is inappropriate to write regulations for this subcategory.

Sodium Chloride

Summary of Determinations

It  has  been determined that no further effort be given to developing
or  revising  BAT  and  NSPS  regulations  for  the  Sodium   Chloride
Subcategory.   The  basis  for  this  determination  is that there are
existing BAT and NSPS regulations that prohibit discharge  of  process
wastewater  (40 CFR 415.163 and 415.165).  The subcategory is excluded
under Paragraph 8 of the Settlement Agreement.  Pretreatment standards
will be developed in Phase II.

Production Processes and Effluents

Sodium chloride is produced by three methods:  1) solar evaporation of
sea water, 2)  solution mining of natural brines,  and 3) mining of rock
salt.

In the solar  evaporation  process,  salt  water  is  concentrated  by
evaporation  in open ponds to yield a saturated brine solution.  After
saturation is reached, the brine is fed  to  a  crystallizer,  wherein
sodium  chloride  precipitates,  leaving  behind  a concentrated brine
solution (bittern) consisting  of  sodium,   potassium,  and  magnesium
salts.  The precipitated sodium chloride is recovered for sale and the
brine is recycled from the operation.

The  brine  in  the second process is first aerated to remove hydrogen
sulfide.  The brine is then pumped  to  settling  tanks  where  it  is
treated  with caustic soda and soda ash to remove most of the calcium,
magnesium, and iron present as insoluble salts.   After  clarification
to  remove these insolubles,  the brine is then sent to multiple effect
evaporators.  As water is removed, salt crystals form and are  removed
as  a slurry.   The slurry is washed with fresh brine to remove calcium
sulfate.  The washed slurry is filtered, the mother liquor is returned
to evaporators, and  the  crystals  from  the  filter  are  dried  and
screened.   Wastes  are generated from the multiple effect evaporators
and driers,  basic brine purification, and from water treatment.   Zero
aqueous   discharge   can  be  accomplished  by  replacing  barometric
condensers with non-contact exchangers,  eliminating  packing  station
wastes by more efficient design, and recycling all process water.

Mining of rock salt produces no process wastewater.
                                   817

-------
related wastewater during the screening of one plant.  The subcategory
is excluded under Paragraph 8 of the Settlement Agreement.

Production Process and Effluents

Sodium  hydrosulfide  is produced by reaction of hydrogen sulfide with
sodium hydroxide.  Sodium hydrosulfide is used in the  manufacture  of
sodium  sulfide,  other chemicals, and paper (Kraft).  It is also used
in dehairing of hides and industrial  wastewater  treatment.   Process
wastewater may be derived from filter backwash water.

The subcategory profile data are given in Table 26.35-1.

Toxic Pollutants

Toxic pollutants found in the waste during screening of one plant were
phenol   (76   pg/1)   and  naphthalene  (90  ^g/1)   which  are  below
treatability levels.  Due to the very  small  flows  and  waste  loads
generated  by  this  industry,  this  subcategory  is  excluded  under
Paragraph 8 of the Settlement Agreement

Status of Regulations

Subpart BD has been reserved for this subcategory.

Sodium Metal

Summary of Determinations

It has been determined that no further effort be given  to  developing
or revising BPT, BAT, NSPS, or Pretreatment regulations for the Sodium
Metal Subcategory.  The basis for this determination is that the small
quantities  of  toxic  pollutants found during screening are far below
accepted treatability levels.   This  subcategory  is  excluded  under
Paragraph 8 of the Settlement Agreement.

Production Processes and Effluents

Sodium  metal  is  manufactured with chlorine by electrolysis of fused
sodium chloride.  It is used in the  production  of  tetraethyl  lead,
sodium  cyanide,  sodium  peroxide, and titanium and zirconium metals.
In liquid form, it is used as a nuclear reactor coolant;  it  is  also
used as a light, thermally conductive solid in various applications.

The industry profile for this subcategory is given in Table 26.36-1.

Toxic Metals

Data has been received on about 60 percent of the industry as a result
of  Section  308  letters.  In addition, sampling surveys were made at
two plants representing 38 percent of the industry.   Toxic  pollutants
found during sampling were as follows:
                                  820

-------
          Pollutant

          Copper

          Zinc

          Dichlorobromomethane

          Chloroform
Maximum Concentration
    Observed

        31

        13

        33

        10
These  pollutants  are  at very low concentrations which are far below
accepted treatability levels.
Status of Regulations

BPT regulations (40 CFR 415.182) were promulgated on March
These regulations have since been remanded by the court.
                         12,  1974
BAT  and  NSPS  regulations  requiring zero discharge (40 CFR 415.183)
were promulgated on March 12, 1974.  These regulations have been since
remanded by the court.  However,  it  has  been  determined  that  the
sodium  metal  subcategory  be  excluded from BAT and NSPS regulations
because data from Section 308 letters and  sampling  surveys  indicate
that  toxic  pollutant concentrations are far below accepted treatable
levels.

Because no significant quantities of toxic pollutants are present,  no
further   effort   will   be  given  to  development  of  pretreatment
regulations for this subcategory.

Sodium Silicate

Summary of Determinations

It has been determined that no further effort be given  to  developing
BPT,  BAT,  NSPS, and Pretreatment regulations for the Sodium Silicate
Subcategory.  The basis for  this  determination  is  that  the  small
quantities  of  toxic  pollutants  found  during  screening  are below
accepted levels of treatability.  This subcategory is  excluded  under
Paragraph 8 of the Settlement Agreement.

Production Processes and Effluents

Sodium  silicate is manufactured both in liquid and anhydrous powdered
iorm.  It has many industrial uses, such as  additives  in  adhesives,
flocculants,  and  cleaning agents.  It is also used in the production
of soap and  household  detergents.   Sources  of  process  wastewater
include contact cooling water, filter backwash, gas scrubbers and tank
cleaning.

The industry profile for this subcategory is given in Table 26.37-1,
                                 821

-------
Toxic Pollutants

Data has been received on about 63 percent of the industry as a result
of  Section  308  letters.  In addition, a sampling survey was made at
one plant which represents about  6  percent  of  the  industry.   The
following  pollutants were detected:  nickel, copper, and zinc.  These
levels are below  accepted  treatability  levels.   In  addition,  the
sampling  data  was  taken  from  wastewaters  receiving  insufficient
treatment.   The  wastes  were  ponded  to  remove  suspended   sol ids
consisting  essentially  of sand and other silicates.  Normally the pH
of the wastes would be lowered to 9 and receive  additional  settling.
However  the  dissolved silicate and high pH are considered beneficial
by sewerage authorities in  the  removal  of  solids  in  primary  and
secondary settling systems.

Maximum concentrations of toxic pollutants found during sampling are:

          Pollutant

          Copper

          Nickel

          Zinc

Status of Regulations
                              347

                              121

                              181
BPT,  BAT,  and  NSPS  regulations  (40  CFR  415.192)  requiring zero
discharge of pollutants were promulgated on  March  12,  1974.   These
regulations  have  since  been  remanded  by  the court and are not in
effect.

Because no significant quantities of toxic pollutants are present,  no
further   effort   will   be  given  to  development  of  pretreatment
regulations for this subcategory.

Sodium Silicofluoride

Summary of Determinations

This subcategory has been excluded from the present study but will  be
included in the Phase II, Inorganic Chemicals, review.

Production Processes and Effluents

Sodium  silicofluoride  is  used in the manufacture of sodium fluoride
and in the light metal industry as a protective  agent.   It  is  also
used  as  an  insecticide, as a fluxing and opacity agent for ceramics
and in detergent products.

The industry profile for this subcategory is given in Table 26.38-1.
                                   822

-------
 TABLE  26.35-1  -
SUBCKTECOKT PROFILE .DATA SUMMARY
 SUBCftTEGOEY
SODIUM HYDRCSULETDE
 Total subcate^ry capacity rate
 Total subcategory production rate
 Number of plants in this subcategory
 308 Data on file for
     With total capacity of
     With total production of
     Representing capacity
     Representing production
     Plant production range:
             Minimum
             Maximum
     Average production
     Median production
     Average capacity utilization
     Plant age range:
             Minimum
             Maximum
     Waste voter flow range:
             Minimum
             Maximum
     Volume per unit product:
             Minimum
             Maximum
                               NA
                               NA
                               12
                                3
                           56,900 kkg/year
                           44,700 kkg/year
                               NA
                               NA


                            3,800 kkg/year
                           36,500 kkg/year
                               NA
                               NA
                               NA

                                5 years
                               14 years

                              NA
                              NA

                              NA
                              NA
 Sources of data are Stanford Research Institute, Directory of Chemical
 Producers, U.S.A., 1977, U.S. Department of Commerce,  Current Industrial
 Reports, December 1977; Energy and Environmental Analysis, Inc.;  Draft
 Report, "Preliminary Economic Assessment of Effluent Limitations  in the
 Inorganic Chemical Industry," June, 1978 and "Economic  Analysis of
Proposed Revised Effluent Guidelines and Standards for the Inorganic
Chemicals  Industry," March, 1980
NA = Not Available
                                   823

-------
Sodium Sulfite

Summary of Determinations

It has been determined that no further efforts be given to  developing
or revising regulations for the Sodium Sulfite Subcategory.  The basis
for  this determination is that there are existing regulations for BAT
and NSPS that require zero discharge of process wastewater  pollutants
(40; CFR  415.203  and  415.205).   The  subcategory is excluded under
Paragraph 8 of the Settlement Agreement.  Pretreatment standards  will
be developed in Phase II.

Production Processes and Effluents

Sodium  sulfite  is  produced  by  two  processes.   One is the direct
reaction of soda ash with  sulfur  dioxide.   There  are  four  plants
manufacturing  sodium  sulfite  using  this  process.   In  the  other
process, sodium sulfite is produced as a by-product in the manufacture
of phenol.  Since this process is used primarily to produce phenol and
its derivatives, it is not considered for this subcategory.

Sodium Thiosulfate

Summary of Determinations

It has been determined that no further effort be given  to  developing
BPT, BAT, NSPS, or Pretreatment regulations for the Sodium Thiosulfate
Subcategory.   The  basis  for  this  determination  is  that no toxic
pollutants were found at significant levels in the  raw  waste  during
screening of one plant.  The subcategory is excluded under Paragraph 8
of the Settlement Agreement.

Production Processes and Effluents

Most  of  the  sodium  thiosulfate  is  produced  by the sulfur-sodium
sulfite process.   It  is  used  extensively  in  the  development  of
negatives and prints in the photographic industry.  It is also used in
medicine, in the paper and dyeing industries, and as a bleaching agent
for  natural  products.   Process  wastewater  sources  include filter
backwash and the discharge from barometric condensers.

The subcategory profile data age given in Table 26.40-1.

Toxic Pollutants

Data has been received on about 33 percent of the industry as a result
of Section 308 letters.  A sampling survey at one plant indicated that
toxic pollutants in the effluent are below treatment levels.

Toxic pollutants identified in the effluent were:
                                    824

-------
TABLE  26.36-1  -
SUBCAOEQORY PROFILE DATA SUMMARY
SUBCATEGORY
SODIUM METAL
Total subcategory capacity rate
Tbtal subcategory production rate
Number of plants in this subcategory
308 Data on file for
    With total capacity of
    With total production of
    Representing capacity
    Representing production
    Plant production range:
            Minimum
            Maximum
    Average production
    Median production
    Average capacity utilization
    Plant age range:
            Minimum
            Maximum
    Waste water flow range:
            Minimum
            Maximum
    Volume per unit product:
            Minimum
            Maximum
                           381,000 kkg/year
                               NA
                                5
                                3
                            96,340 kkg/year
                            78,541 kkg/year
                                25 percent
                               NA

                               NA
                               NA
                               NA

                               NA
                               NA

                               NA

                               NA

                               NA
                               NA

                               NA
                               NA
Sources of data are Stanford Research Institute, Directory of Chemical
Producers, U.S.A., 1979, U.S. Department of Conmerce, Current Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry, " June, 1978 and "Economic Analysis of
Proposed Revised Effluent Guidelines and Standards for the Inorganic
Chemicals Industry," March,  1980

NA=  Not Available
                                  825

-------
           Pollutant

           Copper

           Zinc
Concentration Uq/1)

         91

         94
Status of Regulations

Subpart BG has been reserved for this subcategory.

Stannic Oxide

Summary of Determinations

It  has  been  determined  that  no  additional  effort  be  given  to
developing  revised BAT or NSPS regulations for this subcategory.  The
basis for this recommendation is that existing regulation for  BPT  is
zero discharge.  Pretreatment standards will be developed in Phase II.

Production Process and Effluents

Tin is reacted with air and oxygen in a furnace to form stannic oxide.
The  product  is  recovered  with  dry bag collectors and packaged for
sale.  There is no process wastewater from this process.

Plants

There are three plants producing stannic oxide.

BPT Limitations
  *
Regulations for BPT were promulgated on May 22, 1975 (CFR 415.602) and
require  zero  discharge  of  process  wastewater   pollutants.    The
regulations have not been challenged.

BAT and NSPS Limitations

Zero  discharge  regulations were proposed for BAT and NSPS on May 22,
1975.  Since BPT already requires zero discharge,  BAT  and  NSPS  are
being excluded under Paragraph 8 of the Settlement Agreement.

Strong Nitric Acid

Summary of Determinations

It  has  been determined that no further effort be given to developing
regulations for the Nitric Acid (Strong) Subcategory.  The  basis  for
this  determination  is  that no process related  toxic pollutants were
found at significant levels in the process wastewater during screening
of two plants and verification  of  one  plant.   The  subcategory  is
excluded under Paragraph 8 of the Settlement Agreement.
                                   826

-------
 TABLE  26.37-1  -
SUBCATBQOEY PRCFHE DATA StMMAKf
SUBCftlEQOK*
SODIUM SILICATE
Tbtal subcategory capacity rate  (27 PlantsI
Total subcategory production rate
Number of plants in this subcategory
308 Data on file for
    With total capacity of
    With total production of
    Representing capacity
    Representing production
    Plant production range:
            Minimum
            Maximum
    Average production
    Median production
    Average capacity utilization
    Plant age range:
            Minimum
            Maximum
    Waste water flow range:
            Minimum
            Maximum
    Volume per unit product:
            Minimum
            Maximum
                           927,300 kkg/year
                               NA
                               39
                               21
                               NA
                           431,000 kkg/year
                               47 percent
                               NA
                            12,400 kkg/year
                            57,300 kkg/year
                               NA
                               NA
                               NA

                                  7 years
                                 43 years

                               NA
                               NA

                               NA
                               NA
Sources of data are Stanford Research Institute, Directory of Chemical
Producers, U.S.A., 1979, U.S. Department of Commerce, Current Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry," June,  1978 and  "Economic Analysis of
Proposed Revised Effluent Guidelines  and Standards for the Inorganic
Chemicals Industry," March, 1980
NA = Not Available
                                  827

-------
Production Processes and Effluents

Most  of  the  strong nitric acid is produced by dehydration of dilute
nitric acid.  Strong nitric acid is used in the manufacture of organic
compounds where nitric acid acts as an oxiding  agent  instead  of  an
acid.   It  is  also  used in the manufacture of dye intermediates and
explosives.  The principal wastewater source is derived from equipment
washing.  The industry profile data are given in Table 26.42-.
Toxic Pollutants

Toxic pollutants found in the waste streams during sampling of
nitric acid plants were:
                                                                strong
Pollutant

Chromium
Zinc
Lead
Mercury
Silver
Nickel
Cadmium
Cyanide
                       (2
                                   Maximum
                          Concentration Observed (
                          Screening        Verification
                          Plants)       (1 Plant)      !
                          40,000
                             900
                              70
                               8.6
                               0.69
                             5.0
                             2.0
                               0.020
                                              50
                                             120
                                              10
                                               1 .2
                                              15
                                              50
                                               2.0
                                               0.020
In  a  follow-up,  it was found that the chromium and zinc are used as
corrosion inhibitors  in  the  cooling  tower,  and  are  not  process
related.   Control  of these pollutants should involve best management
practices instead of end-of-pipe treatment.

Status of Regulations

Subpart AV has been reserved for this subcategory.

Sulfur Dioxide

Summary of Determinations

It has been determined that no further effort be given  to  developing
BPT,  BAT,  NSPS,  and Pretreatment regulations for the Sulfur Dioxide
Subcategory.  The basis  for  this  determination  is  that  no  toxic
pollutants  were  found  at significant levels in the raw waste during
screening of one plant.  The subcategory is excluded under Paragraph 8
of the Settlement Agreement.

Production Processes and Effluents

Most of the sulfur dioxide  is produced by  air  oxidation  of  sulfur.
The major portion of sulfur dioxide production is in the gaseous form,
although  a  small percentage is also produced in liquid form.  In the
                                   828

-------
TORT.*!  26.38-1  -    SUBCATBQORy PRCFILE DATA SUMMARY"
SUBCAOEGOK*
SODIUM SILICOFLUOKIDE
Total subcategory capacity rate
Total subcategory production rate
Number of plants in this subcategory
308 Data on file for
    With total capacity of
    With total production of
    Representing capacity-
    Representing production
    Plant production range:
            Minimum
            Maximum
    Average production
    Median production
    Average capacity utilization
    Plant age range:
            Minimum
            Maximum
            *
    Waste vZa'Ssr flow range-:
            Minimum
            Maximum
    Volume per unit product:
            Minimum
            Maximum
                               NA
                           51,800 kkg/year
                                6
                                1
                            7,460 kkg/year
                            3,970 kkg/year
                               NA
                              7.5 percent

                               NA
                               NA

                               NA
                               NA
                               NA

                               NA
                               NA

                               NA
                               NA

                               NA
                               NA
Sources of data are Stanford Research Institute, Directory of Chemical
Producers, U.S.A., 1977, U.S. Department of Commerce, Current Industrial
Reports/ December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry ,"  June,  1978 and  "Economic Analysis of
Proposed Revised Effluent Guidelines  and Standards for the Inorganic
Chemicals Industry," March, 1980
NA = Not Available
                                   829

-------
gaseous form,  it is  predominantly  used  in  on-site  manufacture  of
sulfuric acid.   It is also used in the paper and petroleum industries,
as  well  as  for  fermentation  control  in  the  wine  industry, for
bleaching in the textile and food industries, and in the production of
other chemicals.  The wastewater source at one  plant  was  a  process
effluent from an extraction operation.

The subcategory profile data are given in Table 26.43-1.

Toxic Pollutants

Data has been received on about 33 percent of the industry as a result
of Section 308  letters.  No toxic pollutants were found at significant
levels  in  wastewaters  during  the  screening  of one sulfur dioxide
plant.
                                        has  been  reserved  for  this
Status of Regulations

Subpart BI (40 CFR  415.610,   5/22/75
subcategory.

Sulfuric Acid Industry

Summary of Determinations

It  has  been determined that no further effort be given to developing
BPT, BAT, NSPS,  and Pretreatment regulations  for  the  Sulfuric  Acid
Subcategory.    The  basis  for  this  determination  is that the small
quantities of toxic pollutants found during screening  are  far  below
accepted  treatability  levels.   This  determination  applies  to the
production of sulfuric acid by  the  contact  process  from  elemental
sulfur  only.   This  subcategory is excluded under Paragraph 8 of the
Settlement Agreement.

Production Processes and Effluents

Sulfuric acid is one of the most extensivley used of all  manufactured
chemicals.   The  major  industrial use is in the fertilizer industry,
with on-site captive use of the product as a dominant practice.  It is
also used in the manufacturing of  plastics,  explosives,  detergents,
hydrofluoric   acid,  nuclear  fuel  and  several  other  organic  and
inorganic products.  This industry has no process wastewater, but does
have cooling tower blowdown.
The industry profile data for this
26.44-1.

Toxic Pollutants
                                    subcategory  are  given  in  Table
Data has been received on about 21 percent of the industry as a result
of Section 308 letters.  In addition a sampling survey was made at one
plant  which  represents  less  than  1 percent of the industry.  Only
nickel and copper were detected but were at levels far below  accepted
                                    830

-------
TABLE  26.40-1  -
SUBCAIEGORT PRCFILE DATA SUMMARY
 SUBCATEGOKf
SODIUM THIOSULFATE
Total subcategory capacity rate
Total subcategory production rate
Number of plants in this subcategory
308 Data on file for
    With total capacity of
    With total production of
    Representing capacity
    Representing production
    Plant production range:
            Minimum
            Maximum
    Average production
    Median production
    Average capacity utilization
    Plant age range:
            Minimum
            Maximum
    Waste water flew range:
            Minimum
            Maximum
    Volume per unit product:
            Minimum
            Maximum
                              NA
                              NA
                               6
                               5
                           88,000 kkg/year
                           70,300  kkg/year
                              NA
                              NA


                            4,400 kkg/year
                           27,000 kkg/year
                              NA
                              NA
                              NA

                               3 years
                              51 years

                             NA
                             NA

                             NA
                             NA
Sources of data are Stanford Research Institute, Directory of chemical
Producers, U.S.A., 1977, U.S. Department of Commerce, Current Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry ,"  JXine,  1978 and  "Econonic Analysis of
Proposed Revised Effluent Guidelines  and Standards for the Inorganic
Chemicals Industry," March,  1980
NA = Not Available
                                  831

-------
treatability  concentrations.   They probably result from corrosion and
are concentrated by recycling  the cooling water.  Apart from waterside
corrosion some corrosion products results from acid  leaks.   However,
NPDES  permits  generally  require pH control and this will limit this
problem.
Status of Regulations

BPT regulations (40 CFR 415.212) were promulgated on March
These regulations have been remanded by the court.
12,   1974
NSPS  and  BAT  regulations  requiring zero discharge {40 CFR 415.212)
were promulgated on March  12,   1974.   These  regulations  have  been
remanded  by  the  court.   Because no significant quantities of toxic
pollutants are present no further effort will be given to  development
of pretreatment regulations for this subcategory.

Zinc Oxide

Summary of Determinations

It  has  been determined that no further effort be given to developing
BAT, NSPS, or pretreatment regulations  for  Zinc  Oxide  Subcategory.
The  bases  for  this  determination  are 1) only one plant exists the
generates process liquid effluents from the manufacture of zinc  oxide
using  the  wet  chemical process, and 2) processes using oxidation of
zinc produce no waterborne wastes.  The subcategory is excluded  under
Paragraph 8 of the Settlement Agreement.

Production Processes and Effluents

Two  major  processes  are  used for the manufacture of zinc oxide: 1)
those  involving  oxidation  of   zinc,   and   2)   those   involving
precipitation  from solution followed by calcination.  Two methods are
used to prepare zinc oxide using the oxidation process— the  American
Process and the French Process.

In  the  American  Process,  zinc  ore is dried and converted to crude
oxide by roasting at approximately 1000  degrees  C.  Zinc  sinter  is
introduced  into a reaction kiln with equal amounts of coke.  The zinc
vapor and carbon monoxide formed are oxidized to zinc oxide and carbon
dioxide and drawn off through ducts to cyclone and baghouse collection
equipment.  No waterborne wastes are generated.

In the French Process, crude zinc oxide  sinter  and  dried  coke  are
mixed  with  binder  and  fed  through  briquetting  rolls.   The  raw
briquettes are fed into cokers operating at temperatures  between  500
and  1000  degrees C.  Zinc sinter is converted to zinc metal in vapor
using electrical ore vaporizers or rotary burners.   The  zinc  vapors
are purified to remove lead and cadmium impurities.  The purified zinc
is then vaporized and reacted with oxygen to produce zinc oxide, which
is  recovered  by  dry  collection  methods, cooled, and packaged.  No
waterborne wastes are generated.
                                    832

-------
 TABLE 26.42-1  -
SUBCKEEOOHlf PRCFUE DATA SUMMARY
 SUBCA2EGQKT
STRONG NITRIC ACID
Tbtal  subcatecpry capacity rate
Total  subcategory production rate
Number of plants in this subcategory
308 Data on file for
    With total capacity of
    With total production of
    Representing capacity
    Representing production
    Plant production range:
            Minimum
            Maximum
    Average production
    Median production
    Average capacity utilization
    Plant age range:
            Minimum
            Maximum
    Waste water flow range:
            Minimum
            Maximum
    Volume per unit product:
            Minimum
            Maximum
                                NA
                                NA
                                NA
                                5
                          155,200 kkg/year
                          121,000 kkg/year
                                NA
                                NA

                            5,300 kkg/year
                           60,200 kkg/year
                                NA
                                NA

                                NA

                               11 years
                               49 years

                                NA
                                NA

                                NA
                                NA
Sources of data are Stanford Research Institute, Directory of Chemical
Producers/ U.S.A., 1977, U.S. Department of Ccnroerce, Current Industrial
Reports, December 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chemical Industry, " June,  1978 and "Eooncmic Analysis of
Proposed Revised Effluent Guidelines and Standards for the Inorganic
Chemicals Industry/" March,  1980
NA = Not Available
                                  833

-------
In the wet chemical process, crude  zinc  oxide  recovered  from  lead
smelters  is used as the raw material.  The zinc oxide is leached with
caustic soda solution to remove sulfate and dissolve lead saltsr^ The
undissolved  zinc  oxide  is then recovered from the leaching mixture,
washed, neutralized to remove alkali, dried, calcined,  and  packaged.
Waterborne 'wastes  are generated from the deleading step, desulfating
step, etc.

Plants

There are about 20 zinc oxide producers in  the  United  States.   The
producers  include  both  primary and secondary.  The final product of
the primary producers is zinc oxide,  while  the  secondary  producers
manufacture  zinc  oxide  and  use  it  to make zinc chloride and zinc
nitrate.  About 80-90 percent of the total zinc oxide produced is made
by the American and French Processes, and 10 percent is  made  by  one
plant using the wet chemical process.

Zinc Sulfate

Summary of Determinations

It  has  been determined that no further effort be given to developing
revised BAT or NSPS regulations for the Zinc Sulfate Subcategory.  The
basis for this determination is that existing BPT regulations  specify
zero   discharge   of   process   wastewaters   to  navigable  waters.
Pretreatment standards will be developed in Phase II.

Production Processes and Effluents

Zinc sulfate is produced by reaction of  sulfuric  acid  with  various
crude  zinc  starting  materials,  such  as zinc oxide from brass mill
fumes, zinc metal residues from various sources,  and  zinc  carbonate
by-product  from sodium hydrosulfite manufacture.  The following basic
steps are followed: reaction of the zinc containing raw material  with
refiltration  of  solids, and either evaporation to dryness or sale as
solution grade.  Liquors from the preceding  processes  are,  in  some
cases,  refined  to  recover  by-products  and  other  wastewaters are
recycled.  The only wastes are filter cake residues.

Plants

There are eighteen plants producing zinc sulfate and none are known to
discharge wastes from ,the process system.

BPT Limitations

Regulations were promulgated  on  May  22,  1975  (  40  CFR  415.632)
requiring  no  discharge of process wastewater pollutants to navigable
waters.  The  discharge  of  contaminated  non-process  wastewater  is
permitted.    This   includes  rainfall  runoffs,  accidental  spills,
accidental leaks, and discharges related to personal safety equipment.
All reasonable measures must be taken to prevent, reduce  and  control
                                  834

-------
such  contact  to  the  maximum  degree  feasible, and to mitigate the
effects of such contact once it has occurred.

BAT and NSPS Limitations

BAT and NSPS guidelines were proposed on May 22, 1975, requiring  zero
discharge  of  process  wastes to navigable waters.  Since BPT already
requires zero discharge, BAT and NSPS are excluded under  Paragraph  8
of the Settlement Agreement.
                                  835

-------
 TABLE 26.43-1  -
            PRCFUE DATA SUMMARY
 SUBCATEGOR*
SULFUR DIOXIDE
 Total subcategory capacity rate
 Total subcategory production rate
 Number of plants  in this subcategory
 308 Data on file  for
    With total capacity of
    With total production of
    Representing  capacity
    Representing  production
    Plant production  range;
            Minimum
            Maximum
    Average production
    Median  production
    Average capacity  utilization
    Plant age range:
            Minimum
            Maximum
    Waste water flow  range:
            Minimum
            Maxinum
    Volume  per unit product:
            Minimum
            Maximum
                               NA
                               NA
                               15
                                5
                          453,000 kkg/year
                          364,000 kkg/year
                               NA
                               NA

                            27,800 kkg/year
                          170,000 kkg/year
                               NA
                               NA
                               NA


                                3 years
                               51 years

                              NA
                              NA

                              NA
                              NA
 Sources of data are Stanford Research Institute, Directory of Chemical
 Producers, U.S.A., 1977, U.S. Department of Commerce, Current Industrial
 Reports, December 1977; Energy  and Environmental Analysis, Inc.; Draft
 Report, "Preliminary Economic Assesanent of Effluent Limitations in the
 Inorganic Chemical Industry, " JUne, 1978 and "Economic Analysis  of
Proposed Revised Effluent Guidelines and Standards for the Inorganic Chemicals
Industry," March,  1980
NA = Not. Available
                                   836

-------
TABLE  26.44-1  -
SUBCAIEGORY PROFUSE DATA SUMMARY
SUBCATEGOEY
                      SULFURIC ACID
Total subcategory capacity rate
Total subcategory production rate
Number of plants in this subcategory
308 Data on file for
    With total capacity of
    With total production of
    Representing capacity
    Representing production
    Plant production range:
            Minimum
            Maximum
    Average production
    Median production
    Average capacity utilization
    Plant age range:
            Minimum
            Maximum
    Waste water flow range:
            Minimum
            Maximum
    Volume per unit product:
            Minimum
            Maximum
                       33,619,000 kkg/year
                               NA
                              109
                               52
                        7,758,000 kkg/year
                        6,308,000 kkg/year
                               23 percent
                               NA

                            5,300 kkg/year
                           47,700 kkg/year
                               NA
                               NA
                               NA

                                3 years
                               78 years

                               NA
                               NA

                               NA

                               NA
Sources of data are Stanford Research Institute, Directory of Chemical
Producers, U.S.A., 1979 , U.S. Department of Ocnmerce, Current Industrial
Reports, Decenber 1977; Energy and Environmental Analysis, Inc.; Draft
Report, "Preliminary Economic Assessment of Effluent Limitations in the
Inorganic Chanical Industry ," June,  1978 and "Economic Analysis of
Proposed Revised Effluent Guidelines and Standards for the Inorganic
Chemicals Industry," March,  1980
NA = Not Available
                                   837

-------
838

-------
839

-------
840

-------
                              REFERENCES
1.   U.S.  Environmental Protection Agency.  Major Inorganic Products,
Development Document.  EPA-440/l-74-007a,  1974.

2.  U.S Environmental Protection  Agency.   Development  Document  for
Interim  Final Effluent Limitations Guidelines and Proposed New Source
Performance  Standards  for  the   Significant   Inorganic   Products.
EPA-440/1-75-037, 1975.  358 Pp.

3.   Calspan  Corp.  Addendum  to  Development  Document  for Effluent
Limitations Guidelines and New Source  Performance  Standards.   Major
Inorganic  Products Segment of Inorganic Chemicals Manufacturing Point
Source Category,  Contract No. 68-01-3281,  1978.

4.  Sampling Screening  Procedure  for  the  Measurement  of  Priority
Pollutants.  U.S, Environmental Protection Agency, 1976.  6 Pp.

5.   Coleman,  R.T.,  J.D. Colley, R.F, Klausmeiser, D.A. Malish, N.P.
Meserole,, W.C. Micheletti, and K. Schwitzgebel.  Treatment Methods for
Acidic Wastewater Containing Potenially Toxic  Metal  Compounds.   EPA
Contract  No.  68-02-2608, U.S. Environmental Protection Agency, 1978.
220 Pp.

6.  Kraus, K.A., and H.O.  Phillips.   Processes  for  Removal  and/or
Separation  of  Metals  from  Solutions.   U.S. Patent 3,317,312, U.S.
Patent Office, May 2, 1:967.  9 Pp.

7.  Scott, M.C. Heavy  Metals  Removal  at  Phillips  Plating.   WWEMA
Industrial Pollution Conference, St. Louis, Missouri, 1978.  16 Pp.

8.   Scott,  M.C.  Sulfext,  -  A New Process Technology for Removal of
Heavy Metals from Waste .Streams.  The 32nd  Annual  Purdue  Industrial
Waste Conference, Lafayette, Indiana, 1977.  17 Pp.

9.    Patterson,   J.W.,   and   R.A,  Minear.   Wastewater  Treatment
Technology.  Illinois Institite of Technology,, 1973.

10. Patterson,  J.W.   Wastewater  Treatment  Technology.   Ann  Arbor
Science Publishers, Inc.  Ann Arbor, Michigan, 1975.

11.  Schlauch,  R.M.,  and A.C. Epstein.  Treatment of Metal Finishing
Wastes by Sulfide Precipitaion.  EPA-600/2-75-049, U.S.   Environmental
Protection Agency, 1977.  89 Pp.

12.  Campbell,  H.J.,  Jr.,  N.C. Scrivner, K. Batzar, and R.F. White.
Evaluation of Chromium  Removal  from  a  Highly  Variable  Wastewater
Stream.    tcThe   32nd  Annual  Purdue  Industrial  Waste  Conference,
Lafayette, Indiana, 1977.  38 Pp.

-------
13. Wing, R.E., C.L. Swanson, W.M. Doane,  and  C.R.  Russell.   Heavy
Metal Removal with Starch Xanthate-Cationic Polymer Complex.  J. Water
Pollution Control Federation, 46  (8):  2043-2047, 1974.

14.  Wing,  R.E.   Heavy  Metal  Removal  from  Wastewater with Starch
Xanthate, In:  Proceedings of the 29th Annual Purdue Industrial  Waste
Conference, Lafayette, Indiana, 1974.  Pp. 348-356.

15.  Wing,  R.E.   Removal of Heavy Metals from Wastewater with Starch
Xanthate-Cationic Polymer Complex.  The 46th Annual Conference of  the
Water Pollution Control Federation, Cleveland, Ohio, 1973.  38 Pp.

16.  Wing,  R.E.   Removal of Heavy Metals from Wastewater with Starch
Xanthate.  Presented at the Traces of Heavy Metals in Water:   Removal
and Monitoring Conference, Princeton, New Jersey, 1973.  Pp. 258-273.

17.  Swanson,  C.L., R.E. Wing, W.M. Doane, and C.R. Russell.  Mercury
Removal from Wastewater with Starch Xanthate Cationic Polymer Complex.
Environmental Science & Technology 7(7):  614-619, 1973.

18. Hanway, J.E., Jr., R.G. Mumford, and D.G. Barth.  A Promising  New
Process   for   Removing   Heavy   Metals   from   Wastewater.   Civil
Engineering-ASCE 47(10):  78-79,  1976.

19. Hanway, J.E., Jr., R.G. Mumford, and P.N,  Mishra.   Treatment  of
Industrial  Effluents  for  Heavy  Metals  Removal Using the Cellulose
Xanthate Process.  The 71st Annual Meeting of the  American  Institute
of Chemical Engineers, Miami, Florida, 1978.  21 Pp.

20.  Wing,  R.E.,  L.L.  Navickis,  B.K.  Jasberg,  and  W.E. Rayford.
Removal of Heavy Metals from Industrial  Wastewaters  Using  Insoluble
Starch  Xanthate.   EPA-600/2-78-085,  U.S.  Environmental  Protection
Agency,  1978.   116 Pp.

21. DeJong, G.J., and Ir. C.J.N Rekers.   The  Akzo  Process  for  the
Removal  of  Mercury  from Wastewater.  Journal of Chromatography 102:
443-451, 1974.

22. Van der Heem, P.  The Removal  of  Traces  of  Heavy  Metals  from
Drinking  Water  and  Industrial  Effluent  with  Ion Exchangers.  The
Regional American Chemical Society Meeting, 1977.  16 Pp.

23. Chemical Marketing Recorder,  July 24,  1978.

24. Slen, T.T.,  M.  Chem,  and   J.  Lauber.    Incineration  of  Toxic
Chemical Wastes.  Pollution Engineering  10(10):42,  1978.

25.   TRW   Systems   Group.    Recommended   Methods   of  Reduction,
Neutralization,  Recovery  or  Disposal  of  Hazardous  Waste.    NTIS
PB-224589, 1973.
                              842

-------
26.  Ellerbusch,  F,,  and  H.S.  Skrovronek.   Oxidative Treatment of
Industrial  Wastewater.   Industrial  Water  Engineering  14(5):20-29,
1977.

27. Knopp, P.V., and T.L. Randall.  Detoxification of Specific Organic
Substances  by  Wet  Oxidation.   The  51st Annual Conference of Water
Pollution Control Federation, 1978.

28, Arthur D. Little, Inc.  Treatment Technology Handbook.

29.  Schell,  W.J.   Membrane  Ultrafiltration  for  Water  Treatment.
Envirogenics Systems Co.

30.  Vanderborght,  B.M.,  and  R.E. Van Grieken.  Enrichment of Trace
Metals in Water Adsorption of Activated Carbon.  Analytical  Chemistry
49 (2):311-316, 1977.

31.   Cheremisinoff,  P.N.,  and  F.  Ellerbusch.   Carbon  Adsorption
Handbook.  Ann Arbor Science Publishers, Inc.,  Ann  Arbor,  Michigan,
1978.

32.  Jacobs Engineering Group Inc.  Study of the Application of BAC to
Industrial Wastewater.  Office of Water Research and Technology,  U.S.
Environmental Protection Agency,  1978.

33.  Otsubo,  K.,  S.  Yamazaki,  and  Y.  Sakuraba.   Advanced  Water
Treatment  for   Fluoride-Containing   Wastewater.    Hitachi   Hyoron
58(3):219-, 1976.  Trans. For Rockwell Intl.

34.  Zabban,  W.,  and H.W. Jewett.  The Treatment of Fluoride Wastes.
In:   Proceedings  of  the  22nd  Annual   Purdue   Industrial   Waste
Conference, Layfayette,  Indiana,  1967.  Pp. 706-716.

35. Rubel, F., Jr., and R.D. Woosley.  Removal of Excess Fluoride from
Drinking  Water.   EPA-570/9-78-001.   U.S.  Environmental  Protection
Agency,   1978.  16 Pp.

36.  Wu, Y.C.  Activated Alumina  Removes  Fluoride  Ions  From  Water.
Water Sewage Works 125(6):76-82,  1978.

37.  Maier,  F.J.   Partial  Defluoridation of Water.  Public Works 91
(11), 1960.

38. Maier, F.J.  New Fluoride Removal Method Cuts Costs,   Engineering
News-Record 148 (24):40,  1952,

39. Kennedy, D.C., M.A. Kimler, and C,A. Hammer.  Functional Design of
a  Zero-Discharge  Wastewater Treatment System for the National Center
for Toxicoligical Research.   In:   Proceedings  of  the  31st  Annual
Purdue   Industrial  Waste  Conference,  Lafayette, Indiana, 1976.  Pp.
823-830.
                                843

-------
40. Hannah, S.A., M. Jelus, and J.M. Cohen.  Removal of Uncommon Trace
Metals by Physical and Chemical Treatment  Processes.   Journal  Water
Pollution Control Federation 49( 11 ) .-2297-2309, 1977,

41.  Maruyama,  T.,  S.A. Hannah,  and J.M. Cohen.  Removal of Uncommon
Trace Metals by Physical and Chemical  Treatment  Processes.   Journal
Water Pollution Control Federation  49(11):2297-2309, 1977.

42.  Gulledge,  H.H.,  and J.T. O'Connor.  Removal of Arsenic (V) from
Water by  Adsorption  on  Aluminum  and  Ferric  Hydroxides.   Journal
American Water Works Association 65 (8):548-552, 1973.

43. Gupta, S., and K.Y. Chen.  Arsenic Removal by Adsorption.  Journal
Water Pollution Control Federation  50 (3}:493, 1978.

44.  Larsen,  H.P.,  J.K.  Shou, and L.W. Ross.  Chemical Treatment of
Metal-Bearing  Mine  Drainage.   Journal   Water   Pollution   Control
Federation 45 (8):1682-1695, 1973

45.  Nilsson, R.  Removal of Metals by Chemical Treatment of Municipal
Wastewater.  Water Research 5:51-60, 1971

46. Sorg, T.J., O.T. Love, and  G.S.  Logsdon.   Manual  of  Treatment
Techniques for Meeting the Interim  Primary Drinking Water Regulations.
EPA-600/8-77-005.  U.S. Environmental Protection Agency, 1977.  73 Pp.

47.  Colley,  J.D.,  C.A. Muela, M.L. Owen, N.P. Meserole, J.B. Riggs,
and J.C.  Terry.   Assessment  of   Technology  for  Control  of  Toxic
Effluents from the Electric Utility Industry.  EPA-600/7-78-090.  U.S.
Environmental Protection Agency, 1978.

48. Smithson, G.R., Jr.  An Investigation of Techniques for Removal of
Chromium   from   Electroplating    Wastes.    EPA    12010  EIE,   U.S.
Environmental Protectin Agency, 1971, 91 Pp.

49.  Patterson,   J.W.,  H.E.  Allen,  and   J.J.   Scala.    Carbonate
Precipitation  for  Heavy  Metals Pollutants.  Journal Water Pollution
Control Federation 49{12) : 2397-2410, 1977.

50. Sabadell, J.E.  Traces of Heavy Metals in Water Removal  Processes
and  Monitoring.   EPA-902/9-74-001.   U.S.  Environmental  Protection
Agency, 1973.

51. U.S. Environmental Protection Agency.   Environmental  Multi-Media
Assessment  of  Selected  Industrial Inorganic Chemicals.  EPA Contract
No. 68-03-2403, 1977.

52. JRB Associates,  Inc,  An  Assessment  of  pH  Control  of  Process
Waters  in  Selected   Plants.   Draft  Report  to  the Office of Water
Programs, U.S. Environmental Protection Agency,  1979.

53. Energy and Environmental  Analysts,   Inc.   Economic  Analysis  of
Proposed  Revised Effluent Guidelines and Standards for the  Inorganic
                               844

-------
Chemicals  Industry.   Prepared  for  Office  of  Water  Planning  and
Standards, U.S.  Environmental Protection Agency, 1980.

54.  Kirsch,   E.J., J.E. Etzel  J. Water Pollution Control Federation,
Vol. 45, 359, 1973.

55. E.I.  Dupont  de  Nemours  &  Company,  Development  Document  for
Effluent  Limitations  Guidelines and New Source Performance Standards
for   Simultaneous   Beneficiation-Chlorination   Process   of    Ti02
Manufacture,  Pigments Department, Edge Moor, Delaware, 1977.

56.  U.S.  Environmental  Protection  Agency, Prodeedings:  Seminar on
Analytical Methods for Priority Pollutants, Denver.  1977.

57. SRI International, Report No.  61B,  Chlorine,  Supplement  B,  by
Yen-Chen  Yen,  (a  private  report by the Process Economics Program),
Menlo Park,  California, 1978.

58. U.S. Environmental Protection  Agency,  Development  Document  for
Proposed  Effluent  Limitations  Guidelines,  New  Source  Performance
Standards and Pretreatment Standards  for  the  Steam  Electric  Point
Source  Category,   (Effluent  Guidelines Division, Office of Water and
Waste Management,  Washington, D.C.), 1979.

59, U.S. Environmental Protection Agency, Quality Criteria for  Water,
(EPA-44/9-76-023), Washington, D.C., 1976.

60.  U.S.  Environmental  Protection  Agency, Development Document for
Proposed  Effluent  Limitations  Guidelines  and  Standards  for   the
Inorganci  Chemicals  Manufacturing  Point  Source Category, {Effluent
Guidelines Division, Office of Water and Waste Management, Washington,
D.C.), 1980.

61. U.S. Environmental Protection Agency, Treatability Studies for the
Inorganic Chemicals Manufacturing  Point  Source  Category,  (Effluent
Guidelines Division, Office of Water and^Waste Management, Washington,
D.C.), 1980.

62.   E.  I.   duPont  de  Nemours  &  Company,  Comments  on  Proposed
Development  Document for Effluent Limitations Guidelines and Standards
for the Titanium Dioxide - Chloride Ilmenite  Process  Subcategory  of
the  Inorganic  Chemicals Point Source Category, Wilmington, Delaware,
1980.

63. U.S. Environmental Protection Agency, Economic Impact Analysis  of
Pollution Control  Technologies for Segments of the Inorganic Chemicals
Manufacturing Industry.  EPA 440/2-81-023, 1982.
                                84!

-------
846

-------
                             BIBLIOGRAPHY
Andco  Environmental  Processes,  Inc.  Andco Chromate and Heavy Metal
Removal System.  1977

Anderson, J.R., and C.O. Weiss.  Method  for  Precipitation  of  Heavy
Metal Sulfides.  U.S. Patent 3,740,331.  U.S. Patent Office.  June 19,
1973.
Birkett,  J.D.   Electrodialysis
Engineering 14 (5):6-9, 1977.
An  Overview.   Industrial  Water
Bowen, L.B., J.H. Mallison, and J.H. Cosgrove.  Waste Recovery:   Zinc
Recovery  from  Rayon  Plant  Sludge.   Chemical  Engineering Progress
73(5):50-54, 1977.

Case, O.P.  Metallic Recovery from Wastewaters Utilizing  Cementation.
EPA 670/2-74-008.  U.S. Environmental Protection Agency, 1974.  36 Pp.

Cheng,  M.H.,  J.W. Patterson, and R.A. Minear.  Heavy Metals Uptake by
Activated  Sludge.   Journal  Water   Pollution   Control   Federation
47(2) :362-376, 1975.

Cheremisinoff,   P.N.,  and  Y.H.  Habib.   Cadmium,  Chromium,  Lead,
Mercury: A plenary  Account  for  Water  Pollution.   Part  2  Removal
Techniques.  Water and Sewage Works 119(8):46-51, 1972.
Cohen,   J.M.  Trace  Metal  Removal  by  Wastewater  Treatment.
Technology Transfer, 1977.
                              EPA
Current Industrial Reports Inorganic Chemicals.   U.S.  Department  of
Commerce Bureau of the Census, 1977.

Davis,  H.J., F.S. Model, and J.R. Leal.  PBI Reverse Osmosis Membrane
for   Chromium   Plating   Rinse   Water.    EPA-600/2-78-040.    U.S.
Environmental Protection Agency, 1978.   35 Pp.

Dean,  J.G.,  F.L.  Bosqui, and K.H. Lanouette.  Removing Heavy Metals
from Wastewater.  Environmental Science and  Technology  6(6}:518-522,
1972.

Dow  Chemical  Company.   Selecting  Ion  Exchange  Resins  for  Water
Treatment Applications.   Industrial  Water  Engineering  14(5):11-13,
1977.

Energy  and  Environmental  Analysis, Inc.  Draft Preliminary Economic
Assessment  of  Effluent  Limitations . in  the   Inorganic   Chemicals
Industry.  1978.

FMC  Corp.   Industrial Wastewater Treatment.  A guidebook to Hydrogen
Peroxide for Industrial Wastes.  Philadelphia, Pennsylvania.
                                  847

-------
FMC  Corp.   Hydrogen  Peroxide.
Philadelphia, Pennsylvania.
Chemical
and   Technical   Data.
Ford,   D.L.    Putting   Activated  Carbon  in  Perspective  to  1983
Guidelines.  Industrial Water Engineering 14(3):20-27, 1977.

General Technology Corporation.  Draft Contractor Document  for  Major
Inorganic Chemicals.  1973.

General  Technology  Corporation.  Final Contractor Document for Major
Inorganic Chemicals.  1973.

Greek, B.F., and W.F. Fallwell.  Chlorine,  Major  Alkalies  Still  in
Doldrums.  Chemical and Enginering News 56(6):8-ll, 1978.

Grover,  P.   A  Waste Stream Management System.  Chemical Engineering
Progress 73(12):71-73, 1977.

Haggenmacher, J.H.  Chromate Removal Decreases Slowdown to Below  0,05
ppm Hexavalent Chrome.  Chemical Processing, 1977.

Hoyle,   D.L.    Designing   for  pH  Control.   Chemical  Engineering
83(24):121-126, 1976.

Jacobs Engineering Group Inc.  Estimating Standards.

Jenkins, S.H., D.G. Keight, and R.E.  Humphreys.   The  Solubility  of
Heavy  Metal  Hydroxides   in Water, Sewage and Sewage Sludge - I.  The
Solubi1ity  of  Some  Metal  Hydroxides.   Int.    J.  Air  Wat.  Pol1.
8:537-556, 1964.

Jenkins,  S.H,,  D.G.  Keight,  and A. Ewins.  The Solubility of Heavy
Metal  Hydroxides  in  Water,  Sewage,  and  Sewage  Sludge  -  II  The
Precipitation   of   Metals  by  Sewage.   Int.    3.  Air  Wat.  Poll.
8:679-693, 1964.

Keating, E.J., R.A. Brown,  and  E.S.  Greenberg.   Phenolic  Problems
Solved  with  Hydrogen Peroxide Oxidation.  The 33rd Annual Industrial
Waste  Conference, West Lafayette, Indiana, 1978.  20 Pp.

Kibble, W.H.  Hydrogen  Peroxide  Helps  Solve  Industrial  Wastewater
Problems.  Industrial Wastes 26-29, 1978,

Knocke,  W.R.,  T.  Clevenger, M.M. Ghosh, and J.T. Novak, Recovery of
Metals from Electroplating Wastes.  The 33rd Annual  Industrial  Waste
Conference, West Lafayette, Indiana.  1978.  37 Pp.

Lanuouette, K.H.  Heavy Metals Removal.  Chemical Engineering Deskbook
Issue  84{22):73-80, 1977.

Linstedt, K.D., C.P. Houck, and J.T. O'Connor.  Trace Element Removals
in  Advanced  Wastewater Treatment Processes.  Journal Water Pollution
Control Federation 43(7):1507-1513, 1971.
                                  848

-------
Mir, L.,  W.  Eykamp,  and  R.L.  Goldsmith.   Current  and  Developing
Applications   for   Ultrafiltration.   Industrial  Water  Engineering
14(3):14-19, 1977.

Morton, S.D., and E.W. Sawyer.  Clay Minerals Remove Organics, Viruses
and Heavy Metals from Water.  Water and Sewage  Works  123  (Reference
No. 1976):R117-R120, 1976.

Porter,  J.W.,   Brothers,  G.W.,  and  W.B.  Whitton.  Cost Estimating
Guidelines for Wastewater Treatment Systems.   EPA  17090  DRU.   U.S.
Environmental Protection Agency, 1970.  96 Pp.
Resources  Conservation  Co.
Washington
The  RCC  Brine  Concentrator.  Renton,
Richardsons Estimating Standards.  Means Cost Data.  1978.

Rizzo, J.L,,  and A.R. Shepard.  Treating  Industrial  Wastewater  with
Activated Carbon.  Chemical Engineering 84(1):95-100, 1977.

Rosenzweig,  M.D.  Mercury  Cleanup  Routes  -  I Chemical Engineering
82(2):60-61,  1975.

Sabadell, J.E,, ed. Traces of Heavy Metals in Water Removal  Processes
and  Monitoring.   EPA-902/9-74-001.    U.S.  Environmental  Protection
Agency, 1973.  342 Pp.

Shen, Y.S.  Study of Arsenic Removal   from  Drinking  Water.   Journal
American Water Works Association 65(8):  543-548, 1973.

Sigworth,  E.A., and S.B. Smith.  Adsorption of Inorganic Compounds by
Activated  Carbon.    Journal   American   Water   Works   Association
64(6):386-390, 1972.

Smith,  R.  Cost of Conventional and Advanced Treatment of Wastewater.
Journal Water Pollution Control Federation 40(9):1546-1574, 1968.

Sonksen, M.K., P.M. Sittig, E.F. Maziarz.   Treatment of Oily Wastes by
Ultrafiltration/Reverse Osmosis.  A Case  History.   The  33rd  Annual
Industrial Waste Conference, West Lafayette, Indiana, 1978.  28 Pp.

Stanford  Research  Institute.  Directory of Chemical Producers U.S.A.
1977.

Stinson, G.F.  Electrochemical Wastewater Treatment Process.  Personal
Communication.  1978.

Theim, L.D. Badorek, and  J.T.  0'Connor.    Removal  of  Mercury  From
Drinking  Water  Using Activated Carbon.  Journal American Water Works
Association 68 (8):447-451, 1976.
                                  84 9

-------
Thompson, G.S., Jr.  First Annual  Conference  on  Advanced  Pollution
Control  for  the  Metal  Finishing Industry.  EPA-600/8-78-010.  U.S.
Environmental Protection Agency, 1978.  143 Pp.

S.K. Williams Company.  Wastewater Treatment  and  Reuse  in  a  Metal
Finishing  Job Shop.  EPA-670/2-74-042.  U.S. Environmental Protection
Agency, 1978.  143 Pp.
Wiseman, S.,  and B.T.  Bawden.   Removing  Refractory  Compounds
Wastewater.  Chemical Engineering Progress 73(5):60-64, 1977.
From
Zabban,  W.,   and R. Helwick.  Defluoridation of Wastewater.  The 30th
Annual Industrial Waste Conference, West Lafayette, Indiana, 1975.  38
Pp.

Zievers, J.F., R.E. Wing, S. Grain, and D. Gordon.  Practical  Use  of
Starch Insoluble Xanthates.  Industrial Filter and Pump Mfg. Co.  1977.
20 Pp.
                                  850

-------
                              APPENDIX A

                    ANALYSIS OF LONG-TERM EFFLUENT
                         MONITORING DATA FOR
                   THE INORGANIC CHEMICALS INDUSTRY
Introduction

This   appendix   contains  tabulated  summaries  of  the  statistical
parameters derived from the analysis of long-term effluent  monitoring
data collected by industry and reported to the EPA or State regulatory
agencies  during  the last two or three years.  The particular sets of
data selected for analysis are taken from plants which  apply  a  well
defined  treatment  technology  to  process  wastewaters  from  single
product or product group manufacturing operations  associated  with  a
specific   subcategory.   Data  have  been  excluded  which  represent
wastewaters diluted with noncontact cooling water or  commingled  with
waste  sources  from  unrelated  products.  Each table in the appendix
indicates the actual number of observations on  which  the  calculated
statistical  parameters  are  based.  The derivation of the parameters
was discussed in Section 8.2 of the development document.

The statisitcal performance information presented here was taken  into
consideration  in  the development of limitations for each subcategory
studied in detail in the main report.  These  were  expressed  as  the
Concentration  Bases  (mg/1)  and Effluent Unit Mass Loadings (kg/kkg)
for each pollutant  assuming  the  model  plant  flow  conditions  and
applying the specified pollutant removal technologies at e'ach level of
treatment.   The tables on the following pages summarize the available
historical effluent monitoring results and give the  individual  plant
performance  characteristics  in  concentration  and loading units for
both daily and monthly measurements.   Variability  factors  shown  on
each  table  were  used to calculate the plant "Performance Standards"
shown in  the  right  hand  column  of  each  table.   Similarly,  the
Variability  Factor Ratio (VFR) used in this report is the variability
factor for daily measurements divided by variability factor for 30-day
average data.

In general, the effluent data base acquired included NPDES  monitoring
data collected during the period from January 1, 1975 through June 30,
1976.   Firms  who  monitored  over this time period provided up to 18
months of 30-day average data and as many as 547 measurements of daily
or 24-hour data.   In cases where monitoring was done  less  frequently
than  daily,  perhaps omitted on weekends,  or only weekly measurements,
the actual number of observations used in the calculation is  recorded
for each parameter.

Included  in  Appendix  A  are statistical measures appropriate to the
analysis of long-term monitoring data and the  historical  performance
of  inorganic  chemical  pollutant  discharge  levels.   The statistics
presented include  measures   of  the  amount  or  level  of  pollutant
                                A-l

-------
discharge, such as long-term average, minimum level, and maximum level
for  both  daily,  or  24-hour measurements, as well as 30-day average
measurements.

Also given in the table is the coefficient  of  variation,  CV,  which
reflects  the dispersion of measurements above and below the long-term
average level.  Other measures of variability that may be of interest,
such as range or  standard  deviation  are  also  calculated  for  any
parameter   from   any  information  given  herein.   In  addition  to
statistics of  pollutant  level  and  variation  of  pollutant  level,
variability  factors  are  given  for  each  parameter.  A variability
factor is the ratio of an upper  percentile  of  the  distribution  of
pollutant  measurements to the long-term average pollutant level.  The
basis of  the  particular  upper  percentile  chosen  for  variability
factors is explained as a footnote to the table.

The historical performance of each firm, using the variability factor,
is  given for each parameter and is expressed in the same units as the
long-term average.

For reference, the tables in Appendix A  are  organized  by  inorganic
chemical   subcategory   and   the   manufacturing   process  in  that
subcategory.  For each plant, as many  as  six  tables  are  included.
These tables appear in the following order.

1.   Daily measurements of pollutant concentrations in effluent stream
given in parts per million (ppm).

2.  Daily measurements of - total effluent discharge  load  measured  in
kilograms per day.

3.  30-day averages of pollutant concentration  (ppm).

4.  30-day averages of total effluent pollutant load (kg/day).

5.   Daily  measurements  of  pollutant  unit loadings in the effluent
streams given in kilograms of  pollutant  per  thousand  kilograms  of
product (kg/kkg).

6.   30-day averages of pollutant unit loadings in the effluent stream
given in  kilograms of pollutant  per  thousand  kilograms  of  product
(kg/kkg).
                               A-2

-------
                           Table A-la

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                      Subcategory Chlorine
                      Mercury Cell Process
                            Plant tA
                 Historical Summary      Variability  Performance
 Parameter         Statistics              Factors     Standards

   (mg/1)      No   Min   Avg   Max   CV          *            p


Mercury      530  .006  .014  .021  .286      1.88         .026

TSS          530 1.00  7.4  62.     .581      3.04       22.5

Chlorine     428 0.08   .638  1.50  .463      2.28         1.46

 (Total Residual)


  * - 99% of the daily maximum measurements expected to be less
      than the performance standard, p.

                           Table A-lb
           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                      Subcategory Chlorine
                      Mercury Cell Process
                            Plant #A



               Historical Summary      Variability   Performance
 Parameter         Statistics            Factors      Standards

   (kg/day)   No  Min  Avg   Max  CV          *             P


Mercury      530 .015 .031  .047 .129      1.66          .051

Chlorine     420 .156 1.44 3.40  .463      2.54         3.65
 (Total Residual)
  * _
      99% of the daily maximum measurements expected to be less
      than the performance standard, P.
                              A-3

-------
                     Table A-lc

     Historical Effluent Monitoring  Data Summary
 with Variability Factors and Performance Standards
                   30 Day Averages
                Subcategory Chlorine
                Mercury Cell Process
                      Plant #A
Parameter
(rag/1)
Mercury
TSS
Historical Summary Variability
Statistics Factors
No Min Avg Max CV *
18 .008 .014 .020 ,293 1.47
18 5.1 7.4 12.9 .355 1.58
Chlorine 18 .380 .638 .847 .194 1.38
(Total Residual)
Performance
Standards
P
.021
11.7
0.88
95% of the monthly averages are expected to be within the
performance standard,  P.

                    Table A-ld

     Historical Effluent  Monitoring Data Summary
 with Variability Factors and Performance Standards
                   30  Day Averages
                Subcategory Chlorine
                Mercury Cell Process
                      Plant #A
         Historical Summary
Variability   Performance
Parameter
(kg/day)
Mercury
Chlorine

No
18
18
St
Min
.020
.91 1
atist
Avg
.031
.44 2
ics
Max CV
.037 .197
.23-
Factors
*
1.33
1.50
Standards
P
.041
2.16
95% of the monthly averages are expected to be within the
performance standard, P.
                        A-4

-------
                           Table A-le

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                      Subcategory Chlorine
                      Mercury Cell Process
                            Plant #A
 Parameter

    (g/kkg)  No  Min
         Historical Summary
             Statistics
           Variability   Performance
             Factors      Standards
                Avg
Max  CV
Mercury
      530 .027   .055  .084
Chlorine    420 .00028 .0026 .006
(Total Residual)
                             .090

                             .006
  * -
99% of the daily maximum measurements expected to be less
than the performance standard,  P.

                     Table A-lf

     Historical Effluent Monitoring Data Summary
 with Variability Factors and Performance Standards
                   30 Day Averages
                Subcategory Chlorine
                Mercury Cell Process
                      Plant fA
 Parameter
         Historical Summary
             Statistics
              No  Min  Avg  Max  CV
           Variability
             Factors
Performance
 Standards
Mercury      18  .035  .055 .065

Chlorine     18 1.6   2.52 3.91
                                                  .072

                                                 3.8
  * - 95% of the monthly averages are expected to be within the
      performance standard, P.
                              A-5

-------
                           Table A-2a

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                      Subcategory Chlorine
                      Mercury Cell Process
                            Plant #B
 Parameter

  (ug/1)
   Historical Summary
       Statistics
Variability   Performance
  Factors      Standards
  No  Min  Avg  Max  CV
Mercury
 516 .041 .634 2.87 .910
   4.52
  2.87
      99% of the daily maximum measurements expected to be less
      than the performance standard, P.
                           Table A-2b

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                      Subcategory Chlorine
                      Mercury Cell Process
                            Plant #B
 Parameter

  (kg/day)
   Historical Summary
       Statistics
Variability
  Factors
Performance
 Standards
  No  Min  Avg  Max  CV
Mercury
516 .0005 .011 .088 .818
   4.35
  .046
    - 99% of the daily maximum measurements expected to be less
      than the performance standard, P.
                              A-6

-------
                           Table A-2c

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                      Chlorine Subcategory
                      Mercury Cell Process
                            Plant #B
 Parameter

  (ug/1)
 Historical Summary
     Statistics
Variability   Performance
  Factors      Standards
No  Min  Avg  Max  CV
Mercury
17 .325 .634 1.15 .293
    1.45
0.919
  * - 95% of the monthly averages are expected to be within the
      performance standardr P.
                           Table A-2d

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                      Subcategory Chlorine
                      Mercury Cell Process
                            Plant IB
 Parameter

 (kg/day)
 Historical Summary
     Statistics
Variability   Performance
  Factors      Standards
 No  Min  Avg  Max  CV
Mercury
 17 .005 .011 .019
    1.45
.015
  * - 95% of the monthly averages are expected to be within the
      performance standard, P.
                             A-7

-------
                           Table A-2e

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                      Subcategory Chlorine
                      Mercury Cell Process
                            Plant fB
 Parameter

   (9/kkg)
   Historical Summary
       Statistics
Variability
  Factors
Performance
 Standards
  No  Min  Avg  Max  CV
Mercury
516 .0037 .082 .658
                .344
    - 99% of.the daily maximum measurements expected to be less
      than the performance standard, P.
                           Table A-2f

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                      Subcategory Chlorine
                      Mercury Cell Process
                            Plant tfB
 Parameter

 (g/kkg)
   Historical Summary
       Statistics
Variability
  Factors
Performance
 Standards
  No  Min  Avg  Max  CV
Mercury
  17 .037 .082 .14
                ,11
  * - 95% of the monthly averages are expected to be within the
      performance standard, P.
                              A-8

-------
                           Table A-3a

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                      Subcategory Chlorine
                      Mercury Cell Process
                            Plant fC
 Parameter

   (mg/1)
   Historical Summary
       Statistics
Variability   Performance
  Factors      Standards
  No  Min  Avg  Max  CV
Mercury
349 .0005 .014 .136 2.29
   9.45
  0.132
  * - 99% of the daily maximum measurements expected to be less
      than the performance standard,  P.
                           Table A-3b

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                      Subcategory Chlorine
                      Mercury Cell Process
                            Plant fC
 Parameter

  (kg/day)
   Historical Summary
       Statistics
Variability
  Factors
Performance
 Standards
  No  Min  Avg  Max  CV
Mercury
349 .0001 .003 .088 2.33
   10.22
   .028
  * - 99% of the daily maximum measurements expected to be less
      than the performance standard,  P.
                             A-9

-------
                           Table A-3c

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                      Subcategory Chlorine
                      Mercury Cell Process
                            Plant fC
 Parameter

  (mg/1)
              Historical Summary
                  Statistics
Variability
  Factors
Performance
 Standards
             No  Min  Avg  Max  CV
Mercury
            17 .0009 .014 .062 1.21
   2.99
  .042
      95% of the monthly averages are expected to be within the
      performance standard, P.
                           Table A-3d

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                      Subcategory Chlorine
                      Mercury Cell Process
                            Plant *G
               Historical Summary
                   Statistics
Parameter

 (kg/day)    No  Min  Avg  Max  CV
Variability   Performance
  Factors      Standards
Mercury
            17 .0002 .003 .014 1.33
   3.22
 .0088
  * - 95% of the monthly averages are expected to be within the
      performance standard, P.
                             A-10

-------
                           Table A-3e

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                      Subcategory Chlorine
                      Mercury Cell Process
                            Plant fC
 Parameter
   Historical Summary
       Statistics
Variability
  Factors
Performance
 Standards
              No  Min  Avg  Max  CV
Mercury
349 .0006 .017 .485
                 .154
  * - 99% of the daily maximum measurements expected to be less
      than the performance standard, P.
                           Table A-3f

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                      Subcategory Chlorine
                      Mercury Cell Process
                            Plant #C
 Parameter

   (g/kkg)
   Historical Summary
       Statistics
Variability
  Factors
Performance
 Standards
  No  Min  Avg  Max  CV
Mercury
 17 .0011 .016  .077
               .0485
  * - 95% of the monthly averages are expected  to be within the
      performance standard, P.
                             A-ll

-------
                           Table A-4a

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                      Subcategory Chlorine
                      Mercury Cell Process
                            Plant #D
               Historical Summary
                         Variability   Performance
Parameter Statistics Factors
(mg/1) No Min Avg Max CV *
Mercury
82 .002 .004 .011 .500 2.24
Chlorine 49 2.0 19.1 62 1.01 4.96
(Total Residual)
Standards
P
0.009
94.7
      99% of the daily maximum measurements expected to be less
      than the performance standard,  P.
                           Table A-4b

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                      Subcategory Chlorine
                      Mercury Cell Process
                            Plant #D
 Parameter

  (kg/day)
 Historical Summary
     Statistics
Variability
  Factors
No  Min  Avg  Max  CV
Performance
 Standards
Mercury
82 .021 .047 .118 .383
Chlorine      49 20.5 203  663 1.03
(Total Residual)
   2.20

   5.04
    .104
                                       1026
      99% of the daily maximum measurements expected to be less
      than the performance standard, P.
                             A-12

-------
                     Table A-4c

     Historical Effluent Monitoring Data Summary
 with Variability Factors and Performance Standards
                   30 Day Averages
                Subcategory Chlorine
                Mercury Cell Process
                      Plant #D
         Historical Summary
Variability   Performance
Parameter
(mg/1)
Mercury
Statistics
No
22
Chlorine 14
(Total Residual)
Min
.003
4.0
Avg
.004
19.1
Max
.008
57.8
Factors Standards
CV * p
.250 1.60 0.006
.969 2.91 55.6
95% of the monthly averages are expected to be within the
performance standard, P.
                     Table A-4d

     Historical Effluent Monitoring Data Summary
 with Variability Factors and Performance Standards
                   30 Day Averages
                Subcategory Chlorine
                Mercury Cell Process
                      Plant fD
         Historical Summary
Variability   Performance
Parameter
(kg/day)
Mercury
Statistics Factors Standards
No
22
Chlorine 14
(Total Residual)
Min
.032
39.1
Avg
.047
203
Max cv * P
.098 .340 1.66 .079
616 .945 2.89 588
95% of the monthly averages are expected to be within the
performance standard,  P.
                       A-13

-------
                           Table A-4e

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                      Subcategory Chlorine
                      Mercury Cell process
                            Plant |D
 Parameter

   (gAkg)
 Historical Summary
     Statistics
Variability   Performance
  Factors      Standards
No  Min  Avg  Max  CV
Mercury
82 .386 .864 2.17
                1.91
Chlorine
(Total Residual)
      99% of the daily maximum measurements expected to be less
      than the performance standard, P.
                           Table A-4f

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                      Subcategory Chlorine
                      Mercury Cell Process
                            Plant fD

 Parameter
 Historical Summary
     Statistics
              No  Min  Avg  Max  CV
Variability   Performance
  Factors      Standards
Mercury
22 .588 .864 1.8
Chlorine
(Total Residual)
                1.45
  * - 95% of the monthly averages are expected to be within the
      performance standard, P.
                             A-14

-------
                           Table A-5a

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                      Subcategory Chlorine
                      Diaphram Cell Process
                            Plant fE

 Parameter

 (kg/day)
  Historical Summary
      Statistics
Variability
  Factors
Performance
 Standards
 No  Min  Avg  Max  CV
Lead
153 .045 1.42 5.40
   4.12
  5.85
      99% of the daily maximum measurements expected to be less
      than the performance standard, P.
                           Table A-5b

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                      Subcategory Chlorine
                      Diaphram Cell Process
                            Plant flE
Historical Summary Variability
Parameter Statistics Factors
(kg/day) No Min Avg Max CV *
Performance
Standards
P
Lead
 12 .460 1.42 5.40 .824
   1.58
  2.25
  * - 95% of the monthly averages are expected to be within the
      performance standard, P.
                              A-15

-------
                           Table A-5e

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                      Subcategory Chlorine
                      Diaphram Cell Process
                            Plant #E
 Parameter

 (g/kkg)
  Historical Summary
      Statistics
Variability
  Factors
Performance
 Standards
 No  Min  Avg  Max  CV
Lead
153 .205 6.46 24.6
                26.6
  * - 99% of the daily maximum measurements expected to be less
      than the performance standard, P.
                           Table A-5f

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                      Subcategory Chlorine
                      Diaphram Cell Process
                            Plant fE
 Parameter

 (g/kkg)
  Historical Summary
      Statistics
Variability
  Factors
Performance
 Standards
 No  Min  Avg  Max  CV
Lead
 12 2.09 6.46 24.6
               10.24
  * - 95% of the monthly averages are expected to be within the
      performance standard, P.
                              A-16

-------
                         Table A-7a

         Historical Effluent Monitoring Data Summary
     with Variability Factors and Performance Standards
                       30 Day Averages
               Subcategory Hydrofluoric Acid/
                          Plant $G

             Historical Summary
Variability   Performance
Parameter
(kg/day)
Fluoride
TSS
Statistics
No
15
16
Min
4.54
7.26
Avg
16
28
.7
.6
Max
27
52
.2
.2
Factors
CV *
.449 X.74
.441 1.72
Standards
P
29.0
49.2
* - 95% of the monthly averages are expected to be within the
    performance standard,  P.
                         Table A-7e

         Historical Effluent Monitoring Data Summary
     with Variability Factors and Performance Standards
                       30 Day Averages
               Subcategory Hydrofluoric Acid/
                          Plant #G

             Historical Summary
Variability   Performance
Parameter
(g/kkg)
Fluoride
TSS
Statistics Factors Standards
No
15
16
Min Avg
99.1 365
158.5 624
Max CV *
594
1140
P
633
1074
* - 95% of the monthly averages are expected  to be within the
    performance standard,  P.
                           A-17

-------
                       Table A-8b

       Historical Effluent Monitoring  Data Summary
   with Variability Factors and Performance Standards
                   Daily Measurements
              Subcategory Titanium Dioxide
                    Chloride Process
                        Plant ffH
           Historical Summary
Variability   Performance
Parameter
(kg/day)
Chromium
Copper
Zinc
TSS

No
394
394
394
394
Stc
Min
.000
.000
.000
0.40
atist:
Aver
.013
.027
.028
8.34
Lcs
Max
.210
.190
.108
176.

CV
1.69
1.04
.679
1.92
Fa

7
5
3
8
ctors
*
.78
.20
.42
.35
Standards
P
.097
.139
.097
69.7
- 99% Of the daily maximum measurements expected to be less
  than the performance standard,  P.
                         A-18

-------
                           Table A-8c

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                  Subcategory Titanium Dioxide
                        Chloride Process
                            Plant £H
               Historical Summary
Variability   Performance
Parameter
(mg/1)
Chromium
Copper
Zinc
TSS
Statistics
No
13
13
13
13
Min
.000
.000
.001
1.20
Aver
.004
.010
.012
3.14
Max
.013
.030
.026
8.60
CV
.750
.700
.500
.599
Factors

2
2
1
1
*
.46
.43
.93
.98
Standards
P
0.010
0.024
0.023
6.22
* - 95% of the monthly averages are expected to be within the
    performance standard,  p.
                             A-19

-------
                       Table A-8d

       Historical Effluent Monitoring Data Summary
   with Variability Factors and Performance Standards
                     30 Day Averages
              Subcategory Titanium Dioxide
                    Chloride Process
                        Plant #H
Parameter
(kg/day)
Chromium
Copper
Zinc
TSS
Historical Summary
Statistics
No
13
13
13
13
Min
.002
.000
.004
2.60
Avg
.013
.027
.028
8.34
Max
.043
.100
.051
24.0
CV
.769
.852
4.29
.695
Var iabili ty Per f ormance
Factors Standards
* P
2.62 .033
2.74 .073
1.80 .051
2.14 17.9


- 95% of the monthly averages are expected to be within the
  performance standard, P.
                         A-20

-------
                         Table A-8e

         Historical Effluent Monitoring Data Summary
     with Variability Factors and Performance Standards
                     Daily Measurements
                Subcategory Titanium Dioxide
                      Chloride Process
                          Plant fH
             Historical Summary
Variability   Performance
Parameter
(g/kkg)
Chromium
Copper
Zinc
TSS
Statistics Factors Standards
No
394
394
394
394
Min
.000
.000
.000
5.49
Aver
.178
.37
.384
114
Max CV
2
2
1
.88
.6
.48
2415
*
1
1
1
P
.33
.9
.33
956
* - 99% of the daily maximum measurements expected to be less
    than the performance standard,  p.
                           A-21

-------
                         Table A-8f

         Historical Effluent Monitoring Data Summary
     with Variability Factors and Performance Standards
                       30 Day Averages
                Subcategory Titanium Dioxide
                      Chloride Process
                          Plant fH
             Historical Summary
Variability   Performance
Parameter
(9/kkg)
Chromium
Copper
Zinc
TSS
Statistics
No
13
13
13
13
Min
.027
.000
.055
35.7
Avg
.178
.37
.384
114
Max CV
.59
1.37
.70
329
Factors Standards
* P
.453
1.0
.70
246
* -* 95% of the monthly averages are expected to be within the
    performance standard, P.
                            A-22

-------
                         Table A-9a

         Historical Effluent Monitoring Data Summary
     with Variability Factors and Performance Standards
                     Daily Measurements
                Subcategory Titanium Dioxide
                       Sulfate Process
                          Plant #1
parameter
(mg/1)
Cadmium
Chromium
Iron
(total)
Iron
(diss)
Lead
Nickel
Zinc
TSS
Historical Summary
Statistics
No
26
26
30
153
26
26
26
183
Min
.001
.010
.40
.080
.002
.010
.010

Avg
.009
.021
3.25
.279
.017
.029
.027
35.8
Max
.020
.070
19.1
4.98
.050
.080
.300

CV
.444
.857
1.42
2.01
.765
.690
2.11
1.71
Variability
Factors

2
4
6
8
3
3
9
7
*
.03
.23
.74
.64
.67
.52
.93
.70
= SES = = = 3S=3=S«=ES =
Performance
Standards
P
0.018
0.088
21.9
2,41
0.062
0.102
0.268
276.
* - 99% Of the daily maximum measurements  expected  to be  less
    than the performance standard,  P.
                           A-23

-------
                          Table A-9a-l

           Historical Effluent Monitoring Data Summary
       with Variability Factors and performance Standards
                       Daily Measurements
                  Subcategory Titanium Dioxide
                         Sulfate Process
                            Plant *I
                  April 76 through September 78
               Historical Summary
Variability   Performance
Parameter
(mg/1)
Cadmium
Chromium
Iron**
Lead
Nickel
Zinc
TSS
Statistics
No
109
128
854
128
128
128
899
Min
,001
,010
.010
.002
.010
.010
.000
Avg
.058
.072
.620
.068
.080
.151
21.2
Max
.100
.400
59.9
.100
.680
1.14
975
CV
.762
.755
5.58
.609
.883
1.35
3.11
Factors

3.
3.
13
3.
4.
6.
11
*
85
81
.5
15
39
41
.0
Standards
P
.224
.275
8.39
.214
.354
.966
233
  * - 99% of the daily maximum measurements expected to be less
      than the performance standard,  P.

** 04-76 to 08-78
                             A-24

-------
                            Table A-9a-2

            Historical Effluent Monitoring  Data  Summary
        with Variability Factors and  Performance Standards
                        Daily Measurements
                    Subcategory Titanium  Dioxide
                          Sulfate Process
                             Plant  #1
                 September  78 through February 79
Parameter
(mg/1)
Cadmium
Chromium
Iron
Lead
Nickel
Zinc
TSS **
Historical Summary
Statistics
No
22
22
22
22
22
22
22
Min
.1
.10
212
.1
.1
.1
35.6
Avg
.1
0.27
390
.1
0.15
0.67
257
Max
.1
0.74
605
.1
0.35
1.8
1135
CV
0
0.74
0.31
0
0.47
0.75
1.12
Variability
Factors
*
1
3.73
1.94
1
2.58
3.81
5.46
Performance
Standards
P
.1
1.01
757
.1
0.39
2.55
1403
 * - 99% of the daily maximum measurements expected to be less
     than the performance standard,  P.

** 09-78 to 01-79
                              A-25

-------
                           Table A-9a-3

            Historical Effluent Monitoring Data Summary
        with Variability Factors and Performance Standards
                        Daily Measurements
                   Subcategory Titanium Dioxide
                          Sulfate Process
                             Plant #1
                 September 79 through April 81
Parameter
(mg/1)
Cadmium
Chromium
Iron
Lead
Nickel
Zinc
TSS**
Historical Summary Variability
Statistics Factors
No
82
81
84
82
82
54
30
Min
0.10
0.05
202
0.10
0.05
0.05
11
Avg
0.10
0.23
330
0.10
0.14
0.50
50
Max CV
0.10 0
0.67 0.87
582 0..22
0.10 0
0.53 0.86
1.18 1.2
398 1.18

1
4
1
1
4
5
5
*
.0
.4
.5
.0
.3
.8
.8
Performance
Standards

0.
1.
495
0.
0.
2.
290
P
10
0

10
60
9

 * - 99% of the daily maximum measurements expected to be less
     than the performance standard, P.

** April 81
                                A-26

-------
                        Table A-9b-l

         Historical Effluent Monitoring Data Summary
     with Variability Factors and Performance Standards
                     Daily Measurements
                Subcategory Titanium Dioxide
                       Sulfate Process
                          Plant #l
                April 76 through September 78
             Historical Summary
Variability   Performance
Parameter
(kg/day)
Cadmium
Chromium
Iron**
Lead
Nickel
Zinc
TSS
Statistics
No
109
128
854
128
128
128
899
Min
,004
.045
1.00
.008
.047
.049
26 1,
Avg
.432
.526
4.29
.503
.576
1.07
350
Max
.908
2.65
3,854
.908
3.99
55.1
58,820
CV
.782
.707
5.78
.634
.790
1.33
2.79
Factors
*
3.95
3.61
13.6
3.27
3.98
6.32
10.5
Standards

1
1

1
2
6
14,
P
.70
.90
585
.65
.29
.76
120
* - 99% of the daily maximum measurements  expected  to be less
     than the performance standard,  P.

** 04-76 to 08-78
                            A-27

-------
                           Table A-9b-2

            Historical Effluent Monitoring Data Summary
        with Variability Factors and Performance Standards
                        Daily Measurements
                   Subcategory Titanium Dioxide
                          Sulfate Process
                             Plant #1
                 September 78 through February 79
Parameter
(kg/day)
Cadmium
Chromium
Iron
Lead
Nickel
Zinc
TSS**
Historical Summary Variability
Statistics Factors
No
22
22
22 14
22
22
22
22 2
Min
6.95
6.95
,730
6.95
6.95
6.95
,474
Avg Max CV
6.95 6.95 0
18.76 51.41 0.74
27,097 42,035 0.31
6.95 6.95 0
10.42 24.32 0.47
46.55 125.06 0.75
17,856 78,860 1.12
*
1.0
3.73
1.94
1.0
2.56
3.81
5.46
Performance
Standards
P
6.95
69.98
52,568
6.95
26.88
177.36
97,493
 * - 99% of the daily maximum measurements expected to be less
     than the performance standard,  P.

** 09-78 to 01-79
                                A-28

-------
                           Table A-9b-3

            Historical Effluent Monitoring Data Summary
        with Variability Factors and Performance Standards
                        Daily Measurements
                   Subcategory Titanium Dioxide
                          Sulfate Process
                             Plant #1
                   September 79 through April 81
Parameter
(kg/day)
Cadmium
Chromium
Iron
Lead
Nickel
Zinc
TSS**
Historical Summary Variability
Statistics Factors
No
82
81
84
82
82
54
30
Min
8.
4.
2
1
16,521
8.
4,
4.
2
1
1
900
Avg
8
18
.2
.8
26,990
8
11
40
.2
.4
.9
4,089
Max CV
8.2 0
54.8 0.87
47,599 0.22
8.2 0
43.3 0.86
96.5 1.2
32,551 1.18
*
1.
4.
1.
1.
4.
5.
5.

0
4
5
0
3
8
8
Performance
Standards

8.
81.
P
2
8
40,484
8.
49.
2
0
237
23,718
 * - 99% of the daily maximum measurements expected to be less
     than the performance standard,  P.

** April 81
                               A-29

-------
                          Table A-9c-l

           Historical Effluent Monitoring  Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                  Subcategory Titanium Dioxide
                         Sulfate Process
                            Plant #1
                  April 76 through September  78
               Historical Summary
Variability   Performance
Parameter
(mg/1)
Cadmium
Chromium
Iron**
Lead
Nickel
Zinc
TSS
Statistics
No
26
30
28
30
30
30
30
Min
.003
.010
.060
.004
.010
.010
1.58
Avg
.058
.072
.620
.068
.080
.151
21.2
Max
.100
.130
3.74
.100
.245
.815
74.8
CV
.722
.524
1.51
.578
.594
1.03
1.03
Factors

2.
2.
4.
2.
4.
3.
3.
*
43
04
00
14
39
05
04
Standards
P
.142
.147
2.47
.147
.354
.406
64.3
  * - 95% of the monthly averages are expected to be within  the
      performance standard,  P.

** 04-76 to 08-78
                             A-30

-------
                           Table A-9c-2

            Historical Effluent Monitoring Data Summary
        with Variability Factors and Performance Standards
                          30 Day Averages
                   Subcategory Titanium Dioxide
                          Sulfate Process
                             Plant #1
                 September 78 through February 79
Parameter
(mg/1)
Cadmium
Chromium
Iron
Lead
Nickel
Zinc
TSS**
Historical Summary Variability
Statistics Factors
No
5
5
5
5
5
5
5
Min
.1
.10
244.4
.1
.1
0.30
51.6
Avg
.1
0.27
390
.1
0.15
0.67
257
Max
.1
0.48
502.9
.1
0.25
0.83
595
CV
0
0.48
0.25
0
0.37
0.64
0.750
*
1
1.79
1.41
1
1.61
2.05
2.23
Performance
Standards
P
1
0.48
550
.1
0.24
1,37
573
 * - 95% of the monthly averages are expected to be within the
     performance standard,  P.

** 09-78 to 01-79
                               A-31

-------
                           Table A-9c-3

            Historical Effluent Monitoring Data Summary
        with Variability Factors and Performance Standards
                          30 Day Averages
                   Subcategory Titanium Dioxide
                          Sulfate Process
                             Plant #1
                   September 79 through April 81
Parameter
(mg/1)
Cadmium
Chromium
Iron
Lead
Nickel
Zinc
TSS**
Historical Summary Variability
Statistics Factors
No
20
20
20
20
20
13
20
Min
0.10
0.05
261
0.10
0.05
0.12
48
Avg
0.10
0.23
330
0.10
0.14
0.50
50
Max CV
0.10 0
0.47 0.52
418 0.13
0.10 0
0.41 0.67
0.78 0.45
214 0.35
*
1.
1.
1.
1.
2.
1.
1.

0
9
2
0
1
7
6
Performance
Standards

0.
0.
396
0.
0.
0.
80
P
10
44

10
29
85

 * - 95% of the monthly averages are expected to be within the
     performance standard,  P.

** 09-79 to 04-81
                               A-32

-------
                          Table A-9d-l

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                  Subcategory Titanium Dioxide
                         Sulfate Process
                            Plant #1
                  April 76 through September 78
               Historical Summary
Variability   Performance
Parameter
(kg/day)
Cadmium
Chromium
Iron**
Lead
Nickel
Zinc
TSS
Statistics
No
26
30
Min
.016
.064
28 4.00
30
30
30
30
.021
,065
.074
116 1,
Avg
.432
.526
42.9
.503
.576
1.07
350 4
Max
.780
.862
294
.852
1.49
5.62
,797
CV
.743
.549
1.59
.602
.569
.996
1.01
Factors

2
2

2
2
2
2
*
.47
.09
4.14
.19
.13
.97
.99
Standards

1.07
1.10
P


177
1.10
1.23
3.18
4,041




  * - 95% of the monthly averages are expected to be within the
      performance standard, P.

** 04-76 to 08-78
                             A-33

-------
                           Table A-9d-2

            Historical Effluent Monitoring Data Summary
        with Variability Factors and Performance Standards
                          30 Day Averages
                   Subcategory Titanium Dioxide
                          Sulfate Process
                             Plant #1
                 September 78 through February 79
Parameter
(kg/day)
Cadmium
Chromium
Iron
Lead
Nickel
Zinc
TSS**
Historical Summary Variability
Statistics Factors
No
5
5
5
5
5
5
5
Min
6.95
6.95
16,981
6.95
6.95
2.4
3,585
Avg
6.95
18.76
27,097
6.95
10.42
46.55
17,856
Max CV
6.95 0
33.4 0.48
34,941 0.25
6.95 0
17.4 0.37
57.7 0.640
41,340 0.750
*
1
1.79
1.41
1
1.61
2.05
2.23
Performance
Standards
P
6.95
33.4
38,214
6.95
16.68
95.19
39,812
 * - 95% of the monthly averages are expected to be within the
     performance standard, P.

** 09-78 to 01-79
                               A-34

-------
                           Table A-9d-3

            Historical Effluent Monitoring Data Summary
        with Variability Factors and Performance Standards
                          30 Day Averages
                   Subcategory Titanium Dioxide
                          Sulfate Process
                             Plant #1
                   September 79 through April 81
Parameter
(kg/day)
Cadmium
Chromium
Iron
Lead
Nickel
Zinc
TSS**
Historical Summary Variability
Statistics Factors
No
20
20
20
20
20
13
20
Min
8.
4.
2
1
21,350
8.
4.
9.
2
1
8
3,926
Avg
8.2
18.8
26,990
8.2
11.4
40.9
4,089
Max CV
8.
38.
34,
8.
33.
63.
17,
2 0
4 0.52
186 0.13
2 0
5 0.67
8 0.45
502 0.35
*
1.
1.
1.
1.
2.
1.
1.

0
9
2
0
1
7
6
Performance
Standards
P
8.
36.
2
0
32,387
8.
23.
69.
2
7
5
6,543
 * - 95% of the monthly averages are expected to be within the
     performance standard,  P.

** 09-79 to 04-81
                                A-35

-------
                          Table A-9e-l

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                  Subcategory Titanium Dioxide
                         Sulfate Process
                            Plant #1
                  April 76 through September 78
               Historical Summary
Variability   Performance
Parameter
(g/kkg)
Cadmium
Chromium
Iron**
Lead
Nickel
Zinc
TSS
Statistics
No
109
128
854
128
128
128
899
Min
.041
.464
10.3
.082
.484
.505
.268
Avg
4.
5.
44
5.
5.
11
13
45
42
.2
18
93
.02
.91
Factors Standards
Max CV
9.
27
39
9.
41
36
.3
,711
36
.1
568
606
*
17
19
6,0
17
23
69
145
P
.5
.6
28
.0
.6
.7
.5 (kg/kkg)
  * - 99% of the daily maximum measurements expected to be less
      than the performance standard,  P.

** 04-76 to 08-78
                             A-36

-------
                           Table A-9e-2

            Historical Effluent Monitoring Data Summary
        with Variability Factors and Performance Standards
                        Daily Measurements
                   Subcategory Titanium Dioxide
                          Sulfate Process
                             Plant #1
                 September 78 through February 79
Parameter
(g/kkg)
Cadmium
Chromium
Iron
Lead
Nickel
Zinc
TSS**
Historical Summary Variability
Statistics Factors
No
22
22
22
22
22
22
22
Min
0.67
0.67
152
0.67
0.67
0.67
25.48
Avg
0.67
1.82
279
0.67
1.01
4.51
183.9
Max CV *
0.67
4.99
433
0.67
2.36
12.13
812
Performance
Standards
P
0.67
6.79
542
0.67
2.6
17.2
1004


(kg/kkg)



(kg/kkg)
 * - 99% of the daily maximum measurements expected to be less
     than the performance standard, P.

** 09-78 to 01-79
                               A-37

-------
                           Table A-9e-3

            Historical Effluent Monitoring Data Summary
        with Variability Factors and Performance Standards
                        Daily Measurements
                   Subcategory Titanium Dioxide
                          Sulfate Process
                             Plant #1
                   September 79 through April 81
Parameter
(g/kkg)
Cadmium
Chromium
Iron
Lead
Nickel
Zinc
TSS**
Historical Summary Variability Performance
Statistics Factors Standards
No
82
81
84
82
82
54
30
Min
0,80
0.40
170
0.80
0.40
0.40
9.3
Avg
0.80
1.82
278
0.80
1.11
3.97
42.1
Max CV
0.80
5.32
490
0.80
4.20
9.36
335
* P
0.80
7.93
417
0.80
4.75
23.0
244



(kg/kkg)



(kg/kkg)
 * - 99% of the daily maximum measurements expected to be less
     than the performance standard, P.

** April 81
                               A-38

-------
                          Table A-9f-l

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                  Subcategory Titanium Dioxide
                         Sulfate Process
                            Plant #1
                  April 76 through September 78
Parameter
(g/kkg)
Cadmium
Chromium
Iron**
Lead
Nickel
Zinc
TSS
Historical Summary Variability Performance
Statistics Factors Standards
No
26
30
28
30
30
30
30
Min Avg
.165
.66
41.
.216
.67
.762
1.2
4.45
5.42
2 442
5.18
5.94
11.0
13.9
Max CV
8.
8.
04
88
3209
8.
15
57
49
78
.4
.9
.4
* P
11.0
11.3
1,824
11.3
12.7
32.8
41.6 (kg/kkg)
  * - 95% of the monthly averages are expected to be within the
      performance standard, P.

** 04-76 to 08-78
                             A-39

-------
                           Table A-9f-2

            Historical Effluent Monitoring Data Summary
        with1Variability Factors and Performance Standards
                          30 Day Averages
                   Subcategory Titanium Dioxide
                          Sulfate Process
                             Plant #1
                 September 78 through February 79
Parameter
(g/kkg)
Cadmium
Chromium
Iron
Lead
Nickel
Zinc
TSS**
Historical Summary Variability Performance
Statistics Factors Standards
No
5
5
5
5
5
5
5
Min
0.
0.
174
0.
0.
0.
36
67
67
.9
67
67
23
.9
Avg
0.67
1.81
279
0.67
1.01
4.5
184
Max CV
0.67
3.23
360
0.67
1.69
5.6
426
* P
0.67
3.2
394
0.67
0.17
0.98
410



(kg/kkg)



(kg/kkg)
 * - 95% of the monthly averages are expected to be within the
     performance standard, P.

** 09-78 to 01-79
                               A-40

-------
                           Table A-9f-3

            Historical Effluent Monitoring Data Summary
        with Variability Factors and Performance Standards
                          30 Day Averages
                   Subcategory Titanium Dioxide
                          Sulfate Process
                             Plant #1
                   September 79 through April 81
                Historical Summary
Variability   Performance
Parameter
(g/kkg)
Cadmium
Chromium
Iron
Lead
Nickel
Zinc
TSS**
Statistics
No
20
20
20
20
20
13
20

0
0

0
0
0
40
Min
.80
.40
220
.80
.40
.95
.44
Avg
0
1

0
1
3
42
.80
.82
278
.80
.11
.97
.11
Factors Standards
Max CV
0.
3.
80
72
352
0.
3.
6.
180
80
25
20
.3
*
0.
3.
334
0.
2.
6.
67.
P
80
49

80
30
74
4



(kg/kkg)



(kg/kkg)
 * - 95% of the monthly averages are expected to be within the
     performance standard,  P.

** 09-79 to 04-81
                               A-41

-------
                           Table A-lOa

           Historical Effluent Monitoring Data Summary
       with Variability Factors and performance Standards
                       Daily Measurements
                  Subcategory Aluminum Fluoride
                            Plant fJ
 Parameter

  (mg/1)
  Historical Summary
      Statistics
Variability
  Factors
Performance
 Standards
 No  Min  Avg  Max  CV
Lead
152 0.11 2.28 12.8 .601
   3.12
  7.11
      99% of the daily maximum measurements expected to be less
      than the performance standard,  P.
                           Table A-lOb

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                  Subcategory Aluminum Fluoride
                            Plant fJ
 Parameter

  (kg/day)
  Historical Summary
      Statistics
Variability
  Factors
Performance
 Standards
 No  Min  Avg  Max  CV
Lead
152 0.09 2.15 15.3 .753
   3.82
  8.20
  * - 99% of the daily maximum measurements expected to be less
      than the performance standard, P.
                             A-42

-------
                           Table A-lOc

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                  Subcategory Aluminum Floride
                            Plant fJ
               Historical Summary      Variability   Performance
 Parameter         Statistics            Factors      Standards

   (mg/1)      No  Min  Avg  Max  CV          *             P


Lead          10 1.51 2.28 3.90 .601      1.55         3.54



  * - 95% of the monthly averages are expected to be within the
      performance standard, P.


                           Table A-lOd

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                  Subcategory Aluminum Floride
                            Plant f J



               Historical Summary      Variability   Performance
 Parameter         Statistics            Factors      Standards

   (kg/day)    No  Min  Avg  Max  CV          *             P


Lead          10 1.51 2.15 3.70 .326      1.54         3.30
  * - 95% of the monthly averages are expected to be within the
      performance standard, P.
                             A-43

-------
                           Table A-lOe

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                  Subcategory Aluminum Fluoride
                            Plant #J
 Parameter
         Historical Summary
             Statistics
Variability   Performance
  Factors      Standards
              No  Min  Avg  Max  CV
Lead
       152  .89 21.2  150
                80.7
  * -
99% of the daily maximum measurements  expected  to be  less
than the performance standard,  P.
                           Table A-lOf

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                  Subcategory Aluminum Floride
                            Plant fJ
 Parameter

   (g/kkg)
         Historical Summary
             Statistics
Variability
  Factors
Performance
 Standards
        No  Min  Avg  Max  CV
Lead
        10 14.9 21.2 36.4
                32.5
  * - 95% of the monthly averages are expected to be within the
      performance standard, P.
                             A-44

-------
                         Table A-lla

         Historical Effluent Monitoring Data Summary
     with Variability Factors and Performance Standards
                       30 Day Averages
                 Chrome Pigments Subcategory
                          Plant #K
             Historical Summary
Variability    Performance
Parameter
(mg/1)
Arsenic
Cadmium
Chromium
(hexavalent)
Chromium
(Total)
Copper
Lead
Mercury
Zinc
Cyanide
(Available)
Cyanide
(total)
TSS
Statistics
No
23
23
23
23
23
23
23
23
23
23
23
Min
.0096
.050
.028
.197
.038
.217
.0004
.012
.0003
.025
0.27
Factors
Avg Max CV
.079
.079
.112
.442
.134
.412
.001
.04
.019
.118
11.2
.235
.164
.592
.799
.296
1.635
.0018
.087
.076
.316
33.3
.668
.339
1.04
.404
.529
.681
.400
.437
1.57
.995
.662
2
1
2
1
1
2
1
1
3
2
2
*
.02
.56
.70
.66
.87
.12
.66
.72
.58
.63
.01
Standards
P
.156
.123
.302
.733
.250
.873
.0016
.074
.068
.310
22.5
* - 95% of the monthly averages are expected  to  be within  the
    performance standard,  P.
                           A-45

-------
                           Table A-12a

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                  Subcategory Hydrogen Cyanide
                        Andrussow Process
                            Plant #L
 Parameter
 Historical Summary
     Statistics
Variability
  Factors
Performance
 Standards
              No  Min  Avg  Max  CV
Ammonia
35  14. 113. 188. .335
   2.02
  229
      99% Of the daily maximum measurements expected to be less
      than the performance standard, P.
                           Table A-12b

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                  Subcategory Hydrogen Cyanide
                        Andrussow Process
                            Plant £L
 Parameter

   (kg/day)
 Historical Summary
     Statistics
Variability
  Factors
Performance
 Standards
No  Min  Avg  Max  CV
Ammonia
35  112 1533 2419 .365
   2.14
  3283
  * - 99% of the daily maximum measurements expected to be less
      than the performance standard, P.
                              A-46

-------
                           Table A-12c

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                  Subcategory Hydrogen Cyanide
                        Andrussow Process
                            Plant #L
 Parameter

   (mg/1)
 Historical Summary
     Statistics
Variability
  Factors
Performance
 Standards
No  Min  Avg  Max  CV
Ammonia
 8  80. 113. 134. .335
   1.32
  150
      95% Of the monthly averages are expected to be within the
      performance standard, P.
                           Table A-12d

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                  Subcategory Hydrogen Cyanide
                              Andrussow Process
                            Plant fL
 Parameter

  (kg/day)
 Historical Summary
     Statistics
Variability
  Factors
Performance
 Standards
No  Min  Avg  Max  CV
Ammonia
 8  908 1533 1941 .212
   1.42
  2177
  * - 95% of the monthly averages are expected to he within the
      performance standard, P.
                              A-47

-------
                           Table A-12e

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                  Subcategory Hydrogen Cyanide
                        Andrussow Process
                            Plant £L
 Parameter

   (kg/kkg)
         Historical Summary
             Statistics
Variability
  Factors
Performance
 Standards
        No  Min  Avg  Max  CV
Ammonia
        35 .606 8.29 1309
                17.8
  * _
99% of the daily maximum measurements  expected  to  be less
than the performance standard,  P.
                           Table A-12f

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                  Subcategory Hydrogen Cyanide
                              Andrussow Process
                            Plant fL
 Parameter

   (kg/kkg)
         Historical Summary
             Statistics
Variability
  Factors
Performance
 Standards
        No  Min  Avg  Max  CV
Ammonia
         8 46.9 8.29 10.5
                11.8
  * - 95% of the monthly averages are expected to be within the
      performance standard, P.
                              A-48

-------
                         Table A-13a

         Historical Effluent Monitoring Data Summary
     with Variability Factors and Performance Standards
                     Daily Measurements
                         Subcategory
                      Andrussow Process
                          Plant £M

             Historical Summary
Variability   Performance
Parameter
(mg/1)
Cyanide
(Free)
Cyanide
(Total)
Anunon i a
COD
TOC
SS
Statistics
No
534
25
26
25
26
22
Min
.01
.039
.193
2.71
.783
5
Avg
.202
.192
3.63
15.9
8.30
35
Max
3.27
.460
10.2
45.2
25.6
267
CV
1.58
.667
.636
.552
.845
1.57
Factors

7
3
3
2
4
8
*
.26
.42
.51
.90
.22
.16
Standards
P
1.46
.65
12.7
46.1
35.0
286
* - 99% of the daily maximum measurements expected to be less
    than the performance standard,  P.
                           A-49

-------
                           Table A-13c

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                  Subcategory Hydrogen Cyanide
                        Andrussow Process
                            Plant fM
 parameter

  (mg/1)
 Historical Summary
     Statistics
Variability
  Factors
Performance
 Standards
No  Min  Avg  Max  CV
Cyanide
(Free)
19 .082 .202 .351 .391-
   1.78
 0.359
  * - 95% of the monthly averaged are expected to be within the
      performance standard,  P.
                              A-50

-------
                           Table A-13e

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                  Hydrogen Cyanide Subcategory
                        Andrussow Process
                            Plant ffM
 Parameter

   (g/kkg)
 Historical Summary
     Statistics
Variability
  Factors
Performance
 Standards
No  Min  Avg  Max  CV
Cyanide
19 .457 1.12 1.95
                1.99
  * - 95% of the monthly averages are expected to be within the
      performance standard, P.
                             A-51

-------
                           Table A-15a

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                   Subcategory Nickel Sulfate
                            Plant #0
 Parameter

  (mg/1)
 Historical Summary
     Statistics
Variability
  Factors
Performance
 Standards
No  Min  Avg  Max  CV
Nickel
88 .080 1.83 8.33 1.21
   5.84
  10.7
    - 99% of the daily maximum measurements expected to be less
      than the performance standard, P.
                           Table A-15b

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance standards
                       Daily Measurements
                   Subcategory Nickel Sulfate
                            Plant $0
 Parameter

  (kg/day)
 Historical Summary
     Statistics
Variability
  Factors
Performance
 Standards
No  Min  Avg  Max  CV
Nickel
88 1.02 8.32 44.6  1.31
    6.24
   51.9
  * - 99% of the daily maximum measurements expected to be less
      than the performance standard, p.
                             A-52

-------
                           Table A-15c

           Historical Effluent Monitoring Data Summary
       with Variability Factors and performance Standards
                         30 Day Averages
                   Subcategory Nickel Sulfate
                            Plant £0
 Parameter

  (mg/1)
 Historical Summary
     Statistics
Variability
  Factors
Performance
 Standards
No  Min  Avg  Max  CV
Nickel
 3 1.29 1.83 2.48 1.21
   1.54
  2.82
  * - 95% of the monthly averages are expected to be within the
      performance standard, P.


                           Table A-15d

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                   Subcategory Nickel Sulfate
                            Plant fO
 Parameter

  (kg/day)
 Historical Summary
     Statistics
Variability
  Factors
Performance
 Standards
No  Min  Avg  Max  CV
Nickel
 3 5.04 8.32 11.1 .302
   1.49
  12.4
  * - 95% of the monthly averages are expected to be within the
      performance standard, P.
                             A-53

-------
                           Table A-15e

           Historical Effluent Monitoring  Data  Summary
       with Variability Factors and Performance Standards
                       Daily Measurements
                   Subcategory Nickel Sulfate
                            Plant fO
 Parameter

   (g/kkg)
 Historical Summary
     Statistics
Variability
  Factors
Performance
 Standards
No  Min  Avg  Max  CV
Nickel
88  112  912 4890
                5,691
      99% of the daily maximum measurements expected to be  less
      than the performance standard,  P.
                           Table A-15f

           Historical Effluent Monitoring  Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                   Subcategory Nickel Sulfate
                            Plant #0
 Parameter

   (9/kkg)
 Historical Summary
     Statistics
Variability
  Factors
Performance
 Standards
No  Min  Avg  Max  CV
Nickel
 3  553  912 1217
               1,360
  * - 95% of the monthly averages are expected to be within the
      performance standard, P.
                             A-54

-------
                           Table A-16a

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                 Subcategory Sodium Hydrosulfite
                            Plant #P
 Parameter

    (kg/day)
Historical Summary
    Statistics
Variability
  Factors
Performance
 Standards
  No  Min  Avg  Max  CV
TSS
  36  .91 3.78 41.1 1.69
      3.77
     14.2
      95% of the monthly averages are expected to be within the
      performance standard, P.
                           Table A-16e

           Historical Effluent Monitoring Data Summary
       with Variability Factors and Performance Standards
                         30 Day Averages
                 Subcategory Sodium Hydrosulfite
                            Plant fp
 Parameter

     (g/kkg)
Historical Summary
    Statistics
Variability
  Factors
Performance
 Standards
  No  Min  Avg  Max  CV
TSS
  36 16.3 67.5  734
                    254
  * - 95% of the monthly averages are expected to be within the
      performance standard, P.
*U.S. GOVERNMENT PRINTING OFFICE:  1982-0-361-085/^62
                              A-55

-------


-------

-------

-------