-------
  BEST DEMONSTRATED AVAILABLE TECHNOLOGY  (BOAT)

BACKGROUND  DOCUMENT FOR  F001-F005 SPENT SOLVENTS
                        VOLUME  2
          U.S. ENVIRONMENTAL PROTECTION AGENCY

                  Office of  Solid Waste
                   401 M Street,  S.W.
                 Washington,  D.C.  20460
    James  R. Berlow, Chief             David Pepson
 Treatment Technology Section         Project Manager
                          CJ.S.  Environmental Protoction Agency
                             ie* 5, Library (5FL-16)
                             9.  l»arboril Str«»t, Hooa
                                  5tL   159604
                    November 7,  1986

-------
                       BDAT BACKGROUND DOCUMENT FOR
                      F001-F005 SPENT SOLVENT WASTES

                             TABLE OF CONTENTS
VOLUME 1                    .                                       Page

Executive Summary 	   xxi

SECTION 1:  Background and General Description

1.1    Legal Background 	   1-1
1.2    EPA's Approach to Developing BDAT 	   1-2

       1.2.1  Waste Treatability Groups 	   1-3
       1.2.2  Determination of "Demonstrated" Treatment
              Technologies  	•  1-3
       1.2.3  Determination of "Available" Treatment
              Technologies  	   1-4

              (1)  Treatment technologies that present greater
                   total risks than land disposal methods 	   1-5

              (2)  Proprietary or Patented Processes 	   1-5

              (3)  Substantial Treatment 	   1-5

       1.2.4  Collection and Analysis of Performance Data 	   1-6

              (1)  Collection of Performance Data 	   1-6
              (2)  Treatment Design and Operation 	   1-7

       1.2.5  Identification of "Best" Demonstrated Available
              Treatment 	   1-8

       1.2.6  Variance from the Treatment Standard	   1-8


SECTION 2:  Industries Affected

2.1    Introduction 	   2-1
2.2    Classification of Waste as F001-F005 Spent Solvents 	   2-1
2.3    Industries Which Use Listed Solvents 	   2-1
2.4    Spent Solvent Waste Generation 	   2-10

       2.4.1  Surface Cleaning 	   2-10
       2.4.2  Equipment Cleaning 	   2-11

-------
                       TABLE OF CONTENTS (Continued)


                                                                   Page
SECTION 2 (continued)

2.5    Geographical Distributions 	   2-12

REFERENCES 	   2-45


SECTION 3:  Waste Characterization

3.1    Introduction 	   3-1
3.2    Waste Characterization Data 	   3-1

       3.2.1  Furniture Manufacturing 	   3-3
       3.2.2  Plastics and Resins Industry 	   3-7
       3.2.3  Fiber Industry 	   3-10
       3.2.4  Pharmaceuticals Manufacturing 	   3-11
       3.2.5  Paint Formulation 	   3-15
       3.2.6  Dyes and Pigments Manufacturing 	   3-16
       3.2.7  Organic Chemicals Manufacturing 	   3-17
       3.2.8  Organic Pesticides Manufacturing 	   3-21
       3.2.9  Printing Industry 	   3-24
       3.2.10 Can Coating Industry 	   3-26
       3.2.11 Membrane Production Industry 	   3-31

REFERENCES 	   3-32


SECTION 4:  Applicable Treatment Technologies 	   4-1

4.1    Introduction 	   4-1
4.2    Carbon Adsorption 	   4-1

       4.2.1  Applicability 	   4-1
       4.2.2  Underlying Principles  of Operation 	   4-2
       4.2.3  Description of Activated Carbon Manufacture and
              Carbon Regeneration 	   4-3

              (1) Activated Carbon Manufacture 	   4-3
              (2) Carbon Regeneration 	   4-5

       4.2.4  Design and Operating Parameters Affecting
              Performance 	   4-6

              (1) Design Parameters  	   4-6
              (2) Operating Parameters 	   4-9

-------
                      TABLE OF CONTENTS  (Continued)
SECTION 4:   (continued)

       4.2.5  Bench-Scale Testing 	   4-10
       4.2.6  Pilot-Scale Testing 	   4-11

       Carbon Adsorption References 	   4-13

4.3    Distillation 	   4-14

       4.3.1  Steam Stripping 	   4-14
              (1)  Applicability 	   4-14
              (2)  Underlying Principles of Operation 	   4-14
              (3)  Description of Steam Stripping 	   4-17
              (4)  Design and Operating Parameters Affecting
                  Performance 	   4-17

       4.3.2  Batch Distillation 	   4-20
              (1)  Applicability 	   4-20
              (2)  Underlying Principles of Operation 	   4-21
              (3)  Description of Batch Distillation	   4-21
              (4)  Design and Operating Parameters Affecting
                  Performance 	   4-21

       4.3.3  Thin Film Evaporation 	   4-23
              (1)  Applicability 	   4-23
              (2)  Underlying Principles of Operation 	   4-23
              (3)  Description of Thin Film Evaporation 	   4-24
              (4)  Design and Operating Parameters Affecting
                  Performance 	   4-24

       4.3.4  Fractionation 	   4-24
              (1)  Applicability 	   4-24
              (2)  Underlying Principles of Operation 	   4-26
              (3)  Description of Fractionation 	   4-26
              (4)  Design and Operating Parameters Affecting
                  Performance 	   4-26

       Distillation References 	   4-29

4.4    Biological  Treatment 	   4-30

       4.4.1  Applicability 	   4-30
       4.4.2  Underlying Principles of Operation 	   4-30
              (1)  Anaerobic Biological Treatment 	   4-31
              (2)  Aerobic Biological Treatment 	   4-31
                                    111

-------
                       TABLE  OF CONTENTS  (Continued)
SECTION 4:  (continued)

       4.4.3  Description of.Biological Treatment 	   4-32
              (1)  Activated Sludge 	   4-32
              (2)  Aerated Lagoons 	   4-34
              (3)  Trickling Filters 	   4-35
              (4)  Rotating Biological Contactors 	   4-37

       4.4.4  Design and Operating Parameters Which Affect
              Performance 	   4-39
              (1)  Equalization 	   4-39
              (2)  Nutrients 	   4-39
              (3)  Aeration/Oxygen Supply 	   4-40
              (4)  Wastewater-Biomass Contact 	   4-41
              (5)  Microorganism Growth Phase 	   4-43
              (6)  Temperature 	   4-43
              (7)  ph 	   4-44
              (8)  Selection of Microorganisms 	   4-45

       Biological  Treatment References 	   4-46

4.5    Incineration 	   4-47

       4.5.1  Applicability 	   4-47
       4.5.2  Underlying Principles of Operation 	   4-47
       4.5.3  Description of Incinerators 	   4-48
              (1)  Liquid Injection 	   4-48
              (2)  Rotary Kiln 	   4-48
              (3)  Fluidized Bed 	   4-51
              (4)  Hearth 	   4-51

       4.5.4  Design and Operating Parameters Affecting
              Performance 	   4-51
              (1)  Design Parameters 	   4-51
              (2)  Operating Parameters 	   4-59

       Incineration References 	   4-61

4.6    Wet Air Oxidation 	   4-62

       4.6.1  Applicability 	   4-62
       4.6.2  Underlying Principles of Operation 	   4-62
       4.6.3  Description of Wet Air Oxidation 	   4-63
              (1)  Conventional Wet Air Oxidation 	   4-63
              (2)  Catalyzed Wet Air Oxidation 	   4-65
                                    IV

-------
                       TABLE OP CONTENTS (Continued)
                                                                   Page
SECTION 4:  (continued)

       4.6.4  Design and Operating Parameters Affecting
              Performance 	   4-65

       Wet Air Oxidation References 	   4-67

4.7    Air Stripping 	   4-68

       4.7.1  Applicability 	   4-68
       4.7.2  Underlying Principles of Operation 	   4-68
       4.7.3  Description of Air Stripping 	   4-68
              (1) Mechanical Surface Aerators 	   4-70
              (2) Diffused Aerators 	   4-70

    Air Stripping References 	   4-71

4.8    Fuel Substitution	   4-72

       4.8.1  Applicability 	   4-72
       4.8.2  Underlying Principles of Operation 	   4-72
       4.8.3  Description of Fuel Substitution 	   4-72
              (1) Industrial Boilers 	   4-73
              (2) Industrial Kilns 	   4-73

       4.8.4  Design and Operating Parameters Affecting
              Performance 	   4-74

       Fuel Substitution References 	   4-76

-------
                       TABLE OF CONTENTS (Continued)
VOLUME 2                                                           Page

SECTION 5:  Treatment Performance 	   5-1

5.1    Introduction 	   5-1
5.2    Summary of Treatment Performance Data 	   5-2
5.3    Data Editing Rules 	   5-5
5.4    Statistical Methods for Establishing BOAT 	   5-7

       5.4.1  Variability Factor Calculation 	   5-7
       5.4.2  Outlier Test 	   5-9
       5.4.3  Analysis of Variance 	   5-9

5.5    Development of BOAT Treatment Standards for Wastewaters
       Containing F001-F005 Spent Solvent Wastes 	   5-12

       5.5.1  Transfer of Treatment Data for Wastewaters
              Containing F001-F005 Spent Solvent Wastes 	   5-14
       5.5.2  Derivation of Average Variability Factors for
              Wastewater Treatment 	   5-17
       5.5.3  Acetone Wastewaters 	   5-20
       5.5.4  n-Butyl Alcohol Wastewaters 	   5-21
       5.5.5  Carbon Disulfide Wastewaters 	   5-22
       5.5.6  Carbon Tetrachloride Wastewaters 	   5-23
       5.5.7  Chlorobenzene Wastewaters 	   5-27
       5.5.8  Cresols (Cresylic Acid) Wastewaters 	   5-34
       5.5.9  Cyclohexanone Wastewaters 	   5-37
       5.5.10 1,2-Dichlorobenzene Wastewaters 	   5-38
       5.5.11 Ethyl Acetate Wastewaters 	   5-44
       5.5.12 Ethylbenzene Wastewaters 	.'   5-45
       5.5.13 Ethyl Ether Wastewaters 	   5-57
       5.5.14 Isobutanol Wastewaters 	   5-58
       5.5.15 Methanol Wastewaters  	   5-59
       5.5.16 Methylene Chloride Wastewaters 	   5-62
       5.5.17 Methyl Ethyl Ketone Wastewaters 	   5-71
       5.5.18 Methyl Isobutyl Ketone Wastewaters 	   5-73
       5.5.19 Nitrobenzene Wastewaters 	   5-77
       5.5.20 Pyridine Wastewaters 	   5-83
       5.5.21 Tetrachloroethylene Wastewaters 	   5-84
       5.5.22 Toluene Wastewaters 	   5-91
       5.5.23 1,1,1-Trichloroethane Wastewaters 	   5-110
       5.5.24 l,l,2-Trichloro-l,2,2-Trifluoroethane Wastewaters.   5-115
       5.5.25 Trichloroethylene Wastewaters 	   5-116
                                    VI

-------
                       TABLE OF CONTENTS (Continued)
SECTION 5:   (Continued)
       5.5.26 Trichlorofluoromethane Wastewaters 	   5-123
       5.5.27 Xylene Wastewaters  	   5-126

5.6    Development of BDAT Treatment Standards for
       F001-F005 Spent Solvent Wastes  (Other Than Wastewater) ..   5-129

       5.6.1  Transfer of Incineration Treatment Data 	   5-130
       5.6.2  Derivation of An Average Variability Factor for
              Incineration 	   5-133
       5.6.3  Acetone (Other Than Wastewater) 	   5-135
       5.6.4  n-Butyl Alcohol (Other Than Wastewater) 	   5-139
       5.6.5  Carbon Disulfide (Other Than Wastewater) 	   5-140
       5.6.6  Carbon Tetrachloride (Other Than Wastewater) 	   5-143
       5.6.7  Chlorobenzene (Other Than Wastewater) 	   5-144
       5.6.8  Cresols (Cresylic Acid)  (Other Than Wastewater)...   5-147
       5.6.9  Cyclohexanone (Other Than Wastewater) 	   5-148
       5.6.10 1,2-Dichlorobenzene (Other Than Wastewater) 	   5-149
       5.6.11 Ethyl Acetate (Other Than Wastewater) 	   5-152
       5.6.12 Ethylbenzene (Other Than Wastewater)  	   5-153
       5.6.13 Ethyl Ether (Other Than Wastewater) 	   5-156
       5.6.14 Isobutanol (Other Than Wastewater)  	   5-157
       5.6.15 Methanol (Other Than Wastewater) 	   5-158
       5.6.16 Methylene Chloride (Other Than Wastewater)  	   5-159
       5.6.17 Methyl Ethyl Ketone (Other Than Wastewater) 	   5-163
       5.6.18 Methyl Isobutyl Ketone (Other Than Wastewater)	   5-166
       5.6.19 Nitrobenzene (Other Than Wastewater)  	   5-170
       5.6.20 Pyridine (Other Than Wastewater) 	   5-173
       5.6.21 Tetrachloroethylene (Other Than Wastewater) 	   5-174
       5.6.22 Toluene (Other Than Wastewater) 	   5-177
       5.6.23 1,1,1-Trichloroethane (Other Than Wastewater)  ....   5-181
       5.6.24 1,1,2-Trichloro-l,2,2-Trifluoroethane (Other Than
              Wastewater)  	   5-184
       5.6.25 Trichloroethylene (Other Than Wastewater)  	   5-185
       5.6.26 Trichlorofluoromethane (Other Than  Wastewater)  ...   5-188
       5.6.27 Xylene (Other Than Wastewater)  	   5-189

REFERENCES 	   5-192
                                   VI1

-------
                      TABLE OF CONTENTS  (Continued)


                                                                   Page

VOLUME 3

APPENDIX I  	      1-1
APPENDIX II 	     II-l
                                   Vlll

-------
                              LIST OF TABLES

Table                                                             Page

        BOAT Treatment Standards 	    xxii

2-1     Constituents of Listed.Hazardous Spent Solvent Wastes ..    2-2

2-2     Industries Using Solvents Listed as F001-F005 	    2-3

2-3     Industries Involved in Surface Cleaning and
        Degreasing 	    2-9

2-4     Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Wood Furniture Manufacturing 	    2-14

2-5     Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Metal Furniture Manufacturing 	    2-15

2-6     Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Plastics and Resins Manufacturing 	    2-16

2-7     Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Fiber Manufacturing 	    2-17

2-8     Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Pharmaceutical Manufacturing 	    2-18

2-9     Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Paint Manufacturing and Application 	   2-19

2-10    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Cyclic Crudes and Intermediates Including Dyes
        Manufacturing 	    2-20

2-11    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Pigments Manufacturing 	    2-21

2-12    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Organic Chemicals Manufacturing 	    2-22
                                    IX

-------
                              LIST OF TABLES
                                (Continued)

Table                                                             Page

2-13    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Agricultural Chemicals Manufacturing 	   2-23

2-14    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Printing Industry 	   2-24

2-15    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Commercial Testing Laboratories 	   2-25

2-16    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Electronic Components Manufacturing 	   2-26

2-17    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Semiconductors and Related Devices Manufacture	   2-27

2-18    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Synthetic Rubber Industry 	   2-28

2-19    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Tire Industry 	   2-29

2-20    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Textiles Industry 	   2-30

2-21    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Leather and Tanning Industry 	   2-31

2-22    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Transportation Vehicles Manufacturing 	   2-32

2-23    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Paper Coating Industry 	   2-33

-------
                              LIST OF TABLES
                                (Continued)

Table                                                             Page

2-24    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Adhesives and Sealants Industry 	   2-34

2-25    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Food Industry - Beer, Edible Fats, and Butter	   2-35

2-26    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Dry Cleaning Industry 	   2-36

2-27    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Wool Weaving and Finishing Industry 	   2-37

2-28    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Petroleum Refining Industry 	   2-38

2-29    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Primary Metals Manufacturing 	   2-39

2-30    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Fabricated Metals Manufacturing 	   2-40

2-31    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Non-Electric Machinery Manufacture 	   2-41

2-32    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Electric Equipment Manufacture 	   2-42

2-33    Census Data (1977) for Number of Facilities in Each
        State and EPA Region
        Instruments and Clocks Manufacture 	   2-43
                                    XI

-------
                               LIST OF TABLES
                                (Continued)
Table
                                                                   Paqe
2-34    Census Data  (1977)  for Number  of Facilities  in Each
        State and EPA Region
        Automotive Repair Shops  	    2-44

3-1     Summary of Industries for Which Spent Solvent Waste
        Characterization Data Are Available  	    3-2

3-2     Waste Characterization Data for Spent Thinner and
        Solvent from Furniture Manufacturing - Plant A 	    3-3

3-3     Waste Characterization Data for Spent Thinner and
        Solvent from Furniture Manufacturing - Plant B 	    3-4

3-4     Waste Characterization Data for Spent Thinner and
        Solvent from Furniture Manufacturing - Plant C 	    3-5

3-5     Waste Characterization Data for Spent Thinner and
        Solvent from Furniture Manufacturing - Plant D 	    3-6

3-6     Waste Characterization Data for Still Bottoms and
        Caustic from Plastics and Resins Manufacturing 	    3-7

3-7     Waste Characterization Data for Epoxy Resin Waste
        from Plastics and Resins Manufacturing 	    3-7

3-8     Waste Characterization Data for Phenolic and Polyester/
        Alkyd Resin Waste from Plastics and Resins
        Manufacturing 	   3-9

3-9     Waste Characterization Data for Solvent Recovery
        Bottoms,  Laboratory Solvents and Chrome Plating
        Solution from Fiber Industry 	   3-10

3-10    Waste Characterization Data for Solvent Recovery
        Bottoms from Pharmaceutical Manufacturing 	   3-11

3-11    Waste Characterization Data for Solvent Recovery
        Bottoms from Pharmaceutical Manufacturing 	   3-13

3-12    Waste Characterization Data for Paint Tank Wash from
        Paint Manufacturing 	   3-15
                                   Xll

-------
                              LIST OF TABLES
                                (Continued)

Table                                                             Page

3-13    Waste Characterization Data for Spent Thinner from
        Paint Manufacturing  	   3-15

3-14    Waste Characterization Data for Dyes and Pigments
        Waste from Dyes and  Pigments Manufacturing 	   3-16

3-15    Waste Characterization Data for Still Bottoms and
        Caustic from Organic Chemicals Manufacturing 	   3-17

3-16    Waste Characterization Data for Isocyanates
        Manufacturing Wastes from Organics Chemicals
        Manufacturing 	   3-19

3-17    Waste Characterization Data for Diphenyl Methane and
        Isocyanate Manufacturing Wastes from Organic
        Chemicals Manufacturing 	   3-19

3-18    Waste Characterization Data for Alkenes Manufacturing
        Wastes from Organic Chemicals Manufacturing 	   3-20

3-19    Waste Characterization Data from Aldehyde Furan
        Manufacturing Waste from Organic Chemicals
        Manufacturing 	   3-20

3-20    Waste Characterization Data from Organic
        Pesticides Manufacturing 	   3-21

3-21    Waste Characterization Data from Organic
        Pesticides Manufacturing 	   3-21

3-22    Waste Characterization Data from Organic
        Pesticides Manufacturing 	   3-22

3-23    Waste Characterization Data from Organic
        Pesticides Manufacturing 	   3-22

3-24    Waste Characterization Data from Organic
        Pesticides Manufacturing 	   3-23

3-25    Waste Characterization Data from Organic
        Pesticides Manufacturing 	   3-23
                                  Xlll

-------
                              LIST OF TABLES
                                (Continued)

Table                                                             Page

3-26    Waste Characterization Data for Solvent Recovery
        Bottoms from Printing Industry 	    3-24

3-27    Waste Characterization Data for Spent Ink Wash from
        Printing Industry 	    3-25

3-28    Waste Characterization Data for Spent Can Coating
        Residue from Can Coating Industry 	    3-26

3-29    Waste Characterization Data for Spent Solvents and
        Organics from Membrane Production Industry 	    3-31

5-1     Quantification Levels for F001-F005 Solvents 	    5-6

5-2     BOAT Treatment Standards (as Concentrations in the
        Treatment Residual Extract) 	    5-13

5-3     Grouping of Spent Solvent Constituents for Transfer
        of BOAT Wastewater Treatment Data 	    5-15

5-4     Variability Factors for All Full-Scale Wastewater
        Treatment Data Sets Used in the Derivation of  the

5-5
5-6
5-7
5-8
5-9
5-10
5-11
5-12
5-13
BOAT Treatment Standards 	
Treatment Performance Data for Carbon Tetrachloride ....
Calculation of BOAT for Carbon Tetrachloride 	
Treatment Performance Data for Chlorobenzene 	
Calculation of BOAT for Chlorobenzene 	
Treatment Performance Data for Cresols (Cresylic
Acid) 	
Treatment Performance Data for 1, 2-Dichlorobenzene 	
Calculation of BDAT for 1, 2-Dichlorobenzene 	
Treatment Performance Data for Ethylbenzene 	
Calculation of BDAT for Ethvlbenzene 	
5-18
5-25
5-26
5-30
5-33
5-36
5-40
5-43
5-47
5-56
                                   XIV

-------
                              LIST OF TABLES
                                (Continued)

Table                                                              Page

5-14    Treatment Performance Data for Methanol  	    5-61

5-15    Treatment Performance Data for Methylene Chloride  	    5-65

5-16    Calculation of BOAT for Methylene  Chloride  	    5-70

5-17    Treatment Performance Data for Methyl Ethyl Ketone  	    5-72

5-18    Treatment Performance Data for Methyl Isobutyl Ketone  ..    5-75

5-19-'   Calculation of BDAT for Methyl Isobutyl Ketone 	    5-76

5-20    Treatment Performance Data for Nitrobenzene 	    5-79

5-21    Calculation of BDAT for Nitrobenzene 	    5-82

5-22    Treatment Performance Data for Tetrachloroethylene  	    5-86

5-23    Calculation of BDAT for Tetrachloroethylene 	    5-90

5-24    Treatment Performance Data for Toluene 	    5-94

5-25    Calculation of BDAT for Toluene 	    5-109

5-26    Treatment Performance Data for 1,1,1-Trichloroethane ...    5-112

5-27    Calculation of BDAT for 1,1,1-Trichloroethane 	    5-114

5-28    Treatment Performance Data for Trichloroethylene 	    5-118

5-29    Calculation of BDAT for Trichloroethylene 	    5-122

5-30    Treatment Performance Data for Trichlorofluoromethane...    5-125

5-31    Treatment Performance Data for Xylene 	    5-128

5-32    Grouping of Spent Solvent Constituents  for Transfer
        of BDAT Treatment Data for All Other F001-F005
        Spent Solvents 	    5-131

5-33    Variability Factors for Incineration Data 	    5-134

5-34    Incineration Data for Acetone 	   5-137
                                    XV

-------
                              LIST OF TABLES
                                (Continued)

Table

5-35    Incineration Data for Carbon Disulfide 	    5-142

5-36    Incineration Data for Chlorobenzene 	    5-146

5-37    Incineration Data for 1,2-Dichlorobenzene 	    5-151

5-38    Incineration Data for Ethylbenzene 	    5-155

5-39    Incineration Data for Methylene Chloride 	    5-161

5-40    Incineration Data for Methyl Ethyl Ketone 	    5-165

5-41    Incineration Data for Methyl Isobutyl Ketone 	    5-168

5-42    Incineration Data for Nitrobenzene 	    5-172

5-43    Incineration Data for Tetrachloroethylene 	    5-176

5-44    Incineration Data for Toluene 	    5-179

5-45    Incineration Data for 1,1,1-Trichloroethane  	    5-183

5-46    Incineration Data for Trichloroethylene 	    5-187

5-47    Incineration Data for Xylene 	    5-191
                                   xvi

-------
                              LIST OF TABLES
                                (Continued)

Table                                                             Page

1-1     Index of Plant Treatment Data 	   1-1

II-l    Analysis of Variance Results for Comparing Biological
        and Combined Biological and Activated Carbon Treatments
        at Plant 246 (Chlorobenzene) 	   II-2

II-2    Summary Statistics for the Transformed Data at
        Plant 246 (Chlorobenzene) 	   II-2

II-3    The Outlier Test Results for the Biological Treatment
        Performance Data at Plant 246 (1,2-Dichlorobenzene) ....   II-3

II-4    Analysis of Variance Results for Comparing Biological
        and Combined Biological and Activated Carbon Treatments
        at Plant 246 (1,2-Dichlorobenzene) 	   II-4

II-5    Summary Statistics for the Transformed Data at
        Plant 246 (1,2-Dichlorobenzene)  	   II-4

II-6    Analysis of Variance for Comparing Steam Stripping
        of Pharmaceuticals Industry Treatment Data and
        Biological Treatment Data at Plant 265 (Methylene
        Chloride) 	   II-5

II-7    Analysis of Variance Results for Comparing Air
        Stripping Treatment and Steam Stripping Pilot-Scale
        Treatments (Methyl Isobutyl Ketone)  	   II-6

II-8    The Outlier Test Results for the Combined Steam
        Stripping and Activated Carbon Treatments at Plant 297
        (Nitrobenzene) 	   II-6

II-9    Analysis of Variance Results for Comparing Steam
        Stripping and Combined Steam Stripping and Activated
        Carbon Treatments at Plant 297 (Nitrobenzene)  	   II-7

11-10   Summary Statistics for the Transformed Data at
        Plant 297 (Nitrobenzene) 	   II-7

11-11   The Outlier Test Results for the Biological Treatment
        Data at Plant  225 (Tetrachloroethylene)  	   II-8
                                   XVI1

-------
                              LIST OF TABLES
                                (Continued)

Table                                                             Page

11-12   The Outlier Test Results for the Biological Treatment
        Data at Plant 234 (Toluene) 	    II-9

11-13   Analysis of Variance Results for Comparing Biological
        Treatment and Combined Biological and Activated
        Carbon Treatments at Plant 246 (Toluene) 	    11-10

11-14   Summary Statistics for the Transformed Data at
        Plant 246 (Toluene) 	    11-10

11-15   Analysis of Variance Results for Comparing Pilot-Scale
        Air Stripping and Pilot-Scale Steam Stripping Data
        (1,1,1-Trichloroethane)  	    11-11

11-16   Summary Statistics for the Transformed Data of the
        Pilot-Scale Air and Steam Stripping Treatments
        (1,1,1-Trichloroethane)  	    11-11

11-17   The Outlier Test Results for the Steam Stripping
        Treatment Data at Plant  284 (Trichloroethylene)  	    11-12
                                  xvi 11

-------
                              LIST OF FIGURES




Figure                                                             Page




2-1     EPA Regions 	    2-13




3-1     Resin Production Process  	    3-8




3-2     Organic Phosphate Ester Production Process 	    3-18




3-3     Litho Pressing of Three Piece Cans 	    3-27




3-4     Production of Two Piece Can Bodies 	    3-28




3-5     Pressing of Can Ends  	    3-29




3-6     Assembly of Three Piece Cans  	    3-30




4-1     Plot of Breakthrough Curve 	    4-4




4-2     Moving Bed Carbon Adsorption  	    4-8




4-3     Isotherms for Carbon Adsorption 	    4-12




4-4     Steam Stripping 	    4-18




4-5     Batch Distillation  	    4-22




4-6     Thin Film Evaporation  	    4-25




4-7     Tray Fractionation Column  	    4-27




4-8     Activated Sludge 	    4-33




4-9     Trickling Filter 	    4-36




4-10    Rotating Biological Contactor 	    4-38




4-11    Liquid Injection Incinerator  	    4-49




4-12    Rotary Kiln Incinerator 	    4-50




4-13    Fluidized Bed Incinerator  	    4-52




4-14    Hearth Incinerator  	    4-53




4-15    Wet Air Oxidation 	    4-64




4-16    Air Stripping 	    4-69






                                    xix

-------

-------
                         5.  TREATMENT PERFORMANCE

5.1      Introduction

    This section explains how all of the treatment  standards for
F001-F005 spent solvents were derived.   A summary of the sources of
treatment performance data used to derive BDAT treatment standards for
spent solvent wastes is presented in Section 5.2.  Data editing
procedures are discussed in Section 5.3.  Statistical analyses, including
calculation of variability factors, outlier determination,  and analysis
of variance are discussed in Section 5.4.  Development of BDAT treatment
standards for wastewaters containing the F001-F005  spent solvent
constituents are presented in Section 5.5.  Treatment standards for all
other spent solvent wastes are presented in Section 5.6.  Complete data
sets characterizing wastes used in the  derivation of the treatment
standards are presented in Appendix I.   This appendix should be consulted
when determining whether to submit a petition for a variance from the
treatment standard.  To obtain a variance, a petitioner would have to
show that their F001-F005 spent solvent  waste is sufficiently different
from the wastes considered in the development of the treatment standard,
such that EPA's consideration of this waste during  the rulemaking would
have resulted in a separate treatability subgroup.   All pertinent
statistical parameters and results used  to determine the treatment
standards are presented in Appendix II.
                                   5-1

-------
5.2      Summary of Treatment Performance Data

    EPA collected data on treatment of wastes containing the F001-F005
spent solvent constituents.   Treatment data were examined by EPA from the
following sources for use in development of BDAT treatment standards for
F001-F005 spent solvents:

    a) Organic Chemicals, Plastics, and Synthetic Fibers (OCPSF)
       Industries Data Base  (Reference 13).  EPA collected treatment
       performance data for  the development of OCPSF effluent limitations
       regulations.  For the F001-F005 spent solvents rule, we used data
       from 28 plants in the OCPSF category.  Wastewater treatment
       technologies for which data were collected as part of this program
       include steam stripping, biological treatment, and systems which
       use these technologies in combination with activated carbon
       adsorption.  These data do not necessarily represent treatment of
       spent solvent wastes, but rather treatment of wastes containing
       the constituents.  The Agency may use treatment data from wastes
       that it believes to be similar and that contain constituents of
       concern even though the actual wastes may not fall within an EPA
       code.

    b) Pharmaceuticals Industry Data Base (Reference 14).  EPA collected
       data for the development of effluent guidelines for the
       Pharmaceuticals industry.  We are using data from one plant which
       operates a steam stripper for treatment of methylene chloride
       wastewater in this data base.  These data were presented in EPA's
       Notice of Availability of Data (Reference 16).

    c) Subseguent to proposal, EPA collected incinerator residue samples
       from the incineration of hazardous wastes, including spent
       solvents, from 10 incinerators at 9 sites (Reference 11).
       Analyses of TCLP extracts of the residue and total analyses of the
       residue were performed for these samples.   Analyses were also
       performed on influent wastes fed to the incinerators at all
       sites.  These data were presented in EPA's Notice of Availability
       of Data (Reference 16).

    d) Data on pilot-scale steam stripping and air stripping of
       solvent-contaminated  groundwater are presented in a paper by
       Stover and Kincannon, 1983 (Reference 2).   These data do not
       necessarily represent treatment of spent solvent wastes, but
       rather treatment of similar wastes containing the constituents.
                                   5-2

-------
e) Data on full-scale powdered activated carbon and biological
   treatment (commercially available PACT® process) of organic
   chemical manufacturing wastewater are presented in a paper by
   Hutton, 1979 (Reference 4).  These data do not necessarily
   represent treatment of spent solvent wastes, but rather treatment
   of similar wastes containing the constituents.

f) Data on pilot-scale air stripping of tap water spiked with
   tetrachloroethylene and trichloroethylene and groundwater
   contaminated by industrial discharge are presented in a paper by
   Love and Eilers, 1982 (Reference 6).  These data do not
   necessarily represent treatment of spent solvent wastes, but
   rather treatment of similar wastes containing the constituents.

g) Data on pilot-scale granular activated carbon adsorption of runoff
   water from a waste disposal site's containment dikes are presented
   in a paper by Becker and Wilson, 1978 (Reference 7).  These data
   do not necessarily represent treatment of spent solvent wastes,
   but rather treatment of similar wastes containing the constituents.

h) Data on full-scale granular activated carbon adsorption of
   pesticide wastewater are presented in a report by IT
   Enviroscience, 1983 (Reference 3).  These data do not necessarily
   represent treatment of spent solvent wastes, but rather treatment
   of similar wastes containing the constituents.

i) Data on full-scale biological treatment of wastewaters from the
   synfuels industry are presented in a report by Torpy, Raphaelian,
   and Luthy, 1981 (Reference 5).  These data do not necessarily
   represent treatment of spent solvent wastes, but rather treatment
   of similar wastes containing the constituents.

j) Data on full-scale granular activated carbon adsorption of cresol
   wastewater are presented in a paper by Baker, Clark, and Jeserig,
   1973 (Reference 12).   These data do not necessarily represent
   treatment of spent solvent wastes, but rather treatment of similar
   wastes containing the constituents.

k) Data on bench-scale wet air oxidation of F001-F005 spent solvent
   wastes and on a synthetic waste containing methylene chloride were
   submitted by Zimpro,  Inc., 1986 (Reference 10).   These data do not
   necessarily represent treatment of spent solvent wastes, but
   rather treatment of similar wastes containing the constituents.

1) Iron and Steel Manufacturing Point Source Category Data Base
   (Reference 9).   EPA collected data for the development of effluent
   guidelines for the iron and steel manufacturing industry.   We
   considered data for xylene and toluene from three combined
   treatment systems in this data base.
                               5-3

-------
    m) Data on pilot-scale granular activated carbon adsorption of
       organic contaminants in a drinking water supply are presented in a
       paper by Ruggiero and Ausubel,  1982 (Reference 8).   These data do
       not necessarily represent treatment of spent solvent wastes, but
       rather treatment of similar wastes containing the constituents.

Performance data used to develop BOAT  treatment standards are presented
by constituent for each of the F001-F005 spent solvent wastes in
Sections 5.5 and 5.6.  Complete data sets displaying all constituent
values and all pollutant parameters analyzed for each influent and
effluent point within each plant are included in Appendix I.  The reader
should consult this appendix for information characterizing the wastes
treated in development of BDAT treatment standards.
                                    5-4

-------
5.3      Data Editing Rules

    The following editing rules were applied to all of the available
data.  Changes from proposal are also discussed here:

    a) All sets of influent and effluent concentrations were considered
       to be paired data unless it was known that the  samples were not
       collected so as to fully account for the retention time in the
       treatment system.  Paired data sets were deleted if the influent
       concentration was less than the corresponding effluent
       concentration.  This is a change from proposal  in response to
       comments.  At proposal, all sets of influent and effluent
       concentrations were considered to be paired data regardless of
       treatment system retention time.

    b) For paired data sets, individual data pairs were deleted if the
       influent concentration was below the quantification level for a
       constituent.  Entire data sets were deleted when the majority of
       the influent concentrations for a constituent were below the
       quantification level for that constituent.  Quantification levels
       for the solvents of concern are shown in Table  5-1.  This is a
       change from proposal.  At proposal, the Agency  used screening
       levels as an editing criteria in order to assess whether the
       effluent concentration level represented treatment or simply
       reflected a low influent concentration.  In response to comments,
       the Agency is no longer using screening levels  to develop land
       disposal restrictions standards.  As a consequence, the Agency
       believes it to be more appropriate to use quantification levels as
       an editing criteria for deleting treatment data sets where
       influent concentrations are low.

    c) Treatment concentration levels reported by the  analytical
       laboratories as at or below the analytical detection limit were
       set equal to the detection limit for averaging  and statistical
       analyses.  Setting the concentration level equal to the detection
       limit reported with a data set is a conservative approach because
       the actual concentration of a constituent reported as "not
       detected" is between zero and the detection limit.   Consequently,
       the mean value computed using the detection limit as an estimate
       of the actual value will be somewhat higher than the true mean of
       the data.  This is the same procedure used at proposal when data
       were reported at or below the detection limit.
                                   5-5

-------
                            Table 5-1

         QUANTIFICATION LEVELS FOR F001  - F005 SOLVENTS
                                            Quantification Level
   Constituent                              	(mg/L)	

Acetone                                              0.05
n-Butyl alcohol                                      5.0
Carbon disulfide                                     0.05
Carbon tetrachloride                                 0.05
Chlorobenzene                                        0.05
Cresols (Cresylic acid)                              0.50
Cyclohexanone                                        0.125
1,2-Dichlorobenzene                                  0.125
Ethyl acetate                                        0.05
Ethyl benzene                                        0.05
Ethyl ether                                          0.05
Isobutanol                                           5.0
Methanol                                             0.25
Methylene chloride                                   0.125
Methyl ethyl ketone                                  0.05
Methyl isobutyl ketone                               0.05
Nitrobenzene                                         0.125
Pyridine                                             0.05
Tetrachloroethylene                                  0.05
Toluene                                              0.05
1,1,1-Trichloroethane                                0.05
1,1,2-Trichloro-l,2,2-trifluoroethane                0.05
Trichloroethylene                                    0.05
Trichlorofluoromethane                               0.05
Xylene                                               0.05
                                 5-6

-------
5.4      Statistical Methods for Establishing BOAT

    To develop BOAT treatment standards, the Agency used the following
three statistical methods that were presented in EPA's Notice of
Availability of Data on the land disposal restrictions (Reference 16):

    1) Variability Factor Calculation - to account for variability in
       performance of well-designed and well-operated systems.

    2) Outlier Test - to determine whether a data point within a data set
       is representative of that data set.

    3) Analysis of Variance - to measure whether differences between data
       sets are statistically discernible.

    More detailed discussions of these methods follow below.

5.4.1    Variability Factor Calculation

    The Agency incorporated a variability factor in the development of
BDAT treatment standards.  To obtain the BDAT treatment standard, the
Agency multiplied the long-term average treatment performance value by
the variability factor.  Variability in performance principally arises
from inherent mechanical limitations in maintaining control parameters at
the optimum setting.

    An example would be an automatic pH control system used to maintain
the proper pH range for precipitation of a toxic metal.  In this system,
a pH sensing device provides a signal to the controller that the pH is
not at the set point (i.e., the optimum design point).  The controller
then changes (either pneumatically or electrically) the position of the
valve that supplies the reagent(s) used to adjust pH.  The Agency would
consider such a system to be well-operated provided that it is properly
designed, calibrated, and maintained.  Nevertheless, this system cannot
be operated without any variation in the level of performance.  Control
valves are not manufactured in such a way that they can precisely add the
exact amount of reagent needed to be at the set point: either too much or
too little reagent will be added.  Also, there is a lag time between the
time that the sensing device detects a problem and the time the
controller adjusts the position of the valve to correct the problem.
Additionally, there can be process upsets that reguire greater changes to
the system with corresponding greater variations in performance.  Another
source of variability is the analysis of treatment samples; even EPA
approved methods will have some variability in test results for the same
samples.  All of the above variations can occur even at well designed and
operated treatment facilities.
                                    5-7

-------
    The Agency used the statistical calculation described below to
account for these changes.  This analysis is the same as has been used
for the development of numerous regulations in the Effluent Guidelines
Program.


                              VF = C99
                                   Mean
where,

    VF  =   Estimate of maximum variability factor determined from a
            sample population of data.

    Cgg =   Estimate of performance values which 99 percent of the
            observations will be below.  Cgg is calculated using the
            following equation:

            Cgg = exp(y + 2.33Sy)

            where y and Sy are the mean and standard deviation,
            respectively, of the logtransformed data.

    Mean =  Arithmetic average of the individual performance values.

    Setting standards based on such a variability factor should not be
viewed as "relaxing" BDAT requirements.  Rather, it accommodates the
normal variability of the processes.  A plant will have to be designed to
meet the mean treatment level in order to be assured of not being out of
compliance when the Agency samples the treatment residues.
                                    5-8

-------
5.4.2    Outlier Test

    An outlier in a data set is an observation (or data point) that is
significantly different from the other data.  The measure of difference
is determined by the statistical method known as a Z-score.  Because the
outlier test assumes data to be normally distributed, it is necessary to
transform the data by computing the logarithm of each data point before
performing the outlier test.  The Z-score is calculated by dividing the
difference between the data point and the average of the data set by the
standard deviation.  For data that is normally distributed, 99.5 percent
(or two standard deviations) of the measurements will have a Z-score
between -2.0 and 2.0.  A data point outside this range is not considered
to be representative of the population from which the data are drawn.

    EPA used this statistical method to confirm that certain data do not
represent treatment by a well-operated system.  The Agency used this
method only in cases where data on the design and operation of a
treatment system were limited.  This method is a commonly used technique
for evaluating data sets.

5.4.3    Analysis of Variance

    EPA used the statistical method known as analysis of variance in
determining the level of performance that represents BOAT.   This method
provides a measure of the differences between data sets.  If the
differences are not statistically discernible, the data sets are said to
be homogeneous.

    This method was used in two cases.  The first case was where more
than one technology was used to treat the same waste.  In this case, the
analysis of variance method was used to determine whether BOAT
represented a level of performance achieved by only one technology or
represented a level of performance achievable by more than one or all of
the technologies.  The second case where the analysis of variance was
used was where different wastes with common constituents were treated
with the same technology.  We used this statistical method to determine
whether separate BDAT values should be established for each waste or
whether the levels of performance were homogeneous and,  therefore,
amenable to a single concentration level for a given constituent.

    To determine whether any or all of the treatment data sets were
homogeneous using the analysis of variance method, it was necessary to
compare a calculated "F value" to what is known as a "critical value."
These critical values are available in most statistics texts.
                                   5-9

-------
    Where the F value is less than the critical value, all treatment data
sets are homogeneous.  If the F value exceeds the critical value, it is
necessary to perform a "pair wise F" test to determine if any of the sets
are homogeneous.  The "pair wise F" test would be done for all of the
various combinations of data sets using the same method and equation as
the general F test.

    The F value is calculated as follows:

       (1) All data points are logtransformed.

       (2) The sum of the logtransformed data points (Ti) is computed for
           each data set.

       (3) The statistical parameter known as the sum of the squares
           between data sets (SSB) is computed:

                                k   1^2    T2
                         SSB =  Y,  	  - _
                               i=l  n;     N
where,
         k  = number of treatment technologies
         i\i = number of data points for technology i
         N  = number of data points for all technologies
         T  - sum of logtransformed data points for all technologies

       (4) The sum of the squares within data sets (SSW) is computed.

                         k   ni           k   T2
               SSW =     L   Z  y?,j -   Z  	
where,
         yi -i  =   the logtransformed observation (j) for treatment
                   technology (i)

       (5) The degrees of freedom corresponding to SSB and SSW are
           calculated.  For SSB, the number of degrees of freedom is
           given by k-1.  For SSW, the number of degrees of freedom is
           given by N-k.
                                   5-10

-------
       (6) Using the above parameters, the F value is calculated as
           follows:
               r -_
                   MSW

where,

               MSB = SSBX(k-l) and
               MSW = SSW/(N-k).

    A computational table summarizing the above parameters
is shown below.

                    COMPUTATIONAL TABLE FOR THE F VALUE
Source      Sum of Squares      Degrees of Freedom   Mean Square
                                                           OOT3
Between            SSB               k-1             MSB = ~
Within             SSW               N-k
                                                                    MSB

                                                                    MSW
                                                           N-k
                                   5-11

-------
5.5      Development of BOAT Treatment Standards for Wastewaters
         Containing F001-F005 Spent Solvent Wastes

    BDAT treatment standards for F001-F005 spent solvent wastes in
wastewater are presented in Table 5-2.  Descriptions of how the treatment
standards were derived are presented in this section.  Treatment
performance data for each constituent are also presented in this
section.  Complete data sets including all constituents and pollutant
parameters analyzed in the wastes treated at each plant are included in
Appendix I.  Where wastewater treatment data were not available to the
Agency, data on which the treatment standards were based were transferred
from other spent solvent wastes for which data were available.  The basis
for transfer of treatment standards for wastewaters is presented in
Section 5.5.1, page 5-14.

    The derivation of BDAT treatment standards includes a variability
analysis as discussed in Section 5.4.  For some data sets, data were
insufficient to develop variability factors; in these cases the Agency
used a variability factor that represented the average of the variability
factors from available data sets.  Calculation of the average variability
factors is discussed in Section 5.5.2.

    In some cases, the treatment standard derived from the data was below
the EPA published analytical quantification level for a specific
constituent because of the lower quantification levels associated with
the treatment residuals actually tested.  In these instances, the BDAT
treatment standard was set at the published quantification level, which
is the lowest level at which EPA can support analytical quantification
over the range of wastes that will be subject to this rule.
                                   5-12

-------
                                Table 5-2
                         BDAT TREATMENT STANDARDS
           (As  Concentrations  in the Treatment Residual Extract)
Constituent

Acetone
n-Butyl alcohol
Carbon disulfide

Carbon tetrachloride
Chlorobenzene
Cresols (cresylic acid)

Cyclohexanone
1,2-Dichlorobenzene
Ethyl acetate

Ethylbenzene
Ethyl ether
Isobutanol

Methanol
Methylene chloride
Methylene chloride generated
  at Pharmaceuticals plants

Methyl ethyl ketone
Methyl isobutyl ketone
Nitrobenzene

Pyridine
Tetrachloroethylene
Toluene

1,1,1-Trichloroethane
l,l,2-Trichloro-l,2,2-
  trifluoroethane
Trichloroethylene

Trichlorofluoromethane
Xylene
 Wastewaters
  Containing
Spent Solvents
    (mq/L)

     0.05
     5.0
     1.05

     0.05
     0.15
     2.82

     0.125
     0.65
     0.05

     0.05
     0.05
     5.0

     0.25
     0.20
    12.7
     0.05
     0.05
     0.66
     1.12
     0.079
     1.12

     1.05
     1.05

     0.062

     0.05
     0.05
Non-Wastewater
 Spent Solent
    Wastes
    (mg/L)

    0.59
    5.0
    4.81

    0.96
    0.05
    0.75

    0.75
    0.125
    0.75

    0.053
    0.75
    5.0

    0.75
    0.96
    0.96
    0.75
    0.33
    0.125

    0.33
    0.05
    0.33

    0.41
    0.96

    0.091

    0.96
    0.15
                                     5-13

-------
5.5.1    Transfer of Treatment Data for Wastewaters Containing F001-F005
         Spent Solvent Wastes

    Where wastewater treatment data on spent solvents were not available,
the Agency developed treatment standards based on the treatment of wastes
(e.g. process wastes) containing the constituents listed in F001-F005.
We believe these wastes to be similar to F001-F005 spent solvent
wastewaters.  We have not identified, nor are we aware of, any
constituents in the F001-F005 spent solvent wastewaters that would cause
these wastes to treat differently than the broad array of wastes for
which we have data.  EPA's data base for wastewaters includes wastes
generated in the manufacture of over 200 products at over 30 different
facilities.

    Where wastewater treatment data on a particular spent solvent waste
or constituent were not available to the Agency, treatment data were
transferred from other constituents for which data were available.  For
this rulemaking, treatment data were transferred based on similarity of
chemical structure with the exception of carbon disulfide which is
structurally dissimilar to the other listed F001-F005 hazardous wastes.

    EPA's data transfer criteria represents a change from proposal.  At
proposal, the Agency relied primarily on the physical parameters of
solubility and Henry's Law Constants.  Solubility was used to predict the
effectiveness of biological treatment; where biological treatment was the
technology basis, the treatment standard was set at the level of
detection.  Henry's Law constants were used to predict the effectiveness
of steam stripping.

    Commenters stated that the Agency should base the transfer of data on
average characteristics of wastes in a relatively large and diverse
grouping.  In consideration of the comments received, the Agency believes
that, for the wide range of wastes covered for this particular
rulemaking, a broader approach to transfer of data is warranted.
Accordingly, in the final rule, we are using chemical structure as the
basis for transfer of data.  The Agency believes that chemical structure
allows the consideration of a broader array of physical and chemical
factors affecting treatment, while at the same time relating the transfer
rationale to an indicator that is commonly used to predict how organic
compounds will react with other compounds and under various conditions.
Included in Table 5-3 are the structural groups upon which the transfer
of treatment standards for wastewaters was based.  One F001-F005 spent
solvent constituent, carbon disulfide, was determined not to be
                                   5-14

-------
en
i—>
tn
          Name  of  Structural  Group
          Halogenated  Aliphatics
          Non-halogenated  Aromatics
          Halogenated  Alkenes
          Halogenated  Aromatics
          Ketones
          Alcohols
                                                                             Table 5-3

                                       GROUPING OF SPENT SOLVENT CONSTITUENTS FOR TRANSFER OF BOAT WASTEWATER TREATMENT  DATA
Functi onal
  Group

R-X
R=R'
                                          R-C-R'
                                            II
                                            0
R-OH
    Consti tuent

Carbon tetrachloride
Methylene chloride
  Pharmaceuticals Wastewater
  All Other Wastewaters
1 ,1,1-Tri chloroethane
1,1,2-Trichloro-1,2,2-trifl uoroethane
Trichlorofluoromethane

Ethyl benzene
To!uene
Xylene
Ni trobenzene
Pyridi ne

Tetrachloroethylene
Trichloroethylene

Chlorobenzene
1,2-Di chlorobenzene

Acetone
Cyclohexanone
Methyl ethyl ketone
Methyl isobutyl ketone

n-Butyl alcohol
Isobutanol
Methanol
          transferred  treatment  data.
          "Treatment  Technologies:   B  =  Biological;  SS  =  Steam  Stripping;  AC  =  Activated  Carbon.
          "-Treatment  standard  shown  is  the  quantification  level  for  the  constituent.
          ''Commercially  available patented  PACT®  process.
Treatment
 Value
 (mg/U

  0.05C
Technology"
12.7
0.20
1.05
1.05a
0.05C
0.05C
1.12
0.05C
0.66
1.12a
0.079
0.062
0.15
0.65
0.05a-c
0.125a-c
0.05a-c
0.05°
5.0a-c
5.0a-c
0.25a'c
ss
B
SS
SS
Bd
B
B&AC
AC
SS&AC
B&AC
B
B&AC
B&AC
B&AC
SS
SS
SS
SS
SS
SS
SS
  Constituent From Which
Data Were Transferred
                                                                                                                                1,1,1-Trichloroethane
                                                                                      Toluene
                              Methyl Isobutyl Ketone
                              Methyl Isobutyl Ketone
                              Methyl Isobutyl Ketone
                              Methyl Isobutyl Ketone

                              Methyl Isobutyl Ketone
                              Methyl Isobutyl Ketone
                              Methyl Isobutyl Ketone

-------
                                                                       Table  5-3   (Continued)

                                       GROUPING OF SPENT SOLVENT CONSTITUENTS FOR TRANSFER OF BOAT WASTEWATER TREATMENT DATA
Name of Structural Group
Ethers
Esters
Functional
Group
R-O-R'
R-C-OR1
II
0
Constituent
Ethyl ether
Ethyl acetate
Phenols

Organic Sulfur Compounds
                                          R = S
Cresols

Carbon  disulfide
en
 i
aTransferred treatment standards.
'•'Treatment Technologies:  B = Biological; SS = Steam Stripping;  AC = Activated Carbon.
"-Treatment standard shown is the Quantification Level  for the constituent.
Treatment
Value
(mg/L) Technology"
0.05a'c SS
0.05a'c SS
2.82 AC
1.05a SS
Constituent From Whi
Data Were Transferred
Methyl Isobutyl Ketone
Methyl Isobutyl Ketone

1 ,1 ,1-Trichloroethane

-------
structurally similar to any other F001-F005 constituents.  For this
reason, transfer of treatment data could not be based on chemical
structure.  However, carbon disulfide does have a large Henry's Law
Constant, 1.2 x 10~^ atm m^/mole (Reference 1), indicating that
carbon disulfide is amenable to steam stripping.  Henry's Law Constant
was therefore used as the basis for transferring treatment data to carbon
disulfide.

    To best account for the range of physical and chemical properties
within a structural group that affect treatment by a specific technology,
the Agency transferred data from the compound with the least stringent
treatment standard for any member of that structural group.  If no
treatment data were available for any member of the particular structural
group, data representing the least stringent treatment standard from the
next most similar structural group were transferred.  For example, no
treatment data were available for any member of the alcohols, esters, and
ethers structural groups.  The ketones were considered to be the next
most similar structural group, based on the oxygen containing,
electron-releasing functional groups present in all four structural
groups.  Therefore, data representing the least stringent treatment
standard for constituents in the ketones group were transferred to the
alcohols, ethers, and esters groups.

5.5.2    Derivation of Average Variability Factors for Wastewater
         Treatment

    The derivation of BDAT treatment standards includes a variability
analysis as discussed in Section 5.4.1.  For some data sets, data were
insufficient to develop variability factors; in these cases the Agency
used a variability factor that represented the average of the variability
factors from available data sets.  Calculation of the average variability
factors is shown in Table 5-4, page 5-18.
                                   5-17

-------
                            Table  5-4

VARIABILITY FACTORS FOR ALL FULL-SCALE WASTEWATER TREATMENT DATA
    SETS USED IN THE DERIVATION OF  THE  BOAT TREATMENT  STANDARDS
                      BIOLOGICAL TREATMENT

          Constituent         Plant     Variability Factor

      1,2-Dichlorobenzene     202              2.11
      Methylene Chloride      265              7.58
      Tetrachloroethylene     225              3.65
      Toluene                 234              1.87
                              257              1.89
                              286              3.25
                              AVERAGE          3.39


        BIOLOGICAL TREATMENT FOLLOWED BY ACTIVATED CARBON

          Constituent         Plant     Variability Factor

      Chlorobenzene           246              4.93
      1 ,2-Dichlorobenzene     246              3.68
      Toluene                 246              9.89
                              AVERAGE          6.17

                         STEAM STRIPPING

          Constituent         Plant     VariabilIty Factor

      Methylene Chloride      12003            3.76
      Toluene                 246              1.21
      Trichloroethylene       284              1.81
                              AVERAGE          2.26

          STEAM STRIPPING FOLLOWED BY ACTIVATED CARBON

          Constituent         Plant     VariabilIty Factor

      Nitrobenzene            297              2.65
      Toluene                 297              1.55
                              AVERAGE          2.10

Average variability factor for all BOAT wastewater treatment = 3.56.
                                5-18

-------
                       Table 5-4  (Continued)

VARIABILITY FACTORS FOR ALL FULL-SCALE WASTEWATER TREATMENT DATA
   SETS USED IN THE DERIVATION  OF THE BOAT TREATMENT STANDARDS
               ACTIVATED CARBON ADSORPTION* TREATMENT

          Constituent         Plant     Variability Factor

      Chlorobenzene           246              4.93
      1,2-Dichlorobenzene     246              3.68
      Toluene                 246              9.89
      Nitrobenzene            297              2.65
      Toluene                 297              1.55
                              AVERAGE          4.54
"Includes data sets for biological  treatment  followed  by acti-
 vated carbon adsorption and steam stripping  followed  by acti-
 vated carbon adsorption.
                                5-19

-------
5.5.3    Acetone Wastewaters

    The Agency has no data for wastewater treatment for the removal of
acetone.  For reasons presented in Section 5.5.1,  EPA used chemical
structure as the basis for transferring treatment  data to acetone spent
solvent wastewaters.   Specifically, we transferred the treatment data
from methyl isobutyl  ketone because, like acetone, methyl isobutyl ketone
contains the ketone functional group.  Methyl isobutyl ketone was the
only constituent for which we had data in the ketones structural group.
Using performance data from methyl isobutyl ketone, the BDAT treatment
standard for acetone is 0.05 mg/L.  The technology basis for this
treatment is steam stripping.

    We believe the BDAT treatment standard for acetone spent solvent
wastewaters represents substantial treatment.  We  would expect untreated
acetone wastes to be similar to untreated methyl isobutyl ketone wastes,
from which we transferred treatment data, since they are used in some of
the same manufacturing processes as shown in Section 2 of this document.
As discussed on page 5-73, in reference to methyl  isobutyl ketone, we
believe these constituent reductions substantially diminish the toxicity
of the spent solvent wastes containing acetone and substantially reduce
the likelihood of migration of acetone from spent  solvent wastes.

    [The proposed technology-based BDAT treatment  standard for acetone
was the same as the standard at promulgation although the derivation of
the treatment standard has changed because of the  change in the approach
to data transfer.  (See Section 5.5.1 for a more detailed discussion of
the methodology for data transfer.)]
                                    5-20

-------
5.5.4    n-Butyl Alcohol Wastewaters

    The Agency has no data for wastewater treatment for the removal of
n-butyl alcohol.  For reasons presented in Section 5.5.1, EPA used
chemical structure as the basis for transferring treatment data to
n-butyl alcohol spent solvent wastewaters.  Specifically, we transferred
the treatment data from methyl isobutyl ketone, which contains the ketone
functional group, to n-butyl alcohol, which contains the hydroxyl
functional group.  The alcohols structural group is most structurally
similar to the ketones group based upon their oxygen containing,
electron-releasing functional groups.  Methyl isobutyl ketone was the
only constituent for which we had data in the ketones structural group.
Using performance data from methyl isobutyl ketone, the transferred BOAT
treatment standard for n-butyl alcohol is 0.05 mg/L.  This transferred
standard is below the quantification level and could not be used as the
treatment standard; therefore, the BOAT treatment standard was set at the
quantification level of 5.0 mg/L.  The technology basis for this
treatment standard is steam stripping.

    We believe the BDAT treatment standard for n-butyl alcohol spent
solvent wastewaters represents substantial treatment.  We would expect
untreated n-butyl alcohol wastes to be similar to untreated methyl
isobutyl ketone wastes, from which we transferred treatment data, since
they are used in some of the same manufacturing processes, as shown in
Section 2 of this document.  As discussed on page 5-73, in reference to
methyl isobutyl ketone, we believe these constituent reductions
substantially diminish the toxicity of the spent solvent wastes
containing n-butyl alcohol and substantially reduce the likelihood of
migration of n-butyl alcohol from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard for n-butyl
alcohol was estimated at the detection limit of <0.100 mg/L based on
biological treatment (see Table 13, 51 FR 1725).  The principal
difference between the proposed and promulgated treatment standards is
EPA's consideration of quantification levels in setting the standard (see
the discussion on the use of quantification levels in Section 5.5 on
page 5-12).  To a lesser extent, the Agency's change in the criteria for
data transfer affected the treatment standard.  (See Section 5.5.1,
page 5-14, for a discussion of the Agency's methodology for data
transfer.)]
                                   5-21

-------
5.5.5    Carbon Bisulfide Wastewaters

    The Agency has no data for wastewater treatment for the removal of
carbon disulfide.  For reasons presented in Section 5.5.1,  in most cases
EPA used chemical structure as the basis for transferring performance
data where data are unavailable.   However, carbon disulfide is
structurally dissimilar to the other listed F001-F005 hazardous wastes.
Therefore, transfer of treatment  data could not be based on chemical
structure.

    Carbon disulfide has a large  Henry's Law Constant, 1.2  x 10~^ atm
m^/mole (Reference 1), indicating that carbon disulfide is  amenable to
steam stripping.  Therefore, the  data used to determine the treatment
standard were transferred from the constituent with the closest Henry's
Law Constant and for which BOAT was based on steam stripping.  The data
on which the treatment standard for 1,1,1-trichloroethane was based, 1.05
mg/L, were transferred to carbon  disulfide.

    We believe the BDAT treatment standard for carbon disulfide spent
solvent wastewaters represents substantial treatment.  We would expect
untreated carbon disulfide wastes to be similar to untreated
1,1,1-trichloroethane wastes from which we transferred treatment data.
As discussed on page 5-110, in reference to 1,1,1-trichloroethane, we
believe these constituent reductions to substantially diminish the
toxicity of the spent solvent wastes containing carbon disulfide and
substantially reduce the likelihood of migration of carbon disulfide from
spent solvent wastes.

    [A technology-based BDAT treatment standard was not developed for
carbon disulfide wastewaters at proposal.  The promulgated treatment
standard was based on data transferred from treatment of 1,1,1-trichloro-
ethane.  (See Section 5.5.1, page 5-14 for a discussion of  the Agency's
methodology for data transfer.)]
                                   5-22

-------
5.5.6    Carbon Tetrachloride Wastewaters

    The Agency has biological treatment data for carbon tetrachloride at
plant 225 in the OCPSF data base.  The Agency also has data from
full-scale biological treatment of wastewater from organic chemicals
manufacturing (commercially available patented PACT® process.
Reference 4).  The data are summarized in Table 5-5 and calculation of
the BOAT treatment standard is shown in Table 5-6.

    The following steps were taken to derive the BDAT treatment standard
for carbon tetrachloride:

    1. We evaluated each data set to determine whether any of the data
       represent poor design or operation of the treatment systems.  The
       available data and information did not show any of the data to
       represent poor design and operation.   Accordingly, none of the
       data were deleted on this basis.

    2. We calculated the arithmetic average  treatment concentration and
       the variability factor for each data  set as shown in Table 5-6.
       Process variability could not be calculated for biological
       treatment by the PACT® process because there is only one data
       pair available from this process.   Therefore, the average
       variability factor for BDAT biological treatment, 3.39, was used
       (calculation of the average variability factor is shown in
       Table 5-4, page 5-18).

       Process variability could not be calculated for biological
       treatment at plant 225 because all effluent values were reported
       as less than or equal to the detection limit of 10 ug/L.  We would
       expect some variability in the data because the actual
       concentrations would range from 0 to  the detection limit of 10
       ug/L.  To estimate the variability, the Agency used the average
       variability factor for BDAT biological treatment, 3.39.
       (Calculation of the average variability factor is shown in
       Table 5-4, page 5-18.)

    3. The analysis of variance method was not used to compare different
       treatments of the same waste because  data are available for only
       one type of treatment for each waste.

    4. EPA then analyzed the data to determine if the various treatment
       concentration levels shown in Table 5-6 could be associated with
       separate waste treatability subgroups.  Sufficient data did not
       exist to identify separate waste treatability subgroups;
       therefore, one waste treatability subgroup was established for all
                                   5-23

-------
       sources of wastewaters containing carbon tetrachloride  spent
       solvents.   The least stringent  treatment level  within the
       treatability subgroup was selected for BDAT  (0.034 mg/L from  plant
       206)  to ensure that the standard could be achieved for  all  waste
       matrices within the waste treatability subgroup.   This  calculated
       concentration level is below the guantification level and could
       not be used as the treatment standard; therefore,  the treatment
       standard was set at the guantification level  of 0.05  mg/L.  The
       technology basis was biological treatment.

    5.  The BDAT treatment standard for carbon tetrachloride  represents
       treatment of a variety of waste matrices generated by process
       streams from the manufacture of at least seven  different
       products.   The untreated waste  concentration  of carbon
       tetrachloride ranged from 0.050 mg/L to 44 mg/L in these waste
       matrices.   All of these wastes  can be treated to the  BDAT
       treatment standard or below (0.050 mg/L). We believe these
       constituent reductions substantially diminish the toxicity  of the
       spent solvent wastes containing carbon tetrachloride  and
       substantially reduce the likelihood of migration of carbon
       tetrachloride from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard for carbon
tetrachloride was <0.010 mg/L based on biological treatment  (see Table
13, 51 FR 1725).   The difference between the proposed  and promulgated
treatment standards is primarily due to EPA's consideration  of
quantification levels in setting the promulgated standard (see the
discussion on the use of guantification levels in Section 5.5, page  5-12)
and the incorporation of a variability factor in derivation  of the
promulgated treatment standard.  (See  Section 5.4 for  a discussion of the
variability factor.)  The changes in data editing also contributed to the
change in the treatment standard (data editing rules are presented in
Section 5.3).]
                                   5-24

-------
                            Table  5-5

       TREATMENT  PERFORMANCE DATA  FOR  CARBON  TETRACHLORIDE
     .Plant 225
Biological Treatment3
Influent   Effluent
 (UQ/U     (uq/U
  1,890
    543
    411
    942
  1,730
  1,054
  1,676
  1,813
    874
    832
    896
    842
  2,306
  1,340
     51
    210
 44,000
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10b
10b
           Plant 225
     Products Manufactured

Polyvinyl chloride
Perch!oroetnylene
Chlorinated paraffins
Chlorine
Hydrogen chloride
Sodium methyl ate
  D.G. Hutton, 1979
Biological Treatment3-0
Influent   Effluent
 (UO/L)     (UQ/L)
     95
                 5.5
                         Description  of
                         Waste Treated

                  Wastewater  from organic  chemi-
                  cals  manufacturing.
aThe data do not represent paired data (i.e.,  the samples  were
 not collected so as to fully account for the retention time in
 the treatment system).
bln the data base from which this data was taken, the sampling
 data was designated using a different code,  plant 227, because
 it represented a different sampling episode.
cCommercially available patented PACT® process.
                                5-25

-------
                            Table 5-6

          CALCULATION OF BOAT FOR CARBON TETRACHLORIDE
Plant
 No.    Technology
   Average
  Treatment                 Treatment Con-
Concentration Variability  centration Level
   (uq/L)	Factor     Avg.  x VF  (ua/Ll
225 Biological
lb Biological
10
5.5
3.39a
3.39a
34
19
aAverage variability factor  of  all  BOAT  biological  treatment
 data (see Table 5-4 and the discussion  on  page  5-17).

bCommercially available patented PACT® process.
                                5-26

-------
5.5.7    Chlorobenzene Wastewaters

    The Agency has biological treatment data for chlorobenzene at plants
202, 206, 246, and 263 in the OCPSF data base.   Also available from the
OCPSF data base are data for biological treatment followed by activated
carbon adsorption at plant 246.  The Agency also has data from full-scale
biological treatment of wastewater from organic chemicals manufacturing
(commercially available patented PACT® process. Reference 4).  The data
are summarized in Table 5-7 and calculation of  the BOAT treatment
standard is shown in Table 5-8.

    The following steps were taken to derive the BOAT treatment standard
for chlorobenzene:

    1. We evaluated each data set to determine  whether any of the data
       represent poor design or operation of the treatment systems.  Data
       for biological treatment at plant 263 (consisting of three data
       points) were deleted on the basis of poor design and performance.
       Based on the disproportionately low removals relative to other
       biological treatment systems for wastes  containing chlorobenzene,
       EPA judged this system to be poorly designed and operated.  This
       system achieved a reduction of only 15.5 percent as compared with
       85 to 99 percent for other biological systems treating wastes
       containing chlorobenzene.

       Data for biological treatment at plant 206 were deleted because
       the treatment system at this plant was shown to be poorly designed
       and/or operated based on the wide variation in influent
       concentrations.  The nature of biological treatment systems
       requires sufficient control of influent  concentrations through the
       use of equalization to prevent "shock loading" of the biomass.

    2. We calculated the arithmetic average treatment concentration and
       the variability factor for each data set as shown in Table 5-8.
       Process variability could not be calculated for biological
       treatment by the PACT® process because there is only one data
       pair available from this process.  Therefore, the average
       variability factor for BOAT biological treatment, 3.39, was used.
       (Calculation of the average variability factor is shown in Table
       5-4, page 5-18.)

       Process variability could not be calculated for biological
       treatment at plant 202 because all effluent values were reported
       as less than or equal to the detection limit of 10 ug/L.  We would
       expect some variability in the data because the actual
       concentrations would range from 0 to the detection limit of 10
                                   5-27

-------
   ug/L.  To estimate the variability,  the Agency used the  average
   variability factor for BDAT biological  treatment,  3.39.
   (Calculation of the average variability factor is  shown  in Table
   5-4, page 5-18.)

3.  Biological treatment and biological  treatment followed by
   activated carbon adsorption of chlorobenzene at plant  246 were
   compared with the analysis of variance  method to determine whether
   the performance of one technology was significantly better than
   the other for treatment of the same  waste.   It was shown that the
   addition of activated carbon adsorption to  biological  treatment
   significantly improved treatment performance.  Therefore, the
   treatment concentration level for plant 246 is 149 ug/L  based upon
   biological treatment followed by activated  carbon  adsorption.
   (Refer to the statistical calculations  and  results in  Table II-l,
   Appendix II.)  The analysis of variance method could not be used
   to compare treatments on any other wastes because  data were not
   available for more than one treatment for other wastes.

4.  EPA then analyzed the data to determine if  the various treatment
   concentration levels shown in Table  5-8 could be associated with
   separate waste treatability subgroups.   Sufficient data  did not
   exist to identify separate waste treatability subgroups;
   therefore, one waste treatability subgroup  was established for all
   sources of wastewaters containing chlorobenzene spent  solvents.
   The least stringent treatment level  within  the treatability
   subgroup was selected for BDAT (0.15 mg/L from plant 246) to
   ensure that the standard could be achieved  for all waste matrices
   within the waste treatability subgroup.  The technology  basis was
   biological treatment followed by activated  carbon  adsorption.

5.  The BDAT treatment standard for chlorobenzene represents treatment
   of a variety of waste matrices generated by process streams from
   the manufacture of 31 or more different products.   The untreated
   waste concentration of chlorobenzene ranged from 0.010 mg/L to 7.2
   mg/L in these waste matrices.  All of these wastes can be treated
   to a level of 0.15 mg/L or below; in all cases we  were able to
   treat to the BDAT treatment standard.   We believe  these
   constituent reductions substantially diminish the  toxicity of the
   spent solvent wastes containing chlorobenzene and  substantially
   reduce the likelihood of migration of chlorobenzene from spent
   solvent wastes.
                               5-28

-------
    [The proposed technology-based BDAT treatment standard for
chlorobenzene was 0.062 mg/L based on biological treatment followed by
activated carbon adsorption (see Table 13, 51 FR 1725).   The difference
between the proposed and promulgated treatment standards is primarily due
to the incorporation of a variability factor in derivation of the
promulgated treatment standard.   (See Section 5.4 for a  discussion of the
variability factor.)  Other less significant factors affecting the change
in the treatment standard are the changes in data editing (data editing
rules are presented in Section 5.3), and deletion of some of the data
points used at proposal because  they represented poor operation of the
treatment systems at the time of sampling.]
                                   5-29

-------
                            Table  5-7

           TREATMENT PERFORMANCE DATA FOR CHLOROBENZENE
     Plant 202
Biological Treatment3
Influent
(ua/L)
135
160
140
99
79
284
404
429
361
401
163
152
161
188
304
225
302
214
159
116
Plant
Bioloaical
Influent
(ua/L)
9,206
16,646
49,775
1,414
14,731
3,159
6,756
929
Effluent
(ug/U
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
206
Treatment3
Effluent
(ug/U
1 ,083
710
460
2,781
142
603
153
794
           Plant 202
     Products Manufactured

Disperse dye coupler
Disperse dyes
Naphthalene sulfonic acid
Organic pigments
p-Phenylene diamine
Sulfur dyes
Vat dyes
Xylenesulfonic acid, sodium
  salt
2-Bromo-4,6-dinitroaniline
2,4-Dimtroanil me
2,4-Dinitrochlorobenzene
2 ,4-Dinitrophenol
2,4,6-Trinitrophenol
4-Ch1oro-2,6-dinitrobenzene
  sulfonic acid, potassium salt
                                             Plant 206
                                       Products Manufactured

                                  3,3-Dichlorobenzidine
                                  Polyurethane resins
                                  Orthochloroaniline
                                  Benzophenone
                                  2-Sulfophthalic  acid
                                  2,6-Dichloronitroaniline
aThe data do not represent paired data (i.e.,  the samples were
 not collected so as to fully account for the  retention time in
 the treatment system).
                                 5-30

-------
                       Table 5-7 (Continued)

           TREATMENT  PERFORMANCE DATA  FOR CHLOROBENZENE
     Plant 246
Biological Treatment3
Influent   Effluent
  (ug/L)      (ua/Ll





1


1



3
343
19
729
856
10
,564
10
10
,258
355
287
409
,040
115
36
44
151
19
111
229
233
298
10
38
14
17
         Plant 246
    Biological Treatment3
         Followed by
 Activated Carbon Adsorption
     Influent   Effluent
      (UQ/L)     (uq/U





1



1


3
7
343
19
729
856
10
,564
10
10
836
,258
287
409
.040
,200
21
10
10
10
19
33
30
56
68
10
10
10
10
80b
       6.500
       6.075
                     70C
           Plant 246
     Products Manufactured

Aniline
Dinitrotoluene (mixed)
Methylene diphenyl diisocyanate
Nitrobenzene
Polymeric methylene diphenyl
  diisocyanate
Polyoxypropylene glycol
Toluene diaimne (mixture)
Toluene diisocyanates (mixture)
Polymeric methylene dianiline
Polyurethane resins
           Plant 246
     Products Manufactured

Same as Plant 246 - Biological
Treatment
aThe data do not represent paired data (i.e..  the samples were
 not collected so as to fully account for the  retention time in
 the treatment system).
bln the data base from which this data was taken, the sampling
 data was designated using a different code,  plant 219, because
 it represented a different sampling episode.
                                 5-31

-------
                       Table  5-7  (Continued)

           TREATMENT PERFORMANCE DATA FOR CHLOROBENZENE
       Plant 263                             Plant 263
  Biological Treatment3                Products Manufactured
Influent   Effluent
 (ug/L)     (ug/L)                Methylene diphenyl diisocyanate
                                  Polymeric methylene diphenyl
    515        788                  diisocyanate
    832        404                Polyurethane resins
    443        320                Polyurethane component
                                  Polyurethane prepolymer
                                  Propoxylates, alkylamines
    O.G. Mutton, 1979                    Description of
  Biological Treatment3•b                Waste Treated
   Influent   Effluent
    (ug/L)     (ug/L)             Wastewater from organic chemi-
                                  cals manufacturing.
     1,900         12
aThe data do not represent paired data (i.e.,  the samples were
 not collected so as to fully account for the  retention time in
 the treatment system).
''Commercially available patented PACT® process.
                                 5-32

-------
                            Table 5-8

              CALCULATION OF BOAT FOR CHLOROBEN2ENE
                         Average
                        Treatment                 Treatment  Con-
                                                 Avg.  x  VF  (UQ/L)

                                                       34

                                                      906

                                                      149
Plant
No.
202
246
246



Concentration Variability
Technology (ua/L) Factor
Biological 10 3.39a
Biological 101 8.96
Biological fol- 30 4.93
lowed by
Activated
Carbon
  lb    Biological          12         3.39a            41
aAverage variability factor of all  BOAT biological  treatment
 data (see Table 5-4 and the discussion on  page  5-17).
^Commercially available patented PACT® process.
                                5-33

-------
5.5.8    Cresols (Cresylic Acid) Wastewaters

    The Agency has full-scale granular activated carbon adsorption data
for cresols (cresylic acid) (References 3 and 12).   The Agency also has
biological treatment data (Reference 5).  The data  are summarized in
Table 5-9.

    The following steps were taken to derive the BDAT treatment standard
for cresols (cresylic acid):

    1.  We evaluated each data set to determine whether any of the data
       represent poor design or operation of the treatment systems.   The
       available data and information did not show  any of the data to
       represent poor design and operation.   Accordingly, none of the
       data were deleted on this basis.

       The biological treatment data set shown in Table 5-9 was deleted
       because the data were considered unreliable  for use in developing
       treatment standards.  The confidence  of identification of the
       compounds present in the samples is questionable since the
       identifications were reported as "tentative."  One activated
       carbon data set (Baker,  et.  al.) was  deleted because the influent
       and effluent concentrations  were not  reported individually, but in
       ranges.  The average effluent concentration  and treatment
       concentration level could not be determined  from the data.

    2.  We calculated the arithmetic average  treatment concentration level
       and the variability factor for the data set.   Process variability
       could not be calculated  for  activated carbon adsorption because
       there is only one data pair  available from this process.
       Therefore, the average variability factor for BDAT activated
       carbon adsorption, 4.54, was used (calculation of the average
       variability factor is shown  in Table  5-4, page 5-19).

    3.  The analysis of variance method was not used to compare different
       treatments of the same waste because  data are available for only
       one type of treatment for each waste.

    4.  Sufficient data did not  exist to identify separate waste
       treatability subgroups;  therefore, one waste  treatability subgroup
       was established for all  sources of wastewaters containing cresols
       (cresylic acid) spent solvents.  The  treatment level within the
       treatability subgroup was selected for BDAT  (2.82 mg/L from Torpy,
       Raphaelian,  and Luthy, 1981) by multiplying  the process effluent
       concentration, 0.620 mg/L, by the average variability factor  for
       BDAT activated carbon adsorption, 4.54.   The  technology basis was
       activated carbon treatment.
                                   5-34

-------
    5. The BOAT treatment standard for cresols (cresylic acid) represents
       treatment of a waste matrix generated by a process stream from the
       manufacture of pesticides.   The untreated waste concentration of
       cresols (cresylic acid) was as high as 16.5 mg/L in this waste
       matrix.  This waste was treated to a concentration below the BDAT
       treatment standard (2.82 mg/L).  We believe these constituent
       reductions substantially diminish the toxicity of the spent
       solvent wastes containing cresols (cresylic acid) and
       substantially reduce the likelihood of migration of cresols
       (cresylic acid) from spent  solvent wastes.

    [The proposed technology-based BDAT treatment standard for cresols
(cresylic acid) was estimated at the detection limit of <0.100 mg/L based
on biological treatment (see Table 13, 51 FR 1725).   The principal
difference between the proposed and promulgated treatment standards is
EPA's use of performance data for  activated carbon adsorption treatment
of cresols and the incorporation of a variability factor in derivation of
the promulgated treatment standard.  (See Section 5.4 for a discussion of
the variability factor.)]
                                   5-35

-------
                            Table  5-9

     TREATMENT PERFORMANCE DATA FOR CRESOLS (CRESYLIC ACID)
IT Enviroscience, 1983
 Full-Scale Granular                     Description of
   Activated Carbon                      Waste Treated
 Influent   Effluent
  (ug/L)     (ug/Ll               Pesticide wastewater

 16,500        620a
 Torpy, Raphaelian &
    Luthy,  1981                          Description of
 Biological  Treatment                   Waste Treated
 Influent   Effluent
  (ug/L)     (ug/L)               Wastewater from the synfuels
                                  industry
  1.886         15.3C
  2,536         36.8C
   Baker et. al.,  1973
   Full-Scale Granular                   Description of
     Activated Carbon                    Waste Treated
Influent         Effluent
 (ug/Ll           (ug/L)           Cresol wastewater

3,500,000-    0-7,000,000d
6,500,000
aReference did not specifically identify constituent as
 o-, m-, or p-cresol.
bThe data do not represent paired data (i.e., the samples were
 not collected so as  to fully account for the retention time in
 the treatment system).
Co-Cresol.
dp-Cresol.
                                 5-36

-------
5.5.9    Cyclohexanone Wastewaters

    The Agency has no data for wastewater treatment for the removal of
cyclohexanone.  For reasons presented in Section 5.5.1, EPA used chemical
structure as the basis for transferring treatment data to cyclohexanone
spent solvent wastewaters.  Specifically, we transferred the treatment
data from methyl isobutyl ketone because, like cyclohexanone, methyl
isobutyl ketone contains the ketone functional group.   Methyl isobutyl
ketone was the only constituent for which we had data  in the ketones
structural group.  Using performance data from methyl  isobutyl ketone,
the transferred standard for cyclohexanone is 0.05 mg/L.  The standard
derived from the transferred data is below the quantification level and
could not be used as the treatment standard.  Therefore, the BOAT
treatment standard was set at the quantification level of 0.125 mg/L.
The technology basis for this treatment standard is steam stripping.

    We believe the BDAT treatment standard for cyclohexanone spent
solvent wastewaters represents substantial treatment.   We would expect
untreated cyclohexanone wastes to be similar to untreated methyl isobutyl
ketone wastes, from which we transferred treatment data, since they are
used in some of the same manufacturing processes, as shown in Section 2
of this document.  As discussed on page 5-73, in reference to methyl
isobutyl ketone, we believe these constituent reductions substantially
diminish the toxicity of the spent solvent wastes containing
cyclohexanone and substantially reduce the likelihood  of migration of
cyclohexanone from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard for
cyclohexanone was estimated at the detection limit of  <0.100 mg/L based
on biological treatment (see Table 13, 51 FR 1725). The principal
difference between the proposed and promulgated treatment standards is
EPA's consideration of quantification levels in setting the standard (see
the discussion on the use of quantification levels in  Section 5.5 on
page 5-12).  To a lesser extent, the Agency's change in the criteria for
data transfer affected the treatment standard.   (See Section 5.5.1,
page 5-14, for a discussion of the Agency's methodology for data
transfer.)]
                                   5-37

-------
5.5.10   1,2-Dichlorobenzene Wastewaters

    The Agency has biological treatment data for 1,2-dichlorobenzene at
plants 202, 206, and 246 in the OCPSF data base.  The Agency also has
data for biological treatment followed by activated carbon adsorption at
plant 246 in the OCPSF data base.   The data are summarized in Table 5-10
and calculation of the BDAT treatment standard is shown in Table 5-11.

    The following steps were taken to derive the BDAT treatment standard
for 1,2-dichlorobenzene:

    1. We evaluated each data set  to determine whether any of the data
       represent poor design or operation of the treatment systems.  In
       EPA's judgment, one data point in the data set for biological
       treatment at plant 246 represented poor design and operation.  We
       confirmed this judgment using the the outlier test (refer to Table
       II-3, Appendix II).  The outlying data point was deleted.

    2. We calculated the arithmetic average treatment concentration and
       the variability factor for  each data set as shown in Table 5-11.

    3. Biological treatment and biological treatment followed by
       activated carbon adsorption of 1,2-dichlorobenzene at plant 246
       were compared with the analysis of variance method to determine
       whether the performance of  one technology was significantly better
       than the other for treatment of the same waste.  It was shown that
       the addition of activated carbon adsorption to biological
       treatment significantly improved treatment performance.
       Therefore, the treatment concentration level for plant 246 is 0.65
       mg/L based upon biological  treatment followed by activated carbon
       adsorption.  (Refer to the  statistical calculations and results in
       Appendix II.)  The analysis of variance method could not be used
       to compare treatments on any other wastes because data were not
       available for more than one treatment for other wastes.

    4. EPA then analyzed the data  to determine if the various treatment
       concentration levels shown  in Table 5-11 could be associated with
       separate waste treatability subgroups.  Sufficient, data did not
       exist to identify separate  waste treatability subgroups;
       therefore, one waste treatability subgroup was established for all
       sources of wastewaters containing 1,2-dichlorobenzene spent
       solvents.  The least stringent treatment level within the
       treatability subgroup was selected for BDAT (0.65 mg/L from plant
       246) to ensure that the standard could be achieved for all waste
       matrices within the waste treatability subgroup.  The technology
       basis was biological treatment followed by activated carbon
       adsorption.
                                   5-38

-------
    5.  The BDAT treatment standard for 1,2-dichlorobenzene represents
       treatment of a variety of waste matrices generated by process
       streams from the manufacture of 30 different products.   The
       untreated waste concentration of 1,2-dichlorobenzene ranged from
       0.233 mg/L to 4.4 mg/L in these waste matrices.   All of these
       wastes were treated to the BDAT treatment standard or below (0.65
       mg/L).  We believe these constituent reductions  substantially
       diminish the toxicity of the spent solvent wastes containing
       1,2-dichlorobenzene and substantially reduce the likelihood of
       migration of 1,2-dichlorobenzene from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard for
1,2-dichlorobenzene was 0.053 mg/L based on biological  treatment followed
by activated carbon adsorption (see Table 13, 51 FR 1725).  The
difference between the proposed and promulgated treatment standards is
primarily due to the incorporation of a variability factor in derivation
of the promulgated treatment standard.  Other less significant factors
affecting the change in the treatment standard are the  changes in data
editing (data editing rules are presented in Section 5.3) and deletion of
some data points used at proposal because they represented poor operation
of the treatment systems (see the discussion of the outlier analysis in
Section 5.4).]
                                   5-39

-------
                            Table 5-10

        TREATMENT PERFORMANCE DATA FOR 1,2-DICHLOROBENZENE
     PLant 202
Biological Treatment3
Influent   Effluent
 (ug/L)     (ug/U
  1,350
  1,554
  4.387
  2,444
 21
 20
 15
 10
           Plant 202
     Products Manufactured

Disperse dye coupler
Disperse dyes
Naphthalene sulfonic acid
Organic pigments
p-Phenylene diamine
Sulfur dyes
Vat dyes
Xylenesulfonic acid, sodium
  salt
2-Bromo-4,6-dinitroaniline
2,4-Dinitroanilme
2,4-Din itrochlorobenzene
2,4-Dinitrophenol
2,4,6-Trmitrophenol
4-Chloro-2,6-dinitrobenzene
  sulfonic acid,  potassium salt
     Plant 206
Biological Treatment3
Influent   Effluent
 (uq/U      (UQ/L)
    806
    437
    396
    381
    233
  2,333
    649
  1.247
    555
    847
125
175
121
 89
 77
 55
 63
 61
 72
 44
           Plant 206
     Products Manufactured

3,3-Dichlorobenzidine
Polyurethane resins
Orthochloroanilme
Benzophenone
2-Sulfophthalic  acid
2,6-Dichloronitroanilme
aThe data do not represent paired data (i.e.,  the samples were
 not collected so as to fully account for the  retention time in
 the treatment system).
                                 5-40

-------
                       Table 5-10   (Continued)

          TREATMENT PERFORMANCE DATA FOR 1,2-DICHLOROBENZENE
     Plant 246                               Plant 246
Biological Treatment3                  Products Manufactured
Influent   Effluent
 (ug/L)     (ug/L)                Aniline
                                  Dinitrotoluene (mixed)
  2,081      l,153b               Methylene diphenyl  diisocyanate
    820        681                Nitrobenzene
    914        830                Polymeric methylene diphenyl
  1,558        612                  diisocyanate
  2,801        516                Polyoxypropylene glycol
  1,620        529                Toluene diamine (mixture)
  1,198        626                Toluene diisocyanates (mixture)
  1,182        603                Polymeric methylene dianiline
  1,338        506                Polyurethane resins
  1,157        449
  1,412        470
    768        394
  1,894        512
  1,243        468

aThe data do not represent paired data (i.e.,  the samples  were  not
 collected so as to fully account for the retention time  in  the
 treatment system).
bln EPA's judgment, this data point represented poor  design  and
 operation.   We confirmed this judgment using  the outlier  test  (refer
 to Table II-3, Appendix II) and this data point was  deleted.
                                 5-41

-------
                      Table 5-10 (Continued)

        TREATMENT PERFORMANCE DATA FOR 1,2-DICHLOROBENZENE
        . Plant 246
    Biological Treatment3
         Followed by
 Activated Carbon Adsorption
    Influent   Effluent
     (ug/L)     (UQ/L)
      2,081
        820
        914
      1,558
      2,801
      1,620
      1,198
      1,182
      1,338
      1,157
      1.412
        768
      1,894
      1,243
      3,000
      2,187
      3,275
368
481
126
225
157
158
177
186
191
178
158
136
150
149
 50b
 72b
 35"
           Plant 246
     Products Manufactured
Aniline
Dinitrotoluene (mixed)
Methylene diphenyl diisocyanate
Nitrobenzene
Polymeric methylene diphenyl
  diisocyanate
Polyoxypropylene glycol
Toluene diamine (mixture)
Toluene diisocyanates (mixture)
Polymeric methylene dianiline
Polyurethane resins
aThe data do not represent paired data (i.e.,  the samples  were
 not collected so as to fully account for the  retention time in
 the treatment system).
"In the data base from which this data was taken, the sampling
 data was designated using a different code,  plant 219, because
 it represented a different sampling episode.
                                 5-42

-------
                            Table 5-11

           CALCULATION OF BOAT FOR 1,2-DICHLOROBENZENE


                         Average
                        Treatment                 Treatment  Con-
Plant                 Concentration Variability  centration  Level
 No.       Technology     (ua/Ll	Factor     Avg. x  VF  (ug/Ll

 202       Biological       16.5       2.11              35

 206       Biological       88.2       2.48            219

 246       Biological      554         1.56            862

 246       Biological      176         3.68            648
           followed by
           Activated
           Carbon
                                5-43

-------
5.5.11   Ethyl Acetate Wastewaters

    The Agency has no data for wastewater treatment for the removal of
ethyl acetate.  For reasons presented in Section 5.5.1, EPA used chemical
structure as the basis for transferring treatment data to ethyl acetate
spent solvent wastewaters.  Specifically, we transferred the treatment
data from methyl isobutyl ketone, which contains the ketone functional
group, to ethyl acetate, which contains the ester functional group.  The
esters structural group is most structurally similar to the ketones group
based upon their oxygen containing, electron-releasing functional
groups.  Methyl isobutyl ketone was the only constituent for which we had
data in the ketones structural group.  Using performance data from methyl
isobutyl ketone, the BDAT treatment standard for ethyl acetate is 0.05
mg/L.  The technology basis for this treatment standard is steam
stripping.

    We believe the BDAT treatment standard for ethyl acetate spent
solvent wastewaters represents substantial treatment.  We would expect
untreated ethyl acetate wastes to be similar to untreated methyl isobutyl
ketone wastes, from which we transferred treatment data, since they are
used in some of the same manufacturing processes, as shown in Section 2
of this document.  As discussed on page 5-73, in reference to methyl
isobutyl ketone, we believe these constituent reductions substantially
diminish the toxicity of the spent solvent wastes containing ethyl
acetate and substantially reduce the likelihood of migration of ethyl
acetate from spent solvent wastes.

     [The proposed technology-based BDAT treatment standard for ethyl
acetate was estimated at the detection limit of <0.100 mg/L based on
biological treatment (see Table 13, 51 FR 1725).  The difference between
the proposed and promulgated treatment standards is primarily due to the
Agency's change in the criteria for data transfer.  (See Section 5.5.1,
page 5-14, for a discussion of the Agency's methodology for data
transfer.>]
                                   5-44

-------
5.5.12   Ethylbenzene Wastewaters

    The Agency has biological treatment data for ethylbenzene at plants
202, 211, 215, 221, 230, 234, 238, 242, 244, 251,  253,  257,  293, and 299
in the OCPSF data base.   The Agency also has data from pilot-scale steam
stripping and pilot-scale air stripping of solvent contaminated
groundwater (Reference 2).  The data are summarized in Table 5-12 and
calculation of the BDAT treatment standard is shown in Table 5-13.

    The following steps were taken to derive the BDAT treatment standard
for ethylbenzene:

    1. We evaluated each data set to determine whether any of the data
       represent poor design or operation of the treatment systems.   The
       available data and information did not show any of the data to
       represent poor design and operation.   Accordingly, none of the
       data were deleted on this basis.

       In consideration of the amount of full-scale data available for
       ethylbenzene, we believe it is appropriate  to exclude data for
       pilot-scale air stripping and pilot-scale steam stripping.

    2. We calculated the arithmetic average  treatment concentration and
       the variability factor for each data  set as shown in Table 5-13.
       Process variability could not be calculated for biological
       treatment at plants 202, 211, 215, 221, 230, 234, 238, 242, 244,
       251, 253, 293, and 299 because all effluent values were reported
       as less than or equal to the detection limit of 10 ug/L.   We would
       expect some variability in the data because the actual
       concentrations would range from 0 to  the detection limit of 10
       ug/L.  To estimate the variability, the Agency used the average
       variability factor for BDAT biological treatment, 3.39.
       (Calculation of the average variability factor is shown in Table
       5-4, page 5-18.)

    3. The analysis of variance method was not used to compare different
       treatments of the same waste because  data are available for only
       one type of treatment for each waste.

    4. EPA then analyzed the data to determine if  the various treatment
       concentration levels shown in Table 5-13 could be associated with
       separate waste treatability subgroups.  Sufficient data did not
       exist to identify separate waste treatability subgroups;
       therefore, one waste treatability subgroup  was established for all
       sources of wastewaters containing ethylbenzene spent  solvents.
       The least stringent level within the  treatability subgroup was
                                   5-45

-------
       selected for BOAT (0.034 mg/L  from plants  202,  211,  215,  221,  230,
       234,  238,  242,  244,  251, 253,  293,  and 299)  to  ensure  that  the
       standard could  be achieved for all waste matrices  within  the waste
       treatability subgroup.   This calculated concentration  level is
       below the  quantification level for ethylbenzene and  could not  be
       used as the treatment standard;  therefore,  the  treatment  standard
       was set at the  quantification  level of 0.05  mg/L.  The technology
       basis was  biological treatment.

    5.  The BDAT treatment standard for ethylbenzene represents treatment
       of a variety of waste matrices generated by process  streams from
       the manufacture of at least 160 different  products.  The  untreated
       waste concentration of ethylbenzene ranged from 0.010  mg/L  to  80.0
       mg/L in these waste matrices.   All of these  wastes were treated to
       the BDAT treatment standard or below (0.050  mg/L).   We believe
       these constituent reductions substantially diminish  the toxicity
       of the spent solvent wastes containing ethylbenzene  and
       substantially reduce the likelihood of migration of  ethylbenzene
       from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard  for
ethylbenzene was <0.010 mg/L based on biological  treatment  (see  Table 13,
15 FR 1725).  The principal differences between the proposed  and
promulgated treatment  standards are EPA's consideration of  quantification
levels in setting the  standard (see the discussion on  the use of
quantification levels  in Section 5.5  on page 5-12)  and the  incorporation
of a variability factor in derivation of the promulgated  treatment
standard.  Another less significant factor affecting the  change  in the
treatment standard is  the change in data editing  (data editing rules  are
presented in Section 5.3).]
                                   5-46

-------
                            Table 5-12

            TREATMENT  PERFORMANCE  DATA  FOR  ETHYLBENZENE
     Plant 202
Biological Treatment3
Influent   Effluent
 (uo/L)     (ug/L)
    507
    512
    449
    398
    307
    367
    390
    489
    546
    596
    292
    303
    280
    207
    171
     96
    176
    181
    146
    119
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
           Plant 202
     Products Manufactured

Disperse dye coupler
Disperse dyes
Naphthalene sulfonic acid
Organic pigments
p-Phenylene diamine
Sulfur dyes
Vat dyes
Xylenesulfonic acid, sodium
  salt
2-Bromo-4,6-di n i troan i1i ne
2,4-Dinitroaniline
2,4-0 i n i trochlorobenzene
2,4-Dinitrophenol
2,4,6-Trinitrophenol
4-Chloro-2,6-dinitrobenzene
  sulfonic  acid, potassium salt
     Plant 211
Biological Treatment3
Influent   Effluent
 (ug/L)     (UQ/Ll
 80,000
 36,584
 43.171
 17,902
 14,769
 12,923
 64,154
10
10
10
10
10
10
10
           Plant 211
     Products Manufactured

Coal tar solvent
Coatings
Cresols (mixed)
Ethyl benzene
Methyl naphthalene
Naphthalene
Pitch tar residue
Pyridines (tar bases)
2,4-Xylenol (dimethyl phenol)
Phenol
aThe data do not represent paired data (i.e.,  the samples were
 not collected so as to fully account for the retention time in
 the treatment system).
                                  5-47

-------
                     Table 5-12 (Continued)

           TREATMENT  PERFORMANCE DATA  FOR ETHYLBENZENE
     Plant 215                               Plant 215
Biological Treatment3                  Products Manufactured
Influent   Effluent
 (ug/Ll     (uq/L)                Benzene
                                  Toluene
  1,150         10                Mixed xylenes
    564         10                Cyclohexane
  4,150         10                Isobutylene
                                  Propylene
                                  Polypropylene
                                  Butyl rubber
                                  Paraffins
     Plant 221                               Plant 221
Biological Treatment3                  Products Manufactured
Influent   Effluent
 (ug/L)     (ug/L)                Di-isodecyl  phthalate ester
                                  Ethylene
     64         10                Propylene
     10         10                Isopropanol
    140         10                Petroleum hydrocarbon resins
                                  1,3-Butadiene
                                  Butylenes
                                  Cyclopentadiene dimer
                                  Isobutylene
                                  Isoprene
aThe data do not represent paired data (i.e.,  the samples were
 not collected so as to fully account for the  retention time in
 the treatment system).
                                  5-48

-------
                     Table 5-12 (Continued)

           TREATMENT PERFORMANCE DATA  FOR  ETHYLBENZENE
     Plant 230                               Plant 230
Biological Treatment3                  Products Manufactured
Influent   Effluent
 (ug/U)     (ug/Ll                Benzene
                                  Ethylene
  1,217         10                Hydrogen
    893         10                Propylene
  1,537         10                Pyrolysis gasoline
  2,652         10                Polyethylene resin
  3,040         10                Polypropylene
    101         10                Polypropylene resin
    107         10                1,3-Butadiene
    483         10                Butylenes
    628         10
    578         10
    521         10
    440         10
    699         10
    563         10
    389         10
aThe data do not represent paired data  (i.e.,  the  samples  were
 not collected so as to fully account for  the  retention  time  in
 the treatment system).
                                 5-49

-------
                     Table 5-12 (Continued)

          TREATMENT PERFORMANCE DATA FOR ETHYLBENZENE
     Plant 234
Biological Treatment3
Influent
(ug/U
168
390
108
200
157
480
130
114
110
585
90
150
59
90
608
220
260
490
120
228
227
10
10
10
339
10
250
3,850
336
378
295
640
71
Effluent
(UQ/U
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
           Plant 234
     Products Manufactured

Acetic acid
Acetic anhydride
Acetone
Acetaldehyde
Propiomc acid
PET resins/fibers
Acetoacetanil ide
Terephthalic acid
n-Propyl acetate
Oiethyl phthalate
Dimethyl phthalate
di-n-Butyl phthalate
Bis(2-ethylhexyl)phthalate
Methyl isobutyl ketone
Isopropoacetate
Isobutyl acetate
Hydroquinone
aThe data do not represent paired data (i.e.,  the samples were
 not collected so as to fully account for the  retention time in
 the treatment system).
                                  5-50

-------
                      Table 5-12 (Continued)

           TREATMENT  PERFORMANCE DATA FOR ETHYLBENZENE
     Plant. 238
Biological Treatment3
Influent   Effluent
 (UQ/L)     (UQ/L)
  3,350
    220
10
10
           Plant 238
     Products Manufactured

Formaldehyde
Polystyrene (crystal)
Polystyrene (impact)
Polystyrene latex
Polystyrene oriented sheet
ABS resin
Phenolic resins
Styrene-acrylonitrile resin
Styrene maleic anhydride resins
     Plant 242
Biological Treatment3
Influent   Effluent
 (ua/Ll     (ua/Ll
    553
    190
10
10
           Plant 242
     Products Manufactured

Alkyd resins
Epoxy resins
Glyoxal-urea formaldehyde
  textile resin
Unsaturated polyester resins
Acrylic resins
Melamine resins
Urea resins
     Plant 244
Biological Treatment3
Influent   Effluent
 (ua/L)     (ug/L)
    608
                10
                             Plant  244
                       Products  Manufactured

                  Cyclohexanol
                  C4 Hydrocarbons
                  Ethylene
                  Ethylene-methacrylic  acid
                    copolymer
                  Polyethylene polyvinyl  acetate
                    copolymers
                  Propylene
                  Hexamethylenediamine
                  Polyethylene resins
                  Adiponitrile
aThe data do not represent paired data (i.e.,  the samples were
 not collected so as to fully account for the  retention  time in
 the treatment system).
                                  5-51

-------
                      Table 5-12 (Continued)

            TREATMENT  PERFORMANCE DATA  FOR  ETHYLBENZENE
     Plant.251
Biological Treatment3
Influent   Effluent
 (UQ/U     (uq/L)
  1,281
  1,235
  1,360
10
10
10
           Plant 251
     Products Manufactured

Acetone
Acetonitrile
Acrylonitrile
Benzene
Butylenes (mixed)
Dialkylbenzene, by-product
Diphenyl oxide (diphenyl ether)
Ethane
Ethyl benzene
Ethylene
Formaldehyde
Iminodiacetic acid
Naphthalene
Nitrilotnacetic acid
o-Xylene
Phenol
Propylene
Resin tars
Sorbic  acid,  salts
Toluene
1,3-Pentadiene (piperyliene)
Phenolic resins
Cumene
1,3-Butadiene
Cyclopentadiene dimer
Isoprene
Xylenes (mixed)
     Plant 253
Biological Treatment9
Influent   Effluent
 (uq/L)     (uq/L)
                             Plant  253
                       Products  Manufactured

                  Polypropylene  resins
    144
     10
10
10
aThe data do not represent  paired data  (i.e.,  the samples  were
 not collected so as to fully account for  the  retention  time  in
 the treatment system).
                                   5-52

-------
                     Table 5-12 (Continued)

           TREATMENT  PERFORMANCE DATA FOR ETHYLBENZENE
     Plant 257
Biological Treatment3
Influent   Effluent
 (ug/L)     (ug/L)
     63
     71
     67
     75
    355
    327
    239
    139
    179
    149
    159
    153
     94
    124
    116
     85
    122
    172
    141
     83
    157
    231
    376
    608
  3,648
    970
  1.000
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10"
55"
10b
           Plant 257
     Products Manufactured

Acetone
Ally! chloride
Bisphenol-A
Butylenes (mixed)
Diacetone alcohol
Ethylene
Isobutylene
Phenol
Propylene
Vinyl chloride
Epichlorohydrin
Acetone
Epoxy resins
Isopropanol
Methyl ethyl ketone
Methyl isobutyl ketone
n-Butyl alcohol
Cumene
Ethanol
sec-Butyl alcohol
Butadiene
Isoprene
aThe data do not represent  paired  data (i.e.,  the samples  were
 not collected so as to fully account  for the  retention  time  in
 the treatment system).
"in the data base from  which  this  data was taken,  the  sampling
 data was designated using  a  different code, plant 259,  because
 it represented a different sampling episode.
                                   5-53

-------
                      Table 5-12 (Continued)

            TREATMENT  PERFORMANCE DATA  FOR  ETHYLBENZENE
     Plant 293                               Plant 293
Biological Treatment3                  Products Manufactured
Influent   Effluent
 (UQ/U     (ug/L)                Polystyrene (impact)
                                  Polystyrene & copolymers
  3,565         10                Polystyrene oriented sheet
  2,287         10                ABS resin
                                  SAN resin

     Plant 299                               Plant 299
Biological Treatment3                  Products Manufactured
Influent   Effluent
 (ug/L)     (ug/L)                ABS resin

    114         10
     22         10
    230         10
    112         10
     82         10
    114         10
     85         10
     77         10
     79         10
     81         10
    132         10
     75         10
    124         10
    144         10
     99         10
    105         10

  Stover and Kincannon,  1983              Description  of
  Pilot-Scale Steam Stripper              Waste Treated
     Influent   Effluent
      (ug/L)     (ug/L)            Pilot-scale study of ground-
                                  water near a waste  disposal
      23,500         10            dump site which contained
      23,500         10            household refuse, demolition
      23,500         10            materials,  chemical  sludges,
      23,500        992            and hazardous liguid chemicals.
      23,500         10

aThe data do not represent  paired data (i.e.,  the samples  were
 not collected so as to  fully account for the retention  time in
 the treatment system).
                                   5-54

-------
                    Table 5-12 (Continued)

         TREATMENT PERFORMANCE DATA FOR ETHYLBENZENE
Stover and Kincannon,  1983             Description  of
 Pilot-Scale Air Stripper              Waste  Treated
   Influent   Effluent
    (ug/L)     (uq/U            Pilot-scale study of ground-
                                water  near a  waste  disposal
    23,500         53            dump site which  contained
    23,500         10            household refuse, demolition
    23,500        528            materials, chemical sludges.
    23,500        558            and hazardous liquid chemicals.
    23,500         10
    23,500      1,035
                                 5-55

-------
                          Table 5-13

             CALCULATION OF BOAT FOR ETHYLBENZENE
                        Average
                       Treatment                 Treatment Con-
Plant
No.
202
211
215
221
230
234
238
242
244
251
253
257
293
299
Concentration Variability
Technology (ua/L) Factor
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
10
10
10
10
10
10
10
10
10
10
10
11.7
10
10
3.39a
3.39a
3.39a
3.39a
3.39a
3.39a
3.39a
3.39a
3.39a
3.39a
3.39a
1.98
3.39a
3.39a
centration Level
Ava. x VF (ua/L)
34
34
34
34
34
34
34
34
34
34
34
23
34
34
aAverage variability factor  for  BOAT  Biological  Treatment  (see
 Table 5-4 and the discussion  on page 5-17).
                                5-56

-------
5.5.13   Ethyl Ether Wastewaters

    The Agency has no data for wastewater treatment for the removal of
ethyl ether.  For reasons presented in Section 5.5.1, EPA used chemical
structure as the basis for transferring treatment data to ethyl ether
spent solvent .wastewaters.  Specifically, we transferred the treatment
data from methyl isobutyl ketone, which contains the ketone functional
group, to ethyl ether, which contains the ether functional group.  The
ethers structural group is most structurally similar to the ketones group
based upon their oxygen containing, electron-releasing functional
groups.  Methyl isobutyl ketone was the only constituent for which we had
data in the ketones structural group.  Using performance data from methyl
isobutyl ketone, the BDAT treatment standard for ethyl ether is 0.05
mg/L.  The technology basis for this treatment standard is steam
stripping.

    We believe the BDAT treatment standard for ethyl ether spent solvent
wastewaters represents substantial treatment.   As discussed on page 5-73,
in reference to methyl isobutyl ketone, we believe these constituent
reductions substantially diminish the toxicity of the spent solvent
wastes containing ethyl ether and substantially reduce the likelihood of
migration of ethyl ether from spent solvent wastes.

    [The proposed technology-based treatment standard for ethyl ether was
estimated at the detection limit of <0.100 mg/L based on biological
treatment (see Table 13, 51 FR 1725).  The principal difference between
the proposed and promulgated treatment standards is  the Agency's change
in the criteria for transfer of treatment data (see  Section 5.5.21 for a
discussion of the Agency's methodology for data transfer).]
                                   5-57

-------
5.5.14   Isobutanol Wastewaters

    The Agency has no data for wastewater treatment for the removal of
isobutanol.  For reasons presented in Section 5.5.1, EPA used chemical
structure as the basis for transferring treatment data to isobutanol
spent solvent wastewaters.  Specifically, we transferred the treatment
data from methyl isobutyl ketone,  which contains the ketone functional
group, to isobutanol, which contains the hydroxyl functional group.  The
alcohol structural group is most structurally similar to the ketones
group based upon their oxygen containing, electron-releasing functional
groups.  Methyl isobutyl ketone was the only constituent for which we had
data in the ketones structural group.  Using performance data from methyl
isobutyl ketone, the transferred standard for isobutanol is 0.05 mg/L.
The standard derived from the transferred data is below the
quantification level and could not be used as the treatment standard.
Therefore, the BDAT treatment standard was set at the quantification
level of 5.0 mg/L.  The technology basis for this treatment standard is
steam stripping.

    We believe the BDAT treatment standard for isobutanol spent solvent
wastewaters represents substantial treatment.  We would expect untreated
isobutanol wastes to be similar to untreated methyl isobutyl ketone
wastes, from which we transferred treatment data, since they are used in
some of the same manufacturing processes, as shown in Section 2 of this
document.  As discussed on page 5-73, in reference to methyl isobutyl
ketone, we believe these constituent reductions substantially diminish
the toxicity of the spent solvent wastes containing isobutanol and
substantially reduce the likelihood of migration of isobutanol from spent
solvent wastes.

    [The proposed technology-based BDAT treatment standard for isobutanol
was estimated at the detection limit of <0.050 mg/L based on biological
treatment (see Table 13, 51 FR 1725).  The primary difference between the
proposed and promulgated treatment standards is EPA's consideration of
quantification levels in setting the standard (see the discussion on the
use of quantification levels in Section 5.5 on page 5-12).  To a lesser
extent, the Agency's change in the criteria for data transfer affected
the treatment standard.  (See Section 5.5.1, page 5-14, for a discussion
of the Agency's methodology for data transfer.)]
                                   5-58

-------
5.5.15   Methanol Wastewaters

    The Agency has wet air oxidation treatment data for methanol
(Reference 10).  The data are summarized in Table 5-14.

    The following steps were taken to derive the BDAT treatment standard
for methanol:

    1. We evaluated the data set to determine whether any of the data
       represent poor design or operation of the treatment system.   The
       data for wet air oxidation treatment of methanol were deleted
       because we did not believe the treatment to be substantial.   By
       transferring data from another technology, a BDAT treatment
       standard over 10,000 times (four orders of magnitude) smaller
       could be obtained.  We have no information to conclude,  nor  do we
       believe, that the wastes treated by the wet air oxidation unit are
       sufficiently different from the similarly-treated wastes on  which
       the standard was based to account for this large difference  in
       treatability.  Taking the variability into account, the  standard
       derived from wet air oxidation would also be over 200 times
       greater than any BDAT treatment standard.  We therefore  conclude
       that the treatment represented by this data set for wet  air
       oxidation treatment of methanol does not represent substantial
       reductions in toxicity or likelihood of migration.

    2. Because the Agency has no other data for treatment of methanol,
       treatment data for methanol were transferred from another
       compound.  For reasons presented in Section 5.5.1, EPA used
       chemical structure as the basis for transferring treatment data to
       methanol spent solvent wastewaters.  Specifically, we transferred
       the treatment data from methyl isobutyl ketone, which contains the
       ketone functional group, to methanol, which contains the hydroxyl
       functional group.  The alcohols structural group is most
       structurally similar to the ketones group based upon their oxygen
       containing, electron-releasing functional groups.  Methyl isobutyl
       ketone was the only constituent for which we had data in the
       ketones structural group.  Using performance data from methyl
       isobutyl ketone, the transferred standard for methanol is 0.05
       mg/L.  The standard derived from the transferred data is below the
       quantification level and could not be used as the treatment
       standard.  Therefore, the BDAT treatment standard was set at the
       quantification level of 0.25 mg/L.  The technology basis for this
       treatment standard is steam stripping.

       We believe the BDAT treatment standard for methanol spent solvent
       wastewaters represents substantial treatment.  We would  expect
       untreated methanol wastes to be similar to untreated methyl
                                   5-59

-------
       isobutyl ketone wastes,  from which we transferred treatment
       performance, since they are used in some of the same manufacturing
       processes, as shown in Section 2 of this document.   As discussed
       on page 5-73, in reference to methyl isobutyl ketone, we believe
       these constituent reductions substantially diminish the toxicity
       of the spent solvent wastes containing methanol and substantially
       reduce the likelihood of migration of methanol from spent solvent
       wastes.

    [The proposed technology-based BOAT treatment standard for methanol
was estimated at the detection limit of <0.100 mg/L based on biological
treatment (see Table 13, 51 FR 1725).  The principal difference between
the proposed and promulgated treatment standards is EPA's consideration
of quantification levels in setting the standard (see the discussion on
the use of quantification levels in Section 5.5 on page 5-12).  To a
lesser extent, the Agency's change in the criteria for data transfer
affected the treatment standard.  (See Section 5.5.1, page 5-14, for a
discussion of the Agency's methodology for data transfer.)]
                                   5-60

-------
                       Table 5-14

         TREATMENT  PERFORMANCE DATA FOR METHANOL
Data Submitted by Zimpro,  Inc.,  1986
	Met Air Oxidation	
               Diluted    Oxidation
 Raw Waste      Feed       Product       Description of
   (ug/L)       (ug/L)     (ug/L)         Waste Treated

 36,900,000   9,200,000    800,000       General  organic
                                   5-61

-------
5.5.16   Methylene Chloride Wastewaters

    The Agency has biological treatment data for methylene chloride at
plants 246 and 265 in the OCPSF data base.  Also from the OCPSF data base
are steam stripping data at plant 284 and biological treatment followed
by activated carbon adsorption data at plant 246.  The Agency also has
data from pilot-scale granular activated carbon adsorption (Reference 7)
and data from wet air oxidation treatment (Reference 10).  Data are also
available for steam stripping of methylene chloride wastewater from the
Pharmaceuticals Manufacturing Industry (plant 12003 of the Industrial
Technology Division data base).  The data are summarized in Table 5-15
and calculation of the BOAT treatment standard is shown in Table 5-16.

    The following steps were taken to derive the BOAT treatment standard
for methylene chloride:

    1. We evaluated each data set to determine whether any of the data
       represent poor design or operation of the treatment systems.  Data
       for steam stripping of methylene chloride wastewater at the
       Pharmaceuticals manufacturing facility (plant 12003) were
       evaluated to determine whether the steam stripper could be
       considered well-designed and operated.  The steam stripper was
       designed to operate at 98°C in the overhead.  However, many data
       points were obtained during operation at overhead temperatures
       below 98°C.  Therefore, the data were examined to determine the
       minimum temperature representative of a well-operated system.  As
       a method of evaluating the data, the effluent concentration was
       plotted as a function of overhead temperature.  The data indicate
       that, as the overhead temperature drops below the design
       temperature, there is an increase in the variability in the
       effluent concentrations achieved at a given overhead temperature.
       This increased variability is an indication of increased
       instability or poor control of the steam stripping system.  Since
       the variability in the effluent concentrations increased as the
       overhead temperature dropped below 90°C, the minimum overhead
       temperature for a system that was well-operated was estimated as
       90°C.  Twenty-one data points were deleted from the data set
       because the overhead temperature was below 90°C.

       In consideration of the amount of full-scale data available for
       methylene chloride, we believe it is appropriate to exclude data
       for pilot-scale activated carbon adsorption and bench-scale wet
       air oxidation treatment.
                                   5-62

-------
2. We calculated the arithmetic average treatment concentration and
   the variability factor for each data set as shown in Table 5-16.
   Process variability could not be calculated for biological
   treatment followed by activated carbon adsorption at plant 246
   because all effluent values were reported as less than or equal to
   the detection limit of 10 ug/L.  We would expect some variability
   in the data because the actual concentrations would range from 0
   to the detection limit of 10 ug/L.  To estimate the variability,
   the Agency used the average variability factor for BDAT biological
   treatment followed by activated carbon adsorption, 6.17.
   (Calculation of the average variability factor is shown in
   Table 5-4, page 5-18)

3. Biological treatment and biological treatment followed by
   activated carbon adsorption of methylene chloride at plant 246
   were compared to determine whether the performance of one
   technology was significantly better than the other for treatment
   of the same waste.  Since all effluent values in each data set
   were reported as less than or egual to the detection limit, the
   two data sets were considered statistically homogeneous.  The
   combined biological treatment followed by activated carbon
   adsorption data set was not used to determine the BDAT treatment
   standard because the addition of activated carbon adsorption did
   not significantly improve treatment performance.   Therefore, the
   treatment concentration level for plant 246 is 0.20 mg/L based
   upon biological treatment.

   Data were not available for more than one treatment for other
   wastes; therefore, the analysis of variance method could not be
   used to compare treatments on other wastes.

4. EPA then analyzed the data to determine if the various treatment
   concentration levels shown in Table 5-16 could be associated with
   separate waste treatability subgroups.   Methylene chloride spent
   solvent wastewater generated at a Pharmaceuticals Manufacturing
   facility was identified as a separate waste treatability
   subgroup.  Data were insufficient to identify other waste
   treatability subgroups; therefore, a second waste treatability
   subgroup was established for all remaining sources of wastewater
   containing methylene chloride spent solvents.   We then compared
   the treatment levels for the two waste treatability subgroups by
   the analysis of variance.   The treatment levels are significantly
   different (see Table II-6, Appendix II).   A separate BDAT
   treatment standard (12.7 mg/L from plant 12003) was developed for
   the Pharmaceuticals manufacturing industry based on the data for
   steam stripping of methylene chloride.   The least stringent
                               5-63

-------
       treatment level within the second treatability subgroup for all
       other methylene chloride wastewaters was selected for BDAT (0.20
       mg/L from plant 265) to ensure that the standard could be achieved
       for all waste matrices within the waste treatability subgroup.
       The technology basis was biological treatment.

    5. The BDAT treatment standard for methylene chloride represents
       treatment of a variety of waste matrices generated by process
       streams from the manufacture of over 39 different products.  The
       untreated waste concentration of methylene chloride ranged from
       7,000 mg/L to 10,000 mg/L in pharmaceuticals wastewater.   This
       waste was treated to the BDAT treatment standard developed for
       methylene chloride wastewaters from pharmaceuticals manufacturing
       or below (12.7 mg/L).  The untreated waste concentration of
       methylene chloride ranged from 0.027 mg/L to 12.1 mg/L in all
       other waste matrices.  All of these wastes were treated to the
       BDAT treatment standard or below (0.20 mg/L).   We believe these
       constituent reductions substantially diminish the toxicity of the
       spent solvent wastes containing methylene chloride and
       substantially reduce the likelihood of migration of methylene
       chloride from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard for methylene
chloride was 0.011 mg/L based on biological treatment (see Table 13, 51
FR 1725).  The difference between the proposed and promulgated treatment
standards is primarily due to the use of additional data on treatment  of
methylene chloride and the incorporation of a variability factor in
derivation of the promulgated treatment standard.  The additional data
supported development of a separate waste treatability subgroup for
methylene chloride spent solvent wastewaters from pharmaceuticals
manufacturing as discussed above.  The additional data were presented in
EPA's Notice of Availability of Data (51 FR 31783).  Another less
significant factor affecting the change in the treatment standard is the
change in data editing (data editing rules are presented in Section 5.3).]
                                   5-64

-------
                            Table 5-15

         TREATMENT  PERFORMANCE  DATA  FOR  METHYLENE  CHLORIDE
     Plant 246                               Plant 246
Biological Treatment3                  Products Manufactured
Influent   Effluent
 (ug/L)     (ug/L)                Aniline
                                  Dinitrotoluene (mixed)
     27         10                Methylene diphenyl diisocyanate
     94         10                Nitrobenzene
  1,817         10                Polymeric methylene diphenyl
    717         10                  diisocyanate
    154         10                Polyoxypropylene glycol
    133         10                Toluene diamine (mixture)
    501         10                Toluene diisocyanates (mixture)
    135         10                Polymeric methylene dianiline
    460         10                Polyurethane resins
  1,640         26
  3,907         10
    969         10
    277         10
aThe data do not represent paired data (i.e.,  the samples were
 not collected so as to fully account for the  retention time in
 the treatment system).
                                   5-65

-------
                      Table 5-15 (Continued)

         TREATMENT  PERFORMANCE DATA  FOR METHYLENE  CHLORIDE
         Plant 246
   Biological Treatment3
        Followed by
Activated Carbon Adsorption
    Influent   Effluent
     (uq/L)     (UQ/L)
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10
                    10b
                    10b
                    10b


1





2

3














5
3
27
94
,817
717
154
133
501
135
,062
460
.907
969
277
10
10
10
10
10
10
10
10
10
10
10
10
,550
,005
                                       Plant 246
                                 Products Manufactured

                            Aniline
                            Oinitrotoluene (mixed)
                            Methylene diphenyl  diisocyanate
                            Nitrobenzene
                            Polymeric methylene diphenyl
                              diisocyanate
                            Polyoxypropylene glycol
                            Toluene diamine (mixture)
                            Toluene diisocyanates (mixture)
                            Plymeric methylene  dianiline
                            Polyurethane resins
2,980
aThe data do not represent paired data (i.e.,  the samples were
 not collected so as to fully account for the  retention  time in
 the treatment system).
bln the data base from which this data was taken, the sampling
 data was designated using a different code,  plant 219,  because
 it represented a different sampling episode.
                                    5-66

-------
                     Table 5-15 (Continued)

        TREATMENT PERFORMANCE  DATA FOR  METHYLENE  CHLORIDE
     Plant 265
Biological Treatment3
Influent   Effluent
 (ua/L)     (uq/U
    760
    690
    500
     60
     10
     10
                                  Plant 265
                            Products Manufactured

                       Tar,  tar crudes, and tar
                         pitches
     Plant 284
  Steam Stripping
Influent
 (ug/L)
   .135
   ,600
   ,140
  1 ,760
  2,400
    690
    570
    320
    267
    520
    198
    641
  4,800
 12,100
    469
Effluent
 (ug/U

     10
     10
     10
     10
     10
     10
     10
     10
     10
     10
     10
     10
     10
     10
     18
           Plant 284
     Products Manufactured

Benzene
1,3-Butadiene
Ethylene
Propylene
Methylene chloride
1,1,2-Trichloroethane
Vinylidine chloride
1,2,3-Tnchloropropene
1,2-Dichloropropane
Propylene oxide
Ethylene oxide
Propylene glycol
Dipropylene glycol
Tripropylene glycol
Ethylene glycol
Methyl  chloride
Oiethylene glycol
Triethylene glycol
Tetraethylene glycol
Ethanol amines
Polypropylene
Chloroform
Carbon  tetrachloride
1,2-Dichloroethane
Vinyl chloride
aThe data do not represent paired data (i.e.,  the samples  were
 not collected so as to fully account  for the  retention time in
 the treatment system).
                                    5-67

-------
                    Table 5-15 (Continued)

       TREATMENT PERFORMANCE  DATA  FOR  METHYLENE  CHLORIDE
    Plant 12003
  Steam Stripping
Influent
(ua/U
8,250,000
8,250,000
8,250,000
8,250,000
8,250,000
8,250,000
8,250,000
8,250,000
225,000
225,000
225,000
225,000
225,000
225,000
225,000
225,000
7,000,000
7,000,000
7,000,000
7,000,000
7,000,000
7,000,000
7,000,000
7,000,000
11,200,000
9,900,000
9,100,000
9,400,000
10,200,000
11 ,800.000
10,000,000
12,000,000
9,500,000
9,500,000
9,500,000
9,500,000
9,500,000
9,500,000
9,500,000
9,500,000
Effluent
(ug/L)
926
5,100
4,940
3, OOO3
1.9903
5.7003
22.8003
38,050a
3,900a-b
8,360a-b
20,600a-b
4,070a-b
10,700a-b
20,300a-b
4,800a>b
7,8703-b
1 ,720
1,630
3,600a
14.2503
39.3003
138, OOO3
110, OOO3
60.8003
10.1003
22,850a
57.5003
115, OOO3
59.9003
127.0003
3,180
3,730a
7,200
4,040
4,270
1,470
1.6203
2.630
7.8303
15.8003
           Plant 12003
     Products Manufactured

Pharmaceuticals
aData point deleted in analysis  -  overhead  temperature  less
 than 90°C.
bThese data were deleted because the  document  from which
 they were obtained (Reference 14)  stated that  the data set
 was suspect.

                              5-68

-------
                      Table 5-15 (Continued)

         TREATMENT  PERFORMANCE DATA  FOR METHYLENE CHLORIDE
   Becker and Wilson, 1978
    Pilot-Scale Granular
   Activated Carbon Column
     Influent   Effluent
      (UQ/D     (ua/L)
         190
                          Description of
                          Waste Treated

                   Runoff water from a waste dis-
                   posal site's containment dikes.
      51.0
Data Submitted by Zimpro,  Inc.,  1986
	Viet Air Oxidation	
 Raw Waste
   (ug/L)

 3,600,000
    15,000
   500,000
Diluted
 Feed
 (uo/L)
125,000
Oxidation
 Product
 (ug/L)

   4,000
  <1,000
 <10,000
Description of
Waste Treated

General organic
                                 5-69

-------
                           Table 5-16

           CALCULATION OF BOAT FOR METHVLENE CHLORIDE


                         Average
                        Treatment                 Treatment  Con-
Plant                 Concentration Variability  centration  Level
 No.      Technology     (ug/L)	Factor     Avg. x VF (ug/L)

Pharmaceuticals Manufacturing Industry:

 12003  Steam Stripping  3,375        3.76          12,690
All Other Methylene Chloride Wastewaters:

 284    Steam Stripping     10.5      1.40              15

 246    Biological           11.2      1.78              20

 265    Biological           26.7      7.58             202

 246    Biological  fol-     10        6.17a             62
         lowed by
         Activated
         Carbon
aAverage variability factor for BOAT biological  treatment
 followed by activated carbon adsorption (see Table 5-4 and
 the discussion on page 5-17).
                                5-70

-------
5.5.17   Methyl Ethyl Ketone Wastewaters

    The Agency has wet air oxidation treatment data for methyl ethyl
ketone (Reference 10).  The data are summarized in Table 5-17.

    The following steps were taken to derive the BOAT treatment standard
for methyl ethyl ketone:

    1. We evaluated the data set to determine whether any of the data
       represent poor design or operation of the treatment system.  Data
       on bench-scale wet air oxidation treatment were deleted because
       the concentration of methyl ethyl ketone in the diluted feed to
       the wet air oxidation process was not reported and the detection
       limit was not reported with the data.

    2. Because the Agency has no other data for treatment of methyl ethyl
       ketone, treatment data for methyl ethyl ketone were transferred
       from another compound.  For reasons presented in Section 5.5.1,
       EPA used chemical structure as the basis for transferring
       treatment data to methyl ethyl ketone spent solvent wastewaters.
       Specifically, we transferred treatment data from methyl isobutyl
       ketone because, like methyl ethyl ketone, methyl isobutyl ketone
       contains the ketone functional group.  Methyl isobutyl ketone was
       the only constituent for which we had data in the ketone
       structural group.  Using performance data from methyl isobutyl
       ketone, the BDAT treatment standard for methyl ethyl ketone is
       0.05 mg/L.  The technology basis for this treatment is steam
       stripping.

       We believe the BDAT treatment standard for methyl ethyl ketone
       spent solvent wastewaters represents substantial treatment.  We
       would expect untreated methyl ethyl ketone wastes to be similar to
       untreated methyl isobutyl ketone wastes, from which we transferred
       treatment performance, since they are used in some of the same
       manufacturing processes, as shown in Section 2 of this document.
       As discussed on page 5-73, in reference to methyl isobutyl ketone,
       we believe these constituent reductions substantially diminish the
       toxicity of the spent solvent wastes containing methyl ethyl
       ketone and substantially reduce the likelihood of migration of
       methyl ethyl ketone from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard for methyl
ethyl ketone was estimated at the detection limit of <0.050 mg/L based on
biological treatment (see Table 13, 51 FR 1725).  The principal
difference between the proposed and promulgated treatment standards is
the Agency's change in the criteria for data transfer.   (See Section
5.5.1, page 5-14, for a discussion of the Agency's methodology for data
transfer.)]
                                   5-71

-------
                           Table 5-17

       TREATMENT PERFORMANCE DATA FOR METHYL  ETHYL  KETONE
Data Submitted by Zlmpro,
       Inc.,  1986
     Het Air  Oxidation
              Oxidation
 Raw Waste     Product                   Description  of
   (ug/L)      (UQ/D                    Waste Treated

 8,200,000  Not detected3         Solvent still  bottom waste-
                                  water.
aThe detection limit was not reported with the data.
                                  5-72

-------
5.5.18   Methyl Isobutyl Ketone Wastewaters

    The Agency has treatment data for methyl isobutyl ketone from
pilot-scale steam stripping and pilot-scale air stripping of solvent
contaminated groundwater (Reference 2).  The data are summarized in
Table 5-18 and calculation of the BDAT treatment standard is shown in
Table 5-19.

    The following steps were taken to derive the BDAT treatment standard
for methyl isobutyl ketone:

    1. We evaluated each data set to determine whether any of the data
       represent poor design or operation of the treatment systems.  The
       available data and information did not show any of the data to
       represent poor design and operation.  Accordingly, none of the
       data were deleted on this basis.

    2. We calculated the arithmetic average treatment concentration and
       the variability factor for each data set as shown in Table 5-19.
       Process variability could not be calculated for the pilot-scale
       steam stripper because all effluent values were reported as less
       than or equal to the detection limit of 10 ug/L.   We would expect
       some variability in the data because the actual concentrations
       would range from 0 to the detection limit of 10 ug/L.   To estimate
       the variability, the Agency used the average variability factor
       for BDAT full-scale steam stripping, 2.26.  (Calculation of the
       average variability factor is shown in Table 5-4, page 5-18.)

    3. Air stripping and steam stripping of methyl isobutyl ketone at the
       pilot-scale plant were compared with the analysis of variance
       method to determine whether the performance of one technology was
       significantly better than the other for treatment of the same
       waste.  It was shown that steam stripping provided significantly
       better removal of methyl isobutyl ketone when compared with air
       stripping.  Therefore, the treatment concentration level for the
       pilot-scale plant is 23 ug/L based upon steam stripping.   (Refer
       to Table II-7, Appendix II.)  The analysis of variance method
       could not be used to compare treatments on any other methyl
       isobutyl ketone spent solvent wastes because data were not
       available for more than one treatment for any other wastes.

    4. EPA then analyzed the data to determine if the various treatment
       concentration levels shown in Table 5-19 could be associated with
       separate waste treatability subgroups.   Sufficient data did not
                                   5-73

-------
       exist to identify separate waste treatability subgroups;
       therefore, one waste treatability subgroup was established for all
       sources of wastewaters containing methyl isobutyl  ketone  spent
       solvents.   The least stringent treatment level within the
       treatability subgroup was selected for BOAT (0.025 mg/L from
       pilot-scale steam stripping) to ensure that the standard could be
       achieved for all waste matrices within the waste treatability
       subgroup.   This calculated concentration level is  below the
       quantification level and could not be used as the  treatment
       standard;  therefore, the treatment standard was set at the
       quantification level of 0.05 mg/L.

    5. The BOAT treatment standard for methyl isobutyl ketone represents
       treatment  of a waste matrix generated from the manufacture of four
       different  products.  The untreated waste concentration of methyl
       isobutyl ketone was as high as 76.4 mg/L in this waste matrix.
       This waste was treated to a concentration below the BDAT  treatment
       standard (0.050 mg/L).  We believe these constituent reductions
       substantially diminish the toxicity of the spent solvent  wastes
       containing methyl isobutyl ketone and substantially reduce the
       likelihood of migration of methyl isobutyl ketone  from spent
       solvent wastes.

    [The proposed technology-based BDAT treatment standard for methyl
isobutyl ketone was estimated at the detection limit of <0.100 mg/L based
on biological treatment (see Table 13, 51 PR 1722).  The  principal
differences between the proposed and promulgated treatment standards are
EPA's consideration of quantification levels in setting the standard (see
the discussion on the use of quantification levels in Section 5.5 on page
5-12) and the incorporation of a variability factor in derivation of the
promulgated treatment standard.  Another less significant factor
affecting the change in the treatment standard is the change in data
editing (data editing rules are presented in Section 5.3).]
                                   5-74

-------
                            Table 5-18

      TREATMENT PERFORMANCE DATA FOR METHYL ISOBUTYL KETONE
Stover and Kincannon,  1983
Pilot-Scale Steam Stripper
   Influent   Effluent
    (ug/L)     (ug/L)
    76.400
    76,400
    76,400
    76,400
    76,400
    10
    10
    10
    10
    10
       Description of
       Waste Treated

Pilot-scale study of ground-
water near a waste disposal
dump site which contained
household refuse, demolition
materials, chemical sludges,
and hazardous liquid chemicals.
Stover and Kincannon,  1983
 Pilot-Scale Air Stripper
   Influent   Effluent
    (UQ/D     (ug/L)
    76,400
    76,400
    76,400
    76,400
    76,400
    76,400
45,000
60,000
24,400
42,800
18,500
60,200
       Description of
       Waste Treated

Pilot-scale study of ground-
water near a waste disposal
dump site which contained
household refuse, demolition
materials, chemical sludges,
and hazardous liquid chemicals.
                                  5-75

-------
                           Table 5-19

         CALCULATION OF  BOAT FOR METHYL  ISOBUTYL  KETONE


                         Average
                        Treatment                 Treatment.  Con-
Plant                 Concentration Variability  centration  Level
 No.       Technology     (ug/L)	Factor     Avg.  x  VF  (ug/L)

 PS     Air Stripping     41,817      2.83           118,342

 PS     Steam Stripping        10      2.26a              23
aAverage variability factor for BOAT full-scale steam stripping
 (see Table 5-4 and the discussion on page 5-17).
                                5-76

-------
5.5.19   Nitrobenzene Wastewaters

    The Agency has data for treatment of wastewaters containing
nitrobenzene by biological treatment at plant 246, biological treatment
followed by activated carbon adsorption at plant 246, steam stripping at
plants 246 and 297, and steam stripping followed by activated carbon
adsorption at plant 297 in the OCPSF data base.   The data are summarized
in Table 5-20 and calculation of the BOAT treatment standard is shown in
Table 5-21.

    The following steps were taken to derive the BDAT treatment standard
for nitrobenzene:

    1. We evaluated each data set to determine whether any of the data
       represent poor design or operation of the treatment systems.  In
       EPA's judgment, one data point in the data set for steam stripping
       at plant 297 represented poor design and operation.  We confirmed
       this judgment using the outlier test (refer to Table II-8,
       Appendix II).  The outlying data point was deleted.  Data for
       steam stripping at plant 246 were deleted on the basis of design
       and performance.  Based on the disproportionately low removals
       relative to other treatment systems for wastes containing
       nitrobenzene, EPA judged this system to be poorly designed and
       operated.  This system achieved a reduction of only 35.7 percent
       as compared with 93.8 to 99.9 percent for other systems treating
       wastes containing nitrobenzene.

       Data for biological treatment and biological treatment followed by
       activated carbon adsorption at plant 246 were also deleted.
       During the sampling episode, this plant experienced high
       discharges of polyoxypropylene glycol, or "polyol", a product at
       the plant, into the treatment system.  The discharge of polyol is
       normally closely controlled at this plant since the polyol
       interferes with removals of nitrobenzene in the treatment system.
       The data for this plant were not considered in developing BDAT
       treatment standards since the treatment system was not
       well-operated at the time of sampling.

    2. We calculated the arithmetic average treatment concentration and
       the variability factor for each data set as shown in Table 5-21.

    3. Steam stripping and steam stripping followed by activated carbon
       adsorption of nitrobenzene at plant 297 were compared with the
       analysis of variance method to determine  whether the performance
       of one technology was significantly better than the other for
       treatment of the same waste.  It was shown that the addition of
       activated carbon adsorption to steam stripping significantly
                                   5-77

-------
       improved treatment performance.   Therefore,  the  treatment
       concentration level for plant  297 is  0.66  mg/L based  on  steam
       stripping followed by activated  carbon adsorption.   (Refer  to  the
       statistical  calculations and results  in Table II-9, Appendix II.)
       The analysis of variance method  could not  be used to  compare
       treatments on .any other wastes because data  were not  available for
       more than one treatment for any  other waste.

    4.  EPA then analyzed the data to  determine if the various treatment
       concentration levels shown in  Table 5-21 could be associated with
       separate waste treatability subgroups.   Sufficient data  did not
       exist to identify separate waste treatability subgroups;
       therefore, one waste treatability subgroup was established  for all
       sources of wastewaters containing nitrobenzene spent  solvents.
       The highest  treatment level within the treatability  subgroup was
       selected for BOAT (0.66 mg/L from plant 297) to  ensure that the
       standard could be achieved for all waste matrices within the waste
       treatability subgroup.  The technology basis was steam stripping
       followed by  activated carbon adsorption.

    5.  The BDAT treatment standard for  nitrobenzene represents  treatment
       of a waste matrix generated by process streams from  the
       manufacture  of at least four different products.  The untreated
       waste concentration of nitrobenzene ranged from  87 mg/L  to  330
       mg/L in this waste matrix.  This waste was treated to the BDAT
       treatment standard or below (0.66 mg/L).   We believe  these
       constituent  reductions substantially  diminish the toxicity  of  the
       spent solvent wastes containing  nitrobenzene and substantially
       reduce the likelihood of migration of nitrobenzene from  spent
       solvent wastes.

    [The proposed technology-based BDAT treatment standard  for
nitrobenzene was <0.010 mg/L based on biological  treatment  (see Table 13,
51 FR 1725).  The difference between the proposed and promulgated
treatment standards is primarily due  to the  incorporation of a
variability factor  in derivation of the promulgated treatment standard.
Other lesser factors affecting the change in the  treatment  standard are
the changes in data editing (data editing rules are presented in Section
5.3) and deletion of some data points in the final  rule because they
represent poor operation of the treatment system  (see the discussion  of
the outlier test in Section 5.4.)]
                                   5-78

-------
                            Table  5-20

           TREATMENT PERFORMANCE DATA FOR NITROBENZENE
     Plant  246
  Steam Stripping
Influent
(ua/L)
290,780
226,415
196,530
155,310
363,560
1,965,760
361,510
675,000
290,460
344,720
91,200
201,990
230,540
233,786
237,940
Plant
Bioloaical
Influent
(ug/L)
5,559
2,779
2,405
3,796
4,400
2,838
3,656
1 ,214
2,319
1 ,420
2,062
821
1,145
876
Effluent

-------
                      Table 5-20 (Continued)

            TREATMENT  PERFORMANCE DATA FOR  NITROBENZENE
     Plant 246
Biological Treatment3
   and Activated                             Plant 246
 Carbon Adsorption                     Products Manufactured
Influent   Effluent
 (ua/U     (ug/L.)                Aniline
                                  Dinitrotoluene (mixed)
  5,559        982                Methylene diphenyl diisocyanate
  2,779      1,902                Nitrobenzene
  3,405        141                Polymeric methylene diphenyl
  3,796        538                  diisocyanate
  4,400        537                Polyoxypropylene glycol
  2,838         79                Toluene diamine (mixture)
  3,656        420                Toluene diisocyanates (mixture)
  2,015         16                Polymeric methylene diamline
  1,214         10                Polyurethane resins
  2,319        233
  1,420         10
  2,063         10
    821         10
  1,145         10
    876         10
 87,000        230b
 45,030        179b
 90,500         38b
aThe data do not represent paired data (i.e.,  the samples were
 not collected so as to fully account for the  retention time in
 the treatment system).
bln the data base from which this data was taken, the sampling
 data was designated using a different code,  plant 248, because
 it represented a different sampling episode.
                                  5-80

-------
                      Table 5-20 (Continued)

           TREATMENT PERFORMANCE DATA FOR NITROBENZENE
     Plant 297                               Plant 297
  Steam Stripping                      Products Manufactured
                                  Nitrobenzene
                                  Nitrototuene
                                  Aniline
                                  o-Toluidine
Influent
(ug/L)
330,000
190,000
267,160
309,920
106,995
144,860
139,530
87,000
139,340
189,054
Effluent
(ug/L)
14,377
10,545
8,752
4,600
6.098
11,072
21,992
17,065
12,264
11,163
         Plant 297
Steam Stripping Followed by                  Plant 297
 Activated Carbon Adsorption           Products Manufactured
    Influent   Effluent
     (ug/L)     (ug/L)            Same as Plant 297 - Steam
                                  Stripping
330,000
190,000
267,160
309,920
106,995
144,860
139,530
87,000
139.340
189.054
374
150
143
330
372
140
4,900a
135
331
251
aln EPA's judgment,  this data point  represented poor design  and
 operation.   We confirmed this judgment  using  the outlier  test
 (refer to Table II-8,  Appendix II)  and  this data point  was
 deleted.
                                  5-81

-------
                           Table  5-21

              CALCULATION  OF  BOAT FOR  NITROBENZENE
                         Average
                        Treatment                 Treatment Con-
Plant                 Concentration  Variability  centration Level
 No.       Technology     (ug/L)	Factor     Avg. x VF (ug/L)

 297     Steam Stripping  11,793         2.68         31,605

 297     Steam Stripping     247         2.65            655
         followed by
         Activated
         Carbon
                                5-82

-------
5.5.20   Pyridine Wastewaters

    The Agency has no data for wastewater treatment for the removal of
pyridine.  For reasons presented in Section 5.5.1, EPA used chemical
structure as the basis for transferring treatment standards to pyridine
spent solvent wastewaters.  Specifically, we transferred the treatment
data from toluene because, like pyridine, toluene contains the aromatic
ring functional group.  Toluene had the least stringent treatment
standard in the non-halogenated aromatics structural group.  Using
performance data from toluene, the BDAT treatment standard for pyridine
is 1.12 mg/L.  The technology basis for this treatment is biological
treatment followed by activated carbon adsorption.

    We believe the BDAT treatment standard for pyridine spent solvent
wastewaters represents substantial treatment.  We would expect untreated
pyridine wastes to be similar to untreated toluene wastes, from which we
transferred treatment data.  As discussed on page 5-91, in reference to
toluene, we believe these constituent reductions substantially diminish
the toxicity of the spent solvent wastes containing pyridine and
substantially reduce the likelihood of migration of pyridine from spent
solvent wastes.

    [The proposed technology-based BDAT treatment standard for pyridine
was estimated at the detection level of <0.500 mg/L based on biological
treatment (see Table 13, 51 FR 1725).  The principal difference between
the proposed and promulgated treatment standards is the Agency's change
in the criteria for data transfer.  (See Section 5.5.1, page 5-14, for a
discussion of the Agency's methodology for data transfer.)]
                                   5-83

-------
5.5.21   Tetrachloroethylene Wastewaters

    The Agency has biological treatment data for tetrachloroethylene at
plants 225 and 280 in the OCPSF data base.  Wet air oxidation treatment
data are also available for treatment of wastewater containing
tetrachloroethylene (Reference 10).  The Agency also has data from two
pilot-scale air strippers treating solvent spiked tap water (Reference 6,
Site 1) and industrial discharge contaminated groundwater (Reference 6,
Site 2), and from full-scale biological treatment of wastewater from
organic chemicals manufacturing (commercially available PACT® process,
Reference 4).  The data are summarized in Table 5-22 and calculation of
the BDAT treatment standard is shown in Table 5-23.

    The following steps were taken to derive the BDAT treatment standard
for tetrachloroethylene:

    1. We evaluated each data set to determine whether any of the data
       represent poor design or operation of the treatment systems.  In
       EPA's judgment, one data point in the data set for biological
       treatment at plant 225 represented poor design and operation.  We
       confirmed this judgment using the outlier test (refer to Table
       11-11, Appendix II).  The outlying data point was deleted.  Data
       for pilot-scale air stripping treatment at Sites 1 and 2 and for
       bench-scale wet air oxidation were deleted because, in
       consideration of the amount of full-scale data available for
       tetrachloroethylene, we believe it is appropriate to exclude
       pilot-scale data.

    2. We calculated the arithmetic average treatment concentration and
       the variability factor for each data set as shown in Table 5-23.
       Process variability could not be calculated for biological
       treatment at plant 280 because all effluent values were reported
       as less than or equal to the detection limit of 10 ug/L.  We would
       expect some variability in the data because the actual
       concentrations would range from 0 to the detection limit of
       10 ug/L.  To estimate the variability, the Agency used the average
       variability factor for BDAT biological treatment, 3.39.
       (Calculation of the average variability factor is shown in Table
       5-4, page 5-18.)

       Process variability could not be calculated for biological
       treatment by the PACT® process because there is only one data
       pair available from this process.  Therefore, the average
       variability factor for BDAT biological treatment, 3.39, was used.
       (Calculation of the average variability factor is shown in
       Table 5-4, page 5-18.)
                                      5-84

-------
    3. The analysis of variance method was not used to compare different
       treatments of the same waste because data are available for only
       one type of treatment for each waste.

    4. EPA then analyzed the data to determine if the various treatment
       concentration levels shown .in Table 5-23 could be associated with
       separate waste treatability subgroups.  Sufficient data did not
       exist to identify separate waste treatability subgroups;
       therefore, one waste treatability subgroup was established for all
       sources of wastewaters containing tetrachloroethylene spent
       solvents.  The least stringent treatment level within the
       treatability subgroup was selected for BDAT (0.079 mg/L from plant
       225) to ensure that the standard could be achieved for all waste
       matrices within the waste treatability subgroup.  The technology
       basis was biological treatment.

    5. The BDAT treatment standard for tetrachloroethylene represents
       treatment of a variety of waste matrices generated by process
       streams from the manufacture of over 17 different products.  The
       untreated waste concentration of tetrachloroethylene ranged from
       0.062 mg/L to 31.5 mg/L in these waste matrices.  All of these
       wastes were treated to the BDAT treatment standard or below (0.079
       mg/L).  We believe these constituent reductions substantially
       diminish the toxicity of the spent solvent wastes containing
       tetrachloroethylene and substantially reduce the likelihood of
       migration of tetrachloroethylene from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard for
tetrachloroethylene was <0.010 mg/L based on biological treatment (see
Table 13, 51 FR 1725).  The difference between the proposed and
promulgated treatment standards is primarily due to the incorporation of
a variability factor in derivation of the promulgated treatment
standard.  Other lesser factors affecting the change in the treatment
standard are the changes in data editing (data editing rules are
presented in Section 5.3) and deletion of some data points in the final
rule because they represent poor operation of the treatment system (see
the discussion of outlier test in Section 5.4).]
                                      5-85

-------
                            Table 5-22

        TREATMENT PERFORMANCE DATA FOR TETRACHLOROETHYLENE
     Plant 225                               Plant 225
Biological Treatment3                  Products Manufactured
Influent   Effluent
 (ug/L)     (ug/L)                Polyvinyl  chloride
                                  Perch!oroethylene
  2,251         10                Chlorinated paraffins
     95         10                Chlorine
    132         10                Hydrogen chloride
    482         19                Sodium methylate
    169         10
    186         10
    288         10
    913         10
  1,617         10
    374         10
    746         10
    714         10
    470         10
    252         12
    302         10
 17,500        476b>c
 31,500        150b
 24,000         55b

aThe data do not represent paired data (i.e., the samples were
 not collected so as to fully account for the retention time in
 the treatment system).
bln the data base from which this data was taken, the sampling
 data was designated using a different code,  plant 227, because
 it represented a different sampling episode.
cln EPA's judgment, this data point represented poor design and
 operation.   We confirmed this judgment using the outlier test
 (refer to Table 11-11,  Appendix II) and this data point was
 deleted.
                                  5-86

-------
                     Table 5-22   (Continued)

        TREATMENT PERFORMANCE DATA FOR TETRACHLOROETHYLENE
     Plant 280    .                           Plant 280
Biological Treatment3                  Products Manufactured
Influent   Effluent
 (ug/L)     (uq/L)                Adipic acid,  di(2-ethy1hexyl)
                                    ester
    413         10                Alkylphenols  (incl.  p-t-butyl)
    401         10                Fatty acid esters
    858         10                Phosphate esters, mixed triaryl
    998         10                Phosphate esters, tributyl
    405         10                Phosphate esters, tricresyl
    258         10                Phosphate esters, tris(b-
    110         10                  chloroalkyl)
  1,270         10                Phosphate esters, trixylenyl
  1,748         10                Phosphates, alkyl acid,  pyro-
    572         10                  phosphates  & salts
    729         10                Phosphonates, diethyl  bis(2-
    399         10                hydroxyethyl) aminomethyl

aThe data do not represent paired data (i.e., the samples  were
 not collected so as to fully account  for the retention  time in
 the treatment system).
                                  5-87

-------
                      Table 5-22 (Continued)

        TREATMENT PERFORMANCE DATA FOR TETRACHLOROETHYLENE
 D.G. Mutton, 1979.
Biological Treatment3-13
Influent   Effluent
 (ua/L)     (ua/L)
     62
7.3
                        Description of
                        Waste Treated

                 Wastewater from organic chemi-
                 cal manufacturing.



Love and Eilers, 1982
Pilot-Scale Air Striooer. Site 1
Average
Influent
Concentration
(ug/L)

1 ,025
636
338
114
107

Average Effluent Concentration at
Various Air-to-Water Ratios (ug/L)
1:1 2:1 3:1 4:1 8:1 16:1 20:1

698 416 304 156 16
161 177 46 34 8 <1 <1
139 103 47 34 4 1 2
32 17 7 4 <1 <1 <1
32 17 7 4 <1 <1 <1
Description
of
Waste Treated

Tap water was
spiked with
tetrachloro-
ethylene and
trichloro-
ethylene.



   Love and Eilers, 1982
      Pilot-Scale Air
     Stripper. Site 2	
                 Average
                 Effluent
                  Concen-
  Average       tration for
  Influent      4:1 Air-to-
Concentration   Water Ratio
   (uo/L)         (ug/L)

     94              9
                        Description of
                        Waste Treated

                 Ground water was contaminated
                 by industrial discharge; pilot-
                 scale column was run continu-
                 ously for over one year.
aThe data do not represent paired data (i.e., the samples were
 not collected so as to fully account for the retention time in
 the treatment system).
''Commercially available patented PACT® process.
                                  5-1

-------
                 Table 5-22 (Continued)

   TREATMENT PERFORMANCE DATA FOR TETRACHLOROETHYLENE
Data Submitted by Zimpro,
       Inc.,  1986                        Description of
   Met Air Oxidation                     Waste  Treated
              Oxidation
 Raw Waste     Product            General organic.
   (ug/L)      (uo/L)

    41,000      <1,000
                                  5-89

-------
                           Table 5-23

           CALCULATION OF BOAT FOR TETRACHLOROETHYLENE
Plant
 No.     Technology

 225     Biological

 280     Biological

  I*3     Biological
                         Average
                        Treatment                 Treatment Con-
                      Concentration Variability  centration Level
                         (ug/L)	Factor     Avg. x VF (uo/L)
                           21.5

                           10

                            7.3
3.65

3.39a

3.39a
79

34

25
aAverage variability factor for all BOAT biological treatment
 data (see Table 5-4 and the discussion on page 5-17).
bCoirniercially available patented PACT® process.
                                5-90

-------
5.5.22   Toluene Wastewaters

    The Agency has biological treatment data for toluene at plants 202,
206, 208, 210, 211, 215, 217, 221, 223, 230, 234, 240, 242, 244,  246,
251, 257, 265, and 286 in the OCPSF data base.   Also available from the
OCPSF data base are data for biological treatment followed by activated
carbon adsorption at plant 246, steam stripping data at plant 246, and
data for steam stripping followed by activated carbon adsorption at plant
297.  The Agency also has data from pilot-scale steam stripping and air
stripping of solvent contaminated groundwater (Reference 2), full-scale
biological treatment of wastewater from organic chemicals manufacturing
(commercially available patented PACT® process, Reference 4), and
pilot-scale activated carbon adsorption of runoff water from a waste
disposal site's containment dikes (Reference 7).  The Agency also has
data from the Iron and Steel Manufacturing Development Document
(Reference 9) and wet air oxidation data submitted by Zimpro, Inc.
(Reference 10).  The data are summarized in Table 5-24 and calculation of
the BOAT treatment standard is shown in Table 5-25.

    The following steps were taken to derive the BOAT treatment standard
for toluene:

    1. We evaluated each data set to determine whether any of the data
       represent poor design or operation of the treatment systems.  In
       EPA's judgment, two data points in the data set for biological
       treatment at plant 234 represented poor design and operation.  We
       confirmed this judgment using the outlier test (refer to Table
       11-12, Appendix II,).  The outlying data point was deleted.

       Data for biological treatment at plant 253 (consisting of  three
       data points) were deleted on the basis of poor design and
       performance.  Based on the disproportionately low removals
       relative to other treatment systems for wastes containing  toluene,
       EPA judged this system to be poorly designed and operated.  This
       system achieved a reduction of only 34.6 percent as compared with
       86.3 to 99.9 percent for other systems treating wastes containing
       toluene.

       Individual paired data points for steam stripping treatment at
       plant 246 were deleted when the influent concentrations were below
       the 0.05 mg/L quantification level for toluene.

       Data for biological treatment at plant 206 were deleted because
       the treatment system at this plant was shown to be poorly  designed
       and/or operated based on the wide variation in influent
       concentrations.  The nature of biological treatment systems
       requires sufficient control of influent concentrations through the
       use of equalization to prevent "shock loading" of the biomass.
                                      5-91

-------
   Data on pilot-scale steam stripping, pilot-scale air stripping,
   pilot-scale carbon adsorption, and bench-scale wet air oxidation
   treatment were deleted because, in consideration of the amount of
   full-scale data available for toluene, we believe it is
   appropriate to exclude pilot-scale data.   The data from the Iron
   and Steel Manufacturing Development Document were not used because
   insufficient information exists in some cases to determine the
   concentrations treated and, in other cases, which technology was
   achieving removal.  Also, in some cases,  the treated values
   represented significant dilution of the wastestream.

2. We calculated the arithmetic average treatment concentration and
   the variability factor for each data set  as shown in Table 5-25.
   Process variability could not be calculated for biological
   treatment at plant 217 because there is an insufficient number of
   data points available from this process to allow a meaningful
   estimation of process variability for the plant.  Therefore, the
   average variability factor for BOAT biological treatment,  3.39,
   was used (calculation of the average variability factor is shown
   in Table 5-4, page 5-18).

   Process variability could not be calculated for biological
   treatment at plants 202, 208, 210, 211, 215, 221, 223, 230, 240,
   242, 244, 251, and 265 because all effluent values were reported
   as less than or equal to the detection limit of 10 ug/L.   We would
   expect some variability in the data because the actual
   concentrations would range from 0 to the  detection limit of 10
   ug/L.  To estimate the variability, the Agency used the average
   variability factor for BOAT biological treatment, 3.39.
   (Calculation of the average variability factor is shown in Table
   5-4, page 5-18.}

   Process variability could not be calculated for full-scale
   biological treatment of wastewater from organic chemicals
   manufacturing (the Zimpro PACT® process)  because there was only
   one data point available for this process.  Therefore, the average
   variability factor for BOAT biological treatment, 3.39, was used
   (calculation of the average variability factor is shown in
   Table 5-4, page 5-18).

3. Biological treatment and biological treatment followed by
   activated carbon adsorption of toluene at plant 246 were compared
   with the analysis of variance method to determine whether the
   performance of one technology was significantly better than the
   other for treatment of the same waste.  It was shown that the
   addition of activated carbon adsorption to biological treatment
                                  5-92

-------
       significantly improved treatment performance.  Therefore, the
       treatment concentration level for plant 246 is 1.12 mg/L based
       upon biological treatment followed by activated carbon
       adsorption.  (Refer to the statistical calculations and results in
       Table 11-13, Appendix II.)  The analysis of variance method could
       not be used to compare treatments on other wastes because data
       were not available for more than one treatment for any other waste.

    4. EPA then analyzed the data to determine if the various treatment
       concentration levels shown in Table 5-25 could be associated with
       separate waste treatability subgroups.  Sufficient data did not
       exist to identify separate waste treatability subgroups;
       therefore, one waste treatability subgroup was established for all
       sources of wastewaters containing toluene spent solvents.  The
       least stringent treatment level within the treatability subgroup
       was selected for BOAT (1.12 mg/L from plant 246) to ensure that
       the standard could be achieved for all waste matrices within the
       waste treatability subgroup.  The technology basis was biological
       treatment followed by activated carbon adsorption.

    5. The BOAT treatment standard for toluene represents treatment of a
       variety of waste matrices generated by process streams from the
       manufacture of 150 different products.  The untreated waste
       concentration of toluene ranged from 0.010 mg/L to 160 mg/L in
       these waste matrices.  All of these wastes were treated to the
       BDAT treatment standard or below (1.12 mg/L).  We believe these
       constituent reductions substantially diminish the toxicity of the
       spent solvent wastes containing toluene and substantially reduce
       the likelihood of migration of toluene from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard for toluene
was 0.016 mg/L based on activated carbon adsorption (see Table 13, 51 PR
1725).  The difference between the proposed and promulgated treatment
standards is primarily due to the incorporation of a variability factor
in derivation of the promulgated treatment standard.  Other less
significant factors affecting the change in the treatment standard are
the changes in data editing (data editing rules are presented in
Section 5.3) and deletion of some data points in the final rule because
they represent poor operation of the treatment system (see the discussion
of the outlier test in Section 5.4).]
                                      5-93

-------
                            Table 5-24

              TREATMENT PERFORMANCE DATA FOR TOLUENE
     Plant 202
Biological Treatment3
Influent
(ug/L)
122
130
144
126
107
139
154
150
155
148
81
95
95
73
138
94
107
87
67
60
Plant
Bioloaical
Influent
(ua/L)
3,486
19,707
17,697
4,001
57,475
8,327
49,379
834
14,877
24,264
Effluent
(ug/U
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
206
Treatment3
Effluent
(ua/U
231
212
220
7,411
320
6,087
14
344
17
52
           Plant 202
       Products Manufactured

Disperse dye coupler
Disperse dyes
Naphthalene sulfonic acid
Organic pigments
p-Phenylene diamine
Sulfur dyes
Vat dyes
Xylenesulfonic acid, sodium
  salt
2-Bromo-4,6-dinitroaniline
2,4-Dinitroaniline
2,4-Dinitrochlorobenzene
2,4-Dinitrophenol
2,4,6-Trinitrophenol
4-Chloro-2,6-dinitrobenzene
  sulfonic acid, potassium salt
                                             Plant 206
                                       Products Manufactured

                                  3,3-Dichlorobenzidine
                                  Polyurethane resins
                                  Orthochloroaniline
                                  Benzophenone
                                  2-Sulfophthalic acid
                                  2,6-Dichloronitroaniline
aThe data do not represent paired data (i.e., the samples were
 not collected so as to fully account for the retention time in
 the treatment system).
                                   5-94

-------
                      Table 5-24 (Continued)

              TREATMENT PERFORMANCE DATA FOR TOLUENE
     Plant 208
Biological Treatment3
Influent   Effluent
 (uq/L)     (ug/L)
    370
    285
    251
    283
    140
    345
    514
    587
    472
    449
    635
    724
    593
    640
10
10
10
10
10
10
10
10
10
10
10
10
10
10
           Plant 208
     Products Manufactured

Cyclic (coal tar) intermediates
Tar, tar crudes, and tar
  pitches
     Plant 210
Biological Treatment3
Influent   Effluent
 (ug/L)     (ug/L)
  5,805
    135
10
10b
           Plant 210
     Products Manufactured

Acetic acid
Acetone cyanohydrin
Acrylic acid
Acrylic acid esters
Acrylic resins, oil additives
Alkyl amines
Ethoxylates
Methacrylic acid esters
Methyl  methacrylate
Alkyl phenols
Acetylene
Methacrylic acid
aThe data do not represent paired data (i.e.,  the samples were
 not collected so as to fully account for the  retention time in
 the treatment system).
^In the data base from which this data was taken, the sampling
 data was designated using a different code,  plant 282, because
 it represented a different sampling episode.
                                   5-95

-------
                     Table 5-24 (Continued)

             TREATMENT PERFORMANCE DATA FOR TOLUENE
     Plant 211
Biological Treatment3
Influent   Effluent
 (ug/L)     (ua/L)
  4,000
  2,082
  3,024
  1,220
  1,546
  1,154
  1,315
 10
 10
 10
 10
 10
 10
 10
           Plant 211
     Products Manufactured

Coal tar solvent
Coatings
Cresols (mixed)
Ethyl benzene
Methyl naphthalene
Naphthalene
Pitch tar residue
Pyridines (tar bases)
2.4-Xylenol (dimethyl phenol)
Phenol
     Plant 215
Biological Treatment3
Influent   Effluent
 (ug/L)     (ug/L)
  4,550
  3,300
  3,700
 10
 10
 10
           Plant 215
     Products Manufactured

Benzene
Toluene
Xylenes (mixed)
Cyclohexane
Isobutylene
Propylene
Polypropylene
Butyl rubber
Paraffins
     Plant 217
Biological Treatment3
Influent   Effluent
 (UQ/l)     (UQ/L)
 60,000
 47,300
 34,400
 10
108
102
           Plant 217
     Products Manufactured

Phthalic anhydride
Butyl benzyl phthalate
Benzyl chloride
Tetrachlorophthalic anhydride
Phosphate esters
Phthalate esters
Polybenzyl ethyl benzene
aThe data do not represent paired data (i.e., the samples were
 not collected so as to fully account for the retention time in
 the treatment system).
                                   5-96

-------
                     Table 5-24 (Continued)

             TREATMENT PERFORMANCE DATA FOR TOLUENE
     Plant 221                               Plant 221
Biological Treatment3                  Products Manufactured
Influent   Effluent
 (ua/Ll     (uQ/U                Di-isodecyl phthalate ester
                                  Ethylene
    323         10                Propylene
    190         10                Isopropanol
     10         10                Petroleum hydrocarbon resins
                                  1,3-Butadiene
                                  Butylenes
                                  Cyclopentadiene dimer
                                  Isobutylene
                                  Isoprene

     Plant 223                               Plant 223
Biological Treatment3                  Products Manufactured
Influent   Effluent
 (ug/U     (ug/U                Acrylic acid esters
                                  Caprolactam
    265         10                Cyclohexanone
    179         10                Isobutanol
     99         10                n-Butyl  alcohol
                                  2-Ethyl  hexanol

aThe data do not represent paired data  (i.e.,  the  samples were
 not collected so as to fully account for the retention time in
 the treatment system).
                                  5-97

-------
                     Table 5-24 (Continued)

             TREATMENT PERFORMANCE DATA FOR  TOLUENE
     Plant 230                               Plant  230
Biological Treatmenta                  Products  Manufactured
Influent   Effluent
 (ug/L)     (ug/U                Benzene
                                  Ethylene
 15,891         10                Hydrogen
  4,649         10                Propylene
  4,904         10                Pyrolysis gasoline
 20,065         10                Polyethylene resin
  4,534         10                Polypropylene
 19,848         10                Polypropylene  resin
  3,867         10                1,3-Butadiene
 30,347         10                Butylenes
  4,426         10
  3,806         10
  3,942         10
  3,538         10
  3,882         10
  3,789         10
  3,503         10

aThe data do not represent  paired data (I.e.,  the  samples  were
 not collected so as to fully account for the  retention  time  in
 the treatment system).
                                   5-98

-------
                      Table 5-24 (Continued)

              TREATMENT PERFORMANCE DATA FOR TOLUENE
     Plant 234
Biological Treatment8
Influent   Effluent
 (UQ/Ll     (UQ/L)
 16,000
  6,200
  3,750
  7,600
  8.700
  5,750
  6,800
  6,500
  9,800
  5,300
  4,100
 26,000
  3,825
  3,400
  6,450
 13,000
 35,000
  5,950
  5,310
  5,476
 11,060
  4,700
  2,350
  4,806
  6,650
 15,000
 32,500
  7,700
 25,750
  6,150
 16,000
  4,487
 10
 10
 10
 10
 10
 10
 10
 10
 10
 10
 10
 21
 19
 37
 10
 10
 10
 10
 10
 11
 12
 15
 10
 10
 67"
 10
235b
 12
 10
 10
 10
 10
                                             Plant 234
                                       Products Manufactured

                                  Acetic acid
                                  Acetic anhydride
                                  Acetone
                                  Acetaldehyde
                                  Propionic acid
                                  PET resins/fibers
                                  Acetoacetanil ide
                                  Terephthalic acid
                                  n-Propyl acetate
                                  Diethyl phthalate
                                  Dimethyl phthalate
                                  di-n-Butyl phthalate
                                  Bis(2-ethylhexylJphthalate
                                  Methyl isobutyl ketone
                                  Isopropoacetate
                                  Isobutyl acetate
                                  Hydroquinone
aThe data do not represent paired data (i.e.,  the samples were
 not collected so as to fully account for the  retention time in
 the treatment system).
^In EPA's judgment,  this data point represented poor design
 and operation.   We  confirmed this judgment using the outlier
 test (refer to  Table 11-12, Appendix II) and  this data point was
 deleted.
                                   5-99

-------
                     Table 5-24 (Continued)

             TREATMENT PERFORMANCE DATA FOR  TOLUENE
     Plant 240
Biological Treatment8
Influent   Effluent
 (UQ/L1     (uq/L)  Acetic acid
 22,700
10
                             Plant  240
                       Products  Manufactured
Acetylene
Acrolein
Acrylic acid esters
Benzene
Cyclohexanone
Diethylene glycol
Epoxidized esters
Ethylamines (mono, di,  tri)
Ethylene
Ethylene dimer
Ethylene glycol
Ethylene glycol monomethyl
  ether
Ethylene oxide
Isopropyl amines (mono, di)
Peracetic acid
Polyethylene glycol
Polyethylene polyamines
Propylene
Toluene
1,2-Dichloroethane
Butylenes
Xylenes (mixed)
     Plant 242
Biological Treatment3
Influent   Effluent
 (UQ/L)     (Ud/U
  1,533
  1,200
10
10
           Plant 242
     Products Manufactured

Alkyd resins
Epoxy resins
Glyoxal-urea formaldehyde tex-
  tile resin
Unsaturated polyester resins
Acrylic resins
Melamine resins
Urea resins
aThe data do not represent paired data (i.e., the samples were
 not collected so as to fully account for the retention time in
 the treatment system).
                                   5-100

-------
                      Table 5-24 (Continued)

              TREATMENT PERFORMANCE DATA FOR TOLUENE
     Plant 244                               Plant 244
Biological Treatment3                  Products Manufactured
Influent   Effluent
 (ug/L)     (ug/L)                Cyclohexanol
                                  C4 hydrocarbons
  1,109         10                Ethylene
                                  Ethylene-methacrylic acid
                                    copolymer
                                  Polyethylene polyvinyl acetate
                                    copolymers
                                  Propylene
                                  Hexamethylenedlamine
                                  Polyethylene resins
                                  Adiponitrile
     Plant 246                               Plant 246
  Steam Stripping                      Products Manufactured
Influent   Effluent
 (ug/L)     (ug/L)                Aniline
                                  Dinitrotoluene (mixed)
     98         12                Methylene diphenyl  diisocyanate
     80         10                Nitrobenzene
     57         10                Polymeric methylene diphenyl
     72         10                  diisocyanate
                                  Polyoxypropylene glycol
                                  Toluene diamine (mixture)
                                  Toluene diisocyanates (mixture)
                                  Polymeric methylene dianiline
                                  Polyurethane resins
aThe data do not represent paired data (i.e.,  the samples  were
 not collected so as to fully account for  the  retention  time  in
 the treatment system).
                                   5-101

-------
                      Table 5-24 (Continued)

              TREATMENT PERFORMANCE DATA FOR TOLUENE
         Plant 246
    Biological Treatment3
        Followed by
Activated Carbon Adsorption
     Influent   Effluent
      (ua/L)     (UQ/D
       4,372
          77
       4,881
       2,273
         244
      12,938
       4,166
      10,375
      12,864
         180
       5,397
       1,371
       3,899
       5,400
       5,500
       6,575
 98
  b
  b
  b
 53
 30
 91
330
437
 21
  b
  b
  b
 50C
 10C
 10C
           Plant 246
     Products Manufactured

Aniline
Dinitrotoluene (mixed)
Methylene diphenyl diisocyanate
Nitrobenzene
Polymeric methylene diphenyl
  diisocyanate
Polyoxypropylene glycol
Toluene diamine (mixture)
Toluene diisocyanates (mixture)
Polymeric methylene dianiline
Polyurethane resins
aThe data do not represent paired data (i.e.,  the samples were
 not collected so as to fully account for the  retention time in
 the treatment system).
°The treatment effluent was not sampled on this sampling day.
cln the data base from which this data was taken, the sampling
 data was designated using a different code,  plant 219, because
 it represented a different sampling episode.
                                   5-102

-------
                      Table 5-24 (Continued)

              TREATMENT PERFORMANCE DATA FOR  TOLUENE
        Plant 246                            Plant 246
   Biological Treatment3               Products Manufactured
     Influent   Effluent
      (ug/L)     (ug/L)           Aniline
                                  Dinitrotoluene (mixed)
       4,372        736           Methylene diphenyl  diisocyanate
          77        168           Nitrobenzene
       4,881         14           Polymeric methylene diphenyl
       2,273        308             diisocyanate
         244          b           Polyoxypropylene glycol
      12,938         94           Toluene diamine (mixture)
       4,166        661           Toluene diisocyanates  (mixture)
      10,375      1,453           Polymeric methylene dianiline
      12,864          b           Polyurethane resins
         180      2,136
       4,308          b
       5,397        102
       1,371          b
       3,899          b

aThe data do not represent  paired data  (i.e.,  the samples  were
 not collected so as to fully account for the  retention  time in
 the treatment system).
bThe treatment effluent was not  sampled on this sampling day.
                                   5-103

-------
                      Table 5-24 (Continued)

              TREATMENT PERFORMANCE DATA FOR  TOLUENE
     Plant 251
Biological Treatment3
Influent   Effluent
 (ua/Ll     (ug/U
 15,840
 26,060
 21,700
 10
 10
 10
           Plant 251
     Products Manufactured

Acetone
Acetonitrile
Acrylonitrile
Benzene
Butylenes (mixed)
Dialkylbenzene, by-product
Oiphenyl oxide (diphenyl ether)
Ethane
Ethyl benzene
Ethylene
Formaldehyde
Iminodiacetic acid
Naphthalene
Nitrilotriacetic acid
o-Xylene
Phenol
Propylene
Resin tars
Sorbic acid, salts
Toluene
1,3-Pentadiene (piperylene)
Phenolic resins
Cumene
1,3-Butadiene
Cyclopentadiene dimer
Isoprene
Xylenes (mixed)
     Plant 253
Biological Treatment3
Influent   Effluent
 (UQ/D     (ua/L)
    175
    230
     66
 38
140
130
                              Plant  253
                        Products Manufactured

                   Polypropylene resins
aThe data do not represent paired data (i.e.,  the  samples  were
 not collected so as to fully account  for  the  retention  time  in
 the treatment system).
                                   5-104

-------
                     Table 5-24 (Continued)

             TREATMENT PERFORMANCE DATA FOR  TOLUENE
     Plant 257
Biological Treatmenta
Influent   Effluent
 (ua/U     (ua/U
  1,330
  1,800
  2,090
  1 ,730
  3,365
  3,720
  3,746
   ,660
   ,964
   ,482
   ,040
   ,510
  4,933
  4,665
  4,707
  3,836
  3,160
  2,627
    450
    684
    600
  4,121
  5,290
  4,985
  7,417
 12,900
 11,400
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
16b
45b
10b
           Plant 257
     Products Manufactured

Acetone
Ally!  chloride
Bisphenol-A
Butylenes (mixed)
Diacetone alcohol
Ethylene
Isobutylene
Phenol
Propylene
Vinyl  chloride
Epichlorohydrin
Acetone
Epoxy resins
Isopropanol
Methyl ethyl ketone
Methyl isobutyl ketone
n-Butyl alcohol
Cumene
Ethanol
sec-Butyl alcohol
Butadiene
Isoprene
aThe data do not represent paired data (i.e.. the samples were
 not collected so as to fully account for the retention time in
 the treatment system).
bln the data base from which this data was taken, the sampling
 data was designated using a different code, plant 259, because
 it represented a different sampling episode.
                                    5-105

-------
                     Table 5-24 (Continued)

             TREATMENT PERFORMANCE DATA FOR  TOLUENE
     Plant 265                               Plant 265
Biological Treatment3                  Products Manufactured
Influent   Effluent
 (ug/L)     (ug/L)                Tar,  tar crudes, and tar
                                    pitches
 37,750         10
 44,000         10
 50,000         10
     Plant 286                               Plant 286
Biological Treatment3                  Products Manufactured
Influent   Effluent
 (ug/L)     (ug/L)                Formaldehyde
                                  Phenolic resins
160,000         38                Urea resins
 52,000         80
 24,000        110
     Plant 297
  Steam Stripping
Followed by Activated                        Plant 297
  Carbon Adsorption                    Products Manufactured
Influent   Effluent
 (ug/L)     (ug/L)                Nitrobenzene
                                  Nitrotoluene
  8,650         10b                 Aniline
    640         I4b                 Toluidine
    750         10b
aThe data do not represent paired data (i.e., the samples were
 not collected so as to fully account for the retention time in
 the treatment system).
bln the data base from which this data was taken, the sampling
 data was designated using a different code, plant 248, because
 it represented a different sampling episode.
                                    5-106

-------
                     Table 5-24 (Continued)

             TREATMENT PERFORMANCE DATA FOR TOLUENE
  Stover and Kincannon, 1.983
  Pilot-Scale Steam Stripper
     Influent   Effluent
      (uq/L)     (uq/L)
      92,000
      92.000
      92.000
      92.000
      92,000
    126
     10
     10
     10
     S3
       Description of
       Waste Treated

Pilot-scale study of ground-
water near a waste disposal
dump site which contained
household refuse, demolition
materials, chemical sludges,
and hazardous liquid chemicals
  D.G. Hutton, 1979
Biological Treatmenta-b
Influent   Effluent
 (uq/L)     (uq/L)
  680
             4.1
                         Description of
                         Waste Treated

                  Wastewater from organic chemi-
                  cals manufacturing.
  Stover and Kincannon, 1983
    Pilot-Scale Air Stripper
     Influent   Effluent
      (UQ/L)     (uq/L)
      92,000
      92,000
      92.000
      92,000
      92,000
      92.000
 30.000
 23,300
 19,000
 17,100
  6,600
 44,800
       Description of
       Waste Treated

Pilot-scale study of ground-
water near a waste disposal
dump site which contained
household refuse, demolition
materials, chemical sludges,
and hazardous liquid chemicals
   Becker and Wilson,  1978
Pilot-Scale Granular Activated
	Carbon Column	
     Influent
      (ug/L)

         120
Effluent
 (ug/L)

    0.3
       Description of
       Waste Treated

Runoff water from a waste dis-
posal site's containment dikes
aThe data do not represent paired data (i.e.,  the samples  were
 not collected so as to fully account for the  retention  time in
 the treatment system).
^Commercially available patented PACT® process.

-------
                      Table 5-24 (Continued)

              TREATMENT PERFORMANCE DATA FOR TOLUENE
Iron and Steel Manufacturing
 Development Document, 1980
Multiple Treatment Technologies
Average         Average
Influent        Effluent
 (ug/U          (ug/L)        Plant
                              Description of
                              ijaste Treated
  8,920
  5,450
  6,130
    40
    73
      003        Excess ammonia liquor
                 and miscellaneous
                 wastewaters
      008        Excess ammonia liquor
                 and benzol plant
                 wastewaters
      009        Excess ammonia liquor
                 and benzol plant
                 wastewaters
Data Submitted by Zimpro,  Inc.,  1986
	Wet Air Oxidation	
 Raw Waste
   (ug/L)
Diluted
 Feed
 (ug/L)
Oxidation
 Product
 (uq/L)
Description of
Waste Treated
 34,100,000   8,500,000
            200,000
               General  organic
                             5-108

-------
                           Table 5-25

                 CALCULATION OF BOAT FOR TOLUENE

                         Average
                        Treatment                 Treatment Con-
                                                 centration Level
                                                 Avo.  x VF (ua/Ll

                                                        13
                                                        34
                                                        34
                                                        34
                                                        34
                                                        34
                                                       248
                                                        34
                                                        34
                                                        34
                                                        22
                                                        34
                                                        34
                                                        34
                                                    11,100
                                                        34
                                                        22
                                                        34
                                                       247
                                                     1,118
                                                        18
Plant
246
202
208
210
211
215
217
221
223
230
234
240
242
244
246
251
257
265
286
246



297


l"
Concentration Variability
Technology (ua/L) Factor
Steam Stripping
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological
Biological fol-
lowed by
Activated
Carbon
Steam Stripping
followed by
Activated Carbon
Biological
10.5
10
10
10
10
10
73.3
10
10
10
11.9
10
10
10
630
10
11.6
10
76
113



11.3


4.1
1.23
3.39a
3.39a
3.39a
3.39a
3.39a
3.39a
3.39a
3.39a
3.39a
1.87
3.39a
3.39a
3.39a
17.61
3.39a
1.89
3.39a
3.25
9.89



1.55


3.39a
                                                      14
aAverage variability factor for BOAT biological  treatment  data
 (see Table 5-4 and the discussion on page 5-17).
Commercially available patented PACT® process.
                                5-109

-------
5.5.23   1,1,1-Trichloroethane Wastewaters

    The Agency has full-scale biological treatment data for
1,1,1-trichloroethane from plant 240 in the OCPSF data base.   The Agency
also has data from pilot-scale steam stripping and pilot-scale air
stripping of solvent contaminated groundwater (Reference 2),  pilot-scale
air stripping of industrial discharge contaminated groundwater (Reference
6), and bench-scale wet air oxidation of a general organic waste
(Reference 10).  The data are summarized in Table 5-26 and calculation of
the BDAT treatment standard is shown in Table 5-27.

    The following steps were taken to derive the BDAT treatment standard
for 1,1,1-trichloroethane:

    1. We evaluated each data set to determine whether any of the data
       represent poor design or operation of the treatment systems.  Data
       from pilot-scale wet air oxidation treatment were deleted because
       the concentration of 1,1,1-trichloroethane in the diluted feed to
       the wet air oxidation process was not reported and because the
       detection limit of 1,000 ug/L for 1,1,1-trichloroethane in the
       oxidation product is too high for the data to be meaningful with
       regard to how well the system will perform.

       Data on pilot-scale steam stripping and pilot-scale air stripping
       were not deleted because, in consideration of the amount of
       full-scale treatment data available for this constituent, we
       believe that it is appropriate to include this pilot-scale data in
       derivation of the BDAT treatment standard for 1,1,1-trichloroethane.

    2. We calculated the arithmetic average treatment concentration and
       the variability factor for each data set as shown in Table 5-27,
       Process variability could not be calculated for biological
       treatment at plant 240 because all effluent values were reported
       as less than or equal to the detection limit of 10 ug/L.  We would
       expect some variability in the data because the actual
       concentrations would range from 0 to the detection limit of 10
       ug/L.  To estimate the variability, the Agency used the average
       variability factor for BDAT biological treatment, 3.39.
       (Calculation of the average variability factor is shown in Table
       5-4, page 5-18.)

       To account for full-scale process variability in the pilot-scale
       steam stripping data, the average variability factor for BDAT
       full-scale steam stripping, 2.26, was used.  (Calculation of the
       average variability factor is shown in Table 5-4, page 5-18.)
                                      5-110

-------
       Process variability could not be calculated for the pilot-scale
       air stripper at Site 2 because there is only one data pair
       available from this process.   Therefore, the average variability
       factor for all BOAT wastewater treatment, 3.56, was used.
       (Calculation of the average variability factor is shown in Table
       5-4, page 5-18.)

    3. Steam stripping and air stripping of 1,1,1-trichloroethane at the
       pilot-plant (Reference 2) were compared with the analysis  of
       variance method to determine whether the performance of one
       technology was significantly better than the other for treatment
       of the same waste.  It was shown that steam stripping provided
       significantly better removals of 1,1,1-trichloroethane compared
       with air stripping (refer to Table 11-15, Appendix II).
       Therefore, the treatment standard for the pilot-plant is 1,046
       ug/L based upon steam stripping.

    4. EPA then analyzed the data to determine if the various treatment
       concentration levels shown in Table 5-26 could be associated with
       separate waste treatability subgroups.   Sufficient data did not
       exist to identify separate waste treatability subgroups;
       therefore, one waste treatability subgroup was established for all
       sources of wastewaters containing 1,1,1-trichloroethane spent
       solvents.  The least stringent treatment level within the
       treatability subgroup was selected for BOAT (1.05 mg/L from
       pilot-scale steam stripping)  to ensure that the standard could be
       achieved for all waste matrices within the waste treatability
       subgroup.

    5. The BDAT treatment standard for 1,1,1-trichloroethane represents
       treatment of a variety of waste matrices generated by process
       streams from the manufacture of over 28 different products.  The
       untreated waste concentration of 1,1,1-trichloroethane ranged from
       0.010 mg/L to 150 mg/L in these waste matrices.  All of these
       wastes were treated to the BDAT treatment standard or below (1.05
       mg/L).  We believe these constituent reductions substantially
       diminish the toxicity of the spent solvent wastes containing
       1,1,1-trichloroethane and substantially reduce the likelihood of
       migration of 1,1,1-trichloroethane from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard for
1,1,1-trichloroethane was 0.457 based on steam stripping (see Table 13,
51 PR 1725).  The difference between the proposed and promulgated
treatment standards is primarily due to the incorporation of a
variability factor in derivation of the promulgated treatment standard.
Other less significant factors affecting the change in the treatment
standard are the changes in data editing (data editing rules are
presented in Section 5.3).]
                                     5-111

-------
                            Table 5-26

       TREATMENT PERFORMANCE DATA FOR 1 ,1 ,1-TRICHLOROETHANE
Stover and Kincannon, 1981
Pilot-Scale Steam Stripper
   Influent   Effluent
    (ug/U     (ug/L)
   150,000
   150,000
   150,000
   150,000
   150,000
      10
      10
     150
   2,135
      10
       Description of
       Waste Treated

Pilot-scale study of ground-
water near a waste disposal
dump site which contained
household refuse, demolition
materials, chemical sludges,
and hazardous liquid chemicals.
      Plant 240
Biological Treatment3
Influent   Effluent
 (mq/U     (mq/L)
   215
    10
    95
10
10
10
           Plant 240
     Products Manufactured

Acetic acid
Acetylene
Acrolein
Acrylic acid esters
Benzene
Cyclohexanone
Diethylene glycol
Epoxidized esters
Ethylamines (mono,  di, tri)
Ethylene
Ethylene dimer
Ethylene glycol
Ethylene glycol monomethyl
  ether
Ethylene oxide
Isopropyl amines (mono, di)
Peracetic acid
Polyethylene glycol
Polyethylene polyamines
Propylene
Toluene
1,2-Dichloroethane
Butylenes
Xylenes (mixed)
aThe data do not represent paired data (i.e.,  the samples  were
 not collected so as to fully account for the  retention time in
 the treatment system).
                                   5-112

-------
                      Table 5-26 (Continued)

       TREATMENT PERFORMANCE DATA FOR 1,1,1-TRICHLOROETHANE
Stover and Kincannon,  1983.
  Pilot-Scale Air Stripper
   Influent   Effluent
    (ug/L)     (ug/U
   150,000
   150,000
   150,000
   150,000
   150,000
   150,000
 53,000
 66,000
 60,000
 39,200
  7,600
 66,300
       Description of
       Waste Treated

Pilot-scale study of ground-
water near a waste disposal
dump site which contained
household refuse, demolition
materials, chemical sludges,
and hazardous liquid chemicals.
               Love and Eilers,  1982
         Pilot-Scale Air Stripper.  Site 2
  Average
  Influent
Concentration
   (ug/L)

      237
 Average Effluent Concentration for
   4:1 Air-to-water Ratio (ug/L)
                 23
Data Submitted by Zimpro,
       Inc.,  1986
   Wet Air Oxidation
 Raw Waste
   (ug/L)

   370,000
Oxidation
 Product
 (ug/L)

  <1,000
                   Description
                       of
                  Waste Treated

                  Ground water
                  was contami-
                  nated by
                  industrial
                  discharge;
                  pilot-scale
                  column was
                  run continu-
                  ously for
                  over one
                  year.
       Description of
       Waste Treated
General  organic.
                                   5-113

-------
                           Table 5-27

          CALCULATION OF BOAT FOR  1 ,1 ,1-TRICHLOROETHANE


                         Average
                        Treatment                 Treatment Con-
Plant                 Concentration Variability  centration Level
 No.      Technology     (ug/L)	Factor     Avg. x VF (ug/Ll

 PS     Air Stripping    48,683        5.80           282,361

 PS     Steam Stripping     463        2.26a            1,046

 240    Biological           10        3.39b               34

 PS,2   Air Stripping        23        3.56C               82
aAverage variability factor for BOAT full-scale steam stripping
 (see Table 5-4 and the discussion on page 5-17).
''Average variability factor for BOAT biological treatment (see
 Table 5-4 and the discussion on page 5-17).
aAverage variability factor for all  BOAT wastewater treatment
 (see Table 5-4 and the discussion on page 5-17).
                                5-114

-------
5.5.24   1,1,2-Trichloro-l,2,2-trifluoroethane Wastewaters

    The Agency has no data for wastewater treatment for the removal of
1,1,2-trichloro-l,2,2-trifluoroethane.  For reasons presented in
Section 5.5.1, EPA used chemical structure as the basis for transferring
treatment data to 1,1,2-trichloro-l,2,2-trifluoroethane spent solvent
wastewaters.  Specifically, we transferred the treatment data from
1,1,1-trichloroethane because, like 1,1,2-trichloro-l,2,2-trifluoro-
ethane, 1,1,1-trichloroethane contains the halogen functional group.
1,1,1-trichloroethane had the least stringent treatment standard in the
halogenated aliphatics structural group.  Using performance data from
1,1,1-trichloroethane, the BOAT treatment standard for
1,1,2-trichloro-l,2,2-trifluoroethane is 1.05 mg/L.  The technology basis
for this treatment is steam stripping.

    We believe the BDAT treatment standard for 1,1,2-trichloro-
1,2,2-trifluoroethane spent solvent wastewaters represents substantial
treatment.  We would expect untreated 1,1,2-trichloro-l,2,2-trifluoro-
ethane wastes to be similar to untreated 1,1,1-trichloroethane wastes,
from which we transferred treatment data.  As discussed on page 5-110, in
reference to 1,1,1-trichloroethane, we believe these constituent
reductions substantially diminish the toxicity of the spent solvent
wastes containing 1,1,2-trichloro-l,2,2-trifluoroethane and substantially
reduce the likelihood of migration of 1,1,2-trichloro-l,2,2-tri-
fluoroethane from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard for
1,1,2-trichloro-l,2,2-trifluoroethane was 0.457 mg/L based on steam
stripping (see Table 13, 51 FR 1725).  The principal difference between
the proposed and promulgated treatment standards is the Agency's change
in the criteria for data transfer.   (See Section 5.5.1, page 5-14,  for a
discussion of the Agency's methodology for data transfer.)]
                                     5-115

-------
5.5.25   Trichloroethylene Wastewaters

    The Agency has biological treatment data for trichloroethylene at
plants 213, 217, and 253 in the OCPSF data base.  The Agency also has
data from steam stripping at plant 284 and biological treatment followed
by activated carbon adsorption at plant 246 in the OCPSF data base.  Data
are available for pilot-scale activated carbon adsorption (Reference 8).
Full-scale biological treatment data of wastewater from organic chemicals
manufacturing (commercially available patented PACT® process, Reference
4) and data from pilot-scale air stripping of tap water spiked with the
constituent (Reference 6) are also available.  The data are summarized in
Table 5-28 and calculation of the BDAT treatment standard is shown in
Table 5-29.

    The following steps were taken to derive the BDAT treatment standard
for trichloroethylene:

    1. We evaluated each data set to determine whether any of the data
       represent poor design or operation of the treatment, systems.  In
       EPA's judgment, one data point in the data set for steam stripping
       at plant 284 represented poor design and operation.  We confirmed
       this judgment using the outlier test (refer to Table 11-17,
       Appendix II).  The outlying data point was deleted.  Another data
       point was deleted because the influent was less than the
       quantification level (0.05 mg/L).

       In consideration of the amount of full-scale data available for
       trichloroethylene, we believe it is appropriate to exclude data on
       pilot-scale activated carbon adsorption and pilot-scale air
       stripping treatment.

    2. We calculated the arithmetic average effluent concentration and
       the variability factor for each data set as shown in Table 5-29.
       Process variability could not be calculated for biological
       treatment at plant 253 and biological treatment by the PACT®
       process because there was only one data pair available for each
       process.  Therefore, the average variability factor for BDAT
       biological treatment, 3.39, was used (calculation of the average
       variability factor is shown in Table 5-4, page 5-18).

       Process variability could not be calculated for biological
       treatment at plants 213 and 217 and for biological treatment
       followed by activated carbon adsorption at plant 246 because all
       effluent values were reported as less than or equal to the
       detection limit of 10 ug/L.  We would expect some variability in
                                     5-116

-------
       the data because the actual  concentrations  would range  from 0 to
       the detection limit of 10  ug/L.   To estimate  the variability, the
       Agency used the average variability factor  for BDAT  biological
       treatment,  3.39, for plants  213  and 217  and the average
       variability factor for BDAT  biological treatment followed by
       activated carbon adsorption, 6.17,  for plant  246 (calculation of
       the average variability factors  is  shown in Table 5-4,  page 5-18).

    3.  The analysis of variance method  was not  used  to compare different
       treatments of the same waste because data are available for only
       one type of treatment for  each waste.

    4.  EPA then analyzed the data to determine  if  the various  treatment
       concentration levels shown in Table 5-27 could be associated with
       separate waste treatability  subgroups.   Sufficient data did not
       exist to identify separate waste treatability subgroups;
       therefore,  one waste treatability subgroup  was established for all
       sources of wastewaters containing trichloroethylene  spent
       solvents.  The least stringent treatment level within the
       treatability subgroup was  selected for BDAT (0.062 mg/L from plant
       246) to ensure that the standard could be achieved for  all waste
       matrices within the waste  treatability subgroup.  The technology
       basis was biological treatment followed  by  activated carbon
       adsorption.

    5.  The BDAT treatment standard  for  trichloroethylene represents
       treatment of a variety of  waste  matrices generated by process
       streams from the manufacture of  over 50  different products.  The
       untreated waste concentration of trichloroethylene ranged from
       0.010 mg/L to 10.3 mg/L in these waste matrices.  All of these
       wastes were treated to the BDAT  treatment standard or below (0.062
       mg/L).  We believe these constituent reductions to substantially
       diminish the toxicity of the spent solvent  wastes containing
       trichloroethylene and substantially reduce  the likelihood of
       migration of trichloroethylene from spent solvent wastes.

    [The proposed technology-based  BDAT treatment  standard  for
trichloroethylene was <0.019 mg/L based on steam stripping  (see Table 13,
51 FR 1725).  The difference between the proposed  and promulgated
treatment standards is primarily  due to the incorporation of a
variability factor in derivation  of the promulgated  treatment  standard.
Other less significant factors affecting the change  in the  treatment
standard are the changes in data  editing (data  editing rules are
presented in Section 5.3) and deletion  of some  data  points  in  the final
rule because they represent poor  operation of the  treatment system (see
the discussion of the outlier test  in Section 5.4).]
                                     5-117

-------
                           Table 5-28

        TREATMENT PERFORMANCE DATA FOR TRICHLOROETHYLENE
     Plant 213                               Plant 213
Biological Treatment3                  Products Manufactured
Influent   Effluent
 (ug/L)     (ug/L)                Acetylenic alcohols & diols
                                  Ethylene-vinyl acetate
     16         10                  copolymer
     67         10                Polyvinyl alcohol resin
     76         10                PVC copolymers, ethylene-vinyl
                                    chloride
                                  Polyvinyl acetate resins
                                  Polyvinyl chloride
     Plant 217                               Plant 217
Biological Treatment3                  Products Manufactured
Influent   Effluent
 (ug/L)     (ug/L)                Phthalic anhydride
                                  Butyl  benzyl  phthalate
     98         10                Benzyl  chloride
    200         10                Tetrachlorophthalic anhydride
    224         10                Phosphate esters
                                  Phthalate esters
                                  Polybenzyl ethyl benzene
aThe data do not represent paired data (i.e.,  the samples were
 not collected so as to fully account for the  retention time in
 the treatment system).
                                    5-118

-------
                       Table 5-28 (Continued)

          TREATMENT PERFORMANCE DATA FOR TRICHLOROETHYLENE
          Plant 246
    Biological Treatment3
         Followed by
 Activated Carbon Adsorption
     Influent   Effluent
      (ug/L)     (ug/L)
          SO
          70
          40
     10B
     10b
     10b
           Plant 246
     Products Manufactured

Aniline
Dinitrotoluene (mixed)
Methylene dipnenyl  diisocyanate
Nitrobenzene
Polymeric methylene diphenyl
  diisocyanate
Polyoxypropylene glycol
Toluene diamine (mixture)
Toluene diisocyanates (mixture)
Polymeric methylene dianiline
Polyurethane resins
     Plant 253
Biological Treatment3
Influent   Effluent
 (uo/L)     (up/Li
    484
16
                             Plant  253
                       Products  Manufactured

                  Polypropylene  resins
aThe data do not represent paired data (i.e.,  the samples were
 not collected so as to fully account for the  retention time in
 the treatment system).
''in the data base from which this data was taken, the sampling
 data was designated using a different code,  plant 219, because
 it represented a different sampling episode.
                                   5-119

-------
                     Table 5-28 (Continued)

        TREATMENT PERFORMANCE DATA FOR TRICHLOROETHYLENE
     Plant 284
  Steam Stripping
Influent
(ua/L)
1,650
5,200
5,000
1,720
1,560
59
10,300
90
84
83
210
1 ,600
160
204
10
Effluent
(ug/L)
10
10
10
10
10
10
10
10
10
10
10
27
10
85a
10b
 D.G. Hutton, 1979
Biological Treatmentc'd
Influent   Effluent
 (ug/L)     (ug/L)
           Plant 284
     Products Manufactured

Benzene
1,3-Butadiene
Ethylene
Propylene
Methylene chloride
1,1,2-Trichloroethane
Vinylidine chloride
1,2,3-Trichloropropene
1,2-Oichloropropane
Propylene oxide
Ethylene oxide
Propylene glycol
Dipropylene glycol
Tripropylene glycol
Ethylene glycol
Methyl chloride
Diethylene glycol
Triethylene glycol
Tetraethylene glycol
Ethanol amines
Polypropylene
Chloroform
Carbon tetrachlonde
1,2-Dichloroethane
Vinyl chloride

       Description of
       Waste Treated

Wastewater from organic chemi-
cals manufacturing.
     60
                 5.8
aln EPA's judgment, this data point represented poor design
 and operation.  We confirmed this judgment using the outlier
 test (refer to Table 11-17, Appendix II) and this data point
 was deleted.
^This data point was deleted from analyses because the
influent
 is less than the quantification level (50 ug/L).
cThe data do not represent paired data (i.e., the samples were
 not collected so as to fully account for the retention time in
 the treatment system).
"^Commercially available patented PACT® process.
                                   5-120

-------
                      Table 5-28 (Continued)

         TREATMENT PERFORMANCE DATA FOR TRICHLOROETHYLENE
               Love and Eilers,  1982
         Pilot-Scale Air Stripper.  Site 1
  Average
  Influent     Average Effluent Concentration  at
Concentration  Various Air-to-Water Ratios  (uo/L)
   (ua/L)      HI  111  HI  4j_L  8_il  16:1   20:1

    1,064      796  614  508  319   53
      397      223  273  102   82   22   <1     <1
      241      136  110   61   53    8    2      3
      110       40   28   18    9    3   <1     <1
       73       22   14    8    6    1   <1     <1
                                     Description
                                         of
                                    Waste Treated

                                    Tap water
                                    spiked with
                                    tetrachloro-
                                    ethylene and
                                    trichloro-
                                    ethylene.
 Ruggiero and Ausubel,  1982
Pilot-Scale Granular Activated
	Carbon Column	
     Influent
      (ug/L)

      171
Effluent
 (ug/L)

 0.59
   Description of
   Waste Treated

Contaminated drinking
water supply.
                                  5-121

-------
                           Table 5-29

            CALCULATION OF BOAT FOR TRICHLOROETHYLENE
                         Average
                        Treatment                 Treatment Con-
                                                 Avg. x VF (ug/L)

                                                       20

                                                       34

                                                       34

                                                       54

                                                       62
Plant
No.
284
213
217
253
246
Concentration Variability
Technology (ua/L) Factor
Steam Stripping 11.3
Biological 10
Biological 10
Biological 16
Biological fol- 10
lowed by
Activated
Carbon
1.81
3.39a
3.39a
3.39a
6.17b
  lc    Biological          5.8       3.39a            20
aAverage variability factor for BOAT biological  treatment (see
 Table 5-4 and the discussion on page 5-17).

''Average variability factor for BOAT biological  treatment fol-
 lowed by activated carbon adsorption (see Table 5-4 and the
 discussion on page 5-17).

"-Commercially available patented PACT® process.
                                5-122

-------
5.5.26   Trichlorofluoromethane Wastewaters

    The Agency has trichlorofluoromethane treatment data from full-scale
biological treatment of wastewater from organic chemicals manufacturing
(commercially available patented PACT® process, Reference 4).  The data
are summarized in Table 5-30.

    The following steps were taken to derive the BOAT treatment standard
for trichlorofluoromethane:

    1. We evaluated the data set to determine whether the data represent
       poor design or operation of the treatment system.  The available
       data and information did not show any of the data to represent
       poor design and operation.  Accordingly, none of the data were
       deleted on this basis.

    2. We calculated the arithmetic average treatment concentration and
       the variability factor for the data set.  The average effluent
       concentration is 13 ug/L.  Process variability could not be
       calculated for this plant because there is only one data pair
       available from this process.  Therefore, the average variability
       factor for BOAT biological treatment, 3.39, was used (calculation
       of the average variability factor is shown in Table 5-4, page
       5-18).

    3. The analysis of variance method was not used to compare different
       treatments of the same waste because data are available for only
       one type of treatment for each waste.

    4. Sufficient data did not exist to identify separate waste
       treatability subgroups; therefore, one waste treatability subgroup
       was established for all sources of wastewaters containing
       trichlorofluoromethane spent solvents.  The BOAT treatment level
       for trichlorofluoromethane was selected (0.044 mg/L from Hutton,
       1979) by multiplying the process effluent concentration, 0.013
       mg/L by the average variability factor from BDAT biological
       treatment, 3.39.  This calculated treatment standard is below the
       quantification level and could not be used as the treatment
       standard; therefore, the treatment standard was set at the
       quantification level of 0.05 mg/L.  The technology basis was
       biological treatment.

    5. The BDAT treatment standard for trichlorofluoromethane represents
       treatment of a waste matrix generated by process streams from the
       manufacture of organic chemicals.  The untreated waste
                                     5-123

-------
       concentration of trichlorofluoromethane was as high as 0.920 mg/L
       in this waste matrix.   This waste was treated to a concentration
       below the BDAT treatment standard (0.050 mg/L).  We believe these
       constituent reductions substantially diminish the toxicity of the
       spent solvent wastes containing trichlorofluoromethane and
       substantially reduce the likelihood of migration of
       trichlorofluoromethane from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard for
trichlorofluoromethane was 0.457 mg/L based on steam stripping (see
Table 13, 51 FR 1725).  The principal differences between the proposed
and promulgated treatment standards are EPA's consideration of
quantification levels in setting the standard (see the discussion of the
use of quantification levels in Section 5.5, page 5-12) arid use of
biological treatment performance data at promulgation that were not used
at proposal.  (The data were deleted at proposal because the influent
concentration was below the screening level for trichlorofluoromethane.)
The incorporation of a variability factor in derivation of the
promulgated treatment standard also contributed to the change in the
treatment standard since proposal.]
                                      5-124

-------
                           Table 5-30

     TREATMENT PERFORMANCE DATA FOR TRICHLOROFLUOROMETHANE
 D.G. Mutton,  1979                       Description  of
Biological  Treatment3•^                  Waste  Treated
Influent   Effluent
 (ug/T)     (ug/L)                Wastewater  from organic  chemi-
                                  cals  manufacturing.
    920         13
aThe data do not represent  paired data  (i.e.,  the  samples were
 not collected so as to fully account for  the  retention time  in
 the treatment system).
Commercially available patented PACT®  process.
                                 5-125

-------
5.5.27   Xylene Wastewaters

    The Agency has wet air oxidation data (Reference 10),  activated
carbon adsorption followed by steam stripping data (Reference 9),  and
carbon adsorption data (Reference 7) for xylene.   The data are summarized
in Table 5-31.

    The following steps were taken to derive the  BDAT treatment standard
for xylene:

    1. We evaluated each data set to determine whether any of the  data
       represent poor design or operation of the  treatment systems.  The
       data for wet air oxidation were deleted because the detection
       limit of 500 ug/L for xylene in the oxidation product is too high
       for the data to be meaningful with regard  to how well the system
       will perform.  The data from the Iron and  Steel Manufacturing
       Development Document were not used because insufficient
       information exists in some cases to determine the concentrations
       treated and, in other cases, which technology was achieving
       removal.  Also, in some cases, the treated values represented
       significant dilution of the wastestream.

    2. We calculated the arithmetic average treatment concentration and
       the variability factor for the data set.  Process variability
       could not be calculated for activated carbon adsorption at  the
       pilot-scale plant because there is only one data pair available
       from this process.  Therefore, the average variability factor for
       all BDAT activated carbon adsorption, 4.54, was used (calculation
       of the average variability factor is shown in Table 5-4).

    3. The analysis of variance method was not used to compare different
       treatments of the same waste because data  are available for only
       one type of treatment for each waste.

    4. Sufficient data did not exist to identify  separate waste
       treatability subgroups; therefore, one waste treatability subgroup
       was established for all sources of wastewaters containing xylene
       spent solvents.  The BDAT treatment level  for xylene was selected
       (0.0005 mg/L from Becker and Wilson, 1978) by multiplying the
       process effluent concentration, 0.0001 mg/L by the average
       variability factor from BDAT activated carbon adsorption, 4.54.
       This calculated standard is below the guantification level  and
       could not be used as the treatment standard.  Therefore, the
       treatment standard was set at the quantification level of 0.05
       mg/L.  The technology basis was activated  carbon adsorption.
                                      5-126

-------
    5.  The BDAT treatment standard for xylene represents treatment of a
       waste matrix from a disposal site.   The untreated waste
       concentration of xylene was as high as 0.140 mg/L in this waste
       matrix.   This waste was treated to a concentration below the BDAT
       treatment standard (0.050 mg/L).   We believe these constituent
       reductions substantially diminish the toxicity of the spent
       solvent  wastes containing xylene and substantially reduce the
       likelihood of migration of xylene from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard for xylene was
<0.005 mg/L based on activated carbon adsorption followed by steam
stripping (see  Table 13, 51 FR 1725).  The principal differences between
the proposed and promulgated treatment standards are EPA's consideration
of quantification levels in setting the standard (see the discussion on
the use of quantification levels in Section 5.5 on page 5-12)  and the
incorporation of a variability factor in derivation of the promulgated
treatment standard.  Other less significant factors affecting the change
in the treatment standard are the changes in data editing (data editing
rules are presented in Section 5.3) and deletion of some data points in
the final rule  because they represent poor operation of the treatment
system (see the discussion of the outlier test in Section 5.4).]
                                     5-127

-------
                           Table 5-31

             TREATMENT PERFORMANCE DATA FOR XYLENE
Iron and Steel  Manufacturing
 Development Document,  1980
Multiple Treatment Technologies

Average        Average
Influent       Effluent
 (ug/U         (ug/U      Plant
101,000
                                       Description of
                                       Waste Treated
                           009         Excess ammonia  liquor
                                      and  benzol  plant
                                      wastewaters
Data Submitted by Zimpro,  Inc.,  1986
	Wet Air Oxidation	
 Raw Waste
   (uq/L)
              Diluted
               Feed
               (Ud/U
Oxidation
 Product
 (ug/L)
Description of
Waste Treated
    212,000
                21,200
    <500
General organic
Becker and Wilson,  1978
 Pilot-Scale Granular
Activated Carbon Column
Influent   Effluent
 (ug/U     (ug/U

   140         <0.1
                                         Description of
                                         Waste Treated

                                  Runoff water from a waste dis-
                                  posal site's containment dikes
                              5-128

-------
5.6      Development of BDAT Treatment Standards for F001-F005 Spent
         Solvent Wastes (Other Than Wastewater)

    BDAT treatment standards for F001-F005 spent solvent wastes (other
than wastewater) are presented in Table 5-2.  BDAT treatment standards
for spent solvent wastes other than wastewaters are based on incineration
of the waste.  Treatment standards were calculated from data on the
analysis of the TCLP extract of incinerator residue.  Descriptions of how
the treatment standards were derived for spent solvent wastes other than
wastewaters are presented in this section.  Treatment performance data
for each constituent are also presented in this section.  Data sets
including all constituents and all pollutant parameters analyzed in the
wastes treated at each incinerator are included in Appendix I.  Where
data on the TCLP extract were not available, treatment data were
transferred based on structural similarity.  Transfer of incineration
treatment data is discussed in Section 5.6.1.

    The derivation of BDAT treatment standards includes a variability
analysis as discussed in Section 5.4.  For some data sets, we had
insufficient data to develop variability factors.   Therefore, to account
for process variability, an average variability factor was calculated
from data available from TCLP extracts of incinerator ash from the
burning of waste containing acetone, methylene chloride, and toluene.
Calculation of the average variability factor is discussed in Section
5.6.2.

    In some cases, the treatment standard derived from the data was below
the EPA published analytical quantification level  for a specific
constituent.  In these instances, the BDAT treatment standard was set at
the quantification level,  which is the lowest level at which EPA can
support analytical quantification over the range of wastes that will be
subject to this rule.
                                     5-129

-------
5.6.1    Transfer of Incineration Treatment Data

    Where data on the TCLP extract of incinerator ash were not available
to the Agency, treatment data were transferred from other constituents
for which data were available.  For this rulemaking, treatment data were
transferred based on similarity of chemical structure.  Chemical
structure is commonly used to predict how organic compounds will react
with other compounds and under various conditions.  Constituents
considered to be similar in chemical structure contain the same
functional groups.  Functional groups such as double bonds, hydroxyl
groups, ketone groups, and amino groups, are the parts of the molecule
where most chemical reactions occur (including combustion reactions which
occur during incineration).  A compound's chemical, physical, and
thermodynamic properties are also dependent on chemical structure.
Included in Table 5-32 are the structural groups upon which the transfer
of treatment standards was based.

    Although parameters such as the heat of combustion could be used to
indicate the amenability of a compound to incineration, the Agency
believes that for the wide range of wastes covered for this particular
rulemaking, a broader approach to data transfer is warranted.  Therefore,
the Agency transferred treatment standards based on general chemical
structure rather than on a single physical, chemical, or thermodynamic
property specific to the treatment technology.

    The F001-F005 hazardous wastes were grouped according to chemical
structures as listed in Table 5-32.  To best account for the range of
physical and chemical properties within a structural group that affect
treatment, the Agency transferred data from the compound with the least
stringent treatment standard for any member of that structural group.  If
no treatment data were available for any member of a particular
structural group, data representing the least stringent treatment
standard from the next most similar structural group were transferred.
For example, no treatment data were available for any member of the
alcohols, esters, and ethers structural groups.  The ketones were
considered to be the next most similar structural group, based on the
oxygen containing, electron-releasing functional groups present in all
four structural groups.  Therefore, data representing the least stringent
treatment standard for constituents in the ketones group were transferred
to the alcohols, ethers, and esters groups.
                                      5-130

-------
                                           Table 5-32

                     GROUPING OF  SPENT  SOLVENT  CONSTITUENTS FOR  TRANSFER OF
                         BOAT TREATMENT DATA FOR ALL OTHER  F001  - F005
                                         SPENT  SOLVENTS
    Name of
Structural Group
                                          Functional
                                            Group
     Consti tuent
   Treatment
    Standard
 (mg/L in TCLP
   Extract of
Incinerator Ash)
Constituent From Which
Data Were Transferred
 I
h->
CO
Halogenated       R-X
Aliphati cs
Non-Hal ogenated /("")\-R
Aromati cs
Halogenated       R=R'
Alkenes

Halogenated   /^T\
Aromati cs     \O)~X
Carbon tetrachloride          0.96a
Methylene chloride            0.96
1,1,1-Trichloroethane         0.41
l,l,2-Trichloro-l,2,2-
 trif1uoroethane              0.96a
Trichlorofluoromethane        0.96a

Ethylbenzene                  0.053b
Toluene                       0.33
Xylene                        0.15
Nitrobenzene                  0.125b
Pyridine                      0.33a

Tetrachloroethylene           0.05b
Trichloroethylene             0.091

Chlorobenzene                 0.05b
1,2-Dichlorobenzene           0.125b
                                                                                                  Methylene  Chloride
                                                                          Methylene Chloride
                                                                          Methylene Chloride
                                                                          To!uene
Ketones
                                          R-C-R1     Acetone                        0.59
                                            ||       Cyclohexanone                  0.75a
                                            0       Methyl ethyl ketone            0.75
                                                    Methyl isobutyl  ketone         0.33
                                                                          Methyl  ethyl  ketone
transferred treatment  data.
"Treatment standard shown  is  the  quantification  level  for  the  constituent.

-------
                                    Table 5-32  (Continued)

                     GROUPING OF  SPENT  SOLVENT  CONSTITUENTS FOR TRANSFER OF
                         BOAT TREATMENT DATA FOR ALL OTHER F001 - F005
                                         SPENT  SOLVENTS
    Name of       Functional
Structural Group    Group
Consti tuent
   Treatment
    Standard
 (mg/L in TCLP
   Extract of
Incinerator Ash)
Constituent From Which
Data Were Transferred




en
1— *
CO



Al cohols

Ethers
Esters



Phenol s
Organic Sulfur
Compounds
R-OH n-Butyl alcohol
Isobutanol
Methanol
R-O-R' Ethyl ether
R-C-OR1 Ethyl acetate
II
0

/O\-OH Cresols
R = S Carbon disulfide

5.0a'b Methyl ethyl ketone
5.0a-b Methyl ethyl ketone
0.75a Methyl ethyl ketone
0.75a Methyl ethyl ketone
0.75a Methyl ethyl ketone



0.75a Methyl ethyl ketone
4.81

transferred treatment data.
"Treatment standard shown is  the  quantification  level  for  the  constitutent.

-------
5.6.2    Derivation of An Average Variability Factor for Incineration

    The derivation of BOAT treatment standards includes a variability
analysis as discussed in Section 5.4.1.  For some data sets, we had
insufficient data to develop variability factors; in these cases we used
a variability factor that represented the average of the variability
factors from available data sets.  Calculation of the average variability
factors is shown in Table 5-33, page 5-134.
                                     5-133

-------
                           Table 5-33

           VARIABILITY FACTORS FOR INCINERATION DATA*

   Constituent                Site         Variability Factor

Methylene Chloride              2                9.05
Acetone                         2                1.629
                             AVERAGE             5.34
"The variability factors for methylene chloride and acetone
 shown in the above table are generated from a plant sampled
 subsequent to proposal.  Analysis of the samples were
 completed after the Agency's September 5,  1986 Notice of Data
 Availability (51 FR 31783).  The average variability factor
 calculated from these data is somewhat higher than the
 variability factor generated from data available at proposal
 and presented in the Notice of Data Availability (5.34
 compared to 3.56).  In addition, since this value is based on
 incineration data, it provides a better representation of the
 variability experienced in a full-scale incinerator than does
 the previous value derived from wastewater treatment
 technlogies.  The specific data used to generate the
 individual variability factors for methylene chloride and
 acetone has been claimed to be confidential.
                             5-134

-------
5.6.3    Acetone (Other Than Wastewater)

    The Agency has treatment data for the TCLP extract of incinerator ash
from the treatment of acetone (Reference 11).  The data are summarized in
Table 5-34.

    The following steps were taken to derive the BDAT treatment standard
for acetone:

    1. We evaluated the data to determine whether any of the data
       represented poor design or operation of the incineration system.
       Data from one facility (site 6) were deleted because the
       incineration system control devices were not properly designed and
       operated.  The facility is under a Consent Decree to replace and
       improve the current incinerator control system.  Data from another
       facility (site 2) were deleted because the Agency judged that the
       system was not properly operated at the time the data were
       collected.  A follow-up sampling visit confirmed the Agency's
       judgment.  The new data were not used in the determination of the
       long-term performance average for incineration of acetone;
       however, the data were used to develop a variability factor for
       incineration.

    2. We determined an arithmetic average residue concentration level
       and a variability factor for each data set.  Residue concentration
       levels reported as less than or equal to the reported detection
       limit were set equal to the detection limit for statistical
       analyses.  This is a conservative approach since the actual
       concentration would be between zero and the detection limit.   Two
       residue concentration levels were reported for site  5,  one for
       each incinerator at the site.   These were considered as two
       separate data points.

       Process variability could not be calculated from the incineration
       data because only one  influent and effluent data pair was
       available for each data set.   Therefore,  to account  for process
       variability, an average variability factor was calculated for
       incineration, 5.34 (calculation of the average variability factor
       is shown in Table 5-33).

    3. The analysis of variance  method was not used to compare different
       treatments of the same waste  because data are available for only
       one type of treatment  for each waste.

    4. EPA then analyzed the  data to determine if the various  treatment
       concentration levels shown in Table 5-34  could be  associated with
                                     5-135

-------
       separate waste treatability subgroups.   Sufficient data did not
       exist to identify separate waste treatability subgroups;
       therefore,  one waste treatability subgroup was established for all
       sources of  wastes (other than wastewater)  containing acetone spent
       solvents.  The least stringent treatment level within the
       treatability subgroup was selected for  BDAT (0.59 mg/L for site 7
       obtained by multiplying the variability factor by the highest
       average residue concentration level)  to ensure that the standard
       could be achieved for all waste matrices within the waste
       treatability subgroup.   The technology  basis was incineration.

    5. The BDAT treatment standard for acetone represents treatment of a
       variety of  waste matrices incinerated at six different sites.  The
       untreated waste concentration of acetone ranged from 36 mg/kg to
       160,000 mg/kg in these waste matrices.   All of these wastes were
       treated to  the BDAT treatment standard  or below (0.59 mg/L).  We
       believe these constituent reductions  substantially diminish the
       toxicity of the spent solvent wastes  containing acetone and
       substantially reduce the likelihood of  migration of acetone from
       spent solvent wastes.

    [The proposed  technology-based BDAT treatment standard for acetone
was estimated at the detection limit of <0.050 mg/L based on incineration
(see Table 11, 51  FR 1722).  The difference  between the proposed and
promulgated treatment standards is primarily a result of additional data
gathering subsequent to proposal.  The new data were presented in EPA's
Notice of Availability of Data (51 FR 31783).   In addition, a variability
analysis was incorporated into the development of the treatment  standards
for promulgation.]
                                     5-136

-------
                                                                Table  5-34

                                                       INCINERATION  DATA  FOR  ACETONE
01
 i
                   Si te  Type of Incinerator

                     1    Rotary Kiln with
                         Secondary Combustor

                     3    Rotary Kiln with
                         Secondary Combustor

                     5    Fixed Hearth (Two
                         Separate Incinera-
                         tion Systems)
                     7   Fixed Hearth with
                         Secondary Combustor
Wastes Incinerated

PCB Contaminated Dirt
 Flow-Weighted
    Average
Influent (mg/kg)

         36
Drum Feed Solids            160,000
Liquid Waste Fuel

(From Furniture Manu-        13,500
  facturing Industry)        13,500
Solvent Wastes
High-Btu Liquid Wastes
Low-Btu Liquid Wastes
Lacquer-Coated Cardboard

High-Btu Liquids              3,120
Low-Btu Liquids
Solids Feed
                                                                                              Inci nerator
                                                                                               Residue*
<5
                    <500
                    <500
                    <500
 TCLP
(ug/L)   Footnotes

   <5       a
                      <2.5      <5
          <5
          <5
         110
                         Rotary Kiln with
                         Secondary Liquid
                         Injection Combustor
Liquid Waste Fuel
     86,000
<2.5      <5
                   'Values shown as "<" were reported as below the indicated detection limits.

                   (a)  Influent concentration is flow-weighted average.
                   (b)  Influent concentration is an arithmetic average.

-------
                                                       Table 5-34 (Continued)

                                                    INCINERATION DATA FOR ACETONE
                Site  Type of Incinerator

                  9   Rotary Kiln with
                      Secondary Combustor
                              Wastes Incinerated

                              High-Btu  Liquids
                              Low-Btu Liquids
                              Solids Feedb
Flow-Weighted
Average
Influent (ma/ka)
Incinerator
Residue*
Total TCLP
(mg/kg) (ug/L) Footnotes
223,335
<2.5
67
en
CO
CO
"Values shown as "<" were reported as  below the indicated detection limits.

(a)  Influent concentration is flow-weighted average.
(b)  Gel  and filter press residue.

-------
5.6.4    n-Butyl Alcohol (Other Than Wastewater)

    The Agency has no data on TCLP extracts of residue from incineration
of n-butyl alcohol to use in the derivation of the BDAT treatment
standard.  For reasons presented in Section 5.6.1, EPA used chemical
structure as the basis for transferring treatment data to n-butyl alcohol
spent solvent wastes other than wastewaters.  Specifically we transferred
treatment data from methyl ethyl ketone, which contains the ketone
functional group, to n-butyl alcohol, which contains the hydroxyl
functional group.  The alcohols structural group is most structurally
similar to the ketones group based upon their oxygen-containing,
electron-releasing functional groups.

    The Agency has data on the analysis of the TCLP extract of
incineration residue for three compounds in the ketones structural
group:  acetone, methyl ethyl ketone, and methyl isobutyl ketone.  To
best account for the range of physical and chemical properties within a
structural group that affect treatment by a specific technology, the
Agency transferred data representing the least stringent treatment
standard from the compounds for which data were available in the ketones
structural group.  The data from which the treatment standard for
incineration of methyl ethyl ketone was derived were transferred to
n-butyl alcohol.  The treatment standard is 0.75 mg/L based on the
transferred data.  The transferred value is below the quantification
level for n-butyl alcohol and could not be used as the treatment
standard.  Therefore, the treatment standard is set at the quantification
level of 5.0 mg/L.

    We believe the BDAT treatment standard for n-butyl alcohol spent
solvent wastes (other than wastewater) represents substantial treatment.
We would expect untreated n-butyl alcohol wastes to be similar to
untreated methyl ethyl ketone wastes from which we transferred treatment
data since they are used in many similar manufacturing processes, as
shown in Section 2 of this document.  As discussed on page 5-163, in
reference to methyl ethyl ketone, we believe these constituent reductions
substantially diminish the toxicity of the spent solvent wastes
containing n-butyl alcohol and substantially reduce the likelihood of
migration of n-butyl alcohol from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard for n-butyl
alcohol was estimated at the detection limit of <0.100 mg/L based on
incineration (see Table 11,  51 FR 1722).  The principal difference
between the proposed and promulgated treatment standards is the Agency's
change in the criteria for data transfer (see Section 5.6.1, page 5-130,
for a discussion of the Agency's methodology for data transfer.)]
                                     5-139

-------
5.6.5    Carbon Bisulfide (Other Than Wastewater)

    The Agency has treatment data for the TCLP extract of incinerator ash
from the treatment of carbon disulfide (Reference  11).  The data are
summarized in Table 5-35.

    The following steps were taken to derive the BDAT treatment standard
for carbon disulfide:

    1.  We evaluated the data to determine whether  any of the data
       represented poor design or operation of the incineration system.
       The available data and information did not  show any of the data to
       represent poor design and operation.   Accordingly, none of the
       data were deleted on this basis.

    2.  We determined an arithmetic average residue concentration level
       and a variability factor for each data set.   Residue concentration
       levels reported as less than or equal to the reported detection
       limit were set equal to the detection limit for statistical
       analyses.  This is a conservative approach  since the actual
       concentration would be between zero and the detection limit.
       Process variability could not be  calculated from the incineration
       data because only one influent and effluent data pair was
       available for each data set.  Therefore, to account for process
       variability, an average variability factor  was calculated for
       incineration, 5.34 (calculation of the average variability factor
       is shown in Table 5-33).

    3.  The analysis of variance method was not used to compare different
       treatments of the same waste because data are available for only
       one type of treatment for each waste.

    4.  EPA then analyzed the data to determine if  the various treatment
       concentration levels shown in Table 5-35 could be associated with
       separate waste treatability subgroups.  Sufficient data did not
       exist to identify separate waste  treatability subgroups;
       therefore, one waste treatability subgroup  was established for all
       sources of wastes (other than wastewater) containing carbon
       disulfide spent solvents.  The least stringent treatment level
       within the treatability subgroup was selected for BDAT (4.81 mg/L
       from site 3 obtained by multiplying the variability factor by the
       highest average residue concentration level) to ensure that the
       standard could be achieved for all waste matrices within the waste
       treatability subgroup.  The technology basis was incineration.
                                      5-140

-------
    5.  The BDAT treatment standard for carbon disulfide represents
       treatment of a variety of waste matrices incinerated at two
       sites.   The untreated waste concentration of carbon disulfide was
       as high as 400 mg/kg in these waste matrices.  All of these wastes
       were treated to the BDAT treatment standard or below (4.81 mg/L).
       We believe these constituent reductions substantially diminish the
       toxicity of the spent solvent wastes containing carbon disulfide
       and substantially reduce the likelihood of migration of carbon
       disulfide from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard for carbon
disulfide was estimated at the detection limit of <0.010 mg/L based on
incineration (see Table 11, 51 FR 1722).  The difference between the
proposed and promulgated treatment standards is primarily a result of
additional data gathering subseguent to proposal.  The new data were
presented in EPA's Notice of Availability of Data (51 FR 31783).  In
addition, a variability analysis was incorporated into the development of
the treatment standards for promulgation.]
                                     5-141

-------
                                                             Table 5-35

                                               INCINERATION DATA FOR CARBON DISULFIDE
                Site  Type of  Incinerator

                  3   Rotary Kiln with
                      Secondary Combustor

                  8   Rotary Kiln with
                      Secondary Liquid
                      Combustor
Wastes Incinerated

Drum Feed Sol ids
Liquid Waste Fuel

Liquid Waste Fuel
  Flow-Weighted
     Average
Influent* (mo/kg)

        <400
        <400
                                                                                           Incinerator
                                                                                            Residue*
Total


   2.8
  <2.0
 TCLP
(ug/L)   Footnotes

  900       a
ro
                 •Values  shown  as  "<"  were  reported as below the indicated detection limits.

                 (a)   Influent  concentration  is  flow-weighted average.

-------
5.6.6    Carbon Tetrachloride (Other than Wastewater)

    The Agency has no data on TCLP extracts of residue from incineration
of carbon tetrachloride to use in the derivation of the BDAT treatment
standard.  For reasons presented in Section 5.6.1, EPA used chemical
structure as the basis for transferring treatment data to carbon
tetrachloride spent solvent wastes other than wastewaters.  Specifically
we transferred treatment data from methylene chloride to carbon
tetrachloride; both contain the halogen functional group.

    The Agency has data on the analysis of the TCLP extract of
incineration residue for two compounds in the halogenated aliphatics
structural group:  methylene chloride and 1,1,1-trichloroethane.  To best
account for the range of physical and chemical properties within a
structural group that affect treatment by a specific technology, the
Agency transferred data representing the least stringent treatment
standard from the compounds for which data were available in the
halogenated aliphatics structural group.  The data from which the
treatment standard for incineration of methylene chloride was derived
were transferred to carbon tetrachloride.  The treatment standard is
0.96 mg/L based on the transferred data.

    We believe the BDAT treatment standard for carbon tetrachloride spent
solvent wastes (other than wastewater) represents substantial treatment.
We would expect untreated carbon tetrachloride wastes to be similar to
untreated methylene chloride wastes from which we transferred treatment
data.  As discussed on page 5-159, in reference to methylene chloride, we
believe these constituent reductions substantially diminish the toxicity
of the spent solvent wastes containing carbon tetrachloride and
subsequently reduce the likelihood of migration of carbon tetrachloride
from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard for carbon
tetrachloride was estimated at the detection limit of <0.010 mg/L based
on incineration (see Table  11, 51 FR 1722).   The principal difference
between the proposed and promulgated treatment standards is the Agency's
change in the criteria for  data transfer (see Section 5.6.1, page 5-130,
for a discussion of the Agency's methodology for data transfer.}]
                                     5-143

-------
5.6.7    Chlorobenzene (Other Than Wastewater)

    The Agency has treatment data for the TCLP extract of incinerator ash
from the treatment of chlorobenzene (Reference 11).   The data are
summarized in Table 5-36.

    The following steps were taken to derive the BDAT treatment standard
for chlorobenzene:

    1. We evaluated the data to determine whether any of the data
       represented poor design or operation of the incineration system.
       The available data and information did not show any of the data to
       represent poor design and operation.  Accordingly, none of the
       data were deleted on this basis.  Data were deleted from one site
       because chlorobenzene was reported as below the detection limits
       for both the influent and the TCLP extract of the ash.

    2. We determined an arithmetic average residue concentration level
       and a variability factor for each data set.  Residue concentration
       levels reported as less than or equal to the reported detection
       limit were set equal to the detection limit for statistical
       analyses.  This is a conservative approach since the actual
       concentration would be between zero and the detection limit.
       Process variability could not be calculated from the incineration
       data because only one influent and effluent data pair was
       available for each data set.  Therefore, to account for process
       variability, an average variability factor was calculated for
       incineration, 5.34 (calculation of the average variability factor
       is shown in Table 5-33).

    3. The analysis of variance method was not used to compare different
       treatments of the same waste because data are available for only
       one type of treatment for each waste.

    4. EPA then analyzed the data to determine if the various treatment
       concentration levels shown in Table 5-36 could be associated with
       separate waste treatability subgroups.  Sufficient data did not
       exist to identify separate waste treatability subgroups;
       therefore, one waste treatability subgroup was established for all
       sources of wastes (other than wastewater) containing chlorobenzene
       spent solvents.  The least stringent treatment level within the
       treatability subgroup was selected for BDAT (0.016 mg/L for sites
       8 and 9 obtained by multiplying the variability factor by the
       highest average residue concentration level) to ensure that the
       standard could be achieved for all waste matrices within the waste
       treatability subgroup.  This calculated standard is below the
       quantification level and could not be used as the treatment
                                      5-144

-------
       standard; therefore, the treatment standard is set at the
       quantification level of 0.05 mg/L.  The technology basis was
       incineration.

    5. The BOAT treatment standard for chlorobenzene represents treatment
       of a variety of waste matrices incinerated at two sites.  The
       untreated concentration of chlorobenzene was as high as 1,100
       mg/kg in these waste matrices.  All of these wastes were treated
       to the BDAT treatment standard or below (0.05 mg/L).  We believe
       these constituent reductions substantially diminish the toxicity
       of the spent solvent wastes containing chlorobenzene and
       substantially reduce the likelihood of migration of chlorobenzene
       from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard for
chlorobenzene was estimated at the detection limit of <0.020 mg/L based
on incineration (see Table 11, 51 FR 1722).   The difference between the
proposed and promulgated treatment standards is primarily a result of
additional data gathering subsequent to proposal and use of the
analytical quantification level as the treatment standard since the
standard derived from the data is below the  EPA published analytical
quantification level for chlorobenzene (see  Table 5-1 and the discussion
on page 5-17).   The new data were presented  in EPA's Notice of
Availability of Data (51 FR 31783).  In addition, a variability analysis
was incorporated into the development of the treatment standards for
promulgation.]
                                     5-145

-------
                                             Table  5-36

                                 INCINERATION DATA  FOR CHLOROBENZENE
Site  Type of Incinerator

  8   Rotary Kiln with
      Secondary Liquid
      Injection Combustor

  9   Rotary Kiln with
      Secondary Combustor
Wastes Incinerated

Liquid Waste Fuel
High-Btu Liquids
Low-Btu Liquids
Solids Feedc
 Flow-Wei ghted
    Average
Influent (ma/kg)

      1,100
       ,034
Incinerator
 Residue*
         TCLP
        (ug/L)  Footnotes

 <1.5      <3       a
           <3
"Values shown as "<" were reported as  below the indicated  detection limits.

(a)  Influent concentration is flow-weighted average.
(b)  Influent concentration is an arithmetic average.
(c)  Gel and filter press residue.

-------
5.6.8.   Cresols (Cresylic Acid) (Other Than Wastewater)

    The Agency has no data on TCLP extracts of residue from incineration
of cresols (cresylic acid) to use in the derivation of the BDAT treatment
standard.  For reasons presented in Section 5.6.1, EPA used chemical
structure as the basis for transferring treatment data to cresols
(cresylic acid) spent solvent wastes other than wastewaters.
Specifically we transferred treatment data from methyl ethyl ketone,
which contains the ketone functional group, to cresol (cresylic acid),
which contains the phenol functional group.  The phenols structural group
is most structurally similar to the ketones group based upon their
oxygen-containing, electron-releasing functional groups.

    The Agency has data on the analysis of the TCLP extract of
incineration residue for three compounds in the ketones structural
group:  acetone, methyl ethyl ketone, and methyl isobutyl ketone.  To
best account for the range of physical and chemical properties within a
structural group that affect treatment by a specific technology, the
Agency transferred data representing the least stringent treatment
standard from the compounds for which data were available in the ketones
structural group.  The data from which the treatment standard for
incineration of methyl ethyl ketone was derived were transferred to
cresols (cresylic acid).  The treatment standard is 0.75 mg/L based on
the transferred data.

    We believe the BDAT treatment standard for cresols (cresylic acid)
spent solvent wastes (other than wastewater) represents substantial
treatment.  We would expect untreated cresols (cresylic acid) wastes to
be similar to untreated methyl ethyl ketone wastes from which we
transferred treatment data.  As discussed on page 5-163, in reference to
methyl ethyl ketone, we believe these constituent reductions
substantially diminish the toxicity of the spent solvent wastes
containing cresols (cresylic acid) and substantially reduce the
likelihood of migration of cresols (cresylic acid) from spent solvent
wastes.

    [The proposed technology-based BDAT treatment standard for methyl
ethyl ketone was estimated at the detection limit of <0.100 mg/L based on
incineration (see Table 11, 51 PR 1722).  The principal difference
between the proposed and promulgated treatment standards is the Agency's
change in the criteria for data transfer (see section 5.6.1, page 5-130,
for a discussion of the Agency's methodology for data transfer.)]
                                      5-147

-------
5.6.9    Cyclohexanone (Other Than Wastewater)

    The Agency has no data on TCLP extracts of residue from incineration
of cyclohexanone to use in the derivation of the BDAT treatment
standard.  For reasons presented in Section 5.6.1, EPA used chemical
structure as the basis for transferring treatment data to cyclohexanone
spent solvent wastes other than wastewaters.  Specifically we transferred
treatment data from methyl ethyl ketone because, like cyclohexanone,
methyl ethyl ketone contains the ketone functional group.

    The Agency has data on the analysis of the TCLP extract of
incineration residue for three compounds in the ketones structural
group:  acetone, methyl ethyl ketone, and methyl isobutyl ketone.  To
best account for the range of physical and chemical properties within a
structural group that affect treatment by a specific technology, the
Agency transferred data representing the least stringent treatment
standard from the compounds for which data were available in the ketones
structural group.  The data from which the treatment standard for
incineration of methyl ethyl ketone was derived were transferred to
cyclohexanone.  The treatment standard is 0.75 mg/L based on the
transferred data.

    We believe the BDAT treatment standard for cyclohexanone spent
solvent wastes (other than wastewater) represents substantial treatment.
We would expect untreated cyclohexanone wastes to be similar to untreated
methyl ethyl ketone wastes from which we transferred treatment data since
they are used in many similar manufacturing processes, as shown in
Section 2 of this document.  As discussed on page 5-163, in reference to
methyl ethyl ketone, we believe these constituent reductions
substantially diminish the toxicity of the spent solvent wastes
containing cyclohexanone and substantially reduce the likelihood of
migration of cyclohexanone from spent solvent wastes.

     [The proposed technology-based BDAT treatment standard for
cyclohexanone was estimated at the detection limit of <0.100 mg/L based
on incineration  (see Table 11, 51 FR 1722).  The principal difference
between the proposed and promulgated treatment standards is the Agency's
change in the criteria for data transfer (see Section 5.6.1, page 5-130,
for a discussion of the Agency's methodology for data transfer.)]
                                      5-148

-------
5.6.10   1,2-Dichlorobenzene (Other Than Wastewater)

    The Agency has treatment data for the TCLP extract of incinerator ash
from the treatment of 1,2-dichlorobenzene (Reference 11).  The data are
summarized in Table 5-37.

    The.following steps were taken to derive the BOAT treatment standard
for 1,2-dichlorobenzene:

    1. We evaluated the data to determine whether any of the data
       represented poor design or operation of the incineration system.
       Data from one facility (site 2) were deleted because the Agency
       judged that the system was not properly operated at the time the
       data were collected.  A follow-up sampling visit confirmed the
       Agency's judgment.  The new data were -hot used in the
       determination of the long-term performance average for
       incineration of 1,2-dichlorobenzene; however, the data were used
       to develop a variabiity factor for incineration.  Data from
       another site were deleted because 1,2-dichlorobenzene was reported
       below the detection limits for both the influent and the TCLP
       extract of the ash.

    2. We determined an arithmetic average residue concentration level
       and a variability factor for each data set.  Residue concentration
       levels reported as less than or equal to the reported detection
       limit were set equal to the detection limit for statistical
       analyses.  This is a conservative approach since the actual
       concentration would be between zero and the detection limit.  Two
       residue concentration levels were reported for site 5, one for
       each incinerator at the site.  These were considered as two
       separate data points.

       Process variability could not be calculated from the incineration
       data because only one influent and effluent data pair was
       available for each data set.  Therefore, to account for process
       variability, an average variability factor was calculated for
       incineration, 5.34 (calculation of the average variability factor
       is shown in Table 5-33).

    3. The analysis of variance method was not used to compare different
       treatments of the same waste because data are available for only
       one type of treatment for each waste.

    4. EPA then analyzed the data to determine if the various treatment
       concentration levels shown in Table 5-37 could be associated with
       separate waste treatability subgroups.   Sufficient data did not
       exist to identify separate waste treatability subgroups;
       therefore, one waste treatability subgroup was established for all
                                     5-149

-------
       sources of wastes (other than wastewater)  containing
       1,2-dichlorobenzene spent solvents.   The least stringent treatment
       level within the treatability subgroup was selected for BDAT
       (0.011 mg/L for site 7 obtained by multiplying the variability
       factor by the highest average residue concentration level)  to
       ensure that the standard could be achieved for all waste matrices
       within the waste treatability subgroup.   This calculated standard
       is below the quantification level and could not be used as  the
       treatment standard; therefore, the treatment standard is set at
       the quantification level of 0.125 mg/L.   The technology basis was
       incineration.

    5.  The BDAT treatment standard for 1,2-dichlorobenzene represents
       treatment of a variety of waste matrices incinerated at two
       sites.  The untreated waste concentration of 1,2-dichlorobenzene
       ranged from 92 mg/kg to 1,085 mg/kg in these waste matrices.  All
       of these wastes were treated to the BDAT treatment standard or
       below (0.125 mg/L).  We believe these constituent reductions
       substantially diminish the toxicity of spent solvent wastes
       containing 1,2-dichlorobenzene and substantially reduce the
       likelihood of migration of 1,2-dichlorobenzene from spent solvent
       wastes.

    [The proposed technology-based BDAT treatment standard for
1,2-dichlorobenzene was estimated at the detection limit of <0.010 mg/L
based on incineration (see Table 11, 51 FR 1722).  The difference  between
the proposed and promulgated treated standards is primarily a result of
additional data gathering subsequent to proposal and use of the
analytical quantification level as the treatment standard since the
standard derived from the data is below the EPA published analytical
quantification level for 1,2-dichlorobenzene (see Table 5-1 and the
discussion on page 5-17).  The new data were presented in EPA's Notice of
Availability of Data (51 FR 31783).  In addition, a variability analysis
was incorporated into the development of the treatment standards for
promulgation.]
                                      5-150

-------
                                                            Table 5-37

                                              INCINERATION DATA FOR  1,2-DICHLOROBENZENE
I
I—«
en
                Si te  Type of Incinerator

                  5   Fixed Hearth  (Two
                      Separate Incinera-
                      tion Systems)
                      Fixed Hearth with
                      Secondary Combustor
Wastes Incinerated

(From Furniture Manu-
  facturing Industry)
Solvent Wastes
High-Btu Liquid Wastes
Low-Btu Liquid Wastes
Lacquer-Coated Cardboard

High-Btu Liquids
Low-Btu Liquids
Solids Feed
 Flow-Weighted
    Average
Influent (mg/kg)

      1,085
      1,085
                                                                                           Incinerator
                                                                                           Residue*
 Total
(mg/kg)

 <100
 <100
 TCLP
(ug/L)
Footnotes

    a
    a
         92
                                                                                                      <2
                "Values shown as "<" were reported  as  below indicated  detection limits.

                (a)  The influent concentration is  an  arithmetic average.
                (b)  The influent concentration is  flow-weighted average.

-------
5.6.11   Ethyl Acetate (Other Than Wastewater)

    The Agency has no data on TCLP extracts of  residue from incineration
of ethyl acetate to use in the derivation of the BDAT treatment
standard.  For reasons presented in Section 5.6.1, EPA used chemical
structure as the basis for transferring treatment data to ethyl acetate
spent solvent wastes other than wastewaters.  Specifically we transferred
treatment data from methyl ethyl ketone, which  contains the ketone
functional group, to ethyl acetate, which contains the ester functional
group.  The esters structural group is most structurally similar to the
ketones group based upon their oxygen-containing, electron-releasing
functional groups.

    The Agency has data on the analysis of the  TCLP extract of
incineration residue for three compounds in the ketones structural
group:  acetone, methyl ethyl ketone, and methyl isobutyl ketone.  To
best account for the range of physical and chemical properties within a
structural group that affect treatment by a specific technology, the
Agency transferred data representing the least  stringent treatment
standard from the compounds for which data were available in the ketones
structural group.  The data from which the treatment standard for
incineration of methyl ethyl ketone was derived were transferred to ethyl
acetate.  The treatment standard is 0.75 mg/L based on the transferred
data.

    We believe the BDAT treatment standard for  ethyl acetate spent
solvent wastes (other than wastewater) represents substantial treatment.
We would expect untreated ethyl acetate wastes  to be similar to untreated
methyl ethyl ketone wastes from which we transferred treatment data since
they are used in many similar manufacturing processes, as shown in
Section 2 of this document.  As discussed on page 5-163, in reference to
methyl ethyl ketone, we believe these constituent reductions
substantially diminish the toxicity of the spent solvent wastes
containing ethyl acetate and substantially reduce the likelihood of
migration of ethyl acetate from spent solvent wastes.

     [The proposed technology-based BDAT treatment standard for ethyl
acetate was estimated at the detection limit of <0.100 mg/L based on
incineration (see Table 11, 51 FR 1722).  The principal difference
between the proposed and promulgated treatment  standards is the Agency's
change in the criteria for data transfer (see Section 5.6.1, page 5-130,
for a discussion of the Agency's methodology for data transfer.)]
                                      5-152

-------
5.6.12   Ethylbenzene (Other Than Wastewater)

    The Agency has treatment data for the TCLP extract of incinerator ash
from the treatment of ethylbenzene (Reference 11).  The data are
summarized in Table 5-38.

    The following steps were taken to derive the BOAT treatment standard
for ethylbenzene:

    1. We evaluated the data to determine whether any of the data
       represented poor design or operation of the incineration system.
       Data from one facility (site 6) were deleted because the
       incineration system control devices were not properly designed and
       operated.  The facility is under a Consent Decree to replace and
       improve the current incinerator control- system.  Data from another
       facility (site 2) were deleted because the Agency judged that the
       system was not properly operated at the time the data were
       collected.  A follow-up sampling visit confirmed the Agency's
       judgment.  The new data were not used in the determination of the
       long-term performance average for incineration of ethylbenzene;
       however, the data were used to develop a variability factor for
       incineration.   Data from a third site were deleted because
       ethylbenzene was reported as below the detection limits for both
       the influent and the TCLP extract of the ash.

    2. We determined an arithmetic average residue concentration level
       and a variability factor for each data set.   Residue concentration
       levels reported as less than or equal to the reported detection
       limit were set equal to the detection limit for statistical
       analyses.  This is a conservative approach since the actual
       concentration would be between zero and the detection limit.   Two
       residue concentration levels were reported for site 5,  one for
       each incinerator at the site.   These were considered as two
       separate data points.

       Process variability could not  be calculated from the incineration
       data because only one influent and effluent  data pair was
       available for each data set.   Therefore,  to account for process
       variability, an average variability factor was calculated for
       incineration,  5.34 (calculation of the average variability factor
       is shown in Table 5-33).

    3.  The analysis of variance  method was not used to compare different
       treatments of  the same waste because data are  available for only
       one type of treatment  for each waste.
                                     5-153

-------
    4.  EPA  then analyzed the  data  to  determine  if  the various  treatment
       concentration  levels shown  in  Table  5-38  could be associated with
       separate waste treatability subgroups.   Sufficient  data did not
       exist  to identify separate  waste  treatability subgroups;
       therefore,  one waste treatability subgroup  was established for all
       sources  of  wastes (other  than  wastewater) containing  ethylbenzene
       spent  solvents.   The least  stringent treatment level  within the
       treatability subgroup  was selected for BOAT (0.053  mg/L for site 7
       obtained by multiplying the variability  factor by the highest
       average  residue concentration  level) to  ensure that the standard
       could  be achieved for  all waste matrices  within  the waste
       treatability subgroup.  The technology basis was incineration.

    5.  The  BDAT treatment standard for ethylbenzene represents treatment
       of a variety of waste  matrices incinerated  at five  sites.  The
       untreated waste concentration  of  ethylbenzene ranged  from 780
       mg/kg  to 43,000 rag/kg  in  these waste matrices.   All of  these
       wastes were treated to the  BDAT treatment standard  or below  (0.053
       mg/L).  We  believe these  constituent reductions  substantially
       diminish the toxicity  of  the spent solvent  wastes containing
       ethylbenzene and substantially reduce the likelihood  of migration
       of ethylbenzene from spent  solvent wastes.

    [The proposed  technology-based BDAT  treatment  standard for
ethylbenzene  was estimated at the  detection limit  of <0.010  mg/L based on
incineration  (see  Table 11, 51 FR  1722). The difference between the
proposed and  promulgated treatment standards is primarily  a  result of
additional  data gathering subseguent  to  proposal.   The  new data were
presented in  EPA's Notice of  Availability of Data  (51 FR  31783).   In
addition, a variability analysis was  incorporated  into  the development of
the treatment standards for promulgation.]
                                      5-154

-------
                                                             Table  5-38

                                                  INCINERATION DATA  FOR ETHYLBENZENE
01
 i
tn
tn
                 Site   Type  of  Incinerator

                   3    Rotary Kiln with
                       Secondary  Combustor

                   5    Fixed Hearth  (Two
                       Separate Incinera-
                       tion  Systems)
Fixed Hearth with
Secondary Combustor
                         Flow-Weighted
                            Average
Wastes Incinerated      Influent (mg/kg)

Drum Feed Sol ids              4,048
Liquid Waste Fuel

(From Furniture Manu-           780
  facturing Industry)           780
Solvent Wastes
High-Btu Liquid Wastes
Low-Btu Liquid Wastes
Lacquer-Coated Cardboard

High-Btu Liquids             14,642
Low-Btu Solids
Solids Feed
                                                                     Inci nerator
                                                                      Residue*
                                                                    Total      TCLP
                                                                   (mg/kg)   (ug/L)  Footnotes

                                                                       0.5       2       a
                                                                    <300
                                                                    <300
                                                                                           <300
          <3
          <3
          10       a
                       Rotary Kiln with
                       Secondary  Liquid
                       Injection  Combustor

                       Rotary Ki1n wi th
                       Secondary  Combustor
                        Liquid  Waste  Fuel
                        High-Btu  Liquids
                        Low-Btu  Liquids
                        Solids  Feedc
                             43,000
                              8,591
<1.5      <3       a
<1.5      <3       a
                •Values  shown as  "<" were reported as below indicated detection limits.

                (a)  The  influent concentration is flow-weighted average.
                (b)  The  influent concentration is an arithmetic average.
                (c)  Gel  and filter press residue.

-------
5.6.13   Ethyl Ether (Other Than Wastewater)

    The Agency has no data on TCLP extracts of residue from incineration
of ethyl ether to use in the derivation of the BDAT treatment standard.
For reasons presented in Section 5.6.1, EPA used chemical structure as
the basis for transferring treatment data to ethyl ether spent solvent
wastes other than wastewaters.  Specifically we transferred treatment
data from methyl ethyl ketone, which contains the ketone functional
group, to ethyl ether, which contains the ether functional group.  The
ethers structural group is most structurally similar to the ketones group
based upon their oxygen-containing, electron-releasing functional
groups.

    The Agency has data on the analysis of the TCLP extract of
incineration residue for three compounds in the- ketones structural
group:  acetone, methyl ethyl ketone, and methyl isobutyl ketone.  To
best account for the range of physical and chemical properties within a
structural group that affect treatment by a specific technology, the
Agency transferred data representing the least stringent treatment
standard from the compounds for which data were available in the ketones
structural group.  The data from which the treatment standard for
incineration of methyl ethyl ketone was derived were transferred to ethyl
ether.  The treatment standard is 0.75 mg/L based on the transferred data.

    We believe the BDAT treatment standard for ethyl ether spent solvent
wastes  (other than wastewater) represents substantial treatment.  We
would expect untreated ethyl ether wastes to be similar to untreated
methyl ethyl ketone wastes from which we transferred treatment data.  As
discussed on page 5-163,  in reference to methyl ethyl ketone, we believe
these constituent reductions substantially diminish the toxicity of the
spent solvent wastes containing ethyl ether and substantially reduce the
likelihood of migration of ethyl ether from spent solvent wastes.

     [The proposed technology-based BDAT treatment standard for ethyl
ether was  estimated from  the detection limit of <0.100 mg/L based on
incineration  (see Table 11, 51 FR  1722).  The principal difference
between  the proposed and  promulgated treatment standards  is the Agency's
change  in  the criteria for data transfer  (see Section 5.6.1, page 5-130,
for a  discussion of the Agency's methodology for  data transfer.)]
                                      5-156

-------
5.6.14   Isobutanol (Other Than Wastewater)

    The Agency has no data on TCLP extracts of residue from incineration
of isobutanol to use in the derivation of the BDAT treatment standard.
For reasons presented in Section 5.6.1, EPA used chemical structure as
the basis for transferring treatment data to isobutanol spent solvent
wastes other than wastewaters.  Specifically we transferred treatment
data from methyl ethyl ketone, which contains the ketone functional
group, to isobutanol, which contains the hydroxyl functional group.  The
alcohols structural group is most structurally similar to the ketones
group based upon their oxygen-containing, electron-releasing functional
groups.

    The Agency has data on the analysis of the TCLP extract of
incineration residue for three compounds in the-ketones structural
group:  acetone, methyl ethyl ketone, and methyl isobutyl ketone.  To
best account for the range of physical and chemical properties within a
structural group that affect treatment by a specific technology, the
Agency transferred data representing the least stringent treatment
standard from the compounds for which data were available in the ketones
structural group.  The data from which the treatment standard for
incineration of methyl ethyl ketone was derived were transferred to
isobutanol.  The treatment standard is 0.75 mg/L based on the transferred
data.  The transferred value is below the quantification level for
isobutanol and could not be used as the treatment standard.  Therefore,
the treatment standard is set at the quantification level of 5.0 mg/L.

    We believe the BDAT treatment standard for isobutanol spent solvent
wastes (other than wastewater) represents substantial treatment.  We
would expect untreated isobutanol wastes to be similar to untreated
methyl ethyl ketone wastes from which we transferred treatment data.  As
discussed on page 5-163, in reference to methyl ethyl ketone, we believe
these constituent reductions substantially diminish the toxicity of the
spent solvent wastes containing isobutanol and substantially reduce the
likelihood of migration of isobutanol from spent solvent wastes.

    [The proposed technology-based BDAT treatment standard for isobutanol
was estimated at the detection limit of <0.050 mg/L based on incineration
(see Table 11, 51 FR 1722).   The principal difference between the
proposed and promulgated treatment standards is the Agency's change in
the criteria for data transfer (see Section 5.6.1,  page 5-130, for a
discussion of the Agency's methodology for data transfer.)]
                                      5-157

-------
5.6.15   Methanol (Other Than Wastewater)

    The Agency has no data on TCLP extracts of residue from incineration
of methanol to use in the derivation of the BOAT treatment standard.   For
reasons presented in in Section 5.6.1, EPA used chemical structure as the
basis for transferring treatment data to methanol spent solvent wastes
other than wastewaters.  Specifically we transferred treatment data from
methyl ethyl ketone, which contains the ketone functional group, to
methanol, which contains the hydroxyl functional group.  The alcohols
structural group is most structurally similar to the ketones group based
upon their oxygen-containing, electron-releasing functional groups.

    The Agency has data on the analysis of the TCLP extract of
incineration residue for three compounds in the ketones structural
group:  acetone, methyl ethyl ketone, and methyl isobutyl ketone.  To
best account for the range of physical and chemical properties within a
structural group that affect treatment by a specific technology, the
Agency transferred data representing the least stringent treatment
standard from the compounds for which data were available in the ketones
structural group.  The data from which the treatment standard for
incineration of methyl ethyl ketone was derived were transferred to
methanol.  The treatment standard is 0.75 mg/L based on the transferred
data.

    We believe the BOAT treatment standard for methanol spent solvent
wastes (other than wastewater) represents substantial treatment.  We
would expect untreated methanol wastes to be similar to untreated methyl
ethyl ketone wastes from which we transferred treatment data since they
are used in many similar manufacturing processes, as shown in Section 2
of this document.  As discussed on page 5-163, in reference to methyl
ethyl ketone, we believe these constituent reductions substantially
diminish the toxicity of the spent solvent wastes containing methanol and
substantially reduce the likelihood of migration of methanol from spent
solvent wastes.

     [The proposed technology-based BDAT treatment standard for methanol
was estimated at the detection limit of <0.100 mg/L based on incineration
(see Table 11, 51 FR 1722).  The principal difference between the
proposed and promulgated treatment standards is the Agency's change in
the criteria for data transfer (see Section 5.6.1, page 5-130, for a
discussion of the Agency's methodology for data transfer.)]
                                      5-158

-------
5.6.16   Methylene Chloride (Other Than Wastewater)

    The Agency has treatment data for the TCLP extract of incinerator  ash
from the treatment of methylene chloride (Reference 11).   The data are
summarized in Table 5-39.

    The following steps were taken to derive the BDAT treatment standard
for methylene chloride:

    1. We evaluated the data to determine whether any of  the data
       represented poor design or operation of the incineration system.
       Data from one facility (site 6) were deleted because the
       incineration system control devices were not properly designed and
       operated.  The facility is under a Consent Decree  to replace and
       improve the current incinerator control system.  Data from another
       facility (site 2) were deleted because the Agency judged that the
       system was not properly operated at the time the data were
       collected.  A follow-up sampling visit confirmed the Agency's
       judgment.  The new data were not used in the determination of the
       long-term performance average for incineration of methylene
       chloride; however, the data were used to develop a variability
       factor for incineration.

    2. We determined an arithmetic average residue concentration level
       and a variability factor for each data set.  Residue concentration
       levels reported as less than or egual to the reported detection
       limit were set equal to the detection limit for statistical
       analyses.  This is a conservative approach since the actual
       concentration would be between zero and the detection limit.  Two
       residue concentration levels were reported for site 5, one for
       each incinerator at the site.  These were considered as two
       separate data points.

       Process variability could not be calculated from the incineration
       data because only one influent and effluent data pair was
       available for each data set.  Therefore, to account for process
       variability, an average variability factor was calculated for
       incineration, 5.34 (calculation of the average variability factor
       is showing in Table 5-33).

    3. The analysis of variance method was not used to compare different
       treatments of the same waste because data are available for only
       one type of treatment for each waste.

    4. EPA then analyzed the data to determine if the various treatment
       concentration levels shown in Table 5-39 could be associated with
       separate waste treatability subgroups.  Sufficient data did not
       exist to identify separate waste treatability subgroups;
       therefore, one waste treatability subgroup was established for all

                                      5-159

-------
       sources of wastes (other than wastewater)  containing methylene
       chloride spent  solvents.   The least stringent  treatment  level
       within the treatability subgroup was selected  for BOAT (0.96 mg/L
       for site 5 obtained by multiplying the variability factor by the
       highest average residue concentration level) to ensure that the
       standard could  be achieved for all waste matrices within the waste
       treatability subgroup.   The technology basis was incineration.

    5.  The BOAT treatment standard for methylene  chloride represents
       treatment of a  variety of waste matrices incinerated at  six
       sites.  The untreated waste concentration  of methylene chloride
       ranged from 22  mg/kg to 14,875 mg/kg in these  waste matrices.   All
       of these wastes were treated to the BDAT treatment standard or
       below (0.96 mg/L).  We believe these constituent reductions
       substantially diminish the toxicity of the spent solvent wastes
       containing methylene chloride and substantially reduce the
       likelihood of migration of methylene chloride  from spent solvent
       wastes.

    [The proposed technology-based BDAT treatment standard for  methylene
chloride was estimated at the detection limit of  <0.010 mg/L based on
incineration (see Table 11, 51 FR 1722).  The difference between the
proposed and promulgated treatment standards is primarily a result of
additional data gathering subsequent to proposal. The new data were
presented in EPA's Notice of Availability of Data (51 FR 31783).  In
addition, a variability analysis was incorporated into the development of
the treatment standards for promulgation.]
                                      5-160

-------
                                                            Table 5-39

                                             INCINERATION DATA FOR METHYLENE CHLORIDE
en
Site  Type of Incinerator

  1    Rotary Kiln with
      Secondary Combustor

  3    Rotary Kiln with
      Secondary Combustor

  5    Fixed Hearth (Two
      Separate Incinera-
      tion Systems)
                      Fixed Hearth wi th
                      Secondary Combustor
                                             Wastes Incinerated

                                             PCB Contaminated Dirt
                                             Drum Feed Soli ds
                                             Liquid Waste Fuel
                         Flow-Weighted
                            Average
                       Influent* (mg/kg)

                                 22
                              9,808
(From Furniture Manu-         5,690
  facturing Industry)         5,690
Solvent Wastes
High-Btu Liquid Wastes
Low-Btu Liquid Wastes
Lacquer-Coated Cardboard

High-Btu Liquids               <300
Low-Btu Liquids
Solids Feed
  Incinerator
   Residue*
 Total      TCLP
(mg/kg)    (ug/L)   Footnotes
   <3
                                                                                         <300
                                                                                         <300
                                                                          <300
20
   <1.5      23
             <3
            180
            150
                     Rotary Kiln with
                     Secondary Liquid
                     Injection Combustor
                              Liquid  Waste  Fuel
                              6,600
             26
               "Values  shown as  "<" were reported as below indicated detection limits.

               (a)   Influent is  flow-weighted average.
               (b)   Influent is  an arithmetic average.

-------
                                                       Table  5-39 (Continued)

                                              INCINERATION  DATA  FOR METHYLENE  CHLORIDE
                Site  Type of Incinerator

                  9   Rotary Kiln with
                      Secondary Combustor
                              Wastes Incinerated

                              High-Btu Liquids
                              Low-Btu Liquids
                              Solids Feedb
Flow-Weighted
Average
Influent (ma/ka)
Incinerator
Residue*
Total
TCLP
(ug/L) Footnotes
14,875
100
en
 i
CTi
PO
•Values shown as "<" were reported as below indicated detection limits.

NA - Not Analyzed

(a)  Influent is flow-weighted average.
(b)  Gel and filter press residue.

-------
5.6.17   Methyl Ethyl Ketone (Other Than Wastewater)

    The Agency has treatment data for the TCLP extract of incinerator ash
from the treatment of methyl ethyl ketone (Reference 11).  The data are
summarized in Table 5-40.

    The following steps were taken to derive the BOAT treatment standard
for methyl ethyl ketone:

    1. We evaluated the data to determine whether any of the data
       represented poor design or operation of the incineration system.
       Data from one facility (site 6) were deleted because the
       incineration system control devices were not properly designed and
       operated.  The facility is under a Consent Decree to replace and
       improve the current incinerator control system.  Data from another
       facility (site 2)  were deleted because the Agency judged that the
       system was not properly operated at the time the data were
       collected.  A follow-up sampling visit confirmed the Agency's
       judgment.  The new data were not used in the determination of the
       long-term performance average for incineration of methyl ethyl
       ketone; however, the data were used to develop a variability
       factor for incineration.   Data from a third site were deleted
       because methyl ethyl ketone was reported as below the detection
       limits for both the influent and the TCLP extract of the ash.

    2. We determined an arithmetic average residue concentration level
       and a variability factor for each data set.   Residue concentration
       levels reported as  less than or equal to the reported detection
       limit were set equal to the detection limit for statistical
       analyses.  This is  a conservative approach since the actual
       concentration would be between zero and the detection limit.   Two
       residue concentration levels were reported for site  5,  one for
       each incinerator at the site.   These were considered as two
       separate data points.

       Process variability could not be calculated from the incineration
       data because only one influent and effluent data pair was
       available for each  data set.   Therefore,  to account  for process
       variability, an average variability factor was calculated for
       incineration,  5.34  (calculation of the average variability factor
       is shown in Table 5-33).

    3.  The analysis of variance  method was not used to compare different
       treatments of the same waste  because data are available for only
       one type of treatment  for each waste.
                                     5-163

-------
    4. EPA then analyzed the data to determine if the various treatment
       concentration levels shown in Table 5-40 could be associated with
       separate waste treatability subgroups.   Sufficient data did not
       exist to identify separate waste treatability subgroups;
       therefore, one waste treatability subgroup was established for all
       sources of wastes (other than wastewater) containing methyl ethyl
       ketone spent solvents.  The least stringent treatment level within
       the treatability subgroup was selected for BOAT (0.59 mg/L for
       site 7 obtained by multiplying the variability factor by the
       highest average residue concentration level) to ensure that the
       standard could be achieved for all waste matrices within the waste
       treatability subgroup.  The technology basis was incineration.

    5. The BDAT treatment standard for methyl ethyl ketone represents
       treatment of a variety of waste matrices incinerated at four
       sites.  The untreated waste concentration of methyl ethyl ketone
       ranged from 28,165 mg/kg to 110,000 mg/kg in these waste
       matrices.  All of these wastes were treated to the BDAT treatment
       standard or below (0.75 mg/L).  We believe these constituent
       reductions substantially diminish the toxicity of the spent
       solvent wastes containing methyl ethyl ketone and substantially
       reduce the likelihood of migration of methyl ethyl ketone from
       spent solvent wastes.

    [The proposed BDAT technology-based treatment standard for methyl
ethyl ketone was estimated at the detection limit of <0.050 mg/L based on
incineration (see Table 11, 51 FR 1722).  The difference between the
proposed and promulgated treatment standards is primarily a result of
additional data gathering subseguent to proposal.  The new data were
presented in EPA's Notice of Availability of Data (51 FR 31783).  In
addition, a variability analysis was incorporated into the development of
the treatment standards for promulgation.]
                                      5-164

-------
                                            Table  5-40

                             INCINERATION DATA  FOR METHYL  ETHYL  KETONE
Si te  Type of Incinerator

  3   Rotary Kiln  with
      Secondary Combustor

  5   Fixed Hearth (Two
      Separate Incinera-
      tion Systems)
      Fixed Hearth  with
      Secondary Combustor
      Rotary Kiln with
      Secondary Liquid
      Injection Combustor
                         Flow-Weighted
                            Average
Wastes Incinerated      Influent (mg/kg)

Drum Feed Solids            100,000
Liquid Waste Fuel

(From Furniture Manu-        28,165
  facturing Industry)        28,165
Solvent Wastes
High-Btu Liquid Wastes
Low-Btu Liquid Wastes
Lacquer-Coated Cardboard

High-Btu Liquids             35,000
Low-Btu Liquids
Solids Feed

Liquid Waste Fuel           110,000
                                                                           Inci nerator
                                                                            Resi due*
<300
<300
<300
          TCLP
         (ug/L)  Footnotes

            <3       a
 <3
 25
140
            <3
'Values shown as "<"  were  reported  as  below indicated  detection limits.

(a)  The influent concentration  is  flow-weighted average.
(b)  The influent concentration  is  an  arithmetic average.

-------
5.7.18   Methyl Isobutyl Ketone (Other Than Wastewater)

    The Agency has treatment data for the TCLP extract of incinerator ash
from the treatment of methyl isobutyl ketone (Reference 11).   The data
are summarized in Table 5-41.

    The following steps were taken to derive the BDAT treatment standard
for methyl isobutyl ketone:

    1. We evaluated the data to determine whether any of the  data
       represented poor design or operation of the incineration system.
       Data from one facility (site 6) were deleted because the
       incineration system control devices were not properly  designed and
       operated.  The facility is under a Consent Decree to replace and
       improve the current incinerator control system.  Data  from another
       facility (site 2) were deleted because the Agency judged that the
       system was not properly operated at the time the data  were
       collected.  A follow-up sampling visit confirmed the Agency's
       judgment.  The new data were not used in the determination of the
       long-term performance average for incineration of methyl isobutyl
       ketone; however, the data were used to develop a variability
       factor for incineration.

    2. We determined an arithmetic average residue concentration level
       and a variability factor for each data set.  Residue concentration
       levels reported as less than or equal to the reported  detection
       limit were set equal to the detection limit for statistical
       analyses.  This is a conservative approach since the actual
       concentration would be between zero and the detection  limit.   Two
       residue concentration levels were reported for site  5,  one for
       each incinerator at the site.   These were considered as two
       separate data points.

       Process variability could not be calculated from the incineration
       data because only one influent and effluent data pair  was
       available for each data set.   Therefore, to account  for process
       variability, an average variability factor was calculated for
       incineration, 5.34 (calculation of the average variability factor
       is shown in Table 5-33).

    3. The analysis of variance  method was not used to compare different
       treatments of the same waste because the data are available for
       only one type of treatment for each waste.

    4. EPA then analyzed the data to determine if the various  treatment
       concentration levels shown in Table 5-41 could be associated with
       separate waste treatability subgroups.  Sufficient data did not
                                     5-166

-------
       exist to identify separate waste treatability subgroups;
       therefore, one waste treatability subgroup was established for all
       sources of wastes (other than wastewater)  containing methyl
       isobutyl ketone spent solvents.  The least stringent treatment
       level within the treatability subgroup was selected for BDAT (0.33
       mg/L for site 7 obtained by multiplying the variability factor by
       the highest average residue concentration  level) to ensure that
       the standard could be achieved for all waste matrices within the
       waste treatability subgroup.  The technology basis was
       incineration.

    5. The BDAT treatment standard for methyl isobutyl ketone represents
       treatment of a variety of waste matrices incinerated at six
       sites.  The untreated waste concentration  of methyl isobutyl
       ketone ranged from 15 mg/kg to 32,000 mg/kg-in these waste
       matrices.  All of these wastes were treated to the BDAT treatment
       standard or below (0.33 mg/L).  We believe these constituent
       reductions substantially diminish the toxicity of the spent
       solvent wastes containing methyl isobutyl  ketone and substantially
       reduce the likelihood of migration of methyl isobutyl ketone from
       spent solvent wastes.

    [The proposed BDAT technology-based treatment standard for methyl
isobutyl ketone was estimated at the detection limit of <0.010 mg/L based
on incineration (see Table 11, 51 FR 1722).  The  difference between the
proposed and promulgated treatment standards is primarily a result of
additional data gathering subsequent to proposal.  The new data were
presented in EPA's Notice of Availability of Data (51 FR 31783).  In
addition, a variability analysis was incorporated into the development of
the treatment standards for promulgation.]
                                     5-167

-------
                                                                Table  5-41

                                                INCINERATION DATA  FOR METHYL  ISOBUTYL  KETONE
oo
                    Si te   Type  of  Incinerator

                      1    Rotary Kiln  with
                          Secondary  Combustor

                      3    Rotary Kiln  with
                          Secondary  Combustor

                      5    Fixed Hearth  (Two
                          Separate Incinera-
                          tion  Systems)
                         Fixed Hearth with
                         Secondary Combustor
Wastes Incinerated
Flow-Weighted
Average
Influent (mg/kg)
Incinerator
Residue*
Total TCLP
(mg/kg) (ug/L) Footnotes
PCB Contaminated Dirt             15
Drum Feed Solids             30,000
Liquid Waste Fuel

(From Furniture Manu-           315
  facturing Industry)           315
Solvent Wastes
High-Btu Liquid Wastes
Low-Btu Liquid Wastes
Lacquer-Coated Cardboard

High-Btu Liquids             10,818
Low-Btu Liquids
Solids Feed
                  <2
               <200
               <200
               <200
<2
                            <2
<2
<2
62
                         Rotary Kiln with
                         Secondary Liquid
                         Injection Combustor
Liquid Waste Fuel
32,000
<2
                   "Values shown as "<" were reported as below indicated detection limits.

                   (a)  Influent concentration is flow-weighted average.
                   (b)  Influent concentration is an arithmetic average.

-------
                                                      Table  5-41  (Continued)

                                            INCINERATION DATA FOR  METHYL  ISOBUTYL  KETONE
                Site  Type of Incinerator

                  9   Rotary Kiln with
                      Secondary Combustor
Wastes Incinerated

High-Btu Liquids
Low-Btu Liquids
Solids Feedb
 Flow-Weighted
    Average
Influent (mg/kg)

     24,905
  Incinerator
   Residue*
 Total      TCLP
(mg/kg)    (ug/L)   Footnotes

   <1.0       <2       a
en
i
                "Values shown as "<" were reported  as  below  indicated  detection limits.

                (a)  Influent concentration is  flow-weighted  average.
                (b)  Gel and filter press residue.

-------
5.6.19   Nitrobenzene (Other Than Wastewater)

    The Agency has treatment data for the TCLP extract of incinerator ash
from the treatment of nitrobenzene (Reference 11).   The data are
summarized in Table 5-42.

    The following steps were taken to derive the BOAT treatment standard
for nitrobenzene:

    1. We evaluated the data to determine whether any of the data
       represented poor design or operation of the  incineration system.
       The available data and information did not show any of the data to
       represent poor design and operation.   Accordingly, none of the
       data were deleted on this basis.

    2. We determined an arithmetic average residue  concentration level
       and a variability factor for each data set.   Residue concentration
       levels reported as less than or equal to the reported detection
       limit were set equal to the detection limit  for statistical
       analyses.  This is a conservative approach since the actual
       concentration would be between zero and the  detection limit.

       Process variability could not be  calculated  from the incineration
       data because only one influent and effluent  data pair was
       available for each data set.   Therefore, to  account for process
       variability, an average variability factor was calculated for
       incineration, 5.34 (calculation of the average variability factor
       is shown in Table 5-33).

    3. The analysis of variance method was not used to compare different
       treatments of the same waste  because data are available for  only
       one type of treatment for each waste.

    4. EPA then analyzed the data to determine if the various treatment
       concentration levels shown in Table 5-42 could be associated  with
       separate waste treatability subgroups.   Sufficient data did not
       exist to identify separate waste  treatability subgroups;
       therefore, one waste treatability subgroup was established for all
       sources of wastes (other than wastewater)  containing nitrobenzene
       spent solvents.   The least stringent  treatment level within the
       treatability subgroup was selected for BDAT  (0.011 mg/L for site  7
       obtained by multiplying the variability factor by the highest
       average residue concentration level)  to ensure that the standard
       could be achieved for all waste matrices within the waste
       treatability subgroup.   This   calculated standard is below the
       quantification level and could not be used as the treatment
       standard; therefore, the treatment standard  is set at the
       quantification level of 0.125 mg/L.   The technology was
       incineration.

                                      5-170

-------
    5. The BDAT treatment standard for nitrobenzene represents treatment
       of a variety of waste matrices incinerated at one site.  The
       untreated waste concentration of nitrobenzene was as high as 79
       mg/kg in these waste matrices.  This waste was treated to a
       concentration below the BDAT treatment standard (0.125 mg/L).  We
       believe these, constituent reductions substantially diminish the
       toxicity of the spent solvent wastes containing nitrobenzene and
       substantially reduce the likelihood of migration of nitrobenzene
       from spent solvent wastes.

    [The proposed BDAT technology-based treatment standard for
nitrobenzene was estimated at the  detection limit of <0.010 mg/L based on
incineration (see Table 11, 51 FR  1722).  The difference between the
proposed and promulgated treatment standards is primarily a result of
additional data gathering subsequent to proposal and use of the
analytical quantification level as the treatment standard since the
standard derived from the data is  below the EPA published analytical
quantification level for nitrobenzene (see Table 5-1 and the discussion
on page 5-17).  The new data were  presented in EPA's Notice of
Availability of Data (51 FR 31783).  In addition, a variability analysis
was incorporated into the development of the treatment standards for
promulgation.]
                                      5-171

-------
                                                             Table 5-42

                                                 INCINERATION DATA FOR NITROBENZENE
                Site  Type of Incinerator

                  7   Fixed Hearth with
                      Secondary Combustor
                              Wastes  Incinerated

                              High-Btu  Liquids
                              Low-Btu Liquids
                              Solids  Feed
 Flow-Weighted
    Average
Influent (mg/kg)

         79
                                                                                           Incinerator
                                                                                            Residue*
 TCLP
(ug/L)   Footnotes

   <2       a
en
 i
"Value shown as "<" were reported  as  below  indicated  detection  limits.

(a)  The influent concentration is flow-weighted  average.
t\)

-------
5.6.20   Pyridine (Other Than Wastewater)

    The Agency has no data on TCLP extracts of residue from incineration
of pyridine to use in the derivation of the BOAT treatment standard.  For
reasons presented in Section 5.6.1, EPA used chemical structure as the
basis for transferring treatment data to pyridine spent solvent wastes
other than wastewaters.  Specifically we transferred treatment data from
toluene to pyridine; both which contain the aromatic ring functional
group.

    The Agency has data on the analysis of the TCLP extract of
incineration residue for four compounds in the aromatics structural
group:  ethylbenzene, toluene, xylene, and nitrobenzene.  To best account
for the range of physical and chemical properties within a structural
group that affect treatment by a specific technology; the Agency
transferred data representing the least stringent treatment standard from
the compounds for which data were available in the aromatics structural
group.  The data from which the treatment standard for incineration of
toluene was derived were transferred to pyridine.  The treatment standard
is 0.33 mg/L based on the transferred data.

    We believe the BDAT treatment standard for pyridine spent solvent
wastes (other than wastewater) represents substantial treatment.  We
would expect untreated pyridine wastes to be similar to untreated toluene
wastes from which we transferred treatment data since they are used in
some of the same manufacturing processes, as shown in Section 2 of this
document.  As discussed on page 5-177 in reference to toluene, we believe
these constituent reductions substantially diminish the toxicity of the
spent solvent wastes containing pyridine and substantially reduce the
likelihood of migration of pyridine from spent solvent wastes.

    [The proposed technology based BDAT treatment standard for pyridine
was estimated at the detection limit of <0.500 mg/L based on incineration
(see Table 11, 51 FR 1722).  The principal difference between the
proposed and promulgated treatment standards is the Agency's change in
the criteria for data transfer (see Section 5.6.1, page 5-130, for a
discussion of the Agency's methodology for data transfer.)]
                                      5-173

-------
5.6.21   Tetrachloroethylene (Other Than Wastewater)

    The Agency has treatment data for the TCLP extract of incinerator ash
from the treatment of tetrachloroethylene (Reference 11).  The data are
summarized in Table 5-43.

    The following steps were taken to derive the BOAT treatment standard
for tetrachloroethylene:

    1. We evaluated the data to determine whether any of the data
       represented poor design or operation of the incineration system.
       Data from one facility (site 6) were deleted because the
       incineration system control devices were not properly designed and
       operated.  The facility is under a Consent Decree to replace and
       improve the current incinerator control system.  Data from another
       facility (site 2) were deleted because the Agency judged that the
       system was not properly operated at the time the data were
       collected.  A follow-up sampling visit confirmed the Agency's
       judgment.  The new data were not used in the determination of the
       long-term performance average for incineration of
       tetrachloroethylene; however, the data were used to develop a
       variability factor for incineration.  Data from a third site were
       deleted because tetrachloroethylene was reported as below the
       detection limits for both the influent and the TCLP extract of the
       ash.

    2. We determined an arithmetic average residue concentration level
       and a variability factor for each data set.  Residue concentration
       levels reported as less than or egual to the reported detection
       limit were set equal to the detection limit for statistical
       analyses.  This is a conservative approach since the actual
       concentration would be between zero and the detection limit.

       Process variability could not be calculated from the incineration
       data because only one influent and effluent data pair was
       available for each data set.  Therefore, to account for process
       variability, an average variability factor was calculated for
       incineration, 5.34 (calculation of the average variability factor
       is shown in Table 5-33).

    3. The analysis of variance method was not used to compare different
       treatments of the same waste because data are available for only
       one type of treatment for each waste.

    4. EPA then analyzed the data to determine if the various treatment
       concentration levels shown in Table 5-43 could be associated with
       separate waste treatability subgroups.  Sufficient data did not
       exist to identify separate waste treatability subgroups;
       therefore, one waste treatability subgroup was established for all

                                      5-174

-------
       sources of wastes (other than wastewater)  containing
       tetrachloroethylene spent solvents.   The least stringent treatment
       level within the treatability subgroup was selected for BDAT
       (0.016 mg/L for sites 1, 3,  8, and 9 obtained by multiplying the
       variability factor by the highest average  residue concentration
       level) to ensure that the standard could be achieved for all waste
       matrices within the waste treatability subgroup.  This calculated
       value is below the quantification level and could not be used as
       the treatment standard;  therefore, the treatment standard is set
       at the quantification level  of 0.05 mg/L.   The technology basis
       was incineration.

    5.  The BDAT treatment standard  for tetrachloroethylene represents
       treatment of a variety of waste matrices incinerated at four
       sites.  The untreated waste  concentration of tetrachloroethylene
       ranged from 4 mg/kg to 17,000 mg/kg in these waste matrices.  All
       of these wastes were treated to the BDAT treatment standard or
       below (0.05 mg/kg).  We believe these constituent reductions
       substantially diminish the toxicity of the spent solvent wastes
       containing tetrachloroethylene and substantially reduce the
       likelihood of migration of tetrachloroethylene from spent solvent
       wastes.

    [The proposed BDAT technology-based treatment standard for
tetrachloroethylene was estimated at the detection limit of <0.010 mg/L
based on incineration (see Table 11, 51 FR 1722).  The difference between
the proposed and promulgated treatment standards is primarily a result of
additional data gathering subsequent to proposal  and use of the
analytical quantification level as  the treatment standard since the
standard derived from the data is below the EPA published analytical
quantification level for tetrachloroethylene (see Table 5-1 and the
discussion on page 5-17).  The new data were presented in EPA's Notice of
Availability of Data (51 FR 31783).  In addition, a variability analysis
was incorporated into the development of the treatment standards for
promulgation.]
                                      5-175

-------
                                             Table 5-43

                              INCINERATION DATA FOR TETRACHLOROETHYLENE
Si te  Type of Incinerator

  1    Rotary Kiln with
      Secondary Combustor

  3    Rotary Kiln with
      Secondary Combustor

  8    Rotary Kiln with
      Secondary Liquid
      Injection Combustor

  9    Rotary Kiln with
      Secondary Combustor
Wastes Incinerated

PCB Contaminated Dirt
Drum Feed Solids
Liquid Waste Fuel

Liquid Waste Fuel
High-Btu Liquids
Low-Btu Liquids
Solids Feedb
 Flow-Weighted
    Average
Influent (ma/kg)
        256
     17,000
        466
  Inci nerator
   Residue*
 Total      TCLP
(mg/kg)    (ug/L)   Footnotes

   <3        <3       a
             <3
             <3
             <3
"Values shown as "<" were reported  as  below  indicated  detection  limits.

(a)   The influent concentration is  flow-weighted  average.
(b)   Gel and filter press residue.

-------
5.6.22   Toluene (Other Than Wastewater)

    The Agency has treatment data for the TCLP extract of incinerator ash
from the treatment of toluene (Reference 11).  The data are summarized in
Table 5-44.

    The followiag steps were taken to derive the BDAT treatment standard
for toluene:

    1. We evaluated the data to determine whether any of the data
       represented poor design or operation of the incineration system.
       Data from one facility (site 6) were deleted because the
       incineration system control devices were not properly designed and
       operated.  The facility is under a Consent Decree to replace and
       improve the current incinerator control system.  Data from another
       facility (site 2) were deleted because the Agency judged that the
       system was not properly operated at the time the data were
       collected.  A follow-up sampling visit confirmed the Agency's
       judgment.  The new data were not used in the determination of the
       long-term performance average for incineration of toluene;
       however, the data were used to develop a variability factor for
       incineration.  Data from a third site were deleted because toluene
       was reported as below the detection limits for both the influent
       and the TCLP extract of the ash.

    2. We determined an arithmetic average residue concentration level
       and a variability factor for each data set.  Residue concentration
       levels reported as less than or equal to the reported detection
       limit were set equal to the detection limit for statistical
       analyses.  This is a conservative approach since the actual
       concentration would be between zero and the detection limit.  Two
       residue concentration levels were reported for site 5, one for
       each incinerator at the site.  These were considered as two
       separate data points.

       Process variability could not be calculated from the incineration
       data because only one influent and effluent data pair was
       available for each data set.  Therefore, to account for process
       variability, an average variability factor was calculated for
       incineration, 5.34 (see Table 5-33).

    3. The analysis of variance method was not used to compare different
       treatments of the same waste because data are available for only
       one type of treatment for each waste.

    4. EPA then analyzed the data to determine if the various treatment
       concentration levels shown in Table 5-44 could be associated with
       separate waste treatability subgroups.  Sufficient data did not
       exist to identify separate waste treatability subgroups;

                                      5-177

-------
       therefore, one waste treatability subgroup was established for all
       sources of wastes (other than wastewater)  containing toluene spent
       solvents.   The least stringent treatment level within the
       treatability subgroup was selected for BOAT (0.33 mg/L for site 7
       obtained by multiplying the variability factor by the highest
       average residue concentration level) to ensure that the standard
       could be achieved for all waste matrices within the waste
       treatability subgroup.  The technology basis was incineration.

    5. The BDAT treatment standard for toluene represents treatment of a
       variety of waste matrices incinerated at six sites.  The untreated
       waste concentration of toluene ranged from 3 mg/kg to 100,357
       mg/kg in these waste matrices.  All of these wastes were treated
       to the BDAT treatment standard or below (0.33 mg/L).   We believe
       these constituent reductions substantially diminish the toxicity
       of the spent solvent wastes containing toluene and substantially
       reduce the likelihood of migration of toluene from spent solvent
       wastes.

    [The proposed BDAT technology-based treatment standard for toluene
was estimated at the detection limit of <0.010 mg/L based on incineration
(see Table 11, 51 FR 1722).  The difference between the proposed and
promulgated treatment standards is primarily a result of additional data
gathering subseguent to proposal.  The new data were presented in EPA's
Notice of Availability of Data (51 FR 31783).  In addition,  a variability
analysis was incorporated into the development of the treatment standards
for promulgation.]
                                      5-178

-------
                                             Table 5-44

                                    INCINERATION DATA FOR TOLUENE
Site  Type of Incinerator

  1   Rotary Kiln with
      Secondary Combustor

  3   Rotary Kiln with
      Secondary Combustor

  5   Fixed Hearth (Two
      Separate Incinera-
      tion Systems)
      Fixed Hearth with
      Secondary Combustor
Wastes Incinerated

PCB Contaminated Dirt
 Flow-Weighted
    Average
Influent (mo/kg)
Drum Feed Solids             38,057
Liquid Waste Fuel

(From Furniture Manu-        12,743
  facturing Industry)        12,743
Solvent Wastes
High-Btu Liquid Wastes
Low-Btu Liquid Wastes
Lacquer-Coated Cardboard

High-Btu Liquids              9,562
Low-Btu Liquids
Solids Feed
  Incinerator
   Residue*
 Total      TCLP
(mg/kg)    (ug/L)  Footnotes

   <3         6       a
                       2.5      27
                    <300
                    <300
                    <300
             <3
              7
             61
      Rotary Kiln with
      Secondary Liquid
      Injection Combustor
Liquid Waste Fuel
     43,000
    2.1
13
"Values shown as "<" were reported  as  below indicated detection limits.

(a)   The influent concentration  is  flow-weighted  average.
(b)   The influent concentration  is  an  arithmetic  average.

-------
                                                        Table  5-44  (Continued)

                                                     INCINERATION DATA FOR TOLUENE
                                                                                            Incinerator
                                                                        Flow-Weighted         Residue*
                                                                           Average          Total      TCLP
                 Si te  Type of Incinerator     Wastes  Incinerated       Influent  (mg/kg)    (mg/kg)    (ug/L)   Footnotes

                   9   Rotary Kiln with        High-Btu  Liquids             100,357            <1.5       <3       a
                       Secondary Combustor     Low-Btu  Liquids
                                               Solids  Feedb
                 'Values shown as "<" were reported  as  below  indicated  detection  limits.

                 (a)  The influent concentration is  flow-weighted  average.
                 (b)  Gel and filter press residue.
00
O

-------
5.6.23   1,1,1-Trichloroethane (Other Than Wastewater)

    The Agency has treatment data for the TCLP extract of incinerator ash
from the treatment of 1,1,1-trichloroethane (Reference 11).   The data are
summarized in Table 5-45.

    The following steps were taken to derive the BOAT treatment standard
for 1,1,1-trichloroethane:

    1. We evaluated the data to determine whether any of the data
       represented poor design or operation of the incineration system.
       Data from two sites were deleted because 1,1,1-trichloroethane was
       reported as below the detection limits for both the influent and
       the TCLP extract of the ash.

    2. We determined an arithmetic average residue concentration level
       and a variability factor for each data set.  Residue  concentration
       levels reported as  less than or equal to the reported detection
       limit were set equal to the detection limit for statistical
       analyses.  This is  a conservative approach since the  actual
       concentration would be between zero and the detection limit.  Two
       residue concentration levels were reported for site 5, one for
       each incinerator at the site.  These were considered  as two
       separate data points.

       Process variability could not be calculated from the  incineration
       data because only one influent and effluent data pair was
       available for each data set.   Therefore, to account for process
       variability, an average variability factor was calculated for
       incineration, 5.34  (calculation of the average variability factor
       is shown in Table 5-33).

    3. The analysis of variance method was not used to compare different
       treatments of the same waste because data are available for only
       one type of treatment for each waste.

    4. EPA then analyzed the data to determine if the various treatment
       concentration levels shown in Table 5-45 could be associated with
       separate waste treatability subgroups.   Sufficient data did not
       exist to identify separate waste treatability subgroups;
       therefore, one waste treatability subgroup was established for all
       sources of wastes (other than wastewater) containing
       1,1,1-trichloroethane spent solvents.  The least stringent
       treatment level within the treatability subgroup was  selected for
                                     5-181

-------
       BDAT (0.41 mg/L for site 7 obtained by multiplying the variability
       factor by the highest average residue concentration level) to
       ensure that the standard could be achieved for all waste matrices
       within the waste treatability subgroup.   The technology basis was
       incineration.

    5.  The BDAT treatment standard for 1,1,1-trichloroethane represents
       treatment of a variety of waste matrices incinerated at five
       sites.  The untreated waste concentration of 1,1,1-trichloroethane
       ranged from 463 mg/kg to 29,000 mg/kg in these waste matrices.
       All of these wastes were treated to the BDAT treatment standard or
       below (0.41 mg/L).  We believe these constituent reductions
       substantially diminish the toxicity of the spent solvent wastes
       containing 1,1,1-trichloroethane and substantially reduce the
       likelihood of migration of 1,1,1-trichloroethane-from spent
       solvent wastes.

    [The proposed technology-based BDAT treatment standard for
1,1,1-trichloroethane was estimated at the detection limit of <0.010 mg/L
based on incineration (see Table 11, 51 FR 1722).  The difference between
the proposed and promulgated treatment standards is primarily a result of
additional data gathering subsequent to proposal.  The new data were
presented in EPA's Notice of Availability of Data (51 FR 31783).  In
addition, a variability analysis was incorporated into the development of
the treatment standards for promulgation.]
                                     5-182

-------
                                                               Table  5-45

                                               INCINERATION DATA  FOR  1 ,1 ,1-TRICHLOROETHANE
oo
co
                   Si te   Type  of  Incinerator

                     3    Rotary Kiln  with
                         Secondary  Combustor

                     5    Fixed Hearth (Two
                         Separate Incinera-
                         tion  Systems)
Fixed Hearth with
Secondary Combustor
                         Flow-Wei ghted
                            Average
Wastes Incinerated      Influent (mg/kg)

Drum Feed Solids             29,000
Liquid Waste Fuel

(From Furniture Manu-           463
  factoring Industry)           463
Solvent Wastes
High-Btu Liquid Wastes
Low-Btu Liquid Wastes
Lacquer-Coated Cardboard

High-Btu Liquids              1,920
Low-Btu Liquids
Solids Feed
                                                                                              Incinerator
                                                                                               Residue*
                                                                    Total
                                                                   (mg/kg)
                                                                    <300
                                                                    <300
<300
          TCLP
         (ug/L)  Footnotes

            <3       a
            <3
             3
77
                     8    Rotary  Kiln  with
                         Secondary  Liquid
                         Injection  Combustor

                     9    Rotary  Kiln  with
                         Secondary  Combustor
                        Liquid  Waste  Fuel
                        High-Btu  Liquids
                        Low-Btu  Liquids
                        Solids  Feedc
                             10,000
                             15,792
            <3
            <3
                   'Values  shown  as  "<" were  reported as below  indicated detection  limits.

                   (a)   The influent  concentration is flow-weighted average.
                   (b)   The influent  concentration is an arithmetic average.
                   (c)   Gel  and  filter press  residue.

-------
5.6.24   l,l,2-Trichloro-l,2,2-trifluoroethane (Other Than Wastewater)

    The Agency has no data on TCLP extracts of residue from incineration
of l,l,2-trichloro-l,2,2-trifluoroethane to use in the derivation of the
BDAT treatment standard.  For reasons presented in Section 5.6.1 EPA used
chemical structure as the basis for transferring treatment data to
1,1,2-trichloro-l,2,2-trifluoroethane spent solvent wastes other than
wastewaters.  Specifically we transferred treatment data from methylene
chloride to 1,1,2-trichloro-l,2,2-trifluoroethane; both contain the
halogen functional group.

    The Agency has data on the analysis of the TCLP extract of
incineration residue for two compounds in the halogenated aliphatics
structural group:  methylene chloride and 1,1,1-trichloroethane.  To best
account for the range of physical and chemical properties within a
structural group that affect treatment by a specific technology, the
Agency transferred data representing the least stringent treatment
standard from the compounds for which data were available in the
halogenated aliphatics structural group.  The data from which the
treatment standard for incineration of methylene chloride was derived
were transferred to 1,1,2-trichloro-l,2,2-trifluoroethane.  The treatment
standard is 0.96 mg/L based on the transferred data.

    We believe the BDAT treatment standard for 1,1,2-trichloro-
1,2,2-trifluoroethane spent solvent wastes (other than wastewater)
represents substantial treatment.  We would expect untreated
1,1,2-trichloro-l,2,2-trifluoroethane wastes to be similar to untreated
methylene chloride wastes from which we transferred treatment data since
they are used in many similar manufacturing processes, as shown in
Section 2 of this document.  As discussed on page 5-159, in reference to
methylene chloride, we believe these constituent reductions to
substantially diminish the toxicity of the spent solvent wastes
containing 1,1,2-trichloro-l,2,2-trifluoroethane and substantially reduce
the likelihood of migration of 1,1,2-trichloro-l,2,2-trifluoroethane from
spent solvent wastes.

    [The proposed BDAT technology-based treatment standard for
1,1,2-trichloro-l,2,2-trifluoroethane was estimated at the detection
limit of <0.010 mg/L based on incineration (see Table 11, 51 FR 1722).
The principal difference between the proposed and promulgated treatment
standards is the Agency's change in the criteria for data transfer (see
Section 5.6.1, page 5-130, for a discussion of the Agency's methodology
for data transfer.)]
                                      5-184

-------
5.6.25   Trichloroethylene (Other Than Wastewater)

    The Agency has treatment data for the TCLP extract of incinerator ash
from the treatment of trichloroethylene (Reference 11).  The data are
summarized in Table 5-46.

    The following steps were taken to derive the BDAT treatment standard
for trichloroethylene:

    1. We evaluated the data to determine whether any of the data
       represented poor design or operation of the incineration system.
       Data from one facility (site 2) were deleted because the Agency
       judged that the system was not properly operated at the time the
       data were collected.  A follow-up sampling visit confirmed the
       Agency's judgment.  The new data were not used in-the
       determination of the long-term performance average for
       incineration of trichloroethylene; however, the data were used to
       develop a variability factor for incineration.  Data from two
       other sites were deleted because trichloroethylene was reported as
       below the detection limits for both the influent and the TCLP
       extract of the ash.

    2. We determined an arithmetic average residue concentration level
       and a variability factor for each data set.  Residue concentration
       levels reported as less than or equal to the reported detection
       limit were set equal to the detection limit for statistical
       analyses.  This is a conservative approach since the actual
       concentration would be between zero and the detection limit.
       Process variability could not be calculated from the incineration
       data because only one influent and effluent data pair was
       available for each data set.  Therefore, to account for process
       variability, an average variability factor was calculated for
       incineration, 5.34 (calculation of the average variability factor
       is shown in Table 5-33).

    3. The analysis of variance method was not used to compare different
       treatments of the same waste because data are available for only
       one type of treatment for each waste.

    4. EPA then analyzed the data to determine if the various treatment
       concentration levels shown in Table 5-46 could be associated with
       separate waste treatability subgroups.  Sufficient data did not
       exist to identify separate waste treatability subgroups;
       therefore, one waste treatability subgroup was established for all
       sources of wastes (other than wastewater) containing
       trichloroethylene spent solvents.  The least stringent treatment
                                      5-185

-------
       level within the treatability subgroup was selected for BDAT
       (0.091 mg/L for site 7 obtained by multiplying the variability
       factor by the highest average residue concentration level) to
       ensure that the standard could be achieved for all waste matrices
       within the waste treatability subgroup.  The technology basis was
       incineration.

    5.  The BDAT treatment standard for trichloroethylene represents
       treatment of a variety of waste matrices incinerated at three
       sites.  The untreated waste concentration of trichloroethylene
       ranged from 1,009 mg/kg to 4,700 mg/kg in these waste matrices.
       All of these wastes were treated to the BDAT treatment standard or
       below (0.091 mg/L).  We believe these constituent reductions
       substantially diminish the toxicity of the spent solvent wastes
       containing trichloroethylene and substantially reduce the
       likelihood of migration of trichloroethylene from spent solvent
       wastes.

    [The proposed BDAT technology-based treatment standard for
trichloroethylene was estimated at the detection limit of <0.010 mg/L
based on incineration (see Table 11, 51 FR 1722).  The difference between
the proposed and promulgated treatment standards is primarily a result of
additional data gathering subsequent to proposal.  The new data were
presented in EPA's Notice of Availability of Data (51 FR 31783).  In
addition, a variability analysis was incorporated into the development of
the treatment standards for promulgation.]
                                      5-186

-------
                                                             Table 5-46

                                               INCINERATION DATA FOR TRICHLOROETHYLENE
en
HJ
CO
                 Site   Type  of  Incinerator

                   7    Fixed Hearth with
                       Secondary  Combustor
                   8    Rotary  Kiln with
                       Secondary  Liquid
                       Injection  Combustor

                   9    Rotary  Kiln with
                       Secondary  Combustor
Wastes Incinerated

High-Btu Liquids
Low-Btu Liquids
Solids Feed

Liquid Waste Fuel
High-Btu Liquids
Low-Btu Liquids
Solids Feedb
 Flow-Weighted
    Average
Influent (mg/kg)

      1,009
      4,700
      4,244
                 "Values  shown  as  "<"  were  reported as below  indicated  detection  limits.

                 (a)   The influent concentration  is flow-weighted  average.
                 (b)   Gel  and  filter  press  residue.
  Incinerator
   Residue*
 Total      TCLP
fmg/kg)    (ug/L)   Footnotes
 <300
17
             <3
             <3

-------
5.6.26   Trichlorofluoromethane (Other Than Wastewater)

    The Agency has no data on TCLP extracts of residue from incineration
of trichlorofluoromethane to use in the derivation of the BDAT treatment
standard.  For reasons presented in Section 5.6.1, EPA used chemical
structure as the basis for transferring treatment data to
trichlorofluoromethane spent solvent wastes other than wastewaters.
Specifically we transferred treatment data for methylene chloride to
trichlorofluoromethane; both contain the halogen functional group.

    The Agency has data on the analysis of the TCLP extract of
incineration residue for two compounds in the halogenated aliphatics
structural group:  methylene chloride and 1,1,1-trichloroethane.   To best
account for the range of physical and chemical properties within a
structural group that affect treatment by a specific technology,  the
Agency transferred data representing the least stringent treatment
standard from the compounds for which data were available in the
halogenated aliphatics structural group.  The data from which the
treatment standard for incineration of methylene chloride was derived
were transferred to trichlorofluoromethane.  The treatment standard is
0.96 mg/L based on the transferred data.

    We believe the BDAT treatment standard for trichlorofluoromethane
spent solvent wastes (other than wastewater) represents substantial
treatment.  We would expect untreated trichlorofluoromethane wastes to be
similar to untreated methylene chloride wastes from which we transferred
treatment data since they are used in many similar manufacturing
processes as shown in Section 2 of this document.  As discussed on page
5-159, in reference to methylene chloride, we believe these constituent
reductions substantially diminish the toxicity of the spent solvent
wastes containing trichlorofluoromethane and substantially reduce the
likelihood of migration of trichlorofluoromethane from spent solvent
wastes.

    [The proposed technology-based BDAT treatment standard for
trichlorofluoromethane was estimated at the detection limit of <0.010
mg/L based on incineration (see Table 11, 51 FR 1722).   The principal
difference between the proposed and promulgated treatment standards is
the Agency's change in the criteria for data transfer (see Section 5.6.1,
page 5-130, for a discussion of the Agency's methodology for data
transfer.)]
                                     5-188

-------
5.6.27   Xylene (Other Than Wastewater)

    The Agency has treatment data for the TCLP extract of incinerator ash
from the treatment of xylene (Reference 11).  The data are summarized in
Table 5-47.

    The following steps were taken to derive the BOAT treatment standard
for xylene:

    1. We evaluated the data to determine whether any of the data
       represented poor design or operation of the incineration system.
       Data from one facility (site 6) were deleted because the
       incineration system control devices were not properly designed and
       operated.  The facility is under a Consent Decree to replace and
       improve the current incinerator control system.  Data from another
       facility (site 2) were deleted because the Agency judged that the
       system was not properly operated at the time the data were
       collected.  A follow-up sampling visit confirmed the Agency's
       judgment.  The new data were not used in the determination of the
       long-term performance average for incineration of xylene;  however,
       the data were used to develop a variability factor for
       incineration.  Data from two other sites were deleted because
       xylene was reported as below the detection limits for both the
       influent and the TCLP extract of the ash.

    2. We determined an arithmetic average residue concentration level
       and a variability factor for each data set.  Residue concentration
       levels reported as less than or equal to the reported detection
       limit were set egual to the detection limit for statistical
       analyses.  This is a conservative approach since the actual
       concentration would be between zero and the detection limit.
       Process variability could not be calculated from the incineration
       data because only one influent and effluent data pair was
       available for each data set.   Therefore, to account for process
       variability, an average variability factor was calculated  for
       incineration, 5.34 (calculation of the average variability factor
       is shown in Table 5-33).

    3. The analysis of variance  method was not used to compare different
       treatments of the same waste  because data are available for only
       one type of treatment for each waste.

    4. EPA then analyzed the data to determine if the various treatment
       concentration levels shown in Table 5-47 could be associated with
       separate waste treatability subgroups.   Sufficient data did not
       exist to identify separate waste treatability subgroups;
       therefore, one waste treatability subgroup was established for all
       sources of wastes (other  than wastewater) containing xylene spent
                                     5-189

-------
       solvents.   The least stringent treatment level  within the
       treatability subgroup was selected for BDAT (0.15 mg/L for site 7
       obtained by multiplying the variability factor  by the highest
       average residue concentration level)  to ensure  that the standard
       could be achieved for all waste matrices within the waste
       treatability subgroup.  The technology basis was incineration.

    5.  The BDAT treatment standard for xylene represents treatment of  a
       variety of waste matrices incinerated at four sites.  The
       untreated waste concentration of xylene ranged  from 7,300 mg/kg to
       46,393 mg/kg in these waste matrices.   All of these wastes were
       treated to the BDAT treatment standard or below (0.15 mg/L). We
       believe these constituent reductions  substantially diminish the
       toxicity of the spent solvent wastes  containing xylene and
       substantially reduce the likelihood of migration of" xylene from
       spent solvent wastes.

    [The proposed BDAT technology-based treatment standard for xylene  was
estimated at the detection limit of <0.010 mg/L based  on incineration
(see Table 11, 51 FR 1722).  The difference  between the proposed and
promulgated treatment standard is primarily  a result of additional data
gathering subsequent to proposal.  The new data were presented in EPA's
Notice of Availability of Data (51 FR 31783).  In addition, a variability
analysis was incorporated into the development of the  treatment standards
for promulgation.]
                                      5-190

-------
                                             Table  5-47

                                    INCINERATION  DATA FOR XYLENE
Site  Type of Incinerator

  3   Rotary Kiln with
      Secondary Combustor

  7   Fixed Hearth with
      Secondary Combustor
      Rotary Kiln wi th
      Secondary Liquid
      Injection Combustor

      Rotary Kiln with
      Secondary Combustor
Wastes Incinerated

Drum Feed Sol ids
Liquid Waste Fuel

High-Btu Liquids
Low-Btu Liquids
Solids Feed

Liquid Waste Fuel
High-Btu Liquids
Low-Btu Liquids
Solids Feedb
                                                       Flow-Weighted
                                                          Average
                                             Incinerator
                                              Residue*
                    Total      TCLP
Influent (mg/ka)   (mg/kg)   (ug/L)  Footnotes
     15,863
     46,393
      7,300
     22,039
   1.5
<300
15
28
            <3
            <3
"Values shown as "<" were reported as below indicated detection limits.

(a)  The influent concentration is flow-weighted average.
(b)  Gel and filter press residue.

-------
                          REFERENCES - SECTION 5
1.   Versar, Incorporated.  Physical-Chemical Properties and
     Categorization of RCRA Wastes According to Volatility.
     EPA-450/3-85-007.  Prepared for U.S. EPA, Office of Air Quality
     Planning and Standards, Emissions Standards and Engineering
     Division.  February 1985.

2.   Stover, E.L. and D.F. Kincannon.  "Contaminated Groundwater
     Treatability - A Case Study."  Journal American Water Works
     Association.  June 1983.

3.   IT Enviroscience, Inc.  Survey of Industrial Applications of
     Aqueous-Phase Activated-Carbon Adsorption for Control of Pollutant
     Compounds from Manufacture of Organic Compounds.  Prepared for U.S.
     EPA, Industrial Environmental Research Laboratory.  April 1983.

4.   Button, D.G.  "Removal of Priority Pollutants by the DuPont PACT
     Process."  Proceedings of the 7th Annual Industrial Pollutant
     Conference.  Philadelphia, Pennsylvania.  June 5-7, 1979.

5.   Torpy, M.F., L.A. Raphaelian, and R.G. Luthy.  Wastewater and Sludge
     Control - Technology Options for Synfuels Industries, Volume 2:
     Tar-Sand-Combustion Process Water - Removal of Organic Constituents
     by Activated-Sludge Treatment.  ANL/ES-115.  Argonne National
     Laboratory.  1981.

6.   Love, O.T. and R.G. Eilers.  "Treatment of Drinking Water Containing
     Trichloroethylene and Related Industrial Solvents."  Journal
     American Water Works Association.  August 1982.

7.   Becker, D.L. and S.C. Wilson.  "The Use of Activated Carbon for the
     Treatment of Pesticides and Pesticidal Wastes."  Carbon Adsorption
     Handbook.  P.N. Cheremisinoff and F. Ellerbusch, eds.  Ann Arbor
     Science, Ann Arbor, Michigan.  1978.

8.   Ruggiero, D.C. and R. Ausubel.  "Removal of Organic Contaminants
     from Drinking Water Supply at Glen Cove, New York."  UbEPA, Office
     of Research and Development.  Nebolsine Kohlman Ruggiero Engineers,
     P.C.  NTIS PB82-258963.  1982.
                                 5-192

-------
9.   U.S. EPA.  Development Document for Effluent Limitations Guidelines
     and Standards for the Iron and Steel Manufacturing Point Source
     Category, Volume II.  Office of Water Regulations and Standards,
     Effluent Guidelines Division.  EPA 440/l-80-024b.  December 1980.

10.  Data submitted by Zimpro, Inc.  1986.

11.  Acurex Corporation.  Characterization of Hazardous Waste
     Incineration Residuals.  Contract No. 68-03-3241.  Prepared for U.S.
     EPA, Hazardous Waste Engineering Research Laboratory.  June 1986.

12.  Baker, C.D., E.W. Clark, and W.V. Jeserig, The Sherwin Williams Co.,
     and C.H. Huether, Westvaco Corporation.  "Recovering para-Cresol
     from Process Effluent."  Chemical Engineering.  69(8):  77+ (August
     1973).

13.  Notice of Availability of New Information for Establishment of
     Effluent Guidelines for Organic Chemicals, Plastics, and Synthetic
     Fibers (OCPSF) Industrial Point Source Category.  October 1985.

14.  U.S. EPA.  Development Document for Effluent Limitations Guidelines
     and Standards for the Pharmaceutical Manufacturing Point Source
     Category.  EPA 440/1-83-084.  September 1983.

15.  51 FR 1722, Table 11

16.  51 FR 31783

17.  51 FR 1725, Table 13

18.  Allinger, Norman L., M.P. Cava, D.C. Johgh, C.R. Johnson,  N.A.
     Norman, and C.L. Stevens.  Organic Chemistry, Worth Publishers,
     Inc., New York.  1971.

19.  Reid, Robert C., J.M. Prausnitx, and T.K.  Sherwood.   The Properties
     of Gases and Liguids, 3rd. ed.  McGraw Hill Book Company,  New York.
     1977.
                                 5-193

-------