vvEPA
           United States
           Environmental Protection
           Agency
           Office of Air Quality
           Planning and Standards
           Research Triangle Park NC 27711
EPA-450/3-83-013a
September 7985
           Air
Polymer
Manufacturing
Industry -
Background
Information for
Proposed
Standards
Draft
EIS

-------
                    ENVIRONMENTAL PROTECTION AGENCY
    Background Information and Draft Environmental Impact Statement
                                  for
                     Polymer Manufacturing Industry
               s\  ^
 repared bv:
Jack R. Farmer
Director, Emission Standards and Engineering Division
U.S. Environmental Protection Agency (MD-13)
Research Triangle Park, North Carolina  27711
                                                        /   Date
     1.  The proposed standards of performance would limit emissions
of volatile organic compounds from new, modified, and reconstructed
polymer manufacturing facilities.  Section 111 of the Clean Air Act
(42 U.S.C. 7411), as amended, directs the Administrator to establish
standards of performance for any category of new stationary source of
air pollution that "... causes or contributes significantly to air
pollution which may reasonably be anticipated to endanger public health
or welfare."  Although all regions of the United States will  be affected,
the Southeast and Gulf States will be affected the most.

     2.  Copies of this document have been sent to the following Federal
Departments:  Labor, Health, and Human Services, Defense, Transporta-
tion, Agriculture, Commerce, Interior, and Energy; the National  Science
Foundation; the Council on Environmental Quality; State and Territorial
Air Pollution Program Administrators; EPA Regional Administrators;
Local Air Pollution Control Officials; Office of Management and Budget;
and other interested parties.

     3.  The comment period for review of this document is 75 days from
the date of publication of the proposed standard in the Federal  Register.
Mr. Gil Wood may be contacted at (919) 541-5625 regarding the date of
the comment period.

     4.  For additional information contact:

         Mr. James Berry
         Chemicals and Petroleum Branch
         Emission Standards and Engineering Division (MD-13)
         U.S. Environmental Protection Agency
         Research Triangle Park, North Carolina  27711
         Telephone:  (919) 541-5605

     5.  Copies of this document may be obtained from:

         U.S. Environmental Protection Agency Library (MD-35)
         Research Triangle Park, North Carolina  27711
         Telephone:  (919) 541-2777

         National Technical Information Service
         5285 Port Royal  Road
         Springfield, Virginia  22161

-------
                        EPA-450/3-83-019a
           Polymer
Manufacturing Industry -
Background Information
for Proposed  Standards
    Emission Standards and Engineering Division
    U.S. ENVIRONMENTAL PROTECTION AGENCY
       Office of Air, Noise, and Radiation
    Office of Air Quality Planning and Standards
    Research Triangle Park, North Carolina 27711

            September 1985

-------
This report has been reviewed by the Emission Standards and Engineering Division of the Office of AirQuality Planning
and Standards, EPA, and approved for publication. Mention of trade names or commercial products is not intended to
constitute endorsement or recommendation for use. Copies of this report are available through the Library Services
Office (MD-35), U.S. Environmental Protection Agency, Research Triangle Park, N.C. 27711,orfrom National Technical
Information Services, 5285 Port Royal, Springfield, Virginia 22161.

-------
                            TABLE OF CONTENTS
                                                                Page
1.0  SUMMARY
     1.1  Regulatory Alternatives ..............     1-1
     1.2  Environmental Impact ...............     1-2
          1.2.1  Air Emissions Impacts ...........     1-4
          1.2.2  Water, Solid Waste, and Noise Impacts ...     1-4
          1.2.3  Energy Impacts .........  .  .....     1-5
     1.3  Economic Impact ..................     1-5
2.0  INTRODUCTION
     2.1  Background and Authority  for Standards  ......     2-1
     2.2  Selection of Categories of Stationary Sources.  .  .     2-4
     2,3  Procedure for Development of Standards
          of Performance ..................     2-6
     2.4  Consideration of Costs ..............     2-8
     2.5  Consideration of Environmental  Impacts  ......     2-9
     2.6  Impact on Existing Sources ............     2-10
     2.7  Revision of Standards of  Performance .......     2-11
3.0  THE POLYMERS AND RESINS INDUSTRY
     3.1  Industry Description ...............     3-1
          3.1.1  End-Uses of the Five Polymers
                 Chosen for NSPS Development ........     3-2
     3.2  Polymerization Processes  and Process
          Emissions .....................     3-11
          3.2.1  Polypropylene ................     3-13
          3.2.2  Low Density Polyethylene (LDPE)  ......     3-22
          3.2.3  High Density Polyethylene (HOPE) ......     3-32
          3.2.4  Polystyrene ................     3-40
          3.2.5  Polyester Resin ...........  ...     3-53
     3.3  Fugitive VOC Sources and  Emissions . .  ......     3-63
     3.4  Baseline Emissions ................     3-64
          3.4.1  Process Emissions  .............     3-64
          3.4.2  Fugitive Emissions .............     3-70
     3.5  References for Chapter 3  .............     3-72

-------
4.0  EMISSION CONTROL TECHNIQUES
     4.1  Control Techniques for Process Emissions 	    4-1
          4.1.1  Control by Combustion Techniques  	    4-2
          4.1.2  Control by Recovery Techniques	    4-24
     4.2  Control Techniques for Fugitive Emissions	    4-34
          4.2.1  Leak Detection and Repair Program	    4-34
          4.2.2  Preventive Programs 	    4-36
     4.3  References for Chapter 4	    4-38
5.0  MODIFICATIONS AND RECONSTRUCTIONS
     5.1  Definitions	    5-1
          5.1.1  Modification	    5-1
          5.1.2  Reconstruction	    5-2
     5.2  Modifications and Reconstructions at
          Polymers and Resins Facilities 	    5-3
          5.2.1  Process Emissions 	    5-3
          5.2.2  Fugitive Emissions	  .    5^-5
          5.2.3  Summary	    5-6
     5.3  References for Chapter 5	    5-7
6.0  MODEL PLANTS AND REGULATORY ALTERNATIVES
     6.1  Model  Plants	    6-1
     6.2  Regulatory Alternatives	    6-16
          6.2.1  Baseline Control	    6-16
          6.2.2  Control Techniques	    6-17
          6.2.3  Regulatory Alternatives 	    6-20
          6.2.4  Summary of Regulatory Alternatives	    6-41
7.0  ENVIRONMENTAL IMPACTS
     7.1  Air Pollution Impacts	    7-2
          7.1.1  Average Annual Model Plant
                 VOC Emissions	    7-2
          7.1.2  Industrywide VOC Emission
                 Impacts of New Plants	    7-9
          7.1.3  Secondary Air Quality Impacts of
                 the Regulatory Alternatives  	    7-9

-------
     7.2  Water Pollution Impacts.	
     7.3  Solid Waste Disposal Impacts .......
     7.4  Energy Impacts 	
          7.4.1  Model Plant Energy Impacts	
          7.4.2  Industrywide Energy Impacts . . .   .
     7.5  Other Environmental Impacts	
          7.5.1  Noise Impacts 	
          7.5.2  Irreversible and Irretrievable
                 Commitment of Resources 	
          7.5.3  Environmental Impacts of Delayed
                 Regulatory Action 	
     7.6 References for Chapter 7	
8.0  COSTS
     8.1  Cost Analysis of Regulatory Alternatives  .
          8.1.1  Flare Design and Cost Basis  .  .  .  .
          8.1.2  Thermal  Incinerator Design and
                 Cost Basis	'.....
          8.1.3  Catalytic Incinerator Design
                 and Cost Basis	
          8.1.4  Condenser Design and Cost  Basis  .  .
          8.1.5  Ethylene Glycol  Recovery System
                 Design and Cost  Basis  	
          8.1.6  Fugitive Emission Control  Program
                 Design and Cost  Basis	
          8.1.7  Cost Analysis Results  	
     8.2  Other Cost Considerations	
          8.2.1  Water Pollution  Control
                 Regulations ...  	
          8.2.2  Occupational  Safety  and  Health
                 Regulations 	
          8.2.3  Toxic Substance  Control  Regulations
          8.2.4  Solid and  Hazardous  Waste  .....
                 Regulations 	
          8.2.5  Clean Air Act 	
     8.3  References  for  Chapter  8  	
 7-19
 7-20
 7-24
 7-24
 7-29
 7-34
 7-34

 7-34

 7-35
 7-36

 8-1
 8-5

 8-12

 8-14
 8-16

 8-18

 8-19
 8-36
 8-64

 8-64

 8-65
8-65

8-66
8-67
8-69

-------
9.0  ECONOMIC IMPACT
     9.1  Industry Characterization	
          9.1.1  Industry Structure	
          9.1.2  Industry Performance	
          9.1.3  Five-Year Projections 	
     9.2  Economic Impact Analysis 	 »
          9.2.1  Economic Impact Assessment
                 Methodology:  Revenue and Price .  .  .  . .
          9.2.2  Economic Impact of VOC Potential NSPS
                 Regulatory Alternatives -
                 Polymers and Resins 	
     9.3  Potential Socioeconomic and Inflationary
          Impacts	
          9.3.1  Fifth Year Costs and Benefits ......
          9.3.2  Impacts on Small Facilities 	
          9.3.3  Other Impacts 	
     9.4  References for Chapter 9 	
APPENDIX A - Evolution of the Proposed Standards
APPENDIX B - Index to Environmental Considerations
APPENDIX C - Emission Source Test Data and
             Fugitive Emission Source Counts
     C.I  Flare VOC Emission Test Data 	
          C.I.I  Control Device   	
          C.I.2  Sampling and Analytical Techniques.  .  . .
          C.I.3  Test Results  	
     C.2  Thermal  Incinerator VOC Emission Test Data .  . .
          C.2.1  Environmental Protection Agency (EPA)
                 Polymers Test Data	
          C.2.2  Environmental Protection Agency (EPA)
                 Air Oxidation Unit Test Data	
          C.2.3  Chemical Company Air Oxidation Unit Test
                 Data	
          C.2.4  Union Carbide Lab-Scale Test Data .  .  . .
     C.3  Vapor Recovery System VOC Emission Test Data . .
9-1
9-1
9-20
9-25
9-36

9-37
9-43

9-50
9-50
9-54
9-55
9-56
C-2
C-3
C-3
C-5
C-5

C-9

C-17

C-2 5
C-33
C-33

-------
     C.4  Discussion of Test Results and Technical Basis
          of the Polymers and Resins VOC Emissions Reduction
          Requirement	         C-35
          C.4.1  Discussion of Flare Emission Test
                 Results	         C-35
          C.4.2  Discussion of Thermal  Incineration Test
                 Results . . .	         C-35
     C.5  Fugitive Emission Equipment Inventory	         C-41
     C.6  References for Appendix C	         C-46

APPENDIX D - Emission Measurement and Performance
             Test Methods
     I.   Process VOC Sources	         D-l
          1-D.l  Emission Measurement	         D-2
          I-D.2  Recommended Test Methods	         D-3
          I-D.3  References	? .  . .         D-3
     II.  Fugitive VOC Sources	         D-4
          II-D.l  Emission Measurement Methods 	         D-4
          II-D.2  Continuous Monitoring Systems
                  and Devices  .  .  .  «	         D-7
          II-D.3  Performance Test Method ... 	         D-8
          II-D.4  References	         D-10
APPENDIX E - Detailed Design and Cost Estimation Procedures
     E.I  General	         E-l
     E.2  Flare Design and Cost Estimation Procedure ...         E-l
          E.2.1  Flare Design Procedure	         E-2
          E.2.2  Flare Cost Estimation  Procedure 	         E-6
     E.3  Thermal Incinerator Design  and Cost Estimation
          Procedure	         E-9
          E.3.1  Thermal  Incinerator  Design Procedure.  . .         E-9
          E.3.2  Thermal  Incinerator  Cost Estimation
                 Procedure	         E-22
     E.4  Catalytic Incinerator Design  and Cost Estimation
          Procedure	         E-2 5
          E.4.1  Catalytic Incinerator  Design Procedure. .         E-25
                                    vii

-------
     E.4.2  Catalytic Incinerator Cost Estimation
            Procedure ...» 	
E.5  Surface Condenser Design and Cost Estimation
     Procedure	.	
     E.5.1  Surface Condenser Design	
     E.5.2  Surface Condenser Cost Estimation
            Procedure 	
E.6  Ethylene Glycol Recovery Systems Design and Cost
            Estimation Procedure 	
     E.6.1  Ethylene Glycol  Recovery System Design.  .
     E.6.2  Ethylene Glycol  Recovery System Cost
            Estimation Procedure	
E.7  Piping and Ducting Design and Cost Estimation
     Procedure	
     E.7.1  Piping and Ducting Design Procedure . .  .
     E.7.2  Piping and Ducting Cost Estimation
            Procedure 	
E.8  References 	
E-28

E-36
E-40

E-44

E-44
E-53

E-53

E-60
E-60

E-60
E-66
                              vi i i

-------
                             List of Tables

 1-1   Assessment of Environmental and Economic Impacts for
      Each Regulatory Alternative Considered  	
 3-1   Polypropylene (PP) Plant List	
 3-2   Low Density Polyethylene (LDPE) Plant List 	
 3-3   High Density Polyethylene  (HOPE) Plant List  	
 3-4   Polystyrene (PS) Plant List	'. . .  .
 3-5   Poly(ethylene terephthalate) (PET) Polyester Plant
      List	
 3-6   Characteristics of Vent Streams from the Polypropylene
      Continuous Liquid Phase Slurry Process 	
 3-7   Characteristics of Vent Streams from the Polypropylene
      Gas Phase Process  	
 3-8   Characteristics of Vent Streams from the Low Density
      Polyethylene High-Pressure Process 	
 3-9   Characteristics of Vent Streams from the Low Density
      Polyethylene Low-Pressure Process and the High Density
      Polyethylene Gas Phase Process   .	
3-10  Characteristics of Vent Streams from the High Density
      Polyethylene Liquid Phase Slurry Process 	
3-11  Characteristics of Vent Streams from the High Density
      Polyethylene Liquid Phase Solution Process  	
3-12  Characteristics of Vent Streams from the Polystyrene
      Batch Process	  .
3-13  Characteristics of Vent Streams from the Polystyrene
      Continuous Process 	  	
3-14  Characteristics of Vent Streams from the Expandable
      Polystyrene Post-Impregnation Process	
3-15  Characteristics  of Vent Streams from the Expandable
      Polystyrene In-Situ Process 	
3-16  Characteristics  of Vent Streams from the
      Poly(ethylene  terephthalate) DMT  Process	
3-17  Characteristics  of Vent Streams from the
      Poly(ethylene  terephthalate) TPA  Process	
 Paqe
 1-3
 3-3
 3-4
 3-5
 3-6

 3-7

 3-18

 3-23

 3-27


 3-31

 3-35

 3-39

 3-44

 3-48

 3-54

 3-55

 3-59

3-62

-------
3-18  Vapor Pressures of Major Organic Compounds
      Used in Polymers and Resins Manufacturing 	
3-19  Uncontrolled Fugitive Emission Rates	
4-1   Flare Emission Studies  	  	
6-1   Model Plant Characteristics for Process Emissions
      from the Polypropylene Liquid Phase Process.  .  . .
6-2   Model Plant Characteristics for Process Emissions
      from the Polypropylene Gas Phase Process	
6-3   Model Plant Characteristics for Process Emissions
      from the LDPE High Pressure Process   	
6-4   Model Plant Characteristics for Process Emissions
      from the LDPE Low Pressure and the HOPE Gas Phase
      Processes	•	
6-5   Model Plant Characteristics for Process Emissions
      from the HOPE Liquid Phase Slurry Process 	
6-6   Model Plant Characteristics for Process Emissions
      from the HOPE Liquid Phase Solution Process . .  . .
6-7   Model Plant Characteristics for Process Emissions
      from the Crystal or Impact Polystyrene Continuous
      Process    	
6-8   Model Plant Characteristics for Process Emissions
      from the Expandable Polystyrene Post-Impregnation
      Suspension Process   	
6-9   Model Plant Characteristics for Process Emissions
      from the Expandable Polystyrene In-Situ Suspension
      Process    	
6-10 Model Plant Characteristics for Process Emissions
      from the Poly(ethylene  terephthalate) DMT Process .
6-lla Model Plant Characteristics for Process Emissions
      from the Poly(ethylene  terephthalate) TPA Process
      Using  a Single  End  Finisher	
6-llb Model  Plant Characteristics for Process  Emissions
      from the Poly(ethylene  terephthalate) TPA Process
      Using  Multiple  End  Finishers	
6-12 Fugitive VOC  Emission Model Plant Parameters.  .  .
6-13 Control Specification for Fugitive  Emissions
      Under  Regulatory Alternative  2.  .  .	  • •
3-65
3-66
4-11

6-3

6-4

6-5


6-6

6-7

6-8


6-9


6-10


6-11

6-12


6-13
 6-14
 6-15

 6-19

-------
6-14  Regulatory Alternatives for the Polypropylene
      Liquid Phase Process		          6-22
6-15  Regulatory Alternatives for the Polypropylene
      Gas Phase Procees		          6-24
6-16  Regulatory Alternatives for the LDPE High Pressure
      Process		          6-25
6-17  Regulatory Alternatives for the LDPE Low-Pressure
      and HOPE Gas Phase Process	          6-27
6-18  Regulatory Alternatives for the HOPE Liquid Phase
      Slurry Process	          6-29
6-19  Regulatory Alternatives for the HOPE Liquid Phase
      Solution Process. .	          6-30
6-20  Regulatory Alternatives for the Crystal  or Impact
      Polystyrene Continuous Process	          6-32
6-21  Regulatory Alternatives for the Expandable
      Polystyrene Post-Impregnation Suspension Process. .  .          6-34
6-22  Regulatory Alternatives for the Expandable
      Polystyrene In-Situ  Suspension Process	          6-36
6-23  Regulatory Alternatives for the Poly(ethylene
      terephthalate) DMT Process	          6-38
6-24a Regulatory Alternatives for the Poly(ethylene
      terephthalate) TPA Process Producing Low Viscosity PET
      or High Viscosity PET with a Single End  Finisher. .  .          6-39
6-24b Regulatory Alternatives for the Poly(ethylene
      terephthalate) TPA Process Producing High Viscosity  PET
      with Multiple End Finishers	        6-40
6-25  Summary of Uncontrolled Emissions and Emission Reductions
      per Process Line for Regulatory Alternatives by Model
      Plant	          6-42
7-la  Primary Air Quality  Impacts of the Regulatory
      Alternatives for Polymers and Resins
      Plants (Mg/yr)	          7-4
7-lb  Primary Air Quality  Impacts of the Regulatory
      Alternatives for Polymers and Resins
      Plants (tons/yr)	          7-6
                                     XI

-------
7-2a  Industrywide Primary Air Quality Impacts of the
      Regulatory Alternatives for Polymers
      and Resins Plants (Mg/yr) 	
7-2b  Industrywide Primary Air Quality Impacts of
      the Regulatory Alternatives for Polymers
      and Resins Plants (tons/yr) 	
7-3a  Secondary Air Quality Impacts of the
      Regulatory Alternatives for Polymers and
      Resins Plants 	
7-3b  Secondary Air Quality Impacts of the
      Regulatory Alternatives for Polymers and
      Resins Plants 	
7-4   Industrywide Solid Waste Impacts of the
      Regulatory Alternatives for New Polymer
      and Resin Process Lines 	
7-5   Volume of Biological Sludge Generaged by
      Process Operations in New Polymer and Resin Process
      Lines that Employ Flares, Thermal Incineration,
      or Catalytic Incineration   	
7-6a  Energy Impacts of the Regulatory Alternatives
      for Polymer and Resin Plants	
7-6b  Energy Impacts of the Regulatory Alternatives
      for Polymer and Resin Plants	
7-7a  Industrywide Energy Impacts of the Regulatory
      Alternatives for New Polymer and Resin
      Plants (TJ/yr)	
7-7b  Industrywide Energy Impacts of the Regulatory
      Alternatives for New Polymer and Resin
      Plants (1,000 bbl oil/yr)  	
8-1   Summary of Regulatory Alternatives for
      the Model Plants 	  	
8-2   Installation Cost Factors   	
8-3   Annualized Cost Factors for Polymers and Resins
      NSPS  (June 1980 Dollars) 	
8-4   Fugitive  VOC Regulatory Alternative Control
      Specifications  	
7-10
7-12
7-15
7-17
7-21
7-23

7-25

7-27


7-30


7-32

8-2
8-6

8-7

8-20
                                    xii

-------
8-5   Fugitive VOC Emission Data for the Sources
      in Polymers and Resins Model Plants  	
8-6   Fugitive VOC Regulatory Alternative Costs for
      Polymers and Resins Model Units  	
8-7   Summary of Fugitive VOC Emission Control
      Costs for the Sources in Polymers and
      Resins Model Unit (May 1980 dollars) 	
8-8   Initial Leak Repair Labor-Hours Requirement
      for Valves for the Model Unit  .........
8-9   Total Annual Costs for Initial  Leak Repair
      for Valves for the Model Unit (May 1980 Dollars)
8-10  Annual Monitoring and Leak Repair Labor
      Requirements for Valves for the Model Unit
      (Monthly Leak Detection and Repair Program)  .  .
8-11  Annual Monitoring and Leak Repair Costs for
      Monthly Monitoring of Valves for the Model
      Unit (May 1980 Dollars)	 .  .  .  .
8-12  Initial Leak Repair Labor-Hours Requirement
      for Pump Seals for the Model  Unit	
8-13  Total Annual Costs for Initial  Leak Repair
      for Pump Seals for the Model  Unit (May
      1980 Dollars)	
8-14  Annual Monitoring and Leak Repair Labor
      Requirements for Pump Seals of  the Model
      Unit (Monthly Leak Detection and Repair
      Program) 	  	
8-15  Annual Monitoring and Leak Repair Costs for
      Monthly Monitoring of Pump Seals for the
      Model Unit (May 1980 Dollars)	
8-16  Relief Valve Control  Costs for  Rupture Disk
      Systems with Block Valves and Three-Way
      Valves (May 1980 Dollars)	
8-17  Capital  and Net Annual!zed Costs for Control  of
      Emissions from Safety/Relief Valves for the
      Model Unit (May 1980 Dollars)  	
8-21

8-23


8-24

8-25

8-25


8-26


8-27

8-29


8-29
8-30
8-31
8-33
8-34
                                    xi i i

-------
8-18  Capital and Net Annualized Costs for Control of
      Emissions from Open-Ended Lines (May 1980 Dollars). .          8-35
8-19  Capital and Net Annualized Costs for Control of
      Emissions from Compressor Seals for the
      Model Unit (May 1980 Dollars)	 .          8-37
8-20  Capital and Net Annualized Costs for Control of
      Emissions from Sampling Systems for the Model  Unit
      (May 1980 Dollars)	          8-38
8-21  Polypropylene, Liquid Phase Model  Plant
      Regulatory Alternatives Costs (June 1980
      Dollars)	          8-39
8-22  Polypropylene, Gas Phase Model Plant
      Regulatory Alternatives Costs (June 1980
      Dollars)	          8-40
8-23  Low Density Polyethylene, High-Pressure Model  Plant
      Regulatory Alternatives Costs (June 1980 Dollars)  . .          8-41
8-24  Low Density Polyethylene, Low-Pressure and High Density
      Polyethylene, Gas Phase Model Plant Regulatory
      Alternatives) Costs (June 1980 Dollars) 	          8-42
8-25  High Density Polyethylene, Liquid Phase-Slurry
      Model Plant Regulatory Alternatives Costs (June
      1980 Dollars) . .	          8-43
8-26  High Density Polyethylene, Liquid Phase-Solution
      Model Plant Regulatory Alternatives Costs (June
      1980 Dollars)	          8-44
8-27  Polystyrene Continuous Model  Plant Regulatory
      Alternatives Costs (June 1980 Dollars)	          8-45
8-28  Expandable Polystyrene Post-Impregnation Suspension
      Process Model Plant Regulatory Alternatives Co'st
      (June 1980 Dollars)	          8-46
8-29  Expandable Polystyrene In-Situ Suspension Process
      Model Plant Regulatory Alternatives Costs
      (June 1980 Dollars)	  .          8-47
8-30  Poly(ethylene terephthalate)  (PET)  - DMT Process
      Model Plant Regulatory Alternatives Costs
      (June 1980 Dollars)	          8-48
                                    xiv

-------
 8-31a Poly(ethylene terephthalate)  (PET)  -  TPA  Process
       Model  Plant Regulatory Alternatives Costs,
       Low Viscosity,  or High Viscosity  with Single  End
       Finisher (June  1980 Dollars)  	
 8-31b Poly(ethylene terephthalate)  (PET)  -  TPA  Process
       Model  Plant Regulatory Alternatives Costs,
       High .Viscosity  with Multiple  End  Finishers  ....
 8-32   Costs  and Associated Emission Reductions  of
       Regulatory  Alternatives for Polypropylene,
       Liquid Phase Process	
 8-33   Costs  and Associated Emission Reductions  of
       Regulatory  Alternatives for Polypropylene,
       Gas Phase Process  	
 8-34   Costs  and Associated Emission Reductions  of
       Regulatory  Alternatives for LDPE, High Pressure
       Process  	
 8-35   Costs  and Associated Emission Reductions  of
       Regulatory  Alternatives  for LDPE, Low Pressure and
       HOPE,  Gas Phase Process	 .  .
 8-36   Costs  and Associated Emission Reductions  of
       Regulatory  Alternatives  for HOPE, Liquid
       Phase  Slurry Process	
 8-37   Costs  and Associated  Emission Reductions of
       Regulatory Alternatives  for HOPE, Liquid
       Phase  Solution Process	
 8-38   Costs  and Associated Emission Reductions of
       Regulatory Alternatives for Polystyrene,
       Continuous Process	
 8-39  Costs  and Associated Emission Reductions of
       Regulatory Alternatives for EPS, Post-Impregnation
       Suspension Process	
8-40  Costs and Associated Emission  Reductions of
      Regulatory Alternatives for EPS, In-Situ Suspension
      Process 	
8-41  Costs and Associated Emission  Reductions of
      Regulatory Alternatives for PET/DMT  Process ....
 8-49


 8-50


 8-51


 8-52


 8-53


 8-54


 8-55


 8-56


8-57


8-58


8-59

8-60
                                     xv

-------
8-42a Costs and Associated Emission Reductions of Regulatory
      Alternatives for PET/TPA Process,  Low Viscosity,
      and High Viscosity with Single End Finisher	8-61
8-42b Costs and Associated Emission Reductions of Regulatory
      Alternatives for PET/TPA Process,  High Viscosity  with
      Multiple End Finishers	         8-62
8-43  Total Fifth-Year Net Annualized Cost of
      Process and Fugitive Emission Controls for
      Polymers and Resins Facilities Affected by NSPS . .  .         8-63
9-1   Number of Polymer and Resin Plants by
      Manufacturer and Type; January 1,  1982	         9-4
9-2   Capacity of Polymer and Resin Plants by
      Manufacturer and Type; January 1,  1982  	         9-6
9-3   Size Distribution of Polymer and Resin Plants
      by Type and Capacity; January 1, 1982 ........         9-12
9-4   Number of Polymer and Resin Plants by
      Type and Location; January 1, 1982  .........         9-13
9-5   Employment in Polymer and Resin Plants
      in 1977 and 1981	         9-14
9-6   U.S. Exports of Polymers and Resins, by
      Type and Year, 1976-1981 (gigagrams)	         9-17
9-7   Domestic Consumption of Polymers and Resins
      by End-Use Market and Process Type, 1981 (gigagrams).         9-18
9-8   Shipments of Polymers and Resins by
      Major Market, 1981  (gigagrams)	         9-19
9-9   Production, Capacity, and Capacity Utilization
      of Polymers and Resins  .	         9-21
9-10  Sales and Value of  Production of Polymers
      and Resins, 1978-1981  (million nominal dollars)  ...         9-24
9-11  Projected Required  New Capacity for
      Polymers and Resins, 1984 through 1988	         9-31
9-12  Projected Number of New Polymers and Resins
      Plants, by Process, 1984 through 1988	         9-34
                                    xvi

-------
 9-13  Operational Characteristics of Polymers and
      Resins Model Process Lines Built 1984 Through 1988
      (June, 1980 Dollars)	
 9-14  Polymers and Resins Model Process Line Control
      Costs and Maximum Price Increases for Model Plant
      by Product, Process, and Regulatory Alternative
      (June, 1980 Dollars)	
 9-15  Fifth Year Net Annualized Cost to Society of
      Regulatory Alternatives by Model Process Line Product
      and Process (June, 1980 Dollars)	 .
 9-16  Upper Boundary of Total Annualized Fifth Year
      Cost to Society (June, 1980 Dollars)	
 C-l   Emission Analyzers and Instrumentation
      Utilized for Joint EPA/CMA Flare Testing  	
 C-2   Steam-Assisted Flare Testing Summary  .........
 C-3   Summary of Thermal Incinerator Emission
      Test Results	 .
 C-4   Typical  Incinerator Parameters for ARCO
      Polymers Emission Testing Based on Data
      from August 1981  	
 C-5   ARCO Polymers Incinerator Destruction
      Efficiencies for Each Set of Conditions 	
 C-6   Air Oxidation Unit Thermal  Incinerator
      Field Test Data	
 C-7   Destruction Efficiency Under Stated Conditions
      Based on Results of Union Carbide  Laboratory
      Tests	
 C-8   Comparisons of Emission Test Results for Union
      Carbide Lab Incinerator and ROHM & HAAS Field
      Incinerator 	
C-9   Equipment Counts and Emissions for Fugitive
      VOC Emission Sources in SOCMI  Model  Units  	
C-10  Equipment Inventories and Emission Estimates for
      Fugitive VOC Emission Sources  in Polymers
      and Resins Plants 	
 9-45
 9-47


 9-51

 9-53

 C-6
 C-7

 C-8


 C-l 3

 C-l 6

 C-21


C-34


C-38

C-42


C-44
                                    xvn

-------
E-l   Procedure to Design State-of-the-Art (0.5 Mach)
      Elevated Steam-Assisted Smokeless Flares   	
E-2   Flare Budget Purchase Cost Estimates Provided
      by National Air Oil Burner, Inc., In October 1982
      Dollars 	
E-3   Capital and Annual Operating Cost Estimation
      Procedure for State-of-the-Art Steam-Assisted
      Smokeless Flares  ... 	 	
E-4   Worksheet for Calculation of Waste Gas
      Characteristics (molecular weight, molecular formula,
      lower heating value in Btu/scf)	
E-5   Generalized Waste Gas Combustion Calculations.   .  .  .
E-6   Procedure to Design Thermal Incinerators
      Combusting Streams With Lower Heating Values
      (LHV) Greater than 60 Btu/scf . 	
E-7   Capital and Annual Operating Cost Estimates for
      Thermal Incinerators Without Heat Recovery  	
E-8   Operating Parameters and Fuel Requirements
      of Catalytic Incinerator Systems 	
E-9   Gas Parameters Used for Estimating Capital and
      Operating Costs of Catalytic Incinerators 	
E-10  Catalytic Incinerator Vendor Cost Data	
E-ll  Capital and Operating Cost Estimation
      for Catalytic Incinerator Systems 	
E-12  Procedure to Calculate Heat Load of a
      Condensation System for Styrene in Air	
E-l3  Procedures to Calculate Heat Transfer Area of a
      Condensation System of Styrene in Air 	
E-14  Capital and Annual Operating Cost Estimation for
      Condensers with Refrigeration 	
E-15  EG Recovery Costs for Baseline System, for PET Plants
      Producing a Low Viscosity Product or a High Viscosity
      Product with a Single End Finisher	
E-l6  EG Recovery Costs for Baseline System for PET Plants
      Producing a High Viscosity Product with Multiple
      End Finishers 	
E-3
E-8
E-12
E-l 7
E-19
E-20

E-2 6

E-2 9

E-31
E-34

E-37

E-41

E-45

E-50


E-54


E-57
                                   xvi i i

-------
E-17  Piping and Ducting Design Procedure 	
E-18  Piping Components 	
E-19  Installed Piping Costs.	  	
E-20  Installed Ducting Cost Equations,  December  1977
      Dollars	
E-61
E-62
E-63

E-64
                                   xix

-------
                             List of Figures

                                                                    Page

3-1   General Polymerization Process 	  ....     3-12
3-2   Simplified Process Block Diagram for the Polypropylene
      Continuous, Liquid Phase Slurry Process  	     3-15
3-3   Simplified Process Block Diagram for the Polypropylene
      Gas Phase Process	     3-21
3-4   Simplified Process Block Diagram for the Low Density
      Polyethylene High-Pressure Process 	     3-25
3-5   Simplified Process Block Diagram for the Low Density
      Polyethylene Low-Pressure and High Density Polyethylene
      Gas Phase Processes	     3-29
3-6   Simplified Process Block Diagram for the High Density
      Polyethylene Liquid Phase Slurry Process 	     3-33
3-7   Simplified Process Block Diagram for the High Density
      Polyethylene Liquid Phase Solution Process 	     3-37
3-8   Simplified Process Block Diagram for the Polystyrene
      Batch Process	     3-42
3-9   Simplified Process Block Diagram for the Polystyrene
      Continuous Process   	     3-46
3-10  Simplified Process Block Diagram for the Expandable
      Polystyrene Post-Impregnation Suspension Process  	     3-50
3-11  Simplified Process Block Diagram for the Expandable
      Polystyrene In-Situ Suspension Process 	     3-52
3-12  Simplified Process Block Diagram for the PET/DMT
      Process	     3-57
3-13  Simplified Process Block Diagram for the PET/TPA
      Process	    3-60
4-1   Steam Assisted Elevated Flare System  	    4-4
4-2   Steam Injection Flare Tip	    4-5
4-3   Discrete Burner Thermal Incinerator 	    4-16
4-4   Distributed Burner Thermal Incinerator  	    4-16
4-5   Catalytic Incinerator		    4-20
4-6   Condensation System 	    4-28

                                     xx

-------
4-7  Two Stage Regeneration Adsorption System .  ,  .  .  .  .  .
4-8  Packed Tower for Gas Absorption	  .
C-l  Flare Sampling and Analysis System ..........
C-2  Schematic of Incineration System at ARCO
     Polypropylene Facility	'.."..
C-3  Incinerator Combustion Chamber 	
C-4  Petro-Tex Oxo Unit Incinerator	
C-5  Off-gas Incinerator, Monsanto Co.,
     Chocolate Bayou Plant	
C-6  Thermal Incinerator Stack Sampling System	.  .
E-l  Interpolation of Steam Bleed Rates for
     Intermittent Flares. .................
E-2  Estimated Flare Purchase Cost for 40 ft Height  .  .  .  .
E-3  Approximate Fluidic Seal  Costs ............
E-4  Purchase Costs for Thermal  Incinerator Combustion
     Chambers 	 ....... 	
E-5  Installed Capital  Costs for Inlet Ducts,  Waste  Gas
     and Combustion Air Fans,  and Stack for Thermal
     Incinerator Systems with  no Heat  Recovery	
E-6  Installed Capital  Costs for Catalytic  Incinerators
     With and Without Heat Recovery 	
4-30
4-33
C-4

C-ll
C-18
C-2 6

C-31
C-32

E-7
E-10
E-ll

E-23
E-24
E-35
                                   xxi

-------

-------
                             1.0  SUMMARY

1.1  REGULATORY ALTERNATIVES
     Standards of performance for stationary sources  of volatile
organic compounds (VOC) from process and fugitive  emission  sources  in
the polymers.and resins industry are being developed  under  the  authority
of Section lll'of the Clean Air Act.  These standards would,  in general,
affect new and modified/reconstructed existing  facilities that  produce
the following  basic polymers:  polypropylene, polyethylene, polystyrene,
and poly(ethylene terephthalate).  The fugitive emission standards
would not apply to poly(ethylene terephthalate) facilities.
     Because of production and emission differences,  twelve model plants
and regulatory alternatives specific to each model  plant were developed.
The model plants and their regulatory alternatives are presented  in
detail  in Chapter 6.  Regulatory Alternative 1  for each model plant
represents the levels of control within each industry segment in  the
absence of new regulations.  It provides the basis for comparison of
the impacts of the other regulatory alternatives.
     Regulatory Alternative 2 for each model plant except poly(ethylene
terephthalate) plants examines the control  of fugitive emissions.   These
requirements are as follows:
     o    Monthly monitoring for leaks from valves in gas and light
          liquid service, and pump seals in light  liquid services;
     o    Weekly visual inspection for liquid leakage from  pump seals in
          light liquid service;
     o    Installation of controlled degassing  vents  on compressors,
          rupture disks on relief valves, and caps on open-ended  lines;
          and
     o    Closed-purge sampling on sampling connections.
This particular set of requirements was adopted based upon  the  results
of alternative levels of fugitive emission control  already  analyzed
for the proposed fugitive emission standards for the  synthetic  organic
chemical manufacturing industry (SOCMI).  Similarities between  the  polymers
                                    1-1

-------
 and  resins  industry  and  SOCMI  enable  the  results  to  be  transferred.
 As this  particular set of  fugitive emission controls was already found
 to be  reasonable  and representative of  the best system  for  reducing VOC
 fugitive emissions,  no other fugitive emission control  alternatives
 were considered.
     For the  poly(ethylene terephthalate) model plants, Regulatory
 Alternative 2 examines the reduction  of either methanol emissions using
 a condenser or  ethylene  glycol emissions  from the cooling tower by
 using  a  distillation column or making greater use of an existing dis-
 tillation column.  No additional regulatory alternatives were developed
 for  the  poly{ethylene terephthalate)  model plant using  the  terephthalic
 acid process  making  a low  viscosity product or a high viscosity product
 with a single end finisher per process  line.
     Additional regulatory alternatives for the eleven  remaining model
 plants further  reduce emissions through the additional control of
 process  emissions.   For  the polypropylene, polyethylene, and expandable
 polystyrene model plants,  additional  control was achieved by applying
 combustion  devices to groups of emission  streams that were  combined on
 the  basis of  their emanating from equipment performing a particular
 task,  such  as polymerization or material  recovery, within a production
 line.  For  the  crystal or  impact polystyrene model plant, additional
 control  was obtained  by  applying additional recovery to the process
 emissions.  For poly(ethylene  terephthalate) plants  using the dimethyl
 terephthalate process, additional control was obtained by greater
 recovery  of the methanol  stream from  the methanol  recovery system and
 by recovery of  ethylene  glycol  emissions from the cooling tower with a
 distillation  column.  For poly(ethylene terephthalate) plants using the
 terephthalic  acid process  for  producing a high viscosity product with
multiple end  finishers per process line, additional  control  was obtained
 by making greater use of the existing distillation column.
 1.2  ENVIRONMENTAL IMPACT
     The environmental and energy impacts of each  regulatory alternative
for each model plant are  presented in  Chapter 7.   Table 1-1  presents a
summary of the aggregate  environmental and energy  impacts;  that is,
each model plant's Regulatory Alternative 1 is combined and  the resulting
impact is reported under  Regulatory Alternative I.  Similarly,  Regulatory

                                  1-2

-------
           Table  1-1.   ASSESSMENT  OF  ENVIRONMENTAL AND ECONOMIC IMPACTS
                    FOR EACH REGULATORY ALTERNATIVE CONSIDERED

Administrative
Action
Regulatory
Alternative I
Regulatory
Alternative II
Regulatory
Alternative III
Regulatory
Alternative IV
Regulatory
Alternative V
Regulatory
Alternative VI
Regulatory
Alternative VII

Air
impact
_2**
4-2**
44**
44**
44**
44**
44**
Solid
Water waste
impact impact
0 0
+1** o
4-1** -1*
4-1** -1*
+1** -1*
4-1** -1*
+1** -1*

Energy
impact
0
-1*
-2*
-2*
_2*
-3*
-3*

Noi se
impact
_!**
0
_!**
-2**
_2**
_2**
_2**

Economic
impact
0
-1*
-1*
-!*•
-1*
-2*
_2*
KEY: + Beneficial  impact
     - Adverse impact
0 No impact
1 Negligible impact
2 Small impact
3 Moderate impact
4 Large impact
  * Short term impact
 ** Long-term impact
*** irreversible impact
                                    1-3

-------
 Alternative 2 impacts  for  each  model  plant  are  totaled  as  Regulatory
 Alternative II.   Regulatory  Alternative  III  represents  Regulatory
 Alternative 3 impacts  for  ten model  plants  plus Regulatory Alternative 2
 impacts  for the  polypropylene model  plant using the  gas phase process
 and the  poly(ethyl,ene  terephthalate)  model  plant using  the terephthalatic
 acid process producing a low viscosity product  or a  high viscosity
 product  with a single  end  finisher per process  line  as  no  Regulatory
 Alternative 3 was  developed  for these model  plants.  Regulatory Alternative
 IV  corresponds to  Regulatory Alternative 4  for  eight model plants plus
 Regulatory  Alternative 3 impacts for  two model  plants plus Regulatory
 Alternative 2 impacts  for  two model  plants.  Regulatory Alternatives V,
 VI, and  VII correspond to  regulatory  alternatives combined using the
 same procedure as  outlined above.  Regulatory VII, thus, represents the
 aggregate impact of the most stringent regulatory alternatives considered
 for each model plant.
 1.2.1  Air  Emissions Impacts
     Total  VOC emissions from new process lines  in these industry segments
 in  1988  are  projected  to be  approximately 7.33  gigagrams (Gg) under
 Regulatory  Alternative  I,  compared to 5.84,  3.62, 3.18, 2.87, 2.80, and
 2.77 Gg  under  Regulatory Alternatives II, III,  IV, V, VI, and VII,
 respectively.  The average percent emission  reductions from the Regulatory
 Alternative  I  level achieved by Regulatory Alternatives II, III, IV, V,
 VI  and VII  are 20, 51, 57, 61,  61.8, and 62.2 percent, respectively.
 1.2.2  Water,  Solid Waste, and  Noise  Impacts
     Little  adverse affect on water quality  is expected under any of
 the regulatory alternatives.  Implementation of fugitive controls
 under Regulatory Alternative II would result in a small  positive effect
 on water by curtailment of potential  liquid leaks.
     Minor adverse solid waste impacts could occur under Regulatory
 Alternatives III through VII due to the use of catalytic incinerators.
 Spent catalyst from catalytic incinerator use may be generated at an
annual rate of up to 4.21 m3 (149 ft3) under Regulatory Alternative VII.
     Some noise impact could arise  from increased use of flares  under
the regulatory alternatives.   By employing  noise mitigation techniques,
additional  noise impact on surrounding communities should  be  minimal.
                                  1-4

-------
1.2.3  Energy Impacts
     Under Regulatory Alternative II, implementation-of fugitive
controls in nine of the twelve model plants and of additional  methanol
or ethylene glycol recovery in poly(ethylene terephthal ate)  plants
result in a net increase in energy usage of 66 TJ (11,000 bbl  oil)
from what otherwise would occur under Regulatory Alternative 1.  This
increase is due solely to the energy required to implement Regulatory
Alternative 2 for the poly(ethylene terephthal ate) model  plant using
the terephthalic acid process producing a low viscosity product or a
high viscosity product with a single end finisher per process line.
Excluding this model plant, there would be a net decrease in energy
usage of 65 TJ (10,000 bbl oil) under Regulatory Alternative II because
the energy credit obtained through recovered VOC is greater  than the
energy expended to implement the controls.
     Under Regulatory Alternatives III through VII, there is a net
increase in energy usage as the energy needed to implement process VOC
emission controls exceeds the energy credit obtained through recovered
VOC.  The net annual energy increase ranges from about 370 TJ  (61,000
bbl  oil) under Regulatory Alternative III (approximately  a 37 percent
increase over Regulatory Alternative I energy usage) up to about 940
TJ (154,000 bbl  oil) under Regulatory Alternative VII, (approximately a
93 percent increase over Regulatory Alternative I energy  usage).
1.3  ECONOMIC IMPACT
     As was done for the environmental and energy impacts, the
aggregate economic impacts that result from the costs for each of the
regulatory alternatives are summarized in Table 1-1.  A more detailed
economic analysis is presented in Chapter 9 and a more detailed analysis
of costs for each industry segment is presented in Chapter 8.
     Under Regulatory Alternative II, the industry as a whole would
spend a net annual amount of around $1.9 million in the fifth year
(1988) compared to what the industry as a whole would otherwise spend
under Regulatory Alternative I.  Fifth year annual costs  compared to
Regulatory Alternative I for individual industry segments range from
an annual cost of $600 for a poly(ethylene terephthalate) process
line using the dimethyl  terephthalate process up to $105,000 for
                                  1-5

-------
a poly(ethylene terephthalate) process line using the terephthalic acid
process producing a low viscosity product or a high viscosity product
with a single end finisher.
     Under Regulatory Alternative III, the industry as a whole would
spend an net annual amount of $7.7 million in the fifth year (1988)
over and above what they would otherwise spend under Regulatory
Alternative I.  Fifth year annual costs for individual industry segments
range from an annual cost of $600 for a poly(ethylene terephthalate)
process line using the dimethyl terephthalate process up to $404,000
for a high density polyethylene process line using the liquid phase
solution process.
     Under Regulatory Alternatives IV, V, and VI, fifth year annual
costs of $10.2 million, $12.9 million, and $14.6 million, respectively,
over and above what would be otherwise spent under Regulatory Alternative
I would be realized by the industry as a whole.  Annual costs for
individual industry segments range from net annual costs of $600 for a
poly(ethylene terephthalate) process line using the dimethyl terephthalate
process up to a cost of $1.2 million (under Regulatory Alternative 6)
for an expandable polystyrene process line using the post-impregnation
suspension process.
     Under the most costly combination of individual regulatory alterna-
tives for each model plant, Regulatory Alternative VII, total additional
annualized costs of controls in 1988 are estimated to be $15.4 million.
The potential adverse economic impacts of these regulatory alternatives
are expected to be minor in view of the generally small'price increases
anticipated as a result of control costs.
                                    1-6

-------
                             2.0  INTRODUCTION

2.1  BACKGROUND AND AUTHORITY FOR STANDARDS
     Before standards of performance are proposed as a Federal regulation,
air pollution control methods available to the affected industry and the
associated costs of installing and maintaining the control equipment are
examined in detail.  Various levels of control based on different technolo-
gies and degrees of efficiency are expressed as regulatory alternatives.
Each of these alternatives is studied by EPA as a prospective basis for a
standard.  The alternatives are investigated in terms of their impacts on
the economics and well-being of the industry, the impacts on the national
economy, and the impacts on the environment.  This document summarizes the
information obtained through these studies so that interested persons will
De able to see the information considered by EPA in the development of the
proposed standard.
     Standards of performance for new stationary sources are established
under Section 111 of the Clean Air Act (42 U.S.C. 7411) as amended, herein-
after referred to as the Act.  Section 111 directs the Administrator to
establish standards of performance for any category of new stationary
source of air pollution which "... causes, or contributes significantly
to air pollution which may reasonably be anticipated to endanger public
health or welfare."
     The Act requires that standards of performance for stationary sources
reflect,"... the degree of emission reduction achievable which (taking
into consideration the cost of achieving such emission reduction, and any
nonair quality health and environmental impact and energy requirements) the
Administrator determines has been adequately demonstrated for that category
of sources."  The standards apply only to stationary sources, the construc-
tion or modification of which commences after regulations are proposed by
publication in the Federal Register.
                               2-1

-------
     The 1977 amendments to the Act altered or added numerous provisions
that apply to the process of establishing standards of performance.
Examples of the effects of the 1977 amendments are:
     1.  EPA is required to review the standards of performance every
4 years and, if appropriate, revise them.
     2.  EPA is authorized to promulgate a standard based on design,
equipment, work practice, or operational procedures when a standard
based on emission levels is not feasible.
     3.  The term "standards of performance" is redefined, and'a new
term "technological system of continuous emission reduction" is defined.
The new definitions clarify that the control system must be continuous
and may include a low- or non-polluting process or operation.
     4.  The time between the proposal and promulgation of a standard
under section 111 of the Act may be extended to 6 months.
     Standards of performance, by themselves, do not guarantee protection
of health or welfare because they are not designed to achieve any
specific air quality levels.  Rather, they are designed to reflect the
degree of emission limitation achievable through application of the
best adequately demonstrated technological system of continuous emission
reduction, taking into consideration the cost of achieving such emission
reduction, any nonair quality health and environmental impacts, and
energy requirements.
     Congress had several reasons for including these requirements.
First, standards with a degree of uniformity are needed to avoid
situations where some States may attract industries by relaxing standards
relative to other States.  Second, stringent standards enhance the
potential for long-term growth.  Third, stringent standards may help
achieve long-term cost savings by avoiding the need for more retrofitting
when pollution ceilings may be reduced in the future.  Fourth, certain
types of standards for coal burning sources can adversely affect the
coal market by driving up the price of low-sulfur coal or effectively
excluding certain coals from the reserve base because their untreated
pollution potentials are high.  Congress does not intend that new
source performance standards contribute to these problems.  Fifth, the
standard-setting process should create incentives for improved technology.
                                2-2

-------
     Promulgation of standards of performance does not prevent State
or local agencies from adopting more stringent emission limitations
for the same sources.  States are free under Section 116 of the Act to
establish even more stringent emission limits than those established
under Section 111 or those necessary to attain or maintain the National
Ambient Air Quality Standards (NAAQS) under Section 110.  Thus, new
sources may in some cases be subject to limitations more stringent
than standards of performance under Section 111, and prospective
owners and operators of new sources should be aware of this possibility
in planning for such facilities.
     A similar situation may arise when a major emitting facility is
to be constructed in a geographic area that falls under the prevention
of significant deterioration of air quality provisions of Part C of
the Act.  These provisions require, among other things, that major
emitting facilities to be constructed in such areas are to be subject
to best available control technology.  The term Best Available Control
Technology (BACT), as defined in the Act, means
     ... an emission limitation based on the maximum degree of reduction
     of each pollutant subject to regulation under this Act emitted
     from, or which results from, any major emitting facility, which the
     permitting authority, on a case-by-case basis, taking into account
     energy, environmental, and economic impacts and other costs, determines
     is achievable for such facility through application of production
     processes and available methods, systems, and techniques, including
     fuel  cleaning or treatment or innovative fuel combustion techniques
     for control  of each such pollutant.  In no event shall application
     of "best available control  technology" result in emissions of any
     pollutants which will exceed the emissions allowed by any applicable
     standard established pursuant to Sections 111 or 112 of this Act.
     (Section 169(3))
     Although standards of performance are normally structured in
terms of numerical emission limits where feasible, alternative approaches
are sometimes necessary.  In some cases physical measurement of emissions
from a new source may be impractical or exorbitantly expensive.
Section lll(h) provides that the Administrator may promulgate a design
or equipment standard in those cases where it is not feasible to
prescribe or enforce a standard of performance.  For example, emissions
of hydrocarbons from storage vessels for petroleum liquids are greatest
during tank filling.  The nature of the emissions, high concentrations
                               2-3

-------
for short periods during filling and low concentrations for longer
periods during storage, and the configuration of storage tanks make
direct emission measurement impractical.  Therefore, a more practical
approach to standards of performance for storage vessels has been
equipment specification.
     In addition, Section lll(j) authorizes the Administrator to grant
waivers of compliance to permit a source to use innovative continuous
emission control technology.  In order to grant the waiver, the Administra-
tor must find:  (1) a substantial likelihood that the technology will
produce greater emission reductions than the standards require or an
equivalent reduction at lower economic energy or environmental cost;
(2) the proposed system has not been adequately demonstrated; (3) the
technology will not cause or contribute to an unreasonable risk to the
public health, welfare, or safety; (4) the governor of the State where
the source is located consents; and (5) the waiver will not prevent
the attainment or maintenance of any ambient standard.  A waiver may
have conditions attached to assure the source will not prevent attainment
of any NAAQS.  Any such condition will have the force of a performance
standard.  Finally, waivers have definite end dates and may be terminated
earlier if the conditions are not met or if the system fails to perform
as expected.  In such a case, the source may be given up to three
years to meet the standards with a mandatory progress schedule.
2.2  SELECTION OF CATEGORIES OF STATIONARY SOURCES
     Section 111 of the Act directs the Adminstrator to list categories
of stationary sources.  The Administrator "... shall include a category
of sources in such list if in his judgment it causes, or contributes
significantly to, air pollution which may reasonably be anticipated to
endanger public health or welfare."  Proposal and promulgation of
standards of performance are to follow.
     Since passage of the Clean Air Act of 1970, considerable attention
has been given to the development of a system for assigning priorities
to various source categories.  The approach specifies areas of interest
by considering the broad strategy of the Agency for implementing the
Clean Air Act.  Often, these "areas" are actually pollutants emitted
by stationary sources.  Source categories that emit these pollutants
                               2-4

-------
are evaluated and ranked by a process involving such factors as:
(1) the level of emission control (if any) already required by State
regulations, (2) estimated levels of control that might be required
from standards of performance for the source category, (3) projections
of growth and replacement of existing facilities for the source category,
and (4) the estimated incremental amount of air pollution that could
be prevented in a preselected future year by standards of performance
for the source category.  Sources for which new source performance
standards were promulgated or under development during 1977, or earlier,
were selected on these criteria.
     The Act amendments of August 1977 establish specific criteria to
be used in determining priorities for all major source categories not
yet listed by EPA.  These are:  (1) the quantity of air pollutant
emissions that each such category will emit, or will be designed to
emit; (2) the extent to which each such pollutant may reasonably be
anticipated to endanger public health or welfare; and (3) the mobility
and competitive nature of each such category of sources and the consequent
need for nationally applicable new source standards of performance.
     In some cases it may not be feasible immediately to develop a
standard .for a source category with a high priority.  This might
happen when a program of research is needed to develop control  techniques
or because techniques for sampling and measuring emissions may require
refinement.  In the developing of standards, differences in the time
required to complete the necessary investigation for different source
categories must also be considered.  For example, substantially more
time may be necessary if numerous pollutants must be investigated from
a single source category.  Further, even late in the development
process the schedule for completion of a standard may change.  For
example, inability to obtain emission data from well-controlled sources
in time to pursue the development process in a systematic fashion may
force a change in scheduling.  Nevertheless, priority ranking is, and
will continue to be, used to establish the order in which projects are
initiated and resources assigned.
     After the source category has been chosen, the types of facilities
within the source category to which the standard will apply must be
                               2-5

-------
determined.  A source category may have several facilities that cause
air pollution, and emissions from some of these facilities may vary
from insignificant to very expensive to control.  Economic studies of
the source category and of applicable control technology may show that
air pollution control is better served by applying standards to the
more severe pollution sources.  For this reason, and because there is
no adequately demonstrated system for controlling emissions from
certain facilities, standards often do not apply to all facilities at
a source.  For the same reasons, the standards may not apply to all
air pollutants emitted.  Thus, although a source category may be
selected to be covered by a standard of performance, not all pollutants
or facilities within that source category may be covered by the standards.
2.3  PROCEDURE FOR DEVELOPMENT OF STANDARDS OF PERFORMANCE
     Standards of performance must (1) realistically reflect best
demonstrated control practice; (2) adequately consider the cost, the
nonair quality, health and environmental impacts, and the energy requirements
of such control; (3) be applicable to existing sources that are modified
or reconstructed as well as new installations; and (4) meet these
conditions for all variations of operating conditions being considered
anywhere in the country.
     The objective of a program for developing standards is to identify
the best technological system of continuous emission reduction that
has been adequately demonstrated.  The standard-setting process involves
three principal phases of activity:  (1) information gathering, (2)
analysis of the information, and (3) development of the standard of
performance.
     During the information-gathering phase, industries are queried
through a telephone survey, letters of inquiry, and plant visits by EPA
representatives.  Information is also gathered from many other sources,
and a literature search is conducted.  From the knowledge acquired about
the industry, EPA selects certain plants at which emission tests are con-
ducted to provide reliable data that characterize the pollutant emissions
from well-controlled existing facilities.
     In the second phase of a project, the information about the industry
and the pollutants emitted is used in analytical studies.  Hypothetical
                               2-6

-------
 "model  plants" are defined to  provide  a  common  basis  for analysis.   The
 model  plant definitions, national  pollutant  emission  data,  and  existing
 State  regulations governing emissions  from the  source category  are  then
 used in establishing  "regulatory alternatives."  These regulatory
 alternatives are essentially different levels of emission control.
     EPA conducts studies to determine the impact of  each regulatory
 alternative on the economics of the industry and on the national economy,
 on the environment, and on energy  consumption.  From  several  possibly
 applicable alternatives, EPA selects the single most  plausible  regulatory
 alternative as the basis for a standard of performance for  the  source
 category under study.
     In the third phase of a project,  the selected regulatory alternative
 is translated into a standard of performance, which,  in  turn, is written in
 the form of a Federal regulation.  The Federal regulation,  when applied to
 newly constructed plants, will limit emissions to the  levels  indicated  in
 the selected regulatory alternative.
     As early as is practical  in each  standard-setting  project, EPA
 representatives discuss the possibilities of a standard  and the form it
might take with members of the National Air Pollution  Control Techniques
Advisory Committee.  Industry representatives and other  interested parties
 also participate in these meetings.
     The information acquired in the project is summarized  in the Background
 Information Document (BID).   The BID,  the standard, and  a preamble explain-
 ing the standard are widely circulated to the industry  being considered for
control, environmental groups, other government agencies, and offices
within EPA.  Through this extensive review process, the  points of view of
expert reviewers are taken into consideration as changes  are made to the
documentation.
     A "proposal  package" is assembled and sent through  the offices of EPA
Assistant Administrators for concurrence before the proposed standard is
officially endorsed by the EPA Administrator.  After being  approved by the
EPA Administrator,  the preamble and the proposed regulation are published
 in the Federal  Register.
     As a part of the Federal  Register announcement of the proposed
regulation, the public is invited to participate in the  standard-setting
process.  EPA invites written comments on the proposal and also holds a

                               2-7

-------
public hearing to discuss the proposed standard with interested parties.
All public comments are summarized and incorporated into a second volume
of the BID.  All information reviewed and generated in studies in support
of the standard of performance is available to the public in a "docket" on
file in Washington, D. C.
     Comments from the public are evaluated, and the standard of performance
may be altered in response to the comments.
     The significant comments and EPA's position on the issues raised are
included in the "preamble" of a "promulgation package," which also contains
the draft of the final regulation.  The regulation is then subjected to
another round of review and refinement until it is approved by the EPA
Administrator.  After the Administrator signs the regulation, it is published
as a "final rule" in the Federal Register.
2.4  CONSIDERATION OF COSTS
     Section 317 of the Act requires an economic impact assessment with
respect to any standard of performance established under Section 111
of the Act.  The assessment is required to contain an analysis of:
(1) the costs of compliance with the regulation, including the extent to
which the cost of compliance varies depending on the effective date of
the regulation and the development of less expensive or more efficient
methods of compliance; (2) the potential inflationary or recessionary
effects of the regulation; (3) the effects the regulation might have on
small business with respect to competition; (4) the effects of the regulation
on consumer costs; and (5) the effects of the regulation on energy use.
Section 317 also requires that the economic impact assessment be as
extensive as practicable.
     The economic impact of a proposed standard upon an industry is usually
addressed both in absolute terms and in terms of the control costs that  .
would be incurred as a result of compliance with typical, existing State
control regulations.  An incremental approach is necessary because both new
and existing plants would be required to comply with State regulations in
the absence of a Federal  standard of performance.  This approach requires a
detailed analysis of the economic impact from the cost differential  that
would exist between a proposed standard of performance and the typical
State standard.
                               2-8

-------
     Air pollutant emissions may cause water pollution problems, and
captured potential air pollutants may pose a solid waste disposal
problem.  The total  environmental impact of an emission source must,
therefore, be analyzed and the costs determined whenever possible.
     A thorough study of the profitability and price-setting mechanisms
of the industry is essential to the analysis so that an accurate
estimate of potential adverse economic impacts can be made for proposed
standards.  It is also essential to know the capital requirements for   •
pollution control systems already placed on plants so that the additional
capital requirements necessitated by these Federal standards can be
placed in proper perspective.  Finally, it is necessary to assess the
availability of capital to provide the additional control equipment
needed to meet the standards of performance.
2.5  CONSIDERATION OF ENVIRONMENTAL IMPACTS
     Section 102(2)(C) of the National Environmental Policy Act  (NEPA)
of 1969 requires Federal agencies to prepare detailed environmental
impact statements on proposals for legislation and other major Federal
actions significantly affecting the quality of the human environment.
The objective of NEPA is to build into the decisionmaking process of
Federal agencies a careful consideration of all  environmental aspects
of proposed actions.
     In a number  of  legal challenges to standards of performance for
various industries,  the United States Court of Appeals for the District
of Columbia Circuit  has held  that environmental  impact statements need
not be  prepared by the Agency for proposed actions  under Section 111
of the  Clean Air  Act.  Essentially, the Court  of Appeals has  determined
that the  best system of emission reduction requires the Administrator
to take into account counter-productive environmental effects of a
proposed  standard, as well  as economic costs  to  the industry.   On this
basis,  therefore, the  Court established a  narrow exemption from  NEPA
for EPA determination  under Section  111.
      In addition  to  these judicial  determinations,  the  Energy Supply
and Environmental Coordination  Act  (ESECA) of 1974  (PL-93-319)  specifically
exempted  proposed actions under the Clean  Air Act from  NEPA  requirements.
According to  Section 7(c)(l), "No  action  taken under  the  Clean Air  Act
                                2-9

-------
shall be deemed a major Federal action significantly affecting the
quality of the human environment within the meaning of the National
Environmental Policy Act of 1969." (15 U.S.C. 793(c)(l))
     Nevertheless, the Agency has concluded that the preparation of
environmental impact statements could have beneficial effects on
certain regulatory actions.  Consequently, although not legally required
to do so by Section 102(2)(C) of NEPA, EPA has adopted a policy requiring
that environmental impact statements be prepared for various regulatory
actions, including standards of performance developed under Section 111
of the Act.  This voluntary preparation of environmental impact statements,
however, in no way legally subjects the Agency to NEPA requirements.
     To implement this policy, a separate section in this document is
devoted solely to an analysis of the potential environmental impacts
associated with the proposed standards.  Both adverse and beneficial
impacts in such areas as air and water pollution, increased solid
waste disposal, and increased energy consumption are discussed.
2.6  IMPACT ON EXISTING SOURCES
     Section 111 of the Act defines a new source as  "... any stationary
source, the construction or modification of which is commenced  ..."
after the proposed standards are published.  An existing source is
redefined as a new source if "modified" or "reconstructed" as defined
in amendments to the general provisions of Subpart A of 40 CFR  Part
60,  which were promulgated in  the  Federal Register on December  16,
1975 (40 FR 58416).
     Promulgation of a standard of performance requires States  to
establish standards of performance for existing sources in the  same
industry under Section lll(d)  of the Act  if  the standard for new
sources limits emissions of  a  designated  pollutant  (i.e., a pollutant
for  which air quality criteria  have not been  issued  under Section  108
or which has not been listed as a  hazardous  pollutant under Section  112).
If a State does not act, EPA must  establish  such standards.  General
provisions outlining procedures for control  of existing sources under
Section lll(d) were promulgated on November  17, 1975, as Subpart B of
40 CFR  Part  60  (40  FR 53340).
                                2-10

-------
2.7  REVISION OF STANDARDS OF PERFORMANCE
     Congress was aware that the level of air pollution control achievable
by any industry may improve with technological advances.  Accordingly,
Section 111 of the Act provides that the Administrator "... shall, at
least every 4 years, review and, if appropriate, revise ..." the
standards.  Revisions are made to assure that the standards continue
to reflect the best systems that become available in the future.  Such
revisions will not be retroactive, but will apply to stationary sources
constructed or modified after the proposal of the revised standards.
                               2-11

-------

-------
                 3.0  THE POLYMERS AND RESINS INDUSTRY

     The polymers and resins industry consists of operations that combine
monomer or chemical intermediate materials obtained from the basic
petrochemical industry and the synthetic organic chemical manufacturing
industry (SOCMI) into polymeric or copolymeric products.  (A copolymer
is formed when two different monomers are polymerized together so that
both occur in the same polymer chain.  The copolymer will generally
combine to some extent the properties of the individual polymers and
will often have lower strength and a lower melting point than either of
the polymers.)  Such products include plastic materials, synthetic
resins, synthetic rubbers, and synthetic fibers.  This chapter describes
the polymers and resins industry, its production processes and associated
emissions of volatile organic compounds (VOC), and industry practices
and State and Federal regulations affecting VOC emissions.
3.1  INDUSTRY DESCRIPTION
     A large number of polymers are produced domestically by a variety
of processes.  Polymers can be grouped into two categories:   thermoplastic,
those which melt upon reheating and, thus, can be reshaped after initial
fabrication, and thermosetting, those which do not.  Thermoplastic
polymers are linear chain polymers with little or no crosslinking between
the individual chains.  Common end-uses include safety shields, clothing,
appliance parts, boiling bags, sutures, textiles and woven goods, bottles
for a variety of fluids, toys, and hot/cold insulated drink cups.
Thermosetting polymers are extensively crosslinked, making them far more
rigid and often insoluble.  These resins are often used in applications
where rigidity or heat-resistant characteristics are required.  End-uses
include molding compounds, adhesives and bonding resins, laminating
resins, paper and surface coatings, and as a binder for fiber glass and
other reinforced plastics for construction and transportation applications.
                               3-1

-------
Acrylics
Al kyds
High Density Polyethylene
Low Density Polyethylene
Mel amine Formaldehyde
Nylon 6
Nylon 66
Phenol Formaldehyde
The selection of a suitable polymer for a particular end-use application
depends on the specific properties of the polymer.
     The U.S. Environmental Protection Agency (EPA) study by the
Pullman-Kenogg Company ranked segments of the polymers and resins
industry by magnitude of emissions for the purpose of new source
performance standard (NSPS) development priority setting.1  This study
examined 16 polymers and resins segments with potentially large VOC
emissions:
                                           Polyester Fibers
                                           Polypropylene
                                           Polystyrene
                                           Polyvinyl Acetate
                                           Polyvinyl Alcohol
                                           Styrene-Butadiene Latex
                                           Unsaturated Polyester Resins
                                           Urea Formaldehyde
     The majority of these 16 polymers are of the thermoplastic type
(acrylics, polyethylene, nylon 6 and 66,  (saturated) polyester resin,
polypropylene, polystyrene, polyvinyl acetate, polyvinyl alcohol, and
styrene-butadiene latex); the remainder are thermosetting resins.
     The survey of these 16 polymers and  resins segments showed that
emissions from five of these categories amount to approximately 75 percent
of the current total estimated VOC emissions from these 16 polymers and
resins manufacturing operations.  These five source categories, all of
which were found to be experiencing growth, were chosen for NSPS development,
They are:
              Polypropylene;
              Low density polyethylene;
              High density polyethylene;
              Polystyrene; and
              Polyester resin, poly(ethylene terephthalate), [PET].
     Tables  3-1 to 3-5 list existing production locations and capacities
for plants that produce these five polymers and resins.
3.1.1  End-Uses of the Five Polymers Chosen for NSPS Development
     The  16  commercial polymers and resins covered  by the Pullman-Kellogg
report have  an extremely wide variety of  end-uses and are found in
 1.
 2.
 3.
 4.
 5.
                             3-2

-------
                 Table 3-1.   POLYPROPYLENE (PP) PLANT LIST'

Company
ARCO Polymers, Inc.
Amoco Chemical Corp.
El Paso Polyolefins Co.
Exxon Chemical Co.
Sulf Oil Chemical Co.
Hercules, Inc.
Northern Petrochemical Co.
Phillips Chemical Co.
Shell Chemical Co.
Soltex Polymer Corp.
Texas Eastman Co.
USS Chemicals
Location
La Porte, TX
Chocolate Bayou, TX
Odessa, TX
Pasadena, TX
Bay town, TX
Cedar Bayou, TX
Bayport, TX
Lake Charles, LA
Morris, IL
Pasadena, TX
Norco, LA
Woodbury, NJ
Deer Park, TX
Long view, TX
La Porte, TX
Kenova, WV
Capacity
Gg/yr
181
234
68
68
181,
181
204
395
91
91
136
136
91
64
159
75
aSource:  SRI International, 1982 Directory of Chemical Producers,
 United States.   (Docket Reference Number II-I-82.)
                                 3-3

-------
    Table 3-2.   LOW DENSITY  POLYETHYLENE  (LDPE)  PLANT  LIST'

Company
Allied Chemical Co.b
ARCO Polymers, Inc.
Cheroplex Co.
Cities Service Co.c
Dow Chemical U.S.A.
E.I. du Pont de Nemours & Co., Inc.
El Paso Polyolefins Co.
Exxon Chemical Co.
Gulf Oil Chemical Co.
Mobil Chemical Co.
National Distillers & Chemical Corp.
Northern Petrochemical Co.
Phillips Chemical Co.
Texas Eastman Co.
Union Carbide Corp.
United Foam Corp.
Location
Orange, TX
Port Arthur, TX
Clinton, IA
Lake Charles, LA
Freeport, TX
Plaquemine, LA
Orange, TX
Victoria, TX
Odessa, TX
Pasadena, TX
. Baton Rouge, LAH
Mt. Bel view, TXa
Cedar Bayou, TX
Orange, TX .
Baytown, TX
Beaumont, TX
Deer Park, TX
Tuscola, IL
Morris, IL
Pasadena, TX
Longview, TX
Seadrift, TX
Taft, LA
Penuelas, P.R.
Louisville, KY
Average
capacity,
Gg/yr
-
181
188
329 '
447
254
211
109
181
68
299d
154d
259
129
136
249
75
273e
_f
166
5449
272
141
73
 Source:  SRI International,  1982 Directory of Chemical  Producers,  United States,
 unless otherwise indicated.   (Docket Reference Number II-I-82.)

 Source:  Texas Air Control Board communication.  (Docket Reference Number  II-D-62.

cln a letter dated March'9, 1982, Cities Services Co.  indicated that they were
 closing all their polymer and resins manufacturing plants.   (Docket Reference
 Number II-D-46.)

 Source:  Chemical  Engineering.  October 18, 1982.  p. 26.   (Docket Reference
 Number II-I-102.)

Capacity obtained  from  April 14, 1982, letter from Northern Petrochemical
 Company.   (Docket  Reference  Number II-D-52.)

 Primarily HOPE produced; small portion of 680 Gg total  capacity used for LDPE.

9Capacity is about  245 Gg liquid phase and 299 Gg gas  phase.

 Source: Organic Chemical Producers Data Base.   Product  Data Report-Nationwide.
 February 17,  1981,  p. 926.   (Docket Reference Number  II-I-69.)
                               3-4

-------
        Table 3-3.  HIGH DENSITY POLYETHYLENE  (HOPE) PLANT LIST'
                                                                   Capacity,
             Company                         Location                Gg/yr
Allied Chemical Corp.
American Hoechst Corp.
Amoco Chemical Corp.
ARCO Polymers, Inc.
Chemplex Co.
Dow Chemical U.S.A.

E.I. du Pont de Nemours & Co., Inc.

Gulf Oil Corp.
Hercules, Inc.
Nat'l. Petrochemical Corp.
Phillips Chemical Co.
Soltex Polymer Corp.
Union Carbide Corp.
Baton Rouge, LA
Bayport, TX
Chocolate Bayou, TX
Port Arthur, TX
Clinton, IA
Freeport, TX
Plaquemine, LA
Orange, TX
Victoria, TX
Orange, TX
Lake Charles, LA
La Porte, TX
Pasadena, TX
Deer Park, TX
Seadrift, TX
299
136
172
159
122
 54
150
104
102
261b
  7
283
680
340
 91
 Source:  SRI International, 1982 Directory of  Chemical Producers,
 United States.   (Docket Reference  Number II-I-82.)
 ''Expansion from 191 Gg/yr to 261 Gg/yr completed  in  1982.  Chemical
 Engineering.  October 4, 1982.  p. 25.  (Docket Reference Number II-I-101.)
                                 3-5

-------
                 Table  3-4.    POLYSTYRENE  (PS)  PLANT LIST£
Company
A&E Plastics, Inc.
American Hoechst Corp.
Amoco Chemical Corp.
ARCO Polymers, Inc.
BASF Wyandotte Corp.
Carl Gordon, Ind., Inc.
Cosden Oil & Chemical Co.
Crest Container Corp.0
Dow Chemical Corp.
Gulf Oil Chemical Co.
Huntsman-Goodson Chan. Corp.
Kama Corp.
Mobil Chemical Co.
Monsanto Co.e
Polysar Resins, Inc.
Richardson Company
Shell Chemical Co.
Texstyrene Plastics, Inc.
U.S.S. Chemicals9
Vititek Inc.
Location
City of Industry, CA
Chesapeake, VA
Leominster, MA
Peru, IL
Joliet, IL
Torrance, CA
Willow Springs, IL
Beaver Valley, PA
South Brunswick, NJ
Owensboro, KY
Oxford, MA
Windsor, NJ
Calumet City, IL
Big Spring, TX
Orange, CA
Fort Worth, TX
Gales Ferry, CT
Midland, MI
Torrance, CA
Ironton, OH
Joliet, IL
Marietta, OH ,,
Channel view, TXa
Troy, OH
Hazel ton, PA
Holyoke, MA
Joliet, IL
Santa Ana, CA
Addyston, OH
Decatur, AL
Springfield, MA
Copley, OH
Leominster, MA
Forest City, NC
West Haven, CT
Belpre, OH
Fort Worth, TX
Haverhill, OH
Delano, CA
Capacity,
Gg/yr
16
118
54
113
136
16
41
213
79
23
45
54
122
20
27
3.6
86
100
91
86
64
141
18
9
11
41
18
29
136
45
136
54
54
18
_f
136
23
9
2
Process
-
-
Continuous
Batch
Batch
.
Batch
Batch
Batch ,
Continuous
Continuous
Continuous
Continuous
Continuous
'
.
Continuous
Continuous
Continuous
Continuous
Continuous
Continuous
•
-
aSource:  SRI International,  1982  Directory of Chemical Producers, United States,
 unless otherwise indicated.  (Docket Reference Number II-I-82.)
 Source:  Industry communications.

cSource:  Organic Chemical  Producers Data Base.  Product Data Report - Nationwide.
 February 17, 1981.   p.  943.  (Docket Reference Number II-I-68.)
 This plant is not currently  in  production.  Letter from Gulf Oil  to Texas Air
 Control Board.   July 28,  1982.  (Docket Reference Number II-I-93.)
 lionsanto's Long Beach plant  has been closed.

 In mid-1977, this company  switched its 18 Gg PS plant to production of other
 styrene copolymers.   Small quantities of specialty grade PS are still  being
 produced.

telephone conversation  on  October 12, 1982, with U.S.S. Chemicals indicated that
 this plant has  been  closed.  (Docket Reference Number II-E-33.}
e.,
                                     3-6

-------
 Table 3-5.   POLY(ETHYLENE  TEREPHTHALATE)  (PET)  POLYESTER PLANT  LIST*
Plant
Akzona Inc.

Allied Corp.
American Hoechst Corp.

Avtex Fibers, Inc.
E.I. du Pont de Nemours











Eastman Kodak Co.



Fiber Industries, Inc.



Firestone Tire &
Rubber Co.
Goodyear Tire &
Rubber Co.

ICI Americas Inc.
Minnesota Mining &
Manufacturing Co.
Monsanto Co.
Rohm and Haas Co.
Location
Central , SC
Lowland, TN
Moncure, NC
Spartanburg, SC
Greer, SC
Lewis town, PA
Camden, SC
Charleston, SC
Chattanooga, TN
Kins ton, NC
Old Hickory, TN
Wilmington, NC
Old Hickory, TN
Wilmington, NC
Brevard, NC
Par! in, NJ
Circleville, OH
Florence, SC
Columbia, SC
Kings port, TN
Rochester, NY
Windsor, CO
Fayetteville, NC
Florence, SC
Greenville, SC
Salisbury, NC
Shelby, NC

Hopewell, VA

Scottsboro, AL
Point Pleasant, WV
Hopewell, VA
Decatur, AL
Greenville, SC
Decatur, AL
Fayetteville, NC
Average
b capacity,
Process Gg/yr Product
DMT
DMT
TPA
DMT/TPA
'-
DMT/TPA


DMT/TPA



-
.
.
-
-
-
DMT
DMT
-
-
DMT/TPA0

DMT/TPA0
TPAC
DMT°

TPA

DMT/TPA
25 Fiber
48 Fiber
39 Fiber
261 Fiber, Bottle
32 Bottle Resins
18 Fiber


7-fi Fiber
/OD


249
566
14 Film
7 Film
29 Film
34 Film
204 Fiber
238 Fiber, Bottle
25 Film
11 Film

(36) Fiber, Bottle
680 Fiber


23 Fiber

11 Fiber, Bottle



Resins
, Film














Resins



Resins





Resins
122 Bottle Resins
-
_
-
DMT/TPA
36 Film
25 Film
10 Film
91 Fiber




91 Bottle Resins
Source:  SRI International, 1982 Directory of Chemical Producers, United States.
 (Docket Re'ference Number  II-I-82.)  Does not include manufacturers of unsaturated
 resins or processors using resins as a raw material (generally  to produce fibers).
 Saturated resins listing  is also not included as it is comprised of polyesters
 other than PET.

 DMT - Dimethyl terephthalate process.
 TPA - Terephthalic acid process.

°Industry correspondence.   (Docket Reference Number II-B-53.)
                                          3-7

-------
numerous sectors of the economy.  Forms include shapes for structural
housings or parts, films, sheets, surface coatings, adhesive liquids,
foams, fibers, and filaments.  Many types of manufacturing processes are
used to shape resin into these forms.  The various kinds of shaping
techniques used include blow molding, tubular film blowing, calendaring,
injection molding, rotational molding, casting, coating, extrusion,
foaming, and elongation to orient fibers.  These shaping operations are
noted, but not elaborated on in this document.  (This document discusses
the manufacturing of the above selected polymers but does not include
the fabrication of polymer products.)
     These products are used in every sector of the economy with
particularly large applications in the construction, transportation,
clothing, consumer goods, and electrical industries.  Generally, end-use
functions include structural components in equipment or appliances,
insulation, film for packaging wrap, and fiber for lines or clothing.
     Each major polymer or resin product has its own properties, forms,
and end-use sectors.  The important end-uses of each polymer chosen for
NSPS development are summarized below.
     3.1.1.1  End-Uses of Polypropylene.  Polypropylenes, which are made
by many different processes, are lightweight, water- and chemical-resistant
plastics, somewhat rigi'd, but easy to process.  They are thermoplastic
and belong to the olefins family.  Polypropylene products can be formed
in many ways, including molding, extrusion, rotational molding, powder
coating, thermoforming, foam molding, and fiber orientation.
     Molded applications include bottles for syrups and foods, caps,
auto parts, appliance parts, toys, housewares, and furniture components.
Polypropylene fibers and filaments are used in carpets, rugs, carpet
backing, woven bags, and cordage.  Film uses include packaging for
cigarettes, records, toys, and housewares.  Extrusions include pipes,
profiles, wire and cable coatings, and corrugated packing sheets.2
     Products formed by injection molding consume about 41 percent of
the polypropylene produced domestically.  The second most utilized
form, fibers and filaments, accounts for 31 percent of the total
production.  Other forms account for the remaining 28 percent.3
The major sectors using polypropylenes are consumer/institutional
(19 percent), fuririture/furnishings (18 percent), packaging (16 percent),
                               3-8

-------
transportation  (12 percent), and electrical/electronics  (7 percent).
Other uses account for the remaining 28 percent.3
     3.1.1.2  End-Uses of Polyethylenes:  Low Density and High Density.
Polyethylenes are the largest volume plastics produced,  both domestically
and internationally.  These thermoplastic polymers are valued for their
structural strength, water and chemical resistance, and  easy processing
characteristics.
     There is nearly an infinite variety of polyethylenes that differ in
melting point, clarity, and density.  They are generally divided into
two broad categories, low and high density, both of which are flexible,
although high density polyethylene (HOPE) is more rigid.  Within the
last few years, a new class of LOPE has appeared - linear low density
polyethylene (LLDPE).  LLDPE combines the linear molecular structure of
HOPE with the physical and optical properties of conventional LOPE, and
its overall properties are superior to those of conventional  LDPE.4
Polyethylenes are often extruded into film, sheets, pipe, or profiles,
injection molded, blow molded, rotationally molded, foamed, or formed in
other ways.5
     3.1.1.2.1  Low density polyethylene (LDPE).  Conventional LDPE is
used primarily in packaging.   Specific applications include packaging
film and wrap, trash bags, garment bags, and molded forms (toys, housewares,
containers, and others).*>  End-uses are found in many segments of the
economy, including the packaging industry (62 percent),  consumer/institutional
industries (11 percent), electrical/electronics industries (7 percent),
and other sectors (21 percent).6
     Like conventional LDPE,  LLDPE is suitable for many  end-uses.
Specific applications may include housewares, lids, closures, blow
molded parts (e.g., toys, bottles, and drums), wire and  cable insulation,
extruded pipe and tubing, and industrial and consumer films (e.g., food
packaging, trash bags, and garment bags).?
     3.1.1.2.2  High density  polyethylene (HOPE).   The primary application
for HOPE is the manufacture of blow molded bottles for bleaches, liquid
detergents, milk, and other fluids.  Other blow molded forms  for which
HOPE is used include automotive  gas tanks,  drums,  and carboys.  HOPE is
                               3-9

-------
also used for injection molded forms including material  handling  pallets,
stadium seats, trash cans, and auto parts.   The film is  used  in shopping
bags.5  HOPE is of special value where high impact  resistance is  required.
     Products formed by blow molding represent 40 percent of  the  total
domestic HOPE production.  Another 22 percent is injection molded, while
6 percent is attributed to film and sheet applications.   Other uses
account for 32 percent.8  End-use sectors for HOPE  include packaging
(45 percent), consumer/institutional (11 percent),  building and construction
(9 percent), and other sectors (35 percent).8
     3.1.1.3  End-Uses of Polystyrene.  Polystyrene plastics  are  durable,
provide good electrical insulation, and are easy to process.   This
thermoplastic is used in molded forms, extrusions,  liquid solutions,
adhesives, coatings, and foams.9
     Molded uses include toys, auto parts,  housewares, kitchen items,
appliances, wall tiles, refrigerated food containers, radio and tele-
vision housings, small appliance housings,  furniture, packages, and
building components such as shutters.  Extruded sheets also are used in
packaging, appliances, boats, luggage, and disposable plates.  Foamed
styrene is a good insulator and is used in construction, packaging,
boats, housewares, toys, and hot/cold insulated drink cups.9
     Fifty percent of the domestic polystyrene production is  molded  into
its consumer form.  An additional 33 percent of the domestic  production
is formed by extrusion, while other forming operations are used for  the
remaining 17 percent of polystyrene produced.10  Segments of  the  economy
using products from the polystyrene industry include the packaging
industry  (35 percent), the consumer/institutional industries  (22  percent),
and the building/construction and electrical/electronics industries
(10 percent each).  End-uses in all other sectors account for the remaining
23 percent.10
     3.1.1.4  End-Uses of Polyester Resin. Poly(Ethylene Terephthalate),
[PET].  Poly(ethylene terephthalate) [PET] polyester resins are  spun into
fiber, blown into film, molded into bottles and other forms,  or blended
into adhesive products.  Most of the PET produced in the U.S. is  used
for fiber production.
     Polyester fibers are used widely in clothes, textiles, and woven
goods.  They are thermoplastic polymers that retain their original
                               3-10

-------
 shape, enabling  clothing  to  have  permanent press characteristics.  A
 specialty  PET  fiber with  high density and high tensile strength is used
 for tires,  seat  belts, and other  industrial applications.  Some specialty
 PET molding and  extruding materials are engineering thermoplastics with
 high gloss, hard scratch  resistance, and high rigidity.
 3.2  POLYMERIZATION PROCESSES AND PROCESS EMISSIONS
      All processes for manufacturing the five polymers and resins chosen
 for NSPS development follow  a general series of steps and procedures.
 Figure 3-1  illustrates a  simplified stepwise process for polymer production.
 The manufacture  of a polymer may be considered as a five step operation:
      1.   Raw materials storage and preparation
      2.   Polymerization  reaction
      3.  Materials recovery
      4.  Product finishing
      5.  Product storage
      Raw materials storage and preparation includes methods of storing
 monomers and other raw materials to be used in the polymerization reaction.
 Raw material drying and other purification steps may be taken.  Raw
 materials are then routed to the polymerization reactor.
      In the reactor, raw materials and catalyst are combined with other
 processing materials to produce the polymer.   Reactor conditions, such
 as  temperature and pressure, are specific to the product being made.
After polymerization,  unreacted materials are recovered and returned  to
 raw material storage,  and the polymer is routed to product finishing.
     The product finishing stage of the polymerization process may
 include extruding and pelletizing, cooling and drying, introduction of
 additives, shaping operations, and curing operations.   The polymer is
 then  ready for product storage and shipping.
     Pollutant emissions from any chemical  process, including polymerization,
may be considered in two categories:  (1)  those that can be anticipated based
on the process flow diagram and (2)  those that can be  identified only by
sampling procedures (e.g., leakage at valves,  pumps, compressors,  and
 flanges).  The first type of emissions  will  be referred to as "process"
emissions and the second type as "fugitive"  emissions.   This  NSPS  would
limit VOC emissions from raw materials  preparation, polymer production,
                               3-11

-------
RAW MATERIALS
 STORAGE AND
 PREPARATION
POLYMERIZATION
   REACTION
 MATERIAL
 RECOVERY
REACTANTS AND
  REACTION
   MEDIA
  RECYCLE
   PRODUCT
  FINISHING
    PRODUCT
    STORAGE
                        ORGANICS  TO

                        BY-PRODUCTS
                       WATER TO
                        WASTE
                      TREATMENT
Step 1:  Raw Materials Storage and
  Preparation may include
(a)  Raw materials storage,
(b)  Raw materials purification,
(c)  Recovered raw materials recycle
       return,
(d)  Raw materials drying, and
(e)  Catalyst activation.
Step 2;  Polymerization Reaction may be
(a)  Batch or continuous-homogeneous
       polymerization,
(b)  High or low pressure polymerization,
       and
(c)  Liquid or gas phase polymerization.

Step 3:  Material Recovery may include
(a)  Product/raw materials separation,
(b)  Catalyst deactivation,
(c)  Product recovery and devolatilization,
(d)  Reactants and reaction media recycle, and
(e)  Organic by-product separation and
       recovery.
Step 4:  Product Finishing may include
(a)  Extruding and pelletizing,
(b)  Product cooling and drying,
(c)  Additives introduction,
(d)  Product shaping (.e.g., fiber spinning,
       molding, fabricating), and
(e)  Product curing, annealing or
       modification (e.g., fiber
       stretching and crimping).
Step 5:  Product Storage consists of
(a)  Product storage, and
(b)  Product shipping.
                 Figure 3-1.  General Polymerization Process
                                   3-12

-------
material recovery, polymer extrusion and pelletizing, and product cooling
drying and storage.  The process descriptions in this section are repre-
sentative of most of the polymerization processes used to manufacture
the products of the source categories chosen for NSPS development.
     The remainder of this section presents information on the polymers
and resins listed below, including their production processes and the
process emissions associated with each process.
      1.     Polypropylene - continuous, liquid phase slurry process
      2.     Polypropylene - gas phase process
      3.     LDPE - high-pressure process
      4.     LDPE - low-pressure process
      5.     HOPE - liquid phase slurry process
      6.     HOPE - liquid phase solution process
      7.     HOPE - gas phase process
      8.     Polystyrene - batch process
      9.     Polystyrene - continuous process
     10.     Expandable Polystyrene - post-impregnation suspension process
     11.     Expandable Polystyrene - in-situ suspension process
     12.     PET - Dimethyl terephthalate (DMT) process
     13.     PET - Terephthalic acid (TPA) process
Fugitive emissions, which originate from equipment components common to
all of the processes, are treated in the same manner for the entire
industry and are discussed separately in Section 3.3.
3.2.1  Polypropylene
     Polypropylene is a high molecular weight, crystalline polymer of
propylene.  Witlra density of 0.902 to 0.904 g/cm3, it is one of the
lightest commercial thermoplastics.  The general formula for polypropylene
is:
       ... - CH2 - CH - CH2 - CH - CH2 - CH - CH2 - ...
                   CH3             CH3        CH3

     Polypropylene can be stereospecific, which means that each repeating
methyl  group of the polymer chain can be attached to its neighboring
groups in two different geometrical  arrangements.  Depending on the
geometrical  arrangement of the methyl  groups,  the polymer exists in the
following three forms:   (1) isotactic - with all methyl  groups aligned

                               3-13

-------
on the same side of the chain as shown above, (2)  syndiotactic - with
methyl groups alternating, and (3) atactic - all  other forms in which
the methyl groups are randomly aligned on either side of the chain.
Both isotactic and syndiotactic forms, because of  their regular structure,
are highly crystalline, whereas the atactic form has little crystallinity.
Only the isotactic polypropylene is of commercial  interest.  Atactic
resin, an undesirable byproduct of all polypropylene processes, represents
about 7 percent of the total product.1!
     Polypropylene is produced by either a liquid  phase or a gas phase
process.  Two basic types of liquid phase process  are employed - slurry
and solution.  The slurry process is the predominant liquid phase process,12
and can be a batch or, continuous process.  Batch polymerization is
applicable particularly when low volume specialty resins are to be
produced.  The continuous, liquid phase slurry process and the gas phase
process are described in this section.
     3.2.1.1  Polypropylene Continuous, Liquid Phase Slurry Process.
Polypropylene resins are produced through coordination polymerization
using a heterogeneous Ziegler-Natta type catalyst system, which is
typically a combination of titanium chlorides and aluminum alkyls.12  In
conventional liquid phase processes, the catalyst provides a relatively
low polymer yield on the order of 500 to 1,000 units per unit of catalyst.13
In addition, a significant amount of catalyst or its residue remains in
the reaction product and must be removed.  More recently, some slurry
processes have used high yield catalysts with improved activity.  These
catalysts are known to provide a relatively high polymer yield on the
order of  5,000 to 7,000 units per unit of catalyst.13  In these processes,
the catalyst is present in such small quantities that it can remain in the
product,  eliminating the need for the additional process equipment
required  for catalyst  removal and recovery.  The elimination of this
process operation results in lower VOC emissions.
      3.2.1.1.1  Process description.  Both conventional and high yield
catalyst  continuous slurry processes are represented in Figure 3-2.
(The  identification numbers used for process equipment in this section
and the identification letters used for VOC emission streams in the
following section refer to this figure.)  Liquid phase slurry processes
                                3-14

-------
                                            a  •
                                            (/)  l/l
                                            (/)  (/I
                                            01  01
                                            u  u
                                              IB
                                           -a o
                                            10 -o

                                           U3 O)
                                            U> (U
                                            0)-=
                                            u *•>
                                            O
                                                                    3
                                                                    O
                                                                    (U
                                                                    c:
                                                                    oi
                                                                    Q_

                                                                    e
                                                                    Q.
                                                                    CD  (/)
                                                                   -E  


                                                                   O  (1)
                                                                    U J=
                                                                    o a.
                                                                   I—
                                                                   CQ T3
                                                                       •r-
                                                                    CO  3
                                                                    (/)  CT
                                                                    
-------
may use either liquid propylene or a different organic compound as a
diluent in which the polymer forms a slurry.  The only difference between
the conventional and high yield processes is that process steps 6 and
7 are unnecessary in the high yield process.
     Reactor feed material (1) consists mainly of liquid propylene (the
monomer), comonomer ethylene (if a copolymer product is desired), a process
diluent (which acts as a heat transfer agent), a polymer suspension
medium, and a heterogeneous Ziegler-Natta type catalyst.  Hexane is
used often as a process diluent, although some processes use mixtures of
other aliphatic hydrocarbons.  The catalyst  (2) sometimes is manufactured
on-site.  The catalyst solution is prepared  by mixing the catalyst with
the process diluent.  Propylene is charged to the polymerization reactor
while the catalyst solution and process diluent are metered in separately.
Hydrogen is also introduced into the reactor for molecular weight control.
Spent diluent from the catalyst preparation  operation is sent to the
diluent recovery section for reuse.
     Polymerization is carried out in either of two types of reactors  (3),
a  continuously  stirred, jacketed  vessel or a loop reactor.  Most conventional
slurry  processes employ jacketed, continuous, stirred-tank reactors.
The pipe or loop reactor is more  prevalent  in high yield catalyst plants.
Operating pressures of 2,070 to 2,760 kPa  (300 to 400 psig) are  common,
 but they can  be  as high as 4,140  kPa  (600  psig) when higher operating
temperatures  are used.14  Reaction is carried out at temperatures of
 about  60°C  (140°F) for approximately 8  hours.15  A portion of the reaction
 effluent, which  consists  of  polymer, monomer,'and diluent, is drawn
 continuously  from the reactor  to  a flash tank  (4) in which the unreacted
 propylene  and propane (an  impurity in the  monomer) are  vaporized, and
 subsequently  condensed by  compression and  cooling  (5).
      If the  catalyst  residues  must  be  removed from the  product polymer,
 the residual  slurry  from  the flash tank is fed  to the  deactivation/decanting
 section (6)  for washing with  a methanol-water solution.  Some  processes
 use isopropyl  alcohol instead  of  methanol  to deactivate the  catalyst.
 The washing  with alcohol  decomposes  the catalyst, dissolves  the  residues,
 and results  in two  phases -  a  lighter  diluent/crude  product  phase  and  a
 heavier methanol-water phase.   The  crude methanol from this  latter  phase
 is refined in a distillation column  (7) and leaves  the column  in the
                                3-16

-------
overhead for recycle to the process.  The column bottoms containing
catalyst metals are sent to the plant wastewater treatment facility.
     The crude product slurry, containing isotactic polymer and atactic
polymer-diluent (hexane) solution, is decanted from the methanol-water
phase and fed to a slurry vacuum/filter system (8) where isotactic
polymer solids are separated from the atactic polymer, which is dissolved
in the diluent by vacuum filtration.  (Alternatively, a centrifuge may
be used in place of the slurry vacuum/filter system.)  The atactic-diluent
solution is then introduced into a diluent purification unit (9)  containing
a stripping column in which the diluent is evaporated, condensed, and
purified, after which it is dried for recycle.  The atactic solids may
then be recovered or burned in an incinerator.  Liquid and gaseous waste
streams from the diluent separation and purification unit may be  burned
in the same incinerator with the atactic waste.
     The isotactic product from the slurry filter goes through a  product
dryer (10), extruder and pelletizer (11), and then is sent to storage (12).
The type of polymer dryer used varies with the facility, but the  fluidized
bed dryer with a hot nitrogen or air purge is the most common.
     Except for the high yield catalyst process,  the variations in the
various processes are minor and have little effect on VOC emissions.
The high yield slurry process, however,  does not require catalyst removal.
The absence of deactivation/decanting and alcohol recovery processes
eliminates several  major VOC emission sources.  The units that would not
be required by the high yield process are identified in Figure 3-2.
     3.2.1.1.2  Emissions from the polypropylene  continuous,  liquid  phase
slurry process.   The characteristics of vent streams from this process
are shown in Table 3-6.  Each vent stream is identified (circled  letter)
in Figure 3-2.  The total  process VOC emission rate for a conventional
slurry process is approximately 37 kg VOC/Mg product.   The high yield
process requires neither a decanter nor a neutralizer.   Therefore, the
high yield process would not have emissions from  these sources and.the
emission rate would be about 23.8 kg VOC/Mg product.  The emission
streams are continuous, or nearly so, and consist mainly of propylene,
ethylene, propane,  and a small  amount of the diluent used by  the  process,
usually hexane:   The temperatures of the vent streams vary from ambient
to 104°C (220°F),  and the pressure is about atmospheric.
                               3-17

-------
LU


UJ
D-


§

O.
o
0.


LU CO
0= CO
                I
            o



        €  Sl
        ^H  (*J
        O   CJ
                                      g"   a
                                      d.
              o<
                              s
       o
       ••4

       CO



       s
       o

       s
                                                                                                                  O




                                                                                                               CM  a.
                              •g
                      »  «  «•   §   S   S

                      i  I  i   J   J   I

                      I  1  1   I   I   I
                                           i—     O
                               o
                       B.   Ul   B
                       1   1   s

                       tt   r   t
                       5   I   1
                                                              Sg   "5
                               g   0

                               I   a
                                                                                  £
                                                                                  a.
                                                                                  0>

                                                                                  I
                                                                    g
                                                                    H

                                                                    i
                                                                                              g
                                                                                              o.
                                                                              



                                                                                      I
                StJ









c


4-»
"c
4->
C
at
£
a)
a
s
o
.-4
=
•g
1
*G
1
(A
£
to
£
4->
GO
1
(A
I'
IQ
e
o
.Q

a>

4**
_
£


1


U
e
o
•e
to
g
i
i
o

to
u
o
(U
1
o

CO
o
(O
c
o
.a
i-
«J
(J
3
O
*^
g
3=

c
3


O
U
e
o
•e
to
T3
>»
|

O
s-

o
0)
»
•°
e
0)
u
ai

o

"10
u
i
a>
"ai
_>>


,
I
"x
o
1
3
£
Ol
e
1

1
u-
u-
•5

«
1

(A
V)


t-
01
>»
L.
Tg
N
3
^

at
5

u
**-
w
s
«
ai
^


















(A
e
o
"S
0)
u
i en
*g
1


i
0)
^
ai
•o




(A
|
(A
V)
i
o
Ol
e
to
t.
O)
JS

5
•o
c
ai
&
3
S1
•5
3
.e
                                                            3-18

-------
      The vent from the catalyst preparation section,  Stream A,  releases
 diluent continuously.   Reactor vent Stream B is  .continuous  and  releases
 monomer and diluent.   Reactor vent Stream C is an  intermittent  stream
 that occurs during emergency dumps of  the reactor.  This  stream is
 typically controlled  by a flare.
      The combined total  of the decanter  and neutralizer vents,  Streams D
 and E,  is usually the  largest VOC  emission  source  in  the  process.  The
 constituents are methanol  or isopropyl alcohol (if used), 63 hydrocarbons,
 and diluent.   Newer, high  yield catalyst processes  do not require these
 process steps.   The absence of these vents  significantly reduces the
 overall  VOC emission rates for the high  yield process.
      The slurry  vacuum/filter system vent,  Stream F,  is one of  the
 largest VOC emission streams,  venting  process diluent and alcohol that
 has remained  in  the polymer.   It is common  to both  the conventional and
 high  yield  slurry  processes  and releases  at  atmospheric pressure.
      The vacuum  jet exhaust,  Stream G, from  the  by-product and  diluent
 recovery section,  can be the  second largest  VOC  emission stream in the
 entire  process.  The diluent  recovery  section, which consists of an
 evaporator, an extractor,  and  distillation units, is common to both the
 conventional  and high yield processes.   It emits process diluents and
 traces  of alcohol.
      The vents from the product drying section,  Stream H,  emit diluent,
 methanol, and propane diluted  in large quantities of air at a relatively
 high  temperature, 104°C (220°F), and near atmospheric pressure.
      Emissions are also released from the extrusion/pelletizing  section
 vent, Stream  I.  Significant quantities of hydrocarbon still remain in
 the polypropylene powder as it exits the dryer and enters  the extruder
 feed  chute.  At this point, the powder is in equilibrium with a  vapor
 that can contain up to 25 percent hydrocarbon by  weight.   As a result of
 heating and compression in the extruder,  there is some VOC loss  through
 the extruder/pelletizer section and further losses from the powder/pellet
 transfer system downstream from the product dryer since the transfer medium
acts as a stripping gas.
     The stream properties and VOC  concentrations of these process  vents
can vary depending on  process conditions.  The variation generally
depends on the grade or type of product being manufactured,  process
                               3-19

-------
variables such as temperature, pressure, catalyst concentration,  or
catalyst activity, and the amount of hydrogen used for molecular  weight
control.
     3.2.1.2  Polypropylene Gas Phase Process.  The gas phase process
for producing polypropylene is relatively new and currently is being
used in only one plant.  Processing steps are less complicated than in
the traditional continuous liquid phase slurry process.  The gas  phase
process is similar to the high yield liquid phase slurry process in that
there is no need for catalyst removal.  The product processing techniques,
however, are different due to the gas phase reaction.  The uncontrolled
continuous VOC emission rate from the gas phase and the high yield
liquid phase slurry processes are comparable.  These two newer processes
are expected to predominate in future polypropylene capacity because of
their  simplicity, lower capital  cost, and high yield.
     3.2.1.2.1  Process description.  Figure  3-3  shows a simplified  flow
diagram of the gas phase  process.   (The identification numbers used  for
process equipment in this section and the identification letters used
for VOC emission  streams  in the  following section refer to this  figure.)
In the gas phase  process, catalysts and hexane are premixed  in the
catalyst mix  drum  (1)  from which they are fed to  the  gas phase reactor (2).
Propylene monomer (3)  is  introduced into the  gas  phase reactor by a
separate line.
      The product, which  is  in a  powder  form  containing contaminants  of
 propylene and finely divided  catalyst,  is transferred  from the gas phase
 reactor to  a fluidized bed  catalyst deactivator  (4).   The gaseous components
containing  unreacted propylene  from the reactor  are recovered and purified
 in a material  recovery operation (5)  and then recycled back  to the
 reactor.  The recycle system  contains bag filters, an  entrained  gas
 scrubber,  a  recycle scrubber, and a recycle  gas  compressor.   The gas
 stream from the  fluidized bed catalyst  deactivator passes through  a  bag
 filter system (6)  to recover product.   A nitrogen purge  gas  stream  from
 the bag filters  is then fed to a scrubber (7).   The purge  gas  from the
 HC1  scrubber is sent to a flare for combustion.
      Polymer product from the fluidized bed  deactivator  and  the  bag
 filter system (6) is sent through the product finishing  steps of extrusion,
 pellet blending, and storage (8).
                                3-20

-------
   CATALYST

CO-CATALYST
    HEXANE
         NaOH -
      SOLUTION
         NaCL „
      SOLUTION
                    SCRUBBER
                      (7)
  FLUIDIZED
BED CATALYST
DEACTIVATOR
    (4)
                                                  BAGHOUSES
                                                     (6)
EXTRUSION/
 PELLET
BLENDING &
  STORAGE
   (8)
    Figure  3-3.   Simplified  Process  Block  Diagram for the  Polypropylene
                                    Gas  Phase  Process
                                          3-21

-------
     3.2.1.2.2  Emissions from the polypropylene gas phase process.   Two
VOC offgas streams from the process, the scrubber vent,  which is a con-
tinuous stream (A), and the reactor blowdown vent, which is an intermittent
stream (B), are currently controlled by a flare system according to
one industry source.  The uncontrolled VOC emission rate for the entire
process, based on the average expected flow rate from these two streams,
is 36.5 kg VOC/Mg product.  The VOC emissions consist primarily of
propylene, propane, and hexane.  Table 3-7 summarizes the emission
characteristics of this process.
3.2.2  Low Density Polyethylene (LDPE)
     Ethylene polymerizes in the presence of a suitable catalyst in the
following manner.
         H\                        H   H               H   H
I    I
C = C
  I    I                I    I
-C — C — (C2H4)    — C — C
               Catalyst
                                     H   H               H.   H
     ^Conventional  LDPE resins  have  a high degree of branching with
 densities that range  from 0.910  to  0.935 g/cm3 and are produced in
 either an autoclave or tubular reactor.
      LDPE resins may  also be produced commercially at low pressures in
 both gas phase fluidized bed reactors and liquid phase solution process
 reactors.  The low pressure process yields  a much more linear  structure.
 Linear LDPE can be made in the gas  phase with melt indices  and densities
 over the full  commercial range.   Both processes are described  below.
      3.2.2.1  LDPE High-Pressure Process.   The high-pressure process  is
 currently used more widely than  the low-pressure process.   However, due
 to the more favorable economics  of the  low-pressure process, few,  if
 any, high-pressure plants are  likely to be  built in the  future. 16
 The high-pressure process is a free radical process utilizing  pressures
 of several thousand atmospheres  to polymerize ethyl ene  and  copolymers of
 ethylene, and nonolefinic comonomers such  as vinyl  acetate. 1?   High-pressure
 processes, however, are not capable of polymerizing propylene  or  higher
 olefins  to their corresponding polyolefin.18  Free radical  catalysts, or
 initiators, are usually used and are predominantly oxygen and  peroxides.
 The product grade can be changed by varying the throughput rate of the
                                   3-22

-------
 LU
 Q-


 §
 Q.
 o
 a.
 LU
o«a
Q£ CO
U- CO
    LU
CO O
S O
"=C O£
LU Q_
C£.
(_ LU
CO CO
     co

LU CD
O

CO
o
co
o:
LU
 t
CO
 CD
.0
 ITJ
                    3 O!


                    V) V)
                   0)

                   I
              o  •>— o
              •i- (U CJ 3
              « 4-> O -O
                c
                   en  o
                   » at C    O  >> (O c
                             a. a. , s- 5. •"
                             0-0-2:   	
                                       s
                             U1

                             o
                                       tu
                                       4J
                                       c
                                       

                                                             1
                                                              o
                                                              u
                                                                  o
                                                                  a.

                                                                  u
                                                                  (U  S-
                                                             04-13
                                                                  IO  O1

                                                             c:  =
                                                             13      
-------
polymer in the reactor, the reactor configuration,19 reaction temperature,
or type and concentration of the catalyst.20
     3.2.2.1.1  Process description.  Figure 3-4 is a schematic
representation of the high-pressure process.  In this process, ethylene
is dried (1) and fed to the suction of a first stage compression system
(2) which raises its pressure to about 28 MPa (4,000 psig).  The exact
pressure depends on the pressure employed in the product separation
phase of the process.  The primary compressor discharge is combined
with recycle ethylene and introduced to the suction of a second stage
compression system (3) which raises the pressure of the ethylene stream
to 275 to 345 MPa (40,000 to 50,000 psig).  A comonomer is also added
if a copolymer product is desired.  The compressor discharge effluents
are fed to the reactor (4).  An initiator solution of organic peroxide
in isopropyl alcohol is injected directly into the reactor to initiate
polymerization.  The amount of peroxide catalysts required varies from
about 10 to 100 ppm.21
     Either tubular or autoclave reactors are used.  Temperatures in the
reactors may vary from about 150 to 300°C (300 to 570°F), although
ranges from 200 to 250°C (390 to 480°F) are more common.  Pressures
within the reactors vary from about 100 to 345 MPa (15,000 to 50,000 psig).
For many types of polyethylene, the common pressure range is 130 to
240 MPa (20,000 to 35,000 psig).21  The total  residence time of the
reactants varies from 45 to 60 seconds in a tube reactor and from 25 to
40 seconds in an autoclave reactor.22  polyethylene product from the
reactor is continuously throttled into the high-pressure separator (5)
which operates at a pressure of 6 to 28 MPa (900 to 4,000 psig).  Most
of the unreacted ethylene is flashed and withdrawn overhead from this
separator.  It is cooled, separated from low molecular weight polymers
(wax) in the high-pressure wax knock-out drum (6), and recycled to the
second stage compressor (3) suction.  The separator bottoms pass through
a throttle valve to the low-pressure separator (7) which operates at
35 to 70 kPa (5 to 10 psig).  The remaining ethylene is flashed from the
product and withdrawn overhead from the separator.  The withdrawn gas is
also cooled to condense waxes and routed through the low-pressure wax
knock-out drum (8) to the olefins recovery unit (9).   From this unit the
recovered ethylene is returned to the monomer  storage area.  The degassed
                               3-24

-------






£2
E =
is
33





















to!
\
;





































-



1
« a,
£

UJ
tno
ggls
al°






















}
1
?•
SS
| 	 ^a-
t/1 S
^, £ i *-* ^
pi-

(3)^ 	







r-
id)
© 	 i
h






. uioe
i=°
sssc
U4^
O to
II SS
"~~il "^
-K 10 ^:

-------
homopolymer or copolymer product is extracted as the low-pressure separator
bottoms which still contain residual  ethylene monomer.   These product
resins are routed to the finishing line where antioxidants are added to
resin which is then melted and pelletized in an extruder-pelletizer (10).
The hot pellets are water cooled and conveyed to a hot air dryer (11).
The residual ethylene within the polymer is very low after extrusion and
drying.  Dried pellets are conveyed to storage (12).
     3.2.2.1.2  Emissions from the LDPE high-pressure process.  The
offgas stream characteristics for the high-pressure process are shown
in Table 3-8.  The total uncontrolled VOC emission rate for the entire
process varies significantly, ranging from about 1 kg VOC/Mg product
to about 29 kg VOC/Mg product or higher.  In general, the composition
of the streams varies based on the type and amount of product made,
separation pressures involved, and ambient temperature.  The emission
streams consist mainly of ethylene.  Other compounds, such as ethane,
methane, propane,  propylene, and isopropanol, are also present.  The
temperature of the streams varies from 50 to 260°C  (106 to 500°F),
and the pressure varies from atmospheric to 140 kPa (20 psig).
     The start-up, shutdown, and maintenance vents, Stream A, are
intermittent and consist of VOC gases from a number of sources, such as
compressors, within the plant.  Included in this stream are  the vents
from the wax blowdown system, which can be a source of significant
ethylene losses.   The emission rate is highly dependent upon the design
of the wax blowdown and discharge system.
     The emergency reactor vent, Stream B, is intermittent.   It is
activated either manually  or automatically as an emergency blowdown
during a process upset.  The total mass and instantaneous flow rate can
vary dramatically  depending on the nature of the upset.  This VOC
emission source generally  emits directly to the atmosphere because  it
is an  extremely large quantity emitted at very  high pressure at infrequent
intervals.   It consists of monomer and polymers, with trace  amounts of
catalyst,  in  a three-phase flow.   In  order  for  this stream to be safely
destroyed,  such as in a flare  system,  the solids would have  to be
separated  from the bulk flow  before it reached  the  combustion zone.
      The dryer and storage bin vents,  Streams  C and D, are usually
combined.   They  are  continuous  streams consisting of mostly  air with
                                   3-26

-------











1—

CO
1 1 1
o
-
o
— '
LU
IDO
I— CO
CO

OO
o; o
LL.Q:
Q.
CO
"Z. LU
-=i|^ f*S
LU ID
o: co
H-CO
CO LU
|_ Q.
LU 3C
> CD
1— 4
O
LU
CO Z
C_> LU
t— ( — I
1— >-
oo ic
"— i I—
o: LU
LU >-
i— —i
0 O
«£ Q_
Sc
o
CO
1
00


ai &
a.
f»
ai
s.
k o
Ol o
CL
E
O)
r-


en
e E 4J
0 «~-. 0
•r* Ol O 3
VI 4J O "O
m re ^* o

if! 5"^


2
3
a










i
rt5
=5





Q
i
ai
*^
CO


vi e
$.2
U 4J
0 U
5- 01
a. co



•e
|
o
o





S

n •
re
*



ffi
°u
re



0
C*)
CM

O

+J
1
i
0)
c
^*



-0
re
ui
. 01
X k
O 3
3 0)
J= U
cL ai
7 c
|l
'CO





<




s.
Ol
o
Ol

ai ai
J| £5
£g CMOO
en o en
en ~* en
i i
o o
CO CM


o
CM
ci °



•s,
VI <»)
ai *
re «•



** CM
[g 0 •
t *•* *M
^^ r*. r^.

O o c!

•w
01 VI
4-> 3
*> 0
I I
ai +j
C 0
i-l CJ


Ol
2
5
CA
0)
^ ^->
C IB
ai re £
«* i. c
C 01 O
2 ?> •>-
> S- VI VI
•O *> VI .
i" "si 1
oi e
en i- c i—
tS 5 |2




o
CQ O




CO
a.
a: uT
a. Q.


















w
o
o
O)
VI

in
O)
|
Q.
^^

re
g
u-
v>
01
n
j_
Ol
5
o

«?
V) VI
0) T-
O c
Q} *W
C 4J
a =
in -o
01 0
s r
M .«

•o o
c •»•
I-H 4->
U
re
.. 0)
C i.
o
•^ c
s 5
£ re
O N
•F- Ol
E
•s 3?
g §•
^ "
5 £
re ja









































•
c
o
re
Ol

1
01
s_
VI
t
eg

T
Ol
i.
3
iZ
Ol
01
CO
u


















t
c
o
VI

13.
§
U
g

ai
^
ai
o.
en
CM
4J,
V)
re
01

re
>f
"re
u
•r"
a.
^





























in
o
+>
a
s-
a
S°
v>

at
t.
3
VI
in
ai
i_
JC
Ol
JE CM
^ S
re -a
c
ej *°
Ol o
i S

in «
Ol Ol
VI 1=
re ai
Ol i—
•— >>
01 S
S- 0
>> ct
u
ai sT
en ai
Ol O
a.
in
Ol VI
•o re
3
r— S:
U U
C 3
ai n-



































,
O
o
°o 5
10 r^
o C Ol
1 ^
i -°
+J 01
en
s- re
ai s-
b 5














i
Ol
f
*"
"o
in
f
4J

ai
in
ai
Q.
2
w
^J
S
€
13
C
IA
"c
re
r^

re

5
u-
in 01
•2 o
s- re
c -a
o ai
in o
11"
(O O)
"S t*
^2
a> a>
tf5
§.=
Ol «^-
VI
p— ai

3 a
re *J
VI
ai ai
re .s
o.^1
It
3-27

-------
small amounts of ethylene.  These streams are usually emitted to the
atmosphere.
     A considerable number of LDPE plants are located near or integrated
with a plant which manufactures ethylene.  As a result, the unreacted
ethylene from the polymer plant is purified by recycling through the
ethylene manufacturing unit.  Overall process VOC emissions from these
units can be expected to be 8 to 10 percent lower than those where the
LDPE plant and the ethylene plant are not located at the same site.
     3.2.2.2  LDPE Low-Pressure Process.23  Most, and probably all,
new low density polyethylene plants will use the relatively new low-
pressure process because of its much more favorable economics.  The most
significant cost aspect of the new process is the drastic reduction in
reaction pressure.  The low-pressure process uses reaction pressures
of only 0.69 MPa to 2.1 MPa (100 to 300 psig) in comparison to pressures
as high as 345 MPa (50,000 psig) in the high-pressure process.  Changes
in product grade are accomplished primarily by changing the catalyst
and/or reactor gas composition; reactor operating conditions remain
the same.24
     The new process has reduced capital investment requirements by
50 percent, energy consumption by 75 percent, and the operating cost
of making low density polyethylene by 50 percent.  One significant
technical aspect of the low-pressure process is its capability of
producing high or low density polyethylenes in the same process equipment.
The advantages of the new process are so overwhelming that it is likely
to be the preeminent process for future expansion of the polyethylene
i ndustry.
     3.2.2.2.1  Process description.  The process flow diagram presented
in Figure 3-5 is based on Union Carbide's Unipol process.  Ethylene is
polymerized in the presence of a chain transfer agent and an alpha-olefin
comonomer to produce polymers having desired melt indices, densities,
and molecular weight distributions.  The alpha-olefin comonomer is
usually 1-butene, or the more costly and higher-boiling 1-hexene or
1-pentene.  Before entering the reactor, the monomers (depending on
their sources) are subjected to varying degrees of pretreatment (1)  to
remove impurities that could poison the catalyst.  Monomer is then fed
continuously into the fluidized bed reactor (2).  Catalyst is added
separately (3).
                                  3-28

-------
                                                         O)


                                                         OJ
                                                        r—

                                                         >>



                                                         O) CO


                                                        I— CO
                                                         O CO
                                                        Q.   CO
                                                        O  03
                                                         O
                                                        _1  CO
                                                            03
                                                         OJ CD
                                                        .c
                                                        4->  O)
                                                            OJ
                                                         (O  O
                                                        •r- Q.
                                                        a
                                                         O-i-
                                                         o  co
                                                        r—  C
                                                        ca  a)
                                                           a
                                                         to
                                                         CO -C
                                                         

                                                        -a -P
                                                         0)
                                                         o. a)
                                                         S  s-
                                                        •i-  3
                                                        CO  CO
                                                            CO
                                                            a>
                                                          •  s-
                                                        LO a.
                                                        U.
3-29

-------
     The process uses a fluidized bed reactor technology and a new
family of catalysts that trigger the desired chemical reaction at pressures
between 690 and 2,070 kPa  (100 and 300 psig) and temperatures of about
100°C  (212°F).  The fluid  bed in the reactor is granular polyethylene,
the product of the polymerization reaction.  Circulated up through the
bed, the gas stream containing the unreacted ethylene and comonomer
passes out of the reactor  through an enlarged top section designed to
reduce velocity, thereby disengaging most of the fine particles.  It
then goes to a cycle compressor (4) and through an external cooler (5)
before returning to the reactor.  Dry, free-flowing solid product is
removed intermittently from the continuously growing fluid bed through
a discharge system (6) to  keep the volume of the bed approximately
constant.
     Although most of the  unreacted monomer is recycled, some residual
VOC must be purged (7) from the granular product -before it can be safely
conveyed in air.  As an optional final step, one or more conventional
additives (e.g., antiblocking, antislipping, antioxidizing, ultraviolet-
light-stabilizing) may be  added to the granular product before it is
stored or shipped (8).
     The overall combined  conversion rate of ethylene and comonomer is
97 to 99 percent.  The average residence time of the polymer in the
reactor bed is 3 to 5 hours, during which the particles grow to an
average size of about 1,000 microns.  Polymer density is regulated by
the type and concentration of alpha-olefin comonomer, which controls the
frequency of short chain branches.  Molecular weight is influenced by
the reaction temperature and the concentration of chain-transfer agent
in the circulating gas.   Molecular weight distribution is manipulated
primarily by catalyst type and composition but also, to a much lesser
extent, by reactor operating conditions.^
     3.2.2.2.2  Emissions from the LDPE low-pressure process.   The
waste gas stream characteristics for the LDPE low-pressure process are
shown in Table 3-9.   The combined process VOC emission rate for this
process is approximately 23.4 kg VOC/Mg product.   One process  vent,  the
product discharge vent,  contributes slightly more than 95 percent
(22.3 kg VOC/Mg product) of the total  VOC discharged annually  from the
model  plant for this  process.
                                  3-30

-------

LU

LU CO
	 1 CO
>- LU
:n o
h- O
LU C£
>- D_
O LU
0- CO
>- re
H- 0-
1—4
co co
2? ^
LU CD
Q
LU
IS ^*"
O LU
_J 1

LU rn
a: i —
H- LU
>-
21 —i
o o
C£ 0-
1 1
>-
co i —
< CO
LU -ZZ
fV 1 t 1
\— Q
co
H- CD
LU IE

LU
u- :n
O 1—
t/> Q
o -z.
CO CO
HARACTERI
RE PROCES
O =>
CO
CO

• LU
O1 C£
i a.
CO 1
3

w


fc



tJ
a.
§
*~


O)
c s; 4J
o «-^. o
•*- OJ O 3
(A 4J O "D
£ £ u
E a> a.
UJ .ic



1
to
z





13
J






(J
re
£

to

CA CT
V) O
CJ 40
0 0
££
0 t-

o CM > «r
>ZO CJ O t- U O t_ O O tj CJ O
min >z* Sz1 >5 > 3E1 >< >^ Sz1 oexj ^^ OCM o
o at to tn c\j co ^j co ^H a> *o ^ co r^- PI i^«. in in o ch o o o
1 1 <-4






'O «— I O O CMOi-4
i-^io^^oioea -4d
°V
^




10 'c^Smco IUDO cow
CO




**•
§^^ o o o*) vo in P-H in *— * •«< r*,
^o^ocsoor^no ooco
OOOOOOOOCUO OOfO
CM CM


4J
cr cr cr c cr cr cr
a)ajQ>u)4-> O 4J 4J 4^ 4J 4J O O O 3
EB5EE111II I .|
ocj4jaja)a)ttJaj4-'4-> 4j 4J
^tS°t£,^c:ccoo oo

c cr
o a>
re 4^ c 4_>
u cr o c — *
^ 4^ U C > Z 0)
•^ re o CJ *j o 4J
3 •*- (Q 4-> O OJ CT "^^ (5
Q. It— ^ 4J e f— 4J 4J >• t. (•— Ij—
7- -o c w Q) u c re rewc
w b i'S1* *" rea>-ojj= euwo
^~ 3 ,c > tn  o» O wi e *»—
re Q- O> a. OJ t. L. en  C3f- £»>(«
*• t. C 3 >^ 3: CL "O O «—
» 00 4J W U ^
re 4J § >^o E aJ S?o o c wo) 01 o -«-> 3 > t»cr>tt— -
^ E 4Jj=o u re43
re orereojE
(J
1
u
°*
"
CO
Q»

D>
**"

"c
3
•CJ
O
c_
CL
u_
a_
e:
^
re
s.
c
o
*J
" *O
correspondeno
PR = polynierl
1 I
3 40
I £
«-*  i—
S •»-
C_ L.
O (U
cr re
>0 re
Q)
t! "
o z
to ce







































o
4->
•a
*^"
re
2
40
in
t_
*£

m
i
fo
£
3
•i—













.
aJ
re
C
>.

O)
IA
W-
°


J
S

1
U
5
u
!„
3
U

t.
(U
M
>»
re
O)
re

^3"
„
— '

o
40
S"
o
X
cu

re
£
4>*

i







































(U
c
o
t/1
'E
OJ
OJ
>

4-)
"5)
^

re
«
4->
C
V
>
(A
ji











—
£
>>
V)
1

u
01

o
u
en
•~
t
to

01
a.
40
0)
c
o

£
"O
(O
i s
u. t_
«J 3
O
i~- (/>
II
C. (A
4-> U
cr o
>»5
C QJ

• t_
1 •>

>
t. 4J
O)*C7>

U. 4-
c re
f!S
o -o
"= GJ

-------
     Process emissions consist of VOC with two to six carbon atoms along
with nitrogen or air.  The temperature of the streams varies from 38 to
827°C (100 to 1,520°F), and the pressure varies from 0.7 to 138 kPa (0.1 to
20 psig).
3.2.3  High Density Polyethylene (HOPE)
     HOPE resins are linear thermoplastic polymers of ethylene with
densities higher than 0.96 g/cm3 and copolymers of ethylene with densities
as low as 0.94 g/cm3.  HOPE resins are typically produced at low pressures
by either liquid phase or gas phase processes.  There are two liquid
phase processes, slurry and solution.  In both, the solvent dissolves
the ethylene monomer and comonomer and suspends the solid catalyst.  The
basic difference between them is that in the solution process the solvent
also dissolves the polyethylene product.
     The gas phase process is virtually identical  to the LDPE gas phase
process.  All three HDPE processes are described in this section.
     3.2.3.1  HDPE Liquid Phase Slurry Process.  Of the two liquid
phase processes, the slurry process is predominant and is the only one
capable of producing the whole range of HDPE polymers.25  The slurry
or particle form process of Phillips Petroleum Company serves as the
basis for this description, but it is intended to illustrate all liquid
phase slurry processes.
     3.2.3.1.1  Process description.  As illustrated by the schematic
for this process, Figure 3-6, the feed section (1) consists of the
reactor feed storage and a catalyst purification and activation system.
The chromium oxide catalyst is suspended in a solvent (pentane or isobutane)
and is continuously fed to the reactor (2).  Other slurry processes use
Ziegler catalysts or molybdenum oxide catalysts.  Purified ethylene
monomer and a comonomer (1-butene or hexane) are fed to the reactor
where suspension polymerization takes place.  The  reactor is usually a
closed-loop pipe reactor.
     Temperatures of 20 to 100°C (68 to 212°F) in  the reactor have
been reported, but the polymerization rates at 20°C (68°F) are probably
very low.  Pressures are from about 690 to 3,450 kPa (100 to 500 psig).
Higher pressures are normally required at the higher temperatures  to
dissolve sufficient ethylene in the liquid phase.   The slurry in the
reactor contains 18 to 25 percent solid polyethylene.  Settling legs and
                                  3-32

-------
                                                  -l->
                                                  O)
                                                  O
                                                  Q.
                                                  1/1

                                                  0)
                                                  O
                                                  en
                                                  O)
                                                  s- >
                                                  en C
                                                  (O S-

                                                  5 i—
                                                  O O>
                                                  O (/)
                                                  r- rtS
                                                  CO f
                                                     a.
                                                  w
                                                  10 T3
                                                  OJ -r-
                                                  O 3
                                                  O CT
                                                  S- -r-
                                                  O __ I
                                                  Q.

                                                 •I—
                                                 oo
                                                  cu

                                                  3
                                                  O)
3-33

-------
screw conveyors are used to withdraw concentrated slurries containing 50
to 80 percent solids.14  Unreacted monomer and diluent (3) are separated
from the product by flashing.  Final stripping of the gases from the
polymer is performed using steam.  The wet polymer solids are then
centrifuged (4) to remove water, and dried in a closed-loop nitrogen or
fluidized air drying system (5).  The-resulting polymer fluff is mixed
with various finishing agents (6), and packaged (7).
     Vapors from the flashing vessels are sent through a diluent recovery
unit (8) which condenses the diluent and recycles it through diluent
treaters (9) back to the reactor.  The ethylene-rich stream is then sent
to the ethylene recovery unit (10) for purification and sent to recycle
ethylene treaters (11) and back to the reactor.
     3.2.3.1.2  Emissions from the HOPE liquid phase slurry process.
This process has one intermittent and three continuous process emission
sources.  The major emission source for this process is the recycle
treater vent with an emission rate of about approximately 12.7 kg
VOC/Mg product.  The total emissions from the four streams varies from
13.0 to 13.3 kg VOC/Mg product.  Table 3-10 shows the composition of
these streams.
     The emissions from the drying system (Stream B) are a dilute
mixture of process solvent in nitrogen when a closed-loop nitrogen dryer
is used or a mixture of process solvent and air when an air fluidized
bed dryer is used.  The continuous mixer in which antioxidants are
added to the polymer (Stream C) vents a low VOC emission stream.  Some
of the process solvent that is still in the polymer is emitted along
with a large quantity of nitrogen, usually to the atmosphere.  The
recycle treater system has continuous emissions (Stream D) which are
about 80 weight percent VOC.  Treaters are vessels containing materials
such as adsorbers, dessicants, or molecular sieves which remove water
and other impurities in the recycle ethylene stream.  Emissions occur
when the vessels are purged prior to regeneration.
     Most plants use a separate recycle treater for each individual VOC
component since components are usually recovered by fractional distillation,
Therefore, an HOPE plant recycling ethylene, isobutane, and butene would
generally have three treaters.
                                  3-34

-------




UJ
LU
-J
3J
UJ
O
^
1—
oo
UJ
a
3=
CD

A
LUrB
3T 00
J- 00
LU
s: o
o o
r»* rv*
U_ Q_

00 >-
*3 5?
LU =>
Q£ i
(— OO
1/5
LU
1— 00
^*f^ "^7^
UJ 3=
> 0.
u. a
o >-<
oo cy
O HH
i— i _j
oo
i— t
rv
LU
j^
 4->



£
m
I"
i
*~


0)
o • --. o
•^ fll O 3
in m > o
LU ™ °"


01
L.
3
Z










3>
E
ID
Z








1=
m
£

in



u
u> C
W O
cu .,-
U 4->
O U
C- 01
O_ CO



0>

LU
O
§



CM





CM
O

01
C
Ol


1
01
4->
C








c
o

t?
ID
t
a.
•o
Ol
Ol
u.














a.
s:
o:
01
c c
ID Ol
4J C7)
3 O
0 4->
—i Z
co r^
°s



*— 1
CM


*£..
O
1

O


l/l

o
s
c
*JT
c
o







1
s
ii
£
"c

01

0




m









Lu

at at
c c co c c
10 ai cm oj
4-> o> 0) -w 01 en
30 —SCO
JO t~ >>-O IQ i>
"-" Z LU ~ LU Z
U» «f O O O O
O O^ •— * CD O «•-«
en vo -ncj



l-l ^H
CM CM



CO
O CO

0 CM i
CO


in vi
3 3
O O
3 3
C C
•M 4->
C C
O O
U U






01
•u
10
i- u> en
01 '
x ai c
•g *J O
S In
ML. VI
3 4-> -r-
0 E
3 . 01 UJ
C »—
• •*- O r—
4-> >> ID
CO 4J
O CO O
O Of H-




O O

,







1 1 g^
a- £


































ai
u
1
c
o
n.
VI
£
0
u

£
VI
3
-C
t— t



C
o

0
c

o

ai
u
t-
i
oo
ID







































,1
g
C
c
3

(«
IO
•o
£
tn
0)
L.
Q.
E
(1)
L.

U)

Ul
to

$
o
u
£

•
en
c
^-
M
«n
«
i
c.
Q.
II
U.
a.
»n
C
o
•tJ
m
L.
m
Q
S
u
Q.
in
*
i_
01


'
i.

11

i
u
























•











c
o
1

J;
c
ai
•o
E
ID
01

VI
£

•
CO
ai
3
CD

Lu

01
ai
to
•o










































.

^J
c
i
&.
01
ex

co
u
c
£
3
u
u
o

ai
c
o
ai
















in

>>
•o
at
n

•a
a

^
3
c
J_
**~
to
J_
o
*^
V)

I
01
.c
CJ)
.c
I/I
c.
1?

•o
c
Ol
en
o
c.
•M
C

01
*—
u
£

o

VI
'^"
•o
c-
CO

2
o
<*-
3-35

-------
     The intermittent offgas stream is the feed preparation stream
(Stream A) which consists mostly of ethylene.  Sources for this  stream
include drying/dehydrating and other feed purification operations.  Its
emission rate is 0.2 kg VOC/Mg product.  Emissions occur when the treating
vessels are purged prior to regeneration, usually once a month.
     The HOPE liquid phase slurry process described above has an ethylene
recycle and closed-loop nitrogen drying system.  These greatly reduce
the emissions.  Some plants, however, vent unreacted monomer and use
simple single pass dryers.  These plants have substantially higher
emissions.  The major VOC source for these plants is the flash tank
where the unreacted monomer stream is about 50 percent VOC.  This stream
is often burned in a boiler because of its high heat value.
     A considerable number of HOPE plants are located near or integrated
with olefins distillation trains.  Some of these units do not need
recycle treaters since the ethylene is purified by recycling it  through
the olefins manufacturing unit.  In these cases, overall process VOC
emission from these units can be expected to be only 2 percent of the
emissions from separate HOPE plants.
     3.2.3.2  HOPE Liquid Phase Solution Process.  The solution  process
differs from the slurry process primarily in that the polymerization
process is carried out at a temperature higher than the solution temperature
of the polymer in the selected solvent medium so that the polymer, is
in solution (completely dissolved) rather than in particle form  suspended
in the reaction mixture.  The solvent medium may be cyclohexane, pentane,
hexane, or heptane.  Operating temperatures for the solution process
may range from 100 to 200°C (212 to 392°F), depending upon the solution
temperature of the pol-ymer in the sol vent.26  Emphasis in HOPE manufacturing,
however, is shifting from the solution process to the 'slurry and gas
phase processes.
     3.2.3.2.1  Process description.27,28  As illustrated in Figure 3-7,
ethylene monomer goes through preparation step (1) before being  fed via
a compressor (2) to a stirred reactor (3).  The preparation step assures
removal of acid gases and moisture.  Catalyst is introduced as a slurry
into the reactor from the catalyst preparation section (8).  The reactor
may be operated in a continuous or batch mode.  Operating conditions may
be around 260°C (500°F) and 7 MPa (1,000 psig).  The exact operating
conditions depend in part on the particular polymer being produced.
                                  3-36

-------
                                                     1
                                                     CO
                                                     £
                                                     (O
                                                     S-
                                                     O)
                                                     I
                                                               
                                                              ^: wi
                                                              •»-> 
                                                     oo
                                                     O
                                                               (O
    CO
-i^
 O  O)
 O  
                                                               en
3-37

-------
     From the reactor, the reaction mixture proceeds to the separation
section (4) where the catalyst, solvent mixture,  and polymer are separated.
If desired, the catalyst may be recovered and recycled.  The solvent
mixture is sent to solvent recovery (9).  Vapors  of unreacted monomer
and solvent pass through a condenser (10).  The unreacted monomer is
recycled to the monomer preparation area through  recycle ethylene
treaters (11), while the condensed solvent is returned to the flash
drum.  Alternatively, a reflux condenser (lOa) may be placed on the
reactor to recycle unreacted monomer directly back into the reactor.
If an ethylene plant is located on-site, the ethylene vapors from the
reactor may be sent directly to the ethylene plant, thereby eliminating
the recycle ethylene treaters,in the HOPE plant.
     As it leaves the separator, the polymer is extruded and melt cut (5).
The melt cut polymer still contains solvent that  is subsequently stripped
from the polymer in a steam still "(6), which may  be operated under
pressure or vacuum.  For copolymer processing, the stripping is carried
out under vacuum (1.0 to 700 mm Hg) in order to lower the vaporization
point of the solvent below 71 to 77°C (160 to 170°F) and prevent
agglomeration and melting of the copolymer charge.  The vaporized mixture
of steam and reaction medium from the steam still is sent to a distillation
train (9).  The recovered solvent is recycled to  the catalyst preparation
section and to the reactor.
     The steam stripped extrudate from the extruder is then dewatered,
re-extruded in a semi-dry state, and dried to reduce the water content
to about 1 to 2 weight percent or less.  The final product may be blended
with additives at this time.  The extrudate is then cooled, chopped,  and
transferred to packaging facilities (7).
     3.2.3.2.2 Emissions from the HOPE liquid phase solution process.
This process has numerous offgas streams.  The major one is the recycle
treater vent which has an emission rate of about  13 kg VOC/Mg product,
with other streams contributing about 16 kg VOC/Mg product.  If the
HOPE plant is integrated with an olefin manufacturing operation, the
plant may not have recycle treaters since the ethylene can be purified
in the olefin manufacturing plant.  In this case, overall process VOC
emissions can be expected to be about 55 percent  of the emissions from
those HOPE plants that do not have an adjacent ethylene plant.  Table 3-11
                               3-38

-------
          Table 3-11.   CHARACTERISTICS OF VENT STREAMS  FROM  THE HIGH  DENSITY
                           POLYETHYLENE  LIQUID PHASE  SOLUTION  PROCESS9
Process
Sectionb
RMP

RMP
RMP

PR


PR

MR



MR


MR

PF


PF



Stream^ Name
A Tank vents

8 Cyclohexane treater vents
C Catalyst preparation vent

0 • Compressor vents


E Reactor vents

F Separation vents



Sd Recycle ethylene
treaters


H Recovery distillation

I Extruder vents


J Stripper vents

Z (All indicated streams
to flare)
Total Emission Rate
Emission
rate,
*9 VOC/Mg Temperature, Composition,
Nature product °C % by volume
Continuous; 0.024
cyclic
Intermittent (see stream Z)
Generally (see stream Z)
continuous
Continuous (see stream Z)
and
intermittent
Generally (see stream Z)
intermittent
Some (see stream Z)
continuous
and
intermittent
Continuous 12.7 21


Continuous and (see stream Z)
intermittent
Continuous 0.63 iQ5


Continuous11 0.85 40

14.8
29.0














61.0 etnylene*
1.0 hydrogen
18.0 isobutal
20.0 ethane


VOCf - 0.05
steam and air-
99.959
vocf- o.oi
air - 99.99
VOCi
air

aSource of information:   Industry correspondences.
"Raw materials  preparation PR = polymerization reaction;  PF= product finishing; MR = material  recovery.
csee Figure 3-7 for stream identification.
Emission characteristics are assumed to be the same as for HOPE,  slurry process.
ePercent by weight.
fCyclohexane.
SMostly steam.
"Upsets occur 6 to 7 times per year.
Tlncludes  ethylene, cyclohexane, and  other VOC comonomers.  Ethylene is emitted at about 4  35 ka vor/Mn
 cyclohexane at about 8.7 kg  VOC/Mg product;  and other VOC at  about 1.74 kg VOC/Mg product.          9
                                                 3-39

-------
shows the basic characteristics of the process VOC emissions from
a solution process at an HOPE facility which has ethylene treaters.
     3.2.3.3  HDPE Gas Phase Process.  The HOPE gas phase
process is similar to the LDPE low-pressure process described earlier
(Section 3.2.2.2) in terms of processing steps (Section 3.2.2.2.1),
potential VOC emissions (Section 3.2.2.2.2), and potential growth.
Further, as indicated in Section 3.2.2.2 for the LDPE low-pressure
process, some process equipment can produce both HDPE and LDPE.  Thus,
for the purposes of this project, the process and emission descriptions
for the LDPE low-pressure process serve also for production of HDPE
when the same process is used.
3.2.4  Polystyrene
     Styrene readily polymerizes to polystyrene by a relatively
conventional free radical chain mechanism.  Either heat or a catalyst,
typically benzoyl peroxide or di-tert-butyl per-benzoate, will initiate
the polymerization.  Styrene will homopolymerize in the presence of
inert materials and copolymerize with a variety of monomers.  Pure
polystyrene has the following structure:
     The homopolymers of styrene are also referred to as general  purpose
or crystal polymers.  The copolymers of styrene are generally produced
in the presence of particular elastomers to improve the strength  of the
polymer.  They are called impact or rubber-modified polystyrenes.  For
them, the styrene content varies for impact polystyrenes from about
88 to 97 percent, for styrene-acrylonitrile copolymers (SAN)  from about
70 to 75 percent, and for styrene-butadiene copolymers from 50 percent
and above.  (Styrene-acrylonitrile copolymers and styrene-butadiene
rubbers are not part of this source category.)  Where a blowing (or
expanding) agent is added to the polystyrene, the product is  referred
to as an expandable polystyrene.  This may be done as part of the
process (as in the production of expandable beads) or as part of  the
fabrication process (as in foamed polystyrene applications.)
                               3-40

-------
      Homopolymers and copolymers can be produced by bulk (or mass),
 solution (a modified bulk), suspension, or emulsion polymerization
 techniques.  In solution,  or modified bulk, polymerization  the reaction
 takes place as the monomer is dissolved in a small  amount -of solvent,
 such as ethylbenzene.  Suspension polymerization takes  place with the
 monomer suspended in a water phase.   The bulk and solution  polymerization
 processes are homogeneous  (i.e., take place in one  phase),  whereas the
 suspension  and emulsion polymerization processes are  heterogeneous
 (i.e.,  take place in more  than one phase).  The bulk  (mass)  process  is.
 the most widely used process for polystyrene today.29  The  suspension
 process is  also commonly used, especially in the production  of expandable
 beads.29  The use of the emulsion process for producing homopolymer  of
 styrene has decreased significantly  since the mid-1940's.30
      This section describes  both  the  bulk (mass)  batch  and continuous
 processes for the production of crystal  or impact polystyrene  in Sub-
 sections 3.2.4.1  and 3.2.4.2,  respectively,  and  the suspension process
 for the production  of expandable  polystyrene  beads  in Subsection 3.2.4.3.
      3.2.4.1   Polystyrene  Batch  Process.   Various grades of  polystyrene
 can be  produced by  a variety  of batch  processes.  Batch processes generally
 have a  high conversion  efficiency, leaving  only  small amounts  of unreacted
 styrene to  be  emitted  if the  reactor is purged or opened between batches.
 A typical plant will  have multiple process trains, each usually capable
 of  producing  a  variety  of grades of polystyrene.
      3.2.4.1.1  Process description.   Figure  3-8 is a schematic
 representation of the polystyrene batch bulk  polymerization process.
 Pure  styrene monomer and comonomer (if a copolymer product is desired)
 are  pumped from storage  (1) to the mix feed tank (2), then usually to an
 agitated tank, often a  prepolymerization reactor, for mixing the reactants.
 Small amounts of mineral oil  (as a lubricant and plasticizer), the dimer
 of  alpha-methylstyrene  (as  a polymerization regulator),  and an antioxidant
 are added.  Polybutadiene may be added in the case of production of an
 impact  grade polystyrene.  The blended or partially polymerized feed  is
then pumped into a batch reactor (3).  During the reactor filling process,
some styrene vaporizes and  is vented  through an overflow drum (4).  When
the reactor is charged, the vent is closed and polymerization is  thermally
initiated.  The reaction may also be  initiated by introducing a free
                               3-41

-------
                                                   to
                                                   in
                                                   (D
                                                   O
                                                   O

                                                   a.
                                                   
-------
 radical initiator into the feed tank along with other reactants.  After
 polymerization is complete, the polymer melt, which contains some unreacted
 styrene monomer, ethylbenzene (an impurity from the styrene feed), and
 low molecular weight polymers (dimers, trimers, and other oligomers) is
 pumped to a vacuum devolatilizer (5).  In the devolatilizer the residual
 monomer, ethylbenzene and low polymers are separated, condensed (6), and
 sent to the by-product recovery unit (7).  Overhead vapors from the
 condenser are usually exhausted through a vacuum pump (8).  Molten
 polystyrene from the bottom of the devolatilizer is pumped through a
 stranding dieplate into a cold water bath.  The cooled strands  are
 pelletized (9) and sent to product storage (10).
      3.2.4.1.2  Emissions from the polystyrene  batch process.   The
 process has four major emission  sources,  which  are:   the  monomer storage
 and feed dissolver vent,  the  reactor drum vent,  the styrene  condenser
 vent,  and  the extruder quench  vent.   These are  labeled A,  B, C,  and  D,
 respectively,  in  Figure 3-8.
     The total  emission rate is  estimated  to  range  from 0.61 to  2.5  kg
 VOC/Mg product.   The  major  vent  is the  devolatilizer condenser vent
 (Stream C).   This  continuous offgas  vent has  an  emission  rate of  0.25 to
 0.75 kg VOC/Mg  product.   The emissions  consist of unreacted styrene,
 which  is flashed  from the product polymer  in  the vacuum devolatilizer,
 extremely  diluted  in  air  due to  leakage.   The stream is exhausted through
 a  vacuum system  (i.e.,  vacuum pump and  oil demister), to the atmosphere.
     The VOC emissions  associated with  other  continuous offgas streams
 (A  and  D)  are 0.09 kg  VOC/Mg product and 0.15 to 0.3  kg VOC/Mg product,
 respectively.  Pure styrene is emitted  directly to the atmosphere from
 the monomer storage and feed dissolver  vent (Stream A), whereas steam
 and styrene vapor are usually vented through a forced-draft hood  (Stream D)
 and passed through a mist separator'pad or electrostatic precipitator
 before  venting to the atmosphere from the extruder quench  vent.
     The only intermittent offgas stream from a batch process is the
 reactor drum vent (Stream B).   Its VOC emissions range from 0.12 to
 1.35 kg VOC/Mg product.  Emissions occur from the reactor  drum vent only
during reactor filling periods.  Their filling frequency is once per
day, and the associated offgases  are vented to the atmosphere.   Table 3-12
summarizes  these four sources  of VOC emissions.
                                  3-43

-------











to
CO
co
til
B ' --•
O
o
oi
Q.
O
t
CO

LU
LU
1
2
LU
JC

§
u_
CO
LU
o:
te
i_
LU
LU
O

CO
o
fr~H

co
I—I
o;
LU
O
3
 VI
o> ex
Q.
*
£
3
id
s-o
E"
£

g |V
S.S HI
— Id J-



0)
u
s
re
5£








n
re









"I
L.
{/I






•O
to c
(0 O
0>T-
0 *J
ss
Q- CO
£
tu
OJ Q) (/)
m 0) OJ O °5
S- S-4-> U C

5" 5*35 s- s- c
"* *" ^< 5"5 21

O CT1 • • t-0 >}
o «m co i~- *»
i — voroin co cri t/)










II O 1



co r^ •-*
| «3- CSJ CM


H-
m in
ro r» ro in
r-I C5 0 CSJ
en t ill
o CM in ' in -H
. r-l CM r-H U5
° O O C3 O

CsJ
S 1 S =
5 j-j oo
= "§ J .1
*J Q) 4J *->
1 S 53
OJ

•a re
Di
Q) J=
c> , *£ « =
ro t- c c O
t -iJ fl) O QJ **"
O C N > 3 W
4J o QJ *r~ CT to
to S- > 4-> ^— I- **~
O) C ••- QJ S- E
S- > 1- QJ 4Jt/i  re c TCJ
C O 4-> r— QJ 3 ^
o-awoe o-o !-•»-> re
c oj LO re 3 >c -i-JC -M
OCU-I-GJU OJO XQJ 0
S«4--OC^-0 OU LU> H-









< CD O O




•





C^ Di LU.
•p Cu E CL,






























O
re
OJ
c
o
*^
re
N
i
,_
o
Q.
II


a.
• •«
a> en
u c
esponden
f1n1nsh1
t.
o u
o =
>, o
S- S-
4V O-
to
3 tl
"i LU
^ 0_
O QJ
4-> O

P ^
O
M- f—
c re

t+- Q)
O 4->

§ f
to s
ra .n













































c
4-)
u

jj
E=
a>
•c

i
0)
to
o
M-
CO
t
CO

tu
i.
3

LU
QJ
QJ
to
o



































4->
U
OJ
•r-3
o

Q.
to

•j^

C
^
(U
0
u
o
c
e
re



o
to
10
QJ
Ol
re
s-
o
w

,_
re
s.
QJ

re
E
i
C£
T3






















































,
£.
QJ
Q.
IV
U
c

&-
L.
3
U
o
o
CD
C
o























































to

>.
re

s-
a>
S

a>
CJV
(0
i_

-------
      3.2.4.2   Polystyrene  Continuous  Process.  As with the  batch process,
 various  continuous  processes are  used to make a  variety of  grades of
 polystyrene or copolymers  of styrene.   The chemical  reaction in continuous
 processes  does not  approach completion  as efficiently as the reaction in
 batch  processes.  As  a  result, a  lower  percentage of styrene is converted
 to  polystyrene, and larger amounts of unreacted  styrene may be emitted
 from continuous process sources.  A typical plant may contain more than
 one process train, each producing either the same or different grades of
 polymer  or copolymer.
     3.2.4.2.1 Process description.  The bulk (mass) continuous process
 is  represented in Figure 3-9.  The feed dissolver tank (1)  is charged
 with styrene,  polybutadiene (if an impact grade  product is desired),
 mineral  oil (lubricant and plasticizer), and small amounts of recycled
 polystyrene, antioxidants  and other additives, in proportions that vary
 according  to the grade of  resin to be produced.  Blended feed is pumped
 continuously to the reactor system (2) where it  is thermally polymerized
 to  polystyrene.  A process train usually employs more than one reactor
 in  series. Some polymerization occurs in the initial reactor, often
 referred to as the prepolymerizer.  Polymerization to successively
 higher levels  occurs in subsequent reactors in the series.  Either
 stirred autoclaves or tower reactors are employed, depending on the
 variation  in the process.  The polymer melt,  which contains unreacted
 styrene monomer, ethylbenzene (an impurity from the styrene feed)  and
 low polymers,  is pumped to a vacuum devolatilizer (3).   In the devolatilizer,
 most of the monomer, ethylbenzene, and low molecular weight polymers  are
 separated, condensed (4), and sent to the styrene recovery unit.   Non-
 condensables (i.e.,  overhead vapors)  from the  condenser  are typically
 exhausted through a  vacuum pump.   Molten polystyrene from  the bottom  of
the devolatilizer is pumped by  an extruder through  a stranding  dieplate
 into a cold water bath.  The solidified  strands  are  then pelletized  (5)
and sent to storage  (6).
     In the styrene  recovery unit, the crude styrene monomer recovered
from the condenser (4) is purified in  a  distillation column  (7).  The
styrene overhead from  the tower is condensed  (8)  and returned to the
feed dissolver tank.  Noncondensables  are vented  through a  vacuum system
                                  3-45

-------
                                                              (S)
                                                              CD
                                                              U
                                                              O

                                                              Q_

                                                              t/1
                                                              3
                                                              O
                                                               O
                                                              O

                                                               O)

                                                               cu

                                                               >>
                                                              -1-5
                                                               to
                                                               o
                                                              D.
                                                               cu
                                                               s-
                                                               o
                                                               re

                                                               en
                                                               O
                                                               O
                                                              CO
                                                              ts>
                                                              O)
                                                              O
                                                              O
                                                              -a
                                                              CD
                                                              CL.

                                                              •i—
                                                              oo
                                                              CTl
                                                               I
                                                              oo

                                                              0)

                                                              3
                                                              cn
3-46

-------
 (9).  Column bottoms containing  low molecular  weight  polymers  are  sometimes
 used as a fuel  supplement.
      3.2.4.2.2   Emissions from the  polystyrene continuous  process.  The
 process has  four types  of vent streams,  all  of which  are continuous.
 These are:   the feed dissolver vent,  the devolatilizer  condenser vent,
 the styrene  recovery unit condenser vent, and  the extruder quench  vent.
 These are Streams  A, B,  C,  and D, respectively, in Figure  3-9.  Industry's
 experience with continuous  polystyrene plants  indicates a  wide range of
 emission  rates  from  plant to plant  depending,  in part,  on  the type of
 vacuum system used.   Two types of vacuum systems are  now used in the
 industry.  One  relies on steam ejectors; the other, on  vacuum pumps.
 Where steam  ejectors are used, the  overheads from the devolatilizer
 condenser vent  and the styrene recovery  unit condenser  vent are composed
 mainly of steam.  Some companies have recently  replaced these steam
 ejectors with vacuum pumps.  It is  estimated that the typical total VOC
 emission  rate for plants still using steam ejectors is about 3.25 kg
 VOC/Mg product.  For plants that use vacuum pumps, it is estimated that
 the  total VOC emission rate is about 0.26 kg VOC/Mg product.  Table 3-13
 presents the VOC vent stream characteristics.
      These vent streams differ from those of the batch process.  Emissions
 from the devolatilizer condenser vent tend to be higher from the continuous
 process than from the batch process.  In addition, there is no reactor
 vent  drum vent  in the continuous  process.  The other offgas streams
 (from  storage and the extruder) have emissions comparable to those of
 the  batch processes.
      3.2.4.3  Expandable Polystyrene Suspension Process.  The suspension
 process is a batch polymerization process that may be used to produce
 crystal, impact, or expandable polystyrene beads.   An expandable
 polystyrene (EPS) bead consists typically of high-molecular weight
 crystal grade polystyrene and 5-8 percent low boiling aliphatic
 hydrocarbon blowing agent (typically pentane or isopentane although
 other blowing agents, such  as esters, alcohols, and  aldehydes,  can  be
 used).31  When used to produce an EPS bead,  the suspension process  can
be adapted in one of two ways for the impregnation  of the bead  with the
blowing agent.  One method  is to  add the  blowing agent to a reactor
after polymerization.  The  other  method  is to add  the blowing agent to
                                  3-47

-------

10
to
CO
LU
C_5
O
Q_

CO
O
=3
2=
t—

''C-"
o

LU
LU
C£
1*
co

i
o
0.
LU
3=
r-
i
C£
U_
CO


LU
az
I—
co
r-
z:
LU
^>

LL.
O

CO
CJ
1— 1


)_|
oi
LU

o


«=c

o
oo
r—
co


QJ

ja
(0
l—


















!«
•f" 4J
MS
O
Q.
O

^ "

O "•».4O
f- 0 0
in «o 3
in a»-o
1-40 0
iB t-S a-



s-
s
ra














0)
E
ra











.u

ra
a)
£_
40
01





.O
in c
in o

U 4O
So
0)
a- 

i'o£
40 CU O
OO
u) m
CM • •




i







l






en
g
O


en
o
3
C

•U
cS







£.



O
(A
cn
C3

"S
(U
LU










, Q)
00 01

00 CM
t-1 CO
CM r~




o






o
o
»—


VD
01
CM
1
in
0
o


in
o
3
C

40
 C
0} OJ
Q >










m












^





a>

aj B ra s-
s_ ro i
>, 0) 40 40
40 40 O1 O1
a* cu
»-i rji CTi o
. • . ra
CM r~ 01 s.
Ol Ol r™




O I






O !-<
f— CM



CO , LO
i — CM
O CO
t 1
LO LO IO
O •— CM
O O O


vt in
3 3
O O
1 1

40 40
O O



4O
C C O)
3 CD 40
> ro

0> Q) U C
> > c o
O 0) i-
o s- 3 m
£a> cr in
in T-
= s- E
CO QJ flj LU
C-O -D
OJ = 3 r—
S-O S- ro
40 X O
in LU i—










o a











Of U_
S Q-





























C

in

c
t|~
40
U
O
S.
ex
II

0.

>>
 s-
u
C l —
OJ ro
"O •*-
ir £-
O CO
Q. 40 c
in ro O
0) E i-
s- u ra
g g 1
in o co

1 2 E
fO fO
Q. OJ
• • O) S-
C S- 4->
o Q. tn
fO 
-------
the monomer prior to polymerization.  The former method (which shall  be
referred to as the "post-impregnation" suspension process) is more
common than the latter method3^ (which shall be referred to as the
in-situ suspension process).  Both process are described below.
     3.2.4.3.1  Process description.  The post-impregnation suspension
process is essentially a two-part process using two process trains in
series.  In the first process train, raw styrene monomer is polymerized
and a finished polystyrene bead is produced.  The second process train
takes the finished bead from the first train, impregnates the bead with
a blowing agent, and produces a finished EPS bead.  Figure 3-10 is a
schematic representation of this process.
     In the first train, styrene monomer, water, initiator, and suspending
agents form the basic charge to the suspension reactor (I).33  The
styrene to water ratio, about one-half to one-quarter monomer to water
volume ratio,33 varies depending on the type of polystyrene required.
Initiators are commonly used because the reaction temperature is usually
too low for adequate thermal initiation of polymerization.  Suspending
agents are typically protective colloids and insoluble inorganic salts.
Protective colloids are added to increase the viscosity of the continuous
water phase and insoluble inorganic salts, such as MgCOs, are added to
prevent coalescence of the drops upon collision.33
     In the reactor, the styrene is dispersed as droplets throughout
the water phase.  The droplet size may range from about 0.1 to 1.0 mm.34
The reactor is heated to start the polymerization process, which takes
place within the droplets.  An inert gas, such as nitrogen, is frequently
used as a blanketing agent in order to maintain a positive pressure at
all times during the cycle and prevent air leaks.  Once polymerization
starts, temperature control  is typically maintained through a water-
cooled jacket around the reactor and facilitated by the added heat
capacity of the water in the reactor.  The size of the product bead
depends on the strength of agitation and on the nature of the monomer
and suspending system.33  Between  20 and 70 percent conversion, agitation
becomes extremely critical.35  If  agitation weakens or stops between
these limits, excessive agglomeration of the polymer particles may
occur, followed by a runaway reaction.  Polymerization typically occurs
                                  3-49

-------
                                         cu
                                         c
                                         OJ
                                        o
                                        O_
                                         O)
                                        (O
                                        -o
                                        £=
                                        ro
                                        Q.
                                        X
                                        LU

                                        
                                        -O CU
                                        CU S-
                                        •i- Q.
                                        Q.4J
                                        E  w
                                       •i-  o
                                       OO Q_
                                        I
                                       oo
                                       cu
3-50

-------
within several  hours, the actual  time varying primarily with  the
temperature and the amount and type of initiator(s)  used.36
     Once the reaction has been completed,  the polystryene-water  slurry
is normally pumped from the reactor to a hold tank  (2), which has an
agitator to maintain dispersion of the polymer particles.   Hold tanks
have at least three functions the polymer-water slurry is  cooled  to
below the heat-distortion temperature of the polymer (generally 50 to
60°C (122 to 140°F), chemicals are added to promote  solubilization of
the suspension agents, and the tank serves  as a storage tank  until the
slurry can be centrifuged.37  From the hold tanks, the polymer-water
slurry is fed to a centrifuge (3) where wash water is  added and the
slurry is then centrifuged to separate the  wash water  from the solids.
The wash water is discarded and the polymer product, which may retain
between 1 and 5 percent water,37 are sent to dryers  (4).  From the
dryers, the product beads may be sent to a  classifer (5),  which separates
the beads according to size, and then to storage bins  or tanks (6).
The product beads do not always meet specified criteria for further
processing into expandable beads.  In these instances, the "off-spec"
beads may be processed and sold as crystal  (or possibly impact)
polystyrene.
     In the second train, the product bead  (from the storage  bins of
the first train), water, blowing agent, and any desired additives are
added to an impregnation reactor (7).  The  beads are impregnated  with
the blowing agent through the utilization of temperature and  pressure.
Upon completion of the impregnation process, the bead-water slurry is
transferred to a hold tank (8).  Acid may be fed to  the hold  tank and
part of the water drained as wastewater. From the  hold tanks, the
slurry is washed and dewatered in centrifuges (9) and  then dried  in
low temperature dryers (10).  From the dryers, the  EPS bead may undergo
sizing, if not already done, before being transferred  to storage  silos
(11) or directly to packaging (12) for shipment to  the customer.   In
some instances, additives may be applied to the EPS  bead before shipping
to improve process characteristics.
     The in-situ suspension process is shown schematically in Figure  3-11,
The major difference between this process and the post-impregnation
suspension process is that polymerization and impregnation takes  place
                                  3-51

-------
              0) C
              f-H OJ
              .0 U

              "O 4J
              c in
              n >,
              a..-)
                    to n
              x o  a) 3
              td p*.eq u
                                                O)

                                                CD
                                                o
                                               Q.
                                                O)
                                                (O
                                               -a
                                                a.
                                                x
                                               LlJ

                                                ca
                                                re

                                                cr>
                                               o in
                                                   in
                                               ^: a)
                                                o o
                                               02 a.
                                                in o
                                                
                                               -a 3
                                                0) CO
                                               •I- +J

                                                Q.OO

                                               •i- C
                                               co i— i
                                                I
                                               co
                                                cu
                                                a>
3-52

-------
at the same time in a single reactor.   The reaction mixture,  composed of
styrene monomer, water, polymerization catalysts, and additives,  are
charged to a reactor (1) to which a blowing agent is added.38  The
styrene monomer is polymerized at elevated temperatures and pressure in
the presence of the blowing agent so that the blowing agent is entrapped
in the polymerized bead, which may contain 5-7% blowing agent.38  After
polymerization and impregnation has taken place, the EPS bead-water
slurry follows essentially the same steps as in the post-impregnation
suspension process.  These steps are repeated in Figure 3-11 from Figure 3-10.
     3.4.4.3.2  Emissions from the expandable polystyrene suspension
process.  Table 3-14 summarizes the VOC emission sources and stream
characteristics for the post-impregnation suspension process.  Table 3-15
summarizes the VOC emission sources and stream characteristics for the
in-situ suspension process.
     Available  information on the post-impregnation suspension process
shows  a rather  large variation in emission rates for some of the emission
points.   For  two  plants for which complete information was available,
overall emission  rates of  about  4.2 and 9.5 kg VOC  per Mg of product
were estimated.   An  emission  rate of  about 5 kg  VOC per Mg of product
was  estimated for an in-situ  suspension process.  The  blowing agent,
which  is  continually diffusing out  of the bead  in  the  manufacturing
process and  during storage, constitutes almost  all  of  the VOC emissions
from both processes.   A small  amount  of styrene is  emitted from  the
suspension reactors  in the post-impregnation  process and  from the  mix
tanks and reactors in  the  in-situ  process.
 3.2.5  Polyester Resin
      Polyester resins  used for fiber  production may be classified  chemically
 as either polyethylene terephthalate) [PET]  resins or non-PET  resins.   Of
 these two types, PET resins are the most important polyester as  a
 thermoplastic synthetic fiber.   PET resins may, in turn,  be  classified
 as low viscosity or high  viscosity resins.   Low viscosity resins are
 used in such applications as fiber and bottle production.   High viscosity
 resins are used in such applications as tire cord.
      PET resins are produced commercially from ethylene glycol  (EG) by
 either the dimethyl terephthalate (DMT) process or the terephthalic acid
 (TPA) process.  Both  processes first produce the intermediate
                                   3-53

-------








LU
LU
h-
CO
_J
2
LU
CD
•=c
o
«CM
^
X
LU
rtJ
LU CO
0= CO
ss
a: a.
u_
CO O
3! 1 —
eciSi

CO LU
f_^
LU *-<
u, to
0 0
a.
co
O
E
. CHARACTER]

^T
' -
CO
(U
s
(U
f—










c
o
"^ **
4J
Ul J->
O X
§•
"


0)
3 en
to v)
a.

01
^
•»-»
So
ate
g
|w



cn
§3C 1 *
»-*^, (J
1n

iyi







o"o
U QJ
cu to
t-
o
a.

:>

i.
•o o)
On
CM CO




1





1





CO
r-4
O
1
C3
O

c:
0)
-M
ji
1
•u


IA
c
0)
>
o
Suspension react






<









£





0 I-
>5





i





i






CM
O


1
01
4J

trt
>
C.
O
4J
U
Impregnation rea






CO









LU
CU




tt—
>5
CM CO
-§




O





t
CM






««4
CO
•
CO

c
0)
-M
4->
I
4-3
C



(A
4-»
C
S
c
10
cn
c
•5
I






u









Lu
(X



<4-
C3 ^
>5
Sco
ocn
cn




o





i— <
CM





CO
CM
I
CM
LO
l-H

(A

O
3
+J
O
U




U>
C
(11
c
(O
.c
to






a









LU
CU



t^
o t-
O T-
CM CO
o cn
o cn
cn




o






CM




CO

IO
• o

c:
(A 0)
3 4j
O J->
1 1
4_> rti
CZ 4^
o c
o >—


CA
4J
C
s
c
s
o
L-
in Q.
c •»-
a>
& i
>! O
C. U,
O 0-






LU Lu









LU LU
a- a.



t^.

°5
t*-» ro
l-H CO
o cn
cn




0






CM





CM
CM
1
ft
O

vt
zs
o
3
• c
•M
C
O
O




Storage vents






to









a.



i^.
o t.
O •!-
rt en
^ CO
O cn
cn




0






CM



.C
^***
cn •
CM cn
.-i i
1 ^
CM >-l
• •
O «9-

•p
C
0)
4J

1
4J
C


W
01
V)
V)
0 Ol
(K
Packaging system
Total Emission R<






3?









O_





































0)
U
r*
stry correspondei
•3

_

C
O
4J
(O

O
«4_
c
'^"
o

O)
u
t_
3
O
to
10



























OJ
en
IO
c.
o
t>

u
o
0.
II
CO

• •1
en
c
c/l
"c
; PF = product f1
dentificatlon.

o
4-> 
4-*
re
u

• o

c to
• o> cn
01 cn i-
S *° m
•r- C7)
cn c <4-
*f- -r- O
'cn o E
0) r- 3
2Z CQ tO
d) **- ,JZ
3-54

-------










UJ
LU
C£
1—
oo

i
O
a.

UJ
— i
ca
2
•^f

y-
X
UJ
UJ
JC
I—
258
o oo
oz oo
U. LU
oo o
2: o:
 O "O
tn (O > O
•r- t- [_
E 01 Q.
LU .*





U
3
Z













03
m
z












E
01
£
oo






V) CZ
tn o
u 4J
O U
Q- OO




c a>
01 =
1- 01 T3 -o
4-> CJ -r- U>,(- Oi- O T-
oo«t< o <->•.- >«c »«r
> OO «£
LO CO CM ^O ^- uo LO
ooco cOT*-*cn ocn ocn
en en






0 O 00



fO
i

«— * m ^t* r*^
CM CO CM CM





CO LO «»•
CM «S- IO CM
>-i —« O O
o •-< o o


^ °c ^
a> a) o t/i
4-> 4-> 4J 3
4J J-> 4-> O
1 1 I 1
CD ai ai ' j_>
c c co
i— t-t 1-1 O








ul
c:
OJ u>
VI > 4J
4J « =
c 4J -*: en
(U C C >
> ai 10

-x c
C t- 01 ID
10 O C 4->
O "5 -C
X 10 i— in
•i- 

o>
J_

a








LU












LU
a_





•o -o
§.< §<
CM OO CM CO
Sen o en
en o en
o en o en
en en






0 0



co co
CO CO
l i

r*. I-H
CM CM





T 00
r- en
0) . .
0 
0) V>
•M 3
4J O
1 1
a> -M
4-> C
C O
i-" O



V)
•£
ai

*-> 0)
C 4->
OJ 10
E cc
tu
> c
o to o
t- 4-> •!-
Q. c tn
E 111 tn
•^- > •!-

4-> 0) UJ
0 01
•o i- 're
O O *J
I- 4-1 0
CU OO 1 —








LU 
•o

o
CL

E
o
4-
£Z


M-
O

CU
O
3
O
CO
ra














*
OJ
o

o
t/)

u
TJ
O
t_
CL

II
CO
°"
en
c
IE
"E
o
3
o
L.
CL
n

u.
Q-
C
o
4-«
U
ra
0)
t_
c:
o

(O
N
L.
ai

>•••* CZ
o o
Q. i-
II (O
(J
Q_ t+-
CZ CZ
O 0)
ra *"
t- E
ra ra
CL OJ
OJ t.
Q. vi
CA L.
r— O
ra tf~
L. ^-C
CD t— * *


E CU

-------
bis-(2-hydroxyethyl )- terephthal ate (BHET) monomer and then polymerize  it
to PET under reduced pressure with heat and catalyst.   Emissions  from
the two processes differ in that the DMT process produces methanol  as  a
by-product during esterification, whereas water is a by-product in  the
TPA process.  The production of methanol vapor in the DMT process creates
the need for methanol recovery and purification operations and their
attendant VOC emissions.  Most, if not all, high viscosity PET is produced
in plants using the TPA process.  Both the DMT and TPA processes  are
described in this section.
     3.2.5.1  PET/DMT Process.  The DMT process is the older of the two
processes for making PET.  Currently, most PET is produced using the DMT
process, which may be either batch or continuous.  The basic differences
in going from the batch to the continuous process are (1) the replacement
of the kettle- reactor with a column-type reactor for esterification,
(2) "no-back-mix" (i.e., no stirred tank) reactor designs are required
in the continuous process at the polymerizer, and (3) different additives
and catalysts are used to assure proper product character!" sties. 39
     3.2.5.1.1  Process description.  A schematic diagram of the continuous
DMT process is presented in Figure 3-12.  The primary reaction is:
CH3OOC
COOCH3 + HOCH2CH2OH-*-HO - (OC
                                                  COOCH2CH20)n H + 2nCH3OH
        DMT              EG                PET
     Dimethyl terephthal ate and ethyl ene glycol are purified (1) before
being fed to the esterification reactor (2).  There, in the presence of
a catalyst, they react to form bis-hydroxyethyl terephthal ate (BHET) and
methanol.  The continuous removal  of methanol is necessary  in order to
shift the reaction equilibrium in favor of increased production of BHET.
Therefore, a vent stream is withdrawn from the esterifier.   This stream
is fed to a methanol recovery process (3) where methanol is condensed
and purified by distillation, before being forwarded to the methanol
storage tank (4).  The BHET monomer is polymerized (5) to PET in a
second reaction step under reduced pressure with heat and a catalyst.
The polymerization reaction may be carried out in two or more reactors
in series operated at increasing temperatures and successively lower
pressures.  Unreacted ethyl ene glycol is flashed from the polymer product,
condensed, and, if desired, can be purified and recycled.
                                  3-56

-------
                                                        CO
                                                        CO
                                                        cu
                                                        o
                                                        o
                                                       LU

                                                       Q.
                                                        CU
                                                        i-
                                                        o
                                                        2
                                                        en
                                                        (O
                                                        o
                                                        o
                                                       r—
                                                       CQ

                                                        CO
                                                        CO
                                                        01
                                                        u

                                                        2
                                                       Q.
                                                        Q.

                                                       •r-
                                                       oo
                                                       oo
                                                        CU
                                                        C7)
3-57

-------
     3.2.5.1.2  Emissions from the PET/DMT process.  The methanol recovery
section and the reactor are the major sources of VOC emissions from the
DMT process.  Both are continuous.  These are shown as Streams A and B
in Figure 3-12.  A summary of the vent stream characteristics is provided
in Table 3-16.  Emissions from the methanol recovery section are recovered
by condensers, which results in relatively low emission rates.  This
stream is typically composed of methanol and nitrogen.
     The emission streams from the reactors are composed primarily of
ethylene glycol with small amounts of methanol vapors and volatile feed
impurities.  The amount of ethylene glycol that is emitted to the atmosphere
depends upon the number of polymerizers used and on the type of system
used to recover the ethylene glycol.  A plant may recover the ethylene
glycol by using a spent ethylene glycol spray condenser directly off of
the reactors and before the stream passes through the vacuum system.
The condensed ethylene glycol may then be recovered through distillation.
This type of recovery system results in low emission rates.  Alternatively,
a plant may send the emission stream directly through the vacuum system
(typically composed of steam ejectors) without using a spent ethylene
glycol spray condenser.  The steam ejectors used to produce a vacuum
will result in contaminated water, which is then cooled for reuse.
Ethylene glycol in the cooling tower is recovered from the bottom of
the cooling water with distillation columns.  This system of recovering
ethylene glycol results in much higher emission levels of ethylene
glycol, because of the cooling water's contact with the atmosphere.
     3.2.5.2  PET/TPA Process.  The TPA process, which has been available
only since 1963, is now generally preferred over the DMT process because
it avoids the recovery and purification of the methanol by-product
generated in the DMT process.  The TPA process can also be a batch or
continuous process.
     3.2.5.2.1  Process description.  A schematic diagram of the continuous
TPA process is presented in Figure 3-13.  The primary reaction is:
HOOC -<> COOH + HOCH2CH2OH —^ HO - (OC -<> COOCH2CH20)n H + 2nH20
       TPA
EG
PET
                               3-58

-------
T
— i
o
a.
LU
i
*
S
o
a:
u_
oo
2£

10
z







i
(O




^j-
!
to





m e
I/I o
01 T-
o u
s» o


i— C
is,
15

s: "^
01 i-l
^c in




o







rr>



S
o




in
o
5
e
o






Esterlfler Vent
(Methanol recovery
vent)






in
e
01 o>
IA cr "t5
Polymerizer Reactor
Cooling Tower Water
- with spent ethyl e
glycols spray con





m






oz
Q_










"=














in
d) fO OO
in in to
o> o oi









s-

in
o> > M >> 0.

recov

•2


-u
(O
Q,
3£
.a










































m identification.

0>
i.
in

s-
£
CM
1
CO
01
l_
3
en
uZ

e
0
u
3S
t^
Q.
in
fS
UJ
5

"5
10
k
e
o
in
in
a.
t—
13

3

0)



C 01
*t» C
o
u s.
01
4:52
o
in 4-»
S ^
5«
in
•~ e

v .^*
•— IO

t.
si I-
O) O
i- U.
O.O
Q. 0)
*-•§
0) 3
£ **

c
V) O
10 0.
3

E O)
10 C
in >r-
•o
ja v
Q.
s-s


0) w
•U 10
ie >
in ^»
LU X
0)
3-59

-------
                                CO
                                CO
                                O)
                                O
                                a

                               cu
                               LU
                               Q_

                                0)

                               •*->
                                CD
                                to
                               »^-
                               Q
                                O
                                a
                               r—
                               CO

                                CO
                                CO
                                
-------
     The processing  steps for producing PET by this method are very similar
 to those of the DMT  process except that water rather than methanol is
 the by-product from  the esterifiers.  Thus, the TPA process does not
 have methanol recovery and purification.
     Polymerization  is carried out as with the DMT process - in two or
 more reactors in series operated at increasing temperatures and
 successively lower pressures.  In the production of high viscosity
 PET, there are two processes currently being used that are distinct from
 each other.  One process can be described as essentially the low
 viscosity process with a single, additional end or final polymerizer.
 In this process, the first polymerization reactor, which may be called
 a prepolymizer, pre-polycondensation reactor, or low polymerizer,
 operates at temperatures between 220 and 290°C (430 and 555°F) and
 under vacuum [e.g., between 10 and 20 torr (10 mm Hg to 20 mm Hg)].40.41
 These temperature conditions are slightly higher and the vacuum level
 range is slightly smaller than for the production of low viscosity PET.
 The additional  reactor, which may be. called an end finisher or high
 polymerizer, may operate with a temperature of between 290 and
 300°C (555 and 570°F) and a vacuum level  of approximately 0.9 torr
 (0.9 mm Hg).41,42
     The other high viscosity process contains a greater number of
 process vessels.  The difference of primary importance is the number
 and operating conditions of the polymerization vessels.  In this second
 process, the production of high viscosity PET takes place in two or
more end finishers rather than in one as in the process discussed
 above. The first end finisher is typically operated with an intermediate
 vacuum level of about 2 torr (2 mm Hg).   The polymer leaving this
 reactor then enters a second end finisher, which may have a vacuum
level  typically around 0.5 torr (0.5 mm  Hg)43 ancj as iow as Q.25
torr (0.25 mm Hg).44
     3.2.5.2.2   Emissions from the PET/TPA process.  There are two sources
of VOC emissions from this process as illustrated by Figure 3-13.   These
streams, which  are continuous, are emitted from the esterifiers (Stream A),
and from the polymerizers (Stream B).  Offgas  characteristics  for these
streams are summarized in Table 3-17.
                                  3-61

-------
10
to
to
UJ
o
o
C£
O.

«=C
Qu
r-

UJ
r—
	 |
M^
^^
"j_~
fe
Q.
Ul
cc.
UJ
r—
LU
LU
_j

rc
f™*
UJ


o
Q.
UJ
(— -
i
§
a:
LI-

CO
?3
UJ
C£



J-
Ul
5*

U.
O
CO


1 —
to
Oi
UJ

O



i-T?
(L~5
V™/


»


i
oo
0)
S
to
•k
c
^
4_> ^^
Is
§
s



O 5.4->
t- 00
in -o =3
in Q) >• "O
•r- 4-» O
E ra D>s-
UJ I_J£ C-



e
s
ra
s












i
(O











O
I
01
1.

C/J


.0
in c
in o
ss
0 U
0. CO



O O O S- O
O OJ CM O *f- OJ





E OJ
4-» •
•s o







CO LO
CO OJ




OJ
LO
O CO

o o


in in
I i
4-> 4J
C C
O O
t _\ f_)

0>

O) C
in co
5- J- ttl-O
O  2 >,0
T5 O O *— u
•w ro 1— 4->
c= ,
0) of t" ra
s» a> 4J s-
i. 4-> CO.
s_ a) ra aj in
O) N 2C CL
•^ "u en ""o
•^ 0) C JI U

•2 ^""o Ic "o>
in o o
ii i Cv O 1










«C CO





a.
S £


























LO
"*- IS
tO *^ LO

CT> O en





I_
CU

(US- CU CU
cu v) c c:
r— C OJ o
>> CD X« U
-c?-a a> c
4-> C 4-> O >,
cu o Co u re
O C£ l-
C >j c ie1 W
CU CO OS.

tn Q. to t/) UJ
to 01
O*O LU O
JC O J= J=

?a. 5 ? 1
0
1 r— 1 1
























































CD
U
a>
o
CL
in
£
0
U

>>
J->
to
•o

I-" 1

c
o

J_>
re
E
o

c


4-
O
i
o
re .

































c
o
•f—
u
ra
at
c
o


ra
M
i-
01
|

Q
CL
II

o-

o

ro
ra
CL
£
a.

r~-
re
1-
OJ

re
E

(O
s-
II
ex.
E
o











































K
o

re
u

•r-
C
CU
•o

E
CU

4->
tn

s-
o
M-
fH
t— t
1
ro

CU
S-
3
LL.
a>
O)
00
u





CU

3
4_j
O
V)
*f—
OJ
to
C
CU
•a
c
o
u

«*
CL
V)
CU
.c
c
CU
-i
s_

o

CU
re
3
c
j—
o
o
u
s.
o

u
re
i-
o


c
CU
CA
£
re

re



o
rjj
S-

I


u
3
-o
£
CL

€

tj
o


o>

OJ
0

o


o
CL
3
O)
in
ro
f«










M-
O
O)
.a

^
c:
(U
4J
O

-^
O)
c.
•o
c
o>
CL
OJ
-o
S—
re

r—
s

re
*~
0
V)

OJ
re
Q


CU
-d
" 4J
ro
§r—
^.
 «
O 4_>

N
Include
polymer'
3-62

-------
      The overall  emission rate from the production of low viscosity PET
 is approximately  0.36 kg VOC per Mg of product.   Of this  amount,  about
 0.04 kg VOC per Mg of product is from the esterifiers and about 0.32 kg
 VOC per Mg of product is from the polymerization  reactors via  the
 cooling tower.
      The overall  emission rate from the production of high viscosity
 PET depends upon  the particular process.   In  the  first high viscosity
 PET process discussed above, the overall  emission rate is about 0.36 kg
 VOC per Mg of .product.   Of this 0.36 kg VOC per Mg of product,  about
 0.04 kg VOC per Mg of product is from the esterifiers, which are
 controlled by reflux condensers, and 0.32 kg  VOC  per  Mg of product  is
 from the polymerization  reactors via the  cooling  tower.  In the second
 high viscosity PET process discussed above (i.e.,  where two or  more  end
 finishers are used),  the overall  emission rate has been estimated to be
 about 3.55 kg VOC  per Mg of product.   Of  the  overall  emission rate,
 about 0.15 kg VOC  per Mg product is  from  the  esterifiers,  which are
 controlled by distillation  columns,  and about 3.4 kg  VOC  per Mg product
 from the polymerizers via  the  cooling tower.
      The difference  in the  emission  rates from the  polymerization
 reactors (via the  cooling  tower)  between  the  two  high  viscosity processes
 is  due  to  the ethylene glycol  recovery  system used  with the  two types of
 high  viscosity  PET processes.   In the  first process, ethylene glycol  is
 recovered  directly from  the  reactors  by spray condensers and from the
 esterifiers by  reflux condensers.   In  the  second  process,  ethylene
 glycol  envisions from  the polymerizers  are  allowed  to pass  directly to
 the  vacuum  system  and into  the cooling  tower.  The  ethylene  glycol is
 then  recovered  from the water  in the  cooling  tower.  This  arrangement
 allows  for  a  higher ethylene glycol concentration  in the cooling  tower
 and a larger  loss  of ethylene  glycol  to the atmosphere  from windage.
 3.3   FUGITIVE VOC  SOURCES AND  EMISSIONS
      Fugitive VOC  emissions  result when process fluids  leak from the
 plant equipment.   The potential fugitive VOC  sources in the polymers and
 resins  processes are similar to those in  synthetic organic chemicals
manufacturing and other industries.  Sources include valves, pump
 seals, compressor  seals, safety or relief valves,  flanges, sampling
                                  3-63

-------
connections, and open-ended lines.  Fugitive emission sources are
extensively described in References 45 through 48.
     Table 3-18 lists the vapor pressures of the various organic compounds
used in the polymer and resin processes.  Those compounds with vapor
pressures greater than 0.3 kPa (2.25 mm Hg) at 20°C (68°F) are considered
to be "light liquids," and those with vapor pressures equal to or less
than 0.3 kPa at 20°C are considered to be "heavy liquids."
     Data characterizing the uncontrolled levels of fugitive emissions
specifically for the polymers and resins industry are generally unavailable
at the present time.  However, data of this type have been obtained for
the synthetic organic chemical manufacturing industry (SOCMI) and
several other industries.  Because the operation of the various process
equipment in the polymers and resins industry is not expected to differ
greatly from the operation of the same equipment in the SOCMI industry
and because the chemicals processed in polymer and resin manufacturing
are the same chemicals that form a large portion of the SOCMI data base,
the SOCMI fugitive emission data are used to approximate the levels of
fugitive emissions in the polymers and resins industry.  The SOCMI emission
factors, which are used for the purposes of this project, are presented in
Table 3-19.  {See Reference 48 for a detailed discussion of the derivation
of these emission rates.)
3.4  BASELINE EMISSIONS
     The baseline emission level is that level of emissions achieved in
the absence of additional EPA standards; in this instance, in the absence
of a polymers and resins  NSPS.  This section describes briefly the
various industrial practices and existing regulations that affect baseline
emissions.
3.4.1  Process Emissions
     Process emissions  in the polymers and  resins industry are controlled
through both industrial practices  and government regulations.   Industrial
practices  for polyolefin  production are  distinctly different than for
either polystyrene or polyester production.  The following sections
describe  briefly the relevant industrial practices and  regulations.
     3.4.1.1   Industrial  Practices.   In  general, most polyolefin plants
control the large volume, intermittent  streams with  flares to avoid
                                   3-64

-------
       Table 3-18.  VAPOR PRESSURES OF MAJOR ORGANIC COMPOUNDS USED IN
                    THE POLYMERS AND RESINS MANUFACTURING
Organic Compound Vapor Pressure @20°C (kPa)
Ethylene
Ethane
Propylene
Propane
Iso-butane
Butene
Methanol
Isopropyl Alcohol
Ethyl benzene
Styrene
Ethylene glycol
gas
gas
gas
gas
gas
gas
12.3
4.4
1.0
0.7
0.01
Process
PP,PE
PE
PP,PE
PP
PE
PE
PET (DMT)
PP,PE
PS
PS
PET
Key: PP = Polypropylene
PE = Polyethylene
PS = Polystyrene
PET = Polyethylene terephthalate
DMT = Dimethyl terephthalate
Source:  Weber,  R.C.,  P.A.  Parker,  and M.  Bowser.  Vapor  Pressure
         Distribution of Selected Organic  Chemicals.   EPA-600/2-81-021.
         IERL,  U.S.  EPA,  Cincinnati,  Ohio.   February  1981.   (Docket
         Reference Number II-A-38.)
                                 3-65

-------
              Table 3-19.   UNCONTROLLED  FUGITIVE EMISSION RATES
     Fugitive
  emission source
Uncontrolled emission
 rate,3 kg/hr/source
Valves
  Gas
  Light liquid^
  Heavy liquid^
Pump seals
  Light liquid^
  Heavy liquid0
Compressor seals
Safety or relief valves
    Gas
Flanges
Sampling connections
Open-ended lines
        0.0056
        0.0071
        0.0023
        0.0494
        0.0214
        0.228
         0.1040
        0.00083
        0.015
        0.0017
 These uncontrolled emission levels are based on the data presented in
 Reference 48.
 \ight liquid is defined as a liquid with a vapor pressure greater than
 that of kerosene.
 "Heavy liquid is defined as a liquid with a vapor pressure equal to or
 less than that of kerosene.
                                   3-66

-------
 buildup of explosive concentrations within the plant.  Continuous streams
 may also, on occasion, be controlled by a flare.
      Liquid phase polypropylene plants do not routinely install  VOC
 control equipment on the smaller continuous emission streams.  The
 polymerization reactors and the diluent separation and purification
 units are generally provided with emergency relief valves leading to a
 flare (for safety purposes) in case of an upset.  These emergency vents
 usually pass through "knock-out" drums, which disentrain liquids and
 polymer particles before the vapors are released to the flare.   As
 polypropylene units are subject to plugging,  most are provided with
 emergency relief valves throughout the entire process and,  in the great
 majority of cases,  these relief valves discharge to the flare header.
      Both vent streams from the gas phase polypropylene plant are flared.
      In the high-pressure,  low density polyethylene plant,  flares are
 generally used to control emissions from safety  relief valves, except
 from  the  reactor.  The emergency reactor vent stream typically is not
 flared,  because safety considerations  dictate the need for  a  particulate
 removal  system.   Based on available information,  only  one company has  a
 particulate polymer  removal  technology that can  handle high-pressure
 emergency  vent gas,  and  this system is  used on tubular reactors.   Controls
 are not  routinely applied to the  dryer  and bin storage vents.
      Low-pressure LDPE and  gas  phase HOPE polyethylene  plants (based
 upon  Union  Carbide's Unipol process), currently  flare  both  the continuous
 and intermittent  emission streams.  The flare system has a  liquid seal
 and is air  assisted.49
      In liquid  phase,  high  density  polyethylene plants, a flare is
 generally installed as part of  the  safety system.  Safety relief devices
 leading to  the  flare are utilized to avoid accidents resulting from
 equipment overpressurization or other malfunction.  While recycle treater
 vent streams are  usually flared, dryer vent streams are usually vented
 directly to the atmosphere.   This is done because traditional  dryers
 dilute the organic with large quantities of air,  making the cost  of
burning the organic prohibitive.
     In the polystyrene industry, no routine control is applied  to the
batch or continuous processes used to produce a crystal or impact grade
polystyrene other than  condensation operations in which styrene  is
                                  3-67

-------
recovered, due to its value and ease of recovery.   Offgas from the
styrene condenser and other vents are usually vented directly to  the
atmosphere.  Similarily, no routine control  is applied in expandable
polystyrene plants.  There is some use of condensers or combustion
devices, but control varies widely between plants.  Flares are not
usually installed in any polystyrene plant.
     In the polyester industry, control devices (incinerators or  flares)
are not used.  Rather, as in the polystyrene industry, condensers and
distillation columns are installed to recover methanol and/or unreacted
ethylene glycol because of their value.
     3.4.1.2  Current State VOC Regulations.  Many of the these types of
polymer and resin plants are located in five States (California,  Illinois,
Louisiana, New Jersey, and Texas) which have regulations that limit VOC
emissions.  Almost all of the polyolefin plants are located in Texas  or
Louisiana.  Only one of the polyester plants, however, is located in  any
of these five States; it is in New Jersey.
     The five States have different VOC regulations.  These are summarized
below:
     1. In California, the South Coast Air Quality Management District
        (SCAQMD), which contains five of the six polystyrene plants in
        the State, imposes an emission limitation of 34 kg/day (75  Ibs/day)
     2. Illinois allows new sources to comply with either of two  standards:
        a.   No waste gas stream discharged into the atmosphere in
             excess of 100 ppm equivalent methane (molecular weight of
             16.0) [Rule 205(g)(l)(A)(iii)], or
        b.   A maximum of 8 pounds per hour of organic material
             [Rule 205(g)(l)(C)(1)].
             Emissions of organic material in excess of 8 pounds per
        hour is allowed, provided such emissions are controlled by
        State agency-approved air pollution control methods or equipment
        capable of reducing 85 percent or more of the uncontrolled
        organic material that otherwise would be emitted to the atmosphere
        [Rule 205(g)(l)(C)(ii)].
              In addition, in the case  of emissions from vapor blowdown
        systems or any safety relief valve, except such valves not
                                   3-68

-------
    capable of causing an excessive release (a discharge of more
    than 0.65 pound of mercaptans and/or hydrogen sulfide into the
    atmosphere in any 5-minute period), such emissions must be
    controlled to 10 ppm equivalent methane or less, through combustion
    in a smokeless flare, or by some other State agency-approved
    control device [Rule 205(g)(2)(A-C)].
 3. Under Louisiana law, new sources must burn VOC'waste gas streams
    at a minimum temperature of 704°C (1,300°F) for 0.3 second or
    longer in a direct flame afterburner or any equally effective
    device [Section 22.8(a)].  This law allows the following exemptions
    [Section 22.8(c)(l-3)]:
    a.   Waste gas streams with less than 100 tons/yr VOC emissions.
    b.   Waste gas streams that will not support combustion without
         the addition of auxiliary fuel.
    c.   Waste gas streams where disposal  cannot be practically or
         safely accomplished by other means without causing an
         economic hardship.
         Also significant is that this section (Section 22.8) does
    not apply to safety relief and vapor blowdown systems where
    control  cannot be accomplished because of safety or economic
    considerations.
         In addition, Section 22.9 limits emissions of organic
    solvents from any source which uses organic solvents to
    1.3 kilograms (3 pounds) per hour or 6.8 kilograms (15 pounds)
    per day.  Where emissions exceed these amounts, they must be
    reduced where feasible by one or more of the following methods:
    a.   Incineration,  provided 90 percent of the carbon in the
         organic compounds being incinerated is oxidized to carbon
         dioxide [except as provided in 22.9.3(a)].
    b.   Carbon adsorption of the organic  material.
    c.   Any other equivalent means as may be  approved by the
         Technical  Secretary.
         Where a waste  gas stream may be subject to both Sections  22.8
    and 22.9,  then that stream must show compliance  with both.
4.   New Jersey law regulating VOC emissions uses a  "sliding scale"
    to determine allowable emissions.   Vapor pressure  and concentration

                              3-69

-------
       of VOC in the vent stream are used to determine applicable
       exclusion rates and maximum allowable emissions.  The exclusion
       rates range from 0 to 3.2 kilograms (7 pounds) per hour and the
       control efficiencies required to comply with the maximum allowable
       emissions range from 85 to 99.7 percent.
    5.  The Texas Air Control Board (TACB) regulates new polymer and
       resin plants on a case-by-case basis, requiring the application
       of best available control technology  (BACT).  The TACB follows
       several general "rules of thumb" in making BACT determinations:
       a.   All waste gas streams, including analyzer vents, cannot
            be vented directly to the atmosphere, unless provided for
            in a special provision in the operating permit;
         b.   Reactors are to have pressure and temperature controls to
             minimize pressure excursions that require emergency
             releases;
         c.    Inlet and outlet valves to the  reactors are to be placed
             as close to the reactor as possible to help minimize
             emissions in cases of emergency releases; and
         d.   VOC emissions  from the extruder, pelletizier, and end-product
              storage  should be able to meet  a limit of 350 Ibs of VOC
             per million Ibs of product.  The exact emission limit may
             be  higher or lower,  depending on the  individual case.
3.4.2  Fugitive  Emissions
     There  are  presently  no  Federal  regulations that  specifically reduce
emissions from  polymers and  resins manufacturing  plants.  However, some
fugitive  emission  reduction  is  achieved  by operating  practices currently
followed  by industry  and  applicable  State or  local  regulations.
     3.4.2.1  Industrial  Practices.   The industrial practices used by
the SOCMI industry  are typically  used  in the  polymers  and resins  industry.
Their primary reason  for  controlling  fugitive emissions  is  the economic
loss from leaks.   Such leaks are  usually large enough  to be physically
evident (that is,  can be  seen,  heard,  or smelled)  and are termed  "easily
detectable leaks."   These are normally repaired  to minimize the loss  of
product.   Fugitive  emissions, as  considered  in  this report, are considerably
smaller and less readily  identified than "easily  detectable leaks."   For
                                  3-70

-------
a detailed description of fugitive emission control practices that may
be used, see References 46 and 47.
     3.4.2.2  Existing Regulations.  There are two types of regulations
that affect fugitive VOC emissions from polymer and resin plants.  The
first regulates industrial operating practices on the basis of worker
health and safety.  Because some aspects of these regulations deal with
worker exposure to process emissions, they may have some impact on
fugitive VOC emissions.  The second type is regulations that were
specifically developed to limit fugitive emissions.
     3.4.2.2.1  Health and safety regulations.  Several regulations have
been developed under the direction of the Occupational  Safety and Health
Administration (OSHA) and the National  Institute for Occupational Safety
and Health (NIOSH) to limit worker exposure to chemical substances.
Protecting the workers may be accomplished by limiting emissions or by
providing workers with individual safety protection from the emissions.
Thus, although present health and safety regulations do not mandate a
reduction in fugitive VOC emissions, and any reduction in fugitive
emissions resulting from these regulations is '-incidental," OSHA regula-
tions may affect a company's attitude about leaks and fugitive VOC
emissions.
     3.4.2.2.2  Fugitive emissions regulations.  Currently, only
California and Texas have fugitive regulations and they apply only to
new polymer and resin plants.
     California presently prohibits open-ended process lines to minimize
fugitive VOC emissions.  This State also requires relief valves to
discharge to a flare system, and be monitored and maintained, or a
rupture disk to be used.   In addition to these regulations, the SCAQMD
also requires these plants to vent fugitive emissions from compressor
seals to a fired-heater or flare system.
     Texas requires new polyolefin plants to use enclosed compressors to
vent fugitive emissions to a fired-heater or flare system.
                                  3-71

-------
 3.5  REFERENCES FOR CHAPTER 3

  1.  Click, C.N and O.K.  Webber.   Polymer Industry  Ranking  by VOC  Emissions
      Reduction that would occur from  New  Source  Performance Standards.
      Pullman-KeHogg Company.   EPA Contract  No.  68-02-2619.  1979.
      Docket Reference Number II-A-10.*

  2.  The Society of the Plastics  Industry, Inc.  The Story of the  Plastics
      Industry.  1977. p.  31-32.   Docket  Reference  Number II-I-31.*

  3.  The Society of the Plastics  Industry, Inc.  Facts and Figures of
      the Plastics  Industry.  New  York, New York, 1978.  p. 63.  Docket
      Reference Number II-I-38.*

  4.  Kurtz, S.J.,  L.S. Scarola, and J.C.  Miller.  Convert LDPE Film
      Lines  for LLDPE Extrusion.   Plastics Engineering.  June 1982.
      p.  45.   Docket Reference Number  II-I-92.*

  5.  Reference 2.   p.  29.

  6.  Reference 3.   p.  59.

  7.  Modern Plastics  Encyclopedia, 1981-1982.  McGraw-Hill Inc.  p. 66.
      Docket Reference Number II-I-75.*

  8.   Reference 3.   p.  55.

  9.   Reference 2.   p.  32.

10.   Reference 3.   p.  65.

11.   Bhatia, Jeet  and  Rossi, R.A.  Pyrolysis Process Converts  Waste
      Polymers  to Fuel Oils.  Chemical  Engineering.   October 4,  1982.
      p.  58.  Docket  Reference Number II-I-100.*

12.   Reference  1,  p.  179.

13.   Cipriani,  Cipriano and C.A. Trischman, Jr.  New catalyst  cuts
      polypropylene costs and energy requirements.  Chemical  Engineering.
     April 20,  1981.  pp.  80-81.  Docket Reference  Number  II-I-72.*

14.  Albright, L.F.  Processes  for Major Addition -  Type Plastics
     and Their Monomers.   McGraw Hill  Book Company.   1974.   p.  150.
     Docket Reference Number II-I-19.*

15.  Brydson, J.A.  Polyolefins  Other  Than Polyethylene, and Diene
     Rubbers.  In:  Plastics Materials.   ILIFFE Books,  London,  1970.
     p. 134.  Docket Reference  Number  II-I-13.*

16.  Report of visit to the Texas  Air  Control  Board, September 20-21, 1982,
     from Ken Meardon to  Polymers  and  Resins  NSPS file.  Docket Reference
     Number II-B-50.*

17.  Reference 3, p. 58.
                                  3-72

-------
18.  Sittig', Marshall.  Polyolefin Production Processes—Latest
     Developments.  In:  Chemical  Technology Review,  No.  70.
     Noyes Data Corporation.  1976.  p.  48.   Docket Reference
     Number II-I-21.*

19.  A Leap Ahead in Polyethylene Technology.  Chemical  and  Engineering
     News.  December 5, 1977.  p.  22.  Docket Reference  Number II-I-33.*

20.  New Route to Low-Density Polyethylene.   -Chemical  Engineering.
     December 3, 1979.  p. 82.  Docket Reference  Number  II-I-50.*

21.  Albright, L.F.  High Pressure Processes For  Polymerizing Ethylene.
     Chemical  Engineering.  December 19, 1966.  p.  114.   Docket  Reference
     Number II-I-ll.*

22.  Reference 14.  pp. 92 and 96.

23.  Reference 20, pp. 80-82.

24.  Reference 20, p. 83.

25.  Paschke, E.  The Outlook for High Density  Polyethylene.
     Chemical  Engineering Progress.  January 1980.
     p. 74.  Docket Reference II-I-56.*

26.  Reference 18, pp. 236-238.

27.  Reference 18, pp. 239-243.

28.  Trip Report to DuPont's Sabine River Works Plant.   July 1,  1982.
     Docket Reference Number II-B-41.*

29.  Reference 7, p. 90.

30.  Reference 13, p. 347.

31.  Benning, Calvin J.  Plastic Foams:   The Physics  and  Chemistry of
     Product Performance and Process Technology.  Volume  1:  Chemistry
     and Physics of Foam Formation.  John Wiley and Sons.  New
     York.  1969.  pp. 4-5.  Docket Reference Number  II-I-121.*

32.  Rosen, Stephen L.  Fundamental Principles  of Polymeric Materials.
     John Wiley and Sons.  New York.  1982.   p. 184.   Docket Reference
     Number II-I-125.*

33.  Reference 32, p. 182.

34.  Reference 14, p. 341.

35.  Reference 32, pp. 182-183.

36.  Reference 14, p. 343.
                                3-73

-------
 37.   Reference 14,  p.  345.

 38.   Reference 31,  p.  5.

 39.   Reference 1, p. 123.

 40.   Moncrieff,  R.W.   Man-Made  Fibers.  6th Edition Newnes-Butterworths.
      1975.   p.  441.  Docket  Reference Number II-I-123.*

 41.   Trip Report to  Fiber Industries' Poly(ethylene terephthalate)
      plant.   September 1982.  Docket Reference Number II-B-53.*

 42.   Trip Report to  Fiber Industries' Poly(ethylene terephthalate) plant.
      Octooer 12, 1983.  Docket  Reference Number II-B-80.*

 43.   Letter.  J.W. Torrance, Allied Fibers and Plastics, to Sims
      Roy, U.S.  Environmental Protection Agency.  February 23, 1984.
      Docket  Reference  Number II-D-150.*

 44.   Letter,  R.K. Smith, Allied  Fibers and Plastics, to J.R. Farmer,
      U.S. Environmental Protection Agency.  April 15, 1982.  Docket
      Reference Number  II-D-53.*

 45.   Wetherold,  R.6.,  C.P. Provost, and C.D. Smith.  Assessment of
      Atmospheric Emissions from  Petroleum Refining.  Volume 3, Appendix B
      EPA-600/2-80-075c.  April 1980.  Docket Reference Number II-A-12.*

 46.   U.S. Environmental Protection Agency.  Background Information for
      Proposed Standards for VOC  Fugitive Emissions in Synthetic Organic
      Chemicals Manufacturing Industry.   Research Triangle Park, N.C.
      EPA Publication No. EPA-450/3-80-033a.  November 1980.  Docket
      Reference Number  II-A-16.*

 47.   U.S. Environmental Protection Agency.  VOC Fugitive Emissions in
      Petroleum Refining Industry - Background Information for Proposed
      Standards.  (Draft EIS) Research Triangle Park.  EPA Publication
      No. EPA-450/3-81-015a.   November 1982.  Docket Reference Number  II-A-33.*

 48.   U.S. Environmental Protection Agency.  Fugitive Emission Sources of
     Organic Compounds—Additional Information on Emissions, Emission
      Reductions, and Costs.   Research Triangle Park.  EPA Publication
     No. EPA-450/3-82-010.  April 1982.   Docket Reference Number II-A-32.*

49.  Dedeke, W.C.  Letter to Messrs.  J.C.  Berry and E.J.  Vincent,  U.S.
      Environmental  Protection Agency.   November 11, 1982.   Docket  Reference
     Number II-D-69.*
*References can be located in Docket Number A-82-19  at the  U.S.  Environmental
 Protection Agency Library,  Waterside Mall,  Washington,  D.C.
                                3-74

-------
                      4.0  EMISSION CONTROL TECHNIQUES

     Volatile organic compounds (VOC), used as solvents and key raw
materials in the manufacture of polymers and resins, are emitted to the
atmosphere from a variety of process equipment.  The emissions may be
considered as two large groups:  process emissions, which result from
fundamental operations of the process, and fugitive emissions, those
that escape directly to the atmosphere rather than through a flare or
exhaust system.  Process VOC emissions can be reduced either by installing
emission control devices or by reducing the VOC in the vent streams by a
process modification such as recovery of monomer or solvent.  Fugitive
VOC emissions can be reduced or essentially eliminated by increased
surveillance and maintenance or by installation of specified controls or
leakless equipment.  This chapter describes emission control techniques
that may be used to reduce these emissions from the polymers and resins
industry.  Control techniques for process emissions are discussed in
Section 4.1 and for fugitive emissions in Section 4.2.
4.1  CONTROL TECHNIQUES FOR PROCESS EMISSIONS
     Process emissions from the manufacture of polymers and resins are
diverse in both composition and flow.  Streams contain a wide range of
VOC concentrations, i.e., less than 1 percent to essentially 100 percent,
but most are of high concentration.  Some streams are continuous,  while
others are intermittent.  Process emissions also differ in temperature,
pressure, heating value, and miscibility.  These factors are extremely
important in the selection and design of VOC emission control  equipment.
     Due to this diversity, different control techniques may be appropriate
for different vent streams.  The control  techniques may be characterized
by two broad categories:  combustion techniques and recovery techniques.
Combustion techniques such as flares and incinerators are applicable to a
variety of VOC streams.  Recovery techniques such as condensation,
                                 4-1

-------
 absorption,  and adsorption,  are effective for some select vent streams.
 Economic incentives may encourage the use of either type of VOC control,
 since certain combustion configurations  may permit heat recovery,  and
 recovery techniques permit the conservation and reuse  of valuable  materials.
 The selection of a control system for a  particular application is  based
 primarily on considerations  of technical  feasibility and process economics.
      The most common control  techniques  form the basis for this chapter.
 Basic design considerations  for flares,  thermal  and catalytic  incinerators,
 industrial  boilers, condensers,  absorbers,  and adsorbers,  are  briefly
 described.   The conditions affecting the  VOC, removal efficiency of each
 type of device and its  applicability for  use in  the polymers and resi-ns
 industry are examined.   Emphasis, has been given  to flares,  thermal
 incinerators,  and condensers  because .of their wide applicability to a-
 variety of  VOC streams..  Combustion  techniques are-discussed in
 Subsection  4.1.1 and recovery  techniques  in Subsection 4.1.2.           •  ,
 4.1.1  Control  by Combustion Techniques                         ,
      The four  major combustion devices that ,are  pi? can be  used  to  control
 VOC  emissions  from the  polymers  and  resins  industry  are:   flares,  thermal
 or  catalytic, incinerators, and boilers.   Flares  are  the  most widely used
 control  devices  at  polyethylene  and  polypropylene  manufacturing plants.
 Incinerators and .boilers, are also  used, to  a  lesser  extent, to  control
 continuous vent  streams.  Although these  control devices are founded
 upon  basic combustion principles,  their operating  characteristics  are
 very  different.   While  flares  can  handle  both  continuous and intermittent
 streams, neither boilers  nor incinerators can  effectively handle large
 volume  intermittent  streams.   Subsection  4.1.1 discusses the general
 principles of combustion, and  then the design  and  operation, VOC destruction
 efficiency, and  applicability  of these four combustion devices  at polymers
 and  resins manufacturing  plants.
     Combustion  is  a rapid oxidation process, exothermic in nature,
which results in the destruction of VOC by converting  it to carbon
 dioxide and water.  Poor  or incomplete combustion  results in the production
 of other organic compounds including carbon monoxide.  The chemical
 reaction sequence which takes  place in the destruction of VOC by combustion
 is a complicated process.  It  involves a series of reactions that produce
                                 4-2

-------
free radicals, partial oxidation products, and final combustion products.
Several intermediate products may be created before the oxidation process
is completed.  However, most of the intermediate products have a very
short  life and, for engineering purposes, complete destruction of the
VOC is the principal concern.*
     Destruction efficiency is a function of temperature, turbulence,
and residence time.  Chemicals vary in the magnitudes of these parameters
that they require for complete combustion.  An effective combustion
technique must provide:^
     1.  Intimate mixing of combustible material  (VOC) and the oxidizer
(air),
     2.  Sufficient temperature to ignite the VOC/air mixture and complete
its combustion,
     3.  Required residence time for combustion to be completed,  and
     4.  Admission of sufficient air (more than the stoichiometric
amount) to oxidize the VOC completely.
     4.1.1.1  Flares.  Flaring is an open combustion process in which
the oxygen required for combustion is provided by the air around the
flame.  Good combustion in a flare is governed by flame temperature,
residence time of components in the combustion zone, turbulent mixing of
components to complete the oxidation reaction, and oxygen for free
radical formation.
     There are two types of flares:   ground level flares and elevated
flares.  Kalcevic (1980) presents a detailed discussion of different
types of flares,  flare design and operating considerations,  and a method
for estimating capital  and operating costs for flares.3  Elevated flares
are most common in the polymers and resins industry.  The basic elements
of an elevated flare system are shown in Figures  4-1 and 4-2.  Process
offgases are sent to the flare through  the collection header.  The
offgases entering the header can vary widely in volumetric flowrate,
moisture content,  VOC concentration, and heat value.  The knock-out  drum
removes water or hydrocarbon droplets that could  create problems  in  the
flare combustion zone.   Offgases are usually passed through  a water  seal
before going to the flare.   This prevents a possible flame flashback,
caused when the offgas  flow to the flare is too low and the  flame front
moves down into the stack.
                                 4-3

-------
Nozzles J£_
Flaie
Head t
(6) f
Bamei \
(5)
Flare
Slack
(4)
fia^ finllrrfinti !UiH»r __«_
3nd Tfuisfw 1 infl r PmraA 1
1 1 Gas ~l i
Eaissioa y „— rt-_, f^ fy
Gas 	 41 	 1
OiseflUaiaoeat 	 Water 	 '.'.-.' ' '(]
OH« ' seal *" '.'•'; .'. " \
(2) ^Y^ (3)

1\
z.
^



_J

y
[
f



I

^

*





^

?
b
•

t


\
X
^
•^

k



\
'
^H

J k



\






Butnets
(7)


Sieaa
— — Line.

	 Air Lias
	 Gas Line

                0(310
Figure 4-1.  Steam Assisted  Elevated Flare System
                      4-4

-------
    PILOT
    ASSEMBLY
 STEAM
 HEADER
     STEAM JETS



        FFUSER



     STEAM HEADER
                       INTERNAL
                       STEAM
                       INJECTOR
                       TUBES
                                       PILOT AND
                                       MIXER
STEAM
DISTRIBUTION
RING
 ENTER STEAM
JET
      TIP SHELL
         PLAN
                     CONTINUOUS
                     MUFFLER
                                         CENTER STEAM
                                         JET
                                     ELEVATION
  Figure  4-2.   Steam  Injection  Flare Tip
                      4-5

-------
     Purge gas (N2, C02, or natural gas) also helps to prevent flashback
in the flare stack caused by low offgas flow.  The total volumetric flow
to the flame must be carefully controlled to prevent low flow flashback
problems and to avoid a detached flame (a space between the stack and
flame with incomplete combustion) caused by an excessively high flowrate.
A gas barrier or a stack seal is sometimes used just below the flare
head to impede the flow of air into the flare gas network.
     The VOC stream enters at the base of the flame where it is heated
by already burning fuel and pilot burners at the flare tip.  Fuel flows
into the combustion zone where the exterior of the microscopic gas
pockets is oxidized.  The rate of reaction is limited by the mixing of
the fuel and oxygen from the air.  If the gas pocket has sufficient
oxygen and residence time in the flame zone it can be completely burned.
A diffusion flame receives its combustion oxygen by diffusion of air
into the flame from the surrounding atmosphere.  The high volume of fuel
flow in a flare requires more combustion air at a faster rate than
simple gas diffusion can supply, so flare designers add steam injection
nozzles to increase gas turbulence in the flame boundary zones, drawing
in more combustion air and improving combustion efficiency.  The steam
injection promotes smokeless flare operation by minimizing the cracking
reactions that form carbon.  Significant disadvantages of steam usage
are the increased noise and cost.  The steam requirement depends on the
composition of the gas flared, the steam velocity from the injection
nozzle, and the tip diameter.  Although some gases can be flared smokelessly
without any steam, typically 0.15 to 0.5 kg of steam per kg of hydrocarbon
in the flare gas is required.
     Steam injection is usually controlled manually with the operator
observing the flare (either directly or on a television monitor) and
adding steam as required to maintain smokeless operation.  Several flare
manufacturers offer devices which sense flare flame characteristics and
adjust the steam flowrate automatically to maintain smokeless operation.
     Some elevated flares use forced air instead of steam to provide the
combustion air and mixing required for smokeless operation.  These
flares consist of two coaxial flow channels.  The combustible gases flow
in the center channel and the combustion air (provided by a fan in the
                                 4-6

-------
bottom of the flare stack) flows in the annulus.  The principal advantage
of air assisted flares is that expensive steam is not required.  Air
assist is rarely used on large flares because air flow is difficult to
control when the gas flow is intermittent.  About 600 J/sec (0.8 hp) of
blower capacity is required for each 45 kg/hr (1UO Ib/hr) of gas flared
(Klett and Galeski, 1976).4
     Ground flares are usually enclosed and have multiple burner heads
that are staged to operate based on the quantity of gas released to the
flare.  The energy of the gas itself (because of the high nozzle pressure
drop) is usually adequate to provide the mixing necessary for smokeless
operation and air or steam assist is not required.  The fence or other
enclosure reduces noise and light from the flare and provides some wind
protection.
     Ground flares are less numerous and have less capacity than elevated
flares.  Typically they are used to burn gas "continuously" while steam
assisted elevated flares are used to dispose of large amounts of gas
released in emergencies (Payne, 1982).5
     4.1.1.1.1  Flare combustion efficiency.  The flammability limits of
the gases flared influence ignition stability and flame extinction
(gases must be within their flammability limits to burn).  When flammability
limits are narrow, the interior of the flame may have insufficient air
for the mixture to burn.  Outside the flame, so much air may be induced
that the flame is extinguished.  Fuels with wide limits of flammability
are therefore usually easier to burn (for instance, H2 and acetylene).
However,  in spite of wide flammability limits,  CO is difficult to burn
because it has a low heating value and slow combustion kinetics.
     The auto-ignition temperature of a fuel affects combustion because
gas mixtures must be at high enough temperature and at the proper mixture
strength  to burn.  A gas with a low auto-ignition temperature will
ignite and burn more easily than a gas with a high auto-ignition temperature.
Hydrogen  and acetylene have low auto-ignition temperatures while CO has
a high one.
     The heating value of the fuel  also affects the flame stability,
emissions, and structure,  A lower heating value fuel  produces a cooler
flame which  does not favor combustion kinetics  and also is more easily
                                 4-7

-------
extinguished.  The lower flame temperature will also reduce buoyant
forces, which reduces mixing (especially for large flares on the verge
of smoking).  For these reasons, VOC emissions from flares burning gases
with low heat content may be higher than those from flares which burn
high heat content gases. ,
     Some fuels, also, have chemical differences (slow combustion kinetics)
sufficient to affect the VOC emissions from flares.  For instance, CO is
difficult to ignite and burn, and so flares burning fuels with large
amounts of CO may have greater VOC emissions than flares burning pure
VOC.
     The density of the gas flared also affects the structure and stability
of the flame through the effect on buoyancy and mixing.  The velocity in
many flares is very low, and, therefore, most of the flame structure is
developed through buoyant forces on the burning gas.  Lighter gases thus
tend to burn better, all else being equal.  The density of the fuel also
affects the minimum purge gas required to prevent flashback and the
design of the burner tip.
     Poor mixing at the flare tip or poor flare maintenance can cause
smoking (particulate).  Fuels with high carbon-to-hydrogen ratios (greater
than 0.35) have a greater tendency to smoke and require better mixing if
they are to be burned smokelessly.
     The following review of flares and operating conditions summarizes
five studies of flare combustion efficiency.  Each study can be found in
complete form in the docket.
     Palmer (1972) experimented with a 1/2-inch ID flare head, the tip
of which was located 4 feet from the ground.6  Ethylene was flared at 15
to 76 m/sec (50 to 250 ft/sec) and 0.12-0.62 x 106 J/sec (0.4-2.1 x 106
Btu/hr) at the exit.  Helium was added to the ethylene as a tracer at 1
to 3 volume percent and the effect of steam injection was investigated
in some experiments.  Four sets of operating conditions were investigated;
destruction efficiency was measured as greater than 99.9 percent for
three sets and 97.8 percent for the fourth.  The author questioned the
validity of the 97.8 percent result due to possible sampling and analytical
errors.  He recommended further sampling and analytical techniques
development before conducting further flare evaluations.
                                 4-8

-------
     Siege!  (1980)  made the first comprehensive  study of a  commercial
flare system.7  He  studied burning of refinery gas  on a  commercial  flare
head manufactured by Flaregas Company.  The flare gases  used  consisted
primarily of hydrogen (45.4 to 69.3 percent by volume) and  light  paraffins
(methane to  butane).  Traces of I^S were also present in some runs.   The
flare was operated  with from 130 to 2,900 kilograms of fuel/hr (287  to
6,393 Ib/hr), and the maximum heat release rate was approximately 68.9  x  106
J/sec (235 x 106 Btu/hr).  Combustion efficiency and local  burnout was
determined for a total of 1,298 measurement points.  Combustion efficiency
was greater  than 99 percent for 1,294 points and greater than 98  percent
for all points except one, which had a 97 percent efficiency.  The
author attributed the 97 percent result to excessive steam addition.
     Lee and Whipple  (1981) studied a bench-scale propane flare.8  The
flare head was 2 inches in diameter with one 13/16-inch  center hole
surrounded by two rings of 16 1/8-inch holes, and two rings of 16 3/16-inch
holes.  This configuration had an open area of 57.1 percent.   The velocity
through the head was  approximately 1 m/sec (3 ft/sec) and the heating
rate was 0.09 x 106 J/sec  (0.3 x 106 Btu/hr).  The effects of steam and
crosswind were not  investigated  in this study.  Destruction efficiencies
were greater than 99  percent  for three of four tests.  A 97.8 percent
result was obtained in the only  test where the probe was located off the
centerline of the flame.   The author did not believe that this probe
location provided a valid  gas sample for analysis.
     Howes, et al.  (1981)  studied two commercial flare heads at John Zink's
flare  test facility.9  The  primary purpose of this test  (which was
sponsored by the EPA) was  to -develop a flare testing procedure.  The
commercial flare heads were  an LH air assisted head and  an LRGO  (Linear
Relief Gas Oxidizer)  head  manufactured by John Zink Company.  The LH
flare  burned 1,045  kg/hr  (2,300  Ib/hr) of commercial propane.  The  exit
gas  velocity based  on the  pipe diameter was 8.2 m/sec (27  ft/sec) and
the  firing rate was 12.9  x 106 J/sec  (44 x  106 Btu/hr).  The  LRGO flare
consisted of  three  burner  heads  1  meter  (3  feet) apart.  The  three-burner
combination  fired  1,909  kg/hr (4,200  Ibs/hr)  of  natural  gas.   This
corresponds  to  a firing  rate of  24.5  x 106  J/sec  (83.7  x 106  Btu/hr).
Steam  was not used  for either flare,  but  the  LH  flare head was in  some
                                  4-9

-------
trials assisted by a forced draft fan.  In four of five tests, combustion
efficiency was determined to be greater than 99 percent when sampling
height was sufficient to ensure that the combustion process was complete.
One test resulted in combustion efficiency as low as 92.6 percent when
the flare was operated under smoking conditions.
     An excellent detailed review of the above four studies was done by
Payne, et al. in January 1982,10 and a fifth study  OMcDaniel, et al.
(1982) o determined the influence on flare performance of mixing, heat
content, and gas flow velocity.H  A summary of these studies  is given
in Table 4-1.  Steam assisted and air assisted flares were tested at the
John Zink facility using the procedures developed by Howes.  The test
was sponsored by the Chemical Manufacturers Association (CMA)  with the
cooperation and support of EPA.  All of the tests were with an 80 percent
propylene, 20 percent propane mixture diluted as required with nitrogen
to give different Btu/scf values.  This was the first work which
determined flare efficiencies at a variety of "nonideal" conditions
where lower efficiencies had been predicted.  All previous tests were of
flares which burned gases that were very easily combustible and did not
tend to soot.  This was also the first test which used the sampling and
chemical analysis methods developed for the EPA by Howes.
     The steam assisted flare was tested with exit flow velocities up to
19 m/sec (62.5 ft/sec), with heat contents of 11-81 x 106 J/scm (294 to
2,183 Btu/scf) and with steam-to-gas (weight) ratios varying from zero
(no steam) to 6.86:1.  Flares without assist were tested down  to
7.2 x 106 J/scm (192 Btu/scf).   All of these tests, except for those
with very high steam-to-gas ratios, showed combustion efficiencies of
over 98 percent.  Flares with high steam-to-gas ratios (about  10 times
more steam than required for smokeless operation) had lower efficiencies
(69 to 82 percent) when combusting 81 x 106 J/scm (2',183 Btu/scf) gas.
     The air assisted flare was tested with flow velocities up to
66 m/sec (218 ft/sec) and with  Btu contents of 3.1-81 x 106 J/scm (83 to
2,183 Btu/scf).  Tests at 10.5  x 106 J/scm (282 Btu/scf) and above gave
over 98 percent efficiency.  Tests at 6.3 x 106 J/scm (168 Btu/scf)  gave
55 percent efficiency.
     These results have been recently supplemented by tests conducted
under EPA's flare testing program.  A summary of these results is found
                                 4-10

-------
























GO
tit
1 ' '
!— 1

t__J
GO
O
i — i
CO
GO
1— H
s^.
UJ

UJ
<=c
u_
1
•=1-
OJ
ns
i —






















0)
u
c
11
s-
~
-g, 5
3 CO
O
s- to
.£= O




in
to
CJ3
•o
0)

s-
to
u.









^
0
t/1
C
rt






O
(O
Dl
V)

O (U (/I
,— CL 0
53 ^-*
(B -r-
OJ S. t- -0
4-» 03 U
OJ 4J S- =
to us § i»*.
5-5 O1^-*
_g
S-
c ^~
o s-
>
s- tn 4-»
O Q t/1
S-
C • OJ
o a >

= Ci-=3
r- «
CO
en

(U — .

a. cn
•T— <— *
3 •
o9 OJ
ai
0) GJ





a.
•f- CO


&. i
J: S
o
«3- CO
^3- CM O


V)
03
C3 OJ
C
OJ i— OJ
S S- ">,
a. 3 a.
o +-> o
S- (tj S-
o. z Q.


Q_

•^ "7T7 *^^ >—^'
tfl • 4-1
c "- = « c. cn
•r- vi *^- ~o •»-(/»
in ca wi tn w
OJ (U = OJ (13
a i- o«* o
(Q (O O 
V)
•^

c: v)
VI C=
OJ (Q
O fO
i- -i-
§5r
0 —
























4-n

-------
in Docket Reference Number II-B-94.  One result of these more recent
tests is the determination of a broader range of exit velocity and heat
content parameters under which steam-assisted and nonassisted flares
can achieve 98 percent combustion efficiency.
     After consideration of the results of these tests,  EPA has concluded
that 98 percent combustion efficiency can be achieved by steam assisted
flares if these flares are operated with combustion gas  heat contents
and exit flow velocities within ranges determined by the tests.   Steam
flares obtain 98 percent combustion efficiency combusting gases with
heat contents over 11.2 x 106 J/scm (300 Btu/scf) at velocities of less
than 18.3 m/sec (60 ft/sec).  Steam flares are not normally operated at
the very high steam-to-gas ratios that resulted in low efficiency in
some tests because steam is expensive and operators make every effort  to
keep steam consumption low.  Flares with high steam rates are also noisy
and may be a neighborhood nuisance.  Nonassisted pipe flares obtain
98 percent efficiency with heat contents over 200 Btu/scf at velocities
of less than 18.3 m/sec.  In addition, steam-assisted and nonassisted
flares can obtain 98 percent combustion efficiencies with exit velocities
of up to 120 m/sec (400 ft/sec) provided the heat content of the gas
stream is at least 37.3 MJ/scm (1,000 Btu/scf).  Air assisted flares
obtain 98 percent efficiency with heat contents over 11.2 x 106 J/scm
and at velocities not exceeding that determined by the following formula:

           v(ft/sec) = 28.75 + 0.867 HC
         where:  v   = maximum gas velocity in ft/sec, standard conditions.
                HC   = heat content of the combusted gas in Btu/scf.

     4.1.1.1.2  Applicability.  A typical  polymer plant  produces several
hundred million pounds of product per year.  Because of  this huge throughput,
the VOC emissions that result from frequent process upsets are also
large.  Flares are used mainly to minimize the safety risk caused by
emergency blowdowns where large volumes of gases with variable composi-
tion must be released from the plant almost instantaneously.   Flares are
ideal  for this service and their reliability, as measured by absence of
explosions and plant fires, has been demonstrated repeatedly.   Flares
                                 4-12

-------
 also effectively eliminate the hazard of process streams which, during
 startup or shutdown, would otherwise vent to the atmosphere and could
 also create an explosion or toxic hazard.  Finally, flares are also used
 to burn co-products or by-products of a process that has too little
 value to reclaim, and thus would otherwise be a continuous VOC emission
 during normal  operation of the unit.  This practice, once the norm, has
 abated considerably during the past decade as the value of VOC stream
 components has dramatically increased.
      4.1.1.2 -Thermal  Incinerators.  The design and operation of thermal
 incinerators are influenced by operating temperature,  residence time,
 desired VOC destruction efficiency, offgas characteristics,  and combustion
 air.   Operating temperatures may  typically be between  650°C  (1,200°F)
 and 980°C  (1,800°F)  with a residence time of 0.3  to 1.0 second.12   The
 temperature theoretically required to achieve complete oxidation  depends
 on the nature  of the chemical  involved  and can be determined  from kinetic
 rate  studies.13  The design  of the combustion chamber  should  maximize
 the mixing  of  the VOC  stream,  combustion  air,  and hot  combustion products
 from  the burner.  This  helps  ensure that  the  VOC  contacts  sufficient
 oxygen  while at combustion temperature, for  maximum combustion  efficiency.
      The heating  value  and water  content  of the waste  gas  feed  and the
 excess  combustion air  delivered to  the  incinerator  also affect  incinerator
 design  and  operation.   Heating  value  is a  measure of the heat produced
 by  the  combustion of the VOC  in the waste  gas.  Gases with a heating
 value  less  than  1,860 kJ/scm  (50 Btu/scf) will not  burn and require
 auxiliary fuel  to maintain combustion.  Auxiliary fuel requirements can
 be  reduced  and  sometimes even eliminated by transferring heat from the
 exhaust gas to the inlet gas.  Offgases with a heating value between
 1,860 kJ/scm and 3,720  kJ/scm  (100 Btu/scf) can support combustion but
 require some auxiliary fuel to ensure flame stability, i.e., avoid a
 flameout.  Theoretically, offgases with a heating value above 3,720 kJ/scm
 possess enough heat content to not require auxiliary fuel  (although
 practical experience has shown that 5,580 kJ/scm  (150 Btu/scf) and above
may be necessary,14 and these gases may be used as a fuel gas or boiler
feed gas.15  A thermal incinerator handling offgas streams with varying
heating values and moisture content requires periodic adjustment to
                                 4-13

-------
maintain the proper chamber temperatures and operating efficiency.
Increases in heat content reduce auxiliary fuel requirements, whereas
increases in water content can substantially increase fuel  requirements.
     Incinerators are always operated with excess air to ensure a sufficient
supply of oxygen.  The amount of excess air used varies with the fuel
and burner type but should be kept as low as possible.  Using too much
excess air wastes fuel because this air must be raised to the combustion
temperature but does not contribute any heat by participating in the
oxidation reaction.  Large amounts of excess air also increase the flue
gas volume and may cause an operator to invest in a larger system than
required.
     A thermal incinerator usually contains a  refractory-lined chamber
(Which may vary in cross-sectional size along  its length) containing a
burner at one end.  Because of the risk to the refractory,, incinerators
are neither brought quickly up to nor cooled down quickly from operating
temperatures.  They require a fairly constant  fuel input to maintain
combustion temperature.  A diagram of a thermal incinerator using discrete
burners is shown in Figure 4-3.   (Numbers in parentheses following the
mention of equipment parts or streams denote the numbered items on the
referenced figures.)  Discrete dual fuel burners (1) and inlets for the
offgas (2) and combustion air (3) are arranged in a premixing chamber
(4) to thoroughly mix the hot products from the burners with the offgas
air streams.  The mixture of hot  reacting gases then passes into the
main combustion chamber  (5).  This section is  sized to allow the mixture
enough time at the elevated temperature for the oxidation reaction to be
completed (residence times of 0.3 to 1 second  are common).  Energy can
then be recovered from the hot flue gases with the installation of a
heat recovery section (6).  Preheating of combustion air or the process
waste offgas fed to the  incinerator by the incinerator exhaust gases
will reduce auxiliary fuel usage.  In some instances, the incinerator
exhaust gas may be used  in a waste heat boiler to generate steam.
Insurance regulations require that if the process waste offgas is preheated,
the VOC concentration must be maintained below 25 percent of the lower
explosive limit  (LEL) to minimize explosive hazards.16
                                 4-14

-------
      Thermal  incinerators designed specifically for VOC incineration
 with natural  gas as the auxiliary fuel  may use a grid-type (distributed)
 gas burner similar to that shown in Figure 4-4.  The tiny gas flame jets
 (1) on the grid surface (2) ignite the  vapors as they pass through  the
 grid.   The grid acts as a baffle for mixing the gases entering the
 chamber (3).   This arrangement ensures  burning of all vapors  using  less
 fuel and a shorter burning length in the duct than conventional  forward
 flame  burners.   Overall,  this  system makes possible a shorter reaction
 chamber while maintaining high efficiency.17
     Thermal  incinerators used to burn  halogenated VOC's  often use
 additional  equipment to remove the corrosive combustion products.   The
 flue gases  are  quenched to lower their  temperature and routed through
 absorption  equipment such as spray towers  or liquid jet scrubbers to
 remove the  corrosive gases from the exhaust.18
     Packaged,  single  unit thermal  incinerators  are available in many
 sizes  to  control  streams  with  flowrates  from a  few hundred scfm up  to
 about  50,000  scfm.   A  typical  thermal incinerator  built to handle a VOC
 waste  stream  of  850  scm/min (30,000 scfm)  at a temperature of 870°C
 (1,600°F) with  0.75  second residence time  would  probably  be a refractory-
 lined  cylinder.   With  the typical  ratio  of  flue  gas  to waste  gas of
 about  2.2, the  chamber  volume  necessary  to  provide  for  0.75 second
 residence time  at  870°C (1,600°F) would  be  about 100  m3  (3,500 ft3).   If
 the  ratio of the  chamber  length to the diameter  is  2, and  if  a 30.5 cm
 (1 ft) wall thickness  is  allowed, the thermal incinerator would measure
 8.3  m  (27 ft) long by 4.6  m (15 ft) wide, exclusive of heat exchangers
 and  exhaust equipment.
     4.1.1.2.1  VOC destruction efficiency.  The destruction  efficiency
 of an incinerator can be affected by variations in chamber temperature,
 residence time, inlet concentration, compound type, and flow  regime
 (mixing).  Of these, chamber temperature, residence time, and flow
 regime are the most important.
     When the temperature  exceeds 700°C   (1,290°F), the oxidation reaction
 rate is much faster than the rate at which mixing can take place,  so VOC
 destruction becomes more dependent upon  the fluid mechanics within the
combustion chamber.19  Variations in inlet concentration also  affect the
                                 4-15

-------
    Waste Gas
     Auxiliary
   Fuei Burner
     (discrete)

        (1)


Combustion Air
                                    Stacx
                       Mixing
                      Section
                       (4)
Combustion
  Section
   (5)
Optional
  Heat
Recovery
   (6)
             Figure 4-3.   Discrete Burner Thermal  Incinerator
                              (2)            (1)
                           Burner Plate -i   Fl arae Jets-7
                                 Stack
                                                                      n   Fan
                                                                    Optional
                                                                     Heat
                                                                    Recovery
                                                                     (4)
                           (natural gas)
                          Auxiliary Fuel

           Figure 4-4.   Distributed  Burner Thermal  Incinerator
                                     4-16

-------
VOC destruction efficiency achievable;  kinetics calculations describing
the combustion reaction mechanisms indicate much slower reaction rates
at very low compound concentrations.  Therefore, at low VOC concen-
tration, a greater residence time is required to achieve a high combustion
efficiency.
     Test results show that a VOC control  efficiency of 98 percent can
be achieved consistently for many VOC compounds by well-designed units
and can be met under a variety of operating conditions:20,21 combustion
chamber temperatures ranging from 700 to 1,300°C (1,300 to 2,370°F) and
residence times of 0.5 to 1.5 seconds.   The test results covered the
following VOC compounds:  C] to C$ alkanes and olefins, aromatics (benzene,
toluene, and xylene), oxygenated compounds (methyl ethyl ketone and
isopropanol), chlorinated organics (vinyl  chloride), and nitrogen-
containing species (acrylonitrile and ethylamines).  At chamber temperatures
below 760°C (1,400°F), a wide range of efficiencies were reported for
several VOC compounds.  This information,  used in conjunction with
kinetics calculations, indicates overall that the minimum combustion
chamber parameters for ensuring at least a 98 percent VOC destruction
efficiency are a combustion temperature of 870°C (1,600°F), and a resi-
dence time at combustion temperature of 0.75 second.  A thermal incinerator
designed to produce these conditions in the combustion chamber should be
capable of high destruction efficiency for almost any VOC even at low
inlet concentrations.
     Based on the studies of thermal incinerator efficiency, auxiliary
fuel use, and costs, EPA has concluded that 98 percent VOC destruction,
or a 20 parts per million by volume (ppmv) compound exit concentration
(whichever is less stringent), is the highest reasonable control level
achievable by all new incinerators considering current technology.22
This estimate is predicated on thermal  Incinerators operated at 870°C
(1,600°F) with a 0.75 second residence time.
     4.1.1.2.2  Applicability.  Thermal incinerators can be used to
control a wide variety of continuous waste gas streams (one has been
observed in a polypropylene plant23).  They can be used to destroy VOC
in streams with any concentration and type of VOC.  Although they
accommodate minor fluctuations in flow, incinerators are not well suited
to streams with intermittent flow because of the large auxiliary fuel
                                 4-17

-------
requirements during periods when there is no fuel contribution from the
waste gas, yet the chamber temperature must be maintained to protect the
incinerator lining.
     For extremely dilute streams, a catalytic incinerator might be a
favorable choice over a thermal incinerator if supplemental fuel require-
ments are of principal concern.  However, most waste gas streams in this
industry contain enough heating value to support a flame by itself on a
properly designed flame burner.  Such streams can be considered for use
as fuel gas or boiler feed gas, from which the recovery of energy may
more than compensate for a thermal incinerator's capital costs.
     4.1.1.3  Catalytic Incinerators.  The control principles and equip-
ment used in catalytic incineration are similar to those employed in
conventional thermal incineration.  The VOC-containing waste gas stream
is heated to an appropriate reaction temperature and then oxidation is
carried out at active sites on the surface of a solid catalyst.  The
catalyst increases the rate of oxidation, allowing the reaction to occur
at a lower temperature than in thermal incineration.  This technique may
offer advantages over thermal incineration in auxiliary fuel savings
where low VOC content makes large fuel usage necessary.  Catalytic
incinerators also may produce less NOX because of lower combustion
temperatures and smaller excess air requirements.
     Combustion catalysts are made by depositing platinum or platinum
alloys, copper oxide, chromium, or cobalt on an inert substrate, which
is suitably shaped to fit the mechanical  design of the incinerator.   The
operating temperature of the catalyst is  usually from 315°C (600°F)  to
650°C (1,200°F).  Combustion may not occur below 315°C and temperatures
higher than 650°C may shorten the catalyst life or even evaporate catalyst
from the support substrate.24  Accumulation of particulate matter,
condensed VOC's, or polymerized hydrocarbons on the catalyst can block
the active sites and reduce its effectiveness.  Catalysts can also  be
contaminated and deactivated by compounds containing sulphur, bismuth,
phosphorous, arsenic, antimony, mercury,  lead, zinc, tin, or halogens.
If the catalyst is so "poisoned," VOC's will pass through unreacted  or
only partially oxidized.   Catalytic incinerators can operate efficiently
treating offgas streams with VOC concentrations below the lower explosive
                                 4-18

-------
limit.  This is a distinct advantage over thermal incinerators which
would in this situation require auxiliary fuel.
     A schematic of a catalytic incinerator unit is shown in Figure 4-5.
During operation, the waste gases (1) first enter the mixing chamber
(also called the preheat zone) (3) where they are heated by contact with
the hot combustion products of a burner (2).  The mixing chamber
temperature may vary as a function of the composition and type of
contaminants to be oxidized, but will generally operate in the range of
343°C (650°F) to 593°C (1,100°F).25  The heated mixture then passes
through the catalyst bed (4) where oxygen and VOC's diffuse to the
catalyst and are adsorbed on its surface.  The oxidation reaction takes
place at these "active sites."  Reaction products desorb from the active
sites and diffuse back into the waste gas.  As with the exhaust gases
from thermal incinerators, the products of combustion leaving the bed
may be used in a waste heat recovery device (5) before being exhausted
to the atmosphere.
     4.1.1.3.1  VOC destruction efficiency.  The destruction efficiency
of catalytic incinerators is a function of many variables, including
type of catalyst, its surface area, volume, and pore size distribution,
gas composition, uniformity of flow through the catalyst bed, oxygen
concentration, and temperature in the unit.26,27
     The efficiency of a catalytic incinerator will deteriorate over
time, necessitating periodic replacement of the catalyst.  The replace-
ment time varies widely, depending on the service of the unit, from less
than 1 year up to 10 years,12 with an average life between 3 and 5 years.28
     A 1980 study by Engelhard Industries for the EPA involved testing
of both pilot and full-scale catalytic incineration systems.  The full-
scale unit installed on a formaldehyde plant achieved control efficiencies
ranging from 97.9 to 98.5 percent.  These efficiencies represent overall
control levels for carbon monoxide, methanol, dimethyl ether, and
formaldehyde.  Measurements indicated the ability of the system to
control at this level consistently over a 1-year period.  No trend in
the data points gave indication of a maximum catalyst life.29
     4.1.1.3.2  Applicability.  A catalytic incinerator is best applied
to a continuous stream that is (1) low in VOC  (higher VOC concentrations
                                 4-19

-------
          II! 5
                        s_
                        S
                        to
                        
iZ
4-20

-------
 lead to higher catalyst temperatures, which can seriously damage the
 catalyst activity and possibly create fire hazards) and (2) free from
 solid particles and catalyst "poisons."  A catalytic incinerator in many
 situations may be favored over a thermal incinerator because it can
 destroy the VOC at a lower temperature and, therefore, use less fuel.
 However, since most of the streams involved in the polymers and resins
 industry are high enough in heating value to self-combust without using
 auxiliary fuel, virtually no advantage is achieved by using a catalytic
 unit and their applicability in this industry is very limited.
     4.1.1.4  Industrial Boilers.  Fireboxes of boilers (and process heaters)
 can be used, under proper conditions, to incinerate waste streams that
 contain VOC's.  Combustible contaminants, including smoke, organic
 vapors, and gases can be converted essentially to carbon dioxide and
 water in boiler fireboxes.  As the primary purpose of the boiler is to
 generate steam, all aspects of operation must be thoroughly evaluated
 before this method of air pollution control can be used.  Any breakdown
 in the boiler can result in expensive process downtime.  Consequently,
 the risk of shutdown should be kept small and' only streams that do not
 threaten boiler performance should be introduced.
     For the satisfactory use of boilers as a control  device, there are
 several prerequisites.   Generally, the burner must be modified, the
 boiler must operate continuously and concurrently with the pollution
 source, the contaminants must be completely combustible, and the products
 of combustion must not  corrode the materials used to construct the
 boiler.  Corrosive VOC  compounds can be combusted in a boiler, but
 special attention must  be given to operate above the dew point of the
 flue gases.  If these gases are allowed to condense, severe corrosion
 problems will  occur.   Further,  the volumetric flowrate of low VOC concen-
tration emission streams must be taken into consideration because they
 can reduce thermal efficiencies in the same way as excess combustion air
 does.   The pressure drop caused by additional  products of combustion
should not exceed the draft provided by boiler auxiliaries.   Boiler
 life,  efficiency,  and capacity  can be affected by the presence of con-
taminants  in the VOC  emission streams.   Halogens,  for example, would be
devastating to the life of boiler tubes.  Finally, a personnel safety
                                 4-21

-------
hazard may occur if coal-fired boilers that are not pulverized coal-
fired are used to destroy organic waste.  Any interruption in the air
supply to these types of boilers would release into the boiler house
combustion vapors and any hazardous or toxic substances that may have
been injected.30  Great care, therefore, must be exercised in selecting
this mode of pollution control.
     The large majority of industrial boilers are of water tube design.
Water, circulated through the tubes, absorbs the heat of combustion.
Drums store the superheated water from which steam is directed to external
heat exchangers for use as process steam.
     Both forced and natural draft burners, designed to thoroughly mix
the incoming fuel and combustion air, may be used.  After ignition, the
mixture of hot reacting gases passes through the furnace section that is
sized to allow the oxidation reaction to reach completion and to minimize
abrasion on the banks of the water tubes.  Energy transfer from the hot
flue  gases to form steam can attain  greater than 85 percent  efficiency.
Additional energy can be recovered from the hot exhaust gases by installation
of a  gas-gas heat exchanger to  preheat combustion air.
     Boilers designed specifically for use as a VOC control  device
typically use discrete or  vortex  burners, depending on the heating  value
of the vent stream.  A high  intensity or vortex burner can be effective
for  vent streams with low  heating  values (i.e., streams where a
conventional burner  may not  be  applicable).  Effective combustion of
low  heating value streams  is accomplished in a high intensity burner
by passing the  combustion  air  through a  series of  spin vanes to  generate
a strong  vortex.
      4.1.1.4.1   VOC  destruction efficiency.  Furnace  residence  time and
temperature profiles vary  for  industrial boilers  depending on the
furnace  and burner  configuration,  fuel  type, heat  input,  and excess air
level.31  A mathematical model  has been  developed  which estimates the
furnace  residence time  and temperature  profiles for a variety of
industrial  boilers.32   This  model  predicts  mean furnace residence times
of  from  0.25 to 0.83 seconds for natural  gas-fired water  tube  boilers
in  the size  range  from  15  to 150 x 106  Btu/hr.  Furnace exit temperatures
for  this  range  of  boiler sizes are at or above  1,200°C  (2,200°F) with  peak
                                     4-22

-------
furnace temperatures occurring in excess of 1,540°C (2,810°F).  Residence
times for oil-fired boilers and natural gas-fired boilers are similar.
A boiler (or process heater) furnace can be compared to an incinerator
where the average furnace temperature and residence time determine the
combustion efficiency.  However, when a vent gas is injected as a fuel
into the flame zone of a boiler (or process heater), the required
residence time is reduced due to the relatively high flame zone
temperature.  The following test data,  which document the destruction
efficiencies for industrial boilers and process heaters, are based on
injecting the wastes identified into the flame zone of each combustion
control device.
     An EPA sponsored test was conducted in an effort to determine the
destruction efficiency of an industrial boiler for polychlorinated
biphenyls (PCB's).33  The results of this test indicated that the PCB
destruction efficiency of an oil-fired industrial boiler firing PCB-
spiked oil was greater than 99.9 percent.  This efficiency was determined
based on the PCB content measured by a gas chromatagraph in the fuel
feed and flue gas.
     Firebox temperatures for process heaters show relatively wide
variations depending on the application.  Tests were conducted by EPA to
determine the benzene destruction efficiency of five process heaters
firing a benzene offgas and natural gas nrixture.34»35,36  The units
tested are representative of process heaters with low temperature
fireboxes (reboilers) and medium temperature fireboxes (superheaters).
Sampling problems occurred while testing one of these heaters and, as a
result, the data for that test may not be reliable and are not presented.~
The reboiler and superheater units tested showed greater than a
98 percent overall destruction efficiency for C^ to Cg hydrocarbons,
Additional tests conducted on a second superheater and a hot oil heater
showed that greater than 99 percent overall destruction of GI to Cg
hydrocarbons occurred for both units.3^  These efficiencies were
determined based on the benzene content measured by a gas chromatagraph
in the fuel feed and flue gas.
     4.1.1.4.2  Applicability.  Use of a boiler for VOC emission control
in the polymers and resins industry is not common.  Despite the potential
problems, boilers are being used in at least two polypropylene plants3^
                                   35
4-23

-------
and a high-density polyethylene plant.38  yne polypropylene plants
supplement boiler fuel with waste gas that otherwise would be flared.
The high density polyethylene plant sends the dehydrator regeneration
gas (a mixture of natural gas and nitrogen) and a degassing stream from
the recycle diluent step (mostly ethylene) to steam-generating boilers
as a fuel.
     A boiler would be used as a control device only if the process
generated its own steam or the fuel value of the waste gas was sufficient
to make the process a net exporter of steam.  Whenever either condition
exists, installation of a boiler is an excellent control measure that
provides greater than 98 percent VOC destruction and very efficient
recovery of the heat of combustion of the waste gas.
4.1.2  Control by Recovery Techniques
     The three major recovery devices are condensers, adsorbers, and
absorbers.  These devices permit many organic materials to be recovered
and, in some cases, reused in the process.
     Condensers are widely used for recovering organics from both
continuous and intermittent rich by-product streams in polystyrene
manufacturing processes.  The VOC is mainly styrene which is easily
condensed because of its relatively high condensation temperature.  The
ease of styrene recovery and the ability of a condenser to handle an
intermittent stream makes it a desirable control technology for all
process VOC emissions in the polystyrene industry.  Another application
of condensers occurs in polyethylene manufacturing processes.  In the
DMT process, emissions from the methanol recovery section are minimized
through the use of condensers.  In both the DMT and TPA polyethylene
processes, ethylene glycol (EG) is often recovered using an EG spray
condenser, or with a distillation column incorporating a reflux condenser.
Condensers may also be used in series with other air pollution control
systems.  A condenser located upstream of an incinerator, adsorber, or
absorber will reduce the VOC load entering the downstream control device.
The downstream device will abate most of the VOC that passes through the
condenser.
     Adsorbers are used on gas streams which contain relatively low VOC
concentrations.  Concentrations are usually well below the lower explosive
limit  in order to guard against overheating of the adsorbent bed.
Adsorbers are often neither suitable nor the most efficient means of
                                 4-24

-------
control  for the higher VOC concentration streams  characteristic  of  the
polymers and resins industry.
     Absorbers, which use low  volatility liquids  as  absorbents,  are
another control option.  Their use is generally limited to applications
in which the spent absorbent can be used directly in a process,  since
desorption of the VOC from the absorbent is often prohibitively  expensive.
     Recovery techniques either condense the organic or contact  the
VOC-containing gas stream with an appropriate liquid or solid.   Gases
containing only one or two organic gases are easier to process  by recovery
techniques than multi-component mixtures.  The presence of inert or
immiscible components in the waste gas mixture complicates recovery
techniques.
     4.1.2.1  Condensers.  Condensation devices transfer thermal energy
from a hot vapor to a cooling medium, causing the vapor to condense.
Condenser design thus typically requires knowledge of both heat and mass
transfer processes.  Heat may be transferred by any combination of three
modes:  conduction, convection, or radiation.
     The design of a condenser  is significantly affected by the number
and nature of  components  present  in the  vapor  stream.  The entering
gases may consist  of a single condensable  component or any number  of
gaseous components which  may or may not  all  be condensable or miscible
with one another.  Example  gas  streams  found in the polystyrene industry
may consist  of a  single  condensable component  (styrene);  a mixture of
condensable  and noncondensable  components  (styrene  and air); a  mixture
of condensable, but  immiscible, components (styrene and steam); or a
mixture of  condensable,  but immiscible,  components  with a noncondensable
component  (styrene,  steam,  and  air).
     Condensers are  designed  and  sized  using the  principles  of
thermodynamics.   At  a  fixed pressure, a pure component will  condense
 isothermally at the  saturation  or equilibrium temperature, yielding a
 pure liquid condensate.   A vapor mixture,  however,  does  not  have  a
 single condensate temperature.   As  the  temperature  drops, condensation
 progresses,  and the  composition,  temperature,  enthalpy,  and  flowrate of
 both the  remaining vapor and  the condensate will  change.   These changes
 can be calculated from thermodynamics data, if it is  assumed that the
 vapor and liquid condensate are in equilibrium.   Variations  in  composition
                                   4-25

-------
 and temperature will affect most of the physical and transport properties
 which must be used in condenser design calculations.  Mhen these properties
 change, the calculations governing the heat transfer process are adjusted
 to accommodate these changes.
      In a two-component vapor stream with one noncondensable component,
 condensation occurs when the partial pressure of the condensable component
 is equal  to the component's vapor pressure.  To separate the condensate
 from the  gas at fixed pressure, the temperature of the vapor mixture
 must be reduced.  The liquid will begin to appear when the vapor pressure
 of the condensable component becomes equal to its partial  pressure,  the
 "dew point."  Condensation continues as the temperature is further
 reduced.   The presence of a noncondensable component interferes  with the
 condensation process, because a layer of noncondensable on the condensate
 acts as a heat transfer barrier.
      Two  types of condensers are employed:   contact and surface.   Contact,
 or direct,  condensers cause the hot gas to mingle intimately  with  the
 cooling medium.   Contact condensers usually operate by  spraying  a  cool
 liquid directly  into  the gas stream.   Contact  condensers also may  behave
 as scrubbers  since they sometimes collect  noncondensable vapors which
 are immiscible with the coolant.   The  direct  contact  between  the vapor
 and the coolant  limits  the application of  contact  condensers  since the
 spent  coolant  can  present  a secondary  emission  source or a wastewater
 treatment problem,39  unless it  is  economically  feasible to separate  the
 two in  a  subsequent process.
     Surface,  or indirect,  condensers  are  usually  common shel1-and-tube
 heat exchangers.   The coolant usually  flows through the tubes and the
 vapor  condenses  on the  outside  of the  tubes.   In  some cases,  however, it
 may be  preferable  to condense the vapor inside the tubes.  The condensate
 forms  a film on  the cool tube and drains to storage.40-  The shel1-and-tube
 condenser is the optimum configuration  from the standpoint of mechanical
 integrity, range of allowable design pressures and temperatures, and
 versatility in type of service.  Shell-and-tube condensers may be designed
 to  safely handle pressures  ranging from full vacuum to approximately
41.5 MPa  (6,000 psig), and  for temperatures in the cryogenic range  up to
approximately 1,100°C (2,000°F).41  Surface condensers usually require
more auxiliary equipment for operation  (such as a cooling tower or  a
                                 4-26

-------
refrigeration system) but offer the advantage of recovering valuable VOC
without contaminating the coolant, thereby minimizing waste disposal
problems.  The successively more volatile material returned from the
condenser to the distillation column is termed "reflux," or overhead
product.  The heavier compounds removed at the bottom are often called
column "bottoms."42
     The major pieces of equipment used in a typical refrigerated surface
condenser system are shown in Figure 4-6.43  Refrigeration is often
required to reduce the gas phase temperature sufficiently to achieve low
outlet VOC concentrations.  This type of system includes dehumidification
equipment (1), a shell-and-tube heat exchanger (2), a refrigeration
unit  (3), recovery tank  (4), and operating pumps  (5).  Heat transfer
within a shell-and-tube  condenser occurs through  several material layers,
including the condensate film, combined dirt and  scale, the tube wall,
and the coolant film.  The choice of coolant used depends on the saturation
temperature  of the VOC stream.  Chilled water can be used to cool down
to 4°C  (40°F), brines to -34°C  (-30°F), and  chlorofluorocarbons below
-34°C  (-30°F).44  Temperatures  as  low  as  -62°C  (~80°F) may be  necessary
to condense  some VOC streams.39
      4.1.2.1.1  Condenser  control  efficiency.  VOC removal efficiency of
a condenser  is  dependent upon  the  composition of  the stream.   Single
component  streams with  a relatively  high  boiling  point will easily
condense,  resulting  in  essentially  100 percent  control efficiency.
Thus,  very  high  efficiencies would be  expected  for condensers  controlling
 such  streams in  the  polystyrene industry.   Ethylene glycol spray  condensers
 in  PET polyester production  reduce EG  emissions  to the  atmosphere from
 9.5  to 0.21  kg/Mg of product,  or 97.8  percent (Tables  3-14 and 3-15).   A
 less  condensable component in the stream, however, will  reduce the
 control efficiency because of the lower temperatures required for higher
 percentage removal.   Water-cooled condensers sometimes  cannot achieve a
 sufficiently low temperature to ensure high control efficiency.  Better
 control, of course, is  possible by use of a chilled coolant or even a
 refrigerated condenser  at an increased cost.  Outlet concentrations for
 low boiling organics may be above 10,000 ppmv to 20,000 ppmv.45
      4.1.2.1.2  Applicability of condensers.  Water-cooled condensers
 are effective in reducing potential emissions of  high boiling, easily
                                  4-27

-------
                                         C
                                         o
                                         •I—
                                         •u
                                         (O
                                         (/>
                                        • e
                                         O)
                                         T3
                                        IO
                                         I
                                         (U

                                         3
                                         o>
4-28

-------
condensable organics,  and find broad application in the polystyrene
manufacturing segment.  Surface condensers are used to recover styrene
from polystyrene vents and methanol and ethylene glycol from PET polyester
esterifier vents.  Spray condensers are highly efficient at recovering
EG from PET polyester esterifier and reactor vents.  Condensers cannot
be used to condense low boiling organics such as ethylene or propylene
in streams containing large quantities of inert gases such as nitrogen.
Refrigerated condensers may be a viable option unless the stream contains
water or heavy organics which would freeze and foul the condenser.
     4.1.2.2  Adsorbers.  Vapor-phase adsorption utilizes the ability of
certain solids to preferentially adsorb and thereby concentrate certain
components from a gaseous mixture onto their surfaces.  The gas phase
(adsorbate) is pumped through a packed bed of the solid phase (adsorbent)
where selective components are captured on its surface by physical
adsorption.  The organic molecules are retained at the surface of the
adsorbent  by means of intermolecular or Van-der-Waals  forces.  The
adsorbed organics can be readily removed and the adsorbent regenerated.
     The most common  industrial vapor-phase adsorption systems use beds
of activated carbon.  Carbons made from a variety of  natural materials
(wood, coal, nut shells, etc.) are marketed for their  special adsorbent
properties.  The multiple bed system maintains at least one bed online
while another is being  regenerated.  Most systems direct the vapor
stream downward through  a fixed carbon bed.  Granular carbon is usually
favored because it is not easily entrained in the exhaust  stream.
     Figure 4-7 is a  schematic of  a typical fixed  bed,  regenerative
carbon adsorption system.  The process offgases are  filtered and  cooled  (1)
to minimize  bed  contamination  and  maximize adsorption efficiency.  The
offgas is  directed through the porous  activated carbon bed (2) where
adsorption of the organics progresses  until the activated  carbon  bed  is
 "saturated".  When the  bed is  completely  saturated,  the organic will
 "breakthrough"  the  bed  with  the exhaust  gas  and the  inlet  gases must
then be  routed  to an  alternate bed.   The  saturated  bed is  then  regenerated
to  remove  the  adsorbed  material.
      Low-pressure  steam (3)  is usually used  to  heat  the carbon  bed
 during the regeneration cycle, driving off the  adsorbed organics, which
 are usually recovered by condensing the vapors  (4)  and separating.them
                                  4-29

-------
VOC-tatet
Y«t.Stru«
    VENT TO
  ATMOSPHERE
   Low-pressure
   Stem
                                       AOSCH8EH 2
                                      (REGENERATING)
                                                       DECANTER
                                                         and/ or
                                                     OISTIUJNG TOWEH
Recovered
Solvent


Water
           Figure 4-7.   Two Stage Regeneration  Adsorption System
                                     4-30

-------
from the steam condensate by decanting or distillation (5).   The adsorption/
regeneration cycle can be repeated numerous times, but eventually the
carbon loses its adsorption activity and must be replaced.  The carbon
can sometimes be reactivated by recharring.
     4.1.2.2.1  Adsorber control efficiency.  The efficiency of an
adsorption unit depends on the properties of the carbon and the adsorbate,
and on the conditions under which they contact.  Lower temperatures aid:,
the adsorption process, while higher temperatures reduce the adsorbent's
capacity.46  Removal efficiencies of 95 to 99 percent are achieved by
well-designed and well-operated units.4'
     4.1.2.2.2  Applicability.  Adsorbers effectively control streams
with dilute concentrations of organics.  In fact, to prevent excessive
temperatures within the bed due to the heat of adsorption, inlet concen-
trations of organics are usually limited to about 0.5 to  1 percent.44
The maximum practical inlet concentration is about 1 percent, or
10,000 ppmv.48  Higher concentrations are frequently handled by allowing
some condensate to  remain  from  the  regeneration  process to remove the
heat generated during adsorption.  Also, the inlet stream can be diluted
by use of  a condenser or addition of  air or nitrogen upstream of the
adsorber.  If the organic  is  reactive or oxygen  is present in the vent
stream, then  additional precautions may be  necessary to safeguard the
adsorption system.
     Adsorbers can  foul and hence are not  very  suitable for  streams
containing fine particles  or  polymerizeable monomers.  Both  can contaminate
the  beds  and  result in  poor performance, or even introduce safety problems.
Because  of their  limitations  in certain  gas streams,  carbon  adsorbers
are  not  ideally suited  for most of  the  emission streams encountered  in
the  polymers  and  resins  industry.
      4.1.2.3   Absorbers.   Absorption  is a  gas-liquid  mass transfer
 operation in  which  a gas  mixture is contacted  with a  liquid  (solvent)
 for the purpose of  preferentially  dissolving  one or  more  components
 (solutes) of  the  gas.   Absorption  may entail  only the  physical  phenomenon
 of solution or may  also involve chemical  reaction of the  solute with
 constituents  of the solvent.49
      For any  given  solvent,  solute, and set of operating  conditions,
 there exists a theoretical equilibrium ratio of solute concentration in
                                  4-31

-------
the gas mixture to solute concentration in the solvent.  The driving
force for mass transfer in an operating absorption tower is related to
the difference between the actual concentration ratio and this equilibrium
ratio.50  The solvents used are chosen for high solute (VOC) solubility
and include liquids such as water, mineral oil, nonvolatile hydrocarbon
oils, and aqueous solutions of oxidizing agents like sodium carbonate
and sodium hydroxide.51
     Devices based on absorption principles include spray towers, venturi
scrubbers, packed columns, and plate columns.  Spray towers and venturi
scrubbers are generally restricted to particulate removal and control of
high-solubility gases.52  Most VOC control by gas absorption is by
packed or plate columns.  Packed columns are used mostly for handling
corrosive materials, liquids with foaming or plugging tendencies, or
where excessive pressure drops would result from the use of plate columns.
They are less expensive than plate columns for small-scale or pilot
plant operations where the column diameter is less than 0.6 m (2 ft).
Plate columns are preferred for large-scale operations, where internal
cooling is desired, or where low liquid flowrates would inadequately wet
the packing.5-*
     A schematic of a packed tower is shown in Figure 4-8.  The gas is
introduced at the bottom (1) and rises through the packing material (2).
Solvent flows by gravity from the top of the column (3), countercurrent
to the vapors, absorbing the solute from the gas phase and carrying the
dissolved solute out of the tower (4).  Cleaned gas exiting at the top
is ready for release or final treatment such as incineration.
     The major tower design parameters, column diameter and height,
pressure drop, and liquid flowrate, are based on the specific surface
area of the tower packing, the solubility and concentration of the
components, and the quantity of gases to be treated.
     4.1.2.3.1  Absorber control efficiency.  The VOC removal efficiency
of an absorption device is very dependent on the characteristics of the
solvent and the design and operation of the tower.  Generally, for a
given solvent and solute, an increase in absorber size or a decrease in
the operating temperature can increase the VOC control efficiency of the
system.
                                 4-32

-------
                                                                         CLEANED GAS OUT
                                                                      ^  To Final Control Oevics
ABSORBING  (3)
UQUID IN
                                                                                     VQCLAOEN
                                                                                     GAS IN
                                                  (4)
                                         ABSORBING UQUID
                                          WITH VOC OUT
                                    To Disposal ar VOC/Solvent Recavety
                        Figure 4-8.   Packed Tower  for  Gas Absorption
                                           4-33

-------
     Systems that utilize organic liquids as the solvent usually include
a separate item of equipment to strip the adsorbed gas so that the
solvent can be recycled to the absorber.  The efficiency of the absorber
is affected by the efficiency of the stripper.  For example, a theoretical
absorber calculated to achieve a removal efficiency of 99.9 percent
with once-through solvent usage (equivalent to 100 percent stripping
efficiency), would achieve only 98.5 percent VOC removal if the solvent
were recycled through a stripper that was 98 percent efficient.54
     4.1.2.3.2  Applicability.  The selection of absorption for VOC
control depends on the availability of an appropriate solvent for the
specific VOC.  Absorption is sually not considered when the VOC
concentration is below 200-300 ppmy.55
     The use of absorbers is generally limtied to applications in which
the stripped absorbent can be reused directly or with minimum treatment.
Absorption may not be practical if the waste gas stream contains a
mixture of organics, since all will likely not be highly soluble in the
same absorbent.  Abosrbers have found limited use as a VOC emission
control device in the polymers and resins industry.
4.2  CONTROL TECHNIQUES FOR FUGITIVE EMISSIONS
     This section discusses control techniques that can be appleid to
reduce fugitive VOC emissions in the polymers and resins industry.  Two
approaches are available.  The first involves a leak detection and
repair program in which leaking fugitive emissions sources are located
and repaired at specified intervals.  The second is a preventive approach
whereby fugitive emissions are either captured and vented to a control
device or eliminated through the installation of specified controls or
leakless equipment.  The technical application of these methods is
briefly explained in the following subsections.
4.2.1  Leak Detection and Repair Program
     4.2.1.1  Leak Detection.  The most common types of equipment that
have the potential to release fugitive emissions are valves, pump and
compressor seals, pressure relief devices, flanges, open ended lines,
and sampling connections.  When a leak develops, it can be detected in
several ways - instrument monitoring of the  individual  component, a unit
                                 4-34

-------
 area  survey,  or  by  means  of  a  fixed point monitoring system.  These
 methods would generally yield  only a qualitative indication of the size
 of  a  leak.56
      4.2.1.2   Repair Program.  When a  leak is located, the leaking
 component can be scheduled for repair  or replacement.  Many components
 can be serviced  on-line without disturbing the plant operations.  An
 example would be the tightening of a valve packing gland.  If tightening
 the packing gland does not stop the leak, the valve must be isolated from
 the process.   Control valves can often be isolated, but block valves
 generally cannot be.
      Repairs  of  rotary equipment seals usually require isolation of the
 leaking device.  Almost all process pumps are "spared" (two installed in
 parallel, but only  one required for operation) so that either one can be
 isolated for  repair.  However, most compressors are not spared, and so
 compressor seal  replacements often necessitate a partial  or complete
 shutdown.
     Most leaking flanges can be resealed simply be retightening the
 flange bolts.  A flange leak that requires a gasket seal  replacement
would likely  require a total or partial shutdown of the entire unit.
When the leaks can  be corrected only by a total  or partial  shutdown, the
temporary emissions resulting from a shutdown and startup could be
 larger than the continuous fugitive emissions that would  take place
before a scheduled  shutdown.57  For this reason, the repair of certain
leaks is best  delayed until the next scheduled shutdown.
     4.2.1.3   Effectiveness of Leak Detection and Repair  Programs.   The
emission reduction  achieved by a leak detection and repair program is
dependent on  several factors, including the leak definition,  the inspection
interval, the  allowable delay until repair,  and the effectiveness of the
repair.
     To implement a monitoring program, an instrument meter reading
 (organics concentration)  which will indicate  an  equipment leak must be
defined.   The meter reading selected may vary from 1,000  ppmv to
100,000 ppmv.58  The theoretical  efficiency can  then  be estimated by
relating the  leak definition to the percentage of total mass  emissions
that can  be expected from sources  with  concentrations  at  the  source
greater than the leak definition.   In general, defining a leak by a low
                                 4-35

-------
meter reading results in larger potential emission reductions.  However,
the potential increase in emissions that can occur through attempts at
repairing relatively minor leaks may dictate that these leaks be excluded
from the leak definition.  For the polymer and resins industry, a leak
definition of 10,000 ppm was selected.
     The inspection interval depends on the expected occurrence and
recurrence of leaks after a piece of equipment has been checked and/or
repaired.  This interval can also vary with the type of equipment and
service conditions.  Monitoring may be scheduled on an annual, quarterly,
monthly, or weekly basis.
     If a leak is detected, the equipment should be repaired as soon as
practicable.  The longer the delay, the less effective will be the
overall reduction program.
4.2.2  Preventive Programs
     An alternative approach to the leak repair program is to replace
the potentially leaky equipment with components based on "leakless"
technology, or with equipment that captures the leakage for control.
This approach is referred to as a preventive program.  For example, in
many cases, emissions from a pump seal can be reduced to a negligible
level through the installation of an improved shaft sealing mechanism,
such as dual mechanical seals, or it can be eliminated entirely by
installing sealless pumps, which do not have a shaft/casing junction and
thus do not leak during normal operation.^9  A barrier fluid can be
circulated between the mechanical seals.  Degassing vents in the barrier
fluid system allow the transport of leakage in a closed system to a
control device.  These solutions are sometimes not feasible.  For example,
the maximum service temperature of a dual mechanical seal is usually
about 260°C (500°F).  Mechanical seals also cannot be used on pumps with
reciprocating shaft motion or those handling extremely corrosive or
abrasive fluids.
     As in the case of pumps, compressor emissions occur at the junction
of the moving shaft and the stationary casing.  Emissions from both
centrifugal and rotary compressors can be controlled either with mechanical
seals with barrier fluid systems or with liquid film seals.  As with
pumps, the degassing vents for the seal fluid must discharge into a
                                 4-36

-------
 closed system to prevent process gas from escaping.  Leakage from recip-
 rocating compressors can be controlled by creating a void in the packing
 and  inserting one or more spacer rings into the packing gland or by
 enclosing the seal area and venting the leakage collected by either
 system to a control device.
     Leakage from safety/relief valves' can be minimized by installing a
 rupture disk upstream of each valve.  Such combinations can be spared by
 installation of a two-way valve which assures that one emergency relief
 system is always operational.   The other could then be repaired while
 process operations continue.  An alternative method for controlling
 relief valve emissions in some types of service is to use a soft elastomer
 seat in the valve.
     Caps,  plugs,  and second valves are devices that can essentially
eliminate fugitive emissions from open-ended lines.  VOC emissions  from
the purging of sampling lines  can be minimized by a closed-purge sampling
system that enables  purge organics  to be recycled to the process or
contained for subsequent disposal.
                                     4-37

-------
4.3  REFERENCES FOR CHAPTER 4

1.   Lee, K.C., H.J. Jahnes, and D.C. Macauley.   Thermal  Oxidation
     Kinetics of Selected Organic Compounds.  Journal  of the  Air
     Pollution Control Association.  29:749-751.  July 1979.   p.  750.
     Docket Reference Number II-I-47.*

2.   Perry, R.H. and C.H. Chilton.  Chemical Engineers'  Handbook, Fifth
     Edition.  McGraw-Hill Book Company.  1973.   p.  9-18.   Docket Reference
     Number II-I-16.*

3.   Kalcevic, V.  Organic Chemical Manufacturing Volume 4:   Combustion
     Control Devices, Report No. 4, Control  Device Evaluation,  Flares
     and the Use of Emissions as Fuels.  U.  S. Environmental  Protection
     Agency.  Research Triangle Park, N.C.  Publication No.  EPA-450/3-80-026.
     December 1980.  Docket Reference Number II-A-18.*

4.   Klett, M.G. and J.B. Galeski.  Flare Systems Study.  Lockheed
     Missiles and Space Company.  NTIS Report PB-251664.  Publication
     No. EPA 600/2-76-079.  March 1976.  Docket Reference Number  II-A-2.*

5.   Payne, R., D. Joseph, J. Lee, C. McKinnon, and  J. Pohl.   Evaluation
     of the Efficiency of Industrial  Flares  Used to  Destroy Waste Gases.
     Phase I Interim Report - Experimental Design.  EPA Contract
     No. 68-02-3661.  Draft, January 1982.  p. 2-20.  Docket Reference
     Number II-A-29.*

6.   Palmer, P.A.  A Tracer Technique for Determining  Efficiency  of an
     Elevated Flare.  E.I. duPont de Nemours and Company.   Wilmington,
     DE.  1972.  Docket Reference Number II-I-15.*

7.   Siegel, K.D.  Degree of Conversion of Flare Gas in Refinery  High
     Flares.  University of Karlsruhe, The Federal Republic of Germany.
     Ph.D. Dissertation.  February 1980.  Docket Reference Number II-I-57.*

8.   Lee, K.C. and G.M. Whipple.  Waste Gas  Hydrocarbon Combustion in  a
     Flare.  Union Carbide Corporation.  South Charleston, W.V. (Presented
     at the 74th Annual Meeting of the Air Pollution Control  Association.
     Philadelphia, PA.  June 21-26, 1981.)  Docket Reference  Number II-I-73.*

9.   Howes, J.E., T.E. Hill, R.N. Smith, G.R. Ward,  and W.F.  Herget.
     Development of Flare Emission Measurement Methodology.   EPA  Contract
     No. 68-02-2682.  Draft, 1981.  Docket Reference Number II-A-27.*

10.  Reference 5.

11.  McDaniel, M.  Flare Efficiency Study, Volume I.  Engineering-Science.
     Austin, Texas.  Prepared for Chemical Manufacturers Association,
     Washington, D.C.  Draft 2, January 1983.  Docket  Reference
     Number II-I-107.*

12.  Kenson, R.E.  A Guide to the Control of Volatile  Organic Emissions.
     Systems Division, MET-PRO Corporation.   Technical Page  10T-1.
     Harleysville, PA.  1981.  Docket Reference Number II-I-65.*

                                      4-38

-------
 13.  Reference 1, p. 749.

 14.  Keller, M.  Comment on Control  Techniques Guideline Document  for
      Control of Volatile Organic Compounds  Emissions  from Manufacturing
      of High-Density Polyethylene,  Polypropylene,  and Polystyrene  Resins.
      NAPCTAC Meeting.  June 1,  1981.   p.  6.   Docket  Reference  Number
      II-D-39.*

 15.  Blackburn, J.W.  Organic  Chemical  Manufacturing, Volume 4:
      Combustion Control  Devices, Report 1,  Thermal Oxidation.  U.  S.
      Environmental  Protection  Agency.   Research  Triangle Park, N.C
      Publication No. EPA-450/3-80-026.   December 1980.   p.  1-1.
      Docket Reference Number II-A-18.*

 16.  Basdekis,  H.S.   Organic Chemical Manufacturing,  Volume 4: Combustion
      Control  Devices,  Report 2,  Thermal Oxidation  Supplement (VOC
      Containing Halogens  or Sulfur).  U.  S. Environmental  Protection
      Agency.   Research Triangle  Park, N.C.  Publication  No. EPA-450/3-80-026.
      December 1980.   pp.  1-2 and 1-4.   Docket  Reference  Number II-A-18.*

 17.   North  American  Manufacturing Company.  North American Combustion
      Handbook.   2nd  Edition.  Cleveland,  North American  Mfg. Company
      1978.   p.  264.   Docket Reference Number II-I-37.*

 18.   Reference  16, pp. 1-1  and 1-2.

 19.   Stern,  A.C., ed.  Air  Pollution, Third Edition, Volume IV, Engineering
      Control  of Air  Pollution.   New York, Academic Press.  1977.   p. 368
      Docket  Reference Number II-I-26.*

 20.   Maseone, D.C.   Thermal  Incinerator Performance for NSPS.   U. S.
      Environmental Protection Agency.  Research Triangle Park,  N.C.
      Memorandum  to J.R. Farmer,  Chemicals and Petroleum Branch.  June 11
      1980.   Docket Reference Number II-B-4.*

 21.   Maseone, D.C.  Thermal  Incinerator Performance for NSPS,  Addendum.
      U. S. Environmental  Protection Agency.   Research Triangle  Park,
      N.C.  Memorandum to J.R. Farmer, Chemicals and Petroleum  Branch
      July 22, 1980.  Docket  Reference Number II-B-5.*

 22.   Reference 20, p. 1.

 23.   EEA, Incorporated.  Trip Report to ARCO Polymers, Inc. EPA  Contract
      No. 68-02-3061, Task 2.  August 20, 1980.   Docket Reference  Number
      II-B-10.*
24.
25.
U. S. Environmental  Protection Agency, Office of Air and Waste
Management.  Control  Techniques for Volatile Organic Emissions from
Stationary Sources.   Research Triangle Park, N.C.   Publication
No. EPA-450/2-78-022.  May 1978.   p. 32.   Docket Reference  No. II-A-5.*

U. S. Environmental  Protection Agency, Office of Air and Water
Programs.  Air Pollution Engineering Manual.  Research  Triangle
Park, N.C.  Publication  No.  AP-40.   May  1973.   p.  180.   Docket
Reference Number II-I-18.*
                                    4-39

-------
26.  Reference 25, p. 181.

27.  Reference 24, p. 34.

28.  Key, J. A.  Organic Chemical  Manufacturing, Volume 4:   Combustion
     Control Devices, Report 3, Catalytic Oxidation.   U. S.  Environmental
     Protection Agency.  Research  Triangle Park, N.C.  Publication No.
     EPA-450/3-80-026.  December 1980.  p. II-9.  Docket Reference
     Number II-A-18.*

29.  Engelhard Industries Division, Engelhard Corporation.   Catalytic
     Incineration of Low Concentration Organic Vapors.  Prepared for
     U. S. Environmental Protection Agency.  Washington, D.C.   Contract
     No. 68-02-3133.  January 1981.  Docket Reference Number II-A-21.*

30.  Letter from Monsanto Company  to J.R. Farmer, U.  S. Environmental
     Protection Agency.  April 19, 1982.  p. 17.  Docket Reference Number
     II-D-57.*

31.  U.S. Environmental Protection Agency.  A Technical Overview of the
     Concept of Disposing of Hazardous Wastes in Industrial  Boilers.
     Cincinnati, Ohio.  EPA Contract No. 68-03-2567.   January 1981.  p. 5-2.
     Docket Reference Number II-A-38.*

32.  Reference 31, pp. 5-7 through 5-19.

33.  U.S. Environmental Protection Agency.  Evaluation of PCB Destruction
     Efficiency in an Industrial Boiler.  Research Triangle Park,
     North Carolina.  EPA Contract No. 600/2-81-055a.  April 1981.
     Docket Reference Number II-A-41.*

34.  U.S. Environmental Protection Agency.  Emission Test Report on
     Ethylbenzene/Styrene.  Amoco Chemicals Company.  Texas City, Texas.
     Research Triangle Park, North Carolina.  EMB Report No. 79-OCM-13.
     August 1979.  Docket Reference Number II-A-34.*

35.  U.S. Environmental Protection Agency.  Emission Test Report.
     El Paso Products Company.  Odessa, Texas.  Research Triangle Park,
     North Carolina.  EMB Report No. 79-OCM-15.  April 1981.  Docket
     Reference Number II-A-40.*

36.  U.S. Environmental Protection Agency.  Emission Test Report.  USS
     Chemicals.  Houston, Texas.  Research Triangle Park, North Carolina.
     EMB Report No. 80-OCM-19.  August 1980.  Docket Reference Number
     II-A-35.*

37.  Shell Chemical Company, Woodbury Plant.  Application for Permit to
     Construct, Install or Alter Control Apparatus or Equipment.  To New
     Jersey State Department of Environmental Protection.  March 16,
     1976.  Docket Reference Number II-I-24.*

38.  EEA, Incorporated.  Trip Report to Phillips Chemical Company's
     HOPE plants at Pasadena,' Texas.  Contract No. 68-02-3061, Task 2.
     August 28, 1980.  Docket Reference Number II-B-13.*
                                    4-40

-------
39.  Erikson, D.G.  Organic Chemical  Manufacturing, Volume 5:  Adsorption,
     Condensation, and Absorption  Devices,  Report  2, Condensation.
     U. S. Environmental  Protection Agency.   Research Triangle Park,
     N.C.  Publication No.  EPA-450/3-80-027.   December 1980.  p. II-3.
     Docket Reference Number II-A-19.*

40.  Reference 24, p. 84.

41.  Devore, A., G.J. Vago, and  6.J.  Picozzi.   Heat Exchangers:  Specifying
     and Selecting.   Chemical  Engineering.  87(20):133-148.  October 1980.
     p. 136.  Docket Reference Number II-I-61.*

42.  Kern, D.Q.  Process  Heat  Transfer.  New  York, McGraw-Hill Book
     Company.  1950.  p.  255.  Docket Reference Number II-I-3.*

43.  Reference 39, p. II-4.

44.  Reference 39, -p. IV-1.

45.  Parmele, C.S.,   W.L. O'Connell,  and H.S.  Basdekis.  Vapor-Phase
     Adsorption Cuts Pollution,  Recovers Solvents.  Chemical Engineering.
     86(28) ;58-70.  December 1979.  p. 60.  Docket Reference Number
     II-I-51.*

46.  Basdekis, H.S.  and C.S. Parmele.  Organic  Chemical Manufacturing,
     Volume 5:  Adsorption, Condensation, and Absorption Devices, Report 1,
     Carbon Adsorption.  U. S. Environmental  Protection Agency.  Research
     Triangle Park,  N.C.  Publication No. EPA-450/3-80-027.  December 1980.
     p. II-l.  Docket Reference  Number II-A-19.*

47.  Reference 45, p. 69.

48.  Reference 45, p. 62.

49.  Reference 2, p. 14-2.

50.  Standifer, R.L.  Organic  Chemical Manufacturing, Volume 5:  Adsorption,
     Condensation, and Absorption  Devices,  Report 3, Gas Absorption.
     U. S. Environmental  Protection Agency.   Research Triangle Park,
     N.C.  Publication No.  EPA-450/3-80-027.   December 1980.  p. III-5.
     Docket Reference Number II-A-19.*

51.  Reference 24, p. 76.

52.  Reference 50,  p. II-l.

53.  Reference 2, p. 14-10.

54.  Reference 50, pp. III-6 and III-7.

55.  Reference 50, p. 1-1.
                                 4-41

-------
 56.   U.  S.  Environmental  Protection Agency.  VOC Fugitive Emissions in
      Petroleum Refining Industry  - Background  Information for Proposed
      Standards,  Draft EIS.   Research Triangle  Park, N.C.  Publication
      No. EPA-450/3-81-015a.   November  1982.  p. 4-1.  Docket Reference
      Number II-A-33.*

 57.   Reference 56,  p.  4-7.

 58.   Reference 56,  p.  4-8.
59
      U.  S.  Environmental Protection Agency.  VOC Fugitive Emissions in
      Synthetic  Organic Chemicals Manufacturing Industry - Background
      Information  for  Proposed Standards, Draft EIS.  Research Triangle
      Park,  N.C.   Publication No. EPA-450/3-80-033a.  November 1980.
      p.  4-13.   Docket Reference Number II-A-16.*
*References can be located in Docket Number A-82-19  at the U.S.
Environmental Protection Agency Library,  Waterside Mall, Washington,
•J • w •


                                 4-42

-------
                 5.0  MODIFICATIONS AND RECONSTRUCTIONS

     The provisions of Title 40 Code of Federal  Regulations,  Sections  60.14
and 60.15 (40 CFR 60.14 and 60.15) state that an existing facility  can
become an affected facility and consequently, subject to the  new source
performance standard (NSPS) if there is a modification or reconstruction
to that operation.  An "existing facility," defined in 40 CFR 60.2, is a
facility of the type for which a standard of performance is promulgated
and the construction or modification of which was commenced prior to the
proposal date of the applicable standards.  This chapter gives the
definition of modification and reconstruction as found in 40  CFR 60.14
and 60.15.  A discussion of possible process changes at a polymers  and
resins facility that would constitute a modification or reconstruction,
making it subject to the proposed NSPS, is also  included in the chapter.
5.1  DEFINITIONS
5.1.1  Modification
     Modification is defined in Section 60.14 as any physical or operational
change to an existing facility which results in  an increase in the
emission rate of the pollutant(s) to which the standard applies.  Paragraph  (e)
of Section 60.14 lists exceptions to this definition which will  not be
considered modifications, irrespective of any changes in the  emission
rate.  These changes include:
     1.  Routine maintenance,  repair, and replacement at the  facility,
     2.  An increase in the production rate not  requiring a capital
expenditure as defined in Section 60.2 (bb),
     3.  An increase in the hours of operation,
     4.  Use of an alternative fuel or raw material if the existing
facility was designed to accommodate the alternate fuel or raw material
prior to the date of any applicable standard,
                               5-1

-------
      5.  The addition or use of any system or device whose primary
 function is to reduce air pollutants, except when a system is removed or
 replaced by a system considered to be less efficient, and
      6.  Relocation or change in ownership.
      As stated in paragraph (b), emission factors,  material balances,
 continuous monitoring systems, and manual  emission  tests  are to  be used
 to determine emission rates expressed as  kg/hr of pollutant.  Paragraph  (c)
 affirms that the addition of an affected  facility to a stationary  source
 through any mechanism — new construction, modification,  or reconstruction -
 does  not make any other facility within the  stationary source subject  to
 standards of performance.  Paragraph  (f)  provides for superseding  any
 conflicting provisions and (g) stipulates  that compliance be achieved
 within  180 days  of the completion of  any  modification.
 5.1.2  Reconstruction
      A  "reconstruction" occurs when replacement of  components at an
 existing facility takes place to such extent  that:   (1) the fixed  capital
 cost  of the new  components exceeds  50 percent of the fixed  capital cost
 that  would be required to construct a comparable new facility, and
 (2) it  is economically and technologically feasible  for the facility to
 comply  with the  applicable standards  set forth.   Any existing facility
 undergoing "reconstruction"  becomes an affected facility  subject to
 requirements  of  NSPS,  irrespective  of any  change in  pollutant  emission
 rate.
      The owner or operator of  any existing facility  who proposes to
 replace components  such that the  fixed capital  cost  exceeds  50 percent
 of the  fixed  capital  of a comparable  new facility, must notify the
 Administrator in  writing  60  days  prior to  commencement of the replacement.
 The notice must  include identification of  the  owner  and plant, a description
 of the  replacement  being  made  including present  and  proposed air pollution
 control  equipment,  an  estimate of the proposed  replacement  costs and the
 cost  of  a  comparable new  facility,  an approximation of the  operating
 life  of  the existing facility after replacement, and a discussion of any
economic or technical  limitations the facility may have in complying
with  applicable standards.
                               5-2

-------
     The Administrator will determine whether the proposed replacement
constitutes a "reconstruction" within 30 days of receipt of the operators
notice and any additional information he or she may reasonably require
for making a decision.  The final decision is based upon the costs
involved, the estimated operating life of the facility after replacement
compared to a new facility, the extent to which the components being
replaced cause or contribute to air pollutant emissions from the facility,
and any economic or technical limitations for compliance with applicable
standards.                                              -
     The "reconstruction" provision requires a facility to comply with
the NSPS if it undergoes changes that make it essentially a new source.
The purpose of the provision is to assure that during reconstruction the
facility will install the appropriate emission control equipment.
5.2  MODIFICATIONS AND RECONSTRUCTIONS AT POLYMERS AND RESINS FACILITIES
     The polymers and resins industry is expected to experience some
growth in the coming years.  Within the industry, process technology and
operational procedures are undergoing continual change.  For example, in
the polypropylene industry, catalysts and other technology continue to
improve.^  In addition, some companies are diversifying their product
lines by shifting to copolymers or to new combinations of comonomers.
These factors may lead to process changes at existing facilities.
Changes in operating conditions would mean that an existing facility
would be subject to new source standards of performance if the changes
cause increased emissions.- Under these conditions, the facility becomes
a modified facility.  However, it is difficult to determine what kinds
of physical or operational changes made at polymers and resins facilities
will constitute a modification, owing to wide variations in VOC flow and
concentration in process vent streams.  These variations make it difficult
to assess emission increases under paragraph (a)  of Section 60.14.
Several changes that could be encountered in polymer and resin plants
and their possible effects on emissions are presented below.
5.2.1  Process Emissions
     In general, a number of modifications may be made to polyolefin
plants to increase production.  Changes may include upgrading, adding,
or improving such equipment as electric drivers,  reactor compressors,
                               5-3

-------
 catalyst  addition systems,  refrigeration equipment, heat exchange equipment
 and piping, and the reactors themselves.  In general, these changes will
 result  in increases in baseline emissions, especially those which result
 from the  increase in pellet production.
     Existing high pressure, LDPE facilities and liquid phase, HOPE
 facilities are being, and others will be, converted to the production of
 linear  LDPE.  This conversion may result from (1) replacing existing
 equipment with the new gas phase technology for producing LLDPE, or
 (2) modifying existing equipment in high pressure, LDPE facilities to
 allow LLDPE production with either the gas phase process2 or the liquid
 phase solution process.3  These changes, on the whole, are likely to
 result  in a decrease in baseline level of emissions and emission rates
 as given  in Chapter 6.  The cost of modifying existing facilities to
 accept  the gas phase technology has been reported to be relatively low.4
     Recent advances in LLDPE gas phase technology have resulted in
 capacity  increases in existing facilities by 35 percent with little or
 no capital investment and by 65 percent with "some" capital investment.2
 One company is increasing its capacity by approximately 33 percent
 through changes in the gear train drive on the cycle gas compressor,
 installation of an additional transfer line to the fluff storage vessels,
 and installation of a new Cg unloading system.   No increase in process
 emissions is expected from these changes.  However, fugitive emissions
 from these specific changes are expected to increase.5
     The conversion to copolymer production or to new copolymers may
 result  in increased baseline emissions as the comonomer is more likely
 to be less volatile than the olefins, hence retained longer in the
 process and not released until product finishing or storage steps, which
 are not typically under baseline control.  However, steam stripping of
the product before or immediately after extrusion,  followed by condensation,
 can control these emissions so that overall  emissions  from the process
 line will not be increased.  At least one company has  sought to recover
a comonomer (vinyl acetate) from ethylene recycle streams  by adding a
distillation column and auxiliary equipment.6  The addition of this
 equipment does not create any new process emission sources^ as the
ethylene will  be the overhead product and the vinyl  acetate the bottoms
product.
                               5-4

-------
   -   Other  potential  changes  in  polymer and  resin processes  include
 changes  in  type or  operation  of  the product  dryer, condenser, or distillation
 columns.  In  polypropylene  and high pressure, LOPE processes, most newly
 installed dryers are  fluidized bed, closed loop systems, which recycle
 the  nitrogen  used for drying  polymer products.  However, some existing
 plants use  rotary dryers, which  vent emissions directly to the atmosphere.
 These dryers  may be replaced  with fluidized  dryers because the latter
 are  more efficient.   This conversion is likely to result in a decrease
 in emissions.  Another operational change may involve increasing the
 operating pressure of product dryers, which  may increase emission rates.
      There  are many instances where a polymers and resins facility may
 need  to  replace parts that have  failed or not "performed well with a dif-
 ferent,  improved part.  This  is  often the case with distillation columns
 and their associated  condensers.  The trays  or packing materials in a
 distillation  column in conjunction with the  operation of the condenser
 can have an important impact  on  pollutant emissions from the process.
 Replacing column or condenser parts with improved equipment can reduce
 pollutant emissions.  If the  replacement results in an increase in
 emissions,  it is not  exempt from being considered a modification.  An
 example where such replacement may occur is in the gas phase process  for
 producing LDPE and HOPE.  In this process, the butene (comonomer) used
 for producing these polymers must be purified by distillation before  use
 in the reaction processes.  The  distillation column and condenser used
 in the purification may be replaced.  If emissions could be offset
 elsewhere so that there was no net increase in emissions,  the existing
 facility would not become subject to the standards.
 5.2.2  Fugitive Emissions^
     Routine equipment changes and additions at polymers and resins
 facilities for increased ease of maintenance, plant productivity  or
 plant safety can cause an increase in the fugitive emission rate.
However, fugitive emissions from other sources could be reduced to
 compensate for this increase.
     Potential fugitive emission sources,  such as  pumps or valves,  may
 be replaced.  If such  a source is replaced with an equivalent source
 (such as is  done during routine repair and replacement), the fugitive
                               5-5

-------
emissions from the facility should not increase because the number of
potential sources in the same vapor pressure service .(i.e., handling the
same monomer or comonomer) remains unchanged.
     As noted above, process equipment pieces may be modified or added
to existing facilities to increase the capacity of or to optimize a
process.  The addition of new equipment would normally increase fugitive
emissions from a facility due to the increased number of potential
emission sources (pumps, valves, sampling connections, etc.) that are
associated with the process equipment.
     When an owner or operator replaces several components of an existing
facility,'that facility may become subject to applicable standards of
performance under the provisions of Section 60.15.  For example, if an
owner or operator replaces several fugitive emission sources such as
pumps, compressors, or sampling loops in an existing facility, and if
the fixed capital costs for the new equipment exceeds 50 percent of the
costs of all fugitive emissions sources in the unit, the Administrator
may determine that reconstruction has occurred.  Reconstructions may
occur as a result of damage caused by fires, explosions, hurricanes, or
other catastrophes.  They might also result from feedstock changes,
product changes, or other major process changes which would require
additions or replacement of several fugitive emission sources.
     The process can also be changed without changing the polymer.  One
such case would be a change in catalyst.  In this case, fugitive emissions
would not be expected to change because neither the number of fugitive
sources nor the vapor pressure of the monomer or comonomer(s) would
change.
5.2.3  Summary
     In general, some alterations are likely to be made in existing
polypropylene and polyethylene plants that would be considered modifications
or reconstructions.  However,  most changes likely to be made in existing
polymer plants will result in  reduced emissions and hence will not be a
modification as defined by Section 60.14.   For those changes where there
is a potential for increased emissions, relatively inexpensive equipment
can be installed to control  these emissions, so that there will  be no
increase in emissions from the production  line, and hence no modification.
                               b-6

-------
5.3  REFERENCES FOR CHAPTER 5

1.  "KEY POLYMERS:  POLYPROPYLENE".   Chemical  and  Engineering  News.
     September 6, 1982.   p. 15.   Docket Reference  Number  II-I-96.*

2.   "Union Carbide Unveils an Improved UNIPOL Process  for LLDPE  Resins,"
     Chemical  Engineering.   April  5,  1982.   p. 17.  Docket Reference  .
     Number II-I-88.*

3.   "Dow has Announced  New Linear Low Density Polyethylene (LLDPE)
     Technology."  Chemical Engineering.  October  5, 1981.  p.  35.
     Docket Reference Number II-I-78.*

4.   "A Step Up for LLDPE Know-How."   Chemical Week. March 31, 1982,
     p. 11.  Docket Reference Number  II-I-86.*

5.   Texas Air Control Board.   Permit Amendment No.  8334.  High Density
     Polyethylene Production Expansion.  March 30, 1981.   Docket Reference
     Number II-I-70.*

6.   Texas Air Control Board.   Permit No. 7021.  E.I. duPont,  Sabine
     Power Works, Polyolefins  D & 6 Unit.  October 5, 1978.  Docket
     Reference Number II-I-43.*

7.   VOC Fugitive Emissions in Synthetic Organic Chemicals Manufacturing
     Industry - Background Information for Proposed Standards.   Chapter 5,
     Modification and Reconstruction.  EPA-450/3-80-033a.  November  1980.
     Docket Reference Number II-A-16.*
 *References  can  be  located  in Docket Number A-82-19 at the U.S.
  Environmental Protection Agency, Waterside Hall, Washington, D.C,
                                     5-7

-------

-------
           6.0  MODEL PLANTS AND REGULATORY ALTERNATIVES

     This chapter presents model plants and their parameters and
regulatory alternatives for the reduction of process and fugitive
VOC emissions from the five polymers and resins segments chosen for
NSPS development.  Section 6.1 presents the model plants chosen to
represent the five polymers and resins segments, and Section 6.2 describes
the regulatory baseline (i.e., the level of control  that is likely to
be employed in new plants in the absence of a new source performance
standard) and the individual regulatory alternatives for each model
plant.
6.1  MODEL PLANTS
     A plant is modeled to describe process parameters and offgas stream
characteristics that are representative of a typical new source being
regulated by the standard.  The purpose of developing a model plant is
to study the effect of regulatory alternatives on each type of facility
regulated by the NSPS and to estimate the energy, environmental, and
economic impacts associated with each regulatory alternative.
     As  described in Chapter 3, the five polymer and resin  segments,
especially the polyolefins, share certain fundamental similarities in
terms of their processes.   At  the same  time, however, there are among
the  various  processes many  differences  that affect  process  parameters
and  process  emission characteristics.   Thus, no  single model plant can
adequately characterize  the process emissions  of all five polymers and
resins  segments.  Therefore,  twelve model  plants were developed:
       1.  Polypropylene  -  continuous,  liquid phase  slurry process,
       2.  Polypropylene  -  gas  phase process,
       3.  Low density  polyethylene  -  high-pressure, process,
       4.  Low density  polyethylene  -  low-pressure process  and  high density
 polyethylene - gas  phase process,
                                   6-1

-------
      5.  High density polyethylene - liquid phase slurry process,
      6.  High density polyethylene - liquid phase solution process,
      7.  Polystyrene - continuous process,
      8.  Expandable polystyrene - post-impregnation suspension process,
      9.  Expandable polystyrene - in-situ suspension process,
     10.  Poly(ethylene terephthalate), (PET) - DMT process, and
     lla. Poly(ethylene terephthalate), (PET) - TPA process, for low
          viscosity PET and high viscosity PET with a single end finisher,
     lib. Poly(ethylene terephthalate), (PET) - TPA process for high
          viscosity PET with multiple end finishers.
     The model plants were selected to be representative of basic manu-
facturing processes used in plants making these polymers and resins and
not of any individual processes used by a specific plant.  However, .in
order to provide a basis for economic analysis, data from a specific
plant and its process were used.  No model plant was developed based upon
the polystyrene batch process for the production of crystal or impact
polystyrene because available information indicates that new plants that
produce crystal or impact polystyrene are unlikely to use this process.
     Tables 6-1 through 6-llb summarize the parameters for process emissions
for each model plant used in the study.  The model plants are based upon
the respective plant descriptions in Chapter 3.  Each model plant is
presented on the basis of its process sections; thus, the parameters and
offgas stream characteristics presented in Tables 6-1 through 6-llb are
the combined characteristics of the individual streams in each process
section as identified in Chapter 3.
     As noted in Chapter 6, "Model Process Units and Regulatory Alternatives,"
of the report "VOC Fugitive Emissions in Synthetic Organic Chemicals
Manufacturing Industry - Background Information for Proposed Standards"
(EPA-450/3-80-033a), fugitive emissions are proportional to the number
of potential sources, but are not related to capacity, throughput, or
age.  Based on a qualitative assessment of both the emission source
counts and emission estimates (see Appendix C), a single model unit, as
presented in Table 6-12, was chosen to represent the fugitive emission
characteristics of the polymers and resins industry.  This model  plant,
however, was not applied to the poly(ethylene terephthalate) model plants
because those PET plants using the DMT process are already covered by the
                                  6-2

-------

H
" a
!! -
,g
tn
1

f^
•o
O
2
C
S

— 1 O i —
a. a.

LU
t*~\
o


•
r™~
1


2
ro







I

O
44
fi=
O
C















i
s.
19
*3 4->
O 0 J=
o o> cn
*Ii

c
o

£**
!^

0

cn • **-
c a> u
+3 ,2^
•a eg 3

= m
-
fo
o

V)
"S
a i_
a.


3~i
t2fiT



tn eT S-
«•»- >»
O fl "*»v
2s. cn
t-2E
a.



•— s.
2-S
J-E

0)
(TS
*^*. (J
e o 3
oo-o
"^^ 2
tn cna»
^" =






3 OJ
•M S-
Z (/>


C
0

u
tn
u
1 j


CO





g
>
§



o
cn




Cl
CM

O
d




d




"2
m




d






0
d








*j
e
o
CJ


tn
"« c

E «
cc a.

a. — «



*f CO




J3 .Q CM OJ
O tJ ZZ 3C
O O-
> > LTJ tn _
§cn 
.
cy

«T




T3
^
>
S




O
CM

CM
U-
to


to
^



o





o
S




O
o
U)
^





0
cn







c
o
u





"£ >
So
0
(3 O
E Q£
,
cn

LO


 <
cn P-.
cn d




T-t



«*-



v-4
^*



O
cn
CM





O
cn




o
S





to
CM








e
0




cn
c
•U i-
O JZ
€42
o =
a. u_

**


,
!





I
1


















i


















j
i







i

0)
cn
T3
0
v>
Product

LO
!



































CO
U)




cn
r-.
U)





CM
P*.
cn













VI
1



























•a i—
1 I
(Q C
i" 5
e oj
c
s 1
1 g.
o. cn

g ® o ^
(Q 3 CM C
Is10.".
^ 0>
tn cn ^ e
f— ^- Q_
** ** o o
a> e to s-
e a» 0) * a.
So jsi +*
o» oJ « S e
jc ex u o
•o w o
. •»-»
c o ^ u c
0 S- i- 3 U (U
O 3 * o --o js cn cn
*•».. uo u o o •« ITJ .«
cn rtspN. f to t. *o •-
o  3: t.
o^>- sm> fn^-
1- CLCM C •*- IT} -^ *O -^
*•* **— tn 3 tn i— o> i— o> ** o
i™ it) 0> O C t— U 2 U O) O
o s. u 3: o -os- s. e i_
a. t. cn4J tn a. o H^ a. * cL
tJ V» T— 13 LO I VJ PH. C P"*
a> ^ 4*1 c o * "•*••. o   Cst S- CD
ftj s- a> ca » cx^O tn tn tn 
6-3

-------
           en  — .bi
           2o 3 Hi
           Sg^m
              t*
                            O3 CM

                            cno
LU
CO


O
OO
oo
LU

co
oo
LLJ CO
o oo
O LU
o; o
cc o.

LU LU
   oo
oo   j_>
                                                                 3 VI

                                                                 5§
                       §
                                                            C O
                                                            .2£
                                          6-4

-------












LU
p-


0
a;
LU
CO
o
CO
CO
H- 1
LU
CO CO
CO CO
LU LU
O O
OO
a; a:
a. a_
O£ LU
LU =}
CO
co co
O LU
H-.Q;
1— Q-
co
!-• 1C
tt CD
LU I— i
j— n:
o

•a
 .c
O UJ

cn


o
CO

— "


s

u
o"
U)
CM


O
CM
cn





o







CO
cn
0





Q
Ol

§
- Polymerlzati
Reaction
CM



1




1






f



1





1
I


!



,







j






t


Material
Recovery
ro


CO
CM

Q>

> *C
CM CO

o en

o
CO •





UJ


O
§


o
s
^*

"*



s







CM
0






e
o


Product
Finishing
^


00
CM

4t

> <
CM CO

o cn
cn

o
CO





S


§
10
M


1
to*

u,



CM
CM






CO
o






c
o
u

Ot
o»
s
00
u
o
a.
to


00
CM


O

5..

O
2


o

^>
r-4


O
55


S

O JE JS
4J ^ . "* ,
** j= CM * ai
*P O. O
E E -a s e
i— O C 1 » IQ U
 1? *"
— s. i. "S
M~ Q. 3 t/l
t U • +J
C . +J . 3
e o vi c o» e
v> •*•»  - i) a. n- m
o w ui at
O W) Lfl i. -O
i_ o "a w> co (j .— o •- at s_
cfl — o o C cu -2 to
i- u. i. e e +J a>
>, O --0 0., • C u
CO •*»• "^" i- 0 — fCM
CM «3 0> <** V . O 01
i. a. CM o • cv f~ j? >.•£• o
•• I— «fc- >» *» U ** C
>>=»'>^'-^'e o c aioj w
4->^_ui3vi a> o > a. -a o c v> vi a> t— o
r— OU^*O V) i- f- 3 3 i— g a>
0) rt3 "3 C T- 
-------









LU


f=^
fy
U. CO
LU
CO CO
Z CO
O LU
t— 4 O
CO O
CO C£
1-4 Q.
LU LU
CO
CO «C
co =c
LU Cu

gto
Cu O
ce: LU
O 0.
LU a

CO
<_> LU
H-4 *T*
co
LU 
iQ *•»•»*
,32 w s
"~ ca


-
!>•
||
CU


AJ IB
£sr



VI
vet.
O « "^.

11


J
4kl
ee *j
CO 3
.SSI
5 ""2
i

*_—« 1
O  CM

CM CO
cn


O

"C
CM



cn
cn
to
CM
0


CM
UJ
O





in

m





o
o






1






VI

— §
"C H*
01 JJ
*J (O

cs.
X 01
^al

J
^
—«

_
> z <
p*. co in
to co 9-
in m


CO






CO
CM
o
tn
in



o
*•«




CO
0

10
CM





C




.
b.
0)
c














t
i
i
1

•%

£CM
> z
cr» ^N
P* CM
rv. CM


CTt
in
ro
— <



s
i
IO


I
CM
PO




tn

2





VO
O




*o
c_
a>
c





c
o

"Jo
N
^ §

>t O
"o *°

j
r\j

i
i

(
i
1



,
i
t




i
i
t
{



i





i
i
i

i
i
i





i






i










— ai

at o
JJ U
£ 3C

ro

-§>
CO

O M

in CM
CO VO


o
VO
in




in
CO
in
in
CM



2
tn




CO

in
ro
CO





CM
OJ






1









cn
c:
3 t/1

o "c
t. ••-
Q_ LL.

^ 	 :

tn
to
C_5 C-
CM CO
O Ch
o e^
cn

cn
O






O
If)
g
o
-


0
^"
CM




CO
CO

tfl





in
O
o

	 ' , , , „„,!,




|




a;

re
t_
o
CO

3
T3
O
O-

in




























CM
in
r-
^
O
in
m





PO
CM









I.H"











!j

h—
O















>, « —
*— *j tft ai
ai «j j-i TS
> ID 4J C 0
— • c ai g
4-> ><, Q) >
U C > 0)
a> 4^ L. f
a. tfi a) u *j
in 3 cn N
OJ -0 ^^>, C
* "~ y c T3
u c ui « ai
SO i- -S
. -O 0) 3
>, -o .c >—
O) 4J 4-1 O
t- in u c
01 03 3 "O •»-
a. JD -a c
! ,' • : '»'• " ' ' O TO 4-»
to • £_ O
3 • fi1 ^J" C
O r— 0) * C (L>
x: o > -3 ai t.
gJJ J_> i=
= tj (0 »— *
* CM O OJ OJ f- OJ
U O. (- O t.
T3 Wl 4J O
c c. a> (/I — «Z
ro o L. re o>
t- ^ O) I_
• C in Q)
o ^" re ro (-^
CM a) a> * o
*j >, — » in "
• ffl  a) cr
Q- m a>  o e:
CJ U t. U O
TS 3 t-
g- Z £ i_'S
O -B *O Jj E *S
*-* •»- CM O rtJ 5
tn (O QJ
o; c -a QJ t- a>
J9 O C t- 4J >
o re oo —
O C 4J
*J O) rO O i-:
•O S ""• W 3S 3
ai a> o (o — i4-
3 1J QJ t- 3 <3J
 t- J3 (- O J3
in OJ OJ
re at o e c o
C 4-> >, (0 *J
U. n- f— 4_>
• 4-> -O O i— "O
T- +-> , c: O E E [_
(/> rere t_3a>va>
Q. "O (/) C C/l *O
* 31 in ^ ^~
O >> UJ nj 4-> -Q tn
O C E 3 OJ C
c.o*o 4^ * H33;o*jo
>) L. j_» o
•^^ » (O 3 U. T5 ^~
cnCO t) » j-> c QJ E OJ
ts **^ c co 1/1 (Q ^ a) i-
L. u. ^* 4-1 re
>, in •• o • -o cs — • a>
"^f^UO E«C c J2*
cn«p*» o (a to .w -w
CS - «*- e ajreajore
O C C_> (Q Vi QJ > C ^~
LO CM — u o j-) a> ret- co.
M QJ >— 1 £ ^ « Q) 4J (1) (Q
Q.CNJ 71 *•* (O i— CO CT) O QJ
• • i— in 4J QJ (- jr
•• co trt a> rec_L.3»4->
>) c (. •• . 3:t-"O O Q.-^^
j-i — 3 in o >, **- cnj_>
•p-reoc c, **- >, u t-c3
O t. Z O (O t. C V) O *— O
reH— f— r— QJ 4J OJ fll -1-1 J^ C-
Q. cn4-> 3 E « cn g (j yj.cn
Oin*"T3 QJ 4J ^} QJ 4J 4Jc — ce t_-~(_
C O t- O £ O O"+J
<«(-o) • — c CD -- or 4->
i— . Q_ &TD ^ OJ »*• O >*"- (JTD
Q- O 1- QJ CTI g £3(1)*
«- re E re.— -a Orec
a> rec to QJ 4-1 to t_4_> i_ i_ u ro
•o • 4J(o in > ore -j o 4JQ.O —
oo oxj *t a; H— ja i/) ^ 4/7 *--^— ^
EZ^-cn re^ o "a a*«*-
6-6

-------
o

u_

GO

O
t-^
GO
GO GO
>-> GO
21 LU
LU U

GO C£.
GO a.
LU
O >-
o tt:
a: a;
a. rD

Qi GO
O
U_ LU
   GO

o a:
1—I Q.

GO a
o

           (O   3 .e
           t. c_> u as
            So a; •—
            > ^- 4)
           <   03
O
O
LO
 I
 CD
           0=03
           c. o o -ts
           *J— » O
          15-
20
LD
"O t-
o ^- cn
ro
O


CM



CM
O
S

8

LT)
CO


S
«s-

o





*i

o





en
*j -2
u .e
3 i/l
•o —
o e
Q. lT


-------










LU
3= .


LU
CO
o
CO CO
CO CO
l— i LU
S o
ryr
co a.
CO
LU Z
0 0
O >-«


1
ceo
oco
LU
CO CO
O 
I— o-
O i— i
^c *
cC LU
3T Q.
O C3
s*
<

Q.


LU




«3
1
^O
cu
3
H^









1

» o
»— (TJ
O O)
a. ce
CM

0

C
c c?
"O (O (-
>UJZ*<
co t
(TJ C-
•^ O)
t. >
0) O
J-t O
en

^.
CO

s
"*- ITS
>(/•><
CM <^ CM
o r— CM
en


CM
O


s



§
CM

««

(U
o
o
CO

CO*




to



in
*"






UO






"
c:
o
CJ





c:

O f
11
,;

]
'


i
!





1

!









,







j



,
1






i







;


In

c_
o
VJ
*J
(J
3
•o
O
t_
Q-
LO*
































O
CO



MX
CM






0
CTt
CM

















_J
|













I
-M
0>
U
0)
•^

or
o
Q)
ce
i—
£
0)
(0

^
c
03
C
o
4J

1
§
its
N
"C
1 g
f-» >
0. C-
0)
• J=
e «-)
o o
 UD
Q. E
0) 0) T3

CU 3 • 
•*- in C • •
ex 'n c x c: c

o *C *tl ^ x x
O 4J OJ O  a; -— x: .c
^t^ C « C U O 0
en co en u u U
co *-* ic >, >,
U. (0 0) « U U
u o •• o ec «r
>> ro C. O T3 • X S*
en cu *— o to • •
ca •• >• >*- TO Lrt vn
CO C C CM CVJ
Ocn».L.o vi o ITJ •
en a) i— < u t- ..c
•• *•» « c c c t-
•-viv) i^ E 41 0) OJ J->
4J^j 3 co en u


ITS t- 4) C T3 4J
r— CU Q.T3 O C V) V) V) <*-
OO O *J UJ «— 
-------













LU
— L.




0
ate
U- CO
CO
CO LU
Z 0
o o
CO Q.
CO
>— i CO
s§
co z
CO t— i
LU I—
0 Z
0 0
0.
LU
^^ ^^
O LU
1 1 (^
CO H^
O CO
I-* >-
1— _J
CO O
1— 1 Q.
LU (—
1— O
0 tx
rv «^
— *

o o;

z °
< |*
a. co
—1 OS
LU C_)
Q









1
IO

(U

.a
ro
i —








































__^
g
v)
J
.g
QJ
•M
1
+J
V)
UJ
=
o
~

1
LU
g

I

£
4J
§
0
5

1
i
e
o
u
=

























I— 4J
(J (J OS


*
o

4J »*

|j


O
o


C >t
0) 4J
t. to
>) i-
4J — (U E
CO < (J ra
 3 V)
O) O "O -i—
•u) O O C
fO OJ C. ^*
3T a= a. u_

J fxj







































_^

2




en
o
CM





CM
0

























t/>


2























,
4J
O
2
&
c/t
4J
O
s.
^
4/1
-C
4J
e
•*>>


o

^
0
e

gj
d)
VI
c
o
u
V)

<
0)
5
Ol V)
V) Ut
a. >,

O it}
>,^ -a us
-*^ 1C C
Oi * fO 0)
C3 00 £

to u_
£ £ J5S 1

e3Jf .. ^ **"

LO CM i- U 0 O)
r». ITJ a; <— " 4.)
t. a.CM «
•*(/)[/) O)
>, C V) U •• -0
4J t- VI 3 VI
•^- « « o c c

-------
   oo
LU OO
JC LU
o a.
a:
u. z
   o
CO I—i
Z= CO
o z
H-l LU
oo a.
co oo j

E oo
oo O
CO HH
on (s
a. LU
«*£
oo>i'
o oo
t-< O
I— Q-
oo
l—< LU
C£. -Z.
LU LU
I— C£


§ OO
CJ O
   a.
          en  ^- *j
          283-g,
             i


             S .

             I-
             u
  S1 -w
  •^ QI U
  4-1 3 V)
  VI *
    C t-

  o-S-5
  ; LU

  : ca

  •21   2<
LU     —

*f
o
£o
CM CO

•a
s
i
i
i
m
CM



cn
«


\r>
**"•
o


CM

CM


en



CD

C




C
o


(t)
N
-§


5-t tj




N

t
t

!

i
i
i
i
i

i

i


i
i
i


i
i
i


i
i

i
i
i


i
i
i










~ £
•^ D

01 O

•9 1)
SI a:

^
CM CM
CM CM

o c. -o
S1^ CJ C.
*c o —
tn tO
o en CM co
° s ~ ii
*-••. i-*
O £



CM CM

to o

to co
tO CM

S CO
to in
en co


*r to
in en

~

CM tO
*O CO
-* CM

2 CO

CM O^

• S-
4J IV
e 4-»
0 1=
CJ i— t









e
u ^
2 V)

O C
U •—
Q. U.
.•" • • "!'!i:i|i|i|i" • • JLI'
^
CM CM
CM CM

§5 Is
»-« CO ^N CO
o en o en
en en

CM — <



CM CM

O O
cn O
CO CO
*~t f-l

o o
co en

to to


r- O
LO PO



0 0


CM CM

CM — »

4J Q)

0 ^C




&
(Q
S_
O


w
u
3

o
t_
a.

03
























CM


0




en



















to

5
£

































L.
>>
en


tO PO
.. (rt
>i C
.»- /o
u t.
rtj h-
o.
(O V)
U V>
a
4-J U
c o
a t_



•— 0
« .
££
                                                                                           ~-   —  n
                                                                                                  i   i
                                                                             ra c
                                                                             4-> (O
                                                                             O 4J
                                                6-10

-------















LU
1 —
S 00
O 00
Qi LU
Lu 0
O
00 OIL
•Z. Q_
O
00 O
OO »-H
I-H 00
LU LU
CL
OO 00
OO Z3
LU 00
0
O ID
CL. l— l
00
O£. 1
O Z:
Ll_ i— i

00 LU
l-H LU

00 5-
f-t I—
OS 00
LU >-
0 O
<=c LU
O CD
1— Q
Z Z.
	 1 D_
0- X
LU
	 |
| ! J
Q
O
2Z


rr\
1
^>O


•—
(O






•







1
i.
0)

**



















I
U
(/I
1
u.
Of
o
J

LU


c
o

(A
UJ
o
o
•o
(U
2
e
0
o
=)


(U

o
•£
0
t.

0*3 -C
o o cn
> 0)i-

aB3

.
o
5a«
*tf)
0 •

o


C »*4-
i- O) O
•M 3 
— E


oT
(Q
^^."u
e o 3
O O "O
^-> 0
trt t-
01 cno.
Sl~o,
3



















UJ E



O tn

5 £

2 tO




e
o
CO
V)
(/I

0

Q.

in



0)
(U
>*z t.
4.?tJ*^-
CO «C <
in ro CM
P-. tn P*.

C3 CD CO
en


u>
in





CM

CO




(O
CM
O
•-I






•-H



cn





CM
O








5
^




(A

(TJ C
•*-• o
0) *J
•W fl3
11
(Q i.



T-H

^
CVJ



0)
0)

> OO «t

CO ^ CO
co i-4 en
in co


in
r-.
***
i— i



in
ro

—«
to



u
in
en
vn





>*H
cn



CM
CO





2t
f-H








5
c






la
N
Z|




CM



1
i
i





i

i



i





i
i

i
t
i



i
i
i







i


i
i
i






i
1







i
i
i







•sfr
T- (U
s- >
at o
4-* U
ITS  at
e 4-*
o c
U i-t







«f
u ^
3 > CO t-O
^^ a) P*.
cn of — •
to ->.
c cj
CM cn f- uo
p- »0 — o <^
a. s- cn-M
rt3 Ul Q_ C •*-
*J O (U (13 O
C O CX £_ CJ
(O i. 01
*— CL >j CL*O
Q_ 4? o S-

i— O (J r— "O
•S - 0.2 S
o o 
-------
                O U Ol
                O t— OJ
                   o r
 LU
 CO
 z
 ore
 I-H CO
 CO CO
 co uj
 UJ CC
    0.
 CO
 CO I—
 U-IS
 CJ O
 o

 a. LU
 CO I
 o
 CO O£
 I— I LU
O 2E
et LU
a: _j
eC >-
3COC
O I—
    LU
             8
°I
CD
—1 O
a. a.
LU
a
O
t-i
 I



(U


(O
             r—   '^. O
                         CO


                         o
                                                      0°
                                                      o -a
                         f— >»  — C
                               >, O
                               »— eg
                               O (U
                               CL. c;
                                                      no     r-i
                               6-12

-------




LU
LU
I

oo
o oo
i— i OO
1— .04
00 O


.LU Q.
0 LU
H—



•=C«=C

a- 1—
_J Q.
LU LU
o cc
O LU
sc 1 —


^
(0
1
r— 1

10



3 O>
tn »^
tn t/i
Q) 0.
a.

"
a.
€ <->
O) O
H-

tn
tn c
O) •*-
O *0
O 4-
a.



F— 4J
«3 C
i2 S!






tn *
a> c s-
£ 2"oi
Q. H" SI



p- i-


o en
t— E





^ to



c
o


u
8
l/l
in
0)
U
2
a.



^°


i
i
t

ou
O
1
I
t





i in
1 CSJ



i i





i i
i i
t i





u>
o





1-4










tn

o ro
O O



tn tn
O 3
3 3
C C
C C
o o
o o




in o
S- *t— N
O) 4-> T- C
•M (O i- O
E (Q E *J
a- >, u
S 
81.
r— QJ
^ N
O *^
u- u
a>
•S- f™
t- 0
tn
c o
^-^»
t/l  *J
u
C R3
O (W
1- i_

u o
u- *»
"C N
Ul QJ

i?
4-» 0.

o s

c
o t-
^. 0
(OH- •
u LIJ tn
3 Q_ QJ


<^ 4J (I)
C •*• tn
O v>
U 0 C=
O) o *^
^ tt; in
»« C i- O
d. -t— > Ul
O "~ S "«
O CO
O "O O ^- H-
L. (O C ^- UJ
>» - to 4-» L. a_

CB *^ 3 *- >»
>> •• O O r~ t—
••^. in i. o u a> tn
01 --i m r*. o.  4^
o f*- T- t- o en >
*— t  <-* . QJ C
s- Q.<\J B o ^r
• • tn tn tn +J •»—
>>ct/)^.» tntoJZ
•^ «a o o c *^> s-
<_> S- CJ 3= O *J S_ O
4-> « I
 a. c i- f~ >»—
u tn «^ ^ CL»— s-
> Ct"O O ^- C
Q. 4-» 0 S. E 0>i-

r- O*0 — -0 WH-
aj rtj 
-------






LU
LU
i f
>-

1 ~*
LU
^
_l
O
CL.
LU


1—

§co
cc
l-V 111
CO
CO i— i
!S^ TT
O I—"
l-t LL.
CO
CO Q
2Z LU
LU
LU
CO _J
CO O,
LU I— I
0 t—
0 _I

a. 2:

CC C£J
O 2=
U. I-H
CO
CO =>
o
t— i co
1— CO
CO LU
i— t O
on o
LU Q;
o

•*-*
IB
E
+3
(A
LU


(A
O
(A
l/l
I

O
>•
0)
*iS
4J
g
u
=




01
*o
4J
1
5























S_
rtJ
3 f~
o u a>

^1=*

c
o
4J

(A •
(Lit
O


O
u en
=> •*-
U) (A
tn CL
a>
CU
»
|o"


S£
O Q
0 i-
^



ra c
"M  l-
M a.
tn en
*- ^c en
E E



u- — .
o w
tu's
!• (Q
5 £
^2±;






c
o

u
CO
V)
VI
 4J
C C
3 8




C
(A 0
fO C 4-*
•^ O Ifl
1- ^- M
5 «3 To
(O t. O) •»-
o. >> o
S 
to *tl
a> CD
t.
S> 4^
G O
•*- CA


0) *
V) tA
V) L.

W
C T-
«3
C S
•r*
4> CO
*-» c:

0° ^S
Q "O O C

•^.co o
o> — 3 -a
CD Lu TJ =
S_ .. o O 0»
>» q o i.
•^» oj r*. n r

CO ^M
O OJ t— O tA T-

i- Q.OJ g v-
•• W) CA C
>> C (A •• (A (O
*J t—  fO M
(O IA Q_ t- f— T-
tJ IA "O Q. W
O> i- C OJ
•WOO) 0 r- £
C O CL CJ OJ >,
r— O, >» 0.13 O *O
0. -W S» £ o.
M— -r- (Q QJ
-— O U -0 tA 5-
QJ ra c •— Q.
•o • Q. ra j=
0 O (0 4J H- (0
E a: c_J co ra
6-14

-------
       Table 6-12.  FUGITIVE VOC EMISSION MODEL PLANT PARAMETERS
Equipment Component
Equipment Counts
Uncontrolled
Emission Rate
(kg/hr/Source)
Valves
Vapor service
Light liquid service
Heavy liquid service
Pump seals
Light liquid service
Heavy liquid service
Compressor seals
Safety relief valves
Vapor service
Flanges
Sampling connections
Open-ended lines

402
524
524

29
30
2

42
2,400
104
415

0.0056
0.0070
0.00023

0.0494
0.0214
0.228

0.104
0.00083
0.015
0.0017
TOTAL UNCONTROLLED FUGITIVE EMISSIONS:  151 Mg/yr
                                6-15

-------
 fugitive emission standards for the synthetic organic chemicals manufacturing
 industry (SOCMI) due to their methanol by-product production and those PET
 plants using the TPA process use only "heavy" liquids (i.e., ethylene glycol)
 and solids.
 6.2  REGULATORY ALTERNATIVES
      This section defines various regulatory alternatives or possible
 courses of action EPA could take to reduce VOC emissions from the polymers
 and resins industry.  These alternatives provide a basis for determining
 the air quality and nonair quality environmental impacts, energy requirements,
 and costs associated with varying degrees of VOC emissions reduction and
 represent comprehensive programs for reduction of emissions.
      The regulatory alternatives were developed in two basic steps:
 (1) determination of baseline  control  and (2)  determination of more
 stringent levels of VOC control  based upon applicable VOC control  techniques.
 In addition,  the regulatory alternatives were  developed  by applying
 baseline control  and more stringent control  to emissions from  process
 sections within  a single process line  for each model  plant.
 6.2.1   Baseline  Control
     Baseline control reflects the  level  of  VOC control  that is  likely
 to be employed in a  new plant in  the absence of the new  source performance
 standard.   Its determination is  difficult.   As discussed in  Chapter  3,
 not all  States regulate  VOC emissions  and the  States  that  do regulate
 these emissions  have  regulations  of different  stringency.   In addition,
 the level of control  employed by  two plants  producing  the  same product
 with the  same basic process in the same  State  may vary due to differences
 in  detail of the  process.  Finally, the actual capacity  of the new plant
 may determine the level of emission control employed.  Given these difficul-
 ties, the following methods were used for determining baseline control.
     6.2.1.1  Process Emissions.   For polyolefin plants, individual
 process VOC emission streams from the plant descriptions in Chapter 3
were identified as the streams most likely to be controlled in the
absence of a standard by the following conditions:
         - Intermittent streams;
         - Continuous streams of  mass flow rates greater than
           91  Mg/yr (100 tons/yr); and
         - Exceptions to the above based  on specific  process
           information pertaining to the  plant  process.
                                  6-16

-------
    The streams identified as most likely to be controlled were
assumed to be presently flared.  Intermittent streams are generally
controlled by a flare for safety reasons and large continuous streams
are likely to be controlled, typically by a flare since flares are
acceptable control devices in States with VOC regulations.
     For the polystyrene and the PET model  plants, a different basis was
used to identify the regulatory baseline.  Both industries currently use
various recovery or combustion technologies that achieve varying levels
of VOC emission reductions.  For the crystal or impact polystyrene
segment, baseline control reflects the use of condensers on the material
recovery emission streams (i.e., devolatilizer vent stream and styrene
condenser vent stream).  As current industry practice in controlling
emissions from expandable polystyrene plants varies considerably between
plants, it was assumed for analyses purposes that there would be no
baseline control.  For the poly(ethylene terephthalate) segment, baseline
control reflects the use of spent ethylene glycol condensers to recover
the ethylene glycol from the polymerization reactors and reflux condensers
on the esterifiers, if a low viscosity product is produced or if a high
viscosity product is produced using a single end finisher.  Baseline control
for a plant producing high viscosity PET using multiple end finishers reflects
the use of spent ethylene glycol spray condensers on the initial end finisher
only and distillation columns on the esterifiers and cooling tower.
     6.2.1.2  Fugitive Emissions.  Fugitive emission baseline control,
which applies to all model plants except the PET plants, assumes that
75 percent of all gas safety/relief valves and sampling connections,
most of the open-ended lines, and 0 percent of all other fugitive emission
sources are controlled.  These assumptions are consistent with those
made in the analysis of fugitive emission baseline chosen for SOCMI.
6.2.2  Control Techniques
     Process emissions may be controlled by flares, thermal incinerators,
catalytic incinerators, boilers, condensers, absorbers, and adsorbers.
Fugitive emissions may be controlled through leak detection and repair
programs and equipment, design, and operational requirements.  These
control techniques and their associated emission reductions, which are
discussed in detail in Chapter 4, "Emission Control Techniques," were
examined to determine technical feasibility when applied to each model
                                  6-17

-------
plant and the possible level of VOC emission reduction.  This information
was then used to develop the most effective control  options.
     6.2.2.1  Process Emission Control Devices.  Combustion devices,
such as flares, thermal or catalytic incinerators, and boilers, are the
most prevalent emission control techniques in this industry, especially
in the polyolefin segments.  Combustion devices are not typically used,
however, in either the polystyrene or the poly(ethylene terephthalate)
processes for VOC emission control.  Some combustion of emissions from
expandable polystyrene plants does take place.
     The preference of one combustion device over another is dependent
on the waste gas characteristics of each vent stream.  Flares, the most
commonly used control technique for process offgases, are universally
applicable in controlling upset emissions from polyolefin plants.  They
are capable of handling these emergency releases, as well as low volume
continuous vent streams from these processes.  Thermal incineration in
incinerators or boilers also is applicable to polyolefins vent streams
and is second only to the flare in its frequency of use in the industry.
Thermal incinerators can be used to control continuous streams with a
wide range of concentrations or type of VOC.  Boilers are used as control
devices for continuous flow, high heating value streams.  Catalytic
incineration may be used for continous flow, low heating value streams.
Its lower operating temperatures requires less supplementary fuel than
thermal incinerators to achieve the same level of VOC emission reduction.
     Condensers, absorbers, and adsorbers are sensitive to changes in
VOC flowrate or concentration, and VOC removal efficiencies decrease as
the VOC concentration in the offgas decreases.  Thus, these devices often
are used primarily to recover process materials rather than as an emission
reduction technique.  These devices also are more chemical specific than
other emission control techniques.  Absorbers and adsorbers are not widely
used in the polymer and resin industry and, thus, are not considered in the
regulatory analyses.  Condensers, however, can be used as emission control
devices in the case of polystyrene and poly(ethylene terephthalate).
     6.2.2.2  Fugitive Emission Control  Techniques.   VOC fugitive emissions
control techniques include leak detection and repair programs, and
equipment, design, and operational requirements.  These programs are
discussed in Chapter 4.  One combination of these control techniques
(see Table 6-13) was chosen as the regulatory alternative for controlling
                                  6-18

-------
      Table 6-13.  CONTROL SPECIFICATION FOR FUGITIVE EMISSIONS
                          UNDER REGULATORY ALTERNATIVE 2
Source
Inspection
 Interval
Equipment Specification
Valves
  Vapor service
  Light liquid service
  Heavy liquid service
 Monthly
 Monthly
 None
    None
    None
    None
Pump seals
  Light liquid service
  Heavy liquid service
 Monthly9
 Weekly  visual
 None
    None
    None
Compressor seals

Safety relief valves
  Vapor service
 None
 None
    Controlled degassing
      vents
    Rupture disks  on
      relief valves^
Flanges
 None
    None
Sampling connections
 None
Open-ended lines
(purge, drain, sample lines)  None
    Closed-purge
      sampling
                     Cap
aFor pumps, instrument monitoring would be supplemented with weekly visual
 inspections for liquid leakage.   If liquid is noted to be leaking from the
 pump seal, the pump seal  would be repaired.
     safety relief valves,  instrument monitoring would be necessary after
 an overpressure relief.
                                6-19

-------
 VOC fugitive emissions from the polymers and resins industry.  This
 regulatory alternative was chosen for consistency with the fugitive
 emission regulations for SOCMI.
 6.2.3  Regulatory Alternatives
      The following sections present the regulatory alternatives,  including
 baseline control, for a process line in each of the model  plants.   The
 regulatory alternatives are presented on a process section-by-process
 section basis.  Determination of baseline control  (Regulatory Alternative
 1) was described previously in Section 6.2.1.  Regulatory  alternatives
 of increasing stringency were developed, in general, by first implementing
 fugitive emission controls for each model  plant (except the PET model
 plants).  The fugitive VOC control  program is known to be  reasonable
 and the best technological  system for fugitive VOC emissions based  on
 analyses already performed in the SOCMI NSPS.  Therefore,  this control
 is retained in all  succeeding regulatory alternatives.  This avoids
 unnecessary combinations of the same process controls with and without
 fugitive emission control.
      Additional  regulatory alternatives of increasing stringency were
 then  developed on the  basis of controlling each  remaining  process section
 with  potential  uncontrolled emissions until  all  process sections within
 the process line were  analyzed.
      In  developing  the regulatory alternatives for  the polypropylene and
 polyethylene  model  plants,  flares and incinerators  (thermal  and catalytic)
 were  considered  the control  devices  most likely  to  be  used to  control
 continuous  streams  because  of  the wide  range  of  applicability, current
 use in the  industry, and  favorable costs and  were used to  develop the
 regulatory  alternatives for these model  plants.  Boilers, which some
 plants may  choose to use, were not costed  specifically  for any regulatory
 alternative because not all  polypropylene and polyethylene plants have
 boilers  or  the need for steam.  However, if a plant has a boiler and a
 need for the  steam, boilers  are more cost-effective than either flares
 or thermal  incinerators.  For control of intermittent streams in the
 regulatory alternatives, only flares were used because they are the  only
control device considered feasible to control intermittent  streams.
Combustion devices were also used to develop all  but one of the regulatory
alternatives for the expandable polystyrene model plants.
                                  6-20

-------
     In developing the regulatory alternatives for the crystal or impact
polystyrene and poly(ethylene terephthalate) (PET) plants, recovery
techniques are more likely to be used than combustion techniques on
streams containing styrene monomer or ethylene glycol.  For crystal or
impact polystyrene plants, condensers are the most likely recovery technique
to be used and are used to develop the regulatory alternatives for this
model plant.  For the PET plants, ethylene glycol recovery systems are
the most likely control techniques to be used in the industry and are
used to develop the regulatory alternatives for this model plant.  A
flare was also analyzed for control  of the methanol  vent stream found in
the PET/DMT plants.
     6.2.3.1  Polypropylene - Continuous, Liquid Phase Slurry Process.
(Table 6-14.)  By applying the general criteria outlined in Section 6.2.1,
the process emission streams identified under baseline control (Regulatory
Alternative 1) are Streams B and C (reactor vents),  Stream D (decanter
vent), Stream E (neutralizer vent),  Stream F (slurry filter/vacuum
system vent) and Stream 6 (diluent separation and recovery).  These
streams correspond to the polymerization reaction section (Streams B
and C) and the material recovery section (Streams D  through G), and
are assumed to be controlled by combustion (i.e., a  flare).  Based on
a 98 percent VOC destruction efficiency, the baseline control  of pro-
cess emissions from a process line is equivalent to  an annual  process
emission reduction of about 1,690 Mg.  Fugitive emission baseline
control, as discussed in Section 6.2.1, would reduce uncontrolled
fugitive emissions from 50 Mg/yr to  approximately 35 Mg/yr.  Under
baseline control, an annual  emission reduction of 1,705 Mg is achieved
or approximately 89 percent of the total uncontrolled emissions from a
process line in this model plant.
     Regulatory Alternative 2 represents the application of fugitive
emission controls in addition to baseline control.  Fugitive emission
controls result in annual fugitive emissions of 16 Mg per process line.
Under this alternative, an annual emission reduction of 1,725 Mg per
process line is achieved or approximately 90 percent.
     Regulatory Alternative 3 includes the same control  as in Regulatory
Alternative 2 plus the combustion of the emissions from the product
                                  6-21

-------
                    Table 6-14.   REGULATORY ALTERNATIVES FOR THE
                         POLYPROPYLENE LIQUID PHASE PROCESS
Regul atory
Alternative
1
(baseline)
2
3
4
Process Emissions
Process
Section(s) Control Fugitive
Control! eda Technique Emissions
PR + MR Flare d
PR + MR Flare e
PR, MR, plus Combustion^ e
PF
PR, MR, PF, Combustion^ e
plus RMP
Annual Emission Reduction
Per Process Li neb
Mg/yr Percent^
1,705 89%
1,725 90%
1,852 97.0%
1,856 97.3%
aProcess sections include the following:
 RMP s raw materials preparation
 PR  = polymerization reaction
 MR  s material recovery
 PF  s product finishing
 PS  = product storage

bRepresents reduction in annual  VOC emissions from the uncontrolled level.

CBased on total uncontrolled emissions of 1,908 Mg/yr (1,858 Mg/yr process
 emissions plus 50 Mg/yr fugitive emissions).

fugitive baseline control.

eControl  equivalent to Regulatory Alternative 2 (see  Table  6-13).

^Combustion devices may include  flares, thermal  incinerators,  catalytic
 incinerators, and boilers.
                                      6-22

-------
finishing section.  This alternative reduces annual emissions by 1,852 Mg
per process line or approximately 97 percent.
     Regulatory Alternative 4 includes the same control as in the third
regulatory alternative plus the combustion of emissions from the raw
materials preparation section.  This alternative is somewhat more stringent
than the third, reducing annual emissions from each process line by
1,856 Mg or 97.3 percent from uncontrolled levels in a process line.
     6.2.3.2  Polypropylene - Gas Phase Process.  (Table 6-15.)  As
stated in Table 6-2, the mass flowrate of emissions from the material
recovery section is about 530 times smaller than that from the polymerization
reaction section.  The intermittent emissions occur less than 1 percent
of the time, and is, for safety purposes, most likely to be sent to a
flare during emergencies or process upsets.  Based on information from
the one company that uses this process, the emissions from the material
recovery section are controlled by the same flare as the emissions from
the polymerization reaction section.  Thus, the baseline (Regulatory
Alternative 1) includes control of both the intermittent and continuous
streams by a flare.  This baseline represents a 95 percent reduction
(1,267 Mg/yr) of total uncontrolled VOC emissions from a process line
in this model plant.
     Regulatory Alternative 2 represents the application of fugitive
emission controls in addition to baseline control.  Fugitive emission
controls result in annual fugitive emissions of 16 Mg.  This alternative
achieves an annual emission reduction of about 1,290 Mg from each process
line, or approximately 97 percent.
     6.2.3.3  Low Density Polyethylene, High-Pressure Process.
(Table 6-16.) The baseline control (Regulatory Alternative 1) for this
model plant sends the non-emergency, purge-type intermittent emissions
from all process sections to a flare.  Under baseline control, process and
fugitive emissions from the process line would be reduced by 594 Mg/yr
or about 81 percent of the total uncontrolled emissions in a process line.
     Regulatory Alternative 2, which represents fugitive emission
controls plus baseline control, results in a total annual emission
reduction of 609 Mg per process line or about 83 percent.
     Regulatory Alternative 3 represents Regulatory Alternative 2 controls
plus the combustion of emissions from the product storage section.  This
                                  6-23

-------
                   Table 6-15.  REGULATORY ALTERNATIVES FOR THE
                         POLYPROPYLENE GAS PHASE PROCESS
Process Emissions
Regul atory
Alternative
1
(baseline)
Process
Section(s)
Control! eda
Polymeri-
zation
Control
Technique
Flare
Fugitive
Emissions
d
Annual Emi
Per
Mg/yr
1,267
ssion Reduction
Process Line&
Percent^
95%
              Reactor (PR)
              plus Material
              Recovery (MR)

              PR plus MR     Flare
1,287
97%
aProcess sections include the following:
 RMP = raw materials preparation
 PR  = polymerization reaction
 MR  = material recovery
 PF  - product finishing
 PS  ~ product storage

bRepresents reduction in annual  VOC emissions from the uncontrolled level.

cBased on total uncontrolled emissions of 1,328 Mg/yr (1,278 Mg/yr process
 emissions plus 50 Mg/yr fugitive emissions.)

^Fugitive baseline control.

eControl equivalent to Regulatory Alternative 2 (see Table 6-13).
                                       6-24

-------
     Table 6-16.  REGULATORY ALTERNATIVES FOR THE LDPE HIGH PRESSURE PROCESS
Regulatory
Alternative
1
(baseline)
2
3
Process Emissions
Process /
Section(s) Control Fugitive
Controlled9 Technique Emissions
Purges Flare d
Baseline (B) Flare e
Baseline Combustion^ e
plus PS
\nnual Emission Reduction
Per Process Lineb
Mg/yr Percentc
594 81%
609 83%
664 91%
              Baseline, PS
              plus PF

              Baseline, PS
              PF plus
              Emergency
              Vents from
              Reactor and
              Separators
Combustion^


Combustionf
678


704
93%


96%
aProcess sections include the following:
 RMP = raw materials preparation
 PR  = polymerization reaction
 MR  = material recovery
 PF  = product finishing
 PS  = product storage

bRepresents reduction in annual  VOC emissions from the uncontrolled level.

cBased on total uncontrolled emissions of 730 Mg/yr (692 Mg/yr process
 emissions plus 38 Mg/yr fugitive emissions.)

•^Fugitive baseline control.

eControl  equivalent to Regulatory Alternative 2 (see Table 6-13).

^Combustion devices may include  flares, thermal incinerators,  catalytic
 incinerators, and boilers.
                                        6-25

-------
alternative reduces total uncontrolled emissions from a process line
by 664 Mg/yr or approximately 91 percent.
     Regulatory Alternative 4 is the same as Regulatory Alternative 3
plus the combustion of emissions from the product finishing section.
Total uncontrolled emissions from a process line under this alternative
are reduced by 678 Mg/yr or about 93 percent.
     Regulatory Alternative 5 is the same as Regulatory Alternative 4
plus the combustion of emissions from the emergency reactor vent in
the polymerization reaction section.  {Note: If the high-pressure
reactor vent, Stream B, from this process is sent to a flare, a parti-
culate polymer removal system is a prerequisite.  Based on available
information, only one company has a particulate polymer removal system
technology that can handle high-pressure emergency vent gas.  In
general, it is expected that the emergency vent from the reactor will
be released through a pressure relief system and vented to the
atmosphere.  The exhaust gas contains suspended particulate polymer
that cannot be directly sent to a flare system because of safety
considerations.  Unless the particulate removal system technology
becomes available to most companies, Stream B is not likely to be
controlled.) Total uncontrolled emissions under this alternative are
reduced by 704 Mg/yr per process line or about 96 percent.
     6.2.3.4  Low Density Polyethylene Low-Pressure and High Density
Polyethylene Gas Phase Process.  {Table 6-17.)  The intermittent streams
in the raw materials preparation, the polymerization reaction (except
decompositions), and the product finishing sections and Stream J (product
discharge vent) in the product finishing section were determined to be
controlled by flaring under the baseline assumptions.  Annual emission
reductions from a process line under baseline control (Regulatory
Alternative 1) are 1,725 Mg or about 95 percent of the total uncontrolled
emissions from a process line in this model plant.
     Regulatory Alternative 2 represents the application of fugitive
emission controls in addition to baseline control.  Fugitive emissions
control results in annual fugitive emissions of 23 Mg per process line.
Under this alternative, an annual emission reduction of 1,762 Mg
per process line is achieved, or approximately 96.1 percent.
                                  6-26

-------
         Table 6-17.  REGULATORY ALTERNATIVES FOR THE LDPE LOW-PRESSURE
                            AND HOPE GAS PHASE PROCESS
Process Emissions
Regul atory
Alternative
1
(baseline)
Process
Section(s)
Controlled3
RMP (Inter
mittent
Control
Technique
Flare
Fugitive
Emissions
d
Annual Emission Reduction
Per Process Line*5
Mg/yr Percentc
1,725 94%
     2

     3
streams only),
PR (Non-
emergency
i ntermi ttent
streams only),
and PF

Baseline

Baseline
plus contin-
uous streams
from RMP

Baseline,
continuous
streams from
RMP, plus PS

Baseline,
continuous
streams from
RMP and PS,
plus PR emer-
gency stream
Flare

Combustion^




Combustionf




Combustion^
e

e
1,755

1,762




1,766




1,770
96.1%

96.4




96.7%




96.9%
aProcess sections include the following:
 RMP = raw materials preparation
 PR  = polymerization reaction
 MR  = material  recovery
 PF  = product finishing
 PS  = product storage

bRepresents reduction in annual  VOC emissions from the uncontrolled level.

cBased on total  uncontrolled emissions of 1,827 Mg/yr (1,752 Mg/yr process
 emissions plus 75 Mg/yr fugitive emissions.)

^Fugitive baseline control.

eControl equivalent to Regulatory Alternative 2 (see Table 6-13).

fCombustion devices may include  flares, thermal  incinerators, catalytic
 incinerators, and boilers.
                                       6-27

-------
     Regulatory Alternative 3 represents Regulatory Alternatives 2
controls plus the combustion of continuous emissions from the raw
materials preparation section.  Under Regulatory Alternative 3,
emissions are reduced by 1,762 Mg/yr per process line or about 96.4
percent.
     Regulatory Alternative 4 represents, in addition to the control
achieved in Regulatory Alternative 39 the combustion of continuous
emissions in the product storage section.  This alternative results in
an annual emission reduction per process line of 1,766 Mg or 96.7 percent
of the total uncontrolled emissions.
     Regulatory Alternative 5 represents Regulatory Alternative 4
controls plus the control of emergency vents (i.e., decompositions)
from the polymerization reaction section.  This alternative results in
an annual emission reduction 1,770 Mg per process line or about 96.9
percent of the total uncontrolled emissions from a single process line.
     6.2.3.5  High Density Polyethylene Liquid Phase, Slurry Process.
(Table 6-18.)  Under baseline control (Regulatory Alternative 1)
Stream A (feed preparation) and Stream D (recycle treaters) would be
controlled.  These streams correspond to the raw materials preparation
and material recycle sections, respectively.  Annual uncontrolled
emissions from a process line are reduced by 900 Mg or about 92
percent.
     The next level of control, Regulatory Alternative 2, represents
baseline control plus the fugitive emission controls listed in Table 6-13.
An annual emission reduction of 920 Mg per process line would be achieved,
which is a 94 percent reduction from the uncontrolled level of emissions
in a process line.
     Regulatory Alternative 3 requires the same control  as Regulatory
Alternative 2 plus the combustion of the emissions from the product
finishing section.  This alternative reduces emissions from a process line
by 948 Mg/yr or about 97 percent.
     6.2.3.6  High Density Polyethylene, Liquid Phase, Solution Process.
(Table 6-19)  For this model  plant, baseline control (Regulatory
Alternative 1) was assumed to be reflected by the current level  of
control  being practiced by the plant on which the model  plant was
                                  6-28

-------
          Table 6-18.  REGULATORY ALTERNATIVES FOR THE HOPE LIQUID PHASE
                                  SLURRY PROCESS
Regulatory
Alternative
1
(baseline)
2
3
Process Emissions
Process 1
Section(s) Control Fugitive
Controlled3 Technique Emissions
'RMP plus ' Flare d
MR
RMP plus Flare e
MR
RMP, MR Combustionf e
plus PF
\nnual Emission Reduction
Per Process Line^
Mg/yr Percentc
900 92%
920 94%
948 97%
^Process sections include the following:
 RMP = raw materials preparation
 PR  = polymerization reaction
 MR  = material recovery
 PF  = product finishing
 PS  = product storage

bRepresents reduction in annual  VOC emissions from the uncontrolled level.

cBased on total uncontrolled emissions of 981 Mg/yr (931 Mg/yr process
 emissions plus 50 Mg/yr fugitive emissions.)

^Fugitive baseline control.

eControl equivalent to Regulatory Alternative 2 (see Table 6-13).

fCombustion devices may include  flares, thermal incinerators,  catalytic
 incinerators, and boilers.
                                       6-29

-------
          Table 6-19.  REGULATORY ALTERNATIVES FOR THE HOPE LIQUID PHASE
                                 SOLUTION PROCESS
Process Emissions
Regul atory
Alternative
1
(baseline)
2
3
Process
Section(s)
Controlled3
RMP, PR, MR
RMP, PR, MR
MR
RMP, PR, MR
Ar
Control Fugitive
Technique Emissions
Flare d
Flare e
Combustion^ e
inual Emission Reduction
Per Process Lineb
Mg/yr
824
844
888
Percent0
90%
92%
97%
              plus PF
^Process sections include the following:
 RMP = raw materials preparation
 PR  = polymerization reaction
 MR  = material recovery
 PF  s product finishing
 PS  = product storage

^Represents reduction in annual  VOC emissions from the uncontrolled level.

cBased on total uncontrolled emissions of 920 Mg/yr (870 Mg/yr process
 emissions plus 50 Mg/yr fugitive emissions.)

^Fugitive baseline control.

eControl equivalent to Regulatory Alternative 2 (see Table 6-13).

^Combustion devices may include  flares, thermal incinerators,  catalytic
 incinerators, and boilers.
                                       6-30

-------
based.  Under this assumption, all the streams in the raw materials
preparation, polymerization reaction, and material  recovery sections
were assumed "to be flared.  Baseline control  results in an annual  emis-
sion reduction of 824 Mg per process line or-about 90 percent of the
total uncontrolled emissions from a process line.
     Regulatory Alternative 2 requires the same control as baseline plus
fugitive emission controls.  This alternative reduces process and fugi-
tive emissions from the process sections in a process line under baseline
control by 844 Mg/yr or about 92 percent of the total uncontrolled emis-
sions from a process line.
     Regulatory Alternative 3 includes the controls of Regulatory
Alternative 2 as well as the emissions from the product finishing section
being controlled by combustion.  This regulatory alternative results in
an annual  emission reduction of 888 Mg per process or about 97 percent.
     6.2.3.7  Polystyrene, Continuous Process.  (Table 6-20.)  As
discussed earlier, baseline control (Regulatory Alternative 1) is
considered to be the use of condensers on the material recovery section
steams.  This reflects current industry practices so that no process
emission reduction is achieved beyond the uncontrolled emission rates
reported for this process.  Fugitive emission baseline control results
in an annual emission reduction of 23 Mg per process line.  This results
in a 27 percent reduction from uncontrolled emission levels.
     Regulatory Alternative 2 represents the application of fugitive
emission controls in addition to baseline control.   Fugitive emission
control results in annual fugitive emissions  of 24 Mg.  Under this
alternative, an annual emission reduction of  53 Mg per process line is
achieved,  or approximately 62 percent of the total  uncontrolled emissions
from a process line.
     Regulatory Alternatives 3 through 7 represent fugitive emission
control under Regulatory Alternative 2 plus further recovery of emissions
from the material recovery section by the use of refrigerated condensers
to emission levels ranging from 0.012 to 0.0018 kg VOC/Mg product.
Under these alternatives, emissions from a process line are reduced by
about 57 Mg/yr, or about 67 percent of the total uncontrolled emissions
from a process line.
                                  6-31

-------
Table  6-20.   REGULATORY  ALTERNATIVES FOR THE,  CRYSTAL  OR  IMPACT
                     POLYSTYRENE  CONTINUOUS  PROCESS
Process Emissions
Regulatory
Alternative
1
(baseline)
2
3
4
5
6
7
Process
Section(s)
Control 1 eda
-
-
Materi al
recovery
Material
recovery
Material
recovery
Material
recovery
Material
recovery
Control
Technique
-
-
Recovery6
Recovery^
RecoveryS
Recovery0
Recovery 'i
Fugitive
Emissions
c
d
d
d
d
d
d
Annual Emission
Per Process
Mg/yr
23
53
57.05
57.23
57.37
57.41
57.43
Reduction
Lineb
Percent0
27%
62%
67.1%
67.3%
67.4%
67.5%
67.6%
  Represents reduction  in annual VOC emissions from the uncontrolled level.

  bBased on total  uncontrolled emissions of 85 Mg/yr (10 Mg/yr process
  emissions plus  75 Mg/yr fugitive emissions.)

  °Fugitive baseline control.

  dControl equivalent  to Regulatory Alternative 2 (see Table  6-13).

  eControl to an emission rate of 0.012 kg VOC/Mg product through use of
  refrigerated condensers.

  fControl to an emission rate of 0.0072 kg VOC/Mg product through use of
  refrigerated condensers.

  SControl to an emission rate of 0.0036 kg VOC/Mg product through use of
  refrigerated condensers.

  "Control to an emission rate of 0.0024 kg VOC/Mg product through use of
  refrigerated condensers.

  ''Control to an emission rate of 0.0018 kg VOC/Mg product through use of
  refrigerated condensers.
                                  6-32

-------
     6.2.3.8  Expandable Polystyrene, Post-Impregnation Suspension Process.
(Table 6-21).  As discussed earlier, no baseline control (Regulatory
Alternative 1) is assumed for process emissions.  Fugitive emission
baseline control results in an annual emission reduction per process
line of about 15 Mg, or approximatley 5 percent of total uncontrolled
emissions from a process line.
     Regulatory Alternative 2 reflects increased fugitive emission
controls so that an annual  emission per process line of 35 Mg, or about
12 percent of total uncontrolled emissions, is achieved.
     Regulatory Alternative 3 reflects the fugitive emission control
under Regulatory Alternative 2 plus the control of continuous emissions
from the product storage section.  This alternative results in an
annual emission reduction of 91 Mg per process line, or about 32 percent
of total uncontrolled emissions.
     Regulatory Alternative 4 represents the emission controls of
Regulatory Alternative 3 plus the control of intermittent emissions
from the product finishing section.  This alternative results in an
annual emission reduction of 184 Mg, or about 64 percent of the total
uncontrolled emissions, per process line.
     Regulatory Alternative 5 represents the emission controls of
Regulatory Alternative 4 plus the control of the intermittent streams
from the product storage section.  This alternative results in an annual
emission reduction of 237 Mg, or about 83 percent of the total uncontrolled
emissions, per process line.
     Regulatory Alternative 6 represents the emission controls of
Regulatory Alternative 5 plus the control of continuous emissions from
the product  finishing section.  This alternative results in an annual
emission reduction of 266 Mg, or slightly less than 93 percent of the
total uncontrolled emissions, per process line.
     Regulatory Alternative 7 represents the emission controls of
Regulatory Alternative 6 plus the control of the emissions from the
polymerization  reaction section.  This alternative results in an annual
emission reduction of 267 Mg, or about 93 percent of the total uncontrolled
emissions, per  process line.
                                  6-33

-------
   Table 6-21.   REGULATORY ALTERNATIVES  FOR THE  EXPANDABLE
        POLYSTYRENE  POST-IMPREGNATION  SUSPENSION  PROCESS
Process Emissions
Regul atory
Alternative
1
(baseline)
2
3
4
5
6
7
Process
Section(s)
Control! eda
-
-
PS
(continuous
streams only)
PS
(continuous .
streams only
plus PF
(intermittent
streams only)
PS
(continous.
streams only)
plus PF
(all streams)
PS (all
streams) plus
PF (all
streams
PS, PF
PR
Control Fugitive
Technique Emissions
d
e
Combustion*1 e
Combustion^ . e
Combusti onf e
Combustion^ e
Combustion*1 e
Recovery9
Annual Emission Reduction
Per Process Lineb
Mg/yr Percent^
15 5%
35 12%
91 32%
184 64%
237 83%
266 92.7?
267 93. OX
 PR * polymerization  reaction
 PF * product finishing
 PS = product storage

Represents reduction in annual  VOC emissions  from the uncontrolled level.

ceased on total  uncontrolled emissions of 287  Mg/yr (237 Mg/yr process
 emissions and 50 Hg/yr fugitive emissions).

^Fugitive baseline control.

eControl equivalent to Regulatory Alternative  2 (see Table 6-13).

^Combustion devices may include  flares, thermal incinerators, catalytic
 incinerators, and boilers.

9Recovery to 0.003 kg VOC/Mg product through use of condensers.
                                 6-34

-------
     6.2.3.9  Expandable Polystyrene, In-situ Suspension Process.
 (Table 6-22).  As discussed earlier, no baseline control (Regulatory
Alternative 1) is assumed for process emissions.  Fugitive emission
baseline control results in an annual emission of 5 Mg per process
line, or about 9 percent of the total uncontrolled emissions from a
process line.
     Regulatory Alternative 2 reflects increased fugitive emission
control so that an annual emission reduction per process line of almost
12 Mg, or about 21 percent of the total  uncontrolled emissions, is
achieved.
     Regulatory Alternative 3 represents the fugitive emission controls
of Regulatory Alternative 2 plus the control of intermittent emissions
from the polymerization reaction section.   This alternative results in
an annual emission reduction of 20.6 Mg, or about 36 percent of the
total uncontrolled emission, per process line.
     Regulatory Alternative 4 represents the emission controls of
Regulatory Alternative 3 plus the control  of continuous emissions from
the product finishing section.  This alternative results in an annual
emission reduction of about 43.4 Mg, or about 76 percent of the total
uncontrolled emissions, per process line.
     Regulatory Alternative 5 represents the emission controls of
Regulatory Alternative 4 plus the control  of continuous streams from
the product storage section.  This alternative results in an annual
emission reduction of 49.2 Mg, or about 86 percent of the total uncon-
trolled emissions, per process line.
     Regulatory Alternative 6 represents the emission controls of
Regulatory Alternative 5 plus the controls of intermittent emissions
from the raw material preparation section.  This alternative results in
an annual  emissions,  per process line.
   Regulatory Alternative 7 represents  the emission controls of
Regulatory Alternative 6 plus the control  of intermittent emissions
from the product finishing section.   This  alternative results in an
annual  emission reduction of 50.6 Mg, or about 89 percent of the
total uncontrolled emissions,  per process  line.
                                  6-35

-------
 Table  6-22.   REGULATORY  ALTERNATIVES  FOR THE  EXPANDABLE
             POLYSTYRENE  IN-SITU  SUSPENSION  PROCESS
Process Emissions
Regulatory
Alternative
1
(baseline)
2
3
4
5
6
Process
Section(s)
Controlled3
-
-
PR
PR
plus PF
(continuous
streams only)
PR, PF
(continuous
streams )
plus PS
PR, PF
(continuous
streams), PS,
plus RMP
;
Control Fugitive
Technique Emissions
d
e
Combustion11 e
Combustion^ e
Combustion^ e
Combustion^ e
\nnual Emission Reduction
Per Process Line'5
Mg/yr Percent1-
5 9%
11.7 21%
20.6 36%
43.4 76%
49.2 86%
50.2 88%
     7        PR, PF        Combustionf      e             50.6       89%
             (continuous
             streams), RMP, PS,
             plus PF
             (i ntermi ttent
             streams)

^Process sections include the following:                         ~~~~~~
 RHP  =  raw materials preparation
 PR  =*  polymerization reaction
 PF  =  product finishing
 PS  *  product storage

Represents reduction in annual VOC emissions from the uncontrolled level.

cBased  on total uncontrolled emissions of  57 Mg/yr (40 Mg/yr process
 emissions and 17 Mg/yr fugitive emissions).

fugitive baseline control.

eControl equivalent to Regulatory Alternative 2 (see Table 6-13).

fCombustlon devices may include flares, thermal incinerators, catalytic
 incinerators, and boilers.
                                  6-36

-------
      6.2.3.10  Poly(ethylene terephthal_ate_K_DM^_Proce_ss.   (Table  6-23).
 As discussed earlier,  baseline  control  (Regulatory  Alternative 1)  in
 Table 6-23 represents  the use of spent ethylene  glycol  condensers  to
 recover ethylene  glycol  from the polymerization  reactors and the
 esterifiers.   This corresponds  to a  recovery  system currently in use  in
 the industry;  thus no  emission  reduction  from uncontrolled  emissions
 occurs under baseline  control.
      Regulatory Alternatives 2  through 6  reflect the  use of refrigerated
 condensers  to  control  the  methanol from the material  recovery section
 (i.e.,  methanol recovery)  to emission levels  of 0.018 to 0.0018 kg
 VOC/Mg of  product.  Under  this  alternative, annual  emissions from a
 process line  are  reduced between  2.43 to  2.67  Mg, or  from about 30 to
 33 percent  from uncontrolled levels.
      Regulatory Alternative  7 reflects control under  Regulatory Alterna-
 tive  4 plus  the additional recovery of ethylene glycol  from the cooling
 tower (i.e., the  polymerization  reaction  section) through the use of a
 distillation column.   Under  this  alternative,  annual  emissions are
 reduced by  5.33 Mg, or 66  percent of uncontrolled emissions, from a
 process line.
      6-2.3.11  Poly(ethy1ene  terephthlate), TPA Process - low viscosrty_
 PET  or high viscosity PET with a single  end finisher.  (Table 6-24a.)
As  discussed earlier, baseline control (Regulatory Alternative 1)  is
assumed to be  equivalent to  current industry practice in which ethylene
glycol  is recovered using spent ethylene  glycol spray condensers.   The
resulting emissions are the  same as the uncontrolled emissions and,
thus, no emission reduction  occurs under  baseline control.
      Regulatory Alternative 2 reflects the additional  recovery of
ethylene glycol from the cooling water tower using a distillation  column
that  results in a 50 percent reduction in  emissions  to the  atmosphere
from  the cooling water tower.  This regulatory alternative  would reduce
emissions by 2.35 Mg/yr per process line or about 44 percent from  uncon-
trolled emissions.
   6.2.3.12  Poly(ethy1ene terephthalate), TPA Process - high  viscosity
PET with multiple end finishers.  (Table  6-24b.)   As discussed earlier,
baseline control  (Regulatory Alternative  1) is assumed to be the use  of
spent ethylene glycol  spray condensers on  the initial  end finishers
                                  6-37

-------
         Table 6-23.   REGULATORY ALTERNATIVES FOR THE
            POLYETHYLENE TEREPHTHALATE)  DMT  PROCESS0
Process Emissions
Regulatory
Alternative
1
(baseline)
2
3
4
5
6
7
Process
Section(s)
Controlled"
PR

PR plus
MR
PR plus
MR
PR plus
MR
PR plus
MR
PR plus
MR
PR plus
MR
J
Control
Technique
Recovery6

Recovery8.
Recovery'
Recovery6
RecoveryS
Recovery6
Recovery"
Recovery6
Recovery1
Recovery f
RecoveryJ
Recovery^
RecoveryJ
Annual Emission
Per Process
Mg/yr
0

2.43
2.48
2.59
2.66
2.67
5.33
Reduction
Linec
Percent
0%

30%
31%
32.3%
33. 1%
33.3%
66%
aThese regulatory alternatives apply to a  line producing a low viscosity or
 high viscosity  product (with a single end finisher).
DPR  = polymerization reaction
 MR  - material  recovery
cRepresents reduction in annual VOC emissions from the uncontrolled level.
dBased on total  uncontrolled emissions of  8.03 Mg/yr.
eRecovery of ethylene glycol to an emission rate  of 0.355 kg VOC/Mg product
 through the use of  spent ethylene glycol  spray condensers recovering the  ethylene
 glycol prior to the vacuum system servicing the  polymerizers.
^Includes condenser  on MR controlling emissions to 0.018 kg VOC/Mg of product.
9Includes condenser  on MR controlling emissions to 0.0144 kg VOC/Mg product.
"Includes condenser  on MR controlling emissions to 0.0072 kg VOC/Mg product.
^Includes condenser  on MR controlling emissions to 0.0027 kg VOC/Mg product.
0'Includes condenser  on MR controlling emissions  to 0.0018 kg VOC/Mg product.
^Additional recovery of  ethylene  glycol by recovery  from the cooling water tower
 using  a distillation column  that results in a reduction in atmospheric emissions
 of  50  percent.
                                    6-38

-------
                     Table 6-24a.  REGULATORY ALTERNATIVES FOR
      THE PQLY(ETHYLENE TEREPHTHALATE) TPA PROCESS PRODUCING LOW VISCOSITY PET
                 OR HIGH VISCOSITY PET WITH A SINGLE END FINISHER
                  Process Emissions
Regulatory
Alternative
1
(baseline)
2
Process
Section (s)
Controlled9
RMP, PR
RMP, PR
Control
Technique
Recoveryd
Recovery^
Annual Emission
Per Process
Mg/yr
0
2.35
Reduction
Lineb
Percentc
0%
44%
aRMP = raw materials recovery
 PR  = polymerization reaction

bRepresents reduction in annual  VOC emissions from the uncontrolled level.

cBased on total uncontrolled emissions of 5.3 Mg/yr.

Recovery of glycol  from the raw materials preparation and polymerization reaction
 sections using spent ethylene glycol  spray condensers.

eControl  of emissions from the cooling water tower {i.e.,  polymerization reaction
 section) using a distillation column  that results in a  50 percent reduction in
 atmospheric emissions from the  cooling water tower.
                                      6-39

-------
                     Table 6-24b.  REGULATORY ALTERNATIVES FOR
              THE PQLY(ETHYLENE TEREPHTHALATE) TPA PROCESS PRODUCING
                 HIGH VISCOSITY PET WITH MULTIPLE END FINISHERS
Process Emissions
Regul atory
Alternative
1
(baseline)
2
3
4
5
6
7
Process
Section(s)
Controlled3
RMP,
RMP,
RMP,
RMP,
RMP,
RMP,
RMP,
PR
PR
PR
PR
PR
PR
PR
Control
Technique
Recovery*1
Recovery6
Recovery6
Recovery6
Recovery6
Recovery6
Recovery6
Annual Emission Reduction
Per Process Line"3
Mg/yr
0
7.2
14.7
22.3
26.1
29.8
37.3
Percent1-
Q%
10%
21%
31%
37%
42%
53%
aRMP = raw materials recovery
 PR  ~ polymerization reaction

^Represents reduction in annual VOC emissions from the uncontrolled level.

cBased on total uncontrolled emissions of 71 Mg/yr.

^Baseline recovery includes distillation columns on the esterifiers and
 cooling tower and spent ethylene glycol spray condensers on the initial  end
 finishers.

Additional control result of sending increasing amounts of cooling water from
 the cooling water tower to the distillation column.  Feed rate from cooling
 water tower will determine resulting ethylene glycol  concentration in  the  cooling
 water tower and the resulting level  of ethylene glycol  emissions to the  atmosphere.
                                       6-40

-------
only and distillation columns on the esterifiers and the cooling tower.
The resulting emissions are considered the same as the uncontrolled
emissions and, thus, no emission reduction occurs under baseline control
   Regulatory Alternatives 2 through 7 reflect the increased flow of
cooling water to the distillation column to recover additional  ethylene
glycol from the cooling water tower.  By varying the feed rate of the
cooling water to the distillation column, the ethylene glycol concen-
tration in the cooling water will vary and emissions to the atmosphere
will also vary.  These alternatives reflect increased flow rate to the
distillation column that results in effective emission reductions from
the polymerization reaction section via the cooling water tower ranging
from 7.2 Mg/yr or 10 percent reduction of the uncontrolled emissions,
under Regulatory Alternative 2,  up to 37.3 Mg/yr or about 53 percent
reduction,  under Regulatory Alternative 7, per process line.
6.2.4  Summary of Regulatory Alternatives
     This section summarizes the regulatory alternatives  for each model
plant.  The uncontrolled emission rates,  annual  emission  reductions,  and
percent control  achieved by the  regulatory alternatives are summarized
in Table 6-25.
                                  6-41

-------
Table 6-25.  SUMMARY OF UNCONTROLLED EMISSIONS AND EMISSION REDUCTIONS
     PER PROCESS LINE FOR REGULATORY ALTERNATIVES BY MODEL PLANT
Annual. Emission
Uncontrolled Emissions,
Model Hg/yr
Plant Process Fugitive Total

PP, L 1,858 50 1,908

PP, 6 1,278 50 1,328

LOPE, HP 692 38 730



LDPE, LP 1,752 75 1,827
HOPE, 6



HOPE, SL 931 50 981


HOPE, SO 870 50 920



PS, C 10 75 85



Regulatory
Alternative
1
2
3
4
1
2
1
2
3
4
5 .
1
2
3
4
5
1
2
3
1
2
3
1
2
3
4
5
6
7
Process
Emissions,
Mg/yr
1,690
1,690
1,817
1,821
1,252
1,252
583
583
638
652
678
1,703
1,703
1,710
1,714
1,718
385
885
913
809
809
853
0
0
4.05
4.23
4.37
4.41
4.43
Fugitive
Emissions,
Mg/yr
15
35
35
35
15
35
11
26
26
26
26
22
52
52
52
52
15
35
35
15
35
35
23
53
53
53
53
53
53
Reductions3

Total
Mg/yr
1,705
1,725
1,852
1,856 •
1,267
1,287
594
609
664
678
704
1,725
1,755
1,762
1,766 "
1,770
900
920
948
824
844
888
23
53
57.05
57.23
57.37
57.41
57.43
Percent
89$
90%
97%
97.3%
95%
97%
81%
83%
91%
93%
96%
94%
96.1%
96.4%
96.7%
96.9%
92%
94%
97%
90%
92%
97%
27%
62%
67.1%
67.3%
67.4%
67.5%
67.6%
                               6-42

-------
Table  6-25.   SUMMARY  OF  UNCONTROLLED EMISSIONS AND EMISSIONS REDUCTIONS
PER PROCESS LINE  FOR  REGULATORY  ALTERNATIVES  BY MODEL PLANT  (concluded)
Annual Emission
Uncontrolled Emissions,
Model Mg/yr
Plant Process Fugitive Total



EPS, PI 237 50 287





EPS, IS 40 17 57




PET/DMT 808





PET/TPA0 5.3 0 5.3



PET/TPAC 71 0 71



Regulatory
Alternative
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
1
2
3
4
5
6
7
Process
Emissions,
Mg/yr
0
0
56
149
202
231
232
0
0
8.9
31.7
37.5
38.5
38.9
0
2.43
2.48
2.59
2.66
2.67
5.33
0
2.4
0
7.2
14.7
22.3
26.1
29.8
37.3
Fugitive
Emissions,
Mg/yr
15
35
35
35
35
35
35
5
11.7
11.7
11.7
11.7
11.7
11.7
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Reductl onsa

Total
Mg/yr
15
35
91
184
237
266
267
. 5
11.7
20.6
43.4
49.2
50.2
50.6
0
2.43
2.48
2.59
2.66
2.67
5.33
0
2.4
0
7.2
14.7
22.3
26.1
29.3
37.3
Percent
5%
12%
32*
64?
83%
92.7%
93. OS
9%
Zl%
36%
76%
862
88%
89%
0%
30%
31%
32%
33%
33.3%
66%
0%
44%
0%
10%
21%
31%
37%
42%
53%
aFrom uncontrolled levels.

bLow viscosity, and high viscosity with single end finisher.

cHigh viscosity with multiple end finishers.
KEY:  PP = Polypropylene
   LDPE = Low Density Polyethylene
   HOPE = High Density Polyethylene
     PS = Polystyrene
 L = Liquid Phase Process
 S = Gas Phase Process
SL = Slurry Process
SO = Solution Process

  6-43
HP  = High-Pressure Process
LP  = Low-Pressure Process
PI  = Post-Impregnation Process
IS  = In-situ Process

-------

-------
                      7.0  ENVIRONMENTAL IMPACTS
     This chapter assesses the environmental  impacts of implementing the
regulatory alternatives presented in Chapter 6.   The assessment discusses
these impacts in terms of air quality,  water quality, solid waste generation,
and energy requirements.  Other areas examined include noise impacts,
irreversible and irretrievable commitment of resources, and impacts of
delaying implementation of the regulatory alternatives.
     Process and fugitive VOC emissions from polymers and resins plants
operating under Regulatory Alternative II* are projected to be about
93 percent less than uncontrolled emissions and  about 20 percent less
than estimated baseline emissions (Regulatory Alternative I).   VOC
emissions under the most stringent combination of regulatory alternatives,
Regulatory Alternative VII, are almost 97 percent of uncontrolled emission
levels and about 62 percent of baseline emission levels.
     Secondary air pollutants emitted by VOC emission control  devices
are anticipated to be minimal  in comparison to the quantity of VOC
reduced.  Water pollution and solid waste disposal impacts of  the regulatory
alternatives are expected to be minimal in comparison to the amount of
liquid and solid wastes generated during polymers and resins manufacturing
operations.
     The energy required to implement Regulatory Alternative II (primarily
fugitive VOC control) is estimated to be 66 terajoules (TJ) (11,000
barrels of oil) per year over energy demands in  the absence of any VOC
control.  Most of this increase energy usage is  atrributable to
*In this chapter, arabic numerals are used to describe regulatory
 alternatives when used in the context of a single model  plant,  or model
 plants of a single type; Roman numerals are used to describe the
 total values associated with all model  plants of a given Regulatory
 Alternative.  For example, all model plants under Regulatory Alternative 1
 are collectively described as Regulatory Alternative I.
                          7-1

-------
 one model plant [poly(ethylene terephthalate),  terephthalic acid
 process using a single  end finisher].   Except for the  polyethylene
 terephthalate) plants,  Regulatory Alternative 2 for all  other model
 plants show a decrease  in  energy demand because the energy cost for
 fugitive VOC control  under Regulatory  Alternative 2 is less than  the
 energy credit resulting from the reduced loss of VOC under these
 control measures.   The  most stringent  combination of regulatory
 alternatives, Regulatory Alternative VII,  requires about 1,950 TJ
 (318,000 barrels of oil) per year more than  energy demands without
 VOC controls, and  about 940  TJ  (154,000 barrels  of oil)  per year more
 than the projected requirement under current VOC  control  practices
 (Regulatory Alternative I).
      Projected noise  impacts due to implementation of  any  regulatory
 alternative are expected to  be  minimal.  No  significant  irreversible or
 irretrievable commitments  of resources  are expected  to be  incurred under
 the regulatory alternatives.  Delaying  implementation  of Regulatory
 Alternatives II through  VII  is  anticipated to adversely  impact air quality,
 Detailed  discussion of the assessed environmental  impacts  is presented
 in  the  following sections.
 7.1   AIR  POLLUTION  IMPACTS
      The  air pollution  impact of each regulatory alternative is determined
 by comparison  of uncontrolled VOC emission rates to residual VOC emission
 rates for emission control  systems installed on process operations and
 residual  VOC  emission rates  for fugitive emission control  practices.   In
 order to  analyze the incremental air quality impact of each regulatory
 alternative,  average annual VOC emission rates from process lines  in
 each model plant are determined and used to project industrywide air
quality impacts of new polymers amd resins process lines.
7.1.1  Average Annual Model Plant VOC  Emissions
     Annual  VOC emission rates for a process line in each model plant
are determined through the  following equation:
          Ei.i = Pi.i + P.,-      (1)
                             7-2

-------
    where EJJ = annual VOC emission rate (Mg/yr) for a process line
                in each model plant i_  (for example i = polypropylene/
                liquid phase; for other model plant types see
                Table 6-20) under Regulatory Alternative j_
                (j = uncontrolled, 1,  2, 3, 4, 5, 6, or 7);
          P-JJ = annual VOC process emission rate for a process
                line in model plant i_  under Regulatory Alternative j_;
           Fj = annual VOC fugitive emission rate for a process line
                in each model plant under Regulatory Alternative j_.
The annual VOC process emission rate (P-JJ) is a function of the uncontrolled
emission rate and the effectiveness of the control  technique applied to
each VOC stream or process section.  Thus the annual  average process VOC
emission rate (P-jj) can be expressed as:
          PIJ =   L(Uai x  [1-Caij])        (2)
    where Uai- = the uncontrolled VOC emission rate (Mg/yr) from process
                section a^ (a = raw material preparation, polymerization
                reaction, material  recovery, product finishing, or product
                storage) in a process line in model  plant J_;
         caij = tne voc emission reduction efficiency of the control system
                for process section £ in a process line in model  plant j_
                under Regulatory Alternative j_.
The uncontrolled process emission rates for each process section in a
process line in each model plant are presented in Tables 6-1 through
6-1Ib.  The control techniques employed on appropriate model plant process
sections and their emission control  efficiencies are presented in Tables
6-14 through 6-24b.
     The quantity of fugitive VOC emissions are assumed to be the same
for each model plant.  The uncontrolled fugitive VOC emission rate is
based on equipment component counts and uncontrolled component emission
rates presented in Table 6-12.  The effectiveness of fugitive VOC emission
control through the use of leak detection and repair programs is based
on the. leak detection and repair (LDAR) model  developed for control of
fugitive VOC emissions from SOCMI,1 and is presented in Table 6-13.
     Table 7-1 presents the primary or VOC related air quality impacts
of the regulatory alternatives for a process line in  each model plant.
                                   7-3

-------
Table 7-1 a.  PRIMARY AIR QUALITY IMPACTS OF THE REGULATORY
    ALTERNATIVES FOR POLYMERS AND RESINS PLANTS (Mg/yr)
Regulatory
Model Plant Alternative
PP/Liquid
PP/Gas
LDPE/High Pressure
LDPE/Low Pressure
HDPE/Gas Phase
HOPE/SI urry
HDPE/Solut1on
PS/Continuous
EPS/Post-
•irapregnation
Ub
1
2
3
4
Ub
1
2
Ub
1
2
3
4
5
Ub
1
2
3
4
5
Ub
1
2
3
Ub
1
2
3
Ub
1
2
3
4
5
6
7
Ub
1
2
3
4
5
6
7
VOC Emissions per
Process Line3
Process
1,858
168
168
41
37
1,278
26
26
692
109
109
54
40
14
1,752
49
49
42
38
34
931
46
46
18 '
870
61
61
17
10
10
10
6
5.8
5.63
5.59
5.57
237
237
237
181
88
35
6
5
Fugitive
50
35
15
15
15
50
35
15
38
27
12
12
12
12
75
53
23
23
23
23
50
35
15
15
50
. 35
15
15
75
53
23
23
23
23
23
.23
50
35
35
15
15
15
15
15
Combined
1,908
203
183
56
52
1,328
61
41
730
136
121
66
52
26
1,827
102
72
65
61
57
981
81
61
33
920
96
76
32
85
63
33
29
28.8
28.63
28.59
28.57
287
272
252
196
103
50
21
20
Percent VOC Emission Reduction From
Baseline (Regulatory Alternative 1)
Process
0
76
78
0
0
51
63
87
0
14
22
31
0
61
0
72
0
41
42
43.7
44.1
44.3
0
24
63
85
97.5
97.9
Fugi ti ve
57
57
57
57
56
56
56
564
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
57
• 57
57
57
57
Comtn ned
10
72
74
33
11
52
62
81
29
36
40
44
25
59
21
67
48
54
54.3
54.6
54.6
54.7
7
28
62
82
92.3
92.6
                             7-4

-------
         Table 7-la.   PRIMARY AIR  QUALITY IMPACTS OF  THE  REGULATORY
              ALTERNATIVES FOR POLYMERS AND RESINS PLANTS  (Mg/yr)
                                     (concluded)


Model Plant
EPS/In-situ







PET/DMT^







PET/TPAC


PET/TPAd








Regulatory
Alternative
Ub
1
2
3
'4
5
6
7
Ub
1
2
3
4
5
6
7
Ub
1
2
Ub
1
2
3
4
5
6
7
VOC E
Prot
Process
40
40
40
31
8.3
2.5
1.5
1.1
8
8
5.6
5.55
5.44
5.37
5.36
2.7
5.3
5.S
2.9
71
71
64
56
49
45
41.
34
[missions pei
:ess Line3
Fugitive (
17
12
5.3
5.3
5.3
5.3
5.3
5.3
0
0
0
0
0
0
0
•0
0
0
0
0
0
0
0
0
0
0
0


Combined
57
52
45.3
36.3
13.6
7.8
6.8
6.4
8
8
5.6
5.55
5.44
5.37
5.36
2.7
5.3
5.3
2.9
71
71
64
56
49
45
41
34
Percent '
Baseline
Process
„
_
0
23
79
94
96
97

-
30
31
32
33.1
33.3
66
_
-
44
_
-
10
21
31
37
42
53
/OC Emission
(Regulatory
Fugitive
_
-
56
56
56
56
56
56

-
0
0
0
0
0
0
_
-
0
-
-
0
0
0
0
0
0
i Reduction From
f Al ternati ve 1 )
Combined
_
-
13
30
74
.85
87
88
.
-
30
31
32
33.1
33.3
66
_
-
.44
-
-
10
21
31
37
42
53
 From Table 6-25.

b
 U  = uncontrolled.


 These regulatory alternatives apply to  a process line producing PET with a single end finisher.


d
 These regulatory alternatives apply to  a process line producing PET with multiple end finishers.
                                          7-5

-------
Table 7-1 b.  PRIMARY AIR QUALITY IMPACTS OF THE REGULATORY
    ALTERNATIVES FOR POLYMERS AND RESINS PLANTS (tons/yr)

Model Plant
PP/Uquld




PP/Gas



Regulatory
Alternative
Ub
1
2
3
4
Ub
1
2
LDP£/H1gh Pressure Ub





1
2
3
4
5
VOC
Emissions
Process line3
Process
2,048
185
185
45
41
1,408
29
29
763
120
120
60
44
15
Fugitive
55
39
17
17
17
55
39
17
42
30
13
13
13
13
per

Combined
2,103
224
202
62
57
1,463
67
45
804
150
133
73
57
23
Percent
Baseline
Process

..
0
76
78
—
.
0
_
_
0
51
63
87
VOC Emission
(Regulatory
Fugitive

_
57
57
57
•
_
57
_
_
56
56
56
56
Reduction from
Alternative 1)
Combined

_
10
72
74
.
_
33
—
..
11
52
62
81
LDPE/Low Pressure
HDPE/Gas Phase





HDPE/Slurry



HDPE/Solut1on



PS/Continuous







EPS/Post-
itapregnation







Ub
1
2
3
4
5
Ub
1
2
3
Ub
1
2
3
Ub
1
2
3
4
5
6
7
Ub
1
2
3
4
5
6
7
1,931
54
54
46
42
37
1,026
51
51
20
959
67
67
19
11
11
11
6.6
6.4
6.2
6.16
6.14
261
261
261
199
97
39
6.6
5.5
83
58
25
25
25
25
55
39
17
17
55
39
17
17
83
58
25'
25
25
25
25
25
55
39
17
17
17
17
17
17
2,013
112
79
72
67
63
1,081
89
67
36
1,014
106-
84
35
94
69
36
31.6
31.4
31.2
31.16
31.14
316
300
278
216
114
55
23
22
«
_
0
14
22
31
_
_
0
61
—
_
0
72
_
_
0
41
42
43.7
44.1
44.3

_
0
24
63
85
97.5
97.9
_
_
57
57
57
57.
_
_
57
57
_
_
57
57
.
_
57
57
57
57
57
57

_
57
57
57
57
57
57
_
_
29
36
40
44
«•
_
25
59
_
_
21
67
.
_
48
54
54.3
54.6
54.6
54.7

—
7
28
62
82
92.3
92.6
                            7-6

-------
         Table 7-1b.   PRIMARY  AIR QUALITY  IMPACTS FOR  THE REGULATORY
           ALTERNATIVES  FOR POLYMERS  AND RESINS  PLANTS (tons/yr)
                                     (concluded)


Model Plant
EPS/In-situ







PET/DMT^







PET/TPftC


PET/TPAd








Regulatory
Alternative
Ub
1
2
3
4
5
6
7 .
Ub
1
2
3
4
5
6
7
Ub
1
2
Ub
1
2
3
4
5
6
7
VOC 1
Pro<
Process
44
44
44
34
9.1
2.8
1.7
1.2
9
9
6.2
6.1
6.0
5.92
5.91
2.98
6
6
3
78
78
70
62
54
49
45
37
Emissions p
:ess line3
Fugitive
19
13
6
6
6
6
6
6
0 '
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
o •
er

Combined
63
57
50
40
15.1
8.8
7.7
7.2
9
9
6.2
6.1
6.0
5.92
5.91
2.98
6
6
3
78
78
70
62
54
49
45
37
Percent ^
Baseline
Process

_
0
23
79
94
96
97

_
30
31
32
33.1
33.3
66
_
_
44
_
-
10
21
31
37
42
53
fOC Emission
(Regulatory
Fugitive

_
56
56
56
56
56
56

_
0
0
0
0
0
0

_
0
_
..
0
0
0
0
0
0
Reduction from
Alternative 1)
Combined

_
13
30
74
85 •
87
88

—
30
31
32
33.1
33.3
66

_.
44
_
_,
10
21
31
37
42
53
 From Table 6-25.
b
 U = uncontrolled.

c
 These regulatory alternatives apply to a process line producing PET with a.single end finisher.
d
 These regulatory alternatives apply to a process Ifne producing PET with multiple end finishers.
                                        7-7

-------
(Table 7-1 and other tables, as appropriate, in this chapter have been
separated into two tables; the first table reports the information in
metric units; the second, in English units.)  Process VOC emission
reductions under baseline controls (Regulatory Alternative I) range
from 0 to 98 percent, depending on whether current polymers and resins
industry practices include controls on the type of model  plant examined
and the extent that these controls are used to treat VOC  emissions from
a particular model plant process section.  Fugitive VOC emission control
under current industry practices assumes that 75 percent of all gas
safety relief valves and sampling connections, most of the open-ended
lines, and 0 percent of all other fugitive emission sources are controlled.
This results in a decrease in fugitive emissions from uncontrolled
levels of about 30 percent (from 1,51 Mg/yr to 106 Mg/yr).
     Process VOC emission controls and process VOC emission rates under
Regulatory Alternative II generally remain unchanged from the baseline
levels.  However, fugitive emission control practices implemented under
Regulatory Alternative II result in a decrease from baseline control of
approximately 57 percent in fugitive emissions of VOC from a process
line in each model plant except for the PET process lines.  Regulatory
Alternative 2 for the poly(ethylene terepthalate) process lines reduces
baseline emissions either through the use of  a distillation column or a
refrigerated condenser.  Under Regulatory Alternative II, overall VOC
emission reductions  from baseline control (processes and fugitive
emission reductions) range from 7 to 48 percent.  The differences in
emission reduction efficiency over baseline among process lines in the
model plants is a function primarily of the extent that process controls
are currently employed by each industry segment and the proportion of
fugitive emissions to overall emissions from  a process line in the
model plant.
     Under Regulatory Alternatives III through VII, more stringent controls
are applied  to  process emissions.  Implementation of Regulatory Alternative
VII, the most stringent  combination of regulatory alternatives, would
result  in a  range of overall  VOC emission reduction from baseline
control of about  33  to 93 percent.
                                 7-8

-------
7.1.2  Industrywide VOC Emission Impacts of New Plants
     The effect of the regulatory alternatives on VOC emissions from new
process lines projected to be built over the next 5 years can be estimated
by the equation:
= ]C(E
                     iJ
   where  My  =  total VOC emissions from all new polymer and resin
                 process lines (Mg/yr);
         EJJ  =  annual emission rate from a process line in model plant i
                 under Regulatory Altertative j_ (Mg/yr- from Table 7-1);
          n-j  =  number of new process lines of model plant type i_
                 projected to be built by 1988 (from Table 8-43).
The impact of the regulatory alternatives on VOC emissions from the
85 new process lines expected to be built are presented in Table 7-2.
Baseline emissions from new process lines are projected to be 7,332
Mg/yr (8,080 tons/yr), or a 91 percent overall reduction from uncontrolled
levels.   The additional increment of emission control under Regulatory
Alternative II results in overall VOC emissions of about 5,837 Mg/yr
(6,432 tons/yr):  a 93 percent decrease from the uncontrolled emission
rate, or a 20 percent incremental reduction in emissions over Regulatory
Alternative I.  The additional process emission controls required under
Regulatory Alternative VII, the most stringent combination of regulatory
alternatives, are expected to reduce uncontrolled emissions by almost
97 percent to about 2,770 Mg/yr (3,056 tons/yr):  a 62 percent incremental
reduction in emissions over Regulatory Alternative I.
7.1.3  Secondary Air Quality Impacts of the Regulatory Alternatives
     Secondary air pollutants are those emissions which are not usually
associated with an uncontrolled process, but which result from the use
of pollution control  equipment.  VOC emission control equipment that may
be incorporated into model plant VOC emission reduction systems include
flares,  thermal  incinerators, catalytic incinerators, and condensers.
Secondary air pollutants are not expected to be generated by leak detection
and repair programs for fugitive emissions of VOC.  Secondary air pollutants
that may be generated through the improper use or maintenance of VOC
emission control  systems are not expected to be significant.   However,
pollutants generated by the combustion of fuel to generate steam for the
flares,  to incinerate low VOC content streams, or to generate electrical
                               7-9

-------



CO
LU
s

C£
Ul

I





O
fe-
CD
LU


LU
TT

LL..— >
o £-
1— D>
«c^S
CL.
ECO
t-4U-
SK
n2
t-i cu
- as
§ CO
I-H OS
OC LU
CuS
>•
LU _l
Q O
t—t cu

oso
1— U.
co
=3

S«
t— <
fO
CM
1
'"-
Q)

3
ro
t—






|-
U. 0)
ll
P &.
*£
•ss
tA 
Hj
IA
tA
Q)
2
ex.
=c

1




V

§
Ju

tA
tA
U
O
D.

61
«2
§**

u

<

w
_«
X

1
X


, ,2KS

i i in u> tn







to co
i i o t-- r^.




CM r*» r-.^ eo
r— CM *r o in
•— t co *o tn >«r
* •> M
f^, ,»< *«|
t~t


Sin in in tn
*-4 CO COCO
'
CM CM CM cn co
CM*«< _i  t in







l t O




CM en cn
tn *f 
tA
O-
tA
Ul

in cn
i i cMin
i < in in

o -*
1 1 to

tO IO ID CO
co co to cn
in
oo oo
CO CM

to to 10 co
U) CM CM rH
to



_« _4 ^H CO
co co 
O
to
o_
f O tO
f** CO CM CM CM CM
i i CM to co cn cn
i i in in in in in in
in en
o **• co tn r1- f*-
< i CM to co cn cn

in in in co CM *~*

oooooooo
rH

^••Cf^fCMtoOlNJO
*r«r *f co r-4

CM


r^-CMCMtOCOOi-HO
cor^inenomcMCM
CM CM CM rH rH
ominintninmtn

r-~r-r^rHCOinioin
CM CM CM •-<
u
c
o
(0
tn a»
O S-
D_ a.
cu
Ul
7-10

-------
oo
0£
O
CD
UJ


UJ

I—


o'sT

00 --s.

(_j gr


s: oo
•-> I—



1—1 Q_

«=C oo^"
ID 2:  d)
O*i—< "O
   00  3
IV I i 1 ,—
i—i o;  o
<;     c
   a  o
>- ^  o
S oo
i—i rv
CC LU
Q_ 21

LU —I
Q O
i—« Q.


?«
OS O
I— U.
oo

Q
CM




 0}



 (C
T- « .a
*• = 9
u i. o

D ^
"''a.


1^5
    "ai
             C H-
             

  8WJ CU
  01 >
> o *-
*£

ll
  O

  S- VI
  01 OJ
  O3 O

  12
  za.
    *-   t

co o ^- to t< co
1 1 i— 4 CO f-"- CO CO CO
1 1 IO IO IO IO IO tO
t 1 O CM P*- & CT\ O*l
IO IO OO
f^-lOOOO^CMCMi— t
I— 1 V~* r* f t

SS222222
un to co
CM CM CM Cn CM

«
CO CO \O CO CO ^"
SSSS2---
CO CO CO CO CO CO
r-f t— I
CO IO U) »-i
„,»—
0
3
I
C
oo
UJ
»-H CO
O T-< CM CO CO ^O
1 1 CO CO CO CO CO *O

O _< CM CO CO to
CM CTl _4 tO IO CT>
Sto cn co co r** !"••• co
IO CO CO CO CO CO ,-«

00000000
CM en ,-t to 10 cn
SSSSSRS52

.

..-—..«

00000000.
to io ^r co co r^



1
r-
CU

t 1 «*•
t 1
, ,?.

S S CO

o o o

cn cn co
to to co

CO
co co en
IO U5 CM

o o o
CO CO Cn
to 10 CM
= ^CM

^3
CL.
H-
H-
O.

' 1 t-H CM CO CO *T tO

a —< ;-« r-cM co
1 I ,-( CM CO CO -d" tO

K^ -cr to en to ^H <•
r** to to ^~ Kf ^* co



S^^u,^^^^

-

SSSSSS5S

oooooaoo

^-.^»»lo^o*3•^^^rco


h-
•h^
a.
CO CM
1 1 CM tO IO tO to >
CL
CL
(O
tft
Ol
>
^M
»
CO Ol i-
^f r— C3
i t— *5
co o 
h- C
3 01
Sin
II Ol
i = i£


V)
i.
01
^
(A
C
<*.
^0
01

4)
"cL
•p

£

5


UJ
a.
CO
•^
•o
o
s_
a.
a>
c
r^-

V)
VI
a>
u
o
t.
a.

to
o
4->

>»
a.
a.

5
fl3
3
cn
at

01
V)
?s
0
0)
3

01
•)->


01
>
+J
e:
&.
01
r-
rtj

£•
S
03
's
cn
01 *
•••8
L to
(O 3
3 V)
CJ T-
•^ >>
II
fl3 WJ
ta
S" O
01
•a +J
c e
3 
^u
ia to
> c

O -M

v) »
5o
O 19
C r—
3
42 s*
S_
s- cn


01
Ol 01
S- 0
01 01

                                                          7-n

-------
; I:

* O 4J
CO
11
LU i «<
5 i i£
I—
UJ
f—
^
o
(A IQ
O (X
§"^
j w g
S^
O 01
a. ea
fee i
= i g
g !„
o= • KM
uj i a2
in
u.'tT
CO W
1— c:
0 0
<•*->
CL. ^*-^
»-H CO
>- Z

•a:
g-S
(— 4
OS CO
H-< UJ
>- O
f*S. sy
l~< CO
QSQ:
Qu UJ
UJ^
O —1
3 Q.

So;
CO U_
=3
o
•sc
t— <
*
JD
CVJ
1
*"*"
a>
r™
J3
fO
H-








c^ a>
« i-
^ o~
fl

.5


1
C
1
Cf
5
3
X

M
a.


3
|
"*
at
n
3

ocess
i
^
Of
c
-1
L. (A
f


tO !"»• CJ> CT>


cn in in cn
CM o o to
co"



cn




T to «cf in
»-• O CO CO
o ^~*


Igajr-.^
m co *^ »H



in to to •— <
cn



^ .-H CM CO





o

•g
o
CO
LU
Q_
a
3=


co to vo r-
i i ^- Ln in in m in


i t in in in Ln in in


r-'-Hco
O 1-1 CM co *j- "sy


r*^ ^ co ^* co co PO co
i— t f— i



>— i >-i


CMCMCMCOCOCMCMCM




CM


to^s-
IO ^f CM t-4 f-H
OHOCOCOCOCOfOCO


SCO in in in in in in
in CM CM CM CM CM CM

to ^~
IO *t CM »-4 ^-t
^-t t-4 ^



CJ






o
3
c
c
o
o
-X.
1.


co to
r^ CO CM CM CM cs
i i CM to co cn cn


i i in in in m in in


in cn
O ^ co in r-I i*^
i > CM to co cn cn


to in in *f CM f-t



or-copoporococo


in in in co »-4 "^ *^




CM




toocoto^-inpocM
CO CO CM CM .-H







o u


\.
7-12

-------



CO
UJ


1— *
1—
^
^?
LlJ
|— •
_J



O
1—
^
UJ
f*v^

UJ
in
"""T
Lu >}
t/T
CO C
1— 0



Q_
t— • 1 —

*a— ^F
L_ _J
t-i 0.
<: co -o
=D ^ O)

CO 3
rs 1 1 1 ,__
i— i rv c_»
«^ C
Q 0
>- Z O
S CO
I-H Qi
oi 1 1 1
Q- S
UJ —I
Q O
t—> CL.

>- fy
o; o
I—* * *
co
Q
^^
\ — t



_Q
CM
r^
O)
3
1—







I'-'
u. oj

i!
0) i—
a: <:

o 1^
In S
w» *""'
4-* =
a; i—
o ^
&. VI
(t) (O
a. CD
^-
C i—
0
VI V)
V) 0)
*H o
a.
8






^
1
o
or









I
f*"
j


0)
^ i
F"

LU

V)
V)
O

a.


i
0
S
p


o
5
en
3
•*•

V)
VI
a>
u
o
a.

o
01
_j
V*
vt
0)
g

Q.
I
TJ>


1



0

4^
OJ
3
JU

V)
o
a.




>
5
S-
 U> Ul U) U5




i i o <*5 cn ^ to r*»
CM !*- en en en





g CM 0 0 W JO 1 PO CO tO
1 1 CO CO CO CO CO IO





to to ^ ^* ^3" ^ t±" CM




oooooooo







to vo <• ^- ^- ^r ^r CM







^






CM ;H CO
CM •— • o cn cn cn
cncntototoLoiocM








CM i— t CO
CM «-H o cn o> cn
cncntototo^tocM








=3 CO tOtO





•Q
O
•^*

cu


, . ?





o




i i S





sss




o o o







r^r-rr







CO







to 
3
5
• g

fij tU
"w "w >



t_
C C -tJ
a) a» *—
 c
> > s.
f- T- C (U
C C "tnlo
S_ S_ Ul
^ S *s £*
P- r- 33 0
(0 fl 4-»
O (0
*T* ^ in S S 42 ^
r-. co o ra fl s.
a>  JS *i—
IQ "3 u u s, -M -a
h— J— C QJ
S £ Ui Wl S- o
&. s_ " j= j= jS Si
U. U. =3 t— h— 3 O-
« ja u -a 
-------
power to operate control devices may adversely impact air quality.
Consequently, particulate, sulfur oxide (SOX) , and nitrogen oxide (NOX)
emission rates are estimated, based on energy consumption rates for VOC
emission control presented in Chapter 8, by the equation:       IMg
                           x KGx) + (
where S
X
XJ-
                                     x Kox)] ni x 0.454.   103kg
                                                                         (4)
                  new process  line  industrywide  emission  rate  (Mg/yr)
                  of  secondary air  pollutant _x   (x  = particulate, SOX, NOX)
                  under  Regulatory  Alternative^ (j =1,  2, 3, 4);
                  air pollutant _x  (x  =  particulate, SOX,  Nox)  under Regulatory
                  Alternative^ (j  =  1,  2,  3, 4);
            G.JJ  = natural  gas  requirements  to control VOC emissions from
                  a process  line in model plant  i_ under Regulatory Alternative
                  j_ (106 ft3/yr);
            KQX  = pollutant _x  emission  factor for natural gas  combustion2
                 (lb/106 ft3)
            Bij  = fuel oil required to  generate  electricity and steam
                 for  VOC control equipment  in a  process line in model plant i_
                 under Regulatory Alternative j_  (gal/yr);
            KQX  = pollutant x_  emission  factor for No. 4 fuel oil combustion3
                  (lb/103  gal3);
            n-j  =  number of new process lines of model plant type _i_ to be built
                  in year  1988.
No. 4 fuel  oil  is chosen  to represent  the  approximate midrange of fuel oil
grades available  for  use.  Fuel oil sulfur content of 2 percent by weight
represents  the  upper  limit of sulfur content in No. 4 fuel  oil.4 It is
recognized  that  particulate and SOX emissions from fuel  combustion will  be
reduced through  State Implementation Plan  (SIP) requirements which follow
the guidelines  set forth  in EPA regulations on the preparation of
implementaion plans   [40 CFR 51]. 5  However, it is useful to estimate
uncontrolled secondary  pollutant emission rates to construct a worst-case
scenario.   Potential  secondary air pollutant impacts for new polymers
and resins  process lines are presented in Table 7-3.
     The 85 new process lines that are expected to come  on  line by  1988
are projected to  emit approximately 19.3 Mg/yr (21.3 tons/yr)  of particulate,
822 Mg/yr (907 tons/yr)  of SOX, and 173 Mg/yr (190 tons/yr) of NOX,
if these plants operate under current industry  practices (Regulatory
                              7-14

-------
Table 7-3a.  SECONDARY AIR QUALITY IMPACTS OF THE REGULATORY
         ALTERNATIVES FOR POLYMERS AND RESINS PLANTS
Regulatory
Model Plant Alternatives
PP/liquid



PP/Gas

LDPE/High Pressure




LDPE/Low Pressure
HDPE/Gas Phase



HDPE/Slurry


HDPE/Solution


PS/Continuous






EPS/Post-
impregnation






1
2
3
4
1
2
1
2
3
4
5
1
2
3
4
5
1
2
3
1
2
3
1
2
3
4
5
6
7

1
2
3
4
5
6
7
Process Line Emissions,
kg/yr
Parti cul ate
73
73
78
83
84
84
23
23
35
38
185
100
100
105
120
160
49
49
50
41
41
230
c
c
1 •
1
1
1
2

c
c
21
225
320
1,070
1,070
Number of
Industrywide Emissions,
. Mg/yr
SOV NOY New Process Lines" Particulate SOV
A A A
2,660
2,660
2,690
2,690
3,130
3,130
840
840
1,090
1,160
6,870
3,275
3,275
3,280
3,450
4,690
1,780
1,780
1,820
1,210
1,210
1,470
c
c
36
39
54
56
70

c
c
300
970
2,000
2,140
2,140
720
720 9
790
860
810
810 9
230
230 4
370
410
1,790
1,050
1,050
1,130 20
1,340
1,720
480
.480 6
500
440
440 9
3,300
c
c 2
7
7
10
11
13

c
c
275 2
3,280
4,630
16,100
16,100
0.66
0.66
0.70
0.74
0.75
0.75
0.09
0.09
0.14
0.15
0.74
2.01
2.01
2.11
2.42
3.16
0.29
0.29
0.30
0.37
0.37
2.06
c
c
0.002
0.002
0.002
0.002
0.004

c
c
0.04
0.44
0.64
2.13
2.13
23.9
23.9
24.1
24.2
28.1
28.1
3.36
3.36
4.38
4.61
27.4
65.4
65.4
65.5
68.9
93.7
10.7
10.7
10.9
10.8
10.8
13.2
c
c
0.07
0.08
0.11
0,11
0.14

c
c
0.60
1.94
4.00
4.27
4.27
NOX
6.49
6.49
7.08
7.73
7.29
7.29
0.93
0.93
1.50
1.64
7.14
21.0
21.0
22.5
26.7
34.3
2.89
2.89
2.98
3.99
3.99
29.7
c
c
0.01
0.01
0.02
0.02
0.03

c
c
0.55
6.55
9.24
32.0
32.0
                            7-15

-------
            Table  7-3a.   SECONDARY  AIR  QUALITY  IMPACTS  OF  THE REGULATORY
                        ALTERNATIVES  FOR POLYMERS  AND RESINS  PLANTS
                                              (concluded)
Regulator}
Model Plant Alternat1v<
EPS/In-s1tu






PET/DMT<*






PET/TPA^

PET/TPA®






TOTAL f .






1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
1
2
3
4
5
6
7
I
II
III
IV
V
VI
VII
Process Line Emissions,
r kg/yr
ss Parti cul ate
c
c
12
24
32
46
59
756
756
756
756
757
757
1,000
756
975
74
84
98
115
130
145
185
_
-
-
-
-
-

Industrywide Emissions,
Number of K Mg/yr
MJX NUX New Process Lines" Particulate SOX
c
c
320
480
575
880
1,190
33,910
33,930
33,930
33,930
33,940
33,940
44,990
33,910
43,750
3,330
3,770
4,380
5,230
5,790
6,460
8,410
_
-
-
-
-
-
"
c
c
135 3
290
410
565
720
6,480
6,480 7
6,480
• 6,480
6,490
6,490
8,600
6,480
8,360 13
635
720 1
840
1,000
1,110
1,240
1,610
_
-
85
-
-
-
—
c
c
0.04
0.07
0.10
0.14
0.18
5.28
5.28
5.28
5.28
5.28
5.29
7.02
9.81
12.7
0.07
0.08
0.10
0.12
0.13
0.14
0.19
19. 3<=
22.2C
24.2
25.0
26.6
28.1
30.0
c
c
0.95
1.44
1.72
2.64
3.56
237
237
237
237
237
237
315
440
569
3.32
3.76
4.37
5.22
5.77
6.45
8.39
822 c
952 c
953
964
1,015
1,017
1,098
NQX
c
c
0.40
0.87
1.22
1.69
2.16
38.0
45.2
45.3
45.7
45.3
45.3
60.2
84.1
109
0.63
0.72
0.34
1.00
1.10
1.23
1.60
173<=
197C
227
238
255
278
294
""Emission factors  from References 2 and 3.  Fuel  consumption  under each  regulatory alternative i's derived from
 ubles 3-21 and 8-31b.  Factors used are for combustion of No. 4 fuel oil having 2 percent sulfur content by weight.
 (Source: Reference 4.)                                                                                 a


sFrora Table 3-43.


CEstisiates of secondary pollutants from polystyrene plants are not available, and are not included in  totals  for regulatory
 alternatives.


dThese regulatory  alternatives apply to a process Line producing PET with a single end finisher.


«These regulatory  alternatives apply to a process line producing PET with multiple end finishers.
      there Is no emission value for a model  plant under a particular regulatory alternative, the emission value
 of the preceedlng regulatory alternative for that model  plant category is used.
                                                  7-16

-------
Table 7-3b.  SECONDARY AIR QUALITY IMPACTS FOR THE REGULATORY
         ALTERNATIVES FOR POLYMERS AND RESINS PLANTS
Process Line Emissions
Regulatory
Model Plant Alternatives
PP/L1qu1d



PP/Gas

LDPE/High
Pressure



LDPE/Low Pressure
HDPE/Sas Phase



HDPE/Slurry


HDPE/Solution


PS/Continuous






EPS/Post-
irapregnation





1
2
3
4
1
2
1
2
3
4
5
1
2
3
4
5
1
2
3
1
2
3
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Ibs/yr
Parti cul ate
160
160
170
180
185
185
50
50
75
85
405
220
220
235
265
350
108
108
111
.90
90
505
c
c
2
2
3
3
3
c
c
46
490
705
2,350
2S350
sox
5,860
5,860
5,910
5,930
6,890
6,890
1,850
1,850
2,400
2,540
15,100
7,200
7,200
7,220
7,590
10,330
3,910
3,910
3,990
2,650
2,650
3,230
c
c
80
85
120
125
155
c
c
660 •
2,140
4,410
4,700
4,700
a
Number of
MOX New Process Lines
1,590
1,590 9
1,735
1,890
1,790
1,790 9
510
510
825 4
905
3,930
2,320
2,320 20
2,480
2,940
3,780
1,060
1,060 6
1,100
975
975 9
7,260
c
c 2
15
16
23
24
29
c
c
600
7,220 2
10,200
35,300
35,300
Industrywide Emissions,

Parti cul ate
0.73
0.73
0.77
0.82
0.83
0.83
0.10
0.10
0.15
0.17
0.81
2.22
2.22
2.33
2.67
3.48
0.32
0.32
0.33
0.40
0.40
2.27
c
c
0.002
0.002
0.003
0.003
0.003
c
c
0.05
0.49
0.71
2.35
2.35
tons/yr
sox
26.4
26.4
26.6
26.7
31.0
31.0
3.70
3.70
4.81
5.08
30.2
72.1
72.1
72.2
75.9
103
11.7
11.7
12.0
11.9
11.9
14.6
c
c
0.08
0.09
0.12
0.12
0.15
c
c
0.66
2.13
4.41
4.70
4.70

NOX
7.15
7.15
7.80
8.52
8.04
8.04
1.02
1.02
1.65
1.81
7.87
23.2
23.2
24.8
29.4
37.8
3.18
3.18
3.28
4.39
4.39
32.7
c
c
0.02
0.02
0.02
0.02
0.03
c
c
0.60
7.22
10.2
35.3
35.3
                           7-17

-------
              Table 7-3b.   SECONDARY  AIR QUALITY IMPACTS OF THE  REGULATORY
                          ALTERNATIVES FOR  POLYMERS AND  RESINS PLANTS
                                               (concluded)
Regulatory
Model Plant Alternatives
EPS/In-situ






PET/OMTd






PET/TPA<»

PET/TPAe






TOTALf






1
2
3
4
5
6
7
1
2
3
4
5
6
7
I
2
1
2
3
4
5
6
7
I
II
III
IV
V
VI
VII
Process Line Emissions,
Ibs/yr
Parti cu late
c
c
26
52
71
100
130
1,660
1,660
1,660
1,660
1,660
1,660
2,210
1,660
2,150
165
185
215
255
285
315
410
..
-
-
-
-
-
—
50X
c
c
700
1,060
1,260
1,940
2,620
74,610
74,645
74,646
74,652
74,659
74,667
99,190
74,600
96,450
7,320
8,290
9,640
11,500
12,700
14,200
18,500
_
-
-
-
-
-
—
a
Number of
NOX Hew Process Lines"
c
c
290 3
640
900
1,240
1,590
14,257
14,263 7
14,264
14,265
14,266
14,268
18,954
14,300
18,430 13
1,400
1,580 1
1,840
2,200
2,430
2,720
3,540
—
-
.
-
-
-
—
Industrywide Emissions,
tons/yr
^articulate
c
c
0.04
0.08
0.11
0.15
0.19
5.82
5.82
5.82
5.82
5.83
5.83
7.74 '
10.8
14.0
0.08
0.09
0.11
0.13
0.14
0.16
0.21
21 .3C
24.55
26.7
27.6
29.3
31.0
33.0
sox
c
c
1.05
1.58
1.89
2.91
3.93
261
261
261
261
261
261
347
485
627
3.66
4.15
4.82
5.75
6.37
7.11
9.25
907C
1,049=
1,056
1,063
1,119
1,121
1,210
HOX
c
c
0.44
0.96
1.35
1.87
2.38
49.9
49.9
49.9
49.9
49.9
49.9
66.3
92.7
120
0.70
0.79
0.92
1.10
1.22
1.36
1.77
190C
218C
250
263
281
307
324
aEaission factors  from References 2 and 3.  Fuel consumption under each regulatory alternative is derived from
 Tables 8-21 and 8-31b.  Factors used  are for combustion of Mo. 4 fuel oil having 2 percent sulfur content by weight.
 (Source: Reference 4.)

bFrcw Table 8-43.

cEsti(Mt«s of secondary pollutants from polystyrene plants are not available,  and are not included in totals for
 regulatory alternatives.

dThese regulatory alternatives apply to a process line producing PET with a single end finisher.

•These regulatory alternatives apply to a process line producing PET with multiple end finishers.

f«here there is no enrission value for  a model  plant under a particular regulatory alternative, the emission value
 of the proceeding regulatory alternative for that model  plant category is used.
                                                   7-18

-------
Alternative I).   Under Regulatory  Alternative  II,  secondary air pollutant
emissions will  not increase appreciably  for  all  model  plants  except one
because the new process lines will  employ  fugitive emission control
programs, which do not require the use  of  add-on pollution control
systems.  The new PET process lines using  a  single end finisher under
Regulatory Alternative 2 are estimated,  on a per process  line basis, to
emit an additional 0.2 Mg/yr (0.2  tons/yr) of  particulate, 9.8 Mg/yr
(12.3 tons/yr)  of SOX, and 1.9 Mg/yr (2.3  tons/yr) of  NOX above baseline
control.
     Particulate emissions generated by new  process lines during  control
of VOC containing process streams  under Regulatory Alternative III  are
about 24.2 Mg/yr (26.7 tons/yr), or 25  percent greater than  secondary
particulate emissions under baseline conditions.  SOX  emissions under
Regulatory Alternative III are projected at  958 Mg/yr  (1,056  tons/yr):
a 16 percent increase in SOX emissions  over  Regulatory Alternative  I.
NOX emissions under Regulatory Alternative III are estimated  to be
about 227 Mg/yr (250 tons/yr), which represents a 31 percent increase
over baseline secondary NOX emissions.
     Projected particulate, SOX and NOX emissions from new  process  lines
operating under Regulatory Alternative  VII,  the most stringent combination of
regulatory alternatives, are 30.0 Mg/yr (33.0 tons/yr), 1,098 Mg/yr
(1,210 tons/yr) and 294 Mg/yr (324 tons/yr), respectively.   These emission
levels are equivalent to increases over baseline levels of  about  55
percent  for particulates,  34 percent for SOX, and 41 percent for  NOX.
Again, it should be noted  that actual secondary air pollutant emissions
probably will be  less than  the values presented due to fuel  combustion
requirements in many SIP's.5  To the extent that gas cleaning systems
are  used  (see Section 7.3), actual secondary air pollutant emissions
will be  reduced further.
7.2  WATER POLLUTION  IMPACTS
     Under the regulatory  alternatives  analyzed, increased water use
will occur only in  PET  process lines producing  a high viscosity product.
Water  usage  in a  PET  plant comprised of seven process lines  is estimated
to  increase  by 58,674 m3  (15.5 x  106 gal/year), or  about 7 percent of the
consumptive water requirements of  a continuous  polyester resin plant
with an  annual capacity of 105 Gg  (231  x  106 Ibs).6
                               7-19

-------
     Fugitive VOC emission sources can adversely affect water quality,
as leaking components that handle liquid hydrocarbon streams may increase
the waste level entering wastewater treatment systems.  Implementation
of leak detection and repair programs (Regulatory Alternative II) should
reduce the waste load on wastewater treatment systems by reducing the
amount of VOC that may leak from process equipment and enter the
wastewater systems.
7.3  SOLID WASTE DISPOSAL IMPACTS
     Solid waste impacts are anticipated to arise primarily from the
disposal of spent catalyst from catalytic incinerators.  Catalytic
incinerators are employed in a number of regulatory alternatives for
several different process lines.
     The quantity of spent catalyst expected to be generated annually by
new process lines is estimated by the equation:
          DTj- = [Raij x Q x 0.02382 m3/ft3 x n^ * 3]       (5)
    where Dfj = Annual quantity of spent catalyst generated under
                Regulatory Alternative j_ (j = I, II, III or IV);
         Rajj = Exhaust gas flow rate (scfm) from process section a_
                (a = raw materials preparation, polymerization reaction,
                material recovery, product finishing, or product storage)
                in a process line in model plant T_ (i = liquid phase
                polypropylene and other process types) under Regulatory
                Alternative j_ (refer to Tables 6-1 to 6-11);
            Q = 2.25 ft3 of catalyst/1,000 scfm of catalytic incineration
                throughput (Section 8.1.3.1);
           Nj = number of new process lines of model plant type i_ expected
                to be built by 1988 (Table 8-43).
It is assumed that the catalyst bed has a useful life of 3 years
(Section 8.1.3.1).
     The solid waste disposal impacts of the regulatory alternatives are
presented in Table 7-4.  The total solid waste disposal impact of the
regulatory alternatives for the 44 new process lines expected to employ
catalytic incinerators is projected to be 3.35 m3/yr (119 ft3/yr)
under Regulatory Alternative III, 3.90 m^/yr (138 ft3/yr) under
Regulatory Alternative IV, and 4.21 m3/yr (149 ft3/yr) under Regulatory
Alternatives V through VII.  The volume of solid waste generated by
                               7-20

-------
Table  7-4.   INDUSTRYWIDE SOLID WASTE  IMPACTS  OF THE REGULATORY
      ALTERNATIVES FOR NEW POLYMER AND RESIN PROCESS LINES3
Model Regulatory Flow,'5
Plant Alternative scfm
LOPE, High
Pressure



LDPE, Low
Pressure and
HOPE, Gas
Phase

HOPE, Slurry


HOPE, Solution


EPS/Post-
impregnation





EPS/In-situ






TOTALS






1
2
3
4
5
1
2
3
4
5
1
2
3
1
2
3
1
2
3
4
5
6
7
1
2
3
4
5
6
7
I
,11
III
IV
V
VI
VII
0
0
1,590
1,990
1,990
0
0
0
1,070
1,070
0
0
230
0
0
16,240
0
0
1,890
1,890
8,405
8,405
8,405
0
0
0
1,020
1,615
1,615
1,615
_
-
-
-
-
-
"
Solid Solid
Waste Waste
Generated.e Number of New Generated,
m3(ft3)/yr5 Process Linesd m3(ft3)/yr
0
0
0.034 (1.2) 4
0.042 (1.5)
0.042 (1.5)
0
0
0 20
0.023 (0.8)
0.023 (0.8)
0
0 6
0.005 (0.2)
0
0 9
0.345 (12.2)
0
0
0.04 (1.4)
0.04 (1.4) 2
0.18 (6.3)
0.18 (6.3)
0.18 (6.3)
0
0
0
0.022 (0.77) 3
0.034 (1.2)
0.034 (1.2)
0.034 (1.2)
_
_
-
-
-
-
™
0
0
0.14 (4.8)
0.17 (6.0)
0.17 (6.0)
0
0
0
0.45 (16)
0.45 (16)
0
0
0.03 (1.2)
0
0
3.1 (110)
0
0
0.08 (2.8)
0.08 (2.8)
0.36 (12.6)
0.36 (12.6)
0.36 (12.6)
0
0
0
• 0.07 (2.3)
0.10 (3.6)
0.10 (3.6)
0.10 (3.6)
0
0
3.35 (119)
3.90 (138)
4.21 (149)
4.21 (149)
4.21 (149)
aSolid waste generated through use of catalytic incinerators.

bPer process line.  See Tables 6-1 through 6-llb and  Tables 8-21  through
 8-31b, as appropriate.

cBased on 2.25 ft3 of catalyst/1,000 scfm of catalytic incinerator throughput
 and a three year catalyst life.

dSee Table 8-43.

eWhere there is no value for a model plant under a  particular regulatory
 alternative, the value of the preceding regulatory alternative for that
 model plant category is used.
                                 7-21

-------
the use of spray towers or liquid jet scrubbers employed upstream
of catalytic incineraters to treat corrosive exhaust gases under
Regulatory Alternative VII represents about 0.01 percent of the
volume of biological sludge generated annually as solid waste by
polyethylene process operations.7 Volumes of biological sludge
generated by applicable process operations in new process lines
are presented in Table 7-5.  Disposal of spent catalyst is not
expected to be affected by the Resource Conservation and Recovery
Act of 1976 (RCRA), since catalyst for VOC incinerator is not
listed as a hazardous waste under 40 CFR 261.30.8 Further, the
high price of the platinum or palladium catalyst may encourage
recycling where feasible.
     Solid waste may be generated as a result of implementing the regulatory
alternatives in the form of ash collected from control of the secondary
air pollutants, discussed in Section 7.1.3.  The quantity of fly ash
generated is estimated by assuming the installation of a scrubbing
system on oil-fired boilers to control both sulfur oxides and particulates
that achieve a 60 percent control efficiency for particulate emissions
and a 95 percent removal efficiency for SOY emissions.   Applying these
                                          s\
efficiences to the secondary pollutants presented in Table 7-3 and
assuming a fly ash density factory of 0.72 Mg/m3 (45 lb/ft3),10 the
maximum volume of fly ash generated industrywide by new polymers and
resins'process lines is projected to be about 1,325 m3/yr (46,750
ft3yr) under Regulatory Alternative VII, the alternative having
the greatest potential for secondary pollutant emissions.  This
fly ash generation rate represents 1.2 to 3.4 percent of the total
volume of biological sludge estimated to be generated by product
manufacturing operations, or a worst-case scenario fly ash generation
factor of about 0.01 Mg/Mg process VOC emissions controlled (Table
7-3).  Disposal of fly ash should not be affected by RCRA, as fly
ash from fuel  combustion is specifically exempted from RCRA by 40
CFR 261.4(b)(4).n
   Solid wastes generated by fugitive VOC leak detection and repair
programs include replaced mechanical  seals, seal packing, rupture disks,
and valves.  The solid waste impacts of fugitive VOC emission reduction
                               7-22

-------
Table 7-5.   VOLUME OF  BIOLOGICAL SLUDGE  GENERATED BY  PROCESS OPERATIONS  IN
      NEW POLYMER  AND RESIN PROCESS  LINES THAT EMPLOY  FLARES, THERMAL
                    INCINERATION,  OR CATALYTIC  INCINERATION
Model Plant
PP /Li quid
PP/Gas
LDPE/High
LDPE/Low
Pressure
and HOPE/
Gas Phase
HDPE/Slurry
HDPE/Solution
PS/Continuous
EPS/Post-
impregnation
EPS/In-situ
PET/DMT
PET/TPA
TOTAL
Annual Production Industrywide
Rate Per Process Biological Sludge
Line,0 Number of New Production Rate,d
Gg/yr Ob/yr) Process Linesc n^/yr (ft3/yr)
50
35
70
75
70
30
37.5
25
8
15
ro H-»
o tn
-h n>
-
(no
(77
(154
(165
(154
(66
(83
(55
(18
(33
(33
(44

x 106)
x 106)
x 106)
x 106)
x 106)
x 106)
x 106)
x 106)
x 106)
x 106}
x 1Q6)
x 106
-
9
9
4
20
6
9
2
2
3
7
13
1
85
4,850-12,000(171,000-431,000)
3,400-8,520(120,000-301,000)
3,020-7,580(107,000-268,000)
16,200-40,600(572,000-1,430,000)
4,530-11,400(160,000-403,000)
2,910-7,310(103,000-258,000)
813-2,040(28,700-72,000)
539-1,350(19,000-47,700)
264-664(9,320-23,400)
2,840-11,400(100,000-403,000)
5,280-21,150(186,000-747,000)
541-2,170(19,100-76,600)
39,400-103,000(1.39 x 106-3.64 x 106}9
27,500-68,900(970,000-2.43 x lQ6)9,n
   •	. -—— .._.». ..vv « jrf«( *, v i  *A*IJ  i liyw i a uw i jr  a i uci na (*( vtr  ana ij iuu ror rLI/lrrt prOCSSS
   lines.  For illustration purposes, however, this model plant is  included in this table.

  bFrom Tables 6-1 through 6-llb.

  cFrora Table 8-43.

  d8ased on 49-123 ra3 sludge/yr/10 x 106  Ibs. of polypropylene,  polyethylene, and polystyrene
   produced; 123-493 nr* sludge/yr/10 x 105 Ibs of polyester produced.  (Source:  Reference 7.)

  sSingle end finisher.

  ^Multiple end finishers.

  9Does not include PET/TPA numbers.

  "These totals include only sludge  generated by process operations employing catalytic
   incineration under the regulatory alternatives.
                                       7-23

-------
programs are not anticipated to be significant because of the ability to
recycle metal solid wastes and the small quantity of wastes generated.
7.4  ENERGY IMPACTS
7.4.1  Model Plant Energy Impacts
   The energy impacts of the regulatory alternative for process lines in
each model plant are calculated as the sum of energy expended to control VOC
emissions from process operations and the energy credit for reduced VOC
losses by implementing fugitive VOC emission control practices and process
VOC emission recovery practices.  For the purposes of this analysis, energy
impacts of process emission control are assumed to consist of natural gas
consumption by flares, thermal incinerators, and catalytic incinerators,
electricty consumption by thermal incinerators, catalytic incinerators,
and condensers, and steam consumption by flares.  The methods used to
calculate energy consumption are discussed in Section 8.1.  The energy
impacts of controlling VOC emissions from a process line in each model
plant under each regulatory alternative are presented in Table 7-6.
     Energy impacts associated with process VOC emission control increase
when moving from a less restrictive regulatory alternative to a more
restrictive alternative because additional energy-consuming control
techniques are used to reduce VOC emissions.  Energy impacts associated
with control of fugitive VOC emissions are presented as negative values
(energy credits or positive impacts) in Table 7-6.  The energy credit for
reduced VOC loss under fugitive VOC control practices is greater in
absolute terms than the energy cost of the fugitive VOC control.  Energy
impacts of fugitive VOC emission control are constant once the leak
detection and repair program is implemented, as the leak detection
inspection intervals, equipment specifications, and resulting emission
rates for each model plant type do not change under Regulatory
Alternatives II through VII.
     Table 7-6 shows that the energy impact of process VOC emission
reduction in a model plant increases in moving from a less stringent to a
more stringent regulatory alternative.  The magnitude of the total  energy
impact value is less than the impact of controlling only VOC emissions
from process operations because the energy credit resulting from fugitive
VOC emission control work practices at least partially offsets the  energy
impact of process emission controls.

                                7-24

-------
Table 7-6a.   ENERGY  IMPACTS OF THE REGULATORY  ALTERNATIVES  FOR POLYMER
                               AND RESIN PLANTS3
                   Process Emissions Control Energy Impact (GJ/yr)
   Fugitive       Combined
Emission Control Emission Control
  Energy Impact
Regulatory
Model Plant Alternatives
PP/Liquid



PP/Gas

LDPE/High
Pressure



LDPE/Low Pressure
HDPE/Gas Phase



HDPE/Slurry


HDPE/Solution


PS/Continuous9






EPS/Post-
impregnation




1
2
3
4
1
2
1
2
3
4
5
1
2
3
4
5
1
2
3
1
2
3
1
2
3
4
5
6
7
1
2
3
4
5
6
7


Natural Gasb Electricity0
2,260
2,260
2,910
3,670
2,260
2,260
754
754
1,770
2,020
5,040
4,540
. 4,540
5,290
7,200
8,700
1,510
1,510
1,600
2,260
2,260
32,000
_
-
-
-
-
-
"

-
2,290
32,850
45,000
165,900
165,900
_.
-
23
23
_
"*
.
-
257
321
321
-
-
172
172
_
-
37
_
-
263
_
-
37
40
55
57
72

-
307
307
1,360
1,360
1,360

Steamd
2,720
2,720
2,720
2,730
3,210
3,210
862
862
862
862
6,720
3,360
3,360
3,360
3,360
4,640
1,820
1,820
1,820
1,240 -
1,240
1,240
_
-
-
-
-
-
-

_
0
679
679
785
785

Process Total6
4,990
4,990
5,660
6,430
5,470
5,470
1,620
1,620
2,890
3,200
12,100
7,890
7,890
8,650
10,700
13,500
3,330
3,330
3,460
3,500
3,500
33,500
0
0
37
40
55
57
72

_
2,600
33,800
47,000
168,000
168,000
Mg VOC
Reduced
0
20
20
20
0
20
0
15
15
15
15
0
30
30
30
30
0
20
20
0
20
20
0
30
30
30
30
30
30
Q
12
12
12
12
12
12

GJ/yrf
0
-915
-915
-915
0
-915
0
-706
-706
-706
-706
0
-1,410
-1,410
-1,410
-1,410
0
-952
-952
0
-952
-952
0
-1,230
-1,230
-1,230
-1,230
-1,230
-1,230
0-
-516
-516
-516
-516
-516
-516

GJ/yr
4,990
4,080
4,750
5,510
5,470
4,560
1,620
910
2,180
2,500
11,400
7,890
6,480
7,240
9,320
12,100
3,330
2,380
2,510
3,500
2,550
32,570
0
-1,230
-1,358"
-l,360h
-1,351"
-1,349"
-l,341h

-516
2,080
33,300
46,500
167,500
167,500
                                     7-25

-------
       Table  7-6a.
ENERGY  IMPACTS OF  THE REGULATORY  ALTERNATIVES  FOR POLYMER
           AND  RESIN  PLANTS3  (concluded)
Fugitive Combined
Emission Control Emission Control
Process Emissions Control Energy Impact (GJ/yr)
Model Plant
EPS/In-situ






PET/DMT'






PET/TPA*

PET/TPA*






Regulatory
Alternatives
1
2
3
4
5
6
7
1
Z
3
4
5
6
7
1
2
1
2
3
4
5
6
7
Natural Gasb
_
-
760
2,100
3,165
4,210
5,233
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Electricity0
_
-
0
164
261
261
261
2,400
2,410
2,420
2,420
2,420
2,425
2,425
2,400
2,400
2,090
2,090
2,090
2,090
2,090
2,090
2,090
Steamd
_
-
326
326
326
642
958
32,350
32,350
32,350
32,350
32,350
32,350
43,775
32,350
42,525
1,320
1,770
2,400
3,270
3,840
4,530
6,520
Process Total8
„
-
1,090
2,590
3,750
5,110
6,450
34,750
34,760
34,770
34,770
34,770
34,775
46,200
34,750
44,925
3,410
3,860
4,490
5,360
5,930
6,620
8,610
Energy
Mg VOC
Reduced
0
5.7
6.7
6.7
6.7
6.7
6.7
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
Impact Energy Impacts
GJ/yrf
0
-290
-290
-290
-290
-290
-290
0
0
0
d '
0
0
0
0
•o
0
0
0
0
0
0
0
• GJ/yr
0
-290
800
2,300
3,460
4,820
6,160
34,750
34.720J
34,730k
34,7200
34, 720 J
34, 730 J
46,100J
34,750
44.375J
3,410
3.730J
4.230J
4.950J
5,460J
6.080J
7.940J
al»pacts are reported on  a per process line basis.
     8tu * 1.0551 GO.   This factor includes the energy conversion efficiency; it is  not an energy equivalent.
 (Reference 12.)
cl Mfh  » 9.476 x 10"3  GJ.  This factor includes the energy conversion  efficiency;  it is not an energy equivalent.
 (Reference 12.)
^l.OOO  Ib steam « 1.525 GJ.  This factor  includes the energy conversion efficiency;  it is not an energy equivalent.
[0.2482 bbl distilled  fuel oil/1,000 Ib steam.]
*1 GJ » 0.1628 bbl crude oil.  (Reference 13.)
fSased  on the following:   17,598 8tu/lb of styrene monomer; 19,683 Btu/lb of propylene; 20,276 Btu/lb of ethylene;
 7,310  Stu/lb of athylene  glycol; and 19,499 3tu/lb of pentane.  (Reference 14.)
9Esti»ates of energy use under baseline control are not  available.
^Includes energy credit of 360 Gg for recovered styrene  monomer.
'these  regulatory alternatives apply to a process line producing PET with a single end finisher.
^Includes energy credit for recovered ethylene glycol.
''These  regulatory alternatives apply to a process line producing PET with multiple end finishers.
Mote:  Values on some  lines may not total exactly due to rounding,  negative values  indicate energy credits.
                                                    7-26

-------
Table 7-6b.  ENERGY IMPACTS OF THE REGULATORY ALTERNATIVES FOR POLYMER
                           AND RESIN PLANTS3
Process Emissions Control Energy
Impact
Regulatory Natural Gas,11 Electricity,1- Steam," Process Total ,e
Model' Plant Alternatives 10° Btu/yr kWh/yr 1,000 Ib/yr bbl o1l/yr
PP/L1quid ,



PP/6as

IDPE/High
Pressure



LDPE/Low Pressure
HDPE/Gas



HDPE/Slurry


HDPE/Solut1on


PS/Cont1nuous9






EPS/Post-
impregnation





1
2
3
4
1
2
1
2
3
4
5
1
2
3
4
5
1
2
3
1
2
3
1
2
3
4
5 '
6
7
1
2
3
4
5
6
7
2,145
2,145
2,760
3,475
2,145
2,145
715
715
1,675
1,915
4,775
4,300
4,300
5,015
6,820
8,250
1,430
1,430
1,515
2,145
2,145
30,350

_
_
_
_
_
-
—
-
2,170
31,130
42,650
157,230
157,230
.
-
2,450
2,450
_
-
.
-
27,140
33,880
33,880
„
-
_
18,160
18,160
.
-
3,875
_;
-
27,780
.
_
3,878
4,224
5,837
6,061
7,571
_
-
32,450
32,450
143,880
143,880
143,880
1,789
1,789
1,789
1,794
2,105
2,105
565
565
565
565
4,405
2,200
2,200
2,205
2,205
3,040
1,195
1,195
1,195
810
810
810

_
_
_
_
_
-
_
-
0
445
445
515
515
813
813
922
1,046
891
891
263
263
470
521
1,965
1,285
1,285
1,410
1,750
2,200
542
542
563
570
570
5,460
0
0
6
7
9
9
12
_
-
423
5,510
7,660
27,400
27,400
Fugitive
Emission Control
Energy Impact
Mg VOC
reduced
0
22
22
22
0
22
0
17
17
. 17
17
' 0
33
33
33

0
22
22
0
22
22
0
33
33
33
33
33
33
0
13
13
13
13
13
13
Combined
Emission Control
Energy Impact
bbl oil/yrf bbl /oil /yr
0
-149
-149
-149
0
-149
0
-115
-115
-115
-115
0
-230
-230
-230

0
-155
-155
0
-155
-155
0
-200
-200
-200
-200
-200
-200
0
-84
-84
-84
-84
-84
-84
813
664
773
897
891
742
263
148
355
406
1,850
1,285
1,055
1,180
1,520
1,970
542
387
408
570
415
5,305
0
-200
-221 h
-221"
-220"
-220"
-218&
0
-84
339
5,420
7,570
27,300
27,300
                                7-27

-------
     Table  7-6b.   ENERGY  IMPACTS  OF  THE  REGUALTORY ALTERNATIVES  FOR  POLYMER
                                   AND RESIN  PLANTS3  (concluded)
-ugitive Combined
Emission Control Emission Control
Process Emissions Control Energy
Model Plant
EPS/!n-s1tuf






PET/DMT*






PET/TPA*

PET/TPA*






Regulatory
Alternatives
1
2
3
4
5
6
7
1
2
3
4
5
6
7
I
2
1
2
3
4
S
6
7
Natural Gas," Electricity
10b 3tu/yr kWh/yr
_
-
720
1,990
3,000
3,990
4,960
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
_
-
0
17,350
27,550
27,550
27,550
253,060
254,774
254,836
255,101
255,448
255,856
255,856
253,060
253,060
220,815
220,815
220,815
220,815
220,815
220,815
220,815
,*• Steam,"
1,000 Ib/yr
„
-
214
214
214
421
628
21,215
21,215
21,215
21,215
21,215
21,215
28,705
21,215
27,885
864
1,160
1,573
Z.141
2,516
2,969
4,278
Impact
Process Total ,B
bbl oil/yr
.
.
177
422
611
832
1,050
5,657
5,660
5,560
5,661
5,661
5,662
7,521
5,657
7,313
555
629
731
872
965
1,080
1,400
Energy Impact
Mg VOC
reduced
0
9
9
9
9
9
9
0
0
0
0
0
.0
0
0
0
0
0
0
0
0
0
0
bbl oil/yrf
0
-47
-47
-47
-47
-47
-47
0
0
0
o 	
o 	
0
0
0
0
0
0
0
o •
0
0
0
Energy Impact
bbl/oil/yr
0
-47
130
375
564
785
1,000
5,657
5,653J
5.653J
	 5,553J
5.653J
5.554J
7.505J
5,657
7.306J
555
608
688
806
888
990
1,293
a!i8pacts ar* on a per process line basis.
     8tu » 1.0551 SJ.   This factor includes the energy conversion efficiency; it is not an energy equivalent.
 (Reference 12).
cl icHh » 9.476 x 10~3  GJ.  This factor includes the energy conversion efficiency; it is not an energy  equivalent.
 (Reference 12.)
^1,000 Ib steam * 1.525 GJ.  This factor includes the energy conversion  efficiency; it is not an energy equivalent.
CO. 2482 bbl distilled  fuel oil/1, 000 Ib steam.]
el SO • 0.1628 bbl crude oil.  (Reference 13.)
       on  the following:   17,598 Btu/lb of styrene monomer; 19,683 3tu/1b of propylene;  20,276 3tu/lb  of sthylene;
 7,810 3tu/lb of athylene  glycol ;  and 19,499 3tu/lb of pentane.  (Reference 14.1
9E$t1(iHtes of energy use under baseline control are not available.
"Includes  energy credit for recovered styrene monomer.
'these regulatory alternatives apply to a process line producing PET with a single end  finisher.
•J Includes  energy credit for recovered ethyl ene glycol.
''These regulatory alternatives apply to a process line producing PET with multiple end  finishers.
Mote:   Values on some lines may not total exactly due to  rounding,  negative values indicate energy  credits.
                                                  7-28

-------
 7.4.2   Industrywide Energy Impacts
      Process line energy impacts presented in Table 7-6 are used to
 project industrywide energy impacts of the regulatory alternatives on
 new polymers and resins process lines expected to be built by the
 year 1988.  The industrywide energy impacts of process and fugitive VOC
 emissions control in new process lines operating under each regulatory
 alternative are presented in Table 7-7.
      Under Regulatory Alternative I, industrywide energy demand is
 estimated to be about 1,010 TJ/yr (164,000 bbl crude oil/yr).   Under
 Regulatory Alternative II,  industrywide energy demand is estimated to be
 about 1,140 TJ/yr (186,000 bbl  oil/yr).  The total  energy demand increases
 when shifting from baseline control  (Regulatory Alternative I)  to Regulatory
 Alternative II due primarily to the process controls associated with poly-
 (ethylene terephthalate)  plants using  a single end  finisher in  each  process
 line.   Implementation of  fugitive VOC  control  practices  and increased recovery
 from process  emissions  in  polyethylene terephthalate) process  lines
.saves  about 67 TJ/yr  (11,000 bbl  oil/yr).   Energy impacts  under Regulatory
 Alternative VII,  the  most stringent  combination  of  regulatory alternatives,
 are approximately 2,020 TJ/yr  (329,000 bbl  oil/yr);  the  increase is  due
 to  a greater  energy demand  in controlling  process emissions.
     In  most  instances, the incremental  energy requirements of  the most
 stringent  regulatory  alternatives are  increases  over the energy  required
 under  baseline  (Regulatory  Alternative 1)  control.   Excluding the
 polystyrene process lines because no estimate of energy usage under
 baseline control  is available and the  polyethylene  terephthalate)
 process  lines because no estimate of energy usage for making the polymer
 is  available, the next  largest  incremental percent and absolute  size
 increase associated with the most stringent regulatory alternative over
 baseline control  is for a process line producing high density polyethylene
 using a  solution process.   The absolute size of this energy requirement
 compared to the total  energy used in making the polymer is small.  A
HOPE, solution process model plant comprised of three process lines would
use approximately 23.8 x Ifl3 TJ/yr (3.88 x 10$ bbl  oil/yr).15 Total  energy
consumed by process emission controls under the most stringent
alternative for this model  plant is 100 TJ/yr (16.4  x 106 bbl  oil/yr),
less than 0.5  percent of the energy required for the production  of the
polymer.
                                   7-29

-------
Table 7-7a.  INDUSTRYWIDE ENERGY IMPACTS OF THE REGULATORY ALTERNATIVES
               FOR NEW POLYMER AND RESIN PLANTS (TJ/yr)
Regulatory
Model Plant Alternative
PP/Uqutd



PP/Gas

LDPE/HIgh
Pressure



LDPE/Low Pressure
HOPE/Gas Phase



HDPE/Slurry


HDPE/Solut1on


?S/Cont1nuous






EPS/Post-
Inpregnatlon





1
2
3
4
1
2
1
2
3
4
5
1
2
3
4
.5
1
2
3
1
2
3
1
2
3
4
5
6
7
1
2
3
4
5
6
7
r Process
(S Process
4.99
4.99
5.66
6.43
5.47
5.47
1.62
1.62
2.89
3.20
12.1
7.89
7.89
8.65
10.7
13.5
3.33
3.33
3.46
3.50
3.50
33.5
_
-
0.037
0.04
0.055
0.057
0.072
_
-
2.60
33.8
47.0
168
168
Line Energy Emissions3 Number of New Industrywide Energy Emissions
Fugitive1-
0
-0.92
-0.92
-0.92
0
-0.92
0
-0.71
-0.71
-0.71
-0.71
0
-1.41
-1.41
-1.41
-1.41
0
-0.95
-0.95
0
-0.95
-0.95
0
-1.23
-1.23
-1.23
-1.23
-1.23
-1 .23
0
-0.52
-0.52
-0.52
-0.52
-0.52
-0.52
Combined1- Process Lines0 Process
4.99
4.08 9
4.75
5.51
5.47
4.56 9
1.62
0.91 4
2.18
2.50
11.4
7.89
6.48
7.24 20
9.32
12.1
3.33
2.38 6
2.51
3.50
2.55 9
32.6
0
-1.23. 2
-1.36d
-1.36d
-1.35d
-1.35d
-1.34
-------
       Table  7-7a.   INDUSTRYWIDE  ENERGY  IMPACTS  OF- THE REGULATORY  ALTERNATIVES
                   FOR NEW  POLYMER AND  RESIN  PLANTS (TJ/yr)  (concluded)
Regulatory
Model Plant Alternatives
EPS/In-situ






PET/DMTe






PET/TPAS

PET/TPA9






TOTAL"






40n a per process
1
Z
' 3
4
5
6
7
1
2
3
4
5
5
7
1
2
1
2
3
4
5
6
7
I
II
III
IV
y
VI
YII
me basis.
Process
Process
^
-
1.09
2.59
3.75
5.11
6.45
34.8
34.8
34 .'8
34.8
34.3
34.8
46.2
34.8
44.9
3.41
3.86
4.49
5.36
5.93
5.62
8.61
-
-
-
-
-
-
"
rroffl rabl e /
uine Energy Emissions3
Fugitive*-
0
-0.29
-0.29
-0.29
' -0.29
-0.29

0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
_
-
-
-
-
-
—
r-6.
^OiiiD I (lea
0
-0.29
0.80
2.30
3.46
4.82
6.16
34.8
34. ?f
34. 7*
34. Jf
34. 7f
34. 7f
46. 2*1
34.8
44. 9f
3.41
3.73^
4.23f
4.95f
5.46f
5.08f
7.94f
.
-
-
-
-
-
"

'lumber of Mew Industrywide Energy Emissions
Process Lines'3 Process

..
3.34
3 7.78
11.3
15.3
19.4
243
7 243
243
243
243
243
323
13 • 452
584
1 3.4
3.9
4.5
5.4
5.9
6.6
8.6
1,008
1,141
1,448
85 1,565
1,687
1,934
2,020

-ugi cive1-
0
-0.37
-0.87
-0.87
-0.87
-0.87
-0.87
0
0
0
0'
0
0
0 -
. 0
0
0
0
0
o
0
0
0
0
-66
-66
-66 .
-66
-66
-66

Combined
0
-0.87
2.37
6.9
10.4
14.5
18.5
243
243 f
243 f

243f
243f
323f
452
583 f
3.4
3.7
' 4.2
5.0
5.5
6.1
7.9
1,008
1,074 •
1,380
1,497
1,519
1,866
1,951

bFrom Table 8^3.
cNegative values Indicate energy credits.
dlncludes energy cradit for recovered styrene monomer.
eThese regulatory alternatives apply to a process line producing PET with a single end finisher.
^Includes energy credit for recovered athylene glycol.
SThese regulatory alternatives apply to a process line producing PET with multiple end finishers.
hWhere there is no value for a plant under a particular regulatory alternative, the value of  the orece°dinq requlatorv
 alternative for that model  plant category is used.                                         '
NOTES:  Clashes indicate that the entry is not applicable.
                                               '  7-31

-------
Table 7-7b.  INDUSTRYWIDE ENERGY IMPACTS OF THE REGULATORY ALTERNATIVES
         FOR NEW POLYMER AND RESIN PLANTS (1,000 bbl  oil/yr)
Regulatory
Modal Plant Alternatives
PP/Liquld



PP/Gas

LDPE/High
Pressure



LOPE/Low Pressure
HDP£/Gas Phase



HOPE/SI urry


HDPE/Solutlon


PS/Continuous






EPS/Post-
impregnatlon





1
2
3
4
I
2
1
2
3
4
5
1
2
3
4
5
1
2
3
1
2
3 •
1
2
3
4
5
6
7
1
2
3
4
5
6
7
Process Line Energy Emissions3 Number of Ne
Process
0.813
0.813
0.922
1.05
0.89
0.89
0.263
0.263
0.470
0.521
1.97
1.29
1.29
1.41
1.75
2.20
0.542
0.542
0.563
0.57
0.57
5.46
—
-
0.006
0.007
0.009
0.009
0.012
_
-
0.423
5.51
7.66
27.4
27.4
Fugitive^
0
-0.149
-0.149
-0.149
0
-0.149
0
-0.115
-0.115
-0.115
-0.115
0
-0.23
-0.23
-0.23
-0.23
0
-0.155
-0.155
0
-0.155
-0.155
0
-0.20
-0.20
-0.20
-0.20
-0.20
-0.20
0
-0.084
-0.084
-0.084
-0.084
-0.084
-0.084
Combined0 Process Line
0.812
0.663 9
0.773
0.901
0.89 9
0.741
0.263
0.148
0.355 4
0.406
1.86
1.29
1.06
1.18 20
1.52
1.97
0.542
0.387 6
0.408
0.57
0.415 9
5.31
0
-0.20 2
-0.221
-0.221d
-0.22d
-0.22d
-0.218d
0
-0.084
0.339
5.43 2
7.58
27.3
27-. 3
iw Industrywide Energy Emissions
is^ Process Fugitive1- Combined1-
7.31
7.31
8.30
9.41
8.02
8.02
1.05
1.05
1.88
2.09
7.86
25.7
25.7
28.2
34.9
44.0
3.25
3.25
3.38
5.13
5.13
49.1
0
0
0.012
0.013
0.018
0.019
0.023

_
0.85
11.0
15.3
54.7
54.7
0
-1.34
-1.34
-1.34
0
-1.34
0
-0.46
-0.46
-0.46
-0.46
0
-4.6
-4.6
-4.6
-4.6
0
-0.93
-0.93
0
-1.40
-1.40
0
-0.40
-0.40
-0.40
-0.40
-0.40
-0.40
0
-0.17
-0.17
-0.17
-0.17
-0.17
-0.17
7.31
5.97
6.96
8.07
8.02
6.68
1.05
0.59
1.42
1.63
7.40
25.7
21.1
23.6
30.3
39.4
3.25
2.32
2.45
5.13
3.73
47.7
0
-0.40
-0.442d
-0.443d
-0.440d
-0.439d
-0.437d
0
-0.17
0.68
10.9
15.2
54.6
54.6
                               7-32

-------
       Table  7-7b.   INDUSTRYWIDE  ENERGY  IMPACTS  OF  THE REGULATORY  ALTERNATIVES
            FOR NEW POLYMER AND  RESIN  PLANTS  (1,000 bbl  oil/yr)  (concluded)

Model Plant
£PS/In-situ






PET/OMT6






PET/TPA6

PET/TPA9






TOTAL"






Regulatory
Alternatives
1
2
3
4
5
6
7
1
2
3
4
5
6
7
1
2
1
2
3
4
5
6
7
I
II
III
IV
V
YI
VII
Process Line Energy Emissions3 Mumber of Mew. Industrywide Energy
Process
_
-
0.177
0.422
0.611
0.832
1.05
5.66
5.66
5.66
5.66
5.66
5.66
7.52
5.66
7.31
0.56
0.53
0.73
0.37
0.97
1.08
1.40
_
-
-
-
-
-
-
rugi tive1-
0
-0.047
-0.047
-0.047
-0.047
-0.047
-0.047
0
0
0
0
0
0
0
0
0
0
0
0
0
0 .
0
0
_
-
-
•
.
-
-
Combined1- Process Lines" Process
0
-0.047
0.129
0.374 3
0.564
0.785
. 1.0
5.66
5.65^ 7
5.65f
5.65f
5.65'
5.65f
7.51'
5.66 13
7.3lf
0.56
0.61f
0.69f
0.81f 1
0.89f
0.99f
1.29f
35

-
-
.
-
-
.
-
0.53
1.27
1.33
2.50
3.15
39.6
39.6
39.6
39.6
39.6
39.5
52.6
73.6
95.0
0.56
0.63
0.73
0.87
0.97
1.08
1.40
164
186
236
255
275
315
329
Fugi tive*-
0
-0.14
-0.14
-0.14
-0.14
-0..14
-0.14
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
-10.8
-10.8
-10.8
-10.8
-10.8 .
-10.8
Emissions
Combined'-
0
-0.14
0.39
1.12
1.59
2.36
3.01
39.5
39. of
39. 6f
39. 5f
39. 6f
39. of
52. 6T
73.6
95. Of
0.56
0.61f
0.69f
0.81f
0.89f
0.99f
1.29f
164
175
225
244
264
304
318
 aOn a per process line basis.  From Table 7-6.
 bFrom Table 8-43.
 °Negative values indicate energy credits.
 Includes energy credit for  recovered styrene monomer.
,eThese regulatory alternatives apply to a process line producing PET with a single end  finisher.
 ^Includes energy credit for  recovered ethylene glycol.
 9These regulatory alternatives apply to a process line producing PET with multiple end  finishers.
 "Where there is no value for a plant under a particular regulatory alternative, the value
  of the proceeding regulatory alternative for that model plant category is used.
 NOTES:  Dashes indicate that the entry is not applicable.
        Values on some lines may not total exactly due to rounding.
                                                   7-33

-------
 7.5  OTHER ENVIRONMENTAL IMPACTS
 7.5.1  Noise Impacts
      Flares can be a source of noise pollution.   Noise  generated  during
 flaring results from unsteadiness  in the  combustion  process  and steam
 injection.16.17  Noise levels  in excess of  100  decibels  (dB)  have been
 recorded within 100 meters  (330 feet)  of  the  flare tip,18  but attenuate
 with increasing distance from  the  flare.19
      Noise from combustion  by  a flare  can be  reduced by  partially enclosing
 the flare tip with acoustical  materials,  though  this practice may not be
 feasible for elevated flares:  the  enclosure's useful working  life may be
 shortened because of direct flame  impingement, high  exit gas  velocities,
 and openness to weather conditions.  Noise  from  steam injection can be
 minimized by use of properly positioned multiport steam  jets,  partial
 enclosure of the elevated flare tip, or use of shrouding and  acoustical
 barriers for the steam jets and injectors.17  Noise  impacts also  can be
 mitigated by placing the flare  as  far  as  practicable from the  plant
 boundaries,  though  this practice does  not attenuate lower frequency
 noise  (less  than 500 HZ)  as effectively as higher frequency noise.19
 Thus,  by employing  the proper  flare  design and site selction,  potential
 noise  impacts  on community  areas surrounding each affected polymers and
 resins  plant should be minimal.
 7.5.2   Irreversible and Irretrievable  Commitment of Resources
      In  general,  Regulatory Alternative II will  require no process VOC
 emission controls beyond  those  required under current practices (Regulatory
 Alternative  I).   As  indicated in Section  7.4, the fugitive VOC leak
 detection  and  repair program implemented  under Regulatory Alternative II
 will result  in energy  savings for most plants.  Thus, Regulatory Alternative
 II  appears to  require  no  irreversible  and irretrievable commitment of
 resources beyond what  is currently committed under industry practices.
     Regulatory Alternatives III through VII require installation  of
 additional pollution control equipment to  reduce VOC  emissions from
 process  operations.  Consequently, demand  for metals, refractory,
 electrical equipment, and other raw materials needed  to  manufacture VOC
 emission control equipment will increase,  though it  is  unlikely that
the amount of resources used by the new process  lines to meet Regulatory
Alternatives III through VII will be significant in comparison to
                                  7-34

-------
industrywide consumption of the same resources.   It is possible that
some materials used to construct the additional  VOC controls  ultimately
will be salvaged and recycled at the end of the  control  device's useful
life, thereby reducing the amount of resources permanently committed to
control of VOC emissions.  Guy wire-supported flares and settling ponds
for scrubber wastewater may require considerable land space,20"22 but
are not expected to commit significant additional  areas of land than
what is currently used for manufacturing operations at polymers and
resins plant sites.  Therefore, no irreversible  and irretrievable
commitment of resources is expected in meeting the requirements of
Regulatory Alternatives III through VII.
7.5.3  Environmental Impacts of Delayed Regulatory Action
     The 5-year impact of delaying implementation of any regulatory
action regarding control of VOC from new polymers and resins  process lines
depends on the regulatory alternative considered.   Assuming that current
industry VOC control practices (Regulatory Alternative I) are maintained
through year 1988, nationwide VOC emissions may  increase by as much as
1,495 Mg/yr (1,648 tons/yr) in the absence of Regulatory Alternative II,
and as much as 4,560 mg/yr (5,024 tons/yr) in the absence of  Regulatory
Alternative VII.  Consequently, a delay in implementation of  any regulatory
action beyond that prescribed under Regulatory Alternative I  may effectively
increase the nationwide level of VOC emissions  from polymers  and resins
manufacturing by about 4 to 14 percent.23
     Delay of any regulatory action for VOC emissions from the polymers
and resins industry probably will have a negative impact on water quality,
as leak detection and repair of fugitive VOC emission sources can reduce
the amount of VOC contained in runoff.24  No negative solid waste impacts
are anticipated by delay of regulatory action:  as discussed in Section 7.3,
the quantity of solid waste generated increases  as a result of implementing
the regulatory alternatives.
     Table 7-7 indicates that a delay in implementing Regulatory Alternative
II may result in an energy savings of 66 TJ/yr (11,000 bbl crude oil/yr).
Delay  in implementing Regulatory Alternatives III through VII may result in
energy savings from 370 up to 940 TJ/yr (61,000 to 154,000 bbl oil/yr)
when compared to baseline control.
                                  7-35

-------
      REFERENCES FOR CHAPTER 7
 3

 4
7.

8.



9.

10



11.

12.

13.


14.
 Fugitive Emission Sources of Organic Compounds—Additional  Information
 on Emissions, Emission Reductions, and Costs.  U.S. Environmental
 Protection Agency.  Office of Air Quality Planning and Standards,
 Research Triangle Park, North Carolina.  Publication No
 EPA-450/3-82-010.  April  1982, pp. 4-1 to 4-68.   Docket Reference
 Number II-A-32.*

 Compilation of Air Pollution Emission Factors,  Second Edition.
 U.S. Environmental Protection Agency, Research  Triangle Park, North
 Carolina.  Publication No. AP-42.  August 1977.   pp. 1.4-1  and
 1.4-2.  Docket Reference  Number II-A-3.*

 Reference 2, pp. 1.3-1 and 1.3-2.

 Babcock and Wilcox,  Steam/It's Generation and Use, Thirty-eighth
 Edition.   The Babcock and Wilcox Company.  New  York, N.Y. 1975.
 p. 5-19.   Docket Reference Number II-I-20.*

 "EPA Regulations on  Preparation of Implementation  Plans."
 Environment Reporter,  125:0101-0108.   BNA,  Washington,  D.C.
 July 30,  1982.   Docket Reference Number II-I-95.*

 Development Document for  Effluent Limitation  Guidelines and New
 Source Performance Standards  for the  Synthetic Resins Segment of
 the Plastics  and Synthetic  Materials  Manufacturing  Point Source
 Category.   U.S.  Environmental  Protection  Agency.   Washington, D.C
 Publication No.  EPA-440/l-74-010a.  March 1974.  p.  204.  Docket
 Reference Number II-A-1.*

 Reference 6,  p.  147.

 "EPA  Regulations  for Identifying  Hazardous Waste."   Environment
 Reporter, 161:1864.  BNA, Washington, D.C.  September 17, 1982
 Docket Reference Number  II-I-98.*

 Reference 2,  pp.  1.3-3 and 1.3-4.

 Perry,  R.H.   Chemical  Engineering Manual, Third Edition.  McGraw-Hill
 Book  Company, Incorporated, New York, N.Y.  1976. p.. 3-18.   Docket
 Reference Number  II-I-22.*

 Reference 8,  p.  161:1853.

 Reference 10, pp. 1-3 to 1-7.

 Petroleum Facts and Figures.  American Petroleum  Institute.
Washington, D.C.  1971.  Docket Reference Number  II-I-14.*

Rossini, F.D., K.S. Pitzer, R.L. Arnett, R.M.  Braun, and G.C.  Pimental.
Selected Values of Physical  and Thermodynamic  Properties of  Hydrocarbons
and Related Componds.  American Petroleum Institute.  Washington  D C
1953.  Docket Reference Number II-I-5.*                            '  '
                                 7-36

-------
15.



16.



17.




18.



19.

20.
21.
22.
23.
24.

*
McRae, Alexander and Janice L.  Dudas.   The Energy Source Book.
Aspen Systems Corporation.   Germantown,  Maryland.  1977.   p.  441.
Docket Reference Number II-I-28.*
Shore, D. Towards Quieter Flaring.
West Drayton, Middlesex, England.
Number II-I-17.*
 Flaregas Engineering Limited.
1973.   p. 2.  Docket Reference
Straitz, J.F, III.  Solving Flare-Noise Problems.   National  AirOil
Burner Company, Incorporated.  Philadelphia,  PA.  (Paper presented
at Inter-Noise 78.  San Francisco, CA.   May 8-10,  1978.)   pp.  2-5.
Docket Reference Number II-I-39.*

Oenbring, P.R., and T.R. Sifferman.  Flare Design  ... Are Current
Methods Too Conservative?  Hydrocarbon  Processing.  59J5): 127.   May
1980.  Docket Reference Number II-I-58.*

Reference 16, p. 4.

U.S. Environmental Protection Agency.  Background  Information  for
Proposed Standards on VOC Fugitive Emissions in Synthetic Organic
Chemicals Manufacturing Industry.  EPA-450/3-80-033a.  November
1980.  p. 7-8.  Docket Reference Number II-A-16.*

Keller, M.  Comment on Control Techniques Guideline Document for
Control of Volatile Organic Compound Emissions from Manufacturing
of High-Density Polyethylene, Polypropylene, and Polystyrene Resins.
(Presented before National Air Pollution Control  Techniques Advisory
Committee.  Raleigh, North Carolina  June 1, 1981.)  Docket Reference
Number II-D-39.*

Neveril, R.B.  Capital and Operating Costs of Selected Air Pollution
Control Systems.  U.S. Environmental Portection Agency.  Research
Triangle Park, North Carolina.  Publication No. EPA-450/5-80-002.
December 1978.  p. 4-51.  Docket Reference Number  II-A-7.*

Memo from Meardon, K.R., Pacific Environmental Services,  Inc.  to
Polymers and Resins file.  Estimate of Emissions from Existing
Polymers and Resins Capacity.  September 15, 1983.   Docket Reference
Number II-B-78.*

Reference 20, p. 7-8.

References can be located in Docket Number A-82-19  at the U.S.
Environmental Protection Agency Library, Waterside  Mall,  Washington D.C,
                                  7-37

-------

-------
                               8.0  COSTS

     This chapter presents assumptions,  procedures,  and results of the
analysis to estimate the costs of controlling volatile organic compounds
(VOC) emissions from the polymers and resins industry.  The results are
estimates of capital costs, annualized costs, and incremental  costs of
emission reductions, both from the baseline and from the next  less stringent
regulatory alternative, for each regulatory alternative described in Chapter
6 and summarized in Table 8-1.  The cost impacts of  environmental regulations
other than the NSPS are also discussed.
8.1  COST ANALYSIS OF REGULATORY ALTERNATIVES
     The cost analysis consists of two steps for each control  system:
designing a system that will reliably maintain the desired efficiency
and estimating capital and operating costs for such  a system.   Designing
a control system for process VOC emissions requires  an analysis of the
waste gas characteristics of the combined stream to  each control  device
specified in a regulatory alternative.  The stream characteristics along
with mass and energy balances are the basis for determining the equipment
sizes, operating parameters, and operating requirements (e.g., fuel).
For fugitive VOC emissions, the cost analysis is based on SOCMI Model  Plant
B.L2
     Once these control system parameters have been  determined, then the
capital and annual costs can be calculated.  The capital cost  estimates
for the implementation of the regulatory alternatives include  purchase
and installation of the control or monitoring devices and piping systems
necessary for proper control of process and fugitive VOC emissions from
each model plant.
     All process VOC control capital costs are converted to June 1980
dollars using the plant cost indices published in the Chemical Engineering
Economic Indicators.  The installed capital costs for process  controls
represent the total investment, including indirect costs such  as engineering
and contractors' fees and overhead, required for purchase and  installation
of all equipment and materials for the control systems.  These are battery-
                                  5-1

-------
Table 8-1.   SUMMARY  OF REGULATORY ALTERNATIVES FOR  THE MODEL PLANTS


Model Plant/
Regulatory Process Sections Controlled1'2
Alternative RMP
Polypropylene,
liquid phase
1 (Baseline)
2
3
4 C5
Polypropylene,
gas phase
1 (Baseline)
2
Low density,
polyethylene
high pressure
1 (Baseline)
2
3
4
5
PR MR PF PS Other

C4 C5
C4 C5
C4 C5 C5
C4 C5 C5

p6 p5
p6 jr5



C6
C6
C= C"
C5 c^ C^
C6,7 C5 C5 C6
Fugitive
Emission
Regulatory
Control

No
Yes
Yes
Yes

No
Yes



No
Yes
Yes
Yes
Yes


Annual Emission
Reduction-^
Mg/yr

1,705
1,725
1,852
1,856

1,270
1,290



594
609
664
678
704
percent

89
90
97.0
97.3

95
97

.

81
83
91
93
96
Low density
 polyethylene,
 low pressure
 and high density
 polyethylene,
 gas phase
1 (Baseline)
2
3
4
5
High density
polyethylene,
slurry process
1 (Baseline)
2
3
High density
polyethylene,
solution
process
1 (Baseline)
2
3
C^>>7 c6 ^5
C6 C^ C^
C4 C6 C5
C4 C6 C5 C5
C4 C6»7 C5 C5



C6 c^
G^> C^
C$ C5 C5



C4 C6 C4
C4 C^ C4
C4 C6 C4 C5
No
Yes
Yes
Yes
Yes



No
Yes
Yes



No
Yes
Yes
1,725
1,755
1,760
1,765
1,770



900
920
950



825
845
890
94
96.1
96.4
96.7
96.9



92
94
97



90
92
97
                                       8-2

-------
Table 8-1.  SUMMARY OF REGULATORY ALTERNATIVES FOR THE MODEL PLANTS (concluded)

Model Plant/
Regulatory




Process Sections Controlled1'2
Alternative EMP
Polystyrene,
conti nuous
process
1 (Baseline)8
2
3
4
5
6
7
Expandable Poly-
styrene, post-
impregnation
suspension
process
1 (Baseline)
2
3
4
5
6
7
Expandable Poly-
styrene, in-situ
suspension
process
1 (Baseline)
2
3
4
5
6
7
Poly(ethylene)
terephthalate),
dimethyl
terephthal ate
1 (Baseline)14
2
3
4
5
6
7
Poly( ethyl ene)
terephthal ate), 22
terephthalic
acid
1 (Baseline) 23
2
Poly (ethyl ene)
terephthal ate), 25
terephthalic
acid
1 (Baseline) 26
2
3
4
5
6
7




























C6
C6



R
R
a.
R
R
R
R




R
R



R
R
R
R
R
R
R
PR MR PF PS Other


R
R
R RR9
R • RR10
R RR11
R RR12
R RR13






C5
C6 C5
c4. cf
C4 C4
R5 C4 C4





C6
C6 C6
C6 . C6 C6
C6 C6 C6
C6 C6 C4



R
R Rl<>
R Rl7
R Rl8
R R"
R R20
RR21 R20




R
RR24



R
RR27
RR27
RR27
RR27
RR27
RR27
Fugitive
Emission
Regulatory
Control


No
Yes
Yes
Yes
Yes
Yes
Yes




No
Yes
Yes
Yes
Yes
Yes
Yes



No
Yes
Yes
Yes
Yes
Yes
Yes



N.A.15
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.




N.A.
N.A.



N.A.
N.A.
N.A.
N.A.
N.A.
N.A.
N.A.


Annual Emission
Reduction3
Mg/yr


23
53
57.05
57.23
57.37
57.41
57.43




15
35
91
184
237
266
267



5
12
21
43
49
50.2
50.6



0
2.43
2.48
2.59
2.66
2.67
5.33




0
2.4



0
7.2
14.7
22.3
26.1
29.8
37.3
Percent


27
62
67.1
67.3
67.4
67.5
67.6




5
12
32
64
83
92.7
93.0



9
21
36
76
86
88
89



0
30
31
32
33
33.3
66




0
44



0
10
21
31
37
42
53
                                     8-3

-------



































•
CU
£t
ea
O

*ioL
5-
^u
^
H
^

^
in
.

u
3
•a
s_
Q.

cn
z:
j.
m
0.
8


cn
^
co

o

o
48

£
01
VI
c
cu
•a
c
o
o


01
id


ul
S.
01

c
Ol
1
o
o

•a
01
(O
Ol
cn

cu

J3
o
s_
j»
o
o
CTt
4J
O

•a
0
CL.

a>


cu
CL

o
o

01

CO

o
0

o
o

VI
s-
cu
VI

cu
•o
c
o
0

•o
CO
id
01
cn

01

£
o
t-
c=
o
e\J e
C. U

t/1
cn •
c c
4-9 *f—

3 U
VI fd
01 CU

s- c
cu o
O 4J
4-> IO
N
i. •!-
CU S-
4-> CU
•0 E


Ol O
C CL

"o o
0 i.
O M-
c in
o c
o
C '!—
E VI
3 VI •
*O E Ol
0 cu 3
o
c: c 4->
O -r-
4J c 0)
*— *r- «d
i— 4J 3
•i- 0
4J 3 01
'r- OJ -I-
"o
>,4-> O
-Q C O
2S- J=
Ol 4-1
C °" id
O O -r-
C_3 to >
-H
VI
ra

o

 c

t— -a
c
(O 

10

-j= ,
c:
XJ CU
CU XJ
s~ c: •
CU O CU
> o
O CU
U >! >
CU tt3 •(—
i_ j_ 4-j
O. ftf
01 01 C

"" •— cu
r- 0 4->
O O r—
cj >,<:
^t
cu o
cu c *J
C CU tJ

r- >> 3
>>-c? en
•P CU C£

4-> S-
* C CU
01  3

•P (tJ XJ

S- 0>4->
cu c: to
O.T- 3
0 t/»^-

01 >
Ol Ol CU
S 'at 01
O 01 (O
S— Ol 3
a. eu
> i —
.— o
03 01 £_
eoi -P
0) C
o o o
C 0 CJ
s-
<*- CUi '-

cu <=
•P j= 0
S. 4J'r-
^IS
01 l_ XJ
<<*-<
CU r- C
J= >, O
4-> JZ T-

C CO CJ
O fd
<4- CU
c o s-
E
3 C C
r— O O
O -r- 't-
O 4-> -P
U (d
C 3 N
0 XJ •!-
•i- CU I-
4J S- CU
r^ 4-» >>
r— C r—
•r- CU O
4-> CJ Q.
01 S-
-r- CU CU
XJ O..C
 S. •+•>
.c: .— a.
4-> 3 01 t-
01 O CU
01 CU E 4-»
4^ t. -P (0
cj TJ 3:

C(d CU O>
JZ -C C


i- o o
i— CU 4J O
OS CJ
S- O 01
4-* 4J C CU
C 0 JZ
0 t_ -r- 4J
O CU Ol
4-1 01 rtj



o en c
-r- C r- O
4-» -r- O •.-
>r. ,—  4J
XJ O X O
"O O *™~ CU
«c u en 01
i


cu
c



(—
UJ




01
0

CU S-


O3 Ol

^.^


r— C
3 CU
en

at 4->  -i- cu
s~ 4-1 a.
Ol Ol
01 •,-
•r- XJ XJ
c
i— * t/1 a3
o cu
CJ XJ «
>, 3 t- .
r-*r- CU S-
cn cj 3 cu
C O JZ
CU -r- 4-» 01

OJ r— S- C
r— O at *r-
^i^^1"
•p c 3: -a
a> o c
cj en ai



O -f- O 1-

4-> CU U tu
fO Ol
t. (TJ XJ fl>
at CD c ,c
a- (0 4->
o
. oi C
0> 01 S- O
01 C CU
O) O •«- 01
O -P T- CU
S_ CJ i- 01
O. CU O) C
01 4-> CU
r— 01 XJ
(O 01 CU C
EO1 O
O) O) CJ
0 0 J=
= 2^ 5>
4- a. c i.
o o a.
CU VI
4-» JZ Wl
S- 4J t= r-
aE 0
e 3 u
01 t. *0 r-
< «*- u a>
Ol "
CU -i- CM

4-> CU CU




t- 01 C

4-> . — a>
(O 3 4-*
3: vi ,—
cu 
o 3: at -r-
0 Oi 4J
§4-> ns
S- C
r- t. CU S~
^- Of XJ CU
4-> C 4->
XJ 03 3 r—
O) 3: «=C
01 »
ro cn4-> >,
cu c c s_
t- -r- Ol O
CJ r— O 4->
c o t- ns
•i- O O) r—
CJ CL 3
01 en
4-> cu tn at
o j= i— i a:
CU 4->
•4- c o at
CU 0 S. XJ

C 3
r— E en
0 3 C «
S. r— .,- 4J
•P o cn c
c: o c cu
O fd o
o c t. t.
o cu
r— -T— Ol CL

C <13 O O
O *— *r- O>
•r- r— 4-»
4^ «r- (_) 0
•r- 4-* 3 -M
XJ 01 XJ

< XJ 2! 3"
                  j= en
                 8422
                 •r- O U
                 «- 3 3
                 co -a t}
                                                  _0 o  <1J
                                                     o  >
                                                    . !-  O
                                                   V) CX U
                                                         -

                                                   S- 4->  C
                                                   OJ  S-  OJ

                                                   C  (O -r-
CO

 O)
 o
+J

 to
 (U
•M
 O
           U
           0)
           to
cu
o
      M- (O S-
         J3 O
      Cn
      C S- VI
      •r- 0) 0)
      •o-otj
      3 C T-
              ,
      V) 3 t-
      cu    aj
      O V) >
      •i-  u u
      a) «i- ai
      •o > i-
         0>
      c -a t—
      4J C O
      « 0) -r-
      3 > 4->

      £1 O T-
            "
01
>
Ol
c
cu
4->
                                                                     c
                                                                     o
                                                                    CJ
                                                                            8-4

-------
limit costs and do not include any provisions for bringing utilities,
services, or roads to the site, or for any backup facilities, land, research
and development required, or for any process piping and instrumentation
interconnections that may be required within the process generating the
waste gas.  The installation factors assumed for the various process control
devices are presented in Table 8-2.  Actual  direct and indirect cost factors
depend upon the plant specific conditions and may vary with the size of
the system.  The annualized costs consist of the direct operating and
maintenance costs, including labor, utilities, fuel, and materials for the
control system, and indirect costs for overhead, taxes, insurance,
administration, and the capital recovery charges.  The utilities considered
include natural gas and electricity.  The annualized cost factors that are
used to analyze all of the process (and fugitive) VOC control systems are
summarized in Table 8-3.
     The fugitive control capital cost includes costs for monitoring
instruments, caps for open-ended lines, piping to vent compressor seals
to an existing enclosed combustion device or vapor recovery header, rupture
disk assemblies, closed-purge sampling connections, and initial leak repair.
The derivation of annualized labor, administrative, maintenance, and
capital costs for fugitive control are presented in Section 8.1.6.
     The following sections outline the design and costing procedures
developed for flares, thermal incinerators,  catalytic incinerators,
condensers, ethylene glycol recovery systems, and fugitive control
systems.  Details of these procedures are given in Appendix E.  This
section presents an overview of the procedures and their important
features.  The results of the cost analysis  for the various regulatory
alternatives are also presented.
8.1.1  Flare Design and Cost Basis
     Elevated flares were costed based upon  state-of-the-art industrial
design.  Associated piping and ducting from  the process sources to a
header and from a header to the flare were conservatively designed for
costing purposes.  Operating costs for utilities were based on industry
practice.
     8.1.1.1  Flare Design.  Design of flare systems for the combinations
of waste streams was based on standard flare design equations for diameter
and height presented by IT Enviroscience.^  These equations were simplified
                                8-5

-------
                                 Table  8-2.   INSTALLATION COST  FACTORS
Installation
Cost Component
Major Equipment Purchase
Price (P)
Unspecified Equipment
Total Equipment (A)
Installation Factors
(Multiples of A)
Foundations
Structures
Equipment Erection
Piping
Insulation
Paint
Fire Protection
Instruments
Electrical
Sales Tax
Freight
Contractor's Fee
Engineering
Contingencies
TOTAL
	
Thermal
Flare Incinerator
1.0

0.0
TTTT


0.06
0.01
0.15
0.20
0.02
0.05
0.05
0.06
0.10
0.16
0.09
0.15
2.10 A
2.10 P
1.0

0.2
TT7


0.10
0.03
0.26 .
0.34
0.10
0.09
0.02
0.26
0.09
0.08
0.16
0.39
0.25
0.17
3.33 A .
4.00 P
Catalytic
Incinerator
1.0

0.0
OT


0.03
0.05
0.10
0.01
0.01
0.15
0.05
0.06
0.08
0,12
0.08
0.08
1.82 A
1.82 P
Condenser
1.0

0.0
ITU


-
0.02
0.10
0.04
0.01
-
0.05
0.06
0.05
0.03
0.03
1.39 A
1.39 P
EGRS .
(Baseline)
1.0

0.0
"O"

.
0.03
o ?n
U» L,\J
0.15
0.40
0.05
0.03
0 01
\J o U 1
0.15
0.10
0.08
0.16
0.34
0.27
0.27
3.24A
3.24P
EGRS
(Alternative)0
1.0

0.0
or


0.06
Oon
. o\J
0.25
0.60
0.10
0.05
Om
. U A
0.25
0 15

-------
     TABLE 8-3.  ANNUALIZEO COST FACTORS FOR POLYMERS  AND  RESINS  NSPS
                           (June 1980 Dollars)
Direct Cost Factors

Operating labor price:   $13/hr (including overhead)3

Operating labor requirements (including supervisory labor):

     =   620 labor hours/yr for flare*5
     =  1200 labor hours/yr for thermal  incinerator without  heat recovery0
     =  1500 labor hours/yr for thermal  incinerator with  heat recovery0
     =   620 labor hours/yr for catalytic incinerator without heat recovery'5
         930 labor hours/yr for catalytic incinerator with heat recoveryb
          60 labor hours/yr for condenser6
     =  4300 labor hours/yr for ethylene glycol  recovery  baseline systenr
     =  8600 labor hours/yr for ethylene glycol  recovery  regulatory alternative
          systemS

     Electricity price:   $0.049/kwhn

     Natural gas price:   $5.67/6J US.gS/MMBtu)1

     Steam price:  $13.62/Mg ($6.18/1000 lb)J

     Water price:  $0.079/m3 ($0.30/1000 gal )k

     Styrene recovery credit:   $0.788/kg ($0.357/1b)1

     Ethylene glycol recovery  credit:   $0.33/kg  ($0.15/lb),  for 80 percent  puritym

     Ethylene glycol recovery  credit:   $0.60/kg  ($0.27/lb),  for high purity11

     R-12 Refrigerant:   $2.74/kg ($1.24/lb)°

     R-502 Refrigerant:   $5.13/kg ($2.33/lb)°

     Indirect Cost Factors

     Interest rates:

       10   percent (in the absence of taxes)

Equipment life, N:P

       15 years for flare
       10 years for thermal incinerator, catalytic incinerator, condenser,
          ethylene glycol recovery system, piping, initial labor for fugitive
          emissions control, other fugitive emission  equipment except
          rupture disks and monitoring instruments
        6 years for monitoring instruments
        2 years for rupture disks
                                   8-7

-------
     TABLE 8-3.  ANNUALIZED COST FACTORS FOR POLYMERS AND RES I MS  NSPS
                    (June 1980 Dollars)  (Continued)
Indirect Cost Factors (Con't)
                                  4/1   +  4 \
Capital recovery charge factor =   x        '
                                 ~~"
     0.131 for flare
     0.163 for thermal incinerator, catalytic incinerator,  condenser,
     ethyl ene glycol recovery system, piping

Taxes, insurance, and administration:  0.04 x total  installed capital
cost0!
                                                      r
Maintenance cost:  0.5 x  Total  installed capital  cost
                   $3,000 for monitoring instrument,  calibration,  and
                   maintenance

Operating hours:  8600 hours/yr
h
 Includes  wages  plus  40 percent  for labor-related adminstrative and
 overhead  costs.

 0.5  man-hours/shift  x 3600 hrs/yr * 8 hrs/shift + 15 percent of the
 operating labor for  supervisory costs.

'Blackburn, J.W.  Control  Device  Evaluation:  Thermal Oxidation, Report
 No.  1  in  Organic Chemical Manufacturing, Volume 4.  U.S. LEnvironmental
 Protection Agency.   Research Triangle Park, N.C. Publication No.
 EPA-450/3-80-026.  December 1980.

 0.75 man-hours/shift x 8600 hrs/yr * 8  hrs/shift  +15 percent of the
 operating labor for  supervisory costs.
3
"1 man-hour/week x 8600 hrs/yr r 8 hrs/shift * 21 shifts/week +15 percent
 of operating  labor for supervisory costs.

 4 man-hours/shift x  8600 hrs/yr -f 8 hrs/shift.

 Number of man-hours/year provided by an industry source.

 Memo from Chasko and Porter, EPA, September 17, 1980.  Guidance for
 developing GTGD Cost Chapters.

 Memo from Al  Wehe, to  Information Analysis Working Group for the Industrial
 Boiler Working  Group.  April 23, 1981.   IFCAM Modification:

      Projected  1985  price in 1978 dollars is $4.91 + $.60 delivery charge
      per  MMBtu.

      Projected  1990  price in 1978 dollars is $5.55 + $0.61 delivery charge
      per  MMBtu.

                                     8-8

-------
Footnotes to Table 3-3 Concluded

     By linear interpolation between $4.91  and $5.55/MMBtu;  1938 price in
     1978 dollars = $5.29/MMBtu.

     Using GNP implicit price del".-tor index;   -Hii quarter 1978 of 154.99
     and 2nd quarter 1980 of 175.28; 1988 price in 1980 dollars =
     175.28/154.99 x 5.29 = $5.98/MMBtu.

     Assumed higher heating value of 1040 Btu/scf at 16°C(60°F).

 Neverill, R.B.  Capital  and Operating Costs of Selected Air Pollution
 Control Systems.  tj.S. Environmental  Protection Agency.  Research Triangle
 Park, N.C. Publication No, EPA-450/5-80-002.   December 1978.   pp. 3-12:

     $5.04/1000 Ib steam, 4th quarter 1977.

     Using GNP implicit price deflator index:   4th quarter 19/7 of 142,91
     and 2nd quarter 1980 of 175.28; updated steam price = 175.28/142.91
     x $5.04 = $6.18/1000 Ib steam.
k
 Peters, M.S. and K.D. Timmerhaus.   Plant Design and Economics for Chemical
 Engineers.  McGraw-Hill  Book Co.  New York,TO".'  ThircTTdltion.  19807
 p. 881~
1
m
90 percent of styrene price given in Chemical  Marketing Reporter.

Assumed based on polymer raw material  (high purity)  ethyl ene glycol
price of      '
 Polymer raw material  grade of $0.27/lb of ethyl ene glycol  given in
 Chemical Marketing Reporter.

 Telecon:  K. Meardon, PES to  ARC Supply Co.  May 1984 prices of $1.31/lb
 for Freon-12 and $3. 11/1 Ib for Freon-502 deflated to June 1980 dolalrs
 using Chemical  Engineering Producer Price Industrial  Chemicals Index
 (June 1980 = 3T27.T; May 1984  = 344.8).

 Average equipment lives given by Neverill in reference cited in j.,
 pp. 3-16.

 Fugitive Emission Sources of  Organic Compounds -- Additional  Information
 on Emissions, Emission Reductions, and Costs.   U.S. Environmental
 Protection Agency.  Research  Triangle Park,  N.C. Publication No. EPA-
 450/3-82-010.  April  1982. pp. 5-16.
r
 Per reference cited in footnote q:

     9 percent of tot-.-tl installed capital  costs for maintenance and
     miscellaneous charges - 4 percent of total installed capital costs
     for taxes,  insurance and  administration  (equivalent to miscellaneous).
                                      8-9

-------
to functions of the following waste gas characteristics:   volumetric
flow rate, lower heating value, temperature, and molecular weight for a
state-of-the-art exit velocity of one-half sonic velocity (0.5 Mach).
The diameter equation is based on the equation of flow rate with velocity
times cross-sectional area.  A minimum commercially available diameter
of 2 inches was assumed.  The height correlation premise is design of a
flare that will not generate a nonlethal radiative heat level (1500 Btu/ft2
hr, including solar radiation4) at the base of the flare (considering
the effect of wind).  Heights in 5-foot multiples with a minimum of 30 ft.
were used.5
     Supplemental fuel, natural gas, is added to increase the heating
value to 115 Btu/scf to ensure combustion.6  For flares with diameters
of 24-inches or less, this natural gas was assumed to be premixed with
the waste gas and to exit out the stack.  For larger flares, a gas ring
at the flare tip was assumed because such separate piping is more economical
than increasing the flare stack size for large diameter.
     Purge gas also may be required to prevent air intrusion and flashback.
A purge velocity requirement of 1 fps was assumed during periods of
continuous flow for standard systems without seals.7  For flares handling
only intermittent flows, purge gas requirements were assumed to be
negligible according to the industry practice of not purging or perhaps
purging before a planned intermittent release.^  For combined streams
with very large turndown ratios (intermittent flow :  continuous flow),
supplying purge gas to maintain an adequate continuous flow in a large
flare (designed for the intermittent flow) can become more expensive
than designing a second separate flare for the continuous flare.  For flares
handling very small flows in a minimum-available-size flare, the cost of
supplying sufficient purge gas can be greater than the cost of a fluidic seal
on the flare tip to prevent air intrusion and subsequent  flashback.
In such cases, a fluidic seal, which requires a greatly reduced purge
rate, was used.
     Natural gas consumption at a rate of 80 scfh per pilot flame to
ensure ignition and combustion was assumed.  The number of pilots was
based on diameter according to available commercial equipment.9
     Steam was added to produce smokeless combustion through a combined
mixing and quenching effect.  A steam ring at the flare tip was used to
                               8-10

-------
 add steam at a  rate of 0.4 lb steam/lb of hydrocarbons (VOC plus methane
 and ethane) in  the continuous stream  (or the intermittent stream if no
 continuous flow was present).10  Availability and deliverability of this
 quantity of steam was assumed.  For flares handling large intermittent
 flows, steam requirements were calculated for a small continuous "bleed"
 rate that is necessary to prevent shortened steam nozzle equipment life
 (due to steam condensate contacting the high temperature metal).11*12
     Piping (for flows less than 700 scfm) or ducting (for flows equal
 to or greater than 700 scfm) was designed from the process sources to a
 header combining the streams (via "source legs") and from the header to
 the base of the flare (via "pipelines").  Since it is usual industry
 practice, adequate pressure (approximately 3 to 4 psig) was assumed
 available to transport all waste gas streams without use of a compressor
 or fan.  The source legs were assumed to be 70 feet in length,13 while
 the length of pipelines to the flare was based on the horizontal distance
 required to provide the safe radiation level for continuous working
 (440 Btu/hr-ft2, including solar radiation9).  For flows less than
 700 scfm, an economic pipe diameter was calculated based on an equation
 in the Chemical Engineer's Handbook14 and simplified as suggested by
 Chontos.15»16»17  The next larger size (inner diameter) of schedule
 40 pipe was selected unless the calculated size was within 10 percent of
 the size interval  between the next smaller and next larger standard
 sizes.  For flows  of 700 scfm and greater, duct sizes were calculated
 assuming a velocity of 2,000 fpm for flows of 60,000 acfm or less and
 5,000 fpm for flows greater than 60,000 acfm.  Duct sizes that were
 multiples of 3-inches were used.  (See Section E.7 for detailed design
 and cost procedures for piping and ducting.)
     8.1.2.1  Flare Costing.   Flare  purchase costs were based on costs
 for diameters  from 2 to 24 inches and heights from 20 to 200 feet
 provided by National  Air Oil  Burner, Inc., (NAO) during November 1982.9
A cost was also provided for  one additional  case of 60 inch diameter and
 40 feet height.10   These costs are October 1982 prices of self-supporting
 flares without ladders and platforms for heights of 40 feet and less and
 of guyed flares with ladders  and platforms for heights of 50 feet and
 greater.  Flare purchase costs were  estimated for the various regulatory
 alternatives by either choosing the  value  provided for the  required
                              8-11

-------
height and diameter or using two correlations developed from the  NAO
data for purchase cost as a function of height and diameter.  (One
correlation for heights of 40 feet and less and one for heights of
50 feet and greater.)  An installation factor of 2.1 (see  Table 8-2)  was
used to estimate installed flare costs.
     Piping costs were based on those given in the Richardson Engineering
Services Rapid Construction Estimating Cost System^ as combined  for
70 ft. source legs and 500 ft. and 2,000 ft. pipelines  for the cost
analysis of the Distillation NSPS.^9  Ducting costs were calculated
based on the installed cost equations given in the CARD Manual.20
Installed costs were put on a June 1980 basis using the following Chemical
Engineering Plant Cost Indices:  the overall index for  flares; the
pipes, valves, and fittings index for piping; and the fabricated  equipment
index for ducting.  Annualized costs were calculated using the factors
presented in Table 8-3.
8.1.2  Thermal Incinerator Design and Cost Basis
     For costing purposes thermal incinerator designs were based  on heat
and mass balances for combustion of the waste gas and any  required
auxiliary fuel, considering requirements of total combustion air.
Associated piping, ducting, fans, and stacks were also  costed.
     8.1.2.1  Thermal Incineration Design.  Designs of  thermal incineration
systems for the various combinations of waste gas streams  were developed
using a procedure based on heat and mass balances and the  characteristics
of the waste gas in conjunction with some engineering design assumptions.
In order to ensure a 98 percent VOC destruction efficiency, thermal
incinerators were designed to maintain a 0.75 second residence time at
870°C (1600 °F).21  The design procedure is outlined in this section.
     In order to prevent an explosion hazard and satisfy insurance
requirements, dilution air would be added to any individual or combined
waste stream with both a lower heating value between 13 and 50 Btu/scf  at
0°C  (32°F)  (about 25 and 100 percent of the lower explosive limit)22'23
and an oxygen concentration of 12 percent or greater by volume™  that
required preheating to maintain a combustion temperature of 870°C (1600°F).
Dilution air would be added to reduce the lower heating value of  the  stream
to below 13 Btu/scf.  (Adding dilution air is a more conservative assumption
                               8-12

-------
than the alternative of adding natural  gas and is probably more realistic
as other streams often have enough heat content to sustain the combustion
of the combined stream for the regulatory alternative.)  The only stream
(Stream G in the polypropylene, liquid phase plant) for which a thermal
incinerator was costed in this heat content range, however, did not require
preheating, because^the combined streams for the section had a heat content
greater than 50 Btu/scf.
     The combustion products were then calculated assuming 18 percent
excess air for required additional combustion air,25,26 ^u^ Q percent
excess air for oxygen in the waste gas, i.e., oxygen thoroughly mixed with
VOC in waste gas.  The procedure includes a calculation of auxiliary fuel
requirements for streams (usually with heating values less than 60 Btu/scf)
unable to achieve stable combustion at 870 °C (1600 °F) or greater.  Natural
gas was assumed as the auxiliary fuel as it was noted by vendors as the
primary fuel now being used by industry.  Natural gas requirements were
calculated using a heat and mass balance assuming a 10 percent heat loss
in the incinerator.27,28  Minimum auxiliary fuel requirements for low
heating value streams were set at 5 Btu/scf to ensure flame stability.^9
     For streams able to maintain combustion at 870°C (1600°F) such as the
combined streams costed for the Polymers and Resins industry, fuel was
added for flame stability in amounts that provided as much as 13 percent
of the lower heating value of the waste gas for streams with heating
values of 650 Btu/scf or less.  For streams containing more than 650 Btu/scf,
flame stability fuel requirements were assumed to be zero since coke
oven gas, which sustains a stable flame, contains only about 590 Btu/scf.
In order to prevent damage to incinerator construction materials, quench
air was added to reduce the combustion temperature to below the incinerator
design temperature of 980°C (1800°F) for the cost curve given by IT
Enviroscience.3^
     The total flue gas was then calculated by summing the products of
combustion of the waste gas and natural gas along with the dilution air.
The required combustion chamber volume was then calculated for a residence
time of 0.75 sec, conservatively oversizing by 5 percent according to
standard industry practice.3^  The design procedure assumed a minimum
commercially available size of 1.01 m3  (35.7 ft3) based on vendor
information32 and a maximum shop-assembled unit size of 205 m3  (7,238 ft3).33
                               8-13

-------
      The design procedure could allow for pretreating of combustion air,
 natural gas, and when permitted by insurance guidelines, waste gas using
 a recuperative heat exchanger in order to reduce the natural gas required
 to maintain a 870°C (1600°F) combustion temperature.  However, all streams
 to thermal incinerators for polymers and resins regulatory alternatives
 had sufficient waste gas heating values to combust at greater than 870°C
 (1600 °F), and even at greater than 980°C (1800 °F), without preheating
 the input streams.  If a plant had a use for it, heat could be recovered.
 (In fact, a waste heat boiler can be used to generate steam, generally
 with a net cost savings.)
      8.1.2.2  Thermal  Incinerator Costing.  Thermal incinerator purchase
 costs were taken directly from the IT Enviroscience graph for the calculated
 combustion chamber volume.30 (Essentially equivalent purchase costs
 would be obtained by using data from the GARD manual.20)  An installation
 cost factor of 4.0 was used  based  on the Enviroscience  document (see  Table
 8-2).34 The installed  cost used for the  minimum commercially available
 size of 1.01m3 (35.7 ft3)  was  $217,800 (June  1980)  ($52,000 December
 1979 purchase  cost).
      The installed  cost  of one  150-ft. duct to  the  incinerator  and  its
 associated  fan  and  an  80-ft. stack  were  also  taken  directly from  the  IT
 Enviroscience  study.35  A  minimum  installed cost of $70,000 (December  1979
 dollars) was assumed for waste  gas  streams with flows below 500 scfm.  The
 costs of piping  or  ducting from the process sources  to the  150-ft.  duct
 noted above were estimated as for  flares.
      Installed costs were put on a  June  1980  basis  using the following
 Chemical Engineering Plant Cost Indices:  the overall index for thermal
 incinerators; the pipes, valves, and fittings index  for piping; and the
 fabricated equipment index for ducts, fans, and stacks.  Annualized costs
 were calculated using the factors in Table 8-3.  The electricity required
 was calculated assuming a 1.5 kPa (6-inch H20) pressure drop across the
 system and a blower efficiency of 60 percent.
 8.1.3  Catalytic Incinerator Design and Cost Basis
     Catalytic incinerators are generally cost effective VOC control
devices for low concentration streams.  The catalyst increases the
chemical rate of oxidation allowing the reaction to proceed at a lower
energy level (temperature) and  thus requiring  a smaller  oxidation  chamber,
less expensive  materials, and much  less auxiliary fuel  (especially for
                                8-14 '

-------
low concentration streams) than required by a thermal  incinerator.   The
primary determinant of catalytic incinerator capital  cost is volumetric
flow rate.  Annual operating costs are dependent on emission rates,
molecular weights, VOC concentration, and temperature.  Catalytic
incineration in conjunction with a recuperative heat exchanger can
reduce overall fuel requirements.
     8.1.3.1  Catalytic Incinerator Design.  The basic equipment'components
of a catalytic incinerator include a blower, burner, mixing chamber,
catalyst bed, an optional heat exchanger, stack, controls, instrumentation,
and control panels.  The  burner is used to preheat the gas to catalyst
temperature.  There is essentially no fume retention  requirement.  The
preheat temperature is determined by the VOC content  of the combined
waste  gas  and combustion  air, the VOC destruction efficiency, and  the
type and  amount  of catalyst required.  A sufficient amount  of air  must
be  available  in  the gas  or  be  supplied to  the  preheater  for VOC  combustion.
 (All the  gas  streams  for which  catalytic incinerator  control  system
costs  were developed  are dilute enough  in  air  and therefore require no
additional  combustion  air.)  The  VOC  components  contained  in  the gas
streams  include  ethylene, n-hexane,  and  other  easily  oxidizable  components.
These  VOC components  have catalytic  ignition temperatures  below  315  °C
 (600  °F).  The catalyst  bed outlet temperature is  determined  by  gas VOC
 content.   Catalysts  can  be operated  up  to  a temperature  of 700  °C  (1,300 °F)
 However,  continuous  use  of the catalyst at this high  temperature may
 cause accelerated thermal aging due  to  recrystallization.
      The catalyst bed size required  depends upon the type of  catalyst used
 and the VOC destruction  efficiency  desired.  Heat  exchanger requirements
 are determined by gas inlet temperature and preheater temperature.  A
 minimum practical heat exchanger efficiency is about 30 percent; a maximum
 of 65 percent was assumed for this analysis.  Gas temperature,  preheater
 temperature, gas dew. point temperature, and gas VOC content determine the
 maximum feasible heat exchanger efficiency.  A stack is used to vent the
 flue"gas  to the atmosphere.
       Fuel  gas requirements were calculated  based on  the heat required
 for a preheat temperature  of  315 °C  (600  °F), plus 10 percent for auxiliary
 fuel. The fuel was  assumed to  be natural  gas,  although oil  (No.  1 or 2)
 can be used.  Electricity  demand was based  on pressure  drops of 4 inches
                                 8-15

-------
 water for systems without heat recovery and 10 inches water for systems
 with heat recovery, a conversion rate of 0.0001575 hp/in.  water,  65 percent
 motor efficiency, and 10 percent additional  electricity required  for
 instrumentation, controls, and miscellaneous.   A catalyst  requirement of
 2.25 ft3/!,000 scfm was assumed for 98 percent efficiency.36   Catalyst
 replacement every three years was assumed.
      8.1.3.2  Catalytic Incinerator Costing.   Calculations for capital
 cost estimates were based on equipment purchase costs obtained from
 vendors for all  basic components and the application  of direct and
 indirect cost factors.56,37,38  purchase cost  equations were  developed
 based on vendor third quarter 1982  purchase  costs of  catalyst incinerator
 systems with and without heat exchangers for sizes  from 1,000 scfm  to
 50,000 scfm.  The cost data  are based on carbon steel  material  for
 incinerator systems and stainless steel  for heat exchangers.   Catalytic
 incinerator systems of  gas volumes  higher than  50,000 scfm can be
 estimated by considering two equal  volume units  in  the  system.  A minimum
 available unit size of  500 scfm was assumed39.40; the installed cost of  this
 minimum size unit,  which can be used without addition of gas  or air for
 stream flows greater than about 150 scfm40, was  estimated  to  be $53,000
 (June 1980).   Heat  exchangers  for small  size systems  are costly and may
 not  be practical.   The  direct  and indirect cost  component  factors used for
 estimating capital  costs of  catalytic  incinerator systems with no heat
 exchangers and for  heat exchangers  were  presented in Table 8-2.  Installed
 costs of  piping,  ducts,  fans,  and stacks were estimated by the same procedure
 as for thermal incinerators.   Installed  costs were put on a June 1980
 basis using  the  following Chemical  Engineering Plant Cost indicies:   the
 overall index for catalytic  incinerators; the pipes, valves, and fittings
 index for piping; and the fabricated equipment index for ducts, fans, and
 stacks.  Annualized costs were calculated using the factors in Table 8-3.
 8.1.4  Co_ndenser^Design  and Cost Basis
     This section outlines the general procedures used for  sizing  and
 estimating the costs of surface condenser systems applied to the gaseous
 streams from the continuous process  polystyrene model  plant and the
 poly(ethylene terephthalate), dimethyl terephthate process  model plant.
For illustrative purposes, the following paragraphs describe the condenser
design and costs for the polystyrene model plant.  Details  of  the condenser
                              8-16

-------
design and costs for both model  plants are found in  Docket Reference
Number II-B-93.
     8.1.4.1  Surface Condenser Design.   The condenser system evaluated
consists of a shell  and tube heat-exchanger with the hot fluid in  the
shell side and the cold fluid in the tube side.   The condenser system  is
sized based on the total heat load and the overall, heat transfer coefficient
which is established from individual heat transfer coefficients of the gas
stream and the coolant.
     Although some older polystyrene processes emit styrene in steam,  which
is more readi-ly condensed and thus less costly to control, saturated styrene
in air at 27°C (80°F) was assumed for costing design purposes since new plants
under baseline conditions were assumed to use vacuum pumps.  Total heat
load was calculated using the following procedure:  the system condensation
temperature was determined from the total pressure of the gas and  vapor
pressure data for styrene in air.  As the vapor pressure data are  not
readily available, the condensation temperature was estimated by a regression
equation of available  data points41 using the Clausius Clapeyron equation
which relates the stream pressures  to the temperatures.  The total pressure
of the stream is equal  to the vapor pressures of individual components at
the  condensation temperature.  Once the condensation  temperature was known,
the  total heat load  of the condenser was  determined from the latent heat
content of  the condensed styrene and  the  sensible heat  changes  of styrene
and  air.  The coolant  is selected based on  the  condensation temperature.
      The  styrene-in-air refrigerated  condenser  systems  were  designed
according  to  procedures  for  calculating shell-side42, tube-side43, and
condensation44  heat  transfer coefficients,  mass transfer coefficients,45
and, finally, an  overall  heat  transfer coefficient  for  condensation in the
presence  of a non-condensible  using a multiple  section  analysis.46  Heat
 exchanger47'48  and  refrigeration  unit49 characteristics were  developed from
 vendor  information  in  conjunction with  information  given  primarily in the
 Chemical  Engineers'  Handbook.42'50  Refrigerant characteristics were
 taken primarily from the Chemical Engineers'  Handbook51*52 and
 publications of the American Society  of Heating, Refrigerating, and Air-
 conditioning Engineers (ASHRAE).53   The total  required heat transfer
 area and refrigeration capacity then  were calculated from the total
 heat load, temperature difference,  and overall  heat transfer coefficient
 and commercially available sizes were selected.
                                     8-17

-------
     8.1.4.2  Surface Condenser Costing.  Since the yas volumes of the
two streams are low, the calculated required (selected) heat transfer areas
are also low - about 1.7 (2.3) and 4.5 (7.2), respectively, for 90 and 98
percent reduction of styrene emissions from a single process line).   The
purchase costs of the heat exchanger47'48 and refrigeration systems  were
estimated from data provided by vendors.49  An installation factor of
1.39 (see Table 8-2) was used to estimate installed condenser costs.
The corresponding required refrigeration capacities for 90 and 98 percent
styrene reduction were only 0.056 and 0.080 tons (0.20 and 0.28 kW or
670 and 960 Btu/hr), respectively.  For 90 percent reduction, the required
capacity is much smaller than the 0.117 tons (0.412 kW or 1405 Btu/hr)
available capacity for the minimum available size refrigeration unit of
1/4 compressor horsepower (0.186 kW compressor) and a coolant temperature
of 0°F (-17.8°C).  For 98 percent reduction, however, 1/2 compressor
horsepower (0.373 kW compressor) was required to provide sufficient
available refrigeration capacity for a coolant temperature of -30°F
(-34.4°C).  Installed costs were put on a June 1980 basis using Chemical
Engineering Cost Indexes.  No additional piping was costed since the
condenser unit is so small (£5 in. diameter) that it should be able
to be installed adjacent to the source.  Details of the design and cost
estimates are given in Appendix E (Section E.5).
8.1.5  Ethylene Glycol Recovery System Design and Cost Basis
     This section briefly describes the procedures used to estimate  the
cost of the baseline systems and of using a distillation column to recover
ethylene glycol from a cooling water tower.
     8.1.5.1  Ethylene Glycol Recovery System Design.  The baseline  system
for a PET plant recovers ethylene glycol through spent ethylene glycol
spray condensers, reflux condensers, and distillation columns.   The  equipment
selected to comprise baseline recovery systems was obtained from information
provided by industry sources, which also provided basic operating and
maintenance parameters.  The information for the EG spray condenser  system
was provided for a PET/TPA plant of larger capacity than the model plant.
The basic equipment was considered to be the same regardless of process
(DMT or TPA).  (The flow from the TPA process can be as much as 35 percent
less than that from a DMT process; however,  differences depend  on the
completeness of the polymerization reaction  and the part of the process
where VOC is emitted.  Both of these factors are plant-specific.)
                                    8-18

-------
     Industry sources also provided the basic equipment requirements  and
design considerations for installing a distillation column on a cooling
water tower to recover ethylene glycol from the cooling water.  The
information provided was for a PET/TPA plant of larger capacity than  the
model plant.
     8.1.5.2 EG Recovery System Costing.  The costs of the baseline
system were estimated based on the preliminary design using industry
information in conjunction with standard engineering procedures.  The
costs of the ethylene glycol  spray condenser and recovery system for  new
plants using the DMT and the TPA processes were derived from a base set
of costs provided by an industry source for a similar system on a larger
capacity plant.  The costs of the distillation column were derived from
information provided by an industry source for a similar system on a  larger
capacity plant.  Details of the cost estimates are given in Appendix  E
(Section E.6).  Installation cost factors and annualized cost factors
are  given in Tables 8-2 and 8-3, respectively.
8.1.6  Fugitive Emission Control Program Design and Cost Basis
     As noted in Section 6.2.4, a leak detection and repair program and
equipment specifications was specified as-the regulatory alternative
beyond baseline control.  This regulatory alternative was selected so
that fugitive emission control in the polymers and resins industry would  .
be consistent with the SOCMI fugitive VOC emission regulations.  The
following sections outline the specific requirements of the regulatory
alternative and the procedures used to estimate its capital and annual
costs.
     8.1.6.1  Design  of  Fugitive VOC  Regulatory Alternative.   The  following
equipment in the  polymers and  resins  industry were considered  for  regulation:
process valves, pumps, compressors, safety  relief  valves, flanges, sampling
connections, and  open-ended lines.  Those VOC emissions resulting  from  the
transfer,  storage,  treatment,  and  disposal  of process  wastes  are not covered
by  this regulation.
      The  selected leak  detection  and  repair program  and equipment
 specifications  for  the  regulatory  alternative are  listed  in  Table  8-4.
 The control  specifications were  described  in Chapter  6.
      The  technical  parameters  for  the polymers  and resins model  plant
 for fugitive emissions  are given  in Table  8-5.

                                     8-19

-------
        Table 8-4.  FUGITIVE VOC REGULATORY ALTERNATIVE CONTROL
                            SPECIFICATIONS9


                                  	Control Specification
Emission source
Inspection/
monitoring
 interval
Equipment
Valves

  Gas
  Light liquid
  Heavy liquid

Pumps seals

  Liqht liquid

  Heavy liquid

Safety/relief valves

  Gas
Open-ended, lines (purge,
drain, sample lines)

Compressors
Sampling connections
Flanges
 Monthly
 Monthly
 None
 Month!yc
 Weekly Visual
 None
 None



 None

 None



 None


 None
None
None
None
None

None
Rupture disks on
relief valves
Cap

Controlled
degassing
vents

Closed-purge
sampling

None
aThe regulatory alternative is a combination of Regulatory Alternatives V
 and III for VOC Fugitive Emissions in Petroleum Refining Industry -
 Background Information.
 Sources found to  be  leaking by monitoring would be repaired.
°For pumps, monthly instrument monitoring would be supplemented with
 weekly visual inspections for liquid leakage.  If liquid is noted to
 be leaking from the pump seal, the pump seal would be repaired.
                                   8-20

-------
3 i—
-o o
C !_ C.
t— 0 V) 4-> >,
10 T- C C 	
3 vi o o cn
c: (/>••- us
C •»-•»->
•a: E u o
-a  . CO
03 «-
OO
H-
eC T3
	 1 -o •
— ' CU V)
ci- _;:: c t.
10 i— O >,
	 1 3 0 ••- ~-
TT i c i- vi en
UJ c jj vi s
O 


h—H
OO
UJ
o
C£

O <" "
in t— at
OO ,— ,— c
10 o O i-
li 1 3 S- — >,
!r= c 4-> in ^
•J- c c vi en
 C ^^ 3
00 S 'vi "* "
00 ^ vi

|_. .; £
S: cu
LU



O
• "
M-
li 1 O O

> L. CO
1 — 1 CU u


>— J 30
CD -z.fi
— \

Li_




UD
|
oo
CU

0
ra

O CO
•i- U
VI C-
V) 3
•i- 0
S v)
LU



0 CO
"3- CM

>3- cn o
t-H i— (









CM VO ^O
en en ci
LO CO <~4
"


CO CTi
r^ LO I
• • 1
o o






•o
-^ C
CO
•— c
o
1— 4J t-
-C U ••—

c 4-> a.
O CU CO
s -o t-







C\l Ol ^O
r-. ITS o

o% oo *— c
f-H CO









> (O -F- QJ
,— CJ -J 31
re



CO


^^ ^3










OJ CM
a\ ^o
«3- m



_^
<>D 1
• 1
0






T3
-^ C
CU
r— C
O
>v,-u-
P— -W t-
J= O «r-
J_J QJ (Q
C 4J Q.
o ai aj
2: -o t.







LO CM
LO »JO

CM LO
f— I







S2
*a- CM
0 CD
C3 0













o\ o
oj en














."2 2
3 3
o- o*
(/) -r- -r-
re
CU J-> >,
W JZ >
OT (O
CL'f- QJ
• E -_i or
3
0-



OJ &\ CM O^
O 
O >, t-
*f— (J ^
*-J C VI
V) «f- O)
a. -o w -D
.^ re aj *o < T-
V) U ^- Ol < 3
•r- ro L. ^- 1C O-
"O "O <1) 3 O I O -i—
c tf) Q. K &. tn -f- »—
ai to t/) -M ~o -M
S- *O 13 E C CU >,
3cn
+J Ol 4-> CO 4-) U O 4-) 03
Q. 3 C O to Q.O , O St. JT
ce a. 5> o en 2r o a.
^— -^
U ra
c c c
fO O flj
O> E E
t- C Q.
O O f-
CM Cy»CMLOOJ«t_ 3
O CJ*'s3-^-«3'Qi^.f- o*
• ••»*i-^o> 4)
OOfOcnr^^o CD c
•-* f— • O CO UJ • t.
*-« )— r OJ r- O
.E O • QJ «4—
O 3 • O -O
en o rs -— * f— o
to » jz
 o • L. • e
^- tO OJ 4^ ^~
t#- trt 4-> U C 1 i—
en OirtcnCO COO
Or**. oco -^OOJU u
^-*-4COLOO CQEOU 0)4->
OOCMi-HO CDOJOJ ^C
»-HOOJOO r— "O <*- C JOO
OOOOO 'O>(tJCC*— ^
O ••-'—' OJ O
S •(-> CO CO (U C
••- C (O 0)
co CTO*J3 • C/) .*
O3 3 *r- CO L. 0)
U_ 4-> 1 t- fO U
(usyeo j O) >

to +JOJC 3 O>-^ Q)
t- o; TO c 4-j to
aj o *a -^- (o
JZ Q.CCM 0) 4J C "O
4->  LU O CD >* 4_)
E o c C jz
"O O * CM COO CJ^
eu t_ to co 3 4_> .1—
E **- c o\ c re r—
3 O «-H 4J -r- r—
10 -O .,_ U_ 3 C
CO CD tO i— CJ CO C7) -^-
co co re c to -^ ^ t_ a)
 o cotoEQ-coo c
r— 'r- •r—4->LU'=C(UJZ j2 o jz E
> CO.— U 4JOC t-O-UO.
creaJ .-^- o«3vo T-
**- -^OJC ' COJOCr-^>,3
OJ f— CO C 3 S- C <— 'CO»J2CT
•i— o re o o co  re co o «*— o
^N. CtOCCO O (O £ t L. i*_U-
>j OJOJ-r-OJ E'Ot-tO.D-OT-
+J CO 1 t- r— O> ^— O~^.jQCDU>^
eyre c CLQ.C re ce eu^-O E to aJr—
u-o a; E E re-*-> OOJZ^LO 3 re a.c
GO Oe_3tOLZh-rejZ! U~OCiJ4-
8-21

-------
     8.1.6.2  Fugitive VOC Emission Control  Costs.  This section presents
the cost estimates and the input parameters  affecting the cost estimates
for each of the fugitive emission sources within the polymers and resins
industry.  All the cost information presented here is obtained from the
SOCMI NSPS analysis. 1»2  All costs are updated to represent second
quarter 1980 dollars.  Table 8-6 summarizes  the total cost associated
with the fugitive VOC regulatory alternative and Table 8-7 summarizes
the costs to control the individual equipment.
     The following describes the cost estimation procedures and
assumptions used to derive these costs for each fugitive VOC emission
source.
Valves
     The fugitive emission control specifications for valves include
monthly monitoring and leak repairs.  Therefore, the annual costs
associated with controls include initial leak repair costs and recurring
monthly monitoring and leak repair costs.
     The costs for leak detection and repair programs for valves are
based on the following factors:  (1) monitoring time, (2) repair time
for on-line and off-line repair, and (3) fractions of leaks repaired
on-line and off-line.  The following estimates were used for the above:
     (1)  Monitoring time:  The monitoring time estimate of 2 man-minutes
per valve was used.
     (2)  Repair time:  The repair time estimate of 10 man-minutes per
valve for on-line repair of valves and 4 man-hours per valve for off-
line repair of valves was used.
     (3)  Fraction of leaks repaired on-line and off-line:  Seventy-five
percent of all valves were estimated to be repaired on-line, while
25 percent were estimated to be repaired off-line.
     Based on the results of the leak detection and repair (LDAR) model,54
the annual costs of monitoring and repairing valves have been estimated
for the monthly monitoring program.  The cost calculations are presented
in Tables 8-8 through 8-11.  The input parameters, e.g., occurrence
rate, initial leak frequency, etc., are discussed in the Emission
Reduction section (Section 4) of the SOCMI AID.55
                                  8-22

-------
       Table 8-6.  FUGITIVE VOC REGULATORY ALTERNATIVE COSTS
                FOR POLYMERS AND RESINS MODEL UNITS
        Cost Item
1. Installed capital cost
2. Annual cost .of operation
   a)  Operating labor including
       administration and support
   b)  Maintenance
   c)  Miscellaneous3
   d)  Annualized capital costs
   Total
3. VOC recovery credits^
4. Net annualized cost
May 1980 dollars
     66,800

     20,800
      5,700
      2,600
     14,100
     43,200
    (31,000)C
     12,200
  Taxes, insurance, and administration.
  Based on 58.69 Mg of VOC recovered annually at a credit of $528 per Mg.
  Values in parentheses denote credits.
                              8-23

-------
















CO

CO
f— 1
OJ-
t-H
_1 Z
O Z3
O£.

"3Z, LU
O O
O CD

•y. ^^
O CO
•— < Z
CO »-«
CO CO
I— 1 LU-— >
s: o: t/>
LU S-
f"*\ (O
t_i sc f—
t_»*  O
CO "O
LU CC
> LU O
H- >• cn
CD O "~
Z3 0_ >>
u_ to
LU SS
O
CO
>~ LU
Oi 0
1 =
2 o
ra co
CO
n:
• l*~»
r*"11*
00 CD
U_
0)
r™"
jQ
(0
i
r^







































l/>
CO
O

3
O
10

c
o
•I—
in
1

o
CD
5*


































"io
•P
o
cn-P
C E
*r~ CJ
S_ E
JK! o o.
re 4J •!— •
 -P^ >
re O i~" i—
CD 4- O) 
in


•P "O in
£= »i- Q-r—
CO 3 E «3
i- cr 3  CTi—
W -r- •!- «
(O r— r- >
C3










p
O
+J
•r-
•P
in
O
O









o
o
00
to
to


o
o
CM
A
CT>




O
O
CO
A
CO
f—t





o
o


to




o
o
r-.
A
CO
CO



o
o
<-4


o
o
CO
A
CM








W
o
o

^_
03
1 *
0.
R3
O
-a
O)
P^
2
in
c



,
r-|
O O
o o
LO to
A
co
"""*


0
o
r-l
A
CM




O
O
CM
A
CM






O
O
O
A
r-l




O O
O
to
A
to




o o
o o
IO CM


0 0
o








. % t
in •!-
O -P  r—
u
r— ^~
TU ^3 (O



4
CM























































•P
in
o
o

O)

'•p
A3
s_
cu
CL
O
p_
 C (T3
C T- Q.
•r- i. CU
-P 0 S-
as -P
S- T- J4
O. O CO
O S i—

n*—* v
(O i"H




O 0 O
o o o
o r"» 10
to LO CM



o o
0 0
o •*
A
co




0 0
0 0
P^» to








o o
0 0
CO CO






0 0
O 0
r-- co
A A
r— 1 r— 1




o
o
en
A
r-l


O
O
»-H
A
«sj*




fj
S_
O
CL.
o.

co
°9

C
o m
•r- 3
•P CO O
re o co
s_ c c
4-> re re
in c: i—
'c -p "co
•r- C O
-o re T-
«c 2: s

• ^— « «"•*
CM .0 a




o
o
CM
co



o
0
LO
A
LO




o
o
LO
A
CO






o
o
10
A
r-<




o
o
to
A
en




o
o
A
CO


•o
o
to
^J"
r— 1











•p
in
o
o
3
C
re

"re
•p
o
i—


•
«i-
O
o
o
A
r— 1
co


o







^M,.
O
o
CO
A
,__(
^_,




^'•*
O
o
r-l
A
CM
•—



O
O
co
A
LO
•— *


^_^
o
o
o
A

re
o
0
CO
A
P*1-
T-H









4->
.,—
-o
a>
S-
u
t*
CO
0
o

s-
o
o



•
LO
o
O CO
o o
CM CM
CM
t— 1


O
O
LO 1
A
LO




o o
O CM
1^. LO
A
r-l





^"1* ,*™H
O LO
O CM
LO r-l
*—* *•— • *





0 0
CD CO
CO «5f
A
^~




O LO
o r^
«3- IT)
A


O LO
o en
CM —
A
CO
"""*







s_
CD
Q.
4*^
in in
o in
(J CO
c:
T3  O
IM T- O
•r- 4J >
i— O
re a» <+-
3 M- 0
c <+-
ST co a>
03 s:
4-S
+J W
CO O
Z 0


• •
to r-











































sl
-C*^
siP
Cft
tJ5
• O3
4-> U-)
•5 Ir-
CO O
i.
U «=
O
4-> -P
O . O
C 3
CO "O
•o 
4J i — 
re co re
> ce: CQ
re 02 o
8-24

-------
        Table 8-8.  INITIAL LEAK REPAIR LABOR-HOURS REQUIREMENT
                       FOR VALVES FOR THE MODEL UNIT

No. of valves
per model
402
524
Initial leak
frequency
0.114
0.065
Estimated no. of
initial leaks
45.8
34.1
Repair time,
man-hours3
1.13
1.13
Labor-hours
required,
man-hours
51.8
38.5
90.3
 Based on 75 percent valves repaired on-line in 10 man-minutes and
 25 percent repaired off-line in 4 man-hours.
        Table 8-9.  TOTAL ANNUAL COSTS FOR INITIAL LEAK REPAIR
                     FOR VALVES FOR THE MODEL UNIT
                          (May 1980 Dollars)
Initial leak repair labor charges $18/hour

Admin. & Support costs, 40 percent of labor charges

Total costs

Annualized charges for initial leak repair
16.3 percent of total  cost
1,630

  650

2,280


  370
^Capital recovery factor is 0.163 based on initial leak repair costs
 amortized over 10 years at 10 percent interest.
                                8-25

-------








r-
i |

~^

i
r~|
Q
O

LU

1—
fV
O
I -
f ' j
CO
LU
L j

fe-e
LU O_
il t
C£ T-
1— 1 fO
ra Q.
o- cu
LU C£
oc
TO
o; E
O (0
co

i— i O
«=C CU
Q_ 4J
I tJ QJ
o; o
^ ^
>
^- c--
+J
CD E
*Z O

(^ ^2*
O
t— 4
0
^—
	 I

ra
§
"=£

o
r—
CO
CJ
^
«3









i- co
•r- S-
ra 3 ** CO
a. o -a s-
CU J= CU 3 00 UO CO
S— *H~ _E to CM CO
-i£ o 3 i co i — cn
^ • •
H- 10 CT>
os- r^ en
cu
• Q.
O



cn «co
c , -a s_
•r-  Cn
S- S- S_ 0 .
o o -r- js: co 1.0 oo
+J -Q 3 1 LO O to
•r- ra CT E , — CM OO
Ei — CU ra
^^
^-

E -a
o cu cr> en
•M CU ^ ^
o cu i— .
ra S- r— __
S— O
LJ_ CO
C7)
E ra
O CU S
+•> E 1 CM CM
•1— T- E
E •*-> ra
O E



CD | ^ [ ^
O T- CU CU
^~ E E
CU O 3 3
Q-+-» S- S-
^5  a ^ LO -
ra


"O
(1) .p-
O | ~T
• T— 4J O"
> J=-r-
5- CO Olc— i—
CU ro -i- ra
(/I CD —I 4->
0
1—






































^
ra
cu
4->
E
ra
1
CM
s_
o
4-
cu
-M
3
E
•r-

CO


CU
£
•M
CD
E
S-
O
•(->
E
O

4J
E
CU
E
3
S-
•(->
co
E
HH
ra



p.
o
T3
CU
ffi
s_
cu
Q.
o

CO
cu
o
S-
0
CO
q-
o
J-
o
, ^
U
ra
s-
M-

0

-a
cu
CO
ra
.Q
A
T3
0
E

cc
Q
_l
CU
+->
1
4-
CM
-a
E
O
ra
E
S-
3
+J

s- c7
1*11
UJ
> eu
O CJ
E
-a cu
E S-
3 CU
o "<—
H- CU
CO ^-^
ra
CU •
E
<4- CU
O O
S-
S- CU
CU Q.
-Q
E cn
3 r—
E
O
CU 4->
CD
ra r—
i- ra
CU 3
> cr

-------
        Table 8-11.  ANNUAL MONITORING AND LEAK REPAIR COSTS FOR
            MONTHLY MONITORING OF VALVES FOR THE MODEL UNIT
                          (May 1980 Dollars)
Labor charges of $18 per hour for total 562.7 labor-hours,3 $ 10,130
Admin. & Support costs at 40 percent of labor
charges, $

Annualnzed charge for initial leak
repair , $
Total annual  costs, $

Annual product recovery credit,0 $

Net annualized costs, $

Cost Effectiveness, $/Mg
  4,050


    370

 14,550

(17,750)C

($3,200)

    (95)
 Total of 363.9 monitoring labor hours and 198.8 leak repair labor
 hours from Table 8-10.
bFrom Table 8-9.
cProduct recovery credit is calculated at $528/Mg.  The emission
 reductions are 14.4 Mg/yr from vapor service valves and 19.23 Mg/yr
 from light liquid service valves (see Table 8-5).
 Figures in parenthesis indicate credits.
                              8-27

-------
Pumps
     The fugitive VOC emission control specifications for light liquid
pumps include monthly monitoring supplemented by weekly visual inspections
and leak repairs.  Therefore, the annual costs associated with controls
include initial leak repair, including pump seal replacement costs, and
recurring monthly monitoring and leak repair, including seal replacement
costs.  The factors affecting the costs of a leak detection and repair-
program are (1) monitoring time, (2) repair time, and (3) cost of
replacement seal.  The cost estimates were based on 10 man-minutes for
instrument monitoring, 0.5 man-minutes for visual monitoring, and 16
man-hours per pump for repair.  Every.month 4.2 percent of all pump
seals will be replaced as routine maintenance.  On the average, half of
routinely maintained seals, i.e., 2.1 percent of all  seals, are assumed
to be leaking seals.
     Based on the LDAR model, the annual costs of leak detection and
repair have been calculated for monthly programs for pumps.  The
calculations are presented in Tables 8-12 through 8-15.  The input
parameters, e.g., occurrence rate, initial frequency, and others, are
discussed in the Emission Reduction section (Section 4) of the SOCMI
AID.55
Safety/Relief Valves
     The fugitive VOC emission control specifications for safety/relief
valves include installation of rupture disk systems on the relief valves.
There are no requirements for leak detection and repair.  It is assumed
that leaks would be corrected by routine maintenance with no additional
labor requirement.  Therefore, the only annual costs associated with
controls for safety/relief valves are equipment costs.
Equipment Costs:  Costs were computed for the installation of a rupture
disk upstream of a safety/relief valve in gas service.  These costs were
based on estimates from the Hydroscience (now IT Enviroscience) report.56
The cost estimates were based on the following assumptions:   No piping
modification was required, and the disk and its holder simply could be
inserted between the flanges of the relief valve and  the system it
protects.  To allow in-service disk replacement, a block valve or a
                                    8-28

-------
         Table 8-12.   INITIAL LEAK REPAIR LABOR-HOURS REQUIREMENT
                   FOR PUMP SEALS FOR THE MODEL UNIT
No. of pump
seals per
model unit
29
Initial leak
frequency
0.088
Estimation no. of
initial leaks
2.6
Repair
time,
man-hours
16
Labor-hours
required,
man-hours
41.6
          Table 8-13.   TOTAL ANNUAL COSTS FOR INITIAL LEAK REPAIR
                    FOR PUMP SEALS FOR THE MODEL UNIT
                            (May 1980 Dollars)
Initial leak repair labor charges,
at $18/hour

Admin. & Support costs, 40 percent of
labor charges

Seal costs, $139.4/single seala>b
Total Costs

Annualized charges for initial  leak
repair, 16.3 percent of total  costs
  750
  300

  360

1,410


  230
aSeal cost is $139.4.  The value has been obtained by updating a cost
 value of $113 for last quarter 1978.  The cost includes 50 percent
 credit for old seal.
Calculation = No. of initial  leaks (from Table 8-12) x seal cost
             = 2.6 x 139.4 = $362

GCapital recovery factor is 0.163 based on initial leak repair costs
 amortized over 10 years at 10 percent interest.
                               8-29

-------
1—
1 — 1
2:
~^

i
LU
Q
O
s:

LU
rn
1—

Lt-
0

(S)
	 i
<^
LU
to
CL
s:
Q.
f — ,.
r*s £=
O ro
1 1 ^
0)
oo o
1— S-
z: o.
LU
s: s_
LU T-
CC. to
H-) Q.
^3 01
t y ***>^
LU
rv "O
E
C£ ro
O
CQ E
•=C O
_J •!-
•P
rv CJ
l— l QJ
eC -P
Q. QJ
LU Q
o;
v>*
i£ rO
*3I QJ
LU _1
	 1
>^
Q r—
^y ^"
(^ *P
E
C3 0
z. s:
• — i • — •
c£
o
1—
i — i
?^
0
s:

	 i
> •—
0 r—
S-
• QJ
0 Q.




CD
E «> CO
•i- -0 S-
S- S- QJ 3 CD tn LO
O O S- O • •
-P J2 -r- J^ CO CM O
•r- rO 3 1 LO i — r^
E i— CT E
O QJ ro
S S- E



T3
QJ i-
CO S- ro
QJ O QJ
E -P >5 CM CM
•I- -I- i— LO
1— E S-
O QJ
E 0.



O)
E ro
•r- E LO
S_ -.r-
O QJ E O CD
1 \ ^ | ^
•r- "r™ C"
E -P ra
0 E
s:



en -p
E E
H— •!— QJ
OS- E
0 3 r—
QJ -P S- ro
o «t— ^J ^
>j E CO CO
1— O E -r-
E t~H ^^


Q.
E -P
Z3 S- •!-
CX QJ E
Q. 13
<4- CTl
O CO i — CM
i— a>
• ro -a
O QJ O
'z: ts> &










































•
E
ra
QJ
4-}

E
ra
E
i
CM

rO

S-
O


co
QJ

^
E
•t—
E

LO

CO
• i —


S- 5-P OJ
i— C£
c— o^** 	 ^
•P C
E -I-
O CO •
E = to
a>
-P "O -P
E QJ ro
QJ -P E
CJ 3 •!-
S- Q.-P
QJ E co
Q. O QJ
O
LO E
• CD O
LO S- •!—
QJ CO
E 3 CO
O -r-
co E
T3 Q) QJ
QJ =J
CO i — T3
ra ro E
CQ > (O
J3
8-30

-------
        Table 8-15.  ANNUAL MONITORING  AND  LEAK  REPAIR  COSTS  FOR
          MONTHLY MONITORING  OF  PUMP SEALS  FOR THE  MODEL  UNIT
                           (May 1980 Dollars)
Labor charges of $18 per hour for  total 259.3
labor-hours,  $

Admin. & Support costs at 40 percent of
labor charges, $

Annualized charge for initial leak
repair,  $

Annual replacement cost of leaking
seals, $

Total annual costs

Annual product recovery credit,  $

Net annualized cost, $

Cost effectiveness, $/Mg
 4,670


 1,870


   230


 1.650

$8,420

(4,030)e

 4,390

   575
 Total of 70.5 monitoring hours and 188.8 leak repair labor hours from
 Table 8-14.

DFrom Table 8-13.
*•
'Based on $139.4/seal  and 11.8 leaking seals per year obtained from
 Table 8-14.

 Product recovery credit is calculated at $528/Mg.  The emission
 reductions obtained are 7.63 Mg/yr from pump seals (see Table 8-5).
a
"Figures in parentheses indicate a credit.
                                8-31

-------
3-way valve was assumed to be installed upstream of the rupture disk.
In addition, to prevent damage to the relief valve by disk fragments,  it
was assumed that an off-set mounting would be required.  The rupture
disk life was assumed to be 2 years.
     Equipment cost estimates for control  of fugitive emissions from
safety/relief valves were calculated for two different systems:
(1) rupture disk with block valve and (2)  rupture disk with 3-way valve.
In computing total costs, half of the safety/relief valve sources
(i.e., 5.5 from Table 8-5) were assumed to be installed with rupture
disks with block valves and the remaining half with the rupture disks
with 3-way valves.  These costs are shown in Table 8-16.  The total
estimated installed cost of a new rupture disk system with block valve
was $1,995 in second quarter 1980.  The total estimated installed cost
of a new rupture disk with 3-way block valve was $4,137 in second quarter
of 1980.  Based on equipment installation costs, total capital  and
annualized costs for control of emissions from the model unit's
safety/relief valves were calculated and are presented in Table 8-17.
Open-ended lines
     Although all open-ended lines were assumed controlled under baseline
and, thus, no fugitive emissions would occur (see Table 8-5), cost data
were generated for open-ended lines and are presented in Table 8-18.
As there are no fugitive VOC emission sources, cost data were not
generated for open-ended lines.
Compressors
     The fugitive VOC control specifications for compressors require
control of degassing vents.  The costs of control are presented below.
Equipment Costs:  The cost of control equipment for compressors was
based on installation of closed vents for degassing reservoirs of
compressors.  The estimate was based on information contained in the
Hydroscience (now ITE) report56 and was for the following items per
compressor:
      122 m length of 5.1-cm diameter schedule 40
     carbon steel pipe
     Three 5.1-cm cast steel plug valves and one
     metal gauge flame arrestor
               Total for second quarter 1980
                                    8-32
.$6,400

 $1,600

 $8,000

-------
   Table 8-16.  RELIEF VALVE CONTROL COSTS FOR RUPTURE  DISK  SYSTEMS
                WITH BLOCK VALVES AND THREE-WAY VALVES
                           (May 1980 dollars)
(1)  Rupture disk systems with block valve

     Rupture disk and assembly

       One 7.6 cm stainless steel rupture disk
       One 7.6 cm carbon steel rupture disk holder
       One 0.6 cm dial face pressure gauge
       One 0.6 cm carbon steel bleed valve
       One 7.6 cm gate valve
       One 10.2 cm tee and one 10.2 cm elbow
       Installation at 34 total hours and $13/hr

       Total  cost for second quarter 1980
     Rupture disk at 58 percent of total costs'
     Assembly at 16.3 percent of total costs
     Maintenance & Miscellaneous.
(2)
                      Total $/year

Rupture disk systems with 3-way valve.

Rupture disk and assembly

  One 7.6 cm stainless steel rupture disk
  One 7.6 cm carbon steel disk holder
  One 0.6 cm dial  face pressure guage
  One 0.6 cm carbon steel bleed valve
        6 cm safety/relief valve
       One
       Two
        6 cm elbows
       One 10.2 cm tee and one 10.2 cm elbow

                           Subtotal •

     Three-way valve

       One 7.6 cm, 3-way,  2 port valve
       Total  Installation

                           Subtotal

       Total  cost for second quarter 1980
     Rupture disk  at  58 percent of total  costs
     Assembly at  16.3 percent of total  costs
     Maintenance  & Miscellaneous.
                           Total  ($/year)
                                                        Capital costs
                                                       230
                                                       384
                                                        18
                                                        30
                                                       700
                                                        21
                                                       612
   $1,995

Annualized costs

     $133
      288
      180

     $601

  Capital costs
      230
      384
       18
       30
    1,456
       30
    	21

    2,169
                                                     1,320
                                                       648

                                                     1,968

                                                    $4,137

                                                 Annualized costs
                                                    $1,142
 Capital  recovery  factor is  0.58 based  on  2-year equipment life and
 10 percent  interest.

^Capital  recovery  factor is  0.163 based on 10-year equipment life and
 10 percent  interest.

"Based  on  9  percent  of  total  capital  costs.
                                    8-33

-------
       Table 8-17.   CAPITAL AND NET ANNUALIZEO COSTS FOR CONTROL OF
         EMISSIONS  FROM SAFETY/RELIEF VALVES FOR THE MODEL UNIT
                           (May 1980 Dollars)
      Costs
Rupture disk system0
 Installed  capital  cost,b  $

 Annualized costs,0 $

 Annual operating cost,  $

  Maintenance at 5 percent of capital  cost,  $
  Miscellaneous at 4 percent of capital cost,  $

 Total annual cost,  $

 Annual product recovery credit,  $

 Net annualized costs, $

 Cost effectiveness, $/Mg
       33,730

        6,550
        1,690
        1,350

        9,590

       (5,290)

        4,300

          430
aOne half of the total systems with block valves and the remaining
 half with 3-way valves.

 Based on $3,066 per system and 11 sources.  The cost of one system is
 computed from Table 8-16 as follows:
  0.5 ($1,995 for rupture disk system with block valve and assembly +
  $4,137 for rupture disk system with 3-way valve and assembly).
cBased on annualized cost data in Table 8-16 per system and 11 sources.
 The cost of one system is computed as follows:   0.5 [($133 + 288)
 annualized cost for rupture disk system with block valve and assembly +
 ($133 + 637) annualized cost for rupture disk system with 3-way valve
 and assembly).

 Product recovery credit is calculated at $528/Mg and an emission
 reduction rate  of 10.02 Mg/yr (see Table 8-5).
                                8-34

-------
 Table 8-18.  CAPITAL AND NET ANNUALIZED COSTS FOR CONTROL OF EMISSIONS
                         FROM OPEN-ENDED LINES
                           (May 1980 dollars)
         Costs
    Caps
Installed capital cost, $
Annualized cost at 16.3 percent
  of capital cost, $
Annualized operating costs, $
  Maintenance at 5 percent of
  capital cost, $
  Miscellaneous at 4 percent of
  capital cost, $
Total annual costs before credit, $
Recovery credit3, $
Net annualized costs, $
Cost effectiveness, $/Mg
    53

     8.6


     2.7

     2.1
    13.4
 7,870
(7,857)b
  (527)
aRecovery credit based on uncontrolled VOC emission factor of
 0.0408 kg/day.  Based on 100 percent control efficiency, for caps
 and $528/Mg VOC emission reduction.
 Emission Reduction:
 0.0408 kg/day/open-ended line x 365 day/yr x 1 Mg/1,000 kg = 14.9 Mg/yr.
 Recovery Credit:
 14.9 Mg/yr x $528/Mg VOC = $7,870/yr/open-ended line.
^Values within parentheses denote credit.
                              8-35

-------
The above costs include connection of the degassing reservoir to an existing
existing enclosed combustion device or vapor recovery header.  The cost
of a control device, added specifically to control the degassing vents,
is, therefore, not included.
     Total capital and net annualized costs for control of emissions
from compressor seals for the model units were developed.  These cost
data are presented in Table 8-19.
Sampling Systems
     Equipment costs were computed for closed loop sampling connections.
The cost estimates were based on information from the Hydroscience
(now ITE) report56 and was for the following items per sampling system.
Table 8-20 presents capital  and annual  costs for control  of emissions
from sampling systems in the model  plant.
                                                          $210
                                                          $320
      One  6-m  length of 25-cm  diameter schedule 40
      carbon steel pipe and three 2.5-cm carbon
      steel ball valves.
      Installation (at 18 hours and $18/hr)
          Total cost (second  quarter 1980 dollars)
 8.1.7  Cost Analysis Results
     The  results of the cost  analyses for all regulatory alternatives
and model plants are summarized in this section.  The installed capital
costs by control device and the total operating and other annualized
costs are presented for each  regulatory alternative in Tables 8-21 to
8-31b, for the twelve model plants.  Tables 8-32 to 8-42b show the costs and
associated emission reductions of regulatory alternatives for the eleven
model plants for the implementation of an alternative both from the
baseline control level  and from the next less stringent alternative.
     National  control  costs of the most stringent regulatory alternatives
over baseline controls are estimated, as shown in Table 8-43, based on the
projected number of new process lines,  which are described in Chapter 9
"Economic Impact".   The total  nationwide fifth-year net annualized cost
in these new process lines is estimated to range from $11.2  million for
baseline control to about $26.7 million for the maximum achievable control
level.  The maximum achievable control'level  from baseline is equivalent
to an annual  VOC emission reduction of  about 4,700 Mg.
                                   8-36

-------
 Table 8-19.  CAPITAL AND NET ANNUALIZED COSTS FOR CONTROL OF EMISSIONS
                FROM COMPRESSOR SEALS FOR THE MODEL UNIT
                            (May 1980 dollars)
         Costs
Degassing reservoir vent
Installed capital cost,a>b $
Annualized cost at 16.3 percent
  of capital cost,c $
Annualized operating costs, $
  Maintenance at 5 percent of
  capital cost, $
  Miscellaneous at 4 percent of
  capital cost, $
Total annual costs before credit, $
Annual product recovery creditd
Net annualized costs, $
Cost effectiveness, $/Mg
         6,400

         1,040


           320

           260
         1,620
        (2,110)e
          (490)e
          (120)e
aBased on two compressor sources in the model plant and the assumption
 that controls apply to only 40 percent of the sources.  (60 percent
 of compressors within the industry are known to be controlled and
 need not be considered for the purpose of cost analysis.)   (Reference 2)
bCapital  cost per compressor is $8,000.
cCapital  recovery factor, 0.163, is based on the equipment amortized
 over 10 years at 10 percent interest.
dProduct credit is calculated based on  $528/Mg and emission reduction of
 3.99 Mg/yr.
eValues within parentheses denote credit.
                              8-37

-------
 Table 8-20.  CAPITAL AND NET ANNUALIZED COSTS FOR CONTROL  OF EMISSIONS
                FROM SAMPLING SYSTEMS FOR THE MODEL UNIT
                           (May 1980 dollars)
         Costs
Closed purge sampling
    connections
Installed capital cost,3 $
Annualized cost at 16.3 percent
  of capital cost,b $
Annualized operating costs, $
  Maintenance at 5 percent of
  capital cost, $
  Miscellaneous at 4 percent of
  capital cost, $
Total annual costs, $
Annual product recovery credit0, $
Net annualized costs, $
Cost effectiveness, $/Mg
      13,800

       2,250

         690

         550
       3,490
      (l,810)d
       1,680
         490
aBased on $530 per sampling system and 26 sampling systems in the
 model plant.
bCapital recovery factor, 0.163, is based on the equipment amortized
 over 10 years at 10 percent interest.
cProduct credit is calculated based on $528/Mg and 3.42 Mg/yr of VOC
 recovered annually (see Table 8-5).
dValues within parentheses denote credit.
                               8-38

-------
  Table  8-
21.   POLYPROPYLENE,  LIQUID  PHASE MODEL  PLANT, REGULATORY
     ALTERNATIVES COSTS  (June 1980  dollars)9'0
                                                   Regulatory Alternative
                                                         2          3
 INSTALLED CAPITAL COST ($)

  Hare(s)
  Flare Ducting
  Thermal Incinerator
  Thermal Incinerator Ducts,
    Fans & Stack
  Catalytic Incinerator
  Catalytic Incinerator Ducts,
    Fans & Stack
  Condenser
  Fugitive Leak Detection
    and Repair (LDAR)

  Total

ANNUALIZED COST ($/yr)

  Direct

    Operating Labor
    Operating Materials
      (e.g., catalyst)
    Mai ntenance
                            45,600
                            40,400
 45,600
 40,400
                            85,900
                            33,500
 28,400

114,300
 40,500
 Costs are on a per process line basis.
b
 Some totals do not add up exactly due to rounding.

 Includes miscellaneous operating costs  for fugitive
 45,600
 40,400
217,800.
 62,100
 58,500
 45,800
217,800
                                                      86,700    86,700
 28,400     28,400

418,800    437,100
 73,200
Materials and Labor
Utilities
Natural Gas
Electricity
Steam
Subtotal
Indirect
Taxes, Insurance, &
Administration0
Capital Recovery
Subtotal
Recovery Credit
Total (Direct +
Indirect - Credit)
4,300
12,800
11,100
61,700
3,440
12,400
15,800
77,500
8,200
12,800
11,100
72,500
4,570
18,500
23,000
10,300
85,200
23,500
16,400
120
11,100
113,200
16,800
22,100
84,700
10,300
187,600
24,400
20,700
120
11,120
129,500
17,500
70,600
88,100
10,300
207,300
                                      leak detection  and repair program.
                                         8-39

-------
Table  8-22.   POLYPROPYLENE, GAS PHASE  MODEL PLANT  REGULATORY
            ALTERNATIVES  COSTS  (June 1980 dollars)
                                                            a,o
                                            Regulatory Alternative
                                             1               2
  INSTALLED CAPITAL COST ($)

   Flare(s)
   Flare Ducting
   Thermal Incinerator
   Thermal Incinerator Ducts,
     Fans & Stack
   Catalytic Incinerator
   Catalytic Incinerator Ducts,
     Fans It Stack
   Condenser
   Fugitive Leak Detection
     and Repair (LDAR)

   Total

  ANNUALIZED COST  ($/yr)

   •Direct

     Operating Labor
     Operating Materials
       (e.g., catalyst)
     Maintenance
104,400
 59,300
163,700
 22,300
104,400
 59,300
 28,400

192,100
 29,300
Materials and Labor
Utilities
Natural Gas
Electricity
Steam
Subtotal
Indi rect
Taxes, Insurance, &
Administration0
Capital Recovery
Subtotal
Recovery Credit
Total (Direct +
Indirect - Credit)
8,200.
12,800
13,000
56,400
6,540
. 23,400
29,900
86,300
12,100
12,800
13,000
67,200
7,670
29,500
37,200
10,300
94,000
 d
  Costs are on  a per basis line basis.
 b
  Some totals do not add up exactly due to rounding.

  Includes miscellaneous operating costs for fugftive leak
  repair program.
             detection and
                                       8-40

-------
       Table  8-23.   LOW DENSITY POLYETHYLENE, HIGH  PRESSURE MODEL PLANT

              REGULATORY ALTERNATIVES COSTS  (June  1980 dollars)a'b

INSTALLED CAPITAL COST ($)
Hare(s)
Flare Ducting
Thermal Incinerator
Thermal Incinerator Ducts,
Fans & Stack
Catalytic Incinerator
Catalytic Incinerator Ducts,
Fans & Stack
Particulate Removal
Fugitive Leak Detection
and Repair (LDAR)
Total
ANNUALIZED COST ($/yr)
Direct
Operating Labor
Operating Materials
(e.g. , catalyst)
Mai ntenance
Materials and Labor
Utilities
Natural Gas
Electricity
Steam
• Subtotal
Indirect
Taxes, Insurance, &
Administration0
Capital Recovery
Subtotal
Recovery Credit
Total (Direct +
Indirect - Credit)
1

24,500
71,700








96,200


11,200


4,800

4,280
3,500
23,700


3,850
14,900
18,700


42,500
Regulatory Alternative
2 3

24,500
71,700







23,600
119,800


16,400


8,400

4,280
3,500
32,600


4,750
20,000
24,800
7,750

49,700

24,500
71,700



112,000
101,500


23,600
333,300


33,100
4,290

19,100

10,000
1,330
3,500
71,300


13,300
54,700
68,100
7,750

131,700
4

24,500
71,700



165,000
195,300


23,600
480,100


49,800
5,360

26,400

11,500
1,660
3,500
98,200


19,200
78,600
97,800
7,750

188,300
5

526,900
477,700



165,000
195,300


23,600
1,388,500


61,000
5,360

71,800

28,600
1,660
27,200
195,600


55,500
210,700
266,300
7,750

454,200
 Costs are on a per process line basis.

b
 Some totals do not add up exactly due to rounding.


 Includes miscellaneous operating costs for fugitive leak detection and repair program.
                                          8-41

-------
Table 8-24,. LOW DENSITY POLYETHYLENE, LOW PRESSURE AND HIGH DENSITY
POLYETHYLENE, GAS PHASE MODEL PLANT REGULATORY ALTERNATIVES COSTS3 'b
(June 1980 dollars)
Regulatory Alternative

INSTALLED CAPITAL COST ($)
Flare(s)
Flare Ducting
Thermal Incinerator
Thermal Incinerator Ducts,
Fans & Stack
Catalytic Incinerator
Catalytic Incinerator Ducts,
Fans & Stack
Condenser
Fugitive Leak Detection
and Repair (LDAR)
Total
ANNUAL IZED COST ($/yr)
Direct
Operating Labor
Operating Materials
(e.g., catalyst)
Maintenance
Materials and Labor
Utilities
Natural Gas
Electricity
Steam
Subtotal
Indirect
Taxes, Insurance, &
Administration0
Capital Recovery
Subtotal
Recovery Credit
Total (Direct +
Indirect - Credit)
1 2

45,000 45,000
88,900 88,900








38,000
133,900 171,900


33,500 43,900



6,690 11,000

25,700 25,700

13,590 13,590
79,500 94,200


5,360 6,860
20,400 28,500
25,700 35,300
15,500

IOS.,200 114,100
3

57,900
94,300








38,000
190,100


55,000



12,000

30,000

13,630
110,600


7,590
31,100
38,600
15,500

133,800
4

57,900
94,300



81,600

94,000


38,000
365,700


71,700

2,870

20,700

40,800
890
13,630
150,700


14,600
59,600
74,200
15,500

209,400
5

138,000
169,200



81,600

94,000


38,000
520,800


82,900

2,870

28,500

49,300
890
,18,800
183,300


20,800
82,400
103,200
15,500

270,900
Costs are on a per process line basis.
                                             8-42

-------
Table 8-25.
HIGH  DENSITY POLYETHYLENE, LIQUID PHASE-SLURRY
                                           a,b
           MODEL  PLANT  REGULATORY ALTERNATIVES COSTS
                         (June  1980 dollars)
Regulatory Alternative

INSTALLED CAPITAL COST ($)
Flare(s)
Flare Ducting
Thermal Incinerator
Thermal Incinerator Ducts,
Fans & Stack
Catalytic Incinerator
Catalytic Incinerator Ducts,
Fans & Stack
Condenser
Fugitive Leak Detection
and Repair (LDAR)
Total
ANNUALIZED COST ($/yr)
Di rect
Operating Labor
Operating Materials
(e.g., catalyst)
Mai ntenance
Materials and Labor
Utilities
Natural Gas
Electricity
Steam
Subtotal
Indirect
Taxes, Insurance, &
Administration0
Capital Recovery
Subtotal
Recovery Credit
1

25,100
45,300









70,400


22,300



3,520

8,560

7,390 '
41,800


2,820
10,700


2

25,100
45,300








28,400
98,800


29,300



7,420

8,560

7,390
52,600


3,950
16,800

10,300
3

25,100
45,300



53,000

89,100


28,400
240,900


46,000

620

14,500

9,070
190
7,390
77,700


9,630
39,900

10,300
  Total (Direct +
    Indirect - Credit)                 55,300

a
 Costs are on a per process line basis.
b
 Some totals do not add up exactly due to rounding.

 Includes miscellaneous operating costs  for fugitive
 repair program.
                                 63,000
116,900
                               leak detection and
                                  8-43

-------
Table 8-26.   HIGH  DENSITY  POLYETHYLENE,  LIQUID  PHASE-SOLUTION
          MODEL PLANT REGUALTORY ALTERNATIVES COSTS3'b
                        (June 1980 dollars)
Regulatory Alternative

INSTALLED CAPITAL COST ($)
Flare(s)
Flare Ducting
Thermal Incinerator
Thermal Incinerator Ducts,
Fans & Stack
Catalytic Incinerator
Catalytic Incinerator Ducts,
Fans & Stack
Condenser
Fugitive Leak Detection
and Repair (LDAR)
Total
ANNUALIZED COST ($/yr)
Pi rect
Operating Labor
Operating Materials
(e.g., catalyst)
Maintenance
Materials and Labor
Utilities
Natural Gas
Electricity
Steam
Subtotal
Indirect
Taxes, Insurance, &
Admini strati one
Capital Recovery
Subtotal
Recovery Credit
Total .(Direct +
Indirect - Credit)
1 2

38,000 38,000
29,400 29,400








28,400
67,400 95,800


33,500 40,400



3,370 7,270

12,800 12,800

5,020 5,020
54,700 65,500


2,700 3,830
9,780 15,900
12,500 19,700
10,300

67,200 74,900
3

38,000
29,400



425,100

181,600


28,400
702,500


57,100

43,800

37,600

181,500
13,600
5,020
338,700


28,100
114,600
142,700
10,300

471,000
 Costs are on a per process line  basis.
 b
 Some totals do not add up exactly due to rounding.
 Includes miscellaneous operating costs for fugitive leak detection and
 repair program.
                                 8-44

-------
                 Table 8-27.   POLYSTYRENE CONTINUOUS MODEL  PLANT  REGULATORY

                         ALTERNATIVES COSTS  (June  1980  dollars)3'5
Regulatory Alternative
1 234
INSTALLED CAPITAL COST ($)
Condenser
Fugitive Leak Detection
and Repair (LDAR)
Total
ANNUALIZED COST ($/yr)
Direct
Operating Labor
Operating Materials
(e.g'. , catalyst)
Maintenance
Materials and Labor
Utilities
Natural Gas
Electricity
Steam
Freon
Subtotal
Indirect
Taxes, Insurance, &
Administration1:
Capital Recovery
Subtotal
Recovery Credit
Total (Direct +
Indirect - Credit)

1,550

38,000 38,000
0 38,000 39,550


10,400 10,897



4,350 4,428


192

2
0 14,800 15,519


1,500 1,562
8,100 8,352
0 9,600 9,914
15,500 18,373

0 8,850 7,060

1,550

38,000
39,550


10,897



4,428


207

2
15,534


1,562
8,352
9,914
18,500

6,948
5

1,717

38,000
39,717


10,897



4,436


^286

2
15,621


1,569
8,379
9,948
18,596

6,973
6

2,409

38,000
40,409


10,897



4,470


297

6
15,670


1,596
8,492
10,088
18,628

7,130
7

2,890

38,000
40,890


10,897



4,495


371

8
15,771


1,616
8,570
10,186
18,644 •

7,313
 Costs are on a per process line basis.

b
 Some totals do not add up exactly due to rounding.


 Includes miscellaneous operating costs for fugitive leak detection and repair program.
                                            8-45

-------

o -
_i   ce
CO   O

             Q|
             sc
                           gg
                           IO f-H
                                            O O
                                            o o
                                            CO CO
                                                                                                            IS
                           VO C
                           CO C
                                       g   g
                                       r-.   co
                                                                                                           o o
                                                                                                           o o
                                                                                                           LO CM
                                                                                                           o o   o
                                                                                                           00   O
                                                                                                           t—« .—I   CM
                                             r-4      CO
                                                                                                                       r-4     O
g   g
co   en
                                                    o


                                                    CO
                                                                       g    2
                                                                       CO    ^H
                                                                       O    O    O
                                                                       o    *-<    o
                                                                       VO    r-H    CH
                                                                                     o o o
                                                                                     O CT( «3
                                                                                     t-< LO r**
                                                                                     o ch
                                                                                     o tr>
                                                                                                           o o
                                                                                                           LO O
                                                                                                           en o
                                                                                                           oo   o
                                                                                                           CO O   CO
                                                                                                           _4 r—I   CM
                               	C U U  U U
                               : t~ T- ra c  c ra
                                                                            ro  (U u)   CD -i—
                              ^- O O 4-1 «-* «-• •
                            •— * 3
                            Svi O
                            •^-
                       _j   cu  fa    -u
                                                                       ro  ra  • 4J
                                                                                        (j      0)   i	
                                                                                                                                      ui    E   o
                                                                                                                                      o    o   c
                                                                      8-46

-------


1—
_l
Q.
—I
LU
O
O
^.
oo
00
1 1 1
o
o:
Q.

o
i — i
oo
z:
UJ
0.
oo
Z3
oo

1—

00
'1
z:
UJ
UJ
Qi

| —
OO

	 1
0
°~
UJ
en
a
•^

Q_
X
UJ


•
01
CM
CO
 4-> L. L.
_i s- s- c: c
< OJ O •>> •*- •!—
>— cn c: c o u o
O O 1-H t-H O O
O trt O i— .— 4-> *J
j "a* o g i c^^*
jg U. L. h- »— O CJ
—
0 00
o o o
o  o: H
C CD 4J T3 r- ^ O
U_ C *O) to 4J < &.
O 3 O =3 i-
1
§ S
CM PO
cn ^>
to
So
«*
O PO
CO *f
in




g s
cn PO
% **



o o
O ^!
CM r-*
O CM
PO



O
in
PO




PO
CM


















atlng Labor
atlng Materials
.g., catalyst)
tenance
*- J- CD C
a&— 5
O O E

g §Si g
cn r*. PO co «sr
CM CM *"** PO
§ ggg g
o cn PO 10 CM
PO PO -H CM PO
CM CM -*




O O O O O
o o in CM o
CM r-^.-i PO
CM ^ cn



8000 0
o in CM o
s s - s



o o o o
o cn CM o
in CM co to
*r *? _T co*
CM



SO
*~t
PO tO
PO in*





o










t.
o
re
_j
! 3$
r— O
re u» r— *^
•M -^- -M oj eu re o

2S



o o
PO O
^ CO
oTco*
PO



§1
^ r»>




SI

•o
2
O
"re D
4-1 O)
S S
•i g
CO tt

1
CM
O
O
CM
cn
^"*



§
CM
O*
r-*



O
O
to
o



g
cn
CO
CM



CM
to*





0 -












(Direct +
rect - Credit)
^ -5
re c
r































.
- ; .?
C
1
S
Vt Ol
•»- '3
re
-° >»
SC
o
•*- re
*" 2
tA
VI O.
Ql 3
0 =
o -o
k ^
S- 4->

* S
S „
i/) E
O O















i-
0
o.
b.
re
8-


C
re
e
o
s
•g

re
0)
•—
>
5
•£
u
o

. M
O
cn
c
4-»
£
0)
§•

s miscellaneous
(U
•a
3
O
C
o
8-47

-------
         Table 8-30.   POLY(ETHYLENE TEREPHTHALATE)  (PET)  -  DMT PROCESS  MODEL  PLANT

                    REGULATORY  ALTERNATIVES COSTS  (June  1980  dollars)a'b
Regulatory Alternative

INSTALLED CAPITAL COST ($)
Condenser
E6RSC (Baseline)d
Distillation Column
Total
AHNUALIZEO COST (S/yr)
Direct
Operating Labor
Operating Materials
(e.g., catalyst)
Maintenance
Materials and Labor
Utilities
Natural Gas
Electricity
Steam
Water
Subtotal
Indirect
Taxes, Insurance, &
Administration
Capital Recovery
Subtotal
Recovery Credit
Total (Direct +
Indirect - Credit)
1
1,628,400
1,628,400


22,100
3,100
12,400
131,100
664
169,364
65,100
264,940
330,040
174,000
325,400
2
1,520
1,628,400
1,629,920


22,600
3,176
12,484
131,100
665
170,025
65,161
265,190
330,351
174,328
326,047
3
1,520
1,628,400
1,629,920


22,600
3,176
12,487
131,100
665
170,028
65,161
265,190
330,351
174,342
326,037
4
1,520
1,628,400
1,629,920


22,600
3,176
12,500
131,100
665
170,041
65,161
265,190
330,351
174,350
326,041
5
1,520
1,628,400
1,629,920


22,600
3,176
12,517
131,100
665
170,058
65,161
265,190
330,351
174,359
326,050
6
1,520
1,628,400
1,629,920


22,600
3,176
12,537
131,100
665
170,078
65,161
265,190
330,351
174,361
326,068
7
1,520
1,628,400
239,300
1,869,220


37,350
5,146
12,537
177,400
665
233,100
74,733
304,100
378,860
175,950
436,000
 Costs are on a per process line basis.
b
 Some totals do not add up exactly due to rounding.

 E6RS - ethylene glycol recovery system.
d
 Ethylene glycol is recovered from the polymerization reactors  through use of
 EG spray condensers and additional  recovery equipment and from the esterifiers
 through use of reflux condensers.
                                                8-48

-------
Table  8-31a.   POLYETHYLENE  TEREPHTHALATE)  (PET)  - TPA  PROCESS MODEL

           PLANT  REGULATORY ALTERNATIVES  COSTS, LOW VISCOSITY

                OR HIGH VISCOSITY  WITH  SINGLE  END  FINISHER

                             (June 1980 dollars)9'"3


                                         	Regulatory  Alternative
                                                   1
     INSTALLED CAPITAL  COST (S)

       EGRSC (Base1ine)d
       Distillation Column

       Total

     ANNUALIZED COST (S/yr)

       Direct

        Operating Labor
        Operating Materials
           (e.g.,  catalyst)
        Maintenance
          Materials and Labor
        Utilities
          Natural Gas
          Electricity
          Steam
          Water

       Subtotal

       Indirect

        Taxes, Insurance, &
          Administration
        Capital Recovery

       Subtotal

       Recovery Credit

       Total (Direct +
        Indirect  - Credit)
1,628,400  1,628,400
            239,300

1,628,400  1,867,700
   22,100     36,850
    3,100
   12,400
  131,100
     664
   65,100
  265,400

  330,500

  174,000
  5,070
 12,400
172,300
    664
  169,400    227,300
 74,710
303,900

378,600

175,400
  325,900    430,500
      Costs are on a  per process line basis.

      Some totals do  not add up exactly due to rounding.

     "EGRS - ethylene glycol recovery system.

      Ethylene glycol is recovered from the polymerization reactors through  use of
      EG spray condensers and additional recovery equipment and  from the esterifiers
      through use of  reflux condensers.
                                    8-49

-------
   Table  8-315.   POLYETHYLENE TEREPHTHALATE)  (PET)  -  TPA PROCESS MODEL

                 PLANT  REGULATORY  ALTERNATIVES  COSTS,  HIGH VISCOSITY

                             WITH  MULTIPLE END  FINISHERS

                                (June  1980  dollars)a'b


                                                       Regulatory  Alternative
INSTALLED CAPITAL COST ($)

  EGRSC  (Baseline)d
  Distillation Column

  Total

ANNUALIZED COST (S/yr)

  Direct

    Operating Labor
    Operating Materials
      (e.g., catalyst)
    Maintenance
     Materials and Labor
    Utilities
     Natural Gas
     Electricity
     Steam
     Water
298,500   298,500   298,500
568,000   580,500   597,000
        298,500
        618,900
        298,500
        633,000
        298,500
        649,800
866,500   879,000   895,500    917,400    931,500   948,300
  1,970     1,970
1,970
 10,820    10,820    10,820
  5,340     7,170     9,720
1,970
         10,820
         13,230
1,970
         10,820
         15,550
1,970
         10,820
         18,350
         298,500
         696,300

         994,800
 14,750    14,750    14,750     14,750     14,750    14,750      14,750
1,970
          10,820
          26,440
Subtotal
Ind1 rect
Taxes, Insurance, &
Administration
Capital Recovery
Subtotal
Recovery Credit
Total (Direct t
Indirect - Credit)
32,880
34,660 '
140,980
175,640
123,390
85,130
34,710
35,160
143,010
178,170
125,750
87,130
37,260
35 ,820
145,700
181,520
128,250
90,530
40,770
36,700
149,260
185,960
130,740
95,990
43,090
37,260
151,560
188,820
131,970
99,940
45,890
37,930
154,290
192,220
133,220
104,890
53,980
39,790
161,850
201,640
135,680
119,940
 Costs are on  a per line basis.
b
 Some totals do not add up exactly due to  rounding.

 EGRS - ethylene glycol recovery  system.
d
 Ethylene glycol is recovered from the esterifiers and cooling tower using distillation columns  and from
 the initial end finishers using  spent ethylene glycol spray condensers.
                                         8-50

-------





















o:
o
Lu

CO
LU
^>
t—H
I—
*^C
a:
LU
H-
1
•a:

>;
O
t—
— I
LU
OH

LU
O
CO
•z.
o
»— 1
I —
o

cu
LU
r*>
'o
CO
CO
t— <
III
LJ_1
0
LU
g
CJ
0
CO
co
^C

a
IEZ
-
0-
o
o:
>-
— i
o
D-













































CO

•r->
+J
CO <8
> C

£ CO
<8 +-»
s- <:
CO
*-> +-»
r— C
< CO

C|w g
O -r-
S_
C 4->
O OO

P to
18 CO
4-> CO

CO
J s
*Q.=I
O

Lu











"o
S-

e
o
o
CO
c
o £
CO
^5 *o
+J
<8 E
4-> 0
C S_
CO >4_
E
co 
CL-r-
£ +->
e

co
•^
(«•
^^


























M
c
^+- O -O
O •'*• en '-O >-O LO
•(-> s: oo o CM
+-> o — ~. i M co en
CO 3 
o -a «sr
C_3 ^
as








A
C
o

O •!->>»
o u-C; o r* "S-
=» 3 CT> 1 CM CM
•0 S -H
CO
C£






T3
co
•^ -P 0 0 0
i— oo  CD kO O O
to 3 ^**^ I oo r**» CO
o -o *+
O CO
o:



c
o'2>,
O u ^«. o r*. •— i
^ 3 O> 1 CM »* LO
•OS — < •-<
CO








•o
co
N *
•r- P 000
r— tO O O O
18 O <»* P-. «-H CO
3 O 1 . « •
c r^ o cy*
!= -4 CM





>) CO
&• ^ .^"^
o -i- co
4-» -U c
(8 <8 -P-
^ C >-H r— CM CO «^-
3 S- CO
CD CO CO
CO +•> (8
O£ i— m
^ *•— •















































U)
•r»
CO
•(8
-O

CO
c
to
s
o
s_
a.
1
o

QJ
s_
18

CO
C
o
4^
U
3
•o

C—

C
O

to
to
'g
CO

•o
c
18

to
to
O
O
«S
 .-8
d ^^
•r— U 'f~ O^
 *T~ >
a  O ti_
to i- in •<—
•P O
c s= co en
0 0 v> 2:
•i— O
CO O S-
to £T 4-> CO
••- O Q.
s e to
co e co o
0 to 00
CO O 18 CM
.c cu v»
•*-* <8 &.
U O
14- O CO ••->
0 P -0
•o
co to co c
N e > 18
*^" flj •!"•
IO CO 4-> CO
*•• 4J ^ ••»
^^ ff ^ I^IB
c— CO t. i—
18 CO
E 4- +-> to
to co i— to
-£= <8 CO
CO 4J U
.C O to O
4-> -r- !_
•c: j: o.
0 ** +->
•P -i- r8
3 s-
co o en
3 "O <<_ e
Q CO •••
+j c >
o o v-
. 3 -i- CO
"O X3 4-» to
CO U
S_ >i 3 CO
S- r— T3 U
3-i- CU ••-
0 CO S. >
c 
•»- CO >*- T3
o
CO CO r—
to s-
r=- C O +•»
r— (8 U C
•r- O O
4-* 18 I— C
CO CO O
to co o
I— U U
18 CO -r-
4J -C > 18
= +J CO
co *o o
e . +->
CO tU f—
S. > 0 +->
o -^ s_ c:
c: •*-> 4-» co
•r- 18 C to
S= O
CO i- U CO
•i- CO i.

+J i— 3
 c
(8 tO to O
.C f- +-> -i-
4-> f -r- CO
+-> to
^ 2 'S
CO -i- CO .

•r- -O <8 •—
r— co -c: 18
!= r— 4J =
3 •— O
' O i_ '1- .
CO S- CO +•* •»->
•r- +-> £2 -r- C
S 4-» T3 (8
4-> O «J -0 r—
>— I O l_ <8 CX
J2
8-51

-------









Qi
O
U_

OO
UJ
£

c5
UJ
f—
	 1


^.
a:
o

_i
CD
UJ tO
O£. OO
OO
bill
0
o
oo oi
2: o_
o
1-H UJ
r- 00
o <:
=3 3=
o Q-
LU
Q; oo
«C
O
1—4 »
OO UJ
OO ZE
t-t UJ

1 • I ^M
a.
o o
UJ r*t
|— Q.

H-4 1
sz

oo


Q
§

oo
H-
to
o
o

.
oo
oo
CO

 C
•r- S-
•P 
r— C
< 0)

(JL^ C
O -r-
s-
£= -P
O OO
•r"
•P 00
+-> 
a> x
r— O>
CLZ


s»
U-














fW»
o
S-
4^
- C
O
o

f—
*| tp.
O •—
d)
£? CO
O (O
•r- CD



C S-

is


P^
^^















i
n t
C I
M- 0
O •(— O5 I-O
•p s: co
•P o "•--. i oo
CO 3 ^^
o -a
O O)
«:






A
c
o
•1- i.
0 -P >>
0 0 <. 0
> 3 0) 1 CVJ
•a s
 O
r- to 1.0
(O O 4&~ t"^
3 O • «
c r~
c
i^



„
c
«f- O
O -i-
^ O> LO
•P 0 S CO
10 3 ' — 1 OO
o -o «+
O  3 05 1 CVJ
•o s
0)
a:


•



T3
a>
N »
•r- -P 0
r— CO O
(O O *^" P*«»
3 O 1 •>
c r-«.



S- > <-~
O T-  -P C
CO 
c
• !«•
fMHI

CO
CO
CO
o
O
s_
Q.

S-
(y
a.
to

e:
o


-------





















>_
C£.
O
	 1

LU
°-
LU
O
fO
OO OO
•z. oo
O LU
i— i CJ
1— O
O O£
^3 Q-
Q
LU LU


z: oo
O 00
I— < LU
oo OS'
OO Q-
1— H
El 32
LU 
•r—
4_>
CD CO
> £T
'43 Q
CO +J
C i—
S- 
o oo
4^ co
C8 to
-P 
Cl
'
QJ
co x
*a.z.
E±
u_






"o
s_
0
o

CD

O r—
CD
C CO
O co
•i- CQ
C8 S
+•} O

Q.T-
i — i (O
CD
£
•*














4- O
O -I- O5 O O O O
•PS co en «* o
P O •••-. 1 •* <• O CM
(^ 3 •C^- A A A
o -o • «-H •* o
C_3 CD <— l
Qi






A
0
•t- S-
0 -P >>
O O ' — % LO to ^" ^O
> 3 C7) 1 <-H in <-< CM
•a S
_•*;





-o
N "
•i- -P O O O 0
i— to <&• i o o o o
to o CM co ^o cy^
^3 t ^ A « « A
C P-. CM «3 LTJ
c co to vo
«C CM


A

 o o o o
•P O S CO 1^ «* «sf
CO 3 *^«. 1 *^" CM 1^ r^»
O *^O *&" A' A A
(_3 >
o o ^^ tn o «* o
S» 3 O 1 <— < 1 	 CO r-H
T3 S rH
CD
Q£



T3
CD
N «
•i- -P O O O O
i — CO O O O O
co o <&• CM CM co r~-
3 O 1 A A A A
c: i — 01 to r-i
S= CO •* r-H


>> CD

O T- CD
•P -P C
O3 CO *r~
1 	 C r-Hi 	 CM OO «* tO
31- CD
O) CD tO
CD -P CO
OL i— CQ
^C "*-^





























^
CO
co
CO

CD
fM
to
to
CD
o
Q

CL
t
CD
CL
CO
C
o

S-
03
CO
C
o
•P
o
T3
CD
C
0
CO
to
V
•a
c

CO
•P
CO
O
O
03
8-53

-------


















to
LU
^»
H-i
1—

•z.

LU
1 —
_J


^_
C£.
P
5
__
CD
1 1 1
on
LU
O
to
o
! '
o
a
LU
C£
^
o
CO
CO
HH
s:
LU
a
LU
5
I-H
CJ
o
CO
CO
^c

a
^^
•=C

CO
CO
CO
CJ

.
IT)
CO
1
00

Q)

^
(-»
r_
^_>
CU 03
> c

•IJ 

^  eu
OJ
<1> X
CL3I
E
«-• E
o
Lu







rM»
o
^B,
o
o
CO

t+~ *r*
0 i—

C 00
o  O
C 5^
I1*"
cu cu

CLf-
i— < (O
c

cu

f—
^^
























M
c;
f- O -O
O "r- Ol LO O O O
+-> s: en o o o
+J O 	 1 CM CO O1 «*•
CO 3 fe*)- A M M
o -a CM co LO
o cu < — ' «— t
fy









A
c:
o
•f- S_
O +•» >i
co tj "^^* co r^* <^* ^*
=>- 3 o> I ro
•o s
0)
o;




•o
0)
N •
•!-•!-> O O O O
r— 00 1 O O O O
03 o fe*5- CD r^» LO LO
3 O « « « «
C 00 CD LO «-t
•a: ^

jT
'»- O
0 -r-
4-> O> LO O O O
•(-> cj ^ CD r^ ^- co
CO 3 -^ 1 CM ^* LO LO
o -o <>•> « •
O CO CM ro





A
c
o
••- s_
O •*-* >i
o o<; o p^ • t-H LO
>• 2 O} 1 CO CO ^f ^f
"O ^g"
CU
ctf





•o
cu
N •
•r- •>-> O 0 O O
i— CO O O O O
O3 O W" CD LO CM f^
30 1 ....
£= CO CO «3- LO
C CM O LO
, OJ

o ••- cu
+->•!-> C
03 03 T-
r— C -Hr— CM CO «* LO
3 &. CU
CJ1 CO 00
CU +•* O3
Oi r— CQ
r—
00
00
'r~
"
CU

•a
c
03

00
-f«J
00
o
o
03
CO
cu
jz: oj
•!-> r—
05 O
C M- -C
•r-  ^r*
E CU 01
03 T3 S- C

•f-> O i.
CO S- O CU
+J O co
c: c -H
o o t* <+-
•i— O -i—
00 O
CO C +-> 0)
§™^
CO
cu E cu s.
O 00 CO
CU  03 t. r-.
O CTl
"+- O CU W
o •<-> -o
CU CO CU C
N E > 03
•r- 03 •!-
00 CU +J CO
S- 03 C
i— +•> C T-
r— 00 S- r—
03 CU
E S- +-> to
00 CU i— 00
x: ns co
O) 4-> U
-C O 00 O
+J -i- S-
.c £: CL
o -^ •*-*
4-> t- 03
3 t.
CU O O1
3 -0 1- C
a cu f-
(J O t-
. 3 -i- , 3 
c: 03 cu
.,_ cu 4- -a
o

JZI ^J 4-> O
CO S-
r— C O 4->
f— 03 O C
•i- U O
3 CO 0
E J=

CO CU O
OS. E
O +•» o E
00  -=: > 03
C •(-> CU
CU "O O
E • -t-1
CU OJ i—
S_ > O 4->
u •!- s_ c:
C +J •(->  C 03
+J i— 2
03 O 00
4-3 C
03 OO 00 O
.£;.,_.,_>.,_
4J ^1 -i— CO
+•> 00
^cSl
CU -r- CU
xx C
•i— "O «3 r—
(_ ^ <*" (O
C i— -t-> C
3 i— O
O S_ •<- «
00 S-- CU +•> +0
•f~" +J tr* •)_ f^
c *-> -a 03
•(-> O 03 "O r—
i— i o s_ 03 a.
^
8-54

-------











Q£
0
LU

00
LU
2<
H-
OS
LU
1—

^£

>••
o;
o
I—
Ij
05
LU — 1 >-
i— or
CJ Q£
^3 ^D
0 _J
LU If)
OS
LU
Z 00
o <:
GO 0.
00
i— i C3
1 . i -^

O »—i
LU —1
1—

l-H LU
O 0-
o a
00 3C
OO
 5=
•p" ^
+•> CU
C f—
i- «*
r— C
< CO
05
<*- C
0 ••-
i.
a 4^
O CO
•^
4-> VJ
(O V)
<-» co
CO —
E *->
cu x
'o.z
£
>- S
O

1 i













"o
S-
4-3
c
o
o

cu

t- i-
O r—
CO
C V)
0 03
•^ CO

i
o tJ "^^
=» 3 O)
•o s
CU
Q£





•a
N «

r"* V)
03 OV9-
3 O
C
C
 0 S
o -a *»
O (U
Of






A
c
o
t- S-
0 «-> >i
o w *•*«.
=» 3 OJ
•o s

r»*





T3
CU
N •>
r— «/l
(8 OV»-
3 0
C
c
>i cu
s. >
o ••-
 cu
cu +->
Q£ •—
                  JD
            LO     LO
            CO     CM
            n     en
            o
            CM
co
CM
           O
           o
o
o
05
  ft,
CO
LO
           LO    O
           CO    CO
           CO    CM
           O
           CM
                  CO
           O
           O
O
O
<-o
                  vo
     0)


   I r—    CM     CO
     cu
     v>
                             CO
                             CO
                             u
                             o
                             cu
                             CO
                             s-
                             VI
                             c
                             o
                             CO
                             o
                            •f—
                             01
                  +->     CO
                  o     s-
                  3  CO  O
                  -a  s-  E
                  O  «3  i-
                  s_     cu
                  CX  +-> J=
                      C 4J

                  •—1—3
                                               o  co
                                               *->  (J

                                               +->  >
                                               C  CO
                                               co  ~o
                                    +-> 1—
                                    as  as  •
                                        O O5
                                    1- 'I- S
                                    3  CX
                                    o  >, s_
                                    O  4-> CU
                                    o     ex
                                        03
                                    o     o

                                        o cr>
                                               C  i-
                                               ns  +->
                                                  c
                                               CO  O
                                               c  u
                                o
                            O5  E
                            01  E
                            CO  O
                            o  o
                            o
                            S-  03
                            CX

                            *s
                                    -i

                                        CO 01
                                    V)     CO
                                    •i—  en 05
                                    c
                                    o
                                    T3
                                    o
                     •i-  CO

                     01  U
                     •r-  CO
                     u +•>
                     3  O
                                       ts
                                        O
                 •o
                  a>
                  1.
                  s_
                  3
                                          -o
                                        S-  CO
                                        CX &.
                                        CO  O
                                CO
                            OJ  01
                            s_
                            cu -o
           +J


       •I"  ^—
        3  CX

       •o  cu
        CO  <—
        c  o
        8  *
           c
        cu T-

          C
 C=  O
 n)  E

"a. o
    u
 
 CX
 o  -a
 i_  c:
•a  03

    CO
    u
                  «

                  4->

                  >>
          •r-    ^
           E     CU
           CU     y

          •O    i—
           c     c


           01     VJ
          *->    -r-
           01
           o     +->
          O    "-<
         0      -Q
 VJ
 £=  +->
 o  e
•^  su
+->  VJ
 u
 cu  -o
 01  C

 OJ
 c  -a
•i-  cu

 01  -r-
•f-jQ

•i-  O
t-  o
O  CU
3  -a
13
CU  r—
V.  O


o  c

^8
VJ
o  c
o  o

CO  E
-£=  O
+->  o
8-55

-------






01
O
U.

co
Ul

£
§
Ul
H-
_J


^»
o:
o

5
CD tO
ui co
o; co
Ul
C£
CO Q_
o z
l-« O
1— I-H
0 H-

a —i
ui o
C£. CO

Z Ul
o co
*•"* ^
CO £
CO OL.
H- 1
Sj i— i
^•fr
a 0-
It 1 HH

^£
1— 1 "
BUI
Q.
co a
CO 31


a


CO

CO
o
U

.
CO
i
CO

0)

f~\
to
1—



















O>

•r—
-P
O>  c
•r- S-
•P (U
CO -P
CD
t— C
< a;
0)
C|- C
O •!-
j_
cz «t^
o co
•(^
4-> co
CO CO

c
4- T-
0 i—
CD
£= to
O ,
O O "~~- O «d-
>• 3 O) 1 CM «*•
•o s
a>
D;





"S
N «
•i- +J O O
i— CO 1 O O
co O V* t-*- t-H
3 O » «
C t~- VO
d C?1
< CO




•*
c
(*- O
O •!-
+j cn irj o
•P o S co ^-t
CO 3 	 1 CO CO
f^ ^y OO ft
o o *o






A
C
o
•i- S.
B-p >> ,-,_..
o ~-» o •*
5» 3 Cn 1 CM IO
•a s
cu
Di






•a
O)
N «
•r- 4J 00
i — co O O
ro O *O- P— CO
3 O 1 " "
C P>- CO
e o
> CD
c > ^.
O -i— 
+J -P C
i — C t-l i — CM CO
3 S- CL>
en 
CL
CO

c
o

cu
L>
CO
CO
J^
o
•r""
•P
o
3

cu
S-.

c
o
co
CO
's
tU

-a
c
to
CO
•P
co
o
o
CO
8-56

-------







o^
0

00
LU
*^P
*s»
LU
1—
_1

>™
a;
o
I—

=3 O
O =D
LU z
z S
o o
/t-H O
oo
oo •>
t-H LU

LU LU
as
ca >-
UJ 1—
1— oo
^c ^™
*— t 	 j
8£
oo
oo

•r"
^^
CU 03
•i- &-
-P CU
CU
r— C
^c cu
O)
f ] f"
°£
C -P
O 00

•P V)
03 in
-P CU
E _J
cu
E -P
cU x
i— CU
O.Z
« E
0
i.
Lu













r^>
O
t
^>
c
o
0
CU
c
M- -t-
O r—
CU
c in
O 
f—
^^



















c ^a
q_ 0 •—
O -i— O> LO CM
•PS CD "3-
•P O *>>. 1 CM *d-
in 3 ^^- •*— *
O T3
O CU
oi




„
c
O LO
•t- S_ 0
O-l-^ ^* «
•»^ .^^ *
o o *-» i o ^j-
5» 3 O> OO
-a s
cu
OS .







•o
CU <-i
IS| « -—
•r- -P O O
i — in I LO 01
«s o «o- co r-.
3 C_5 «
C CO r-H
c ^<
^^


„
c:
<4- O
O •!-
4-» O) LO r~»
•PUS en o
in 3 ~--. i CM CM
o -o &>
C_3 ,
O O — O «*•
3» 3 O> 1 OO OO
"O S
cu
o:








-o
cu
N «
•r- -P O O
i— in LO to
-
c





>> CU ^

O -i- CU
-P -P C
(13 (O »r-
r— C f- 1 i — CM OO
3 S- CU
O> CU (/)
CU -P 03
ni c— eo
JD
rf— X
CM en LO o
CM [-^ CM LO
10 r-H en t— i
«M«- «\ *\
oo en








CO «3° ' ^t CM
<— i ^-< O O
0000












.a
^-^
CM LO !-»• OO
r-H CM LO . CO
.—1 ^H <— 1










OO OO t^ CM
O O O ,-H
CM CM CM CM










oo r** * — i oo
CM OO •* <*

•^ *^h *^" ^f
oo oo oo oo














CO OO O OO
«5j- r*- oo * — i
en en i— i oo
*» *» *» «
(^ (^ -I — . t*^>











•* LO IQ I~~












































tn
in
rt3
f*i

eu
c
•r—
p—

in
in
CU
o
o
CL

S-
eu
CL

OS

s=
0

cu
S.
rt3

in
c
o

•p
o
3

CU
i.

g~
o
'01
I/)
•r«
^
CU

T3

03

in
•P
in
O
O









































.

-------






rv*
O


oo
LU

P

T?f
01
LU
t—
_J

^_
01
O
«a:— "
o oo
I-H 2E
J— LU
yQ.
00
0 =>
LU OO
=Z O
O l-<
t-4 f—
oo <
oo 2:
M CD
SLU
01
d.
Q s:
Lu P!
l_ 1

>-< oo
80
0.
oo
oo •>
«C 00
Q.
a LU

^£

oo
1—
00
0
o

.
cn
CO
i
CO

QJ
(•M
.O

H*



















CD

•i—
4-*
0) <3
> C
•I™" &—
•P CO
ID -P
S- <
CU
•p -p
r- e
 LO
•PS CO
•p o •»». i co
to 3 V=t
o -a
o co
ex:








**
c
o
•i- S-
B-p >>
o -^ o
S" 3 O> 1 CM
-o s
CO







-o
CO
N «
•r- 4-> O
r— CO 1 O

£=
^^


«\
C
(4- O
O •!-
•PS- LO
•p o >> co
CO 3 ^. 1 CO
O "O ^^~
o co
a:





r>
c
o
*«•« <
O -P >>
O O "~-. O
>• 3 C35 1 CM
"^ ^*
CO
Q£







T3
CO
N •>
•r- -P O
I— CO O
as O <&• r^-
c r-.

§.



>> CO
£! > ^~.
•S'-p e
CO tQ »r—
i— C t-l r— CM
3 i- CO
CO CO CO
CO 4-> (0
Qi r— CO


LO

to
•s
r— 1














to
in











o
0
CM
CM
cn






LO
1— I
CO
I— 1











to
1 —













o
o
en
cn
cn








CO






0000
.-H 0 t-H 0
.— i cn oo CM
•\ A n A
CM CO •* t-~
CM













co co cn T-H
cn LO CM











0000
o o o o
to i — o CM
to to LO i^
cn o o
t-l CM C~





o o o o
LO P^ t— 1 CM
r- CM CD co
t-l CM «* «*•











cn CM <— i CM
tO CM LO LO
i— 1 CM CM CM



*







-
'O 0 0 0
o o o o
LO CM CM «3-
to co co LO
en o o .-H
CM LO CM CM

t-H T— 1





•3- LO tO 1 —











































.
•r-
CO
(O
.0
CO
r""
CO
CO
CO
o
o
s_
a.

i.
co
a.

to

c
o

CO
^
to
CO
c
o
•r™
4^
O
13
•a
CO
S-
c
o
•r—
CO
CO
•i—
co

-a
to

• CO
•p
CO
O
O
to
8-58

-------
1
t
1
II
li







>_
C£.
0
1 —
_J
•=3
CD
UJ
oo
b_ oo
o LU
c_>
00 0
•Z Qi
o a. ;
1 — "Z. '
o o
1^ 1 — t
a oo
LU Z
01 LU
a.
2: oo
0 =>
t-H OO
oo
00 => '
r-H |—
LU 00
1
C2 2
LU l-H
| —
l-H OO
o a.
O LU
oo
OO C£.
< 0
u_
a
=: oo
< LU

oo >— <
I — i —
oo S:
o a;
LU
i—
d5l
t
co

cu

-O

•~



















cu
^_>
cu re
> c
•i- S-
•P CU
03 4->
S- <
CU
•P -P
5 cu
01
O •!-
C 4->
o oo
4J CO
(O to
•P CU
c *
CD
CU X
r— CU
O.Z

LL.








•^
S-
•P
o
o

cu
c

O r^
= 35
! o re
'43 ^
C S-
C^ *t—

cu cu

"Q-T-
E -P
I-H a)
£1
s_


r—










C
(i Q
*1— ^rf1
O T- CT)
P O ^^
O "O
c_> cu
o;








<=
o
0 -P >>
o u -•-.
S- 3 01
-o s:
cu







•a
cu
IM •>
r— tO <»">•
(T3 O
3 O
C
c
• 3 CT>
•a s
cu
Di







•a
cu
M "
i — to
03 O *0-
3 U
C
,cu
O *T—
•p 4-^
(TJ5 fO
"3 s_
01 cu
CU 4->
oi i—
 T-H O
CM








o o o o o o
*i- o to o o o
CM I-~ T-I T-H l-» LO |
tO CM 1 — •* CM CM
CM r*^ to CM CM







O LT) O O O O
CO LO tO LO P~- CM
o*> co r^- co CM r"*~
,_, CM CO •* •*



*




r- to ** CM CM to
tO LO CO •* "-0 lj°











o o o o o o
•* o o o o o
CM CT> <—l CM C31 >*
to co to o CM "^
CM o r~- en T-H
r-l i-H I—I CM


cu
c
T-H^-cMCO'^-i-otor-.
cu
1 CO
i ^
































in

-------







r>s
O
Lu

CO
LU

I — I
1—
ae
LU

_J

^-


H-

•r—
•P
cu re
> c
•r- S-
•P cu
re -P
C i—
5- <
•P -P
t~~ * C

CL'i-
e -p
i-t re
g*
^
CD
-P

^^



















t\
C /"^
q_ O -—
O *r— Ol tO C^ tO CTl CO LO
+->S tOcMrOCMOI~~
•PO^~ CM- — • r-HCOOl
CO 3 tO- 1 •> «
o -o ^-i cn
C_> <1) CO
C£








cT
O CO LO r-t 1^ ,-H LT)
•t-s- «* o i— i o o r~-
B-p >, 1 ......
O 	 CMOOOOCM
S> 3 CO
•o s
cu
a:




-a
cu ^
N « — »
•r- -P (--•O'd-cricoo
f— CO ^" ^H t— i CO
re o *&• i to — - o">
3 O «
C CTt
c o
•=C r-H

„
C
<»- O
O -i-
•pco tor-i — «* o o
Pos voLn«d-^-ir>o
CO3' 	 CMCMCMCMCM-*
o -o v> i
o cu o
OZ CM


~


««
c:
O CO CO CTl tO I^- CM

CJ -P >, 	
OO^^ CMCMCMCMCM^n
>• 3 CO
•o s: i

f>^








T3
CU
N "
•i-+-> r— r-- ,— i o co o
r— CO "d"CO<=d-LniOO
re o <*o- to to to to to to
3 O «
c: i o
c: <— i
 —
o -i- o;
•P -P £=
re re «r-
l^ C i j— rvi rv} ^H- LO tO f^
3 S- CU
CO CU CO
oj +j re
Di i — CO








































co
CO
re
.n

cu
c
•r—
^_

CO
CO
cu
o
o
o.

i~
CD
Q-
re

c
o

cu
S-
re

co
c
o
•r-*
| *
O

•a
cu


c
o
•1—
CO
CO
•1—
i
•o
c
re

co
-p
CO
o
t ^
re


































•
CO
•P
EI
0
re
j_
re

P— .
o
-o

a>

•i—
P
re
a>
cu

s-
0

CO
p
-5
cu
s_
0

4-^
C
CU
CO
0)
s-
a.
cu
s-

co
cu
CO
cu

-p
c
cu
s_
re
o.
.=

co
s_
cu

ps
3
•z.
£1
8-60

-------






oi
o


OO 03
UJ Qi
> UJ
i— l 3C
H- OO
^Z
^C -tfi
o£ i — i
UJ U_

_I Q
UJ

Qi UJ
0 _1
1— CD
< -z
	 1 i— i
•=> oo
CD

as 1—
i — i
U- 3
O

OO 1—
~Z i — i
O 00
H-< 0
1— O
O OO
^D t—t
>""** ^>
UJ
a: z
CD
:z i— i
o a:

00 Q
oo -z
t — i ^

1 I 1 A
>»
?-*, h—
UJ i— l
1— 00
< O
t-i O
CJ OO
O i— i
oo >•
00
**" o
a _i
•z
00
oo oo
1— UJ
oo o
0 0
0 Qi
a.

• «^
re ^
CM J—

co (i i
a.
cu
o
03
| —





















CU
>
•r-
^^
cu  c
•I— i_
-p co
03 -P
C i —
i- <
•P -P
,_— c
< 0)
01
(4— d
0 •^-
S-
c: -P
o oo
•r—
•P OO
03 
CO X
i— CO
ft ~z*
E
1-1 S
o
i_
U-














^^
o
S-
•p
c
o
o

CO

M— "r-
O r—
CO
sr to
O 03
•r- 03
-p

-P O
c s_
CU M-
E
CO CO
I— >

E -P"
l — I 03
s_
CO
+J
^^ '


















m
C 0
4- O O
O T- CT) t-O
•p s: i
-P 0 --^ 00
to 3 W ^1-
O T3
O CO
a:








•>
c
0
0 -P >> «* .
o o ---. i
> 3 Ol CM
T5 21
cu
f*v^








T3
CU
N O
•i- -P 1 O
i — If) tO
03 O •fc^' "
3 O ^t*
C O
C i— 1
e£




o
C
M- O O
O •!- O
-P CD VO
^ o s: i
to 3 ~-- OO
o -o to- «d-
c ^ CO
Di






cf
O ^3"
•r- S_
0 -P >, CM
O 0 --- 1
>• 3 CO
-a s:
CO
Qi







-a
cu
N - O
•,- +-> 0
i — 1/5 U3
03 O <^S- 1
3 C_3 ' «d-
!= 0
C i — 1



>j CU

O -i- CU
•P -P c
03 O3 *r—
1 	 C r-4 \ 	 CM
3 i- CU
CO CU I/I
CO -P 03
Di r— 02




























^

















•
to
•r-
to
03
_n

CO
c.
•r—
r—

to
to
CO
o
o
&_
a.
s_
cu
Q.
03

C
O

CO
S-
03

to
c:
O
'^J
0
3
•a
CO
S-

c:
o
to
e
co

-a
c~
03

to
-P
to
O
O
03
8-61

-------




^
£

oo
LU
~^
*~*
i£
z
LU
|— • fl3
_1 O"
^C l*v*
u.
>- a:
PS

^t *~~
—1 U_
:=>
CD 0
LU Z
Q£ LL
U- LU
Q-
00 HH
0 !~
1— £
O
=3 in
0 1—
LU HH
z >-
0 h-
1— 1 I— 4
oo oo
oo o
>-! 0
LU 1— I
^»
Q
gl
S3!
oo oo
oo oo
eC LU
08
Z Qi
<: Q.
co <:
1— Q.
00 1—
o ^
0 H-
LU
Q-
rv
CM
1
CO
CU
^™
n3


















CU
•M
CU  C
•r- S-
4-> CU
(O +J
i- <
cu
Jf "c
•a: co
en
t|-n £*
o -^
o oo
•r*
•P 10

(O H
•»-> 0
C S-

£z
tu cu
"o.'!-
E •»->
HH (O
s_
cu
4-5
r~
"*














c"
4- 0
O •r— C7>
•(-•S i cor-.cocoor-~
•*-* o "•>» r~- >^ f— i to o o
CO 3 *^S- CM «* 1 — O OO O
O TJ n «
0^ rH r-1 CM







1
f—
0
•r- S_
cj-P>, cMio«5r--cOLr>
s»3cr> i — i — r--oooor~
"O 3£
cu
a:






•a
cu
N « OOOOOO
•r- +> | OOtOLOLOLO
r"~ CO CO ^^ ^f O"i CT^ CO
3CJ CMOOLOOO-si-Un
C *~^
<

C
 O> CM OO «J- LO VO CTl
CO 3 -~~.
o -a -&0-
o cu

A
C
.O^ CMr~-OO^HCOOO
o-»->>, i — •3-cMiocnr--
gO^I r-HCMCMCMOO
"O S
CU
C£






T3
cu
N « OOOOOO
-1~ +J OO^O^HtO,-!
i — co O-^-cOCOi — CO
= <-> CM LO O «3- CD <*
t— 1 r-H r— 4 CO
C


>)  ^-~.
O M- CU

f(3 (O *r—
3 E -"; ~ °° * w « ^
CD CD CO
CU 4J (0
Oi i— CQ






1
1
1
1
























»
CO
nj
.a
cu
s
**"*
CO
CO
cu
o
o
a.
cu
o.
ITS

o

t
(0

CO
o
•M
o
3
T3
cu
s-
c:
o
CO
CO
•1—
i
__
C
(t3

V)
to
Q
o
8-62

-------
      Table 8-43.  TOTAL FIFTH-YEAR NET ANNUALIZED  COST OF  PROCESS
         AND FUGITIVE EMISSION CONTROLS FOR POLYMERS  AND RESINS
                      FACILITIES AFFECTED BY NSPS


Annual i zed
Regulatory Costs Per
Polymer/ Alternative Process Line3
Process Number $/yr
PP
Liquid
Gas
LPDE
High Pressure
Low Pressure
HOPE
Slurry
Solution
Gas
PS
Continuous
Post-Impregnation
Suspension Process
In-situ Suspension
Process
PET
DMT
TPAb
TPAC
TOTAL
4
2

5
5

3
3
5

7
7
7

7
2
7

129,800
7,700

411,700
165,700

61,600
403,800
165,700

7,310
1,215,300
215,400

110,600
104,600
34,800

Number
of
Total
Nationwide
Fifth-Year
Net
Annual i zed
New Process
Process and Fugitive
Lines Costs, $/yr
9
9

4
10

6
9
10

2
2
3

7
13
1
85
($)
1,168,200
69,300

1,646,800
1,657,000

369,600
3,634,200
1,657,000

14,620
2,430,600
646,200

774,200
1,359,800
34,800
15,462,000
 From  baseline  annualized  costs.

 Low viscosity  PET  line  or high viscosity PET line with a single end finisher.

'High  viscosity PET line with multiple  end finishers.'
                                 8-63

-------
 8.2  OTHER COST CONSIDERATIONS
 8.2.1   Hater  Pollution  Control  Regulations
      8.2.1.1   Federal Water  Pollution  Control Act  (FWPCA).   Polymers and
 resins  industry (a  subcategory  of  SIC  2821) facilities are required by
 the FWPCA to  comply with effluent  limitation guidelines.  Under the
 guidelines, existing  sources must  apply the best practical control
 technologies  available  (BPCTA)  and new sources must apply best available
 demonstrated  control  technology (BADCT).
      The  Clean Water  Act of 1977 amended the FWPCA and required that the
 best available technology economically achievable  (BATEA) be implemented
 by  1984 for nonconventional and toxic  pollutants.  For conventional
 pollutants, best conventional technology (BCT) is required.  The development
 of  BATEA  and  BCT guidelines take into  account different cost considerations.
      EPA  has  developed  water quality criteria documents for 64 toxic
 water pollutants.   These documents contain recommended maximum permissible
 pollutant concentrations for the protection of aquatic organisms, human
 health, and some recreational activities.  These documents do not consider
 treatment technology, costs, or other  feasibility factors.
     The  National Pollution Discharge  Elimination System (NPDES) authorizes
 States to  issue discharge permits.  Approximately 85 percent of the
 chemical  products industry (SIC 28) is in compliance with the Federal
 water pollution reporting regulations  required under NPDES.
     The  capital cost to the plastics  and synthetics industry of water
 pollution  control totalled $308 million from 1972 through 1977 (updated
 to  second  quarter 1980  dollars  from second quarter 1977 dollars using
 the  fixed  nonresidential investment part of implicit price deflator of
 the  gross  national  product).  The  cumulative capital  costs from 1977
 through 1986 were projected to be  $470 million.   The total  annualized
 costs for  1972 - 1977 was $117 million and the projection for
 1977 - 1986 was $850 mi 11 ion.57
     8.2.1.2  Safe Drinking Water Act  (SDWA).   The Safe Drinking Water
 Act  requires EPA to establish primary and secondary drinking  water
 standards.  Primary regulations are aimed at protecting public  health.
 They establish maximum allowable contaiminant  levels  in drinking water
and provide for water supply system operation.   Secondary regulations
                                    8-64

-------
are designed to protect public welfare and to control the taste, odor,
and appearance of drinking water.  The Act also controls underground
injection through permitting.  In establishing maximum control  levels
(MCL), the technological and economic feasibility is considered as well
as the health effects.  Currently, the MCL for VOC in groundwater is
being developed; therefore, control  costs are unknown.  Since there are
very few MCLs at this time, States have the option of controlling toxic
pollutants when a MCL does not exist.
8.2.2  Occupational  Safety and Health Regulations
     The Occupational Safety and Health Administration (OSHA) is
responsible for protecting workers against hazardous materials found in
the work place.  There are two types of OSHA regulations affecting the
organic chemical industry.  The first type requires general  work practice
and engineering controls for hazardous substances.  If engineering
controls and work practice standards are not capable of achieving full
compliance, protective equipment is  to be used.
     A second type of OSHA regulation, for more significant hazardous
air pollutants, involves comprehensive requirements for administrative
practices and engineering controls specific to a particular pollutant.
     The average cost of OSHA regulations on the entire chemical industry
is estimated to be $208.40 per worker per year.  The type of worker
protection is dependent on the chemical  produced at each distillation
facility.  In those facilities where only general  controls are required,
the costs would vary with the control method(s) employed by each facility.
     OSHA also has specific regulations, under Section 29 CFR 1910.106,
for chemical  facilities which handle, store, or use flammable and combustible
liquids with a flash point less than 90°C (200°F).  OSHA develops these
standards for toxicity levels and not based upon cost criteria.
8.2.3  Toxic Substance Control Regulations
     Toxic Substance Control Act (TSCA)  requirements are based on the
need to provide necessary information concerning the toxicity of new and
existing chemicals.   In order to develop a chemical inventory,  TSCA
requires reporting of the manufacturing, importing, or processing of any
chemical substance used for a commercial purpose.   Any substance not on
the inventory will be considered new and premanufacture notice and
testing will  be required.  Reporting and premanufacture notification
                                    8-65

-------
 (PMN)  requirements  include:   (1) the cost of using screening and testing
 to gain appropriate  information for new chemicals, (2) the cost of
 testing existing chemicals, and (3) the cost of the delay caused by the
 testing/reporting process.  PMN could have a significant impact on the
 entire chemical industry, with cost estimates ranging from $78.5 million
 to $2 billion.
     Small companies will probably suffer more than the larger firms
 since small firms have minimal access to the information necessary to
 develop a  PMN.  The  impact of PMN requirements also will be greater to
 the small  firms because the cost per unit product will be higher for low
 volume, low revenue chemicals.  The cost of preparing notices for new
 chemicals  is estimated to be between $820 and $7400 per chemical.
     EPA has been concentrating its efforts on new chemicals being
 developed  rather than on existing chemicals; therefore, the actual  cost
 of meeting TSCA for the polymers and resins industry is unknown.
 8.2.4  Solid and Hazardous Waste Regulations
     8.2.4.1  Resource Conservation and Recovery Act (RCRA).  RCRA
 establishes a national program to improve solid waste management including
 the control of hazardous waste, the promotion of resource conservation
 and recovery, and the establishment of a solid waste disposal  program.
     The hazardous waste program regulates wastes from generation to
 disposal ("cradle to grave") requiring EPA to produce standards for
 generators, transporters, and those who transport, store, and dispose
 (TSD facilities).  The wastes are identified and listed by industry.   At
 the time of generation, a manifest is issued to record the movement of
 the wastes from cradle to grave.
     The management of nonhazardous wastes is essentially a State and
 local  function implemented under State and regional  solid waste plans.
     As the cost of handling wastes increases,  some firms will  reduce
 their costs by changing their process to eliminate wastes or by recycling
 or reclaiming the waste.  New plant and equipment expenditures  for  solid
waste control  were $42-$45 million  for the entire chemical  industry in
 both 1978 and 1979.   The annual  cost imposed by RCRA on 45 organic
chemical plants generating hazardous wastes is  estimated to be  $10.9  million
 or an average annual cost of $240,000 per plant.   These estimates are
based on model  plants.

                                    8-66

-------
     8.2.4.2  Superfund.  The Comprehensive Environmental Response,
Compensation, and Liability Act, or Superfund, regulates the cleanup of
hazardous waste dumpsites and chemical spills.  Superfund provides
adequate funding, liability, standards, and authority to the government
to recover costs from the responsible parties.  Any person in charge of
a facility is required to report any "release" of a specified quantity
of hazardous waste into the environment immediately.  The emphasis of
the regulation is to report the release of the wastes and to clean them
up first and then to recover costs later.  The Act also authorizes a tax
on all hazardous wastes received at a disposal facility that is to be
deposited in a trust fund for use after the facility closes.  The fee to
the chemical industry for this trust fund is less than 2 percent of
their profits.
8.2.5 Clean Air Act
      There are only three potential Federal air pollution standards that
might impose costs on the facilities that will be impacted by this potential
polymers and resins NSPS.  All three are based on Section 111 of the
Clean Air Act, and have been initiated by EPA since the Act's Amendments
of 1977.  First, of course, is this polymers and resins NSPS.  Second is
the fugitive emission NSPS for the synthetic organic chemical manufacturing
industry (SOCMI).  It will affect only PET/DMT facilities, because all
other fugitive emissions from polymer and resin facilities will  be covered
by this polymer and resin NSPS.  Third is the volatile organic liquid (VOL)
storage NSPS.  It will affect only polystyrene and PET/DMT facilities.  Of
these standards, only the SOCMI fugitive emissions NSPS has been promulgated.
     The SOCMI fugitive emission NSPS will apply to PET/DMT because
methanol is produced, and methanol is a chemical the production of which is
expressly covered by the fugitive emission NSPS.  (All other chemical
products produced by the polymerization facilities of concern are not
covered by the fugitive emission NSPS.) However, any costs incurred in
compliance with the fugitive emission NSPS will  be substantially offset
by credits for the value of organics saved.  Thus, no significant costs
will accrue to PET/DMT facilities due to the fugitive emission NSPS.
     The VOL storage NSPS will apply to the storage of inputs to polystyrene
production facilities.  In situations where styrene is produced on site,
the costs of complying with the VOL storage NSPS will  be shared between
                                    8-67

-------
the styrene and the polystyrene units.  The VOL storage NSPS also will apply
to the storage of methanol at PET/DMT facilities.  In all other cases the
inputs and outputs from polymerization, facilities have vapor pressures
under 3.4 kPa, and therefore are not covered, or over 76.6 kPa, and
therefore would routinely be stored in pressure vessels meeting the VOL
storage NSPS.  In any event, costs incurred in compliance with the VOL
storage NSPS will  be substantially offset by credits for the value of
inputs and outputs saved.  Thus, no significant costs will accrue to
polmerization facilities due to the VOL storage NSPS.
                                    8-68

-------
 8.3  REFERENCES  FOR CHAPTER 8

 1.   Memo  from Siebert, P., Pacific Environmental Services, Inc. (PES),
     to  Polymers & Resins NSPS Project File.  September 8, 1982.  Selection
     of  SOCMI Fugitive Analysis Model Plant B to represent fugitive
     emissions characteristics of polymers and resins plants.
     Docket  Reference Number II-B-44.*

 2.   U.S.  Environmental Protection Agency.  VOC  Fugitive Emissions in
     Synthetic Organic Chemical Manufacturing Industry - Background
     Information for Proposed Standards.  Research Triangle Park,
     N.C.  Publication No. EPA-450/3-80-033a.  November 1980.
     Docket  Reference Number II-A-16.*
 3.
 4,

 5,
 8.
 9.
10.
11.
Kalcevic, V. Control  Device Evaluation:  Flares  and  the Use of
Emissions as Fuels.  In:  Organic  Chemical  Manufacturing Volume 4:
Combustion Control  Devices.  U.S.  Environmental  Protection Agency.
Research Triangle Park, N.C.  Publication  No. EPA-450/3-80-026.
December 1980.  Docket Reference  Number  II-A-18.*

Reference 3, p. IV-4.

Memo from Sarausa,  A.I.,  Energy and  Environmental Analysis,  Inc.
(EEA), to Polymers  and Resins File.   May 12,  1982.   Flare costing
program (FLACOS).  Docket Reference  Number II-B-39.*
     Telecon.  Siebert,  Paul,  PES with Straitz, John III
     Oil  Burner Company,  Inc.  (NAO).  November 4, 1982.
     of Flare  Cost  Data and  Flaring of High Air-Content
     Docket Reference Number  II-E-59.*
                                                  ,  National Air
                                                    Availability
                                                  Streams.
Telecon.  Seibert Paul, PES,  with  Keller,  Mike,  John  Zink,  Co.
August 13, 1982.  Clarification of comments  on draft  polymers and
resins CTG document.  Docket  Reference Number  II-E-18.*

Telecon. Siebert, Paul, PES with Fowler,  Ed, NAO.   November 12,
1982.  Flare Design and Operating  Parameters.   Docket Reference
Number II-E-60.*'

Telecon. Siebert, Paul, PES with Fowler,  Ed, NAO.   November 5,
1982.  Purchase Costs and Design and Operating Criteria  for
Steam-assisted, Elevated Flares.  Docket  Reference Number II-E-58.*

Telecon. Siebert, Paul, PES with Fowler,  Ed, NAO.   November 15,
1982.  Additional Flare Cost  Estimates and Flare Design  Criteria
and  Procedures.  Docket Reference  Number  II-E-61.*

Telecon. Siebert, Paul, PES,  with  Knock,  Cor,  NAO. May 2, 1983.   Steam
requirements for intermittent flares.  Docket  Reference  Number  II-E-68.*
12.  Telecon. Siebert, Paul, PES,  with  Keller,  Mike,  John Zink, Co. May 12.
     1983.  Steam- and air-assisted  intermittent  flare  guidelines.  Docket
     Reference Number II-E-69.*
                                      8-69

-------
13.  Memo from Senyk, David, EEA, to EB/S Files.  September 17, 1981.
     Piping and compressor cost and annualized cost parameters used in
     the determination of compliance costs for the EB/S industry.  Docket
     Reference Number II-B-33.*

14.  Perry, R.H. and C.H. Chilton, eds. Chemical Engineers' Handbook,
     fifth edition.  New York, McGraw-Hill Book company. 1973. p. 5-31.
     Docket Reference Number II-I-16.*

15.  Chontos, L.W. Find Economic Pipe Diameter via Improved Formula.
     Chemical Engineering.  87(12):139-142.  June 16, 1980.
     Docket Reference Number II-I-59.*
16.


17.



18.



19.



20.




21.



22.




23.
     Memo from Desai, Tarun, EEA, to EB/S Files.  March 16, 1982.
     Procedure to estimate piping costs.  Docket Reference Number II-B-37.*

     Memo from Kawecki, Tom, EEA, to SOCMI Distillation File.  November 13,
     1981.  Distillation pipeline costing model documentation.
     Docket Reference Number II-B-36.*
     Richardson Engineering Services.
     Estimating Standards, 1980-1981.
     Number II-I-52.*
Process Plant Construction Cost
1980.  Docket Reference
     Memo from P. Siebert, PES, to Polymer and Resin File.  March 16, 1983.
     Distillation NSPS Pipeline Costing Computer Program (DMPIPE), 1981.
     Docket Reference Number II-B-66.*

     Neveril, R.B.  Capital and Operating Costs of Selected Air Pollution
     Control Systems.  U.S. Environmental Protection Agency, Research
     Triangle Park, N.C.  Publication No. EPA-450/5-80-002.  December 1978.
     Docket Reference Number II-A-7.*

     Memo from Mascone, D.C., EPA, to Farmer, J.R., EPA.   June 11, 1980.
     Thermal incinerator performance for NSPS.  Docket Reference Number
     II-B-4.*

     Air Oxidation Processes in Synthetic Organic Chemical  Manufacturing
     Industry - Background Information for Proposed Standards.   U.S.
     Environmental Protection Agency, Research Triangle Park,  N.C.
     Draft EIS.  August 1981.  p.  8-4.  Docket Reference  Number II-A-26.*
     Blackburn, J.W. Control Device Evaluation:  Thermal  Oxidation.   In:
     Chemical Manufacturing Volume 4:   Combustion Control  Devices.   U.S.
     Environmental  Protection Agency,  Research Triangle  Park,  N.C.
     Publication No. EPA-450/3-80-026, December 1980.  p.  1-1.
     Docket Reference Number II-A-18.*

24.  Reference 14,  p. 8-9.

25.  Steam: Its Generation and Use.  New York, Babcock & Wilcox Company,
     1975.  p. 6-10.  Docket Reference Number II-I-20.*
                                    8-70

-------
26.  Memo from P. Siebert, PES,  to Polymers  and  Resin  File,  March  16, 1983.
     Distillation NSPS Thermal  Incinerator  Costing  Computer  Program
     (DSINCIN).  May 1981.  p.  2.   Docket Reference Number II-B-67.*

27.  Reference 22, p. 8-13.

28.  Reference 23, pp. V-3 and  V-15.

29.  Reference 23, p. III-8.

30.  Reference 23, Fig. A-l,  p.  A-3

31.  Reference 22, p. 8-9.

32.  Reference 26, p. 4.

33.  Reference 23, p. 1-2.

34.  Reference 22, pp. G-3 and  G-4.

35.  Reference 23, p. V-18.

36.  Telecon. Katari, Vishnu, Pacific Environmental Services,  Inc. with
     Tucker, Larry, Met-Pro Systems Division.   October 19, 1982.   Catalytic
     incinerator system cost estimates.  Docket Reference Number  II-E-41.*

37.  Telecon.  Katari, Vishnu,  Pacific Environmental Services, Inc.,  with
     Kroehling, John, DuPont, Torvex Catalytic Reactor Company.
     October 19, 1982.  Costs for typical size catalytic incinerators.
     Docket Reference Number II-E-40.*

38.  Letter from Kroehling, John, DuPont, Torvex Catalytic Reactor
     Company, to Katari, V., PES.  October 19, 1982.  Catalytic  incinerator
     system cost estimates.  Docket Reference  Number II-D-66.*

39.  Key, J.A. Control Device Evaluation: Catalytic Oxidation.  In:  Chemical
     Manufacturing Volume 4: Combustion Control  Devices.  U.S. Environmental
     Protection Agency, Research Triangle Park, N.C. Publication  No.  EPA-
     450/3-80-026.   December 1980.  Docket Reference Number  II-A-18.*

40.  Telecon. Siebert, Paul, Pacific Environmental  Services, Inc., with
     Kenson, Robert,  Met-Pro Corporation, Systems Division.   July 22, 1983.
     Miminum size catalytic incinerator units.  Docket Reference  Number II-
     E-73.*

41.  Reference 14, p. 3-59.

42.  Reference 14, pp. 10-25 through 10-28.

43.  Reference 14, pp. 10-12 through 10-15.

44.  Ludwig,  E.E.  Applied Process Design for Chemical and Petrochemical
     Plants, Volume  3.  Houston, Gulf Publishing Company.  1965.   p.  80.
     Docket Reference Number II-B-93, Attachment D.*

                                 8-71

-------
45.   Reference 44, pp. 100 and 104.

46.   Reference 44, pp. 100 through 106.
47.  Telecon.  Meardon, Ken, Pacific Environmental Services, Inc., with
     Mahan, Randy, Brown Finntube Company.  July 30, 1984.  Cost estimates
     for various size condensers (4.6 ft2 up to 34 ft2).  Docket Reference
     Number II-E-90.

48.  Telecon.  Meardon, Ken, Pacific Environmental Services, Inc., with
     Kurtz, Ned, American Standard Heat Transfer Division.  July 30,  1984.
     Cost estimates for various size condensers.  Docket Reference
     Number II-E-91.

49.  Memo from K. Meardon, PES to Polymer Manufacturing NSPS File,
     August 17, 1984.  Refrigeration Units for Condensers.  Docket
     Reference Number II-B-88.*

50.  Reference 14, pp. 11-1 through 11-18.

51.  Reference 14, pp. 3-71, 3-126, 3-214, and 12-46 through 12-48.

52.  Weast, R.C., ed. Handbook of Chemistry and Physics, fifty-third  edition.
     Cleveland, The Chemical Rubber Company. 1972. p. F-36.   Docket Reference
     Number II-I-122.*
53.
54.
55.
56
57.
Thermophysical Properties of Refrigerants.  New York,  American
Society of Heating, Refrigerating and Air-conditioning Engineers,
Inc.  1976.  pp. 9 through 11, 105, and 106.  Docket Reference
Number II-B-93, Attachments AF and AG.*

Memo.  Dimmick, F., EPA:SDB, to Wyatt, S., EPA:SDB.   September 4,  1980.
Minutes of meeting between EPA and Texas Chemical  Council  representatives
about TCC comments on recommended NSPS for fugitive  VOC emissions  in
SOCMI.   Docket Reference Number II-E-4.*

U.S. Environmental Protection Agency.  Fugitive Emission Sources of
Organic Compounds -- Additional Information on Emissions,  Emission
Reductions, and Costs.  Office of Air Quality Planning and Standards.
Research Triangle Park, North Carolina.  Publication No. EPA-450/3-82-010.
April 1982.  Docket Reference Number II-A-32.*

Erikson, D.G. and V. Kalcevic.  Fugitive Emissions.  Report 2
In:  Organic Chemical Manufacturing Volume 3:   Storage,  Fugitive,
and Secondary Sources.  U.S. Environmental  Protection  Agency,  Research
Triangle Park, N.C.  EPA-450/3-80-025.  December 1980.
Docket Reference Number II-A-37.*
The Cost of Clean Air and Water:  Report to  Congress.   U.S.
Environmental  Protection Agency,  Washington,  D.C.   August  1979.  p.
Docket Reference Number II-A-9.*
                                                                       27,
*References can be located in Docket Number A-82-19  at  the  U.S.
 Environmental  Protection Agency Library,  Waterside  Mall, Washington, D.C.

                                    8-72

-------
                            9.0 ECONOMIC IMPACT
9.1  INDUSTRY CHARACTERIZATION
     The five polymers and resins segments selected for potential  NSPS
development account for about 75 percent of the current total  estimated
VOC process emissions from 16 major polymers and resins manufacturing
operations.  The five are:
     1.  Polypropylene (PP),
     2.  High Density Polyethylene (HOPE),
     3.  Low Density Polyethylene (LDPE),
     4.  Polystyrene (PS), and
     5.  Poly(ethylene Terephthalate) (PET) or Polyester Resin
9.1.1  Industry Structure
     Polymers are chemically prepared through polymerization,  a process
that converts monomers or intermediate materials obtained from the
synthetic organic chemical manufacturing industry (SOCMI) into polymer
products.  Such products include plastic materials, synthetic  resins,
synthetic rubbers, and synthetic organic fibers. It is for this
polymerization process with its large volume emission of volatile  organic
compounds (VOC) that the potential NSPS is being developed.
     LDPE, HOPE, and PS—polymers used in plastics—have characteristics
similar to those of PET, a polymer used in fibers, bottles and film.
Because of data limitations, only data for PET fiber are used  in this
analysis; however, all PET production is being regulated.  PP,
additionally, is a polymer used both as a fiber and as a plastics
substance.
     9.1.1.1  Industries.  The fibers portion of the industry, producing
both cellulosic (rayon, acetate) and noncellulosic (polyester) fibers,
is composed of relatively few plants, and those that produce
                                    9-1

-------
noncellulosic fibers often carry out both the polymerization process and
the process of converting the polymers into noncellulosic fibers in the
same plant.  The plastic materials portion of the industry is much larger
and more diverse than the fibers portion and includes two SIC industries.
     In the polymers and resins industry, products are typically
manufactured in four stages (Figure 9-1):  (1) polymerization, (2)
compounding, (3) processing, and (4) and fabricating and finishing.  As
indicated above, polymerization is the initial stage in which monomers
are converted to polymers and resins, and again as indicated above, it
is this stage that is the primary focus of this analysis.  This stage
involves the five process steps described in Chapter 3.
     After polymerization, the polymers and resins are usually combined,
in the second stage, with various compounding materials, frequently
coloring agents, plasticizers for flexibility, fillers (inert mineral
powders) for firmness and rigidity, and to decrease production costs,
reinforcing agents for product strength and abrasive resistance.  Other
additives may also be used to promote product flame retardance and curing.
For some polymers and resins, the compounding stage is omitted.
     The third  stage in  polymers and resins manufacturing is referred to
as processing.  This stage involves the molding, casting, calendaring
(pressing), or  extruding of the polymers and  resins and  their compounding
additives  into  films, sheets, fiber, or rigid plastics.  Such product
reinforcement materials  as fiberglass or various synthetic fibers may also
be added during this stage.
     The fourth and final stage in  polymers and  resins products
manufacturing is that of fabricating and finishing.  At  this stage, the
sheets, rods, tubes, and special  shapes are finished or  fabricated  into
final  products  such as packing materials,  containers,-  housings, pipes,
and toys.   After this stage,  the  products  are shipped  to the major
end-use markets—construction, packaging,  transportation, electronics,
                                                   •
appliances,  textiles, furniture,  and housewares.
     Although the  potential NSPS  development  is  germane  only to the
polymerization  stage of  production, the  other stages are discussed  in
this report because of  their  interrelationships.   Polymerization  takes
place  in plants of varying characteristics.   In  some,  only the
                                     9-2

-------
polymerization stage is completed.  In others, however, polymerization and
one or all of the three additional stages may constitute plant operations.
Of all polymers and resins produced, approximately 20 percent is processed
(stage three) in the same plants in which the polymers and resins are
produced. Thirty percent is processed by about 5,600 independent custom
and proprietary processors and fabricators and finishers.  The remaining
50 percent is processed and fabricated at 9,100 plants that additionally
manufacture such nonplastics products as automobiles, appliances,
textiles, and housewares.
     The plants that produce the five polymers and resins selected for
the potential NSPS development are predominantly classified in three SIC
Industries:
     SIC 2821, Plastics materials and resins
     SIC 2824, Organic fibers, noncellulosic
     SIC 3079, Miscellaneous plastics products
     Industry 2821 comprises those plants that produce polymers and
resins for sale or shipment to other plants.  These plants limit
production to stage 1.   Plants combining stage 1 with any other stage
are classified in either SIC 3079 or SIC 2824.  Industry 3079, comprised
of the various types of processors, includes the largest number of plants
in the plastics industry.  Industry 2824 includes those PET plants
producing polyester fibers.  PET plants that produce polyester for film
and other nonfiber materials are classified in either SIC 2821 or SIC
3079.
     9.1.1.1.1  Ownership.   Table 9-1 lists the producers of the five
considered polymers and resins and indicates the number of U.S.  plants
owned by each producer.  About a fourth of the plants producing the five
polymers and resins are subsidiaries of large petroleum manufacturers.
Exxon with total  sales  in excess of $100 billion in 1981 had plastics
sales of about $500 million in its subsidiary, the Exxon Chemical
        2
Company.   This company operates three plants that produce polymers.   Gulf
Oil's subsidiary, Gulf  Oil  Chemicals, operates six polymer plants, half of
which produce polyethylene.  The Mobil  Chemical  Company, a subsidiary of
Mobil  Oil, owns four polymer plants, three of which produce polystyrene.
Other major oil  companies owning separate subsidiaries  that operate
                                    9-3

-------








3
OJ
«-H
ra
i
IO
"3
C?

S
BY MAMUFACTURER AJ
to
1
s!
to
UJ
cc
a
S

ce

§

C1-
u.
o
ce
UJ
1

f
i
en
co

/^
19
t—






ro
,2

i tt
cu
to
cu
3:

a
Cu



L
Manufacturei

ro
£



i—
UJ
CU



co
cu


UJ
cu
=


cu
a




n
CU




Manufacturer
CO r-4 r-* Ovl t-H


»— t

T-< •-<
•
-

1— *
OJ r-4




§
en C
TJ *— 4
a in
0 ra "
<3 U J=
« i Z t
O *T



CM »-« CM




r-4 CO •*


" •
1—4 «— 1















C ra
0 C
•f— 4J -r-
4-a wl *+-
ra J= O
!_ U S-
in o co 4->
u a. o co
•r- !_ 3: Cu
4-> O
v> cj c c
ra ra ra
r— ra -a u o
CU C CU ••-•'-
o •*- s- i-
11 1 tsj r— CO CO
CM -3-


CM

CO



1—4





Minnesota Mining
Mobil Chemical

^ r-H



r-4




f—l



f— »



f— 4





f—l





Atlantic Richfield
Avtex Fibers
«^- OJ •— *


t— t

CO

-*

CM


ra
u
$_ *g
CU CO
Monsanto
National Distill
National Petroch

CM CM r-l








CM



r-4



r-H r-l











BASF Wyandotte
Chemplex
Cities Service
CM






r-4


«— 1


r—
3
Phillips Petrole

CM








CM



















Crest Container
ro r-i co


!— 1

CO r-l




CM




CU in rO
3 ro U
O ra f-
S- 3C E
^n QJ
•Q J=
t. C U
"o o j=
a. ce to

i-i en •*
r-4



0




r-l LO



CM CM



CM CM











Dart Industries
Dow Chemical
Du Pont
CM






r-1


r-4




i.
O
a.
X
2
o
00

-



«3-












rH





r-l





ryl
o
's
co
JZ
o
4J
in
UJ
LO




CO

--1


r— i

ra
c
IfH
•a
•r—
O
•o
i.
ra
•a
c
ro
4->
CO

nr
















CM





CM




in
ra
3
4->
ra
^
o
I/I
a.
LU
r-l LO r-4




f—l I— 1

1—4

CO r-l



U

Texstyrene Plast
Union Carbide
United Foam

CO LO »-l



If) •-•












CM





r-4


CU
jd
-§
Exxon
Fiber Industries
Firestone Tire & R
CO i-l CO
T— 4
CO


- CM

i2

S
CM j£>




ro
U
J= CO «t
O 4-> 1 —
CO 4-> 1—
CO T-

CM CM IO



CM




CM CM



r-4



CM





'->


S.
CO
_a
3
O£
S_
•r- C
-§
S- S- r—
a o •<-
Q) CD ^D
-0* r- M-
0 !- r—
O ra 3
f rs (_J r n





















.
LO
1
CO

o
r-l

CO
0)
.a
ro
r—
C
T3
ai
.^
ro
1 *

O
u

ra
ra
•a
1
M-
•a
(U
'I
o
o
ai
o
s-
o
ra
9-4

-------
polymer plants are Standard Oil (Indiana), Atlantic Richfield, and Shell
Oil; and two major oil companies—Phillips Petroleum and Cities
Service—operate plants that are integral parts of their parent corporate
structures.
     About 10 percent of the polymers and resins plants are owned by
companies in industries as diverse as food processing, textiles,
packaging, and photographic equipment manufacturing.  For the most part,
these plants operate as subsidiaries or as parts of subsidiaries owned by
the major companies.
     The remaining two-thirds of the plants are owned by corporations
classified in the basic chemical industry and are generally organized as
divisions of the parent company.  Only a few plants operate as
subsidiaries.  Du Pont, the largest chemical  manufacturer, with total
sales exceeding $20 billion in 1981 and plastics sales (including
fibers) of over $3 billion, operates 14 plants, most of which produce
    2
PET.   Dow Chemical, the second largest chemical manufacturer, owns nine
plants producing polymers and resins.  All of Dow's plants produce either
polyethylene or PS; none produces PET.  Other large chemical companies are
Union Carbide, American Hoechst, Dart Industries, and Monsanto.  Enka, a
subsidiary of Akzona, operates two polymer plants producing PET fibers.
All but a few polymer plants are owned by large, multi-plant corporations.
     9.1.1.1.2  Capacity share.  Table 9-2 lists the 1981 production
capacities of those firms that produced the five polymers and resins.
     The total capacity of the plants producing the five polymers and
resins exceeded 15 thousand gigagrams in 1981.  The basic chemical
manufacturers provided approximately one third; major oil companies
another third; and the remaining third was accounted for by plants owned
by such diverse companies as Eastman Chemicals, USS, Bordens, and National
Distillers.
     The largest, single manufacturer of PP was Hercules, which accounted
for nearly 25 percent of the total  capacity.   Oil companies accounted for
about 40 percent of capacity.
     Phillips Petroleum had the largest production capacity (about 25
percent) of HOPE.  Oil companies, in general, accounted for approximately
60 percent of the HOPE, and chemical companies provided 30 percent.
                                    9-5

-------












>a
CM
CO
~
,_(

J^
ro
s
e
ro
rs
LU
Q.
1

LU


1™ *

l£
S
s

CO

CO
1
af
=
to
LU
ce
cc
5
g
u.
0
|_«


^
°

CM
cn

£

tu
ex.


to
^u


LU
a.
§


LU
cu
o
«J

a.
a.






(U

5
.£
§
03
*£*




"«
4J
O
1—
t:
a.


10
a.


LU
a.
a
2=


LU
a.
o
~"

a.
Q-







s-
£
3
O
3
C





in o in in o
§«-« CO VO I-H
CO

in
CO

o o



1 1
IS
in


IT)

CM


1 8







C
o
•a
J 8 .a
in c 'C t!


3 in <2 u

3; ^ " .5 2

in in o in in
co r» CM

CD O ID
CM

LO LO LO
v-1 CO CM
CM CM
in
0
o m
O CO
CO i-«













~ (O
O C
4-J I/I ii-
ia .c o

u a. o cu
•r- S- 3= D-
4J O
(O n3 fO


0 i- S- S-

< 
CO CM

LO
CO

0








in
CO
«—*










cn
c:
E "io
s
to (U
O CJ
(/>

C *i^
•^- o

tn o
CO CM
r^

O
CM

LO
fH
CM


0
U)
l-H


0
CO


0
CO
t— 4







•a

•o s_
£= 'D
ro x co
>, (1)
3 r— C/l
rv 





£
3
CU

S a.
CU O
CO
t/1


r— W>
r— >•)
s. ci

0 0
CM r^.




0 0
CM r*.



















&. in
 cn
o -a
CJ C


V) 4-1
S- (O

o in o
m ° 5

0


in

—<



o
CO







o o .
p»* cn
CM









r— S-
ui fO O)
« *^ >»
5 o


r— CU
E ^~ +J
•§ J: 'o

o m o
CO CM i— »
co **• r**

o o
O CO
crt «3~

in
CM


in in
o o
CM CM


o O in


in
vo







in

u
0 .E
ai -w e


3 cn
o = «

o in





in in

"-•



o
t— i







in
CO
CM


*~*
C
co tn
•t— O

C 4J

i— Q_
o cu
c
T3 (U


^) 4-*
C in
4-> CU

in in
CO CO
CO IO












o in
in in
CM "3-


in o
ro CO






in
O

£
Z3
4^
(O
O


Q- O
X
.— X

in in
O
i— i
1


in





o
cn


o in
m r-.
cn












ai
-^
!5 1
S- 0
10 U-


C 01
O 4->
c c

o in
CO CM
<£>

O in
CO CM
IO
















S-
cu

.a

C£
in 03

•^ CU
4J •»-
3
1— O


O (U

Li. LU
in in
Si




o in













in
CO
CM









in
n3
o
'g

.c: 
•^
t/1 4J
CO -f-
=> =»

LO O O
co r^ cn
t—t cn

m
CO

0 0
r^. 10
r-H


0

^


O
cn
ro


o
CO
.-H



s-
ai
^3

3
01

01
i-
•i- CZ
h- O
"S ^
03 O -t-
0) d O

•O i — M-
0 S- r—
O -03
O CJ O
o
CO
m
in
i— i
o
o
PI
CM
LO
CM
IO
CM


in
in
cn
CM

0
in
p^
<3-

0
in
CO

CM












_J
4-

•a
a>
9-
t=
o
(_>


-------
      Oil  companies  accounted  for about  30  percent  of  the  total  LDPE
 capacity.   The  chemical  companies accounted  for  close to  30  percent with
 the  two  largest chemical  producers  of this polymer producing about 20
 percent  of this amount.
      The  basic  chemical  companies accounted  for  close to  40  percent of the
 PS market  with  the  two  largest  producers providing about  30  percent of the
 capacity—about the same  as that of all oil  companies.
      PET  capacity was essentially—close to  90 percent—devoted to
 manufacturing fiber.  Du  Pont had the largest capacity for
 production—30  percent  of the total.  About  10 percent of the PET
 capacity was in film and  bottle  resins, and  40 percent of that was
 produced by du  Pont.
      9.1.1.1.3   Vertical  integration.  Vertical  integration is the
 operation  of a  single firm at more  than one  stage  of  production.  For the
 purpose of this analysis, it is  an  indicator of  the ability of firms to
 invest in  pollution  control, and to use or produce substitutes that could
 involve less air pollution.  Also,  vertically integrated firms may find it
 easier to  pass  compliance costs  forward or backward.  Vertical integration
 in polymers and resins follows two  distinct  patterns:   "backward-to-
 suppliers"  is characteristic of  oil companies; "forward-to-final-product
 markets" is typical of chemical   companies and of the  subsidiaries of other
 types of manufacturers.
     Backward-to-suppliers integration.   Over a third of the polymers and
 resins production capacity is owned by oil  companies that provide organic
 chemical inputs  to production.  The greatest portion of this backward
 integration involves large petroleum producers such as Exxon and Gulf.   A
 relatively small portion involves industrial  organic chemical producers
 that produce both the monomers and polymers.   Essentially all of the
 backward integration is  employed by the  manufacturers  of PP, HOPE,  LDPE
 and PS; very few PET manufacturers are so organized.
     Forward-to-final product market integration.  Forward integration,
which involves two or more of the stages of manufacturing, occurs among
 all  of the different types of firms that produce  one or all  of the  five
 polymers and resins.  Firms that employ  such  integration account for
                                    9-7

-------
one-third of all polymers and resins production capacity.  Chemical firms
frequently employ forward integration: Dow Chemical, for instance produces
polymers and resins and such end products as food wrap and insulated cups;
Union Carbide carries its polymers and resins production through to food
wrap products.  Forward integration is also characteristic of particular
divisions or subsidiaries of firms in such diverse industries as food
processing, textiles, photographic equipment, and aerospace.  In addition
to its PET production, for instance, Eastman Chemicals manufactures
plastics materials for molding camera housings.  (Additionally, Eastman
Chemicals maintains approximately 30 percent of all polyester film
capacity.)  Dart Industries supplies polymer to its household products
manufacturing subsidiaries.  Forward vertical integration is found also
among other PET fiber producers: Firestone and Goodyear produce both PET
fibers and the tires in which they are used; Monsanto produces the PET
fibers used in its production of artificial turf.  Texfi previously
produced PET fibers for its textile mills; however, it discontinued this
production in 1980.
     9.1.1.1.4  Horizontal integration—product diversification.
Horizontal integration can be found among (1) firms that are essentially
petroleum companies concerned with polymers and resins production, (2)
firms that are primarily chemical manufacturers, and (3) firms within
other, more diverse industries, e.g., food processing and textiles.  Some
60 percent of all polymers and resins plants are owned by diversified
and/or horizontally integrated firms which produce two or more polymers
and resins.
     All of the major petroleum corporations have horizontally integrated
facilities for the production of at least two polymers.  Atlantic
Richfield and Gulf Oil operate separate plants to produce each of the
polymers and resins except PET.  Among chemical firms, all major
corporations produce two or more polymers and resins.  National Distillers
with such diverse operations as distilling, metal fabrication, and
textiles is typical of other firms whose operations include the
horizontally integrated production of two or more polymers.
                                    9-8

-------
     The polymers and resins industry's single-plant operations are
generally those producing PET.   These constitute some 70 percent of the
single-plant operations in PET.
     9.1.1.1.5  Total production and capacity.   Total production of the
five polymers and resins increased from 11.3 thousand gigagrams in 1980 to
                      •j
11.7 thousand in 1981.    (These  data vary from data contained in Table
9-9 because of difference in the scopes of the sources.)  Total polyester
fiber production accounted for about 45 percent of all  synthetic fibers
produced.  Of the five polymers, LDPE accounted for the greatest
capacity—close to 30 percent while PET accounted for the lowest at 16
percent.  In 1981, capacity utilization varied from a high of 90 percent
in the production of PET (fibers) to a low of 70 percent for PS (Table
9-9).
     9.1.1.1.6  Value of shipments.  Values of shipments for the five
polymers and resins are included in the total shipments data for three
SIC industries:  Plastics materials and resins (SIC 2821), Organic fibers,
noncellulosic (SIC 2824), and Miscellaneous plastics products (SIC 3079).
Values of shipments for SICs 2821 and 2824 are listed below for the period
1977-1981.  Shipments for SIC 3079 are not shown, because the five
polymers and resins are a minor portion of that sector's shipments and
cannot be separated from the total.
     The total value of shipments for industry 2821 increased from $10.8
billion (nominal) in 1977 to $17.5 billion (nominal) in 1981.4  SIC 2821
does not show shipments by individual type polymers, but it does show them
by  categories—thermosetting and thermoplastic resins.  Shipments for
thermoplastic resins (SIC 28213), those which include PP, HOPE, LDPE, PS,
PET plastics, and polyvinyl chloride, amounted to $9.3 billion in 1977.
Although no values of shipments for thermoplastics have been published
since 1977, estimates of the 1981 value can be made by assuming that
thermoplastics grew at the same rate as did overall shipments for SIC
2821.  Based on this assumption, the 1981 value of shipments for
thermoplastics is estimated to be about $15.0 billion.  Although the value
of shipments of thermoplastics includes polyvinyl chloride, for purposes
of this study the value of the four polymers produced in the plastics
products industry (SIC 3079) are assumed to be about the same as that of
                                    9-9

-------
the polyvinyl chloride; consequently, the total SIC 2821 and 3079
shipments in 1981 for PP, HOPE, LDPE, and PS is estimated to be $15.0
billion.
     Shipments for SIC 2824 Organic fibers, noncellulosic increased from
$6.4 billion in 1977 to $11.0 billion in 1981.6  Assuming that the
growth of PET approximated that for the total industry, the shipments for
PET fibers (SIC 28244) are estimated to have increased to $3.8 billion in
1981.  The sum of the shipments of thermoplastics and PET fibers, as shown
below, is $18.8 billion, an approximation of the value of the five
polymers and resins.
     The industries' values of shipments for 1977 and estimated values (of
the five polymers) for 1978, 1979, 1980, and 1981 are tabulated below:
     Industry
1977
1978
1979
1980
                                      •(million nominal dollars)-
1981
SIC 2821
Plastic materials
and resins
10,818
11,998    14,282    15,570    17,500
SIC 28213
Thermoplastic materials
and resins
 9,266
10,277a   12,233a   13,340a   14,990a
SIC 2824
Organic fibers
noncellulosic
 6,380
 6,921    8,227
          8,811
          11,035
SIC 28244                   2,187      2,373a   l,821a    3,021a     3,783a
PET fibers
a Estimated by EPA
     9.1.1.2  Plants.  The considered polymers and resins  are  produced  in
over 100 plants with varying characteristics.  Some of the more important
characteristics—size, age, location, and employment—are  discussed  in
this section.
     9.1.1.2.1  Size.  The annual production capacities of plants
producing the polymers and resins vary from less than  10 gigagrams for  the
                                    9-10

-------
smaller plants, which produce PET fiber and film, to more than 400
gigagrams for the largest polyethylene plants.  Except for the large
du Pont and Fiber Industries plants with capacities in excess of 100
gigagrams, most PET plants have capacities of less than 50 gigagrams.
     Table 9-3 shows the size distribution of the polymer and resin plants.
The distribution of the LDPE and HOPE plants is similar—about 70 percent
of the plants of each type have capacities in excess of 100 gigagrams.
     9.1.1.2.2  Age.   Data are not published on the ages of individual
plants listed in Table 9-1; consequently, plant age distributions cannot
be developed.  Information is available that indicates when the different
types of polymers and resins were first introduced.  PS was introduced  in
1938 and used in manufacturing housewares.  LDPE was introduced in 1942  for
use in packaging.  HOPE and PP were introduced in 1957; PET was first
produced commercially in the U.S. in 1953.
     9.1.1.2.3  Location.  The 128 known plants producing the five
polymers and resins are located in 21 states and Puerto Rico (Table 9-4).
The high capacity plants producing the polyolefins (polyethylene and PP)
are predominately located in the petroleum producing regions of Texas  and
Louisiana.  The PS plants are located primarily in industrial states:
Illinois, Massachusetts, Ohio, and New Jersey.  The PET plants are located
in large textile mill areas in order to supply fibers to that industry,
and over 60 percent are located in the Southeast.
                           4 fi
     9.1.1.2.4  Employment. '   The 1981 employment data for plants
producing the five considered polymers and resins are included in the
Census of Manufactures data by SIC code.
     As shown in Table 9-5, in 1981 an estimated 66,100 workers were
employed in plants producing the five polymers and resins: 34,100 in
plants producing polymers only; 10,700 in processing plants captively
producing polymers; and 21,300 in PET fiber plants.  Total employment  in
thermoplastics increased slightly from 1980 to 1981; however, employment
in PET fibers decreased significantly, a condition reflecting an overall
reduction in textile industry production.  The ratio of production workers
to all workers is significantly greater in the fiber plants (SIC 2824)
than in the plastics plants (SIC 2821), a difference that reflects the
labor intensiveness of these plants.
                                    9-11

-------
       Table 9-3.   SIZE DISTRIBUTION OF POLYMER AND RESIN PLANTS
                BY TYPE AND CAPACITY; January 1, 1982a
                               (percent)
Capacity in gigagrams
Product
PP
LDPE
HOPE
PS
PET
100
and
below

43.8
10.0
25.0
75.7
59.3
101
to
200

37.5
40.0
50.0
21.6
22.2
201
to
300
______ i °/. } 	 	
	 \/0) 	
12.5
35.0
12.5
2.7
14.8
301
and
above

6.3
15.0
12.5
-
3.7
Total

100
100
100
100
100
Source: Compiled from data contained in Tables 3-1 to 3-5.
                                9-12

-------
            Table 9-4.  NUMBER OF POLYMER AND RESIN PLANTS
                BY TYPE AND LOCATION; January 1, 1982a
Location
Texas
Illinois
South Carolina
Louisiana
North Carolina
Ohio
California
Massachusetts
New Jersey
Alabama
Tennessee
Pennsylvania
Virginia
Iowa
Kentucky
West Virginia
Colorado
Connecticut
Michigan
New York
Puerto Rico
TOTAL
PP LDPE HOPE PS
11 13 11 7
1 2 7

2 4 3
1
6
6
5
1 3
1

2
1
1 1
1 1
1

1
1

1
16 22 15 42
PET


10

8
1


1
3
4
1
2


1
1


1

33
Total
42
10
10
9
9
7
6
5
5
4
4
3
3
2
2
2
1
1
1
1
1
128
Source:   Compiled from data contained in Tables 3-1 to 3-5.
                               9-13

-------















ID

«*
in

**
*^dh

t-H
CO
Ul
i— 4

Q
"3^
i^C

(*s»
p->
I— t
1—4
t—
2=
l^c
, 1
ex.
•z.
to
LU
C£
O
•ss.

•4*^
to
3
•o
1— 1
!





4-
O
S-
QJ

S
Z3
d

a>
.c
4*}

*+—
o

E
O
•r—
.{_}
£-
O
a.
o
s-
a.

CO
-E
(—
O O
O O
in r^
« ^
oo oo
m «*




o o
0 0
CM r***
«» ft
f-» U3
in «*

„



«— 1 OO
CM <—4
CO CM
CM 00
CM




•a
to

to
^«
ro
•r-
s_
O)
4J
(X3
E
o
tO •!-
O 4->
•i— CO
•»-> (O
W i—
CO f*
i — to o
a. E E
•i- S-
i — to CO
r- CO .E
< i- I-

• •
i— 1 CM

CO
E
CO (O
.E to
4->
CO
O -E

to co
O •!-
•r-
+-> t-l
to CM
fO OO
r— CM
a.
0 O
E i— i
i™ co
CO
f* c~
•»-> •!-

E S-
•r~ CO
JD
to E
CO 3
CO E
>>
O i—
r— tO
CX4->
E 0
CO -t-3








































































•
(>•»,
[••^

g^
CO
F
>»
O

"a.
E
CO

to
CO
-a
3
^_
O
X
LU
O
O
r-4
n
^J-
OO




*

•*.^*

CO
i.
CO
E
>j
"o
Q-


oo
.0
OO
1—4
CM
00
CM

O
co

E
•r-

f_
IO
•f->
O


o
c*7
-(->

t^
o

-(->
E
CO
o
s-
O)
CL

CM
CM



























































c^.
O

E
O
1J3
£.. ^«
o to
O-4->
O O
i- 4J
CL
CO
CO -E

•i-J

E -P
O
E
•a o
CO -r*O

tO O E
ja 3 o
-a -r-
to O 4->
•r- i. U
a. 3
to ~a
•i- 0 O

-M a. a.
























































cy> -4->
r-«. E
O CO
oo E
>^
c-j o
H-H r~" •
CO Q.
E CO
•r™
r—
4-) fO
c: -t-J
 CM
•r- S- CO
- r— CO CM
cx a.
Q. 0
tO O >— <
CM CO
aj

H^ ••- -r-
o
o

ft
o





t
«a:
»
22:







01
r~-.
o
00










>,^
, —
CO to

•r- CO
-t-> E
Q. >>

cj o
C5.
CO
^» C7)
O E
tO •!—
to u
O) 3
u -a
o o
S- S-
Q- CX


•*



































O
O
o

CT^
U3




O
o
o

*^I~
r*^>






*=xt"
CM
co
CM

















s_
CO

•r«
C|_

O
•i^
E
tO
CD
i_
o


un
E
S- •!-
CO
^2 to
E to
• 3

0 CO
i- (O t— 4
CO 4->
JD 0 E
E -P •!-
3
E CO CO
j= E
CO -l-> (O
j= to
•M O
+J CO

O 1— -4->
LU
E a. to
O -i-

1 ^ .f^ «^>
S- CM
O to CO
CX S- CM
O CO
S- >»O
Q. O >— < •
r~ CO f*^
CO f"* i^^
g~ E E O"l
1— CO -r- ^H
O O
o o
00 1-4
•» #1
1—4 tO
CM «D




o
o
CO

CM
CM






«3j~
«>J-
CM
CO
CM

I —
LU
a_

oo
Q_

1— •>
LU • I i I
Q- Q.
• 	 Q
n:
CO
i- «
CO LU
l~l O
•r~ f~>
t[ i

S». ft
CO CL.
•M 0.
co — ^^
CO U3
>> •— +
O 4^ -j-
Q- O CO
h— *^^*

UD








































































•
CO

J3

t
•F—
fO

fQ

4-J
O
E
||

•

-------
                      Q
     9.1.1.3  Markets.   The versatility of plastics has made them .
suitable for use in such diverse markets as furniture, automobiles,
housing, packaging, toys, and electronics.  Their special properties and
high-volume processing capabilities combine to give them a competitive
edge over wood, metals, paper, and glass, and most especially in the
packaging and consumer products industries.  The largest product by volume
of the plastics and resins is LDPE, a product used as film in the
packaging industry.  PP has realized the most rapid production growth in
recent years—generally at the expense of PS—with significant applica-
tions in both the auto industry and the fibers market.
     9.1.1.3.1  Substitutes.  Plastics materials and fibers derived from
the five polymers and resins are used as substitutes for wood, metal,
natural fibers, paper, glass, and other plastics and synthetic fabrics in a
variety of industries.  PET derived fibers, for instance, are widely used
substitutes for cotton fabrics; PP is used as a nylon substitute for
automotive fabrics and home furnishing (especially carpeting) materials;
LDPE is an extensively used packaging material substitute for paper; and
HOPE substitutes for steel and other plastics in the fabrication of pipes.
One of the most rapidly developing markets is the automotive industry where
the emphasis on fuel efficiency has lead to the use of lighter-weight
substitutes for metal; PP, for instance, is now used extensively in the
fabrication of such automotive body components as fenders and doors.
     9.1.1.3.2  Imports-Exports.  The U.S. balance of payments has been
strengthened by the consistent export surplus of both plastics materials
and resins and synthetic organic fibers.  The 1981 export surplus in
plastics was about $3.0 billion; that in synthetic organic fibers was $576
        9
million.   However, as U.S. raw materials prices gravitate toward the
world prices as a result of natural gas and oil deregulation, domestic
producers will lose much of their past competitive advantage over foreign
producers.  Additionally, developing countries with extensive gas and oil
exports are planning to bring petrochemical complexes on line in the 1980s
to process their raw materials.  Saudi Arabia, Mexico, and some Far
Eastern nations, for instance, plan major world-scale plants.  Among the
more industrialized nations, Great Britain, Norway, and Canada plan to
develop processing plants to exploit their newly discovered oil and gas
supplies.
                                    9-15

-------
     Table 9-6 shows the exports of the five polymers and resins for the
                 9  10
period 1976-1981. '     The greatest increase in exports for the period
occurred for PET fibers—over 500 percent.  The next greatest increase is
shown for LDPE—an increase of about 76 percent.  No increase is shown for
PS.  The greatest volume of exports—over 400 gigagrams exported in
1981—was of LDPE, and the volumes of PP and PET exports were 311 and 400,
respectively.  The lowest exports were for PS—fewer than 100 gigagrams
exported.
     Polymer imports have been relatively minor—less than 100 gigagrams
in 1981—and of this 50 Gg was HOPE.  In PET fibers, imports reached about
15 gigagrams, an increase over the approximately 8 gigagrams reported in
1980.10
     9.1.1.3.3  Processes and uses.   The 1981 consumption of polymers and
resins by type of processing is shown in Table 9-7.   About 3,000 gigagrams
of LDPE were processed with over 60 percent of this  amount extruded into
film.  Of the 1,400 gigagrams of PP processed, about a third was extruded
as fibers, another third processed by injection molding, .and the remainder
was processed into film and other products.  The largest portion of the
1,900 gigagrams of HOPE processed was molded, 40 percent by blow molding
and about 25 percent by injection molding.  About one-half of the PS
processed was injection molded.  The primary use of  PET was for fibers and
fi1aments.
     9.1.1.3.4  End-use markets.  Shipments to end-use markets in 1981 of
PP, HOPE, LDPE, PS, and nonfiber PET are shown in Table 9-8.  Over 4,000
gigagrams of these polymers and resins were shipped  to the packaging
industry for fabrication into such products as bottles, jars, food
containers, refuse bags, and baskets and LDPE represented almost 50
percent of the plastics so shipped.   Another 1,000 gigagrams entered the
consumer (toys, kitchenware) and institutional markets: PS accounted for
one-third of this, and the remaining two-thirds was  split about equally
among the polyolefins.  Approximately 190 gigagrams  were supplied to the
transportation industry with PP accounting for about 60 percent of this
amount.  About 400 gigagrams entered the furniture and home furnishings
market with nearly 90 percent of this as PP, essentially for use in textile
products such as carpets and drapes.  HOPE contributed about half of the
400 gigagrams utilized in the building and construction industry.
                                    9-16

-------
 Table 9-6.  U.S. EXPORTS OF POLYMERS AND RESINS, BY TYPE AND YEAR,

                          1976-1981 9' 10
                             (gigagrams)
Type a
PP
LDPE
HOPE
PS
PET

1976
161
254
166
70
63

1977
128
260
193
53
80

1978
168
343
223
71
150
Year
1979
324
407
276
73
238

1980
308
520
239
76
330

1981
311
446
209
68
400
Exports represent sales of plastic resins for all  types except PET;
exports of polyester indicated are fibers.  Less than 20 gigagrams of
polyester resins were exported in 1981.
                              9-17

-------
o
*"*
4%
CD

(••'I
CO
en
UJ
Q-
i^^
i^
CO
CO
LU
O
0
C£.
a.
o
•z.
s
C£
•[£

LU
CO

1
0
UJ
(/)
co E
z: to
H-4 S-
CO CO
LU ra
C£. CO
•r-
O CO

CO
LU
O
a.
u.
o

Z:
O
1— 1
(—
a.
!§
co
z:
0
CJ
(_J
r— (
t—
CO
LU

f"*i
Q


•
y*^
1
CTt

CU

JD
(O
f-






(— -
LU
Cu







co
a.







LU
a.
a
in









LU
Q.
0
_J






a.
o













4J
CU

.«• s-
co ra
10 E
cu
o cu
O t/)
S- 3
Cu 1
•a
c
LU







o Qi D; oi o oi CD
I-H z: 2: z: CT> z: o
T— < ^" CO
" "
'"~4





o; o; QC oi c£ oilco
z: z: z: z: z: z: 1-1
LO
i






1^. "^- O LO Di «=J"|O
CM CM LO CO Z: CM i-H
1— 1 t— 1 ^1-










cr* to r*^ t— i <"v^ ^~ fr^»
^d- LO r— i z: COKO
CO T-I CM CO
A I **
i-H (CM




r*^ vo fo ro CM co |co
co 1—1 co co ko
r— < ^* KO
1
1








^J
o
3
(/) "O
•»-> o
c: s-
4-3 CU Q-
•r— E
CU 3 
lOECUCUCUCUCUO
3 r— CU S- O-JD -C I—
S- -i- J= T- ••- -r- -»->
•»-> U_ CO 3 Q- U_ O
X
LU


O O O 0
i"""*, r*«*
T-I r-.
t — 1





CO O LO VO
•sf O 10
r^. co LO
A
1— f





CM 01 «^ LO
LO VO CO i-H
•^ r*^ CM CTV
n
T—l








00 1^- r-H CO
CO CM CT> VO
CM co cr>
«*
CM




«d" CTV i— < CM
VO CM CD LO
LO t-H ^f
•%
t— 1















CO
c
•r-»
T3
r—
O CO
E £=
•r- t/1
C T3 S-
O r- CU
•r- O -G
+-> E +->
0 O t—
0) 3: ro
•1-3 O r— 4-> .
C i— r— O
i— i CQ <: i —









































=
CO
.4^
o
3
-a
o
s_
Q.
T3
CU
•a
3
1 *
X
cu

^,
cu

•4-3 •
o -o
= cu
4-3
C S-
•r- O
Q.
T3 a>
cu a: •
-a
3 4->
r— O
0 Z.
c:
i— i n

o^*
(O Z!
9-18

-------










3
-I
«t
n

t— i
CO
cn
i— i
1—
LU
^
^
rv
0


_J









Q_
Q_

















4-i

S- O
4-) tU
£= 10 i—
o c: LU
•r- O
-4-J C_J ••>
res i —
•4"^ O^ • ** fO
s- E cn o
O -i- C -i-
Q. cn T- s-
to to -o +->
C sx r— CJ

•t—
53
S-
3
LJ_


LO T3 "O -a cn «*
t-H tO
CM







co .a JD «* co LO
«* CM tO OO
OO CM tO

t— 1




cn to js >— i cn «d-
»-l Sf tO O CM
CM CM CM i— 1
n
CM










•-H JQ -Q i— 1 VO Cn
r*^ »^ *sd" f^
CM CO st- «*

oo








r-t 1C) ^1 tO r-H OO
CM CM CO t— 1 to
CM CM oo r--.
A
I— 1






c/)
o cn
c:
^^ •?"•
fO +->
c >, res
0 i- 0
•r-  J=
•i— O CO
+•>  — i
1— H ••»
r^ • A
••» (O C/l i-
i_ T- CD  ^: to
E 1 ' 'r* •!••} +J
3 CO t/J O S- i —
co 3 cu , o ta
c -a _c i— CL •*->
o c: -a r— x o
<_) i— i «C cc LU f—
to
T3
0
O
cn

cn
c:
o
a.
to
in
.(->
o
3
•o
O
S-
a.
ta
0
^
(O
) O •!-
i — 4->
C i — tO
O.i— £=

H- = ^
LU -a
Q- £= C
•i- ta •
S- 0
a) -a s_ -M
-Q CD OJ O>
•r- T3 E
<4- 3 ' 3 »
1 i — tO tO
£= O C >,
o s= oo
^r i— i t_3 -t-3


ta ^2 o
























































•
cu
3
to
o

o
to
•f—
T3

•o

o
^
rO

o


"C5
1—
O)

C"
-M
•1^
3


T3
9-19

-------
     Not shown in Table 9-8 is the shipment of polyester fibers.
Approximately 90 percent of these fibers were shipped to the textile
industry for processing into yarn for the apparel and related fabric
industries in 1981.  PET continued to be the predominant noncellulosic
fiber.  It accounted for over 50 percent of the noncellulosic output and
over 30 percent of the total of all fibers, both man-made and natural,
consumed in the textile industry.
9.1.2  Industry Performance
     The initial commercial development of the major thermoplastics
occurred in the period 1935-1945 with the introduction of such plastics as
PS and LDPE.  In the 1950's HOPE, PP, and PET were introduced.  Large
scale productions of these reduced their cost and they began to compete
with materials such as wood, cotton, paper, metal, and glass.  The current
general production marketing and financial characteristics of the industry
are examined below.
                                      9  10
     9.1.2.1  Capacity and Production.  '     Yearly production and
capacity data for the period 1976-1981 are shown in Table 9-9.  Although
industry statistics may vary according to source, the data shown in Table
9-9 represent the general trends in the industry.  During the period, PP
had the highest average annual rates of increase both in capacity, about
13 percent, and production, about 10 percent.  PP capacity notably
increased about 40 percent in 1978-1979; however, it decreased about 5
percent 1980-1981.  HOPE averaged annual increases of about 9 and 10
percent in production and capacity, respectively.  Production and capacity
in LDPE increased at average rates of about 6 and 8 percent, respectively,
and the rate of average annual increase for PS was 2 and 1 percent for
production and capacity, respectively.  During the period PET fibers had
an average annual 3 percent increase in production and less than one
percent in capacity.  PET fiber capacity decreased 4 percent during
1980-1981—essentially because of the closing of the Texfi operation.
     Utilization rates were generally highest for LDPE, varying from a low
of 79 percent in 1981 to a high of 93 percent in 1979.  The lowest
utilization rates occurred in the manufacturing of PS—varying from a low
of 64 percent in 1976 to a high of 77 percent in 1979.  Such low rates
reflect the competition in this relatively mature sector.
12
                                    9-20

-------
   Table 9-9.  PRODUCTION, CAPACITY, AND CAPACITY UTILIZATION OF
                      POLYMERS AND RESINS 9' 10

1976
1977
1978
1979
1980
1981

Production (Gg)
Capacity (Gg)
Utilization (%)
Production (Gg)
Capacity (-Gg)
Utilization (%)
Production (Gg)
Capacity (Gg)
Utilization (%)
Production (Gg)
Capacity (Gg)
Utilization (%)
Production
Capacity (Gg)
Utilization (%)
Production
Capacity
Utilization (%)
PP
1,174
1,230
97
1,246
1,450
86
1,394
1,540
91
1,740
2,180
80
1,650
2,400
69
1,790
2,290
78
LDPE
2,640
3,000
88
2,935
3,220
91
3,226
3,610
89
3,530
3,810
93
3,310
3,950
84
3,490
4,430
79
HOPE
1,417
1,680
84
1,657
1,815
89
1,906
2,150
99
2,270
2,470
92
2,000
2,540
79
2,130
2,720
78
PS
1,450
2,260
64
1,563
2,385
66
1,730
2,360
73
1,820
2,360
77
1,600
2,450
65
1,640
2,350
70
PET a
1,630
2,070
79
1,640
2,040
80
1,720
2,090
82
1,890
2,180
87
1,810
2,180
83
1,890
2,100
90
Fiber only.
                              9-21

-------
     In 1980, the utilization rate for PP dropped to a low of 69 percent as
a result of that year's recession in the automobile and housing
industries.  In 1981, the rate increased to 78 percent, an increase
brought about by increased production and a decrease in capacity from
plant closings.  The utilization rates for PET varied between 79 and 90
percent during the six-year period.  As in the case of PP, the high rate
of 1981 reflects both a decrease in capacity and an increase in
production.
     9.1.2.2  Prices.  Polymer and resin prices have fluctuated
considerably during the last ten years.  As measured by the Producer Price
Index for SIC 2821, Plastics materials and resins, nominal prices
increased 56 percent during 1973-74 as consumer inventory buildups
occurred as a result of the OPEC oil embargo and anticipated ensuing
shortages.  Prices remained relatively stable from 1974 to 1978.  In 1979,
however, they jumped by 18 percent as oil prices increased.
     The 1978 and 1981 prices are compared below for each of the polymers
and resins.  The prices, based on data contained in Chemical and
Engineering News, indicate price fluctuations during each of the
years.  '''    The greatest increases occurred in the prices for PP
and PS with increases of about 45 percent for each.  These increases
primarily reflect the post-1977 rising demand for these polymers in the
automobile and home furnishings industries.  The price increase for the
other polymers amounted to about 30 percent.
                           Price ranges 13.14,15,16
                                            (nominal)
                                   1978
                        1981
            Product
            PP
            LDPE
            HOPE
            PS
            PET
  .60-.73
  .64-.73
  .68-.73
  .56-.66
1.43-1.54
            •($/kg)«
1.10-1.23
1.01-1.06
 .99-1.08
1.10-1.14
1.85-2.58
                                    9-22

-------
                       g
     9.1.2.3  Finances.   Plastics have experienced relatively good
profits over the past two decades because of their exceptionally high
demand as a replacement for such materials as metal, wood, glass, and
rubber; indeed, this demand has fueled relatively high growth rates in
production of all types of plastics.  The Federal Reserve Board's
production index indicates that the annual growth for all plastics
materials averaged 9.5 percent per year between 1973 and 1979 compared
with the 3.1 percent growth rate for all industrial products.  However, in
1980, plastics production fell 9 percent while the overall industrial
output declined by only 3.5 percent.  This comparatively large drop
reflected the severely depressed state of two key plastics markets—the
automobile and construction industries.  Because many of the plastics
markets are expected to mature within the next ten years, long-term growth
is not expected to exceed 6 percent, an amount well below past growth
rates.  Recent trends in sales and profits for the five polymers and
resins are outlined below.
     9.1.2.3.1  Sales and value of production.   The trends in sales
(nominal dollars) and value of production (nominal dollars) over the past
several years is shown in Table 9-10.  The sales data reflect the actual
sale of polymers and resins on the open market and exclude interplant
transfers and captive consumption; however, data for the estimated value of
production of polymers and resins produced (for sale and inventory) do
include captive consumption and interplant transfers.
     The highest domestic dollar sales during the 1978-81 period for the
four polymers and resins (sales for PET fibers  are not available) are
shown for LDPE.   Its sales increased from $1.5  billion in 1978 to over
$1.9 billion in 1980.  Sales for this polymer then increased only slightly
from 1980 to 1981, from $1.9 billion to about $2.1 billion.  The greatest
sales increase during the period occurred for HOPE with the level
increasing from less than $1 billion dollars  in 1978 to over $1.4 billion
in 1981.17
     The value of production for all  five polymers increased significantly
between 1978 and 1979, stabilized between 1979  and 1980 and increased
between 1980 and 1981.  The greatest value of production is shown for PET
fibers.  The value for this polymer increased from $2.4 billion  in 1978 to
                                    9-23

-------
        Table 9-10.   SALES AND  VALUE OF PRODUCTION  OF  POLYMERS  AND
                      RESINS  12,13,14,15,16,18,19,20
                    1978-1981 (million  nominal  dollars)
Item
Domestic sales
PP
LDPE
HOPE
PS
PET a
Value of production
PP
LDPE
HOPE
PS
PET
1978

612
1,492
828
978
NA

800
2,000
1,200
1,000
2,400
1979

695
1,901
1,148
1,387
NA

1,250
2,800
1,700
1,500
3,400
1980

793
1,974
1,237
1,358
NA

1,250
2,800
1,750
1,500
3,500
1981

1,015
2,077
1,414
1,446
NA

2,000
3,300
2,250
1,800
3 ,850
NA = Not available
a
   Domestic sales data are not available for polyester fibers.
                                  9-24

-------
 close to $3.9 billion in 1981.   The increases  in values  for the polymers
 generally varied between 10 and 20 percent from 1980  to  1981 except for
 PP.   Its value increased from $1.25 billion in 1980 to $2  billion  in
 1981—a 60 percent  increase.13'14'15'16'18'19'20
      9.1.2.3.2  Profits.21   The profitability  of firms classified  in SIC
 Industry 2821, Plastics  materials  and  resins,  is shown below for the
 period 1974-1981.   These before-tax profit margins are based on Robert
 Morris Associates Annual  Statement Studies,  a  publication  that
 incorporates  the-financial  data of about  130 firms in the  plastics
 industry;  however,  because  this publication  excludes data  from  firms
 having assets  in excess  of  $50  million, the  profitability  values shown  do
 not  reflect the  profiles  of the larger firms that produce  polymers,  e.g.,
 du Pont,  Dow  Chemical, and  the  large petroleum companies.   It may,
 however,  cover some of the  small subsidiaries  of these large corporations.
 For  the  most part the returns reflect only the  profits of  the smaller,
 single-plant operations.
                       BEFORE-TAX PROFIT MARGINS
                                                 21
1974
8.6
1975
6.0
   (Profit as a percent of sale)
1976      1977      1978      1979
7.8       3.2       4.8       3.7
1980
3.6
1981
3.6
Composite returns in 1974 approached 9 percent of sales and reflected the
effects of the general price increases and economic expansion in 1973.  In
1975, the profits dropped to 6 percent of sales.  Since 1976, profits have
varied between 3 and 5 percent.
9.1.3  Five-Year Projections
     Capacity increases are projected below for the period January, 1984 to
January, 1989 and are expressed in terms of required new model  plants or
process lines.  Eighty-five new model process lines equivalent  in size to
27 model plants (defined in Chapter 6) are projected.   The numbers of model
plants and process lines by type of polymer and process are:
                                    9-25

-------
                    Polymer/Process
Number of Plants
  Number of
Process Lines
                  PP
             Liquid phase
             Gas phase

                 LPDE
             High pressure
             Low pressure

                 HOPE
             Liquid phase-slurry
             Liquid phase-solution
             Gas phase

                  PS
             Continuous
             EPS,  PI
             EPS,  IS

                 PET
             DMT
             TPA - single end  finisher
                 - multiple end  finishers
       3
       3
       1
       5
       2
       3
       5
       1
       0.7
       0.3
       1
       1.9
       0.5
      9
      9
      4
     10
      6
      9
     10
      2
      2
      3
      7
     13
      1
       The methodology used to project capacity increases expressed in terms
of the number of new model plants  is discussed below.  No explicit locational
component was included in these projections, for the model plant construction
and the baseline control estimation (see Chapter 8) take into account the cur-
rent geographic distribution (SIP  and non SIP states) of the existing plants
and assumes new plants would mirror the same distribution, capacity, and emis-
sion controls.  The numbers of plants by process result from a growth analysis
of each polymer, as described below.  The projection is subject to some uncer-
tainty because a few markets, as indicated in Section 9.1.2 above, are currently
unstable, and such instability begets caution.  Chemical  industry opinion indi-
cates that there may be few new grass roots polymers and resins facilities
built over the next several years.18  Such assumptions imply that unstable
markets mean that some growth in production will  come from the upgrading or
demothballing of existing production units in ways that may not technically
be considered modifications under the Clean Air Act, and,  therefore,  not subject
to regulation by an NSPS.  (See Chapter 5 for a definition of modification.)  .
If this assumption proves correct and the projection of 27 new plants  is
                                     9-26

-------
 excessive,  the projected national  annualized cost of the proposed standards
 may be overstated.   However, in Section 9.2 the economic effects of the
 individual  regulatory alternatives were found to be so small  that a more
 detailed study of future growth is not warranted in order to  guide the
 selection of proposed regulatory alternatives.
      New capacity and plants are projected using the two equations:
 NC  =

  where
       1
FP    —
      CU
- CC + RC - 1C
                                                 (1)
       NC.
       FP
       CU
       CC
       RC
       1C
   new capacity
   1988 production
   1988 capacity utilization rate
   current capacity (1981)
   retired capacity,  1984-88
   interim capacity,  1982-83
and
 MP
NP = NC
where
     NP = new plants
     NC = new capacity
     MP = model plant size
The factors and assumptions used and the results of the calculations,
tabulated in Tables 9-11 and 9-12, are discussed below.
     9.1.3.1  Projected Production.  The 1988 production of each of the
five polymers and resins is projected by applying the applicable growth
rate as determined from published projections23'24 to the 1981 domestic
production and net exports and summing.   The values so calculated are
tabulated in Table 9-11.
                                                 (2)
                                    9-27

-------
      The  published  growth  rates  are  developed from historical  trends  and
 end  uses  data,  from communications with  industry  personnel,  and  from
 considerations  of worldwide  developments.   In general,  new production
 capacity  announced  for  Canada  and other  petroleum producing  countries has
 led  some  analysts to conclude  that U.S.  plastics  exports will  decrease as
 worldwide supply and consequent  U.S.  imports  increase.   In addition,  price
 decontrols  have decreased  the  feedstock  advantage of  U.S. producers.
 Assessing the import-export  situation is difficult because it  can  change
 quickly in  response to  governmental  policy  decisions, especially trade
 restrictions.    Domestically, the decline  in demand  in 1979 has resulted
 in the industry's carefully  examining its capacity and adjusting it by
 temporary and permanent closures.
      In order to determine the appropriate  annual  growth rates used in
 projecting  the  1988 production of each of the polymers and resins, growth
 rates published within  the last  three years were  assessed.  To more
 completely  specify  the  projected demand, domestic  growth rates are
 determined  for  both domestic markets  and net  exports.  The annual growth
 rates assumed for each  polymer and resin are  discussed individually below
 and shown in Table  9-11.
     PP.  Published annual growth rates for domestic consumption of PP for
 the period  to 1990  vary from 5.4 to 8.0 percent depending in part on  the
 time at which the projections were made.    An intermediate value of  7.0
 percent is  used as  the  basis for the  projection of domestic demand.
     Published  net  exports growth rates to 1990 were -4.7 percent and  -7.2
 percent.  For this  analysis the higher value  of -4.7 percent is  used  in
 order that  the  projections would be conservative and the NSPS cost
 estimates would not be underestimated.
     LDPE.  A complicating factor in the projections for LDPE is the
 production of the relatively new polymer linear low density polyethylene
 (LLDPE).  For these projections a combined LDPE and LLDPE production  is
considered.  In assessing these projections,  it should be noted that with
but a small capital  investment, LDPE producers can switch to  LLDPE
production.
     Published growth rates for the  combined LDPE and LLDPE consumption to
1990 varied from 5.0 to 6.0 percent.     For this analysis a value of 5.6
is selected as most representative of values given.
                                    9-28

-------
     Annual growth rates for net exports ranged from -19.9 to -1.1
        19
percent.    Due to the variation, a growth rate for net exports of -10.0
is selected for use in the analysis.
     HOPE.  Published annual growth rates for HOPE range from 6 to 10
                                                                  po pc
percent with most analysts projecting values slightly less than 8.  '
A growth rate of 7.9 percent is used.  In assessing the growth of HOPE, the
                                                   27
secondary capacity in LLDPE plants is acknowledged.
     Since the published growth rates for net exports ranged from -15.0 to
+8.4 percent, net exports are assumed to remain constant in this analysis.
     £S  Since less than five percent of the polystyrene production is
exported and imports are negligible, no differential annual growth rates
were determined.  PS is a mature industry and is not expected to show a
high growth rate.  A concensus annual growth rate of 4.0 percent is assumed
                                                                      po
on the basis of published projections ranging from 3.6 to^4.5 percent.
     PET.  Growth in the polyester fiber portion of the PET industry is
predicted to turn positive again in 1984 and to increase at an annual rate
                                28
of about 4 percent through 1990.    Polyester resins for films and bottles
account for about 10 percent of the total PET production.  This portion of
                                                           24
the industry is expected to increase somewhat more rapidly.    For this
analysis no growth is assumed to 1984.  From 1984 on, a growth rate of 4
percent is used.
     The situation for net exports is not clear cut.  The traditional
export outlet for PET fibers has been Western Europe, but since that area
now experiences considerable overcapacity, some limitations have been
placed on imports.  In 1981 exports to the People's Republic of China were
large, but declined in 1982.  Some industry analysts indicate that the
1982 decrease in U.S. exports probably signals a long-term trend.-
However, due to the uncertainties in the import/export situation, net
exports are assumed to remain constant in order to arrive at a conservative
projection.
     9.1.3.2  Projected Capacity.  Projected capacity required in 1988 is
calculated by applying a utilization rate to the projected 1988 production.
                                    9-29

-------
     In order to project the capacity, a capacity utilization rate that
will trigger additional capacity must be assumed.  The capacity utilization
                                                              .29
                                                                 and
rates used are estimated on the basis of published information
                                            30
discussions with an industry representative.    These sources indicate that
because the industry is generally profitable at an overall industry
utilization rate of 85 percent, a utilization rate in the high 80's will
trigger capacity expansion.  Thus, the arbitrary assumption was made that
an 88 percent utilization rate would trigger expansion in the PP, LDPE,
HOPE and PET industries.  Since PS is a mature industry, but one with new
specialty uses being developed, a slightly lower utilization rate of 85
percent was assumed to trigger expansion.  It is well to note, of course,
that individual companies do not base their expansion plans solely on
consideration of overall industry utilization rates, for other
company-specific factors also affect expansion decisions.  However, for
these industry projections the overall industry utilization rates are used.
     The projected 1988 capacities for the five polymers and resins,
calculated from the assumed utilization rates and the projected 1988
production, are shown in Table 9-11.
     9.1.3.3  Retired Capacity.  In calculating the retired capacity, the
useful  life of all plants is assumed to be 20 years.   Therefore, any
capacity built between January, 1964 and January, 1969,  is assumed to be
retired during the five-year period examined in the analysis.   The portion
of this retired capacity that will  be replaced by plants coming under the
potential NSPS varies by polymer process.   For this analysis,  all  20-year
old PP, LDPE, HOPE and PS plants are assumed to be replaced.   For the PET
industry where there are no known plans to rebuild a  portion  of the retired
capacity, it is assumed that only half the retired capacity would be
replaced by 1989.   The assumed retired capacities, then, are  those units
constructed between January, 1964 and January,  1969.   These values,
tabulated from the "Construction Alerts"  published in the April  and
October 1964-1968  issues of Chemical  Engineering, are shown in  Table
9-11.31
     9.1.3.4  Interim Capacity.  Interim capacities are  the capacities
of plants to be completed in 1982 and 1983.   Values for  PS, LDPE,  HOPE
and PS  are taken from the Facts and Figures  of  the U.S.  Plastics
                                    9-30

-------




















§
cn


CD

<-*•>

:c
t—
CO
cn
£

CO
01
a

(O
I—


























•o >i
t— 
IO 1- S "-
4-> -t- (U U
O 3 C 10
1— cr a.
OJ IO
1- U
E >>co
'^ -I— ^?
aj o oj
•M 10 CO

•-H (O t-H
u
•o >>co
££2

•u loco
ai o-cr.
a: lOt-H
u

•a
0) >)

O CO ••-
(U CO U

o t-i a.
s- ro
a. u
c
0
OO N 4->
cn ••- 10
t-HI— S-
1 *
3

•o c
QJ O

U CO -M
OJ CO U
•f-30^ 3
o t-H -a
s- o
a. s-
a.
in

4J 0) 4J V-
§•»->  OJ 4->
3 *J cn
o (O  tf
0) 0
Z Q-
X
0)

o
c
o
O 'f-

t-H . 4J Q.
OO oo tn E


O C
•o o
u

_a
c
o


CO U
cn 3
^H T3
O
s-
a.

IO
^
t-H -r-
CO U
cn 10
t*H Q,
to
u





u
3
-a
o
S-
a_

^
cn
J



"
«,^
cn
_3



^
•^
cn
3





_
>J
cn
-D




M








cn
CD






&^












cn
CD









cn
CD








cn
CD





s-

cn
CD













LO
i —






<— i





LO
t-H
T— 1






O
LO
a>

OJ




§






o
o
IO

OJ



f^

^>
i



o

r-




o








o
CO

M
F-H





o
cn


fH




o
cn
OJ

OJ












Q-
O
O
0
"
1 — 1



o
o^
n




o
CJ
to






o
S

LO




CO
00






o

to

•a-



o

o

1


to

LO




0
LO
•3-







O

C3

CO





o
cn


CO




O
CO
•t
^f









LLJ
Q_
a
	 i
o
1— 1
«g-
•*




ID

r—t




o
CO
CO






0
cn
cn

CO




oo
co






o
t-H
LO

CO





o




cn

i —




0
t-H







O

cn







O
CO
l-H

OJ




o
OJ

OJ









LU

O

o
LO
1 — 1





0
to
t— 1




0
OJ
r— 1






0
•3-
LO
•t
OJ




LO
oo






o
to
t-H

OJ



o

^}-




o

•9-




O
1 —







o

LO
«
t-H





O

to

r— i




O
LO
ro

OJ











oo
Q-
O
CO





O
cn





t*-
O
CO







O
•3-

OJ




oo
CO






0
LO
l-H
91
OJ





' o




 CO
U CO >>
 +->
OO t-H -t—
JZ IO
tn tn cn a.
4-> •»-> 3 10
S- S-' 0 0
o o s-
Q. Q.-C T3
X E *-> 
E . «3- t»-
cn oo i—
cz cn 10
o  ja ^ >,
u 10 cni —
3 1— 3 C
cn cn "d ^-* o o
. . o s-
cn cn s- tn jz -a

-------
  Industry.   Values for PET are taken from Textile Organon.10  The interim
  capacities are shown in Table 9-11.
       9.1.3.5  New Capacity.  New capacities are calculated by using
  equation (1) i.e., the 1981 capacity and the interim capacity are
  subtracted from the projected 1988 capacity, and the retired capacity is
  added to obtain the total  required new 1988 capacity.   A sensitivity
  analysis was not conducted in light of the low level of price effects as
  discussed below.
       In projecting the additional  capacity required in 1988 it should be
  noted that this capacity does not  necessarily translate into new plants;
  indeed, it may not translate completely to other affected facilities, i.e.,
  modifications and reconstructions.   Additional  capacity may be obtained in
  a variety of ways including:
                 debottlenecking existing plants
                 adding a process train or process section
                 converting  facilities now or formerly used to produce
                 other polymers to the production of high demand polymers
                 bringing back on stream plants now on standby or
                 mothballed
                 modifying production processes in order to enhance
                 conversion  rates, e.g., changing the catalyst
                 modifying existing  plants
                 reconstructing existing plants,  and
                 constructing new facilities
       Only new plants and older plants modified  or reconstructed (as
  defined by the Clean Air Act)  will  be affected  by this potential  NSPS and,
  thus, only they are  of interest in  this study.   For purposes of this
  economic impact analysis it was assumed that all  affected facilities  will
  be new model  size plants,  i.e., no  modifications  and reconstructions  are
  included.  (See Chapter 5.)
       9.1.3.6   New Plants.   The projected capacity requires  the  equivalent
_ of 27 new model  size plants or 85  new model  process lines.   Industry
  sources indicate that some  of this  capacity  may be in  the form  of new grass
  roots plants  and some capacity will  stem from adding new process  trains  to
  existing plants.   Sufficient information is  not available,  however, to
                                      9-32

-------
allow the projection of the amount of new capacity that will be added by
new plants and the amount that will be added by additional process trains
at existing plants.
     Another factor to take into consideration in projecting plant growth
by process is that technology is now available to shift both conventional
HOPE and high-pressure LDPE plants over to low-pressure operations that can
make both LLDPE or HOPE.  Thus, data on plants for the two traditional
types of polyethylene, LDPE and HOPE, are becoming increasingly unreliable.
In the next few -years, individual sources of high or low density
polyethylene will probably be difficult to ascertain.  Capacity reporting
could.change to reflect combined polyethylene data, and eventually the
density differentiation could end as have property distinctions among other
         32
polymers.
     Additionally, although some new capacity might be expected to come
from modifications and reconstructions, the analysis makes no projections
of the extent of modifications or reconstruction, i.e., changes in a plant
that would bring an existing plant under the potential NSPS (see Chapter 5).
     For purposes of this economic impact analysis, therefore, the
projected new capacities are expressed in terms of equivalent new model
plants.
     9.1.3.7  Projections by Process.  In order to categorize these
projected plants by process, it is necessary to determine the probable
proportion of these plants that will be devoted to each process.  The
assumptions underlying those determinations, based on discussions with
industry and trade associations representatives, are discussed below.  The
projected number of new plants by process is tabulated in Table 9-12.  (For
ease of comparison with other information given in this document,
capacities are also expressed in terms of model process lines.)
     9.1.3.7.1  PP plants.  Both liquid-phase and gas-phase process
plants will be built in the future.  Companies will construct gas-phase
plants if they already have the appropriate proprietary technology;
however, if they do not, they are more likely to build plants using
liquid-phase process technology.  It was assumed that approximately equal
use will be made of the two processes; therefore, three new gas phase and
                                    9-33

-------






CO
CO
1— 1
CD
O

m
^J.
CO
CT>
r~4

co
CO
UJ
8
CO
co
 CT>
.p t_ 3 r-4 i—l
O CD CD
CD .a z: co jr
•r-3 E -P en
O 3 >
•a o c 03 -•-.
o o -i— o. en
S C 	 1 03 CD
Q- 0



t=.a
03 >,
1 — -p t-
a_ ••- >,
O "" —
i — 03 en
CO O.CD
•a 03
00








CO
CO
cu
o
o
£_
Q.











03
>> "O
-P O) £-
3-r- £_ >,
CU O T- --~
2i 03 3 Ol
Q. CTCD
O3 CU
C-> D^

1 ^
(_
~-
^
c

a.




cricn "^o tocno CMCMCO
r— 1 i— 1







r> co
• •
COCO f-ILO CMCOLO i— 1 O O






LO
•
OLO OLO OOLO r — LOOO
LOCO r~-r^ r-»cor--. COCM





OLO OO OOO LOLOCM
LOO OOLO r-icr>LO r-~r--r-
«-H i-l CM i— 1 CM i— 1








c:
0
^"'•p
{- 3
3 r—
r— O
CO CO
cu
CU £_ QJ CU CU
CO 3 £- CO CO
03 CO 3 03 03 CO
JZ CU COCO JZJZCU 3
CL CO CUCO Q.Q-CO O
03 £_ CU 03 3 >— I OO
-O-C Q.C- "OT3J= CQ-t— I
•i— CX Q. T- T— O. -i—
3 J= 33 -P « •>
crco C7>3 crcrco ccooo
•i— O3 'r— O "l— "1 — O3 O Cu CL.
_JCD 3C_1 _I_JCD OUJUJ




LO O O O
p~. O i— 1 LO
t A
I— 1 I—)



LU UJ
a. a. o
a. a o oo
Q. _J IE CL.






r»- co <— t
i — i







en LO
,-H r-H CD







LO LO O
r-H ^H CM





LO LO O
O O «*
I— I T-t




CO
£_
CL
CU CO
Jta •!—
CO C
•i— -r-
C C(_
3— "O
c
-a cu
c
cu cu
1 —
cu a.
r— pr"
en *i-!
CT r—
'r~ ^
co E

1 1

1— «C
Q 1—




0
co






1—
LU













•
f^~^
CO
1 — 1

•5
•p—
C/l
1
c
"^
•o
a.
o
•r—
•4-1
nj
O^
}
r— •
O
CL

QJ
^~
_o
03
TD
d
03
O.
• X
o CU
>— 1
t-H t.
1 0
O CO
-P CU
• CO
i— 1 t— 1 CO
i— i i cu
1 ^O O
CT> O
CO i-
CU CU Q.
f~) fl Q
03 03 3
1— 1— -p

CD CU O)
CU CU -C
OO CO 1—
O3 J3 O
9-34

-------
three new liquid phase process plants are projected.  The capacity of
these six plants, 765 Gg/yr, is essentially equal to the projected required
new capacity.
     9.1.3.7.2  LDPE plants.  Most new construction will employ gas-phase,
low pressure, Unipol or similar proprietary process technology; however,
the use of the high pressure liquid process cannot be ruled out.  The
consequent projections assume that one high pressure plant will be built
and that the remainder of the capacity added will be low pressure plants.
These assumptions result in one high-pressure plant and five low-pressure
plants having a combined capacity of 1030 Gg/yr, somewhat more than the
1000 Gg/yr required capacity projected.
     9.1.3.7.3  HDPE plants.  Both liquid-phase and gas-phase process
plants will  be constructed.  The gas-phase process plants are cheaper to
build and operate since the process catalyst remains in the polymer.
However, if companies do not have the appropriate, proprietary, gas phase
technology,  they will build liquid phase plants.  The choice between  these
two depends  upon technology licensing costs.  The projections assume  equal
capacity for each process and result in five gas phase plants.
     Further, the liquid phase plants may be either slurry or solution.
Essentially equal capacity was assumed for each process to yield two  liquid
phase slurry plants and three liquid phase solution plants.  The projected
new plants have a combined capacity of 1440 Gg/yr; the projected required
capacity is  1410 Gg/yr.
     9.1.3.7.4  PS plants.  Two types of PS plants exist today.  Production
of bulk plastics, i.e., large quantities of one formulation, is most
efficiently accomplished in a continuous plant.  On the other hand, if a
firm's market requires many grades and specialized materials such as
expandable PS (EPS), then a batch plant is more suitable.  Indications are
that new capacity will probably be added in both forms; therefore, one
continuous plant and EPS process lines constituting the equivalent of one
EPS plant are projected.  The projected process lines result in 345 Gg/yr;
the projected required capacity is 340 Gg/yr.
     9.1.3.7.5  PET plants.  New PET plants may use either the terephthalic
acid (TPA) process or dimethyl terephthalate (DMT) process.  Industry
representatives indicate the TPA process is more likely; therefore, since
                                    9-35

-------
the TPA and DMT model plants are the same size, two TPA process plants  and
one DMT plant are projected.  Of the required 21 model  process lines,  20 are
projected to be for fiber resins and one for industrial resins.  The indus-
trial resins process line is expected to be a TPA line.
     The combined capacity of the projected PET plants  is 320 Gg/yr, slightly
less than the 340 Gg/yr projected.  However, considerations of the current
industry conditions in which plants are being closed down33 make it un-
likely that a fourth new plant would be built by January, 1989.  Overall PET
plant capacity remained constant in 1979 and 1980 and was reduced in 1981
from 2,180 Gg to 2,100 Gg.  Although the PET industry has experienced growth
in the bottle segment and more rapid growth is projected for that segment,  it
constitutes less than 10 percent of the 1981 capacity and was operated at
less than 50 percent of capacity.   Industry representatives indicate that by
modifying present facilities, capacity can be increased to meet demand
without new plants until 1986.34  The projection assumes, therefore, that
some new capacity required for PET  production would be obtained by increases
in capacity that would not be subject to the potential  NSPS.

9.2  ECONOMIC IMPACT ANALYSIS
     Section 9.2 discusses the economic impact analysis methodology and the
potential impacts of regulatory alternatives controlling VOC emissions  from
new  source polymers  and resins manufacturing processes.  Generally speaking
and  as the impact analysis methodology discussion will  show, the potential
economic impacts of  VOC emission controls, though real  and of measurable
impact, will have little significant impact on the  27  projected plants.
     The additional  annualized costs of controls in  1988 (Table 9-16),  the
fifth year of controls, for the 85  projected model  process lines (27 pro-
jected polymers and  resins manufacturing plants) are estimated to  be but $15.5
million  under the combination of the most  stringent  regulatory alternatives
for  each plant.  Thus,  a detailed  "Regulatory  Impact Analysis" as  prescribed
by Executive Order  12291 is not  required.
                                     9-36

-------
      The potential economic  impacts of the  regulatory  alternatives  are
expected to be very minor  in view of the  small price increases anticipated
as a  result of the control costs.  Assuming that the incremental costs can
be passed forward by each  of the new source facilities, price increases
required by the facilities for most of the  alternatives would be less than
five-tenths of a percent.  The maximum price increase  required to fund any
alternative would not exceed 3.3 percent.   Because of  the minor increases
required by these plants,  no significant  economic impacts of the potential
NSPS  are expected.
      Section 9.2.1 will discuss this study's revenue and price impact
assessment methodology.  It will be followed by 9.2.2—-the economic impacts
projected by that methodology.
9.2.1  Economic Impact Assessment Methodology:  Revenue and Price
      As explained in Chapter 8, each regulatory alternative is associated
with  incremental levels of investment, operating costs, recovery credits,
and VOC emission control levels.  The incremental costs alter the total
cost  structure of each affected plant and potentially affect its pricing,
profitability, and economic viability.  In this analysis, the expected
economic effects of these regulatory alternatives are analyzed.
     The methodology used in this analysis of expected economic impacts
on the polymers and resins industry involves first a quantitative financial
analysis utilizing a model plant approach to compare prices before and
after implementation of the standards, and second, a qualitative analysis
of expected industry and macroeconomic effects.  The price impact
methodology is based on a simplified price impact analysis.  This
methodology calculates the revenue and price increases required by model
plants to maintain the same net present values (NPV) before and after the
installation of emission-control equipment.   This revenue calculation
relies on a single derived equation that requires several  types  of input
data for each 'model  plant.  The macroeconomic analysis consists  of an
evaluation of aggregate industry and macroeconomic impacts based on an
understanding of the market structure and dynamics in the polymers and
resins industry, the background of which is  discussed in Section 9.1.
     The purpose of this analysis is to determine the revenue increase that
exactly offsets emission control costs so that the NPV of the model  plant
remains constant or the NPV of the incremental  cash flow is zero at the
stated weighted average cost of capital.   In this analysis, capital  costs,
operating and'maintenance costs, recovery credits,  investment life,  income
taxes, and inflation need to be taken  into account.   A nominal  discount
                                    9-37

-------
rate is used for the analysis because most equity capital cost data are
available in nominal terms.  The use of a nominal discount rate requires
that revenues and operating costs be properly inflated.  The revenue
increase that the analysis calculates is expressed in base-year dollars, in
this case June, 1980.  The required revenue increase is converted to a
required unit price increase by dividing the revenue by the annual sales
volume.  This step requires the assumption that annual sales volume is
constant, indicating perfectly inelastic demand.
     The derivation of the basic formula for the price impact analysis
requires the following assumptions:
     o    Emission control investments have -zero salvage value.
     o    No differential inflation occurs among the cost or revenue items.
     o    The weighted average cost of capital and the marginal income tax
          rate remain constant during the life of the investment.
     o    Depreciation is based on the Tax Equity and Fiscal Responsibility
          Act of 1982.
     o    A 10 percent investment tax credit is applicable on emissions
          control investment and is realized in the year following the
          investment.
     o    NPV of the model plant remains constant at the stated weighted
          cost of capital.
     o    Where control equipment lifetime is less than that of the process
          unit, replacement investment in control equipment can occur
          automatically and the basic formula derived below will  hold.
          Theoretically, a problem could arise if the last control
          equipment replacement outlives the process unit itself.  However,
          such a situation is not predictable and is unlikely to enter into
          the initial decisions to invest in control equipment and to
          adjust product price to recoup that investment.
     The derivation of the basic formula of the methodology is presented in
the context of the standard definition of NPV:
                      CF.
     NPV =
-I.
                                    9-38

-------
where,
     NPV = net present value (in base year dollars)
     y   = time period
     n   = investment life
     CF  = projected operating cash flows in period y
     d   = nominal interest rate or cost of capital used for discounting
     I   = investment (period y=0)

     Cash flow, CF, is defined as revenues, R, less operating and
maintenance costs, OM, less income taxes, T.
              - OMy - Ty
                                                      (2)
     Revenue and operating costs will inflate at the same rate, thus,
nominal revenue (and O&M) equals the product of the inflation factor,
(l+inf)y, and constant dollar value in y=0 dollars.  That is,
and,
     Ry  = (R0}
     OMy = (OMQ) (l+inf)y
                                                      (3)

                                                      (4)
where, inf = annual inflation rate.
     Income taxes, T , are computed on the basis of nominal  current
income.  Income taxes are defined to equal the tax rate, t,  times taxable
net income.  The investment tax credit, -ITC, is subtracted  directly from
income taxes.  Taxable income equals revenue less operating  and maintenance
expense, and less depreciation.  Thus,
V t(Ry -
                           - ITC
     Substituting the income tax Equation (5) into the cash flow Equation
(2) yields the following expanded expression for cash flow.
                                    9-39

-------
              - OMy - Ty
           Ry - OMy - t(Ry - OMy - Dy) + ITCy
                    - tRy + tOMy + tDy + ITCy
           (l-t)Ry - (l-t)OMy + tDy + ITCy
                                              (6a)

                                              (6b)

                                              (6c)

                                              (6d)
     Further, by substitution of Equations (3) and (4) into (6d), operating
cash flow, CF , may be expressed by:

     CFy = (l-t)R0(l+inf)y - (l-t)OM0(l+inf)y + tDy + ITCy            (7)
     Net present value can now be defined in terms of revenue, operating
and maintenance costs, and income tax effects, including depreciation and
the investment tax credit.

     By substitution of Equation (7) into Equation (1),
                      (l-t)R0(l+inf)y - (l»t)OM0(l+inf)y + tDy + ITCy
     NPV = -IQ +  Z
                 .y=i
                                                         (8)
     Equivalently, each term under summation can be summed individually so
that
n   (l-t)R0(l+inf)y
                                                                      (9)
                                        (l-t)OM0(l-Hnf)y     n   ITCy + tDy
NPV = -ln +  Z
        0   y=i
                     -  E
                       y=i
     By imposing the constraint that incremental NPV=0, it is possible to
solve for the annual revenue requirement, R , (in base year dollars) that
is equivalent to incremental emissions control investment, I , and annual
                                    9-40

-------
 operating and maintenance  costs,  OMQ.   Rearranging  terms  on  the  right
 hand side of Equation  (9)  so  that revenue,  operating  costs and investment
 related  items are  grouped,  and  setting  NPV=0 yields
                                                                       (10)
     n
0 =  X
    y=i
(l-t)R0(l+inf)y
                                  (l-t)OMo(l+inf)y
                            -  £
                            y=i
ITC
                                                 y=i
                                                                        tD
     Next, the terms  related to emissions control investment are isolated.
Assuming a 10 percent investment tax credit in the year after investment
and five-year ACRS rates, the investment, investment tax credit, and
depreciation terms on the right hand side of Equation (10) can be expressed
as a product of a constant, TAXF, and I .
                                                                      (11)
                    =  I
                                              .22t
                                              .211
                 1  -
                                  (1+d)
                                 .21t
                                   (1+d)2     (1+d)3
                                                                   .21t
                                                                  (1+d)
                      I0 {TAXF}
                                 (1+d)'
                                                           (12)
where TAXF is the sum of terms in brackets.  TAXF is a constant that can be
repeatedly applied to different investments so long as the tax rate,
interest rate, and ACRS life remain the same.
     Next, Io {TAXF} is substituted into Equation (10), the constant
terms are moved outside their summations, and RQ is isolated on the left
hand side of the equation.   R  is then expressed as
                                    9-41

-------
     Ro =
                                (l+inf)y
               (l-t)OMQ    Z
                         y=i
                                (l+inf)y
                  (1-t)
                         y=i
                          {TAXF}
                                                                 (13)
     Simplifying Equation (13) results in
                         TAXF
     Ro=OMo
           (l+inf)y
                   (1-t)   z
                          y-i
                                           (14)
     Equation (14) is the equation used to calculate the annual revenue
increase that exactly offsets NSPS capital and operating costs so that the
NPV of the firm remains constant.
     The above equation is applicable in those cases in which the total
investment has a single economic life of n.  When the investments cover
capital equipment with different lives, the equation needs to be expanded
to reflect capital costs associated with each life.  For example, when a
portion of the investment has a life of 15 years and another portion 10
years, R  is then expressed as:
=  OM,
                        TAXF
                      + I
      15  (l+inf)y
(1-t)   Z
      y=l  (1
                                              10
TAXF
                                                       10 (1+inf)y
                                                  (1-t) z       —
                                                       y=l
(15)
where, I-,c = investment with an economic life of 15 years
       I,  = investment with an economic life of 10 years.
     The investment lives of the various equipment items are indicated in
Chapter 8.
                                    9-42

-------
      The  price  increase  that  the  model  plant  needs  to  realize  this  annual
 revenue increase  is
                 Ro
           PI  =  —
                 Qn
                                                            (16)
where,    PI = the unit  price  increase  in base year dollars
          R  = the required annual revenue increase in base year dollars
          QQ = annual sales volume in units
Finally, the percentage increase in unit price is calculated using the
formula
        PI
PPI =   —  x 100
          o
                                                                      (17)
where,    PPI = percent unit price increase

           PI = unit price increase in base year dollars

           PQ = pre-emission control unit price in base year dollars

     Equations (15), (16), and (17) were used to calculate the revenue and
price effects of emission controls on the affected polymers and resins
plants.
9.2.2  Economic Impact of VOC Potential NSPS Regulatory Alternatives -
       Polymers and Resins
     Industry and macroeconomic impacts of the potential NSPS regulatory
alternatives include effects on prices, profitability, plant viability,
employment, and other economic measures.   The price analysis here is
quantitative.  The other characteristics  are analyzed qualitatively on the
basis of the industry price analysis.   A  more involved analysis of
                                    9-43

-------
aggregate effects is not warranted, since the quantitative price impacts
described in the next section are minor.
       9.2.2.1  Required Revenue and Maximum Price Increases.   On the plant
level, the net present value (NPV) method is used to estimate  required
revenue increases and maximum percentage product price increases for each
model and regulatory alternative.  These measures are based on the assumption
that incremental emission control costs are fully passed forward.
       The model plants represent the new facilities that are  expected to  be
constructed during the analysis' five-year period.  The model  process lines
and model plant product type, plant capacity, and processes were discussed in
Chapters 6 and 8.  The numbers of new plants by type, size, and process were
projected in Section 9.1.  The number, capacity, production, and annual sales
for the projected new model polymers and resins process lines  are summarized
in Table 9-13.
       The model plants represent 13 product-process combinations.  Twenty-
seven affected plants consisting of 85 process lines are projected for the
first five years, i.e., 1984 through 1988.  The sales level for each model is
the product of the average 1980 polymer price and the model plant's production,
The production levels of the model plants are based on a capacity utilization
rate of 85 percent, the assumed optimal rate for new facilities.
       The average prices for plant products were derived from one of two
sources.  Prices for PP, LDPE, HOPE, and PS were determined from data con-
tained in the "Plastics Resins Domestic Merchant Sales" table in Facts and
Figures.35  That table reflects both quantities of sales and net 1980
dollar values which represents actual selling prices after deductions for
discounts; consequently, the division of the dollar value by the quantity  of
the products sold provides a valid measure of typical plant average annual
prices.  PET prices were not available from the above source.
       The PET price represents the price of fiber processed within the PET
plants.  Prices of PET used for bottles and film, which amounts to only a
small portion of the total PET produced, are not available in published
sources.  Determining an average annual price for the PET fibers plant was
more involved than determining that for the other plants, because the PET
plant produces a mix of product fibers of varying prices.  Chemical and
Engineering News (C&EN)36 reports 1980 prices for three polyester fibers:
                                     9-44

-------








OO
UJ
t^.
1 — 1
	 1

oo
oo
LU
o
0
^v
CL, -— s
co
LU =t
Q _!
0 _J
"Z. 0
Q
CO
I-" 00
co en
LU t — 1
C£
O LU

*~O
co • — •
cc.
LU 00
2L. CO
>- en
— 1 i-H
o
CD
t 1 — 3
c*^ C^
Q~
CO JZ
C_> I—
i— i
oo oo
i — i en
CU «— I
LU
i— i—
O —1
^^ ^— H
cC CQ
_C
O

_J
^^
^—
o
(— H
1—

o;
LU
Cu
O


CO
r-H
en

QJ
r—
-O
re
1—









t_
OJ --^
r— CCO O
re QJ o
3 tO C O
c , — _Lj fas
e=C re ^— f
00
u !•— -
QJ 01
O ^
"si w
o «^-^

c-
0 QJ
-Q T- C •
r— 4-> -i- — • .
fO CJ 	 1 ^T
3 3 • 05
E -0 I- _-
E O 
•< £_ CL
Q.

O
•i—
fO QJ *""**
IM 4-3 6^
•i- re —'
"^ °^
±?
ID


^>
f— - +J
03 T- £_ QJ — x
a u a> c en
C 03 O-'i- CI3
CO. — I — '
^c re
o


-o  QJ
QJ QJ O C
•r-3 ^1 Q; -r—
0 E •»- — 1
t- 3 q-
Q. H: ea:











to
t/5
QJ
O
O
£_
^










4— )
u
3
-a
o
£_
CU


O LO
LO CM
r-. co
en o
CM CM


o o
r^. r>--






LO CO
CM en
•51- CM







LO LO
oo oo








O LO
LO CO






en en


















QJ
to
re
-c a>
CL to
re
~O -E
•t- a.
3
O" tO
•i- re
—1 CD





CL.
a.




O LO
en r-
r-. CM
CO CM
=3- LO


CM CM
CO 00






LO 00
en co
LO IO







LO LO
oo oo








O LO
r--. i--.






•5t O
i— i


















C- QJ
3 i_
in 3
in to
CU to
c, CD
CL £_
0.
f"
0) S
•t- 0




LU
Q_
Q
— 1




LO LO OO
o ^- 10
o i— i oo
r-- o 0
<3- CM LO


en en en
r~- r~- r~-






LO LO OO
en LO oo
LO CM tO







LO LO LO
OO OO 00








O O LO
!"*«• oo r*-«






vo en o










c:
O
b'43
£_ 3
3 i —
•— 0
in in

CD CD
to to
re re
r- t— Q}
a. CL to
re
T3 *"C3 ^~
•r- T- a_
3 3
O" O" to
•i- -r- re
_l _1 CD



LU
^
c^
'




oo oo .— i
»— H ^J* O^
LO oo r--.
CM tO •— 1
OO OO r-H


CM OO OO
o r~-. r---
1 	 1 t 	 1 T 	 1





en oo oo
r— i r— I to
CO CM







LO LO LO
OO CO 00







LO
r-~ LO oo
OO CM






CM CM OO



















in
3
o
3 I— 1 00
C Q- I-H

| * A A
c: oo co
O Q- Q-
C_> LU LU




QJ
CO
Q-




o o o
LO LO O
en en to
CM CM O
CM CM 00


CD CD O
co co oo
1 — 1 1 — I f — 1





00 OO O
CM CM r->-
1 — 1 1 — 1 1 — 1







LO LO LO
00 OO CO








LO LO O
t-H fH CM






r-^ oo ,-H
1 — 1



to
£_
CD
C_ f—
QJ to
-C -i-
to c:
•1— -I—
c: H-
4— "O
1—
"O QJ
c~
CD CD

QJ CL
"01+3
g^ ,__
•r- Z3
 E

1 1

H~ ^C
^r* Q^
Q 1—




1—
LU
a.




















.
»
£U
o
QJ
CL
LO
00
a>
o
4-3

-a
QJ


to
to
re
CD

re

to
QJ
re
c:
o
4-5
re
N
•t—
^w
•r—
4_>
3

^^
4-5
O
re
CL
re
o

QJ
J2
re
4-5
•r-
i[
0
£_
Cu

>,
^~
r—
re

CD
£1
O)
CD
re


























w
-2
re
£_

f
0

,J
03
NI
•f—
r—
•r-
4->


V)
QJ
E

>>
»^-
o
re
CL
re
o

r— «
re
3
C
re

to

re

cr
QJ
c: .
0
•r-

o
3

o

Ct,

, ,
to
3
C

•a:









CO
r>

<<-
O
t ^
re
-^
~*~^
to
CD
•^
4—
p^
< [

QJ
^\
i T 2
a
[__\

CD
re

•r-
[ ^
. to
QJ

to
•i—
oo
O-
LU

«l
o
QJ
O
£_
Cu
CD
^

• «\
to
QJ
U
•r—

CL
{t_
0

c:
o
«f—

re
.1^
£_
QJ


t
O
t»_

4-5
X
QJ


QJ
QJ

O





































v
CD

•r—
t.
Q.
to
g
•r—

C
O

J ^
O
3
T3
O
£_
CL
'i 	
"re
3
C
£
re

to
re
O"


to
QJ

"re
to

r^~
03
3
C

e^
"O





e
11^^l
CO
1— H
_
+^
-r-

C
*^*
"O

03
i— i
• o


0
•i—
re
£
QJ
CL
•r-
|

to
O
0

QJ
re

^— -^
oo
Q-
LU

CD
£
£_

to
4?
o
CL

QJ
i —
f*i
re
T3
re
Q.
x
QJ


O
14-
to
QJ
to
to
CD
o
O

Cu



• 1^

QJ

1—
QJ
9-45

-------
staple, textured filament, and feeder filament.  To establish a price for
the plant's product mix, the analysis assumed that its sales were similar
to those of total shipments shown in the 1977 Census of Manufactures  for
these three types of PET fibers (on a percentage basis).  Consequently, the
model plant product price used to determine annual sales is the average
price of the three types of fibers reported in C&EN weighted by the product
shipments of these fibers as reported in the Census of Manufactures.
     Required revenue increases for polymer and resin model process lines
and percent required maximum price increases for process lines (and also
model plants) to comply with potential NSPS regulatory alternatives are
shown in Table 9-14.  These measures are shown with the associated process
line estimated investment, operating and maintenance costs, and recovery
credits for each regulatory alternative.  For each model, Regulatory
Alternative 1 is baseline control.  Other regulatory alternatives reflect
different levels of control and associated cost as described in Chapter 8.
     For these computations, the corporate income tax rate on marginal
income was assumed to be 50 percent.  Inflation was assumed to be 8 percent
per annum.  The nominal weighted cost of capital was estimated to be 14
percent.  The after-tax cost of capital reflects a 12.5 percent nominal
interest rate, a capital structure of 35 percent debt and 65 percent
equity, and a 17.9 percent nominal return on equity.  Because of the
capital structure and tax effect, the size of the interest rates accounts
for only a small portion of the total cost of capital, and changes in the
rate result in relatively minor changes in the cost of capital.  For
example, changing the interest rate by 5 percentage points (10 to 15
percent) changes the cost of capital by only one percentage point.  The
interest rate is based on a 2 percent real risk-free component which was
estimated from an analysis of the historical real yields on U.S. Treasury
securities (10 year maturities) and an 8 percent inflation component.  The
product of these two components (1.02 X 1.08) represents the nominal
risk-free interest rate—10 percent.  This rate is increased by 25 percent
(to approximate the interest on a Baa (Moody's) corporate bond) in order to
incorporate an appropriate risk component.  The capital  structure and cost
of equity are based on an analysis of data reported in Value Line for nine
chemical firms involved in the manufacture of plastics.     The cost of
                                                op
equity was derived by the dividend yield method.
                                    9-46

-------
Table 9-14.  POLYMERS AND RESINS MODEL PROCESS LINE CONTROL COSTS3 AND MAXIMUM PRICE INCREASES FOR MODEL  PLANT
                     BY  PRODUCT, PROCESS, AND REGULATORY ALTERNATIVE  (JUNE, 1980 DOLLARS)
Model process lines F
by product and process i

PP
o Liquid phase


o Gas phase
LDPE
o High pressure

,

o Low pressure



HOPE
o Liquid phase slurry

o Liquid phase solution

o Gas phase



PS
o Continuous





o EPS, PI





o EPS, IS





PET
o DMT





o TPA
- Single end finisher
— Multiple end finishers





iegulatory
ilternative


2
3
4
2

2
3
4
5
2
3
4
5

2
3
2
3
2
3
4
5

2
3
4
5
6
7
2
3
4
5
6
7
2
3
4
5
6
7

2
3
4
5
6
7

2
2
3
4
5
6
7
Investment0


28.4
332.9
351.2
28.4

23.6
237.1
383.9
1,292.3
38.0
56.2
231.8
386.9

28.4
170.5
28.4
635.1
38.0
56.2
231.8
386.9

38.0
39.6
39.6
39.7
40.4
40.9
28.4
249.1
289.4
677.6
712.0
738.3
15.6
40.4
235.5
390.8
409. 3
427.9

1.5
1.5
1.5
1.5
1.5
240.8

239.3
12.5
29.0
50.9
65.0
81.8
128.3
Recovery
credit^
	 / tinnn

10.3
10.3
10.3
10.3

7.8
7.8
7.8
7.8
15.5
15.5
15.5
15.5

10.3
10.3
10.3
10.3
15.5
15.5
15.5
15.5

15.5
18.4
18.5
18.6
18.6
18.6
10.3
10.3
10.3
10.3
10.3
10.9
3.4
3.4
3.4
3.4
3.4
3.4

0.3
0.3
0.4
0.4
0.4
2.0

1.4
2.4
4.9
7.4
8.6
9.8
12.3
0&Hd
\\ _ .
t) — _ 	 _ —
12.0
64.9
81.9
11.9

9.8
57.1
89.8
223.6
16.3
33.4
80.5
119.2

11.9
42.8
12.0
309.4
16.3
33.4
80.5
119.2

16.3
17.1
17.1
17.2
17.3
17.4
12.0
68.2
258.4
402.6
1,102.5
1,105.8
6.3
25.2
70.7
109.5
129.6
149.5

0.7
0.7
0.7
0.8
0.8
73.4

67.5
2.3
5.5
9.9
12.8
16.3
26.2
Required
revenue
increase8


11.5
110.7
130.5
7.4

9.9
89.6
144.6
418.0
8.4
27.8
> 101.6
160.9

7.4
59.9
7.5
397.1
8.4
27.8
101.6
160.9

10.4
8.5
8.4
8.4
8.6
8.8
7.5
97.3
292.8
496.0 '
1,200.6
1,207.3
6.2
29.0
104.2
166.6
189.1
211.3

0.6
0.6
0.6
0.6
0.6
39.2

38.5
1.9
5.1
10.3
14.1
18.9
33.4
Maximum
price
1ncreaser


0.04
0.37
0.44
0.04

0.02
0.18
0.30
0.86
0.02
0.05
0.19
0.31

0.02
0.13
0.04
1.97
0.02
0.05
0.19
0.31

0.03
0.03
0.03
0.03
0.03
0.03
0.02
0.26
0.79
1.35
3.26
3.28
0.05
0.25
0.88
1.41
1.60
1.79

0.00
0.00
0.00
0.00
0.00
0.17

0.17
0.01
0.02
0.03
0.05
0.06
0.11
                                                9-47

-------
a   Control  costs  and  recovery credits  are  based on data contained in Tables
    8-21 to  8-31b.

b   Investment for each alternative represents the sum of the incremental
    capital  costs  (over baseline, Alternative I) for the various types of
    equipment as aggregated by economic life (costs were rounded to the
    nearest  $100).  For example, the investment for Regulatory Alternative 5,
    LDPE low pressure  is $386,900 (shown above as 386.9).  This is obtained
    from Table 8-24 and represents the sum  of incremental costs of $1,250 for
    equipment with an  economic life of 2 years, $9,200 for equipment with an
    economic life of 6 years, $283,000 for  equipment with an economic life of
    10 years, and $93,000 for equipment with an economic life of 15 years.
    The sum  of the incremental costs indicated above also equals the dif-
    ference  of the total cost under baseline of $133,900 and the cost under
    Alternative 5 of $520,800.

c   Except for PET, recovery credits are obtained directly from the tables
    in Chapter 8; incremental costs are not involved since there are no
    credits  under baseline.  However, for PET there are recovery credits under
    baseline; consequently, incremental credits are used.

d  O&M represents the incremental  annualized costs (excluding capital
    recovery which is  incorporated in the formula shown below).  In the
   example, the incremental OHM cost for LDPE low pressure is $119,200 which
    is the difference between the annualized cost of $84,800 under baseline
    (105,200 less 20,400) and $204,000 under Alternative 5 (286,400 less
    82,400).

e  The calculation of the required revenue increase is based on Equation
    (15) (from the text).  With the substitution  of the parameters indicated
   in the text,  Equation (15) simplifies  to
        R0 * OM0 + .577Ix + .230I2 + .15213,
   where

        R0  s Required revenue increase
        OMg = O&M less recovery credit
        II  * Incremental  capital  costs of equipment with economic
                life of 2 years.
        12  * Incremental  capital  costs of equipment with economic
                life of 6 years.
        13  = Incremental  capital  costs of equipment with economic
                life of 10 years.
        14  * Incremental  capital  costs of equipment with economic
                life of 15 years.

   For Alternative 5 of LDPE low pressure, the calculations  are:

        R0  = (119.2 - 15.5) + .577(1.3)  + .230(9.2) + .152(283.0) +  .115(93.0)

            =< 160.9
   The calculation of the  maximum price increase  is  based on Equations
   '16) and  (17)  from the  text.   By  substituting  Equation (16) into Equation
    17), the latter simplifies to:
        PPI
 Rq
P6T3o
100 or Required revenue increase
                              do!Tar  sales  volume
                                                        100
   The dollar sales  volume  for each model  process  line  is shown in Table
   9-13.   Percentage price  increases  are the  same  for both the model process
   line and  the  multiple  process  line model plant.
                               9-48

-------
     Maximum price increases for all alternatives of the models are
relatively insignificant except in the few cases indicated below.  The
maximum price increase for Alternative 2 of all model plants except PET is
less than .05 percent; for PET plants the increase is .4 percent.
Increases for the remaining alternatives are less than 1 percent for all
model plants except the HOPE liquid phase solution model and the EPS
models.  The price increase under Alternative 3 for the HOPE model is 2.0
percent; for the EPS models the price increases range up to 3.3 percent
under Alternatives 6 and 7 for the EPS, PI process line.
     9.2.2.2  Expected Price and Profitability Impacts.  The elasticity of
demand determines the extent to which cost increases can be passed forward
to the consumer in terms of higher prices.  An elastic demand implies that
sales revenue will be reduced if costs are passed forward.  In such a case,
the producers can be expected to absorb some of the costs in order to
minimize the impact on profits."  On the other hand, an inelastic demand
implies that costs can be passed forward.  In general, the demand for
plastic products tends to be inelastic in the major end-use markets
discussed in Section 9.1.  Although, as discussed in Section 9.1,
substitutes exists for plastics in many of the markets (e.g., building,
packaging, and transportation), there are no competitively priced
substitutes available in the short run.  In a number of the markets
(primarily consumer and institutional) plastic products represent a small
portion of the end user's budget.   Both of these determinants suggest an
inelastic demand.  In addition to these determinants, other factors exist
in the industry that would affect the ability of the industry to pass costs
forward.  For example, the extensive vertical integration that exists
within the large petroleum and chemical firms facilitates substantial
cost pass-through.  All of these demand determinants and market factors
indicate that costs imposed industry-wide can very likely be passed
forward.  However, in the case of the potential NSPS costs, the ability of
the 27 projected new plants to pass the costs forward may be limited in
some instances because production in these plants constitutes a minor
                                    9-49

-------
portion of the total industry's output.  The potentially impacted PP, LDPE,
and HOPE plants constitute about 30 percent of their respective projected
1988 polymer and resin capacities, while the PS and PET plants constitute
less than eight percent of theirs.  Consequently the PS and PET plants
could be expected to absorb a portion of the cost of the regulatory
alternatives.  (This will depend also on the relative costs of production
of the new and existing plants.)  Even though these plants do absorb the
costs, the costs are for the most part insignificant as shown above and the
impact on profitability would be insignificant.
     9.2.2.3  Other Economic Effects.  Since no significant impacts on
prices or profits are anticipated, the potential NSPS is not expected to
have any significant effects on the industry or economy.  Although capital
availability may be of considerable concern to the industry in its efforts
to modernize, the small incremental costs of NSPS controls are not expected
to have any effect on the generation of the required capital for the
regulatory alternatives.  Little or no postponement of plant construction
is expected to occur.  The potential NSPS is not expected to have
significant aggregate effects on output, employment, competition, industry
structure, productivity, or foreign trade.

9.3  POTENTIAL SOCIOECONOMIC AND INFLATIONARY IMPACTS
     The socioeconomic and inflationary impacts of the potential  NSPS are
examined in terms of the fifth year costs and benefits to society of each
regulatory alternative, the impacts on small facilities, the level  of
inflation, and the balance of trade.
9.3.1  Fifth Year Costs and Benefits
     The total annualized costs to society for each regulatory alternative
in the fifth year of implementation, 1988, are presented .in Table 9-15.
Projected regulatory costs are customarily summed for the fifth year to
facilitate comparison of cost impacts among various environmental
standards.  The need for and effects of regulations are reconsidered every
four years.  Costs were determined as the product of the annualized  costs
(or net annualized costs where there are product recovery credits) and the
projected number of plants expected to be affected in the fifth year.
These costs to society are based on a 10 percent real  social  interest rate
                                    9-50

-------
Table 9-15.   FIFTH YEAR  NET ANNUALIZED COST3 TO SOCIETY OF REGULATORY
        ALTERNATIVES BY  MODEL  PROCESS LINE  PRODUCT AND PROCESS
                         (JUNE,  1980 DOLLARS)
Net
Model process lines Regulatory annual ized
by product and process alternative cost
($1000)
PP
o Liquid phase


o Gas prase
LDPE
o High pressure



o Low pressure



HDPE
o Liquid phase slurry

o Liquid phase solution

o Gas phase



PS
o Continuous





0 EPS, PI





o EPS, IS





PET
0 DMT





0 TPA
- Single end finisher
- Multiple end finishers






2
3
4
2

2
3
4
5
2
3
4
5

2
3
2
3
2
3
4
5

2
3
4
5
6
7
2
3
4
5
6
7
2
3
4
5
6
7

2
3
4
5
6
7

2
2
3
4
5
6
7

7.7
110.1
129.8
7.7

7.2
89.2
145.8
411.7
8.9
28.6
104.2
165.7

7.7
61.6
7.7
403.8
8.9
28.6
104.2
165.7

8.9
7.1
7.0
7.0
7.2
7.3
7.7
99.8
296.0
503.4
1,208.4
1,215.3
6.2
28.9
106.1
107.2
192.9
215.4

0.6
0.6
0.6
0.6
0.7
110.6

104.6
2.0
5.4
10.9
14.8
19.8
34.8
Total
Number of annual ized
process cost
lines ($1000)

9 69.3
990.9
1,168.2
9 69.3

4 28.8
356.8
583.2
1,646.8
10 89.0
286.0
1,042.0
1,657.0

6 46.2
369.6
9 69.3
3,634.2
10 89.0
286.0
1,042.0
1,657.0

2 17.8
14.2
14.0
14.0
14.4
14.6
2 15.4
199.6
592.0
1,006.8
2,416.8
2,430.6
3 18.6
86.7
318.3
321.6
578.7
646.2

7 4.2
4.2
4.2
4.2
4.9
774.2

13 1,359.8
1 2.0
5.4
10.9
14.8
19.8
34.8
These costs represent the costs  of the  regulatory alternatives over
and above the baseline as described in  Chapter 8.
                             9-51

-------
as opposed to the 14 percent nominal weighted cost of capital used in deter-
mining the price changes (Section 9.2).  The 10 percent rate represents costs
to society and is generally used in benefit-cost studies.  The highest
alternative cost for each of the model plants is shown in Table 9-16.  The
total of these costs, which provides a view of the upper boundary of aggre-
gate cost to society, is $15.5 million.
     Executive Order 12291 specifies that a regulatory action, to the
extent permitted by law, must not be undertaken unless the potential  benefits
to society from the regulation outweigh the potential costs to society.  An
exhaustive benefit-cost analysis is not appropriate here because the poten-
tial NSPS will not constitute a major rule within the meaning of the Execu-
tive Order since the cost of the standards and their overall impacts on the
economy are minor.  A qualitative enumeration of the benefits follows.
     The potential standards will reduce the rate of VOC emissions to the
atmosphere.  These compounds are precursors of photochemical oxidants,
particularly ozone.  The EPA publication, AIR QUALITY CRITERIA FOR OZONE AND
OTHER PHOTOCHEMICAL OXIDANTS (EPA-600/8-78-004, April 1978), explains the
effects of exposure to elevated ambient concentrations of oxidants.  (The
problem of ozone depletion of the upper atmosphere and its relation to this
standard are not addressed here.)  These effects include:
     o  Human health effects.  Ozone exposure has been shown to cause in-
        creased rates of respiratory symptoms such as coughing, wheezing,
        sneezing, and shortness of breath; increased rates of headache and of
        eye and throat irritation; and physiological  damage to red blood
        cells.  One experiment links ozone exposure to human cell  damage
        known as chromosomal aberrations.
     o  Vegetation effects.  Ozone reduces citrus, grape, cotton,  and
        other crop yields by damaging leaves and plants.  The reduction has
        been linked to both the level and duration of ozone exposure.
                                    9-52

-------
      Table  9-16.  UPPER BOUNDARY OF TOTAL ANNUALIZED FIFTH YEAR COST
                        TO SOCIETY (JUNE, 1980 DOLLARS)a
Model process lines
by product and process
PP
o Liquid Phase
o Gas Phase
LDPE
o High Pressure
o Low Pressure
HDPE . .
o Liquid Phase Slurry
o Liquid Phase Solution
o Gas Phase
PS
o Continuous
o EPS, PI
o EPS, IS
PET
o DMT
o TPA
- Single end finishers
- Multiple end finishers
TOTAL
Number
of
process
1 i nes

9
9

4
10

6
9
10

2
2
3

7
13
1
Regulatory
alternative

4
2

5
5

3
3
5

7
7
7

7
2
7
Total
annual ized
cost
($)

1,168,200
69,300

1,646,800
1,657,000

369,600
3,634,200
1,657,000

14,600
2,430,600
646,200

774,200
1,359,800
34,800
15,462,000
a  Annualized fifth year costs  are based on  the highest  alternative  cost
   for each model  process line  as shown  in Table 9-15.
                                     9-53

-------
     o    Materials effects.  Ozone exposure has been shown to accelerate
          the deterioration of such organic materials as plastics and
          rubber (elastomers), textile dyes, fibers, and certain paints and
          coatings.
     o    Ecosystem effects.  Continued ozone exposure has been shown to be
          linked to structural changes in forests—the disappearance of
          certain tree species (Ponderosa and Jeffrey pines) and the death
          of predominant vegetation.  Continued ozone exposure, hence,
          causes a stress on the ecosystem.
     In addition to the evidence of the physical and biological effects
enumerated above, a reduction of VOC emissions is likely to improve the
aesthetic and economic value of the environment through:  (1) beautifica-
tion of natural forests and undeveloped land through increased vegetation;
(2) increased visibility; (3) reduced incidence of noxious odors; (4)
increased length of life for works of art, including paintings, sculpture,
architecturally important buildings, and historic monuments; (5) improved
appearance of structures, sculptures, and paintings, and (6) the improved
productivity of workers, especially farm laborers.
9.3.2  Impacts on Small Facilities
     The Regulatory Flexibility Act (Public Law 96-354, September 19, 1980)
directs Federal agencies to pay close attention to minimizing any
potentially adverse impacts of a standard on small businesses, small
governments, and small organizations.  This standard will  have no known
effects on small governments and small organizations.  It  may affect some
small businesses but the impacts will be few and minor.  Essentially, all
firms that will be required to comply with the standard either are not
small businesses, or are subsidiaries of large firms.  The businesses that
are expected to own or operate polymers and resins producing plants during
the first five years following proposal are those currently in the field.
(See Table 9-2).  The Small  Business Administration (SBA)  classifies small
businesses in SIC 2821 (PP, LDPE, HOPE, and PS) as those with 750 or fewer
employees and in SIC 2824 (PET) as those with 1,000 or fewer.   These levels
were set as criteria for extending SBA loans and related assistance (13 CFR
Part 121, Schedule A).  Only two of the firms in Table 9-2 are believed to
                                    9-54

-------
 be  small  businesses, and both of  these produce PET.  Because of the
 competitive  nature of the  industries and the  relatively high levels of
 capital  required to construct polymers and resins manufacturing plants, it
 is  unlikely  that any businesses constructing  new plants would be small or,
 if  small, would remain small.  Furthermore, as the analysis in Section 9.2
 explains, any potential adverse economic impacts on new plants, regardless
 of whether they are owned or operated by large or small businesses, would
 be minor.
 9.3.3  Other Impacts
     No other impacts are expected as a result of the costs of the
 regulatory alternatives.   There should be no significant pressure on the
 level of inflation.   As the construction of the new source plants will not
be adversely affected,  neither will the employment level  in the industry.
No effects are expected on the balance of trade.
                                   9-55

-------
9.4  REFERENCES FOR CHAPTER 9

1.   Kline Guide to the Chemical  Industry.  Fair-field, New York,
     Charles H. Kline & Co., Inc., 4th Edition,  1980.   pp. 150-161.
     Docket Reference Number II-I-53.*

2.   The Society of the Plastics  Industry, Inc.   Facts and Figures
     of the Plastics Industry, 1982.  p. 21.   Docket Reference
     Number II-I-81.*

3.   Big-volume Chemicals'  Output Fell Again  in  '81.  Chemical  and
     Engineering News.  6>0_:12.  May 3, 1982.   Docket Reference
     Number II-I-89.*

4.   U.S. Department of Commerce.  U.S. Industrial  Outlook, 1982.
     Washington, D.C., U.S. Government Printing  Office, January 1982.
     p. 120.  Docket Reference Number II-I-84.*

5.   U.S. Department of Commerce.  Census of  Manufacturers, 1977.
     Washington, D.C., U.S. Government Printing  Office, July 1980.
     Docket Reference Number II-I-60.*

6.   Reference 4, p. 316.

7.   Reference 2, p. 1-3.

8.   Standard & Poors.  Industry Surveys:  Chemicals,  November 5,  1981.
     p. C22-C26.  Docket Reference Number II-I-80.*

9.   Reference 2.  p. 12-67.

10.  Textile Economics Bureau, Inc.  Textile  Organon.   December 1982.
     p. 249-251.  Docket Reference Number II-I-105.*

11.  Polyester Fiber Makers Pick Up the Loose Ends.  Chemical  Business.
     Chemical Marketing Reporter.  222(14):9-16.  April 6, 1981.   Docket
     Reference Number II-I-71.*

12.  Business.  Chemical and Engineering News.  58^:10.  December 1,  1980.
     Docket Reference Number II-I-64.*

13.  Reference 12.  56_: 12-16.  September 4, 1978.   Docket Reference
     Number II-I-42.*

14.  Reference 12.  ^6_:10.   December 4, 1978.  Docket Reference
     Number II-I-44.*

15.  Reference 12.  j59_:13-22.  August 31, 1981.   Docket Reference
     Number II-I-74.*

16.  Reference 12.  59_:11.   November 2, 1981.  Docket Reference
     Number II-I-79.*
                                  9-56

-------
 17.   Reference 2.   p.  8.   1980-1982.   Docket  Reference Numbers  II-I-54,
      -66,  and  -81.*

 18.   Reference 12.   57j 12-16.   September  3, 1979.   Docket Reference
      Number  II-I-48.*

 19.   Reference 12.   58^:12-16.   October 6, 1980.  Docket Reference
      Number  II-I-62.*

 20.   Reference 12.

 21.   Robert  Morris  Associates.  Annual Statement Studies.  Plastics
      materials and  synthetic resins.   1976-1982.  Docket Reference
      Numbers IT-I-23,  -29, -36, -46, -55, -67, and -83.*

 22.   Plastics  World.   40_:61.  April 1982.  Docket Reference
      Number  II-I-87.*

 23.   Predicasts Forecasts, 1982 Annual Cumulative Edition.  July 29, 1982.
      p. B-236-248.   Docket Reference Number II-I-94.*

 24.   Predicasts, Inc.   Industry Study:  Thermoplastics to 1995.  T67.
      Cleveland, Ohio.  March 1982.  p. 3.  Docket Reference Number II-I-85.*

 25.   Reference 22.  40:64.

 26.   Chemical  Profile.  Chemical Marketing Reporter.  222:54.
      December  13, 1982.  Docket Reference Number II-I-106.*

 27.   Reference 22.  40:59.

 28.   Polyester Makers Say all Signs Point to Market Rebound in '83,
      Though Current Outlook is Dim.  Chemical  Marketing Reporter.
      222^:3-30.   November 29, 1982.  Docket Reference Number II-I-104.*

 29.   Reference  12.  60:17.  May 24, 1982.   Docket Reference Number II-I-90.*

 30.   Telecon.  Hanamaker, 6., DPRA to Symuleski, Richard,  Standard Oil
      Company of. Indiana, Chicago,  IL.  Chemical  Manufacturers Association
      representative for Polymers and Resins  Study,  December 6, 1982.
      Industry Utilization Rates.  Docket Reference  Number  II-E-52.*

 31.   Chemical Engineering.  April  and October 1964-1968.   Docket Reference
      Numbers II-I-110 through II-I-119

 32.   Reference  12.  60_:11.  September 6,  1982.  Docket Reference
      Number II-I-97.*

 33.   Reference  12.  60^:7.  May 31, 1982.   Docket Reference Number II-I-91.*

34.   Reference  11, p. 16.
                                  9-57

-------
35.  The Society of the Plastics Industry, Inc.   Facts  and  Figures
     of the Plastics, Industry, 1981.  p. 8.   Docket Reference
     Number II-I-66.*  '

36.  Reference 12.  58:9-10.

37.  Arnold Bernhart and Company, Inc«   The  Value Line  Investment
     Survey.  New York, November 19, 1982.  p.  1238-1251.   Docket
     Reference Number II-I-103.*

38.  Weston, O.R. and E.F. Brigham.  Managerial  Finance.   Hinsdale,  111.
     The Dryden Press.  1977. p. 617.  Docket Reference Number II-I-27.*
*References can be located in Docket Number A-82-19  in  the U.S.
 Environmental  Protection Agency Library,  Waterside  Mall,
 Washington, D.C.

                                  9-58

-------
             APPENDIX A
EVOLUTION OF THE PROPOSED STANDARDS
                 A-l

-------
                              APPENDIX A
                 EVOLUTION OF THE PROPOSED  STANDARDS
     The purpose of this study was to develop new source performance

standards for the polymers and resins industry.   Work  on the  study  was

begun in January 1980 by Energy and Environmental  Analysis,  Inc.,  (EEA)

under the direction of the Office of Air Quality Planning and Standards
(OAQPS), Emission Standards and Engineering Division (ESED).   In June 1982,

this study was transferred from EEA to Pacific Environmental  Services,
Inc. (PES).  The decision to develop this standard was made  on the

recommendation of EEA based on a source category survey study.  In

performing the standard development, previous EPA study reports,

responses to requests for information under Section 114 of the Clean

Air Act, plant visit information, and industry comments were  used.

     The following chronology lists the important events that have
occurred in the development of background information  for the new  source

performance standards for the polymers and resins industry.

     Date                                   Activity

August 30, 1979
May 19, 1980



July 15, 1980


August 15, 1980


August 19, 1980



August 28, 1980


September 2, 1980
Pullman Kellogg Report,  "Polymer Industry Ranking
by VOC Emissions Reduction that would occur from
New Source Performance Standards."

Meeting of CPB, EMB, SDB.'EAB, and EEA representa-
tives to discuss recommendations for the development
of the NSPS for polymers and resins industry.

Meeting of EPA and EEA representatives to discuss
scope of the polymers and resins NSPS.

Plant visit to USS Novamont's polyproylene facility
at LaPorte, Texas.

Plant visit to Soltex Polymer Corporation's
high-density polyethylene facility at Deer Park,
Texas.

Plant visit to Phillips Chemical Company's high-
density polyethylene facility at Pasadena, Texas.

Meeting between EEA, EPA, and Union Carbide in
South Charleston, West Virginia.
                                   A-2

-------
    Date

September 2, 1980



September 9, 1980



September 11, 1980


September 15, 1980



September 16, 1980


September 22, 1980



September 29, 1980




October 2, 1980


October 9, 1980



November 7,  1980




November 21, 1980




December 16, 1980
                      Activity

 Union Carbide Corporation's response to Section 114
 request for  information on Union Carbide's low density
 polyethylene facility at Port Lavaca, Texas.

 Mobil Chemical Company's response to Section 114
 request for  information on Mobil's styrenics plant
 at Santa Ana, California,,

 Plant visit to Union Carbide Corporation's low-density
 polyethylene facility at Port Lavaca, Texas.

 Union Carbide's response to information requested
 at September 2, 1980, meeting on low pressure and
 high pressure polyethylene.

 Plant visit to Mobil Chemical Company's styrenics
 facility at Santa Ana, California.

 USS Novamont's response to Section 114 request for
 information on USS Novamont's polypropylene facility
 at LaPorte, Texas.

 Gulf Oil Chemicals Company's reply to EPA's September 2,
 1980, letter requesting information on Gulf's gas
 phase, high-density polyethylene process at
 Orange, Texas.

 Plant visit to Tennessee Eastman Company's polyester
 resin facility at Kingsport, Tennessee.

American Hoechst Corporation's response to Section 114
 request for information on American Hoechst's
 polyester resin facility at Greer, South Carolina.

Tennessee Eastman Company's response to Section 114
 request for information on Tennessee Eastman's
 poly(ethylene terephthalate) facility at Kingsport,
Tennessee.

Northern Petrochemical Company's response to
Section 114 letter request for information on
Northern Petrochemical's low-density polyethylene
 plant at Morris,  Illinois.

Meeting of CMA,  EPA, and EEA representatives on
status of the polymers and resins NSPS development.
                           A-3

-------
   Date

December 18, 1980



March 17, 1981



July 6, 1981


August 24, 1981



September 1981


November 17, 1981
February 26, 1982


March 4, 1982




March 26,  1982



April 2, 1982


April 13,  1982


April 13,  1982
                     Activity

Phillips Chemical Company's response to EPA request
for additional information on Phillips Chemical's
high-density polyethylene facility at Pasadena,  Texas.

Standard Oil Company (Indiana) response to Section 114
request for information on Amoco Chemicals Corporation's
gas phase polypropylene process.

Meeting of CMA, EPA, and EEA representatives on
status of the polymers and resins NSPS development.

Meeting of EPA and EEA representatives on status
of and future plans for cost analysis for polymers
and resins NSPS.

Model Plant parameters package was sent to CMA for
comments.

Meeting of CPB, EAB, SDB, and EEA representatives
to review the proposed model plant parameters,
process baseline control and regulatory alternatives,
and fugitive  regulatory alternatives for the
polymers and  resins NSPS.

BID Chapters  3 to 6 were sent to the industry for
comments.

Meeting of Allied Chemical Company, EPA, and EEA
representatives to discuss vacuum system design
alternatives  to  reduce air emissions in polyester -
TPA process plant.

Meeting of CPB,  EAB, SDB, and EEA representatives
to discuss the regulatory approach and recommendations
for the standard.

Phillips Chemical Company's  comments on draft BID
Chapters 3-6.

Texas  Chemical Council's comments on draft BID
Chapters 3-6.

Union  Carbide Corporation's  comments on draft BID
Chapters 3-6.
                                 A-4

-------
    Date
                     Activity
April 14,  1982


April 14,  1982


April 14,  1982

April 15,  1982


April 15,  1982


April 15,  1982

April 19,  1982

April 19,  1982

June 1982

June 30, 1982


July 1, 1982


September 14, 1982


September 15, 1982
September 20, 1982
September 20, 1982
September 22, 1982



September 29, 1982


November 2, 1982
Gulf Oil Chemicals Company's comments on draft BID
Chapters 3-6.

Northern Petrochemical Company's comments on draft
BID Chapters 3-6.

Tennessee Eastman's comments on draft BID Chapters 3-6.

Allied Fibers and Plastics' comments on draft
BID Chapters 3-6.

Chemical Manufacturing Association's comments on
draft BID Chapters 3-6.

USS Corporation's comments on draft BID Chapters 3-6.

DuPont's comments on draft BID Chapters 3-6.

Monsanto's comments on draft BID Chapters 3-6.

Project transferred from EEA to PES.

Plant visit to Soltex Polymer Corporation's high-densit'
polyethylene facility at Deer Park, Texas.

Plant visit to DuPont Corporation's high-density
polyethylene facility at Orange, Texas.

Plant visit to Gulf Oil Chemicals Company's polystyrene
facility at Marietta, Ohio.

Plant visit to Monsanto Plastics and Resins Company's
polystyrene facility at Port Plastics (Addyston),
Ohio.

Trip to Texas Air Control Board, Austin, Texas.
114 Letter to Fiber Industries  requesting information
on emissions and their control  at Fiber Industries'
Polyester resin plant, Salisbury,  N.C.

Letter to Gulf Oil  Chemical requesting  additional
information regarding Gulf's polystyrene plant,
Marietta,  Ohio.

Plant visit to Fiber Industries'  polyester  facilities
at Salisbury, North Carolina.

Letter to  Tennessee Eastman requesting  information on
ethylene glycol recovery  and costs  at polyester  resins
plants.
                                A-5

-------
      Date

November 2, 1982


November 10, 1982


November 11, 1982



November 22, 1982



November 23, 1982


December 16, 1982



February 18, 1983


March 16,  1983


March 30,  1983


April 19,  1983


April 20,  1983
 April  20 and 21,
 1983
 April  26,  1983


 April  27,  1983


 May 4; 1983


 May 12, 1983
                          Activity

Letter to Cosden Oil  requesting information on expandable
polystyrene industry.

Letter from Cosden Oil responding to 11/2/82 letter
requesting information on expandable polystyrene industry.

Letter from Gulf Oil  responding to 9/22/82 letter
requesting additional information on polystyrene
facility at Marietta, Ohio.

Letter from Tennessee Eastman responding to 11/2/82
letter requesting information on ethylene glycol
recovery and costs at polyester resins plants.

Meeting with Fiber Industries to discuss poly(ethylene
terephthalate) manufacture and emission control.

114 Letter sent to four manufacturers of expandable
polystyrene requesting information on production,
emissions, and emission control.

Letters from ARCO and BASF Wyandotte  responding to 114
letter on expandable  polystyrene.

Mai lout of material  for National Air  Pollution Control
Techniques Advisory  Committee meeting.

Letter from Cosden Oil Company responding to  a 114
letter requesting information on  expandable polystyrene.

Letter from Union Carbide  Corporation commenting on
draft NSPS regulation and  background  document.

Letter from DuPont commenting  on  draft NSPS regulation
and  background  document.


Trip to  Union Carbide Corporation,  South  Charleston,  W.V.,
to discuss control of emergency  releases  from high
 pressure,  LDPE  tubular  reactors.

 Presentation  made to the  National Air Pollution  Control
 Techniques Advisory  Committee.

 Letter  from  DuPont commenting  on  draft  NSPS regulation
 and background  document.

 Letter  from  American Hoechst responding to a  114 letter
 requesting information  on expandable polystyrene.

 Letter  from Monsanto commenting on  draft NSPS regulation
 and background  document.
                                  A-6

-------
    Date

May 19, 1983


May 19, 1983


May 23, 1983



May 31, 1983

June 2, 1983


June 16, 1983


June 21, 1983



June 24, 1983


June 30, 1983
      /             Activity

Meeting with Allied Fibers to discuss their concerns with
the draft  NSPS  regulation.

Letter from Hercules  Inc., commenting on draft NSPS
regulation and  background document.

114 Letter to Gulf Oil Chemicals requesting information
on emissions and emission control from high pressure,
LDPE manufacturing.

Letter from Dow Chemical commenting on draft NSPS.

Meeting with Gulf Oil Chemicals to discuss control of
emergency releases from high pressure LDPE autoclave reactors,

Letter to USS Chemicals requesting additional information
concerning intermittent emissions from their plant.

Letter from Union Carbide Corporation transmitting
safety report excerpts concerning high velocity vents
for emergency releases from UNIPOL reactors.

Letter from Texas Chemical  Council  commenting on the
draft NSPS.

Letter from Gulf Oil  Chemicals transmitting TACB operating
permits for Gulf's Cedar Bayou plant.
                            A-7

-------

-------
                              APPENDIX  B
                 INDEX TO ENVIRONMENTAL CONSIDERATIONS

     This appendix consists of a reference system which  is cross-indexed
with the October 21, 1974, Federal Register  (39 FR 37419) containing
the Agency guidelines for the preparation of Environmental Impact
Statements.  This index can be used to  identify sections of the document
which contain data and information germane to any portion of the
Federal Register guidelines.
                               B-l

-------
       co
       zr
       O
       an
       LU
       o
       1—4
       CO

       O
       CJ
       O
m     cf

       CC
•^i*     *«C~
^^*     ^.—





4-)
E
O
S
3
O
0
0

E
0
•r—
-| ^
re
£••
o
q-
E
i — i

T3
E
3
O
S_
CT

O
re
CO

cu
_E
40

E
•r—
-E

•r—
3
E
O
•i —
4-3
re
CJ
o
_i






E
•i —

-a
cu
IM
•r-
c .
re
^
^3
CO

CU
s_
re

CO
cu
>
•i —
i *
re
E
S-
CU

r~~
re

>.
i~
o

re
3
CO
CU
S-

cu
.E
r-










•
CM
1
i— -t

JZ
CO

0
S-
5

t— 1
1
t— 1

CO
cu
CO
re
O-

«t
t— 1
•
t— 4

E
o

•Jj
o
cu
CO

t-4

S-
cu
4-^
CL
re
.E
CJ










CO
•a
s.
re
•o
E
re
4-5
CO

"S
CO
0
a.
o
s_
CL

CU
.E
,{_3

S-
o
q-

co
•r—
CO
re


>1
s-
0
4J
•4_3
re

CO

cu
e~
h-






«— 1
|
CM

CO
CU
CO
re
CL

n
1->.
•
CM
E
O
•r—
4->
CJ
CU
co

A
CM

S-
cu
4>^
Q.
re
-E
CJ

E
•r—

-a
cu •
N «*
il CM
re
B _E
e co
3 3
CO O
S_
CO .E
•r- 4->
•O
E
re

A
P^

n
ID

>•> **
S-OO
o
4*^ n
re CM
(^
3 CO
coi-
cu cu
^- 4^
Q.
cu re

4->CJ

COE

o

re cu
CO
CO CO
0.3
•r- O
.E CO
CO'r-
E-O
o
•r- CU
4-* S—
re re

CU co
S- E
o
CO-r-
34->
O O
•r- re
s-
re >>
> 0

cu cu
•E CO
l— re












cu
t —
^_3

>•


T3
O)
O
CU

<{ —
«


s_
[ ^
CO
3
•c

•p-

cu

.(_>

q-
Q

f-
O
•r-
CO
CO
3
O
CO
•I —
-a

 E
O. O
re E
.E O
0 0
cu
E ~~--
•F- CO
CO
-a cu
CU E

E CO
CU 3
CO .Q
cu =
S-
CL CU
g—
CO 4->
•r—
q-
co o
cu
> CO
•r— r~~
4>^ Br~
re re
E 4-5
i- CU
CU Q
[ %
P«
re •
1— !
Coo
o
4J E
re o

— 2 t >
CO O
cu cu
s- co



t-H
1
cr>

CO
cu
CO
re
CL

««
CT>

S-
cu

CL
re
o

E
•1 —

-a
cu
[ ^
£^
O)
co
cu
S-
CL

cu
S-
re

>^
i_
1 ^ •
CO 10
3 CO
•a i
E a->
•i—
c~
CU CO
-E 3
4-> O
C.-
'[ e-"
O 4->







•a
cu

O E
CU T-

q- TD r- 1
re cu i— i
i/i i
CU CO CO
^2 3
O CO
O CO CU
4-> -r- CO
-a re
CO CL
cu cu
CO S- «<
co re oo
cu •
o cu co
o >
S_ T- -o
Q-4-> E
re re
•a E
E 5- CM
re cu •
4J CO
CO i —
co re co
O E
S- >> 0
3 C/T-
0 0 4->
CO 4-> O
re cu «
O r— CO OO
•r- 3 CO
q- co « i
•t— CU CO CO
0 S_
a> s- -E
Q. CU CU CO
CO .E 4-> 3
4J Q. 0
cu re s-
c~ >sjc; <~
1— .a o 4->




r—
O
^.
4-^
<^
O
CJ

*;
O

^^
4->
•f—
f—
•r-
re

•r~
re

re

•o
E
re

CO
cu
CL
>^
4-^

CU

.(_>

C"
O

E
O
-M
re

s^.
o
q—
E
1 — I

A
CM
•
*^1"

-o
E
re

j— i
•
*^d"

CO
E
0

[ '
CJ
cu
co

n
•*

S-
cu

CL
re
r*
CJ

E r***
•i- CO
i

cu
> -E
•r- CO
CO 3
O
CO S-
•r- .E

COt— 1
O 1

"o
E CO
J= CU
O CO
co re
4-> Q-
UJ
o.
Q-
^
s_
O CO
4>> | ^
re E
•— cu

co a>
d) 4-^
csi re
4^
coco
E
•r- 4J
i- 0 —
re re CD
CL Q.I — i
CU E •*
S— H~< r*^
a. co
s- "re a:
0 4-> U_
q- E
a> CT>
CO E CO
CU E^--
E O
•r- S-
l — «r—
cu >
•a E
•r- LU
3
CD E
O
>">»r— •
O 4->
E O
CU «=C
CO



>^
5—
o

re

3
CO
CU
S-

q-
0

>^
S-
«3
g=
E
CO

•a
E
re co
cu

E •!-
3 4J
o re
S- E
CO S-
^£ CO
0 4->
re r—
CQ re




CO
-o
S-
re
•a
E
re
4J
CO

CO
E
•I—
CO
O
Q.
O
£_
CL

S-
o
lf~
CO
•r—
CO
re


>^
S-
o
4^
3
4J
re

00




                                                       o
                                                       •(->
                                                       re
                                                       r—
                                                       3
                                                       CO
                                                       CU


                                                       S-
                                                       cu
                                                       _E
                                                       o
                                                       4J CO
                                                          E
                                                       CLO
                                                       CO O
                                                       E re
                                                       o
                                                      •r- >>
                                                      •(-> CJ
                                                       re E
                                                      r— CU
                                                       cu co
                                                      a; re
 CO
 3
-o
 O
 CU
                                                                                     CO
                                                                                     CU
 o
 re
4-

-a
 E
 re

 CO
 cu
 co
 co
 CU
 O
 o

 CL


"8
-i->
 o
 CU
                                      o
                                      E
                                     JE
                                      O
                                      CU
                                      O
                                      S-
                                     •»->
                                      E
                                      O
                                      o

                                     q-
                                      o
                                     re
                                     re

                                     «=c
                                                                     B-2

-------
T™"
c
J'
•a i
co to
oo
oo .c:
3 CD
O 3
00 O
•i— &—
TD -E
4-5
CO
S- LO
ta «— i
oo to
cu
> 00
•i- cu
4-5 cn
03 03
E Q.
S-
CU •*
4-5 CM
r— •
to to

S- O
O T-
-1^ J^>
03 O
i — CU
3CO
cn
CU «
i- to

00 S-
3 a>
0 4-5
•1- Q.
S_ 03
03 _E
> O

««
•
>•} i*^
S-
O oo
4-> E
03 O
i "H—
3 1 ^
cn o
cu cu
s- co

00 «
3 r~~-
o
•f— £„
S- CU .
fQ 4—^ ^J"
> Q.CM
03 1
cu .E r^
-C C_3
4-5 _E
E CD
4- •!- 3
0 0
00 CU .E
JjJ _| ^ * S
O E
03 CO CM
a. oo i
•r- S_
Q. oo
r— a;
OS CU CD
4-5 S- 03
E 03 CO.
CU
E OO •>
E CU CO
0 > .
S- *r— f^
•r- 4-5
> 03 JC
E E CD
OJ S- 3
CU 0
CU 4J S-
JE I— J=
1— 03 4-5










>.
S-
o

03 "

*3
CD S- •
CU CO CO
1- 4-5 CO
Q- I
oo 03 r-»
3 -E
Otic-
•r- cn
S- E 3
03 -I- O
> i-
•O .E
> CO
S- 4->
O Q.
4-5 O3 •
03 -E <—
i— CJ> tO
3 1
cn E co
CU -i-
-a CD
00 CO 3
3 OO O
O oo S-
S- O 4->
03 OO
> V- 1— 1
CO CO
-E CU
4-> S_ 00
03 CU
4- CD
O 00 O3
4-5 0-
OO E

OO i — «— '•

I— CO
03 r-
E CU E
03 T5 O
O T-

OO O
O E CU
O O CO


"^i
C7^
•
OO
E
O

^/ 1 ^
O O
4-5 CU
to co

3 •>
CD CO
CO
S- S-
co
00 4-5
3 a.
o 03 •
•r- j= un
5- C_3 LO
03 1
> E ai
•r—
QJ _E
^: -a cn
4-5 CO 3
4-5 O
4- E S-
O CU _E
00 4-5
00 CU
4-5 S- tO
O ELCO
03 1
CL a> cn
•r- O3 OO
CU
O oo cn

E > Q-
O T-
E 4-5 •>
O O3 CO
O E •
CU S- CT>
QJ
CU 4-5 -o
-E 1 — E
1— 03 03
O
03

O 00
03 E
i— CU


CU 4-5
C£ 03
4-5
CD CO
E
•i- 4J
S- O* — •
O3 O3 CT>
a. Q.I-H
a> E-*
S- i— i 1^.
Q- CO
r—
s- 03 a;
O 4-5 U_
4- E
co cr>
oo E co
CO E — •
E O
•r- S-
co >
•0 E
•i- UJ
3
o
^) >r~
O 4-5
c o
CO •<
cn
o.
•1™
o
'E
o
E
O
o
CO
oo
T3 CU
E >
03 -r-
4-5
« 03
>>E
CD S_
i- CO
CO 4-5
E i—
CO 03

«v
1— ^^
03 t-
4-> O
E 4-5
d) 03
^ 1 ' '
E 3
O CD
&- CU
•i- S_
E 4-
UJ O















00
4-5
O
03
Q.
E
•r~

^_
03
4-5
E
CU
E

O
i-
•i —
E
UJ








 00
4-5
 O
 03
 Q.
 CU
 E
UJ
 CO
 E
 03

4-5
 00
 O
O
              oo
             4-5
              O
              tO
              Q-
              E
                           O
 o
 E
 O
 o
UJ
                         B-3

-------

-------
          APPENDIX C
   EMISSION SOURCE TEST DATA
              AND
FUGITIVE EMISSION SOURCE COUNTS
            C-l

-------
              APPENDIX C:  EMISSION SOURCE TEST DATA
                AND FUGITIVE EMISSION SOURCE COUNTS

     The purpose of this appendix is to describe the test results of
flare and thermal incinerator volatile organic compounds (VOC) emissions
reduction capabilities and the equipment inventory used in the development
of fugitive emission estimates for the background information document
(BID) for this industry.  Background data and detailed information
which support the emission levels, reduction capabilities and the
fugitive emission model are included.
     Section C.I of this appendix presents the VOC emissions test data
including individual test descriptions for control of process sources
by flaring.  Sections C.2 and C.3 present the VOC emissions test data
for control of process sources by thermal incineration and vapor
recovery system, respectively.  Section C.4 consists of comparisons of
various VOC test results and a discussion exploring and evaluating the
similarities and differences of these results.  Section C.5 contains
the available fugitive emission inventories for the polymers and
resins industry and discusses the selection of a single model plant to
best represent the characteristics of the fugitive emissions from the
industry.
C.I  FLARE VOC EMISSION TEST DATA
     The design and operating conditions and results of the five
experimental studies of flare combustion efficiency that have been
conducted were summarized in Section 4.1.1.1.1.  This section presents
more detailed results of the first flare efficiency emissions test to
encompass a variety of "non ideal" conditions that can be encountered
in an industrial setting.  These results represent the first phase of
an extended EPA study.  More recent test results are found in Docket
Reference Number II-B-94.
                              C-2

-------
      The aforementioned  experimental  study was performed during a
 three week period  in  June 1982 to determine the combustion efficiency
 for both air-  and  steam-assisted  flares under different operating
 conditions.  The study was  sponsored  by the U.S.  Environmental  Protection
 Agency and the Chemical  Manufacturers Association (CMA).  The test
 facility and flares were provided by  the John Zink  Company.   A total
 of  23 tests were conducted  on  the steam-assisted  flares and  11 tests
 on  air-assisted flares.   The values of the following  parameters were
 varied:   flow  rate of flare gas,  heating valve of flare gas,  flow rate
 of  steam,  and  flow rate  of  air.   This section describes the  control
 device and the sampling  and analytical  technique  used and  test results
 for  the  steam-assisted flare.
 C.I.I   Control Device.
     A John Zink standard STF-S-8 flare tip was used  for the
 steam-assisted flare test series.  This  flare tip has an inside diameter
 of 0.22 m  (8 5/8 in.) and is 3.7  m (12 ft.  3.5 in.) long with  the
 upper  2.2  m (7 ft 3 in)  constructed of stainless  steel  and the  long
 1.5 m  (5 ft 0.5 in) constructed of carbon  steel.  Crude propylene was
 used as the flare gas.   The maximum capacity  of the flare tip was
 approximately  24,200 kg/hr  (53,300 Ib/hr)  for crude propylene at
 0.8 Hach exit velocity.  Variations in  heating valves  of flare  gas
were obtained  by diluting the propylene  with  inert nitrogen.
 C.I.2  Sampling and Analytical  Techniques
     An extractive sampling system was  used to collect  the flare
 emission samples and transport these  samples  to two mobile analytical
laboratories.   Figure C-l is a diagram of  the  sampling  and analysis
system.  A specially designed 8.2 m (27  ft) long sampling probe was
suspended over the flare flame by  support  cables from a  hydraulic
crane.
     Gaseous flare emission samples entered the sampling system via
the probe tip,  passed through the particulate filter, and then were
carried to ground  level.   The sampling system temperature was maintained
above 100°C (212°F) to prevent  condensation of water vapor.  The flare
emission sample was divided into three possible paths.   A fraction of
the sample was  passed through an EPA Reference Method 4  sampling train
to determine moisture content of the sample.  A second fraction was

                                C-3

-------
                                      e
                                      IO

                                      •o
                                      c
                                      ro
                                      e.
                                      E
                                      10
C-4

-------
 directed through a moisture removal cold trap and thence, into a
 sampling manifold in one of the mobile laboratories.  Sample gas in
 this manifold was analyzed by continuous monitors for (L, CO, (XL, NO
 and THC on a dry sample basis.  A third sample was directed into a
 sampling manifold in the other mobile laboratory.  Sample gas in this
 manifold was analyzed for S02 and hydrocarbon species on a wet basis.
      Data collection continued for each test for a target period of
 20 minutes.   Ambient air concentrations of the compounds of interest
 were measured in the test area before and after each test or series of
 tests.
      Flare emission  measurements of carbon monoxide (CO), carbon
 dioxide (C02),  oxygen (C02),  oxides of nitrogen (N0x),  total  hydrocarbons
 (THC) and sulfur dioxide (S02) were measured by continuous analyzers
 that responded  to real  time changes in concentrations.   Table C-l
 presents a summary of the instrumentation used during the tests.
 C.I.3   Test  Results
     Twenty  three tests were  completed on the steam-assisted  flare.
 Table C-2 summarizes  the results of these tests.   The results  indicate
 that the combustion  efficiencies of the  flare plume  are  greater  than
 98 percent under varying condition  of  flare  gas flow rate, including
 velocities as high as  18.2  m/s (60  fps)  flare gas, heat  content over
 11.2 MJ/m3  (300  Btu/scf), and  steam flow rate below  3.5  units of  per
 unit of  flare gas.  The concentrations of  NO  emissions  which were
 also measured during  the testing ranged  from 0.5  to  8.16 ppm.
 C.2  THERMAL  INCINERATOR VOC EMISSION  TEST DATA
     The  results  of six  emission tests and one laboratory study were
 reviewed  to evaluate the  performance of  thermal incinerators under
 various operating conditions in  reducing VOC  emissions from the different
 process waste streams generated  during the manufacture of polymers and
 several  synthetic organic chemicals.  The variable parameters under
which the  incinerator tests were  performed include combustion temperature
and residence time, type of VOC,  type and quantity of supplemental
fuel, and feedstocks  (solid, liquid, and gaseous waste streams).  The
test results, which are summarized in Table C-3S in combination with a
theoretical analysis  indicate that high VOC reduction efficiencies (by
weight)  can be achieved by all new incinerators.
                                C-5

-------


cu
CD "a.
t— « O
f— c
OO •!-
LU S-
h- Q-

LU cn
ce c
€^ •!-"
1 4.)
LU tO
S_
eC CU
s: D-
cj o

^O£
1—4 g—
_i to cn
<-< EC
1— •r-T-
=3 S- 4->
O_ ca
z: i-
O CU
>-« a.
h- 0
z:
LU
o;
H-
oo

H-4

O i-
^^ cu
- O.
_J

^y
z
oo
oo
HH
§

• ,__
T-( O)
ci •§
^~(
cu
r— T3
JD C
03 <*S
|— -
cu

CO
y~(














cu
o

cu
0
W)
cu
c:

pz
3

•r—
E
cu
C"
O











E
O.
CL

LO
CM
I
0

















X
0




o
.— (
,_
cu

5

c
o
S-
4^
O
CU
^_
LU

O
E
SZ
cu
f^
1—
T3
fd
O!
5-
4-> CU

c c e o
o o o a.
•r- -I- (0 •!- X
4->4->-i-C C>, 4-> CU
Q- Q. to O OS- C
S- S- >> ••— .p- 4J CU i —
OOr— 4J 4JCUS-4-> CU
t/>to(o O-S-i — CL
O C C 0 OCO«=C3
•o-o o o^rj=o+-> 10
CU CU O -i- •!- CL O -r- to i — O
S- S- S- (0 -r- CU O
03 o34-» CU CU CU O-i-E EE
s- s- o E E E4->os- os-
<4-t- CU 03 (0 (0 OCUCU S-CU
E Ci— i— i— •— _C S- -C J= -C
I— l >-H LU U- LU U- Q.Q-1 — CJ4J




1 i
o
O
CO
^ CO
CL CM
Q- S
a. -c LU o
O CL E Cu o 4->
o a. E o o
o ss o a. o CM o
•» ^^ LO o o *3" i — i o
t-HLOCMr-i eCLOLOLOl O
1 1 1 1 --» 1 1 1 O CO
OOOO Z O O O ^f 1

to cu
CU S-
C •!- 3 CU
00 4J S_
J3 CU 03 3
S- CL E S- 4->
03 OO O CU 03
O f-* T- Q- S_
O C CM 4-> E CU
S- O O T3 O CU CL
•a _a oo cu cu h— E
>, S_ «—- CU S- CU
-E 03 CUT- 4-> H-
O S- OO Q C
i — o cu cu cu
03 S- O T3 T3 -r- J3
CM 4J T303CC-Q O
OO CMO >>S_-r-T-E S-
o o o (— in i— 3 3  CU
00 CU CL
CD O X LU i— C 3
O O eC 	 LU O O
O O to -C T— O
CMCMO 03CuLO 5-
1— 1 i— i • VO T— 1 O> COO CU
Q-Q.CU'— I i— lOeCO J=>i
C i-H CM4-» OO S-S- |— i—
03 03 >> 03 4J O) J2
J3^3T34J CUE >,03J= 03E
•r- -i— 
-------
> OJ OJ
> ••- o
J U S-
) -r- QJ
• M-Q_
                             *" trj
                            CO    ^
                                                   QJ   CT   cn  
                                                                                                OJ   QJ


                                                                                                5   (O
                                                                                                •o
                                                                                                QJ
                                                                  (U   QJ
                                                                  4->  -U
                                                                  to  uo
                                                                                       o  ^i-

                                                                                       ro  ro
                                                                                       CO  CM  .-H   Lf)
                                                                                       cr.  ^o  «s-   oj
                                                                                       ^r  tn  cn   i-<
                                                                                 m   *a-  in   CD  i—


                                                                                 CM   CM  OJ   CM  CM
                                                                                 CM   CM  CO
                                                                                 CM   CM  CM
                                                                                 CM   CM  CM  CM   CM
                                                                                      CD  "^  i— i
                                                                                                               cncncoco
                                                                                                               cncncncncncricricr>cncncncn
S
                                                                                                                       mrocnfOCMcnom
                                                                                                                                       -
                                                                                                                                     csi  in  CM
                                                                                                                                       -
                                                                                                              CMCMCMCMCMr-(.-l
                                                                                                              cccMcnco
                                                                                                              CM  »— <  CNJ   i — 1
                                                                                                                                                  vo  CM   CM  ro
                                                                                                                                                  OJ   • — i  CD  CD
                                                                                                                                                                          O   T-*  CM
                                                                                                                                                                          cn   o  co   cn
                                                                                                                                                                          CTi   CD  CT»   Oi
                                                                                                                                                                          CM  CO   CO  CM
                                                                                                                                                                          CD  O   CD
                                                                                                                                                                              CM   CM  ^^
                                                                                                                                           1—I   O  CM  "•t
                                                                                                                                           ^-   50  OO  ^~
                                                                                                                                           co   to  to  co
                                                                                                                                           O   CD  CD  CD

                                                                                                                                           O   CD  O*  CD
                                                                                                                                                                         G   O  CD   O
                                                                                                             j—  in   i—»  cn   I-H  \o

                                                                                                             CO  CO   CO  CM   CO  CM
                                                                                                                                                                         CM   CM  CM   CM
                                                                                                                                                                              CM  CM   •— "
                                                                                                                                                                         LTJ   cn  in   LO
                                                                                                                                                                               -
                                                                                                                                            .
                                                                                                                                LO  LO   tO  LO
                                                                                                                                            . i — i   to  to  to   to
                                                                                                            C-7

-------
5[ CM CM CO CO t-*CM CM (





















v>
1-
LU
o:

l^

Ul
1—
o

(/)
CO
5
CC
o
1
UJ
zz
t— <
If?
I— «
g
UJ
j«t
1
S
EA



o
O
OJ
S























ID
V7




4^
•p-
(
LL

w

cr j=
cu cn
u ei M
4= $c.
LU JD

Of *—
U (/
O CU £
in |— rj
cc *<—






IE







"S
4->
2
CU
E
•r-
O
C
i— t
CU
tn
s






+J
U
2
•o
O
ex
o
h-














o
T^i
*Mi
cot
°!
°',
c
(O
fO
ex
o

1


.J
J|  4-> j-j
in in ro ra
ro 
S3CO CD
CM r-* r-* CM




ro co ro
CD CU CU
tn tn tn

U O (J
ro ro ro.
4-> 4-» -M
4->
ro ro ro
UCD O CD
«->_ro ro co





































cu
t-t 1






LO
CO




10
o







o
o
f— 1








•a
tn
ro
O
cu
JJ






cu
-o
£
TD
0
CU

















X
h-
c
o
** in
c ur
cu
CD.





£
LU






*.o co r*-«.
CM co cn
co cn cn




000
i— 1 i-H i— 1







LO O LO
CM ,-t«3-
^ LO LO
r- 1 t— 1 i-*








in in tn
ro ro ro
CD C3 CS
QJ CD CU
4-J 4J 4->
tn in in








•a
0
•=£
tn
o t.
•r- CU
S- UJ
u

cncn




coco
CM CM







C3LO
vo r —
*— 4 r-H








ro 
cL
i-
o
o
aj
•a
is
ro
o *
4-J
czu-
o ro
•r-h-
X


1
o
S-
a>
a.





cu
^o
cn
cn




vo
C5







O
o
»-H








rd
Ul
(O
cs
OJ
Ul
J









cu
c:

3
CO











Q.
i-
o
°
ro
u
Sx
JZ H-
CJ
x c
cu o
4-» 4J
t tn
O 3
S- 0
4-> 2T
CU
a.

o
4-)


tn
E
T~







cn
A*
'(0

4-J
E
OJ
C
o
o



ro

cu

E
O
CJ





ro
in
ro
CD
cu
4J
tn
ra







cu

s_
4-1
E
O
B
3














o

ro «
U IT.
•r- CU
E 4->
cu ro
-C •<— X
0 TO r—
CU
o = «
4-3 S- C
cr cu •<-
fO 4J >
tn C i —
E f-4 cC
0
CO t-H
t-r O

-a

4-> ro
cu
2: o
cn
Q. in
'cu
-a -a
CU 0
tn E:
o
Q. m
O 4-J
1- E
a. cu
•» 3
CO S-
O in

S- >—•
ro
U c
£ ?
>- ^~
•a >)
,c
•o
i — c
o «
-t-» Ul
c
!- 0
O J2
1- i-
10
-— - o
QJ O
•f-> S-
•r- -0
Ul >,
1 _£=
C
O f—
• — (0
CO O
i-H 4J
•0 S-
0 O
, S-
> -Q 4->
o in
s- -o E
CL CU •-«
•i= = C

in O
O ra S-


in tn
ro in •» E
cn 4-* in o

r— 3 0 S-
ro in _o ro
S- CO S- U
3 I- (0 O
4-J U t-
ro 4-> o -o
E in s_ >>
^_ cu -a j:
4J .C CU
3 ro . — ro
cu ro j=
•o E 3 4->
eu ~o co
4-) q_ -i- cr
E C > E
CU T— O
£ CU "O C
cu cn E
i — ro ••- , —
CL s- ro
CL CU S- *->
=J > 0 O
ro .a
cu
•o
^


in
QJ
S
t-
tn

•"•
o
s-

tn
0 •
.a tn
i- E
o .a
o s-
1,8
J= S-
-o
O Q)
•W C
ro

O 4->
QJ O

Ul I —
1 ro
O 0
CO i.
r-< O
-a
0 i-l
.E: o
01
E: T3
cr
«a: ro
Q_
UJ O
01
•a
OJ Ul
Ul r—
0 OJ
o.-a
0 O
S- Tr
o,
Ul
Ul C
•O Q)
0 H
J= p
aj -M
E ui
c:
QJ .— ,
ai
s- c
5£
>»
O
JT E
o ro •
ro E
cu * o
tn •?-
-Q o tn
_a =s
•0 S- JO
cu ro ,=
s- u 6
=J O U
in i_
ro -o cu
• cu >> >
2 -E 0
s_
tn r— CL
4-> ro =
, — 4-1 .r-
3 O
V) 4-> O
CU 4->
t- S-
o in
4-> M- ra
tn cn
CU E
•*-> O -E

E 4-» -r-
ra ra 5
i£ ^
,O Q.I
14- = J_J
oo c:
U CJ
QJ H
cn E co
ro -i— i —
CO i— 1 CL
> O =J
«=C ^f CO
u -o

































CT)
E
X
cu
o
t_
CL
o
4-)
in
o
4-3
ro
u
•i^.
o


1.
0
1

o
E
O
4J
CU
'a.
u
s.
cu


ro

in

in
CU

i_
o
14—

E
CO
•r-
O
<4-
M-
tu
cu
cn
ra
S-
cu
>
<
cu
                                      C-8

-------
      Three sets  of  test data are available.   These are emission tests
 conducted  on  (1)  incinerators at polymers  and resins  plants  by EPA,
 (2)  incinerators  for  waste  streams  from air  oxidation processes conducted
 by EPA  or  the  chemical  companies, and  (3)  laboratory  unit data from
 tests conducted  by  Union Carbide Company on  incinerated streams containing
 various  pure organic  compounds.   (No adequately  documented data were
 found for  tests  of  incinerators  at  polymers  and  resins plants  that
 were conducted by the companies.)
     The EPA test studies represent the most in-depth work available.
 These data  show  the combustion efficiencies  for  full-scale incinerators
 on process  vents  at four chemical plants.  The tests  measured  inlet
 and outlet  VOC, by compound,  at  different  incineration temperatures.
 The reports include complete  test results, process  rates,  and  descriptions
 of the test method.   The four plants tested  by the  EPA are:
     1.  ARCO  Polymers,  Deer  Park,  Texas,  polypropylene unit,
     2.  Denka Chemicals, Houston,  Texas, maleic anhydride unit,
     3.  Rohm  and Haas,  Deer  Park,  Texas, acrylic acid unit, and
     4.  Union Carbide,  Taft, Louisiana, acrylic acid  unit.
 The data from ARCO Polymers  include test results based on  three different
 incinerator temperatures and  three  different waste  stream  combinations.
 The data from Rohm and Haas also include results for  three temperatures.
 The data from Union Carbide include test results based  on  two  different
 incinerator temperatures.   In all tests, bags were  used  for collecting
 integrated samples and a gas  chromatagraph with flame  ionization
detector (6C/FID) was used for obtaining an organic analysis.
C.2.1  Environmental  Protection Agency  (EPA) Polymers  Test Data2
     EPA conducted emission tests at the incinerator at  the ARCO
Polymers, Inc., LaPorte polypropylene plant in Deer Park, Texas (listed
as ARCO Chemical, Co., in LaPorte, Texas, in the 1982  Directory of
                  O                                - •"    i - j—   **
Chemical Producers ) to assess emission levels and VOC destruction
efficiency.
     The ARCO polypropylene facility has a nameplate capacity of
181,000 Mg/yr (400 million lbs/yr).3  The facility produces polypropylene
resin by a liquid phase polymerization process.  The facility includes
two "plants" (Monument I and Monument II) comprised of a total  of six
                                C-9

-------
process trains  producing a  variety  of  polypropylene  resins.   Both
plants discharge their gaseous,  liquid,  and  solid  process  wastes to
the same  incinerator  system where they undergo  thermal  destruction.
The wastes in the plants occur from:
     a)   processing  chemicals and  dilution  solvents  for the  catalyst,
     b)   spent catalyst,
     c)   waste polymeric material  (by-product  atactic  polymer), and
     d)   nitrogen-swept propylene  from  the  final  stages (product
          resin purge columns) of the  process.
The feed rates of these wastes to the  incinerator  vary  according to
which trains are running and what startups are  occurring in the two
plants.  Feed rate variations were  observed  during the  two weeks of
the incinerator test.
     The waste heat boiler  associated  with the  incinerator provides a
major portion of the  process steam  needed by the two  polymer  plants.
Natural gas is used as an auxiliary fuel to  fire the  incinerator.  If
necessary, fuel oil can also be  used.  Under full  production  conditions,
the atactic waste provides  approximately 50  percent of  the energy
needed to produce the steam, and natural gas use is reduced.
     C.2.1.1  Control Device.  The  incinerator  and associated equipment
were designed by John Zink, Company.   The system was  put into operation
on August 16, 1978.  The incinerator's two main purposes are  to destroy
organic waste from the polymer processes (primary) and  to provide heat
to generate steam (secondary).   Figure C-2 depicts a  flow diagram of
the incinerator and associated equipment.  Each inlet stream  has its
own nozzle inside the incinerator.  Combustion  air is fed into the
incinerator at the burner nozzles located approximately 4 feet beyond
the incinerator entrance.   The combustion air flow rate is regulated
manually.   The quench air enters the incinerator within 3 feet of the
burner nozzles.  It is used to maintain  a constant temperature and
provide excess combustion air.  The quench air  flow rate is automatically
regulated  by an incinerator temperature  controller.
     During normal  operation with all waste  streams entering  the
incinerator, the natural  gas is cut back and the atactic waste becomes
the major fuel source.  The purge gas, which has a low fuel value because
                                C-10

-------
                                                       o
                                                       OJ


                                                       O!



                                                      1
                                                       CS.
                                                       O
                                                       O-
                                                      -u
                                                      03
                                                      trt

                                                      in
                                                      O)
                                                      O
                                                      o

                                                      o
                                                      o
                                                      oo
                                                      CM
                                                       I
                                                      o

                                                      cu

                                                      3
                                                      en
c-n

-------
it is 95 percent nitrogen, is fed continuously to the incinerator for
destruction of the VOC since there is no gas storage capacity in the
system.  During an upset of the incinerator this stream is sent to a
flare.  ARCO provided data to illustrate normal operating parameters
of the incinerator.  These are listed in Table C-4 and represent the
averages for the month of August 1981.  The following are considered
design parameters:
     a)   heat input  =2.18 MJ/s (7,.45 x 106 Btu/hr),
     b)   air supply   =15.1 standard m3/s at 0°C (33,900 scfm, at 60°F)
     c)   firebox temperature  =980°C average and 1,200°C maximum (1,800°F
          average and 2,200°F maximum),
     d)   firebox residence time  =1.5 seconds, and
     e)   pressure  =19 kPa (78 in. H20).
     C.2.1.2  Sampling and Analytical Techniques.  A secondary purpose
of the ARCO incinerator test was to compare results of different
analytical methods for to the measurement of VOC emissions.  During
the testing phase of this program, three different methods were used
for the collection and analysis of hydrocarbons.  These were:
     a)   EPA Method 25,
     b)   Proposed EPA Method 18 (both on-site and off-site analyses
          performed), and
     c)   Byron instruments Model 90 sample collection system and
          Model 401 hydrocarbon analyzer sampling system and instrument
          combination.
     To characterize the VOC destruction efficiency across the thermal
incinerator, liquid, solid, and gas phase sampling was performed.  The
sampling locations were:
     a)   Incinerator inlet - waste gas stream
                            - natural gas stream
                            - atactic waste stream
     b)   Waste heat boiler outlet, and
     c)   Scrubber stack outlet (volumetric flow rate).
     The sampling system used for Method 25 consisted of a mini-impinger
moisture knockout, a condensate trap, flow control system, and a
sample tank.  Both pre- and post-sampling leak tests were performed to
ensure sample integrity.  In the case of Method 18, samples were
collected using a modification of EPA Method 110 for benzene.  This

                                C-12

-------

















a

LJ_
«£
1—
•a:
a

z.
o

o
LU
CO
•=c
CD

CD
•z. ro

1— 
1 §
a: >-


a: a
O LU
H- D
S 5
LU CO
= ct:
.-H Q_
O 	
2T




O
1 — t
a.

h-
i
o
QJ
to
[_

















QJ C 4-> IO
4-> O C 4J
tO -f- CO E
(O 4-> £ 03
3 rO 3i— •
•— CD.
(J 3 £

•£> S- *~ t— «
U 'I— E

O C
-M rtS
i— •
•i- ro *-*
__J S-
aj c c:
. 4J »i- 0)
cn o E
to c: 3
3 i— * C


4^









O 4->

M- S

•t- S- CL
13 O

•i- tQ

Q) E
0) E OJ
4->-r- E
CO O 13


*r*
O


I










s-
OJ

OJ
£
ro
s-
m
D_


Ll_
O
O

«— I



t-H

(_)
0


vo



u_
0
0
CM
1
LO
CM
CM

O
o

i— i



O
.—i












O
o
I— 1
1

CM
i— 1


o
O
t
CM











^—^,

o
in

.-H
o

(-H


o
CO



«3








OJ


re
cc
e

4-5
rO

O)
Q.

QJ
I—


cn cn
in to
0. D-

LO CM
CO 1-1

ro *o
D_ a.

LO O
CO UD
in co



en
cn
o.

i


ro
1 D_ 1 O I


t-H


r-f
CM








*—*
cn

in
CL

LO
LO
1
LO

j— f

ro
1 Q_ i II

O
O


1
CO
CD
cn



^— ^
cn
to
c.

o
LO
r-H
1
O

«^_*.
i ra i ii
D-


O
CO
O

,_!
A
to
cn



CJ £ O) E=
cn 3 cn 3
n3 QJ E ra E
s_ cn-i- s- *r-
QJ C X OJ X
> vo £3 > ro

OJ QJ
S- 4->
3 OJ
c/i oi

OJ ^
S- fl3
0- OJ

OO









1 1 1









^j^ 	 t
a> x;
in M—
•^. o
ro in
Eo
OJ O
IO O
o »
II "CO
O 1
1 CO
f-- O
•3- O
O "
. 10
o * 	 •






















1 1 I























1 1 1














OJ E
cn 3
re E o>
s_ T- cn
01 X C


O i—

~— CL
1 U O
Ol S_
in ^L



ro
t-H
1
(—1
»— 1










i-


ja
LO
ro

u
01
in i

cn












•^
jT*




r-l
co
(_j
OJ 1
to

cn



CO







QJ


s- cn
QJ c

«t a:
QJ
C/)
(O
cn

u

c
ro
cn

O
C-13


-------
modification was necessary due  to  the  high moisture  content  of the
incinerator gases and the positive pressure  of  the emissions.   To
ensure  that a  representative, integrated  sample was  collected  using
the modified Method 18, three validation  tests  for sample  flow rate
and sample volume into the Tedlar  bag  were performed.
     The principle underlying the  Byron method  is the same as  EPA
Method  25.  However, rather than using a  modified standard GC,  the
Byron method uses a process analyzer.  This  instrument speciates C2
from higher hydrocarbons, but gives a  single value for all nonmethane
hydrocarbons.  After separation, all carbonaceous material is  combusted
to C02 which is then converted  to  CH4  before being measured  by an  FID.
Thus, the variable response of  the  FID to different  types  of organics
is eliminated  in the Byron 401  as  it is in EPA  Method 25.
     The oxides of nitrogen (NO ) content of the flue gas  was  determined
                                A
using the methodology specified in  EPA Method 7.  A  detailed description
of all  these sampling and analytical techniques can  be found in the
ARCO test report.
     The total  flue gas flow rate was determined two or three  times
daily using procedures described in EPA Method  2.  Based on  this
method, the volumetric gas flow rate was determined  by measuring the
cross-sectional area of the stack and the average velocity of  the  flue
gas.   The area  of the stack was determined by direct measurements.
     The work performed during this program incorporated a comprehensive
quality assurance/quality control   (QA/QC) program as an integral part
of the overall  sampling and analytical  effort.   The major  objective of
the QA/QC program was to provide data of known quality with  respect to
completeness,  accuracy, precision,  representativeness, and comparability.
     C.2.1.3  Test Results.   The VOC measurements were made  by at
least four of five independent methods  for each of eight different
combinations of incinerator temperature and waste streams.  Table  C-5
summarizes the  results of measured destruction efficiencies  (DE's) for
each  of these conditions.
     The results indicate that  the values for the DE's by Method 25
are consistently lower and of poorer quality.  The poorer quality  is
indicated by the imprecision reflected  by the much larger standard
                                C-15

-------
             Table C-5.   ARC!)  POLYMERS  INCINERATOR DESTRUCTION EFFICIENCIES FOR EACH SET OF EDITIONS
Percent Destruction Efficiency3
Calculated for Each Method
Canal tions
.'.jaW
W-'iS/WS
1,*W«F
AW/NS/HG
1,6QO"F
if8'
»* *'';c

UoOO'F

Method 18 (on-site)
HCC
•99.99777 r .00000
VJ3.9979 - .0004
>99.99721 = .00009
99.3 t .1
>92.76 i .07
?j. 996:4 = .00007
""99. 990 = .304
>99.9975 i .0001

Byron
THCd
99.994 - .002
99.996 - .001
99.9961 ± .0003
99.9 ± .1
99.3 r .10
99.9941 -. .0001
99.983 - .007
99.994 i .002
,n qC in Stack
Method 18 (off-site)
Byron Speciated
NMHCe Method 25f HC9
99.997 i .002 99.344 - .006
99.998 - .001 99.3 t .4
99.9957 ± .0002 99.6 ± .2
99.6 ± .4 76 ± 20
99.88 ± .04 66 ± 10 99.88 - .04
99.99796 i .00005 96.32 t .08
99.983 = .007 98 z 3
99.995 ± .003h 99+1 99.9979 = .0001
qas
                                        (gC in Atactic Waste + gC in Waste Gas)
  where:  gC » grans of organic carbon
  The nuwber following the = sign is the standard deviation (statistically expected true value would fall between the
  reported value nlnus the standard deviation and the reported value plus the standard deviation).
 Cawlltfons of test given are materials burned and the temperature of the incinerator.  Material codes are AW = Atactic
 Waste, N3 • Natural Gas, and WG » Waste Gas.  Incinerator design parameters are about 2.18 MJ/s (7.45 MMBtu/hr), 15.1 sm3/s
 {33,900 scfn) afr supply, 980°C; 1200°C(1800°F; 2200°F maximum) firebox temperature, 1.5 seconds residence time, and
 19 kPa (78 In. HgO pressure).
c;le»Sijrs'J using proposed EPA Method 18 (on-site) for hydrocarbons (HC) utilizing gas chromatography (GC) with a flame
 fonlzation detector (FID).  The values with "greater than" signs (>) indicate that the VOC was below the detectable
 Hrit and the detection level was used to calculate the DE's.
 •teasjred using the Byron Instruments Model 90 sample collection system and the Sryon Model 401 Hydrocarbon Analyzer
 s»apl1ng system and instrument combination (utilizing reduction to methane and FID) in the total hydrocarbon (THC)
e"!«asurei uslnj the Byron I tod els 90 and 401 combination (utilizing reduction to methane and FID) in the nonmethane
 hydrocarbon node.
 !(easjred jslnj EPA Method 25 for total gaseous nonmethane organics (TGNMO) utilizing GC-FID.  Data not believed to
 represent true values.
                proposed EPA Method 18 (off-site) for individual hydrocarbon species utilizing GC-FID.
 Difficulties with analysis - Based on most probable value.
                                                                   C-16

-------
deviations for this measurement method.  The  accuracy  and  representa-
tiveness of these values obtained  from Method 25  is, thus,  questionable.
If Method 25 results are disregarded, the DE's  for  all  testing  combinations
are found to be consistently above 99 percent.
C.2.2  Environmental Protection Agency (EPA)  Air  Oxidation  Unit Test
       Data
     The EPA test study represents the most in-depth work  available
for full-scale incinerators on air oxidation  vents  at  three chemical
plants.  Data includes inlet/outlet tests on  three  large incinerators.
The tests measured inlet and outlet VOC concentrations  by compound for
different incinerator temperatures.  The referenced test reports
include complete test results, process rates, and test  method descriptions.
The three plants tested are Denka's maleic anhydride unit  in Houston,
Texas, Rohm and Haas's acrylic acid unit in Deer  Park,  Texas, and
Union Carbide's acrylic acid unit in Taft, Louisiana.   The  data from •
Union Carbide include test results for two different incinerator
temperatures.  The data from Rohm and Haas include  results  for  three
temperatures.  In all  tests, bags were used for collecting  integrated
samples and a GC/FID was used for organic analysis.
     C.2.2.1  Denka Test Data.   The Denka maleic anhydride facility
has a nameplate capacity of 23 Gg/yr (50 million  Ibs/yr).  Maleic
anhydride is produced  by vapor-phase catalytic oxidation of benzene.
The liquid effluent from the absorber, after undergoing recovery
operations, is about 40 weight percent aqueous solution of maleic
acid.  The absorber vent is directed to the incinerator.  The thermal
incinerator has a primary heat recovery system to. generate process
steam and uses natural  gas as supplemental  fuel.  The plant was operating
at about 70 percent of capacity when the sampling was conducted.  The
plant personnel  did not think that the lowered production rate would
seriously affect the validity or representativeness of the results.
     1.  Control  Device.   The size of the incinerator combustion
                2          2
chamber is 204 m  (2,195 ft ).  There are three thermocouples used to
sense the flame temperature, and these are averaged to give the temperature
recorded in the control  room.  A rough sketch of the combustion chamber
is provided in Figure  C-3.
                                C-17

-------
12ft
                                               FLOW
                                             SIDE VIEfl
          (Inlet)
                                             23ft-3 Jin
                                                                                            17 ft-6 in
                                                                                  (Outlet)
            There are Three Thermocouples Spaced Evenly Across the Top of the Firebox.
            The Width of the  Firebox is 6ft-6in.
                       Figure C-3.   Incinerator  Combustion Chamber
                                                  C-18

-------
      2.    Sampling and Analytical  Techniques.   Gas samples of total
 hydrocarbons (THC),  benzene,  methane,  and ethane were obtained according
 to the September 27,  1977,  EPA draft benzene method.   Seventy-liter
 aluminized  Mylar  bags were used  to  collect samples over periods of
 two to three hours for each sample.   The insulated sample box and bag
 were heated to  approximately  66°C  (150°F) using  an electric drum
 heater.   During Run  1-Inlet,  the  rheostat used to control  the temperature
 malfunctioned so the  box was  not  heated  for this run.   A stainless
 steel  probe was. inserted into the  single port at the  inlet and connected
 to the gas  bag  through a "tee."  The other leg of the  "tee" went to
 the total organic  acid (TOA)  train.  A TeflonR line connected the bag
 and the  "tee."   A  stainless steel  probe  was  connected  directly to the
 bag at the  outlet.  The lines  were kept  as  short as possible and not
 heated.   The boxes were transported  to the  field lab  immediately upon
 completion  of sampling.  They  were heated until  the GC analyses  were
 completed.
     A Varian model 2440 gas  chromatograph  with  a Carle gas sampling
 valve, equipped  with matched  2 cm3 loops, was used  for the  integrated
 bag  analysis.   The SP-1200/Bentone 34 GC  column  was operated at  30°C
 (176°F).  The instrument has a switching  circuit which allows  a  bypass
 around the  column through a capillary tube  for THC  response.   The
 response curve was measured daily  for benzene (5,  10,  and  50 ppm
 standards) with  the column and in  the bypass (THC)  mode.   The  THC mode
was also calibrated daily with propane (20,  100,  and 2000 ppm  standards).
The calibration  plots  showed moderate nonlinearity.  For sample  readings
that fell within the range of  the  calibration standards, an  interpolated
response factor was used from  a smooth curve drawn  through  the calibration
points.  For samples above or  below the standards,  the  response  factor
of  the nearest standard was assumed.   THC readings  used peak  height
and column readings used area  integration measured  with an  electronic
 "disc" integrator.
     Analysis for carbon monoxide was done on samples  drawn  from  the
same integrated gas sample bag used for the THC,   benzene, methane, and
ethane analyses.  Carbon monoxide analysis was  done following  the  GC
analyses using EPA Reference Method 10 (Federal  Register, Vol. 39,
                                C-19

-------
No. 47, Harch 8, 1974).  A Beckman Model 215 NDIR analyzer was used to
analyze both the inlet and outlet samples.
     Duct temperature and pressure values were obtained from the
existing inlet port.  A thermocouple was inserted into the gas sample
probe for the temperature while a water manometer was used for the
pressure readings.  These values were obtained at the conclusion of
the sampling period.
     Temperature, pressure, and velocity values were obtained for the
outlet stack.  Temperature values were obtained by a thermocouple
during the gas sampling.  Pressure and velocity measurements were
taken according to EPA Reference Method 2 (Federal Register, Vol. 42,
No. 160, August 18, 1977).  These values also were obtained at the
conclusion of the sampling period.
     3.   Test Results - The Denka incinerator achieved greater than
98 percent reduction at 760°C (1400°F) and 0.6 second residence time.
These results suggest that 98 percent control is achievable by properly
maintained and operated incinerators under operating conditions less
stringent than 870°C (1600°F) and 0.75 second.  Table C-6 provides a
summary of these test results.
                                     5
     C.2.2.2  Rohm and Haas Test Data .  The Rohm and Haas plant in
Deer Park, Texas, produces acrylic acid and ester.  The capacity of
this facility has been listed at 181 Gg/yr (400 million Ibs/yr) of
acrylic monomers.  Acrylic esters are produced using propylene, air,
and alcohols, with acrylic acid produced as an intermediate.  Acrylic
acid is produced directly from propylene by a vapor-phase catalytic
air oxidation process.  The reaction product is purified in subsequent
refining operations.  Excess alcohol  is recovered and heavy end by-products
are incinerated.  This waste incinerator is designed to burn offgas
from the two absorbers.  In addition, all  process vents (from extractors,
vent condensers, and tanks) that might be a potential source of gaseous
emissions are collected in a suction vent system and normally sent to
the incinerator.  An organic liquid stream generated in the process is
also burned, thereby providing part of the fuel  requirement.  The
remainder is provided by natural  gas.
                                C-20

-------
O  (J  >
   O UJ J2
                         ,-«,-( ^H       co
                                                                                 ^ m to o o
                                                                                 O «-H »-H ~H ^H
                                                                                 O CM CVJ
                                             «  CM CVJ CM CV       i-      t-
                                                                                 O C3 O C3 O
                                                                 3 1C
                                                                 :ei
                                                                                   §p
                                                                                   V
                                                                                  •
                                                                                "^-.^^*^.--^"^.
                                                                                tn O^ ^-« o^ r*«
                                                                                CNJ O O »-H evi
                                                                                **»^- "^. •"•-. -^.
                                                                                trt CT» CM «W CT^

                                                                                 OJ
                                                                                to <
          
          01
                                                                                                                       QJ   •   CO
                                                                                                                  —  .—   ,__
                                                                                O  S-  3
                                                                                I-  O  O
                                                                               •*-» u 3:
                                                                                                                                O)   4-'

                                                                                                                               <=:   z:
                                                   C-21

-------
     1.  Control Device - Combustion air  is added to  the  incinerator
in an amount to produce six percent oxygen in the effluent.  Waste
gases are flared during maintenance shutdowns and severe  process
upsets.  The incinerator unit operates at relatively  shorter residence
times (0.75-1.0 seconds) and higher combustion temperatures (650° -
850°C) [1200°-1560°F] than most existing  incinerators.
     The total installed capital cost of  the incinerator  was $4.7 million.
The estimated operating cost due to supplemental natural  gas use is
$0.9 million per year.
     2.   Sampling and Analytical Techniques - Samples were taken
simultaneously at a time when propylene oxidations, separations, and
esterifications were operating smoothly and the combustion temperature
was at a steady state.  Adequate time was allowed between the tests
conducted at different temperatures for the incinerator to achieve
steady state.  Bags were used to collect  integrated samples and a
6C/FID was used for organic analysis.
     3.   Test Results - VOC destruction efficiency was determined at
three different temperatures and a residence time of  1.0  second at
each temperature.  The test results are summarized in Table C-6.
Efficiency is found to increase with temperature and, except for 774°C
(1425°F), is above 98 percent.  Theoretical calculations  show that
greater efficiency would be achieved at 870°C (1600°F) and 0.75 second
than at the longer residence times but lower temperatures represented
in these tests.
     C.2.2.3  Union Carbide Corporation (UCC) Test Data6.  The total
capacity for the UCC acrylates facilities is about 90 Gg/yr (200
million Ibs/yr) of acrolein, acrylic acid, and esters.  Acrylic acid
comprises 60 Gg/yr (130 million Ibs/yr) of this total.  Ethyl  acrylate
capacity is 40 Gg/yr (90 million Ibs/yr).  Total heavy ester capacities
(such as 2-ethyl-hexyl  acrylate) are 50 Gg/yr (110 million Ibs/yr).
UCC considers butyl acrylate a heavy ester.
     The facility was originally built in 1969 and utilized British
Petroleum technology for acrylic acid production.  In 1976 the plant
was converted to a technology obtained under license from Sohio.
     1.   Control  Device - The thermal  incinerator is one  of the two
major control devices used in acrylic acid and acrylate ester manufacture.

                                C-22

-------
 The UCC incinerator was installed in 1975 to destroy acrylic acid and
 acrolein vapors.  This unit was constructed by John Zink Company for
 an installed cost of $3 million and incorporates a heat recovery unit
 to produce process steam at 4.1 MPa (600 psig).  The unit operates at
 a relatively constant feed input and supplements the varying flow and
 fuel  value of the streams  fed to it with inversely varying amounts of
 fuel  gas.  -Energy consumption averages  15.5 fU/s (52.8 million Btu/hr)
 instead of the designed level  of 10.5 to 14.9 MJ/s (36 to 51 million
 Btu/hr).   The operating cost in 1976,  excluding capital  depreciation,
 was  $287,000.  The unit is run with  nine percent excess  oxygen instead
 of the designed three to five  percent excess oxygen.   The combustor is
 designed  to handle a maximum of four percent propane  in  the oxidation
 feed.
      The  materials of construction  of  a  nonreturn block  valve in  the
 4.1 MPa (600 psig) steam line, from  the boiler section  require that the
 incinerator be operated at 650°C  (1200°F)  instead of  the designed
 980°C  (1800°F).   The residence time  is three to four  seconds.
      2.    Sampling and  Analytical Techniques -  The  integrated  gas
 samples were obtained  according  to the September 27,  1977,  EPA draft
 benzene method.
     Each  integrated  gas sample was  analyzed on a Varian Model  2400
 gas chromatograph  with  FID,  and a heated Carle  gas  sampling  valve with
 matched 2-cm   sample  loops.  A valved capillary bypass is used  for
 total  hydrocarbon  (THC) analyses and a 2 m  long,  3.2 mm  (1/8-in.)
 outer  diameter nickel column with PORAPAK   P-S,  80-100 mesh  packing  is
 used for component analyses.
     Peak area measurements were used for the individual component
 analyses.  A Tandy TRS-80, 48K floppy disc computer interfaced  via the
 integrator pulse output!of a Linear  Instruments Model 252A recorder
 acquired, stored, and analyzed the chromatograms.
     The integrated gas,, samples were analyzed for oxygen and ^carbon
 dioxide by duplicate Fyrite readings.  Carbon monoxide concentrations
were obtained using a Becknian Model  215A nondispersive infrared (IR)
 analyzer using the integrated samples.  A three-point calibration
 (1000, 3000, and 10,000 ppm CO standards) was used with a linear-log
 curve fit.
                                C-23

-------
      Stack  traverses  for outlet flowrate were made using EPA Methods 1
 through  4  (midget impingers)  and NO  was sampled at the outlet using
 EPA Method  7.
      3.   Test Results  - VOC  destruction efficiency was determined at
 two different  temperatures.   Table  C-6 provides  a summary of these
 test results.   Efficiency was found to increase  with temperature.   At
 (800°C)  1475°F,  the efficiency was  well  above 99 percent.   These tests
 were,  again, for residence times greater than 0.75 second.   However,
 theoretical calculations show that  even  greater  efficiency  would be
 achieved at 870°C (1600°F)  and 0.75 second  than  at the  longer residence
 times  but lower  temperatures  represented in these tests.
     All actual  measurements  were made as parts  per million (ppm)  of
 propane with the other  units  reported  derived from the  equivalent
 values.  The values were measured by digital  integration.
     The incinerator combustion  temperature for  the first six runs  was
 about  630°C (1160°F).   Runs 7 through  9  were  made at an incinerator
 temperature of about 800°C  (1475°F).   Only  during Run 3 was  the  acrolein
 process operating.  The  higher temperature  caused most  of the compounds
 heavier than propane to  drop  below  the detection  limit  due  to the wide
 range  of attenuations used, nearby  obscuring  peaks,  and baseline noise
 variations.  The  detection  limit  ranges  from  about  10 parts  per  billion
 (ppb)  to 10 ppm,  generally  increasing  during  the  chromatogram, and
 especially near  large peaks.   Several  of the  minor  peaks were  difficult
 to measure.  However, the compounds  of interest,  methane, ethane,
 ethylene, propane, propylene,  acetaldehyde, acetone, acrolein, and
 acrylic acid, dominate the chromatograms.   Only acetic  acid was  never
 detected in any sample.
     The probable reason for  negative  destruction efficiencies for
 several light components is generation by pyrolysis  from other components,
 For instance, the primary pyrolysis  products  of acrolein are carbon
monoxide and ethylene.  Except for methane and, to a much lesser
 extent, ethane and propane, the fuel gas cannot contribute hydrocarbons
 to the outlet samples.
     A sample taken from the  inlet line knockout trap showed 6 mg/g of
acetaldehyde, 25 mg/g  of butenes, and  100 mg/g of acetone when analyzed
by gas chromatography/flame ionization detection  (GC/FID).

                                C-24

-------
 C.2.3  Chemical Company Air Oxidation  Unit Test Data
      These data are from tests performed by chemical companies on
 incinerators at two air oxidation units:  the Petro-Tex oxidative
 butadiene unit at Houston, Texas, and  the Monsanto acrylonitrile unit
 at Alvin, Texas.  Tests at a third air oxidation unit, the Koppers
 maleic anhydride unit at Bridgeville,  Pennsylvania,7 were disregarded
 as not accurate because of poor sampling technique.8
      c-2-3'1  Petro-Tex Test Data9.  The Petro-Tex Chemical Corporation
 conducted emission testing at its butadiene production facility in
 Houston,  Texas, during 1977 and 1978.  This facility was the "Oxo" air
 oxidation butadiene process.   The emission tests were conducted during
 a period  when Petro-Tex was modifying the incinerator to improve
 mixing  and,  thus,  VOC  destruction efficiency.
      1.    Control  Device -  The Petro-Tex incinerator for the 'Oxo1
 butadiene process  is designed  to  treat 48,000  scfm waste gas containing
 about 4000 ppm  hydrocarbon  and 7000 ppm carbon dioxide.   The use of
 the  term  hydrocarbon in this discussion indicates  that besides  VOC, it
 may  include  nonVOC  such as  methane.   The waste gas  treated  in this
 system results  from air used to oxidize butene to  butadiene.  After
 butadiene has been  recovered from air oxidation  waste  gas in  an  oil
 absorption system,  the  remaining gas  is combined with  other process
 waste gas  and fed to the incinerator.   The combined waste gas stream
 enters the incinerator  between seven  vertical  Coen duct  burner assemblies.
 The  incinerator design  incorporates flue  gas recirculation  and a waste
 heat  boiler.  The benefit achieved by  recirculating flue gas  is to
 incorporate the ability to generate a constant 100,000 Ibs/hr of
 750 psi steam with variable waste gas flow.10  The waste gas flow can
 range from.10 percent to 100 percent of the design production rate.
     The incinerator measures 72 feet by 20 feet by 8 feet, with an
 average firebox cross-sectional area of 111 square feet.  The installed
 capital  cost was $2.5 million.
     The waste gas stream contains essentially no oxygen; therefore,
 significant combustion  air must be supplied.  This incinerator is
 fired with natural  gas  which supplies 84 percent of the firing energy.
The additional  required energy is  supplied by the hydrocarbon content
of the waste  gas stream.  Figure C-4 gives a rough sketch of this unit.

                                C-25

-------
                                  Augmenting
                                  (Supplemental)
                                  Air Duct
WASTE
  GAS
                             &:a*:*S:SS^^


                        Retircuiation -^
                           Air Duct
BOILER
                                                 RECIRCULAT10N
                                                    AIR FAN
                                                                          mm
                                                                              ~
             Figure  C-4.    Petro-Tex oxo unit incinerator.
                                          C-26

-------
      2.    Sampling  and  Analytical  Techniques.   Integrated  waste gas
 samples  were  collected  in  bags.  The  analysis was  done  on  a  Carle
 analytical  gas  chromatograph  having the  following  columns:
      1.    6-ft  OPN/PORASILR (80/100).
      2.    40-ft 20  percent SEBACONITRILER on gas chrom.  RA 42/60.
      3.    4-ft  PORAPAKR N 80/100.
      4.    6-ft  molecular sieve bx  80/100.
      Stack  gas  samples  were collected  in 30 to  50  cc syringes via a
 tee on a long stainless steel probe, which can  be  inserted into the
 stack, at nine  different locations.  They were  then transferred to a
 smaller 1 cm3 syringe via a small  glass coupling device  sealed  at both
 ends  with a rubber  grommet.  The 1-cm3 samples  were injected into a
 Varian 1700 chromatograph for hydrocarbon analysis.  The chromatograph
 has a 1/8-in. x 6-ft column packed with 5A molecular sieves and a
 1/4-in. x 4-ft  column packed with  glass beads connected  in series with
 a bypass before and after the molecular sieve column, controlled by a
 needle valve to split the sample.  The data are reported as ppm total
 HC, ppm methane, and ppm non-methane hydrocarbons  (NMHC).  The  CO
 content in the stack was determined by using a  Kitagawa sampling
 probe.  The 02 content in the stack was determined via a Teledyne
 02/combustible analyzer.
     3.   Test Results.   Petro-Tex has been involved in a modification
 plan for its  'Oxo' incinerator unit after startup.   The facility was
 tested by the company after each major modification to determine the
 impact of these changes  on the VOC destruction efficiency.   The  incinerator
 showed improved performance after each modification and the destruction
 efficiency increased from about 70 percent to above 99 percent.   Table
 C-4 provides a summary of these test results.   The modifications made
 In the incinerator are described below.
November 1977
     Test data prior to  these changes  showed the incinerator was not
destroying hydrocarbons  as well  as it  should (VOC destruction efficiency
 as low as 70 percent), so the following changes  were made:
     1.  Moved the duct  burner baffles from back of the burner  to the
front;
                                C-27

-------
     2.  Installed spacers to create a continuous slot for supplemental
air to reduce the air flow through the burner pods;
     3.  Installed plates upstream of the burners so that ductwork
matches burner dimensions;
     4.  Cut slots in recycle duct to reduce exit velocities and
improve mixing with Oxo waste gas;
     5.  Installed balancing dampers in augmenting (supplemental) air
plenums, top and bottom;
     6.  Installed balancing dampers in three of the five sections of
the recycle duct transition; and
     7.  Cut opening in the recirculation duct to reduce the outlet
velocities.
March 1978
     After the November changes were made, a field test was made in
December 1977, which revealed that the incinerator VOC destruction
efficiency increased from 70.3 percent to 94.1 percent.  However, it
still needed improvement.  After much discussion and study the following
changes were made in March 1978:
     1.  Took the recirculation fan out of service and diverted the
excess forced draft air into the recirculation duct;
     2.  Sealed off the 14-cm (5-1/2-in.) wide slots adjacent to the
burner pods and removed the 1.3 cm (1/2-in.) spacers which were installed
in November 1977;
     3.  Installed vertical baffles between the bottom row of burner
pads to improve mixing;
     4.  Installed perforated plates between the five recirculation
ducts for better waste gas distribution in the incinerator; and
     5.  Cut seven 3-in. wide slots in the recycle duct for better
secondary air distribution.
July 1978
     After the March 1978 changes, a survey in April 1978, showed the
Oxo incinerator to be performing very well (VOC destruction efficiency
of 99.6 percent) but with a high superheat temperature of 450°C (850°F).
So, in July 1978, some stainless steel shields were installed over the
superheater elements to help lower the superheat temperature.  A
                                C-28

-------
 subsequent  survey  in  September  1978,  showed  the  incinerator to  be
 still  destroying 99.6 percent of  the  VOC  and with  a  lower  superheat
 temperature of  400°C  (750°F).
     This study pointed  out  that  mixing is a critical  factor in  efficiency
 and  that incinerator  adjustment after startup  is the most  feasible and
 efficient means of  improving mixing and,  thus, the destruction efficiency.
     C.2.3.2  Monsanto Test  Data.     Acrylonitrile is  produced by
 feeding propylene,  ammonia,  and excess air through a fluidized,  catalytic
 bed  reactor.  In the  air oxidation process,  acrylonitrile,  acetonitrile,
 hydrogen cyanide, carbon dioxide, carbon  monoxide, water,  and other
 miscellaneous organic compounds are produced in the  reactor.  The
 columns in  the  recovery section separate  water and crude acetonitrile
 as liquids.  Propane, unreacted propylene, unreacted air components,
 some unabsorbed organic products, and water  are emitted as  a vapor
 from the absorber column overhead.  The crude acrylonitrile  product is
 further refined in  the purification section  to remove  hydrogen cyanide
 and the remaining hydrocarbon impurities.
     The organic waste streams from this  process are incinerated in
 the absorber vent thermal oxidizer at a temperature  and residence time
 sufficient  to reduce stack emissions below the required levels.  The
 incinerated streams include  (1)  the absorber vent vapor (propane,
 propylene,   CO, unreacted air components,  unabsorbed  hydrocarbons), (2)
 liquid waste acetonitrile (acetonitrile, hydrogen cyanide, acrylonitrile),
 (3) liquid  waste hydrogen cyanide, and (4) product column bottoms
 purge  (acrylonitrile, some organic heavies).   The two separate acrylonitrile
 plants at Chocolate Bayou, Texas,  employ  identical  thermal oxidizers.
     1.   Control  Device - The Monsanto incinerator burns both liquid
and gaseous wastes  from the acrylonitrile unit and is termed the
absorber vent thermal  oxidizer.   Two identical  oxidizers are employed.
The primary purpose of the absorber vent thermal  oxidizers is hydrocarbon
emission abatement.
     Each thermal  oxidizer is a  horizontal, cylindrical, saddle-supported,
end-fired unit consisting of a primary burner vestibule attached to
the main incinerator shell.  Each  oxidizer measures 18 feet in diameter
by 36 feet  in length.
                                C-29

-------
     The thermal oxidizer is provided with special  burners and burner
guns.  Each burner is a combination fuel-waste liquid unit.  The
absorber vent stream is introduced separately into the top of the
burner vestibule.  The flows of all waste streams are metered and
sufficient air is added for complete combustion.  Supplemental natural
gas is used to maintain the operating temperature required to combust
the organics and to maintain a stable flame on the burners during
minimum gas usage.  Figure C-5 gives a plan view of the incinerator.
     2. Sampling and Analytical Techniques.  The vapor feed streams
(absorber vent) to the thermal oxidizer and the effluent gas stream
were sampled and analyzed using a modified analytical reactor recovery
run method.  The primary recovery run methods are Sohio Analytical
Laboratory procedures.
     The modified method involved passing a measured amount of sample
gas through three scrubber flasks containing water and catching the
scrubbed gas in a gas sampling bomb.  The samples were then analyzed
with a gas chromatograph and the weight percent of the components was
determined.
     Figure C-6 shows the apparatus and configuration used to sample
the stack gas.  It consisted of a sampling line from the sample valve
to the small water-cooled heat exchanger.  The exchanger was then
connected to a 250 ml sample bomb used to collect the unscrubbed
sample.  The bomb was then connected to a pair of 250 ml bubblers,
each with 165 ml of water in it.  The scrubbers, in turn, were connected
to another 250 ml sample bomb used to collect the scrubbed gas sample
which is connected to a portable compressor.  The compressor discharge
then was connected to a wet test meter that vents to the atmosphere.
     After assembling the apparatus, the compressor was turned on
drawing the gas from the stack and through the system at a rate of
about 90 cm3/s ( 0.2 ft3/min).  Sample gas was drawn until at least
0.28 m3 (10 ft3) passed through the scrubbers.  After the 0.28 m3
(10 ft3) was scrubbed, the compressor was shutdown and the unscrubbed
bomb was analyzed for Cfy, C2's, C^., and 0,3%, the scrubbed
bomb was analyzed for N2, air, 03, COg, and CO, and the bubbler
liquid was analyzed for aerylonitrile., acetonitrile, hydrogen cyanide,
and total organic carbon.  The gaseous samples were analyzed by gas
chromatography.
                                  C-30

-------
jjr >i--'-1"-^-"- - ^i- "-^-9
                                                    uu
                                                    55
                                                                    CD

                                                                     0)
                                                                    •M
                                                                     o
                                                                     O
                                                                     o
                                                                     o

                                                                     o
                                                                     in
                                                                     S
                                                                     fa
                                                                     s_
                                                                     0}
                                                                    ID
                                                                     I

                C-31

-------
                       03 .33
                       03 a
                       S =
                       3 «3
                     —t & ">


                     lal
                     ~   "S
                                          cu
                                         -M
                                          CO
                                          >>
                                          t/)

                                          en
                                          to

                                         .*
                                          u
                                          o
                                         +->
                                          «3
                                          O)
                                          e
                                          u
                                          c
                                         0}

CL_
1 T
or
UJ
X
UJ



V.


T
A
r »
r
i.
                                         O)


                                         u.
C-32

-------
     3.  Test Results.  The Monsanto Chemical Intermediate Company
conducted emissions testing at its Alvin (Chocolate Bayou), Texas,
acrylonitrile production facility during December 1977.  The VOC
destination efficiency reported was 99 percent.  (Residence time
information was not available and the temperature of the incinerator
is considered confidential information by Monsanto.)
C.2.4  Union Carbide Lab-Scale Test Data12
     Union Carbide test data show the combustion efficiencies achieved
on 15 organic compounds in a lab-scale incinerator operating between
430° and 830°C (800° and 1500°F) and 0.1 to 2 seconds residence time.
The incinerator consisted of a 130 cm, thin bore tube, in a bench-size
tube furnace.  Outlet analyzers were done by direct routing of the
incinerator outlet to a FID and GC.  All inlet gases were set at
1000 ppmv.
     In order to study the impact of incinerator variables on efficiency,
mixing must first be separated from the other parameters.  Mixing
cannot be measured and, thus, its impact on efficiency cannot be
readily separated when studying the impact of other variables.  The
Union Carbide lab work was chosen since its small size and careful
design best assured consistent and proper mixing.
     The results of this study are shown in Table C-7.  These results
show moderate increases in efficiency with temperature, residence
time, and type of compound.  The results also show the impact of flow
regime on efficiency.
     Flow regime is important in interpreting the Union Carbide lab
unit results.  These results are significant since the lab unit was
designed for optimum mixing and, thus, the results represent the upper
limit of incinerator efficiency.  As seen in Table C-7, the Union
Carbide results vary by flow regime.  Though some large-scale incinerators
may achieve good mixing and plug flow, the worst cases will  likely
require flow patterns similar to complete backmixing.  Thus, the
results of complete backmixing would be relatively more comparable to
those obtained from large-scale units.
C.3  VAPOR  RECOVERY SYSTEM VOC EMISSION TEST DATA13
     On July 14, 1980, Mobil Company collected samples of hydrocarbon
emissions from the exhaust vent of the Vapor Recovery/Knockdown System
                                C-33

-------
               Table C-7.   DESTRUCTION EFFICIENCY UNDER STATED CONDITIONS
                   BASED ON RESULTS OF UNION CARBIDE LABORATORY TESTS
Destruction Efficiency of Compound in Percent
at Residence Time
0.75 second
Flow b
Regime
Two-stage
Backraixing


Complete
Backraixing


Plug Flow



Femperature
TF>
1300
1400
1500
1600
1300
1400
1500
1600
1300
1400
1500
1600
Ethyl
Acryl ate
99.9
99.9
99.9
99.9
98.9
99.7
99.9
99.9
99.9
99.9
99.9
99.9
Ethanol
94.6
99.6
99.9
99.9
86.8
96.8
99.0
99.7
99.9
99.9
99.9
99.9
Ethyl ene
92.6
99.3
99.9
99.9
84.4
95.6
98.7
99.6
99.5
99.9
99.9
99.9
Vinyl
Chloride
78.6
99.0
99.9
99.9
69.9 •
93.1
98.4
99.6
90.2
99.9
99.9
99.9
0.5/1.5 sec
Ethyl ene
87.2/97.6
98.6/99.8
99.9/99.9
99.9/99.9
78.2/91.5
93.7/97.8
98.0/99.0
99.4/99.8
97.3/99.9
99.9/99.9
99.9/99.9
99.9/99.9
aThe results of the Union Carbide work are presented as a series of equations.  These
 equations relate destruction efficiency to temperature, residence time, and flow
 regine for each of 15 compounds.  The efficiencies in this table were calculated
 from these equations.
bThree flow regimes are presented:  two-stage backmixing, complete backmixing, and
 plug flow.  Two-stage backmixing is considered a reasonable approximation of actual
 field units, with complete backmixing and plug flow representing the extremes.
                                               C-34

-------
 at  its  Santa  Ana,  California  polystyrene  plant.   The samples  were
 taken using a MDA-808 Accuhaler   pump  while  velocity was  determined
            P
 using a Kurz  Model  441  air velocity meter.   Samples were taken  while
 the  plant was in normal  operation.  One set  of samples  was  taken while
 a vacuum was  drawn on dissolver  tanks.  Another  set  of  samples was
 taken while a vacuum was drawn on the  flash  tank.  Both sets  of  samples
 were analyzed for  styrene  and ethylbenzene by an independnet  laboratory.
 Computations  for emission  rates  were made based  on velocity,  sample
 volume  and sample  time.  The  test results, submitted by the company,
 indicate that 0.942  kg/day of ethylbenzene and 10.018 kg/day  of  styrene
 are  emitted from the exhaust  vent of the vapor recovery/knockdown
 system.  No more information was provided regarding  the sampling and
 analysis procedure used  by Mobil or the laboratory.   It is  assumed
 that standard industrial practices were used, thus generating valid
 estimates of  emissions.  However, the  data should not be  used as a
 significant basis for emission limitation.
 C.4  DISCUSSION OF TEST  RESULTS  AND THE TECHNICAL BASIS OF  THE POLYMERS
     AND RESINS VOC  EMISSIONS REDUCTION REQUIREMENT
     This section discusses test results as  well  as  available theoretical
'data and findings on flare and incinerator efficiencies,  and presents
 the  logic and the technical basis behind the choice  of  the  selected
 control  level.
 C.4.1   Discussion of Flare Emission Test Results
     The results of  the five flare efficiency studies summarized in
 Section  4.1.1.1.1 showed a 98 percent  VOC destruction efficiency
 except  in a few tests with excessive stream,  smoking, or  sampling
 problems.  The results of  the Joint CMA-EPA  study, summarized in
 Table C-2, confirmed that  98 percent VOC destruction efficiency  was
 achievable for all tests (including when smoking  occurred)  except when
 steam quenching occurred within  the range of flare gas  velocities and
 heating  values tested.  Therefore, flare gas  velocity as  high as
 18.2 m/s (60  fps) and lower heating values as low as 11.2 MJ/m3  (300  Btu/scf)
 were selected as the range of operating conditions that would ensure
 achievement of the 98 percent VOC reduction  efficiency.
 C.4.2   Discussion of Thermal  Incineration Test Results
     Both the theoretical  and experimental data  concerning  combustion
 efficiency of thermal incinerators are discussed  in  this  section.  A
                                 C-35

-------
 theoretical consideration of VOC combustion  kinetics  leads  to  the
 conclusion that at 870°C (1600°F) and 0.75 second  residence  time,
 mixing is the crucial design parameter.14  Published  literature  indicates
 that any VOC can be oxidized to carbon dioxide and water  if  held at
 sufficiently high temperatures in the presence of  oxygen  for a sufficient
 time.  However, the temperature at which a given level of VOC reduction
 is achieved is unique for each VOC compound.  Kinetic studies indicate
 that there are two rate-determining (i.e., critically slow)  steps in
 the oxidation of a compound.  The first slow step of the  overall
 oxidation reaction is the initial  reaction in which the original
 compound disappears.   The initial  reaction of methane (CHJ  has been
 determined to be slower than that  of any other nonhalogenated organic
 compound.  Kinetic calculations show that,  at 870°C (1600°F), 98
 percent of the original  methane will  react  in 0.3 seconds.  Therefore,
 any nonhalogenated VOC will  undergo  an  initial  reaction step within
 this  time.  After  the initial  step,  extremely rapid free radical
 reactions occur until  each carbon  atom  exists as  carbon monoxide (CO)
 immediately before oxidation is complete.   The  oxidation of CO  is the
 second  slow step.   Calculations show  that,  at 870°C (1600°F), 98
 percent of an  original  concentration  of  CO  will react  in 0.05 second.
 Therefore, 98  percent  of any VOC would  be expected  to  undergo the
 initial and final  slow reaction steps at 870°C  (1600°F)  in about  0.35
 second.   It is  very unlikely that the intermediate  free  radical  reactions
 would take nearly  as long as 0.4 seconds to convert 98 percent  of the
 organic molecules  to CO.  Therefore, from a theoretical  viewpoint, any
 VOC should undergo complete  combustion at 870°C (1600°F)  in  0.75
 second.  The calculations on which this conclusion  is  based  have  taken
 into account the low mole fractions of VOC and oxygen which would be
 found in the actual system.  They have also provided-for the  great
 decrease in concentration per unit volume due to the elevated temperature.
 However, the calculations assume perfect mixing of  the offgas and
 combustion air.  Mixing has  been identified as the crucial design
 parameter from a theoretical  viewpoint.
     The test results both indicate an achievable control  level  of
98 percent at or below 870°C  (1600°F) and illustrate the importance of
                                C-36

-------
mixing.   Union  Carbide  results  on  lab-scale  incinerators  indicated  a
minimum  of  98.6 percent efficiency at  760°C  (1400°F).   Since  lab-scale
incinerators  primarily  differ from field  units  in  their excellent
mixing,  these results verify the theoretical  calculations  and  suggest
that a full-size field  unit can maintain  similar efficiencies  if
designed  to provide good mixing.   The  tests  cited  in Table C-6 are
documented  as being conducted on full-scale  incinerators  controlling
offgas from air oxidation process  vents of a  variety of types  of
plants.  To focus on mixing, industrial units were selected where all
variables except mixing were held  constant or accounted for in other
ways.  It was then assumed any changes in efficiency would be  due to
changes in mixing.
     The case most directly showing the effect  of  mixing  is that of
Petro-Tex incinerator.  The Petro-Tex data show the efficiency changes
due to modifications on the incinerator at two  times after startup.
These modifications (see Section C.2.3.1, 3. Test  Results) increased
efficiency from 70 percent to over  99 percent,  with no  significant
change in temperature.
     A comparison of the Rohm and Haas test versus the  Union Carbide
lab test, as presented  in Table C-8, indirectly shows the  effect of
mixing.  The UCC lab unit clearly outperforms the  R&H unit.  The data
from both units are based on the same temperature, residence time, and
inlet stream conditions.  The more complete mixing of the  lab  unit is
judged the cause of the differing efficiencies.
     The six tests of in-place incinerators do  not, of  course,  cover
every feedstock.  However, the theoretical discussion given above
indicates that any VOC compound should be sufficiently  destroyed at
870°C (1600°F).   More critical  than the type of VOC is  the VOC  concentration
in the offgas.  This is true because the kinetics  of combustion are
not first-order at low VOC concentrations.  The Petro-Tex  results are
for a butadiene plant, and butadiene offgas tends  to be lean in VOC.
Therefore, the test results support the achievability of 98 percent
VOC destruction efficiency by a field incinerator designed to  provide
good mixing, even for streams with low VOC concentrations.
                                C-37

-------
     Table C-8.  COMPARISONS OF EMISSION TEST RESULTS FOR UNION CARBIDE
             LAB INCINERATOR AND ROHM & HAAS FIELD INCINERATOR9
                Rohm and Haas Incinerator
                 Inlet         Outlet
                      Union Carbide Lab Incinerator
                         Inlet         Outlet
Compound
Propane
Propyl ene
Ethane
Ethyl ene
TOTAL
(Ibs/hr)
900
1800b
10
30
2740
(Ibs/hr)
150
150b
375
190
865
(Ibs/hr)
71.4
142.9
0.8
2.4
217.5
(Ibs/hr)
0.64
5.6
3.9
3.4
13.54
Overall VOC
Destruction
Efficiency:
68.4%
 Table shows the destruction efficiency of the four listed compounds for the
 Rohm & Haas (R&H) field and Union Carbide (UC) lab incinerators.  The R&H
 results are measured; the UC results are calculated.  Both sets of results
 are based on 1425 F combustion temperature and one second residence time.
 In addition, the UC results are based on complete backmixing and a four-step
 combustion sequence consisting of propane to propylene to ethane to ethylene
 to C0« and H90.  These last two items are worst case assumptions.
3
 Are not actual  values.  Actual  values are confidential.   Calculations with
 actual values give similar results for overall VOC destruction efficiency.
                                   C-38

-------
      The EPA tests  at Union  Carbide and  Rohm and Haas  were for residence
 times greater than  0.75 second.   However,  theoretical  calculations
 show that greater efficiency would  be  achieved  at 870°C  (1600°F)  and
 0.75 second  than at the longer residence times  but lower temperatures
 represented  in these two  tests.   The data  on which the achieveability
 of  the 98 percent VOC destruction efficiency is  based  is test  data  for
 similar control systems:   thermal incineration  at various  residence
 times  and temperatures.   If  98 percent VOC  reduction can be  achieved
 at  a lower temperature, then  according to  kinetic theory it  can certainly
 be  achieved  at 870°C (1600°F), other conditions  being  equal.
      A control  efficiency  of  98 percent  VOC  reduction, or  20 ppm  by
 compound,  whichever is  less stringent, has  been  considered to  be
 acheivable control   level for  all  new incinerators,  considering available
 technology,  cost and  energy use.     This is  based  on incinerator  operation
 at  870°C  (1600°F) and on adjustment  of the  incinerator after start-up.
 The  20  ppm (by  compound) level was chosen after  three different incinerator
 outlet  VOC concentrations, 10 ppm, 20  ppm, and 30  ppm, were  analyzed.
 In addition  to  the  incinerator tests cited earlier  in this Appendix,
 data from  over  200  tests by Los Angeles  County (L.A.) on various waste
 gas  incinerators were considered  in  choosing the 20 ppm  level.  However,
 the  usefulness  of the L.A. data was  limited  by three factors:  (1)  the
 incinerators tested are small units  designed over a decade ago; (2) the
 units were designed, primarily, for  use  on coating operations; and
 (3) the units were  designed to meet  a  regulation requiring only 90  percent
 VOC reduction.
     The 10 ppmv level was judged to be too stringent.   Two of the  six
 non L.A. tests and  65 percent of the L.A. tests fail this criteria.
 Consideration was given to the fact  that many of the units tested were
 below 870°C  (1600°F) and did not have good mixing.  However, due to
 the large percent that failed, it is judged that even with higher
temperatures and moderate adjustment, a large number of units would
still not meet the  10 ppmv level.
     The 20 ppm level was judged to  be  attainable.  All of the non L.A.
and the majority of the L.A.  units met  this criteria.  There was
concern over the large number of L.A. tests that failed,  i.e. 43 percent.
However, two factors outweighed this concern.

                                C-39

-------
      First,  all  of  the  non  L.A.  units  net  the  criteria.   This  is
 significant  since,  though the  L.A.  units represent  many  tests,  they
 represent  the  same  basic design.  They all  are small  units  designed
 over  a decade  ago to meet a rule  for 90 percent reduction.   They  are
 for similar  applications for the  same  geographic  region  designed  in
 many  cases by  the same  vendor.  Thus,  though many failed, they  likely
 did so due to  common factors and  do not represent a widespread  inability
 to meet 20 ppm.
      Second, the difference between 65 percent failing 10 ppmv  and
 43 percent failing  20 ppm is larger than a  direct comparison of the
 percentages  would reveal.   At  20  ppm,  not only did  fewer units  fail,
 but those that did  miss the criteria did so by a  smaller margin and
 would require  less  adjustment.  Dropping the criteria from  10 ppm to
 20 ppm drops the failure rate  by  20 percent, but  is judged  to drop  the
 overall time and cost for adjustment by over 50 percent.
      The difference between the two levels  is  even  greater  when the
 adjustment effort for the worst case is considered.  The crucial point
 is how close a 10 ppm level  pushes actual field unit efficiencies to
 those of the lab unit.  Lab unit  results for complete backmixing
 indicate that  a 10  ppm level would force field units to  almost  match
 lab unit mixing.  A less stringent 20  ppm level increases the margin
 allowed for nonideal incinerator  operation, especially for  the  worst
 cases.  Given  that an exponential increase may occur in  costs to
 improve mixing enough for field units  to approach lab unit  efficiencies,
 a drop from 10 ppm to 20 ppm may  decrease costs to  improve  mixing in
 the worst case by an order  of magnitude.
     The 30 ppm level  was judged  too lenient.   The  only  data indicating
 such a low efficiency was from L.A.   All other data showed  20 ppm.
The non-L.A. data and lab data meet 20 ppm and  the  Petro-tex experience
showed that moderate adjustment can increase efficiency.   In addition,
the L.A. units were judged  to have poor mixing.   The mixing deficiencies
were large enough to mask the effect of increasing temperature.  Thus,
 it is judged that 20 ppm could be reached with moderate adjustment and
that a 30 ppm level  would represent a criteria not based  on the best
available control technology cost, energy,  and environmental impact.
                                C-40

-------
C.5   FUGITIVE  EMISSION  EQUIPMENT  INVENTORY
      The  fugitive  VOC emission  and  control  cost  analysis  for  the
polymers  and resins  industry  is based  upon  the analysis for the synthetic
organic chemical manufacturing  industry  (SOCMI).   The  SOCMI analysis
is reported in EPA-450/3-80-033a, Background  Information  for  Proposed
Standards for  VOC  Fugitive  Emissions in  Synthetic  Organic Chemicals
Manufacturing  Industry, and EPA-450/3-82-010, Fugitive Emission Sources
of Organic Compounds - Additional Information on Emissions, Emission
Reductions and Costs.  Table C-9 summarizes the model  units and emissions
estimates developed  in these studies.  The available fugitive  emission
equipment inventory  from the polymers  and resins industry,  corresponding
emission estimates (based on SOCMI  emission factors),  and  plant capacities
are presented in Table C-10.  The majority of the  plants  for which
data are available are most similar to SOCMI Model Unit B.  One SOCMI
model  unit (Model Unit B) was chosen to  represent  all  polymers and
resins facilities with regard to fugitive VOC emissions.15
                                C-41

-------
  Table C-9.
FUGITIVE VOC
 EQUIPMENT COUNTS AND EMISSIONS FOR
EMISSION SOURCES IN SOCMI MODEL UNITS





Pump Seals
Light Liquid
Service
Heavy Liquid
Service
Val ves
Vapor Service
Light Liquid
Service
Heavy Liquid
Service
Safety/relief
val ves
Vapor Service
Light Liquid
Service
Heavy Liquid
Service
Open-ended
1 i nes
Compressor
seals
Sampling
connections
Flanges
Total
Emissions
(kg/hr)
- baseline:
- uncontrolled:
Revi
Equipment
Model
Unit
A
15
8

7
362
99

131

132

13
lid

1

1

1046

1
26f
600



3.2
4.5
sed SOCMI
count for
Model
Unit
B
60
30

30
1,450
402

524

524

50
42d

4

4

4156

2
104f
2,400



12.2
17.3
Fugitive Ana
Model Unitc
Model
Unit
C
185
92

93
4,468
1,232

1,618

1,618

157
13Qd

13

14

1,2776

8
320f
7,400



37.9
53.8
lysisb
Emission
Factor,
kg/hr/
source

0.0494

0.0214

0.0056

0.0071

0.00023


0.104





0.0017

0.228
0.0150
0.0083





C-42

-------
                      FOOTNOTES FOR TABLE C-9
U
 Equipment components in VOC service only.
b
 From Fugitive Emission Sources of Organic Compounds - Additional
 Illforniation  on Missions, Emission Reductions, and Costs.	
 EPA-450/3-82-010, April 1982.                ~	—
c
 52 percent of existing SOCMI units are similar to Model Unit A-
 33 percent of existing SOCMI units are similar to Model Unit B;
 15 percent of existing SOCMI units are similar to Model Unit C.
d
 Seventy-five percent of gas safety/relief valves are assumed to be
 controlled at baseline; therefore, the baseline emissions estimates
 are based on the following counts:  A, 31; B, 11; C, 33.
e
 All open-ended lines are 100 percent controlled at baseline.
f
 Seventy-five percent of sampling connections are assumed to be controlled
 at baseline; therefore, the baseline emissions estimates are based  on
 the following counts:  A, 7; B,  26; C,  80.
                                 C-43

-------
        Table C-10.   EQUPIMENT  INVENTORIES AND EMISSION ESTIMATES FOR
Equipment
Component3
              Polymers and Resins Plant Data by Type of  Plant

A:PP-LP  B:PP-6  C:LDPE-HP  D:LDPE-LP  E:HDPE-GP  F:HDPE-6P  G:HDPE-LP  H:HDPE-LP

          24b                               7        lb          2t>         6&
 Pump Seals
  Light Liquid
   Service
  Heavy Liquid
   Service

Valves
  Vapor Service
  Light Liquid
   Service
  Heavy Liquid
   Service

Safety/relief
  valves
  Vapor Service
  Light Liquid
    Service
  Heavy Liquid
    Service

Open-ended
  lines
  Vapor Service
  Light Liquid
   Service
  Heavy Liquid
   Service

Compressor
  seals

Sampling
  connections

Flanges

Total Estimated
   Emissions
    (kg/hr)h
 -  baseline:
 - uncontrolled:
                          3,000
                                          49QC
                                                               igd.e
                                                             485
                                              293
                                               78

                                              215
                                             2.313C
                                                                 7d,e
                                                                 69
                                                                 3

                                                               695
                     24.01"
                     25.1
          19.6
          19.6
22.4J
34.8
20.lk
28.7
5.1
6.5
0.5
0.5
3.5
4.0
15.9
16.4
  Equipment components in VOC service only.
 b
  Assumed all pumps, in light liquid service.

  Assumed one-half of valves in vapor service  and one-half in light
  liquid service.
 d
  Assumed all safety/relief valves in vapor service.

  Seventy-five percent of gas safety/relief valves are assumed to be
  controlled at baseline; therefore,  the  baseline emissions estimates
  are based on 11 safety/relief valves.

  All open-ended lines are 100 percent controlled at baseline.
                                               C-44

-------
      FUGITIVE  VOC  EMISSION  SOURCES  IN  POLYMERS AND  RESINS  PLANTS
                  Polymers and Resins  Plant Data by Type of Plant

I:HOPE-LP  J:HDPE-LP  K:PS-Cont  L:EPS-PI  M:£PS-PI  N:EPS-IS  0:PET-OMT  P:PET-TPA

  66b        67°      '   22°        4*>        3b       iflb         I00b     lib
                                                                 Assumed
                                                                   For
                                                                 Polymers
                                                                   And
                                                                 Resins
                                                                 Analysis
 2,010
100C
13       175
                                                               1.500C     43C
                                            2d
                                                                         42
   29

   30


1,450
  402

  524

  524


   426
                                                  unknown
                                         250f
                                               415f
2,000
16.3 5.7
16.3 5.7
149
100
1.8
2.7
1009
4,000
0.24 0.43 1.47 25.2 0.82
0.24 0.43 1.47 47.8 0.82
1049
2,400
12.1
17.2
 Seventy-five percent  of sampling connections are  assumed to be controlled
 at baseline; therefore, the baseline emissions  estimates are based on
 26 sampling counts.
h
 Calculated using SOCMI emission factors for each  equipment component
 given in Table C-10.

 Based on a total  equipment count of about 9,400.

 Based on a total  equipment count of about 8,200.
k
 Based on a total  equipment count of about 10,200.
                                               C-45

-------
C.6  REFERENCES FOR APPENDIX C

 1.  McDaniel , M.  Flare Efficiency Study,  Volume I.   Engineering-Science.
     Austin, Texas.  Prepared for Chemical  Manufacturers  Association,
     Washington, D.C.  Draft 2, January 1983.   Docket Reference  Number
     II-I-107.*

 2.  Lee, K.W. et a!., Polymers and Resins  Volatile Organic  Compound
     Emissions from Incineration:  Emission Test  Report,  ARCO  Chemical
     Company, LaPorte Plant, Deer Park, Texas,  Volume I Summary  of
     Results.  U.S. Environmental Protection Agency,  Research  Triangle
     Park, North Carolina.   EMB Report No.  81-PMR-l.   March  1982.
     Docket Reference Number II-A-30.*

 3.  SRI International, 1982 Directory of Chemical  Producers.
     Docket Reference Number II-I-82.*

 4.  Maxwell, W. and G. Scheil.  Stationary Source  Testing of  a  Maleic
     Anhydride Plant at the Denka Chemical  Corporation, Houston,
     Texas.  U.S. Environmental Protection  Agency,  Research  Triangle
     Park, North Carolina.   EMB Report No.  78-OCM-4,  March 1978.
     Docket Reference Number II-A-4.*

 5.  Blackburn, J.   Emission Control  Options for  the  Synthetic Organic
     Chemicals Manufacturing Industry, Trip Report  to Rohm and Haas
     Acrylic Acid and Esters Plant.  U.S. Environmental Protection
     Agency, Research Triangle Park,  North  Carolina.   EPA Contract No.
     68-02-2577, November 1977.  Docket Reference Number  II-B-2.*

 6.  Scheil, G.  Acrylic Acid and Esters Production.   Emission Test Report,
     Union Carbide  Corporation, Taft, Louisiana.  U.S. Environmental
     Protection Agency.  EMB Report No. 78-OCM-8.  September 1980.
     Docket Reference Number II-A-36.*
 7.  Letter from Lawrence,  A.,  Koppers  Company,  Inc., to Goodwin, D.,
     EPA.  January 17,  1979.   Docket  Reference Number II-D-2.*

 8.  Air Oxidation Processes  in Synthetic  Organic  Chemical Manufacturing
     Industry-Background Information  for Proposed  Standard Preliminary
     Draft EIS.  U.S.  Environmental Protection Agency.  Research
     Triangle Park, North Carolina.   August  1981.   pp. C-7 and C-8.
     Docket Reference  Number  II-A-26.*

 9.  Letter from Towe,  R.,  Petro-Tex  Chemical Corporation, to Farmer, J.,
     EPA.  August 15,  1979.  Docket Reference Number II-D-4.*

10.  R.D. Pruessner and Broz, L.D.  Hydrocarbon  Emission Reduction
     Systems.  CEP. August 1977.  pp.  69-73,  Docket Reference Number
     II-I-30.*
                                  C-46

-------
 11.   Letter  from Weishaar, M., Monsanto Chemical Intermediates Co., to
      Farmer, J., EPA, November 8, 1979.  Docket Reference Number II-D-5.*

 12.   Lee, K., J. Hansen and D. Macauley.  Thermal Oxidation Kinetics
      of Selected Organic Compounds.  (Presented at the 71st Annual
      Meeting of the APCA, Houston, Texas, June 1978.)  Docket Reference
      Number  II-I-41.*

 13.   Letter  and attachments from Bowman, V.A. Jr., Mobil  Chemical
      Company, to Farmer, J.R., EPA.  September 9, 1980.  pp. 13-16.
      Docket  Reference Number II-D-9.*

 14.   Mascone, "D.C. Thermal Incinerator Performance for NSPS.
      U.S. Environmental Protection Agency.  Research Triangle Park, N.C.
      Memorandum to J.R. Farmer, Chemicals and Petroleum Branch.
      June 11, 1980.  Docket Reference Number II-B-4.*

 15.   Memorandum from Siebert, P.  Pacific Environmental Services,  Inc.
      to Polymers and Resins NSPS Project File.  September 8, 1982.
      Selection of SOCMI Fugitive Analysis Model  Plant B to Represent
      Fugitive Emissions Characteristics of Polymers and Resins Plants.
      Docket Reference Number II-B-44.*
*References can be located in Docket Number  A-82-19  at  the  U.S.
 Environmental  Protection Agency Library,  Waterside  Mall, Washington, D.C.

                              C-47

-------

-------
APPENDIX D:  EMISSION MEASUREMENT AND PERFORMANCE TEST METHODS
                    I.  Process VOC Sources
                            D-l

-------
  APPENDIX D:  EMISSION MEASUREMENT AMD PERFORMANCE TEST METHODS
                       I.   Process VOC Sources

 I-D.l  EMISSION MEASUREMENT
 I-D.1.1   Introduction
     A new source performance standard for process sources in the
 polymers  and resins manufacturing industry could be in three different
 formats.  A regulation could be based on emission concentration,
 emission  rate, or percentage emission reduction.  The purpose of this
 appendix  is to discuss and recommend measurement methods acceptable
 for determination of VOC concentration and emission flow rates, and
 procedures for calculation of emission reduction efficiency.
 I-D.1.2   VOC Concentration Measurement
     Numerous methods exist for the measurement of organic emissions.
Among these methods are gas chromatograph (GC, draft Method 18),
direct flame ionization detection (FID, Method 25A), and EPA Reference
Method 25 (EPA 25)-Determination of Total  Gaseous Nonmethane Organic
Emissions as Carbon.  Each method has advantages and disadvantages.
Of the three procedures, GC has the distinct advantage of identifying
and quantifying the individual  compounds present.  Disadvantages are
that GC systems are expensive and determination of the column required
and analysis of samples can be time consuming.
     The FID technique is the simplest procedure.  However, the FID
responds differently to various organic compounds and can yield highly
biased results depending upon the compounds  involved.   Another dis-
advantage of the FID is that a separate methane measurement is required
to determine nonmethane organics.   The direct FID procedure does not
identify or quantify individual  compounds.
     Method 25 sampling and analysis  provides a single nonmethane
organic measurement on a carbon basis;  this  is convenient for establishing
control  device efficiencies on  a consistent  basis.   However,  EPA 25
                              D-2

-------
 does not provide any qualitative or quantitative information on
 individual  compounds present.
 I-D.1.3  Emission Test Experience
      The EPA conducted an emission test  (1)  using these  three methods
 at a polypropylene production  facility.   This  facility was  equipped
 with an incinerator-heat  recovery boiler system that was fired with  a
 process waste liquid for  primary fuel.   A gaseous vent stream was
 directed to  this unit for emission control.  The results of testing
 show that Method 25  resulted in  the  highest  results for  outlet concentration
 and lowest  emission  reduction  efficiency.  It  is believed that the C02
 separation column  specified in Method 25 was not removing completely
 the high levels  of C02  in the  exhaust samples  prior to oxidation  of
 the VOC  fractions.   This  incomplete  separation  would yield  high results.
 The results  of Method  18  and Method  25A  testing are similar.   No
 difficulties  were  experienced  in  the performance of Methods  25A or 18.
 I-D.2  RECOMMENDED TEST METHODS
     The EPA  Method  18  is  recommended as  the test procedure  for determining
 the VOC  concentration in  emissions from  polymers and resins  facilities.
 If  a mass flow rate  is  required,  this result can  be multiplied by
 appropriate molecular weights to  obtain  mass concentrations.  EPA
 Method 2  can  be  used to measure exhaust  flow rates so that VOC mass
 rates can be  calculated.  EPA Method 2A,  process flow instrumentation
 or,  if appropriate, material balances can be used to calculate inlet
 vapor flow rates which, when combined with inlet VOC concentration,
will allow calculation of the emission reduction efficiency of the
control  device.
     The  cost of performing an emission test will vary depending on
the  format of a  regulation.  If it is assumed that the emission reduction
efficiency must  be measured, then a test is estimated  to  cost from
$10,000 to $15,000 per source.
I-D.3  REFERENCES
     (1)  Emission Test Report:  Arco Chemical  Company.   Deer Park.
Tests.  EMB  Report No. 81-PMR-l.   March  1982.  Docket  Reference
Number II-A-30.*
                              D-3

-------
                     II.  Fugitive VOC Sources

II-D.l  Emission Measurement Methods
II-D.1.1  General Background
     A test method was not available v/hen EPA began the development of
control technique guidelines, new source performance standards, and
hazardous pollutant standards for fugitive volatile organic compounds
from industrial categories such as petroleum refineries, synthetic
organic chemical manufacturing, and other types of processes that
handle organic materials.
     During the development and selection of a test method, EPA reviewed
the available methods for measurement of fugitive leaks with emphasis
on procedures that would provide data on emission rates from each
source.  To measure emission rates, each individual piece of equipment
must be enclosed in a temporary cover for emission containment.  After
containment, the leak rate can be determined using concentration
change and flow measurements.  This procedure has been used in several
studies(1»2) and has been demonstrated to be a feasible method for
research purposes.  It was not selected for this study because direct
measurement of emission rates from leciks is a time-consuming and
expensive procedure, and is not feasible or practical  for routine
testing.
     Procedures that yield qualitative or semi-quantitative indications
of leak rates were then reviewed.  There are essentially two alternatives:
leak detection by spraying each component leak source with a soap
solution and observing whether or not bubbles were formed; and, the
use of a portable analyzer to survey for the presence of increased
organic compound concentration in the vicinity of a leak source.
Visual, audible, or olefactory inspections are too subjective to be
used as indicators of leakage in these applications.   The use of a
portable analyzer was selected as a basis for the method because it
would have been difficult to establish a leak definition based on
bubble formation rates.  Also, the temperature of the component,
physical configuration, and relative movement of parts often interfere
with bubble formation.   However, where soaping is possible, it can
                              D-4

-------
 be used as a preliminary screening technique.  Soap solution would
 be applied to the potential leak surfaces and if no bubbles are
 observed, the source is presumed not to be leaking.
      Once the basic detection principle was selected, it was then
 necessary to define the procedures for use of the portable analyzer.
 Prior to performance of the first field test, a procedure was reported
 that conducted surveys at a distance of 5 cm from the components. (3)
 This information was used to formulate the test plan for initial
 testing. (4)   In addition,  measurements were made at distances of 25 cm
 and 40 cm on  three perpendicular lines around individual sources.  Of
 the three distances, the most repeatable indicator of the presence of
 a  leak was  a  measurement at 5 cm,  with a leak definition concentration
 of 100 or 1000  ppmv.  The localized meteorological  conditions  affected
 dispersion  significantly at greater distances.  Also,  it was  more
 difficult to  define  a  leak  at  greater distances because  of  the small
 changes  from  ambient concentrations  observed.  Surveys were conducted
 at 5  cm  from  the source  during the  next  three facility tests.
      The  procedure was distributed  for  comment in  a  draft control
 techniques  guideline document. (5)  Many commentors  felt that  a  measurement
 distance of 5 cm could not  be accurately repeated  during screening
 tests.  Since the concentration  profile  is rapidly changing between 0
 and about 10  cm  from the source, a small variance  from 5  cm could
 significantly affect the concentration measurement.  In  response to
 these comments,  the  procedures were changed so that measurements were
 made  at the surface  of the  interface, or essentially 0 cm.  This
 change required  that the leak definition level be increased.  Additional
 testing at two refineries and three chemical  plants was performed by
 measuring volatile organic concentrations at the interface surface.
     A complication that this change introduces is that a small mass
 emission rate leak ("pin-hole leak") can be totally captured by the
 instrument and a high concentration result will be obtained.  This has
occurred occasionally in EPA tests, and a solution to this problem has
not been found.
     The calibration basis for the analyzer was evaluated.  It was
recognized that  there are a  number of potential  vapor stream components
and compositions that can be expected.  Since all  analyzer types do

                              D-5

-------
not respond equally to different compounds, it was necessary to establish
a reference calibration material.  Based on the expected compounds and
the limited information available on instrument response factors,
hexane was chosen as the reference calibration gas for EPA test programs.
At the 5 cm measurement distance, calibrations were conducted at
approximately 100 or 1000 ppmv levels.  After the measurement distance
was changed, calibrations at 10,000 ppmv levels were required.  Commentors
pointed out that hexane standards at this concentration were not
readily available commercially.  Consequently, modifications were
incorporated to allow alternate standard preparation procedures or
alternate calibration gases in the test method recommended in the
Control Techniques Guideline Document for Petroleum Refinery Fugitive
Emissions.
     Since that time, studies have been completed that measured the
response factors for several instrument types(6*7,8)^  The results of
these studies show that the response factors for methane and hexane
are similar enough for the purposes of this method to be used inter-
changeably.  Therefore, in later NSPS, the calibration materials were
hexane £r_ methane.
     The alternative of specifying a different calibration material
for each type stream and normalization factors for each instrument
type was not intensively investigated.  There are at least four instrument
types available that might be used in this procedure, and there are a
large number of potential stream compositions possible.  The amount of
prior knowledge necessary to develop and subsequently use such factors
would make the interpretation of results prohibitively complicated.
Additionally, based on EPA test  results, the measured frequency of
leak occurrence in a process unit was not  significantly different when
the leak definition was based on meter reading using a reference
material and when  response factors were used to correct meter readings
to actual concentrations for comparison to the leak definition.
     An alternative approach to  leak detection was evaluated by EPA
during field testing. (9>10) The  approach used was an area'survey, or
walkthrough, using a  portable  analyzer.  The  unit area was surveyed by
walking through the unit positioning the instrument probe within
1 meter of  all  valves  and  pumps.  The concentration readings were
                               D-6

-------
 recorded on a portable strip chart recorder.   After completion  of  the
 walkthrough,  the local wind conditions  were used with the  chart data
 to locate the approximate source of any increased ambient  concentrations.
 This  procedure was  found to yield mixed results.   In  some  cases, the
 majority of leaks  located by individual  component testing  could be
 located  by walkthrough surveys.   In other.tests,  prevailing  dispersion
 conditions and local  elevated ambient concentrations complicated or
 prevented the interpretation of  the results.   Additionally,  it  was not
 possible to develop  a general  criteria  specifying how much of an
 ambient  increase at  a distance of 1  meter  is  indicative of a 10,000
 ppm concentration at  the leak  source.   Because of the potential  variability
 in results from site  to  site,  routine walkthrough surveys were  not
 selected as a reference  or  alternate test  procedure.
 11-D.1.2  Emission Testing  Experience                                   :
      During the development of the  new  source  performance standard for
 fugitive VOC  emissions in the  synthetic organic manufacturing industry,
 a  screening program using Method  21  was conducted  at 24 process units.01)
 Several  of the  process units  included in this  survey were monomer
 production and  polymerization  sections  of  plants  included in the
 polymers  and  resins industry  category.  The instruments used were
 flame ionization, catalytic  oxidation, and, in one case, photoionization.
 The flame  ionization  and catalytic oxidation instruments were calibrated
 with methane  standards.  The  photoionization instrument was calibrated
 with isobutylene.  The response factors  for these compounds are similar
 for use  in  this  application.
 II-D.2  CONTINUOUS MONITORING SYSTEMS AND DEVICES
     Since  the  leak determination procedure is not a direct emission
 measurement technique, there are no continuous monitoring approaches
 that are  directly applicable.  Continual surveillance is achieved by
 repeated  monitoring or screening of all  affected potential  leak  sources.
A continuous monitoring system or device could serve as  an  indicator
that a leak has developed between inspection intervals.   The  EPA per-
formed a  limited evaluation of fixed-point monitoring systems for
their effectiveness in leak detection. (9J2J3)  Tne systems  consisted
of both remote sensing devices with a central  readout and a central
analyzer system  (gas chromatograph) with remotely collected samples.
                              D-7

-------
The results of these tests indicated that fixed point systems were not
capable of sensing all leaks that were found by individual component
testing.  This is to be expected since these systems are significantly
affected by local dispersion conditions and would require either many
individual point locations, or very low detection sensitivities in
order to achieve similar results to those obtained using an individual
component survey.
     It is recommended that fixed-point monitoring systems not be
required since general specifications cannot be formulated to assure
equivalent results, and each installation would have to be evaluated
individually.
II-D.3  PERFORMANCE TEST METHOD
     The recommended fugitive emission detection procedure is Reference
Method 21.  This method incorporates the use of a portable analyzer to
detect the presence of volatile organic vapors at the surface of the
interface where direct leakage to atmosphere could occur.  The approach
of this technique assumes that if an organic leak exists, there will
be an increased vapor concentration in the vicinity of the leak, and
that the measured concentration is generally proportional to the mass
emission rate of the organic compound.
     An additional  procedure provided in Reference Method 21  is for
the determination of "no detectable emissions."  The portable VOC
analyzer is used to determine the local ambient VOC concentration in
the vicinity of the source to be evaluated, and then a measurement is
made at the surface of the potential leak interface.  If a concentration
change of less than 5 percent of the leak definition is observed, then
a "no detectable emissions" condition exists.  The definition of 5
percent of the leak definition was selected based on the readability
of a meter scale graduated in 2 percent increments from 0 to 100
percent of scale, and not necessarily on the performance of emission
sources.
     Reference Method 21 does not include a specification of the
instrument calibration basis or a definition of a leak in terms of
concentration.  Based on the results of EPA field tests and laboratory
studies, methane or hexane are recommended as the reference calibration
bases for fugitive emission sources in the polymers and resins manufacturing
industries.
                              D-8

-------
     There are at least four types of detection principles currently
available in commercial portable instruments.  There are flame ionization,
catalytic oxidation, infrared absorption (NDIR), and photoionization.
Two types (flame ionization and catalytic oxidation) are known to be
available in factory mutual certified versions for use in hazardous
atmospheres.
     The recommended test procedure includes a set of design and
operating specifications and evaluation procedures by which an analyzer's
performance can be evaluated.  These parameters were selected based on
the allowable tolerances for data collection, and not on EPA evaluations
of the performance of individual instruments.  Based on manufacturers'
literature specifications and reported test results, commercially
available analyzers can meet these requirements.
     The estimated purchase cost for an analyzer ranges from about
$1,000 to $5,000 depending on the type and optional  equipment.  The
cost of an annual  monitoring program per unit, including semiannual
instrument tests and reporting is estimated to be from $3,000 to
$4,500.  This estimate is based on EPA contractor costs experienced
during previous test programs.  Performance of monitoring by plant
personnel may result in lower costs.   The above estimates do not
include any  costs  associated with leak repair after detection.
                              D-9

-------
II-D.4  REFERENCES

 1.  Joint District,  Federal,  and State  Project  for  the  Evaluation of
     Refinery Emissions.   Los  Angeles  County  Air  Pollution  Control
     District, Report Nos. 2,  3,  5,  6, and  8.  1957-1958.   Docket
     Reference Numbers II-I-6  through  II-I-10.*

 2.  Wetherold, R. and L.  Provost.   Emission  Factors  and  Frequency of
     Leak Occurrence  for  Fittings in Refinery  Process  Units.  U.S.
     Environmental Protection  Agency,  Research Triangle  Park, NC.
     Report Number EPA-600/2-79-044.  Docket  Reference Number II-A-8.*

 3.  Telecon. Hustvedt, K.C.,  EPA, CPB,  to  Harrison,  P.,  Meteorology
     Research, Inc.  Regulating  leaks  and determining  emission  rates
     from petroleum refinery.  December  22, 1977.  Docket Reference
     Number II-E-1.*

 4.  Miscellaneous Refinery Equipment  VOC Sources  at  ARCO,  Watson
     Refinery, and Newhall Refining  Company.   U.S. Environmental
     Protection Agency, Emission  Standards  and Engineering  Division,
     Research Triangle Park, NC.   EMB  Report  Number  77-CAT-6.   December
     1979.  Docket Reference Number  II-A-11.*

 5.  Hustvedt, K.C.,  R.A.  Quaney, and  W.E.  Kelly.  Control  of Volatile
     Organic Compound Leaks from  Petroleum  Refinery  Equipment.  U.S.
     Environmental Protection  Agency,  Research Triangle  Park, NC.
     OAQPS Guideline  Series.  Report Number EPA-450/2-78-036.   June
     1978.  Docket Reference Number  II-A-6.*

 6.  DuBose, D.A., and G.E. Harris.  Response  Factors  of  VOC Analyzers
     at a Meter Reading of 10,000 ppmv for  Selected  Organic Compounds.
     U.S. Environmental Protection Agency,  Research  Triangle Park,
     N.C.  Publication No. EPA 600/2-81-051.   March  1981.   Docket
     Reference Number II-A-22.*

 7.  Brown, G.E., et  al.   Response Factors  of VOC  Analyzers Calibrated
     with Methane for Selected Organic Compounds.  U.S.  Environmental
     Protection Agency, Research  Triangle Park,  N.C.   Publication No.
     EPA 600/2-81-022.  September 1980.  Docket  Reference Number II-A-13.*

 8.  DuBose, D.A., et al.   Response  of Portable  VOC  Analyzers to
     Chemical Mixtures.  U.S.  Environmental Protection Agency,  Research
     Triangle Park, N.C.   Publication  No. EPA  600/2-81-110.  June 1981.
     Docket Reference Number II-A-25.*

 9.  Emission Test Report:  Dow  Chemical Company,  Plaquemine, LA.  EMB
     Report No. 78-OCM-12C, December 1980.  Docket Reference Number
     II-A-20.*

10.  Weber, R.C., et al.   "Evaluation  of the  Walkthrough  Survey Method
     for Detection of Volatile Organic Compound  Leaks,"  EPA Report No.
     600/2-81-073, EPA/IERL Cincinnati,  Ohio,  April  1981.   Docket
     Reference Number II-A-24.*
                                  D-10

-------
  11.  Blacksmith, J.R., et al.  "Frequency of Leak Occurrence for
      Fittings in Synthetic Organic Chemical Plant Process Units"  EPA
      Report No. 600/2-81-003.  September 1980.  EPA/IERL  RTP  NC
      Docket Reference Number II-A-14.*                  '    '


  12'  S1nSlon TeSt ReP°rt:   Sun Petroleum Products Co., Toledo, OH,"
      II-A-150** N°' 78~°CM~12B' °ctober 198°-   Docket Reference Number


  13.  "Emission Test Report:   Union Carbide Corp., Torrance,  CA."  EMB
      Report No.  78-OCM-12A,  November 1980.   Docket Reference Number
      J. Jl *"M— JL / *
*References can be located in Docket  Number A-82-19  at  the
 U.S. Environmental  Protection Agency Library, Waterside Mall
 Washington,  D.C.                                             '
                               D-ll

-------

-------
       APPENDIX E:  DETAILED DESIGN AND COST ESTIMATION PROCEDURES

 E.I  GENERAL
      This appendix consists of a more detailed presentation of the
 bases,  assumptions, and procedures used to estimate equipment designs
 and corresponding capitalwand operating costs for flares,  thermal
 incinerators,  catalytic incinerators, shell-and-tube condensers,  ethylene
 glycol  recovery systems,  and piping and ducting.   The basis of design and
 cost estimates are presented in the following sections:   E.2,  flares; E.3,
 thermal  incinerators;  E.4,  catalytic incinerators;  E.5,  shell-and-tube
 condensers;  E.6,  ethylene glycol  recovery  systems;  and E.7,  piping and
 ducting.   Sufficient detail  was presented  in  Chapter 8 for  VOC fugitive
 emissions control.   The installation cost  factors used in each analysis  and
 the  annualized cost factors  used  in all  of the  cost analysis are  given in
 Tables 8-2 and 8-3,  respectively.
 E.2   FLARE DESIGN  AND  COST ESTIMATION PROCEDURE
      Flares  are open  combustion devices  that  can  be used to effectively
 and  inexpensively  reduce  VOC  emissions.  The  polypropylene and polyethylene
 industries commonly  use flares  to  control  large emergency releases and
 some  high  VOC  streams.  Elevated flares were  costed based upon  state-of-
 the-art industrial  design  (one-half of  sonic  velocity  and a minimum of
 115 Btu/scf).  Flare height and diameter,  which are the primary
 determinants of capital cost, are  dependent on flare flow rate, heating
 value, and temperature.  Associated  piping and ducting from the process
 sources to a header and from a  header to the flare  were conservatively
 designed for costing purposes.  Operating  costs for utilities were
 based on industry practice (1 fps purge of waste  gas plus natural  gas
 for continuous flow flare; 80 scfh  natural  gas per  pilot,  number of
pilots based on flare  tip diameter; 0.4 Ib steam/lb hydrocarbon at
maximum smokeless rate).
                                    E-l

-------
     E.2.1  Flare Design Procedure
     Design of flare systems for the various combinations of waste
streams was based on standard flare design equations for diameter and
height presented by IT Enviroscience.l These equations were simplified
to functions of the following waste gas characteristics:  volumetric flow
rate, lower heating value, temperature, and molecular weight for a
state-of-the-art exit velocity of one-half sonic velocity (0.5 Mach).
The diameter expression is based on the equation of flow rate with
velocity times cross-sectional area.  A minimum commercially available
diameter of 2 inches was assumed.  The height correlation premise is
design of a flare that will not generate a lethal radiative heat level
(1500 Btu/ft2 hr, including solar radiation2.) at the base of the flare
(considering the effect of wind).  Heights in 5-foot multiples with a
minimum of 30 ft. were used.3  Natural gas to increase the heating
value to 115 Btu/scf was considered necessary to ensure combustion of
streams containing no sulfur or toxic materials.4  For flares with
diameters of 24-inches or less, this natural gas was assumed to be
premixed with the waste gas and to exit out the stack.  For larger
flares, a gas ring was assumed if large amounts of gas were required
because separate piping to a ring injecting natural gas into the existing
waste gas is more economical than increasing the flare stack size for
large diameters.  The flare height and diameter selection procedure is
detailed in Table E-l.
     Purge gas also may be required to prevent air intrusion and
flashback.  A purge velocity requirement of 1 fps was assumed during
periods of continuous flow for standard systems without seals.?  For
flares handling only intermittent flows, purge gas requirements were
assumed to be negligible according to the industry practice of not
purging or perhaps purging before a planned intermittent release.°
For combined streams with very large turndown ratios  (intermittent
flow : continuous flow), supplying purge gas to maintain an adequate
continuous flow in a large flare  (designed for the intermittent flow)
can become more expensive than designing a second separate flare for
the continuous flare.  For flares handling very small flows in a minimum-
available-size flare, the cost of supplying sufficient purge gas can be
                                  E-2

-------
        a
        LU
        LU
        —i
        LU
        0
        «=c
        LO
             LU
        1—   OS
        as   <
        

                                                                       3 "
                                   •   X   XCO   X
r  ~  **     
-------
Footnotes for  Table  E-l
     Standard conditions  for flare  design  and  cost  calculations:  70°F  and  1 a tin.

     ''Auxiliary natural  gas  requirement  assumes  115  Btu/scf minimum  lower heating
      value necessary  to ensure  combustion  (Reference  4) and  lower heating  value
      of natural  gas of  930  Btu/scf  at 70°F (lOOCTBtu/scf  at  32°F).  From an energy
      balance                                                     •  •
      «
        ng
* "V * % x
Qng (scfm) -
L1V * <'„ + V * (
'115 - LHV (Btu/scf)
wg
LHV (Btu/scf) - 115
115 Btu/scf)
x O^g (scfm)
(2.72 x 10"3) x
i-r
(^ (1b/hr) x^O
*•
fwg (°F) +460)/MWfl>g
,/Ap tin. H20)
     cTenperature of mixture approximated  by  assuming  uniform  specific  heats
      per unit mass.
     dFron Envirosdence (Reference 1,  Appendix  A)
     D(1n)
          Using an approximation for sonic velocity,  c,  given  fay  Straitr
     (Rsftrence 5) that by comparison with the  relationship  given
     1n Reference 6 implies that the ratio of specific heats,  c /c   is
     about 1.15, which 1s reasonable for the compounds and temperatures
     of Interest, and using an exit velocity of one-half sonic velocity.
          V.(fps)  - 0.5 x c =• 125 x
     Frcra the Envlronscience equation for flame angle,
        Ap (in. H20)  « 55 [Vg (fps)/550]2 - (1.818 x  iti"4)

     Substituting,
0 (1n.)-
Wh1d
(2.72 x 10"3) x Qfl
-------
Footnotes  for Table  E-l  (.concluded)
     9FromAp (in. H20) = 55 [Ve (fps)/550]2, (as in c.)
V , of 60 mph
w
9 = TAN"1
"l.47 (fps/mph)xVw (nph)"
ssu yAp/ss
= TAN"1
"1.47 (60)"
V
e
     Vrom Enviroscience (Reference  1, Appendix A), assuming  flama eraissivity,
     e, of 0.12, flame radiation  intensity, I, of 1200 Btu/hr  ft
                              LHVl.g (Btu/1b)xT
                           (12.56) I
           -  /CQfi.q(scf/m1n) x <-HVfl>q (Btu/scf)^x  60 mln/hr x 0.12]  -3.33 x 0 x (Ve/550) COS 8

            V              4 IT (1200 Btu/hr - ft2)          ~	


     Safe pipe length  Is  the pipe length necessary to reach the horizontal  distance    ,
     from the  flare  where the flame radiation intensity,  I, is reduced to 440 Btu/hr-ft
     Including solar radiation £-300 Btu/hr-ft"), a safe  working level, (Reference 9 ):,

              L -/(r2 -  H2)
                _  /Wfi.g  (scfn|) x LHVn.g {Btu/scf) x 60 m1n/hr x 0.12]
                 / ———^~-^^—-^—-^^—_^_^^^^^^^^^^^^^_	     - H
                                12.56 (140 Btu/hr-ft2)
                                           E-5

-------
greater than the cost of a fluidic seal on the flare tip to prevent air
intrusion and subsequent flash back.  In such cases, a fluidic seal,
which requires a greatly reduced purge rate, was used.
     Natural gas was also assumed at a rate of 80 scfh per pilot flame
to ensure ignition and combustion.  The number of pilots was based on
diameter according to available commercial equipment.9
     Steam was added to produce smokeless combustion through a combined
mixing and quenching effect.  A steam ring at the flare tip was used   ..
to add steam at a rate of 0.4 Ib steam/1b of hydrocarbons  (VOC plus
methane and ethane) in the continuous stream (i.e., the maximum continuous
rate or the intermittent stream if no continuous flow was  present).10
Availability and deliverability of this quantity of steam  was assumed.
For flares handling large intermittent flows, steam requirements were
approximated  (see Figure E-l) for a small continuous  "bleed" rate that
is necessary to prevent shortened steam nozzle equipment life (due  to
steam condensate contacting the high temperature metal.)11'12
     Piping (for flows less than 700 scfm) or ducting  (for flows equal
to or greater than 700 scfm) was designed from the  process sources  to
a header combining the streams and from the  header  to  the  base of the
flare.  Since it is usual industry practice, adequate  pressure (approxi-
mately 3 to 4 psig) was assumed available to transport all waste gas
streams without use of a compressor or fan.  The source legs from the
various sources to the flare  header were  assumed to be 70  feet in
length,13 while the length  of pipelines to  the flare  was based on the
horizontal  distance required  to provide a tolerable and safe radiation
level for continuous working  (440 Btu/hr-ft2, including solar radiation9).
Piping and  ducting were selected  and costed  as outlined in Section  E.7.
     E.2.2  Flare Cost Estimation Procedure
     Flare  purchase costs were  based on costs for  diameters  from 2  to
24  inches and heights  from  20 to  200 feet provided  by National Air  Oil
Burner,  Inc.,  (NAO) during  November  1982  and presented in  Table E-2.9
A cost was  also provided  for  one  additional  case of 60 inch  diameter
and 40  feet height.10 These costs are  October 1982  prices  of self-
 supporting  flares without  ladders and  platforms  for heights  of 40  feet
and less  and  of guyed  flares  with ladders and platforms  for  heights of
 50  feet and greater.   Flare purchase  costs  were  estimated  for  the
                                   E-6

-------
                                                                     
-------














>•
ca

CM
V3 CO
LU ^
H- «— «
^£
sg
£§
v« *— i^
III UM
LU P™ "
to
o 22
O H-4
H^.| *
CO •
CJ H-4
"TJ 0^
a. a*
LU
(— 2=
LU Q£
CD ""*'
CQ CO
jjj^j
03 _1
LU O
/v*
•^ fX
_i H- 1
LU<

1
• *T
CM Z
t O
LU i— •

41 z
J3
ea
j—










A











O
1/1
3
ca
o
0
0
CM
CD
U
•r- <*-
£- U
O. in
CD 3
- tf> 4J
(O ca
"o o
s. o
3 m
a. i-«


<*-
U
in
3

CQ
o
0
o
r-4


,.
u
in
3
ca
o
S
CM

.*— »
4^ M"*
(fa, CJ
— • w
4-> 3
JS 4J
cn co
0) O
3= O
in
O) >-H
s_
ns
1— (4-
iZ o
in

3

CO
o
g
**H



^1-1
J,
g*

3^
o-^^
U_ 0)
4J
«
C£
S_
(U

^0. g
H* (Q
•r™
O



CO
cn
to
to
-«>=>•






CO
cn
to
to
•^>>






co
cn
to
to








0
CM








0
CM










O
CM










O
O
CO
CM






CM






CO
O
CM
CO







O
to
in
P^







o
to
in
P^








in
CM








O
CM










O
CM










§
CO
to






CO






to
in
01
cn







CM
cn
to
CO








cn
o
CO








in
co








in
CM










o
CM










3
CM
l-H
1— 1





^:






CM
CO
CO
CM






0
cn
in
CO







in
to
in
r— <
»— 1







O
in








o
•«*•










o
co










o
o
CM
m
CM





to






r--. cn
o «*•
O 0
01 CO
CM CO






«a- cn
tO CM
in o
in '3-
CM CO






CM to
to co
cn i*»
co r^*
i-l CM





,

in in
10 CO








in o
in r«»










o o
<^- in










0 0
a o
CO O
•«• o
•<3- f-





CO O
1— 1





l-~ CO
CM «*
1 — 1^.
^r m






i— I CO
tO *3"
o to
i— i cn
«3- •=*•






co o
in co
to to
*-< in
co co







o in
O !-H








O in
co cn










o o
to f*»










s §
CO CM
^3 ^^
O CO
' r* i-H




CM *3-
i— 1 i— 1





1 —
.3-
to






in
CO
r^.
in






in
CM
CO
,3.
^»







3








0
1—1










o
CO










s
CM
en

r— 1




to
f— t





•=1-
to
0
to
co






cn
f— 4
r-*.






CM
to
CO
CO
in







0
in








o
CM
'""'










o
cn










§
CO
to
CM
CM




co






O CM
i-- co
in in
cn CM






in CM
in co
r^N c?
O in
CO O
V**





to co
CM CM
r-. CM
"3* <3*
to CO







in o
to o








m o
CO to










o o
O CM










0 0
o o
O CM
O CO
CO O
CM ^f




O »3-
CM CM



£_
0
4-
10
0)
o
^,
a.
in

^

,,
4—
4->
03
r"~
a.
.



^
HI
VL*
-o
-a
03 •
i— (/I

3 O
O 4—
•^ *•*
«^ •••
3*0.
 "O
0) SZ
S- 03
03
r— 10
(U
c -a
•f— ea
4-> r—
s»
O -C
Q.4-J
O-'i—
3 2
V)
i cn
q_ flj
•
0) 4-»
(U R3
14— 4j
s_
o a>
•3-
•a
4- C
^3 fO
10 4->
4-) 0)
(— ^i
C35M-
•r—
CD O
j= in
4? °
cn
cn 4J
(U -=
o a>
•r" «r~
i- ai
a. -c:
ns
£-8

-------
  various regulatory alternatives by either choosing the value provided
  for the required height and diameter or using two correlations developed
  from the MAO data for purchase cost as a function of height and diameter.
  (One correlation for heights of 40 feet and less, i.e., self-supporting
  flares and one for heights of 50 feet and greater, i.e.,  guyed flares.)
  Purchase costs of large diameter,  40-ft.  high flares were approximated
  using a curve developed from the NAO  data (see Figure E-2).   Purchase
  costs for fluidic seals were approximated using a curve based on data
  provided by NAQiO (See Figure E-3).   An installation factor  of 2.1  (see
  Table 8-2}  was used  to estimate installed flare costs.   Installed costs
  were put on a June  1980 basis using  the following Chemical Engineering
  Plant Cost Indices:   the overall index  for  flares;  the  pipes,  valves,
  and  fittings  index for piping;  and the  fabricated equipment  index for
  ducting.  Annualized  costs were calculated  using  the  factors  presented
  in Table 8-3.  The flare cost estimation  procedure  is presented  in
  Table  E-3.
"E.3   THERMAL  INCINERATOR DESIGN AND COST  ESTIMATION PROCEDURE
      Thermal  incinerator designs for costing purposes were based  on
  heat  and mass  balances  for combustion of  the waste gas  and any required
  auxiliary fuel, considering requirements  of total combustion air.
 Costs of associated piping, ducting,  fans, and stacks were also estimated.
 E.3.1  Thermal Incinerator Design Procedure
      Designs of thermal  incineration systems for  the various combinations
 of waste gas streams were developed using a procedure based on heat
 and mass balances and the characteristics of the waste gas in conjunction
 with some engineering design assumptions.   In order to ensure a 98 percent
 VOC destruction efficiency,  thermal incinerators were designed to
 maintain a 0.75 second residence time at 870°C (leoo'F).1^ The design
 procedure is outlined in this section.
      Streams with low heat contents,  which require auxiliary  fuel  to
 ensure combustion and sometimes require  air  dilution or  fuel  enrichment
.to  prevent an explosive hazard,  are often  able  to  utilize  recovered
 waste heat by preheating inlet air, fuel,  and  perhaps, waste  gas.  The
 design considerations  for such streams are noted in  the  following
 discussion,  -but the combustion calculations, etc.  are  not  detailed
                                  E-9

-------
     90,000—j
     80,000-
     70,000-
^   60,000-

CM
£   50,000-
 o    40,000-
 V)
 10

 u

 3    30,000-
 o.
      20,000-
      10-,000-
                      10
20         30        40        50


   Flare Tip  Diameter (in.)
                                                                       60
                                                                                 70
            Figure E-2.   Estimated Flare Purchase Cost for 40  ft Height
                                        E-10

-------
                                                                      o
                                                                      VO
                                                                      
                                                                                        3
             o
             a
             a
             CO
(286L  '
                        E-11

-------
    Table E-3.  CAPITAL AND ANNUAL OPERATING COST ESTIMATION
               PROCEDURE FOR STATE-OF-THE-ART STEAM-ASSISTED
                              SMOKELESS FLARES
          Item
               Value
Capital Costs

Flare purchase cost, C'1'
     (Oct. 1982 $)
Fluidic seal purchase cost, C'"f| s
      (Oct. 1982 $)

Flare system purchase cost, C''f]
 Select from Table E-2 if value given
 or use equations:
   (3905.7)  + (35.054) H x D + (900.36) D
 - (126.08)02, for 20 < H < 40 ft and D < 8 in,
   (6275.6)  + (224.10) H + (12.782) H x D
 + (24.856)02, for 50 < H < 200 ft.
 or from Figure E-2 if H = 40 ft and D > 8 in.

 See footnote a.-
jf-i
fl
              ^i
              fl .s.
Flare installed cost, C'f|
     (Oct. 1982 $)

Total installed piping costs, C'D
     (Aug. 1978 $)

Total installed ducting costs, C'
     (Dec. 1977 $)

June 1980 Installed costs

       Piping0, Cp

       Ductingd, Cd

       Flare5, Cfl

       Total  flare system cost,
         csys
 c' 'fl  x 2.1
 Method of Appendix E.7
 Method of Appendix E.7
 C'p x 1.206

 C'd x 1.288
 C1
   fl  x 0.818
(Cp 4- Cd + Cf!
                                  E-12

-------
     Table E-3.   CAPITAL AND ANNUAL OPERATING COST ESTIMATION
                PROCEDURE FOR STATE-OF-THE-ART STEAM-ASSISTED
    	SMOKELESS FLARES (Concluded)
           Item
               Value
 Annual 1 zed  Costsf
 Operating labor,
 Maintenance,  Cm
 Utilities
       scfh
      scfh
Cost natural gas, C_ _
                   ri . y
Cost steam,
Taxes, admin. & insurance,
Capital recovery, Ccr
Total  annualized,
 620 hr/yr x $18/hr = $11,160
 0.05 x C
         sys
 80 scfh, for 2 < D < 8;
160 scfh, for 10 < D < 20;
240 scfh, for D = 24;
320 scfh, for D = 60.
[(0.3272)(D 1n)2 -
0.366l£ C(Qn.n.). x
       i         ""
                                      ".45
                        aux
                                             (Qscfh) purge)]
                                               n .g.
See footnote j
Csys x 0.04
0.1315 Cfl  + 0.1627 (Cp + Cd)
cl
                                                                        x 60
                                             m
           n.g
                                                                 cr
                               Ctax
                                  E-13

-------
Footnotes for Table E-3
aFluidic seal  is costed only if cost of purge  gas  without  seal  is greater
 than the annualized cost of the seal  plus  any purge  gas required with
 the seal, i.e., taking the October 1982 purchase  cost  of  a seal, Cf|.s.
 from Figure E-3 for D, if
 53.45   y£ x 60
       1scf¥
(°
.372
                                                        scfm
                                     (Dim.))' -(Q^.
         >  (0.1315 + 0.05 + 0.04)
                                    $ capital
         x 0.818 jun.  '80$ x 2.1  installed  x  c     1
                 Oct.  '80$       purchase      rl.s.j


                          (°'451£T^)X [D(1n0]2  '  (QSfl!3.)cont.
   + 53.45
           IcW
 or, simplifying,
      if    1169|D(in.)1  - 3154 (Qscfm )         >  (0.3805  x Cfl  ^  )
                L     J         V  fl.gJcont.                 tl
-------
 Footnotes  for Table  E-3  (Concluded)

  Ensures sufficient  continuous  flow per vendor  information  (See Docket
  Item No.  II-E-20)for  flare with any continuous flow using  a fluidic seal:

     0.45  scfh  >

              in

      _g  (scfm)^ x 60 min/hr x /t1 operating hours per year  \
                                \     at stream combination i/

     +  CQpilot(scfh) + Qpurge(scfh)] x 8760 hr/yr

     x 520°R  scf at 60°F  x 1,040 Btu (HHV) x       $5.98
       '-*•""* scf at 70°R      scf at 60"F(10° 6tu (HHV)
           x (1Qf Btu)

              10 BTU


JFor continuous streams only,  assumes steam at 0.4 Ib/lb of hydrocarbon
 at maximum continuous flaring rate for 8600 hr/yr:
     cont.
                      x MWcont.  x
                                    wt.% HC
                                     x 8600  hr/yr  x 60 min/hr
                                                          1000 (Ib steam)
                                   v   100 ~ I cont.

        x (lbHirole/387  scf at 70°F)  x (0.4  Ib steam/lb  HC)  x  (1000  lb  steam)

        x $6. 18/( 1000 Ib  steam)

 or simplifying,

        3.296[~QW  Q  (scfm)  x MW x  ^|
             L w'9-                  100
                                         fl.g.
 For intermittent streams  only and for combinations of  continuous and
 intermittent  streams,  determine  approximate  steam bleed  rate  (SBR)
 for flare  diameter, D, using  Figure  E-l  (which extrapolates to about
 800 Ib/hr  at  60 in. and 925 Ib/hr at 66  in.)  annual cost for  intermittent
 flow duration, t-}n-t:

            SBR, lb steam/hr  x   (1000 lb steam) •  x    $6.18
               /      ,_,  x       luuu (lb steam)    (1000 Ib steam)
            x  (8600 -Eti) hr/yr
 .
S600  i
           x MW x
                                  HC
                               100
                                          gl
                                           Ji
where i indicates those stream combinations for which the design hourly
steam cost (based on waste gas only since natural  gas does not require
steam for smokeless combustion) exceeds the hourly steam bleed cost,.i.e.,
       r"

         £L. x 3.296  (MW x wt-%
                                       x Q.  - (0.00618 x-SBR) > 0.
                             E-l 5

-------
because all combined streams to thermal incinerators for polymers and
resins regulatory alternatives had sufficient waste gas heating values
to combust at 870°C (1600°F) without preheating the input streams.
Therefore, only the design procedure for high heat content streams,
independently able to sustain combustion at 870°C (1600°F), is detailed
in this section.
     The first step in the design procedure was to calculate the physical
and chemical characteristics affecting combustion of the waste gas
stream from the model plant characteristics given in Chapter 6, using
Table E-4.  In order to prevent an explosion hazard and satisfy insurance
requirements, dilution air would be added to any individual or combined
waste stream with both a lower heating value between 13 and 50 Btu/scf
at 0°C (32°F) (about 25 and 100 percent of the lower explosive limit)15>16
and an oxygen concentration of 12 percent or greater by volume^? that required
preheating to maintain a combustion temperature of 870°C (1600°F).
Dilution air would be added to reduce the lower heating value of the
stream to below 13 Btu/scf.  (Adding dilution air is a more conservative
assumption than the alternative of adding natural gas and is probably
more realistic as other streams often have enough heat content to
sustain the combustion of the combined stream for the regulatory
alternative.)
     The combustion products were then calculated using Table E-5
assuming 18 percent excess air for required additional combustion
air,1***19 but 0 percent excess air for oxygen in the waste gas, i.e.,
oxygen thoroughly mixed with VOC in waste gas.  The procedure would
include a calculation of auxiliary fuel requirements for streams (usually
with heating values less than 60 Btu/scf) unable to achieve stable
combustion at 870°C (1600°F) or greater.  Natural gas was assumed as
the auxiliary fuel as it was noted by vendors as the primary fuel now
being used by industry.  Natural gas requirements would be calculated
using a heat and mass balance assuming a 10 percent heat loss in the
incinerator.20*21  Minimum auxiliary fuel requirements for low heating
content streams would be set at 5 Btu/scf to ensure flame stability.22
     The design procedure for streams, such as the combined streams
costed for the polymers and resins industry, able to maintain combustion
at 870GC (1600°F) is presented in Table E-6.  Fuel was added for
flame stability in amounts that provided as much as 13 percent of the
                                  E-16

-------









oo
0
p
oo

5
UJ
1—
o

fy
Jg^
3=

c

-M
IB

JZ

i.

^"™ *****

S- -U
fO CQ
3 C
0 T-
r- 
-4->
01
o
3

S.
(O
r^
u
CJ
i












X
• C
>i ;£ *"
o» moo
CD • O «— •
• c
" 2
1
f— J3 X
"s" ^ =»
S ^L 5
J3
J2
•i- c o}»*^> a
X^f^s a
O « (O _J
1 "=="~5
•" "*»» •
>, 0 01 ^
*•> V) « trt
o ^ ^j** ^i ^
S ?« ? 2 2 .



" " " " « » „ ,, „
'3~~'5 'S'x "5 -5 -5 "3




..
^
"* — • — .
>0 «?• — . ^5 1^»T "* ° **>
2 o o => as o o_ ^5 S § o S 2 2i ~ 2
~ 2 fS'S gj S S iif S Si ?g" s g g g
CVJ
°1 " »

^5^ '•O C3
-a- S _voS ". ^ ya co""". "^e 2 2 *U
oaz CM esa o "^sj s*' tj^ tjf^ ^— -^_ , ^ — — — ~
= uzo ^^ ii , ," , ^ u 
-------
X














CO
CJ
C-l O)
1— £=
CO T-
t— I •<->
c£ id
o
S &.
IcE •*
as 0
CJ r—

CO •
«C «J
C3 r—
J t.t ^ r~^r
H- E <»-
CO O O
§ i. § ?
u_ ,tJ 4-> -a
O r- 03 3
3 I—
z u e o
o a T- c
>-» i— O
f**» £j (jj CJ
^1* f™ ^jjj ^H^
s ' »  >
«c en
CJ 1—
(U
0 *
U_ J-
It)
tu S
LU U
S 
*
01 ^
u ^^. •
e z 01 -N. a
O V) • tfl
g O »C 91 I
. 5 5

I s.lls§




_
« x c: sow
u ^ o -a u & v * ajso x

?
o e u ^ u x •









5s e^
' »4
irt «

. *«-i5
^ ti

• S •

^ X w* I
I 0-0-




















V
vt
I
0)
o

^
1
c/1
4)

s-
u
x

y
41
• O •
** b* cs
s°^ =•
X U ui
u 3 2.
u as u
2 01 -o
f— 4rf 0>
1 2 1
4) V» « •
i £ t §

. » s "3
w •••• o o
• 1 1 §
§c <2 2
a S
0 u *; *
ita — O —
01 01 • •*•• O
cate an approximate
lues taken from Ain
heats of fonnation
on approximate hea
rapolatlon to appr
1 * i 1 S
•»- CT U wi
- c ^ 2 S
w> •«— d w
Si «I "« -0 «^

QJ o) *j a>
5 *"" — « w
e u 3 3 s
01 
-------
                Table  E-5.   GENERALIZED  WASTE  GAS COMBUSTION CALCULATIONS
 Basis:  per 100  Ib waste gas  (w.g.)
                                                                                     Stream I.O.:
 Assumptions:    (1)  18 percent  excess air (1 02:3.76 N2 by volume or by mole);
               (2)  known waste-gas composition  of C, H, N, 0,  (in Ib-moles of atom  per  100 Ib w.g.};
               (3)  known amount of AIR (in lb-moles/100 Ib w.g.) in waste gas stream  (also included in  N & 0 compositions);
               (4)  oxygen in waste gas (in air  or hydrocarbons) thoroughly mixed  with VOC so only stoichiometric requirement.
 Reaction (in  dry air):
 Cc Hh Hn 0Q + (1.18) (0-f-- f) (02 * 3.76 N2)—*-C(C02) * £ (H.,0) +  £ + (C + f - f)(1.18)(3.76) X(N2)  + (0.18) (C + j - £)  (
Products of  Combustion in  lb/100 Ib w.g.  (wet):
C02:
44.01  Ib y - /  lb-fnole CO,
Ib-mole  * C l—•rr	~
                                    w.g.
                                            (44.01) C
                                                     (4.76) (1.18)]  (0.013)
                                                                                      (29
                                                                                                 air,
                      - (2.118) C + (9.S39) H - (1.059)  0 +  (0.377) AIR
                               ¥
     if no Jtr r-rtr-s
     i.e., (C +• -) S
                      -  (124.3) C * (31.08) H +• (14.01)  N  -  (62.15) 0 «
                                   >  .  (14.01) „
                              (0-u)  (c -
                                                                                           (0.377) AIR
     if no air required.  32.00 Ib rO    ,r  . H., » (-32.0Q)C - (8.00) H + (16.00) 0
     1.«» (c+JijsJ.-   UwnoU t7 '  J
               4    2
Total:   prod.  m Ib  products (wet) ^ lb(CO. + H.Q + N. -rt).)
        w.g.         100 Ib w.g.         100 Ib w.g'l! .......
aPounds  of water per pound of dry air  at SOT and SOS relative humidity.
                                                    E-19

-------
r
















CO
C£
O «•"•»
I— >
S 5
LU •— '
™2J"
l-« CO
O LU
« (O
S CO H-
3 Z 0
LU >— < IO
oc f— ->.
'"" LU 4-»
z =c oa
S « o
CO LU U3
1 o£
O 3=
H- nz i—
LU H-« Q±
CC r£ LU
=3 I—
C3 CO «C
LU Sp. LU
O LU O
a: ce
Q. 1—
CO


* r*S

I t— i
LU (—
CO
(U =3


,
t— CJ -

















•




























































































T»
j= a:^-
*x" Cio
3 CM •
IS CO O
M in o voo





















^.
z
_J
CM Z
O — J
Ort"
CMO
2S°°




'C in a otn o a
u r»O Or*» o in
"3 vi VI VI VI VI VI
S S553S55
=> — * ^ «^ -J
3 V V V
1 £11









_l— 1^
V)
IS
o>
Is
3

Jn^
x

u ia
j^
^^

•B-S*


X Jl
a 2
VI
s
"1
VI f^

u
•^» o

»
^ «I«J
V V V
in oo
^^ i*^ in











































O
CM
=
J3 OO


3 — '

CO X-

= "t
^ ^^
xl • i~t
II *CM 5^ ?
+ 8 -C ^
— > 1 W
— i «r o ea
^ _^ ii ^ ^te
S" "" X K
en x to -j
• in • x
o en m ^^^
in x

<•»• CM >«
en .— •_ ^ ^
Is a x o o
_ >_ — M
"g 1 X X
^-1 O CM LO
— • a en 10
U CO
•*» »^ in **



-
=
a
TT
VI


S vl

U U

i«
^••v*
VI
O A


CO
oca


33


^ *^ «*
(O (O -M
»— S. «
3 a> e
JiS-S
fO 35 vi
CJ •*•> OJ

^•^
ca
E-20









ipj
^jP.





fcj
VJ W
l^r
n



-7
01
1
X
3
IS
>« —
2^
"7 i
O is

ja
X r—

^O
2
**^
JS
1—
O)
X
IS
_:
•a
sr
9
•g

en a»
a>
£i •
r-» C
_ cr


*•*
'i. "i
•^ t-

















j=
(•
n* h^a K
II X •£
— , ^^11 O) VI
en
X
|
tn
to

1
i
•CJ CMlCM
o o loo
O. •*•

en S
? -r o?
x o> •— •
5 * «r
0 £ O
o —<
— i o u
-^. O i- X
VI i— t IS
0) •-» CO
F« (
2 £
/t a) en
- "o in
± II 7
— » ^ ^.
. C- "^ 1"
^ •— 

tu
•W VI

. 3
U
IS IS
u«-^
^•^
PO








(^
o

VI

u
^
Q.

V)
IS
O)

IS
I
IS ••
CO)
s

U i—
f— VI

11 '
is U

^2
•«^























• • fO
i. cn
^>

— o

-> —
ij -o





o-
1
a
X
«n
CM
u

u
u







e»




•o


-------
mo

         » g  t. o  i-
 s_   u
 4)   3
 0.  4J
     19
 VI   C
4J
 U   <*-
 3   O
^!
 O   4)
 i.   i.
 a.  3
                                               §   _
                                  C   19   4J

                                  O   U   Ul
                                  •—   41   3
                                  4-1   Q.  ^



                                  ja.  4J   U
 U   4)   4J

 Ul  "c   S.
 19  1-   n-
 01      19
     o
 41   ui   e
*•>       o
 Ul  4J   -r-
 19   i«   4J
 3   OJ   3

*.  -s   -

 0   &.  •»
                                  ja   o  —  —
           c
 c     •   o
 41   r—   -r-
 OJ    O   4-1
 S    >   ul
 4->   —•   3
 41        J2
                                                    i-   t.
                                               ^   ul   C
                                               i—   O  t_
                                                3
                                               -W
                                               03
      O   4)

      u   S
 E:    ui   u-
t-    19   41
      en  s-
«^       41
 19  <9  >»-
 41  e   u
Ib

no
02
         ^
          c
     -•   19
     >w^
          VI
     • •   19
  •  vi   en
  CM s
 Z  O   41
     •*»  4>J
  •  4J   Ul
 o  a.   19

 ^•s   *
     ui   s
  "  US  •»—

 ^"   c
 u  en   41
     e   en

^  s   S-
 ^-  o   o
  I   r—

     *O   41
 ai  vt.   s.

 3  c  "3
 a  o   cj
                                                u
                                                41   **—
                                                Q.  a
00
                                                L- 1
                                               t-   19
                      O
                      vo
                      VO

                      cs"
                     I CM
                                                                   CM


                                                                    X
                                                                   CM
                                                                   Ol
                                                                    CM

                                                                   S
                                                                   CJ


                                                                   X
                                                                   CM
                                                                   O
                                                                   CM


                                                                   X
                                                                   CM

                                                                   01
                                                                  u^







                                                                    CM
                                                                  S

                                                                  va
                                                                                  01

                                                                                - c

                                                                                  3

                                                                                  CO


                                                                                  g
                                                                                  vo

                                                                                  o"       "s
                                                                                  CM
                                                                                  —       O
                                                                                                                            X


                                                                                                                            c
                                                                                                                   o
                                                                                                                   o

                                                                                  f
                                                                                                                              en


                                                                                                                            c?

                                                                                                                            X


                                                                                                                            en

                                                                                                                            2
ed
                                           c
                                  e   vi   3
                                  19   19
                                      .a   4)
                                 CM       4J
                                      Q.   US

                                 *   S   a
                                 U
                                               u   s-

                                               3  £

                                               •w  en
                                               e  *
                                               4)  CM
                                                4)
                                                a.
LHV
eq
                                  §  S.
                                  S-  4->
                                      ?  t
             •w   o
              19   <*-


              X  01
             4?  VO
             i-   CM
                                  S   .  &
                                  41  VO  vo
                                      4i   o
                                                u  —
                                                O  LO
                                                         41

                                                         a
                                                         41
                       S  ••
                       19  VI

                        CO VI

                       =5,5
LU


 0)



 A3
 

 O
 O
                                  s-
                                  41
                                  O,
                                               Ol   CM

                                               ^  S
                                               19
                                                         CM  e
                                                        u   o
         O   >»  CM   ^  19
         t-   C   .       en
         4J   19  O    »
         3  •—        ^r F—


         •5  S-           3
              X'  U.   »«  4J
         •8   3  5_   CM  19
              19  O   O>  C
         vl       O
   . 	   19  Vb.  CO   Vi-  vt-
    —>   en   o  —•    o  o
                                                                   CO  CM
                                                                   0  0
                                                                   CM,  as
                                                                   a  o
                                                                    PO
                                                                   0
                    U  CM
                        O

                    o  o

                    o  "~~

                    •*•
                                                                   0  0
                                                                                    19
                                                                                    e
                                                                                   09
                                                                                   VO
of
on
                                           I.

                                          v£



                                          "8
                                           VI
                                           3

                                           I_

                                           19
o> «
• in
S 19
1— Ol

i 2
CQ VI
** 19
— - S
Ul H-
•w o
u
3 41
•g s.
O 3
U +»
Q-io
1_
Ul 41
4*»

LU
a
O
VS
E.
O
Vk.
Ul
19
U.
o
i
CO
»B4
o
u

A
Ul
19
en
pM,
«
t,
3
S

1


                                                                                                                            3E

                                                                                                                            .a
                                                                                                                                      41
                                                                                                                                      U
                                                                                                                                      0)
                                                                                                                                      u
                                                                                                                                      0)
                                                                                                                                     01
                                                                                                                                     o:
                                                                                                   -S1     .
                                                                                                   I    I1
                                                                                                    *    Is
                                                                                                    §1   •$
ore
he
abrica
field
                                                                                                         1     3
equ
                                                                                 X
                                                                                 3
                                                                                 t)
                                                                              9)

                                                                              3
                                                                                                                                     ~   1
                                                                                                   a    LO
                                                                                                        CM
                                                                                                   —-    41
                                                                                                   O     w
                                                                                                   VO    •—
                                                                                                   ««•     vi
would r
un
arge
                                           f

                                           i

                                           i
                                            19
                                                                              Q.
                                                                              s.
                                                                              o
                                    O


                                    °~
                                                                                     CM

                                                                                   S
                                                                                   co
                                                                                   VO
                                                                             E-21

-------
lower heating value of the waste gas for streams with heating values of
650 Btu/scf or less.  For streams containing more than 650 Btu/scf,
flame stability fuel requirements were assumed to be zero since coke
oven gas, which sustains a stable flame, contains only about 590 Btu/scf.
In order to prevent damage to incinerator construction materials,
quench air was added to reduce the combustion temperature to below the
incinerator design temperature of 980°C (1800°F) for the cost curve
given by IT Enviroscience.23
     The total flue gas was then calculated by summing the products of
combustion of the waste gas and natural gas along with the dilution
air.  The required  combustion chamber volume was then calculated for a
residence time of 0.75 sec, conservatively  oversizing by 5 percent
according to standard industry practice.24  The design procedure
assumed  a minimum commercially available size of 1.01 m3  (35.7  ft3)
based on vendor information25 and a  maximum shop-assembled unit size
of 205 m3 (7,238 ft3).26
     The design procedure  could allow for pretreating of  combustion
air, natural gas, and when permitted by insurance guidelines, waste
gas  using a  recuperative  heat exchanger in  order to  reduce the  natural
gas  required to maintain  a 870°C  (1600°F) combustion temperature.
However, all streams  to  thermal  incinerators  for polymers and  resins
regulatory alternatives  had  sufficient waste  gas heat contents  to
combust at greater  than  870°C  (1600°F), and even at  greater  than 980°C
 (1800°), without  preheating  the  input streams.   If a plant had  a use
for  it,  heat could be  recovered.   (In fact, a waste  heat boiler can  be
used to generate  steam,  generally with  a  net  cost  savings.)
E.3.2   Thermal  Incinerator Cost Estimation  Procedure
     Thermal incinerator purchase  costs  for the calculated  combustion
 chamber volume were taken directly  from Figure  E-4,  (Figure  A-l in the
 IT Enviroscience  document,  Reference 23).   An installation  cost factor
 of 4.0  (see  Table 8-2)  was used based  on  the  Enviroscience  document.27
 The installed cost used for the minimum commercially available size of
 1.01 m3 (35.7 ft3) was $217,800 (June 1980; based  on a  December 1979
 purchase cost of $52,000).  The installed cost of  one 150-ft.  duct to the
 incinerator and its associated fan and stack  were  taken directly
 from Figure E-5 (Figure IV-15,  curve 3  in the IT Enviroscience study28).
                                   E-22

-------
                                                                     n

                                                                     H-
                                                                     U.
                                                                     U
 O
 >

 IU


 o

 £
 U.
 ui

 a:
 in
<

«
o

en

a

o
o
                                                                          in
                                                                          s_
                                                                          
-------
                          o
                          o
                                                                 CO
                                                                  CO
                                                                           o
                                                                           
                                                                         (/> rC
                                                                         (T3
                                                                            O)
(000*13)  IVlldVD  d2~~iVlSNI 6161
                                                                         CO
                                                                         +J OO
                                                                         Si.
                                                                          CO S-
                                                                          •(-> 
                                                                         4-» 00
                                                                         •r»
                                                                          Q.-O
                                                                          «3 C
                                                                         O (O

                                                                         T3  •»
                                                                          O) 
-------
A minimum  installed  cost  of  $70,000  (in December  1979) was assumed for
waste  gas  streams with  flows  below 500 scfm.  The costs of piping or
ducting  from  the process  sources  to  the 150-ft. duct  noted above were
estimated  as  for flares.   Installed  costs were put on a June 1980 basis
using  the  following  Chemical  Engineering Plant Cost Indices:  the
overall  index for thermal  incinerators; the pipes, valves, and fittings
index  for  piping; and the  fabricated equipment index  for ducts, fans,
and stacks.   Annualized costs were calculated using the factors in
Table  8-3.  The electricity required was calculated assuming a l.SkPa
(6-inch  H20)  pressure drop across the system and a blower efficiency of
60 percent.   The cost calculation procedure is given  in Table E-7.
E.4  CATALYTIC INCINERATOR DESIGN AND COST ESTIMATION PROCEDURE
     Catalytic incinerators are generally cost effective VOC control
devices  for low concentration streams.  The catalyst increases the
chemical rate of oxidation allowing  the reaction to proceed at a lower
energy level  (temperature) and thus requiring a smaller oxidation
chamber, less expensive materials, and much less auxiliary fuel
(especially for low  concentration streams) than required by a thermal
incinerator.  The primary determinant of catalytic incinerator capital
cost is  volumetric flow rate.  Annual operating costs are dependent on
emission rates, molecular weights, VOC concentration, and temperature.
Catalytic  incineration in conjunction with a recuperative heat exchanger
can reduce overall fuel  requirements.
E.4.1  Catalytic Incinerator Design Procedure
     The basic equipment components of a catalytic incinerator include
a blower, burner,  mixing chamber,  catalyst bed,  an optional  heat
exchanger,  stack,  controls, instrumentation, and control  panels.   The
burner is used to preheat the gas  to catalyst temperature.   There is
essentially no fume  retention requirement.   The preheat temperature is
determined by the VOC content of gas, the VOC destruction  efficiency,
and the  type and amount of catalyst required.   A sufficient amount of
air must be available in the gas or be supplied to the preheater for
VOC combustion.   (All the gas streams for which catalytic  incinerator
control  system costs were developed are dilute enough in  air and
therefore require no additional  combustion air.)   The VOC  components
contained in the gas streams include ethylene, n-hexane,  and other
                                  :-25

-------
        TABLE E-7.  CAPITAL AND ANNUAL OPERATING COST ESTIMATION
        PROCEDURE FOR THERMAL INCINERATORS WITHOUT HEAT RECOVERY
        ITEM
               VALUE
Capital  Costs

Combustion Chamber

 Purchase cost
 Installed cost
 Installed cost, June 1980*

Piping & Ducting (from sources
 to main incinerator duct)

 Installed cost

 Installed cost, June 1980b
Ducts, Fans & Stacks (from
 main duct to incinerator
 and from incinerator to
 atmosphere)

 Installed cost0
 Installed cost, June 1980d

Total Installed Cost, June 1980



Annualized Costs6

 Operating labor
 Maintenance material & labor
 Utilities

   natural gas^


   electricity 9

 Capital recovery"

 Taxes, administration & insurance

 Total Annualized Cost
from Figure E-4 for Vcc
purchase cost x 4.0
installed cost x 1.047
see Section E.7 for Qw.g. (scfm)

installed cost x 1.206 for piping
installed cost x 1.288 for ducting
from Figure E-5 for Qw>q.;
  use $70,000 minimum

installed cost x T.064

sum of combustion chamber,
piping & ducting, and ducts,
fans, & stacks
1200 hr/yr x $18/hr = $21,600
0.05 x total installed cost


(5.639 x ID'4) (% aux) x LHVW _
       x Qw.g. db/hr)      W'9'

(0.4955) x Qf>g> (scfm)

0.1627 x total installed cost

0.04 x total installed cost

operating labor + maintenance
  + utilities + capital recovery
  + taxes, administration &
  insurance
                              E-26

-------
 Footnotes  for Table E-7

 allpdated using Chemical Engineering Plant Cost  Index from December 1979
  (247.6) to June  1980  (259.2).

 bPiping updated using .Chemical Engineering Plant Cost pipes, valves,
  and fittings index from August 1978 (273.1) to June 1980 (329.3).
  Ducting updated  using Chemical Engineering Plant Cost fabricated
  equipment index  from December 1977 (226.2) to June 1980 (291.3).

 cFrom Figure E-5  for no heat recovery from Enviroscience (Reference 28),
  which assumed 150-ft of round steel inlet ductwork with four ells,
  one expansion joint, and one damper with actuator; and costed according
  to the 6ARD Manual (Reference 29).  Fans were assumed for both waste
  gas and combustion air using the ratios developed for a "typical
  hydrocarbon" and various estimated pressure drops and were costed
  using the Richardson Rapid System (Reference 30).  Stack costs were
  estimated by Enviroscience based on cost data received from one
  thermal  oxidizer vendor.

  Although these Enviroscience estimates were developed for lower
  heating value waste gases using a "typical  hydrocarbon" and no dilution
 to limit combustion temperature, the costs were used directly because
  Enviroscience found variations in duct, etc., design to cause only
  small  variations in total  system cost.  Also, since the duct, fan,
 and stack costs  are based on different flow rates (waste gas, combustion
 air and waste gas, and flue gas, respectively) the costs can not be
 separated to be  adjusted individually.

^Updated using Chemical Engineering Plant Cost fabricated equipment
 index from December 1979 (273.7) to June 1980 (291.3).

eCost factors presented in Table 8-3.

f[(% aux)  x LHVwgj x Qw>g.  (Ib/hr) x

  (100 lbw.g.)/100(lbw.g.)  x (8600 hr/yr) x (lb-mole/17.4 lbn.gj x

  (379 scf at  60°F/lb-mole)  x (1040 Btu(HHV)/scf at 60°F) x  $5.98/106

 Btu (HHV)  x  (106 Btu)/K)6 (Btu).

SElectricity  = (6 in.  H20 pressure drop) x Qf>g< (scfm)  x (8600 hrs/yr)

 x (0.7457 kW/hp) x (5.204 Ib/ft2/in.  H20) -h [(60 sec/min)  x (550 ft-lb/

 sec/hp)  x (0.6 kW blower/1 kW electric) x $0.049/kWh].

h!0 percent interest (before taxes)  and 10 yr. life.
                              E-27

-------
easily oxidizable components.  These VOC components have catalytic
Ignition temperatures below 315°C (600°F).   The catalyst bed outlet
temperature is determined by gas VOC content.  Catalysts can be operated
up to a temperature of 700°C (1,300°F).  However, continuous use of
the catalyst at this high temperature may cause accelerated thermal
aging due to recrystallization.
     The catalyst bed size required depends upon the type of catalyst
used and the VOC destruction efficiency desired.  About 1.5 ft3 of
catalyst for 1,000 scfm is required for 90 percent control efficiency
and 2.25 ft3 is required for 98 percent control efficiency.31  As
discussed earlier many factors influence the catalyst life.  Typically
the catalyst may loose its effectiveness gradually over a period of
2 to 10 years.  In this report the catalyst is assumed to be replaced
every 3 years.
     Heat exchanger requirements are determined by gas inlet temperature
and preheater temperature.  A minimum practical heat exchanger efficiency
is about 30 percent.  Gas temperature, preheater temperature, gas dew
point temperature and gas VOC content determine the maximum feasible
heat exchanger efficiency.  A maximum heat exchanger efficiency of 65
percent was assumed for this analysis.  The procedure used to calculate
fuel requirements is presented in Table E-8.  Estimated fuel requirements
and costs are based on using natural gas, although either oil (No. 1
or 2) or gas can be used.  Fuel requirements are drastically reduced
when a heat exchanger is used.  Total heat requirements are based on a
preheat temperature of 600°F.  A stack is used to vent flue gas to the
atmosphere.
E.4.2  Catalytic Incinerator Cost Estimation Procedure
     The capital cost of a catalytic incinerator system is usually
based on gas volume flow rate at standard conditions.  For'catalytic
incineration, 70°F and 1 afm (0 psig) were taken as standard conditions. '
The operating costs are determined from the gas flow rate and other
conditions such as gas VOC content and temperature.  Table E-9 presents
the basic gas parameters required for estimating system c.osts.
     As noted earlier, equipment components of a catalytic incineration
system include blower, preheater with a burner, mixing chamber, catalyst
bed, an optional heat exchanger, stack, controls, and internal ducting
                                  E-28

-------
     Table E-8.  OPERATING PARAMETERS AND FUEL REQUIREMENTS
                     OF CATALYTIC INCINERATOR SYSTEMS
     Item
  Source of information  or calculation
Waste Gas Parameters

(1)  Flow rate (Q2)» scfm

(2)  Amount of air present in
     the gas, scfm
(3)  Amount of air required
     for combustion at 20%
     excess, scfm

(4)  Net amount of additional
     air required (0.3), scfm

(5)  Total  amount of gas to be
     treated (0.4), scfm

(6)  Waste  gas Temperature at
     the inlet of PHR&, °F

(7)  Waste  gas temperature at
     preheater outlet or
     catalyst bed inlet, °F

(8)  Temperature rise in the
     catalyst bed, °F

(9)  Flue gas temperature at
     catalyst bed outlet, °F

(10)  Minimum possible temperature
     of flue gas at PHR outlet, °F

(11)  PHR efficiency at maximum
     possible heat recovery01, %
(12)  PHR design  efficiency,  %
      From Table E-9

  0, if the waste gas contains VOC and
  nitrogen or other inert gas; and
  C(l - volume percent VOC) * (volume
  percent VOC)] x VOC volume flow  (Qj),
  scfm, if the waste gas contains VOC
  and air

  See footnote a.
 Item (3) - Item  (2); and 0 if
 [Item (3) - Item  (2)] is negative

 Item (1) + Item  (4)
 From Table E-9
 600°F
(25°F/1% LEL) x (%LEL from Table E-9)
Item (7) + Item (8)
See footnote C.
[Item (1) x (Item (7) - 25°F -
Item (6))] * [Item (5) x (Item (9) -
Item (6))]e

See footnote f
                                 E-29

-------
          Table E-8.   OPERATING PARAMETERS  AND  FUEL  REQUIREMENTS
                     OF CATALYTIC INCINERATOR SYSTEM (concluded)
     Item
Source of information or calculation
(13) Waste gas  temperature at
     PHR outlet,°F

(14) Amount of  heat required by
     preheater  at additional 10%
     for auxiliary, Btu/min

(15) Amount of  heat required
     for preheater and auxiliary
     fuel, 106  Btu/h

(16) Amount of  natural gas
     required per year, 10^ cfm
0.65 [Item (9) - Item (6)] + Item (6)
Item (5) x [Item (7) - Item (13)] x
[Gas specific heat9, Btu/scf, °F] x
[Item (14) x 60 minutes/hour] x (10%)"
x (106 Btu)/lQ6 Btu
[Item (14) x (8,600 x 60) minutes/year]
x 10-3 r (1,040 Btu/cfm)
aOn volume basis (scfm/scfm):  11.45 for methane, 20.02 for ethane, 28.58 for
 propane, 54.31 for hexane, 17.15 for ethylene, and 45.73 for pentane.
 Values taken from p. 6-2 in Steam (Reference 18) for 100% total.air and
 multiplied by 1.2 for 120% total air or 20% excess air.

^Primary heat recovery unit.

cHeat exchanger should be designed for at least 50°F above the gas dew point.

dThe heat exchanger will be designed for 25°F lower than the preheater
 temperature so as to not cause changes in catalyst bed outlet temperature.

eThough the heat recovery to the temperature level of inlet gas is the
 maximum heat efficiency possible, in some cases this may not be possible
 due to gas dew point condition.

fCost estimates are based on calculated maximum possible heat recovery
 up to an upper limit of 65 percent heat recovery.

9Gas specific heat varies with composition and temperature.  Used 0.019 Btu/ft3°F
 based on average specific heat of air for calculation purpose.

"Auxiliary fuel requirement is assumed to be 10 percent of total.
                                     E-30

-------
        TABLE E-9.  GAS PARAMETERS USED FOR ESTIMATING CAPITAL AND
               OPERATING COSTS OF CATALYTIC INCINERATORSa
          ITEM
                   VALUE
Stream identification

Stream conditions
  Temperature,°F
  Pressure,  psig
  VOC content:
     Emission  factor, kg/Mg
       of  product
     Weight  %  of total  gas
     Mass  flow rate,  kg/h
          Ib/h

    Organic constituents, wt %
    Average mol. wt.  (MI), Ibs
    Volume flow  (Qi), scfm
    Heat content  (HI),
      Btu/scf

 Total gas:
    Constituents
    Mass flow rate, Ib/h
    Molecular weight  (M2)
    Volume flow  (Qg), scfm
    Air volume flow rate, scfm
    VOC concentration (A),
      of LEL

    Heat content
      Btu/total scf
 Identify the vent and the polymer
 industry from Chapters 3 and 6
 (Emission factor, E, kg/Mg) x 1000 Mg/Gg
 (Plant production rate, P, Gg/yr) *
 (8,600 h/yr)
 (kg/h) x (2.205 Ib/kg)
 (VOC mass rate,  Ib/h)  * (60 min./h)  }
 (Molecular weight (MI), Ibs/lb mole) x
 (385 scf/lb-mole at 68°F)= 1.645 (EP/MX)
 (174.273)(2.521NC  + NH)C
 VOC,  air and  others
 (VOC  rate,  Ib/h)  *  (wt% of VOC  in
 gas,  Wi/lOOX)
(Gas  mass  rate,  Ib/h)  *  (60  min/h)  •=•
 (Gas molecular  weight  (M2 ),  Ib/lb mole)
 x  (385  ft3  /I b  mole)  =  1.645
 (Total  gas  flow  (Q2),  scfm)  -  (VOC  volume
 flow  (Qi),  scfm)

 (100)  [(Volume flow  of VOC,  scfm) *
 (Volume flow of  air, scfm] * LELd
From Chapter
                                 E-31

-------
Footnotes for Table E-9

^Obtain gas parameters from Chapter 3 of the BID, except those to be
 calculated.

bCalculate using weight  percent values of VOC components.

cif the VOC heating value is not available, calculate it using on heat of
 combustion values of 14,093 Btu/lb from carbon converted to C02 and
 51,623 Btu/lb from hydrogen converted to water.  Nc and NH denote number
 carbon and hydrogen atoms in VOC.

dLower explosion levels  of ethylene, hexane, methanol, propane, butane,
 and pentane are 3.1, 1.32, 7.3, and  2.5, 1.9, and 1.4, respectively.

eTotal gas heat content averages 50 Btu/scf at 100 percent LEL.
                               E-32

-------
 including bypass.   Calculations  for capital  cost estimates  are  based
 on equipment purchase costs obtained from vendors 31,32,33  and  application
 of direct and indirect cost factors.   Table  E-10 presents third quarter
 1982  purchase costs of catalyst  incinerator  systems  with and without
 heat  exchangers  for sizes  from 1,000 scfm to 50,000  scfm.   The  cost
 data  are  based on  carbon steel for  incinerator  systems  and  stainless
 steel  for heat exchangers.   The  heat exchanger  costs are based  on
 65 percent heat  recovery.   Catalytic  incinerator systems of gas volumes
 higher than  50,000  scfm can be estimated  by  considering two equal
 volume units in  the system.   A minimum available unit size  of 500 scfm
 was assumed.34'35   The installed  cost of  this minimum size  unit (which can
 be used without  addition of gas or  air for stream flows greater than
 about  150 scfm35) was  estimated to  be $53,000 (June  1980).  The heat
 exchangers  for small  size  systems would be costly  and may not be practical.
 Table  8-2 presents  the direct and indirect installation cost component
 factors used for estimating capital  costs of catalytic  incinerator
 systems.  The  geometric mean  of the  two vendor estimates for each flow
 rate was  multiplied  by the  ratio of  total installed  costs to equipment
 purchase  costs of 1.82 developed for  a skid-mounted  catalytic incinerator.
 Actual  direct  and indirect  cost factors depend upon  the plant specific
 conditions and may  vary with  system sizes.
     Since the equipment purchase cost presented in Table E-10
 represents the third quarter  of 1982,  the cost data was adjusted to
 represent June 1980 by using  a cost index multiplying factor of
 82.3 percent (based on Chemical Engineering plant cost indices of
 259.2  for June 1980 and 315.1 for August 1982).   The direct and
 indirect  capital cost  factors were applied to the adjusted purchase
 costs  and the  resultant estimates of  catalytic incinerator installed
 capital costs  as of June 1980 are presented in Figure E-6.
     Installed costs of piping, ducts, fans,  and stacks were estimated
 by the  same  procedure  as for  thermal  incinerators.  Installed  costs
were put  on  a June 1980 basis using the following Chemical  Engineering
 Plant Cost indices:   the overall  index for catalytic incinerators;  the
 pipes, valves, and fittings index for piping; and the fabricated
equipment index for ducts,  fans,  and stacks.
                                  E-33

-------










CM

M
_^^
(D
SI
O

f—
oo
o
u

^•H

UJ
>
GC

|_
s


2
O
z



u
*•" *
J^
3:
u.
s
c^


o
UJ

LU

CQ
^^








































































1
u
£
o
SI


o
u
i.
is
V*-

4^
VI
3
OJ
VI

J=

3
^
e
^

3
a
u
u
o
IS
i.
G
U
s

^J
4J



**
u
«
OJ


»«
VO
4^

3
S
01



CQ


o
^3
e


c
A3
=
U


>J
i.
0)
o
u
0)
w




«K
&.
|

J






-^
St^
U
U--—

80 O
0 0
CM CO CM










O O O
« • *
o in o
e> co o
— 1 VO



o o o
§0 0
o o
* •» •
fn CM o
CM O
1 *
i 7
CM g
a



§o o
o o


09 CM 
CM » a








o o o

o" vn" o"
^ CM CO
*•« *^


000
• * *
•O Ul O
CM r^ o
— m
i
f^
vn





0 S °
o o o

"* 2 2.












































•
W1
/a
%
J
ffli
-"
s.
4=2
s.
ns
3
TS
t.
JS
trt
1.1
a
«°
E-34


-------
     5,000
o
o
o
i-H
•fcfl-
+J
«/>
o
 (O
 Q.

 IQ
•o
 


^







































-








Key:
With 65%
Without h








^
•^

r
X
/
S






teat r
;at re








.X
^

s'.
^









2COV
cove








/












jry
•y








^x






















-X
/






















•
x






















X
























^x*




















X
X X
tj? >^ —
fill













>





X
X
/:





^













X

















•7*-


x

















X
b

.
1


















/






























































.51 10 100
Gas Flow Rate (1000 scfm)
Figure E-6. Installed Capital Costs for Catalytic Incinerators
.-.' With and Without Heat Recovery
                                                E-35

-------
     Table 8-3 presents cost bases used for annualized cost estimates.
The operating labor requirement value is based on conversations with
vendors.  The capital recovery factor is based on capital recovery
period of 10 years and an interest rate (before taxes) of 10 percent.
(Actually the current tax regulations allow the control system owners
to depreciate the total capital expenditure in the first 5 years.)
Fuel cost is the major direct cost item.
     The total annual operating costs are calculated using the cost
bases shown in Table 8-3 and the fuel requirements calculated in
Table E-8.   Table E-ll presents a procedure for calculating total
annualized cost estimates of catalytic  incinerators.
     The amount of catalyst required usually depends upon the control
efficiency.  According to a vendor,33 typical catalyst costs are about
$3,000 per ft3.  Indirect additional costs involved in replacing the
catalyst every 3 years are assumed to be 20 percent.  Therefore, for
98 percent efficient systems, the annual catalyst replacement costs
amount to $2.70/scfm.
     Electricity cost calculations are  based on pressure drops of
4 in. water for systems with no heat recovery and 10 in. water for
systems with heat recovery, and at 10 percent additional electricity
required for instrumentation, controls, and miscellaneous.  Therefore,
at the conversion rate of 0.0001575 hp  per inch of water pressure
drop per cubic foot per minute, 65 percent motor efficiency, and $0.049/kWh
electricity unit cost, the total annual electricity costs amount to
$0.335/scfrn for units with no heat recovery (i.e., for 4 in. H20 pressure
drop) and $0.838/scfm for units with heat recovery (i.e., for 10 in. ^0
pressure drop).
E.5  SURFACE CONDENSER DESIGN AND COST  ESTIMATION PROCEDURE
     This section presents the details  of the procedure used for
sizing and estimating the costs of condenser systems applied to the
combined material recovery streams from the continuous process polystyrene
model plant and from the DMT-process polyethylene (terephthalate) model
plant.  An outline of the design and costing of the condenser system
for  the polystyrene model plant is presented in this section as an
example of the procedure used.  Details of either application are given
in the docket.36
                                  E-36

-------
           Table E-ll.  CAPITAL AND OPERATIONS COST ESTIMATION FOR
          	           CATALYTIC INCINERATOR SYSTEMS
            Item
                                                       Value
 Capital  Costs
     Incineration system
       Installed cost, June 1980

     Piping & ducting (from sources
       to main incinerator duct)

       Installed cost
       Installed  cost,  June  1980a
    Ducts, fans & stacks  (from main duct
      to incinerator and  from incinerator
      to atmosphere)

      Installed costb
      Installed cost, June 1980C

    Total Installed Cost, June 1980
Annualized Costs
  Direct costs
    Operating labor
    Maintenance material  and
      labor

    Catalyst  requirement
    Utilities:
      Fuel  (natural  gas)
      Electricity
  From  Figure  E-6
 See Section E-7 for  source flow
 rates, scfm.

 Installed cost x  1.206 for piping
 Installed cost x  1.288 for ducting
 From Figure E-5 for waste gas flow
 (0,2), scfm; use $70,000 minimum

 Installed cost x 1.064

 Sum of incineration systems,
 piping & ducting,  and ducts,
 fans, & stacks
 $11,200 for systems  with  no heat
 recovery;  and  $16,700 for systems
 with  heat  recovery

 (0.05)  x  (Total  installed capital
 cost, $ from Figure  E-6)

 $2.7  x  (Total  gas volume  flow(Qd)a,
 scfm, item  5 from Table E-8) =
 ($2.7 x Q4)

 ($6.22/lQ3ft3) x (Amount  of natural
 gas required,  I03ft3,  item  16 of
 Table E-8)e

 ($0.335/scfm) x (Total gas volume
 flow rate (Q4), scfm,  Item 5 from
 Table E-8) for units with no heat
 recovery; and  ($0.838/scfm) x (Total
 gas volume flow rate (0.4), scfm,
 Item 5 from Table E-8) for units with
heat recovery
                                    E-37

-------
          Table E-ll.  CAPITAL AND OPERATIONG COST ESTIMATION FOR
                       CATALYTIC INCINERATOR SYSTEMS (Concluded)
           Item
         Value
Indirect Costs
    Capital recovery
    Taxes, insurance and
      administrative charges
Total Annualized Costs
(0.1627) x (Total installed capital
cost, $ from Figure E-6)

(0.04) x (Total installed capital
cost, administrative charges
$ from Figure E-6)

Sum of total  di rect costs and
total indirect costs
aUpdated using Chemical Engineering Plant Cost Index from December 1979
 (247.6) to June 1980 (259.2).

^Piping updated using Chemical Engineering Plant Cost pipes, valves, and
 fittings index from August 1978 (273.1) to June 1980 (329.3).  Ducting
 updated using Chemical Engineering Plant Cost fabricated equipment index
 from December 1977 (226.2) to June 1980 (291.3).

GSee footnote c, Table E-7 for discussion on application of these costs
 developed by Enviroscience (Reference 28).

^Updated using Chemical Engineering Plant Cost fabricated equipment index
 from December 1979 (273.7) to June 1980 (291.3).

eTotal gas flow including waste gas and additional  combustion air.
                                    E-38

-------
Two  types of condensers  are  In use  in  the  industry:   surface
condensers  in which  the  coolant does riot contact the  gas or condensate;
and  contact condensers in which coolant, gas, and condensate are intimately
mixed.
     Surface condensers  were evaluated for the combination of the
following two streams from the polystyrene model plant:  the styrene
condenser vent and the styrene recovery unit condenser vent.  These
streams may consist  of styrene and  steam, which are immiscible, or of
styrene in air, a non-condensable.  The nature of components present in
the  gas stream determines the method of condensation:  isothermal or
non-isothermal.  The condensation method for streams  containing either
a pure component or  a mixture of two immiscible components is isothermal.
In the isothermal condensation of two immiscible components, such as
styrene and steam, the components condense at the saturation temperature
and yield two immiscible liquid condensates.  The saturation temperature
is reached when the  vapor pressure  of the components  equals the total
pressure of the system.   The entire amount of vapors  can be condensed
by isothermal  condensation.  Once the condensation temperature is
determined, the total heat load is  calculated and the corresponding
heat exchanger system size is estimated.
     The condensation of styrene mixed with a non-condensable, such as
air,  can not be considered to be isothermal.  Therefore, systems to
condense in the presence of a non-condensable are usually designed
by considering the heat exchanger to be divided into a number of sections,
each  with a portion of the total  temperature drop.   In general, the
condensation of styrene   in air is accomplished less readily, and thus
more  expensively, than the condensation of styrene in steam.
     As new plants were  assumed to  use vacuum pumps, which result in
styrene-in-air emissions, the costs of condensing styrene in air were
estimated for  regulatory alternatives of 90 percent through 98.5 percent
styrene emission reduction.  The condenser inlet stream was assumed to
contain saturated styrene in air at 27°C (80°F)  since most industry
data  showed the emission stream to  be near that temperature, and since
the stream was the outlet from previous process or emission control
condensers that would emit at or near saturation.  The following general
procedure and  assumptions were used in evaluating the condensation
                                  E-39

-------
systems for the combined streams containing styrene in air from the
continuous polystyrene model plant.  The presented procedure, however,
is for the entire heat exchanger while the actual procedure36 utilized
an iterative, multiple section analysis.
E.5.1  Surface Condenser Design
     The condenser system evaluated consists of a shell and tube heat
exchanger with the hot fluid in the shell side and the cold fluid in
the tube side.  The system condensation temperature is determined from
the total pressure of the gas and vapor pressure data for styrene 'in air.
As complete vapor pressure data are not readily available, the conden-
sation temperature is estimated by a regression equation of available
data points^? using the Clausius Clapeyron equation, which relates the
stream pressures to the temperatures.  The total pressure of the stream
is equal to the vapor pressures of individual components at the conden-
sation temperature.  Once the condensation temperature is known, the
total heat load of the condenser is determined from the latent and
sensible heat contents of styrene and air (see Table £-12).  The design
requirements of the condensation system are then determined based on
the heat load and stream characteristics.  The coolant is selected
based on the condensation temperature.  The condenser system is sized
based on the total heat load and the overall heat transfer coefficient
which is established from individual heat transfer coefficients of the
gas stream and the coolant.  An accurate estimate of individual coef-
ficients can be made using such data as viscosity and thermal conductivity
of the gas and coolant and the standard sizes of shell and tube systems
to be used.
     The styrene-in-air refrigerated condenser systems were designed
according to procedures for calculating shell-side,40 tube-side,41
and condensation^ heat transfer coefficients, mass transfer coef-
ficients,^ and, finally,  an overall heat transfer coefficient for
condensation in the presence of a noncondensable using a multiple
section analysis.44 Heat exchanger45»46 an(j refrigeration unit4?
characteristics were developed from vendor information in conjunction
with information given primarily in the Chemical  Engineers'
Handbook.40,48 Refrigerant characteristics were taken primarily from
the Chemical  Engineers'  Handbook49'50 and publications of the American
                                  E-40

-------
                Table E-12.  PROCEDURE TO CALCULATE HEAT LOAD
                 OF A CONDENSATION SYSTEM FOR STYRENE IN AIR
         Item
              Value
Heat exchanger type
Source identification
Source production capacity
 (CAP), Gg/yr

Source emission factor (E),
 kg VOC/Mg product

Desired mass emission reduction,
 (% Red'n), %

Gas stream condition
Partial  pressure of styrene
 at inlet (Pin)

Composition of gas stream
 at inlet

Styrene mass fl owrate
 (Ws), lb/hrd

Gas stream volumetric
 flowrate (V), acfrn6

Gas stream mass flowrate
 (W), lb/hrf

Partial  pressure of styrene
 at outlet (PoutK mm M9
Shell and tube heat exchanger
 with hot fluid in the shell
 side and the cold fluid in the
 tube side

Identify the polymer industry and
 the vent from Chapters 3 and 6
From model plant in Chapter 6
From model plant in Chapter 6

From regulatory alternative in
 Chapter 6

Assumed saturated styrene in
 air at 80°F, 1 atm.
7.959 mm Hga
0.01047 ft3 styrene/ft3 gasb;
0.002767 Ib styrene/ft3
0.2564 x CAP x E
6.022 x W,
4.400 x V

100%-.(I Red'n)  x 7.959 „„„ „ 9
   TDD                     3
Temperature required for reduction
 fri   \  oi/h
 ('  out'5  *•

Temperature required for reduction
 (Tout). °P

Latent beat change of styrene (Qstv)
 Btu/hr1                          y
 4847.95 * [18.2440 - In (Pout)3


(1.8 x T'out) - 459.67
166.36 x Ws x (% Red'n)
                                     E-41

-------
                Table E-12.  PROCEDURE TO CALCULATE HEAT LOAD
                OF A CONDENSATION SYSTEM FOR STYRENE IN AIR (Concluded)
         Item
            Value
Average (bulk) gas temperature
                             i k
Density of air (Pair), lb/ft3

Specific heat of air((cD)al>),
 Btu/lb-°F

Sensible heat change of air (Qair)»
 Btu/hr
 Specific heat of styrene ((c0)   ),
  Btu/lb-°F   ,                 sty
 (80 + Tout)  * 2

 1 * [(0.002517 x Tb)  + 1.157]


 From API Report 441

 V x pal> x (cp) .   x  (80-Tout)  x
              K air
   60 min/hr


 From API Report 44"1
 Sensible heat change of styrene (Q'sty)
  Btu/hr
 Ws x (cp)     x (80-Tout)
          sty
Total design heat load (Qtot)» Btu/hr
(Qsty + Qair + Q'sty)
^Calculated from Clausius Clapeyron curve fit

   (In p = -4847.95 (   1  \ + 18.2440) of styrene vapor pressure versus
                         ~
 temperature data given on p. 3-59 of the Chemi cal  Engi neers '  Handbook
 (Reference 37) for 80°F (see temperature required for reduction).
DVolume fraction of styrene = 7.959 mm Hg = 0.01047 ft3 styrene/ft3 gas.
                                760 mm Hg

cAssuming ideal  gas:

    v  = RT =  1545 ft lbf/lbm - °R x §40 JR  , 394.13ft3/lb-mole;
    F    p~     14.7 lbf/in/ x 144 in^/ft^

    styrene content (Ib/ft3 gas) =

        0 .010 47 ft3 styrene x  Ib-mole    x 104.14 Ib styrene
             ft3 gas          394.13 ft3        Ib-mole     ~
                                 E-42

-------
Footnotes for Table E-12 (concluded)


dCAP.Gg product/yr x 1000 Mg/Gg x E kg VOC/Mg product
              86UO nr/yr x 0.4536 kg/lb

eCAP,Gg product/yr x 1000 Mg/Gg x E kg VOC/Mg product    = i 544 x CAP x E
 8t>uu nr/yr x u.453b kg/lb x u.uoz/67 ib/ft3 x bu min/hr

fV.acfm @ 80°F x «28.9 Ib/lb-mole x 60 mln/hr
         394.13 acf/lb-mole @ 80"F

9Using mass emission reduction as initial  approximation of volume
(partial pressure) reduction.  After outlet temperature is approximated,
the outlet gas molar volume and partial  pressure can be calculated and
iterated upon until an acceptable outlet temperature is calculated.

"Solving Clausius Clapeyron curve fit of styrene vapor pressure data
 (r2 = 0.99995) referred to in footnote a for temperature.

isiope, m, of Clausius Clapeyron curve fit = - \/R

 latent heat of styrene, \= -m x R =
                                                 Btu-lb-mole
      4847.95 (°K) x 1.9853 cal/g-mole-°K x 1.8  cal/g-mole
                        104.14 lb/lb-mole

JT,°K = T,°C + 273.15 = 5 (T,°F-32) + 273.15 = 0.5556 T,°F + 255.37   -
                        ?

kFor an ideal gas (pV = mRT/MW), P± =  mi/Vj_= T£ ;  pa1r = 0.08081b/ft3
                                 P2
 at 0°C (ChE Hndbk, p. 3-72) (Reference 38)
    Pair @ T,°K = 0.0808 x 273.15°K =    0.808 x 237.15	
                             TT'KU.5556 X T,"F + 255.37
1(cP}
'°'769
     air
°-231
                                         where
and
                                                                 are
 specific heats of nitrogen and air, respectively,  available by
 interpolation from API Report 44,  p. 652 (Reference 39).

m(cp)    vs T,°F, values are available for interpolation on  p.  682
     sty

 of API Report 44 (Reference 37).
                              E-43

-------
Society of Heating, Refrigerating, and Air-Conditioning Engineers
(ASHRAE).5l  The total required heat transfer area and refrigeration
capacity then were calculated from the total heat load, temperature
difference, and overall heat transfer coefficient, and commercially
available sizes were selected.  A tabular procedure for calculating
heat exchanger and refrigeration system size for a single section heat
exchanger and a refrigeration unit is presented in Table E-13.
E.5.2  Surface Condenser Cost Estimation Procedure
     Since the gas volumes of the two streams are low, the calculated
required (selected) heat transfer areas are also low  (about 1.7 (2.3)
and 4.5 (7.2) ft2, respectively, for 90 and 98 percent reduction of
styrene emissions from a single process line).  The purchase costs of the
heat exchanger43,44 anc[ refrigeration systems 45 were estimated from data
provided by vendors.  An installation factor of 1.39  (see Table 8-2) was
used to estimate installed condenser costs.
     The corresponding required refrigeration capacities for 90 and
98 percent styrene reduction were only 0.056 and 0.080 tons (0.20 and
0.28 kW or 670 and 960 Btu/hr), respectively.  For 90 percent reduction,
the required capacity is much smaller than the 0.117  tons (0.412 kW or
1405 Btu/hr) available capacity for the minimum available size
refrigeration unit of 1/4 compressor horsepower (0.186 kW compressor)
and a coolant temperature of 0°F (-17.8°C).  For 98 percent reduction,
however, 1/2 compressor horsepower (0.373 kW compressor) was required
to provide sufficient available refrigeration capacity for a coolant
temperature of -30°F (-34.4°C).
     Installed costs were put on a June 1980 basis using Chemical
Engineering Plant cost indexes.  No additional piping was costed since
the condenser unit is so small  (_< 5 in.) that it should be able to be
installed adjacent to the source.  Table E-14 presents the procedure
for estimating capital and annual operating costs for condensation
systems.
E.6  ETHYLENE GLYCOL RECOVERY SYSTEMS DESIGN AND COST ESTIMATION PROCEDURE
     This section outlines the  basis and procedures used to design and
estimate costs of the baseline  and regulatory alternative ethylene glycol
recovery systems.  The resulting costs for the two systems are presented.
                                  E-44

-------
                  Table E-13.   PROCEDURES  TO  CALCULATE  HEAT TRANSFER
                    AREA OF  A  CONDENSATION SYSTEM  OF STYRENE  IN AIR
Heat exchanger configuration

Source Identification


Coolant temperature,  TC,°F


Shell-side heat transfer
 Coefficient (h0),       Btu
                                                   Assume  appropriate size unit.3

                                                   Identify the polymer industry and
                                                   the vent from Chapters 3 and 6

                                                   Tout-10, rounded to nearest
                                                   multiple of 10°F.

                                                   Calculate using procedure in
                                                   Chemical Engineers' Handbook,
                                                   pp. 10-25 thru 10-28
                                                   (Reference 38)b
Coolant
                                                    Select  chilled water at Tc > 60°F;
                                                    50% ethylene glycol - 50% water
                                                    solutions  at Tc = 40,50°F;
                                                    Freon-12 at -40°F < Tc < 30°F;
                                                    and Freon-502, at Tr<-50°F.c
Tube-side Reynold's Number (NRe)'
                                                    (12 x  rH x  p  ) * JJL
Tube-side heat transfer
 Coefficient (),       Btu
                    hr-ft^-T
                                                    Calculate  using  appropriate equations
                                                    for  forced convection  in pipes.6
Coolant flow (Wc),  Ib/hr
Temperature change of coolant
Coolant flow (Vc),  gpm

Across-tube heat transfer
 coefficient (ht),       Btu
                                                    Calculate  for  assumed heat exchanger
                                                    and  cool ant.f

                                                    Qtot *  (Cp x Wc)


                                                    Calculate  for  assumed heat exchanger.9
                                                    Calculate  for assumed  tube
                                                    size."
                     hr-ft^-T
                                        E-45

-------
                  Table £-13.  PROCEDURES TO CALCULATE HEAT TRANSFER
                  AREA OF A CONDENSATION SYSTEM OF STYRENE IN AIR (Concluded)
Condensation heat transfer
 Coefficient (hc),    Btu
                   hr-ft2-"F
Calculate using procedure in
Applied Process Design for
Chemical and Petrochemical Plants,
vol. 3, p. so (Reference 42).'
 Overall tube-side heat
  transfer coefficient (hj)
C(l/hc)
 Mass transfer coefficient (Kg).
                   1 b- mole
                 hr-ft2-mra Hg
Calculate using procedure in
Applied Design for Chemical  and
Petrochemical Plants. Vol. 3,
pp. 100, 101, 104 (Reference 43).
  Clean overall heat transfer
   coefficient (Uc), Btu/ft^hr-T
Dirty overall heat transfer
 coefficient (Ud), Btu/f^-hr-TJ

Log mean temperature difference (LMTD),  °F


Required heat transfer area (A), ft2

Required heat transfer area with 10% safety
 margin (A1), ft*
Required refrigeration capacity (RC1),  tons1

Selected refrigeration capacity (RC),
 compressor horsepower
Horsepower per ton of refrigeration (Hp/ton),
 Hp/ton
Calculate from h-j, Kg, and hj
using procedure in Applied Process
Design for Chemical and Petro-
chemical plants, Vol. 3, pp.  100-106
(Reference 44), which iterates on the
condensate film temperature to balance
the tube-side and shell-side  heat
transfer within ±5% and then uses
the average to calculate the  clean
overall heat transfer coefficient.
[(1/UC) + 0.001 + 0.001]

     - AT2) - In
                        "1
Qtot
           x LMTD>k
A x 1.1
if A1 > Aj (available area for
assumed size), try a larger size
heat exchanger (see footnote a)

Qtot r 12,000

Select from vendor information
(such as Reference 47)  for design
heat load and coolant temperature

Based on vendor information for
coolant temperature (see  Reference 36)
                                          E-46

-------
  Footnotes for Table E-13
  w»
   If no information is available on approximate condenser size required,
   assume an overall heat transfer coefficient between 3 and 15 Btu/ft2-hr-°F,
   calculate required heat transfer area as noted near end of this table, and
   select an appropriate size condenser.  Condenser and tube characteristics
   for large units  (> 34 ft2) can be found in pp. 11-1 thru 11-18 of the
   Chemical Engineers' Handbook (Reference 48).  Characteristics of smaller
   units can be obtained from vendor information.45,46  Characteristics of
   units developed  for this analysis can be found in the docket.36

       Characteristics needed for design calculations:

       Tube:  outer diameter, D0 (in.);   inner diameter, D-j  (in.);
              thickness, Xw (in.);  cross sectional area, Ax  (ft'/tube);
              specific external surface area, Ae (ft2/ft of tube);
              tube side hydraulic radius, r^ (ft) = D-j/(4 x 12 in./ft).

       Condenser:  shell  diameter, Ds (in.);
                   tube count, NT (no. of tubes);
                   tube pitch, p (in.);
                   length, L (ft);
                   effective length, Leff (ft) = L - (Nts x Lred/12 in/ft)
                      where Nts = no. of tube sheets (can assume 2)
                            Lrec| (in.) = reduction in effective length per
                                         tube sheet (can assume = 0.25 in.,
                                         for Ds < 8 in.; = 1.5 in.,
                                         for Ds >. 8 in.);
                   total  tube area,  AT (ft')  = Ae x Leff x NT.

b
 Assumed baffle cut,  lc = 0.25 Ds, Ds < 10  in.; = 0.33 Ds, 10  < Ds < 18 in.;
 and = 0.45 Ds, Ds 2 18 in.; radial  clearance between shell and outer
 tube limit = 1/4 in.,  Ds < 8 in.; = 7/16 in., for 8 < Ds < 25 in.;
 = 1/2 in., for Ds 2 25 in.; baffle  spacing,  ls = 0.6 Ds; diametrical  shell-
 to-baffle clearance  =  0.05 in.,  for Ds < 4 in.;  = 0.075 in.,  for 4 < Ds
 < 7.5 in.; = 0.10 in.,  for 7.5 < Ds < 14 in.; =  0.125 in., for 14 $ Ds
 < 18 in.;  = 0.150 in.,  for 18 < Ds  < 25 in.; = 0.30 in., for  25  < Ds
 < 42 in.;  = 0.35 in.  for Ds 2 42 in.; number of  sealing strips ^ 0.5 x Nc

GCoolant characteristics  can be interpolated  or extrapolated for  the
 coolant temperature, Tc, from the Handbook of Chemistry and Physics,
 p. F-36 (Reference 50)  and from The Chemical  Engineers'  Handbook:
 pp. 3-71,  126, & 214 (Reference  WF Tor water; and pp.  12-46  thru  12-48
 (Reference 49) for ethylene'glycol-water solutions; and from  Thermophysical
 Properties of Refrigerants pp. 9 thru 11 for Freon-12 and pp.  105  and 106
 for Freon-502 (Reference 51).   Characteristics required are dynamic viscosity
 ( H-C),  Ib/ft-hr; density (  PC),  lb/ftj;  specific  heat  C(cD)c] Btu/lb-°F;
 thermal  conductivity  (kc),  Btu/hr-°F-ft;   and specific  gravity ( Ych
 dimensionless =  Pc/62.42 lb/ftj.
                               E-47

-------
 Footnotes  for Table  E-13  (Continued)

 dFor coolant  velocity, V  = 3 fps  (3-10 fps recommended by Kern in Process
  Heat Transfer (Reference 52)).

 eFrom The  Chemical Engineers' Handbook, =pp. 10-12 thru 10-15 (Reference 41)


        ,   /    \ 0-14
  Assumed   / P-b \       =1 since  little change in coolant temperature.

           \^w/
    (1)  For turbulent flow (NRe >  10,000) (from Eq. 10-51):


        h   =    0.023 x V, ft/hr x  P lb/ft3 x cp, Btu-1b-°F  x  (ib °-14

                             (NRe)0'2
       if  0.7 < NPr < 700 & L/D > 60;
    (2) For transition flow (2000 < NRe < 10,000) (from Eq.  10-49):
       h
      0.029 k
-125)Npr1/3 [l  WDi)2/3]/!^0-
           L    vu/     J\RW/
                                                                14
   (3) For laminar flow (NRe < 2100) (from Eq.  10-40):


       h.  = 0.465 k   Np.1/3  /fibX0-14 x 0.87  (1  + 0.015
               rH
where
                   (NRe x NPr x 4 x rH)  f L.
     coolant velocity of 3 fps,
 Wc, Ib/hr = Ax x NT x 10,800 ft/hr x Pc

9For coolant velocity of 3 fps.
 Vc, gpm = Ax x NT x 180 ft/mi n  x 7.48 gal /ft3.
    where:   K^ = thermal  conductivity  of  tube = 64.2 Btu/ft-hr-°F for
                 Admiralty brass  given on pp. 23-49 of The Chemical
                 Engineers'  Handbook  (see Reference 53)1

            DL = (Do - °i) * In  (

""Must assume temperature  of condensate film on tubes, which is the
 variable that is iterated upon to  balance overall shell -side and
 overall  tube-side heat transfer.

                               E-48

-------
Footnotes for Table E-13 (Concluded)

JAssuming both shell-side and tube-side fouling coefficients  equal
 to 0.001 as given for cooling tower  water or refrigerants  and  for
 industrially clean gases, clean hydrocarbon vapors,  or atmospheric
 air in Applied Process Design for Chemical  and Petrochemical Plants,
 Vol. 3, Tables 10-10 and 10-11  (Reference 54).                 :	

kMore complicated relationships, considering the average and  end point
 values for individual  sections  and the entire exchanger, were  used  for
 the multiple section analysis {see References 36 and 44).
                               E-49

-------
          Table E-14.  CAPITAL AND ANNUAL OPERATING COST ESTIMATION
                        PROCEDURE FOR CONDENSERS WITH REFRIGERATION
           Item
       Value
Capital Costs
    Purchase cost, July 1984
    Purchase cost, June 1980b
    Total Installed Cost, June 1980C
Annual 1 zed Costsd
    Operating labor
    Maintenance materials & labor
    Utilities
      Electricity, pumping'
      Electricity, refrigeration
      Coolant, make-up
    Capital recovery1
    Taxes, administration
      & insurance
    Total annualized cost
      without recovery credit

    Styrene recovery credit-3
    Net Annualized Cost
       after recovery credit
[(16.816 x AT) + 125.151]
if Freon coolant, x >1.3a
Purchase cost, Sept. 1982 x 0.871
Total Purchase Cost x 1.39

See footnote e
0.05 x total installed cost

6.104 x Yc x Vc
See footnote g
See footnote h
0.1627 x total installed cost
0.04 x total installed cost
Operating labor & maintenance
  + utilities + capital recovery
  + taxes, administration &
    insurance
2767 x Ws x (% Red'n. * 100)
Total annualized cost - styrene
  recovery credit
                                     E-50

-------
Footnotes for Table E-14

aBased on vendor information (References 45 and 46).

bDeflated using Chemi cal Engi neeri ng Fabricated Equipment Cost Index from
 July 1984 (estimated) (334.6) to June 1980 (291.3).

cBreakdown of installed cost factor given in Table 8-2.

dCost factors presented in Table 8-3.
     20 to 34 ft2 units:  Operating labor cost = 1 hr/wk x 52 wk/yr
 x 1.15 (with supervision/without supervision) x $18/hr (including
 overtime).  For < 20 ft2 units:  Operating labor cost = 2 hr/mon
 x 12 mon/yr x 1.15 (with supervision/without supervision) x $18/lir
 (including overtime).

fUsing Equation 6-2, p. 6-3 in The Chemical  Engineers' Handbook
 (Reference 55) for V = 3 fps , assuming a pumping height of
 of 50 ft. and a pump efficiency of 65%:
50 ft x Vc x Vc x 0.7457 kW
     gpm Tt/hp       h"p
                                           8600 hr/yr
                                      0.65 pump efficiency

                                              x $0.049/kwh
          where Yc = specific gravity of coolant = P~  * 62.42 lb/ft3
                                                  (the density of water)

9        RC'  x (hp/ton of refrigeration for Tr)	
     u.85 compressor efficiency x 0.85 motor efficiency

        x 0.7457 KW  x 8600 hr/yr x $0.049/kwh
             np

"For coolant velocity of 3 fps:

 For ethylene glycol-water solutions, Freon-12,  and Freon-502, assume  one
 replacement per year of coolant in condenser and refrigeration system
 and coolant volume in condenser and refrigeration unit twice that of
 condenser alone.

       coolant volume, gal  = Ax x NT x L x  (2 vol.  in
                             cond & refrig/vol.  in  cond)
                           = 2 x Ax x NT x  L.

 For ethylene glycol-water solutions:

       cost of coolant = j[Xw ($0.30/1000 gal x  7.48 gal/ft3)]
                         + [XEG ($0.27/lb x PEG  Ib/ft3)]}  x coolant
                         volume,ft3 x 1 replacement per year
                       = ($0.00224 Xw + $0.27 X  EG pEG)
                         x annual  coolant replacement,  ft3/yr.
                                E-51

-------
Footnotes for Table E-14 (Concluded)
             where,
                  Xyj  = volume fraction of water in  solution
                      = volume fraction of ethylene  glycol  in  solution
       For Freon-12:
              Cost of coolant = (coolant volume,ft3 x 1 replacement/yr
                  x $1.31/lb) *  PC, lb/ftj

       For Freon-502:

              Cost of coolant = (coolant volume,ft3 x 1 replacement/year
                  x $3.11/lb) *  PC, lb/ft3

       For chilled water, assume 0.1% make-up (99.9% recycle):

              Cost of coolant = $0.30/1000 gal x Vc,gpm x 60 min/hr
                                x 8600 hr/yr x 0.001 make-up/total
                              = 0.1548 x Vc


       ilO percent interest  (before taxes) and 10 yr. life.

       JVL.lb styrene emitted/hr x 8600 hr/yr x (% Red'n in condenser * 100)
              x 0.90,fraction of reduction recovered x $0.3575/lb styrene.
                                      E-52

-------
E.6.1  Ethylene Glycol Recovery System Design
     The  baseline  system  for  a plant  producing a  low  viscosity product
or  high viscosity  product with a  single  end finisher  was represented
by  a system that recovers ethylene  glycol  (EG) emitted from the
polymerization reactors through use of EG  spray condensers and from
the esterifiers through use of reflux condensers.  The baseline
system for a plant producing  a high viscosity product was represented
by  a system that recovers ethylene  glycol  emitted from the polymerizers
through the use of EG spray condensers on  the initial end finishers
and a distillation column on  the  cooling water tower  and from the
estrifiers through the use of reflux  condensers.
     The  regulatory alternative control  systems utilized distillation
columns.  For those PET plants producing a low viscosity product or a
high viscosity product with a single end finisher, further control of
EG emissions was obtained by  the  installation of a distillation column
that reduces the EG concentration in the cooling water tower.  For
those PET plants producing a  high viscosity product with multiple end
finishers, further control of EG emissions was obtained by increasing
the flow  rate of cooling water to the existing distillation column,
which results in a lower  EG concentration  in the cooling tower.
     The  equipment selected to comprise  the ethylene  glycol recovery
systems,  as well as the design and operating parameters, was based on
information provided by industry sources.  The industry information
(much of which was considered confidential) was used  in conjunction
with standard engineering references such as the Chemical Engineers'
Handbook17, and McCabe and Smith's Unit Operations of Chemical
Engineering56,  and engineering judgment.57 These design procedures are
summarized in the footnotes to Tables E-15 and E-16 for the baseline
systems.
E.6.2  Ethylene Glycol Recovery System Cost Estimation Procedure
     The cost estimates and their bases are presented in Table E-15
for the baseline ethylene glycol  recovery system for PET plants producing
a low viscosity product or a high viscosity product with a  single end
finisher and in Table E-16 for PET plants producing a high  viscosity
product with multiple end finishers.  The costs  of the baseline systems
                                  E-53

-------
    Table E-15.  EG RECOVERY COSTS FOR BASELINE SYSTEM FOR PLANTS PRODUCING
  A LOW VISCOSITY PRODUCT OR A HIGH VISCOSITY PRODUCT WITH A SINGLE END FINISHER
                               (June 1980 dollars)
        Item (Number of Item)
    Model
    Plant
 Process
  Line
Capital Costs
    Spray Condensers (28 @ $17,384
                            each)
    Reflux Condensers (14)
    Pumps (28)
    Heat Exchangers (42)
    EG Recovery System (1)
    Refrigeration System (14)
    Installed Capital  Cost Factor
    Total Installed Capital Cost
  $486,750*

  $243,400b
   $23,30QC
   $64,300d
  $386,3006
  $128,700f
         4.249
$5,234,000
$1,628,400^
Annual i zed Costs
Operating Labor
Operating Materials
Maintenance Materials and Labor
Electricity
Steam
Water
Taxes, Insurance
and Administration
Capital Recovery
Recovery Credit
Total (Annual Costs -
Recovery Credit)
$173, SOO1
0
$21,70Qk
$86,6501
$917,930"!
$4,650n
$209,400°
$853,100P
($1,218,300)3
$1,048,700
$22, 100 J
0
$3,100J
$12,4000
$13 1,100 J
$66 4J
$65,100
$265,400
( $174,000 )J
$325,900
                                     E-54

-------
Footnotes for Table E-15

aBased on size estimate from Tennessee Eastman and cost estimate from
 Missouri Boiler.  Two spray condensers plus two spares per line in
 model plant.

bGoing to this system replaces feed lines from estifiers to distillation
 column (CL-2) with reflux condensers, one per line plus one spare per line
 The size and cost were assumed to be the same as for EG spray condensers
 on the reactors.

                                            A 1 hp pump is  required per
      Reference  57,  p.  14,  for  pump  sizing.
  line at  a  cost of  $831  per pump.
 dSee  Reference  57, pp.  7-9, for exchanger sizing.  Only  sizing done
  for  reactors in which  industrial  resins are produced.   Assume size and
  costs for heat exchangers associated with reactors producing textile
  resins would be the same.  Seven  scraped surface exchangers (plus 7
  spares) for end finishers at $1,130 each; 7 scraped surface exchangers
  for  prefinishers (plus 7 spares)  at $1,930 each; 7 exchangers (plus
  7 spares) for  esterifiers at $1,530 each (average of $1,130 plus $1,930).

 eCosts obtained from a industry source were considered confidential.  A
  correction factor of 0.764 was obtained from this source to scale the
  costs down to  our model plant capacity.  Using the total equipment and
  steam jet ejector system cost yielded the EGRS cost given.
     Reference 57.  Seven systems for prefinisher at $9,283 each, and
 7 for finishers at $9,100 each.

9See Table 8-2, factors based on sum of piping, insulation, painting,
 instruments, and electrical factors equaling 1.12 A for larger capacity
 system given in confidential industry information.  Does not apply to
 refrigeration systems.

"Model  plant is comprised of seven process lines.  Estimate of total
 installed capital cost of equipment for just one process line was
 obtained with the following equation:
                ,0.6
                    x $5,234,000 = $1,628,400
1 Based on 9,640 man-hours per year at $18 per year.  The number of
 man-hours has been scaled down from the number of man-hours provided by
 a confidential industry source.

JEstimate of cost for a single process line was obtained by the following
 equati on:

             (I) * ACi
      where:
              3-j = operating labor, operating materials, maintenance
                   materials and labor, electricity, steam, and water.
                               E-55

-------
Footnotes for Table E-15 (Concluded)

^Based on maintenance labor requirements and maintenance materials cost
 given in confidential industry information for a larger system.
      cost comes from 4 sources: pumps, recovery system, heat exchanger
 to chill water, and for the reflux condensers.   The 14 pumps at 1  hp use
 91,500 kWh per year.  Recovery system was estimated to use 412,800 kWh
 per year based on confidential industry information for a larger system.
 Chilled water is necessary for chilling the spent EG used in the EG
 spray condensers and is necessary for reactors  producing high tenacity
 (high viscosity) industrial resins.  The electricity requirement to
 maintain the chilled water is 660,100 kWh per year for a plant producing
 a high viscosity product.  Electricity usage by the exchangers for the
 reflux condensers was assumed to be the same as that required by the
 spray condensers {i.e., 603,900 kWh per year).   Assuming all the lines
 in a model plant produce high tenacity (high viscosity) industrial
 resins, total electricity usage in the model plant- is 91,500 plus
 412,800 plus 660,100 plus 603,900, which is equal to 1,768,300 kWh per year.
 Cost of electricity is $0.049/kWh.

mBased on confidential industry information and  scaling steam usage in
 the EG recovery system and its vacuum system proportionately according
 to plant capacity, steam usage was estimated to be 1.34 x 10° Ibs/year.
 An estimated additional 14.532 x 106 Ibs of steam per year would be
 required over the baseline system by the vacuum system servicing the
 polymerization reactors.  A cost of $6.18/1,000 Ib of steam was used.

"Based on water consumption of 15.5 x 106 gallons per year (from confidential
. industry information scaled down by proportioning relative capacities).

°Based on 0.04 x Installed Capital Cost.

PBased on a capital recovery factor of 0.163.

QBased on a total EG recovery of 19.5 kg of EG/Mg of product and a
 recovery credit of $0.595/kg  ($0.27/lbs of EG from Chemical  Marketing
 Reporter).  The 19.5 kg of EG/Mg of product is  an increase in the
 recovery of ethylene glycol from the polymerization reactors plus
 ethyl ene glycol recycled from the estifiers that would otherwise have
 to be replaced with fresh feed if the baseline  system was used.
                              E-56

-------
      TABLE E-16.  EG RECOVERY COSTS FOR BASELINE SYSTEM FOR PET
PLANTS PRODUCING A HIGH VISCOSITY PRODUCT WITH MULTIPLE END FINISHERS
                         (June 1980 dollars)
Item (Number of Item)
Capital Costs
Spray Condensers (4)
Pumps (4)
Heat Exchangers (4)
Reflux Condensers (4)
Distillation Column, cooling tower
Refrigeration system (2)
Installed Capital Cost Factor
Total Installed Capital Cost
Annuali zed Costs
Operating Labor
Operating Materials
Maintenance Materials and Labor
Electricity
Steam
Water
Taxes, Insurance
and Administration
Capital Recovery
Recovery Credit
Total (Annual Costs -
Model
Plant
$82,640*
3,945&
8,220?
82,640d
661,8006
22,065^
3.249
1,258,800
29,500"
-
3,940i
21, 640 J
5,170k
-
50, 350 1
204,800"!
(233,490)"
81,910
Process
Line
$41,320
1,970
4,110
41,320
568,000
11,030
3.24
866,480
14,750
-
1,970
10,820
2,635
-
34,660
140,980
(116,745)
89,070
                                     E-57

-------
Footnotes for Table E-16
     number of spray condensers is based on one per line with one spare
 per line and two lines per plant.  The cost per spray condenser was
 estimated by multiplying the spray condenser cost in Table E-15 by the
 relative process line sizes (20 Gg vs. 15 Gg) raised to the 0.6 power
 as follows:
           $17,384 x /20\°-6  = $20,660
                     \T57

bTwo pumps plus two spares per process line,
 from Table E-15 as follows:
                                              Cost per pump adjusted
            $830 x/20\°-6  =
                  YI57
                              $986
cTwo heat exchangers plus two spares per process line.
 exchanger adjusted from Table E-15 as follows:
                                                        Cost per heat
    [2 scraped surface exchanges for prefinishers at $1,930 each plus 2
     exchangers at $1,530 each for esterifiers] x/20\0-°  =  $8,220
                                                 \T5V

dOne reflux condenser plus one spare per process line.  Cost per reflux
 condenser assumed the same as the spray condenser in footnote a.

^Based on adjusting size and cost information claimed confidential  from
 an industry source.  In general, the size was adjusted based upon
 proportioning the flow to the distillation column directly on plant
 capacity and then sizing the distillation column on the basis of the
 new flow divided by the flow to the original distillation column and
 the result raised to the 0.6 power.  The cost was adjusted to June 1980
 dollars using a cost index factor of 0.8208 based upon the Chemical
 Engineering Plant Cost Index of 320.3 for January 1984 and 262.9 for
 June 1980.

      per refrigeration system adjusted from Table E-15 as follows:

                    0.6
     $9,283 x
                        = $11,030
9See Table 8-2.  Applies to spray condensers, pumps, heat exchangers,
 and reflux condensers.  Does not apply to distillation column and
 refrigeration system.

hconfidential industry information on operating labor scaled down to
 assume  1,640 hours per year at $18 per hour.

''Based on maintenance labor requirements and maintenance materials cost
 given in confidential industry information for a larger system.
 Assumes $1,330 for materials and 145 hours for maintenance labor.

JElectricity usage per source was assumed to be proportional to plant
 capacity.  The 4 pumps at 1.33 hp each would use 17,385 kWh/year.  The
 electricity requirement to maintain the chilled water (refrigeration

                                  E-58

-------
Footnotes for Table E-16 (Concluded)

 system) is 212,070 kWh/year.   Electricity used by the exchangers for
 the reflux condensers was assumed to be the same as that required by
 the spray condensers (i.e., 212,070 kWh/year).  Total electrical usage
 in the model plant for these  controls is 17,385 plus 212,070 plus
 212,070, which is equal to 441,525 kWh/year.  The cost of electricity
 is $0.049/kWh.

k Includes only steam used in distillation column.  Distillation steam
 total  equals 836,625 Ib/yr (97,575 Ib/yr to heat the water; 5,853 Ib/yr
 to heat the ethylene glycol;  and 733,197 Ib/yr to vaporize the
 water).  Steam cost of $6.18  per 1000 Ib of steam.

1 Total  installed capital cost  x 0.04.

•"Total  installed capital cost  x 0.1627.

"Based on total EG recovery of 17.65 Ibs of EG/1,000 Ibs of product
 and a recovery credit of $0.15 per Ib. Of the  total recovered EG,
 about 16.68 Ibs EG/1,000 Ibs  comes from the initial end finishers and
 0.97 Ib EG/1,000 Ibs product  from the distillation column.
                               E-59

-------
 were estimated based on the design estimate developed and standard
 engineering procedures.
      The costs of the regulatory alternative distillation columns were
 derived from confidential  cost data provided by an industry source for
 a similar system on a larger capacity plant.  In general, the distil-
 lation columns were sized and costed on the basis of the flow rate
 from the cooling tower to the column using the ratio of the target flow
 rate divided by the base flow rate and the result raised to the 0.6 power.
 Utility costs, which are primarily steam costs, were calculated based
 on the flow rate from the  cooling tower to the distillation column.
 E.7  PIPING AND DUCTING DESIGN AND COST ESTIMATION PROCEDURE
      Control costs for flare and incinerator systems included costs of
 piping or ducting to convey the waste gases (vent streams)  from the
 source to a pipeline via a source leg and through a pipeline to the
 control device.  All vent  streams were assumed to have  sufficient
 pressure to reach the control  device.   (A fan  is included on the duct,
 fan,  and stack system of the incinerators.)
 E.7.1  Piping and Ducting  Design Procedure
      The pipe or duct diameter for each waste  gas stream (individual
 or  combined) was determined by the procedure given in Table E-17.   For
 flows less  than 700  scfm,  an economic  pipe diameter was  calculated
 based on an equation in  the Chemical Engineer's  Handbook59  and  simplified
 as  suggested by Chontos.60»6l>62  The  next larger size  (inner diameter)
 of  schedule 40 pipe  was  selected unless the calculated size  was  within
 10  percent  of the difference between the  next  smaller and next  larger
 standard size.   For  flows  of 700 scfrn  and  greater,  duct  sizes were
 calculated  assuming  a velocity  of 2,000  fpm for  flows of 60,000  acfm
 or  less and  5,000 fpm for  flows  greater  than 60,000 acfm.  Duct  sizes
 that  were multiples  of 3-inches were used.
 E.7.2   Piping and Ducting  Cost Estimation  Procedure
      Piping  costs were based on  those  given  in the Richardson Engineering
 Services Rapid Construction Estimating Cost System30 as combined for
 70 ft.  source legs and 500 ft. and 2,000 ft. pipelines for the cost
analysis of  the Distillation NSPS.63 (See Tables E-18 and E-19).
Ducting  costs were calculated based on the installed cost equations
given in the GARD Manual.64  (See Table E-20).
                                  E-60

-------
                 Table E-17.   PIPING AND DUCTING DESIGN PROCEDURE
           Item
                       Value
(1)  Pipe diameter, D

     (a) Piping^
For Source Legs:
D (in.) = 0.042 x Q (scfm)  + 0.472,  for Q < 40  scfm
D (in.) = 0.009 x Q (scfm)  + 2.85,  for 40 12 in. or Q>700scfir
and Q<60,000 acfm  ^______
D (in.) = (0.1915)  VQ(acfm), for Q > 60,000 acfm
Select size that is a multiple of 3 inches.
Assumed 70-ft. source leg from each source to the
pipeline.  Assumed separate pipelines for large
{ > 35,000 scfm) intermittent streams and for all
continuous streams together.  Selected pipeline  ;   ,
length of 70, 500 or 2,000 ft. if calculated safe
pipeline length within 10 percent of standard length;
if not, selected calculated length between standard
values.

Assumed 70-ft. source legs from each source to the
pipeline.  Used duct, fan, and stack cost from
Enviroscience (Reference 28) which assumes a 150-ft.
duct cost based on the GARD Manual  (Reference 64).
aEconomic pipe diameter equations from Reference  62  (which  is based upon References 5S
 and 60 ) .

bFrom continuity equation  Q _  IT  D2y ;  assumed  velocity,  V, of  2,000 fpm for lower
                               ~~~~
 flows and 5,000 fpm for higher flows.
                                           E-61

-------
                        Table E-18.  PIPING COMPONENTS3
Equi pment
Type
Check Valves
Gate Valves
Control Valves
Strainers
El bows
Tees
Flanges
Drip Leg Valves
Expansion Fittings
Bolt and Gasket Sets
Hangers
Field Welds
Pipe Length,
(Schedule 40) (ft)
Number of Equipment Type in Pipe Leg Type
Source
1
4
1
1
8
6
15
1
2
. 15
9
18
70 .
Compressor
1
2
-
1
6
2
10
1
1
12
4
12
20
Pipeline (500 ft)
1
3
1
1
6
2
20
1
5
21
50
40
500
Pipeline (2,000 ft)
1
3
1
1
6
3
35
1
20
38
200
120
2,000
From Reference 62.
                                               E-62

-------
(O
oo

oo
o
o
a.
»«t
Q.

a
CO
 t

UJ
CU
 ro
              CU
              Q.
 C5
              CU
              0.
 s-
 o
             CO
             r-
•P
cpotf)otf)otr>QooLngooooQtr>




                              »-li-«.-)»-)»Hi-ICMeN4COOO'5l-LOr^.r-4COOVOClJ'-''*
                    CU
                    cu
                    CL

                                                                                                       O
                                                                                                       o
                                                                                                       O)
                                                                                          o
                                                                                          CL
                                                                                                       O
                                                                                                       O
                                                                                                      •o
                                                                                                       cu
                                                                                         CO
                                                                                   CU


                                                                                   cu
                                                                                  M-
                                                                                   

                                                                                         z  cu

                                                                                         o  cu
                                                                                         oo ac



                                                                                          l|

                                                                                         Q_ (4«
                                                                                                    re
                                                           E-63

-------
(O
 oo
 cc
 cr>
 LU
 ca
 LU

 CJ
 00

 o
 §
 cr
 LU



 o
 o

 CD



 CJ
 o
 LU
 c/)
 O
 CM
  I
 O)
 r—


 fO
                

q^
c^
co
i

CM

O
CM


4-

a
«^
en
VO
CM
4»

^
rn
vo
cn



CM
a

o
CO
^.

4-
Q

-VO

VO

in
vo
vo
CO




CM
^**
•""*

O

VI
a
VI
o


CM
a

in
•— <

CO
4-
a
CM
in

T
4-
O
o
CO
T

CM


trt

•
cn

+

o

CD
cn

^

CM

iri
o

i



'a
o
o

CO

+
a
P^

ro
CM

4»
in
g
"*


CM
a

o
in
Q.

4-
a

^f

VO
vo
"'
VO
VO
en




co
**5"


CM
CM
«~4
VI
a
VI
CO
CM

CM
a

co
to

~-4
4-
a
CO
f—1

ro
4-
1^

t*~t
*7

Co

co
vo

,a^

4*

a

g
o


+

in

f«^
VO





=>
S
VO

^^

4-
a
in

S
CM

4^
in
o
"•


CM
a

0

CM

4-
a

co

vo
en

o
CM



*r
^^
^^


o
cn
VI
a
VI
0
cn
CM
a

in
CO


•-*
4-
a
l
o
cn
CM
4-
VO
CO

«v
CM
Q
in

^^

i

4*

a

"*;
VO

CO


CO

^^
o
CM




=5
VO


•Ht

*
a
in

CO


4»
K
o
~*

CM
a

o

o
CM

4-
Q

CM

CM

m
i
CM


VO

*^>



CO
VI
a
VI
CM
(—4
CM
a

CM
CO
CM

—4
4-
O
en
o

CM
4-
O

p».
1

CM


CM
m
CM

i-H



a
en


U3

^B

O
CM





=
^


^^

+
a
P^

vo
^^

4-
CM
'-"


fJ
a

VO

_,

4-
a

o

in

CM
CO
cn
CM



3D
— i
•^ 1











' IO
3

10

a
<

0)

4-1
C
•
c:
CD

Ut
O CO

4") |
•*• UJ
V)


S. JS
•W 10


e e
It* 1-

 c;

cu o
4-> x:
V)

(A V)

O
.£> 0)
r- CL
* 4J"
4J cn
u ei
3 ^«
4-> O.

O O.

ie j=
I- U
4J 10
Ul OJ
"~" c

c: ui


O Cil
^ c
§§.
u =
S
— cn
o. c

r— O.
0) <^

4J
O

o
U S)
lO C


^ —

4- 0
O 10
(O (J
=3 C

aj
= 
u
3
V)
t/1


•a
c:
03
a.
a
•o

c
5
X
CJ
^_
o
4**
^
^
                                              to

                                              o

                                              o
                                                                ai
                                                                ^
                                                F
                                                                           • o
                                                                         CO _
                                                                         AlO
                                                        o «S
                                                          o '
                                                        in o
                                                        Q-O O
                                                          O O
                                                     vn    O-


                                                     §uof
                                                     a> e u- a
                                                                       a;

                                                                       "5
                                                                       
-------
     Costs of source legs were taken or calculated directly from the
tables.  Costs of pipelines for flares were interpolated for the safe
pipeline lengths differing by more than 10 percent from the standard
lengths of 70, 500, and 2,000 ft.
                                  E-65

-------
E.8  REFERENCES

 1.  Kalcevic, V.  Control  Device Evaluation:   Flares  and  the  Use  of
     Emissions as Fuels.  In:   Organic Chemical  Manufacturing  Volume  4:
     Combustion Control Devices.   U.S. Environmental Protection  Agency.
     Research Triangle Park, N.C.  Publication No.  EPA-450/3-80-026.
     December 1980.  Docket Reference Number II-A-18.*
                                     *
 2.  Reference 1, p. IV-4.

 3.  Memo from Sarausa, A.I.,  Energy and Environmental Analysis,  Inc.
     (EEA), to Polymers and Resins File.  May 12,  1982.  Flare costing
     program (FLACOS).  Docket Reference Number II-B-39.*

 4.  Telecon.  Siebert, Paul,  Pacific Environmental Services,  Inc.  (PES)
     with Straitz, John III, National AirOil Burner Company,  Inc.
     (NAO).  November 4, 1982.  Availability of Flare  Cost Data  and
     Flaring of High Air-Content Docket Reference  Number II-E-59.*

 5.  Straitz, J.F. III.  Make the Flare Protect the Environment.
     Hydrocarbon Processing.  56.  October 1977.  Docket Reference
     Number II-I-32.*
 6.
 7.
 8.
 10.
 11.
12.
Oenbring, P.R. and T.R. Sifferman.  Flare Design...  Are Current
Methods Too Conservative?  Hydrocarbon Processing.   59:124-129.
May 1980.  Docket Reference Number II-I-58.*

Telecon.  Siebert, Paul, PES, with Keller, Mike,  John Zink Co.
August 13, 1982.  Clarification of comments on draft polymers  and
resins CTG document.  Docket Reference Number II-E-18.*
Telecon.  Siebert, Paul, PES with Fowler, Ed,
1982.  Flare Design and Operating Parameters.
Number II-E-60.*
NAO.  November 12,
 Docket Reference
Telecon.  Siebert, Paul, PES with Fowler, Ed, NAO.   November 5,
1982.  Purchase costs and Design and Operating Criteria for Steam-
assisted, Elevated Flares.  Docket Reference Number II-E-58.*

Telecon.  Siebert, Paul, PES with Fowler, Ed, NAO.   November 15,
1982.  Additional Flare Cost Estimates and Flare Design Criteria  and
Procedures.  Docket Reference Number II-E-61.*

Telecon. Siebert, Paul, PES, with Knock, Cor, NAO.   May 2,  1983.
Steam requirements for intermittent flares.  Docket Reference
Number II-E- 68.*

Telecon.  Siebert, Paul, PES, with Keller, Mike, John Zink, Co.
May 12, 1983.  Steam- and air-assisted intermittent flare guidelines.
Docket Reference Number II-E-69.*
                                  E-66

-------
 13.   Memo  from Senyk, David, EEA, to EB/S Files.  September 17, 1981.
      Piping and  compressor  cost and annualized cost parameters used in
      the determination of compliance costs for the EB/S industry.
      Docket Reference Number II-B-33.*

 14.   Memo  from Mascone, D.C., EPA, to Farmer, J.R., EPA.  June 11,
      1980.  Thermal incinerator performance for NSPS.  Docket Reference
      Number II-B-4.*

 15.   Air Oxidation Processes in Synthetic Organic Chemical Manufacturing
      Industry - Background  Information for Proposed Standards.  U.S.
      Environmental Protection Agency, Research Triangle Park,  N.C.
      Draft EIS.  August 1981.  p. 8-4.   Docket Reference Number II-A-26.*

 16.   Blackburn, J.W.  Control Device Evaluation:   Thermal  Oxidation.
      In:  Chemical Manufacturing Volume 4:  Combustion Control Devices.
      U.S.  Environmental  Protection Agency,  Research Triangle  Park,  N.C.
      EPA-450/3-80-026.  December 1980.   p. 1-1.   Docket Reference Number
      II-A-18.*

 17.   Perry, R.H. and C.H. Chilton, eds.  Chemical  Engineers'  Handbook,
      fifth edition.  New York,  McGraw-Hill Book  Company.  1973.
      p. 8-9.  Docket Reference Number II-I-16.*

 18.   Steam:  Its Generation and Use.   New York,  Babcock &  Wilcox Company,
      1975.   p. 6-10.  Docket Reference  Number II-I-20.*

 19.  Memo from P. Siebert,  PES  to Polymers and Resins File.  March 16,
      1983.   Distillation  NSPS Thermal  Incinerator  Costing  Computer
      Program (DSINCIN).   May 1981.  p.  2.   Docket  Reference  Number
      II-B-67.*

20.  Reference 15, p.  8-13.

21.  Reference 16, pp.  V-3,  V-15.

22.  Reference 16, p.  III-8.

23.  Reference 16, Fig. A-l, p.  A-3.

24.  Reference 15, p.  8-9.

25.  Reference 19, p.  4.

26.  Reference 16, p.  1-2.

27.  Reference 15, p.  G-3 and 6-4.

28.  Reference 16, Fig. V-15,  curve  3,  p.  V-18.

29.  Neveril,  R.B.  Capital  and  Operating  Costs of Selected Air
     Pollution Control  Systems.   U.S. Environmental  Protection Agency,
     Research  Triangle  Park, N.C.   Publication No. EPA-450/5-80-002.
     December  1978.  Docket  Reference Number  II-A-7.*

                                   E-67

-------
30.
31.
32.
33.
34.
Richardson Engineering Services.
Estimating Standards, 1980-1981.
II-I-52.*
                                       Process Plant  Construction Cost
                                       1980.   Docket  Reference  Number
35.




36.




37.

38.

39.




40.

41.

42.



43.

44.
Telecon.  Katari, Vishnu,  Pacific Environmental  Services,  Inc.
with Tucker, Larry, Met-Pro Systems Division.   October  19,  1982.
Catalytic incinerator system cost estimates.   Docket Reference
Number II-E-41.*

Telecon.  Katari, Vishnu,  Pacific Environmental  Services,  Inc.,
with Kroehling, John, DuPont, Torvex Catalytic Reactor  Company.
October 19, 1982.  Catalytic incinerator system cost estimates.
Docket Reference Number II-E-40.*

Letter from Kroehling, John, DuPont, Torvex Catalytic Reactor
Company, to Katari, V., PES.  October 19, 1982.   Catalytic incinerator
system cost estimates.  Docket Reference Number II-D-66.*

Key, J.A.  Control Device Evaluation:  Catalytic Oxidation.   In:
Chemical Manufacturing Volume 4:  Combustion Control Devices.
U.S. Environmental Protection Agency™  Research Triangle Park,
N.C. Publication No.  EPA-450/3-80-026.  December 1980. Docket
Reference Number II-A-18.*

Telecon.  Siebert, Paul, Pacific Environmental Services, Inc.,
with Kenson, Robert, Met-Pro Corporation, Systems Division.   July  22,
1983.  Minimum size catalytic incinerator units.  Docket Reference
Number II-E-73.*

Memo from Paul Siebert, PES to Polymers Manufacturing Industry  NSPS
File.  October 5, 1984.  Condensation System Design and Cost Computer
Program for Polystyrene and Poly(ethylene terephthalate).   Docket
Reference Number II-B-93.*

Reference 17, p. 3-59.

Reference 17, pp. 3-72.

Rossini, F.D. et al. Selected Values of Physical and Thermodynamic
Properties of Hydrocarbons and Related Compounds, Comprising the
Tables of API Research Project 44.  Pittsburgh,  Carnegie Press,
1953.  pp. 652 and 682.  Docket Reference Number II-I-5.*

Reference 17, pp. 10-25 through 10-28.

Reference 17, pp. 10-12 through 10-15.

Ludwig, E.E.  Applied Process Design for Chemical and Petrochemical
Plants, Volume 3.  Houston, Gulf Publishing Company.  1965.   p. 80.
Docket Reference Number II-B-93, Attachment D.*

Reference 42, pp. 100, 101, and 104.

Reference 42, pp. 100 through 106.

                             E-68

-------
 45.   Telecon.   Meardon,  Ken,  Pacific  Environmental  Services,  Inc., with
      Mahan,  Randy,  Brown Finntube  Company.   July  30,  1984.  Cost estimates
      for various size condensers  (4.6 ft2  up to 34  ft2).  Docket Reference
      Number  II-E-90.*

 46.   Telecon.   Meardon,  Ken,  Pacific  Environmental  Services,  Inc. with
      Kurtz,  Ned, American Standard Heat Transfer  Division.  July 30, 1984.
      Cost estimates  for  various size  condensers.  Docket Reference
      Number  II-E-91.*

 47.   Memo from  K. Meardon,  PES to  Polymer Manufacturing NSPS  File.
      August  17,  1984.  Refrigeration  Units for Condensers.
      Docket  Reference Number  II-B-88.*

 48.   Reference  17, pp.   11-1  through  11-18.

 49.   Reference  17, pp. 3-71,  3-126, 3-214, and 12-46 through
      12-48.

 50.   Weast,  R.C., ed.  Handbook of Chemistry and Physics, fifty-third
      edition.  Cleveland, The Chemical Rubber Company, 1972.  p. F-36.
      Docket  Reference  Number  II-I-122.*

 51.   Thermophysical Properties of Refrigerants.  New York, American
      Society of  Heating, Refrigerating and Air-conditioning Engineers,  Inc.
      1976.   pp.  9 through 11, 105, and 106.  Docket Reference Number II-B-93
      Attachments AF and AG.*                                                '

 52.   Kern, .D.Q.  Process Heat Transfer.  New York, McGraw-Hill Book
      Company, 1950.    p. 306.  Docket Reference Number II-I-3.*

 53.   Reference 17, pp. 23-49.

 54.  Reference 42, pp. 57 and 58.

 55.  Reference 17, pp. 6-3.

 56.  McCabe,  W.L. and J.C. Smith.   Unit Operations of Chemical Engineering,
     second edition.   New York,  McGraw-Hill  Book Company.   1967.
     1007 p.   Docket Reference Number  II-I-12.*

 57.  Meardon, Ken, Pacific Environmental  Services, Inc.  to E.J.  Vincent,
     EPArCAS.  January 1983.  Designs  and  cost estimates  for ethylene
     glycol recovery  systems.   Docket  Reference Number II-B-62.*

58.  Telecons.   Meardon,  Ken,  Pacific  Environmental  Services,  Inc.,
     with R.  Smith,  Allied Fibers.   December  1982  and January  1983.
     Design and operating parameters of ethylene glycol  recovery systems
     Docket Reference Number II-E-54.*
                               E-69

-------
59.  Reference 17, p. 5-31.

60.  Chontos, L.W.  Find Economic Pipe Diameter via Improved Formula.
     Chemical Engineering.  £7_( 12): 139-142.   June 16,  1980.   Docket
     Reference Number II-I-59.*

61.  Memo from Desai, Tarun, EEA, to EB/S Files.   March 16,  1982.
     Procedure to estimate piping costs.  Docket Reference Number
     II-B-37.*

62.  Memo from Kawecki, Tom, EEA, to SOCMI Distillation File.  November 13,
     1981.  Distillation pipeline costing model documentation.   Docket
     Reference Number II-B-36.*

63.  Memo from Paul Sie,bert, PES, to Polymers and Resin File.  March 16,  1983.
     SOCMI Distillation NSPS Pipeline Costing Computer Program (DMPIPE),
     1981.  Docket Reference Number II-B-66.*

64.  Reference 29, Section 4.2, pp. 4-15 through 4-28.
* References  can  be  located in Docket Number A-82-19 at the
  U.S. Environmental Protection Agency Library, Waterside Mall,
  Washington, D.C.

                                  E-70

-------
                                   TECHNICAL REPORT DATA
                            (Please read Instructions on the reverse before completing)
1. REPORT NO.
      EPA-450/3-83-019a
                                                            3. RECIPIENT'S ACCESSION NO.
4. TITLE AND SUBTITLE
                                                            5. REPORT DATE
    Polymer Manufacturing Industry
    tion for  Proposed Standards
- Background  Informa-
                                                                       September 1985
                       6. PERFORMING ORGANIZATION CODE
7. AUTHOR(S)
                                                            8. PERFORMING ORGANIZATION REPORT NO.
9. PERFORMING ORGANIZATION NAME AND ADDRESS
   Pacific  Environmental Services,  Inc.
   1905 Chapel  Hill  Road
   Durham,  N.C.  27707
                                                            10. PROGRAM ELEMENT NO.
                        11. CONTRACT/GRANT NO.

                         68-02-3060
12. SPONSORING AGENCY NAME AND ADDRESS
                                                            13—T.YPE.OF REPORT AND PERIOD COVERED
   U.S.  Environmental Protection Agency
   Office  of Air Quality Planning and Standards
   Research Triangle Park,  North Carolina  27711
                        14. SPONSORING AGENCY CODE
                         EPA/200/04
15. SUPPLEMENTARY NOTES
16. ABSTRACT
 Standards  of performance  for  the control of volatile organic compound  emissions from
 the polymer manufacturing industry are being proposed under the authority of
 Section  111 of the Clean  Air  Act.  These standards  would apply to  new, modified,
 and reconstructed facilities  that manufacture polypropylene, polyethylene, polystyrene,
 or poly(ethylene terephthalate).  This document  contains background  information
 and environmental and economic impact assessments of the regulatory  alternatives
 considered in developing  the  proposed standards.
17.
                                KEY WORDS AND DOCUMENT ANALYSIS
                  DESCRIPTORS
                                               b. IDENTIFIERS/OPEN ENDED TERMS
                                     c.  COS AT I Field/Group
 Air  Pollution
 Volatile Organic Compounds
 Polymers
 Resin
 Polyethylene              *
 Polypropylene
 Polystyrene
 Poly(ethylene terephthalate)
           Air Pollution  Control
 13B
18. DISTRIBUTION STATEMENT

 Unlimited
          19. SECURITY CLASS /This Report)'
            Unclassified
21. NO. OF PAGES

     512
                                               20. SECUR! TY CLASS /This page)
                                                 Unclassified
                                                                          22. PRICE
EPA. Form 2220-1 (Rev. 4-77)
                       PREVIOUS EDITION !S OBSOLETE

-------

-------
-------
















5
e


g

?s


s


CO


-E£ CO
LU «g
IE CO
O CO
•— I LU
CO O
CO O
21 sf
LU
COTpB
cc 01 a:
LU-ULU
^-3
g"* ^P
O Cu.
O^-S
.
O. CO
a: o
O LU
I — CO
LU O
z a:
*-* a.



u
C-

*~

t
o

»2J
JS
*~

















C
a.' o
4-> i — CL-
IO •*->
QJ 3 en
3T O ^C
0) S- 1^
4-> 
•a QJ
0 3T

Cu QJ


r3 (O
ji» s

to E

s-



o
ro
to o
CD -M
2
ro QJ
t. c
3 f-
4-> U
ro c
£1 1—4


•g*.
ss
3 «3
cr s-
 U
c
c *-
o
£.3
tn
3 I.
' "1"*
O


O


a> t-
tss
ro ro
"* aj '

JJ *0 1
u c
10 *— «
-M











QJ
*£
ro
5-
ro








0
OJ
i-r 1
^^,
O
O

1— 1
OJ






^^
Lu
0
o
CO
CO
— I
o
o
CO
en
*— i








u?
o I
CO
o
0
cn
OJ



o
in
o
«-H t

O
o
«3-





tr
o
in
tn
i— i
i
o
in
1 r-l

EJ
o
CO

10


CJ
en
fl3 Q)
cw c
> ro
< O£

QJ


ro
QJ
a.

QJ
h—





=


1

* 	 f
ro
a.


CO
T— *
I
to
CO*
^
en
en
CL
O
CO
1— 1
a. t i


o
OJ

•— t



'en
CL
O
t— 1
in
ro
' ^ '

en
i

CO




i i i








*— •«
•i—
t/i
CL

O
10
to i i
Q_

CO
1—1
"*



aj n
en 3
a) c: x
> (O (fl

a>
a.












ii it








, 	 l: 	 ,
^;=
t~i J3
<—•—
O O
o o
CO O

LO n
OJ LO
U U II
ai ai
t/> C/1

en en
O 0
LO CO
OJ to

co 




it ii





"S"E
rj O
in tn
o o
00
o o
OJ CO
(J O
OJ QJ
II V) V)

CO CO
E E
 ro > ro
«=c sr  O
ro r—
or u.
= I-
QJ <5

t/)










1 II II


























u_^
CJ M-
cn u
CO
E 0
CO O
T— 1 *
i—i un
1 f 1 | • r-*
? O
cn co

• o
CD rH




1 I i t I














1 I 1 t 1







CJ Q) CJ
en cn en
o ro ro ro aj
en i- t- s_ en
C Q) CJ OJ C
ro > > >  -t-
QJ _J U
m u *r-
O -r~ U C
S- C f- ro
-M to c en
•i- en ro s-
= s- en o
O S-
CD
C-14




















































CO
en
_
ro •
s- o
ZI OJ


c= = c.
o o
O 4J
u_ to
en s_
•er ro QJ
QJ ^ C

3 QJ G
4J S- C
•M W ''"
tn en OJ
^D CJ -M
in CL ro
QJ in
JD -r- 0
C JJ 4-J
ro u
u « ro
•— O eC
o S- en
QJ C
i — Q- *i—
QJ S-
M— en "D
QJ
O
S- en (0
QJ c: j^

p ^ 4->

= Q 0
ro JD o