905R83116
United States
Environmental Protection
Agency
Office of Air Quality
Planning and Standards
Research Triangle Park NC 27711
March 1983
Air
Polymer
Manufacturing
Industry -
Background
Information for
Proposed
Standards
Preliminary Draft
Draft
EIS
-------
NOTICE
This document has not been formally released by EPA and should not now be construed to represent
Agency policy. It is being circulated for comment on its technical accuracy and policy implications.
Polymer
Manufacturing Industry -
Background Information
for Proposed Standards
Preliminary Draft
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
March 1983
-------
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 1-3
1.2.3 Energy Impacts 1-4
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 3-48
3.3 Fugitive VOC Sources and Emissions 3-55
3.4 Baseline Emissions 3-57
3.4.1 Process Emissions 3-57
3.4.2 Fugitive Emissions 3-62
3.5 References for Chapter 3 3-64
-------
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 4-23
4.2 Control Techniques for Fugitive Emissions 4-33
4.2.1 Leak Detection and Repair
Program 4-33
4.2.2 Preventive Programs 4-35
4.3 References for Chapter 4 4-37
5.0 MODIFICATIONS AND RECONSTRUCTIONS 5-1
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 6-13
6.2.1 Baseline Control 6-13
6.2.2 Control Techniques 6-14
6.2.3 Regulatory Alternatives 6-15
6.2.4 Summary of Regulatory Alternatives . . . 6-31
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 7-14
-------
7.2 Uater Pollution Impacts 7-20
7.3 Solid Waste Disposal Impacts 7-21
7.4 Model Plant Energy Impacts 7-25
7.4.1 Model Plant Energy Impacts 7-25
7.4.2 Industrywide Energy Impacts 7-30
7.5 Other Environmental Impacts 7-35
7.5.1 Noise Impacts 7-35
7.5.2 Irreversible and Irretrievable
Commitment of Resources 7-35
7.5.3 Environmental Impacts of Delayed
Regulatory Action 7-36
7.6 References for Chapter 7 7-38
8.0 COSTS
8.1 Cost Analysis of Regulatory Alternatives .... 8-1
8.1.1 Flare Design and Cost Basis 8-9
3.1.2 Thermal Incinerator Design and
Cost Basis 8-11
8.1.3 Catalytic Incinerator Design
and Cost Basis 8-13
8.1.4 Condenser Design and Cost Basis. . . . 8-15
8.1.5 Ethylene Glycol Recovery System
Design and Cost Basis 8-16
8.1.6 Fugitive Emission Control Program
Design and Cost Basis 8-17
8.1.7 Cost Analysis Results 8-33
8.2 Other Cost Considerations 8-54
8.2.1 Water Pollution Control
Regulations 8-54
8.2.2 Occupational Safety and Health
Regulations 8-56
-------
8.2.3 Toxic Substance Control Regulations. . 8-57
8.2.4 Solid and Hazardous Waste
Regulations 8-57
8.2.5 Clean Air Act 8-58
8.3 References for Chapter 8 8-63
9.0 ECONOMIC IMPACT
9.1 Industry Characterization 9-1
9.1.1 Industry Structure 9-1
9.1.2 Industry Profile 9-20
9.1.3 Five-Year Projections 9-25
9.2 Economic Impact Analysis 9-36
9.2.1 Economic Impact Assessment
Methodology: Revenue and Price. . . . 9-37
9.2.2 Economic Impact of VOC Potential
Polymers and Resins 9-43
9.3 Potential Socioeconomic and Inflationary
Impacts 9-50
9.3.1 Fifth Year Costs and Benefits 9-50
9.3.2 Impacts on Small Facilities 9-54
9.3.3 Other Impacts 9-55
9.4 References for Chapter 9 9-56
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-2
C.I.I Control Device C-3
C.I.2 Sampling and Analytical Techniques. . . . C-3
C.I.3 Test Results C-5
C.2 Thermal Incinerator VOC Emission Test Data ... C-5
C.2.1 Environmental Protection Agency (EPA)
Polymers Test Data C-9
C.2.2 Environmental Protection Agency (EPA)
Air Oxidation Unit Test Data C-17
IV
-------
C.2.3 Chemical Company Air Oxidation Unit Test
Data C-25
C.2.4 Union Carbide Lab-Scale Test Data .... C-33
C.3 Vapor Recovery System VOC Emission Test Data . . 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
APPENDIX D - Emission Measurement and Performance
Test Methods
I. Process VOC Sources D-l
I-D.l Emission Measurement D-2
I-D.2 Recommended Test Methods D-3
I-D.3 References D-3
II. Fugitive VOC Sources 0-4
11-D.I Emission Measurement Methods D-4
II-D.2 Continuous flonitoring Systems
and Devices D-7
I I-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 F-6
E.3 Thermal Incinerator Design and Cost Estimation
Procedure E-8
E.3.1 Thermal Incinerator Design Procedure. . . E-8
E.3.2 Thermal Incinerator Cost Estimation
Procedure ..... E-17
-------
E.4 Catalytic Incinerator Design and Cost Estimation
Procedure E-17
E.4.1 Catalytic Incinerator Design Procedure. . E-22
E.4.2 Catalytic Incinerator Cost Estimation
Procedure E-23
E.5 Surface Condenser Design and Cost Estimation
Procedure E-31
E.5.1 Surface Condenser Design E-32
E.5.2 Surface Condenser Cost Estimation
Procedure E-32
E.6 Ethylene Glycol Recovery Systems Design and Cost
Estimation Procedure E-34
E.6.1 Ethylene Glycol Recovery System Design. . E-34
E.6.2 Ethylene Glycol Recovery System Cost
Estimation Procedure E-34
E.7 Piping and Ducting Design and Cost Estimation
Procedure E-41
E.7.1 Piping and Ducting Design Procedure . . . E-41
E.7.2 Piping and Ducting Cost Estimation
Procedure E-41
E.8 References E-46
-------
List of Tables
1-1 Assessment of Environmental and Economic Impacts for
Each Regulatory Alternative Considered 1-3
3-1 Polypropylene (PP) Plant List 3-3
3-2 Low Density Polyehtylene (LDPE) Plant List 3-4
3-3 High Density Polyethylene (HOPE) Plant List 3-5
3-4 Polystyrene (PS) Plant List 3-6
3-5 Polyethylene Terephthalate (PET) Plant List 3-7
3-6 Characteristics of Vent Streams from the
Polypropylene Continuous Liquid Phase Slurry
Process 3-18
3-7 Characteristics of Vent Streams from the
Polypropylene fias Ph-ase
Process 3-23
3-3 Characteristics of Vent Streams from the
Low Density Polyethylene High-Pressure,
Liquid Gas Process 3-27
3-9 Characteristics of Vent Streams from the the
Low Density Polyethylene Low-Pressure, Gas
Phase Process 3-31
3-10 Characteristics of Vent Streams from the
High Density Polyethylene Low-Pressure, Liquid
Phase Slurry process 3-35
3-11 Characteristics of Vent Streams from the
High Density Polyethylene Low-Pressure Liquid
Phase Solution Process 3-39
3-12 Characteristics of Vent Streams from the
Polystyrene Batch Process 3-44
3-13 Characteristics of Vent Streams from the
Polystyrene Continuous Process 3-47
VII
-------
3-14 Characteristics of Vent Streams from the
Polyester DMT Process ........ 3-51
3-15 Characteristics of Vent Streams from the
Polyester TPA Process 3-54
3-16 Vapor Pressures of Major Organic Compounds
Used in the Polymers and Resins Segments Chosen for
NSPS Development 3-56
3-17 Uncontrolled Fugitive Emission Rates 3-58
4-1 Flare Emission Studies 4-11
6-1 Model Plant Characteristics for Process Emissions
from the Liquid Phase Polypropylene,
Process 6-3
6-2 Model Plant Characteristics for Process
Emissions from the Gas Phase Polypropylene,
Process 6-4
6-3 Model Plant Characteristics for Process Emissions
from the High Pressure Liquid Phase LDPE
Process 6-5
6-4 Model Plant Characteristics for Process Emissions
from the Low Pressure, Gas Phase LDPE/HDPE
Process 6-6
6-5 Model Plant Characteristics for Process
Emissions from the Low Pressure, Liquid Phase
HOPE Slurry Process 6-7
6-6 Model Plant Characteristics for Process
Emissions from the Low Pressure, Liquid
Phase HOPE Solution Process 6-8
6-7 Model Plant Characteristics for Process Emissions
from the Continuous Polystyrene
Process 6-9
6-8 Model Plant Characteristics for Process
Emissions from the DMT Poly(ethylene Terephthalate)
Process . . 6-10
vi n
-------
6-9 Model Plant Characteristics for Process
Emissions from the TPA Polyester
Process 6-11
6-10 Fugitive VOC Emission Model Plant Parameters 6-12
6-11 Control Specification for Fugitive Emissions
Under Regulatory Alternative 2 6-16
6-12 Regulatory Alternatives for the Liquid Phase
Polypropylene Process 5-19
6-13 Regulatory Alternatives for the Gas Phase
Polypropylene Process 6-21
6-14 Regulatory Alternatives for the High Pressure,
Liquid Phase LDPE Process 6-22
6-15 Regulatory Alternatives for the Low Pressure,
Gas Phase LDPE/HDPE Process 6-24
6-16 Regulatory Alternatives for the Liquid Phase
High Density Polyethylene Slurry Process 6-26
6-17 Regulatory Alternatives for the Liquid Phase
HOPE Solution Process 6-26
6-18 Regulatory Alternatives for Process Emissions
for the Continuous Polystyrene Process 6-29
6-19 Regulatory Alternatives for Process Emissions
from the DMT Poly(ethylene Terephthalate)
Process 6-30
6-20 Regulatory Alternatives for Process Emissions
from the TPA Poly(ethylene Terephthalate)
Process 6-32
6-21 Summary of Uncontrolled Emissions and Emission
Reductions for Regulatory Alternatives by
Model Plant 6-33
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
7-2a Industrywide Primary Air Quality Impacts of
the Regulatory Alternatives for Polymers
and Resins Plants (Mg/yr) 7-10
7-2b Industrywide Primary Air Quality Impacts of
the Regulatory Alternatives for Polymers
and Resins Plants (tons/yr) 7-12
7-3a Secondary Air Quality Impacts of the
Regulatory Alternatives for Polymer and
Resins Plants 7-16
7-3b Secondary Air Quality Impacts of the
Regulatory Alternatives for Polymer and
Resins Plants 7-18
7-4 Industrywide Solid Waste Impacts of the
Regulatory Alternatives for New Polymer
and Resin Plants 7-23
7-5 Volume of Biological Sludge Generaged by
Process Operations in New Polymer and Resins
Plants that Employ Flares, Thermal Incineration,
or Catalytic Incineration .... 7-24
7-6a Energy Impacts of the Regulatory Alternatives
for Polymer and Resin Model Plants 7-26
7-6b Energy Impacts of the Regulatory Alternatives
for Polymer and Resin Model Plants 7-28
7-7a Industrywide Energy Impacts of the
Regulatory Alternatives New Polymer
and Resins Plants 7-31
7-7b Industrywide Energy Impacts of the
Regulatory Alternatives New Polymer
and Resins Plants 7-33
-------
8-1 Sunmary of Regulatory Alternatives for
the node! Plants 3-2
8-2 Installation Cost Factors 8-5
8-3 Annualized Cost Factors for Polymers and
Resins NSPS (June 1980 Dollars) 8-6
8-4 Fugitive VOC Regulatory Alternative Control
Specifications 8-19
8-5 Fugitive VOC Emission Data for the Sources
in Polymers and Resins (P&R) Model Plants 8-20
8-6 Fugtiive VOC Regulatory Alternative Costs for
Polymers and Resins Model Units 8-21
8-7 Summary of Fugitive VOC Emission Control
Costs for the Sources in Polymers and
Resins Model Unit 8-22
8-8 Initial Leak Repair Labor-Hours Requirement
for Valves for the Model Unit 8-24
8-9 Total Annual Costs for Initial Leak Repair
for Valves for the Model Unit (May 1980 Dollars . . . 8-24
8-10 Annual Monitoring and Leak Repair Labor
Requirements for Valves for the Model Unit
(Monthly Leak Detection and Repair Program) 8-25
8-11 Annual Monitoring and Leak Repair Labor
Costs for Monthly Monitoring of Valves for
the Model Unit (May 1980 Dollars) 8-26
8-12 Initial Leak Repair Labor-Hours Requirement
for Pump Seals for the Model Unit 8-27
8-13 Total Annual Costs for Initial Leak Repair
for Pump Seals for the Model Unit (May
1980 Dollars) 8-27
8-14 Annual Monitoring and Leak Repair Labor
Requirements for Pump Seals of the Model
Unit (Monthly Leak Detection and Repair
Program) 8-28
-------
8-15 Annual Monitoring and Leak Repair Labor
Costs for Monthly Monitoring of Pump Seals
for the Model Unit (May 1980 Dollars) 8-29
8-16 Relief Valve Control Costs for Rupture Disk
Systems with Block Valves and Three-Way
Valves (May 1980 Dollars) 8-31
8-17 Capital and Met Annualized Costs for Control of
Emissions from Safety/Relief Valves for the
Model Unit (May 1980 Dollars) 8-32
8-18 Capital and Net Annualized Costs for Control of
Emissions from Compressor Seals for the
Model Unit (May 1980 Dollars) 8-34
8-19 Capital and Net Annualized Costs for Control of
Emissions from Sampling Systems in Model Unit
(May 1980 Dollars) 8-35
8-20 Polypropylene, Liquid Phase Model Plant
Regulatory Alternatives Costs (June 1980
Dollars) 8-36
8-21 Polypropylene, Gas Phase Model Plant
Regulatory Alternatives Costs (June 1980
Dollars) 8-37
8-22 Low Density Polyethylene, Liquid Phase Model
Plant Regulatory Alternatives Costs (June
1980 Dollars) 8-38
8-23 Low and High Density Polyethylene, Gas Phase Model
Plant Regulatory Alternatives Costs (June
1980 Dollars) 8-39
8-24 High Density Polyethylene, Liquid Phase-Slurry
Model Plant Regulatory Alternatives Costs (June
1980 Dollars) 8-40
8-25 High Density Polyethylene, Liquid Phase-Solution
Model Plant Regulatory Alternatives Costs (June
1980 Dollars) 8-41
3-26 Polystyrene-Continuous Model Plant Regulatory
Alternatives Costs (June 1980 Dollars) 8-42
-------
8-27 Polyester (PET) - DMT Process Model Plant
Regulatory Alternatives Costs (June 1980 Dollars) . . 8-43
8-28 Polyester (PET) - TPA Process Model Plant
Regulatory Alternatives Costs (June 1980 Dollars) . . 3-44
8-29 Costs and Associated Emission Reductions of
Regulatory Alternatives for Polypropylene,
Liquid Phase Process 8-45
8-30 Costs and Associated Emission Reductions of
Regulatory Alternatives for Polypropylene,
Gas Phase Process 8-46
8-31 Costs and Associated Emission Reductions of
Regulatory Alternatives for LDPE, High Pressure,
Liquid Phase Process 8-47
8-32 Costs and Associated Emission Reductions of
Regulatory Alternatives for LDPE/HDPE, Low Pressure,
Gas Phase Process 8-48
8-33 Costs and Associated Emission Reductions of
Regulatory Alternatives for HOPE, Liquid
Phase Slurry Process 8-49
8-34 Costs and Associated Emission Reductions of
Regulatory Alternatives for HOPE, Liquid
Phase Solution Process 8-50
8-35 Costs and Associated Emission Reductions of
Regulatory Alternatives for Polystyrene,
Continuous Process 8-51
8-36 Costs and Associated Emission Reductions of
Regulatory Alternatives for PET/DMT
Process 8-52
8-37 Costs and Associated Emission Reductions of
Regulatory Alternatives for PET/TPA
Process 8-53
8-38 Total Fifth-Year Net Annualized Cost of
Process and Fugitive Emission Controls for
Polymers and Resins Facilities Affected
by NSPS 8-55
-------
8-39 Number and Capacity of Plants to be Affected
by Both VOL Storage and Polymers and Resins
Standards Through 1988 8-60
8-40 Storage Tank Compliance Costs for Plants to be
Affected by Both VOL Storage and Polymers and
Resins Standards Through 1988 8-61
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, 1975-1981 9-17
9-7 Domestic Consumption of Polymers and Resins
by End-Use Market and Process Type, 1981 9-18
9-8 Shipments of Polymers and Resins by
Major Market, 1981 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 Estimated Required New Capacity for
Polymers and Resins, 1988 9-31
9-12 Projected Number of New Polymers and Resins
Plants, by Process to 1988 9-34
9-13 Operational Characteristics of Polymers and
Resins Model Plants Built 1984 Through 1988
(June, 1980 Dollars) 9-45
xiv
-------
9-14 Polymers and Resins Model Plant Control Costs
and Maximum Price Increases by Plant Product,
Process, and Regulatory Alternative (June, 1980
Dollars) 9-47
9-15 Fifth Year Net Annualized Cost to Society of
Regulatory Alternatives by Model Plant Product
and Process (June, 1980 Dollars) 9-51
9-16 Upper Boundary of Total Annualized Fifth Year
Cost to Society (June, 1980 Dollars) 9-53
C-l Emission Analyzers and Instrumentation
Utilized for Joint EPA/CMA Flare Testing C-6
C-2 Steam-Assisted Flare Testing Summary C-7
C-3 Summary of Thermal Incinerator Emission
Test Results C-8
C-4 Typical Incinerator Parameters for ARCO
Polymers Emission Testing Based on Data
from August 1981 C-13
C-5 ARCO Polymers Incinerator Destruction
Efficiencies for Each Set of Conditions C-16
C-6 Air Oxidation Unit Thermal Incinerator
Field Test Data C-21
C-7 Destruction Efficiency Under Stated Conditions
Based on Results of Union Carbide Laboratory
Tests C-34
C-8 Comparisons of Emission Test Results for Union
Carbide Lab Incinerator and ROHM & HAAS Field
Incinerator C-38
C-9 Equipment Counts and Emissions for Fugitive
VOC Emission Sources in SOCMI Model Units C-42
xv
-------
C-10 Equipment Inventories and Emission Estimates for
Fugitive VOC Emission Sources in Polymers
and Resins Plants C-44
E-l Procedure to Design State-of-the-Art (0.5 Mach)
Elevated Steam-Assisted Smokeless Flares E-3
E-2 Flare Budget Purchase Cost Estimates Provided
by National Air Oil Burner, Inc., In October 1982
Dollars E-7
E-3 Capital and Annual Operating Cost Estimation
Procedure for State-of-the-Art Steam-Assisted
Smokeless Flares E-9
E-4 Worksheet for Calculation of Waste Gas
Characteristics (molecular weight, molecular formula,
lower heating value in Btu/scf) E-ll
E-5 Generalized Waste Gas Combustion Calculations .... E-14
E-6 Procedure to Design Thermal Incinerators
Combusting Streams With Lower Heating Values
(LHV) Greater than 60 Btu/scf E-15
E-7 Capital and Annual Operating Cost Estimates for
Thermal Incinerators Without Heat Recovery E-20
E-8 Operating Parameters and Fuel Requirements
of Catalytic Incinerator Systems E-24
E-9 Gas Parameters Used for Estimating Capital and
Operating Costs of Catalytic Incinerators E-26
E-10 Catalytic Incinerator Vendor Cost Data E-27
E-ll Calculation Procedure for Estimation of Annualized
Costs for Catalytic Incinerator Systems E-30
E-12 Procedure to Calculate Heat Transfer Area of An
Isothermal Condenser System E-33
E-13 Capital and Annual Operating Cost Estimates for
2
a 20 ft Condenser System for the Streams from
the Continuous Polystyrene Model Plant E-35
E-14 EG Recovery Costs for Baseline System E-36
E-15 EG Recovery Costs for Regulatory Alternative
System E-33
xvi
-------
E-16 Piping and Ducting Design Procedure E-42
E-17 Piping Components E-43
E-18 Installed Piping Costs E-44
E-19 Installed Ducting Cost Equations, December 1977
Dollars E-45
xvn
-------
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, Liquid Phase Process. . . 3-21
3-5 Simplified Process Block Diagram for the Low Density
and High Density Polyethylene Low-Pressure, Liquid
Phase Process 3-29
3-6 Simplified Process Block Diagram for the High Density
Polyethylene Low-Pressure, Liquid Phase Process
Slurry 3-33
3-7 Simplified Process Block Diagram for the High
Density Polyethylene, High-Pressure, Liquid Phase
Solution Process 3-37
3-3 Simplified Process Block Diagram for the Polystyrene
Batch Process 3-42
3-9 Simplified Process Block Diagram for the Polystyrene
Continuous Process 3-45
3-10 Simplified Process Block Diagram for the Polyester DMT
Process 3-50
3-11 Simplified Process Block Diagram for the Polyester TPA
Process 3-53
4-1 Steam Assisted Elevated Flare System 4-4
4-2 Steam Injection Flare Tip 4-5
4-3 Distributed Burner Thermal Incinerator 4-16
4-4 Catalytic Incinerator 4-20
4-5 Condensation System 4-27
4-6 Two Stage Regenerative Adsorption System 4-30
4-7 Packed Tower for Gas Absorption 4-32
xvi n
-------
C-l Flare Sampling and Analysis System C-4
C-2 Schematic of Incineration System at ARCO
Polypropylene Facility C-ll
C-3 Incinerator Combustion Chamber C-18
C-4 Petro-Tex Oxo Unit Incinerator C-26
C-5 Off-gas Incinerator, Monsanto Co.,
Chocolate Bayou Plant C-31
C-6 Thermal Incinerator Stack Sampling System C-32
E-l Purchase Costs for Thermal Incinerator Combustion
Chambers E-18
E-2 Installed Capital Costs for Inlet Ducts, Waste Gas
and Combustion Air Fans, and Stack for Thermal
Incinerator Systems with no Heat Recovery E-19
E-3 Installed Capital Costs for Catalytic Incinerators
With and Without Heat Recovery E-29
xix
-------
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 111 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, nine 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:
Monthly monitoring for leaks from valves in gas and light
liquid service, and pump seals in light liquid services;
Weekly visual inspection for liquid leakage from pump seals in
light liquid service;
Installation of controlled degassing vents on compressors,
rupture disks on relief valves, and caps on open-ended lines;
and
Closed-purge sampling on sampling connections.
1-1
-------
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 and the petroleum refining industry.
Similarities between the polymers and resins industry and these other
two industries 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 ethylene glycol emissions from
the process by using an alternative ethylene glycol recovery system
that is more efficient than the system under Regulatory Alternative 1.
No additional regulatory alternatives were developed for the poly(ethylene
terephthalata) model plant using the terephthalic acid process.
Regulatory Alternative 3 for the eight remaining model plants
further reduces emissions through the additional control of process
emissions. For the polypropylene and polyethylene 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 polystyrene model plant, Regulatory Alternative 3 attains additional
control by applying additional recovery to the process emissions.
Regulatory Alternative 3 for poly(ethylene terephthalate) plants using
the dimethyl terephthalate process achieves additional control by
combustion of the methanol stream from the methanol recovery system.
A fourth regulatory alternative (Regulatory Alternative 4) was
developed for three of the nine model plants. This alternative applied
combustion control to additional process emissions.
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
Air
impact
_2**
+2**
+4**
+4**
Water
impact
0
+1**
+1**
+1**
Sol id
waste
impact
0
0
-1*
-1*
Energy
impact
0
+2*
+3*
+3*
Noise
impact
_!**
0
-1**
_2**
Economic
impact
0
-1*
-1*
-1*
KEY: + Beneficial impact
- Adverse impact
0 No impact
1 Negligible impact
2 Smal1 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 eight model plants plus Regulatory Alternative 2
impacts for the poly(ethylene terephthalate) model plant using the
terephthalatic acid process as no Regulatory Alternative 3 was developed
for it. Finally, Regulatory Alternative IV corresponds to Regulatory
Alternative 4 for three model plants plus Regulatory Alternative 3
impacts for five model plants plus Regulatory Alternative 2 impacts
for one model plant.
1.2.1 Air Emissions Impacts
Total VOC emissions from new plants in these industry segments in
1983 are projected to be approximately 17.9 gigagrams (Gg) under
Regulatory Alternative I, compared to 13.6, 3.4, and 3.3 Gg under
Regulatory Alternatives II, III, and IV, respectively. The average
percent emission reductions from the Regulatory Alternative I level
achieved by Regulatory Alternatives II, III, and IV are 24, 81, and
82 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 and IV due to the use of catalytic incinerators.
Spent catalyst from catalytic incinerator use may be generated at an
annual rate of 2.9 m3 (102 ft3) and 3.0 m3 (135 ft3), respectively,
under these two alternatives.
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.
1.2.3 Energy Impacts
Under Regulatory Alternative II, implementation of fugitive
controls in seven of the nine model plants and of a more efficient
ethylene glycol recovery system in poly(ethylene terephthalate) plants
result in a net decrease in energy usage from what otherwise would
1-4
-------
occur under Regulatory Alternative 1. This occurs because the energy
credit obtained through recovered VOC is greater than the energy
expended to implement the controls.
Under Regulatory Alternative III, a net decrease in energy usage
is also obtained. Six of the nine industry segments realize a net
decrease, while the other three segments realize a net increase in
energy usage.
Under Regulatory Alternative IV, a net decrease in energy usage
across the industry is again obtained, with six segments realizing a
net decrease and three segments realizing a net increase in 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 dhole would
realize a net annual credit of around $0.2 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 credit of $0.6 million in the poly(ethylene terephthalate)
plants using the terephthalatic acid process up to $1.6 million in
polyethylene plants using the gas phase technology.
Under Regulatory Alternative III, the industry as a whole would
spend an net annual amount of $2.8 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 annual credits in poly(ethylene terephthalate) and polystyrene
plants up to $3.2 million for high density polyethylene plants using a
solution process.
Under Regulatory Alternative IV, a fifth year annual cost of
$4.3 million over and above what would be otherwise spent under Regulatory
Alternative I would be realized by the industry as a whole. Again,
1-5
-------
the range of annual costs for individual industry segments range from
net annual credits in the poly(ethylene terephthalate) and polystyrene
segments to a cost of $3.2 million in high density polyethylene,
solution process segment.
Under the ,-nost costly combination of individual regulatory
alternatives for each model plant, total additional annualized costs
of controls in 1988 are estimated to be $4.9 million. The potential
adverse economic impacts of these regulatory alternatives are expected
to be very minor in view of the 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 amendnents 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 thesa 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 ;nay 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-control led 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 deternination 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), "Mo 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
-------
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
Pull man-Kellogg 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. This study
examined 16 polymers and resins categories with potentially large VOC
emissions:
Acrylics Polyester Fibers
Alkyds Polypropylene
High Density Polyethylene Polystyrene
Low Density Polyethylene Polyvinyl Acetate
Mel amine Formaldehyde Polyvinyl Alcohol
Nylon 6 Styrene-Butadiene Latex
Nylon 66 Unsaturated Polyester Resins
Phenol Formaldehyde 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 categories 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:
1. Polypropylene
2. Low density polyethylene
3. High density polyethylene
4. Polystyrene, and
5. 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
3-2
-------
Table 3-1. POLYPROPYLENE (PP) PLANT LIST3
Capacity
Company Location Gg/yr
ARCO Polymers, Inc. La Porte, TX 181
Amoco Chemical Corp. Chocolate Bayou, TX 234
El Paso Polyolefins Co. Odessa, TX 68
Pasadena, TX 68
Exxon Chemical Co. Baytown, TX 181
Gulf Oil Chemical Co. Cedar Bayou, TX 181
Hercules, Inc. Bayport, TX 204
Lake Charles, LA 395
Northern Petrochemical Co. Morris, IL 91
Phillips Chemical Co. Pasadena, TX 91
Shell Chemical Co. Norco, LA 136
Woodbury, NJ 136
Soltex Polymer Corp. Deer Park, TX 91
Texas Eastman Co. Longview, TX 64
USS Chemicals La Porte, TX 159
Kenova, WV 75
aSource: SRI International, 1982 Directory of Chemical Producers,
United States.
3-3
-------
Table 3-2. LOW DENSITY POLYETHYLENE (LDPE) PLANT LIST"3
Company
Allied Chemical Co.D
ARCO Polymers, Inc.
Chenplex Co.
Cities Service Co.0
Dow Chemical U.S.A.
E.I. du Pont de "lemours & Co., Inc.
El Paso Polyolefins Co.
Exxon Chemical Co.
Gulf Oil Chemical Co.
ilobil Chemical Co.
National Distillers & Chemical Corp.
Northern Petrochemical Co.
Phillips Chemical Co.
Texas Eastnan Co.
Union Carbide Corp.
United Foam Corp.
Location
Orange, TX
Port Arthur, TX
Clinton, I A
Lake Charles, LA
Freeport, TX
Plaquemine, LA
Orange, TX
Victoria, TX
Odessa, TX
Pasadena, TX
Baton Rouge, LA .
fit. Bel view, TX°
Cedar Sayou, 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,
3g/yr
-
181
188
329
447
254
211
109
131
68
299d
154a
239
129
-
136
249
75
273e
_f
166
5449
272
141
73
aSource: SRI International, 1982 Directory of Chemical Producers,
United States, unless otherwise indicated.
Source: Texas Air Control Board communication.
cln a letter dated March 9, 1982, Cities Services Co. indicated that
they were closing all their polymer and resins manufacturing plants.
dSource: Chemical Engineering. October 18, 1982. p. 26.
eCapacity obtained from April 14, 1982, letter from Northern
Petrochemical Company.
^Primarily HOPE produced; small portion of 580 Gg total capacity used
for LDPE.
^Capacity 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. 925.
-------
Table 3-3. HIGH DENSITY POLYETHYLENE (HOPE) PLANT LIST'
Company
Location
Capacity,
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
261fc
7
283
680
340
91
Source: SRI International, 1982 Directory of Chemical Producers,
United States.
Expansion from 191 Gg/yr to 261 Gg/yr completed in 1982.
Engineering. October 4, 1982. p. 25.
Chemical
3-5
-------
Table 3-4. POLYSTYRENE (PS) PLANT LISTa
Conpany
A&E Plastics, Inc.
American Hoechst Corp.
Amoco Chemical Corp.
ARCO Polymers, Inc.
BASF Wyandotte Corp.
Carl Gordon, Inri., Inc.
Cosden Oil & Chemical Co.
Crest Container Corp.
Dow Chemical Corp.
Gulf Oil Chemical Co.
Huntsman-Goodson Chem. 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, '/A
Leominster, MA
Peru, IL
Jo! let, IL
Torrance, CA
Mil low 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, TX
Troy, OH
Hazel ton, PA
Hoi yoke, MA
Joliet, IL
Santa Ana, CA
Addyston, OH
Decatur, AL
Springfield, MA
Copley, OH
Leoninster, 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
36
100
91
86
64
141
13
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
-
-
-
Source: SRI International, 1982 Directory of Chemical Producers, United States,
unless otherwise indicated.
e"
Industry communications.
Source: Organic Chemical Producers Data Base. Product Data Report - Nationwide.
February 17, 1981. p. 943.
This plant is not currently in production. Letter from Gulf Oil to Texas Air
Control Board. July 28, 1982.
lonsanto's Long Beach plant has been closed.
In rmd-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 nas been closed.
3-6
-------
Table 3-5. POLYETHYLENE TEREPHTHALATE (PET) POLYESTER PLANT LISr
01
ant
Akzona Inc.
Allied Corp.
American
Hoechst Corp.
Avtex Fibers, Inc.
E.I. du
Eastman
Pont de Nemours
Kodak Co.
Fiber Industries, Inc.
Location Process
Central, SC DMT
Lowland, TN DMT
Moncure, NC TPA
Spartanburg, SC OMT/TPA
Greer, SC
Lewistown, ?A OMT/TPA
Camden, SC
Charleston, SC
Chattanooga, TN DMT/TPA
Kins ton, NC
Old Hickory, TN
Wilmington, NC
Old Hickory, TN
Wilmington, NC
Brevard, NC
Parlin, NJ
Circleville, OH
Florence, SC
Columbia, SC DMT
King sport, TN DMT
Rochester, NY
Windsor, CO
Fayetteville, NC OMT/TPAC
Florence, SC
Greenville, SC DMT/TPAC
Salisbury, NC TPAC
Shelbv, NC DMTC
Average
capacity,
3g/yr Product
25 ciber
43 Fiber
39 ciber
261 Fiber, Bottle Resins
32 Bottle Resins, Film
18 Fiber
736 Fiber
249
566
14 Filn
7 Film
29 Film
34 Film
204 Fiber
238 Fiber, Bottle Resins
25 Filn
11 Film
(36) Fiber, Bottle Resins
680 Fiber
Firestone Tire &
Rubber
Goodyear
Rubber
Co.
Tire &
Co.
ICI Americas Inc.
Minnesota Mining &
Manufacturing Co.
Monsanto
Ronm and
Co.
Haas Co.
Hopewel 1 , VA TPA
Scottsboro, AL DMT/TPA
Point Pleasant, WV
Hopewell, VA
Decatur, AL
Greenville, SC
Decatur, AL DMT/TPA
Fayetteville, "1C
23 Fiber
11 Fiber, Sottle Resins
122 Bottle Resins
36 Filn
25 Him
10 Filn
91 Fiber
91 Bottle Resins
Source: SRI International, 1982 Directory of Chemical Producers, United States.
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 manufacturers of polyesters other than PET.
DMT - Dimethyl terephthalate process.
TPA - Tareohthalic acid process.
Industry correspondence.
Does
c.
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 rigid, 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.
fielded 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,
2
profiles, wire and cable coatings, and corrugated packing sheets.
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
3
production. Other forms account for the remaining 28 percent.
The major sectors using polypropylenes are consumer/institutional
(19 percent), furniture/furnishings (13 percent), packaging (16 percent),
3-8
-------
transportation (12 percent), and electrical/electronics (7 percent).
Other uses account for the remaining 28 percent.
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 LDPE has appeared - linear low density
polyethylene (LLDPE). LLDPE combines the linear molecular structure of
HOPE with the physical and optical properties of conventional LDPE, and
4
its overall properties are superior to those of conventional LDPE.
Polyethylenes are often extruded into film, sheets, pipe, or profiles,
injection molded, blow molded, rotationally molded, foamed, or formed in
5
other ways.
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).
Like conventional LDPE, LLDPE is suitable for many end-uses.
Specific applications may include housewares, lids, closures, blow
molded parts (such as toys, bottles, and drums), wire and cable insulation,
extruded pipe and tubing, and industrial and consumer films such as food
O
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
6
bags. 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
Q
account for 32 percent. End-use sectors for HOPE include packaging
(45 percent), consumer/institutional (11 percent), building and construction
(9 percent), and other sectors (35 percent).'
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.
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.
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. 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.11
3.1.1.4 End-Uses of Polyester Resin, Po1y(Ethy1ene 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 which 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; those that can be anticipated based
on the process flow diagram and those that can be identified only by
sampling procedures, such as 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 material preparation, polymer production,
3-11
-------
RAW MATERIALS
STORAGE AND
PREPARATION
POLYMERIZATIO
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,
(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
and drying. The process descriptions in this section are representative
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, liquid phase process.
4. LDPE - low-pressure, gas phase process.
5. HOPE - low-pressure, liquid phase slurry process.
6. HOPE - low-pressure, liquid phase solution process.
7. HOPE - low-pressure, gas phase process.
8. Polystyrene - batch process.
9. Polystyrene - continuous process.
10. PET - Dimethyl terephthalate (DMT) process.
11. 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
3
propylene. With a density of 0.902 to 0.904 g/cm , it is one of the
lightest of the commercial thermoplastics. The general formula for
polypropylene is:
... CH0 CH CH0 CH CH0 CH CH0 ...
2 i 2 | 2 i 2
CH-3 CH-j CHo
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 sane 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 crystal!inity.
Only the isotactic polypropylene is of commercial interest. Atactic
resin, an undesirable byproduct of all polypropylene processes, represents
12
about 7 percent of the total product.
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,
and can be a batch or continuous process. Batch polymerization is
particularly applicable 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
13
typically a combination of titanium chlorides and aluminum alkyls. In
conventional liquid phase processes, the catalyst provides a relatively
14
low polymer yield on the order of 500 to 1,000 units/unit of catalyst.
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 recently developed 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/unit of catalyst. 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 previously required for catalyst removal and recovery. The
elimination of a 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 by 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
-------
*-i O
C >i
i, (J
3 01
«- C
f V
,
a.
ai -c
"-> en
tn 01
OJ
CO
3
O
C
O
O)
c:
OJ
Q.
O
a.
OJ i/l
OJ
> u
O
s_ s_
o a-
S s-
-------
may use either a different organic or simply liquid propylene as a
diluent in which the polymer forms a slurry. The only difference between
the two 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), process
diluent (which acts as a heat transfer agent), a polymer suspension
medium, and a heterogeneous Ziegler-Natta type catalyst. Hexane is
often used as a process diluent, although some processes use mixtures of
other aliphatic hydrocarbons. The catalyst (2) is sometimes 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
15
temperatures are used. Reaction is carried out at temperatures of
about 60°C (140°F) for approximately 8 hours. A portion of the reaction
effluent, which consists of polymer, monomer, and diluent, is continuously
drawn 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 nethanol 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 is identified in Figure 3-2. The total
process VOC emission rate for a conventional slurry process is almost
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 22.8 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
-------
i- O
<
o o o o o
53
=5 !
0)
"<-> 3
*-»> O
i Wl 4-» >
w e 3 s-
a. w > a)
> NI
4_> 1_
5 c
3
25 *
* |
o w}
§ 8
o ^ *-»
3J O
i- 1. .3
*2 5 ti
vi t !/> o o
V) M O. 1=
, =S :u
CJ C) CJ * W
O£ S5 Z wl A £
m " (j
-------
The vent fron the catalyst preparation section, Stream A, releases
diluent continuously. The reactor vent, Stream B, is continuous and
releases monomer and diluent.
The combined total of the decanter and neutralizer vents, Streams C
and D, is usually the largest VOC emission source in the process. The
constituents are methanol or isopropyl alcohol (if used), C3 hydrocarbons,
and diluent. Hewer, 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 E, 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 F, 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 G, 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 H. 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 as 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
variables such as temperature, pressure, catalyst concentration, or
catalyst activity, and the amount of hydrogen used for molecular weight
control.
3-19
-------
3.2.1.2 Polypropylene Gas Phase Process. The gas phase process
for producing polypropylene is relatively new and is currently 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).
3.2.1.2.2 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
3-20
-------
CATALYST
CO-CATALYST
HEXANE
PROPYLENE
MONOMER
STORAGE
(3)
CATALYST
MIX
DRUM
(1)
GAS
PHASE
REACTOR
(2)
NaOH
SOLUTION
NaCL
SOLUTION
'ATE'IALS
FLUIDIZED
BED CATALYST
DEACTIVATOR
(4)
8AGHOUSES
(6)
EXTRUSION/
PELLET
BLENDING 4
STORAGE
(8)
Figure 3-3.
Simplified Process Block Diagram for the Polypropylene
Gas Phase Process
3-21
-------
stream (B), are currently controlled by a flare system which handles
both streams, according to one industry source. The uncontrolled VOC
emission rate for the entire process, based on the average expected
flowrate from these two streams, is 36.5 kg VOC/flg 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
Catalyst I I II
w _ r r . (r H ^ r r
^m^^H^^MtMBM^H^^^HiBBMHHMB^p w \j - V w n* it f *_* - w »
n-2 | |
H H H H
Conventional LDPE resins have a high degree of branching with
densities that range from 0.910 to 0.935 g/cm 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, Liquid Phase 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.
The high-pressure process is a free radical process utilizing pressures
of several thousand atmospheres to polymerize ethylene and copolymers of
18
ethylene, and nonolefinic comonomers such as vinyl acetate. High-pressure
processes, however, are not capable of polymerizing propylene or higher
19
olefins to their corresponding polyolefin. 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
?0
polymer in the reactor, the reactor configuration," reaction temperature,
21
or type and concentration of the catalyst.
3-22
-------
c
c.
eg - 01
c e c
Ol Ol Oi Ol O:
>, 5 s S >i S c
C. CL -T3 r ^. CL fT3
O O X T3 C C X
1_ !- SJ >i ' S- Ol
a. o. = a- c.
c.
o
o
a.
3m
C£ to
U- 00
LU
Sc^
< ce.
» 00
LL. O
O
(/i >^
t/1 (/)
01 a.
i.
c.
i-
OJ
O «^ U
ai
U
«
Ol
C
c
o
CO
O E
§ .2 .
Q. 4^ C
(/l ITS O
01 M -
I- _J
i- L. fl
o ai u
! £" "5 L
! <-> n. c
tfl Ol
3 II -3
o
C C£
>> s-
= U 4J
o at vi
*- >
4J O 1.
« U O
g £ *
O ro
C '-J
'- 01
C C£
co 5;
3-23
-------
3.2.2.1.1 Process description. Figure 3-4 is a schematic
representation of the liquid phase, 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
22
varies from about 10 to 100 ppm.
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).22 The total residence time of the
reactants varies from 45 to 60 seconds in a tube reactor and from 25 to
23
40 seconds in an autoclave reactor. 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
homopolymer or copolymer product is extracted as the low-pressure separator
bottoms which still contain residual ethylene monomer. These product
3-24
-------
>,
-t->
(/>
3 <"
O co
_! ,
'o
-------
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, liquid phase process.
The offgas stream characteristics for the liquid phase process are shown
in Table 3-8. The uncontrolled VOC emission rate for the entire process
is 2.9 kg VOC/Mg product. 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 emergency reactor vent, Stream A, is intermittent. It is
activated either manually or automatically as an emergency blowdown
during a process upset. The total mass and instantaneous flowrate can
vary dramatically depending on the nature of the upset. This, the major
VOC emission source in the process, 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 B and D, are continuous
and consist of mostly air with small amounts of ethylene. These streams
are usually emitted to the atmosphere.
The emergency vents, Stream C, are intermittent and consist of VOC
gases from a number of sources from the plant other than the reactors.
The wax blowdown system can be a source of significant ethylene losses.
The emission rate is highly dependent upon the design of the wax blowdown
and discharge system.
A considerable number of LDPE plants are located near or integrated
with a plant which manufactures ethylene. As a result, the unreacted
3-26
-------
1
i
i
1 1
I !
1
|
1
1
)
i
1
i
i
1
1
i
i
f
I
1
I
UJ
z:
1 1 1
_i
5-
h-
LLJ !
>-
I [
£
>~
1
CO
2T
LU
C
2re
o on
1 00
LU
LU <_J
=£ O
h tt
Q-
O LU
ce on
LU <
oo SI
^ o
LU > i
cz r:
fcs
_i i
= !
1_1LU
*" =
U-£
O 00
UJ
00 ~ ]
I *.
1 c
I 0
^
4_>
»*
O
O.+J
c 3
c
!
1
oT
i-
c/i en
tn -r-
OJ (/l
S- C.
D_
*
eu
L.
3
M
(O
S-
O) O
Q.O
S
K-
c cn
o r:
to (U O O
tn *-> O 3
r- (O > -O
E s- O
LU cn i_
JaC C.
1 OJ
1 s.
3
-M
TO
ZT
1
t
C
0 C
C C3 OJ
OJ r- C CJ
(U C 2S t. i
ro CL O. >^
3sj= o c j:
.C 4-J 1^ tn 4-31-
M LU Q^ i i LU
LU «C
O tr> cvj csj
ro .00
i t t-H CSJ O
LT> CTl
c^ d
0
CM O
1
O
U)
r^ CM
r-f" 0
4_>
c
OJ (/>
4-> 3
jJ O
r- 3
s =
0) J«)
4-> =
c o
(J
^ i i
- = I
oo t3 !
. . I 1 1 i \
rw i 1
ui i 1
1
c:
o;
>
0 ' f '
=c '
ex
^
1
< ' ' OJ
1 ! t=
0 |
1
1
CO
ro
u
ro
ai
4J
>» =
U OJ
c >
i a>
CTl i.
~ 0)
i i £?
UJ 0
:
|
!
1
1 1 1
i ' '
1
i !
1 1 i
i j i
1 LJ '
1
1 1
1 ' 5 !
1 I 01
<£ C2
! ' *- !
M +j >
j ! !
> i i
< i i
p i i
i i i
i i
1 ! 1
' ' -C 1
1 1 uT= '
1 = 1
! a)
u ^-» ; c£ u_
M o ' ' ' Q- C_
1 '_ i) '
' c_ ^o
i !
r
O
OJ C
OJ C «
C CJ C.
CJ QJ r C
r- C >, fc.
>, n3 Q_ O.
£ J= O C
+J -M S_ l^
LU LU O_ i *
PO O LD CNJ
,jp _ ^H CM
C^
0
CM
i
O
(/I
a>
s_
«
>
CM
a
c
OJ
4-5
4->
^
C
a>
4_>
C
CT!
C
^
3
U
X
i;
si
>
>,
u
c s-
a> o
CH4-3
!- 'J
fJJ fO
= 01
UJ -
r^r
EC
UJ
"~
O
O!
c
(U
.-
>,
J? !-
--J -r-
UJ «£
CM CO
C 01
CT1
t-
c
4_J
'->
ra
01
U
>,
(J
c
OJ
CT
1.
GJ
t=
GJ
a>
! 5
0
«J-
uo
co cr>
o cv
i^
4_J
a;
U
X
OJ
w
c
o
4->
u
0)
Wl
I/)
w
Of
U
o
1_
cu
_
*
o
1_
<4-
to
4_)
CJ
>
n
^,
5
1
=
c
= a>
c
4_)
C
o
o
cn
fa
u
c
4->
l/>
*J
u
-D
o
^
II
0)
re
w Q^
^
a> c:
> o
C f/l
(/i
J2 "-
O} UJ
CT
T3 P
S_ ro
O *J
4J O
UO 1
oo
D.
cn
^
to
01 -
0 =
C !-
0! M- .
T3 C
C *J C
O *-> '^
C. 3 4J
t/1 T3 U
OJ O OJ
^ !_ 1^.
s- c.
O E
U II C
1 >^ U_ *J
1. C. 0
! " *
' t/1 ' OJ
1 3 C i.
' "O O
- 'Z 0
1 *a -*j
| .. 0) ro
! c s- M
C '~
0 j C s.
4J o gj
ro ~
;=">,
> j ro ^~
CMC
| y_ -r- C.
' *- a) a1
' i*. ->,43
! O d
' OJ Q.-I-
i U
^ 1. H 4->
c_ z: c
O fi£ GJ
L^ C*. >
na ^
c
J_J
fO
a
!ti
c
c
1
__
5
I1
L.
-t-J
un
*~
*-
^T
^
c-.
^
i
1)
L^
U
3-27
-------
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.
24
3.2.2.2 LDPE Low-Pressure, Gas Phase Process. Most, and probably
all, new low density polyethylene plants will use the relatively new gas
phase 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, gas phase 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
25
catalyst composition; reactor operating conditions remain the same.
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 gas phase 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 industry.
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).
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
3-28
-------
I/I
a co
c ai
(0 O
Q_
o
a.
en
a>
^r «
4-> OJ
s_
S- 3
O (/I
4- 1/1
0)
C s_
ra o_
i- i
cn 5
re o
O)
-ii C
u a
o i
>-.
03 -C
4->
tO O)
co >,
>
+J
TJ -r-
-------
bed, the gas stream containing the unreacted ethylene and cononomer
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 fron 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) fron 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
approximately 97 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
21
extent, by reactor operating conditions.
3.2.2.2.2 Emissions from the LDPE low-pressure, gas phase process.
The waste gas stream characteristics for the LDPE gas phase process are
shown in Table 3-9. The combined process VOC emission rate for this
process is about 23 kg VOC/Mg product. One process vent, the product
discharge vent, contributes almost 96 percent (22.3 kg VOC/Mg product)
of the total VOC discharged annually from the model plant for this
process.
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
35°C (100 to 185°F), and the pressure varies from 0.7 to 138 kPa (0.1 to
20 psig).
3-30
-------
1
1
1
1
1
1
. 1
!_.!
cc
i
LU
c.
3
_i
UJ
UJ
5;
P !
UJ
>
2
H-
Z
C
§.
UJ UJ
So
5£
as yj
U). gO
en 31
5 °*
UJ V)
<
go
^ [
UJ
^
°
i
:2
2
5
C5 '
o
i
^ i
-C
^
I
[
1
)
1 *
1 £
I o *->
1 3-3
5
t
i
*
u
| ^ £*
oi a.
u
«
s_
3
U U
at o
=
5
^*
' C £ *J
O "%. U
^ * O 3
trt o* O ^
w* +J > O
-*- 'O U,
Jwf=-
4)
3
s
i
1
c
*5
i
t
1 **
* u
£
I
i
-
vi 3
u C
u o ~ --
c ?v o 2: =» «c
> Z u; <-J ^j i. o u w u <-j '^J :» t_;
O csj O ^ o "^ O CM O "^ C (N' O c^J d CNJ *» fsj CO O
°'s ss s" ~s ?s ^s s- ss ss 2g ='§i §
.
O . OO O?J PU-*
i * I O ' O " i i C* C*
i ^ i ^/ n ro co ^i cc i ui ^
CO
i
oco ooooofn^^i |
O O O* O O O O O «SJ ^*-
CM C2 O O ^"
eg
C C C C C C =
0)0)t/f O)OfO)O)(AMO)
*J*JO *J*rf*J*Joo*JOO
III 111III1II
4J.WC ***rf*-*JCC**CC
_c c o ccccooeoc
*rf *J
c c
O) 0)
> >
c c c
_O 01 S
"S ** e "5 *-
_u e o c c
<^>Wt. =Sc> 3J
*- U>O) fl
*^ -C Oi 4< U ifl * C v? C
(/)«*!» >t*^ > Wl « 41 « OJ *J C
^3 C- W V **, S> L. ^ >vi
^ k. 3 >»O-OC
2 >, O >» = 4) 01 01 U 0! -g
£ _ C 2>31U = 0)>>,
*2«-^3Owci^r> , . ^
flflOflo^-flOiHSus r c
^ (J
5 c z
! .. *e 2*
a.
*- 5 "i*
-S Z S
-C
c
J? U
w £
i -2 "r =
C *-
i ~~ c ^'
'! U -C
| U. > T
' C W I/1
! c ** 5
i -5 u ^
S, i ~-
| TS -0
' » §
) -,' ^ ^ ^
;£ ^ >
S '" ' *
' C = M t_
C X >*
ui c _= r "±
u u c >
S - - = i.
: i 1 1 * 1
; -5 - <
' ~~ "~
; c ££=""* =
| -2 *£ r =- 1
; o D j*> x ,
, -*- ** i ^
: 'c 1 = § § "
| u *" ' ^ ^ '/
0 '1 '_ ^ J- 1
3-31
-------
3.2.3 High Density Polyethylene (HOPE)
HOPE resins are linear thermoplastic polymers of ethylene with
3
densities higher than 0.96 g/cm and copolymers of ethylene with densities
3
as low as 0.94 g/cm . 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 HOPE processes are described in this section.
3.2.3.1 HOPE Low-Pressure. Liquid Phase Slurry Process. Of the
two liquid phase processes, the slurry process is predominant and is the
O C.
only one capable of producing the whole range of HDPE polymers. 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
screw conveyors are used to withdraw concentrated slurries containing 50
to 80 percent solids. Unreacted monomer and diluent (3) are separated
from the product by flashing. Final stripping of the gases from the
3-32
-------
OJ
r^
QJ
CD
O
Q_
l/l
ST
CD l/l
r- o;
:z cj
^^
OJ i.
5- S_
O i-
s- o;
CD (/)
Q SI
CJ !-
O 3
r- CT
SQ -r-
OJ CU
u s-
O 3
S_ VI
Q. 1/1
O)
CJ i-
-------
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 low-pressure, 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 13 kg VOC/Mg product.
The total emissions from the four streams is about 14 kg VOC/Mg product.
Table 3-10 shows the composition of these streams.
The emissions from the closed-loop drying system (Stream B) are a
dilute mixture of process solvent in nitrogen. 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.
Host 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.
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
3-34
-------
^
1
So i
UJ *
0(0
0"
i3 uj
= 0 \
~"~ c:
UJ O_
~*~
( >-
ee.
|§
OO UJ
i* 2:
o: 5!
00 Q
r l |
2: o-
LU .
=* _;
U_ "
O LU
on ^
I LU
t/> ce
Cu
LU 3
( O
o __:
< t
o: LU
;r LU
C-J _1
O LU
Ag
0)
r
^
h
i
1
f
1
1
I
!
1
I
1
1
1
i
o
1 ^ ^
f ..- ' .
'/) -M
0 3
c.
! ^
! "
i
!
I
1
O)
' -3
T3
i- <_;
OI 0
o.
OJ
i
i
c £' *J
o \ u
^ o;o rj
in j->5> o
r- to 1.
= s- cn o.
UJ ^
01
i-
^
T3
^
-2
-c
5
V
£.
t-1 '
LO 1
1
u
w c
i/7 O
D .^
O '->
u
1
*J
c
Cl
^
+J
o
o
c
rt
00
CM
O
o>
c
c
13
C
c
o
,
u
a
CQ
.2.
~
CJ
o
4_,
r-
^
CT>
S-
01
o
-a
i! C!
IT C ^U C Z!
^ ^n i) -J 01 ^i
3 O r - C O
O -^ -CT C JC "3
O «T COCO
o a-. co o
CM CXI
m
O r-v c-"
r^ ^^
CM 1
f*S
Ul 1/1
3 3
0 O
3 3
C C
c c
o o
.£
t_ Ul CC
01 1.
X 01 C
j; -M O
~ 01 i/i
o '=
3 01 UJ
C
M >i 1o
C U -M
O Ol O
O Oi 1
^ ^
CU
tj
01
T3
C
a
t/i
n
s_
c
U
>>
L
M
jg
-
g
2
C
0
^
U
^
p
13 -
C
s
c
=
3
'w
"°
01
cyl
11
2.
~
1J
U
yi
^
"3
a
^*
i_
Ol
o
U
Ol
i_
tl
01
~
3-35
-------
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 Low-Pressure, 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
27
upon the solution temperature of the polymer in the solvent. 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. ' 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 HPa (1,000 psig). The exact operating
conditions depend in part on the particular polymer being produced.
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
3-36
-------
a
33
LJ
UJ
a: o
UJ
1 CC.
>- UO
O
OO
C oo
OJ co
CU OJ
U
-C O
en i.
r- 0.
03
-LJ
ai
o
4-)
01
OJ
-i-)
«3
2
T3
O C
5 O)
fO oo
S- T3
cn^:
-------
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 low-pressure, 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 19 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 60 percent of the
emissions from those HOPE plants which do not have an adjacent ethylene
plant. Table 3-11 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 HOPE Low-Pressure. Gas Phase Process. The HOPE gas phase
process is similar to the LDPE gas phase process described earlier
3-38
-------
!
i
I
'
i
i
L- 1
s 1
^ '
>- '
>- '
>- 1
~
^1 '
'_ro '
d u~ '
"n
r; I
c !
= =2 !
'
^£ 1
s* <
±£ _1 '
'-* C '
-° 1
I*
2 - !
^- c£ ]
^
z ^ \
> ^
= !
il
S ^
.
OJ
^
=
a
i.
u o
0.0
55
>
7*
^ ^ W
0 -v. U
«-_; 3
^ ^ 5 ^
-a w
= w =n 0.
LbJ ^^
11
'-
|
5
*"
1
a;
^i
j^
.rt O
-J -
_' *>
fc U
C. wO
'
1
fvj
w s
C^J rfl
c u
u
O ,
u
4_(
trt a;
i "
3
c =
1_
^ 01
1 =
f>
^
yi U
3; 1-
> u
C 3J
0 L.
^ (~
< aa
S 2
1
i
i*j
S
o»
u
t/i
i>
u
/i
1/1
1
3
C
W
C
o
(J
>»
iT)
s_
s
1
c
o
4-> J_l
^1 -fl
** 4) C
« 1- i)
o a. >
t_j
u
T3-
S S W 4J £
>, C J ^ ~
^ o .= c
> -3
! 1 1 OJ >)« 4J 1 '
C "*~ O O 01
- O '
> . cc c: cr
^O ^- -- Og >
y in
t O
1 1 1 CSJ | «
r^j ix* r**( »NJ
^ 1 « ( % (n
u u it . u ^o
i_ t. u ?M u. .
w *J *J « MO
I/I (/) W1 W1
OJ O( 4) 0>
OJ O) OJ OJ
ui (/) i/l 1/1
4->
i_ oS
« *J -0
*j *J C
C tj
£ *
C OJ 3 "3
c *-» o c
fl C 3 *> ifl J-»
* c c e
t/> v 4J k/> i/» OJ VI
3 >*J*J 3 3 *J S
o , e ** o o-*-»o
3-W r-O 3 3-^- 3
c= ^»fjj: c cs c
^ X U
u» cs > LO> a; w -w oc -3 *j LU>
y
I £ I £ = s:
cc
o
U
^j
^ -g
OJ Q£
i.
^ ^
.-5
^i :
u -a
VI '
S^
>
s- -a
0)
S . S
J VI
u v -r
s_ 4- u
_ y t
o C
T ~~ s
w u1
a i.
33
II VI C
ii
a- *
r: u. >,
i w
0 = ""
r c
'J **- vl
-O '_
£ S |J
5 31 5 =
Si i"g
-e ^; i.
U «s a;
u it S \
^ = 3J
= >^ -c i-o
i. 1 = |1
U C -I
S- '!
1- i ~J »
^1 O. 3 fl ^ -^
V «j
>, - vi X -
'- r * * r
vi 4J i- 5 J:
3 *- - S. X -3
t3 -o -o u
= i-
2 a u » c i i,
!- « "3l I) ^
o a = - -o -
= t S^.s^i-^
= 1' « 3 =
w >. c ^J y -3
"c 1 =^3^5 ^-
3-39
-------
(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, gas
phase process, some process equipment can produce both HOPE and LDPE.
Thus, for the purposes of this project, the process and emission
descriptions for the LDPE low-pressure, gas phase process serve also for
production of HOPE 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 rubber modified 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.)
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
3-40
-------
30
the most widely used process for polystyrene today. The suspension
process is also commonly used, especially in the production of expandable
30
beads. The use of the emulsion process for producing homopolymer of
styrene has decreased significantly since the mid-1940's. This section
describes both the bulk (mass) batch and continuous processes.
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
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).
3-41
-------
l/l
l/l
o;
u
o
O
-P
s-
o
a
o
ca
o
0.
-a
01
a.
oo
CO
I
ro
OJ
3
3-42
-------
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 0.64 to 2.5 kg
VOC/Mg product. The major vent is the devolatilizer condenser vent
(Stream B). 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 devolatil izer,
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 C) are 0.09 kg VOC/Hg 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 C)
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 D). 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.
3.2.4.2 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
3-43
-------
V
oo
t/1
UJ
0
o
Qi
Q.
z
CJ
I
<:
i
UJ
0£
>-
[
00
>-
_l
o
a.
; i i
P
^
O
as
u.
c/l
i
ua
OS
»
u_ '
o
t
CJ
K
C/1
5
UJ
CHARACT
CM
1
n
OJ
3
/~
i
j
1
i
i
I
t
i
!
.
g
1 =
j '«3
i.
i
i °
i
1
i
1
i
f
OJ
s-
3 cn
Wl-r-
tO
a) a.
u
i
I
s-
1 1
1 *-<-'
0)0
o.
£
a
en
C °" 4J
O ~ O
- « O
.- flj S-
= s- en a.
UJ -^
i
t
01
^
3
fa
z:
i
1
u
£
«
s:
I
1
"_
5
00
i
i
i
1
I
.O
'.O «
t/l O
0)
U -w |
£ 3 !
0. 00 !
t
_
$
X1
OJ 0) ti
C C S-
OJ O) OJ 0) cC
S_ S_4-> t. =
>1 >» «T3 't -M 13 >-r- OJ
O ^^ *-> «r L
0 a* c^ >L
O »ur co r--, 4j
t UD on UD ro 1> 4-> 4J
C 4J C C
0 C 00
O CJ t_)
3 vi
*-» U 0 -r- C- «
VI S_ > 4-> r t. ._.
OJ C U i. =
^ > S- flj +-> C 4JC 4J
oaj-^a;i- cuo xaj o
S>+--oo£-a au uj> i
1
< CC C_J O !
'
1
ci u_ '
a r: =. ;
i
I
|
1
t
1
1
1
1
I
i
i
i
)
i
i
i
i
i
i
i
i
j
t
i
i
i
1
1
i
1
i
i
i
I
1
i
i
1
!
|
1
1
(
1
!
c
o
4«>
U
fO
^
sj
- -M
S- 0
0) 0)
5, 0
i S-
o c.
a.
II -r-
.c
o: 4j
a.
c
V ^1
U C -c
C "- CU
o o
a. c c o
vi O
qj <+- -^ 4->
S- 4-> 0
I. +-> ra c
O O U
J 3 .- !L
0 <4- S-
>, O a. c vi
vi aj c
3 II "C O
-3 .f-
Z U- VI
k " fl_ = VI
^ -f-
- ^ ? s i
C -f-» ^
c CJ '/i cu T;
^- > en
U Q 4. T3 s_
^ '0 O i. OJ
= 01 U_ 0 CL
i- S- *J
O CO vn U
;*- . l 'j
c T3 ro ^- c
-f- <- -a aj
i- OJ -r- 1_
^- a) '-. i. s»
0 ^> 3 U 3
ts en w u
3J S ;- 2 u
in ~"
3 0) 3 OJ
s a; ^T oo :r c
*W J2 C_J T3 CJ
t
S-
-a
OJ
>^
^
i_
2J
>
1)
,
CJ
>
<<
4
3-44
-------
GJ
O
O
in
3
O
C
O
O)
C
O)
O
Q.
O)
.C
i-
o
(O
S-
cn
O
O
I/I
O)
O
p
o>
I
ro
CD
3
CD
3-45
-------
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, ethyl benzene (an impurity from the styrene feed) and
low polymers, is pumped to a vacuum devolatilizer (3). In the devolatilizer,
most of the monomer, ethyl benzene, 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
(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. It is estimated that the typical
total VOC emission rate is about 3.25 kg VOC/Mg product. Table 3-13
presents the VOC vent stream characteristics.
3-46
-------
o
LO
LO
LU
O
O
an
o.
LO
3
3
1
O
O
LU
LU
C£.
LO
1
O
Q.
LU
a:
i
§
i.
on i
ZL" '
LU
on
LU
u_
O
«
^*
14
LU
C_?
*C 1
<£
;_j
n
1
r"1 1
n> '
;
1
1
i
i
1
i
i
!
i
»
t
Q
3**.
O ~~
c.
1 ^
i o
i
i
i-
3 CT>
t/i »-
to in
i- "~
D_
n
i-
^j
S- 0
Ojo
C.
5
c ^
O "-^4J
i O (J
VI "O 3
tn 0) > T3
-M 0
^ (Q CT i_
LU l- j«: CL
i
Qj
s»
13
i
t
f
1
1
' QJ
t j-»
'
'
i
]
i
j
<_?___
' S
' «
4^
LO
I
f
yi0""
j- O
L? --'
'- i
r~ i/7
OJ
C
CL
-5 OJ
fD OJ OJ _ C
O~" ajc Qjc; 5l-
cj5s_ s-T3 s_fO o; >»
C'>,aj >> aj >,aj AJ4J
!_, _c t-'*-' +J4-) on on
>^ o -M uo on on co
4J a. o a^ e
(jO CO CSJ i O O*v O
LOLO «
aj c c a;
T3 3 OJ 4-»
o >>+^ a:
C L. C JZ
i. 0) > C O
> D O GJ *i»
f IM 0^-3 ^1
t/1 i t_ L." 'I
y» ^- -T S- =
t- 4-> a; c1 a? LJ
C3 (O C "O "CJ
i c; c 3 r
-a o JJ s- o s- ft?
OJ 0) OJ »-> X o
U_ O > (/I 1 i 1
^
CJ
2
u
. c
S § -2
t. II fD
O U
'_) C^ »
u* -* -Ij
4-> C C
i/l C Qj
*C *J -f-
^ ^
CL <£;
OJ S.
O C. ui
n cr: on
3-47
-------
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.
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. When vacuum pumps are used, emissions as well as energy
consumption are lower than with steam ejectors.
3.2.5 Polyester Resin
Polyester resins used for fiber production may be classified chemically
as either poly(ethylene terephlate [PET] resins or non-PET resins. Of
these two types, PET resins are the most important polyester as a
thermoplastic synthetic fiber.
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
bis-(2-hydroxyethyl)-terephthalate (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. 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
3-48
-------
in the continuous process at the polynerizer, and (3) different additives
32
and catalysts are used to assure proper product characteristics.
3.2.5.1.1 Process description. A schematic diagram of the continuous
DMT process is presented in Figure 3-10. The primary reaction is:
CH3OOC -O COOCH3 + HOCH2CH2OH-*-HO - (OC -£>- COOCH2CH20) H
n
DfTT EG PET
Dimethyl terephthalate and ethylene 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 terephthalate (BHET) and
nethanol . 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 ethylene glycol is flashed from the polymer product,
condensed, and, if desired, can be purified and recycled.
3.2.5.1.1 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 3-10. A summary of the vent stream characteristics is provided
in Table 3-14. 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 stream from the reactor is 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 where the ethylene glycol is recovered. 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
3-49
-------
i in
an-
s:
_i
o
25
IIS
O oe
Z LU
S 3« <
^Z ^3 ^
h- CJ
to
V)
Q.
o
I
ro
a>
s_
CT)
3-50
-------
I ^
>> o
**- C W Ti
" ffl a; ~
o S *- >,^-
*J H = O
^ z: oj o o
^ ** E
'O C >
3-51
-------
rates. Alternatively, a plant nay 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 3-11. The primary reaction is:
HOOC
COOH + HOCH.CH.OH *» HO - (OC -<~~VCOOCH0CH00) H + 2nH.O
C. C. IL C. U t.
TPA EG PET
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.
3.2.5.4 Emissions from the PET/TPA process. There are two sources
of VOC emissions from this process as illustrated by Figure 3-11. These
continuous streams are emitted from the esterifier (Stream A), and from
the polymerizer (Stream B). Offgas characteristics for these streams
are summarized in Table 3-15.
As with the DMT process, the overall emission rate from this process
depends on the type of system used to recover the ethylene glycol. If
the ethylene glycol is recovered from the cooling water, emissions occur
from the cooling tower. In this system, offgases from the estarifiers
are sent through a distillation column to recover the ethylene glycol.
3-52
-------
<
c
«
I
s
z
o
1
a:
UJ
K-
t/1
LU
^
Z
t/1 ^D
< t
I a; o:
£ UJ <£
1 £.
< UJ
S. SC.
Q.
U
I LU
< 0
* *^* ^?
I OJO
: t h
k J
2
^
^-
l/J
V)
3-53
-------
t
1
I
oo
01
LU
O
O
CL.
r\
*~
* ^
LU
t
I
^£
p
e.
LU
S
z i
UJ |
1
>.
^
UJ
_J
s
LU
31
H-
»-
O
o:
U-
S
UJ
C£
|
CO
V
UJ
>
u_
0
'-O
~~
(
CO
o:
UJ
t
«C
Q£
Z
LT>
,-_!
co
_-
j=
TO
I j
t
i
i
1
c
o
1 ! C .
VI -M
C IS
s~
s
1
1
a-
3 en
CO -
VI (/I
£
O C_J
CLO
c
0)
en
c n
o *--. +>
t t^ u
co "O 3
vi OJ > -O
^- 4-J 0
' = «» en i_
UJ i- J* CL
OJ
i_
3
T3
y
vi
a- c
c a;
i. S- OJ -O
O OJ f C
MS >, 0
U 0 j? U
<0 h- 4-»
O (U >i
o: t- w
OJ *J S_
S- 4J C CU
OJ fO QJ VI
N 3 O.
i- f
i- en o
i-5 5 >>
F 'o 5 *CT
0 0
C- CJ 1
C2
UO OJ Lrt
en o o
!.
01
1- C
cu i- a; a;
CO) in T3
oj wi c: c
i c (DO
>> 0) "CO
jz -a a) c
+> C *J O >,
0) o ITS o re
u c: t-
C ^i C
S 2
O
i 1 l i
a>
u
u
T3
C
O
O.
01
i-
-
0
u
£>
U1
3
C
C
0
4J
<0
^
c
c
l^
c
a;
u
i-
5
(D
U
.^
c
*~
E
OJ
c
'*-
^-4
,-H
OJ
^
en
O!
a;
u
.
w«
c
3
O
C
VI
QJ
I/I
C
o
T3
c
0
u
TO*
^
CL
VI
aj
JC
"
0>
^
QJ
4-)
s^
4-)
3
c1
o
0
o
s.
0
re
i.
o
4-}
C
CJ
i/i
in
S
£
c:
O
c/l
Ul
c:
5
r
C
OJ
J-J
c
o
p
o
.
t-
2
c
4-J
J_
OJ
en
r
O
o
5
i_
u-
^
u
3
o
1_
d.
_.
^~
^^
O
en
^
.
o
"2
re
c
i
.,_
3
O"
O)
u
OJ
o
u
OJ
§
U-
J-J
u
"O
o
o.
2^
'O
o
en
^
OJ
^2
O
C
CL
QJ
CO
cc
a;
«=:
OJ
t.
co
cr
5
V)
c
o
V)
VI
2;
r^
ro
4-J
c
-------
The overheads from this distillation column are fed to the cooling
tower. Under this setup, there is only one major emission sourcethe
cooling tower. The ethylene glycol emission rate has been calculated to
be about 9.5 kg VOC/Mg of product. Alternatively, where a spent ethylene
glycol spray condenser is used to recover the ethylene glycol directly
from the reactors, reflux condensers may be used on the esterifiers.
This set-up results in emissions of about 0.04 kg VOC/Mg of product from
the esterifiers and about 0.21 kg VOC/Mg of product from the reactors.
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 petroleum refining. Sources include valves, pump
seals, compressor seals, safety or relief valves, flanges, sampling
connections, and open-ended lines. Fugitive emission sources are
extensively described in References 33 through 37.
Table 3-16 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." For the
purposes of this project, only those organic compounds considered as
"light liquids" under this classification scheme are of concern with
regard to fugitive emissions.
Data characterizing the uncontrolled levels of fugitive emissions
in 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 the petroleum
refining industry. 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, the
SOCMI fugitive emission data can be used to approximate the levels of
fugitive emissions in the polymers and resins industry. The final SOCMI
data, which are used for the purposes of this project, are presented in
3-55
-------
Table 3-16. VAPOR PRESSURES OF MAJOR ORGANIC COMPOUNDS USED
IN THE POLYMERS AND RESINS SEGMENTS CHOSEN FOR NSPS DEVELOPMENT
Organic Compound Vapor Pressure Process
Ethylene >38,000 mm Hg @ 8.9°C PP, PE
Ethane >30,000 mm Hg @ 23.6°C PE
Propylene 7,600 mm Hg @ 19.8°C PP, PE
Propane 7,600 mm Hg @ 26.9°C PP
Iso-butane 1,520 nm Hg 0 7.5°C PE
Butene 1,410 mm Hg @ 21°C PE
Hexane 100 mm Hg @ 15.8°C PP
Methanol 60 mm Hg @ 12.1°C PET (DMT)
Isopropyl Alcohol 20 mm Hg @ 12.7°C PP, PE
Butylene 5 mm Hg @ 11.6°C PP
Ethylbenzene 5 mm Hg @ 13.9°C PS
Styrene 5 mm Hg @ 18°C PS
Light/Heavy Liquid Break 2.25 mm Hg @ 20°C
Ethylene glycol 0.05 mm Hg @ 20°C PET
Key: PP = Polypropylene
PE = Polyethylene
PS = Polystyrene
PET = Polyethylene terephthalate
DMT = Dimethyl terephthalate
Source: Perry, Robert H. and Cecil H. Clinton, Chemical Engineers
Handbook. 5th Edition. McGraw Hill. New York, 1973.
3-56
-------
Table 3-17. (See References 36 and 37 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
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 liquid phase, 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
3-57
-------
Table 3-17. UNCONTROLLED FUGITIVE EMISSION RATES
Fugitive Uncontrolled emission
emission source rate, kg/hr/source
Valves
Gas . 0.0056
Light liquicT 0.0070
Heavy liquid0 0.00023
Pump seals
Light liquidb 0.0494
Heavy liquid 0.0214
Compressor seals 0.228
Safety or relief valves
Gas 0.1040
Flanges 0.00083
Sampling connections 0.015
Open-ended lines 0.0017
aThese uncontrolled emission levels are based on the
data presented in Reference 35.
Light liquid is defined as a petroleum liquid with a
vapor pressure greater than that of kerosene.
GHeavy liquid is defined as a petroleum liquid with a
vapor pressure equal to or less than that of kerosene.
3-58
-------
emergency vent gas, and this system is used on tubular reactors. Controls
are not routinely applied to the dryer and bin storage vents.
Gas phase polyethylene plants (based upon Union Carbide's Unipol
process), currently flare both the continuous and intermittent emission
38
streams. The flare system has a liquid seal and is air assisted.
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 other than condensation operations in
which styrene is recovered, due to its value and ease of recovery.
Offgas from the styrene condenser and other vents are usually vented
directly to the atmosphere. Flares are not usually installed in these
plants.
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)
3-59
-------
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)(i)].
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
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],
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.
3-60
-------
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.
New Jersey law regulating VOC emissions uses a "sliding scale"
to determine allowable emissions. Vapor pressure and concentration
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.
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
3-61
-------
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
a detailed description of fugitive emission control practices that may
be used, see References 34 and 35.
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 regulations
may affect a company's attitude about leaks and fugitive VOC emissions.
3-62
-------
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.
3-63
-------
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-Kailogg Company. EPA Contract No. 68-02-2619. 1979.
2. The Society of the Plastics Industry, Inc. The Story of the Plastics
Industry. 1977. p. 31-32.
3. The Society of the Plastics Industry, Inc. Facts and Figures of
the Plastics Industry. New York, New York, 1978. p. 63.
4. Kurtz, S.J., L.S. Scarola, and J.C. Miller. Convert LDPE Film
Lines for LLDPE Extrusion. Plastics Engineering. June 1982.
p. 45.
5. Reference 2. p. 29.
6. Reference 2. p. 29.
7. Reference 3. p. 59.
8. Modern Plastics Encyclopedia, 1981-1982. McGraw-Hill Inc. p. 66.
9. Reference 3. p. 55.
10. Reference 2. p. 32.
11. Reference 3. p. 65.
12. Bhatia, Jeet and Rossi, R.A. Pyrolysis Process Converts Waste
PC
P.
13. Reference 1, p. 179.
Polymers to Fuel Oils. Chemical Engineering. October 4, 1982.
n. 58.
14. Cipriani, Cipriano and C.A. Trischman, Jr. El Paso Polyolefins Co.
Chemical Engineering. April 20, 1981. p. 80-81.
15. Albright, L.F. Chapter 3. In: Processes for Major Addition - Type
Plastics and Their Monomers. McGraw Hill Book Company. 1974.
16. Brydson, J.A. Polyolefins Other Than Polyethylene, and Diene
Rubbers. In: Plastics Materials. ILIFFE Books, London, 1970.
p. 134.
17. Trip Report to the Texas Air Control Board, September 20-21, 1982,
from Ken Meardon to Polymers and Resins NSPS file.
18. Reference 8, p. 58.
3-64
-------
19. Sitting, Marshall. Polyolefin Production ProcessesLatest
Developments. Noyes Data Corporation. 1976. p. 48.
20. A Leap Ahead in Polyethylene Technology. Chemical and Engineering
News. December 5, 1977. p. 22.
21. New Route to Low-Density Polyethylene. Chemical Engineering.
December 3, 1979. p. 82.
22. Albright, L.F. High Pressure Processes For Polymerizing Ethylene.
Chemical Engineering. December 19, 1966.
23. Reference 15. p. 92 and 96.
24. Reference 21, p. 80-82.
25. Reference 21, p. 83.
26. Pasche, E. The Outlook for High Density Polyethylene. CEP.
January 1980. p. 74.
27. Reference 19, p. 236-238.
28. Reference 19, p. 239-243.
29. Trip Report to DuPont's Sabine River Works Plant. July 1, 1982.
30. Reference 8, p. 90.
31. Reference 14, p. 347.
32. Reference 1, p. 123.
33. Wetherhold, R.G., 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.
34. 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.
35, 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. January 1982.
36. U.S. Environmental Protection Agency. Fugitive Emission Sources of
Organic CompoundsAdditional Information on Emissions, Emission
Reductions, and Costs. Research Triangle Park. EPA Publication
No. EPA-450/3-82-010. April 1982.
3-65
-------
37. U.S. Environmental Protection Agency. VOC Fugitive Emissions in
Synthetic Organic Chemicals Manufacturing IndustryBackground
Information for Promulgated Standards. Preliminary Draft. Research
Triangle Park, N.C, EPA Publication No. EPA-45Q/3-80-033b. June 1982,
38. Dedeke, W.C. Letter to Messrs. J.C. Berry and E.J. Vincent, U.S.
Environmental Protection Agency. November 11, 1982.
3-66
-------
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 resins
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 or 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
2
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. 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
pulls down into the stack.
4-3
-------
Sleara
Nozzles
("0
Gas Collection Heaoa
and Tians/ei Liu
Emission
Souice
Gas
i
y
OisefltiaiaMflt
Qma
(2)
Ignition
Device
Air Line
Gas Line
0(310
Figure 4-1. Steam Assisted Elevated Flare System
4-4
-------
PILOT
ASSEMBLY
STEAM
HEADER
PILOT AND
MIXER
STEAM'
DISTRIBUTION |
RING
TIP SHELL
INTERNAL
STEAM
INJECTOR
TUBES
CENTER STEAM
JET
CONTINUOUS
MUFFLER
CENTER STEAM
JET
PLAN
ELEVATION
Figure 4-2. Steam Injection Flare Tip
4-5
-------
Purge gas (N,,, C02, or natural gas) also helps to prevent flashback
in the flare stack caused by low offgas flow. The total volunetric 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 smoke!essly
without any steam, typically 0.15 to 0.5 kg of steam per kg of 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 (100 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).
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 flanmability
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, H^ 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 smoke!essly.
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. 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. 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 H?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 10°
J/sec (235 x 10 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.
o
Lee and Whipple (1981) studied a bench-scale propane flare. 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 10 J/sec (0.3 x 105 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 ^as located off the
center!ine 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 link's
flare test facility. 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 !b/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 deternined 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, and a fifth study [McDaniel , et al.
(1982)] deternined the influence on flare performance of mixing, heat
content, and gas flow velocity. 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 stean) 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 10 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 10° J/scm (168 Btu/scf) gave
55 percent efficiency.
4-10
-------
1 1
i ,
0
c
I 1 'o "
!l s: g
i ^ 3
-U £
' 3J
! ! P =-
i! .2
I ^
ji
j
i « "sT
3 J=
OL ---
B 3
3 2Q
0
S- US
f O
"~ Ct
f
oo I ' en
UJ 13
CO
0
= -3
d 0)
s>
UJ
UJ
o; c
eC Ol
_i *
U. Ift
1 ^
1
0 CO f^-
O CTv
0 CT,
*l
00
r-. y3 r^.
o^ o cr.
r-t CO
r*.
CM ^H
I 1
r?
*± o
a *r
c
(/i
+-> 0
-c .a
OJ CMC! U
C 0) = 03
QJ C i cj
f fO >9 O
>> a. O in '.
jz o u~> 3 -a
^j i p*" >,
UJ Q_ Q. JT
=
CM
c c c
«- CT fa
-f- CJ
C/) VI 4-J
U CD OJ !^
s_ *- -o
i- 4-J 0) U
-a a; 4j s- =
i- a; a; ys
3 (J S *
Lf! t/1 fl £ 1 .
. -r- .r- O CM
o a -o LJ '
1
5
Is
-j ^
OJ -4-J **-
4_) -o ^- O
C -f- O)
O J3 trt >,
G_ i. yi +*
ra -r
3 CJ Q i/l
"C i-
.
C JZ C
UJ IZJ 0-13
CO
en
""* CU ^-~v
CM O
r Q. ca
cr\ ex en
^ ^ , ,
3
a; oc 'cj
3 CD
i O) CJ
<0 OJ -f-
a. __i oo
en ^
en
0
O Cl c^
I 0
o-.
1 1 O
O U3 1
OJ O CT> m
C3> OT L.O CO
'S
«"' CO r^-
*J uo m
- i i
ex ro cr»
- o o
C 0
«3- GD .
^r CM o o
t/i
^
0 J-> 0
£u 2r a.
S- 4-J
* > c/i
4_) . * * -r-
C1- CfG Cc/) C(/l
CJ) VI CT5 *» CT *" d ^O
.,_ ^ -r- 13 *- ^l -r-
VI (O yl i/l Ul (/I =
ty a; = cu
^-fl r **- r -i- f t/1
a « o »o « *o
CJ-O U tO O- CJfO
L. 5- C_ S- fO 1- -r-
OJ T3 OI -T- OJ CJ T3
= = 1^ s ? 1 =
Ovo om o(^3 o^
CO CJ" CJ ' CJ^ ''
1
1
i
1
1
1
1
1
1
1
1
]
c '
r- -W J
TS O3 ]
4^ , ]
Oj CU |
-0 r CNJ I
O) IS CO
2 d C^
.9 o
4-11
-------
After consideration of the results of these five 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 10 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. Air assisted flares obtain 98 percent
efficiency with heat contents over 11.2 x 10 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.
The EPA has a program underway to determine more exactly the
efficiencies of flares used in the petroleum/SOCMI industries and a
flare test facility has been constructed. The combustion efficiency of
four flares (3.3 to 30.5 cm dia.) will be determined and the effect on
efficiency of flare operating parameters, weather factors, and fuel
composition will be established. The efficiency of larger flares will
be estimated by scaling. A final report of this work should be available
in the summer of 1983.
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. 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
14
may be necessary, and these gases may be used as a fuel gas or boiler
i r-
feed gas. ° 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 4-2. (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.
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 4-3. 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.
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
1R
remove the corrosive gases from the exhaust.
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
19
combustion chamber. Variations in inlet concentration also affect the
4-15
-------
Waste Gas
Auxiliary
ei 3urner
discrete)
(1)
^omoustion Air
Mixing
Section
(4)
Combustion
Section
(5)
Ootionat
Heat
Recovery
(6)
Figure 4-2. Discrete Burner Thermal Incinerator
iVaste Gas
U) (1)
Burner Plate-. Flame Jets-r
Stacx
Gotional
Heat
Recovery
(4)
(natural gas)
Auxiliary Fuel
Figure 4-3. 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 tine 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
20 21
and can be met under a variety of operating conditions: ' 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 Cg 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
22
achievable by all new incinerators considering current technology.
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
9 ^
observed in a polypropylene plant ). 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 NO because of lower combustion
X
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
74.
from the support substrate. 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 4-4.
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
pc 07
concentration, and temperature in the unit. '
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
1 ? ?
than 1 year up to 10 years, with an average life between 3 and 5 years.
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
29
the data points gave indication of a maximum catalyst life.
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
-------
s_
o
-*->
0)
o
(O
o
3
o>
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 nay 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 lim'ted.
4.1.1.4 Industrial Boilers. Fireboxes of boilers and fired 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
30
been injected. 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. Boilers typically operate at
combustion chamber temperatures above 1,650°C (3,000°F) with a residence
time of about 1 second.
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. For vent streams with heating values between
1,100 kJ/scm (300 Btu/scf) and 1,850 kJ/scm (500 Btu/scf), a discrete
burner would be best suited. Streams with lower heating values would
probably require vortex burners to ensure the desired VOC destruction.
4.1.1.4.1 VOC destruction efficiency. VOC destruction efficiency
achievable by boilers depends on the same factors that affect any combustion
technique. Since boiler furnaces typically operate at higher peak
temperatures and with longer combustion residence times than thermal
incinerators, the VOC destruction efficiency usually would be expected
to match or exceed the 98 percent efficiency demonstrated in incinerators.
4.1.1.4.2 Applicability. Use of a boiler for VOC emission control
in the polymers and resins industry is uncommon. Despite the potential
32
problems, boilers are being used in at least two polypropylene plants
33
and a high-density polyethylene plant. The polypropylene plants
4-22
-------
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.
4-23
-------
Adsorbers are often neither suitable nor the most efficient means of
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 or
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
4-24
-------
can be calculated from thermodynamics data, if it is assumed that the
vapor and liquid condensate are in equilibrium. Variations in composition
and temperature will affect most of the physical and transport properties
which must be used in condenser design calculations. .When 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
34
treatment problem, unless it is economically feasible to separate the
two in a subsequent process.
Surface, or indirect, condensers are usually common shell-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
35
forms a film on the cool tube and drains to storage. The shell-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
4-25
-------
41.5 MPa (6,000 psig), and for temperatures in the cryogenic range up to
oc
approximately 1,100°C (2,000°F). Surface condensers usually require
more auxiliary equipment for operation (such as a cooling tower or a
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."37
The major pieces of equipment used in a typical refrigerated surface
OQ
condenser system are shown in Figure 4-5. 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 she!1-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).39 Temperatures as low as -62°C (-80°F) may be necessary
34
to condense some VOC streams.
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/rig 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
4-26
-------
o en
10
-a i-
ai o
L. J->
a) tn
>
o o
O l-
ai
or
LT>
I/)
a.
z:
o _ i;
l/» O UJ
3 1 s
s
^
O
u/
v»
z
IU
o
zu
o *
Z
(U
oo
c
o
s=
O)
-o
c
o
LO
I
N-
8IU
a:
CD
»~~t
o;
3
z
UJ
a
ci
4-27
-------
refrigerated condenser at an increased cost. Outlet concentrations for
low boiling organics may be above 10,000 pprnv to 20,000 ppmv.
4.1.2.1.2 Applicability of condensers. Water-cooled condensers
are effective in reducing potential emissions of high boiling, easily
condensable organics, and find broad application in the polystyrene
manufacturing segment. Surface condensers are used to recover styrene
from polystyrene vents and nethanol 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 4-6 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
4-28
-------
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 then
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. Removal efficiencies of 95 to 99 percent are achieved by
42
well-designed and well-operated units.
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-
40
trations of organics are usually limited to about 0.5 to 1 percent.
The maximum practical inlet concentration is about 1 percent, or
43
10,000 ppmv. 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 polynerizeable 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)
4-29
-------
VOC-Uaw
Vent Stream
FAN
OMn
CiOMd
Low-pressure
Steam
(3)
AOSORBES 1
> ADSORBING)
AQSOR8ER 2
(3EGSERATING;
Z'.atta
Oo«n
iil
VEHT TO
A7HOSPHEHE
( CONOEJBEH j
DECANTER
ana/or
01STIUJNG TOWES
Recovered
Solvent
Hatet
(5)
Figure 4-6. Two Stage Regenerative Adsorption System
4-30
-------
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
44
constituents of the solvent.
For any given solvent, solute, and set of operating conditions,
there exists a theoretical equilibrium ratio of solute concentration in
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. 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
46
and sodium hydroxide.
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
47
high-solubility gases. 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
48
the packing.
A schematic of a packed tower is shown in Figure 4-7. 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.
4-31
-------
GLEANED GAS OUT
^" To Final Csntroi Cevica
ABSORBING (3 ]
LIQUID IN
VQC LAOBI
GAS IN
(4)
ABSORBING LIQUID
WITH VOC OUT
To OtSQOsai or VOC/Solvent Recovery
Figure 4-7. Packed Tower for Gas Absorption
4-32
-------
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.
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
49
through a stripper which was 98 percent efficient.
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 usually not considered when the VOC
concentration is below 200-300 ppmv.
The use of absorbers is generally limited 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. Absorbers 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 applied 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 fugitive emission sources are located and repaired
at specified intervals. The second is a preventive approach whereby
fugitive emissions never materialize because of 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
4-33
-------
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
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.51
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 seal around the packing
material. 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
52
before a scheduled shutdown. 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.
In order to implement a monitoring program, an instrument meter
reading (VOC concentration) which will be presumed to indicate an equipment
leak must be defined. The meter reading selected may vary from 1,000 ppriv
4-34
-------
53
to 100,000 ppmv. 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
meter reading results in larger potential emission reductions. However,
the difficulty of repairing relatively minor leaks may dictate that
these leaks be excluded from the leak definition.
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 emissions for control.
This approach is referred to as a preventive program. For example, in
many cases, leakage 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
54
thus do not leak during normal operation. A barrier fluid can be
circulated between the mechanical seals. Degassing vents in the barrier
fluid system allow the transport of emissions 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
4-35
-------
seals with barrier fluid systems or with liquid film seals. As with
pumps, the degassing vents for the seal fluid must discharge into a
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 and venting
the void to a collection system.
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 double block and bleed valves (that vent to closed
systems) are devices that can reduce 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.
4-36
-------
4.3 REFERENCES FOR CHAPTER 4
1. Lee, K.C., H.J. Oahnes, and O.C. Hacauley. Thermal Oxidation
Kinetics of Selected Organic Compounds. Journal of the Air
Pollution Control Association. 29:749-751. July 1979. p. 750.
2. Perry, R.H. and C.H. C'nilton. Chemical Engineers' Handbook, Fifth
Edition. McGraw-Hil1 Book Company. 1973. p. 9-18.
3. Kalcevic, V. Emission Control Options for the Syntheti; Organic
Chemicals Manufacturing Industry, 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.
4. Klett, M.G. and J.B. Galeski. Flare Systems Study. Lockheed
Missiles and Space Company. NTIS Report PB-251664. Publication
No. 600/2-76-079. March 1976.
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. .
6. Palmer, P.A. A Tracer Technique for Determining Efficiency of an
Elevated Flare. E.I. duPont de Nemours and Company. Wilmington,
DE. 1972.
7. Siege! , 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.
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.)
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.
10. Reference 5. p .
11. McDaniel, M. Flare Efficiency Study, Volume I. Engineering-Science.
Austin, Texas. Prepared for Chemical Manufacturers Association,
Washington, D.C. Draft 2, January 1983.
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.
4-37
-------
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.
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.
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. p. 1-2 and 1-4.
17. North American Manufacturing Company. North American Combustion
Handbook. Cleveland, North American Mfg. Company. 1978. p. 264.
18. Reference 16, p. 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.
20. 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.
21. Mascone, 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.
22. Reference 20, p. 1.
23. EEA, Incorporated. Trip Report to ARCO Polymers, Inc. EPA Contract
No. 68-02-3061, Task 2. 1980.
24. 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.
25. 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.
26. Reference 25, p. 181.
4-38
-------
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. 11-9.
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.
30. Letter from Monsanto Company to J.R. Farmer, U. S. Environmental
Protection Agency. April 19, 1982. p. 17.
31. Memo from Senyk, D., EEA, Inc., to Distillation file.
32. 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.
33. EEA, Incorporated. Trip Report to Phillips Chemical Company. EPA
Contract No. 68-02-3061, Task 2. August 8, 1980.
34. Erikson, D.G. Organic Chemical Manufacturing, Volume 5: Adsorption,
Condensation, and Absorption Devices, Report 2, Condensation.
U. S. Environmental Protection Agency. Research Triangle °ark,
N.C. Publication No. EPA-450/3-80-027. December 1980. p. 11-3.
35. Reference 24, p. 84.
36. Devore, A., G.J. Vago, and G.J. Picozzi. Heat Exchangers: Specifying
and Selecting. Chemical Engineering. 87(20):133-148. October 1980.
p. 136.
37. Kern, D.Q. Process Heat Transfer. New York, McGraw-Hill Book
Company. 1950. p. 255.
38. Reference 34, p. 11-4.
39. Reference 34, p. IV-1.
40. Parmele, C.S., W.L. O'Connell, and H.S. Basdekis. Vapor-Phase
Adsorption Cuts Pollution, Recovers Solvents. Chemical Engineering.
36(28):58-70. December 1979. p. 60.
41. 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.
4-39
-------
42. Reference 40, p. 69.
43. Reference 40, p. 62.
44. Reference 2, p. 14-2.
45. Standifer, R.L. Organic Chemical Manufacturing, Volume 5: Adsorption,
Condenstation, 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. II1-5.
46. Reference 24, p. 76.
47. Reference 45, p. II-l.
48. Reference 2, p. 14-10.
49. Reference 45, p. III-6 and III-7.
50. Reference 45, p. 1-1.
51. 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.
52. Reference 51, p. 4-7.
53. Reference 51, p. 4-8.
54. 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.
4-40
-------
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 130 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 nay 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
2
allow LLDPE production with either the gas phase process or the liquid
phase solution process. 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
4
accept the gas phase technology has been reported to be relatively low.
Recent advances in LLDPE gas phase technology have resulted in
capacity increases in existing facilities by 35 percent with little or
2
no capital investment and by 65 percent with "some" capital investment.
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 C~ unloading system. No increase in process
emissions is expected from these changes. However, fugitive emissions
5
from these specific changes are expected to increase.
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. 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, LDPE 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 fluidizad 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.
In some cases a facility can be converted, as noted above, from the
production of one polymer or copolymer to another. In such a case,
whenever either the number of fugitive emission sources or the vapor
pressure of the new comonomer is higher than the original, the emissions
could be expected to increase. As shown in Table 3-17, emission rates
for equipment in vapor service are higher than the rates for the sane
equipment in light liquid service which, in turn, are higher than those
rates for equipment in heavy liquid service. So that, if the vapor
pressure of the new monomer or comonomer is higher than the vapor pressure
of the original, the fugitive emissions could be expected to increase.
However, the most common change is to a comonomer of lower vapor pressure.
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.
5-6
-------
5.3 REFERENCES
1. "KEY POLYMERS: POLYPROPYLENE". Chemical and Engineering News.
September 6, 1982. p. 15.
2. "Union Carbide Unveils an Improved UNIPOL Process for LLDPE Resins,"
Chemical Engineering. April 5, 1982. p. 17.
3. "Dow has Announced New Linear Low Density Polyethylene (LLDPE)
Technology." Chemical Engineering. October 5, 1981. p. 35.
4. "A Step Up for LLDPE Know-How." Chemical Week. March 31, 1982,
p. 11.
5. Texas Air Control Board. Permit No. 8334. High Density Polyethylene
Production Expansion. March 30, 1981.
6. Texas Air Control Board. Permit No. 7021. Polyolefins D & G Unit.
October 5, 1978.
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.
5-7
-------
6.0 MODEL PLANTS AMD 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 categories chosen for NSPS development.
Section 6.1 presents the model plants chosen to represent the five
polymers and resins categories, 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 categories,
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 polymer and
resin categories. Therefore, nine model plants were developed:
1. Polypropylene - continuous, liquid phase slurry process,
2. Polypropylene - gas phase process,
3. Low density polyethylene - high-pressure, liquid phase process,
4. Low density and high density polyethylene - low-pressure, gas
phase process,
6-1
-------
5. High density polyethylene - low-pressure, liquid phase slurry
process,
6. High density polyethylene - low-pressure, liquid phase solution
process,
7. Polystyrene - continuous process,
8. Poly(ethylene terephthalate), (PET)-DMT process, and
9. Poly(ethylene terephthalate), (PET) - TPA process.
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 because available information indicates
that new plants are unlikely to use this process.
Tables 6-1 through 6-9 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 6-9 are
the combined characteristics of the individual streams in each process
section as identified in Chapter 3.
As noted in Chapter 5, "Model Process Units and Regulatory Altarnatives,"
of the report "VOC Fugitive Emissions in Synthetic Organic Chemicals
Manufacturing Industry - Background Information for Proposed Standards"
(EPA-450/3-80-Q33a), 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 6-10, was chosen to represent the fugitive emission
characteristics of the polymers and resins industry. This model plant,
however, does not apply to the poly(ethylene terephthalate) model plants
because only a "heavy" liquids (i.e., ethylene glycol) and solids are
used in these plants.
6-2
-------
.- ; =,_. i_. i w._. 5«;
- ** i § : ? ! I i >-> -
O ' 0-0
='v u | s ! e ! = i
3 m cv >o !
= >/> ] o ! o
cvj «r ! fn
j | - j M
I v> c - « tn c c
ml a> ^ o c i <*i
fj -^
! I
~Z < i
i i:2:
5: ^
i
ra i;
3 _ T3
6-3
-------
^ !:
Oi
c.
at
; 5
s '
>
) ev
' C
i O O
Is
* G;
*-> u
c -
O> O)
u
u
a; tr
i- a. a*
o so ^'
c «*
u- aj a.
15 = ' '
'
3 - tr " ' ~
TJ , ' *-
6-4
-------
1
1
!
V) ]
i
<"
UJ
UJ
O
cc
c.
X
o
"
«/> !
c_; :
~
1/1 !
~ !
LU ;
<_J
<
se
<: ;
2
_
g
3j
2 '
7
o
&
o
1
1 ^
c
' 0)
. >
J «C
1
1
1
I
1
1
t
I
j
1
1
I
1
1
i
1
;
i
i
1
1
2
|
"
">
1 ^
|
£
1 °
' V>
! r
=
LA*
a
' ;*
', "S
1 ~
1 o
| u
i
i ^
-r
j
i 5
, ~
1 C
! 3
, _r
t
i ,
u ! i
«
3 ! I
3 o -^ ; . ; =
*|* i ' : "
[ |
i
5~ j ^=
« . ! i ; 5
i» ; ' | o
| ! ! °
"" ' 1
j i
i i
.*-. ' '
cr> w j
C 4J W | C
Z -C ! | 5
TS IB =
*" i! i ""
'*-' }
1 1
1
=L-
?"
(Do i {
1
*J
i '<"=
i=I ! s
U ' 1 C
kk. n t/t r «
*J^ Cv
0
>
<^ ^.
wi C '-.
o >»
U «< [ G
O '- Ol t cs.
Wt- S ) t | -
'* i !
1 1
1 1
1 1
i i
w ; !
fl X '
-» *v. ^£
d ' '"*
S , ' ' ^
1 1
1
V - i |
« 1*x. w ' J
c 5 5 ' ' ~-
O W ' 1
nc. | ! -
.£ '5* '
= ' '
uk.
C «/ , '
ij "=" ' J J
Si i : "2
*-»,(
S v-. ,
i
, (
1 '
- ^ 5
i' ' TS , ^
^" ^ i
^ T =
t/< ^_ ;
^ ' 3V'-
i, T ^ j:
i ' '
i |
| '
' 1 cc ; ec
t j (X j CSJ
i ;
i i
a j (j i w
= _; i- « ' u
* o ' o -
5 !» * »*
U. < | CSJ CO | CM C£
% ' ' O ^ l O CT
1 , C^ | &
i i
i >
i )
; o ] c
, | ~" ! -'
. ; ! i
' !
) 1
1 1
1 1
1 !
1 j
i *r 1 v
i tf) "^
1 i
i i i
: | oo
3 0 Cft
51 ^ ID
1
> 1 1 ^- VD
i
1
1
I
|
|
|
«r ! «
. : - r
i '
1
1 1
i i
t 1
! ! 1
1 1 1
! i 1
1 ' I
s ' ' \c \ *r
; > : i.-5 cv
" i ' ! * ^^
i ' !
i i
i ! :
i i i
- ; I t c ; c
i
i
[
: i 3 c
* L , , > '
\
>
o i>
: > r*
SO | -C
~ r', ' S
I ' " ri ^
. c -c '
J "" ~" i r
CO
c^
j
^ 5
£S
5. u.
g-
3
*.
, ~ .- r^ ^
ri ^ S ^ -= >.
-j; a)--^
C s *" - "a;
CC *T ** i -
CNJ ^ U v"
.. ^ o-^ s -
. . ^ jl ^ -J
>, C in '»_ Ji C
t3 O) 3 ~ -
^5 ^ c "~ i; i-
^" ^» cl *"* * ~~
i - ' Z
T3 ".
-. ~ ~ ^ -^ ~? |
6-5
-------
O .-a 3 '
Ol =» -w
33
2 ! j
! I
SI ~
II >5|££i
VI , '
- i! i! 5
\ 5
O cv
2* S
Otl CC
< ! =o
S5
OJ C2
O ^
s 3 ..r
^v. -TWO
D -a --.
'J i;^-.
VI ; ^3 _
6-5
-------
5 "2 CSJ! i
> «j ~ ' co »*- '
in o ! o er !
! i !
cL !
U.
- ! ! °
5 Hi
S- ! ' c
VJ , ' =
< , -
-------
i'u'
p ! i
?
ce
~':
01 <" i
! 1/1 C i.
ult Oi >i I
!i SI
ec
a. r '
u ' _:
6-8
-------
tu
UJ II
^ I!
± I!
-
u~i ~
-------
w- '
Irt *-> I
c 2 !
JZ I '
a.
U.1
ce
S u vn
i - i
(g g i
3
ce
O i. Ol
W ^ E
C. ~»
O B
i. " -
<->! O
I
T a
6-n
-------
Table 6-10. FUGITIVE VOC EMISSION MODEL PLANT PARAMETERS
Equipment Component
Equipment Counts
Uncontrolled
Emission Rate
(kg/hr/Source)
Valves
Vapor service
Liqht liquid service
Heavy liquid service
Pumps 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
2400
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 HUSSIONS: 151 Mg/yr
6-12
-------
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.
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
difficulties, 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
infomation pertaining to the plant process.
The streams identified as nost 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.
6-13
-------
For the polystyrene and the PET model plants, a different basis was
used to identify the regulatory baseline. Both industries currently use
various recovery technologies that achieve varying levels of VOC emission
reductions. For the 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). For the
poly(ethylene terephthalate) segment, baseline control reflects the use
of distillation columns to recover the ethylene glycol from the cooling
tower water.
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 the SOCMI
and petroleum refinery industries.
6.2.2 Control Techniques
Process emissions may be controlled by flares, thermal incinerators,
catalytic incinerators, boilers, condensers, absorbers, and adsorbers.
Fugitive emissions nay 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
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 ara not typically used,
however, in either the polystyrene or the polyester processes for VOC
emission control and, thus, are not considered as control techniques for
the purpose of defining regulatory alternatives for either of these two
segments.
6-14
-------
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. Themal 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 polyester.
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 6-11) was chosen as the regulatory alternative for controlling
VOC fugitive emissions from the polymers and resins industry. This
regulatory alternative was chosen so that this NSPS would be consistent
with fugitive regulations for the petroleum refining industry and the
synthetic organic chemical manufacturing industry (SOCMI).
5.2.3 Regulatory Alternatives
The following sections present the regulatory alternatives, including
baseline control, for 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
6-15
-------
Table 6-11.
CONTROL SPECIFICATION FOR FUGITIVE EMISSIONS UNDER
REGULATORY ALTERNATIVE 2
Source
Inspection
Interval
Equipment Specification
Valves
Vapor service
Light liquid
Heavy liquid
Pump seals
Light liquid
Heavy liquid
Compressor seals
Safety relief valves
Vapor service
Flanges
Sampling connections
Open-ended lines
(purge, drain, sample lines)
Monthly
Monthly
None
Monthly0
Weekly visual
None
None
None
None
None
None
None
None
None
None
None
Controlled degassing
vents
Rupture disks on
relief valves
None
Closed-purge
sampling
Cap
For punps, 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.
6-16
-------
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 SOCfll and Petroleum Refinery Fugitive VOC NSPS's. 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 the process section remaining
with the highest level of potential uncontrolled emissions until all
process sections were analyzed. For each model plant, if a whole process
section was found to be unreasonable to control as a whole on the basis
of emission reduction achieved and the associated cost, that process
section was then examined to determine whether an individual stream
could be controlled at a reasonable cost.
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.
In developing the regulatory alternatives for the polystyrene and
poly(ethylene terephthalate) (PET) plants, recovery techniques are inori
likely to be used than combustion techniques on streams containing
styrene monomer or ethylene glycol . For polystyrene plants, condensers
are the most likely recovery technique to be used and are used to develop
6-17
-------
the regulatory alternatives for the polystyrene 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 6-12.) By applying the general criteria outlined in Section 6.2.1,
the process emission streams identified under baseline control (Regulatory
Alternative 1) are Stream B (reactor vents), Stream C (decanter vent),
Stream D (neutralizer vent), Stream E (slurry filter/vacuum system vent)
and Stream F (diluent separation and recovery). These streams correspond
to the polymerization reaction section (Stream B) and the material
recovery section (Streams C through F), and are assumed to be controlled
by combustion (i.e., a flare). Based on a 98 percent VOC destruction
efficiency, the baseline control of process emissions is equivalent to
an annual process emission reduction of about 5,010 Mg/yr. Fugitive
emission baseline control, as discussed in Section 6.2.1, would reduce
uncontrolled fugitive emissions from 151 Mg/yr to approximately 106 Mg/yr.
Under baseline control, an annual emission reduction of 5,053 Mg/yr is
achieved or approximately 89 percent of the total uncontrolled emissions
from 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 47 Mg/yr. Under this
alternative, an annual emission reduction of 5,118 Mg/yr is achieved or
approximately 90 percent of the total uncontrolled emissions.
Regulatory Alternative 3 includes the sane control as in Regulatory
Alternative 2 plus the combustion of the emissions from the product
finishing section. This alternative reduces emissions by 5,499 Mg/yr or
97 percent of the total uncontrolled emissions.
Regulatory Alternative 4 includes the same control as in the third
regulatory alternative plus the combustion of emissions from the raw
material preparation section. This alternative is somewhat mors stringent
than the third, reducing annual emissions by 5,510 Mg/yr or 97.2 percent
fron uncontrolled levels.
6-18
-------
Table 6-12.
REGULATORY ALTERNATIVES FOR THE LIQUID PHASE
POLYPROPYLENE PROCESS
Process Emissions
Regulatory
Alternative
1
(Baseline)
2
3
4
Process
Section(s)
Controlled
PR + MR
PR + MR
PR, MR, plus
PF
PR, MR, PF,
plus RMP
Control
Technique
Flare
Flare
Combustion
Combustion
Fugitive
Emissions
d
e
e
e
Annual Emiss
Mg/Yr
5,058
5,118
5,499
5,510
ion Reductions
Percent0
89%
90%
97.0%
97.2%
Process sections include the following:
RMP = Raw Material Preparation
PR = Polymerization Reaction
MR = Material Recovery
PF = Product Finishing
PS = Product Storage
K
Represents reduction in annual VOC emissions from the uncontrolled level.
C8ased on total uncontrolled emissions of 5,667 Mg/yr (5,516 Mg/yr
process emissions plus 151 Mg/yr fugitive emissions.)
Fugitive baseline control.
i-)
Control equivalent to Regulatory Alternative 2 (see Table 6-11).
Combustion devices may include flares, thermal incinerators, catalytic
incinerators, and boilers.
6-19
-------
6.2.3.2 Polypropylene - Gas Phase Process. (Table 6-13.) As
stated in Table 6-2, the mass flowrate of emissions from the material
recovery section is about 460 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. The continuous emissions
may or may not be sent to the same flare. Based on information from one
company, the emissions from the material recovery section may be controlled
by the same flare as the emissions from the polymerization reaction
section. However, considering similar processes by other companies, the
continuous stream may not be controlled in the absence of the NSPS.
Thus, the baseline (Regulatory Alternative 1) includes control of only
the intermittent stream, by a flare. This baseline represents a 31 percent
reduction of total uncontrolled VOC emissions for 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 47 Mg/yr. This alternative
achieves an annual emission reduction of 1,290 Hg/yr, or approximately
32 percent of the total uncontrolled emissions.
Regulatory Alternative 3 includes the same control as in Regulatory
Alternative 2 plus the combustion of the emissions from the material
recovery section. This alternative results in an annual emission reduction
of 3,863 Mg/yr or 97 percent of total uncontrolled emissions.
6.2.3.3 Low Density Polyethylene, High Pressure, Liquid Phase Process.
(Table 6-14.) The baseline control (Regulatory Alternative 1) for this
model plant sends the emissions from the emergency vents other than from
the reactor to a flare. If the high-pressure reactor vent, Stream B, from
this process is sent to a flare, a parti oil ate 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 systsn and
vented to the atmosphere. The exhaust gas contains suspended particulate
polymer that cannot be directly sent to a flara system because of safety
considerations. Unless the particulate removal system technology becomes
6-20
-------
Table 6-13. REGULATORY ALTERNATIVES FOR THE GAS PHASE
POLYPROPYLENE PROCESS
Process Emissions
Regulatory
Alternative
1
(Saseline)
2
3
Process
Section(s)
Controlled
Polymerization
Reactor (PR)
PR
PR Plus
Material
Recovery
Control
Technique
Flare
Flare
Combustion
Fugitive
Emissions
d
e
e
Annual Enissi
Mg/Yr
1,231
1,290
3,863
on Reductions
Percent
31%
32%
97%
aProcess sections include the following:
RMP = Raw Material Preparation
PR = Polymerization Reaction
MR = Material Recovery
PF = Product Finishing
PS = Product Storage
Represents reduction in annual VOC emissions from the uncontrolled level.
C3ased on total uncontrolled emissions of 3,986 Mg/yr (3,835 Mg/yr process emissions
plus 151 Mg/yr fugitive emissions).
Fugitive baseline control.
eControl equivalent to Regulatory Alternative 2 (see Table 6-11).
f
'Combustion devices may include flares, thermal incinerators, catalytic incinerators,
and boilers.
6-21
-------
Table 6-14.
REGULATORY ALTERNATIVES FOR THE HIGH PRESSURE
LIQUID PHASE, LDPE PROCESS
Process Emissions
Regulatory
Alternative
1
(Baseline)
2
3
4
Process
Section(s)
Controlled
(Emergency
Vents Other
than from
Reactor)
Baseline (B)
B
plus PS
Baseline, PS
plus PF
Control Fugitive
Technique Emissions
Flare d
Flare e
Combustion e
Combustion e
Annual Emi
Mg/Yr
100
159
379
434
ssion Reductions
Percent
10%
17%
39%
45%
aProcess sections include the following:
RMP = Raw Material Preparation
PR = Polymerization Reaction
MR = Material Recovery
PF = Product Finishing
PS = Product Storage
^Represents reduction in annual VOC emissions from the uncontrolled level.
°Based on total uncontrolled emissions of 963 Mg/yr (812 Mg/yr process emissions plus
151 Hg/yr fugitive emissions).
Fugitive baseline control.
eControl equivalent to Regulatory Alternative 2 (see Table 6-11).
"^Combustion devices may include flares, thermal incinerators, catalytic incinerators, and
boilers.
6-22
-------
available to most companies, Stream B is not likely to be controlled.
Therefore, Stream 8 is not included in the baseline or any other regulatory
alternative. Under baseline control, process and fugitive emissions
would be reduced by 100 Mg/yr or about 10 percent of the total uncontrolled
emissions.
Regulatory Alternative 2, which represents fugitive emission controls
plus baseline control, results in a total annual emission reduction of
159 Mg/yr or about 17 percent of the total uncontrolled emissions.
Regulatory Alternative 3 represents Regulatory Alternative 2 controls
plus the combustion of emissions from the product storage section. This
alternative reduces total uncontrolled emissions by about 380 Mg/yr or
39 percent.
Regulatory Alternative 4 is the same as Regulatory Alternative 3
plus the combustion of emissions from the product finishing section.
Total uncontrolled emissions under this alternative are reduced by about
434 Mg/yr or 45 percent.
6.2.3.4 Low Density and High Density Polyethylene, Low Pressure, Gas
Phase Process. (Table 5-15.) The intermittent streams in the raw
materials preparation, the polymerization reaction and the product
finishing section and Stream I (product discharge vent) in the product
finishing section were determined to be controlled by flaring under the
baseline assumptions. Annual emission reductions under baseline control
(Regulatory Alternative 1) are about 3,455 f'g/yr or 95 percent of the
total uncontrolled emissions from 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 47 Mg/yr. Under this
alternative, an annual emission reduction of 3,514 f1g/yr is achieved, or
approximately 96.5 percent of the total uncontrolled emissions.
Regulatory Alternative 3 represents Regulatory Alternatives 2
controls plus the combustion of continuous emissions from the raw material
preparation and polymerization reaction sections. Under Regulatory
Alternative 3, emissions are reduced by 3,517 Mg/yr or 96.6 percent.
Regulatory Alternative 4 represents in addition to the control
achieved in Regulatory Alternative 3, the combustion of continuous
6-23
-------
Table 6-15. REGULATORY ALTERNATIVES FOR THE LOW PRESSURE,
GAS PHASE LDPE/HDPE PROCESS
Process Emissions
Regulatory
Alternative
1
(Baseline)
Process
Section(s)
Control! eda
RMP (Inter-
mittent
Control
Technique
Flare
Fugitive
Emissions
d
Annual Enission Reductions"
Mg/Yr Percent
3,455 95%
streams only),
PR (Inter-
mittent
streams only),
and PF
2 Baseline Flare e 3,514 96.5%
3 Baseline Combustion e 3,517 96.6%
plus contin-
uous streams
from RMP and
PR
4 Baseline, Combustionf e 3,524 96.8%
continuous
streams from
RMP and PR,
plus PS
aProcess sections include the following:
RMP = Raw Material Preparation
PR = Polymerization Reaction
MR = Material Recovery
PF = Product Finishing
PS = Product Storage
^Represents reduction in annual VOC emissions from the uncontrolled level,
C3ased on total uncontrolled emissions of 3,641 Mg/yr (3,490 Mg/yr process emissions
plus 151 Mg/yr fugitive emissions).
Fugitive baseline control.
eControl equivalent to Regulatory Alternative 2 (see Table 6-11).
fCombustion devices -nay include flares, themal incinerators, catalytic incinerators,
and boilers.
6-24
-------
emissions in the product storage section. This alternative results in
an annual emission reduction of 3,524 Mg/yr or 96.8 percent of the total
uncontrolled emissions.
6.2.3.5 High Density, Polyethylene, Low Pressure, Liquid Phase,
Slurry Process. (Table 6-16.) Under baseline control (Regulatory
Alternative 1) Stream A (feed preparation) and Stream D (recycle treaters)
would be controlled. These streams correspond to the raw material
preparation and material recycle sections, respectively. Annual uncontrolled
emissions are reduced by 2,700 Mg/yr or 92 percent.
The next level of control, Regulatory Alternative 2, represents
baseline control plus the fugitive emission controls listed in Table 6-11.
An annual emission reduction of about 2,760 Mg/yr would be achieved,
which is a 94 percent reduction from the uncontrolled level of emissions.
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 by about 2,840 Mg/yr
or 97 percent of the total uncontrolled emissions from this model plant.
6.2.3.6 High Density Polyethylene, Low Pressure, Liquid Phase,
Solution Process. (Table 6-17) 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 based. Under this assumption, all the streams in the raw material
preparation, polymerization reaction, and material recovery sections
were assumed to be flared. Baseline control results in an annual emission
reduction of 2,700 Mg/yr or 90 percent of the total uncontrolled emissions.
Regulatory Alternative 2 requires the sane control as baseline plus
fugitive emission controls. This alternative reduces process and fugitive
emissions from the process sections under baseline control by about
2,760 i"1g/yr or 92 percent of the total uncontrolled emissions.
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 2,891 Mg/yr or 97 percent.
6-25
-------
Table 6-16. REGULATORY ALTERNATIVES FOR THE LIQUID PHASE
HIGH DENSITY POLYETHYLENE SLURRY PROCESS
Process Emissions
Regulatory
Alternative
1
(Baseline)
2
3
Process
Section(s)
Controlled
RMP plus
MR
RMP plus
MR
RMP, MR
plus PF
Control
Technique
Flare
Flare
Combustion
Fugitive
Emissions
d
e
e
Annual Emi
Mg/Yr
2,700
2,759
2,842
ssion Reductions
Percent
92%
94%
97%
aProcess sections include the following:
RMP = Raw Material Preparation
PR = Polymerization Reaction
MR = Material Recovery
PF = Product Finishing
PS = Product Storage
Represents reduction in annual VOC emissions from the uncontrolled level.
C3ased on total uncontrolled emissions of 2,945 Mg/yr (2,794 Mg/yr process emissions
plus 151 Mg/yr fugitive emissions).
Fugitive baseline control.
eControl equivalent to Regulatory Alternative 2 (see Table 6-11).
'Combustion devices may include flares, thermal incinerators, catalytic incinerators,
and boilers.
6-26
-------
Table 5-17. REGULATORY ALTERNATIVES FOR THE LIQUID PHASE
HOPE SOLUTION PROCESS
Process Emissions
Regulatory
Alternative
1
(Baseline)
2
3
Process
Section(s)
Controlled
RMP,
RMP,
RMP,
plus
PR,
PR,
PR,
PF
MR
MR
MR
Control
Technique
Flare
Flare
Combustion
Fugitive
Emissions
d
e
e
Annual Emission Reductions
Mg/Yr
2
2
2
,700
,759
,891
Percent1"
90%
92%
97%
Process sections include the following:
RMP = Raw Material Preparation
PR = Polymerization Reaction
MR = Material Recovery
PF - Product Finishing
PS = Product Storage
^Represents reduction in annual VOC emissions from the uncontrolled level.
^Based on total uncontrolled emissions of 2,995 Mg/yr (2844 Mg/yr process emissions
plus 151 Mg/yr fugitive emissions).
A
Fugitive baseline control.
eControl equivalent to Regulatory Alternative 2 (see Table 5-11).
Combustion devices iiay include flares, thermal incinerators, catalytic incinerators,
and boilers.
6-27
-------
6.2,3,7 Polystyrene. Continuous Process. (Table 6-18.) 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 45 flg/yr. This results in a 11 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 47 Mg/yr. Under this
alternative, an annual emission reduction of 104 Mg/yr is achieved, or
approximately 26 percent of the total uncontrolled emissions.
Regulatory Alternative 3 represents fugitive emission control under
Regulatory Alternative 2 plus further recovery of emissions from the
material recovery section by the use of condensers to an emission level
of 0.06 kg VOC/f1g product. Under this alternative, emissions are reduced
by 331 Mg/yr or 84 percent of the total uncontrolled emissions.
6.2.3.8 Poly(ethylene terephthalate), DMT Process. (Table 6-19).
As discussed earlier, baseline control (Regulatory Alternative 1) reflects
the recovery of ethylene glycol from the water in the cooling tower
servicing the polymerizer vacuum system. This system corresponds to a
recovery system currently in use in the industry, thus no emission
reduction from uncontrolled emissions occurs under baseline control.
Regulatory Alternative 2 reflects the use of a different recovery
system (based on the use of a spent ethylene glycol spray condenser
placed between the polymerizers and the vacuum system) that controls
emissions from the polymerization reaction section to an emission level
of 0.21 kg VOC/Mg product. Under this alternative annual emissions are
reduced by about 976 Mg, or about 96 percent from uncontrolled levels.
Regulatory Alternative 3 reflects control under Regulatory Alternative 2
plus the combustion of the methanol emissions from the material recovery
section (i.e., methanol recovery section). Under this alternative,
annual emissions are reduced by 995 Mg, or about 98 percent of uncontrolled
emissions.
6-23
-------
Table 6-18 REGULATORY ALTERNATIVES FOR PROCESS EMISSIONS
FOR THE CONTINUOUS POLYSTYRENE PROCESS
Regulatory
Alternative
1
(Baseline)
2
3
Process
Process
Section
Control 1 ed
-
-
Material
Recovery
Emissions
Control
Technique
Recovery6
Fugitive
Emissions
c
d
d
Annual Emi
Mg/Yr
45
104
331
ssion Reductions3
Percent
11%
26%
84%
Represents reduction in annual VOC emissions from the uncontrolled level.
Based on total uncontrolled emissions of 394 Mg/yr (243 Mg/yr process emissions
plus 151 Mg/yr fugitive emissions).
r>
Fugitive baseline control.
Control equivalent to Regulatory Alternative 2 (see Table 6-11).
rt
Control to an equivalent emission rate of 0.06 kg VOC/"1g product through use of
recovery techniques such as condensers.
6-29
-------
Table 6-19. REGULATORY ALTERNATIVES FOR PROCESS
EMISSIONS FROM THE DMT POLYETHYLENE TEREPHTHALATE) PROCESS
Process Emissions
Regulatory
Alternative
1
(Baseline)
2
3
Process
Section
Controlled
PR
PR
PR
MR
Control
Technique
Recovery
Recovery6
Recovery6 ,
Combustion
Annual Emission
Mg/Yr
0
976
995
Reductions
Percent
0
96%
98%
aPR = Polymerization Reaction
MR = Material Recovery
Represents reduction in annual VOC emissions from uncontrolled levels,
cBased on total uncontrolled emissions of 1,017 Mg/yr.
Recovery of ethylene glycol from the cooling water tower.
Q
Recovery of ethylene glycol to an emission rate of 0.21 kg VOC/Mg
product through the use of a more efficient recovery system such
as using a spent ethylene glycol spray condenser recovering the
ethylene glycol prior to the vacuum system servicing the polymerizers,
Combustion devices include flares, incinerators, and boilers.
6-30
-------
6.2.3.9 Poly(ethy1ene terephthlate), TPA Process. (Table 6-20.)
As discussed earlier, baseline control (Regulatory Alternative 1) is
assumed to be equivalent to current industry practice in which ethylene
glycol is recovered from the water in the cooling tower serving the raw
material preparation polymerization reaction sections. The resulting
emissions are the same as the uncontrolled emissions, and, thus, no
emission reduction occurs under baseline control.
Regulatory Alternative 2 reflects the use of a spent ethylene
glycol spray condenser and recovery system controlling emissions from
the polymerization reactors. The spray condensers would be placed
between the reactors and the vacuum system evacuating the reactors.
This regulatory alternative would reduce emissions by 972 Mg/yr or
97 percent from uncontrolled emissions.
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 6-21.
6-31
-------
Table 6-20. REGULATORY ALTERNATIVES FOR PROCESS EMISSIONS
FROM THE TPA POLYETHYLENE TEREPHTHALATE) PROCESS
Process Emissions
Regulatory
Alternatives
1
(Baseline)
2
Process
Section
Controlled3
RMP, PR
RMP, PR
Control
Technique
Recovery
Recovery6
Annual Emission
Mg/Yr
0
972
Reductions
Percent
0
97%
aRHP = Raw Material Preparation
PR = Polymerization Reaction
Represents reduction in annual VOC emissions from uncontrolled levels.
C8ased on total uncontrolled emissions of 988 Mg/yr.
Recovery of elthylene glycol from the cooling water tower servicing
both the raw material preparation and polymerization reaction section.
eControl of the raw material preparation section to an emission rate
of 0.04 kg VOC/Mg product through the use of recovery techniques
such as reflux condensers and the polymerization reaction section to
0.21 kg VOC/Mg product through the use of a more efficient recovery
system such as a spent ethylene glycol spray condenser recovering
the ethylene glycol prior to the vacuum systai servicing the polymizers.
6-32
-------
'able 5-21. SUMMARY OF JNCONT30LLEO EMISSIONS AND EMISSION REDUCTIONS FOR REGULATORY ALTERNATIVES
3Y MODEL PLANT
UNCONTROLLED
EMISSIONS
MODEL (Mq/yr)
PLANT -rocess Fugitive Total
pp, -
5,516 151 5,667
?P, G 3,335 151 3,986
^DPE, . 312 151 963
LDPE/ 1DPE, G 3,490 151 3,541
"CPE, £L a, 794 151 2,945
HOPE, 50 2,344 151 2,995
JS, : 243 151 394
-" :t'- 1,317 0 1,017
=ET.*3i 998 J 998
REGULATORY
ALTERNATIVE
1
2
3
4
1
2
3
1
2
3
4
1
2
j
4
i
2
j
1
2
3
i
7
3
i
-
J
i
2
A
Process
Emissions
O'g/yr)
5,013
5,013
5,395
5,406
1,185
1,185
3,760
55
55
275
330
3,410
3,410
3,417
3,420
2,655
2,655
2,733
2,655
2,555
2,787
0
0
227
u
976
995
n
9/2
NNUAL EMISSION
Fugi tive
Emissions
(Mg/yr,i -'
45
104
104
104
45
104
104
45
104
104
104
45
104
104
104
45
104
104
45
104
104
45
104
104
i]
0
J
0
0
REDUCTIONS
TOTU
:;ig/y)
5,058
5,113
5,499
5,510
1,230
1,290
3,863
100
159
379
434
3,455
3,514
3,517
3,524
2,T00
2,759
2,342
2,700
2,759
2,891
45
104
331
'j
976
995
j
?72
a
(")
39-;
90°;
97-.
97.2".
3U
32",
97i
10?
17:i
39 -<
452
95 %
96.5?,
96 .5 i
96.3%
92",
94S
971
90S
92s;
97"
in
26:;
34°;
96-;
?8-;
0%
97-;.
jncontrolled levels.
\EV : 3D = Polypropylene
LDPE = Low Density Polyethylene
iOpE = ^^gn Density °o!yethylene
PS = 3o!ystyrene
^" = 3olyethylene terepthaiate
L = Liquid Phase
3 = 3as Dhase
SL = Slurry Process
SO = Solution Process
C = Continuous Process
6-33
-------
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
34 percent less than uncontrolled emissions and about 25 percent less
than estimated baseline emissions (Regulatory Alternative I). VOC
emissions under Regulatory Alternative III are almost 96 percent of
uncontrolled emission levels and 81 percent of baseline emission levels.
VOC emissions under Regulatory Alternative IV are over 96 percent of
uncontrolled emissions and 82 percent of baseline emissions.
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 (fugitive
VOC control) is estimated to be 1,340 terajoules (TJ) (218,000 barrels
of oil) per year over energy demands in the absence of any VOC control.
However, Regulatory Alternative II requires about 78 TJ (13,000 barrels of
*In this chapter, arabic numerals will be used to describe regulatory
alternatives when used in the context of a single model plant, or model
plants of a single type; Roman numerals will be 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
will be collectively described as Regulatory Alternative I.
7-1
-------
oil) per year less than baseline controls (Regulatory Alternative I),
because the energy cost for fugitive VOC control under Regulatory
Alternative 2 for seven model plants and for the more efficient ethylene
glycol recovery system under Regulatory Alternative 2 for the PET model
plants is less than the energy credit resulting from the reduced loss of
VOC under these control measures. Therefore, there is a net reduction
in the energy requirement under Regulatory Alternative II when compared
to Regulatory Alternative I. Regulatory Alternatives III and IV require
about 1,630 TJ and 1,610 TJ, respectively, (265,000 and 262,000 barrels
of oil) per year more than energy demands without VOC controls. Regulatory
Alternatives III and IV require about 213 TJ and 192 TJ, respectively,
(34,000 and 31,000 barrels) per year more than the projected requirement
under current VOC control practices (Regulatory Alternative I) in the
polymers and resins industry.
Projected noise impacts due to implementation of any regulatory
alternative are expected to be minimal. Mo significant irreversible or
irretrievable commitments of resources are expected to be incurred under
the regulatory alternatives. Delaying implementation of Regulatory
Alternatives II through IV 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 model plant VOC emission rates are determined
and used to project industrywide air quality impacts of new polymers amd
resins plants.
7.1.1 Average Annual Model Plant VOC Emissions
Annual VOC emission rates for each model plant are determined
through the following equation:
E = P.. + F. (1)
U 1J J
7-2
-------
where E.. = annual VOC emission rate (Mg/yr) for model plant i (for
|J ~
example i = polypropylene/liquid phase; for other model
plant types see Table 6-20) under Regulatory
Alternative j_ (j = uncontrolled, I, II, III or IV);
P.. = annual VOC process emission rate for model plant j_
under Regulatory Alternative j_;
F. = annual VOC fugitive emission rate for each model
J
plant under Regulatory Alternative j_.
The annual model plant VOC process emission rate (P,-J 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 model process VOC emission rate (P..) can be expressed
' J
as:
where U . = the uncontrolled VOC emission rate (Mg/yr) from process
31
section a_ (a = raw material preparation, polymerization
reaction, material recovery, product finishing, or product
storage) in model plant j_;
C .. - the VOC emission reduction efficiency of the control system
a ij
for process section a^ in model plant j_ under Regulatory
Alternative j_.
The uncontrolled process emission rates for each process section in each
model plant are presented in Tables 6-1 through 6-9. The control techniques
employed on appropriate model plant process sections and their emission
control efficiencies are presented in Tables 6-11 through 5-19.
The quantity of fugitive VOC emissions are assumed to be the sane
for each model plant type. The uncontrolled fugitive VOC emission rate
is based on equipment component counts and uncontrolled component emission
rates presented in Table 6-10. The effectiveness of fugitive VOC emission
control through the use of leak detection and repair programs (Regulatory
Alternatives II, III, or IV) is based on the leak detection and repair
(LDAR) model developed for control of fugitive VOC emissions from SOCHI,
and is presented in Table 6-11.
Table 7-1 presents the primary or VOC related air quality impacts
of the regulatory alternatives for each model plant. (Table 7-1 and
other tables, as appropriate, in this chapter have been separated into
7-3
-------
Taale
-la °RK1ARY AIR QUALITY IMPACTS OF THE REGULATORY ALTERNATIVES FOR
POLYMERS AND RESINS PLANTS Clg/yr)
Model 31ant
PP/Liquid
PP/Gas
LDPE/Liquid
LDPE/HOPE/
Gas
HDPE/Liauid
SluTy
riDPE/liquid
Solution
PC
SET/DTT
,--/-DA
Regulatory
Alternative
ua
1
7
3
4
a
U
1
2
3
Ua
1
2
3
4
Ua
^
0
3
4
Ua
1
2
3
a
U
1
2
3
Ja
1
2
^
Ja
1
3
u
1
0
voc
f
Process
5,516
504
504
122
ill
3,835
2,650
2,650
n
812
757
757
537
482
3,490
80
30
77
70
2,794
139
139
56
2,844
139
189
57
243
243
243
16
l,0i:
1,017
41
22
QQ£
992
26
Emissions
1odel Plant
Fugitive
151
106
47
47
47
151
106
47
47
151
106
47
47
47
151
106
47
47
47
151
106
47
47
151
106
47
47
151
105
47
47
c
0
0
0
o
"1
Q
ger
Percent VOC Emission
Fron Baseline (Regulatory
Combined
5,667
610
551
169
158
3,986
2,756
2,697
124
963
863
804
534
529
3,641
136
127
124
117
2,945
245
186
103
2,995
295
236
104
394
349
290
62
1,017
:,oi7
41
22
99S
993
25
Process
-
0
76
73
.
-
0
97
-
0
29
36
-
.
0
10
13
-
-
0
60
-
,
a
70
_
-
93
^
96
93
.
9"
Fugitive
_
_
56
55
56
.
56
56
-
56
56
56
-
.
56
56
56
-
-
56
56
-
-
DC
- c
_
-
55
DO
_
.
3
3
-
j
Reduction
Alternative '.)
Combined
_
.
U
72
74
-
2
96
_
-
J
32
39
-
.
32
7 7
3"
-
-
24
58
-
-
20
? ~
_
-
1~
G'_
_
-
96
op
.
-
9"
7-4
-------
Table 7-lD PRIMARY AIR QUALITY IMPACTS OF THE REGULATORY ALTERNATIVES FOR
POLYMERS AND RESINS PLANTS (tons/yr)
Model Plant
PP/Liquid
PP/Gas
LDPE/Liquid
LDPE/HDPE/
Gas
HDPE/Liquid
Slurry
HDPF/Liquid-
Solution
3s
PET/DflT
JCT /TDi\
Regulatory
Alternative
ua
1
2
3
4
ua
1
2
3
'Ja
1
2
3
4
ua
i
2
3
4
Ua
1
2
3
'Ja
1
9
3
J3
I
2
3
,3
1
7
3
U*
1
2
vo:
Emissions
Model P1anth
Process
6,068
554
554
134
122
4,227
2,921
2,921
85
895
334
834
591
530
3,339
88
88
85
77
3,080
153
153
62
3,135
208
208
63
263
268
268
18
1,120
1,120
45
24
1,100
1,100
24
Fugitive
166
117
52
52
52
166
117
52
52
166
117
52
52
52
166
117
52
52
52
166
117
52
52
166
117
52
52
166
117
52
52
0
o
Q
0
0
0
c
oer Percent VOC Enission Reduction
From Baseline (Regulatory
Combined
6,234
671
606
136
174
4,393
3,033
2,973
137
1,061
951
886
643
582
4,005
205
140
137
129
3,246
270
205
114
3,301
325
260
115
434
335
320
7u
1,120
1.120
45
24
1,100
1,130
24
Process
_
-
0
76
78
_
.
0
97
-
0
29
36
.
.
0
10
13
-
-
0
50
-
.
0
70
_
.
0
93
_
0
96
98
_
97
Fugitive
_
_
56
56
56
_
.
56
56
_
.
56
56
56
_
_
56
56
56
-
_
56
56
.
_
56
56
_
_
36
56
_
o
56
56
_
0
Alternative 1)
Combined
_
.
10
72
74
_
.
2
96
_
-
7
32
39
_
.
32
33
37
-
-
24
53
-
_
21
33
_
.
1"
3:
_
^
96
92
.
T
7-5
-------
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 95 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. For example, under baseline control only a 10 percent reduction
in uncontrolled process VOC emissions from a liquid phase, low density
polyethylene (LDPE/Liquid) manufacturing plant, because the major emitting
process section -- the polymerization reactor can not be fitted with
a VOC control system because of safety considerations. (Safety problems
associated with control of VOC from the LDPE/Liquid polymerization
reactor are discussed in Section 6.2.3.3.) Similarly, gas phase polypropylene
(PP/Gas) process emissions under baseline control are reduced from
uncontrolled levels by 31 percent, because only polymerization reactor
exhaust streams are controlled. 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 151 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 56 percent in fugitive emissions of VOC from each model
plant except for the PET model plants. Regulatory Alternative 2 for the
poly(ethylene terepthalate) model plants, reduces baseline emissions
through the use of an ethylene glycol spray condenser and recovery
system. Under Regulatory Alternative II, overall VOC emission reductions
from baseline control (processes and fugitive emission reductions) range
from 2 to 97 percent. The differences in emission reduction efficiency
over baseline among 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 the model plant.
7-6
-------
Under Regulatory Alternative III, controls are applied to process
emissions. Implementation of Regulatory Alternative III would result in
a range of overall VOC emission reduction from baseline control of about
32 to 98 percent. (Note that there is no Regulatory Alternative 3 for
the PET/TPA Model Plant.)
Under Regulatory Alternative IV, more stringent control requirements
for process emissions were analyzed for the following model plants: liquid
phase polypropylene (PP/Liquid), liquid phase low density polyethylene
(LDPE/Liquid) and gas phase low density/high density polyethylene
(LDPE/HDPE/Gas). Combined process and fugitive VOC emission reduction
levels of 74, 39, and 37 percent from baseline emissions, respectively,
occurred under each model plant's Regulatory Alternative 4.
7.1.2 Industrywide VOC Emission Impacts of New Plants
The effect of the regulatory alternatives on VOC emissions from new
plants projected to be built over the next 5 years can be estimated by
the equation:
(3)
i
where NT = total VOC emissions from all new polymer and resin
plants (Mg/yr);
E. . = annual emission rate from model plant j_ under Regulatory
Alternative j_ (Mg/yr-from Table 7-1);
n. = number of new model plant j_ projected to be built by 1988
(f"om Table 8-38).
The impact of the regulatory alternatives on VOC emissions from the
27 new plants expected to be built are presented in Table 7-2. Baseline
emissions from new plants are projected to be 17,910 Mg/yr (19,740 tons/yr),
or a 79 percent overall reduction fron uncontrolled levels. The additional
increment of emission control under Regulatory Alternative II results in
overall VOC emissions of about 13,570 Mg/yr (14,950 tons/yr): an 34 percent
decrease from the uncontrolled emission rate, or 24 percent incranental
reduction in emissions over Regulatory Alternative I. The additional
process emission controls required under Regulatory Alternative III are
expected to reduce uncontrolled emissions by 96 percent to about 3,430 Mg/yr
(3,770 tons/yr): an 75 percent increase in reduction efficiency over
Regulatory Alternative II. Additional or mora effective process emission
controls employed in some plants under Regulatory Alternative IV result
7-7
-------
t/1
Ol
!
O C
'V .5
^ = , -
^ ^ O -"
= -wtJ i
s-i i
trt O
O CO O O
O rou") ' r--
o cc^o to ^r
- i 5^
0000
-------
c a; ! o o o o IOGO o o o o a
O C CM CM -CT CM OOU^ _i^_tfv.fv1P^.
joo ;oo OO>UD
-------
-
° *"" -9
*-j r c
) O '^D ' I I O vC un
^ r*» p>* CM
CM CO CO
1JD *ff ^ t O
co co co in u->
o o o o c
cr> co co m <
rocc CO CO r
J O CM I O CM CM
*-< 'S (^ CM
O Oi CO
'- I
£
<-C O <-O ^
CM CM CM '
^P^CMCMCM ^-cr-^CMrM
^O -^ CM CV
' -^ ^^ Lf)
CM CM CM CO
7-10
-------
~5 == CT !
I o o o
00000
o o o c; o
CO ~* Lrt 01 UT
cr cc CM csj oo
o o o o o
< >-
5s
7-n
-------
in an overall VOC emission rate of 3,270 Mg/yr (3,520 tons/yr): this
value represents an emission reduction from uncontrolled levels of
slightly greater than 96 percent and a 4 percent increase in emission
reduction efficiency over Regulatory Alternative III emission reduction
level s.
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
power to operate control devices may adversely impact air quality.
Consequently, particulate, sulfur oxide (SO ), and nitrogen oxide (NO )
A A
emission rates are estimated, based on energy consumption rates for VOC
emission control presented in Chapter 8, by the equation:
sxj E C(6,j x 'GX> + l "i * °-454- --
where S . = new plant industrywide emission rate (Mg/yr) of secondary
xj
air pollutant x_ (x = particulate, SO , No ) under Regulatory
X X
Alternative j_ (j = 1, 2, 3, 4);
G.. = natural gas requirements to control VOC emissions from model
plant 1 under Regulatory Atlernative j_ (10 ft /yr);
2
Kr = pollutant x emission factor for natural gas combustion
bX c o
(lb/100 ftj)
B-. = fuel oil required to generate electricity and steam
for VOC control equipment in model plant j_ under
Regulatory Alternative j_ (gal/yr);
o
K = pollutant x emission factor for No. 4 fuel oil combustion
ox p , -,
gal
n = number of new plants of model plant type i to be built
i ~
in year 1988.
7-12
-------
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.
It is recognized that particulate and SO emissions from fuel combustion
X
will be reduced through State Implementation Plan (SIP) requirements
which follow the guidelines set forth in EPA regulations on the preparation
of implementation plans [40 CFR 51]. 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 plants are presented in Table 7-3.
The 27 new plants that are expected to come on line by 1988 are
projected to emit approximately 30 Mg/yr (33 tons/yr) of particulate,
1,323 Mg/yr (1,458 tons/yr) of SO. and 258 Mg/yr (284 tons/yr) of NO.
A X
if these plants operate under current industry practices (Regulatory
Alternative I). Secondary air pollutant emissions will not increase
appreciably under Regulatory Alternative II, because 24 of the 27 plants
will employ fugitive emission control programs which do not require the
use of add-on pollution control systems. The new PET plants under
Regulatory Alternative 2 are estimated to emit an additional 1.3 Mg/yr
(1.4 tons/yr) of particulate, 66 Mg/yr (72.6 tons/yr) of SO , and 12.6 Mg/yr
A
(13.9 tons/yr) of NO above baseline control.
A
Particulate emissions generated by new plants during control of VOC
containing process streams under Regulatory Alternative III are about
33.2 Mg/yr (36.6 tons/yr), or 11 percent greater than secondary air
pollutant emissions under baseline conditions. SO emissions under
X
Regulatory Alternative III are projected at 1,400 Mg/yr (1,542 tons/yr):
a 5 percent increase in emissions over Regulatory Alternative I. NO
X
emissions under Regulatory Alternative III are estimated to be about
299 Mg/yr (330 tons/yr), which represents a 16 percent increase over
baseline secondary air pollutant emissions.
Projected particulate, SO and NO emissions from new plants operating
X X
under Regulatory Alternative IV are 32.8 Mg/yr (36.1 tons/yr), 1,380 Mg/yr
(1,521 tons/yr) and 295 Mg/yr (325 tons/yr) respectively. Under Regulatory
Alternative IV, the 2 percent across-the-board reduction in secondary air
pollutant emissions from the levels estimated for Regulatory Alternative III
7-13
-------
i
i
1
;O J GOG-- ~^r-^
' ' *=T «T *T G O O CM
1 1 >
t ! :
t t '
i/l 1^1 * * * d*1 ' i-D d <~!
OS- 'CO ' CM CM CM C ' GOO
- -5 i i
iH " ! 1 1
U i
\4-> 1
|« ; ^_
i.§ ! ^ ^ ^ S o G £
I-J ! OOGO ' GOO
1 W 1 '
I10 I
JQ_ j i
1 ' 1
' I
t !
1 !
I
I
I {
-, I i
~7>
»i_ ) ' 1
i. ^ f*i ' co
s°-
i s
3 J
i i
1 1
i I
i
1 t
1 j f-» i ~O OC ' r~* r*^ o^
i x j T-^rLDun ro r^ CM; CM CO
i i ~* ^^ !
i (
j-j '
| CO CC CO '-0 SO CO O
j« x: -^^r*^ o r-~,r*-^-
- ^_^ on | * " - ^
>> |
*" "*** <
o -^ i
i i-2 '
^ ^ ' '
" 5 '^-*T-T^ 20COO
i. i
t»
d-
1 i
i j
^ ;
>^ 0)
"" > t
-o -a \
^ t ; ^ CM c- ^- -cMcn
5 ' 1
^ ' '
"° 1 "^ 1
1 ~ "^
CM CM CM ^J
c\i ^j CM c>;
-S^ £ ^
^ ^ ^D U3
CM CM CM CV
^y. ^J- ^3- ,
|
0 0 O 0
iO U3 O >
i-* CM C\J
CM ^M CM C~O
0 0 0
O G O C
^^"i^vo"
2 3 3 UD
0000
r CM CM
CM r- T
' 5-
; f
- ^-i a> o^
r^ (_,* ^ p^T
1 ^ ^ri
*T cr G G
_n u-' G* G
1
1
1
I
^ ^^-^3-
i ^- ~^ i >
|
1
j
I
i
0
G G CT1 CTi
~H ,- ^-^-<
1
|
I r- .-. cr.cn
| J ? ^m
^r^r^V
!
I
I
G 0 GO
1
I
| -!M««-
! -a
J
; ?
GO G
U C G CO
t
1
-O O '
u o c ; o
1 CM
I
1
G O ]
2 °. | ^r
^
CM
|
t
o uJ ^' I cr,
G
CO
ro
CM CM
(J CC CO t
CM CM "M
i '
! ^
CM CM
O G G ' 3*i
cv. -« : -
I
. . 1 ._.
GO ^ .
"^*
Ni IM CO -^
-t ^" 3*> CM
CM CM ^ *T
1
I
t
I
t
r^ (-^ 0«f
^^ j « ^
I
I
i
t
CM
t
I
I
1
1
!
uo 3 | 0 £
O G ' CO G
,_- ,Q ' , ^
OJ "VJ ' "M CM
CM "si | CM
CM CM CT1 %!
CM C^,' ( i CNJ
55 | 5S
CM ^ CM
7-14
-------
F
I I
^z. i
' I
X
a
*-
X
C
VI
4-J
fO
3
«
t?
Q_
X
C
3
X
Zn
3J
T3
5
s_
fC
O_
0 O -C T3
0s O 1 ro
*-,"""
-f) vO 5" 3~>
-TV r-j c-' cv
u o -a -c
-" O O ~
-> «3- «=r -o
" * * *
CO ^ CM CO
cj\ o n c\j
CM ro m f!
CM
I t t I
if t t
J I 11
1 ^- =>
i
3
^
1
i_
2
t;
-"0
CM
0
CM
t
CO
l/l
QJ
^
OJ
o
i/1
r
"H
*
C
W
0)
^J
T3
>>
|
fO
i
1_
i13
OJ
i.
1)
T3
-
C
j_)
U
1J
)
"i
CJ
a)
u
i)
^
_
5
yi
O
J
ji
^
«r
'w
CD
t.
ȣ
0)
cc
u
u
i.
o
t^j
^
~i
,
J2
w
ii
4-0
Q
U
U
a
c:
cu
'J
i_
OJ
a.
CM
CT'
>
13
~
"
O
01
«3-
i
o
~
^
Jl
^~
o
o
u
a
"O
3D
U
-a
5
^-
^
OJ
>
'_
"J
=c
>,
X
(T3
"tr
5i
OJ
CXL
L.
u-
t/i
-fl
O
*-*
c
c
3J
T7
^
U
C
o
c
OJ
w
t3
T3
C
fD
OJ
"^
TJ
>
i-S
O
OJ
-a
s>
*_)
-a
'o.
0)
i_
>!
>,
1.
5
i^
ji
a
3
__
O
C.
>,
o
"2
~
_J
u
'71
j_
O
71
u
3
=
'
J"*
>-,
"
^S
c
,
i
51
u
n
u
a;
i/
o
OJ
3
T3
>
C
c
VI
ul
i
^
O)
>
4J
(O
C
OJ
-o
>,
5
a
31
2J
-^
sJ
5_
o
3.
13
J_
~
3
a U
a. 3
yi
if '-
i D
^3 ^~
aj
5 «
<4- O
1) -4~J
3 C
, 13
a r
> t=-
C r
C U
-D
'-o £
-!->
o 2
- *3
s
'71 ~
1; C.1
'*- ;>
3^ ^
*3 T3
S- 1'
-5-:5
7-15
-------
I i-"^ j^ I
-.->--> CO CT ' "- CM I-- !
*r CO | ,-* 0
1 t
i
1
I
i
g = 3~ 55 2 2§Si §§S£!
CT. CT ^ *~ * -D VD C5* CM !M O fl { CO CC O O
CMCMr^J -H COCOC1C3* CMCMr^f^l
CM CM CMCM
1
OOO^J ^l^O OOOO OOOO
-^ | ^om^ ! oooc oooo
n-rr«r '^-'rsj^jCMf^ o O " '
^ ^ ^^
i i
i
COCOO-^ C^CTiCTi OOOO '-O^O^CJ^
*" n -o | r-.|cMCMf^rn CMC-v.ro""
; q -" £ £ |
I
I f t
1
1 ' 1 t
, ,
O o *r ' O O ' '
O O O i ~^ "~ ' '-C ! OO cnr--CO ^cn*rr'
r-j CM CM ~ r- _o " ^ o* 0* ^-«erW ; rn cr.
~° 1 *3*rr '^ 1 ~
S5£ ==,- ,,a8 '
en s^ en ~ -f> o ' o o o ^T "^ | c?i cc
| f'J CM CM | *3" *T
1 i !
1 i '
1 1 1
I T
.-M ,- -1 i
CM CM CM f^ n r-* O O CO CM CM ' CO *3*
OOO OOCM OO
I
«3-^-un ' i^--^ : "" JD so . 'OO ' ' o
i^'occ ooo c^crcnlocr
C^CTiCTi OOf-^1 cT1^-; 31 ."*
-------
jcx I ~ ^->r
r: 1 *r -N; c-. u
'3 o
CO CO U3 i-t
O =
T3
S_ i
r CH
31 OJ
b !
! 5
a.
3J
3 CJ
O) f-
T;
9 >>
31 OJ
0) -J
SC i
i ,_ w ai *-
c D
Jl C
7-17
-------
is due to switching from the use of separate flare and thermal incineration
control systems to venting all VOC to thermal incinerators in new PP/liquid
plants. Consequently, the energy requirements for VOC emission control
systems in PP/Liquid plants under Regulatory Alternative IV are less
than the energy requirements for the other regulatory alternatives. The
energy requirement aspect of the regulatory alternatives is discussed in
detail in Section 7.4.
Comparison of the total mass of secondary pollutant emissions
(particulate + SO + NO ) presented in Table 7-3 to total process VOC
j\ y\
emissions presented in Table 7-2 indicates that under Regulatory
Alternative I, a 66,100 Mg/yr (73,630 ton/yr) reduction in VOC emissions
is obtained with a concomitant potential increase in secondary pollutant
emissions of 1,611 Mg/yr (1,775 tons/yr). Under Regulatory Alternative III,
the alternative having the greatest potential for secondary air pollutant
emissions, uncontrolled VOC emissions are reduced by 80,620 Mg/yr
(39,650 tons/yr), while increasing the potential amount of secondary
pollutants by about 1,730 Mg/yr (1,910/tons/yr). 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. 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
With the exception of LDPE/Liquid plants, (Section 3.2.2), there
are no water effluents from the model plants. In the case of the LDPE/
Liquid process, centrifugal scrubbers are used upstream of the VOC
control device (catalytic incinerator) to remove particulates that may
plug or foul the catalyst bed. The cooling water, used in the scrubber
to separate product resin from the waste gases, is routed to a knock-out
drum where contained gas is removed. The resin is separated and the
water recirculated to the scrubber. Thus, the system essentially is
closed and results in no water pollution.
About 4,900 m /yr (1.3 x 10 gallons/yr) of water are consumed by
condensers used to control VOC emissions from a PS production unit
(Table 8-26). This represents approximately 4 percent of the consumptive
water requirements of a PS plant based on the bulk process.0 Increased
7-18
-------
water use will occur under certain regulatory alternatives because of
the ethylene glycol condensers and recovery systems used in PET plants.
For Regulatory Alternative 2 at PET/TPA model plants, and Regulatory
Alternatives 3 and 4 at PET/DMT model plants, water usage will increase
by 58,674 m3 (15.5 x 10^ gal/year) each, or about 7 percent of the
consumptive water requirements of a continuous polyester resin plant.
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 LDPE/Liquid processes under Regulatory
Alternative 3 and 4, LDPE/HDPE/Gas processes operating under Regulatory
Alternative 4, and HDPE/Liquid solution processes under Regulatory
Alternative 3.
The quantity of spent catalyst expected to be generated annually by
new plants is estimated by the equation:
D,. = [(R .. x Q x 0.02382 m3/ft3 x n.) * 3] (5)
IJ a IJ 1
where Dj. = Annual quantity of spent catalyst generated under
Regulatory Alternative^ (j = I, II, III or IV);
R .. = Exhaust gas flow rate (scfm) from process section a
a IJ ~
(a = raw materials preparation, polymerization reduction,
material recovery, product finishing, or product storage)
in model plant j_ (i = liquid phase polypropylene and other
process types) under Regulatory Alternative j_ (refer to
Tables 6-1 to 6-8);
Q = 2.25 ft of catalyst/1,000 scfm of catalytic incineration
throughput (Section 8.1.3.1)
N. = number of new plants of node! plant type j_ expected to be
built by 1988 (Table 8-38).
It is assumed that the catalyst bed has a useful life of 3 years
(Section 8.1.3.1).
7-19
-------
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 14 new plants expected to employ catalytic
incinerators is projected to be 2.88 m /yr (102 ft /yr) under Regulatory
3 3
Alternative III and 3.01 m /yr (135 ft /yr) under Regulatory Alternative IV.
The volume of solid waste generated by the use of spray towers or liquid
jet scrubbers employed upstream of catalytic incineraters to treat
corrosive exhaust gases under Regulatory Alternative IV (Section 4.2.2.3)
represents about 0.01 percent of the volume of biological sludge generated
annually as solid waste by process operations in LDPE/Liquid, LDPE/HDPE/Gas,
and HDPE/Liquid processes. Volumes of biological sludge generated by
applicable process operations in new plants 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
g
incinerator is not listed as a hazardous waste under 40 CFR 261.30.
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
Q
and a 95 percent removal efficiency for SO emissions. Applying these
X
efficiences to the secondary pollutants presented in Table 7-3 and
assuming a fly ash density factory of 0.72 Mg/m (45 Ib/ft ), the
maximum volume of fly ash generated industrywide by new polymers and
3 3
resins plants is projected to be about 1,157 m /yr (40,660 ft yr) under
Regulatory Alternative III, the alternative having the greatest potential
for secondary pollutant emissions. This fly ash generation rate represents
0.8 to 2.2 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/f1g 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).U
7-20
-------
Table 7-4. INDUSTRYWIDE SOLID WASTE IMPACTS OF THE REGULATORY ALTERNATIVES FOR
NEW POLYMER AND RESIN PLANTS
Model
Plant
LDPE/Liquid
LDPE/HDPE/Gas
HDPE/Liquid-
Solution
TOTAL
Regulatory
Alternative
1
2
3
4
1
2
3
4
1
2
3
I
II
III
IV
Solid
waste
generated
m3(fr)/yrb
0
0
0.13 (4.8)
0.17 (6.0)
0
0
0
0.09 (3.2)
0
0
0.92 (32.4)
-
-
-
Solid waste
generated
Number of iqdustrywide-
New Plants0 m3 (ft^/yr^
0
1 0
0.13 (4.8)
0.17 (6.0)
0
10 Q°
0.09 (32.0)
0
3 0
2.75 (97.2)
0
Q
4 2.88 (102) ,
14 3.01 (135)°
Solid waste generated through use of catalytic incinerators.
K "3
Based on 2.25 ft of catalyst/1,000 scfm of catalytic incinerator throughput
and a three year catalyst life. See Tables 6-1 through 6-13, Section 4.2.2.4.2
and Section 8.1.3.1.
cSee Table 3-3S.
Where there is no value for a model plant under a particular regulatory
alternative, the value of the preceeding regulatory alternative for that
model plant category is used.
7-21
-------
Table 7-5. VOLUME OF BIOLOGICAL SLUDGE GENERATED BY
PROCESS OPERATIONS IN NEW POLYMER AND RESINS PLANTS THAT EMPLOY
FLARES, THERMAL INCINERATION, OR CATALYTIC INCINERATION
Model Plant
Model Plant Annual
Production Rate
Sg/yr (Ib/yr)
.Number of ,
New Plants
Industrywide
Biological Sludge
Production Rate - rn3/yr (ftVyr)
PP/Liquid
PP/Gas
LDPE/liquid
LOPE/HOPE/
Gas
HOPE/'-iauid
Slurry
HDPE/Liquid
Solution
PS
PET/DMT
PET/TPA
Total
150
105
280
150
210
90
75
105
105
_
(331 x
(231 x
(617 x
(331 x
(463 x
(198 x
(165 x
(231 x
(231 x
_
105)
106)
105)
106)
106}
106)
io6)
105)
IO6)
3
3
1
10
2
3
2
1
2
27
4,870-12,200(172,000-431,000)
4,530-11,370(160,000-402,000)
3,396-8,528(120,000-302,000)
16,231-40,692(573,000-1,433,000)
4,540-11,400(150,000-403,000)
2,910-7,310(103,000-258,000)
1,618-5,680(57,200-200,000)
2,340-11,400(100,000-403,000)
5,680-22,800(200,000-806,000)
47,500-128,000(168,000-453,000)
22,161-55,592(1.59 x 106-4.51 x IO6)'
From Tables 6-1 through 6-9.
bFrom Table 8-38.
C3ased on 49-123 m sludge/yr/10 x 10 Ibs. of polypropylene, polyethylene, and polystyrene produced;
123-493 m sludge/yr/10 x 10 Ibs of polyester produced. (Source: Reference 7)
These totals include only sludge generated by process operations employing catalytic
incineration under the regulatory alternatives.
7-22
-------
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
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 each model
plant are calculated as the sum of energy expended to control VOC emissions
from process operations and the energy saved by implementing fugitive
VOC emission control 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, electricity 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 process emissions of VOC from each
model plant under each regulatory alternative are presented in Table 7-6.
Energy impacts associated with process VOC emission control generally
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. However, energy impacts
associated with control of process emissions from PP/Liquid model plants
operating under Regulatory Alternative 4 are less than Regulatory
Alternative 3 impacts because control techniques that are more energy-
efficient are used (thermal incinerators rather than flares).
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 IV.
7-23
-------
c o
9-
S- CJ
T3 OJ
C !_l
1?
O *J
3
0
4-J
§. ^5-
>> O
OJ
Ul ^ ->3
O U
C ri 1-
O
,_.
O
!
0
C_
1
^
5
-jj
13
-3
-a
o
"J
£"*-
D C
r U
IS
1^
!
Q O O O O O C
j
O 0 O 0 OOO
000 0 0
CM CM CM CM CM
o en =n a, o en en
1
t
)
oooo ooo
r*-» p»x (^- CM CM <^t
i
i
o o o ooo
j
1
I 1 O ill
1
o o o i CM CM CM
* <*-. '** C\l CM CM
.
(
1 *
- ^ - ^ : - ^ -,
] -a
1 rt it
oooo
oooo
oooo
CO CO CO
CM CM CM
( 1 1
0 en en C*
oooo
oooo
^r *» oo o
if) in ui '-o
oooo
oooo
CM CM CM CM
LT) -D Lf) J^
1 t
> O
^
OOOO
^ " GO ~
- - " *
_ .VJ -, ^
'; 3
-
i =
r*^ LT3 O O csj ^~ '-T5 ^~ '-J3 *T
O"s '-C P^- r^ in "M c\j '-O CM CP>
en
1
1
1
oooo ooo ooo
CO CO CO CO CO CO CO
CM CM CM CM CM CM CM
1 t 1 1 1 It
t 1
omcicn oc^o> ccnc*
,
oooo ' ooo ooo
CMCMCnCT? U1LO-'"' CMCMO
^^2
oooo ooo ooo
1
!
, , , , 1 lilt i i ~c
' !
]
oooo; ! o
* * * * 555 ^ * 5
1
1
I !
!
1
: p
^ J ^3 ^^ 1
= 3 ~ 5" ; 5 H"^
7-24
-------
c: o c; . o =
000 00
i c o o
CN. O)
> UJ XJ
i- O O;
M i~ O U
CT i- T3
3 *j c- a;
H
I
l!°
£ !
o o
0 0
-C
o o
o c
CM O
T "
0 O
CM CM
o o
X X
o o
CM CM
0
o
-C
0
o
n
o
o
CM
U">
0
X
0
CM
O O
o o
-C
o ~"o
c o
CM 0
T -""
0 0
CM CM
iT! LO
O O
X X
c c
CM CM
1
X'
C£
c
C!
>
3
>,
C71
i.
OJ
C
0,'
c
^
c
c
-
>,
OJ
"oi
OJ
OJ
a:
" '
g
OJ
"^
>
3
CT
i
O)
OJ
c
T
o
4/1
2
>,
1_)
C
OJ
'£
5J
C
CM
i
jj
CJ
>
3
a;
a
c
CJ
c
o
*J
o
c
.JC
*^
LJ
Ol
E
.^
O [ CM O O
I "" "^ ^
i- C C
U wi
c~ c o
- S- CJ
C O .E
OJ '-
CM n ] ^i CM
~ ^1
3 c 3 "- a/
j C2 J= O
7-25
-------
z
_ 1
Q.
1
UJ
UJ
o
UJ
_J
2
a:
£
t/j
UJ
t
Si
oc.
UJ
5
o
~^>
LjJ
ac |
UJ
LJ_
0
£
<
s
^.
frt
d
UJ
1
£
P^
OJ
S
1
wl
o
to
tn
UJ
0)
4-1
cn
u_
j_,
u
g
s-
i ^
UJ
1 ^
.w
c
j =
t
£
o
' vi
on
i es
t/i
to
OJ
U
1 c
a_
i
1
i
i
i
i
i
i
J
i
i
^j
U
c «
o o.
i- £ S.
UJ 1. O
OJ
o c i
OJ UJ J2
S "°
5 +J
0 C
o
(U
o -^
Q. ^
M °
>, J3
0> J=
tu
C
UJ
o
O 13
O > HI
1- U
*J (/I 3
C C T3
O O <1J
(J 1 i-
T-,
rO
^j ;_
."
Ifl '-
i/l O
(U
0 S
°-
^
O -C
5
0) 0
J-J O
wo C
"
_Q
>,
sU
r ^v.
0 3
__
LU
^
C3 >>
- ^
2 S
J-4O
fO O
Z ^
OJ
i. -t-
O J-
j_) ro
^ U
o) ^
C£L <
CJ
O £-
o o o o
CM CO CO ro
*> i
O O O O
^" ^f1 "V
1 1 1
O LO LO UO
^O US VO
O O O O
^ ^" LO LO
CM CM CM
» « i < . t
LO LO LO 1
^j- ^- r^
O O
1 1 CM O
*y ro
O O O 1
CM CM CM
r*- r-» r*.
'
'
I
I
t
7^
_
i
it
000
0 «3- 0
*3- LO
1
o o o
^* ^f
1 1
O Lf> LO
yD vo
o o o
O O ^3-
O O O
CM CM f (
. < ^H ro
CM
l i i
O O O
» » .
CM CM CM
>
!
!
^
CD
&
0000
o o o o
«5 -^ CO O
0000
o o o o
in in LD
1 1 1
O Lf) LT> tn
o o o o
0000
\O VJ0 CM LO
vo cr> o
c\j ^j ro
i i i
000
o CM r^;
_ -x, ~
S-
3
3o
LU 3
C_ CT
H _i
o o o
CO ro O
CO *3- CM
^
O 0 0
LO LO
f I
O LO LO
o o o
co co ^r
CO CO LD
VjD
14
000
'.C LO '-O
000
ro ro oo
o
i i f-
«^r
CO
000
CM CM '
T^v. ^ ^C1
O*
cc
^ CM -0
O
o >-
UJ 3 ^
Q -1- O
7-26
-------
!
'
C/5
1
z:
- U
0 0
( y
5 *~*
c^
or
1
U.
o
^
'_J
^
nr
5
£
LU
LU
.a
i
r-^.
O)
'
^c
I
!
'
j
C
O
t/t
l/l
UJ
E?
c
UJ
o
c
o
I/}
0
',0
(/I
I
s
CJ
0
o
U
a.
1
1
4->
O
c *
o a.
i- = s.
m> >,
' >n ^~
E 01-
LU S- O
U
T3 C i
aiuj ja
J3 0
= s_
<-) C
o
CU
in S-
TO r~
Q. i-
= 0
O) J3
cu
c
(J
o -a
O3" Q)
S- 0
4^ *fi 3
c c -a
o o cu
01 S-
1
*J S-
O >i
(/> O
QJ
O f
O JO
^. ^3
t-
^
U= J2
5 """"
*J O
LO O
*
>^
£
>f*» >^
^ "^^
ij jr
<-) 3
CD >>
^ 3
fl j->
*^O
fO O
2T «"
>^ >
;_
-U *5j
/o c
i i-
3 ai
a; r
QT cC
, Jj
GJ =
"O fO
sr c_
H-
o o o
ro LTJ
l i
O O O
C^ O">
I 1
o t^) tn
o o o
in
i-H
1
1 t 1
^0
1 1 O
ro
( i t
. i c\j ro
<4-
Q.
ai en
go 0
o o
ro CM csj
ro ro ro
o o o
O 0 0
000
o o o
ro VjD ^
CO CM CM
ro ro ro
ID LO U^
000
XXX
ro ro ro
0 0
O X X
o
CM *? ^r
CT"
O O GO
O
« CM ro
1
5
-^^
LU
O-
cn
O O
0 0
ro \£>
CO CM
ro ro
0 O
0 0
0 0
o o
CO "^
ro CM
CO CO
ir> un
0 0
X X
<^>
0
O X
CM *3-
CP>
o o
« rsj
*t
^
--v^
H-
UJ
Q.
O
U
. en
3 QJ
*4- C
OJ
r^ 4_)
-«- OJ
J-J
Wl 4-
r~ Q
-a
ja
ji ^w
^
CM CD
CO
*r o
CM
CD
O
4J CJ
we c
cm ai
+J » 'TO *>, in
fl) > 1- 4-> .^
i i- 3 CU -C
10 3 O- 0)
> CT QJ H- S.
i flj O O
= :»>
cr >, ra j= >,
(U C7> * *" O*l
S- CD !.
>i ai c 3 cu
Cf> C 4> 4J c
t- CO CD CD
d) C
C C *C *^ QJ
0) TO r- jj
+-> CM fO
C 4J 0 - U
fO O C O f-
c CM -o
^ vi c
O (/> *'- " -
c ,- a;
+-> C c/l
if) 4J -i- OJ QJ
1- -^ r >
j-> .. ^ a! ^
f- >> U 0 >
uc: L.
- C QJ d. a)
>>
CO"*- O *J
QJ *^- "4- ns
U M- QJ r 0>
**- C 3
U- C O -M
QJ O CC
c ^ t. ro a; c
O S- OJ CO i -t-
t/i > c « re c:
S- C 0 C^ 3
O) C O r-t -t- O
> (J (t! i-
c >^ >
:_) rji C a; -*^
S- OJ = +J
>, OJ C O O OJ
C OJ O "" C 'c -0
a a> E QJ z <->
QJ un 53 ^ ^ u
jc LO a; i_ o a> re
"O 3 -"- * 4-J j_) c 5 C.'
QJ^-U < O &- >> t
3C*^ QJO +J+JJJ
i »- o a; aj o
U !^ C JD C 4->
r Q -kJ 1_ -«^. r O QJ t-'
4J (j QJ 3 QJ i- ^ ^
L. (j flQ M J^J VI QJ QJ "
C fO 4- QJ C£ fO > >
jj u- a: J2 o o >,
tj i/l * ' CO t_) w T3
«4- JE Lnaji-i."
JT 1 -C VI
wit r-vci-i_Qj
f- ^ 1-1 3 O O C
JT cr c «- ^ M-
h- .. aj r-
3 CM . § s* 3 £ i. i_i/i
i t i; P i. u tj
. i !| jj i P QJ C
i-^X t/iJ2C^^->-,C
O UD ^ ^ il i- IP
.p-^. QJ f c? <1)
r ii - o O-^-GJCU
*-> ^ I-H ^3 "c ~c
8^ ° ^ " i] .^ ^ ^ v,
* ^ !/l *J LJ O *^
da -C U ~a (ii u- ^n ^
7-27
-------
Table 7-6 shows that the energy impact of process VOC emission
reduction in a model plant generally increases in moving fron 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.
7.4.2 Industrywide Energy Impacts
Model plant energy impacts presented in Table 7-6 are used to
project industrywide energy impacts of the regulatory alternatives on
new polymers and resins plants expected to be built by the year 1988.
The industrywide energy impacts of process and fugitive VOC emissions
control in new plants operating under each regulatory alternative are
presented in Table 7-7.
Under Regulatory Alternative I, industrywide energy demand is
estimated to be about 1,418 TJ/yr (231,000 bbl crude oil/yr). Under
Regulatory Alternative II, industrywide energy demand is estimated to be
about 1,340 TJ/yr (218,000 bbl oil/yr). The total energy demand decreases
when shifting from baseline control (Regulatory Alternative I) to Regulatory
Alternative II, because implementation of fugitive VOC control practices
and increased recovery from process emissions in the polystyrene and
poly(ethylene terephthalate) plants saves about 118 TJ/yr (19,000 bbl
oil/yr). Energy impacts under Regulatory Alternative III are approximately
1,630 TJ/yr (265,000 bbl oil/yr); the increase is due to a greater
energy demand in controlling process emissions. Total energy impacts on
the industry under Regulatory Alternative IV are about 20 TJ/yr (3,000 bbl
oil/yr) less than Regulatory Alternative III levels because of the lower
energy demands to control process emissions from PP/Liquid plants.
For seven of the nine model plants, the incremental energy requirements
of the most stringent regulatory alternatives are either net decreases
or an increase of less than 0.1 percent of the energy required under
baseline (Regulatory Alternative 1) control. For the polypropylene, gas
phase and high density polyethylene solution process model plants, the
incremental percent increase associated with the most stringent regulatory
alternative over baseline control is large. However, the absolute size
of the energy requirement compared to the total energy used in making
7-28
-------
u en cu
, ''.
r^^T'ST'^' r-- CD CT ' *3" ' O r~-
:M ' 't \ u~. -"> -" -I
\ 's
1 !
; re cc co
CD cc co co ' ococo CD CM CM CM
ill; ii iff
t
t
1
f"! ' CM CNJ rv.
CM CM CO O ' rv. r^^ r^, «j- ^ CO O
CM CM CM *-* uT; Lfl vjT) ^C
1
! i
t >
f !
I 1
j
1 l
1
I
I
I
i
1 f
I
SSSSi ?Sg ^SS^'
1 i ' i^SS
i
; j
i i
i i
C00| CDO COCQCO
O CM CM C\J ' ^ CSJ CM O CM fM CM
1 1 1 ( 11 \ III
! ,
| ; --; -T un -o
\
i 1
1
1 J
t j
; ;
'
ssss : =. --? =TS in ||
i '
i i
coccco -nS "or->l cccc
CD^-^-r^- I Our;.n OCOco' CD'O'^r O C: ' OO
OJ Oj CM J If 1 I II J
1
! !
/ 1
1 1
1 I "^
1 f
' 1
1
O CM ro TJ ' j rj
i
1
1 I
i : '
ro in O O CM \ -J^ \ 3 !
7-29
-------
y, -o
D
-
^ n m o
-
un oo CM *
r^ *r ^ u
' O O%P
I. .»!"
T3 .3 OJ
T3 3
3 U -O
!_)
=^
-"or-
0) F
3 > 3 _) C
-3 JT ^a :
D *J -.
^- ~ -O
1) -C UJ
-s
7-30
-------
CM O c-*1 I O CM O
^ " ' 1 '
i
CM CM CM ' CM CM
. . . t .
0 o«
lit t 1
1
CM CM w un CM
r-.r-rv.r-t CM CM CO
r*"> f* CO C * * CM
1
t
l
1
tr o i
-MO J CD C2 O
j
1
!
1
|
j
'
|
"H i
3 '
CJ" LO
cT i
v^ CM CO O
o o o c:
un u^ LT:
CD O O O
1 1 1
^^CM ^
iO >^ P-. r-.
O O O O
u? CM CO C
U3 iO i£ r-.
O CD O O
O 0 0 0
t 1 1
^UDCM^
O O O O
3.
^.
_j
5
; ^og
!
'
0 C
o *r«s- rr o o o
, , , ,
t
'
1
CM CM O* en r- r-^ rv
Ol Uf7 UD LO ' «" « « --H
i-, _,_J ^ , )
1
t
1
1
J
CM r-*
-------
1
5
of
s
t-o
^
§
2£
>:
i
3
VI
s
i
UJ
0)
>- "C
ai ~
0 r-
-J O
a;
=
"~
55
;
£
3£
^
2
.-
3
S
1
^
l
u
CJ
1 '-
1 ~
5
(/i <_'
^
a
= i-
" ^ i
>,'
do !_>
= r- >
2 -*-1
Is 1
20 u-
iid
s
i
c
"^
t/i
1/1
OJ
J
a.
.e
°i
S» p
^Q_
_5 a;
ju
=
^^ i
u 5
^ >>
>>^
si o ' D
u >
^3 ! ^j
no r
-ao u_
SI *
'ij ' '
' 73
"" ^
0)
-j
' £
t
' >> QJ
- >
i ^2 "^
'C T3
,
CO -^ 00
OJ C\J CM CM
'
!
I
]
0 O 0 C
i I r
< 01 *o (^
ro CM f"-* (*
C^J CM CVJ CM
CM
1 1 1 1
1111
till
.
OJ
t QJ
!J O t" ^
c c »- c =
= 1J ZH-C 3
= -3 *3 C ^n
3 5 n -J 1J -D ;
7-32
-------
the polymer is small. For example, the HOPE, solution process model
plant would use approximately 23.8 x 106 Gg/yr (3.88 x 109 bbl oil/yr).15
Total energy consumed by process emission controls under the most stringent
alternative for this model plant is 102 TJ/yr (16.6 x 106 bbl oil/yr),
less than 0.5 percent of the energy required for the production of the
polymer.
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
1 f\ 17
injection. ' Noise levels in excess of 100 decibels (dB) have been
18
recorded within 100 meters (330 feet) of the flare tip, but attenuate
19
with increasing distance from the flare.
. 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. 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
19
noise (less than 500 HZ) as effectively as higher frequency noise.
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. Thus, Regulatory Alternative II appears
to require no irreversible and irretrievable commitment of resources
beyond what is currently committed under industry practices.
7-33
-------
Regulatory Alternatives III and IV 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 27 new plants to meet Regulatory Alternatives III and IV will be
significant in comparison to 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. Guywire-supported
flares and settling ponds for scrubber wastewater may require considerable
20-22
land space, 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 require-
nents of Regulatory Alternatives III and IV.
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 plants
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
4,340 mg/yr (4,790 tons/yr) in the absence of Regulatory Alternative II,
and as much as 14,640 mg/yr (16,220 tons/yr) in the absence of Regulatory
Alternative IV. 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
23
manufacturing by 20 to 65 percent.
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. 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.
7-34
-------
Table 7-7 indicates that a delay in implementing Regulatory
Alternative II may result in an adverse energy impact of 78 TJ/yr
(13,000 bbl crude oil/yr). However, delay in implementing Regulatory
Alternatives III and IV may result in energy savings of 213 to 192 TJ/yr
(34,000 to 31,000 bbl oil/yr) when compared to baseline control.
7-35
-------
7.6 REFERENCES
1. Fugitive Emission Sources of Organic CompoundsAdditional 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 .*
3.
4.
5.
6.
7.
8.
9.
10.
11.
Compilation of Air Pollution Emission Factors, Second Edition.
U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina. Publication No. AP-42. April 1977. pp. 1.4-1 and
1.4-2. Docket Reference Number .*
Reference 2, pp. 1.3-1 and 1.3-2. Docket Reference Number .*
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 .*
U.S. Environmental Protection Agency. Code of Federal Regulation.
Title 40, Chapter 1, Part 51. Washington, D.C. Office of the
Federal Register. November 25, 1971. Docket Reference
Number .*
[TO BE ADDED - Consumptive water requirements]
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. 147. Docket
Reference Number .*
U.S. Environmental Protection Agency. Code of Federal Regulations.
Title 40, Chapter 1, Part 261.30. Washington, D.C. Office of the
Federal Register. May 19, 1980. Docket Reference Number .*
Reference 2, pp. 1.3-3 and 1.3-4. Docket Reference Number .^
Perry, R.H. Chemical Engineering Manual, Third Edition. McGraw-Hill
Book Company, Incorporated, New York, N.Y. 1976. p. 3-18. Docket
Reference Number .*
U.S. Environmental Protection Agency. Code of Federal Regulations.
Title 40, Chapter 1, Part 261.4(b)(4). Washington, D.C. Office of
the Federal Register. Hay 19, 1980. Docket Reference Number .*
12. Reference 10, pp. 1-3 to 1-7. Docket Reference Number
7-36
-------
13. Petroleum Facts and Figures. American Petroleum Institute. Washington,
D.C. 1971. Docket Reference Number .*
14. Rossini, F.D., K.S. Pitzer, R.L. Arnett, R.M. Braun, and G.C. Pinental.
Selected Values of Physical and Thermodynamic Properties of Hydrocarbons
and Related Componds. American Petroleum Institute. Washington,
D.C. 1953. Docket Reference Number .*
15. McRae, Alexander and Janice L. Dudas. The Energy Source Book.
Aspen Systems Corporation. Germantown, Maryland. 1977. p. 441.
16. Shore, D. Towards Quieter Flaring. Flaregas Engineering Limited.
West Drayton, Middlesex, England. 1973. p. 2. Docket Reference
Number .*
17. Straite, J.F. Solving Flare-Noise Problems. National Air Oil
Burner Comapny, Incorporated. Philadelphia, PA. (Paper presented
at Inter-Noise 78. San Francisco, CA. May 8-10, 1978.) pp. 2-5.
Docket Reference Number .*
18. Oenbring, P.R., and T.R. Sifferman. Flare'Design ... Are Current
Methods Too Conservative? Hydrocarbon Processing. j>9_(5):127. May
1980. Docket Reference Number .*
19. Reference 16, p. 4. Docket Reference Number .*
20. U.S. Environmental Protection Agency. Background Information for
Proposed Standards or VOC Fugitive Emissions in Synthetic Organic
Chemicals Manufacturing Industry. Office of Air Quality Planning
and Standards. Research Triangle Park, North Carolina. Publication
Number EPA-450/3-80-033a. November 1980. p. 7-8. Docket Reference
Number Section 4.5. Docket Reference.*
21. 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 i.*
22. Nevril, 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 .*
23. [TO BE ADDED - Nationwide VOC emissions-uncontrolled]
24. Reference 20, p. 7-8. Docket Reference Number .*
References can be located in Docket Number A-82-19 at the U.S.
Environmental Protection Agency Library, Waterside flail, Washington O.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 each succesively
less stringent 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, requirements and costs of a leak detection
and repair (LDAR) program were developed for an assumed equipment cojnt
based on the analysis for control of fugitive emissions from the synthetic
2
organic chemical manufacturing industry (SOCMI).
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
8-1
-------
SUMMARY OF REGULATORY ALTERNATIVES FOR THE MODEL PLANTS
''ode! 31ant/ Process Sections Controlled *
Regulatory Alternative RMP PR MR 3F PS Other
Polypropylene, liquid
phase
1 (Baseline) F F
2 F F
3 C C C
4 C C C C
- jlypropylene, gas
snase
1 'Baseline) F
2 F
3 C C
LOW density,
polyethylene.
liquid pnase
1 \Saseline) F
2 F
3 C C
4 C 3 C
-;w density and iiign
Density polyethylene,
^as ohase
1 Baseline) F^ F^ F
2 F F r
3 3 C C
4 C C 3 3
JiiM density
polyethylene,
slurry process
Basel me: - -
1 F F
3 3 3 3
Hign Density
oolyethylene,
solution process
'. 3asalne) F F F
2 F F F
3 C C C C
Fugitive
Ernssiofi
Regulatory
Cont-ol
No
Yes
Yes
Yes
No
Yes
Yes
No
Yes
Yes
Yes
No
Yes
Yes
Yes
No
''es
"es
MO
Yes
Yes
Annual Emssion
'"3/y
5,060
5,120
5,500
5,510
:,230
1,290
3,365
100
150
330
435
3,455
"j , 2 I Z
3,520
3,525
2, "30
2, "50
2,34C
2,;oo
2,750
2,390
Reduction"
"ercent
39
90
97
97.2
31
32
97
10
17
39
45
95
36.5
96.:
96.3
g?
94
96
90
9?
96
8-2
-------
Table 3-1. SUMMARY OF REGUL5TORY ALTERNATIVES FOR THE MODEL PLANTS 'Concluded)
Model 3'ant/
Regj'atory Alternative
Polystyrene, continuous
process
1 (Baseline)
2
3
Poly(ethylene
tereprtthalate),
dinethyl
terephthalate
1 (Baseline)
9
3
Poly(ethylene
terephthalate) ,
terephthalic acid
1 (Baseline)
p
i o
Process Sections Controlled1'
RHP PR MR PF PS Other
R
R
R RR
R R
R RR
R RR C
R R
3R RR
Fugitive
Emission
Regulatory
Control
No
Yes
Yes
O.5
N.A.
N.A.
N.A.
n.A.
Annual Emssion
Reduc
Mg/yr
45
105
330
0
?75
995
0
970
i
tion"
Dercent
11
26
34
0
96
98
0
97
"Pricess Sections:
RMP - raw materials preparation ?R - polymerization reaction
(including esteriflcation PF - product finishing
in polyethylene terephthalate) PS - product storage
plants
jther - a1.1 emergency vents other than
from the reactor
"Control Devices:
- = flare.
C = conbustion devices including flares, incinerators, ooilers.
R = recovery devices used under baseline as part of process.
RR = additional recovery devices or more efficient recovery systen.
"-ron uncontrolled levels,
-lanng of intermittent streams only.
"".A. = not applicable.
8-3
-------
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-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 3-3.
Although fugitive control costs are in May 1980 rather than June
1980 dollars, the costs are considered, for all practical purposes, to
2
be on the same time basis as the ones for process VOC controls. The
fugitive control capital cost includes costs for fnonitoring instruments,
caps for open-ended lines, piping for the connections to an existing
enclosed combustion device or vapor recovery header, rupture disk
o
assemblies, closed-loop sampling connections, and initial leak repair.
The derivation of annualized labor, adninistrative, 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 ara also presented.
8-4
-------
I
1
C£
o
0
> t-
(O C
4-1 -r-
13 0
t_> C
»-.
0
1 i
0) C
1 0
c
i
0)
o
r^
I
3J
ll
-^ 0
-a o
i -4-*
*- ' 3
2 °
«r a.
^DOU^OOLD" LOLT^O^'OrOLDun '-jO^D
o cr)o omcNi^or-ioocsJ'--O'-^Lriroco cs;cj
^ ot- o o o o o c c o o o o o" o o ^ «r
^- r^ r^ ** =L
PS S i2 5S SS S -i§ oRjcsi S c\I
ooooooocoooooo mro
^c ^
O Cf~1 C
ex 5. OJ i- o
i- - o^H -~ d
*J 3CL-u» t/i
S C"u_oJ i- -tj c/»cj
*»~ Q_ U C k/li (/> £Z 1) ^^ f 5 ZZ tj
3" M -*j oJ 4J3D *j i-H''~H-*jtjajcri
i_ .^ 5]t2 ^2 "5 -StjIL c1^ ^ ? i ^ 2 .S
O»- 2. r i ' C 3 <» C 3J -^ _: 3J -r ^-) -^ iJ
fQ C ^ *?" i O *J "*" C ^ <" C r~ ftt O C O ^
= = C^^ Lj-t-OLtJ^ C.U. UJC^LU^LUO ^
u
c
(13
3
O
(TJ
C
(3
o
a)
i
3
a
«
"2
<^
0
s.
r
a
CJ
r '
rj^
OJ
r
5"
1J
^
>
'^
a
=
j_i
^
>>
n
o
0)
S-
r
u
^
*\
c.
V
>)
1)
2
iJ
S
^
D
<
8-5
-------
TABLE 8-3. ANNUALIZED COST FACTORS FOR POLYMERS AND RESINS NSPS
(June 1980 Dollars)
Direct Cost Factors
Operating labor price: $18/hr (including overhead)
Operating labor requirements (including supervisory labor):
= 620 labor hours/yr for flare c
= 1200 labor hours/yr for thermal incinerator without heat recovery
= 1500 labor hours/yr for thermal incinerator with heat recovery b
= 620 labor hours/yr for catalytic incinerator without heat recovery
= 930 labor hours/yr for catalytic incinerator with heat recovery
= 60 labor hours/yr for condenser ^
= 4300 labor hours/yr for ethylene glycol recovery baseline system
= 8600 labor hours/yr for ethylene glycol recovery regulatory alternative
system^
Electricity price: $0.049/kwh
Natural gas price: $5.67/GJ
Steam price: $13.62/Mg ($6.18/1000 lb)J
Water price: $0.079/m3 (SO.30/1000 gal)k
Styrene recovery credit: $0.788/kg
Ethylene glycol recovery credit: $0.33/kg ($0.15/lb), for 80 percent purity
Ethylene glycol recovery credit: $0.60/kg ($0.27/lb), for high purity11
Indirect Cost Factors
Interest rates:
8.5 percent (after taxes)
10 percent (before taxes)
Equipment life, N:°
15 years for flare, ethylene glycol recovery system
10 years for thermal incinerator, catalytic incinerator, condenser,
piping
Capital recovery charge factor =
(1 + D -1
8-6
-------
TABLE 8-3. ANNUALIZED COST FACTORS FOR POLYMERS AND RESINS NSPS
(June 1980 Dollars)
(Concluded)
Indirect Cost Factors (Con't)
0.131 for flare
0.163 for thermal incinerator, catalytic incinerator, condenser,
ethylene glycol recovery system, piping
0.244 overall for fugitive (varies with individual equipment
components)
Taxes, insurance, and administration: 0.04 x Total installed capital
costp
Maintenance cost: 0.05 x Total installed capital cost^
0.085 x Total installed capital cost, overall for
fugitive VOC control (varies from zero to estimates
of actual materials costs, although the 0.05 factor
is used in general)
Operating hours: 8600 hours/yr
Includes wages plus 40 percent for labor-related administrative and
overhead costs.
0.5 man-hours/shift x 8600 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. Environmental
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 -f 8 hrs/shift + 15 percent of the
operating labor for supervisory costs.
"1 man-hour/week x 8600 hrs/yr * 3 hrs/:
of operating labor for supervisory costs.
Q
1 man-hour/week x 8600 hrs/yr * 3 hrs/shift * 21 shifts/week + 15 percent
f4 nan-hours/shift x 8600 hrs/yr f 8 hrs/shift.
^Number of nan-hours/year provided by an industry source.
Memo from Chasko and Porter, E!
developing CTGD Cost Chapters.
Memo from Chasko and Porter, EPA, September 17, 1980. Guidance for
8-7
-------
from AT 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.
By linear interpolation between $4.91 and $5.55/MMBtu; 1988 price
in 1978 dollars = $5.29/MMBtu.
Using GNP implicit price deflator index: 4th 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).
JNeveri11, 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.
p. 3-12:
$5.04/1000 Ib steam, 4th quarter 1977.
Using GNP implicit price deflator index: 4th quarter 1977 of 142.91
and 2nd quarter 1980 of 175.28; updated steam price = 175.28/142.91
x $5.04 = $6.13/1000 Ib steam.
i/
Peters, M.S. and K.D. Timmerhaus. Plant Design and Economics for
Chemical Engineers. McGraw-Hill Book Co. New York, N.Y. Third Edition.
T91KT:p. s8l.
90 percent of styrene price given in Chemical Marketing Reporter
nAssumed based on polymer raw material (high purity) ethylene glycol price
of 27?/lb.
nPolymer raw material grade of $0.27/lb of ethylene glycol given in
Chemical Marketing Reporter.
°Average equipment lives given by Neverill in reference cited in i.,
p. 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. p. 5-16.
reference cited in footnote n:
9 percent of total installed capital costs for maintenance
and miscellaneous charges - 4 percent of total installed capital
costs for taxes, insurance and administration equivalent to
miscellaneous).
8-8
-------
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
to functions of the following waste gas characteristics: volumetric
flow rate, lower heating value, temperature, and molecular weight. 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
2 4\
Btu/ft hr, including solar radiation ) at the base of the flare
(considering the effect of wind). Heights in 5-foot multiples with a
5
minimum of 30 ft. were used.
Supplemental fuel, natural gas, is added to increase the heating
value to 115 Btu/scf to ensure combustion. 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. For flares handling
only intermittent flows, purge gas requirements were assumed to be
negligible according to the industry practice of not purging or perhaps
o
purging befora 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. In
such cases, a fluidic seal , which requires a greatly reduced purge rate,
was used.
8-9
-------
Natural gas consumption at a rate of 30 scfh per pilot flame to
ensure ignition and combustion was assumed. The number of pilots was
based on diameter according to available commercial equipment.
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 lb steam/1b of hydrocarbons (VOC plus methane
and ethane) in the continuous stream (or the intermittent stream if no
continuous flow was present). Availability and deliverability of this
quantity of steam was assumed.
Piping (for flows less than 700 scfm) or ducting (for flows equal
to or greater than 700 scfn) 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, 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-ft , including solar radiation1^). For flows less than
700 scfm, an economic pipe diameter was calculated based on an equation
13
in the Chemical Engineer's Handbook and simplified as suggested by
Chontos. ' * 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 acfn. 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
g
provided by National Air Oil Burner, Inc., (NAO) during November 1982.
A cost was also provided for one additional case of 60 inch diameter and
40 feet height. 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
8-10
-------
greater. Flare purchase costs were estimated for the various regulatory
alternatives by either choosing the value provided for the required
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
at
19
1 8
analysis of the Distillation NSPS. Ducting costs were calculated
based on the installed cost equations given in the GARD Manual.
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).20 The design procedure is outlined in this section.
In order to prevent an explosion hazard and satisfy insurance
requirements, dilution air was 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) and an
oxygen concentration of 12 percent or greater by volume. Dilution air
was added to reduce the lower heating value of the stream to below
8-11
-------
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 assuming 18 percent
excess air for required combustion air, but 0 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. Minimum auxiliary fuel requirements for low heating value
streams were set at 5 Btu/scf to ensure flame stability.
For streams able to maintain combustion at 870°C (1600°F), 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
... 22
tnviroscience.
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
tine of 0.75 sec, conservatively oversizing by 5 percent according to
23
standard industry practice. The design procedure assumed a minimum
commercially available size of 1.01 m (35.7 ft ) based on vendor
24 3
information and a maximum shop-assembled unit size of 205 m
(7,238 ft3).25
The design procedure would 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
8-12
-------
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 870°C
(1600°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.)
3.1.2.2 Thermal Incinerator Costing. Thermal incinerator purchase
costs were taken directly from the IT Enviroscience graph for the
23
calculated combustion chamber volume. (Essentially equivalent purchase
20
costs would be obtained by using data from the GARD manual. ) An
installation cost factor of 4.0 was used based on the Enviroscience
IJC
document (see Table 8-2). The installed cost of one 150-ft. duct to
the incinerator and its associated fan and stack were also taken directly
27
from the IT Enviroscience study. A minimum 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 costed 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
28
from an equation developed for the Distillation NSPS cost analysis.
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
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.
8-13
-------
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 VOC gas
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; 65 percent was assumed for this analysis. Gas temperature,
preheater temperature, gas dew point temperature, and gas VOC content
determine the maximum possible 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
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
3 ?9
2.25 ft /I,000 scfm was assumed for 98 percent efficiency." Catalyst
replacement every three years was assumed.
8-14
-------
8.1.3.2 Catalytic Incinerator Costing. Calculations for capital
cost estimates were based on equipment purchase costs obtained fron
vendors for all basic components and the application of direct and
29 30 31
indirect cost factors. ' ' 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 scfn 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 scfn can be
estimated by considering two equal volume units in the system. 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 Condenser Design and Cost Basis
This section outlines the procedures used for sizing and estimating
the costs of surface condenser systems applied to the gaseous streams
fron the continuous process polystyrene model plant.
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,
which condenses the vapors by isothermal condensation, 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 strean
and the coolant.
Total heat load was calculated using the following procedure: the
system condensation temperature was detarmined from the total pressure
of the gas and vapor pressure data for styrene and steam. As the vapor
8-15
-------
pressure data are not readily available, the condensation temperature
was estimated by trial-and-error 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 contents of styrene and steam. The coolant is
selected based on the condensation temperature.
For our purpose, no detailed calculations were made to determine
the individual and overall heat transfer coefficients. Since the streams
under consideration contain low amounts of styrene, the overall heat
transfer coefficient was estimated based on published data for steam.
Then the total heat transfer area was calculated from the known values
of total heat loads and overall heat transfer coefficient using Fourier's
general equation.
8.1.4.2 Surface Condenser Costing. Since the gas volumes of the
two streams are very low, the calculated heat transfer areas are also
o
very low (about 10 ft ). The heat exchanger costs for each stream were
obtained for minimum available size of 20 ft from vendors. ' ' '
An installation factor of 1.39 (See Table 8-2) was used to estimate
installed condenser costs. Mo additional piping was costed since the
condenser unit is so small ( 1-2 ft. diameter) that it should be able to
be installed adjacent to the source.
8.1.5 Ethylene Glycol Recovery System Design and Cost Basis
This section briefly describes the procedures used to estimate the
cost for recovering ethylene glycol (EG) using a baseline system and
using an alternative ethylene glycol recovery system that is composed of
an ethylene glycol spray condenser and associated equipment.
8.1.5.1 Ethylene Glycol Recovery System Design. The equipment
selected to comprise the ethylene glycol spray condenser and both recovery
systems was obtained from information provided by industry sources. The
baseline system recovers ethylene golycol from the downstream of the
cooling water tower. Ethylene glycol emissions from the polymerization
reactor and from the distillation column recovering EG from the esterifier
emissions accumulate in the cooling water and are emitted from the
8-16
-------
cooling tower. The regulatory alternative system recovers ethylene
glycol from the polymerization reactor through use of an EG spray condenser
and recovers ethylene glycol from the esterifier through use of a reflux
condenser. Thus, very little EG is emitted to be collected in the
cooling water and later emitted to the atmosphere from the cooling water
tower. Industry sources also provided basic operating and naintenance
parameters. The design and operating parameters for the baseline system
were used to design a system for the model plant. 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 DMT
process can be as much as 35 percent less than that from a TPA process;
however, differences depend on the completeness of the polymerization
reaction and the part of the process where VOC is enitted. Both of
these factors are plant-specific.)
8.1.5.2 EG Recovery System Costing. The costs of the baseline
systan were estimated based on the preliminary design using standard
engineering procedures as well as some industry information. The costs
of the ethylene glycol spray condenser and recovery system for new
plants using the OUT 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. 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
oe consistent with other fugitive VOC emission regulations for the
petroleum refining industry and the synthetic organic chemical manufacturing
industry.
The following sections outline the specific requirements of the
regulatory alternative and the procedures used to estimate its capital
and annual costs. Detailed descriptions are found in Appendix F.
8-17
-------
8.1.6.1 Design of Fugitive VOC Regulatory Alternative. The following
equipment in the polymers and resins industry were considered for regulation:
in-line 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 are a combination of Regulatory Alternatives III
2
and V for VOC fugitive emissions in the petroleum refining industry.
The technical parameters for the polymers and resins model plant
for fugitive emissions are given in Table 8-5.
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 BID for Model Unit B. 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 parts describe the cost estimation procedures and
assumptions used to derive these costs as well as more detailed cost
tables for each fugitive VOC emission source.
Valv es
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) nonitoring 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.
8-18
-------
Table 3-4. FUGITIVE VOC REGULATORY ALTERNATIVE CONTROL
SPECIFICATIONS
Control Specification
Emission source
Inspection/
moni toring
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
Hanges
Monthly
Monthly
None
Monthly^
Weekly Visual
None
None
None
None
None
None
None
None
None
None
None
Rupture disks on
relief valves
Cap
Controlled
degassing
vents
Closed-purge
sanpling
None
The regulatory alternative is a combination of Regulatory Alternatives V
and III for VOC Fugitive Emissions in Petroleum Refining Industry -
Background Information.
Fraction of sources found to be leaking by monitoring would be repaired.
GFor 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.
3-19
-------
_,-
A
00
^~
i
UJ
5
r:
Oi
0.
'
00
2
oo
UJ
o
?»
^
on
a;
L-l
i
o
CX
Z
G/^
UJ
o
T3
o
UJ
zz
H-
a:
o
^
Q
O
'^0
r_'1
y*
'-U
£_j
=0
"3 '
C C
r O VI i- S-
ro-i- C 4-1 >,
3 VI O C ^
c i/i o cn
C 4-J U IZ
«: S u
0) 3 O
15
s-
-o
o
oj vi
i i = i.
<0 i O >,
3 O -^
C i- in 01
C 4-1 VI E
< c -
o =
S s
>,
u
o ^
t. >
*J U
c *-
O M-
O 4-
r-
c 4~>
O 0)
0 S
"O
CDO
i (/I
0 C S.
13 i. O >,
3 4J ! ..
c c vi cn
C O V) E
< U -i-
C E
3 01
«
i.
o
ai u
V
_ i) ^ CU
o *- u
!- C JS I.
O *-> O v. 3
-> C * 01 C
O vi -* vi
UJ U
~
CD
3
-S '
CU
J-L. ('
'1
au
-yi
CO
s- o>
O) U
ja s.
S 3
3J 3 O
< 2Z c/l
.a
a
i
O CO
*r CM
aj o- o
n c^> co
IT) CO «-H
m en
f^- in i
i
o c
c
0
(J to
CD a.
4-> O)
u '
c
a
^ c
UJ 10
01
CM CTl lO
r~ in o
in CM ^H
n
l£3 O CM
uo r- o
000
o o o
o' o' o"
CM "sr ^f
O CM CM
i
t
'
c
o u
~ u
! vi i.
VI 3
r- C
LU
a -a
3 '3
cr a-
vi ^* >
OJ m en fn
> ,T3 - 0)
T3
^»
-^ CM
1C O
r-^ o O
CM CM
CT! '-C
^" m O
^
10 i o
i
o ^*
c
0
.,_ ^_ ^£
4-J -r- 1.1
a? a. -o
^j gj
OJ 1L. (U
TT t.
O 3
^4 C *-*
13 ro Q.
OJ 3
-j cr:
LO CM CM
*^> ^O O
c-j tn o
f-H ] *
a- « O
CJ^ i"-" *f
^* CM C
O O
0 0 O
CTi O > *
CM CO I
,/)
OJ
r
^
T3 -3
33 U
Q- -r
_, r- 0)
"O S-
tn -C :> >-i
^^.p QJ 3J -Q
Q- I/I
O^ CM ' 0s
o> ^a- <~c
o on o? o co
UD
»
o
VI
VI
t=
Uj
1 ,
vim
in[ m
O O o r- r^
O O O 1
. . 1
VI
o.
c
VI Ol OJ OJ VI O
C71 4J S- VI >, u
3 C fQ O VJ
r 01 O
a. => o e
CTI CM in CM
CJ1 ^* ^* ^f
O CO CO 1^ IO
0
CO
~- O CO
i4 CC i^J O
O CM rt O
O CM O O
o o o o
O CM ys o
CM O
CM
VI
Q
r
in 4->
OJ u
~ -
^ c
vl O
T3 S. O
OJ O
~ yl Cl
C '^1 C VI I
OJ O) OJ T3
i i. i en 4-1
sr a. a. c o
C 1
0 CC
VI C
UJ CM
1
C CC
o
= 0.
o
4J1
2 CM
C CT.
0 <
C i
s-
^ «c
o
4-iO
5 o
< CM
1 CO
i i
in m
-o
c o
3 WO
O "S-
a. i
' E <£
5 Q.
U UJ
U 1-
r- OJ
C J2
"3 S
s^ S.
0
ei <4- s.
00
CO CL
VI OJ
t < qi fy
y ,
OO U
0) C
J= COJ
4-1 O Ol
«£
4- in
O vl C
r*> O
ta S "
UJ 4«1
r U
CJ OJ OJ
5 - "o
<- Q.
U. -S
OJ = 4-1
I «1
i; o 5
j= a. s.
4J CO >r-
S- >
0) C
ja cu LU
o *->
4-1 1^1
c; *
T3 S =J
OJ !.
= t-
^ 2
Jl -O
VI OJ V»
0 =4-1
*- VI
VJ V 0
4J O-3
= oj 5
3 I-
^O L/l
CJ T5 O
~ ^ '-J
rg jj -, j
~ ,9 ,5 1~ J3" -T" ^
,_>
OJ
OJ
VI
OJ
u
3
O
VI
o
1)
j2
o
OJ
_,
^J
o
i.
u
0)
c
o
c
L/1
L.
3
O
g7
O
&
A
cc
c
-3
OJ
./I
c
-^
t
CO
CU
-O
I
cy
OJ
,
CU
.>
4_>
(IS
OJ
4->
^.
o
^
^_-
3
cu
v
j_J
_^7
T3
0>
^
-U
aj
8-20
-------
Table 3-6. FUGITIVE VOC REGULATORY ALTERNATIVE COSTS
FOR POLYMERS AND RESINS MODEL UNITS
Cost item May 1980 dollars
1. Installed capital cost 66,800
2. Annual cost of operation
a) Operating labor including
administration and support 20,700
b) Maintenance 5,700
c) Miscellaneous3 2,600
d) Annualized capital costs 16,300
Total 45,300
3. VOC recovery credits5 (29,600)c
4. Net annualized cost 15,700
aTaxes, insurance, and administration.
Based on 56.12 fig of VOC recovered annually at a credit of $528 per Mg.
(This amount, 56.12 Mg, of VOC recovered represents 96 percent of the
total VOC reduction under the fugitive regulatory alternative, i.e., a
96 percent recovery efficiency is assumed.)
Values in parentheses denote credits.
8-21
-------
CO
1
CO
o
o
_J
o z
ce rs
i
2: _j
O LU
Q
o
CO
CO Z
CO -*
< CO
UJ
LU cr.
V)
0 C. S-
O Z fO
^
LU CO O
t LU
H- S O
>- CO
o i 01
U. C.
LL. z io1
0 S
>- CO
CU LU
S CC
S 0
CO CO
LU
T
u
^
0
c
o
VI
.,_
1
<_,
o
_
'O
o
1
^ a;
-ii O Q.
(T3 4_) .f
OJ 3
_i c cr
o co
V)
c
C7> O
C
t 4->
i U
CL OJ
tj C
CO O
u
V)
o
VI
V)
OJ
a.
§
CJ
>> cu aj
o £- 'cu *fo
2? *" *
4-> -O VI
^» -r~ O,^~
CT> 3 = «
i-
VI '^ '^ fQ
C3
~
5
£
VI
o
o
o
o
CO
'*O
UD
0
O
CM
O°t
o
o
co
oo
t 4
0
o
*f
viT
o
o
f^
CO
oo
0
o
*-*
g
CO
CM
4J
VI
O
f_
rT3
>t«
a.
(j
-a
u
_
T3
4-1
VI
^
^
(/I
O
^
.^
r>
U
at
NJ
^n
^
3
C
C\J
O O
0 0
r^ LO
UD
0
O
CM
^f
O
0
CM
CM
0
O
0
^
S °
^o
<*o
0 G
0 0
^O CM
<
0 0
o
.,_
4_) T3
c a.
OJ OJ
I. u
3 TO
CT ^J
Ol r
p.. ^^
o «
o c
'.'^ ^-^
C3 O O O
o o o o
co en r~~ o
«3- LD LO CM
O O
0 0
O ^t"
ro
0 0
O O
O O
0 0
co ro
0 0
0 0
f^ CO
*-4 ~-
0 O
0 0
r*^ CT>
«« it
rT r~l
o o
o o
o *r
4->
o
a.
a.
CO
O S.
u o -a c
^ CO '/I
CZ f~~ S^ 4J CU O
f- OV-- « O CU
4J CT C 10 i- CZ C
n3 C !- CL 4J 10 «
S- *- i- CD V) C r
U 4-> OS--'- CU "
5" S. -t- ^£ .^ c 0
i a. o cu -5 10 i-
"3 O S f cc S S
3
£3 ^~« < ^-^ **,
r 4J C_J
fQ U 1>
C -4- M-
e aj o
T3
4J Ol
+* >" ^"
OJ O
,0 r^
J
"O
O)
i.
u
OJ
o
QJ
T3
VI
OJ
^
£
aj
SI
H
r-
OJ
^
a
8-22
-------
(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.
2
Based on the results of the leak detection and repair (LDAR) model,
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 BID.37
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 3-12 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
BID.
Safety/Relief Valves
The fugitive VOC emission control specifications for safety/relief
valves include installation of rupture disk systems on the relief valves.
8-23
-------
Table 3-8. 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 $13/hour
Adinin. & 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
aCapital recovery factor is 0.163 based on initial leak repair costs
amortized over 10 years at 10 percent interest.
8-24
-------
u.
=
J
i
-
»-
K
£
1/5
LU
>
O £
LL. eg
t.
oo o>
t o
^£ !i_
uj a.
UJ i.
C£ -r-
i f^
ZD O.
c- a)
UJ Q£
Q£
=e c
O tl
CD
«c c
_1 0
OS »->
i i U
TJ
i . t ^
v- _
! .-i
'o a) o
0. £ -C
OJ 1
Q£ *J C
2
-~
i/)
V"
« !-
a.1
CO uo fn
IT) O
r-» r^-
^-, ^H
c\j rsj
i
!
!
4J 4-9
C =
Q; 1J
3 5
L. 1-
4J 4J
VI LO
C C
I
;
C\J f F I
O CM T3 f
<3- LH -4-*
c
1
! "*^
' IB
1 =
*
c
; °
! =
; s
\ ?
-
i o
1
' ^
«u
1 ^J
I_
2
0
l/l
(4^
0
c
o
4J
o
^
i.
*«-
i §
I
5
!/l
fl
J2
"
a>
^J
o
e
Q_
0£
C
cu
S j=
(T3 4_>
^J
*J CT
^ 0
i ^
1 CM
CV1
^3
(0 C
^
t~ o
o s_
I*- T3
u C
*J 3
3 *J
c
= u
^ QJ
>
-- 0
c/l "^;
| J
-^
^1 ^3
C 4J
^
O '-w
*J 0
'c 5-
a 11
= -s
c S c
v a)
E; U _J
^n s-
S- -D OJ
u '-a.
yi OJ
c > a>
i
-------
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, $ 4,050
Annualjjzed charge for initial leak
repair0, $ 370
Total annual costs, $ 14,550
Annual product recovery credit,0 $ (17,750)
Net annualized costs, $ ($3,200)
Cost Effectiveness, $/Mg (95)
aTotal 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.
8-26
-------
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,
nan- 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, 750
at $18/hour
Admin. & Support costs, 40 percent of
labor charges 300
Seal costs, $139.4/single seala'b 360
Total Costs 1,410
Annualized charges for initial leak
repair, 16.3 percent of total costs 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
GCapita1 recovery factor is 0.163 based on initial leak repair costs
amortized over 10 years at 10 percent interest.
8-27
-------
1
1 H
z.
_J
LU
Q
O
Z^
LU
^~
1
U_
o
oo
1
«=C
LU
oo
CL
2Z
ZD
Q.
>~-%
<*v* E
O "3
U_ '_
CD
oo o
t i.
Z Q.
LU
s: s-
LU -r-
t*^ O3
<> a.
Z3 V
O" O£
LU
ct: -o
c
QC 03
0
co c
ce o
H4 QJ
Q. 0)
LU O
rf"*^
^
i»
Q r
Z -C
03
r ^
*4
c s-
QJ
Q.
O
Z.
C7> "CO
c -as.
r- S_ O) 3
S- 0 S_ 0
»-> (O 3 1
i- r 0- C
C 0) 03
O 1- S
S
-o s-
0) IB
S- OJ
01 0 >,
OJ +J
E s.
i- C d>
1 0 Q.
E
CD
S= 03
f C"
S_ " -^
O O) =
+J E i
r- -I- C
e !-> us
O E
en
c
4- !-
O S-
O
O) -M
Q.-I-
>> «=
1 0
Ez
CL
C^ 4^
3 i- -T-
CX O) C
0. 3
M_
O co t
i Ol
fQ "^
O 0) 0
Z co E
1
CO
CO*
CO
1 I
^o
^~H
CO
t
f-H
o LO tr>
00 CM O
Lf) ^-11^-
CM CM
t i un
LO
*
0 0
_
4>}
c
OJ
3 i
1- 03
+-> 3
CO co
C *-
^ « ^>
CT>
CM
T3
J S
3
Q.
c
0
CD
S-
O)
co co
QJ GJ
£ 1c
.p
* *
O CO
S- 0
O
0 .=
<4- -M
O
4-) .O
.^.
T3 C
OJ T-
O "^3
4-) T3
0) F
U CJ
i- C
E cu -t
03 CL
O) O!
4-> r-l S-
. m
C CM 3
E co co
1 CO Q.
CM 0) E
. 3
03 a.
OJ
S- +-> C-)
O 03 C
l(_ S_ ,
CO 0)03
01 o a>
3 S '
£= S_ 0_
r- V 0
C 3
cj co
un u c
0 0
CO !
^ ^^ 4~)
r- U
O) J= 03
E 4-> S-
i- C LU
4~* O
^
CD
C +J t
t- CO)
t- 0> T3
o u o
4-> i- E
r- 0)
C O-CC
0 cz _c
E 0 4J
3
S- T5 CD
4-J DC
CO CO -r-
c:
-------
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, $ 4,670
Admin. & Support costs at 40 percent of
labor charges, $ 1,870
Annualized charge for initial leak
repair, $ 230
Annual replacement cost of leaking
seals, $ 1.650
Total annual costs $8,420
Annual product recovery credit, $ (4,030)e
Net annualized cost, $ 4,390
Cost effectiveness, $/Mg 575
aTotal of 70.5 monitoring hours and 188.8 leak repair labor hours from
Table 8-14.
bFrom Table 8-13.
C8ased on $139.4/seal and 11.8 leaking seals per year obtained from
Table 3-14.
Product recovery credit is calculated at $528/Mg. The emission
reductions obtained are 7.63 Mg/yr from pump seals (see Table 8-5).
g
Figures in parentheses indicate a credit.
8-29
-------
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
OQ
based on estimates from the Hydroscience (now IT Enviroscience) report.
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
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 3-16. The total
estimated installed cost of a new rupture disk system with block valve
was $1,995 in second quarter 1930. 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
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.
8-30
-------
Table 8-16. RELIEF VALVE CONTROL COSTS FOR RUPTURE DISK SYSTEMS
WITH BLOCK VALVES AND THREE-WAY VALVES
(May 1980 dollars)
Capital costs
(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.
Total $/year
(2) Rupture disk systems with 3-way valve.
Rupture disk and assembly
One 7.6 cm stainless steel rupture disk
One 7.6 cm carbon steel disk holder
One 0.6 cm dial face pressure guage
One 0.6 cm carbon steel bleed valve
One 7.6 cm safety/relief valve
Two 7.6 cm elbows
One 10.2 cm tee and one 10.2 cm elbow
Subtotal
Three-way valve
One 7.6 cm, 3-way, 2 port valve
Total Installation
Subtotal
Total cost for second quarter 1980
Rupture disk at 58 percent of total costs'
Assembly at 16.3 percent of total costs
Maintenance & Miscellaneous.
Total ($/year)
230
384
30
700
21
612
$1,995
Annualized costs
$133
288
180
$601
Capital costs
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.
C3ased on 9 percent of total capital costs.
8-31
-------
Table 8-17. CAPITAL AND NET ANNUALIZED COSTS FOR CONTROL OF
EMISSIONS FROM SAFETY/RELIEF VALVES FOR THE MODEL UNIT
(Hay 1980 Dollars)
Costs Rupture disk system9
Installed capital cost,b $ 33,730
Annualized costs,0 $ 6,550
Annual operating cost, $
Maintenance at 5 percent of capital cost, $ 1,690
Miscellaneous at 4 percent of capital cost, $ 1,350
Total annual cost, $ 9,590
Annual product recovery credit, $ (5,290)
Net annualized costs, $ 4,300
Cost effectiveness, $/Mg 430
aOne half of the total systems with block valves and the remaining
half with 3-way valves.
bBased 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).
C8ased 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).
8-32
-------
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
00
Hydroscience (now ITE) report and was for the following items per
comoressor:
122 m length of 5.1-cm diameter schedule 40 $6,400
carbon steel pipe
Three 5.1-cn cast steel plug valves and one $1,600
metal gauge flame arrestor
Total for second quarter 1980 $8,000
The above costs include connection of the degassing reservoir to an
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 8-18.
Sampling Systems
Equipment costs were computed for closed loop sampling connections.
The cost estimates were based on information from the Hydroscience
38
(now ITE) report and was for the following item per sampling system.
Table 8-19 presents capital and annual costs for control of emissions
from sampling systems in the model plant.
One 6-m length of 25-cm diameter schedule 40 $210
carbon steel pipe and three 2.5-cm carbon
steel ball valves.
Installation (at 18 hours and $18/hr) $320
Total cost (second quarter 1980 dollars) $530
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 3-20 to
8-28, for the nine model plants. Tables 8-29 to 8-37 show the costs and
8-33
-------
Table 3-18. 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 $ 6,400
Annualized cost at 16.3 percent of
capital cost,c $ 1,040
Annual operating cost for
Maintenance at 5% of capital cost, $ 320
Miscellaneous at 4% of capital cost, $ 260
Total annual cost, $ 1,620
Annual product recovery credit ' (850)e
Net annualized costs, $ 770
Cost effectiveness, $/Mg 485
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.)
Capital cost per compressor is $8,000.
cCapital recovery factor, 0.163, is based on the equipment amortized
over 10 years at 10 percent interest.
Product credit is calculated based on $528/Mg and emission credit of
1.6 Mg/yr (i.e., 40 percent of the 3.99 Mg/yr from the model fron
Table 8-5).
eValues within parentheses denote credit.
8-34
-------
Table 8-19. CAPITAL AND NET ANNUALIZED COSTS FOR CONTROL
OF EMISSIONS FROM SAMPLING SYSTEMS IN MODEL UNIT
(flay 1980 Dollars)
Closed loop sampling
Costs connections
Installed capital cost,3 $ 13,800
Annualized cost, at 16.3 percent of
capital cost, $ 2,250
Annual operating costs for
Maintenance at 5% of capital cost, $ 690
Miscellaneous at 4% of capital cost, $ 550
Total annual cost, $ 3,490
Annual product recovery credit, $ (1,800)
Net annualized costs, $ 1,690
Cost effectiveness, $/Mg 520
Based on $530 per sampling system and 26 sampling systems in the
model plant.
Capital recovery factor of 16.3 percent, is based on the equipment
amortized over 10 years at 10 percent interest.
cProduct recovery credit is calculated based on $528/Mg and 3.24 Mg/yr
of VOC recovered annually (see Table 8-5).
Values within parentheses denote credit.
8-35
-------
Table 3-20. POLYPROPYLENE, LIQUID PHASE MODEL PLANT REGULATORY
ALTERNATIVES COSTS (JUNE 1980 DOLLARS)
Regulatory Alternative
INSTALLED CAPITAL COST ($)
Flare(s)
Flare Ducting
Thermal Incinerator
Thernal Incinerator Ducts,
Fans & Stack
Catalytic Incinerator
Catalytic Incinerator Ducts,
Fans & Stack
Condenser
Fugitive Leak Detection
and Repair (LDAR)
Total
ANNUAL I ZED COST ($/yr)
Direct
Operating Labor
Operating
Materials
(e.g., catalyst)
Maintenance
Materials and
Labor
Utilities
Natural gas
Electricity
Steam
Subtotal
Indirect
Taxes, Insurance. &
Administration
Capital Recovery
Subtotal
Recovery Credit
Total (Direct +
Indirect - Credit)0
12,500 12,500 12,500
36,400 36,400 36,400
213,000 368,000
95,200 125,000
66,300 66,300 66,800
49,000 116,000 429,000 561,000
11,200 31,900 53,500 42,300
2,450 8,150 23,800 30,400
4,280 4,280 4,290
210 1,620
27,800 27,300 27,300
45,700 72,100 110,000 74,300
1,960
7,580
4,560
23,900
9,540 23,400
29,600
17,100 19,800
74,300 96,600
91,900 116,000
29,600 29,600
55,200 70,900 171,800 161,100
alncludes miscellaneous operating costs for fugitive leak detection
and repair program.
Some totals do not add up exactly due to rounding.
5-36
-------
Table 8-21. POLYPROPYLENE, GAS PHASE MODEL PLANT REGULATORY
ALTERNATIVES COSTS (JUNE i960 DOLLARS)
Regulatory Alternative
1
INSTALLED CAPITAL COST ($}
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 ($/yr)
Direct
Operating Labor
Operating
Materials
(e.g., catalyst)
Maintenance
Materials and
199,000 199,000 206,000
166,000 166,000 181,000
66,800 56,800
366,000 432,000 454,000
11,200 31,900 31,900
Laoor
Utilities
Natural gas
Electricity
Steam
Subtotal
Indirect
Taxes, Insurance, &
Adninistration
Capital Recovery
Subtotal
Recovery Credit
Total (Direct +
Indirect - Credit)
18,300
12,800
740
43,000
14,500
53,300
67,900
110,900
24,000
12,300
740
59,400
17,200
69,600
86,800
29,600
126,600
25,100
12,800
14,300
34,100
18,100
72,900
90,900
29,600
145,400
alncludes miscellaneous operating costs for fugitive leak detection
and repair program.
3-37
-------
Table 3-22. LOW DENSITY POLYETHYLENE, LIQUID PHASE MODEL PLANT REGULATORY
ALTERNATIVES COSTS (JUNE 1980 DOLLARS)
Regulatory Alternative
3
INSTALLED CAPITAL COST (S)
Flare(s)
Flare Ducting
38,400
99,400
88,400
99,400
88,400
99,400
38,400
99,400
Thermal Incinerator
Thermal Incinerator Ducts,
Fans & Stack
Catalytic Incinerator
Catalytic Incinerator Ducts,
Fans & Stack
Condenser
Fugitive Leak Detection
and Repair (LDAR)
Total
ANNUALIZED COST ($/yr)
Direct
Operating Labor
Operating
Materials
(e.g., catalyst)
Maintenance
Materials and
Labor
Utilities
Natural gas
Electricity
Steam
Subtotal
Indirect
Taxes, Insurance, &
Administration
Capital Recovery
Subtotal
Recovery Credit
Total (Direct +
Indirect - Credit)
188,000
219-,000 257,000
139,000 152,000
66,800 66,800 66,800
255,000 613,000 663,000
11,200 31,900 48,600 48,600
9,390
3,550
15,100
8,550
nb
17,200
33,000
31,500
21,400
35,500
37,300
K h h .h
2,676,200° 2,547,100° 2,647,100° 2,647,100°
2,696,900 2,702,600 2,782,700 2,796,500
7,510
27,300
35,300
10,100 24,400 26,400
44,100 102,000 111,000
54,200 127,000 137,000
29,500 29,600 29,600
2,711,500 2,727,200 2,379,800 2,903,900
Includes miscellaneous operating costs for fugitive leak detection
and repair program.
Excess steam nay oe available at zero cost to the plant.
totals nay not add up exactly due to Bounding.
-------
Table 3-23. LOW AND HIGH DENSITY POLYETHYLENE, GAS PHASE MODEL
PLANT REGULATORY ALTERNATIVES COSTS
(June 1980 Dollars)
Regulatory Alternative
INSTALLED CAPITAL COST ($)
Flare(s)
Flare Ducting
Thermal Incinerator
Thernal Incinerator Ducts,
Fans A Stack
Catalytic Incinerator
Catalytic Incinerator Ducts,
Fans 4 Stack
Condenser
-ugitive Leak Detection
and Repair (LDAR)
Total
ANNUALIZED COST ($/yr)
Direct
Operating Labor
Operating
Materials
(e.g., catalyst)
Maintenance
Materials and
Labor
Utilities
Natural gas
Electricity
Stean
Subtotal
Indirect
Taxes, Insurance. &
Administration
Capital Recovery
Subtotal
Recovery Credit
Total (Direct +
Indirect - Credit)
158,000 158,000 158,000 137,000a
921,000 921,000 927,000 938,000
66,800 66,300 56,300
1,080,000 1,146,000 1,151,000 1,142,000
11,200 31,900 31,900 31,900
54,000 59,700
17,100 17,000
18,200 18,200
59,900 59,500
17,000 88,600
18,200 19,500
100,500 126,900 127,100 199,400
43,200 43,200 43,400 43,000
170,700 187,000 187,800 137,000
213,900 232,800 233,300 232,500
29,600 29,600 29,600
314,300 333,000 331,400 402,400
aFlare capital cost lower as no seal necessary to prevent inflow with
higher continuous flow.
Includes miscellaneous operating costs for fugitive leak detection
and repair program.
"Some totals do not add up exactly due to rounding.
3-39
-------
Table 3-24. HIGH DENSITY POLYETHYLENE, LIQUID PHASE-SLURRY MODEL
REGULATORY ALTERNATIVES COSTS
(June 1980 Dollars)
Regulatory Alternative
3
INSTALLED CAPITAL COST ($}
Flare(s) 16,700 16,700 19,900
Flare Ducting 35,100 35,100 49,800
Thermal Incinerator
Thermal Incinerator Ducts,
Fans & Stack
Catalytic Incinerator
Catalytic Incinerator Ducts,
Fans & Stack
Condenser
Fugitive Leak Detection
and Repair (LDAR) 66,800 66,300
Total 51,800 118,600 136,500
ANNUALIZED COST (S/yr)
Direct
Operating Labor 11,200 31,900 31,900
Operating
Materials
(e.g., catalyst)
Maintenance
Materials and
Labor
Utilities
Natural gas
Electricity
Steam
Subtotal
Indirect
Taxes, Insurance, 4
Administration
Capital Recovery
Subtotal
Recovery Credit
Total (Direct +
Indirect - Credit)0
2,590
4,280
18,200
36,200
2,070
7,910
46,200
8,290
4,280
18,200
62,600
4,670
24,200
29,600
51,900
9,180
4,280
18,600
63,900
5,390
27,000
29,600
66,700
Includes miscellaneous operating costs for fugitive leak detection
and repair program.
Some totals may not add up exactly due to rounding.
8-40
-------
Table 2-25. HIGH DENSITY POLYETHYLENE, LIQUID PHASE-SOLUTION
.'IODEL PLANT REGULATORY ALTERNATIVES COSTS
(JUNE 19GO DOLLARS)
Regulatory Alternative
3
INSTALLED CAPITAL COST ($)
Flare(s)
Flare Ducting
Thermal Incinerator
Thermal Incinerator Ducts,
Fans It Stack
Catalytic Incinerator
Catalytic Incinerator Ducts,
Fans J Stack
Condenser
Fugitive Leak Detection
and Repair (LDAR)
Total
ANNUALIZED COST ($/yr)
Direct
Operating Labor
Operating
Materials
(e.g., catalyst)
Maintenance
Materials and
Labor
Utilities
Natural gas
Electricity
Steam
Subtotal
Indirect
Taxes, Insurance, &
Administration3
Capital Recovery
Subtotal
Recovery Credit
Total (Direct +
Indirect - Credit)0
12,500 12,500 12,500
30,100 30,100 30,100
1,070,000
343,000
66,800 66,300
42,700 109,000 1,530,000
11,200 31,900
2,130
4,230
4.3,500
131,300
7,330 73,700
4,280 507,900
40,900
18,900 18,900 18,900
36,400 52,300 325,300
1,710 -1,310 61,000
5,550 22,300 254,000
3,260 27,200 315,000
29,600 29,600
44,700 50,400 1,111,300
Includes miscellaneous operating costs for fugitive leak detection
and repair program.
Some totals do not add up exactly due to rounding.
3-41
-------
Table 8-25. POLYSTYRENE-CONTINUOUS MODEL PLANT REGULATORY
ALTERNATIVES COSTS (JUNE 1980 DOLLARS)
Regulatory Alternative
1
INSTALLED CAPITAL COST ($)
Flare(s)
Flare Ducting
Thermal Incinerator
Thermal Incinerator Ducts,
Fans & Stack
Catalytic Incinerator
Catalytic Incinerator Ducts,
Fans & Stack
Condenser 2,780
Fugitive Leak Detection
and Repair (LDAR) 66,800 66,800
Total 0 65,800 69,600
ANNUALIZED COST ($/yr)
Direct
Operating Labor
Operating
Materials
(e.g., catalyst)
Maintenance
Materials and
Labor
Utilities
Natural gas
Electricity
Stean
Water
Subtotal
Indirect
Taxes, Insurance, &
Administration
Capital Recovery
Subtotal
Recovery Credit
Total (Direct + b
Indirect - Credit)
20,700
5,700
Q 26,400
2,600
16,300
0 18,900
29,600
0 15,700
21,800
5,840
150
390
28,200
2,710
16,800
19,500
190,600
(142,900)c
alncludes miscellaneous operating costs for fugitive leak detection
and repair progran.
Sone totals do not add up exactly due to rounding.
cParentheses denote a negative total net control cost.
8-42
-------
Table 8-27. POLYESTER (PET) - DMT PROCESS MODEL PLANT REGULATORY
ALTERNATIVES COSTS (JUNE 1980 DOLLARS)
Regulatory Alternative
1 2 3
INSTALLED CAPITAL COST (S)
Flare(s) 12,500
Flare Ducting . 3,600
EGRSa (Baseline)0 r 234,000
EGRSa (Alternative)0 2,514,900 2,514,900
TOTAL 234,000 2,514,900 2,536,100
4NNUALIZED COST (S/yr)
Direct
Operating Labor
Operating
Material s
(e.g., catalyst)
Maintenance
Materials and
Labor
Utilities
Natural gas
Electricity
Stean
Water
Subtotal
Indirect
Taxes, Insurance, &
Administration
Capital Recovery
Subtotal
Recovery Credit
Total (Direct - .
Indirect - Credit)
77,400
10,400
940
828,100
916,800
9,360
38,100
47,500
407,100
557,200
154,800
27,700
70,500
823,100
4,650
1,079,800
100,500
408,900
510,500
1,292,000
298,400
166,000
22,800
4,280
70,600
328,200
4,650
1,096,400
101,400
423,900
525,300
1,292,000
329,700
aEGRS - ethylene glycol recovery system.
The baseline system recovers ethylene glycol (EG) from downstream of the cooling water
tower. Ethylene glycol emissions from the polymerization reactor and f^on the distillation
column ^ecovering EG emitted from the esterifiers accumulate in the cooling water ana
are emitted from the cooling tower.
GThe regulatory alternative system recovers ethylene glycol from the polymerization reactor
througn use of an EG spray condenser and recovers ethylene glycol fron the esterifier
through use of a reflux condenser. Thus, very little EG is emitted to be collected in
the cooling water and later emittea to the atmosphere from the cooling water tower.
Some totals do not add UP exactly due to rounding.
8-43
-------
Table 3-28. POLYESTER (PET) - TPA PROCESS MODEL PLANT REGULATORY
ALTERNATIVES COSTS (JUNE 1980 DOLLARS)
Regulatory Alternative
1 2 3 4
INSTALLED CAPITAL COST ($)
EGRSa (Baseline)b 234,000
EGRSa (Alternative)0 2,514,900
Total 234,000 2,514,900
ANNUALIZED COST ($/yr)
Direct
Operating Labor 77,400 154,800
Operating
Materials
(e.g., catalyst)
Maintenance
Materials and
Labor 10,400 21,700
Utilities
Natural gas
Electricity 940 70,600
Steam 828,100 823,100
Water 4,550
Subtotal 916,800 1,079,800
Indirect
Taxes, Insurance, &
Administration 9,360 100,600
Capital Recovery 38,100 409,900
Suototal 47,500 510,500
Recovery Credit 407,100 1,292,000
Total (Direct & Indirect
- Credit) 557,200 298,400
aEGRS - ethylene glycol recovery systen.
The baseline system recovers ethylene glycol (EG) from downstream of the cooling water
tower. Ethylene glycol emissions from the polymerization reactor and fron the distillation
column recovering EG emitted from the esterifiers accumulate in the cooling water and
are emitted from the cooling tower.
cThe regulatory alternative system recovers ethylene glycol from the polymerization reactor
through use of an EG spray condenser and recovers ethylene glycol fron the esterifier
through use of a reflux condenser. Thus, very little EG is emitted to be collected in
the cooling water and later emitted to the atmospnere from the cooling water tower.
8-^4
-------
Table 3-29. COSTS AND ASSOCIATED EMISSION REDUCTIONS OF REGULATORY
ALTERNATIVES FOR POLYPROPYLENE, LIQUID PHASE PROCESS
Regulatory
Alternative
Implementation of
Alternative from Baseline Control
Implementation of Alternative
from Next Less Strinaent Alternative
Annualized
Cost
($)
voc
Reduction
(Mg/yr)
Cost of
Reduction
(S/Mg)
Annualized
Cost
(S)
VOC
Reduction
(Mg/yr)
Cost of
Reduction
(S/Mg)
(Baseline)
2
3
4
S 15,700
5117,000
S106.000
60
440
450
$250
$265
S235
S 15,700
S 101,000
S (10,700)a
50
330
10
S
S
S
260
255
(l,070)a
aParenthesas indicate a negative incremental cost.
8-45
-------
Table 3-30. COSTS AMD ASSOCIATED EMISSION REDUCTIONS OF REGULATORY
ALTERNATIVES FOR POLYPROPYLENE, GAS PHASE PROCESS
Implementation of
Alternative from Baseline Control
Implementation of Alternative
from Next Less Stringent Alternative
Regulatory
Alternative
Annual i zed
Cost
(S)
voc
Reduction
(Mg/yr)
Cost of
Reduction
($/Mg)
Annual ized
Cost
($)
VOC
Reduction
(Mg/yr)
Cost of
Reduction
($/»g)
1
(Baseline)
2
3
$ 15,700
S 34,500
60
2,635
$260
$ 13
S 15,700
$ 18,300
60
2575
S260
S 7
8-46
-------
Fable 3-31. COSTS AND ASSOCIATED EMISSION REDUCTIONS OF REGULATORY
ALTERNATIVES FOR LOPE, HIGH PRESSURE, LIQUID DHASE ACCESS
Implementation of
A1ternafive from Baseline Control
Implementation of Alternative
from Next Less Strinoent Alternative
Regulatory
Alternative
Annual ized
Cost
(S)
voc
Reduction
(Mg/yr)
Cost of
Reduction
(S/Mg)
Annual izsd
Cost
(S)
VOC
Reduction
:^g/yr)
Cost of
Reduction
(S/Mg)
1
(Baseline)
2
3
4
3 15,700
3168,300
5192,400
60
275
335
S2GO $ 15,700 50
$610 3152,600 215
$570 S 24,100 50
S710
3400
8-47
-------
Table 3-32. COSTS AND ASSOCIATED EMISSION REDUCTIONS OF REGULATORY
ALTERNATIVES FOR LDPE/HDPE LOW PRESSURE, 3AS PHASE PROCESS
Implementation of Inple-nentation of Alternative
Alternative from Baseline Control from ?text Less Stringent Alternative
Regulatory
Alternative
1
(Baseline)
2
3
4
Annual ized
Cost
($)
^
S 15,700
S 17,100
S 08,100
VOC
Reduction
(fig/yr)
.
60
63
71
Cost of
Reduction
(S/Mg)
S260
S270
$1,240
Annual ized
Cost
($)
.
515,700
SI, 400
$71,000
VOC
Reduction
(Mg/yr)
.
60
3
8
Cost of
Reduction
($/?1g)
.
S260
S470
53,800
8-48
-------
fadle 3-33. COSTS AND ASSOCIATED EMISSION REDUCTIONS OF REGULATORY
ALTERNATIVES FOR HOPE, LIQUID PHASE SLURRY PROCESS
Implementation of
Alternative from Baseline Control
Implementation of Alternative
from Next Less Strinaent Alternative
Regulatory
Alternative
Annualized
Cost
(S)
voc
Reduction
(Wg/yr)
Cost of
Reduction
(S/Mg)
Annualizsd
Cost
(3)
VOC
Reduction
Cost of
Reduction
(S/Mg;
(Easeline)
0
315,700
320,500
SO
140
S250
S150
315,700
3 4,300
oU
30
3250
3 50
8-49
-------
Table 3-34. COSTS AfID ASSOCIATED EMISSION REDUCTIONS OF REGULATORY
ALTERNATIVES FOR HOPE, LIQUID PHASE SOLUTION PROCESS
Implementation of
Alternative from Baseline Control
Implementation of Alternative
from Next Less Stringent Alternative
Annuali zed
Regulatory Cost
Alternative ($)
1
(Baseline)
2 $ 15,700
2 $1,067,100
VOC Cost of Annual ized VOC Cost of
Reduction Reduction Cost Reduction Reduction
(Mg/yr) ($/Mg) ($) (Mg/yr) (S/Mg)
60 $ 260 $ 15,700 50 S 260
190 $5,620 SI, 051, 400 130 $8,090
8-5C
-------
Table 3-35. COSTS AND ASSOCIATED EMISSION REDUCTIONS OF REGULATORY
ALTERNATIVES FOR POLYSTYRENE, CONTINUOUS PROCESS
Implementation of
Alternative from Baseline
Regulatory
Alternative
1
(Baseline)
2
3
Annual ized
Cost
($)
-
S 15,700
(S142,900)a
VOC
Reduction
(Mg/yr)
-
60
285
Control
Cost of
Reduction
($/Mg)
-
$260
($500)a
Implementation of Alternative
from Next Less Strinqent Alternative
Annual ized
Cost
(S)
-
$ 15,700
($158,600)a
VOC
Reduction
(Mg/yr)
-
50
225
Cost of
Reduction
(S/Mg)
-
S260
(S700)3
Parentheses denote a negative net or incremental cost.
8-51
-------
Table 3-36. COSTS AND ASSOCIATED EMISSION REDUCTIONS OF REGULATORY
ALTERNATIVES FOR PET/DMT PROCESS
Implementation of
Alternative from Baseline
Regulatory
Alternative
1
(Baseline)
2
3
Annual i zed
Cost
($)
-
(S258,800)a
($277,500)a
VOC
Reduction
(Mg/yr)
-
975
995
Control
Cost of
Reduction
($/Mg)
-
($270}a
(S230)a
Inpl mentation of Alternative
froin Next Less Strinqent Alternative
Annual ized
Cost
($)
-
(S267,800)a
$31,300
VOC
Reduction
(Mg/yr)
-
975
20
Cost of
Reduction
($/Mg)
-
($270)a
$1,570
Parentheses indicate a negative net or incremental cost.
8-52
-------
Table 8-37. COSTS AND ASSOCIATED EMISSION REDUCTIONS OF REGULATORY
ALTERNATIVES FOR PET/TPA PROCESS
Implementation of
Alternative from Baseline
Regulatory
Alternative
1
(Baseline)
2
Annual ized
Cost
($)
-
(S258,800)a
VOC
Reduction
(Mg/yr)
-
970
Control
Cost of
Reduction
($/Mg)
-
($270)
Implementation of Alternative
fron Next Less Stringent Alternative
Annual ized
Cost
($)
-
($258,800)a
VOC
Reduction
(Mg/yr)
-
$970
Cost of
Reduction
($/Mg)
-
(270)a
Parentheses indicate a negative net or incremental cost.
8-53
-------
associated emission reductions of regulatory alternatives for the nine
model plants for the implementation of an alternative both from the
baseline control level and from the next less stringent alternative.
National control costs are estimated, as shown in Table 8-38, based
on 27 projected new facilities, described extensively in Chapter 9
"Economic Impact". The total nationwide fifth-year net annualized cost
ranges from $8.3 million for baseline control to $3.7 million for the
maximum achievable control level. The maximum achievable control level
is equivalent to a 14,950 Mg VOC reduction from baseline per year.
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 (NPOES) 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
8-54
-------
Table 3-38. TOTAL FIFTH-YEAR NET ANNUALIZED COST OF PROCESS
AND FUGITIVE EMISSION CONTROLS FOR POLYMERS AND RESINS
FACILITIES AFFECTED BY NSPS
Regulatory
Polymer/ Al
Process
PP
Liquid
Gas
LDPE
Liquid
Gas
HOPE
Liquid Slurry
Liquid, Solution
Gas
PS a
Continuous
PET
DMTa
TPAa
TOTAL
ternative
Number
4
3
4
4
3
3
4
3
3
2
Annualized
Per Facility
Costs, $/yr
105,900
34,500
192,400
88,100
20,500
1,067,100
88,100
(142,900)b
(227,500)b
(258,800)°
Number of
Facilities
3
3
1
5
2
3
5
2
1
2
27
Total
Nationwide
Fifth-Year
Net
Annualized
Process
and Fugitive
Costs, $/yr
($)
317,700
103,500
192,400
440,500
41,000
3,201,300
440,500
(285,800)b
(227,500)b
(517,600)
3,706,000
From uncontrolled baseline, others from controlled baseline
Parentheses denote a negative net or incremental cost.
Source: Tables 8-20 through 8-28 of this report.
8-55
-------
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
on
1977 - 1986 was $850 million. y
8.2.1.2 Safe Drinking Water Act (SDVJA). 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
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 nethod(s) employed by each facility,
8-56
-------
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
(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 wil"! 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
8-57
-------
(TSD facilities). The wastes are identified and listed by industry. At
the tine 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.
8.2.4.2 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 two regulatory actions begun since 1977 under Sections 111
and 112 of the Clean Air Act that will, if promulgated, impose costs on
firms with new, modified, or reconstructed polymerization facilities.
These actions are the potential NS?S for control of VOC emissions from
both process and fugitive sources (the subject of this study), and the
potential NSPS for control of VOC emissions from raw materials storage
at the production site. The control costs for the former have been
tabulated above (Section 8.1); the control costs for the latter are
8-58
-------
estimated, as discussed below, primarily from information contained in
the draft BID for volatile organic liquid (VOL) storage tanks.
For the number and type of affected facilities discussed in Section 9.1,
the sum of the potential fifth-year net annualized costs of the VOL
storage is estimated (Table 9-15). The control cost estimation for VOC
emissions from monomer and raw materials storage requires estimations of
(1) the number of polymers and resins plants at which storage tanks are
required, (2) the number of storage tanks required at each plant, and
(3) the VOC control costs per storage tank.
First, the number of plants at which storage tanks are required is
estimated. The polymerization plant is either a part of a larger chemical
complex at which the monomer is both produced and consumed or it is an
isolated plant. Where the polymer is produced at a major chemical
complex storage tanks serve as both monomer storage following production
and the onsite monomer storage for the polymerization process. The
costs for these storage emission controls are not assigned here to the
polymerization facilities because the costs would be incurred even in
the absence of the polymerization facilities. The proportion of the new
plants requiring onsite storage is that proportion of plants shown in
Table 9-1 that are not owned by large oil or chemical companies. The
results are shown in Table 8-39.
Second, the required number of storage tanks is estimated. It is
assumed that the required storage capacity is equal to a 30-day supply
or about 3.4 percent of annual capacity and that 2.4 tanks are required
for each gigagram of plant production capacity. The number of tanks
required per gigagram is based on data for the model producer/consumer
contained in the draft VOL storage BID, indicated above, and was
calculated by dividing the number of tanks for the nodel by the plant
41
production capacity. Thus, the number of tanks, shown by type of
polymer in Table 8-40, totals 116.
Third, the cost for emission control per storage tank is calculated
from cost data contained in the draft VOL storage BID. These costs,
given in first quarter 1980 dollars, are converted to June 1980 dollars
by use of the Engineering News Record Construction Index (101.8). At
the tine the draft EIS was prepared, EPA recommended Regulatory
8-59
-------
Table 8-39. NUMBER AND CAPACITY OF PLANTS TO BE AFFECTED BY BOTH
VOL STORAGE AND POLYMERS AND RESINS STANDARDS THROUGH 1988a
Polymer
PP
Liquid
Gas
LDPE
Liquid
Gas
HOPE
Liquid,
Slurry
Liquid,
Solution
Gas
PS
Continuous
PET
DflT
TPA
Projected
number
of new
plants
3
3
1
5
2
3
5
2
1
2
Percent of
these
expected to
be nonmajor
chemical or
oil company
20
20
10
25
10
Number
of new
nonmajor
company
plants
0.6
0.6
0.1
0.5
0.2
0.3
0.5
0.5
0.1
0.2
Capacity
of the
plants
(Gg/y)
90
63
28
75
42
27
75
38
10
20
This table is based on data contained in Section 9.1 of this report.
8-60
-------
Table 8-40. STORAGE TANK COMPLIANCE COSTS FOR PLANTS TO BE AFFECTED
BY BOTH VOL STORAGE AND POLYMERS AND RESINS
STANDARDS THROUGH 1988
Polymer
PP
Liquid
Gas
LDPE
Liquid
Gas
HOPE
Liquid, Slurry
Liquid, Solution
Gas
Storage,3
8
5
2
6
4
2
6
Number
tanks
19
12
5
14
10
5
14
Credit
@ $255/tank
4,800
3,100
1,300
3,600
2,600
1,300
3,600
PS
Continuous
1,800
PET
DMT
TPA
TOTAL
1
2
39
2
5
93
Oc
0C
22,100
Production capacity requiring storage (from Table 8-39) times 8.4
percent (see text).
Storage times 2.4 (see text).
CPET/DMT and PET/TPA is excluded since only heavy liquids and solids are
used in these processes; heavy liquids and solids are excluded from
control requirements.
8-61
-------
Alternative III. This option would require each storage vessel storing
a volatile organic liquid (VOL) with a true vapor pressure of less than
76.6 kPa to be equipped with a contact internal floating roof with a
liquid-mounted primary seal and a continuous secondary seal. A vapor
control system would be required for all storage vessels storing a VOL
with a true vapor pressure of greater than or equal to 76.6 kPa. Thus,
all polymers and resins node! plants except PET/DMT and PET/TPA, which
use heavy liquids and solids, will be covered.
The VOC control costs for monomer and raw materials storage are
estimated from costs developed for SOCMI in the following manner.
Alternative III for all of the SOCMI was estimated to result in a total
42
annualized credit of $6,220,000 due to product recovery credits In
order to obtain the credit per tank, the annualized credit was divided
4^
by the total SOCMI storage tank population of 24,287. Finally, the
emission control costs for these storage tanks are assumed to be comparable
to these required for the polymers and resins industry. The control
costs for the polymers and resins industry are shown in Table 3-40. The
resulting fifth-year cost to the polymers and resins industry under the
above assumptions is a negative $22,100.
8-62
-------
8.3 REFERENCES
1. Memo from Siebert, P., Pacific Environmental Services, Inc. (PES),
to Polymers & Resins NSPS Project File. September 3, 1982. Selec-
tion of SOCMI Fugitive Analysis Model Plant B to represent fugitive
emissions characteristics of polymers and resins plants.
2. VOC Fugitive Emissions in Petroleum Refining Industry - Background
Information for Proposed Standards, Internal Review Draft. U.S.
Environmental Protection Agency, Research Triangle Park, N.C.
Publication No. EPA-450/3-80-026. November 6, 1981.
3. 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.
4. Reference 3, p. IV-4.
5. Memo from Sarausa, A.I., Energy and Environmental Analysis, Inc.
(EEA), to Polymers and Resins File, flay 12, 1982. Flare costing
program (FLACOS).
6. Telecon. Siebert, Paul, PES with Straitz, John III, National Air
Oil Burner Company, Inc. (NAO). November 1982. Design, operating
requirements, and costs of elevated flares.
7. Telecon. Seibert Paul, PES, with Keller, Mike, John Zink, Co.
August 13, 1982. Clarification of comments on draft polymers and
resins CTG document.
8. Telecon. Siebert, Paul, PES with Fowler, Ed, NAO. November 12,
1982. Operating requirements of elevated flares.
9. Telecon. Siebert, Paul, PES with Fowler, Ed, NAO. November 5,
1982. Purchase costs and operating requirements of elevated flares.
10. Telecon. Siebert, Paul, PES with Fowler, Ed, NAO. November 17,
1982. Purchase costs and operating requirements of elevated flares.
11. 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.
12. Reference 6. (May be a separate reference in future.)
13. Perry, R.H. and C.H. Chilton, eds. Chemical Engineers' Handbook,
fifth edition. New York, McGraw-Hill Book Company. 1973. p. 5-31.
8-63
-------
14. Chontos, L.W. Find Econonic Pipe Diameter via Improved Formula.
Chemical Engineering. 87(12):139-142. June 16, 1980.
15. Memo from Desai, Tarun, EEA, to EB/S Files. March 16, 1982.
Procedure to estimate piping costs.
16. Hemo from Kawecki, Tom, EEA, to SOCMI Distillation File. November 13,
1981. Distillation pipeline costing model documentation.
17. Richardson Engineering Services. Process Plant Construction Cost
Estimating Standards, 1980-1981. 1980.
18. EEA. Distillation NSPS Pipeline Costing Computer Program (DflPIPE),
1981.
19. Neverill, 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.
20. Memo from Mascone, D.C., EPA, to Farmer, J.R., EPA. June 11, 1980.
Thermal incinerator performance for NSPS.
21. 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. Fig III-2, p. III-8.
22. Reference 21, Fig. A-l, p. A-3
23. 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. 8-9.
24. EEA. Distillation NSPS Thermal Incinerator Costing Computer Program
(DSINCIN). May 1981. p. 4.
25. Reference 21, p. 1-2.
26. Reference 23, p. 6-3 and 6-4.
27. Reference 21, Fig. V-15, curve 3, p. V-18.
28. Reference 24. p. 8.
29. Telecon. Katari, Vishnu, Pacific Environmental Services, Inc. with
Tucker, Larry, Met-Pro Systems Division. October 19, 1982. Catalytic
incinerator system cost estimates.
30. Telecon. Katari, Vishnu, Pacific Environmental Services, Inc.,
with Kroehling, John, DuPont, Torvex Catalytic Reactor Company.
October 19, 1982. Catalytic incinerator system cost esitmates.
8-64
-------
31, Letter from Kroehling, John, DuPont, Torvex Catalytic Reactor
Company, to Katari, V., PES. October 19, 1982. Catalytic incinerator
system cost estimates.
32. Telecon. Katari, Vishnu, Pacific Environmental Services, Inc.,
with Redden, Charles, Artisan Company. September 29, 1982. Heat
exchanger system cost estimates.
33. Telecon. Katari, Vishnu, Pacific Environmental Services, Inc.,
with Mr. Ruck, Graham Company. September 29, 1982. Heat exchanger
system cost estimates.
34. Telecon. Katari, Vishnu, Pacific Environmental Services, Inc.,
with Clower, Dove, Adams Brothers, a representative of Graham
Company. September 30, 1982. Heat exchanger system cost estimates.
35. Telecon. Katari, Vishnu, Pacific Environmental Services, Inc.,
with Mahan, Randy, Brown Fintube Company. October 7, 1982. Heat
exchanger system cost estimates.
36. Memo. Dimmick, F., EPA:SDB, to Wyatt, S., EPA:SDB. August 11,
1980. Minutes of meeting between EPA and Texas Chemical Council
representatives about TCC comments on recommended NSPS for fugitive
VOC emissions in SOCMI.
37. VOC Fugitive Emissions in Synthetic Organic Chemical Manufacturing
Industry - Background Information for Proposed Standards. Draft
EIS. U.S. Environmental Protection Agency, Research Triangle Park,
N.C. Publication No. EPA-450/3-80-033a. November 1980.
38. Organic Chemical Manufacturing Volume 4: Combustion Control Devices.
U.S. Environmental Protection Agency, Research Triangle Park, North
Carolina. Publication No. EPA-450/3-80-026. December 1980.
39. The Cost of Clean Air and Water: Report to Congress. U.S. Environmental
Protection Agency, Washington, D.C. August 1979. p. 27.
40. VOC Emissions from Volatile Organic Liquid Storage Tanks - Background
Information for Proposed Standards. U.S. Environmental Protection
Agency, Research Triangle Park, N.C. Publication No. EPA-450/3-31-003a
Draft EIS. April 1981.
41. Reference 40, p. 6-5.
42. Reference 40, p. 9-48.
43. Reference 40, p. 9-46.
8-65
-------
9. 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 PSpolymers used in plasticshave characteristics
similar to those of PET, a polymer used in fibers. 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
Note: Last minute changes in some of the orocess parameters and
renulatory alternatives, and in the olant lists in Chanter 3,
have not yet been incorporated in this chapter. These changes,
however, do not sinnificantly alter the economic analysis
nor change the conclusions of this chapter.
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 marketsconstruction, 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. 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
2
$500 million in its subsidiary, the Exxon Chemical Company. This
company operates three plants that produce polymers. Gulf Oil's
subsidiary, Gulf Oil Chemicals, operates six polymer plants, most 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 polymer
9-3
-------
TO
CM
CO
CT,
f_(-
i-
TO
3
C
TO
~3
UJ
5^
\
a
<^
ce
UJ
cs
1
^£
ZD
^
y
^.
a
00
1
§
ex
z
^
J
=
1
p
~
o
UJ
03
t
11
TO
*~
TO
2
5
c_
UJ
3
UJ
a.
o
i
a.
C_
O TO M y t^ *^ QJ
f n cj ^ 2 C3 Q-
onCl-i. 4JjCO^-
= = Z m ccc
3 (A < S. 0) i « O O
U = w = C J3 = *J -J
QJ 3 tj C TO '^ O O TO TO
T* ^ i^« IBM xx ^3T »£ y ^y ^^
CM CM US *T X3- « CM CM <
CM - CM
O
~-« P-. ^ ^H
^ "* *"*
1
C TO -C
O Z
, » l/l i, -r-
TO JC O ^*- CJ GJ
1- 0 i. J= -J U
I/I O 0) *^ U l/l *J -
o a. c ii ;- o >
-f- i-;zc- aiG;" i-
4J O J3 C D
^1 OCC U---TOXUO
^ TO TO U, >> U
TO"3UU ' 3» i^
Q. = D -r- =X £.0)
S -2 = = 3 ^ < j= -tl
<_>
CM CO rt CO CM
-
^H CM
2
OJ
'O r- S-
^. & 1/1 TO U
4-1 3 TO U =
0- S- I E
CJ CiJ O
t/i *7 j= a.
a. u c u
|^ l/l r D
^ 1^ *" aj ^~
J= O O J= O
c. a. K i/> uo
CM ~ CJ1 ^ VO
<
C «
CM ^* LO
CM CM
CM CM <
_
on
S» ^1 TO
C
= TO
J f_^ Q ^
'- TO C 3 TO
^ O d CS UJ
UT> ^ LO
ro
^H
TO
C
TO in
5 "
C 4-1
* 1/1
* TO
i 0)
i G. 13
5 g 1
"O Of ro
«3 >i
Q -*J C
C en O
TO X -i-
M a> c
l/> H- 3
*3- cn m
ITJ
CM CM
CM .*
(/)
"0
CJ
I/)
^ 2J
fO r-
3 W
T5 3
C
o
*/)
£- O CJ
x «;
X -r-
UJ UJ I
^-4
T3
0
U_
T3
OJ
4-1
^
C
^3
t
a
g
=
3J
t
M
c
J
u
LU
ro
CM
r»
tJ
U
£
OJ
JC
w
v^
l/^
3
CM
CM
0)
.-.
Qi
<*3
OJ
j_
1
<_
-o
^
o
c
o
o
9-4
-------
plants are Standard Oil (Indiana), Atlantic Richfield, and Shell Oil; and
two major oil companiesPhillips Petroleum and Cities Serviceoperate
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 produced about
40 percent.
Phillips Petroleum had the largest production capacity (about 25
percent) of HOPE. Oil companies, in general, accounted for approximately
60 percent of the HOPE market, and chemical companies provided 30 percent.
9-5
-------
o
CM
CO
f 1
^
rc
3
C
13
-3
UJ
Q_
g
LU
o*"
1
^£
U.
z
>.
CO
LO
I
z
O
O ^"* f") *G m
^>
CO
0 O
CD
13
in
in
r^
CM
O 0
CD Cn
LO
c
o
-o
O C i- i-
CU ro CU O
3 m «2 U
O « CJ ro
_CU _3 O C 5
in in o m in
i » r^. *r < * CM
ro r-. CM
in o in
r^ "Q* en
CM
in in in
-H 00 CM
CM CM
cn
C3
o in
O ro
CO
ro
O C
-J i/l M-
.^ 5)
O Cv
r*^ ^H.
tl *O
c c:
O 0
z s:
o o
ro n
O
r^j
i *
O 0
ro
OJ
,^
>
!_
X (/I
i"- (/)
Q. OJ
S **~
J=
< - (J
O uo
r csj
in
CM
O
00
LO
O
E
i)
O
S- CL
*J 3
1) O
ex s.
o
in
0. U
- ro
* Ul
SI o.
0 0
CM r^
0 0
CM r^
1- in
CU CU
i»
^^ /!
c: 3
,° "2
_
in *^
OI &-
'- O
o
ro
g
s_
a>
c
o
Q.
X
o
o
r**i
o
CO
LO
"
in
\o
in
ro
u
s
5
-o
^
-o
LU
0 LO
O CM
in LO
en CM
O
-*
in
ro
CM
ro in
u
c -w
>-- ro
i rx
.,^
O 0)
C
o 01
u w
,
o -w
C tn
ro X
w CU
OO 1
in in
co ro
ro 10
O in
in in
CM
o
O
,__
^O
i.
2;
^
C
w
n. Q
X
LU uJ
LD Lrt tjO LO O
uO rxi «a- CO
O c\J u-^
r-H LO
~
o
o
en
CM
in o in in
' CM
LO
rM
O in
en in
en
CM
O in o
LO r^ in
en r^
4-> 1
c *j <:
S C 1/1 -r- O
o in in o o
co CM ro i en
LO ^^ en
o in in
co CM ro
LO "
0 O
r LO
O
LO
CM
O
cn
ro
O
co
1) V.
o cy
~ "^
c^ "3
ce
in oo
u <
t- CJ
S- !_ 0,
3 1 O
^ 5j ^3
C C *. ^ ^*
C ro C
-J CU O O
CJ rjj ^ f . 4»
J2 i. O S-
._ ._ o -03
1_ LL. CD C_J O
in
i
ro
O
«^
1
ro
in
0)
ro
h~
=
o
OJ
.^
a
_i
c
o
u
rO
*J
-0
-3
S
l_
14.
"S
-
.**
^
^
LJ
11
U
3
O
ro
9-6
-------
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
capacityabout the same as that of all oil companies.
PET capacity was essentiallyclose to 90 percentdevoted to
manufacturing fiber. Du Pont had the largest capacity for
production30 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-supoliers 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 is produced by 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 integrationproduct 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 in separate plants.
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
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
capacityclose 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 categoriesthermosetting and thermoplastic resins. Shipments for
thermoplastic resins (SIC 28213), those which include PP, HOPE, LDPE, PS
and polyvinyl chloride, amounted to $9.3 billion in 1977.s 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 chlorides, the value of the four
polymers produced in the plastics products industry (SIC 3079) are
assumed to be about the same as that of the polyvinyl chloride;
9-9
-------
consequently, S15.0 billion is a reasonable estimate of the total SIC 2281
and 2079 shipments in 1981 for PP, HOPE, LDPE, and PS.
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 1981
(million nominal dollars)
SIC 2821 10,818 11,998 14,282 15,570 17,500
Plastic materials
and resins
SIC 28213 9,266 10,277a 12,233a 13,340a 14,990a
Thermoplastic materials
and resins
SIC 2824 6,380 6,921 8,227 8,811 11,035
Organic fibers
noncellulosic
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
characteristicssize, age, location, and employmentare 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 similarabout 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 introduced
in 1970.
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 Mew 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.
9.1.1.2.4 Employment.4'6 The 1980 employment data for plants
producing the five considered polymers and resins are included into the
SIC data.
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 AMD 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
12.5
35.0
12.5
2.7
14.8
301
and
above
6.3
15.0
12.5
-
3.7
Total
100.0
100.0
100.0
100.0
100.0
a 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
243
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
a Source: Compiled from data contained in Tables 3-1 to 3-5.
9-13
-------
0
ft
f
r-H
CO
cn
Q
CL
=
3
to
CO
<^
r- 1
CO
O1
.I
r*^
^x.
en
, i
CO
-a
o
c ^
c ^
t i
tr>
^^
S-
+J
to
3
o
c
t 1
CO
g*
-M
t^_
O 0
1_
CU to
JD U
CO
^
re
co
CO
to
^
i
^*
co i
r- CM
CO
-C CM
O
i- O
~3 oo
o c
4-
O
C
o
'£
i_
o
r
re
en -M
o S
co E
>.*
<_J O
( * r~
OO CL
c cD
*f
f
c
S- -i-
OJ
jQ (/I
~ ^o
3
4- C r-(
O CO
S- re < «
CO t->
..a o c
E <- > -r- CL-I-) -M re E +-><-
3 +J
c
re
cu <
C~ Q_
-(-> O
E
4- S_
O CU
c~
C -(->
o
r- C
-|_> !-
s_
O t/*
CL cu
o cu
S- >.
Q. 0
CO Q.
-E E
h- CU
O O
O O
LO l~-
» ft
co co
LO <*
O O
0 0
CM r-
ft ft
f*^ ^O
LO
O
4->
r->.
P*^
G-I
r i
c
r"
CO
re
i i
CO
cr»
i i
c
""
Q.
f.
c re
*^ 4-2
O
W 4-^
c
CU O)
S -=
o
CL O
£
CU +->
c
co CU
CU U
-o v-
3 CU
r CL
O
X CM
LU CM
O
o
. I
ft
^J-
co
cC
7^
CO
, 1
CM
CO
CM
OO
Q.
ft
LU
0.
Q
J
tft
**
UJ
Q_
< -N
" "
M
Q.
Q.
K^^>-
to
s_
CO
>J
,
Q
0.
co
o
i.
CL
CO
C"
^_J
c:
0
o
CO
CO
re
r*i
CO
r-
CO
*r^
-C
M
o
CO
j=
4^
o
-M
IT"
O
r-
-M
0 C
3 O
-o <-
O 4J
OL 3
"O
O O
> S-
QL. CL
r~ ^J
CO O
e ^
O CO
^~ g~
CL4->
w
O) 4-
O
CO
S c
re cu s i
u a CM
r- S- CO
r- CU CM
Q. CL
Q. 'O
re o *
CM OO
CO
-C CO C
h- -^ -r-
o
o
«»
o
i-H
*
^.c.
z.
en
P*""*
o
CO
>,
1
0) to
> s-
r- OJ
4^ "~
CL >,
re r
0 O
CL
to
S- CD
o r
CO !
CO O
O) 3
U "O
o o
i- 5-
CX CL
CU
4" r*
O 1 -M
LU
c a. co
O -r-
i- C
-i-) -f ^J*
5- CM
O 00 CO
CL S- CM
0 CU
S >iO
a. o i > -
i oo r^
^J C\ ^*^
-c E c en
1 CO T- T-I
o o
0 0
co <t
ft ftl
T-* l£>
CM «3
O
O
CO
ft
CM
CM.
^^
^J-
CM
CO
CM
LU
a.
ft
oo
a.
l *
LU LU
a. Q-
^^^ r~\
31
to
5^ ^
CO LU
l"*} f^
o
4- _J
L. **
CU O_
*-> a.
CO x_^» »
CU >-D
>, i 1-
O -M +
Q. O CO
f^ *^^*
^
O
("
o
re
T
r
re
>
re
4-*
O
c
II
-------
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 yearsgenerally at the expense of PSwith 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
Q
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
Q
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 fibersover 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 exportsover 400 gigagrams exported in
1981was of LDPE, and the volumes of PP and PET exports were 311 and 400,
respectively. The lowest exports were for PSfewer than 100 gigagrams
exported.
Polymer imports have been relatively minorless than 100 gigagrams
in 1981and most of it was of HOPE. In PET fibers, imports reached about
15 gigagrams, an increase over the approximately 8 gigagrams reported in
I960.10
9.1.1.3.3 Product differentiation. 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 filaments.
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 divided 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
HDPE
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
a 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
A
Cn
CO
en
LU
Q.
L^
^^
OO
oo
LU
c ^
0
Q_
a
LU
s:
LU
oo
^
Q
LU
CQ - -x
oo E
i i S_
OO C7>
LU fO
cz cn
r~
O CD
^j
,-Q
1_
J
LU
a.
oo
a.
LU
CL
Q
3=
LU
a.
a
'
a.
Q_
4_3
0)
.« S_
O
C ^-
4-5 0) CL
r~ E
T3 "D C X
C C E (T3 OJ r
O fO 03 T3
r- 4-> S- S_ 4->
t/iEojojaioJaio
3 r QJ S- O._Q ^ H"
i_ !- .C -1- -1- !- 4->
4-> U- OO 3 Q_ LU O
x
LU
0000
p"^ [**»
f^H P*^
«
11
OO O ID 1^3
^J- Q CO
1~~ OO Lf>
r1 ' t
CM CTl «sj- LO
LO ^O CO <~ i
<^ P^- CM C^
n
r 1
CO f** r 1 OO
OO CM CTl VO
CM oo cy>
*»
CM
«3" CTi r-H CM
>*O CM CTl Lf5
LO r-H 53-
t
< 1
C7>
c:
r
-a
,_«
o en
S c
1 ^
c -a s.
O r CU
r- O J=
4J S -M
U Or-
QJ 2 ^O
r-l O r- 4J
C r- r- O
i i CO «C 1
"
t/1
4-J
O
13
"^5
o
i.
Q.
T3
O)
T3
3
S.
4->
X
OJ
5_
OJ
g"
4«}
O T3
=
-------
o
^'1
A
CT>
ri
CO
cn
r-H
t
LU
NX
O£
s
cv
o
«c
"*
^^
CO
00
OO to
LU £
cc. to
Q cn
2 H5
_H
~~
00
CO
en
^
r~
^
03
1
h-
UJ
C_
00
Q_
LU
Q.
Q
^
LU
Q-
Q
J
a.
C-
4_)
O)
_^
s-
03
2:
O CM T3 LO -O LO
i i O > '
CM
r^* co ">f co co co
r-. CM <*o CM **
VO *"^ *"~l OO
cn *^ CD t«o o cn
CM r~~ CM i^ 1
O CM CM
*
r 1
cr r^* i^ ^o r*^ «-H
CM co **o cn r-~
O t I CM
*
CM
r-l CO J3 CM O~l r-H
CM *yO ""* tQ CM
r-H 00 t 1 OO CM
(J
r
CO 03
O ! cn c
r- C C O
4-> O <- <-
(J i- -C -!->
34->^3
i- CJ T- +J
+-> 1) C -i
C CO i S_ -)->
O C LU 3 to
r- O U_ C
4-J f ^ A H t
03 i .«
+J Cn « f8 4) *
s- c cn u s- i.
O -I- C -i 3 O)
Q. cn -i- i. 4J e
uo 03 T3 *-> >- 3
C -i
.-I «5J- O
oo ^i- «=r
*»
oo
UO -Q CO t 1 OO
CM OO t 1 VO
CM OO 1
*
r I
(/)
cn
c
^*
4_>
>, (0
S- 0
-= CO
^_> .,- +J 4-1
O 03
T3 -C r Q. -t->
c -o t x o
H-, <£ e£ LU t
-o
O
O
C7>
cn
c:
S-
o
a.
A
CO
4-J
U
3
"^
O
^
a.
CL)
s-
fO
o
J-
^
(O
, 0 -^
i »->
C r (/I
O r C
03 !-
) =
LU TD
Q_ C C
i 03
S_ U
0) "O S_ 4J
o Qj DO)
r- -a E
<<- 3 3 «
1 1 CO I/I
C 0 C >,
0 C 0 O
z: i i o -»->
fX3 j^ (J
*
0)
i-
3
co
O
^~
O
CO
r*
-a
"^
^
o
>
(O
o
4)
-a
qj
^^
L^
^_j
.,
3
13
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 Profile
The initial commercial development of the major thermoplastics
occurred in the period 1930-1940 with the introduction of such plastics as
PS and LDPE. In the 1950's HOPE and PP 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. 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. PET fibers had the
lowest increase during the period with 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
12
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 PSvarying from a low
of 64 percent in 1976 to 77 percent in 1979. Such low rates reflect the
competition in this relatively mature sector.
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 (X)
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 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 ($/kg)
PP .60-.73 1.10-1.23
LDPE .64-.73 1.01-1.06
HOPE .68-.73 .99-1.08
PS .56-.66 1.10-1.14
PET 1.43-1.54 1.85-2.58
9-22
-------
o
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 marketsthe
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 SI.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 S50 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
(Profit as a percent of sale)
1974 1975 1976 1977 1978 1979 1980 1981
8.6 6.0 7.8 3.2 4.8 3.7 3.6 3.6
Composite returns in 1974 approached 9 percent of sales and reflected the
effects of the general price increases and inventory reductions 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 plants. Twenty-
seven new plants equivalent in size to the model plants (defined in
Chapter 6) are projected. The numbers of plants by type of polymer and
process are:
9-25
-------
Polymer/Process Number of plants
PP
Liquid phase 3
Gas phase 3
LDPE
Liquid phase 1
Gas phase 5
HOPE
Liquid phase-slurry 2
Liquid phase-solution 3
Gas phase 5
PS
Continuous 2
PET
DMT 1
TPA 2
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) took into
account the current geographic distribution (SIP and non SIP states) of the
existing plants and assumed new plants would mirror the same distribution,
capacity, and emission controls. The numbers of plants by process result
from a growth analysis of each polymer, as described below. The projection
is subject to some uncertainty because a few markets, as indicated in
Section 9.1.2 above, are currently unstable, and such instability begets
caution. Chemical industry opinion indicates that there may be few new
grass roots polymers and resins facilities built over the next several
18
years. Such assumptions imply that unstable markets mean that some
growth in production will come from the upgrading or demothballinq 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
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
9-26
-------
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 = FP - CC + RC - 1C ,n
\cu I (L)
where
NC = new capacity
FP = 1988 production
CU = 1988 capacity utilization rate
CC = current capacity (1981)
RC = retired capacity, 1984-88
1C = interim capacity, 1982-83
and
NP = NC * MP (2)
where
MP = 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 was projected by applying the 1981 domestic
production and net exports growth rates as determined from published
23 24
projections. ' These published growth rates were developed from
historical trends and end uses data, from conversations 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.
9-27
-------
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
25
governmental policy decisions, especially trade restrictions.
Domestically, the continuing decline in demand since 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 concensus growth rates
were determined for both domestic markets and net exports. (Consensus
rates were used in order that the NSPS cost estimates would be conservative
and not underestimated.) The annual growth rates assumed for each polymer
and resin are discussed individually below.
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
23
time at which the projections were made. An intermediate value of 7.0
percent was 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 was 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 was
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
23
1990 varied from 5.0 to 6.0 percent. For this analysis a value of 5.6
was selected as most representative of values given.
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
was selected for use in the analysis.
9-28
-------
HOPE. Published annual growth rates for HOPE range from 6 to 10
~~"^~ ")"\ ?(~i
percent with most analysts projecting values slightly less than 8. '
A growth rate of 7.9 percent was used. In assessing the growth of HOPE,
27
the 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 were assumed to remain constant in this
analysis.
_PS 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 was
assumed on the basis of published projections ranging from 3.6 to 4.5
23
percent.
PET. Growth in the polyester fiber portion of the PET industry is
predicted to turn positive again in 1984 and increase at an annual rate of
28
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 was assumed to 1984. From 1984 on, a growth rate of 4
percent was 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 were assumed to remain constant in order to arrive at a
conservative projection.
A sensitivity analysis was not conducted in light of the low level of
price effects as discussed below.
9.1.3.2 Projected Capacity. Projected capacity required in 1988 was
calculated by applying a utilization rate to the projected 1988 production.
9-29
-------
In order to project the required capacity, a capacity utilization
rate that will trigger additional capacity must be assumed. The capacity
utilization rates used were estimated on the basis of published
information and discussions with 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 were
used.
The projected 1988 capacities for the five polymers and resins,
calculated using the assumed utilization rates, are shown in Table 9-11.
9.1.3.3 Retired Capacity. In calculating the retired capacity, the
useful life of all plants was assumed to be 20 years. Therefore, any
capacity built between January, 1964 and January, 1969, was 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 were assumed to be replaced. For the PET
industry where there are no known plans to rebuild a portion of the retired
capacity, it was 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 were taken from the Facts and Figures of the U.S. Plastics
9-30
-------
»
3 C IB
O- Q.
CO
CO
en
c/i
LLJ
a:
a
o
a.
o
a.
o
3:
a; *J en
.- u i
V 0.33
Q£ ^O Q"\
U *t
£ >,ro
ri8?
OJ
I
01
O 10
en
-------
9 10
Industry. Values for PET were taken from Textile Organon. The
interim capacities are shown in Table 9-11.
9.1.3.5 New Capacity. New capacities are calculated by using
equation (1). The 1988 production for all polymers, except PS, was
estimated by applying the applicable growth rate to the 1981 consumption
and net exports, and summing the two. For PS, the 1981 production level
was increased by its growth rate. These values are tabulated in Table
9-11.
In projecting the new capacity required in 1988 it should be noted
that new 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
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 study
it was assumed that all affected facilities will be new plants, i.e., no
modifications and reconstructions are included. (See Chapter 5.)
9.1.3.6 New Plants. The projected new capacity requires the
equivalent of 27 new model plants. However, industry sources indicate that
not all the new capacity will be in the form of new grass roots plants--
some capacity will stem from adding new process trains to existing plants.
Insufficient information is available, however, to allow the projection of
9-32
-------
the amount of new capacity that will be added by new plants and that which
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 to make
either LLDPE or HOPE. Thus, data on plant source 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 reflecting 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 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 analysis, therefore, the projected new capacities
are expressed in terms of 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 are tabulated in Table 9-12.
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
three new liquid phase process plants are projected. The capacity of
these six plants, 765 Gg/yr, is 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;
9-33
-------
CO
CO
01
0
oo
00
LU
O
SI
>.
C2
*
OO
t
CTi
O C CO
K3 4-> CTi
i. r- ^- ^-t
O) Q.T- 1
£ 3 -Q CO
3D CTi
'Z. C i i
>)
>r-
O
4J Q. >>
"T33 ^3 ^^^
o o en
2: o
4J
C
rC
Q.
CO
CO
o
o
o
s_
a.
>^"0
r- S- S_
3 o-i- >>
CU (T3 3 -^.
z a. cr CD
13 0) O3
fc "^
S- 4->
S- 3
3 r
O
CO OO
QJ tt) <1) Q)
CO CO CO CO
fO fO fQ fO CO
g"
crco crco crcrco c I «ac
r rO *r- fO * '" fO O ^" Q.
_1CO _1C3 _J_1C3 O Cir
in o o o o
VO O <-t LO «3-
rv o <* ^
«t ««
i i ii
LU UJ
Q. Q. (
Q- O Q OO UJ
EX 13; Q. Q.
< <
1
CTi
OJ
H-
QJ
^J
oo
fO
9-34
-------
however, the use of the high pressure liquid process cannot be ruled out.
The consequent projections assume that one liquid-phase plant will be
built and that the remainder of the capacity will be added with gas-phase
plants. These assumptions result in one high-pressure, liquid-phase plant
and five low-pressure, gas-phase plants having a combined capacity of 1030
Gg/yr, somewhat more than the 1000 Gg/yr required capacity projected.
9.1.3.7.3 HOPE 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, then a batch
plant is more suitable. Since indications are that most new capacity will
probably be added in the form of continuous process plants, only a
continuous process model plant is described. The projected new capacity
would require two plants.
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
the TPA and DMT model plants are the same size, two TPA process plants and
one DMT plant are projected. The combined capacity of the PET plants is
315 Gg/yr, slightly less than the 330 Gg/yr projected. However,
considerations of the current industry conditions in which plants are being
33
closed down make it unlikely that a fourth new plant would be built by
January, 1989.
9-35
-------
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
utilization. Industry representatives indicate that by modifying present
facilities, capacity can be increased to meet demand without new plants
34
until 1986. 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. By 1988, three new plants
should meet the additional demand.
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
considered plants.
Total additional annualized costs of controls in 1988 (Table 9-16),
the fifth year of controls, for the projected 27 polymers and resins
manufacturing plants are estimated to be but $4.9 million dollars under the
combination of the most stringent regulatory alternatives for each plant.
Thus, detailed "Regulatory Impact Analysis" as prescribed by Executive
Order 12291 is not required.
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 any of the alternatives would be less than
four-tenths of a percent, except for the one HOPE liquid phase, solution
facility. The costs of Regulatory Alternative 3 for this plant is
substantially higher than those for the other projected plants. However,
even though these costs are significantly higher, the maximum price
9-36
-------
increase required to fund the most stringent regulatory alternative would
not exceed 2 percent; price increases required for the least stringent
alternative for this plant is less than 0.1 percent. Because of the minor
increases required by this and the other plants, no significant economic
impact 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 9.2.2the 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, 1S80. 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:
Emission control investments have zero salvage value.
No differential inflation occurs among the cost or revenue items.
t The weighted average cost of capital and the marginal income tax
rate remain constant during the life of the investment.
t Depreciation is based on the 1981 Economic Recovery Act,
Accelerated Cost Recovery System (ACRS) five-year rates.
t A 10 percent investment tax credit is applicable on emissions
control investment and is realized in the year following the
investment.
NPV of the model plant remains constant at the stated weighted
cost of capital.
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:
n CFy
NPV = I -- 1 (1)
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.
CFy - Ry - 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,
(1+inf)^, and constant dollar value in y=0 dollars. That is,
and,
Ry = (R0) (l+lnf^ (3)
OMy = (OMQ) (l+1nf)y (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, ITCv , is subtracted directly from
income taxes. Taxable income equals revenue less operating and maintenance
expense, and less depreciation. Thus,
Ty = t(Ry - OMy - Dy) - ITCy (5)
Substituting the income tax Equation (5) into the cash flow Equation
(2) yields the following expanded expression for cash flow.
9-39
-------
CFy = Ry - OMy - Ty (6a)
= Ry - OMy - t(Ry - OMy - Dy) + ITCy (6b)
= Ry - 0My - tRy + tOMy H- tDy + ITCy (6c)
= (l-t)Ry - (l-t)OMy + tDy + ITCy (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)OMo(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),
n (l-t)R0(l+inf)y - (l-t)OMQ(l+inf)y + tOy + ITCy
NPV = -IQ + Z - (8)
Equivalently, each term under summation can be summed individually so
that
(9)
(I-t)R(l+inf)y (l-t)OM(l+inf)y n ITCy + tDy
Q
NPV = -I + Z - Z
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, OM . 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)
(l-t)R(l+inf)y
(l-t)OM(l+inf)y
n o
n - v
y=1 c[+*\y
' n '
_ V
y=i
M+H^
- l
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 IQ.
(11)
n ITCy+t°y
y=i
= i.
i -
I0 (TAXF}
(1+d)-
.22t
.21t
.21t
(1+d) (1+d)2 (1+d)3 (1+d)4
.21t
(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, I {TAXF} is substituted into Equation (10), the constant
terms are moved outside their summations, and R is isolated on the left
hand side of the equation.
R is then expressed as
9-41
-------
n (l+inf)y
(l-t)OMQ E + I {TAXF}
RQ = (13)
n (l+inf)y
y-i
Simplifying Equation (13) results in
TAXF
R°= OM°
(1-t) E
y=i
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:
R = ONL + lie TAXF + Iin TAXF (15)
0 0 15 10 -
15 (l+inf)y 10 (1 + inf)y
(1-t) l (1-t) E
y=l (l+d)y y=l (l+d)y
where, LC = investment with an economic life of 15 years
I,g = 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
PI - -
Qn
where, PI = the unit price increase in base year dollars
R = the required annual revenue increase in base year dollars
Q = annual sales volume in units
Finally, the percentage increase in unit price is calculated using the
formula
PPI = x 100
P_
where, PPI = percent unit price increase
PI = unit price increase in base year dollars
P = 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 (MPV) method is used to estimate required
revenue increases and maximum percentage product price increases for each
model plant 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 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 plants are summarized in Table
9-13.
The model plants represent ten product-process combinations. Twenty-
seven affected plants are projected for the first five years, i.e.,
1984-1988. 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 the model plant products were derived from one
of two sources. Prices for PP, LDPE, HOPE, and PS were determined from
data contained in the "Plastics Resins Domestic Merchant Sales" table in
Facts and Figures. 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 fiber prices were not available from the above
source.
It is assumed that the PET new facility will be a fibers plant.
Determining an average annual price for the PET fibers model 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
oc
Engineering News (C&EN) reports 1980 prices for three polyester fibers:
9-44
-------
oo
h-
CL
_j
LU
Q
O
OO
^^
h 4
oo ^--
LU OO
C£. CH
Q _J
Z _J
>- < 1
o «
Q^ 1 i 1
1 1 *
1-i- _J
O rs
oo
O CO
i CO
( en
OO r-H
h t
cr: 3Z
UJ CJ3
^ ~~i
tj o
< eel
cn -s.
i _J
r i i
< =>
rx CO
LU
Qu
O
ro
I-H
1
cn
OJ
p
/**>
ro
r
o
r tO
ro O) i. 4->
3 r OJ C
C ro CL ro
C tO r-
<: CL
o
OJ
u
r-
S-
o_
C
JZt O 4->
r- C
p 4J (0
ro O r
3 3 Q.
C ^3
c o s-
eC i. aj
CL CL
C
0
r-
-t-XO
r ro
r i
40
"*^s
>«j
r 4-J 4-)
ro T- S- C
3 U O) ro
C ra CL.
C CL CL
«=C ro
U
O M-
OJ O T3
4-> OJ
ai o 4->
15 r^ rXJ ^"
O £ H- ro
S_ 3 4- r-
CL, C ro Q.
to
to
O)
o
o
S«-
a.
4-J
o
3
a
o
s,.
a.
,^ ^
o
o
o
i-H
be*
0>
^
^^^
fe^
.»^.
cn
^D
*HH«'
-*^»
c^
* '
p *«1
cn
O
*_^«
O O
0 0
« *»
CO CO
O LO
LO O
f-H i 1
CO CO
a;
to
(O
JZ OJ
CL U1
ro
-a -c
r- CL
3
G" 'Si
r- ra
_J 0
a.
Q-
o o
o o
CM \J2
M r,
LO }*r~
i 4- >
S- 3
3 i
O
to to
a) a>
10 to
ro fl3
JC J= O)
3 0
LO !-
CO to S-
, -P-
~O *P*
OJ 0 C
E <0 O
3 CL !-
(/I ro 4->
to o to o
ro O) 3
i O TJ
0) ro !- O
t. 3 i- S-
ro C CL ZL.
c
to ro M- i
CU O rO
4-) to 3
ra P C C
i- ro O C
3 -i- ro
C ^^ t '
o ai ra (/i
p- > .
4J C T- ro
ro O i- 3
rsi -i- 0) CT
P 4^ "T3 OJ
i O
r- 3 i- tO
4J -O C OJ
3 O M- p
S- ro
>, CL4-> to
4-> X
O ro 4-> ro
ro 3 3
CL C OJ c
ra C
-------
staple, textured filament, and feeder filament. To establish a price for
the plant's product mix, the analysis assumed that its sales were similar
c
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 and maximum price increases for polymer and resin
model plants to comply with potential NSPS regulatory alternatives are
shown in Table 9-14. These measures are shown with the associated
estimated investment, operating and maintenance costs, and recovery credits
for each regulatory alternative. For each model plant, 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 (Moody1s) 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
38
equity was derived by the dividend yield method.
9-46
-------
Taole 3-1-1. :QLYMERS AND RESINS MODEL PLANT CC'CRC'- :OST3a AND MAXIMUM PRICE INCREASES 3Y PLANT PRODUCT,
PROCESS, AND REGULATORY ALTERNATIVE (JUNE, I960 DOLLARS)
Model plant categories R
by product anc r^ocess a
PP
Liquid phase
Gas phase
LDPE
Liquid pnase
Gas phase
HDPE
Liquid phase slurry
Liquid phase solution
Gas phase
PS
Continuous
PET
t DMT
« TPA
egulatory
1 tar^ative
2
3
4
2
3
2
3
4
2
3
4
2
3
2
3
2
3
4
2
3
4
2
3
4
2
Investment
66.3
380.0
510.9
56.3
88.3
66.3
424.8
475.3
66.8
29.8
330.8
66.8
84.7
66.8
1,484.8
66.8
29.8
330.8
66.8
69.6
59.5
67.0
2,347.7
2,368.8
2.2S0.9
Recovery"
crecit
(.51000)
29.6
29.6
29.6
23.6
29.5
29.6
29.5
29.5
29.6
29.6
29.6
29.6
29.5
29.6
29.6
29.6
29.6
29.6
29.5
185.5
189.1
29.5
914.6
914.5
884.9
CiMd
29.0
79.0
46.5
29.0
44.5
29.0
123.7
138.8
29.0
27.5
129.8
29.0
31.0
29.0
848.7
29.0
27.5
129.8
29.0
30.9
30.9
29.0
283.1
311.1
254.2
Reauired"
revenue
increase
15.0
113.6
101.4
16.0
34.6
16.0
165.1
187.9
16.0
8.8
156.9
16.0
20.6
16.0
1,051.8
16.0
8.8
156.9
16.0
(137.6)
(141.2)
16.0
(268.2)
(237.5)
(284.0)
Maximum
price
increase
"""""
.02
.13
.11
.03
.06
.01
.08
.10
.02
.01
.15
.01
.01
.03
1.74
.02
.01
.16
.02
-.21
-.22
.01
-.17
-.15
-.18
Control costs and recovery credits are based on data contained in Taoles 8-20 to 8-28.
Investment for each alternative represents the sum of the incremental capital costs (over baseline,
Alternative 1) for the various types of eauipment as aggregated by economic life (costs were rounded to
the nearest 5100). For example, the investment for Alternative 3 of PP gas phase is 388,300 (snown above
as 88.8). This is obtained from data in Taole 8-21 and represents the sum of a 57,000 incremental cost
for flares (with an economic life of 15 years as indicated in "aole 8-3), a SIS,000 increnental cost for
flare ducting *ith an economic life of 10 years, and 356,300 for LDAR with an economic life of 5t years.
Because of rounding, the sum of the incremental costs (88,300) does not eaual the total incremental cost
of $88,000 ootainea by subtracting $366,000 from $454,000 (See Taole 3-21).
Recovery credits are taken directly from the tables in Chapter 3 for each alternative (since there are no
credits uncer baseline).
O&M represents the sum of the incremental direct costs and indirect costs (excluding capital recovery).
From Taole 8-21, the incremental direct cost for P? gas chase Alternative 3 is $41,100 (84,100 less
43,000) anc the incremental indirect cost is 33,500 U3.COO less 14,600). the total incremental O&M
cost is $--,500 (shown as -14.5 above).
9-47
-------
The calculation of the required revenue increase is based on Equation (15) (from the text). With the
substitution of tne parameters indicated in tne tsxt, Eauation (15) simplifies to
30 = OMn - .1151, - .1521- - .2481,,
where ° ° 1 ~ 2
OM * O&M (from above) minus recovery credit
I, = Incremental caoital costs of equipment with economic life of 15 years.
Ij = Incremental cacital costs of eauiorcent with economic life of 10 years.
if = Incremental caoital costs of tAQR equipment (w-th average economic life cf 5-j years).
For PP gas pnase Alternative 2, tne calculations are:
S = (44.5 - 29.5) -r .115(7.0) * .152(150) - .248(56.3)
0 = 34.5
Parentheses indicate negative requires revenue increases.
The calculation of the maximum price increase is based on Equations (15) and (17) from the text. By
substituting Equation (16) into Equation (17), the latter simplifies to:
PP1 = R or Required revenue increase
-2 x 100 x 100
P Q dollar sales volume
The dollar sales volume for each model plant is shewn in Table 9-13
9-48
-------
Required revenue decreases and price decreases indicate situations in
which recovery credits exceed annual capital and operating and maintenance
costs. These include Regulatory Alternatives 3 and 4 for both the PS and
PET/DMT model plants and Alternative 2 for PET/TPA model.
Required revenue increases and maximum price increases for all other
alternatives of the model plants are relatively insignificant except in the
single case indicated below. The maximum price increase for Alternative 2
of all model plants is less than .05 percent. Increases for the remaining
alternatives are less than .2 percent for all model plants except the
HOPE liquid phase solution model. The price increase under Alternative 3
for this model is 1.7 percent; its after-tax annualized required revenue
increase is $1.1 million.
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 COST a TO SOCIETY OF REGULATORY
ALTERNATIVES BY MODEL PLANT PRODUCT AND PROCESS (JUNE, 1980 DOLLARS)
Net
Model plant categories Regulatory annual ized
by product and process alternative cost
($1000)
PP
Liquid phase
Gas phase
LDPE
Liquid phase
Gas phase
HOPE
Liquid phase slurry
Liquid phase solution
Gas phase
PS
Continuous
PET
DMT
t TPA
2
3
4
2
3
2
3
4
2
3
4
2
3
2
3
2
3
4
2
3
4
2
3
4
2
15.7
116.6
105.9
15.7
34.5
15.7
168.3
192.4
15.7
8.2
159.5
15.7
20.5
15.7
1,067.1
15.7
8.2
159.5
15.7
(137.9)
(141.5)
7.0
(298.0)
(279.0)
(306.0)
Total
Number of annual ized
plants cost
(S1000)
3 47.1
349.8
317.7
3 47.1
103.5
1 15.7
168.3
192.4
5 78.5
41.0
797.5
2 31.4
41.0
3 47.1
3,201.3
5 78.5
41.0
797.5
2 31.4
(275.8)
(283.0)
1 7.0
(298.0)
(279.0)
2 (612.0)
These costs represent the costs of the regulatory alternatives over
and above the baseline as described in Chapter 8.
Parenthesis indicate negative required revenue increases.
9-51
-------
as opposed to the 14 percent nominal weighted cost of capital used in
determining 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, is $4.9 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 potential NSPS will not constitute a major rule within the
meaning of the Executive 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:
« Human health effects. Ozone exposure has been shown to cause
increased 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 those human cell damages known as chromosomal aberrations.
Vegetation effects. Some rates can result in reduced crop yields
from damages to leaves and/or plants have been shown for several
crops including citrus, grapes, and cotton. The reduction in
crop yields was shown to be 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 plant categories
by product and process
PP
Liquid Phase
t Gas Phase
LDPE
Liquid Phase
Gas Phase
HOPE
Liquid Phase Slurry
0 Liquid Phase Solution
t Gas Phase
PS
Continuous
PET
DMT
TPA
TOTAL
Number
of
plants
3
3
1
5
2
3
5
2
1
2
Regulatory
alternative
3
3
4
4
3
3
4
2
2
2
Total
annual i zed
cost
(S)
349,800
103,500
192,400
797,500
41,000
3,201,300
797,500
31,400
7,000
(612.000)
4,909,400
Annualized fifth year costs are based on the highest alternative cost
for each model plant as shown in Table 9-15.
9-53
-------
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.
f Ecosystem effects. Continued ozone exposure has been shown to be
linked to structural changes in foreststhe 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) and SIC 3079 as those with 1,000 and 250 or
fewer, respectively. These levels were set as criteria for extending SBA
loans and related assistance (13 CFR Part 121, Schedule A). Only two of
9-54
-------
the firms in Table 9-2 are believed to 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 small businesses would
undertake construction of new plants. Furthermore, as the analysis in
Section 9.2 explains, any potential adverse economic impacts on new plants,
regardless of whether they are 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 or, the balance of trade.
9-55
-------
9.4 REFERENCES FOR CHAPTER 9
1. Kline Guide to the Chemical Industry. Fairfield, New York, Charles H.
Kline & Co., Inc., 4th Edition, 1980. p. 150.
2. The Society of the Plastics Industry, Inc. Facts and Figures of the
Plastics Industry, 1982. p. 21.
3. Big-volume Chemicals' Output Fell Again in '81. Chemical and
Engineering news. 6G_:12. May 3, 1982.
4. U.S. Department of Commerce. U.S. Industrial Outlook, 1982.
Washington, D.C., U.S. Government Printing Office, January 1982. p.
120.
5. U.S. Department of Commerce. Census of Manufactures, 1977.
Washington, D.C., U.S. Government Printing Office, July 1980.
6. U.S. Department of Commerce, U.S. Industrial Outlook, 1982.
Washington, D.C., U.S. Government Printing Office, January 1982.
p. 316.
7. Reference 2, p. 1-3.
8. Standard & Poors. Industry Surveys: Chemicals, November 5, 1981.
p. C22-C26.
9. Reference 2. p. 12-67.
10. Textile Economics Bureau, Inc. Textile Organon. December, 1982. p.
249-251.
11. Polyester Fiber Makers Pickup the Loose Ends. Chemical Business.
Chemical Marketing Reporter. _220_(14) :9-16. April 6, 1981.
12. Business. Chemical and Engineering News. 58_:10. December 1, 1980.
13. Reference 12. _56_:12-16. September 4, 1978.
14. Reference 12. _56_:10. December 4, 1978.
15. Reference 12. 59;. 13-22. August 31, 1981.
16. Reference 12. 5£:11. November 2, 1981.
17, Reference 2. p.8. 1980-1982.
18. Reference 12. _57_:12-16. September 3, 1979.
19. Reference 12. 58:12-16. October 6, 1980.
9-56
-------
20. Reference 12. _58_:10. December 1, 1980.
21. Robert Morris Associates. Annual Statement Studies. Plastics
materials and synthetic resins. 1976-1982.
22. Plastics World. 40:61. April 1982.
23. Predicasts Forecasts, 1982 Annual Cumulative Edition. July 29, 1982.
p. B-236-248.
24. Predicasts, Inc. Industry Study: Thermoplastics to 1995. T67.
Cleveland, Ohio. March 1982. p. 3.
25. Reference 22. 40_:64. April 1982.
26. Chemical Profile. Chemical Marketing Reporter. 2!22_:54. December 13,
1982.
27. Reference 22. 40:59. April 1982.
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.
29. Reference 12. 60:17. May 24, 1982.
30. Telecon. Symuleski, Richard, Amoco, Chicago, IL. Chemical
Manufacturers Association representative for Polymers and Resins
Study, December 6, 1982. Industry Utilization Rates.
31. Chemical Engineering. April and October 1964-1968.
32. Reference 12. 60_:11. September 6, 1982.
33. Reference 12. 60:7. May 31, 1982.
34. Polyester Fiber Makers Pick Up the Loose Ends. Chemical Business,
Chemical Marketing Reporter. _2_19:9-14. April 6, 1981.
35. The Society of the Plastics Industry, Inc. Facts and Fiaures of the
Plastics Industry, 1981. p. 8.
36. Reference 12. 58_:9-10. December 1, 1980.
37. Arnold Bernhart and Company, Inc. The Value Line Investment Survey.
New York, November 19, 1982. p. 1238-1251.
38. Weston, J. R. and E. F. Brigham. Managerial Finance, Hinsdale, 111.,
The Dryden Press. 1977. p. 617.
9-57
-------
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 which have
occurred in the development background information for the new source
performance standards for the polymers and resins industry.
Date Activity
May 19, 1980 Meeting of CPB, EMB, SDB, EAB, and EEA representatives
to discuss recommendations for the development of
the NSPS for polymers and resins industry.
July 15, 1980 Meeting of EPA and EEA representatives to discuss
scope of the polymers and resins NSPS.
August 15, 1980 Plant visit to USS Novamont's polyproylene facility
at LaPorte, Texas.
August 19, 1980 Plant visit to Soltex Polymer Corporation's
high-density polyethylene facility at Deer Park,
Texa s.
A-2
-------
Date
August 28, 1980
September 2, 1980
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 15, 1980
Activity
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.
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
oly(ethylene terephthalate) facility at Kingsport,
ennessee.
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 rssins NSPS development.
A-3
-------
Date
Activity
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
April 14, 1982
April 14, 1982
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.
Gulf Oil Chemicals Company's comments on draft BID
Chapters 3-6.
Northern Petrochemical Company's comments on draft
BID Chapters 3-6.
A-4
-------
Date Activity
April 14, 1982 Tennessee Eastman's comments on draft BID Chapters 3-6.
April 15, 1982 Allied Fibers and Plastics' comments on the draft
BID Chapters.
April 15, 1982 Chemical Manufacturing Association's comments on
draft BID Chapters 3-6.
April 15, 1982 USS Corporation's comments on draft BID Chapters 3-6.
April 19, 1982 DuPont's comments on draft BID Chapters 3-6.
April 19, 1982 Monsanto's comments on draft BID Chapters 3-6.
June 1982 Project transferred from EEA to PES.
June 30, 1982 Plant visit to Soltex Polymer Corporation's high-density
polyethylene facility at Deer Park, Texas.
July 1, 1982 Plant visit to DuPont Corporation's high-density
polyethylene facility at Orange, Texas.
September 14, 1982 Plant visit to Gulf Oil Chemicals Company's polystyrene
facility at Marietta, Ohio.
September 15, 1982 Plant visit to Monsanto Plastics and Resins Company's
polystyrene facility at Port Plastics (Addyston),
Ohio.
September 29, 1982 Plant visit to Fiber Industries' polyester facilities
at Salisbury, North Carolina.
A-5
-------
APPENDIX 8
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.
5-1
-------
en
co
X
> I
s
s
o_
Q.
z.
o
t *
1
«c
Q£
LU
Q
i t
OO
z.
O
l_J
1
o
«c
a.
i i
<
i
z.
Document
E
O
4-J
re
!=
C
Q
<4-
E
l i
T3
E
3
O
1_
cr
^e.
u
re
CO
cu
JE
4J
E
P^
JE
4->
^
3
E
O
r
4->
re
o
o
E CM
f 1
.-H
a
CU JE
N CD
r- 3
S- .O
re i_
S JE
E 4->
3
tO i I
1
CO '
s_
re to
cu
to cn
cu re
> 0-
p-
4-> >
re i i
E
S_ l-H
cu
4-> E
r- O
re !-
4-J
>> <->
C co
O OO
4-J
rO »
r i 1
3
cn s.
cu cu
1_ 4J
Q.
cu re
JE JE
H- <-J
to
"0
re
a
E
re
4-J
VI
T3
CO
1/1
O
CL
O
s_
Q.
cu
JE
4->
S-
o
4-
to
r-
to
re
03
>>
c
o
4->
3
4->
re
4->
tn
CO
JE
1
<-i
1
CM
to
CU
cn
«
CL
A
I-
*
CM
E
O
r
4->
(J
CO
OO
A
CM
s_
CO
4J
C-
re
-E
CJ
E
T3
a
re
t
p^
un
>> "
Coo
0
4-> »
recM
3 V)
cni-
co cu
S-4->
a.
co re
JE -E
4->0
C75 E
O
E-o
re co
to
to to
CL 3
r- o
-E V)
lO-r-
ET3
O
i CO
4-> S-
re re
cu to
S- E
o
t/v-
3 4->
O 0
r- re
s_
re >,
> 0
E
CO CO
JE cn
1 re
CO
s_
4->
re
« E
CO
s- ~o
CU T-
4J E
cu a. o
JE re E
4J JE O
O O
>) QJ
-C? E \
r- to
T3 to
cu -a co
4-> CO E
O 4-J -i
CU E V)
<*- a> 3
M- V) -Q
re cu =
s_
>, a. cu
C JE
4-> tO 4J
tO -i
3 <4-
o to o
E CU
r- > tO
0) 4-> T-
JE re re
4J E 4->
S- CU
M_ a> a
O 4->
E re
o < '
r- >,
01 S- CO
to o
3 4-> E
o re o
tO 1 !-
- 3 4J
T3 cn O
CO CU
<: s- oo
tH
1
cn
V5
CO
cn
re
Q-
ft
cn
s_
cu
4->
ca-
re
JE
o
E
r
T3
CU
4->
E
CU
to
cu
S-
CL
CU
re
>,
4->
to <£>
3 CO
T3 1
E cn
r"
JE
CU CD
-E 3
4-> O
S-
4- -E
O 4->
TD
CO
4->
(J E
CU T-
t|-
<*- -d
re co
to
a; v>
-C 3
(J
O V)
4-> -r-
a
to
CO CO
V) S-
v> re
CO
o cu
0 >
1- !-
Q.4-J
re
-o c
E i-
re co
4->
V) r-
cu re
o
s- >>
3 C
o o
V) 4->
re
O i
r- 3
4- cn
r- CU
0 5.
CO
a. co
00 JE
4->
cu
J= >)
i- 3
ft
< 1
1
CO
V)
CO
cn
re
CL
M
co
co
o
E
re
CM
CO
to
E
O
+J
O
CO
oo r--.
LO
« i
CO CO
S- JE
cu cn
4J 3
Q. o
re s_
JE JE
O 4->
A
CM
r- <3"
O
s- -o
4-> C
E re
o
(J r- 1
14- Tf
0
00
v> c:
4? 0
P- *f
r 4-»
r- 0
J3 CO
re oo
r^
r- *»
re *3-
>
re i-
co
O 4->
E a.
re re
JE
to o
QJ
CX E <£>
>,T- CO
4-> 1
c <
cu cu
JE > -E
4-J -I cn
CD 3
C O
O tO J-
r- JE
E 4J
0 >,
r- CDi 1
4-> O 1
re i *$
EC
E 00
O JE CO
M- o cn
E cu re
11 4-J CS.
2T
LU
O
1
X
UJ
o
z.
\ i
>>
s_
O t/>
4-> 4-J
re E
i CO
3 E
cn co
CO 4-J
c£ re
4->
cnoo
E
i- 4->
'_ U
re re cn
Q. CL<-H
CO E *
l_ i i r*.
CL CO
^_
s- re ce:
O 4-> U_
4- E
cu cn
to E f">
CO E
E O
r- 1_
r 'i
CU >
T3 E
i UJ
3
O E
O
>>"-
O 4 J
E (J
CO «=C
cn
>>
i.
o
4->
re
r
3
cn
cu
i-
<*-
o
>,
C
re
c:
E
3
o
E
re to
CO
a >
E -r-
3 4-s
o re
S- E
cn s-
^. cu
O 4-J
re r
cc re
to
"O
S-
re
T3
E
re
4-J
v>
cn
E
r
to
0
CL
O
S_
Q.
S_
0
4-
to
*f
V>
re
_G
>,
C
o
4-J
3
4-)
re
4-J
oo
>,
s_
o
4->
re
r
3
cn
cu
s_
s_
CO
-C
4-J
o
o
4-> (J)
E
CuO
t-~ *r
JE 4->
to <_>
E re
o
- >>
4-> '_!
re E
i CU
co cn
cs: re
>,
sT
4-J
to
3
-a
.,
-o
cu
4-J
(J
CU
4-
<4
<
in
CU
p-
4-5
r
r
r
O
re
4-
c
E
re
to
cu
to
to
CO
o
c
i.
Cu
T3
CU
4->
(J
CO
4-
u_
^~-
Q
s_
4->
E
O
o
<4-
o
>,
4->
f
r
_
jz
re
^~
r
re
>
-------
JT
(_»
0
Q
C
O
4_>
ro
E
s-
0
M-
C
i i
T3
C
3
O
S-
CT
\^
O
ro
CO
CO
f
40
c
^~
4_>
*^
3
C
o
r
4_)
ro
0
O
E
r- CO
CO
"C* 1
O> IO
00
oo .E
3 en
O 3
oo o
r- i_
T3 -E
I ^
CO
t- CO
ro <i
1
OO *sC
CO
> c/1
r- O)
4-> CD
ro ro
E CL
S-
d) *
4-> CM
^ «
fO <*O
>s C7
S- 0
O -r-
i ^ t ^
ro O
r- CO
3 CO
CD
CO «
S- 10
00 i_
3 O)
0 4-5
i- CL
S- ro
ro _E
> O
t 1
>, r*^
5-
O 00
4-> E
ro O
r '^
3 4->
CD O
CO CO
t- CO
00 «
3 r-
O
r- <_
i. CO
ro 4-> CO
> Q.CM
ro 1
CO SZ t*~~
C" ( 1
4-» S2
E cn
4- -r- 3
O O
-o s-
oo oi ^:
4-) 4-} 4»^
O E
ro CO CM
Q. 00 1
E CO P**
1- 1-
Cv 00
r CO
ro CO CD
4-> S- ro
E ro CL
CO
E oo «
E CO CO
O >
s~ '^ r***
r- 4J
> re .E
E E cn
CO S- 3
CO O
co 4-> i-
-E r- -E
H- ro 4->
>».
c
O
4-^
fO M
r r~^
3
CD i-
QJ OJ CO
S. 4-> CO
0. 1
oo ro f^-
3 -E
O C_3 .E
r- CD
S- E 3
rt3 -r- O
> S-
-o ^
0) CO 4-»
-C 00
4-> 00 CO
3 CM
<4- 0 1
O oo 1
.^
OO "O 00
4-J CO
O CO CD
ro S- ro
Q. ro CL
E
I 00 >
o> «=t
CD-I r--
i. 4->
CO ro E
CEO
CO S- -
CO 4->
CO 4-> O
JT r- CO
t ro CO
i/>
CO
>
r"
40
ro
E
S-
OJ
40 A
CO
fO
i.
>, CO
i. 4->
O CL
4-> ro
ro JT ^t
i O LO
r: i
31 E CO
CO -r-
Sw J*
"O O")
oo (U 3
3 00 O
O oo S-
r- 3 .£=
S- O +->
ro oo
> »r- r-H
o i
CO CO
4-> S_ 00
ro CO
H- CD
O oo ro
1 i rS
OO E
r- ro >
OO r t 1
>i CL
r CO
ns r-
E CO E
ro T3 O
O -
4O g 4_)
00 O
O = CO
O O CO
CM
Ol
oo
E
O
>^ 'r
C- 4_}
0 0
40 CO
ro CO
^~
3 «
CD CD
Co
S_ !-
CO
VO 4->
3 CL
0 ro
i JC LT5
i- C-3 LO
ro I
Z> E CT^
r
O
H- C l-
O CO -E
00 4->
00 CO
4J i- UD
O CLCO
ro 1
Q. CO CTi
E S-
r- ro oo
CO
U oo CD
i- CO ro
E > CL
O T-
E -M >
O ro CO
U E .
CO S- CJi
CJ
CO -M T3
-E r E
1 ro ro
o
ro
Q_
O oo
+-> 4->
T3 =.
r <3J
3 S
CD O!
O) 4-1
C£. ro
4->
CT1CO
C
i- 4_>
S- 0-->
T3 ro O^i
a. cx<i
Q_ CO
i. ro a;
O 4O Ll-
<+- E
CO CT>
00 CO
C O
1^.^ i
0) > I
T: E
i LU
3
O
(J
CO
oo
T3 CO
E >
ro -i
* rO
CD S-
t- CO
CO 40
C r
CU ro
4-1 o
E +->
0) rO
E 3
O CD
i. CO
r- S_
LU O
01
4-3
O
Q.
p:
CO
c
o
-------
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 only the first
phase of an extended study of which a final report should be available
in 1983.
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.1.1 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 Mach 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
a
o
c
ra
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 CL, CO, C0?, NO
t- c. /\
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 SOo 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 (C09), oxygen (C09), oxides of nitrogen (NO ), total hydrocarbons
C. C. X
(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
J\
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-3, in combination with a
theoretical analysis indicate that high VOC reduction efficiencies (by
weight) can be achieved by all new incinerators.
C-5
-------
C5
s:
^-i
i
oo
UJ
!
LU
<*v
^£
I
U.
^£
T~
tj
^s^
^£
0.
LU
1
Z.
o
0
o
u_
*~*
LU
i i
I
i_^
H-
Z.
O
1
z.
ce.
\
GO
z.
i-^
f~^
z.
GO
se.
LU
JVJ
>
t
e^
Z.
Z.
o
GO
z2
ST"
LU
r I
|
O
cu
o
rC
1
0)
^
CL
r
U
c
^
s_
Q.
CD
C
«r~
4-)
(T3
S_
CU
CL
O
0)
cn
fO
i_
(O C1
£ c
* 'i"~
^* 4-J
o. to
cu
CL
O
£_
CU
cu
S
fO
S_
fO
a.
^
cu
T3
O
^^
"O
c
d)
_*y
fQ
^~
cu
(j
c
o
CJ
f .^_ ^
Q. CO O
i- >S *^~
0 f 4J
co ra ITS
J2 +-> N
rO ^O *r
u c
a o
OJ O 'r-
s. s-
ro -M CU
s. u e
4- CU ra
C r i
i i LU U.
(=
5.
a.
>« o
*s un o
tO CM r-l
1 1 1
o o o
c
o
f*"l
t.
rO
(J
0
s-
a
^a
f~m
fO
CM -^
O CM O
C_3 O t
o x
0 e£
o
CM O
CM
OS ro
t * <^O
Q. OJ ^
c * '
fQ >«)
JD T3 -M
.»- CU 4->
S. i 0
o cu o
= 100
c
o
r
^J
fQ
IN
< i O
CM 4->
(Q
r^J §
S- U
(T3 J=
O O
5^.
Sk»
J^J
C^
»
o
J.J
0
a.
OJ
g
(Q
^
Q-
E
Q.
CL
LO
i
0
^--^,
CM
0
CO
t.
cu
u
i-
1
"O
cu
<-
cu o
a. CL
CL
O C
-C O
(J !-
CO
O -
4-> U
o cu
J= i.
a. CL.
r^
a.
S o
0
0 -3-
LO 10
1 1
o o
c
o
4->
a u
cu cu
cu &-
Q--t-
OO Q
-O T3
C C
3 3
U
C
o
i-
4^
(J
cu
i C
LU O
r»
CO 4->
O ra
i 4J
c: LO
O
^ ^>
4J CU
S 4J
^ ?t3
r CU
0 3
i.
O
4->
CO
f"
E
di
_e
Lu
o
o
CM
1
O
3-
cu
3
4->
rO
S_
CU
a.
a>
i
4-1
C
cu
r
-a
5
T3
CU
_Q
T3
CU
CO
O
Q.
X
cu
r
CU
E CU
3 r
i CL
-------
!' -s\
= ~
1 (/I
O!
I/) l/l
GJ O> C
-^ -^f Ol CD
CO CO CO
cc c: «=r
<& i-o ^r i
o c C
-
o ro CsJ
0^0
^c ^- CM c
CM CM CC Cf> U^ CT~
CM X) VC *3-
ite
1 i. C OJ :/> CC CCCC X CC CC CO CC
G; 3-^,1 - -*
'^^J,
C *3 i^3 J ' JM CNJC-J CM OJ CM CM '"Si
_1 0) => CC 1
CM CM Oj
CSJ CM tM CM CM
«3- "C- ^- "sr
CM CV CM CM CM
^_, ^ fv CM
C-7
-------
en
c
« t/i
r C
«*(X
5
on
**
"c
UJ
>>
CJ -U
QJ G">
jj fl,> 5-?
4
<+- >>
UJ _O
QJ -»
O tn
c -o
(U 4) C
T? £ O
in I OJ
o; vi
GO Q£ *
l».
J
00
LU
ce
^~
c/>
LU £.- -
QJ 0
o *~" *"""' "
(/^
t-^
3L"
O
h-
<
C£
LU
-z.
CJ
^ T3
4J
_t rt3
< S-
r: oj
Q£ C
1_U T-
X U
i_ c
H -1
U.
O OJ
>- 1/1
| 3
ZD
.
H"
c:
c
j3
re
u
c
TT
5
>.
j;
CL
O
csjCNirorot «C\JCSJCNJ uf> oo >o*>cr»crt un ^onr^v 1-^01
C^C^CTiC-'OOIC'vC"' CO CNJ GC C^ U>CTi
CT> O"> ff> O1^ O^ C7^ O"v O^ O^ CO O*> CT* CT1 CTi
uo o ift LO mo in uo 10 o o o c^m
oooooooo o LOOLO OLD
OOOO OOOO O cv^^^f i^r^.
^OCO^OCOO^COO ^ ^LOLO » ^T
GJ a> QJ
4-> 4>J 4_J
Ul tfl ^l
t3 tJ (O
333
u u o
-4J 4-> 4J
tJ tj O
*O fO »O
re rn ro »-) -4-> 4-J
Q^ GJ flJ ^C OC ^T
VI u-i yi 4> 4- 4
33^ c «t or «
a. )
C. 4J i S-
>i a; >< in >, o
, i- LU ^<
O (O O U
C- S «t «C
*
^
"
o
5< X (_>
* - H
^ X OJ
s- - i - -c
Cu ^i V. ^x '
Z s- « -D s- -O «r
>^ T3 C ZZ
0)
I/)
4^
c
CsJ
O
4^
C
fO
tn
c
^o
en
en
^«
(O
.u)
C
c
o
^_
rO
^«
4->
C
ai
^
c
o
o
fO
in
(O
cs
a)
y>
3
0)
u
4^
^
O
U
0
<
«
c
i_J
c »
U tr
.1 Q;
53 ti
^r x
CJ 13 h
OJ
*j C r
C7 CJ
3 4-> >
yi C P
O
4J
VI
u
M- S.
O O
-^-- <*-
CO ^^
^* o
^t
o ^
jr c
*-> re
OJ
s o
=£
a. t/i
UJ P~
at
a ~o
CJ O
Q
Q. ui
t- C
CL 0)
S
trt ^
C -4-J
O t/i
J3 C
0 0
-o >>
o
i C
>
1 J=
O i
»-- fO
CO 0
I 4_>
c s-
o o
JC U-
4J
-1
^C 4>J
uu c
a -Q
a s
in O
0 U
a.
O 3
cu
. 1
o
o c
cu c
= en
L- oo
C O S-'
0 u- -C
p- O
in C
3 m
£U =
O 10 CU
t_) aj ^
cu >> s-
O tn
s- -o c
c. ai >
3 C
in c
o -o s.
" 1(5-
« m -
^
* 3 ^g
S. QJ S-
= i- "3
4-J U
fc ^^ o
c: w) s-
_ QJ -3
Zj "^ jz1
f C
3 ft)
QJ T:
-0=3
flj T3
r c >
u
= QJ ~rt
QJ en c
C, i-
a, QJ s-
= > o
LO < <*-
.
=
L.
*O
U
O
i.
^
>!
JT
ai
c
_^
c
o
c
r
t3
4_)
4_>
"2
*D
o
cr%
w
'QJ
"O
^
tn
*->
C
5
i.
4-J
t/1
C
c
o
i.
cc
}£
o
s_
f-*
TJ
O
*J
i-
0
*4-
QJ
^
i/)
1
C
o
«»-
CO
* 1
c
o
4J
^QJ
t^
CL.
UJ
T3
a
o
o.
o
s-
l/l
^3
O
CJ
CJ
4-J
u
O
u
QJ
>^
OJ
1_
(/I
fC
i
en
cu
L.
4_J
if
CU
~
OJ
Q
C71
fC
i.
CJ
>
^
c
o
JZ
t.
u
T3
>*,
JfT
(U
C
(C
4_)
C
o
c
f
ft3
4->
o
-4-J
w
o
u-
i-H
0
nr
o
c
0
a-.
r^
'aj
X!
J2
tn
C
§
4J
1/1
C
c
c
r.
C
W
U
c
^
~?
o
i-
U
~
.£.
4-J
-o
c
o
'J
c
*
1 1
*T
C
C!
.*_)
I/1
O
GJ
0
=
j_)
1/5
fO
CD
^3
-i
~
QJ
p
^,
OO
en
c
X
S
QJ
>
O
S-
Q.
O
W
VI
C
o
4-)
fO
o
u_
TD
O
i»
:
3
(-_
O
C
M
QJ
0
u
i.
QJ
*J
U
ts
>
^
5
o
OJ
a;
cri
^
3J
>
<
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 (GC/FID) was used for obtaining an organic analysis.
2
C.2.1 Environmental Protection Agency (EPA) Polymers Test Data
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
Chemical Producers ) to assess emission levels and VOC destruction
efficiency.
The ARCO polypropylene facility has a nameplate capacity of
2
131,000 Mg/yr (400 million Ibs/yr). 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 link, 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
-------
Jt
u
«
w
" T
u
X"
x.
w
u
UJ
w
tl 91
w w »
« s
O 91 O
a
3 a.
0) w
V =
»1
30
W
£
01 44
11 =
11
3 |
O
rO
O)
c
0)
o
c.
o
o
Ci
i->
re
o;
-(->
1/1
>,
(/)
O)
u
c
o
o
4->
-------
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
-------
1
1
1
1
1
5
C£
Lu
- >-
I 2£
Or-*
Lu O
7^- C£
oi !
^ I
t
i *
«
S
. _, 1
^_ !
£ i
l
, i
r* ]
O \
° i
O C +->
VI -- O) C
ftJ 4_) = (O
3; fQ 33 ,-_
^ CD.
U 3 O
.r- (j r- i .
j -M £ -H
<->-£
ro o o <»e
4_j a) i-
<£ Q£ M-
|
f
I
1 =
I £
"-«,
^. -4J
O C
*-> 13
(O i
s- a.
OJ
^ ^^
o
C oJ
o
J-> JJ
tn CU
t5 S
O 3
C
OJ O
i "-> r:
1 "
ID
3
I
'
O 4-»
U C
14- (0
c
.^ 1-Q.
3 O
O" *J > ^
ID
1 i»
OJ *->
1) = C
<-> .1- 01
i en (J H
' (D C 3
! 3 1-1 c
0
C S
4-)
1
1
j
I
i
1 5 *j
' ^ C
j ^- ro
[ -X i_ a.
j .^ fl
! _J 5- *J
QJ
I a? c a)
' V U 3
fO C C
3 | 0
1 ^S
1 f~>
1 ^
1
t
\
<
I
1
1
1
I
1
1
1
1
| s_
5
1 SJ
! ?
i «
! °"
i
LL.
o
O
, i
i
r^
o
LL.
o
O
**
i-H
t
LO
CM
-
'^J
O
O
LD
1
CM
LO
c-
LO
LO
7
TT
'
O
0
CO
LD
I
O
UD
OJ
en
i
«
QJ
t-
?
GJ
^
1
cn cn
I/I Ul
a. o.
LTJ
LO CM
CO --1
x | ^-^ II
«
(/I
CL
VD
n
m
i a. i o t
.^
( <
*3-
r_4
CM
CD
W)
Q.
LO
LO
LO
«a-
CO
c.
o
>-H
1
^
«a-
t ns i i '
Q_
JM:
C
m
O
14
1
£2
LD
cn
GJ = T3 (C > rC
<: ctr:
S d
yi Ci
I/1 _
QJ =
1- «C
D- ^
4->
OO
1 1 1
^ ra
ec a
^ 3
O O
*" ^
^- 5
51
i_
^.
\
1
i.
Si
J3
<
r t
**
U
0)
l/l
en
CO
I
LO
1_
Jd
J3
CM
«-)
-,'
U
QJ
*S>
cn
CM
«3-
OJ
en
(t3
i-
QJ
>
«=C
(/i
'o
u
c
en
^D
CM
i t
t
S_
JZ
JD
cn
LO
CM
0
<:
U1
er
_j
c
r^j
en
i |
i
i_
-Q
O OJ
*-H C
i a;
o »
cn >>
^ fi
1 T3
-------
i
cr
u_
<£
1
Q
O
~~\
LU
00
cc
CD
~ ^-
OO l
LU <
1 Q
Z 00
o oo
.-« LU
oo o
00 O
Z 0.
LU
00 73 SS
C£.
z o c
T5 OJ
O I
^
Q_ OJ
£- t
CD +>
r i.
re CU
&- £
-w u
'O C
Z '
i
_
! "" s-
= 0
OJ 4-^
o- s-
OJ
"2 =
' re u
i ,- =
I .£ ^
' ~* i_
i § *
c_j
i
i
0
, <
OJ S-
jj 0
in j->
; 5 i.
CJ
1 ,^ .p-
' -1^' t)
f fj
' 4_l ""
1 "^
1
J
1
' i-
CU
4-J
0)
I ~
(T3
t.
03
C^
1
, ,
O
CM
i 1
^_s
(_}
0
,_,
i <
C\J
^^.^
!_]_
O
O
=0
CO
l
0
ro
en
u_
O |
uo
CO
ej
0
CT\
CM
u_
0
L^)
4 I
o
*^"
,,-«»
u_
o
IT)
LO
-H
1
O
-T"
1 i »
>_j
O
^
is
CU
C1
re a;
a) £
<: ce
OJ
^_>
itJ
^
2-
aJ
h-
-O
o"
5^
c
""
r*^.
i
n
re
0-
cc
7
i
CO
*s
«
Q.
0
CO
T 1
^*
*o
£- 1 l
0
CM
cn
*[n
CL
0
r-H
1
LT3
re
i a- t
s
1
<<5J-
n
t i i
cr.
1/1
O
-u_^
re t i
Q_
n
^
a; ~-
cn 3
ro ^) s
C7^ Ol
0 0
in oo
CM ^
ro kC
i i
l f
i i
OJ =
re
Q X
=i rr
O3
-i-^
ce
~
T
a;
^
00
11 1 I 1 11
II 1 1 1 II
u
cu C
(/I (J
CO
E C
C
cc o
« < LO
II t 1 1 1 T-H
O l
1 O
cn o
c -
o
O ,-4
" "
. «* ^
S E
U U
(/I LO
§0
0
o o
esj m
u u
OJ CU
t/l V) t 1 1 (1
^VV
<3- CM
S2
1 1 1 1 1 ' 1
01= c cu c
cr. 5 ci zn ~
s- -= ^> i. i- il ci
tu x c CL' cu D =
-a r: cr.
cn re i-
= o ? °
O
*\ M j>
U"
2*
"""
re
s- C
^ CsJ
re
o c
c -*^
"*- re
i^ i_
re o
i1 2 =
c u
*-» 1/5
in
js a> *j
^ i- LA
^ a. re
3
CtJ VI
C JJ 4^
*n '^ ;
U T3
c <:
0 S- ^
OJ =
OJ U
3 *-> 3
u- J* -3
GJ
LQ 4-J ^
U
' ^T> "C
cu ^
-S T
5 3 3
1= C2 C_J
re .c u
C-14
-------
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 C~
from higher hydrocarbons, but gives a single value for all nonmethane
hydrocarbons. After separation, all carbonaceous material is combusted
to CCL which is then converted to CH. 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
X
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 3-5. ARCQ POLYMERS "ICIUERATOR DESTRUCTION EFFICIENCIES FOR EACH SET OF ;ONDITIONS
Percent Destruction Efficiency3
Calculated for Each Method
'lethod 13 (on-site)
b c
Conations HC
AV, .'iG Via ?9.3?""7 r .00003
^ J J ^ J r
-.-I tS/'-.a >'3?.9?79 : .0004
l.SOG-F
AW/NG/WG >99. 99721 i .00009
l,oOO°F
'JG'WG 99.3 - A
',3 ,.3 -33.75 t .07
-,, ',, ?3.396'4 r .30007
-i '. j ^ 9 3 . 390 ~ . 304
-,,/u -99.3975 ± .0001
-pr-pnf 1P9 'rurTl'on pfficlpnrv = TOO - -^
Hethod 13 (off-site)
Syron Byron Speciated
THCd NflHC6 Metiod 25f HC9
39.994 - .302 99.397 r .002 99.344 - .006
99.996 - .001 99.993 r .001 39.3 r .4
99.9961 - .0003 99.9957 ± .0002 99.6 ± .2
99.9 ± .1 99.6 ± .4 76 r 20
99.3 - .10 99.33 : .04 56 : 10 99.33 r .04
39.9941 t .0001 99.99796 : .30005 96.32 - .08
39.983 t .007 39.983 - .307 98-3
99.994 ± .002 99.995 ± .003h 99 r 1 99.9979 = .0001
qC in Stack qas
(gC in Atactic Kaste + gC in Waste Gas)
.vnere: -jC = grans of organic carbon
'-.s numbs - following the : sign is the standard deviation (statistically expected true value would fall between the
-eoorrea ^alue Tinus the standard deviation and the reported value plus the standard deviation).
D7)'i) indicate that the 70C /ias below the detectable
'"it ana ~.ne detection level .vas used to calculate the DE's.
""easj'-ec jsing :he Syron Instruments 'loael 90 sample collection system and the Bryon Model 401 Hydrocarbon Analyze*-
SaTj'ing s/sten and instrument conbination (utilizing reduction to methane and ~IQ) in the total Hydrocarbon ,'THC)
oce.
"''aasu-eJ jsm: tie 3yon "oaels 90 and 401 combination (utilizing reduction to lethane and -ID) in the nonmethane
lyarDcaraon -oae.
''ea3j--ecj jsn: EPA Metnoo 25 for total gaseous nonnethane organics (TGNMO) utilizing GC-FID. Data not believea to
-ecresent true /alues.
3"633.''9a js^ig ornoosea EPA letnod 13 ;of-f-site) for individual hydrocarbon soecies utilizing GC-FID.
"j1"*^/ t ies vi th analvsis - Based on ^ost jrobable 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.
4
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 naleic
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.
i. 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
-------
15ft- 6in'
12 ft
FU3W
SIDE VIEW
(Inlet)
17ft-Sin
(Outlet)
23 Ft-3.5 in
There are Three Thermocouples Spaced Evenly Across the Too of the Firebox.
The Width of trie Firebox is 6ft-6 in.
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
D
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
n
the total organic acid (TOA) train. A Teflon 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, March 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.
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
-------
i =
CM CM CM CM CM
( Ol OH- O
O O O O O
o o o o" o
i *
= -o
* CT> <*
O O -H C
CM CM
£ r!
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
GC/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
Petroleun 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 flJ/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
3
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 Beckman 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°FS 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 (ppn) 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
7
>
8
naleic anhydride unit at Bridgeville, Pennsylvania, were disregarded
as not accurate because of poor sampling technique.
C.2.3.1 Petro-Tex Test Data^. 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. 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
Redrcalation
Air Ouct
REC:RCUUTION
AIR FAN
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 §2 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 'Oxo1 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-2
-------
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
90 m/s ( 0.2 ft /min). Sample gas was drawn until at least 0.28 m
(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 CH,, C2's, C^Hg, and CoFL, the scrubbed bomb was analyzed
for N?, air, 0?, CCL, and CO, and the bubbler liquid was analyzed for
acrylonitrile, acetonitrile, hydrogen cyanide, and total organic
carbon. The gaseous samples were analyzed by gas chromatography.
C-30
-------
^^fessssss.--" '"-z-L-^X
TJ
31
O
>>
«
ca
-------
O UJ *>
~2|
2| g
03 .22
u
OJ
c
o
s_
u
t
:l
»
<
uu
1
1 EXCHANGEH
S? ^
T
3
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 sone 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
at Residence Tine
0.75 second
Flow . Temperature
RegineD (°F)
Two-stage
Backmixing
Complete
Backmixing
Plug Flow
1300
1400
1500
1600
1300
1400
1500
1600
1300
1400
1500
1600
Ethyl
Aery late
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
Ethylene
92.6
99.3
99.9
99.9
84.4
95.5
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
in Percent
0.5/1.5 sec
Ethylene
37.2/97.6
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 taDle tiers calculated
frvn these equations.
^Three flow regimes are presented: two-stage backmixing, complete backmixing, and
plug flow. Two-stage backnixing is considered a reasonable approximation of actual
field units, with complete bacKnixing and plug flow representing the extremes.
C-34
-------
at its Santa Ana, California polystyrene plant. The samples were
taken using a MDA-808 AccuhalerR pump while velocity was determined
^
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 ethyl benzene 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 ethyl benzene 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
13.2 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. 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 (CH.) 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 370°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*7). 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 INCINERATOR
Rohm and Haas Incinerator
Inlet Outlet
Union Carbide Lab Incinerator
Inlet Outlet
Compound
Propane
Propylene
Ethane
Ethyl ene
TOTAL
(Ibs/hr)
900
1800b
10
30
2740
(Ibs/hr)
150
150b
375
190
365
(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%
93.8%
aTable 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 C09 and H90. These last two items are worst case assumptions.
h
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
14
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 met the criteria. This is
significant since, though the L.A. units represent many tests, they
represent the sane 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 natch
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 nay 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-Q33a, 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.
C-41
-------
Table C-9. EQUIPMENT COUNTS AND EMISSIONS FOR
FUGITIVE VOC EMISSION SOURCES IN SOCMI MODEL UNITS
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
Emissions
(kg/hr)
- baseline:
- uncontrolled:
Revi
Equipment
Model
Unit
A
15
8
7
362
99
131
132
13d
lld
1
1
rt
104e
1
f
26f
600
3.2
4.5
sed SOCMI
count for
Model
Unit
B
60
30
30
1,450
402
524
524
5°d
42d
4
4
Q
415e
2
f
104r
2,400
12.2
17.3
Fugitive Anal
Model Unit0
Model
Unit
C
185
92
93
4,468
1,232
1,618
1,618
157
130d
13
14
O
l,277e
8
f
320T
7,400
37.9
53.3
ysis
Average
Revised
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
Equipment components in VOC service only.
From Fugitive Emission Sources of Organic Compounds - Additional
Information on Emissions, Emission Reductions, and Costs, EPA-450/3-82-010,
April 1982.
C52 percent of existing SOCHI units are similar to Model Unit A;
33 percent of existing SOCHI units are similar to Model Unit B;
15 percent of existing SOCMI units are similar to Model Unit C.
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,33; 8,11; C,33.
eAll open-ended lines are considered together with a single emission
factor; 100 percent controlled at baseline.
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
-------
- =
S i
CT> f"** "*
3 JZ O! > 1)
tr^-c^C'aaj -c
"O OH i.' V"1 ^"^
C-44
-------
FOOTNOTES FOR TABLE C-10
Equipment components in VOC service only.
Assumed all pumps in light liquid service.
cAssumed one-half of valves in vapor service and one-half in light
liquid service.
A
Assumed all safety/relief valves in vapor service.
eSeventy-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,33; B,ll; C,33.
All open-ended lines are considered together with a single emission
factor; 100 percent controlled at baseline.
^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.
Calculated using SOCMI average revised emission factors for each
equipment component given in Table C-10.
C-45
-------
C. 6 REFERENCES FOR APPENDIX C
1. McDaniel, M. Flare Efficiency Study, Volune I. Engineering-Science.
Austin, Texas. Prepared for Chemical Manufacturers Association,
Washington, D.C. - Draft 2, January 1983.
2. Lee, K.W. et al., Polymers and Resins Volatile Organic Compound
Emissions from Incineration: Emission Test Report, ARCO Chemical
Company, LaPorte Plant, Deer Park, Texas, Volume I Sunnary of
Results. U.S. Environmental Protection Agency, Research Triangle
Park, North Carolina. EMB Report No. 81-PHR-l. March 1982.
3. SRI International, 1982 Directory of Chemical Producers.
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. Contract No. 68-02-32814, March 1978.
5. Blackburn, J. Emission Control Options for the Synthetic Organic
Chemicals Manufacturing Industry, Trip Report. U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina. EPA
Contract No.
68-02-2577, November 1977.
6. Scheil, G. Emission Control Options for the Synthetic Organic
Chemicals Manufacturing Industry, Trip Report. U.S. Environmental
Protection Agency, Research Triangle Park, North Carolina.
Contract No.
68-02-2577, November 1977.
7. Letter from Lawrence, A., Koppers Company, Inc., to Goodwin, D.,
EPA. January 17, 1979.
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. p. C-7 and C-8.
9. Letter from Towe, R., Petro-Tex Chemical Corporation, to Farmer, J.,
EPA. August 15, 1979.
10. Broz, L.D. and R.D. Pruessner. Hydrocarbon Emission Reduction
Systems Utilized by Petro-Tex. Presented at 83rd National Meeting
of AIChE, 9th Petrochemical and Refining Exposition, Houston,
Texas, March 1977.)
11. Letter from Weishaar, M., Monsanto Chemical Intermediates Co., to
Farmer, J., EPA, November 8, 1979.
12. Lee, K., J. Hansen and D. Macau!ey. Thermal Oxidation Kinetics
of Selected Organic Compounds. (Presented at the 71st Annual
Meeting of the APCA, Houston, Texas, June 1978.)
C-46
-------
13. Letter and attachments from Bowman, V.A. Jr., Mobil Chemical
Company, to Fanner, J.R., EPA. September 9, 1980. p. 13-16.
Response to Section 114 letter on polystyrene manufacturing
plants.
14. Memorandum from Mascone, D.C., EPA. June 11, 1980. Incinerator
efficiency.
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.
C-47
-------
APPENDIX D: EMISSION MEASUREMENT AND PERFORMANCE TEST METHODS
I. Process VOC Sources
D-l
-------
APPENDIX D: EMISSION MEASUREMENT AND 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.l.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 Monmethane 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.
1-0.1.3 Emission Test Experience
The EPA conducted an emission test(l) 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 COp 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. Mo
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.
D-3
-------
II. Fugitive VOC Sources
II-D.l Emission Measurement Methods
II-D.1.1 General Background
A test method was not available when 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 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 leaks 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.
Once the basic detection principle was selected, it was then
necessary to define the procedures for use of the portable analyzer.
D-4
-------
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
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
D-5
-------
approximately TOO or 1000 ppmv levels. After the measurement distance
was changed, calibrations at 10,000 ppmv levels were required. Conmentors
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^ ' ' . 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 p_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. The
variation in response factor is not a significant problem because
ambient concentrations around leaks are usually much higher than the
leak definition and much lower when no leak exists.
An alternative approach to leak detection was evaluated by EPA
during field testing/ ' ) 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
recorded on a portable strip chart recorder. After completion of the
walkthrough, the local wind conditions were used with the chart data
D-6
-------
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.
II-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. '
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.
11-0.2 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
(9 12 13)
their effectiveness in leak detection. ' ' ' The systems consisted
of both remote sensing devices with a central readout and a central
analyzer system (gas chromatograph) with remotely collected samples.
The results of these tests indicated that fixed point systems were not
capable of sensing all leaks that were found by individual component
D-7
-------
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 ara 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.
0-9
-------
II-D.4 REFERENCES
1. Joint District, Federal, and State Project for the Evaluation of
Refinery Emissions. Los Angeles County Air Pollution Control
District, Nine Reports. 1957-1958.
2. Wetherold, R. and L. Provost. Emission Factors and Frequency of
Leak Occurrence for Fittings in Refinery Process Units. Radian
Corporation, Austin, TX. For U.S. Environmental Protection
Agency, Research Triangle Park, NC. Report Number EPA-600/2-79-044.
3. Telecon. Harrison, P., Meteorology Research, Inc., with
Hustvedt, K.C., EPA, CPB. December 22, 1977.
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. DIB Report Number 77-CAT-6. December
1979.
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.
6. DuBose, D.A., and G.E. Harris. Response Factors of VOC Analyzers
at a Meter Reading of 10,000 pprw for Selected Organic Compounds.
U.S. Environmental Protection Agency, Research Triangle Park,
N.C. Publication No. EPA 600/2-81-051. September 1981.
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. May 1981.
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. September 1981,
9. Emission Test Report: Dow Chemical Company, Plaquemine, LA. EMB
Report No. 78-OCM-126, December 1980.
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.
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. May 1981. EPA/IERL, RTP, NC.
D-10
-------
12. "Emission Test Report: Sun Petroleum Products Co., Toledo, OH,"
EMB Report No. 78-OCM-12B, October 1980.
13. "Emission Test Report: Union Carbide Corp., Torrance, CA." E?1B
Report No. 78-OCM-12A, November 1980.
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 capital and operating costs for flares, thermal
incinerators, catalytic incinerators, shell-and-tube condensers, 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; and E.6,
piping and ducting. Sufficient detail was presented in Chapter 3 for
ethylene glycol recovery systems and 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 3-2
and 8-3, respectively.
E.2 FLARE DESIGN AMD 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 3tu/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 */ere 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.
These equations were simplified to functions of the following waste
gas characteristics: volumetric flow rate, lower heating value,
temperature, and molecular weight. 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
2
generate a lethal radiative heat level (1500 Btu/ft 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. Natural gas to increase the heating value to 115 Btu/scf was
considered necessary to ensure combustion of streams containing no
4
sulfur or toxic materials. 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
Q
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. In such cases, a fluidic seal, which requires a
greatly reduced purge rate, was used.
E-2
-------
c. c. a.
y- *- o
^
c
- o-
Ji O "***
3: fi
1 C
O U h-2
o
A " ° X
A en
= » 77
Fc *- o* ^>
C- u.
+ "" "cr> ~ x
0> 01* =>S ~- "
?" i2 2 o" j;,1
cr « _j r^ i i
<*.
1
* c
c1 °"
cr x
X V
r^ c C-
* it *
4. 31 »
o- ?
01 3
Z^ x
X 0
CT cr
5 3
S £ H
fl O
ng »
fl CM
-So"
' ""ci o co"
. u
iS -^S
i =n u »*
, *- CM
'O- 01^
-^T" -^- *
CTI if- * ^
C CM fl CO
Cr i -o
o c -
cr> x (/i
CO (J *
uo Q) m
VI CM
CSI_^
"
S
*
cn
J^
X
CTICM
F a;
cr ^*
x x
O O
X X
'O CC
f** CO
LD -^
^v
>
CM
CO
CO
5
i : >
u1!*?
03
E-3
-------
Footnotes for Table E-l
Standard conditions for flare design and cost calculations: 70°F and 1 atm.
3Auxi"Hary natural gas requirement assumes 115 3tu/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 (1000 Btu/scf at 3Z°F). From an energy
balance
(Q x LHV ) + (Q x LHV ) = (Q + Q ) x (115 Btu/scf)
v^ng ng wg wg ng wg
(scfni)
'115 - LHV (Btu/scf)
wg
LHV (Btu/scf) - 115)
x Qwg (scfm)
^Temperature of mixture approximated by assuming uniform specific heats
per unit mass.
From Enviroscience (Reference 1, Appendix A)
(2.72 x 10'3) x Q.. (Ib/hr) x /ff (°F) + 460)/MW
D(in)
/e (fps)/550]2 = (1.818 x 10~4) Va2
Substituting,
(2.72 x 10~3) x Qfl (scf/min) x MWf] (Ib/lb mole)x(60 min/hr)x(Va/125)
(387 scf at 70°F/lb-r,o1e) x,/( 1.818 x 10'4)
simplifies to the relationship given.
eSelect next larger standard size than calculated diameter unless calculated diameter is
within 10« of interval between next smaller and next larger size, if so, select next
smaller standard size.
fActual exit velocity V (fps) = Q (acfs)/A(ft2)
'°F) * 460) acf min
= Q (scfm) x
' 'g
(7D°F -r 460)
TT x [D(in.)]' x
scf
U4
E-4
-------
Footnotes for Table E-l
-c-om p (in. H,,0) = 35 [VQ (fps}/550]2, fas in c.)
'Fron Enviroscience (Reference 1, Appendix A) assuning wind velocity,
V , of 50 npn
IAN
-1
1.47 (fps/mph)xV, (nph)
bou y^p/sb
= TAff1
"1.47 ;50)"
V
e
'From Enviroscience (Reference 1, Appendix A), assuning flams enissivity.
s, of 0.12, flame radiation intensity, I, of 1200 Btu/hr ft
H /qf1.q Ob/hr)xUVf1jg (3tu/lb)xc
V (12.56) I
[q,, (scf/min) x LHV,. (Btu/scf) x 60 nin/hr x 0.12] -3.33 x 0 x (V /550) CDS
1 1 .g _ 1 1 ,g _ ,, _ e
4;r (1200 3tu/hr - ft )
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 3tu/hr-ft
including solar radiation ( 300 Btu/hr-ft ), a safe working level, (Reference 12):,
/[Qfl q (scfm) x LHVf1 (Btu/scf) x 60
-------
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
q
diameter according to available commercial equipment.
Steam was added to produce smokeless combustion through a combined
mixing and quenching effect. A stean 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 (or the intermittent
stream if no continuous .flow was present). Availability and
deliverability of this quantity of steam was assumed.
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 (approximately
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,
while the length of pipelines to the flare was based on the horizontal
distance required to provide a tolerable and safe radiation level for
2 9
continuous working (440 Btu/hr-ft , including solar radiation ).
Piping and ducting were selected and costed as outlined in Section E.6.
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
>vi
10
Q
November 1982 and presented in Table E-2. A cost was also provided
for one additional case of 60 inch diameter and 40 feet height.
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 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, i.e., self-supporting flares and one
for heights of 50 feet and greater, i.e., guyed flares.) 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
E-6
-------
>-
CQ
to
o co
LU C£
0 t
0 0
cc: o
o_
CM
LU CTi
\ -H
-.
LU "
«£ O
~T" "^
O i i
CC
ID >
a- cc
LU
I 2:
LU CC
c n -~)
o cc
22 _J
LU O
CC
_] i i
1 1 ^^
j
«c
CM ^
1 0
LU i
\
co ec
i ~^
jz
os
1
CD
.J£
S_
o_
cy
co
'O
0
s-
3
0.
, «.
4>J
q-
N^_^
4_}
e'-
en
_O)
CO
fl3
Ll_
<
b
r
i_
CT-
0
CO
3
4->
CQ
CD
0
0
CM
M-
0
CO
»-^^
3
-I-J
CQ
O
o
LO
t 1
M-
(J
^^
3
cc
o
o
o
t 1
(J
co
-^^
3
-l->
CQ
0
o
0
CM
c^.
0
CO
^^
3
4->
CQ
0
O
LO
t t
r I
0
CO
3
4J
CC
CD
O
O
T-H
X «
-C
^~
5 i
"^ "**-^*1
-
2. o;
- it!
r
/*~^
CO CO LO CM !^*
tr. o LO LO o
LC CM Cn CO O
LO CO CTI CO O
faO- CM CM
CO O OJ O ^J-
Cn ^O O^ C7^ LO
LO LO LO LO LO
LO l~-~ CO CO LO
V*- »-H CM
co o r~- LO CM
en LO en LO LO
LO LO o LO en
LO r-^ co > i co
V3- ^H f-H
O LO LO O LO
CM CM CO LO LO
O O LO O LO
CM CM CM *f LO
O O 0 O O
CM CM CM CO <3-
o o o o o
o o c o o
CO CO CM CM CO
CM LO I 1 LO *3-
i i CNJ *3-
CXI CO ^ LO CO
en
^f-
o
co
CO
en
CM
o
,3-
CO
LO
CO
f-*
^^
CM
LO
CO
o
1
o
LO
0
o
o
o
f-*^
o
T-H
1^,
r-.
CM
r^.
*^-
r 1
LO
o
T-H
*^-
co
LO
LO
t-H
CO
o
o
o
CO
o
LO
o
o
CO
0
0
( 1
CM
f-H
CO
LO
«*
p^
LO
CO
^J-
LO
en
^j-
o
CO
LO
LO
CO
LO
T-H
LO
en
o
f-v.
0
?~~.
CM
1^
CO
T-H
<^-
« 1
f^
f^
r^
^j-
LO
LO
CO
r^
LO
LO
CM
CO
,-j.
^»
O
CO
0
T-H
o
CO
o
CM
en
[*-^
T-H
LO
t-H
^J-
LO
0
LO
co
^^
g- (^
*-) i
t- a.
^
-r;
CO (O
5-
IT3 CO
1 1_
CT-. CU
TQ
CT)-O
t^ r^
4.^
5- jc
O ->
Q.-I-
CL 2
3
CO CO
1 CO
ci- S_
i fO
CO r
CO Cl_
;_ T,
O CO
^ ^)
ai Ei
^.
(O i-
0
CO
CD CO
r- t-
13
S_
o s_
CO
-t-J (->
CO (T3
CO CO
4- S_
CT1
O
*± T3
Cu. (^
C
CO O!
JT L|-
t^ 0
CO LO
S- 0
c
<4- CO
co JCT
Lj C71
r *i
S- CO
O_ -C
(O
E-7
-------
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
Themal 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 (1600°F).12 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
discjssion , but the combustion calculations, etc. are not detailed
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 was added to any individual or
CDmbined waste stream with both a lower heating value between 13 and
E-8
-------
Table E-3. CAPITAL AND ANNUAL OPERATING COST ESTIMATION
PROCEDURE FOR STATE-OF-THE-ART STEAM-ASSISTED
SIIOKELESS FLARES
Item
Value
(1) Flare purchase cost, C'-,
(Oct. 1982S)
(2) Flare installed cost, Cf1
(Oct. 1982 S)
(3) Total installed piping costs, C
(Aug. 1973 S) p
(4) Total installed ducting costs, C,
(Dec. 1977 S) a
(5) June 1980 Installed costs
(a) Piping
(b) Ductingc
(c) Flared
(d) Total flare system cost, C
Select fron Taola E-2 if value given
or jse equations:
(3905.7) * 35.054) H x D + (900.36) 3
- (126.38)0% for 20 H 40 ft.
(6275.6) * (224.10) H + (12.782) H x D
+ (24.856)0% for 50 H 200 ft.
C1
fl x 2.1
Method of Appendix E.6
Method of Appendix E.6
C x 1.206
C. x 1.288
a
Cfi x 0.318
sys
Annual i zed Costs9
(I) Operating labor,
(2} Maintenance, C^
(3) Utilities
(a) (Q) Pilot
(b) (Q) purge9
(c) Cost natural gas, C
n.g.
(d) Cost steam, C ]
s LIU
(3) Taxes, adnin. & insurance
(5) Total operating,C
620 hr/yr x SiS/hr = 311,160
0.05 x C
svs
30 scfh, for 2 < 0 < 3;
160 scfh, for 10 < D < 20;
240 scfh, for D = 24;
320 scfh, for- 0 = 60.
[(0.3272)(Dft(in))2- (Q^
3.149 (Q«J>x + 53.45 C(Q
+ (Q^g.) Purge)]
3.296 0 scfin x MW x wtj^']C cont.
L J -ri.g
x 50
op
sys
:,
!
x 0.04
n.g.
stn
'tax
(7) Total annualized, C
tot
E-9
-------
Footnotes for Table E-3
a
For installation cost factor breakdown, see Table 8-2.
Updated using Chemical Engineering Plant Cost pipes, valves and fittings
index from August 1970 (273.1) to June 1980 (329.3).
c'Jpdated using Chemical Engineering Plant Cost fabricated equipment index
from December 1977 (226.2) to June 1980 (291.3).
Adjusted using Chemical Engineering Plant Cost Index from October 1982
(317 estimated) to June 1980 (259.2).
eFor annualized cost factors, see Table 8-3.
Based on vendor information for pilots without energy conservation
(Reference 9).
^Ensures continuous' flow of at least 1 fps for flare with any continuous
flow:
f {(D(in.))2 x .n2. 2 x (1 fps) x (60 sec/nrin) - [Qfl n (scfm)] .ieo min/hr
t v itt in i i .y CUMLJ
(scfm) x 60 min/hr x 8600 hr/yr + [Q .. .(scfh) + Q (scfh)]lx 8760 hr/y
ng pilot purge / J
520 °R scf at 60°F x 1.040 Btu (HHV) x $5.98 x (1 3tu)
530°F scf at 70°R scf at 60°F (106 Btu (HHV)) 106 Btu
Assumes steam at 0.4 Ib/lb of hydrocarbon at maximum continuous
flaring rate for 8600 hr/yr:
Q . (scfm) x MW . x f w>° x 8600 hr/yr x 60 min/hr
cont cont
f wt>y° HC \
I 100S )
V /cont.
x (lb-mole/387 scf at 70°F) x (0.4 Ib steam/lb HC) x ^QOQ I ^s team)
x $6.18/(1000 Ib steam)
c_
-------
= - b
= £ to
2 c r
CJ 3
O CT '**>
OJ ^3&< 0) -C j
^Z0 5 « I
U C II J3 O
I
OJ »< CJ 3
ff"ff
^ O !i ^ O
Ci -^
te = en ^. ai
S- -n wi .]
^"
. cc ;> g
c\j ^- «
O O O
< CO
*r o
<£> 0s CO
CT"i O^ O^
^ cc <"*
3- ys LT-
o> o^
// ff II
C C
XXX
lai
cxj O ^.
9-oj]
' a; c a> = ^J
- :
, "
o.- *^ i ^ S
E-n
-------
r
O T5 ^3 i
.
ul . i/i 3
(B C
ic To
JZ O II .3 G
_ T3 C
'»; > c_;
01 »<
-------
50 Btu/scf at 0°C (32°F) (about 25 and 100 percent of the lower explosive
limit) and an oxygen concentration of 12 percent or greater by volume.
Dilution air was 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 combustion air, 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. Minimum
auxiliary fuel requirements for low heating content streams would be
set at 5 Btu/scf to ensure flame stability.
The design procedure for streams able to maintain combustion at
370°C (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 lower
heating value of the waste gas for streams with heating values of
650 Btu/scf or less. For streams containing more than 550 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
14
given by IT Enviroscience.
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
15
according to standard industry practice. The design procedure
E-13
-------
TA3LE --5. GENERALIZED WASTE JAS :Or!BUSTICN CALCULATIONS
Sasis: oer 100 'b waste gas (w.g.) Strean I.O.:
Assuiiotions: (1) 18 percent excess air ,'1 0,:3.76 'I,, by volune or by nole);
known waste gas coinposi fon of C, 4, N, 0, (in Ib-ioles of Jton per 100 Ib f.n.i;
2
i 3 ' ^'6> d.18)] (0.013) JiV^a (29 ft£fe^)
i.e., ;C -> j) > ^ " Ib dry air
= (2.113) C + (9.539) H - (1.059) 0 * (0.377) AIR =
if TO air required. 13.02 Ib ,H, ,.. , ,. .,., Ib H.O a ,. Ib dry air, _ ,Q ,., , ,. -77. ...
HA i u - - - - \ * I in ii\ I lUBUiJ) C. \ ^J 1 K mi-TI n "~~) ~ \Jf>Jl} n ~ i J. J / / I -^In
\ j. U D^'TlO f3 i a1, . ' .' ! D~TIO I c
i.e., 1,1, + ji-T " ^ dry air
,, . i" air required 23.02 Ib rN + f(. + H _ 0, ,3 7g> ,, ig,-,
i.e., (C * j) > -j-'
= (124.3) C * (31.08) H + (14.01) N - (62.15) 0 =
if noair reauired: 2S^
i. e., ;C * T; < 7
-.e., L - j)>3
i* "0 air reauired. 32.30 Ib -0 ,r + ^, = (-32.00)C - (3.00) -\ * (16.00) 0
3:a; : prog. _ 'b products (wet) . Ibl CO., - HjO T '(,, +0,,)
w.q. 100 Ib w.g. 100 Ib w.g.
"ojias OT" ^ater pe" pouna of dry air at 30°F and 60" relative Humidity.
E-14
-------
re
14-
in
3
4-i
ca
r~\
US
^
oo 3:
z: H-
UJ CX
5 i i
uj 'J cr
co z c:
o i^
1 00 =»
33 3:
UJ CC _J
33 S'
CO O 00
UJ UJ
O OO 33
O at -J
C£ O
C£ O
^O ^ *-"H
IS LU
1
o: O
_LJ __1
-_
2
GJ
3
^>
es
53
^j
i
CT
3
t^
x
^_
w *
2 ~
S T7
~~;
- 4-
X
o 53
CNJ 3-* " - i-
_r - en c.
CSj ^
^ O 0 r*. S* 3
ro csj i ?- CT
l .Z CO O tlLC
0 ^H ,-H r- C
*-» J3 Or
-f X i C
X , O -t-
^f CSJ O - U3 -
- > ^ O T3
> « o * < i s- o -c
i^-* _j * - T3 e. E=
1 ;» ^^- * X r < it II 3 4-
'^-3=CM>30-h^-^ '-)
S. j CSJ ~ ro O CVJ - - - * il 3") (/»
^1 SI » ' O _J *S * ' CNJ O -
- -J cn .^-- O o en ^ 3
x -o o + o *r cr
3 csj t -~- i .5 3 01 - 01
'OfOCOCNJO' '^O
i iu i ^H ^ ^3 -^-» H-
*«| LDO'-OOCOiDO O r CMO*
* CO ZO O» O -~-
4- I-H CO t i t C 1- 1
O UT) X ***. "***. O >f- X O
VI O Cjs LO (/) i/l r-H ro i ^
-^, >_^ . ^ flj OJ^ .CO
3 If) X c ( WIOJO^X
4-> i^-'+3o o a; r
S2 L J fO E = r O LO CTi
- . CNJ 5* 1 1 C E - CNJ
Cl «.2E^.Q J3S( O
U") C3 C^ LO CO CO '^*i *"» »^ ("~ 1 :>>:>>>> Oi x x i- ^ T- 1/1 - .
- - o cr> ^ (CM 4-
y^/syVV ^.CO * O^ lO CT> r C^ d -
f* ^-H IT) » «- CX! U
urs O O urs O O 113 - ' - - ' 3 CM CO C "O u- t_j
^SS^SS 3 ' + "*" c-sicvj o-cr^ cr>
+
c
^
4->
S c "c o
- * £
T3 = = 0 U
i* *- v» i u a. s_
3 * *- O U
!D O fO O "O H
C22 «U ^ ;/> O^ <13
X i. O>*J r U
P-» 3 T- u_ CJ
'->> *-> Ct3O3
TJ j-J 3 C T) ai >>O 3 J3
-r- r- 4-J 4.J r
3 *O (U 3 (yi GJ *^ "" J2 3 fl il
^ ^ ^ ^j fl rO 3O »" O ^
C f '--a f 3 -DU-3 4J ^ I
Cl _^. ^^^ (j '
^ . O * ~"* - -
-* 4- CSJ - *T
E-15
-------
3
4J
T~
a
V)
1_
^
^
S
cr
5
4«J
C
o
V)
3
-C
S
u
U_
o ^
o .c
O re
to
I-H * »
C TJ
re re
4-) LL.
re o
CM
c n
re
u i>
s- re
re i.
01 3
s- -o
Ol 01
s- c
O 1.
a.
<+_
(_j ^
^s^ .,_
3 VJ
cn -o
0 i-
kO O
^J
4-> (XI
3 S-
o aj
TO »-
O <-
3= E
J=
VI -U
a»
« o
03 <*-
VI
Cn C
3 O
4) 4_>
TO -
CJ T3
~"~ C
1_ (j
CJ
c £
( re
-C C
4-9 as
. 4J
3 VI
VI
TO f-
aj ai
>- 3
i/l
.
VI
4_J
C
o
^
OJ
3
^
TO
^
X
3
TO
_Q
"S
r
C_
a.
vi
O ^_J
O SJ
C
< s-
1 OJ
ui M-
OJ
ai Q£
.0 u-
TO O
I
CJ
J_ !__)
<4- .
-^^,
01
VI U
re 3
Ol Ol
01 LL.
4-J
1/1 C
re «-
01
O !~
3
4-J (j
O
O '*-
<4- 0
5
.a i-
3 4-J
u re
-a -^
i- X
re o
-a s-
c a.
re a.
VI
Ol
i- >
Ol -r-
O.-W
re
c: >
-J Ol
4- C
O O
0
o> re
4J *~
= O
Ol
a -c
- ai
CL re
«
VI
re
Cl
c
01
o
01
'o
u
]
o
I
c
re
4-1
01
JT
01
t
c
0)
u
c:
V)
3
22
0
LO
^^
~
1
*J
a
re
>_
p_
TO
VI
t.
TO
L.
O
4
j_)
cj
33 **-
^ V
^ VI
3 *-s.
CT 3
CJ 4->
U CQ
CJ O">
3 LO
^j
« 3
TO O
C -3
O TO
]
TO
O) 3
0. 4->
vi C
U <4-
3 0
O OJ
!*. i
C. 3
C TO
o i.
*- D
3 'U
J2 4->
1 4->
O Ol
VI C
Ol
o
01 Ifl
re re
2 01
<4_ Ol
O !-
a.
JD O
,«H
CM~
C
«t l/l
CM C
Z 0
°CM &
Z 3
VI
CM re
O
o 01
c
**" ^ 3
<: o
1 r
LU .
O
Ol t»-
S c
1
c
-O 01
VI
jO
t _t
c
S o
0
M- re
Ol
VI O^
Ol
01 to
4^ 1
VI LU
~s 01
*4 ^
1
^_ f
3 O
re H-
01 VI
CL re
= Ol
4J Ol
4J _/l
Ol re
r; 3
C
3
O
^
"re
C
t/1
4-)
VI
3
jQ
5
u
o
4J
L.
(Q
c
o
3
T3
C
o
TO
VI
TO
4J
(/I
i
c
c
01
Ol
X
0
01
i.
.^
3
cr
Ol
c
3
01
re
3
cr
O)
re
*-**
t M
u_
0
*o
01
o
4-J
o
22
5
V)
re
.^
re
^
o
4_)
3
a
c
re
U-
o
o
to
^
01
1
4-J
01
u_
o
^_
3
C9
C
C
~-
re
01
u
14-
,,
o
Ol
O.
1
CO
.«
i.
TO
VI
V)
0)
u
X
4)
4_)
C
U
01
CL
CO
re
^j
^n.
re
^
re
o
VI
13
Ol
re
4J
re
^
re
c
O
0
> VI
* 3
o 5
4_>
V> O
O «-
c.
5 oi
o s-
o
VI U-
re oi
Ol i-
Ol
L. .C
re j
3
re <*-
c u
T 3
^_* 4-1
CO
IT c
o
re o
» i
O II
<4-
-Ct
to ~~.
CM 3
O CO
o o
c to
re ^o
«
" O
CM CM
O
II
^
O :»
i
^f >
°°. §
CM
Z It
O ^
CTi CM
'O Z
CM
4-J
0 C
a>
- u
O s-
CM CJ
= Q.
1- CM
O
1^. -o
LO CO
O cn
«
CM ^*
O CM
CJ
! ijQ
U» CM
O
CM ^r
O "
^~
VI
V)
TO
0)
O
TO
^
O
Vl
TO
r
_
CM
X
CM
C*i
O
< i
CM
f^
t" i
ro
X
CM
O
*
O
HH*
-4-
CS1
X
o
o
+
CM
O}
o
1
CM
r^
₯
CM
O
co"
^
cc
c
CM
+
CM
CM
O
a
^
n
CM
o
o
-t-
-~
CM
0
0
-i-
CM
CM
O
o
CM
0
CO
o
CM
CO
c
-^
o
CM
*"
^«v
X
^4
o
0
+
VI
VI
TO
-4_>
CJ
3
TO
c
0
c
4_)
3
a.
TO
^)
>^
"4-
V)
TO
cn
TO
3
TO
C
^
CM
0
-c
«
0
CM
^
'O
O
0
.4.
CM
*7"
j3
r-H
CM
-i.
CM
O
o
a>
TO
TO
~j
t.
CJ
GJ
*^O
. «
TO
,_
TO
3
TO
C
M-
0
_
H-
O
0
4*J
3
§
U
O
3
3
£-
re
c
0
r
T3
-a
TO
TO
JJ
TO
3
CM
O
J3
?
01 5
s- =>
-t- 3
'yi I
VI
Ol Ol "C
i- C
C CL re
4-J
re o
* - 4-J - ^
VI O
" C 1
S 0 LL.
3* 0
i ^ '^
X '/I 4-J
3 "-> 3
re re o
** Ol -2
-c re
X U >,
Ol1* I*
c -r- re
- -OS.
r- Ol 01
CL O. C
O VI 01
01
3 re
0) Ol
f ^ II K-
^ "* CL 3J
n ^j ;_j
O ^: vi
o 01
CO U-
^H 01 0
~-- 3 O
O
+ II CC
o
Ol 4-i
c/l
3 re u-
1 01 0
1 0
O 01 '-O
O 4->
00 VI 5.
re o
^- 3 i+-
VI ^* VI
*-» O TO
(J
3 CJ U_
T3 !- 0
C 3 O
S_ 4-* O
O. TO CO
i. -H
TO a. o
C7l £ J->
a)
a> j *
4-> cn
VI d-J
TO a) 3
3^1-
o
CL. « CD
u cn c
3*" 3 J=
^* h U
' ^, i a) aj
a) 3
^ i ~ A.J
r i 3 TO
O * O>
O ,*> *. Q.
r~* r i
-^ a)
X O *J
3 *""*'>
fo a. s-
0 O
-t- -n aj
vi
i
,__
>
4-J
OJ
a.
OJ
M
TO
^_
^
3
4_>
TO
C
C
TO
VI
TO
Cn
ai
VI
TO
J_J
0
VI
CJ
3
TO
>
C
4J
TO
a>
^
a;
3
o
ii
cn
Cn
4J
23
o
^
o
CM
1*
VI
TO
cn
U
VI
TO
3
n
§
i-H
T3
CJ
ai
VI
TO
,__
TO
3
TO
C
3
CS
i l
CT
3
-J
x
3
TO
ii
X
VI
TO
cn
,__
TO
TO
^
a>
"o
c=
1
,
- k
CO
t
O
-f
CM
O
O
^
OJ
cn
c
a>
o
T
ft
,__
cn
c
.2
^r
r*.
«_t
x
.5
r^
O
s_
X
i-
^x_
^_
cn
3
cr
x
rj
^
?~>
,__
O
c:
i <
^^
cn
3
^2
i
0
a
- -
x
* ^
Ol
3
S
'VI
CJ
'g
T
* -.
TO
P_
o-
^.
-3
o a>
IU i
2
. T
! "[^lO
C .
u.
-*r 3
r^j
O -J
^ *,
ur;
^
a>
u
c
a>
t.
CJ
OJ
cr
^
o
4-1
U
TO
4-
-S
a;
u_
TO
C
cn
V)
ai
-o
LO
0
)_^
X
^y
O
CM
cn
-^
c
0
O
f
X
*~*
(_)
CJ
o
I-C
(
-
X
>*v
u
a)
VI
-O
r-.
*
x
^-^
=
c:
cn
VI
11
CJ
S-
TO
^_
i
^
TO
4-J
VI
3
£
A
^
CJ
01
a;
,2
TO
^
T3
TO
>^
!
TO
!_J
i.
S
'-J
C
d
^-^
m
^^
_o
n
«
33 CM
re O 31
01
s_ c c: 4-
3 '"'- S
= -~ 3 Iw o
E-16
-------
3 3
assumed a minimum commercially available size of 1.01 m (35.7 ft )
based on vendor information and a maximum shop-assembled unit size
of 205 n3 (7,238 ft3).17
The design procedure would 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. 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
Themal incinerator purchase costs for the calculated combustion
chamber volume were taken directly from Figure E-l, (Figure A-l in the
IT Enviroscience document, Reference 14). An installation cost factor
1 Q
of 4.0 (see Table 8-2) was used based on the Enviroscience document.
The installed cost of one 150-ft. duct to the incinerator and its
associated fan and stack were also taken directly from Figure E-2
19
(Figure IV-15, curve 3 in the IT Enviroscience study. A minimum
cost of $70,000 (in December 1979) was assumed for waste gas streams
with flows below 500 scfm. The costs of piping or ducting fron the
process sources to the 150-ft. duct costed 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 6-inch FLO 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 AMD 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
-17
-------
o
o
o
o'
o
o
o
o
o
rt
»-
u.
tu
i
5>
u>
o
Ul
u.
u.
Ul
ec
UJ
i
2
g
?
00
i
o
o
i.
O)
J3
(O
c
o
I
-M
3
C_5
O
S-
o
0)
i-
o
to
O
o
OJ
I/)
fO
"o
1-
0.
-------
O
CO
<
O 4->
r- (13
+J -
C 3
10
in
E
ro GJ
CD 4->
O
tO <4-
o
o
1o +J
Q.TJ
05 C
O ro
ra U.
4->
to i-
-* <
CM
I
-------
TABLE E-7. CAPITAL AND ANNUAL OPERATING COST ESTIMATES FOR
THERMAL INCINERATORS WITHOUT HEAT RECOVERY
Item
Value
Capital Costs
Combustion Chamber
Purchase cost
Installed cost
Installed cost, June 1980a
Piping & Ducting (from sources
to main incinerator duct)
Installed cost
Installed cost, June 1930b
Ducts, Fans & Stacks (from
main duct to incinerator
and fron incinerator to
atmosphere)
Instal led cost0
Installed cost, June 1980U
Total Installed Cost, June 1980
Annualized Costs6
Operating labor
Maintenance material & labor
Utilities
f
natural gas
electricity^ u
Capital recovery
Taxes, administration & insurance
Total Annualized Cost
from Figure E-l for V
purchase cost x 4.0
installed cost x 1.047
see Section E.7 for Q,
w.g,
(scfro)
installed cost x 1.206 for piping
installed cost x 1.288 for ducting
from Figure E-2 for Q ;
use $70,000 minimumw>9-
installed cost x 1.064
sun of combustion chamber,
piping & ducting, and ducts,
fans, & stacks
1200 hr/yr x $18/hr = $21,500
0.05 x total installed cost
(5.639 x 10" ) (% aux) x LHVw>r1_
(0.4955) x Qf (scfm)
0.1627 x totaT9installed cost
0.04 x total installed cost
operating labor + maintenance
+ utilities + capital recovery
+ taxes, administration &
insurance
E-20
-------
FOOTNOTES TO TABLE E-7
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).
GFrom Figures E-2 for no heat recovery from Enviroscience (Reference 19),
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 GARD Manual (Reference 20). 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 21). 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 LHVw/LHVn>g>] (Ib^/lOO Ib^) x Qw>g< (Ib/hr) x
(100 lbw )/100(lbw ) x (8600 hr/yr) x (lb-inola/17.4 lbn ) x
(379 scf'at 60°F/lb-moie) x (1040 8tu(HHV)/scf at 50°F) x $5.98/10°
Btu (HHV) x (106 Btu)/106 (Btu).
9Electricty = (6 in. H?0 pressure drop) x Qf ( (scfm) x (8600 hrs/yr)
x (0.7457 kW/h)p x (5.204 Ib/ft2/in. H20) +[(60 sec/min) x (550 ft-lb/
sec/hp) x (0.6 kW blower/1 kvJ electric).
10 percent interest (before taxes) and 10 yr. life.
E-21
-------
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
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 ft of
catalyst for 1,000 scfm is required for 90 percent control efficiency
and 2.25 ft is required for 98 percent control efficiency. As discussed
earlier many factors influence the catalyst life. Typically the
catalyst nay 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
E-22
-------
is about 30 percent. Gas temperature, preheater temperature, gas dew
point temperature and gas VOC content determine the maximum possible
heat exchanger efficiency. A 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 (Mo. 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 Catlaytic 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 costs.
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
including bypass. Calculations for capital cost estimates are based
22 23 24
on equipment purchase costs obtained from vendors ' ' and application
of direct and indirect cost factors. Table E-10 presents third quarter
1932 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. The heat exchangers for small
size systems would be costly and may not be practical. Table 3-2
presents the direct and indirect installation cost component factors
used for estimating capital costs of catalytic incinerator systems.
The geometric nean of the two vendor estimates for each flow rate was
multiplied by the ratio of total installed costs to equipment purchase
E-23
-------
Table E-8. OPERATING PARAMETERS AND FUEL REQUIREMENTS
OF CATALYTIC INCINERATOR SYSTEMS
Item
Source of information or calculation
Waste Gas Parameters
IT)Flow rata~((L)s 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 (Q-), scfm
O
(5) Total amount of gas to be
treated (Q4)> 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
From Table E-9
0 if the waste gas contains VOC and
nitrogen or other inert gas; and
[(1 - volume percent VOC) * (volume
percent VOC)] 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/U LEL) x (%LEL fron Table E-9)
Item (7) + Item (3)
(10) Minimum possible temperature See footnote C.
of flue gas at PHR outlet, °F
(11) PHR efficiency at maximum
possible heat recovery , %
(12) PHR design efficiency , I
[Item (1) x (Item (7) - 25°F -
I ten (6))] 4 [Hem (5) x (Item (9)
Item fS))]e
65
E- 24
-------
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
0.65 [Item (9) - Item (6)] + Item (6)
(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
Item (5) x [Item (7) - Item (13)] x
[Gas specific heat9, Btu/scf, °F] x
[Item (14) x 60 minutes /hour] x (10%)
[Item (14) x (3,600 x 60 minutes/year)]
-3
required per year, 10 cfm x 10 -f (1,040 Btu/cfm)
aOn volume basis (scfm/scfm): 9.53 for methane, 16.68 for ethane,
23.82 for propane, 45.26 for hexane, and 14.3 for ethylene.
Primary heat recovery unit.
cHeat exchanger should be designed for at least 50°F above the gas dew
point.
The 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.
Cost estimates are based on 65 percent heat recovery.
specific.,heat varies with composition and temperature. Used
0.019 Btu/ft °F based on average specific heat of air for calculation
purpose.
Auxiliary fuel requirement is assumed to be 10 percent of
total .
E-25
-------
TABLE E-9. GAS PARAMETERS USED ^OR ESTIMATING CAPITAL AND
OPERATING COSTS OF CATAL/TIC INCINERATORS*
ITEfl
VALJE
Stream identification
Strean conditigns
Temperature, F
Pressure, psig
VOC content:
Emission factor, kg/Mg
of product
Weight ** of total gas (W.)
Mass flow rate, kg/h
Ib/h
Organic constituents, wt *
Average mol. wt. (M..), Ibs
Volune flow (0^), scfm
Heat content (H
Btu/scf
l'1
Identify the vent and the polymer
industry fron the BID
(Emission factor, E, kg/Hg) x
(Plant production rate, P, Gg/yr) -s
(3,600 h/yr)
(kg/h) x (2.205 Ib/kg)
(VOC mass rate, Ib/h) * (60 nin./h) 4
(Molecular weight (M ), Ibs/lb mole) =
1.645 (EP/f^) L
(174.273)(2.521N + Nu)c
C n
Total gas:
Constituents
Mass flow rate, Ib/h
Molecular weight (MJ
Volume flow (Qj, sCfm
VOC, air and others
(VOC rate, Ib/h) i (wtZ of VOC in
gas, i^)
Gas mass rate, Ib/h) T (50 nin/h) *
(Gas molecular weight (M,), lb/lb mole)
x (385 ft3/lb mole) = 1.645 (EP/H^ )
Air volume 'low rate, scfm (Total gas flow, scfm) x (volume
percent air in total gas)
VOC concentration (A), ", (100)[(Volune flow of VOC, scfnh*
of LEL (Volume flow of air, scfm] * LEL
Heat content (H ),
Btu/total scf^
aObtain gas parameters from Chapter 3 of the 310, except those to be
calculated.
Calculate using weight percent values of VOC components.
clf the VOC heating value is not available calculate it using on heat
of combustion values of 14,093 Btu/lb from carbon converted to CO
and 51,623 Btu/lb fron hydrogen converted to rfater. N and N,, denote
number carbon and hydrogen atoms in VOC c
Lower explosion levels of ethylene, hexane, riethanol, prooane and
butane are 3.1, 1.32, 7.3 and 2.5, and 1.9 respectively.
eTotal gas heat content averages 50 Btu/scf at 100 percent LEL.
E-26
-------
o o
o o
>.-§
o
o
o
o
o
o
at
o
o
o
c
OJ
o
o
o
o
o
o
c
o
o"
o
o
o
co
0)
c
l§
o g
o o
= ]
s~
g
5
o
o
' ~ c; c
E-27
-------
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-3.
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.
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, typical catalyst costs are about
$3,000 per ft . 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
E-28
-------
5,000
o
o
o
r-H
<=<=>
CO
O
1,000
T3
O>
to
100
10
Key:
With
Without h
eat r
at re
2COV
^
-<>
X
0.5 1 10
Gas Flow Rate (1000 scfm)
100
Figure E-3. Installed Capital Costs for Catalytic Incinerators
With and Without Heat Recovery
E-29
-------
Table E-ll. CALCULATION PROCEDURE FOR ESTIMATION OF ANNUALIZED
COSTS FOR CATALYTIC INCINERATOR SYSTEMS
Cost component
Direct costs
Operating labor
Maintenance material and
labor
$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-3)
Catalyst requirement
Utilities:
Fuel (natural gas)
Electricity
Indirect costs
Capital recovery
Taxes, insurance and
administrative charges
fotal annualized costs
$2.7 x (Total gas volume flow(QA)a,scfm,
item 5 from Table E-8) = ($2.7 j? Q4)
($6.22/103ft3) x3(Araount of natural
gas required, 10 ft , Item 16 of
Table E-8)a
($0.335/scfn) x (Total gas volume
flow rate (0.), scfm, Item 5 from
Table E-8) f3r units with no heat
recovery; and ($0.838/scfm) x (Total
gas volume flow rate (Q.), scfm, Item 5
from Table E-8) for units with heat recover
(0.1627) x (Total installed capital cost,
$ from Figure E-3)
(0.04) x (Total installed capital cost,
$ from Figure E-3)
Sum of total direct costs and total
indirect costs
Total gas flow including waste gas and additional combustion air.
E-30
-------
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, 65 percent motor efficiency, and $0.049/kVJh electricity unit
cost, the total annual electricity costs amount to $0.335/scfm for
units with no heat recovery (i.e., for 4 in. I-LO pressure drop) and
$0.834/scfm for units with heat recovery (i.e., for 10 in. HoO 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
gaseous streams from the continuous process polystyrene model plant.
Two types of condensers are in use in the industry: surface condensers
in which the coolant does not contact the gas or condensate; and
contact condensers in which coolant, gas, and condensate are intimately
mi xed.
Surface condensers were evaluated for the following two streams
from the polystyrene model plant: the styrene condenser vent and the
styrene recovery unit condenser vent. These streams consist of styrene
and steam which are immiscible. The nature of components present in
the gas stream determines the -nethod of condensation: isothermal or
non-isothermal. The condensation method for streams containing either
a pure component or a mixture of two imm'scible 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 following procedure and
assumptions were used in evaluating the isothermal condensation systems
for the two streams containing the immiscible styrene and steam from
the continuous polystyrene model plant.
E-31
-------
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 and
steam. As the vapor pressure data are not readily available, the
condensation temperature is estimated by trial-and-error 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 is known, the total heat load of the
condenser is determined from the latent heat contents of styrene and
steam. 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 coefficients 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. For this
study no detailed calculations were made to determine the individual
and overall heat transfer coefficients. Since the streams under
consideration contain low amounts of styrene, the overall heat transfer
coefficient is estimated based on published data for steam. Then the
total heat transfer area is calculated from the known values of total
heat loads and overall heat transfer coefficient using Fourier's
general equation. A tabular procedure for calculating heat exchanger
size is presented in Table E-12.
E.5.2 Surface Condenser Cost Estimation Procedure
Since the gas volumes of the two streams are very low, the
?
calculated heat transfer areas are also very low (about 10 ft"). The
heat exchanger costs for each stream were obtained for the minimum
available size of 20 ft2 from vendors.25'26'27'28 An installation
factor of 1.39 (See Table 8-2) was used to estimate installed condenser
costs. No additional piping was costed since the condenser unit is so
small (~l-2 ft. diameter) that it should be able to be installed
E-32
-------
Table £-12. PROCEDURE TO CALCULATE HEAT TRANSFER AREA OF AN
ISOTHERMAL CONDENSER SYSTEM
Item
Va 1 u e
Heat exchanger type
Cocurrent shell and tube heat
exchanger with the hot fluid
in the shell side and the cold
fluid in the tube side
las stream condi tion
(including tenperature
(T )°F pressure (P.),
psTg, and compositton)
Condensation temperature
(T2),°F
Total heat load (H), Btu/h
Coolant used
Temperature rise of
coolant, (AT),°F
Coolant outlet temperature
(T3),°F
Log nean tenperature
difference (Lf1TD),°F
Heat transfer coefficient (U)
o
Heat transfer area (A), ft
Obtain fron Chanters 3 and 6
water at 85°F, 25 gpn
H(Btu/h)+ [(25 gpm x 500 Ib/h/gpm) x
(1 Btu/lb°F)]
85°F +^T
C(TrT3) - (T2 - 35)] 4 In
240 Btu/h ft2°F
(H)/U(LMTD)
Determine from vapor pressures of styrene ana steam and the total
gas pressure. Calculate the styrene vapor pressure using Clausius
Clapeyron equation:
In P/Po . (\/R) (1/T0 -
and T are
where P and P are stream pressures, mm Hg; and
corresponding°temperatures, °K; V is latent neat, cal/g-mole;
and R is universal gas constant = 1.99 cal/g rnoie°K.
The same equation can be rearranged to eliminate \ and R:
ln(P/
(1/TQ -
ln(P
l/Pc
Using two known values of pressure and temperature, calculate
the pressure for an assumed temperature. Proceed by trial-and-error
until the temperature .vhicn gives a total value of styrene oressure
and steam pressure equals to the toal gas pressure.
fa-
Total of latent heat of styrene and steam in stream per unit time
(Btu/hr): Calculate the latent heat of styrene from Clausius-Clapeyron
equation using pressure and temperature values of gas and condensation
condition, multiply by Ib/hr styrene in strean and add to product of x
(970.3 Btu/lb steam) and Ib/hr stean in stream.
''Fixed amount of 25 gpm is used in order to maintain turbulent flow.
Obtained usiqg a clean overall heat transfer coefficient (U ) of
366 Btu/h ft'"0F and a dirt factor (D.F.) of 0.003. The clein overall
neat transfer coefficient is obtained using weignted averages (86"
steam and 16» styrene) of pure fluid heat tsansfer coefficients,
1,000 3tu/h ft^'F for steam and 35 3tu/h Jt£°F for styrene and tne
following relationship:
E-33
-------
adjacent to the source. Table E-13 presents the estimated total
o
capital and annual operating costs for the condenser system of 20 ft
heat transfer area.
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.6.1 Ethylene Glycol Recovery System Design
The equipment selected to comprise the two ethylene glycol recovery
systems, as well as the design and operating parameters, were based on
information provided by industry sources. The baseline system recovers
ethylene glycol (EG) from downstream of the cooling water tower. In
this system, ethylene glycol emissions from the polymerization reactor
and from the distillation column recovering EG from the esterifier
emissions accumulate in the cooling water and are emitted from the
cooling tower. The regulatory alternative system recovers ethylene
glycol emitted from the polymerization reactor through use of an EG
spray condenser and from the esterifier through use of a reflux condenser.
The industry information (much of which was considered confidential)
was used in conjunction with standard engineering references such as
pq
the Chemical Engineers' Handbook , and McCabe and Smith's Unit
Operations of Chemical Engineering , and engineering judgement.
These design procedures are summarized in the footnotes to Tables E-14
and E-15 for the baseline and regulatory alternative systems, respectively.
E.6.2 Ethylene Glycol Recovery System Cost Estimation Procedure
The cost estimates and their bases are presented in Table E-14
for the baseline ethylene glycol recovery system and Table E-15 for
the regulatory alternative system. The costs of the baseline system
were estimated based on the design estimate developed and standard
engineering procedures. The costs of the regulatory alternative
ethylene glycol spray condenser and recovery system for new plants
using the DflT and TP*\ process were derived fron confidential cost data
provided by an industry source for a similar system on a larger capacity
E-34
-------
Table E-13. CAPITAL AND ANNUAL OPERATING COST ESTIMATES
FOR A 20 ft2 CONDENSER SYSTEM FOR THE
STREAMS FROM THE CONTINUOUS POLYSTYRENE MODEL PLANT
item
Value
Control system
Capital Cost:
Purchase cost
Installed capital cost3
Annualized cost:
Operating labor
Maintenance0
Utilities:
Watera
Electricity8
Taxes, insurance.and
administration
Capital recovery^
Total annualized cost
without recovery credit
Total amount of styrene
recovered from X Ib/hr
of styrene
Annual styrene recovery credit
at 30.3575/lb
Heat exchanger with a rtaxinum
capacity of 20 ft heat transfer
area
32,000
2,780
SI, 080
140
3 390
3 150
$ 110
S 450
$2,300
(X Ib/hr x 3,600 hr/yr x 99S heat exchanger
efficiency x 90% recovery efficiency
from the separator)
* 2,000 Ib/ton = Y tons/year
Y tons x 2,000 Ib/ton x S0.3575/lb = SZ
Total annualized cost after credit (32,300 - SZ)
Cost effectiveness of emission
reduction (S/Mg)
(32,300 - SZ)/[X Ib/hr x 3,600 hr/yr
x 99% heat exchanger (VOC reduction)
efficiency)/2,205 Ib/Mg]
Purchase cost tines installation cost factor of 1.39 (see Table 3-2).
Operating labor cost = 1 hr/wk x 52 wk/yr x 1.15 (with supervision/
without supervision) x S18/hr (including overtime).
c'1aintenance cost = 0.05 :< (installed capital cost).
Water cost = 25 gpm x 60 min/hr x 8,600 nr/yr x 0.1 make-up/total
x S0.30/(l,000 gal) x (1,000 gal)/l,000 gal.
"Electricity consunption (eauations from Reference 29) and cost:
hydraulic horsepower = 50 ft x (1.0 specific gravity) x 25 gpm/3960 = n.3157 ho
bra
-------
Table E-14. EG RECOVERY COSTS FOR BASELINE SYSTEM
__ Value
Capital Costs
1 Distillation Column (CL-1) 24,000a
1 Distillation Column (CL-2) 24,000b
1 Pump 350C
1 Condenser 39,180d
Descalation factor to bring to 1980 dollars 0.82e
Installed Capital Cost Factor 3.26
Total Installed Capital Cost 233.985
Annualized Costs
Operating Labor 77,4009
Operating Materials 0
Maintenance Materials and Labor 10,350
Electricity 9401
Steam 828,120°'
Water 0
I/
Taxes, Insurance and Administration 9,360
Capital Recovery 38,1401
Recovery Credit (407,100)m
Total (Annual Costs - Recovery Credit) 557,210
-------
FOOTNOTES FOR TABLE E-14.
aDistillation column (CL-1) size of 4' diameter and 36' high, with
11 trays (Reference 32). Rough cost estimate of $24,000 obtained from
Perry's (Reference 29), p. 18-55 in 1968 dollars adjusted to June 1980
dollars.
Assumed to be same as CL-1.
cDeterrnined to be 3 hp with cost of $350.
See Reference 31, page 12.
eFrom Chemical Engineering indices.
See Table 8-2, developed based on confidential industry information for
alternative system.
^Confidential industry information on operating labor scaled down to
assume 4 man-hours/shift, 3 shifts per day (= 4,300 hours per year) at
$18 per hour.
Confidential industry information on maintenance labor and material
costs were scaled down to assume 520 hours per year for maintenance
labor and $1,000 per year for material costs.
119,239 kWh per year for pump at $0.049/kWh
''Assumes steam required in distillation columns is equivalent to steam
required in EG recovery steam in the regulatory alternative system (see
footnote k for Table E-15).
Based on 0.04 x Installed Capital Cost.
Based on capital recovery factor of 0.163 for 10% interest and 10 year
life.
mEthylene glycol of 80 percent purity is recovered at a rate of g
11.8 Ibs/thousand Ibs product (Reference 32). Thus, for a 230 x 10
capacity plant, total annual EG recovery would be 2,714,000 Ibs. Raw
material EG, which has a high purity, is worth about 2.1 i per Ib. The
EG recovered from the baseline system is about 30 percent pure and was
assumed to have a value of 15
-------
Table E-15. EG RECOVERY COSTS FOR REGULATORY ALTERNATIVE SYSTEM
Item Value
Capital Costs
7 Spray Condensers 148,400a
7 Reflux Condensers 148,000
7 Pumps 2,450C
7 Heat Exchangers 37,800d
1 EG Recovery System 386,300e
Descalation Factor 0.82
Installed Capital Cost Factor 4.24g
Total Installed Capital Cost 2.514,943
Annualized Costs
Operating Labor 154,800h
Operating Materials 0
Maintenance Materials and Labor 21,7001
Electricity 70,550
Steam 828,120k
Water 4.6501
Taxes, Insurance and Administration 100,600n
Capital Recovery 409,940n
Recovery Credit (1,292,000)°
Total (Annual Costs - Recovery Credit) 298,360
E-3G
-------
FOOTNOTES FOR TABLE E-15.
aBasad on size estimate from Tennessee Eastman and cost estimate from
Missouri Boiler. One spray condenser per line in model plant.
Going to this system replaces feed lines from estifiers to distillation
column (CL-2) with raflux condensers, one par line. The size and cost
were assumed to be the same as for EG spray condensers on the reactors.
'See Reference 31, pp. 10 and 11, for pump sizing. A 3 hp pump is
required per line at a cost of $350 per puinp.
' See Reference 31, pp. 7-9, for heat 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 sane.
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.
From Chemical Engineering indices.
3See 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.
L_
Based on 8,600 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.
""Based on maintenance labor requirements and maintenance materials cost
given in confidential industry information for a larger system.
^This cost comes from 3 sources: pumps, recovery system, and heat exchanger
to chill water. Pumps at 3 hp use 19,239 kWh per year. Recovery
system was estimated to use 412,800 kWh per year based on confidential
industry information for a larger system. Cost of electricity is
$0.049AM. 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 watergin each heat exchanger is calculated to
be 177,000 Btu/hr or 1,522 10 BTU per year. Using a conversion efficiency
of 0.72 and converting to kWh, this is equal to 4.461 x 10° kW'n per
year. Assuming 2 of the 7 lines in a model plant produce high tenacity
(high viscosity) industrial resins, total electricity usagg in the
model plant is 19,239 x 7 plus 412,800 plus 2 x 4.461 x 10° equal to
1,439,673 kWh per year.
I/
Based on confidential industry information and scaling steam usage in
the EG recovery system and its vacuum system proportionatelygaccording
to plant capacity, steam usage was estimated to be 1.34 x 10 Ibs/year.
A cost of $6.18/1,000 Ib of steam was used.
'Based on water consumption of 15.5 x 10 gallons per year (from confidential
industry information scaled down by proportioning relative capacities).
m3ased on 0.04 x Installed Capital Cost.
nBased on a capital recovery factor of 0.163.
E-39
-------
°Based on a total EG recovery of 20.8 Ibs of EG/1,000 IDS of product and a
recovery credit of $0.27/lbs of EG (from Chemical Marketing Reporter).
The 20.8 Ibs of EG/1,000 Ibs of product is based on 99 percent recovery
of the 18.2 Ibs of EG/1,000 Ibs of product estimated entering the EG
spray condenser (equal to 18 lbs/1000 Ibs) plus 2.3 Ibs of EG/1,000 Ibs
of product recycled from the estifiers that would have to be replaced
with fresh feed if the baseline system was used.
E-40
-------
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-16. For
flows less than 700 scfm, an economic pipe diameter was calculated
based on an equation in the Chemical Engineer's Handbook and simplified
34 3"B*~~36
as suggested by Chontos. ' ' 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 scfn 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
71
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.37 (see Tables E-17 and E-18)
Ducting costs were calculated based on the installed cost equations
TO
given in the SARD Manual. (see Table E-19)
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-41
-------
[
J
LLJ
i^*""
1 i-J
!_J
ce
O-
^T1
1/7
UJ
f_J
E
O
3
o
0
-1-
.
1
UJ
CD
J3
j2
f
I
j
0)
3
fO
i
1
1
i
i
1 c
' OJ
1 4J
[
=
t>
c:
<4l 0
U 0 O
c/i i*^ ^r
o v o>
V /I -g
o-o o>
S- 0
O t- (/> C
cro-
X X 0)
en
CM cn s-
X
1! 11 U
ecu
f i_ C= , t£ , CTi OJ
r-i I- 4- u 3 vi s- »-
j34_ >2^-^O ^ * -O IT
c o- c u = j: =
aj jr: -r- 3
CL.+J ^--^'* - r V* VJ
.,_ .,«. C CJ i/i O vi
Ct IS Q "3 C^ (/I ^C +^ ^C
n -1
CT>
d .c '/?
^ ti, ^
(_) d iT3
3 3J r
a>
^2 f- T3
' C_
^-»,
O
^j
C
o
p_
fl
^
c
-^
s
c
cu
VI
*->
i
jj
'§
OJ
'r-
M~
O
5
C
3J
C
(
OJ
CL
a.
-o
^j
y
^
^
0)
J=
Ol
o1
1
D
vt
O)
c u
r_
O-J=
CL ai
u- ^~
Q
VI ^3
5_
^ .-a
a? -a
t - j
3 VI
U
r <4
O
4- C
.1- O)
M- CL
O C3
O
* c
L. 5
2
0 ^
O^
1 . 1
u
+J
0)
JD
_^
^
C7>
C
i*
"S
J_)
r
3
U
U 3J
3
OJ *«
ty -a
r i-
c
C vi
OJ H
O o
3 **
O V) VI
t/» OJ
^ ~ y^
f~ c; ~ n?
(J ft) VI <
OJ *
-3 (J «0
QJ a; c ^Q *
U VI OJ CO
3 U VI
O vi O U
VI O U U
0) i- C
C -*-» OJ
4J .f- > (j ^.
U- » C 3 O)
1 OJ UJ -O <4-
O Cu OJ
c. o *-' * *
^ 1J <4_ ) _
3 *J 4J ID 3
VI V) C
VI O O T3
-------
re
GO
zz
LJ
r^~
O
Q.
y
O
o
O3
t?»
b <
-^
i i
o_
r-~T
I
LU
OJ
>""^
re
f
O
C
CM
0>
c
,_
OJ
r
.
1
j_j
C
cu
cr
Q.
0
0
in
O)
c
Q.
T
o_
s-
o
1/1
(/J
a
Q_
1| §
t . 1 CJ
I
°
S-
O)
JD
TIT
J
O!
O
S-
3
O
j ;
CO CXJ OO O CM C
CM t 1 O
CM
^^^^y^roo^LO^-iooo
CM CM LD «a~ O
tn
f-^ CM I »~ ^ *.Q CM CD *"""* 1-1 CM *^ CM CO
^M I < 1 CM
f i *3- t t t iCO^Du^< *OJi-n OCCO
T i r I < I r-~
S"
4 J - -*
i/i cy
ZDUO O
1/1 -f -i->
c/i CD -^ CU CJ
CU > J~> ^. i
t/1 > 1 -I l/l 3
ri) '^0 r * ro ' ' ro (/I ^3
> CU -a > r_n -Q CO
!>>(/! C ^- -E
' C ro r i- CD o -o cu u
OJO1 > ro , O) ul O! <- C 00 3 GO
1 = C. > O c 1/1 oj i './i ro 1-
'""* >i ' s/ ^ .f 5 CD ^
u
!l)
o;
^
^
u_
E-43
-------
a;
(C
o
o
CL.
H-4
Q.
CO
1
LU
*f LT>
o
CM
I* I
OJ
iy
I o JcMCMCMCMCM'<
! ""*
CO
a;
<5
CL
r
e_
i CM
< i co
o
o
j-j
c
CD
O
O.
e=
o
o
c
o
-o
a>
e/i
ns
O
C
O)
a>
if-
ai
I en
! 2
a.
5-
O!
3
CL
O
C_3
oo
Cu
tO
c
o
(B
tn
-------
re
i/}
t
i
£3
£
^"H
cx
UJ
CC
t_;
UJ
Q
K
gr
o
£;
)
cr
UJ
1
00
o
CD
z.
\
JJ
£
UJ
1
<£
1/1 !
>_*
. 1
cr.
*7
1
, i
"re
I
i
t
'
I
!
CD
re
(/)
=
CL.
I
1
j
C
"
Q
i-
CD
"i
T-
o
o
u
tu
re
re
ja
>.
H-
is
c_
i
t
'
t
'
i^
i re
1
0.
f "
c_
, k
c
^__^
LO
u
OJ
O)
re
f"**
CD
C
^
a.
ex.
.
4J
o
0
o
CSI
CD
i
OJ
a.
c_
^_
o
o
OJ
tn
CJ
ex.
2
Ol
a;
CO
3
(X)
^ ^
£;
<-
L/l
CD
J5
'w
~
-C
1
cs;
cs;
- «
,j
"
o
CO
CSJ
CSJ
o
CO
^
+
IT)
P^
4,
0
o
1
CSJ
2
eo
I 1
+
a
CO
LD
0
1 1
4-
LO
p^s.
*jfl
CSI
o
0
o
1-4
.
^
CSJ
J.
LO
*
CSJ
o
^
CSi
CO
'-O
CTi
^
^^*
o
OJ
^_
r .
o
cr>
o
v
c
ro
CSJ
c;
LT)
^
^
4-
O
,_,
O
cr.
CS)
-f.
on
^f
CSJ
i
CSJ
o
ID
^>
+
Q
^r
CO
+
ro
*
O
CSJ
csi
Q
fs^
*
LP
CO
1 «
+
i-O
o
CSJ
o
c^
CSJ
4-
CSJ
CS.1
ro
"
CS!
1-C
*-^.
^^\
I «
CO
c
CSJ
II
CSJ
CSJ
ro
CSJ
*
+
C2
cr»
o
CSJ
4.
o
^^
i
CSJ
gT*.
CSJ
CSJ
4-
.
m
<*D
O
CS,
*CT
CSI
,-g.
*
^J-
1
-t-
CSJ
-H
CSJ
o
f*^.
^_
a
CO
LT)
CSJ
CC
O"*
CNJ
~^_
-1
1
JD
o
^-^
CC
OJ
u
OJ
D
OJ
2
c
E
a
CD
a;
j=
^
c:
.>
01
f ^
c
0
4-1
10
re
a
re
VI
a>
t/i
3
_§
01
,r
u
3
t
^~
re
j^
I/)
' '
t/)
s
5
s
0
0)
Q.
o
'jn
§
re
u
i.
O
o
~
re
x>
_
O
u_
re
c
,^-,
^
3J
OJ
E
re
c
re
»
*i/i
f^*- o_
i -
_|J n
OJ C
CD S-
r 3 -
.O t/1
I OJ
c a.
~
3 Q-
c
-C
Wl X
OJ to
CJ
en
a; j=
i 4->
a. ~
^ re
Q. 2
C~ I,
re
c = -
*2 !
-«-) CJ -w.
o cu *
c -c
o
a. en
o -^~
' / cr
~' ^
i- X
a. u
^
L. j_j
O lj
"*" ^
D
-t
^
h
I 1 4-1
^ (4-
CL
UT5
X
^^
,
O Al
f >
4-
^
i i o
o
LO re
O t/l
a.
X
c?
csj r^
CO 3 A1
^ 2
CSJ -^ ^~~
Lfl O
re :s
3
ajir
= =
* t/1 '1
X'
J 1_
a.
OJ
OJ
"s
I/1
CL1
"C ^
- ^
s_ -yi
O '.0
u_ it,
w"
i/'
O
I
O
"~~*
=
ro U
re
o
o o
o o
o *
n
Al
^2>
i-
'»
^ CC
^^r
:i 11
E-45
-------
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.
2. Reference 3, 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).
4. Telecon. Siebert, Paul, PES with Straitz, John III, National Air
Oil Burner Company, Inc. (NAO). November 1982. Design, operating
requirements, and costs of elevated flares.
5. Straitz, J.F. III. Make the Flare Protect the Environment.
Hydrocarbon Processing. 56. October 1977.
6. Oenbring, P.R. and T.R. Sifferman. Flare Design... Are Current
Methods Too Conservative? Hydrocarbon Processing. 59:124-129.
May 1980.
7. Telecon. Siebert, Paul, PES, with Keller, Mike, John Zink Co.
August 13, 1982. Clarification of comments on draft polymers and
resins CTG document.
8. Telecon. Siebert, Paul, PES with Fowler, Ed, NAO. November 12,
1982. Operating requirements of elevated flares.
9. Telecon. Siebert, Paul, PES with Fowler, Ed, NAO. November 5,
1982. Purchase costs and operating requirements of elevated
flares.
10. Telecon. Siebert, Paul, PES with Fowler, Ed, NAO. November 17,
1982. Purchase costs and operating requirements of eleveated
flares.
11. Memo from Senyk, David, EEA, to E3/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.
12. Memo from Mascone, D.C., EPA, to Farmer, J.R., EPA. June 11,
1980. Thermal incinerator performance for NSPS.
13. 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.
Fig. III-2, p. III-8.
E-46
-------
14. Reference 13, Fig. A-l, p. A-3.
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. 8-9.
16. EEA. Distillation NSPS Thermal Incinerator Costing Computer
Program (DSINCIN). flay 1981. p. 4.
17. Reference 13, p. 1-2.
18. Reference 15, p. G-3 and 6-4.
19, Reference 13, Fig. V-15, curve 3, p. V-18.
20. Neverill, 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.
21. Richardson Engineering Services. Process Plant Construction Cost
Estimating Standards, 1980-1981. 1980.
22. Telecon. Katari, Vishnu, Pacific Environmental Services, Inc.,
with Kroehling, John, DuPont, Torvex Catalytic Reactor Company.
October 19, 1982. Catalytic incinerator system cost estimates.
23. Telecon. Katari, Vishnu, Pacific Environmental Services, Inc.
with Tucker, Larry, Met-Pro Systems Division. October 19, 1982.
Catalytic incinerator system cost estimates.
24. Letter from Kroehling, John, DuPont, Torvex Catalytic Reactor
Company, to Katari, V., PES. October 19, 1982. Catalytic incinerator
system cost estimates.
25. Telecon. Katari, Vishnu, Pacific Environmental Services, Inc.,
with Redden, Charles, Artisan Company. September 29, 1982. Heat
exchanger system cost estimates.
25. Telecon. Katari, Vishnu, Pacific Environmental Services, Inc.,
with Mr. Ruck, Graham Company. September 29, 1982. Heat exchanger
system cost estimates.
27. Telecon. Katari, Vishnu, Pacific Environmental Services, Inc.,
with Glower, Dove, Adams Brothers, a representative of Graham
Company. September 30, 1982. Heat exchanger system cost estimates,
28. Telecon. Katari, Vishnu, Pacific Environmental Services, Inc.,
with Mahan, Randy, Brown Fintube Company. October 7, 1982. Heat
exchanger system cost estimates.
29. Perry, R.H. and C.H. Chilton, eds. Chemical Engineers' Handbook,
fifth edition. New York, McGraw-Hill Book Company. 1973. p. 5-3.
E-47
-------
30. McCabe, W.L. and J.C. Smith. Unit Operations of Chemical Engineering,
second edition. New York, McGraw-Hill Book Company. 1967.
1007 p.
31. Memo from Norwood, Tom, Pacific Environmental Services, Inc., to
Meardon, Ken, Pacific Environmental Services, Inc. January 1983.
Designs and Cost Estimates for ethylene glycol recovery systems.
32. Telecons. Meardon, Ken, Pacific Environmental Services, Inc.,
with Allied Fibers. December 1982 and January 1983. Design and
operating parameters of ethylene glycol recovery systems.
33. Reference 29, p. 5-31.
34. Chontos, L.W. Find Economic Pipe Diameter via Improved Formula.
Chemical Engineering. 87_(12):139-142. June 16, 1980.
35. Memo from Desai, Tarun, EEA, to EB/S Files. March 16, 1982.
Procedure to estimate piping costs.
36. Memo from Kawecki, Tom, EEA, to SOCMI Distillation File. November 13,
1981. Distillation pipeline costing model documentation.
37. EEA. SOCMI Distillation NSPS Pipeline Costing Computer Program
(DMPIPE), 1981.
38. Reference 20, Section 4.2, p. 4-15 through 4-28.
E-48
-------
i
T)
O
O O O
^^
< 5 ° "*
(5 W) C O ^
lilfl
W OQ. y»
^ u
^IM -
3 O
as
(A
S
a
323 =
mo o
m" " 2
Qo =3
- 3 J"
2.S 5 £
o *
_» ffl o
o ^ c.
o-S as
MS £
QJ y) CD
a| §
2 « S.
= H ?
^ 03_ <
5°
Tjm-TTo
5*3 2 3 S 8
u« 8 S ^,1
a»-< r 3 J«
01 03r,
= 5a§
S a
------- |