United States Office of Water EPA-821 -R-97-011
Environmental Protection (4303) October 1997
Agency Washington, DC 20460
&EPA Supplemental Technical
Development Document for
Effluent Limitations Guidelines
and Standards for the Pulp,
Paper, and Paperboard
Category
Subpart B (Bleached Papergrade Kraft
and Soda)
and
Subpart E (Papergrade Sulfite)
-------
DCN 14487
SUPPLEMENTAL TECHNICAL DEVELOPMENT DOCUMENT FOR
EFFLUENT LIMITATIONS GUIDELINES AND STANDARDS
FOR THE PULP, PAPER, AND PAPERBOARD CATEGORY
SUBPART B (BLEACHED PAPERGRADE KRAFT AND SODA)
AND
SUBPART E (PAPERGRADE SULFITE)
Engineering and Analysis Division
Office of Science and Technology
U.S. Environmental Protection Agency
Washington, D.C. 20460
October 15, 1997
-------
DISCLAIMER
This report has been reviewed and approved for publication by the Engineering
and Analysis Division, Office of Science and Technology. This report was prepared with the
support of Eastern Research Group, Inc. (Contract No. 68-C5-0013) under the direction and
review of the Office of Science and Technology. Neither the United States Government nor any
of its employees, contractors, subcontractors, or their employees make any warrant, expressed or
implied, or assumes any legal liability or responsibility for any third party's use of or the results
of such use of any information, apparatus, product, or process discussed in this report, or
represents that its use by such party would not infringe on privately owned rights.
-------
TABLE OF CONTENTS
Page
CONCLUSIONS 1-1
1.1 Introduction 1-1
1.2 Subcategorization 1-1
1.3 Scope of Rules 1-1
1.4 Best Practicable Control Technology Currently Available
(BPT) 1-2
1.5 Best Conventional Pollutant Control Technology (BCT) 1-2
1.6 Best Available Technology Economically Achievable
(BAT) 1-2
1.7 New Source Performance Standards (NSPS) 1-3
1.8 Pretreatment Standards for Existing Sources (PSES) 1-3
1.9 Pretreatment Standards for New Sources (PSNS) 1-3
1.10 Best Management Practices (BMP) 1-3
SCOPE OF RULEMAKING 2-1
2.1 Introduction 2-1
2.2 Effluent Limitations Guidelines and Standards 2-1
2.3 NESHAPs 2-3
2.4 References 2-5
SUMMARY OF DATA COLLECTION METHODS 3-1
3.1 Introduction 3-1
3.2 Data Received Since Proposal 3-1
3.3 Integrated Regulatory Development 3-1
3.4 References 3-4
INDUSTRY PROFILE 4-1
4.1 Introduction 4-1
4.2 Bleached Papergrade Kraft and Soda Subcategory 4-2
4.3 Papergrade Sulfite Subcategory 4-3
4.4 Trends in the Industry 4-3
4.5 References 4-4
SUBCATEGORIZATION 5-1
5.1 Introduction 5-1
5.2 Description of the Industry Subcategorization in Effect
Prior to the Promulgation of this Rule 5-1
5.3 Revised Industry Subcategorization 5-2
5.3.1 All Mills 5-5
5.3.2 Subpart B - Bleached Papergrade Kraft and Soda Subcategory 5-5
-------
/
TABLE OF CONTENTS (Continued)
Page
5.3.3 Subpart E - Papergrade Sulfite Subcategory 5-7
5.4 References 5-8
WATER USE AND WASTEWATER CHARACTERISTICS 6-1
6.1 Introduction 6-1
6.2 Mill Water Use - Bleach Plant Effluent Portion of Final
Effluent 6-1
6.3 Mill Water Use For Option A and Option B Mills 6-2
6.4 Definition of Process Wastewater 6-4
6.5 Use of Biocides 6-5
6.6 References 6-6
POLLUTION PREVENTION AND WASTEWATER
TREATMENT TECHNOLOGIES 7-1
7.1 Introduction 7-1
7.2 Pollution Prevention Controls Used in Pulping and
Delignification Processes 7-2
7.2.1 Chip Quality Control 7-2
7.2.2 Defoamers and Pitch Dispersants 7-3
7.2.3 Extended Cooking 7-4
7.2.4 Effective Brown Stock Washing 7-6
7.2.5 Closed Screen Room Operation 7-8
7.2.6 Oxygen Delignification 7-9
7.2.7 Steam Stripping 7-10
7.2.8 Spent Pulping Liquor Management, Spill
Prevention, and Control 7-11
7.2.9 Maximizing Recovery Boiler Capacity 7-12
7.3 Pollution Prevention Controls Used in the Bleach Plant 7-15
7.3.1 Ozone Bleaching 7-15
7.3.2 Improved Mixing and Process Control 7-17
7.3.3 Chlorine Dioxide Substitution 7-17
7.3.4 Enhanced Extraction 7-19
7.3.5 Elimination of Hypochlorite 7-21
7.3.6 Strategies to Minimize Kappa Factor and
DBD and DBF Precursors 7-21
7.3.7 Enzyme Bleaching 7-25
7.3.8 Peroxide Bleaching 7-25
7.3.9 Totally Chlorine-Free Bleaching of Papergrade Kraft Pulps .. 7-26
7.3.10 Totally Chlorine-Free Bleaching of Papergrade
Sulfite Pulps 7-27
7.4 End-of-Pipe Wastewater Treatment Technologies 7-28
in
-------
TABLE OF CONTENTS (Continued)
Page
7.5 References 7-29
DEVELOPMENT OF CONTROL AND TREATMENT
OPTIONS 8-1
8.1 Introduction 8-1
8.2 Toxic and Nonconventional Pollutants 8-2
8.2.1 Bleached Papergrade Kraft and Soda Subcategory (BPK) .... 8-2
8.2.2 Papergrade Sulfite Subcategory (PS) 8-4
8.2.3 Point of Compliance Monitoring 8-5
8.3 Conventional Pollutants 8-14
8.3.1 Approach to Option Development 8-15
8.3.2 Identification of Mills Representing Secondary
Wastewater Treatment Performance in Each
Subcategory 8-17
8.3.3 Control Options and Performance Levels 8-22
8.3.4 Description of Technology Bases 8-24
8.4 BAT 8-24
8.5 BPT 8-24
8.6 BCT 8-24
8.7 NSPS 8-25
8.7.1 Conventionals 8-25
8.7.2 Toxics and Nonconventionals 8-26
8.8 PSES 8-29
8.8.1 Performance of End-of-Pipe Secondary
Biological Treatment Systems 8-32
8.8.2 Bleached Papergrade Kraft and Soda Pass-Through
Analysis for PSES and PSNS 8-34
8.8.3 Papergrade Sulfite Pass-Through Analysis 8-41
8.9 PSNS 8-42
8.10 References 8-42
POLLUTANT REDUCTION ESTIMATES 9-1
9.1 Introduction 9-1
9.2 Pollutant Loading Calculations and Data Sources 9-3
9.2.1 Pollutant Loading Calculations 9-3
9.2.2 Data Sources 9-6
9.3 Industry Baseline Pollutant Loadings 9-8
9.3.1 AOX, Chlorinated Phenolic Compounds,
TCDD, and TCDF 9-9
9.3.2 Chloroform 9-13
9.3.3 COD and Color 9-14
IV
-------
TABLE OF CONTENTS (Continued)
Page
9.4 Pollutant Loadings After Implementation of the Control Options .... 9-14
9.5 Pollutant Reductions 9-16
9.6 References 9-16
10 BAT, PSES, NSPS, and BMP FINAL COMPLIANCE COSTS 10-1
10.1 Introduction 10-1
10.1.1 BAT and PSES Cost Estimation 10-1
10.2 Compliance Cost Estimates for the Bleached Papergrade
Kraft and Soda Subcategory 10-4
10.2.1 Technology Options 10-4
10.2.2 Costing Revisions 10-7
10.2.3 Baseline Status 10-11
10.2.4 Compliance Cost Estimates 10-12
10.2.5 Corporate Commitments to Install BAT Elements 10-16
10.2.6 Estimated Costs of TCP Bleaching Options 10-18
10.2.7 Voluntary Advanced Technology
Incentives Program Costing 10-19
10.3 Compliance Cost Estimates for the Papergrade
Sulfite Subcategory 10-20
10.3.1 Technology Options 10-20
10.3.2 Cost Model Revisions 10-21
10.3.3 Compliance Cost Estimates 10-21
10.4 NSPS Compliance Costs 10-23
10.4.1 NSPS Compliance Costs for the BPK Subcategory 10-24
10.5 References 10-26
11 NON-WATER QUALITY ENVIRONMENTAL IMPACTS 11-1
11.1 Impacts of BAT, PSES, and BMPs on the Bleached
Papergrade Kraft and Soda Subcategory 11-1
11.1.1 Summary of Impacts on Wood Consumption 11-1
11.1.2 Summary of Impacts on Wastewater Flow, BOD5,
and Solid Waste Generation 11-1
11.1.3 Summary of Impacts on Energy Consumption 11-2
11.1.4 Summary of Impacts on Atmospheric Emissions 11-2
11.2 Wood Consumption 11-3
11.2.1 BAT and PSES Option A 11-3
11.2.2 OptionB 11-4
11.3 Effluents and Solid Waste 11-4
11.3.1 Effluent Flows 11-4
11.3.2 Solid Wastes 11-6
11.4 Energy Impacts 11-10
-------
TABLE OF CONTENTS (Continued)
Page
11.4.1 Overview of Energy Impacts 11-10
11.4.2 Estimation of Energy Impacts 11-11
11.5 Atmospheric Emissions 11-14
11.5.1 Emissions Due to Mill Process Changes 11-14
11.5.2 Emissions Due to Burning Increased Quantities of
Black Liquor 11-15
11.5.3 Emissions Due to Changes in Energy
Consumption 11-17
11.5.4 Greenhouse Gases 11-17
11.5.5 Carbon Monoxide Emissions from
Oxygen Delignification 11-21
11.5.6 Carbon Monoxide from Chlorine Dioxide
Bleaching 11-21
11.6 Impacts of New Source Performance Standards
and Pretreatment Standards for New Sources
(NSPS and PSNS) on the Bleached Papergrade
Kraft and Soda Subcategory 11-24
11.7 Impacts of Totally Chlorine Free (TCP)
Technology on the Bleached Papergrade Kraft
and Soda Subcategory 11-25
11.8 Impacts of BAT, PSES, and BMPs on the
Papergrade Sulfite Subcategory 11-25
11.8.1 Wood Consumption 11-25
11.8.2 Solid Waste and Effluents 11-26
11.8.3 Energy Impacts 11-27
11.8.4 Atmospheric Emissions 11-27
11.9 Impacts of New Source Performance Standards
and Pretreatment Standards for New Sources (NSPS and
PSNS) for the Papergrade Sulfite Subcategory 11-27
11.10 References 11-27
11.11 Other References 11-30
12 BEST CONVENTIONAL POLLUTANT CONTROL
TECHNOLOGY 12-1
12.1 Introduction 12-1
12.2 Summary of the Final BCT Methodology 12-1
12.3 Revisions to the Proposed BCT Technology
Options and Cost Estimates 12-1
12.3.1 Conventional Pollutant Control Option Performance 12-2
12.3.2 Accounting for Cluster Rules Impacts on BCT Costs 12-3
12.3.3 Engineering Cost Estimates 12-3
VI
-------
TABLE OF CONTENTS (Continued)
Page
12.4 Final BCT Methodology 12-6
12.4.1 BCT Technology Basis 12-6
12.4.2 End-of-Pipe Treatment Costs 12-7
12.4.3 Conventional Pollutant Removals 12-7
12.4.4 BCT Cost Test Results 12-7
12.5 References 12-10
13 ABBREVIATIONS AND CONVERSIONS 13-1
13.1 Abbreviations 13-1
13.2 Units of Measure 13-4
13.3 Unit Conversions 13-6
vn
-------
LIST OF TABLES
Page
1-1 Comparison of the Final Codified Sub categorization
Scheme With the Previous Subcategorization Scheme 1-4
1-2 Application of Rules to Pulp, Paper, and Paperboard
Subcategories 1-6
2-1 Application of Rules to Pulp, Paper, and Paperboard
Subcategories 2-6
3-1 Major Post-Proposal Data Sources 3-5
4-1 Mills Omitted from EPA's Mid-1995 BAT/PSES Cost
and Loadings Estimates 4-5
4-2 Mid-1995 Applicability and Mill Counts by Subcategory 4-5
4-3 Applicability by mill for the Bleached Papergrade Kraft
and Soda and Papergrade Sulfite Facilities 4-6
8-1 Mills Representing the Performance of Secondary Wastewater
Treatment Bleached Papergrade Kraft and Soda Subcategory 8-45
8-2 Mills Representing the Performance of Secondary
Wastewater Treatment Papergrade Sulfite Subcategory 8-46
8-3 Data Available on the Use of TCF-Bleached Kraft Pulp 8-47
8-4 POTWs Receiving Chemical Pulp Mill Wastewaters 8-48
8-5 Pollutant Removals at POTWs Receiving Chemical Pulp Mill Wastewaters
(Calculated from Comment Submittals and EPA Permit Compliance System
Data) 8-49
8-6 National Council of the Paper Industry for Air and Stream Improvement, Inc.
(NCASI) Data Pollutant Removals at POTWs Receiving Chemical Pulp Mill
Wastewaters 8-52
8-7 Comparison of Pollutant Removals at POTWs Receiving Wastewaters from
Bleached Papergrade Kraft and Soda Mills and Direct-Discharging Bleached
Papergrade Kraft and Soda Mills 8-53
viii
-------
LIST OF TABLES
Page
9-1 Baseline Technology Groups for Bleached Papergrade
Kraft and Soda, and Papergrade Sulfite Mills: AOX,
9-2
9-3
9-4
9-5
9-6
9-7
9-8
9-9
9-10
9-11
9-12
9-13
9-14
9-15
9-16
9-17
9-18
9-19
9-20
Chlorinated Phenolic Compounds, TCDD, and TCDF
AOX Baseline Loadings
Production-Normalized Kraft Bleach Plant Flow Ratesa
TCDD Baseline Concentrations
TCDF Baseline Concentrations
Trichlorosyringol Baseline Concentrations
2,4,5-Trichlorophenol Baseline Concentrations
2,4,6-Trichlorophenol Baseline Concentrations
3,4,5-Trichlorocatechol Baseline Concentrations
3,4,5-Trichloroguaiacol Baseline Concentrations
3,4,6-Trichlorocatechol Baseline Concentrations
3,4,6-Trichloroguaiacol Baseline Concentrations
4,5,6-Trichloroguaiacol Baseline Concentrations
Tetrachlorocatechol Baseline Concentrations
Tetrachloroguaiacol Baseline Concentrations
2,3,4,6-Tetrachlorophenol Baseline Concentrations
Pentachlorophenol Baseline Concentrations
Baseline Technology Groups for Chloroform
Chloroform Baseline Loadings
Baseline Technology Groups for COD and Color
9-18
9-19
9-20
9-20
9-21
9-21
9-22
9-22
9-23
9-23
9-24
9-24
9-25
9-25
9-26
9-26
9-27
9-28
9-28
9-29
IX
-------
LIST OF TABLES
Page
9-21 COD and Color Baseline Loadings for Bleached Kraft
and Papergrade Sulfite Operations Only 9-29
9-22 Final Effluent Long-Term Average Loadings After
Implementation of the BPK and PS Options (kg/kkg) 9-30
9-23 Sources of Estimated Long-Term Average Pollutant
Loadings For Papergrade Sulfite BAT Options 9-31
9-24 Summary of Subcategory Loads and Reductions 9-32
10-1 Baseline Technology Groups for BPK Mills 10-28
10-2 Bleached Papergrade Kraft and Soda Process Technologies Costed 10-29
10-3 Baseline Status of Bleached Papergrade Kraft and Soda Mills 10-30
10-4 Comparison of Bleached Papergrade Kraft and Soda
BAT, PSES, andBMPs Compliance Cost Estimates 10-31
10-5 BAT, PSES, and BMPs Compliance Cost Estimates for
Direct and Indirect Discharging BPK Mills 10-32
10-6 Range of Estimated BAT, PSES, and BMPs Costs for
the 84 Bleached Kraft Mills 10-33
10-7 Component Capital Costs as Percentage of Total Capital
Cost for the Bleached Papergrade Kraft and Soda Mills 10-34
10-8 Component Operating Costs as Percentage of Total Operating Cost for the
Bleached Papergrade Kraft and Soda Mills 10-35
10-9 Bleached Papergrade Kraft and Soda Mill Technology
Upgrades Costed 10-37
10-10 Corporations Announcing Commitments to Upgrade Process
Technologies to Include BAT Elements After July 1, 1995 10-39
10-11 Baseline Status of Bleached Papergrade Kraft and Soda
Mills Including Adjustment for Corporate Commitments
to Install BAT Elements 10-40
-------
LIST OF TABLES
Page
10-12 Bleached Papergrade Kraft and Soda Total BAT, PSES,
and BMPs Cost Estimates with Process Upgrades
Announced but not Underway by Mid-1995 10-41
10-13 TCP Process Technologies Costed 10-42
10-14 Comparison of BAT and PSES Option Costs for the
Bleached Papergrade Kraft and Soda Subcategory 10-43
10-15 Case Study Mill Incentive Tier Costs 10-44
10-16 Papergrade Sulfite Technology Process Technologies Costed 10-45
10-17 ECF Versus TCP Costing for the Papergrade Sulfite Mills 10-46
10-18 Total BAT, PSES, and BMPs Papergrade Sulfite
Compliance Cost Estimates (All Segments) 10-47
10-19 Papergrade Sulfite Technology Upgrades 10-48
10-20 NSPS Compliance Costs 10-49
10-21 NSPS Limitations 10-50
11-1 Summary of Impacts of Options A and B Relative to Baselinea
for the Bleached Papergrade Kraft and Soda Subcategory 11-33
11-2 Bleach Plant Effluent Flow for Bleached Papergrade Kraft and
Soda Mills With and Without Extended Delignification 11-34
11-3 Black Liquor Solids, BOD5, and Sludge Generation for
Option A and Option B for the Bleached Papergrade
Kraft and Soda Subcategory 11-35
11-4 Effect of Options A and B on Energy Consumption
Relative to 1995 Base Case for the Bleached Papergrade
Kraft and Soda Subcategory 11-36
11-5 Process Changes Affecting Energy Consumption at
Bleached Papergrade Kraft and Soda Mills 11-37
XI
-------
LIST OF TABLES
Page
11-6 Impact of BAT, PSES, and BMPs: Bleached
Papergrade Kraft and Soda Subcategory Air Emissions 11-38
11-7 Atmospheric Emission Changes Due to Burning
Recovered Black Liquor, Bleached Papergrade Kraft
and Soda Subcategory Before MACT II is Applied 11-39
11-8 Atmospheric Emission Changes Due to Burning
Recovered Black Liquor, Bleached Papergrade Kraft
and Soda Subcategory After MACT II is Applied 11-41
11-9 Atmospheric Emissions: Oil Combustion and Oil Combustion
Plus BLS Combustion (Before MACT II is Applied) (Mg/yr) 11-43
11-10 Changes to Carbon Dioxide Emissions Resulting From
Changes in Consumption of Fossil Fuel and Wood
Consumption for Option A and Option B for the
Bleached Papergrade Kraft and Soda Subcategory 11-44
11-11 Average CO Emissions From Kraft Bleaching With
Chlorine Dioxide 11-45
11-12 Carbon Monoxide Emissions from Bleach Plants for
Option A and Option B for Bleached Papergrade Kraft
and Soda Subcategory 11-47
11-13 Comparison of NSPS/PSNS to Conventional Technology 11-48
11-14 Comparison of Two TCF Technologies to Conventional
Technology (Option A) for the Bleached Papergrade
Kraft and Soda Subcategory 11-49
12-1 Best Conventional Pollutant Control Technology (BCT) Costs 12-11
12-2 Conventional Pollutant Reductions Associated With BCT 12-12
12-3 Results of the Final BCT Cost Test 12-13
13-1 Units of Measurement 13-7
xn
-------
LIST OF FIGURES
Page
7-1 U.S. and Worldwide Increase in Kraft Pulp Produced by
Extended Cooking, 1983-1992 7-34
7-2 Extended Cooking Continuous Digester System
(EMCC®) 7-35
7-3 U.S. and Worldwide Increase in Kraft Pulp Produced by
Oxygen Delignification, 1970-1992 7-36
7-4 Typical Medium-Consistency Oxygen Delignification System 7-37
7-5 Typical High-Consistency Oxygen Delignification System 7-38
7-6 Continuous Steam Stripper System 7-39
11-1 Energy Impacts of Proposed Regulations, Bleached
Papergrade Kraft and Soda Subcategory 11-51
11-2 Emissions of CO Measured by NCASI 11-52
xni
-------
Section 1 - Conclusions
SECTION 1
CONCLUSIONS
1.1 Introduction
The new regulations for the pulp, paper, and paperboard industry, also known as
the "Cluster Rules," include effluent limitations guidelines and standards for the control of
wastewater pollutants and national emission standards for hazardous air pollutants. Information
and rationale supporting the proposed effluent limitations guidelines and standards were
provided in "Proposed Technical Development Document for the Pulp, Paper, and Paperboard
Category Effluent Limitations Guidelines, Pretreatment Standards, and New Source Performance
Standards," (the TDD) October 1993, EPA-821-R-93-019. Technical information and rationale
supporting the proposed air emission standards were provided in "Pulp, Paper, and Paperboard
Industry - Background Information for Proposed Air Emission Standards," (the Background
Information Document, or the BID) October 1993, EPA-453-R93-050a.
To support the final regulations, EPA chose not to rewrite the entire TDD and the
BID but instead to prepare several technical support documents to supplement the TDD and the
BID. This document, and several others referenced herein, support the final effluent limitations
guidelines and standards. The TDD is superseded to the extent that this document or the other
technical documents supporting the final rule are inconsistent with it. This document is referred
to in the other support documents as the Supplemental Technical Development Document
(STDD), Document Control Number (DCN) 14487. This section of the STDD highlights key
aspects of the final effluent limitations guidelines and standards.
1.2 Subcategorization
EPA is revising the existing Subcategorization scheme for effluent limitations
guidelines and standards for this industry (40 CFR Parts 430 and 431). The new effluent
limitations guidelines and standards that are being promulgated today affect only those mills in
the new Bleached Papergrade Kraft and Soda Subcategory (Subpart B) and the new Papergrade
Sulfite Subcategory (Subpart E). EPA has reprinted in their entirety the current effluent
limitations guidelines and standards that remain applicable to mills subject to the new
subcategories. Table 1-1 summarizes the new subcategories and the corresponding subcategories
from the existing regulations.
1.3 Scope of Rules
The proposed rules applied to mills within the U.S. Department of Commerce,
Bureau of the Census Standard Industrial Classifications (SIC) 2611 (pulp mills), 2621 (paper
mills except building paper mills), 2631 (paperboard mills), and 2661 (building paper and
building board mills). All of the mills affected by the new effluent limitations guidelines and
standards are in SIC 2611 (pulp mills). Since the proposal, the Office of Management and
1-1
-------
Section 1 - Conclusions
Budget began to use the North American Industry Classification System (NAICS). The
applicable NAICS numbers are 32211 (pulp mills), 322121 (paper mills, except newsprint mills),
322122 (newsprint mills), and 32213 (paperboard mills). The components of these rules
applicable to each subcategory of the Pulp, Paper, and Paperboard Point Source Category are
shown on Table 1-2.
1.4 Best Practicable Control Technology Currently Available (BPT)
EPA proposed revisions to the existing BPT effluent limitations guidelines for
five-day biochemical oxygen demand (BOD5) and total suspended solids (TSS) for all
subcategories of the pulp, paper, and paperboard industry. The proposed revisions were based on
the application of secondary wastewater treatment with appropriate water use and reuse.
However, for the reasons set forth in the preamble to the final rules, EPA in the exercise of its
discretion has decided not to revise the BPT effluent limitations guidelines for conventional
pollutants for Subparts B and E. The existing BPT guidelines will continue to apply.
1.5 Best Conventional Pollutant Control Technology (BCT)
EPA proposed revisions to the BCT effluent limitations guidelines for BOD5 and
TSS for all subcategories of the pulp, paper, and paperboard industry. After proposal, EPA
considered further whether technologies are available for Subparts B and E that achieve greater
removals of conventional pollutants than the current BPT effluent limitations guidelines, and
whether those technologies are cost reasonable according to the BCT cost test. After evaluating
the candidate BCT technologies for both Subparts B and E, EPA concluded that none of the
candidate options passed the BCT cost test; therefore, more stringent BCT effluent limitations
guidelines are not being promulgated for Subparts B or E.
1.6 Best Available Technology Economically Achievable (BAT)
EPA is promulgating BAT effluent limitations guidelines for Subparts B and E to
control toxic and nonconventional pollutants in the bleach plant effluent and in the end-of-pipe
effluent. For Subpart B mills, the technology basis for these effluent limitations guidelines is
complete (100 percent) substitution of chlorine dioxide for chlorine in the bleaching process
along with the other elements as presented in Section 8 of this document. For Subpart E mills,
the technology basis for these effluent limitations guidelines is either complete (100 percent)
substitution of chlorine dioxide for chlorine in the bleaching process and other elements, or
totally chlorine-free (TCF) bleaching. Section 8 of this document explains these options in more
detail.
In addition to the effluent limitations guidelines based on complete substitution,
EPA is promulgating for Subpart B the Voluntary Advanced Technology Incentives Program for
direct discharging mills that have or plan to install advanced technology beyond that which forms
the basis of today's BAT and New Source Performance Standards (NSPS). The Incentives
1-2
-------
Section 1 - Conclusions
Program is discussed in more detail in a separate document entitled "Technical Support
Document for the Voluntary Advanced Technology Incentives Program, DCN 14488."
1.7 New Source Performance Standards (NSPS)
EPA is revising the NSPS for toxic, nonconventional, and conventional (BOD5
and TSS) pollutants for Subpart B. For Subpart B mills, the technology basis for these
limitations is extended delignification followed by complete (100 percent) substitution of
chlorine dioxide for chlorine in the bleaching process, well-operated biological treatment, and
other elements as presented in Section 8 of this document. For Subpart E mills, EPA is revising
the NSPS for toxic and nonconventional pollutants. For Subpart E mills, the technology basis for
these limitations is the same as the BAT limitations.
1.8 Pretreatment Standards for Existing Sources (PSES)
EPA is revising the PSES for toxic and nonconventional pollutants for Subparts B
and E. The technology basis of these standards is the same as the basis for the BAT limitations
promulgated today, with the exception of biological treatment. Mills must monitor for
compliance with these standards at the bleach plant effluent.
1.9 Pretreatment Standards for New Sources (PSNS)
EPA is revising the PSNS for toxic and nonconventional pollutants for Subparts B
and E. The technology basis of these standards is the same as the basis for the NSPS limitations
promulgated today, with the exception of biological treatment. Mills must monitor for
compliance with these standards at the bleach plant effluent.
1.10 Best Management Practices (BMP)
EPA is promulgating BMPs for direct and indirect discharging mills regulated
under Subparts B and E. These BMPs are intended to prevent or otherwise contain leaks and
spills and to control intentional diversions of spent pulping liquor, soap, and turpentine. The
BMPs will reduce wastewater loadings of nonchlorinated toxic compounds and hazardous
substances and, as an incidental matter, loadings of other pollutants. These BMPs are discussed
in more detail in a separate document entitled "Technical Support Document for Best
Management Practices for Spent Pulping Liquor Management, Spill Prevention, and Control,
DCN 14489."
1-3
-------
Section 1 - Conclusions
Table 1-1
Comparison of the Final Codified Subcategorization Scheme
With the Previous Subcategorization Scheme
New
Subpart
A
B
C
D
E
F
G
H
I
J
New Subcategorization Scheme
Dissolving Kraft
Bleached Papergrade Kraft and Soda
Unbleached Kraft
Dissolving Sulfite
Papergrade Sulfite
Semi-Chemical
Mechanical Pulp
Non-Wood Chemical Pulp
Secondary Fiber Deink
Secondary Fiber Non-Deink
Existing Subcategorization Scheme (With
Existing 40 CFRPart 430 Subparts Noted)
Dissolving Kraft (F)
Market Bleached Kraft (G),
BCT Bleached Kraft (H),
Fine Bleached Kraft (I),
Soda (P)
Unbleached Kraft (A)
- Linerboard
- Bag and Other Products
Unbleached Kraft and Semi-Chemical (D, V)
Dissolving Sulfite (K)
- Nitration
- Viscose
- Cellophane
- Acetate
Papergrade Sulfite (J,U)
- Blow Pit Wash
- Drum Wash
Semi-Chemical (B)
- Ammonia
- Sodium
GW-Thermo-Mechanical (M),
GW-Coarse, Molded, News (N)
GW-Fine Papers (O)
GW-Chemi-Mechanical (L)
Non-Wood Chemical Pulp Mills
Deink Secondary Fiber (Q)
- Fine Papers
- Tissue Papers
- Newsprint
Tissue from Wastepaper (T)
Paperboard from Wastepaper (E)
- Corrugating Medium
- Non-Corrugating Medium
Wastepaper-Molded Products (W)
Builders' Paper and Roofing Felt (40 CFR Part
431 Subpart A)
1-4
-------
Section 1 - Conclusions
Table 1-1 (Continued)
New
Subpart
New Subcategorization Scheme
Existing Subcategorization Scheme (With
Existing 40 CFR Part 430 Subparts Noted)
K
Fine and Lightweight
Papers from Purchased Pulp
Non-Integrated Fine Papers (R)
- Wood Fiber Furnish
- Cotton Fiber Furnish
Lightweight Papers (X)
- Lightweight Papers
- Lightweight Electrical Papers
L
Tissue, Filter, Non-Woven, and
Paperboard from Purchased Pulp
Non-Integrated
- Tissue Papers (S)
- Filter and Non-Woven (Y)
- Paperboard (Z)
See the Code of Federal Regulations (CFR), Title 40, Chapter I, volume including Parts 425 to 699, Part 430, edition
as of July 1, 1997, for the Subcategorization scheme being supercoded. See the CFR edition to be published as of
July 1, 1998, for the final codified Subcategorization scheme.
1-5
-------
Section 1 - Conclusions
Table 1-2
Application of Rules to Pulp, Paper, and
Paperboard Subcategories
New Effluent Guidelines
Subcategory
Dissolving Kraft
Bleached Papergrade Kraft
and Soda
Unbleached Kraft
Dissolving Sulfite
Papergrade Sulfite
Semi-Chemical
Mechanical Pulp
Non-Wood Chemical Pulp
Secondary Fiber Deink
Secondary Fiber
Non-Deink
Fine and Lightweight
Papers from Purchased Pulp
Tissue, Filter, Non- Woven,
and Paperboard from
Purchased Pulp
New Effluent
Guidelines
Subpart
A
B
C
D
E
F
G
H
I
J
K
L
Clean
Air
Act
NESHAP"
X
X
X
X
X
X
xe
xe
xe
xe
xe
xe
Clean Water Act
Toxic &
Nonconven-
tional":
BAT, NSPS,
PSES, and
PSNS
X
X
Conven-
tional0:
NSPS
X
xd
BMP
X
X
aNational Emission Standards for Hazardous Air Pollutants.
bToxic and nonconventional pollutants in this rulemaking include: 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD),
2,3,7,8-tetrachlorodibenzofuran (TCDF), chloroform, 12 chlorinated phenolic compounds and adsorbable organic
halides (AOX). The 12 chlorinated phenolic compounds include trichlorosyringol, 2,4,5-trichlorophenol, 2,4,6-
trichlorophenol, 3,4,5-trichlorocatechol, 3,4,5-trichloroguaiacol, 3,4,6-trichlorocatechol, 3,4,6-trichloroguaiacol,
4,5,6-trichloroguaiacol, tetrachlorocatechol, tetrachlorguaiacol, 2,3,4,6-tetrachlorophenol, and pentachlorophenol.
Conventional pollutants in this rulemaking include BOD5 and TSS.
dEPA is not promulgating NSPS for conventional pollutants for Subpart E. Existing standards continue to apply.
eRule applies if these mills operate a bleach plant that uses chlorine or chlorine dioxide.
1-6
-------
Section 2 - Scope of Rulemaking
SECTION 2
SCOPE OF RULEMAKING
2.1 Introduction
Approximately 565 pulp, paper, and paperboard mills operate in the United States
(1). Most of these mills are subject to effluent limitations guidelines or standards as a result of
regulations promulgated in the 1970s and 1980s. These existing regulations are based on the
division of the mills into 26 subcategories. Each subcategory is defined by processes employed
and/or products manufactured, and has a separate set of effluent regulations.
In 1993, EPA proposed revisions to the existing effluent regulations for all 565
mills and new regulations for some mills. EPA also proposed to consolidate the existing 26
subcategories into 12 new subcategories, based primarily on the pulping process used at each mill.
In today's rulemaking, EPA is promulgating the revised subcategorization scheme. However, at
this time, EPA is not revising the existing effluent regulations for any mills except those mills in
two of the revised subcategories: Bleached Papergrade Kraft and Soda mills (Subpart B) and
Papergrade Sulfite mills (Subpart E). The number of mills in each of these subcategories is
discussed in Section 4 of this document.
In 1993, EPA also proposed national emission standards for hazardous air
pollutants (NESHAPs) for some of the 565 mills. EPA's proposed NESHAPs would affect pulp
mills in six effluent guideline subcategories (see Tables 1-2 and 2-1). Although the industry was
not originally subcategorized, EPA has subcategorized the industry since the proposal for the
purposes of selecting the maximum achievable control technologies (MACT) which form the basis
of the NESHAPs. EPA established four subcategories for mills that chemically pulp wood fiber:
kraft, sulfite, soda, and semichemical. EPA's Office of Air and Radiation (OAR) also separated
the MACT standards into three components: MACT I, MACT II, and MACT III. Each
component of the MACT standards is described in Section 2.3.
2.2 Effluent Limitations Guidelines and Standards
The Clean Water Act (CWA) authorizes EPA to develop the regulations to control
the amount of pollutants discharged to navigable waters of the United States by industrial
dischargers. In 1993, for the Pulp, Paper, and Paperboard Point Source Category, EPA proposed
the following regulations:
BPT (best practicable control technology currently available);
BCT (best conventional pollutant control technology);
BAT (best available technology economically achievable);
NSPS (new source performance standards);
PSES (pretreatment standards for existing sources); and
PSNS (pretreatment standards for new sources).
2-1
-------
Section 2 - Scope of Rulemaking
These regulations establish quantitative limits on the discharge of pollutants from
industrial point sources. As explained in the preamble and in Section 12 of this document, EPA
decided not to promulgate proposed regulations for BPT and BCT; however, previously
promulgated BPT and BCT regulations remain in effect. Regulations for BAT, NSPS, PSES, and
PSNS are being promulgated. The applicability of the various regulations is summarized below:
BPT
BCT
BAT
NSPS
PSES
PSNS
Direct
Discharge
X
X
X
X
Indirect
Discharge
X
X
Existing
Source
X
X
X
X
New
Source
X
X
Conventional
Pollutants
X
X
X
Toxic and
Nonconventional
Pollutants
X
X
X
X
All of these regulations are based upon the performance of specific technologies
but do not require the use of any specific technology. The regulations applicable to direct
dischargers are effluent limitations guidelines, applied to pretreatment individual facilities through
National Pollutant Discharge Elimination System (NPDES) permits issued by EPA or authorized
states under Section 402 of the CWA. The regulations applicable to indirect dischargers are
pretreatment standards, administered by local permitting authorities (i.e., the government entity
controlling the Publicly Owned Treatment Works (POTW) to which the industrial wastewater is
discharged). The pretreatment standards are designed to control pollutants that pass through or
interfere with POTWs.
EPA is now promulgating BAT, NSPS, PSES, and PSNS for two subcategories of
the Pulp, Paper, and Paperboard Point Source Category: the Bleached Papergrade Kraft and Soda
Subcategory (Subpart B) and the Papergrade Sulfite Subcategory (Subpart E). The Bleached
Papergrade Kraft and Soda Subcategory is comprised of 86 mills that use a kraft or soda pulping
process followed by a bleach plant (see Section 4). The Papergrade Sulfite Subcategory is
comprised of 11 mills that use a sulfite pulping process; ten of the these mills have a bleach plant
while one mill makes unbleached pulp.
In 1993, EPA had also proposed regulations requiring the implementation of
BMPs. EPA is now promulgating BMP regulations applicable to direct and indirect discharging
mills with pulp production in Subpart B or Subpart E.
2-2
-------
Section 2 - Scope of Rulemaking
2.3 NESHAPs
In 1990, Congress passed comprehensive amendments to Section 112 of the Clean
Air Act. The objective of these amendments is to reduce nationwide air toxic emissions.
Congress identified 189 "hazardous air pollutants" (HAPs) to be controlled by a regulatory
structure based on source categories. NESHAPs based on Maximum Achievable Control
Technology (MACT) must be established for new and existing sources. In no case can the
NESHAPs be less stringent than the "MACT floor" for existing sources, which, roughly
paraphrasing, requires the standard to be at least as stringent as the average emission limitation
achieved by the best performing 12 percent of sources (Clean Air Act (CAA) Section 112(d)(2)).
For more detailed discussion, see Section VIA of the preamble for the final rules.
NESHAPs for the pulp and paper manufacturing source category are divided into
three parts:
MACT I controls emissions from noncombustion sources from pulping and
bleaching operations at chemical (kraft, soda, and sulfite) and semi-
chemical wood pulping mills.
MACT II (which is being proposed) controls emissions from combustion
sources (e.g., recovery furnaces, lime kilns, smelt dissolving tanks) from
chemical recovery operations at wood pulping mills (kraft, soda, sulfite,
and semichemical).
MACT III addresses emissions from noncombustion sources from mills
that mechanically pulp wood, pulp secondary fibers, or pulp nonwood
materials, and those that use papermachine additives and solvents.
MACT I controls primarily volatile HAPs, while MACT II controls metal HAPs.
For bleached papergrade kraft and soda mills, MACT I (for the most important
emission points) is based on the performance of the following control technologies:
Collection and destruction of organic HAPs emitted by pulping area vents;
Controls on condensate streams from the digester system, evaporator
system, turpentine recovery system, and high volume low concentration
(HVLC) vents (brown stock washers and oxygen delignification (OD)), if
any, but under a delayed compliance schedule (eight years) to encourage
upgrade of brown stock washers and installation of OD;
2-3
-------
Section 2 - Scope of Rulemaking
Collection and treatment (caustic scrubber) of bleach plant vents to control
hydrogen chloride and other chlorinated HAPs (other than chloroform);
Compliance with BAT limits on chloroform or elimination of hypochlorite
and complete substitution of chlorine dioxide for elemental chlorine; and
Collection and treatment of pulping area condensates, including evaporator
"foul" condensates by steam stripping or biological treatment.
For sulfite mills, MACT I is based on the performance of:
Collection and destruction of organic HAPs emitted by digester system
vents, evaporator system vents, and pulp washing system vents;
Collection and neutralization (caustic scrubber) of organic HAPs emitted
by bleach plant vents to control hydrogen chloride and other chlorinated
HAPs (note that the scrubber does not control chloroform); and
Compliance with BAT limits on chloroform (these limits are based on
elimination of hypochlorite and complete substitution of chlorine dioxide
for elemental chlorine or totally chlorine free bleaching).
The MACT II proposal is based on:
Control of particulate HAPs by electrostatic precipitators (for recovery
furnaces and lime kilns) and venturi scrubbers (for lime kilns and smelt
dissolving tanks).
MACT I was proposed on December 17, 1993 (the same time as the effluent
limitations guidelines) and is being promulgated today. MACT II is being proposed today. On
September 29, 1995 a Presumptive MACT report was issued for the MACT III source category.
Presumptive MACT is an estimate of MACT based on an assessment of readily available
information and information gathered through consultation with experts in state and local
agencies, EPA, environmental groups, and the regulated industry. No information was identified
during the Presumptive MACT process to suggest that sources associated with the MACT III
source category warrant listing pursuant to Section 112(c)(3) of the CAA.
Further, MACT III sources have no air pollution control devices, so the floor for
these sources is no control. In addition, available information indicates add-on controls would not
be cost effective for most emission points. EPA, therefore, decided in most instances not to
require controls beyond the floor. Mills covered by MACT III must still collect and treat bleach
2-4
-------
Section 2 - Scope of Rulemaking
plant vents if they operate bleach plants that use chlorine or chlorinated compounds.
EPA has taken no action on MACT standards for chemical additives and solvents
at the paper machines. If information becomes available regarding cost-effective HAP controls
beyond the floor for these sources, EPA will propose a MACT standard in the future.
2.4 References
1. Development Document for Proposed Effluent Limitations Guidelines and
Standards for the Pulp. Paper, and Paperboard Point Source Category.
EPA-821-R-93-019, U.S. EPA, Washington DC, October 1993.
2-5
-------
Section 2 - Scope of Rulemaking
Table 2-1
Application of Rules to Pulp, Paper, and
Paperboard Subcategories
NESHAP Subcategory
Dissolving Kraft Pulping
Papergrade Kraft and Soda
Pulping
Unbleached Kraft Pulping
Dissolving Sulfite Pulping
Papergrade Sulfite Pulping
Semi-Chemical Pulping
Mechanical Pulping
Non-Wood Pulping
Secondary Fiber Pulping
Papermaking Systems
New Effluent
Guidelines
Subpart
A
B
C
D
E
F
G
H
I,J
K-LC
Today's
Clean
Water
Act
Rules
X
X
Clean Air Act Rules
MACT
I
X
X
X
X
X
X
MACT
II
X
X
X
X
X
X
MACT
III
X
X
X
X
X
X
xa
xa
xa
xb
alf these mills operate a bleach plant that uses chlorine or chlorine dioxide.
bConsidered under MACT III but no controls are required.
°Applicable to stand-alone papermaking systems; such systems at integrated mills are covered by the effluent limitations
guidelines for the applicable subparts.
2-6
-------
Section 3 - Summary of Data Collection Methods
SECTION 3
SUMMARY OF DATA COLLECTION METHODS
3.1 Introduction
EPA collected information necessary for the development of the effluent
limitations guidelines and standards from many sources, including several Agency and industry
sampling programs, an industry-wide census questionnaire, questionnaire surveys submitted to
mills in other countries, industry trade associations, public meetings, mill site visits, conferences,
literature reviews, and other EPA offices. The data sources utilized for the proposed rules were
described in the TDD. This section describes the data collected since proposal.
3.2 Data Received Since Proposal
EPA has gathered a substantial amount of new information and data since
proposal. Much of this information was collected with the cooperation and support of the
American Forest and Paper Association (AF&PA) and the National Council of the Paper Industry
for Air and Stream Improvement Inc. (NCASI). Many individual mills in the U.S. and abroad, as
well as environmental groups have also assisted in gathering this information. The comments
received on the proposal and the July 1996 notice of data availability were also an important
source of information. The data gathering activities for this final rule are summarized in detail in
the proposal (58 FR at 66096), and in the July 15, 1996 notice of data availability (61 FR at
36837).
Some of the new information and data were generated through EPA-sponsored
field sampling or visits at individual mills in the U.S., Canada, and Europe. Additional sampling
data were voluntarily supplied by many facilities, along with information from laboratory and
pilot-scale studies. In order to respond to and clarify comments on the proposal, the Agency also
gathered and received voluntarily submitted information including data on secondary fiber mill
processes, recovery furnace capacities, best management practices, capital and operating costs,
process operations, and impacts of pollution prevention technology on the recovery cycle. Some
of the major data collection activities are listed in Table 3-1 with references to the rulemaking
record section(s) where the information can be found.
3.3 Integrated Regulatory Development
In 1990, EPA established the Pulp and Paper Regulatory Cluster, comprised of
representatives from various EPA offices. One role of the Pulp and Paper Regulatory Cluster
was to identify optimal approaches to solving environmental problems associated with the pulp
and paper industry through regulatory coordination. As a result of the Cluster's efforts, the
effluent limitations guidelines and the NESHAP rulemakings for the pulp and paper industry
were integrated and jointly proposed. Regulation of land application of pulp and paper mill
3-1
-------
Section 3 - Summary of Data Collection Methods
sludge also was considered in the Agency's coordinated regulatory strategy. The joint proposal of
the effluent limitations and NESHAPs eventually became known as the Cluster Rules.
The first step in developing the Cluster Rules was to collect mill-specific
information from all facilities subject to both the effluent limitations and the NESHAPs. As
described in Section 3.0 of the TDD, EPA used information from many sources to develop the
integrated regulatory options proposed in 1993. The information collected includes the process
and control technologies in use, data representing the performance of these technologies, and
financial information used in the analysis of the economic impact of these options. This
information was compiled in a mill-specific database for use in developing the effluent
limitations and NESHAPs. Estimated costs, pollutant reductions, and other environmental
impacts for each regulatory alternative were then developed and various combinations of these
alternatives were analyzed. See the Economic Analysis (DCN 14649) for a detailed presentation
of the economic and related analyses (3).
The control technologies considered as the bases for BAT, PSES, NSPS, and
PSNS are described in Sections 7.0 and 8.0. The control technologies considered as the bases for
BMP and NESHAP are described in separate documents (1,2). The control options for BAT and
PSES involve pulping and bleaching process changes. The performance of existing secondary
biological wastewater treatment systems employed by direct dischargers and POTWs also were
considered in developing these options. The BMP requires prevention and control of leaks,
spills, and intentional diversions of spent pulping liquor, soap, and turpentine. The NESHAP
control technologies include steam strippers, combustion, and caustic scrubbing. Steam strippers
are used to remove HAPs from pulping area condensates. The "clean condensate alternative" in
the MACT I standards also allow mills as a compliance alternative to treat hard pipe condensates
in end-of-pipe secondary biological treatment systems. See the preamble for the final rules at
Section VI.A.3(d). Combustion devices are used to destroy non-chlorinated HAPs removed by
steam strippers and hard-piped air emission streams. Combustion devices include stand-alone
devices such as thermal incinerators or existing devices such as lime kilns, power boilers, and
recovery furnaces. Caustic scrubbers and process changes are used to reduce chlorinated HAP
emissions in the bleaching area.
EPA developed regulatory alternatives based on pulping and bleaching process
changes alone, air emission control options alone, and combinations of process changes and air
emission controls. Each regulatory alternative also included secondary wastewater treatment,
and spill prevention and control components. The alternatives were designed to evaluate the
most efficient application of control technologies to minimize the cross-media transfer of
pollutants between air and water, and partitioning of pollutants (e.g., dioxins) to sludges and
pulps. EPA's economic analysis summarizes the costs and benefits of each regulatory alternative
evaluated by EPA for the final cluster rules (3).
EPA evaluated whether the pulping and bleaching process changes that form the
basis of BAT and PSES reduce HAP emissions sufficiently to satisfy CAA requirements. Based
on available data, the analyses showed that the use of the bleaching process technologies
3-2
-------
Section 3 - Summary of Data Collection Methods
decrease uncontrolled emissions of chlorinated HAPs (including chloroform, chlorine, and
hydrochloric acid), but increase others. This decrease in uncontrolled air emissions of
chlorinated HAPs is attributable to the elimination of hypochlorite as a bleaching agent and use
of complete chlorine dioxide substitution. However, uncontrolled air emissions of some non-
chlorinated HAPs, including methanol, methyl ethyl ketone, and formaldehyde, show modest
increases as a result of the bleaching process changes (see Section 11). EPA decided that the
bleaching process changes and existing bleach plant caustic scrubbers sufficiently reduced and
controlled emissions of chlorinated HAPs. However, additional controls were needed in the
pulping area to satisfy the CAA requirements for non-chlorinated HAPs.
EPA also considered the effect of the air pollution controls on effluent loadings of
toxic and nonconventional pollutants. The analyses showed that the major air pollution controls
that form the basis for NESHAP (steam stripping, combustion, and caustic scrubbing) did not
significantly affect effluent loadings of these pollutants. Steam stripping systems remove
compounds from pulping area condensates. Combustion destroys the compounds removed from
the condensates along with most compounds emitted from process vents. Steam stripping and
combustion reduce the amount of pollutants that could enter surface waters due to deposition and
the volume of wastewater discharged to the wastewater treatment system. Chlorinated HAPs that
remain in bleaching area wastewaters after process changes are implemented react with caustic in
the scrubber, neutralizing the caustic effluent. Non-chlorinated HAPs that absorb into the caustic
are bio-degradable, and are not estimated to significantly increase the pollutant load to the
wastewater treatment system. Caustic scrubbing operations are also not expected to significantly
increase the volume of wastewater discharges to the wastewater treatment system.
The analyses of multiple regulatory alternatives showed that no single control or
process change technology is currently available to reduce pollutant discharges to the air and
water to levels required by the respective statutes. The demonstrated control technologies that
can serve as bases for BMP, BAT, PSES, NSPS, and PSNS pose no significant adverse impacts
to and have some benefits for air quality. Similarly, the BMPs reduce leaks and spills and are
capable of reducing intentional diversions of pulping liquors, soaps, and turpentine while
increasing recovery of important process chemicals and energy (1); the air emissions control
technologies that can serve as the basis for NESHAP pose no significant adverse impacts on and
have some benefits for water quality. Therefore, combining the best control technology options
for effluent limitations with the best control technology options for air emission standards
represents a reasonable method for constructing the final regulatory alternative. EPA selected
control options for the final rulemaking based on evaluation of pollutant reductions; costs; and
economic, nonwater quality and non-air quality, and energy impacts (3). EPA also considered
cost effectiveness and environmental impacts.
3-3
-------
Section 3 - Summary of Data Collection Methods
3.4 References
1. Technical Support Document for Best Management Practices for Spent Pulping
Liquor Management Spill Prevention, and Control. EPA, Washington DC,
Record Section 30.9, DCN 14489, 1997.
2. Anderson, Donald F., Memorandum on Data Available for Development of COD
Limitations. Record Section 22.4, DCN 14788, September 30, 1997.
3. Economic Analysis for the National Emission Standards for Hazardous Air
Pollutants for Source Category: Pulp and Paper Production: Effluent Limitations
Guidelines, Pretreatment Standards, and New Source Performance Standards:
Pulp, Paper, and Paperboard Category - Phase I. Prepared by ERG for EPA.
Record Section 30.5, DCN 14649, 1997.
3-4
-------
Section 3 - Summary of Data Collection Methods
Table 3-1
Major Post-Proposal Data Sources
Data Source
Public Comments
Voluntary Data Submissions from Industry (NCASI, AF&PA)
Secondary Fiber Questionnaires
Recovery Furnace Capacity Surveys
BMP Questionnaire
Capital and Operating Cost Requests
Operating Data Requests for Recently Installed Pulping and
Bleaching Technologies
Request for Information About the Impact of Technologies on the
Recovery Cycle
Data Supplied by Individual U.S. Mills
Data Collected by EPA at U.S., Canadian, and European Mills
Data Supplied by NCASI from Canadian Mills
Laboratory Trial Data Submitted by Individual Companies (Dissolving
Mills and Papergrade Sulfite Mills)
Other Data Supplied by NCASI
Record Section
19.1
21.1
21.1.1
21.1.2
21.1.3
21.1.4
21.1.5
21.1.6
21.6.1.1
21.6.1.2
21.6.1.7
21.11
21.12
3-5
-------
Section 4 - Industry Profile
SECTION 4
INDUSTRY PROFILE
4.1 Introduction
This section discusses the number of pulp and paper mills included in the effluent
limitations guidelines analyses. The mill counts in this section have been updated for the Bleached
Papergrade Kraft and Soda and Papergrade Sulfite subcategories only.
The Office of Water promulgation baseline is mid-1995. The EPA mid-1995
database includes 86 bleached papergrade kraft and soda mills and 11 papergrade sulfite mills.
The total number of bleached kraft mills was reduced from 88 at proposal to 86 now, due to the
closing of one bleached papergrade kraft mill (Simpson Paper, Fairhaven, CA) and the
reclassification of another mill as unbleached kraft (Port Townsend, Port Townsend, WA). All 11
papergrade sulfite mills (one mill received BMP cost estimates only, Great Northern Paper,
Millinocket, ME) and 84 of the 86 bleached papergrade kraft and soda mills have been included in
EPA's mid-1995 cost and loading estimation efforts (See Section 10 and Section 9, respectively).
Note that one mill manufactures both bleached papergrade kraft and papergrade sulfite pulp, so
there are a total of 96 mills in the two subcategories.
Table 4-1 lists the mills not included in the costs and loadings analyses presented in
the July 1986 notice. Costs and loadings estimates for the Georgia-Pacific specialty grade sulfite
mill in Bellingham, WA were not included in the July 1996 Notice. Estimates for this mill are
included, however, in the results presented in this document. As of mid-1995, the Bellingham mill
was the only mill in the Specialty Grade segment of Subpart E. EPA expects another papergrade
sulfite facility to enter the specialty-grade market in the near future. Also, Badger Paper's
papergrade sulfite mill in Peshtigo, WI shut down its pulping process in September 1996 (paper
making operations have continued) but is still included in the analyses because EPA's baseline
remained mid-1995.
Table 4-2 provides a summary of applicability and mill counts by subcategory. The
table shows each component of the joint rulemaking (MACT, BAT, PSES, and BMPs) and
illustrates that all components apply to mills in the bleached papergrade kraft and soda and
papergrade sulfite subcategories. Because EPA decided not to revise BPT and BCT, these
regulations have been removed from this analysis. Table 4-3 shows the applicability of each
component for each of the 97 mills currently included in EPA's database.
Additional details on applicability for each of the effluent guideline subcategories
included in the rulemaking are presented in the following subsections.
4-1
-------
Section 4 - Industry Profile
4.2 Bleached Papergrade Kraft and Soda Subcategory
The bleached papergrade kraft and soda mills subject to BAT, PSES, MACT, and
BMPs as well as specific mills that did not receive costs or loadings estimates are discussed
below.
Of the 86 bleached papergrade kraft and soda mills, 84 mills produce
bleached papergrade kraft products and 2 produce bleached papergrade
soda products.
MACT applies to these 86 mills.
One mill (James River, Camas, WA) produces both bleached papergrade
kraft and bleached papergrade sulfite products (and is counted in both
subcategories).
One mill (Stone Container Corp., Snowflake, AZ) did not receive costs and
loadings estimates even though BAT/PSES and BMPs apply. The mill has
announced it will cease bleached kraft production. The mill currently has a
functional bleach plant and is counted in the bleached kraft subcategory,
but is expected to be reclassified as unbleached kraft after promulgation.
Costs and loadings will be estimated in the phase II rulemaking for
unbleached kraft production.
One mill (Port Townsend Paper, Port Townsend, WA) does not have a
conventional bleach plant, and only brightens with peroxide and sodium
hydroxide. This mill was counted as bleached papergrade kraft at proposal
but has now been reclassified as unbleached kraft.
One mill (Longview Fibre, Longview, WA) shut down its chlorine-based
bleach plant in March 1994. In 1995 and through October 1996, a small
amount of semi-bleached pulp using one stage of peroxide bleaching was
processed. EPA did not estimate compliance costs and loadings reductions
for this mill. But, because the mill has a functional bleach plant, it has not
been reclassified as unbleached kraft. BAT/PSES and BMPs apply but no
costs or loadings were estimated, on the assumption that the mill would
cease bleached kraft production rather than invest in new bleaching
technology. Costs and loadings will be estimated in the phase II
rulemaking for unbleached kraft production.
Of the 77 mills to which BAT applies, 75 received BAT cost and loading
estimates. (As mentioned previously, cost and loading estimates were not
prepared for Stone Container, Snowflake, and Longview Fibre, Longview.)
4-2
-------
Section 4 - Industry Profile
PSES applies to 9 mills.
BMPs apply to 86 mills. BMPs apply to Stone Container, Snowflake and
Longview Fibre, Longview, but no BMP costs were estimated, based on
the assumption they have or will cease bleached kraft production.
4.3 Papergrade Sulfite Subcategory
The papergrade sulfite mills subject to BAT, PSES, MACT, and BMPs along with
the one mill that only received BMP costs are described below.
In the U.S., 11 mills currently produce papergrade sulfite products (the
papergrade sulfite subcategory covers mills with both bleached and
unbleached production).
One mill produces only unbleached sulfite products. BAT applies, but they
are assumed to have no cost. BMP costs were estimated.
One mill (James River, Camas, WA) produces both bleached papergrade
kraft and bleached papergrade sulfite products (and is counted in both
subcategories).
One of the mills is both a direct and an indirect discharger. However,
wastewater from its pulping and bleaching operations is discharged to a
POTW, so this mill is covered by PSES for the papergrade sulfite
subcategory, not BAT.
MACT applies to 11 mills.
Of the 10 mills to which BAT applies, 9 received cost and loadings
estimates.
PSES applies to 1 mill.
BMPs apply to 11 mills.
4.4 Trends in the Industry
The development of increasingly more advanced process technologies that
minimize the discharge of wastewater and wastewater pollutants is a critical step toward the
Clean Water Act's ultimate goal of eliminating the discharge of pollutants into the nation's
waters. EPA is interested in encouraging development of advanced technologies for broader
commercial applications (1). As these technologies become proven and their efficiencies
publicized, EPA hopes that they will become standard industry practice. Thus, EPA believes it is
4-3
-------
Section 4 - Industry Profile
in the public interest to encourage mills today to develop environmentally beneficial technology
and to provide incentives for mills that are innovative and forward-looking in their use of new
technologies that are more environmentally and cost effective despite their greater initial capital
cost.
4.5 References
1. Technical Support Document for the Voluntary Advanced Technology Incentives
Program. EPA, Washington DC, Record Section 22.8, DCN 14488.
4-4
-------
Section 4 - Industry Profile
Table 4-1
Mills Omitted from EPA's Mid-1995 BAT/PSES Cost and Loadings Estimates
Subcat.
B
B
E
C
B
Company
Stone Container Corp.
Simpson Paper Co.
Great Northern Paper
Port Townsend Paper
Longview Fibre Co.
Location
Snowflake, AZ
Fairhaven, CA
Millinocket, ME
Port Townsend, WA
Longview, WA
Reason for Omission
Mill is expected to cease bleached pulp production
before Cluster Rules take effect. Removed from
costs and loadings estimates.
Closed in March 1993
Unbleached sulfite (BMP costs only)
No bleach plant but was misclassified as BPK at
proposal
Chlorine-based bleaching curtailed March 1994
Table 4-2
Mid-1995 Applicability and Mill Counts by Subcategory
Effluent Subcategory
Bleached Papergrade Kraft and Soda
Papergrade Sulfite
Number of Mills Affected8
Number of
Mills in this
Subcategory
86
11
96
Clean Air
Act
MACT
86
11
96
Clean Water Act
BAT
77
10
86
PSES
9
1
10
BMP
86
11
96
aOne mill has production in both subcategories so total mills affected is one less than the sum of the mills in the two
subcategories.
4-5
-------
Section 9 - Title
Table 4-3
Applicability by mill for the Bleached Papergrade Kraft and Soda and Papergrade Sulfite Facilities
Company
Alabama Pine Pulp
Alabama River Pulp Co. Inc.
Appleton Papers Inc.
Badger Paper Mills Inc.
Boise Cascade Corp.
Boise Cascade Corp.
Boise Cascade Corp.
Boise Cascade Corp.
Boise Cascade Corp.
Bowater Inc.
Bowater Inc.
Champion International Corp.
Champion International Corp.
Champion International Corp.
Champion International Corp.
Champion International Corp.
Champion International Corp.
Chesapeake Paper Products Co.
Consolidated Papers Co.
Container Corp. of America
(Jefferson Smurfit)
Mill
Claiborne (Perdue Hill)
Claiborne (Perdue Hill)
Roaring Spring
Peshtigo
Deridder
International Falls
Jackson
St. Helens
Wallula
Calhoun
Catawba
Canton
Cantonment (Pensacola)
Courtland
Houston (Sheldon)
Lufkin
Quinnesec (Norway)
West Point
Wisconsin Rapids
Brewton
State
AL
AL
PA
WI
LA
MN
AL
OR
WA
TN
SC
NC
FL
AL
TX
TX
MI
VA
WI
AL
Subcat
B
B
B
E
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
Discharge
Status
Direct
Direct
Direct
Direct,
Indirect
Direct
Direct
Direct
Indirect
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
MACT
'
'
'
'
'
'
'
1
BAT
'
'
'
'
'
'
'
1
PSES
BMP
'
'
'
'
'
'
'
1
Comments
Recently shut
down its pulping
process.
-------
Table 4-3 (Continued)
Section 9 - Title
Company
Federal Paper Board Co.
(International Paper)
Federal Paper Board Co.
(International Paper)
Finch Pruyn & Co Inc.
Fraser Paper (Cross Pointe)
Georgia-Pacific Corp.
Georgia-Pacific Corp.
Georgia-Pacific Corp.
Georgia-Pacific Corp.
Georgia-Pacific Corp. (Nekoosa)
Georgia-Pacific Corp. (Leaf River)
Georgia-Pacific Corp.
Georgia-Pacific Corp. (Nekoosa)
Georgia-Pacific Corp.
Georgia-Pacific Corp.
Oilman Paper Co.
Great Northern Paper Co.
Gulf States Paper Corp.
International Paper Co.
International Paper Co.
International Paper Co.
International Paper Co. (And'scogn)
International Paper Co.
International Paper Co.
Mill
Augusta
Riegelwood
Glens Falls
Park Falls
Ashdown
Bellingham
Brunswick
Crossett
Nekoosa
New Augusta
Palatka
Port Edwards
Woodland
Zachary (Port Hudson)
St. Marys
Millinocket
Demopolis
Bastrop
Erie
Georgetown
Jay
Mobile
Moss Point
State
GA
NC
NY
WI
AR
WA
GA
AR
WI
MS
FL
WI
ME
LA
GA
ME
AL
LA
PA
SC
ME
AL
MS
Subcat
B
B
E
E
B
E
B
B
B
B
B
E
B
B
B
E
B
B
B
B
B
B
B
Discharge
Status
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Indirect
Direct
Direct
Direct
Indirect
MACT
1
'
'
'
'
'
'
'
'
'
'
'
'
'
'
'
'
'
BAT
1
'
'
'
'
'
'
'
'
'
'
'
^
^
^
'
PSES
^
^
BMP
1
^
^
^
^
'
^
^
^
^
^
^
^
^
^
'
^
^
Comments
Only unbleached
suffite.
-------
Table 4-3 (Continued)
Section 9 - Title
Company
International Paper
Co. (Hammermill)
International Paper Co. (Riverdale)
International Paper Co.
International Paper Co.
James River Corp. (Crown Paper
Co.)
James River II Inc
James River II Inc
James River Corp. (Wauna Mill)
James River Corp.
James River Corp. (Naheola Mill)
James River II Inc. (Crown Paper)
Kimberly Clark Corp.
Kimberly-Clark Corp.
Lincoln Pulp & Paper Co.
Longview Fibre
Louisiana-Pacific Corp.
Mead Corp.
Mead Corp.
Mead Corp.
P. H. Glatfelter Co.
Pope & Talbot Inc.
Mill
Pine Bluff
Selma
Texarkana
Ticonderoga
Berlin
Camas
Camas
Clatskanie
Old Town
Pennington
St. Francisville
Coosa Pines
Everett
Lincoln
Longview
Samoa
Chillicothe
Escanaba
Rumford
Spring Grove
Halsey
State
AR
AL
TX
NY
NH
WA
WA
OR
ME
AL
LA
AL
WA
ME
WA
CA
OH
MI
ME
PA
OR
Subcat
B
B
B
B
B
E
B
B
B
B
B
B
E
B
B
B
B
B
B
B
B
Discharge
Status
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
MACT
'
'
'
1
'
'
1
'
'
'
'
'
BAT
'
'
'
1
'
'
1
'
'
'
'
'
PSES
BMP
'
'
'
1
'
'
1
'
^
^
'
^
Comments
No costs
estimated.
TCP bleaching
process.
oo
-------
Table 4-3 (Continued)
Section 9 - Title
Company
Port Townsend Paper
Potlatch Corp.
Potlatch Corp.
Potlatch Corp.
Procter & Gamble Paper
S.D. Warren (SAPPI)
S.D. Warren (SAPPI)
S.D. Warren (SAPPI)
Scott Paper Co./SAPPI
Simpson Paper Co.
Simpson Paper Co.
Simpson Paper Co.
Simpson Tacoma Kraft Co.
St. Joe Forest Products Co.
Stone Container Corp.
Stone Container Corp.
Stone Container (Savannah River)
Stone Container Corp.
Temple Inland Forest Products
Union Camp Corp.
Mill
Port Townsend
Cloquet
Lewiston
McGehee
Mehoopany
Hinckley (Skowhegan)
Muskegon
Westbrook
Mobile
Anderson
Fairhaven
Pasadena
Tacoma
Port St. Joe
Missoula
Panama City
Pt. Wentworth
Snowflake
Evadale (Silsbee)
Eastover
State
WA
MN
ID
AR
PA
ME
MI
ME
AL
CA
CA
TX
WA
FL
MT
FL
GA
AZ
TX
SC
Subcat
B
B
B
B
E
B
B
B
B
B
B
B
B
B
B
B
B
B
B
B
Discharge
Status
Direct
Indirect
Direct
Direct
Direct
Direct
Indirect
Direct
Direct
Direct
Direct
Indirect
Direct
Indirect
Direct
Indirect
Direct
N
Direct
Direct
MACT
'
'
BAT
'
PSES
'
BMP
'
'
Comments
No bleach
plant/Reclassified
as unbleached
kraft.
Closed in March
1993.
No costs and
loadings
estimated.
-------
Table 4-3 (Continued)
Section 9 - Title
Company
Union Camp Corp.
Wausau Paper Mills Co.
Westvaco Corp.
Westvaco Corp.
Wesrvaco Corp.
Weyerhaeuser Paper Co.
Weyerhaeuser Paper Co.
Weyerhaeuser Paper Co.
Weyerhaeuser/Flint River Mill
Weyerhaeuser Paper Co.
Weyerhaeuser Paper Co.
Willamette Industries Inc.
Willamette Industries Inc.
Willamette (Penntech Papers Div.)
Willamette Industries Inc.
Mill
Franklin
Brokaw
Covington
Luke
Wickliffe
Columbus
Longview
New Bern
Oglethorpe
Plymouth
Rothschild
Bennetsville
Hawesville
Johnsonburg
Kingsport
State
VA
WI
VA
MD
KY
MS
WA
NC
GA
NC
WI
sc
KY
PA
TN
Subcat
B
E
B
B
B
B
B
B
B
B
E
B
B
B
B
Discharge
Status
Direct
Direct
Direct
Indirect
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
Direct
MACT
'
'
'
'
'
'
'
'
'
'
'
BAT
'
'
'
'
'
'
'
'
'
'
PSES
^
BMP
^
^
^
^
^
^
'
^
^
^
^
Comments
One OZ-ECF
SWD line.
NA - Not Applicable, no costs and loadings were determined for this mill.
Note: James River, Camas, WA has production in both subcategories.
-------
Section 5 - Subcategorization
SECTION 5
SUBCATEGORIZATION
5.1 Introduction
EPA proposed a new Subcategorization scheme to replace the former
Subcategorization scheme found in 40 CFR Part 430 (Pulp, Paper, and Paperboard Point Source
Category) and 40 CFR Part 431 (The Builder's Paper and Board Mills Point Source Category).
The new Subcategorization scheme consolidates into 12 subcategories what had once been 26
subcategories. EPA's reasons for combining and reorganizing the subcategories are described in
the proposal (see 58 FR 66098-66100) and in a document entitled "Selected Issues Concerning
Subcategorization" (1). EPA solicited comment on whether any specific subcategories proposed
should be divided into smaller subcategories and whether any specific subcategories proposed
should be combined to form larger subcategories. This section provides a description of the
industry Subcategorization in effect prior to the promulgation of this rule, describes EPA's
methodology in developing the proposed Subcategorization scheme, provides summaries of
comments received on the proposed Subcategorization scheme, provides EPA's assessment of the
Subcategorization comments, and presents EPA's final determinations on the Subcategorization
scheme.
5.2 Description of the Industry Subcategorization in Effect Prior to the
Promulgation of this Rule
Manufacturing processes and untreated wastewater characteristics (i.e., pollutant
loadings which varied somewhat by final product produced) were the principal factors used to
subcategorize the industry prior to this rulemaking. Data used to determine that
Subcategorization represented the state of the industry during the early-to-mid 1970s. At that
time, the overall level of wastewater treatment provided by the industry was not consistent
among mills with similar manufacturing processes. EPA concluded at that time that untreated
wastewater pollutant loadings provided a reasonable basis to subcategorize the industry
principally because the costs of compliance for mills with similar untreated wastewater pollutants
loadings to achieve uniform effluent levels were similar.
The Subcategorization in effect prior to the promulgation of this rule was:
40 CFR Part 430
Subpart A - Unbleached Kraft;
Subpart B - Semi-Chemical;
Subpart C - Reserved;
Subpart D - Unbleached Kraft-Neutral Sulfite Semi-Chemical (Cross
Recovery);
Subpart E - Paperboard from Wastepaper;
5-1
-------
Section 5 - Subcategorization
Subpart F - Dissolving Kraft;
Subpart G - Market Bleached Kraft;
Subpart H - Board, Coarse, and Tissue (BCT) Bleached Kraft;
Subpart I - Fine Bleached Kraft;
Subpart J - Papergrade Sulfite (Blow Pit Wash);
Subpart K - Dissolving Sulfite Pulp;
Subpart L - Groundwood-Chemi-Mechanical;
Subpart M - Groundwood-Thermo-Mechanical;
Subpart N - Groundwood-Coarse, Molded, and News (CMN) Papers;
Subpart O - Groundwood-Fine Papers;
Subpart P - Soda;
Subpart Q - Deink Secondary Fiber;
Subpart R - Non-Integrated-Fine Papers;
Subpart S - Non-Integrated-Tissue Papers;
Subpart T - Tissue from Wastepaper;
Subpart U - Papergrade Sulfite (Drum Wash);
Subpart V - Unbleached Kraft and Semi-Chemical;
Subpart W - Wastepaper-Molded Products;
Subpart X - Non-Integrated-Lightweight Papers;
Subpart Y - Non-Integrated-Filter and Non-Woven Papers;
Subpart Z - Non-Integrated-Paperboard; and
40 CFR Part 431
Subpart A - Builders' Paper and Roofing Felt.
5.3 Revised Industry Subcategorization
Since the early-to-mid 1970s, all but one of the direct discharging mills have
installed secondary wastewater treatment systems. End-of-pipe discharge data supplied in the
1990 Census Questionnaire for most mills show that the degree of end-of-pipe wastewater
treatment provided by the industry is much more uniform than it was during the 1970s. In
consideration of the factors in CWA Section 304(b), EPA has determined that the
Subcategorization analysis for this pulp, paper, and paperboard industry is more appropriately
conducted based on manufacturing processes employed and engineering aspects of the
application of various types of control techniques rather than raw waste loads. EPA believes that
these factors more accurately represent a mill's ability to comply with effluent limitations
guidelines and standards and achieve pollutant reductions.
As discussed in Section 5.3.1 of the Proposed Technical Development Document,
the pulp, paper, and paperboard industry can be classified by major production processes. These
production processes and the applicable former subcategories are listed below.
5-2
-------
Section 5 - Subcategorization
Integrated Pulp and Paper Mills
Chemical Pulp Mills
Kraft and Soda Mills
— Dissolving Kraft (Subpart F)
— Bleached Papergrade Kraft (Subparts G, H, I)
— Bleached Papergrade Soda (Subpart P)
— Unbleached Kraft (Subparts A, D, V)
Sulfite Mills
— Dissolving Sulfite (Subpart K)
— Papergrade Sulfite (Subparts J and U)
Non-Wood Fiber Pulp Mills
Semi-Chemical Pulp Mills (Subparts B, D, V)
Mechanical Pulp Mills
Stone Groundwood (Subparts N and O)
Refiner
Thermo-Mechanical (Subpart M)
Chemi-Mechanical (Subpart L)
Chemi-Thermo-Mechanical
Secondary Fiber Mills
Deink Mills (Subpart Q)
Non-Deink Mills (Subparts E, T, W, and Part 431 Subpart A)
Non-Integrated Paper Mills (Subparts R, S, X, Y, Z) (producers of products
from purchased pulp)
The classification of the industry by major production processes addresses many
of the statutory factors set forth in CWA Section 304(b), including manufacturing processes and
equipment (e.g., chemical, mechanical, and secondary fiber pulping; pulp bleaching; paper
making); raw materials (e.g., wood, secondary fiber, non-wood fiber, purchased pulp); products
manufactured (e.g., unbleached pulp, bleached pulp, finished paper products); and, to a large
extent, untreated and treated wastewater characteristics (e.g., BOD5 loadings, presence of toxic
chlorinated compounds from pulp bleaching) and process water usage and discharge rates. EPA
determined that other factors such as size, age, and geographical locations were not significant
factors to explain the technical feasibility or economic achievability of effluent limitations
guidelines and standards for this industry. As a result, the Agency used the production process
classifications described above as a starting point for reviewing Subcategorization.
Most manufacturing processes at pulp, paper, and paperboard mills generate
wastewaters that contain substantial quantities of the conventional pollutants BOD 5 and TSS.
5-3
-------
Section 5 - Subcategorization
Furthermore, efficient BOD5 removal is a principal design objective for pulp, paper, and
paperboard mill wastewater treatment systems. For these reasons, BOD5 and TSS are important
measures of pollution generation and wastewater treatability for the pulp and paper industry.
Although EPA is making no changes to BPT and BCT conventional pollutant limitations
previously promulgated for any subcategories at this time, EPA is revising NSPS for the
conventional pollutants BOD5 and TSS for the Bleached Papergrade Kraft and Soda Subcategory.
In addition, effective secondary biological treatment is a component of the revised BAT and
PSES technology bases for the Bleached Papergrade Kraft and Soda and the Papergrade Sulfite
Subcategories. Therefore, EPA has determined that BOD5 loadings are an important component
of the Agency's Subcategorization analysis.
EPA examined the status of the industry with respect to treatment of BOD 5 to
determine if the former subcategories adequately represent current industry characteristics. The
Agency determined that, based upon the present status of the industry, many of the former
subcategories are no longer necessary because mills with similar production processes have, at
reasonable costs, achieved similar production normalized effluent quality, notwithstanding
differences in untreated wastewater pollutant loadings. Accordingly, EPA used effluent quality,
in terms of final effluent production normalized BOD5 load, as a basis to further subcategorize
the industry beyond the major process classifications set forth above.
Using the methodology described in Section 5.0 of the Technical Development
Document for the proposed rule (2), the Agency compared average and range of production-
normalized final effluent BOD5 loadings of the mills selected to represent each former
subcategory. Subcategories with similar process technologies were compared. Based upon these
comparisons, the Agency determined that several former subcategories exhibited similar treated
wastewater characteristics, and that these subcategories might be grouped or combined into
revised subcategories. EPA's revised Subcategorization scheme is listed below (see also Table 1-
1).
40 CFR Part 430
Subpart A - Dissolving Kraft;
Subpart B - Bleached Papergrade Kraft and Soda;
Subpart C - Unbleached Kraft;
Subpart D - Dissolving Sulfite;
Subpart E - Papergrade Sulfite;
Subpart F - Semi-Chemical;
Subpart G - Mechanical Pulp;
Subpart H - Non-Wood Chemical Pulp;
Subpart I - Secondary Fiber Deink;
Subpart J - Secondary Fiber Non-Deink;
Subpart K - Fine and Lightweight Papers from Purchased Pulp;
Subpart L - Tissue, Filter, Non-Woven, and Paperboard from Purchased
Pulp.
5-4
-------
Section 5 - Subcategorization
5.3.1 All Mills
Some commenters on the proposed rule indicated that EPA should retain the
former Subcategorization scheme, which was based on an evaluation of raw waste loads and
flows. One commenter stated that if EPA insists on revising the subcategories, the Agency must
also evaluate raw waste loads, product type, fiber furnish, and raw waste treatability in addition
to effluent loads.
EPA has determined that the groupings in Subparts A through L (excluding
Subparts B and E discussed in Sections 5.3.2 and 5.3.3, respectively) are appropriate since they
are comprised of mills using similar processes and attaining similar effluent quality. EPA will
consider evaluating the relationship between raw waste loads and treated effluent waste loads
along with other factors suggested by commenters that may affect the reasonableness of the
groupings when EPA determines whether new or revised effluent limitations guidelines and
standards are appropriate for these subcategories. At that time, EPA would likely consider in
addition to other factors identified in CWA Section 304(b):
Specific processes, including papermaking processes;
Products;
Fiber furnish;
Grade changes;
Chemical usage;
Non-fibrous material (fillers, additives, coatings, etc.);
Shrinkage;
Raw material (fiber) quality;
Product quality/requirements;
Water requirements;
Pollutant loadings; and
Effluent quality.
If after further analysis EPA determines that certain types of mills within a
subcategory cannot achieve the same effluent quality without undue economic impact, EPA will
consider further segmenting the subcategory as appropriate to better respond to material
differences between facilities. In the interim, the Subcategorization scheme for these
subcategories is simply a redesignation of the old subcategories into the new subcategories. The
limitations and standards promulgated under the old Subcategorization scheme are recodified
under the new Subcategorization scheme in the form of segments corresponding to the old
subparts. (In recodifying these limitations and standards, EPA has not changed the substance of
the existing regulations.)
5.3.2 Subpart B - Bleached Papergrade Kraft and Soda Subcategory
Commenters on the proposed rule indicated that EPA should retain the existing
Subcategorization scheme which was based on an evaluation of raw waste loads and flows. One
5-5
-------
Section 5 - Subcategorization
commenter further stated that if EPA insists on revising the subcategories, raw waste loads,
product type, fiber furnish, and raw waste treatability must also be evaluated in addition to
effluent loads.
In response, for the Bleached Papergrade Kraft and Soda Subcategory, EPA
undertook an analysis of the relationship between conventional pollutant loading (BOD5 and
TSS) in raw wastewater and treated effluent. No soda mills were included in the analysis
because none had more than 85 percent of their final production in the subcategory. (EPA
selected a final production cut-off of 85 percent within a single subcategory for the wastewater
from the mill to be considered representative of that subcategory's wastewater. This approach is
described in the Technical Development Document for the proposed rule (2) and in "Selected
Issues Concerning Development of Conventional Pollutant Control Options" (3).) Specific
analyses performed are documented and described in "Analysis of the Relationship Between
Conventional Pollutant Loadings in Raw Wastewater and in Treated Effluent at Papergrade Kraft
Mills" (4). The findings of this analysis include:
The distribution of final effluent and raw wastewater BOD5 and TSS loads
support EPA's consolidation of the four current papergrade kraft and soda
subcategories into a single subcategory;
Final effluent conventional pollutant loads are not dependent on raw
wastewater conventional pollutant loads;
Final effluent conventional pollutant loads are dependent on treatment
system removal efficiencies; and
The data support EPA's assumption and demonstrate that the long-term
average BOD5 and TSS performance levels are achievable, regardless of
raw wastewater loads.
The purpose of industry Subcategorization is to provide a mechanism for
addressing variations among raw materials, processes, products, and other parameters that can
result in distinct effluent characteristics. Regulation of a category by subcategory ensures that
each subcategory has a uniform set of effluent limitations that take into account technical
achievability and economic impacts unique to that subcategory. EPA considered the processes,
raw materials, wastewater treatability, and other factors unique to bleached papergrade kraft and
soda mills as the basis for combining bleached papergrade kraft and soda mills in a unique
subcategory separate from other pulping processes (e.g., papergrade sulfite and dissolving kraft
and sulfite processes).
EPA evaluated whether additional subcategory segmentation was appropriate for
development of effluent limitations guidelines and standards for toxic and nonconventional
pollutants based primarily on process changes (e.g., BAT and PSES). EPA considered further
segmentation of the Bleached Papergrade Kraft and Soda Subcategory by product brightness
5-6
-------
Section 5 - Subcategorization
(high versus low) and fiber furnish (hardwood versus softwood) because these factors may
influence the technical feasibility of bleaching process technologies. EPA decided not to further
segment the Bleached Papergrade Kraft and Soda Subcategory by product brightness because
available data do not demonstrate that variation in this parameter results in significant differences
in effluent characteristics. Moreover, at some mills, final product characteristics vary sufficiently
(on a day-to-day basis) to make permitting and compliance impracticable. EPA was also
concerned that mills might strive for higher brightness than their product required in order to
qualify for less stringent limits, with the unintended result of having mills use more rather than
less bleaching chemicals and hence discharging more pollution than they otherwise would.
EPA also considered fiber furnish (hardwood versus softwood) as a basis for
further segmenting the Bleached Papergrade Kraft and Soda Subcategory. For toxic and
nonconventional pollutants with compliance points at the bleach plant, EPA found no difference
in achievability of limitations due to fiber furnish. However, for the bulk parameter adsorbable
organic halides (AOX) with compliance point at the end-of-pipe, EPA found higher effluent
loadings for softwood mills than for hardwood mills. EPA set limitations for AOX based on data
for softwood mills to ensure that the limitations would be achievable for all furnishes. Since
many mills pulp both hardwood in variable combinations, or "swing" between hardwood and
softwood, effluent limitations based on fiber furnish would be very difficult to administer. For
this reason, EPA found it unnecessary and inappropriate to further segment the Bleached
Papergrade Kraft and Soda Subcategory by fiber furnish.
5.3.3 Subpart E - Papergrade Sulfite Subcategory
Several comments were submitted concerning the feasibility of "totally chlorine-
free" (TCF) technology-based limits for certain sulfite pulping processes and products.
Specifically, comments indicated that TCF bleaching processes are not technically feasible for
manufacture of ammonium-based papergrade sulfite pulp from softwood and for manufacture of
specialty papergrade sulfite products such as pulps for photographic papers and plastic molding
products.
After reviewing the comments, EPA concurs that additional segmentation of the
Papergrade Sulfite Subcategory is necessary to better reflect product considerations, the variation
of manufacturing processes, and the demonstration of pollution prevention process changes
within the category for the purpose of establishing BAT, PSES, NSPS, and PSNS. See
"Segmenting the Papergrade Sulfide Subcategory" (5) for additional information concerning
EPA's rationale for segmenting the Papergrade Sulfite Subcategory. The segments for the
Papergrade Sulfite Subcategory are:
(a) Production of pulp and paper at papergrade sulfite mills that use an acidic
cooking liquor of calcium, magnesium, or sodium sulfite unless those
mills are specialty-grade sulfite mills.
5-7
-------
Section 5 - Subcategorization
(b) Production of pulp and paper at papergrade sulfite mills that use an acidic
cooking liquor of ammonium sulfite, unless those mills are specialty-grade
sulfite mills.
(c) Production of pulp and paper at specialty-grade sulfite mills. Specialty-
grade sulfite mills are those mills where (1) 25 percent or more of
production is characterized by pulp with a high percentage of alpha
cellulose and high brightness sufficient to produce end products such as
plastic molding compounds, saturating and laminating products, and
photographic papers; or (2) those mills where 50 percent or more of
production is 91 ISO (International Organization for Standardization) units
brightness and above.
EPA is not revising NSPS for conventional pollutants for the Papergrade Sulfite
Subcategory. Because the NSPS for conventional pollutants for former Subparts J and U were
the same, EPA has recodified these standards in a single table for Subpart E (without
distinguishing between the former subparts in the form of segments).
5.4 References
1. Comment Response Document, Volume I, "Selected Issues Concerning
Subcategorization." U.S. Environmental Protection Agency, Washington DC,
Record Section 30.11, DCN 14497, 1997.
2. Development Document for Proposed Effluent Limitations Guidelines and
Standards for the Pulp, Paper, and Paperboard Point Source Category. EPA-821-
R-93-019, U.S. Environmental Protection Agency, Washington DC, October
1993.
3. Comment Response Document, Volume I, "Selected Issues Concerning
Development of Conventional Pollutant Control Options." U.S. Environmental
Protection Agency, Washington DC, Record Section 30.11, DCN 14497, 1997.
4. Radian Corporation. Analysis of the Relationship Between Conventional
Pollutant Loadings in Raw Wastewater and in Treated Effluent at Papergrade
Kraft Mills. Herndon, Virginia, Record Section 22.1, DCN 14039, November 16,
1995.
5. Comment Response Document, Volume I, "Segmenting the Papergrade Sulfite
Subcategory." U.S. Environmental Protection Agency, Washington DC, Record
Section 30.11, DCN 14497, 1997.
5-8
-------
Section 6 - Water Use and Wastewater Characteristics
SECTION 6
WATER USE AND WASTEWATER CHARACTERISTICS
6.1 Introduction
Section 6.0 of the TDD describes water use and wastewater recycle practices, and
the general characteristics of wastewater at mills that manufacture pulp, paper, and paperboard in
the U.S. This information was gathered in EPA's 1990 National Census Questionnaire of the
industry. All pulp and papermaking processes use water; in fact, the pulp and paper industry is
the largest industrial process water user in the U.S. Except as noted below, EPA believes that the
information presented in Section 6.0 of the TDD is still representative of the industry today. This
section discusses specific topics related to water use.
Water use in the industry decreased approximately 30 percent between 1975 and
1990, reflecting significant effort by the industry to reduce consumption and increase wastewater
reuse and recycle (1). EPA believes that this trend has continued because mill modernizations
have occurred and continue to occur. These projects generally include the installation of
equipment that uses water more efficiently. The total effluent flow from an integrated bleached
kraft mill is normally between about 50 to 150 m3/kkg of pulp produced, although a few mills
discharge significantly lower or higher flows (2). In 1995, the average U.S. bleached kraft and
soda mill discharged approximately 95 m3/kkg of pulp (3).
6.2 Mill Water Use - Bleach Plant Effluent Portion of Final Effluent
The final rule preamble and Section 8.2.3 of this document set forth EPA's
reasons for regulating some pollutants at the bleach plant effluent rather than at the final effluent
discharge to receiving waters. One reason is that most of the regulated chlorinated compounds
are not present at detectable concentrations in the final effluent, and may or may not be present at
detectable concentrations in the bleach plant effluent. Because the bleach plant effluent flow rate
is a fraction of the influent to treatment flow rate (which is assumed to be the same as the final
effluent flow rate for this analysis), EPA is concerned that the dilution of the bleach plant
effluent in the final effluent will make it more difficult to detect the regulated chlorinated
compounds in the final effluent. This section describes the fraction of the final effluent that is
composed of bleach plant effluent.
The fraction of the final effluent that is composed of bleach plant effluent is
described in various references as shown below. Although the figures listed in each reference
differ, they agree that the final effluent is composed of a significant fraction of bleach plant
effluent as well as other wastewaters from other areas of the mill. The data are believed to be
similar for bleached papergrade kraft mills and for papergrade sulfite mills.
6-1
-------
Section 6 - Water Use and Wastewater Characteristics
Source
EPA Sampling Database (derived from bleached papergrade kraft mills)
(4)
Technical Development Document, (1990 Industry Census
Questionnaire) (5)
Average of Bleached Papergrade Kraft Mills
Average of Papergrade Sulfite Mills
Water Use Reduction in the Pulp and Paper Industry (6)
Pulp Bleaching - Principles and Practice (7)
Bleach Plant Effluent
Flow Percentage
of Final Effluent Flow
7% to 43%
30 to 50%
30%
35%
50%
50 to 70%
The figures derived from the EPA sampling database are discussed in more detail
below but the other references provide little more information than is listed in the table. The
EPA sampling database contains information from mills sampled by EPA, and information
supplied by NCASI and individual mills.
The mills in the EPA sampling database were divided into two groups: mills that
make market pulp only, and mills that make pulp and paper. The amount of bleach plant effluent
in the final effluent for market pulp mills ranged from 21 to 43 percent (among nine mills). For
the pulp and paper mills the range was 7 to 38 percent (among nine mills). An integrated mill
has more wastewater sources, mainly from the paper making operations, so the bleach plant
effluent proportion is smaller. The lowest values in each group (7 and 21 percent) were reported
by mills that make unbleached pulp in addition to bleached pulp. The mills included in this
analysis used a variety of bleaching sequences. The values for the nine mills using elemental
chlorine-free bleaching sequences to make only bleached pulp were 30 to 43 percent.
6.3 Mill Water Use For Option A and Option B Mills
Of the mills included in the EPA sampling database, mills using extended cooking
and/or oxygen delignification (Option B technologies) generally discharged less wastewater than
mills using conventional cooking (Option A technology). In addition, those mills using totally
chlorine-free bleaching discharged less wastewater than the mills using elemental chlorine-free
bleaching sequences, because in chlorine-free bleaching processes, the bleach plant filtrates are
returned to the recovery system. Several reasons exist for these differences. The following table
summarizes the average production-normalized bleach plant flow rates that EPA calculated using
its sampling database. The following production-normalized bleach plant flow rates were used to
calculate some of the pollutant loadings and reductions described in Section 9.0.
6-2
-------
Section 6 - Water Use and Wastewater Characteristics
Production-Normalized Kraft Bleach Plant Flow Rates"
TypeofMill11
Mills Without EC or OD
Mills With EC and/or OD
TCP Mills
Hardwood Lines
Average
(m3/kkg)
24.7
19.7
11.6
No. Of Lines
12
4
1
Softwood Lines
Average
(m3/kkg)
37.1
24.7
18.3
No. Of
Lines
13
12
2
The average flow rates presented in this table were derived from bleached papergrade kraft mills.
bEC = extended cooking, OD = oxygen delignification, TCP = totally chlorine-free bleaching
Although retrofitting an oxygen delignification system (which would be a
practical necessity for compliance with Option B) has no direct effect on effluent flows by itself,
the data presented above indicate that mills with extended cooking and/or oxygen delignification
have lower bleach plant flows than mills with conventional pulping. Several possible reasons
exist for the lower flow. Some mills have reported reductions in effluent flow due to oxygen
delignification projects because it is normal practice to close the screen room process and return
these filtrates to the recovery cycle when oxygen delignification is installed. In some cases, when
unbleached pulp kappa number-into-bleaching is reduced, it is possible to retire one or two
complete bleaching stages (e.g., convert a C/DEoDED bleach plant to OD-DEopD). Such action
could reduce effluent flows by about 15 m3/kkg pulp. In rare cases, oxygen delignification will
result in some water conservation if lower unbleached pulp kappa number-into-bleaching allows
the use of reduced wash water flow in the first bleaching stage.
The following discussion explains several reasons why EPA expects mills to
continue to reduce effluent flow rates. Other changes in effluent flow rates as a result of the
promulgated regulations are discussed in Section 11.3.
Water use in the industry decreased approximately 30 percent between 1975 and
1990 (1), reflecting significant effort by the industry to reduce consumption and increase
wastewater reuse and recycle. These reductions may have resulted from specifically planned
water conservation projects or they may have been secondary benefits of other mill
modernization projects. During mill renovation, new equipment is not installed in isolation.
Instead, it is common practice to modernize the entire mill area involved with the new
equipment, at least to some extent. Modern equipment is generally designed to conserve water
more effectively than older designs. Many details can be involved, such as the replacement of
packing on shafts with modern mechanical seals that use little or no water, or reduction in
cooling water requirements by more efficient design and increased use of cooling towers with
subsequent recycle. These modifications will generally reduce effluent discharges, but it is
difficult to provide realistic numeric estimates.
6-3
-------
Section 6 - Water Use and Wastewater Characteristics
Two elements of the two BAT options that will reduce effluent flows directly are
closing screen rooms and implementing BMPs. The application of current engineering practices
to the design of new systems and equipment will result in conservation of water. The greatest
improvements are likely to be seen in mills currently using relatively high quantities of water.
The kappa number of unbleached pulp entering the bleach plant is reduced by
employing two types of extended delignification, extended cooking (EC) and oxygen
delignification (OD). Mills that have recently installed EC and/or OD have also made related
improvements. For example, new or upgraded washers use water more efficiently and may
therefore generate less wastewater.
When upgrading the first chlorine/chlorine dioxide stage to high or 100 percent
chlorine dioxide substitution for chlorine, it is common to convert from low consistency
operation to medium consistency, or increase the use of recycled bleach filtrates for pulp dilution
to raise the temperature without incurring the cost of direct steam heating. These changes can
lead to an effluent reduction of about 12 m3/kkg pulp in softwood mills and 5 m3/kkg in
hardwood mills. Such improvements are most likely to be made in mills which have high
effluent flows.
6.4 Definition of Process Wastewater
The effluent limitations guidelines and standards for the pulp and paper industry
are applicable to discharges of process wastewaters directly associated with the manufacturing of
pulp and paper. In 1993, EPA proposed a definition of process wastewater as any water which
during manufacturing or processing comes into direct contact with or results from the production
or use of any raw material, intermediate product, finished product, byproduct, or waste product.
The proposed definition also specifically included boiler blowdown, wastewaters from water
treatment and other utility operations, blowdown from high rate (e.g., greater than 98 percent)
recycled non-contact cooling water systems to the extent they are mixed and co-treated with other
process wastewaters, and stormwaters from the immediate process areas to the extent they are
mixed and co-treated with other process wastewaters. The proposed definition specifically stated
that contaminated groundwaters from on-site or off-site groundwater remediation projects are not
considered process wastewaters. Separate permitting was proposed to be required for the
discharge of such groundwaters.
The proposed definition also specifically excluded certain process materials from
the definition of process wastewater. These process materials included: green liquor at any
liquor solids level, white liquor at any liquor solids level, black liquor at any liquor solids level
resulting from processing knots and screen rejects, black liquor after any degree of concentration
in the kraft or soda chemical recovery process, reconstituted sulfite and semi-chemical pulping
liquors prior to use, any pulping liquor at any liquor solids level resulting from spills or
intentional diversions from the process, lime mud and magnesium oxide, pulp stock, bleach
chemical solutions prior to use, and paper making additives prior to use (e.g., alum, starch and
size, clays and coatings). Because these materials were excluded from the proposed definition of
6-4
-------
Section 6 - Water Use and Wastewater Characteristics
process wastewater, they would have been prohibited from discharge into POTWs or waters of
the United States without a National Pollutant Discharge Elimination System (NPDES) permit
and effluent limitation or other authorization.
In response to the comments opposing the exclusion of these process materials
(8), EPA revised the proposed definition of process wastewaters to eliminate the exclusion of the
named process materials (40 CFR Part 430.01(m)). As the commenters contended, the proposed
language would have effectively required "closed-cycle" mills, which was not EPA's intent. The
proposed language was intended to prevent discharge of process materials during clean up in
preparation for permanent mill closure. EPA's definition continues to specify that process
wastewater is generated "during manufacturing or processing." Thus, the definition of process
wastewater as promulgated today does not allow for discharges from a mill that is not engaged in
manufacturing. Any mill wishing authorization to discharge in this manner must obtain
authorization in an NPDES permit or individual control mechanism administered by a POTW.
The distinction that process wastewater is generated "during manufacturing or
processing" should not be taken to exclude wastewaters generated during routine maintenance,
including maintenance during a scheduled temporary mill shut-down. Maintenance wastewaters
were not explicitly excluded from the definition of process wastewater at proposal, nor are they
excluded from the definition finally promulgated. Wastewaters generated during routine
maintenance are a result of pulp manufacturing processes and as such are included in the
definition of process wastewater. Many mills commingle leachates from landfills receiving
wastes associated with the processing or manufacturing operations. Therefore, EPA also has
included these leachate wastewaters in the definition of process wastewaters.
6.5 Use of Biocides
The existing BAT regulations for Subparts G, H, I, J, P, and U (now the Bleached
Papergrade Kraft and Soda and Papergrade Sulfite subcategories) establish effluent limitations
guidelines and standards for pentachlorophenol and trichlorophenol when used as biocides. One
way for dischargers to comply with these limitations is to certify that they do not use these
compounds as biocides. Using the data collected in the 1990 National Census Questionnaire,
EPA compared the names of biocide products reported as used by pulp mills in the Bleached
Papergrade Kraft and Soda and Papergrade Sulfite subcategories with readily available
information from the Office of Pesticide Programs to determine if these products contain
pentachlorophenol or trichlorophenol. EPA was unable to identify any biocide products used by
those pulp mills that contain these chemicals. Therefore, EPA expects that mills in these
subcategories will be able to comply with these limitations by certifying that they do not use
these compounds as biocides.
6-5
-------
Section 6 - Water Use and Wastewater Characteristics
6.6 References
1. Miner, R. and J. Unwin. "Progress in Reducing Water Use and Wastewater Loads
in the U.S. Paper Industry." TAPPI Journal. 74(8): 127-131, August 1991.
2. Mannisto, H., E. Mannisto, and M. Krogerus. Proceedings of the Minimum
Effluent Symposium. TAPPI Press, Atlanta, 1996.
3. Bryant, P., E.W. Malcolm, C.P. Woitkovich. "Pulp and Paper Mill Water Use in
North America" In: TAPPI International Environmental Conference, 1996.
4. EPA Sampling Database (Electronic File). Record Section 21.10, DCN 14499,
1997.
5. Development Document for Proposed Effluent Limitations Guidelines and
Standards for the Pulp, Paper, and Paperboard Point Source Category. EPA-821-
R-93-019, U.S. EPA, Washington DC, October 1993, p. 6-6.
6. H. A. Simons, NLK Consultants, and Sandwell Inc. "Water Use in the Pulp and
Paper Industry" for the Canadian Pulp and Paper Association, 1994, p. 60.
7. Histed, J., N. McCubbin, P.L. Gleadow. "Water Reuse and Recycle" in Pulp
Bleaching - Principles and Practice. C.W. Dence and D.W. Reeve, eds., 1996, p.
629.
8. Comments Submitted on Behalf of the American Forest and Paper Association.
Volume II, Section 13.1, pg 151-161, Record Section 19.1, DCN 20026, 1994.
6-6
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
SECTION 7
POLLUTION PREVENTION AND WASTEWATER TREATMENT TECHNOLOGIES
7.1 Introduction
This section describes technologies that are in use at pulp, paper, and paperboard
mills to prevent the formation of wastewater pollutants or reduce the discharge of wastewater
pollutants. Various combinations of these technologies were considered as the basis for the
effluent limitations guidelines and standards for the industry.
Two major approaches may be used to improve effluent quality at pulp, paper, and
paperboard mills: (a) in-process technology changes and controls to prevent or reduce the
formation of wastewater pollutants of concern, and (b) end-of-pipe wastewater treatment
technologies to remove pollutants from process wastewaters prior to discharge.
The Agency has defined pollution prevention as source reduction and other
practices that reduce or eliminate the formation of pollutants. Source reduction includes any
practice that reduces the amount of any hazardous substance or pollutant entering any waste
stream or otherwise released into the environment, or any practice that reduces the hazards to
public health and the environment associated with the release of such pollutants. Such practices
may include equipment or technology modifications; process or procedure modifications;
reformulation or redesign of products; substitution of raw materials; and improvements in
housekeeping, maintenance, training, and inventory control. Other pollution prevention practices
include increased efficiency in the use of raw materials, energy, water, and other resources.
The Agency has developed a model pollution prevention plan for the pulp and
paper industry as part of the Agency's effort to encourage pollution prevention programs in U.S.
industries. The model plan is discussed in a series of reports:
Pollution Prevention Opportunity Assessment and Implementation Plan for
Simpson Tacoma Kraft Company, Tacoma, Washington (1);
Model Pollution Prevention Plan for the Kraft Segment of the Pulp and
Paper Industry (2); and
Pollution Prevention for the Kraft Pulp and Paper Industry, Bibliography
(3).
Pollution-preventing process changes may be implemented in the pulping,
bleaching, chemical recovery, and papermaking areas of a mill. Many of the in-process controls
that prevent or reduce wastewater pollution also result in improved product quality and/or fiber
yield, as well as reduced operating costs through more efficient use of process materials and
prevention and control of leaks and spills of spent pulping liquor, soap, and turpentine. Sections
7-1
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
7.2 and 7.3 describe applicable pollution prevention controls and technologies for the industry.
These sections also provide information on the performance of each technology and the number
of mills using each technology.
Additional information on pollution prevention technologies for the pulp and paper
industry is available in these EPA documents:
Summary of Technologies for the Control and Reduction of Chlorinated
Organics from the Bleached Chemical Pulping Subcategories of the Pulp
and Paper Industry (4);
Pollution Prevention Technologies for the Bleached Kraft Segment of the
U.S. Pulp and Paper Industry (5); and
Technical Development Document for the Pulp, Paper, and Paperboard
Category Effluent Limitations Guidelines, Pretreatment Standards, and
New Source Performance Standards (6).
End-of-pipe wastewater treatment includes physical, chemical, and biological
processes that remove pollutants from mill effluent prior to discharge to a receiving stream or
POTW. Section 7.4 describes end-of-pipe wastewater treatment technologies applicable to the
industry.
7.2 Pollution Prevention Controls Used in Pulping and Delignification Processes
This section describes applicable technologies for reducing and preventing
pollutant discharges from the pulping area of a chemical pulp mill. Pulping area processes include
chipping, cooking, pulp washing, and screening. Pollution prevention technologies applicable to
pulping area processes include chip quality control, use of dioxin precursor-free defoamers and
pitch dispersants, extended cooking, effective brown stock washing, closing the screen room,
oxygen delignification, steam stripping of condensates, and spent pulping liquor spill prevention
and control.
7.2.1 Chip Quality Control
Chip thickness control is an important component of improving yield, reducing
bleaching chemical requirements, and optimizing pulp quality. Chip thickness can be controlled
by close control of chipping equipment tolerances or the use of chip thickness screens.
In preparation for chemical pulping, wood is reduced to chips approximately 2 to 5
millimeters thick and 10 to 30 millimeters long. Some mills use chips obtained from an off-site
source such as a sawmill, although most mills perform at least some chipping on site. Several
chipper designs are in use today, but the most common is the flywheel-type disc chipper, in which
7-2
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
logs are fed to one side of the disc at a pre-determined angle through a vertical directing chute.
After chipping, the chips pass a set of vibratory screens to separate chips of
acceptable length and width from fines and oversized pieces. Oversized chips are rechipped, and
fines are usually burned in an on-site hog fuel boiler.
Good quality chipping and screening equipment provides a uniform supply of chips
to the digester, leading to more uniform cooking and reduced digester chemical consumption.
The uniform pulp produced results in less variation in bleach plant feed and better control of the
bleaching process, which reduces overbleaching of the pulp to remove shives and colored fiber.
The effluent quality from the bleach plant is improved because when overbleaching is avoided,
lower levels of chlorinated organics are formed. A more uniform pulp also reduces the amount of
rejects from screening following brown stock washing.
Mills can improve the quality of chips going to the digester in several ways. Mills
that receive chips from an off-site source can develop a chipping quality control program for
suppliers to ensure that uniform, high-quality chips are received. Mills that chip on site can
closely monitor the operation of the chipper to maintain optimum settings (e.g., consistent and
effective blade settings) in order to produce chips of consistent size. This approach minimizes the
quantity of off-specification chips produced, and eliminates the need for chip thickness screens
(7).
Mills can also provide uniform chip dimensions with chip thickness screens. Chip
thickness screens are rotating disc screens that separate overly thick chips, which are then sliced
or crushed to the desired thickness for pulping. Chip thickness screening also removes knots and
compression wood, in which dioxin precursors are reported to be concentrated (8). Providing
chips of uniform thickness, free of knots and compression wood, results in improved pulping,
reduced screen rejects, and improved bleaching, and may also reduce the load on the mill's
wastewater treatment system. When knots and compression wood are removed prior to cooking,
they are typically burned in the mill's power boiler. If these components are instead removed from
the pulp by screening after cooking, they may be sent to the wastewater treatment system. New
mills in the U.S. typically use chip thickness screening and many existing mills have installed chip
thickness screens, usually as part of an overall woodyard or pulp mill upgrade. Therefore, EPA
considers chip quality control to be a part of the baseline technology used today at bleached
papergrade kraft mills and additional cost for this technology has not been included in the
BAT/PSES options.
7.2.2 Defoamers and Pitch Dispersants
Defoamers are used to break and inhibit the formation of black liquor surface foam
formed when air is entrained in the pulp during brown stock washing. Pitch dispersants are added
to pulp to prevent wood resins and fatty acids from depositing on the paper at the paper machine.
Both of these chemical additives are introduced into the pulp flow prior to brown stock washing
and are carried with the pulp into the bleach plant. Defoamers and pitch dispersants have been
7-3
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
shown to contain the chlorinated dibenzo-p-dioxin (CDD) and chlorinated dibenzofuran (CDF)
precursors dibenzo-p-dioxin (DBD) and dibenzofuran (DBF) (5,10). In the first (chlorine or
chlorine dioxide) stage of bleaching, the DBD and DBF are chlorinated to form 2,3,7,8-TCDD
and 2,3,7,8-TCDF.
Defoamers can be mineral oil based or water based. Mineral oil-based defoamers
are a blend of 90 percent oil and 10 percent additives. Defoamers made from re-refined oil are
particularly high in DBF concentration (11). DBD and DBF can be essentially eliminated from
defoamer oils (i.e., to a level of less than 1 ppb of both) by using two-stage severe hydrotreating
technology (11). Alternatively, completely oil-free defoamers that do not contain CDD and CDF
precursors may be used (12). Switching to non-contaminated defoamers can contribute to
reducing the content of 2,3,7,8-TCDF in bleached pulp and in mill effluent by at least 90 percent
(13). The Agency believes that the U.S. pulp industry has generally converted to the use of
precursor-free additives.
As evidenced by the drop in measured TCDD and TCDF discharges from
bleaching pulp mills, use of either water-based defoamers or defoamers made with precursor-free
mineral oils has been common industry practice since the early 1990s. At that time, it became
widely known in the industry that use of precursor-free defoamers in the brown stock or bleach
plant areas substantially reduces the dioxin formed in bleaching. Consequently, EPA assumes use
of precursor-free defoamers to be part of the baseline technology used today at bleached
papergrade kraft mills and costs have not been included in the BAT/PSES options.
7.2.3 Extended Cooking
At chemical pulping mills, wood chips or a non-wood fiber furnish are cooked in a
chemical solution in a digester at elevated temperature and pressure to dissolve the lignin that
holds the cellulose fibers together. Chemical pulping occurs in either a batch or continuous
digester system.
The most common continuous digester is the vertical downflow type. The wood
chips are preheated in a steaming vessel to remove air and some of the volatile wood constituents
before the chips enter the digester. The chips are mixed with the cooking liquor and are fed into
the top of the digester so that they move downward by gravity through the tower. The hydraulic
pressure in the tower is kept at approximately 1,140 kPa (165 psi). After the chips have been
impregnated with liquor, the temperature is raised to approximately 105 to 130
time of one to one and a half hours, until the pulping reaction is complete. The reaction is
stopped in the lower region of the tower, where diffusion washing of the pulp is carried out,
normally using a countercurrent flow of the filtrate from the first brown stock washer. The pulp
is blown from the bottom of the digester at about 1,380 kPa (200 psi) to a tank at atmospheric
pressure.
For batch cooking, a mill normally uses several vessels. Mills in the U.S. with
batch digesters have reported using anywhere from 3 to 24 vessels. Wood chips and cooking
7-4
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
liquor are added simultaneously to the top of the digester, after which the digester is sealed and
raised to a target operating pressure of approximately 700 to 900 kPa (102 to 131 psi) and
temperature of approximately 170
cooking for one to two hours, the pulp is blown into an atmospheric tank, from which it is
pumped to the brown stock washing system. Usually a mill staggers the digester cycles to
maintain a continuous flow of pulp through its washing and pulping sections.
The continuous process produces pulp at a consistent rate and with lower energy
requirements; however, a batch pulping system enables a mill to pulp several different grades at
once by using different digesters for different pulping conditions or fiber furnishes. Since the
continuous process was commercialized in the 1950s, the amount of chemical pulp produced by
continuous digesters has increased. Most new installations are now continuous systems.
Chemical pulp mills that bleach must remove enough residual lignin from the pulp
prior to bleaching to achieve their required final pulp brightness. The kappa number (a measure
of a pulp's lignin content) of the pulp entering the bleach plant dictates the amount of bleaching
chemicals needed. Since decreasing bleaching chemical use lowers both the cost for bleaching
chemicals and the environmental impact of the effluent from the bleach plant, it is generally
desirable for mills that bleach to lower the prebleaching kappa number as much as possible,
without seriously affecting pulp yield and strength.
Through work done by the Swedish Forest Products Research Institute (STFI) in
the late 1970s (14), the concept of "extended cooking" for papergrade kraft pulps was developed
and commercialized in the late 1980s. Extended cooking enables a mill to lower the kappa
number of the pulp entering the bleach plant further than is possible with a traditional kraft
pulping digester, while increasing pulp strength and maintaining or increasing pulp yield. During
extended cooking, the pulp is mixed with the cooking liquor for a longer time than in traditional
cooking, under modified temperature and alkalinity conditions. The process can be performed
using either a batch or continuous pulping system. Thirteen of the 86 mills in the Bleached
Papergrade Kraft and Soda Subcategory used extended cooking as of mid-1995. Extended
cooking is not typically used for papergrade kraft pulps that will not be bleached, because
achieving a low kappa number out of the digester is not as important as it is for pulps that will be
bleached.
In 1992, approximately 11 million kkg/year of kraft pulp was produced worldwide
using extended cooking (15). This figure represents about 20 percent of world bleached kraft
capacity. Figure 7-1 shows the increase in the amount of kraft pulp produced by extended
cooking from 1983 through 1992. Capacities shown are as reported by mills and vendors, and
have been normalized to air dry kkg/day. Extended cooking for dissolving kraft and sulfite pulps
has not been demonstrated on an industrial scale.
The types of continuous extended cooking processes most commonly used in the
U.S. are the Modified Continuous Cooking (MCC®) developed by Kamyr Inc. and Kamyr AB
and Iso Thermal Cooking (ITC®) process developed by Kvaerner, and Extended Modified
7-5
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
Continuous Cooking (EMCC®) processes developed by Kamyr Inc. Figure 7-2 shows a typical
EMCC® installation. These processes are similar, in that fresh cooking liquor, comprised of
sodium hydroxide and sodium sulfide, is added at several points in the digester, instead of at just
one point as with traditional continuous cooking. More lignin is dissolved with these processes
than is dissolved in the traditional digester, because the active chemical concentration is kept more
uniform throughout the cooking process. At the same time, less damage is done to the wood
fiber cellulose, because a high initial cooking liquor concentration is avoided. The resulting pulp
is stronger and has a lower kappa number than traditional pulps, while the pulp yield is
maintained.
Extended cooking in batch digesters yields similar results. Two systems are
available commercially: the Rapid Displacement Heating (RDH®) System, sold by Beloit Inc.,
the Super Batch® System, sold by Sunds Defibrator, Inc., and VISBATCH® and
ENERBATCH®, sold by Voest-Alpine. These processes maintain a more uniform cooking liquor
concentration throughout the cook than traditional batch cooking. The wood chips are initially
impregnated with warm black liquor under pressure to remove air in the chips. The warm black
liquor is then displaced with hot black liquor and white liquor to begin the cook. After cooking,
the spent cooking liquor is displaced with wash liquor from the first brown stock washer. The
spent cooking liquor becomes the warm black liquor used for impregnation during another cook.
The batch extended cooking process requires several large holding tanks to store liquor between
the various stages of the cooking process, and many older mills do not have the space to install a
batch extended cooking system. Most of the extended cooking installations in the world are
continuous rather than batch, although several batch systems have recently been installed.
The unbleached kappa number of softwood kraft pulps typically ranges from 30 to
32 for traditional cooking. Extended cooking by either the continuous or batch processes can
achieve softwood pulp unbleached kappa numbers ranging from 12 to 18. For hardwood kraft
pulps, the unbleached kappa number of approximately 20 for traditional cooking can be reduced
to 8 to 10 using extended cooking (5). As explained in Section 8, EPA has used higher kappa
number targets for the development of the regulatory options.
7.2.4 Effective Brown Stock Washing
In a chemical pulping mill, after the pulp leaves the digester it is cleaned through a
series of knotters, screens, and countercurrent washers to remove impurities and uncooked fiber
and to recover as much spent cooking liquor as possible. In brown stock washing, spent pulping
chemicals, along with any dissolved wood components, are separated from the pulp and sent to
the recovery boiler for recovery of chemicals and energy (steam generation). Effective brown
stock washing minimizes the amount of pulping liquor carried over to the bleach plant with the
pulp. The mill's effluent quality is also improved because residual black liquor that is carried over
to the bleach plant includes unchlorinated toxic substances, some of which appear to resist
degradation in biological treatment plants (16). If chlorine-based bleaching chemicals are used,
7-6
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
the organic compounds that are carried over with the pulp to the bleach plant also react with the
bleaching chemicals and, therefore, increase the mill's effluent load of chlorinated organics.
Effective brown stock washing also reduces the amount of bleaching chemicals
required to bleach the pulp to a given brightness, because well-washed pulp carries less organic
material, which competes with the pulp fiber for reaction with the bleaching chemicals. Finally,
effective brown stock washing is essential for satisfactory operation of an oxygen delignification
system.
Mills try to minimize the amount of water used for brown stock washing, because
all water added at this stage is typically evaporated in the black liquor recovery cycle. Although
using more water increases the removal of pulping liquor and dissolved organic material from the
pulp, the maximum amount of water that can be used depends on the capacity of the black liquor
evaporators and the additional energy requirements necessary to evaporate more black liquor.
Brown stock washing effectiveness at kraft mills is conventionally expressed as
saltcake (Na2SO4) loss per mass of pulp, and is considered to be effective if the washing loss is
less than 10 kg Na2SO4/kkg of pulp (17,18,19). A loss of less than 10 kg Na2SO4/kkg is
approximately equivalent to 99 percent recovery of spent pulping chemicals. The average
washing loss for the Bleached Papergrade Kraft and Soda Subcategory was 13.5 kg Na2SO4/kkg
in 1989, as reported in the 1990 census questionnaire. This washing loss is much lower than
losses were ten years ago.
The traditional method of pulp washing using a rotary vacuum drum washer has
been supplemented or replaced in many mills with other, high-efficiency washers, including
pressure diffusion washers, belt washers, and presses of various types. All are capable of
providing well-washed pulp.
Pressure diffusion washers are enclosed and operate at elevated pressure and
temperature, resulting in good washing efficiency. The pulp enters a pressure diffusion washer at
the top and moves downward as a fiber mat between the stationary central body of the washer
and the moving perforated cylindrical screen surrounding it. The wash water flows from the
center of the washer through the pulp mat and outer screen, and is extracted continuously from
the system. The pulp continues through the vessel and is removed at the bottom. Pressure
diffusion washers require relatively less floor space than other types of washers and are therefore
often selected for upgrading a mill's washing system where little space is available.
In a belt washer, the pulp flows onto an endless moving horizontal filter cloth and
is drawn off at the opposite end. Wash water is applied to the top of the pulp mat and is drawn
through the pulp by vacuum boxes located beneath the cloth. Each wash water addition point is
considered to be one "stage" of washing, and the wash water moves countercurrently from the
final stage through to the first stage on the belt. A belt washer can provide up to seven stages of
washing. The system can be enclosed to minimize air emissions. Belt washers are not currently
7-7
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
used to incrementally increase brown stock washing capacity, but are efficient systems for
replacing an entire washing line.
A wash press consists of a cylindrical washer and a press roll. Pulp that is washed
in a wash press leaves the system at a higher consistency than with other washing systems. The
geometric configuration of the cylindrical washer causes the pulp to be dewatered during washing,
because the pulp is forced into a smaller space and ultimately passes through a nip between the
washing cylinder and the press roll. Instead of leaving the washer at the usual consistency of
between 10 and 15 percent solids, the pulp mat leaves at a consistency of between 30 and 40
percent. This type of washer is beneficial when high consistency is required downstream of the
brown stock washing area.
7.2.5 Closed Screen Room Operation
After brown stock washing, pulp is usually screened to remove oversized particles.
The pulp is first diluted with fresh water, screened using gravity or pressure screens, and then
thickened in a decker to an appropriate consistency for the next process operation. In an open
screen room, the filtrate from the decker goes to the sewer. This sewered stream carries residual
organics and cooking liquor solids from the pulping operation. Closing the screen room
eliminates the overflow of decker filtrate to the sewer. This operation optimizes the water
balance around the washing and screening operations, because all of the decker filtrate is reused
as dilution water for the screening operation, or as brown stock wash water. Residual organics
and cooking liquor are thus returned to the chemical recovery cycle. Implementing closed
screening effectively allows the decker to be used as an extra stage of brown stock washing.
The closed screen room concept has been discussed in literature for many years.
The use of closed screening lowers the overall waste load to the mill wastewater treatment
system, including the chemical oxygen demand (COD) load. Closed screening is standard
equipment for mills using oxygen delignification because it is an integral part of the brown stock
area in which an appropriate flow balance must be maintained for the efficient operation of
oxygen delignification. The use of closed screening is becoming the common industry practice.
Based on information collected by EPA in mid-1995, 44 of 86 bleached papergrade kraft mills in
the United States used closed screen rooms.
The ability to operate the screen room as a closed process depends on a systematic
optimization of the pulping, washing, screening, and liquor recovery cycles, and the type of
washing and screening equipment available. Effective brown stock washing should be used to
minimize the amount of cooking liquor solids carried to the screening operation. In addition,
many mills are replacing gravity flow screens with pressure screens to prevent air entrainment and
resultant foaming problems (20).
7-8
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
7.2.6 Oxygen Delignification
Oxygen delignification uses oxygen gas to remove lignin from pulp after brown
stock washing and prior to bleaching. Using oxygen delignification between the kraft or sulfite
pulping processes and the bleach plant results in lower bleaching chemical demands than a
traditional bleaching sequence, because the unbleached kappa number drops by approximately 50
percent and the subsequent bleaching chemical requirements also drop in a relative manner (5). In
addition, bleaching to a particular brightness can often be accomplished using fewer bleaching
stages than a traditional bleach line if oxygen delignification is used prior to bleaching. Decreased
bleaching chemical use reduces pollutant levels in the mill's bleach plant effluent. Although the
operation of an oxygen delignification system in itself does not decrease the effluent flow from the
bleach plant, it can lessen water use if older, less efficient bleaching towers are bypassed and if
filtrates are sent to recovery boilers.
Over one-half of the world's bleached kraft production is subject to oxygen
delignification. This number includes 100 percent of mills in Sweden and Japan, all but one mill in
Finland, and essentially all new mills built worldwide in the last 10 years. As of mid-1995, 22
mills in the Bleached Papergrade Kraft and Soda Subcategory reported using oxygen
delignification. The amount of bleached papergrade kraft pulp produced by these mills using
oxygen delignification represented approximately 30 percent of the total U.S. production. One
mill in the Papergrade Sulfite Subcategory reported using oxygen delignification. Figure 7-3
illustrates the rate of increase in the use of oxygen delignification systems in the U.S. and
worldwide. Oxygen delignification has been adopted much more extensively in foreign mills than
in U.S. mills, while U.S. mills have adopted extended cooking more rapidly than foreign mills.
Figures 7-4 and 7-5 show medium- and high-consistency oxygen delignification
systems, respectively, both of which are in use today. In both cases, oxygen delignification must
start with well-washed pulp, and magnesium salt (MgSO4) must be added to protect the cellulose
fibers from degradation. After oxygen delignification, the pulp must be washed well to remove
organic material so that the subsequent bleaching stages can operate effectively.
High-consistency oxygen delignification is accomplished at a consistency of
approximately 25 to 30 percent, which is attained using a press prior to the oxygen delignification
tower. The pulp is then fed to a pressurized reactor into which oxygen and sodium hydroxide (or
oxidized white liquor) are added. The pulp is fluffed using baffles inside the tower to achieve a
more consistent reaction, and gaseous reaction products are purged from the vessel to avoid a fire
hazard. Pulp degradation has been a problem with high-consistency systems, even though
magnesium salt is added for pulp stabilization.
Medium-consistency oxygen delignification takes place at a consistency of between
10 and 20 percent, which is attained using a brown stock decker. The decker averts the need for
the more expensive press that is required for a high-consistency oxygen delignification system.
7-9
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
Prior to entering the reaction tower, the pulp is mixed with oxygen and sodium hydroxide (or
oxidized white liquor). Since fewer gaseous compounds are formed, the risk of fire is eliminated
with the medium-consistency system. The medium-consistency system has less potential for pulp
degradation than in a high-consistency system, but slightly less delignification occurs than with a
high-consistency reaction due to a lower reaction rate.
The filtrate from the post-oxygen delignification washers is sent to the recovery
boiler, marginally increasing the load on the boiler, but concurrently increasing the amount of
recovered chemicals and energy (21). Recycling the filtrate from the oxygen delignification
washers, rather than sending it to wastewater treatment, reduces the bleach plant effluent flow,
load of BOD5 by 30 to 50 percent, COD by 40 percent, color by approximately 60 percent, and
chlorinated organics by approximately 35 to 50 percent (5,8). Currently, all kraft and sodium-
based sulfite mills with oxygen delignification recycle the associated filtrate.
7.2.7 Steam Stripping
Wastewater streams in the pulping and chemical recovery areas of a chemical pulp
mill contain organic and sulfur compounds that may be emitted to the air or conveyed to the
wastewater treatment system. Condensate streams from evaporators, digester blow tanks, and
turpentine recovery systems at kraft mills contain the highest loadings of these compounds, with
evaporator condensate representing the major volume of pulping area condensate flow. Steam
strippers are used to control air emissions of organic and sulfur compounds from pulping area
condensate streams, and at the same time reduce the organic load of the stripped wastewater on
the wastewater treatment system.
Steam stripping is a fractional distillation process that involves the direct contact of
steam with wastewater. Figure 7-6 presents a schematic of a continuous steam stripper system.
Wastewater is pumped into the top of the stripping column, and steam is injected near the bottom
of the column. The column is typically equipped with perforated trays or packing to increase
contact between the vapor and the liquid. Heat from the steam vaporizes the volatile compounds
in the wastewater, which are carried out the top of the column with the steam. This overhead
vapor stream is typically incinerated on site with attendant energy recovery (steam generation)
(22). In the future, the overhead vapor stream may be concentrated or rectified to produce a
methanol-rich stream that can be used to replace other fuels burned on site. Wastewater leaving
the steam stripper passes through a heat exchanger to preheat the unstripped wastewater entering
the steam stripper. The stripped wastewater is then discharged to the wastewater treatment
system or reused in the mill for fresh hot water applications. The removal efficiency of volatile
compounds is determined by the steam-to-feed ratio in the column. The steam stripper may be a
stand-alone piece of equipment, or, at some mills, it may be integrated into the evaporator set.
A properly designed steam stripper can reduce the BOD5 load to a mill's
wastewater treatment system by removing organic compounds, primarily methanol, from the
pulping area condensate streams. Mills that currently use steam stripping to reduce the load of
organic constituents discharged to wastewater treatment report using steam-to-feed ratios ranging
7-10
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
from 145 to 215 kg/m3. A steam-to-feed ratio of 180 kg/m3 achieves approximately 90 percent
removal of methanol. Total reduced sulfur (TRS) compounds can be removed at lower steam
rates. Steam stripping is not part of the BAT/PSES model technology, but is part of MACT.
Therefore, costs for steam stripping were incorporated in the basis for the MACT I standards
rather than the BAT/PSES options. See the preamble at Section VIA.
7.2.8 Spent Pulping Liquor Management, Spill Prevention, and Control
Mills that perform chemical or semi-chemical pulping of wood or other fibers
generate spent pulping liquors that are generally either recovered in a chemical recovery system or
treated in a wastewater treatment system. These mills may lose pulping liquor through spills,
equipment leaks, and intentional diversions from the pulping and chemical recovery areas of the
mill. Spills and intentional diversions of pulping liquor are a principal cause of upsets in biological
wastewater treatment systems, the type of treatment system used at most chemical and semi-
chemical pulp mills. Spent pulping liquor losses also increase the need for pulping liquor make-up
chemicals, decrease energy generated from pulping liquor solids combustion, and increase
hazardous air pollutant emissions.
Unintentional pulping liquor losses at pulp mills are most commonly caused by
process upsets, equipment breakdowns, and tank overfillings. Maintenance and construction in a
mill's pulping and chemical recovery areas may cause intentional diversions of pulping liquor to
the wastewater treatment system. Spent pulping liquor may also be lost during normal mill
operations, such as planned shutdowns and start-ups and pulp grade changes.
In addition to the potential harm to biological wastewater treatment systems and
possible increased releases of toxic and hazardous substances associated with spills and other
releases of spent pulping liquor, the discharge of substantial quantities of soap and turpentine has
been shown to be extremely harmful to biological wastewater treatment systems and has resulted
in violations of NPDES permit levels. The provision of secondary containment for turpentine
storage tanks and other curbing, along with diversion and containment measures in soap and
turpentine handling areas, constitute another element of a management program aimed at
protecting wastewater treatment effectiveness and protecting the receiving waters.
Management programs, combined with engineered controls and monitoring
systems, can prevent or control spent pulping liquor losses. These efforts should be both
proactive to prevent pulping liquor losses and reactive to control spills after they have occurred.
Practices to prevent or control spent pulping liquor losses at chemical pulp mills
include the following:
Management of process operations to minimize variability.
Preventative maintenance programs for equipment in spent pulping liquor
service.
7-11
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
Automated spill detection, such as conductivity sensors in sewers in the
pulping and chemical recovery areas.
Frequent operator surveillance of pulping and chemical recovery areas to
quickly detect and repair leaks.
Secondary containment or annual integrity tests and high-level alarms on
pulping liquor bulk storage tanks.
Secondary containment for turpentine storage tanks and other collection,
containment, and control measures for soap and turpentine areas.
Spill collection systems for the pulping and chemical recovery areas with
sufficient capacity to store collected spills and planned liquor diversions.
The collected liquor may then be recovered in the chemical recovery
system or slowly released to the wastewater treatment system at a rate that
does not adversely impact the wastewater treatment system.
Mills with effective pulping liquor spill prevention and control programs have
instituted a combination of these practices to substantially eliminate black liquor losses. It has
been reported that the practical maximum reduction in BOD5 raw wastewater loading that can be
attained from spill prevention is 5 kg/kkg (23). Based upon site visits by the Agency, it appears
that sulfite mills are less likely than kraft or soda mills to have engineered controls for collecting
spills and leaks of pulping liquors at the immediate process areas.
Details of the practices listed above, and the associated estimated costs and
effluent reduction benefits for mills that chemically pulp wood or other fibers are in the document
entitled, "Technical Support Document for Best Management Practices for Spent Pulping Liquor
Management, Spill Prevention, and Control" (24).
7.2.9 Maximizing Recovery Boiler Capacity
At kraft mills, spent cooking liquor (black liquor) from the digester and from
brown stock washing is sent to the chemical recovery area, where it is concentrated in multiple-
effect evaporators and then burned as part of the chemical recovery process. Section 4.2.4 in the
TDD describes this process in detail. The organic fraction of the concentrated black liquor solids
generates heat (captured as steam generated in the boiler) as it is burned, and the inorganic
material produces a molten smelt in the hearth that is dissolved to regenerate the cooking
chemicals.
When a mill improves its brown stock washing or installs oxygen delignification or
extended cooking, it is common practice to recycle the resulting filtrates to the recovery boiler,
thus increasing the amount of organics sent to the recovery boiler. The heating value of these
7-12
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
additional organics is an important factor in determining whether a mill needs to increase the
burning capacity of its recovery boiler when it makes these process changes.
The increase in the heating load on the recovery boiler (referred to in this section
as an increase in boiler capacity) depends not only upon the amount of organics coming from the
process, but also upon their heating value. The heating value of the organics differs depending
upon the wood (hardwood or softwood) and the process from which the organics are obtained.
For example, the solids recycled from an oxygen delignification process have been oxidized and
therefore have a lower heating value than those solids recycled from a brown stock washer or
digester. Also, more organics are recovered from softwood pulp than from hardwood pulp.
In U.S. mills, the increase in recovery boiler capacity that results from improving
brown stock washing alone is minimal (generally less than 1 percent) because, as indicated in
Section 7.2.4, U.S. mills, on average, have fairly good brown stock washing. This contribution to
the recovery boiler is negligible when the accuracy of boiler flow measurements is taken into
account. However, the impact of extended delignification processes such as extended cooking
and oxygen delignification is an important consideration in determining increases in recovery
boiler capacity.
EPA prepared a detailed analysis of the impacts of extended cooking, oxygen
delignification, screen room closure, improved brown stock washing, and other components of
BAT and BMP on the kraft recovery system. This analysis is presented in the Analysis of Impacts
of BAT Options on the Kraft Recovery Cycle (21). The main conclusion is that heat load on the
boiler, not black liquor solids (BLS), is the key parameter determining boiler capacity. EPA's
BAT cost model was adjusted to account for the impacts on the recovery system of additional
black liquor recovered.
EPA estimated that the increase in thermal load from Option A will average 1.5
percent, with a maximum of 5 percent. For Option B, the average increase in thermal load is 2.2
percent, with a maximum of 8.7 percent. In estimating mill-specific costs, if the increase in
thermal load to the boiler was estimated to be less than 1 percent, EPA provided no capital costs,
assuming that this increase in load could be accommodated by improving boiler operation. For
thermal load increases exceeding 1 percent, three options were available: the addition of
anthraquinone, implementation of oxygen-based black liquor oxidation, and boiler upgrades (air
system upgrades, improving liquor delivery, and firing high concentration black liquor, among
others). Boiler upgrades are high-cost modifications that typically increase boiler capacity by 10
percent, and thus are not appropriate for accommodating the smaller thermal load increases EPA
estimated would result from BAT. The TDD (Section 8.2.9) also discussed several other
methods of increasing recovery boiler capacity.
Anthraquinone and oxygen-based black liquor oxidation are not recovery boiler
modifications; they are means of reducing the thermal load on the boiler that can be employed
even if a mill is recovery-boiler-limited. Anthraquinone is a catalyst that increases pulp yield and
thus decreases the quantity of BLS to be burned for a fixed quantity of production.
7-13
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
Anthraquinone is only an option for reducing the thermal load on the boiler for mills not already
adding it to the digester. Oxygen-based black liquor oxidation is not a common practice in the
United States, but equipment for this process is sold by two major industrial gas vendors. In this
process, black liquor is partially oxidized before it is fired to the recovery boiler, reducing the
thermal load of the black liquor by about 5 percent. This technology could be applied only to
mills with non-direct contact evaporator recovery boilers (because most direct contact
evaporation recovery boilers are already equipped with black liquor oxidation to reduce odor
emissions).
Because of concerns about which boilers at bleached papergrade kraft mills can
accommodate the increase in thermal load that will result from BAT, NCASI surveyed the
industry in 1995. NCASI provided the results of this survey to EPA. A recovery boiler has a
margin of capacity to accommodate an increased black liquor load unless all possible
modifications to a boiler have been made. Of 190 boilers represented in the survey, 78 (41
percent) show evidence of being operated at their maximum capacity. Even for these boilers,
however, the reductions in thermal load provided by anthraquinone or oxygen-based black liquor
oxidation can accommodate the small increased loadings from BAT and BMP.
While estimating the costs of BAT, EPA used the status of each boiler reported in
the 1995 survey to determine if an adjustment to recover boiler capacity was required. For the 84
bleached kraft and soda mills for which compliance costs were estimated, EPA estimated the
following recovery boiler capacity adjustments would be required:
None
Anthraquinone
Oxygen Black Liquor Oxidation
Recovery Boiler Air System Upgrade
Option A
(Number of mills requiring
recovery boiler capacity
adjustment)
61
9
13
1
Option B
(Number of mills requiring
recovery boiler capacity
adjustment)
57
10
16
1
Thus, EPA concluded that most mills would not require adjustments to the thermal
load to the recovery boiler (or upgrades to the boiler) to accommodate the increase in black
liquor thermal load to the boiler that will result from BAT. No mills require recovery boiler
rebuilds or replacements.
7-14
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
7.3 Pollution Prevention Controls Used in the Bleach Plant
This section describes applicable technologies for reducing and preventing
pollutant discharges from bleach plants at chemical pulp mills. For most facilities, the Agency
defines the bleach plant as including the stage where bleaching agents (e.g., chlorine, chlorine
dioxide, ozone, sodium or calcium hypochlorite, peroxide) are first applied, each subsequent
extraction stage, and each subsequent stage where bleaching agents are applied to the pulp. A
limited number of mills produce specialty grades of pulp using hydrolysis or extraction stages
prior to the first application of bleaching agents. For those mills, EPA considers the bleach plant
to include those pulp pretreatment stages. Although oxygen delignification systems are integrated
with pulping and chemical recovery systems, the convention in the industry is to include oxygen
delignification when specifying bleach sequences (e.g., O D/C Eop D). The Agency is using that
convention in this document, although it considers oxygen delignification to be part of pulping
(prebleaching) rather than bleaching. Section 4.2.6.1 of the TDD presents an overview of bleach
plant operation.
7.3.1 Ozone Bleaching
Ozone, a powerful oxidizer, has been studied for over 20 years as a potential
replacement for chlorine and chlorine dioxide in the first stage of pulp bleaching. Historically,
two major drawbacks have inhibited the adoption of industrial-scale ozone bleaching: high cost
and poor selectivity (i.e., a high degree of carbohydrate degradation and therefore viscosity drop)
(25). Recent technological developments, such as the introduction of medium-consistency ozone
bleaching and improvements in ozone efficiency and selectivity, have removed these
disadvantages to using ozone (25,26). Ozone bleaching technology continues to evolve. The first
full-scale ozone bleaching systems in a sulfite mill and a kraft mill started up in 1991 and 1992,
respectively.
Ozone is generated either from oxygen or air, though it is normally produced using
oxygen. The oxygen (O2) passes through a series of tubes and a high voltage is applied, causing
the oxygen molecules to dissociate. The dissociated molecules recombine to form ozone (O3),
which is relatively unstable. Since ozone can easily decompose back to oxygen, ozone must be
generated on site for immediate use in the bleach plant.
Ozone bleaching has been researched at low, medium, and high consistencies. No
full-scale low-consistency systems are in existence at this time. Most currently operating full-
scale systems process the pulp at medium consistency (10 to 15 percent). Medium-consistency
systems have lower capital costs than high-consistency systems do (26).
Oxygen delignification with effective post-oxygen washing is necessary prior to
ozone bleaching to lower the lignin content of the pulp and therefore reduce the ozone charge
required. No "target" kappa numbers are currently defined for the pulp entering the ozone stage;
for ozone bleaching, target kappa numbers tend to be site-specific. Ozone oxidizes the
carbohydrates in pulp as well as the lignin, so the ozone charge must be optimized to achieve the
7-15
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
maximum pulp delignification while minimizing the effects on pulp viscosity. Mills currently
operating ozone bleaching systems use between 5 and 12 kg ozone per kkg of pulp, although
actual application rates and operating conditions for this technology are usually confidential. A
high-consistency system allows a higher ozone application rate than a medium-consistency
system.
The processes and equipment used for ozone bleaching and oxygen delignification
are similar. Prior to being fed to the ozone bleaching tower, the pulp is treated with either acetic
or sulfuric acid to lower the pH. As with oxygen delignification, the pulp may be fluffed in the
reactor to facilitate a more uniform reaction. Ozone is delivered to the reactor with an oxygen
carrier gas. This carrier gas can be recovered after ozone bleaching, cleaned, and recycled to the
ozone generator, or used elsewhere in the mill (e.g., in an oxygen delignification system or an
oxygen-enhanced extraction stage).
The reaction of the pulp with the ozone normally takes a few minutes, as opposed
to a few hours with other bleaching agents. The ozone bleaching reactor is therefore much
smaller than other reaction vessels in traditional bleach plants.
Process effluents from ozone bleaching can be recycled to the recovery boiler,
which decreases the volume of bleach plant effluent and the amount of non-chlorinated
compounds discharged from the bleach plant. Because chlorine and chlorine derivatives are not
used for first stage bleaching, chlorinated organic compounds (e.g., CDDs/CDFs, chlorinated
phenolics) are not formed. The increase in the load of solids on the recovery boiler from recycling
the ozone stage filtrate is lower than the increase from an oxygen delignification stage. The
increase in solids for the ozone and subsequent extraction stages cause an additional heating load
on the recovery boiler of approximately 1 percent (27).
In September 1992, a U.S. kraft mill began to produce lower brightness kraft pulp
using ozone bleaching. This mill pulps and bleaches softwood to approximately 82-83 ISO;
however, the bleaching sequence at this mill (OZED) includes a final chlorine dioxide brightening
stage. The following description of this mill is based upon a recent publication (28).
This mill reports significant environmental benefits from the OZED bleaching
sequence compared to sequences CEDED and OD/CED for both hardwood and softwood for
parameters such as color, BOD5, COD, chloroform, and AOX. Although chlorine dioxide is used
in the final bleaching stage, 2,3,7,8-TCDD has not been detected in D-stage filtrate or bleached
pulp. Another advantage of this sequence is the potential to recycle filtrates from the oxygen,
ozone, and extraction stages to the recovery system. Compared to bleaching sequences of
CEDED and OD/CED, the OZED sequence achieves equivalent pulp properties, except for
viscosity, for both hardwood and softwood. Since pulp viscosity and strength have a different
correlation for oxygen/ozone bleached pulps than for chlorine compound bleached pulps, the mill
reports that the decrease in viscosity has not been a problem.
7-16
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
Additional environmental benefits and further recycling of filtrates could be
achieved if the final chlorine dioxide stage were to be converted to use a non-chlorine containing
compound. The mill has studied combinations of oxygen and ozone with peroxide and believes
that an acceptable softwood TCP pulp can be made. The mill does not currently use a TCP
bleaching sequence because it would substantially increase operating costs over the cost of the
OZED sequence.
7.3.2 Improved Mixing and Process Control
To realize the full benefits of technologies such as high chlorine dioxide
substitution, oxygen-enhanced extraction, and oxygen delignification on the bleach plant effluent,
the pulp and bleaching agents must be well mixed and the chemical addition rate controlled as
precisely as possible. Normally, when these technologies are installed, mixing and process control
are also upgraded.
High shear mixers, introduced in the late 1970s, dramatically increased the contact
of the pulp with gases such as oxygen, making the oxygen-enhanced extraction stage a practical,
effective bleaching technology (29). High shear mixing has similarly become a vital part of a
medium-consistency oxygen delignification system. To ensure uniform application of the
chemicals in a high chlorine dioxide substitution stage, the pulp must be well-mixed with the
chlorine dioxide. EPA has included costs for high shear mixers at individual mills as necessary.
7.3.3 Chlorine Dioxide Substitution
In the late 1980s and early 1990s, bleached kraft mills began substituting chlorine
dioxide (C1O2) for some or all of the molecular chlorine normally used in the first bleaching stage.
This process became known as chlorine dioxide substitution, and became common practice in the
industry at that time because C1O2 substitution reduces the formation of chlorinated organics in
the bleach plant effluent and lowers bleach plant chemical consumption (4).
As mills used more and more C1O2, additional benefits were realized such as
further reductions in chlorinated organics in bleach plant effluent and more consistent and
improved pulp quality because it minimizes cellulose degradation. Then mills began to use
complete (100 percent) substitution of C1O2 for chlorine. By mid-1995, over 32 percent of
bleached kraft production in the United States was made using complete C1O2 substitution, and
many companies had committed to installing the technology on additional bleach plants. By 1996,
67 percent of bleached kraft production in Canada was made using complete C1O2 substitution
(30).
7-17
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
The amount of C1O2 used is expressed as percent substitution and is defined as the
percentage of the total chlorine bleaching power of the first bleaching stage that is provided by
chlorine dioxide. It is calculated by the following formula:
Percent Substitution =
2.63 (C1O2 in kg/kkg)
2.63 (C1O2 in kg/kkg) + (CL in kg/kkg)
(1)
where 2.63 equals the oxidizing power of chlorine dioxide compared to chlorine. Chlorine
dioxide is a stronger oxidizing agent than chlorine. Consequently, less chemical is required when
chlorine dioxide is substituted for chlorine.
Chlorine dioxide must be generated on site because it is unstable and cannot be
transported in a pure form by truck or rail. As of January 1, 1993, most mills that bleach
chemically pulped wood pulps generated chlorine dioxide on site, and more than 60 percent of
those mills used chlorine dioxide substitution in the first bleaching stage, as shown below:
Subcategory
Bleached Papergrade Kraft and Soda
Papergrade Sulfite
Total Number
of Mills
87
10
Number With
CIO2 Generation
On Site
79
5
Number With C1O2
Substitution in First
Bleaching Stage
60
4
In generating chlorine dioxide, byproducts are produced. Sodium sulfate (Na2SO4)
and chlorine are generated in different amounts, depending on the chlorine dioxide generator, as
shown below (31):
Chlorine Dioxide Generator
Solvay
Mathieson
R2
R3
R8
C12 (kkg/kkg CIO2)
0
0
0.66
0.66
0
Na2SO4 (kkg/kkg C1O2)
3.5
3.5
7
2.4
1.4
It is important to control the ratio of chlorine applied in the first bleaching stage to
lignin content of the pulp entering the first bleaching stage, as well as the amount of C1O2
substitution, to most effectively reduce the formation of chlorinated compounds (4). The ratio of
total chlorine (from molecular chlorine and chlorine dioxide) in the first bleaching stage
7-18
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
(expressed as percent on pulp) to kappa number of the pulp entering the first bleaching stage is
referred to as the active chlorine multiple (ACM) or kappa factor:
[CL (kg/100 kg brown stock) + 2.63 CIO, (kg/100 kg brown stock)]
Kappa Factor (ACM) = (2\
Pre-chlorination kappa number ^ '
A Canadian study evaluated the formation of 2,3,7,8-TCDD and 2,3,7,8-TCDF in
the final effluent from bleached kraft mills relative to C1O2 substitution and ACM (29). The study
developed an equation for the relationship between C1O2 substitution and ACM which is described
in Section 8.3.5 of the TDD. For complete C1O2 substitution, the study equation predicts that
TCDDs and TCDFs will not be formed if an ACM of 0.48 or less is used. The detection limits in
the study were 10 ppq for 2,3,7,8-TCDD and 30 ppq for 2,3,7,8-TCDF.
EPA's data, described in a separate document (32), show that 2,3,7,8-TCDD is
normally not detected at mills using complete C1O2 substitution but that 2,3,7,8-TCDF is
occasionally detected below 30 ppq (even when the ACM is below 0.48). This document also
shows that the 12 chlorinated phenolic compounds regulated by this rule are normally not
detected at mills using complete C1O2 substitution and that chloroform generation is substantially
reduced for these mills compared to conventional bleaching sequences.
7.3.4 Enhanced Extraction
In the alkaline extraction stages of the bleach plant, lignin reaction products from
the preceding acid stages are dissolved, or extracted, by applying sodium hydroxide (caustic)
solutions to the pulp. These dissolved products are then washed from the pulp. The first caustic
extraction stage of the bleach plant can be enhanced with oxygen, hydrogen peroxide, or both
(shown as E0, Ep, and Eop, respectively) to replace equivalent quantities of chlorine-based
bleaching chemicals in other bleaching stages (34). Enhanced extraction is a low capital cost
measure that improves effluent quality by reducing chlorine consumption, therefore reducing the
amount of chlorinated organics in the bleach plant effluent. Mills implementing enhanced
extraction typically reduce molecular chlorine use in the first bleaching stage, while keeping the
chlorine dioxide addition rate constant (resulting in a higher level of chlorine dioxide substitution).
Oxygen-enhanced extraction became commercially feasible in the early 1980s due
to the development and introduction of high shear mixers for pulp stock. A high shear mixer is
required to ensure good mixing of the gaseous oxygen with the pulp. The extraction must be
carried out in either an upflow extraction tower or a downflow tower preceded by a small upflow
pre-retention tube to maintain the pressure required to keep the oxygen gas in solution until it has
reacted with the pulp. Adding oxygen to the extraction stage improves delignification by
approximately 25 percent (35), while allowing the mill to use less chlorine or chlorine dioxide in
the overall bleaching sequence. Adding between 4 and 6 kg of oxygen per kkg of pulp saves
approximately 2 kg of active chlorine per kg of oxygen (29).
7-19
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
Oxygen-enhanced extraction normally reduces overall bleaching chemical costs,
thus justifying, in many cases, the capital cost of the additional mixing power and piping required.
Data from many mills in the U.S. that converted their bleaching sequences from E to E0 operation
during the 1980s support this fact.
Hydrogen peroxide is usually added at the inlet to the oxygen mixer when Eop is
used, or at the inlet to the stock pump for Ep alone. Adding hydrogen peroxide in the first
extraction stage improves delignification and reduces chlorine-based chemical requirements either
in the first chlorination stage or further along in the bleaching sequence. Applying 1 kg of
peroxide per kkg of pulp results in an active chlorine savings of approximately 2 to 3 kg (31). If
hydrogen peroxide-enhanced extraction is used following 100 percent chlorine dioxide
substitution, a higher final brightness can be achieved (29).
Mills that use hydrogen peroxide-enhanced extraction are able to reduce the
amount of either molecular chlorine or chlorine dioxide in other bleaching stages. The cost of
hydrogen peroxide is currently much higher than the cost of molecular chlorine and slightly lower
than the cost of chlorine dioxide (see Section 10). Therefore, a mill operating hydrogen peroxide
enhancement increases its operating costs if it reduces molecular chlorine use, but decreases
operating costs if it reduces chlorine dioxide use.
Adding oxygen to an extraction stage is more capital intensive than adding
peroxide because, as described above, a high shear mixer and other equipment must be used for
the bleaching stage to operate effectively. Therefore, the simplest way for a mill to enhance its
extraction stage is with hydrogen peroxide. Unlike oxygen, hydrogen peroxide is also effective in
enhancing the second extraction stage. In addition, the bleaching power of the Eop stage can be
increased by raising the temperature, which is another strategy to reduce bleaching chemical
charge at relatively low cost.
A significant number of U.S. bleached pulp mills have implemented enhanced
extraction. As of January 1, 1993 and mid-1995, the number of mills using some form of
enhanced extraction were:
Subcategory
Bleached Papergrade Kraft and Soda
Papergrade Sulfite
Total Number of
Mills
87
10
Number With
Enhanced Extraction
in 1993
(ED, ED, or Eon)
65
4
Number With
Enhanced Extraction
in 1995
(E0, En, or Enn)
77
6
7-20
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
7.3.5
Elimination of Hypochlorite
Sodium hypochlorite and calcium hypochlorite are effective bleaching agents that
attack lignin. Sulfite pulps are more easily bleached with hypochlorite than kraft pulps because
the lignin is more easily solubilized (19). Hypochlorite can also degrade cellulose, decreasing pulp
viscosity. To limit cellulose degradation, hypochlorite is usually applied in an intermediate
bleaching stage for kraft pulps.
Chloroform is generated when pulp is bleached with hypochlorite (36). Mills that
use sodium or calcium hypochlorite in one or more bleaching stages generate approximately ten
times as much chloroform as mills using a CEDED bleaching sequence (37). Controlling
chloroform releases generally entails eliminating hypochlorite as a bleaching agent. The bleaching
power of hypochlorite can be replaced by chlorine, chlorine dioxide, peroxide, and/or oxygen.
However, replacing hypochlorite with chlorine is counterproductive if the purpose is to reduce
chloroform generation, because bleaching with molecular chlorine also generates chloroform.
For some mills, particularly mills with short bleaching sequences (e.g., CEH),
eliminating hypochlorite requires replacing the hypochlorite bleaching tower with a new chlorine
dioxide tower, washer, and auxiliaries made of materials resistant to the more corrosive
environment of chlorine dioxide bleaching. Some mills may be able to modify the bleaching
chemical additions to other stages (i.e., adding oxygen and/or peroxide to the first extraction
stage) and abandon the hypochlorite stage, rather than replacing it. Mills with a CEHDED-type
of bleaching sequence and mills that use hypochlorite only in extraction stages may be able to
eliminate the hypochlorite stage.
Replacing hypochlorite reduces direct operating costs, because hypochlorite is
more expensive than chlorine dioxide, oxygen, and peroxide. Hypochlorite also has a lower
chlorine equivalence factor (see Section 10). The number of mills using a hypochlorite stage or
hypochlorite-reinforced extraction as of January 1, 1993 and mid-1995 are shown below:
Subcategory
Bleached Papergrade Kraft and Soda
Papergrade Sulfite
Total Number
of Mils
87
10
Mills Using
Hypochlorite in 1993
36
9
Mills Using
Hypochlorite in 1995
20
7
7.3.6
Strategies to Minimize Kappa Factor and DBD and DBF Precursors
As noted in Section 7.3.3 in this document, the control of the kappa factor (or
ACM) in the first bleaching stage is an important process control parameter. EPA's data and the
data of other researchers indicate that particular chlorinated organic pollutants are not detected in
bleach plant effluent when bleaching kappa factors are maintained below certain values. In
7-21
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
particular, Shariff el al. (38) studied TCDF and, for C1O2 bleaching, developed the following
relationship:
TCDF ~ (precursor concentration) (kappa factor)4
Assuming this relationship is correct, to minimize the formation of TCDD/F, process operators
should minimize:
1. Kappa factor; and
2. Precursors contained in the pulp entering the bleach plant.
Each strategy is discussed separately below. This discussion also appears in EPA's Comment
Response Document (39). Some of the strategies for minimizing kappa factor are discussed in
more detail in separate subsections of this section but are briefly repeated here for completeness
of this discussion.
Minimizing Kappa Factor
Strategies for minimizing kappa factor include:
Reducing first stage kappa factor and shifting bleaching to later stages that
are not implicated in TCDD/F formation;
Optimizing conditions in the first bleaching stage for most efficient use of
the C1O2 applied;
Improving process control so that kappa factors are not permitted to
fluctuate;
Modifying the extraction stages by peroxide and/or oxygen reinforcement
and operating at a higher temperature;
Reducing knots, compression wood, dirt, and shives by improving
screening or operating oxygen delignification; and
Reducing black liquor carryover by improving pulp washing.
Each of these strategies is discussed below.
Shifting bleaching to later stages - Bleaching is a multi-stage process in which,
historically, acid chlorine bleaching stages are alternated with alkaline extraction stages. This
multi-stage process provides considerable flexibility to mill operators in terms of where they make
the bleaching reactions happen. Mills producing high brightness pulp generally have five
bleaching stages (D0ED1ED2). They can reduce the kappa factor in the D0 stage and increase the
7-22
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
bleaching power of later stages by reinforcing the extraction stages with oxygen and peroxide and
if necessary, increasing the C1O2 charge in the Dl and D2 stages.
Optimizing first stage bleaching conditions - Reeve (40) reports that C1O2
required to reach a target kappa number (CEK) is a function of pH, chloride ion concentration,
the type of pulp, and extraction stage conditions. If pH is too high, for example, required kappa
factor will increase. Georgia-Pacific reported that control of first stage pH was key to effective
low kappa factor bleaching (41). Insufficient chloride ion concentration will also increase the
required kappa factor. Pulp mill bleaching conditions such as pulp consistency, temperature,
retention time, and pulp cleanliness are extremely varied. In addition, the objective of bleaching,
in terms of target brightness, cleanliness and strength differ among products and mills. Because of
this variety, bleaching conditions must be optimized for each product produced at each mill.
Improving process control - At some mills, the quality of unbleached pulp
entering the first stage of bleaching is poor and inconsistent. The kappa number may vary widely
over space and time. If a constant bleaching chemical application rate is used, the kappa factor
also varies widely. To control kappa factor thus means controlling the kappa number. Extended
cooking and oxygen delignification produce pulps with a more uniform and constant kappa
number than the pulp produced by many conventional pulping operations. Improved mixing also
produces a more uniform pulp. In addition, in-line kappa number measuring instruments with
feed-back controls are used to adjust the chemical application rate to the measured kappa number.
A charge of bleaching chemical can also be applied stepwise to improve process control (42).
Peroxide and oxygen reinforced extraction - It has become common practice to
compensate for a lower kappa factor by using peroxide and/or oxygen in the second stage (Ep or
Eop). This strategy can also reduce operating costs, because peroxide is less expensive than
chlorine dioxide. In addition, the bleaching power of the Eop stage can be increased by raising the
temperature, which will reduce the required kappa factor at relatively low cost.1
Reducing dirt and shives entering the bleach plant - Conventionally pulped
wood contains dirt (colored particulate matter) and shives (intact fiber bundles). Dirt and shives
can be removed by bleaching. Often, the C1O2 charge is established to meet dirt specifications,
and may exceed the charge necessary to meet brightness specifications. Improved chip size
control (to remove knots and compression wood for improved pulping), improved brown stock
screening, oxygen delignification, and extended cooking reduce dirt and shives entering the bleach
plant making possible the use of lower bleaching chemical charge (a lower kappa factor).
'In addition to reducing the potential for the formation of TCDD/F, reduced kappa factor minimizes bleaching costs.
Reeve and Dence (1996) report that after balancing the C1O2 applied in the first stage with the C1O2, peroxide, and
oxygen used in later stages, the most cost effective kappa factor is 0.14 to 0.16, depending on wood species.
7-23
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
Reducing black liquor carry-over by improving pulp washing - Poorly washed
pulp carries a large quantity of black liquor into the bleach plant. Black liquor contains the non-
fibrous fraction of raw wood that is separated from the fiber during pulping, as well as the spent
cooking chemicals. This black liquor carry-over consumes oxidizing (bleaching) chemicals. For a
poorly washed pulp, a higher kappa factor is required to achieve the same degree of
delignification in the first bleaching stage than is required for a well-washed pulp.
Minimizing Precursors
The results reported by EPA and Shariff et al. indicated that TCDF can still be
formed at mills that have eliminated obvious sources of DBD and DBF precursors such as
contaminated defoamers and kerosene. Thus, some TCDF precursors must be contained in brown
stock pulp. This pulp is a mixture of cellulose fiber, lignin, and black liquor carry-over. Hise (43)
and Hrutfiord (44) have speculated that precursors could originate from lignin fragments that
accumulate in the black liquor after pulping.
Berry (45) found that precursor concentration can be decreased by steam stripping
brown stock pulp to remove volatile components. Berry (46) explains that steam stripping is an
inherent part of high consistency oxygen delignification and also occurs when brown stock pulp is
released from the digester at a temperature at which steam is driven from the pulp. Berry (45)
found that processing pulp through an oxygen delignification stage using nitrogen-gas instead of
oxygen removed precursors. Because no oxidation occurred under these test conditions, Berry
concluded that the precursor removal was a result of the physical actions of the washing and
steam stripping. Others conclude that oxygen delignification will oxidize a significant portion of
the lignin monomers (47), again, resulting in reduced precursor levels.
Hrutfiord (44) found that greater quantities of TCDD/F are formed in the
chlorination of compression wood than normal wood. Compression wood, prevalent in knots, is
the main source of/>-hydroxyphenyl which can form DBD and DBF.
Pulping digester and evaporator condensates, derived from black liquor, contain
many of the same constituents as the black liquor itself. The condensates are frequently treated to
remove foul-smelling TRS and then to wash pulp prior to bleaching. It has been hypothesized
that inadequately treated condensates used to wash oxygen delignified pulp are a source of
TCDD/F precursors (Francis, Personal Communication, 1992).
Removal of TCDD/F precursors, including lignin monomers2 contained in black
liquor, could be accomplished through:
2If the TCDF precursors are actually in the lignin, a relationship should exist between kappa number into bleaching and
TCDF. Data collected by EPA and data submitted by Shariff et al. do not demonstrate this conclusion, possibly because
kappa factor has a much more significant impact on the formation of TCDF than precursor concentration.
7-24
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
Brown stock washing that achieves extremely low carry-over of organic
compounds;
Oxygen delignification (through increased washing, steam stripping, or
other physical mechanisms or, possibly, by chemical oxidation of
precursors);
Removal of knots (compression wood) from brown stock pulp prior to
bleaching; and
Use of only precursor-free pulping and evaporator condensates in post-
oxygen delignification washing.
7.3.7 Enzyme Bleaching
Enzymes are organic compounds that act as catalysts in reactions. Xylanase
enzymes improve the bleachability of wood pulps by partially hydrolyzing the xylan (the primary
bonding agent between the cellulose and the lignin) although the exact mechanism by which they
aid in bleaching is not known (48). The lignin is therefore more easily removed in subsequent
bleaching stages. Xylanase may be added to the pulp after brown stock washing or after oxygen
delignification to reduce or eliminate the need for bleaching with chlorine compounds. The
optimum conditions for the xylanase reaction are temperatures between 40 and 55
between 4 and 6, and retention time between 0.5 and 3 hours (48). The xylanase is applied at less
than 1 kg/kkg of pulp.
Several mills worldwide have conducted full-scale trials with xylanase on kraft
pulps with resulting increases in brightness and viscosity and no loss of pulp strength (48). More
experimental work has been done using enzymes to bleach kraft pulps than to bleach sulfite pulps.
7.3.8 Peroxide Bleaching
Though hydrogen peroxide is primarily used to reinforce caustic extraction stages,
hydrogen peroxide can replace chlorine compounds in bleaching chemical pulps. The brightness
achievable using peroxide can be increased by lowering the lignin content of the pulp as much as
possible prior to bleaching (e.g., by using oxygen delignification).
While bleaching stages that use chlorine compounds inherently remove metal ions
(e.g., Ca, Mg, Na, etc.) from the pulp, these ions will react with peroxide to form hydroxyl
radicals that can degrade cellulose. Therefore, the metal ions must be removed from solution by
using chelating agents, followed by effective pulp washing prior to applying peroxide (49). When
a peroxide stage follows a chlorine compound bleaching stage, separate addition of a chelating
agent is not required. One method of peroxide bleaching using chelating agents is the Lignox®
process developed by Eka Nobel (50).
7-25
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
The peroxide charge required for a full peroxide stage is approximately 2.5 percent
on pulp. To fully use the peroxide charge and achieve high brightness requires a temperature of
between 70 and 90
decrease pulp viscosity (49).
Peroxide bleaching has been demonstrated in full-scale applications at both kraft
and sulfite mills outside the U.S. The capital cost of implementing peroxide bleaching is minimal,
assuming use of existing bleaching towers. Because the unit cost for hydrogen peroxide is higher
than for chlorine, operating costs for peroxide bleaching may be higher depending upon the
amount of peroxide used. The cost of the large amount of peroxide that is necessary to bleach a
pulp to full brightness has limited the use of peroxide bleaching.
7.3.9 Totally Chlorine-Free Bleaching of Papergrade Kraft Pulps
Totally chlorine-free bleaching is performed without the use of chlorine, sodium or
calcium hypochlorite, chlorine dioxide, chlorine monoxide, or any other chlorine-containing
compound. TCP bleaching is performed using a combination of enzymes, ozone, oxygen, and/or
peroxide. In the last several years, numerous TCP bleaching processes have been developed and
are now used at bleached papergrade kraft mills worldwide.
Section 8.3.11 of the TDD reported briefly the status of TCP bleaching at
papergrade kraft mills in 1993. At that time about 15 mills worldwide were making some TCP
bleached pulp but few of the mills were dedicated to TCP production due to a lack of market
demand. Also, most of the TCP pulps were bleached to a lower brightness than market kraft
grades (75-80 ISO vs 88-90 ISO). Therefore, EPA concluded that TCP was not an available
pollution prevention technology because of limited worldwide experience with this process and a
lack of data for TCP bleaching of softwood to full market brightness.
EPA received many comments that it should continue to investigate TCP bleaching
because dioxin and furan are not generated at any level with TCP bleaching, thus assuring that
these pollutants are not released to the environment. The Agency conducted two sampling
programs at the one U.S. mill that produces softwood TCP bleached kraft pulp. EPA collected
samples of bleach plant filtrates but could not collect samples of treated effluent because the mill
does not employ secondary treatment. The Agency also conducted a sampling program at a
Nordic mill that produces hardwood and softwood kraft pulp on two bleach lines that alternate
between elemental chlorine-free (ECF) and TCP bleaching. Samples collected at this mill could
not be used to characterize treated TCP bleaching effluents, because the TCP bleaching effluents
are combined with ECF bleaching effluents for treatment.
Both of the sampled softwood TCP fiber lines employed oxygen delignification
followed by multiple stages of peroxide bleaching. The U.S. mill's unbleached pulp kappa
number was between 7 to 10. The bleach sequence for the U.S. mill is QEopPPPS (51). Q
represents an acidic chelant stage, followed by an enhanced extraction stage (Eop), three alkaline
peroxide stages (PPP), and the addition of sodium bisulfite (S). Bleached pulp brightness was
7-26
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
approximately 79 during the first sampling episode at the U.S. mill, but by the time of the second
sampling episode pulp brightness had increased to 83 ISO. The Nordic mill uses extended
cooking, and was able to reduce the lignin content of unbleached pulp to a low kappa number of
four. The Nordic mill bleach sequence is QPPP. At the time of sampling, this mill bleached pulp
to a brightness of 83 ISO.
At neither mill was chloroform or chlorinated phenolic pollutants detected in
samples collected by EPA. At the U.S. mill, dioxin, furan, and AOX were not detected above the
analytical minimum level during sampling fully representative of TCP operations. The average
bleach plant AOX loading measured by EPA at the Nordic mill was 0.002 kg/ADMT (compared
to a long-term average of 0.51 kg/ADMT for Option A). EPA's dioxin sampling results for the
Nordic mill were surprising. Dioxin was detected at a concentration just above the minimum level
in one sample of combined bleach plant filtrate, when the mill was bleaching without the use of
chlorine or any chlorinated compounds. Furan was not detected. EPA believes the dioxin results
were unique to the operation of this mill and does not conclude that TCP bleaching generates
dioxin.
Neither of the two sampled mills produced softwood pulp at full market
brightness. By the end of 1996, two mills were producing exclusively TCP pulp that was fully
bright and fully strong (see Section 8.7.2 of this document). Data in EPA's record are insufficient
to confirm that TCP bleaching processes are technically demonstrated for the full range of
products made with bleached kraft pulp. Despite these impediments, EPA believes that the
progress being made in TCP process development is substantial, and that additional data may
demonstrate that TCP processes are indeed available for the full range of market products. TCP
mills will qualify for at least Tier I of the Voluntary Advanced Technology Incentives Program.
7.3.10 Totally Chlorine-Free Bleaching of Papergrade Sulfite Pulps
Section 8.3.14 of the TDD reported briefly the status of TCP bleaching at
papergrade sulfite mills in 1993. At that time at least 10 mills worldwide were making TCP
bleached sulfite pulp. EPA visited one of these mills to collect bleach plant effluent samples. As
of 1995, two U.S. sulfite mills were using TCP bleaching processes.
The bleaching sequences at most of the TCP papergrade sulfite mills are based on
oxygen delignification, followed by one or more peroxide bleaching stages. Many of these mills
describe their oxygen delignification stage as an enhanced extraction stage (Eop); however, in form
and function it is the same as oxygen delignification (i.e., a pressurized tower is used to introduce
oxygen to lower the pulp lignin content). Further delignification, or bleaching, is performed with
one or more peroxide stages. The peroxide stages are operated at consistencies ranging from 12
to 30 percent; peroxide charges vary between 30 and 40 kg/ADMT. To prevent side reactions of
metal ions in the peroxide stages, some mills use acid washes or add chelating agents before,
between, or in the peroxide stages. Some mills also add sodium silicate or nitrilamine to peroxide
stages to further brighten the pulp.
7-27
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
The furnish used by papergrade sulfite mills using TCP bleaching include a variety
of hardwoods (predominately birch and beech) and softwoods (mostly spruce). These TCP sulfite
mills also make a variety of products including market pulp, tissue, and printing and writing
grades. Bleached pulp properties vary by product. Most mills report brightnesses near 85 ISO;
the range reported is 70 to 90 ISO.
7.4 End-of-Pipe Wastewater Treatment Technologies
This section describes the BAT component of efficient biological treatment. More
detailed information about the wastewater treatment processes in use at pulp and paper mills is
presented in Section 8.5 of the TDD.
According to EPA's 1990 Census Questionnaire, more than 80 percent of direct
discharging pulp and paper mills in the United States use primary and secondary (biological)
wastewater treatment, while only 2 percent use tertiary treatment. Indirect discharging mills
discharge to POTWs, which also use secondary wastewater treatment.
Primary treatment is the removal of suspended solids. Primary treatment may also
include other pre-biological treatment processes such as equalization, neutralization, or cooling.
Secondary treatment involves a biological process to remove organic matter through biochemical
oxidation. In the pulp and paper industry, activated sludge systems and aerated/non-aerated basin
systems are the most commonly used biological processes. Tertiary treatment is advanced
treatment, beyond secondary, to remove particular contaminants. Examples of tertiary treatment
are the removal of phosphorus by alum precipitation and removal of toxic refractory organic
compounds by activated carbon adsorption.
Common elements of wastewater treatment, as practiced in the pulp and paper
industry, include (but are not limited to):
Primary sedimentation;
Neutralization;
Equalization;
Precooling;
Nutrient addition;
Aeration;
Multi-basin systems, some of which act as polishing ponds;
Mixed or hybrid treatment systems (activated sludge and basin systems
operated in series or parallel); and
7-28
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
Addition of flocculants to secondary clarifiers to improve settling.
As noted in Section 8.5 of the TDD, the operation of treatment systems is important to achieving
optimum effluent quality, particularly the activated sludge systems which require careful day-to-
day attention to numerous important operating parameters. EPA defined efficient biological
treatment as that which removes 90 percent or more of the influent BOD5. Most papergrade kraft
and sulfite mills are achieving 90 percent BOD5 removal (see TDD Table 9-8 based on data from
the 1990 Census Questionnaire data). NCASI also reported greater than 90 percent BOD5
removals for direct discharging mills and for POTWs receiving chemical pulp mill wastewater.
7.5 References
1. Pollution Prevention Opportunity Assessment and Implementation Plan for
Simpson Tacoma Kraft Company. Tacoma. Washington. EPA 910/9-92-027, U.S.
EPA Region 10, August 1992.
2. Model Pollution Prevention Plan for the Kraft Segment of the Pulp and Paper
Industry. EPA 910/9-92-030, U.S. EPA Region 10, September 1992.
3. Pollution Prevention for the Kraft Pulp and Paper Industry. Bibliography. EPA
910/9-92-031, U.S. EPA Region 10, September 1992.
4. Summary of Technologies for the Control and Reduction of Chlorinated Organics
from the Bleached Chemical Pulping Subcategories of the Pulp and Paper Industry.
U.S. Environmental Protection Agency, Washington DC, April 27, 1990.
5. Pollution Prevention Technologies for the Bleached Kraft Segment of the U.S.
Pulp and Paper Industry. EPA/600/R-93/100, U.S. Environmental Protection
Agency, Washington DC, August 1993.
6. Development Document for Proposed Effluent Limitations Guidelines and
Pretreatment Standards for the Pulp. Paper, and Paperboard Point Source
Category. EPA-821-R-93-019, U.S. EPA, Washington DC, October 1993.
7. Confidential Mill Visit Report. Report prepared by Radian Corporation for EPA.
Record Section 7.4, DCN 07654, October 29, 1992.
8. McCubbin, N., H. Edde, E. Barnes, J. Folke, E. Bergman, and D. Owen. Best
Available Technology for the Ontario Pulp and Paper Industry. ISBN 0-7729-
9261-4. Ontario Ministry of the Environment, Canada, February 1992.
7-29
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
9. Allen, L.H., R.M. Berry, B.L. Fleming, C.E. Luthe, and R.H. Voss. "Evidence
That Oil-Based Additives are Potential Indirect Source of the TCDD and TCDF
Produced in Kraft Bleach Plants" In: Proceedings for the Eight International
Symposium on Chlorinated Dioxins and Related Compounds. Umea, Sweden,
August 21-26, 1988.
10. Voss, R.H., C.E. Luthe, B.L. Fleming, R.M. Berry, and L.H. Allen. "Some New
Insights into the Origins of Dioxins Formed During Chemical Pulp Bleaching" In:
Proceedings for the 1988 CPPA Environment Conference. Vancouver, BC,
Canada, October 25-26, 1988.
11. Cully, T.G. and S.C. Cohen. "Lowering DBF and DBD Levels in Defoamer Oil"
In: Proceedings of the TAPPI 1990 Pulping Conference. Toronto, Ontario,
Canada, October 14-17, 1990. pp. 981-983.
12. Twomey, L.F. "New Oil-Free Brownstock Washer Defoamer Can Help Mills
Reduce Dioxin Related Problems" In: Proceedings of the TAPPI 1990 Pulping
Conference. Toronto, Ontario, Canada, October 14-17, 1990. pp. 984-987.
13. Rempel, W., D.C. Pype, and M.D. Ouchi. Mill Trials of Substantial Substitution
of Chlorine Dioxide for Chlorine-Part III: Medium Consistency. Unpublished.
14. Harder, N. "Extended Delignification in Kraft Cooking." Svensk Paperstidning.
81(15):483, 1978.
15. Macleod, M. "Extended Delignificati on in Kraft Mills Science, Technology, and
Installations Worldwide" Presented at: 1992 Conference on Emerging
Technologies in Non-Chlorine Bleaching. Hilton Head, South Carolina, March 2-
5, 1992.
16. European Environmental Research Group (MFG) and O. Duoplan. Chlorine
Dioxide in Pulp Bleaching — Technical Aspects and Environmental Effects.
Literature Study. CEFIC, Sodium Chlorate Sector Group, Brussels, Belgium,
February 1993.
17. Galloway, L.R., P.I. Helminen, D.N. Carter. "Industry's Effluent Problems Spawn
New Engineering Technology, Design" In: Pulp & Paper. September 1989. pp.
91-97.
18. Allen, L. "Pitch Control in Pulp Mills" In: PAFRICAN Report. MR 344,
December 1996.
7-30
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
19. Smook, G.A. Handbook for Pulp & Paper Technologists. Joint Textbook
Committee of the Paper Industry, TAPPI, Technology Park, Atlanta, Georgia and
CPPA, Montreal, Quebec, Canada, 1982 and reprints until 1989.
20. Grace, T.M., B. Leopold, E.W. Malcolm, and MJ. Kocurek, eds. Pulp and Paper
Manufacture: Volume 5 - Alkaline Pulping. Joint Textbook Committee of the
Paper Industry, TAPPI, Technology Park, Atlanta, Georgia and CPPA, Montreal,
Quebec, Canada, 1989.
21. Analysis of Impacts of BAT Options on the Kraft Recovery Cycle. Prepared by
Eastern Research Group and N. McCubbin for EPA. Record Section 23.1.2, DCN
14490, 1997.
22. Pulp. Paper, and Paperboard Industry - Background Information for Proposed Air
Emission Standards (Manufacturing Processes at Kraft. Sulfite. Soda, and Semi-
Chemical Mills). EPA 453/R93-050a, U.S. Environmental Protection Agency,
Research Triangle Park, North Carolina, October 1993.
23. Springer, A.M. Industrial Environmental Control-Pulp and Paper Industry. John
Wiley & Sons, Inc., New York, New York, 1986.
24. Technical Support Document for Best Management Practices for Spent Pulping
Liquor Management. Spill Prevention, and Control. EPA, Washington DC,
Record Section 30.9, DCN 14489, 1997.
25. Colodette, J.L., U.P. Singh, A.K. Ghosh, and R.P. Singh. "Ozone Bleaching
Research Focuses on Reducing High Cost, Poor Selectivity." Pulp and Paper.
67(6): 139-147, June 1993.
26. Henricson, K. "New Generation Kraft Pulping and Bleaching Technology".
PaperijaPuu. 74(4), 1992.
27. Shackford, L.D. "Commercial Implementation of Ozone Bleaching Technology."
Presented at: 1992 Conference on Emerging Technologies in Non-Chlorine
Bleaching. Hilton Head, South Carolina, March 2-5, 1992.
28. Nutt, W.E., B.F. Griggs, S.W. Eachus, and M.A. Pikulin. "Developing an Ozone
Bleaching Process." TAPPI Journal. 76(3): 115-123, March 1993.
29. Fredstrom, C., K. Idner, and C. Hastings. "Current State-of-the-Art of E0, Ep and
Epo Technologies" Presented at: 1992 Conference on Emerging Technologies in
Non-Chlorine Bleaching. Hilton Head, South Carolina, March 2-5, 1992.
7-31
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
30. Pryke, D.C., and D.W. Reeve, "Chlorine Dioxide Delignification Practices in
Canada" In: Proceedings of the 1996 Pulping Conference. TAPPI, 1996.
31. Fredette, M.C. "New C1O2 Generation Capacity" In: Proceedings of the TAPPI
1990 Bleach Plant Operations Short Course. Hilton Head, South Carolina, June
17-22, 1990. pp. 159-168.
32. Luthe, C.E., P.E. Wrist, and R.M. Berry. "An Evaluation of the Effectiveness of
Dioxins Control Strategies on Organochlorine Effluent Discharges from the
Canadian Bleached Chemical Pulp Industry." Pulp & Paper Canada. 93(9):40-49,
1992.
33. Data Available for Limitations Development for Toxic and Nonconventional
Pollutants. EPA, Washington DC, Record Section 22.6, DCN 14494, 1997.
34. Hastings, C. and K. Idner. "Current State-of-the-Art of Eo, Ep and Eop
Technologies." Presented at: 1992 Conference on Emerging Technologies in
Non-Chlorine Bleaching. Hilton Head, South Carolina, March 2-5, 1992.
35. O'Reardon, D. "Review of Current Technology Vital to Bleach Plant
Modernization Study." Pulp & Paper. 66(4): 124-129, April 1992.
36. Crawford, R.J., M.N. Stryker, S.W. Jett, W.L. Carpenter, R.P. Fisher, and A.K.
Jain. "Laboratory Studies of Chloroform Formation in Pulp Bleaching." TAPPI
Journal. November 1987. pp. 123-128.
37. Crawford, R.J., V.J. Dallons, A.K. Jain, and S.W. Jett. "Chloroform Generation at
Bleach Plants with High Chlorine Dioxide Substitution or Oxygen Delignification."
TAPPI Journal. 74(4): 159-163, April 1991.
38. Shariff, A., U. Ahlborg, P. Axegard, C. Rappe, and A. Van Heiningen. "An
Assessment of the Formation of 2,3,7,8-TCDD and 2,3,7,8-TCDF when Chlorine
Dioxide is the Oxidizing Agent in the First Stage of Bleaching of Chemical Pulp."
Submitted in comments on EPA's July 1996 Notice of Data Availability, DCN
25539A2, January 1996.
39. Comment Response Document. Volume I, "Justification for Establishing
Limitations and Standards for AOX." Record Section 30.11, DCN 14497, 1997.
40. Reeve, D.W. "Chlorine Dioxide in Delignification." In: Pulp Bleaching. Principles
and Practice. Carlton W. Dence and Douglas W. Reeve, ed. TAPPI Press,
Atlanta, Georgia, 1996.
7-32
-------
Section 7 - Pollution Prevention and Wastewater Treatment Technologies
41. Conner, T. J. "Panel 2 - Low Kappa Factor Bleaching - Plant Experience." In:
Proceedings TAPPI 1996 Pulping Conference. Vol 1:301, 1996.
42. Rappe, C., S. Swanson, B. Glass, K.P. Kringstad, F. DeSousa, L. Johansson, and
Z. Abe. Pulp and Paper Canada. 90(8):42, 1989.
43. Hise, R.G. and H.L. Hintz. "The Effect of Brown Stock Washing on the
Formation of Chlorinated Dioxins and Furans During Bleaching." TAPPI Journal.
73(1): 185-190, 1990.
44. Hrutfiord, B. F. and A.R. Negri. "Chlorinated Dibenzofurans and Dibenzodioxins
from Lignin Models." TAPPI Journal. 75(8): 129, 1992.
45. Berry, R.M., B.I. Fleming, R.H. Voss, C.E. Luthe, and P.E. Wrist. "Toward
Preventing the Formation of Dioxins During Pulp Bleaching." Pulp and Paper
Canada. 90(8) 48, 1989.
46. Berry, Richard M. "Dioxins and Furans in Effluents, Pulp, and Solid Waste" In:
Pulp Bleaching. Principles and Practice. Carlton W. Dence and Douglas W.
Reeve, ed. TAPPI Press, Atlanta, Georgia, 1996.
47. Hrutfiord, B.F., and A.R. Negri, "Dioxin Sources and Mechanisms During Pulp
Bleaching." Chemosphere. 25(l-2):53-56, 1992.
48. Farrell, R.L. "Status of Enzyme Bleaching R&D and Mill Work." Presented at:
1992 Conference on Emerging Technologies in Non-Chlorine Bleaching. Hilton
Head, South Carolina, March 2-5, 1992.
49. Anderson, R. Peroxide Delignification and Bleaching. Presented at: 1992
Conference on Emerging Technologies in Non-Chlorine Bleaching. Hilton Head,
South Carolina, March 2-5, 1992.
50. Basta, J., L. Andersson, W. Hermansson. Lignox and Complementary
Combinations. Presented at: 1992 Conference on Emerging Technologies in Non-
Chlorine Bleaching. Hilton Head, South Carolina, March 2-5, 1992.
51. Young, J. "Louisiana-Pacific's Samoa Mill Establishes TCF Pulp Production".
Pulp & Paper. 67(8):61-63, August 1993.
7-33
-------
100.000
80.000
60.000
40.000
20.000
0
'ADM/toy
WorMwtatoToul
Figure 7-1. U.S. and Worldwide Increase in Kraft Pulp Produced by Extended Cooking, 1983-1992
-------
CNp am
Bin Actlvdw
Figure 7-2. Extended Cooking Continuous Digester System (EMCC®)
Courtesy of Kamyr, Inc., Glens Falls, New York.
-------
150.000
125,000
100.000
75,000
50.000
25,000
0
ADMt/day
(M
Worldwtd* Total
8 I i I i i i
Figure 7-3. U.S. and Worldwide Increase in Kraft Pulp Produced by Oxygen Delignification, 1970-1992
-------
LO
Oxygen
Mixer
Vent
Post
Oxygen
Washers
Brownstock
JBrownstock
Washer
To Washers
and Recovery
Medium
Consistency
Pump
'Wash Water
To Chlorination
OXY8LMC.DWC
Figure 7-4. Typical Medium-Consistency Oxygen Delignification System
-------
NoOH
Vent
^1
oo
To Washers
and Recovery
Brown stock
Washer
Figure 7-5. Typical High-Consistency Oxygen Delignification System
-------
Vent to
Control Device
u>
Stream From
Rafl<«T«nk
Vent to
Control Davica
Raflux Stream
ToWMtcwaier
TrMtnwnt
StrlpiMrFMd
FMoVBottomt
Heat Exchanger
Steam
Stripping
Column
Staam
Figure 7-6. Continuous Steam Stripper System
-------
Section 8 - Development of Control and Treatment Options
8.1
SECTION 8
DEVELOPMENT OF CONTROL AND TREATMENT OPTIONS
Introduction
This section describes the combinations of pulping and bleaching technologies,
in-process water conservation practices, and end-of-pipe wastewater treatment that the Agency
configured as technology options for consideration as bases for the following regulations:
BPT (best practicable control technology currently available);
BCT (best conventional pollutant control technology);
BAT (best available technology economically achievable);
NSPS (new source performance standards);
PSES (pretreatment standards for existing sources); and
PSNS (pretreatment standards for new sources).
These regulations establish quantitative limits on the discharge of pollutants from
industrial point sources. As explained in the preamble and in Section 12 of this document, EPA
decided not to promulgate the proposed regulations for BPT and BCT. BPT and BCT limitations
for the pulp and paper industry, therefore, are based on the formerly promulgated BPT and BCT
limitations. EPA is promulgating BAT, NSPS, PSES, and PSNS today. The applicability of the
various regulations is summarized below:
BPT
BCT
BAT
NSPS
PSES
PSNS
Direct
Discharge
X
X
X
X
Indirect
Discharge
X
X
Existing
Source
X
X
X
X
New
Source
X
X
Conventional
Pollutants
X
X
X
Toxic and
Nonconventional
Pollutants
X
X
X
X
All of these regulations are based on the performance of specific technologies but
do not require the use of any specific technology. The regulations applicable to direct
dischargers are effluent limitations guidelines which are applied to individual facilities through
NPDES permits issued by EPA or authorized states under Section 402 of the CWA. The
regulations applicable to indirect dischargers are standards, and are administered by local
permitting authorities (i.e., the government entity controlling the POTW to which the industrial
8-1
-------
Section 8 - Development of Control and Treatment Options
wastewater is discharged). The pretreatment standards are designed to control pollutants that
pass through or interfere with POTWs.
8.2 Toxic and Nonconventional Pollutants
The toxic and nonconventional pollutants consist of AOX, chloroform, 2,3,7,8-
TCDD, 2,3,7,8-TCDF, and 12 chlorinated phenolic compounds. Regulations are being
promulgated for these pollutants for mills in the Bleached Papergrade Kraft and Soda (Subpart
B) and Papergrade Sulfite (Subpart E) subcategories.
8.2.1 Bleached Papergrade Kraft and Soda Subcategory (BPK)
EPA developed four options to reduce wastewater discharges of toxic and
nonconventional pollutants. Two options focused on elemental chlorine-free bleaching
technologies and two focused on totally chlorine-free bleaching technologies. The ECF options
are identified as Option A and Option B. The TCP options are identified as TCF-peroxide and
TCF-ozone. Each option is described in the following sections.
8.2.1.1 BPK Option A
Option A consists of conventional pulping followed by complete substitution of
chlorine dioxide for elemental chlorine, as well as the nine elements identified below:
(i) Adequate chip thickness control;
(ii) Closed brown stock pulp screen room operation, such that screening
filtrates are returned to the recovery cycle;
(iii) Effective brown stock washing, i.e., washing that achieves a soda loss of
less than or equal to 10 kg Na2SO4 per ADMT of pulp (equivalent to 99
percent recovery of pulping chemicals from the pulp);
(iv) Use of TCDD- and TCDF-precursor-free defoamers (water-based
defoamers or defoamers made with precursor-free oils);
(v) Elimination of hypochlorite, i.e., replacement of hypochlorite with
equivalent bleaching power in the form of additions of peroxide and/or
oxygen to the first extraction stage and/or additional chlorine dioxide in
final brightening stages;
(vi) Use of strategies to minimize kappa factor and TCDD- and TCDF-
precursors in brown stock pulp;
8-2
-------
Section 8 - Development of Control and Treatment Options
(vii) High shear mixing to ensure adequate mixing of pulp and bleaching
chemicals;
(viii) Oxygen and peroxide enhanced extraction, which allows elimination of
hypochlorite and/or use of a lower kappa factor in the first bleaching stage;
and
(ix) Efficient biological wastewater treatment, achieving removal of 90 percent
or more of influent BOD5.
8.2.1.2 BPK Option B
Option B is identical to Option A, with the addition of extended delignification
(oxygen delignification and/or extended cooking). In a slight change from the definition of the
proposed option, EPA has defined Option B not only in terms of the presence of extended
delignification technology (i.e., oxygen delignification or extended cooking) but also by the pre-
bleaching kappa number achieved by extended delignification. EPA defines extended
delignification as the operation of such technologies that result in a kappa number of 20 or less
for softwoods and less than 13 for hardwoods. Sections 7.2.3 and 7.2.6 contain more detailed
descriptions of each technology.
8.2.1.3 TCF-Peroxide
EPA evaluated a peroxide-based TCP bleaching option which is performed
without the use of chlorine, sodium hypochlorite, calcium hypochlorite, chlorine dioxide,
chlorine monoxide, or any other chlorine-containing compound. This option contains all but two
of the elements listed for Option A. The two elements not included in this option are: 1) use of
dioxin- and furan-precursor-free defoamers, and 2) strategies to minimize kappa factor and
TCDD- and TCDF-precursors in brown stock pulp because the option uses peroxide, and not
chlorine-based, bleaching. For the purpose of estimating costs, the TCF-Peroxide option
included anthraquinone pulping and oxygen delignification to achieve unbleached pulp kappa
numbers of 10 for softwood and 6 for hardwood.
8.2.1.4 TCF-Ozone
EPA evaluated an oxygen-based TCF bleaching option which is performed
without the use of chlorine, sodium hypochlorite, calcium hypochlorite, chlorine dioxide,
chlorine monoxide, or any other chlorine-containing compound. This option contains all but two
of the elements listed for Option A. The two elements not included in this option are: 1) use of
dioxin- and furan-precursor-free defoamers, and 2) strategies to minimize kappa factor and
TCDD- and TCDF-precursors in brown stock pulp because the option uses ozone and peroxide,
and not chlorine-based, bleaching. For the purpose of estimating costs, TCF-ozone option
included anthraquinone pulping and oxygen delignification to achieve unbleached pulp kappa
numbers of 10 for softwood and 6 for hardwood.
8-3
-------
Section 8 - Development of Control and Treatment Options
8.2.2 Papergrade Sulfite Subcategory (PS)
EPA had proposed a TCP bleaching option for all PS mills. Some commenters
questioned the achievability of TCP bleaching to produce all the existing PS products. In
response to these comments, EPA divided the PS subcategory into three segments:
A) Calcium, magnesium, or sodium sulfite pulping;
B) Ammonium sulfite pulping; and
C) Production of pulp and paper at specialty-grade sulfite mills. These mills
produce 25 percent or more pulp with a high percentage of alpha cellulose
and high brightness for end products such as plastic molding compounds,
saturating and laminating products, and photographic papers or these mills
produce 50 percent or more pulp with 91 ISO brightness and above.
After the proposal, EPA focused on one option for each segment. Each option has
the following elements in common:
(i) Use of TCDD- and TCDF-precursor-free defoamers (water-based
defoamers or defoamers made with precursor-free oils);
(ii) For segments with ECF bleaching, elimination of hypochlorite, i.e.,
replacement of hypochlorite with equivalent bleaching power in the form
of additions of peroxide and/or oxygen to the first extraction stage and/or
additional chlorine dioxide in final brightening stages; and
(iii) Efficient biological wastewater treatment.
Segment A (calcium, magnesium, or sodium sulfite) includes:
Totally chlorine-free bleaching (bleaching with peroxide);
Oxygen and peroxide enhanced extraction; and
Improved pulp cleaning.
Segment B (ammonium sulfite) includes:
Complete substitution of chlorine dioxide for chlorine;
Peroxide enhanced extraction; and
High shear mixing.
Segment C (specialty-grade sulfite) includes:
Complete substitution of chlorine dioxide for chlorine;
8-4
-------
Section 8 - Development of Control and Treatment Options
Oxygen and peroxide enhanced extraction; and
High shear mixing.
Thus, EPA has developed one option for each PS subcategory segment.
8.2.3 Point of Compliance Monitoring
EPA is requiring mills in Subparts B and E to demonstrate compliance with
effluent limitations guidelines and standards on dioxin, furan, 12 chlorinated phenolic pollutants,
and (for Subpart B mills) chloroform inside the discharger's facility at the point where the
wastewater containing those pollutants leaves the bleach plant. EPA is also requiring indirect
dischargers in Subpart B to demonstrate compliance with AOX pretreatment standards at the
bleach plant. For direct dischargers, EPA is authorized by the Clean Water Act and EPA's
regulations at 40 CFR Parts 122.44(1), 122.45(h), and 125.3(e) to specify an in-plant point of
compliance monitoring for technology-based limitations. For indirect dischargers, EPA relies on
40 CFR Part 403.6(d), which prohibits dilution to achieve a categorical pretreatment standard.
Hereafter, EPA refers to the limitations and standards for which compliance must be
demonstrated in-plant as "in-plant limitations."
As set forth in more detail below, EPA is establishing in-plant limitations on
bleach plant effluent because limitations imposed on those pollutants at the point of discharge are
impractical and infeasible as measures of the performance of process technologies representing
the technology-based levels of control. Moreover, in-plant effluent limitations are consistent
with the MACT standards for chloroform, which independently require achievement of BAT
limitations on dioxin, furan, chloroform, the 12 chlorinated phenolic pollutants, and (for indirect
dischargers) AOX at the bleach plant in order to ensure that the removals represented by the
MACT technology floor — complete substitution of chlorine dioxide for elemental chlorine and
elimination of hypochlorite — are attained.
EPA's data show that mills using the model BAT, NSPS, PSES and PSNS
technologies for Subparts B and E are able to achieve at the bleach plant concentrations of dioxin
and the 12 chlorinated phenolic pollutants at levels below the minimum levels of currently
available analytical methods. Furan concentrations, in turn, are very near the analytical minimum
levels at Subpart B mills using the applicable model technologies and are below the minimum
level for Subpart E mills using the model technologies corresponding to the segments in that
Subpart. (At the end of the pipe, furan cannot be detected by available analytical methods in the
effluent of many Subpart B mills.)
Because only 10 to 40 percent of the wastewater discharged by mills in Subpart B
originates in the bleach plant, the concentrations of pollutants in the mills' final effluent would
be one-tenth to two-fifths of their concentrations at the bleach plant. Substantial dilution can
also occur at Subpart E mills. In the biological wastewater treatment system, the pollutants may
be present but in concentrations below the applicable analytical minimum levels. When they are
discharged to receiving streams, however, dioxin and furan bioaccumulate in aquatic organisms.
8-5
-------
Section 8 - Development of Control and Treatment Options
Were EPA to allow compliance monitoring of the final effluent, no way would be available to
determine whether the bleach plant effluent has been adequately controlled, or whether the
effluent has simply been diluted to below the analytical minimum level by other flows. Diluting
pollutants in this manner rather than preventing their discharge is inconsistent with achieving the
removals represented by the technology-based levels of control, and hence with the purpose of
the resulting limitations and standards. Dilution is also inconsistent with the goals of the Clean
Water Act in general. Sections 101(a) and 301(b)(2)(A). While no mill is required to install
EPA's model technologies, establishing limitations at the bleach plant is the only way EPA can
ensure that none of these pollutants will be discharged at concentrations greater than the levels
achievable through implementation of the best available technology. See E.I, du Pont de
Nemours & Co. v. Train. 430 U.S. 112, 129(1977).
With respect to the 12 chlorinated phenolic pollutants, EPA acknowledges that
these pollutants could be degraded by biological treatment of the facility's combined wastewater.
However, the same process technologies necessary to address dioxin and furan also reduce the
levels of chlorinated phenolic pollutants to concentrations below minimum levels at the bleach
plant. Commenters have supplied no data showing that the chlorinated phenolic pollutants
should or indeed, as a practical matter, could be segregated from the dioxin- or furan-bearing
wastestreams in order to fully use a mill's secondary treatment system. Nor is there any
assurance that BAT limitations for these pollutants, if monitored at the end of the pipe, would be
achieved by treatment rather than simply by the effects of dilution. See 40 CFR Part 122.45(h).
Thus, EPA concludes that it is appropriate to require compliance monitoring for the limitations
and standards on the 12 chlorinated phenolic pollutants at the point they most easily can be
achieved and measured — at the bleach plant.
In the case of chloroform, in-plant limits at Subpart B mills are authorized by 40
CFR Part 122.45(h) (for direct discharging mills) and 40 CFR Part 403.6(d) (for indirect
discharging mills) because they offset the effects of dilution, in this case, the occurrence of
uncontrolled volatilization. (Section 403.6(d) prohibits indirect dischargers from employing
dilution as a substitute for treatment.) Limitations and standards at the point of effluent
discharge are impractical and infeasible because chloroform would be lost as air emissions in
wastewater conveyances and treatment facilities (e.g., collection boxes and aeration tanks). (For
a discussion of the volatility of chloroform at Subpart B and E mills, see DCN 03815 and DCN
09323.) Such incidental air stripping not only would cause chloroform to be present in the final
effluent at levels below detection (thus complicating compliance monitoring), but it also would
be inconsistent with the model technology EPA has used as a basis for its limits. See CWA
Section 301(b)(2)(A), 304(b)(2)(A) & (B). As is the case with dilution, EPA would have no way
of knowing whether reductions in wastewater discharges are being achieved by application of
BAT- or NSPS-level technologies or by air emissions (or dilution) in the wastewater conveyance
and treatment facilities. In-plant limitations for chloroform also allow EPA to evaluate the
environmental effectiveness of each mill's treatment and process technologies, information that
could assist EPA at a later date in revising these effluent limitations guidelines and standards as
the Clean Water Act contemplates. See CWA Section 308(a)(l). Moreover, in other regulatory
contexts, EPA recognizes that dilution includes not only mixing a pollutant of concern with other
8-6
-------
Section 8 - Development of Control and Treatment Options
wastestreams, but also mixing it with excess air in the form of uncontrolled volatilization. See
52 FR 25760, 25778-79. Volatilization, like dilution, does nothing to remove, destroy, or
immobilize pollutants, and for this reason is not in itself a form of treatment. Id. at 25779. The
policy reasons supporting that principle in the hazardous waste context similarly apply here.
Finally, EPA is setting effluent limitations and standards at the bleach plant in
order to avert the non-water quality environmental impacts caused by the volatilization of
chloroform to the air and in order to be consistent with its Clean Air Act determination that the
MACT floor for chloroform consists of bleach plant process modifications, i.e., complete
chlorine dioxide substitution and elimination of hypochlorite as bleaching agents. Specifically,
EPA is requiring under the Clean Air Act that chloroform emissions be controlled by complying
with the BAT requirements for all regulated pollutants. See 40 CFR Part 63.445(d). Therefore,
EPA has determined under its Clean Air Act authority that bleach plant technologies — and
bleach plant limitations on dioxin, furan, chloroform, the 12 chlorinated phenolics, and (for
indirect dischargers) AOX — are necessary to regulate air emissions of chloroform. The situation
presented here is very different from the situation EPA faced when promulgating effluent
limitations guidelines and standards for the organic chemicals, plastics, and synthetic fibers
industrial category in 1987. See 52 FR 42522, 42658-62. In that rulemaking, the issue before
EPA was whether to use in-plant limitations and standards to regulate air emissions of certain
volatile and semi-volatile pollutants; EPA chose not to set in-plant requirements for that purpose
because it determined that the regulation of such emissions was best accomplished in a Clean Air
Act proceeding, which EPA was commencing at that time. See id. at 42560-62. In contrast,
EPA in this rulemaking integrated its decision-making under the Clean Water Act and the Clean
Air Act expressly to address these cross-media issues. (Indeed, the statutory provisions
authorizing establishment of BAT limitations and NSPS, PSES, and PSNS specifically require
EPA to consider non-water quality environmental impacts in promulgating such limitations and
standards.) Taking into account both the air and water objectives of these Cluster Rules, EPA
therefore concludes that it is highly appropriate for EPA to set effluent limitations and standards
under the Clean Water Act to correspond to and support its concurrent regulation of air emissions
under the Clean Air Act.
As noted above, EPA is requiring indirect dischargers subject to Subpart B to
demonstrate compliance with AOX pretreatment standards at the bleach plant. (EPA is not
specifying a point of compliance monitoring for AOX for direct discharging mills in Subpart B in
this rule.) Like the pretreatment standards for dioxin, furan, chloroform, and the 12 chlorinated
phenolic pollutants, the pretreatment standards for AOX are based exclusively on process
changes. EPA expects that removals achieved by indirect dischargers employing the PSES or
PSNS model technology (or its equivalent), in combination with removals achieved by biological
treatment systems at POTWs, will be comparable to the AOX removals achieved by direct
dischargers complying with BAT limitations or NSPS. Because biological treatment is not part
of the PSES or PSNS model technology, EPA concluded that an end-of-pipe standard for AOX
would simply reflect the effects of dilution, not treatment. Dilution is expressly prohibited by
EPA's pretreatment regulations as a substitute for treatment. See 40 CFR Section 403.6(d).
Therefore, EPA determined that the only practicable way of enforcing that provision and,
8-7
-------
Section 8 - Development of Control and Treatment Options
incidentally, measuring the effectiveness of a discharger's chosen PSES and PSNS technology is
to require mills to demonstrate compliance with the AOX pretreatment standard at the bleach
plant.
EPA received comments inquiring why EPA believed it was infeasible to measure
compliance with a limitation of below the minimum level (or "
-------
Section 8 - Development of Control and Treatment Options
presenting long-term risks to human health, aquatic life and wildlife; consequently, each unit of
mass removed from wastewater effluent is beneficial to the environment. For these reasons, EPA
believes it has acted reasonably in interpreting the Clean Water Act to require dischargers to
demonstrate compliance with technology-based limitations and standards at the bleach plant in
this instance.
EPA exercised this authority in the effluent limitations guidelines and standards
EPA promulgated in 1984 for the nonferrous metals manufacturing point source category
(secondary aluminum smelting subcategory) as can be found in 40 CFRPart 421.32(e) (BAT),
Part 421.34(e) (NSPS), Part 421.35(e) (PSES), Part 421.36(e) (PSNS); see 49 FR 8742, 8758-59.
Under those regulations, dischargers must demonstrate compliance with the BAT limitations,
new source performance standards, and pretreatment standards for total phenolics "at the source,"
which EPA defined as "at or before the commingling of delacquering scrubber liquor blowdown
with other process or nonprocess wastewater." 40 CFR Part 421.3 l(c); see 49 Fed. Reg. at 8758-
59. EPA based this decision on the possibility of significant dilution. See 49 Fed. Reg. at 8758-
59. As explained in the preamble of that rule, EPA was concerned there, as here, that plants
would be able to meet the limits for these toxic organic pollutants through dilution because, as is
the case here, the pollutants are present in wastewater only from certain unit operations and in
concentrations that could be reduced below analytical detection levels after commingling with
other process wastewater. 49 Fed. Reg. at 8758-59; 48 Fed. Reg. 7032, 7056 (Feb. 17, 1983).
EPA concluded that requiring compliance at the end of the pipe would contravene the strong
policy of the Clean Water Act that pollutants be removed, not diluted, and for that reason
proposed limitations on internal wastestreams. This provision was not contested and continues
to apply today.
EPA received support from commenters, including at least one state, for its
proposal to establish in-plant limitations in the Cluster Rules. The state commenter supported
the use of in-plant limitations for dioxin and other pollutants as a way of ensuring that these
pollutants have been "eliminated." While EPA cannot say on today's record that mills using
chlorine or chlorine-containing compounds as part of their bleaching processes have "eliminated"
dioxin discharges (even if the effluent is reported at below the applicable minimum level for
dioxin), EPA agrees for the reasons set forth above that bleach plant limitations are necessary in
order to ensure that limitations expressed as "
-------
Section 8 - Development of Control and Treatment Options
assert that, even if EPA possessed the legal authority to regulate internal wastestreams, EPA has
failed to identify any exceptional circumstances that would justify in-plant limitations.
EPA disagrees with the first comment that it lacks the legal authority to impose
limitations on internal wastestreams. EPA's regulation at 40 CFR Part 122.45(h) explicitly
authorizes such limitations on a case-by-case basis; that regulation, which was promulgated in
1979, has been upheld by both U.S. Courts of Appeals that considered it. See Public Service
Company of Colorado. Fort St. Vrain Station v. EPA. 949 F.2d 1063 (10th Cir. 1991); Texas
Municipal Power Agency (IMPA) v. EPA. 836 F. 2d 1482 (5th Cir. 1988). EPA believes it is
reasonable to exercise this case-by-case authority on a subcategory-wide basis because EPA's
record shows that every permit writer is likely to encounter the same facts that EPA has
identified to justify these in-plant limitations; indeed, no commenter has come forward with
information demonstrating that the removals expected by the effluent limitations and standards
for dioxin, furan and chloroform can be achieved by treatment after the bleach plant, or that the
12 chlorinated phenolic pollutants can be segregated from other wastestreams to avoid the
problem of dilution. For these reasons, EPA foresees no situation at this time when a permit
writer would reach a different conclusion on an individual permit basis. If over time, however,
mills are able to demonstrate that they can achieve the applicable removals at the end of the pipe
by treatment rather than by dilution or volatilization, or if new, approved analytical methods
make it feasible to demonstrate compliance with the limitations or standards measured at the end
of the pipe (taking dilution into account), then EPA will consider amending the rule to authorize
those dischargers to demonstrate compliance at the end of the pipe rather than at bleach plant.
Commenters do not deny that, as a consequence of bleach plant operations, an
"addition" of pollutants occurs to navigable waters or that such "addition" constitutes a discharge
of pollutants subject to Clean Water Act regulation. However, they assert that EPA is authorized
to control that "addition" only at the point at which the pollutant physically enters the receiving
water. EPA disagrees that the Clean Water Act must be read so narrowly — especially when
applied to technology-based limitations. To the contrary, Congress defined the term "effluent
limitations" to mean "any restriction" on "quantities, rates, and concentrations of... constituents
which are discharged from point sources into navigable waters." Section 502(11). The term
"discharge," in turn, applies to "any addition," and is to be broadly construed. Section 502(12);
see United States v. Earth Sciences. Inc.. 599 F.2d 368, 374 (10th Cir. 1979); SEP. Inc. v. City of
Dayton. 519 F. Supp. 979, 989 (S.D. Ohio 1981). Like section 301(a)'s prohibition on the
discharge of "any pollutant," these definitions authorize EPA to impose any reasonable form of
restriction on pollutants that will "eventually" be discharged into waters of the United States.
Public Service Company of Colorado. 949 2d at 1065: see TMPA. 836 F. 2d at 1488-90. As the
Fifth Circuit recognized in upholding EPA's internal wastestream rule, "it is sometimes
necessary to regulate discharges within the treatment process to control discharges at the end."
TMPA, 836 F.2d at 1488. The authority to establish effluent limitations thus includes, where
appropriate, the authority to impose limitations on internal wastestreams. This authority is
especially broad when, as here, the in-plant restrictions are designed to promote the Act's goals
of eliminating, and not simply diluting or dispersing, pollution in effluent through technology-
based limitations. See Chevron. U.S.A.. Inc. v. NRDC. 467 U.S. 837, 843-44, 861 (1984); see.
8-10
-------
Section 8 - Development of Control and Treatment Options
e.g.. Sections 101(a)(l), 101(a)(3), 301(b)(2)(A). The legislative history of the Clean Water Act
also emphasizes that dischargers are not to employ dilution as an alternative to treatment.
Establishment of in-plant limitations is also authorized by section 402(a)(2),
which directs EPA to prescribe conditions for NPDES permits to assure compliance with the
requirements of section 402(a)(l), including section BAT limitations promulgated under section
301. See TMPA, 836F.2dat 1489: Montgomery Environmental Coalition v. Costle. 646 F.2d
568, 586-87 (D.C. Cir. 1980). As the D.C. Circuit recognized, EPA's broad authority under
section 402(a)(2) includes the authority to regulate inflows to treatment works, e.g., in the form
of moratoria on sewer hook-ups to POTWs. Montgomery Environmental Coalition v. Costle,
646 F.2d at 588. EPA also relies on the regulation prohibiting the use of dilution to meet
categorical pretreatment standards, 40 CFRPart 403.6(d), for authority to impose in-plant
limitations on indirect dischargers.
EPA also acknowledges comments asserting that EPA lacks authority to regulate
internal wastestreams because they are not waters of the United States. Because EPA is not
basing its authority on the possible status of internal wastestreams as waters of the United States,
this additional argument by commenters requires no response.
EPA acknowledges that the U.S. Court of Appeals for the District of Columbia
Circuit recently ruled that EPA lacks the legal authority to impose "water quality-based standards
upon internal facility waste streams." American Iron and Steel Institute (AISI) v. EPA, 115 F.3d
979, 996 (D.C. Cir. 1997). The issue arose in the context of a challenge to EPA's regulations
implementing the Great Lakes Critical Programs Act, 33 U.S.C. Part 1268. In its regulations,
EPA promulgated a procedure, applicable only to states in the Great Lakes System, that provides
that when a permit includes a water quality-based effluent limitation below the level of
quantification, the permit must also require the permittee to develop and conduct a pollutant
minimization program (PMP), which includes among other things, a control strategy. See 40
CFR Part 132 App. F, Procedure 8.D. Procedure 8.D. 1. states that the goal of a PMP is "to
reduce all potential sources of the pollutant to maintain the effluent at or below the Water Quality
Based Effluent Limitations." It further states that the control strategies should be "designed to
proceed toward the goal of maintaining all sources of the pollutant to the wastewater collection
system below the WQBEL." See Procedure 8.D.3. Industry litigants challenged these
provisions, claiming that they would be used to set internal waste stream WQBELs. The Court
held that although EPA has the authority to require monitoring of internal wastestreams, see
AISI, 115 F.3d at 995, the Clean Water Act does not authorize EPA to require compliance with a
WQBEL at a point inside the facility and thereby deprive a permittee of the ability to choose its
own control system to meet the WQBEL, see id. at 996. Therefore, the Court vacated Procedure
8.D. "insofar as it would impose the point-source WQBEL upon a facility's internal waste
streams." Id.
EPA does not believe that decision controls here. First, the court did not consider
or rule upon the regulations upon which EPA bases the in-plant limitations for these Cluster
Rules, to wit, 40 CFRPart 122.44(1), 122.45(h), 125.3(e), and 403.6(d). Therefore, those
8-11
-------
Section 8 - Development of Control and Treatment Options
regulations, upheld by the Fifth and Tenth Circuits, continue to have force and effect. Second,
the AISI court did not consider the question whether EPA has authority to regulate internal
wastestreams in the context of technology-based controls. Unlike water quality-based effluent
limitations, which are calculated to ensure that water quality standards for the receiving water are
attained, technology-based limitations and standards are derived to measure the performance of
specific model technologies that EPA is required by statute to identify. In identifying these
technologies, EPA is directed to consider precisely the type of internal controls that are irrelevant
to the development of water quality-based effluent limitations, such as the processes employed,
process changes, and the engineering aspects of applying various types of control techniques.
See CWA Sections 304(b)(2)(B), 306(a)(l), 307(b) & (c). Indeed, EPA's regulations must
identify the degree of effluent reduction attainable through the application of BAT-level process
and procedure innovations, operating methods, and other control measures and practices for the
subcategory being regulated. CWA Section 304(b)(2)(A). Limitations applied at the bleach
plant represent the degree of effluent reduction EPA has determined is attainable by the BAT,
NSPS, PSES, and PSNS model technologies. Thus, they are consistent with the long-recognized
principle that technology-based limitations are intended to reflect, for each industrial category or
subcategory, the "base level" of technology (including process changes) and to ensure that '"in
no case . . . should any plant be allowed to discharge more pollutants per unit of production than
is defined by that base level." E.I, du Pont de Nemours & Co. v. Train, 430 U.S. at 129, quoting
S. Rep. No. 414, 92d Cong., 1st Sess 50, reprinted in A Legislative History of the Water
Pollution Control Act Amendments of 1972, at 1468 (1973).
EPA also rejects as inapposite the commenters' second objection — that in-plant
limitations may ignore or fail adequately to account for reductions in pollutant loadings that will
occur after the bleach plant but before discharge, such as reductions attributable to secondary
treatment. Under Sections 301(b)(2) and 304(b)(2), EPA is required to establish effluent
limitations guidelines and standards that reflect, for example, the best available technology
economically achievable for the particular pollutants of concern. While the model BAT and
NSPS technologies for Subcategory B includes secondary biological treatment to address AOX
loadings, the model BAT and NSPS technologies for dioxin, furan, chloroform and the 12
chlorinated phenolic compounds (and, for PSES and PSNS, AOX) consist exclusively of a
sequence of in-plant processes, including improved brown stock washing, elimination of
hypochlorite and complete substitution of chlorine dioxide for elemental chlorine, which are
employed at the pulp mill and the bleach plant. The bleach plant limitations and standards
codified today thus fully account for the model technologies and processes upon which they were
based. In EPA's view, these processes, not end-of-pipe biological treatment, represent the best,
most efficient methods of minimizing the discharge of dioxin, furan, chloroform, the 12
chlorinated phenolics, and (for indirect dischargers) AOX, and moving toward the ultimate BAT
goal of eliminating their discharge altogether. See CWA Sections 101(a)(l) and 301(b)(2)(A).
Moreover, commenters have failed to supply any data demonstrating that dioxin and furan can be
effectively treated by biological treatment processes. Results of the Five-Mill Study and the 104-
Mill Study show that dioxin and furan are not removed by biological treatment systems at direct
discharging BPK and PS mills (8,9). Rather, these pollutants were found to either partition to the
secondary sludge of a biological treatment system or pass through untreated. The partitioning
8-12
-------
Section 8 - Development of Control and Treatment Options
was neither consistent nor predictable. (If dischargers ultimately can make this showing, EPA
would consider amending the regulation.) With respect to the 12 chlorinated phenolic
compounds, EPA acknowledges that these pollutants can be treated using biological treatment.
However, as noted above, the same process technologies necessary to address dioxin and furan
also, as an ancillary matter, reduce the levels of chlorinated phenolics to nondetectable
concentrations at the bleach plant. Moreover, without bleach plant limitations on these
pollutants, EPA would have no assurance that limitations are being achieved by treatment rather
than simply dilution, and that the pollutants are being removed.
EPA also disagrees with the third comment. Contrary to the commenters'
assertions, the establishment of in-plant limitations does not have the effect of unlawfully
dictating a particular technology or process. While EPA's model technologies indisputably rely
on a particular sequence of process changes, which is expressly authorized under section
304(b)(2)(A)&(B), mills remain free to develop and implement any combination of processes
and technologies, including but not limited to chemical substitution, to achieve the limitations
and standards. The only "treatment" essentially precluded by the in-plant limitations and NSPS,
PSES, and PSNS is dilution of bleach plant effluent with effluent from other mill processes,
which is not BAT/NSPS and, in the case of pretreatment standards, is explicitly prohibited by
other regulations. See 40 CFR Part 403.6(d).
EPA concludes that commenters' fourth objection also has no merit. They argue
that EPA's determination in this rulemaking that end-of-pipe limitations BAT on pulp and paper
effluent are infeasible or impracticable is inconsistent with 1990 guidance to permit writers that
characterized such limitations as appropriate. EPA disagrees with commenters' interpretation of
that guidance. Entitled "Strategy for the Regulation of Discharges of PHDDs and PHDFs from
Pulp and Paper Mills to Waters of the United States" (May 21, 1990) (DCN 03960), the guidance
specifically urged permit writers to impose limitations on dioxin at the bleach plant when
measuring discharges of those pollutants at the end of the pipe was impracticable or infeasible.
This guidance was consistent with EPA's custom of giving the permit writer discretion to
determine the appropriate point of regulation on a case-by-case basis and reflected EPA's belief
at the time that some mills could demonstrate dioxin removals by wastewater treatment facilities.
DCN 03960 at 20. Based on the data assembled as part of this rulemaking, however, EPA has
determined that no mills today have made that showing. Therefore, as discussed in more detail
above, EPA concluded that it is appropriate to impose in-plant limitations on dioxin on a
categorical, rather than case-by-case, basis.
Several commenters also challenge EPA's finding that chloroform volatilizes en
route to and during wastewater treatment, although they offer no data to contradict the data and
analyses supporting EPA's conclusion. Indeed, EPA's data show that chloroform levels are
easily quantified at the bleach plant but cannot be reliably measured in the final effluent. These
data refute some commenters' unsupported assertion that chloroform levels can be reliably
measured at the end of the pipe. EPA has found that bleach plant chloroform loadings are lower
at mills that have greater contact between bleach plant filtrates and air (e.g., mills that use high
air flow vacuum drum washers) compared to those mills that have less air contact (e.g., low air
8-13
-------
Section 8 - Development of Control and Treatment Options
flow pressure diffusion washers or compaction baffle washers). This suggests that much of the
chloroform generated in the bleach plant is easily stripped by exposure to air. Thus, EPA has
concluded that the majority of chloroform removed in biological treatment systems is removed
by air stripping, not biological degradation. NCASI studied air and water concentrations of
chloroform around four bleached kraft mill wastewater treatment systems and concluded that
"significant reductions in aqueous chloroform concentrations were observed across flumes and
other points of turbulence in the treatment system. Typically the majority of the chloroform was
removed in the first third of the effluent treatment systems." (NCASI Technical Bulletin 642,
DCN 09323).
In addition, EPA disagrees with comments asserting that volatilization is not an
"exceptional circumstance" justifying limitations on internal wastestreams. While EPA agrees
that volatilization is not listed as an "exceptional circumstance" identified in the internal
wastestream rule to justify in-plant limitations, see 40 CFR Part 122.45(h)(2), EPA notes that the
list is merely illustrative and not exhaustive. Moreover, volatilization is reasonably equivalent to
the dilution example provided in the regulation, insofar as the particular pollutant is reduced to
non-quantifiable levels for reasons unrelated to treatment. See supra. Therefore, EPA believes
that volatilization, like dilution, constitutes an exceptionable circumstance within the scope of 40
CFR Part 122.45(h).
Finally, EPA disagrees that EPA lacks the legal authority to consider air impacts
when setting BAT effluent limitations. EPA is clearly authorized under Section 304(b)(2)(B) to
consider non-water quality environmental impacts when choosing BAT and, consequently,
setting effluent limitations based on that model BAT technology. See also CWA Sections
306(b)(l)(B), 307(b) & (c). It follows therefore that EPA can take non-water quality
environmental impacts into account when choosing the point of regulation for its BAT limits.
Indeed, the legislative history and case law interpreting this statutory authority emphasize that
EPA was not to achieve effluent reductions at the expense of other environmental media. See,
e.g.. BP Exploration & Oil. Inc. v. EPA. 66 F.2d 784, 800-02 (6th Cir. 1995); Weyerhaeuser Co.
v. Costle. 590 F.2d 1011, 1052-53 (D.C. Cir. 1978). Thus, in this instance, not only is
chloroform reliably and feasibly measured at the bleach plant (when this is not the case at the end
of the pipe), but in-plant limitations also prevent adverse air impacts that might ensue from end-
of-pipe limits. Finally, chloroform's volatile nature and the significance of air impacts from
volatilization is evidenced by EPA's decision to impose MACT standards for chloroform at the
bleach plant.
8.3 Conventional Pollutants
EPA proposed revised effluent limitations guidelines for control of conventional
pollutant discharges via the following regulations: BPT, BCT, and NSPS. This section provides
a description of EPA's approach to control option development for conventional pollutants for
the proposed rule, revisions to EPA's approach and data set used to develop the control options,
revised control option performance levels for the final rule, and descriptions of the technology
bases for the final conventional pollutant control options.
8-14
-------
Section 8 - Development of Control and Treatment Options
8.3.1 Approach to Option Development
Secondary treatment includes biological or chemical processes used to remove
suspended and soluble organic matter from wastewater. Secondary treatment, in conjunction
with good water conservation practices, constitutes the technology basis for the current
conventional pollutant limitations for bleached papergrade kraft and soda mills and papergrade
sulfite mills. With the exception of one bleached papergrade kraft mill that discharges to
territorial seas of the United States and one which uses land disposal for its primary treated
waste, all of the direct-discharging bleached papergrade kraft and soda mills and direct-
discharging papergrade sulfite mills operate secondary biological wastewater treatment systems.
Secondary biological wastewater treatment removes organic matter through biochemical
oxidation. Activated sludge systems and aerated/non-aerated basin systems are the most
commonly used biological processes. As needed, secondary biological treatment in the pulp and
paper industry is preceded by pretreatment (e.g., equalization, neutralization, cooling) and
primary treatment (e.g., clarifiers) for the removal of settleable solids.
Tertiary treatment is advanced treatment, beyond secondary biological treatment,
to remove particular contaminants. Common tertiary treatment operations in municipal and
industrial wastewater treatment systems are the removal of phosphorus by alum precipitation,
removal of toxic refractory organic compounds by activated carbon adsorption, and removal of
TSS by multimedia filtration. Tertiary treatment is not common in the pulp and paper industry.
The Agency proposed to develop effluent limitations guidelines for conventional
pollutants based on secondary biological wastewater treatment with appropriate water use and
reuse because almost all direct-discharging mills used secondary treatment. Elements of a
secondary biological wastewater treatment train, as practiced in the pulp and paper industry, may
include (but are not limited to):
Equalization;
Neutralization;
Precooling;
Primary sedimentation;
Nutrient addition;
Aeration;
Addition of flocculants to secondary clarifiers to improve settling;
Multi-basin systems, some of which act as polishing ponds; and
8-15
-------
Section 8 - Development of Control and Treatment Options
Mixed or hybrid treatment systems (activated sludge and basin systems
operated in series or parallel).
Some of the aerated stabilization basin treatment systems in use in the industry
cover large areas. Some mills operate their treatment systems to meet water quality-based
discharge limitations. Other treatment systems were enlarged in an add-on manner as the
capacity of the mill increased over time. The Agency considers all of these scenarios as typical
of pulp and paper industry wastewater treatment practice, and included all of them in
characterizing pulp and paper industry secondary biological wastewater treatment performance.
The proposed revised effluent limitations guidelines were developed in seven
steps:
1. Identification of mills representing the performance of secondary
wastewater treatment in each subcategory;
2. Analysis of the performance (in terms of production-normalized final
effluent BOD5 load) of the representative mills to determine "the average
of the best existing performance";
3. Identification of combinations of in-process flow reduction and end-of-
pipe wastewater treatment used by the industry to achieve these
performance levels;
4. Estimation of the cost of applying these identified technologies at mills
that do not currently achieve "the average of the best existing
performance";
5. Estimation of the benefits of applying the identified technologies, in terms
of the reduction in the mass of conventional pollutants discharged;
6. Comparison of the estimated costs and benefits; and
7. Review of non-water quality environmental impacts.
The remainder of this section describes Items 1, 2, and 3: the identification of
mills representing secondary treatment performance, the determination of conventional pollutant
control option performance levels, and identification of technologies used by the industry to
achieve the conventional pollutant control option performance levels.
8-16
-------
Section 8 - Development of Control and Treatment Options
8.3.2 Identification of Mills Representing Secondary Wastewater Treatment
Performance in Each Subcategory
To identify mills representing the performance of secondary wastewater treatment
in each subcategory, the Agency began with the list of mills that responded to the 1990 census
questionnaire (1) and deleted mills from the list that were not considered representative of
secondary biological treatment system design, operation, and performance as practiced by mills
in each subcategory. Deleted mills included primarily indirect-discharging mills and "multiple"
subcategory mills (discussed below).
Ideally, only mills with all of their final off-machine production in a single
subcategory would be used to represent the performance of secondary biological wastewater
treatment for that subcategory. After application of the mill selection parameters, both the
Bleached Papergrade Kraft and Soda and the Papergrade Sulfite subcategories included few or no
mills with all of their production in a single subcategory, reflecting the actual composition of the
U.S. pulp and paper industry (a typical mill has operations in more than one subcategory). To
account for this situation, while at the same time reflecting the characteristics of the principal
subcategory at a mill, the Agency chose to use mills with a large percentage of their final
production in a single subcategory to represent the subcategory. For the Bleached Papergrade
Kraft and Soda Subcategory, the Agency selected a final production cut-off of 85 percent within
this subcategory for a mill to be considered representative of the subcategory. For the Papergrade
Sulfite Subcategory, the Agency selected mills with papergrade sulfite pulp comprising 37 to 96
percent of their final production.
8.3.2.1 Summary and Analysis of Comments Submitted Concerning Identification of
Mills Representing the Performance of Secondary Biological Wastewater
Treatment
Several comments were submitted concerning the selection of mills representing
the performance of secondary biological wastewater treatment. Commenters expressed concern
that EPA used several mills to represent the Bleached Papergrade Kraft and Soda Subcategory
that commenters believed had wastewater treatment operations that are not common practice in
the industry, or that commenters believed achieved performance levels better than those levels
achieved by the BPT/BCT technology basis. Specific issues are described below.
EPA included mills with multiple basins, particularly holding and
polishing ponds, that have residence times longer than those typical in the
industry.
EPA included mills with equalization of wastewater prior to secondary
biological treatment, which were asserted not to be common industry
practice.
8-17
-------
Section 8 - Development of Control and Treatment Options
EPA included mills that have mixed treatment systems (i.e., treatment
systems including both activated sludge treatment and basins), which was
asserted as not common industry practice.
EPA included mills that have chemically assisted clarification to enhance
pollutant removals, which was asserted not to be common industry
practice. In particular, mills using unusually large amounts of coagulant
should have been excluded.
EPA included mills that discharge to water quality limited streams.
Because several commenters criticized EPA for using mills with "atypical"
treatment to represent the subcategory, the Agency undertook an analysis to determine what
constituted typical wastewater treatment for the Bleached Papergrade Kraft and Soda
Subcategory. To describe "typical" wastewater treatment at bleached kraft mills, EPA reviewed
and summarized the treatment systems for all direct-discharging mills that have the highest
percentage of their production in the bleached kraft subcategory. EPA also analyzed treatment
operation and performance data in response to the commenters' criticisms mentioned above. Data
used for the analysis were obtained from mill responses to the 1990 census questionnaire and
supplemental data from comments and site visit reports, where available. Note that for this
analysis, treatment performance is measured by final effluent production-normalized loads,
consistent with development of BPT/BCT performance levels, not treatment efficiency (i.e.,
percent removal) or effluent concentrations. The results of these analyses are discussed in the
rulemaking record (2) and are summarized below.
Basin System Residence Times - EPA analyzed the relationship between total
basin (aerated plus non-aerated basins) residence time versus effluent BOD5 and TSS loads by
plotting residence time versus load. The plots do not suggest any obvious relationship between
basin residence time and treatment performance; the best performing mills have a wide range of
residence times. BOD5 and TSS loads were also plotted versus aerated basin residence time.
These plots also show that no well-defined correlation exists between aerated basin residence time
and performance.
Polishing/Holding Ponds - EPA ranked mills in order of both effluent BOD5 and
TSS loads, distinguishing between mills with and without polishing/holding ponds. A mill was
considered to have a polishing/holding pond if the last treatment unit in its wastewater treatment
system before discharge was a non-aerated basin. The rankings indicate that use of
polishing/holding ponds is common practice - 46 percent of mills have at least one such pond -
and that mills without polishing ponds perform as well (or as poorly) as mills with polishing
ponds.
Several mills have large holding ponds to control effluent flow during low flow
seasons. Data from one mill that controls effluent flow indicate that performance was
approximately the same when the holding pond was bypassed as when it was used, suggesting
that holding ponds function as flow-control devices, not treatment units (3).
8-18
-------
Section 8 - Development of Control and Treatment Options
Equalization - EPA ranked mills in order of both effluent BOD5 and TSS loads,
distinguishing between mills with and without equalization. Equalization in the industry is used
predominantly prior to activated sludge treatment to dampen the effect of shock loads on
activated sludge systems which are more sensitive to upsets than basins. The rankings indicate
that approximately 15 percent of activated sludge systems industry wide and 20 percent of
bleached kraft activated sludge systems have equalization, and that mills without equalization
achieve the same level of performance as mills with equalization.
Mixed Treatment Systems - EPA ranked mills in order of both effluent BOD5
and TSS loads, distinguishing between mills with and without mixed treatment systems. More
than one-third of bleached papergrade kraft activated sludge systems and about 24 percent of
activated sludge systems industry wide are mixed systems, indicating that mixed systems are
common. The range of treatment performance at mills with mixed treatment systems is
comparable to the range at mills without mixed treatment systems.
Chemically Assisted Clarification - EPA ranked mills in order of both effluent
BOD5 and TSS loads, distinguishing between mills with and without chemically assisted
clarification. Mills having chemically assisted clarification achieve BOD5 and TSS levels across a
smaller and overall lower range than mills without chemically assisted clarification; however, the
three and six best performing mills in terms of BOD5 and TSS, respectively, do not use chemically
assisted clarification.
EPA further investigated treatment systems that include addition of coagulants/
flocculants to secondary clarifiers to learn more about why this chemical addition was practiced.
Data used for the analysis included NCASI's 1993 survey of coagulant used at 33 bleached kraft
mills (4) and phone contacts with nine mills (5). Based on this analysis, EPA determined that the
secondary biological wastewater treatment systems of two mills are not considered typical
secondary treatment as practiced by the Bleached Papergrade Kraft and Soda Subcategory
because they added very high quantities of chemicals (i.e., chemical addition costs in excess of
$5,000 per day) for the removal of color—a treatment that is not necessary to achieve the
proposed TSS limitations. Treatment performance data for these systems were eliminated from
the data set used to characterize typical treatment effluent loadings of conventional pollutants.
Discharge to Water Quality Limited Streams - Simply because a mill is required
to meet water quality-based effluent limits for BOD5 and/or TSS does not mean that its treatment
system is atypical. Unusual equipment or unusual operating practices do not appear to be
required to meet the water quality-based effluent limits. Rather, secondary biological treatment
systems are capable of discharging at levels that will achieve water quality-based effluent limits in
certain streams.
Based on these analyses, EPA believes that it would be inappropriate and arbitrary
to eliminate mills from the secondary wastewater treatment performance data set simply because
they operate wastewater treatment systems with relatively long aerated basin residence times, or
8-19
-------
Section 8 - Development of Control and Treatment Options
operate common biological treatment components such as polishing/holding ponds, equalization,
mixed treatment systems, or chemically assisted clarification.
8.3.2.2 Adjustments to the Selection of Data to Represent the Performance of
Secondary Biological Wastewater Treatment
This section discusses the adjustments EPA made after proposal to the selection of
data used to represent the performance of secondary biological wastewater treatment.
Adjustments include revisions to the performance data, clarification of the conventional pollutant
control option technology bases, and revision of the data selection methodology for the
Papergrade Sulfite Subcategory.
Revisions to the Performance Data
One mill was not used to represent secondary wastewater treatment effluent loads
for the Bleached Papergrade Kraft and Soda Subcategory during the development of the proposed
conventional pollutant limitations because it was incorrectly classified as having less than 85
percent of its production in the Subcategory. In fact, 100 percent of its production is bleached
kraft. Therefore, performance data for this mill were included in the development of the final
secondary biological wastewater treatment performance levels for the Bleached Papergrade Kraft
and Soda Subcategory.
EPA determined that the production data used at proposal to characterize the
loadings from another mill were not correct. This sulfite mill sold virtually all of its production in
the form of slush pulp to a neighboring paper mill (that had no other source of pulp). The effluent
from the two mills was combined for treatment. For the proposed rule, EPA evaluated the
performance data for the combined wastewater treatment system based on the combined final
production from both mills (i.e., slush pulp and pulp sold off-site from the pulp mill plus paper
from the paper mill) consistent with the methodology used for other combined wastewater
treatment systems in the industry. Because the pulp is transported to the second mill as slush
pulp, EPA determined that a more appropriate methodology for evaluating performance data for
this combined wastewater treatment system would be to use the total off-machine production
from the two mills (i.e., pulp sold off-site from the pulp mill and paper from the paper mill). With
this change, more than 85 percent of the combined mill final production was derived from sulfite
pulp manufactured at the mill complex.
EPA determined that one mill whose data were used at proposal to represent the
performance of the proposed conventional pollutant control options for the Papergrade Sulfite
Subcategory was not representative of the Subcategory as a whole because it treats wastewater
from liquor by-products manufactured on site. Because this mill is unique among papergrade
sulfite mills, performance data from this mill were eliminated from development of the final
secondary biological wastewater treatment performance levels for the Papergrade Sulfite
Subcategory.
8-20
-------
Section 8 - Development of Control and Treatment Options
Clarification of the Conventional Pollutant Control Option Technology Basis
As discussed previously, EPA further investigated treatment systems that include
addition of coagulant/flocculants to secondary clarifiers. Based on this analysis, EPA determined
that the secondary biological wastewater treatment systems at two mills are not considered typical
secondary treatment as practiced by the Bleached Papergrade Kraft and Soda Subcategory
because of the addition of large quantities of chemicals for the removal of color. Treatment
performance data for these two systems were eliminated from the data set used to characterize
typical treatment effluent loadings of conventional pollutants for this subcategory.
EPA also further evaluated wastewater treatment operations at one mill to
determine whether the treatment system represents typical secondary biological wastewater
treatment for the Papergrade Sulfite Subcategory. This facility uses a high rate activated sludge
treatment system to treat only a portion of their total mill wastewater effluent. The facility
decided to treat only a portion of its wastewater flow due to land availability limitations, and
treats only the most concentrated BOD5 wastewater streams to a higher removal efficiency. The
treated effluent is then combined with the untreated wastestream resulting in a combined effluent
which meets the facility's permit limitations. This practice of treating only the most concentrated
wastewater streams results in a great reduction in the size of the wastewater treatment facility
required, saving both space and cost. Because the total effluent BOD5 load discharged from this
facility is less than the current effluent limitations guidelines which were the basis for the design of
the wastewater treatment facility, EPA considers this treatment system to be representative of the
performance attainable through secondary treatment as typically practiced at similar facilities, and
considers performance data from this facility as appropriate for use developing the conventional
pollutant control option performance levels.
Data Selection Methodology for the Papergrade Sulfite Subcategory
The Papergrade Sulfite Subcategory is characterized by mills whose final
production is often comprised of a large portion of purchased pulp. For the proposed rule, BPT
and BCT option performance levels for the subcategory were calculated using data from mills
with 37 to 96 percent on their final production in the subcategory in order to expand the
population of data used. This approach was a relaxation of the criteria used for other
subcategories which included data from only mills with at least 85 percent of their final production
in the subcategory.
After proposal, EPA reassessed the impact of purchased pulp on the final effluent
BOD 5 load at papergrade sulfite mills. EPA performed further analyses of the data set, after
incorporating the revisions discussed above, to evaluate the relationship between final effluent
BOD5 and TSS loads and the percentage of sulfite production at each mill. EPA determined that
mills with 85 percent or more of final off-machine production derived from sulfite pulp produced
on site discharged substantially higher BOD5 loads from secondary biological wastewater
treatment than mills with less than 85 percent of final off-machine production derived from sulfite
pulp produced on site. Consequently, EPA revised the methodology used to identify mills
8-21
-------
Section 8 - Development of Control and Treatment Options
representing the performance of secondary wastewater treatment for the Papergrade Sulfite
Subcategory to include a final production cut-off of 85 percent for this subcategory.
Final List of Mills Identified as Representing Secondary Biological Wastewater
Treatment
Tables 8-1 and 8-2 present the final list of mills identified as representing
secondary wastewater treatment performance for both the Bleached Papergrade Kraft and Soda
and the Papergrade Sulfite Subcategories, respectively. Using performance data from these mills,
EPA calculated final conventional pollutant control option performance levels for both
subcategories as discussed in the following subsection.
8.3.3 Control Options and Performance Levels
EPA's methodology for calculating conventional pollutant control option
performance levels is described in Section 9.2.3 of the Proposed Technical Development
Document. This methodology is unchanged from the proposed rule with the exceptions described
below.
8.3.3.1 BPT
For the proposed rule, EPA developed two options based on the average of the
best existing performance. These options were:
Option 1: The performance level representing the average of the best 90
percent of mills in each subcategory; and
Option 2: The performance level representing the average of the best 50
percent of mills in each subcategory.
Although the Agency has the statutory authority to revise BPT, the Agency also
has the discretion to determine whether to revise BPT effluent limitations guidelines in particular
circumstances. For the final rule, the Agency is exercising its discretion not to revise BPT for
conventional pollutants for Subpart B and E at this time.
8.3.3.2 BCT
For the proposed rule, EPA developed four options based on the best conventional
pollutant control technology. These options were:
Option A. 1: The performance level represented by the best-performing mill in
each subcategory assuming the baseline performance is equal to the
proposed BPT Option 2;
8-22
-------
Section 8 - Development of Control and Treatment Options
Option A.2: Multimedia filtration assuming the baseline performance is equal to
the proposed BPT Option 2;
Option B.I: The performance level representing the average of the best 90
percent of mills in each subcategory assuming the baseline
performance is equal to current industry performance; and
Option B.2: The performance level representing the average of the best 50
percent of mills in each subcategory assuming the baseline
performance is equal to current industry performance.
Two of these options, Options A.I and A.2, assumed the baseline performance to
be equal to the proposed BPT Option 2. Because EPA has decided not to revise BPT limitations
for conventional pollutants, these two BCT options are no longer applicable. Therefore, for the
final rule for the Bleached Papergrade Kraft and Soda Subcategory, EPA only considered BCT
Options B.I and B.2, known now simply as BCT Options 1 and 2.
For the final rule for the Papergrade Sulfite Subcategory, EPA revised the BCT
development methodology to include only one treatment option as follows:
Option 1: The performance level representing the average performance of all
mills identified as representing the performance of secondary
wastewater treatment for the subcategory.
EPA believes this revision is appropriate since the revised data set for this subcategory contains
performance data from only three mills.
The proposed BCT option performance levels and the final BCT option
performance levels are listed below. The best performing mills whose long-term average
production-normalized mass loadings were used to calculate BCT Options 1 and 2 performance
levels for the Bleached Papergrade Kraft and Soda Subcategory and BCT Option 1 performance
levels for the Papergrade Sulfite Subcategory are indicated in Tables 8-1 and 8-2, respectively.
Bleached Papergrade Kraft and Soda Subcategory
Long-Term Average Loads
BCT Option
Option 1
Option 2
BOD5 Performance Level
(kg/OMMT)
Proposed
2.65
1.57
Final
2.73
1.73
TSS Performance Level
(kg/OMMT)
Proposed
4.46
2.72
Final
4.40
2.72
8-23
-------
Section 8 - Development of Control and Treatment Options
Papergrade Sulfite Subcategory
Long-Term Average Loads
BCT Option
Option 1
Option 2
BOD5 Performance Level
(kg/OMMT)
Proposed
4.97
3.60
Final
7.05
Not Applicable
TSS Performance Level
(kg/OMMT)
Proposed
5.46
4.74
Final
8.37
Not Applicable
8.3.4
Description of Technology Bases
For the proposed and final rules, the conventional pollutant technology bases
include two components, good water conservation practices and secondary biological wastewater
treatment. These two technology components are discussed in detail in Section 9.2.5 of the
Proposed Technical Development Document.
8.4
BAT
After re-evaluating technologies for mills in the Bleached Papergrade Kraft and
Soda Subcategory, EPA determined that the model technology for effluent limitations guidelines
based on BAT should be Option A. The key process technology for Option A is complete (100
percent) substitution of chlorine dioxide for chlorine, along with other in-process technologies and
existing end-of-pipe biological treatment technologies listed in Section 8.2.1.1. Section 7 contains
a description of the pollution prevention technologies considered for Option A. The reasons for
EPA's selection of Option A as the basis of BAT for the Bleached Papergrade Kraft and Soda
Subcategory are stated in the preamble.
The basis for BAT for the three segments of the papergrade sulfite Subcategory is
as described in Section 8.2.2, i.e., the option that EPA developed for each segment also serves as
the basis for BAT for each segment.
8.5
BPT
Although the Agency has the statutory authority to revise BPT, the Agency also
has the discretion to determine whether to revise BPT effluent limitations guidelines in particular
circumstances. For the final rule, the Agency is exercising its discretion not to revise BPT for
conventional pollutants for Subpart B and E at this time.
8.6
BCT
EPA evaluated two technology options for BCT for the Bleached Papergrade
Kraft and Soda Subcategory and one technology option for the Papergrade Sulfite Subcategory.
8-24
-------
Section 8 - Development of Control and Treatment Options
EPA is not revising BCT limitations for these subcategories because more stringent effluent
limitations for conventional pollutants did not pass the BCT cost test.
8.7 NSPS
NSPS for conventional and toxic and nonconventional pollutants are discussed
below.
8.7.1 Conventionals
For the control of conventional pollutants in the proposed rule, EPA considered
the best demonstrated end-of-pipe treatment. The performance level for this option was
equivalent to that for proposed BCT Option A. 1 (the performance level represented by the best-
performing mill in each subcategory).
For the final rule for the Bleached Papergrade Kraft and Soda Subcategory, the
Agency considered two options as follows:
Option 1: The performance level represented by the best-performing mill in
the subcategory (equivalent to the proposed NSPS option); and
Option 2: The performance level representing the average of the best 50
percent of mills in the subcategory (equivalent to BCT Option 2).
For the final rule, EPA selected NSPS Option 2 because the Agency determined that the
performance of the single best mill does not account for all sources of process-related variability
in conventional pollutant generation and treatability expected in the entire subcategory, including
raw materials (i.e., furnish), process operations, and final products. In selecting the final NSPS
technology basis for conventional pollutants, EPA found it necessary to consider the secondary
wastewater treatment performance of the best 50 percent of existing mills in this subcategory in
order to ensure that the resulting standards reflect the full range of processes and raw materials to
produce the full range of products covered by this subcategory. Therefore, EPA is promulgating
NSPS for the conventional pollutants BOD5 and TSS, based on efficient biological treatment
achieving removal of 90 percent or more of influent BOD5 and is retaining the pH NSPS
promulgated in 1982.
For the final rule for the Papergrade Sulfite Subcategory, the Agency concluded
that data in the record are not representative of the performance that can be achieved in the
papergrade subcategory as a whole. Therefore, the conventional pollutant limitations in the 1982
NSPS regulation will be retained in the final NSPS regulations.
8-25
-------
Section 8 - Development of Control and Treatment Options
8.7.2 Toxics and Nonconventionals
For bleached papergrade kraft and soda mills, the Agency focussed its analysis on
Option B as the technology basis for defining NSPS for toxics and nonconventional pollutants.
This option includes all the elements of Option A with the addition of oxygen delignification
and/or extended cooking. Section 8.2.1.2 describes this key difference between Option A and
Option B (Sections 7.2.3 and 7.2.6 contain more detailed descriptions of each technology).
EPA received comments that NSPS should be based on TCP bleaching
technologies and flow reduction technologies such as those technologies that form the basis of the
Voluntary Advanced Technology Incentives Program Tiers II and III. EPA does not intend that
NSPS should prevent the manufacture of any products currently made by the U.S. pulp and paper
industry. Thus, EPA assumed at proposal, and continues to assume that the same grades of pulp
will be made by new source mills as are made by existing mills. In addition, these new source
mills will produce pulp for the same range of paper and paperboard products as do existing pulp
producers. As discussed in this section, EPA has determined that data available in the record are
not sufficient to confirm that TCP bleaching processes are technically demonstrated for the full
range of products made with bleached kraft pulp. For the majority of applications, the most
important bleached kraft pulp quality requirements are strength and brightness. EPA's record
discussed in Sections 8.7.2.2 to 8.7.2.4, confirms that fully bright and strong kraft pulps can be
made using TCP bleaching processes (7,8). EPA however, lacks data on the use of TCP for
certain applications with other quality requirements, specifically:
Tissue, which requires soft, absorbent pulp, with good runnability on tissue
machines; and
Food-grade liner board, which requires pulp with low extractives content
to prevent taste and odor transfer.
The data are not sufficient to confirm that pulp made with TCP processes is or is
not feasible for these applications, i.e., EPA does not have sufficient data to establish
subcategories based on end products for which TCP pulp is usable. EPA is inviting interested
parties to supply more data on the full range of products currently made with TCF-bleached kraft
pulps. EPA will evaluate these data, and determine whether to propose revisions to NSPS based
on TCP and, if appropriate, flow reduction technologies.
EPA evaluated TCP bleaching as part of the technology basis for the Voluntary
Advanced Technology Incentives Program. EPA has identified a "Toward-TCF" option as one of
the technology bases of Incentive Tier II, and a TCP option as one of the technology bases of
Incentive Tier III. See the Voluntary Advanced Technology Incentives Program Technical
Support Document for additional detail on these options.
8-26
-------
Section 8 - Development of Control and Treatment Options
The basis for NSPS for the three segments of the papergrade sulfite subcategory is
the same as the basis for BAT as described in Section 8.2.2.
8.7.2.1 Properties of Paper Products
Paper is made from pulp fiber and fiber fragments bonded together into a web-like
structure. Paper is typically made from a mixture of two or more different types of fibers
representing different wood species and/or different pulping methods. Paper also contains non-
fibrous components such as starch, clay, titanium dioxide, and a variety of other additives. Prior
to forming into paper, wood fiber undergoes a mechanical process called beating. The objective
of beating is to cause structural changes in the pulp fibers, such as, development of new fiber
surfaces; fiber cutting, which modifies the fiber size distribution; and partial dissolving of
polysaccharides from the fiber wall. The properties of the final paper product are determined to a
large extent by the composition and qualities of the furnish, such as:
Pulp wood species;
Pulping method (kraft, sulfite, mechanical, etc);
Beating conditions; and
Fillers and additives.
The shortcomings of one component of a furnish are compensated for by
adjustments in the other components. For example, hardwood fibers are relatively small and form
a paper that is smooth and dense, but not very strong. Softwood fibers are longer and contribute
to the strength of the finished paper. In general, bleached kraft pulps are important components
of many paper products because they contribute strength and brightness.
8.7.2.2 TCF Bleaching at Metsa-Rauma and SCA-Ostrand
By the end of 1996, two mills were producing exclusively TCF pulp that was fully
bright and fully strong. The SCA-Ostrand pulp mill in Timra, Sweden began to produce TCF
pulp in May 1995 and stopped all production of ECF pulp in June 1996. The Metsa-Rauma pulp
mill, in Rauma, Finland was constructed in 1994-1996, and began to produce TCF pulp in 1996.
Full details of the history and operations of these mills are available in the record (7,8).
Characteristics of the TCF softwood pulp produced at Metsa-Rauma and SCA-Ostrand are
presented below, with characteristics of oxygen delignified ECF softwood pulp, for comparison.
8-27
-------
Section 8 - Development of Control and Treatment Options
TCF Pulp Manufactured at SCA-6strand and Metsa-Rauma
Product
Oxygen Delignified ECF1
Ostrand TCF (7)
BotnmVerde 85 (8)
BotniaVerde 88 (8)
Wood
Type
SW
sw
SW
sw
Brightness
(ISO)
88.7 to 89.4
88
-85
87.1 to 90.7
Tensile Strength
(Nm/g)
70
80
70
70
Tear Strength
(NmVkg)
14.2 to 15.1
up to 13
over 15
13. 7 to 14
BotniaVerde is the trademark for Rauma's TCF pulps.
As discussed below, the 75 to 85 ISO TCF softwood pulp made at SCA-Ostrand
is used in the manufacture of wood-containing printing papers (lightweight coated and
supercalendared papers), while 88+ ISO TCF softwood pulp is used to manufacture fine paper.
Similarly, the lower brightness TCF pulp made at Metsa-Rauma is used in the manufacture of
wood-containing printing papers, while the high brightness TCF pulp, BotniaVerde 88, is used in
the manufacture of wood-free fine papers.
8.7.2.3 EPA Concludes that TCF Bleaching is An Available, Demonstrated
Technology for Some Products
After examining the data available from SCA-Ostrand and Metsa-Rauma, EPA has
determined that TCF bleaching is an available, demonstrated technology for the production of
high brightness and high strength hardwood and softwood kraft pulps used for the manufacture of
wood-containing printing papers and wood-free fine papers.
8.7.2.4 EPA Concludes that Its Record is Not Sufficient to Determine if TCF
Bleaching is an Available, Demonstrated Technology for All Products
Data available to EPA about the use of TCF-bleached kraft pulp for various paper
products is summarized in Table 8-3. EPA has determined that these data are not sufficient to
confirm that TCF bleaching processes are technically demonstrated for the full range of products
made with bleached kraft pulp. EPA lacks data on the use of TCF kraft pulp for certain
applications with other quality requirements, specifically:
Tissue, which requires soft, absorbent pulp, with good runnability on tissue
machines; and
'From: (9), Table 13. Data taken from Malinen, R., T. Rantanen, R. Rauronen, and L. Toikkanen. "TCF Bleaching to
high brightness-bleaching sequences and pulp properties," International Pulp Bleaching Conference Proceedings,
CPPA, Montreal, 1994.
8-28
-------
Section 8 - Development of Control and Treatment Options
Food grade, which requires pulp with low extractives content to prevent
taste and odor transfer. EPA has no data on the extractives content, after
TCP bleaching, of kraft pulp made from wood species that U.S.
manufacturers use for food grade liner board.
Further, EPA concludes that the data are not sufficient to determine if pulp made
with TCP processes is or is not feasible for these applications, i.e., EPA does not have sufficient
data to define a range of product specifications that can be made with TCP-bleached kraft pulp,
and thus is not able to define a subcategory for which TCP bleaching is an available, demonstrated
technology.
8.8 PSES
Under the CWA, EPA is authorized to establish categorical pretreatment standards
for existing sources and new sources that discharge pollutants that pass through POTWs or
interfere with treatment processes or sludge disposal methods at POTWs. EPA's definition of
POTW pass through is presented in a January 28, 1981 Federal Register Notice. The Federal
Register Notice defines pass through for the purposes of developing national categorical
standards as follows:
"In determining whether a particular pollutant is passing
through the POTW and is, therefore, appropriately
subject to regulation through categorical pretreatment
standards, the Agency compares POTW removal with
removal obtained by a direct discharger. A pollutant
will be deemed to pass through a POTW, and will thus
be characterized as incompatible, where the average
treatment provided by POTWs nationwide does not
realize the same percentage of removal of the regulated
parameter as would be required of direct dischargers
with national effluent standards for that pollutant. Thus,
if in order to comply with their direct discharge BAT
standards, direct dischargers in Category Y were
required to remove 85 percent of pollutant X, then
POTWs must achieve an average of at least 85 percent
removal of pollutant X in order to avoid the conclusion
that pollutant X presents a Pass-Through problem."
Based on this guidance, pass through is determined by comparing the average
treatment provided by POTWs nationwide (expressed as a percentage removal) to the average
treatment provided by direct discharging bleached papergrade kraft and soda (BPK) and
papergrade sulfite (PS) mills that control pollutants to the level of BAT (or NSPS, for new
sources).
The ten indirect discharging facilities in the BPK and PS subcategories each
contributes the majority of flow or pollutant loadings to a POTW. EPA refers to these POTWs as
8-29
-------
Section 8 - Development of Control and Treatment Options
"industrial POTWs." A list of these mills and their corresponding industrial POTW is shown in
Table 8-4.
At proposal, using very limited data, EPA concluded that biological treatment
systems used at these POTWs were not designed to the same standards as the biological treatment
systems installed and operating at direct-discharging BPK and PS mills. As a result, EPA
concluded that the treatment systems at direct discharging mills achieve greater removals of BOD5
and TSS than are achieved by the industrial POTWs, that is, BOD5 and TSS pass through the
industrial POTWs. By extension, EPA concluded that other pollutants more difficult to
biodegrade than BOD5 (such as AOX, dioxin, furan, and other chlorinated compounds) also pass
through the industrial POTWs.
In order to prevent pass through, EPA determined that PSES (and PSNS, for new
sources) were necessary. EPA proposed PSES and PSNS based on the same technologies that it
used as the basis for BAT and NSPS, respectively. EPA also proposed that compliance with
these standards be demonstrated at the same points as it proposed for BAT and NSPS. That is,
EPA proposed PSES and PSNS for dioxin, furan, 12 chlorinated phenolic compounds, and certain
volatile compounds and required compliance monitoring at the bleach plant effluent. In addition,
EPA proposed PSES and PSNS for AOX, COD, and color and required compliance monitoring
at the point of discharge to the industrial POTW. These proposed standards would have required
indirect-discharging mills to construct complete secondary treatment facilities duplicating the
treatment systems currently operated by POTWs. In 58 FR at 66123-66125, (December 17,
1993) EPA also discussed other PSES options intended to obviate the need for complete
secondary treatment at the indirect discharging mills.
Several commenters on the proposed PSES reported that POTWs adequately treat
BOD5 and TSS from pulp mill wastewaters. Further, commenters asserted that POTWs also
remove AOX from pulp mill wastewaters. These commenters concluded that pretreatment
standards for BOD5, TSS, and AOX set at the point of discharge to the POTW are unnecessary.
As discussed below, EPA reviewed the comment submittals and other available
data and determined that they could be used for a pass-through analysis for conventional
pollutants. EPA had no data characterizing POTW removals of the toxic and nonconventional
pollutants considered for regulation (other than AOX). Such data were not provided by the
commenters and were not available from other sources. Because available data were sufficient,
however, to indicate that POTW and mill treatment system control of BOD5 and AOX appear to
be comparable, EPA used mill treatment system data to characterize POTW removals of toxic and
nonconventional pollutants.
The remainder of this section makes the following points:
1) The control of BOD5, TSS, and AOX by biological wastewater treatment
systems at POTWs receiving BPK and PS mill wastewaters appears to be
8-30
-------
Section 8 - Development of Control and Treatment Options
comparable to the control provided by biological wastewater treatment
systems at direct discharging mills (8.8.2).
2) EPA concluded that BOD5 and TSS do not pass through POTWs (8.8.2).
3) Because (as presented in 8.8.2) POTW and mill treatment system removals
of BOD5 and AOX appear to be comparable, EPA assumed that the
treatment of other pollutants generated at BPK and PS mills (i.e., dioxin,
furan, chlorinated phenolic compounds, and chloroform) in POTWs is
comparable to their treatment in mill-operated biological treatment systems
(8.8.3).
4) Dioxin and furan are not removed in mill treatment systems, thus, EPA
concludes dioxin and furan pass through POTWs (8.8.3).
5) Chloroform is extremely volatile and as such is air stripped during
conveyance to and initial stages of biological wastewater treatment, thus
EPA considers chloroform to pass through POTWs (8.8.3).
6) Model BAT technologies remove all 12 of the chlorinated phenolic
compounds to concentrations less than the minimum level at the bleach
plant. Mill treatment systems achieve less removal, thus, EPA concludes
that the 12 chlorinated phenolic compounds pass through POTWs (8.8.3).
7) Model BAT technologies reduce AOX discharges by approximately 82
percent of the AOX generated at a conventionally operated bleach plant.
Mill-operated biological treatment systems achieve only 43 percent
removal, thus, EPA concludes that AOX passes through POTWs (8.8.3).
8) Limited data characterizing the performance of biological treatment
systems receiving PS wastewaters appear to indicate that control of BOD5
and TSS is similar to the control provided by treatment systems receiving
BPK wastewaters. For the purpose of conducting a pass-through analysis,
EPA assumed that the treatment of other pollutants generated at PS mills
(i.e., dioxin, furan, chlorinated phenolic compounds, and chloroform) in
POTWs is comparable to their treatment in mill-operated biological
treatment systems treating BPK wastewaters (8.8.4).
9) As discussed in Section 8.8.4, below, EPA concluded that dioxin, furan,
chlorinated phenolic compounds, chloroform, and AOX pass through
POTWs for segments of the Papergrade Sulfite Subcategory for which it is
setting BAT limitations at this time.
8-31
-------
Section 8 - Development of Control and Treatment Options
EPA's pass-through determinations are summarized in the following table. EPA is establishing
PSES and PSNS for the pollutants it has determined pass through POTWs.
Pollutant
dioxin
furan
chlorinated
phenolics
chloroform
AOX
BOD5
TSS
Bleached Papergrade
Kraft and Soda
Subcategory
pass through
pass through
pass through
pass through
pass through
does not pass through
does not pass through
Papergrade Sulfite;
Calcium, Magnesium,
or Sodium Segment
pass through
pass through
pass through
pass through
pass through
does not pass through
does not pass through
Papergrade Sulfite;
Ammonium Segment
pass through
pass through
pass through
no determination
no determination
does not pass through
does not pass through
Papergrade Sulfite;
Specialty-Grade
Segment
pass through
pass through
pass through
no determination
no determination
does not pass
through
does not pass
through
8.8.1
Performance of End-of-Pipe Secondary Biological Treatment Systems
EPA compared pollutant control provided by direct-discharging mill biological
treatment systems and POTWs accepting similar wastewaters. Because BOD5 control is the
primary objective of secondary treatment, secondary treatment systems are designed for optimal
BOD5 removal and may not be optimized for TSS removal. TSS are generated during biological
treatment, thus TSS percent removal must be considered along with BOD5 removal and final
effluent TSS concentrations, to completely evaluate the performance of a secondary biological
treatment system. Data available to EPA indicate that the removals of BOD5 and AOX at direct-
discharging mill treatment systems and POTWs accepting similar wastewaters appear to be
comparable. TSS control is also similar.
POTWs - Summaries of removals achieved by secondary biological treatment
systems operated by POTWs receiving BPK and PS mill wastewaters are presented in Table 8-5.
The removals presented in Table 8-5 were calculated from data provided in comments on the
proposed rule, and from EPA's Permit Compliance System (PCS). Table 8-5 also presents the
average removals achieved by secondary biological treatment systems supplied by NCASI in
comments on the proposed rule (DCN 20026 A31).
Commenters supplied BOD5, TSS, and AOX data for the ten POTWs receiving
BPK or PS mill wastewaters, as shown below:
8-32
-------
Section 8 - Development of Control and Treatment Options
Pollutant
BOD,
TSS
AOX
Number of POTWs for which
Average Pollutant Removals were
Submitted
5
3
4
Number of POTWs for which
Influent/Effluent Data were Supplied to
Support Average Pollutant Removals
o
J
2
2
The average removals for each pollutant for the POTWs were calculated by first
arithmetically averaging all data supplied for each POTW in comment submittals and from EPA
PCS data (as presented in Table 8-5). (Note that the low AOX removal data from the Jackson
County Port Authority were deleted before the average for this facility was calculated because the
treatment system was disrupted during sampling due to changes in the process at the mill.) Then,
average pollutant removals for all POTWs were arithmetically averaged to yield one average
pollutant removal for each pollutant. These averages were compared to the NCASI data
(presented in Table 8-6) and to average pollutant removals from direct-discharging BPK mills.
Mill Treatment Systems - For the purpose of comparing the performance of
POTWs and mill biological treatment systems, EPA used data from three mills employing Option
A technology and four mills employing Option B technology for which treatment system influent
and effluent data were available. EPA combined these data because it believed that treatment
system removals of BOD5, TSS, and AOX did not differ significantly between mills using Option
A and Option B pulping and bleaching technologies. The average removals were calculated by
arithmetically averaging removals achieved by the three Option A mills (Georgia-Pacific's
Brunswick and Leaf River mills (see Record Section 21.6.1.3) and the James River Wauna Mill
(see Record Section 21.6.1.5)) and the removals from four Option B mills sampled by EPA (12).
Comparison - Table 8-7 shows a comparison of pollutant removals between
POTWs receiving BPK wastewaters and treatment systems operated by direct-discharging BPK
mills. These results are also summarized in the table below.
8-33
-------
Section 8 - Development of Control and Treatment Options
Average Treatment System Removals (percent)
POTW
NCASI POTWs
Mill Treatment
Systems
Average
Range
Average
Range
Average
Range
BOD,
93
(6 POTWs)
86 to 97
94.3
(9 POTWs)
86 to 98.8
95
91 to 98
(5 mills)
TSS
86
(5 POTWs)
56 to 95
80.5
(9 POTWs)
4ato98.5
74
31 to 98
(6 mills)
AOX
49
(4 POTWs)
41 to 58
53.4
(6 POTWs)
30 to 100
49
1 1 to 78
aWithout this value, the range of TSS removals reported by NCASI is 77 to 98.5 percent. The POTW reporting this low
removal has a relatively low influent concentration (55 mg/L), contributing to the low percent removal.
These data appear to indicate that the removals of BOD5, TSS, and AOX achieved by secondary
biological treatment systems at these industrial POTWs are comparable to the removals achieved
by secondary biological treatment systems operated by direct-discharging BPK mills. The
comparability of the effectiveness of POTW and mill-operated biological treatment systems in the
control of TSS is further supported by TSS effluent concentrations. EPA compared the POTW
final effluent TSS concentrations reported by NCASI to the concentrations achieved at 32
bleached kraft mills EPA used to characterize the performance of biological treatment. POTW
TSS effluent concentrations ranged from 4.5 to 56 mg/L, and averaged 31 mg/L. Mill treatment
system TSS effluent concentrations ranged from 4.6 to 140 mg/L, and averaged 55 mg/L. EPA
concluded that BOD5 and TSS do not pass through POTWs and thus is not promulgating PSES
or PSNS for BOD5 or TSS. EPA's pass-through determination for AOX is described in 8.8.3
below.
8.8.2
Bleached Papergrade Kraft and Soda Pass-Through Analysis for PSES and
PSNS
EPA compared the pollutant removals achieved by BPK mills implementing the
model BAT technologies (process changes and biological wastewater treatment) to the pollutant
removals achieved by biological wastewater treatment systems treating BPK mill wastewater. As
discussed above, EPA had data from POTWs treating BPK and PS wastewater for three
pollutants (BOD5, TSS, and AOX). For the purpose of conducting the pass-through analyses
discussed below, EPA assumed that the treatment of other pollutants generated at BPK and PS
mills (i.e., dioxin, furan, chlorinated phenolic compounds, and chloroform) in secondary biological
treatment systems at POTWs is comparable to their treatment in mill-operated end-of-pipe
secondary biological treatment systems. This assumption is reasonable because:
1) The majority of the wastewater flow and pollutant loading at these POTWs
is contributed by BPK or PS mills; and
8-34
-------
Section 8 - Development of Control and Treatment Options
2) The data indicate that the control of BOD5, TSS, and AOX pollutants in
POTW treatment systems receiving BPK and PS mill wastewaters appears
to be comparable to the control of these pollutants by biological treatment
systems at direct discharging mills.
Commenters supplied AOX removal data for four POTWs. EPA, however, chose to use its
database of mill treatment system AOX removals to characterize POTW performance. EPA
chose to use its database because it had full information to evaluate the accuracy, precision, and
representativeness of its data. This approach was reasonable because the data supplied by
commenters showed that removals of AOX in POTWs were similar to removals in mill treatment
systems. The conclusions of EPA's pass-through analysis are presented below for each regulated
toxic and nonconventional pollutant.
Dioxin and furan - Results of the Five-Mill Study and the 104-Mill Study show
that dioxin and furan are not removed by biological treatment systems at direct-discharging BPK
and PS mills (13,14). Rather, these pollutants were found to either partition to the secondary
sludge of a biological treatment system or pass through untreated. The partitioning was neither
consistent nor predictable. In contrast, as discussed in Section 9.0 of this document, EPA
estimated that for BPK mills, the model BAT technology would remove 91 percent of the baseline
bleach plant loading of dioxin and furan. In addition, EPA estimated that the model NSPS
technology would remove 93 percent of the baseline bleach plant loading of dioxin and furan.
Compared to the 0.0 percent removal of dioxin and furan achieved by biological wastewater
treatment systems employed by POTWs, the model BAT and NSPS technologies achieve
substantially greater removals. For this reason, EPA concluded that dioxin and furan pass
through POTWs.
Dioxin and furan that remain untreated are discharged to receiving streams. In
contrast, mills implementing the model BAT technologies achieve substantial reductions of dioxin
and furan prior to secondary biological treatment. EPA found that in bleach plant wastewaters
dioxin is removed to less than the minimum level, and furan is removed to less than the minimum
level or to concentrations slightly above the minimum level. EPA also has determined that use of
the model BAT technologies and compliance with limitations for these pollutants at the bleach
plant will reduce concentrations of dioxin and furan in sludges from those found in the Five-Mill
and 104-Mill Studies to levels near the minimum level for the analytical method for solids (15).
As a result of EPA's finding that dioxin and furan pass through POTWs, EPA is
promulgating PSES and PSNS for these pollutants. In addition to preventing pass through of
dioxin and furan, PSES and PSNS will also reduce possible interference with a POTWs sludge
disposal options. The technology basis, numerical limitations, and point of compliance for PSES
are equivalent to BAT while PSNS is equivalent to NSPS.
8-35
-------
Section 8 - Development of Control and Treatment Options
Chloroform - Chloroform is an extremely volatile compound, with a Henry's Law
Constant1 of 3.39><10"3 atm/gmole/m3 (slightly less volatile that benzene, toluene and the xylene
isomers, but more volatile than isopropyl ether or diethyl ether). The more volatile a pollutant,
the higher its Henry's Law Constant, and the less likely that it will be treated by a POTW. The
pollutant is likely to be volatilized as it flows to the POTW, in either the piping system, the head
works, or the collection systems at the POTW. When the pollutant volatilizes before treatment,
the amount of pollutant influent to the POTW and therefore the amount being biodegraded at the
POTW is reduced. EPA considers that very volatile compounds pass through POTWs because a
significant portion of the compound is air stripped and not biodegraded by the POTW.
NCASI studied air and water concentrations of chloroform around four bleached
kraft mill wastewater treatment systems and concluded "significant reductions in aqueous
chloroform concentrations were observed across flumes and other points of turbulence in the
treatment system. Typically, the majority of the chloroform was removed in the first third of the
effluent treatment systems" (16). NCASI's finding confirms EPA's position that volatile
compounds such as chloroform are air stripped, not degraded in POTWs. EPA has consistently
refused to regard transfers of pollutants from wastewater to air as treatment. Thus, EPA
concludes that chloroform removal in POTWs approaches 0 percent.
EPA compared the negligible chloroform removal achieved in POTWs to the
percent removal expected from a mill using the model BAT technologies. EPA made this
evaluation based on bleach plant effluent loadings, summarized below.
Mill Type
Conventional bleaching (hypochlorite and various levels
of chlorine dioxide substitution)
Model BAT Technologies (6,17)
percent removal
Bleach Plant Effluent
Chloroform Loading (g/kkg)
> 140
3.09
>98
EPA concluded that chloroform removals at mills implementing the model BAT technologies are
vastly greater than the removal achieved by POTWs (>98 percent compared to approaching 0
percent); chloroform, therefore, passes through POTWs. As a result of EPA's findings that
chloroform passes through POTWs, as well as because of potential unacceptable non-water
quality environmental impacts from air emissions, EPA is promulgating PSES and PSNS for
chloroform. The technology basis, limitations, and point of compliance for PSES and PSNS are
equivalent to BAT.
'Henry's Law Constants reflect the partitioning of chemical compounds between a liquid phase (in this case water) and a
gas phase. A compound with a high Henry's Law Constant has a high vapor pressure and a low solubility in water and
thus will preferentially partition to air.
8-36
-------
Section 8 - Development of Control and Treatment Options
Chlorinated phenolic compounds - EPA determined that chlorinated phenolic
compounds pass through POTWs. This determination is based on data in the record (18,19,20)
showing that the model BAT technologies remove all 12 of the chlorinated phenolic compounds
to concentrations less than minimum levels for these pollutants in bleach plant wastewaters, prior
to end-of-pipe biological wastewater treatment. In comparison, mills employing conventional
pulping and bleaching technologies were found to discharge one or more of the chlorinated
phenolic pollutants at measurable levels (6).
To assess whether the chlorinated phenolic compounds pass through, EPA used
the total subcategory baseline and option discharge loading estimates presented in Section 9 of
this document. EPA used this approach in order not to overstate the removals of pollutants
reduced to concentrations less than analytical method minimum levels. Loadings of pollutants
measured at less than the sample-specific detection limit were estimated using one-half the
minimum level of the analytical method (see Section 9.2.1). If EPA had assumed that loadings of
compounds not detected were zero, Option A (the model BAT technology) and Option B (the
model NSPS technology) would achieve 100 percent removal of chlorinated phenolic compounds.
The following table summarizes EPA's estimated removals of chlorinated phenolic
compounds. EPA assumed, based on an NCASI study (21), that 45 percent of the estimated
bleach plant load was removed in biological treatment systems. As discussed in Section 8.8.2 and
in Section 9.3, EPA has assumed that POTW removals will be the same as removals achieved by
mill-operated wastewater treatment systems. EPA estimated the baseline load of chlorinated
phenolic compounds discharged by the BPK subcategory prior to the imposition of new
limitations and standards. EPA also estimated the overall chlorinated phenolic compound
removals achieved by the model BAT technologies (Option A) and the model NSPS technologies
(Option B). (See Section 9.3.1.)
Overall Chlorinated Phenolic Compounds Removals at Baseline and for Model Technologies
Baseline
Model BAT Technologies
(Option A)
Model NSPS Technologies
(Option B)
Bleach Plant Effluent
Loading
(kg/yr)
100,000
18,000
14,000
Treated Effluent
Loading
(kg/yr)
55,000
10,000
7,900
Removal of Baseline
Bleach Plant Loading
(Percent)
45
90
92
Source: Table 9-24 of this document.
For the model BAT technologies (Option A), the estimated bleach plant
chlorinated phenolic compounds loading is 18,000 kg/year and the final effluent chlorinated
phenolic compounds loading is 10,000 kg/year. Similarly, for mills using the model NSPS
technologies (Option B), the estimated bleach plant chlorinated phenolic compounds loading is
14,000 kg/yr and the final effluent chlorinated phenolic compounds loading is 7,900 kg/yr.
8-37
-------
Section 8 - Development of Control and Treatment Options
To assess whether chlorinated phenolic compounds pass through, EPA compared
the overall percent removal achieved by biological wastewater treatment systems treating the
estimated baseline bleach plant loadings to the percent removal expected if the BPK subcategory
mills implemented the model BAT technologies (Option A). Option A includes well-operated
biological treatment. The calculated baseline removal is 45 percent (comparing 100,000 to 55,000
kg/yr). The BAT removal is 90 percent (comparing 100,000 to 10,000 kg/yr). The BAT removal
includes the combined effect of the chlorinated phenolic compounds reduction attributable to in-
plant process changes and the chlorinated phenolic compounds removal due to biological
treatment. Furthermore, the estimated subcategory bleach plant loading (18,000 kg/yr) for
Option A is lower than baseline treated effluent load without Option A technology. Therefore,
EPA concluded that because overall chlorinated phenolic compounds removals with
implementation of the model BAT technologies are substantially greater than the removals
achieved by POTWs, chlorinated phenolic compounds pass through POTWs.
Similarly, EPA compared the overall percent removal achieved by biological
wastewater treatment systems treating the estimated baseline bleach plant loadings to the percent
removal expected if the BPK subcategory mills implemented the model NSPS technologies
(Option B). Option B includes well-operated biological treatment. Again, the calculated POTW
removal is 45 percent (comparing 100,000 to 55,000 kg/yr). The NSPS removal is 92 percent
(comparing 100,000 to 7,900 kg/yr). Furthermore, the estimated subcategory bleach plant
loading (14,000 kg/yr) for Option B is lower than baseline treated effluent load with neither
Option A nor Option B technology. Therefore, EPA concluded that because overall chlorinated
phenolic compounds removals with implementation of the model NSPS technologies are
substantially greater than the removals achieved by POTWs, chlorinated phenolic compounds
from new sources ahopass through POTWs.
As a result of EPA's finding that the 12 chlorinated phenolic compounds pass
through POTWs, EPA is promulgating PSES and PSNS for those pollutants. The technology
basis, limitations, and point of compliance for PSES are equivalent to BAT while PSNS are
equivalent to NSPS.
AOX - EPA also determined that AOX passes through industrial POTWs. EPA
bases this conclusion on its review of data (presented and discussed below) comparing the
removals of AOX achieved by POTWs treating wastewater from mills using conventional
bleaching to the AOX removals achieved by direct dischargers meeting limitations based on the
model BAT process technologies. The mills using the model BAT process technologies
consistently achieve far greater AOX removals than indirect discharging mills using conventional
bleaching technologies that discharge to industrial POTWs. Therefore, in the absence of PSES,
the affected industrial POTWs cannot achieve the same overall removals of AOX as achieved by
direct dischargers complying with the BAT limitations for AOX. The same is also true when
considering removals achieved by new sources complying with NSPS. Thus, EPA concludes that
AOX passes through POTWs and is setting pretreatment standards for AOX for existing and new
indirect-discharging mills.
8-38
-------
Section 8 - Development of Control and Treatment Options
The following table lists final effluent AOX loadings for three softwood bleach
lines sampled by EPA (during the short term study sampling program) that use conventional
bleaching (e.g., less that complete chlorine dioxide substitution). The average bleach plant AOX
loading for these three mills is 2.8 kg/kkg (kilograms of AOX per metric ton of unbleached pulp
entering the bleach plant). This AOX loading is used in the following discussion to represent the
typical bleach plant AOX loading from a softwood mill using conventional pulping and bleaching
technologies.
Bleach Plant Loadings for Mills With Conventional Pulping and Bleaching
Mill
A
B
C
Furnish
softwood
softwood
softwood
Bleach Sequence
C/DEopDD
C/DEHED
C/DEDED
Approximate
Chlorine Dioxide
Substitution (Percent)
50
60
30
Average
Bleach Plant AOX
Loading (kg/kkg)
1.6
2.4
4.3
2.8
The following table summarizes the overall removal of the conventional bleaching
AOX load (2.8 kg/kkg) achieved by biological wastewater treatment, and achieved by the model
BAT technologies (Option A) and the model NSPS technologies (Option B).
Overall AOX Removals for Conventional Bleaching and Option Mills
Mill Type
Conventional Bleaching
Model BAT Technologies
(Option A)
Model NSPS Technologies
(Option B)
Long-Term Average
Bleach Plant Effluent
AOX Loading
(kg/kkg)
2.8
1.3
0.77
Long-Term Average
Treated Effluent
AOX Loading
(kg/kkg)
1.6
0.51
0.21
Removal of
Conventional Bleach
Plant Loading (Percent)
43
82
93
Final effluent AOX loadings were not available from the three mills for which
bleach plant AOX loadings are presented. Using other data collected at proposal, EPA
determined that the final effluent AOX loading after secondary biological treatment for
conventional bleaching mills is 1.6 kg/kkg (22). For mills using the model BAT technologies
(Option A), the average bleach plant AOX loading is 1.3 kg/kkg and the final effluent AOX
loading is 0.51 kg/kkg (6,17). Similarly, for mills using the model NSPS technologies (Option B),
the average bleach plant AOX loading is 0.77 kg/kkg and the final effluent AOX loading is 0.21
kg/kkg (6,17).
8-39
-------
Section 8 - Development of Control and Treatment Options
To assess whether AOX passes through, EPA compared the overall percent
removal achieved by a biological wastewater treatment system treating bleach plant effluent from
a mill using a conventional bleaching technology, to the percent removal expected from a mill
using the model BAT technologies (Option A). Option A includes well-operated biological
treatment. (For the reasons discussed in Section 8.8.2, EPA has assumed that POTW loadings
would be comparable to the mill treated effluent loadings presented above.) The calculated
POTW removal is 43 percent (comparing 2.8 to 1.6 kg/kkg). The BAT removal is 82 percent
(comparing 2.8 to 0.51 kg/kkg). The BAT removal includes the combined effect of the AOX
reduction attributable to in-plant process changes and the AOX removal due to biological
treatment. Furthermore, the average bleach plant loading (1.3 kg/kkg) for a mill employing
Option A is lower than treated effluent from secondary biological treatment without Option A
technology. Therefore, EPA concluded that because overall AOX removals at mills implementing
the model BAT technologies are substantially greater than the removals achieved by POTWs,
AOX passes through POTWs.
Similarly, EPA compared the overall percent removal achieved by a biological
wastewater treatment system of treating bleach plant effluent from a mill using a conventional
bleaching technology, to the percent removal expected from a mill using the model NSPS
technologies (Option B). Option B includes well-operated biological treatment. Again, the
calculated POTW removal is 43 percent (comparing 2.8 to 1.6 kg/kkg). The NSPS removal is 93
percent (comparing 2.8 to 0.21 kg/kkg). Furthermore, the average bleach plant loading (0.77
kg/kkg) for a mill employing Option B is lower than treated effluent from secondary biological
treatment with neither Option A nor Option B technology. Therefore, EPA concluded that
overall AOX removals at mills implementing the NSPS process technologies are substantially
greater than for mills without these technologies. Therefore, EPA concluded that because overall
AOX removals at new source mills implementing the model NSPS technologies are greater than
the removals achieved by POTWs, AOX from new sources also passes through POTWs.
As a result of EPA's findings that AOX passes through POTWs, EPA is
promulgating PSES and PSNS for AOX. The technology basis for PSES is equivalent to the
model BAT technologies, except the PSES basis does not include biological wastewater
treatment. Similarly, the technology basis for PSNS is equivalent to the model NSPS
technologies without biological wastewater treatment. Because secondary biological wastewater
treatment is used at industrial POTWs, neither the model pretreatment technology for PSES nor
PSNS includes secondary biological wastewater treatment.
The pretreatment standards promulgated today for AOX reflect the AOX loadings
present in the bleach plant wastewaters prior to biological treatment at direct-discharging mills
that employ model process technologies. EPA expects that AOX reductions achieved by indirect
dischargers employing the PSES or PSNS model process technology, in combination with
removals achieved by biological treatment systems at industrial POTWs, will be comparable to the
overall removals achieved by direct dischargers complying with BAT limitations or NSPS. AOX
limitations based on the performance of the PSES/PSNS process technology are appropriately set,
and compliance demonstrated, at the bleach plant, prior to mixing with other wastestreams.
8-40
-------
Section 8 - Development of Control and Treatment Options
8.8.3 Papergrade Sulfite Pass-Through Analysis
For the purpose of establishing BAT and NSPS limitations for the Papergrade
Sulfite (PS) subcategory, EPA divided the subcategory into three segments: (a) calcium,
magnesium, or sodium sulfite pulping; (b) ammonium sulfite pulping; and (c) specially-grade
sulfite pulping. To conduct a pass-through analysis for the PS subcategory, EPA compared the
pollutant removals achieved by PS mills implementing the model BAT technologies for each
segment, to the removals achieved by biological wastewater treatment systems treating pulp mill
wastewater. Commenters provided EPA with data from one POTW treating PS wastewater for
only three pollutants: AOX, BOD5, and TSS (see Table 8-6). For the purposes of conducting the
pass-through analyses discussed below, EPA assumed that the treatment of pollutants generated
by PS mills in POTWs is comparable to the treatment of these pollutants in mill operated
treatment systems treating BPK wastewaters. This assumption is reasonable because the limited
data characterizing PS mill treatment systems show that removals of pollutants are similar to the
removals in BPK mill treatment systems.
Calcium, magnesium, or sodium sulfite pulping - For the calcium-, magnesium-
, or sodium-based segment, the model BAT technology is based on TCF bleaching. Mills
employing this model technology will achieve the maximum possible reduction in the discharge of
chlorinated pollutants from bleaching. Because chlorine or chlorine-containing bleaching
chemicals are not used, chlorinated pollutants are not generated during bleaching. EPA concluded
that dioxin, furan, the 12 chlorinated phenolic compounds, chloroform, and AOX removals at
mills implementing the model BAT technologies are greater than the removals achieved by
POTWs. EPA finds that TCF bleaching will reduce AOX discharge loads from the 1 to 3 kg/kkg
typically found at conventional bleaching mills to less than minimum levels, even at indirect-
discharging facilities with no on-site biological treatment. This reduction is greater than 99
percent, which far exceeds the AOX reduction that can be demonstrated by POTW treatment.
Thus EPA concludes that these pollutants all pass through POTWs.
EPA is establishing PSES and PSNS for AOX (expressed as below the minimum
level of the analytical method) for mills in this segment of the Papergrade Sulfite Subcategory,
with the limitation expressed as less than the minimum level. One reason EPA is not establishing
specific pretreatment standards for dioxin, furan, the 12 chlorinated phenolic compounds, or
chloroform is that when AOX is controlled to this level, these pollutants will not be generated by
calcium, magnesium, or sodium sulfite bleaching processes.
Ammonium sulfite and specialty-grade sulfite segments - EPA concluded that
dioxin, furan, and the 12 chlorinated phenolic pollutants pass through or interfere with POTW
operations for the ammonium and specially-grade segments for the same reasons described in
Section 8.2.3, for the BPK Subcategory (i.e., EPA concludes that dioxin, furan, and the 12
8-41
-------
Section 8 - Development of Control and Treatment Options
chlorinated phenolic compounds from PS mills will be removed to the same extent in POTWs as
dioxin, furan, and the 12 chlorinated phenolic compounds from BPK mills). The BAT and NSPS
model technologies for both the BPK and PS Subcategories (ammonium and specialty-grade
segments) are based on ECF bleaching process technologies. As a result of EPA's finding that
dioxin, furan, and the 12 chlorinated phenolic compounds pass through POTWs, EPA is
promulgating national pretreatment standards for new and existing sources for those pollutants for
those segments. The technology basis, numerical limitations, and point of compliance for PSES
and PSNS are equivalent to BAT for these segments.
With respect to chloroform and AOX in the ammonium and specialty-grade
segments of the PS Subcategory, EPA has insufficient data at this time to characterize the
performance of the model BAT technologies (12). EPA needs these data to conduct a pass-
through analysis. When these data become available, EPA will make pass-through determinations
and (if warranted) will set pretreatment standards for chloroform and AOX.
8.9 PSNS
For bleached papergrade kraft and soda mills, EPA is promulgating PSNS based
on the model technology for NSPS, which is Option B, excluding effective biological treatment
(which is presumed to occur at the receiving POTW). The basis for PSNS for the three segments
of the PS subcategory is the same as the basis of BAT and NSPS for these mills, as described in
Section 8.2.2.
8.10 References
1. 1990 National Census of Pulp, Paper, and Paperboard Manufacturing Facilities
Questionnaire. EPA. Record Section 3.1, DCN 08410, August 1992.
2. Recommended Revisions to List of Mills to Represent Secondary Treatment Final
Effluent Loads from the Bleached Kraft and Soda and Papergrade Sulfite
Subcategories. Prepared by ERG for EPA, Record Section 22.2, DCN 13539,
September 1, 1995.
3. D. Falatko. Calculation Sheet. Prepared by ERG for EPA. Record Section 22.2,
DCN 14323CBI, June 26, 1997.
4. NCASI 1993 Survey of Coagulant Use at 33 Mills, Record Section 21.12, DCN
13680, September 27, 1995.
5. Telephone conversations with mill personnel, Record Section 22.2, DCNs 13681,
13682, 13683, 13684, 13685, 13690, and 13726, October 1995.
6. Statistical Support Document for the Pulp and Paper Industry: Subpart B. EPA,
Washington DC, Record Section 22.5, DCN 14496, 1997.
8-42
-------
Section 8 - Development of Control and Treatment Options
7. Site Visit Report. SCA Graphic Sundsvall AB. Ostrand Pulp Mill. Report
prepared by Danforth G. Bodien, U.S. EPA. Record Section 21.5.1, DCN 14786,
10/8/96.
8. Site Visit Report. Oy Mestra-Rauma Ab. Report prepared by Danforth G. Bodien,
U.S. EPA. Record Section 21.5.1, DCN 14787, 10/6/96.
9. Annergren, GE. "Strength Properties and Characteristics of Bleached Chemical
and (Chemi)Mechanical Pulps." In: Pulp Bleaching Principles and Practice.
Carlton W. Dence and Douglas W. Reeve, ed. TAPPI Press, Atlanta, Georgia,
1996.
10. Memorandum to the Record and Sodra Cell Marketing Materials. Record Section
19.2.4, DCN 13823, 7/10/95.
11. Site Visit Report. WisaForest Oy. Pietarsaari. Finland. Report prepared by
K.M. Vice, Radian Corporation, For U.S. EPA. Record Section 21.5.1, DCN
12859, 7/14/94.
12. Summary Report for Pulp and Paper Mill Sampling Program. Prepared by Radian
Corp. for EPA. Record Section 21.6.2, DCN 13968, June 1996.
13. G. Amendola, D. Barna, R. Blosser, L. LaFleur, A. McBride, F. Thomas, T.
Tiernan, and R. Whittemore. USEPA/Paper Industry Cooperative Dioxin
Screening Survey: The Occurrence and Fate of PCDDs and PCDFs in Five
Bleached Kraft Pulp and Paper Mills." Chemosphere. 18(1-6): 1181-1188,1989.
14. R.C. Whittemore, L.E. LaFleur, WJ. Gillespie, G.A. Amendola, and J. Helms.
"U.S. EPA/Paper Industry Cooperative Dioxin Study: The 104 Mill Study."
Chemosphere. 20(10-2): 1625-1632, Record Section 8.2, DCN 05267, 1990.
15. Gillespie, W. Progress in Reducing the TCDD/TCDF Content of Effluents. Pulps
and Wastewater Treatment Sludges from the Manufacturing of Bleached Chemical
Pulp (also known as the NCASI 1994 Dioxin Profile! NCASI. Record Section
19.2.3, DCN 13838, October 1995.
16. "Ambient Chloroform Concentrations in the Vicinities of Bleached Pulp Mill
Effluent Treatment Systems" In: NCASI Technical Bulletin No. 642. October
1992.
17. Data Available for Limitations Development for Toxic and Nonconventional
Pollutants. EPA, Washington DC, Record Section 22.6, DCN 14494, 1997.
8-43
-------
Section 8 - Development of Control and Treatment Options
18. "Effects of Alternative Pulping and Bleaching Processes on Production and
Biotreatability of Chlorinated Organics, Special Report No. 94-01" Presented at:
NCASI Technical Workshop. Record Section 20.2.1, DCN 12415, February 1994.
19. Spengel, D.B., B. Bicknell, D.F. Anderson, M. Smith, and D.G. Bodien. "A
Comparison of Chlorinated Phenolic Compound Concentrations and Loadings in
Bleach-Plant and Treatment-System Samples at Eight Mills" Tappi Journal. Vol.
77, No. 11, November 1994.
20. Stinchfield, A.E., and M.G. Woods. "Reducing Chlorinated Organic Compounds
from Bleached Kraft Mills Through First-Stage Substitution of Chlorine Dioxide
for Chlorine" Tappi Journal. Vol. 78, No. 3, March 1995.
21. Wiegand, Paul, Doug Barton, and Larry LaFleur. "Factors Influencing the
Generation and Treatability of Chlorinated Phenolic Compounds and AOX in
Bleached Chemical Pulp Mill Waste Waters" In: NCASI Special Report No. 94-
01 - NCASI Technical Workshop - Effects of Alternative Pulping and Bleaching
Processes on Production and Biotreatability of Chlorinated Organics. NCASI.
Record Section 20.2.1, DCN 12415, February 1994.
22. Development Document for Proposed Effluent Limitations Guidelines and
Standards for the Pulp. Paper, and Paperboard Point Source Category. EPA-821-
R-93-019, U.S. Environmental Protection Agency, Washington DC, October
1993.
8-44
-------
Section 8 - Development of Control and Treatment Options
Table 8-1
Mills Representing the Performance of Secondary Wastewater Treatment
Bleached Papergrade Kraft and Soda Subcategory
Observation
Number
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
Final
Conventionals
Option 1
Mill?
(Y/N)
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
N
N
Final
Conventionals
Option 2
Mill?
(Y/N)
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
Y
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
N
Long-Term
Average BODS
Effluent Load
(kg/OMMT)
0.26
0.61
0.70
1.06
1.16
1.45
1.55
1.60
1.60
1.92
1.98
2.42
2.57
2.80
2.97
3.02
3.25
3.16
3.35
3.57
3.58
3.68
3.78
4.34
4.34
4.82
5.41
5.53
5.55
5.69
5.70
6.68
Long-Term
Average TSS
Load Effluent
(kg/OMMT)
0.24
1.31
2.23
2.74
3.97
2.11
1.40
3.11
3.48
2.29
2.25
1.95
5.44
2.05
3.04
6.07
8.43
3.55
7.01
9.16
4.60
7.13
8.37
8.73
3.31
2.91
9.67
6.88
4.56
9.64
8.62
9.79
Wastewater
Treatment
Type
B
S
S
S
S
B
B
S
B
B
B
B
S
B
B
B
S
B
B
B
S
B
S
B
B
B
B
B
B
S
B
B
B - Mills that operate secondary wastewater treatment in basins.
S - Mills that operate secondary wastewater treatment in activated sludge systems or a combination of activated
sludge systems and basins.
8-45
-------
Section 8 - Development of Control and Treatment Options
Table 8-2
Mills Representing the Performance of Secondary Wastewater Treatment
Papergrade Sulfite Subcategory
Observation
Number
1
2
3
Long-Term Average BOD5
Effluent Load
(kg/OMMT)
nd
nd
nd
Long-Term Average TSS
Load Effluent
(kg/OMMT)
nd
nd
nd
Wastewater Treatment
Type
nd
nd
nd
nd - Not disclosed to prevent compromising confidential business information.
8-46
-------
Section 8 - Development of Control and Treatment Options
Table 8-3
Data Available on the Use of TCF-Bleached Kraft Pulp
Major Product Type
Newsprint
(wood-containing)
Magazine Printing
(wood-containing, uncoated,
super-calendared)
Light Weight Coated
(wood-containing,
e.g.,catalog grade)
Uncoated Free Sheet (wood
free; e.g., office paper)
Coated Wood-Free and
Other Fine Graphics Papers
Packaging Papers
Tissue:
-cellulose wadding
-facial & napkin
-toilet
-towel
Board Grades
Food-Grade Liner Board
Bleached Kraft Pulp Contribution
up to 25% semi-bleached SW kraft used for
reinforcement
up to 30% bleached or semi-bleached SW kraft
40 - 50% bleached chemical pulp for strength
>90% bleached kraft, HW&SW mixed at variable
proportions, for strength and brightness stability
up to 70% HW kraft for uniformity
SW kraft for strength
added for strength; must not detract from
absorbency, softness, runnability contributed by
sulfite or CTMP pulps or secondary fiber
top layer made from bleached SW&HW kraft for
printability and tensile strength
bleached kraft of semi-bleached with bleached kraft
overlay for strength mechanical pulp needed for
stiffness
Key Bleached Kraft Pulp Qualities
Strength
Strength
web strength, uniformity, low freeness
Strength, high brightness and brightness stability ,HW
for light scattering power (opacity), and formation
Same as uncoated free sheet and more HW for
uniformity of base sheet
strength, stretchability (from high shrinkage potential)
strength, absorbency, softness
strength, brightness
strength, extractives-free to prevent migration of odor
and taste
TCF
Kraft Data?
no; but LWC properties
exceed the quality needed for
newsprint
yes: Ostrand (7)
yes: Ostrand and Metsa-
Rauma (7,8)
yes: Ostrand and Metsa-
Rauma (7,8)
yes: Sodra(lO)
yes: Wisaforest reports food
contact sackgrade can be
made from 100% TCF pulp
(11)
None identified in record
None identified in record
None identified in record
oo
HW - hardwood
SW - softwood
CTMP - chemi-thermo-mechanical pulp
-------
Section 8 - Development of Control and Treatment Options
Table 8-4
POTWs Receiving Chemical Pulp Mill Wastewaters
POTW Name
POTW
Permit No.
Location
Mill Discharging
to the POTW
Bleached Papergrade Kraft and Soda Subcategory
Gulf Coast Waste
Disposal Authority
Muskegon County
Wastewater Management
System
Upper Potomac River
Commission
City of St. Helens
Jackson County Port
Authority
Western Lake Superior
Sanitary District
Bay County Wastewater
Treatment Plant
Erie City Wastewater
Treatment Facility
City of Port St. Joe
Wastewater Treatment
Plant
TX0052591
MI0027391
MD0021687
OR0020824
MS0002674
MN0049786
FL0002631
PA0026301
FL0020206
Pasadena, Texas
Muskegon, Michigan
Westernport, Maryland
St. Helens, Oregon
Pascagoula, Mississippi
Duluth, Minnesota
Panama City, Florida
Erie, Pennsylvania
Port St. Joe, Florida
Simpson Pasadena Paper Co.
S. D. Warren (SAPPI)
Westvaco Corporation
(Luke, MD mill)
Boise Cascade Corporation
International Paper Co.
(Moss Point, MS)
Potlatch Corporation
(Cloquet, MN mill)
Stone Container
Corporation*
International Paper Co.
St. Joe Forest Products Co.*
Papergrade Sulfite Subcategory
Peshtigo Joint
Wastewater Treatment
Plant
WI0030651
Peshtigo, Wisconsin
Badger Paper Mills, Inc.
*Also produces unbleached kraft pulp
8-48
-------
Section 8 - Development of Control and Treatment Options
Table 8-5
Pollutant Removals at POTWs Receiving Chemical Pulp Mill Wastewaters
(Calculated from Comment Submittals and EPA Permit Compliance System Data)
POTW Name
BODS
TSS
AOX
COD
COLOR
Bleached Papergrade Kraft and Soda Subcategory
Gulf Coast Waste Disposal
Authority
Pasadena, Texas
Muskegon County Wastewater
Management System
(Muskegon, Michigan)
Upper Potomac River
Commission
(Westernport, Maryland)
No influent/effluent data
provided in DCNs 20030 or
20021.
City of St. Helens
(St. Helens, Oregon)
Influent/effluent data provided
in DCN 20099.
*
*
94.4% for 1993(DCN
20030-UPRC
comments;
DCN 20021 -
Westvaco comments)
86% (PCS**);
90% for 7/93-6/94
(DCN 20099-City of
St. Helens comments)
*
95%(PCS**)
93.0% for 1993
(DCN 20030-UPRC
comments;
DCN 20021 -
Westvaco comments)
*
*
*
44.6% for 1993
(DCN 20030-UPRC
comments;
DCN 20021 -
Westvaco comments)
41% for 7/93-6/94
(DCN 20099-City of
St. Helens comments)
*
*
72.2% for 1993 (DCN
20030-UPRC
comments;
DCN 20021 -
Westvaco comments)
*
*
*
12.1% for 1993
(DCN 20030-UPRC
comments;
DCN 20021 -
Westvaco comments)
*
oo
-k
VO
-------
Section 8 - Development of Control and Treatment Options
Table 8-5 (Continued)
POTW Name
Jackson County Port Authority
(Pascagoula, Mississippi)
Influent/effluent data supplied in
DCN 12724; no influent/effluent
data supplied in DCN 20042;
influent/effluent data supplied in
DCN 13832.
Western Lake Superior Sanitary
District
(Duluth, Minnesota)
No influent/effluent data
supplied in DCNs 20028 or
20044.
Bay County Wastewater
Treatment Plant
(Panama City, Florida)
No influent/effluent data
supplied in DCN 20066A1.
Erie City Wastewater Treatment
Facility
(Erie, Pennsylvania)
City of Port St. Joe Wastewater
Treatment Plant
(Port St. Joe, Florida)
BODS
90% for 1993
(DCNs 12724 and
20042-Port of
Pascagoula
comments)
97% (PCS**);
97.4% for 2/94
(DCN 20028-Potlatch
comments);
96.7% for 1993
(DCN 20044-Western
Lake SD comments)
95%
(DCN20066A1-
Stone Container
comments)
*
92% (PCS**)
TSS
56% for 1993
(DCNs 12724 and
20042-Port of
Pascagoula
comments)
93% (PCS**)
97. 8% for 2/94
(DCN 20028-Potlatch
comments);
94.9% for 1993
(DCN 20044-Western
Lake SD comments)
*
*
92% (PCS**)
AOX
20% for 11/93
(DCN 12724-Port of
Pascagoula
comments-Note that
treatment system
disruption occurred
because of process
changes; 51% for
3/95 through 11/95
(DCN 13832-IP data
submittal (see
Attachment 2B)
58% for 3/94
(DCN 20028-Potlatch
comments)
*
*
*
COD
47% for 1993
(DCNs 12724-Port of
Pascagoula
comments)
81% for 3/94
(DCN 20028-Potlatch
comments);
80%
(DCN 20044-Western
Lake SD comments)
*
*
*
COLOR
*
36% for 3/94
(DCN 20028-
Potlatch comments);
30%
(DCN 20044-
Western Lake SD
comments)
*
*
*
oo
I
-------
Section 8 - Development of Control and Treatment Options
Table 8-5 (Continued)
POTW Name
BODS
TSS
Papergrade Sulfite Subcategory
Peshtigo Joint Wastewater
Treatment Plant
(Peshtigo, Wisconsin)
94% (PCS**)
62% (PCS**)
AOX
*
COD
*
COLOR
*
* = No data available.
** = Calculated from Environmental Protection Agency Permit Compliance System (PCS) data, included in Attachment 2A.
oo
-------
Section 8 - Development of Control and Treatment Options
Table 8-6
National Council of the Paper Industry for Air and Stream Improvement, Inc. (NCASI) Data
Pollutant Removals at POTWs Receiving Chemical Pulp Mill Wastewaters
Bleached Papergrade Kraft and
Soda mills
Papergrade Sulfite mill
(one mill)
Conventional Pollutant Removals
(Average Removals)
BODS
94.3 %*
(for 9 mills)
94.6 % ***
(for 7 mills)
93 % ****
(for 5 mills)
95 %*
TSS
80.5 %*
(for 9 mills)
90 %*
(for 8 mills-after
removing one mill
with unusually low
TSS removal)
84 %*
Nonconventional Pollutant Removals
(Average Removals)
AOX
53.4 %**
(for 6 mills)
44 6 %*****
(for 6 mills)
25.9 %**
COD
81.6%**
(for 6 mills)
74 7 %*****
(for 5 mills)
no data available
Color
-4.8 %** and *****
(for 4 mills)
no data available
oo
(^
to
All data presented are from NCASI comments "An Analysis of the Relative Performance of POTW and Paper Industry Wastewater Treatment Systems on
Conventional and Non-Conventional Pollutants," April 1994, DCN 20026 A31:
= Appendix B, page 3. Calculated by arithmetically averaging removal data reported in table.
= Appendix B, page 4. Calculated by arithmetically averaging removal data reported in table.
= page 5, Table 2. Influent/effluent data used for this table not presented.
= page 6, Table 3. Influent/effluent data used for this table not presented.
***** = page 8, Table 5. Influent/effluent data used for this table not presented.
Note that the subcategories of the mills used to calculate the averages in Tables 2, 3, and 5 are not specified. Therefore, the papergrade sulfite mill could be
included in these averages. Average influent/effluent data for the 10 POTWs were presented in Appendix B.
**
****
-------
Section 8 - Development of Control and Treatment Options
Table 8-7
Comparison of Pollutant Removals at POTWs Receiving Wastewaters from
Bleached Papergrade Kraft and Soda Mills and Direct-Discharging Bleached
Papergrade Kraft and Soda Mills
Pollutant Removal
Percentages
BODS
TSS
AOX
POTWs Receiving Wastewaters from Mills in the Bleached Papergrade Kraft and Soda Subcategory*
Average
Range
93
(6 POTWs)
86 to 97
86
(5 POTWs)
56 to 95
49
(4 POTWs)
41 to 58
NCASI Data for POTWs Receiving Wastewaters from Mills in the Bleached Papergrade Kraft and Soda
Subcategory from DCN 20026 A31 Appendix B, page 3-4
Average
Range
Direct-discharging Mil
of 3 Mills with O
Average
Range
94.3
(9 POTWs)
86 to 98.8
80.5
(9 POTWs)
4 to 98.5
53.4
(6 POTWs)
30 to 100
Is in the Bleached Papergrade Kraft and Soda Subcategory - 7 Mills (Combination
ption A Technology Basis** and 4 Mills with Option B Technology Basis***)
95
91 to 98
(5 mills)
74
31 to 98
(6 mills)
49
11 to 78
Averages were calculated by arithmetically averaging the removal data for each POTW in Table 2 and then
arithmetically averaging the removals for all POTWs (Note: Low AOX removal data from Jackson Co. Port
Authority due to treatment system upset were excluded from average).
Averages were calculated by arithmetically averaging removal data from 1995 data supplied by Georgia-
Pacific Corporation for the Brunswick and Leaf River mills (see Record Section 21.6.1.3) and by James
River for the Wauna mill (see Record Section 21.6.1.5) and then arithmetically averaging the removals for
the three mills.
Averages were calculated by arithmetically averaging the removal efficiencies from EPA-sponsored sampling
data for 4 mills. The sampling was conducted in 1993-1994. Data are presented in the June 1996 "Summary
Report for Pulp and Paper Mill Sampling Program" in Record Section 21.6.2 DCN#13968.
8-53
-------
Section 9 - Pollutant Reduction Estimates
SECTION 9
POLLUTANT REDUCTION ESTIMATES
9.1 Introduction
After the 1993 proposal of effluent limitations guidelines and standards for the
pulp and paper industry (58 FR 66078), EPA updated the calculation of effluent loadings
reductions for each bleached papergrade kraft and soda mill and each papergrade sulfite mill
potentially subject to those guidelines and standards to establish a new baseline of mid-1995. In
addition, EPA revised and simplified the methodology used to estimate the baseline. These
revised estimates were reported in the July 1996 Notice of Data Availability (61 FR 36835).
After the notice, EPA recalculated the effluent loadings reductions using the same methodology
and the same base year of mid-1995, but made minor changes in loadings for a few particular
mills (consistent with cost model changes). For the final loadings estimates presented here, EPA
also revised the methodology used to interpret concentration measurements for TCF bleaching
processes reported as less than the method minimum level.
Mills in the Bleached Papergrade Kraft and Soda and Papergrade Sulfite
Subcategories will be subject to revised discharge limitations based on BAT or revised PSES.
EPA is promulgating BAT limitations and PSES for TCDD, TCDF, chloroform, 12 chlorinated
phenolic compounds and AOX. The 12 chlorinated phenolic compounds include
trichlorosyringol, 2,4,5-trichlorophenol, 2,4,6-trichlorophenol, 3,4,5-trichlorocatechol, 3,4,5-
trichloroguaiacol, 3,4,6-trichlorocatechol, 3,4,6-trichloroguaiacol, 4,5,6-trichloroguaiacol,
tetrachlorocatechol, tetrachloroguaiacol, 2,3,4,6-tetrachlorophenol, and pentachlorophenol.
Indirect dischargers subject to Subpart B must demonstrate compliance with PSES
for all regulated pollutants at the bleach plant; direct dischargers subject to Subpart B must
demonstrate compliance with BAT limitations for all regulated pollutants (except AOX) at the
bleach plant.
EPA is promulgating BAT limitations and PSES for a subset of these pollutants
for the Papergrade Sulfite Subcategory (Subpart E). For mills in the calcium-, magnesium-, or
sodium-based sulfite pulp segment, EPA is promulgating BAT limitations and PSES for AOX
only, with compliance demonstrated at the end of pipe. For mills in the ammonium-based sulfite
pulp and specialty-grade pulp segments, EPA is promulgating BAT limitations and PSES for the
same pollutants covered by Subpart B except for chloroform and AOX, with compliance
demonstrated at the bleach plant.
EPA had proposed limitations for COD and color. The in-plant process changes
that form the bases of the BAT limitations, PSES, and BMPs, reduce final effluent COD and
color loadings. For this reason, EPA estimated the effluent loadings reductions for COD and
color. For reasons stated in the preamble, EPA intends to develop COD limitations for kraft and
sulfite mills in a future rulemaking and EPA has decided not to develop a national regulation for
9-1
-------
Section 9 - Pollutant Reduction Estimates
color. Instead, permit writers are expected to continue to develop local color effluent limitations
based on applicable water quality standards.
This section describes the approach used to estimate the baseline mass (kg/yr) of
these pollutants in bleach plant effluents and final effluents discharged from bleached kraft and
soda and papergrade sulfite mills. The estimates are based on information available from each
mill as of mid-1995. This section also describes the estimate of pollutant mass that would be
discharged after implementation of the technology options EPA considered in selecting the basis
of the final effluent limitations guidelines and standards. The difference between the baseline
mass discharge and the mass that would be discharged after implementation of a technology
option is referred to here as the pollutant reduction. Pollutant reductions were estimated for all
the pollutants named above.
At present, 96 mills are subject to these regulations. Eighty-six mills discharge
wastewater directly and are regulated under BAT; 10 mills discharge wastewater indirectly and
are regulated under PSES. (See Section 4 for more details about the number and type of mills.)
For Subparts B and E, the BAT limitations and PSES are identical for all regulated pollutants
except AOX. Loadings and reductions calculated for indirect-discharging mills are included in
subcategory and industry totals. Of the ten indirect-discharging mills, nine are in the Bleached
Papergrade Kraft and Soda Subcategory and one is in the Papergrade Sulfite Subcategory.
For those pollutants for which all Subpart B and E mills must demonstrate
compliance at the bleach plant (i.e., all pollutants identified above except AOX in the case of
direct dischargers), EPA calculated baseline loadings, loadings after implementation of the
technology options, and pollutant reductions based on bleach plant effluent data. For AOX, EPA
made the estimates described above using final effluent data (i.e., after secondary treatment at the
mill's wastewater treatment plant or at a POTW). EPA employed this approach even for
indirect-discharging mills in Subpart B (which must demonstrate compliance with the AOX
pretreatment standard at the bleach plant) because EPA expects that the combination of
pretreatment for AOX consistent with the promulgated AOX standard, coupled with additional
removals achieved by the POTW, will produce the same final effluent loading reductions
achieved by direct-discharging facilities subject to Subpart B.
A series of computer programs was used to calculate production-normalized mass
loadings of each pollutant from EPA or industry-supplied mill sampling data. The production-
normalized loadings were then incorporated into a series of spreadsheets. The series of
spreadsheets estimates a baseline pollutant discharge rate (kg pollutant/kkg brown stock pulp into
bleaching) for each mill and compares these loads to production-normalized loadings that would
be achieved after implementation of the technology options. After making this comparison, the
production-normalized loadings and the pollutant reductions are multiplied by each mill's annual
brown stock (unbleached pulp) production to convert to units of kg/yr.
9-2
-------
Section 9 - Pollutant Reduction Estimates
9.2 Pollutant Loading Calculations and Data Sources
EPA does not have data from each mill subject to Subpart B or E that
characterizes the discharge of all pollutants for which EPA is establishing limitations and
standards. Instead, EPA calculated pollutant loadings using data characterizing the generation of
these pollutants by a variety of pulping and bleaching technologies. EPA also used information
about the pulping and bleaching technologies in place at each mill.
The data used to calculate production-normalized pollutant loadings were derived
from six sources:
1) Short-term studies (1988-1993);
2) EPA/industry long-term variability study (1991-1992);
3) Self-monitoring data (supplied by mills subject to Subparts B or E);
4) Data collected by EPA and industry since the proposal of these
regulations;
5) The 1994 NCASI dioxin survey; and
6) Bleach plant chloroform data collected by NCASI.
The short-term sampling database contains the results of two- to three-day sampling episodes at
thirteen mills. These sampling episodes took place in 1988 through 1993. The long-term
variability study database contains the results of intensive sampling efforts at eight mills in 1991
and 1992. The self-monitoring database contains the results of analyses performed by individual
mills which were collected from responses to Question 49 of the EPA 1990 National Census of
Pulp, Paper, and Paperboard Manufacturing Facilities Questionnaire (reflecting the period from
1985 through 1991). The data collected by EPA and industry since the proposal of these
regulations were mainly from mills using complete substitution of chlorine dioxide for chlorine,
including mills using extended cooking and/or oxygen delignification. EPA also collected data
from three mills using TCP bleaching. The 1994 NCASI dioxin survey is, in part, a tabulation of
end-of-pipe effluent loadings for TCDD and TCDF reported by mills during 1994, or in earlier
years if the mill did not analyze for these pollutants during 1994.
9.2.1 Pollutant Loading Calculations
The data in each database are used to calculate production-normalized pollutant
loadings for the bleach plant and/or final effluent for each mill. Three types of data are needed to
calculate a production-normalized loading: a pollutant concentration, a wastewater flow rate, and
a brown stock pulp flow rate. For example, the concentration of AOX in the final effluent from a
mill may be 15,000 " g/L. The final effluent flow rate from this mill may be 25,000 m3/day. The
9-3
-------
Section 9 - Pollutant Reduction Estimates
brown stock pulp flow rate into bleaching for this mill may be 1,200 kkg brown stock pulp/day
(kkg/day). These data are used to calculate daily (kg AOX/day) and production-normalized (kg
AOX/kkg brown stock pulp) mass loadings for the final effluent for this mill.
Daily Mass
Loading
15,000 £3.
I j
1000
L
25,000
nr
m
day
io9M
ng
375
kg_
day
Production-
Normalized
Mass
Loading
375
day
1,200
kkg
day
0.313
kkg
Another example is shown below for a bleach plant effluent. Depending on the
mill, the bleach plant effluent may be characterized by one or more separate samples. At some
mills, all bleach plant wastewaters are discharged to a single sewer. For these mills, production-
normalized mass loadings are calculated as in the previous example. More commonly, however,
bleach plant wastewaters are discharged to two separate sewers: one that handles acidic
wastewaters (e.g., discharges from chlorine, chlorine dioxide, hypochlorite, and ozone stages) and
one that handles alkaline wastewaters (e.g., discharges from extraction stages). At these mills,
separate samples of acid and alkaline sewers (or filtrates that are discharged to sewers) were
collected and analyzed. For these mills, EPA calculated separate mass loadings for each pollutant
in each sewer for each day and then summed the production-normalized loadings to obtain a
"bleach plant effluent" loading for that day, as shown below. The flow rates of the acid and
alkaline sewers are usually measured (or estimated) separately, but only one brown stock
production value is applicable to the calculation.
9-4
-------
Section 9 - Pollutant Reduction Estimates
Acid Filtrate
Alkaline Filtrate
Acid Filtrate
Mass Loading
Alkaline Filtrate
Mass Loading
Bleach Plant
Mass Loading
Production-
Normalized
Bleach Plant
Mass Loading
Chloroform Wastewater Flow
Concentration Rate
( (mVday)
25 18,000
15 15,000
ic ^g v if)3 L v i Q nnn m v ir>9
L m3 day
1 C ^g Y 1 Q3 L v 1
-------
Section 9 - Pollutant Reduction Estimates
9.2.2
Data Sources
The data sources used to calculate production-normalized pollutant loadings are
described in the following sections.
9.2.2.1
Short-Term Studies
For the short-term studies, loadings for each pollutant were calculated at the
bleach plant and the final effluent for 11 mills. These sampling episodes were either two or three
days in length. Specific details varied from mill to mill. Depending on the mill, sampling location,
and pollutant, two- or three-day composite samples were collected, three consecutive 24-hour
composite samples were collected, or three nonconsecutive 24-hour composite samples were
collected. Where multi-day composite samples were collected, average wastewater and brown
stock pulp flow rates for the multi-day period were also obtained. Where 24-hour composite
samples were collected, 24-hour average wastewater and brown stock pulp flow rates were
obtained. From these data, EPA calculated an average pollutant mass loading for each stream at
each mill regardless of the number of days of samples collected or the number of analytical
measurements available. EPA calculated each mass loading based on between one and three
analytical data values.
For the short-term studies, EPA performed a preliminary review of the data quality
associated with each of the 11 mills and discarded from the database some individual analytical
data points that did not meet the criteria of the preliminary review (1). A second review was
performed by EPA's Sample Control Center (SCC) for four mills (of the 11) that represented
technology options EPA considered at proposal (2). SCC used the same data quality review
criteria for data from the short-term sampling episodes as for the variability study.
9.2.2.2
EPA/Industry Long-Term Variability Study
The long-term variability study consisted of nine 24-hour composite samples
collected in the summer of 1991 and nine 24-hour composite samples collected in the winter of
1991-92 at eight mills. Like the short-term sampling episodes, the average wastewater and brown
stock pulp flow rates were obtained for each 24-hour sampling period. Again, SCC reviewed the
quality of the individual analytical data points and discarded from the database the analytical
results that did not meet the method quality control criteria (3). In general, when data were not
discarded, 18 data points were available for each pollutant at each sampling point. From the
available data, an average mass loading was calculated for each pollutant, at each sampling point,
for each mill.
9.2.2.3
Self-Monitoring Data
The self-monitoring data consisted of a wide range of analyses varying from one
pollutant in one stream at some mills to many pollutants in many streams at other mills. The
analytical results submitted by each mill for various pollutants in various streams over the period
9-6
-------
Section 9 - Pollutant Reduction Estimates
from 1985 through 1992 were entered into a database. Average annual flow rates for each mill
(bleach plant effluent, final effluent, and brown stock pulp) and fiber furnish and other operating
data were obtained from responses to EPA's 1990 Census Questionnaire (reflecting data for 1989)
and follow-up letters for mills that made subsequent process changes (primarily reflecting data for
1990 and 1991, but also 1992 for some mills). Because of changes in the industry's pulping and
bleaching practices since 1985-92 and the availability of more recent data, EPA limited the use of
the self-monitoring data. Self-monitoring AOX data were used to estimate the baseline loadings
of mills that have not made any major process changes in recent years.
9.2.2.4 Data Collected Since Proposal of the Regulations
Since the December 1993 proposal of these regulations, EPA, NCASI, and various
mills have collected additional wastewater monitoring data. Most of the data were collected at
mills using complete substitution of chlorine dioxide for chlorine (ECF bleaching) on at least one
bleach line. Some of the mills also used extended cooking, oxygen delignification, and/or TCP
bleaching.
Many of these additional studies followed the format of the short-term studies, i.e.,
three consecutive days of sampling at the bleach plant and final effluents with analysis of the
samples for the pollutants proposed for regulation at each sampling point. However, specific
details varied from mill to mill and several mills collected samples over longer periods of time.
A thorough review of the results of the pollutant analyses was performed by SCC
for all results submitted with supporting quality assurance/quality control data. SCC reviewed the
pollutant analytical data quality and discarded analytical results that did not meet method quality
control criteria. SCC used similar data quality review criteria for data from these sampling
episodes as for the long-term variability study. From the resultant database, EPA calculated an
average pollutant concentration or mass loading for each stream at each mill. The results of the
SCC review were reported in a series of quality assurance data review memoranda which can be
found in Section 21.6 of the docket supporting these regulations.
Since proposal, EPA also received some final effluent AOX data from bleached
papergrade kraft mills in Alberta, Canada. While some of these data were used to develop the
AOX limitations for each option, these data were not used to estimate the baseline AOX loadings
of the industry.
9.2.2.5 1994 NCASI Dioxin Survey
Each year since the 104-Mill Study (4) was conducted in 1988, NCASI has asked
each mill for its most recent TCDD and TCDF data for bleached pulps, final mill effluents, and
wastewater treatment sludges. If a mill does not respond to the request in a particular year,
NCASI uses data from a previous year for its annual compilation. EPA used the final effluent
9-7
-------
Section 9 - Pollutant Reduction Estimates
data from the 1994 survey to estimate the baseline discharges of TCDD and TCDF for the
industry (5). The list of currently operating bleached papergrade kraft and sulfite pulp mills in
NCAST's 1994 survey matches the list in EPA's baseline database (reflecting mid-1995
operations). The survey also includes data for dissolving kraft and dissolving sulfite mills which
EPA did not use in the analysis reported here.
NCASI calculates effluent mass loadings using detected concentrations or one-half
the reported detection limits. (Note that this approach is slightly different than the approach EPA
used to make a separate estimate using its own database and one-half of the minimum level for
non-detect values.) NCASI reports results in units of milligrams/day. EPA multiplied the
detected concentrations or one-half the reported detection limits by the 1994 average daily
wastewater discharge for each mill (as reported in the survey) and by 350, an estimate of the
number of days that each mill produced bleached pulp. After several unit conversions, EPA
reported results in units of g/yr of TCDD and TCDF discharged by each mill.
9.2.2.6 NCASI Bleach Plant Chloroform Data
In December 1988, NCASI published Technical Bulletin No. 558, which presented
total bleach plant chloroform generation rates for a variety of kraft and sulfite bleaching sequences
(6). The report also provided information that could be used to estimate the bleach plant effluent
chloroform loading. These data were used to supplement EPA's sampling data for the bleach
sequences studied.
9.3 Industry Baseline Pollutant Loadings
In support of the 1993 proposal, EPA developed a procedure for estimating the
baseline bleach plant and final effluent pollutant mass loadings for each mill. The procedure used
all available data from each mill. EPA created a model with which to estimate loadings where
data were not available. The model estimated the average pollutant loadings achieved by mills
using several combinations (one combination for each option evaluated for each subcategory) of
pulping and bleaching technologies. Even though the procedure was complicated and labor-
intensive, EPA received few public comments on this baseline estimation procedure.
Instead, commenters objected to EPA's use of data dating, in some cases, to the
1988 104-Mill Study, to characterize the industry's 1993 pollutant discharges. As discussed
above, EPA has addressed this comment by updating the estimate of the baseline pollutant
loadings. Changes to the baseline estimate include:
Updated data collected by EPA, NCASI, and individual facilities;
Limited use of self-monitoring data dating from 1985-1992 (where more
recent data were not available); and
9-8
-------
Section 9 - Pollutant Reduction Estimates
Incorporation of the results of NCASI's 1994 Dioxin Survey into the
analysis, made available to EPA in late 1995.
EPA has simplified the baseline estimation procedure used at proposal. The
revised procedure uses three models to reflect process technology status and thereby estimate the
baseline loadings for all mills. These models replace the complicated baseline-estimation
procedure used at proposal where sampling data were used to represent the mills from which
sampling data were available, and models were used to fill data gaps for mills from which no data
(or out-of-date data) were available. The three models were developed for different pollutant
groups:
1) AOX, chlorinated phenolic compounds, TCDD, and TCDF;
2) Chloroform; and
3) COD and color.
EPA determined that separate models were necessary because, for each group of
pollutants, different process criteria more accurately predict pollutant loadings. For example, for
final effluent AOX loads, the furnish pulped, the use of extended pulping technologies, and the
percent chlorine dioxide substitution were strongly related to the pollutant loading. The three
models are summarized below, and are described in detail in subsequent sections.
Model
1
2
O
Pollutants
AOX, chlorinated phenolic
compounds, TCDD, TCDF
Chloroform
COD, color
Main Process Criteria Predictive
of Pollutant Loadings
Furnish pulped;
Extended pulping status;
Percent C1O2 substitution
Hypochlorite use;
Percent C1O2 substitution
Screen room status;
Pre-bleaching kappa number
9.3.1
AOX, Chlorinated Phenolic Compounds, TCDD, and TCDF
The model used for AOX, chlorinated phenolic compounds, TCDD, and TCDF
("AOX model") is shown in Table 9-1. Each bleach line at each papergrade kraft and sulfite mill
was assigned to one of the eleven baseline groups. These groups are similar to the groups used at
proposal to estimate baseline loadings (although fewer groups exist now than at proposal). The
groups are based on the furnish pulped, extended pulping equipment in place, pre-bleaching kappa
number achieved, and bleach sequence of each bleach line.
9-9
-------
Section 9 - Pollutant Reduction Estimates
Because a mill may have more than one type of bleach line, assigning a complete
mill to a group is more subjective than assigning a single bleach line. Sixteen of the 84 mills in the
Bleached Papergrade Kraft and Soda Subcategory for which EPA estimated pollutant reductions
operate with bleach lines in more than one group. In these cases, judgment was used to select a
group that would result in the best estimate of the combined mill effluent.
EPA assigned the AOX model groups at the same time that EPA made preliminary
estimates of the costs of compliance with the revised options. Because capital costs are related to
equipment, group assignments were based on pulping equipment in place rather than strict
adherence to the pre-bleaching kappa number ranges reported in Table 9-1.
AOX Group F was formed to aggregate the data from mills that did not achieve
the kappa numbers expected from the pulping technologies they employ. Three softwood bleach
lines with some form of extended delignification and ECF bleaching did not achieve kappa
numbers as low as expected for mills in Group H. (EPA believes that these mills were not
operating optimally.) Conversely, two hardwood mills achieved kappa numbers lower than
expected. As shown in Table 9-2, to estimate AOX baseline loads, these bleach lines were
grouped with bleach lines employing no extended pulping and ECF bleaching (i.e., Groups E and
F were combined).
9.3.1.1 AOX
The average AOX loadings for papergrade kraft and sulfite baseline groups are
summarized in Table 9-2. Sampling data were not available for all the groups identified in Table
9-1; therefore, several groups identified in Table 9-1 were consolidated (as shown in Table 9-2).
EPA assumed that the effluent characteristics from mills in groups with similar bleach plant
operating conditions were similar. EPA also used self-monitoring data to characterize the effluent
AOX loadings for the consolidated Group A, B, and C.
In general, the AOX baseline loadings listed in Table 9-2 represent the average of
the values available from mills producing either hardwood or softwood but not from mills
producing both hardwood and softwood. Average final effluent AOX production-normalized
loadings were also calculated for mills in each group (where data were available) that produce
both hardwood and softwood, and usually fell within the range of the values listed above.
However, the only data available from mills in Groups G and I were from mills producing both
hardwood and softwood, so this loading (0.38 kg/kkg) was applied to both the hardwood and
softwood columns in Table 9-2. For TCF mills (Group K), final effluent AOX loadings were
assumed to be zero.
In the baseline model, each mill was designated as a mill normally producing only
hardwood, only softwood, or as a mill producing both hardwood and softwood. Mills producing
only hardwood or softwood were assigned the baseline loadings listed in Table 9-2 (in accordance
with its assigned grouping). Mills producing both hardwood and softwood were assigned a
baseline loading by multiplying the loadings listed in Table 9-2 by the percent of hardwood and
9-10
-------
Section 9 - Pollutant Reduction Estimates
softwood brown stock pulp that is bleached at each mill. (The hardwood and softwood brown
stock pulp rates that were used here were also used in the compliance cost model.) For these
mills, the baseline loadings would be between the two loadings listed for mills in each
consolidated AOX group.
9.3.1.2 TCDD, TCDF, and the 12 Chlorinated Phenolic Compounds
EPA used the same AOX model but a slightly different procedure to estimate the
baseline loadings of TCDD, TCDF, and the chlorinated phenolic compounds. These pollutants
are not consistently detected in pulp mill effluents, particularly at higher levels of chlorine dioxide
substitution or at TCF mills. The mills that use extended delignification technologies have lower
bleach plant effluent flows than mills using conventional pulping because they have modernized
and use less water to wash pulp more efficiently. To account for these flow differences, EPA
estimated average bleach plant concentrations of TCDD, TCDF, and chlorinated phenolic
compounds. The concentrations were calculated by averaging any detected values with the
number of non-detected values multiplied by one-half the minimum level of the analytical method
for each pollutant. For example:
ML\
n
, 2 j
1
average concentration
where a, b, and c are detected concentrations, n is the number of results reported as "not
detected", and ML is the minimum level for the analytical method. The ML for the analytical
method was used rather than using various sample-specific detection limits (some of which may
be greater than the method ML, due in most cases to reduced sample volume), in order to simplify
the procedure. This calculation was used for all data except for TCF mills, EPA assumed that the
concentration and loading of these pollutants at TCF mills was zero.
To calculate a mass loading, the average group concentration was multiplied by an
average production-normalized bleach plant flow rate for the mills in each group. Production-
normalized bleach plant flow rates (based on sampling data collected at bleached papergrade kraft
mills) are shown in Table 9-3. Although the flow rates were derived from kraft mill data, because
of limited sampling data for papergrade sulfite mills, these flow rates were used to estimate
pollutant loadings from papergrade sulfite mills. The group average concentrations for TCDD,
TCDF, and the 12 chlorinated phenolic compounds are shown in Tables 9-4 through 9-17. Group
average concentrations are reported as "ND" when the compound was never detected among the
data available for that group of mills. For these cases, a concentration of one-half the minimum
level for the compound was used to calculate the mass loading for that group of mills.
EPA calculated average concentrations and production-normalized flow rates
using sampling results from mills bleaching either hardwood or softwood but not from mills
bleaching a mixture of hardwood and softwood. For mills bleaching a mixture of hardwood and
softwood, EPA calculated baseline loadings by multiplying the appropriate concentrations and
9-11
-------
Section 9 - Pollutant Reduction Estimates
flows by the percent of hardwood and softwood brown stock pulp that is bleached at each mill.
After a baseline mass loading was calculated for each bleach line at each mill, the
individual pollutant loads were summed for mills with multiple bleach lines to give the total bleach
plant pollutant load for the mill. Final effluent mass loadings were calculated from the estimated
bleach plant loadings. For TCDD and TCDF, the final effluent loads were assumed to be the
same as the bleach plant loads for each mill because EPA's data do not show that TCDD or TCDF
are degraded by biological treatment. This method provides an upper-bound estimate of the final
effluent loads because some of the TCDF (and perhaps TCDD if present) may adsorb to the
sludge.
For the 12 chlorinated phenolic compounds, 45 percent of the bleach plant load at
each mill was assumed to be removed in the treatment system (or the associated POTW for
indirect discharging mills). The 45 percent removal efficiency was based on an NCASI study of
the removal of specific chlorinated phenolic compounds in various wastewater treatment systems
(7).
For TCDD and TCDF, a second baseline calculation was made using data
compiled by NCASI for 1994. EPA used the NCASI data to calculate effluent mass loadings for
each mill using the detected concentrations or one-half the reported detection limits multiplied by
the 1994 average daily wastewater discharge for each mill (as reported in the survey) and by 350,
an estimate of the number of days that each mill produced bleached pulp. After several unit
conversions, the results were reported in units of g/yr of TCDD and TCDF discharged by each
mill. (Note that one-half the reported detection limits were used rather than one-half the
minimum level because 1) the analytical methods used are not reported and 2) the reported
detection limits are generally lower than the minimum level for Method 1613.)
The two dioxin baseline estimates are presented below:
Baseline Final Effluent Discharge
from All Bleached Papergrade
Kraft and Soda Mills
TCDD
TCDF
Estimate Using Bleach
Plant Loads (g/yr)
15.2
115
Estimate Using 1994
NCASI Dioxin Survey
Results (g/yr)
13.1
41.7
The TCDD baseline estimates using the two calculation procedures are very close.
The TCDF baseline estimates are not as close because TCDF is occasionally detected just above
the method minimum level in some bleach plant effluents but is diluted to concentrations below
the minimum level at the final mill effluent. Because of this fact, EPA believes the TCDF baseline
estimate based on the bleach plant data is more appropriate.
9-12
-------
Section 9 - Pollutant Reduction Estimates
After July 1996, EPA again recalculated the effluent reductions. The baseline
remained mid-1995, but the methodology outlined above was modified slightly. EPA used two
different procedures for handling concentrations of chlorinated pollutants reported as less than the
detection limit. For mills that used chlorine-containing bleaching agents, EPA used one-half the
method minimum level to estimate effluent discharge loadings, i.e., EPA assumed the chlorinated
pollutants were present at concentrations too low to measure by current analytical methods. (This
approach is the same one used previously for handling non-detected concentrations.) For mills
that used TCF bleaching, EPA assumed that chlorinated pollutants were not present; EPA
assumed that TCF mills would discharge zero kilograms per year of AOX and the individual
chlorinated pollutants (chloroform, TCDD, TCDF, and the 12 chlorinated phenolic compounds).
This minor methodology change affects the baseline discharges for two mills (one
kraft and one sulfite) that currently use TCF bleaching sequences and seven mills (one kraft and
six sulfite) that are expected to use a TCF bleach sequence after promulgation of these
regulations. These same baseline and post-promulgation loading changes were also applied to the
1994 NCASI dioxin profile data (for TCDD and TCDF) for the affected mills to adjust EPAs
effluent reduction estimates based on these data in the same manner.
9.3.2 Chloroform
The model used for chloroform is shown in Table 9-18. Kraft and sulfite mills
were divided into six baseline groups. The groups are defined by the bleach sequence used at
each mill, specifically whether mills used hypochlorite in the bleach sequence and what level of
chlorine dioxide substitution was used in the first bleaching stage. For the July 1996 Notice, EPA
used the same value for ECF and TCF bleach lines because chloroform was not detected in final
effluents from either type of mill. For this recalculation, EPA used the same loading for ECF mills
(Group E) but used zero for TCF mills (Group F). Mills with more than one bleach line were
assigned a group based on whether hypochlorite was used on any bleach line and what the
average production-normalized chlorine dioxide substitution in the first bleaching stage was for all
the fiber lines at the mill.
In addition to bleach sequence, the type of washers used in the bleach plant (8)
may affect bleach plant effluent chloroform loadings. Bleach plant effluent chloroform loadings
are somewhat greater at mills using low air-flow washers (e.g., pressure or diffusion washers)
than at mills using high air-flow washers (e.g., vacuum-drum washers). This effect on bleach
plant chloroform loadings was not accounted for in the baseline model because EPA does not
know what type of washers are used on each bleach line at each mill in the country. In general,
most mills use vacuum-drum washers and the most data are available from mills using these
washers.
The average bleach plant and final effluent loadings for mills in the various groups
are summarized in Table 9-19. For bleach plants, EPA and NCASI data were used (6,9). For
final effluents, no NCASI data were available so only EPA data were used. (Note: Although EPA
calculated bleach plant effluent chloroform reductions, they were not used directly in the benefit
9-13
-------
Section 9 - Pollutant Reduction Estimates
analyses. Air emission reductions of chloroform were estimated to assess an important non-water
quality environmental benefit (see Section 11 of this document) while final effluent pollutant
reductions were used to evaluate aquatic environmental impacts.)
9.3.3 COD and Color
EPA proposed limitations for COD and color. The in-plant process changes that
form the bases of the BAT limitations, PSES, and the BMPs reduce final effluent COD and color
loadings. For this reason, and to fully evaluate the options considered, EPA estimated effluent
loadings reductions for COD and color. For reasons stated in the preamble and a separate
document (10), EPA intends to develop COD limitations for kraft and sulfite mills in a future
rulemaking. EPA also decided not to develop national effluent limitations guidelines and
standards for color for Subparts B and E; instead, permit writers must continue to develop site-
specific effluent limitations for color as necessary to achieve applicable water quality standards
(see33U.S.C. 13
The model used for estimating baseline COD and color loadings was based on
screen room status and pre-bleaching kappa number. EPA assumed that the available final mill
effluent loadings were derived from the pulping and bleaching operations at each mill. The model
did not specifically account for unusual loadings from papermaking or other on-site pulping (e.g.,
mechanical or secondary fiber) operations although these operations were used at some of the
mills for which data were available.
As shown in Table 9-20, kraft and sulfite mills were divided into four baseline
groups. Mills with more than one fiber line were assigned a group based on whether any screen
room at the mill was open and on the average production-normalized pre-bleaching kappa number
for all the lines at the mill. COD and color baseline loadings and reductions were calculated for
pulp mill final effluents, but not for bleach plant effluents. The average final effluent baseline
loadings for mills in the four groups are presented in Table 9-21.
9.4 Pollutant Loadings After Implementation of the Control Options
After estimating baseline loadings for each mill, EPA estimated the reduction in
pollutants discharged to receiving streams attributable to the three principal BAT and PSES
technology options considered by EPA. The long-term average (LTA) performance of each
option for each pollutant was subtracted from each mill's baseline discharge. The LTA pollutant
loadings for each option are presented in Table 9-22.
See Section 8 for a description of the technology options considered for BAT and
PSES for the Bleached Papergrade Kraft and Soda and Papergrade Sulfite Subcategories.
EPA evaluated the performance of each technology option described above by
calculating a LTA loading for each pollutant of concern. EPA used data from facilities that
employed processes most similar to the technology components of each option. The sampling
9-14
-------
Section 9 - Pollutant Reduction Estimates
data that were used to calculate the performance of each option are presented in a separate
document (11). These data were a subset of the data used to develop baseline pollutant loadings.
As described previously, for the estimation of baseline, EPA estimated pollutant loadings for
several combinations of pulping and bleaching operations in addition to the BAT options.
The LTAs used in these estimates for AOX, TCDF, and chloroform are not exactly
the same values as the LTAs used to develop the limitations but they are reasonably close. The
reason that EPA employed slightly different LTAs to estimate loading reductions than to calculate
limitations is that the two LTAs served difference purposes. EPA used the LTAs for load
reductions to help in its evaluation of the technology options being considered; while highly
informative, the reductions were not statutory decision criteria. In contrast, EPA refined its LTAs
before using them to calculate limitations, because the calculations would lead to the imposition
of enforceable permitting and pretreatment requirements. Therefore, EPA concluded that it
needed to exercise more analytical rigor in determining the limitations-related LTAs than the
LTAs used ultimately for the non-statutory benefits analysis. This document presents the
estimated loadings and reductions used to select the options. The final LTAs used to develop the
limitations are described in the Statistical Support Document for the Pulp and Paper Industry:
SubpartB(12).
The long-term averages for AOX and TCDF used in the pollutant reduction
calculations for kraft Options A and B were calculated in early 1997. The AOX LTAs are based
on more data than the LTAs presented in EPA's July 1996 Notice. The TCDF long-term average
is based on the data presented in EPA's July 1996 Notice.
The chloroform LTA calculated for kraft Option A, 0.0003 kg/kkg (calculated
from non-detect results using one-half the method minimum level), was also applied to kraft
Option B because chloroform is not expected to be detected in the final effluent from mills with
either technology. TCDD and the 12 chlorinated phenolic compounds are also not expected to be
detected in the final effluent from kraft Option A or B mills. The LTAs for these pollutants were
calculated in the manner described in Section 9.1.2: one-half the minimum level for each pollutant
was multiplied by a group-average production-normalized bleach plant effluent flow rate. For
kraft Options A and B, the LTA loadings are the same as the baseline loadings for Groups E and
H/J, respectively. The COD and color LTAs for kraft Options A and B were calculated by EPA
in February 1996(13).
EPA had little end-of-pipe data with which to calculate LTA pollutant loads for
sulfite mills. For the calculation of pollutant reductions, EPA assumed that TCF mills would not
discharge any chlorinated compounds, including AOX, and the pollutant concentrations measured
at ECF sulfite mills would be similar to those concentrations at ECF kraft mills (i.e., mostly not
detected). The sources of data used for sulfite mills are summarized in Table 9-23. For the
calculation of pollutant reductions, pollutant concentrations for the sulfite ECF option were
transferred from kraft Option A.
Table 9-22 presents the estimated loadings for each pollutant. The loadings in
Table 9-22 were subtracted from the baseline loadings to determine the pollutant reductions for
9-15
-------
Section 9 - Pollutant Reduction Estimates
each mill and then were summed for all mills in each subcategory. If the baseline loading for any
mill(s) was lower than the option LTA, the removal was set to zero.
9.5 Pollutant Reductions
The pollutant reductions calculated using the methodology described in this
document are summarized in Table 9-24. Mill-by-mill estimates are confidential business
information (CBI) and are included as such in the CBI portion of the rulemaking record (14). In
Table 9-24, for bleached papergrade kraft mills, the column headings in the table differentiate
between Options A and B. For papergrade sulfite mills, the same results are shown in each
column because only one option was evaluated for each segment: TCP for the calcium-,
magnesium-, or sodium-based segment and ECF for the ammonium-based and specialty-grade
segments. Note that the values in each column may not add, due to rounding.
For the kraft mill TCF options that were evaluated by EPA, the estimated pollutant
reductions for AOX and the individual chlorinated pollutants are the same as the baseline loading
estimates, because EPA assumes that these compounds would not be present in measurable
quantities in the effluent from these mills. These values are not shown in Table 9-24. For COD
and color, EPA would expect the pollutant reductions for the TCF options (including oxygen
delignification, closed screen rooms, and improved brown stock washing) to be at least equivalent
to those that were estimated for kraft Option B.
9.6 References
1. Spengel, D., "Memorandum to the Record: Short-Term Study Data Review."
Prepared by Radian Corporation for EPA. Record Section 5.1, DCN 05833,
October 27, 1993.
2. Compilation of Data Quality Review Documentation for the Pulp and Paper Short
Term Study: Episodes 1375. 1376. 1691. and 1692. U.S. EPA Sample Control
Center. Record Section 5.2, DCN 40221, November 11, 1992.
3. Compilation of Data Quality Review Documentation for the Summer and Winter
Phases of the Pulp and Paper Variability Study. U.S. EPA Sample Control Center.
Record Section 5.2, DCN 40216-40220, October 1, 1992.
4. R.C. Whittemore, L.E. LaFluer, WJ. Gillespie, G.A. Amendola, and J. Helms.
"U.S. EPA/Paper Industry Cooperative Dioxin Study: The 104 Mill Study."
Chemosphere. 20(10-2): 1625-1632, Record Section 8.2, DCN 05267, 1990.
5. Gillespie, W. Progress in Reducing the TCDD/TCDF Content of Effluents. Pulps
and Wastewater Treatment Sludges from the Manufacturing of Bleached Chemical
Pulp (also known as the NCASI 1994 Dioxin Profile! NCASI. Record Section
19.2.3, DCN 13838, October 1995.
9-16
-------
Section 9 - Pollutant Reduction Estimates
6. Results of Field Measurements of Chloroform Formation and Release From Pulp
Bleaching. NCASI Technical Bulletin No. 558. Record Section 2.3.2, DCN
03815, December 1988.
7. Wiegand, Paul, Doug Barton, and Larry LaFleur. "Factors Influencing the
Generation and Treatability of Chlorinated Phenolic Compounds and AOX in
Bleached Chemical Pulp Mill Waste Waters" In: NCASI Special Report No. 94-
01 - NCASI Technical Workshop - Effects of Alternative Pulping and Bleaching
Processes on Production and Biotreatability of Chlorinated Organics. NCASI.
Record Section 20.2.1, DCN 12415, February 1994.
8. Examination of Data Relevant to EPA's Proposed Effluent Limitations Guideline
for Chloroform at Bleached Papergrade Kraft Subcategory Mills. NCASI. Record
Section 21.12, DCN 13967, February 1996.
9. Chloroform Generation at Bleach Plants with Low Molecular Chlorine Usage or
Split Chlorination. NCASI Technical Bulletin No. 605. Record Section 2.3.2,
DCN 05495, March 1991.
10. Analysis of Data Available for Development of COD Limitations. Prepared by
Eastern Research Group for EPA. Record Section 22.4, DCN 13958, July 1996.
11. Data Available for Limitations Development for Toxics and Nonconventional
Pollutants. EPA, Washington DC, Record Section 22.6, DCN 14494, 1997.
12. Statistical Support Document for the Pulp and Paper Industry: Subpart B. EPA,
Washington DC, Record Section 22.5, DCN 14496, 1997.
13. M. Smith. Revised Pulp and Paper LTAs. EPA, Washington DC, Record Section
24.0, DCN 13966, February 20, 1996.
14. Mill-by-Mill Pollutant Loading and Reduction Estimates. Prepared by ERG for
EPA. Record Section 22.6, DCN 14501, July 1997.
9-17
-------
Section 9 - Pollutant Reduction Estimates
Table 9-1
Baseline Technology Groups for Bleached Papergrade
Kraft and Soda, and Papergrade Sulfite Mills:
AOX, Chlorinated Phenolic Compounds, TCDD, and TCDF
AOX
Group
A
B
C
D
E
F
G
H
I
J
K
Example Bleach
Sequence(s)
CEHd
CEHD, CED
C/DEH, C/DEHDED,
C/DED, C/DEDED
D/CEDED,
D/CEopDEpD
DEDED, DEopDD
HW: C/DEDED,
C/DEoDEP
SW:ECorODwith
DEDED, DEopDD
EC or OD with
C/DEDED, D/CEDED
EC or OD with
DEDED, DEopDD
EC and OD with
C/DEDED, D/CEDED
EC and OD with
DEDED, DEopDD
TCFd
EC or
OD?b
No
No
No
No
No
No
Yes
Yes
Yes
Both
Both
Maybe
Chlorine
Used?
Yes
Yes
Yes
Yes
No
Yes
No
Yes
No
Yes
No
No
Hypo-
chlorite
Used?c
Yes
Maybe
Maybe
Maybe
No
Maybe
No
Maybe
No
Maybe
No
No
Percent
ofC!O2
Used
none on
site
0 in first
stage
<70
70 to 100
100
<100
100
<100
100
<100
100
none
Kraft Mill Kappas
HW
>13
10 to 13
-
10 to 13
<10
sw
>27
-
20 to
27
15 to
<20
<15
aGroups E and H include the bleached papergrade kraft mills that represent the two ECF options under consideration
for that subcategory. Groups E and K include the papergrade sulfite mills that represent the TCP and ECF options
under consideration for that subcategory.
bEC is extended cooking (e.g., MCC, EMCC, RDH or SuperBatch) and OD is oxygen delignification.
°"Maybe" indicates that the pollutant loadings are not significantly different for mills using or not using hypochlorite.
dMills using this bleaching sequence do not usually bleach to full brightness.
9-18
-------
Section 9 - Pollutant Reduction Estimates
Table 9-2
AOX Baseline Loadings
Consolidated AOX Groups
Kraft A, B, C
Kraft D
Kraft E,F
Kraft G, I
Kraft H, J
Kraft & Sulfite K
Sulfite A
Sulfite B, C
Final Effluent (kg/kkg)
Hardwood
1.61
0.56
0.27
0.38a
0.153
0.00
5.82
1.61
Softwood
3.00
1.50
0.39
0.38a
0.153
0.00
5.82
3.00
The only data available from mills in Groups G and I were from mills producing both hardwood and softwood so
this loading was applied to both the hardwood and softwood columns.
9-19
-------
Section 9 - Pollutant Reduction Estimates
Table 9-3
Production-Normalized Kraft Bleach Plant Flow Rates3
AOX Groups"
A, B, C, D, E, F(HW)
F(SW), G, H, I, J
Kc
Type of Mill
Mills Without EC or OD
Mills With EC and/or OD
TCP Mills
Hardwood Mills
(m3/kkg)
24.7
19.7
11.6
Softwood Mills
(m3/kkg)
37.1
24.7
18.3
The average flow rates presented in this table were derived from bleached papergrade kraft mills. However, these
flow rates were used to estimate mass loads of chlorinated phenolic compounds, TCDD, and TCDF from papergrade
sulfite mills as well as bleached papergrade kraft and soda mills.
bPapergrade sulfite mills fall in groups A, B, C, and K.
These flow rates were not used to calculate the final set of reduction estimates in 1997; instead, EPA assumed that
TCP mills discharged no chlorinated organic pollutants.
Table 9-4
TCDD Baseline Concentrations
Consolidated
AOX Groups3
A,B, C
D
E
F
G,I
H,J
K
Bleach Plant Effluent (pg/L)
Minimum Level = 10 pg/L
Hardwood
Concentration
5.63
5.24
ND
7.80
5.27
9.54
ND
Detects
2
1
0
1
1
4
0
Non-detects
70
41
6
4
43
24
4
Softwood
Concentration
64.6
6.38
ND
ND
8.09
ND
ND
Detects
12
3
0
0
6
0
0
Non-detects
6
37
29
30
39
86
10
Tapergrade sulfite mills fall in groups A, B, C, and K.
9-20
-------
Section 9 - Pollutant Reduction Estimates
Table 9-5
TCDF Baseline Concentrations
Consolidated
AOX Groups3
A, B, C
D
E
F
G,I
H, J
K
Bleach Plant Effluent (pg/L)
Minimum Level = 10 pg/L
Hardwood
Concentration
6.99
8.83
NDb
52
4.87
10.6
ND
Detects
12
10
0
2
4
8
0
Non-detects
60
32
6
3
40
20
4
Softwood
Concentration
593
6.47
NDb
ND
7.93
9.62
ND
Detects
15
5
0
0
3
13
0
Non-detects
2
35
29
24
41
73
13
Tapergrade sulfite mills fall in groups A, B, C, and K.
bTCDF was not detected at the mills for which data are included in EPA's database but comments on the July 1996
Notice indicate that TCDF is detected at some mills of this type. Therefore, the reductions may be underestimated.
Table 9-6
Trichlorosyringol Baseline Concentrations
Consolidated
AOX Groups"
A,B, C
D
E
F
G,I
H, J
K
Bleach Plant Effluent C g/L)
Minimum Level = 2.5 " g/L
Hardwood
Concentration
12.2
16.4
ND
20.9
7.63
ND
ND
Detects
57
22
0
5
26
0
0
Non-detects
25
22
17
8
15
44
4
Softwood
Concentration
1.30
ND
ND
ND
ND
ND
ND
Detects
1
0
0
0
0
0
0
Non-detects
31
38
79
28
43
100
5
Tapergrade sulfite mills fall in groups A, B, C, and K.
9-21
-------
Section 9 - Pollutant Reduction Estimates
Table 9-7
2,4,5-Trichlorophenol Baseline Concentrations
Consolidated
AOX Groups3
A, B, C
D
E
F
G,I
H, J
K
Bleach Plant Effluent C g/L)
Minimum Level = 2.5 " g/L
Hardwood
Concentration
ND
4.18
ND
8.92
ND
ND
ND
Detects
0
5
0
5
0
0
0
Non-detects
82
40
17
8
41
44
9
Softwood
Concentration
1.80
ND
ND
ND
1.35
ND
ND
Detects
5
0
0
0
1
0
0
Non-detects
27
38
79
28
42
102
10
Tapergrade sulfite mills fall in groups A, B, C, and K.
Table 9-8
2,4,6-Trichlorophenol Baseline Concentrations
Consolidated
AOX Groups3
A,B, C
D
E
F
G,I
H, J
K
Bleach Plant Effluent C g/L)
Minimum Level = 2.5 " g/L
Hardwood
Concentration
18.2
8.34
ND
25.8
7.42
ND
ND
Detects
59
31
0
11
35
0
0
Non-detects
23
14
17
2
9
44
9
Softwood
Concentration
46.3
12.8
ND
ND
44.5
1.36
ND
Detects
20
36
0
0
38
1
0
Non-detects
10
0
79
28
7
101
10
Tapergrade sulfite mills fall in groups A, B, C, and K.
9-22
-------
Section 9 - Pollutant Reduction Estimates
Table 9-9
3,4,5-Trichlorocatechol Baseline Concentrations
Consolidated
AOX Groups3
A,B, C
D
E
F
G, I
H, J
K
Bleach Plant Effluent C g/L)
Minimum Level = 5.0 " g/L
Hardwood
Concentration
8.37
47.8
ND
ND
14.6
ND
ND
Detects
14
19
0
0
17
0
0
Non-detects
56
24
17
13
18
44
4
Softwood
Concentration
112
82.2
ND
ND
64.3
ND
ND
Detects
10
19
0
0
20
0
0
Non-detects
17
17
79
28
19
98
5
Tapergrade sulfite mills fall in groups A, B, C, and K.
Table 9-10
3,4,5-Trichloroguaiacol Baseline Concentrations
Consolidated
AOX Groups3
A,B, C
D
E
F
G, I
H, J
K
Bleach Plant Effluent C g/L)
Minimum Level = 2.5 " g/L
Hardwood
Concentration
5.61
1.55
ND
2.00
7.83
ND
ND
Detects
24
3
0
1
17
0
0
Non-detects
54
42
17
12
23
44
4
Softwood
Concentration
36.8
49.4
ND
ND
52.7
ND
ND
Detects
11
37
0
0
27
0
0
Non-detects
20
0
79
28
15
103
5
Tapergrade sulfite mills fall in groups A, B, C, and K.
9-23
-------
Section 9 - Pollutant Reduction Estimates
Table 9-11
3,4,6-Trichlorocatechol Baseline Concentrations
Consolidated
AOX Groups3
A,B, C
D
E
F
G, I
H, J
K
Bleach Plant Effluent C g/L)
Minimum Level = 5.0 " g/L
Hardwood
Concentration
ND
9.39
ND
ND
2.65
ND
ND
Detects
0
24
0
0
1
0
0
Non-detects
54
15
17
0
30
44
4
Softwood
Concentration
ND
2.96
ND
ND
3.00
ND
ND
Detects
0
2
0
0
5
0
0
Non-detects
0
34
79
28
28
98
5
Tapergrade sulfite mills fall in groups A, B, C, and K.
Table 9-12
3,4,6-Trichloroguaiacol Baseline Concentrations
Consolidated
AOX Groups3
A,B, C
D
E
F
G, I
H, J
K
Bleach Plant Effluent C g/L)
Minimum Level = 2.5 " g/L
Hardwood
Concentration
1.49
4.38
ND
3.04
ND
ND
ND
Detects
6
17
0
3
0
0
0
Non-detects
64
27
17
10
35
44
4
Softwood
Concentration
6.33
1.72
ND
ND
1.61
ND
ND
Detects
5
8
0
0
5
0
0
Non-detects
17
30
79
28
32
100
5
Tapergrade sulfite mills fall in groups A, B, C, and K.
9-24
-------
Section 9 - Pollutant Reduction Estimates
Table 9-13
4,5,6-Trichloroguaiacol Baseline Concentrations
Consolidated
AOX Groups3
A,B, C
D
E
F
G, I
H, J
K
Bleach Plant Effluent C g/L)
Minimum Level = 2.5 " g/L
Hardwood
Concentration
4.04
62.4
ND
8.19
2.97
ND
ND
Detects
17
19
0
3
14
0
0
Non-detects
62
24
17
10
28
44
4
Softwood
Concentration
25.9
14.0
ND
ND
10.7
1.27
ND
Detects
16
31
0
0
16
1
0
Non-detects
16
6
79
28
24
103
5
Tapergrade sulfite mills fall in groups A, B, C, and K.
Table 9-14
Tetrachlorocatechol Baseline Concentrations
Consolidated
AOX
Groups(a)
A,B, C
D
E
F
G, I
H, J
K
Bleach Plant Effluent C g/L)
Minimum Level = 5.0 " g/L
Hardwood
Concentration
3.21
6.59
ND
10.6
4.55
ND
ND
Detects
5
13
0
6
6
0
0
Non-detects
69
26
17
7
33
44
4
Softwood
Concentration
28.1
8.57
ND
ND
9.20
ND
ND
Detects
21
14
0
0
16
0
0
Non-detects
10
20
79
28
26
100
5
Tapergrade sulfite mills fall in groups A, B, C, and K.
9-25
-------
Section 9 - Pollutant Reduction Estimates
Table 9-15
Tetrachloroguaiacol Baseline Concentrations
Consolidated
AOX Groups3
A,B, C
D
E
F
G, I
H, J
K
Bleach Plant Effluent C g/L)
Minimum Level = 5.0 " g/L
Hardwood
Concentration
2.81
2.54
ND
8.19
ND
ND
ND
Detects
4
2
0
2
0
0
0
Non-detects
79
43
17
11
38
44
4
Softwood
Concentration
13.9
4.31
ND
ND
2.89
ND
ND
Detects
13
13
0
0
5
0
0
Non-detects
19
26
79
28
34
104
5
Tapergrade sulfite mills fall in groups A, B, C, and K.
Table 9-16
2,3,4,6-Tetrachlorophenol Baseline Concentrations
Consolidated
AOX Groups3
A,B, C
D
E
F
G, I
H, J
K
Bleach Plant Effluent C g/L)
Minimum Level = 2.5 " g/L
Hardwood
Concentration
1.63
1.47
ND
ND
1.38
ND
ND
Detects
9
1
0
0
5
0
0
Non-detects
66
43
17
13
36
44
9
Softwood
Concentration
ND
ND
ND
ND
2.88
ND
ND
Detects
0
0
0
0
4
0
0
Non-detects
32
38
79
28
39
102
10
Tapergrade sulfite mills fall in groups A, B, C, and K.
9-26
-------
Section 9 - Pollutant Reduction Estimates
Table 9-17
Pentachlorophenol Baseline Concentrations
Consolidated
AOX Groups3
A,B,C
D
E
F
G,I
H, J
K
Bleach Plant Effluent 0 g/L)
Minimum Level = 5.0 " g/L
Hardwood
Concentration
ND
ND
ND
2.44
2.62
ND
ND
Detects
0
0
0
1
2
0
0
Non-detects
83
46
17
12
42
44
9
Softwood
Concentration
2.26
2.74
ND
ND
10.8
ND
ND
Detects
6
1
0
0
1
0
0
Non-detects
26
39
79
28
44
103
10
Tapergrade sulfite mills fall in groups A, B, C, and K.
9-27
-------
Section 9 - Pollutant Reduction Estimates
Table 9-18
Baseline Technology Groups for Chloroform
Chloroform Group
A
B
C
D
E
F
Hypochlorite Use
Yes
Yes
No
No
No
No
Chlorine Dioxide Substitution
0%
>0%
<50%
50 to 99%
100%
TCP
Table 9-19
Chloroform Baseline Loadings
Chloroform Group3
A
B
C
D
E
F
Bleach Plant (g/kkg)
220
142
61
19
0.70
0.0
Final Effluent (g/kkg)
5.4
5.4
1.1
0.90
0.30b
0.0
Tapergrade sulfite mills fall in groups A, C, and F.
bCalculated from non-detect results using one-half the minimum level.
9-28
-------
Section 9 - Pollutant Reduction Estimates
Table 9-20
Baseline Technology Groups for COD and Color
COD/Color Group
A
B
C
D
Screen Room Status
Open
Closed
Open
Closed
Pre-Bleaching Kappa Number
' 20
' 20
<20
<20
Table 9-21
COD and Color Baseline Loadings
for Bleached Kraft and Papergrade Sulfite Operations Only
COD/Color Group3
A
B
C
D
Final Effluent COD Loading
(kg/kkg)
51
37
33
28
Final Effluent Color Loading
(kg/kkg)
89
85
60
43
Tapergrade sulfite mills fall in groups A, C, and D.
9-29
-------
Section 9 - Pollutant Reduction Estimates
Table 9-22
Final Effluent Long-Term Average Loadings
After Implementation of the BPK and PS Options
(kg/kkg)
Pollutant
AOX
Chloroform
TCDDC
TCDF
Trichlorosyringol0
2,4,5-TrichlorophenoP
2,4,6-TrichlorophenoP
3,4,5-Trichlorocatechol0
3,4,5-TrichloroguaiacoP
3 ,4,6-Trichlorocatechol°
3 ,4,6-TrichloroguaiacoP
4,5,6-TrichloroguaiacoP
Tetrachlorocatechol0
Tetrachloroguaiacol0
2,3 ,4,6-Tetrachlorophenolc
Pentachlorophenol0
All 12 chlorinated phenolics0
COD
Color
Units are kg/kkg where kkg is air-dry metric tons of pulp into bleaching
except for COD and color where it is air-dry metric tons of brown stock
pulp produced on site.
Kraft
Option A
0.512
0.0003
1.85 x 10'10
4.18 x 10'10
4.63 x 10'5
4.63 x 10'5
4.63 x 10'5
9.25 x 10'5
4.63 x 10'5
9.25 x 10'5
4.63 x 10'5
4.63 x 10'5
9.25 x 10'5
9.25 x 10'5
4.63 x 10'5
9.25 x 10'5
8.33 x 10'4
38.2
84.5
Kraft
Option B
0.208
0.0003
1.25 x 10'10
2.83 x 10'10
3.13 x 10'5
3.13 x 10'5
3.13 x 10'5
6.25 x 10'5
3.13 x 10'5
6.25 x 10'5
3.13 x 10'5
3.13 x 10'5
6.25 x 10'5
6.25 x 10'5
3.13 x 10'5
6.25 x 10'5
5.63 x 10'4
25.5
53.4
Sulfite
TCFa
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
0.0
25.5
53.4
Sulfite
ECF"
0.512
0.0003
1.85 x 10'10
4.18 x 10'10
4.63 x 10'5
4.63 x 10'5
4.63 x 10'5
9.25 x 10'5
4.63 x 10'5
9.25 x 10'5
4.63 x 10'5
4.63 x 10'5
9.25 x 10'5
9.25 x 10'5
4.63 x 10'5
9.25 x 10'5
8.33 x 10'4
38.2
84.5
aTCF is the BAT option for papergrade sulfite mills using calcium-, magnesium-, or sodium-based cooking liquor.
bECF is the BAT option for papergrade sulfite mills using ammonium-based cooking liquor, or mills that make
specialty-grade sulfite products. (The only difference between the ECF options for these two segments is that the
ECF option for the ammonium-based mills includes peroxide-enhanced extraction, while the ECF option for the
specialty-grade sulfite mills includes peroxide- and oxygen-enhanced extraction.)
The LTAs for these pollutants for Kraft Options A and B and the Sulfite ECF Option are based on one-half the
minimum level for each pollutant multiplied by a group-average production-normalized bleach plant effluent flow
rate.
9-30
-------
Section 9 - Pollutant Reduction Estimates
Table 9-23
Sources of Estimated Long-Term Average Pollutant
Loadings For Papergrade Sulfite BAT Options
Pollutant
AOX
Chloroform
TCDD & TCDF
Chlorinated phenolic compounds
COD
Color
Sulfite TCFa
Zero
Zero
Zero
Zero
Kraft Option B
Kraft Option B
Sulfite ECF"
Kraft Option A
Kraft Option A
Kraft Option A
Kraft Option A
Kraft Option A
Kraft Option A
TCP is the BAT option for papergrade sulfite mills using calcium-, magnesium-, or sodium-based cooking liquor.
bECF is the BAT option for papergrade sulfite mills using ammonium-based cooking liquor or those making
specialty grade sulfite pulps.
9-31
-------
Table 9-24
Section 9 - Title
Summary of Subcategory Loads and Reductions
Bleach Plant
Baseline Load
Final Effluent
Baseline Load
Kraft Option A
and Sulfite
Discharge Load3
Kraft Option B
and Sulfite
Discharge Load3
Kraft Option A
and Sulfite
Reduction"
Kraft Option B
and Sulfite
Reduction"
AOX (kkg/yr)
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
41,000
36,000
33,000
3,000
4,400
4,000
380
37,000
3,400
12,000
12,000
11,000
910
370
370
0
12,000
910
6,200
5,800
5,400
420
370
370
0
5,800
420
28,000
24,000
22,000
2,100
4,000
3,600
380
26,000
2,500
34,000
30,000
28,000
2,600
4,000
3,600
380
32,000
3,000
COD (kkg/yr)
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
1,200,000
1,200,000
1,100,000
97,000
70,000
59,000
3,300
1,100,000
100,000
1,100,000
1,000,000
960,000
88,000
51,000
42,000
2,500
1,000,000
90,000
830,000
800,000
740,000
65,000
51,000
42,000
2,500
780,000
67,000
130,000
110,000
100,000
8,900
20,000
17,000
830
120,000
9,700
370,000
350,000
320,000
32,000
20,000
17,000
830
330,000
33,000
VO
OJ
to
-------
Table 9-24 (Continued)
Section 9 - Title
Bleach Plant
Baseline Load
Final Effluent
Baseline Load
Kraft Option A
and Sulfite
Discharge Load3
Kraft Option B
and Sulfite
Discharge Load"
Kraft Option A
and Sulfite
Reduction"
Kraft Option B
and Sulfite
Reduction"
Color (kkg/yr)
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
2,200,000
2,100,000
1,900,000
180,000
120,000
120,000
5,800
2,000,000
190,000
2,100,000
2,000,000
1,800,000
180,000
94,000
100,000
5,500
1,900,000
180,000
1,700,000
1,600,000
1,400,000
130,000
94,000
100,000
5,500
1,500,000
140,000
55,000
41,000
38,000
3,400
13,000
14,000
290
52,000
3,600
510,000
490,000
440,000
47,000
13,000
14,000
290
460,000
47,000
Chloroform (kg/yr)
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
1,400,000
1,300,000
1,200,000
140,000
140,000
120,000
14,000
1,300,000
160,000
54,000
48,000
44,000
4,900
5,400
5,000
350
49,000
5,200
9,000
8,800
8,100
610
210
210
0
8,400
610
9,000
8,800
8,100
610
210
210
0
8,400
610
45,000
40,000
35,000
4,300
5,200
4,800
350
40,000
4,600
45,000
40,000
35,000
4,300
5,200
4,800
350
40,000
4,600
-------
Table 9-24 (Continued)
Section 9 - Title
Bleach Plant
Baseline Load
Final Effluent
Baseline Load
Kraft Option A
and Sulfite
Discharge Load3
Kraft Option B
and Sulfite
Discharge Load"
Kraft Option A
and Sulfite
Reduction"
Kraft Option B
and Sulfite
Reduction"
2,3,7,8-TCDD (g/yr)c
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
16
15
14
1.3
0.78
0.78
0.0092
15
1.3
16
15
14
1.3
0.78
0.78
0.0092
15
1.3
4.5
4.4
4.1
0.33
0.13
0.13
0.0
4.2
0.33
3.6
3.4
3.2
0.25
0.13
0.13
0.0
3.3
0.25
12
11
9.9
0.92
0.65
0.64
0.0092
11
0.93
12
12
11
1.0
0.65
0.64
0.0092
11
1.0
2,3,7,8-TCDF (g/yr)c
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
120
120
110
9.5
6.7
6.7
0.011
110
9.5
120
120
110
9.5
6.7
6.7
0.011
110
9.5
7.9
7.6
7.1
0.54
0.30
0.30
0.0
7.4
0.54
6.3
6.0
5.6
0.43
0.30
0.30
0.0
5.9
0.43
110
110
98
8.9
6.4
6.4
0.011
100
9.0
120
110
100
9.0
6.4
6.4
0.011
110
9.1
VO
oo
-------
Table 9-24 (Continued)
Section 9 - Title
Bleach Plant
Baseline Load
Final Effluent
Baseline Load
Kraft Option A
and Sulfite
Discharge Load3
Kraft Option B
and Sulfite
Discharge Load"
Kraft Option A
and Sulfite
Reduction"
Kraft Option B
and Sulfite
Reduction"
Trichlorosyringol (kg/yr)c
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
6,800
6,600
6,000
520
270
250
20
6,300
540
3,800
3,600
3,300
290
150
140
11
3,500
300
680
660
610
49
18
18
0.0
630
49
490
480
440
35
18
18
0.0
460
35
3,100
3,000
2,700
240
130
120
11
2,800
250
3,300
3,100
2,900
250
130
120
11
3,000
260
2,4,5-Trichlorophenol (kg/yr)c
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
1,100
1,100
980
86
33
31
2.0
1,000
88
600
580
540
47
18
17
1.1
550
48
550
540
500
43
9.6
9.6
0.0
510
43
450
440
410
35
9.6
9.6
0.0
420
35
52
43
39
4.6
8.8
7.7
1.1
46
5.7
150
140
130
13
8.8
7.7
1.1
140
14
VO
oo
-------
Table 9-24 (Continued)
Section 9 - Title
Bleach Plant
Baseline Load
Final Effluent
Baseline Load
Kraft Option A
and Sulfite
Discharge Load3
Kraft Option B
and Sulfite
Discharge Load"
Kraft Option A
and Sulfite
Reduction"
Kraft Option B
and Sulfite
Reduction"
2,4,6-Trichlorophenol (kg/yr)c
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
16,000
15,000
14,000
1,200
440
420
30
15,000
1,200
8,700
8,500
7,800
650
240
230
16
8,100
670
680
660
610
49
18
18
0.0
630
49
500
480
440
35
18
18
0.0
460
35
8,100
7,800
7,200
600
230
210
16
7,400
620
8,200
8,000
7,400
620
230
210
16
7,600
630
3,4,5-Trichlorocatechol (kg/yr)c
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
39,000
38,000
36,000
2,000
1,300
1,300
14
37,000
2,000
22,000
21,000
20,000
1,100
740
730
7.5
20,000
1,100
1,400
1,300
1,200
97
36
36
0.0
1,300
97
990
950
880
69
36
36
0.0
920
69
20,000
20,000
19,000
1,000
700
700
7.5
19,000
1,000
20,600
20,000
19,000
1,000
700
700
7.5
20,000
1,000
VO
00
-------
Table 9-24 (Continued)
Section 9 - Title
Bleach Plant
Baseline Load
Final Effluent
Baseline Load
Kraft Option A
and Sulfite
Discharge Load3
Kraft Option B
and Sulfite
Discharge Load"
Kraft Option A
and Sulfite
Reduction"
Kraft Option B
and Sulfite
Reduction"
3,4,5-Trichloroguaiacol (kg/yr)c
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
13,000
13,000
12,000
730
470
460
9.1
12,000
740
7,300
7,000
6,600
400
260
250
5.0
6,900
410
670
660
610
49
18
18
0.0
630
49
490
480
440
35
18
18
0.0
460
35
6,600
6,400
6,000
350
240
240
5.0
6,200
360
6,800
6,500
6,200
370
240
240
5.0
6,400
370
3,4,6-Trichlorocatechol (kg/yr)c
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
2,000
2,000
1,800
160
54
50
4.1
1,900
160
1,100
1,100
1,000
85
30
28
2.2
1,000
88
1,000
1,000
950
81
17
17
0.0
960
81
910
890
830
69
17
17
0.0
840
69
72
59
55
3.9
13
11
2.2
66
6.1
200
190
180
16
13
11
2.2
190
18
VO
-------
Table 9-24 (Continued)
Section 9 - Title
Bleach Plant
Baseline Load
Final Effluent
Baseline Load
Kraft Option A
and Sulfite
Discharge Load3
Kraft Option B
and Sulfite
Discharge Load"
Kraft Option A
and Sulfite
Reduction"
Kraft Option B
and Sulfite
Reduction"
3,4,6-Trichloroguaiacol (kg/yr)c
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
2,100
2,000
1,800
160
87
84
2.4
1,900
160
1,100
1,100
1,000
89
48
46
1.3
1,100
91
590
580
530
46
14
14
0.0
550
46
470
460
430
35
14
14
0.0
440
35
550
510
470
44
33
32
1.3
500
45
670
630
580
54
o o
JJ
32
1.3
610
56
4,5,6-Trichloroguaiacol (kg/yr)c
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
8,400
8,100
7,500
530
330
330
6.6
7,900
530
4,600
4,400
4,200
290
180
180
3.6
4,300
290
680
660
610
49
18
18
0.0
630
49
490
480
440
35
18
18
0.0
460
35
3,900
3,800
3,500
240
170
160
3.6
3,700
240
4,100
4,000
3,700
250
170
160
3.6
3,900
260
VO
oo
oo
-------
Table 9-24 (Continued)
Section 9 - Title
Bleach Plant
Baseline Load
Final Effluent
Baseline Load
Kraft Option A
and Sulfite
Discharge Load3
Kraft Option B
and Sulfite
Discharge Load"
Kraft Option A
and Sulfite
Reduction"
Kraft Option B
and Sulfite
Reduction"
Tetrachlorocatechol (kg/yr)c
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
7,800
7,400
6,900
570
350
340
5.2
7,200
570
4,300
4,100
3,800
310
190
190
2.9
4,000
320
1,300
1,300
1,200
93
36
36
0.0
1,200
93
980
950
880
69
36
36
0.0
910
69
3,000
2,800
2,600
220
160
150
2.9
2,700
220
3,300
3,100
2,900
240
160
150
2.9
3,000
250
Tetrachloroguaiacol (kg/yr)c
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
4,300
4,100
3,700
340
190
180
4.6
3,900
340
2,300
2,200
2,100
190
100
100
2.5
2,200
190
1,200
1,100
1,000
90
30
30
0.0
1,100
90
940
910
840
69
30
30
0.0
870
69
1,200
1,100
1,000
97
72
70
2.5
1,100
100
1,400
1,300
1,200
120
72
70
2.5
1,300
120
VO
oo
VO
-------
Table 9-24 (Continued)
Section 9 - Title
Bleach Plant
Baseline Load
Final Effluent
Baseline Load
Kraft Option A
and Sulfite
Discharge Load3
Kraft Option B
and Sulfite
Discharge Load"
Kraft Option A
and Sulfite
Reduction"
Kraft Option B
and Sulfite
Reduction"
2,3,4,6-Tetrachlorophenol(kg/yr)c
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
1,100
1,100
980
87
31
29
2.7
1,000
89
600
590
540
48
17
16
1.5
550
49
590
580
530
46
8.8
8.8
0.0
540
46
470
460
430
35
8.8
8.8
0.0
430
35
16
7.5
5.5
1.9
8.3
6.9
1.5
12
3.4
130
120
110
13
8.3
6.9
1.5
120
14
Pentachlorophenol (kg/yr)c
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
2,300
2,200
2,100
160
54
50
4.1
2,100
160
1,200
1,200
1,100
85
30
28
2.2
1,200
88
1,100
1,100
1,000
81
17
17
0.0
1,000
81
930
910
840
69
17
17
0.0
860
69
130
120
110
3.9
13
11
2.2
120
6.1
320
310
290
16
13
11
2.2
300
18
VO
-k
o
-------
Table 9-24 (Continued)
Section 9 - Title
Bleach Plant
Baseline Load
Final Effluent
Baseline Load
Kraft Option A
and Sulfite
Discharge Load3
Kraft Option B
and Sulfite
Discharge Load"
Kraft Option A
and Sulfite
Reduction"
Kraft Option B
and Sulfite
Reduction"
All 12 Chlorophenolics (kg/yr)c
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
100,000
100,000
94,000
6,500
3,700
3,600
100
98,000
6,600
57,000
55,000
52,000
3,600
2,000
2,000
57
54,000
3,700
10,000
10,000
9,400
770
240
240
0.0
9,700
770
8,100
7,900
7,300
590
240
240
0.0
7,500
590
47,000
45,000
42,000
2,800
1,800
1,700
57
44,000
2,900
49,000
47,000
44,000
3,000
1,800
1,700
57
46,000
3,000
NCASI 2,3,7,8-TCDD (g/yr)
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
14
13
12
0.88
0.59
0.59
0.0025
13
0.89
4.5
4.3
4.0
0.33
0.13
0.13
0.0
4.2
0.33
3.3
3.2
3.0
0.23
0.13
0.13
0.0
3.1
0.23
9.3
8.8
8.2
0.56
0.46
0.46
0.0025
8.7
0.56
10
10
9.3
0.66
0.46
0.46
0.0025
8.7
0.56
-------
Table 9-24 (Continued)
Section 9 - Title
Bleach Plant
Baseline Load
Final Effluent
Baseline Load
Kraft Option A
and Sulfite
Discharge Load3
Kraft Option B
and Sulfite
Discharge Load"
Kraft Option A
and Sulfite
Reduction"
Kraft Option B
and Sulfite
Reduction"
NCASI 2,3,7,8-TCDF (g/yr)
All Mills
All Kraft
BAT Kraft
PSES Kraft
All Sulfite
BAT Sulfite
PSES Sulfite
All Direct
All Indirect
45
42
41
1.1
3.4
3.4
0.013
44
1.1
8.6
8.4
7.8
0.60
0.28
0.28
0.0
8.0
0.60
6.7
6.5
6.0
0.44
0.28
0.28
0.0
6.3
0.44
37
33
33
0.44
3.2
3.1
0.013
36
0.46
38
35
35
0.61
3.2
3.1
0.013
38
0.62
VO
-k
to
Note: Columns may not add, due to rounding.
Tor papergrade sulfite mills, results from analysis of only one option is presented for each segment: TCP for the calcium-, magnesium-, or sodium-based segment
and ECF for the ammonium-based and specialty-grade segments. The same results are shown in the columns labeled for kraft Options A and B.
Deduction = Baseline Load minus Option Discharge Load (if Option Discharge Load is less than the Baseline Load).
°EPA found no measurable differences between Kraft Options A and B in the concentrations of these pollutants in bleach plant effluent. The slightly greater
removals of the bleach plant pollutants by Option B mills are a result of the reduced bleach plant flow rates found at mills employing Option B technology.
-------
Section 10 - Final Compliance Costs
SECTION 10
BAT, PSES, NSPS, and BMP FINAL COMPLIANCE COSTS
10.1 Introduction
This section describes EPA's final compliance cost estimates for BAT, PSES,
NSPS, and BMPs for mills in the Bleached Papergrade Kraft and Soda subcategory and mills in
the Papergrade Sulfite subcategory. EPA cost estimates for the proposed BAT, PSES, and BMPs
were presented in the Proposal TDD (1). Following the proposal, EPA collected and analyzed
additional data and issued a Notice of Data Availability (Notice) on July 15, 1996 (61 FR 36835).
In this notice, EPA presented revised cost estimates for two of the options evaluated at proposal.
Thus, the mid-1995 status of each mill served as the "baseline" to estimate the compliance cost to
implement BAT, PSES, and BMPs option technologies presented in the July notice. The costs
presented in the Notice were based on the estimates provided in the June 18, 1996 version of the
BAT and BMP Compliance Cost Estimates Report (2).
EPA estimated final costs of BAT, PSES, and BMPs taking into account
comments received on the costs presented in the Notice. In this section, EPA presents the final
compliance costs for all mills in each subcategory (direct and indirect dischargers) to complete the
process changes that comprise the BAT and PSES model technologies and to implement the
BMPs. In addition, several additional analyses, including costs reflecting corporate commitments
(Section 10.2.5), costs for two TCF option costs (Section 10.2.6), costs for EPA's Voluntary
Advanced Technology Incentives Program (Incentives Program) (Section 10.2.7), and costs for a
typical mill to install the technology that forms the basis of NSPS (Section 10.4) are included in
this section.
10.1.1 BAT and PSES Cost Estimation
EPA developed the BAT cost model to estimate costs for each mill in the BPK and
PS subcategories. For PSES, EPA evaluated the same process change technology options as it
evaluated for BAT. EPA determined the same in-plant BAT process changes are appropriate to
achieve effective pollutant reductions for preventing "pass through" at a POTW. Therefore, EPA
used the same cost model to estimate the costs of PSES and BAT. EPA's estimates include the
costs for new, upgraded, or modified process units (i.e., evaporators, recovery boiler, and
recausticizing) that are incidentally affected by the implementation of BAT and PSES options.
Using the BAT cost model, EPA estimated costs in two ways: (1) an
extrapolation of cost model results for ten mills that represent the 86 bleached papergrade kraft
and soda mills covered by the rule and (2) a mill-by mill cost estimate for the 86 mills covered by
the rule. EPA estimated costs for BAT/PSES Options A and B for the bleach papergrade kraft
and soda subcategory based on the mill-by-mill method, and estimated TCF costs based on the
extrapolation often mills to represent the 86 mills covered by the rule. For papergrade sulfite
mills, EPA used the mill-by-mill method.
10-1
-------
Section 10 - Final Compliance Costs
10.1.1.1 Preliminary Evaluation of Cost Model
In response to comments on the BAT, PSES, and BMPs regulations proposed in
1993, the cost model used to estimate compliance costs was revised significantly. To ensure the
proper function of the revised cost model, it was used to obtain a preliminary, order-of-magnitude
estimate of the costs of the revised options. In order to perform this preliminary evaluation, EPA
developed 10 technology groups, or "cost groups," to represent the range of baseline pulping and
bleaching operations used by BPK mills. The 10 groups, labeled A through K are depicted in
Table 10-1. In general, mills in lower groups (i.e., A through D) require the most extensive
process technology changes and thus are projected to incur the highest costs to comply with the
proposed rule. Mills in the higher groups (i.e., E through K) generally have process technologies
already in place equivalent to or better than the BAT options.
EPA classified each BPK mill into a costing group. First, each pulping and
bleaching line at every mill was classified as Group A through Group K based on the technologies
already in place. If a mill had only one bleach line, the group classification applied to the mill as
well as the bleach line. Because a mill may have more than one type of bleach line, assigning a
complete mill to a group is more subjective than assigning a single line. For such mills,
engineering judgment was used to assign the mill to a cost group that would provide the most
reasonable estimate of the cost that the mill would incur to implement each option.
EPA chose 10 model mills, one from each group, to estimate the average cost of
compliance for all the mills in a group. The estimated compliance costs for each model mill were
extrapolated for the entire technology group by multiplying the model mill's costs by the total
annual brown stock production for all mills in that group. After summing the results for all 10
groups, EPA obtained total estimated compliance costs. Results of this preliminary cost model
evaluation that were compared to the subsequent mill-by-mill (mid-1995) costs discussed below
demonstrated results that were within 30 percent accuracy for capital costs. Therefore, as
discussed in Section 10.2.6, model-mill costing was used for estimating compliance costs for the
two TCF options EPA considered.
Model mill costing was not performed for PS mills. Because this subcategory
consists of only 11 mills, compliance costs for this subcategory were estimated using mill-by-mill
costing, as described below (Section 10.1.1.2).
10-2
-------
Section 10 - Final Compliance Costs
10.1.1.2 Mill-by-Mill Costing
Four information sources were used to determine each mill's status from which
mill-by-mill costs were estimated:
BAT Baseline Database;
Mill cost estimates from proposal;
BMPs status file; and
Recovery boiler questionnaire.
The BAT Baseline Database (3) was the main source of updated information for
direct- and indirect-discharging mills for both subcategories. Inputs to the BAT Baseline
Database include the 1990 census questionnaire information (and subsequent 1991 and 1992 letter
updates); industry comments on the proposed rule; industry-supplied data; site visit reports;
sampling episodes; meetings and telephone conversations EPA had with industry, environmental
groups, and public; and technical articles and conference proceedings.
Mill-by-mill costing proceeded in Quattro™Pro for Windows. One electronic file,
or "millsheet," was created for each mill. Once mill-by-mill costing was completed, the results for
BPK and PS mills were compiled using macros (a series of command statements that perform
compilation tasks) created in Quattro™Pro for Windows. The macros were used to tally the
baseline technology in place at the mills, the estimated compliance costs (capital and operating)
for each mill, and the mill technology installations or upgrades necessary to meet a particular cost
option's requirements. EPA evaluated total costs, annualized costs, and individual technology
costs for each option for each subcategory.
10.1.1.3 Final Mill-by-Mill Costing
EPA made minor revisions to the costs presented in the July 1996 Notice in
response to comments. EPA made additional changes to the cost model and corrected mill-
specific information to estimate final costs for the BPK and PS subcategories (see Section 10.2.2
and Section 10.3.2).
EPA also created modified versions of the cost model to analyze on a model-mill
basis the costs of two TCP options and EPA's Incentives Program. Sections 10.2.6 and 10.2.7
contain a discussion of these analyses, including a comparison of the final mill-by-mill cost
estimates of BAT, PSES, and BMPs for all options.
10.1.1.4 Final Compliance Cost Estimates
Since the 1993 proposed rule, AF&PA and other commenters submitted
compliance cost estimates for BMPs, effluent limitations guidelines (BPT, BCT, and BAT), and
standards (NSPS, PSES, and PSNS) that widely diverged from EPA's compliance cost estimates.
A number of factors were responsible for those differing cost estimates, but they were primarily
driven by differences in costs for BAT process technologies (e.g., brown stock washing, screen
10-3
-------
Section 10 - Final Compliance Costs
rooms, oxygen delignification, etc.), costs for upgrading chemical recovery systems (most notably
recovery boilers) to accommodate BAT process technologies and BMPs, and costs for BMPs and
wastewater treatment system upgrades.
EPA's cost bases for BAT/PSES process technologies and BMPs are described in
the BAT Cost Model Support Document (4) and accompanying memorandum, "Memorandum:
Costing Revisions Made Since the Publication of July 15, 1996 Notice of Data Availability (61 FR
3687)" (5). EPA believes that it has adequately addressed site-specific factors such as site
preparation, piping, power requirements, and overhead costs such as project management,
insurance, and taxes in estimating compliance costs for BAT technologies, BMPs, and affected
process units (i.e., recovery systems) because the costs are based on mill projects supplied by
industry that account for such factors.
EPA continues to disagree with industry claims about the extent to which recovery
systems will require upgrades as a result of the options considered for the final rule. This is
discussed more fully in the Analysis of Impacts of BAT Options on the Kraft Recovery Cycle (6),
and Comment Response Document, Volume I, "Issues Concerning EPA's Estimate of the Effect
of BAT Options on Recovery Boilers" (7).
Because EPA is not revising conventional pollutant limitations in the final rule and
existing treatment systems are generally believed to be adequately designed to comply with the
toxic and nonconventional pollutant limitations promulgated by this rule (assuming BAT/PSES
process changes are implemented), mills are no longer expected to incur significant treatment
system upgrade costs attributable to this rule. As a result, EPA refined its baseline analysis of the
existing spill controls already in place at mills and revised its BMP compliance cost estimates
based on cost data provided by industry.
10.2 Compliance Cost Estimates for the Bleached Papergrade Kraft and Soda
Subcategory
The derivation of compliance cost estimates for the BPK Subcategory is described
in the following sections.
10.2.1 Technology Options
EPA's final analysis of BAT and PSES for the BPK Subcategory focused on two
ECF technology options identified as Option A and Option B. These two options have nine
common elements. Sections 8.2.1.1 and 8.2.1.2 detail the technology elements associated with
Options A and B, respectively.
10.2.1.1 Process Technologies Costed
Table 10-2 lists the BAT technologies for which EPA estimated costs for each
option. EPA also evaluated the impacts to each incidentally affected process unit (i.e.,
10-4
-------
Section 10 - Final Compliance Costs
evaporators, recovery boilers, and recausticizing systems) to determine the total costs a mill
would incur to implement each option. EPA evaluated process technologies in place as of mid-
1995 on a mill-specific basis to determine which mills require costs for new or upgraded process
unit(s) as a result of implementing BAT technologies. Because EPA regulations require bleach
plant and final effluent monitoring, the costs for monitoring equipment are also included in the
estimates.
Kappa number targets used for estimating costs were lower than the kappa
numbers used to define Option B (see Section 8.2.1.2). The costing targets reflect the fact that
current pulping practice produces softwood at kappa number 30 and hardwood at kappa number
20-25, and that modern OD systems achieve approximately 50 percent delignification. Therefore,
for costing, the targets were 15 for softwood and 10 for hardwood because a mill intending to
achieve limitations based on OD would likely purchase a modern OD system, which is capable of
achieving those targets (even though its performance would exceed the delignification achieved by
the OD systems used at mills supplying the data EPA used to represent the performance of Option
B).
10.2.1.2 BAT Technology Processes Not Costed
EPA considers the use of precursor-free defoamers as part of current baseline
technology used today at BPK mills. Defoamers are mineral oil- or water-based products used to
break and inhibit the formation of black liquor surface foam. Unchlorinated dioxin precursors are
particularly prevalent in certain mineral oils used in these defoamers, but the precursors can be
removed by a process called hydrorefming. As evidenced by the drop in measured TCDD and
TCDF discharges from bleaching pulp mills, it became known that the use of either water-based
defoamers or defoamers made with precursor-free mineral oils in the brown stock or bleach plant
areas substantially reduces the dioxin formed in bleaching. Accordingly, most mills began to
employ these types of defoamers (8). Consequently, EPA assumes use of precursor-free
defoamers to be part of industry's process baseline. Further, any mill not currently using such
defoamers can use them without incurring significant cost.
EPA considers chip quality control part of the BAT technology basis because it is
an important component of improving yield, reducing bleaching chemical requirements, and
optimizing pulp quality. EPA has in fact found that mills can attain adequate control of chip
thickness by either the use of thickness screens or by "upstream" controls such as improving the
mill's on-site chip-making processes (i.e., maintaining better tolerances on equipment) and
imposing tighter quality control standards on chips delivered by off-site sources. Thus, it is
possible for mills to achieve adequate chip size control through low or negligible cost beyond
current practices to improve quality and uniformity without the need to purchase chip thickness
screens (9). Alternatively, mills with poor chip thickness control may choose to install thickness
screens, which will pay for themselves by improving yield and reducing bleaching chemical
requirements.
10-5
-------
Section 10 - Final Compliance Costs
Similarly, EPA considers the use of efficient biological wastewater treatment to be
part of baseline technology used today at BPK mills. With the exception of one mill that
discharges to territorial waters of the United States, wastewater from all mills in the subcategory
is treated by biological wastewater treatment. These treatment systems are typically operated to
remove in excess of 90 percent of the influent BOD5 load (1), and are also capable of substantial
reductions in COD and AOX. Therefore, EPA did not estimate costs for improving wastewater
treatment systems to this level of efficiency.
In addition, implementation of BAT options typically reduce wastewater treatment
load (see Section 11.0). EPA, however, did not calculate the savings in operating costs
attributable to reduced wastewater treatment system influent BOD5 load, such as aerator
horsepower and electrical use, in estimating the costs of BAT. EPA assumed that operating cost
savings at least offset any capital investments in the treatment system required to achieve the
reduction in energy consumption. Where a substantial reduction in effluent flow is realized by the
BAT technologies in the mill, minor modifications to the effluent treatment systems may be
required so that the mill could take advantage of the energy savings mentioned above. These
changes might involve installing partitions or baffles to direct flow of effluent in an aerated
stabilization basin, or bypassing part of parallel sets of equipment.
10.2.1.3 Basis for Costing Technology Options that also Incidentally Achieve COD
Control
EPA's cost estimates include technologies that capture spent pulping liquors and
return them to the recovery process. Recovery of spent pulping liquors also incidentally achieve
effective COD control. These technologies include:
Spent pulping liquor spill control (i.e., BMPs, discussed in 2.4.4);
Effective brown stock washing; and
Closed brown stock screening.
EPA estimated the cost impact of returning captured spill material to the recovery
process under the promulgated BMP program (see Section 10.2.4.4 for details). While the BMP
requirements do not mandate the degree to which captured spill material is recovered, EPA has
determined that good engineering practice is to recover the material to the maximum degree
possible. Such an approach maximizes the value of recovered chemicals and energy (steam), and
minimizes the discharge load to the wastewater treatment system, including the COD load. EPA
will develop and promulgate COD limitations and standards in a future rulemaking. These
limitations and standards will have the effect of reflecting substantial reductions in COD
discharges achieved by BMPs for pulping liquor spills and leaks.
Costs for effective brown stock washing are included in the BAT and PSES cost
estimate. This technology is used to produce pulp with little black liquor carry-over and is part of
the strategy for reducing AOX and dioxin generation. EPA notes that wastewater from upgraded
or new washing systems installed to comply with BAT and PSES limitations may be recycled to
10-6
-------
Section 10 - Final Compliance Costs
the recovery system. While recovery of the washwater does not enhance a mill's ability to comply
with the limitations and standards being promulgated at this time, doing so is considered good,
and a common pollution prevention practice. As a result, EPA accounted for any increase in the
recovery of black liquor from improved washing in the BAT cost model.
Closed screening was also included in EPA's cost estimate for the BAT and PSES
options. EPA determined that closed screening, which may be designed to operate as a washing
stage, could be implemented under either Option A or Option B at the same cost at mills that
require improved washing instead of additional brown stock washers (the supporting information
on cost comparison, although protected as CBI, can be found in Table 1 in DCN 145008 (10)).
In addition, closed screening is critical for the efficient operation of oxygen delignification, an
Option B BAT element. Use of closed screening lowers the overall waste load on the mill
wastewater treatment system, including COD load, and is becoming common industry practice.
In order to determine the appropriate costs for Option A, EPA compared the cost
of closed screening plus any additional brown stock washing to the cost of improved brown stock
washing only. The results were not clear cut. Although the total capital costs for closed
screening were less, the annual operating costs for closed screening were somewhat higher.
Implementing closed screening allows the decker to be used as an extra, "free" stage of washing,
resulting in lower capital costs at mills that require a washing upgrade. To determine the least
cost alternative, EPA compared the pre-tax and post-tax present value of Option A with open
screening to Option A with closed screening. In several different scenarios, closed brown stock
screening was slightly (1 percent or less) less expensive than open screening. As a result, closed
screening was retained as a component of Option A.
10.2.2 Costing Revisions
The following section details changes to the cost estimation methodology made
since the proposed regulations in 1993.
10.2.2.1 Costing Revisions as a Result of Comments on the Proposal
In response to comments on the proposed rule and as a result of additional
information about the industry collected after proposal, EPA modified several assumptions used in
estimating costs. A description of the changes that affected costs are summarized below:
1) EPA incorporated the costs for BMPs and closed screening into the model.
Costs for these elements were estimated separately at proposal, using very
limited data. EPA received a substantial amount of new data voluntarily
submitted by NCASI on BMPs. EPA also collected the data necessary to
determine the capital and operating costs for the technologies and for the
combined impacts to the recovery system. As described in Section
10.2.1.3, EPA determined that closed screening, which is standard
equipment for water balance at mills operating OD, may be implemented to
10-7
-------
Section 10 - Final Compliance Costs
essentially act as an additional brown stock washer stage at an equivalent
cost at mills that do not use OD (although protected as CBI, this
information can be found on Table 1 of DCN 14508 (10)). As a result, the
total number of new or upgraded brown stock washers for which EPA
estimated final compliance costs is less than the number at proposal.
2) EPA accounted for evaporator modifications when improved brown stock
washing and BMP modifications resulted in additional process wastewater
(i.e., hydraulic load) sent to the recovery area, thus increasing the capital
and operating costs for mills that at baseline operated this affected process
unit at capacity.
3) EPA accounted for increased heat load to the recovery boiler from
additional black liquor recovered by BMPs, screen room closure, and
improved brown stock washing (heat load from OD was accounted for at
proposal). The cost model was revised to estimate the increased capital
cost for mills currently operating recovery boilers at capacity and the
operating cost savings for reduced fuel requirements as a result of
additional steam generation.
4) EPA provided an allowance to increase the capacity of the recausticizing
system which may require as much as 7 percent increase (refer to Chapter 8
of the Analysis of Impacts of BAT Options on the Kraft Recovery Cycle
(6)) due to implementation of OD, if the mill had reported to EPA that they
operated their recausticizing system at capacity. EPA determined from the
NCASI Recovery Furnace Survey that 20 percent of BPK mills do not
have spare capacity. The cost model was revised to estimate additional
capital and operating costs associated with a recausticizing system upgrade
at these mills.
5) EPA revised the assumption that certain continuous cooking digesters
could be modified to extended cooking and achieve the costing target
kappa into bleaching. EPA notes that some mills use certain continuous
cooking digesters for purposes other than kappa reduction (e.g., some mills
use partial extended cooking to affect pulp quality characteristics). These
mills cannot retrofit the digesters to achieve the target kappa number for
the purpose of complying with BAT and PSES limitations without loss of
yield. As a result, EPA estimated the costs for new OD systems that
achieve the costing target kappa number at the affected mills (this revised
assumption resulted in the apparent decrease in the baseline number of mills
and lines using extended cooking presented in Table 10-7).
10-8
-------
Section 10 - Final Compliance Costs
6) EPA adjusted equations used to predict the caustic application rate for
modified bleaching using data provided by one commenter. This revision
increased caustic demand, which, in turn, increased operating costs.
7) EPA revised cost curves for most equipment items in accordance with the
data collected by voluntarily submitted surveys and comments from
industry (see BAT Cost Model Support Document (4)). All cost curves
were adjusted for inflation from a 1993 to a 1995 basis, increasing capital
costs slightly.
8) EPA revised estimates of costs for compliance monitoring, energy, wood,
and chemicals.
9) EPA estimated costs a mill would incur to accommodate any increased
thermal load to individual recovery boilers by estimating costs for addition
of anthraquinone to the pulping digester or oxygen-based black liquor
oxidation. EPA assumed that mills would most likely reduce the load on
the recovery boiler rather than to increase the capacity of an existing boiler
for the incremental increase associated with BAT and BMPs (see Section 6
of the Analysis of Impacts of BAT Options on the Kraft Recovery Cycle
(6)). To obtain realistic costs, EPA assumed mills would use the most
economically feasible upgrade for reducing thermal load.
10.2.2.2 Costing Revisions as a Result of Comments on the Notice
In response to comments on the July 15, 1996 Notice, EPA corrected mill-specific
information and made additional changes to assumptions to estimate final costs. These changes
were not major revisions to the model. A description of the changes and the effect on costs are
summarized below. For a complete description of the following changes, refer to "Memorandum:
Costing Revisions Made Since Publication of July 15,1996 Notice of Data Availability (61 FR
36837)" (5).
1) EPA corrected errors in the application of climate factor. If mills require
installation of peroxide storage facilities, chlorine dioxide storage facilities,
or new or greenfield chlorine dioxide generators, the capital costs were
revised to include the climate factor (i.e., an allowance for enclosing these
facilities in a building in cold climates). The changes to capital costs were
minor and varied depending on mill location.
2) EPA revised the assumption that an increase in chlorine dioxide use of one
kkg per day would require cost for a chlorine dioxide generator conversion
or installation. Instead, EPA assumed that an increase of one kkg per day
would trigger an upgrade to the existing generator (i.e., an allowance for
minor generator upgrades), while an increase in chlorine dioxide use of
10-9
-------
Section 10 - Final Compliance Costs
greater than 20 percent above current capacity would trigger either a
generator conversion or installation of a new chlorine dioxide generator,
which is a more capital-intensive improvement. EPA revised the
assumption because several mills operating large generators are capable of
upgrading them to expand capacity at a lower cost than installing a new
generator to accommodate the amount of chlorine dioxide necessary for
ECF bleaching. This revision decreased the capital cost for several mills.
3) EPA corrected the errors in the unit costs of caustic and hydrogen peroxide
that resulted from a unit conversion error (this error carried through the
proposal and the notice cost estimates). The correct unit costs for caustic
and hydrogen peroxide are $0.29/kg and $0.62/kg, respectively, which are
approximately half as much as the erroneous costs of $0.54/kg and
$1.15/kg, respectively. Although the chemical costs decreased by
approximately one half, the effect on overall operation costs (and, likewise,
on individual mill operating costs) was a net increase. The reason for this
apparent discrepancy is because both ECF technology options result in less
chemical consumption overall; therefore, the Notice estimate of net savings
from reduced chemical use when using high chemical costs was
unreasonably high. Once the correct, lower chemical costs were used, the
chemical cost savings decreased, resulting in an overall slight increase in
operating costs. Therefore, compared to the costs presented in the Notice,
fewer mills now demonstrate overall operating cost savings. This error
particularly affected the Option B operating cost estimate, which still
displays significant reduction of chemical reduction compared to Option A
because of implementation of OD (e.g., operating costs were erroneously
estimated as a $7 million savings at the time of the Notice versus a $2
million cost for final estimates presented in 10.1.4).
4) EPA corrected the double counting of taxes and insurance by adjusting the
factor used to calculate the cost of maintenance and repair from 4 percent
of capital to 2 percent. Taxes and insurance are accounted for in an
additional economic analysis described in Economic Analysis for the
National Emissions Standards for Hazardous Air Pollutants for Source
Category: Pulp and Paper production; Effluent Limitations Guidelines,
Pretreatment Standards, and New Source Performance Standards: Pulp,
Paper, and Paperboard Category—Phase 1 (11). As a result, operating
cost estimates presented in this section decreased slightly compared to the
cost estimates at the time of the Notice.
5) EPA revised the assumption that all softwood lines with OD in place could
operate them to achieve a kappa number less than 20 by providing costs for
an upgraded OD system when a mill's baseline kappa number was greater
than 19. EPA revised this assumption based on mill data, demonstrating
10-10
-------
Section 10 - Final Compliance Costs
that mills not currently pulping to an unbleached kappa number of less than
20 are not capable of achieving the promulgated limitations and standards
without incurring costs. As a result, several mills received capital cost
allowances to upgrade OD systems.
6) EPA revised the assumptions that all hardwood lines that operate OD to
achieve kappa numbers of 11 or 12 would incur an operating cost savings
in complying with BAT. Previously, EPA assumed these mills could
optimize OD to achieve an unbleached kappa number of 10 without
incurring capital costs and receiving operating cost savings. Instead, EPA
is now assuming that these mills could continue to operate at their current
unbleached pulp kappa number without necessarily experiencing any
operating cost savings. EPA revised this assumption because mill data
demonstrate that these mills are capable of meeting BAT limitations at
kappa numbers of 11 or 12.
10.2.3 Baseline Status
At proposal, EPA evaluated BAT, PSES, and BMPs for 88 mills in the BPK
subcategory. Since proposal, the total number of mills in the subcategory has decreased from 88
to 86 because one mill closed and another was reclassified as unbleached kraft. Two other mills
informed EPA that they would cease bleached kraft production, rather than invest in any new
bleaching technology after promulgation of this rule, opting to produce unbleached kraft pulp. As
a result, EPA used the cost model to analyze the costs of BAT, PSES, and BMPs for 84 mills
(refer to Section 4). However, for EPA's economic analysis (11), these two mills are included in
the subcategory profile because mills that produce kraft pulps are subject to MACT I and MACT
II. In EPA's economic analysis, EPA estimated MACT I and MACT II costs for these two mills
and assumed that they would not incur BAT, PSES, and BMPs costs to determine the economic
achievability of the Cluster Rules.
The baseline status of BPK mills both at the time of proposal and in mid-1995 is
shown in Table 10-3. The table presents the number of mills and lines that use certain elements of
the BAT options at proposal and mid-1995. The table also lists the associated percentage of the
total pulp production of the mills using these technologies at proposal and mid-1995. In general,
the number of mills and lines and the production percentages have increased, indicating many mills
have incorporated elements of the proposed BAT options in this time period. An exception is the
number of mills and lines using extended cooking, which is an artifact of overcounting those lines
at proposal (see Section 10.2.2.1). Table 10-3 shows the percentage of ECF production has
increased by a factor of five while the percent of production using hypochlorite has been cut in
half.
10-11
-------
Section 10 - Final Compliance Costs
10.2.4 Compliance Cost Estimates
The following section presents the compliance cost estimates used in this
rulemaking.
10.2.4.1 Total Capital and Operating Compliance Cost Estimates
After revising the model as described in Section 10.2.2, EPA used mill-by-mill
costing to obtain final compliance cost estimates for Option A and Option B. Table 10-4
compares the final compliance costs for Option A and B to the costs of the proposed BAT/PSES
option. The cost components reported in this section are engineering estimates of the cost of
purchasing and installing equipment and the annual operating and maintenance costs associated
with that equipment. Proposal costs have been adjusted from 1993 to 1995 dollars to facilitate
comparison.
EPA also calculated the annualized costs of Options A and B to facilitate a
preliminary comparison of the options prior to performing the economic impact analysis. All cost
estimates in this section are expressed in 1995 dollars. Annualized costs, which were calculated
on a "per year" and "per ton" basis, are also shown Table 10-4. EPA estimated engineering
annualized costs based on a 13 percent nominal (9 percent real) interest rate over 15 years that
accounts for tax/depreciation shield using the following equations:
Annualized Cost/yr = I x ((C x Capital Cost) + (P x O x Operating Cost))
where:
I = Interest factor = 0.1241
C = Capital tax/shield depreciation = 0.792
P = Present value for operating and maintenance = 9.823
O = Operating and maintenance tax shield = 0.593
Capital cost = Final capital compliance cost
Operating cost = Final operating compliance cost
Annualized Cost Annualized Cost/yr
t t/yr
where:
t/yr = kkg/yr = metric tons (kkg) of unbleached pulp produced by BPK
subcategory in 1995 = 29,200,000 kkg
Table 10-5 presents the final Option A and Option B capital and operating cost
estimates broken down to distinguish between direct- and indirect-discharging mills. Nine BPK
mills discharge to a POTW (i.e., discharge indirectly) and are subject to PSES, while 75 direct-
discharging BPK mills are subject to BAT. Table 10-6 shows the capital and operating cost
10-12
-------
Section 10 - Final Compliance Costs
ranges for all 84 mills for which costs were developed and the average cost per mill by option (as
mentioned previously, two mills in the subcategory were estimated to incur no BAT, PSES, or
BMPs costs).
EPA notes that the engineering cost estimates presented in this section, specifically
the operating and maintenance costs and annualized costs, differ slightly from the estimates used
to calculate this rule's economic achievability. In EPA's economic analysis, operating and
maintenance costs include an additional four percent of capital to account for non-plant overhead
costs that are not accounted for in the engineering cost estimates. Annualized costs differ slightly
because the variables used to calculate the rules economic achievability are more detailed (11).
Capital costs in both analyses are equivalent.
10.2.4.2 Technology Components Costs
Tables 10-7 and 10-8 illustrate the cost breakdown for Options A and B, listing the
capital and operating contribution of each costed BAT element and each affected process unit to
total capital and operating costs, respectively. Table 10-7 shows the capital cost of Option B is
approximately twice that of Option A, while Table 10-8 shows the total operating cost for Option
B is a fraction of the Option A costs. The total capital costs presented in Table 10-7 do not
match the costs presented in Section 10.2.4.1 because the costs in Table 10-7 do not include
regional cost factors. Regional cost factors adjust capital costs to account for a mill's
geographical location (i.e., rural locations adjusted for lower cost of capital). These factors were
applied to the total mill cost after summing all component capital costs and had the net effect of
lowering total subcategory capital compliance costs.
The capital and operating costs for several BAT elements and affected process
units differ between Option A and B for several reasons. First, Option B includes extended
delignification to reduce kappa number prior to bleaching. This, in turn, decreases the amount of
chemical required for bleaching to achieve equivalent brightness; therefore, the capital costs for
chlorine dioxide generators and the operating costs for chemical demand are considerably lower.
The capital and operating costs for hypochlorite elimination are also lower because the decreased
chemical demand for Option B enables some mills to avoid replacing hypochlorite stages with a
new D-stage. Second, operating oxygen delignification generates more black liquor, which is sent
to the recovery system, leading to increased costs for evaporator and recovery boiler upgrades for
Option B. Increased black liquor recovery, however, also leads to additional steam generation
which reduces operating costs for some mills by decreasing fuel requirements. Lastly, oxygen
delignification also consumes more white liquor, creating increased demand on recausticizing
systems; therefore, additional recausticizing upgrade costs are associated for Option B and not
Option A.
10.2.4.3 Mill Cost Breakdown
Even mills that use Option B technologies (or better) were estimated to incur some
costs to comply with BMPs as well as bleach plant and final effluent monitoring costs (although
10-13
-------
Section 10 - Final Compliance Costs
protected as CBI, Table II itemizes capital and operating costs for each mill, cost group, and the
BPK subcategory as a whole in DCN 14508 (10)).
The estimated capital costs are much higher for some mills than others within the
same cost group. Many of these high-cost mills require installation of new chlorine dioxide
generators and, for Option B, new oxygen delignification systems, which impact capital costs
significantly. But, in almost all cases, high costs are attributable to baseline use of hypochlorite.
The installation of new D-towers for the elimination of hypochlorite results in high capital costs to
comply with either option compared to other mills in the same cost group, although it reduces
operating costs by lowering chemical costs.
For both Option A and Option B, EPA's mill-by-mill costing resulted in some mills
incurring a net savings in operating and maintenance costs compared to their current status.
Three main reasons are responsible for this net savings:
1) Bleaching chemical costs are the most influential factor determining
whether or not individual mill(s) would achieve a net savings for operating
costs. Most mills for which EPA estimated operating cost savings, for
either option, are estimated to save significantly on chemical costs
compared to their mid-1995 operations. Furthermore, as a result of pre-
bleaching lignin reduction, the bleaching chemical requirements are less for
Option B than Option A. Hence, the opportunity for savings is greater for
OptionB.
2) EPA's mill-by-mill costing resulted in a net decrease in evaporator
hydraulic load for some mills. Because improving brown stock washing
decreases the hydraulic load on evaporators and implementation of BMPs
increases the weak black liquor (includes process wastewater) hydraulic
load on evaporators, if the reduction in hydraulic load due to improved
brown stock washing is greater than the increase in hydraulic load due to
implementation of BMPs, an operating cost savings will occur at the
evaporator because less steam is required to evaporate a smaller total
volume of weak black liquor.
3) Improving brown stock washing, closing the screen room, and
implementing BMPs increases the black liquor recovered per ton of pulp.
Burning the additional recovered black liquor in the recovery boiler
generates steam, thereby decreasing the amount of steam the mill must
generate using supplemental fuel, a net savings.
10.2.4.4 Costs for BMPs
Rationale for Including BMPs - EPA included costs for BMPs as part of the
BAT costs for several reasons. First, a portion of the BAT and BMPs compliance costs are
10-14
-------
Section 10 - Final Compliance Costs
largely inseparable because improved pulping liquor controls can trigger replacement, upgrade, or
modification of process units that are also affected by the BAT limitations (e.g., depending on
choices made in implementing BAT and BMPs requirements, a mill may determine that the
requirements, either singly or together, result in a need to increase evaporator capacity).
Incorporating BMPs costs in the BAT estimate recognizes the degree to which BAT and BMPs
compliance decision-making is intertwined at mills.
BMPs Cost Estimate - The output of the cost model is an estimate of the cost of
complying with BAT or PSES while complying with BMPs. The associated costs of BMPs were
removed from the original cost estimates in order to reassess evaporator upgrade costs and
recovery boiler capacity adjustment costs, (i.e., anthraquinone pulping or black-liquor oxidation),
if such costs would be necessary to comply with BAT or PSES. The "associated costs" incurred
through implementation of BMPs include evaporator upgrade costs due to an increased amount of
wastewater to evaporate and higher recovery boiler costs due to capacity adjustments to
accommodate the incremental increases in thermal load from the recovered black liquor (see
Recovery Impacts Document). After removing the impacts of BMPs, recovery boiler costs that
may be necessary for the implementation of BAT technologies (e.g., closed screening, new or
improved brown stock washing, and extended delignification (Option B only)) were reassessed.
The resulting "BAT\PSES-only" capital and operating cost estimates were subtracted from the
original cost model output to estimate the BMPs costs for each mill (Table III lists the
BAT\PSES-only; BMPs-only; and BAT, PSES, and BMPs capital and operating cost estimates by
mill for each option in DCN 14508 (10)).
The total estimated cost of BMP implementation under Option A is slightly greater
than BMP implementation cost under Option B (see Table 10-5). This difference lies in the
number of mills that would experience the need for recovery capacity adjustment as a result of the
two BAT options. Because Option B includes extended delignification, a greater amount of black
liquor (albeit of lower heat content) is sent to recovery as compared to Option A. Of the
population of mills that currently operate near the maximum thermal capacity of their recovery
boilers, a greater number would require capacity increases under Option B than under Option A.
When the cost of BMP implementation is examined as an addition to BAT implementation, the
added recovery capacity required by BMP implementation is, in the case of a number of mills
under Option B, simply an incremental increase on a capacity adjustment that is already needed.
In those cases where BAT implementation, rather than BMP implementation, is the "trigger" for
recovery capacity adjustment, the cost burden attributable to BMPs is reduced by economy-of-
scale and initial cost considerations. In spite of the fact that equivalent thermal load increases are
assumed for BMP application to both BAT options, the result is a somewhat reduced BMP
implementation cost for Option B when total costs for the BPK subcategory are calculated. The
fact that the difference in the calculated total cost for the two options is small demonstrates that
this reduced cost of BMP implementation is confined to just a few mills in the subcategory.
10-15
-------
Section 10 - Final Compliance Costs
10.2.4.5 Total Number of New, Upgraded, or Modified Pulping and Bleaching
Technologies
Table 10-9 shows the number of mills for which EPA costed new, upgraded, or
modified BAT technologies and other process units (i.e., evaporators, recovery boilers, and
recausticizing systems) that are incidentally affected by each option on at least one fiberline. As
described in Section 10.2.3, many mills have implemented elements of the BAT options between
proposal and mid-1995. As a result, the total that EPA assumes will be necessary for several of
these elements has decreased. Several numbers changed (i.e., extended cooking and brown stock
washers) because EPA revised costing assumptions (see Section 10.2.2).
10.2.5 Corporate Commitments to Install BAT Elements
After proposal of BAT and PSES in 1993, a number of pulp mill owners and
operators announced plans to install new technologies at their facilities. EPA excluded the
incurred costs of process changes that were already implemented as of mid-1995 in the cost
estimates used to analyze the economic achievability of the rules. However, EPA included the
costs of the announced process changes not underway as of July 1, 1995 in the cost estimates
used to analyze the economic achievability of the rule. Although EPA included the costs of the
process changes announced but not yet underway as of mid-1995 in its final cost estimates, EPA
nevertheless evaluated the impact of these costs in an alternative analysis reflecting announced
corporate commitments that were not underway as of mid-1995.
Six corporations announced plans to install new technologies at their facilities after
proposal of BAT and PSES in 1993. The announced plans involved a total of 24 mills. The
process changes were implemented at 12 of these mills by mid-1995, and the costs for these
changes were excluded from EPA's analysis of the economic achievability of this rule. Process
changes at the other 12 mills were not underway as of July 1, 1995. The costs anticipated for
these 12 mills were included in EPA's economic achievability analysis and were also subject to the
alternative analysis described below. Table 10-10 lists the corporations announcing commitments
for those 12 mills, the process change planned, the number of mills affected by corporate plans,
and the reference that contains the corporation decision.
10.2.5.1 Alternative Analysis Reflecting Corporate Commitments
In its alternative analysis, EPA evaluated the impact of corporate commitments by
assuming that mills had incurred the costs of the projects that were announced. EPA, therefore,
revised a mill's "baseline" technology status to include corporate decisions to install a technology
by the end of 1995 if the commitment was confirmed by a corporate statement. EPA performed
this alternative analysis in order to determine whether the lower estimated capital costs and
operating costs would change EPA's economic impact projections for Options A and B. EPA
found that even under the alternative analysis, EPA's projected economic impacts did not change
for the two options (12).
10-16
-------
Section 10 - Final Compliance Costs
Of the 86 mills in the BPK subcategory, corporate commitments were announced
but not underway as of July 1, 1995 for process changes at 12 mills. Under this alternative
analysis, these mills were credited for operating the announced equipment to implement either
ECF bleaching or oxygen delignification followed by ECF bleaching.
The revised "baseline" status of the entire BPK subcategory reflecting this
alternative analysis is shown in Table 10-11. The baseline status during proposal and mid-1995
are included for comparison. Assuming all planned process changes are made, the percentage of
ECF production will represent approximately half of the total production in the BPK subcategory
by the end of 1995.
10.2.5.2 Option A Commitments - ECF Bleaching
Process changes to operate ECF bleaching were announced for 10 mills that
operated at less than 100 percent chlorine dioxide substitution as of mid-1995. Under the
alternative analysis, the baseline status of each of the 10 affected mills was modified to credit
these mills for using 100 percent chlorine dioxide substitution while ensuring other operating
conditions remained equivalent (i.e., equivalent chemical charge during each step of the bleach
sequence). Other elements of Option A that were not in operation at the affected mills, such as
closed screening room, improved brown stock washing, sufficient peroxide or oxygen use for
extraction, or elimination of hypochlorite, were not assumed to be instituted as a result of a
corporate-level decision to implement ECF bleaching; therefore, costs for these upgrades were
included in the alternative analysis cost estimates.
10.2.5.3 Option B Commitments - Oxygen Delignification and ECF Bleaching
Process changes to operate extended delignification followed by ECF bleaching
were announced for two mills that did not operate these technologies as of mid-1995. Under the
alternative analysis, the baseline status of the two affected mills was modified to credit these mills
for using extended delignification with 100 percent chlorine dioxide substitution while ensuring
other operating conditions remained equivalent. This alternative analysis did not assume that
other elements of Option B that were not in operation during mid-1995 would be implemented as
a result of a corporate commitment to implement oxygen delignification and ECF bleaching (i.e.,
EPA estimated costs for those other elements to calculate the total alternative analysis compliance
costs).
10.2.5.4 Compliance Cost Estimates with Corporate Commitments
Overall, under the alternative analysis, adjusting the baseline of the 12 mills
affected by corporate commitments resulted in lower estimated capital and operating compliance
costs for each option because these mills were credited for already installing the announced BAT
elements. Table 10-12 presents a comparison of the final compliance cost estimates of BAT,
PSES, and BMPs (from Section 10.2.4.1) and the alternative analysis costs (although protected as
CBI, Table IV lists the corporations announcing commitments, the process change planned, the
10-17
-------
Section 10 - Final Compliance Costs
specific mills affected by corporate plans, and the reestimated compliance cost of each mill in
DCN 14508 (10)).
EPA did not use the alternative analysis cost estimates to determine the economic
impacts presented in Economic Analysis for the National Emission Standards for Hazardous Air
Pollutants for Source Category: Pulp and Paper Production; Effluent Limitations Guidelines,
Pretreatment Standards, and New Source Performance Standards: Pulp, Paper, and Paperboard
Category—Phase 1(11) because the mill-specific baseline data used by EPA reflected
technologies in place prior to July 1, 1995. Rather, EPA used the final compliance cost estimates
presented in Section 10.2.4.1 to determine the economic impacts of this rule.
10.2.6 Estimated Costs of TCF Bleaching Options
The data available to EPA at promulgation of this rule were insufficient to confirm
that TCF processes were technically available for the full range of market products currently
served by ECF processes. EPA nevertheless evaluated the costs of retrofitting the U.S. bleached
papergrade kraft and soda mills to TCF bleaching to provide perspective on the likelihood of TCF
processes being found to be economically achievable when they are shown to be technically
available.
10.2.6.1 TCF Options
EPA investigated the costs of two TCF bleach sequences, listed in Table 10-13.
TCF bleaching, because it eliminates all use of chlorine-containing compounds, also eliminates the
possibility for the formation of TCDD, TCDF, or other chlorinated pollutants. Thus the TCF
options do not require the elements of Options A and B that minimize the likelihood of generating
TCDD/F during bleaching, i.e., use of dioxin- and furan-precursor free defoamers, and strategies
to minimize kappa factor and TCDD- and TCDF-precursors in brown stock pulp. The other
common elements of Option A and Option B (adequate chip thickness control, closed brown
stock pulp screen room operation, effective brown stock washing, elimination of hypochlorite,
oxygen and peroxide enhanced extraction, adequate mixing, and efficient biological wastewater
treatment) are necessary for successful operation of a TCF bleach sequence and/or for the control
of COD discharges. The TCF bleaching sequences also include medium consistency oxygen
delignification. The first TCF bleach sequence, identified in this document as Option C, was
based on ozone and peroxide bleaching (OZEopQPZP); the second TCF-bleaching sequence,
hereafter called Option D, was based primarily on peroxide bleaching (OQPP).
10.2.6.2 TCF Costing Methodology
Compliance costs for TCF Option C and Option D were estimated by using the
model-mill costing approach (as described in Section 10.1.1.1). EPA used a modified version of
the cost model to estimate the costs for the 10 model mills to implement the two TCF options.
The costs for each model mill were extrapolated for the entire group. The sum of all the groups
provided total compliance cost estimates.
10-18
-------
Section 10 - Final Compliance Costs
10.2.6.3 TCF Estimated Capital, Operating, and Annualized Costs
The capital and operating costs of the two TCF options differ significantly. Option
C is characterized by high capital costs and low operating costs. Conversely, Option D is less
capital cost intensive and more operating cost intensive. For Option C, an expensive capital cost
technology (ozone delignification) is offset with a decrease in peroxide consumption, thereby
decreasing operating costs. Option D incurs high operating costs resulting from increased
peroxide consumption. In general, the same mill would need at least twice as much peroxide for
Option D as for Option C.
Table 10-14 compares the costs of the TCF options to the final costs of Option A
and Option B (although protected as CBI, Table V lists the model-mill costing estimates of each
mill for Option C and D and the mill-by-mill cost estimates of each mill for Option A and B in
DCN 14508 (10).
10.2.7 Voluntary Advanced Technology Incentives Program Costing
To encourage mills to implement and develop more environmentally beneficial
technologies, EPA established the Voluntary Advanced Technology Incentives Program to afford
direct-discharging mills an opportunity to comply with more stringent regulations in return for
regulatory- and enforcement-related incentives, as well as public recognition. The technology
bases for the stringent limitations established for this multiple-tier program include elements of
BAT Option B, other advanced technologies, and not-yet-developed processes and technologies
that will challenge the industry in the future. The Technical Support Document for the Voluntary
Advanced Technology Incentives Program (13) provides a detailed description of this program,
including the methodology used to estimate costs.
10.2.7.1 Description of Tiers
EPA's program establishes three sets of Advanced Technology BAT limitations
for the voluntary program. Each successive tier is characterized by decreased pollutant discharge
and effluent flow. Like the baseline BAT regulation, mills entering this program may implement
any technology or process change to achieve the Advanced Technology BAT limitations;
however, for costing purposes, EPA has identified specific technologies for each tier that would
assist mills in achieving the limitations. The first tier, Tier I, employs the same technologies
proposed for Option B; therefore, the costs for this tier are assumed to equal the estimates for
Option B. Tiers II and III limitations may be achieved by implementing advanced technologies
and flow reduction measures using either an ECF or TCF bleaching process (see Technical
Support Document for the Voluntary Advanced Technology Incentives Program (13)).
10-19
-------
Section 10 - Final Compliance Costs
10.2.7.2 Case-Study Mill
The cost estimates for this program are based on a case-study mill. This mill
represents a typical Group C mill (see Table 10-1) that produces approximately 1,000 UBMT/d of
pulp using a CDEopD bleach sequence. The case-study mill is representative of the type of mill
that may commit to this program. Mills of this size benefit from the economy of scale associated
with implementing advanced technologies (e.g., two-stage oxygen delignification and ozone
delignification that are more costly than either Option A or Option B).
10.2.7.3 Compliance Costs for the Voluntary Advanced Technology Incentives
Program
Table 10-15 shows the approximate costs for the case-study mill to achieve each
incentive tier. These costs represent maximum estimates since many mills will likely combine
elements of the Voluntary Advanced Technology Incentives Program with other modernization
projects, thereby reducing actual costs. Since undeveloped technologies and processes may be
used in the future to achieve the Advanced Technology BAT limitations, EPA expects these
approximate cost projections will be reduced over time. Estimated Option A costs for the case-
study mill are included in the table for comparison.
10.3 Compliance Cost Estimates for the Papergrade Sulfite Subcategory
In response to comments received after Proposal, EPA divided the 11 mills in the
PS subcategory into three segments:
Segment A: Calcium-, magnesium-, or sodium-based sulfite pulping;
Segment B: Ammonium-based sulfite pulping; and
Segment C: Production of pulp and paper at specialty-grade sulfite mills.
10.3.1 Technology Options
EPA's final analysis of BAT and PSES focused on ECF and TCP technology-
based options for three different segments. EPA analyzed one BAT and PSES option for each of
the three segments. The BAT option for Segment A (calcium, magnesium, or sodium sulfite) is
based on TCP bleaching. The BAT and PSES options for Segment B (ammonium sulfite) and
Segment C are based on ECF bleaching. Section 8.2.2 details the technology elements associated
with each PS segment.
10-20
-------
Section 10 - Final Compliance Costs
10.3.1.1 Technology Processes Costed
Table 10-16 lists the process technologies for which EPA estimated costs for each
option. EPA also evaluated the costs to upgrade incidentally affected process units (i.e.,
evaporators for PS mills). For the reasons discussed in Section 10.2.1.2, EPA excluded the costs
for the use of precursor-free defoamers. Similarly, EPA considers the use of efficient biological
wastewater treatment to be part of baseline technology used today at PS mills. Wastewater from
all mills in the subcategory is treated by biological wastewater treatment. These treatment
systems are typically operated to remove most of the influent BOD5 load (see Proposal TDD (1)),
and are also capable of substantial reductions in COD and AOX. Therefore, EPA did not include
costs for improving wastewater treatment systems to this level of efficiency in its BAT cost
estimates.
10.3.2 Cost Model Revisions
EPA used a modified version of the cost model developed for the BPK
Subcategory to estimate the costs for the PS mills to implement the BAT and PSES options and
BMPs. EPA's proposed BAT and PSES for the PS Subcategory was based on TCP bleaching and
oxygen delignification. In response to comments on the proposed rule, EPA eliminated oxygen
delignification from the final options. Based on information from PS mills employing TCP
bleaching, EPA has determined that oxygen delignification is not necessary for TCP bleaching of
papergrade sulfite pulp.
In response to comments on the July 1996 Notice, EPA increased the unit cost of
caustic and peroxide (detailed in Section 10.2.2 because this change affected BPK mills as well),
leading to increased operating cost estimates for PS mills. In addition, the capital costs for the PS
subcategory were recalculated to include regional climate factors for the installation of peroxide
storage facilities, chlorine dioxide storage facilities, and new or greenfield chlorine dioxide
generators (also detailed in Section 10.2.2).
10.3.3 Compliance Cost Estimates
The following sections detail the results of EPA's compliance cost estimates.
10.3.3.1 Total Costs
EPA used the cost model to estimate BAT compliance costs for each segment with
a few exceptions. For one hardwood ammonium sulfite mill, the cost of conversion to TCP
bleaching was used to estimate the cost of compliance with the revised BAT/PSES rather than the
cost of conversion to ECF bleaching. EPA did so because this mill commented that it could
feasibly convert to TCP bleaching without altering its product line. (EPA rejected TCP as a viable
BAT option for Segments B and C because it was not shown to provide the full range or products
made by all mills in these segments including softwood.) Another hardwood ammonium sulfite
mill currently employs TCP bleaching. Mills already operating at TCP were estimated for BMPs
10-21
-------
Section 10 - Final Compliance Costs
capital costs only. Although protected as CBI, Table VI in DCN 14508 (10) lists the 11 PS mills
by segment and the type of bleaching employed at each mill (one PS mill does not employ
bleaching). Table 10-17 presents a breakdown of the number of mills costed for each BAT option
for each segment.
The cost model was not used to estimate total BAT costs for Segment C. The one
mill in this segment provided EPA with an estimate of its cost to convert to ECF bleaching (EPA
notes that at the time of promulgation, one Segment A mill prepared business plans to produce
specialty-grade pulp; however, EPA estimated its compliance costs as a Segment A mill because
that was the mill's status as of mid-1995). EPA reviewed this estimate and determined it was
reasonable. All cost information associated with the Segment C mill, including a description and a
cost breakdown, is CBI; however, this information is listed in located in Table IX in DCN 14508
(10).
Table 10-18 compares EPA's final BAT costs for all three segments to the
estimated costs of the proposed BAT option. EPA estimated that all 11 PS mills will incur costs
to comply with BAT, PSES, and BMPs. Proposal costs have been adjusted from 1993 to 1995
dollars to facilitate comparison.
EPA also calculated the annualized costs for the PS subcategory, as discussed in
Section 10.2.4.1. Annualized costs, which were calculated on a "per year" and "per ton" basis,
are also shown Table 10-18. EPA estimated annualized costs based on a 13 percent nominal (9
percent real) interest rate over 15 years that accounts for tax/depreciation shield using the same
equations used to calculate costs for the BPK subcategory (see Section 10.2.4.1) except the total
of unbleached pulp production equals 1,280,000 kkg the PS subcategory.
10.3.3.2 Technology Component Costs
EPA estimated capital and operating costs for all mills in the PS subcategory.
(Table VII presents the technology component costs and the component costs as a percentage of
total costs in DCN 14508 (10)).
10.3.3.3 Mill Cost Breakdown
The final capital and operating costs for each of the 11 to comply with BAT,
PSES, and BMPs, although protected as CBI, are listed in Table VIII of DCN 14508 (10).
10.3.3.4 Costs for BMPs
Unlike BPK mills, the BMPs cost estimates for PS mills can be easily separated
from the BAT costs. For PS mills, the BAT options had no cost impact on recovery systems.
The only estimated recovery system costs result from the impact of BMPs on evaporators. EPA
assumed implementation of BMPs does not significantly affect recovery boilers for those mills that
operate a recovery boiler. Therefore, no costs for recovery boiler capacity adjustments were
10-22
-------
Section 10 - Final Compliance Costs
estimated. As a result, estimated cost of implementing BMPs include only the capital and
operating costs of BMPs and evaporator upgrades. (The BMPs cost estimates for Segment A and
Segment B (without regional factors) and the BMPs cost estimate for Segment C, although
protected as CBI, are shown in Tables VIII and IX, respectively, in DCN 14508 (10)).
10.3.3.5 Total Number of New or Upgraded Pulping and Bleaching Technologies
EPA estimated the costs of installing technology options. Table 10-19 shows the
number of installations of new or upgraded BAT technologies and affected process units
necessary for each PS mills to comply with BAT, PSES, and BMPs for Segment A and
Segment B. Segment C is excluded to protect CBI; however, a cost breakdown is located in
Table IX in DCN 14508 (10).
10.4 NSPS Compliance Costs
EPA evaluated compliance costs for new sources covered under NSPS limitations.
In 40 CFR Part 430.01Q), EPA has set forth a definition of "new source" for the BPK and PS and
subcategories. In tailoring a "new source" definition specifically for these subcategories, EPA
considered what type of fiber line modifications should be subject to NSPS. A fiber line is a series
of operations employed to convert wood or other fibrous raw material into pulp. For the BPK
and PS subcategories, the fiber line encompasses pulping, deknotting, brown stock washing, pulp
screening, centrifugal cleaning, and multiple bleaching and washing stages. EPA has defined a
BPK or PS source as a new source if:
1) It is constructed at a site at which no other source is located (i.e., a
greenfield mill).
2) It completely replaces an existing source. For example, if a fiber line
completely replaces an existing fiber line. This definition does not include
fiber lines enrolled in the Voluntary Advanced Technology Incentives
Program (13) or fiber lines modified to comply with baseline BAT.
EPA notes that the following changes do not cause an existing fiber line to
be considered a new source:
Upgrades of existing pulping operations;
Upgrades or replacement of pulp screening and brown stock pulp
washing operations;
Installation of extended cooking and/or oxygen delignification
systems or other post-digester, pre-bleaching delignification
systems;
10-23
-------
Section 10 - Final Compliance Costs
Changes in methods or amounts of bleaching chemical applications;
Changes in the types of bleaching chemicals used;
Installation of new bleaching towers to facilitate replacement of
sodium or calcium hypochlorite; and
Installation of new bleached pulp washing systems.
3) It is substantially independent of an existing source at the same site (i.e., if
an existing mill builds and operates an entirely new fiber line that
supplements the capacity of an existing fiber line).
10.4.1 NSPS Compliance Costs for the BPK Subcategory
EPA analyzed the costs of two NSPS options for the control of toxic and
nonconventional pollutants: Option B and a TCP option (the latter is discussed in Section
10.4.1.1). Option A costs are also presented for comparison. EPA estimated NSPS capital and
operating compliance costs using a modified version of the cost model. EPA's costs are for the
complete replacement of a fiber line at a case study mill from Group C (see Table 10-1) that
produces approximately 1000 UBAD kkg/day of fully bleached pulp using two lines. (EPA notes
that the compliance costs estimated for this fiber line replacement would be the same for a fiber
line built at a greenfield mill.) EPA estimated compliance costs assuming most of the recovery
process units (i.e., recovery boilers and evaporators) are already in place and capable of
accommodating the pollutant load contributed from NSPS technologies (except for recausticizing
upgrades for which costs were estimated to account for the increased oxidized white liquor
requirement of OD, which is part of both Option B and the TCP option).
Table 10-20 presents EPA's estimated capital and operating costs. Unlike the
situation for retrofitting existing sources, Option B capital costs are very close to Option A capital
costs (i.e., $202 million versus $201 million) because the capital cost required for installation of
oxygen delignification for Option B at new sources is only slightly greater than the capital cost
required for a larger (and, thus, more costly) chlorine dioxide generator for Option A. However,
the reduced chemical demand for Option B results in lower operating costs and leads to lower
overall annualized cost compared to Option A.
For the BPK subcategory, EPA is also promulgating NSPS for the conventional
pollutants BOD 5 and TSS based on the performance of a secondary wastewater treatment system
as characterized by the average performance of the best 50 percent of the existing mills in the
subcategory using the appropriate level of control (see Section 8.7). EPA estimated the increased
capital cost required to comply with the promulgated NSPS for Subpart B rather than the old
NSPS discharge limitations. The new BPK subcategory encompasses four old subcategories
(former Subparts G, H, I, and P). Table 10-21 compares the old NSPS discharge limitations for
two of the former subcategories to the promulgated NSPS BOD5 discharge limitations. The two
10-24
-------
Section 10 - Final Compliance Costs
former subparts shown, Subparts I and G, had the lowest and highest BOD5 limitations of the four
subparts that EPA combined into the new Subpart B. EPA estimated additional cost of
compliance with the promulgated NSPS over the cost of compliance with the former NSPS for
Subparts I and G, in order to estimate the range of increased cost the promulgated NSPS would
require.
As discussed in Section 11 of this report, compared to Option A, implementation
of Option B technology will reduce the discharge of BOD5 from the pulping and bleaching
processes. EPA accounted for this BOD5 reduction when estimating the incremental cost of the
promulgated NSPS for BOD5.
EPA determined that the incremental capital cost of complying with the selected
NSPS for all pollutants (i.e., Option B plus conventional pollutant control) is 0.50 to 2.0 percent
greater than the capital cost of a new fiber line using Option A that complies with the previous
NSPS limitations for conventional pollutants.
10.4.1.1 TCF Technology as the Basis of NSPS for the BPK Subcategory
While Section 10.2.5 shows that retrofitting TCF technology is much more costly
than retrofitting ECF technologies, EPA notes that recent data from the construction of greenfield
TCF mills outside the U.S. suggest that the costs of such a fiber line may be less than that of a
greenfield ECF fiber line.
Table 10-20 includes the capital and operating costs required for a case-study mill
to install a greenfield fiber line using TCF technologies as the basis for minimizing new source
toxic and nonconventional pollutants. EPA notes that greenfield TCF fiber lines are less
expensive because they:
Obviate the costs required for chlorine dioxide manufacturing;
Require less physical space; and
Use bleaching towers that may be constructed of more inexpensive grades
of stainless steel compared to ECF bleaching towers which require more
expensive alloys and plastics to resist the corrosive action of chlorine
dioxide (and degradation products).
TCF technologies, however, are not demonstrated for the full range of bleached
kraft pulp production at this time. As a result, EPA intends to gather additional data to determine
whether TCF technologies may be available for the full range of market products subsequent to
this rulemaking. EPA will determine whether to propose revisions to NSPS based upon TCF and,
if appropriate, flow reduction technologies.
10-25
-------
Section 10 - Final Compliance Costs
10.4.1.2 NSPS Compliance Costs for PS Subcategory
The technology basis of NSPS for the three segments of the PS subcategory are
the same as the model BAT for those segments. At this time, EPA is not promulgating NSPS for
the control of conventional pollutants. As presented in Section 10.4.1 above, EPA found that for
the BPK subcategory the cost of NSPS technology is an insignificant fraction of the capital cost
of a new fiber line (i.e., 0.50 to 2.0 percent). Although EPA had no data specific to papergrade
sulfite mills with which to estimate the costs of a new fiber line for a PS mill, based on the analysis
of NSPS costs for the BPK subcategory, EPA expects that the NSPS costs for a PS fiber line
would also be an insignificant fraction of the capital costs.
EPA also notes that typical costs of including NSPS technology at a new source
mill are substantially less than the costs of retrofitting existing mills. Moreover, the reduced
operating costs for the NSPS option allow firms to recover the capital cost associated with the
NSPS technology.
10.5 References
1. Development Document for Proposed Effluent Limitations Guidelines and
Standards for the Pulp. Paper, and Paperboard Point Source Category. EPA-821-
R-93-019, U.S. Environmental Protection Agency, Washington DC, October
1993.
2. BAT and BMP Compliance Cost Estimates Report. Report prepared by ERG for
EPA. Record Section 23.1.3, DCN 13947, June 18, 1996.
3. BAT Baseline Database. Data on mill characteristics and operations collected by
EPA through the 1990 census questionnaire, subsequent contacts with mills by
phone, fax, and site visits. Record Section 21.10, DCN 13590.
4. BAT Cost Model Support Document. Report prepared by Radian Corporation for
EPA. Pulp, Paper, and Paperboard Rulemaking, Record Section 23.1.2, DCN
13953, 1996.
5. Cartwright, G. Memorandum: Costing Revisions Made Since Publication of July
15. 1996 Notice of Data Availability (61 FR 36871 Prepared by ERG for EPA.
Record Section 23.1.2, DCN 14493, 1997.
6. Analysis of Impacts of BAT Options on the Kraft Recovery Cycle (Recovery
Impacts Document). Report prepared by ERG and N. McCubbin for EPA.
Record Section 23.1.2, DCN 14490, 1997.
7. Comment Response Document. EPA, Washington DC, Record Section 30.11,
DCN 14497, 1997.
10-26
-------
Section 10 - Final Compliance Costs
8. Telephone conversation with Mr. W. Gillespie of NCASI. Record Section 23.1.1,
DCN 14651, May 5, 1997.
9. EPA Site Visit Report. Prepared by Radian Corporation for EPA. Record Section
7.4, DCN 7654.
10. Classified Appendix for Section 10 of the Supplemental Development Document
(Costs). Prepared by ERG for EPA. Record Section 23.1.3, DCN 14508.
11. Economic Analysis for the National Emissions Standards for Hazardous Air
Pollutants for Source Category: Pulp and Paper Production: Effluent Limitations
Guidelines. Pretreatment Standards, and New Source Performance Standards:
Pulp. Paper, and Paperboard Category - Phase I. Prepared by ERG for EPA.
Record Section 30.5, DCN 14649, 1997.
12. Kaplan, M. Effects of Reductions in Costs from Corporate Commitments.
Prepared by ERG for EPA. Record Section 27.4.4, DCN 14375, January 1997.
13. Technical Support Document for the Voluntary Advanced Technology Incentives
Program. EPA, Washington DC, Record Section 22.8, DCN 14488, 1997.
10-27
-------
Section 10 - Final Compliance Costs
Table 10-1
Baseline Technology Groups for BPK Mills
Group(a)
A
B
C
D
E
G
H
I
J
K
Number
of Mills
3
5
32
8
12
9
7
3
4
1
Example Bleaching
Sequences
CEH (c)
CEHD, CED
C/DEH,
C/DEHDED,
C/DED, C/DEDED
D/CEDED,
D/CEopDEpD
DEDED, DEopDD
EC or OD with
C/DEDED,
D/CEDED
EC or OD with
DEDED, DEopDD
EC and OD with
C/DEDED,
D/CEDED
EC and OD with
DEDED, DEopDD
TCF(c)
Costing Criteria
EC or
OD?(b)
No
No
No
No
No
Yes
Yes
Both
Both
Maybe
Chlorine
Used?
Yes
Yes
Yes
Yes
No
Yes
No
Yes
No
No
Hypo-
chlorite
Used?
Yes
Maybe
Maybe
Maybe
No
Maybe
No
Maybe
No
No
Percent of C1O2
Used
none on site
0 in first stage
<70
70 to 100
100
<100
100
<100
100
none
(a)Group E = BPK BAT Option A, Group H = BPK BAT Option B. Group F was eliminated because no mills belong
to the group.
is extended cooking (e.g., MCC, EMCC, RDH, or SuperBatch) and OD is oxygen delignification.
(c)Mills using this bleaching sequence do not usually bleach to full brightness.
10-28
-------
Section 10 - Final Compliance Costs
Table 10-2
Bleached Papergrade Kraft and Soda Process Technologies Costed
Mid-1995 Costing
Effort Option Number
Process Technologies Costed
(a)
Improved brown stock washing
Closed brown stock screening
Hypochlorite elimination
Oxygen and peroxide enhanced caustic extraction (Eop)
100% chlorine dioxide substitution(b) (ECF bleaching)
Implementing strategies to minimize kappa factor and brown stock precursors
B
Improved brown stock washing
Closed brown stock screening
Hypochlorite elimination
Oxygen and peroxide enhanced caustic extraction (Eop)
100% Chlorine Dioxide Substitution*' (ECF bleaching)
Implementing strategies to minimize kappa factor and brown stock precursors(c)
Kappa number of 15 for softwood and 10 for hardwood entering the first
bleaching stage through addition of oxygen delignification and/or extended
cooking(d)
(a)BAT/PSES technology options also include use of TCDD and TCDF precursor free defoamers, adequate chip
thickness control, and efficient biological wastewater treatment; however, costs were not included (see Section
10.2.1.2).
^The costs for high shear mixing are included in the capital costs for increased chlorine dioxide substitution.
(c)Mills may use many strategies to achieve this technology element. EPA estimates include oxygen and peroxide
reinforced extraction; improved brown stock washing; closed screening; and high shear mixing and control, which
are technologies integral for implementing strategies for minimizing kappa factor and brown stock precursors.
(d)Option B is defined as extended delignification resulting in a kappa number below 20 for softwood and below 13
for hardwood. Lower targets were used for costing to reflect the capability of modern OD systems (refer to Section
8.2.1.2).
10-29
-------
Section 10 - Final Compliance Costs
Table 10-3
Baseline Status of Bleached Papergrade Kraft and Soda Mills
100% Substitution
Option A (w/o OD or EC)
100% Substitution
Option B (w/ OD and/or EC)00
Total ECF Production (100%
Substitution)*'
Hypochlorite on Site (on at
least one line)
Oxygen Delignification Only
(OD)00
Extended Cooking Only (EC)(d)
ODandEC
Total Extended Delignification
(OD,EC,orODandEC)(e)
Percent of
Total Kraft
Production at
Proposal (%)
NC
NC
6.6
37.2
11.7
17.3
10.7
NC
Percent of
Total Kraft
Production at
Mid-1995 (%)
17.6
15.6
33.2
17.6
17.3
5.2
9.4
32.6
Baseline Estimate at
Proposal
# Mills
#Line
# Mills
# Lines
# Mills
# Lines
# Mills
# Mills
# Lines
# Mills
# Lines
# Mills
# Lines
# Mills
# Lines
Total
Mills®
NC
NC
NC
NC
6
9
37
9
13
12
22
8
10
NC
NC
87
Mid-
199500
14
24.5
14
16.5
27(c)
41
20
14
20
6
6.75
7
8
28
35.75
84
NC = not counted.
(a)Fractions denote the amount of time a technology is used on a swing line.
^Includes ozone-ECF production.
(c)Because one mill has a line at Option A and one line at Option B, the number of mills at Option A plus the number
of mills at Option B does not equal the total number of mills with ECF production.
(d)As noted in Section 10.2.2, the number of mills and lines using EC was overcounted at proposal.
(e)Includes ozone-ECF and TCP production.
(f)Refer to Section 4 for a description of the subcategory profile (i.e., total number of mills) at proposal and mid-
1995.
10-30
-------
Section 10 - Final Compliance Costs
Table 10-4
Comparison of Bleached Papergrade Kraft and Soda BAT, PSES, and BMPs
Compliance Cost Estimates
Cost
Capital [$ million]
Engineering O&M [$
million/yrl(b)
Annualized Cost [$ million/yr|
Annualized Cost [$/UBMT](c)
BAT, PSES, BMPs, and Closed
Screening Cost Estimated at
Proposal*8'
2,160
10.6
223
7.50
Option A (BAT,
PSES, and BMPs)
966
113
176
6.04
Option B (BAT,
PSES, and BMPs)
2,130
2.02
211
7.22
(a)See discussion of BMP proposal cost estimate in Section 10.2.4.4.
(b)The engineering operating and maintenance costs presented in this table differ from the annual costs presented in EPA's
economic analysis (10) because the annual costs include an additional four percent of engineering capital cost to account for non-
plant overhead costs.
(c)Using 29.2 million UBMT/yr for Options A and B, which is the mid-1995 total production for 84 BPK mills.
10-31
-------
Section 9 - Title
Table 10-5
BAT, PSES, and BMPs Compliance Cost Estimates for Direct and Indirect Discharging BPK Mills
PSES
BAT
BMPs(a) - direct
discharging mills
BMPs(a) - indirect
discharging mills
Total(c)
Option A
Capital
[$ million]
85.5
697
162
21.1
966
O&M
[$ million/yr]
4.56
79.3
26.9
2.05
113
Annualized
Cost [$/yr]
137.6
38.9
176
Annualized
Cost [$/t](b)
4.71
1.33
6.04
Option B
Capital
[$ million]
255
1,690
159
21.2
2,130
O&M
[$ million/yr]
(7.87)
(18.1)
26.0
2.0
2.02
Annualized
Cost [$/yr]
173
38.0
211
Annualized
Cost [$/t](b)
5.92
1.30
7.22
oo
to
() Represents cost savings.
(a)BMPs cost estimates for Option A and Option B differ. Refer to Section 10.2.4.4 for explanation.
^sing 29.2 million UBMT/yr for Options A and B, which is the total mid-1995 production for the BPK Subcategory.
(c)Total may not equal sum of column costs due to rounding.
-------
Section 10 - Final Compliance Costs
Table 10-6
Range of Estimated BAT, PSES, and BMPs Costs for the 84 Bleached Kraft
Mills
Range ($/mill)
Total Cost
Average Cost Per Mill
Option A
Capital
[$ million]
0.380 to 67.7
966
11.5
O&M
[$ million/yr]
(2.75) to 8.31
113
1.35
Option B
Capital
[$ million]
0.380 to 95.8
2,130
25.3
O&M
[$ million/yr]
(6.08) to 6.70
2.02
0.02
() Represents cost savings.
10-33
-------
Section 10 - Final Compliance Costs
Table 10-7
Component Capital Costs as Percentage of Total Capital Cost for the Bleached
Papergrade Kraft and Soda Mills
Capital Cost Component
Kappa Reduction (OD/EC)
BMPs(a)
Recovery Boiler
Evaporator
Closed Screening/Brown Stock Washing
C1O2 Generator
Adding Eop
Adding D-Towers (Eliminate Hypochlorite)
Recausticizing
Monitoring
Total Capital Cost00
Total Capital Cost for
Component [$]
Option A
NA
151,230,000
8,688,694
53,588,584
199,812,266
429,048,659
35,765,522
220,220,635
NA
16,008,404
1,114,362,762
Option B
1,514,011,241
151,230,000
10,638,969
53,588,584
199,812,266
262,360,284
35,765,522
164,556,880
37,824,295
16,008,404
2,445,796,444
Percent of Overall
Capital Cost
Option A
NA
13.6
0.8
4.8
17.9
38.5
3.2
19.8
NA
1.4
100
Option B
61.9
6.2
0.4
2.2
8.2
10.7
1.5
6.7
1.5
0.7
100
NA = Not applicable to option.
(a)See Section 10.2.4.4 for BMPs cost estimate discussion. The BMPs line item costs only reflect capital and
operating costs for implementing spill prevention and control systems. The cost impact on the recovery boiler or the
evaporator set from implementing BMPs is included in the line items for those areas.
°°Because the regional cost factor is not applied at the component cost level, to estimate component costs as a
percentage of total cost the total capital cost estimates are shown without the regional cost factor. As a result, the
total capital cost estimates shown above appear higher than the BAT, PSES, and BMPs final compliance cost
estimates shown throughout this section. In addition, totals may not equal the sum of component costs due to
rounding.
10-34
-------
Section 9 - Title
Table 10-8
Component Operating Costs as Percentage of Total Operating Cost for the Bleached Papergrade Kraft and
Soda Mills
Operating Cost Component
Kappa Reduction (OD/EC)
BMPs
Recovery Boiler
Evaporator
Closed Screening/Brown
Stock Washing
C1O2 Generator
Adding Eop
Adding D -Towers (Eliminate
Hypochlorite)
Recausticizing
Monitoring (Bleach Plant and
Final Effluent)
Additional Chemical Cost
Over Base
Supervision and Technical
Support
Total Operating Cost Component [$/yr]
Option A
Cost
NA
$6,384,600
$9,416,761
$2,021,791
$10,538,157
$1,822,350
$4,764,793
NA
$9,421,664
$72,059,428
$4,828,048
Savings
NA
$8,461,087
—
—
NA
—
—
—
Option B
Cost
$95,782,769
$6,384,600
$9,416,761
$2,021,791
$7,204,390
$1,851,941
$3,482,789
$756,486
$9,421,664
—
$10,645,282
Savings
$9,702,373
—
—
—
$135,243,178
—
Percent of Overall Operating Cost (%)
Option A
Cost
NA
5.3
7.8
1.7
8.7
1.5
3.9
0.0
7.8
59.4
4.0
Savings
NA
100.0
—
—
—
—
—
Option B
Cost
65.2
4.3
6.4
1.4
4.9
1.3
2.4
0.5
6.4
—
7.2
Savings
6.7
—
—
—
93.3
—
o
oo
-------
Table 10-8 (Continued)
Section 9 - Title
Operating Cost Component
Subtotal for Operating Costs
Subtotal for Operating
Savings
Total Operating Cost and
Percentages*'
Total Operating Cost Component [$/yr]
Option A
Cost
$121,257,592
—
Savings
$8,461,087
$112,796,505
Option B
Cost
$146,968,473
—
Savings
$144,945,551
$2,022,922
Percent of Overall Operating Cost (%)
Option A
Cost
100
—
100.0
Savings
100
100.0
Option B
Cost
100
—
100.0
Savings
100
100.0
NA = Not applicable for the option.
(a)See Section 10.2.4.4 for BMPs cost estimate discussion. The BMPs line item costs only reflect capital and operating costs for implementing spill prevention
and control systems. The cost impact on the recovery boiler or the evaporator from implementing BMPs is included in the line items for those areas.
^Totals may not equal the sum of component costs due to rounding.
-------
Section 10 - Final Compliance Costs
Table 10-9
Bleached Papergrade Kraft and Soda Mill Technology Upgrades Costed
Proposal'*0
(number costed)
Final
(number costed)
Screen Room Upgrades (Number of Mills)
Option A
Option B
CS
cs
40
40
Brown Stock Washer Upgrades (Number of Mills)
Option A
Option B
Add'l Stages
New Washers
Add'l Stages
New Washers
37
5
37
5
22
1
22
1
Evaporator Upgrades (Number of Mills)
Option A
Option B
NC
NC
20
20
Chlorine Dioxide Generator Upgrades (Number of Mills)
Option A
Option B
Greenfield
New
Upgrade
Conversion
Greenfield
New
Upgrade
Conversion
6
56
NC
18
6
48
NC
17
5
40
8
5
5
28
o
5
4
Oxygen Delignification Installations
Option A
Option B
# Mills
# Lines
# Mills
# Lines
NA
NA
56
96
NA
NA
65
96
Extended Cooking Installations (All Retrofits)
Option A
Option B
# Mills
# Lines
# Mills
# Lines
NA
NA
3
o
J
NA
NA
3
o
3
10-37
-------
Table 10-9 (Continued)
Section 10 - Final Compliance Costs
Proposal^
(number costed)
Final
(number costed)
Recovery Boiler Capacity Adjustments (Number of Mills)
-------
Section 10 - Final Compliance Costs
Table 10-10
Corporations Announcing Commitments to Upgrade Process Technologies to
Include BAT Elements After July 1, 1995
Corporation
International Paper
Champion International
Georgia-Pacific
Westvaco
Willamette Industries
Number of Affected
Mills
8
1
1
1
1
Announced
Commitment
100% C1O2
OD/100% C1O2
OD/100% C1O2
100% C1O2
100% C1O2
Reference
DCN 14326 Section 23. 1.1
DCN 13632 Section 23. 1.1
DCN 13 102 Section 23. 1.1
DCN 13600 Section 23. 1.1
DCN 13641 Section 23. 1.1
10-39
-------
Section 10 - Final Compliance Costs
Table 10-11
Baseline Status of Bleached Papergrade Kraft and Soda Mills Including Adjustment for
Corporate Commitments to Install BAT Elements
100% Substitution
Option A (w/o OD or EC)
100% Substitution
Option B (w/ OD and/or
EC)(b)
Total ECF Production
(100% Substitution)*'
Hypochlorite on Site (on at
least one line)
Oxygen Delignification
Only (OD)
Extended Cooking Only
(EC)(d)
ODandEC
Total Extended
Delignification (OD, EC, or
ODandEC)(e)
Percent of Production (%)
At
Proposal
NC
NC
6.6
37.2
11.7
17.3
10.7
NC
Mid-
1995
17.6
15.6
33.2
17.6
17.3
5.2
9.4
32.6
After
Commitments
25.5
21.4
46.9
17.6
18.8
5.2
9.4
34.1
Baseline Estimate
at Proposal
# Mills
#Line
# Mills
# Lines
# Mills
# Lines
# Mills
# Mills
# Lines
# Mills
# Lines
# Mills
# Lines
# Mills
# Lines
Total
Mills®
NC
NC
NC
NC
6
9
37
9
13
12
22
8
10
NC
NC
87
Mid-
1995(a)
14
24.5
14
16.5
27(c)
41
20
14
20
6
6.75
7
8
28
35.75
84
Mid-1995 w/
Commitments'3'
22
38.75
17
22.5
38
61
20
15
21
6
6.75
7
8
29
36.75
84
NC = Not counted.
(a)Fractions denote the amount of time a technology is used on a swing line.
^This includes ozone-ECF bleaching mill.
(c)Because one mill has a line at Option A and one line at Option B, the number of mills at Option A plus the number of mills at Option
B does not equal the total number of mills with ECF production.
(d)As noted in Section 10.2.2, the number of mills and lines using EC was overcounted at proposal.
(e)This includes ozone-ECF and TCP production.
(f)Referto Section 4 for a description of the subcategory profile (i.e., total number of mills) at proposal and mid-1995.
10-40
-------
Section 9 - Title
Table 10-12
Bleached Papergrade Kraft and Soda Total BAT, PSES, and BMPs Cost Estimates with Process Upgrades
Announced but not Underway by Mid-1995
Mid- 1995 Baseline
Baseline Adjusted for
Alternative Analysis
Option A
Capital
[$ million]
966
882
O&M
[$ million/yr]
113
74.1
Annualized
Cost [$/yr]
176
140
Annualized
Cost [$/t](a)
6.04
4.80
Option B
Capital
[$ million]
2,130
2,050
O&M
[$ million/yr]
2.02
(31.4)
Annualized
Cost [$/yr]
210
179
Annualized
Cost f$/t](a)
7.22
6.14
() Represents cost savings.
(a)Using 29.2 million UBMT per year for Options A and B, which is the mid-1995 total production for the Bleached Papergrade Kraft and Soda Subcategory.
-------
Section 10 - Final Compliance Costs
Table 10-13
TCF Process Technologies Costed
TCF Option
Process Technologies Costed
Ozone-based
(Option C)
Improved brown stock washing
Closed brown stock screening
Kappa number of 10 for softwood and 6 for hardwood entering the first bleaching
stage though oxygen delignification AND anthraquinone addition to the digester
Ozone bleaching (delignification)
Oxygen and peroxide enhanced caustic extraction
Substitution of peroxide bleaching for all chlorinated bleaching compounds (TCF
bleaching)(a)
Chelant addition
Bleach Sequence: OZEopQPZP
Peroxide-based
(Option D)
Improved brown stock washing
Closed brown stock screening
Kappa number of 10 for softwood and 6 for hardwood entering the first bleaching
stage though addition of oxygen delignification AND anthraquinone addition to the
digester
Substitution of peroxide bleaching for all chlorinated bleaching compounds (TCF
bleaching)(a)
Chelant addition
Bleach sequence: OQPP
''The costs for mixing are included in the capital costs for totally free chlorine bleaching.
10-42
-------
Section 10 - Final Compliance Costs
Table 10-14
Comparison of BAT and PSES Option Costs for the Bleached Papergrade
Kraft and Soda Subcategory
Capital Cost [$ million]
Operating Costs [$ million/yr]
Annualized Cost
[$ million/yr]
Annualized Cost [$/UBMT]
Option A(a)
(ECF)
966
113
176
6.04
Option B(a) (OD-
ECF)
2,130
2.02
211
7.22
Option C(b)
(Ozone-TCF)
5,630
849
1,170
40.0
Option D(b)
(Peroxide-TCF)
3,090
660
780
26.7
(a)Estimated using mill-by-mill costing approach.
Estimated using the model-mill costing approach detailed in Section 10.1.1.1.
10-43
-------
Section 10 - Final Compliance Costs
Table 10-15
Case Study Mill Incentive Tier Costs
Option A
Tier I
Tier II
Tier III
TCF Alternative
Capital Cost [$ million]
O&M [$ million/yr]
Annualized Cost [$/UBMT]
NA
NA
NA
NA
NA
NA
88.0
0.56
23.9
108
2.65
33.8
ECF Alternative
Capital Cost [$ million]
O&M [$ million/yr]
Annualized Cost [$/UBMT]
18.5
3.34
11.5
44.0
0.876
13.4
51.6
(0.682)
12.4
70.9
(0.134)
19.1
() Represents a savings.
NA = Not applicable because option/tier is not based on this type of bleaching process. However, EPA expects that
mills employing a TCF bleaching process will be able to achieve at least the Tier I Advanced Technology BAT
limitation for AOX and may be able to achieve the other ultimate Tier I limitations as well.
10-44
-------
Section 10 - Final Compliance Costs
Table 10-16
Papergrade Sulfite Technology Process Technologies Costed
Segment
A - Calcium-,
Magnesium-, or
Sodium-Based Sulfite
Pulping
B - Ammonium-Based
Sulfite Pulping
C - Specialty-Grade
Sulfite Pulping
Bleaching Option
TCP
ECF
ECF
Process Technologies Costed
Totally chlorine free bleaching (bleaching with
peroxide)
Elimination of hypochlorite
Oxygen and peroxide enhanced extraction
Improved pulp cleaning
100% chlorine dioxide substitution (ECF bleaching)(a)
Hypochlorite elimination
Peroxide enhanced extraction
100% chlorine dioxide substitution (ECF bleaching)(a)
Hypochlorite elimination
Oxygen and peroxide enhanced extraction
''The costs for high shear mixing are included in the capital costs for increased chlorine dioxide substitution.
10-45
-------
Section 10 - Final Compliance Costs
Table 10-17
ECF Versus TCF Costing for the Papergrade Sulfite Mills
Segment A - Calcium, Magnesium, or Sodium Sulfite
BAT Option = TCF
Total Mills
Number of mills costed for conversion to TCF
Number of mills currently bleaching at TCF (estimated costs for BMPs,
evaporator, monitoring)
Number of mills producing papergrade sulfite pulp, without bleaching
(estimated costs for BMPs, monitoring)
600
4
1
1
Segment B - Ammonium Sulfite
BAT Option = ECF
Total mills
Number of mills costed for conversion to ECF
Number of mills costed for conversion to TCF (one mill commented they
could feasibly convert to TCF bleaching with lower cost)
Number of mills currently bleaching at or committed to ECF (estimated costs
for BMPs, monitoring)
Number of mills currently bleaching at TCF (estimated costs for BMPs,
monitoring)
400
1
1
1
1
Segment C
BAT Option = ECF
Total mills
Number of mills costed for conversion to ECF
1
1
(a)One mill currently producing papergrade sulfite pulp has prepared business plans to produce specialty-grade pulp
but was costed as a Segment A mill because that was the mill's status as of mid-1995.
^One mill recently ceased papergrade sulfite operations; however, EPA includes the costs of this mill because as of
mid-1995 the mill was producing papergrade sulfite pulp. EPA also estimated the costs for the nine direct
discharging PS mills that are subject to BAT, which are CBI; however, this information is listed in Table VIII in
DCN 14508 (8)).
10-46
-------
Section 10 - Final Compliance Costs
Table 10-18
Total BAT, PSES, and BMPs Papergrade Sulfite Compliance Cost Estimates
(All Segments)
Cost Estimated at Proposal
Cost Estimates for Mid-
1995(a)
BAT, PSES, and BMPs Compliance Cost Estimates
Capital
[$ million]
88.3
73.8
O&M
[$ million/yr]
17.8
4.59
Annualized Cost
[$ million/yr]
21.5
10.6
Annualized Cost
[$/t]
19.06
8.24
(a)Although a cost breakdown for each segment is not presented because this information is CBI, this information is
listed on Table VIII in DCN 14508 (8).
10-47
-------
Section 10 - Final Compliance Costs
Table 10-19
Papergrade Sulfite Technology Upgrades
Installation of Final P-Stage
Chlorine Dioxide Generator
Upgrades
Add New D-Tower (Eliminate
Hypochlorite)
Add Eop
Evaporator Upgrades
Segment A
(Calcium, magnesium, or
sodium sulfite)
BAT Option = TCF
4
0
0
3
3
Segment B
(Ammonium sulfite)
BAT Option = ECF
1
1
1
2
3
Segment C
(Specialty
Grade)
ND
ND
ND
ND
ND
ND = Not disclosed to protect confidential business information. The technologies for the Segment C mill are
identified in Table IX in DCN 14508 (8).
Note: Technology upgrades included in proposal cost estimates are not presented because the proposed option is not
comparable to the revised options.
10-48
-------
Table 10-20
NSPS Compliance Costs
Section 10 - Final Compliance Costs
Toxic and Conventional Pollutant Control
Typical Bleach Sequence
Option A
DEonDnD
Option B
OODEonD
TCF
00(QW)OP(ZQ)(PO)
Capital Costs ($ million)
Unbleached Pulp Mill
New Continuous Digester
New Brown Stock Washing Line
New Closed Screening System
Building and Infrastructure
53.0
19.4
5.94
6.00
53.0
19.4
5.94
6.00
53.0
19.4
5.94
6.00
Bleach Plant
OD System
New D-Stage Tower and Washer
New Eop Stage with Washer
New D-Stage Tower and Washer
New E2 Stage with Washer
New D-Stage Tower and Washer
Chelant Stage with Press Washer
Pressurized PO stage with Washer
High Consistency Ozone System
Pressurized PO Stage with Washer
Chelant Supply System
Peroxide Unloading and Storage
Monitoring
Buildings
Miscellaneous Infrastructure
Greenfield C1O2 Generator
C1O2 Storage
Upgrade Recausticizing
Total Capital Cost
—
15.5
11.3
15.5
10.2
15.5
—
—
—
—
—
0.125
0.124
12.0
13.6
21.6
1.47
—
201
29.4
15.5
11.3
15.5
—
—
—
—
—
—
—
0.125
0.124
12.0
14.4
16.2
1.06
3.10
202
29.4
—
—
—
—
—
4.77
9.55
25.7
9.55
0.200
0.125
—
6.00
15.9
—
—
4.65
190
Operating Costs
Annualized Costs ($/t of pulp)
Annual Cost ($/year)
112
39,200,000
102
35,600,000
96.9
33,900,000
Conventional Pollutant Control ($)
Capital Cost ($)
162,000 to 3,400,000
10-49
-------
Section 10 - Final Compliance Costs
Table 10-21
NSPS Limitations
NSPS
Conventional Pollutants
Old Subpart I - Fine Bleached
Kraft(a)
Old Subpart G- Market Bleached
Kraft(a)
New Subpart B - Bleached
Papergrade Kraft
Former NSPS Limitations
(kg/kkg)
BODS
2.22
4.02
—
TSS
—
—
—
Promulgated NSPS (kg/kkg)
BODS
—
—
1.73
TSS
—
—
—
(a)These subcategories were distinct in the previous rulemaking; however, all previous bleached kraft subcategories
have been reorganized in one subcategory, Subpart B, by this rulemaking. These two former subcategories represent
the range of limitations that are being revised by the promulgation of the limitation for Subpart B in this rulemaking.
10-50
-------
Section 11 - Non-Water Quality Environmental Impacts
SECTION 11
NON-WATER QUALITY ENVIRONMENTAL IMPACTS
11.1 Impacts of BAT, PSES, and BMPs on the Bleached Papergrade Kraft and
Soda Subcategory
This section describes EPA' s estimate of the water quality environmental
impacts of the final options considered for effluent limitations guidelines and standards for the
bleached papergrade kraft and soda subcategory of the pulp and paper industry. The non-water
quality environmental impacts for the incentives tiers are presented in detail in the Technical
Support Document for the Voluntary Advanced Technology Incentives Program (1). Major
impacts are summarized in Table 11-1 and discussed further below.
The estimated effects of BAT, PSES, and BMPs on mill effluents, atmospheric
discharges, and energy consumption discussed in this section are based on the assumption that all
mills adopt the Option A or Option B process technologies described in Sections 8.2.1.1 and
8.2.1.2, respectively. This section also presents EPA's analysis of non-water quality
environmental impacts for TCP bleaching processes.
11.1.1 Summary of Impacts on Wood Consumption
As discussed in detail in Section 11.2, EPA estimates wood consumption could be
reduced by up to 0.3 percent by the implementation of either Option A or Option B and BMPs.
The reduction in wood consumption is a result of the reduction in losses of useful fiber
associated with the recovery of spills (BMPs) and improvements in washing and screening of
pulp.
11.1.2 Summary of Impacts on Wastewater Flow, BOD5, and Solid Waste
Generation
As discussed in detail in Section 11.3, the BAT, PSES, and BMP options analyzed
by EPA will result in progressive reductions in process wastewater flows and pollutant loadings.
Generally, the reductions for Option B are greater than Option A.
The average US bleached kraft and soda mill discharges approximately 95 m3/kkg
of treated effluent. EPA estimated that Option A could result in process wastewater flow
reductions ranging from 10 to 50 m3/kkg. The greater reductions would be realized in mills
presently discharging the highest flows. Option B would result in an additional process
wastewater reduction of up to 15 m3/kkg at mills with the highest effluent flows. See Section
11.3 for a detailed discussion.
EPA also estimates that raw (untreated) BOD5 loads would be reduced by 21
percent through implementation of Option A, and further (31 percent) by Option B. This
11-1
-------
Section 11 - Non-Water Quality Environmental Impacts
reduction would have only a modest effect on treated effluent discharges, but would reduce
energy consumption and solid waste generation. Energy equivalent to approximately
680,000 bbls oil would be saved in the waste treatment plants by Option A and 1,000,000 bbls
oil by Option B (see Section 11.4.2.2).
EPA estimates that the reduction in BOD5 load to activated sludge wastewater
treatment plants (WWTP) will result in a 2 percent reduction in the generation of secondary
wastewater treatment sludge for Option A. Option B will result in a 3 percent reduction.
11.1.3 Summary of Impacts on Energy Consumption
Section 11.4 of this report provides the results of EPA' s detailed analysis of the
energy requirements of BMPs combined with Option A and Option B. As detailed in Section
11.4, bleached kraft mills generate a significant proportion of the energy necessary to operate
pulping and bleaching processes through the chemical recovery process. Implementation of
Option A or Option B in combination with BMPs would increase recovery of organic material
with a resulting increase in energy generated at the mill.
The most useful measures of kraft mill energy performance are the quantity of
energy purchased, including energy associated with off-site manufacturing of bleaching
chemicals, and energy needed for mill effluent treatment. Implementing Option A would
increase purchased energy consumption by approximately one percent, while implementing
Option B would reduce it by one percent.
11.1.4 Summary of Impacts on Atmospheric Emissions
Section 11.5 of this report provides the results of EPA' s detailed analysis of the
air pollution impacts of Option A and Option B. The process changes related to these options
decrease the emissions of some HAPs but have little impact on others. Overall, the emission of
total HAPs from the sources controlled by MACT I decrease by 7 percent compared to baseline
for BAT Option A.
Implementation of Option A may marginally increase the emission of HAPs from
the recovery furnace by up to 1.5 percent, while Option B may result in a marginal increase of up
to about 2.2 percent. However, capacity adjustments and upgrades1 to recovery boilers as part of
mill modernization programs could also result in reduction of emissions below current levels.
'Capacity adjustments and upgrades that would generally reduce emissions of some or most pollutants include
raising the solids concentration of the black liquor fired, improved turbulence and control of air system, improved
boiler instrumentation, automatic port rodding, extraction of lignin from black liquor, heat treatment of black liquor,
and improvement of pulping yield. However, EPA estimates that only one recovery boiler would require an air
system and control upgrade in order to achieve the effluent limitations guidelines and standards promulgated for
Subpart B. See Analysis of Impacts of BAT Options on the Kraft Recovery Cycle (DCN 14490) (2).
11-2
-------
Section 11 - Non-Water Quality Environmental Impacts
As explained in Section 11.5.4, the pollution prevention measures implemented at
mills in response to BAT and PSES would have no direct effect on emissions of carbon dioxide,
the greenhouse gas of concern in this analysis. However, emissions of carbon dioxide will
change approximately in proportion to the changes in energy consumption mentioned above due
to the secondary effects of modifying the quantity of fossil fuel burned to serve the pulp industry.
EPA concludes that the increased CO2 emission attributable to Option A pose no unacceptable
non-water quality environmental impact.
Sections 11.5.5 and 11.5.6 present EPA' s analysis of carbon monoxide emissions
related to effluent limitations guidelines. Emissions of carbon monoxide will increase from
bleach plants (1,500 kkg/yr over 1995 status) and from combustion of black liquor solids will
increase (1,440 kkg/yr over 1995 status, for a total increase of 2,940 kkg/yr over 1995 status) if
Option A technologies are implemented. If all mills implement Option B technologies, the
increase in carbon monoxide emissions would be approximately 240 kkg/yr and 2,120 kkg/yr
over 1995 status, respectively.
11.2 Wood Consumption
This section describes the effect this rulemaking will have on wood consumption
at pulp mills.
11.2.1 BAT and PSES Option A
EPA analyzed the impact of Option A on wood consumption. The effluent flow
from a typical open screening system ranges from 10 to 25 m3/kkg pulp, as discussed below in
Section 11.3. This screen room effluent normally contains 20 to 50 mg/1 wood fibers, in addition
to whatever screen rejects are discharged with the water. The quantity of fiber lost continuously
will therefore range from 0.2 to 1.25 kg/kkg pulp or approximately 0.1 percent of production.
When converting a screen room to closed operation, this loss to effluent discharge is eliminated
by capture and recycle to the brown stock washers.
In any mill, incidents occur from time to time that result in all of the production
being dumped on the floor for a short period. The most common reason is an overflow of a
vacuum drum washer, but any pipe can burst or tank overflow. Operators react quickly to such
significant malfunctions, but 5 to 30 minutes is often necessary to stop the discharge. Assuming
a typical tank overflow lasts 15 minutes, one percent of one day' s production would be lost.
Traditionally, the pulp and associated black liquor lost in this way is washed down the sewer, and
ultimately lost from the process. With the implementation of BMPs, however, most of the
material which escapes the production equipment will be recovered. Sufficient data are not
available to quantify such losses, but losses of 2 kg/kkg would not be unusual, so it can be
assumed that implementation of BMPs will result in a small improvement in wood yield due to
recovery of this pulp.
11-3
-------
Section 11 - Non-Water Quality Environmental Impacts
Thus, mills with open screening at baseline that implement Option A technology
(including closing the screen room) after the implementation of BMPs, would experience at least
0.1 percent reduction in wood consumption and more likely up to 0.3 percent while maintaining
current production.
11.2.2 Option B
EPA estimated that Option B would also result in up to 0.3 percent reduction in
baseline wood consumption. Option B differs from Option A only by the inclusion of extended
delignification (oxygen delignification or extended cooking). EPA concluded that the
installation of oxygen delignification without changing pulping conditions (EPA Option B)
would not incrementally affect overall process yield when compared to Option A (3).
11.3 Effluents and Solid Waste
Implementation of BAT, PSES, and BMPs will reduce effluent flow, as well as
the load of organic substances and suspended solids discharged to the mills" effluent treatment
systems. The reduction in BOD5 and suspended solids discharges will lower energy consumption
and sludge generation in mill wastewater treatment systems and POTWs receiving mill
wastewater, as discussed below.
11.3.1 Effluent Flows
The total effluent flow from an integrated bleached papergrade kraft and soda mill
is normally between about 50 and 150 m3/kkg pulp produced, although a few mills discharge
significantly lower or higher flows (4). The average US bleached kraft and soda mill discharges
approximately 95 m3/kkg (5), which corroborates Mannisto" s graphs (4). For a 1,000 kkg/day
mill, the average effluent flow is similar to that from a city of 250,000 people.
EPA found that bleach plant flows differ by furnish. The average flow for a
hardwood line, not employing extended delignification, was 25 m3/kkg pulp. The flow for a
comparable softwood line was 37 m3/kkg pulp. Bleach plant flows differ because the quantity of
organic material removed from the pulp in bleaching hardwoods is approximately 50 percent less
than that removed from bleaching softwoods. Thus, hardwood bleaching lines often use a smaller
number of stages than softwood lines use and generate less wastewater.
11.3.1.1 Flow Reduction Resulting from Screen System Closure
The BAT element that has the largest affect on effluent flows is closure of the
brown stock screening systems ("closure" eliminates all planned effluent discharges from the
screen room). EPA" s records show that approximately half the bleached kraft mills still
operated open screen rooms in 1995 (EPA BAT Baseline Database) (6). EPA does not have data
tabulating the effluent discharges from these screen rooms. Therefore, EPA estimated flow from
open screen rooms using a mass balance assuming normal pulp consistencies. Results show that
11-4
-------
Section 11 - Non-Water Quality Environmental Impacts
an open screening system could contribute up to 70 m3/kkg pulp, if all dilution is by fresh water.
EPA ultimately rejected this mass balance analysis, however, because it concluded that screening
systems were unlikely to operate with such high discharges today.
Springer (1986) states that a poorly designed and operated open screen room
could require up to 150 m3/kkg fresh water, thus causing a similar amount of effluent to be
discharged (7). He also states that an older open screen room can be operated with 20 to 25
m3/kkg discharge. In the TDD, EPA showed that the average effluent flow from the pulping area
of a group of mills that included both mills with open screening and mills with closed screening
systems was 16.4 m3/kkg pulp. This amount includes the unbleached white water from the
screen room, digester condensates, and miscellaneous flows. Based on these data, EPA
concluded that discharges typical for open screen rooms could range from 10 to 25 m3/kkg pulp.
Thus, the conversion of a screen room from open to closed will typically reduce mill effluent
flow by 10 to 25 m3/kkg pulp, or approximately 10 to 20 percent from the average mill flow (95
m3/kkg).
11.3.1.2 Flow Reduction Resulting from BMPs
BMP implementation will reduce effluent flow in three ways:
The recovered black liquor will likely be reused instead of discharged to
the mill effluent treatment system. The objective of BMPs is to reduce
discharge of organic substances by improving the degree of process
closure of the mill. Although up to 34 kg black liquor solids per ton pulp
may be recovered as a result of BMPs, the effects on flow are modest. The
quantities have been estimated for each mill and are generally in the order
of 1 m3/kkg pulp.
The attention paid to miscellaneous discharges and the efforts that will be
made to avoid clean water discharges diluting the recovered spills will
result in a further reduction, as estimated by EPA, in effluent discharges of
1 to 2 m3/kkg in most mills.
Since mills are expected to segregate clean cooling water to avoid dilution
of spilled black liquor, it will be possible to either discharge these clean
waters separately from contaminated wastewater, or to reuse the clean
water. Reuse of clean water could result in about 2 percent reduction in
effluent flow from mills that choose to make use of this potential benefit
of BMPs.
11.3.1.3 General Effects of BAT on Effluent Flows
Other than closing screen rooms and BMPs described above, no elements of the
two BAT options will reduce effluent flows directly. However, the application of current
11-5
-------
Section 11 - Non-Water Quality Environmental Impacts
engineering practices to the design of new systems and equipment will result in conservation of
water. The greatest improvements are likely to be seen in mills currently using relatively high
quantities of water.
Retrofitting an oxygen delignification system (which would be a practical
necessity for compliance with Option B) has no direct effect on effluent flows by itself. Some
mills have reported reductions in effluent flow in oxygen delignification projects because it is
normal practice to close the screen room process by recycling the screen decker filtrate to brown
stock washing when oxygen delignification is installed. In some cases, as noted in the TDD,
when the unbleached pulp kappa number into bleaching is reduced, one or two complete
bleaching stages can be retired (e.g., convert a CD Eo DED bleach plant to an O D EopD). Such
action could reduce effluent flows by about 15 m3/kkg pulp. In rare cases, oxygen delignification
will result in some water conservation if lower unbleached pulp kappa number into bleaching
allows the use of reduced wash water flow in the first bleaching stage.
Mills reduce the kappa number of unbleached pulp entering the bleach plant by
two types of extended delignification: extended cooking and oxygen delignification. EPA' s
data, presented in Table 11-2, show lower bleach plant effluent flows in mills with extended
cooking or oxygen delignification.
When upgrading the first chlorine/chlorine dioxide stage to high or 100 percent
chlorine dioxide substitution for chlorine, low consistency operations are usually converted to
medium consistency, or increase the use of recycled bleach filtrates for pulp dilution to raise the
temperature without incurring the cost of direct steam heating. These changes can lead to a
reduction in bleach plant effluent flows of about 12 m3/kkg pulp in softwood mills and 5 m3/kkg
in hardwood mills. Such improvements are most likely to be made in mills which have high
effluent flows.
During mill renovation, new equipment is not installed in isolation. Instead, it is
common practice to modernize the mill area involved, at least to some extent. Modern
equipment is generally designed to conserve water more effectively than older designs. Many
details can be involved, such as the replacement of packing on shafts with modern mechanical
seals that use little or no water, or reduction in cooling water requirements by more efficient
design. These modifications will generally reduce effluent discharges modestly, but it is difficult
to provide realistic numeric estimates.
11.3.2 Solid Wastes
EPA estimates that implementation of Option A and Option B would result in a
reduction in the generation of sludge in the effluent treatment systems. The reduction in
generation of wastewater treatment sludge results from the decrease in organic load discharged to
the effluent treatment system.
11-6
-------
Section 11 - Non-Water Quality Environmental Impacts
Commenters have expressed concerns that modifying mills to approach closure of
the water cycle would result in large increases in solid waste requiring disposal. This issue
always requires careful consideration, since improving effluent discharges by simply transferring
wastes to another medium is clearly undesirable. The only available study supported by detailed
engineering analysis and mill experience which considers the technologies involved in the
present discussion shows that the rate of solid waste generation for a closed cycle mill would be
lower than the current industry average by a factor of about three (8).
EPA has not found any detailed analysis in the literature which would suggest that
solid waste generation would increase as a result of partial mill closure. The process changes
that are elements of Options A and B are not expected to cause the generation of additional
quantities of solid waste. None of the "very low effluent" mills discussed in the Technical
Support Document for the Voluntary Advanced Technology Incentives Program generates large
quantities of solid waste (1). Review of the pollution prevention technologies being developed
for flow reduction for the pulp and paper industry suggests that increased solid waste generation
can be readily avoided.
Because of impacts on energy use, implementation of BAT would cause some
small change in solid waste generation at utility power stations burning coal. Option A would
increase generation of solid waste, while Option B would cause a reduction. EPA has considered
these changes to be negligible and has not attempted to estimate quantities of this material.
11.3.2.1 Current Sludge Disposal
An analysis of sludge disposal practices in the late 1980s showed that mills
bleaching chemical pulp were disposing of 2.5 million dry kkg/year (9). Since the bleached
papergrade kraft subcategory makes up about 90 percent of the bleached chemical pulp
production, for this analysis, EPA assumed that this subcategory also contributes approximately
90 percent of the sludge. The sludge generation rate is equivalent to approximately 80 kg/kkg
pulp produced.
As reported in 1991, approximately 52 percent of the sludge was being landfilled,
20 percent stored in surface impoundments, 9 percent incinerated, and 7 percent applied
(presumably beneficially) to land. The rest was disposed of by various means including ocean
outfall, selling, and mixtures of two or more of the above.
11.3.2.2 Primary Sludge
The overall tightening of mill systems due to closing screen rooms, BMPs, and the
improvement of washing systems will reduce fiber losses. The reduction in fiber discharges will
vary, and will be the greatest in mills presently discharging relatively large amounts of fiber. As
discussed in Section 11.2.1, fiber losses of 0.1 to 0.3 percent would be eliminated. Assuming, on
average, that 0.2 percent fiber is eliminated, approximately 140,000 kkg/yr fiber would be
recovered if all screen rooms previously open were closed. This figure includes the general
11-7
-------
Section 11 - Non-Water Quality Environmental Impacts
tightening up that would be associated with implementation of BAT, PSES, and BMPs due to
reductions of spills and also reductions in flows of the weak white water and filtrates that
typically contain 20 to 100 mg/L fiber.
NCASI estimated that the average bleached kraft mill generated 57 kg of sludge
per ton product in 1989, on the basis of an industry survey (10). Consideration of mill practices
suggest that this quantity has been reduced somewhat since 1989, but EPA has no definitive data.
The average reduction of 0.2 percent (2 kg/kkg) derived from above sources would represent a
reduction in primary sludge discharges of 4 percent from the 1989 discharges. Because closed
screen rooms, BMPs, and effective brown stock washing are common to both Option A and
Option B (and because the extended delignification process unique to Option B does not affect
fiber losses), EPA estimates no difference of primary sludge between the options.
11.3.2.3 Secondary Sludge and BAT
All but one of the bleached papergrade kraft mills in the US employ secondary
wastewater treatment either on site or through a POTW. (The mill without secondary treatment
discharges to the ocean.) Mills either have activated sludge treatment systems (AST), aerated
stabilization basins (ASB), or some combination of these types. In order to consider the effects of
Options A and B on secondary sludge, EPA considered not only secondary wastewater treatment
systems at direct-discharging mills with ASTs, but also POTWs with ASTs receiving
predominantly bleached papergrade kraft and soda effluent. All secondary treatment systems
create sludge by converting dissolved organic material (BOD5) into biomass. However, much
more sludge is generated by AST than by ASB. In addition, sludge is routinely wasted from
AST while it is typically left to degrade biologically in an ASB. Therefore, the estimate of the
reduction in solid waste generation resulting from BAT focused on mills employing AST.
The quantity of solid waste produced by activated sludge or similar wastewater
treatment processes is proportional to the BOD5 load on the treatment system. Secondary
wastewater treatment sludge can be the major source of solid waste in a mill. Four of the nine
POTWs that process wastewater from indirect-discharging bleached kraft mills use aerated
stabilization basins, and, therefore, generate little sludge for disposal. The other five POTWs use
AST (11). Totaling direct and indirect dischargers that use AST, about 30 percent of the
bleached papergrade kraft mills use AST and produce secondary wastewater treatment sludge.
Black liquor solids have a BOD5 of approximately 0.3 kg BOD5/kg BLS.
Approximately 0.6 kg of biological (secondary) sludge is generated in an activated sludge system
for each kg BOD5 applied (12). This relationship was used along with an estimate of the
reduction in BLS that would result from implementation of Option A and Option B and 1995
baseline sludge generation estimates to calculate the associated changes in sludge generation
(13). Additional BLS combusted and decreases in BOD5 and sludge generation are shown in
Table 11-3.
11-8
-------
Section 11 - Non-Water Quality Environmental Impacts
In 1991, EPA determined that the 104 bleached chemical pulp mills discharged
approximately 2.5 million dry tons per year of sludge from wastewater treatment plants (14). No
reason exists to suppose that a large change has taken place since that time. The bleached
papergrade kraft subcategory produces approximately 90 percent of the pulp produced by
bleached chemical pulp mills, and consideration of the processes generally used indicates that the
quantity of sludge discharged will be approximately in proportion to production.
The foregoing quantities of solid waste include only the dry material. If the
sludge is landfilled, it will probably be about 40 percent dry concentration, so the total weight
basis requiring disposal will be about 2.5 times the dry quantity.
11.3.2.4 Dioxin and Furan in Sludge
Sludge generated at bleached papergrade kraft and soda mills may contain dioxin
and furan if these pollutants are found in wastewater treated at these mills. At proposal, the
Agency estimated that the mills in these two subcategories generated 177 g/yr toxic equivalent
(TEQ) dioxin in their wastewater treatment sludge. Since the proposal, industry has significantly
reduced the level of dioxin and furan in its wastewater. The Agency estimates that the dioxin
and furan content of the sludge has decreased similarly, to approximately 50 g/yr TEQ.
The control technologies that form the basis of the BAT limitations and PSES
promulgated today limit the concentration of dioxin and furan allowed to be discharged. As a
result, the Agency estimates that when fully implemented, the combined application of BAT
limitations and PSES will reduce the present sludge loading of dioxin TEQ by 43 g/yr,
approximately an 85 percent reduction from current levels.
11.3.2.5 Aerated Stabilization Basins
Approximately 70 percent of mills in the bleached papergrade kraft subcategory
use ASBs, some in combination with activated sludge treatment (6). Though ASBs generate
much less sludge than activated sludge treatment, they often become partially filled with sludge
after a number of years of operation, and require dredging. Lightly loaded ASBs have the ability
to mineralize organic sludge, and operate for many years without cleanout. As discussed above,
the BAT options will reduce the discharge of BOD 5 and suspended solids to treatment and thus
reduce ASB dredging frequencies.
11.3.2.6 Potassium and Chloride Purges
In a conventional, relatively "open" kraft mill, non-process elements such as
potassium and chloride are eliminated from the system by discharge in the milF s wastewater.
Many authors, including Tran (1990) (15), have shown that as mills approach process closure,
the concentrations of chloride and potassium throughout the liquor system rise, and can cause
plugging on the surfaces of the chemical recovery boilers exposed to the products of combustion
(i.e., fireside).
11-9
-------
Section 11 - Non-Water Quality Environmental Impacts
Potassium and chloride concentrate in the dust caught in the electrostatic
precipitator of the kraft mill recovery boiler. To control the concentrations of potassium and
chloride in the mill" s cooking cycle, some mills with excellent BMPs and operating practices
which minimize losses from the green/white/black liquor cycle have to remove and discharge a
portion of the precipitator dust, which is a mixture of inorganic salts of sodium and potassium.
The total quantities of these substances discharged with the precipitator dust is identical to the
quantity previously discharged with the pulp mill and bleach plant effluents. The point of
discharge from the cycle has simply moved.
The precipitator dust discharge, which may be up to 20 kg/kkg pulp, has been
described as a solid waste discharge in some documents. However, in many mills, the dust never
exists in dry form except between the plates of the precipitator, and is normally discharged as a
solution in the effluent2. Today, mills commonly discharge this material with the effluent.
Most of the potassium in a mill system enters with the wood and purchased
chemicals (15). The potassium entering with the wood will be discharged by any mill, whether
operating like a pre-1970 mill, or in accordance with the most advanced BAT criteria. The
quantity of potassium entering with the chemicals, and hence discharged, will be less in the more
advanced mills, since the quantity of chemicals purchased will drop due to recycle. The mill
operator is also likely to avoid purchasing contaminated chemicals to minimize the problems
caused by potassium in the mill.
11.4 Energy Impacts
11.4.1 Overview of Energy Impacts
Sections 304(b) and 306 of the Clean Water Act specifically direct EPA to
consider the energy requirements of effluent limitations guidelines and standards it establishes.
EPA estimated the impacts of BAT, PSES, and BMPs on the energy use of the 86 mills with
production in the bleached papergrade kraft and soda subcategory. For Option A and Option B,
combined with BMPs, EPA analyzed the following changes in energy use:
On-site electrical demand within the mill;
Electrical demand for wastewater treatment;
Pulp mill and bleach plant process steam demand; and
Off-site electrical demand resulting from manufacture of bleaching
chemicals (primarily raw materials for on-site C1O2 generation).
Quantities are small. The BFR process at Canton, NC, which is the largest chloride removal system operating in the
US, discharges approximately 30 mVday, or 0.03% of total mill discharge flow.
11-10
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-4 presents EPA's estimate of the effect of Option A and Option B on
energy consumption relative to consumption in 1995. The estimated energy impacts were
converted to an "oil equivalent," to conveniently present the combined changes in thermal energy
and electric power. As depicted in Figure 11-1, EPA estimated that Option A would result in an
increase in oil consumption of 840,700 bbl/year while Option B would result in a decrease in oil
consumption of approximately 1,535,000 bbl/year. The energy savings demonstrated by Option
B is primarily due to replacement of a portion of chlorine dioxide bleaching chemicals by oxygen
(in oxygen delignification). Manufacture of oxygen requires substantially less electrical energy
that the manufacture of chlorine dioxide of equivalent bleaching power.
11.4.2 Estimation of Energy Impacts
Estimates of the energy impacts of implementing the technology required to meet
Option A and Option B are discussed in this section.
11.4.2.1 Calculation Methodology of Energy Impacts for Option A and Option B
Process Changes and BMPs
EPA evaluated the effect of each process change element of Option A and
Option B in each mill on demand for steam and electrical energy. The process changes which
have a significant effect are listed in Table 11-5. Items described as "insignificant" or "minor"
were excluded from calculations of changes in energy consumption because they have no
discernible impact within the accuracy of the estimate. In addition to the explicit process
changes, the consequential effects of reducing effluent flow and BOD load have an effect on
energy consumption in the mills" wastewater treatment plants.
EPA estimated, on a mill-specific basis, the process changes that each BPK mill
would need to make in order to implement Option A and, separately, Option B. Based on these
estimates, EPA calculated the mill-specific electricity involved. Details of the assumptions and
associated equations are defined in the BAT Cost Model Support Document (16). The cost
model equations used for the calculations in the report reflect comments received by EPA from
the pulp and paper industry and the public on the 1993 and 1996 versions of the cost model. The
calculations included changes in power demand for both the mill site and for the manufacture of
the principal bleaching chemicals used for each process variation.
The manufacture of sodium chlorate for mill-site chlorine dioxide generation is a
major factor in off-site electrical energy demand. Production of chlorine dioxide requires
approximately 11 kWh/kg, whereas the equivalent quantity of chlorine requires only about
5 kWh/kg, and the equivalent quantity of oxygen about 1 kWh/kg. All of the potential bleach
plant modifications will reduce the demand for electrolytically produced caustic, thus reducing
demand for off-site electrical energy. The difference in power required for the various alternative
bleaching processes are calculated in the cost model, and are included in the data presented in
Figure 11-1 and Table 11-4.
11-11
-------
Section 11 - Non-Water Quality Environmental Impacts
For both Option A and Option B, the need to generate steam by burning fossil fuel
at the mill site will be reduced by the heat generated from burning black liquor recovered by
improved washing, closing the screen rooms, and BMP. For Option B, a small additional
increase in recovery of heat energy will occur due to the incorporation of oxygen delignification,
which recovers organic material otherwise discharged to the effluent treatment system in mills
that do not use oxygen delignification. EPA estimated that the recovered heat would be
approximately one percent of the base case heating value of the black liquor burned at the mill.
Heat energy is also consumed in evaporating the recovered black liquor and providing heat to
oxygen delignification systems. The net effect was calculated for each mill, and the industry-
wide total is shown in Figure 11-1 and Table 11-4.
11.4.2.2 Calculation of Energy Impacts from Effluent Treatment System Operation
Wastewater from all but one US bleached kraft mill is treated in a biological
wastewater treatment system prior to discharge to the environment. These treatment systems are
equipped with aerators to facilitate biochemical oxidation of the wastewater BOD 5 load. As
described in Section 11.3.2.3, implementation of BAT or further pollution prevention technology
will reduce the BOD5 load requiring treatment.
Biodegradation of BOD5 requires approximately 1.25 kWh per kg BOD5 to
adequately aerate the wastewater (derived from Kocurek, 1992; also a widely accepted value)
(12). As described in Section 11.3.2.3, the effects of Option A and Option B on BOD5 loads to
the effluent treatment systems were calculated mill by mill on the basis of the quantity of
recovered organic material (black liquor solids). EPA assumed that each kg of black liquor
solids would exert 0.3 kg BOD5 and the reduction in BOD5 load was calculated for each mill.
Where the BOD5 load is reduced, mills can generally reduce the electrical energy used for
aeration of the biological treatment systems. This reduction in energy consumption in effluent
treatment is included in the total energy impacts for Option A and Option B shown in Figure 11-
1 and Table 11-4.
Where a substantial reduction in effluent flow is realized by the pollution
prevention measures in the mill, minor modifications to the effluent treatment systems may be
required so that the mill could take advantage of the energy savings mentioned above. These
modifications might involve baffles to direct flow of effluent in an ASB, or bypassing part of
parallel sets of equipment. See Section 10 for further discussion.
11.4.2.3 Equivalence of Various Forms of Energy
EPA calculated an "oil equivalent" to conveniently present the combined effects
of the changes in thermal energy and electric power. The oil equivalent is based on the
assumption that all nuclear, hydro-electric, waste fuel, natural gas, coal, co-generation, and wind
power systems across the country are operated at their maximum capacity, and that any increase
or decrease in fuel electric power demand caused by the effluent guidelines regulations is
supplied by conventional condensing-type oil fired power stations. (If EPA assumed that
11-12
-------
Section 11 - Non-Water Quality Environmental Impacts
additional electrical demand would be supplied by coal or natural gas burning facilities, then the
predicted effect on fossil fuel consumption would be quite similar. It is expressed in terms of oil
equivalents here for convenience of the reader. Coal equivalents could also reasonably be used.)
For example, a mill burning all its black liquor and hog fuel would normally also
burn some purchased fossil fuel (oil, coal, or natural gas) to generate additional steam not
produced by the recovery boiler and power boiler. All the black liquor must be burned, but the
mill cannot normally increase the quantity of black liquor generated, since it is directly related to
the pulp production rate. The hog fuel is relatively inexpensive, so all available material will be
burned at all times, subject to any limitations in wood burning equipment. Any change in the
requirement for process steam will be supplied by changing the quantity of fossil fuel purchased
and burned.
Many mills also generate some or all of the electric power they require by passing
steam through turbines prior to its use as process heat. This power (known as co-generated
power) is relatively inexpensive, so mills normally operate their co-generation equipment to its
maximum potential. Some generate more power than is required on site, and sell the surplus to
the local utility or other customer. Whether the mill is a net buyer or seller of power, any change
in on-site power demand will be passed on to the national electrical power grid, reflecting
ultimately in the load on utility stations.
The overall efficiency of conversion of thermal energy in fossil fuels to electricity
delivered to consumers is approximately 25 percent, because thermal power stations ultimately
reject approximately two-thirds of the thermal energy derived from combusted fuel due to the
thermodynamic properties of steam. Energy losses to the stack gas and mechanical and electrical
losses occur in the turbines, generators, and distribution system. In addition, a small fraction of
the power generated is used in the utility plant itself for motors, electrostatic precipitators, and
other necessary auxiliary equipment.
To convert the steam demand calculated as tons per day to equivalent barrels of
oil, EPA made the following assumptions. EPA assumed a steam plant operating efficiency of
75 percent, a useful enthalpy of one ton of process steam at a typical mill as 2.7 GJ, and a heat
content of 1 barrel of oil equal to 6 GJ. The exact values vary up to several percent from those
values assumed from mill to mill, but such variations are minor since the actual change in energy
consumption which would result from implementation of the effluent guidelines is only a few
percent total, as shown in Table 11-4.
11.4.2.4 Changes in Energy Consumption Relative to Industry Total
In order to determine whether the estimated energy requirements of Option A and
Option B pose unacceptable impacts, EPA compared them to the total energy consumption of the
bleached papergrade kraft and soda subcategory. EPA estimated the total oil equivalent by
adding the purchased electricity and fossil fuels reported in AF&PA's 1995 annual report (17)
with EPA' s estimate of off-site power consumption for chemical manufacturing (DCN 14510).
11-13
-------
Section 11 - Non-Water Quality Environmental Impacts
For the bleached papergrade kraft and soda subcategory, EPA estimated the total oil equivalent
energy consumption to be 117 million bbl/year. The fractional change in this total energy
consumption for Options A and B are shown in Table 11-4. Option A represents a 1 percent
increase, while Option B would result in a 1 percent reduction in subcategory energy
consumption.
11.5 Atmospheric Emissions
Sections 304(b) and 306 of the Clean Water Act specifically direct EPA to
consider the air pollution impacts of effluent limitations guidelines and standards it establishes.
EPA estimated the impacts of BAT, PSES, and BMPs on the generation and emission of air
pollutants by the 86 mills with production in the bleached papergrade kraft and soda subcategory.
As detailed in this section, EPA analyzed the air emissions impacts of Option A and Option B.
These options will affect atmospheric emissions in a number of ways, as follows:
Control technologies that form the basis of Option A and Option B involve
changes in processes used to produce bleached pulp. As discussed in
Section 11.5.1, air emissions decrease for some air pollutants and remain
unchanged for others.
Mills will be burning material in the recovery boiler that was previously
discharged with the effluent because of the substantial improvements in
overall mill closure discussed in Section 11.3. This practice will tend to
marginally increase emissions of many substances to the atmosphere by up
to one to two percent, as discussed in detail in Section 11.5.2.
The location of points of emissions of carbon dioxide from mill sites will
change, as discussed below, but the total emission will not.
The changes in overall energy consumption discussed in Section 11.4 will
change atmospheric emissions from on-site and off-site energy production
facilities (increase for Option A and decrease for Option B).
A localized increase in emissions of carbon monoxide will occur due to
increased chlorine dioxide substitution.
11.5.1 Emissions Due to Mill Process Changes
The control technologies that form the basis of the effluent limitations guidelines
and standards involve changes in the processes used to produce bleached kraft pulp. These
changes affect the rate at which air pollutants, including HAPs, are emitted from pulping and
bleaching processes. As shown in Table 11-6, the process changes at bleached papergrade kraft
facilities subject to BAT, PSES, and BMPs decrease the emissions of some HAPs but have little
impact on others. For example, the elimination of chlorine and hypochlorite from bleaching
11-14
-------
Section 11 - Non-Water Quality Environmental Impacts
processes as part of the basis for BAT and PSES will reduce the emission of chloroform in the
bleached papergrade kraft subcategory by 64 percent but will have little impact on the emission
of methanol. The application of BAT, PSES, and BMPs for the bleached papergrade kraft and
soda subcategory will reduce the emission of total HAPs from 149,000 Mg/year to 139,000
Mg/year (7 percent reduction). The application of BAT, PSES, and BMPs plus MACT I, II, and
III for the bleached papergrade kraft and soda subcategory will reduce total HAP emissions from
149,000 Mg/year to 59,200 Mg/year (60 percent reduction).
11.5.2 Emissions Due to Burning Increased Quantities of Black Liquor
Option A or Option B, combined with BMPs, will result in recovery and burning
of increased quantities of black liquor, as discussed in Section 11.3.2.2. EPA calculated the
changes in quantities of black liquor generated for each mill for Options A and B (16). EPA
estimated the change in atmospheric emissions by applying emission factors (18,19,20,21)
developed in support of EPA' s MACT IINESHAP to these changes in black liquor firing rates.
These estimates, before and after MACT II is applied, are presented in Tables 11-7 and 11-8.
Emissions after MACT II controls are implemented are only presented in Table 11-8.
The underlying assumption for calculation of the marginal air emission increases
presented in Table 11-7 and Table 11-8 is that the emissions from a recovery boiler are
proportional to the quantity of fuel fired. EPA believes that this assumption will generally lead
to an overestimate of the actual emissions for the reasons discussed below.
Depending on the current status of a mill, three alternative scenarios3 exist:
Recovery Boiler Operating Below Capacity
If the recovery boiler is operating below its maximum capacity, then the
introduction of additional black liquor will raise the bed temperature, and
the associated increase in feed of combustion air will increase turbulence.
As discussed by many authors, increasing boiler load will normally reduce
emissions of organic pollutants, provided the proper combustion
conditions are maintained.
Particulate emissions would perhaps increase due to the increased gas flow
through the precipitator, but increasing the bed temperature in a recovery
boiler improves retention of sodium in the bed, thus reducing particulate
formation. SO2 emissions would drop because raising the bed temperature
reduces sulfur emission. One characteristic of recovery boiler combustion
conditions is that when SO2 emissions drop, HC1 emissions also drop.
Many investigations have shown that SO2 and HC1 emissions drop to
3Discussed in detail in the Analysis of Impacts of BAT Options on the Kraft Recovery Cycle (1997) (2) and the BAT
Cost Model Support Document (1996) (16).
11-15
-------
Section 11 - Non-Water Quality Environmental Impacts
essentially zero in boilers with hot smelt beds and adequate combustion
conditions. While the increased liquor firing rates that result from
implementation of Option A and Option B with BMPs, will not
necessarily cause sufficient rise in bed temperature to eliminate the SO2
and HC1 emissions, increased rates will present mills with the opportunity
to modify recovery boiler operation to reduce emissions. Thus, a simple
assumption that emissions are proportional to firing rate results in an
apparent overestimate of emissions from boilers currently operating below
capacity.
Boiler Operating at Full Capacity: Mill Chooses to Upgrade the
Boiler to Accommodate the Additional Black Liquor
Recovery boilers are typically upgraded by air system modifications and
firing higher concentration liquor. Both result in raising the furnace
temperature while upgraded air systems also improve turbulence. The
results are the same likely reductions in air emissions discussed above
(22).
Boiler Operating at Full Capacity: Mill Chooses to Oxidize the Black
Liquor to Reduce Its Heating Value
In this case, the solids feed to the boiler will rise, but the heat input will
not since the heating value of the liquor is reduced by oxygen black liquor
oxidation. The demand for combustion air will drop (due to less organic
feed and since the products of oxidation introduce oxygen to the stream),
so the stack gas flow will drop. Because gas flow through the boiler is the
key to emissions of several pollutants, emissions will not likely rise in
proportion to the fuel feed. Black liquor oxidation can accommodate a
thermal load increase of up to 5 percent. Therefore, if a milF s thermal
load increases by less than 5 percent, black liquor oxidation can reduce
total thermal load below the milF s baseline thermal load.
In all cases, TRS emissions are likely to be reduced by the above-mentioned
increases in temperature and upgrades to the boiler. In addition, sulfur dioxide emission increase
estimates are likely overstated because they do not account for the fact that some mills in
sensitive areas for sulfur dioxides already have sulfur dioxide controls in place or may choose
alternative controls available in the final MACT rule that mitigate these increases.
The discharge of carbon dioxide from the recovery boiler stack will increase in all
three scenarios, but will be balanced by a corresponding reduction in emissions from the effluent
treatment system and receiving water, as discussed in Section 11.5.3.
11-16
-------
Section 11 - Non-Water Quality Environmental Impacts
The increases in discharges of particulate HAPs due to changes in black liquor
firing are overshadowed by the effects of changes in quantities of oil fired in power boilers, as
shown in Table 11-9.
11.5.3 Emissions Due to Changes in Energy Consumption
As discussed in Section 11.4 and summarized in Table 11-4, Option A and Option
B will have an effect on total energy consumption. For the analysis presented here, EPA
estimated changes in on-site steam demand, on-site electric power consumption, and off-site
electric power consumption for each mill individually. On-site steam demand is met by power
boilers that burn black liquor, wood, coal, or oil and by recovery boilers burning black liquor.
Electrical demand is typically met by off-site electric power generating stations that burn coal,
oil, natural gas, or use nuclear or hydro energy. For the purpose of this analysis, EPA calculated
an oil equivalent to combine the effects of all energy changes.
As discussed in Section 11.4 and summarized in Table 11-4, both Option A and
Option B, in combination with BMPs, result in a net increase in combustion of black liquor
solids and a corresponding fuel benefit. The decrease in steam demand from the fossil fuel fired
boilers will result in less combustion in on-site power boilers and lower emissions from those
sources that offset the increased emissions from the recovery boilers discussed in Section 11.5.2.
Option A results in a net increase in off-site electric power consumption and a net decrease in on-
site power consumption. The on-site electrical savings is further decreased by the decrease in
steam demand so that, as shown in Table 11-4, the mill realizes a net energy savings. However,
the energy demand associated with Option A results in a net global increase.
Combustion of oil causes emissions of SO2, carbon dioxide, and trace quantities
of various metals (paniculate HAPs). Paniculate HAPs are also associated with combustion of
black liquor, as discussed in Section 11.5.2. The changes in air emissions due to estimated
changes in energy consumption are shown in Table 11-9. Table 11-9 also presents total
emissions for SO2 and paniculate HAPs resulting from BLS combustion (shown in Table 11-7)
plus oil combustion. Changes in carbon dioxide emissions are discussed in Section 11.5.4.
11.5.4 Greenhouse Gases
The earth radiates long-wavelength radiation that is absorbed by water vapor and
carbon dioxide (CO2) in the atmosphere near the earth' s surface. Because both water vapor and
CO2 are transparent to the incoming, warming, solar radiation but absorb the long-wave radiation
from the earth' s surface, the net effect of increases in atmospheric CO2and water vapor is a
warming of the earth" s atmosphere. This effect has been termed the "greenhouse effect" and
CO2 and water vapor have been termed "greenhouse gasses." Anthropogenic generation of water
vapor is minuscule relative to atmospheric recycling and is normally ignored in greenhouse gas
analysis. Therefore, water vapor is ignored here.
11-17
-------
Section 11 - Non-Water Quality Environmental Impacts
CO2 is an ultimate product of all combustion processes, including the combustion
of fossil fuels to generate electricity and the combustion of wood and black liquor at a pulp mill.
CO2 is also the ultimate product of the biodegradation of water-borne organic wastes generated
by a pulp mill. This biodegradation occurs in the biological wastewater treatment system, with
the ultimate disposal of sludge, and in the receiving stream.
The generation of CO2 attributable to the production of bleached pulp equals the
carbon taken into the mill with the wood and other raw materials, less the carbon that leaves the
mill as product plus the CO2 generated during the production of energy needed to produce the
product. Thus, to minimize the generation of greenhouse gasses, yield of product from the wood
must be maximized and energy use minimized. As discussed in Section 11.4, production of
bleached pulp consumes energy not just at the pulp mill, but also during the production of
bleaching chemicals, with chlorine dioxide requiring the most energy.
EPA examined the effect of Options A and B combined with BMPs on the
generation of CO2 by considering the overall mill carbon balance and the energy balance. As
discussed below, EPA concluded that neither option would have an impact on the total emission
of greenhouse gasses from mills due to pulp processing. However, the changes in energy
consumption will have the effect of increasing CO2 emissions for Option A while they will be
reduced for Option B. EPA concludes that the increased CO2 emissions attributable to Option A
pose no unacceptable non-water quality environmental impact.
11.5.4.1 Mill Carbon Balance
In this context, the mill carbon balance includes all pulping, bleaching, recovery
cycle, effluent treatment, and residual effects on the final receiving water. The effects of fuel
burned for energy production are discussed separately below.
11-18
-------
Section 11 - Non-Water Quality Environmental Impacts
All carbon that enters the mill as a component of raw material, wood, or
purchased chemicals leaves by one of the following paths:
As part of the product
Will not be affected by either option.
As CO2 from combustion of black liquor or
recalcination of lime in the recovery cycle
Will be slightly increased by both options but offset by
reduction in organic load in mill effluents and in energy
demand on power boilers.
As organic material in mill effluents
Oxidized to CO2 in biological treatment systems and by
subsequent biological action in the receiving waters. Some
of the material may be recovered as sludge from waste
treatment plants, and ultimately converted to CO2 by
incineration or degradation in landfills.
As organic gases from digester
Will be in the order of 1 kg/kkg pulp. MACTI combustion
control devices will reduce these emissions by
approximately 98 percent. The rest will oxidize to CO2 in
time in the atmosphere.
Thus, nearly all the carbon entering the mill eventually reaches the atmosphere as
carbon dioxide, except for the carbon component of the product.
The quantity of carbon entering the mill will not be modified significantly by
implementation of BMPs and Option A or Option B, because pulp yield is not affected by these
technologies (see Section 11.2.2). However, as discussed in Section 11.2, mills may experience
decreased wood consumption up to 0.3 percent for both options. Minor changes in carbon
entering the mill will include increased use of methanol as a reductant for manufacture of
chlorine dioxide. The recausticizing cycle will have to process additional quantities of lime mud
(calcium carbonate) because more black liquor will have to be processed as a result of the
implementation of BAT. The additional carbon dioxide released in the calcination reaction in the
lime kiln has its origin in the wood used in the digester, and is balanced by a reduction in carbon
dioxide released by biological oxidation of the milF s waste waters.
To put these assumptions in perspective, the quantity of carbon entering the mill
with the wood should be compared with the above-mentioned minor sources.
The worst case for increased carbon input with methanol would be conversion of
a softwood mill using no chlorine dioxide in the first bleaching stage to 100 percent substitution,
and installing one of the methanol reduction processes to produce chlorine dioxide. Total
chlorine dioxide consumption would be approximately 40 kg ClO2/kkg pulp which would require
a feed of 150 kg methanol per ton C1O2 (equivalent to 6 kg/kkg pulp). The methanol would be
discharged to the effluent as an organic by-product of C1O2 manufacture and would be oxidized
11-19
-------
Section 11 - Non-Water Quality Environmental Impacts
to carbon dioxide in the biological treatment system, producing approximately 8 kg carbon
dioxide/kkg pulp. The end result would be under 0.3 percent of the total mill emission. The
incremental emission of CO2 would be less for mills that currently operate methanol reduction
C1O2 plants and for mills already using high C1O2 substitution. If a mill uses extended cooking
or oxygen delignification then the amount would be cut further because of reduced chlorine
dioxide consumption.
11.5.4.2 Effect of Energy Production and Carbon Sequestration on Carbon Dioxide
Emissions
Fossil Fuel Consumption - As discussed in Section 11.4 and summarized in
Table 11-4, both Option A and Option B will have an effect on total energy consumption that can
be represented as a quantity of oil burned for electric power generation. The carbon content of
fuel oil is typically in the range of 83 percent to 88 percent. EPA assumed that the average
carbon content of fuel oil burned is 85 percent, and calculated the effect of changes in energy
consumption on carbon dioxide emissions. These changes, presented in Table 11-10, would
occur primarily at electric utility stations remote from the pulp mills.
Assuming 85 percent carbon, and a typical heating value of 42 MJ/kg, fuel oil
contains approximately 20 grams carbon per MJ. In the case of a boiler firing coal, the fuel
would typically contain approximately 60 percent carbon, and have a heating value of
approximately 28 MJ/kg. The carbon content of the coal is therefore approximately 21 g/MJ.
Therefore, the carbon dioxide generation for production of a given amount of electricity varies
little, regardless of the fuel used.
Carbon Sequestration - Mills may reduce wood consumption by up to 0.3
percent for both options, as discussed in Section 11.2. Approximately 2,500 kg wood is
consumed to manufacture 1,000 kg of fully bleached kraft pulp (oven dry basis). Wood contains
approximately 50 percent carbon, so that about 1,250 kg carbon are fed to a mill per ton product.
Some of this carbon is incorporated into the product, while the remainder as carbon dioxide
(approximately 760 kg carbon/kkg pulp (61 percent) or 2,800 kg ClO^kkg pulp) is emitted to the
atmosphere by the pulping and bleaching process, including stack emissions (the majority) and
biodegradation of effluents.
If a mill reduces wood consumption by 0.3 percent, carbon input to the mill is
reduced by 3.75 kg carbon per ton product (13.75 kg CO2 per ton product). With a yearly
bleached papergrade kraft production of 29.2 million tons per year, 401,500 tons CO2 per year is
sequestered by decreased wood use. Sixty-one percent of the sequestered CO2 would have been
emitted to the atmosphere. Therefore, the net reduction in CO2 emissions is 245,000 tons CO2
per year.
Carbon sequestration lessens the impact of fossil fuel consumption on CO2
emissions for Option A and further reduces CO2 emissions for Option B. EPA concludes the
increased CO2 emissions pose no unacceptable environmental impacts.
11-20
-------
Section 11 - Non-Water Quality Environmental Impacts
11.5.5 Carbon Monoxide Emissions from Oxygen Delignification
Someshwar (1997) presented data on numerous tests of emissions from four full-
scale oxygen delignification systems showing that they generated from 25 to 200 grams CO per
ton pulp processed (23). The quantity of CO generated correlated loosely with the oxygen
charge.
The proposed air emission control regulations under MACT I require that the
vents from oxygen delignification system be incinerated, so that the rate of emission to the
atmosphere will be substantially below the rate of generation of CO. Because such an incinerator
would be burning various organic gases including methanol, the CO emission to the atmosphere
will depend on the design and operation of the incinerator rather than the rate of formation of CO
in the oxygen delignification system.
MACT I requirements will ensure efficient oxidation of CO from this source.
Keeley (1997) suggests that at least 95 percent conversion of CO to carbon dioxide would be
attained (24). Consequently, EPA does not consider that the exact rate of CO emission from the
oxygen delignification reactors is important.
11.5.6 Carbon Monoxide from Chlorine Dioxide Bleaching
Downe (1996) expressed concerns that the increase in emissions of carbon
monoxide (CO) associated with increases in use of chlorine dioxide (as would be encouraged by
guidelines based on either option) would create difficulties for mills requesting permits under air
emission control regulations (25).
11.5.6.1 Information Available Prior to 1996
Traditionally, carbon monoxide emissions from bleaching have not been
considered significant relative to combustion sources in a pulp mill. However, Van der Merwe
(1980) reported that bleaching with chlorine dioxide generated sufficient CO to kill a
maintenance worker inside a bleaching vessel (26). Van der Merwe's supposition was
subsequently discussed and confirmed by other authors in the context of being an occupational
safety issue. As a common practice today, mills test vessels associated with chlorine dioxide
bleaching for carbon monoxide prior to entry for maintenance and inspection.
Van der Merwe concluded:
Chlorine bleaching of unbleached softwood did not generate CO;
Sodium hypochlorite bleaching of softwood generated trace amounts of
CO4;
"Elimination of hypochlorite bleaching is a process technology component of both Option A and Option B.
11-21
-------
Section 11 - Non-Water Quality Environmental Impacts
Chlorine dioxide delignification5 of oxygen delignified hardwood and
softwood pulp generated a concentration of CO of up to 3.2 percent in
head space of a laboratory reactor; and
Chlorine dioxide delignification of softwood pulp that had not been
previously delignified with oxygen generated much higher levels of CO
than oxygen delignified pulp.
Van der Merwe also found that CO was generated when unbleached pulp (kappa
32) was bleached with hydrogen peroxide. However, unbleached pulp is never processed with
hydrogen peroxide on an industrial scale unless the kappa number is first reduced to well below
20.
In independent work associated with investigations of bleach process yields,
Kutney (1983) showed that CO is formed in bleaching, and that the formation is approximately
proportional to the extent to which chlorine dioxide is substituted for chlorine (27). He
confirmed the trends noted by Van der Merwe.
11.5.6.2 Recent NCASI Study of CO Emissions from Bleaching
Someshwar (1997) reviewed the literature on emissions for CO from chlorine
dioxide bleaching (23). He also reported on measurements of CO emissions from six full-scale,
operating bleach plants carried out by NCASI using Continuous Emission Monitoring (CEM)
equipment and reported on various measurements for other operating mills by third parties.
The measurements reported by Someshwar include mass flows of CO per unit
pulp production, whereas most other authors report only concentrations of CO generated.
Someshwar concluded:
The major contributors to CO emissions from the bleach plants were the
chlorine dioxide bleach towers;
Emissions of CO from bleach plants were from 90 to 750 g/kkg pulp
processed, with an overall average of 390 g/kkg;
Emissions of CO are roughly proportional to the C1O2 charge applied to
the entire bleach plant; and
Contributions to total CO emission from the bleach plant by hydrogen
peroxide and extraction stages were small.
5The term "chlorine dioxide delignification" is often used to refer to the first chlorine dioxide stage of a modern
bleach plant.
11-22
-------
Section 11 - Non-Water Quality Environmental Impacts
The data reported by Someshwar are summarized in Table 11-11. EPA analyzed
these data to determine which of the known variables in the bleaching process correlate best with
the measured emissions of CO. Correlation with unbleached kappa number, wood type
(softwood or hardwood), and level of chlorine dioxide substitution were examined. EPA
attempted to discern a relationship between CO emissions and chlorine dioxide charge. The two
sets of data which form a coordinated series of experiments (mill AA and mill SG) were
examined separately and found to exhibit poor linear regression coefficients relative to
equation (1). The best linear regression equation found for CO emission rate vs. total chlorine
dioxide charge is:
CO emission g/kkg pulp = 7,780 x %C1O2 + 220 (1)
Equation (1) is shown in Figure 11-2. An alternative regression line which is
forced to pass through the origin is also shown in Figure 11-2.
CO emission g/kkg pulp = 1,637 x %C1O2 (2)
The regression coefficient (0.146) is clearly poor. Although a definite trend
toward increasing CO emissions with increasing use of chlorine dioxide exists, the relationship is
loose, and not necessarily linear.
Increased use of chlorine dioxide results from mills electing to operate with high
substitution rates (an element of both Option A and Option B) and by bleaching high kappa pulp.
One could therefore expect mills using oxygen delignification to emit less CO than similar mills
without oxygen delignification, as suggested by Van der Merwe (1980) (26). Someshwar does
not indicate which mills in Table 11-11 use oxygen delignification (23). EPA assumed that mills
showing unbleached kappa numbers which corresponded to the use of oxygen delignification or
extended cooking are operating one (or both) of these processes.
EPA assumptions for individual data sets are shown in Table 11-11. On this
basis, EPA calculated average carbon monoxide emissions for softwood mills with and without
oxygen delignification as follows:
Mills with traditional cooking
Mills with oxygen delignification or extended cooking
439 g/kkg pulp
3 52 g/kkg pulp
The few data on hardwood bleaching showed no significant difference between
mills with or without oxygen delignification.
11-23
-------
Section 11 - Non-Water Quality Environmental Impacts
11.5.6.3 Effect of BAT on Total Emission of CO
To assist in estimating the costs for Options A and B, EPA calculated the chlorine
dioxide consumption for each mill. The total is shown in Table 11-12.
Based on the equations presented in Figure 11-2 and the 1995 bleached kraft
subcategory production, the total estimated emissions of carbon monoxide from bleaching for
1995 baseline conditions, Option A, and Option B were calculated as shown in Table 11-12. The
values shown in Table 11-12 for CO emissions per ton pulp differ from the values above because
they refer to the total of hardwood and softwood pulp.
11.5.6.4 Carbon Monoxide Emissions from Bleach Plants vs. Total Mill Emissions
The total CO emissions from combustion of pulping liquors in bleached,
unbleached, sulfite, and semi-chemical pulping mills is 274,000 tons of carbon monoxide per
year. Approximately half the total production of chemical/semi-chemical pulp is bleached
papergrade kraft.
Assuming that approximately half of the above-mentioned carbon monoxide
emissions are from bleached papergrade kraft, the bleach plant emissions for Option A would
represent approximately 9 percent of the total carbon monoxide emission from the liquor burning
sources in the bleached papergrade kraft subcategory.
EPA concluded that implementation of Option A would increase emissions of CO
from bleach plants by approximately 50 g/kkg pulp (about 30 kkg/yr for a typical mill), while
Option B would cause an increase of approximately 7 g/kkg pulp. The emission increase due to
Option A is approximately 10 percent of the average emission of CO from combustion sources in
a mill. Emission control technology to reduce CO from boilers is well known and available. So,
wherever "Prevention of Significant Deterioration" concerns exist, it would be feasible to reduce
CO emissions from combustion sources to counter the increase in emissions from the bleach
plant.
11.6 Impacts of New Source Performance Standards and Pretreatment Standards
for New Sources (NSPS and PSNS) on the Bleached Papergrade Kraft and
Soda Subcategory
EPA analyzed the projected non-water quality environmental impacts of Option A
for the bleached papergrade kraft and soda subcategory for BAT, PSES, and BMPs based on
complete substitution of chlorine dioxide for chlorine and other technology elements (see Section
8.2.1.1 for Option A technology description) in Sections 11.1 through 11.5 above. This section
presents the non-water quality environmental impacts of a second technology configuration
(NSPS and PSNS) which is equivalent to BAT, PSES, and BMPs with the addition of extended
delignification (oxygen delignification or extended cooking) on a new 1000 kkg/d bleached
papergrade kraft fiber line (See Section 8.2.1.2 for Option B technology description).
11-24
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-13 presents the non-water quality environmental impacts of the selected
technology basis for NSPS and PSNS, compared to conventional pulping and bleaching
technology. These estimates are based on the same calculation methodology described under
BAT and PSES, applied to a 1000 tpd model mill. Based on these estimates, EPA concludes that
the process technologies that form the basis for NSPS and PSNS for the bleached papergrade
kraft and soda subcategory pose no unacceptable non-water quality environmental impacts.
11.7 Impacts of Totally Chlorine Free (TCF) Technology on the Bleached
Papergrade Kraft and Soda Subcategory
EPA analyzed the projected non-water quality environmental impacts of Option A
for the bleached papergrade kraft and soda subcategory for BAT, PSES, and BMPs based on
complete substitution of chlorine dioxide for chlorine and other technology elements (see Section
8.2.1.1 for Option A technology description) in Sections 11.1 through 11.5. This section presents
the non-water quality environmental impacts of two TCF options, TCF-Peroxide and TCF-Ozone
described in Section 8 of this document.
Table 11-14 presents the non-water quality environmental impacts of the two TCF
options, compared to conventional pulping and ECF bleaching technology (Option A). These
estimates are based on the same calculational methodology described under BAT. Based on these
estimates, EPA concludes that TCF process technologies for the bleached papergrade kraft and
soda subcategory pose no unacceptable non-water quality environmental impacts.
11.8 Impacts of BAT. PSES. and BMPs on the Papergrade Sulfite Subcategory
EPA analyzed the non-water quality impacts that result from implementing BAT,
PSES, and BMPs at the 11 papergrade sulfite mills. Because the number of mills in this
subcategory is significantly fewer and the size of these mills is generally smaller than the bleached
papergrade kraft and soda subcategory, the non-water quality impacts are lesser in magnitude in
the papergrade sulfite subcategory compared to the bleached papergrade kraft and soda
subcategory.
11.8.1 Wood Consumption
EPA notes that the impacts of BAT, PSES, and BMPs results in up to 0.3 percent
decrease in wood consumption for the bleach papergrade kraft and soda subcategory.
Approximately two-thirds of this decreased demand, or 0.2 percent, can be attributed to reduced
fiber loss due to the implementation of BMPs. The remaining 0.1 percent of decreased demand
can be attributed to closed screening. EPA assumes that the demand for wood at papergrade
sulfite mills will decrease up to 0.2 percent due to the implementation of BMPs. Closed screening
is not a BAT, PSES, and BMPs technology element of papergrade sulfite subcategory and
therefore will not result in decreased wood consumption.
PULP97.TDD/17 November 1997 11-25
-------
Section 11 - Non-Water Quality Environmental Impacts
11.8.2 Solid Waste and Effluents
The effects of this rulemaking on solid waste and effluents are discussed below.
11.8.2.1 BOD5 and Sludge Generation
Implementation of BMPs will result in decreased solid waste generation from the
recovery and rerouting of an incremental amount of red liquor to other processes that was once
sent to wastewater treatment. Eight of the 11 mills can use a recovery furnace or incinerator to
combust the recovered red liquor which was previously sent to wastewater treatment. The other
three mills use other proprietary processes to recover chemicals and/or byproducts which are
assumed to be able to accommodate the incremental red liquor that was previously sent to
wastewater treatment. For estimation of solid waste impacts, EPA assumed that all 11 mills
would be able to reduce the total organic load sent to wastewater treatment resulting from
additional recovered red liquor by an amount equivalent to that achieved by mills employing a
recovery furnace or incinerator. This assumption is reasonable because the three mills that do not
use a recovery furnace or incinerator do employ various techniques to convert the organic load in
the red liquor to a usable byproduct.
EPA estimates BOD5 in untreated wastewater will decrease by 26,700 kg/d
through the implementation of BMPs for the papergrade sulfite subcategory. EPA calculated an
associated decrease in sludge generation of 2,470 kkg/yr for papergrade sulfite mills that use
activated sludge treatment (several mills use aerated stabilization basins which do not produce
sludge) based on the same calculation methodology described above for the bleached papergrade
kraft and soda subcategory. EPA is projecting no change in BOD5 and sludge generation as a
result of implementation of BAT and PSES in the papergrade sulfite subcategory.
Since proposal, the dioxin and furan content of sludge at papergrade sulfite
facilities has decreased significantly. As discussed in Section 11.3.2.4 for the bleached papergrade
kraft and soda subcategory, the control technologies that form the basis of the BAT limitations
and PSES promulgated today limit the concentration of dioxin and furan allowed to be
discharged. As a result, these limitations will also reduce the sludge loading of dioxin at
papergrade sulfite facilities by approximately 85 percent.
11.8.2.2 Effluent Flows
EPA assumed the reduction of effluent flow for papergrade sulfite mills will be
comparable to the reductions achieved by bleached papergrade kraft and soda mills from the
implementation of BMPs. EPA, therefore, estimates papergrade sulfite mills may achieve an
effluent flow reduction of approximately 1 m3/kkg pulp. EPA is projecting no additional change
in effluent flows as a result of implementation of BAT and PSES in the papergrade sulfite
subcategory.
PULP97.TDD/17 November 1997 11 -26
-------
Section 11 - Non-Water Quality Environmental Impacts
11.8.3 Energy Impacts
As a result of implementation of BAT, PSES, and BMPs, EPA estimates off-site
electrical energy consumption will decrease by 38,100 MMBTU/yr, primarily because of reduced
bleaching chemical requirement. EPA estimates on-site electrical energy consumption will
decrease by 1,050,000 MMBTU/yr, primarily because of decreased demand for operating
wastewater treatment.
EPA estimates the change in on-site steam demand for the papergrade sulfite
subcategory to be 10,300 MMBTU/yr. EPA calculated the change in steam demand by
calculating the increase in steam demand required for the evaporation of the recovered red liquor
at all 11 mills minus the additional steam that could be generated from the combustion of red
liquor recovered at the seven mills that use recovery boilers. (Ingruber, Kocurek, and Wong
(1985) report a range of red liquor heating values that average 14.9 MJ/kg. This value was used
to estimate the amount of steam generated (28)).
The total impact of BAT, PSES, and BMPs for the papergrade sulfite subcategory
decreases energy consumption by 1,080,000 MMBTU/yr.
11.8.4 Atmospheric Emissions
EPA estimates emissions of HAPs from papergrade sulfite mills will increase by
2.6 percent as a result of implementation of BAT, PSES, and BMPs. This increase in HAPs is
projected to occur from implementation of BMPs that will capture of additional red liquor that
was previously discharged, and burning of that red liquor in a chemical recovery boiler or
incineration device at 8 of the 11 mills. The other three mills do not use a combustion-based
process unit and therefore the capture of additional red liquor at these mills is assumed to not
result in increased HAP emissions. The increased air emissions from the recovery of additional
pulping liquor are based on the same calculation methodology described for the bleached
papergrade kraft and soda subcategory with emission factors changed to reflect sulfite operation.
The estimated emission does not represent a significant increase.
11.9 Impacts of New Source Performance Standards and Pretreatment Standards
for New Sources (NSPS and PSNS) for the Papergrade Sulfite Subcategory
NSPS and PSNS that EPA is promulgating for the papergrade sulfite subcategory
are equivalent to BAT and PSES, respectively. Therefore, EPA projects no non-water quality
environmental difference between NSPS/PSNS technology and BAT technology.
11.10 References
A number of EPA documents involved in the development of the effluent
guidelines are referred to herein. They are available in the public docket, except for portions
PULP97.TDD/17 November 1997 11 -27
-------
Section 11 - Non-Water Quality Environmental Impacts
claimed as Confidential Business Information. This list includes documents prepared by the EPA
and its contractors, as well as comments submitted to EPA concerning the rulemaking.
All other references listed below can be found in the open literature of the pulp and
paper industry. Copies of the smaller documents are also in the public docket.
1. Technical Support Document for the Voluntary Advanced Technology Incentives
Program. EPA, Washington DC, Record Section 22.8, DCN 14888, 1997.
2. Analysis of Impacts of BAT Options on the Kraft Recovery Cycle (Recovery
Impacts Document). Reported prepared by ERG and N. McCubbin for EPA.
Record Section 23.1.2, DCN 14490, 1997.
3. Effect of Oxygen Delignification on Yield of the Bleached Kraft Pulp
Manufacturing Process. EPA, Washington DC, Record Section 23.1.2, DCN
14491.
4. Mannisto, H, E. Mannisto, and M Krogerus. Proceedings of Minimum Effluent
Mils Symposium. TAPPI Press, Atlanta 1996.
5. Bryant, Patrick, E.W. Malcolm, and C.P. Woitkovich. "Pulp and Paper Mill Water
Use in North America." Presented at: TAPPI International Environmental
Conference. 1996.
6. BAT Baseline Database. Data on mill characteristics and operations collected by
EPA through the 1990 Census Questionnaire, subsequent contacts with mills by
phone, facsimile machine, and site visits.
7. Springer, Alan. Industrial Environmental Control. Pulp and Paper. John Wiley
and Sons, New York, 1986.
8. Jaegel, A., "Multimedia Environmental Performance of TCP Closed Bleach Plant
Kraft Pulp Production." In: Proceedings of Non-Chlorine Bleaching Conference.
Orlando, Florida, March 1996.
9. Nutt, W. E., S.W. Eachus, B.F. Griggs, and M.A. Pikulin. "Development of an
Ozone Bleaching Process" In: Proceedings of TAPPI Pulping Conference.
Boston, MA, 1992.
10. NCASI, "Solid Waste Management and Disposal Practices in the US Paper
Industry." Technical Bulletin 641. National Council for Air and Stream
Improvement. 1992.
PULP97.TDD/17 November 1997 11-28
-------
Section 11 - Non-Water Quality Environmental Impacts
11. NCAST. An Analysis of the Relative Performance of POTW and Paper Industry
Wastewater Treatment Systems on Conventional and Non-Conventional
Pollutants. DCN20026A31. April 1994.
12. Kocurek, M. J. ed. Pulp and Paper Manufacture. Vol. 9 Mill Control. Quality.
Environmental. Corrosion. Electrical. TAPPI Press, Atlanta, 1992.
13. Lynde-Maas, Mary Kay, J.P. Unwin, and R.A. Miner. "Preliminary Results from
the NCASI 1995 Wastewater and Solid Waste Survey." In: 1997 Environmental
Conference and Exhibit. Book 1. Minneapolis, MN, May 5-7, 1997.
14. Assessment of Risks from Exposure. Terrestrial and Avian Wildlife, and Aquatic
Life to Dioxins and Furans from Disposal and Use of Sludge from Bleached Kraft
and Sulfite Pulp and Paper Mills. EPA 560\5-90-013. U.S. Environmental
Protection Agency, July 1990.
15. Tran, H. N, D. Barham, and D. W. Reeve. "Chloride and Potassium in the Kraft
Recovery Cycle." Pulp and Paper Canada. 91:5, May 1990.
16. BAT Cost Model Support Document. Report prepared by Radian Corporation for
US Environmental Protection Agency. Record Section 23.1.2, DCN 13953, 1996.
17. Hicks, J., and L. Wolfe. US Pulp and Paper Industry
Year 1995. Report prepared by AF&PA. Washington DC, 1996.
18. T. Holloway. "Presentation of the HAP Emissions DataBase." Midwest Research
Institute, June 12, 1995.
19. T. Holloway. "Summary of PM and HAP Metals Data." Prepared by Midwest
Research Institute for EPA, June 14, 1996.
20. R. Nicholson. "Addendum to Summary of Responses to the 1992 NCASI MACT
Survey." Prepared by Midwest Research Institute for EPA, May 29, 1996.
21. T. Holloway and R. Nicholson. "Nationwide Costs, Environmental Impacts and
Cost-Effectiveness of Regulatory Alternatives for Kraft, Soda, Sulfite, and
Semichemical Combustion Sources." Prepared by Midwest Research Institute for
EPA, October 9, 1996.
22. Adams, T. N., W. J. Frederick, T. W. Grace, M Hupa, K.Iisa, A. K. Jones, and H.
Tran. Kraft Recovery Boilers. TAPPI Press, Atlanta, Copyright by AF&PA,
ISBN 0-9625985-9-3, 1997.
PULP97.TDD/17 November 1997 11 -29
-------
Section 11 - Non-Water Quality Environmental Impacts
23. Someshwar, A. V., D. S. Dillard, A. K. Jain, and A. L. Caron. "Carbon Monoxide
Emissions from Oxygen Delignification and Chlorine Dioxide Bleaching of Wood
Pulp" In: Proceedings of TAPPI Environmental Conference. Minneapolis, MN,
1997.
24. Keeley, S. A., K. T. Hiltgen, and J. Orynawka. "Permitting Carbon Monoxide
Emissions from Bleaching Operations at Pulp and Paper Mills." In: Proceedings
of TAPPI Environmental Conference. Minneapolis, MN, May 1997.
25. Downe, S. "EPA Makes "Cluster" Modifications as Effluent Rule Promulgation
Nears." Pulp and Paper. December 1996, pp 53-58.
26. Van der Merwe, A. J. W., F. J. Viljoen, B. D. Thorn, and G. J. Lourens. "Carbon
Monoxide Generation During Chlorine Dioxide Bleaching." TAPPI Journal. 63:8,
August 1980.
27. Kutney, G. W., M Mallais, and J. R. du Manoir. "Understanding the Bleaching
Process: Pulp Yield Loss and Carbon Monoxide Generation in the CEDED
Sequence." Journal of Pulp and Paper Science. June 1983.
28. Ingruber, O.V., M.J. Kocurek, and A. Wong, eds. Pulp and Paper Manufacture
Volume 4: Sulfite Science and Technology. Joint Textbook Committee of the
Paper Industry, Atlanta, 1985.
29. T. Holloway. "Documentation of Methods Used to Estimate Nationwide
Environmental Impacts of Control Options for Kraft and Soda Combustion
Sources." Prepared by Midwest Research Institute for EPA, July 1, 1996.
30. Water Use Reduction in the Pulp and Paper Industry. Prepared by H.A. Simons
Ltd., NLK Consultants Inc. for and Sandwell Inc., for Canadian Pulp and Paper
Association, Montreal, Canada, 1994.
31. Rodden, G., Finnish Tour, Pulp and Paper Canada. 98:8, August 19, 1997.
11.11 Other References
The following is a list of references used in the development of Section 11 but not
directly cited:
Adams, Terry N., and W. James Frederick. Kraft Recovery Boiler Physical and Chemical
Processes. AF&PA, Washington, 1988.
Ahlenius, Lars. "Closing up the Bleach Plant - MoDo Experience" Presented at: TAPPI Pacific
Section Conference. Seattle, September 16-17, 1993.
PULP97.TDD/17 November 1997 11-30
-------
Section 11 - Non-Water Quality Environmental Impacts
Albert, R. J. "Technical and Economic Feasibility of the Effluent Free Bleached Kraft Mill"
Proceedings from International Non-Chlorine Bleaching Conference. Hilton Head, NC, 1993.
Bicknell, B., D.B. Spengel, and TJ. Holdsworth. Comparison of Pollutant Loadings from ECF.
TCF. and Ozone/Chlorine Dioxide Bleaching Process International Non-Chlorine Bleaching
Conference. Amelia Island, Florida, 1995.
Technical Support Document for Best Management Practices for Spent Pulping Liquor
Management. Spill Prevention, and Control. EPA, Washington DC, Record Section 30.9, DCN
14489, 1997.
Bodien, D. Report of Visit to Oy Metsa Rauma AB. Rauma. Finland. EPA, Seattle, Washington,
November 6, 1996.
Bodien, D. Report of Visit to SCA Graphics AB. Ostrand. Sweden. Finland. EPA, Seattle,
Washington. November 1996.
Caron, J. R., and L. D. Williams. "Design and Startup of the Bleach Filtrate Recycle Process"
Presented at: Proceedings from TAPPI Environmental Conference. Orlando, Florida. TAPPI
Press, 1996.
Dence, C. W., and D. W. Reeve, eds. Pulp Bleaching. Principles and Practice. TAPPI Press,
Atlanta, 1996.
Dunn, K. C., S. Stratton, and P. F. Earl. "Update on the Bleach Filtrate Recycle Process at
Champion, Spring 1996" In: Proceedings from PPI World Pulp Symposium. Brussels, 1996.
Development Document for Proposed Effluent Limitations Guidelines and Standards for the Pulp.
Paper and Paperboard Point Source Category. EPA-821-R-93-019, U.S. Environmental
Protection Agency, Washington DC, October 1993.
Methodology for Determining BAT Pollutant Loadings for the Revision of the Pulp and Paper
Effluent Limitations Guidelines. U.S. Environmental Protection Agency, Record Section 24.0,
DCN 13952, June 1996.
Estimates of BCT Cost. U.S. Environmental Protection Agency, Record Section 23.2.
Notice of Availability for The Pulp and Paper Industry Effluent Guidelines Portion of the "Cluster
Rules". Federal Register, July 15, 1996.
Homer, G., S. Johnson, and M. Epiney. "State-of-the-art ECF: Pulping and Bleaching with
Oxygen, Ozone and Chlorine Dioxide" In: Proceedings from TAPPI Pulping Conference.
Nashville, TN (TAPPI Press, Atlanta), 1996.
PULP97.TDD/17 November 1997 11-31
-------
Section 11 - Non-Water Quality Environmental Impacts
Jamieson, A.G., and G. Possum. "Influent of Oxygen Delignification on Pulp Yields" In:
Proceedings from 29th APPITA Conference. 1976.
Kocurek, M. J. ed. Pulp and Paper Manufacture. Vol. 5. Alkaline Pulping. TAPPI Press,
Atlanta, 1989.
Leader, J.P., Harry Lim, and Gary Byron. "Medium Consistency Oxygen Delignification in an
(CD) EoD Bleaching Process for Radiata Pine Pulp." Appita Vol 39, No. 6, November 1986.
Lindstrom, L-A., International Papermaker. December 1996, p. 14.
Paperi & Puu. Reinforced Multi-Stage Oxygen Delignification - Recent Experiences. Vol 9,
1994.
Parsad, B., J. Gratzl, A. Kirkmann, H. Jameel, T. Rost, and V. Magnotta. "High-Kappa Pulping
and Extended Oxygen Delignification Decrease Recovery Boiler Load." TAPPI Journal.
November 1994.
Pauksta, P. "Safari 96 Report on Mills in Finland." TAPPI Journal. September 1996.
PAPRICAN. "Safari 96 - A Report on Visits to Five Finnish Bleached Kraft Mills." Pulp and
Paper Canada. June 1996
Regulatory Impact Assessment for Land Application of Bleached Pulp and Paper Mill Wastewater
Treatment Sludges. Report prepared by ERG for EPA. April 1991.
Stratton, S. C., and G. E. Maples. "Overview of the BFR Process and Demonstration Project"
In: Proceedings of TAPPI Environmental Conference. Atlanta, GA, 1995.
Technical Support Document: Chemical Recovery Combustion Sources at Kraft and Soda Pulp
Mills. EPA-453/R-96-012, U.S. Environmental Protection Agency, Research Triangle Park, NC,
October 1996.
Vice, K. M., R. E. Sieber, and B. Bicknell. "Cost of Upgrading Bleach Plants to Minimize COD
Discharges." In: Proceedings of International Non-Chlorine Bleaching Conference. Amelia
Island, Florida, 1995.
PULP97.TDD/17 November 1997 11-32
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-1
Summary of Impacts of Options A and B Relative to Baseline"
for the Bleached Papergrade Kraft and Soda Subcategory
Option A"
Option B"
Reduction in Wood Consumption
Up to 0.3%
Marginal reduction from screen
room closure and spill recovery
Up to 0.3%
Marginal reduction from screen
room closure and spill recovery
Water Conservation
Typically 5 to 10% due to closed
screening, better spill collection
Typically 10 to 15% due to
closed screening, better spill
collection, other mill
modernization
Solid Waste Generation0
Primary sludge
Secondary sludge at mills with
activated sludge treatment
About 4% reduction due to fiber
recovery
2% reduction due to reduced BOD5
to treatment
About 4% reduction due to
fiber recovery
3% reduction due to reduced
BOD5 to treatment
Energy Consumption
(as change in bbls of oil)
Increase 1% due to energy for off-
site chemical manufacture, with
offsets in mill.
Decrease 1% due to
replacement of C12 and C1O2
with O2; reduction in WWTP
power
Air Emissions
From burning increased quantities of
black liquor
Total gaseous HAPs, kkg/yr
Total paniculate HAPs, kkg/yr
Carbon monoxide, kkg/yr
Sulfur dioxide, kkg/yr
Increase 220 (1.1%)
Increase 0.71 (1.1%)
Increase 1,440 (1.2%;
Increase 784 (1.2%)
Increase 308 (1.6%)
Increase 1.03 (1.6%)
Increase 2,120 (1.7%)
Increase 1,150 (1.7%)
From changes in energy
consumption
Carbon dioxide, kkg/yr
Sulfur dioxide, kkg/yr
Total particulate HAPs, kkg/yr
Increase 154,000
Increase 1,800
Increase 4.64
Decrease 1,650,000
Decrease 6,300
Decrease 16.3
From changes in chlorine dioxide
bleaching
Carbon monoxide, kkg/yr
Increase 1500
Increase 220
aBaseline is technology in place as of mid-1995.
bData presented are simplified. Refer to detailed discussion in body of report. Individual mills vary.
"Percentage reductions for solid waste are all shown relative to total solid waste discharge from same source in late
1980s.
11-33
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-2
Bleach Plant Effluent Flow for Bleached Papergrade Kraft and
Soda Mills With and Without Extended Delignification
Mills without oxygen delignification or extended cooking
Mills with oxygen delignification or extended cooking
Reduction in effluent flow
Hardwood3
24.7 nfVkkg
19.7 nrVkkg
SmVkkg
Softwood3
37.1 mVkkg
24.7 nrVkkg
12.4 nrVkkg
aDCN 13952, Section 24.
11-34
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-3
Black Liquor Solids, BOD5, and Sludge Generation for Option A and
Option B for the Bleached Papergrade Kraft and Soda Subcategory
Additional heat from black liquor
solids combustion15 (GJ/d)
Additional BLS to combustion
resulting from process changes0 (kkg/d)
Decreased BOD5 to treatment (kg/d)
Decreased sludge generation in ASTs
(kkg/yr)
Option A
29,600
2,160
649,000
36,000
% Change
from Baseline
Option A
NC
1.5
21
2
Option Ba
45,600
3,260
979,000
62,000
% Change
from Baseline
Option B
NC
2.2
31
3
aHeat in BLS from extended delignification assumed to have HHV of 14.5 MJ/kg.
bSee BAT Cost Model Support Document (4).
Increase over baseline estimated assuming 1,750 kg BLS/kkg pulp for baseline.
NC = Not calculated.
11-35
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-4
Effect of Options A and B on Energy Consumption Relative to 1995 Base Case
for the Bleached Papergrade Kraft and Soda Subcategory
Option A
Option B
Note
Electrical Power Consumption
Increase (decrease) in off-site power consumption
Increase (decrease) in on-site power consumption
Total increase (decrease) in power consumption
Total increase (decrease) in power consumption
Increase (decrease) in thermal energy to generate power
Increase (decrease) in thermal energy to generate power
Increase (decrease) in oil equivalent to generate power
Increase (decrease) in oil equivalent to generate power
MW
MW
MW
kWh/kkg
GJ/d
GJ/kkg
trillion
BTU/yr
bbl/yr
87
(21)
67
19
23,100
0.28
7.666
1,347,000
(161)
66
(95)
(27)
(32,800)
(0.39)
(10.88)
(1,911,000)
a
b
Steam Consumption
Net steam requirements (savings)
Fossil fuel requirement (savings) from steam demand
(generation)
kkg/d
bbl/yr
(2,410)
(506,200 )
1,790
376,400
c
Total Energy Consumption
Total increase (decrease) in fuel consumed
Total increase (decrease) in fuel consumed
Total increase (decrease) in fuel consumed
Increase (decrease) in fuel consumed relative to total
energy consumption by bleached kraft subcategory in
1995
trillion
BTU/yr
bbl/yr
number of
households
4.785
840,700
46,100
1%
(8.734)
(1,535,000)
(84,300)
(1%)
e
e
d
e
aOff-site power consumption is for manufacture of bleaching chemicals.
bEstimate of thermal energy required assumes overall efficiency of condensing power station and distribution system
of 25 percent.
"Conversion of fuel oil to useful steam assumes overall steam plant efficiency of 75 percent.
dAssumes 103.6 million Btu/household/yr (Energy Information Administration (DOE) 1993)
eSee DCN 14510 for baseline energy calculations of 116 million bbl/yr.
11-36
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-5
Process Changes Affecting Energy Consumption at
Bleached Papergrade Kraft and Soda Mills
Process modification
Improve brown stock washing and
screen room closure
Extended cooking
Oxygen delignification
High chlorine dioxide substitution
Best Management Practices
Evaporator upgrade
Evaporator load reduction
Measures to compensate for
increased load on recovery boiler:
Recovery boiler upgrade
Anthraquinone pulping
additive
Black liquor oxidation
Recausticizing upgrade
Reduction in effluent flow due to
above
Reduction in effluent BOD5 due to
above
Steam demand
Reduced demand for fossil fuel
corresponding to fuel value of
recovered black liquor.
Reduced demand from reduction in
water to evaporate.
Reduced demand for fossil fuel
corresponding to fuel value of
recovered black liquor.
Reduced demand for fossil fuel
corresponding to fuel value of
recovered black liquor.
Heat demand for oxygen reactor.
Minor increase
Reduced demand for fossil fuel
corresponding to fuel value of
recovered black liquor.
Steam demand to evaporate
recovered water.
Steam demand increase
Steam demand decrease
Steam generated from above
mentioned black liquor replaces
some steam from fossil fuel.
None
Reduction in net demand since
steam generated in reaction is used
for evaporator.
Insignificant
None
None
Electrical demand
Minor, may be plus or minus.
Insignificant in fiber line.
Net reduction in off-site power for
bleach chemical manufacture.
Additional mixing energy in
fiberline.
Net reduction in power for bleach
chemical manufacture.
Increased energy for pulp mixing.
Increased energy off site for bleach
chemical manufacture.
Insignificant
Insignificant
Insignificant
Minor change
None
Increase
Minor change
Minor reduction in pumping
energy.
Reduction in WWTP power.
11-37
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-6
Impact of BAT, PSES, and BMPs:
Bleached Papergrade Kraft and Soda Subcategory Air Emissions
Air Pollutants
Hazardous air pollutants (HAP)
Chloroform"
MethanoP
Volatile organic compounds
Total reduced sulfur
Bleached Papergrade Kraft BAT Baseline Emissions and Reductions
(Mg/yr)
Baseline
Emissions
149,000
9,510
96,400
569,000
100,000
Emission Reductions
from
BAT/PSES/BMPs
10,000
6,060
3,100
11,000
1,300
Emission Reductions
from BAT/PSES/BMPs
Plus MACT I, II, and III
89,800
6,240
66,080
301,200
59,800
"Baseline emission is a subset of baseline HAP emissions (149,000 Mg/yr).
11-38
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-7
Atmospheric Emission Changes Due to Burning Recovered Black Liquor,
Bleached Papergrade Kraft and Soda Subcategory Before MACT II is Applied
Pollutant
Total Reduced Sulfur
(TRS)
Criteria Pollutants
Carbon Monoxide
Paniculate Matter
Nitrogen Oxides
Sulfur Dioxides
Volatile Organic
Compounds
Emissions, Mg/yra
Before
MACT II
Baseline
2,650
123,700
31,370
36,120
67,770
19,500
Increase from
Option A
27.4
1,440
356
423
784
213
Increase from
Option B
36.0
2,120
514
623
1,150
295
Change from Baseline
Option A
1.03%
1.17%
1.14%
1.17%
1.16%
1.09%
Option B
1.36%
1.71%
1.64%
1.73%
1.69%
1.51%
Gaseous HAPs
Acetaldehyde
Benzene
Formaldehyde
Hydrochloric Acid
Methanol
Methyl Ethyl Ketone
Methyl Isobutyl Ketone
Phenol
Toluene
Xylenes
Total Gaseous HAPs
1,150
580
421
6,890
6,810
469
556
1,330
482
1,159
19.860
12.5
6.43
4.99
79.1
72.5
5.26
6.26
14.6
5.45
13.0
220
17.0
9.10
7.43
115
95.9
7.45
8.91
20.4
7.79
18.5
308
1.08%
1.11%
1.19%
1.15%
1.06%
1.12%
1.13%
1.09%
1.13%
1.12%
1.11%
1.48%
1.57%
1.76%
1.67%
1.41%
1.59%
1.60%
1.53%
1.62%
1.60%
1.55%
11-39
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-7 (Continued)
Pollutant
Emissions, Mg/yra
Before
MACTII
Baseline
Increase from
Option A
Increase from
Option B
Change from Baseline
Option A
Option B
Participate HAPs
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Cobalt
Lead
Manganese
Mercury
Nickel
Selenium
Total Paniculate HAPs
Total HAPsb
2.72
8.16
0
1.81
2.72
1.81
20.0
11.8
0.907
5.44
8.16
62.6
19,920
0.02
0.10
0.00
0.02
0.02
0.02
0.22
0.14
0.01
0.06
0.10
0.71
220
0.03
0.14
0.00
0.02
0.03
0.03
0.32
0.20
0.02
0.08
0.14
1.03
308
0.89%
1.17%
0.00%
0.86%
0.89%
1.14%
1.13%
1.16%
1.49%
1.05%
1.20%
1.14%
1.11%
1.29%
1.69%
0.00%
1.25%
1.29%
1.65%
1.62%
1.67%
2.15%
1.51%
1.73%
1.64%
1.55%
aAll nationwide baseline emissions estimates received from MRI (MRI July 1996, October 1996) (21,29)
bTotal HAPs include gaseous HAPs and paniculate HAPs.
11-40
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-8
Atmospheric Emission Changes Due to Burning Recovered Black Liquor,
Bleached Papergrade Kraft and Soda Subcategory After MACT II is Applied
Pollutant
Total Reduced Sulfur
(TRS)
Criteria Pollutants
Carbon Monoxide
Paniculate Matter
Nitrogen Oxides
Sulfur Dioxides
Volatile Organic
Compounds
Emissions, Mg/yra
After MACT
II Baseline
2,650
Increase from
Option A
27.4
Increase from
Option B
36.0
Change from Baseline
Option A
1.03%
Option B
1.36%
123,700
18,500
36,120
67,770
19,500
1,440
209
423
784
213
2,120
301
623
1,150
295
1.17%
1.13%
1.17%
1.16%
1.09%
1.71%
1.62%
1.73%
1.69%
1.51%
Gaseous HAPs
Acetaldehyde
Benzene
Formaldehyde
Hydrochloric Acid
Methanol
Methyl Ethyl Ketone
Methyl Isobutyl Ketone
Phenol
Toluene
Xylenes
Total Gaseous HAPs
1,150
580
421
6,890
6,810
469
556
1,330
482
1,159
19,860
12.5
6.43
4.99
79.1
72.5
5.26
6.26
14.6
5.45
13.0
220
17.0
9.10
7.43
115
95.9
7.45
8.91
20.4
7.79
18.5
308
1.08%
1.11%
1.19%
1.15%
1.06%
1.12%
1.13%
1.09%
1.13%
1.12%
1.11%
1.48%
1.57%
1.76%
1.67%
1.41%
1.59%
1.60%
1.53%
1.62%
1.60%
1.55%
11-41
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-8 (Continued)
Pollutant
Emissions, Mg/yra
After MACT
II Baseline
Increase from
Option A
Increase from
Option B
Change from Baseline
Option A
Option B
Participate HAPs
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Cobalt
Lead
Manganese
Mercury
Nickel
Selenium
Total Paniculate HAPs
Total HAPsb
0.91
5.44
0.00
0.91
0.91
0.91
11.8
7.3
0.00
2.72
5.44
37.2
19,890
0.01
0.06
0.00
0.01
0.01
0.01
0.13
0.08
0.01
0.03
0.06
0.42
220
0.02
0.08
0.00
0.01
0.02
0.02
0.19
0.12
0.01
0.05
0.08
0.60
308
1.57%
1.03%
0.00%
1.02%
1.57%
1.34%
1.12%
1.11%
0.00%
1.23%
1.05%
1.13%
1.11%
2.25%
1.48%
0.00%
1.46%
2.25%
1.92%
1.61%
1.59%
0.00%
1.77%
1.51%
1.62%
1.55%
aAll nationwide baseline emissions estimates received from MRI (MRI July 1996, October 1996) (21,29)
bTotal HAPs include gaseous HAPs and paniculate HAPs.
11-42
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-9
Atmospheric Emissions: Oil Combustion and Oil Combustion Plus BLS
Combustion (Before MACT II is Applied) (Mg/yr)
Pollutant3
Oil Combustion11
Increase for
Option A
Decrease for
Option B
Combustion: Total Oil Plus BLS
Increase for
Option A
Decrease for
Option B
Criteria Pollutants
Sulfur Dioxides0
1,794
6,312
2,578
5,166
Particulate HAPs
Antimony
Arsenic
Beryllium
Cadmium
Chromium
Cobalt
Lead
Manganese
Mercury
Nickel
Selenium
Total Particulate
HAPs
0.08
0.14
0.009
0.25
0.16
0.21
0.23
0.10
0.04
3.3
0.08
4.64
0.27
0.51
0.032
0.87
0.57
0.74
0.82
0.36
0.13
11.8
0.28
16.32
0.10
0.24
0.009
0.27
0.18
0.23
0.45
0.24
0.05
3.36
0.18
5.35
0.24
0.37
0.03
0.85
0.54
0.71
0.50
0.16
0.11
11.7
0.14
15.29
"Emissions based on energy consumption in Figure 11-1, Values shown are total kkg/yr for the bleached papergrade
kraft and soda subcategory.
bEmission factors from AP-42, Section 1.3, 5th Edition, 1995.
°Assumes that the average sulfur content of residual oil burned in the US is 0.7 percent (or where high sulfur oil is
used, SO2 emission control equipment is installed to reduce emissions to the equivalent of 0.7 percent sulfur oil).
11-43
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-10
Changes to Carbon Dioxide Emissions Resulting From Changes in
Consumption of Fossil Fuel and Wood Consumption for Option A and Option
B for the Bleached Papergrade Kraft and Soda Subcategory
Option A
Option B
Fossil Fuel Consumption
Increase by 399,000 kkg/yr
Reduce by 1,405,000 kkg/yr
Wood Consumption
Reduce by 245,000 kkg/yr
Reduce by 245,000 kkg/yr
Net Change
Increase by 154,000 kkg/yr
Reduce by 1,650,000
kkg/yr
11-44
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-11
Average CO Emissions From Kraft Bleaching With Chlorine Dioxide
Mill
Code
Bleaching
Sequence
Wood Type
%
Sub.
Total
C1O2
charge
Bleach
Plant
Inlet
Kappa
OD
or
EC
SF1
SD2
SE1
C
c
AA
SE2
C
AA
SF2
SA1
SA2
SD1
AA
SG2
E
E
AA
SGI
B
G
B
AA
F
Cd (Eop)DEpD
CD(Eo) DP
D(Eop)DP
D(Eop)D
D(Eop)D
CdEDED
D( Eo)DP
D(Eop)D
CdEDED
Cd (Eop)DEpD
Cd( Eo) D
CD( Eo)D
Cd( Eo) D
CdEDED
CEDED
Cd(Eop)DED
Cd(Eop)DED
CdEDED
CEDED
D(Eop)D
D(Eop) PD
D(Eop)D
CdEDED
Cd(Eop) D
Hwd
Swd
Swd
Swd
Hwd
Swd
Hwd
Swd
Swd
Swd
Hwd
Swd
Swd
Swd
Swd
Swd
Swd
Swd
Swd
Swd
H/S/Swdust
Swd
Swd
Swd
6%
15%
100%
100%
100%
0%
100%
100%
10%
13%
35%
55%
50%
30%
30%
60%
60%
50%
15%
100%
100%
100%
70%
85%
0.74%
1.08%
1.12%
1.18%
1.20%
1.30%
1.32%
1.43%
1.53%
1.58%
1.65%
1.66%
1.97%
1.98%
2.02%
2.14%
2.19%
2.44%
2.49%
2.67%
2.77%
2.82%
2.90%
3.06%
8.0
11.3
13.3
15.0
10.0
30.0
8.2
12.5
30.0
13.0
9.0
11.3
10.9
30.0
nk
27.5
27.5
30.0
nk
27.0
22.5
27.0
30.0
27.0
y
y
y
y
y
n
y
y
n
y
y
y
y
n
n
n
n
n
n
n
n
n
n
n
Total CO Emissions
g/kkg
Avg.
268
264
295
282
400
332
286
423
314
586
245
195
355
323
532
323
309
364
650
386
423
432
441
455
Min.
250
241
214
136
309
264
327
550
350
459
264
218
641
223
245
91
286
Max.
295
282
345
395
550
309
495
623
295
218
364
564
409
414
236
614
527
750
627
Source of
Data
Mill
Mill
Mill
NCASI
NCASI
Lab
Mill
NCASI
Lab
Mill
Mill
Mill
Mill
Lab
Mill
NCASI
NCASI
Lab
Mill
NCASI
NCASI
NCASI
Lab
NCASI
11-45
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-11 (Continued)
Mill
Code
Bleaching
Sequence
Wood Type
%
Sub.
Total
C1O2
charge
Bleach
Plant
Inlet
Kappa
OD
or
EC
B
G
SB
AA
SG3
SG4
D(Eop)D
D(Eop) PD
D Eop DD
DEDED
CEDED
CEDED
Hwd
Swd
Swd
Swd
Swd
Swd
100%
100%
100%
100%
70%
100%
3.07%
3.39%
3.41%
3.58%
3.63%
3.98%
17.0
30.0
13.0
30.0
nk
nk
n
n
y
n
n
n
Maximum
Minimum
Average
Total CO Emissions
g/kkg
Avg.
295
464
414
527
614
568
650
195
392
Min.
195
414
409
577
568
641
91
329
Max.
423
550
418
650
568
750
218
455
Source of
Data
NCASI
NCASI
Mill
Lab
Mill
Mill
Data in column "OD or EC" were inferred by EPA from bleaching conditions. All other data after Someshwar
(1997) (23).
11-46
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-12
Carbon Monoxide Emissions from Bleach Plants for Option A and Option B
for Bleached Papergrade Kraft and Soda Subcategory
Chlorine dioxide use, kkg/day
Carbon monoxide emissions, kkg/yr
Carbon monoxide emissions, g/kkg pulp
Baseline
1,660
10,900
374
Option A
2,200
12,400
425
Option B
1,740
11,200
381
11-47
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-13
Comparison of NSPS/PSNS to Conventional Technology
Wood Consumption
Effluent Flow
BOD, to Treatment
Sludge Generation
Carbon Dioxide Emissions
1000 tpd Fiber Line
No Difference
Moderate Decrease"
Decrease by 1 1,300 kg/day
Decrease by 890 kg/day
Decrease by 21,700 Mg/yr
Energy Impacts:
Total Electricity Demand
Total Steam Demand
Total Energy Demand
Decrease by 222,600 million BTU/yr in oil equivalent
Increase by 60,180 million BTU/yr in oil equivalent
Decrease by 162,400 million BTU/yr in oil equivalent
Air Emissions:
Hazardous Air Pollutants
Chloroform
Volatile Organic Compounds
Total Reduced Sulfur
Particulate Matter
Carbon Monoxide
Nitrogen Oxides
Sulfur Dioxides
Increase by 407 Mg/yr
No Difference
Increase by 707 Mg/yr
Increase by 28 Mg/yr
Decrease by 12 kg/yr
Decrease by 3 Mg/yr
Decrease by 28 Mg/yr
Decrease by 56 Mg/yr
aSee Section 11.4.1.3
11-48
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-14
Comparison of Two TCF Technologies to Conventional Technology
(Option A) for the Bleached Papergrade Kraft and Soda Subcategory
Wood Consumption
Final Effluent Flow
BOD5 to Treatment
Sludge Generation from BODS
Carbon Dioxide Emissions
TCF-Peroxide
No Difference
Decrease by approximately
90 percent (a)
Decrease by approximately
90 percent (a)
Decrease by approximately
90 percent (a)
Moderate Decrease
TCF-Ozone
No Difference
Decrease by approximately
90 percent (a)
Decrease by approximately
90 percent (a)
Decrease by approximately
90 percent (a)
Moderate Decrease
Energy Impacts:
Total Electricity
Total Steam Demand
Total Energy Demand
Decrease by 4.73 trillion
BTU/yr in oil equivalent
Decrease by 35.1 trillion
BTU/yr in oil equivalent
Decrease by 39.9 trillion
BTU/yr in oil equivalent
Decrease by 7. 1 1 trillion
BTU/yr in oil equivalent
Decrease by 83.8 trillion
BTU/yr in oil equivalent
Decrease by 90.9 trillion
BTU/yr in oil equivalent
Air Emissions:
Hazardous Air Pollutants - Chlorinated
Hazardous Air Pollutants - Non-Chlorinated
Chloroform
Volatile Organic Compounds - Chlorinated
Volatile Organic Compounds - Non-Chlorinated
Total Reduced Sulfur
Paniculate Matter
Carbon Monoxide from C1O2 Application
Elimination (b)
No Difference (c)
Elimination (b)
Elimination (b)
No Difference (c)
No Difference
No Difference
Elimination
Elimination (b)
No Difference (c)
Elimination (b)
Elimination (b)
No Difference (c)
No Difference
No Difference
Elimination
11-49
-------
Section 11 - Non-Water Quality Environmental Impacts
Table 11-14 (Continued)
Nitrogen Oxides
Sulfur Dioxides
TCF-Peroxide
Slight Decrease (d)
Slight Decrease (d)
TCF-Ozone
Slight Decrease (d)
Slight Decrease (d)
(a) Final effluent flow rates are typically 95 m3/kkg (see Section 11.3), of which 25 to 60 m3/kkg typically come
from bleach plant effluents. While TCP operation itself does not require recycling of bleach plant effluents, TCP
facilities operating today typically have bleach plant effluent discharges of 5 to 15 m3/kkg, a bleach plant reduction
of approximately 30 m3/kkg (30). TCP facilities recycle bleach plant effluents to recover bleaching chemicals and to
reduce BOD5 load to and sludge generation in secondary treatment. Industry practice in mills using TCP technology
is to reduce effluent flow throughout the mill. For example, the Metsa-Rauma mill at Rauma, Finland discharges 9
m3/kkg effluent, of which 5 m3/kkg is from the bleach plant (31). 90 percent final effluent flow reduction is
attainable but actual reduction at each mill depends on local site constraints and the priority given to flow reduction.
It is reasonable to assume that comparable decreases would occur for BOD 5 and sludge generation from BOD5.
(b) Any emission results from process water chlorination.
(c) See Revised Draft Chemical Pulping Emission Factor Development Document, DCN A9240IVA8.
(d) Due to decrease in fossil fuel consumption.
11-50
-------
Section 11 - Non-Water Quality Environmental Impacts
5,000,000
4,000,000
3,000,000
2,000,000
1,000,000
0
-1,000,000
-2,000,000
Energy Saved by Alternative Regulatory Scenarios,
barrels Oil Equivalent
Based on 1 00% Implementation of each scenario,
relative to 1 995 status of mills, in bleached papergrade kraft sub-sector
Option B/Tier1
i*l
I iifi^
i NOMNXimMJS
Option A
Figure 11-1. Energy Impacts of Proposed Regulations,
Bleached Papergrade Kraft and Soda Subcategory
11-51
-------
Section 11 - Non-Water Quality Environmental Impacts
700
> 600
5 500
400
300
200
3
f
100
0.00% 0.50% 1.00% 1.50% 2.00% 2.50% 3.00% 3.50% 4.00% 4.50%
Chlorti* Dioxid* charge % on pulp
Figure 11-2. Emissions of CO Measured by NCASI
11-52
-------
Section 12 - Best Conventional Pollutant Control Technology
SECTION 12
BEST CONVENTIONAL POLLUTANT CONTROL TECHNOLOGY
12.1 Introduction
The 1977 amendments to the Clean Water Act established BCT for discharges of
conventional pollutants from existing industrial point sources. BCT is not an additional
limitation, but replaces BAT for the control of conventional pollutants.
This section presents a summary of the final BCT methodology, describes the
revisions made to the BCT technology options and cost estimates for the end-of-pipe treatment
technologies, and discusses the results of the final BCT cost test. This section does not represent
the methodology used to estimate end-of-pipe treatment costs since it has not changed since
proposal. This methodology is described in the Proposed Technical Development Document
(EPA 821-R-93-019) (TDD).
12.2 Summary of the Final BCT Methodology
In considering whether to promulgate revised BCT limits for the Bleached
Papergrade Kraft and Soda (Subpart B) and the Papergrade Sulfite (Subpart E) subcategories,
EPA considered whether technologies are available that achieve greater removals of conventional
pollutants than the current BPT effluent limitations guidelines. EPA also considered whether
those technologies are cost-reasonable according to the BCT cost test, which compares the
incremental removals and costs associated with BCT limitations to a baseline associated with
BPT.
For the final rule, EPA chose to conduct the BCT analysis using estimates of
industry's 1989 discharge of conventional pollutants (based on data from the 1990 industry
census) as the BPT baseline against which BCT technology options are analyzed. EPA evaluated
the candidate BCT technologies and concluded that none of the options passed the BCT cost test.
Therefore, at this time, EPA is not promulgating more stringent BCT limitations for Subparts B
and E of this industry category. BCT limitations for former Subparts G, H, I, and P (now Subpart
B) and former Subparts J and U (now Subpart E) remain in effect.
12.3 Revisions to the Proposed BCT Technology Options and Cost Estimates
Based on comments received on the proposed rule and the July 15, 1996 Notice of
Data Availability (61 FR 36835) and correction of double-counting some costs, EPA revised the
inputs to the BCT cost test related to BCT option performance, the costs to install or upgrade
end-of-pipe treatment systems, and the annual operating costs for end-of-pipe treatment systems.
This section describes only these revisions (see also Comment Response Document, DCN 14497,
Volume IV, "BCT Cost Test").
12-1
-------
Section 12 - Best Conventional Pollutant Control Technology
12.3.1 Conventional Pollutant Control Option Performance
At proposal, EPA developed four candidate BCT options:
(1) Option A.I - Performance level of the best-performing mill in each
subcategory assuming the baseline performance is equal to the proposed
BPT Option 2;
(2) Option A.2 - Multimedia filtration assuming the baseline performance is
equal to the proposed BPT Option 2;
(3) Option B.I - Performance level representing the average of the best 90
percent of mills in each subcategory assuming the baseline performance is
equal to current industry performance;
(4) Option B.2 - Performance level representing the average of the best 50
percent of mills in each subcategory assuming the baseline performance is
equal to current industry performance.
Two of these options, Options A.I and A.2, assumed the baseline performance to
be equal to the proposed BPT Option 2. Because EPA has decided not to revise BPT limitations
for conventional pollutants, EPA chose not to evaluate these two options. For the final rule for
the Bleached Papergrade Kraft and Soda Subcategory, EPA only considered Options B.I and
B.2, known now simply as BCT Options 1 and 2.
EPA identified a new BCT option for the Papergrade Sulfite Subcategory, which
is the average performance level achieved by the three mills with at least 85 percent of their
production in the subcategory whose wastewater treatment performance reflects BCT candidate
level performance. Final production of most mills in the Papergrade Sulfite Subcategory is
comprised of a large portion of purchased pulp. For the proposed rule, BCT option performance
levels for this subcategory were calculated using data from mills with 37 to 96 percent of their
final production in the subcategory. After proposal, EPA reassessed the impact of purchased
pulp on the final effluent BOD5 load and determined that four mills with 85 percent or more of
final off-machine production derived from sulfite pulp produced on site discharged substantially
higher BOD5 loads from secondary biological wastewater treatment. EPA determined that
effluent from these few mills more appropriately represented wastewater from the subcategory.
EPA used data from three of these mills to characterize BCT, because EPA did not consider the
treatment performance of the fourth mill to be representative of the subcategory as a whole. This
fourth mill treats wastewater from liquor by-products manufactured on site, which is unique
among papergrade sulfite mills.
Section 8.3 of this document presents a detailed discussion of the development of
options for the control of conventional pollutant discharges.
12-2
-------
Section 12 - Best Conventional Pollutant Control Technology
12.3.2 Accounting for Cluster Rules Impacts on BCT Costs
In order to conduct the BCT cost test, discussed in Section 12.4.4. EPA estimated
the pollutant removals achieved by the industry in upgrading from BPT to BCT. EPA also
estimated the cost for upgrading from BPT to BCT.
EPA estimated the pollutant removals achieved by the BCT Options by
calculating the reduction in each mills discharged pollutant load at Options 1 and 2, relative to
the baseline (pollutant load discharged in 1989, as reported in the 1990 industry census). At
proposal, EPA attributed the entire increment to BCT, even though some conventional pollutant
removals result from other components of the Cluster Rules.
EPA estimated the cost for each mill to upgrade its treatment system from the
baseline 1989 performance to the Option 1 and Option 2 level of performance. Before EPA
estimated the costs of BPT at proposal it accounted for the pollutant load reduction that will
result from implementation of the cluster rules (i.e., BAT, BMP, and NESHAP). Applying these
load reductions reduced the load of BOD5 requiring treatment, and therefore also reduced the
costs associated with end-of-pipe treatment system upgrades. At proposal, EPA used these BPT
costs to represent the costs of BCT Options B.I and B.2.
EPA received comments questioning this approach. When EPA revised the inputs
to the BCT cost test, it reevaluated this approach and, for the final rule, decided not to account
for cluster rule impacts on costs because they were not also taken into account when estimating
loads.
12.3.3 Engineering Cost Estimates
EPA received several comments pertaining to the estimate of costs to install
and/or upgrade end-of-pipe treatment systems for the removal of conventional pollutants. The
Agency has not changed its approach to designing end-of-pipe treatment systems, but has revised
certain unit costs which comprise the capital and annual operating and maintenance costs for
end-of-pipe treatment and some assumptions based on comments received on the proposed rule.
This section presents a summary of the individual unit costs and assumptions that were revised or
reevaluated prior to their inclusion in the final BCT cost test, including aerated stabilization basin
liner costs, activated sludge aeration and sludge handling costs, flow reduction costs, indirect
cost factors, land costs, and polymer addition. All other costs that comprise end-of-pipe
treatment system installation and operation have been described previously in the TDD
supporting the proposed rule.
12.3.3.1 Aerated Stabilization Basin Liner Costs
Based on comments received on the proposed rule, EPA reevaluated the unit cost
to install liners for aerated stabilization basins. At proposal, EPA estimated clay liner costs to be
$0.37 per square foot (1991 dollars). Commenters questioned whether these costs accounted for
12-3
-------
Section 12 - Best Conventional Pollutant Control Technology
transportation of the clay to mills that were not located near a vendor. Therefore, EPA
reevaluated the transportation costs associated with installing clay liners and found that, in some
cases, transportation would add over $2.00 per square foot to the cost of installing clay liners.
Because of the relatively high clay transportation costs, EPA investigated
alternative materials for use as aerated stabilization basin liners, including plastic (HDPE) liners
and geosynthetic liners. Based on the information provided by vendors, HDPE liners would cost
approximately $0.35 per square foot, including transportation and installation, while geosynthetic
liners would cost $0.57 per square foot, including transportation and installation. EPA chose to
estimate basin costs assuming installation of a geosynthetic liner, which includes excavating the
lagoon, laying the liner, and covering the liner with 6 to 8 inches of backfill from the original
excavation.
12.3.3.2 Activated Sludge Aeration Costs
EPA reevaluated aeration costs for mills with existing activated sludge systems.
When designing upgrades to achieve the target BOD5 and TSS loads, EPA considered additional
aeration tank volume and/or aeration, as well as operational modifications. In some cases, EPA
determined that the mill has sufficient existing aeration capacity, but requires additional retention
time to meet the target long-term average conventional pollutant concentrations. In these cases,
additional aeration costs were eliminated from the compliance cost estimates.
12.3.3.3 Activated Sludge Handling Costs
As described in the TDD, EPA estimated costs for the management and disposal
of sludge from new or upgraded activated sludge units. These incremental costs were based on
the increased amounts of sludge resulting from upgrades as determined by the activated sludge
design model.
At proposal, the Agency calculated mill-specific unit costs to estimate the capital
and annual operating and maintenance (O&M) costs for sludge handling and disposal. EPA
assumed that sludge increases of 5 percent or less could be handled with no additional cost.
Following proposal, EPA determined that sludge increases up to 15 percent could be handled
with no additional cost. However, EPA also determined that, for mills that required increased
sludge handling capacity, a minimum capital cost would be incurred that was greater, in some
cases, than the incremental capital cost that was previously estimated.
12.3.3.4 Flow Reduction Costs
As described in Section 11.6.1.1 of the TDD, some mills could not achieve target
BOD 5 or TSS loads for the candidate BCT options with end-of-pipe treatment alone because their
target BOD5 and TSS concentrations were lower than the lowest currently demonstrated
concentrations. Therefore, as at proposal, these mills required flow reduction to increase their
12-4
-------
Section 12 - Best Conventional Pollutant Control Technology
target concentrations. The technologies that reduce flow also reduce BOD5 and TSS loads. Flow
reduction, if required, was applied to a mill first; the revised final effluent flow rates, BOD 5 load,
and TSS load achieved with flow reduction were entered into models to design and cost end-of-
pipe wastewater treatment upgrades that would enable the mill to achieve the target loads.
For some mills, the cost to meet the more stringent candidate BCT option (Option
2) was lower than the cost to meet Option 1 because Option 1 costs were based on end-of-pipe
treatment upgrades only and not on flow reduction. For these mills, EPA decided for the final
rule to cost flow reduction for both candidate BCT options, which resulted in a lower cost to
comply with BCT Option 1.
12.3.3.5 Indirect Cost Factors
In response to comments, EPA reviewed the indirect cost factors used to estimate
overhead and profit, engineering, and contingency costs. The indirect cost factors used at
proposal totaled 30 percent of the total direct capital costs. Commenters believed that indirect
costs could range from 39 percent to 85 percent of the total direct capital costs. EPA reviewed
data submitted by commenters and consulted a standard chemical engineering design text. (1)
EPA found that an indirect cost factor of 30 percent was reasonable for the final design stage;
however, an indirect cost factor of 45 percent was more appropriate for cost estimates at the
conceptual stage. Therefore, because EPA's costs are "conceptual design" not "final design" EPA
revised the end-of-pipe compliance costs using an indirect cost factor of 45 percent.
12.3.3.6 Land Costs
While in the process of revising end-of-pipe cost estimates, EPA considered
whether certain mills actually required additional capital costs to purchase land on which to build
or expand end-of-pipe treatment systems. For certain mills for which EPA had estimated land
purchase costs at proposal, EPA reviewed information collected during recent site visits.
Although this information suggested these mills may have sufficient room to accommodate the
estimated treatment upgrades, because this information contradicted information reported in the
1990 Census Questionnaire and similar information was not available for all mills, for the final
rule, EPA chose not to revise (i.e., reduce) land purchase costs. The result is more conservative
(i.e., higher) compliance cost estimates.
12.3.3.7 Polymer Addition
At proposal, the polymer addition upgrade was developed based on best
professional judgment. As discussed in the Proposed TDD, polymer addition was costed at a rate
of 5 mg/L based on final effluent flow rate. Comments received on the proposal from pulp mills
stated that EPA had overcosted polymer addition by using too high a rate of addition. EPA
contacted additional pulp mills and determined that a polymer addition rate of 1.1 mg/L was
more appropriate. Therefore, EPA revised the cost of polymer addition, which resulted in lower
compliance costs.
12-5
-------
Section 12 - Best Conventional Pollutant Control Technology
12.3.3.8 Calculation of Total Annualized Costs
One of the components of total annualized costs for both the proposed and final
rulemakings is annual general and administrative costs (GAC), such as insurance, which is
included in annual O&M costs. For proposal, EPA estimated GAC as 4 percent of capital costs
plus 60 percent of O&M costs for purposes of determining BCT-candidate option costs. After
proposal, EPA determined that the 60 percent component was already included in the engineering
estimates. For the final rule, therefore, EPA estimated GAC for BCT-candidate option costs as
only 4 percent of capital. (For more discussion of their impact on this change see, Comment
Response Document, DCN 14497, Volume IV, "BCT Cost Test.")
12.4 Final BCT Methodology
Components of the BCT methodology used for EPA's final evaluation of BCT
Options 1 and 2 are described in the following sections.
12.4.1 BCT Technology Basis
As discussed in Section 12.2.1, EPA identified two final candidate BCT options
for the Bleached Papergrade Kraft and Soda Subcategory and one final candidate BCT option for
the Papergrade Sulfite Subcategory. As at proposal, EPA calculated the BCT performance levels
as a function of effluent pollutant concentration, mill production, and effluent flow rate. Both in-
process flow reduction technologies and end-of-pipe wastewater treatment can be combined to
achieve the BCT performance levels.
The technologies used to estimate the cost to comply with BCT Options 1 and 2
are the same technologies used at proposal. Technologies were selected based on the operations
present at each mill, as reported in the 1990 census questionnaire. Costed technologies include:
Flow reduction technologies
Paper machine vacuum pump seal water recycle;
Screen room closure; and/or
Reuse of deinking washwater after flotation clarification.
End-of-pipe wastewater treatment
Primary clarification;
Aerated stabilization basins; and/or
Activated sludge systems.
12-6
-------
Section 12 - Best Conventional Pollutant Control Technology
12.4.2 End-of-Pipe Treatment Costs
The final BCT compliance cost estimates for the Bleached Papergrade Kraft and
Soda (BPK) and the Papergrade Sulfite (PS) subcategories are shown in Table 12-1. EPA
estimated that BPK mills would incur an average capital cost of $1.2 million (and a maximum
cost of $14.5 million) for Option 1 and an average capital cost of $2 million (and a maximum
cost of $17.4 million) for Option 2. Annual O&M costs for the two options ranged from zero to
$3.5 million. PS mills would incur an average capital cost of $1.2 million (and a maximum cost
of $3.9 million). Annual O&M costs ranged from zero to $0.3 million.
12.4.3 Conventional Pollutant Removals
The final BCT option conventional pollutant removals for the Bleached
Papergrade Kraft and Soda and the Papergrade Sulfite subcategories are shown in Table 12-2.
EPA estimated that BPK mills would remove an average of 1.6 million pounds of BOD5 and TSS
from pulp mill effluents for Option 1 and an average of 2.5 million pounds of BOD5 and TSS for
Option 2. PS mills would remove an average of 57,000 pounds of BOD5 and TSS under the
candidate option.
12.4.4 BCT Cost Test Results
The background, application, and results of the BCT cost test are discussed in the
following sections.
12.4.4.1 Background
The 1977 Clean Water Act amendments added Section 304(b)(4), which
established BCT for direct discharges of conventional pollutants from existing industrial point
sources. Effluent limitations based on BCT may not be less stringent that the limitations based
on BPT. Thus, BPT effluent limitations are a "floor" below which BCT effluent limitations
cannot be established.
The Clean Water Act amendments that created BCT also specify that the cost
associated with BCT effluent limitations be "reasonable" with respect to the effluent reductions.
Accordingly, the BCT methodology was developed to answer the question of whether it is "cost-
reasonable" for industry to control conventional pollutants at a level more stringent than that
already required by BPT effluent limitations.
As promulgated in July 1986 (51 FR 24974), the first step in establishing BCT
effluent limitations for an industry (or a subcategory within an industry) is to identify candidate
technologies that provide conventional pollutant control beyond the level achieved by the BPT
effluent limitations. The next step is to evaluate these technology options by applying the two-
part BCT cost test. To "pass" the POTW test (the first part of the test), the cost per pound of
conventional pollutant removed by industrial dischargers in upgrading from BPT to BCT must be
12-7
-------
Section 12 - Best Conventional Pollutant Control Technology
less than the cost per pound of conventional pollutant removed in upgrading POTWs from
secondary treatment to advanced secondary treatment. The POTW upgrade cost is referred to as
the POTW benchmark; its derivation is described in the 1986 final BCT methodology notice.
The second part of the test that the BCT technology must pass is the industry cost-
effectiveness test. This test is actually a ratio of two incremental costs: (1) the cost per pound
removed by the BCT technology relative to BPT; and (2) the cost per pound removed by BPT
relative to no treatment (i.e., "raw" waste load or "baseline"). The ratio of these two costs is a
measure of the BCT technology's cost effectiveness. As in the POTW test, this ratio is compared
to a calculated industry cost benchmark. If the industry ratio is lower than the benchmark, the
BCT technology passes the cost test.
EPA evaluates both tests as measures of reasonableness. As such, if the BCT
technology passes both the POTW and industry cost test, than the most stringent technology
option among them becomes the basis for setting BCT effluent limitations. Alternately, if no
candidate technology more stringent than BPT passes, then BCT effluent limitations are set equal
to BPT effluent limitations.
12.4.4.2 Application of the Final BCT Cost Test
EPA reviewed the control and treatment technology alternatives available for
application in the pulp, paper, and paperboard industry for the control of conventional pollutants.
As mentioned in Section 12.2.1, the result was to define two final candidate technologies for
BPK mills and one final candidate technology for PS mills.
The BCT cost test calculations rely on cost and performance data from the 1990
industry census questionnaire and on results from the wastewater treatment model used at
proposal (with the changes noted in Section 12.2.3). Since no information is available that
accurately depicts the costs for the removal of conventional pollutants from raw waste load
(RWL) to the current BPT regulation, the BCT cost test was performed using the LTA as derived
from the 1989 performance data collected in the 1990 industry census questionnaire to represent
BPT level of performance of biological treatment (biological treatment constitutes the basis of
the current BPT). The source of cost estimates for wastewater treatment upgrades to achieve
BCT was the end-of-pipe treatment system cost model used at proposal, which estimates capital
and operating engineering costs. These costs were annualized using a cost annualization model
that estimates the cost actually incurred by the mill to upgrade its pollution controls. This model
takes into account tax savings the business accrues through depreciation and other tax shields.
Given these inputs, the BCT cost test was performed for two option for BPK mills
and one option for PS mills:
BPK - Option 1 - Current LTA to the average of the best performing 90
percent of BPK mills;
12-8
-------
Section 12 - Best Conventional Pollutant Control Technology
BPK - Option 2 - Current LTA to the average of the best performing 50
percent of BPK mills; and
PS Option - Current LTA to the average of PS mills.
For each option, the pounds of pollutant removed and cost incurred for the POTW cost test are
the incremental pounds and costs associated with secondary wastewater treatment upgrades and
necessary flow reduction. The ratio of incremental costs compared to incremental pounds
constitutes the first part of the BCT cost test.
The second part, the industry cost-effectiveness test, requires the computation of a
ratio of two incremental costs. The first incremental cost is the cost per pound for the removal of
conventional pollutants incurred by industry in upgrading from BPT to BCT (i.e., the ratio from
the first part of the cost test). The second incremental cost is the cost per pound for the removal
of conventional pollutants incurred by industry to meet BPT relative to no treatment (i.e., RWL to
current LTA). The ratio of the first cost to the second cost is the measure of the BCT
technology's cost effectiveness.
The next step in the BCT cost test is to compare the two tests' results (or ratios)
to the POTW and industry benchmarks. As explained above, the ratios calculated for the BCT
cost test must be less than the POTW and industry benchmarks, respectively, to pass the BCT
cost test. For a more detailed explanation of the benchmarks, refer to the 1986 notice of final
regulation for the BCT methodology. In this analysis, the benchmarks are as defined in that
notice, but indexed to comparable year data as the costs of treatment. As at proposal, EPA has
indexed all costs to 4th quarter 1991 dollars.
Finally, the costs and pollutant removals for each mill were apportioned according
to subcategory production. Baseline (RWL to current LTA) costs and pollutant removals are
apportioned because mills are meeting their current limits based on all production at the facility,
not just BPK and PS production. BCT option costs and pollutant removals are apportioned
because the mill LTAs are calculated using proposed LTAs for all subcategories, not just the new
LTAs for the BPK and PS subcategories. Mill LTAs for each option were calculated by
multiplying the mill's percent production for each subcategory by the subcategory proposed LTA
and then summing. Therefore, the calculations for both baseline and BCT option costs and
pollutant removals are:
Costs = Total annualized cost x percent production in BPK (or PS)
Removals = Total pollutant removals (BOD5 + TSS) x percent production in BPK (or PS)
PULP97.TDD/17 November 1997 12-9
-------
Section 12 - Best Conventional Pollutant Control Technology
12.4.4.3 Cost Test Results
The final BCT cost test results for the Bleached Papergrade Kraft and Soda and
the Papergrade Sulfite Subcategories are shown in Table 12-3. None of the candidate technology
options passed the BCT cost test. Therefore, EPA is not promulgating more stringent BCT
effluent limitations guidelines for Subparts B and E at this time. Rather, the BCT limitations
promulgated for former Subparts G, H, I, and P (now Subpart B) and former Subparts J and U
(now Subpart E) remain in effect.
12.5 References
1. Means. Building Construction Cost Data. 54th Annual Edition, 1996.
PULP97.TDD/17 November 1997 12-10
-------
Section 12 - Best Conventional Pollutant Control Technology
Table 12-1
Best Conventional Pollutant Control Technology (BCT) Costs
Subcategory1
Bleached Papergrade Kraft and
Soda
Papergrade Sulfite
BCT Option
Option 1
(best 90%)
Option 2
(best 50%)
Option 1
(average)
Capital ($)
102,006,505
172,405,961
11,030,865
Engineering
O&M ($/yr)
12,059,708
18,724,362
593,202
Total
Annualized
Costs (1995 $)
17,181,639
28,092,172
1,451,886
'Costs for mills with operations in more than one subcategory have been apportioned based upon annual production.
PULP97.TDD/17 November 1997
12-11
-------
Section 12 - Best Conventional Pollutant Control Technology
Table 12-2
Conventional Pollutant Reductions Associated With BCT
Subcategory1
Bleached Papergrade Kraft and Soda
Papergrade Sulfite
BCT Option
Option 1
(best 90%)
Option 2
(best 50%)
Option 1
(average)
BOD5 Reductions
(Ib/yr)
47,108,171
74,873,613
3,115,156
TSS Reductions
(Ib/yr)
73,738,529
118,476,733
4,065,428
PULP97.TDD/17 November 1997
12-12
-------
Section 12 - Best Conventional Pollutant Control Technology
Table 12-3
Results of the Final BCT Cost Test
Bleached
Papergrade Kraft
and Soda - Option 2
(best 50%)
Bleached
Papergrade Kraft
and Soda - Option 1
(best 90%)
Papergrade
Sulfite - Option 1
(average)
A. POTW Test
Total Annualized BCT Costs (1995 $)
Pounds Removed by BCT
Industry BCT Benchmark (cost per pound
removed relative to current LTA)
POTW Benchmark (cost per pound to
upgrade to advanced secondary treatment)
Is BCT cost < POTW cost?
Pass/Fail Test
$28,092,172
193,350,346
0.15
0.48
YES
PASS
$17,181,639
120,846,700
0.14
0.48
YES
PASS
$1,451,886
7,180,583
0.20
0.48
YES
PASS
B. Industry Cost Test
RATIO 1
Total Annualized Costs (RWL to Current
LTA)
Pounds Removed (RWL to Current LTA)
Bl. Industry BCT Cost (from Part A)
B2. Industry Current Cost (RWL to Current
LTA)
RATIO 1: Ratio of Bl to B2
$241,694,766
3,375,724,754
0.15
0.07
2.03
$241,694,766
3,375,724,754
0.14
0.07
1.99
$19,439,167
277,287,784
0.20
0.07
2.88
RATIO 2
Cl. POTW Benchmark
C2. POTW Cost to upgrade from no
treatment to secondary
RATIO2: Ratio of Cl to C2
Is Ratio 1 < Ratio 2?
Pass/Fail Test
0.48
0.37
1.29
NO
FAIL
0.48
0.37
1.29
NO
FAIL
0.48
0.37
1.29
NO
FAIL
PULP97.TDD/17 November 1997
12-13
-------
Section 13 - Abbreviations and Conversions
SECTION 13
ABBREVIATIONS AND CONVERSIONS
13.1
Abbreviations
2,3,7,8-TCDD
2,3,7,8-TCDF
ACM
AF&PA
AOX
ASB
AST
BAT
BCT
BID
BFR
BLS
BMP
BOD5
BPK
BPT
C
CAA
CBI
CDD
CDF
CEK
CEM
CFR
2,3,7,8-tetrachlorodibenzo-p-dioxin
2,3,7,8-tetrachlorodibenzofuran
active chlorine multiple
American Forest and Paper Association
Adsorbable organic halides. A bulk parameter which measures
the total chlorinated organic matter in wastewater.
aerated stabilization basin
activated sludge treatment
Best Available Technology Economically Achievable
Best Conventional Pollutant Control Technology
Background Information Document: Pulp, Paper, and
Paperboard Industry—Background Information for Proposed Air
Emission Standards (October, 1993)
bleach filtrate recycle
black liquor solids
Best Management Practices
Five-day biochemical oxygen demand
bleached papergrade kraft and soda mills
Best Practicable Control Technology
bleach sequence symbol for chlorine stage
Clean Air Act
confidential business information
chlorinated dibenzo-p-dioxin
chlorinated dibenzofuran
target kappa number
continuous emission monitor
Code of Federal Regulations
13-1
-------
Section 13 - Abbreviations and Conversions
C1O2
CMN
CO2
COD
CTMP
CWA
D
DBD
DBF
DCN
E
EA
EAD
EC
ECF
EMCC®
EPA
FR
GAC
H
HAP
HVLC
HW
ISO
ITC®
chlorine dioxide
corrugated, molded and newsprint
carbon dioxide
chemical oxygen demand
chemi-thermo-mechanical pulp
Clean Water Act
bleach sequence symbol for chlorine dioxide stage
dibenzo-p-dioxin
dibenzofuran
document control number
bleach sequence symbol for extraction stage
Economic Analysis for the National Emission Standards for
Hazardous Air Pollutants for Source Category: Pulp and Paper
Production; Effluent Limitations Guidelines, Pretreatment
Standards, and New Source Performance Standards: Pulp,
Paper, and Paperboard Category - Phase I, Record Section 30.5,
DCN 14649.
Engineering and Analysis Division
extended cooking
elemental chlorine-free
extended modified continuous cooking, a registered trademark of
Kamyr, Inc.
U.S. Environmental Protection Agency
Federal Register
general and administrative costs
bleach sequence symbol for hypochlorite stage
hazardous air pollutant
high concentration low volume
hardwood
International Organization for Standardization
Iso Thermal Cooking, a registered trademark of Kvaerner
13-2
-------
Section 13 - Abbreviations and Conversions
LTA
LTS
MACT
MCC®
ML
N
NA
Na2SO4
NAICS
NC
NC
NCASI
ND
ND
NESHAP
NPDES
NRDC
NSPS
O
O&M
OAR
OD
P
PCS
PH
PMP
long-term average
long-term study
Maximum Achievable Control Technology
Modified Continuous Cooking, a registered trademark of Kamyr,
Inc.
minimum level
bleach sequence symbol indicating the absence of a washing
stage
not applicable
sodium sulfate
North American Industry Classification System
not costed
not counted
National Council of the Paper Industry for Air and Stream
Improvement, Inc.
not detected
not disclosed to prevent compromising confidential business
information
National Emission Standards for Hazardous Air Pollutants
National Pollutant Discharge Elimination System
Natural Resources Defense Council
New Source Performance Standards
bleach sequence symbol for oxygen stage
operating and maintenance
Office of Air and Radiation
oxygen delignification
bleach sequence symbol for peroxide stage
permit compliance system
negative logarithm of the effective hydrogen-ion concentration in
moles per liter, a measure of acidity
pollutant minimization program
13-3
-------
Section 13 - Abbreviations and Conversions
POTW publicly owned treatment works
PS papergrade sulfite mills
PSES Pretreatment Standards for Existing Sources
PSNS Pretreatment Standards for New Sources
Q bleach sequence symbol for acid chelant stage
QA quality assurance
QC quality control
RDH® Rapid-Displacement Heating, a registered trademark of Beloit
Corp.
RWL raw waste load
S bleach sequence symbol for sodium bisulfite
SCC Sample Control Center
SIC Standard Industrial Classification
STDD Supplemental Technical Development Document for Effluent
Limitations Guidelines and Standards, for the Pulp, Paper, and
Paperboard Category Subpart B (Bleached Papergrade Kraft and
Soda) and Subpart E (Papergrade Sulfite), October 1997
STFI Swedish Forest Products Research Institute
SW softwood
TCP totally chlorine-free
TDD Technical Development Document for Proposed Effluent
Limitations Guidelines and Standards for the Pulp, Paper, and
Paperboard Point Source Category, October 1993
TEQ toxic equivalent
TRS total reduced sulfur
TSS total suspended solids
UBK unbleached kraft mills
Z bleach sequence symbol for ozone stage
13-4
-------
Section 13 - Abbreviations and Conversions
13.2
ADMT
ADT
atm
bbl
BTU
d
g
G
kg
kkg
kPa
kWh
J
L
m3
mg
M
MOD
ng
OMMT
OMT
Pg
ppb
ppm
ppq
ppt
psi
Units of Measure
air dry metric ton
air dry (short) ton
atmosphere
barrel
British Thermal Unit
day
gram
giga
kilogram
1,000 kilograms = 1 metric ton = 1 mega gram
kilopascal
kilo Watt hour
joule
liter
cubic meter
milligram
mega
million gallons per day
nanogram
off-machine metric ton
off-machine (short) ton
picogram
part per billion
part per million
part per quadrillion
part per trillion
pounds per square inch
microgram
13-5
-------
Section 13 - Abbreviations and Conversions
UBADt unbleached air dry ton
UBMt unbleached metric ton
W watt
yr year
13.3 Unit Conversions
Table 13-1 presents mass and concentration unit conversions used throughout this
document.
13-6
-------
Section 13 - Abbreviations and Conversions
Table 13-1
Units of Measurement
Mass Units
Unit
Gigagram
Megagram or Metric ton
Kilogram
Gram
Milligram
Microgram
Nanogram
Picogram
Femtogram
Unit Abbreviation
Gg
Mg or kkg
kg
g
mg
ng
Pg
fg
Equivalent Mass in Grams
1,000,000,0000
1,000,000
1,000
1
0.001
0.000001
0.000000001
0.000000000001
0.000000000000001
Concentration Units
Unit Abbreviation
ppm (10-6)
ppb (10-9)
ppt(10-12)
ppq (10-15)
Liquids
mg/L
ng/L
Pg/L
Solids
mg/kg =
ng/kg = pg/g
pg/kg = fg/g
Notes: (1) For liquids, conversions from metric concentration unit to ppm, ppb, ppt, and ppq are
approximate.
(2) 1.0 kg = 2.2046 Ibs.
Source: American Petroleum Institute
Publication No. 4506, March 1990
13-7
------- |